Advances in Insect Physiology
Volume 13
This Page Intentionally Left Blank
Advances in Insect Physiology edited by...
4 downloads
407 Views
19MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Advances in Insect Physiology
Volume 13
This Page Intentionally Left Blank
Advances in Insect Physiology edited by
J E TREHERNE M. J. BERRIDGE and V. 6. WIGGLESWORTH Department ofZoology, The University Cambridge, England
Volume 13
1978
ACADEMIC PRESS LONDON NEW YORK SAN FRANCISCO A Subsidiarv of Harcourr Brace Jovanovich. Publishers
ACADEMIC PRESS INC. (LONDON) LTD 24/28 Oval Road London NW 1
United States Edition published by ACADEMIC PRESS INC. 11 1 Fifth Avenue New York. New York 10003
Copyright @ 1978 by ACADEMIC PRESS INC. (LONDON) LTD
AN Rights Reserved
No part of this book may be reproduced in any form by photostat, microfilm, or any other means, without written permission from the publishers
Library of Congress Catalog Number: 63-14039 ISBN: 0-12-024213-3
PRINTED IN GREAT BRITAIN AT THE SPOTTISWOODE BALLANTYNE PRESS BY WILLIAM CLOWES AND SONS LIMITED LONDON, COLCHESTER AND BECCLES
Contributors Robert P. Bodnaryk
Canada Agriculture, Research Station, 195 Dafoe Road, Winnipeg, Manitoba R3T2M9, Canada Norbert Elsner
Zoologisches Institut der Universitatzu Koln, 5 Koln-Lindenthal, WeyertalII9,K67n, Germany Bernd Heinrich
Division of Entomology, University of California, Berkeley, California 94720, USA Ann E. Karnrner
Division of Biology, Kansas State University,Manhattan, Kansas 66506, USA Dennis R. Nelson
Metabolism and Radiation Research Laboratory, Agricultural Research Service, US.Department of Agriculture, Fargo, North Dakota 58102, USA Andrej V. Popov
Sechenov Institute of Evolutionary Physiology and Biochemistry, Leningrad, USSR Richard H. White
Biology Department, University of Massachusetts at Boston, Boston, Massachusetts 02125, USA
This Page Intentionally Left Blank
Contents
Contributors
V
Long-Chain Methyl-Branched Hydrocarbons: Occurrence, Biosynthesis, and Function DENNIS R. NELSON
1
Insect Visual Pigments RICHARD H. WHITE
35
Structure and Function of Insect Peptides ROBERT P. BODNARYK
69
Insect Flight Metabolism ANN E. KAMMER AND BERND HEINRICH
133
Neuroethology of Acoustic Communication NORBERT ELSNER AND ANDREJ V. POPOV
229
Subject Index
351
Cumulative List of Authors
373
Cumulative List of Chapter Titles
375
This Page Intentionally Left Blank
Long-Chain Methyl-Branched Hydrocarbons: Occurrence, Biosynthesis, and Function Dennis R. Nelson Metabolism and Radiation Research Laboratory, Agricultural Research Service, US.Department of Agriculture. Fargo, North Dakota, USA
1 Introduction 1 2 Occurrence 2 2.1 n-Alkanes and n-alkenes 2 2.2 Cycloalkanes 3 2.3 2- and 3-methylalkanes 3 2.4 Internally branched methylalkanes:Analysis 4 2.5 Monomethylalkanes 6 2.6 Dimethylalkanes 13 2.7 Trimethylalkanes 16 3 Biosynthesis 17 4 Functions 21 References 25
1
Introduction
Surface waxes or lipids of all organisms are responsible for the water-repellent character of their surfaces. For example, the skins of higher animals are kept soft, smooth, and free of cracks by lipids [largely squalene, mono-, di-, and triacyl glycerols, wax esters, and fatty acids in man (Nicolaides, 1974)l. By keeping the skin pliable and continuous, microorganisms are unable to penetrate and cause infections, and the skin surface is prevented from drying out and becoming rough and scaly. Also, the surface lipids of plants (see reviews by Caldicott and Eglinton, 1973 and Kolattukudy, 1975) and insects are important because (1) they allow the uptake of water but prevent excessive water loss when available moisture is low (Beament, 1964, 1967; Browning, 1967); (2) they prevent the penetration of inorganic chemicals (Beament, 1964); (3) they act as a bamer against microorganisms (David, 1967); (4) they affect the absorption of agricultural 1
2
DENNIS R NELSON
chemicals (Ebeling, 1964) [in plants, their formation is inhibited by some herbicides (Still et al., 1970; Kolattukudy and Brown, 1974)l; (5) they may serve as a sex attractant (Evans and Green, 1973); and (6) they may serve as a kairomone for insect parasites and predators (Lewis et al., 1975a,b, 1976). In the present review, I have restricted myself to a consideration of the hydrocarbon components of the surface lipids, particularly to the long-chain internally branched methylalkanes and methylalkenes. These compounds have been extensively investigated since 1970 when di- and trimethylalkanes were identified in an insect (Nelson and Sukkestad, 1970) and the technique of identifying mixtures of the methylalkanes from their mass spectra was elucidated (McCarthy et al., 1968; Nelson and Sukkestad, 1970). The majority of studies of the occurrence and function of the long-chain hydrocarbons has been done with insects. The studies of biosynthesis of alkanes and the origin of the methyl groups have been done largely with plants and microorganisms though some of the more recent investigations have involved insects and other arthropods.
2
2.1
Occurrence n-ALKANES AND n-ALKENES
Although in some insects, the surface lipids are mainly long-chain alcohols (Bowers and Thompson, 1965; Bursell and Clements, 1967), ketoesters (Meinwald et al., 1975), and wax esters (Gilby, 1957a,b; Faurot-Bouchet and Michel, 1964, 1965; Brown, 1975), alkanes are a common component of both insect and plant surface lipids and are ubiquitous hydrocarbons in nature. The hydrocarbons of insects usually occur as mixtures, however only n-alkanes were reported from the hydrocarbon fraction of the lipids from larval cast skins of the beetle, Tenebrio molitor L. (Bursell and Clements, 1967). In addition to the alkanes, alkenes have been reported in the wax of bees, Apis mellifera L. (Streibl et al., 1966), the little house fly, Fannia canicularis (L.) (Uebel et al., 1975a), the house fly, Musca domestica L. (Louloudes et al., 1962; Carlson et al., 1971), the house cricket, Acheta domesticus L. (Hutchins and Martin, 1968), the boll weevil, Anthonomus grandis Boh. (Hedin et al., 1974), the stonefly, Pteronarcys californica Newport (Armold et al., 1969), the cockroaches Periplaneta australasiae (F.) and P. brunnea Burmeister, and P. fuliginosa (Serville) (Jackson, 1970), P. japonica Karny and P. americana L. (Jackson, 1972), the Argentine ant, Zridomyrmex humilis (Mayr) (Cavill and Houghton, 1973), the bull ant, Myrmecia gulosa (F.) (Cavill and Williams, 1967), the fleshfly, Sarcophaga bullata Parker (Jackson et al., 1974), the stable
LONG-CHAIN METHYL-BRANCHED HYDROCARBONS
3
fly, Stomoxys calcitrans (L.) (Uebel et al., 1975b), the face fly, Musca autumnalis De Geer (Uebel et al., 1975~).The pecan weevil, Curculio caryae (Horn), has alkenes and akadienes from 20 to 28 carbons in length (Mody et al., 1975) and volatiles of the confused flour beetle, Tribolium confusum Jacquelin duVal, contained both 1-alkenes and heptadecadiene (Keville and Kannowski, 1975). Tridecene constitutes 90 per cent of the defensive secretion from the prothoracic glands of the lacewing, Chrysopa oculata Say (Blum et al., 1973). The major hydrocarbon component of the surface lipids of the American cockroach, Periplaneta americana L., is cis,cis-6,9-heptacosadiene(Baker er al., 1963; Beatty and Gilby, 1969), whereas all other cockroaches studied have methylalkanes as the major components. This diene is changed by ultraviolet light and oxygen into conjugated unsaturated and oxygenated compounds (Beatty and Gilby, 1969), and antioxidants [polyhydric phenols such as 3,4dihydroxybenzoic acid (protocatechuic acid), which are also involved in tanning] present on the cuticle prevent degradation and the subsequent polymerization (Atkinson and Gilby, 1970; Atkinson et al., 1973). Also, ultraviolet light increases the hydrocarbon content of the cuticular wax (Gingrich, 1975). Alkenes, alkadienes, and alkatrienes and their methyl-branched isomers make up about 90 per cent of the hydrocarbon fraction of the millipede, Graphidostreptus tumuliporus (Karsch) (Oudejans, 1973). n-Alkenes, 2,(m 1)-, 2,(m2)-, and 3,(~-2)-dimethylalkenes,and 2- and 3-methylalkenes have been identified in bacteria (Albro and Dittmer, 1969a; Tornabene and Markey, 1971), and polyolefins have been reported from algae (Youngblood and Blumer, 1973) and mosses (Karunen, 1974). 2.2
CYC L O AL K ANES
Cycloalkanes were reported in M. domestica (Louloudes et al., 1962), wool wax (Mold et al., 1964) and tobacco (Enzell et al., 1969); 1-cyclohexylakanes were found in Nonesuch seep oil (Johns et al., 1966), and 1-cyclopentyl-and 1cyclohexylalkanes, 7- and 9-~yclohexylalkanes,dicyclohexylalkanes and diphenylalkanes were found in paraffin wax (Levy et al., 1961). In G. tumuliporus, cyclopropane alk-1-enes were found only in the female (Oudejans, 1973).
2.3
2-
AND
3-METHYLALKANES
Both 2- and 3-methylalkanes (is0 and anteisoalkanes) have been found in meteorites (Or6 et al., 1968) [no alkanes were found in lunar samples
4
DENNIS R . NELSON
(Meinschein et al., 1970) or in parts of graphitetroilite nodules of iron meteorites not exposed to the earth’s atmosphere (Or6 et al., 1968)1, in petroleum (Hills and Whitehead, 1966), in numerous plants (Eglinton and Hamilton, 1967; Weete, 1972; Kolattukudy and Walton, 1973; Kolattukudy, 1975; and references cited therein), in the land snail, Cepaea nemoralis (L.) (Van der Horst and Oudejans, 1972), and in the millipede, G. tumuliporus (Oudejans, 1972). 2-Methylalkanes were reported from the common house cricket, A. domesticus (Hutchins and Martin, 1968; Blomquist et al., 1976), the female tiger moth, Holomelina opella nigricans (Reakirt) (Roelofs and Carde, 1971), the silkworm, Bombyx mori L. (Shikata et al., 1974), and the crickets, Allonemobius fasciatus (De Geer) and Gryllus pennsylvanicus Burmeister (Blomquist el al., 1976). 3-Methylalkanes were reported from the surface lipids of the big stonefly, P. californica (Armold et al., 1969), the cockroaches, P . australasiae, P. brunnea, and P . fuliginosa (Jackson, 1970), P . japonica and P. americana (Jackson, 1972), L. maderae and B. orientalis (Tartivita and Jackson, 1970), the Mormon cricket, Anabrus simplex Haldeman (Jackson and Blomquist, 1976), the fleshfly, S. bullata (Jackson et al., 1974), and the fire ants, Solenopsis invicta Buren and S. richteri Fore1 (Lok et al., 1975). 3-Ethylhexacosane was reported from the silkworm (Murata et al., 1974), however, their published mass spectrum is more compatible with that of 3-methylheptacosane when compared with spectra of 3-methyl- and 3-ethylalkanes published by the American Petroleum Institute. It should be noted that in plants (Wollrab et al., 1967), Mollusca and Arthropoda, the majority of the 2-methylalkanes has an odd number of carbon atoms, and the majority of the 3-methylalkanes has an even number of carbon atoms, which would be expected if the methyl branch is derived from the amino acids valine and isoleucine, respectively. Also, 2,(w- 1)-dimethylalkanes were reported in the waxes of the horehound, Marrubium vulgare L. (Brieskorn and Feilner, 1968), and dimethylalkenes were reported in bacteria, as noted above (Albro and Dittmer, 1969a; Tornabene and Markey, 1971).
2.4
INTERNALLY B R A N C H E D M E T H Y L A L K A N E S : A N A L Y S I S
Recent reports of the occurrence and structural identification of internally branched mono-, di-, and trimethylalkanes have depended upon the use of molecular sieves (1/16 in. pellets of Linde type 5A) to separate the branched alkanes from the n-alkanes (O’Connor et al., 1962) and the increased use of improved gas-liquid chromatographic and mass spectrometric methods of
LONG-CHAIN M ETHYL-BRANCHED HYDROCARBONS
5
analysis. Monomethylalkanes with the methyl branch located on about carbon 7 to over 18 elute from gas-liquid chromatographic columns such as SE-30, OV-17, and OV-101 with an equivalent chain length (Miwa, 1963) 0.6 to 0.7 carbon atoms less than the n-alkane with the same number of carbon atoms (Mold et al., 1966; Nelson and Sukkestad, 1970, 1975). Additional internal methyl branches have an additive effect. Thus, two internal methyl branches with isoprenoid spacing decrease the equivalent chain length about 1.4 carbon atoms less than the total number of carbon atoms in the molecule, and three methyl branches cause the equivalent chain length to be about 2.2 carbon atoms less (Nelson and Sukkestad, 1970, 1975). If the branch point is closer to the end of the chain, the effect of the branch on the equivalent chain length is less (Mold et al., 1966). However, on polar columns such as cyclohexanedimethanol succinate, iso- and anteisomethyl branches decrease the equivalent chain length 0.65 and 0.75 carbon atoms, respectively, and a centrally located double bond and a terminal double bond decrease it by 0.2 and 0.5, respectively (Albro and Dittmer, 1969a). The equivalent chain length in conjunction with the carbon number determined by mass spectrometry, gives the number of methyl branches, and the position of the methyl branches is then deduced from the mass spectral fragmentation patterns by comparing the relative intensities of significant adjacent even and odd mass peaks. Methylalkanes give relatively simple mass spectra, and some mass spectra have been analyzed by plotting the carbon number of the fragment ion vs. the intensity of the fragment ion (Mold et al., 1966; Hutchins and Martin, 1968; Nishimoto, 1974). However, on the basis of such mass spectra alone, one cannot distinguish between an isomeric mixture of internally branched monomethylalkanes and internally branch di- and trimethylalkanes or isomeric mixtures of di- and trimethylalkanes (McCarthy et al., 1968; Nelson and Sukkestad, 1970, 1975). Biemann (1962) and Hood (1963) noted that internally branched alkanes tended to fragment at the branch point to give a secondary carbonium ion of [C,Hz,+ll+. Formation of the secondary carbonium ion was also accompanied to some degree by the loss of a hydrogen atom to give another secondary carbonium ion of [C,H,,l' (i.e., a doublet appeared in the mass spectrum that corresponded to the odd-mass secondary carbonium ion and to the even-mass secondary carbonium ion, one mass unit less). Of the two competing reactions (cleavage of the carbon-carbon bond on one side or the other of the branch point) for the formation of the two possible [C,,H2,,+,I+ secondary carbonium ions, the preferred cleavage is that which results in the loss of the larger of the alkyl chains (Pomonis et al., 1978). Also, the formation of the primary (straight-chain) carbonium ion is accompanied to some degree by the loss of a hydrogen atom. However, the significance of the loss of the hydrogen atom as
DENNIS R NELSON
6
an aid to the interpretation of mass spectra was not realized until McCarthy et al. (1968) deduced the effects of the size of the straight-chain tail of the secondary carbonium ion and of the presence of other- branch points in the secondary carbonium ion on the intensity of the [C,H,,lt ion (i.e., other branches on the secondary carbonium ion suppressed the formation of the [C,,H,,lt ion). These observations were used to distinguish between the mass spectra expected for 79-dimethylhexadecane and that of a mixture of 7- and 8methylheptadecane (McCarthy et al., 1968) and were later used by Nelson and Sukkestad (1970, 1975), in conjunction with gas chromatographic retention times expressed as equivalent chain lengths, to identify for the first time internally branched di- and trimethylalkanes in insects.
2.5
MONOMETHYLALKANES
Methyl branched alkanes have been identified mainly in arthropods but also in algae, higher plants, and gastropods, in which the methyl branch is located towards the center of the molecule. Alkenes with similar methyl branching have been found in S . calcitrans (Uebel et al., 1975b). Similar methylalkenes were found in the millipede, G. tumuliporus (Oudejans, 1973). The identified monomethylalkanes and their sources are summarized in Table 1. The GLC peak number given there is equal to the number of carbons in the backbone of the molecule, and the letter A designates one internal methyl branch. The shorter chain monomethylalkanes (less than 20 carbon atoms such as 5-methylpentadecane and 7- and 8-methylheptadecanes) are present in meteorites (Or6 et al., 1968), in a number of algae (Gelpi et al., 1970), and blue-green algae (Han et al., 1968; Fehler and Light, 1970). In Hymenoptera, they were identified in the secretions of Dufour’s gland in the ants, Formica nigricans Emery, F. rufa L., and F. polyctena Foerster (Bergstrom and Lofqvist, 1973), Camponotus intrepidus (Brophy et al., 1973), Pogonomyrmex rugosus var. fuscatus Emery and P. barbatus rugosus Emery (Regnier et al., 1973) and in whole body extracts of the Argentine ant, Iridomyrmex humilis (Mayr) (Cavil1 and Houghton, 1973). The 7- and 8-methylheptadecanes were also reported in the lichen, Siphula ceratites (Wg.) Fr. though they may have been from the algal symbiont (Gaskell et al., 1973). Although n-alkanes and 2- and 3-methylalkanes have been identilied in the waxes of a large number of higher plants, the only internally branched monomethylalkanes reported were in the leaf wax of the walnut tree, Juglans regia L. (Stranski et al., 1970), and of wheat (Nishimoto, 1974). A series of methylalkanes from 17 to 34 carbon atoms was present in the walnut leaf wax. In the GLC peaks identified, GLC peak 27-A was a mixture of 7-, 9-, 11-, and 13methylheptacosane, and GLC peak 29-A was a mixture of 11-, 13-, and 15-
7
LONG-CHAIN METHYL-BRANCHED HYDROCARBONS
TABLE 1 Occurence and structure of internally-branched monomethylalkanes GLC peak no.*
Sourcea
Methyl-branched components
Major isomer"
METEORITES
Carbonaceous chondrites P L A N T S (Algae) C. turgidus C. turgidus A. cyanea L . aestuarii NOSIOC sp. C .fritschii N . muscorum P. luridum A . nidulans A . variabilis
I
PLANTS
16-A
6- and 7-methylhexadecane
17-A
7- and 8-methylheptadecane
17-A
4-, 7- and 8-methylhepadecane
17-A
7- and 8methylheptadecane
similar
17-A
7-, and 8-methylheptadecane
similar
27-A 29-A
7-, 9-, 1 I-, and 13-methylheptacosane 1 I-, 13-, and 15-methylnonacosane
13-methyl 1 1 - & 13methyl
23-A 25-A 27-A 28-A 29-A 30-A 3 I-A 32-A 33-A
1 1 -methyltricosane 1 1methylpentacosane
1 I-, and 13-methylheptacosane lo-, and 12-methyloctacosane 1 I - , and 13-methylnonacosane 10, and 12-methyltriacontane 11-, 13-, and 15-methylhentriacontane lo-, and 12-methyldotriacontane 1 1, 13-, and 15-methyltritriacontane
1 I-methyl
35-A
1 I-, 13-, 15-, and 17-methylpentatriacontane
11- & 13-
(Trees)
J. regia PLANTS
5- and 6methylpentadecane 5methylhexadecane
(Lichen)
S. ceratites PLANTS
15-A 16-A
(Wheat)
T. aeskiv um
12-methyl 1 I-methyl
12-methyl 1 I-methyl
12-methyl 11- & 13-
methyl methyl I N S E C T A (Coleoptera)
A. grandis
P. japonica
C. caryae
20-A 22-A 23-A 24-A 25-A 26-A 27-A 28-A 29-A
10-methyleicosane 1 I-methyldocosane 1 I-methyltricosane lo-, 1 I - , and 12-methyltetracosane 1 I - , and 13-methylpentacosane 4-methylhexacosane 5-methylheptacosane 4-methyloctacosane 5-, 11- 13-, and 15-methylnonacosane
12-methyl 1 I-methyl
?
8
DENNIS R. NELSON
TABLE 1 (cont.)
GLC Sourcea
peak
Methyl-branched components
Major isomef
no.b INSECTA (Diptera)
r 25-A
S.bullata (31-A
S. calcitrans
)
5-, 7-, 9-, 1 1-, and 13-methylpentacosane 5-,7-, 9-, 11-, and 13-methylheptacosane 5-, 7-, 9-, 11-, 13-, and 15-methylnonacosane 5-, 7-, 9-, 11-, 13-, and 15-methylhentriacontane 11-, 13-, and 15-methylhentriacontane 13-, and 15-methyltritriacontane 13-, 1 5 , and 17-methylpentatriacontane 13-, and 15-methylheptatriacontane
INSECTA (Hymenoptera) F. nigricans
F.mfa F. polyctena C. intrepidus P. rugosus F. nigricans C . intrepidus P. rugosus P. barbatus F. nigricans F. rufa F. polyctena C. intrepidus P. barbatus P. rugosus P. rugosus P. barbatus F. nigricans C . intrepidus I . humilis I . humilis C. intrepidus I. humilis
1
1
1
}
S.richteri A . mellvera (Beeswax)
5-methylundecane
11-A 12-A 12-A
5-, and 6-methylundecane 4-methyldodecane 5-methyldodecane
12-A
6-methyldodecane
13-A
5-methyltridecane
13-A
5-, and 6-methyltridecane
14-A
6methyltetradecane
15-A
5-methy lpentadecane
16-A 16-A 17-A 23-A 25-A
4-methylhexadecane 5-methylhexadecane 5-methylheptadecane 9-, and 11-methyltricosane 11-, and 13-methylpentacosane 9-, and 11-methylpkntacosane 5-, 7-, 9-, 11-, and 13-methvlhe~tacosane - . 11-, and 13-methylheptacosane 1 1-, and 13-methylheptacosane 9-, 11-, 13-, and 15-methylheptacosane 11-, 13-, and 15-methylnonacosane 9-, 11-, 13-, and 15-methylnonacosane 9-, 11-, 13-, and 15-methylhentriacontane
{ { 25-A 27-A
S.richteri S.invicta M . gulosa A. mell~era M . gulosa
11-A
{
27-A 27-A 27-A 29-A 29-A 31-A
?
11-methyl 1I-methyl 11-methyl 13-methyl 13-methyl 13-methyl ? 15-methyl ? ?
LONG-CHAIN METHYL-BRANCHED HYDROCARBONS
9
TABLE 1 (cont.) I N S E C TA
(Lepidoptera)
H . zea
31-A
7-, 9-, 11-, 13-, and 15-methylhentriacontane 13-methyl ? 9-, 11-, 12-, 13-, and 15-methylhentriacontane lo-, 12-, 13-, 14-, 15-, and 16-methyldotriacontane ? 9-, 11-, 13-, 15-, and 17-methyltritriacontane ? 15-, and 17-methylpentatriacontane 17-methyl 13-, 15-, 17-, and 19methylheptatriacontane 15-methyl 13-, 15-, 17-, and 19-methylnonatriacontane 15-methyl
23-A
26-A
11methyltricosane 12-methyltetracosane 13-methylpentacosane 13-methylhexacosane
27-A
11-, and 13-methylheptacosane
similar
27-A 27-A 27-A 27-A 29-A 29-A 29-A 29-A 30-A 30-A 30-A 3 1-A 3 1-A 3 1-A 32-A 32-A 33-A 33-A 33-A 33-A 33-A
5-, 7-, 9-, 11-, 13-, and 15-methylheptacosane 5-, 7-, 9-, 11-, and 13-methylheptacosane 13-methylheptacosane 5-, 7-, 11-, and 13-methylheptacosane 5-, 7-, 9-, 11-, and 13-methylnonacosane 7-, 9-, 11-, 13-, and 15-methylnonacosane 13-methylnonacosane 5-, 7-, 9-, 1 1-, 13-, and 15-methylnonacosane 8-, 9-, lo-, and 11-methyltriacontane 8-, 9-, lo-, 11-, 12-, and 13-methyltriacontane 13-methyltriacontane 7-, 9-, 11-, 13-, and 15-methylhentriacontane 5-, 7-, 9-, 1 1-, 13-, and 15-methylhentriacontane 5-, 7-, 9-, 11-, 13-, and 15-methylhentriacontane 9-, lo-, 11-, and 12-methyldotriacontane 9-, lo-, 11-, 12-, and 13-methyldotriacontane 11-, and 13methyltritriacontane 11-, 13-, and 15-methyltritriacontane 7-, 9-, 11-, and 13-methyltritriacontane 13-, 15-, and 17methyltritriacontane 11-, 13-, and 15-methyltritriacontane
11-methyl 13-methyl
34-A 34-A 35-A 35-A 35-A 36-A 36-A 37-A 37-A
13-, 15-, and 17-methyltetratriacontane 13methyltetratriacontane 11-, 13-, and 15-methylpentatriacontane 13-, 15-, and 17-methylpentatriacontane 11-, 13-, 15-, and 17-methylpentatriacontane 13-, 15-, and 17-methylhexatriacontane 14-methylhexatriacontane 11-, 13-, 15-, and 17-methylheptatriacontane 13-, 15-, 17-, and 19methylheptatriacontane 12-, and 13-methyloctatriacontane 13-, 15-, 17-, and 19methylnonatriacontane 13-methylhentetracontane
(3i;
H. virescens
M . sexta I N S E C TA
(Orthoptera)
P. australasiae P. brunnea P.$uliginosa L. maderae B. orientalis M . sanguinipes M. packardii P. japonica A . simplex M.sanguinipes M.packardii P. japonica A . simplex M . sanguinipes M . packardii P. japonica M . sanguinipes M . packardii A . simplex M . sanguinipes M.packardii M . sanguinipes M . packardii A . simplex A . domesticus S. vaga
A . domesticus S. vaga A . simplex A . domesticus S . vaga A . domesticus S. vaga A . simplex
S. vaga
1
11-methyl 11-methyl 9-methyl 5-methyl 9-methyl 9-methyl 11-methyl 11-methyl 7-methyl 11-methyl 11-methyl 11-methyl 11-methyl 9-methyl ?
13- & 15methyl ? 11methyl ? 13-methyl ? 13-methyl 13-methyl Similar 13-methyl
10
DENNIS R NELSON
TABLE 1 (cont.)
Source"
GLC peak no.b
Methyl-branched components
Major isomer'
I N S E C T A (Tricoptera)
27-A 29-A
9-methylheneicosane 8-, lo-, snd 12-methyltricosane 9-, lo-, and 12-methylpentacosane 9-, 1 1-, and 13-methylheptacosane 7-, 9-, lo-, 12-, and 14-methylnonacosane
29-A 31-A 33-A 35-A 36-A 37-A 38-A 39-A 40-A 43-A
13-methylnonacosane I I - , 13-, and 15-methylhentriacontane 1 I - , and 13-methyltritriacontane 13methylpentatriacontane 12-, and 14-methylhexatriacontane 1 I-, 13-, 15-, and 17-methylheptatriacontane 12-methyloctatriacontane 1 I -,13-, 15-, and 17-methylnonatriacontane 12-, and 14-methyltetracontane 13-methyltritetracontane
27-A
4-, and 5-methylheptacosane 4-, and 5-methyloctacosane 4-, and 5-methylnonacosane 4.. and 5-methyltriacontane 4-, and 5-methylhentriacontane
P. calfornica
? ? 1 1
CHORDATA
Wool wax
13-methyl Similar Similar 13-methyl 13-methyl 12-methyl
PETROLEUM
Paraffin wax 31-A
4-methyl Similar 4-methyl 4-methyl Similar
a Meteorites: Or0 et a/., 1968, C . turgidus, A. cyanea, L. aestuarii, Nostoc sp: Gelpi el a/., 1970; N . muscorum, A. nidulans, P. luridum, C. fritschii: Han el al., 1968; A . uariabilis: Fehler and Light, 1970; S. ceratiles: Gaskell et a/., 1973; J . regia: Stransky et al., 1970; T. aestivum: Nishimoto, 1974; A . grandis: Hedin et al., 1972; P . japonica: Bennett et al., 1972 and Nelson, D. R. (unpublished); C. caryae: Mody et al., 1975; S . bullata: Jackson et al., 1974; S . calcitrans: Uebel el al., 1975b; F. nigricans, F. rufa, F. polyctena: Bergstrom and Lijfqvist, 1973; C. intrepidus: Brophy el al., 1973; P. rugosus and P . barbatus: Regnier et al., 1973; I . humilis: Cavill and Houghton, 1973; S. invicta and S . richteri: Lok et al., 1975; Beeswax: Stransky et al., 1966, Streibl et al., 1966; M . gulosa: Cavill et al., 1970; H . zea: Jones et a/., 1971; H . virescens: Vinson el al., 1975; M . sexta: Nelson and Sukkestad, 1970, Nelson el a/., 1972; P . australasiae, P. brunnea, and P . fuliginosa: Jackson, 1970; L. maderae and B . orientalis: Tartivita and Jackson, 1970; P. japonica: Jackson, 1972: M . sanguinipes and M . packardii: Soliday et a/., 1974; A . simplex: Jackson and Blomquist, 1976; A. domesticus: Hutchins and Martin, 1968; S. vaga: Nelson and Sukkestad, 1975; P. calfornica: Armold et al., 1969; wool wax: Mold et al., 1966; paraffin wax: Levy et a/., 1961. G L C peaks designated as described herein and in Nelson and Sukkestad, 1970, 1975. The number is equal to the number of carbons in the backbone of the molecule, and the letter A designates one internal methyl branch. The monomethylalkanes eluted with an equivalent chain length 0.6 to 0.7 carbon atoms less than the n-alkane with the same number of carbon atoms (Mold et a/., 1966; Nelson and Sukkestad, 1970; 1975). Determined from the relative intensities of the major characteristic fragmentation peaks in the mass spectra.
LONG-CHAIN METHYL-BRANCHED HYDROCARBONS
11
methylnonacosane. Other internally brdnched monomethylalkanes from 23-A to 35-A were present, including some in which the methyl branch occurred on an even numbered carbon atom (12 or 14). In wheat, the internally branched. methylalkane series was from 21 to 37 carbon atoms and consisted of even carbon numbered 11-, 13-, and 15-methylalkanes and odd carbon numbered 10- and 12-methyl alkanes. Other Hymenoptera and the Coleoptera had longer chain methylalkanes (over 20 carbon atoms). 10-Methyleicosane was the only methylalkane found in the boll weevil, A. grandis (Hedin et al., 1972). The monomethylalkanes 26A, 27-A, 28-A, and 29-A were identified in whole-body extracts of the pecan weevil, C. caryae, (Mody et al., 1975); 27-A, 29-A, and 3 1-A (each GLC peak consisted of a mixture of isomers) were identified in whole-body extracts of the bull ant, Myrmecia gulosa (F.), (Cavil1 et al., 1970), and the monomethylalkanes 25-A, 27-A, and 29-A were identified in beeswax (Strhsky et al., 1966; Streibel et al., 1966); 23-A, 25-A, and 27-A were identified in S . richteri, and 27-A was identified in S . inuicta (Lok et al., 1975) (Table 1). In Diptera, internally branched monomethylalkanes and monomethylalkenes have been reported from the stable fly, Stomoxys calcitrans, (Uebel et al., 1975b), which contained methylalkanes from 32 to 38 carbons in chain length and from the fleshfly, Sarcophaga bullata, (Jackson et al., 1974), which contained methylalkanes from 26 to 32 carbons in chain length (no methylalkanes with an even-numbered carbon backbone were found) (Table 1). One of the major alkane components of the surface lipids of the female tsetse fly, Glossina morsitans Westwood, is 2-methyltriacontane (personal communication, D. A. Carlson, USDA, Insects Affecting Man and Animals Laboratory, Gainesville, Fla.). Among the Orthoptera, the cockroaches, Periplaneta australasiae, P. brunnea, P. fuliginosa (Jackson, 1970), P. japonica (Jackson, 1972), Leucophaea maderae, and Blatta orientalis (Tartivita and Jackson, 1970), have the smallest internally branched monomethylalkanes (between 20 and 3 1 carbon atoms), followed by the grasshoppers, Melanoplus sanguinipes (F.), and M . packardii Scudder (Soliday et al., 1974) (between 28 and 38 carbon atoms). The common house cricket, Acheta domesticus (Hutchins and Martin, 1968), has monomethylalkanes, from 27 to 39 carbon atoms, and the Mormon cricket, A. simplex has monomethylalkanes from 28 to 38 carbon atoms with all the branch points on odd-numbered carbons (Jackson and Blomquist, 1976). The longest chain monomethylalkanes (33 to about 50 carbon atoms) from an orthopteran insect were in the grasshopper, Schistocerca uaga (Scudder) (Nelson and Sukkestad, 1975). Internally branched monomethylalkanes were identified in four Lepidoptera: the tobacco hornworm, Manduca sexta (L.) (Nelson and Sukkestad, 1970; Nelson et al., 1971, 1972), the corn earworm Heliothis zea (Boddie) (Jones et a[., 1971), the tobacco budworm, H. virescens (F.) (Vinson et al., 1975), and
12
DENNIS R. NELSON
the silkworm, B. mori (Murata et al., 1974). M. sexta had monomethylalkanes ranging in chain length from about 20 to 44 carbon atoms, and the major components were GLC peaks 35-A, 37-A, and 39-A. One GLC peak, 3 1-A, of three GLC hydrocarbon peaks of H. zea was identified as a mixture of 7-, 9-, 11-, 13-, and 15-methylhentriacontanes, and the 13-methyl isomer was shown to be a kairomone for the H. zea larval parasite, Microplitis croceipes (Cresson). Murata et al. (1974) reported finding 9-methyltriacontane in B. mori, but their mass spectra leave doubt as to this identification. However, mass spectra that they deduced as coming from 11,12-dimethyloctacosane was completely compatible with the mass spectra expected for a mixture of 11-, 13-, and 15methylnonacosane. The GLC retention time also appeared to be compatible with that expected for a monomethylalkane chromatographed on OV- 1. Similar homologous series of monomethylalkanes were present in both M. sexta and S . vaga. A comparison of GLC peaks 35-A, 37-A, and 39-A showed that the peaks from both insects contained the same mixture of isomers, but the major component of each peak from S. vaga had the methyl branch on carbon 13, and the major component of each peak from M. sexta had the methyl branch on carbon 17 for peak 35-A and on carbon 15 for peaks 37-A and 39A. Whether this difference is of any significance is not known at present. The majority of the internally branched monomethylalkanes has the methyl branch located on an odd-numbered carbon atom, and in plants and insects, this is usually either on carbon 11, 13, or 15. Monomethylalkanes with the methyl branch on an even-numbered carbon atom have been reported in only six insects: the boll weevil, Anthonomus grandis Boheman, with 10methyleicosane (Hedin et al., 1972), the grasshopper, Melanoplus sanguinipes, with GLC peak 30-A a mixture of 8-, 9-, lo-, and 11-methyltriacontanes and GLC peak 32-A a mixture of 9-, lo-, 11-, and 12-methyldotriacontanes (Soliday et al., 1974), the grasshopper, M. packardii, with GLC peak 30-A a mixture of 8-, 9-, lo-, 11-, 12-, and 13-methyltriacontanes and GLC peak 32-A a mixture of 9-, lo-, 11-, 12-, and 13-methyldotricontanes (Soliday et al., 1974), the stonefly, Pteronarcys californica, with GLC peak 23-A being a mixture of 8-, lo-, and 12-methyltricosanes, GLC peak 25-A a mixture of 9-, lo-, and l2-methylpentacosanes, and GLC peak 29-A a mixture of 7-, 9-, lo-, 12-, and 14-methylnonacosanes (Armold et al., 1969a), Popillia japonica, with 12-methyltetracosane (Bennett et al., 1972), and Heliothis virescens, with possibly 9-, 11-, 12-, 13-, and 15-methylhentriacontanesand lo-, 12-, 13-, 14-, 1 5 , and 16-methyldotriacontanes (Vinson et al., 1975). The only report of internally branched monomethylalkanes from a chordate was the finding in wool wax of methylalkanes from 17 to 44 carbon atoms (Mold et al., 1966) (Table 1). The methyl branch occurred mainly at the 13 position for the even-carbon numbered series.
LONG-CHAIN METHYL-BRANCHED HYDROCARBONS
13
Paraffin wax from petroleum contained 4- and 5-methylalkanes (Levy et al., 1961). In meteorites (carbonaceous chondrites) and algae, the internally branched monomethylalkanes occurred as single GLC peaks (in plants and insects, they occurred as homologous series), which were a mixture of two methylalkanes, one with the branch point on an odd-carbon atom and the other with the branch point on an evencarbon atom (Table 1). 2.6
DIMETHYLALKANES
Two homologous series of methylalkanes in addition to the n-alkanes and monomethylalkanes were reported in the tobacco hornworm, Munduca sexta (Nelson and Sukkestad, 1970; Nelson et al., 1971; Nelson et al., 1972) and in the grasshopper, Schistocercu uaga (Nelson and Sukkestad, 1975) in which two (B series) or three (C series) methyl branches, respectively, were located toward the center of the molecule. In both M. sextu and S. vaga, the major methylalkanes in the A-series were 35-A; in the B-series, 35-B; and in the Cseries, 35-C. The homologous B-series (dimethylalkanes) ranged from about 35 to 55 carbons in S . uagu and from about 21 to 47 carbons in M. sextu. As previously noted, for the monomethylalkanes, M. sexta had the methyl branching located farther down the chain than did S. vaga, and this was also true for the dimethylalkanes. The major component of 33-B in S . ougu was 13,17-dimethyltritriacontane,but in M. sexta, the only two isomers present (13,17- and 15,19-dimethyltritriacontane)were present in about equal amounts (Table 2). Likewise, the major component of 35-B in S. uugu was 13,17dimethylpentatriacontane and in M. sexta, it was 15,19-dimethylpentatriacontane; the major components of 37-B were 13,17-, 15,19-, and 17,21dimethylheptatriacontane in S. ouga and 15,19-dimethylheptatriacontane in M . sextu; the major component of 39-B was 13,17-dimethylnonatriacontane in S. vaga, but in M . sextu the three isomers 13,17-, 15,19-, and 17,21dimethylnonatriacontane were present in about equal amounts. Similar homologous series have recently been reported to be present in the stable fly, Stomoxys culcitrans (Uebel et al., 1975b), in the Mormon cricket, A. simplex (Jackson and Blomquist, 1976), and in the grasshoppers, Melanoplus sunguinipes and M. puckardii (Soliday et ul., 1974). The major isomers in M. sunguinipes were the 11,15-dimethyl isomers, but in M. packardii, the 13,17dimethyl isomer was the major isomer in three of five GLC fractions identified. The female tsetse fly, G. morsituns, was reported to have a homologous series of dimethylalkanes, one of which was a mixture of 15,19- and 17,21-dimethylheptatriacontane (personal communication, D. A. Carlson). Bennett et al. (1972) established that the dimethylalkanes 9,13dimethyltricosane and 11,15-dimethylpentacosane were present in the
TABLE 2
a
P
Occurrence and structure of internally branched dimethylalkanes
Sourcea
GLC peak no!
Methyl-branched components
Major isomef
INSECTA (Coleoptera)
P. japonica
23-B
9,13-dimethyltricosane 11,15-dimethyIpentacosane
31-B
11,15- and 13,17-dimethylhentriacontane 11,15-, 13,17- and 15,19dimethyltritriacontane 11,15-, 13,17- and 15,19dimethylpentatriacontane 11,15-, 13,17- and 15,19-dimethylheptatriacontane 15,19- and 17,21dimethylheptatriacontane
{ 25-B
INSECTA (Diptera)
S . calcitrans
G. morsitans INSECTA (Hymenoptera) P. rugosus P. barbatus I N s EC TA (Lepidoptera)
37-B
3,5-dimethyldodecane
1
M.sexta
{ 3,4-dimethyltridecane 27-B 33-B 35-B 37-B 39-B 41-B
9,13-dimethylheptacosane 13,17- and 15,19-dimethyltritriacontane 13,17- and 15,19-dimethylpentatriacontane 13,17-, 15,19- and 17-21dimethylheptatriacontane 13,17-, 15,19- and 17,21dimethylnonatriacontane 13,17- and 15,19-dimethylhentetracontane
Similar 15,19-methyl 15J9-methyl Similar Similar
33-B
9,13-, 11-15-, 13,17-, and 15,19dimethyltritriacontane
llJ5-methyl
33-B 33-B 34-B 35-B
11,15-, and 13,17-dimethyltritriacontane 9,13-, llJ5-, and 13,17-dimethyltritriacontane 12,16-, 13,17-, and 14,18dimethyltetratriacontane 11,15-, 13,17-, and 15,19-dimethylpentatriacontane
13,17-methyl 13~7-methyI SimiiU 11,15-methyl
INSECTA (Orthoptera)
M . packardii M . sanguinipes A . simplex S . vaga S. vaga M . packardii
1
zz z
v)
P Z rn r v)
0
z
*
6 Z
c
? 0 M . sanguinipes A . simplex S . vaga S . vaga M . packardii M . sanguinipes A . simplex S . vaga .S. vaga M . packardii S . vaga M . packardii M . sanguinipes
1
~
_
37-B 37-B 38-B 39-B 39-B 41-B 41-B 41-B 43-B 45-B 47-B 49-B 51-B _
~
e
z
5
14,18-dimethylhexatriacontane
37-B
S . vaga
I
9,13-, 11,15-, 13,17-, and 15,19-dimethylpentatriacontane 11,15-, 13,17-, and 15,19-dimethylpentatriacontane 9.13-, 1 1,15-, 13,17-, and 15,19-dimethylpentatnacontane
35-B 35-B 35-B 36-B
11,15-, 13,17-, 15,19-, and 17,21dimethylheptatriacontane
13,17-methyl
11,15-, 13,17-, 15,19-, and 17,2 1-dimethylheptatriacontane
15,19-methyl 13,17- & 15,19-methyl Similar 13,17-methyl 13,17-methyl 13,17-methyl 11,15-methyl 13,17-methyl 13,17-methyl
11,15-, 13,17-, 15,19-, and 17,21dimethylheptatriacontane 1 I,l5-, and 14,18-dimethyloctatriacontane 1 1,15-, 13,17-, 15,19-, and 17,21-dimethylnonatriacontane 13,17-, 15,19-, and 17,21-dimethylnonatriacontane 11,15-, 13,17-, 15,19-, 17,21-, and 19,23-dimethylhentetacontane 11,15-, 13,17-, 15,19-, 17,21-, and 19,23-dimethylhentetracontane 13,17-, and 19,23-dimethylhentetracontane 13,17-, and 19,23-dimethyltritetracontane 13,17-dimethylpentatetracontane 13,17-dimethylheptatetracontane 13,17-dimethylnonatetracontane 13,17-dimethylhenpentacontane -
-I
I
_ppp._____p-p
“P.japonica: Bennett el al., 1972; S.calcirrans: Uebel er al., 1975b; G. morsitans: Carlson (personal communication); P. rugosus and P. barbarus: Regnier el al., 1973; M . sexfa: Nelson and Sukkestad, 1970, Nelson er al., 1972; M. packardii and M . sanguinipes: Soliday et al., 1974; S . vaga: Nelson and Sukkestad. 1975; and A . simplex: Jackson and Blomquist, 1976. bGLC peaks designated as described herein and in Nelson and Sukkestad, 1970, 1975. The number is equal to the number of carbons in the backbone of the molecule, and the letter A designates one internal methyl branch. The dimethylalkanes with isoprenoid spacing eluted with an equivalent chain length about 1.4 carbon atoms less than the n-alkane with the same number of carbon atoms (Nelson and Sukkestad, 1970, 1975). Determined from the relative intensities of the major characteristic fragmentation peaks in the mass spectra.
< 7 W a D
Z 0 I
rn 0
I
<
0
a
0
n
% m
0 Z v)
16
DENNIS R NELSON
Japanese beetle, P. japonica. Thus, to date the eight insects: M. sexta, S. vaga, M . sanguinipes, M . packardii, A . simplex, P . japonica, G. morsitans, and S. calcitrans, are the only known sources of the long-chain internally branched dimethylalkanes with isoprenoid spacing of the branch points. Bennett et al. (1972) also reported the presence of dimethylalkanes in which the methyl branches did not have isoprenoid spacing: 9,l l-dimethyldocosane and 9 , l l dimethyltetracosane. However, no mass spectral data were published, and we have been unable to confirm these structures. Lok et al. (1975) reported the presence of 10,12- and 12,14-dimethylalkanes in the fire ants, S . invicta and S. richteri. However, their published mass spectrum does not appear to support the structure when compared with spectrum 2 114, published by the American Petroleum Institute, as fragment ions are expected at 154 and 182 rather than at 155 and 183, respectively, and at 197 and 225 rather than at 196 and 224, respectively. Also, 11,12-dimethyloctacosane was reported in B. mori (Murata et al., 1974). However, as noted earlier, the mass spectra were more compatible with that expected for a mixture of monomethylalkanes. Two short-chain dimethylalkanes (3,5-dimethyldodecane and 3,4-dimethyltridecane) were tentatively identified in the Dufour’s gland of the harvester ant, Pogonomyrmex rugosus (Regnier et al., 1973), and homologous series of 3,9-dimethylalkanes were reported in leaf wax (Brieskorn and Beck, 1970; Brieskorn and Feilner, 1968). A homologous series of 2,6dimethylalkanes was found in crude oils and bituminous shales (Gohring et al., 1967). 2.7
TRIMETHYLALKANES
Another series of hydrocarbons, the internally branched trimethylalkanes in which the methyl branches have isoprenoid spacing, have been reported in only three insects: M. sexta (Nelson and Sukkestad, 1970; Nelson et al., 1972), S . vaga (Nelson and Sukkestad, 1975), and G. morsitans (personal communication, D. A. Carlson). The major trimethylalkane in M. sexta was 13,17,21trimethylpentatriacontane, and in S. vaga, it was a mixture of 11,15,19- and 13,17,21-trimethylpentatriacontanes.The major alkane in female G. morsitans was 15,19,23-trimethylheptatricontane,and a lesser amount of 15,19,23-trimethyloctatriacontane was also present. Trimethylalkanes with isoprenoid spacing but with the first branch on carbon atom 3 have been found in Atta cephalotes isthmicola Weber and with the first branch on carbon atom 3 or 4 in the ants, A . colombica Guerin and A. sexdens (L.) (MacConnell, 1969; Martin and MacConnell, 1970). A short-chain trimethylalkane, 2,6,10-trimethyltetradecane, was found in the Antrim shale (McCarthy and Calvin, 1967). Multiple methyl-branched alkanes such as pristane (2,6,10,14-tetramethylpentadecane)
LONG-CHAl N M ETHY L-B RANCHED HYDROCARBONS
17
and phytane (2,6,10,14-tetramethylhexadecane)have been identified in human, rat, shark, and bovine tissues (Avigan et al., 1967), butterfat (Urbach and Stark, 1975), wool wax (Mold et al., 1963), crustacean surface waxes (Hamilton et al., 1975), fish (Ackman, 197 l), Precambrian sediment (Eglinton et al., 1966), Costa Rican seep oil (Haug and Curry, 1974), and oil shale (Gibert et al., 1975). The isoprenoid alkanes found in geological sources are believed to be formed from phytol, a component of chlorophyl (Avigan and Blumer, 1968). It is likely that in the future, many more branched alkanes similar to the internally branched mono-, di-, and trimethylalkanes will be identified. For example, unidentified internally branched monomethylalkanes were found in Rosmarinus oficinalis (Brieskorn and Beck, 1970). Unidentified branched alkanes (other than 2- and 3-methylalkanes and the isoprenoid paraffins squalane, pristane, and phytane) were reported in bovine brain (Nicholas and Bombaugh, 1965) and liver (Nagy et al., 1969), the millipede, Graphidostreptus tumuliporus (Oudejans, 1972, 1973), the banded wood snail, Cepaea nemoralis (L.) (Van der Horst and Oudejans, 1972), the pea aphid, Acyrthosiphon pisum (Harris) (Stransky et al., 1973), plant waxes (Jarolimek et al., 1964; Wollrab et al., 1965a,b; Wollrab, 1968; Strimsky and Streibl, 1969; Streibl el al., 1974), olive oil (Eisner et al., 1965), bitumens (KovaEev et al., 1972), shale (Gallegos, 197 l), and the alga, Scenedesmus quadricauda (Turpin) Brkbisson (Stransky et al., 1968).
3
Biosynthesis
The biosynthesis of alkanes has been extensively investigated in bacteria and plants. The classical pathway considered for n-alkane biosynthesis was the head-to-head condensation of fatty acids followed by decarboxylation and reduction (reviewed by Kolattukudy, 1968, 1975). [Caldicott and Eglinton (1973) refer to this pathway as a tail-to-tail condensation.] In Corynebacterium diphtheriae (Fluege) Lehmann and Neumann, corynomycolic acid (15carboxyt 16-hydroxyhentriacontane)was synthesized by head-to-head condensation of two molecules of palmitic acid (Gastambide-Odier and Lederer, 1959), and this mechanism was also suggested by the finding of olefins with is0 and/or anteiso branches on both ends in Sarcina lutea (Schroeter) (Albro and Dittmer, 1969a). However, Kolattukudy showed that in plants, the biosynthesis of long-chain dkanes occurred by elongation of fatty acids, followed by decarboxylation (Kolattukudy and Walton, 1973; Kolattukudy, 1975). A similar pathway is believed to be operative in insects. The finding of internally branched methylalkanes has led to increased ,
18
DENNIS R. NELSON
interest in the biosynthesis of alkanes and in the origin of the methyl groups. Lederer (1964, 1969) reviewed four mechanisms that can lead to the synthesis of branched fatty acids and/or alkanes: (1) incorporation of branched chains derived from valine, leucine, or isoleucine; (2) incorporation of propionic acid (as methylmalonic acid); (3) C-methylation with the S-methyl of methionine; and (4) incorporation of mevalonic acid. The amino acids, leucine, isoleucine, and valine, may serve as precursors of 2- and 3-methyl-branched compounds (Fig. 1). The majority of 2-methylalkanes has an odd number of carbon atoms, and the majority of 3-methylalkanes has an even number of carbon atoms,
-
Leucine
-
0
-NH,
y
It
3
40,
CH,-CH-CH,-C-COOH
+ “acetate”
a-ketoisocaproic acid
Odd carbon fatty acid
-NH,
Isoleucine
-
CH, 0 I II CH,-CH,-CH-C-COOH
-co,
-co,
+ “acetate”
a-keto-gmethylvaleric acid
-
Odd-carbon -co, fatty acid
Valine
-Nn,
CH, 0
I
II
CH,-CH-C-COOH
-co,
+ “acetate”
Even carbon isoalkane
Even-carbon fatty acid
-co,
Even-carbon anteisoalkane Odd-carbon isoalkane
a-ketoisovaleric acid Fig. 1.
indicating that they might be synthesized from valine and isoleucine, respectively. Valine and isoleucine, respectively, have been shown to be incorporated into the 2- and 3-methyl fatty acids and alkanes of Sarcina lutea (Albro and Dittmer, 1969b). Valine, after elimination of the carboxyl groups, was incorporated into the 2-methyl fatty acids of Bacillus subtilis Cohn (Kaneda, 1963) and 3-methyl fatty acids were formed from a-keto-p-methylvalerate or from isoleucine (Kaneda, 1966). In tobacco plants, valine, leucine, and isoleucine were incorporated into 2- and 3-methylalkanes although the carboxyl carbon was not incorporated (Kaneda, 1967). The crickets, Allonemobius fasciatus and Glyllus pennsylvanicus, appear to synthesize 2methylalkanes by an elongation-decarboxylation pathway (Blailock et al., 1976). Valine and isobutyric acid were incorporated by A.fasciatus and valine,
LONG-CHAIN METHYL-BRANCHED HYDROCARBONS
19
leucine and isovaleric acid by G. pennsylvanicus into 2-methylalkanes at a greater rate than into n-alkanes. Propionic acid as a source of methyl branching was reported for the biosynthesis of mycocerosic acid (2,4,6,8-tetramethyloctacosanoicacid) and related acids by tuberculosis bacillus (Gastambide-Odier et al., 1963). The biosynthesis was proposed to occur by the sequential addition of four molecules of propionic acid to eicosanoic acid. Conrad and Jackson (1971) and Blomquist et al. (1975) studied the synthesis of 3-methylpentacosane in the American cockroach and found that neither isoleucine nor the S-methyl of methionine were incorporated into the 3-methylalkane but that propionate was incorporated into the alkanes of tanned adults but not of teneral adults. They proposed that 3-methylpentacosane was synthesized via fatty acid elongation and that propionate (as methylmalonate) was incorporated at the penultimate step, followed by the incorporation of one acetate and then reductive decarboxylation to give the methylalkane. Blomquist and Kearny (1976) showed that propionate and methylmalonate were almost exclusively incorporated into 3-methyltricosane and 13-methylpentacosane by Periplaneta fuliginosa, but acetate was incorporated into all alkanes in proportion to their percentage of composition, and there was no incorporation of the S-methyl of methionine. Thus, propionate may be the precursor for both 3-methylalkanes and for the internally branched methylalkanes. Recently, Buckner and Kolattukudy (1 975) demonstrated the incorporation of methylmalonyl-CoA into multi-branched fatty acids such as 2,4,6,8-tetramethyldecanoic acid by cell-free preparations from the uropygial gland of the goose. In addition to branched amino acids and propionate serving as the source of the methyl branches in fatty acids and alkanes, other mechanisms exist for the formation of methyl branching that involve the S-methyl of methionine. Tuberculostearic acid (10-methylstearic acid) is synthesized by Mycobacterium smegmatis (Trevisan) Lehmann and Neumann from oleic acid and methionine (Jaureguiberry et al., 1965). It was postulated that 9,lO-methyleneoctadecanoic acid might be involved as an intermediate because the methyl of methionine was incorporated with only two of its original hydrogen atoms. However, 9,lO-methylenestearic acid did not serve as a precursor of 10methylstearic acid (Lederer, 1964). Similarly, the C-28 methyl of ergosterol was formed by Neurospora crassa Shear et Dodge from methionine, and in the process, one of the methyl hydrogens was lost (Jaureguiberry et al., 1965). The formation of cyclopropyl compounds from methionine had been demonstrated by the synthesis of lactobacillic acid (1 1,12-methyleneoctadecanoicacid) by Lactobacillus arabinosus in which the S-methyl of methionine added across the double bond of cis-vaccenic acid, and in the process lost one of its hydrogen atoms (Liu and Hofmann, 1962). However, it was later shown that the synthesis of 10-methylstearic acid involves 10-methylenestearic acid as an
20
DENNIS R. NELSON
intermediate rather than 9,lO-methylenestearic acid (Jaureguiberry et al., 1966) and that the oleic acid precursor exists as an ester of a phospholipid (Akamatsu and Law, 1970). The involvement of phospholipids (plasmalogen) also has been proposed for the biosynthesis of dimethylalkanes by S. lutea (Albro and Dittmer, 1970). However, plasmalogens (alkyl glyceryl ethers) are not involved in alkane biosynthesis in plants (Kolattukudy and Walton, 1972). The retention of the vinyl hydrogens of oleic acid was demonstrated in the synthesis of dihydrosterculic acid (9,lO-methyleneoctadecanoic acid) from oleic acid and methionine (Polacheck et al., 1966). In G. tumuliporus, the Smethyl of methionine added across the double bond of oleic acid to form a cyclopropane fatty acid (Van Der Horst et al., 1973), which in turn can presumably be used to form cyclopropyl alk- 1-enes by an elongation-decarbopylation-a-oxidation mechanism, followed by reduction and dehydration (Oudejans and Zandee, 1973). The S-methyl of methionine can also form internal methyl branches without the loss of a hydrogen atom. Blue-green algae have a 50: 50 mixture of 7- and 8-methylheptadecane in which the methyl group is derived from methionine and the methyl group is incorporated intact (Fehler and Light, 1970, 1972). The precursor is probably cis-vaccenic acid (Han et al., 1969), and the methyl of methionine is added to the carbons of the double bond. However, it cannot add across the double bond to form a cyclopropyl intermediate or involve a methylene intermediate because these would result in the loss of one of the hydrogens from the S-methyl group. Therefore, the reaction must not be specific for carbons 7 or 8 but must involve either equally. The fourth mechanism for the incorporation of methyl groups involves mevalonic acid. Although mevalonic acid is the source of the methyl groups in such isoprenoid compounds as squalene and phytanic acid, and juvenile hormone (Schooley et al., 1973) [however, the methyl ester group of juvenile hormone is derived from methionine (Metzler et al., 1971)1, it has not been shown to be a source of methyl groups for the long-chain, internally branched methylalkanes. Insects can synthesize both their n-alkanes and branched alkanes though the majority of the n-alkanes probably originates in the diet (Blomquist and Jackson, 1973). Several sites have been proposed for the biosynthesis of surface waxes and hydrocarbons in insects. They include the hypodermis or integument, the fat body, and the oenocytes. Piek (1964) compared the incorporation of water, acetate, and glucose into lipids and proposed that wax acids and hydrocarbons were synthesized in the oenocytes but that wax esters and their component acids and alcohols were synthesized in fat cells. The surface lipids must be replaced at each molt, and Philogene and McFarlane (1967) found that lipids in vacuoles of the oenocytes of the house cricket, Acheta domesticus, were discharged during molting and that at about the same
LONG-CHAIN METHYL-BRANCHED HYDROCARBONS
21
time a lipid-staining layer appeared in the epicuticle. Nelson (1969) showed that acetate and palmitate were incorporated in vitro into hydrocarbons by the integument but not by the fat body of Periplaneta americana and Manduca sexta and concluded that synthesis was by the hypodermis rather than the oenocytes or fat body. Jackson and Baker (1970) also found that synthesis occurred in excised integument of P. americana and that palmitate was a better precursor for the synthesis of 3-methylpentacosane than were acetic, decanoic, or linoleic acids. Armold and Regnier (1975a) reported that isolated pupal integuments of the flesh fly, Sarcophaga bullata, incorporated acetate into hydrocarbons better than did the internal tissues and that synthesis in the pupal integument was stimulated by prior injections of the larva with ecdysterone. In the desert locust, Schistocerca gregaria (Forskd), acetate was incorporated into paraffins by both the hypodermis and the oenocyte-rich peripheral fat body (Diehl, 1973). However, the hypodermis incorporated only about one per cent as much acetate into paraffis as did the oenocyte-rich fat body. The more oenocytes present in the fat body, the greater was the incorporation of acetate into paraffins, and the oenocytes could be stimulated to release their paraffins into the incubation medium by the addition of hemolymph to the medium (Diehl, 1975).
4
Functions
The majority of the known chemicals involved in insect communication are fatty acid esters (Calam, 1971) and acids, esters, ketones, and alcohols (Evans and Green, 1973). However, female Fannia canicularis and Musca domestica produced sex attractants (Z)-9-pentacosene (Uebel et a[., 1975a) and (27-9tricosene (Carlson et al., 1971), respectively. In the stable fly, Stomoxys calcitrans, gas-liquid chromatography was used to obtain several active fractions that contained 11-, 13-, and 15-methylhentriacontanes and 11,15and 13,17-dimethylhentriacontanes, 13- and 15-methyltritriacontanes and 11,15-, 13,17-, and 15,19-dimethyltritriacontanes,@)-9-hentriacontene, (27-9tritriacontene, and methyl-branched hentria- and tritriacontenes, respectively (Uebel et al., 1975b). The sex pheromone of the face fly, Musca autumnalis, was present in both sexes. However, the biological activity of the active components ( Z ) 14-nonacosene, ( Z ) -13-nonacosene, and (Z)13-heptacosene was masked in the male by a higher proportion of nonacosane and heptacosane than were present in the female (Uebel el al., 1975~).The sex pheromone of the mushroom fly, Lycoriella mali (Fitch) was a mixture of 15- to 26-carbon n-alkanes with heptadecane the most active component and from five to six times more was present in females than in males (Kostelc et al., 1975).
N N
TABLE 3 Biological activity of alkanes and alkenes Source"
Active chemical(s)
Function
COLEOPTERA
1-pentadecene 1-pentadecene, hexadecane, and 1-heptadecene
Surfactant Sex pheromones
F. canicularis M . autumnalis M . domestica L . mali S. calcitrans
(Z)-9-pentacosene (2)-13- and (Z)-14-nonacosene and (a-13-heptacosene (2)-9-tricosene n-alkanes from C- 15 to C-26 11-, 13-, and 15-methyl and 11,15- and 13,17-dimethyhentriacontane;13and 15-methyl and 1l , l 5 - , 13,17-, and 15,19-dimethyltritriacontane
S. calcitrans
(Z)-9-hentriacontene, (Z)-9-tritriacontene
S . calcitrans
methyl-branched hentria- and tritriacontenes
Sex pheromone Sex pheromone Sex pheromone Sex pheromone complex GLC fractions containing these components initiated males to copulate GLC fractions containing these components initiated males to copulate GLC fractions containing these components initiated males to copulate
T. confusum T. confusum DIPTER A
HYMENOPTERA
A . claoiger L . alienus
z Z
undecane undecane
Alarm-defense substance Alarm-defense substance
z
cn
P Z m r
cn
0 Z
h
I
eZ
LEPIDOPTERA
H. nigricans H . Zea H. zea H. virescens
2methylheptadecane tricosane 13-methylhentriacontane 1 1-methylhentriacontane, 16-methyldotriacontane, 13-methyltritriacontane, and other isomers
Sex pheromone Kairomone for T. evanescens Kairomone for M. croceipes Kairomone for C. nigriceps
undecane plus tetradecane
Aggregation pheromone
ORTHOPTERA
B. craniifer
5< I
r m
sZ 0
I
t; I
< 0
a
00
T. confusum: von Endt and Wheeler, 1971 and Keville and Kannowski, 1975. F. canicularis: Uebel et al., 1975a. M . autumnalis: Uebel ef al., 197%. M . domestica: Carlson et al., 1971. L . mali: Kostelc ef al., 1975. S . calcifrans: Uebel el al., 1975b. A . claoiger: Regnier and Wilson, 1968. L. g alienus: Regnier and Wilson, 1969. H. nigricans: Roelofs and Carde, 1971. H. zea: Jones, ef al., 1971, 1973. H. virescens: Vinson ef al., 1975. B. Z craniifer: Brossut et al., 1974. (I
cn
N
w
24
DENNIS R NELSON
Also, an unidentified sex pheromone of the tsetse fly, Glossina morsitans, was found in the hydrocarbon fraction of female cuticular lipids (Langley et al., 1975). In the gregarious cockroach, Blaberus craniifer, an aggregation pheromone was secreted by the mandibular glands of larvae (except at ecdysis) and adults (Brossut et al., 1974). The volatile components of the mandibular gland secretions were undecane, tetradecane, and ethyl-caproate. A mixture of undecane and tetradecane (1 : 1) produced all the effects of the natural pheromone and had a perception distance of 40 cm and 10 cm for the adults and first-instar larvae, respectively, and a threshold amount of 0.2 ng and 0.4 ng for the adults and first-instar larvae, respectively. 1-Pentadecene, hexadecane, and 1-heptadecene induced copulatory behaviour in males of the confused flour beetle, Tribolium confusum (Keville and Kannowski, 1975). Also, 1-pentadecene was proposed to function as a surfactant, facilitating the absorption of quinones from the defensive secretions of T. confusum by its predators (von Endt and Wheeler, 1971). In the ants, Acanthomyops claviger (Roger) (Regnier and Wilson, 1968) and Lasius alienus (Foerster) (Regnier and Wilson, 1969), undecane was the principle volatile component of Dufour’s gland and was shown to be an alarmdefense substance and a spreading agent for formic acid. The secretion of the prothoracic glands of the lacewing, Chrysopa oculata, was 90 per cent tridecene, and this material may function as a vehicle for the defensive component 3-methylindole (skatole) (Blum et al., 1973). Tricosane was shown to be a kairomone that elicited a host-seeking response (for eggs of the corn earworm, Heliothis zea) in the hymenopteran parasite, Trichogramma euanescens Westwood (Jones et al., 1973) (Table 3). Docosane, tetracosane, and pentacosane were also present in the active kairomone fraction from moth scales but were less active than tricosane. The kairomone 13-methylhentriacontane from larval H. zea and its frass attracted the larval parasite, Microplitis croceipes (Cresson) (Jones et al., 1971). The monomethylalkanes, 1 1-methylhentriacontane, 16-methyldotriacontane, and 13-methyltritriacontane from the mandibular glands of larvae of Heliothis virescens were optimally active in attracting the parasite, Cardiochiles nigriceps Viereck (Vinson et al., 1975). 2-Methylheptadecane was isolated from female tiger moths, Holomelina opella nigricans, and was found to function as a sex attractant in this insect and in eight other species of tiger moths (Roelofs and Cardk, 1971). Although of unknown significance, hydrocarbons are influenced by sex, age, photoperiod, and diapause or development. Hydrocarbons were twice as predominant in the hemolymph of female Periplaneta americana as in males (Baker et al., 1963). However, in Blatta germanica, no difference was found between males and females (Acree et al., 1965). The amount of hydrocarbon fraction from hemolymph varied with age in both P. americana and B.
LONG-CHAIN METHYL-BRANCHED HYDROCARBONS
25
germanica (Acree et al., 1965). In P. americana, the hydrocarbon content of hemolymph fluctuated in a circadian manner: The fluctuation was depressed by constant light, and the fluctuation was greatest in the male (Turner and Acree, 1967). Sublethal levels of ultraviolet radiation to tanned P. americana caused an increase in the amount of cuticular hydrocarbons (Gingrich, 1975). Manduca sexta reared in diapause-inducing photoperiods had a prolonged period of secretion of surface wax after pupation and produced three times as much as nondiapausing pupae (Bell et al., 1975). The content of cuticular hydrocarbons of S . bullata has been determined during all developmental stages (Armold and Regnier, 1975b). The greatest increase in hydrocarbons occurred during the third larval instar, following pupariation and again following the pupal-adult ecdysis. A linear rate of formation of cuticular hydrocarbons was observed for 7 0 h following pupariation. The largest increases were of heptacosane and nonacosane. The amount of plant epicuticular wax was affected by light and temperature, and the effect was not directly correlated with leaf expansion (Wilkinson and Kasperbauer, 1972; Giese, 1975). In Brussels sprouts, Brassica oleracea var. gemmifera, increased light, decreased temperature or decreased humidity resulted in the deposition of more leaf wax (Baker, 1974). Thus, although the alkanes have a number of functions, the most universal function is to provide a waterproofing layer as part of the surface lipids and prevent desiccation of the organism. Although surface lipids have been considered end products of metabolism, Cassagne and Lessire (1975) showed that aqueous dispersions of stearic acid, lignoceric acid, and octadecane applied to the leaf surface of Allium porrum L. rapidly entered the epidermal cells and were further metabolized. Over 20 per cent of administered fatty acids were found in the internal lipids in 30 min; and some of these were excreted in the surface lipids as alcohols and alkanes. About 15 per cent of administered octadecane entered the epidermal cells, and after 40min, the surface lipids contained 10 per cent of the administered octadecane as wax esters, 6 per cent as alcohols, and 4 per cent as fatty acids. Thus, the surface lipids appear to be in a dynamic equilibrium with the epidermal cells. Such a dynamic state has not yet been demonstrated for insect surface lipids. A function of the surface lipids that has not been considered is that they may serve as a high-energy food source. This food source could be used by predators and also by the next development stage of those insects that consume their exuviae. References Ackman, R. G. (1971). Pristane and other hydrocarbons in some freshwater and marine fish oils. Lipids, 6, 520-522. Acree, F. Jr, Turner, R. B., Smittle, B. J. and Burden, G. S. (1965). Hydrocarbons in haemolymph of cockroaches of different ages. J. Insect Physiol. 11,905-9 10.
26
DENNIS R. NELSON
Akamatsu, Y.and Law, J. H. (1970). Enzymatic alkylenation of phospholipid fatty acid chains in extracts of Mycobacteriumphlei. J. Biol. Chem. 245,701-708. Albro, P. W. and Dittmer, J. C. (1969a). The biochemistry of long-chain nonisoprenoid hydrocarbons. 1. Characterization of the hydrocarbons of Sarcina lutea and the isolation of possible intermediates of biosynthesis. Biochemistry, 8, 394-404. Albro, P. W. and Dittmer, J. C. (1969b). The biochemistry of long-chain, nonisoprenoid hydrocarbons. 11. The incorporation of acetate and the aliphatic chains of isoleucine and valine into fatty acids and hydrocarbons by Sarcina lutea in vivo. Biochemistry, 8,953-959. Albro, P. W. and Dittmer, J. C. (1970). Bacterial hydrocarbons: Occurrence, structure and metabolism. Lipids, 5,320-325. Armold, M. T. and Regnier, F. E. (1975a). Stimulation of hydrocarbon biosynthesis by ecdysterone in the flesh fly Sarcophaga bullata. J. Insect Physiol. 21, 1581-1586. Armold, M. T. and Regnier, F. E. (1975b). A developmental study of the cuticular hydrocarbons of Sarcophaga bullata. J. Insect Physiol. 21, 1827-1833. Armold, M. T., Blomquist, G. J. and Jackson, L. L. (1969). Cuticular lipids of insects. 111. The surface lipids of the aquatic and terrestrial life forms of the big stonefly, Preronarcys californica Newport. Comp. Biochem. Physiol. 31,685-692. Atkinson, P. W. and Gilby, A. R. (1970). Autoxidation of insect lipids: Inhibition on the cuticle of the American cockroach. Science, 168, 992. Atkinson, P. W., Brown, W. V. and Gilby, A. R. (1973). Phenolic compounds from insect cuticle: Identification of some lipid antioxidants. Insect Biochem. 3,309-3 15. Avigan, J. and Blumer, M. (1968). On the origin of pristane in marine organisms. J. Lipid Res. 9, 350-352. Avigan, J., Milne, G. W.A. and Highet, R. J. (1967). The Occurrence of pristane and phytane in man and animals. Biochim. Biophys. Acta, 144, 127-131. Baker, E. A. (1974). The influence of environment on leaf wax development in Brassica oleracea var. gemmi4era. New Phytol. 73,955-966. Baker, G. L., Vroman, H. E. and Padmore, J. (1963). Hydrocarbons of the American cockroach. Biochem. Biophys. Res. Commun. 13,360-365. Beament, J. W. L. (1964). The active transport and passive movement of water in insects. Adv. Insect Physiol. 2,67-129. Beament, J. W. L. (1964). The active transport and passive movement of water in insects. In “Advances in Insect Physiology”, Vol. 2,67-129. Academic Press, New York and London. Beatty, I. M. and Gilby, A. R. (1969). The major hydrocarbon of a cockroach cuticular wax. NaturwissenschaJfen, 54 373-374. Bell, R. A., Nelson, D. R., Borg, T. K. and Cardwell, D. L. (1975). Wax secretion in nondiapausing and diapausing pupae of the tobacco hornworm, Manduca sexta. J. Insect PhySiOl. 21, 1725-1729. Bennett, G. A., Kleiman, R. and Shotwell, 0. L. (1972). Hydrocarbons in haemolymph from healthy and diseased Japanese beetle larvae. J. Insect Physiol. 18, 1343-1350. Bergstrom, G. and Lofqvist, J. (1973). Chemical congruence of the complex odoriferous secretions from Dufour’s gland in three species of ants of the genus Formica. J. Insect Physiol. 19,877-907. Biemann, K. (1962). “Mass Spectrometry Organic Chemical Applications”, pp. 46-161. McGraw-Hill, New York. Blailock, T. T., Blomquist, G. J. and Jackson, L. L. (1976). Biosynthesis of 2-methylalkanes in the crickets Nemobius fasciarus and Gryllus pennsylvanicus. Biochem. Biophys. Res. Commun. -84 1-849. Blomquist, G. J., BlailOck, T. T., Scheetz, R. W. and Jackson, L. L. (1976). Cuticular lipids of Insects, VII. Cuticular hydrocarbons of the crickets, Acheta domesticus, Gryllus pennsylvanicus, and Nemobiusfasciatus. Comp. Biochem. Physiol. 54B,38 1-386.
LONG-CHAIN METHYL-BRANCHED HYDROCARBONS
27
Blomquist, G. J. and Jackson, L. L. (1973). Incorporation of labelled dietary n-alkanes into cuticular lipids of the grasshopper Melanoplus sanguinipes. J. Insect Physiol. 19, 1639-1647. Blomquist, G. J. and Kearney, G. P. (1976). Biosynthesis of internally branched monomethylalkanes in the cockroach Periplaneta fulinginosa. Arch. Biochem. Biophys. 173,546-553. Blomquist, G. J., Major, M. A. and Lok, J. B. (1975). Biosynthesis of 3-methylpentacosane in the cockroach Periplaneta americana. Biochem. Biophys. Res. Comm. 64,43-50. Blum, M. S., Wallace, J. B. and Fales, H. M. (1973). Skatole and tridecene: Identification and possible role in a chrysopid secretion. Insect Biochem. 3,353-357. Bowers, W. S. and Thompson, M. J. (1965). Identification of the major constitutents of the crystalline powder covering the larval cuticle of Samia Cynthia ricini (Jones). J. Insect Physiol. 11, 1003-101 1. Brieskorn, C. H. and Beck, K. R. (1970). Die Kohlenwasserstoffe des Blattwachses von Rosmarinus oflcinalis. Phytochemistry, 9, 1633- 1640. Brieskorn, C. H. and Feilner, K. (1968). Zum Aufbau des pflanzlichen Abschlussgewebes: die normalen und verzweigten Alkane von Marrubium vulgare L. Phytochemistry, 7,485-492. Brophy, J. J., Cavill, G. W. K. and Shannon, J. S. (1973). Venom and Dufow’s gland secretions in an Australian species of Camponotus. J. Insect Physiol. 19,791-798. Brossut, R., Dubois, P. and Rigaud, J. (1974). Le gregarisme chez Blaberus craniifer: Isolement et identification de la phkromone. J. Insect Physiol. 20,529-543. Brown, K. S., Jr. (1975). The chemistry of aphids and scale insects. Chem. SOC.Rev. 4, 263288.
Browning, T. 0. (1967). Water, and the eggs of insects. In “Insects and Physiology” (Eds J. W. L. Beament and J. E. Treherne), pp. 315-328. Oliver and Boyd, London. Buckner, J. S. and Kolattukudy, P. E. (1975). Lipid biosynthesis in the sebaceous glands: Synthesis of multibranched fatty acids from methylmalonyl-coenzyme A in cell-free preparations from the uropygial gland of goose. Biochemistry, 14,1774-1782. Bursell, E. and Clements, A. N. (1967). The cuticular lipids of the larva of Tenebrio molitor L. (Coleoptera). J. Insect Physiol. 13, 1671-1678. Calam, D. H. (1971). Natural Occurrenceof fatty acid ethyl esters. Science, 174, 78. Caldicott, A. B. and Eglinton, G. (1973). Surface waxes. In “Phytochemistry” (Ed. L. P. Miller) Vol. 111, pp. 162-194. Van Nostrand Reinhold Co., New York. Carlson, D. A., Mayer, M. S.,Silhacek, D. L., James, J. D., Beroza, M.and Bierl, B. A. (1971). Sex attractant pheromone of the house fly: Isolation, identification and synthesis. Science, 174,76-78.
Cassagne, C. and Lessire, R. (1975). Studies on alkane biosynthesis in epidermis of AIlium porrum L. leaves. IV. Wax movement into and out of the epidermal cells. Plant Sci. Let. 5 , 26 1-268.
Cavill, G. W. K. and Houghton, E. (1973). Hydrocarbon constitutents of the Argentine ant, Iridomyrmex humilis. Aust. J. Chem. 25, 1131-1 135. Cavill, G. W. K. and Williams, P. J. (1967). Constituents of Dufour’s gland in Myrmecia gulosa. J. Insect Physiol. 13, 1097-1 103. Cavill, G. W. K., Clark, D. V., Howden, M. E. H. and Wyllie, S. G. (1970). Hydrocarbon and other lipid constituents of the bull ant, Myrmecia gulosa. J. Insect Physiol. 16, 17211728.
Conrad, C. W. and Jackson, L. L. (1971). Hydrocarbon biosynthesis in Periparteta americana. J. Insect Physiol. 17, 1907-1916. David, W. A. L. (1967). The physiology of the insect integument in relation to the invasion of pathogens. In “Insects and Physiology” (MsJ. W. L. Beament and J. E. Treherne), pp. 1735. Oliver and Boyd, London. Diehl, P. A. (1973). Para& synthesis in the oenocytes of the desert locust. Nature, Lond. 243, 468-470.
DENNIS R. NELSON
28
Diehl, P. A. (1975). Synthesis and release of hydrocarbons by the oenocytes of the desert locust, Schistocerca gregaria. J. Insect Physiol. 21, 1237-1246. Ebeling, W. (1964). The permeability of insect cuticle. In “The Physiology of Insecta” (Ed. M. Rockstein), Vol. 111, pp. 507-556. Academic Press, New York. Eglinton, G. and Hamilton, R. J. (1967). Leafepicuticular waxes. Science, 156, 1322-1335. Eglinton, G., Scott, P. M., Belsky, T., Burlingame, A. L., Richter, W. and Calvin, M. (1966). Occurrence of isoprenoid alkanes in a Precambrian sediment. In “Advances in Organic Geochemistry” (Eds. G. D. Hobson and M. C. Louis), pp. 41-74. Pergamon Press, New York. Eisner, J., Iverson, J. L., Mozingo, A. K.and Firestone, D. (1965). Oils, fats, and waxes. Gas chromatography of unsaponifiable matter. 111. Identification of hydrocarbons, aliphatic alcohols, tocopherols, triterpenoid alcohols, and sterols present in olive oils. J. Assoc. Of. Agric. Chem. 48,4 17-433. von Endt, D. W. and Wheeler, J. W. (1971). 1-Pentadecene production in Tribolium confusum. Science, 172, 60-6 1. Enzell, C. R., Rosengren, A. and Wahlberg, I. (1969). Tobacco chemistry. I. The occurrence of C,, to C,, iso-paraffins and C,, to C,, anteiso-paraffins in Greek tobacco. Tob. Sci. (New York), 13, 127-129. Evans, D. A. and Green, C. L. (1973). Insect attractants of natural origin. Chem. Soc. Rev. 2, 75-97.
Faurot-Bouchet, E. and Michel, G. (1964). Composition of insect waxes. I. Waxes of exotic Coccidae: Gascardia madagascariensis, Coccus ceriferus, and Tachardia lacca. J. Am. Oil Chem. Soc. 41,418-421. Faurot-Bouchet, E. and Michel, G. (1965). Composition des cires d’insectes. 11. Cires des cochenilles Ceroplastes rusci, Zcerya purchasi, Pulvinaria floeifera, et Quadraspidiotus perniciosus. Bull. Soc. Chim. Biol. 47,93-97. Fehler, S. W. G. and Light, R. J. (1970). Biosynthesis of hydrocarbons in Anabaena variabilis. Incorporation of [methyP4Cl- and [methyl-2H31methionineinto 7- and I-methylheptadecanes, Biochemistry, 9,418-422. Fehler, S . W. G. and Light, R. J. (1972). Biosynthesis of methylheptadecanes in Anabaena Biochemistry, 11, variabilis. In vitro incorporation of S-Imethyl-14Cladenosylmethionine. 241 1-2416.
Gallegos, E. J. (1971). Identification of new steranes, terpanes, and branched paraffins in Green River shale by combined capillary gas chromatography and mass spectrometry. Anal. Chem. 43, 1151-1160. Gaskell, S. J., Eglinton, G. and Bruun, T. (1973). Hydrocarbon constituents of three species of Norwegian lichen: Cetraria nivalis, C. crispa, Siphula ceratites. Phytochemistry, 12, 11741176.
Gastambide-Odier, M. and Lederer, E. (1959). Biosynthesis of corynomycolic acid from two molecules of palmitic acid. Nature, Lond. 184, 1563-1564. Gastambide-Odier, M., Delaumbny, J-M. and Lederer, E. (1963). Biosynthese de I’acide C3,mycocerosique. Incorporation d‘acide propionique. Biochim. Biophys. Acta, 70,670-678. Gelpi, E., Schneider, H., Mann, J. and Oro, J. (1970). Hydrocarbons of geochemical significance in microscopic algae. Phytochemistry, 9,603-6 12. Gibert, J. M., De Andrade Bruning, I. M. R., Nooner, D. W. and Oro, J. (1975). Predominance of isoprenoids among the alkanes in the Irati oil shale, Permian of Brazil. Chem. Geol. 15, 209-215.
Giese, B. N. (1975). Effects of light and temperature on the composition of epicuticular wax of barley leaves. Phytochemistry, 14,92 1-929. Gilby, A. R. (1957a). Studies of cuticular lipids of arthropods. 11. The chemical composition of the wax from Ceroplastes destructor (Newt.). Arch. Biochem. Biophys. 67,307-319.
LONG-CHAIN METHYL-BRANCHED HYDROCARBONS
29
Gilby, A. R. (1957b). Studies of cuticular lipids of arthropods. 111. The chemical composition of the wax from Boophilus microplus (Can.). Arch. Biochem. Biophys. 67,320-324. Gmgrich, J. B. (1975). Ultraviolet-induced changes in cuticular waxes of American cockroaches, Periplaneta americana (L.) (Dictyoptera: Blattaria: Blattidae). Can. J. 2001.53,
1238-1 240. Gohring, K. E. H., Schenck, P. A. and Engelhardt, E. D. (1967). A new series of isoprenoid isoalkanes in crude oils and cretaceous bituminous shales. Nature, Lond. 215,503-505. Hamilton, R. J., Raie, M. Y.,Weatherston, I., Brooks, C. J. and Borthwick, J. H. (1975). Crustacean surface waxes. Part I. The hydrocarbons from the surface of Ligia oceanica. J.C.S. Perkin I, 354-357. Han, J., Chan, H. W-S. and Calvin, M. (1969). Biosynthesis of alkanes in Nostoc muscorum. J. Am. Chem. SOC.91,51 5 6 5 159. Han, J., McCarthy, E. D., Calvin, M. and Benn, M. H. (1968). Hydrocarbon constituents of the blue-green algae Nostoc muscorum. Anacystis nidulans, Phormidium luridum and Chlorogloeafritschii. J. Chem. SOC.(C) 2785-2791. Haug, P. and Curry, D. J. (1974).Isoprenoids in a Costa Rican seep oil. Geochim. Cosmochim. Acta, 38,601-610. Hedin, P. A., Thompson, A. C., Gueldner, R. C. and Minyard, J. P. (1972). Volatile constituents of the boll weevil. J. Insect Physiol. 1879-86. Hedin, P.A., Gueldner, R. C., Henson, R. D. and Thompson, A. C. (1974). Volatile constituents of male and female boll weevils and their frass. J. Insect Physiol. 20,2135-2142. Hills, I. R. and Whitehead, E. V. (1966). Triterpanes in optically active petroleum distillates. Nature, Lond. 209,977-979. Hood, A. (1963).The molecular structure of petroleum. In “Mass Spectrometry of Organic Ions” (Ed. F. W. McLafferty), pp. 597-635.Academic Press, New York. Hutchins, R. F. N. and Martin, M. M. (1968). The lipids of the common house cricket, Achera domesticus L. 11. Hydrocarbons. Lipids, 3,250-255. Jackson, L. L. (1970). Cuticular lipids of insects. 11. Hydrocarbons of the cockroaches Periplaneta australasiae, Periplaneta brunnea and Periplaneta fuliginosa. Lipids, 5,3841. Jackson, L. L. (1972). Cuticular lipids of insects. IV. Hydrocarbons of the cockroaches Periplaneta japonica and Periplaneta americana compared to other cockroach hydrocarbons. Comp. Biochem. Physiol. 41B,33 1-336. Jackson, L. L. and Baker, G. L. (1970).Cuticular lipids of insects. Lipids, 5,239-246. Jackson, L. L. and Blomquist, G. L. (1976). Cuticular lipids of insects. VIII. Alkanes of the Mormon cricket Anabrus simplex. Lipids, 11,77-79. Jackson, L. L., Armold, M. T. and Regnier, F. E. (1974). Cuticular lipids of adult fleshflies, Sarcophaga bullata. Insect Biochem. 4,369-379. Jarolimek, P., Wollrab, V., Streibl, M. and Sorm, F. (1964). Multi-branched alkanes. newly identified substances from plant waxes. Chem. Ind. 237-238. Jaureguiberry, G., Law, J. H., McCloskey, J. A. and Lederer, E. (1965). Studies on the mechanism of biological carbon alkylation reactions. Biochemistry, 4,347-353. Jaureguiberry, G., Lenfant, M., Toubiana, R., Azerad, R. and Lederer, E. (1966). Biosynthesis of tuberculostearic acid in a cell-free extract. Identification of 10-methylenestearic acid as an intermediate. Chem. Commun. 23,855-857. Johns, R. B., Belsky,T., McCarthy, E. D., Burlingame, A. L., Haug, P., Schnoes, H. K., Richter, W. and Calvin, M. (1966). The organic geochemistry of ancient sediments. Part 11. Geochim. Cosmochim. Acta, 30, 1191-1222. Jones, R. L., Lewis, W. J., Bowman, M. C., Beroza, M. and Bierl, B. A. (1971).Host-seeking stimulant for parasite of corn earworm: Isolation, identification, and synthesis. Science, 173,
842-843. Jones, R. L., Lewis, W. J., Beroza, M., Bierl, B. A. and Sparks, A. N. (1973). Host-seeking
30
DENNIS R . NELSON
Stimulants (kairomones) for the egg parasite, Trichogramma evanescens. Environ. Entomol. 2,593-596. Kaneda, T . (1963). Biosynthesis of branched chain fatty acids. 11. Microbial synthesis of branched long chain fatty acids from certain short chain fatty acid substrates. J. Biol. Chem. 238,1229-1235. Kaneda, T. (1966). Biosynthesis of branched-chain fatty acids. V. Microbial stereospecific syntheses of D- 12-methyltetradecanoic and D- 14-methylhexadecanoic acids. Biochim. Biophys. Acta, l2S, 43-54. Kaneda, T. (1967). Biosynthesis of long-chain hydrocarbons. I. Incorporation of L-valine, Lthreonine, L-isoleucine, and Lleucine into specific branched-chain hydrocarbons in tobacco. Biochemistry, 6,2023-2032. Karunen, P. (1974). Polyunsaturated hydrocarbons from Polytrichum commune spores. Phytochemistry, 13,2209-2213. Keville, R. and Kannowski, P. B. (1975). Sexual excitation by pheromones of the confused flour beetle. J. Insect Physiol. 21,8 1-84. Kolattukudy, P. E. (1968). Biosynthesis of surface lipids. Science, 159,498-505. Kolattukudy, P. E. (1975). Biochemistry of cutin, suberin and waxes, the lipid barriers on plants. In “Recent Advances in the Chemistry and Biochemistry of Plant Lipids” (Eds. T. Galliard and E. I. Mercer), Vol. 12, pp. 203-246. Academic Press, New York. Kolattukudy, P. E. and Brown, L. (1974). Inhibition of cuticular lipid biosynthesis in Pisum salivum by thiocarbamates. Plant Physiol. 53,903-906. Kolattukudy, P. E. and Walton, T. J. (1972). Metabolism of alkyl glyceryl ethers and their noninvolvement in alkane biosynthesis in plants. Arch. Biochem. Biophys. 150, 3 10-3 17. Kolattukudy, P. E. and Walton, T. J. (1973). The biochemistry of plant cuticular lipids. In “Progress in the Chemistry of Fats and Other Lipids” (Ed. R. T. Holman), Vol. XIII, pp. 121-175. Pergamon Press, New York. Kostelc, J. G.,Hendry, L. B. and Snetsinger, R. J. (1975). A sex pheromone complex of the mushroom-infesting sciarid fly, Lycoriella mali Fitch. J. New York Entomol. SOC.LXXXIII, 255-256. Kovacev, G., Delova, D. and StranskL, K. (1972). Paraffinic hydrocarbons from Devonian, Triassic and Jurassic sediments of Northern Bulgaria. Collect. Czech. Chem. Commun. 37, 4 106-41 11. Langley, P. A., Pimley, R. W. and Carlson, D. A. (1975). Sex recognition pheromone in tsetse fly Glossina rnorsitans. Nature. Lond. 254,s 1-53. Lederer, E. (1964). The origin and function of some methyl groups in branched-chain fatty acids, plant sterols and quinones. Biochem. J. 93,449-468. Lederer, E. (1969). Some problems concerning biological C-alkylation reactions and phytosterol biosynthesis. Quart. Rev. 23,453-48 1. Levy, E. J., Doyle, R. R., Brown, R. A. and Melpolder, F. W. (1961). Identification of components in paraffin wax by high temperature gas chromatography and mass spectrometry. Anal. Chem. 33,698-704. Lewis, W. J., Jones, R. L.,Gross,H.R., Jr. and Nordlund, D. A. (1976). The role of kairomones and other behavioral chemicals in host finding by parasitic insects. Behavioral Biol. 16, 267289. Lewis, W. J., Jones, R. L., Nordlund, D. A. and Sparks, A. N. (1975a). Kairomones and their use for management of entomophagous insects: I. Evaluation for increasing rates of parasitization by Trichogramma spp. in the field. J. Chem. Ecol. I, 343-347. Lewis, W. J., Jones, R. L.,Nordlund, D. A. and Gross, H.R. Jr. (1975b). Kairomones and their use for management of entomophagous insects: 11. Mechanisms causing increase in rate of parasitization by Trichogramma spp. J. Chem.Ecol. 1,349-360.
LONG-CHAIN METHYL-BRANCHED HYDROCARBONS
31
Liu, T. Y.and Hofmann, K. (1962). Cyclopropane ring biosynthesis. Biochemistry, 1, 189-191. Lok, J. B., Cupp, E. W. and Blomquist, G. J. (1975). Cuticular lipids of the imported fire ants, Solenopsis inuicta and richteri. Insect Biochem. 5,82 1-829. Louloudes, S. J., Chambers, D. L., Moyer, D. B. and Starkey, J. H., 111. (1962). The hydrocarbons of adult house flies. Ann. Entomol. Soc. Am. 55,442-448. MacConnell, J. G. (1969). The alkanes of the ant, Attica colombica. Ph.D. Thesis, University of Michigan, Ann Arbor, 125 pp. Martin, M. M. and MacConnell, J. G. (1970). The alkanes of the ant, Atta colombica. Tetrahedron, 26,307-3 19. McCarthy, E. D. and Calvin, M. (1967). The isolation and identification of the C,, saturated isoprenoid hydrocarbon, 2,6,10-trimethyltetradecane from a Devonian shale. The role of squalene as a possible precursor. Tetrahedron, 23,2609-26 19. McCarthy, E. D., Han, J. and Calvin, M. (1968). Hydrogen atom transfer in mass spectrometric fragmentation patterns of saturated aliphatic hydrocarbons. Anal. Chem. 40, 14751480. Meinwald, J., Smolanoff, J., Chibnall, A. C. and Eisner, T. (1975). Characterization and synthesis of waxes from homopterous insects.J. Chem. Ecol. I, 269-274. Meinschein, W. G., Cordes, E. and Shiner, V. J., Jr. (1970). Search for alkanes of 15 to 30 carbon atom length. Science, 167,753. Metzler, M., Dahm, K. H., Meyer, D. and Roller, H. (1971). On the biosynthesis of juvenile hormone in the adult cecropia moth. Z . Nafurforsch. 826, 1270-1276. Miwa, T. K. (1963). Identification of peaks in gas-liquid chromatography. J. Am. Oil Chem. SOC.40,309-3 13. Mody, N. V., Hedin, P. A., Neel, W. W. and Miles, D. H. (1975). Hydrocarbons from males, females, and larvae of pecan weevil: Curculio caryae (Horn). Lipids, 10, 117-1 19. Mold, J. D., Means, R. E., Stevens, R. K.and Ruth, I. M. (1964). The paraffin hydrocarbons of wool wax. Normal, iso, anteiso, and other brenched isomers. Biochemistry, 3, 1293-1298. Mold, J. D., Means, R. E., Stevens, R. K. and Ruth, J. M. (1966). The p a r a n hydrocarbons of wool wax. Homologous series of methyl alkanes. Biochemistry, 5,455-46 1. Mold, J. D., Stevens, R. K., Means, R. E. and Ruth, J. M. (1963). 2,6,10,14-TetramethyIpentadecane (pristane) from wool wax. Nature, Lond. 199,283-284. Murata, T., Takahashi, S. and Shikata, M. (1974). Studies on the structure of the integument of the silkworm, with special reference to its cuticular lipids. XI. Identification by the gas chromatography-mass spectrometry of hydrocarbons from exuviae in 5th instar larvae. J. Sericult. Sci. Japan, 43, 105-1 10. Nagy, B., Modzeleski, V. E. and Scott, W. M. (1969). Saturated hydrocarbons in bovine liver. Biochemistry J. 114,645-648. Nelson, D. R. (1969). Hydrocarbon synthesis in the American cockroach. Nature, Lond. 221, 854-855. Nelson, D. R. and Sukkestad, D. R. (1970). Normal and branched aliphatic hydrocarbons from the eggs of the tobacco hornworm. Biochemistry, 9,4601-46 11. Nelson, D. R. and Sukkestad, D. R. (1975). Normal and branched alkanes from cast skins of the grasshopper Schistocerca uaga (Scudder). J. Lipid Res. 16, 12-18. Nelson, D. R., Sukkestad, D. R. and Terranova, A. C. (1971). Hydrocarbon composition of the integument, fat body, hemolymph, and diet of the tobacco hornworm. Life Sci. 10, Part 11, 411-419. Nelson, D. R., Sukkestad, D. R. and Zaylskie, R. G. (1972). Mass spectra of methyl-branched hydrocarbons from eggs of the tobacco hornworm. J. Lipid Res. 13,4 13-421. Nicholas, H. J. and Bombaugh, K. J. (1965). Constituents of the hydrocarbon fraction of beef brains. Biochim. Biophys. Acta, 98, 372-379.
32
DENNIS R. NELSON
Nicolaides, N. (1974). Skin lipids: Their biochemical uniqueness. Science, 186, 19-26. Nishimoto, S. (1974). Branched chain alkane in surface waxes of waxy wheats and nowaxy wheats. J. Sci. Hiroshima Univ., Ser. A , 38, 165-174. O’Connor, J. G., Burow, F. H. and Norris, M. S. (1962). Determination of normal paraffins in C,, to C,, paraffin waxes by molecular sieve adsorption. Anal. Chem. 34,82-85. Orb, J., Gelpi, E. and Nooner, D. W. (1968). Hydrocarbons in extraterrestrial samples. J. Brit. Interplanetary Soc. 21,83-98. Oudejans, R. C. H. M. (1972). Composition of the saturated hydrocarbons from males, females, and eggs of the millipede, Graphidostreptus tumuliporus. J. Insect Physiol. 18,857-863. Oudejans, R. C. H. M. (1973). Composition of the unsaturated hydrocarbons of the millipede Graphidostreptus tumuliporus (Karsch) (Diplopoda: Spirostreptida). Netherlands J. Zool. 23, 330-342. Oudejans, R. C. H. M. and Zandee, D. I. (1973). The biosynthesis of the hydrocarbons in males and females of the millipede, Graphidostreptus tumuliporus. J. Insect Physiol. 19,2245-2253. Philogene, B. J. R. and McFarlane, J. E. (1967). The formation of the cuticle in the house cricket, Acheta domesticus (L.) and the role of oenocytes. Can. J. Zool. 45, 181-190. Piek, T. (1964). Synthesis of wax in the honeybee (Apis rnellifera L.). J. Insect Physiol. 10,563572. Polacheck, J. W., Tropp, B. E., Law, J. H. and McCloskey, J. A. (1966). Biosynthesis of cyclopropane compounds-VIII. The conversion of oleate to dihydrosterculate. J. Biol. Chem. 241,3362-3364. Pomonis, J. G., Nelson, D. R., Fatland, C. L. and Zaylskie, R. G. (1978). Graphic relationships between pairs of adjacent fragment ion peaks from the mass spectra of mono- and dimethyl substituted alkanes obtained from insect waxes and by synthesis. Anal. Chem. (In preparation). Regnier, F. E. and Wilson, E. 0. (1968). The alarm-defense system of the ant Acanthomyops claviger. J. Insect. Physiol. 14,955-970. Regnier, F. E. and Wilson, E. 0. (1969). The alarm-defense system of the ant Lasius alienus. J. Insect Physiol. 15,893-898. Regnier, F. E., Nieh, M. and Holldobler, B. (1973). The volatile Dufour’s gland components of the harvester ants Pogonomyrmex rugosus and P. barbatus. J. Insect Physiol. 19,98 1-992. Roelofs, W. L. and Card&,R. T. (I97 1). Hydrocarbon sex pheromone in tiger moths (Arctiidae). Science, 171,684-686. Schooley, D. A., Judy, K. J., Bergot, B. J., Hall, M. S. and Siddall, J. B. (1973). Biosynthesis of the juvenile hormones of Manduca sexta: Labeling pattern from mevalonate, propionate, and acetate. Proc. Nat. Acad. Sci. 10,2921-2925. Shikata, M., Murata, T. and Kaihatu, M. (1974). Studies on the structure of the integument of the silkworm, with special reference to its cuticular lipids. MI. Change of the hydrocarbons in the cuticular lipids during the metamorphosis. J. Sericulf. Sci. Japan, 43, 157-162. Soliday, C. L., Blomquist, G. J. and Jackson, L. L. (1974). Cuticular lipids of insects. VI. Cuticular lipids of the grasshoppers Melanoplus sanguinipes and Melanoplus packardii. J. Lipid Res. 15,399405. Still, G. G., Davies, D. G. and Zander, G. L. (1970). Plant epicuticular lipids: Alteration by herbicidal carbamates. Plant Physiol. 46,307-3 14. Stransky, K. and Streibl, M. (1969). On natural waxes. XII. Composition of hydrocarbons in morphologically different plant parts. Collect. Czech. Chem. Cornmun. 34, 103-1 17. Stransky, K., Streibl, M. and Kubelka, V. (1970). On natural waxes. XV. Hydrocarbon constituents of the leaf wax from the walnut tree (Juglans regia L.). Collect. Czech. Chem. Cornmun. 35,882-891. Strimsky, K., Streibl, M. and Sorm, F. (1966). uber Natunvachse IV. uber einen neuen Typ
LONG-CHAIN METHYL-BRANCHED HYDROCARBONS
33
verzweigter Paraffine aus dem Wachs der Honigbiene (Apis mellifera L.). Collect. Czech. Chem. Commun. 31,4694-4702. Stransky, K., Streibl, M. and Sorm, F. (1968). On natural waxes VII. Lipid hydrocarbons of the alga Scenedesmus quadricauda (Turp.) Breb. Collect. Czech. Chem. Commun. 33,4 16-424. Stransky, K., Ubik, K., Holman, J. and Streibl, M. (1973). Chemical composition of compounds produced by the pea aphid Acyrthosiphon pisum (Harris): Pentane extract of surface lipids. Collect. Czech. Chem. Commun. 38,770-780. Streibl, M., KoneEny, K., Trka, A., Ubik, K. and Pazlar, M. (1974). Composition of the wax of rye straw. Collect. Czech. Chem. Commun. 39,475-479. Streibl, M., Stransky, K. and Sorm, F. (1966). Uber einige neue Kohlenwasserstoffe im Wachs der Honigbiene (Apis melliJca L.) Fette Seifen Anstrichm. 68,799-805. Tartivita, K. and Jackson, L. L. (1970). Cuticular lipids of insects. I. Hydrocarbons of Leucophaea maderae and BIatta orientalis. Lipids, 5,35-37. Tornabene, T. G. and Markey, S. P. (1971). Characterization of branched monounsaturated hydrocarbons of Sarcina lutea and Sarcinaflava. Lipids, 6, 190-195. Turner, R. B. and Acree, F., Jr. (1967). The effect of photoperiod on the daily fluctuation of haemolymph hydrocarbons in the American cockroach. J. Insect Physiol. 13,5 19-522. Uebel, E. C., Menzer, R. E., Sonnet, P. E. and Miller, R. W. (1975a). Identification of the copulatory sex pheromone of the little house fly, Fannia canicularis (L.) (Diptera: Muscidae). J. New York Entomol. Soc. LXXXIII, 259. Uebel, E. C., Sonnet, P. E., Bierl, B. A. and Miller, R. W. (1975b). Sex pheromone of the stable fly: Isolation and preliminary identification of compounds that induce mating strike behavior. J. Chem. Ecol. 1,377-385. Uebel, E. C., Sonnet, P. E., Miller, R. W. and Beroza, M. (197%). Sex pheromone of the face fly, Musca autumnalis De Geer (Diptera: Muscidae). J. Chem. Ecol. I, 195-202. Urbach, G. and Stark, W. (1975). The C-20 hydrocarbons of butterfat. J. Agr. Food Chem. 23, 20-24.
Van Der Horst, D. J. and Oudejans, R. C. H. M. (1972). Hydrocarbons in the land snail Cepaea nemoralis (L.) (Gastropoda, Pulmonata). Comp. Biochem. Physiol. 41B,823-829. Van Der Horst, D. J., Oudejans, R. C. H. M., Plug, A. C. and Van Harmelen, H. J. M. (1973). Biosynthesis of cyclopropane fatty acids in the millipede Graphidostreptus tumuliporus (Karsch) (Diplopoda: Spirostreptida). Comp. Biochem. Physiol. 46B,3 9 5 4 0 4 . Vinson, S. B., Jones, R. L., Sonnet, P. E., Bierl, B. A. and Beroza, M. (1975). Isolation, identification, and synthesis of host-seeking stimulants for Cardiochiles nigriceps, a parasitoid of the tobacco budworm. Entomol. Exp. Appl. 18,443-450. Weete, J. D. (1972). (Review article) Aliphatic hydrocarbons of the fungi. Phytochemistry, 11, 120 1-1 205.
Wilkinson, R. E. and Kasperbauer, M. J. (1972). Epicuticular alkane content of tobacco as influenced by photoperiod, temperature and leaf age. Phytochemistry, 11,2439-2442. Wollrab, V. (1968). Uber Naturwachse VIII. Olefine und Paraffine aus den Wachsen einiger Manzen der Familie Rosaceae. Collect. Czech. Chem. Commun. 33, 1584-1600. Wollrab, V., Streibl, M. and Sorm, F. (1965a). a e r Manzenstoffe XXI. f h e r die Zusammenseztung der Kohlenwasserstoffe aus dem Wachs der Rosenblutenblatter. Collect. Czech. Chem. Commun. 30,1654-1669. Wollrab, V., Streibl, M. and Sorm, F. (1965b). Uber Pflanzenstoffe. XXII. Uber die Kohlenwasserstoffzusammensetzung des Wachses aus der Hopfendolde (Humulus lupulus L.).Collect. Czech. Chem. Commun. 30, 1670-1675. Wollrab, V., Streibl, M. and Sorm, F. (1967). Iso- and anteiso-alkanes in natural waxes. Chem. Indust. (London), 1872-1873. Youngblood, W. W. and Blumer, M. (1973). Alkanes and alkenes in marine benthic algae. Marine Biol. 21, 163-172.
This Page Intentionally Left Blank
Insect Visual Pigments Richard H. White Biology Department, University of Massachusetts at Boston, Boston, Massachusetts, USA
1 Introduction 35 2 Extraction and measurement of insect visual pigments 38 3 Rhodopsin and metarhodopsin 4 0 4 Chromophore and photochemistry 47 5 Regeneration in insect visual systems 5 1 6 Insect color vision and ultraviolet sensitivity 53 7 The problem of the visual pigments of the higher flies 55 8 Transduction and adaptation 57 9 Insect photorgceptor membranes 60 10 Fmalcomments 62 Acknowledgements 62 References 62
1
Introduction
The work of the past decade has begun to outline the particular features of insect visual pigments. Until recently, we knew these photopigments only by inference from electrophysiological and behavioral measurements of spectral sensitivity. The photochemical interpretations of those physiological measurements were typically drawn in terms of the well-known characteristics of vertebrate photopigments. We have now come to realize, however, that extrapolation from vertebrate to invertebrate photoreceptors can be misleading in some important respects. With the characterization of insect visual pigments, the physiology of insect vision is at last being provided with a proper foundation. Although they are different in certain respects, insect visual pigments are similar in their basic features to the photopigments of vertebrates, and studies of insect visual pigments at the present time are necessarily comparative. Recent trends of research into the molecular basis of vertebrate vision, and the 35
36
RICHARD H. WHITE
biochemistry of photoreceptor membranes have been reviewed by Daemen (1973) and Ebrey and Honig (1975). These reviews should be taken as the citations for general statements about visual pigments in the present article. For background, I need only summarize the well-established characteristics of the vertebrate visual pigments that may be broadly designated “rhodopsins”.* Vertebrate rhodopsins consist of a glycoprotein, opsin, and a chromophore, retinaldehyde (more commonly retinal; formerly retinene), the aldehyde of vitamin A. Rhodopsins are hydrophobic proteins, whose natural environment is the specialized membrane of a photoreceptor cell. The chromophore is bound covalently to the opsin in a Schiff base linkage (-C=N-) between the carbonyl of the retinal and an &-aminogroup of a lysine in the protein. Direct spectra of purified vertebrate rhodopsins consist of a main band (the a-band) peaking in the visible, a secondary peak (the P-band) with much lower extinction in the near ultraviolet and a y-band at 280nm due to the absorbance of the aromatic amino acids of opsin (Fig. 1). The main absorption band of the free chromophore lies in the near ultraviolet, at about 380nm (Fig. 1). When it binds to an opsin, forming rhodopsin, the main absorption maximum (Amax) shifts into the visible region of the spectrum. The basis of this “bathochromatic shift” of chromophore absorbance to longer wavelengths is a central problem in current research on the molecular structure of visual pigments. Retinal is a polyene, characterized by a backbone of alternating double and single bonds (Fig. 1). Delocalization of the n-electron system of such a polyene would be expected to shift its absorbance to longer wavelengths. Such a modification of the chromophore’s electronic structure is thought to be accomplished by protonation of the Schiff base linkage between the opsin and the chromophore, and by additional poorly characterized interactions between the chromophore and the protein. The main absorption maxima of the visual pigments belonging broadly to the rhodopsin class range among the vertebrates from 430nm in the blue to 580nm in the yellow-orange. In these various rhodopsins, the chromophore is the same; only the opsins differ. Therefore, it is thought that the absorption maxima of different visual pigments are fine-tuned by the particular disposition of the charged opsin groups at the site of the chromophore. This is an inference, however, for the detailed protein structure of a visual pigment has not been determined as yet. Retinal, as a polyene, can exist as a number of geometric isomers. The chromophore of rhodopsin is specifically the 11-cis isomer of retinal, in which the polyene backbone is bent and twisted around carbon 11 (Fig. 1). The essential action of light absorbed by a molecule of rhodopsin is the photo*“Porphyropsins” form a second class of vertebrate visual pigments. They differ from rhodopsins only in their chromophore, which is 3-dehydroretinal. Porphyropsins are formed mainly in fresh water fishes and larval amphibians. Porphyropsin pigments have not been found in invertebrates.
INSECT VISUAL PIGMENTS
37
chemical isomerization of the chromophore from the bent 11-cis to the straight all-trans configuration. In vertebrates, the protein then undergoes conformational changes not requiring light that lead to the dissociation of the all-trans chromophore from the protein, leaving free retinal and opsin. Since rhodopsin absorbs in the visible, while free retinal and opsin absorb in the ultraviolet, which we cannot see, this process is called bleaching. When slowed at low temperature, or followed by fast photometric techniques, the bleaching process can be resolved into a series of spectrally distinct stages, the intermediates of ISOMERS OF RETINAL
E
Wavelength-nm Fig. 1. Absorbance spectra of retinal and cattle rhodopsin. (A,,,, all-trans and 1 1-cis isomers of retinal.
5 0 0 nm). The inset shows the
bleaching. A number of lines of evidence indicate that the stages of bleaching reflect the progressive “opening up” of the opsin’s structure, and the loosening of the association of the chromophore with it. For instance, intact rhodopsin is stable to a number of reagents that successfully attack opsin or retinal. All vertebrate visual pigments progress through similar bleaching stages, ending with the hydrolysis of the chromophore from the opsin. As rhodopsins are membrane bound proteins, they can be brought into solution without denaturation only in detergent extracts. Digitonin has long been the standard extractant; other detergents have been introduced in recent
RICHARD H. WHITE
38
years. Since vertebrate rhodopsins are bleached by light, difference spectra calculated by subtracting bleached from dark spectra accurately trace the abands of those rhodopsins whose Lax lie at wavelengths greater than 450 nm. In this way, vertebrate visual pigments can be measured in relatively crude extracts or in situ; in membrane suspensions, in whole retinas, in single photoreceptor cells by microspectrophotometry, and in intact functioning eyes by reflectance photometry. 2
Extraction and measurement of insect visual pigments
Well before arthropod rhodopsins were convincingly localized by direct measurement it had been inferred that they must be associated with arrays of microvilli elaborated from the plasma membranes of photoreceptor cells (see review by Win, 1972). The mass of microvilli of a particular receptor cell is called its rhabdomere. Where the rhabdomeres from several contiguous cells cluster together they are collectively termed a rhabdom [Fig. 3(c)l. The association of photopigments with rhabdomeres has been shown directly by microspectrophotometry (Langer and Thorell, 1966; Hays and Goldsmith, 1969; Brown and White, 1972). Like other visual pigments, insect rhodopsins can be brought into solution intact only in detergent micelles (Woken and Scheer, 1963; Gogola et al., 1970; Marak et al., 1970; Hamdorf et al., 1971b; Paulsen and Schwemer, 1972; Schwemer and Paulsen, 1973; Fernandez and Bishop, 1973; Ostroy et al., 1974). Therefore, they must be bound within the microvillus membranes of the rhabdomeres. Digitonin has generally been the detergent used for extracting insect rhodopsins in procedures that are modifications of the techniques developed for vertebrate and squid retinas. These standard procedures may be found in the review of Hubbard et al. (197 1). Vertebrate retinas are large, and photoreceptor membrane is contained within isolated cellular appendages, the outer segments of the receptor cells. Because of these features the outer segment membrane can be easily isolated in substantial amounts prior to rhodopsin extraction. Insect rhabdoms, however, are bound into the complex cellular fabric of small retinas, and consequently the extraction of insect rhodopsins is technically more dimcult. The accessory ommochrome and pteridine pigments densely contained within insect photoreceptors and contiguous cells are particularly bothersome in extraction procedures. They not only raise background extinction, but they may undergo pH or light-induced absorbance changes that can confuse photometric measurements (Bowness and Woken, 1959). Accessory pigments can be removed by repeated preliminary buffer washes of eye homogenates, or in flotation procedures that concentrate rhabdomere membrane prior to detergent extraction (Schwemer et al., 1971; Paulsen and
INSECT VISUAL PIGMENTS
39
Schwemer, 1972). Recently Weber and Zinkler (1974) have devised a method for isolating intact rhabdomeres from Culliphoru eyes. It has generally been found that the absorption spectra of visual pigments are not altered by digitonin extraction (however, see Bruno and Goldsmith, 1974, for a possible example of such an alteration). Detergents other than digitonin are more likely to lead to partial or complete denaturation (Daemen, 1973; Paulsen and Schwemer, 1973) and must be used with caution. Although the work of the past few years has convinced us that insect rhodopsins are orthodox hydrophobic visual pigment proteins bound within membranes, a buffer soluble protein with retinal chromophore was found in one of the earliest attempts to identify an insect photopigment (Goldsmith, 1958a,b). It was extracted from honeybee heads, accounting for half the retinal in the tissue. It behaved like a typical vertebrate rhodopsin, bleaching with the release of retinal to yield a difference spectrum with maximal absorbance change at about 440 nm. Although there is evidence from spectral sensitivity measurements that drone bees have 440 nm receptors, the extract was taken from a population of mostly worker bees in which units sensitive at 535 nm predominate. Thus Goldsmith (1972) and Bruno and Goldsmith (1974) subsequently suggested that the pigment he had extracted was in some respect an artifact. The question of the buffer soluble pigment has been reopened by Pepe et al. (1976). Extracts of bee retinas exposed to tritiated vitamin A were subjected to electrophoresis. The label migrated as a single peak among the faster buffer soluble proteins. The label introduced as vitamin A appeared to be associated with the protein in the form of retinal as in a rhodopsin. However, the retinalprotein complex has not been spectrally characterized, nor has its relationship to the photosensitive pigment extracted by Goldsmith been established. Pepe et ul. (1976) suggest that it might be a precursor of a membrane bound rhodopsin. Alternatively it might function in transport of retinaldehyde. A number of vitamin A binding proteins that act as carriers have been identilied in vertebrates (Heller and Bok, 1976). Microspectrophotometry (MSP), in which a preparation is mounted on the stage of a microscope inserted in the beam of a spectrophotometer, is another procedure that has been particularly useful for the analysis of insect visual pigments. With MSP-applications and limitationshave recently been reviewed by Liebman (1972)-photopigments can be measured in whole eyes, slices of retina, whole cells or in isolated rhabdoms. MSP offers the advantage of measuring photopigments in situ, and the possibility of localizing particular pigments within particular cells. Visual pigments are measured by MSP either in difference spectra, taken between dark and illuminated shples, as will be discussed in some detail below, or by subtracting the baseline spectrum of an adjacent cellular or extracellular region from the rhabdom spectrum (Langer and Thorell, 1966; Hays and Goldsmith, 1969).
RICHARD H. WHITE
40
Various complications, such as the orientation of rhodopsin molecules within their membranes, and optical effects associated with narrow, long or connected rhabdomeres, may alter absorption spectra measured in uiuo in comparison with measurements made on extracts under conditions in which the Beer-Lambert law may be rigorously applied (Snyder and Pask, 1973; Hamdorf and Schwemer, 1975). Such effects are of course important in the actual response of the photoreceptor, but a detailed discussion of them is beyond the scope of the present review. This important area has recently been covered in a review by Goldsmith and Bernard (1974) and by Snyder and Menzel (1975). It is particularly true for insects that the most convincing and informative studies have been those in which photopigments have been measured both in extracts and in situ by MSP. Other techniques that have been used to measure insect photopigments are reflectance photometry in a living butterfly (Stavenga, 1975), and measurement of a fast electrical response from the Drosophila eye (Pak and Lidington, 1974). The latter is similar to the early receptor potential (ERP) that has been recorded from vertebrate (Cone, 1967), squid (Hagins and McGaughy, 1967) and arthropod eyes (Brown et al., 1967; Minke et al., 1973). ERP responses arise directly from charge displacements during photopigment transitions, allowing direct measurements of visual pigments in intact eyes. The technique apparently has not been widely attempted with insects. 3
Rhodopsin and rnetarhodopsin
Squid rhodopsin was the first invertebrate photopigment to be well characterized (Hubbard and St. George, 1958). As in vertebrate rhodopsins, its chromophore is 11-cis retinal. It was found, however, that squid rhodopsin does not bleach. When light is absorbed the initial molecular events are similar to those described above for vertebrate rhodopsins: chromophore isomerization leads to conformational changes in squid opsin. But rather than proceeding to hydrolysis, the reaction culminates with the formation of a colored intermediate, metarhodopsin, with the chromophore in the all-trans configuration still attached to the opsin. Squid metarhodopsin was so named because of its biochemical similarity to the metarhodopsin I bleaching intermediate of vertebrate visual pigments. The difference is that squid metarhodopsin is stable at physiological temperature, whereas vertebrate metarhodopsins decay through additional intermediates that lead finally to hydrolysis. It has subsequently been found that metarhodopsin thermostability is a characteristic of invertebrate photopigments generally, and of insect visual pigments in $articular. Consequently, both the rhodopsin and metarhodopsin states of insect visual pigments are found in illuminated photoreceptors. The
INSECT VISUAL PIGMENTS
41
two thermostable states differ in isomeric form of the chromophore and configuration of the opsin, and generally differ in I,,, and molar extinction as well. A quantum of light absorbed by a molecule of insect rhodopsin converts it through isomerization of the chromophore to metarhodopsin; light absorbed by metarhodopsin re-isomerizes the chromophore regenerating rhodopsin. In continuous light of such spectral composition that it is absorbed by both rhodopsin and metarhodopsin, insect visual pigments flip back and forth between their two stable states, and a photoequilibrium is established. The
Fig. 2. MSP absorption spectra from a larval mosquito (Aedes aegypti) ocellus. (a) Spectra from an ocellus mounted in insect Ringer. Curves 1 and 2 were dark scans whose superposition shows baseline stability. Curve 3 was recorded after an intense flash of yellow light. The change in absorbance resulted from the photoconversion of some rhodopsin to metarhodopsin. (b) Spectra from the ocelli of a living animal. Curves 1 and 2 were measured in the dark; curve 3 was recorded after a yellow flash, and curve 4 was recorded after a subsequent blue flash. The yellow flash converted a portion of the rhodopsin to metarhodopsin; the blue flash reconverted some of the metarhodopsin to rhodopsin. (c) Difference spectrum calculated by subtracting curve 3 from curves 1 and 2 in (a).
concentrations of rhodopsin and metarhodopsin in such a photosteady state will depend on their respective I,,,, absorbance coefficients and quantum efficiencies, and on the spectral quality of the light. Some of the practical consequences of metarhodopsin thermostability for the measurement of insect photopigments are exemplified in a study of the larval mosquito ocellus (Brown and White, 1972). Larvae were grown in darkness in order to ensure that their ocelli would contain only rhodopsin and no metarhodopsin. They were then prepared for MSP in dim red light that hopefully would convert little or none of the rhodopsin to metarhodopsin. Figure 2(a)
42
RICHARD H. WHITE
shows measurements taken from an ocellus mounted in insect Ringer in the microspectrophotometer. Traces 1 and 2 are dark scans; curve 3 was recorded after irradiation with yellow light. As a result of irradiation, absorbance dropped at longer wavelengths and rose at somewhat shorter wavelengths with an isosbestic point at 500 nm. As will be later proven, the decline in absorbance was due to loss of rhodopsin, the increase resulted from the formation of a metarhodopsin photoproduct absorbing at shorter wavelengths, and curve 3 represents the photoequilibrium established by the yellow light. Figure 2(b) shows a similar experiment with an intact living animal. Once again, 1 and 2 are dark scans. Scan 3 followed yellow irradiation as in the preceding experiment. The animal was then illuminated with blue light that is absorbed more strongly by metarhodopsin, so that the photoequilibrium was shifted back somewhat towards rhodopsin (curve 4). The difference spectrum calculated by subtracting curve 3 from curves 1 and 2 in the experiment of Fig. 2(a) is shown in Fig. 2(c). Where the absorption spectra of rhodopsin and metarhodopsin overlap, they mutually subtract in difference spectra. Consequently, in Fig. 2(c) the rhodopsin spectrum to the right is cut off at shorter wavelengths whereas the metarhodopsin spectrum to the left is cut off on its long wavelength side. Measurements such as these show how the visual pigment behaves in situ but they do not accurately characterize either state of the pigment because their spectra interfere. This problem can be overcome, and the spectra of insect rhodopsins determined more accurately under conditions that promote hydrolysis of metarhodopsin. Illumination then results in bleaching, as with vertebrate photopigments. This is often the case when insect visual pigments are brought into solution; metarhodopsins tend to be less stable in digitonin extracts than in uiuo (Schwemer and Paulsen, 1973; Schwemer, personal communication). For MSP, fixation of receptor cells with glutaraldehyde, the histological fixative, generally preserves rhodopsins but renders metarhodopsin labile (Hays and Goldsmith, 1969; Brown and White, 1972). Glutaraldehyde fixation serves the additional purpose of stabilizing cellular structure by cross-linking proteins. It therefore helps to maintain constant spectral baselines and allows the use of other reagents that also promote hydrolysis such as hydroxylamine and potassium borohydride. The former reacts with retinal yielding retinaldehyde oxime, the latter reduces retinal to retinol (or reduces it on site with the same spectral result). To continue with the mosquito ocellus as an example, in the experiment of Fig. 3(a) a dark adapted larva was dissected into glutaraldehyde and mounted in the spectrophotometer in 0.1 M hydroxylamine. Curves 1 and 2 were recorded in the dark, curve 3 after the ocellus had been exposed to intense yellow light for 10 min. As a result, absorbance dropped in the spectral region around 5 15 nm, and rose at 360 nm. The large drop in absorbance at longer
INSECT VISUAL PIGMENTS
43
wavelengths presumably resulted from the bleaching of the main band of mosquito rhodopsin, the increase at 360nm from the formation of retinaldehyde oxime. The difference spectrum is plotted in Fig. 3(b). The rhodopsin 515 nm, is accurate at wavelengths longer than 450 nm. spectrum, A,, Below that its spectrum is obscured by that of the oxime.
Fig. 3. (a) Spectra from a ventral ocellus fixed in a glutaraldehyde and mounted in neutralized hydroxylamine. Curves 1 and 2 were recorded in the dark, curve 3 after the ocellus had been bleached for 10 min with yellow light. The difference spectrum plotted in (b) represents the true absorption spectrum of rhodopsin at wavelengths greater than 450 nm. The inset (c) is a photomicrograph of the ocellus that was measured as it appeared in the microspectrophotometer. For this study a white eye mutant was used that lacked screening pigment. The scalloped rosette at the center of the ocellus is its rhabdom. A 30 pn central area of the rhabdom was measured.
Several lines of evidence indicate that the dif'ference spectrum of Fig. 3(b) truly represents the spectrum of mosquito rhodopsin (R5 15). The most cogent reason for accepting it as a visual pigment is that it matches the spectral sensitivity of mosquito larvae measured from the electroretinogram (ERG) (Seldm et d., 1972). It is also matched well by the theoretical spectrum for a 515 nm rhodopsin calculated from Dartnall's nomogram (Dartnall,1953). Such a theoretical reGnance spectrum for retinal based visual pigments is a useful yardstick for judging the accuracy of measured spectra. The experiment in Fig. 3 also provides evidence that a rhodopsin with retinal chromophore has been measured rather than some other light sensitive substance in the cell. The formation of a photoproduct at about 360nm in the presence of hydroxyl-
44
RICHARD H WHITE
amine indicates that retinal was released by light to form retinaldehyde oxime. Finally, R5 15 was found only in the rhabdom, not in the bodies of the receptor cells. To summarize the general criteria for characterizing an insect rhodopsin: its spectra should reasonably match spectral sensitivity, there should be evidence that it has retinal for chromophore, and it should be localized to a rhabdom and/or shown to be membrane bound by its solubility characteristics in vitro. In measuring the spectra of insect visual pigments one must be aware that metarhodopsin stability can introduce serious distortions. Fig. 2(c) illustrates how difference spectra may fail to correspond closely with absorbance spectra when the spectrum of a photoequilibrium is subtracted from a dark spectrum. A more subtle artifact occurs when one starts with a mixture of rhodopsin and metarhodopsin. In the presence of hydroxylamine or under other conditions that promote bleaching, the difference spectrum will be the sum of the components. The result is a broadened spectrum whose peak lies somewhere between the A,, of the two pigments. The problem is the same when the spectrum of a rhabdomere is compared with a clear area of cytoplasm in MSP. An example is offered by the pioneering microspectrophotometric analysis of Calliphora visual pigment by Langer and Thorell (1966) done before insect metarhodopsin stability was recognized. These first measurements of the presumed absorption maximum of fly rhodopsin varied between 490 nm and 540 nm; most were around 515 nm. The spectra were broader than predicted by the Dartnall nomogram and did not match spectral sensitivity. Calliphora rhodopsin was subsequently found to lie at 490nm, its metarhodopsin at 575 nm (Hamdorf et al., 1973b; Stavenga et al., 1973). It is now clear that the early measurements were of various rhodopsin-metarhodopsin mixtures (see the comments of Langer following the paper of Stavenga et al., 1972). This problem can arise as the result of incomplete dark adaptation or because the MSP scanning beam is bright enough to convert measurable amounts of rhodopsin to metarhodopsin. For the measurements of mosquito rhodopsin described above we were fairly confident that we started with pure rhodopsin, since the larvae were hatched and reared in darkness and dissected in deep red light. In retrospect, however, we cannot be certain that even our best rhodopsin spectra were not contaminated with small amounts of metarhodopsin. It is evident from the foregoing discussion that it is easiest to deal with insect visual pigments whose rhodopsin and metarhodopsin spectra are widely separated. This is true, for instance, of the ultraviolet sensitive pigment of the Neuropteran Ascalaphus macaronius. The main absorption band of Ascalaphus rhodopsin lies at 345 nm, that of its metarhodopsin at 475 nm [Fig. 4(a)l. In this pigment system, light between 440nm and 600nm is absorbed only by M475. M475, like all visual pigments also absorbs light at
INSECT VISUAL PIGMENTS
45
C I
C
pp@@ .
,
.
.
.
& [a]---
Loo Wavelength-nm SQO (00
Fig. 4. Absorption spectra calculated from difference spectra for the rhodopsins and metarhodopsins of three insect species: (a) Ascalaphus macaronius, (b) Culliphora erythrocephala, (c) Deilephila elpenor. The rhodopsin spectra have been normalized to 1.0, and the metarhodopsin spectra have been calculated accordingly. The latter have relatively higher absorbances because their molar extinctions are higher. The curves labeled [R]represent the concentrations of rhodopsin in the photoequilibria that would be established by monochromatic irradiation assuming that the quantum efficiencies of rhodopsin and metarhodopsin are the same. The curves in (b) were recalculated by J. Schwemer for this paper.
TABLE 1
The absorption maxima (A, Order and species ORTHOPTERA
nm) of insect rhodopsins and their metarhodopsin photoproducts
Rhodopsin
Metarhodopsin
Reference Woken and Scheer (1963)
500
Perfplaneta americana
345 440 520 345 440 520 5 10 535
475 (acid) 380 (alkaline) 480 480 480 490 490 490 484 480
Gogala et al. (1970); Schwemer el al. (1971) Schwemer and Paulsen (1973); Hoglund et a/. (1973b)
Aedes aegvpti Drosophila melanogaster
515 480
480 580
Callfphora erythrocephala
490
575
Sarcophaga bullata Musca domestica
490 5 10 512 334
575
Brown and White (1972) Pak and Lidington (1974);Ostroy et a/. (1974) Hamdorfet a/. (1973b);Stavenga et al. (1972) Schwemer (personal communication) Marak et al. (1970) Murietal. (1976)
NEUROPTERA
345
Ascalaphus macaronius LEPIDOPTERA
Deilephila elpenor Manduca sexta" Galleria mellonella Aglais urticae
Brown and Schwemer (personal communication) Goldman et a/. (1975) Stavenga (1975)
DIPTERA
HYMENOPTERA
Apis mellifera (drone)b
415 420
Carlson and Philipson (1972)claimed to have measured four pigments in Manduca, A, 350,450,490,530.Their work has been criticized for poor technique (Goldsmith and Bernard, 1974). It is likely that they were confusing metarhodopsin with rhodopsin . These data are from an abstract; no spectra were published. A water soluble pigment (Amax, 440 nm) with retinal chromophore has also been extracted from honeybee eyes (Goldsmith, 1958a,b). a
3!
0
I
%
?
sz
2
INSECT VISUAL PIGMENTS
47
wavelengths shorter than where its main absorption band lies. Hence ultraviolet light up to 400 nm is absorbed strongly by R345 and to a lesser extent by M475 as well. Irradiation at 345 nm establishes a photoequilibrium consisting of about 1R:3M. Subsequent saturation of the system with blue light shifts the system back to 100 per cent R345 by photoregeneration (Schwemer et al., 1971;Hamdorf et al., 1973b). Figure 4(a) shows the spectra of Ascalaphus R345 and M475 calculated from difference spectra of digitonin extracts. Note that the extinction of metarhodopsin is higher than that of rhodopsin, presumably because the molar extinction of all frans-retinal is higher than that of the 11-cis isomer (Hubbard et al., 1971). From the relative absorbance of each pigment at a particular wavelength, one can calculate the photoequilibrium established by monochromatic irradiation at that wavelength, assuming that the two forms of the pigment have the same quantum efficiencies (Hamdorf et al., 1971a; Hamdorf et al., 1973b). The amount of rhodopsin in such theoretical equilibria is shown in Fig. 4(a). Experimental measurements of photoequilibrium concentrations of R345 and M475 in both extracts and by MSP measurements of Ascalaphus retinas agree well with the calculated photoequilibria. Similar photokinetic studies on the visual pigments of Lepidoptera and Diptera [Fig. 4(b), (c)] have given similar results: the relative concentrations of rhodopsin and metarhodopsin in a brightly illuminated photoreceptor appear to depend only upon the spectra and molar extinctions of the two states of the photopigment, and on the spectral quality of the light (Hamdorf et al., 1973b; Hoglund ef al., 1973a). The absorbance maxima of most insect metarhodopsins lie around 480 nm490 nm (Table 1). However, in certain pigment systems of the higher flies the metarhodopsins peak around 575nm, while the rhodopsins lie at about 490nm. Some cephalopods have similar pigment systems with long wavelength metarhodopsins (Brown and Brown, 1958; Schwemer, 1969; Hamdorf et al., 1972). Before leaving this general discussion of insect rhodopsins, I should mention that the molecular weight of Ascalaphus opsin has been measured by Paulsen and Schwemer (1973). From its relative mobility in sodium dodecylsulfate (SDS)polyacrylamide gel electrophoresis they estimated its molecular weight to be 35 OOO k 1800. This is similar to the molecular weights of frog and cattle opsins determined by the same method, but somewhat less than the molecular weights of cephalopod rhodopsins (Hagins, 1973; Paulsen and Schwemer, 1973). 4
Chromophore and photochemistry
The chromophore of insect visual pigments has been identified from both direct and indirect evidence as retinaldehyde (retinal). In the first successful bio-
48
RICHARD H WHITE
chemical study of insect vision (Goldsmith, 1958a,b), substantial amounts of retinal, identified by the standard reaction with antimony trichloride (CarrPrice procedure: Hubbard et al., 1971) were found in the heads of dark adapted honey bees. None was detected in their bodies. Retinal has subsequently been measured by the Carr-Price reaction in extracts from the heads, but not the bodies, of a number of insects including Orthoptera, Odonata, Lepidoptera, Coleoptera, Diptera and Hymenoptera (Wolken et al., 1960; Briggs, 1961; Wolken and Scheer, 1963). As far as we know, insects like other animals only obtain retinol (vitamin A) and retinal from their food, either directly or from other carotenoids. Retinol is not a vitamin with an essential systemic function for insects as it is for vertebrates. Consequently, insects can be grown, for generations if desired on carotenoid-free diets. Carotenoid deficiency produced ultrastructural changes in the photoreceptor cells of the mosquito (White and Jolie, 1966; Brammer and White, 1969). These morphological changes were accompanied by loss of visual sensitivity. The visual threshold of houseflies grown for several generations on a sterile defined diet without carotenoid rose at least four log units (Goldsmith et al., 1964; Goldsmith and Fernandez, 1966). Sensitivity returned to normal when they were fed fi-carotene. Thus even before insect visual pigments had been well characterized there was evidence pointing to retinal as the chromophore. More direct evidence was provided by Schwemer et al. (1971) and Brown and White (1972), when Ascalaphus and mosquito rhodopsins were shown to react with hydroxylamine and potassium borohydride. Only the chromophore of Ascalaphus rhodopsin (R345) has been well characterized. Retinal was measured in extracts of Ascalaphus eyes by the antimony trichloride method. It was bound exclusively to insoluble debris, presumably membrane fragments (Paulsen and Schwemer, 1972). Both the alltrans and the 11-cis isomers of retinal were identified by thin layer chromatography. Other isomers were not present in measurable amounts. The configuration of the chromophores of vertebrate and squid visual pigments have been identified by denaturing rhodopsin in the darkness, in order to release retinal while avoiding the isomerizing effect of light (Hubbard and Kropf, 1958). In a similar procedure, the chromophore of R345 was released after denaturation with Ag2+. Opsin prepared from bleached cattle retinas provides a sensitive and specific assay for the 11-cis isomer of retinal, since only this isomer promotes the regeneration of cattle rhodopsin, A,, 500 nm (Hubbard et al., 1971). When the chromophore released from denatured R345 was mixed with purified cattle opsin, cattle rhodopsin was regenerated. Therefore the chromophore of R345 is 11-cis retinal, like that of all other rhodopsins. Does light isomerize the chromophore to all-trans retinal as in other
INSECT VISUAL PIGMENTS
49
rhodopsins? Denaturation of a photoequilibrium mixture consisting of onethird R345 and two-thirds M475 released only a third as much 11-cisretinal as from 100 per cent R345. Since thin layer chromatography had demonstrated that the only retinal isomers present in the eye were 1 1 4 s and all-trans, the chromophore of M475 must be all-trans retinal. Hence the basic photochemistry of the Ascalaphus photopigment is similar to that of the visual pigments of other animals. Metarhodopsins, both vertebrate and invertebrate, typically act as pH indicators. The spectrum of Ascalaphus metarhodopsin also shifts with pH, changing from acid M475 to alkaline M380 with a pK of 9.2 (Schwemer et al., 1971; Hamdorf et al., 1973b). Obviously, M475 predominates at physiological pH; this was shown by MSP of Ascalaphus retinas. The pH sensitivity of metarhodopsins has been attributed to the reversible protonation of the Schiff base link between retinal and opsin. These observations indicate that the chromophore of Ascalaphus visual pigment is also bound as a Schiff base, at least in metarhodopsin. This is an important conclusion, since unlike the generality of rhodopsins the spectrum of retinal is shifted to shorter wavelengths when it is associated with its opsin. Speculations that this hypsochromatic spectral shift might result from a fundamentally different association of the chromophore and protein (cf. Goldsmith, 1972) now seem unlikely. Presumably the resonant structure of the chromophore is shortened by way of interactions with its opsin that are in addition to the covalent SchifT base link. This most interesting pigment system needs to be studied with the contemporary biophysical techniques, such as resonance Raman spectroscopy, that are now being applied to vertebrate visual pigments (Ebrey and Honig, 1975). After absorbing light visual pigments pass through a sequence of intermediate states. Interest in these intermediates is grounded in the notion that one of the transitions must lead to transduction, that is, to the generation of an electrical signal across the photoreceptor membrane. The intermediates of vertebrate and squid rhodopsins have been spectrally defined in low temperature experiments in which their rates of decay are slowed (Ebrey and Honig, 1975). The results of similar experiments with the visual pigments of Ascalaphus (Hamdorf et al., 1973b) are summarized in Fig. 5. At physiological temperature only the two thermostable states of the pigment, R345 and M475, can be measured with conventional slow speed spectrophotometry. Irradiation of R345 at -50° C produced a new intermediate, stable at that temperature, peaking at 375 nm. This was designated Lumirhodopsin (L375) in accordance with the terminology adopted for vertebrate visual pigments. When the temperature was subsequently raised to -15O C in darkness, L375 decayed to M475. These observations can be interpreted as follows. At -5OO C cis to
50
RICHARD H WHITE
trans chromophore isomerization occurs, but changes in opsin configuration presumably are prevented. The initial absorbance shift of 30nm to longer wavelengths is thought to reflect the change in chromophore conformation and the differing interactions of its cis and fruns forms with the unaltered protein. Raising the temperature then allows the protein to change conformation with an additional 100 nm shift. In other words, these data support the idea that the rhodopsin to metarhodopsin reaction includes conformational changes of opsin
1
Rhodopsin 345 nm ( 1 1 4 s )
Lumirhodopsin 375 nm (all-trans)
Alkaline metarhodopsin 380 (dl-trans)
pK 9.2
f
Acid metarhodopsin 475 nm (al-trans)
Acid rnetarhodopsin 460 nm ( 1 1 4 s ) -15°C
Fig. 5. Intermediates of the ultraviolet sensitive visual pigment of Ascalaphus macaronius. Wavy lines represent photoreactions, straight lines, dark reactions.
provoked by the initial photoisomerization of the chromophore. In the opposite experiment, irradiation of M475 at -50 OC produced another intermediate at about 460 nm. When warmed in darkness that intermediate regenerated to R345. Since rhodopsin regenerates in darkness from the new intermediate, the chromophore of the intermediate must be 11-cis retinal. Moreover, the opsin presumably retains its metarhodopsin configuration in the new intermediate. Hence it has been designated 11-cis metarhodopsin. The large spectral shift to shorter wavelength that occurs when M460 regenerates to R345 is presumably due to thermal changes in the opsin that can take place after the chromophore has assumed the 1 1-cis configuration, and reflects the additional interactions between protein and chromophore that then becomes possible.
INSECT VISUAL PIGMENTS
51
The spectra and rates of decay of intermediates have been important data for understanding the molecular events that follow upon the absorption of light by rhodopsin. Photoreversal reactions, that is, photoregeneration of rhodopsin from intermediates via frans to cis isomerizations have also been studied in efforts to characterize the interactions of opsin and chromophore (Williams ef al., 1973). Ascalaphus 11-cis metarhodopsin is a remarkable intermediate of photoregeneration. Such metarhodopsins with cis chromophores have been only poorly characterized in vertebrate systems (see comments of Williams after the paper of Hamdorf et al., 1973). One further aspect of rhodopsin photochemistry that deserves comment: energy transfer from opsin to chromophore has been demonstrated in vertebrate rhodopsins. That is, light absorbed by the y-band at 280 nm can initiate bleaching (Kropf, 1967). When measurement of housefly spectral sensitivity was extended into the middle ultraviolet, a shoulder was found at 280nm (Goldsmith and Fernandez, 1968). The authors have suggested that this response at very short wavelengths is due to energy transfer from the opsin to the chromophore.
5
Regeneration in insect visual systems
The spectral sensitivity curves of insects and other invertebrates match rhodopsin but not metarhodopsin spectra (Hamdorf et al., 1973b), that is, only light absorbed by rhodopsin is transduced into a visual response. Therefore, there must be regenerative mechanisms for maintaining adequate concentrations of rhodopsin in illuminated eyes. From the preceding discussion it is clear that photoregeneration from stable metarhodopsin holds rhodopsin at a constant level in photoequilibria. Two features of insect photopigment systems tend to favor high rhodopsin levels in the photoequilibriaestablished by normal polychromatic daylight. Insect metarhodopsins generally have higher absorbance than their rhodopsins and most metarhodopsins peak in the blue-green near the emission maximum of daylight (Hamdorf et al., 1973b). For example, the daylight photoequilibrium of the ultraviolet sensitive system of Ascalaphus is 90 per cent rhodopsin (Hamdorf and Gogala, 1973; Hamdorfet al., 1971a). In other naturally illuminated insect photoreceptors, rhodopsin must be maintained well above the 50 per cent level (Hoglund et al., 1973a,b; Hamdorf and Schwemer, 1975). The predominant photopigment systems of the cyclorraphous flies with their blue-green sensitive rhodopsins and red absorbing metarhodopsins would seem perversely adapted to favor low rhodopsin levels. Nevertheless this is an exception that proves the rule. These flies have red accessory pigments that
52
RICHARD
H
WHITE
absorb the shorter visible wavelengths but leak red light (Goldsmith, 1965; Langer, 1967; Stavenga el al., 1973). Under normal conditions, their rhabdomeres are bathed in scattered red light absorbed selectively by metarhodopsin, and photoequilibria are pushed well over toward rhodopsin. Rhodopsin concentrations in photoequilibria as a function of wavelength are shown in Fig. 4. In vertebrates, regeneration is accomplished by an enzymically mediated visual cycle. All-trans retinal released from rhodopsin by bleaching is reduced to retinol, enzymatically isomerized, and reconverted to 1 1-cis retinal. The chromophore can then recombine with opsin to regenerate rhodopsin (Bridges, 1976). Since regeneration via the visual cycle does not depend on illumination-it occurs in both light and darkness-it is spoken of as “dark regeneration”. Does any sort of dark regeneration take place in insects? None was found in the mosquito ocellus where rhodopsin-metarhodopsin ratios set up by illumination did not change after more than an hour of subsequent darkness (Brown and White, 1972). Complete regeneration of rhodopsin in the moth Galleria required several days of darkness (Goldman ef al., 1975). In Drosophila the time constant of dark regeneration is about 6 h (Pak and Lidington, 1974), whereas in Calliphora it is 25 min (Stavenga et al., 1973). Stavenga (1975) has also reported a dark regeneration half time between 15 and 45 min in the butterfly Aglais. In no instance do we know the mechanism of dark regeneration, and the wide variation in the rates of regeneration among the few insects investigated suggests that there may be a variety of mechanisms. The only direct studies on the visual cycle in insects are contradictory. Taking as precedent the vertebrate visual cycle of bleaching and enzymatic regeneration, Goldsmith and Warner (1964) sought and found a retinal-retinol oxidation-reduction system in the head of the bee. As in the vertebrate retina, retinal predominated in darkness, retinol in light. They also found retinal reductase activity in the heads of bees. On the other hand, Paulsen and Schwemer (1972) found retinal in the eyes of Ascalaphus but no retinol in either light or dark adaptation. In light of our present understanding of insect visual pigments, their results are not surprising. There would be no reason to expect a light4ark retinal-retinol cycle in systems with stable metarhodopsins that do not bleach. I have mentioned the puzzling water soluble protein-retinal complexes that have been extracted from bees. The visual cycle characterized by Goldsmith and Warner is another interesting feature of the bee eye that demands further study. There are several obvious ways by which dark regeneration might be accomplished in insects. Enzymatic trans to cis isomerization of the free chromophore occurs in vertebrates, apparently within the outer segments (Bridges, 1976). An isomerase could account for those instances of relatively
INSECT VISUAL PIGMENTS
53
rapid dark regeneration in insects. It presumably would have to act upon retinal still attached to the metarhodopsin in the photoreceptor membrane. Very slow regeneration, as in Galleria (Goldman et al., 1975) might reflect turnover of visual pigment, that is, complete renewal of rhodopsin via protein synthesis and renewal of photoreceptor membrane. In this regard there is indirect evidence that some insect photoreceptor membranes may have high rates of turnover, rates sufficient to accomplish dark regeneration in a matter of hours (White and Lord, 1975). From the standpoint of physiological adaptation it is hard to see why insects-at least diurnal insects, in which some of the highest regeneration rates have been measured-should have any need to supplement photoregeneration with special mechanisms of dark regeneration. At best, a period of regeneration in darkness following illumination can increase absolute sensitivity by less than a factor of 2 because rhodopsin is already maintained at levels above 50 per cent in photoequilibria (Hoglund et al., 1973a,b). 6
Insect color vision and ultraviolet sensitivity
The existence in insect eyes of units sensitive to different regions of the spectrum is well known from electrophysiological measurements (Autrum and v. Zwehl, 1964; Burkhardt, 1964). Behavioral experiments, most notably with bees and Lepidoptera, have demonstrated true color vision based upon trichromatic perceptual systems (v. Frisch, 1965; Knoll, 1924). The only insects with demonstrated color vision whose visual pigments have been well characterized are the sphinx months Deilephila elpenor and Manduca sexta (Hamdorf et al., 1972a; 1973a; Schwemer and Paulsen, 1973; Schwemer and Brown, personal communication). Their visual systems are similar, and I will discuss only that of Deilephila. Three photopigments have been measured in intact retinas by MSP and in digitonin extracts. Their respective absorption maxima lie at about 345 nm (ultraviolet), 440 nm (blue) and 520 nm (green). Deilephila eyes contain four or five times as much R520 as R345 and R440, assuming that the molar extinctions of the three pigments are similar. The metarhodopsins of all three peak in the vicinity of 480 nm. Components of the Deilephila electroretinogram (ERG) corresponding to each of the three photopigments can be selectively light adapted, indicating that the pigments are localized in separate receptor cells (Hoglund et al., 1973a,b). Thus the essential basis for color perception is present: distinct visual pigments differentially sensitive across the spectrum housed in physiologically separate receptors. Hoglund et al. (1973a,b) have suggested that photoregeneration is important for maintaining this system of color discrimination in balance. Ambient sky light should tend to maintain the three pigments at similar relative concentrations in their respective photoequilibria since their metarhodopsins all
54
RICHARD H WHITE
have similar spectra centered on the wavelength of maximal daylight emission. In this way the relative sensitivities of the three types of receptor might be kept constant and color discrimination might simply be based on the comparison of the response amplitudes of the three receptors. The emission spectrum of the sun extends well into the ultraviolet, to about 300 nm, with a minor peak around 350 nm (Hamdorf el al., 1971a). It has been often pointed out that insects visually exploit these shorter wavelengths whereas vertebrates do not (Goldsmith and Bernard, 1974). Most vertebrates are blind below 400 nm because their lenses are yellow; they function as cut-off filters (Zigman, 1971). When their lenses are removed, vertebrates can see ultraviolet light via the P-band short wavelength absorbance of their visual pigments. I have discussed above the ultraviolet sensitive rhodopsin of Ascalaphus in which the main band is shifted to about 345 nm. It does not follow that all responses of insects to ultraviolet light must depend upon such pigments. As long as their lenses are transparent to ultraviolet light it will be absorbed by the minor bands of longer wavelength pigments. And if a visual pigment whose main band lies in the visible is confined within a very narrow rhabdomere, on the order 1 pm, its spectrum may be distorted so as to relatively increase its extinction at shorter wavelengths (Snyder and Miller, 1972; Snyder and Pask, 1973). There is a problem that must be particularly associated with ultraviolet sensitivity, one that has not yet been experimentally explored in insects: the pathological effect of light. It has been recognized in recent years that visible light can harm vertebrate photoreceptors even at moderate intensities. It is not certain what causes the damage, but photo-oxidation of membrane lipids may be one of a number of photochemical mechanisms. The antioxidant tocopherol (vitamin E) is a notable constituent of vertebrate photoreceptor membranes (Daemen, 1973). The action spectrum of retinal damage in vertebrates rises sharply at the shorter wavelength, higher energy part of the spectrum that for vertebrates ends at 400 nm (Ham et al., 1976). Thus insects, with their ultraviolet transparent eyes, would seem particularly vulnerable to photochemical pathology. While insects see at shorter wavelengths than do vertebrates, they have not extended their visual sensitivity as far into the red. Those vertebrate photopigments absorbing at the longest wavelengths (with A,, as far out as 620 nm) are porphyropsins whose chromophore is 3-dehydroretinal. This form of the chromophore, with its enhanced bathochromatic shift, is unknown among invertebrates. There are vertebrate rhodopsins whose spectra extend well into the red, with A,, to 580 nm. Spectral sensitivity measurements suggest that some butterflies may have red receptors peaking around 600nm (Swihart and Gordon, 1971). However, no attempt has been made to measure directly a visual pigment that might lie behind this long wavelength sensitivity.
INSECT VISUAL PIGMENTS
7
55
The problem of the visual pigments of the higher flies
The visual systems of the higher flies belonging to the cyclorraphous families are under particularly intense investigation, and it is worthwhile considering some of the unresolved problems having to do with their visual pigments. Each ommatidium of the fly compound eye is made up of 8 rhabdomeres, with those designated 1-6 standing in isolation in an outer ring around rhabdomeres 7 and 8 at the center. Rhabdomere 7 lies distal to 8 along the optical axis. Spectral sensitivity measurements on various fly species have shown large response maxima in the blue-green and ultraviolet, and in some instances lesser peaks in the blue, yellow-green and red (Autrum and Burkhardt, 1961; Burkhardt, 1962; Burkhardt and de la Motte, 1972; McCann and Arnett, 1972; Horridge and Mimura, 1975; Horridge et al., 1975; Rosner, 1975; Meffert and Smola, 1976). Rhabdomeres 1 - 6 have been identilled by intracellular electrophysiology as blue-green receptors, A,, 490 nm, in Calliphora (McCann and Arnett, 1972; Horridge and Mimura, 1975, Rosner, 1975). The sensitivity of the peripheral rhabdomeres is clearly due to R490, whose main band has been well characterized (Langer and Thorell, 1966; Hamdorf et al., 1972b; Stavenga et al., 1973). Receptors 1-6 also respond strongly to ultraviolet light (Burkhardt, 1962; McCann and Amett, 1972; Horridge and Mimura, 1975; Rosner, 1975). In fact they are as sensitive at 350 nm as at 490 nm. Although rhodopsin spectra extend into the ultraviolet, the extinction of the secondary short wavelength maximum @-band) of all visual pigments that have been adequately characterized-mainly vertebrate rhodopsins-is much lower than that of the main absorption band (a-band) in the visible (Fig. 1). Recently Horridge and Mimura (1975) and Rosner (1975) have sought to explain the peculiar action spectra of cells 1-6 by suggesting that their rhabdomere membranes carry two distinct photopigments, an ultraviolet sensitive rhodopsin as well as R490. Rosner argued from the adaptation characteristics of these cells. Horridge and Mimura found that the ultraviolet and blue-green responses are differentially sensitive to the plane of polarized light. The possibility of two rhodopsins in the same receptor is intriguing because so far no photoreceptor cell in any animal has been shown to synthesize more than one species of opsin. An alternative explanation of the high ultraviolet sensitivity of rhabdomeres 1-6 has been offered by Snyder and Miller (1972) and Snyder and Pask (1973). The rhabdomeres of flies are dielectric waveguides, whose geometric and optical properties should affect the absorption spectra of their visual pigments. Theoretically, confining a rhodopsin within a rhabdomere of small diameter would increase its ultraviolet absorbance relative to that of the a band at longer wavelengths. Finally, it should be pointed out that separate receptor cells with different
56
RICHARD H. WHITE
sensitivities may be electrically coupled. In that event the location and absorption spectra of the resident photopigments cannot be deduced with certainty from action spectra, even those measured by intracellular recording (Shaw, 1969). In this regard, MSP absorption spectra of single peripheral rhabdomeres in Cafliphora show only a relatively small ultraviolet peak and give no suggestion of two pigments (Langer and Thorell, 1966). Absorption and action spectra do not match in the near ultraviolet. The high sensitivity of rhabdomeres 1-6 at the shorter wavelengths remains a fascinating puzzle. A new approach to identifying the spectral responses of particular cells has been taken by Harris et a f . (1976). They have examined Drosophifa mutants in which particular rhabdomeres either do not develop or degenerate. The spectral sensitivities and modes of adaptation of these mutant flies suggest that cells 1-6 are combined blue-green ultraviolet receptors, that cell 7 is an ultraviolet receptor and that cell 8 is a blue receptor. MSP measurements of Ostroy et a f . (1974) confirm that, as in Cafliphora,the peripheral rhabdomeres carry a bluegreen sensitive rhodopsin. Flies lose their blue-green sensitivity with the genetic elimination of rhabdomeres 1-6. They retain high ultraviolet sensitivity and relatively lower blue sensitivity. The response to ultraviolet light can be reduced by short wavelength adaptation, a result consistent with the assignment of high ultraviolet sensitivity to only one of the remaining central cells 7 or 8. Elimination of all rhabdomeres except number 8 renders the flies ultraviolet insensitive, leaving only a blue response that cannot be altered by adaptation. Hence, blue sensitivity must be associated with cell 8, ultraviolet sensitivity with cell 7. An ultraviolet sensitive pigment, Amax 370 nm was extracted by Harris et a f . (1976) from mutant Drosophifa retaining only rhabdomeres 7 and 8. Its presumed metarhodopsin lay at 470nm. However, as the pigment was measured in crude aqueous suspensions, further confirmation that it is truly a rhodopsin is desirable. Harris et af. (1976) were unable to isolate a blue sensitive rhodopsin from cell 8. It may have been measured by Langer and Thorell (1966) in an early study with MSP, but it has not been well characterized. Snyder and Pask ( 1973) suggested that the central rhabdomeres of fly retinulae might contain the same rhodopsin as that of rhabdomeres 1-6. They argued that the central rhabdomeres are narrow enough to suppress the a-band of R490 and shift its maximum into the blue, while enhancing the pband at 350nm. The results of Harris et al. (1976) do not support that suggestion. In Drosophifa the evidence suggests that there are separate ultraviolet and blue sensitive photopigments. However, the predicted effects of the physical properties of fly rhabdomeres upon the absorption spectra of their rhodopsins need to be experimentally tested by comparing their in vivo and in vitro spectra. The full characterization of the rhodopsins that reside in the small central rhabdomeres is a formidable challenge.
INSECT VISUAL PIGMENTS
8
57
Transduction and adaptation
The control of sensitivity in photoreceptor systems is a classical problem of vision physiology. Several factors are involved in setting sensitivity including the amount of visual pigment that is present, pupil mechanisms-such as migration of screening pigments in insect eyes-that regulate the amount of light entering a receptor, and mechanisms that have been designated electrical or neural adaptation. I will be concerned here with the relationship of rhodopsin concentration to sensitivity, but the particular complication of electrical adaptation needs to be dealt with first. Sensitivity changes of this sort reflect electrical modifications of membranes or of their ionic environments such that background illumination decreases the size of the electrical response evoked by a given stimulus. Electrical adaptation can have considerable effect at very low light levels. Vertebrate visual systems show several log units of change in threshold, that is, in the amount of light required to provoke a given receptor response, even at levels of background illumination that bleach only a few percent of the visual pigment (Weinstein et al., 1967). Dark recovery from such electrical adaptation takes place rapidly and does not depend upon pigment regeneration. When rhodopsin regeneration is prevented threshold drops quickly in darkness but only to a level that reflects the amount of photopigment remaining. Thus the relationship between rhodopsin concentration and the absolute sensitivity of a receptor is manifest after a short period of dark recovery that eliminates the larger effects of electrical adaptation, if during that period the amount of visual pigment remains constant. One might expect-indeed this was the early hypothesis-that absolute sensitivity would then be simply proportional to the amount of rhodopsin, to the quantum catching capacity of the receptor. This proved not to be the case in vertebrates, however; the relationship between rhodopsin concentration and sensitivity is logarithmic. In the rat, for instance, when 17 per cent of the rhodopsin is bleached, threshold increases by a factor of 10, a bleach of 34 per cent elevates threshold a hundredfold (Weinstein et al., 1967). This nonlinear relationship has not been explained. Similar experiments relating absolute sensitivity to rhodopsin concentration became possible in insects when their visual pigments could be reliably measured. The most appropriate insect photoreceptor systems for study are those with the greatest spectral separation between rhodopsin and metarhodopsin, in which the rhodopsin concentration can be widely varied. In such systems, photoequilibria containing different amounts of rhodopsin can be established by monochromatic light adaptation at different wavelengths, or different amounts of rhodopsin can be converted to metarhodopsin by varying either the intensity or duration of illumination. The latter procedure is comparable to the bleaching away of a vertebrate visual pigment except that a "
58
RICHARD H. WHITE
minimal rhodopsin concentration is reached when the system has been driven to photoequilibrium. Since dark regeneration is negligible in insect eyes over short periods, rhodopsin concentration does not change over the few minutes of darkness required for recovery from electrical adaptation. During that short period of dark adaptation, sensitivity rises to a plateau whose level depends upon the rhodopsin concentration in the photoreceptor. The relationship between sensitivity and amount of rhodopsin has now been determined in several insects : Ascalaphus, Deilephila, Manduca and Calliphora (Hamdorf and Rosner, 1973; Hamdorf and Schwemer, 1975; Rosner, 1975). In contrast to the logarithmic relationship found in vertebrates, dark adapted sensitivity has been found in these insects to be a simple linear function of rhodopsin concentration. This has been particularly wellestablished for the blue-green sensitive system of Calliphoru, in which the concentration of R490 was varied from 100 per cent to 30 per cent, and receptor potentials were measured by both extracellular and intracellular methods. The most straightforward interpretation of the linear relationship is that each rhodopsin molecule activated by absorbing a quantum of light contributes equally to the generation of the receptor potential. An extensive theoretical discussion of visual pigment-absolute sensitivity relationships can be found in the article by Hamdorf and Schwemer (1975). In the experiments summarized above, metarhodopsin concentration necessarily varied in reciprocal relationship with the amount of rhodopsin. Hence the question arises of a possible role for metarhodopsin in setting sensitivity. With that possibility in mind, Razmjoo and Hamdorf (1976) studied carotenoid deficient Calliphora with reduced amounts of visual pigments in their photoreceptor membranes. The aim of these experiments was to assess the effect on sensitivity of lowered rhodopsin concentrations independent of altered rhodopsin-metarhodopsin ratios. Once again, an approximately linear proportionality between sensitivity and amount of rhodopsin was found. In another experiment, both carotenoid-rich and carotenoid-depleted flies were irradiated with blue light in order to set up similar photoequilibria with low proportions of rhodopsin. Although the relative change in rhodopsin concentration was the same, the carotenoid-rich flies suffered a relatively greater loss of sensitivity. The authors suggest that in photoreceptor membranes where visual pigment molecules are more densely packed, as in the carotenoid-rich Calliphora, metarhodopsin molecules may have some inhibitory influence on the membrane’s response to light activated rhodopsin. The preceeding discussion has introduced the fundamental question of how the absorption of light by a visual pigment provokes a potential change in the receptor cell membrane. In insects as in other invertebrates, the receptor membrane responds with depolarization, whereas vertebrate receptors hyper-
INSECT VISUAL PIGMENTS
59
polarize. Beyond that almost nothing is known about visual transduction specifically in insects. Compared with the visual cells of various vertebrates and other invertebrates, e.g. Limulus (Lisman and Brown, 1975), insect receptors have not seemed particularly favorable as experimental systems. In general, theories of visual transduction and adaptation involve changes in sodium conductance regulated by calcium ions or by cyclic nucleotides at sodium permeability sites (Ebrey and Honig, 1975). The hypothesis of Razmjoo and Hamdorf (1976) regarding an influence of metarhodopsin on transduction calls to mind a peculiar response of Drosophila photoreceptors to intense blue light that sets up a photoequilibrium favoring metarhodopsin. The receptors are thrown into a state of prolonged depolarization, so that they cannot respond to subsequent stimuli even though rhodopsin is still present. Although the cells recover only very slowly in darkness, they can be abruptly restored to normal function by red light that pushes the photoequilibrium back toward rhodopsin (Cosens and Briscoe, 1972; Minke et af., 1975; Stark and Zitzman, 1976). Thus accumulation of metarhodopsin in the membrane seems to be associated with its prolonged depolarization. However, the effect is apparently not tied directly to the presence of metarhodopsin, nor is the proportion of metarhodopsin to rhodopsin the significant parameter. Prolonged depolarization does not occur in Drosophila receptors whose total amount of visual pigment has been reduced by carotenoid deficiency (Stark and Zitzman, 1976). Stark and Zitzman suggest that there are a limited number of membrane sites, distinct from photopigment molecules, that control depolarization, that these sites are activated by photoconversion of rhodopsin to metarhodopsin, perhaps by way of a transmitter, and that such action persists for some time after stimulation. According to their hypothesis, prolonged depolarization occurs when these limited sites are fully activated by massive conversion of rhodopsin to metarhodopsin. Carotenoid deprivation is seen as lowering the ratio of visual pigment molecules to depolarization sites, so that the latter cannot be saturated. Drosophifuis also unique in offering mutants with impaired transduction for analysis. A group of such mutants on the x-chromosome, designated norp-A, is characterized by reduced or absent photoreceptor potentials, and differences in eye protein composition (Ostroy and Pak, 1974). The blue-green sensitive rhodopsin, R480, in one of these phototransduction mutants, norpAP'*, has been measured in extracts and by MSP (Ostroy et al., 1974). The spectra of R480 and its photoproduct, M580, were found to be similar in mutant and normal flies, but the mutant flies contained only a third as much photopigment. The reduced amount of visual pigment cannot in itself be advanced as an adequate explanation of the mutant phenotype, since the receptor
60
RICHARD H. WHITE
potential is much more severely affected; it is either completely absent or nearly so. The primary cause of blocked transduction in these mutant flies remains unknown.
9
Insect photoreceptor membranes
The photoreceptor membranes of vertebrates are large flat sheets (outer segment disks), while those of insect and most other invertebrates are rolled into tight cylindrical microvilli ranging around 50 nm in diameter. Vertebrate rod outer segment disks are very fluid membranes, with the viscosity of a light oil, owing to their high content of unsaturated phospholipids and low content of cholesterol. The predominantly hydrophobic rhodopsin molecules are embedded in the membrane, but are oriented by hydrophilic interactions with the surrounding medium so that their chromophores are held parallel to the membrane plane. Otherwise, vertebrate rhodopsin molecules are free to rotate and move laterally within the plane of the fluid membrane (Ebrey and Honig, 1975).
Are invertebrate rhodopsins similarly mobile within their microviUus membranes? The question arises with particular pertinence when we consider the sensitivity of many insects and other arthropods to polarized light (Waterman, 1975). Perception of the plane of polarization is possible in the first place because retinal is a highly dichroic linear chromophore (Fig. 1). Vertebrates are generally insensitive to polarized light because the chromophores of their mobile rhodopsins are randomly oriented in the plane of the disk membrane perpendicular to the axis of incoming light. If such a flat membrane were rolled into a cylinder, however, it would be more sensitive to light polarized with the electric vector parallel to the long axis of the cylinder than to light polarized perpendicular to it (Hays and Goldsmith, 1969; Snyder and Laughlin, 1975; Laughlin et al., 1975). Since rhabdomeric photoreceptors are composed of cylindrical microvilli generally oriented perpendicular to the optical axis, they are inherently sensitive to polarized light. Invertebrates with rhabdomeric eyes are preadapted for the evolution of true polarized light perception in which the plane of polarization is behaviorally distinguishable from intensity variation (Waterman, 1975). If photopigments are organized in arthropod microvilli as they are in vertebrate disks, with their chromophores parallel to the plane of the membrane but otherwise oriented at random and free to rotate, the dichroism arising from microvillus geometry could provide dichroic absorbance ratios no greater than 2 : 1, in fact rather lower (Snyder and Laughlin, 1975). But much higher polarization sensitivities have been measured in arthropod eyes, e.g., 9 :1 in the bee (Menzel and Snyder, 1974). Hence it has been inferred that rhodopsin must
INSECT VISUAL PIGMENTS
61
be more rigidly oriented in arthropod microvilli-at least in animals with high polarization sensitivity-than in vertebrate disks. It has been argued that even if photopigments are mobile within the microvillus membrane they could be oriented by preferential axial streaming within the tightly curved phospholipid bilayer (Snyder and Laughlin, 1975). Direct measurements of chromophore orientation in rhabdomeres have been dimcult. The best data are from MSP measurements of crustacean rhabdoms. The earliest measurements recorded absorbance ratios no higher than 2 :1 with highest absorbance parallel to the axes of the microvilli (Waterman et al., 1969). These data seemed compatible with random photopigment orientation. More recently the ratio has been raised to 3 : l by better measurements (Goldsmith, 1975). These more recent data appear to demand nonrandom orientation. Another fruithl approach to the question of photoreceptor membrane organization is to measure rhodopsin mobility. Goldsmith and Wehner (1975); Wehner and Goldsmith (1975) have reported that the rhodopsin in crustacean rhabdomeres neither rotates nor moves along the microvillus axis. The data remain open to question, however, because formaldehyde was used to stabilize the preparations. Although this reagent does not impede the movement of rhodopsin in vertebrate disks, the possibility remains that it might crosslink rhodopsin in rhabdomeric membranes and so hinder normal mobility. Measurements of rhodopsin orientation with the same technical quality have not been made on insect rhabdomeres. Langer (1965) measured somewhat higher rhodopsin absorbances parallel to the axes of the rhabdomere microvilli in Culliphoru. Kirschfleld and Snyder (1975) have presented data suggesting that the chromophores may be oriented perpendicular to the microvillous axes in rhabdomeres 7 or 8 of the fly Muscu. Although rhodopsin mobility has not been directly measured in insect photoreceptors, some relevant inferences have been made from the lipid composition of their membranes. The proportions of phospholipids in Deilephilu and Asculuphus retinas were found to be similar to those in vertebrate disks (Zinkler, 1975). However, fatty acid components were less polyunsaturated, and there was twice as much cholesterol. It should be pointed out, however, that extracts of whole insect retinas are here being compared with purified vertebrate photoreceptor membrane. In any case, the lipid composition of these insect retinas is similar to that of squid (Benoken et ul., 1975) and Limulus (Mason et ul., 1973) rhabdoms. It has been argued from these data that the photoreceptor membranes of insects and other invertebrates are more viscous than vertebrate outer segment membranes. However, the relationship of lipid saturation and cholesterol content to membrane fluidity is not certain. There can be no substitute for direct measurements of the mobility and orientation of rhodopsin molecules and of the mobility of artificial probes introduced into
62
RICHARD H. WHITE
rhabdomere membranes. To summarize, theory supported by some evidence suggests that in comparison with vertebrates, arthropod photoreceptor membranes are less fluid, and that photopigments are less mobile and more highly oriented. The structure of rhabdomere membranes has been examined directly by freeze-fracture electron microscopy. In this procedure membrane bilayers split and membrane proteins are revealed as bumps upon the separated phospholipid leaflets. Densely packed particles 70-90 A in diameter have been found associated with the cytoplasmic halves of rhabdomere membrane in the honey bee (Perrelet et ul., 1972), a crayfish (Fernandez and Nickel, 1976), and a snail (Brandenburger et ul., 1976).
10
Final comments
The recent studies that have begun to characterize insect visual pigments have given us few surprises. They appear to differ from vertebrate photopigments mainly in having metarhodopsins of greater thermostability, a characteristic shared by invertebrate photopigments in general. One insect visual pigment, however, is uniquely interesting. Asculuphus rhodopsin is the only ultraviolet sensitive rhodopsin that has been isolated. The shift of its chromophore’s absorbance to shorter wavelengths is an intriguing problem. The analysis of chromophore-opsin interaction in Asculuphus rhodopsin has begun with the low temperature characterization of its intermediates. A unique 1 1-cis metarhodopsin photoregeneration intermediate is a significant feature of its photochemistry. The ultraviolet sensitive rhodopsin of Asculuphus has joined the photopigments of vertebrates and cephalopods as one particularly suited to photochemical analysis. From it we may expect important insights into the molecular basis of photoreception.
Acknowledgements
I thank Paul K. Brown, Reinhardt Paulsen and Joachim Schwemer for their helpful comments and criticism.
References Autrurn, H. and Burkhardt, D. (1961). Spectral sensitivity of single visual cells. Nature, Lond. 190,639. Autrurn, H. and von Zwehl, V. (1964). Die spektrale Empfindlichkiet einzelner Sehzellen des Bienenauges. Z.vergl. Physiol. 48,351-384.
INSECT VISUAL PIGMENTS
63
Benolken, R. M., Anderson, R. E. and Maude, M. B. (1975). Lipid composition of Limulus photoreceptor membranes. Biochim. Biophys. Acta, 413,234-242. Bowness, J. M. and Wolken, J. J. (1959). A light-sensitive yellow pigment from the housefly. J. Gen. Physiol. 42, 779-792. Brammer, J. D. and White, R. H. (1969). Vitamin A deficiency: effect on mosquito eye ultrastructure. Science, 163,821-823. Brandenburger, J. L., Eakin, R. M. and Reed, C. T. (1976). Effects of light-and dark-adaptation on the photic microvilli and photic vesicles of the pulmonate snail Helix uspersa. Vision Res. 16, 1205-1210. Bridges, C. D. B. (1976). Vitamin A and the role of the pigment epithelium during bleaching and regeneration of rhodopsin in the frog eye. Exp. Eye Res. 22,435-455. Briggs, M. H. (1961). Retinine-1 in insect tissues. Nature, Lond. 192,874-875. Brown, J. E., Murray, J. R. and Smith, T. H. (1967). Photoelectric potential from photoreceptor cells in the ventral eye of Limulus. Science, 158,665-666. Brown, P. K. and Brown, P. S. (1958). Visual pigments of the octopus and cuttlefish. Nature, Lond. 182,1288-1290. Brown, P. K. and White, R. H. (1972). Rhodopsin of the larval mosquito. J. Gen. Physiol. 59, 401-414.
Bruno, M. S. and Goldsmith, T. H. (1974). Rhodopsin of the blue crab Callinectes:evidence for absorption differences in vitro and in vivio. Vision Res. 14,653-658. Burkhardt, D. (1962). Spectral sensitivity and other response characteristics of single visual cells in the arthropod eye. Symp. SOC.Expt. Biol. 16,86-109. Burkhardt, D. (1964). Colour discrimination in insects. In “Advances in Insect Physiology”, Vol. 2, 131-173. Academic Press, New York, London. Burkhardt, D. and de la Motte, I. (1972). Electrophysiological studies on the eyes of Diptera, Mecoptera and Hymenoptera. In “Information Processing in the Visual Systems of Arthropods” (Ed. R. Wehner), pp. 147-153. Springer-Verlag, Berlin, Heidelberg, New York. Carlson, S. D. and Philipson, B. (1972). Microspectrophotometry of the dioptic apparatus and compound rhabdom of the moth (Manduca sexta) eye. J. Insect Physiol. 18,1721-173 1. Cone, R. (1967). Early receptor potential: photoreversible charge displacement in rhodopsin. Science, 155, 1128-1131. Cosens, D. and Briscoe, D. (1972). A switch phenomenon in the compound eye of the whiteeyed mutant of Drosophila melanogaster. J. Insect Physiol. 18,627-632. Daemen, F. J. M. (1973). Vertebrate rod outer segment membranes. Biochim. Biophys. Acta, 300,255-288.
Dartnall, H. J. A. (1953). The interpretation of spectral sensitivity curves. Brit. Med. Bull. 9, 24-30.
Eakin, R. M. (1972). Structure of invertebrate photoreceptors. In “Handbook of Sensory Physiology, Photochemistry of Vision” (Ed. H.J. A. Dartnall), Vol. VII/l, pp. 625-684. Springer-Verlag, Berlin, Heidelberg, New York. Ebrey, T. G. and Honig, B. (1975). Molecular aspects of photoreceptor fmition. Quart. Rev. Biophys. 8, 129-184. Fernandez, H. R. and Bishop, L. G. (1973). Photosensitive pigment from the worker honeybee, Apis mellvera. Vision Res. 13, 1379-1381. Fernandez, H. R. and Nickel, E. E. (1976). Ultrastructural and molecular characteristics of crayfish photoreceptor membrane. J. Cell Biol. 69,72 1-732. Frisch, K. von (1965). “Tanzsprache und Orientierung der Bienen.” Springer-Verlag, Berlin, Heidelberg, New York. Gogala, M., Hamdorf, K. and Schwemer, J. (1970). UV-Sehfarbstoff bei Insekten. Z . vergl. Physiol. 70,4 1 0 4 13. Goldman, L. J., Barnes, S. N. and Goldsmith, T. H. (1975). Microspectrophotometry of rhodopsin and metarhodopsin in the moth Galleria. J. Gen. Physiol. 66,383404.
64
RICHARD H. WHITE
Goldsmith, T. H. (1958a). The visual system of the honeybee. Proc. Nut. Acad. Sci. 44, 123126. Goldsmith, T. H. (1958b). On the visual system of the bee (Apis mellifea).Ann. N . Y. Acad. Sci. 74,223-229. Goldsmith, T. H. (1965). Do flies have a red receptor? J. Gen. Physiol. 49,265-287. Goldsmith, T. H. (1972). The natural history of invertebrate visual pigments. In “Handbook of Sensory Physiology. Photochemistry of Vision” (Ed. H. J. A. Dartnall), Vol. VII/l, pp. 6857 19. Springer-Verlag, Berlin, Heidelberg, New York. Goldsmith, T. H. (1975). The polarization sensitivity-dichroic absorption paradox in arthropod photoreceptors. In “Photoreceptor Optics” (Eds A. W. Snyder and R. Menzel), pp. 392-409. Springer-Verlag, Berlin, Heidelberg, New York. Goldsmith, T. H., Barker, R. J. and Cohen, C. F. (1964). Sensitivity of visual receptors of carotenoid-depleted flies: a vitamin A deficiency in an invertebrate. Science, 146,65-67. Goldsmith, T. H. and Bernard, G. D. (1974). The visual system of insects. In “The Physiology of Insecta” (Ed. M. Rockstein), pp. 165-272. Academic Press, New York, London. Goldsmith, T. H. and Fernandez, H. R. (1966). Some photochemical and physiological aspects of visual excitation in compound eyes. In “The Functional Organization of the Compound Eye” (Ed. C. G. Bernhard), pp. 125-143. Pergamon Press, New York. Goldsmith, T. H. and Fernandez, H. R. (1968). The sensitivity of housefly photoreceptors in the mid-ultraviolet and the limits of the visible spectrum. J. Exp. Biol. 49,669-677. Goldsmith, T. H. and Warner, L. T. (1964). Vitamin A in the vision of insects. J. Gen. Pkysiol. 47,433-441. Goldsmith, T. H. and Wehner, R. (1975). Photo-induced dichroism in a rhabdomeric photoreceptor: evidence for restricted rotation of pigment molecules. Biol. Bull. 149,427. Hagins, F. (1973). Purification and partial characterization of the protein component of squid rhodopsin. J. Biol. Chem. 248,3298-3304. Hagins, W. A. and McGaughy, R. E. (1967). Molecular and thermal origins of fast photoelectric effects in the squid retina. Science, 157,813-816. Ham, Jr., W. T., Mueller, H. A. and Sliney, D. H. (1976). Retinal sensitivity to damage from short wavelength light. Nature, Lond. 260, 152-153. Hamdorf, K. and Gogala, M. (1973). Photoregeneration und Bereichseinstellung der Empfindlichkeitbeim W-Rezeptor. J. Comp. Physiol. 86,23 1-245. Hamdorf, K., Gogala, M. and Schwemer, J. (1971a). Beschleunigung der Dunkeladaptation eines UV-Rezeptors durch sichtbare Strahlung. Z. Vergl. Physiol. 75, 189-199. Hamdorf, K., Hoglund, G. and Langer, H. (1972a). Mikrophotometrische Untersuchungen an der Retinula des Nachtschmetterlings Deilephila elpenor. Vehr. dtsck. zool. Ges. 65, 276280. Hamdorf, K., Hoglund, G. and Langer, H. (1973a). Photoregeneration of visual pigments in a moth. A microphotometric study. J. Comp. Physiol. 86,247-263. Hamdorf, K., Paulsen, R. and Schwemer, J. (1973b). Photoregeneration and sensitivity control of photoreceptors of invertebrates. In “Biochemistry and Physiology of Visual Pigments” (Ed. H. Langer), pp. 155-166. Springer-Verlag, Berlin, Heidelberg, New York. Hamdorf, K., Paulsen, R., Schwemer, J. and Taeuber, U. (1972b). Photoreconversion of invertebrate visual pigments. In “Information Processing in the Visual Systems of Arthropods” (Ed. R. Wehner), pp. 97-108. Springer-Verlag, Berlin, Heidelberg, New York. Hamdorf, K. and Rosner, G. (1973). Adaptation und Photoregeneration im Fliegenauge. J. Comp. Physiol. 8 6 2 8 1-292. Hamdorf, K. and Schwemer, J. (1975). Photoregeneration and the adaptation process in insect photoreceptors. In “Photoreceptor Optics” (Eds A. W. Snyder and R. Menzel), pp. 263-289. Springer-Verlag, Berlin, Heidelberg, New York.
INSECT VISUAL PIGMENTS
65
Hamdorf, K., Schwemer, J. and Gogala, M. (1971b). Insect visual pigment sensitive to ultraviolet light. Nature, Lond. 231,458-459. Harris, W. A., Stark, W. S. and Walker, J. A. (1976). Genetic dissection of the photoreceptor system in the compound eye of Drosophila melanogaster. J. Physiol. 256,415439. Hays, D. and Goldsmith, T. H. (1969). Microspectrophotometry of the visual pigment of the spider crab Libinia emarginala. Z . vergl. Physiol. 65,218-232. Heller, J. and Bok, D. (1976). Transport of retinol from the blood to the retina: involvement of high molecular weight lipoproteins as intracellular carriers. Exp. Eye Res. 22, 403-410. Hoglund, G., Hamdorf, K., Langer, R., Paulsen, R. and Schwemer, J. (1973a). The photopigments in an insect retina. In “Biochemistry and Physiology of Visual Pigments” (Ed. H. Langer), pp. 167-1 74. Springer-Verlag, Berlin, Heidelberg, New York. Hoglund, G., Hamdorf, K. and Rosner, G. (1973b). Trichromatic visual system in an insect and its sensitivity control by blue light. J. Comp. Physiol. 86, 265-279. Horridge, G. A. and Mimura, K. (1975). Fly photoreceptors. I. Physical separation of two visual pigments in Calliphora retinula cells 1-6. Proc. Roy. SOC.Lond. B , 190,211-224. Horridge, G. A., Mimura, K. and Tsukahara, Y. (1975). Fly photoreceptors. 11. Spectral and polarized light sensitivity in the drone fly Eristalis. Proc. Roy. SOC.Lond. B, 190,225-237. Hubbard, R., Brown, P. K. and Bownds, D. (1971). Methodology of vitamin A and visual pigments. In “Methods in Enzymology” (Eds D. B. McCormick and L. D. Wright), Vol. 18, pp. 615-653. Academic Press, New York, London. Hubbard, R. and Kropf, A. (1958). The action of light on rhodopsin. Proc. Nat. Acad. Sci. 130, 13G139. Hubbard, R. and St. George, R. C. C. (1958). The rhodopsin system of the squid. J. Gen. Physiol. 41, 501-528. Kirschfeld, K. and Snyder, A. W. (1975). Waveguide mode effects, birefringence and dichroism in fly photoreceptors. I n “Photoreceptor Optics” (EdsA. W. Snyder and R. Menzel), pp. 5677. Springer-Verlag, Berlin, Heidelberg, New York. Knoll, F. (1924). Lichtsinn und Bliitenbesuch des Falters von Deilephila livornica. Z . vergl. Physiol. 2,329-380. Kropf, A. (1967). Intramolecular energy transfer in rhodopsin. Vision Res. 7,8 11-8 18. Langer, H. (1965). Nachweis dichroitischer Absorption des Sehfarbstoffes in den Rhabdomeren den Insektenauges. Zeit. vergl. Physiol. 51,258-263. Langer, H. (1967). h e r die Pigmentgranula im Facettenauge von Calliphora erythrocephala. Z . vergl. Physiol. 55,354-377. Langer, H. and Thorell, B. (1 966). Microspectrophotometry of single rhabdomeres in the insect eye. Exp. Cell Res. 41,673-677. Laughlin, S. B., Menzel, R. and Snyder, A. W. (1975). Membranes, dichroism and receptor sensitivity. In “Photoreceptor Optics” (Eds A. W. Snyder and R. Menzel), pp. 237-259. Springer-Verlag,Berlin, Heidelberg, New York. Liebman, P. A. (1972). Microspectrophotometry of photoreceptors. In “Handbook of Sensory Physiology. Photochemistry of Vision” (Ed. H. J. A. DartnaU), Vol. VII/l, pp. 481-528. Springer-Verlag, Berlin, Heidelberg, New York. Lisman, J. E. and Brown, J. E. (1975). Effects of intracellular injection of calcium buffers on light adaptation in Limulus ventral photoreceptor. J. Gen. Physiol. 66,489-506. Marak,G. E., Gallik, G. J. and Cornesky, R. A. (1970). Light-sensitive pigments in insect heads. J. Ophthal. Res. 1,65-71. Mason, W. T., Fager, R. S. and Abrahamson, E. W. (1973). Characterization of the lipid composition of squid rhabdom outer segments. Biochim. Biophys. Acta, 306,67-73. McCann, G. D. and Amett, D. W. (1972). Spectral and polarization sensitivity of the Dipteran visual system. J. Gen. Physiol. 59,534-558.
66
RICHARD H. WHITE
Meffert, P. and Smola, U. (1976). Electrophysiological measurements of spectral sensitivity of central visual cells in eye of blowfly. Nature, Lond. 260,342-344. Menzel, R. and Snyder, A. W. (1974). Polarized light detection in the bee, Apis mellifra. J. Comp. Physiol. 88,247-270. Minke, B., Hochstein, S. and Hillman, P. (1973). Early receptor potential evidence for the existence of two thermally stable states in the barnacle visual pigment. J. Gen. Physiol. 62, 87-104.
Minke, B., Wu, C-F. and Pak, W. L. (1975). Isolation of light-induced response of the central retinula cells from the electroretinogram of Drosophila. J. Comp. Physiol. 98,345-355. Muri;R., Coles, J. and Baumann, F. (1976). Microspectrophotometry of rhabdomes in the honeybee drone. Experientia, 32, 759. Ostroy, S. E. and Pak, W. L. (1974). Protein and electroretinogram changes in the alleles of the norp API2Drosophila phototransduction mutant. Biochim. Bfophys. Acta, 368,259-268. Ostroy, S . E., Wilson, M. and Pak, W . L. (1974). Drosophila rhodopsin: photochemistry and extraction differences in the norp APL2phototransduction mutant. Biochem. Biophys. Res. Comm. 59,960-966. Pak, W. L. and Lidmgton, K. J. (1974). Fast electrical potential from a long-lived, longwavelength photoproduct of fly visual pigment. J. Gen. Physiol. 63,740-756. Paulsen, R. and Schwemer, J. (1972). Studies on the insect visual pigment sensitive to ultraviolet light: retinal as the chromophoric group. Biochim. Biophys. Acta, 283,520-529. Paulsen, R. and Schwemer, J. (1973). Proteins of invertebrate photoreceptor membranes. Characterization of visual-pigment preparations by gel electrophoresis. Eur. J. Biochem. 40, 577-583.
Pepe, I. M., Perrelet, A. and Baumann, F. (1976). Isolation by polyacrylamide gel electrophoresis of a light-sensitive vitamin A-protein complex from the retina of the honeybee drone. Vision Res. 16,905-908. Perrelet, A., Bauer, H. and Fryder, V. (1972). Fracture faces of an insect rhabdome. J. Microscopie, 13,97-106. Razmjoo, S . and Hamdorf, K. (1976). Visual sensitivity and the variation of total photopigment content in the blowfly photoreceptor membrane. J. Comp. Physiol. 105,-279-286. Rosner, G. (1975). Adaptation und Photoregeneration im Fliegenauge. J. Comp. Physiol. 102, 269-295.
Schwemer, J. (1969). Der Sehfarbstoff von Eledone moschata und seine Umsetzung in der lebenden Netzhaut. Z . vergl. Physiol. 62, 121-152. Schwemer, J., Gogala, M. and Hamdorf, K. (1971). Der UV-Sehfarbstoff der Insekten: Photochemie in vitro und in vivo. Z . vergl. Physiol. 75, 174-188. Schwemer, J. and Paulsen, R. (1973). Three visual pigments in Deilephila elpenor (Lepidoptera, Sphingidae). J. Comp. Physiol. 86,215-229. Seldin, E., White, R. H. and Brown, P. K. (1972). Spectral sensitivity of larval mosquito ocelli. J. Gen. Physiol. 59,415-420. Shaw, S. R. (1969). Interreceptor coupling in ommatidia of drone honeybee and locust compound eyes. Vision Res. 9,999-1030. Snyder, A. W. and Laughlin, S. B. (1975). Dichroism and absorption by photoreceptors. J. Comp. Physfol. 100, 101-1 16. Snyder, A. W. and Menzel, R., Eds. (1975). “Photoreceptor Optics.” Springer-Verlag, Berlin, Heidelberg, New York. Snyder, A. W.and Miller, W. H. (1972). Fly colour vision. Vision Res. 12, 1389-1396. Snyder, A. W. and Pask, C. (1973). Spectral sensitivity of Dipteran retinula cells. J. Comp. Physiol. 04,59-76. Stark, W. S . and Zitzmann, W. G. (1976). Isolation of adaptation mechanisms and photopigment spectra by vitamin A deprivation in Drosophila. J. Comp. Physiol. 105, 15-27.
INSECT VISUAL PIGMENTS
67
Stavenga, D. G . (1975). Dark regeneration of invertebrate visual pigments. In “Photoreceptor Optics” (Eds A. W. Snyder and R. Menzel), pp. 290-295. Springer-Verlag, Berlin, Heidelberg, New York. Stavenga, D. G., Zantema, A. and Kuiper, J. W. (1973). Rhodopsin processes and the function of the pupil mechanism in flies. In “Biochemistry and Physiology of Visual Pigments” (Ed. H. Langer), pp. 175-180. Springer-Verlag, Berlin, Heidelberg, New York. Swihart, S. L. and Gordon, W. C. (1971). Red photoreceptor in butterflies. Nature, Lond. 231,
126-127. Waterman, T. H. (1975). The optics of polarization sensitivity. I n “Photoreceptor Optics” (Eds A. W. Snyder and R. Menzel), pp. 339-371. Springer-Verlag, Berlin, Heidelberg, New York. Waterman, T. H., Fernandez, H. R. and Goldsmith, T. H. (1969). Dichroism of photosensitive pigments in rhabdoms of the crayfish Orconectes. J. Gen. Physiol. 54,415432. Weber, K. M. and Zinkler, D. (1974). Praparation und Isolierung der Rhabdome aus der Augen eines Insekts (CaNiphora erythrocephala (mutante “chalky”), Diptera). Cytobiologie, 9,5965. Wehner, R. and Goldsmith, T. H. (1975). Restrictions on translational diffusion of metarhodopsin in the membranes of a rhabdomeric photoreceptor. Biol. Bull. 149,450. Weinstein, G. W., Hobson, R. R. and Dowhg, J. E. (1967). Light and dark adaptation in the isolated rat retina. Nature, Lond. 215, 134-138. White, R. H. and Jolie, M. A. (1966). The effects of light and betacarotene upon the endoplasmic reticulum of the mosquito photoreceptor cell. J. Cell Biol. 31, 122A. White, R. H. and Lord, E. (1975).Diminution and enlargement of the mosquito rhabdom in light and darkness. J. Gen. Physiol. 65,583-598. Williams, T. P., Baker, B. N. and Eder, D. J. (1973). Interconversion of metarhodopsins. I n “Biochemistry and Physiology of Visual Pigments” (Ed. H. Langer), pp. 83-88. SpringerVerlag, Berlin, Heidelberg, New York. Wolken, J. J., Bowness, J. M.and Scheer, I. J. (1960). The visual complex of the insect: retinene in the housefly. Biochim. Biophys. Acta, 43,531-537. Wolken, J. J. and Scheer, I. J. (1963). An eye pigment of the cockroach. Exp. Eye Res. 2, 182-
188. Zigman, S. (1971).Eye lens color: formation and function. Science, 171,807-809. Zinkler, D. (1975).Zum Lipidmuster der Photorezeptoren von Insekten. Verh. dtsch. zool. Ges.
67.28-32.
This Page Intentionally Left Blank
Structure and Function of Insect Peptides Robert P. Bodnaryk CanadaAgriculture, Research Station. 195 Dafoe Road, Winnipeg, Manitoba, Canada 1 2
3
4
5
6 7
8
9
1
Introduction 69 Dipeptides 70 2.1 Basic dipeptides 70 2.2 /3-Alanyl-tyrosine 71 2.3 y-Glutamyl-phenylalanine 73 2.4 Other sequestered aromatic compounds 74 Glutathione 75 3.1 The y-glutamyl cycle 75 3.2 y-Glutamyl cycle enzymes in M . domestica 77 3.3 Glutathione in detoxication mechanisms 80 Metabolic aspects of peptide pools 88 Sex peptides from Drosophila 9 1 5.1 The sex peptide from D. melanogaster 92 5.2 PS-I and PS-2 from D. funebris 93 5.3 Sex peptides in other Diptera 94 Proctolin, a proposed neurotransmitter in insect visceral muscle 94 Physiologically active peptides from the corpus cardiacum 96 7.1 Heart-accelerating peptides 97 7.2 The hyperglycaemic hormone 101 Peptides in insect venoms 105 8.1 Bee venom peptides 106 8.2 Kinins from wasps and hornets 116 Concluding remarks 118 Acknowledgements 119 References 119
Introduction
The peptides that have been discovered in the blood, glands or tissues of insects have widely diverse metabolic and physiological function. Cuticle sclerotization, amino acid transport, detoxification of insecticides, neurotransmission, metabolic regulation, regulation of female sexual behaviour and the pharmacological activity of venoms involve peptides in one insect species or another. Insect peptides that have been characterized to date range in size from 69
70
ROBERT P. BODNARYK
dipeptides (section 2) of uncommon amino acid composition or atypical peptide linkage to a 27-amino-acid-residue sex peptide found only in the male accessory gland of Drosophila funebris (section 5). Amounts of individual peptides in insects vary tremendously. The pentapeptide, proctolin, for example, occurs in the hindgut of the cockroach at approximately dhindgut, an amount that is consistent with its proposed function 1.6 x as an excitatory neurotransmitter in insect visceral muscle (section 6). At the other end of the scale, the dipeptide sarcophagine (B-ala-tyr) occurs in the glml, blood of the fully grown larva of Sarcophaga bullata at about 1.3 x an amount that is consistent with its role as a sequestered storage form of tyrosine (section 2). These features of the insect peptides offer a challenge to physiologist and biochemist alike, and much progress in this field has been made to date. An encouraging advance in our knowledge of the insect peptides has been the steadily increasing rate at which their amino acid sequences are becoming known. No fewer than four of the ten insect peptides (excluding dipeptides) whose amino acid sequences are known were sequenced in 1975 or later. All but three of the ten peptides have been synthesized. Given the technical advances in isolation, sequence determination and synthesis of peptides, the day is rapidly disappearing when physiologists remain confined to experimentation with ill-defined or semi-pure extracts of insect peptides. The process of discovery, purification, sequence determination and synthesis often follow one another with a delightful precision of rhythm in many laboratories. Yet, the most satisfying aspect of a study of insect peptides is seen in the diversity of metabolic and physiological roles they play in the life-style of the insects. This review is concerned with the structure and function of insect peptides.
2
2.1
Dipeptides
BASIC DIPEPTIDES
The first unequivocal account of insect dipeptides was given by Levenbook (1966). Nine of nineteen small, soluble peptides found in the blood of young larvae (second and third instars) of the blowfly Phormia regina were isolated and analysed. Eight of these were composed of two amino acids only, while the ninth was a tripeptide. The remaining ten peptides occurred in low concentration and were not investigated. The amino acid sequences of the peptides were reported by Bodnaryk and Levenbook (1968) (Fig. 1). This unusual group of peptides all have a basic C-terminal residue; two have histidine and the remaining seven, lysine.
STRUCTURE A N D FUNCTION OF INSECT PEPTIDES
71
The biological significance of the blowfly peptides remains unknown at present. They occur in the blood of the late second instar larva, reach their maximum concentration in the early third instar and decrease rapidly thereafter. The peptides are not unique to P. regina; several of them have been detected in extracts of second and third instar larvae of the fleshfly Sarcophaga Asp-His; /?-Ala-His (carnosine) Asp-Lys; Asn-Lys; Glu-Lys; Ile-Lys; Ser-Lys; Val-Lys; Gly-Lys-Lys Fig. 1. Structure of dipeptides and one tripeptide from the blood of young larvae of the blowfly, P. regina (Bodnaryk and Levenbook, 1968). Peptide linkage involves the alpha-carboxyl and alpha-aminogroup of each peptide; amino acids are in the L-form.
bullata and the housefly Musca domestica. It is unlikely that they are mere digestive products from the diet, since labelled peptides were detected in P. regina larvae reared on a semi-defined diet containing uniformly labelled Chlorella protein hydrolysate, indicating their synthesis f?om free amino acids. 2.2
/I-ALANYL-TYROSINE
A novel dipeptide, p-alanyl-tyrosine (termed “sarcophagine”) has been isolated from larvae of the fleshfly, Sarcophaga bullata (Levenbook et al., 1969; Bodnaryk and Levenbook, 1969). Unlike the transitory, basic dipeptides described in the preceding section, p-ala-tyr accumulates throughout larval growth until it is the major nonprotein, ninhydrin-positive substance in the blood of the fully grown larva. The dipeptide is found almost exclusively in the blood, at nearly 50 pnoleslml in a fully grown larva. It sequesters more than 70 per cent of the larva’s nonprotein tyrosine, and more than 97 per cent of its p-alanine. 2.2.1
Metabolic fate and function
The transition of the fully grown fleshfly larva to the “white puparium” stage is accompanied by a slight reduction in the amount of p-ala-tyr. Thereafter, p-alatyr decreases precipitously during the first few hours of hardening and darkening (sclerotization) of the puparium. Twelve hours after the onset of sclerotization, only traces of the dipeptide can be found. Several lines of evidence indicate that the dipeptide is hydrolysed by a peptidase to p-alanine and tyrosine during sclerotization and that each component enters the structure of the puparium, without further modification in the case of p-alanine (Bodnaryk and Levenbook, 1969; Bodnaryk, 1970a; 197la,b,c), and after
72
ROBERT P. BODNARYK
extensive metabolism to quinones in the case of tyrosine (Bodnaryk and Levenbook, 1969; see Karlson and Sekeris, 1964; Karlson and Herrlich, 1965). Incorporation of palanine and tyrosine into sclerotized puparium structure quantitatively accounts for the amount of /3-ala-tyr consumed during sclerotization. The dipeptide 8-ala-tyr is about 200 times as soluble as free tyrosine in distilled water at neutral pH, and it is noteworthy that the corresponding alphaalanyl derivative is only sparingly soluble. The combination of 8-alanine with tyrosine serves as a possibly unique biochemical adaptation by the larva of Sarcophaga for the accumulation of high concentrations of readily available tyrosine for use in forming the structure of the puparium. 2.2.2
Hormonal control
Larvae of S. bullata ligatured behind the brain prior to the release of the moulting hormone ecdysone fail to pupate posterior to the ligature (Fraenkel, 1935). The level of /3-ala-tyr remains at its larval level in these unpupated, unsclerotized posterior sections, suggesting that ecdysone is required to initiate the metabolism of the dipeptide, perhaps through the induction of a p-ala-tyrspecific peptidase (Bodnaryk and Levenbook, 1969). If precocious sclerotization of the larval cuticle is induced by a large dose of ecdysone given to larvae, the level of 8-ala-tyr in the artificially created “pupa” is depleted in a manner similar to that observed during normal sclerotization. Analyses of larvae in which precocious, ecdysone-induced sclerotization was blocked by a DOPA-decarboxylase inhibitor indicated that a 8-ala-tyr peptidase and DOPA-decarboxylase are ecdysone-induced enzymes involved in the utilization of /3-ala-tyr (Bodnaryk, 197lc). The enzyme system responsible for /3-ala-tyr synthesis and the peptidase involved in its hydrolysis have been isolated recently by Dunn, Fader and Regnier (Dunn, personal communication, 1976). 8-Ala-tyr synthetase is a soluble enzyme found in the fat body of S . bullata and requires Mg++and ATP for activity. P-Ala-tyr hydrolase is found predominantly in the fat body but also occurs in the integument. No cofactors, cosubstrates or metal ions were required for in vitro hydrolysis of pala-tyr. Fat body /3-ala-tyr hydrolase begins to increase sharply after the formation of the white puparium stage, reaches a maximum 12-18 h later and declines rapidly thereafter. 2.2.3
Occurrence in other Diptera
A survey of fully grown larvae of 46 species (3 1 genera, 14 families) of Diptera by Bodnaryk (1972a) revealed that all members of the genus Sarcophaga synthesize /3-ala-tyr. The occurrence of this dipeptide was not genus-specific, but
STRUCTURE A N D FUNCTION OF INSECT PEPTIDES
73
nearly so. Of the 40 non-Sarcophaga species examined, only the tachinid Voria ruralis contained P-ala-tyr. However, four other tachinids that were examined, namely Lespesia archippivora, Leschenaultia adusta. Exorista mella and Eucelatoria armigera, contained no trace of P-ala-tyr. These observations vitiate an earlier proposal (Bodnaryk, 1970b) that synthesis of P-ala-tyr might be a specific taxonomic criterion of the genus Sarcophaga. Nevertheless, an expanded survey of Diptera for /I-ala-tyr and other aromatic compounds (see following sections) may assist in tracing evolutionary relationships among the Diptera, P-Ala-tyr has not been detected in the very few cases that it has been searched for in nondipterous insects. It is absent from newly ecdysed nymphs and adults of the cockroach, P. americana (Brunet, personal communication) and newly formed pupae of the noctuid Mamestra configurata (Bodnaryk, unpublished). 2.3
7-GLUTAMYL-PHENYLALANINE
A dipeptide, y-glu-phe, with the uncommon y-carboxyl a-amino peptide linkage has been isolated from the larva of the housefly, Musca domestica (Bodnaryk, 1970~).Much like p-ala-tyr in S. bullata described in the preceding section, yglu-phe accumulates during housefly larval growth, becoming the predominant nonprotein, ninhydrin positive substance in the blood of the fully grown larva. This dipeptide is also confined almost entirely to the blood and occurs at 26pmoles/ml in fully grown larvae. y-Glu-phe sequesters more than 80 per cent of the nonprotein phenylalanine and 40 per cent of nonprotein glutamic acid in the newly formed “white puparium” stage (Bodnaryk, 1970). Moreover, y-glu-phe is highly restricted to the larva of members of the genus Musca, but its occurrence is not genus-specific. Of 40 non-Musca dipterans surveyed, only the muscid Stomoxys calcitrans was found to contain y-glu-phe (Bodnaryk, 1972a). 2.3.1
Metabolic fate and function
y-Glu-phe is rapidly and extensively consumed during hardening and darkening of the housefly puparium. Some idea of the magnitude of the rate of dipeptide consumption can be had from the following data: the amount of y-glu-phe at the newly formed “white puparium” stage is about 275 nmoles/insect. Six hours later, the amount is 25 nmoleshnsect. After 12 h the dipeptide is virtually undetectable (Bodnaryk, 1974). The rapid disappearance of y-glu-phe from the pupa coincides exactly with the hardening and darkening of its puparium. Ligature and ecdysone injection experiments leave little doubt that the utilization of y-glu-phe is under the control of the moulting hormone, ecdysone (Bodnaryk, 1970; Bodnaryk and Skillings, 1971).
74
ROBERT P. BODNARYK
There appear to be two simultaneously occurring reactions which break down y-glu-phe in the pupa: (1) conversion of y-glu-phe to 5-oxoproliie and free phenylalanine, catalysed by y-glutamyl cyclotransferase; and (2) hydrolysis of yglu-phe, likely catalysed by an ecdysone-induced y-glutamyl transpeptidase. Virtually none of the y-glutamylresidue of injected [G-3Hl y-glu-phe is incorporated into puparium structure, whereas there is extensive incorporation of the radioactivity of phenylalanine (Bodnaryk, 1974). Most probably, phenylalanine released from the dipeptide is converted to tyrosine which is then hydroxylated to dihydroxyphenylalanine (DOPA), and this in turn is decarboxylated by DOPA-decarboxylase and N-acetylated to form Nacetyl-dopamine. This latter is oxidized by the insect phenoloxidase system to the corresponding quinone which is the actual sclerotizing agent of the cuticular protein, according to the scheme proposed by Karlson and Herrlich ( 1965) for a related fly, Calliphora erythrocephala. However, experimental evidence for these reactions involving tyrosine in M. domestica has not been obtained to date. The biosynthesis of y-glu-phe from phenylalanine and glutathione catalysed by y-glutamyl transpeptidase (Bodnaryk and Skillings, 1971) and the breakdown of y-glu-phe catalysed by y-glutamyl cyclotransferase. (Bodnaryk and McGirr, 1973; Bodnaryk, 1974) are discussed in thelarger context ofthe yglutamyl cycle in section 3.1.
2.4
OTHER SEQUESTERED AROMATIC COMPOUNDS
The dipeptides P-ala-tyr in Sarcophaga and y-glu-phe in Musca represent but two strategies evolved by insects for sequestering large quantities of aromatic compounds in the larval stage for subsequent utilization in cuticle sclerotization. In Drosophila larvae, tyrosine is sequestered as its phenol phosphate ester, tyrosine-O-phosphate (Mitchell and Lunan, 1964; Lunan and Mitchell, 1969), a compound that is highly restricted to members of this genus (Bodnaryk, 1972a). In the cockroach Periplaneta americana, dopamine 3-0sulphate may serve as a protected and sequestered form of dopamine (Bodnaryk and Brunet, 1974; Bodnaryk ef al., 1974). Recently, an interesting compound called “celerin” has been obtained in crystalline form from the blood of the caterpillar, Celerio euphorbiae (Sienkiewicz and Piechowska, 1973). The authors propose that celerin is L-tyrosyl-O-acetyldopamineand call it a dipeptide. In my opinion, the exact structure of celerin has not yet been proven. The stoichiometry of tyrosine, dopamine and the acetate group has not been established by the authors in their report. The location of the acetate group(s) was not specified: either the 3-OH or the 4-OH of dopamine could be
STRUCTURE A N D FUNCTION OF INSECT PEPTIDES
75
acetylated, and the possibility of the tyrosine OH being acetylated is not ruled out. A structure of the type given by Sienkiewicz and Piechowska (1973) is not a true dipeptide. The blood of the larva of Pieris brassicae and four other randomly chosen lepidopterans contains a ninhydrin-positive, phenolic compound with properties similar to those of celerin (Junnikkala, 1968, 1976). Celerin may be an important precursor of the sclerotizing agent(s) in Lepidoptera, emphasizing the need for a clarification of its structure.
3
3.1
Glutathione THE
7-GLUTAMYL CYCLE
The tripeptide y-glutamyl-cysteinyl-glycine (reduced glutathione, GSH) occurs widely in animal tissue at appreciable concentrations (about 3-5 mM for kidney). Until quite recently, a major function for glutathione had not been discovered. Several lines of converging evidence now suggest that glutathione participates in a y-glutamyl cycle, the operation of which may mediate the transport of many (but not all) amino acids in the kidney and perhaps in a variety of other tissues (Orlowski and Meister, 1970). Experimental evidence in support of this intriguing hypothesis has come principally from extensive biochemical studies of the enzymes involved in glutathione metabolism by Meister and collaborators (Orlowski and Meister, 1970; revs. Meister, 1973, 1974; Meister and Tate, 1976), from clinical studies of patients with apparent inborn metabolic errors involving the y-glutamyl cycle (rev. in Meister and Tate, 1976) and from physiological and biochemical studies of the metabolism of the dipeptide, y-glu-phe, in the larva of the housefly (Bodnaryk, 1970c,d; 1972a,b; 1974; Bodnaryk and Skillings, 1971; Bodnaryk and McGirr, 1973; Bodnaryk et al., 1974). Indeed, this latter research seems to have provided the most convincing evidence to date for the translocation of a specific amino acid (phenylalanine) via a y-glutamyl cycle mechanism. The operation of the y-glutamyl cycle as originally proposed by Orlowski and Meister (1970) and updated by Meister (1973, 1974) is given in Fig. 2. Six enzymes are involved, five of which are soluble enzymes found in the cytosol. The sixth and key enzyme of the cycle is the membrane-bound y-glutamyl transpeptidase which is intimately involved in the translocation of the amino acid from the exterior to the interior of the cell. A scheme has been proposed by Meister (1973) in which the membrane-bound y-glutamyl transpeptidase interacts with the y-glutamyl moiety of intracellular glutathione to yield a y-glutamyl-enzyme complex. The extracellular amino acid to be translocated is brought in contact with the membrane at the transport site, perhaps
76
ROBERT P. BODNARYK
by noncovalent binding. There follows an attack of the amino acid nitrogen atom on the y-carbon atom of the yglutamyl-enzyme complex to yield a yglutamyl-amino acid. The formed y-glutamyl-amino acid is then released into the interior of the cell, facilitated perhaps by a conformational change in the cell membrane. The amino acid is finally released from its carrier y-glutamyl group by the action of y-glutamyl cyclotransferase. The ATP-dependent
Amino acid (outside cell)
I
4 Cell membrane
y-glutamyl transpeptidase (membrane bound)
/I
1 II
~ - g l u t a m y l - a h n o acid
/
y-glutarnyl-cysteinyl-glycine (glutathione)
7
Cysteinyl-glycine
7-glutamyl c yclotransferase
\
Peptidase ADP+Pi Amino acid (inside cell)
ATP
A(5-oxoproline)
~
p
~
ADP +PI 5-oxaprolinas ADPtPi
synthetase Glutomic acid
Fig. 2. The pglutamyl cycle (Orlowski and Meister, 1970; Meister, 1973, 1974). 1, recognition and translocation of extracellular amino acid; 2, release of amino acid from its pglutamyl carrier within the cell; 3,4,5, energy-recovery steps.
decyclization of 5-oxoproline and resynthesis of glutathione are the energyrequiring recovery steps in the cycle. Three moles of ATP are utilized to bring one mole of amino acid into the cell by this mechanism. This would appear to be a metabolically “expensive” transport system, but, according to Meister (1974) the high energy requirement may reflect the need for high efficiency in such tissues as kidney and brain.
STRUCTURE AND FUNCTION OF INSECT PEPTIDES
3.2 3.2.1
7-GLUTAMYL CYCLE
77
ENZYMES IN “ M . DOMESTICA”
y-Glutamyl transpeptidase
Discovery of the synthesis and accumulation of y-glu-phe in the larva of M. domestica (Bodnaryk, 1970c, section 2.3) led to a consideration of the mechanisms of its biosynthesis. A particulate y-glutamyl transpeptidase was found in great abundance in the larvae of this species. It was solubilized with sodium deoxycholate and purified about 100-fold. The enzyme catalysed the formation of y-glu-phe in a classical transpeptidation reaction (Hanes et al., 1950) in which the y-glutamyl moiety of glutathione was transferred to phenylalanine (Bodnaryk and Skillings, 1971). Many of the common ac-L-amino acids were as effective or more effective than phenylalanine as acceptors of the yglutamyl residue in vitro, resulting in y-glutamyl-amino acid formation. In vivo, however, only y-glu-phe has been observed in housefly larvae to date. The broad specificity of y-glutamyl transpeptidase indicates that many y-glutamylamino acids might be formed in vivo but turned over too quickly for detection. No completely satisfactory explanation to account for the specific accumulation of y-glu-phe in the housefly larva has yet been provided. The low level in the larva of y-glutamyl cyclotransferase, an enzyme involved in the breakdown of y-glu-phe (Bodnaryk and McGirr, 1973; Section 3.2.2), and the physical compartmentalization of y-glu-phe in the blood are undoubtedly important factors. The tissue distribution of y-glutamyl transpeptidase activity has been studied by histochemical methods by Bodnaryk et al. (1974). Most of the y-glutamyl transpeptidase activity of actively-feeding third instar housefly larvae is located on membranes, specifically on the brush border of the proximal half of the Malpighian tubules (the distal half is completely inactive) and the brush border of epithelial cells of the anterior and posterior portions of the midgut. The localization of y-glutamyl transpeptidase activity on membranes is consistent with biochemical studies which indicated its particulate nature. In effect, y-glutamyl transpeptidase and y-glu-phe form a highly specific system for the absorption and reabsorption of phenylalanine from the lumen of the midgut and Malpighian tubules. The membrane-bound enzyme combines with glutathione (intracellular) and phenylalanine (extracellular) and the resulting y-glu-phe is translocated across the cell membrane and released within the cell. The process is identical to the first step of the y-glutamyl cycle. However, phenylalanine is not released from its y-glutamyl linkage within the cell and y-glu-phe enters the blood, presumably by simple diffusion in response to the concentration gradient generated by its build-up within the cell. A high rate of formation of y-glu-phe combined with low y-glutamyl cyclotransferase activity (Bodnaryk and McGirr, 1973) in larval tissues likely accounts for the accumulation of the dipeptide. Formation of y-glu-phe in housefly larvae may
ROBERT P. BODNARYK
70
represent a special adaptation of the y-glutamyl cycle enzyme, y-glutamyl transpeptidase, for the efficient and selective accumulation of phenylalanine during larval growth (see section 2.3). Remarkable changes occur in y-glutamyl transpeptidase activity during the transition of the actively feeding larva to a post-feeding larva and pupa. The most evident change in the post-feeding larva is the loss of y-glutamyl transpeptidase activity throughout most of the gut, undoubtedly coinciding with a progressive loss of gut function. The Malpighian tubules, however, retain most of their activity. Transition of the post-feeding larva to the inert, barrel-shaped “white puparium” stage is accompanied by an abrupt appearance of very intense y-glutamyl transpeptidase activity on the epidermal cell membrane at the epidermis-cuticle interface. The appearance of this new, epidermal cell membrane transpeptidase is induced by ecdysone, as determined by ligature and ecdysone-injection experiments (Bodnaryk et al., 1974). Epidermal cell yglutamyl transpeptidase activity is maximal 1 to 2 hr after the white puparium stage forms, and thereafter diminishes rapidly. It is virtually undetectable at the time of larval-pupal apolysis. Maximal activity of the transpeptidase coincides with the period of most rapid breakdown of y-glu-phe. Apparently, the ecdysone-induced epidermal-cell transpeptidase catalyses the transfer of the yglutamyl residue of y-glu-phe on to water (presumably in the absence of other suitable y-glutamyl acceptors) and thereby effects the hydrolysis of the dipeptide. Detailed kinetic analyses of the breakdown of y-glu-phe during hardening and darkening of the housefly puparium have in fact provided evidence for a y-glutamyl transpeptidase mediated hydrolysis of y-glu-phe (Bodnaryk, 1974). 3.2.2
y- Glutamyl cyclotransferase
y-Glutamyl cyclotranferase, first described in pig liver by Connel and Hanes in 1956, catalyses the second step of the y-glutamyl cycle-the release of the amino acid from its y-glutamyl carrier within the cell (Fig. 2). A y-glutamyl cyclotransferase which catalyses the conversion of y-glu-phe into 5-oxoproEne (=5-oxo-~-proline, L-5-0x0-pyrrolidine-Zcarboxylic acid, ~-2-pyrrolidone-5carboxylic acid, L-pyroglutamic acid) and phenylalanine has been isolated (4 1O@-foldpurification) from housefly pupae (Bodnaryk and McGirr, 1973). The housefly enzyme exhibits a very narrow and unique specificity, compared with purified mammalian y-glutamyl cyclotransferase (Orlowski et al., 1969). y-Glu-phe is the most active substrate for the housefly cyclotransferase, whereas human and sheep brain cyclotransferase attack this substrate extremely slowly. The preferred substrate for the mammalian enzyme appears acids. According to Orlowski et al. (1969), to be y-glutamyl-y-glutamyl-amino y-glutamyl amino acids are metabolized in a two-step reaction, involving the initial reaction of two y-glutamyl-amino acid molecules catalysed by y-glutamyl
STRUCTURE AND FUNCTION OF INSECT PEPTIDES
79
transpeptidase to yield a y-glutamyl-y-glutamyl-amino acid and a free amino acid, followed by the conversion by y-glutamyl cyclotransferase of the yglutamyl-y-glutamyl-amino acid to a y-glutamyl-amino acid and 5-oxoproline. In the housefly, no evidence exists for such a two-stage reaction: y-glu-phe appears to be converted directly to phenylalanine and 5-oxoproliie (Bodnaryk and McGirr, 1973; Bodnaryk, 1974). The activity of y-glutamyl cyclotransferase in the housefly larva is relatively low, but undergoes a several-fold increase during transition to the “white puparium” stage. Levels of 5-oxoproline in the larva are also very low (about 10 nmoles/insect) and suddenly increase to about 80 nmoles/insect at this time, only to return again to a low level when all of the y-glu-phe has been utilized. Formation of labelled 5-oxoproline from [G-3Hly-glu-phe has been established, leaving little doubt that y-glutamyl cyclotransferase participates in the breakdown of y-glu-phe during hardening and darkening of the puparium. The sudden appearance at the white puparium stage of a highly active y-glutamyl cyclotransferase with specificity directed towards y-glu-phe is probably under the control of ecdysone, although the point has not been confirmed by experimentation.
3.2.3
5-Oxoprolinase
5-Oxoprolinase, first described in rat kidney by van der Werf et al. (1971), catalyses the third step of the y-glutamyl cycle-the ATP-dependent decyclization of 5-oxoproline to glutamic acid. The reaction is one of three energy-recovery steps of the cycle (Fig. 2). 5-Oxoprolinase has not yet been isolated from housefly tissue, but there is evidence to suggest its presence (Bodnaryk, 1974). Radioactivity from [UJ4C1 5-oxoproline injected at the “white puparium” stage is recovered in glutamic acid, and is expired as I4CO,. Kinetic data indicate that conversion of 5-oxoproline to glutamic acid is a relatively slow reaction in the housefly “white puparium” stage, thereby accounting for the accumulation of 5-oxoproline during the breakdown of y-glu-phe. However, the accumulation is of short duration, and levels of 5oxoproline return to “normal” when y-glu-phe is finally consumed. 3.2.4
y-Glutamyl-cysteine synthetase and glutathione synthetase
The remaining two enzymes which complete the y-glutamyl cycle involve the resynthesis of glutathione from glutamic acid, cysteine and glycine, as originally described by Block and coworkers (ref. in Meister, 1973). Two moles of ATP are required for the formation of one mole of glutathione and the reactions are energy-recovery steps of the cycle (Fig. 2). Glutathione biosynthesis has not been investigated systematically in insects. There appear to
80
ROBERT P. BODNARYK
be no accounts of y-glutamyl-cysteine synthetase or glutathione synthetase in any insect; hard data on glutathione levels and glutathione turnover are lacking for the most part. The housefly larva with its highly developed capacity to form and store yglu-phe, may prove to be an interesting experimental animal for such studies. 3.3 3.3.1
GLUTATHIONE I N DETOXICATION MECHANISMS
Glutathione as a specijic co-factor of DDT-dehydrochlorinase
DDT-dehydrochlorinase (E.C. 4.5.1.1 ; commonly DDT-ase) from insects catalyses the conversion of DDT [2,2,bis-(p-chlorophenyl)-1,l 1trichloroethanel to DDE [2,2-bis-(p-chlorophenyl)1, l-dichloroethylenel. DDT-ase has an essential and specific requirement for GSH that cannot be met by a wide variety of sulfhydryl compounds (Lipke and Kearns, 1959a,b; 1960) although the DDT-dehydrochlorination reaction also occurs in the presence of cysteinylglycine but at a slower rate. The requirment of DDT-ase for GSH is remarkably high, with a Km for GSH between 2.5 x 1 0 - 4 ~(Lipke and Kearns, 1959b), 1-3 x 1 0 - 4 ~(Balabaskaran and Smith, 1970) and 5 x M (Maccioni et al., 1970) determined in various strains of houseflies. DDT-ase is a “sulfhydryl enzyme” with 32 cysteines per 120 OOO molecular weight tetramer, the cysteines begin unequally buried in hydrophobic regions of the protein matrix. There are no disulfide bonds and cysteine residues do not appear to be associated with the enzyme active centre. The enzyme likely contains a phospholipid moiety (Dinamarca et al., 1971). DDT-ase is composed of four subunits of approx. mol. wt. 30 OOO which in the presence of DDT become aggregated into a tetrameric structure of approx. mol. wt. 120000. The tetramer is maximally active, the trimer being 10 per cent as active and the dimer and’ monomer inactive. The tetramer is stabilized in its aggregated state by GSH, but not by /$mercaptoethanol or dithiothreitol which in fact disaggregate the tetramer (Dinamarca et al., 1969; 1971). These intriguing observations, however, still do not explain the GSH requirement for enzymatic dehydrochlorination of DDT. The role of GSH as a specific cofactor of DDT-ase is obviously more complex than maintenance of protein sulfhydryls in the reduced state (Dinamarca et al., 1971). Until recently, no reaction of any type between GSH and DDT-ase had been uncovered. It has now been found that the enzyme can act as a glutathione oxidase, converting GSH to GSSG, albeit at a slow rate and at a temperature (30° C) that is sub-optimal for the elimination reaction. Again, the significance of this observation for the dehydrochlorination reaction is obscure. The nature of the essential participation of GSH in DDTdehydrochlorination is being pursued actively by the Chilean group (Dinamarca et al., 1974).
STRUCTURE AND FUNCTION OF INSECT PEPTIDES
3.3.2
81
Glutathione S-transferases
Mercapturic acid biosynthesis represents one (conjugation) of many types of chemical transformation that foreign compounds in living organisms may undergo. The initial step in mercapturic acid formation is conjugation of the foreign compound with glutathione, a reaction catalysed by glutathione Stransferase. The topic has been reviewed by Boyland and Chasseaud (1969) and various accounts of glutathione conjugation reactions in insects and other organisms have appeared (Smith, 1955; 1962; Perry, 1964; Parke, 1968; Perry and Agosin, 1974; Eto, 1974; Habig et al., 1974a,b; Pabst et al., 1974; Plapp, 1976). 3.3.3
Glutathione S-aryltransferase
a Conjugation with xenobiotics Glutathione S-aryltransferase from rat liver catalyses reactions of glutathione with aromatic and other cyclic compounds containing labile halogen or nitro groups (Booth et al., 196 1). Locusts rapidly metabolize p-nitrobenzyl chloride to S-(9-nitrobenzyl) glutathione. The glutathione conjugate is subsequently hydrolysed in the locust gut, malpighian tubules and excreta to S-(9-nitrobenzyl) cysteine which in turn is converted to unidentified products (Fig. 3). Glutathione conjugates and cysteine conjugates of phenoltetrabromophthaleindisulphonate, benzyl chloride, p-chlorobenzyl chloride, p-nitrobenzyl bromide, 1-chloro-2,4-dinitrobenzene,1-fluoro-2,4dmitrobenzene and 3,4-dichloro-1-nitrobenzene were identified in locusts dosed with these compounds (Cohen and Smith, 1964). Detoxication of aromatic halogen compounds is thus fundamentally the same in the locust and vertebrates. However, the GSH derivatives are excreted as mercapturic acids (compounds containing an N-acetyl-L-cysteine residue) in vertebrates as the kidney does not allow in most animals the passage of unacetylated metabolites. In locusts, mercapturic acids are insignificant metabolic products and both the GSH derivative and cysteine derivative formed by hydrolysis are excreted (Cohen and Smith, 1964). The activity of glutathione S-aryltransferase (called “glutathiokinase” by the authors) towards four of the above compounds was highest in fat body, malpighian tubules and gut, with small amounts in other tissues. Enzyme activity was found in all seven of the insects tested and in the tick Boophilus. The insect glutathione S-aryltransferase was distinguished from the analogous enzymes in rat, rabbit and tick by its marked sensitivity to inhibition by phthaleins (Cohen et al., 1964). Clark et ul. (1967) also found that glutathione S-aryltransferase from the grass grub, Cosfelytru zealundicu, was inhibited by phthaleins, sulphonphthaleins and some dicarboxylic acids whereas these compounds had no detectable action on the enzyme from sheep liver. Compounds such as the phthaleins and sulphonphthaleins are isosteric
82
ROBERT P. BODNARYK
CO.NH.CH,.COOH / CH,CI + HS.CH,.CH \
NH .CO.CH,.CH,. CH .COOH I
I
NH, y-glut amylcysteinylglycine (glutathione)
p-nitrobenzylchloride
NO,
(0)
I
glutathione S-aralkyltransferase
CO.NH.CH,.COOH CH,-S.CH,.CH/
/
NH.CO.CH,.CH,.CH .COOH
I NH,
S (p-nitrobenzyl) glutathione
I
r NO,
(0)
y-glutamyl transpeptidase
CO.NH.CH,.COOH CH,-S.CH,.CH/
\ NH,
1 NO,
S(p-nitrobenzyl) cysteinylglycine
I (0)
peptidase (?)
CH,-S.CH,.CH
locust: excretion andlor metabolism to unidentified products
\ NH,
S (p-nitrobenzyl) cysteine
I NO,
(0)
vertebrate liver
,COOH CHI-S.CH,.CH
\
NH .CO .CH, S(p-nitrobenzyl) mercapturic acid
+
STRUCTURE A N D FUNCTION OF INSECT PEPTIDES
83
with GSH and inhibit the aryltransferase by competing with GSH for its binding site. The active site of the grass-grub enzyme differs from that of the sheep-liver enzyme in that it has two binding groups with pK 9.2. The structural requirements of various phthaleins for inhibition of glutathione Saryltransferase from grass-grub and housefly preparations has been examined by Balabaskaran and Smith (1970). Their results support the suggestion of Clark et al. (1967)that the two oxygen atoms in the inhibitor structure which are isosteric with oxygen atoms in the carboxyl groups of GSH are concerned in the binding of the inhibitors to the enzyme. Highly purified glutathione Saryltransferase has been obtained from housefly homogenates by electrophoresis and electrofocusing. The enzyme migrates as a single band that is separate and distinct from the multiple forms of DDT-dehydrochlorinase (Goodchild and Smith, 1970). b Conjugation with y-BHC, a chlorohydrocarbon insecticide Interest in insect glutathione S-transferases was stimulated by the discovery that the insecticide y- 1,2,3,4,5,6-hexachlorocyclohexane(gammexane, lindane, y-BHC) is detoxified by formation of a glutathione conjugate. The possibility was first indicated by the work of Bradbury and Standen (1959,1960). They exposed yBHC-resistant houseflies to ['*CI y-BHC and extracted them after 24 h. The extract, after strong alkaline hydrolysis, was found to contain a mixture of labelled dichlorothiophenols, indicating that the metabolism of y-BHC to watersoluble metabolites involved formation of a C--S bond. The situation was analogous to the metabolism in rats of chlorobenzene to p-chlorophenyl mercapturic acid, which in vitro is converted to p-chlorothiophenol by alkaline hydrolysis. Bradbury and Standen (1959, 1960) thus proposed a similar scheme for the metabolism of y-BHC in insects, with GSH as the likely conjugator. Their proposal was soon verified. Partially purified enzyme preparations from houseflies metabolized pBHC to water-soluble metabolites which yielded isomers of dichlorothiophenol after alkaline hydrolysis. The reaction required GSH (Bradbury and Standen, 1960). The glutathione Stransferase was further purified from flies by Ishida and Dahm (1965a,b)and shown to have a specific requirement for GSH; GSH could not be replaced by several other sulfhydryl compounds. Sims and Grover (1965) demonstrated that housefly supernatant preparations could carry out mercapturic acid Fig. 3. Scheme for the detoxication of p-nitrobenzylchloride in the locust, S. gregaria (Cohen and Smith, 1964. The enzymes y-glutamyl transpeptidase, peptidase and the intermediate S ( p nitrobenzyl) cysteinylglycine have not been determined and are suggested by the reviewer. Both S(p-nitrobenzyl) glutathione and S(p-nitrobenzyl) cysteine may be excreted directly by the locust. In vertebrates, S(p-nitrobenzyl) cysteine undergoes N-acetylation to form S(pnitrobenzyl) mercapturic acid, the main excretory product. Glutathione and cysteine conjugates of arylhalides such as l-chloro-2,4dinitrobenzene,l-fluoro-2,4-dinitrobenzeneand 3,4-dichloro1-nitrobenzene have also been identified in locusts dosed with these compounds.
84
ROBERT P. BODNARYK
synthesis. As final proof, isomeric S-dichlorophenylglutathioneswere detected as major metabolites of y-BHC in houseflies, locusts and ticks (Clark et al., 1966). However, it remained to be determined whether the primary detoxification reaction of y-BHC was conjugated with GSH, or dehydrochlorination of y-BHC to a pentachlorocyclohexene (PCCH) isomer followed by conjugation of the latter with GSH. The problem was resolved decisively by Clark et al. (1969). Their use of inhibitors and colorimetric assays led to the conclusion that a pentachlorocyclohexene is not a major intermediary metabolite of y-BHC in houseflies, blowflies or grass-grubs. y-BHC in these insects is metabolized mainly to a substance having identical chromatographic properties with those of S-2,4-dichlorophenylglutathione.The isotope studies of Bridges (1959) also indicate that y-PCCH is not a major metabolite of y-BHC. Similarly, Reed and Forgash (1970) conclude that the principal metabolic pathway of y-BHC is probably not by way of chlorinated benzenes since these metabolites were found in relatively small amounts in three strains (susceptible, moderately resistant and highly resistant) of houseflies treated with y-BHC. Unfortunately, a standard work on the physiology of insects (Perry and Agosin, 1974, p. 30) still shows the formation of y-PCCH and chlorinated benzenes as major products of y-BHC metabolism in spite of all data to the contrary. Attempts to synergize y-BHC with inhibitors of glutathione Saryltransferase, such as phthaleins and sulphonphthaleins, have not been successful. Although these dyes are excellent inhibitors of the insect enzyme in vitro (Cohen et al., 1964;Clark et al., 1967;Balabaskaran and Smith, 1970) they would not be expected to be effective in vivo since they are ionized at tissue pH and readily excreted (Clark et al., 1969). A causal relationship between resistance to y-BHC and the ability of the resistant strain to detoxify y-BHC by way of glutathione conjugation has never been firmly established in the literature. Glutathione S-transferase activity is found widely in insects, with the housefly being outstanding for its prodigious capacity to metabolize y-BHC. Yet, Ishida and Dahm (1965a,b)have found no correlation between glutathione S-transferase activity and resistance of houseflies to y-BHC. It is possible that detoxication of y-BHC by the action of glutathione S-transferase is an important but supplementary defense mechanism that acts in conjunction with other mechanisms of resistance. The possibility that the uniquely high capacity of houseflies to conjugate y-BHC with GSH is in some way related to their unique capacity to form large amounts of y-glutamyl-phenylalanine from GSH and phenylalanine via the yglutamyl transpeptidase reaction (Bodnaryk, 1970; Bodnaryk and Skillings, 1971;see section 3.2)has never been examined. Indeed, a major question in the study of detoxication enzymes is the relationship of the enzymes of normal intermediary metabolism to those concerned with the detoxication of
STRUCTURE AND FUNCTION OF INSECT PEPTIDES
85
insecticides and other xenobiotics (Clark et al., 1973). Some progress has been made in the identification of the physiological substrates of the glutathione transferases, at least in mammals. Speir and Barnsley (1971) have suggested that 2,3-unsaturated acyl-COA thiol esters might be the normal intermediary metabolites which could serve as substrates in the formation of S-substituted glutathiones commonly found in plant and animal tissue. Recently, the identity of glutathione S-transferase B with ligandin, a major binding protein of rat liver has been established (Habig et al., 1974), suggesting that the protein performs a physiological role as a catalyst as well as in the binding function of cellular transport. It has been pointed out by Boyland et al. (1969) that, since the chlorocyclohexanes and chlorocyclohexens are not planar, it is possible that these and other cycloalkanes are conjugated with glutathione by an enzyme different from glutathione S-aryltransferase (which is concerned with planar, aromatic molecules), and also different from glutathione S-alkyltransferase, which is concerned with relatively simple aliphatic molecules. The enzyme distribution patterns obtained by Ishida (1968) clearly indicate that the housefly glutathione S-aryltransferase which conjugates GSH with l-chloro-2,4dinitrobenzene is not the same as the transferase conjugating GSH with y-BHC and PCCH. Perhaps the pBHC metabolizing enzyme can be appropriately called a glutathione S-cycloalkane transferase. Several glutathione transferases from vertebrates and insects have been isolated by ion-exchange chromatography and electrofocusing by Clark et al. (1973). The activities of glutathione S-crotonyl thioester transferase, glutathione S-aryltransferase and phosphoric acid triester-glutathione-S-methyltransferase were present in a number of forms having different isoelectric points. Although each transferase acts preferentially on its own substrate, some degree of cross-specificity was evident with all three substrates. c Conjugation with organophosphorus insecticides Conjugation reactions involving certain organophosphorus insecticides and glutathione also occur in insects, and are catalysed by a type of glutathione S-aryltransferase. Parathion (0,O-diethyl 0-p-nitrophenylphosphorothioate) is degraded to S(p-nitrobenzene)glutathione and diethyl phosphorothioic acid by a soluble enzyme requiring glutathione (dearylation) (Dahm, 1970; Fig. 4b). Resistant strains of the housefly have an increased capacity to degrade paraoxon (sufficient to account for resistance), attributable to their increased ability to dearylate paraoxon (Nolan and O’Brien, 1970). Diazinon [O,O-diethyl0-(2-isopropyl-4methyl-6-pyrimidyl) phosphorothioatel and diazoxon are degraded efficiently by the soluble fraction from houseflies in the presence of GSH to produce diethylphosphorothioic acid and diethylphosphoric acid and unidentified compounds (Yang et al., 1971). Shishido et al. (1972) found a glutathione S-
86
+
L
+
Pa u
A O? ?X X
u
0
h
W
zX
U?
u I
s
(I)=
0I
rn=a
0-0
u 4
x
c!
2
%c X
u)
0"
8+
a 3
*
s t :
eo 4
X
6
0"
+
G
I
u? X u
X-
ROBERT P. BODNARYK
t b
h
STRUCTURE A N D FUNCTION OF INSECT PEPTIDES
a7
aryltransferase in the fat body of the cockroach (P.americana) and in rat liver glutathione that degraded diazinon to S-(2-isopropyl-4-methyl-6-pyrimidinyl) and diethyl phosphorothioic acid in the presence of GSH (Fig. 4b). Diazoxon and some higher alkyl homologs of diazinon were degraded similarly by the transferase. The requirement for GSH was specific, and GSH could not be replaced by other sulfhydryl compounds. The fat body and rat liver transferases had different properties. A chelating agent, 0-phenanthroline, inhibited the fat body enzyme, but had only a slight effect on the liver enzyme. The relation between the enzyme catalysing the reaction of diazinon with glutathione and glutathione S-aryltransferase, the “typical” substrates of which are aromatic and other cyclic compounds containing labile halogen or nitro groups, is uncertain (Eto, 1974). Glutathione S-aryltransferase is a soluble enzyme found in supernatant preparations of insects. It is by no means the only system that metabolizes a given insecticide. For example, diazinon is also degraded by the microsomal mixed function oxidase system. The increased capacity of this system and of the conjugation reaction in resistant houseflies may largely account for resistance in this species (Folsom et al., 1970; Yang et al., 1971). 3.3.4
Glutathione S-alkyltransferase
Glutathione S-alkyltransferase from rat liver catalyses reactions of glutathione with relatively simple aliphatic molecules such as alkyl halides (Johnson, 1963; rev. Boyland et al., 1969). In insects and mammals a type of glutathione Salkyltransferase is of great importance in the detoxication (dealkylation) of certain alkyl phosphorothioate insecticides. The term phosphoric acid triesterglutathione S-alkyltransferase has been used for the enzyme that catalyses the transfer of a methyl group from a donor (dimethyl phosphoric acid triesters or dimethyl phosphorothionic acid triesters) to the acceptor, glutathione. The transferase type reaction may be regarded as an enzymically catalysed alkylation of glutathione (Hutson et al., 1972). Dealkylation as a detoxication mechanism for alkyl phosphorothioate insecticides was early recognized by Plapp and Casida (1958a,b). The reaction involving glutathione S-alkyltransferase and glutathione that metabolizes methyl parathion to desmethyl parathion and S-methylglutathione in insects and rat liver was firmly established by Fukami and Shishido (1963, 1966) (Fig. 4c). Glutathione-dependent demethylation of several organoFig. 4. Detoxication of some organophosphorus insecticides by glutathione S-aryltransferase (reactions a and b) and glutathione S-alkyltransferase (reactions c, d, e) in insects. (a) Dearylation of parathion (Dahm, 1970); (b) dearylation of diazinon (Yanget al., 1971; Shishido et al., 1972); (c) dealkylation of methyl parathion (Fukami and Shishido, 1963, 1966; Funkunaga et al., 1969); (d) dealkylation of diazinon in resistant houseflies (Lewis and Sawicki, 1971); (e) dealkylation of parathion in resistant houseflies (Lewis and Sawicki, 1971; Oppenoorth et al., 1972).
ROBERT P BODNARYK
88
phosphorus methyl esters has been confirmed in mammalian liver by many authors (Morello et al., 1968; Stenersen, 1969; Hollingworth, 1970; Donninger, 1971). Phosphorothioate insecticides having methyl phosphorus esters are more readily degraded than are those having ethyl or isopropyl phosphorus esters, and especially so in mammals (Plapp and Cassida, 1958a,b; Dauterman et al., 1959; Knaak and O’Brien, 1960; Bull et al., 1963; Fukunaga et al., 1969; Hollingworth, 1969, 1970; Hutson et al., 1972). The substrate specificity of mammalian glutathione S-alkyltransferase may be responsible, at least in part, for the fact that the methyl phosphorothionate insecticides are in general less toxic to mammals than corresponding ethyl homologs (see Eto, 1974). The low mammalian toxicity of fenitrothion [sumithion, 0-0-dimethyl 0-(4-nitro-m-tolyl) phosphorothioatel may in part depend on the efficient 0-demethylation mechanism in mammals (Hollingworth et al., 1967). In the reaction catalysed by glutathione S-alkyltransferase, only one of the two 0-methyl groups is removed from the insecticide; the resulting monodesmethyl compound is not a substrate (Fig. 4c,d,e). Certain geometric isomers of methyl phosphate esters are not degraded by mouse liver glutathione Salkyltransferase, likely for steric reasons (Morello et al., 1968). a Glutathione S-alkyltransferase activity and insecticide resistance Although demethylation of dimethyl phosphoric acid triesters and dimethyl phosphorothionic acid triesters is now a well-recognized detoxication mechanism for many insecticides having this structure, the exact contribution of demethylation relative to other types of degradation mechanisms found in resistant species is not known in quantitative terms. Glutathione S-alkyltransferase activity is apparently higher in organophosphorus-resistant strains of houseflies (Lewis, 1969; Lewis and Sawicki, 1971). However, Oppenoorth et al. (1972) conclude that in houseflies “glutathione-dependent degradation seems to confer only little resistance, at least to parathion”. In studies of the mechanism of azinphosmethyl resistance in a mite, Motoyama et al. (1971) state that the higher rate of demethylation by the resistant mite appears to be responsible in part for resistance. Bull and Whitten (1972) list several factors that are found to be enhanced in organophosphorus resistant tobacco budworms, including mixed-function microsomal oxidases, soluble phosphotriesterase, glucosidic conjugation mechanisms and glutathione-dependent alkyl transferase.
4
Metabolic aspects of peptide pools
In living organisms proteins are continuously being synthesized and degraded. Synthesis begins with free amino acids which are ultimately returned to the free
STRUCTURE A N D FUNCTION OF INSECT PEPTIDES
89
state as the degradation of the protein species proceeds to completion. It seems natural to suppose that during a period of rapid protein turnover, a rather large and heterogeneous pool of peptides must exist, reflecting perhaps both nascent peptides and peptides as hydrolysis fragments from the catabolism of existing proteins. Such “peptide poolsy’have been sought and found in microorganisms and rat liver (Herp et al., 1970) and in Drosophila (Mitchell and Simmons, 1962), although interpretations of their significance have been either highly controversial as in the case of the work by Herp et al. (1970) or unsatisfying. Collett (1976) has reported the presence of a large and heterogeneous pool of small peptides in the blood of adult males of a blowfly, Calliphora erythrocephala, supposedly derived from protein catabolism. Several biochemical facets of the blowfly peptide pool have been examined and some conclusions regarding its physiological status have been drawn. The peptides do not appear to be artifacts of the extraction procedure: clear blood was simply collected into 10 per cent TCA, the precipitated protein removed by centrifuging and the supernatant passed through a Millipore filter to remove final traces of precipitate. They are small peptides, perhaps in the range of 2-5 residues, as judged by their elution profile from a column of Sephadex G-10 and their behaviour on the amino acid analyser. The number of different peptides is undetermined but appears to be very large (in the hundreds, according to Collett) as judged by the multiplicity of [14C1-labelled, hydrolysable peaks observed on the amino acid analyser when extracts of whole flies injected with [l4Cl-glycinewere run. The situation is reminiscent of an earlier study of peptides in whole extracts of Drosophila larvae where the number of peptides is said to be at least 600 (Mitchell and Simmons, 1962; Simmons and Mitchell, 1962). The blowfly [14C1-labelledpeaks do not have corresponding ninhydrin-reacting peaks on the amino acid analyser, presumably because the individual peptide species are present in very low concentrations. In my estimation, many of the [l4C1-labelledpeaks may not be peptides at all, but acid-labile compounds that fortuitously elute with amino acids and small peptides and which derive their carbon label from the metabolically ubiquitous 14C1-glycine.However, the small size of the peptides and their apparent large numbers argue against the peptide material being merely trace amounts of protein not precipitated by TCA. Amino acid analysis of the bulk peptides indicates that all amino acids are present, but in widely different amounts and that glycine and glutamic acid and/or glutamine together account for about 50 per cent of the peptide amino acid. The peptides undergo rapid turnover as seen from a 60 per cent decline in peptide radioactivity during a 12 h period after radioactivity had been introduced into the pool using [I4CI-glycine. (Undoubtedly, some recycling of label occurs during this type of experiment, giving an underestimate of the actual turnover rate.) The likely source of the blood peptides is protein
90
ROBERT P. BODNARYK
catabolism, and the dynamics of protein turnover appear to correspond to increases in peptide levels in the blood. Collett (1976) has also studied the blowfly peptidases. Significantly, the blood contains peptidase activity. Five electrophoretically distinct molecular species of peptidase capable of hydrolysing leucyl-alanine were detected in the blood. Di- and tripeptides injected into the blood are also hydrolysed and at widely different rates, reflecting the specificity of the blowfly peptidases. The peptidases are inhibited noncompetitively by free amino acids as determined in vitro studies of the hydrolysis of lysine-p-nitroanilide and leucine-p-nitroanilide i n the presence of individual free amino acids. Greatest inhibition was observed with essential amino acids, notably leucine, isoleucine, phenylalanine and methionine. Collett has calculated that the combined effects of the essential amino acids at their measured concentration in the blood would inhibit peptidase activity by 15 per cent, whereas the nonessential amino acids would inhibit activity by only 6 per cent. Collett (1976) suggests that the peptide pool in blowfly blood represents a reservoir of essential amino acids, protected from metabolic transformation and excretion by virtue of their peptide linkage, and held in reserve until metabolic demand for them arises. The principle of storage of essential amino acids in the protected form of a peptide has been established for the amino acids tyrosine and phenylalanine in two insect genera by other workers. The dipeptides /I-alanyl-tyrosine in S. bulluta (Levenbook et ul., 1969; Bodnaryk and Levenbook, 1969) and y-glutamyl-phenylalanine in M. domesticu (Bodnaryk, 1970c; see section 2.3) represent sequestered forms of aromatic amino acids that are accumulated in the blood of the growing larva of these species and are utilized only at the end of the larval period to produce quinones for the sclerotization of the puparium. Collett’s work would thus seem to extend this principle to include the more general case of sequestration of essential amino acids in peptide form in insect blood, the sequestered amino acids being eventually released to the free form to participate again in protein synthesis. Nonessential amino acids are of course also sequestered, adding further economy to the system. The pool of di- and tripeptides in the blood of late second and early third instar larvae of the blowfly, P. reginu, may also serve a similar function, although no evidence has been presented in support of this possibility. Since the structure of nine of these peptides has been established (Bodnaryk and Levenbook, 1968; see section 2.1), the young blowfly larva may be a useful experimental animal for further studies on the storage of amino acids as blood peptides. According to Collett, the activity of blood peptidases is governed by the level of amino acids in the blood, essential amino acids making the largest contribution to noncompetitive inhibition of peptidase activity. This is an appealing hypothesis, for it comes to grips with the problem of amino
STRUCTURE AND FUNCTION OF INSECT PEPTIDES
91
acid/peptide homeostasis. Stated simply, periods of active feeding on meat would tend to raise the level of essential amino acids in the blowfly’s blood, thereby depressing peptidase activity and sparing blood peptides. Nonfeeding periods would lead to a decrease in blood amino acids, activation of blood peptidases and hydrolysis of peptides. However, the experimental evidence that supports such a homeostatic mechanism, while convincing, is based largely on in uitro studies. It has not been demonstrated in viuo that a change in the concentration of amino acids in the blood leads to a change in peptidase activity and altered rate of peptide hydrolysis. Thus, crucial tests of Collett’s hypothesis have not been made in uiuo. These tests are, however, well within the scope of future experimentation. One need only mention the possibilities of examining the effects of experimental alteration of the level of blood amino acids by injection, or of dietary manipulation of the amino acid contents of chemically defined diets, on the activity of blood peptidases and the turnover rate of the peptide pool. Future studies may also concentrate on the structure of individual peptides in the pool and the specificity of the various multiple forms of blood peptidases towards these peptides. 5 . Sex Peptides from Drosophila
The act of mating causes fundamental changes in the physiology and behaviour of the female insect. The two principal alterations that have been documented in mated females from many species are: (a) increased egg-laying, and (b) decreased receptivity towards the male (Chen, 1971; de Wilde and de Loof, 1973; Leopold, 1976). In Diptera, the search for the mechanisms whereby mating enhances fecundity and decreases receptivity has centred about the accessory gland fluid transferred by the male during copulation. The active principles in the accessory gland secretion have been investigated in only a few species. In Musca domestica a low-molecular-weight polar substance isolated from male copulatory ducts prevented mating when it was injected into virgin females (Adams and Nelson, 1968; Nelson et al., 1969). The chemical identity of the monocoitic substance(s) remains unknown. In Aedes aegypti a protein dimer termed “matrone” occurs in the male accessory glands. Its subunits, a and p, have approximate molecular weights of 60 0oO and 30 000 daltons (Fuchs et al., 1968, 1969; Fuchs and Hiss, 1970). The a subunit stimulates egg-laying whereas both subunits are needed to cause monogamy (Hiss and Fuchs, 1972). The large sue of the matrone molecule would seem to render its further characterization in terms of amino acid sequence and threedimensional configuration a formidable task. A somewhat simpler situation exists in some species of Drosophila, where peptides and other low-molecularweight substances have been identified as important factors in the accessory glands of males.
92
5.1
ROBERT P BODNARYK
THE SEX PEPTIDE FROM “ D . MELANOGASTER”
The discovery of a male specific peptide in Drosophila can be attributed to Fox (1956) who observed a hydrolysable, ninhydrin-positive spot on chromatograms of the body fluid of males but not females of D. melanogaster. Fox et al. (1959) termed this new male specific substance a “sex peptide” and progress on its purification, amino acid composition and genetics was presented in abstract by Fox et al. (1962). Chen and Diem (1961) pinpointed the male accessory glands (paragonia) as the site of production of the sex peptide. This discovery had important consequences for further studies of the physiological function of the sex peptide, for, a year earlier, Kummer (1960) had suggested that the secretion of the accessory glands probably had a stimulatory effect on oviposition. His view has been confirmed by either transplanting these glands (Garcia-Bellido, 1964; Leahy, 1966; Merle, 1969) or injecting their extracts (Leahy and Lowe, 1967) into virgin females. It became a matter of course to test the sex peptide for possible effects upon Drosophila females. Such tests required sex peptide in sufficient amounts and purity. Chen and Biihler (1970) isolated the sex peptide from a large number of male D. melanogaster adults by preparative ion-exchange chromatography. Injection of the isolated and purified peptide into virgin females resulted in a two- to three-fold increase in the number of eggs laid as compared to virgin females without injection or with saline injection. The stimulatory activity appeared to depend on the amount of material injected, and a single injection was sufficient to maintain fecundity at a high level during the entire period of egg-laying. Moreover, small amounts of the sex peptide could be detected in mated females (but never in virgins), suggesting that the peptide is introduced into the female by the male at mating. Structural studies of the sex peptide from D. melanogaster are incomplete to date. Substantial amino acid heterogeneity has been found in the material (Chen, personal communication), unlike the minor heterogeneity found in the peptide PS-1 from D.funebris (section 5.2.1). 5.1.1
Genetic control of synthesis
The relative ease of genetic manipulation in Drosophila has led to studies of the genetic control of sex peptide synthesis. Sex peptide synthesis occurs in 1X/2A males and 2X/2A females transformed to sterile males by the recessive mutant transformer (Fox et al., 1962). It also occurs in 1 and 2X zygotes converted into intersex adults by the third chromosome gene doublesex (dsx) (Smith and Bischoff, 1969). These results indicate that sex peptide synthesis is not dependent on the presence of the Y chromosome, a conclusion reached earlier
STRUCTURE A N D FUNCTION OF INSECT PEPTIDES
93
by Fox (1956). Chen and Diem (1961) have demonstrated the autonomous synthesis of the sex peptide in female D. melanogaster adults which had received male genital discs in the larval stage by transplantation. 5.2
ps-1
AND
PS-2 FROM
“D. FUNEBRIS”
Two sex-specific, ninhydrin-positive substances, PS- 1 and PS-2, have been isolated from the accessory glands of D. funebris (Baumann and Chen, 1973; Baumann, 1974a,b). PS-1, a 27-residue peptide (Baumann et al., 1975), reduces the receptivity of virgin females towards males, whereas PS-2, a glycine-carbohydrate derivative (incompletely characterized to date) stimulates egg-laying. The sex peptide from D. melanogaster, in contrast to the PS-1 peptide from D. funebris, stimulates egg-laying and reduces virgin female receptivity (Chen, 1975, personal communication). The situation with respect to accessory gland substances in the Drosophila subgenus is evidently complex, and there is no reason to expect that these compounds function in an identical manner in all species. Indeed, in a survey of more than ten Drosophila species, no trace of PS-1 and PS-2 was found in many of the species (Chen, personal communication). PS-1 and PS-2 in D. funebris males have been [l4CI-labelled and their fate after introduction into the female by natural mating has been followed (Baumann, 1974b). Two hours after copulation, the [l4C1-labelledPS-1 and PS2 were found throughout the female, indicating their distribution via the blood. Both substances have a relatively short half-life in the female, and are virtually undetectable 24 h after being introduced during mating. The target($ of PS-1 and PS-2 in the female remain unknown. Identification of the target tissue(s) of these accessory gland substances will be an essential step in determining their mode of action and in understanding the integration of this action with the activity of the brain and corpora allata. 5.2.1
Structure of the PS-1peptide from D. funebris
The complete amino acid sequence of PS-1 from wild type D . funebris has been established by Baumann et al. (1975). In their study, 26.9 mg of peptide were isolated and purified from an extract of 40 g of male flies, representing some 68.8 per cent total recovery of biological activity. The amino acid sequence was determined by manual sequence analysis of tryptic peptides, automated Edman degradation and carboxypeptidase A digestion. The sequence of PS-1 (Fig. 5 ) contains some notable features. The eicosaheptapeptide has a high alanine content with an alternating sequence of Ala-Asn which occurs four times within the 27 amino acid residues. Absent from the structure
ROBERT P. BODNARYK
94
are aromatic and sulfur-containing amino acids, glycine, histidine and isoleucine. The Val :Leu heterogeneity in the second position of the peptide occurs in the ratio of 7 :3. Partial separation of the two forms of the peptide has been achieved during chromatography on Dowex 50. Fractions differing in their ratios of Val :Leu elicit identical biological response. Both chains, which may be designated valine-PS-1 and leucine-PS-1 are synthesized by each individual fly (Baumann, 1974). V a1 Asp
Pro Ser Ala Asn Ala Asn Ala Asn Asn Gln Arg Thr Ala Ala Ala Lys Pro
Leu Gln Ala Asn Ala Glu Ala Ser Ser
Fig. 5. Amino acid sequence of P S I , a peptide from the accessory gland secretion of D.funebris (Baumann et al., 1975).
The evidence that PS-1 is responsible for reducing the receptivity of the female of D.funebris is now very convincing. However, the definitive statement that biological activity is due solely to PS-1 (and not a contaminating peptide, for instance) obviously must await chemical synthesis of the peptide structure of PS- 1 and its bioassay.
5.3
SEX PEPTIDES I N OTHER “DIPTERA”
The apparent absence of sex peptides in several Drosophila species (Chen, 1975, personal communication) would seem to indicate4hat peptides of this nature might not be found in other insect genera. However, Balogun (1974) has reported the preliminary identification of a sex-specific, ninhydrin positive component in the accessory gland of the tsetse flies Glossina morsitans and G. palpalis. Further reports on the nature of this substance and its function in the male accessory gland of the tsetse fly are awaited with interest.
6
Proctolin, a proposed neurotransmitter in insect visceral muscle
Research carried out over the past decade by Brown and his coworkers has culminated in the isolation, identification and synthesis of a pentapeptide, termed proctolin, which might function as an excitatory neurotransmitter in the visceral muscles of insects. It was originally suggested by Brown (1967) that a myotropic substance associated with an efferent pathway of the proctodeal (hindgut) innervation of the cockroach Periplaneta americana might function as an excitatory
STRUCTURE A N D FUNCTION OF INSECT PEPTIDES
95
transmitter substance in insect visceral muscle. The active substance did not appear to be any of the established or putative transmitters including glutamate, the proposed excitatory mediator in the fast, skeletal fibres of insects (rev. Usherwood, 1974; McDonald, 1975). Holman and Cook (1972) identilied glutamic acid, aspartic acid and a small basic peptide (in all probability, proctolin) in active extracts of the hindgut of the roach Leucophaea maderae. However, these authors maintained that their evidence favoured glutamate as the most likely prospective chemical mediator at the visceral excitatory myoneural junction (Holman and Cook, 1970; Cook and Holman, 1975) and that the peptide (termed “hindgut-stimulating neurohormone”, HSH, by them) functioned as a neurohormone involved in the regulation of visceral muscle activity (Holman and Cook, 1972), perhaps by regulating calcium entry into the muscle fibre (Cook and Holman, 1975). The resolution of these conflicting points of view depended, among other things, upon the identification of the hindgut peptide. (a) Arg Pro Lys Pro Gln Gln Phe Phe Gly Leu Met NH,
(b) Arg Tyr Leu Pro Thr Fig. 6. Structure of two putative neurotransmitter peptides. (a) Substance P from mammals, a proposed excitatory transmitter of the primary afferent neurones in the spinal cord (Chang et al., 1971); (b) proctolin from insects, a proposed excitatory transmitter in insect visceral muscle (Starratt and Brown, 1975).
Proctolin was isolated by Brown and Starratt (1975), and sequenced and synthesized by Starratt and Brown (1975). Moreover, the nature of the innervation and the bioelectrics underlying the mechanical activity of rectal muscle fibres in Periplaneta has been thoroughly established (Belton and Brown, 1969; Brown and Nagai, 1969; Nagai and Brown, 1969; Nagai, 1970, 1972, 1973). These studies, in combination with several lines of pharmacological evidence recently obtained with pure proctolin (Brown, 1975), have strengthened Brown’s position that proctolin functions as an excitatory neurotransmitter, and that glutamate is not the excitatory transmitter, at least not in those motor neurones which originate in the central nervous system and constitute the main or sole excitatory innervation of the longitudinal muscle straps on the rectum of Periplaneta. Final acceptance of the view that insects possess a distinct class of motor nerve cell in which synaptic transmission is mediated by proctolin will no doubt rest on the results of iontophoretic experiments and confirmation of the existence of peptidergic terminals at the ultrastructural level (Brown, 1975). The structure of proctolin as determined by Starratt and Brown (1975) is given in Fig. 6. Their preliminary structureactivity studies of closely related
96
ROBERT P. BODNARYK
synthetic analogues of the basic pentapeptide indicate that the C-terminal threonine residue is essential for full activity. Removal of threonine or replacement by serine or threonine amide led to substances with less than 1 per cent of the activity of proctolin. Removal of arginine also resulted in substantial loss of activity (Starratt and Brown, 1975). Tyramine (but not dopamine) behaves as a competitive antagonist, giving rise to the speculation that the phenol group of tyramine has considerable affinity for the same receptor site as the phenol group of the tyrosine residue of proctolin (Brown, 1975). There is a growing body of evidence in mammals that peptides may function as neurotransmitters (von Euler and Gaddum, 1931; Bargmann et al., 1967; Chang et al., 1971; Tregear, 1971; Nicoll and Barker, 1971; Otsuka et al., 1972; Takahashi et al., 1974; Konishi and Otsuka, 1974; Iversen, 1974). Proctolin as the putative excitatory transmitter in insect visceral muscle is the first such peptide to be characterized in insects. Representative species from six insect orders all contained a substance having pharmacological, chromatographic and electrophoretic properties identical to those of proctolin, suggesting that proctolin may be a universal constituent of the Insecta (Starratt and Brown, 1975).
7
Physiologically active peptides from the corpus cardiacum
The insect corpus cardiacum is a neuro-haemal organ (Scharrer, 1952) analogous to the crustacean x-organ-sinus gland complex and to the vertebrate hypothalamo-hypophyseal system (Hanstrom, 1953). Both of these latter systems contain pharmacologically active substances, the most thoroughly characterized being the neurohypophyseal peptides oxytocin and vasopressin. The corpus cardiacum of the roach Periplaneta americana and the locusts Schistocerca gregaria and Locusf a migratoria migratorioides has been widely studied and found to contain and release into the haemolymph a variety of peptides with different physiological activities. Activities attributed to corpus cardiacum peptides are: (1) heart-accelerating; (2) increasing the rate of excretion in the Malpighian tubules; (3) “SFl” activity, which can be defined as increasing the frequency of the spontaneous firing recorded from nerves leaving the sixth abdominal ganglion of P. americana (see Natalizi et al., 1970); (4) hyperglycaemic and (5) adipokinetic. Until recently, it was quite impossible to decide from the literature exactly how many different physiologically active peptides had actually been uncovered in corpora cardiaca extracts, the extent of their purity, or even whether there existed single peptide species with, for example, dual heartaccelerating and hyperglycaemic activity or whether these activities were due to an unresolved mixture of two or more species. The answers to some of these
STRUCTURE A N D FUNCTION OF INSECT PEPTIDES
97
questions can now be given with confidence as progress continues to be made in the chromatographic resolution of cardiaca peptides. Unfortunately there exists no accepted and systematic nomenclature to specify the identity of these peptides, a factor that continues to be counterproductive to efforts to compare and evaluate results from various laboratories. None of the structures of the cardiaca peptides is known to date, in spite of a voluminous literature dealing with them (recent revs. by Maddrell, 1974; Goldsworthy and Mordue, 1974). One seldom sees the trees for the thickness of the forest in dealing with their literature. Nevertheless, it seems probable that the structures of the heartaccelerating peptides and the hyperglycaemic peptide hormone will become known before too long, and progress in this area is dealt with in this review. Much less is known about the adipokinetic hormone. Its peptide nature can only be surmized in the near total absence of data concerning its molecular properties. The reader is referred to Mordue and Goldsworthy (1974) for a summary of its physiological activities and possible function. 7.1 7.1.1
HEART-ACCELERATING PEPTIDES
Neurohormone D; peptide P,; peak 1;factor C
A cardioaccelerator termed neurohormone D has been isolated from the corpora cardiaca of P. americana initially by paper chromatography and later by a combination of paper chromatography and gel filtation. It has been studied in detail by Gersch and his associates (Unger, 1957; Gersh et al., 1960, 1963, 1969;. Gersch and Richter, 1963; Richter, 1967; Gersch and Sturzebecher, 1967; Bauman and Gersch, 1973). Neurohormone D probably corresponds to Brown’s (1 965) heart-accelerating peptide P, and undoubtedly is the same substance obtained by gel filtration and termed “peak 1” by Natalizi et al. (1970) and Traina et al. (1974). Other reports of cardioaccelerator peptides from corpora cardiaca extracts from P. americana (Davey, 1961a; Ralph, 1962) are insufficient in detail for direct comparison with neurohormone D. It is evident, however, that neurohormone D must have been present in their extracts. “Factor C” from the storage lobe of the corpus caridacum of the locusts S. gregaria and L. migratoria is physiologically similar in its effects to neurohormone D (Mordue and Goldsworthy, 1969). Neurohormone D has been characterized as a peptide on the basis of its inactivation by trypsin. It has an approximate molecular weight of 2000 (Gersch and Sturzebecher, 1967) and is relatively heat-labile and acid-labile (Baumann and Gersch, 1973); Traina et al., 1974, 1976). Amino acid analysis of “peak 1” which is said to be identical with neurohormone D by Natalizi et al. (1970) and Traina et al. (1974) has proved to be premature, for “peak 1” has been resolved into two peaks (la and lb) using a long (80 cm) column of
98
ROBERT P EODNARYK
Sephadex G-15 by Traina et al. (1974, 1976). The amino acid analysis is not without value, however, as it demonstrates the absence of sulfur-containing amino acids from this cardioaccelerator preparation (see also Baumann and Gersch, 1973). The point is an important one, for several authors have speculated on the presence of sulfur in neurosecretory peptides of insect corpora cardiaca on the basis of histochemical reactions. Further progress on the structure of neurohormone D, which now can be resolved by efficient gel filtration into two distinct entities, each having heartaccelerator activity and each being inactivated by trypsin, will without doubt be soon forthcoming. 7.1.2
Peak 2; neurohormone C (?);peptide P,;factor B
A second heart-accelerating peptide from corpora cardiaca extracts from P. americana which elutes as “peak 2” from columns of Sephadex G-15 has been investigated intensively by Natalizi et al. (1970) and Traina et al. (1976). It is difficult to assign a correspondence of peak 2 to the earlier literature on cardioaccelerator peptides. The material may correspond to Brown’s (1965) peptide P, since partially purified preparations of both peak 2 and peptide P, appear to have been contaminated by another peptide having hyperglycaemic activity. Peak 2 may correspond to “neurohormone C” of Gersch et al. (1960) but their identities has not been substantiated experimentally. “Factor B” from the glandular lobe of the corpus cardiacum of S . gregaria and L. migratoria is physiologically similar in its effects to neurohormone C (Mordue and Goldsworthy, 1969). a Isolation and amino acid composition Peak 2 has been resolved into two peaks (2a and 2b) by gel filtration on Sephadex G-15 and further purified by chromatography on a column of Sephadex LH-20. It is noteworthy that these preparations are devoid of hyperglycaemic activity (Traina et al., 1976). The heart-accelerating activity of both 2a and 2b is unaffected when they are incubated with trypsin containing a chymotrypsin inhibitor (TPCK), but activity is completely destroyed when incubated with chymotrypsin. An earlier report (Natalizi et al., 1970) indicating that “peak 2” was inactivated by trypsin is thus in error, no doubt due to the use of a trypsin preparation containing uninhibited traces of chymotrypsin. Amino acid analyses of peaks 2a and 2b gave nonintegral values for the amino acid content of each peptide, indicating a certain heterogeneity of the peptide preparations. If leucine is taken as unity in the molar proportions and it is assumed that there is only one tryptophan residue, factor 2b contains 12 amino acid residues including lysine, aspartic acid, serine, glutamic acid, proline, glycine, alanine, valine, and leucine and has a molecular weight of
STRUCTURE A N D FUNCTION OF INSECT PEPTIDES
99
1287.5. Factor 2a is quite similar, containing 17 residues including also threonine and phenylalanine and has a molecular weight of 1795.1. If isoleucine is taken as unity in the molar proportions, the peptides would be much larger, and also contain histidine, arginine, isoleucine and tyrosine as well as threonine and phenylalanine in both. Traina et al. (1976) favour the first interpretation of their amino acid analyses, as it gives a molecular weight consistent with the elution behaviour of the peptides on Sephadex G-15. Until the question of the heterogeneity of 2a and 2b is settled, however, no firm statement on the size or amino acid composition of these peptides can be given. Earlier it was mentioned that various semipurified fractions from paper chromatography or gel filtration columns possessed both heart-accelerating and hyperglycaemic activity, raising the possibility that there existed single peptide species with both activities. Such dual activity is not without precedent and one need only mention noradrenaline, 5-hydroxytryptamine and the polypeptide hormone glucagon as examples of substances active on the heatbeat frequency and glycogenolysis in vertebrates (Farah and Tuttle, 1960). However, heart-accelerating activity has been separated from hyperglycaemic activity in cardiaca extracts (Gersch et al., 1960; Natalizi and Frontali, 1966) and it is now clear that the heart-accelerating peptides 2a and 2b are entities quite distinct from the hyperglycaemic peptides (Traina et al., 1976). 7.1.3
Origin, site ofsynthesis and release
The problem of the origin of the corpus cardiacum peptides has been investigated from various angles by several authors. Mordue and Goldsworthy (1969) have taken advantage of the fact that in locusts the corpora cardiaca are composed of two distinct lobes, which have been designated histologically as “glandular” and “storage” by Highnam (196 1). The glandular lobe is the source of “factor B” (=physiological activity of neurohormone C in P. americana) which increases the heart-beat frequency and decreases the amplitude. The storage lobe is the source of “factor C” (=physiological activity of neurohormone D in P. americana) which produces a smaller increase in frequency but increases the amplitude. The site of synthesis of the heart-accelerating peptides is less clear. Mordue and Goldsworthy (1969) maintain that “factor C” is produced by the neurosecretory cells of the pars intercerebralis, a situation that would parallel the synthesis of neurohormone D by the cerebral neurosecretory cells according to Penzlin (1966). On the other hand if the peptides found within the corpus cardiacum are wholly or partly synthesized within the neurosecretory cells of the pars intercerebralis, some biological activity would be expected from extracts of brain containing these cells and their axons. On this subject Mordue and Goldsworthy (1969) find little or no activity in such extracts. A comparison between the amount of
100
ROBERT P. BODNARYK
heart-accelerating activity in the corpus cardiacum, brain, and heart of P. americana gave values of 10,2.5 and 0.6 units respectively per organ or pair of organs (Natalizi el al., 1970). Such data of course do not establish the site of synthesis at all: the high activity found in the corpus cardiacum may be the result of an accumulation in axons and nerve endings located in the corpus cardiacum of peptides synthesized in the brain and transported to the cardiaca. Another approach to the site of production of the heart-accelerating peptides has been that of Kater (1968) and Scharrer and Kater (1969) who observed that electrical stimulation through the nervous corporis cardiaci I elicited a massive release of heart-accelerating activity into the incubation medium. Ultrastructural studies of stimulated glands indicated that the heartaccelerating activity seemed to be derived from large neurosecretory granules stored within the corpus cardiacum, with the site of origin being the perikarya of intrinsic or extrinsic neurosecretory neurons or possibly both. Gersch (1 974a) has found that stimulation of the various ganglia of the nerve-cord of P. americana in vitro causes selective release of myotropic neurohormones. Stimulation of the thoracic ganglia causes the release of a heart-decelerating factor, and stimulation of the sixth abdominal ganglion releases the peptide neurohormone D. The presence of eserine in the bathing medium increases the release of neurohormone D during stimulation, whereas atropine inhibits release, suggesting that the release of neurohormone D is regulated by cholinergic agents. Neurohormone D exhibits a dose-dependent stimulatory effect on the activity of the neurosecretory axons, but has no effect on the motor axons of the lateral cardiac nerve of Blaberus craniifer. The action of neurohormone D is said to occur only in the presence of an intact cholinergic system (Richter and Gersch, 1974). 7.1.4
Physiologicalfunction
It seems a paradox to caution against assuming that the in ubo physiological role of the heart-accelerating peptides is in the regulation of heart beat. Such a caveat has been expressed by Brown (1965) and Mordue and Goldsworthy (1969) and not without justification: proof is lacking for such a role. In this connection Brown (1965) notes that the mammalian cyclopeptides oxytocin and vasopressin have characteristic smooth muscle activity, but also possess antidiuretic and milk-ejecting activity (Kleeman and Cutler, 1963). Significantly, injection of corpus cardiacum extracts into insects has no effect on heart beat (Mordue and Goldsworthy, 1969; Roussel and Cazal, 1969: Norman, 1972). Extracts of the corpus cardiacum affect the contraction of many insect visceral muscles (Davy, 1964) and these effects, as well as those upon the beating of isolated hearts in uitro, may be pharmacological properties of the peptides. The function of the peptides from the corpus cardiacum with heart-
STRUCTURE AND FUNCTION OF INSECT PEPTIDES
101
accelerating activity, whether it be in the regulation of heart beat, or in the regulation and integration of various other processes, remains unproven.
7.2
THE HYPERGLYCAEMIC HORMONE
A hyperglycaemic hormone comparable to the vertebrate peptide hormone glucagon was first demonstrated by Steele (1961; 1963) in the corpus cardiacum of Periplaneta americana. His observations have been extended to many other insects, notably: the cockroaches P. americana (Ralph and McCarthy, 1964; Brown, 1965; Natalizi and Frontali, 1966; Natalizi et al., 1970), Blaberus discoidalis (Bowers and Friedman, 1963), Leucophaea maderae (Wiens and Gilbert, 1967); the stick-insect Carausius morosus (Dutrieu and Gourdoux, 1967); the locust L . migratoria (Goldsworthy, 1969; Highnam and Goldsworthy, 1972); the black blowfly Phormia regina (Friedman, 1967), the blowfly Calliphora erythrocephala (Norman and Duve, 1969; Vejbjerg and Norman, 1974; Norman, 1975); the bee Apis mellifera (Natalizi and Frontali, 1966) and the moth Manduca sexta (Tager et al., 1975). Recent reviews on insect hormones (Wyatt, 1972) and neurosecretory hormones in insects (Goldsworthy and Mordue, 1974) have dealt with the hyperglycaemic hormone but without much emphasis on its structure. 7.2.1
Structure
Structural studies of the insect hyperglycaemic hormone have lagged far behind those of its vertebrate counterpart, glucagon. Like glucagon, the activity of the hyperglycaemic hormone is destroyed by trypsin (Natalizi and Frontali, 1966; Natalizi et al., 1970). Studies of the elution behaviour of the insect hormone on BioGel P-10 or Sephadex G-25 indicate a low molecular weight, although chromatographic effects have prevented an accurate estimation (Natalizi and Frontali, 1966). The elution volume of immunoreactive glucagon from the corpora cardiaca of the moth, Manduca sexta, on BioGel P-10 corresponds to that of a 4500-dalton peptide, some 30 per cent larger than mammalian glucagon (Tager et al., 1975). The hyperglycaemic activity of the corpus cardiacum of P . americana has been resolved into two fractions by paper chromatography (Brown, 1965) or by ion-exchange chromatography on a column of SE-Sephadex. About one half of the activity was eluted from the column by 0.01 M formic acid, the remainder not until a 1.0 M concentration was reached, indicating the highly-acidic nature of this fraction. No hyperglycaemic activity was retained by the anion-exchanger DEAE Sephadex (Natalizi and Frontali, 1966). In L. migratoria, two fractions with hyperglycaemic activity are also resolved by paper chromatography, one each from the glandular and storage lobes of the corpus cardiacum (Mordue and
102
ROBERT P. BODNARYK
Goldsworthy, 1969). Differences in the physicochemical properties and biological activities of the two fractions are virtually unexplored. An indirect approach to the structural properties of the hyperglycaemic hormone has been made by Tager et af. (1975) who have identified a highly acidic peptide with glucagon-like immunoreactivity from the corpora cardiaca of M. sexta. Although it would seem unlikely at first glance that the insect hyperglycaemic peptide would be reactive towards antibodies directed towards mammalian glucagon, Tager et af. (1975) argue skillfully on the basis of the extreme conservation of structure of vertebrate glucagon during evolution (Bromer et af., 1957, 1971; Thomsen et af., 1972; Markussen et af., 1972; Sundby et af., 1972; Fig. 7) that a glucagon-like peptide might have changed by only 15 per cent during the approximately 900 Myr since the divergence of insects and mammals. Their conclusion that neither the estimated extent of change nor the apparently higher molecular weight of the insect peptide would His Ser Gln Gly Thr Phe Thr Ser Asp Tyr Ser Lys Tyr Leu Asp Ser Arg Arg Ala Gln Asp Phe Val Gln Trp Leu Met Asn Thr Fig. 7. Amino acid sequence of porcine glucagon (Bromer et al., 1957). The amino acid sequences of bovine (Bromer et al., 1971) and human (Thomsen et al., 1972) glucagon are identical. Turkey glucagon differs only at position 28, where an asparagine residue is substituted by a serine residue (Markussen et al., 1972). Duck glucagon (Sundby et al., 1972) differs only at positions 16 (threonine substitutes for serine) and 28 (serine substitutes for asparagine). An extreme conservation of structure during evolution is evident. Glucagon-like immunoreactivity towards porcine anti-glucagon serum has been detected in the corpus cardiacum of the adult sexta by Tager et al. (1975). tobacco hornworm,7M.
~
be likely to preclude reactivity with antibodies against mammalian glucagon is supported by their finding of glucagon-like immunoreactivity in insect corpora cardiaca with properties and tissue distribution as follows: (1) The tissue distribution of the glucagon-like immunoreactivity in M. sexta parallels the distribution of the hyperglycaemic hormone in other insects in so far as it is known (Steele, 1961; Goldsworthy, 1969; Ralph and McCarthy, 1964; Highnam and Goldsworthy, 1972). (2) Both immunoreactivity and hyperglycaemic activity are sensitive to tryptic digestion and behave similarly during gel filtration at neutral pH (Natalizi and Frontali, 1966). (3) The high acidity of the immunoreactive peptide is also in keeping with the known acidity of the insect hyperglycaemic hormone (Natalizi and Frontali, 1966; Natalizi et af., 1970). The chemical and physical similarities thus suggest that the insect hyperglycaemic hormone and the glucagon-like peptide of Tager et al. (1975) are the same. Final proof, however, awaits the isolation and sequencing of the hyperglycaemic peptide. Tager’s immunoreactivity study, apart from
STRUCTURE AND FUNCTION OF INSECT PEPTIDES
103
emphasizing the structural similarity between an insect hyperglycaemic hormone and mammalian glucagon, may have its ultimate value in the development of a simple in vitro assay for the hyperglycaemic hormone. The nearly identical physiological action of the insect and vertebrate hormones will be described in the following sections. 7.2.2
Site of synthesis
The extra-cardiacal synthesis of the hyperglycaemic hormone runs like a theme through the literature dealing with neurosecretory hormones in insects. In P. americana it has been suggested that the hyperglycaemic hormone is synthesized in the neurosecretory cells of the brain, transported via the nervous corporis cardiaci internus (NCCI) to the corpus cardiacum where it is stored, and eventually released into the blood (Ralph and McCarthy, 1964; Wiens and Gilbert, 1967). Although the experimental evidence for this proposed sequence of events is meagre, the scheme is consistent with other experimental findings, especially the fact that glycogen accumulates following neurosecretory cell cautery in many insects (Wyatt, 1967). If the brain is a major site of synthesis of the hyperglycaemic hormone, transport of the hormone must be rapid and continuous, for brain invariably contains very little hyperglycaemic activity (Steele, 1961; Ralph and McCarthy, 1964; Luscher and Leuthold, 1965; Mordue, 1969; Mordue and Goldsworthy, 1969). In L. migratoria there is experimental evidence suggesting that the hyperglycaemic, phosphorylaseactivating peptide found in the storage lobes of the corpora cardiaca is probably synthesized in the neurosecretory cells of the brain, whereas the peptide found in the glandular lobes may be an intrinsic product of the cardiacum (Mordue and Goldsworthy, 1969; Goldsworthy, 1970; Highnam and Goldsworthy, 1972). 7.2.3
Physiological activity
The hyperglycaemic hormone appears to exert its effect by activating fat body phosphorylase (a1,4-glucan :orthophosphate glucosyltransferase; EC 2.4.1.1) (Steele, 1963; Mordue and Goldsworthy, 1969) which in its turn causes a depletion of fat body glycogen as evident in vivo (Steele, 1963) or in vitro (Ralph, 1962; Wiens and Gilbert, 1967a; Friedman, 1967). Glycogen of muscle and gut tissue is not mobilized, but glycogen deposits of the ventral nerve cord of P. americana are also depleted by corpus cardiacum extracts (Steele, 1963; see below). Fat body phosphorylase is also activated by cyclic AMP (Steele, 1964). Increased glycogenolysis in hormone-treated fat body in vitro is accompanied by reduced glycogen synthesis from [l4C1glucose (Wiens and Gilbert, 1967a), a situation analagous to the mammalian cyclic AMP-
104
ROBERT P. BODNARYK
regulated system (Robison et al., 1968). Hormone induced glycogenolysis results in elevation of blood trehalose, the predominant sugar in most insects (Wyatt, 1967), the magnitude and time course of the response being dependent on the species studied, its nutritional state and the amount of hormone used. In P. americana there is an increase in blood trehalose within 30min of the injection with a maximal response (about 2.5-fold elevation in blood trehalose produced by 0.1 gland-pairs of corpora cardiaca) occurring within one hour. Elevated levels of trehalose are maintained for nearly 12 h and return to normal after 48 h (Steele, 1963). In L. migratoria, a minimum of 1.5 glandpairs are needed to produce a maximal rise in blood carbohydrate which occurs two hours after the injection. Blood carbohydrate levels return to near normal after four hours (Goldsworthy, 1969). In the latter case it is not known whether the short duration of the hyperglycaemic response is due to a return of active phosphorylase levels to normal after five hours, or due simply to exhaustion of the glycogen reserves of the fat body. If P. americana (Ralph and McCarthy, 1964) or L. migratoria (Goldsworthy, 1969) are starved before administering the hormone, no hyperglycaemic response can be detected, and it is thought that this is due to a lack of fat body carbohydrate reserve. 7.2.4
Physiologicalfunction
According to Steele (1963) the hyperglycaemic hormone in insects may make blood sugar available at a rapid rate during periods of vigorous exercise or intense metabolic activity. This hypothesis has been inadequately tested. A noteworthy exception is the recent work of Vejbjerg and Norman (1974) with the blowfly C. erythrocephala. In this species, cutting the cardiac recurrent nerve or cardiacetomy results in a marked reduction (seven-fold) in blood trehalose levels after prolonged flight (45 min) and much reduced flight efficiency. These results suggest that the hyperglycaemic hormone is in fact secreted from the corpus cardiacum during flight so as to maintain a sufficiently high trehalose level in the blood. Goldsworthy and Mordue (1974) in their review challenge the assumption that the hyperglycaemic hormone in locusts has a major function in elevating blood trehalose and state that “A greater significance of the hyperglycaemic peptides in the corpus cardiacum of locusts may lie in their capacity to influence other metabolic pathways”. The nature of these pathways is speculative at present. There are metabolic regulatory mechanisms for the control of trehalose synthesis (Murphy and Wyatt, 1965; Friedman, 1967) and these may account for blood sugar homeostasis in many insects. This appears to be the case for the Lepidopterans Hyalophora cecropia (Wiens and Gilberg, 1967b) and Protoparce sexta (Bowers, 1963) where no hyperglycaemic hormone is present in the corpora cardiaca nor do the adults respond to active hormone
STRUCTURE A N D FUNCTION OF INSECT PEPTIDES
105
preparations from other species. Wiens and Gilbert ( 1967b) have concluded that there is no hormonal control of phosphorylase activity in H. cecropia. Thus the evolution of a glucagon-like peptide appears to have occurred in some insect groups and not in others. An understanding of the adaptive value of the hyperglycaemic hormone in physiological terms remains a challenge for insect physiology. The hyperglycaemic hormone also increases the rate of glycogenolysis in the ventral nerve cord of P. americana. Indirect evidence suggests that cyclic AMP mediates the effect of the hormone on phosphorylase activity as it does in the fat body (Steele, 1963; 1964; Hart and Steele, 1973). Robertson and Steele (1972) have reported that low doses of octopamine lead to a cyclic AMP mediated activation of nerve cord phosphorylase activity. Octopamine occurs in relative abundance in the nerve cord of P. americana (Robertson and Steele, 1973a). The significance of nerve cord phosphorylase activation either by the hyperglycaemic hormone or by monophenolic amines is uncertain. Robertson and Steele (1973b) point out that insect nervous tissue contains about ten times more glycogen than mammalian nerve tissue (Treherne, 1965) and that the prominent nutrient reserve of cockroach nerve cord is glycogen in perineurium cells (Wigglesworth, 1960). However it is not known at present how the regulation of phosphorylase activity and glycogenolysis may be related to the trophic function of these cells.
8
Peptides in insect venoms
There are several pharmacologically active substances in the venoms of bees, wasps and hornets: biogenic amines, such as histamine, serotonin, dopamine, noradrenaline, acetylcholine; enzymes, notably phospholipase A, phospholipase B, and hyaluronidase; the peptide toxins melittin, apamin, and MCDpeptide; and kinins. The composition of insect venoms has been thoroughly reviewed by Habermann in 1968 and again in 1972. The broader subject of insect toxins, including venoms, has been covered by Beard (1963). The present work deals only with the peptides in insect venoms, but this should not be taken to imply that these are the most important components of venom. It is probable that no single substance is responsible for the total effect of a sting, nor can the harmful reaction be the simple sum of individual activities (Yoshida et al., 1976). From a clinical standpoint it is generally accepted that the antigenic proteins of the venom are of greatest consequence. The major antigen of bee venom may be the enzyme phospholipase A, (Sobotka et al., 1974; King et al., 1976). Until quite recently, venom peptides were the only insect peptides with known identity and structure-thanks largely to the keen interest which bio-
ROBERT P. BODNARYK
106
chemical pharmacologists and other biochemists have shown towards these powerful agents. A corresponding interest in the biosynthesis of the peptides in the venom gland is not as evident. 8.1 8.1.1
B E E VENOM PEPTIDES
Melittin
The peptide melittin represents 50 per cent of the dry weight of bee venom and with respect to activity it is the main toxin of the venom. Melittin is one of two hemolytic principles in venom. The enzyme phospholipase A is the “indirect (a) Gly Ile Gly Ala
Val
Leu Lys Val Leu
Thr
(b) Gly Ile Gly Ala
Val
Leu Lys Val Leu
Thr
(c) Gly Ile Gly Ala
Ile
Leu Lys Val Leu
Ala
(d) Gly Ile Gly Ala
Ile
Leu Lys Val Leu
Ser U
I
-
Gln NH,
Leu Ile Ser Trp Ile Lq
Lys Arg
Gin
Leu Ile Ser Trp Ile Ly
Lys Arg
Gln Gln NH,
Leu Ile Ser Trp Ile Ly
Lys Arg
LYS Gln NH,
Leu Ile Ser Trp Ile Ly
Lys Arg
Gln Glu NH,
Fig. 8. Amino acid sequences of melittins from the four species of honey bees: (a), Apis mellifera (Habermann and Jentsch, 1967; Kreil, 1973); (b), A. cerana; (c), A.florea (Kreil, 1973); (d), A . dorsata (Kreil, 1975).
hemolysin” in that it is only directly active by causing the formation of lysolecithin from lecithin. Melittin, by contrast, is the “direct hemolysin” and hemolyses erythrocytes by itself. The identification, isolation and amino acid sequencing of melittin have been accomplished by Habermann’s group. a Structure and structure-activity relationships The amino acid sequence of melittin from venom of the widely distributed honey bee of middle Europe, Apis mellvera carnica, has been determined by Habermann and Jentsch (1967) (Fig. 8a). These authors also recognized a second melittin in venom which had a substituted N-terminus. The substituted group was identilied as formyl glycine by Kreil and Kreil-Kiss (1967). The isolation, identification and
STRUCTURE A N D FUNCTION OF INSECT PEPTIDES
107
synthesis of N-terminal formylated melittin (called N"-formyl melittin) was reported by Lubke et al. (1 97 1). Elucidation of the primary structure of melittin has led to an understanding of the physico-chemical and most of the pharmacological properties of the peptide. The amino acid sequence of the hexacosapeptide (Fig. 8a) is atypical in that positions 1 to 20 are occupied by largely hydrophobic amino acid residues, and positions 2 1 to 26 by hydrophilic residues. Therefore, melittin in solution can be considered a cationic detergent (Habermann and Jentsch, 1967). Because of the unequal distribution of hydrophilic residues in melittin and its small size, the peptide does not accommodate itself to a globular configuration. Instead, coacervation of monomers to form tetrameric micelles in aqueous media occurs (Habermann and Kowallek, 1970). Melittin has strong surface activity, decreasing the interfacial tension between air and salt solutions to a degree comparable with the hemolysins lysolecithin or digitonin (Habermann, 1958). The extreme surface-active nature of melittin is evident from its ability to spread readily as a film from water, to form a film from its aqueous solutions at a high rate, and from its ability to penetrate lecithin and mixed lipid films at a rate and to an extent which exceed that of any known biological surfactant. The lytic action of melittin on erythrocytes may thus rest on its capacity to penetrate and disrupt the three-dimensional structure of phospholipid arrays (Sessa et al., 1969). Melittin penetrates lipid monolayers avidly, irrespective of the surface charge of the lipid film, and disrupts artificial phospholipid spherules (liposomes) regardless of whether the liposomes are prepared with a net negative (dicetyl phosphate) or net positive (stearylamine) charge (Sessa et al., 1969). These observations, plus the extraordinary affinity of melittin for lipid membranes suggest that the surface activity of the peptide plus the convenient apolar associations between hydrophobic portions of its structure and the acyl chains of phospholipids are of greater importance than are simple ionic interactions. However, as noted by Habermann (1972), the lack of a strict parallelism between surface activity and hemolytic potency indicates that besides surface activity, other molecular parameters are involved in the hemolytic activity of melittin, one of them probably being its strong basicity. Melittin damages not only erythrocytes but also leucocytes and their lysosomes (releasing lysosomal enzymes), thrombocytes (releasing serotonin), mast cells (releasing histamine), and striated musculature (releasing potassium ions and organic and inorganic phosphates). Like other detergents, melittin disrupts membrane bound enzyme systems, diminishes electron transport in mitochondria and uncouples oxidation from phosphorylation (see Habermann, 1972). The concerted action effected by the several constituents of venom, tissue factors and disrupted biochemical systems is very complex, and when confined locally in a sting painfully effective.
108
ROBERT P. BODNARYK
b Biosynthesis of melittin Promelittin, a precursor of melittin. There are good reasons for supposing that melittin biosynthesis in the secretory cells of the venom gland proceeds via a less active precursor. The ability of melittin to penetrate and disrupt phospholipid membrane structure (Sessa et al., 1969) and its damaging effects on many biochemical systems (Habermann, 1972) suggest that a “safe” precursor ought to be formed, and converted to the active toxin only after being removed from the ribosomes and secreted to a storage site. Indeed, a precursor of melittin (termed promelittin) has been discovered (Kreil and Bachmayer, 1971) and its amino acid sequence determined (Kreil, 1973). The biosynthesis of melittin was studied in vivo by feeding radioactive amino acids to honey bees. Radioactivity was first incorporated into another peptide, and, as the supply of radioactive amino acids became exhausted, the labelled peptide disappeared while the amount of labelled melittin increased and finally reached a plateau. Many structural similarities between the transitory peptide and melittin were established, suggesting that it was a precursor of melittin (Kreil and Bachmayer, 197 1). Extensive analysis of the precursor peptide revealed that it contained the entire sequence of melittin including the Cterminal glutamic acid diamide. The amino end of the promelittin was heterogenous, with peptide species of different chain lengths present in varying amounts. The structure of the main component is: Glu-Pro-Glu-Pro-Asp-ProGlu-Ala-melittin. A second species shorter by two residues has been identified, and another minor species with an extra residue probably exists (Kreil, 1973). The sequence of alternating acidic amino acid residues and proline at the Nterminus of promelittin was found by Kreil to be highly resistant to all proteases except pronase, in marked contrast to the susceptibility of the melittin portion of the molecule to cleavage by common endopeptidases. Activation of promelittin to melittin should therefore be catalysed by enzyme(s) of narrow specificity that cannot attack peptide bonds within melittin. An account of the activating enzyme(s) from the venom gland has not appeared to date. Gauldie el al. (1976) have failed to detect the acidic fragment, even though it must be produced from promelittin in the same molar amount as melittin itself. There must be some efficient mechanism for disposing of the fragment as it is released from the precursor molecule. The physico-chemical properties of promelittin have not yet been investigated, but inferences about its activity can be made (Kreil, 1973). The extra sequence of amino acids at the N-terminal end of promelittin is rich in acidic amino acids and proline. This may lead to a greater solubility of the promelittin monomer and hence to a lesser tendency to form surface-active aggregates, thereby reducing the ability of the protoxin to disrupt membranes. However, further work is needed to establish the properties of promelittin. Kreil (1973) envisages others functions for the N-terminal portion of promelittin. It may be required for packaging or storage of the precursor, or
STRUCTURE AND FUNCTION OF INSECT PEPTIDES
109
may serve as a recognition site in some step of the secretory process. These concepts have been applied to secretory processes in general by Kreil and coworkers (see following section). Evidence for a ‘>re-promelittin ’* (or protomelittin) as a translation product of promelittin mRNA. As seen in the preceding section, the venom gland of honey bees is a highly specialized tissue synthesizing mainly one peptide, melittin, via its precursor promelittin (Kreil and Bachmayer, 1971, 1973). Consequently, the messenger for promelittin is likely to be one of the most abundant species of mRNA in the gland. Kreil and co-workers have turned their attention to the translation of the mRNA of promelittin in frog (Xenopus) oocytes and in mammalian cell-free system. Their work has given insight into the fundamental nature of the translation process and the molecular biology of secretory polypeptides. Unfractionated RNA prepared from young queen bee venom glands and injected into Xenopus oocytes directs the synthesis of a promelittin-like substance. About half of the peptide chain made in oocytes has been sequenced: the 17 amino acid residues identified correspond exactly with the sequences found in promelittin from the venom gland (Kindas-Mugge et al., 1974). The authors draw the r‘ollowing conclusions: (a) the informational content of the codons for a variety of amino acids is the same in cells from different phyla; (b) the results yield final proof that at least part of an injected messenger RNA can be translated with great fidelity and without translational error; (c) the translation of a gland cell insect messenger in Xenopus oocytes demonstrates that at least some of the translational systems of the frog cells are neither phylum nor cell-type specific. Some of the post-translational modifications of promelittin that occur in the venom gland were not observed in the oocytes. Conversion of promelittin to melittin was not detected and the conversion of the carboxyl terminal amino acid of promelittin to the amide form probably does not occur in the oocyte (Kindas-Miigge et al., 1974). Since vertebrates do not synthesize melittin, it is hardly surprising to find that these post-translational mechanisms are lacking in Xenopus. However, the oocyte-derived promelittin, like venom gland promelittin, was heterogeneous at the amino end, suggesting that either the oocyte contains proteases of the correct specificity which can catalyse the activation of the precursor of promelittin, or that the total RNA preparation used in the study contained messengers coding for these proteases. Kreil’s group turned to a cell-free, protein-synthesizing system for further answers, arguing that post-translational modification might be absent or much reduced in vitro. The ability of total RNA preparations from the venom glands of young queen bees to serve as a template in a cell-free system prepared from mammalian sources was investigated by Suchanek et al. (1975). The cell-free system consisted of purified ribosome subunits, rat liver pH-5 fraction and partially purified initiation factors from rabbit reticulocytes and was made
110
ROBERT P BODNARYK
optimal for the translation of rabbit globin mRNA. The heterologous system was found to translate the venom gland mRNA with approximately the same efficiency as hemoglobin mRNA. A polypeptide was synthesized by this in uitro system that had amino acid sequences characteristic of promelittin and which liberated a melittin-like peptide after digestion with a bacterial protease. The polypeptide was larger than promelittin and probably contained the whole of the promelittin sequence, plus a number of additional amino acids at the amino end. Suchanek et al. (1975) note that synthesis of a larger, “prepromelittin” (or protomelittin) by a cell-free system has certain analogies with other observations on the translation of messenger RNAs in heterologous systems. These are: (1) the synthesis of a murine immunoglobin “light” chain containing about 20 additional amino acid residues at the amino end in frog oocytes and a cell-free system from reticulocytes (Stevens and Williamson, 1972; Milstein et al., 1972; Mach et al., 1972; Schechter, 1973) and (2) the synthesis of a polypeptide larger than proparathyroid hormone (“preproparathyroid hormone”) in a cell-free system from wheat germ (Kemper et al., 1974). It can be postulated that in all three cases the large polypeptide products represent precursors, too short-lived to be detected in intact cells but relatively stable in cell-free systems. Suchanek et al. (1975) suggest the intriguing possibility that secretory polypeptides, including melittin, may generally start with some “leading amino acid sequence” which acts as a signal in one of the complex steps involved in secretion. Patterns of promelittin and melittin biosynthesis during bee maturation. Bee stings are used in the defence of the colony against both vertebrate and invertebrate animals [melittin, for example, is highly toxic to Drosophila larvae (Mitchell et al., 1971) and probably to most other invertebrates]. Worker bees undertake a variety of tasks in the hive, followed by a transition from house-bee activity to field-bee activity (Lindauer, 1952). Aging of the worker bee is accompanied by structural changes in the venom apparatus and increased production of venom which reaches its full capacity within about two weeks after emergence. The poison sac is full of venom by this time and the bee can serve as a fearsome guard (Autrum and Kneitz, 1959; da Cruz Landim and Kitajima, 1966). The pattern of synthesis of promelittin and melittin during the maturation process has been followed by Bachmayer el al. (1972). In worker bees, production of promelittin increases slowly from the time of emergence to reach a maximum eight to ten days later, and then declines again. Conversion of promelittin to melittin does not occur during the first two days after emergence, but by the ninth day is proceeding at a maximal rate. Evidently, synthesis of the precursor and its conversion to product are independently controlled. The situation in queen bees is quite different. Both synthesis of promelittin and its conversion to melittin operate close to full capacity in the newly emerged
STRUCTURE A N D FUNCTION
OF INSECT PEPTIDES
111
queen. Bachmayer et al. (1972) point out that full production of melittin, the major toxin of venom, may be needed for imminent duels with other newly emerging queen bees in the colony. Although melittin is the major component of bee venom, its synthesis in worker bees cannot be said to mirror the production of the complete venom. The histamine content of venom has been studied by Owen and Braidwood (1974) and was found to be relatively low (approx. 100 ng/venom extract) in 1week-old worker bees. It continues to increase, reaching approx. 400 ng in 2week-old bees and 1600ng (the maximum) in 5-week-old bees. Owen and Braidwood (1974) conclude that the rise in the histamine content of worker bees is concomitant with the transition from house-bee activity to field-bee activity, and suggest that the venom is only fully completed at the time at which bees start to leave the hive. c Phylogenetic relationships between honeybees as deduced from melittin sequence data There are four real species of Apis: A . mellifera; the Indian bee, A . cerana; and the free-nesting forms A . dorsata and A.jlorea. The amino acid sequences of melittins isolated from the venom apparatus of these species has been obtained by Kreil (1973, 1975) (Fig. 8). All four melittins are composed of 26 amino acid residues; all four begin with the sequence Gly-IleGly-Ala and the carboxyl terminus of each is in the amide form. Differences exist among the species only at positions 5, 10, 15, 22 and 26, the remaining being identical for each species. Kreil (1972, 1975) has used the sequence data to deduce phylogenetic relationships between the various species; these have proven to be consistent with morphological, immunological and other biological criteria. Thus, the sequences of melittin from A . mellifera and A . cerana are identical (Fig. 8a,b) and these two species are very closely related. Cross-fertilization between them is possible, but embryonic development stops at an early stage (Rutner and Maul, 1969). Apparently, in the evolution of honey bees, A . mellifera and A . cerana diverged only recently and it is perhaps not surprising that their melittins are identical. On the other hand, melittin from A . dorsata differs by five residues from that of A.florea, but only by three residues from those of the A. melliferalA. cerana pair. At the nucleic acid level the remote relationship of A.florea to the other three species is even more evident: no fewer than six base changes would be required to convert thejlorea to either the dorsata or the mellueralcerana sequence, using the genetic code of E. coli (Kreil, 1975). A . jlorea is a more distant relative, having only half the number of chromosomes (N = 8) of the A . mellfera1A. cerana pair, is much smaller than the other bees and has many other distinctive features (Kreil, 1973). Based on the structure of melittin, Kreil (1975) proposes a phylogenetic tree for the genus Apis, where
ROBERT P. BODNARYK
112
the line of descent for A.jlorea branches off first from the trunk representing the ancestor common to all honey bees. An identical picture of the evolutionary history of honey bees has been obtained from electrophoretic and immunological cross-reactivity studies of the blood proteins of the different species by Engels et al. (1973). 8.1.2
Mast cell degranulating peptide (MCD-peptide)from bee venom
Mast cells (basophil leucocytes) of mammals are a prominent source of histamine, and in many animals there is a good correlation between the histamine content of a tissue and the number of mast cells (see Bell et al., 1965). Mast cell destruction is the earliest visible process resulting from the sub-cutaneous injection of bee venom in mice (Habermann, 1972). Both melittin and lysolecithin are mastocytolytic agents (lysolecithin is the reaction product of phospholipase A, a key component of bee venom, Habermann,
7 s s ’
Ile Lys Cys Asn Cys Lys Arg His Val Ile Lys Pro His Ile Cys Arg Lys Ile
Cys Gly Lys Asn NH,
1
s
-
s
-
Fig. 9. The structure of MCD-peptide, a mast cell degranulating peptide from bee venom (Haux, 1969; Vernon el al., 1969).
1968, 1972). However, early experimental work indicated that the amount of melittin and phospholipase A in venom did not account entirely for the histamine released in various test tissues. A histamine-releasing fraction was obtained from bee venom (Fredholm, 1966; Fredholm and Hagermark, 1967, 1969) and a peptide (MCD-peptide) with specific histamine-releasing properties was purified by Breithaupt and Habermann (1968). The primary sequence of MCD-peptide was reported by Haux (1969) and confirmed by Vernon et al. (1969) who also determined the position of the two disulphide bridges in the peptide (Fig. 9). MCD-peptide resembles apamin (Fig. 10, section 8.1.3) in that both have two disulphide bridges and both are extremely basic, the MCD-peptide having five lysine and two arginine residues in a total of 22 amino acid residues. Positions three to six in MCD-peptide correspond to positions one to four in apamin, and the carboxyl terminus of both is in the amide form. The MCD-peptide differs from melittin in its specific activity as a histamine releaser, the MCD-peptide being 10-100 times more active. Since bee venom consists of about 50 per cent of melittin and only 1-2 per cent of MCD-
STRUCTURE A N D FUNCTION OF INSECT PEPTIDES
113
peptide, both factors contribute significantly to mast cell destruction caused by whole venom (Breithaupt and Habermann, 1968). MCD-peptide has no hemolytic activity and is practically nontoxic when given by intravenous injection, in marked contrast to the modes of action of melittin and apamin (Habermann, 1972). In the rat, degranulation of mast cells results in the release of histamine and 5-hydroxytryptamine. The MCD-peptide therefore is strongly inflammatory in this species. Recent interest in the MCD-peptide has centred about its potent anti-inflammatory (sic) properties and its ability to suppress the development of adjuvant arthritis and reduce the severity of primary and secondary lesions in established adjuvant arthritis in the rat (Billingham et al., 1973; Hanson et al., 1974). Whether these two apparently opposite biological activities are related is unclear at present (Gauldie et al., 1976). Nevertheless, these intriguing observations, in addition to their obvious potential practical
H Cys Asn Cys Lys M a Pro Glu Thr Ala Leu Cys Ala Arg Arg Cys Gln Gln His NH, I
I
s
-
s
I
I
Fig. 10. The structure of apamin, a neurotoxin from bee venom (Haux et al., 1967; Shipolini et
al., 1967; Callewaert et al., 1968).
significance, may relate to an ancient and apocryphal belief that the venom of honey bees is beneficial in certain arthritic and rheumatoid conditions (Beck, 1935; Broadman, 1963). 8.1.3
Apamin, a neurotoxic peptide from bee venom
Habermann’s group, while screening the peptide fractions of whole bee venom by gel filtration, obtained a potent neurotoxin that was further purified on carboxymethylcellulose and subjected to amino acid analysis. The neurotoxin-a relatively small peptide-was called apamin (Habermann and Reiz, 1964,1965aYb;Habermann, 1972). a Structure The amino acid sequence of apamin (Fig. 10) has been established by two independent groups (Haux et al., 1967; Shipolini et al., 1967). The peptide consists of 18 amino acids, four of them being half-cystines. One disulfide bridge connects a half-cystine in position 1 and a half-cystine in position 11. The second bridge connects a half-cystine in position 3 and a halfcystine in position 15 (Callewaert et al., 1968). Apamin is a basic peptide containing one lysine, one histidine and two arginine residues. The apamin octadecapeptide is the smallest neurotoxic peptide known. Its basicity and cross-linking by disulfide bridges are features common to the snake toxins
114
ROBERT P. BODNARYK
which are generally classified as short neurotoxins (60-62 residues; four disulphide bridges) and long neurotoxins (7 1-74 residues; 5 disulfide bridges) (rev. Yang, 1974). b Synthesis Synthesis of apamin has been achieved recently by van Rietschoten et al. (1975) using a solid-phase procedure. Synthetic apamin reported by the Marseille group represents the first synthesis of neurotoxin with full toxic activity. It is also the first laboratory-synthesized neurotoxin for which the chemical purity and identity with the natural peptide have been demonstrated. By comparison, Aoyagi et al. (1972) reported the synthesis of a peptide with cobrotoxin activity; this synthetic peptide had only about 20 per cent of the activity of the natural snake toxin. van Rietschoten et al. (1975) attribute their remarkably successful synthesis to the use of solid-phase methodology (Merrifield, 1969) with high yields of incorporation of each amino acid (99.3 per cent on average over 15 steps), to the use of the fluorescamine test to check completeness of coupling (Felix and Jimenez, 1973) and to the high quality of the solid support resin (Tregear, 1972). c Structurefunction relationships Specific chemical modifications of natural apamin have been used to study the residues involved in its toxic action (Vincent et al., 1975). Their work has shown that the alpha-amino group of Cys 1, the epsilon-amino group of Lys 4, the carboxylate side chain of Glu 17 and the imidazole group of His 18 are not essential for the toxic activity of apamin. A synergistic effect was observed when several functions were modified. Thus, apamin in which the alpha- and epsilon-amino groups of Cys 1 and Lys 4 have been acetylated and the imidazole of His 18 has been carbethoxylated is devoid of activity. Complete loss of toxicity also results from reduction and alkylation of both disulphide bridges, a finding recently confirmed by Gauldie et al. (1976). Evidently, the disulphide bridges are necessary for biological activity. According to Vincent et al. (1975), the most important part of the apamin sequence for neurotoxic activity appears to be the C-terminal region containing the two arginine residues. Chemical modification of Arg 13 and Arg 14 eliminated toxicity, as did removal of Arg 14 of acetylated apamin by digestion with trypsin. Further progress on structure-function relationships can be expected from work on synthetic structural analogs of apamin (van Rietschoten et al., 1975). d Pharmacological activity and site of action A lethal dose (LD50 = 4 mg/kg) (Habermann, 1972; Gauldie et al., 1976) of apamin given to mice by intravenous injection induces uncoordinated hypermotility within
STRUCTURE A N D FUNCTION OF INSECT PEPTIDES
115
15 min, culminating in generalized convulsions followed by respiratory distress and death. Sublethal doses cause extreme hyperexcitability which can last for up to 60 h. The site of action is localized in the spinal cord as judged from neurophysiological studies on the effect of apamin on spinal reflexes (Wellhoner, 1969). Radioactive apamin is found predominantly in the spinal cord after its injection into mice (Vincent et al., 1975). The effect of apamin is to augment polysynaptic reflexes and render excitatory polysynaptic pathways more effective than inhibitory polysynaptic mechanisms (Wellhoner, 1969). Current research interest is directed towards understanding the molecular mechanisms of toxic action. Little is anything is known about the in vivo biosynthesis of apamin in the venom gland of the bee. 8.1.4
Melittin F, tertiapin and secapin
A large-scale fractionation of 700 g (sic)of crude bee venom has resulted in the isolation of three new peptides (Gauldie et al., 1976). Melittin F, a 19-amino acid residue peptide, is evidently a fragment of melittin consisting of residues 8-26. Tertiapin is a 20-residue basic peptide. Both peptides are present in venom in very small amounts. Their pharmacological properties have not yet been explored. The third newly discovered peptide, secapin, comprises about 1 per cent of lyophilized venom and apparently has been overlooked by other workers. It is a 24-residue basic peptide containing a large proportion of proline and one disulphide bridge. Secapin has a very low mammalian toxicity. At high doses in mice it produces marked hypothermia and signs of sedation. 8.1.5
Other bee venom peptides
Nelson and O’Connor (1968) have reported two small peptides from bee venom with histamine at the C-terminus. They have been characterized as alagly-pro-gln-histamine and ala-gly-gln-gly-histamine (procamine) by Peck and O’Connor (1974). Synthetic procamine is said to have the same chromatographic properties as the natural peptide by the authors. The presence of histamine peptides in bee venom could not be confirmed by Gauldie et al. (1976) in their large-scale fractionation of venom (section 8.1.4), and no explanation of this discrepancy has been offered. Minimine, a basic peptide reported by Lowy et al. (1971) was not found by Gauldie et al. (1976). Probably, minimine activity as described by Lowy et al. (197 1) was due to phospholipase A activity (Habermann, 1972). The total number of bona Jide bee venom peptides discovered to date, including variants of apamin and the MCD-peptide, appears to be eight (Gauldie et al., 1976).
116
ROBERT P. BODNARYK
1
2
3 4
5
6
7
8 9
Arg Pro Pro Gly Phe Ser P r o Phe Arg Bradykinin Arg P r o Pro Gly Phe Ser P r o Phe Arg Lysylbradykinin (kallidin) Met Lys
Arg Pro Pro Gly Phe Ser Pro Phe Arg Methionylly iylbrady kinin
pGlu Thr Asn Lys Lys Lys Leu Arg Gly
Arg Pro Pro Gly Phe Ser Pro Phe Arg Polisteskinin
c1 c2 I 1 Thr Ala Thr Thr Arg Arg Arg Gly
Arg Pro Pro Gly Phe Ser Pro Phe Arg Vespulakinin 1
c 1 c2
I
I
Thr Thr Arg Arg Arg Gly
Arg Pro Pro Gly Phe Ser Pro Phe Arg Vespulakinin 2 Arg Pro Pro Gly Phe Thr P r o Phe Arg Thr6-bradykinin
Ala Arg
Arg Pro Pro Gly Phe Thr P r o Phe Arg
Alanylargininyl-Thr6-bradykinin Fig. 11. A comparison of the structures of bradykinin and bradykinin derivatives with bradykinin-like peptides from the venom of wasps. Polisteskinin is from the venom of wasps of the genus Polistes (Nakajima el al., 1967; Pisano, 1968). Vespulakinins 1 and 2 are from the venom of the yellow jacket Vespula maculifons (Yoshida et al., 1976). The carbohydrate prosthetic group of the vespulakinins has not been fully characterized to date. C : N. Ac. 2, galactose 1; C2: N. Ac. Galactosamine 2 3, galactose 2. Thr6Galactosamine 1 bradykinin and alanylargininyl-Thr6-bradykininare from the Japanese wasp P. rothneyi iwatai (Watanabe et al., 1975).
-
8.2
Y
K I N I N S FROM W A S P S A N D HORNETS
The venoms from wasps and hornets contain bradykinin-like peptides but apparently lack melittin, apamin and MCD-peptide. Bradykinin and the structurally related kallidin (lysylbradykinin) and methionyllysylbradykinin
STRUCTURE A N D FUNCTION OF INSECT PEPTIDES
117
(Fig. 11) are smooth muscle active, hypotensive agents. Bradykinin is among the most potent pain-producing agents (revs. Erdos, 1966; Ryan et al., 1970). Venom kinins have pharmacological actions similar to bradykinin and kallidin. Vasoactive peptides in animal venoms are concisely reviewed by Pisano (1968) whose bibliography provides reference to many general works. Insect peptides similar to bradykinin were first found in venom of the European wasp Vespa vulgaris (Jaques and Schachter, 1954; Schachter and Thain, 1954; Holdstock et al., 1957). Crude venom from V. vulgaris was resolved into three peaks of kinin activity (a major and two minor peaks) by ion-exchange chromatography (Mathias and Schachter, 1958). The activity of the wasp kinins was much reduced by trypsin (bradykinin is resistant to trypsin digestion) indicating that structural differences exist between these two types of kinins. The hornet Vespa crabo contains a single, trypsin-resistant kinin that is clearly different from the wasp kinins by chromatographic and enzymic tests and is about one-tenth as active as bradykinin in contracting guinea pig ileum (Bhoola el al., 1961). The structure of insect venom kinins is known in three cases: polisteskinin from wasps of the genus Polistes, P-11-1 and P-111 from the Japanese wasp, P. rothneyi iwatai, and vespulakinin from the yellow jacket Vespula maculfrons.
8.2.1 Polisteskinin
Polisteskinin was purified from extracts of the terminal three abdominal segments from 6000 wasps of mixed species, Polistes annularis, P. fuscatus and P. exclamans (Prado et al., 1966; Pisano, 1968) and its structure determined (Nakajima et al., 1967; Pisano, 1968). The octadecapeptide contains the nonapeptide bradykinin at its carboxy-terminal end (Fig. 11). Polisteskinin is a strongly basic peptide containing three arginine and three lysine residues out of a total of 18 amino acid residues. Laboratory synthesis has not been reported to date, although this would seem straightforward if synthesis was started from bradykinin. Polisteskinin has pharmacological properties that distinguish it from mammalian bradykinin, kallidin and methionyllysylbradykinin. Unlike these kinins, polisteskinin is not inactivated by passage through the rat pulmonary vascular bed (Ryan et al., 1970), it has a longer acting hypotensive effect and it is the most potent naturally occurring releaser of histamine from rat mast cells (Johnson and Erdos, 1973). Very little is known about the biosynthesis of polisteskinin in the venom gland. In mammals, bradykinin is formed from an inactive protein precursor in blood plasma (bradykininogen) by the action of specific proteases at the site of its action (Erdos, 1966). No kininogen has been found for polisteskinin either
118
ROBERT P. BODNARYK
in venom or whole wasp extracts, at least not in the preliminary experiments of Prado et al. (1966), thus leaving open the question of insect kinin biosynthesis. 8.2.2 Thr6-bradykininand ala-arg-thr6-bradykinin
Two new bradykinin analogues have been reported recently in the venom of the Japanese wasp P. rothneyi iwatai by Watanabe et al. (1975). These are Thr6bradykinin and Ala-Arg-Thr6-bradykinin.The discovery of these bradykinin analogues in Polistes indicates that active peptides in wasp venom may be different from species to species. 8.2.3
Vespulakinins I and 2
Vespulakinins are newly discovered carbohydrate-containing bradykinin derivatives from the venom sac of the yellow jacket Vespula maculifons (Yoshida et al., 1976). They are the first reported naturally occurring glycopeptide derivatives of bradykinin and the first reported vasoactive glycopeptides. Vespulakinins are similar in structure to polisteskinin (Fig. l l ) in that bradykinin is at the carboxy-terminal end. The heptadecapeptide vespulakinin 1 and pentadecapeptide vespulakinin 2 are also highly basic peptides containing no fewer than five arginine residues per 17 and 15 amino acid residues, respectively. Their most distinctive feature is the carbohydrate prosthetic group (which has not yet been fully characterized). The extent to which the carbohydrate moiety contributes to biological activity is not known at present. Vespulakinin 1 is at least twice as potent as bradykinin (on a weight basis) in lowering rat blood pressure, but the duration of the response is not significantly longer (Yoshida et al., 1976). Further pharmacological testing of the potency of the vespulakinins in releasing histamine from mast cells and leucocytes and their ability to produce pain will be of interest.
9
Concluding remarks
The list of peptides found in insects undoubtedly will continue to grow as our knowledge of peptide-mediated processes continue to increase. It is evident that enormous experimental benefit accrues to the individual or group of individuals who, early in the course of their work with peptides, devote their attention to establishing primary structure. Synthesis, structure-function relationships, isotope studies and precise, definitive physiological experimentation are then all possible, and soon follow with the psychological advantage of dealing with a known substance.
STRUCTURE A N D FUNCTION OF INSECT PEPTIDES
119
Acknowledgements
I wish to thank Dr Leo Levenbook for generous and unfailing advice throughout the writing of the review and Dr P. S. Chen for valued guidance on the peptides from Drosophila. I am indebted to Drs H. Baumann, B. E. Brown and A. N. Starratt who provided me with copies of their manuscripts prior to publication. Special thanks are due to Mr. K. D. Oliver and library staff for locating reference material. This chapter is Contribution No. 762, Canada Agriculture. References Adams, T. S. and Nelson, D. R. (1968). Bioassay of crude extracts for the factor that prevents second mating in female Musca domestica. Ann. ent. SOC.Am. 61, 112-1 16. Autrum, H. and Kneitz, H. (1959). Die Giftsekretion in der GiRdriise der Honigbienen in Abhangigkeit vom Lebensalter. Biol. Zentralbl. 18,595-602. Aoyagi, H., Yonazawa, H., Takahasi, N., Kato, T., Izumiya, N. and Yang, C. C. (1972). Synthesis of a peptide with cobrotoxin activity. Biochim. Biophys. Acta, 263,823-826. Bachmayer, H., Kreil, G. and Suchanek, G. (1972). Synthesis of promelittin and melittin in the venom gland of queen and worker bees: patterns observed during maturation. J. Insect. Physiol. 18, 1515-1521. Balabaskaran, S. and Smith, J. N. (1970). The inhibition of l,l,l-trichloro-2,2-bis-(pchlorophenyl) ethane (DDT) dehydrochlorinase and glutathione S-aryltransferase in grassgrub and housefly preparation. Biochem. J. 111,989-996. Balogun, R. A. (1974). A sex-specific ninhydrin-positive component detected in the accessory glands of adult male tsetse flies (Diptera, Glossinidae). Nigerian J. Ent. 1, 13-16. Bargmann, W., Lindner, E. and Andres, K. H. (1967). ~ e Synapsen r an endokrinen Epithelzellen und die Definition sekretorischer Neurone. Untersuchungen am Zwischenlappen der Katzenhypophyse. Z . Zellforsch. 11,282-298. Baumann, H. (1974a). The isolation, partial characterization, and biosynthesis of the paragonial substances, PS-I and PS-2, of Drosophilafunebris. J. Insect Physiol. 20,2181-2194. Baumann, H. (1974b). Biological effects of paragonial substances PS-1 and PS-2, in females of Drosophilafunebris. J. Insect Physiol. 20,2347-2362. Baumann, H. and Chen, P. S. (1 973). Geschlechtsspezifische Ninhydrin-positive Substanzen in Adultmannchen von Drosophila funebris. Rev. Suisse 2001.80,685-690. Baumann, E. and Gersch, M. (1973). Untersuchungen zur Stabilitat des Neurohormons D. Zool. Jb. Physiol. 71, 153-160. Baumann, E. and Gersch, M. (1974). Versuche zur Markierung von Neurohormon D aus Periplaneta americana mit Hilfe von Dansylchlorid. Zool. Jb. Physiol. 18,533-54 1. Baumann, H., Wilson, K. J., Chen, P. S. and Humbel, R. E. (1975). The amino acid sequence of a peptide (PS-1) from Drosophila funebris: A paragonial peptide from males which reduces the receptivity of the female. Eur. J. Biochem. 5 5 521-529. Beard, R. L. (1963). Insect toxins and venoms. Ann. Rev. Ent. 8, 1-18. Beck, B. F. (1935). “Bee venom therapy.” D’Appleton-Century Co., Inc., New York. Bell, G. H., Davidson, J. N. and Scarborough, H. (1965). “Textbook of Physiology and Biochemistry.” 6th Edn. E. & S. Livingstone Ltd., Edinburgh and London. Belton, P. and Brown, B. E. (1969). The electrical activity of cockroach visceral muscle fibres. Comp. Biochem. Physiol. 28,853-863.
120
ROBERT P. BODNARYK
Bhoola, K. D., Calle, J. D. and Schachter, M. (1961). Identification of acetylcholine, 5hydroxytryptamine, histamine and a new kinin in hornet venom. J. Physiol. Lond. 159, 167182. Billingham, M. J. E., Morley, J., Hanson, J. M., Shipolini, R. A. and Vernon, C. A. (1973). An anti-inflammatory peptide from bee venom. Nature, Lond. 245, 163-164. Bodnaryk, R. P. (1970a). Effect of DOPA-decarboxylase inhibition on the metabolism of Balanyl-L-tyrosine during puparium formation in the fleshfly Sarcophaga bullata Parker. Comp. Biochem. Physiol. 35,221-227. Bodnaryk, R. P. (1970b). Chemical taxonomy: an application to the genus Sarcophaga (Diptera: Sarcophagidae). Can. Ent. 102, 349-353. Bodnaryk, R. P. (1970~).Biosynthesis of gamma-L-glutamyl-L-phenylalanineby the larva of the housefly Musca domestica. J. Insect. Physiol. 16,919-929. Bodnaryk, R. P. (197Od). Levels of free glutamic acid, phenylalanine, and y-glutamyl-lphenylalanine during pupal sclerotization in the housefly, Musca domestica L. Comp. Biochem. Physiol. 35,499-502. Bodnaryk, R. P. (1971a). Studies on the incorporation of B-alanine into the puparium of the fly, Sarcophaga bullata. J. Insect Physiol. 17, 1201-1210. Bodnaryk, R. P. (1 97 1b). N-terminal palanine in the puparium of the fly, Sarcophaga bullata: evidence from kinetic studies of its release by partial acid hydrolysis. Insect. Biochem. 1,228236. Bodnaryk, R. P. (1971~).Effect of exogenous molting hormone (ecdysterone) on B-alanyl-Ltyrosine metabolism in the larva of the fly Sarcophaga bullata Parker. Gen. Comp. Endocrinol. 16,363-368. Bodnaryk, R. P. (1972a). A survey of the occurrence of 8-alanyl-tyrosine, y-glutamylphenylalanine and tyrosine-0-phosphate in the larval stage of flies (Diptera). Comp. Biochem. Physiol. 438, 587-592. Bodnaryk, R. P. (1972a). A preparative-scale enzymic synthesis of y-L-glutamyl-Lphenylalanine. Insect. Biochem. 2,49-52. Bodnaryk, R. P. (1972b). Membrane-bound pglutamyl transpeptidase. Evidence that it is a component of the “amino acid site” of certain neutral amino acid transport systems. Can. J. Biochem. 50,524-528. Bodnaryk, R. P. (1974). Kinetic aspects of the breakdown of y-L-glutamyl-L-phenylalanine during sclerotization of the puparium of Musca domestica. Insect Biochem. 4,439-454. Bodnaryk, R. P. and Brunet, P. C. J. (1974). 3-0-hydrosulphato-4-hydroxyphenethylamine (dopamine 3-O-sulphate), a metabolite involved in the sclerotization of insect cuticle. Biochem. J. 138,463-469. Bodnaryk, R. P., Brunet, P. C. J. and Koeppe, J. K. (1974). On the metabolism of Nacetyldopamine in Periplaneta americana. J. Insect Physiol. 20,9 11-923. Bodnaryk, R. P. and Levenbook, L. (1968). Naturally occurring low-molecular-weightpeptides from the blowfly Phormia regina. Biochem. J. 110, 77 1-773. Bodnaryk, R. P. and Levenbook, L. (1969). The role of 8-alanyl-L-tyrosine (sarcophagine) in puparium formation in the fleshfly Sarcophaga bullata. Comp. Biochem. Physiol. 30, 909921. Bodnaryk, R. P. and McGirr, L. (1973). Purification, properties and function of a unique yglutamyl cyclotransferase from the housefly, Musca domestica L. Biochim. Biophys. Acta, 315,352-362. Bodnaryk, R. P. and Skillings, J. R. (1971). y-Glutamyl transpeptidase catalyses the synthesis of y-glutamylphenylalanine in the larva of the housefly, Musca domestica. Insect Biochem. 1, 467-479, Bodnaryk, R. P., Bronskill, J. F. and Fetterly, J. R. (1974). Membrane-bound y-glutamyl
STRUCTURE A N D FUNCTION OF INSECT PEPTIDES
121
transpeptidase and its role in phenylalanine absorption-reabsorption in the larva of Musca domestica. J. Insect Physiol. 20, 167- 1 81. Booth, J., Boyland, E. and Sims, P. (1961).An enzyme from rat liver catalysing conjugations with glutathione. Biochem. J. 79,516-524. Bowers, W. S. (1963). The relationship between hormonal activity and sugar metabolism in Protoparce sexta (Johannsen) and Blaberus craniifer, Burmeister. Ph.D. Thesis, Dissertation Abstr. 23,2575. Bowers, W. S. and Friedman, S. (1963). Mobilization of fat body glycogen by an extract of corpus cardiacum. Nature, Lond. 198,685. Boyland, E. and Chasseaud, L. F. (1969).The role of glutathione and glutathione S-transferases in mercapturic acid biosynthesis, Adv. Enzymol. 32, 173-219. Bradbury, F. R. and Standen, H. (1959). Metabolism of benzene hexachloride by resistant houseflies Musca domestica. Nature, Lond. 183,983-984. Bradbury, F. R. and Standen, H. (1960).Mechanisms of insect resistance to the chlorohydrocarbon insecticides. J. Sci. Food Agric. I I, 92-100. Breithaup, H. and Habermann, E. (1968). Mastzelldegranulierendes Peptid (MCD-Peptid) aus Bienengift: Isolierung, biochemische und pharmakologische Eigenschaften. NaunynSchmiedebergs Arch. Pharmakol. Exp. Pathol. 261,252-270. Bridges, R. G. (1 959).Pentachlorocyclohexene as a possible intermediate metabolite of benzene hexachloride in houseflies. Nature, Lond. 184, 1337-1338. Broadman, J. (1963). “Bee venom: the natural curative for arthritis and rheumatism.” G. Putmans Sons,New York. Bromer, W. W., Boucher, M. E. and Koffenberger, J. E. (1971). Amino acid sequence of bovine glucagon. J. Biol. Chem. 246,2822-2827. Bromer, W. W., Sinn, L. G. and Behrens, 0. K. (1957). The amino acid sequence of glucagon. V. Location of amide groups, acid degradation studies and summary of sequential events. J. Amer. Chem. SOC.79,2807-2810. Brown, B. E. (1 965). Pharmacologically active constituents of the cockroach corpus cardiacum. Resolution and some characteristics. Gen. Comp. Endocrinol. 5,387-401. Brown, B. E. (1967). Neuromuscular transmitter substance in insect visceral muscle. Science,
155,595-597. Brown, B. E. (1975). Proctolin: a peptide transmitter candidate in insects. Life Sci. 17, 1241-
1252. Brown, B. E. and Nagai, T. (1969). Insect visceral muscle: neural relations of the protodeal muscles of the cockroach. J. Insect Physiol. 15, 1767-1 783. Brown, B. E. and Starratt, A. N. (1975). Isolation of proctolin, a myotropic peptide from Periplaneta americana. J. Insect Physiol. 21, 1879-1881. Bull, D. L. and Lindquist, D. A. (1964). Metabolism of 3-hydroxy-N,N-dimethylcrotonamide dimethyl phosphate by cotton plants, insects, and rats. Agric. Food Chem. 12,3 10-3 17. Bull, D. L. and Whitten, C. J. (1972). Factors influencing organophosphorus insecticide resistance in tobacco budworms. J. Agr. Food Chem. 20,561-564. Callewaert, G. L., Shipolini, R. and Vernon, C. A. (1968). The disulphide bridges of apamin. FEBS Lett. 1, 1 1 1-1 13. Chang, M. M., Leeman, S. E. and Niall, H. D. (1971). Amino-acid sequence of substance P. Nature, New Biol. 232, 86-87. Chen, P. S. (1971). Biochemical aspects of insect development. In “Monographs in Developmental Biology” (Ed. A. Wolsky), 3,pp. 116-128. Karger, Basel. Chen, P. S. and Biihler, R. (1970a). Isolierung und Funktion des Sexpeptids bei Drosophila melanogaster. Rev. Suisse Zool. 77,548-554. Chen, P. S . and Buhler, R. (1970b).Paragonial substance (sex peptide) and other free ninhydrin-
122
ROBERT P. BODNARYK
positive components in male and female adults of Drosophila melanogaster. J. Insect Physiol.
16,615-627. Chen, P. S . and Diem, C. (1961). A sex-specific ninhydrin-positive substance found in the paragonia of adult males of Drosophila melanogaster. J. Insect Physiol. 7,289-298. Clark, A. G., Darby, F. J. and Smith, J. N. (1967). Species differences in the inhibition of glutathione S-aryltransferase by phthaleins and dicarboxylic acids. Biochem. J. 103,49-54. Clark, A. G.,Hitchcock, M. and Smith, J. N. (1966). Metabolism of gammexane in flies, ticks and locusts. Nature, Lond. 209, 103. Clark, A. G.,Murphy, S. and Smith, J. N. (1969).The metabolism of hexachlorocyclohexanes and pentachlorocyclohexenes in flies and grass grubs. Biochem. J. 113,89-96. Clark, A. G., Smith, J. N. and Speir, T. W. (1973). Cross specificity in some vertebrate and insect glutathione-transferases with methyl parathion (dimethyl p-nitrophenyl phosphorothionate), l-chloro-2,4-dinitrobenzene and S-crotonyl-N-acetylcysteamine as substrates. Biochem. J. 135,385-392. Cohen, A. J. and Smith, J. N. (1964). Comparative detoxication. 9. The metabolism of some halogenated compounds by conjugation with glutathione in the locus. Biochem. J. 90, 449-
456. Cohen, A. J., Smith, J. N. and Turbert, H. (1964). Comparative detoxication. 10.The enzymic conjugation of chloro compounds with glutathione in locusts and other insects. Biochem. J. 90,
457-464. Collett, J. I. (1976). Peptidase-mediated storage of amino acids in small peptides. Insect Biochem. 6, 179-185. Connell, G.E.and Hanes, C. S. (1 956). Enzymic formation of pyrrolidone carboxylic acid from yglutamyl peptides. Nature, Lond. 177,377-378. Cook, B. J. and Holman, G. M. (1975a).The neural control of muscular activity in the hindgut of the cockroach Leucophaea maderae: prospects of its chemical mediation. Comp. Biochem. Physiol. SOC, 137-146. Cook, B. J. and Holman, G. M. (1975b).Sites of action of a peptide neurohormone that controls hindgut muscle activity in the cockroach Leucophaea maderae. J. Insect Physiol. 21, 1187-
1192. Cook, B. J., Holman, G. M. and Marks, E. P. (1975). Calcium and cyclic AMP as possible mediators of neurohormone action in the hindgut of the cockroach, Leucophaea maderae. J. Insect Physiol. 21, 1807-1814. daCruz Landim, C. and Kitajima, E. W. (1966). Ultraestrutura do aparelho venenifero de Apis (Hymenoptera, Apidae). Mem. Inst. Butantan Sao Paul0 33,701-7 10. Dahm, P. A. (1970). In “Biochemical Toxicology of Insecticides” (Eds R. D. OBrien and I. Yamamoto), pp. 5 1-63. Academic Press, New York and London. Dauterman, W. C., Casida, J. E., Knaak, J. B. and Kowalczyk, T. (1959).Bovine metabolism of organophosphorus insecticides. Metabolism and residues associated with oral administration of dimethoate to rats and three lactating cows. J. Agr. Food Chem. 7, 188-193. Davey, K. G. (1961a). Substances controlling the rate of beating of the heart of Periplaneta. Nature. Lond. 192 284. Davey, K. G. (1961b). The mode of action of the heart accelerating factor from the corpus cardiacum of insects. Gen. comp. Endocrin. 1,2429. Davey, K. G. (1964). The control of visceral muscles in insects. In “Advances in Insect Physiology”, Vol. 2,219-245. Academic Press, New York and London. Dinamarca, M. L., Levenbook, L. and Valde, E. (1971). DDT-dehydrochlorinase. 11. Subunits, sulfhydryl groups, and chemical composition. Archs. Biochern. Biophys. 147, 374-
383. Dinamarca, M. L., Ramirez, A., W a r , E. D. and Capdevila, J. (1974). DDT-dehydrochlorinase. 111. Enzymic oxidation of glutathione. Int. J. Biochem. 5, 1-9.
STRUCTURE A N D FUNCTION OF INSECT PEPTIDES
123
Dinamarca, M. L., Saavedra, I. and Valdes, E. (1969). DDT-dehydrochlorinase. I. Purification and characterization. Comp. Biochem. Physiol. 31, 269-282. Donninger, C. (197 1). Species specificity of phosphate triester anticholinesterases. Bull. W.H.O. 44,265-267. Cutrieu, J. and Gourdoux, L. (1967). Le controle neuro-endocrinien de la trehalosemie de Carausius morosus. C.r. hebd. Seanc. Acad. Sci. Paris 265, 1067-1070. Engels, W. (1973). Evolution of the protein spectra in the haemolymph, a physiological feature of the development of female castes in the 4 real species of Apis:florea, dorsata, cerana and mellijica. Proc. 24th Int. Apicultural Congress, Buenos Aires, 327-332. Erdos, E. G. ( 1966). Hypotensive peptides: bradykinin, kallidin and eledoisin. Aduan. Pharmacol. 4, 1-90. Eto, M. (1974). “Organophosphorus pesticides: organic and biological chemistry.” CRC Press Inc., Cleveland, Ohio. von Euler, U. S. and Gaddum, J. H. (193 1). An unidentified depressor substance in certain tissue extracts. J. Physiol. 12, 74-87. Farah, A. and Tuttle, R. (1960). Studies on the pharmacology of glucagon. J. Pharmacol. 129, 49-55. Felix, A. M. and Jimenez, M. H. (1973). Rapid Fluorometric detection for completeness in solid phase coupling reactions. Anal. Biochem. 52,377-38 1. Folsom, M. D., Hansen, L. G., Philpott, R. M., Yang, R. S. H.,Dauterman, W. C. and Hodgson, E. (1 970). Biochemical characteristics of microsomal preparations from diazinonresistant and susceptible houseflies. Life Sci. 9, 869-875. Fox, A. S. (1956). Chromatographic differences between males and females in Drosophila melanogaster and role of X and Y chromosomes. Physiol. Zool. 29,288-298. Fox, A. S., Mead, C. G. and Munyon, I. L. (1959). Sex peptide of Drosophila melanogaster. Science, 129, 1489-1490. Fox, A. S., Munyon, I. L., Singh, P. and Sweeney, E. A. (1962). Genetic determination, amino acid composition, and synthesis of the sex peptide of males in Drosophila melanogaster. Genetics, 47,95 3. Fraenkel, G. (1935). A hormone causing pupation in the blowfly Calliphora erythrocephala. Proc. Roy. Soc. B118, 1-12. Fredholm, B. (1966). Studies on a mast cell degranulating factor in bee venom. Biochem. Pharmacol. 15,2037-2043. Fredholm, B. and Hagermark, 8.(1967). Histamine release from rat mast cells induced by a mast cell degranulating fraction in bee venom. Acta Physiol. Scand. 69,304-3 12. Fredholm, B. and Hagermark, 8.(1969). Studies on the histamine releasing effect of bee venom fractions and compound 48/80 on skin and lung tissue of the rat. Acta Physiol. Scand. 76, 288-298. Friedman, S . (1967). The control of trehalose synthesis in the blowfly, Phormia regina Meig. J. Insect Physiol. 13,397-405. Fuchs, M. S., Craig, G. B. and Despommier, D. D. (1969). The protein nature of the substance inducing female monogamy in Aedes aegypti. J. Insect Physiol. 15,701-709. Fuchs, M. S.,Craig, G. B. and Hiss, E. A. (1968). The biochemical basis of female monogamy in mosquitoes. I. Extraction of the active principle from Aedes aegypti. Life Sci. 7,835-839. Fuchs, M. S. and Hiss, E. A. (1970). The partial purification and separation of the protein components of matrone from Aedes aegypti. J. Insecf Physiol. 16,931-939. Fukami, J. and Shishido, T. (1963). Studies on the selective toxicities of organic phosphorus insecticides (111). The characters of the enzyme system in cleavage of methyl parathion to desmethyl parathion in the supernatant of several species of homogenates (Part 1). Bofyukagaku, 28,77-83. Fukami, J. and Shishido, T. (1966). Nature of a soluble, glutathione-dependent enzyme system
124
ROBERT P. BODNARYK
active in cleavage of methyl parathion to desmethyl parathion. J. Econ. Entomol. 59, 1338-
1346. Fukunaga, K., Fukami, T. and Shishido, T. (1969). The in vitro metabolism of organophosphorus insecticides by tissue homogenates from mammal and insect. Residue Rev.
25,223-249. Garcia-Bellido, A. (1964).Das Sekret der Paragonien als Stimulus der Fekunditat bei Weibchen von Drosophila melanogaster. Z. Naturf. 19b,49 1495. Gauldie, J., Hanson, J. M., Rumjanek, F. D., Shipolini, R. A. and Vernon, C. A. (1976).The peptide components of bee venom. Eur. J. Biochem. 61,369-376. Gersch, M. ( 1974). Experimentelle Untersuchungen zur Ausschiittung von Neurohormonen aus Ganglien des Bauchmarks von Periplaneta americana nach elektrischer Reizung in vitro. Zool. Jb. Physiol. 78, 138-149. Gersch, M., Fischer, F., Unger, H. and Koch, H. (1960). Die Isolierung neurohormonaler Faktoren aus dem Nervensystem der Kiichenschabe Periplaneta americana. Z. Naturjorsch. 15b, 319-322. Gersch, M. and Richter, K. (1963). Auslosung von Nervenimpulsen durch ein Neurohormen bei Periplaneta americana. Zool. Jb. Physiol. 70,3011308. Gersch, M., Richter, K., Stiirtzebecher, J. and Fabian, B. (1969). Experimentelle Untersuchungen zum Wirkungsmechanismus des Neurohormons D von Periplaneta americana am “biologischen Modell” der Harnblase von Bufo bufo. Gen. Comp. Endocrinol. 12,40-50. Gersch, M. and Stiirzebecher, J. (1967). Zur Frage der Identitat und des Vorkommens von Neurohormon D in verschiedenen Bereichen des Zentralnervensystems von Periplaneta americana. Z. Naturforsch. 22b,563. Gersch, M., Unger, H., Fischer, F. and Kapitza, W. (1963). Identifizierung einiger Wirkstoffe aus dem Nervensystem der Crustaceen und Insekten. Z. Naturjorsch. 18b,587-588. Goldsworthy, G. J. (1969). Hyperglycaemic factors from the corpus cardiacum of Locusta migratoria. J. Insect Physiol. 15, 213 1-2140. Goldsworthy, G. J. (1970).The action of hyperglycaemic factors from the corpus cardiacum of Locusta migratoria on glycogen phosphorylase. Gen. comp. Endocr. 14,78-85. Goldsworthy, G. J. and Mordue, W. (1974). Neurosecretory hormones in insects (Review). J. Endocr. 60,529-558. Goodchild, B. and Smith, J. N. (1970). The separation of multiple forms of housefly l,l,ltrichloro-2,2-bis-(p-chlorophenyl)ethane (DDT) dehydrochlorinase from glutathione Saryltransferase by electrofocusing and electrophoresis. Biochem. J. 117, 1005-1009. Habermann, E. (1958). Zur Wirkung tierischer Gifte und von Lysocithin auf Grenzflachen. Z . Exp. Med. 130,19-23. Habermann, E. (1968). Biochemie, Pharmakologie und Toxikologie der Inhaltsstoffe von Hymenopterengiften. Ergeb. Physiol. Biol. Chem. Exp. Pharmakol. 60,220-325. Habermann, E. (1972). Bee and wasp venoms. Science, 177,314-322. Habermann, E. and Jentsch, J. (1966). Uber die Struktur des toxischen Bienengiftpeptides Melittin und deren Beziehung zur pharmakologischen Wirkung. Naunyn-Schmiedebergs Arch. exp. Path. Pharmak. 253,4041. Habermann, E. and Kowallek, H. (1970). Modificationen der Aminogruppen und des Tryptophans im Melittin als Mittel zur Erkennung von Struktur-Wirkungs-Beziehungen. Hoppe-Seyler’s Z. Physiol. Chem. 35 1,884-890. Habermann, E. and Reiz, K. G. (1964). Apamin: ein basisches, zentral erregendes Polypeptid aus Bienengift. Naturwissenschaften, 5 1,61. Habermann, E. and Reiz, K. G. (1965a).Ein neues Verfahren zur Gewinnung der Komponenten von Bienengift, insbesondere des zentralwirksamen Peptids Apamin. Biochem. Z. 341, 35 1-
466.
STRUCTURE AND FUNCTION OF INSECT PEPTIDES
125
Habermann, E. and Reiz, K. G. (1965b). Zur Biochemie der Bienengiftpeptide Melittin und Apamin. Biochem. Z. 343, 192-203. Habig, W. H., Pabst, M. J. and Jakoby, W. B. (1974a). Glutathione S-transferases. The first enzymatic step in mercapturic acid formation. J. Biol. Chem. 249, 7 130-7 139. Habig, W. H., Pabst, M. J., Fleischner, G., Gatmaitan, Z., Arias, I. M. and Jakoby, W. B. (197413). The identity of glutathione S-transferase B with ligandin, a major binding protein of liver. Proc. Nut. Acad. Sci. U.S.A. 71,3879-3882. Hanes, C. S., Hird, F. J. R. and Isherwood, F. A. (1950). Synthesis of peptides in enzymic reactions involving glutathione. Nature, Lond. 166,228-292. Hanson, J. M., Morley, J. and Soria-Herrera, C. (1974). Anti-inflammatory property of 410 (MCD-peptide), a peptide from the venom of the bee Apis mellifea (L). Brit. J. Pharmacol. 50,383-392. Hanstrom, B. (1953). Neurosecretory pathways in the head of crustaceans, insects and vertebrates. Nature, Lond. 171,72-73. Hart, D. E. and Steele, J. E. (1973). The glycogenolytic effect of the corpus cardiacum on the cockroach nerve cord. J. Insect Physiol. 19,927-939. Hartter, P. and Weber, U. (1975). Basische Peptide des Bienengifts. I. Isolierung, Reduktion und Reoxidation von Apamin und MCD-Peptid. Hoppe-Seyler’s Z. Physiol. Chem. 356,693-699. Haux, P. (1969). Die Aminosaurensequenz von MCD-Peptid, einem spezifisch mastzellendegranulierenden Peptid aus Bienengift. Hoppe-Seyler’s Z. Physiol. Chem. 350,536-546. Haux, P., Sawerthal, H. and Habermann, E. (1967). Sequenzanalyse des Bienengift-Neurotoxins (Apamin) aus seinen tryptischen und chymotryptischen Spaltstiicken. Hoppe-Seyler’s Z. Physiol. Chem. 348,737-738. Herp, A., Liska, M., Payza, N., Pigman, W. and Vittek, J. (1970). Ultratilterable peptides from rat liver and E. coli homogenates. FEBS Lett. 6,321-325. Highnam, K. C. (1961). The histology of the neurosecretory system of the adult female desert locust Schistocerca gregaria. Quart. J. Microscop. Sci. 102,27-38. Highnam, K. C. and Goldsworthy, G. J. (1972). Regenerated corpora cardiaca and hyperglycaemic factor in Locusta migratoria. Gen. comp. Endocr. 18,83-88. Hiss, E. A. and Fuchs, M. S. (1972). The effect of matrone on oviposition in the mosquito, Aedes aegypti. J. Insect Physiol. 18,2217-2227. Holdstock, D. J., Mathias, A. P. and Schachter, M. (1957). A comparative study of kinin, kallidin and bradykinin. Brit. J. Pharmacol. 12, 149-158. Hollingworth, R. M. (1970). The dealkylation of organophosphorus triesters by liver enzymes. In “Biochemical Toxicology of Insecticides” (Eds R. D. O’Brien and I. Yamamoto), p. 75. Academic Press, New York. Hollingworth, R. M., Metcalf, R. L. and Fukuto, T. R. (1967). The selectivity of Sumithion compared with methyl parathion. Metabolism in the white mouse. J. Agr. Food Chem. 15, 242-246. Holman, G . M. and Cook, B. J. (1970). Pharmacological properties of excitatory neuromuscular transmission in the hindgut of the cockroach, Leucophaea maderae. J. Insect. Physiol. 16, 1891-1907. Holman, G. M. and Cook, B. J. (1972). Isolation, partial purification and characterization of a peptide which stimulates the hindgut of the cockroach, Leucophaea maderae (Fabr.). Biol. Bull. 142,446-460. Hutson, D. H., Pickering, B. A. and Donninger, C. (1972). Phosphoric acid triester-glutathione alkyltransferase. A mechanism for the detoxification of dimethyl phosphate triesters. Biochem. J. 127,285-293. Ishida, M. (1968). Comparative studies on BHC metabolizing enzymes, DDT dehydrochlorinase and glutathione S-transferases. Agr. Biol. Chem. 32,947-955.
126
ROBERT P. BODNARYK
Ishida, M. and Dahm, P. A. (1965a). Metabolism of benzene hexachloride isomers and related compounds in vitro. I. Properties and distribution of the enzyme. J. Econ. Entomol. 58,383392.
Ishida, M. and Dahm, P. A. (1965b). Metabolism of benzene hexachloride isomers and related compounds in vitro. 11. Purification and stereospecificity of housefly enzymes. J. Econ. Entomol. 58,602-607. Iversen, L. L. (1974). Neuropharmacology of peptides. Nature, Lond. 252,630. Jaques, R. and Schachter, M. (1954). The presence of histamine, 5-hydroxytryptamine and a potent, slow contracting substance in wasp venom. Brit. J. Pharmacol. 9,53-58. Johnson, M. K. (1963). A distinct enzyme of rat liver and kidney coupling glutathione with some aliphatic halogen compounds. Biochem. J. 87,9-10P. Johnson, A. R. and Erdos, E. G. (1973). Release of histamine from mast cells by vasoactive peptides. Proc. SOC.Exp. Biol. Med. 142, 1252-1256. Junnikkala, E. (1968). A preliminary study of the character of peptide I (PI) in the hemolymph of Pieris brassicae L. Ann. Acad. Sci.fenn. A4, 132, 1-1 1. Junnikkala, E. (1976). A possible storage form of tyrosine in the haemolymph of Pieris brassicae L. and four other Lepidoptera. J. Insect Physiol. 22,95-100. Karlson, P. and Herrlich, P. (1965). Zum Tyrosinstoffwechsel der Insekten-XVI. Der Tyrosinstoffwechsel der Heuschrecke Schistocerca gregaria Forsk. J. Insect Physiol. 11, 7989.
Karlson, P. and Sekeris, C. E. (1964). Biochemistry of insect metamorphosis. I n “Comparative Biochemistry” (Ed. M. Florkin and H. S. Mason), Vol. 6, pp. 221-243. Academic Press, New York. Kater, S. B. (1968). Cardioaccelerator release in Periplaneta americana (L). Science, 160, 765767.
Kemper, B., Habener, J. F., Mulligan, R. C., Potts, J. T. and Rich, A. (1974). Preproparathyroid hormone: a direct translation product of parathyroid messenger RNA. Proc. Nut. Acad. Sci. U S A . 71, 3731-3735. Kindis-Mugge, I., Lane, C. D. and Kreil, G. (1974). Insect protein synthesis in frog cells: the translation of honey been promelittin messenger RNA in Xenopus oocytes. J. Mol. Biol. 87, 45 1-462.
King, T. P., Sobotka, A. K., Kochoumian, L. and Lichtenstein, M. (1976). Allergens of honey bee venom. Arch. Biochem. Biophys. 172,661-671. Kleeman, C. R. and Cutler, R. E. (1963). The neurohypophysis. Ann. Reo. Physiol. 25, 385432.
Knaak, J. B. and O’Brien, R. D. (1960). Effect of EPN on in uiuo metabolism of malathion by the rat and dog. J. Agr. Food Chem. 8, 198-202. Konishi, S. and Otsuka, M. (1974). Excitatory action of hypothalamic substance P on spinal motoneurones of newborn rats. Nature, L m d . 252,734-735. Kreil, G. (1973a). Structure of melittin isolated from two species of honey bees. FEBS Lett. 33, 241-244.
Kreil, G. (1973b). Biosynthesis of melittin, a toxic peptide from bee venom. Amino-acid sequence of the precursor. Eur. J. Biochem. 33,558-566. Kreil, G. (1975). The structure of Apis dorsata melittin: phylogenetic relationships between honey bees as deduced from sequence data. FEBS Lett. 54, 100-102. Kreil, G. and Bachmayer, H.(1971). Biosynthesis of melittin, a toxic peptide from bee venom. Detection of a possible precursor. Eur. J. Biochem. 20, 344-350. Kreil, G. and Kreil-Kiss, G. (1967). The isolation of N-formylglycine from a polypeptide present in bee venom. Biochem. Biophys. Res. Commun. 27,275-280. Kummer, H. ( 1960). Experimentelle Untersuchungen zur Wirkung von Fortpflanzungs-
STRUCTURE A N D FUNCTION OF INSECT PEPTIDES
127
Faktoren auf die Lebensdauer von Drosophila melanogaster Weibchn. Z . vergl. Physiol. 43, 642-679.
Leahy, M. G. (1966). Egg deposition in Dmelanogaster increased by transplant of male paragonia. Drosophila InJ Serv. 41, 145-146. Leahy, M. G. (1966). Egg deposition in D . melanogaster increased by transplant of male deposition in D. melanogaster. Life Sci. 6, 15 1-156. Leopold, R. A. (1976). The role of male accessory glands in insect reproduction. Ann. Rev. Entomol. 21, 199-22 1. Levenbook, L. (1966). Hemolymph amino acids and peptides during larval growth of the blowfly Phormia regina. Comp. Biochem. Physiol. 18, 341-35 1. Levenbook, L., Bodnaryk, R. P. and Spande, T. F. (1969). 8-alanyl-L-tyrosine. Chemical synthesis, properties and occurrence in larvae of the fleshfly Sarcophaga bullata Parker. Biochem. J. 113,837-841. Lewis, J. B. (1969). Detoxification of diazinon by subcellular fractions of diazinon-resistant and susceptible houseflies. Nature, Lond. 224, 917-9 18. Lewis, J. B. and Sawicki, R. M. (1971). Characterization of the resistance mechanisms to diazinon, parathion and diazoxon in the organophosphorus-resistant SKA strain of houseflies (Musca domestica L.). Pestic. Biochem. Physiol. 1,275-285. Lindauer, M. (1952). Ein Beitrag zur Frage der Arbeitseilung im Bienenstaat. Z . vergl. Physiol. 31,391-432.
Lipke, H. and Kearns, C. W. (1959a). DDT-dehydrochlorinase. I. Isolation, chemical properties and spectrophotometric assay. J. Biol. Chem. 234,2123-2128. Lipke, H. and Kearns, C. W. (1959b). DDT-dehydrochlorinase. 11. Substrate and cofactor specificity. J. Biol. Chem. 234, 2 129-2 132. Lipke, H. and Kearns, C. W. (1960). DDT-dehydrochlorinase. Adv. Pest. Control Res. 3, 253281.
Lowy, P. H., Sarmiento, L. and Mitchell, H. K. (1971). Polypeptides and melittin from bee venom: effects on Drosophila. Arch. Biochem. 145,338-343. Lubke, K., Matthes, S. and Kloss, G. (1971). Isolation and structure of N”-formyl melittin. Experientia, 21, 765-767. Lunan, K. D. and Mitchell, H. K. (1969). The metabolism of tyrosine-0-phosphate in Drosophila. Archs. Biochem. Biophys..!JZ, 450-456. Luscher, M. and Leuthold, R. (1965). Uber die hormonale Beeinflussung des respiratorischen Stoffwechsels bei der Schabe Leucophaea maderae. Revue Suisse Zool. 12, 618623.
Maccioni, R., Dinamarca, M. L. and Villar, D. E. (1970). DDT-dehidroclorinasa. Estudio cinetico. Arch. Biol. med. exp. I,25. Mach, B., Faust, C. and Vassali, P. (1973). Purification of 14s messenger RNA of immunoglobulin light chain that codes for a possible light-chain precursor. Proc. Natl. Acad. . Sci. U S A A10,451-455. Maddrell, S. H. P. (1974). Neurosecretion. In “Insect Neurobiology” (Ed. J. E. Treherne), Frontiers ofliology, 35, pp. 307-351. Markussen, J., Frandsen, E., Heding, L. G. and Sunby, F. (1972). Cited in Sundby, F., Frandsen, E. K., Thomsen, J., Kristiansen, K. and Brunfeldt, K. (1972). Crystallization and amino acid sequence of duck glucagon. FEBS Lett. 26,289-293. Mathias, A. P. and Schachter, M. (1958). The chromatographic behaviour of wasp venom kinin, kallidin and bradykinin. Brit. J. Pharmacol. 13,326-329. McDonald, T. J. (1975). Neuromuscular pharmacology of insects. Ann. Rev. Entomol. 20, 15 1166.
Meister, A. (1973). On the enzymology of amino transport. Science, 180, 33-39.
128
ROBERT P. BODNARYK
Meister, A. (1974). Glutathione: metabolism and function via the y-glutamyl cycle. Life Sci. 15, 177- 190. Meister, A. and Tate, S. S. (1976). Glutathione and related pglutamyl compounds: biosynthesis and utilization. Ann. Rev. Biochem. 45, 559-604. Merle, J. (1969). Nature des stimulations apportees par le male sur la physiologie des Drosophila femelles. Abst. 1st European Drosophila Res. Con$ The Hague, Netherlands. Merrifield, R. B. (1969). Solid-phase peptide synthesis. Adv. Enzymol. 32,221-296. Mordue, W. (1969). Hormonal control of Malipighian tube and rectal function in the desert locust, Schistocerca gregaria. J. Insect Physiol. 15,273-285. Mordue, W. and Goldsworthy, G. J. (1969). The physiological effects of corpus cardiacum extracts in locusts. Gen. Comp. Endocrinol. 12,360-369. Morella, A., Vardanis, A. and Spencer, E. Y. (1968). Mechanism of detoxication of some organophosphorus compounds: the role of glutathione-dependent demethylation. Can. J. Biochem. 46,885-892. Motoyama, N., Rock, G. C. and Dauterman, W. C. (1971). Studies of the mechanism of azinphosmethyl resistance in the predaceous mite, Neoseiulus (T.) fallacis (Family: Phytoseiidae). Pestic. Biochem. Physiol. 1, 205-215. Milstein, C., Brownlee, C. G., Harrison, R. M. and Matthews, M. B. (1972). A possible precursor of immunoglobulin light chains. Nat. New Biol. 239, 117-120. Mitchell, H. K., Lowy, P. H., Sarmiento, L. and Dickson, L. (1971). Melittin: toxicity to Drosophila and inhibition of acetylcholinesterase. Arch. Biochem. Biophys. 145,344-348. Mitchell, H. K. and Lunan, K. D. (1964). Tyrosine-0-phosphate in Drosophila. Archs. Biochem. Biophys. 106,2 19-222. Mitchell, H. K. and Simmons, J. R. (1962). In “Amino acid pools” (Ed. J. T. Holden), pp. 136146. Elsevier, Amsterdam. Murphy, T. A. and Wyatt, G. R. (1965). The enzymes of glycogen and trehalose synthesis in silkmoth fat body. J. biol. Chem. 240, 1500-1508. Nagai, T. (1970). Insect visceral muscle. Response of the proctodeal muscles to mechanical stretch. J. Insect Physiol. 16,437-448. Nagai, T. (1972). Insect visceral muscle. Ionic dependence of electrical potentials in the proctodeal muscle fibers. J. Insect Physiol. 18,2299-23 18. Nagai, T. (1973). Insect visceral muscle. Excitation and conduction in the protodeal muscles. J. Insect Physiol. 19, 1753-1764. Nagai, T. and Brown, B. E. (1969). Insect visceral muscle. Electrical potentials and contraction in fibers of the cockroach proctodeum. J. Insect Physiol. 15,215 1-2167. Nakajima, T., Pisano, J. J. and Udenfriend, S. (1967). Structure of the major kinin in wasp (Polistes) venom. Proc. Int. Congr. Biochem. 7th Col. Vol. VIII, 4. Natalizi, G. M. and Frontali, N. (1966). Purification of insect hyperglycaemic and heart accelerating hormones. J. Insecf Physiol. 12, 1279- 1287. Natalizi, G. M., Pansa, M. C., d’Ajello, V., Casaglia, O., Bettini, S. and Frontali, N. (1970). Physiologically active factors from corpora cardiaca of Periplaneta americana. J. Insect Physiol. 16, 1827-1836. Nelson, D. R., Adams, T. S. and Pomonis, J. G. (1969). Initial studies on the extraction of the active substance inducing monocoitic behaviour in houseflies, black flies, and screworm flies. J. econ. Ent. 62,634-639. Nelson, D. A. and O’Connor, R. (1968). The venom of the honeybee (Apis mellifera): free amino acids and peptides. Can. J. Biochem. 46, 1221-1226. Neuman, W. and Habermann, E. (1954). Beitrage zur Charakterisierung der Wirkstoffe des Bienengiftes. Naunyn-Schmeidebergs Arch. Exp. Path. Phormak. 222,367-387. Nicoll, R. A. and Barker, J. L. (1971). The pharmacology of recurrent inhibition in the supraoptic neurosecretory system. Brain Res. 35,501-5 11.
STRUCTURE A N D FUNCTION OF INSECT PEPTIDES
129
Nolan, J. and O’Brien, R. D. (1970). Biochemistry of resistance to paraoxon in strains of houseflies. J. Agr. Food Chem. 18,802-808. Norman, T. C. (1972). Heart activity and its control in the adult blowfly, Calliphora erythrocephala. J. Insect Physiol. 18, 1793-18 10. Norman, T. C. (1975). Neurosecretory cells in insect brain and production of hypoglycaemic hormone. Nature, Lond. 254,259-261. Norman, T. C. and Dave, H. (1969). Experimentally induced release of a neurohormone influencing hemolymph trehalose level in Calliphora eryfhrocephala (Diptera). Gen. comp. Endocr. 12,449-459. Oppenoorth, F. J., Rupes, V., El Bashir, S., Houx, N. W. H. and Voerman, S. (1972). Glutathione-dependent degradation of parathion and its significance for resistance in the housefly. Pestic. Biochem. Physiol. 2,262-269. Orlowski, M. and Meister, A. (1970). The yglutamyl cycle: a possible transport system for amino acids. Proc. Nut. Acad. Sci. U S A . 67, 1248-1255. Orlowski, M., Richman, P. G. and Meister, A. (1969). Isolation and properties of pL-glutamylcyclotransferase from human brain. Biochem. 8, 1048-1055. Otsuka, M., Konishi, S. and Takahashi, T. (1972). A further study of the motoneurondepolarizing peptide extracted from dorsal roots of bovine spinal nerves. Proc. Jap. Acad. 48, 747-752. Owen, M. D. and Braidwood, J. L. (1974). A quantitative and temporal study of histamine and histidine in honey bee (Apis mellifra L.) venom. Can. J. Zool. 52,387-392. Pabst. M. J., Habig, W. H. and Jacoby, W. B. (1974). Glutathione S-transferase. A. A novel kinetic mechanism in which the major reaction pathway depends on substrate concentration. J. Biol. Chem. 249, 7140-7150. Parke, D. V. (1968). “The Biochemistry of Foreign Compounds.” Pergamon Press, Oxford. Peck, M. L. and O’Connor, R. (1974). Procamine and other basic peptides in the venom of the honeybee (Apis mellifera).J. Agr. Food Chem. 22,51-53. Penzlin, H. (1966). The effect of neurohormone D on nuclear volumes of the prothoracic glands in Periplaneta americana L. (Blattaria). In “Insect Endocrines” (Ed. V. J. A. Novak). Brno. Perry, A. S. (1964). The physiology of insecticide resistance by insects. In “The Physiology of Insecta” (Ed. M. Rockstein), 1st Edn., Vol. 3, pp. 285-378. Academic Press, New York, London. Perry, A. S. and Agosin, M. (1974). The physiology of insecticide resistance by insects. In “The Physiology of Insecta” (Ed. M. Rockstein), 2nd Edn., Vol. 6, pp. 3-121. Academic Press, New York, London. Pisano, J. J. (1968). Vasoactive peptides in venoms. Fed. Proc., Fed. Am. SOC.Exp. Biol. 27, 38-62. Plapp, F. W. (1976). Biochemical genetics of insecticide resistance. Ann. Rev.Entomol. 21, 179197. Plapp, F.W. and Casida, J. E. (1958a). Hydrolysis of the alkylphosphorothioate insecticides by rat, cockroaches, and alkali. J. Econ. Enfomol.51, 800-803. Plapp, F. W. and Casida, J. E. (1958b). Bovine metabolism of organophosphorus insecticides. phosphorothioate in rats and a cow. Metabolic fate of O,O-dimethylO-(2,4,5-trichlorophenyl) J. Agr. Food Chem. 6,662-667. Prado, J. L., Tamura, Z., Furano, E., Pisano, J. J. and Udenfriend, S. (1966). Characterization of kinins in wasp venom. Zn “Hypotensive Peptides” (Eds E. G. Erdos, N. Back and F. Sicuteri), pp. 93-104. Springer-Verlag, Berlin, Heidelberg, New York. Ralph, C. L. (1962). Heart accelerators and decelerators in the nervous system of Periplanefa americana. J. Insect PhysioI. 8,43 1-439. Ralph, C. L. and McCarthy, R. (1964). Effects of brain and corpus cardiacum extracts on
130
ROBERT P. BODNARYK
haemolymph trehalose of the cockroach, Periplaneta americana. Nature, Lond. 203, 1 195-
1196. Reed, W. T. and Forgash, A. J. (1970). Metabolism of lindane to organic-soluble products by houseflies. J. Agr. Food Chem. 18,475-481. Richter, K. (1967). Untersuchungen zum Wirkungsmechanismus von Neurohormon D am Herzen von Periplaneta americana. Zool. Jb. Physiol. 73,261-275. Richter, K. and Gersch, M. (1974). Untersuchungen iiber die Wirkung cholinerger, aminerger und peptiderger Substanzen am lateralen Herznerven und den funktionellen Zusammenhang dieser Wirkungen bei der Herzregulation von Blaberus craniifer Burm. (Insecta: Blatteria). Zool. Jahrb.. Abt. Allg. Zool. Physiol. Tiere, 78, 16-32. van Rietschoten, J., Granier, C., Rochat, H., Lissitzky, S. and Miranda, F. (1975).Synthesis of apamin, a neurotoxic peptide from bee venom. Eur. J. Biochem. 56,3540. Roberston, H. A. and Steele, J. E. (1972). Activation of insect nerve cord phosphorylase by octopamine and adenosine 31,51-monophosphate.J. Neurochem. 19,1603-1 606. Robertson, H. A. and Steele, J. E. (1973a). Octopamine in the insect central nervous system: distribution, biosynthesis and possible physiological role. J. Physiol. 237,34-35P. Robertson, H. A. and Steele, J. E. (1973b). Effect of monophenolic amines on glycogen metabolism in the nerve-cord of the American cockroach, Periplaneta americana. Insect Biochem. 3,53-59. Robison, G. A., Butcher, R. W. and Sutherland, E. W. (1968). Cyclic AMP. Ann. Reo. Biochem. 37,149-174. Roussel, J. P.and Cazal, M. (1969). Action, in oiuo, d’extraits de corpora cardiaca sur le rythme cardiaque de Locusta migratoria L. C.r. hebd. Seanc. Acad. Sci., Paris 26858 1-583. Rutter, F. and Maul, V. (1969).The cause of the hybridization barrier between Apis mellifera L. and Apis cerana Fabr. (=syn. A. indica Fabr.). 2. Egg fertilization and embryonic development. Transct. Beekeeping Congress, Munich Vol. XXI, pp. 5 10-5 1 1. Ryan, J. W.,Roblero, J. and Stewart, J. M. (1970). Inactivation of bradykinin in rat lung. In “Bradykinin and Related Kinins, Cardiovascular, Biochemical and Neural Actions” (Eds F. Sicuteri, M. Rochaesilva and N. Back), Ado. Exptl. Med. Biol. 8,263-271. Schachter, M. (1963). Kinins of different origins. Ann. New York Acad. Sci. 104,108-1 16. Schachter, M. and Thain, E. M. (1954). Chemical and pharmacological properties of the potent, slow contracting substance (kinin) in wasp venom. Brit. J. Pharmacol. 9,352-359. Scharrer, B. (1952). Neurosecretion. XI. The effects of nerve section on the intercerebraliscardiacum-allatum system of the insect Leucophaea maderae. Biol. Bull. Wood’s Hole, 102,
261-272. Scharrer, B. and Kater, S. B. (1969). Neurosecretion. XV.An electron microscope study of the corpora cardiaca of Periplaneta americana after experimentally induced hormone release. Z . Zellforsch. 95, 177-186. Schechter, I. (1973). Biologically and chemically pure mRNA coding for a mouse immunoglobulin L-chain prepared with the aid of antibodies and immobilized oligothymidine. Proc. Nut. Acad. Sci. U S A . 70,2256-2260. Schroder, E.,Liibke, K., Lehmann, M. and Beetz, I. (1971). Haemolytic activity and action on the surface tension of aqueous solutions of synthetic melittins and their derivatives. Experientia, 27,764-765. Sessa, G., Freer, J. H., Colacicco, G. and Weissmann, G. (1969). Interaction of a lytic polypeptide, melittin, with lipid membrane systems. J. Biol. Chem. 244, 3575-3582. Shipolii, R., Bradbury, A. F., Callewaert, G. L. and Vernon, C. A. (1967). The structure of apamin. Chem. Commun. 1967,679480. Shishido, T., Usui, K., Sato, M. and Fukami, J. (1972). Enzymatic conjugation of diazinon with glutathione in rat and American cockroach. Pest. Biochem. Physiol. 2 , s 1-63.
STRUCTURE A N D FUNCTION OF INSECT PEPTIDES
131
Sienkiewicz, Z. and Piechowska, M. J. (1973). L-tyrosyl-0-acetyldopamine, a new dipeptide found in the haemolymph of caterpiller of Celerio euphorbiae L. (Lepidoptera). Bull. Acad. Pol. Sci.. Ser. Sci. Biol. 21, 797-802. Simmons, J. R. and Mitchell, H. K. (1962). Metabolism of peptides in Drosophila. I n “Amino acid pools” (Ed. J. T. Holden), pp. 147-157. Elsevier, Amsterdam. Sims, P. and Grover, P. L. (1965). Conjugations with glutathione. The enzymic conjugation of some chlorocyclohexenes. Biochem. J. 95, 156-160. Smith, J. N. (1955). Detoxication mechanisms in insects. Biol. Revs. Cambridge Phil. SOC.30, 455-475. Smith, J. N. (1962). Detoxication mechanisms. Annu. Reo. Entomol. 7,465480. Smith, P. D. and Bischoff, W. L. (1969). Presence of sex peptide in Drosophila melanogaster intersexes. Drosophila In$ S e w . 44, 122. Sobotka, A., Franklin, R., Valentine, M., Adkinson, N. F. and Lichtenstein, L. (1974). Honeybee venom: phospholipase A as the major allergen. J. Allergy Clin. Immunol. 53, 103 abstr. Speir, T. W. and Barnsley, E. A. (1971). The conjugation of glutathione with unsaturated acyl thiol esters and the metabolic formation of S-carboxyalkylcysteines. Biochem. J. 125, 267273. Starratt, A. N. and Brown, B. E. (1975). Structure of the pentapeptide proctolin, a proposed neurotransmitter in insects. Life Sci. 17, 1253-1256. Steele, J. E. (1961). Occurrence of a hyperglycaemic factor in the corpus cardiacum of an insect. Nature, Lond. 192, 680-68 1. Steele, J. E. (1963). The site of action of insect hyperglycaemic hormone. Gen. comp. Endocr. 3, 46-52. Steele, J. E. (1964). The activation of phosphorylase in an insect by adenosine 3I,5’-phosphate and other agents. Am. Zool. 4,237. Stenersen, J. (1969). Demethylation of the insecticide bromophos by a glutathione-dependent liver enzyme and by alkaline buffers. J. Econ. Entomol. 62, 1043-1045. Stevens, R. H. and Williamson, A. R. (1972). Specific IgG mRNa molecules from myeloma cells in heterogeneous nuclear and cytoplasmic RNA containing ply-A. Nature, Lond. 239, 143146. Suchanek, G., Kindas-Mugge, I. and Kreil, G. (1975). Translation of honeybee promelittin messenger RNA. Formation of a larger product in a mammalian cell-free system. Eur. J. Biochem. 60,309-3 15. Sundby, F., Frandsen, E. K., Thomsen, J., Kristiansen, K. and Brunfeldt, K. (1972). Crystallization and amino acid sequence of duck glucagon. FEBS Lett. 26,289-293. Tager, H. S., Markese, J., Spiers, R. D. and Kramer, K. J. (1975). Glucagon-like immunoreactivity in insect corpus cardiacum. Nature, Lond. 254,707-708. Takahashi, T., Konishi, S., Powell, D., Leeman, S. E. and Otsuka, M. (1974). Identification of the motoneuron-depolarizing peptide in bovine dorsal root as hypothalamic substance P. Brain Res. 73,5949. Thomsen, J., Kristiansen, K. and Brunfeldt, K. (1972). The amino acid sequence of human glucagon. FEBS Lett. 21, 315-319. Traina, M. E., Bellino, M. and Frontalii N. (1974). A comparison between neurohormone D and “peak 1” extracted from cockroach corpora cardiaca. Zool. Jb. Physiol. 78, 424428. Traina, M. E., Bellino, M., Serpietri, L., Massa, A. and Frontali, N. (1976). Heat-accelerating peptides from cockroach corpora cardiaca. J. Insect Physiol. 22,323-329. Tregear, G. W. (1972). In “Chemistry and Biology of Peptides” (Ed. J. M. Meienhofer), pp. 175-178. Ann Arbor Science Publ., AM Arbor, Michigan.
132
ROBERT P. BODNARYK
Tregear, G. W., Niall, H. D., Potts, J. T., Leeman, S. E. and Chang, M. M. (1971). Synthesis of substance P. Nature, New Biol. 232,87-89. Treherne, J . E. (1965). The neurochemistry of arthropods. Cambridge Monographs in Experimental Biology. No. 14. Cambridge University Press, London. Unger, H. (1957). Untersuchungen zur neurohormonalen Steuerung der Herztatigkeit bei Schaben. Biol. Zbl. 16,204-225. Unger, H. (1965). Der Einfluss der Neurohormone C and D auf die FarbstoffAbsorptionsfahigkeit der Malpighischen Gefhse (und des Dames) der Stabheuschrecke Carausius morosus (Br.) in vitro. Zool. Jb. Physiol. 71, 7 1 G 7 17. Usherwood, P. N. R. (1974). Nerve-muscle transmission. I n “Insect Neurobiology” (Ed. J. E. Treherne). Frontiers of Biology, 35, pp. 245-305. Vejbjerg, K. and Norman, T. C. (1974). Secretion of hyperglycaemic hormone from the corpus cardiacum of flying blowflies, Calliphora erythrocephala. J. Insect Physiol. 20, 1189-1 192. Vernon, C. A., Hanson, J. M. and Brimblecombe, R. W. (1969). Peptides, British Patent No. 1324823. Vincent, J.-P., Schweitz, H. and Lazdunski, M. (1975). Structure-function relationships and site of action of apamin, a neurotoxic polypeptide of bee venom with an action on the central nervous system. Biochem. 14,2521-2525. Watanabe, M., Yasuhara, T. and Nakajima, T. (1975). New bradykinin analogues in wasp (Polistes rothneyi iwatai) venom. Toxicon, 13, 130. Wellhoner, H. H. (1 969). Spinale Wirkungen von Apamin. Nauyn-Schmiedeberg Arch. Pharmakol. Exp. Pathol. 262,29-41. van der Werf, P., Orlowski, M. and Meister, A. (1971). Enzymatic conversion of 5-0x0-Lproline (L-pyrrolidine carboxylate) to L-glutamate coupled with cleavage of adenosine triphosphate to adenosine diphosphate, a reaction of the y-glutamyl cycle. Proc. Nat. Acad. Sci. U S A . 68,2982-2985. Wiens, A. W. and Gilbert, L. I. (1967a). Regulation of carbohydrate mobilization and utilization in Leucophaea maderae. J. Insect Physiol. 13,779-794. Wiens, A. W. and Gilbert, L. I. (1967b). The phosphorylase system of the silkmoth, Hyalophora cecropia. Comp. Biochem. Physiol. 21, 145-159. Wigglesworth, V. B. (1960). The nutrition of the central nervous system in the cockroach Periplaneta americana L. The role of the perineurium and glial cells in the mobilization of reserves. J. Exp. Biol. 31, 500-5 13. de Wilde, J. and de Loof, A. (1973). Reproduction. In “The Physiology of Insecta” (Ed. M. Rockstein), I, pp. 12-158. Academic Press, New York. Wyatt, G. R. (1967). The biochemistry of sugars and plysaccharides in insects. In “Advances in Insect Physiology”, Vol. 4,287-360. Academic Press, New York and London. Wyatt, G. R. (1972). Insect hormones. In “Biochemical Actions of Hormones” (Ed. G. Litwack), Vol. 2, pp. 385-490. Academic Press, New York and London. Yang, C. C. (1974). Chemistry and evolution of toxins in snake venoms. Toxicon 12, 1-43. Yang, R. S. H., Hodgson, E. and Dauterman, W. C. (1971). Metabolism in vitro of diazinon and diazoxon in susceptible and resistant houseflies. J. Agr. Food Chem. 19, 14-19. Yoshida, H., Geller, R. G. and Pisano, J. J. (1976). Vespulakiniins:new carbohydrate-containing bradykmin derivatives. Biochem. 15,61-64.
Insect Flight Metabolism Ann E. Kammer' and Bernd Heinrich2
' Division of Biology, Kansas State University, Manhartan. Kansas, USA
Division of Enromology. University of California, Berkeley, California, USA
1 Introduction 134 2 The metabolic rate during flight 134 2.1 Ambient temperature 137 2.2 Wing-loading and body mass 139 2.3 Flightspeed 143 2.4 Cost of transport 146 2.5 Ecology and evolution of high metabolic rates 146 3 Neural control of power output 147 3.1 Insects with neurogenic rhythms 147 3.2 Insects with myogenic rhythms 151 3.3 Novel aerodynamic mechanisms 154 3.4 Some comparisons and conclusions 155 4 Supplying the energy demanded: Control of flight metabolism 156 4.1 Insect flight muscles 156 4.2 Oxygen supply 157 161 4.3 Biochemical processes in flight muscle 4.4 Mobilization of stored fuels 169 4.5 Methods for studying flight muscle metabolism 171 4.6 Hormonal control mechanisms 173 4.7 Hemolymph circulation 178 4.8 Substrate availability and flight speed 179 5 Interrelations of flight muscle temperature and metabolic rate 180 5.1 Effects of temperature on the flight motor 181 5.2 Pre-flight warm-up 184 5.3 Stabilization of thoracic temperature during flight 190 5.4 Shivering and nonshivering thennogenesis 19 1 5.5 Why require a high muscle temperature? 195 6 Development and senescence 197 6.1 Hemimetabolous insects 198 6.2 Lepidoptera 200 6.3 Holometabolous insects with fibrillar muscles 203 6.4 Adult diapause, regeneration and polymorphism 206 6.5 Maturation related to use and disuse 208
133
ANN E. KAMMER AND BERND HEINRICH
134
7
1
6.6 Hormonal control 209 6.7 Age and flight metabolism Conclusions 211 Acknowledgements 21 1 References 2 12
210
Introduction
Actively flying insects achieve the highest metabolic rates known, and they do so in the fraction of a second required to shift from quiescence to flight. The various adaptations that make possible the high metabolic rates necessary for flight constitute the subject of this review. Flight depends on the biochemical and mechanical work done by the flight muscles, which must be continually supplied with oxygen and fuel. The work of the muscles is under neural control and therefore the metabolic rate is also under neural control. Hormones participate as part of the biochemical mechanisms by which the neural commands are executed and also as part of the internal milieu supportive of flight. In larger insects, high metabolic rates and the associated heat production result in elevated body temperatures; temperature effects and temperature regulation are thus closely related to flight and they are considered extensively in the following discussion. Much has been written about the flight of insects. Recent reviews have already covered the biochemistry of the flight fuels (Bailey, 1975) and their utilization (Sacktor, 1970, 1975; Crabtree and Newsholme, 1975), nervous coordination (Wilson, 1968), ventilation (Miller, 1966, 1974; Kammer, 1976) and temperature control of the flight muscles (Heinrich, 1974a). Other reviews have considered aerodynamic problems (Lighthill, 1975; Weis-Fogh, 1975), the comparative physiology-anatomy of flight mechanisms (Pringle, 1957, 1968, 1974), and migration (Johnson, 1969,1974; Rainey, 1976). We shall not attempt to provide another review of the above-mentioned aspects of insect flight, but shall draw on these areas insofar as they concern intensity and regulation of flight metabolism.
2
The metabolic rate during flight
Energy expended in flight appears partly as aerodynamic work and partly as heat. The work output must first of all be sufficient to counteract the force of gravity, and secondly, the work done must provide forward thrust. The energy input for flight, on the other hand, must be much greater than the work output of the wings on the air, because of biochemical and mechanical inefficiencies. Typically the muscles are no more than 20 per cent biochemically efficient (i.e.
135
INSECT FLIGHT METABOLISM
TABLE 1 Rates of oxygen consumption of insects at rest and in flight* Metabolic rate
ml O,/g body wt/h
Species
At rest
W * N-' in flight
Reference
In flight
DICTYOPTERA
Periplaneta americana
0.36
36
21
Polacek and Kubista (1960)
ORTHOPTERA
Schistocerca gregaria
10-30 (45) 6-18 (27)
Krogh and Weis-Fogh (1951);Weis-Fogh (1952)
-
17-24
10-14
Weis-Fogh (1967)
0.55 0.7 0.75 0.73 -
54 92 56 51 40-90 64 55 59 82 (105) 29 43 7
32 55 33 30 24-53 38 33 35 49 (62) 17 25 4
Zebe (1954) Zebe (1954) Zebe (1954) Heinrich (1971) Casey (1976a) Zebe (1954) Heinrich and Casey (1973) Heinrich and Casey (1973) Zebe (1954) Zebe (1954) Zebe (1954) Nayar and Van Handel (197lb)
57 (1 11) 36 (61) 12 (18)
0.63
ODONATA
Aeschna grandis LEPIWPTERA
Vanessa w Metopsilus procellus Mimas tiliae Manduca sexta Hyles lineata Hyles euphorbia Hyles euphorbia Deilephila elpenor Saturnia pavonia Antheraea pernyi Triphaenapronuba Spodopterafmgiperda
1
~
DIPTERA
Lucilia sericata Musca vomitoria Drosophila repleta
-
1.8
96 (188) 60 (103) 21 (30)
Drosophila melanogaster Drosophila melanogaster Drosophila gibberosa
5 .O -
12 (22) 33 (40) 19
7 (13) 20 (24) 11
Davis and Fraenkel(l940) Axenfeld (1911) Chadwick and Gilmour ( 1940) Wigglesworth (1949) Hocking (1953) Sotavalta and Laulajainen
Drosophila hydei
-
20
12
Sotavalta and Laulajainen
Drosophilafunebris
-
10
6
Sotavalta and Laulajainen
Tabanus aflnis Tabanus septentrionalis A edesflavescens Aedes nearcticus Simulium venustum
0.8 1.1 3.8 5.6 4.4
22 (63) 56 (68) 22 21 (58) 27 (37)
(1961) (1961) (1961)
13 (37) 33 (40) 13 12 (34) 16 (22)
Hocking (1953) Hocking (1953) Hocking (1953) Hocking (1953) Hocking (1953)
136
ANN E. KAMMER AND BERND HEINRICH
TABLE l-continued Metabolic rate Species
ml O,/g body wtlh At rest
In flight
W * N-' in flight
Simulium vittatum Eristalis tenax
4.8 -
50 (53) 23
30 (3 1) 14
Fannia caninrlaris
-
23
14
Glossina morsitans
-
90
53
Apis meififera
2.0
87 (100)
52 (59)
Apis mellifera Apis mellifera Apis mellifera Apis mellifera Bombus uosnesenskii
3.2 0.5 1.3
60 (98) 312 70 60 55-66
36 (58) 185 42 36 33-39
Bombus spp. Vespa crabro
-
80 ( 110) 17-24
47 (65) 10-14
Reference
Hocking (1953) Sotavalta and Laulajainen (1961) Sotavalta and Laulajainen (1961) Hargrove (1976)
HYMENOPTERA
Jongbloed and Wiersma (1934) Hocking (1953) Kosmin el al. (1932) Bastian and Esch (1970) Sotavalta (1954b) Kammer and Heinrich (1974) Heinrich (1975) Weis-Fogh (1967)
In most cases oxygen consumption was measured; in some cases it was calculated from other data [see Hocking (1953) for discussion of the older work]. We have converted these values into SI units, using 1 L 0, = 5 kcal for all substrates, and 1 cal-h-'-gf" = 0.1186 W-N-'. Flight values are means and, in parentheses, maxima. In this table we have not distinguished among free, tethered on a flight mill, or fixed flight; the oxygen consumption in free flight can be double that in fixed flight.
the ratio of mechanical work done to metabolic cost is approximately 0.2, although values can range from 0.35 to -1.2 (Hill,1939; Weis-Fogh, 1972; Tucker, 1973, 1975)). Of the mechanical work performed by the muscles, only a fraction is aerodynamically useful. The efficiency varies among different kinds of flight, such as fast forward flight, gliding, or hovering. For the latter, WeisFogh (1972) calculated an aerodynamic efficiency (momentum imparted to the aidtotal aerodynamic power) of 50 per cent for a hummingbird and 30 per cent for Drosophilu. Thus for a hovering Drosophilu, for example, of every 100 calories expended, only about 6 calorites (100 x 0.2 x 0.3) result in useful work. The work input, or total energy expended during flight, is reflected in the metabolic rate. Measured and calculated metabolic rates of flying insects vary
INSECT FLIGHT METABOLISM
137
-
over a large continuum. In general, most values fall between about 12 to 60 W N-I (Le., 100 to 500 cal g-' * h-l) (Table 1). These metabolic rates are among the highest known. They represent 50- to 100-fold increases over the resting rate. In comparison small mammals running at maximal speed have metabolic rates of 1-4 W. N-', only 7 times greater than resting rates, and flying birds show a similar 7- to 14-fold increase (Schmidt-Nielsen, 1972). The metabolic rate of a hovering hummingbird, 24 W-N-' (Lasiewski, 1963), is comparable to that of many insects, however. Hovering flight places heavy energetic demands on small birds and bats as well as on insects (Weis-Fogh, 1972), and it is this mode of locomotion, not the systematic position of the animal, that demands high metabolic rates. The large range in the measured metabolic rate reflects intrinsic differences between species as well as different conditions of measurement. The full range of possible rates is probably not yet known. For example, very low values of metabolic rate possible during gliding flight such as that of some butterflies (Nachtigall, 1967) and some dragonflies (May, 1976) have not been measured in the laboratory, where flight metabolism has largely been analysed in relatively small respirometers under conditions such that gliding flight is not possible. Free hovering flight, which is most expensive, has been measured, however, and these results probably represent maximal values for these species. Metabolic rate during flight may be influenced by ambient temperature, load, and flight speed. These factors will be discussed in the following sections.
2.1
AMBIENT TEMPERATURE
The maximum rate at which a muscle can transform chemical energy into mechanical work depends on muscle temperature. If ambient temperature determines the thoracic temperature, then the ambient temperature will strongly influence the metabolic rate. If an insect can regulate its temperature behaviorally or physiologically, then its metabolic rate will be relatively independent of ambient temperature. In a number of species of sphinx moths (Heinrich, 1971a; Heinrich and Casey, 1973; Casey, 1976a), bumblebees (Kammer and Heinrich, 1974; Heinrich, 1975) and syrphid flies (Heinrich and Pantle, 1975), muscle temperature is stabilized during flight over a relatively wide range of ambient temperature, and metabolic rate remains constant (Fig. 1). In these large insects, then, metabolic rate during flight is independent of ambient temperature, but energy is expended prior to flight to raise the thoracic temperature to a level at which flight is possible (section 5). The locust Schistocercu greguriu does not regulate its temperature during flight, at least during tethered flight (Weis-Fogh, 1956a). One might therefore expect that in this insect metabolic rate would be influenced by ambient
138
ANN E. KAMMER AND BERND HEINRICH
60
- :
-_ 0
0
::
.
.
640-
:
- * 20
1
I
I
I
I
I
.
? d
3
I
I
I
a
0
0
-.---
0 0
-------5-? 0
0
00
8
c
0 .c
I
0
0
Y
Q
i
(
3 1 80-
0
.
100
9, \
*
.5
-
0
<
.-
-
L
-2-0 0
60-
v)
c
0 0
c
40-
m
Y = 81.4
0
20-
0
- 0.24X
8. edvardsii 8. vosnesens&ij
temperature. However, because variables that influence metabolic rate, namely flying speed, lift, and wingstroke frequency, were independent of ambient temperature, Weis-Fogh (1956a) concluded that the metabolic rate of flying Schistocerca gregaria was independent of ambient temperatures between 2 5 O and 3 5 O C.
139
INSECT FLIGHT METABOLISM
In the case of small insects or some large, uninsulated ones, body temperatures probably equal or parallel ambient temperatures during flight. We know of no studies showing directly that metabolic rates during flight depend on ambient temperature in these insects. Indirect evidence comes from Soltavalta's studies (1947, 1954a) and Sotavalta and Laulajainen (1961), showing that the wingbeat frequency of some myogenic fliers is a function of ambient temperature (Fig. 2), and metabolic rates reflect wingbeat frequency. Wingbeat frequency increases with increasing ambient and thoracic temperature in 600
500. 400. 350.
300 250 200
300,
150
p 250,
100 80
c
I! :z'
60
160 140
40
I20 1
5
6
7 8 9 10
15
I
I
I
I
I
I
20 25 3 0 3 5 4 0 4 5
- 1 10
15 20
30 4050
Ambient temp. ("C)
Fig. 2. Effect of temperature on wingbeat frequency. (a) Culicidae spp. males, (b) Culicidae spp. females, (c) Drosophila melanogasler, (d) D. funebris, (e) D . repleta, (f) Apis mellifera worker, (g) Vespa vulgaris, (h) Bombus agrorum, (i) Cantharis sp., (j) Absyrtus luteus (from Soltavalta, 1947).
cockroaches, Periplunefa spp. (Farnworth, 1972a,b), in contrast to locusts, and ambient temperature probably interacts with flight effort to determine metabolic rate in these insects. 2.2
WING-LOADING A N D BODY M A S S
The smaller the surface area of the wings relative to body weight, the more rapidly the wings must beat, or the greater the force per wingstroke, or both, in order to keep the animal in flight. Wing-loading (body weighthing area) varies both among species having differently sized wings, as well as among individuals of the same species carrying different loads. Because of the number of factors that influence flight, neither mass nor wing-loading correlate well with metabolic rate in flight, when different kinds of insect are compared (Table 2).
2
P
0
TABLE 2 Metabolic rates of insects with different mass, wing-loading (body weighdwing area), and wingbeat frequency Wing-loading
Species Schistocerca gregaria Manduca sexta Bombus spp. Eristalis tenax Tabanus afinis Apis mellifera T. septentrionalis Syrphus spp. Simulum venustum Aedes nearcticus S. vittatum Drosophila spp. D. melanogaster
2 1.2 0.2 0.14-0.18 0.16 0.105 0.06 0.027 0.003 0.003 0.002 0.002 0.0009
(N.m-*)
Wingbeat frequency (s-9
Metabolic rate W .N-'
7 8 10 10 14 19 10 7 4 4 2 3.5 3
18-20 24-26 14C170 166-185 96-149 200 (178-208) 98 (58-128) 180-200 258 (198-280) 3 18 (305-380) 209 (170-224) 240 208 (185-224)
12 (6-17) 28-33 48 14 13 37 34 52 17 12 30 13
20
Ref.
192 3,4 5,6 7, 8 9 9 9 10,ll Z 9 9 m 9 12, 13 6 9
4
5...
3
p
References: (1) Weis-Fogh, 1952; (2) Weis-Fogh, 1956; (3) Heinrich, 1971a; (4) Casey, 1976b; ( 5 ) Soltavalta, 1947; (6) Heinrich, 1975; (7) Soltavalta, 1952; (8) Soltavalta and Laulajainen, 1961; (9) Hocking, 1953; (10) Heinrich and Pantle, 1975; (1 1) Weis-Fogh, 1973; (12) Vogel, 1966; 0 W (13)Chadwick and Gilmour, 1940. rn 3 Values are means or typical figures, unless the original data provided a range of values. Metabolic rates were determined with different methods but in most cases represent free flight. T
6
INSECT FLIGHT METABOLISM
141
The poor correlation between mass per se and metabolic rate (Weis-Fogh, 1964b, 1975) can be illustrated by some examples. Both Drosophilu ( z 2 mg) 0 have metabolic rates of about 14 W * N-'. and Schistocercu ( ~ 2 0 0 mg) Within the family of sphinx moths (Sphingidae) the calculated (not measured) average aerodynamic power of hovering in 8 species weighing 1.0 to 1.4 g was 0.89 W.N-', whereas the average aerodynamic power of 10 species weighing 0.1 to 0.6 g was 0.79 W.N-' (Casey, 1976b); the difference is small and probably insignificant inasmuch as many of the intermediate-sized moths had the highest hovering costs. Similarly, the calculated aerodynamic power for 23 species of different orders does not vary systematically with size (Weis-Fogh, 1973, 1975). The highest value in Weis-Fogh's sample was 4.7 W-N-' for a heavy beetle, Heliocorpus sp. (12.8 g), the lowest was 1.3 W.N-' for the hawkmoth Munducu sextu (2.1 g), whereas several small insects (<0.3 g) had intermediate values of 2.3 WaN-'. This independence of metabolic rate and body weight of flying insects contrasts with the inverse relationship seen in flying birds and bats (Thomas, 1975) and running mammals (Schmidt-Nielsen, 1972). One explanation for the similar metabolic rates of large and small flying insects is that over a wide size range the power-producing muscles occupy a relatively constant fraction (1Ck20 per cent) of the body mass (Greenewalt, 1962). If there are no intrinsic differences among the flight muscles, power output depends on muscle volume, that is, on the product of cross-sectional area and length, since the former determines the force produced and the latter the shortening velocity, and power equals force times shortening velocity (Weis-Fogh, 1975; Josephson, 1975). But there may be intrinsic differences (Josephson, 1975). Details may modify the above generalities and explain the 5-fold differences in observed metabolic rates. For example, in some Lepidoptera only 5-10 per cent of the body mass is muscle (Greenewalt, 1962). Also the metabolic rate per gram muscle differs in different insects (Table 3). Metabolic rate may vary with body weight, however, when one individual or one species is considered. Within one species of sphingid moth, Hyles lineutu, oxygen consumption was inversely related to body weight. The average weightspecific rate of oxygen consumption of 1.5 g moths was about 38 ml OJg .h whereas moths weighing only 0.5 g consumed about 73 ml O,/g-h (Casey, 1976a). Since, in H. lineutu, approximately 45 per cent of the body mass of 0.5 g moths consists of thorax, while only 20 per cent of 1.5 g moths consists of thorax (Casey, 1976b), the inverse correlation may be primarily a reflection of the percentage of inactive tissue. This idea gains support from metabolic data of freely flying bumblebees (Heinrich, 1975). The heavier an individual bee due to loading of the honey stomach in the abdomen, the lower its overall metabolic rate on the basis of body weight, but the greater the
ANN E. KAMMER AND BERND HEINRICH
142
metabolic rate as a function of thoracic weight. Oxygen consumption varies from about 170 mug thorax-h when the abdomen weighs 150 mg, to about 330 ml/g thorax. h when the abdomen is loaded to 300 mg (Fig. 3). These data indicate that, within any one animal or animals of the same species, the effects of body mass or wing-loading on metabolic rate may not be apparent unless metabolic rate of the wing muscles, rather than whole body weight, is considered. TABLE 3 Oxygen consumption during flight in relation to differences among flight muscles ml O,/
PI OJ
Animal
g muscle min
g muscle. spike
Schistocerca gregaria Locust half-thorax (extrapolated to flight) Manduca sexta Hyles lineata large dipteran flies Drosophila melanogaster Bombus spp. Apis mellSfera
0.9-2.8 2.5 4.9 4.2 8-16 2.2-2.8 8 6.7-8
1.3 1.o 1.6 1.7 18-22 ca. 5 3.3-3.7 8.3
.
Ref. 192 3 495
6,7 8,9, 10 11,12 13, 14 11, 15
References: (1) Weis-Fogh, 1952; (2) Wilson and Weis-Fogh, 1962; (3) Candy, 1970; (4) Heinrich, 1971a; (5) Kammer, 1971; (6) Casey, 1976a; (7) Kammer, 1968; (8) Davis and Fraenkel, 1940; (9) Wyman, 1970; (10) Nachtigall and Wilson, 1967; (1 1) Hocking, 1953; (12) Harcombe, 1975; (13) Heinrich, 1975; (14) Kammer and Heinrich, 1974; (15) Bastian and Esch, 1970.
Muscle masdbody mass from Greenewalt (1962) if not given in studies cited; confirmed by our unpublished measurements on M.sexta, H.lineata, and Bombus.
The effects of wing-loading on metabolic rate may be apparent on the basis of whole body weight when animals of different species having similar modes of flight are compared. For example, both H. Iineata and Manduca sexta have similar power inputs as a function of given wing-loading, even though the average metabolic rate of H. lineata is higher than that of M. sexta (Heinrich and Casey, 1973; Casey, 1976b). For 5 species of sphingid moths the power input per unit mass during hovering increases with wing-loading (Casey, 1976b). However, when insects of different orders are compared (Table 2), there is no correlation between wing-loading and metabolic rate. Additional factors such as flight speed and wingbeat frequency need to be considered. It would be interesting to have metabolic and flight data for a number of species within groups with similarly shaped wings but a large range of wing-loading, e.g. within the large Diptera (Brachycera and Cyclorrhapha) or the Odonata, in which wing-loading varies by a factor Of 4 (Weis-Fogh, 1973). It would also
143
INSECT FLIGHT METABOLISM
be interesting to have metabolic rates of insects at the extremes of the spectrum. At the upper end of the spectrum are lamellicorn beetles with a wingloading of 12-10 N m-2 (Weis-Fogh, 1973, 1975). Analysis of this group would be complicated but perhaps illuminating because of the structural and functional differences between forewings (elytra) and hindwings. Under tethered conditions the elytra produce lift equal to 14 per cent of the body weight primarily passively, although during the active downstroke they provide
-
/
1
0
I
I00
I
I
200
I
I
1
300
Weight of abdomen (mg) Fig. 3. Oxygen consumption per gram thoracic weight as a function of abdominal weight, in a single Bornbus edwardsii fed on different amounts of sugar solution (from Heinrich, 1975).
an additional 3 per cent lift (Schneider and Hermes, 1976). At lower end of the spectrum are the Lepidoptera R hophalocerca with a wing-loading of 0.42 M ~ r n -(Weis-Fogh, ~ 1973, 1975). Since these butterflies with low wing-
loading can glide, they can probably fly with a low metabolic rate, but if they choose to fly at high speed, the rate of energy expenditure would be correspondingly high. 2.3
FLIGHT SPEED
There are no data available on the metabolic rate of freely flying insects as a function of flight speed. However, data on rate of fuel utilization as a function of flight speed on a flight mill point out the large effect of flight speed on
ANN E. KAMMER AND BERND HEINRICH
144
flying speed (rills)
I
I00
1
INSECT FLIGHT METABOLISM
145 (km/h)
2 O0
I(c)
.
,,_hovering
5
10
I
I
15 I
20
2
I
I
:.,
Forward flying speed (m/s) Fig. 4. Effect of flight speed on metabolic rate. (a) Schistocercu greguriu (from Weis-Fogh, 1952). (b) Tubanus aflnis (from Hocking, 1953). (c) Expected aerodynamic power per unit weight, based on the effect of velocity on induced power (to support the insect’s weight), parasite power (to overcome drag of the body) and profile power (to overcome profile drag of the wings) (from Weis-Fogh, 1975).
metabolic rate. In the locust Schistocercu greguria the metabolic rate increased approximately as the second power of the speed (Fig. 4a) (Weis-Fogh, 1952). Similar increases but a wider range of inexpensive speeds were observed in Apis and a variety of Dipreru (Hocking, 1953). Hocking (1953) found for all species examined a basic minimum rate of fuel consumption necessary to maintain the animal in the air. There is very little increase in fuel consumption above this rate at low cruising speeds, but fuel consumption increases steeply at high flight speeds (Fig. 4b). The honeybee, Apis melltfera, for example, increased its metabolic rate 2.5 times from cruising to maximum fight speed. A horsefly, Tubunus uflnis, varied its rate of flight metabolism by a factor of nearly 10 times from cruising to maximum flight speed. At very low forward speeds and during hovering, the metabolic rate will again be high (Fig. 4c). Faster flight is more expensive because air resistance or extra-to-wing drag increases steeply with velocity (Fig. 4c) (Hocking, 1953; Weis-Fogh, 1956a; Gewecke, 1975). Similar effects are observed with propeller-driven airplanes.
146
ANN E. KAMMER AND BERND HEINRICH
In addition, the costs due to wing inertia and thoracic damping become increasingly high at greater velocities. A complete and quantitative analysis of the factors responsible for increased metabolic rate at higher speeds has, however, not been made. 2.4
COST OF TRANSPORT
The metabolic rates of flying insects and of insect muscle are among the highest known (Weis-Fogh, 196 1). Throughout this review we emphasize the high cost of flight per unit time. When the cost per distance travelled is considered, however, flying is seen to be economical as well as fast. The cost of transport can be calculated as P,/wu, where P, = power input or total metabolic rate in watts, w = body weight in newtons, and u = velocity in m/s. Such calculations show that the cost of transport has a minimum value at a certain speed, that the same relationship between cost of transport and body weight holds for both flying insects and birds, and that the cost is much less for a flying animal than for a running animal of the same weight (Tucker, 1970, 1973; Schmidt-Nielsen, 1972). Comparisons of cost of transport can be made only in approximate terms, because published comparisons do not include data on the metabolic rate and speed of walking or running insects. In addition, the actual cost will be increased if the insect, unlike vertebrate homeotherms, expends energy in a pre-flight warm-up. For short-duration flights the additional costs may be substantial. 2.5
ECOLOGY AND EVOLUTION OF HIGH METABOLIC RATES
High rates of metabolic expenditure can presumably only evolve under conditions providing sufficient compensation, such as the ability to escape predators, to mate, and more proximally to accumulate energy profit during foraging. Metabolic rate is thus a pervasive aspect of the overall survival strategy of an animal. Hovering flight is one of the most demanding of metabolic activities. This mode of locomotion can be shown to be instrumental in increasing the rate of intake of food energy. For example, hovering flies, Bombilius spp., visited 2 1.1 flowers of Houstonia caerulea per minute, whereas Syrphus spp. (which were not hovering at the flowers) visited only 5.2 per minute. Hovering moths, Hemaris spp. visited 50.4 Kalmia angustifolium flowers per minute, whereas nonhovering Bombus spp. visited only 15 per minute. Hovering hummingbirds, Archilochus colubris visited 37.1 flowers of Impatiens biji'ora per minute whereas Bombus spp. visited only 8-10 per minute (Heinrich, unpublished
INSECT FLIGHT METABOLISM
147
observations). The caloric cost of hovering for one minute is generally less than the caloric contents of each of the above flowers, provided they have not been depleted by other foragers. Apparently stopping at each flower requires more time than does feeding “on the wing”. In the extremes there are two basic metabolic strategies of harvesting food energy from flowers. In the one, exemplified by many butterflies, energy expenditure during foraging is low, and relatively low-energy food sources can be harvested. The food rewards of a composite inflorescence, for example, are generally individually too small to be economically harvested by hovering. But they can be gathered by a butterfly or a bee that lands on the flowers and reduces its energy expenditure. Low-energy food sources can generally not be harvested by high-energy foragers, but these animals can make much more rapid energy profits than the butterfly or the bee from high-energy food sources. High metabolic rates during flight are in many cases supported by food reserves accumulated in the larval stage. It is perhaps noteworthy that saturniid moths (which do not feed as adults) have evolved large wings. Having relinquished energy intake and having shifted to reliance on fixed energy reserves, they have minimized energy expenditure of flight by decreasing wingloading. Within the sphingid moths there are species with varying reliance on stored vs. acquired energy. It would be of interest to determine the correlation between wing-loading and foraging method within this group.
3
Neural control of power output
To fly faster or to carry additional weight, an insect must increase its output of aerodynamically useful power. In principle this could be accomplished by increasing the wingbeat frequency, increasing the stroke amplitude, or changing other variables of the wingstroke, e.g. angle of attack. Such changes in wing movements would be expected to be accompanied by alteration in the excitation of the power-producing flight muscles. The available evidence shows that the latter expectation is fulfilled, but there are few complete accounts of the wing movements and resulting aerodynamic forces produced by changes in the motor pattern. 3.1
INSECTS WITH NE URO G E NIC RHYTHMS
In the classical study of insect flight, the locust Schistocerca gregaria was flown on a force balance in a wind tunnel (Weis-Fogh and Jensen, 1956; WeisFogh, 1956a; Jensen, 1956). Wing movements were determined from films,
148
ANN E. KAMMER AND BERND HEINRICH
and the angle of attack (angle between a chord of the wing and the relative wind) was calculated for two sections of the forewing and one section of the hindwing. Coefficients of lift and drag were measured by positioning isolated wings in a laminar air flow. The wings were described as moving in a plane inclined to the long axis of the body. Calculations of lift using measured parameters and assuming that steady-state aerodynamics were applicable gave values consistent with the average lift produced by a locust flying in the wind tunnel. In locusts, metabolic rate (estimated from temperature excess) increases linearly with lift produced (Fig. 5a) (Weis-Fogh, 1964a). Changes in wingbeat frequency are small and do not account for changes in this aerodynamic force (Weis-Fogh, 1956a, 1964a). Power output per wingbeat can be increased partly by recruiting additional motor units (although the small number of motor units limits the change) and partly by increasing the excitation of each motor unit (Fig. 5b) (Wilson and Weis-Fogh, 1962). A contraction elicited by paired stimuli to a flight muscle produces 2 to 3 times more work than a twitch elicited by a single stimulus, without greatly increasing the duration of the contraction (Neville and Weis-Fogh, 1963). In addition to altered amounts of excitation of the flight muscles, small changes in the phase relationships among motor units (Mohl and Zarnack, 1975) may contribute to the control of total power output. In locusts changes in muscle activity can result in a small increase in stroke amplitude but changes in forewing twisting appear to be the most important mechanism for controlling lift (Weis-Fogh, 1956a,b, 1964b; Gettrup and Wilson, 1964). A similar explanation may apply to control of thrust. Locusts on a roundabout fly at different speeds without changes in wingbeat frequency (Goldsworthy and Coupland, 1974). Changes in stroke amplitude or wing-twisting could explain the variations in flight speed but these variables were not examined in the latter study. Recent studies have shown that the power-control mechanism employed by a locust depends in part on the experimental conditions. In the case of tethered Locustu migrutoriu stimulated by air currents of certain fixed speeds, wingstroke amplitude and wingbeat frequency were negatively correlated, at least during the first minutes of flight (Gewecke, 1972). In contrast, positive correlations among wingbeat frequency, stroke amplitude, lift and flight speed were observed during longer flights of animals that chose their own air speed (Gewecke, 1975). Control of flight speed depends on input from hair patches on the head, which stimulate flight, and from the antennae, which reduce flight speed (Gewecke, 1974). These observations taken together with the earlier study by Weis-Fogh (1956a) suggest that in locusts several variables are involved in the control of power output, including wingbeat frequency, wingstroke amplitude, and wing-twisting. Although changes in wing-twisting are an important control mechanism,
INSECT FLIGHT METABOLISM
149
100
444 I
100mr
I
Fig. 5. Neural control of power output in Schislocerca gregaria. (a) Metabolic rate increases with lift. P = total metabolic rate = net heat production (P,+ P,) + net aerodynamic power (Pa) (from Weis-Fogh, 1964a). (b) Different patterns of extracellularly recorded muscle potentials correlated with different lift (from Wilson and Weis-Fogh, 1962).
150
ANN E. KAMMER AND BERND HEINRICH
these changes are difficult to quantify without careful measurement. Measurements on the forewing of Locusta migratoria (Zarnack, 1972) show that the basic assumptions according to which previous analyses have been made (Osborne, 1951; Weis-Fogh and Jensen, 1956; Jensen, 1956) are not fulfilled. Therefore, the aerodynamic problems of active insect flight remain as yet unsolved (Zarnack, 1972, 1975). “Simple” approximations have been useful in pointing out the main features of force production under steady-state conditions and in identifying insects that operate with novel mechanisms (Weis-Fogh, 1973, 1975, 1976), but future progress, especially in understanding control mechanisms, may require more exact measurements of wing movements and simultaneous recording of the activity of many flight muscles. There are several complications involved in measuring angle of attack. The aerodynamic angle of attack varies along the length of the wing (Zarnack, 1972, 1975). Rotation of the wing about its longitudinal axis results in an additional velocity component and can give rise to an error of k5O in the angle of attack; both could result in an error in the calculated lift of k 2 0 per cent (Zarnack, 1975). Such considerations may be particularly important in the analysis of control mechanisms. Furthermore, the applicability of coefficients of lift and drag as functions of angle of attack under conditions of a stationary rectilinear air current has been questioned (Bennett, 1973, particularly as regards the upstroke (Zarnack, 1975). In addition changes in wing-twisting may affect not only the angle of attack but also the camber of the wing. The concave profile of the locust forewing during the downstroke is converted into a Z-shape profile during the upstroke (Jensen, 1956). Because the posterior flap substantially increases the lift and drag (Jensen, 1956), it may not be sufficient to treat the locust forewing as a flat plate (Weis-Fogh, 1976, p. 71) as has been done by Zarnack (1972, 1975). However, changes in the shape of the forewing are small relative to the changes in their position, and Zarnack (1972) claims that inclusion of the former does not alter his calculations significantly. With respect to angle of attack, similar analytical difficulties occur in investigating bird flight. The direction of air flow over a flapping wing may be different from that over a gliding wing, the shape of the wing may change during flapping, and force coefficients change markedly with a l o change in angle of attack, which probably cannot be measured with an accuracy of 1 O (Tucker, 1975). In summary, although flight of locusts has been more thoroughly studied than that of other insects, mechanisms by which changes in the motor patterns generate different aerodynamic forces are still poorly known. In other insects that, like locusts, have ordinary muscles and a neurogenic flight rhythm, descriptions of mechanisms controlling the metabolic rate and aerodynamic forces are even more incomplete. Some large Lepidoptera, e.g. saturniid moths, can vary both stroke amplitude and wingbeat frequency, and these changes are correlated with changes in the number of muscle potentials
INSECT FLIGHT METABOLISM
151
per wingbeat in a motor unit (Kammer, 1967); however, the associated effects on metabolic rate have not been determined. In the hawkmoth, Manduca sexta, the stroke amplitude in free flight was 30 per cent larger than in tethered flight, wingbeat frequency was unchanged, and metabolic rate in free flight was double that in fixed flight (Heinrich, 1971a). Electrical recordings from Manduca sexta in fixed flight typically show only an occasional double-firing (Kammer, 1970a, 1971), but with more vigorous flight (slightly higher wingbeat frequency and larger amplitude) the indirect flight muscles are more often
Fig. 6. Changes in stroke amplitude with changes in the pattern of muscle potentials in Manduca sexta. Time exposure, the left wing was illuminated continuously, the right wing was briefly lit
with stroboscopic Rashes (Rash interval indicated in middle trace of each record). Muscles: d , dorsal longitudinal depressor; e, elevator. Time mark on bottom trace, 25 ms.
excited with a pair of impulses (Fig. 6). It is likely that the higher metabolic rate in free flight results from this increase in the neural excitation of the muscles, and that the power output depends in large part on stroke amplitude.
3.2
INSECTS WITH MYOGENIC RHYTHMS
Among the insects with a myogenic type of flight rhythm, wing movements during fixed flight in a wind tunnel have been described for Phormia regina (Nachtigall, 1966) and Drosophila uirilis (Vogel, 1966, 1967). In both species
152
ANN E. KAMMER AND BERND HEINRICH
there are marked differences in pronation-supination between upstroke and downstroke, accomplished by rapid rotation of the wing about its longitudinal axis at the top and bottom of the stroke. In Phormia positive aerodynamic forces are produced under conditions approximating steady-state flow during the middle of the downstroke and the first part of the upstroke, although these
e
(111111111111
300
f 200 j
-
s G
100 el
Time ( 8 )
Fig. 7. Neural control of power putput in a muscoid fly. Air speed was constant, and lift varied with the frequency of muscle potentials (from Nachtigall and Wilson, 1967).
forces are not known in detail (Nachtigall, 1976). Predictions from the kinematic analysis are consistent with measurements of air flow made with a hot-wire anemometer (Wood, 1970). In both species, however, analysis is complicated by nonsteady-state forces that are generated at the top and bottom of the stroke (Weis-Fogh, 1973, 1975; Nachtigall, 1976).
INSECT FLIGHT METABOLISM
153
In various large dipterans, lift (measured with an aerodynamic balance at a constant air speed) is strongly correlated with the frequency of action potentials in the indirect flight muscles (Fig. 7) (Nachtigall and Wilson, 1967). Changes in action potential frequency are also positively correlated with changes in wingbeat frequency (Wilson and Wyman, 1963; Nachtigall and Wilson, 1967; Heide, 1974). Wingbeat frequency is controlled in part by the tonic contraction of small muscles that alter the resonant frequency of the thorax (Nachtigall and Wilson, 1967; Heide, 1971). In addition to changes in wingbeat frequency, control may involve changes in stroke amplitude and in angle of attack (Heide, 1971). CaZZiphora normally has a smaller stroke amplitude at higher air speeds; reduction in amplitude depends on sensory information from Johnston’s organ in the antennae (Gewecke, 1967,1974). In Drosophila lift can be changed by alteration of the body angle (Vogel, 1966), but the magnitude of the output force depends primarily on stroke amplitude (Vogel, 1967). Observations made during fixed flight in an optomotor apparatus also show that visually induced changes in thrust and torque are due to changes in the stroke amplitude (Gotz, 1968). Wing-twisting is a result of the construction of the wing articulation and is not subject to neural control (Vogel, 1967). In most cases studied by Vogel(l967) there was little variation in wingbeat frequency, but in one case lift was proportional to the square of wingbeat frequency. The neural mechanisms controlling power output are not known, but wide variations in the frequency of action potentials in the indirect muscles have been observed (Harcombe, 1975). Another myogenic flier, the honeybee, Apis mellifera appears to employ a variety of control mechanisms, including changes in wingbeat frequency, stroke plane, body angle, and wing incidence (Stellwaag, 1916; Neuhaus and Wohlgemuth, 1960; Wohlgemuth, 1962; Freund (1969) and Herbst (1969) as cited by Nachtigall, 1976; Nachtigall et al., 1971; Esch et al., 1975). Control of stroke plane and wingbeat frequency results from the action of small, direct muscles of the mesothorax and the tonic indirect muscles of the metathorax (Pringle, 1974). Maintained stretch of fibrillar muscles increases the magnitude of tension produced (Abbott, 1973), thus making more power available for flight. In honeybees control of wingbeat frequency is an important mechanism for controlling metabolic rate. Sotavalta (1954b) showed that in tethered flight, fuel consumption is a direct function of number of wingbeats. The average mg. amount of sugar (glucose) consumed per wingstroke was 10.4 x Therefore, 1 mg of sugar supported flight (at the normal frequency of 220/s) for 8-9 min. [Sotavalta’s (1954b) mean value for food consumption per hour is close to those of Beutler (1936) and Hocking (1953).1 Changes in wingbeat frequency are usually accompanied by changes in the frequency of action potentials in the indirect (power-producing) flight muscles. During fixed flight,
154
ANN E. KAMMER AND BERND HEINRICH
wingbeat frequency increased with action potential frequency (Esch and Bastian, 1968). In honeybees flying on a roundabout, there was a positive correlation between the flight speed and the action potential frequency in the dorsoventral flight muscles (Bastian, 1972). In honeybees flying in a wind tunnel, rapid changes in lift and thrust elicited by visual stimuli were associated with a change in action potential frequency in the indirect fibrillar flight muscles and with a change in the number of motor units active, but there was little change in wingbeat frequency (Esch et al., 1975). During longer, steady flights, however, there was an increase in wingbeat frequency as thrust and lift increased; these changes were correlated with higher thoracic temperatures and an action potential frequency increase of about 50 per cent (Esch, 1976). Since the oxygen consumption per action potential is constant (Bastian and Esch, 1970), the higher frequency of action potentials reflects the mechanism by which muscle metabolism changes to provide the increased power output. Similarly in bumblebees, oxygen consumption per action potential is independent of spike frequency (Kammer and Heinrich, 1974), and oxygen consumption increases with load (Heinrich, 1975), suggesting that variations in spike frequency in the fibrillar muscles are part of the control mechanism in bumblebees as in other insects. In myogenic fliers, the wingbeat frequency is dependent on the mechanical properties of the oscillating system and the specialized properties of fibrillar muscle, not on the timed excitation of the indirect muscles by the nervous system. Why then does the spike frequency change with changes in wingbeat frequency? In fibrillar muscle, the duration of the delay between changes in length and tension determines the range of oscillation frequencies into which the muscle can deliver mechanical energy (Pringle, 1974). This delay is shortened by increasing the concentration of calcium ions (Abbott, 1973). Since action potentials cause an increase in intrafibrillar calcium ion concentration, higher spike frequencies will allow the muscles to oscillate efficiently at higher frequencies. Other studies have shown directly that impulse frequency controls the amplitude of oscillation and the mean tension in a graded manner (Machin and Pringle, 1959). In addition, experiments on glycerinated fibers have shown that under given conditions the work output per cycle is maximal at a lower frequency of oscillation when the amplitude of oscillation is larger (Pybus and Tregear, 1973). Thus, to maintain or increase the stroke amplitude while increasing the wingbeat frequency, an increase in action potential frequency is required. 3.3
NOVEL AERODYNAMIC MECHANISMS
Recent studies have added a new mechanism for producing and perhaps controlling aerodynamic forces. Although most insects depend mainly on
INSECT FLIGHT METABOLISM
155
normal aerofoil action, making it possible to analyse the flight in terms of steady-state aerodynamics (Weis-Fogh, 1973), in some insects nonsteady-state effects are essential for production of the necessary lift. These effects are generated by rapid rotation of the wings at the top and bottom of the stroke (Bennett, 1970; Weis-Fogh, 1973, 1975; Lighthill, 1973, 1975; Norberg, 1975), and perhaps throughout the entire stroke (Ellington, 1975). Small variations in the rotation of the wing relative to the body of the insect and rotation of the wing about its axis can have marked affects on the direction of the aerodynamic forces and on their magnitude. Whether or not these novel, nonsteady mechanisms provide another control mechanism for drawing different amounts of aerodynamic power from the flight motor remains to be established. 3.4
SOME COMPARISONS A N D CONCLUSIONS
Both myogenic and neurogenic fliers increase the rate of energy input by increasing the excitation (average action potential frequency) of the powerproducing flight muscles. In order to compare the amount of energy mobilized by changes in spike frequency in different insects we have calculated the amount of oxygen consumed per gram muscle per minute and per action potential (Table 3). These data are approximations, although more confidence can be placed in the two cases in which oxygen consumption and spike frequency were measured simultaneously (Bastian and Esch, 1970; Kammer and Heinrich, 1974). Comparison among insects with neurogenic rhythms shows that two hawkmoths, M . sexta and H . lineata, mobilize more energy per unit time by flying at a higher wingbeat frequency than does S. gregaria. Differences between insects with neurogenic and myogenic rhythms with respect to oxygen consumption per action potential may reflect the specialized properties of fibrillar muscle, which contracts relatively independently of spike frequency, or they may reflect other cellular differences such as proportion of mitochondria or amount of transmitter or calcium released per action potential. The additional energy mobilized by increases in average spike frequency increases the rate of energy output, or more importantly, the rate of doing useful aerodynamic work. The energy input results in increased wingbeat frequency, or increased force per wingbeat as a result of changes in wingstroke amplitude and possibly angle of attack. Changing wingbeat frequency is a useful mechanism in myogenic systems, because the resonant properties of the thorax can be changed by tonic muscles. In locusts and perhaps other neurogenic fliers, changes in wingbeat frequency are less important. In these insects increasing the power output by increasing the wingbeat frequency may be energetically expensive, for two reasons: (1) the wings and thorax form a
156
ANN E. KAMMER AND BERND HEINRICH
resonant system, and forcing the system to oscillate at other than the resonant frequency requires additional energy (Greenewalt, 1960); (2) if the contractions of antagonistic muscles overlap at the higher wingbeat frequencies, work must be done to stretch the muscles, and this work decreases the efficiency of the system (Neville and Weis-Fogh, 1963; Tucker, 1975). Wingbeat frequency is easily measured: less is known about changes in power output per wingstroke. Particularly lacking is information about the rotation of the wing around its long axis and about instantaneous lift and thrust. In addition, contributions of nonsteady effects need to be quantitatively assessed.
4
Supplying the energy demanded: Control of flight metabolism
Initiation of flight involves the rapid activation of neural and metabolic processes. The flight motor is presumably activated by interneurons descending from the brain to the thoracic ganglia. Signals from the flight motor neurons activate the muscles, the wings are brought forward, and flapping flight begins. Initially the muscles utilize fuel within the muscle cells themselves and in the nearby hemolymph. For longer flights fuels in other parts of the body must be mobilized and transported to the muscles. Concommitantly the muscles are continuously supplied with ample amounts of oxygen. Insect flight depends entirely on aerobic metabolism, in contrast to vertebrate locomotion during which the muscles may become partly anaerobic and accumulate an oxygen debt. Flight may continue for several hours (as in locusts and mosquitoes) or for only a few minutes (as in tsetse flies or foraging bees). In the following sections we comment on some properties of flight muscles and then review the mechanisms by which the muscles are supplied with oxygen, energy and fuel. 4.1
INSECT FLIGHT MUSCLES
The high metabolic rates of insect flight muscles can be explained in part by the fact that a large fraction of the muscle cell is occupied by mitochondria (Fig. 8a). Specializations for rapid shortening and lengthening are of two kinds. In insects with a neurogenic flight rhythm, the muscle fibers typically have short sarcomeres, an abundance of sarcoplasmic reticulum, and an ordered array of thick and thin filaments, with an orbit of 6 thin around a thick filament (Fig. 8a; Elder, 1975). More tonic or slowly contracting muscles in the same insect have longer sarcomeres, less sarcoplasmic reticulum, fewer mitochondria, and more thin filaments around each thick filament (Fig. 8b). In insects with a myogenic flight rhythm and fibrillar muscles, the muscles are specialized to
INSECT FLIGHT METABOLISM
157
oscillate rapidly because of a stretch-sensitivity and a delay in tension rise in response to stretch (Armitage et al., 1973; Pringle 1974). In these muscles, the myoflaments are highly ordered, with a 3 : 1 ratio of thin to thick filaments, mitochondria are abundant, and the sarcoplasmic reticulum is reduced (Pringle, 1974; Elder, 1975). These generalities are well known, but some specializations should be mentioned, in part because their functional significance warrants further investigation. We have already discussed differences among fibrillar muscles with respect to metabolic rates (Table 3). The apparently unique enzyme activities of bumblebee muscle will be considered later (section 5). Not all flight muscles exhibit the typical ratio of thick to thin filaments but some have a higher number, typical of slow muscles (Elder, 1975). However, there is no obvious correlation with wingbeat frequency. For example, the typical arrangement of 6 thin filaments surrounding a thick filament is found in the dorsal longitudinal muscle of Munduca sexfa, which has a wingbeat frequency of 30-35/s, and of Anfheraeu polyphemus, which has a wingbeat frequency of 5-8/s (Rheuben, unpub.; Bienz-Isler, 1968a), yet butterflies with an intermediate wingbeat frequency of 8-1O/s have 7-9 thin filaments around a thick filament (Auber, 1967). Another potential source of diversity is the content of paramyosin. Insect muscles that have been examined contain paramyosin in addition to actin and myosin (Bullard et al., 1973), but the function of this protein is not yet known (Levine et al., 1976). Other morphological differences such as the proportion of muscle mass devoted to contractile proteins, mitochondria, sarcoplasmic reticulum or tracheolar supply can be given functional interpretations (Josephson, 1975), but there is as yet little quantitative information on these differences, particularly in relation to mechanical properties and work output. A physiological difference that has not been well studied in relation to flight is the innervation pattern. Some muscle fibers are multiply-innervated rather than supplied by a single “fast” axon (Ikeda and Boettiger, 1965a,b; Kutsch and Usherwood, 1970). Multiple innervation may permit more graded control of muscle work and metabolism. 4.2
OXYGEN SUPPLY
Oxygen is conveyed to the flight muscles via a tracheal system, the fine branches of which penetrate deep into the muscle fibers (Fig. 8a). This configuration reduces the intracellular diffusion pathway, an adaptation important because flight muscle fibers typically are large in diameter and have mitochondria interspersed among the columns of myofibrils (Elder, 1975). An exception to the usual arrangement of tracheoles occurs in the Odonata; no
158
ANN E KAMMER AND BERND HEINRICH
Fig. 8a. Fine structure of a phasic flight muscle, the mesothoracic dorsal longitudinal muscle of the moth Manducu sexra. Cross-sections show that each myofibril is nearly completely surrounded by mitochondria. Tracheoles penetrate the muscle within the lumen of the t-tubule system. x20 OOO. The inset shows at higher magnification a portion of a mitochondrion, a ttubule and associated sarcoplasmic reticulum, and the low ratio (6-8 thin around each thick) of myofilaments. x 100 OOO. (Aldehyde fixation and osmium post-fixation;Tr, tracheole; t, t-tubule; m, mitochondrion.)
INSECT FLIGHT METABOLISM
159
Fig. 8b. Fine structure of a tonic flight muscle, the lower third axillary muscle of M. sexta. Although this muscle has an ample system of t-tubules and sarcoplasmic reticulum, it possesses fewer and smaller mitochondria than the faster flight muscles. Tracheoles were not seen in the ttubules, but rather lay next to the outer membrane of the fiber. x49 OOO. The higher thin to thick ratio of the myofilaments (10-12 thin around each thick) can be seen in the lower portion of the figure. Micrographs by M. B. Rheuben.
160
ANN E. KAMMER AND BERND HEINRICH
tracheolar invaginations have been seen (Smith, 1966). In these insects an oxygen supply adequate for continuous aerobic metabolism is provided by the dense tracheolar plexus that covers the surface of the fibers, which are small in diameter (1Cb20 pm, compared with 10-100 pm in Orthopteru and Lepidoptera or more than 100 pm in fibrillar muscles (Elder, 1975)). The tracheal system is not only efficent in delivering oxygen at high rates, but it is light-weight and thus increases the payload that the insect can carry. In contrast, a significant fraction of the mass of a vertebrate muscle well-supplied with oxygen is composed of capillaries laden with blood. In small insects or in inactive large insects diffusion suffices to supply oxygen at the rate required, but in large, active insects the main tracheae and air sacs are ventilated (Weis-Fogh, 1967; Miller, 1966, 1974; Kammer, 1976). The initiation of flight and also the pre-flight warm-up in large insects is accompanied by activation of ventilatory mechanisms, including changes in the thoracic volume caused by rhythmic contractions of the flight muscles and, in some insects, abdominal pumping movements. Since these abdominal movements begin at the same time as, or in some cases before, the flight muscles are active (Kammer and Heinrich, unpublished observations on Bombus bees), the respiratory control mechanism probably is not a feedback system, dependent on changes in oxygen or CO,, but a system parallel to and perhaps interacting with the neural system controlling flight. [Interaction between the flight control system and the respiratory control system has been observed in the locust Schistocercu greguriu, in which flight motor neurons are modulated at the respiratory frequency (Burrows, 1975), but the functional significance of this modulation is not known.] Although it is generally accepted that the rate at which oxygen reaches the muscles is sufficient to maintain aerobic metabolism without accumulation of an oxygen debt, there are a few observations showing a slightly elevated oxygen consumption after flight. For example, while quiescent after a flight the sphingid moth Metopsilus procellus consumed oxygen at a rate equivalent to 1.5-3 min of flight; however, the rate was only 3 per cent that observed during flight, and similar increases occurred after feeding (Zebe, 1954). Similarly in locusts the oxygen consumption immediately after flight has ceased is slightly higher than at rest (Krogh and Weis-Fogh, 195 1). Possibly this elevated postflight metabolism can be attributed to biochemical activity in tissues other than the muscles, for example, the synthesis of triglyceride or glycogen to replenish the fuel depots. Alternatively, elevated post-flight metabolism could be caused by high thoracic temperatures. Large insects cool slowly (moths, Bartholomew and Epting, 1975a,b) and the resting metabolism is markedly temperaturedependent (locusts, Krogh and Weis-Fogh, 195 1; bumblebees, Kammer and Heinrich, 1974); therefore thoracic temperature and metabolic rate will parallel the cooling curve. Taken together the available data support the conclusion
INSECT FLIGHT METABOLISM
161
that the mechanisms supplying oxygen are sufficient to meet the high metabolic demands during flight, and no oxygen debt accumulates. 4.3
BIOCHEMICAL PROCESSES IN FLIGHT MUSCLE
The 50-to-100-fold increase in metabolic rate required by flight (section 2) presents a severe challenge to the cellular metabolic machinery and to biochemists who would understand metabolic regulation. There must be a rapid and 50-to- 100-fold increase in enzyme activity, substrate flux, oxidative phosphorylation, and ATP production. The general outlines of the metabolic scheme employed by insect muscle are described succinctly by Hochachka and Somero (1973, pp. 68-75). Details and references to the original literature are provided by Sacktor (1970, 1975) and Crabtree and Newsholme (1975). We shall here reiterate only some of the main features in order to provide a biochemical perspective to the high metabolic rates of flight. 4.3.1
Intramuscular carbohydrate
Initially the flight muscles are dependent on their own metabolic reserves. Muscle contains only a small amount of ATP, e.g., 7pmoleslg wet weight of thorax in Phormia regina (Sacktor and Hurlbut, 1966), enough for only about 0.1 s of flight (Crabtree and Newsholme, 1975). To maintain the supply of ATP, glycogen and other carbohydrates in the muscle fibers are metabolized immediately. Both contraction and metabolism are activated by the nervous system (Newsholme and Start, 1973; Sacktor, 1975). The motor neuron triggers an impulse in the muscle cell, depolarizing the muscle cell membrane and causing the release of Ca++ from the sarcoplasmic reticulum in the usual fashion (Maruyama, 1974). In fly muscle (Sarcophaga bullata) the current for the muscle action potential is carried primarily by the inward movement of Ca++ (Patlack, 1976), thus providing in a few milliseconds an additional increase in the cytoplasmic calcium concentration. The calcium ions activate the myofibrillar ATPase and initiate contraction. Calcium ions also initiate a series of reactions setting in motion the breakdown of glycogen (Fig. 9), and Ca++ promotes the activity of another enzyme involved in the processing of carbohydrates, glycerol-3-phosphate dehydrogenase (GPDH). In addition, the products of ATP breakdown, formed when the muscle contracts, activate phosphofructokinase (PFK), an enzyme important in controlling the flux of substrate through the glycolytic pathway. (Additional but unknown control mechanisms are apparently required to account fully for the 100-fold increase in flux through the glycolytic pathway (Newsholme and Start, 1973; Sacktor, 1975). Sugar metabolism proceeds in the usual fashion; the fructose
ANN E. KAMMER AND BERND HEINRICH
162
Glycogen
Trehalose
Glucose I- phosphate
Glucose
I
I
b-ATP
\ Glucose 6-phosphate I
Fructose 6-phosphate
IFDpl Fructose I, 6 - dphosphate
G Dihydroxyacetate phosphate
1W
+
Glyceraldehyde 3-phosphate
NADH I
I
I
I I
I I b i
t
+ I
Glycerol 3-phosphate Z H d
1
.1
i
I
Dlhydmyacetone phosphate
J-
Phosphoenolpyruvate ATP
Pyruvate
-
Fig. 9. Outline of carbohydrate metabolism. The glycerol 3-phosphate (GP) shuttle of insect flight muscle is included. Abbreviations: are: ALD, aldolase; FDP, fructose 1,6-diphosphatase; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; GDH, NAD-glycerol 3-phosphate dehydrogenase; GPDH, mitochondrial glycerol 3-phosphate dehydrogenase: HK, hexokinase; LDH, lactate dehydrogenase; Pase, phosphorylase; PDH, pyruvate dehydrogenase; PFK, phosphofructokinase;PGI, phosphoglucose isomerase; PK, pyruvate kmase; TH, trehalase.
INSECT FLIGHT METABOLISM
163
diphosphate formed by the phosphofructokinase reaction is cleaved into two triose phosphates, glyceraldehyde-3-phosphateand dihydroxyacetone phosphate, which are then converted into pyruvate and ultimately into CO, and H,O via the mitochondrial enzymes of the Krebs tricarboxylic acid cycle. The net yield of ATP is 38 mol per mol of glucose oxidized completely (Crabtree and Newsholme, 1975). Continued operation of the glycolytic pathway requires a continuous supply of NAD+, which is reduced during the oxidation of glyceraldehyde-3phosphate. In vertebrate muscle under anaerobic conditions, NAD+ is regenerated by reducing pyruvate to lactate. In insect muscle the necessary enzyme, lactate dehydrogenase, is absent or inactive, and little lactate is formed. NAD+ cannot be regenerated by the mitochondrial enzymes because the mitochondrial membrane is impermeable to pyridine nucleotides. In active insect muscle, NAD+ is continuously regenerated in the cytosol by reducing dihydroxyacetone phosphate to glycerol-3-phosphate, which penetrates the mitochondrial membrane and is re-oxidized and then returned to the cytoplasm (Fig. 9). In other words, glycerol-3-phosphate acts as a shuttle carrying electrons and protons into the mitochondria, where their transfer to 0, is coupled to oxidative phosphorylation. This glycerol-3-phosphate shuttle ensures that glycolysis continues aerobically, with 2 moles of pyruvate formed from each mole of glucose and available for further oxidation by the Krebs cycle. The role of the a-glycerophosphate cycle in supporting flight is illustrated by the plight of Drosophila mutants deficient in glycerol-3-phosphate dehydrogenase (GPDH) (O’Brien and MacIntyre, 1972). These mutants could not sustain flight and after a few aborted flights refused to take off. Interestingly, mutants with only 10 per cent of the enzyme activity of wild-type flies flew normally. Furthermore, after 25 generations some mutants recovered the ability to fly, although they had no measurable GPDH activity (O’Brien and Shimada, 1974). These “adapted” mutants, unlike the nonflying mutants, could oxidize NADH via a malate-dependent pathway at rates comparable to wildtype flies (O’Brien and Shimada, 1974). The pathway supplying energy for these “adapted” mutants is not known. The activity of one of the enzymes involved in the glycerol-3-phosphate shuttle, GPDH, is stimulated by increasing concentrations of Ca++ions. As a result, increased activity of the shuttle follows the muscle action potential and the release of Ca++ from the sarcoplasmic reticulum. As mentioned above, glycogenolysis is also stimulated by Ca++. Activation of both the contractile machinery and the metabolic flux by the same signal, Ca++released as a result of neural excitation, seems to be a satisfactory mechanism for simultaneous control of two parallel but interdependent processes. In addition to the evidence from insect muscle summarized here, studies on frog muscle show that
ANN
164
E. KAMMER AND BERND HEINRICH
depolarization and the release of intracellular calcium stimulates respiration as well as producing contractions (Van der Kloot, 1967). In the case of insect muscle, however, there is a puzzling gap in our understanding of control mechanisms. Trehalose is another readily available source of carbohydrate for oxidation, since it is present in the muscle and the hemolymph of most insects, but the mechanism by which trehalase activity is controlled is not understood. Trehalase in the hemolymph of Phormia is inhibited by a protein and by 0.01 M Mg++ (Friedman, 196 l), but it is not known if the same factors affect muscle trehalase. Trehalase activity in locust flight muscle increases during flight or during electrical stimulation of an isolated nerve-muscle preparation; therefore the control mechanism is neural and not hormonal (Candy, 1974). However, control of trehalase apparently does not depend on changes in Ca++ (studies on Sarcophaga and Bombus (Vaughan et al., 1973) and on Phormia (Reed and Sacktor, cited in Sacktor, 1975)). 4.3.2
Fuels forfright
The preceding discussion has considered the role of intramuscular carbohydrate as a fuel for the initial minutes of flight. The description was derived primarily from studies on blowfly (Phormia regina) flight muscles, which initially contain large quantities of glycogen (Sacktor and Wormer-Shavit, 1966) and which continue to metabolize carbohydrates during long flights (Clegg and Evans, 1961; Sacktor, 1965). Similarly the flight muscles of the cockroach Periplaneta americana are rich in glycogen, which contributes to the first 10min of flight (Downer and Matthews, 1976a). Locusta migratoria muscles contain large quantities of glucose and trehalose (Bucher and Klingenberg, 1958) and some glycogen; these carbohydrates plus trehalose in the blood and other stores are used during the first 20-30 min of flight, after which lipids are used in continuous flights lasting several hours (Weis-Fogh, 1952; Mayer and Candy, 1969a). Male Douglas-fir beetles, Dendroctonus pseudotsugae also use carbohydrate during early flight and lipid for prolonged flight (Thompson and Bennett, 1971). The moths Prodenia eridania and Spodoptera frugiperda can metabolize both carbohydrates and lipid (Stevenson, 1968a,b; Van Handel and Nayar, 1972b; Van Handel, 1974) and perhaps protein (Van Handel and Nayar, 1972a), but the time course of the utilization is not known. Other Lepidoptera depend primarily on lipids during flight (Zebe, 1954; Domroese and Gilbert, 1964; Gilbert, 1967), as do aphids (Cockbain, 1961). Among the Lepidoptera dependence on lipids appears to be characteristic of those which do not feed as adults, whereas those that do feed have a higher glycolytic capacity and can utilize both carbohydrates and lipids (Beenakkers, 1969); however, the number of species that have been examined is small, and
INSECT FLIGHT METABOLISM
165
generalization may be premature. Lipid also provides a substantial reserve for the migratory flight of the monarch butterfly, Danaus pleXppus (Brown and Chippendale, 1974). Similarly, individuals of flying swarms of the tettigoniid Homorocaryphus nitidulus contains a quantity of fat that is depleted during flight, whereas nonflying individuals of a related species, H. subvitlatus, have a low fat content (Kerukize, 1972). Flight forms of a weevil Callosobruchus maculatus have twice the total body lipid as did normal weevils (Nwanze et al., 1976). It is generally assumed that insects such as bees and flies that feed on nectar use the dietary carbohydrate for flight fuel. Recently Clark (1976) demonstrated that bumblebees can fly at high ambient temperatures although their carbohydrate metabolism had been partially blocked by deoxy-glucose, and he suggested that these insects can metabolize lipids during flight. This interpretation is surprising in light of the low level of lipase activity in Bombus muscles (one third that of Locusta) (Crabtree and Newsholme, 1972b). Since the deoxyglucose inhibition was about 85 per cent and not 100 per cent complete, and since the metabolic rate during flight can vary (Heinrich, 1975), perhaps enough sugar trickled through the usual metabolic pathway to support minimal flight. In a few insects proline is an important energy source, e.g. in the tsetse fly, Glossina morsitans (Bursell, 1963, 1966; Hargrove, 1976) and the beetle Leptinotarsa decemlineata (deKort et al., 1973). Proline may also be used occasionally as an energy source by Schistocerca gregaria (Kirsten et al., 1963; Mayer and Candy, 1969a), Phormia regina (Sacktor and WormserShavit, 1966) and Apis rnellifera (Barker and Lehner, 1972). In the latter insect, however, the rate of consumption is sufficient to supply only about 0.1 per cent of the energy during flight (Barker and Lehner, 1972). 4.3.3
Proline metabolism
For most insects the role of proline metabolism in flight is not known (Crabtree and Newsholme, 1975). Bursell (1975) has suggested that in Diptera the ability to metabolize proliie rapidly enough to support flight may be an adaptation to feeding on mammalian blood. Tsetse flies are obligatory bloodsuckers, and their mitochondria oxidize proline almost 100 times as fast as pyruvate. The mitochondria of facultative bloodsuckers such as Stomoxys oxidize proline and pyruvate at approximately equal rates. The mitochondria of species which do not feed on blood, or of which only the female is bloodsucking (presumably to obtain amino acids for oogenesis), oxidize proline only at low rates (Bursell, 1975). Most insects of other orders fall into the latter category, but as Bursell (1975) points out, there are puzzling exceptions to this scheme. The potato beetle Leptinotarsa decemlineata oxidizes proline at a high rate (deKort et al., 1973) and uses proline as a flight fuel, and another beetle Melolontha
ANN E. KAMMER AND BERND HEINRICH
I66
melolontha has an active proline dehydrogenase (Crabtree and Newsholme, 1970). The adaptive significance of proline metabolism in these two beetles is not known, nor is it clear why the bloodsucking flies specialize on proline, since mammalian blood contains other amino acids.
a'- Pyrrolne - 5 - carboxylate
Glutamate
Alanine
Pyruvate
acetyl Co A
\
o (
- ketoglutarde I I I
TCA cycle
Fig. 10. Outline of proline metabolism. Abbreviations: CS, citrate synthase; FP, flavoprotein; GPT, glutamate-pyruvate transaminase; ODC, oxaloacetate decarboxylase; PCAD, pyrroline carboxylic acid dehydrogenase; PrDH, proline dehydrogenase.
The metabolism of proline (Fig. 10) produces glutamate that, after transamination with pyruvate, enters the Krebs cycle as a-ketoglutarate. If the muscle has an active oxaloacetate decarboxylase, pyruvate is regenerated, the sequence continues, and alanine accumulates. The yield of this pathway is 14 mol of ATP per mol of proline oxidized to alanine (Crabtree and
167
INSECT FLIGHT METABOLISM
Newsholme, 1975). Tsetse flies feed on blood and make rapid use of the abundant amino acids it contains (McCabe and Bursell, 1975). These flies use proline as the main energy source for flight (Bursell, 1963, 1966). Injected labelled glutamate is converted into alanine (Bursell, 1967), and the alanine concentration increases during flight (Fig. 11) (Hargrove, 1976), as would be predicted from the biochemical pathways described above. However, the total amino N in proline, glutamate and alanine decreases during flight (Bursell, 1963; Hargrove, 1976); perhaps some of the a-ketoglutarate formed in the oxidation of proline is not returned as pyruvate but is oxidized completely to CO, and H,O to give extra energy particularly at the beginning of flight, when wingbeat frequency is high (Hargrove, 1975b, 1976).
t
31
i 1
Y
::
0 2E\
C E
0
I
a
0
\
00-0
h.*p%..ejQ.C.-.---.----J2 I
0
2
4
I
0 0 0 -
,
I
l
l
6
8
1012
1
14 0 2 Flight duration(min)
4
I
I
I
6
8
1 0 1 2 1 4
I
I
Fig. 1 1 . Changes in amino acid concentrations during flight in Glossina pallidipes. Curves 1, proline; 2, glutamate; 3, alanine; 4, total amino nitrogen in proline, glutamate, and alanine (from Hargrove, 1976).
Another proposed function of proline is to increase the activity of the Krebs cycle (Sacktor and Childress, 1967). The a-ketoglutarate that enters the Krebs cycle can increase the supply of oxaloacetate available for combination with acetate, thus increasing the activity of citrate synthase, the first enzyme of the Krebs cycle. This increase in Krebs cycle activity could minimize a build-up of pyruvate produced by glycolysis. Provision of a-ketoglutarate to increase the flux of glycolytically derived pyruvate through the Krebs cycle would depend on a supply of glutamate, which can be obtained by oxidation of proline. Support for this role of proline metabolism is provided by the recent observation that supplying proline and pyruvate to Phormiu mitochondria increased the total content of Krebs-cycle intermediates (Johnson and Hansford, 1975). Furthermore, isolated mitochondria from a beetle, Popilliu juponicu, oxidize proline but produce more NH, than alanine, and the activity of glutamate dehydrogenase is high and increased by ADP (Hansford and
ANN E. KAMMER AND BERND HEINRICH
168
Johnson, 1975). These findings suggest that in this beetle proline metabolism proceeds by a pathway different from that demonstrated in tsetse flies. In tsetse flies proline is probably more important as an energy source, but it may also serve as a source of Krebs-cycle intermediates (Hargrove, 1976). In blowflies, which metabolize primarily carbohydrates, the latter role may be more important.
Fatti acid
j~:: 2L-
lFnsl
+ PPi
Fatty acyl- CoA CwMine
p J
Fuf fy my{- cornlfine
cytoso/
! 9
Fatty acyl- carnitine
mkcoA
Carnitino
Fatty acyl- CoA
I
d
- axidotim
Acetyl- CoA
I
TCA cycle Fig. 12. Pathway for metabolism of diglycerides. Enzyme abbreviations are; CPT, carnitine palmitoyltransferase: FAS, fatty acyl-CoA synthetase; DGL,diglyceride lipase. In most muscles that have been examined, carnitine is necessary for the transport of fatty acid into the mitochondria.
INSECT FLIGHT METABOLISM
4.3.4
169
Lipid metabolism
The energy yield from lipids is more than twice the energy yield of an equivalent mass of carbohydrates, and lipids are the usual storage form of fuel, especially for insects that migrate long distances (Weis-Fogh, 1967; Downer and Matthews, 1976b). Lipid metabolism by flight muscle follows the typical metabolic pathway (Fig. 12) (Sacktor, 1975). In all tissues, lipid metabolism is aerobic; there is no known mechanism for partial processing analogous to the formation of lactate (Crabtree and Newsholme, 1975). In both vertebrates and insects, transport of fatty acids into the mitochondria is typically accomplished by combination of the fatty acid with carnitine. [Some moths provide an exception to this rule, however. For example, mitochondria from Prodenia eridania oxidize palmitate at a very high rate in the absence of carnitine (Stevenson, 1968b).l Within the mitochondria the fatty acids are oxidized, entering the Krebs cycle as acetyl-CoA. The yield per mol of long chain fatty acid, e.g. palmitate, completely oxidized is 129 mol of ATP (Crabtree and Newsholme, 1975). 4.4
MOBILIZATION OF S T O R E D FUELS
In order to maintain a high metabolic rate in flight, it is necessary to supply the muscles with ample fuel. The energy stores in the muscles are usually exhausted after several minutes of flight, and then fuel must be mobilized from the fat body, the gut, or both (Gilbert, 1967; Wyatt, 1967; Bailey, 1975). The main blood sugar is a disaccharide, trehalose. In those insects utilizing carbohydrate as a flight fuel, trehalose is maintained at a fairly high concentration in the blood, 0.5-5 per cent, a higher concentration of blood sugar than in vertebrates (Wyatt, 1967). The concentration of glucose and other monosaccharides in the blood is generally low because the fat body rapidly converts glucose into trehalose and other metabolites. The high glucose gradient across the gut wall facilitates the rapid diffusion of glucose from the gut into the hemolymph. Some insects, e.g. Schistocerca gregaria and Periplaneta americana, apparently lack mechanisms for the active transport of monosaccharides from the gut (House, 1974). Whether insects that ingest nectar and other sources of sugar also depend only on diffusion remains to be established; in Phormia regina, for example, the concentration of glucose in the hemolymph may equal or exceed the concentration of trehalose (Evans and Dethier, 1957). In many animals glucose uptake is coupled with Na+ transport, and such a mechanism, resulting in either facilitated diffusion or active transport, would be expected to occur in insect gut. During long flights of insects that depend on carbohydrates for fuel, carbohydrate stores are mobilized and transported to the muscles (Bailey,
170
ANN E. KAMMER AND BERND HEINRICH
1975). The common storage form is glycogen, which when mobilized from the fat body is transported as trehalose. Little quantitative information is available about the relative importance of glycogen and trehalose as energy reserves. For example, in larvae of Manduca sexta the glucose equivalents of trehalose are 8 times those of glycogen (Tager et al., 1976); the concentrations in adults are unknown. Another reserve supply of carbohydrates is provided by sugar in the gut, particularly in well-fed Diptera. Hymenoptera and some Lepidoptera. In insects that also metabolize lipids, e.g. the moth Spodoptera frugiperda, dietary sugar can be utilized directly, rather than being first converted into lipid (Van Handel and Nayar, 1972b). Lipids are typically stored as triglycerides in the fat body. In most species that have been examined, lipids are transported as lipoprotein, i.e. diglycerides bound to a carrier protein, but they can also be transported as triglycerides or fatty acids (Bailey, 1975; Downer and Matthews, 1976b). The lipid content of the hemolymph is usually much lower than that of the fat body; nevertheless blood lipid may be a significant reserve for flight. The concentration of lipid in the hemolymph of insects that metabolize fats for flight is typically high and increases during flight. For example, diglyceride in the blood of locusts Schistocerca gregaria is 9 mglg hemolymph before flight and increases threefold during flight (Mayer and Candy, 1969a). This hemolymph diglyceride is utilized during flight (Spencer and Candy, 1974). These high concentrations of flight fuels in the hemolymph may be necessary to provide a constant supply of metabolites to the muscles (Weis-Fogh, 1964b). Mobilization of fatty acids may depend on the use of glycerol as a carrier between the fat body and the flight muscles. In the locust Schistocerca gregaria the concentration of glycerol in the hemolymph increased 10-fold during a 1 h flight (Candy et al., 1976). [14Cl-glycerolinjected into flown locusts appeared in the blood as trehalose (27 per cent) and diglyceride (36 per cent) (the remainder was not identified). Candy et al. (1976) suggest that glycerol is continuously moved as diglyceride from the fat body to the muscle, then released from the muscle as free glycerol, and then returned to the fat body where it is re-esterified to more diglyceride. This interpretation is supported by the fact that locust muscle has active lipases (Crabtree and Newsholme, 1972b) and that isolated fat body incubated with substrate and adipokmetic hormone rapidly converts [14C1-glycerolinto diacyl glycerol (59 per cent of the total radioactivity) (the remainder was triglyceride, 11 per cent, trehalose, 8 per cent, and unidentified 22 per cent) (Candy et al., 1976). This glycerol shuttle increases the amount of diglyceride available (1 glycerol + 2 triglycerides = 3 diglycerides)to maintain the blood concentration high during long flights. Insects that depend primarily on proline for energy during flight have analogous shuttle mechanisms for providing the thoracic muscles with substrate from reserves stored elsewhere in the body. Tsetse flies (Glossina) are the best studied insect in this regard (Hargrove, 1976). In teneral flies much of
INSECT FLIGHT METABOLISM
171
the fuel, proline, in the hemolymph is consumed during the first two minutes of flight. But the flies can fly for several minutes longer, as the flight muscles continue to be supplied with proline by a circuitous sequence of reactions. Alanine formed as a result of proline metabolism by the flight muscles is transported to the abdomen where it is converted back to proline. The additional carbon required for proline synthesis is obtained from fat body lipids. The proline is then returned to the thoracic flight muscles. The process is rapid: [*4Cl-alanine injected into the thorax appeared in the abdomen in 2 s (Hargrove, 1976). Both thoracic and abdominal tissues can convert alanine into proline but the abdominal mechanisms are faster. In fed flies the residual blood meal provides an additional source of proline and possibly other compounds (McCabe and Bursell, 1975; Hargrove, 1976), and fed flies fly about twice as long as teneral flies (Hargrove, 1975b). 4.5
METHODS FOR STUDYING FLIGHT MUSCLE METABOLISM
The preceding summary of flight metabolism is based on data obtained by a variety of procedures: (1) measurement of the disappearance of a compound after a long flight, (2) determination of the fate of radioactively labeled compounds injected or ingested normally, (3) measurement of changes in the concentration of a metabolite in the hemolymph or other tissues, (4) measurement of the substrate or oxygen consumed by a stimulated half-thorax, by isolated muscles, by muscle homogenates, or by isolated mitochondria, or (5) measurements of enzyme activity. (For a review of these approaches, see Crabtree and Newsholme, 1975.) Information about the activity of key enzymes in various metabolic pathways provides good first approximations of the types of fuels used in flight by different insects, as well as clues to specializations. For example, insects that have enzyme activities suggestive of both carbohydrate and fatty acid metabolism are insects that both feed and migrate and so metabolize both classes of compounds (Beenakkers, 1969a). Insects that depend almost entirely on carbohydrates have high hexokinase activity and little carnitine palmitoyltransferase activity (Crabtree and Newsholme, 1975). The later authors have provided several examples of the correspondence between rates of fuel utilization calculated from oxygen consumption during flight and rates determined by enzyme activity. Data from Bombus bumblebees provide another example. The oxygen consumption of bees in free flight is 170330 ml 02/g thoraxh, depending on the load of sugar syrup in the abdomen (Heinrich, 1975). On the basis that approximately 70 per cent of the thorax by weight is muscle (determined by dissolving the muscle in KOH and weighing the exoskeleton; Kammer, unpubl.), and 1 ml 0, = 7.6 p o l C, compound, we calculate that the maximum rate in flight is equivalent to 60pmol Cdg
172
ANN E. KAMMER AND BERND HEINRICH
muscle/min. The phosphofructokinase activity in Bombus flight muscle is 33pmoUg fresh weighthin at 25O C (Crabtree and Newsholme, 1972a). Assuming a Q,,,of 2 and a thoracic temperature of 35O C during flight, the predicted enzyme activity during flight is 66pmol CJg muscle/min, a value sufficient to account for the maximum metabolic rate observed during flight. If an enzyme activity is low, then probably the insect is not rapidly metabolizing the compound normally processed by that enzyme. For example, locust muscle oxidizes glycerol slowly (Candy, 1970), and the activity of glycerol kinase, the first enzyme on the pathway of glycerol metabolism, is low relative to the rate at which diglycerides are oxidized during flight (Candy et al., 1976). Locusts do not in fact oxidize the glycerol at a high rate but (as discussed above) use most of it to transport fatty acids from the fat body to the muscle (Candy et al., 1976). Some complications that can arise in using enzyme activities to assess in vivo muscle metabolism should be mentioned. Firstly, measurements of enzyme activity are usually made at 25O C (Crabtree and Newsholme, 1975), whereas the working temperature of the muscle may be 35-40° C. Since different enzymes may have different temperature coefficients (Hochachka and Somero, 1973), assuming a Q,, equal to 2 and extrapolating activity at 35O C from measurements made at 25O C may give an incorrect impression of the relationship between enzyme activity and oxygen consumption during flight. For example, in the locust Schistocerca gregaria, the hexokinase activity of flight muscle was determined to be 11.5 pmoUmin per g fresh muscle at 25O C (Crabtree and Newsholme, 1972a). This figure compares reasonably well with a calculated consumption during flight of 14 pmol of C, carbohydrate unit/min per g fresh muscle at 25O C (Crabtree and Newsholme, 1972a). The latter quantity was calculated from the following values: oxygen consumption during flight = 40 ml of OJh per g of insect (Weis-Fogh, 1952); weight of flight muscles = 18 per cent of the body weight (Zebe et al., 1959); thoracic temperature during flight assumed to be 35O C, and Q,,= 2. However, if the Q,,is only 1.4 as suggested by measurements of glucose consumption in halfthorax preparations (Candy, 1970) and by the effects of temperature on muscle twitch (Neville and Weis-Fogh, 1963), then the measured oxygen consumption is equivalent to 20 pmoles C, unit/g muscle/min at 25O C, a figure substantially greater than the measured rate of hexokinase activity (1 1.5 pmollrninlg muscle; Crabtree and Newsholme, 1972a). More data on temperature effects on enzyme activity in vivo and in vitro are needed, however, before the significance of this difference can be assessed. Secondly, measurements of enzyme activity under optimal conditions provide upper limits for the rate of metabolism of a substrate. In vivo the muscle may work within these limits at lower rates determined by substrate availability and by various controls including hormones and ions. For example, with corrections so that
INSECT FLIGHT METABOLISM
173
rates of ATP production are equivalent, the flight muscles of Locustu migrutoriu have approximately equal activities of hexokinase and carnithe palmitoyl-transferase (Crabtree and Newsholme, 1975). This result suggests that locusts can use either carbohydrates or lipids or both as fuels for flight. As studies on flying locusts have shown, both classes of compounds are in fact used, but not concurrently nor equally; carbohydrate is the major fuel for short flights and lipid the major fuel for flights lasting several hours. Lastly, as Crabtree and Newsholme (1975) point out, factors such as diet, age, sex, and season may influence maximum activity of enzymes so that information obtained at one time may not reflect fuel utilization at other times in the life of an insect. 4.6
HORMONAL CONTROL MECHANISMS
Mobilization of fuel reserves from storage sites distant from the flight muscles is controlled, at least in some cases, by hormones released from the corpora cardiaca. Typically the corpus cardiacum is comprised of several kinds of intrinsic secretory cells as well as the enlarged secretory endings of extrinsic cells, neurosecretory neurons that have their somata in the brain (Goldsworthy and Mordue, 1974). Nerves running to the corpora cardiaca from the brain include axons of these neurosecretory cells and possibly axons of ordinary neurons that may synapse on the neurosecretory endings or on the intrinsic cells. The corpus cardiacum releases several different hormones controlling development, behavior, and hemolymph composition (Goldsworthy and Mordue, 1974; Truman and Riddiford, 1974). 4.6.1
Control of hemolymph sugar
The concentration of blood sugar is influenced by at least two hormones; one hyperglycaemic, the other hypoglycaemic. Extracts of corpora cardiaca injected into cockroaches cause a breakdown of fat body glycogen and an increase in blood trehalose (Steele, 1961, 1963; Wiens and Gilbert, 1967). Extracts of corpora cardiaca-corpora d a t a in Peripluneta contain a glucagonlike factor, as indicated by immunoreactivity (Tager et al., 1976). Locust corpora cardiaca contain two distinct hyperglycaemic peptides, one in the storage lobe of the extrinsic cells (Highnam and Goldsworthy, 1972) and a second in the glandular lobe, which has greater hyperglycaemic activity (Mordue and Goldsworthy, 1969). One of these could be the hyperglycaemichormone releasing-factor, an extrinsic cell product, and the other could be the hyperglycaemic hormone, an intrinsic cell product that would be expected to have greater activity (Kramer, pers. comm.). The corpora cardiaca of the blowfly Phormiu regina also have hyperglycaemic activity (Friedman, 1967).
174
ANN E KAMMER AND BERND HEINRICH
In the blowfly Calliphora erythrocephala a hyperglycaemic hormone is produced by the intrinsic neurosecretory cells of the corpus cardiacum (Normann and Duve, 1969), and a hypoglycaemic hormone is produced by the medial neurosecretory brain cells, the axons of which end in a neurohaemal organ near the corpora cardiaca (Normann, 1975). In the moth Manduca sexta the corpora cardiaca-corpora allata contain a hypoglycaemic factor and a glycogenolytic factor that are functionally, immunologically and physically similar to vertebrate insulin and glucagon respectively (Tager et al., 1975, 1976). The roles of these hormones in regulating developmental processes and longterm carbohydrate balance vs. short-term, homeostatic control of blood sugar levels during flight are not yet understood. For example, in Locusta migratoria injection of an extract of the glandular lobes of the corpora cardiaca into animals from which the glandular lobes had been removed did not elicit an increase in the concentration of trehalose in the blood even when blood carbohydrate levels were low (Jutsam and Goldsworthy, 1975). During flight in this locust, the concentration of total carbohydrate in the blood decreased markedly during the first 10 min and after 30 min of flight stabilized at half the resting level. Fat body glycogen was gradually depleted but blood carbohydrate was still low 4.5 h after a 30 min flight (Jutsam and Goldsworthy, 1976). In Schistocerca gregaria the concentration of blood trehalose fell almost to zero during an extended flight (Mayer and Candy, 1969a). The fact that flying locusts do not regulate blood sugar suggests that the hyperglycaemic hormones in the corpora cardiaca (Goldsworthy, 1969) play no direct role during flight in these insects (Goldsworthy, 1976). Since in locusts the major fuel for long flights is lipid, regulation of blood trehalose may not be necessary. Alternatively, the hyperglycaemic hormone may mobilize sugar, but inadequate supplies of glycogen or food may preclude regulation at resting levels. Mobilization may be important even though the rate of trehalose utilization is higher than the rate of glycogenolysis and gluconeogenesis. If the glycogen supply is small relative to the blood trehalose and to the sensitivity of the assay, glycogenolysis may occur in response to the hormone without a detectable increase in hemolymph trehalose. Such is the case in Manduca larvae injected with a glucagon-like factor (Tager et al., 1976). In addition, by analogy with vertebrates, the insulin-like and glucagon-like factors may play a role in lipid metabolism. For example, application of an extract of corpora cardiaca to fat body of a cockroach Leucophaea maderae accelerated both glycogen degradation and lipid mobilization (Wiens and Gilbert, 1967); however, the number of active factors in the extract was not known. Furthermore, in some experiments the mobilization of carbohydrates matches its utilization but the rate of utilization is low (Jutsam and Goldsworthy, 1976; Robinson and Goldsworthy, 1976). The latter authors suggest that a minimum rate of
175
INSECT FLIGHT METABOLISM
trehalose oxidation may be necessary for efficient lipid oxidation. These experiments employ aqueous extracts of corpora cardiaca and do not distinguish between glycaemic hormones and adipokinetic hormone. A role for the corpora cardiaca hormones in the regulation of hemolymph sugar during flight has been demonstrated in the blowfly Calliphora erythrocephala. Intact corpora cardiaca are necessary for the maintenance of blood trehalose during 45-min flights (Fig. 13) and are probably the source of the hyperglycaemic hormone. If the corpora cardiaca are removed or denervated, the trehalose concentration drops during the first 15 min of flight and within 45 min flies are unable to fly although their glycogen reserves are
1
.\
-\ I 0
I 5
I 10
I I I 25 I5 20 T i m e , rnin
I
I
I
30
35
40
I 45
Fig. 13. Role of hyperglycaemic hormone in maintenance of hemolymph trehalose during flight in blowflies. Open circles indicate intact flies and solid circles indicate flies from which the corpora cardiaca have been removed (from Vejbjerg and Normann, 1974).
not depleted (Vejbjerg and Normann, 1974). Differences between blowflies and locusts with respect to control of blood sugar during flight may be related to the fact that blowflies metabolize primarily carbohydrates whereas locusts are adapted to use lipids for long migratory flights and carbohydrates primarily for “trivial flights” or walking (Goldsworthy, 1976). 4.6.2
Control of hemolymph lipid
Lipid concentrations in the hemolymph and fat body are influenced by one or more hormones from the corpus cardiacum. Extracts of the corpus cardiacum produce elevated blood lipids in Locusta migratoria, Schistocerca gregaria, and Tenebrio molitor (Beenakkers, 1969a; Mayer and Candy, 1969b;
176
ANN
E. KAMMER AND BERND HEINRICH
Goldsworthy et al., 1972b). Such an extract injected into Periplaneta americana causes a decrease in the level of blood glycerides (Downer and Steele, 1972). Recently a lipid-mobilizing factor was isolated from the glandular lobes of locust corpora cardiaca and chemically characterized; the hormone was found to be a blocked nonapeptide (Stone et al., 1976). The adipokinetic hormone of the corpus cardiacum is clearly important in the flight of locusts, stimulating the release of diglyceride from the fat body (Beenakkers, 1969; Mayer and Candy, 1969b; Goldsworthy et al., 1972a). Conversion from utilizing carbohydrate to using lipids as flight fuel requires that lipid be mobilized from storage, since resting locusts have low levels of haemolymph lipid (Beenakkers, 1965). Diglycerides are released from triglycerides stored in the fat body and circulated in a lipoprotein complex (Tietz, 1962; Beenakkers, 1969; Mayer and Candy, 1969b; Peled and Tietz, 1973; Jutsum and Goldsworthy, 1976). During flight there is a rapid increase in haemolymph lipid, and an elevated concentration may persist after the cessation of short flights (Jutsum and Goldsworthy, 1976). Locusts without the glandular lobes of the corpora cardiaca fly poorly, presumably because lipid is not mobilized and used. Injections of a small fraction of the corpora cardiaca restore, to some extent, “normal” flight in these operated animals (Goldsworthy ef al., 1973a). Adipokinetic hormone also facilitates use of lipid by flight muscles, since locusts injected with diglyceride appear to use the lipid as a substrate for flight only when corpora cardiaca extract is also injected (Goldsworthy et al., 1973a; Robinson and Goldsworthy, 1974,1976). 4.6.3
Other hormones
In locusts neurosecretory cells in the brain produce one or more hormones which are released from the storage lobes of the corpora cardiaca, but their functions are not known (Goldsworthy, 1976). Such brain hormones may be important in long-term flight in Schistocerca (Michel, 1973b; Michel and Bernard, 1973). Additional hormonal control mechanisms undoubtedly remain to be discovered. Cells that are neurosecretory on the basis of their morphology are distributed through the central nervous system of insects as well as along peripheral nerves (Maddrell, 1974). Diuretic and antidiuretic hormones may be released during flight, keeping the animal in water balance and allowing the removal of metabolic wastes (Goldsworthy, 1976). Neurosecretory endings within skeletal muscles have been described (Osborne et al., 1971; Hoyle, 1974; Hoyle et al., 1974) but it is not known if they have a function in flight. 4.6.4
Neural control of hormone release
Release of hormones that mobilize flight fuels appears to be under neural control. In the blowfly Calliphora erythrocephala stimulation of the brain
INSECT FLIGHT METABOLISM
177
caused release of hyperglycaemic hormone from the corpora cardiaca only if the nerve connecting the two was intact (Normann and Duve, 1969). Intracellular recording from the intrinsic neurosecretory cells of the corpora cardiaca showed that these cells received excitatory synaptic input (Normann, 1973). Electrical stimulation and high K+ solutions that depolarize neurons caused Ca++-dependent exocytosis (Normann, 1974). Flies in which the nerves to the corpora cardiaca had been cut did not maintain blood sugar concentrations during flight as did intact flies (Vejbjerg and Normann, 1974). In addition, intracellular recordings from neurosecretory cells with somata in the brain and axons in the nerves to the corpora cardiaca showed that these neurosecretory neurons received synaptic input from various sensory pathways and from thoracicoabdominal pathways (Bruce and Wilkens, 1976). The observed inputs, however, only weakly modulated ongoing pacemaker activity, but it is possible that the full range of inputs was not observed. Release of the hyperglycaemic hormone apparently occurs early in flight, as shown by rapid adjustment of blood sugar supplies. In Culliphoru hemolymph trehalose was greater than resting levels 6 min aRer flight began, and no decrease below resting levels was observed (Fig. 11) (Vejbjerg and Normann, 1974). In Phormiu trehalose and glucose concentrations in the thorax dropped during the first minute of flight and then remained stable (Sacktor and Wormser-Shavit, 1966). These initial changes may have reflected intramuscular rather than hemolymph sugar concentrations, but the supply rate rapidly equalled the rate of utilization. Perhaps the neural signals that initiate flight in blowflies also stimulate release of hyperglycaemic hormone. Similarly, in locusts mobilization of lipids begins soon after flight starts. Usually haemolymph lipid was elevated within 5-10min after the start of flight, and in some cases, only 2 min of flapping flight was sufficient to initiate lipid mobilization for over 1 h of subsequent rest (Jutsum and Goldsworthy, 1976). What controls the release of adipokinetic hormone is not known. It appears not to be simply a response to lowered blood trehalose or lipid (Houben and Beenakkers, 1975; Jutsum and Goldsworthy, 1976). In fact it appears that adipokinetic hormone is responsible for increasing the concentration of hemolymph lipids, and the resulting vigorous lipid oxidation competitively inhibits trehalose oxidation (Robinson and Goldsworthy, 1976). Release of adipokinetic hormone, however, can be influenced by blood sugar. Injection of trehalose immediately before a flight delays the release of hormone and lipid mobilization (Cheeseman et ul., 1976). Sucrose injections have a similar effect, although sucrose does not support flight. Perhaps the neural mechanisms that initiate flight also initiate the release of adipokinetic hormone, and except when the response threshold is raised by high levels of blood metabolites, release accompanies the commencement of flight. Neural signals triggering release of hormones from the corpora cardiaca could be carried from
178
ANN E. KAMMER AND BERND HEINRICH
the brain via axons of the neurosecretory cells themselves or via nonneurosecretory axons that also may be present in the nerves to the corpora cardiaca. The importance of connections to the brain has been demonstrated by sectioning these nerves, an operation that prevents release of adipokinetic hormone and impairs flight (Goldsworthy et ul., 1972a). These effects were observed in animals tested 2, 6, and 15 days after the operation; the data do not exclude the possibility that severing the nerves interrupted the transport of neurosecretory material from cell bodies in the brain to the corpora cardiaca, rather than interrupting conduction of action potentials. However, severance of the ventral nerve cord in the neck region prevents lipid mobilization (Mordue, unpub., cited in Goldsworthy, 1976). Further evidence for the neural control of the release of adipokinetic hormone is provided by pharmacological studies on Schistocercu greguriu (Samaranayaka, 1976). Acetylcholine stimulated isolated corpora cardiaca to release hormone. In intact locusts reserpine, which depletes neurons of their monoamines, blocked the insecticide-induced release of hormone. Therefore a chain of neurons, both cholinergic and aminergic, is implicated in the control of the corpora cardiaca and release of adipokinetic hormone. Thus, in the two best-studied cases, locusts and blowflies, release of the hormone that mobilizes the flight fuel occurs early in flight and is dependent on intact connections to the central nervous system. Perhaps the thoracic fight control center commands the release of hormones in addition to coordinating the flight movements.
4.7
HEMOLYMPH CIRCULATION
Movement of mobilized fuels from storage sites in the abdomen to the thoracic flight muscles depends on the circulation of the hemolymph. As noted previously, the circulation may be relatively inefficient compared to the capillary system of vertebrates, and the high concentration of metabolites in the blood may compensate for this inefficiency (Weis-Fogh, 1964b; Crabtree and Newsholme, 1975). Flow of hemolymph from abdomen to thorax is accomplished by orderly circulation (Wigglesworth, 1965). Typically the blood is pumped forward by the dorsal vessel and returned ventrally (Brocher, 1920). In some orders flow is assisted by rhythmic contractions of the ventral diaphragm (Wigglesworth, 1965). Definite separation between dorsal and ventral flows is shown by the fact that a temperature difference of 4 O C can be developed between the dorsum and ventrum of the first abdominal segment of an active moth (Heinrich, 1971b). In addition to circulation caused by contractions of the heart and ventral diaphragm, hemolymph may be pumped by movements of the abdomen. Such
INSECT FLIGHT METABOLISM
179
abdominal pumping may occur during pre-flight warm-up, during flight (Miller, 1966; Kammer, 1976), or as a corollary of temperature regulation (Heinrich, 1976b). Although the primary function of abdominal pumping is probably ventilation, these movements may also supply blood-borne nutrients to the thoracic muscles. Hemolymph flow between the thoracic muscles is unidirectional in a resting moth (Brocher, 1920). The effects of flight on this flow, as well as on the intramuscular circulation, have not been described. Circulation between muscle fibers is usually assumed to be predominantly tidal, and the vigorous contractions of the flight muscles would contribute to this tidal flow. In addition, channels of hemolymph extend into the fiber in the T-tubules (Smith, 1966; Smith and Saktor, 1970), but the small dimensions of these extracellular spaces make rapid replenishment of nutrients unlikely. Little information is available about hemolymph circulation during flight, but two studies suggest that the rate is not controlled to match the metabolic demands of the flight muscles. In the blowfly Calliphora eryvthrocephala heartbeat is variable, but during short flights the rate is usually slightly lower than the rate in resting flies (Normann, 1972). In Manduca sexfa the heartbeat rate varies with thoracic temperature, and changes in rate of flow between thorax and abdomen provide a mechanism for regulation of thoracic temperature, since the abdomen rapidly loses excess heat (Heinrich, 1971b). Over a range of ambient temperatures, the metabolic rate during flight is uniformly high, but at low ambient temperature, the blood flow between thorax and abdomen (as inferred from temperature differences) is less than the flow at high ambient temperatures. It is unlikely that the circulation in these moths can be regulated to serve two different functions (heat exchange and nutrient supply) simultaneously. Bumblebees, however, may be able to regulate both blood flow and heat exchange between thorax and abdomen. Under conditions of experimentally induced heat stress, the counter-current heat-exchange that tends to retain heat in the thorax is reduced by altering the timing of contractions in the heart and ventral diaphragm (Heinrich, 1976). It is not known if this mechanism operates during flight at various ambient temperatures.
4.8
SUBSTRATE AVAILABILITY A N D FLIGHT S P E E D
Unlike an automobile that abruptly stops when it runs out of fuel, insects do not fly at top speed and then suddenly stop when their fuel is exhausted. Rather, substrate supply can influence flight speed and wingbeat frequency over a range of intermediate values. For example, in Phormia regina there was a correlation between wingbeat frequency and trehalose concentration in the hemolymph (Clegg and Evans, 1961). In tsetse flies wingbeat frequency declined rapidly during the first few minutes of flight, and at the same time the
180
ANN E KAMMER AND BERND HEINRICH
proline concentration dropped steeply (Hargrove, 1975b). The flight speed of Locusta migratoria flying on a roundabout was reduced by starvation or removal of the corpora cardiaca and restored briefly in starved locusts by injection of trehalose (Goldsworthy and Coupland, 1974). Unexpectedly, injection of 6 mg dipalmitin emulsion slowed the flight of Schistocerca for at least 30-40 min (Robinson and Goldsworthy, 1976). Injection of corpora cardiaca alone also slowed flight, an effect that could be caused by hypoglycaemic hormone with anti-lipolytic action. Injection of both dipalmitin and corpora cardiaca extract resulted in normal flight. These effects were attributed to the action of adipokinetic hormone on the flight muscle, suppressing carbohydrate utilization and enhancing lipid utilization (Robinson and Goldsworthy, 1974,1976). These authors also showed that an extract of the glandular lobe of the corpora cardiaca increased the oxygen consumption of isolated muscles incubated with dipalmitin. Since the muscles were deprived of their normal oxygen supply and innervation, their oxygen consumption was low [the highest value reported was 0.082 ml/min/g, which can be compared with an estimate of 1-2.5 ml/min/g muscle for flying locusts (Weis-Fogh, 1952)l. Hence some question remains about the effect of the hormone on intact muscles during flight. Nevertheless the results are suggestive of a direct effect of hormones on flight muscle metabolism and of a causal effect on flight performance. If flight is controlled by the central nervous system, how do hormones and substrates influence the power output? For insects such as flies in which the wingbeat frequency is determined by the properties of the fibrillar muscles and the mechanical resonance of the wing-thorax load, the effect of limited substrate may be analogous to that of low temperature (cf. section 5.1) (Machin et al., 1962). Possibly with limited substrate the rate at which the flight muscles can do work is less, and thus the oscillatory contractions of the flight muscles occur at a lower frequency, although the neural output continues unchanged. Low substrate concentrations could also alter the force exerted by tonic muscles that control the stiffness of the thorax and thereby control wingbeat frequency. For insects such as locusts, in which the flight rhythm is neurogenic, the explanation may be similar. The flight muscles may contract more slowly in a substrate-deficient flyer than in a well-fed animal. The neural output rhythm does not change, as indicated by the observation that the wingbeat frequency of starved Locusta was not reduced although the insects flew more slowly (Goldsworthy and Coupland, 1974). It is likely that stroke amplitude was smaller as a consequence of the inability of the energy-deficient muscles to generate force as rapidly as in normal flight. 5
Interrelations of flight muscle temperature and metabolic rate
The metabolic rate of flying insects may be influenced by many factors, including morphological constraints, neural and hormonal control mechan-
INSECT FLIGHT METABOLISM
181
isms, and substrate availability. The rate is also strongly influenced by thoracic temperature. As discussed in section 2.1, in small insects thoracic temperature is determined by ambient temperature, but in large insects during flight the metabolic rate is independent of ambient temperature. These large insects produce the high thoracic temperatures required for flight, 30-35 O C, by a preflight warm-up. This endothermic control of thoracic temperature is an indirect and energetically expensive method of controlling flight metabolism. Because of the important influence of temperature on flight metabolism, we shall discuss the effects of temperature and the mechanisms of thermogenesis in some detail. 5.1
EFFECTS OF TEMPERATURE ON THE FLIGHT MOTOR
Thoracic temperature limits the rate at which the muscles can do useful work. Temperature also influences the production of motor patterns by the central nervous system. Both of these effects influence the metabolic rate during flight.
201
' 25
30
35
40
45
Muscle temp.('C)
Fig. 14. Durations of single twitches of locust dorsal longitudinal (hindwing depressor) muscle as a function of muscle temperature (from Neville and Weis-Fogh, 1963).
The temperature dependence of maximum work output is in large part related to twitch duration. For example, in Schistocerca greguriu the twitch duration of hindwing depressor muscles is about 50 ms at 25O C and about half that at 40° C (Fig. 14; Neville and Weis-Fogh, 1963). A similar decrease of twitch duration at the higher temperatures required for flight occurs in the depressor muscles of Munducu sextu (Rheuben and Kammer, in MS). Short twitch durations are required for rapid rhythmic movements. Some of the fastest neurogenic muscles known, the singing muscles of the katydid
182
ANN E. KAMMER AND BERND HEINRICH
Neoconocephalus robustus, go into tetanus when stimulated at 150/s (their working frequency) at 25O C, although they exhibit discrete twitches when stimulated at the same frequency at 35' C (their working temperature) (Josephson, 1973). Both synaptic transmission and the chemical reactions of contraction can be temperature-sensitive, but these effects have not been carefully examined in insects. Experiments on voltage-clamped crayfish (Orconectes limosus) muscle have shown directly that temperature affects the mechanical properties of the muscle independently of membrane potential (Dude1 and Rudel, 1968); cooling reduced the maximum force of contraction, the rate of rise of tension, and the rate of relaxation. In another arthropod, the crab Ocypode ceratophthalma, amplitudes of both tension and of excitatory post-synaptic potentials were maximal between 22O and 28O C and decreased at temperatures below and above this range; the environmental temperatures to which the crabs were exposed were 26O-27.5O C (Florey and Hoyle, 1976). In this experiment and in the studies on insects cited above, it is not possible to distinguish between the direct effects of temperature on the contractile mechanism and changes in tension resulting from the influence of temperature on neuromuscular transmission. Regardless of mechanism, however, temperature-dependent changes in the rate of contraction and relaxation can alter the rate at which work can be done. Prolonging twitch duration can result in less efficient flight. In neurogenic flyers, in which each contraction of one of the power-producing muscles of the flight motor is initiated by an action potential, the frequency of muscle contractions is controlled by the nervous system. However, if the thoracic temperature is low and the rate of muscle shortening and relaxing is slow, the wing elevators may still be contracting while the wing depressors are being activated, and vice versa. As a result, unless wingbeat frequency is low, the antagonistic flight muscles will exert their force primarily against each other rather than on the wings. Conversely, when muscle temperature is high, the energy of the muscular contractions can be applied to the wings, thus increasing the mechanical efficiency of the flight motor. At temperatures suitable for flight, the muscles do not stretch their antagonists until after these have completed most of their contraction (Neville and Weis-Fogh, 1963). The rate of work output of the flight muscles, and the efficiency of the flight motor, are not only related to the temperature-dependent intrinsic rate of muscle shortening. In neurogenic flyers, the neural flight pattern generator is also influenced by temperature. For example, in moths stimulated to fly at low thoracic temperatures, the wingbeat frequency (which is determined by the CNS) increased as thoracic temperature rose (Heinrich and Bartholomew, 1971). Similar results were obtained in other moths (Dorsett, 1962; McCrea and Heath, 1971) and in monarch butterflies (Kammer, 1970b) (Fig. 15). Experiments in which the thoracic ganglia were heated with a small thermode
183
INSECT FLIGHT METABOLISM
show directly that the flight-pattern generator is temperature sensitive (Hanegan and Heath, 1970). In insects with a myogenic flight rhythm, the wingbeat frequency and by inference the metabolic rate are probably more strongly influenced by the effects of temperature on the fibrillar muscles than by the effects of temperature on the central nervous system. The oscillatory power output of isolated fibrillar
-1
** 0 0 0
40
E
L
.+ I E
f
A b
I
A0 A
*
o
I
4
*
0
*
*
A
I
/' /'
A 0
* 4
h
0
/' /-
I
Thoracic temp. ("C)
Fig. 15. Wingbeat frequencies in relation to thoracic temperature of a butterfly, Danaus plexippus, during fixed flight [--I (Kammer, 1970b) and of various sphingid moths during warm-up. 0,Euchloron magaera ; A, Deilephila nerii; 0,Pseudoclaris postica (adapted from Dorsett, 1962); -----, Manduca sexta (adapted from McCrea and Heath, 1971); -----, M . sexta (adapted from Heinrich and Bartholomew, 1971).
muscle increases as the temperature rises (Machin and Pringle, 1959; Machin et al., 1962). In bumblebees, the action potential frequency in the flight muscles increases with thoracic temperature during warm-up (Heinrich and Kammer, 1973; Kammer and Heinrich, 1974), but these changes indicate a correlation not a causal relationship. In flying honeybees the spike frequency increased approximately 50 per cent as the thoracic temperature rose from 24O to 38O C, but the change was nonlinear and greatest for the small increment from 24O to
184
ANN E. KAMMER AND BERND HEINRICH
26' C (Esch, 1976). We know of no studies on Hymenoptera or Diptera in which the effects of temperature on the neural rhythm generator have been examined directly. If spike frequency in the fibrillar muscles does vary with body temperature, then the low wingbeat frequencies of small flies at low ambient temperatures (Fig. 2) would be caused by reduced neural activation as well as by reduced rates of mechanical oscillation directly influenced by muscle temperature. In addition to the effect of temperature on the flight muscles and the central nervous system, the duration of the action potentials themselves is temperature dependent. In honeybees there is a very rapid increase in the duration of action potentials as muscle temperature approaches 18O C (Esch and Bastian, 1968). In bumblebees distinct action potentials can be recorded at a thoracic temperature of 13O C, and the potentials are also of long duration (Heinrich and Kammer, 1973). Similarly, the duration of action potentials in the leg muscles of a locust are increased at low temperatures (Del Castillo et al., 1953), and the locust flight muscles probably have similar properties. The significance of these changes in duration of action potential for flight metabolism is not clear. In bumblebees the oxygen consumption per spike did not vary systematically with thoracic temperature (Kammer and Heinrich, 1974 and unpubl.), suggesting that rate at which events proceeded did not alter the total work done. As noted above, most experiments do not distinguish between the effects of temperature on neuromuscular transmission and on the processes of contraction and relaxation. More data are required before the effects of prolonging the action potential by lowering the temperature can be understood. It is clear, however, that temperature can limit the rate at which a muscle adapted to a particular temperature range can do useful work.
5.2
PRE-FLIGHT WARM-UP
Because of the effect of temperature on the flight motor, an insect forced to attempt flight at a low thoracic temperature will not be able to become airborn. If the insect is sufficiently large and insulated, wingbeating at low muscle temperature would eventually elevate thoracic temperature until muscle twitches are brief and flight is possible. Wingbeating, however, has disadvantages during warm-up. Firstly, it results in convection currents of air that would remove some of the endogenously generated heat. Secondly, a stationary insect might fray its wings on the substrate on which it is perched. Thirdly, the energy expended in moving the wings is wasted and would be more profitably invested in heating the muscles to temperatures at which flight is possible. Insects that fly with an elevated thoracic temperature generally do not
INSECT FLIGHT METABOLISM
185
encounter the above difficulties but have a pre-flight warm-up. During warmup the tendency of the muscles to contract upon each other is accentuated. It occurs under neural control even after the muscles are warm. The muscles are activated synchronously, and thus they contract nearly synchronously, rather than alternately as in flight. The pattern of activation of the power-generating muscles during warm-up varies between neurogenic and myogenic flyers, as well as between species
Celerio lineal0
Bombus sonorus
Fig. 16. Extracellularly recorded action potentials from the indirect flight muscles during warmup (left) and fixed flight (right). (1) The saturniid moth, S.cynlhiu (from Kammer, 1968). Upper trace, wing depressor muscle; lower trace, wing elevator muscle. During warm-up the wingvibration rate is high and the muscle units are activated by single spikes per wingbeat. During flight the wingbeat frequency is low and the muscles are activated by several spikes per wingbeat. (2) The sphinx moth C. (=Hyles) lineura (from Kammer, 1970). The wing depressor muscles (d) and wing elevators (st, e, and t) are activated usually by single spikes that occur synchronously during warm-up and alternately during flight. (3) The bumblebee B. sonorus (from Kammer and Heinrich, 1972). The various indirect flight muscles are activated with phase preferences during warm-up but not during flight. Time marks, 100 ms.
within each of the groups (Fig. 16). During warm-up in Lepidopteru the wings execute low-amplitude wingbeats, analogous to shivering in vertebrates. In sphinx moths, each wingbeat both during warm-up and during flight is associated with one or two spikes for each motor unit of the major powerproducing muscles (Kammer, 1968, 1970a). Wingbeat frequency and hence frequency of muscle contractions increase throughout warm-up until flight temperature and flight wingbeat frequency are reached (Heinrich and Bartholomew, 1971). The animals than change the phasing of the activation
186
ANN E. KAMMER AND BERND HEINRICH
between wing depressors and wing elevators, making them alternate, but maintain approximately the same wingbeat frequency. In saturniid moths, which have low wingbeat frequencies during flight, the wings beat relatively rapidly during warm-up, and each wingbeat is correlated with one or two action potentials from each power-producing motor unit (Fig. 16; Kammer, 1968, 1970b; Hanegan and Heath, 1970). After the transition from warm-up to flight, the wingbeat frequency is low, and the wings move through a large
5
=.-E
i -
4-
3
"
1
c
.o 3 t;
40
30
0
en
- ss I
-
L
2-
20
0
z
-
-
1-
0 10
20
30
-
.0
>
10
40
Thoracic temporaturr ('C)
Fig. 17. Rates of wing-vibration and calculated rates of heat production during pre-flight warmup in the sphinx moth Manducu sextu in relation to thoracic temperature. Arrows at 41' C indicate the average wingbeat frequency and rate of heat production of a 1.5 g moth during free flight (from Heinrich, 1974a).
arc; each wingbeat involves a burst of several muscle spikes per single muscle contraction (Kammer, 1967, 1970b; Hanegan and Heath, 1970). Whether or not these different motor patterns result in different metabolic rates at the same thoracic temperature is not known. Metabolic rates and rates of heat production increase substantially during warm-up. From the changes of thoracic temperature during warm-up in the sphinx moth Munducu sexfu, it has been calculated that heat production increases linearly from about 1 caVmin at a muscle temperature of 15O C to about 4caVmin at 4OoC in a 1.5g moth (Fig. 17) (Heinrich and
INSECT FLIGHT METABOLISM
Bartholomew, 197 1). Wingbeat frequency during warm-up is also a direct function of muscle temperature (Figs 15, 17) and parallels the increase in metabolic rate until muscle temperature, metabolic rate and wingbeat frequency are close to those observed in flight (Fig. 17). Similarly, in myogenic flyers (syrphid flies (Heinrich and Pantle, 1975) and a bumblebee, Bombus vosnesenskii (Heinrich and Kammer, 1973)) heat production calculated on the basis of thoracic (primarily muscle) weight increases from about 2 cal/g thorax per minute at 10" C to about 15 cal/g thorax at flight temperature (36" C) The latter temperature is close to that observed during free flight when the abdomen is unloaded (Heinrich, 1975). The rate of increase in thoracic temperature in Manduca sexta is similar whether or not the moths are warming up by contracting their powerproducing muscles against each other in warm-up, or whether they are contracting them alternately by moving their wings in fixed flight (Heinrich and Bartholomew, 1971). It appears, therefore, that the timing of the activation of the muscles with respect to one another has little or no bearing on their rate of heat production. The rate of heat production during warm-up and flight is determined by the action potential frequency. During warm-up in myogenic flyers, as in neurogenic flyers, the flight muscles are activated such that heat but not wing motion is produced. In bumblebees, for example, antagonistic muscles are activated simultaneously (Kammer and Heinrich, 1972; Fig. 16). In myogenic flyers the relationships between the kinetics of muscle contraction, metabolic rate and neural control of muscle contraction during warm-up are not clearly known. One of the primary difficulties in an analysis of muscle contraction in the intact insect is that the wings do not move during warm-up, so that in general the contraction kinetics of the muscles in the thorax are difficult to ascertain. It has been shown for flies (Wilson and Wyman, 1963; Nachtigall and Wilson, 1967) and honeybees, Apis mellifera (Esch and Bastian, 1968; Bastian and Esch, 1970) that wingbeat frequency during flight is directly related to action potential frequency. Furthermore, Bastian and Esch (1970) and Kammer and Heinrich (1974) showed for honeybees and bumblebees respectively that the rate of oxygen consumption, during both warm-up and flight, is a constant function of action potential frequency (Fig. 18). The honeybees consume about 1.15 pl O,/g body weight per action potential, and the bumble bees consume about 0.8 pl O,/g body weight per action potential. These results suggest that the metabolic rate during warm-up depends on action potential frequency and hence is controlled in part by the central nervous system. However, the methods used did not allow for differences among motor units and thus other possible mechanisms influencing metabolic rate would not have been resolved. Relationships between contraction and action potential frequency are more
188
ANN E. KAMMER AND 8ERND HEINRICH
I
0
' 2
.
4
.
6
.
.
8
.
.
10 12
.
14
&
16 18
02 uptake ( p l / g r n h i n x lo2)
3
Fig. 18. Relationship between metabolic rate and spike frequency in the fibrillar muscles of Hymenoptera. (a) Apis melliferu; 0, warm-up; A, flight (after Bastian and Esch, 1970). (b) Bombus vosnesenskii; H,0,warm-up; 0, flight (from Kammer and Heinrich, 1974).
189
INSECT FLIGHT METABOLISM
apparent during flight than during warm-up. During flight in Calliphora the 120 wingbeats per second are accompanied by 3 action potentials (Pringle, 1949), in Syrphus flies the action potentialhngbeat ratio is about one to twenty (Heinrich and Pantle, 1975), and in the honeybee it is about one to twelve (Esch, 1964). During warm-up in honeybees (Esch, 1964), syrphid flies (Heinrich and Pantle, 1975) and bumblebees (Kammer and Heinrich, 1972), thoracic and wing vibrations are usually not detectable. Tension production by fibrillar muscles can be influenced by spike frequency, however. Esch and Bastian (1968) showed a linear correlation between spike frequency and Warm-up A
-
A
A 4
1
-
. . ..
--------....... ~
....-....... _..
Flight
Fig. 19. Records from a Syrphus fly showing that thoracic vibrations (bottom trace) may be produced or may be absent during warm-up (A-C) but vibrations are continuous during flight (D). Muscle potentials were recorded from the same wing elevator (top trace) and wing depressor muscles (middle trace) throughout the records. Time mark, 200 ms (from Heinrich and Pantle, 1975).
tension generated by dorsoventral muscles of honeybees. Usually all observable elements fire with phase differences, but when apparently accidental synchronization occurs the muscles respond with small twitches 2 to 3 pm in excess of the contraction characteristic of the stimulation level. Boettiger (1 957) examined tension changes in semi-isolated Bombus muscle artificially stimulated. The muscle was loaded with a weight, and shortening then occurred isotonically. Stimulation with a sufficiently high frequency shortened the muscle to a specific point beyond which it started oscillatory cycles of contractions. These results suggest that the muscles in bumblebees during warm-up would also be stretched upon electrical stimulation occurring during
190
ANN E. KAMMER AND BERND HEINRICH
this process. The bursts of impulses (Mulloney, 1970; Kammer and Heinrich, 1972) must result in small muscle twitches, but since the antagonistic muscles are activated synchronously and since the wing articulations are inoperative, it is likely that the muscle contractions of the antagonistic muscles oscillate independently in the manner observed by Boettiger (1957) (see also Pringle, 1957, p. 59). In Syrphus flies thoracic vibrations are sometimes conspicuous during warm-up and absent at other times (Fig. 19); vibration rate during warm-up is higher than the wingbeat frequency during flight at a given temperature and spike frequency (Heinrich and Pantle, 1975). In myogenic insects, therefore, contraction frequency is influenced by neural excitation, as in neurogenic flyers, but other factors such as load on the muscle are also important. But since metabolic rate and heat production are independent of the proportioning of muscle energy between shortening and isometric force generation, heat production is a function of action potential frequency in myogenic as well as neurogenic flyers. 5.3
STABILIZATION OF THORACIC TEMPERATURE DURING FLIGHT
It is now well-established that during flight the thoracic temperature of large, well-insulated insects, such as moths and bees, is independent of ambient temperature over a wide range (Heinrich, 1973, 1974a). As a corollary, metabolic rate in these insects is independent of ambient temperature over that range in which thoracic temperature can be stabilized (Fig. 1). During flight, at least in those insects that stay aloft primarily as a consequence of their muscular energy, the neural input to the flight muscles appears to be committed primarily for power output and for manoeuvring, rather than for temperature control. Recent controversy in this area, relating primarily to thermoregulation in sphinx and saturniid moths, has been reviewed (Heinrich, 1974a). Subsequent data from several species of sphinx moths have corroborated the conclusion that large insects stabilize their thoracic temperature during continuous flight by regulating heat loss (Casey, 1976a). Metabolic rate in these moths (Heinrich, 1971a; Heinrich and Casey, 1973; Casey, 1976a) and in thermoregulating bumblebees (Heinrich, 1975) remains independent of ambient temperature during uninterrupted flight. It is quite possible, however, that thoracic temperature stabilization may also be achieved by adjustment of metabolic rate, particularly in those insects that can continue flight while interrupting their wingbeats. Nachtigall (1967) has pointed out that some butterflies regularly sail. He has observed ZphicZides pOduZirius L. for ;p to 1 h, sailing about 80 per cent of the time, with periods of up to 30 s without wingbeats. Obviously such an insect has the opportunity to avoid overheating itself with endogeneously produced heat (which must be of
INSECT FLIGHT METABOLISM
191
relatively small amount) during flight at high ambient temperatures simply by increasing its percent gliding time. Anecdotal evidence suggests that some dragonflies which remain continuously in flight, and which regulate thoracic temperature near 40° C independently of ambient temperature, may glide more at high than at low ambient temperatures (May, 1976). Those insects which are unable to glide (i.e. insects with high wing-loading) do not have gliding available as a behavioral option, and they must either cease flight or initiate physiological cooling mechanisms to continue activity. Heavy-bodied insects such as sphinx moths and bumblebees initiate active cooling at ambient temperatures greater than 20° C (Heinrich, 1971a, 1975). In at least two types of bulky insects with a high rate of continuous wingbeat during flight, sphinx moths and bumblebees, the heat generated by flight metabolism is not dissipated passively from the thorax rapidly enough to prevent overheating. Excess heat is transported by the hemolymph to the abdomen, from which it is lost by convection to the environment (Heinrich, 1970; Casey, 1976a). In the sphinx moth Manduca sexta, the flow of blood between thorax and abdomen appears to be restricted during warm-up (Heinrich and Bartholomew, 1971). [Interestingly, for those dragonflies that bask there is suggestive evidence that the heat-transfer mechanism may act in reverse. May (1976) suggests that the dragonflies absorb solar radiation by way of the large flat abdomen, and transport the heat by way of the hemolymph to the thorax.] It has recently been shown in bumblebees that heat transfer between thorax and abdomen can be, in part, independent of blood flow because of countercurrent heat exchange in the abdominal petiole. During heat transfer to the abdomen from the thorax, the counter-current heat exchange is physiologically minimized by shunting the blood through the petiole in alternate pulses, rather than having it flowing in both directions at the same time (Heinrich, 1976). It is not known if this mechanism contributes to the regulation of flight metabolism. 5.4
SHIVERING A N D NONSHIVERING THERMOGENESIS
Bumblebees, like other Hymenoptera, metabolize primarily carbohydrate as an energy source (Sacktor, 1965, 1975). The primary pathway of carbohydrate utilization (Fig. 9) involves the conversion of fructose 6-phosphate to fructose 1,6-diphosphate, a reaction catalysed by phosphofructokinase. The fructose 1,6-diphosphate may then be converted to 3-C compounds that are utilized in aerobic metabolism by the mitochondria. Alternatively fructose 1,6-diphosphate may be converted to fructose 6-phosphate by another enzyme, fmctodiphosphatase (Fig. 20) in the first steps of glycogen synthesis. In most insects utilizing carbohydrate as fuel for flight, the activity of phosphofruc-
192
ANN E. KAMMER AND BERND HEINRICH
tokinase in the flight muscle is high relative to that of fructose diphosphatase, indicating that catalysis proceeds predominantly in the direction of carbohydrate degradation. It has recently been determined, however, that the maximum catalytic activities of fructose diphosphatase from the flight muscles (dorsal longitudinal) of Bombus spp. are at least 30-fold those of other tissues (Newsholme et al., 1972). In bumblebee muscle the activities of fructose diphosphatase, hexokinase and phosphofructokinase are similarly highly elevated relative to muscles of most other animals. In some insect muscles the maximum activity of hexokinase is about 10 times that of most vertebrate
Glucose
Fructose 6 phospate
Fructose diphosphatase
fructokinase
H20
k7uctose 46 diphosphate Fig. 20. Substrate cycle catalyzed by fructose diphosphatase, FDP, and phosphofructokinase, PFK.
muscles, and in Bombus hortorum the hexokinase activity of the flight muscles is the highest known. But phosphorylase (assayed in direction of glycogen synthesis) was low (Crabtree and Newsholme, 1972a). Newsholme et al. (1972) hypothesize that the uniquely high activity of fructose diphosphatase in Bombus flight muscle is functional in nonshivering thermogenesis. They propose that both fructose diphosphatase and phosphofructokinase are simultaneously active and catalyse the cyclic interconversion of fructose 6-phosphate and fructose diphosphate in resting muscle. Operation of this substrate cycle (Fig. 20) results in the following net reaction:
ATP + H,O + ADP + P, + heat Hydrolysis of ATP in the resting muscle should release heat, helping to maintain a high thoracic temperature. Several lines of indirect evidence (Newsholme ef al., 1972) are consistent
INSECT FLIGHT METABOLISM
193
with the hypothesis. (1) There is no detectable activity of fructose diphosphatase in nonflight tissues of Bombus. (2) It is improbable that fructose diphosphatase is present to lower the content of glycolytic intermediates after a period of flight, since there is no significant difference between resting and active muscle in fructose 6-phosphate and fructose 1,6-diphosphate activity. (3) It is unlikely that the fructose diphosphatase is involved in gluconeogenesis since the tissues contain little phosphoenolpyruvate carboxykinase or glucose 6-phosphatase. (4) Fructose diphosphatase from flight muscle of Bombus is not inhibited by AMP at concentrations up to 5 mM, as it is in other muscles examined. The hydrolysis of ATP can thus presumably proceed at a high rate at the same time that fructose diphosphate is reconverted to fructose 6phosphate. ( 5 ) The activities of both fructose diphosphatase and phosphofructokinase vary inversely with body weight of the bee, whereas hexokinase activity is independent of body weight. Since the rate of heat loss is inversely related to body size this result suggests that the more rapidly the bees lose heat the greater the rate of substrate cycling. The above in vitro studies were subsequently extended by in vivo studies. Clark et al. (1973), using a method developed by Bloxham et al. (1973) to measure the rate of substrate cycling determined that there was no evidence for substrate cycling during flight at any temperature. In nonflying bumblebees (Bombus afinis) the rate of cycling was high, but only if ambient temperature was low. Recently' Clark (1976) found that 2-deoxy-~-glucoseinhibited both flight muscle glycolysis and substrate cycling of fructose 6-phosphate. At high ambient temperatures (30-33O C) the treated bumblebees could fly but at low ambient temperatures (14O C) they could not. He proposed that flight at high temperatures was supported by oxidation of lipid but that at low temperatures flight was impaired because the necessary pre-flight heat production, which is dependent on fructose 6-phosphate cycling, was inhibited. However, inhibition was not complete but rates were decreased 60-73 per cent, allowing some contribution of carbohydrate metabolism to energy production. The above results are highly indicative of temperature-dependent substrate cycling in inactive bumblebees. The evidence suggests that bumblebees maintain an elevated thoracic temperature while stationary by nonshivering thermogenesis. However, there are no measurements showing that bumblebees have elevated thoracic temperatures in the absence of mechanical activity in the flight muscles. Low ambient temperature does not, as such, predict whether or not the bees are thermoregulating, since the animals may remain in torpor for many hours. As discussed in section 5.2, thermogenesis by shivering is another mechanism for elevating body temperature in bumblebees (Kammer and Heinrich, 1972, 1974; Heinrich and Kammer, 1973). The power-producing muscles are activated when the bees warm up and when they maintain an
194
ANN E KAMMER AND BERND HEINRICH
elevated thoracic temperature. In no case in which we simultaneously measured thoracic temperature and electrical activity of the flight muscle did we observe an elevated temperature in the absence of muscle activity. In thermoregulating, nonflying bumblebees there are no visible movements of wings or thorax. It is reasonable to suppose that the muscles are caused to contract when they are depolarized during recorded muscle potentials, but the muscle contractions are not transmitted as movement to the wings when the wings are folded dorsally over the abdomen. Similarly, in syrphid flies, another insect with a myogenic flight rhythm, thoracic temperature is maintained above ambient in stationary (“resting”) individuals that activate their flight muscles. During warm-up flies may move their wings through a very small amplitude (so that the wings outlines appear slightly “fuzzyyyto the unaided eye) and thus produce a humming sound, or they may hold their wings motionless, folded over the dorsum of the abdomen, without any detectable vibrations. In both situations the electrical activity recorded from the flight muscles is identical (Fig. 23, Heinrich and Pantle, 1975). It can be concluded that the electrical activity of the flight muscles during warm-up in myogenic flyers is associated with muscle contractions. At present the contraction kinetics of the muscles during warm-up are unknown, but the data suggest that the mechanism of heat production during shivering is the same as during flight. Boettiger (1955) isolated dorsal longitudinal muscle from bumblebees and examined contraction kinetics following applied electrical stimuli. As with other muscles, those of bumblebees contract following electrical stimulation. The dorsal longitudinal muscles of the bumblebee, however, contract in a series of twitches when they are under constant load (Boettiger, 1957). In the intact animal the load on the muscles can be controlled by accessory muscles, the intersegmentals and the pleurosternals (Pringle, 1957, 1974), so that both the contraction kinetics of the muscles as well as the transmission of their power to the wings are under regulatory control. The electrical activity of Bombus indirect fight muscle during warm-up, thoracic temperature stabilization, and fight is a fair indicator of the rate of aerobic metabolism (Kammer and Heinrich, 1974; Heinrich and Kammer, 1973). Regardless of the precise nature of the mechanical activity of the muscles during warm-up, one action potential on the average is associated with the consumption of 2.3-2.6 pl OJg thorax/h. During pre-flight warm-up (lasting a minimum of about one minute to 17 min at 30 and 6O C respectively; Heinrich, 1974a) the rate of heat production is directly related to action potential frequency (Kammer and Heinrich, 1974). The calculated rate of heat production at various temperature excess (TTh- 7’’)is also directly related to the action potential frequency (Heinrich and Kammer, 1973). Since the aerobic metabolism proceeds by way of the Krebs cycle that is fed glycolytic intermediates derived from the fructose diphosphate reaction, it can be strongly
INSECT FLIGHT METABOLISM
195
inferred that during continuous warm-up and thoracic temperature stabilization there is no, or only very little, substrate cycling like that discussed above (Fig. 20). It is not known under what circumstances bumblebees use nonshivering thermogenesis instead of shivering, and how the former process is controlled. The rates and amount of heat production are also not clear. The estimated rate of cycling and hence the rate of ATP hydrolysis in bees exposed to 5 O C ambient was 10.4 pmol.min-’.g-’ (Clark et al., 1973). For fly muscle at 25O C the maximum rate of ATP production from glycogen could be 2400 pnol. min-’ g-’ (Crabtree and Newsholme, 1975). Simultaneous measurements of thoracic temperature, oxygen consumption, muscle potential frequency, and fructose phosphate cycling could help answer these questions. It is of interest that both the cockoo-bumblebees, Psithyrus spp., and the honeybee, Apis mellifera, have very low activities of fructose diphosphate relative to phosphofructokinase and hexokinase, and thus apparently lack substrate-cycling capability (Newsholme et al., 1972). Both kinds of bees are able to forage at low ambient temperature by maintaining an elevated thoracic temperature while in flight and while stationary, but their thoracic temperatures and their rates of foraging are generally conspicuously lower than Bombus. In addition, they do not forage at as low temperatures as Bombus. For example, at an ambient temperature of 30° C Psithyrus ashtoni stopped on the average 58.3 s (10-185 s, N = 16) at each group of Chemaedaphne calyculata (Ericaceae) blossoms. Bombus ternarius foraging on the same species stopped on the average 4.3 s (0.5-10 s, N = 20) at the flowers, even at 15O C. Bombus visited 8.2 inflorescences per minute while Psithyrus only visited 2.2. The average flight duration of the foraging Bombus at 15O C was 2.5 s (0.5-8.0 s, N = 23). The average thoracic temperature of seven Bombus queens at 19O C was 34.8O C (33.2-37.1O C), whereas that of a Psithyrus ashtoni was 3 1.5O C at the same time and place (B. Heinrich, unpublished observations). These preliminary results do not prove that Psithyrus forage more slowly than Bombus because they have a lower thoracic temperature, or because they have low fructose diphosphatose activity. But they do suggest that Bombus has experienced strong selective pressures for rapid foraging and that temperature regulation during foraging involves very rapid on-off activity during which nonshivering thermogenesis could be involved.
-
5.5
W H Y R E Q U I R E A H I G H MUSCLE TEMPERATURE?
The question “why require a high muscle temperature?” can be recast from the standpoint of both proximal and ultimate causes. Why has the biochemical machinery evolved to operate in a particular range of temperature, and how
196
ANN E KAMMER AND BERND HEINRICH
does temperature affect the functioning of that machinery at any one time? The latter question relating to the effects on temperature on flight has already been discussed. We are here left to ponder why the muscles have evolved to function at one or another range of temperatures. The answer is beyond proof, but an analysis of muscle mechanics, heat transfer, and comparative physiology may provide insights. In all endothermic insects a high muscle temperature is not only a necessity for activity at high rates, but it is also a consequence of this activity. It is generally accepted that the major portion of the energy expended during flight is degraded to heat in the flight muscles. Weis-Fogh (1972) has calculated that in insects, as well as in birds, approximately 20 per cent of the energy expended by the flight muscles appears as mechanical power. The rest of the energy immediately appears as heat within the muscle. Of the 20 per cent utilizable energy only about 10 per cent is applied by the wings to the air. The other 10 per cent also appears as heat in the body. One would expect, therefore, that at least 90 per cent of the considerable energy expenditure of a flying insect is degraded to heat in the thorax. The high body temperatures during activity depend primarily on high rates of heat production and thus on the high metabolic rates of the flight muscles, rather than on especially effective insulation. On the basis of thoracic weight the weight-specific conductance of saturniid and sphingid moths is comparable to that of birds and mammals (Bartholomew and Epting, 1975a,b). Body size as such does not appear to be a large factor affecting the intensity of flight metabolism (heat production). But rates of heat loss vary as a function of body size. Small insects such as mosquitoes (Aedes), black flies (Sirnuliurn), and fruit flies (Drosophila) weighing about 2 mg or less on theoretical grounds probably cannot heat up more than 1O C during flight (Heinrich, 1975). Bulky fliers, on the other hand, inevitably heat up during flight. The sphinx moth, Manduca sextu, for example, heats up about 15O C above ambient temperature even when the insulating scales from the thorax are removed (Heinrich, 1970). High metabolic rates, such as those occurring during flight, presumably imply specialized enzyme systems (Hochachka and Somero, 1973). The enzymes must be designed to operate at peak efficiency at those temperatures experienced by the tissues when the animal is active. In large insects during flight this temperature-will be determined less by environmental temperature than by the flight activity itself. Small insects can -specialize their enzyme systems to operate more closely to prevailing environmental temperatures during activity times. As suggested previously’ (Kammer and Heinrich, 1974) and discussed in more detail elsewhere (Heinrich, 1977), it is probable that those animals which inevitably heat up due to high-rate continuous activity have evolved their biochemical machinery to operate near the upper temperatures experienced.
INSECT FLIGHT METABOLISM
197
The enzymes can evolve to operate at high rates either at high or low temperatures, but not at both simultaneously (Hochachka and Somero, 1973). If the enzymes in an endothermic insect (one that heats up in flight) are adapted to function at a relatively low temperature, then the animal would “overheat” in perhaps seconds or minutes of continuous flight, and endurance would be severely limited. What good is insulation? The modified scales of moths (Heinrich, 1970; Casey, 1976a)and the pile of bumblebees (Church, 1960)effectively retard the rate of heat loss and help to maintain a high elevated thoracic temperature. However, those insects which are heavily insulated are also the same ones that have superb mechanisms of active heat dissipation. The insulation allows the insects to fly at low air temperatures where they might otherwise be unable to be active, but the heat-loss mechanisms allows them to be also active at high temperatures, despite the insulation. Small insects would appear to require the most insulation in order to keep warm from their endogenously produced heat. However, for the most part they have essentially no insulation. A layer of insulation presumably increases air resistance in flight, and the relative increase in the energetic cost of flight should be greatest in small-bodied insects. For example, a bumblebee with a thoracic diameter of 6 mm can generate a 100 per cent increase in temperature excess by means of a 1 mm layer of insulation (Church, 1960). Flying at an ambient temperature of 5 O C it might thereby achieve the normal flight thoracic temperature of 3 5 O C, rather than 20° C, a thoracic temperature that is insufficient for flight. If the increase in resistance in moving through the air is proportional to thoracic surface area, then air resistance should be increased by only 36 per cent. The same thickness of insulation in a small insect having a 1 mm thoracic diameter would cost it 125 per cent more air resistance, but the 100 per cent greater temperature excess generated would raise the body temperature only l o C. In order to increase its body temperature to 3 5 O C at an ambient temperature of 5 O C, and yet maintain the same metabolic rate, this small insect would have to increase its air resistance 227 times. Alternately, if it were to remain uninsulated and maintain the same thoracic temperature as a bee, it would have to increase its metabolic rate by 227 times. It appears, therefore, that energetic costs of flight would increase exhorbitantly for only minor potential increases of thoracic temperature in small insects. Flying insects must be large before they can feasibly employ insulation. Small size consigns insects to adapt biochemically to operate at temperatures close to those of the environment. 6
Development and senescence
The metabolic specializations required for flight can be illuminated by considering their development. A pupa is relatively inactive, but the adult
198
ANN E. KAMMER AND BERND HEINRICH
insect will have a metabolic rate one or two orders of magnitude greater when it begins to fly. Development of flight capability entails changes in cuticle form and mechanical properties, in muscle structure and biochemistry, and in neural geometry and connectivity. There is a considerable literature dealing with the development of insect motor systems (reviews: Nuesch, 1968; Bentley, 1973; Finlayson, 1975). There is also an extensive literature on senescence, because flight muscles, particularly of the short-lived, easily cultured higher Dipfera,have been used as a model system for studies on ageing (Rockstein and Miguel, 1973; Baker, 1976; Sohal, 1976). In the following sections we draw on this literature to discuss the development and loss of flight capability. Development of the flight neuromuscular system is a gradual process. Some aspects begin in the embryo. Near the time of the imaginal moult growth rates accelerate as the biochemical apparatus for producing power at high metabolic rates is constructed. Development may continue in the adult, after which the properties of the flight muscles remain relatively stable for a time. Finally, the flying ability of the insect declines. The rates at which maturation and senescence proceed may be genetically determined, or influenced by environmental factors such as temperature or diet, or governed by some combination of these plus factors yet to be discovered (Rockstein and Miguel, 1973). The general features of the development of the flight motor system are similar in various insects, but for convenience we shall consider development in 3 groups; (1) hemimetabolous insects, (2) holometabolous insects with a neurogenic flight rhythm, and (3) holometabolous insects with a myogenic flight rhythm. Next we shall consider some specialized developmental sequences and then the effects of use and disuse. Hormonal control mechanisms will be briefly discussed. Finally flight metabolism will be examined in relation to these changes with age. 6.1
HEMIMETABOLOUS I N S E C T S
In insects that undergo a gradual metamorphosis, the basic neural network and the precursors of the flight muscles are laid down early in development, probably during embryogenesis. For example, motor neurons supplying the pterothoracic dorsal longitudinal muscles have been seen in cobalt-filled preparations of the third instar of a cricket, Teleogryllus oceanicus (Bentley, 1973). This muscle and the nerve supplying it can also be discerned in freshly hatched nymphs, suggesting that both arise embryologically. This suggestion is supported by recent work in which neurons have been stained and identified during embryogenesis (Bentley, pers. comm.). In another cricket, Achaefa domesficus the dorsal longitudinal muscles are present in embryos (Voss, 19 11,
199
INSECT FLIGHT METABOLISM
1912). In Schistocerca these muscles are used in hatching, but they have no function in nymphs and persist as rudiments that later develop into the major wing depressor of adults (Thomas, 1954; Bernays, 1972). In the Australian locust Chortoicetes terminifra precursors of all the adult muscles are present in first instar nymphs (Tiegs, 1955). In the hemipteran Oncopeltusfasciatus the dorsal longitudinal muscles are present in the first instar, disappear early in the
Calhphora
p
GPCYCLE,
Locusto
Philosamia Cynthia
migratora
L
I
4lo2
-
I
GP-CVCLE
TCA-CYCLE
P ‘*-a LDH
I
Days after pupariation/pupal ecdyris
I
I
t
l
l
l
l
I
I
I
I
2 5 8 2 4 6 8 4 7 10 12 17 3 5 Days after fifth larval/larval- Day after larval-pupal/pupaladult ecdysis adult ecdysis
Fig. 21. Development of the main catabolic pathways in flight muscles of 3 insect species. The activities of the enzymes at the indicated developmental days are related to the activity of the respective enzyme on the day of maximal lactate dehydrogenase (LDH) activity. On the left the enzymes measured are given, on the right the pathway they represent. Arrows indicate the day of adult emergence. HOAD, 3-hydroxyacyl-CoA dehydrogenase; other abbreviations as in Figs 10, 12, 14 (from Beenakkers et al., 1975).
second instar, and redevelop in later instars from an aggregation of myoblasts (Scudder and Hewson, 1971). Some muscles used in both walking and flying (Wilson, 1962) develop gradually as the nymphs grow (Tiegs, 1955; Altman, 1975). Other muscles, however, develop only late in the postembryonic period (Teutsch-Felber, 1970; Bernays, 1972). Structural and biochemical differentiation is sharply accelerated during the several days before and after imaginal ecdysis in locusts (Brosemer et af., 1963; Vogell, 1965; Walker et af., 1970; Chari and Hajek, 1971; Beenakkers
200
ANN E KAMMER AND BERND HEINRICH
et al., 1975) and in crickets (Chudakova and Bocharova-Messner, 1965; Bocharova-Messner and Yanchuk, 1966). Substantial increases occur in the amount of contractile protein, the size of mitochondria, and the activity of enzymes important in flight metabolism. For example (Fig. 21) in Locusta migratoria the flight muscles grow exponentially in weight and in enzyme content during the week before the terminal ecdysis, and growth continues during the following week (Brosemer et al., 1963; Beenakkers et al., 1975). The final enzyme patterns stabilize at about Day 6 after adult emergence (Beenakkers et al., 1975), which parallels the differentiation of the flight muscles (Brosemer et al., 1963). During the first 10-14 days of adult life the weight of the cuticle also increases (Hill et al., 1968). Parallel to these morphological and biochemical improvements in the flight machinery is a gradual development of the neural elements controlling flight. Branches of motor neurons and endings of sensory neurons grow as the ganglionic neuropile enlarges in successive instars and more fine branches become accessible to cobalt staining (Bentley, 1973; Altman and Tyrer, 1974; Tyrer and Altman, 1974). Motor activity resembling that of adults can be produced by nymphs. For example, Teleogtyllus commodus nymphs in the seventh instar, which must moult four times before adulthood, assume the flight posture for short periods when suspended in a wind tunnel (Bentley and Hoy, 1970). These and later nymphal instars produce patterned muscle activity that resembles in some respects flight motor patterns produced by adults. In nymphs the cycle-time is longer and phase relationships are less precise than in adults, and patterns improve with age. Similar results have been obtained with Gryllus campestris (Weber, 1972), T. oceanicus (Bentley, 1973), Schistocerca gregaria (Kutsch, 197 1, 1974a), and Chortoicetes terminifera (Altman and Tyrer, 1974; Altman, 1975), with differences in the ages at which adult phase relationships and cycle-times are established. Development of motor patterns continues after the imaginal ecdysis in the locusts. Wingbeat frequency is low at first, rises rapidly during the first few days of adult life and then rises more slowly during the next 2+3 weeks (Kutsch, 1973; Altman, 1975). Similar exponential increases of wingbeat frequency with age have been reported in Periplaneta (Farnworth, 1972a) and various Diptera (see below). The increase in wingbeat frequency has been attributed to development of the central rhythm generator (Kutsch, 1973, 1974b), but Altman (1975) has provided evidence that it depends on sensory input from the beating wings. 6.2
“LEPIDOPTERA”
In the Lepidoptera as in other holometabolous insects the change in body form and musculature during development from a larva to an adult is dramatic. It is possible to distinguish five types of muscles according to their fate during
INSECT FLIGHT METABOLISM
20 1
metamorphosis (Snodgrass, 1954): (1) larval muscles that go unchanged into the adult, (2) larval muscles that are reconstructed into adult muscles, (3) larval muscles that degenerate completely, (4) muscles that are newly formed for the imago and that replace degenerated larval muscles, and ( 5 ) newly formed muscles that have no distinguishable precursors. The most thoroughly studied flight muscle, the mesothoracic dorsal longitudinal muscle of large moths (Antheraea pernyi, A . polyphemus and Manduca sexta), belongs to category (2). Early in the pupal stage the precursor larval muscle, which is located in the dorsal region of the mesothorax (Heinertz, 1976), degenerates to narrow strands of fibers lacking myofilaments (Eigenmann, 1965; Bienz-Isler, 1968b). This anlage is subsequently joined by myoblasts of unknown origin. The myoblasts fuse with the anlage (Stocker, 1974) and the fibers grow in size and increase in number (Basler, 1969). Cross-striations and contractile protein appear in the middle of the 3-week pupal period (Eigenmann, 1965; Basler, 1969). Growth, particularly of number of filaments and mitochondria, is rapid during the last week of the pupal (or pharate moth) stage (Bienz-Isler, 1968b). The nervous system also undergoes considerable restructuring during the transformation of a caterpillar into a moth. The interganglionic connectives shorten (Pipa, 1963, 1967, 1973; Heywood, 1965; Tung and Pipa, 1972) and additional neurons, especially those associated with adult receptors, are added (Edwards, 1969; Edwards and Palka, 1976; Sanes and Hildebrand, 1976a,b; Kammer and Athey, unpubl.). However, many motor neurons persist (Panov, 1963; Taylor and Truman, 1974; Truman and Reiss, 1976) including those supplying the dorsal longitudinal muscles of the audlt mesothorax (Casaday, 1975; Rheuben, pers. comm.). These motor neurons retain structural connections with the muscle fibers during development. Adult neuromuscular junctions begin to form as larval neuromuscular junctions are reduced, and more axonal branches and junctions form as the muscle grows (Kaufmann, 1971; Stocker and Nuesch, 1975; Heinertz, 1976). This innervation is necessary for normal development (Niiesch, 1968; Basler, 1969; Nuesch and Bienz-Isler, 1972). The adult flight muscles supplied by these motor neurons are structurally (Bienz-Isler, 1968a; Rheuben, 1975; Fig. 8) and mechanically phasic muscles, whereas the larval muscles supplied by the same motor neurons are tonic (Rheuben, 1975; Kammer and Rheuben, 1976a; Rheuben and Kammer, in prep.). Adult motor patterns develop gradually and are spontaneously produced by a pharate moth during the 2-4 days before eclosion (Kammer and Rheuben, 1976a,b; Fig. 22). The motor patterns produced by pharate moths, both A . polyphemus and Manduca sexta, have a longer and more variable cycle time than adult motor patterns (Kammer and Rheuben, 1976b), and in M . sexta the pattern improves with age (Kammer and Kinnamon, in prep.).
ANN E. KAM,MER AND BERND HEINRICH
202
B
C
Rdl
200 ms
INSECT FLIGHT METABOLISM
203
Along with the morphological and physiological changes taking place in the pharate moth, there are biochemical changes, as in the hemimetabolous insects (Fig. 21, Beenakkers et al., 1975). However, there is little change after eclosion compared to the continued increases in locusts. In a silkmoth, Philosamia Cynthia, the final enzyme pattern is reached shortly after eclosion (Beenakkers et al., 1975). In a hawkmoth Manduca sexta trehalase activity in the thoracic muscles doubled on the first (males) or second (females) day after eclosion, compared with activity present on the day of eclosion, but there were no further increases (Dahlman, 1972). In another moth, Heliothis virescens, no post-eclosion increase in mitochondria1 enzyme activities was observed, and these moths can fly for extended periods shortly after eclosion (Holmes and Keeley, 1975). In Pieris butterflies flight capability increased in successive trials, probably because of hardening of the cuticle (Peterson et al., 1957).
6.3
HOLOMETABOLOUS INSE CT S WITH FIBRILLAR MUSCLES
In Diptera and Hymenoptera the developmental changes in the pupa and pharate adult are similar to those in the Lepidopfera.Some muscles arise early in the larva. In the dipteran Simulium ornatum, anlagen of the adult dorsal longitudinal muscle could be seen in the first instar larva, and other flight muscles were observed in the second instar (Hinton, 1961). In Apis mellifera, one of the few cases in which all of the muscle types have been examined in one species, 39 per cent of the thoracic muscles of the adult bee are derived in some way from larval muscles and 6 1 per cent are formed from aggregations of myoblasts without any association with a larval precursor; no larval muscles persist unchanged in the adult (Daly, 1964). The origin of the myoblasts is in most cases not known-(Finlayson, 1975). In Drosophila (Shatoury, 1956) and Calliphora (Crossley, 1965) the myoblasts are thought to be derived from imaginal discs. In the case of Drosophila leg muscles, precursors of the muscles are present in the leg discs of late third instar larvae, but it is not known whether they originate from the disc itself or migrate into the developing disc (Gehring, 1976). According to Tiegs (1955), the flight muscles arise out of free myoblasts that in older larvae adhere to the certain larval muscle fibers. These myoblasts multiply in the young pupa as the larval muscle fibers degenerate. According to Shatoury (1956) the myoblasts are derived from two imaginal discs, one that forms the wing and the other that forms the mesothoracic leg. Fig. 22. Development of motor patterns in the moth Anrheraea polyphemus. (A) Muscle potentials recorded during flight in adults (dl, dorsal longitudinal wing depressor; rc, wing elevator muscle). (B) Warm-up in adult, muscle potentials as in A. (C) Flight motor pattern from pharate moth 3 days before ecdysis. (D) Warm-up motor pattern from same pharate moth (from Kammer and Rheuben, 1976b).
2 04
ANN E. KAMMER AND BERND HEINRICH
Much of myoblast differentiation is genetically programmed, because cells isolated from early embryos and allowed to differentiate into myotubes in culture followed the same timetable as myoblasts in vivo (Seecof et al., 1973). Myoblast fusion with flight muscle rudiments has been observed in electron micrographs of Lucilia (Peristianis and Gregory, 1971) and Ips (Bhakthan et af., 1970). One of the most thorough studies of the role of myoblasts is that of Crossley (1972a,b) on a intersegmental (nonflight) muscle of Calliphora. He showed that the RNA synthesis of the second set of nuclei continues while that of the original larval muscle nuclei declines as development proceeds, suggesting that nuclei derived from myoblasts control the synthesis of adult muscle proteins. During the development of the fibrillar flight muscles in the cyclorrhaphous Diptera, the remnants of larval muscles that form the anlagen of the adult muscles split longitudinally and increase in both diameter and length (Shatoury, 1956; Beinbrech, 1968; Auber, 1969; Peristianis and Gregory, 1971). As in the flight muscles of other insects there is marked growth in myofibrillar content (described in detail for Calliphora by Auber, 1969) and in mitochondria size and number. The period before and after eclosion is one of active protein synthesis (Campbell and Birt, 1972). Actomyosin synthesis was substantial on the first day after eclosion but then ceased, an unexpected finding in the light of electron-microscopical evidence that suggests substantial formation of myofilaments after Day 1 (Campbell and Birt, 1975). Enzymes of carbohydrate metabolism increase 5- to 100-fold in the pharate and 1-day-old adult fly (Beenakkers et al., 1975; Fig. 21). Maximum a-glycerophosphate dehydrogenase activity was reached in the 4-day-old mature fly (Campbell and Birt, 1972). The extent to which enzyme levels at emergence are less than the maximum in the mature adult can be correlated with the length of time before the insect begins to fly (Campbell and Birt, 1972; Rockstein and Miguel, 1973). Particular attention has been paid to developmental changes after flight begins. The flight performance of recently emerged flies, measured in terms of wingbeat frequency and duration of flight is low and gradually improves during the first several days. During this time there are changes in the flight muscles; the mitochondria increase in size and in density of cristae (Johnson and Rowley, 1972). The activity of mitochondria1 enzymes and of cytoplasmic enzymes also increases (Rockstein and Miguel, 1973; Baker, 1976). Usually there is a mature period during which flight ability is relatively unchanged. Then with advancing age, flight performance, particularly duration of flight, declines (Baker, 1976). Striking changes are seen in the post-eclosion development of tsetse fly flight muscles. After the insect has fed several times on blood, there is an increase in muscle mass (Bursell, 1961, 1973) and cytochrome-c content (Bursell, 1973), a
INSECT FLIGHT METABOLISM
205
two-fold increase in mitochondrial protein (Hargrove, 1975a), a 40 per cent increase in contractile protein (Hargrove, 1975a), a doubling of the diameter of the myofibrils (Anderson and Finlayson, 1973), and a 50-70 per cent increase in the dry weight of the cuticle (Hargrove, 1975a); As a result of this posteclosion growth, the muscle fibers occupy more of the thoracic volume and become densely packed with mitochondria and myofibrils. These morphological changes correlate with the improved flying abilities of mature Glossina. As the fly develops after 1 or more blood meals, the maximum wingbeat frequency increases, the lift produced doubles, and the numbers of wingbeats performed approximately doubles (Hargrove, 1975b). The lift, measured by the amount of added lead weight a fly could carry in free flight, is 4-5 times the body weight (Hargrove, 1975b). Improved flight capability in mature flies may also be due to the increased reserves of proline in the thorax (Hargrove, 1975a,b). Towards the end of the life span, various degenerative changes appear, including structural alterations of mitochondria, loss of myofibrils and decline in enzyme activity (Rockstein and Miguel, 1973; Baker, 1976). These changes do not occur in parallel in any one species, nor are they always present. For example Tribe and Ashhurst (1972) reported there were no degenerative changes in the flight muscle of Calliphora erythrocephala as old as 8&95 days, although the flight ability of flies of this age had declined and the mean survival time was 55 days. Causal relationships among the various functional, morphological, and biochemical changes are difficult to establish, in part because of temporal discrepancies. For example, in male Musca domestica, duration of fight was maximal at 1 day after eclosion, and wingbeat frequency was maximal at 4 days after eclosion and remained high for about 9 days (Rockstein and Bhatnagar, 1966). On Day 1 the activity of the two enzymes measured was still low, about 25 per cent of maximum. One enzyme, APK (arginine phosphokinase, i.e., ATP :1-arginine phosphotransferase) had reached its maximum activity on Day 2 and declined by Day 4 (Baker, 1975b). The other enzyme, extra-mitochondrial a-glycerophosphate dehydrogenase, was maximally active on Day 4 and declined thereafter (Rockstein and Brandt, 1963). This sequence of enzyme changes is not paralleled by changes in the mitochondria. Instead the mitochondria increase in size and number during the first 8 days (Rockstein and Bhatnager, 1965). In newly emerged flies the mitochondria are dispersed between the myofibrils and surrounded by cytoplasm containing large amounts of glycogen; in older flies the mitochondria and myofibrils are tightly packed (Simon et al., 1969). Similar temporal discrepancies are apparent in the maturation of male Phormia regina. Wingbeat frequency was twice as high on the second day after eclosion as on the first day; wingbeat frequencies were even higher on Day 5 and remained so until Day 47 (Baker,
206
ANN E. KAMMER AND BERND HEINRICH
1976). Mitochondrial mass and cytochrome-c content increased during the first week (Levenbook and Williams, 1956). The activity of arginine phosphokinase increased steeply up to Day 4 but then declined, remaining approximately constant from Day 9 to Day 40 (Baker, 1975a). The activity of a-glycerophosphate dehydrogenase showed a similar sequence of changes but with a maximum at Day 9 (Baker, 1975a). The time to 50 per cent mortality was 3 1 days (Baker, 1975a). Again, there is no clear correlation between flight ability, enzyme changes, and mortality. The increase in wingbeat frequency from the first to second day could be due to increased energy supply from developing mitochondria but otherwise changes in flight ability, both duration and wingbeat frequency, may depend more on changes in the hormonal and nervous systems than on changes in the flight muscles. One common change associated with aging and decline in flight ability is depletion of glycogen reserves or loss of the ability to utilize glycogen or both (Williams et al., 1943; Rowley and Graham, 1968; Nayar and Sauerman, 1973; several other studies, see Baker, 1976). Hormonal mechanisms could be involved in this change since glycogen mobilization and perhaps storage is under hormonal control (section 4).
6.4
ADULT D I A P A U S E , REGENERATION AND POLYMORPHISM
More complicated developmental strategies than the linear and uninterrupted sequence of maturation, flight and senescence, described above, are employed by other insects. These strategies are exemplified by life cycles during which the flight muscles regress and then regenerate, and by polymorphic populations, in which the flight apparatus is developed to different degrees in different individuals. In the latter cases, the flying forms are often migratory, and flight is followed by concurrent histolysis of the flight muscles and oocyte development (Johnson, 1969, 1976; Dingle, 1974). Post-eclosion maturation in the beetle Leptinotarsa decemlineata can be interrupted by an adult diapause or can continue in nondiapausing adults (deKort, 1969; deKort and Bartelink, 1972). In nondiapausing adults emerging from the soil (2 days after eclosion), myofibrils are small, 0.3-0.4pm in diameter. Three days later fibril diameter is about 1.5 pm, increasing to 2.4 pm after 8 days, when growth as seen with the electron microscope is complete. The mitochondria also increase in number and size. Mitochondrial enzymes and cytochomes are synthesized at a rapid rate during the first four days after emergence from the soil (Bartelink et al., 1975). Enzyme activity per mg protein increases 4-7-fold, and the protein content per thorax more than doubles. Enzyme development is complete in about 12 days. Animals raised on a photoperiod that induces diapause eclose with muscles that resemble those of nondiapausing forms. The post-eclosion growth of fibers, myofibrils, and mito-
INSECT FLIGHT METABOLISM
207
chondria begins but is not completed; instead the muscles degenerate and in a diapausing insect are structurally and biochemically the same as the muscles of a recently eclosed adult. After diapause is terminated development of the flight muscle starts immediately and follows a time course similar to that observed after emergence of nondiapausing adults (deKort, 1969). More extreme changes occur in scolytid beetles (Finlayson, 1975). Functional muscles may regress and later develop again to full size (Bhakthan et al., 1970, 197 1). In the bark beetle Ips confusus, ihe muscles mature and the insects emerge to attack a new log. Next the muscles degenerate rapidly as the adults become reproductive, and then the muscles redevelop and the insects reemerge (Borden and Slater, 1969). Polymorphism in aphids has been much studied with respect to the factors that influence the production of winged or wingless forms. Various external factors may induce the appearance of alates, including lack of food or water, low temperature, and presence of other individuals in a small area (Lees, 1967; Sutherland, 1969; Shaw, 1970; Judge and Schaefers, 1971; Schaefers and Judge, 1971). Winged forms may be migratory, may make only short flights, or may reproduce without flying (Kring, 1972). In several aphid species, muscle histolysis precedes or accompanies development of embryos (Johnson, 1957). In the Gerridae, there are seasonal, latitudinal, populational, and species differences in the time of occurrence and the proportions of long-winged and short-winged adults (Anderson, 1973; Darnhofer-Demar, 1973; Vepsalainen, 1974a,b). Not all macropterous forms have functional flight muscles (Darnhofer-Demar, 1969). During the teneral period of adults that will become able to fly (1 1 days in Gerris lacustris), the diameter of flight muscle fibers increases 4-fold and the cuticle grows (Andersen, 1973). Later when the females are producing mature eggs, the flight muscles undergo histolysis (Darnhofer-Demar, 1969; Andersen, 1973). In another group of aquatic insects, the Corixidae, morphs have similarly developed wings but differ in muscle development (Young, 1965a,b). In forms capable of flight, the muscle fibers increase in diameter from lOpm after the imaginal ecdysis to 40-60pm at the end of the teneral period (1-4 weeks depending on temperature). Higher temperatures are an important factor inducing development of flight muscles. Another group in which reproduction and flight are incompatible are the pyrrhocorid bugs of the genus Dysdercus (Edwards, 1969a, 1969b; Dingle and Arora, 1973; Davis, 1975). If newly eclosed adults encounter a lack of food, the ovaries do not mature, the flight muscles develop, and the insects fly readily. At the onset of egg production the flight muscles rapidly degenerate. In milkweed bugs (Oncopelhts) there is a behavioral polymorphism: a difference in flight duration although wings and flight muscles are equally well developed (Dingle, 1974).
208
6.5
ANN E KAMMER A N D BERND HEINRICH MATURATION R E L A T E D TO USE AND DISU SE
A question related to the unknown mechanisms controlling maturation and aging concerns the influence of use and disuse on these processes. The available evidence provides no clear answers. Rockstein cut off the wings of newly emerged M. domestica, presumably preventing the flies from using their flight muscles, and followed changes in muscle enzymes (Rockstein and Miguel, 1973). In both operated and intact flies the activity of Mg2+-activated ATPase rose during the first 6 days after eclosion and then declined. The activity of a-glycerophosphate dehydrogenase, however, rose normally during the first 3 - 4 days but then enzyme activity remained high during the next two weeks instead of declining as in intact flies. In a similar experiment on Drosophila melanogaster Sohal (1975) found that mitochondria1 fission, which normally occurs during aging in these flies, is more extensive in flies from which the wings had been removed. Baker ( 1 9 7 5 ~ ) found that in vestigal-winged mutants of D. melanogaster the pattern of changes in APK activity was the same as in wild-type flies, suggesting that flightlessness did not affect this enzyme. In a study involving light- and electron-microscopical examinations of several flightless mutants of D. melanogaster, Deak (1976) also found no indication of muscle degeneration as a result of inactivity. In most of the mutants the muscles appeared to be normal at eclosion and no deterioration was observed during the life of the flies. Shortterm paralysis (up to 9 days) obtained by using temperature-sensitive mutants also produced no gross abnormality. Smaller changes such as the differences in the size and number of mitochondria would not have been seen in this study. The results of the experiments on these flies, which eclose with their muscles well-developed, are conflicting: flightlessness either had no apparent effect, postponed the decline in activity of an enzyme, or accelerated an agingcorrelated fission of mitochondria. In tsetse flies (Glossina), which eclose with markedly immature flight muscles (section 6.3), flying apparently enhances the growth of the muscles. Laboratory raised tsetse flies develop flight muscle more slowly than wild flies (Bursell, 1961). When captive flies are made to fly, the muscles grow more rapidly (Bursell and Kuwengwa, 1972). Other changes involving use of the flight system may depend on development of neural control mechanisms instead of or in addition to maturation of the flight muscles. Motor patterns resembling flight are produced by stages too young to fly, e.g. by grasshopper and cricket nymphs and by pharate moths (discussed above), but the functional significance of this activity is not known. In Glyllus campestris nymphs the flight pattern of potentials from thoracic muscles become more stable and more like that of the adult with successive short flights (Weber, 1972), suggesting that practice improves the
INSECT FLIGHT METABOLISM
209
neural pattern generator. On the other hand, adult locusts with their wings waxed shut and thus prevented from making flight movements showed the same increase in wingbeat frequency during the first three weeks after the imaginal ecdysis as do unrestrained locusts (Altman, 1975). In this case use of the flight mechanisms was not necessary for maturation. These suggestions that use of disuse can influence maturation and senescence are perhaps more puzzling than informative. Further research on the causal mechanisms involved is necessary. Such research may elucidate important mechanisms controlling the rate of aging. 6.6
HORMONAL CONTROL
Development of the flight muscles, like that of other adult structures, is under hormonal control, but the factors that determine which larval muscles degenerate and which grow are not known. Furthermore, there is no simple correlation between the balance or amounts of ecdysone and juvenile hormone (JH or neotenin) and muscle development (Finlayson, 1975). For example, in Leptinotarsa decemlineata lack of JH induces an adult diapause, and both the reproductive system and the flight muscles become reduced (deWilde and deBoer, 1961). In the bug Dysdercus intermedius implantation of corpora allata and corpora cardiaca into starving females (which untreated retain their flight muscles) induces egg production but muscle degeneration (Edwards, 1970; Davis, 1975). Similarly in the bark beetle Dendroctonus pseudotsugae, flight muscles degenerate and acid phosphatase activity in the muscles increases as the ovaries develop; topical application of JH increased acid phosphatase activity (Sahota, 1975). In Oncopeltus fasciatus on the other hand, application of synthetic juvenile hormone or implantation *of corpora allata increased the proportion of males and females making long flights rather than reproducing (Caldwell and Rankin, 1972; Rankin, 1974). Similarly in locusts hormones from the corpora allata appear to facilitate the development of migratory flight. Allatectomized adult Locusfa flew less than shamoperated controls (Wajc and Penner, 1971). The effect, however, is agedependent (Lee and Goldsworthy, 1975). Normal animals exhibit their top speed on Day 18 after the imaginal moult, but allatectomized animals perform poorly at this time. Older animals that had been allatectomized on Day 3 flew faster on Days 25 and 40 than did controls of the same age. Thus, allatectomy retards the development of normal flight capability, but it also slows down the decline of flight performance that is characteristic of aging (Lee and Goldsworthy, 1975). The cerebral neurosecretory cells and corpora cardiaca have also been implicated in the maintenance of long-term flight in Schistocerca (Michel, 1972, 1973b; Michel and Bernard, 1973), but the nature
210
ANN E. KAMMER AND BERND HEINRICH
of the hormone(s) is not clear (Goldsworthy, 1976). There are other examples of the influence of hormones on flight capability and “flight tendency”, but a complete review is beyond our scope. Hormones from the corpora allata and perhaps elsewhere can be viewed as part of long-term control over the intensity of flight metabolism and energy expended by an individual, in contrast to the hormonal (principally from the corpora cardiaca), biochemical, and neural mechanisms discussed in earlier sections. The latter provide primarily short-term controls over individual flights.
6.7
AGE A N D FLIGHT METABOLISM
The observations summarized above illustrate changes in flight metabolism in relation to the developmental stage of an insect. Elements of the flight behavior develop before the ecdysis to an adult in both Orthoptera and Lepidoptera. In the former order no measurements of oxygen consumption have been made to determine if this immature flight activity is correlated with a higher metabolic rate. In the pupa of holometabolous insects oxygen consumption rises before eclosion but it is much less than the metabolic rate of flying adults (Burkett, 1962; Agrell and Lundquist, 1973). Furthermore, the body temperature of a pharate moth (Munducu sexta) does not rise during the production of flight motor patterns, suggesting that the metabolic rate is considerably less than that of adults (Kammer, unpub.). The post-eclosion developmental changes in mitochondria and protein clearly suggest changes in flight metabolism during maturation. Little direct evidence, based on measurements of oxygen consumption, is available to assess the magnitude of the differences. The suggestion that age-related changes in flight metabolism occur is supported, however, by the observed changes in wingbeat frequency in locusts and flies (section 6.1 and 6.3). These temporal changes along with polymorphism within a population add complexity to problems of measuring energy expenditures of a species. Along with the variation in metabolic rate that is under neural control (section 3) they render suspect the presentation of a single value for the flight metabolism of a species, as is often done (Table 1). In addition to differences in flight intensity, the duration of sustained flight varies among the individuals of a population or species (Johnson, 1976). These individual differences, which may be related to age, food supply, hormone levels, or environmental conditions, are particularly important in relation to migration and adaptive change in the dispersal of a species (Johnson, 1976).
INSECT FLIGHT METABOLISM
7
21 1
Conclusions
Control of flight metabolism is accomplished by a complicated meshwork of neural, hormonal, and biochemical mechanisms. Control of the rate of metabolism is primarily neural, since the frequency of excitation of the flight muscles provides the primary control on the demand for metabolic performance. The metabolic machinery of the muscle is also influenced by these neural signals, since Ca++ released by excitation modulates enzyme activity. Muscle metabolism is also influenced by hormones that control the uptake of substrates and perhaps other biochemical reactions in the muscles. The contractile machinery is provided with fuels supplied in part by intramuscular supplies and in part as a result of the action of hormones stimulating mobilization of fuels from storage depots. When hormones intervene as messages in the control system, their release may depend on neural signals, although this has not been clearly established. Although the nervous system can thus be viewed as controlling the demand for fuel and activating the mechanisms of supply, the nervous system in turn may be subject to hormonal influences. It is probable that the hormones controlling hemolymph lipid and sugar content or other hormones such as juvenile hormone and ecdysone influence neuronal excitability or increase the duration of flight. Relatively little is as yet known about this problem, including the mechanism of action of hormones on the central nervous system or hormonal influences on maturation and aging. Insect flight can be examined at many levels from ecological and evolutionary to molecular. In this review we have ranged widely, in an attempt to give an integrated picture of these different aspects. In recent years rapid progress has been made in understanding the aerodynamics of insect flight, the hormonal mechanisms controlling carbohydrate and lipid metabolism, and the mechanisms of temperature regulation before and during flight. In addition, increasing attention is being given to the development of the flight neuromuscular system. Additional progress in these areas can be expected. The results will further illuminate the specializations required to support the high metabolic rates characteristic of insect flight.
Acknowledgements
We thank Dr Karl Kramer, U.S. Grain Marketing Research Laboratory, Manhattan, Kansas, for a critical reading of portions of the manuscript and ,informative discussion, and Dr Mary B. Rheuben, Department of Biology, The Pennsylvania State University, for the electron micrographs and for helpful
212
ANN E. KAMMER AND BERND HEINRICH
discussion of developmental aspects. Unpublished research by the authors was supported by N.S.F. Grants No. BNS75-18569 (to A.E.K.) and BMS-7418897 (to B.H.).
References
Abbott, R. H. (1973). The effects of fibre length and calcium ion concentration on the dynamic response of glycerol extracted insect fibrillar muscle. J. Physiol. 231, 195-208. Agrell, I. P. S. and Lundquist, A. M. (1973). Physiological and biochemical changes during insect development. In “The Physiology of Insecta” (Ed. M. Rockstein) 2nd Edn., Vol. 1, pp. 159-247. Academic Press, New York. Altman, J. S. (1975). Changes in the flight motor pattern during the development of the Australian plague locust, Chortoicetes terminifera. J. comp. Physiol. 97, 127-142. Altman, J. S. and Tyrer, N. M. (1974). Insect flight as a system for the study of the development of neuronal connections. In “Experimental Analysis of Insect Behaviour” (Ed. L. Barton Brown), pp. 159-1 79. Springer-Verlag, Heidelberg and New York. Andersen, N. M. (1973). Seasonal polymorphism and developmental changes in organs of flight and reproduction in bivoltine pondskaters (Hem. Gerridae). Ent. Scand. 4, 1-20. Anderson, M. and Finlayson, L. H. (1973). Ultrastructural changes during growth of the flight muscles in the adult tsetse fly, Glossina austeni. J. Insect Physiol. 19, 1989-1997. Armitage, P., Miller, A., Rodger, C. D. and Tregear, R. T. (1973). The structure and function of insect muscle. Cold Spring Harbor Symp. Quant. Biol. 37,379-387. Auber, J. (1967). Distribution of the two kinds of myofilaments in insect muscles. Amer. Zool. 7, 451-456. Auber, J. (1969). La myofibrillogentse du muscle strib. I.-Insectes. J. Microscopie, 8,197-232. Bailey, E. (1975). Biochemistry of insect flight. Part 2. Fuel Supply. In “Insect Biochemistry and Function” (Eds D. J. Candy and B. A. Kilby), pp. 89-176. Halsted Press, Wiley, New York. Baker, G. T. (1975a). Identical age-related patterns of enzyme activity changes in Phormia regina and Drosophila melanogaster. Exp. Gerontol. 10,23 1-237. Baker, G. T. (1975b). Age-related activity changes in arginine phosphokinase in the house fly, Musca domestica L. J. Geront. 30, 163-169. Baker, G. T. (1975~).Age-dependent arginine phosphokinase activity changes in male vestigal and wild-type Drosophila melanogaster. Gerontologia, 21,203-2 10. Baker, G. T. (1976). Insect flight muscle: maturation and senescence. Gerontology, 22,234-26 1. Barker, R. J. and Lehner, Y. (1972). Free amino acids in thoraces of flown honey bees, Apis mellifera L. (Hymenoptera: Apidae). Comp. Biochem. Physiol. 43B, 163-169. Bartelink, A. K. M., Gijzel, W. P., deKort, C. A. D. and Kroon, A. M. (1975). Mitochondria1 formation in the developing flight muscles of the Colorado beetle. Devel. Biol. 42,262-273. Bartholomew, G. A. and Epting, R. J. (1975a). Rates of post-flight cooling in sphinx moths. In “Perspectives of Biophysical Ecology” (Eds D. M. Gates and R. B. Schmerl), pp. 405-415. Springer-Verlag, New York and Heidelberg. Bartholomew, G. A. and Epting, R. J. (1975b). AUometry of post-flight cooling rates in moths: a comparison with vertebrate homeotherms. J. Exp. Biol. 63,603-613. Basler, W. (1969). Untersuchungen der Nervenwirkung bei Antheraea pernyi Guer. (Lep.) unter besonderer Beriicksichtigung der dorso-longitudinalen Flugmuskeln. Rev. Suisse Zool. 76, 297-362. Bastian, J. (1972). Neuro-muscular mechanisms controlling a flight maneuver in the honeybee. J. Comp. Physiol. 72, 126-140.
INSECT FLIGHT METABOLISM
213
Bastian, J. and Esch, H. (1970).The nervous control of the indirect flight muscles of the honeybee. Z . Vergl. Physiol. 67,307-324. Beenakkers, A. M. T. (1965). Transport of fatty acids in Locusta migratoria during sustained flight. J. Insecr Physiol. 11, 879-888. Beenakkers, A. M. T. (1969). Carbohydrate and fat as a fuel for insect flight. A comparative study. J. Insect Physiol. 15,353-361. Beenakkers, A. M. T., Van Den Broek, A. T. M. and de Ronde, T. J. A. (1975). Development of catabolic pathways in insect flight muscles: A comparative study. J. Insect Physiol. 21,849-
860. Beinbrech, G. (1 968). Elektronenmikroskopische Untersuchungen iiber die Differenzierung von Insektenmuskeln wiihrend der Metamorphose. 2.Zellforsch. Mikrosk. Anat. 90,463494. Bennett, L. (1970).Insect flight: lift and rate of change of incidence. Science, 167, 177-179. Bennett, L. (1975). Insect aerodynamics near hovering. In “Swimming and Flying in Nature” (Eds T. Y.-F. Wu, C. J. Brokaw and C. Brennan), Vol. 2,pp. 815-828. Plenum, New York. Bentley, D. R. (1973). Postembryonic development of insect motor systems. In “Developmental Neurobiology of Arthopods” (Ed. D. Young), pp. 147-177. Cambridge University Press. Bentley, D. R. and Hoy, R. R. (1970). Postembryonic development of adult motor patterns in crickets: a neural analysis. Science, 170, 1409-141 1. Bernays, E. A. (1972). The muscles of newly hatched Schistocerca gregaria larvae and their possible functions in hatching, digging, and ecclysial movements (Insecta: Acrididae.). J. ZOO^. 166,141-158.
Beutler, R. (1936).h e r den Blutzucker der Bienen. Z . uergl. Physiol. 24,71-I 15. Bhakthan, N. M. G., Borden, J. H. and Nair, K. K. (1970). Fine structure of degenerating and regenerating flight muscles in a bark beetle, Ips confusus. I. Degeneration. J. Cell Sci. 6,807-
819. Bhakthan, N. M. G., Nair, K. K. and Borden, J. H. (1971). Fine structure of degenerating and regenerating flight muscles in a bark beetle, Ips confusus. 11. Regeneration. Can. J. Zool. 49,
85-89. Bienz-Isler, G. (1968a).Elektronenmikroskopische Untersuchungen iiber die imaginale Struktur der dorsolongitudinalen Flugmuskeln von Antheraea pernyi Guer. (Lepidoptera). Acta Anat.
70,416-433. Bienz-Isler, G. (1968b). Elektronenmikroskopische Untersuchungen iiber die Entwicklung der dorsolongitudinalen Flugmuskeln von Antheraea pernyi Guer. (Lepidoptera) 11: Ted. Acta Anat. 70,524-553. Bloxham, D. P., Clark, M. G., Holland, P. C. and Lardy, H. A. (1973). A model study of the fructose diphosphase-phosphofructokinasesubstrate cycle. Biochem. J. 134,581-587. Bocharova-Messner, 0. M. and Yanchuk, K. (1966).Ontogenetic changes in the ultrastructure of the wing muscles of Acheta domesticus. Doklady Akad. Nauk SSSR, 170,948-95 1. Boettiger, E. G. (1955).Triggering the contractile process in insect fibrillar muscle. J. cell. comp. physiol. 46,370-371. Boettiger, E. G. (1957). Triggering of the contractile process in insect fibrillar muscle. In “Physiological triggers” (Ed. T. H. Bullock), pp. 103-1 16. American Physiological Society, W ashmgton. Borden, J. H. and Slater, C. E. (1969). Flight muscle volume change in Ips confusus (Coleoptera: Scolytidae). Can. J. Zool. 47,29-32. Brocher, F. (1920). etude expkrimentale sur le fonctionnement du vaisseau dorsal et sur la circulation du sang chez les insectes. IIIe Partie, Le Sphinx convolvuli. Arch. Zool. Exper. et GPn. 60,145. Brosemer, R. W., Vogell, W. and Biicher, T. (1963). Morphologische und enzymatische Muster bei der Entwicklung indirekter Flugmuskeln von Locusia migratoria. Biochem. Z . 338,854-
910.
214
ANN E. KAMMER AND BERND HEINRICH
Brown, J. J. and Chippendale, G. M. (1974). Migration of the monarch butterfly, Danaus plexippus: energy sources. J. Insect Physiol. 20, 1117-1 130. Bruce, I. C. and Wilkens, J. L. (1976). Neurosecretory cells of the fleshfly, Sarcophaga bullata. J. Comp. Physiol. 112, 109-122. Biicher, T. and Kliigenberg, M. (1958). Wege des Wasserstoffs in der lebendigen Organization. Angew. Chem. 10,552-570. Bullard, B., Luke, B. and Winkelman, L. (1973). The paramyosin of insect flight muscle. J. Mol. Biol. 1 5 , 359-367. Burkett, B. (1962). Respiration of Galleria mellonella in relation to temperature and oxygen. Ent. Exp. Appl. 5,305-3 12. Burrows, M. (1975). Co-ordinating interneurones of the locust which convey two patterns of motor commands: their connexions with flight motoneurones. J. Exp. Biol. 63,713-733. Bursell, E. (1961). Post-teneral development of the thoracic musculature in tsetse flies. Proc. Roy. Ent. SOC.Lond. A 36,69-74. Bursell, E. (1963). Aspects of the metabolism of amino acids in the tsetse fly, Glossina (Diptera). J. Insect Physiol. 9,439-452. Bursell, E. (1966). Aspects of the flight metabolism of tsetse (Glossina). Comp. Biochem. Physiol. 19,809-8 18. Bursell, E. (1967). The conversion of glutamate to alanine in the tsetse fly (Glossina morsitans). Comp. Biochem. Physiol. 23, 825-829. Bursell, E. (1973). Development of mitochondria1 and contractile components of the flight muscle in adult tsetse flies, Glossina morsitans. J. Insect Physiol. 19, 1079-1086. Bursell, E. (1975). Substrates of oxidative metabolism in dipteran flight muscle. Comp. Biochem. Physiol. 52,235-238. Bursell, E. and Kuwengwa, T. (1972). The effect of flight on the development of flight musculature in the tsetse fly (Glossinamorsitans). Entomologia exp. appl. 15,229-237. Caldwell, R. L. and Rankin, M. A. (1972). Effects of a juvenile hormone mimic on flight in the milkweed bug, Oncopeltusfasciatus. Gen. Comp. Endocrinol. 19,60 1-605. Campbell, A. J. and Birt, L. M. (1972). Studies on the appearance of soluble a-glycerophosphate dehydrogenase activity during the development of the sheep blowfly Lucilia. Insect Biochem. 2,279-296. Campbell, A. J. and Birt, L. M.(1975). Thoracic muscle protein biosynthesis in Lucilia cuprina. Insect Biochem. 5,43-52. Candy, D. J. (1970). Metabolic studies on locust fight muscle using a new perfusion technique. J. Insect Physiol. 16,53 1-543. Candy, D. J. (1974). The control of muscle trehalase activity during locust flight. Biochem. Soc. Trans. 2, 1107-1 109. Candy, D. J., Hall, L. J. and Spencer, I. M. (1976). The metabolism of glycerol in the locust Schistocerca gregaria during flight. J. Insect Physiol. 22,583-587. Casaday, G. B. (1975). Neurodevelopment in a thoracic ganglion of a moth: identilied motor neuron and microanatomical structure. Thesis, Cornell University. Casey, T. M.(1976a). Flight energetics of sphinx moths: heat production and heat loss in Hyles lineata during free flight. J. Exp. Biol. 64,545-560. Casey, T. M.(1976b). Flight energetics of sphinx moths: power input during hovering flight. J. EXP.Biol. 64,529-543. Chadwick, L. E. and Gilmour, D. (1940). Respiration during flight in Drosophila repleta Wollaston: The oxygen consumption considered in relation to the wing rate. Physiol. Zool. 13,398-410. Chari, N. and Hajek, I. (1971). Metabolic differentiation of flight and jumping leg muscles of locust. Physiol. Bohemoslovaca, 20,57.
INSECT FLIGHT METABOLISM
21 5
Cheeseman, P.,Jutsum, A. R. and Goldsworthy, G. J. (1976). Quantitative studies on the release of locust adipokinetic hormone. Physiol. Ent. 1, 115-121. Chudakova, I. V. and Bocharova-Messner, 0. M. (1965). Changes in the functional and structural features of the wing muscles of the domestic cricket (Achetu domesticus L.). Dokl. Akud. Nauk. SSR 164,656-659. Church, N. S. (1960). Heat loss and body temperatures of flying insects. J. Exp. Biol. 37, 187212. Clark, M. G. (1976). Effect of 2-deoxy-D-glucose on flight and flight muscle substrate cycling by the bumble-bee. Comp. Biochem. Physiol. B55,409-415. Clark, M. G., Bloxham, D. P., Holland, P. C. and Lardy, H. A. (1973). Estimation of the fructose diphosphatase-phosphofructokinasesubstrate cycle in the flight muscle of Bombus ufinis. Biochem. J. 134,589-597. Clegg, J. S. and Evans, D. R. (1961). The physiology of blood trehalose and its function during flight in the blowfly. J.Exp. Biol. 38, 771-792. Cockbain, A. J. (1961). Fuel utilization and duration of tethered flight in Aphis fubae. J. Exp. Biol. 38, 163-174. Crabtree, B. and Newsholme, E. A. (1970). The. activities of proline dehydrogenase, glutamate dehydrogenase, aspartate-oxoglutarate aminotransferase and alanine-oxoglutarate amino transferase in some insect flight muscles. Biochem. J. 117, 1019-1021. Crabtree, B. and Newsholme, E. A. (1972a). The activities of phosphorylase, hexokinase, phosphofructokinase, lactate dehydrogenase and glycerol 3-phosphate dehydrogenase in muscles from vertebrates and invertebrates. Biochem. J. 126,49-58. Crabtree, B. and Newsholme, E. A. (1972b). The activities of lipases and carnitine palmitoyltransferase in muscles from vertebrates and invertebrates. Biochem. J. 130, 697705. Crabtree, B. and Newsholme, E. A. (1975). Comparative aspects of fuel utilization and metabolism by muscle. In “Insect Muscle” (Ed. P. N. R. Usherwood), pp. 405-501. Academic Press, London. Crossley, A. C. S. (1965). Transformations in the abdominal muscles of the blue blow-fly, Calliphoru erythrocephala (Meig), during metamorphosis. J. Embryol. exp. Morph. 14, 89110. Crossley, A., C. S. (1972a). Ultrastructural changes during transition of larval to adult intersegmental muscle at metamorphosis in the blowfly Culliphora erythrocephulu. I. Dedifferentiation and myoblast fusion. J. Embryol. exp. Morph. 27,43-74. Crossley, A. C. S. (1972b). Ultrastructural changes during transition of larval to adult intersegmental muscle at metamorphosis in the blowfly Calliphoru erythrocephulu. 11. The formation of adult muscle. J. Embryol. exp. Morph. 27,75-101. Dahlman, D. L. (1972). Age-dependent trehalase activity in adult tobacco hornworm tissues. Insect Biochem. 2, 143-149. Daly, H. V. (1964). Skeleto-muscular morphogenesis of the thorax and wings of the honeybee Apis mellifera (Hymenoptera: Apidae). California Publ. Ent. 39, 1-77. Darnhofer-Demar, B. (1969). Zur Funktionsmorphologie der Wasserlaufer. I. Die Morphologie des Lokomotionsapparates von Gerris locustris L. (Heteroptera: Gerridae). Zool. Jb. Anut. 86,28-66. Darnhofer-Demar, B. (1973). Zur Populationsdynamik einer univoltinen Population von Gerris lucustris (L.) (Heteroptera, Gerridae). Zool. Anz. Leiprig, 190, 189-204. Davis, N. T. (1975). Hormonal control of flight muscle histolysis in Dysdercus fuluoniger. Ann. ent. Soc. Am. 68,710-714. Davis, R. A. and Fraenkel, G. (1940). The oxygen consumption of flies during flight. J. Exp. Biol. 17,402-407.
216
ANN E. KAMMER AND BERND HEINRICH
Deak, I. I. (1976). Use of Drosophila mutants to investigate the effects of disuse on the maintenance of muscle. J. Insect Physiol. 22, 1159-1 165. deKort, C. A. D. (1969). Hormones and the structural and biochemical properties of the flight muscles in the Colorado beetle. Med. Landbouwhogeschool Wageningen, 69, 1-63. deKort, C. A. D. and Bartelink, A. K. M. (1972). Biochemical maturation of flight muscle mitochondria in the Colorado beetle, Leptinotarsa deCemh?QtQ Say. Comp. Biochem. Physiol. 43B, 1009-1017. deKort, C. A. D., Bartelink, A. K. M. and Scuurmans, R. R. (1973). The significance of Lproline for oxidative metabolism in the flight muscles of the Colorado beetle, Leptinotarsa deCemlint?QtQ.Insect Biochem. 3, 11-1 7. Del Castillo, J., Hoyle, G. and Machne, X. (1953). Neuromuscular transmission in a locust. J. Physiol., Lond. 121, 539-547. deWilde, J. and deBoer, J. A. (1961). Physiology of diapause in the adult Colorado beetle: 11. Diapause as a case of pseudo-allatectomy. J. Insect Physiol. 6, 152-161. Dingle, H. (1974). The experimental analysis of migration and life-history strategies in insects. In “Experimental Analysis of Insect Behaviour” (Ed. L. Barton Browne), pp. 329-342. Springer-Verlag, New York and Heidelberg. Dingle, H. and Arora, G. K. (1973). Experimental studies of migration in bugs of the genus Dysdercus. Oecologia, 12, 119-140. Domroese, K. A. and Gilbert, L. I. (1964). The role of lipid in adult development and flight muscle metabolism in Hyalophora cecropia. J.Exp. Biol. 41,573-590. Dorsett, D. A. (1962). Preparation for flight by hawk-moths. J. Exp. Biol. 39,579-588. Downer, R. G. H. and Matthews, J. R. (1976a). Glycogen depletion of thoracic musculature during flight in the American cockroach, Periplaneta americana L. Comp. Biochem. Physiol. SSB, 501-502. Downer, R. G. H. and Matthews, J. R. (1976b). Patterns of lipid distribution and utilization in insects. Amer. Zool. 16, 733-745. Downer, R. G. H. and Steele, J. E. (1972). Hormonal stimulation of lipid transport in the American cockroach, Periplaneta americana. Gen. Comp. Endocrinol. 19,259-265. Dudel, J. and Rudel, R. (1968). Temperature dependence of electromechanical coupling in crayfish muscle fibers. Pfltgers Archiv, 301, 16-30. Edwards, F. J. (1969a). Development and histolysis of the indirect flight muscles in Dysdercus intermedius. J. Insect Physiol. 15, 1591-1599. Edwards, F. J. (1969b). Environmental control of flight muscle histolysis in the bug Dysdercus intermedius. J. Insect Physiol. IS, 20 13-2020. Edwards, F. J. (1970). Endocrine control of flight muscle histolysis in Dysdercus intermedius. J. Insect Physiol. 16,2027-2033. Edwards, J. S. (1969). Postembryonic development and regeneration of the insect nervous system. In “Advances in Insect Physiology”, Vol. 6,97-137. Academic Press, New York and London. Edwards, J. S. and Palka, J. (1976). Neural generation and regeneration in insects. In “Simpler Networks and Behavior” (Ed. J. C. Fentress), pp. 167-185. Sinauer Associates, Sunderland, Massachusetts. Eigenmann, R. (1 965). Untersuchungen uber die Entwicklung der der dorsolongitudinalen Flugmuskeln von Antheruea pernyi Guer. (Lepidoptera). Rev. Suisse Zool. 72,789-840. Elder, H. Y. (1975). Muscle structure. In “Insect Muscle” (Ed. P. N. R. Usherwood), pp. 1-73. Academic Press, New York. Ellington, C. P. (1975). Non-steady-state aerodynamics of the flight of EnCQrSiQformosa. In “Swimming and Flying in Nature” (Eds T. Y.-F. Wu, C. J. Brokaw and C. Brennan), Vol. 2, pp. 783-796. Plenum, New York. Esch, H. (1964). ober den Zusammenhang zwischen Temperatur, Aktionspotentialen und Thoraxbewegungen bei der Honigbiene (Apis mellifca L.) Z . oergl. Physiol48,547-55 1.
INSECT FLIGHT METABOLISM
217
Esch, H. (1976). Body temperature and flight performance of honey bees in a servomechanically controlled wind tunnel. J. Comp. Physiol. 109,265-277. Esch, H. and Bastian, J. (1968). Mechanical and electrical activity in the indirect flight muscles of the honey bee. 2. vergl. Physiol. 58,429-440. Esch, H., Nachtigall, W. and Kogge, S. N. (1975). Correlations between aerodynamic output, electrical activity in the indirect flight muscles and wing position of bees flying in a servomechanical controlled wind tunnel. J. Comp. Physiol. 100, 147-159. Evans, D. R. and Dethier, V. G. (1957). The regulation of taste thresholds for sugars in the blowfly. J. Insect Physiol. 1,3-17. Farnworth, E. G. (1972a). The effects of ambient temperature, humidity, and age on wing-beat frequency of Periplaneta species. J. Insect Physiol. 18,827-839. Farnworth, E. G. (1972b). Effects of ambient temperature and humidity on internal temperature and wing-beat frequency of Periplaneta americana. J. Insect Physiol. 18,359-372. Finlayson, L. H. (1975). Development and degeneration. In “Insect Muscle” (Ed. P. N. R. Usherwood), pp. 75-149. Academic Press, London. Florey, E. and Hoyle, G. (1976). The effects of temperature on a nerve-muscle system of the Hawaiian Ghost crab, Ocypode ceratophthalma (Pallas). J. Comp. Physiol. 110,5 1-64. Friedman, S. (1961). Inhibition of trehalase activity in the hemolymph of Phormia regina Meig. Arch. Biochem. Biophys. 93,550-554. Friedman, S . (1967). The control of trehalose synthesis in the blowfly, Phormia regina Meig. J. Insect Physiol. 13,397-405. Gehring, W. J. (1976). Developmental genetics of Drosophila. Ann. Rev. Genet. 10,209-252. Gettrup, E. and Wilson, D. M. (1964). The lift-control reaction of flying locusts. J. Exp. Biol. 41, 183-190.
Gewecke, M. (1967). Die Wirkung von Luftstromung auf die Antennen und das Flugverhalten der Blauen Schmeissfliege(Calliphora erythrocephala). 2.vergl. Physiol. 54, 12 1-164. Gewecke, M. (1972). Antennen und Stirn-Scheitelhaare von Locusta migratoria L. als Luftstromungs-Sinnesorganebie der Flugsteuerung. J. Comp. Physiol. 80,57-94. Gewecke, M. (1974). The antennae of insects as air-current Sense organs and their relationship to the control of flight. In “Experimental Analysis of Insect Behavior” (Ed. L. Barton Browne), pp. 100-1 13. Springer-Verlag, New York and Heidelberg. Gewecke, M. (1975). The influence of the air-current sense organs on the flight behaviour of Locusta migratoria. J. Comp. Physiol. 103,79-95. Gilbert, L. I. (1967). Lipid metabolism and function in insects. In “Advances in Insect Physiology”, Vol. 4,69-2 11, Academic Press, New York and London. Goldsworthy, G. J. (1969). Hyperglycaemic factors from the corpus cardiacum of Locusta migratoria. J. Insect Physiol. 15,213 1-2140. Goldsworthy, G. J. (1976). Hormones and flight in the locust. In “Perspectives in Experimental Biology”, Vol. 1, “Zoology” (Ed. P. S. Davis), pp. 167-177. Pergamon Press, Oxford. Goldsworthy, G. J. and Coupland, A. J. (1974). The influence of the corpora cardiaca and substrate available on flight speed and wing-beat frequency in Locusta. J. Comp. Physiol. 89, 359-368.
Goldsworthy, G. J., Coupland, A. J. and Mordue, W. (1973a). The effects of corpora cardiaca on tethered flight in the locust. J. Comp. Physiol. 82,339-346. Goldsworthy, G. J., Johnson, R. A. and Mordue, W. (1972a). In vioo studies on the release of hormones from the corpora cardiaca of locusts. J. Comp. Physiol. 79,85-96. Goldsworthy, G. J. and Mordue, W. (1974). Neurosecretory hormones in insects. J. Endocrinol. 60,529-558.
Goldsworthy, G. J., Mordue, W. and Guthkelch, J. (1972b). Studies on insect adipokinetic hormones. Gen. comp. Endocrinol. 18,545-55 1. Gotz, K. G. (1968). Flight control in Drosophila by visual perception of motion. Kybernetik, Berlin, 4, 199-208.
218
ANN E. KAMMER AND BERND HEINRICH
Greenewalt, C. H. (1960). The wings of insects and birds as mechanical oscillators. Proc. Amer. Philosoph. SOC.104,605-6 11. Greenewalt, C. H. (1962). Dimensional relationships for flying animals. Smithsonian Misc. Coll. 144, no. 2, 1-46. Hanegan, J. L. and Heath, J. E. (1970). Temperature dependence of the neural control of the moth flight system. J. Exp. Biol. 53,629-639. Hansford, R. G. and Johnson, R. N. (1975). The nature and control of the tricarboxylate cycle in beetle flight muscle. Biochem. J. 148,389-402. Harcombe, E. S. (1975). “Neurophysiology of flight motoneurons in Drosophila melanogaster.” Thesis, Yale University. Hargrove, J. W. (1975a). Some changes in the flight apparatus of tsetse flies, Glossina morsitans and G. pallidipes, during maturation. J. Insect Physiol. 21, 1485-1489. Hargrove, J. W. (1975b). The flight performance of tsetse flies. J. Insect Physiol. 21, 13851395. Hargrove, J. W. (1976). Amino acid metabolism during flight in tsetse flies. J. Insect Physiol. 22, 309-313. Heide, G. (1971). Die Funktion der nicht-fibrillaren Flugmuskeln von Calliphora, Ted I. Lage, Insertionsstellen und Innervierungsmuster der Muskeln. Teil 11. Muskulare Mechanismen der Flugsteuerung und ihre nervose Kontrolle. Zool. Jb. Physiol. 76,87-13 1. Heide, G. (1974). The influence of wing-beat synchronous feedback on the motor output systems in flies. Z . Naturforsch, 29, 739-744. Heinertz, R. (1976). Untersuchungen am thorakalen Nervensystem von Antheraea polyphemus. Cr. (Lepidoptera) unter besonderer Beriicksichtigung der Metamorphose. Revue suisse Zool. 83,2 15-242. Heinrich, B. (1970). Thoracic temperature stabilization in a free-flying moth. Science, 168,580582. Heinrich, B. (1971a). Temperature regulation of the sphinx moth, Manduca sexta. I. Flight energetics and body temperature during free and tethered flight. J. Exp. Biol. 54, 141-152. Heinrich, B. (1971b). Temperature regulation of the sphinx moth, Manduca sexta. 11. Regulation of heat loss by control of blood circulation. J. Exp. Biol. 54, 153-166. Heinrich, B. (1973). Mechanisms of insect thermoregulation. In “Effects of Temperature on Ectothermic Organisms: Ecological Implications and Mechanisms of Compensation” (Ed. W. Weiser), pp. 139-1 50. Springer-Verlag, Heidelberg. Heinrich, B. (1974a). Thermoregulation in endothermic insects. Science, 185,747-756. Heinrich, B. (1974b). Thermoregulation in bumblebees: I. Brood incubation by Bombus vosnesenskii queens. J. Comp. Physiol. 88, 129-140. Heinrich, B. (1975). Thermoregulation in bumblebees. 11. Energetics of warm up and free flight. J. Comp. Physiol. B95, 155-166. Heinrich, B. (1976). Heat exchange in relation to blood flow between thorax and abdomen in bumblebees. J. Exp. Biol. 64,561-585. Heinrich, B. (1977). Why have some animals evolved to regulate a high body temperature? Am. Not. 111,623-640. Heinrich, B. and Bartholomew, G. A. (1971). An analysis of pre-flight warm-up in the sphinx moth, Manduca sexta. J. exp. Biol. 55,223-239. Heinrich, B. and Casey, T. M. (1973). Metabolic rate- and endothermy in sphinx moths. J. Comp. Physiol. 83, 195-206. Heinrich, B. and Kammer, A. (1973). Activation of the fibrillar muscles in the bumblebee during warm-up, stabilization of thoracic temperature and flight. J. Exp. Biol. 58,677-688. Heinrich, B. and Pantle, C. (1975). Thermoregulation in small flies (Syrphus sp.): basking and shivering. J. Exp. Biol. 62,599-610.
INSECT FLIGHT METABOLISM
219
Heywood, R. B. (1965). Changes occurring in the central nervous system of Pieris brassicae L (Lepidoptera) during metamorphosis. J. Insect Physiol. 11,4 13-430. Highnam, K. C. and Goldsworthy, G. J. (1972). Regenerated corpora cardiaca and hyperglycemic factor in Locusta migratoria. Gen. Comp. Endocrinol. 18,83-88. Hill, A. V. (1939). The mechanical efficiency of frog muscle. Proc. Roy. SOC.Lond. B127,434451. Hill, L., Luntz, A. J. and Steele, P. A. (1968). The relationship between somatic growth, ovarian growth, and feeding activity in the adult desert locust. J. Insect Physiol. 14, 1-20. Hinton, H. E. (196 1). The role of the epidermis in the disposition of tracheae and muscles. Sci. Progr. 49,329-339. Hochachka, P. W. and Somero, A. N. (1973). “Strategies of Biochemical Adaptation.” W. B. Saunders, Philadelphia; London, Toronto. Hocking, B. (1953). The intrinsic range and speed of flight of insects. Trans. Roy. Enr. SOC. Lond. 104,223-345. Holmes, E. A. and Keeley, L. L. (1975). Mitochondria1 development in the flight muscles of the moth Heliothis virescens. Insect Biochem. 5, 15-24. Houben, N. M. D. and Beenakkers, A. M. T. (1975). The influence of carbohydrate concentration on the release of adipokinetic hormone during locust flight. J. Endocrinol. 64, 66P. House, H. L. (1974). Nutrition. In “The Physiology of Insects” (Ed. M. Rockstein), 2nd Edn., Vol. 5, pp. 1-62. Academic Press, N.Y. Hoyle, G. (1974). A function for neurons (DUM) neurosecretory on skeletal muscle on insects. J. Exp. Zool. 189,401406. Hoyle, G., Dagan, D., Moberly, B. and Colquhoun, W. (1974). Dorsal unpaired median insect neurons make neurosecretory endings on skeletal muscle. J. Exp. Zool. 187, 159-165. Ikeda, K. and Boettiger, E. G. (1965a). Studies on the flight mechanism of insects. 11. The innervation and electrical activity of the fibrillar muscles of the bumblebee, Bombus. J. Insect Physiol. 11,779-789. Ikeda, K. and Boettiger, E. G. (1965b). Studies of the flight mechanism of insects. 111. the innervation and electrical activity of the basalar fibrillar flight muscle of the beetle, Oryctes rhinoceros. J. Insect Physiol. 11,79 1-802. Jensen, M. (1956). Biology and physics of locust flight. 111. The aerodynamics of locust flight. Phil. Trans. Roy. SOC.London, B239,5 11-552. Johnson, B. (1957). Studies on the degeneration of the flight muscles of alate aphids. I. A comparative study of the occurrence of muscle breakdown in relation to reproduction in several species. J. Insect Physiol. 1,248-256. Johnson, B. G., Jr. and Rowley, W. A. (1972). Age-related ultrastructural changes in the flight muscle of the mosquito, Culex tarsalis. J. Insect Physiol. 18,2375-2389. Johnson, C . G. (1969). “Migration and Dispersal of Insects by Flight.” 763 pp. Methuen, London. Johnson, C. G. (1974). Insect migration: aspects of its physiology. In “The Physiology of Insecta” (Ed. M. Rockstein), 2nd Edn., Vol. 111, pp. 280-334.Academic Press, New York and London. Johnson, C. G. (1976). Lability of the flight system: a context for functional adaptation. In “Insect Flight” (Ed. R. C. Rainey), Symp. Roy. Enr. SOC.Lond. 7,217-234. Johnson, R. N. and Hansford, R. G. (1975). The control of tricarboxylate-cycle oxidations in blowfly flight muscle, the steady-state concentrations of citrate isocitrate 2-oxoglutarate and malate in flight muscle and isolated mitochondria. Biochem. J. 146,527-535. Jongbloed, J. and Wiersma, C. A. G. (1934). Der Stoffwechsel der Honigbiene wahrend des Fliegens. Z. vergl. Physiol. 21, 519-533.
220
ANN E. KAMMER AND BERND HEINRICH
Josephson, R. K. (1973). Contraction kinetics of the fast muscles used in singing by a katydid. J. Exp. Biol. 59,781-801. Josephson, R. K. (1975). Extensive and intensive factors determining the performance of striated muscle. J. Exp. Zool. 194, 135-153. Judge, F. D. and Schaefers, G. A. (1971). Effects of crowding on alary polymorphism in the aphid Chaetosiphonfragaefolii. J. Insect Physiol. 17, 143-148. Jutsum, A. R. and Goldsworthy, G. J. (1975). Some problems of homeostasis of haernolymph metabolites in the locust. J. Insect Physiol. 21, 1899-1908. Jutsum, A. R. and Goldsworthy, G. J. (1976). Fuels for flight in Locusta. J. Insect Physiol. 22, 243-249. Kammer, A. E. (1967). Muscle activity during flight in some large Lepidoptera. J. Exp. Biol. 47, 277-295. Kammer, A. E. (1968). Motor patterns during flight and warm-up in Lepidoptera. J. Exp. Biol. 48,89-109. Kammer, A. E. (1970a). A comparative study of motor patterns during pre-flight warm-up in hawkmoths. Z. vergl. Physiol. 70,45-56. Kammer, A. E. (1970b). Thoracic temperature, shivering, and flight in the monarch butterfly, Danaus plexippus (L.).Z . vergl. Physiol. 68,334-344. Kammer, A. E. (1971). The motor output during turning flight in a hawkmoth, Manduca sexta. J. Insect Physiol. 17, 1073-1086. Kammer, A. E. (1976). Respiration and the generation of rhythmic outputs in insects. Fed. Proc. 35,1992-1999. Kammer, A. E. and Heinrich, B. (1972). Neural control of bumblebee fibrillar muscles during shivering. J. Comp. Physiol. 78,337-345. Kammer, A. E. and Heinrich, B. (1974). Metabolic rates related to muscle activity in bumblebees. J. Exp. Biol. 61,219-227. Kammer, A. E. and Rheuben, M. B. (1976a). Metamorphosis of the motor system during the development of moths. In “Neural Control of Locomotion” (Eds R. M. Herman, S.Grillner, S. G. Stein and D. G. Stuart), pp. 775-779. Plenum, New York. Kammer, A. E. and Rheuben, M. B. (1976b). Adult motor patterns produced by moth pupae during development. J. Exp. Biol. 65,65-84. Kaufmann, L. (197 1). Untersuchungen zur Funktion der Flugmuskulatur von Antheraea polyphemus (Lep.) wahrend der Imaginalentwicklung. Rev. suisse Zool. 78, 1007-1036. Kerukize, G. R. (1972). Utilization of fat reserve substances by Homorocoryphus (Orthoptera: Tettigoniidae) during flight. Comp. Biochem. Physiol. 943,563-569. Kirsten, E., Kirsten, R. and Arese, P. (1963). Das Verhalten von freien Aminosauren, energiereichen Phosphorsaure-Verbidungen und einigen Glykolyse- und Tricarbonsaurecyclus-substraten in Muskeln von Locusfa migratoria bei der Arbeit. Biochem. Z. 337,167-178. Kosmin, N. P., Alpatov, W. W. and Resnitschenko, M. S. (1932). Zur Kenntnis des Gaswechsels und des Energieverbrauchs der Biene in Beziehung zu deren Aktivitat. Z. uergf. Physiol. 17,408-42 1. Krogh, A. and Zeuthen, E. (1941). The mechanism of flight preparation in some insects. J. Exp. Biol. 18, 1-10. Krogh, A. and Weis-Fogh, T. (1951). The respiratory exchange of the desert locust (Schistocercagregaria) before, during and after flight. J. Exp. Biol. 28,344-357. Kring, J. B. (1972). Flight behavior of aphids.Ann. Rev.Ent. 17,461-492. Kutsch, W. (1971). The development of the flight pattern in the desert locust, Schistocerca gregaria. Z . vergl. Physiol. 74, 156-168. Kutsch, W. (1973). The influence of age and culture femperature on the wingbeat frequency of the migratary locust, Locusta migratoria. J. Insect Physiol. 19, 763-772.
INSECT FLIGHT METABOLISM
22 1
Kutsch, W. (1974a). The development of the flight pattern in locusts. I n “Experimental Analysis of Insect Behaviour” (Ed. L. Barton Brown), pp. 149-158. Springer-Verlag, New York, Heidelberg, Berlin. Kutsch, W. (1974b). The influence of the wing sense organs on the flight motor pattern in maturing adult locusts. J. comp. Physiol. 88,413-424. Kutsch, W. and Usherwood, P. N. R. (1970). Studies of the innervation and electrical activity of flight muscles in the locust, Schistocerca gregaria. J. Exp. Biol. 52,299-3 12. Lasiewski, R. C. (1963). Oxygen consumption of torpid, resting, active and flying hummingbirds. Physiol. Zool. 36, 122-140. Lee, S. S. and Goldsworthy, G. J. (1975). Allatectomy and flight performance in male LOCUSIU migratoria. J. Comp. Physiol. 100,35 1-359. Lees, A. D. (1967). The production of the apterous and alate forms in the aphid Megoura oiciae Buckton, with special reference to the role of crowding. J. Insect Physiol. 13,289-3 18. Levenbook, L. and Williams, C. M. (1956). Mitochondria in the flight muscles of insects. 111. Mitochondrial cytochrome c in relation to aging and wing-beat frequency of flies. J. gen. Physiol. 39,497-5 12. Levine, R. J. C., Elfvin, M., Dewey, M.M. and Walcott, B. (1976). Paramyosin in invertebrate muscles. 11. Content in relation to structure and function. J. Cell Biol. 71, 273-279. Lighthill, J. (1973). On the Weis-Fogh mechanism of lift generation. J. FluidMech. 60, 1-17. Lighthill, J. (1975). Aerodynamic aspects of animal flight. I n “Swimming and Flying in Nature” (Eds T. Y.-F. Wu, C. G. Brokaw and C. Brennan), Vol. 2, pp. 423-491. Plenum, New York. Machin, K. E. and Pringle, J. W. S. (1959). The physiology of insect fibrillar muscle. 11. Mechanical properties of a beetle flight muscle. Proc. Roy. SOC.Lond. B15 1,204-225. Machin, K. E., Pringle, J. W. S. and Tamasige, M. (1962). The physiology of insect fibrillar muscle. IV. The effect of temperature on a beetle flight muscle. Proc. Roy. SOC.Lond. B155, 493-499. Maddrell, S. H. P. (1974). Neurosecretion. I n “Insect Neurobiology” (Ed. J. E. Treherne), pp. 307-357. North-Holland Publ. Co., Amsterdam. Maruyana, K. (1974). The biochemistry of the contractile elements of insect muscle. I n “Physiology of Insecta” (Ed. M. Rockstein), Vol. IV, 237-269. Academic Press, N.Y. May, M. L. (1976). Thermoregulation and adaptation to temperature in dragonflies. Ecol. Monog. 461-32. Mayer, R. J. and Candy, D. J. (1969a). Changes in energy reserves during flight of the desert locust, Schistocerca gregaria. Comp. Biochem. Physiol. 3 I, 409-4 18. Mayer, R. J. and Candy, D. J. (1969b). Control of haemolymph lipid concentration during locust flight: an adipokinetic hormone from the corpora cardiaca. J. Insect Physiol. 15, 61 1620. McCabe, C. T. and Bursell, E. (1975). Metabolism of digestive products in the tsetse fly, Glossina morsitans. Insect Biochem. 5,769-779. McCrea, M. J. and Heath, J. E. (1971). Dependence of flight on temperature regulation in the moth, Manduca sexta. J. Exp. Biol. 54,415-435. Michel, R. (1972). Influence des corpora cardiaca sur les possibilites de vol soutenu du criquet p6lerin Schistocerca gregaria. J. Insect Physiol. 18, 1811-1827. Michel, R. (1973a). Effect of corpora allata and mimetic juvenile hormone on the sustained flight ability of the pilgrim locust Schistocerca gregaria bred in dense groups and in isolation. Arch. Zool.Exp. Gen. 114, 187-211. Michel, R. (1973b). Variations de la tendance au vol soutenu du criquet pblerin Schistocerca gregaria aprks implantations de corpora cardiaca. J. Insect Physiol. 19, 1317-1325. Michel, R. and Bernard, A. (1973). Influence de la pars intercerebralis sur l’induction au vol soutenu chez le criquet perlerin Schistocerca gregaria. Acrida, 2,139-149. Miller, P. L. (1966). The regulation of breathing in insects. I n “Advances in Insect Physiology”, Vol. 3,279-354. Academic Press, New York and London.
222
ANN E. KAMMER AND BERND HEINRICH
Miller, P. L. (1974). Respiration-aerial gas transport. In “The Physiology of Insecta” (Ed. M. Rockstein), 2nd Edn., Vol. VI, pp. 346-402. Academic Press, New York and London. MOM, B. and Zarnack, W. (1975). Flugsteuerung der Wanderheuschrecke durch Verschiebung der Muskelaktivitat. Natunvissenschqften 65 441-442. Mordue, W. and Goldsworthy, G. J. (1969). The physiological effects of corpus cardiacum extracts in locusts. Gen. Comp. Endocrinol. 12, 360-369. Mulloney, B. (1970). Impulse patterns in the flight motor neurons of Bombus californicus and Oncopeltusfasciatus. J. Exp. Biol. 52,59-77. Nachtigall, W. (1966). Die Kinematik der Schlagfliigelbewegungen von Dipteran. 2. oergl. Physiol. 52, 155-211. Nachtigall, W. (1967). Aerodynamische Messungen am Tragfliigelsystem segelnder Schmetterlingen. 2.vergl. Physiol. 54,2 10-23 1. Nachtigall, W. (1976). Wing movements and the generation of aerodynamic forces by some medium-sized insects. In “Insect Flight” (Ed. R. C. Rainey), pp. 3 1 4 7 . J. Wiley, New York. Nachtigall, W., Widmann, R. and Renner, M. (1971). h e r den “ortsfesten” freien Flug von Bienen in einem Saugkanal. Apidologie, 2,271-282. Nachtigall, W. and Wilson, D. M. (1967). Neuro-muscular control of dipteran flight. J. Exp. Biol. 47, 77-97. Nayar, J. K. and Sauerman, Jr., D. H. (1973). A comparative study of flight performance and fuel utilization as a function of age in females of Florida mosquitoes. J. Insect Physiol. 19, 1977-1988. Nayar, J. K. and Van Handel, E. (1971). Flight performance and metabolism of the moth Spodopterafmgiperda. J. Insect Physiol. 17,2475-2480. Neuhaus, W. and Wohlgemuth, R. (1960). h r das Facheln der Bienen und dessen Verhaltnis zum Fliegen. Z . vergl. Physiol. 43,6 1 5 6 4 1 . Neville, A. C. and Weis-Fogh, T. (1963). The effect of temperature on locust flight muscle. J. Exp.Biol.40, 111-121. Newsholme, E. A. and Start, C. (1973). “Regulation in Metabolism.” 349 pp. J. Wiley, New York. Newsholme, E. A., Crabtree, B., Higgins, S. J., Thornton, S. D. and Start, C. (1972). The activities of fructose diphosphatase (EC-3.1.3.11) in flight muscles from the bumblebee and the role of this enzyme in heat generation. Biochem. J. 128,89-97. Norberg, R. A. (1975). Hovering flight of the dragonfly Aeschna juncea L., kinematics and aerodynamics. In “Swimming and Flying in Nature” (Eds T. Y.-F. Wu, C. J. Brokaw and C. Brennan), Vol. 2, pp. 763-781. Plenum, New York. Normann, T. C. (1972). Heart activity and its control in the adult blowfly, Calliphora erythrocephala. J. Insect Physiol. 18, 1793-18 10. Normann, T. C. (1973). Membrane potential of the corpus cardiacum neurosecretory cells of the blowfly, Calliphora erythrocephala. J. Insect Physiol. 19,303-3 18. Normann, T. C. (1974). Calcium dependence ofneurosecretion by exocytosis. J. Exp. Biol. 61, 401410. Normann, T. C. (1975). Neurosecretory cells in insect brain and production of hypoglycaemic hormone. Nature, Lond. 254,259-26 1. Normann, T. C. and Duve, H. (1969). Experimentally induced release of a neurohormone influencing hemolymph trehalose level in Calliphora erythrocephala (Diptera). Gen. Comp. Endocr. 12,449-459. Niiesch, H. (1968). The role of the nervous system in insect morphogenesis and regeneration. Ann. Rev. Entomol. 1 3 , 2 7 4 4 . Niiesch, H. and Bienz-Isler, G. (1972). Die Entwicklung denervierter Imaginalmuskeln bei
INSECT FLIGHT METABOLISM
223
Antheraea polyphemus (Lep.) und die Frage der entwicklungsfordernden Nervenwirkung. Zwl.Jb. Anat. 89,333-350. Nwanze, K. F., Maskarinee, J. K. and Hopkins, T. L. (1976). Lipid composition of the normal and flight forms of adult cowpea weevils, Callosobruchus maculatus. J. Insect Physiol. 22, 897-899. O’Bricn, S . J. and Machtyre, R. J. (1972). The crglycerophosphate cycle in Drosophila melanogaster. 11. Genetic aspects. Genetics, N.Y.71, 127-138. O’Bricn, S. J. and Shimada, Y. (1974). The a-glycerophosphate cycle in Drosophila melanogaster. IV. Metabolic, ultrastructural, and adaptive consequences of u G p d h - 1 “Null” mutations. J. Cell Biol. 63, 864-882. Osborne, M. F. M. (1951). Aerodynamics of flapping flight with application to insects. J. Exp. Biol. 23,221-245. Osborne, M. P., Finlayson, L. H.and Rice, M. J. (1971). Neurosecretory endings associated with striated muscles in three insects. (Schistocerca, Carausius and Phormia) and a frog (Rana). Z . Zellforsch. 116,391-404. Panov, A. A. (1963). The origin and fate of neuroblasts, neurons, and neuroglid cells in the central nervous system of the China oak silk worm Antheraea pernyi Guer. (Lepidoptera, Attacidae). Ent. Review. 42,186-191. Patlak, J. B. (1976). The ionic basis for the action potential in the flight muscle of the fly, Sarcophaga bullata. J. Comp. Physiol. A107, 1-1 1. Peled, Y. and Tietz, A. (1973). Fat transport in the locust, Locusla migratoria: the role of protein synthesis. Biochem. Biophys. Acta, 2%, 499-509. Peristianis, C. S. and Gregory, D. W. (1971). Early stages of flight muscle development in the blowfly Lucilia cuprina: a light and electron microscopic study. J. Insect Physiol. 17, 10051012. Petersen, B., Lundgren, L. and Wilson, L. (1957). The development of flight capacity in a butterfly. Behauiour, 10,324-339. Pipa, R. L. (1963). Studies on the hexapod nervous system. VI. Ventral nerve cord shortening: a metamorphic process in Galleria mellonellu (L.) (Lepidoptera, Pyrallidae). Biol. Bull. 124, 293-302. Pipa, R. L.(1967). Insect neurometamorphosis. 111. Nerve cord shortening in a moth, Galleria mellonella (L.)may be accomplished by humeral potentiation of neuroglial motility. J. Exp. Z w l . l64,47-60. Pipa, R. L. (1973). Proliferation, movement and regression of neurons during the postembryonic development of insects. I n “Developmental Neurobiology of Arthropods” (Ed. D. Young), pp. 105-129. Cambridge University Press. Polacek, I. and Kubista, V. (1960). Metabolism of the cockroach Periplaneta americana during flight. Physiol. Bohemoslov. 9,228-234. Pringle, J. W. S. (1949). The excitation and contraction of the flight muscles of insects. J. Physiol. 108,226-232. Pringle, J. W. S. (1957).“Insect Flight.” 132 pp. Cambridge University Press. Pringle, J. W. S. (1968). Comparative physiology of the flight motor. I n “Advances in Insect Physiology”, Vol. 5, 163-227. Academic Press, New York and London. Pringle, J. W. S. (1974). Locomotion: Flight. I n “The Physiology of Insecta” (Ed. M. Rockstein), Vol. 111, pp. 433-476. Academic Press, New York and London. Pybus, J. and Tregear, R. (1973). Estimates of force and time of actomyosin interaction in an active muscle and of the number interacting at any one time. Cold Spring Harbor Symp. Quant. Biol. 37,655-660. Rainey, R. C., Ed. (1976).“Insect Flight.” Roy Ent. SOC.Symp. 7,287 pp. Rankin, M. A. (1974). The hormonal control of flight in the milkweed bug Oncopeltusfasciatus.
224
ANN E. KAMMER AND BERND HEINRICH
In “Experimental Analysis of Insect Behaviour” (Ed. L. Barton Browne), pp. 317-328. Springer-Verlag, New York. Rheuben, M. B. (1975). Change in structure from “slow” to “fast” of moth muscle fibers during development. Neurosci. Abs. 1,778. Robinson, N. L. and Goldsworthy, G. J. (1974). The effects of locust adipokinetic hormone on flight muscle metabolism in vivo and in vitro. J. Cornp. Physiol. 89,369-377. Robinson, N. L. and Goldsworthy, G. J. (1976). Adipokinetic hormone and flight metabolism in the locust. J. Insect Physiol. 22, 1559-1564. Rockstein, M. and Brandt, K. F. (1963). Enzyme changes in fight muscle correlated with aging and flight ability in the male housefly. Science, N.Y. 139, 1049-1051. Rockstein, M. and Bhatnagar, P. L. (1965). Age changes in size and number of the giant mitochondria in the flight muscle of the common housefly (Musca dornestica L.). J. Insect Physiol. 11,481-491. Rockstein, M. and Bhatnagar, P. L. (1966). Duration and frequency of wing beat in the aging housefly (Musca dornestica L.). Biol. Bull. (Woods Hole), 131,479486. Rockstein, M. and Miguel, J. (1973). Aging in insects. In “The Physiology of Insecta” (Ed. M. Rockstein), 2nd Edn., Vol. 1, pp. 371-478. Academic Press, New York. Rowley, W. A. and Graham, C. L. (1968). The effect of age on the flight performance of female Aedes aegypti mosquitoes. J. Insect Physiol. 14,7 19-728. Sacktor, B. (1965). Energetics and respiratory metabolism of muscular contraction. In “The Physiology of Insecta” (Ed. M. Rockstein), Vol. 2, pp. 483-580. Academic Press, New York. Sacktor, B. (1970). Regulation of intermediary metabolism, with special reference to the control mechanisms in insect flight muscle. In “Advances in Insect Physiology”, Vol. 7, 268-347. Academic Press, New York and London. Sacktor, B. (1974). Biological oxidations and energetics in insect mitochondria. In “The Physiology of Insecta” (Ed. M. Rockstein), Vol. 4, 2nd Edn., pp. 271-353. Academic Press, New York. Sacktor, B. (1975). Biochemistry of insect flight. Part I, Utilization of fuels by muscle. In “Insect Biochemistry and Function” (Eds D. J. Candy and B. A. Kilby), pp. 1-88. Wiley, NewYork. Sacktor, B. and Childress, C. C. (1967). Metabolism of proline in insect flight muscle and its significancein stimulating the oxidation of pyruvate. Arch. Biochern. Biophys. 120,583-588. Sacktor, B. and Hurlbut, E. C. (1966). Regulation of metabolism in working muscle in oivo. 11. Concentrations of adenine nucleotides, arginine phosphate, and inorganic phosphate in insect flight muscle during flight. J. biol. Chern. 241,632-634. Sacktor, B. and Wormser-Shavit, E. (1966). Regulation of metabolism in working muscle in vivo. I. Concentration of some glycolytic, tricarboxylic acid cycle, and amino acid intermediates in insect flight muscle during flight. J. biol. Chem. 241,624-63 1. Sahota, T. S. (1975). Effect of juvenile hormone on acid phosphatases in the degenerating fight muscles of the Douglas-fir beetle, Dendroctonus pseudotsugae. J. Insect Physiol. 21,471478. Samaranayaka, M. (1976). Possible involvement of monoamines in the release of adipokinetic hormone in the locust Schistocerca gregaria. J. Exp. Biol. 65,415425. Sanes, J. R. and Hildebrand, J. G. (1976a). Structure and development of antennae in a moth, Manduca sexta. Devel. Biol. 5 1,282-299. Sanes, J. R. and Hildebrand, J. G. (1976b). Origin and morphogenesis of sensory neurons in an insect antenna. Devel. Biol. 5 1, 300-3 19. Schaefers, G. A. and Judge, F. D. (.197 1). Effects of temperature, photoperiod, and host plant on alary polymorphism in the aphid Chaetosiphonfragaefolii. J. Insect Physiol. 11,365-379. Schmidt-Nielsen, K. (1972). Locomotion: energy cost of swimming, flying, and running. Science, 111,222-228. Schneider, P. and Hermes, M. (1976). Die Bedeutung der Elytren bei Vertretern des MelolonthaFlugtyps (Coleoptera). J. Cornp. Physiol. 106,3949.
INSECT FLIGHT METABOLISM
225
Scudder, G. G. E. and Hewson, R. J. (1971). The postembryonic development of the indirect flight muscles in Oncopeltus fasciatus (Dallas) (Hemiptera: Hygaeidae). Canad. J. Zool. 49, 1377-1386. Seecof, R. L., Gerson, I., Donady, J. J. and Teplitz, R. L. (1973). Drosophila myogenesis in vitro: The genesis of "small" myocytes and myotubes. Devel. Biol. 35,250-261. Shatoury, H. H. El. (1956). Developmental interactions in the development of the imaginal muscles of Drosophila. J. Embryol. exp. Morph. 4,228-239. Shaw, M. J. P. (1970). Effects of population density on alienicolae of Aphisfabae Scop. I. The effect of crowding on the production of alatae in the laboratory. Ann. appl. Biol. 65, 191-196. Simon, J., Bhatnagar, P. L. and Milburn, N. S. (1969). An electron microscope study of changes in mitochondria of flight muscle of aging houseflies (Musca domestica). J. Insect Physiol. 15, 135- 140. Smith, D. S. (1966). The organization and function of the sarcoplasmic reticulum and T-system of muscle cells. Prog. Biophysics Mol. Biol. 16, 107-142. Smith, D. S. and Sacktor, B. (1970). Disposition of membranes and the entry of haemolymphborne ferritin in flight muscle fibers of the fly Phormia regina. Tissue and Cell, 2,355-374. Snodgrass, R. E. (1954). Insect metamorphosis. Smithsonian Misc. Coll. 122(9), 1-24. Sohal, R.D. (1975). Mitochondrial changes in flight muscles of normal and flightless Drosophila melanogaster with age. J. Morph. 145,337-354. Sohal, R.D. (1976). Aging changes in insect flight muscle. Gerontology, 22,317-333. Sotavalta, 0.(1947). The flight-tone (wingstroke frequency) of insects. Acta ent. Fenn. 4, 1-1 17. Sotavalta, 0. (1952). The essential factor regulating the wing-stroke frequency of insects in wing mutilation and loading experiments and in experiments at subatmospheric pressure. Ann. Zool. SOC.Zoo1.-Bot. Fenn. Vanamo, 15, 1-67. Sotavalta, 0. (1954a). The effect of wing inertia on the wing-stroke frequency of moths, dragonflies and cockroaches. Ann. ent. Fenn. 20,93-101. Sotavalta, 0. (1954b). On the fuel consumption of the honeybee (Apis mellifca L.) in flight experiments. Ann. Zool. SOC.2001.-Bot.Fenn. Vanamo, 16, 1-27. Sotavalta, 0. and Laulajainen, E. (1961). On the sugar consumption of the drone fly (Eristalis tenax L.) in flight experiments. Ann. Acad. Sci. Fenn., Series A, IV. Biol., No. 53, 1-25. Spencer, I. M. and Candy, D. J. (1974). The effect of flight on the concentrations and composition of haemolymph diacylglycerides in the desert locust. Biochem. SOC.Trans. 2, 1093-1096. Steele, J. E. (196 1). Occurrence of a hyperglycaemic factor in the corpus cardiacum of an insect. Nature. Lond. 192, 680-681. Steele, J. E. (1963). The site of action of insect hyperglycemic hormone. Gen. Comp. Endocrinol. 3,46-52. Stellwaag, F. (19 16). Wie steuern die Insekten wiihrend des Fluges. Biol. Zbl. 3 6 , 3 0 4 4 . Stevenson, E. (1968a). Carbohydrate metabolism in the flight muscle of the Southern armyworm moth, Prodenia eridania. J. Insect Physiol. 14, 179-198. Stevenson, E. (1968b). The carnitine-independent oxidation of palmitate plus malate by moth flight muscle mitochondria. Biochem. J. 110, 105-1 10. Stocker, R. F. (1974). Elektronenmikroskopische Beobachtungen iiber die Fusion myogener Zellen bei Antheraea polyphemus (Lepidoptera). Experientfa, 30,896-898. Stocker, R. F. and Nuesch, H. (1975). Ultrastructural studies on neuromuscular contacts and the formation of junctions in the flight muscle of Antheraea polyphemus (Lep.) I. Normal adult development. Cell Tiss. Res. 159,245-266. Stone, J. V., Mordue, W., Batley, K. E. and Morris, H. R. (1976). Structure of locust adipokinetic hormone, a neurohormone that regulates lipid utilization during tlight. Nature, Lond. 263,207-21 1.
1
226
ANN E KAMMER AND BERND HEINRICH
Sutherland, 0. R. W. (1969).,The role of the host plant in the production of winged forms by two strains of the pea aphid, Acyrthosiphon pisum. J. Insect Physiol. 15,2779-2201. Tager, H. S., Markese, J., Speirs, R. D. and Kramer, K. J. (1975). Glucagon-Eke immunoreactivity in insect corpus cardiacum. Nature, Lond. 254,707-708. Tager, H. S., Markese, J., Kramer, K. J., Speirs, R. D. and Childs, C. N. (1976).Glucagon-like and insulin-likehormones of the insect neurosecretory system. Biochem. J. 156,s15-520. Taylor, M. and Truman, J. W. (1974). Metamorphosis of the abdominal ganglia of the tobacco hornworm, Manduca sexta. J. Comp. Physiol. 90,367-388. Teutsch-Felber, D. (1970). Experimentelle und histologische Untersuchungen i n der Thoraxmuskulatur von Periplaneta americana L. Revue Suisse Zool. 77,481-523. Thomas, G. J. (1954). The post-embryonic development of the flight muscles of Lamarckiana sp. (Orthoptera) and a brief comparison of these with those of Saussurea stuhlmanniana (Karsch) and Tanita dispar (Miller). Proc. Roy. Enr. Soc. Lond. A, 23-30. Thomas, S. P. (1975).Metabolism during flight in two species of bats, Phyllostomus hastatus and Pteropus gouldii. J. Exp. Biol. 63,273-293. Thomson, S . N. and Bennett, R. B. (1971). Oxidation of fat during flight of male Douglas-fir beetles, Dendroctonus pseudotsugae. J. Insect Physiol. 17, 1555-1564. Tiegs, 0.W.(1955). The flight muscles of insects- their anatomy and histology; with some observations on the structure of striated muscle in general. Phil. Trans. Roy. Soc. Lond.
B238,221-348. Tietz, A. (1962).Fat transport in the locust. J. Lipid Res. 3,421-426. Tribe, M. A. and Ashhurst, D. E. (1972). Biochemical and structural variations in the flight muscle mitochondria of ageing blowflies, Calliphora erythrocephela. J. Cell Sci. 10,443-469. Truman, J. W.and Reiss, S. E. (1976). Dendritic reorganization of an identified motor neuron during metamorphosis of the tobacco hornworm moth. Science, 192,477-479. Truman, J. W. and Riddiford, L. M. (1974).Hormonal mechanisms underlying insect behaviour. In “Advances in Insect Physiology”, Vol. 10,297-352. Academic Press, New York and London. Tucker, V. A. (1970). Energetic cost of locomotion in animals. Comp. Biochem. Physiol. 34,
841-846. Tucker, V. A. (1973). Aerial and terrestrial locomotion: a comparison of energetics. In “Comparative Physiology: Locomotion, Respiration, Transport, and Blood” (Eds L. Bolis, K. Schmidt-Nielsen and S.H. P. Maddrell), pp. 63-76. North Holland Publ. Co., Amsterdam. Tucker, V. A. (1975). Aerodynamics and energetics of vertebrate fliers. In “Swimming and Flying in Nature” (Eds T. Y.-F. Wu,C. J. Brokaw, and C. Brennan), Vol. 2, pp. 845-867. Plenum, New York. Tung, A. S.-C. and Pipa, R. L. (1972). Insect neurometamorphosis. V. Fine structure of axons and neuroglia in the transforming interganglionic connectives of Galleria mellonella (L.) (Lepidoptera). J. Ultrastruct. Res. 39,556-567. Tyrer, N. M. and Altman, J. S. (1974). Motor and sensory flight neurones in a locust demonstrated using cobalt chloride. J. comp. Neurol. 157, 117-138. Van der Kloot, W. C. (1967). Membrane depolarization and the metabolism of muscle. Am. 2001. 7,661-669.
Van Handel, E. (1974).Lipid utilization during sustained flight of moths. J. Insect Physiol. 20,
2329-2332. Van Handel, E. and Nayar, J. K. (1972a).Tu-m-over of diglycerides during flight and rest in the moth Spodopterafrugiperda. Insect Biochem. 2,8-12. Van Handel, E. and Nayar, J. K. (1972b).Direct use of carbohydrates during sustained flight in the moth, Spodopterafrugiperda. Insect Biochem. 2,203-208. Vaughan, H., Thornton, S. D. and Newsholme, E. A. (1973).The effects of calcium ions on the activities of trehalase, hexokinase, phosphofructokinase, fructose diphosphatase, and pyruvate kinase from various muscles. Biochem. J. 132,527-535.
INSECT FLIGHT METABOLISM
227
Vejbjerg, K. and Normann, T. C. (1974). Secretion of hyperglycaemic hormone from the corpus cardiacum of flying blowflies, Calliphora erythrocephala. J. Insect Physiol. 20, 1189-1 192. Vepsaliiinen, K. (1974a). The wing lengths, reproductive stages and habitats of Hungarian Gerris Fabr. species (Heteroptera, Gerridae). Ann. Acad. Sci. Fenn. A , IV Biologica, 202, 1-
18. Vepsalainen, K. (1974b). The life cycles and wing lengths of Finnish Gerris Fabr. species (Heteroptera, Gerridae). Acta Zool. Fennica, 141, 1-73. Vogel, S. (1966). Flight in Drosophila. I. Flight performance of tethered flies. J. Exp. Biol. 44,
567-578. Vogel, S . (1967). Flight in Drosophila. 11. Variations in stroke parameters and wing contour. J. Exp. Biol. 46,383-392. Vogell, W. (1965). Phasen der Bildung morphologischer und enzymatischer Muster des Flugmuskels der Wanderheuschrecke. Naturwissenschajien, 52,405-418. Voss, F. (1911). Uber den.Thorax von GryNus domesticus. Die nach embryonale Metamorphose im ersten Stadium (Abdomen-Metathorax). Z. wiss. Zool. 100,589-834. Voss, F. (1912). Uber den Thorax von Gryllus domesticus. Die nachembryonale Metamorphose im ersten Stadium (Mesothorax-Kopf). Z. wiss. Zool. 101,445-682. Wajc, E. and Pener, M. P. (1971). The effect of the corpora allata on the flight activity of the male African migratory locust, Locusta migratoria migratorioides (R. and F.). Gen. comp. Endocr. 11,327-333. Walker, P. R., Hill, L. and Bailey, E. (1970). Feeding activity, respiration, and lipid and carbohydrate content of the male desert locust during adult development. J. Insect Physiol.
16,1001-1015. Weber, T. (1972). Stabilisierung des Flugrhythmus durch “Erfahrung” bei der Feldgrille. Naturwissenschajien, 59,366. Weis-Fogh, T. (1952). Fat combustion and metabolic rate of flying locusts. Phil. Trans. Roy. Soc. Lond. B237,l-36. Weis-Fogh, T. (1956a). Biology and physics of locust flight. 11. Flight performance of the desert locust (Schistocerca gregaria). Phil. Trans.Roy. SOC.Lond. B239,459-5 10. Weis-Fogh, T. (1956b). Biology and physics of locust flight. IV. Notes on sensory mechanisms in locust flight. Phil. Trans. Roy. Soc. Lond. B239,553-584. Weis-Fogh, T. (1961). Power in flapping flight. In “The Cell and the Organism” (MS. J. A. Ramsay and V. B. Wigglesworth), pp. 283-300. Cambridge University Press, London. Weis-Fogh, T. (1964a). Functional design of the tracheal system of flying insects as compared with the avian lung. J. Exp. Biol. 41,207-227. Weis-Fogh, T. (1964b). Biology and physics of locust flight. VIII. Lift and metabolic rate of flying locusts. J. Exp. Biol. 41,257-271. Weis-Fogh, T. (1967). Respiration and tracheal ventilation in locusts and other flying insects. J. Exp. Biol. 47,561-587. Weis-Fogh, T.(1972).Energetics of hovering flight in hummingbirds and in Drosophila. J. Exp. Biol. 5679-104. Weis-Fogh, T. (1973). Quick estimates of flight fitness in hovering animals including novel mechanisms for lift production. J. Exp. Biol. 59, 169-230. Weis-Fogh, T. (1975). Flapping flight and power in birds and insects, conventional and novel mechanisms. In “Swimming and Flying in Nature” (Eds T. Y.-F. Wu, C. J. Brokaw and C. Brennen), Vol. 11, pp. 729-762. Plenum, New York. Weis-Fogh, T. (1976). Energetics and aerodynamics of flapping flight: a synthesis. In “Insect Flight” (Ed. R. C. Rainey), pp. 48-72. Wiley, New York. Weis-Fogh, T. and Jensen, M. (1956). Biology and physics of locust flight. I. Basic principles in insect flight. A critical review. Phil. Trans. Roy. Soc. Lond. B239,415-458. Wiens, A. W. and Gilbert, L. I. (1967).Regulation of carbohydrate mobilization and utilization in Leucophaea maderae. J. Insect Physiol. 13,779-794.
228
ANN E. KAMMER AND BERND HEINRICH
Wigglesworth, V. B. (1949). The utilization of reserve substances in Drosophila during flight. J. Exp. Biol. 26, 150-163. Wigglesworth, V. B. (1965). “The Principles of Insect Physiology,” 6th Edn. 741 p. Methuen, London. Williams, C. M., Barness, L. A. and Sawyer, W. H. (1943). The utilization of glycogen by flies during flight and some aspects of the physiological aging of Drosophila. Biol. Bull. Woods Hole, 84,263-272. Wilson, D. M. (1962). Bifunctional muscles in the thorax of grasshoppers. J. Exp. Biol. 39,669677.
Wilson, D. M. (1968). The nervous control of insect flight and related behaviour. In “Advances in Insect Physiology”, Vol. 5,289-338. Academic Press, New York and London. Wilson, D. M. and Weis-Fogh, T. (1962). Patterned activity of co-ordinated motor units, studied in flying locusts. J. Exp. Biol. 39, 643-667. Wilson, D. M. and Wyman, R. J. (1963). Phasically unpatterned nervous control of dipteran flight. J. Insect Physiol. 9,859-865. Wohlgemuth, R. (1962). Die Schlagform des Bienenflugels beim Sterzeln in Vergleich zur Bewegungsweisebeim Fleigen und Facheln. 2. oergl. Physiol. 45,581-589. Wood, J. (1970). A study of the instantaneous air velocities in a plane behind the Wings of a certain Diptera flying in a wind tunnel. J. Exp. Biol. 52, 17-25. Wyatt, G. R. (1967). The biochemistry of sugars and polysaccharides in insects. In “Advances in Insect Physiology”, Vol. 4,287-360. Academic Press, New York and London. Wyman, R. (1970). Patterns of frequency variation in dipteran flight motor units. Comp. Biochem. Physiol. 35, 1-16. Young, E. C. (1965a). Flight muscle polymorphism in British Corixidae: Ecologicalobservations. J. Animal Ecol. 34,353-389. Young, E. C. (1965b). Teneral development in British Corixidae: Proc. Roy. Ent. SOC.Lond. A40, 159-168. Zarnack, W. (1972). Flugbiophysik der Wanderheuschrecke (Locusta migratoria L.) I. Die Bewegungen der Vorderflugel. J. Comp. Physiol. 78,356-395. Zarnack, W. (1975). Aerodynamic forces and their calculation in insect flight. In “Swimming and Flying in Nature” (Eds T. Y.-F. Wu, C. J. Brokaw and C. Brennan), Vol. 2, pp. 797-801. Plenum, New York. Zebe, E. (1954). a e r den Stoffwechsel der Lepidopteran. Z. uergl. Physiol. 36,290-3 17. Zebe, E., Delbruck, A., Biicher, T. (1959). a e r den Glycerin-I-P-Cyclus im Flugmuskel von Locusta migratoria. Biochem. 2. 33 1,254-272.
Ne uroethology of Acoustic Cornmunicat ion Norbert Elsner’ and Andrej V. Popov2
’ Zoologisches lnstitut der
Sechenov Institute of Evolutionary Physiology and Biochemistry, Leningrad, USSR
Universitat zu Koln, 5 Koln-Linderthal, Germany
1 Introduction 229 2 The neuronal basis of sound production 23 1 2.1 Behavioural and biophysical background 23 1 2.2 Neuromuscular basis of sound production 241 2.3 Central vs. peripheral control of sound production 25 1 2.4 The central nervous organization of sound production 260 3 The neuronal basis of sound reception and sound recognition 268 3.1 Innate releasing mechanisms 268 3.2 Sensory mechanisms of sound reception 28 1 3.3 Information processing by auditory neurons of higher order 296 4 Development of acoustic communication 3 16 4.1 Postembryonic development of sound production (larval “stridulation”) 3 16 4.2 Postembryonic development of hearing and innate releasing mechanisms 3 19 5 Genetics of acoustic communication 320 5.1 General ideas underlying a genetic approach 320 5.2 Song patterns of interspecific hybrids 321 5.3 Song-specific innate releasing mechanisms of interspecific female hybrids 326 6 Evolution of acoustic communication 329 6.1 Song-patterns and innate releasing mechanisms: Co-evolution or genetic coupling? 329 6.2 Evolution of song patterns 332 6.3 Evolution of auditory mechanisms 338 7 Concluding remarks 338 Acknowledgements 34 1 References 34 1
1
Introduction
Insects were among the first terrestrial animals that began to use acoustic signals for communication between conspecifics. This happened over 150-200 229
2 30
NORBERT ELSNER AND ANDREJ V. POPOV
million years ago during the Jurrassic period (Alexander 1966). Since that time their system of acoustic communication has progressively developed under the pressure of increasing complexity of the acoustic environment because of ( i ) speciation of other sound producing insects, (ii) differentiation of intraspecific communication, and (iii) emergence of other groups of animals (anurans, birds, and mammals) producing their own sounds and becoming predators of insects. Therefore, nowadays insects which rely on acoustic cues in looking for a mate must have highly specialized sensory systems capable of extracting the conspecific signal from the acoustic environment and for detecting predators. The latter task has become very important for sound producing insects because they demask themselves by their own sound emissions. Stridulatory organs have been developed by members of many insect orders, only a few of which can be mentioned here (review: Dumortier, 1963a). Orthopterans and Cicadidae are the best known singing insects, and most studies were concentrated for a long time on these groups. More recently a well-developed acoustical communication system was found also in fruit flies (Diptera) and significant progress was made in studying male courtship songs and the phonotactic reactions of the females (reviews: Bennet-Clark and Ewing, 1969, 1970; Bennet-Clark, 1971, 1975). Moths (Lepidoptera) seem to have no, or poor, intraspecific sound communication, but they have a very efficient auditory system for detecting the cries of hunting bats (review: Roeder, 1967, 1970). Finally, insects do not only use air-borne sounds: communication by substrate vibrations is frequently found in many groups such as ants and long-horn beetles (review: Markl, 1972, 1973). In this review we will mainly concentrate on Orthopterans (gryllids, tettigonioids and acridids) in an attempt to illustrate the basic principles of nervous mechanisms underlying acoustic behaviour. For comparison, where necessary, we will also refer to data obtained from other insect groups. We choose Orthopterans for this neuroethological approach, because their acoustic behaviour, the neural mechanisms of both sound production and sound reception, their development, genetics and evolution are better known than in all other insects. We have not attempted to present a general and comprehensive survey of all aspects of acoustic communication in orthopteran insects: but we believe that substantial neuroethological work is impossible without first knowing the behaviour of the animal, the physical parameters of acoustic signals, and the environmental and ecological conditions to which they are adapted. Therefore, in the following sections nonelectrophysiological investigations are also considered, but we have focused mainly on those species which are subjects of major neurophysiological studies. There are several books and reviews which cover the field, or parts of it, and a comprehensive book is in preparation which will cover not only Orthopterans but also other insect groups as well as additional aspects of their bio-acoustics
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
231
(Michelsen, Elsner and Popov, “Sound Communication in Insects”, Academic Press, London). The reader can find earlier literature in the following books and articles: “L’Acoustique des Orthopteres”, 1955; “Acoustic Behaviour o_f Animals”, 1963, both edited by R. G. Busnel; Bennet-Clark, 1971; 1975; Elsner and Huber, 1973; Haskell, 1961, 1964; Huber, 1970, 1975; Michelsen, 1971a,b,c; Michelsen and Nocke, 1974; Schwartzkopff, 1964; Frings and Frings, 1960; Worden and Galambos, 1972. We will concentrate mainly on new information which has appeared during the last five to ten years and on questions not covered by recent reviews or which received new interpretations in the light of neuroethology.
2
The neuronal basis of sound production
2.1 2.1.1
BEHAVIOURAL A N D BIOPHYSICAL B A C K G R O U N D
Stridulatory mechanisms
Insects have evolved a wide range of sound-producing mechanisms which have been described in detail elsewhere (Kevan, 1955; Dumortier, 1963a; Haskell, 1964; Bennet-Clark, 1971, 1975; Michelsen and Nocke, 1974). Among Orthopterans, most members of the sub-order Ensifera (Tettigonioidea and Grylloidea) use acoustic signals for intraspecific communication, whereas in the sub-order Caelifera (Acridoidea) sound production seems less common than is often thought (Uvarov, 1966). Neuroethological work has concentrated exclusively on wing and leg stridulation which are the most frequently, although by no means the only, methods found in orthopterans. These stridulatory techniques are commonly known, and, therefore, need to be explained only briefly: Ensifera. They use an elytro-elytral mechanism of sound production: a denticulated vein on the under-surface of one elytron is rubbed by the inside edge (plectrum)- of the other. In crickets both elytra have equally welldeveloped stridulatory files, although in most cases only the right file is used since usually the right elytron covers the left. In tettigonioids, the file of the right elytron (which is usually covered by the left) is normally smaller and less chitinized than the other. Sound is always produced by the closing movements, in some species (more common in tettigonioids than in gryllids), also during the opening, which, however, is normally of less intensity. Acridoidea. Some groups have developed a femuro-elytral method of stridulation: the hind-femora are equipped on their inner face with a row of small pegs which is rubbed against an elytral vein (vena radialis media). Sound is produced during both the upwards and the downwards movement, the latter being mostly more sonorous. It is frequently overlooked that this particular
232
NORBERT ELSNER AND ANDREJ V. POPOV
way of sound production is only found in the subfamily Gomphocerinae (also in Truxalinae and Emerogryllinae which, however, cover only few species). Oedipodinae have a reversed sound producing device: the hindfemura have a nondenticulated crest which is rubbed against a row of pegs (vena intercalata) on the elytra. Apart from this conventional method of leg stridulation, in several acridids sound production by the hindwings has been found, and has recently attracted neuroethological attention. Wing crepitation is well-known in oedipodine grasshoppers where it is thought to originate from the fan-like opening and closing of the hindwings (Isely, 1936). In some gomphocerine grasshoppers such as Stauroderus scalaris and Stenobothrus rubicundus another mechanism is used: the costal margins of the hindwings are strongly sclerotized and are vigorously beaten against each other both during flight and during courtship while the animal is sitting on the ground (see p. 334). 2.1.2
The physical parameters of orthopteran sounds
The physical characteristics of orthopteran sounds and the biophysics of their emission have been reviewed by Dumortier (1963b), Bennet-Clark (1971) and Michelsen and Nocke (1974). Looking in a somewhat simplified way on the sound emissions produced by the passing of the elytral or femural file over the plectrum or the vena radialis media, respectively, one is faced with two different types of sounds, according to the speed of the stridulatory movements and the resonator properties of the elytra. a “Resonant” sound emissions: In all crickets studied so far (for example Gryllus campestris, Nocke, 1971; Oecanthus pellucens, Busnel, 1955) and in several but not in all tettigonioids (for example: Drepanoxiphus modestus, Suga, 1966; Homorocoryphus nitidulus, Bailey, 1970; Neoconocephalus robustus, Morris and Pipher, 1972) the tooth impact rate almost perfectly corresponds to the natural frequency of the elytral resonator. The result is a sound emission consisting of a continuous train of sine waves, the number of which accords to the number of stridulatory teeth used (Fig. la). In other words: each cycle (wave) is reinforced by the next tooth impact before any damping can set in. The frequency spectra of these songs are characterized by sharp peaks marking the carrier frequency. They depend upon the physical properties of the sound emitting structures, especially the size and the elasticity of the elytral resonators (for details see: Nocke, 1971; Michelsen and Nocke, 1974). For the calling song of Gtyllus campestris this peak frequency is 4.5 kHz, small supplementary peaks are found at the first and second harmonics (the latter is also the peak frequency of the courtship song). The song spectrum, normally, has generic specificity (Dumortier, 1963b; Haskell, 1964; Popov et al., 1974b)
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
233
and is often very similar even in closely sympatric species (Fig. 2), although there are some exceptions: The carrier frequencies of the two European mole crickets Gryllotalpa gryllotalpa and G. vinea are 1.6 kHz and 3.5 kHz, respectively (Bennet-Clark, 1970); those of the American mole crickets Scapteriscus acletus and S. vicinus are 2.6 kHz and 3.2 kHz, respectively (Ulagaraj, 1976). These sympatric species, therefore, have a further possible parameter for intraspecific recognition besides the temporal patterning of the songs. The tettigonioids emitting “resonant” sounds have a higher carrier frequency than is found in the calling songs of most crickets (for example: Neoconocephalus robustus: 10 kHz; Homorocoryphus nitidulus: 16 kHz; Drepanoxiphus modestus: 22-24 kHz). b “Non-resonant” sound emissions: In many other orthopterans the tooth impact rate is much lower than the carrier frequency; i.e. during one movement of the pars stridens a series of short impulses and not one long-lasting sinewave is produced. Each impact of a stridulatory peg or tooth at the vena radialis media (acridids) or at the plectrum (tettigonioids) elicits a heavily damped oscillation (Fig. 1b). These highly-transient, rapidly decaying wave trains can easily be identified in the oscillograms and have been called “pulses” (Broughton, 1963), “impulses” (Elsner, 1974a), or “intrasyllabic spikes’’ (Lewis et al., 1971)’ This type of sound emission is found in all acridid grasshoppers studied so far (Gomphocerinae: Elsner, 1974a) and in many tettigonioids (Listrocaelinae and Conocephalinae: Suga, 1966; Tettigoniidae and Decticinae: Broughton et al., 1975; Ephippigeridae: Busnel, 1955). The spectrograms usually show a ’The term “pulse” has produced some confusion in the terminology of insect songs since several glossaries exist which partly contradict each other (for example: Jacobs, 1953; Faber, 1953; Alexander, 1957; Dumortier, 1963b; Broughton, 1963; Elsner, 1974a). Most of the confusion arises from the different use of the term “pulse”. Some authors define it as a biologicalparameter and call the sound emission resulting from a passing of the stridulatory file over the plectrum or the vena radialis a ‘‘pulse” (Dumortier, 1963b; Helversen, 1972). Others define the term as an acoustic parameter and call a train of sine waves characterized by a certain rise and decay a “pulse” (Broughton, 1963; Moms and Pipher, 1972) or “impulse” (Elsner, 1974a). By the latter definition, sound produced by one wing-stroke of a cricket is undoubtedly a “pulse” (in coincidence with the biological definition) but the sound emissions elicited by one leg stroke of a grasshopper would have to be called a “train of pulses” (Elsner, 1974a: “syllable”). In acridid grasshoppers a particular problem arises if the term “pulse” is used as a physical parameter: At the end of a long song and/or when temperature is going above 30’ C the tooth impact rate may oRen reach the natural resonance frequency. That means that one movement of the pars stridens may elicit a “series of pulses” (“syllable”) or one “pulse” only. For most practical purposes it is, certainly, inconvenient to use different terms for sound emissions produced by homologous movements. As far as song elements of higher order are concerned the situation is absolutely hopeless. Terms such as “chirp”, “echeme”, “trill”, “phonatone”, “phrase”, “note”, “syllable”, “buzz”, “sequence”, “strophe”, “verse”, etc. can often be defined only for particular cases. It is much too early to suggest a better terminology, and, therefore, in the present review the terms are generally used as the authors did in their original publications, explaining their meaning where necessary and possible.
Omocestus viridulus
lOOms
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
235
very complex frequency distribution, which in some cases covers nearly 100 kHz (Dumortier, 1963b). c Frequency and impulse-rate modulation: In most cases there is no frequency modulation within the song, although there are some exceptions, especially in the courtship songs of gryllids, but also in the calling songs of some tettigonioids (Broughton et a[., 1975; Pipher and Morris, 1974; Morris and Pipher, 1972) and Cicadidae (Popov, 1975). This frequency modulation may be linked with the structure of the stridulatory file, with the change of action of the sound producing organ, with the varying of the stridulatory speed or with the activation of another organ. For example, the decticine Metrioptera sphagnorum produces different frequency spectra, alternating between the “resonant” and the “nonresonant” type of sound emission (each sequence lasts 250 ms). The first part of the song is characterized by an ultrasonic dominated spectrum with a carrier frequency of 33 kHz. Tooth impact rate and elytral resonance frequency coincide. The second part has an audio-dominated spectrum with the most intense frequencies between 15 and 20 kHz. It consists of a train of rapid-decay “impulses” each elicited by the impact of one tooth. Different nonoverlapping file regions are used for the generation of the audio and the ultra-sonic songparts (Morris, 1970; Morris and Pipher, 1972). The highly transient “impulses” characterizing the songs of acridids and many tettigonioids have recently attracted attention as a further speciesspecific acoustic parameter: the impulse rate is modulated to a large extent by the varying stridulatory speed. To give two examples: In the decticine Platycleis intermedia it ranges between 0.4 and 1.6 kHz Broughton et al., 1975) and in the gomphocerine grasshopper Omocestus viridulus it varies from 0.25 to 4 kHz (Elsner, 1974a). Broughton ef al. (1975) have shown that this modulation is represented in the sonagrams as a clearly visible low-frequency sweep (Fig. lc). This component has long been overlooked since students of orthopteran sounds usually have reduced the playback speed considerably in order to analyse the high-frequency parts of the spectrum (which otherwise Fig. 1. Stridulatory files of orthopterans and micro-structure of the songs. (a) Scanning electrons micrograph of the elytral file and oscillogram of the initial part of a sound pulse (calling song) of the cricket Gryllus cumpestris. Each wave is elicited by the impact of a stridulatory tooth and, thus, the emitted sound frequency corresponds well to the tooth impact rate. (b) Scanning electron micrograph of the femoral file and oscillogram of sound emission (part of a downstroke “syllable”) of the acridid grasshopper Stuuroderus scalaris. Each impact of a stridulatory tooth elicits a highly transient “impulse” which rapidly decays before the next impact. (c) Part of the courtship song of the acridid grasshopper Omocesfus viridulus. The oscillogram (top) illustrates the impulse rate modulation which is expressed as a low-frequency sweep in the sonagram (bottom). Courtesy of E. Eibl (a, top); P. van Dohlen (a, bottom); S. Koppers (b), and D. B. Lewis (c, bottom). (c, top) from Elsner (1974a).
236
NORBERT ELSNER AND ANDREJ V. POPOV
would be lost, since the upper frequency limit of common sonagraphs is only 8 kHz). By doing so the low-frequency sweeps, expressing the tooth-impact rate, inevitably were lost in the background noise. 2.1.3
Stridulatory movements and sound patterns
Although the frequency spectrum might not be as meaningless for intraspecific recognition as has long been assumed, the temporal pattern (amplitude modulation) is much more conspicious. Interspecifically it is the most variable parameter of insect songs. Proof of the species-specificity of the amplitude modulation patterns of insect songs have been obtained for all the groups studied (Dumortier, 1963b; Haskell, 1964; recent papers: Gryllidae: Walker, 1964, 1969a,b; Popov, 1972; Popov et al., 1974a; Tettigonioidea: Dubrovin and Zhantiev, 1970; Spooner, 1964; Walker and Gurney, 1972; Walker et al., 1973; Walker, 1975; Acrididae: Helversen, 1972; Elsner, 1974a; Cicadidae: Popov, 1969, 1975; Young, 1972, 1973; Diptera: Ewing and Bennet-Clark, 1968). In many orthopterans the song patterns have now been described on three different levels: (3 sounds, (ii) stridulatory movements and (iii) central motor patterns underlying stridulation. In the following, the song patterns of crickets, tettigonioids and acridid grasshoppers are briefly described on the first or the first two levels, concentrating on those species which have also been studied neurophy siologically. The need for recording not only the sound pattern but, in addition the singing movements, is obvious: (i) most species stridulate at a fast rate and individual sound emissions cannot easily be associated with certain phases of wing or leg movements. (ii) Furthermore, it is impossible to deduce the complex structure of the movements from sound recordings since the zeropoint of the movement often changes during the performance. The complex superimposition of various rhythms can only be revealed by recordings of singing movements; this is most important for avoiding false homologies. High-speed photography has been used to record singing movements of crickets and tettigonioids (Pasquinelly and Busnel, 1955; Walker et al., 1972; Walker, 1975) and of acridid grasshoppers (Loher and Huber, 1964). More recently, new methods have been introduced which make it possible to record the movements instantaneously. First, Hall-generators (magnetive sensitive semi-conductors) were mounted on the back of the animal and driven by a small magnet attached to the pars stridens (grasshoppers: Elsner, 1970,1974a; crickets: Innenmoser, 1974). This method has now been replaced completely by an opto-electronic device: a tiny piece of reflective sheeting is attached to the stridulating leg or wing and illuminated via a semi-reflective mirror. The reflected light impinges on the surface of a position-sensing photodiode, i.e. the
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
237
stridulatory movements can be recorded instantaneously without loading the animal with the recording device (Helversen and Elsner, 1977). The most important advantage of this method is the fact that it is now possible in acridids to record the singing movements of both hindlegs simultaneously. a Song patterns of crickets The species-specificityof amplitude modulation is illustrated in Fig. 2 which shows the male calling songs of several sympatric cricket species from Azerbaijan, USSR.It can be seen that these songs differ in complexity and, therefore, by the number of time parameters which are potentially available to the song recognizing system. The most primitive amplitude modulation pattern is a continuous trill as found in Gryllodinus odicus, consisting of equal sound pulses repeated for seconds or even minutes with a highly stable rate. A first step in increasing the complexity of the calling song is done by a modest amplitude modulation leading to the waxing and waning of the trill (for example: Pteronemobius heydeni). In many species this amplitude modulation is enhanced in such a way that the trill is broken into more or less stable groups of pulses, usually called “chirps”. These are separated by much longer intervals than those between “pulses” (see footnote, p. 233). In species such as Oecanthuspellucens the chirps are not yet stabilized in duration in contrast to species like Gryllus campestris and Tartarogryllus tartarus. The biological relevance of these time paramefers for species recognition will be discussed in detail below (see p. 268). A few remarks should be made on the songs of the European field cricket Gryllus campestris which is the most thoroughly studied singing insect (reviews: Huber, 1970, 1975). In this species, in addition to the sounds, the stridulatory movements have also been recorded recently (Innenmoser, 1974). The acoustic repertoire contains three different songs: ( i ) Calling song: composed of chirps (following each other at a rate of 3 - 4 Hz) each consisting of 3-4 pulses (called “syllables” by some authors) produced during the inward movement of the elytra (the outward movement is silent). Sound intensity increases from the first to the last pulses (Fig. 2, see also Fig. 12a). (ii) Rivalry song: consists of long sequences of pulses following each other at the same rate as during a calling song (ca. 30 Hz) but they are not arranged into short chirps; sound intensity is considerably increased (Fig. 12a). (iii) Courtship song: this song is completely different from the other two: it consists of “tick” sounds following at intervals of ca. 300-350ms during which extremely faint “Zwischensilben” are produced (Figs 5a, 12). b Song patterns of tettigonioids Neurophysiological investigation of tettigonioid sound production has been carried out to date in only one species, Neoconocephalus robustus (Josephson and Halverson, 1971, see p. 249) which stridulates at an unusually high rate of 15’&200 wing strokes per second.
238
NORBERT ELSNER AND ANDREJ V. POPOV
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
239
Walker and Dew (1972), Walker (1975), Suga (1966), Morris (1970), Morris and Pipher (1972), Bailey and Robinson (1971) and others have analysed the song structure of various other species which often produce very complex movement patterns. c Song patterns of acridid grasshoppers The sound production of acridid grasshoppers has been described in detail by Faber (1953), Jacobs (1953), and Otte (1970). More recent investigations on individual species have been performed by Otte (1972), Loher and Chandrashekaran (1972), and Elsner (1974a,b); female sound production (which is more common than in crickets and tettigonioids) has been studied by Loher and Huber (1964) and Kohsen (1976). Neurophysiological work has been restricted so far to the Gomphocerinae, the males of this sub-family produce most elaborate calling and courtship songs. These performances have attracted neurobiological attention for two reasons. (i) In contrast to crickets and tettigonioids sound is produced by two stridulatory elements simultaneously, i.e. the two hindlegs each rubbing the stridulatory file against the ipsilateral elytron. (ii) In several species, hindleg stridulation is accompanied by movements of other parts of the body, for example the head, the antennae, the palps and the wings. Therefore, gomphocerine grasshoppers are excellent subjects for hvestigations of the intra- and intersegmental organization of invertebrate behaviour. The stridulatory movements of many European grasshoppers have now been recorded using Hall generators (Elsner, 1970, 1974a) and the new optoelectronic device (Halfmann, Elsner and Helversen, unpublished). These recordings have revealed an enormous variety of stridulatory patterns ranging from “simple” up and down strokes (for example: Stenobothrus lineatus) to complex structures with several layers of rhythmicity superimposed upon each other (for example: Stauroderus scalaris). Some examples are given in Fig. 3. In most species studied so far the movements of the two hindlegs are phaseshifted, one hindleg leading the other with a fixed latency. The lead changes abruptly between the two legs from one sequence to the next. But even if, occasionally, a change occurs during the sequence, the transition is not gradual. As a consequence of this phase-lag the elaborate sound patterns produced by each hindleg are partly camouflaged. In particular the clear structure of upstroke and downstroke “syllables” is erased in many cases (Elsner 1974a, see Fig. 18, insert). Surprisingly, several gomphocerine grasshoppers have now been found to produce direrent patterns with each hindleg (Figs 3, 36, 38). In Omocestus viridulus, one hindleg performs the upstroke without interruption, although slowing down the speed of the movement during the middle part (pattern I), Fig. 2. Oscillograms and frequency spectra of the calling songs of various sympatric crickets native to the southern regions of Azerbaidjan, USSR.From Popov, unpublished.
Fig. 3. Sound patterns and stridulatory movements of acridid grasshoppers (Gomphocerinae).Top and middle traces: stridulatory movements of the left and right hindleg, respectively. Bottom traces: sound pattern produced by the right leg only (both hindlegs are intact, but the left tegmen has been removed in order to prevent superposition of two sound emissions). From Halfmann and Elsner, unpublished.
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
24 1
while the other leg (stridulating also with much lower amplitude) interrupts the upward movement (pattern 11). By doing so, the remarkable impulse-rate modulation is preserved even in the sounds of animals stridulating with both hindlegs (Fig. 3). In Chorthippus mollis one hindleg begins each chirp with an accentuated downstroke followed by a vibratory up and downward movement (pattern I). The contralateral leg omits the initial downstroke, performing only the vibratory movement (pattern 11, see Fig. 36). In Chorthippus biguttulus the chirps produced by one hindleg are clearly separated from each other by pauses, during which the leg is kept in its upper position (pattern I). The other hindleg sonorously extends the last upstroke of each chirp in such a way that the clear chirp separation is erased (pattern 11, Fig. 36). In all three species the leg producing pattern I leads the other one; the legs change their role from time to time. These characteristic features of grasshopper songs have attracted attention from both neuronal and behavioural aspects. From a neuronal point of view, the bistability of the stridulatory patterns, and its central and peripheral control, is one of the most interesting attributes of gomphocerine sound production. It will be discussed in detail in one of the following sections (see p. 257). Furthermore, the phase lag of the two hindlegs and the simultaneous production of different stridulatory patterns has to be discussed in the light of present theories on pattern recognition and innate releasing mechanisms (p. 275). 2.2
NEUROMUSCULAR BASIS OF S O U N D P R O D U C T I O N .
Graham Hoyle (1970) has pointed out the three different ways in which the neuroethologist may study the neuronal mechanisms underlying behaviour. He may (i) start at the periphery, recording the central motor pattern sent to the effector organs, (ii) directly attack the central nervous system to investigate the generation of this pattern and (iii) approach via the sense organs, studying the changes of the motor program after sensory interaction. The following three sections will show the extent to which these different approaches have been successful in revealing the neuronal basis of sound production in orthopteran insects. We shall start with the analysis of the motor output pattern deduced in an indirect way, although the direct approach to the central nervous system inaugurated by Franz Huber in the fifties is much older. But for the purpose of this review it may be more advantageous to first of all describe the behaviour in terms of neuronal activity, before discussing its central nervous generation. 2.1
Motoneuronal activity and electro-myography
The central nervous output patterns underlying arthropod behaviour can easily be monitored indirectly in freely moving and normally behaving animals by
242
NORBERT ELSNER AND ANDREJ V POPOV
means of electrodes chronically implanted into the muscles. The motoneuronal activity on which muscle excitation is based can be deduced from such recordings, since a fixed 1 : 1 relationship exists between the action potentials of muscle fibres and the motoneurons driving them (Hoyle, 1957). Of course, only information about motoneuronal action potentials is forthcoming in this way. No references about preceding synaptic events can be made. Electromyograms have now been used to record the motor patterns underlying various behaviours in insects (for example: walking: Hoyle, 1964; Delcomyn, 1973; Pearson, 1972; flight: Wilson and Weis-Fogh, 1962; Wilson, 1968; sound production in cicada: Young, 1972). Studying orthopteran stridulation in this way was started by Huber (1965) and continued (i) in crickets by Ewing and Hoyle (1965), Bentley and Kutsch (1966), Kutsch (1969), Kutsch and Huber (1970), Weber (1974), and Innenmoser (1974), (ii) in tettigonioids by Josephson and Halverson (1971) and (iii) in acridid grasshoppers by Elsner (1967, 1968, 1973, 1974b, 1975), Elsner and Huber (1969), Kohsen (1976) and Koppers (1977). Analysing the motor output underlying stridulation is considered as the first step towards an understanding of the pattern generating neuronal mechanisms. Also, electromyograms provide an excellent basis for describing the songs, since clear activity patterns can be recorded, even at times when only slight movements occur which may be silent. Finally, recording the neuromuscular activity leading to a rich variety of movement patterns, reveals new aspects of functional morphology. 2.2.2
Technique and selectivity of electromyograms
Insulated steel or silver wire electrodes (020-40 pn) have been used for these recordings which can be performed simultaneously on many channels and displayed together with the monitored sound and the stridulatory movements. Even in small gomphocerine grasshoppers (length 15-20 mm) it has been possible to implant up to 30 electrodes into the various pterothoracic muscles without interfering with the behaviour of the animals (Elsner, 1974a). Since these wires were very flexible the insects could walk freely; the range of movements was, of course, limited by the length of the leads (50-70cm). Studying long-term effects following deprivation experiments (see p. 257) the motor pattern underlying stridulation has been monitored over periods of up to 6 weeks via the same chronically inserted electrodes (Elsner, Hirth and Lindberg, unpublished results). In most cases, individual motor units of muscles could be identified confidently since the area from which noticeable activity was recorded turned out to be small (Elsner, 1967, 1974, see Fig. 4 for further explanation). Given this surprisingly high resolution capability, sound production of orthopteran insects could be described in terms of the activity of individual neuromuscular units.
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
243
Fig. 4. Electromyographic recordings from the right subalar muscle of the grasshopper Chorthippus biguttulus during courtship song. (A) Electrical muscle activity (top trace, recorded near site 2) and sound pattern (bottom trace). (B) Identification of individual motor units by simultaneous recordings from the anterior (site 1) and the posterior part (site 2) of the subalar muscle. Each electrode monitors mostly the activity of the unit it is inserted in; the potentials originating from the other unit are recorded with low amplitudes (arrows). 129, subalar muscle; 120, 2nd posterior tergocoxal muscle; SaS, subalar sclerite; PL, pleural ridge. From Elsner (1967,1975, modified).
2.2.3
Co-ordination of motor activity
All behaviour results from motor activity which is specifically co-ordinated at various functional levels and which can be considered to be arranged hierarchically (Weiss, 194 1). In orthopteran stridulation these levels are (i) motor units of the same muscle, (ii) the set of antagonistic and synergistic motor units moving one hindleg or elytron and (iii) the homologous sets of units belonging to the left and right hindleg or elytron, respectively. Using electromyography, the motor patterns underlying stridulation have been analysed on these different levels.
a Motor units belonging to the same muscle The behavioural characterization of single motor units has been investigated in detail in gomphocerine grasshoppers (Elsner, 1967, 1968, 1975). Each unit was found to be individually characterized by a discrete type of activity: (i) phasic: giving 1-3 action potentials per leg or wing stroke and being silent for the rest of the cycle, (ii) bursting: the activity covering exactly the duration of the movement it belongs to and (iii) phasic-tonic: being continuously active, the rate of
244
NORBERT ELSNER AND ANDREJ V. POPOV
activation being increased and decreased rhythmically according to the song pattern. Both in crickets (Kutsch, 1969) and in grasshoppers (Elsner, 1967, 1968) the individual motor units of a given muscle were always found to be activated in a certain order, which was the same in all species and in all types of behaviour in which they are involved (stridulation, walking, flight). For example, in muscles which have both bursting and phasic units the latter were always seen to be activated after the former. In muscles having two phasic units their order of activation was also strictly fixed. This co-ordination of motor units could easily be measured by inserting more than one electrode in the same muscle. It was demonstrated that specific features of the sound patterns (for example: length and intensity of the emitted sounds) were already reflected by the motor unit co-ordination. Optimal synchronization resulted in short lasting and fast movements, producing intense sounds; poor co-ordination led to long-lasting and slow movements, producing weak sounds. According to the specific sound patterns, the coordination was often found to be increasing and decreasing rhythmically (Elsner, 1974a). b Motor co-ordination at the level of muscles Electromyograms recorded simultaneously on many channels have revealed neuromuscular activity patterns of surprising complexity. These motor scores are often far from being “simple” analogues of the stridulatory movements. Certainly, for any muscle involved in stridulation, very precise temporal relationships exist between its individual activation and the recruitment of other muscles, as well as to the corresponding movement and sound emission. Each muscle, however, has to be characterized individually both as far as its recruitment and co-ordination among the whole set is concerned as well as in its relationship to the stridulatory movements. Classifying muscles as “synergists” and “antagonists” becomes extremely dubious when their co-ordination during the performance of dzyerent song patterns is investigated. Therefore, motor activation at the level of muscles has to be considered under the aspects of both neuronal coordination and functional morphology. The “personality” of individual muscles (better: individual sets of motoneurons driving a given muscle) is already evident when their order of activation is studied. As in the case of individual motor units, the muscles as a whole are always recruited in a strictly stereotyped manner. This order of activation is the same in all species, regardless of the specific motor patterns underlying stridulation (Elsner, 1968, 1975). It has never been observed that muscles so far regarded as “synergists” by the anatomists (for example first and second posterior tergocoxal muscles) change their order of recruitment. Considering the co-ordination of muscle activity in tettigonioids
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
245
(Neoconocephalus robustus: Josephson and Halverson, 197 1) and in crickets (Gryllus campestris, G. bimaculatus, Acheta domesticus: Bentley and Kutsch, 1966, Kutsch, 1969; Weber, 1974; Innenmoser, 1974) the situation at first sight appears simple, not justifying the introductory remarks just made: Considering the calling and rivalry songs, two clearly separated sets of antagonistic muscles were found active at the times expected: (i) the opener muscles (among them the basalar and the subalar muscles) being recruited at the end of the inward movement of the elytra and (ii) the closer muscles (among them the anterior and posterior tergocoxal muscles) being activated at the end of the outward movement. The co-ordination within each set appears rather simple since only very slight phase shifts were found. The two antagonists are firmly phase-locked, the interval between the activation of the opener and closer group is held constant, regardless of the increasing pulse intervals (Kutsch, 1969; Innenmoser, 1974; Fig. 5a). During cricket flight (its relation to song will be discussed later, see p. 332) the opener and closer muscles are working as depressors and elevators in a similarly antagonistic manner. Little is known about the mechanisms needed for the transition from opening and closing to depression and elevation. Small auxillary muscles such as the pleuro-alar muscle 85 are thought to serve this function. Recordings show a bursting activity at certain phases of the stridulatory cycle which might help to keep the elytra in an elevated position during singing and, thus converting elevation and depression into closing and opening movements respectively (Kutsch, 1969). Extremely complex motor patterns underlying stridulation have been found in gomphocerine grasshoppers (Elmer, 1968, 1975). The co-ordination between the muscles cannot be condensed into a basic scheme generally valid for all stridulatory patterns. This becomes especially evident when the neuromuscular activity patterns of slowly and rapidly singing species are compared. As mentioned earlier, the duration of individual up- and downstrokes differ widely from one species to the next, ranging from 5 ms’in Chorthippus mollis to about 500 ms in Stenobothrus lineatus. The motor patterns of fast stridulating species (basic frequency ca. 50 Hz: Gomphocerippus rufus, Chorthippus biguttulus, Ch. mollis) appear as rather perfect analogues of the corresponding song patterns. Two sets of antagonistic muscles are coupled in a simple and latency locked, unimodal manner. The pattern of upward and downward movements can rather easily be predicted from this clear alternating recruitment of the two muscle sets. The muscles belonging to each set (“synergists”) are not as well synchronized as has been found in crickets and tettigonioids. However, in no cases does “jittering” occur: the phaseshifts of 1-3 ms are fixed. With respect to the length of the period (ca. 20 ms) these delays are small enough to justify the impression of two antagonistically working muscle sets (Fig. 6a).
NORBERT ELSNER AND ANDREJ V. POPOV
246
m c o u r t s h i p Song
I
G ryl lus bimacu latus
5 0 rnl
Stauroderus scalaris Fig. 5. Simultaneous recordings of sound patterns, stridulatory movements, and neuromuscular activity during calling and courtship song of the cricket Gryllus birnaculatus (a) and the grasshopper Stauroderus scalaris (b). In (a) a downward deflection on the 2nd traces denotes a closing and an upward deflection an opening movement of the elytra. In (b) upward and downward deflections correspond to up and downstrokes of the hindlegs (RHB, LHB-right and left hindleg, respectively). 90, mesothoracic posterior tergoxoxal muscle (wing closer); 99, mesothoracic subalar muscle (wing opener); 128, metathoracic 2nd basalar muscle (leg depressor according to insertion); 129, subalar muscle (leg elevator according to insertion). Note that in this case the muscles 128 and 129 (which are “antagonists” according to their anatomical insertion at the anterior and posterior coxal rim) are synchronously recruited. Further details in the text. (a) from Innenmoser (1974); (b) from Koppers (1977).
In all slowly singing grasshoppers (for example: Stenobothrus lineatus: 1.1 Hz;Gomphocerus sibiricus: 5 Hz;Omocestus uiridulus: 14 Hz) motor coordination is more complex and at certain times contradicts the concept of synergistic-antagonistic relationships based on anatomical studies. The characteristic feature of these motor patterns is a bimodal relationship between
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
247
Fig. 6. Motor scores underlying sound production of a fast (a) and a slowly (b) singing acridid grasshopper. The activity patterns of the stridulatory muscles are displayed against the background of the stridulatory movements they initiate. The changing number of activated motor units belonging to a given muscle is symbolized by different screen densities of the “notes”, each of which illustrates the recruitment of one or more motor units. 120, 129, 119, 132, elevators; 126, abductor; 125, 118, 128, 133, depressors of the hindleg. For further explanation see Elsner (1975).
muscle sets associated with up- and downstroke, respectively. Surprisingly, rather few muscles are recruited at the beginning of the upward movement; an additional and much stronger activation of these units and others (not activated at the beginning) is found at the end of the upstroke, sometimes even overlapping with the recruitment of the downstroke muscles. In Omocestus
248
NORBERT ELSNER AND ANDREJ V. POPOV
viridulus the downstroke is based on a similar scheme of muscle activation: the most powerful muscles are recruited only when the leg has already been lowered half-way (few of them also at the beginning of the downward movement). In Stenobothrus lineatus, certainly, the maximum of depressor activity is found a few ms before the start of the downstroke (as in fast singing species). However, strong units of at least one depressor are additionally activated at the end of the movement, i.e. immediately before the beginning of the upstroke (Elsner, 1975; Fig. 6b). A most striking neuromuscular activity pattern has been found underlying sound production in the grasshopper Stauroderus scalaris. Courtship song is composed of “chirps” each consisting of a vibratory downstroke followed by an upward movement performed in steps (Fig. 5b). During the downstroke presumably “slow” motor units of elevator and depressor muscles are antagonistically activated as expected. During the upstroke, however, strong “fast” units of anatomically antagonistic muscles (for example: second basalar and subalar muscles) are recruited precisely synchronously. Antagonistic activity, although thoroughly looked for, has not been observed (Koppers, 1977). Some aspects of functional morphology: The motor scores of slowly stridulating grasshoppers illustrate most obviously the difficulty of deducing the singing movements from the neuromuscular activity patterns alone. This difficulty stems from our poor understanding of the mechanics of leg and wing movements. Although less spectacular, the problem exists also in crickets and fast stridulating grasshoppers, the motor patterns of which have been somewhat oversimplified so fat. For example, looking more closely at the neuromuscular activity underlying cricket calling song one observes the openers being recruited later and later from one pulse to the next within a chirp. This means that the exact duration of the opening and closing phase cannot be deduced from measuring the corresponding intervals of neuromuscular activity (Kutsch, 1969). In all these cases, the poorly understood elastic characteristics of the wing and leg joints have also to be taken into account. Considerations from this point of view have been made by Innenmoser (1974) for the courtship song of crickets. As described above (see p. 237) this song is characterized by the socalled “tick”-sounds produced by fast closing movements of the elytra. Strong phasic closer muscle activity accounts for this movement. After this “active” closing the elytra are “passively” opened about half-way before any opener activity comes in to complete this movement. During the long inter-“tick” intervals the elytra are kept open; strong opener activity is observed which obviously works against the elastic forces tending to move the elytra inwards again. Consequently, only low-amplitude vibrations are produced, leading to almost inaudible sounds (Innenmoser, 1974; see Fig. 5a).
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
249
Only a few investigations of the contraction kinetics of stridulatory muscles have been performed so far. The best analysed examples are the mesothoracic muscles of the two tettigonioids Neoconocephalus robustus and Euconocephalus nasutus. These are extremely fast-stridulating species, rubbing their elytra together at rates up to 200 Hz. The contraction kinetics, ultrastructure, and the neural control of these extraordinary muscles has been thoroughly investigated by Heath and Josephson (1970); Josephson and Halverson (1 97 1); Elder (1 97 1); Josephson (1973); Josephson et al. (1975). Surprisingly, even these fast-contracting muscles have a neurogenic rhythm, i.e. each contraction is associated with a preceding muscle action potential as in all other Orthopterans. The high contraction rate can only be achieved because these muscles work at elevated temperatures of 33-36" C, i.e. 12O C higher than the environment. In order to reach this high body temperature, singing is preceded by a warm-up, during which antagonistic muscles are contracted synchronously, producing heat but no obvious movement (for further discussion on warm-up in insect muscles see the review by Kammer and Heinrich in this volume). At 35' C the fusion frequency is greater than 400 H z , whereas it is only 150 Hz at 25O C. Therefore, at higher temperatures, the muscles contract in an unfused tetanus and relax about half-way between the tension peaks when stimulated at stridulation rates (Josephson, 1973). c Contralateral co-ordination Ensifera: The two elytra of stridulating crickets and tettigonioids operate as a functional unit. Thus, the left and right neuromuscular sub-systems are almost perfectly synchronized: contralaterally homologous muscles are activated at the same time, even as far as the number of recruited units and their co-ordination are concerned (Kutsch, 1969; Fig. 7a). This synchronization of the two contralateral sets has led to the suggestion that corresponding motoneurons might be directly (electrical?) coupled. Recent neuroanatomical studies disprove this hypothesis: as in locusts the dendritic arborization of cricket motoneurons is restricted to the ipsilateral half of the ganglion (Elepfandt, 1975). Therefore it seems more likely that the two contralateral sets are driven by a common interneuron(s).
Acrididae: Grasshoppers have two morphologically and functionally separate stridulatory organs, i.e. the two hindlegs each rubbing its file against the ipsilateral elytron. Although, theoretically, these two instruments might act independently from each other, they are as strongly coupled as the elytra of stridulating crickets. However, synchrony between the left- and right-side has been found in very few cases only. In most species the two contralateral . neuromuscular systems are activated with a distinct phase-lag, i.e. one leg is leading the other. The essential feature of this co-ordination is its multistability: the lead may change abruptly from time to time. Normally, the co-ordination is
2 50
NORBERT ELSNER AND ANDREJ V. POPOV
9OL
&
V
(a) 9 0 R
Ih/
Gryllus campest ris
I
Fig. 7. Co-ordination of contra-laterally homologous muscles during sound production. (a) Synchronous activation of the left and right posterior tergocoxal muscles (90, wing closer) during calling song of the cricket Gryllus campestris. (b) Multistable co-ordination of the left and right subalar muscles (1 29, elevator of the hindleg) during courtship song of the grasshopper Gomphocerippus mfus.Due to the change of lead which may occur between the left and right
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
251
kept stable between consecutive “chirps” and the switch occurs from one song sequence to the next. A more frequent rearrangement has been found in the grasshopper Gomphocerippus rufus. At each “chirp” the lead may be changed at the first two cycles of elevator and depressor activity. Consequently, 4 different modes of co-ordination can be observed: (i, ii) the right- or the leftside leading at both cycles in question and (iii, iv) the right- (left-) side leading at one cycle but being delayed at the other (Elsner, 1968; Fig. 7b). Apart from Omocestus viridulus (see below), in all grasshopper species studied so far, the two ipsilateral antagonistic sets of motor units (i.e. elevators and depressors) are tightly coupled in such a way that either both are leading or both are phase-lagged with respect to the corresponding contralateral sets. In 0. viridulus the strong mutual couplings are expressed in a reciprocal manner: if on one side the elevators lead the contralaterally homologous set, the depressors are always delayed and vice-versa. Consequently, on this side, the downstroke lasts longer whereas the upstroke is shorter than the corresponding contralateral movement (Elsner, unpublished). Regardless of the complexity of contralateral co-ordination illustrated by these examples the transition from one state to the other is always abrupt. The phenomenon of “relative co-ordination” has not been observed in intact animals. It has recently been shown that the maintenance of this multistable, but otherwise “absolute” co-ordination depends on peripheral interactions with the central nervous system (see the following section). 2.3
CENTRAL
vs.
P E R I P H E R A L CONTROL OF S O U N D PRODUCTION
During the last two decades strong evidence has been accumulated for the theory that orthopteran stridulatio- is programmed predominantly by central nervous networks. External and internal stimuli were thought to have little influence upon the centrally generated patterns. For example, sensory feedback was regarded as dispensible, at least as far as the temporal patterning was concerned (review: Elsner and Huber, 1973). This concept generally holds true if one looks at proprioceptive information on a short-term basis, i.e. considering the phasic nature of sensory feedback. More recently, however, long-term changes in the central motor patterns following peripheral operations have been demonstrated (Lindberg and Elsner, 1977). External stimuli may also have a greater importance for the patterning than was previously thought: it has been shown in bush crickets that the song of one species may have strong modulatory effects upon the song pattern of another. muscle at the 1st and 2nd cycle of activation (1, 2) of each “chirp” 4 different modes of contralateral coordination exist. Potentials of the posterior unit of the left subalar muscle (which are monitored with low amplitude due to the particular recording site, see Fig. 4) are marked by dots. (a) from Kutsch (1969); (b) from Elsner (1968).
2 52
NORBERT ELSNER AND ANDRW V POPOV
The present picture is rathei incomplete; therefore, the following section is more an enumeration of single experimental effects than a comprehensive theory of central and peripheral interactions underlying stridulation. However, the data which are available at present, show that the concept of centrally programmed behaviour has now to be considered anew. 2.3.1 Modifications of stridulatory patterns by external stimuli Orthopteran song patterns are often generally described as “highly stereotyped”, thus overlooking the fact that this holds true only for certain parameters. In crickets, for example, the pulse repetition rate is very stable and independent of most external conditions, while other parameters such as the chirp rate are highly variable, reflecting the state of the external environment. and internal motivation (Popov, 1971, 1972; Popov et al., 1974a). For example, single males of Acheta domesticus stridulate at a rate of 1.5 chirpds which is abruptly doubled to 3 chirps/s as soon as a conspecific female approaches. The pulse repetition rate, however, remains stable (Popov et al., 1974b). Weber (1974) has shown that a male Gryllus campestris making mechanical contact (via the antennae) with a conspecific female produces a highly stereotyped courtship song characterized by an extremely low variation of the “tick”-intervals. As soon as the contact is interrupted these intervals become more irregular, being generally increased until, finally, the song is interrupted. External stimuli do not only influence the excitatory state of the singer in a general way, but may affect even the timing of individual chirps. It has long been known that in all orthopteran families, singing males of those species producing short chirps may alternate, or occasionally synchronize, their chirps in a regular manner if they are within hearing range (review: Dumortier, 1963~).This alternation can best be explained by a mutual inhibition: acoustic signals such as the song of another male (but artificial sound stimuli as well) can have an inhibitory effect which results in a resetting of the chirp rhythm (Jones, 1966a,b, 1974; see also Heiligenberg, 1966, 1969). It has been demonstrated in the field cricket Gryllus campestris, that not only acoustic, but also vibratory and optical stimuli cause this inhibitory effect (Dambach, pers. comm.). With regard to interspecific acoustic interactions Broughton (1965) found that the bush-cricket (tettigonioid) Platycleis intermedia Serville (“P.sabulosa” sensu Broughton) modifies its song when the closely related species P. afinis Fieber chirps nearby. More recently, this interaction, and others, have been studied in more detail, both in the laboratory (McHugh; Samways, 1975, 1976a,b; Samways and Broughton, 1976) and in the field (McHugh, 1972; Samways, 1977). In simple terms, the long chirp of P. afinis either causes P.
A
tt tt tt tt tt tt tt tt t t t t t t t t t t t t t t t t t t t t t t t t
w
tt tt
tf tt t t tt tt tt tt
tt
tt
tt
tt tt tt t t tt t t
Fig. 8. Level recordings of the influence of single Plafycleis afinis “chirps” upon the regular disyllabic emissions of P . intermedia. (a) P. intermediu inhibited for the whole “chirp” of P. &is. but recommences afterwards with increasing disyllabic rate. (b) P. intermedia duets with P. ufinfs, but produces only disyllabic “chirps”. Immediately P. aJinis stops, it sings at an abnormally high but decreasing rate. The normal disyllabic “chirp” rhythm breaks down temporarily (arrow head). (c) There is a trough in P. intermedia’s rate of singing late in the post-interaction after a duet, as manifested by a monosyllabic “chirp” (arrow head). The long horizontal arrows in all three traces represent the extent of the P. afinis emission. Individual syllables of P. intermedia are marked by the small vertical arrows. Courtesy of M. J. Samways.
2 54
NORBERT ELSNER AND ANDREJ V POPOV
intermedia temporarily to cease producing its regular series of short chirps (inhibitory interaction, Fig. 8a) or allows it to duet, but with a characteristic syllable patterning (duet interaction, Fig. 8b,c). Although there are gradations between the two categories, they remain distinct. In some song sequences there is a highly significant alternation of inhibition and duets. After inhibition, P. intermedia recommences singing at a slow and increasing disyllabic chirp rate (Fig. 8a), which, after ca. 6 s can lead to a peak rate of singing (i.e. syllables produced per unit time, irrespective of chirp length), sometimes manifested by a tri- or other polysyllabic chirps. After ca. 9 s P. aflnis chirps again but this time P. intermedia duets with it. At the start of the duet, P. intermedia is partially inhibited, but breaks through to produce only disyllabic chirps at a slow but increasing and erratic rate (Fig. 8b). Immediately the macrosyllables of P. aflnis end, P. intermedia sings at an abnormally high rate with a temporary, but complete breakdown of its normal disyllabic rhythm, with the production of polysyllabic chirps. The rate of singing-then decreases as the normal disyllabic rhythm is resumed. About 5 s after the end of the P. afinis emissions there is sometimes a trough in the P . intermedia rate of singing, occasionally marked by a monosyllabic chirp (Fig. 8c). After ca. 9 s P. ajJinis sings again, resulting in the inhibition of P. intermedia. Thus, the normal fairly stable rate of singing of P. intermedia is thrown into a distinct oscillation by the P. aflnis song.
2.3.2 Proprioceptive control of sound production a The role of acute peripheral feedback Possibilities for phasic sensory interactions are undoubtedly present since numerous sense organs are found in the thoracic region which are able to monitor precisely the stridulatory pattern acoustically or mechanically. These might, therefore, be parts of the feedback loops controlling song performance. First one has to consider acoustic feedback, since acoustic information is not only signalled to “higher” integrative areas of the brain (Adam, 1969; Rheinlaender, 1975) but also directly to the thoracic motor system (Rehbein, 1975, 1976). However, the CNS does not seem to depend on song information to generate the specific motor output. The micropterous cricket individuals which occur occasionally in cultures show stridulatory movements, although they cannot produce any sound because the rudimentary elytra are not able to touch each other. The same motor output pattern underlying stridulation in normal animals is recorded from the mesothoracic muscles of these individuals (Kutsch and Huber, 1970). As expected, cutting the tympana1 nerves (experiments were performed both in crickets and in grasshoppers) has no effect on stridulation (Huber, 1963).
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
255
As far as mechanoreceptive control of stridulation is concerned only a few investigations, on distinct receptors, have been carried out. The most thoroughly analysed example is the wing stretch receptor of caeliferan (Pabst, 1965; Gettrup, 1963) and ensiferan insects (Moss, 1967). In crickets, this sense organ is phasically activated during stridulation. Moss (197 1) working with freely moving animals demonstrated a phasic discharge of the mesothoracic receptor at a rate of 1-3 spikes per pulse. The receptor fires shortly before the activation of the subalar muscle, i.e. at the end of the closing movement (Fig. 9b,c). In the locust Schistocerca gregaria, monosynaptic connections between the stretch receptor and thoracic motoneurons have been found (Burrows, 1974): these might also exist in crickets. If they do, a feedback control mechanism, acting along the shortest route could easily be established. However, just as in the case of acoustic feedback, the phasic monitoring of the wing movements is apparently not essential for the generation of song patterns: section of the nerves which carry this information (and also all other sensory inputs from the wing area) does not cause any change in the song rhythm (Fig. 9a). The minor importance of acute proprioceptive feedback seems to be a general principle, although most of the sense organs in question have still not been investigated in detail. Both in grasshoppers and in crickets numerous experiments have been carried out on gross peripheral areas (Loher and Huber, 1966; Elsner and Huber, 1969; Elsner, 1967, 1973; Kutsch and Huber, 1970). For example, the stridulatory elytra or hindlegs were fixed in various positions, or loaded, or amputated, while in other cases certain muscles were cut at their skeletal insertion. These animals, although heavily restrained by these operations, continued to “stridulate” at the motoneuronal level. Even after complete removal of the wings or the hindlegs, the muscles (located in the body stem) still received the specifically patterned excitation which remained basically unchanged in the short term. This applies also to the co-ordination of the two hindlegs, which in most grasshoppers move with a considerable phase lag. In the grasshopper Gomphocerippus rufus, the hindlegs were coupled mechanically, but nevertheless contralateral homologous units remained activated in their out-of-phase-manner. Consequently, the stridulatory movements were suppressed (Elsner, 1967; Huber, unpublished). Phasic sensory feedback, which obviously is not obligatory for the generation of the basic song rhythms, seems to play a more important role in the fine control of abduction and adduction of the stridulatory organs. For example, when the hindleg of a grasshopper was loaded so that it pressed against the elytron and had to overcome a greater frictional resistance, the activity of certain abductor muscles was increased. That of the major elevator muscles remained unchanged (Elsner and Huber, 1969). If one summarizes all these observations, one concludes that the basic motor
256
NORBERT ELSNER AND ANDREJ V POPOV
n n
200
900 600
100
300
(a)
0
0
n n
200
900I 600
100
300 0
0 0 80
240
400
AP(mr)
Ap
27 31 35 39 43 47 (nu)
250 ms Fig. 9. The role of the wing stretch receptor during cricket stridulation (Gryllus campeslris). (a) Histograms of chirp (left) and pulse intervals (right) before (top) and after (bottom) destroying all stretch receptors. Insert: activity of the tergocoxal muscle 90 (wing closer) during calling song. (b, c) Simultaneous recordings of the stretch receptor (SR, marked by arrows in b) and muscle activity (90, wing closer; 99, wing opener) during calling (b) and courtship song in a freely moving animal. From Moss (1971).
patterns underlying orthopteran stridulation are under central nervous control. Phasic sensory feedback appears to be important for some of the details only. However, the feedback also seems to guarantee a certain level of excitation which is needed for the activation of certain motor units (normally
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
257
characterized by a rather high threshold). If the femur is amputated, such motor units are less likely to be activated. In these animals, the central motor output-which remains unchanged in its temporal patterning-is similar to that recorded in normal grasshoppers when they produce low sound intensities. These findings are mainly based on studies of the field cricket Grylfus campestris and the acridid grasshopper Gomphocerippus rufus. More important interactions between peripheral and central mechanisms have been observed during sound production in the grasshoppers Chorthippus mollis, Chorthippus biguttulus, and Omocestus viridulus. As mentioned earlier, these three species produce different sound patterns with each of their hindlegs (see p. 239). Intact hindlegs change their role from time-to-time, but never produce the same pattern simultaneously. This change depends on peripheral inputs from both sides: as soon as one of the two legs has been amputated distally (i.e. at the coxo-trochanteral joint), the remaining one then produces only one of the two patterns (always the so-called pattern I, see Figs 3,36 and p. 239) while the motor output underlying the other pattern occurs only on the operated side (Elsner, 1974a). If both hindlegs are cut, the motor patterns change from side to side, as in normal animals. During the ensuing days serious long-term effects occur which will be reported below. Surprisingly, the unilateral or bilateral loss of proprioceptive sense organs located near the pleurocoxal and the coxo-trochanteral joint (which are directly stimulated during stridulation) does not seem to be responsible for these phenomena. In the grasshopper Chorthippus biguttulus, sensory structures at the tip of the femora play the most important role. When the peripheral nerve (no. 5 ) is cut in that region, either unilaterally or bilaterally, the resultant effects are as described above (Lindberg, 1974). So far, it is not known precisely which sense organs of the distal part of the femur are involved (for example: the chordotonal organ described by Usherwood et al., 1968 and Burns, 1974 or the stretch receptors studied by Coillot and Boistel, 1968, 1969 and Coillot, 1974, 1975). Recordings have not yet been performed in stridulating animals, and, consequently, no conclusions can be drawn about the patterned form of this peripheral input. b Peripheral long-term eflects upon motor patterns Apart from the effects immediately following peripheral ablation experiments in the grasshoppers Omocestus viridulus, Chorthippus mollis and Chorthippus biguttulus severe long-term alterations occur after loss of both hindlegs. For example, in Ch. mollis the simultaneous performance of different stridulatory patterns by the left and the right hindleg depends on long-term interactions of peripheral mechanisms with the central nervous system. As mentioned above, after the loss of both hindlegs the two motor patterns are produced-Enchanged for a while as in normal animals. However, during the ensuing days pattern I
258
NORBERT ELSNER AND ANDREJ V POPOV
(characterized by strong depressor activity at the beginning of each chirp, see Fig. 36) becomes increasingly dominant on both sides, whereas pattern I1 (which lacks the initial depressor activity) begins to disappear. At first this phenomenon is observed during the last part of the sequences only, a few days later also in the middle and, finally, in the initial parts. After 4-7 days (a considerable inter-individual variability exists), motor pattern I is always simultaneously present on both sides, while pattern I1 disappears completely (Elsner, unpublished). In Chorthippus biguttulus the “chirp”-separation of the sequences cannot be maintained for more than a few days without participation of peripheral mechanisms. If both hindlegs are amputated, or if sensory structures in the region of the tip of the femora are destroyed, the “chirps” are increasingly prolonged; i.e. the pauses (pattern I) and the extended upstrokes (pattern 11) normally separating the chirps after every third cycle of elevator and depressor activity become less frequently inserted (Lindberg and Elsner, 1977). Figure 10 illustrates the time course of this effect for one particular animal (again, as in Ch. mollis, the inter-individual variation is rather high). In this case, on the second day after cutting both hind-femora the metor sequences were subdivided into “chirps” each containing 4-5 cycles of elevator and depressor activity (instead of 3 as in intact animals and shortly after operation). On the third and fourth days the number increased to 6-10 cyclesPchirp” and on the fifth to eighth days 10-20 cycles/”chirp” and more were observed. Finally, after 10 days the subdivision of song sequences into “chirps” completely disappeared. Furthermore, in all three species, the co-ordination of the motor output to both sides is seriously affected by such bilateral ablation experiments. As mentioned earlier, contralaterally homologous muscles are normally activated out of phase (see p. 249). After the loss of both hindlegs, the exact phase shift can only be maintained for not more than 1-2 days. Thereafter, irregularities are observed, generally with a tendency towards synchrony. In Omocestus viridulus in particular, contralaterally homologous units are mainly recruited synchronously after 5-7 days (Elsner, unpublished). Long-term effects following deafferentation have been observed also in crickets (Acheta domesticus, Melanogryllus desertus: Popov, 1911). On the surfaces of the elytra 4 hair plates are present which are stimulated during stridulation. Elimination has the following effects: after operation the crickets sing for several days keeping the wings in normal position. Thereafter, the Fig. 10. Long-term change of the motor pattern underlying sound production in the grasshopper Chorthippus biguttului after amputation of both hindlegs (i.e. autotomy at the coxo-trochanteral joint). The recordings are from the left (upper traces) and the right (lower traces) pleurocoxal muscle (125, depressor). The pauses subdividing the sequences into “chirps” are marked by arrows. From Elsner (unpublished).
69 I
NOllV31NflL/UL/U03 3 l l S f l O 3 V 4 0 h9010H130~Il3N
2 60
NORBERT ELSNER AND ANDREJ V. POPOV
amplitude of pulses decreases and becomes less stable, because the adduction of the elytra is less stabilized. The form of the pulses varies because the startposition of closing is now more uncertain. Hence the pulse rhythm becomes less stable also. Considering these surprising findings in the light of the discussion on central vs. peripheral control one can still consider stridulation to be centrally programmed, but only in the short-term: acute sensory information is not necessary for the generation of song patterns (except for guaranteeing the normal bistability of the patterns in grasshoppers). However, the central nervous “stabilization” of the motor patterns underlying stridulation has only reached an intermediate state, as far as the independence from peripheral mechanisms is concerned. The exact contralateral co-ordination and sometimes even specific elements of the pattern (for example: “chirps”separation in Ch. biguttulus) cannot remain established over a long period without peripheral participation. At present, the long-term effects cannot be explained in a satisfying way. The time course suggests that degeneration of sensory fibres may occur which possibly have trophic influence upon the central nervous networks generating the specific stridulatory patterns. 2.4
THE CENTRAL NERVOUS ORGANIZATION OF SOUND PRODUCTION
2.4.1
Intersegmental networks
Under normal conditions many parts of the central nervous system are involved in the control of stridulation. This is particularly striking on the afferent side: the different songs can be released by external and internal stimuli detected by sense organs which are connected to different, widely separated ganglia. For example, in crickets, the presence of a spermatophore, signalled by proprioceptors in the region of the copulatory apparatus to the last abdominal ganglion, serves as a stimulus for the production of the calling and courtship songs. Rivalry song (performed by many crickets and grasshoppers) may be released by the song of a conspecific male received by prothoracic (in crickets) or abdominal (in grasshoppers) tympana1 organs. In crickets, strong mechanical stimulation of the antennae, transmitted to the protocerebrum, acts as a further releaser of rivalry song and associated behaviour. However, these afferent inputs are not obligatory: they may replace each other, and indeed, if the males have been isolated for some time, songs may even be produced in uucuo (review: Huber, 1963). On the efferent side, the question whether the pattern generator can be localized in a distinct area of the CNS has been discussed extensively. Evidently, those ganglia which contain the motoneurons driving the stri-
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
261
dulatory muscles are the candidates most suspected of being such “centres”. These are the meso- and the prothoracic ganglion in crickets and tettigonioids and the metathoracic ganglion in acridids. Are they able to generate the song rhythms autonomously or do they require some general or even patterned excitation from other ganglia? In those species where stridulation is accompanied by movements of other parts of the body (see p. 239), the presence of intersegmental “song” fibres is clearly evident. For example, in the grasshopper Gomphocerippus rufus movements of the head, the palps and the antennae are performed synchronously with stridulatory movements of the hindlegs. Muscle recordings simultaneously carried out in the head, the pro-, meso-, and metathoracic region demonstrate an exact co-ordination of this multisegmentalmotor output. Each chirp produced by the hindlegs is accompanied by a sideways movement of the head (Fig. 11). Phasic motor activity of the head muscles is also recorded during those parts of the courtship where stridulation, but no visible head movements are performed. Similarly, activity of mesothoracic flight muscles-also well-co-ordinated with the sound pattern-is observed during stridulation (Elmer, 1968). This exact plurisegmental co-ordination of the motor output is probably controlled by descending interneurons running from the brain to the metathoracic ganglion. Stridulation stops immediately after both connectives are cut anterior to the metathoracic ganglion, but the movements of the head and its appendages continue normally. The time required for those parts of the courtship consisting of stridulation which is not accompanied by head movements is taken into account (Loher and Huber, 1966). Another experiment shows that the thoracic song system is not only switched on and off, but can be influenced by commands from the cephalic region. If the head is fixed to the prothorax, the repetition rate of the bursts recorded from the head muscles (each of which normally elicits a sideways movement) and the chirp rate is lowered. Consequently, the motor co-ordination of the head and the thoracic region remains unchanged (Elmer and Huber, J969). These experiments and the analysis of the co-ordination of head and hindleg movements at the chirp level suggest the hypothesis that the cephalic part of the CNS not only determines the general onset of stridulation but also the timing of single chirps. Only the very fine pattern, i.e. the intra-chirp “syllable” rhythm, is thought to be generated by a metathoracic network. We cannot yet answer the question whether the metathoracic ganglion of other grasshopper species which do not accompany stridulation with head movements also depend upon such distinct chirp-commands (Elmer, 1968, 1973). In crickets, the thoracic system is able to generate the pulse and the chirp pattern autonomously. If the connectives posterior to the mesothoracic ganglion are cut, all three songs reappear after some days (Huber, 1960).
262
NORBERT ELSNER AND ANDREJ V POPOV
125
I I, 0,5s
0.2s 99
(C) Ton
0,ls Fig. 11. Intersegmental co-ordination of the motor pattern underlying courtship behaviour in the grasshopper Gomphocerippus rufus. (a, b) Co-ordination of motor commands from the subesophageal and the metathoracic ganglion controlling head movements (49s, f) and hindleg stridulation (125, 129) during the 1st (a) and the 3rd (b) subsequence of courtship. (c) Coordination of motor commands from the mesothoracic ganglion controlling the position of the elytra (99) and sound production (Ton) of the hindleg. 49s, f, slow and fast units of the prothoracic dorso-longitudinal muscle (moving the head); 125, metathoracic pleurocoxal muscle (depressor of the hindleg); 129, metathoracic subalar muscle (elevator of the hindleg); 99, mesothoracic subalar muscle (depressor of the fore-wing); OSG, USG, I, 11, 111, supra-, subesophageal, pro-, meso-, metathoracic ganglion. Further explanation (concerning the black connectives between I1 and 111) is given in the text. From Ekner and Huber (1973).
Similarly, with the cephalic ganglia isolated by cutting the neck connectives, “spontaneous” calling song as well as transitions to both courtship and rivalry song occur after a silent period of 1-2 weeks (Kutsch and Otto, 1972). Longer sequences of calling song (as well as courtship song) have been elicited by electrical stimulation of the cut neck connectives (Otto, 1967). Kutsch and Otto (1972) conclude from these experiments that “the neural basis responsible for patterning the specific song is restricted to the pro- and mesothoracic ganglia alone”. However, it must be stated that these ganglia depend on at least some excitation from the brain or the abdominal region:
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
2 63
while spontaneous sound production occurred after cutting the connections either to the abdominal or the cephalic part of the CNS, the animals were never seen stridulating-even if stimulated electrically (Otto, 1 9 6 7 t a f t e r the connectives to both regions were severed (Kutsch and Otto, 1972). Although the thoracic ganglia are clearly able to generate the motor patterns underlying sound production in the absence of all neural input from the head ganglia, one has nevertheless to consider what role the brain plays in the stridulatory control system under natural conditions where it acts as an important integrating centre for external stimuli. Using focal electrical stimulation Huber (1960) demonstrated that the central body, the mushroom bodies and their connections to the antenna1 and visual system are involved in song control (review: Huber, 1963). Otto (1971) considerably improved this technique and performed electrical brain stimulation in free-moving animals. The electrodes (extremely flexible steel wires of 2 0 p diameter) could be left chronically implanted for up to 4 weeks. Threshold and latency of the response elicited by electrical stimulation did not vary significantly during this time. The areas of the protocerebrum mentioned above were again found to be important for the release of stridulation. But there were clearly no distinct “centres” for each song type. Only in two cases (from a total number of 41) could one song pattern only (calling song) be elicited. In all other experiments either both calling and courtship song (21 x) or both calling and rivalry song (16x) were released; in two cases all three songs were produced (Fig 12a). The elicitation of courtship and rivalry songs required greater stimulation or a higher stimulus repetition rate than that of calling song. The activation of stridulation did not depend on the patterns of stimulation. Therefore it is not surprising that the song patterns, especially the pulse rhythm, could not be altered. Within certain limits, however, it was possible to speed up the chirp rate by increasing the frequency or the current strength of the stimuli, an effect also observed after repeated inhibitory stimulus bursts. The observation that 90 per cent of all loci stimulated gave two song types (and not one or all three), suggests that the brain interacts (with the otherwise autonomous thoracic song system) via two channels: one command system triggering the generation of calling and courtship song and another system releasing calling and rivalry song (according to the strength of activation of each). Bentley (1977) has located the axon of a command interneuron in the neck connectives of crickets. If stimulated at an appropriate frequency it caused the thoracic network to generate a perfectly normal calling song. There is some evidence that a single calling song command fibre is involved: ( i ) effective filaments are located in the same region bilaterally in the same animal, from animal to animal, and from genus to genus (Teleogryllus and Gryllus), (ii) the smallest filaments appear to contain a single axon, (iii) the filaments have a
NORBERT ELSNER AND ANDREJ V. POPOV
264
Brain Stimulation
Calling Song
\
Courtship Song
Gryllus campestris
\ SCD
OC
Vent ra I Cord Recordi ng co
Cb)
Calling Song
Courtship Song
co
0.2s
Fig. 12. Brain stimulation and ventral cord recordings in the cricket Gtyllus carnpestris. (a) Sagital section of the brain illustrating the stimulation sites. The following songs were elicited (oscillograms on the left): calling and courtship song (+); calling and rivalry song (a);calling song only (0); calling, courtship and rivalry song (*); song inhibition (9.P, pons; CB, central body; MB, mushroom body; DC, TC, deuto- and trotocerebrum; SCo, connective to the subesophageal ganglion. (b) Activity of descending interneurons recorded from the neck-connectives (Co) which had been cut (unilaterally) posterior to the recording site. Calling and courtship song was evoked by brain stimulation. 90, 99, activity of the posterior tergocoxal (90, wing closer) and the subalar muscle (99, wing opener). (a) from Otto 1971 (modified); (b) courtesy of D. Otto.
very sharp threshold for effective stimulation, (iu)similar sized bundles (but so far not the effective one) can be shown, by all-or-none spike failure, to contain a single driven unit. The results of previous ablation and stimulation have indicated that speciJcally patterned activity in descending interneurons need not be necessary. Consequently, recent observations made by Otto (in prep.) are surprising. He has recorded from the neck-connectives of Gryllus campestris
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
265
during the calling song and during the transition from calling to courtship song when these were elicited by electrical stimulation or lesions of the brain. The connective was cut between the recording electrode and the thoracic ganglion to ensure that descending activity only was recorded. The contralateral connective was left intact. Otto observed several descending interneurons whose activity was correlated with the song patterns. For example, one fibre appeared to encode the calling song chirps and the respiratory rhythm. This neuron fired tonically during both the chirp (Fig. 12b, upper recording) and the breathing intervals, but remained silent during the chirps and the respiratory movements. Kutsch (1969) found that the cricket breathes mainly in phase with the chirps. The present results show that both of these rhythms are already coupled in the descending pathways from the head ganglia since they are carried by the same neuron. This unit might, therefore, belong to a system controlling the chirp and respiratory rhythms at an intersegmental level. Also, it is remarkable that Bentley (1969b) found a unit with a similar activity pattern with respect to the calling song in the neuropile of the mesothoracic ganglion (see below). Another intersegmental descending neuron observed by Otto was found to be phasically active before each “tick”-sound which characterizes courtship song (see p. 237). Obviously, this neuron plays an important part in the control of the rhythm of this type of song (Fig. 12b, lower oscillogram). It is still not clear how the co-ordination between the rhythmical activity of these descending interneurons and the particular behavioural rhythms are brought about. One cannot exclude the possibility that a second generator for these rhythms is actually located in the brain. However, as soon as the contralateral connective was also cut, the activity of the neuron encoding the respiratory rhythm and the calling song chirp rhythm, changed immediately. It now fired tonicdly even during the breathing movements themselves, i.e. the descending rhythms seem to depend in part on ascending commands originating from the thoracic (and abdominal with respect to respiration?) ganglia. 2.4.2
Intrasegmental networks
Considering the enormous progress that has been made in unravelling the wiring of central nervous networks generating behaviour in invertebrates (for reviews see: Fentress, 1976) intracellular investigations of the orthopteran song generator have proceeded extremely slowly. Studies performed by Bentley (1969a,b), nearly a decade ago, are still the only direct approach made to the problem. He used the brain stimulation technique developed by Huber (1952, 1955, 1960) and evoked stridulatory patterns in Gryllus campestris by small heat lesions made in the dorsal areas of the mushroom body. Although the
266
NORBERT ELSNER AND ANDREJ V POPOV
animal was pinned down on its back, and was unable to move its wings, the CNS produced the specific motor patterns for calling song continuously over ten hours. During this performance, intracellular recordings were made from various processes of mesothoracic motoneurons driving the second basalar (wing opener) and one of its antagonists, the anterior tergocoxal muscle (wing closer), as well as from several interneurons. In both sets of motoneurons, activity was always restricted to the chirps; during the intervals neither postsynaptic potentials nor any slow depolarizing or hyperpolarizing waves were observed. The onset of chirps was sharply marked by a rapid depolarization in the opener neurons followed by the first spike in about 3 ms (Fig. 13a,b). Closer motoneuron activity began with a hyperpolarization occurring 3-7 ms after the first opener spike. Thereafter the closers depolarized repetitively according to the number of pulses per chirp (in this preparation usually 4-5) being firmly latency-locked (about 16 ms) to each preceding depolarization (spike) of the openers. Each closer spike was preceded by IPSPs which correspond 1 :1 to the opener activity. Remarkably, this inhibition almost perfectly coincided with hyperpolarizations observed in the openers after each spike, except for the first in each chirp (Fig. 13c). Bentley (1969b) has drawn a model based on these recordings, which explains the generation of the pulse pattern: the opener neurons receive input from a chirp-timing, slow oscillator which is physiologically separate from the pulse-generating mechanism. As soon as a certain excitatory level is reached, they begin firing rhythmically. Positive couplings between them might be one of the factors important for the production of the bursts. The openers drive interneurons which inhibit the closers and the openers themselves. In this way (i) the closers are prevented from firing simul'taneously with their antagonists and (ii) the activity of the openers is terminated (possibly supported also by accumulating refractoriness, see Wilson, 1964). Appropriate units were found which fired in bursts during the period of common inhibition of the two antagonistic sets of motoneurons (Fig. 13d). As soon as the opener activity stops, their inhibition upon the closers is terminated and is probably followed by sufficient rebound to excite these neurons without the need for further external excitation (which, of course, cannot be excluded). In this way an alternating latency-locked activation of the two sets might be established for aslong as the openers are excited by the slow oscillations. Theoretically, there is no need for any feedback from the closers to the opener neurons. At present it is impossible to propose a similar model for the chirpgenerating mechanism. However, several units have been found which certainly belong to this network. One interneuron undergoes a slow hyperpolarization during the chirp intervals being depolarized and, thus firing bursts during the chirps. Remarkably, it does not fire in phase with the opener or closer motoneurons (Fig. 130. Another unit shows a complimentary activity, being
267
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION I
lOOrns
Fig. 13. Intracellular activity in mesothoracic motoneurons (a-f, lower traces) during generation of motor patterns (elicited by brain lesion) underlying calling song of Gryllus cnmpestris. Extracellular recordings from the 1st and 2nd units of the 2nd basalar nerve (initially upward spikes, shown separately in a) and the 1st tergocoxal nerve (initially downward spikes, shown separately in b) are displayed on the upper traces. The lower traces show: (a, b) Intracellular recording from the 1st motoneuron of the 2nd basalar muscle (wing opener); IPSPs indicated by arrows. (c) Intracellular recording from a (presumed) closer motoneuron; arrows indicate the onset of IPSPs. (d) Intracellular activity of a (presumed) interneuron firing during the period (bar) between the onset of the opener motoneuron burst and the onset of the closer motoneuron burst during a chirp. (e) Intracellular recording from a unit which fires in bursts during the interchirp interval and is probably inhibited during the chirp. (f) Intracellular recording from a unit which fires in a burst during the chirp but not in phase with the pulses and is probably inhibited during the interchirp interval. From Bentley (1969b).
inhibited during the chirps and firing during the intervals. As mentioned earlier, it corresponds precisely to the activity of an intersegmental neuron found by Otto (in prep.) in the neck connectives (Fig. 13e, see also Fig. 12b, upper oscillogram).
NORBERT ELSNER AND ANDREJ V. POPOV
260
3
The neuronal basis of sound reception and sound recognition
3.1
INNATE R E L E A S I N G MECHANISMS
The phonoresponse of female Orthopterans (phonotaxis and agreement song) Commonly several species of insects call in the same biotope and at the same time, so that a female has to extract and recognize the call of a conspecific male from the background of noise produced by other animals. How is this done? What are the features of the signals which the females use for this process of recognition? What are the cue parameters of the song which are selected by the innate releasing mechanisms of the females? How different are these mechanisms in different species having different song structures but living in one biotope? These problems can be investigated, since sexually mature females of Orthopterans show clear behavioural responses to the song emitted by a conspecific male: female tettigonioids and crickets move to the male, being attracted by his calling song. This acoustically dependent movement reaction towards the sound source is called “phonotaxis”. Female gomphocerine grasshoppers produce an “agreement song” when hearing a conspecific male stridulating. This song is considered as the expression of her willingness to mate. Phonotaxis and female stridulation are the behavioural tools which can be used to reveal the nature of the innate releasing mechanisms. The history of the experimental study of phonotactic reactions of insects began with the pioneering experiments of Regen (19 13, 1923) who showed that in the field cricket, Gryllus campestris, the calling song of a male is a single cue which attracts the responsive female. Later, this was confirmed for other Orthopterans and Cicadidae (Baier, 1930; Busnel and Busnel, 1955; Alexander, 1960). In the following, experiments are reported which have been carried out to test the biological efficiency of various physical characteristics of the male calling songs. 3.1.1
3.1.2
The role of amplitude modulation
a Gryllidae Crickets, in particular, are ideal subjects for investigating innate releasing mechanisms by using phonotactic behaviour because their sound signals can easily be synthesized and normal phonotaxis can be observed in the laboratory under experimental conditions. Trilling species: The most primitive cricket song pattern is a continuous trill composed of long series of sound pulses linked together. Walker (1957) showed by behavioural experiments with trilling species of Oecanthus that the most important parameter releasing phonotaxis of conspecific females is the pulse rate. Similar data were recently obtained for trilling mole crickets ’
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
269
(Scapteriscus acletus) in field experiments, which showed that females had a clear preference to sound models with pulse rates corresponding to that of the male calling song (Ulagaraj and Walker, 1975). But this is not always the case. The calling song of Gryllodinus kerkennensis Finot. is also a continuous trill but due to the nearly rectangular form of the pulses and the short intervals between them, it sounds to the human ear, and looks like, a continuous tone. Ymaze experiments with females of this species (Popov and Shuvalov, 1977) showed that the pulse rate has no significance in this case. On the contrary, if the intervals between the pulses are made longer, the pulse rate being the same, females reject this model. A continuous tone of 5 kHz is equally as effective as the calling song, both during separate and simultaneous presentation (Fig. 14). Thus, the main parameter releasing positive phonotaxis in G. kerkennensis females is the continuity of the song (Popov and Shuvalov, 1977). Chirping species: The complexity of the calling song is considerably increased by subdividing it into chirps, i.e. more or less stable groups of pulses separated from each other by intervals longer than those between pulses. In the simplest case each chirp consists of an equal number of pulses and the pulse rate is more or less constant (for example: calling songs of Gryllus campestris and G. bimaculatus). In these two species the female phonotaxis has been extensively studied (Shuvalov and Popov, 1973a,b; 1976; Popov et ul., 1974b, 1975). The following conclusions have been made: (a) The specificity of the female’s response depends upon the conditions of stimulation. If there is no choice (one signal in silence) crude axtificial stimulation (for example a continuous tone of 5 kHz) can induce phonotaxis. The closer the artificial stimulation is to the conspecific calling song, the higher the probability of positive reactions. But if females have a choice between 2 or more sounds, their response becomes much more selective and they respond with a clear preference to sound which is closer to the calling song. They reject sound patterns they responded positively to during separate presentation. (b) The phonotaxis of females is sharply tuned to a pulse rate corresponding to that of the calling song of conspecific males (Fig. 15, upper graph). (c) Chirp duration is not very important for Gryllus bimuculutus females. Stimulation with 4-9 pulses/chirp (chirp duration 130-3 10 ms) were as effective as the calling song; stimulation with 3 pulses/chirp were rejected (Fig. 15, lower graph, black column). The minimal pause between chirps was found to be about 100 ms. If this “window” between the chirps was smaller the signal was rejected. In Gryllus campestris, on the contrary, chirp duration is the second most important parameter of the calling song. Females showed clear preference to 4-pulse chirps. 3- and 5-pulse chirps were rejected in most cases (Fig. 15, lower graph, striated columns). (d) Chirp rates in the biological range (1.6-3.3 Hz), intensity (which reflects the distance to a singing male), pulse form and duration, are not critical for the
270
NORBERT ELSNER AND ANDREJ V POPOV
process of recognition of the song, but influence greatly the motivational level of a female. This can be appreciated indirectly by the speed of her movement. The louder the sound and the higher the chirp rate, the greater the speed (i.e. the shorter the time of reaction). Dathe (1974) showed that the movement of G. bimaculatus females towards the sound source has a constant acceleration.
Fig. 14. Phonotactic response of Gryllodinus kerkennensis females. Top: Oscillogram of the male calling song; bottom: calling song of “Species 1” (not yet described) which is sympatric to G. kerkennensis. The songs are similar in pulse rate but differ in pulse form and interpulse intervals. (a) Probability of response (P) to separate presentation of the conspecific calling song and 5 kHz continuous tone (black columns) and to simultaneous presentation of both signals (white columns). (b) Probability of response to simultaneous presentation of the conspecific calling song and a model imitating the pattern of the song of “Species 1”. From Popov and Shuvalov (1977).
When confronted with a choice between 2 conspecific calling songs, the females always choose the louder song and the one with the higher chirp rate. All these parameters are highly variable in the calling songs of males within one population, as well as in one individual. They reflect the state of motivation of a male and the conditions around him (for example the presence or the calling of other conspecifics); the distance to the female (intensity and spectrum); and the
27 1
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
state of his sound-producing organs (pulse form and duration). The chirp rate is controlled by a variety of different stimuli (Popov, 1971; Popov and Shuvalov, 1974; Heiligenberg, 1966, 1969; Jones and Dambach, 1973). Crvl lus bimaculatus
P
I .o
(a1
0.5
0
h
I
AS ms
36
P
130
170
310 ms
Fig. 15. Phonotactic response of Gryllus bimaculatus and G.campestris females. The latter were from a Ukrainian population allopatric to G. bimaculatus. The calling songs of G. campestris males contained chirps with 4 pulses. (a) Probability of response (P) of G.bimaculatus females to models with pulse periods of 28, 36 and 48 ms presented separately (black columns) and to artificial stimulation with 36 ms pulse period (most similar to the calling song) presented simultaneously with stimulation of 48 ms or 28 ms pulse period (white columns). (b) Probability of response of G.bimaculatus (black columns) and G.campestris (striated columns) females as a function of chirp duration (the stimulations were separately presented). From Popov and Shuvalov (1977).
Walker (1957) studied another chirping species (Oecunthus niueus) and showed that females -respond topulseless chirps. He concluded that the chirp rhythm is of prime importance, the pulse rhythm having only a supplementary effect in this species. However, he presented the signals separately which might
272
NORBERT ELSNER AND ANDREJ V. POPOV
have caused a less specific response of the female (see above). Also, Dathe’s ( 1974) failure to demonstrate that Glyllus bimaculutus females cannot discriminate their conspecific calling song from that of G. campestris may have the same explanation.
Fig. 16. Phonotactic reaction of cricket “Species 2” females. Top: Oscillogram of the male calling song. Left graph: Probability of response (P) to artificial stimulation with pulse periods of 50, 70, 80, 200 and 300 ms (shown on the oscillograms below) presented separately (black columns) and to stimulation with a 80 ms pulse period (most similar to the natural calling song) presented simultaneously with stimulations of 50, 70, 200 or 300 ms (white columns). Right graph; The reaction time to stimulation with pulse periods of 80, 200 and 300ms: the stimulation with the 80 ms pulse period is most effective although all 3 forms of stimulation give 100 per cent of positive reactions when separately presented.
In the evolution of cricket song there is a clear tendency to decrease and stabilize the number of pdses per chirp. In such species as Oecanthuspel+cens chirp duration is highly variable and the number of pulses per chirp is large. In G. bimaculatus the number of pulses per chirp can vary only in the range from
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
273
3 to 5 and in the songs of individual males it is even more stable. In G. campestris (allopatric to G . bimaculatus) the number of pulses per chirp is stabilized at 4 and in populations sympatric to G. bimaculatus (for example in Azerbaijan) at 3 (Popov, 1972). In the calling song of Acheta domesticus chirps contain 2-3 pulses, while the calling song of the black turkmenian cricket (sp. nov. “Species 2”) consists of chirps containing only 1 pulse. Thus, the pulse rhythm in this song corresponds to the chirp rhythm of the calling song of another species (Popov and Shuvalov, 1977). It is, certainly, of interest to know which parameters of such a song are evaluated by the females. Experiments have shown that pulse period (=chirp period of other species) and pulse duration are most important (Fig. 16). Again, in this cricket, a specific reaction was obtained only during simultaneous presentations of signals. The next step in increasing the complexity of the calling songs of crickets was the appearance of amplitude-modulation of pulses within the chirps and/or rate-modulation of pulses. Melanogryllus desertus uses both parameters, increasing gradually the amplitude of the pulses and the pulse period within the chirps (Popov, 1972; Popov et al., 1974a; see Fig. 2). Y-maze experiments with females of this species showed that the rate of increase of pulse amplitude within the chirps and the mean pulse rate determines the accuracy of the phonotactic behaviour (Fig. 17). The tuning of phonotaxis to these parameters is nearly the same for separate or simultaneous presentation of the signals. The modulations of pulse rate and duration within chirps are not important. The mean values of these parameters are analysed by the females, since corresponding artificial stimuli were as effective as the natural calling song (Popov and Shuvalov, 1977). Chirp duration is not a critical parameter (Fig. 17, lower graph). This is in good agreement with the high natural variability of this parameter. In Scapsipedus marginatus studied by Zaretsky (1972) the chirps of the calling song are characterized by a definite modulation of pulse rate, which appeared to be of prime importance for the phonotaxis of the females. In Teleogryllus commodus, pulse rate is also the most essential parameter for the phonotaxis of females (Hill 1975). b Tettigoniidae In trilling tettigonioids pulse repetition rate is also the most important parameter (Bailey and Robinson, 197 1). Spooner (1964) showed that females of Scudderia texensis (Phaneropterinae) can discriminate between slow-pulsed and fast-pulsed phrases of the male calling songs: They give a phonoresponse (song) to the first and a phonotaxis or orientation to the second. He also studied the phonotaxis of males of different Phaneropterinae using an imitation of the female’s response song. He found that the interval between the calling song of a male and the response of a female is most essential and varies in different species between 140 ms and 1.4 s. Thus, the auditory system of these insects must be capable of analysing such long intervals.
2 74
NORBERT ELSNER AND ANDREJ V POPOV
Melanogryllus desertus
I .o
(a)
0.5
I 9
I I8
I
I (b)
9.5
OL
I .o
(C) 0.5
0
22
I I
Fig. 17. Phonotactic reaction of Melanogryllus desertus females. Top: Oscillogram of the male calling song. (a) Probability of response to separate (black columns) and simultaneous (white columns) presentations of stirnulation with different increases in amplitude within thk chirps (the 18 dB model is most similar to the male calling song). (b) Probability of response to separate and simultaneous presentation of stimulation with different pulse rate. The stimulation with a period of 42 ms corresponds best to the mean value of this ptlrameter within the chirps of the male calling song. (c) Probability of response to separate preientation of stimuli with different chirp duration. From Popov and Shuvalov (1977).
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
275
c Acrididae The song of agreement emitted by receptive gomphocerine females is an unambiguous yes-or-no decision, since the females answer very selectively only to the song of conspecific males. Therefore, the female’s innate releasing mechanism (IRM) can easily be examined by recording the phonoresponse elicited by different artificial stimuli. The percentage of answers to different stimuli, given randomly one/min during several hours of experiment, is the index of the efficiency of certain combinations of stimulus parameters. In this way the female’s IRM corresponding to the male calling song has been studied in several species of the genus Chorthippus, most thoroughly in Ch. biguttulus (Helversen, 1972, 1975, 1977). Receptive females of Ch. biguttulus respond to artificial stimuli consisting of rectangularly modulated white noise (3-40 kHz) if the durations of the individual signals and the inLervals between them have certain values; the length of the whole sequence is of minor importance as long as it exceeds 1.2 s. If the duration of the signals is held constant (for example 40ms) but the intervals are varied, a reaction curve sharply tuned to one interval value is obtained (Fig. 40a). If the duration of the noise signal is changed, then the length of the intervals has also to be changed in order to elicit a maximum phonoresponse. When plotted in a signal-to-interval field the most efficient combinations fall on a nearly straight line. If one includes all signal-to-interval combinations eliciting a response of more than 50 per cent with regard to the maximum reaction value, then the female IRM is expressed in the diagram as an elongated eliptic area (Fig. 18). Comparing this result with the natural song pattern of the male one immediately recognizes that the female IRM is well tuned to the “chirp”-tointerval ratio of the male song (see insert in Fig. 18). The combination of these two parameters is most important for recognition, whereas number and duration of the whole sequence appear to be less important. Also, the pattern of “syllables” (“pulses”) of which the chirps are composed (each “pulse” lasting less than 15 ms) has no value for recognition, since the females need unitary signals of at least 40 ms duration. Females tested against artificial stimuli resembling pulsed “chirps”, fail to respond as soon as the interpulse intervals are increased beyond 2 ms; i.e. exceed the normal pulse intervals found in the male calling songs. The fact that the duration of the “chirps” and not of the “syllables” (“pulses”) as one of the two relevant parameters provides a plausible interpretation of the phase shift between the stridulating legs (see insert in Fig. 18, also Fig. 36). This guarantees the camouflage of the short-pulse intervals which might otherwise impede recognition. However, another characteristic feature of the male song is far less understandable when considered in the light of the female IRM. As described previously (see p. 241) one of the hindlegs extends sonorously the last upward movement of each “chirp” and, consequently, the intervals separating the “chirps” are covered (see Fig. 36 for
NORBERT ELSNER AND ANDREJ V. POPOV
276
the stridulatory movement). Certainly, the sound filling the pauses has a rather low intensity. When measured against the beginning of the following “chirp” the attenuation is 8-14 dB which is sufficient to guarantee “chirp” recognition: females tested against effective artificial stimuli upon which continuous noise is superimposed do respond, as long as the signal-to-noise ratio remains at least 8 dB (Helversen, pers. comm.). Nevertheless, the question about what
-
--rrrrr..,,,,,-
* ! o o c
,S-q ,,,\\*a%% 0. ’Ch. brunneus
*
\
w
1
I
Fig. 18. Structure of the innate releasing mechanisms (IRM) of female grasshoppers belonging to three Chorthippus species. Artificial stimuli of rectangularly modulated white noise (insert C) have been used to test the responsiveness of the females. The specific IRMs are shown as shaded or encircles areas in the stimu1us:interval field. They are deduced from individual reaction curves (see Fig. 40) and include all stimulus-to-interval combinations which release at least halfmaximum reaction (agreement song). All reaction curves were recorded at 3 5 O C. At this temperature the female IRM covers the stimulus-to-interval combinations of natural songs produced at all temperatures between 20’ and 40° which is shown for Chorthippus biguttulus songs containing chirps composed of 6 (n) and 8 (A) “syllables” (“pulses”). Insert: Natural calling song of male Ch. biguttulus stridulating with both hmdlegs (A) and after amputation of one hindleg (B). In this individual the “chirps” contain the normal number of 6 “syllables” each, in other cases the number may be 4 or 8. From D. and 0. von Helversen (1977) modified.
biological relevance lies in the simultaneous production of two stridulatory patterns one of which endangers the recognition of the important “chirp”intervals, remains open. The innate releasing mechanism revealed at 35O C covers “chirps” of 40160 ms and interval5 of 8-30 ms.This range is much wider than the variability of the corresponding song parameters at this temperature (“chirps”: 50-80 ms; intervals 8-15 ms). The females consider not only the variability at a given
NEUROETHOLOGY
OF ACOU-STIC COMMUNICATION
277
temperature but all song patterns emitted by a male between 17O and 40° C: the elongated eliptic area covering the IRM (Fig. 18) corresponds best to all “chirp”-to-interval combinations occurring in natural songs at different temperatures. In other words, a female Ch. biguttulus at 35 O C “understands” conspecific males singing at any temperature between 17O and 40° C. In other species, for example Chorthippus longicornis and Ch. montanus the female IRM covers a smaller range, following the temperature dependence of the song. Thus, the females count in their own temperature whenlecognizing a calling male (Helversen, 1977). The female IRMs of Chorthippus brunneus and Ch. mollis-two closely related species to Ch. biguttulus (see Fig. 3 6 t h a v e been investigated in the same way using artificial stimuli of rectangularly modulated white noise. As illustrated in Fig. 18 the IRMs of these species (which are all sympatric) do not overlap. Female Ch. brunneus respond to rather short stimuli (signal: 1030 ms; interval: 2-8 ms), and in this species the length of the whole sequence is of greater importance than in Ch. biguttulus: sequences longer than 400 ms elicit no response. The IRM of Ch. mollis covers a much wider range than that of the two other species: the females respond to signals above 150 ms and intervals above 50ms, the ratio between these two parameters being less important. 3.1.3
The role of thefrequency spectra
Since the work of Pumphrey and Rawdon-Smith (1939) who suggested that the auditory system of insects had no means of frequency discrimination, the recognition of sounds by these animals was thought to be based only on the analysis of the amplitude modulation pattern. Many authors subsequently denied the importance of the frequency spectrum of the calling songs for the phonoresponse (song) or phonotaxis of the females. The fact that the spectra of sympatric species often overlap considerably .and that females are rather insensitive to spectral distortions of the calling songs, strengthened this point of view (review: Dumortier, 1963b,c; Haskell, 1961). On the other hand, experiments with various crickets (Hill, 1974, 1975; Popov et al., 1975) and mole crickets (Ulagaraj and Walker, 1975) have shown that the phonotaxis of females is sharply tuned to the main frequency of the male calling song (first low-frequency peak of its spectrum), which is perceived by a specialized low-frequency channel of the auditory system (Nocke, 1972; Popov et al., 1974b, 1975; Rheinlaender et al., 1976; Hill, 1975). High-frequency sounds outside the range of this channel are not effective even if they have the same amplitude modulation pattern as the natural calling song (Popov et al., 1975). The presence of high-frequency components in addition to low-frequency ones has no influence on the phonotactic reaction.
278
NORBERT ELSNER AND ANDREJ V. POPOV
On the contrary, high-frequency sounds elicit negative phonotaxis, especially when crickets are in flight (Popov et al., 1975; Popov and Shuvalov, 1977). Figure 19 gives an example for Gryllus bimaculatus. Similar data were obtained on Melanogryllus desertus. As soon as the signal is within the range of the low-frequency channel the changes in the spectrum of the calling song are not important for the process of recognition. But the more the spectrum diverges from the characteristic frequency of this channel the less loud the signal appears to the female, hence higher intensities are necessary to evoke a response.
dB
90
Gryllus bimaculatus *
. - 70 80
40. 60: 50
30
Positive Phonotaxis
x
\.\/
\.
( 9 to d)
Negative Phonotaxis ( 9 , d)
*\*
/O
Fig. 19. Threshold-frequency curves of positive and negative phonotaxis of Gryllus bimaculatus adult females. The positive phonotaxis was tested in Y-maze experiments. Artificial sounds which irnitatedjhe amplitude modulation pattern of the calling song (but having different carrier frequency) were presented from one side of the maze. Intensity (dB above 0.00002 N/m2) of each signal was changed in steps of 5 dB up to the threshold. The negative phonotaxis was tested to square pulses of pure tone (duration: 100 ms). The animals were fixed at a holder and placed in a wind-stream. The loud-speaker was located on one side of the animal. Changes in flight activity during sound presentation were recorded. Each point of the curves represents a mean value of thresholds of 3 females.
Hill (1975) showed in Teleogryllus commodus, that frequencies below the dominant frequency of the calling song inhibit the positive phonotaxis of females when presented simultaneously with it (pure continuous tone of 2 kHz was presented together with the 5 kHz calling song). Morris (1975) studied the phonotaxis of females in the tettigonioid Metrioptera sphagnorum, whose calling song contains alternating components of sound (15-20 kHz) and ultra-sound (33 kHz, see p. 235). He demonstrated that the 33 kHz component alone can evoke phonotaxis, though only in 42 per cent of all females. The sonic component alone was ineffective. The author
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
279
suggested two different functions for these two parts of the song, the sonic component being important at close range (courting) and the ultrasonic component being necessary for distant recognition and localization. It has been mentioned earlier that a low-frequency sweep is present in the song spectra of several tettigonioids and acridid grasshoppers which resembles the impulse rate and its modulation (see p. 235, Fig. lc). In some species (for example in the grasshopper Omocestus viridulus and Gomphocerus sibiricus) the range of impulse rate modulation (1 : 15) considerably exceeds the range of amplitude modulation (Elsner, 1974a). It has been suggested, therefore, that this parameter might be used for species-specific recognition. Broughton et al. (1975) using data from Zhantiev and Dubrovin (1968, 1971) discussed the possibility that in tettigonioids the low-frequency component could be perceived by the intermediate organ. In the case of acridids Michelsen (in Elsner, 1974a) has suggested that these sweeps might be signalled by the a, b, c groups of sensory cells of the Muller-organ (see Fig. 2 1). Skovmand and Petersen (pers. comm.) have performed behavioural experiments on females of Omocestus viridulus to test this hypothesis, and have demonstrated that the impulse rate has a meaning to the animal. The impulse rate modulation must be correct compared to the chirp duration at a certain temperature. 3.1.4
Final remarks on the acoustic behaviour of thefemales and their innate releasing mechanisms
The following conclusions about the acoustic behaviour of orthopteran insects are important for our understanding of the structure and function of their auditory systems. ( i ) In experiments with chirping species it was shown that the different parameters of the calling songs are not equal in importance for the phonotactic behaviour of the females. Those that are necessary for the process of recognition (i.e. decision that the signal belongs to the calling song of a conspecific male) are known as “essential recognition parameters”. They are definite, few (1-3), normally most stable and/or “unusual” in comparison to other sounds of a biotope, and are time parameters of the amplitude modulation pattern of the calling song. For example, pulse repetition rate, pulse duration, chirp duration, amplitude or rate modulation of pulses within the chirps can serve as essential parameters of the calling song in different species. Even in cases where the song can potentially be described by many time parameters (for example in Melanogryllus desertus) only a few appear to be essential and this is evidently enough for the recognition tasks the animal meets in a biotope. Other parameters (“motivational parameters”) can influence the
2 a0
NORBERT ELSNER AND ANDREJ V. POPOV
effectiveness or the attractive force of the signal and thus determine the level of response activity of a female. Chirp rate and intensity are the most important among them. It was suggested above that these two groups of parameters are controlled, both on the sound emitting and the sound receiving sides, by different parts of the central nervous system. The essential, stable parameters are mediated by the segmental thoracic centres, the motivational parameters by additional centres in the brain which integrate the different inputs (Popov, 1971). The essential and motivational parameters, working together, determine the specificity and plasticity of the response. It was shown that the specificity of the response, in many cases, depends upon the conditions of stimulation. Without choice, females can respond to very rough models of the calling songs, but in choice situations, they demonstrate clear preference to signals which are closer to the conspecific calling song. This type of preference we call “recognition preference”. It is proposed that the processes of recognition and decision for reaction are working on a probability principle (Popov and Shuvalov, 1977). In every case, females choose the signals which have the higher probability of being the conspecific song, the “ideal” properties of which are in some way imprinted in the central nervous system. When two signals having the same essential parameters are presented simultaneously, females demonstrate a preference of another type (“motivational preference”), based on the value of the motivational parameters. In this case, louder signals and signals with a higher chirp rate become more attractive. There should be different and separate regions in the central nervous system for the control of these 2 types of preference. (ii) Comparative analysis shows that the combinations of the essential parameters of the calling songs are different in different species. Normally, these are the parameters that are most characteristic of the species, both qualitatively and quantitatively. Does it mean that insects of a given species can discriminate only the conspecific signals and measure only restricted numbers of essential parameters, or, if properly trained, can they learn to classify another sound? Closely related species such as Gryllus campestris and G. bimaculatus have different innate releasing mechanisms of phonotaxis, but it is difficult to believe that their auditory systems are very different in their absolute capabilities. One may expect that, potentially, all crickets have similar auditory capabilities, or at least that closely related species do. If this expectation is true, then in the auditory system of insects we have to find specialized mechanisms that can appreciate various time-parameters of sound in addition to the genetically fixed, species-specific “filters” of the essential parameters of the conspecific songs. In any case there should be additional more or less independent mechanisms for precise analysis of intensity, direction and chirp rate of the sounds.
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
281
(iii) Experiments on crickets, tettigonioids and acridid grasshoppers show that only virgin, sexually mature, females show positive phonotaxis or phonoresponse to the calling songs of conspecific males. This responsive state is blocked immediately after copulation, and prior to, during and for a short time after, oviposition (Regen, 1923; Baier, 1930; Renner, 1952; Haskell, 1958, 1960; Loher, 1966). Haskell (1960) suggested that a blood-born chemical factor from the gonads is responsible for the onset of the female responsive state, because castrated female grasshoppers loose this state after 24-48 h. Loher (1966) showed that the hormone of the corpora allata of the brain is engaged in the control of this state. Following allatectomy of mature Gomphocerus rufus females, sexual receptivity disappears within 6 days, and defensive behaviour is shown. Implantation of corpora allata into young, allatectomized females causes maturation and sexual receptivity. On the other hand, the level of corpora allata activity, is regulated by nervous control and by the neurosecretory cells of the pars intercerebralis of the brain. The sudden change from sexual receptivity to defensive behaviour after copulation is due to the presence of a spermatophore which stimulates the receptaculum seminis mechanically. After denervation of the receptaculum seminis, females copulate repeatedly within a short time. All these phenomena must be taken into account in any attempt to study the neuronal mechanisms underlying specific responses to acoustic stimuli. (iv) There must be at least a minimal frequency discrimination in the auditory system of Orthopterans, because positive and negative phonotactic reactions show completely different tuning curves (Fig. 19). The first is tuned to the best frequency of the calling song and the second to the high-frequency (mostly ultrasonic) sounds, air disturbances and vibrations produced by predators.
3.2
SENSORY MECHANISMS OF SOUND RECEPTION
Insects have developed several types of sense organs that may be used for sound reception. Working together, these receptors can cover a surprisingly wide frequency range, fiom very slow air currents or infrasound to far ultrasound (above 100 kHz). Low frequencies, from a few Hz up to 1-2 kHz, are detected by different types of hair sensilla and Johnstone’s organs, while frequencies above 1 kHz are in most cases perceived by specialized tympana1 organs. These latter are especially well developed in those groups which show complex acoustic behaviour such as Orthoptera, Cicadidae or Lepidoptera. Very often, each of the several types of sense organs present in one insect will, on stimulation, initiate different types of behaviour. In crickets, for example,
282
NORBERT ELSNER AND ANDREJ V. POPOV
stimulation of the hair sensilla of the cerci by air puffs, evokes escape reaction, while the tympanal organs serve both for intraspecific acoustic communication and to initiate the evasive behaviour produced in response to the highfrequency sounds emitted by some predators. However, this separation of functions is not absolute. After the removal of the tympanal organs, crickets still alternate or synchronize their singing with artificial sounds (Jones and Dambach, 1973). Haskell (1958) showed that the stimuli received by hair sensillae elicited the same stridulation response movements as they gave to stimuli received via the tympanal organs. Since the structure and functions of these sense organs are very different, we shall consider them separately.
3.2.1
Hair sensilli
The best known examples of hair sensilla used as an auditory organ are the cerci of crickets and cockroaches. The ultrastructure of cricket cercal hairs has been described in detail by Gnatzy and Schmidt (1971a,b, see also for earlier literature). Each sensillum is innervated by 1 bipolar sensory cell the dendrite process of which is attached to the base of the hair by means of the receptor cilium, surrounded by the extracellular space formed by the outer sheath cell (tormogen cell). Thurm (1970, 1974) suggested that the tormogen cells of insect epithelial mechanoreceptors are responsible for the development of a transepithelial potential due to an active pumping of K+ into the extracellular space. Using ion-sensitive electrodes, Kuppers (1974) showed that in the campaniform sensilla of the fly Phormia the concentration of K+ and C1- ions in the receptor space is 115 T 11 mM/1 and 28 T 4.0 md1, respectively, i.e. about 13 and 0.4 times greater than that of the blood, and this can give a transepithelial potential difference of about -65 mV and -25 mV, respectively. Functionally, the receptor potentials are closely linked with this transepithelial potential. A model for the behaviour of the mechanoreceptor epithelium during stimulation is given by Thurm (1974). The length of the cercal hairs varies from ca. 0.1 to 1.6 mm. Each hair has one “preferential plane” of vibration due to the asymmetrical structure of its articulation with the chitinous socket. In cockroaches and crickets, at least, the hair sensillae are arranged in several distinct longitudinal zones. Within one zone all the hairs have the same orientation of the preferred plane of vibration (Nicklaus, 1965; Petrovskaja e? al., 1970; Edwards and Palka, 1974). A deflection of the cercal filiform hair of cockroaches and crickets in one direction in the preferred plane causes depolarization of the receptor cell, while deflection in the other direction hyperpolarizes the cell (Nicklaus, 1965; Knjazev, 1976). In Gryllus bimaculatus the dorsal receptors are depolarized when the hairs are deflected to the lateral side; the ventral receptors, when the
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
283
hairs are deflected to the medial side; and medial and lateral receptors, when the hairs are deflected to the base of the cercus. So, with any stimulus direction there is a specific pattern of activated receptors, and the animal receives precise information about the location of the source (Knjazev, 1976). Pumphrey and Rawdon-Smith (1936) were the first to show that the cercal hairs can be stimulated by sound in the range up to 3-4 kHz. Direct measurements of the oscillations of the filiform hairs of Acheta domesticus have been made by Petrovskaja et al. (1970). It was shown that the hair always vibrates in phase with the sound frequency and behaves as a mechanical vibratory system with one degree of freedom. The longer the hair, the greater its sensitivity at low frequencies and the narrower its working frequency-range. All the hairs are most sensitive at the lowest frequencies studied (20-40 Hz). The frequency range for which these authors observed activity in “single” axons in the cercal nerve, after the elimination of most of the hairs, was much narrower than that of the mechanical response of the hair. Each unit they recorded was narrowly tuned to a definite frequency (1 10, 250, 600, 1000 or 1200 Hz). By selective elimination of hairs of different sizes Petrovskaja (1969) showed by recording from the cercal nerve, that the receptors associated with the longest hairs are most sensitive to the lowest frequencies studied (20-40 Hz); the receptors of the hairs of intermediate size are tuned to 50-100Hz, and the receptors with the shortest hairs are most sensitive to about 1 kHz. Knjazev (1977) and Knjazev and Popov (1977) studied the response of single cercal receptor cells in Gryllus campestris by means of very fine tungsten electrodes introduced through the cuticle close to the cell body. A variety of stimuli such as air puffs, both linear and sinusoidal mechanical deflections and sounds in the range of 25 Hz to 4 kHz were used. They found that even in the first larval stage, the cercal hair receptors function normally as in adults. Figure 20 summarizes the results obtained in adults. The working frequencyrange of each receptor is in good agreement with the mechanical response of the corresponding hair, described by Petrovskaja et al. (1970). The longest hairs (1.5-1.6 mm) respond to frequencies up to 300-400 Hz, the intermediate hairs (0.6mm) up to 600-700Hz and the shortest up to 2 kHz, all the receptors being tuned to the lowest frequencies studied (25-40 Hz) (see Fig. 20a). In each case the threshold frequency curve falls from the highfrequency side towards the low frequencies without an optimum. This suggests that air puffs and very low-frequency particle-movements (infrasound) are more adequate stimuli for these receptors than pressure waves. The longer the hair, the larger is its deflection to the same air stream, and the higher is the sensitivity of the corresponding receptor. Having sensilla with different hair lengths, and hence with different sensitivity of receptors, the system is able to measure the stimulus intensity over a wide range.
2 a4
NORBERT ELSNER AND ANDREJ V. POPOV
/
LB m s.st
40
0.02
0.1
3kHz
1
500 100
300
100 .
1
0
.
3
5
7
d
Fig. 20. The responses of hair sensilla on the cerci of Gryllus bimaculutus. (a) Thresholdfrequency curves of cercal hair sensilla to sound stimulation. Sound frequency is plotted in logarithmic scale. L,, L,, L,, length of the hairs (200, 600, 1500pn, respectively). L,m.s.st., the threshold-frequency curve of the long-hair sensilla response to mechanical sinusoidal stimulation. (b) The response of the medium size hair sensilla (600 pn). Discharge frequency is plotted-as a function of sound intensity at different frequencies of sound stimuli. The inserted oscillograms show the response to 40 Hz pure tone of 360 ms duration (intensity in dB above 0.00002 N/mz). (c) The response of the cercal long-hair sensilla (1000pm) to trapezoidal mechanical stimulation. The discharge frequency is plotted as a function of displacement amplitude at different displacement velocities (V,, V,, V,, V,, 4O/s, 8 O / s , 54O/s, 127O/s). The mean discharge frequency during sequential time intervals (corresponding to certain steps of hair deflection), starting from the beginning of stimulation, were calculated and plotted at each displacement velocity. The inserted oscillograms show the response to stimulation in the excitatory and inhibitory direction of hair deflection in the preferential plane. Displacement velo~ity-54~/s; time calibration, 100 ms. (a, b) from Knjazev and Popov (1977); (c) from Knjazev (1977).
The responses of one individual receptor cell to sound stimulation depends upon the sound frequency (Fig. 20b). All the spikes are linked to the same phase of the sine wave even at low intensities. At high intensities, lowfrequency stimulation produces a burst of 2-3 spikes in that phase. The number of spikes per stimulus changes with intensity in the range 15-25 dB above threshold.
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
285
When a sinusoidal mechanical stimulus was used, the threshold of the longhair receptor was independent of frequency at least up to 300 Hz (Fig. 20a, L, m. s. st.). That means that the tuning of the receptors to low-frequency sound is due to the mechanical properties of the hair itself and its articulation in the socket, but not due to the properties of the receptor cell. The receptor cell responds to a rectangular deflection of the hair with a short initial discharge, adapting quickly. It gives a long-lasting after-discharge on being returned to the resting position, especially after a deflection to large angles. When stimulated by trapezoidal mechanical displacements of the hair, with different slope rise times, the receptor cell “on”-response is a function of the velocity and angle (Fig. 20c). At slow velocities (2-4O/s) the frequency of discharge increases gradually, reaching saturation at deflection amplitude of about 3 4 O . At very high velocities (100-150°/s) the discharge frequency starts at about 500 impulses/s and decreases quickly as the angle is increased. At angles above 3-5O the discharge frequency is independent of velocity, due to developing adaptation. The properties of cercal receptors show that this system is ideal for measuring the direction, intensity and velocity changes of air displacements or air streams produced by the movements of any (especially flying) objects in the vicinity of an insect, or by its own movements. Sound reception in the range above 100 Hz can be regarded as a by-product of this function, having little, if any, biological significance, since it is doubtful that important high-intensity, low-frequency sounds are present in their biotope. 3.2.2
Tympana1organs
Among sound receiving organs of insects, tympanal organs are the most specialized. Their location on the body, structure and function varies from family to family indicating their independent evolution in several groups. Each tympanal organ is characterized by the presence of a thin, specialized membrane (tympanum), associated air cavities, and chordotonal sensilla attached either to the tympanum directly (Acrididae, Cicadidae, Lepidoptera) or to the wall of a small trachea, mechanically coupled to the tympanum (Glyllidae, Tettigonioidea). The anatomy and the ultrastructure of these organs are reviewed in detail elsewhere (Acrididae: Schwabe, 1906; Gray, 1960; Gryllidae: Schwabe, 1906, Friedman, 1972a,b; Michel, 1974; Zhantiev, 1969; Schumacher, 1973a,b, 1975; Cicadidae: Vogel, 1923; Young, 1973; Michel 1975; Lepidoptera: Eggers, 1919; Kennel and Eggers, 1933; Ghiradella, 1971; Reviews: Schwartzkopff, 1964; Autrum, 1963; Michelsen, 1974). In the present review only the details necessary for the understanding of physiological data obtained from Orthopterans will be discussed. The reader is referred to the above articles and reviews for further information.
286
NORBERT ELSNER AND ANDREJ V POPOV
The number of chordotonal sensilla within the tympanal organ varies from 1-2 in Lepidopteru to 1300-1500 in Cicudidue. The fine structure of these
sensilla is surprisingly similar in the tympanal organs of all the families studied, and identical within one tympanal organ. Each sensillum contains one bipolar sensory neuron and several sheath cells which isolate the neighbouring sensilla one from another. The distal process of each receptor cell bears one cilium with a 9 + 0 arrangement of longitudinal microtubules. The cilium projects into a scolopale formed by a specialized scolopale cell and filled with extracellular fluid, and ends in the channel of the scolopale cap which is attached to the hypodermis by attachment cells. The method and position of the attachment of the sensilla to the tympanum or the tracheal wall is different in different tympanal organs. Such functional specialization of tympanal receptors is family-specific. a Acrididue Suga (1960) was the first to succeed in-recording the activity of single tympanal receptors in a locust. He found that the electrical events which appear in a single sensillum during sound stimulation are typical of those of other insect mechanoreceptors. He described the properties of the slow (presumably receptor) potential and spike activity of a single cell. The amplitude of the slow potential follows the envelope of the sound signal and spike activity is not synchronous with the sine waves, in contrast to cercal mechanoreceptors. The receptor output is a function of the logarithm of stimulus intensity in the range 40-45 dB above threshold. No saturation was observed. Romer and Schwartzkopff (1975), on the other hand, recording from the axons of tympanal receptors, found that the single units have dynamic ranges of only ca. 20-25 dB and are limited by saturation. All the units Suga (1960) recorded from had the same characteristic frequency (4-9 kHz) but different sensitivity. Popov (1965, 1968) by selective destruction of separate groups of receptors, Michelsen (1966, 1971) by single unit recording from the cell bodies of tympanal organ receptors and Romer and Schwartzkopff (1975) by recording from single receptor axons in the metathoracic ganglion of a locust, found that each of the 4 groups of tympanal organ receptors (Gray, 1960, see Fig. 21) is specialized and different from the others. Three groups (a, b, c) are most sensitive to low frequencies (3.5-6 kHz in Locustu and 1.5-3.7 kHz in Schistocercu), but have different sensitivity. For example, in Locustu the corresponding thresholds of the low-frequency receptor groups are 50-60 dB, 25-35 dB and 35-45 dB, their characteristic frequencies being, 3.5-4.0 kHz, 4 kHz, and 5.5-6.0 kHz, respectively. In Schistocercu the characteristic frequencies of the a-, b- and c-receptors were found to be 3.7 kHz, 3.5 kHz and 1.5-3.2 kHz, respectively (Fig. 21). The fourth group (d) is sensitive to high frequencies (5-40 kHz); the optimum range is 1&20 kHz where the thresholds are about 35-45 dB.
287
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
All tympanal organ receptors are tonic (very slow adaptation) so that under physiological conditions the duration of their discharge precisely codes the duration of sound pulses. Low-frequency receptors can synchronize their spikes during the rising phase of the signal in spite of their threshold differences. Popov and Svetlogorskaja (197 1) found a sub-receptor plexus /
nerve
folded
' body
J t f u s i f o r m body
u/
el evated
proci
pyritorh I vesicle (SOUND)
Fig. 21. The anatomy of the tympana1 organ and the threshold curves of the four receptor groups (a, b, c, d) in the isolated ear of the locust Schisrocercu greguriu. Broken lines on the graph indicate variations in threshold curves for different cells within each group. From Michelsen (197 1).
formed by the collaterals of the axons of the tympanal organ receptors in the tympanal nerve of Locusta migratoria. This plexus may be the place where the electronic interaction between receptors leading to this synchronization occurs. Romer and Schwartzkopff, 1975 (cf. Rheinlaender, 1975) observed coupled discharges of the tympanal fibres at the level of the metathoracic ganglion in locusts and Rheinlaender (1975) described this phenomenon in detail for
288
NORBERT ELSNER AND ANDREJ V POPOV
tettigonioids. Both the frequency of the occurrence of coupling and the degree of synchronization increased with increasing stimulus intensity. Coupled discharges were observed throughout the whole response, during spontaneous activity and during after-discharge. Thus, it is clear that the coupling cannot be the result of a purely mechanical phenomenon. An electrotonically synchronizing interaction between the auditory fibres will lead to a more precise temporal coding of the amplitude modulation pattern of the signal than can be done by simply summing the responses of independent receptors. Romer and Schwartzkopff (1975) showed that the tympanal organ receptors have sharply asymmetrical directional diagrams (Fig. 22). These are in good agreement with results obtained by Autrum et al. (1960) for the whole tympanic nerve response. The directionality of a single sensory unit is a function of sound intensity and frequency. The whole organ can code the direction of the sound source over a wide range of intensity levels due to the sensitivity distribution of receptors. Thus, the locust ear, due to a functional specialization of the different receptor groups, can code all of the most important properties of natural sounds. These are the intensity, the frequency spectrum and the amplitude modulation pattern. They provide the information necessary for the localization of the sound source. b Tettigonioidea The tympanal organ of tettigonioids consists of two separate groups of receptors-the intermediate organ and the crista .acoustics, both of which are involved in sound reception. The chordotonal sensilla of the crista acoustica occur in a long row one behind the other along the dorsal wall of the anterior tympanal air space. The length of their distal processes and the size of the cell bodies of the sensory neurons decrease gradually in a distal direction. The first anatomists suggested that this system was capable of frequency discrimination (Schwabe, 1906) but physiological proof was only obtained much later. Autrum (1 940, 194 1) demonstrated that the crista acoustica of Decticus is a specialized auditory organ working over a frequency range of 1-90 kHz with its maximal sensitivity in the ultrasound range. He also showed that- lowfrequency sounds of high intensity can be perceived by the subgenual organ (not strictly part of the receptor) indirectly by induced vibrations of the ground or the leg. Katsuki and Suga (1960) first showed 2 types of auditory units entering the prothoracic ganglion of Gampsocleis buergeri-ne, most sensitive to 10 kHz and the other to 6-7 kHz. Autrum (1960) suggested that these units belonged to the tympanal and subgenual organ, respectively. Zhantiev (1971) by selectively cutting the corresponding branches of the tympanal nerve in the vicinity of the receptor organ, showed in Decticus and in Tettigonia that after
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
289
elimination of .the crista acoustica, the summed responses of the tympanal nerve remained unchanged in the frequency range 1-10 kHz but sensitivity to frequencies above 8-10 kHz disappeared. Section of the nerve branch belonging to the subgenual organ had no effect. He concluded that ( i ) the subgenual organs could not perceive sound and (ii)the intermediate organ was responsible for the perception of low frequencies (1-lOkHz, optimum 4-
P
(a)
0’
(c)
30-1
1
’
30
T
I
I
LO
50
60
T
70
6
80
1
1
90
0-0
3b
O;
SO
W
,
.
,
70
80
90
Stimulus intensity CdBl
180’
Fig. 22. (a, b) Intensity curves of single low- and high-frequency receptors of Locusta migratoria (signal: pure tone of 4 kHz and 20 kHz, 100 ms duration). (c, d) Corresponding directional diagrams for 2 levels of intensity (indicated by arrows on the intensity curves). From Romer and Schwartzkopff (1975).
7 kHz with threshold about 55dB) and (iii) the crista acoustica was responsible for the perception of high frequencies (above 8-10 kHz, optimum at 10-12 kHz with corresponding threshold about 35 dB). By recordings at the level of the prothoracic ganglion Rheinlaender (1975) made a detailed analysis of the functional properties of single tympanal organ units in Decticus verruciuorus. He described two main groups of receptors (“a”, “b”) which are very different in many respects. Receptors of the “a”group have a V-shaped threshold frequency curve, each unit being tuned to
290
NORBERT ELSNER AND ANDREJ V POPOV
“just one frequency”. This characteristic frequency appeared to be different in different “a”-units, so that the spectrum is covered by a number of “a”receptors with overlapping frequency characteristics (Fig. 23). Further, “a”units with a similar characteristic frequency response have different absolute sensitivity, thresholds being in the range 25-55 dB SPL (Sound Pressure Level). Each “a”-receptor has a dynamic range of about 30 dB where the spike repetition rate is a function of the logarithm of the intensity. But since a number of “a”-receptors have different and nonoverlapping tuning curves, a
-1
3
5
I
10
12
16
20
30
LO
Stimulus frequency C kHz I
Fig. 23. Threshold curves of several receptor cells (primary auditory fibres) of the tettigonioid Decticus verrucivorus. Courtesy of J. Rheinlaender.
much wider intensity range (up to 70 dB) can be coded at any frequency. “a’,units have pronounced directional sensitivity at low and intermediate intensities (Fig. 24). On the other hand, “b”-units have very flat threshold-frequency curves without clear optima, but their absolute sensitivity is not lower than that of “a”-units. They have no directional sensitivity at any sound frequency or intensity. Both “a”- and “b”-receptors are tonic. On morphological and physiological evidence Rheinlaender ( 1975) suggested that “a”- and “b”-units belong to the crista acoustica and the intermediate organ, respectively. The low-frequency receptors are thought to be located in the proximal part and the high-frequency receptors in the distal part of the crista acoustica, though direct proof is still lacking. Rheinlaender (1975) thinks that “side effects of the
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
29 1
surgical procedure affected the results of Zhantiev (1971), since the small size and relatively close association of the 2 organs make it difficult to destroy one without damage to the other”. c Gryllidae Two separate sense organs are located in the forelegs of crickets in close proximity to the tympanal membranes. These are the subgenual and tracheal organ, each of which can be further separated into proximal and distal parts. The attachment cells of the proximal receptors of the tracheal organ insert on large accessory cells, and those of distal receptors insert on modified dorsal hypodermal cells. According to Young and Ball
+-
response [spikes/20rns stimulus]
response [spi kes/20 m s stimulus]
- - -
/
-
,-I
7 la
- 1
-
sound pressure level [dB] 05d0 75dB
12kHz
6540
180° Fig. 24. Intensity characteristic and directional diagrams for a group a receptor unit of the tettigonioid Decticus uerruciuorus. The directional characteristic was measured for 3 different intensities (65,75 and 85 dB, see arrows). Repetition rate 2/s. From Rheinlaender (1975).
(1974) these groups of the tympanal (tracheal) organ receptors can be further divided into 5 main types, each containing a relatively constant number of scolopidia (each tracheal organ has about 70 scolopidia). These types differ consistently in the location of the sensory neuron, the orientation of the dendrite and scolopale cell, the structure of the scolopale cell, the shape and orientation of the attachment cell and its connection to the secondary attachment cell. Physiological correlates of these morphological specializations are not yet completely established. The functional differentiation of the tympanal organ
292
NORBERT ELSNER AND ANDREJ V POPOV
receptors of crickets into 2 low-frequency groups (optimum around 0.82.5 kHz and 4-6 kHz, respectively) and a high-frequency group (optimum above 10 kHz) was first suggested by investigations of central auditory units which showed different frequency responses (Horridge, 1960; Popov 1969, 1971, 1973). Nocke (1972) recorded from bundles of tympanal nerve fibres in Gryllus campestris and came to similar conclusions. He showed additionally, that low-frequency receptors with an optimum below 4 kHz responded to vibrations as well and suggested that they belonged to the subgenual organ. By selective destructidn of the tympana, he found that the activity of highfrequency units is functionally dependent upon the small tympanal membrane and that of low-frequency units on the large tympanum. Zhantiev and Tshukanov (1972) showed by selective dissection of tympanal nerve branches that 4 kHz receptors originate from proximal, and highfrequency receptors from distal groups of tracheal organ receptors. Markovich (1976) recorded single and quasi-single unit activity from the tympanal nerve fibres of Gryllus bimaculatus at the level of the prothoracic ganglion. Two types of units, similar to those found by Rheinlaender (1975) in tettigonioids, were definitely identified: receptors with a sharp optimum around 4-5.5 kHz and broad-band receptors with wide and flat frequency curves extending from at least 3 - 4 kHz to 35-40 kHz (Fig. 25a). Pure high-frequency receptors, if present, have not yet been found. All the units appeared to be tonic though the high-frequency (=broad band) units showed clear adaptations to very long stationary sounds (Nocke, 1972). Since 5 kHz units are tuned to the spectrum of the conspecific calling song their properties have been studied in detail. Figure 25b shows the sensitivity curves (=action spectra) of such a receptor unit and Fig. 25c shows the efficiency curves (=response spectra) of the same unit for intensities of 60, 70, 80 and 90 dB SPL. The sharp tuning of the receptor can be seen only at low and moderate intensities; the unit becomes less selective to frequency at high intensities. The receptor is purely tonic in its responses to all frequencies and the number of spikes per stimulus increases as a logarithmic function of the stimulus intensity in the range of 35-40 dB above threshold. The latency period increases from 5-6 ms to 2CL25 ms when the intensity decreases from 35-40 dB to 5-10 dB above threshold. d Conclusions In summarizing the data above the functional organization of orthopteran auditory organs one sees that although the anatomy is very different in the different families, the principles used for the coding of the information about the particular properties of the sound signals are common and similar to those of vertebrates. The intensity of the sound is coded in two ways. First, the output of each receptor is linear with the logarithm of sound intensity in a definite range
293
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
dB 100
.
3\
ao 60
4c
0
4
8
20
16
12
25
30
35
40
50 20
5
0
4
8
1
2
7
6
2
0
kHz
Fig. 25. (a) Threshold curves of primary auditory fibres of the two crickets Gryllus cumpestris (curve 1, from Nocke, 1972) and G. bimuculatus (curves 2-4, from Markovich, 1976). (b) The sensitivity curves of a low-frequency receptor fibre of Gryllus bimuculatus. On the ordinate: sound intensity (in dB SPL) necessary to evoke definite response of the receptor cell (5, 20, 30 and 40 imp/stim of 100 ms) at different frequencies. (c) The efficiency curves of a low-frequency receptor of G. bimaculutus (recording at the level of the prothoracic ganglion). On the ordinate: number of impulses (n in the response of the receptor unit to signals of different frequency and intensity levels of 60,70,80 and 90 dB SPL;signal duration 100 ms.
(normally 25-40 dB above threshold). Second, the receptors are distributed over a range of absolute sensitivity, so that the whole organ can cover about 70-85 dB. The maximal absolute sensitivity of tympanal organs in different families of Orchopterms is similar. In all these groups we find species with thresholds as low as 5-15 dB SPL at main frequencies. Different techniques have been used for measuring the sensitivity of tympanal organs of insects (i.e. the summed response of the tympanal organ, single unit recording from
294
NORBERT ELSNER AND ANDREJ V POPOV
receptor axons or from central auditory interneurons, behavioural measurements), and the values obtained are often very different. Single unit recording andlor behavioural tests are most sensitive. For example, in Acheta domesticus the lowest threshold of the tympanal nerve response is about 35-40 dB SPL, but some central neurons have thresholds as low as 20-25 dB SPL and the behavioural response threshold for the same 4 kHz signal is about 15-18 dB SPL (Popov, 1969; Shuvalov and Popov, 1971 and unpublished data). In Locusta migratoria, the lowest threshold of the tympanal nerve response measured by Adam and Schwartzkopff (1967) by recording from the nerve close to the metathoracic ganglion was 32-40 dB, while in Popov’s measurements (1965, 1968) when the electrode was close to the Miiller-organ this threshold was about 13-15 dB SPL. Romer and Schwartzkopff (1975) tested the sensitivity of single receptors in the same animal and found the lowest thresholds to be about 25 dB SPL, while some ventral cord neurons have thresholds as low as 20 dB SPL (Popov, 1971a). Michelsen ( 1971~ ) showed that the sensitivity of the Schistocerca ear can vary by about 2030 dB, depending upon the amount of fat between the ears. Other explanations are possible, especially concerning the experimental procedure in each case. This high sensitivity and wide dynamic range of the tympanal organs can ensure that intraspecific acoustic communication occurs over large distances (up to tens of meters) even in cases when the intensity of the emitted sound is low as, for example, in acridid grasshoppers. At close range, when the sound intensity is high (about 70-90dB SPL), other sense organs may also be involved (Jones and Dambach, 1973). The frequency range covered by the tympanal organ of a given species is normally much wider than is necessary for the perception of intraspecific signals, but in all cases there is at least one group of receptors precisely tuned to the best frequencies of conspecific signals (Orthoptera, Cicadidae) or the signals of predators (Lepidoptera). In all Orthoptera several receptors with different frequency responses occur, so that a certain degree of frequency discrimination is possible in the tympanal organ. The amplitude modulation pattern of insect sounds can be precisely followed by the tympanal organ receptors of Orthopterans and Cicadidae because they discharge tonically without after-discharge and with good time resolution, at least to signals with a pulse repetition rate below 100 Hz and intervals of more than 5-10 ms. The high repetition rate of pulses in the calling songs of some cicadas (above 100-200 Hz) or the occurrence of signals with a high pulse repetition rate in some Acrididae and Tettigonioidea may have no value for the recognition process, because these properties may already be lost at the receptor level. Synchronization of the spike discharges of the receptors may occur at moments of sharp increase in sound intensity. This greatly improves the time-
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
295
resolution of the amplitude modulation pattern by the succeeding levels of the auditory system. This synchronization may be brought about by electrotonic coupling of the receptors in a subreceptor plexus of axon collaterals as described for the locust (Popov and Svetlogorskaja, 197 1) and recently found in the prothoracic auditory organ of crickets (Svetlogorskaja, unpublished). Due to the functional differentiation of receptors within one tympanal organ, i.e. the different orientation of their response areas in the intensity-frequency field, the combination of receptors activated, and the correlation between the levels and patterns of their activity, appear to be specific for any sound in a biotope. This serves as a basis for further discrimination processes in the CNS. No one receptor gives an invariant description of any one parameter of the sound signal, because its output can be modified by several parameters. However, this information can be extracted by central neurons which integrate the output of several receptors. Recent studies of the biophysics of the tympanal organs of insects show that this functional differentiation of receptors, and the tuning of the tympanal organ as a whole, depends, not on the specificity of the receptors themselves, but on the structure and acoustic properties of the mechanical soundtransforming system to which the receptors are connected. This was first clearly demonstrated by Michelsen (1971, 1974), Michelsen and Nocke (1974) for the locust ear, then by Adams (1972) for the moths and recently by Lewis (1974), Lewis er al. (1975) and Nocke (1974, 1975) for the ensiferans. Frequency discrimination in the tympanal organs of locusts was found to be based on resonances of the tympanum, the thin part of which can vibrate independently of the entire tympanum. The resulting sets of resonances interact in a definite way on the surface of the tympanum, so that the position of maximal vibrations induced by the sound is dependent upon its frequency. In so far as 4 groups of receptors are attached to 4 separate specialized areas on the tympanum, their tuning curves are different. Thus, frequency discrimination seems to be a purely physical phenomenon, like that in the cochlea of vertebrates. The dynamic properties of the receptors seem to depend upon the receptor itself. The sensitivity of the locust ear to low frequencies can be significantly modified by several factors, such as fat content of the animal, position of internal tissue behind the ears, etc. In the range 2-5 kHz, the ear appears to respond as a combined pressure and pressure-gradient receiver (Miller, 1974). At frequencies above 10 kHz the ear acts as a pressure receptor due to the high absorptbn of sound by thejnternal tissues, and in this range, the directionality of the ear depends only upon the diffraction pattern of sound at the body surface. In tettigonioids and gryllids it has been found that the prothoracic spiracle is the preferential site of sound entry to the tympanal organ and nor the tympanal
NORBERT ELSNER AND ANDREJ V. POPOV
296
slits or membranes. The tuning of the entire organ is determined by the properties of the acoustic trachea. Lewis (1974) regards this trachea as an exponential horn acting as a broad-band receiver, thus amplifying all frequencies above the lower cut-off frequency. Nocke (1975), on the other hand, considers it to be a resonator intermediate between an open and closed tube and amplifying only a narrow band of frequencies. The presence of the pronotal shield close to the acoustic spiracle is an additional factor in determining the high-frequency directionality of the ear. When it is removed this directionality is eliminated (Lewis et al., 1975). The appearance of the tympanal slits in tettigonioids might be connected with the necessity to maintain high sensitivity of the organ at high frequencies comparable to the sensitivity of the open membranes at low frequencies (Lewis, 1974b). Recently, biophysical measurements of the vibration patterns of the tympanal membranes of Tettigonia viridissima, Homorocoryphus nitidulus vicinus and Gryllus campestris have confirmed the importance of the spiracular entry for sound and clearly established that these systems work as pressure-gradient receivers. They, therefore, have an inherent directionality in addition to that imposed by the pronotal shield. However, we are still far from understanding all physiological mechanisms of sound reception in the tympanal organs of tettigonioids and gryllids since the oscillatory properties of the trachea wall on which the receptors of the crista ucoustica are situated have not been studied in detail as yet. Dragsten et al. (1974) measured the vibrations of the large tympanal membranes of crickets using very sensitive optical heterodyne spectroscopy. They showed that the mechanical response of the posterior membrane is sharply tuned to 5.5 kHz and is linear over vibration amplitudes from 0.01 to 50nm for frequencies from 2 to 20 kHz. The neural threshold response appears at a displacement of about 0.1 nm. The membrane vibrates in a simple mode at both 5 and 18 kHz. But how these vibrations are transformed at the level of small anterior trachea is still unknown. When the tympanal organ is stimulated by direct mechanical vibrations applied to the tympanum (Adams, 1972 on Lepidoptera; Lewis et al., 1975 on Tettigonioidea; Michelsen, 1974 on Acrididae), the receptor cells show clear responses to much lower frequencies than are effective during normal sound stimulation. This is an additional indication that the tuning of the organ is determined by the mechanical system. 3.3
INFORMATION PROCESSING BY AUDITORY NEURONS OF HIGHER ORDER
3.3.1
Central projections of the tympanal nervefibres
Central projections of the tympanal nerve fibres were recently studied in Orthopterans (Popov, 1967a; Rehbein, 1973, 1975; Rehbein et al., 1974; Eibl,
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
297
Fig. 26. The anatomy of two auditory ventral-cord neurons of Locusta migratoria, and the course of the receptor fibres in the thoracic ventral cord (shaded areas), B,, ascending neuron; GI, T-shape neuron. SEG,TH,+ subesophageal, pro-, meso-, metathoracic ganglion. Courtesy of H. Rehbein.
1974; Svetlogorskaja and Popov, 1975) and noctuid moths (Paul, 1973). In acridids, the tympana1 nerve fibres enter the metathoracic ganglion and there form 2 distinct synaptic areas, the caudal and the frontal acoustic neuropile (Fig. 26). About 80 per cent of the fibres send collaterals to the mesothoracic ganglion to form a third, frontal neuropile on the same side. Some receptors also send collaterals to the prothoracic ganglion to form, probably, a fourth acoustic neuropile. These neuropiles are not purely acoustic, since sensory fibres of other segmental nerves also end there (Popov, 1967a).
298
NORBERT ELSNER AND ANDREJ V POPOV
By extracellular staining of single receptor fibres, whose activity was recorded by microelectrodes filled with CoCl,, Rehbein el al. (1974) found that low-frequency tympanal organ receptors of Locusta have projections to both caudal and frontal acoustic neuropiles of the metathoracic ganglion while those of high-frequency receptors end only in the frontal neuropile. By selective staining of separate groups of the tympanal nerve fibres with CoCI,, Rehbein (1 975) showed that auditory fibres with different physiological properties have different projections within the frontal metathoracic neuropile and are not evenly distributed there. All projections of the tympanal nerve fibres are ipsilateral. Svetlogorskaja and Popov ( 1975) described the ultrastructural organization of the frontal acoustic neuropile of the locust. Two types of synaptic endings were found. Most of them contain clear round vesicles of 40-45 nm in diameter. Such endings form typical synapses with dense pre- and postsynaptic membranes and synaptic clefts of about 15-20 nm, similar to those found in mushroom bodies (Mancini and Frontali, 1967; Lamparter et al., 1969; Schurmann, 1970), abdominal (Smith, 1965) and thoracic (Mandelshtam, 1974) ganglia, and deutocerebral glomeruli (Boeckh et al., 1970) of insects, and are supposed to be cholinergic. Four types of contacts of these endings with postsynaptic fibres were found: 1 :1 synapses, convergent, divergent and serial, thus complex interactions can take place here between the receptor fibres and second-order neurons. Some endings contain both dense-core and clear vesicles in different relations. They are distributed evenly on the ventral side of the frontal acoustic neuropile close to the border with the ventral nonauditory neuropile which is filled with endings with dense-core vesicles. This close association of the auditory neuropile with a presumed neurosecretory region may be the basis of a neuro-hormonal control of the afferent system. Such an accumulation of neurosecretory endings was never seen in the middle or dorsal (mainly motor) areas of the ganglion. In noctuid moths (Paul, 1973), the axon of the A,-receptor cell has ipsilateral projections to the ventral neuropiles of all 3 thoracic ganglia, as in the locust, but in tettigonioids and in crickets all the tympanal nerve fibres end in a well-defined ipsilateral ventro-medial neuropile of the prothoracic ganglion (Rehbein, 1973). This acoustic neuropile consists of smaller posterior and larger anterior parts, which are occupied, according to Rehbein (1973) and Eibl (1974) by the same bifurcating tympanal nerve fibres. However, Popov et al. (1976a) using extracellular staining of single tympanal nerve fibres with CoC1,-filled electrodes have found that the 4-5 kHz receptors at least have 2 modifications of their terminal branches. One group ends only in the anterior part of the auditory neuropile while the others bifurcate giving terminal branches to the anterior region and to a region intermediate between the anterior and posterior parts of the auditory neuropile (Fig. 27a). Broad-band or
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
299
Fig. 27 (a). Tympana1 nerve projections in the prothoracic ganglion of Gryllus bimaculatus revealed by extracellular staining of single receptor fibres (left-side) and passive filling of axons with CoCI, (right-side). Left-side fibres and probably the medial-right-side fibre belong to the 45 kHz receptors; the right-side bifurcating fibre belongs to the broad-band receptor. (b) Large segmental auditory neuron in the prothoracic ganglion of G. bimaculatus. Somatic side is postsynaptic. The axon crosses the midline over the anterior commissural bridge. AK, PK, anterior and posterior connectives; LN, leg nerve. From Popov et al. (1976a).
NORBERT ELSNER AND ANDREJ V. POPOV
300
high-frequency receptors were not identified by this method, but it is reasonable to suppose that they have extensive branching in the posterior region of the auditory neuropile and laterally along the course of the tympanal nerve bundle, because all high-frequency- and broad-band higher-order neurons have branches there (see below). By passive filling of the tympanal nerve fibres with CoCl,, 2 main morphological types were found (Fig. 27a, right side). These are believed to represent 4-5 kHz and broad-band receptors respectively. At least some receptors give numerous, very short, side branches on their way from the nerve entry to the acoustic neuropile. Although looked for, in no case have direct connections of the tympanal nerve fibres to motor or associate neuropiles been found. This is contrary to the situation found for the wind stretch receptor of locusts (Burrows, 1974) and for the tarsal receptors (Popov and Svetlogorskaja, unpublished) which have profound monosynaptic connections with motoneurons. 3.3.2
Ventral cord auditory neurons
General anatomy: Due to the development of new techniques of intracellular (Pitman et al., 1972) and extracellular (Rehbein et al., 1974) staining of insect central neurons during electrophysiological experiments it has become possible in some cases, to identify single neurons and to correlate their neuronal topography with their physiological properties. These methods were recently applied to the auditory neurons (including cercal giants) of the locust (Rehbein et al., 1974; Kalmring, 1975a; Rehbein, 1975; Potente 1975) and of crickets (Rheinlaender et al., 1976; Popov et al., 1976a,b,c). The ventral cord auditory neurons which process the information coming from the tympanal or cercal receptors of both sides can be classified into 5 separate types according to their connections to other parts of the central nervous system. These are (i) segmental auditory neurons which distribute auditory information within one ganglion, (ii) ascending auditory neurons which send axons to the anterior ganglia, in most cases up to the brain (iii) descending auditory neurons sending axons to the posterior ganglia, ( i v ) Tshape neurons having descending and ascending axons and thus supplying both thoracic ganglia and the brain with auditory information, ( u ) “throughpassing” neurons, coming from the brain or ascending from the abdominal cord, which receive additional inputs from the auditory neuropiles through lateral branches of their axons (see below). Suga and Katsuki (1961) described large central auditory neurons of tettigonioids and interpreted them physiologically as T-shaped. However, morphologically they are ascending neurons which have their somata in the abdominal cord, presumably in the last abdominal ganglion (Kalmring, pers. comm.). They receive auditory input in the prothoracic ganglion through lateral axonal branches and transmit in both
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
30 1
directions. According to our classification (in relation to auditory input) they have to be considered as “through-passing” neurons.
Segmental auditory neurons: A large segmental auditory neuron identified by Popov et al. (1976a) in the prothoracic ganglion of Gryllus campestris is shown in Fig. 27b. This neuron receives information from all parts of the auditory neuropile of one side (ipsilateral to the soma) and sends it to all parts of the contralateral auditory neuropile. Dendritic terminals are much thinner than the axon and lack boutons which are characteristic of presynaptic endings (similar relations were found in acridids, see Rehbein, 1975). Anterior and posterio-lateral parts of the contralateral auditory neuropile are supplied by different axonal branches of this neuron. A symmetrical neuron of the other side lies very close to this one, so that they can be recorded together by the same electrode. This neuron has no branches outside the auditory neuropiles and does not respond to other modalities of mechanical stimulation. The neuron is responsive in the range 2-3 to 40 kHz, the tuning curve being highly dependent upon both sound intensity and the direction of incidence of sound (Fig. 28a). At low intensities, at frontal, ipsi- and contralateral stimulation it is sharply tuned to 4.5-5.5 kHz; during caudal stimulation a second peak in the region of 20-25 kHz appears. The higher the intensity, the broader is the response curve. Directional properties are well expressed only for optimal low frequencies (4-5 kHz, ipsilateral, caudal and frontal stimulation being more effective than contralateral), and for optimal high-frequencies (18-25 kHz, caudal and contralateral stimulation being more effective than ipsilateral and frontal). This neuron is inhibited by frequencies below 3.5 kHz. The inhibition is mediated through the corresponding low-frequency receptors of the leg auditory organ. When a chirping sound is used as a stimulus, this neuron responds clearly to every pulse of the chirp, when these are only a few dB above the threshold intensity (Fig. 28b). The number of spikes per pulse increases as a linear function of the logarithm of sound intensity in the range at least 50 dB above threshold, the neuron being most sensitive to a pulse rate corresponding to that of the conspecific calling song. The activity level of this neuron can be modulated by central phasic inputs. Thus, it is sharply inhibited during voluntary movements of the animal and during corresponding bursts of activity in descending fibres and motoneurons (Fig. 28b, arrow). Sometimes inhibition, induced by sound and mediated by other auditory neurons, is switched on and then suddenly dissappears. Spontaneous activity, if any, is low and irregular. By means of these large segmental neurons, each half of the ganglion receives integrated information from both tympana1 organs, though receptor
302
NORBERT ELSNER AND ANDREJ V. POPOV
(a)
(b) 28
frontal
38
46
94
caudal 84
74
'I
A
cont ralateral
Fig. 28. (a) Constant-intensity tuning curves of the large segmental auditory neuron of the cricket Gryllus birnnculatus (recording from the prothoracic ganglion). On the ordinates: the number of spikes (n) per stimulus (pure tone, 100 ms) of given intensity (dB) and direction of sound incidence. (b) The response of the large segmental. auditory neuron to chirping sounds. Intensity is indicated on the left side; pulse period in ms is indicated above each column. Pulse duration is 26 ms, carrier frequency is 4.5 kHz; chirp rate is 2.5 Hz; time calibration is 5 0 Hz. Dots of the left row on each oscillogram indicate the beginning of the chirps. From Popov et al. (1976a).
projections are strictly ipsilateral, and subsequent comparison of bilateral inputs can be done even by the unilateral higher-order neurons of each side.
Ascending and T-shape auditory neurons: Anatomy and general physiology. These neurons transmit auditory information to the brain, integrating inputs from different afferent systems and controlling behaviour. Hence, the information processing they do is of great importance for understanding the phonoresponsive specificity described above. The structure of such neurons in locusts are shown in Fig. 26. According to Rehbein (Rehbein et al., 1974; Rehbein, 1975) and Kalmring (1975a) dendritic branches of these neurons are situated in the segment where the soma occurs and are close to it. The axons ascend the nerve-cord on the side contralateral to
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
303
the soma giving side branches, in each ganglion, to the medial associative neuropiles and, sometimes (G-neuron), to the lateral dorsal neuropiles of thoracic ganglia where they have direct connections with motoneurons. Cercal giants have similar connections (Potente, 1975). Rehbein (1975) and Kalmring (1975a) claim that in most ventral-cord auditory neurons, dendritic branches are located outside the primary auditory neuropiles and hence they consider them to be not second-order neurons, but higher-order auditory neurons and suggest that a number of segmental interneurons should be intercalated between them and the receptor neurons. To account for the response patterns of these neurons Kalmring (1975a) assumes the presence of at least 7-9 interneurons, though none of them has been identified. In crickets, at least to date, identified ventral-cord auditory neurons have direct connections with tympana1 organ receptors and can be regarded as second-order auditory neurons, though additional connections with interneurons are highly probable (Popov et al., 1976a,b,c). The same is true for the cercal giants of the locust (Potente, 1975). The number of ascending and T-shape auditory fibres entering the brain in each animal is uncertain since, first, in most cases only the largest fibres have been studied and, second, very often only physiological criteria were used for classification. Bearing in mind the large variability of response due to many internal and external factors, physiological criteria alone are not sufficient. Only few neurons have been identified with certainty, i.e. both physiologically and morphologically (locusts: Rehbein et al., 1974; Rehbein, 1975; Kalmring, 1975a; crickets: Popov et al., 1976a,b,c). Rough estimates give numbers around 10 on each side of the ventral cord. Axonal side branches of ascending and T-shape auditory neurons have probably an additional filtering function, the branching point having a low safety factor. Thus, Kalmring (1976 pers. comm.) succeeded to record simultaneously from the axon and axonal side branch of the same neuron and found that at certain spike frequencies the branch fails to respond (Fig. 29). If this phenomenon is widespread, different “addresses” connected to this neuron can receive different inputs from the same neuron. In most cases, these axonal branches seem to be presynaptic (delivering), but this is probably not always the case. Some “through-passing” neurons of tettigonioids receive auditory inputs in the prothoracic ganglion through such axonal branches (Kalmring, 1976, pers. comm.). If this is true for at least some side branches of auditory neurons, the information going to the brain or to posterior ganglia can be modified at certain regions in the cord by other inputs. At least some of the ascending auditory neurons are polymodal, i.e. can be activated by stimuli other than sound. For example, B- and C-ascending neurons of the locust are excited by cercal stimulation (Kalmring, 1975a); vibratory stimulation of the legs causes an inhibition of the B-neuron and an
NORBERT ELSNER AND ANDREJ V. p o p o v
304
activation of the G-neuron through the same dendritic branches as auditory stimulation does (Kalmring, 1976, pers. comm.). Yanagisawa et al. (1967) demonstrated inhibition of the locust ascending auditory neurons by
Ampl. 100 (%A1
Spike-Filtering
90 dB 80 dB
70 dB
50
I I
10
60 dB
50 dB 40 dB
Interval [ms] I
I
I
I
1
20
30
40
50
60
Fig. 29. Response characteristics of the main axon and the axonal side branch of the B, auditory neuron of Locusta migrutoria to pure tones of 12 kHz and 20 ms duration. Recordings are from the axonal branching in the subesophageal ganglion (arrow). Inserted oscillogrums: Small spikes (arrows) belong to the axonal branch; they occur only, when the instantaneous impulse frequency is lower than 100 Hz.On the graph: percentage of occurrence of small spikes (A,) as a function of interspike intervals. At intervals shorter than 10 ms, the branch fails to respond at all. Courtesy of K. Kalmring.
unidentified low-frequency receptors. Some ventral-cord auditory neurons of crickets can be activated or inhibited by stimulation with an air-stream directed at the thorax (Popov and Rehbein; Popov and Markovich, unpubl.) Therefore, it might not be surprising that a stimulus received by the hair sensilla of the locust may elicit from females the same stridulation response movements as
-
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
305
given to stimuli received via the tympanal organs (Haskell, 1958) or that male crickets without tympana1 organs can alternate singing activity (Jones and Dambach, 1973), because several sense organs can stimulate (and normally stimulate at close range) the same ascending pathways with the same timepatterns. As mentioned above, these neurons are connected to the motor system. Burrows (cf. Rehbein, 1975) found 1 : 1 correlation between spikes of auditory neurons of the locust and postsynaptic potentials of certain motoneurons indicating monosynaptic interactions between them. Weber and Popov (unpubl.) showed that sound stimulation can activate leg and flight muscles in crickets with both neck connectives cut, when the excitability of the animal was
Fig. 30. Activation of respiratory interneurons by auditory input. The recordings were made in the prothoracic ganglion of GryIlus birnaculatus. SA, spontaneous bursting activity of large and small units. Note that EPSPs and spike activity of the large unit last much longer than the sound signal, especially during high-frequency stimulation, and is followed by IPSPs. From Popov and Markovich (unpublished).
high enough. This supports Burrows’ observations on the regulation of motoneurons by thoracic auditory interneurons. Besides direct monosynaptic interaction with motoneurons these auditory neurons can control motor activity polysynaptically, by modifying the activity of descending command fibres. Figure 30 shows an example of the activation of 2 command respiratory fibres by sound stimulation at the level of the prothoracic ganglion of the cricket. In this case it was possible to reset respiratory activity or to modify respiratory rate by appropriate chirp rate and intensity (Popov and Markovich, unpubl.). The activity of the ascending auditory neurons is under central nervous control. McKay (1970) found in a tettigonioid, that the response level of a Tfibre is controlled by an inhibitory input originating within the posterior thoracic ganglion and acting synaptically within the prothoracic ganglion. The rate of habituation is also under the control of central factors which may be
306
NORBERT ELSNER AND ANDREJ V. POPOV
different from those which affect the response level of the interneuron. Rowell and McKay (1969b) showed that the brain controls the rate of habituation and response level of ascending auditory neurons of the locust. Low- and highfrequency ventral cord neurons of crickets can be respectively inhibited and excited phasically during the animal’s own movements (Popov et al., 1976b,c). Stout and Huber (1972) found that the activity of some ascending fibres is periodically suppressed during certain respiratory phases.
Vc-Sup E G - l e r d
Vc-Sup EG-level
(b)
100
50
I
’,-
/ I
/
35
55
I5
95
35
55
- 75
95
sound pressure level [dB]
Fig. 3 1. Basic shapes of intensity curves of receptor units (a) and neurons of the ventral cord and the supraesophageal ganglion (b, c) of Decticus uerruciuorus under ipsilateral stimulation. From Rheinlaender (1 975).
Functional characteristics. The functional properties of ascending and T-shape ventral-cord auditory neurons of insects were studied by several authors (Acrididae: Horridge, 1961; Popov, 1967a,b, 1969, 197lb; Kalmring, 1971, 1975a,b; Kalmring et al., 1972a,b; Rowell and McKay, 1969; Tettigonioidea: Suga and Katsuki, 1961; Suga, 1963; McKay, 1969, 1970; Zhantiev, 1971; Rheinlaender and Kalmring, 1973; Rheinlaender, 1975; Gryllidae: Horridge, 1960; Suga, 1963; Popov, 1969, 1971, 1973; Popov et al., 1974, 1975, 1976b,c; Zhantiev and Chukanov, 1972; Zhantiev et al., 1975; Stout and Huber, 1972; Lepidoptera: Roeder, 1966; Paul, 1974). Each neuron is specific in felation to the stimulus parameters it can measure and normally every functional type is represented by a single neuron on each side of the cord. The processing of information about certain parameters of the sound stimuli (intensity frequency spectrum, direction and time-parameters)
307
NEU ROETHO LOGY OF ACOU STlC COMMUNICATION
performed by these neurons, are similar enough in the different families of insects studied, to consider them together. Intensity. Intensity coding at this level is characterized by additional, more complex means compared to that at receptor cells (Fig. 31). Some neurons have very limited intensity dynamic ranges (10-15 dB) but very steep intensity curves, rising monotonically up to saturation, so that a sharper discrimination of intensity changes in this narrow range is possible (for examples see: Rheinlaender, 1975; Popov et al., 1975; see also Fig. 31b, continuous lines).
12 kHz
13 12 11
10 9 8 7 6 5
4 3 2 1 1
30
40
50
60
70
80
90
100
Fig. 32. Intensity curves of an F,-neuron (Locustu rnigrutoriu) with ipsilateral and contralateral sound presentation before (heavy, solid curve i/o and dashed curve c/o, respectively) and after (thin, solid curve i/c and dashed curve c/c) removal of the contralateral tympanal organ. Vertically shaded area: inhibitory effect of the contralateral tympanal organ during ipsilateral stimulation; diagonally shaded area: inhibitory effect of the contralateral tympanal organ during contralateral stimulation. From Kalmring (1975).
Other neurons have intensity curves rising monotonically without saturation over a very large intensity range (50-70 dB) (see Fig. 3 lb, dotted lines). The precision of intensity coding by these neurons is lower than in the receptor units since they cover the whole biological intensity range. This type of transferfunction is the most common, and has been found in all groups of insects studied so far. The mechanism seems to depend on the integration of the excitatory inputs from all the receptor units of one or both sides.
308
NORBERT ELSNER AND ANDREJ V POPOV
Still other neurons have bell-shaped, nonmonotonical, intensity curves (Fig. 3 lc), indicating a preferential response in a very limited intensity range. Such neurons are common in the ventral cord of acridids (Popov, 1967a,b; Kalmring, 1975a) tettigonioids (Rheinlaender, 1975) and cicadas (Popov, unpubl.) but to much lesser extent, if at all, in crickets. The bell-shape of the intensity curve can be the result of contralateral inhibition (as in F-neurons of the locust; Kalmring, 1975a, see Fig. 32) or ipsilateral inhibitory interactions (,‘L”- and “S”-neurons of the locust; Popov, 1967a). In some broad-band ventral-cord auditory neurons of locusts and tettigoniids intensity transfer-functions can be different in different frequency ranges due to differential inhibitory interactions (Popov, 1967a, b; Kalmring, 1975; Rheinlaender, 1975). As a result the forms of the preferred response
85
65.
65.
65.
rg.
45.
25
25-
areas of such neurons in the intensity-frequency field are complex; Fig. 33 gives an example for tettigonioids. In most (especially the narrow-band) central auditory neurons of crickets and cicadas (Popov et al., 1976b,c; Popov, unpubl.) the frequency dependence of the intensity transfer-functions is minimal, and is expressed only by a slightly increased steepness of the intensity curves at side bands compared with that for optimal frequencies. Frequency-spectrum. Ventral-cord auditory neurons of orthopterans can be classified, according to their frequency response areas, into several distinct groups. In crickets, 6 types of neurons were found at the level of the prothoracic ganglion. They showed different tuning curves, due to their differential connections with certain groups of receptors (Popov, 1973), at least two of them ascend to the brain. Thus, information about different sound frequency bands is transmitted by different channels, similar to the situation in colour vision.
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
309
In locusts and tettigonioids most central auditory neurons are broadband, integrating the activity of frequency-selective receptors. However, due to differential excitatory and inhibitory interactions in different intensity ranges, they have different preferred areas in the intensity-frequency field (see above). In addition to broad-band neurons, a narrow-band 5 kHz-unit, and a noisedetector unit were found in the locust (Popov, 1969, 1971b, 1967b). The latter does not respond to pure tones but only to noises which stimulates both lowand high-frequency receptors. Adam and Schwartzkoff (1967) also described a high-frequency unit in the protocerebrum of the locust. The relative loss of frequency selectivity in the central regions of the auditory pathways in tettigonioids can be regarded, according to Rheinlaender (1975), as an adaptation for the improvement of other auditory functions, i.e. broad-band signal transmission and high sensitivity in the whole frequency range, when the receptor organ works on the resonance principle. This interpretation seems to be reasonable for tettigonioids with nonresonant songs, such as Decticus verrucivorus, but it is questionable whether the situation is the same in other tettigonioids with resonant songs.
Sound source localization. Information about the localization of the sound source is processed by a set of ventral-cord neurons differing in their directional sensitivity, from direction-insensitive neurons to sharply directional units responding only to ipsilateral stimulation (Kalmring, et af., 1972b; Kalmring, 1975a; Rheinlaender, 1975). The highly directional sensitivity is the result of the receptor’s directional properties strongly enhanced by inhibition from the contralateral -tympana1 organ. This phenomenon was first described by Katsuki and Suga (1961) for the tettigonioid Gampsocleis buergeri, and later confirmed for other Orthopterans (Suga, 1963;McKay, 1969; Rheinlaender el al., 1972; Rowel1 and McKay, 1969; Kalmring et af., 1972b; Zhantiev et af., 1975). The directional sensitivity of most ventral-cord auditory neurons is frequency- and intensity-dependent but can be represented in a considerably broader range of intensity and frequency than that of the receptor units, due to integration of the activity of several receptor groups. As a result of peripheral directionality and central bilateral interactions, the 20 dB difference in excitation between ipsilateral and contralateral stimulation found at the level of the receptor organ can be increased in ventral-cord neurons to 50-60 dB (Rheinlaender, 1975). Insects with both ears intact instantly locate the sound source and move directly to it; crickets with one t a r intact and the other destroyed can still find the sound source, utilizing the directional properties of one ear. However, they show greater deviations from the straight path than do the intact animals: movements to the operated side alternates with corrective movements (Popov, et al., in prep.).
310
NORBERT ELSNER AND ANDREJ V POPOV
Amplitude modulation pattern. The time patterns of complex communication songs are encoded by different ventral-cord neurons in different ways (Popov 1973; Popov et al., 1974, 1975; Rheinlaender, et al., 1976; Zhantiev and Chukanov, 1972; Stout and Huber, 1972). Four types of neurons appear to be important for the recognition processing of these sounds: (1) The first type is specialized for the detailed description of all the important features of the amplitude-modulation pattern of conspecific calling and aggressive songs, and is sharply tuned to the best frequency of these sounds (4-5.5 kHz). This is the so-called low-frequency neuron (LF) described by Popov et al. (1975) and Rheinlaender et al. (1976) in Gyllus bimaculatus, see Fig. 34. This neuron has a high level of spontaneous activity (up to 20-30 spike&) which is totally suppressed during sound presentation, so improving signal-to-noise-ratio to few dB above threshold. The duration of this suppression depends on the signal duration and the intensity, in such a way that it is optimally suited for the chirp rate and duration of the conspecific song. The neuron has no after-discharge, even in response to short (10-15 ms) signals and can encode pulse rates up to 100 Hz. It is characterized by slow adaptation and lack of habituation. Due to these properties and its sharp tuning to 4-5 kHz, this neuron can precisely encode the conspecific calling song over the whole intensity range (as the large segmental neuron, see Fig. 26b). However, it is not so effective in responding to conspecific courtship song and to the calling songs of other, sympatric crickets (Fig. 34c). This neuron seems to be identical to the pulse-coder neuron found in the neckconnectives of Gryllus campestris (Stout and Huber, 1972). The geometry of the branches of this neuron is not yet completely revealed, but the preparations we have at present show that it has one main dendritic branch which is restricted to the anterior part of the prothoracic auditory neuropile, ipsilateral to the axon and contralateral to the soma (Popov et al., 1976~).This branch probably has direct monosynaptic connections with lowfrequency (4-5 kHz) receptors. The axon ascends to the frontal protocerebrum, where it ends by numerous terminal branches, in the neuropile lateral to the calyces. On the way through the deuto- and tritocerebrum, it seems to have small side branches, not yet completely mapped (Rheinlaender et al., 1976). (2) The second type of ascending neurons is specialized for the detecting of high-frequency pulsed sounds, such as cries of bats or the “ticks” of the conspecific courtship song. This is the so-called high-frequency (HF) neuron found again in Gyllus bimaculatus (Popov et al., 1975; Rheinlaender et al., 19763. It has a broad tuning curve from 3-4 kHz to at least 40 kHz, with best frequency above 10 kHz (Fig. 34a). In contrast to the LF-unit, the HF-unit has no spontaneous activity, has pronounced after-discharges to short (less than 100 ms) signals, and fast adaptation to long sounds (more than 100 ms). It shows pronounced habituation to repeated stimulation and can only
NEUROETHOLOGY
OF ACOUSTIC COMMUNICATION
31 1
Fig. 34. (a) Threshold curves of the LF, and the HF, neuron in the cricket Gryllus bimaculatus (G. b.). (b) Power spectra of the conspecific calling song (G. b.) and of the calling songs of 5 sympatric crickets (0. t., Oecanthus turanicus; T. t., Tartarogryllus tartarus; S . u., Unknown species; G. k., Gryllodinus kerkennensis; T. b., Tartarogryllus bucharicus). The arrows in (a) mark the 1st harmonics of the corresponding calling songs related to the threshold curves of the 2 neurons (b and a). (c, d) Response patterns of the LF, neuron (c) and the HF, neuron (d) to stimulation with the calling songs specified above each column at different sound intensities. From Rheinlaender et al. (1976).
312
NORBERT ELSNER AND ANDREJ V POPOV
effectively reproduce rhythms below 10 Hz. This neuron responds poorly to the conspecific calling song (Fig. 34d), but clearly encodes the “ticks” of the courtship song. It probably plays an important role in courtship and evasive behaviour. Such a striking combination of functions is not so surprising and contradictory as it may seem at first sight, because in both cases (courtship of the male and cries of hunting bats), the female is brought into an arousal state. During courtship, high-frequency “ticks” activating the HF-unit and arousing the female, are alternated with short low-frequency trills, activating the LF-unit and attracting the female. So, the arousal produced by the “ticks” attains a certain level which is lower than the arousal necessary for evasive behaviour. During evasive behaviour only the HF-unit (and not the LF-unit) is activated. The topography of the branching of this neuron is in good agreement with this function. It receives its main inputs in the lateral, posterior and anterior parts of the axonal ipsilateral (and probably the posterior part of the contralateral) auditory neuropile of the prothoracic ganglion (Fig. 35c). Excitatory inputs are received from the broad-band receptors and inhibitory inputs (probably via an interneuron) from the low-frequency receptors of the tympana1 organ (Popov et al., 1976b). In addition, the HF-unit has dendritic branches in nonacoustic, posterior medio-ventral neuropiles of both sides. The large axon ascends to the brain, where it has projections in several regions, both on the ventral and dorsal sides. Terminal branches are limited entirely to the ipsilateral half of the brain and primarily to the region of the protocerebrum (Fig. 35a,b; Rheinlaender et al., 1976). There, numerous branches lie lateral to the mushroom body and are distinctively concentrated in the fronto-dorsal areas of this part of the brain. No connections with the mushroom body itself were found. The functional significance of this neuropile and its connections with other parts of the brain are unknown. In the deutocerebrum, a few sidebranches run toward the mid-line, while others extend laterally into the boundary region between deuto- and protocerebrum (branch SJ, which is known to be the main projection area of the ascending auditory fibres in acridids and moths (Rehbein, 1975; Roeder, 1969). It seems that the HF-unit also has a very thin, descending axon, although it was not followed morphologically in the pro-mesothoracic connectives. But a fibre with very similar properties has been found in the mesothorax of crickets (Elepfandt and Popov, unpubl.). Extracellular cobalt staining of this fibre showed that it has rich connections to all three medio-ventral associative neuropiles, and 2 branches to the medio-dorsal neuropile where it can probably have direct connections to mononeurons. The descending axon proceeds further, to the metathorax, but has not been stained there yet (Fig. 35d). Being polymodal by input, and poly-addressed by output, this neuron is well suited to a warning or arousal function.
N E U R 0ETH 0 LOGY 0F ACO U ST IC CO M MUN ICATI0N
313
Fig. 35. The anatomy of the HF-unit in the cricket Gryllus birnacularus. (a, b) Projections in the supraesophageal ganglion (a, frontal; b, lateral view). S,_,, side branches; CP, corpora pedunculata. Arrows indicate orientation (D, dorsal; F, frontal). (c) The anatomy of the HF-unit in the prothoracic ganglion. (d) Projections of the HF-unit in the mesothoracic ganglion AK, PK, anterior and posterior connectives; Th,,,, pro- and mesothoracic ganglion. AsA, DsA, ascending and descending axons. (a, b) from Rheinlaender et al. (1976); (c) from Popov et al. (1976b); (d) from Elepfandt and Popov (unpublished).
314
NORBERT ELSNER AND ANDREJ V POPOV
(3) The third type of ascending auditory neurons is incapable of describing each pulse of the conspecific calling song because of its pronounced afterdischarge to short sounds, but it can clearly reproduce the chirp rate and duration and hence was named “chirp-coder” by Stout and Huber (1972). Popov et a f .(1974) found this neuron at the level of the prothoracic ganglion of Gryflus bimaculatus. It is similar to the LF-unit described above in having a high sensitivity in the 5 kHz-region and high spontaneous activity, which is suppressed during sound presentation. However, it differs from the LF-unit by having a moderate sensitivity to high frequencies and in its pronounced adaptation and after-discharge to short signals. (4) The fourth type of ascending neurons has a phasic discharge to sound stimulation. When the calling song is presented, it responds with 1-2 spikes/pulse of the chirp in the whole intensity range. Consequently, it transmits information about the pulse rate independently of sound intensity and frequency (Popov and Markovich, unpubl.). Similar methods of coding of the amplitude-modulation patterns have been found in other insects too (Acrididae: Kalmring, 1975b; Lepidoptera: Roeder, 1966; Cicadidae: Popov, unpubl.). In no case have neurons been observed in the ventral nerve cord which respond specifically to conspecific signals only, or to one definite parameter of these signals. Each neuron has a preference for certain combinations of sound parameters, and these preferential combinations are different in different auditory neurons. Hence, each conspecific signal and the sounds of predators, activate different sets of ascending channels, the information requiring further processing by the brain centres. 3.3.3
Brain auditory neurons
Three neuropiles within the brain are known to be engaged in processing information supplied by ventral cord auditory neurons in insects. (i) The later-ventral neuropile on the border between deuto- and protocerebrum; the so-called “Adam’s region” first found in acridids (Adam and Schwartzkopff, 1967) and then in moths (Roeder, 1969) and in tettigonioids (Rheinlaender and Kalmring, 1973). (ii) The optic lobe, where large multimodal neurons have been found in the locust (Homdge el al., 1964). (iii) The frontal protocerebral neuropile situated under the cups of the calyces and lateral to them, on the midline of each half of the brain. This auditory region was first described in crickets (Rheinlaender et al., 1976), but seems to be present in locusts as well, since Horridge et al. (1974) noted sound-induced activity lateral to the mushroom bodies. Our knowledge of the functional properties of brain neurons is still poor. It was concluded that the brain auditory neurons are more specialized in their filtering properties than the ventral cord neurons since units responding only to
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
315
certain combinations of parameters (pulse rate, intensity, frequency, etc.) were found (Adam and Schwartzkopff, 1967; Adam, 1969). However, many of the functions Adam (1969) ascribed to the brain were later shown to be present already at the thoracic 1evel.bnly few brain neurons were found (in “Adam’s region” of tettigonioids) which were more selective than the ventral cord neurons. For example, neurons insensitive to direction and, thus, ideally suited for the transmission of the actual sound intensity irrespective of direction, and neurons with an intensity-independent coding of direction were found in the brain, thus demonstrating an independent representation of these 2 parameters in the brain (Rheinlaender, 1975). However, from the point of view of information processing, the brain neurons found to date are not specific enough to account for phonotactic specificity. On the contrary, at least some of them have properties that indicate that they do not serve as “specific song detectors”, but participate in complex integrative reactions. For example, the optic lobe neurons of the locust appear to be multimodal: they are directioninsensitive and also respond to vibration, various touch, and visual stimuli, these inputs being mostly independent of each other (Horridge ef al., 1964). Unaccountable bursts of activity, not precedkd by any noticeable stimulus but sometimes associated with movement, were observed in these neurons. Clearly they respond to any stimulus which would “startle” the locust, and being multimodal they cannot be of much use for specific responses. Roeder (1969a,b, 1973) described the instability of the responses of brainneurons in moths reflecting some peculiarities of the evasive behaviour demonstrated by these animals. These are: abrupt and marked lapses in acoustic excitability lasting from a few seconds to an hour; mutual inhibitory interaction between auditory and visual inputs; differential suppression by high sound intensities. He suggested that the acoustic units in the brain of moths may be concerned in arousal, reinforcement and sustainment of the local reflex mechanisms operating in the thoracic ganglia, so that direction measurements and corresponding differential commands to the muscles are formed at the thoracic level (Roeder, 1970, 1973). One cannot exclude the fact that similar relationships are true for other insects, at least in relation to direction and loudness measurements. Wuchenich (1974) and Wohlers (1974) even suggest that a specific recognition process for the calling song pattern can occur at the thoracic level. They propose a model in which the dominant role is as‘cribed to a pulse-rate oscillator monitoring the excitability of some segmental auditory neurons, which, in turn, can reset the oscillator activity. This combination of oscillator, modulated and not modulated oscillating pulse-coder neurons serve as a “time-filter” tuned to the pulse-rate of the conspecific calling song. But the gap between this model and the experimental data available to date is too wide to consider this suggestion as being near to the true situation. A method for measuring the chirp rate was suggested by Stout and Huber
316
NORBERT ELSNER AND ANDREJ V POPOV
(1972), who found that some ascending auditory units of intact crickets are periodically suppressed during certain phases of the respiratory cycle. These cyclically responsive auditory neurons have the highest output when the chirp rate is close to the respiratory rate. Bearing in mind that one and the same “slow oscillating neuronal system” may control both the respiration and the chirp rate of a singing male (Kutsch, 1969) it is reasonable to suppose ( i ) that a similar oscillating mechanism is present in the female and participates in chirp rate recognition and (ii) that the observed interaction between the auditory units and the respiratory cycle can serve as a time-filter for the chirp rates. It is interesting that such oscillations in the responsiveness of ascending auditory neurons were never observed, neither in the prothoracic ganglion nor the brain, in acute electrophysiological experiments on Gryllus bimaculafus (Popov, 1973; Popov ef al.. 1974, 1975, 1976b,c). This indicates that these systems (auditory neurons and slow oscillators) can probably be easily decoupled. Somewhere in the brain, however, there should be neurons that can “read” the messages of these oscillating units. 4
4.1
Development of acoustic communication POSTEMBRYONIC
DEVELOPMENT
OF
SOUND
PRODUCTION
“ STRIDULATION”) (LARVAL
In all orthopteran insects, sound production is not possible before the final moult to adulthood; the elytra which have to be rubbed against each other (in crickets and tettigonioids) or against which hindlegs have to be stroked (acridids) are not fully developed in the larvae. Even after adult emergence, it normally takes 4-7 days before stridulation starts, a delay which is probably controlled by hormonal factors. Nevertheless, it has been shown in several species, that most of the neuronal circuitry necessary for generating the specific song rhythms is already complete during the last instar. This has been demonstrated first in last instar nymphs of the cricket Teleogryllus cornmodus by making heat lesions in the mushroom bodies (an operation known to evoke stridulation in adult animals, Huber, 1952, 1955). In several cases, fully developed calling, agressive or courtship motor patterns have been released in these larvae (Bentley and Hoy, 1970). The movements of the small wing pads were silent, of course, since these did not touch each other. But it could be shown by recording the neuromuscular activity, that the larval motor pattern almost perfectly matches that produced by adult males (Fig. 36a). In the nymphs, normally, the song patterns are obviously suppressed by the brain, since “stridulation” could not be elicited unless lesion were made. When last instar nymphs were placed in natural situations which would normally evoke stridulation in adults (for
N EUROETHOLOGY OF ACOUSTIC COMM U NICATION
317
example, confronting them with conspecilk males or females), no song-like wing bud movements were ever observed in Teleogryllus commodus. A similar situation has been reported for Gryllus campestris. But here, in forty cases of aggressive interaction, at least two attempts to stridulate were seen. Recordings of the neuromuscular activity in these cases clearly demonstrated the similarity between the nymphal motor pattern and the adult aggressive song (Weber, 19 74). The strong inhibition exerted by the brain upon the, otherwise, fully developed song generators of the last instar nymphs, impedes detailed studies on the maturation of the neuronal circuits, since it is very difficult to make lesions in the supraesophageal ganglion of younger instars. The larvae of acridid grasshoppers are more suitable for such investigations: it has long been known that the third and fourth (last) instar nymphs of several Chorthippus species perform stridulatory movements with their hindlegs which, of course, are silent because of the missing elytra (Weih, 1951). These movements are normally performed in response to the song of conspecific adult males (and can, therefore, easily be evoked), but have also been seen to occur spontaneously. Using the new opto-electronic device mentioned in a previous chapter (see p. 236) the singing movements of Chorthippus instars have been recorded (Halfmann and Elsner, in prep.). In all three species (Ch. brunneus, Ch. biguttulus, Ch. mollis), the stridulatory movements of the fourth instars already bear a considerable resemblance to the adult pattern. For example, the specific features characterizing the different patterns produced by each hindleg are already present in the nymphs (Fig. 36b,c,d). Further investigations are now being performed on the neuromuscular level to follow in detail the maturation of grasshopper stridulation through all larval stages. About 2-3 months after the final moult, i.e. a few weeks before death, individuals of Ch. mollis have been observed producing a stridulatory pattern which was simplified in several respects: both hindlegs simultaneously performed pattern I, i.e. on both sides the “chirps” were initiated by steep downstrokes; furthermore, songs were recorded which were characterized, in addition, by highly synchronous movements of both hindlegs (Fig. 36e, lower oscillogram). Simultaneous production of pattern 11, on the other hand, has never been observed. Remarkably, these animals could also stridulate with two patterns in a normal phase-shifted manner if the excitation rises above a certain level (Elsner and Helversen, 1976). Similar changes occur in the motor output of normal animals after amputation of both hind-femora. After 2-4 days, pattern I becomes more and more dominant on both sides (see p. 257). It is assumed, therefore, that during post-larval life, sensory connections to the central nervous system which have importance upon the pattern generating networks (i.e. guaranteeing the normal bistable output, see p. 24 1) may degenerate.
1 L.
3r Teleogryllus cornmodus
4
I
1
(b)
1
I
0.2s
Chorthippus brunneus
-
CC) n
L
0.2 s
Chorthippus biguttulus
I 2
11
Chorthippus mollis
0.2s
0.2s
. .
Chorthippus mollis Fig. 36. Adult and larval stridulation in crickets and grasshoppers. (a) Comparison of motor output of a last instar nymph (1, 3) with the actual song of an adult cricket (Teleogryllus cornmodus). I, activity of the 2nd basalar muscle (wing opener); 3, push-pull recording from the subalar muscle (initially downward spikes, wing opener) and the posterior tergocoxal muscle (upward spikes, wing closer). (b, c, d) Comparison of stridulatory movements of adult (I) and
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
4.2
319
POSTEMBRYONIC DEVELOPMENT OF HEARING A N D INNATE RELEASING MECHANISMS
The development of phonoresponsive behaviour has been studied so far only in adults. In all three orthopteran families the females newly moulted to adulthood, when hearing a conspecific male calling, either show no reaction at all, or display defensive behaviour and escape reactions. Usually, positive phonotaxis or agreement song cannot be observed before the fifth to sixth day (Acrididue: Loher and Huber, 1964, 1966; Loher, 1966; Gryllidae: Shuvalov and Popov, 1971; Tettigonioideu:Morris et al., 1975). For example, in young crickets (several days after final ecdysis) all sounds perceived by the tympanal organs elicit escape reactions (negative phonotaxis) when the animals are out of their holes. Remarkably, this reaction is less consistent to low-frequency than to high-frequency sounds. With sexual maturation, escape reactions to low-frequency sounds is replaced by positive phonotaxis, whereas negative phonotaxis to high-frequency sounds remains unchanged. A possible explanation is that the innate releasing mechanism of positive phonotaxis, if already present, is inhibited until “maturation signals” arrive at the brain (Shuvalov and Popov, 1971). Loher (1966) has shown that in grasshoppers at least, hormonal factors (corpora d a t a hormone) are involved in this process. Larval “stridulation” which occurs in several grasshoppers in response to the song of adults (see above) indicates that in male nymphs at least, innate releasing mechanisms are already present (female larvae have not yet been studied). How well-tuned to the conspecific- song these larval-releasing mechanisms are, compared to those of adult animals, is not yet known. Certainly, the maturation of an innate releasing mechanism is not only a central nervous, but also a sensory problem and, therefore, the postembryonic development of both sound receptor organs and auditory neurons of higher order will have to be investigated in more detail. Ball and Young (1974) have studied the postembryokc development of the tympanal organ in the cricket Teleogryllus commodus. The hearing organ remains undifferentiated until the third larval stage, when the first scolopales appear. All groups of scolopidia are present in the seventh instar but the number of scolopidia is increased until adulthood.
~
last instar nymphs (2) of the grasshoppers Chorthippus brunneus, Ch. biguttulus and Ch. mollis. (e) Stridulation of adult male Ch. mollis: normal phase-shifted production of patterns I and I1 (3) and synchronous display of pattern I by both hindlegs (4) observed in old individuals. In all oscillograms (b-e) the stridulatory movements of the left and right hindleg are monitored on the 1st and 2nd traces, respectively. The sound produced by the right hindleg (the left elytron has been removed) is displayed on the bottom traces. (a) from Bentley (1970); (b-d) from Halfmann (1977); (e) from Elsner and Helversen (unpubl.).
320
N O R B E R T E L S N E R A N D A N D R E J V POPOV
The tympana do not appear before the eighth (posterior tympanum) and the tenth (anterior tympanum) instar. In the larval stages, they are hairless areas of cuticle; the translucent appearance is not found before final ecdysis. It is not known at what stage the connections are made between the tympanal organs and the central nervous system, and, consequently, nothing can be said about the extent of precocious hearing in Teleogryllus. In the acridid Locusta migratoria, the low-frequency “a”-cells of the tympanal organs appear very early (prolarval stage) but high-frequency units appear only at the third stage (Beckert, 1962). The low-frequency auditory channel is already functional, from the receptor level up to the brain in the last instars, whereas high-frequency representation appears in the protocerebrum only in adults (Adam and Schwartzkopff, 1967).
5
5.1
Genetics of acoustic communication GENERAL IDEAS UNDERLYING A GENETIC APPROACH
Acoustic behaviour of orthopteran insects is under strict genetic control: crickets and grasshoppers raised under very different environmental conditions (varying, especially, their acoustical experience) always produce the distinct calling and courtship songs without any modification. The females always respond only to the conspecific songs, even if they were exposed exclusively, to heterospecific sound patterns during their postembryonic development. Obviously, the phenotype, i.e. the sound patterns and the innate releasing mechanisms, very precisely reflects the genotype. Such correlations have led to a genetic approach to the neuronal basis of acoustic communication. Pioneer work has been done by D. Bentley and R. Hoy (on crickets) and by D. and 0. von Helversen (on acridid grasshoppers) with the following views in mind. Beyond the task of revealing the fundamental mechanisms of the genetic control of behaviour, genetic techniques are used to test hypotheses of the structure of the neuronal pattern-generating networks. By hybridization of closely related species, the inherent, centrally programmed motor patterns underlying stridulation may be altered. By such genetic manipulation, the motor patterns can be split into sub-units which may be recombined in a new manner. By investigating both the stridulatory patterns and the innate releasing mechanisms, it can be determined whether the transmitter and the receiver are genetically coupled or may be separated. Hybridization experiments were first performed by Fulton (1933), HormanHeck (1957), Bigelow (1960) and Leroy (1966) using crickets, and by Perdeck (1 958) working with acridid grasshoppers of the genus Chorthippus. These investigations have been applied to the study of neuroethological aspects, in
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
32 1
crickets by Bentley (1971); Bentley and Hoy (1972); Hoy and Paul (1973); Hoy (1974); Hoy et al. (1977) and in acridid grasshoppers, by D. and 0. von Helversen (1975a,b,c) and Helversen and Elsner (1975). 5.2
5.2.1
SONG PATTERNS OF I N T E R S P E C I F I C H Y B R I D S
Gryllidae
In crickets, the most comprehensive genetic analysis has been performed on the Australian species Teleogryllus commodus and T. oceanicus. The calling songs of these crickets are highly complex. Two kinds of pulses are produced by the closing movements of the elytra, the so-called A- and B-pulses, which differ both in duration and intensity. These are grouped into chirps and trills, respectively. Chirps and trills form a phrase, the largest repeating unit of the song. In T. commodus, each phrase is composed of one chirp followed by one or two long trills. The T. oceanicus phrase consists of a chirp followed by about ten short trills, each containing 2 pulses (Fig. 37a). Hybrids from the two reciprocal crosses were produced: ( i ) T. oceanicus 9 x T. commodus 8;(ii) T. commodus Q x T. oceanicus 8.Numerous song characters were measured (including the numbers of pulses/chirp, /trill, and /phrase, the number of trills/phrase, the phrase repetition rate, and the intra- as well as the inter-trill or inter-chirp intervals). Most of these parameters proved to be intermediate between the parental values, indicating polygenic control, an assumption which was confirmed by backcrosses (Bentley, 1971; Bentley and Hoy, 1972). The two reciprocal F1-hybrids were similar in most song pattern characteristics. They differ, however, from each other in a few parameters, such as in the intervals between the trills and in the phrase repetition rate. Genes controlling these characters probably reside on the X-chromosome, since crickets and grasshoppers have XO sex-determination (lacking the Ychromosome, i.e. the males always get Their single X-chromosome from their mother). Of course, one might also consider other maternal effects (cytoplasmic factors). Furthermore, a character such as the inter-trill interval may not be exclusively controlled by genes located on the X-chromosome or by hypothetic maternal factors. If it were so, this parameter in the hybrid should be indistinguishable from that of the mother, which is not the case. Therefore, a multichromosomal control has to be postulated. Summarizing the results, one sees that the neuronal circuits generating the motor patterns for stridulation are determined by complex genetic mechanisms. None of the various characters is determined by monofactorial inheritance on a simple dominant-recessive basis. Rather, the central network is constructed at the direction of a polygenic and multichromosomal system.
322
NORBERT ELSNER AND ANDREJ V. POPOV
TRILL
TRILL
!
..- , I--'
IHlLL
(WILD W PEI
F, AN0 T. OCEANICUS
,- - - - - ---.FEMALE
AND T COMMODUS MALE
MALE AND T COMMODUS FEMALE
Ft AND T. COMMODUS
TELEOORYLLUS OCEANICUS
LtJJ
0.5s
Fig. 37. Genetics of cricket stridulation. (a) Oscillograms of the calling songs of pure species, hybrid and back-cross crickets. (b) Change from the normal Teleogryllus oceanicus two-pulse trill to the three-pulse trill which is characteristic of the back-cross hybrid between T. oceanicus x T. cornmodus and T. oceanicus (see 2nd oscillogramm in a). In each pair of records the top trace shows the stimulus applied to a ventral cord command interneuron and the bottom trace shows the activity of song motoneurons. In the bottom pair of traces the firing rate of the command interneuron has been increased about 10 per cent causing the shiR from wild-type to back-cross trill. From Bentley and Hoy (1974).
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
323
At first sight such a complex situation is, certainly, discouraging for the task of tracing the chain from gene to behaviour on a single unit level. However, it has been demonstrated that the effects of genetic manipulation can be tested with very high resolution capability, possibly on the level of single, identified neurons. By hybridization, animals could be obtained which differed from each other in one character only. For example, the calling song of T. oceanicus is characterized by trills each consisting of two pulses; the backcross between the F,-hybrid and the mother, T. oceanicus Q x (T. oceanicus Q T. commodus 8)8,has trills with three pulses (Fig. 37a). Motor units are activated only once per pulse, i.e. on the motoneuronal level the difference is just one additional action potential. In other words: the genotype specifies the motor output pattern with the highest possible accuracy, i.e. on the basis of single impulse increments (Bentley, 1971). It has been possible to obtain an initial insight into the way in which this specific control might be exerted on the motoneurons. Bentley and Hoy (1974) have stimulated the appropriate command interneuron descending from the brain to the thoracic ganglia in wild-type T. oceanicus. Stimulation at about lOO/s evoked the normal calling song, with trills each containing two pulses. By increasing the rate of excitation by about 10 per cent only, the motor pattern could be changed from the wildtype two-pulse trill to the three-pulse trill of the backcross (Fig. 37b). The genetic manipulation of motor patterns can serve as a useful tool for revealing the basic structure of the neuronal circuits underlying behaviour. As shown earlier, the calling songs of many crickets are subdivided into chirps separated from each other by silent intervals. A fundamental problem, still outstanding, is whether this song structure (i) is based on one single oscillator which underlies phenomena such as accumulating refractoriness leading to regular pauses, or (ii) based on two separate, superimposed oscillators, one generating the pulse rhythm and the other producing the chirp rhythm. Bentley and Hoy (1972) have tested these alternative hypotheses by crossing two chirping crickets (Gryllus campestris and G. armatus) with a non-chirping species (Gryllus rubens). Both G. campestris and G. armatus produce chirps consisting of three pulses each, but differ considerably in rate which is about 3.3 Hz in G. campestris and 14 Hz in G. armatus. G. rubens transmits uninterrupted trains (lasting 3-5 s) of pulses which are not arranged into short chirps. Unfortunately, the hybrids G. campestris 9 x G . rubens 8 had a very low viability, and only a single adult male was obtained. The calling song of this individual consisted of chirps which were intermediate in length between the pulse trains of G. rubens and the chirps of G. campestris. Remarkably, in the hybrid, the chirp rate was not significantly different from G. campestris, despite the much larger number of pulses/chirp and correspondingly shorter silent period. This result is difficult t o understand on the basis of the one-oscillator
324
NORBERT ELSNER AND ANDREJ V POPOV
hypotheses which explains the chirp intervals as the effect of accumulating refractoriness in the pulse-generating oscillator. It seems more likely that the chirp rhythm is generated by a second, inherent oscillator which is separate from the pulse-producing mechanism. This view is strongly supported by other observations which have been reported earlier (see p. 266). Similar experiments on G. armatus and G. rubens suggest another mechanism. In G. armatus a single, pulse-rhythm, oscillator is assumed which ceases periodically due to accumulating refractoriness. This accumulation is thought to occur over three pulses, whereas in G. rubens, it follows on a pulseby-pulse basis. Such a model would also explain the occasional occurrences of chirps of irregular length in G. armatus Q x G . rubens 8 hybrids, which is much more difficult to understand with a two-oscillator hypothesis. 5.2.2 Acrididae D. and 0. von Helversen (1975a,b,c) have hybridized the three closely related sympatric grasshopper species Chorthippus biguttulus, Ch. mollis, and Ch. brunneus. Their extensive study on Ch. biguttulus x Ch. mollis hybrids will be reported below; the unpublished work on Ch. biguttulus x Ch. brunneus is briefly mentioned in the chapter on evolution (see p. 336). As described in detail earlier, the calling and courtship songs of Ch. biguttulus and Ch. mollis are rather complex and differ from each other in nearly all characters (see p. 241, Figs 36, 38). It is very difficult to homologize the various sub-units. Only the sound emissions produced by one up- or downstroke (“syllables”, “pulses”) can be called homologous without doubts; their duration is intermediate in the songs of the hybrids. Other pattern elements such as the “chirps” are probably nonhomologous. In the songs of the hybrids they appear more or less independently superimposed on each other (Fig. 38). Consequently, the complexity of the hybrid song structure is almost doubled compared with that of either parent. The two reciprocal crosses Ch. biguttulus 9 x Ch. mollis 8 and Ch. mollis Q x Ch. biguttulus 8 differ considerably from each other. As a general rule, the hybrid songs usually resemble more the song pattern of the maternal species. In the males, sex-linked genes residing on the X-chromosome might be responsible for this effect. However, the same maternal influence is observed in the song pattern of female hybrids, which have identical genomes in the two reciprocal crosses. Thus, generally unknown, extra-chromosomal, “maternal factors’’ must account for this phenomen. Unlike the rather consistent songs of Teleogryllus hybrids, the Chorthippus hybrids were characterized by an enormous inter- and intra-individual variability of the sound patterns such as is never observed in the parental species. The intra-individual variability, seems to depend partly on the
325
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
motivation of the hybrid. Calling and rivalry song were often heard more biguttulus-like while the courtship song was similar to the mollis song. Even within the performance of a single song sequence, the parental pattern elements were often represented in a changing manner: frequently, songs were observed
Chorthippus mollis
Chorthippus biguttulus
(C)
(d) ,-.-II
.
.. .
I I 11.1
-.
,.1-..1..
.-.-I-. I
,
..
,.-I-...-..,.-..-..
..... .
7..
.I.
‘ 1
(el Ch biguttulus X Ch mollis
04s
Fig. 38. Stridulatory movements (top, left hindleg; middle, right hindleg) and song patterns (bottom traces, right hindleg) of Chorthippus mollis, Ch. biguttulus and hybrids between these two species. In all cases both hindlegs were intact, but the left elytron had been removed in order to obtain a clear sound recording from the right side only. (a, b) Stridulatory patterns of Ch. mollis (a) and Ch. biguttulus (b), (the occasional performance of a “chirp” produced by only 2 cycles of upward and downward movements is marked by an arrow). I, 11, Stridulatory patterns I and 11; in all recordings the initial downstroke which characterizes pattern I of Ch. mollis is marked by white arrows. (c) The left hindleg of the hybrid produces a combination of Ch. mollis pattern I (weakly expressed) and Ch. biguttulus pattern I. Arrows indicate the characteristic elements of Ch. biguttulus patterns I and I1 (pauses and extended upstrokes). (d) The left hindleg of the hybrid stridulates in a way very similar to Ch. mollis pattern I (white arrows), whereas the other leg subdivides the song units by producing “chirps” resembling Ch. biguttulus pattern I1 (black arrows). (e) The right hindleg produces a Ch. mollis “pulse”-pattern whereas the left leg inserts a pause clearly resembling Ch. biguttulus pattern I (black arrow). (f) Pauses (which characterize Ch. biguttulus pattern I) inserted after a downward movement (black arrows) and not (as usual) before (for comparison see (b), middle trace). Vertical calibration, 3 mm. From Helversen and Elsner (1977).
resembling, at the start, a typical mollis structure, which was then superimposed by biguttulus “chirps” towards the end. Remarkably, the two parental patterns were sometimes displayed simultaneously for a short time, each on one side: one hindleg produced a continuous pulse sequence like Ch. mollis whereas the other leg performed a sequence subdivided into the biguttulus “chirps” (Fig. 38d,e; Helversen and Elsner, 1977).
326
NORBERT ELSNER AND ANDREJ V. POPOV
Helversen (1975a) concludes from these observations that a consistent, intermediate pattern-generating circuit has not been fully established in the hybrids. Rather, two parallel, and to some extent independent, neuronal networks are postulated, corresponding to the specific information of the two parental genomes. The patterns are thought to be more or less independently generated by each circuit and then converging in a final common pathway (the motoneurons or elements driving them) where they are superimposed on each other. Remarkably, a suitable “compromise” is not always easily found: quite often the song is performed in a “stuttering” manner, i.e. abortive pulses are produced, and pauses are irregularly inserted. 5.3
SONG-SPECIFIC I N N A T E RELEASING MECHANISMS OF INTERSPECIFIC H Y B R I D S
So far, the genetic approach to acoustic communication has been directed only at the signal-emitting system. Hoy and Paul (1973), Hoy el al. (1977) and D. and 0. von Helversen (1975b,c, 1977) have extended the investigations to the signal-receiving side. Their studies on the structure of hybrid innate releasing mechanisms, and the possible genetic coupling of the signal-emitting and the signal-receiving systems, showed that direrent mechanisms might be present in crickets and grasshoppers, respectively. 5.3.1
Gryllidae
Hoy and Paul (1973) tested the responses of female hybrids, from the cross Teleogryllus oceanicus Q x T . commodus d,which were exposed to the soslnd patterns of their parents and of their hybrid brothers. The crickets walked on a Styrofoam Y-maze-ball where they had to make repeated decisions to turn right or left (Fig. 39). Sound was played alternately from loud-speakers mounted symmetrically on the left- and the right-hand side of the cricket at angles of 40° to the midline. The different sound patterns were broadcast separately. The females had to make 40 choices, 20 when the sound pattern was played from the lefthand loudspeaker and 20 when it came from the righthand. The results, obtained under this experimental situation, showed that the hybrid females preferred the song patterns of their hybrid brothers and responded only rather weakly to the parental songs (Fig. 39). The evolutionary implications of this finding will be discussed below (see p. 330). 5.3.2 Acrididae A completely different situation has been found in acridids. The investigations provide evidence for a more or less independent existence of both parental
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
327
1 FEMALE HYBRID RESPONSES TO I MALE COMMODUS~OCEANICUS HYBRID
--
0
0
,
0,
I-
0
P a:
0
I
0
5
LEFT
t
o
5
t
LEFT
Fig. 39. The preference of female hybrid crickets (Teleogryllus oceanicus x T. commodus) for male songs of hybrids and pure species. Each female (walking on a Y-maze ball) was required to make 20 choices each while the sound was played through the left and right speaker, respectively. If the females always turned toward a particular song whether played on her left or her right, the song scored 1 on both axes. Note that the different song patterns were presented separafely.From Bentley and Hoy (1974).
innate releasing mechanisms in hybrid females (D. and 0. von Helversen, 1975b,c). Most F1-females of the two reciprocal crosses between Chorthippus mollis and Ch. biguttulus answered very weakly to the song patterns of their hybrid brothers. Instead, they responded much better to the songs of one or even both parental species. Tests performed with artificial sound stimuli revealed innate releasing mechanisms of the same structure as was known from non-hybrid females of Ch. mollis and Ch. biguttulus (Fig. 40). Responses were optimal to stimuli combining the essential features of both parental songs.
I I I I I
1
I
Cb)
0 0
1"
: Q 114 : Q 119
0 :
Q 265
A'Q
271
o 9 Ch
-60
mollis
LO
20
100
200
300
400
500
m m s
St imuIus Int ervaI Fig. 40. Comparison of the innate releasing mechanisms of pure species of female grasshoppers (a, Chorthippus biguttulus; b, Ch. mollis) and of individual female hybrids (Q 102, Ch. biguttulus Q x Ch. mollis 8; 9 114, 1 19, 268, 271, 273, 9 Ch. mollis x Ch. biguttulus 8). The reaction (agreement song) of these females was tested by presenting rectangularly modulated white noise. The duration of these sound stimuli was held constant (s = 4 0 ms (a); s = 300 ms (b)) while the intervals were varied. Each test point represents the reactions to 14-24 (a) and 2040 (b) presentations of a particular stimulus :interval combination. The averaged response curves of pure species females (heavy lines) were drawn from 87 (a) and 85 (b) presentations of each combination. Further explanation in the text (see also Fig. 18). From D. and 0. von Helversen (1975b).
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
329
Intra-individual variability of female responses was very small, but individuals differed from each other: some females responded equally to both parental songs, most to both patterns but with different thresholds for each type, and some individuals preferred only one type (Fig. 41a). The distinct patterns of responsiveness characterizing each female remained constant through life. Generally, the innate releasing mechanism of the maternal species was more frequently found in the hybrid than that of the paternal species. As in the case of hybrid song patterns, maternal extrachromosomal factors are thought to be responsible for this phenomenon, since the genotypes of female hybrids from reciprocal crosses are identical. The nonresponsiveness of female hybrids to the songs of their hybrid brothers, and their clear preference for the parental song patterns, hardly favour the idea that an intermediate filtering system is present in the CNS. Rather, the hypothesis is supported that both parental innate releasing mechanisms are formed in parallel during postembryonic development. As mentioned above, a corresponding hypothesis had also been suggested for the hybrid songgenerating system. There, support came from indirect sources, since the output of the two proposed networks could not be recorded before combining in a common final pathway (apart from the few cases where separation was obvious for a short time, when one hindleg performed a mollis-like pattern, whereas the other displayed a biguttulus-like structure). In the case of the innate releasing mechanisms, the investigators were in a more favoured position: the responses of the two hypothetical networks could be tested independently. The results (Fig. 40) give strong evidence for the actual existence of two separate neuronal circuits in individual female hybrids.
6
6.1
Evolution of acoustic communication SONG-PATTERNS AND INNATE RELEASING MECHANISMS: EVOLUTION OR GENETIC COUPLING?
co-
The song-generating system and the corresponding pattern-recognizing mechanisms are well-tuned intra-specifically. Relying on this observation, for which numerous examples have been gathered in the past, evolutionary biologists have postulated the appealing hypothesis of a genetic coupling of the song-patterning system and the corresponding innate releasing mechanism (Alexander, 1962, 1968). The attractiveness of this concept lies in the consequences it has for speciation, which would be enormously accelerated if any alternation of the transmitter system (caused by hybridization or mutation) was accompanied by a corresponding change in the receiver. An
330
NORBERT ELSNER AND ANDREJ V. POPOV
alternative hypothesis explains the tuning of the two systems by a coevolutionary process, during which the song-generating and the songrecognizing mechanisms have been adapted to each other. By hybridization of closely related species these two alternatives can be tested, although a final decision can only be made in the case of a behavioural separation of the two systems (see below). In hybrid crickets the male song-generating system and the female innate releasing mechanism remained behaviourally coupled: the F, females of the cross Teleogryllus oceanicus Q x T. commodus d preferred the songs of their hybrid brothers and rather weakly responded to the sound patterns of the parental species (Hoy and Paul, 1973). As described above, in these experiments the different song patterns were presented separately: the animals
TABLE 1 Hybrid type performing discrimination
% response to
Response to T- 1 song
Response to T-2 song
T-1: ( T . oceanicus Q x T. commodus 8) Series A Series B Total
12 85 97
1 49 50
66
T-2: (T.commodus Q x T. oceanicus d) Series A Series B Total
19 21 40
66 59 125
76
sibling
From Hoy etal. (1977).
walking on the Y-maze-ball did not have the choice of discriminating between two patterns presented simultaneously. Undoubtedly, a considerable amount of uncertainty remained in this experimental situation (see p. 280 for discussion on the phonotactic behaviour of female crickets with or without a choice). Well aware of this argument, Hoy et al. (1977) have recently made auditory discrimination tests by allowing freely walking females to choose between different songs. These experiments have confirmed earlier results: Song patterns and recognition mechanisms turned out to be behaviourally coupled even if the offspring of the two reciprocal crosses (T. commodus 9 x T. oceanicus d and T. oceanicus Q x T. commodus 8)were tested against each other. Female F, hybrids showed a phonotactic preference for the song patterns of their hybrid brothers over that of the males of the reciprocal cross (Table 1, see Fig. 37 for the hybrid song patterns). F S m these tests, the
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
33 1
authors have concluded that the "results imply that the production of song by males and its detection by females have a common genetic basis" (Hoy et al., 1977).
The question is open whether this evidence is strong enough. Thefailure to demonstrate a behavioural separation of the two mechanisms by hybridization does not appear to us as the decisive criterion for the verification of a suggested p mollis
OResponse of biqultu!us
x @biguttulus
1
100
(a)
"moiiis
81
',~I~~II~UII+ dadIU .an_~ , 0 9 Nr
11L
123
121
119
121
115
116
111
113
L2
L5
L7
272 268 271
26L 275
273
286 287 288
I
Ch. mollis X Ch. biguttulus 1.
(b) Chorthippus mollis
11.11
If1
'.
11'1'
1111
1
T 11111
Chorthimus biguttulus
yi)"
IIIVI
r l
r r i i i
IN''
IIIII
0.4 S
11..
I I
Fig. 41. Independence of sound recognizing and sound producing mechanisms in a female hybrid grasshopper. (a) Response (agreement song) of individual female hybrids (Chorthippus mollis 9 x Ch. biguttulus d) to male parental songs (black columns, response to Ch. mollis song; white columns, response to Ch. biguttulus song). Sisters having the same parents are embraced by parentheses. (b) Song of 9 112 compared with typical female songs of pure species Ch. mollis and Ch. biguttulus. This female (which responds in 94 per cent of all tests to the song of Ch. biguttulus) produces a song pattern very similar to female Ch. mollis and not to Ch. biguttulus. The oscillograms show Ch. mollis pattern I1 produced by the hybrid and Ch. mollis pattern I produced by the pure species. (a) from D. and 0. von Helversen (1975b); (b) courtesy of D. von Helversen.
genetic coupling. Bearing in mind that almost all cricket hybrid features are intermediate one would expect the same behavioural coupling, even if the two systems are genetically separated from each other. In crickets the question can only be approached inter-individually, since the females are silent. Working with acridid grasshoppers offers several advantages, since the females are sound producing too. Thus, the question of whether the two systems are coupled can be examined intra-individually. As reported above, both male and female Chorthippus biguttulus x Ch. mollis hybrids were
332
NORBERT ELSNEP AND ANDREJ V POPOV
characterized by a considerable inter-individual variability, resembling the parental phenotypes to different degrees. Strikingly, several hybrid females were found which exclusively responded to the song of one parental species, but produced a song which retained predominantly the song pattern of the other parent (Fig. 41). These examples clearly demonstrate that in the female grasshoppers the song pattern-generating and the recognizing system for the male song are not coupled. Rather, in Chorthippus, the two parts of the acoustic communication system must have undergone co-evolution during phylogeny to adapt to each other (D. and 0. von Helversen, 1975b).
6.2
EVOLUTION OF SONG PATTERNS
The stridulatory patterns of many Orthopterans have a highly complex structure, containing several levels of rhythmicity (Figs 2, 3, 36, 38). The question arises whether these rhythms have been formed de nouo or whether central nervous oscillators developed earlier in evolution have been used as components for the construction of the song-generating neuronal networks. Pre-existing motor patterns, such as flight and walking are among the most likely cyclically recurrent behaviours that might have been incorporated. In Ensifera, especically in crickets, phylogenetic relationships between flight and stridulation have often been suggested, since it appears rather obvious that singing by using the elytra is derived from casual sound production during flight (Zeuner, 1934, 1939; cf. Tuxen, 1966). Strong evidence for this hypothesis comes from the work of Huber (1962) and Kutsch (1969) in which the two behaviours have been compared in some detail: it has been demonstrated that in Gryllus campestris and some closely related species (G. bimaculatus, Acheta domesticus) during flight, the wing beat frequency of 20-30 Hz is almost precisely the same as the frequency of elytral movements during stridulation (Fig. 42b). The central nervous and neuromuscular control system for flight is very similar to that for stridulation, which is not surprising since the same motoneurons, muscles, and appendages are involved in both behaviours. Only minor changes of the motor programmes are necessary to convert the up and down movements of the wings (typical for flight) into the stridulatory to-and-fro movements. These changes probably concern the activity patterns of some small auxiliary muscles only (pleuro-alar muscles) and not the main mesothoracic muscles such as the basalars, subalars, tergo-coxals, and tergo-sternals (Kutsch, 1969). In crickets such as Gryllus campestris, the main difference between flight and stridulation is the chirping structure of the calling songs, whereas flight consists of a continuous series of wing movements. Obviously, these species have evolved further; trilling species (for example Pteronemobius heydeni or
333
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
Gomphocerippus rufus
50 ms
(b) Flight A
1
1
Gryllus campestris
250ms
Fig. 42. Comparison of motor patterns underlying flight and stridulation in an acridid grasshopper (a) and in a cricket (b). 119, metathoracic 1st posterior tergocoxal muscle [wing elevator and elevator (remotor) of the coxal; 129, metathoracic subalar muscle [wing depresser and elevator (remotor) of the coxa]; 89, mesothoracic anterior tergocoxal muscle (wing elevator and wing closer); 90, mesothoracic posterior tergocoxal muscle (wing elevator and wing closer); L, R, left right. (a) from Elsner (1968); (b) courtesy of W. Kutsch.
Tartarogryllus bucharicus) are probably closer to the hypothetical archetype of cricket stridulation. In the course of evolution, further oscillators must have been added step-by-step, to modulate this primitive basic pattern of stridulation. At present only limited speculation can be made about the origin of these rhythms (see below). In acridids, stridulation by using the hindlegs is often considered to be very common, although it seems to be a more recent addition to the behavioural repertoire of this group. Only very few, and phylogenetically rather young, subfamilies (Truxalinae, Oedipodinae, Gomphocerinae) are sound-producing by a
334
NORBERT ELSNER AND ANDREJ V POPOV
femuro-elytral method (Uvarov, 1966). Ethologists have mainly regarded certain leg movements such as walking, jumping and various defensive movements, as a possible basis for the emergence of leg stridulation in these groups (Jacobs, 1953; Otte, 1970). This view is suggested by itself to some extent, and, certainly, might be true in many cases, but little neurophysiological evidence can be added at present. Surprisingly, such support can be given for a hypothesis stating phylogenetic relationships between sound production and flight: an idea which is obvious for ensiferans (see above), but at first sight appears far from leg stridulation. Evidence has been accumulated recently, that the fastest of those rhythms found in grasshopper song patterns has been adapted from flight. In several species (for example: Chorthippus mollis, Ch. biguttulus, Gomphocerippus rufus), the basic rate of stridulatory movements is 50 Hz and, thus, resembles accurately the flight frequency (Elsner, 1968, 1978). In many other species (for example: Gomphocerus sibiricus, Chorthippus brunneus, Ch. montanus, Stauroderus scalaris), stridulation is performed at a much slower rate, but individual movements are split into steps which follow each other at full, or exactly half or double flight frequency (Helversen and Elsner, unpubl.). A rather spectacular three-fold use of the flight pattern for sound production is exhibited by the alpine grasshopper Stenobothrus rubicundus (Elsner, 1974b). ( i ) Flight itself is sonorous since the sclerotized costal areas of the hindwings are vigorously beaten against each other. (ii) The wings also stridulate at the same rate during the second part of calling and courtship songs, while the animal is sitting on the ground. (iii) This part is preceded by leg stridulation which is performed at a much slower basic rhythm upon which flight frequency is partly superimposed, resulting in a splitting of the downward movements into sub-pulses (Fig. 43). In all these cases the recruitment of the flight pattern for leg stridulation is less difficult to understand than might appear at first sight, since the same motor units are involved. The neuromuscular mechanisms of leg and wing movements do overlap considerably, since several pterothoracic muscles and, consequently, the motoneurons driving them are bifunctional. Motor confusion is prevented by a switchable synergistic-antagonistic relationship: bifunctional muscles which insert on the same side of the wing joint are connected to direrent sides of the leg joint and vice versa. Consequently, Fig. 43. Leg and wing stridulation in the grasshopper Stenobothrus rubicundus. (a) Male grasshopper performing wing stridulation. (b) Simultaneous recording of the stridulatory movements and the sound pattern produced by the right hindleg during the 2nd part of leg stridulation. (c, d) Motor activity underlying the 2nd part of leg stridulation (c) and wing stridulation (d). 125, 1st pleurocoxal muscle (depressor and abductor of the coxa); 128, 2nd basalar muscle (depressor of the coxa and depressor of the wing): 129, subalar muscle (elevator of the coxa and depressor of the wing). From Elsner (1974b) modified.
336
NORBERT ELSNER AND ANDREJ V POPOV
whether the leg or the wing is moved, depends upon the coordination of these motor units (Wilson, 1962). For example, in the genus Chorthippus (especially in the group: Chorthippus biguttulus-mollis-brunneus)a basic form of leg stridulation could have evolved simply by changing the co-ordination of bifunctional muscles, acting during flight as synergists and antagonists, to antagonism and synergism, respectively. Such a hypothetic archetypal song pattern should have shown all the characteristic features of grasshopper flight, i.e. (i) consist of sequences not subdivided into “chirps” and (ii) with contra-lateral homologous units being activated synchronously. Among recent grasshopper species, Chorthippus mollis most nearly resembles this hypothetical stridulation pattern. The song sequences are built up from sub-units lasting 0.5 s which appear as images of short “jumping flights” (Flugsprunge) typical for these grasshoppers. These units (60-100 of them are linked together to form one song-sequence) begin with a jump-like downstroke followed by a vibratory movement at flight frequency of 50 Hz (Fig. 36e). In most other species the stridulatory patterns are more elaborate. In Chorthippus biguttulus the basic rhythm retaining flight frequency is interrupted after every third cycle of up- and downstrokes leading to a “chirped” structure; in Ch. brunneus flight frequency is superimposed upon a slower rhythm of 25 Hz, i.e. half the flight frequency. In an attempt to revert the elaborate stridulation patterns of these species to the hypothetical archetype two rather successful approaches have been made. ( i ) F, hybrids obtained by crossing Chorthippus biguttulus x Ch. brunneus produced a surprisingly primitive song pattern resembling Ch. mollis. The most conspicious characters of the parental songs were completely lost, i.e. the slow 25 Hz rhythm of Ch. brunneus and the insertion of pauses which characterizes the song of Ch. biguttulus. Instead, the hybrids displayed uninterrupted song sequences, the hindlegs vibrating at flight frequency of 50 Hz (Helversen and Elsner, in prep.). ( i i ) A similar uninterrupted song pattern has been observed in Ch. biguttulus after cutting sensory structures in the tips of the femora on both sides. As reported earlier (see p. 258), the pauses subdividing the “chirps” are inserted less and less frequently during the ensuing days and eventually disappeared almost completely (Lindberg and Elsner, 1977). If the Fig. 44. Interval histograms (a, c, e) and sequential interval plots (b, d, f ) of normal leg stridulation (a, b); flight (c. d), and “stridulatory” motor patterns recorded after amputation of both hind-femora (e, f, see also Fig. 10) in the grasshopper Chorthippus biguttulus. Each of the interval histograms (a, c, e) contains the data collected from 35-40 song sequences or flight bouts (i.e. recordings from muscles 125 and 129), only 4 (f) and 5 (b, d) of them are documented as sequential interval plots. The inserted oscillograms show pieces of 160 ms duration cut from recordings of the left (upper traces) and the right (lower traces) pleurocoxal muscle 125 (oscillograms 1, 3,4) and the subalar muscles (oscillogram 4). From Elsner (1978).
337
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
FLIGHT 120
"STRIDULATlON"
Serial no.of interval
338
NORBERT ELSNER AND ANDREJ V POPOV
two femora were totally amputated, motor patterns were occasionally observed, which were characterized, not only by the absence of any pauses, but also by a synchronous activation of contralaterally homologous units (Fig. 44). These patterns accurately resembled the hypothetical archetype of stridulation suggested above (Elmer, 1978). 6.3 EVOLUTION OF AUDITORY M E C H A N I S M S At present, the phylogeny of audition and innate releasing mechanisms can only be speculated upon. During evolution, auditory organs appeared, often independently, in different families and even within the same family. Due to the surprisingly similar fine structure of the sensilla in all tympanal-, vibrationsensitive, and proprioceptive organs, one is led to the conclusion that hearing organs were generally derived from proprioceptors monitoring body movements. As a first step towards the reception of air-borne sound, vibrationsensitivity might have been added. Even nowadays the two systems are not completely separated: one can still find receptor organs sensitive to both sound and vibration (Autrum, 1940; Nocke, 1972). Such a dual function exists at higher levels as well. In locusts, many big ventral neurons are sensitive to both vibration and air-borne sound. Obviously, at the time when a song-patternrecognizing mechanism had to be constructed, the animals could have made use of a pre-existing, information-processing system which had already evolved for the perception of body movements and vibrations. A comparative analysis of phonotaxis in crickets and phono-response (agreement song) in grasshoppers, shows that in the course of evolution, parallel to increasing song complexity, additional filters appeared at the highest auditory levels. At the lower centres (receptor level and ventral cord), the auditory neurons are very similar in their properties, i.e. the pre-processing of auditory information is the same in different species of the same family and even in closely related families such as Gryllidae and Tettigonioidea. The final filters should be flexible from the evolutionary point of view, because such closely related species as Gryllus campestris and G. bimaculatus have different innate releasing mechanisms-ne is very precise in measuring chirp duration, the other is not: but at the ventral nerve cord level information processing is nearly identical in both species. 7
Concluding remarks
The long-range goal of the neuroethological approach discussed in this review is to understand the functions and interactions of nerve cells, receptors and effector organs during the performance of acoustic behaviour. Progress towards this goal has occurred on a broad front during the past decade,
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
339
including developmental, genetic and evolutionary aspects. The connections between neurophysiology and classical ethology have been strengthened in many promising ways. Sound production. Electromyographic recordings in freely moving animals have provided a detailed description of stridulatory patterns in terms of neurophysiological events at the level of individual motor units. Because of the high recording selectivity and the possibility of monitoring neuromuscular activity on many channels simultaneously, the central motor output underlying orthopteran stridulation is now better known than that of any other comparably complex behaviour. On the basis of this knowledge any information obtained from direct central nervous recordings or from experimental interventions can be interpreted in a neuronally relevant manner. However, the high standard of electromyographic descriptions of motor patterns should not divert attention from the fact that no information about the circuitry of the neuronal networks generating these patterns is forthcoming. Intracellular recordings from the moto- and interneurons are the only way to attack this problem. Although the first intracellular recordings during the performance of any elaborate insect behaviour were carried out with crickets during song production (Bentley, 1969b), no further progress has occurred. In other invertebrates (crustaceans, leeches, molluscs), in the meantime, the central nervous networks generating simple behavioural activities have been analysed in great detail.’ In singing Orthopteruns, on the other hand, this work has ceased almost completely. Since David Bentley performed his recordings in 1966-67, not a single micropipette has been inserted into a cricket motoneuron during stridulation. Certainly, one reason for this stagnation is the difficulty of eliciting sound production in animals pinned down and dissected for micropipette recordings. But brain stimulation and lesion techniques, which can be successfully used for this purpose had already been developed by Franz Huber in the fifties (Huber, 1952, 1955, 1960). The present work of Otto (in prep.) on interganglionic “song fibres” suggests that the direct approach to the CNS is moving again. His findings of songspecific commands descending from the brain to the thorax show that the gross structure of the central nervous organization of cricket song is more complex than was suggested by Kutsch and Otto’s (1972) observation of stridulation by crickets which had all connections between brain and thoracic ganglia cut. An appealing hypothesis is that the thoracic pattern generators send an “efference copy” to the brain where integration with all kinds of sensory inputs occurs in order to “tune” the pattern to a specific behavioural situation. This consideration is in accord with many behavioural measurements which show that external stimuli conducted to the brain not only switch behaviours on and off but also modulate their pattern (see p. 252). Reviews: Usherwood and Newth, 1975; Fentress, 1976.
3 40
NORBERT ELSNER AND ANDREJ V POPOV
It has now become evident that the previous conception of rigid centrally programming is inadequate. Certainly, the CNS is able to produce the specific patterns of stridulation even after deafferentation, but often only for a limited time. At least in acridid grasshoppers the co-ordination of contralaterally homologous units, the change between different stridulatory patterns, and sometimes even chirp separation depend upon interactions between peripheral sense organs and the central nervous system over time (Lindberg and Elsner, 1977). Future research will certainly benefit from these findings. The elaborate motor patterns underlying stridulation can no longer be considered as monolithic structures immune to attack. It appears to us that it might not be any longer a conditio sine qua non to have the complex pattern with all its details displayed by a preparation set up for intracellular recordings. In many cases the central programme sensu strictu might turn out to be a rather simple pattern which is distinctively modifiable by sensory interactions. It might be possible to evoke this pattern and to set up appropriate recording devices. Sound reception and recognition. Unlike the investigations of neuronal mechanisms underlying stridulation which are performed by a few scattered scientists, the analysis of sound perception is carried out by several large research groups around the globe. As a result of their combined efforts our knowledge has increased considerably during the last ten years and now ranges from the biophysics of hearing and the signal processing within the CNS to the structure of the innate releasing mechanisms. So far, however, it is still an open question where we should look for the final detectors and where in the CNS the final decision about the conspecific or heterospecific nature of a received sound pattern is made. Auditory information processing on the level of thoracic and known brain neuropiles is performed by a population of individually tuned neurons. Each neuron is characterized by definite preferential areas in the intensity-frequency field, certain directional sensitivity and time resolution. The number and tuning of neurons is such that nearly any sound occurring in a given biotope evokes a unique mosaic of activated units. These mosaics can serve as a basis for discrimination between conspecific and heterospecific sound patterns by the subsequent brain structures. We do not know at present where and how this discrimination is realized. One cannot exclude the possibility that descending neurons have specific connections to certain ascending thoracic neurons as well as to intrinsic brain auditory units and hence would be activated only by certain input mosaics. If we assume that the threshold and circuitry of these connections can be regulated by other inputs (for example: hormones, different sensory and central nervous inputs, learning or mutations) it would be easier to imagine how and why the specificity of phonotaxis and phonoresponse (agreement
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
34 1
song) can be modified both in individuals through development and in species evolution. It seems probable that the sound localization and the measurement of loudness are performed by neuronal circuits which are independent of the recognition networks. The question where these functions are located remains open. The findings of Rheinlaender (1975) that invariant encoding of intensity and direction can be found only at the brain level might suggest that the circuits for sound localization and measurement of the intensity are located there, however, a (additional?) participation of thoracic networks has not yet been completely excluded. Thus, the question of how and where in the brain (or other parts of the CNS) specific recognition processes are realized, is still open. So too, is how these mechanisms then trigger the observed phonoresponsive behaviour. These neuronal circuits should occur somewhere between the known auditory projections of ascending auditory neurons and descending pathways. In searchiiig for these circuits we must not forget that clear phonotactic behaviour is realized only during certain external and internal conditions. In order to be more successful these need to be fulfilled, at least partly, during our electrophysiological experiments. Acknowledgements
We gratefully acknowledge the assistance of many colleagues who helped us to prepare the text and the figures and generously provided unpublished data. Our thanks are due to: J. Anders, D. R. Bentley, K. Halfmann, D. and 0. von Helversen, F. Huber, K. Kalmring, J. Innenmoser, A. N. Knjazev, S. Koppers, W. Kutsch, D. B. Lewis, A. M. Markovich, D. Otto, H. Rehbein, J. Rheinlaender, M. J. Samways, V. F. Shuvalov, I. D. Svetlogorskaja. Unpublished work from N. Elmer, K. Halfmann and S. Koppers was supported by the programme “Neurale Mechanismen des Verhaltens” of the Deutsche Forschungsgemeinschaft (El 35/& 10). References Adam, L.-J. (1 969). Neurophysiologie des Horens und Bioakustik einer Feldheuschrecke (Locusta migratoria). Z . vergl. Physiol. 63, 227-289. Adam, L.-J. (1972). Die Darstellung der akustischen Reizsituation im Gehirn der Wanderheuschrecke. Verh. Dtsch. Zool. Ges. 66, 176-184. Adam, L.-J. and Schwartzkopff, J. (1967). Getrennte nervose Reprasentation fur verschiedene Tonbereiche im Protocerebrum von Locusta migratoria. Z . vergl. Physiol. 54, 246-255. Adams, W. B. (1971). Intensity characteristics of the noctuid acoustic receptor. J . gen. Physiol. 58,562-579. Adams, W. B. (1972). Mechanical tuning of the acoustic receptor of Prodenia eridania (Cramer) (Noctuidae). J. exp. Biol. 57, 297-304.
342
NORBERT ELSNER AND ANDREJ V POPOV
Aidley, D. J. (1969). Sound production in a Brazilian cicada. J. exp. Biol. 51,325-338. Alexander, R. D. (1957). Sound production and associated behaviour in insects. Ohio J. Sci. 57, 101-113.
Alexander, R. D. (1960). Sound communication in Orthoptera and Cicadidae. In “Animal Sounds and Communication” (Eds W. E. Lanyon and W. N. Tavolga), pp. 38-92. Am. Inst. Biol. Sci., Washington. Alexander, R. D. (1962). Evolutionary change in cricket acoustical communication. Evolution, 16,443-467.
Alexander, R. D. (1966). The evolution of cricket chirps. Nat. Hist. 7 5 , 2 6 3 1. Alexander, R. D. (1968). Arthropods. In “Animal Communication” (Ed. T. Sebeok). Indiana Univ. Press, Bloomington. Autrum, H. (1940). a e r Lautauaerungen und Schallwahmehmung bei Arthropoden 11. Das Richtungshoren von Locusta und Versuch einer Hortheorie fur Tympanalorgane vom Locustidentyp. Z . vergl. Physiol. 28,326-352. Autrum, H. (1941). h e r Gehor und Erschutterungssinn bei Locustiden. Z . vergl. Physiol. 28, 580-637.
Autrum, H. (1960). Phasische und tonische Antworten vom Tympanalorgan von Tettigonia viridissima. Acoustica, 10, 339-348. Autrum, H. (1963). Anatomy and physiology of sound receptors in Invertebrates. I n “Acoustic Behaviour of Animals” (Ed. R. G. Busnel), pp. 412-433. Elsevier, Amsterdam, London, New York. Autrum, H., Schwartzkopff, F. and Swoboda, H. (1961). Der EinfluO der Schallrichtung auf die Tympanal potentiale von Locusta migratoria L. Biol. Zbl. 80,385-402. Baier, L. J. (1930). Contribution to the physiology of the stridulation and hearing of insects. Zool. Jb. Abt. 3, Allg. Zool. u. Physiol. 47, 15 1-248. Ball, E. and Young, D. (1974). Structure and development of the auditory system in the prothoracic leg of the cricket Teleogryllus commodus (Walker). 11. Postembryonic development. Z . Zellforsch. 147,3 13-324. Bailey, W. J. (1970). The mechanics of stridulation in bush crickets (Tettigoniidae, Orthoptera) I. The tegminal generator. J. exp. Biol. 5 5 495-505. Bailey, W. J. and Broughton, W. B. (1970). The mechanics of stridulation in bush crickets (Tettigonudae, Orthoptera). 11. Conditions for resonance in the tegminal generator. J. exp. Biol. 52,507-5 17. Bailey, W. J. and Robinson, D. (1971). Song as a possible isolating mechanism in the genus Homorocoryphus (Tettigonioidea, Orthoptera). Anim. Behav. 19,390-397. Bennet-Clark, H. C. (1970). The mechanism and efficiency of sound production in mole crickets. J. exp. Biol. 5 5 6 1 9 - 6 5 2 . Bennet-Clark, H. C. (1971). Acoustics of insect song. Nature, Lond. 234, 255-259. Bennet-Clark, H. C. (1975). Sound production in insects. Sci. Progr. Oxf. 62,263-283. Bennet-Clark, H. C. and Ewing, A. W. (1969). Pulse interval as a critical parameter in the courtship song of Drosophila melanogaster. Anim. Behav. 17,755-759. Bennet-Clark, H. C. and Ewing, A. W. (1970). The love song of the fruit fly. Sci. Amer. 223,8592.
Bentley, D. R. (1969a). Intracellular activity in cricket neurons during the generation of behavior patterns. J. Insect Physiol. 15,677-699. Bentley, D. R. (1969b). Intracellular activity in cricket neurons during the generation of song patterns. Z . vergl. Physiol. 62,267-283. Bentley, D. R. (1971). Genetic control of an insect neuronal network. Science, 174, 1139-1 141. Bentley, D. R. (1977). Control of cricket song patterns by descending interneurons. J. Comp. Physiol. 116, 19-38. Bentley, D. R. and Hoy, R. R. (1970). Postembryonic development of adult motor patterns in crickets: A neural analysis. Science, 170, 1409-141 1.
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
343
Bentley, D. R. and Hoy, R. R. (1972). Genetic control of the neuronal network generating cricket song patterns. Anim. Behav. 2 0 , 4 7 8 4 9 2 . Bentley, D. R. and Hoy, R. R. (1974). The Neurobiology of cricket song. Scientific American, 231,3644. Bentley. D. R. and Kutsch, W. (1966). The neuromuscular mechanism of stridulation in crickets (Orthoptera: Gryllidae). J. exp. Biol. 45, 15 1-164. Bigelow, R. S. (1960). Interspecific hybrids and speciation in the genus Acheta Can.J. Zool. 38, 509. Boeckh, J., Sandri, C. and Akert, K. (1970). Sensorische Eingange und synaptische Verbindungen im Zentralnervensystem von Insekten. Experimentelle Degeneration in der antennalen Sinnesbahn im Oberschlundganglion von Fliegen und Schaben. 2. Zellforsch. 103,429-446. Broughton, W. B. (1963). Method in bio-acoustic terminology. In “Acoustic Behaviour of Animals” (Ed. R. G. Busnel), pp. 3-24. Elsevier, Amsterdam. Broughton, W. B. (1965). Song learning among grasshoppers? New Scientist, 27,338-34 1. Broughton, W. B. (1976). Proposal for a new term “echeme” to replace “chirp” in animal acoustics. Physiol. Ent. 1, 103-106. Broughton, W. B., Samways, M. J. and Lewis, D. B. (1975). Low-frequency sounds in nonresonant songs of some bush crickets (Orthoptera, Tettigonioidea). Ent. exp. appl. 18,44-54. Burns, M. D. (1974). Structure and physiology of the locust femoral chordotonal organ. J. Ins. Physiol. 20, 1319-1339. Burrows, M. (1975). Monosynaptic connexions between wing stretch receptors and flight motoneurons of the locust. J. exp. Biol. 62, 189-219. Busnel, M. C. and Busnel, R. G. (1955). La directive acoustique des deplacements de la femelle d’oecanthus pellucens Scop. In “Colloque sur I’Acoustique des Orthopteres” (Ed. R. G. Busnel), pp. 356-364, Fasc. Hors. Ser. Ann. Epiphyties, Pans. Busnel, R. G., Ed. (1955). “L’Acoustique des Orthopteres,” Fasc. Hors. Ser., Ann. Epiphyties, Paris. Busnel, R. G., Ed. (1963). “Acoustic Behaviour of Animals.” Elsevier, Amsterdam, London, New York. Coillot, J. P. (1974). Analyse du codage d’un mouvement periodique par des recepteurs a I’ktuement d’un insecte. J. Insect Physiol. 20, 1101-1 116. Coillot, J. P. (1974). La conversion analogique numerique des recepteurs a I’etirement de la platte rndathoracique du criquet Schisrocerca gregaria. J. Insect Physiol. 21,423433. Coillot, J. P. and Boistel, J. (1968). Localisation et description de recepteurs a I’etirement au niveau de I’articulation tibio-femorale de la platte sauteuse du criquet, Schistocerca gregaria. J. Insect Physiol. 14, 1661-1667. Coillot, J. P. and Boistel, J. (1969). Etude de l’activite electrique propagee de recepteurs a l’etirement de la platte mdathoracique du criquet, Schistocerca gregaria. J. Insect Physiol. 15, 1449-1470. Counter, S . A. (1976). An electrophysiological study of sound sensitive neurons in the “Primitive ear” of Acheta domesticus. J. Insect Physiol. 22, 1-8. Dathe, H. H. (1974). Studies on the phonotactic behavior of the cricket Gryllus bimaculatus (Insecta, Orthopteroidea). Forma er Funcrio, 7, 7-20. Delcomyn, F. (1973). Motor activity during walking in the cockroach Periplaneta americana. J. exp. Biol. 59, 643-654. Dragsten, P. R., Webb, W. W., Paton, J. A. and Capranica, R. R. (1974). Auditory membrane vibrations: measurement at subangstrnm levels by optical heterodyne spectroscopy. Science, 185,55-57. Dubrovin, N. N. and Zhantiev, R. D. (1970). Sound signals of grasshoppers of the family Tettigoniidae (Orthoptera). Zool. J. (in Russian), 49, 1001-1014. Dumortier, B. (1963a). Morphology of sound emission apparatus in arthropoda. In “Acoustic
344
NORBERT ELSNER AND ANDREJ V POPOV
Behaviour of Animals” (Ed. R. G. Busnel), pp. 277-345. Elsevier, Amsterdam, London, New York. Dumortier, B. (1963b). The physical characteristics of sound emissions in Arthropoda. In “Acoustic Behaviour of Animals” (Ed. R. G. Busnel), pp. 346-373. Elsevier, Amsterdam, London, New York. Dumortier, B. (1963~).Ethological and physiological study of sound emission in arthropoda. In “Acoustic Behavior of Animals” (Ed. R. G. Busnel), pp. 583-654. Elsevier, Amsterdam, London, New York. Edwards, J. S. and Palka, J. (1974). The cerci and abdominal giant fibres of the house cricket, Acheta domesticus I. Anatomy and physiology of normal adults. Proc. Roy. Soc. Lond. B 185,83-103. Eggers, F. (1919). Das thorakale bitympanale Organ einer Gruppe der Lepidoptera Heterocera. Zool. Jb. Abt. Anat. Ontog. Tiere, 41,273-376. Eibl, E. (1974). Verlauf des Hornerven und Projektion der im Hornerven verlaufenden Fasern im Prothorakalganglion von Grillen mit Hilfe der CoS-Methode. Staatsexamensarbeit, Univ. Koln. Elder, H. Y. (1971). High frequency muscles used in sound production by a katydid. 11. Ultrastructure of the singing muscles. Biol. Bull. 141,434-448. Elepfandt, A. von (1975). Neurale Koordinationsmechanismen im Mesothorax von Gryllus campestris L. Diss. Univ. Koln. Elsner, N. (1967). Muskelaktivitat und Lauterzeugung bei einer Feldheuschrecke. Zool. Anz. 3 I, SUPPI.Bd. 592-601. Elsner, N. (1 968). Die neuromuskularen Grundlagen des Werbeverhaltens der Roten Keulenheuschrecke Gomphocerippus rufus L. Z . vergl. Physiol. 60,308-350. Elsner, N. (1 969). Kommandofasern im Zentralnervensystem der Heuschrecke Gastrimargus africanus. Zool. Anz. 33, Suppl. Bd. 465-471. Elsner, N. (1970). Die Registrierung der Stridulationsbewegungen bei der Feldheuschrecke Chortippus mollis mit Hilfe von Hallgeneratoren Z. vergl. Physiol. 68,417-428. Elsner, N. (1973). The central nervous control of courtship behaviour in the grasshopper Gomphocerippus rufus. L. In Symp. Invert. Neurobiology, Tihany, Hungary, Hung. Acad. Sciences. Elsner, N. (l974a). Neurothology of sound production in gomphocerine grasshoppers. I. Song patterns and stridulatory movements. J. comp. Physiol. 88,67- 102. Elsner, N. (1974b). Neural economy: Bifunctional muscles and common central pattern elements in leg and wing stridulation of the grasshopper Stenobothrus rubicundus. J. comp. Physiol. 89,227-236. Elsner, N. (1978). Evolutionary relationships of flight and leg stridulation in acridid grasshoppers (in prep.). Elsner, N. and Helversen, 0. v. ( I 976). Altersbedingte Abwandlungen der Stridulationsmuster bei Feldheuschrecken. Verh. Dtsch. Zool. Ges. Hamburg, p. 280. Elsner, N. and Huber, F. (1969). Die Organisation des Werbegesangs der Heuschrecke Gomphocerippus rufus L. In Abhangigkeit von zentralen und peripheren Bedingungen. Z. vergl. Physiol. 65, 389-423. Elsner, N. and Huber, F. (1973). Neurale Grundlagen artspezifischer Kommunikation bei Orthopteren. Fortschr. Zool. 22, 1-48. Enger, P. S., Aidley, D. J. and Szabo, T. (1969). Sound reception in the brazilian cicada Fidicina rana Walk. J. exp. Biol. 51, 339-345. Ewing, A. W. and Bennet-Clark, H. C. (1968). The courtship songs of Drosophila. Behaviour 31,288-301. Ewing, A. and Hoyle, G. (1965). Neuronal mechanism underlying control of sound production in a cricket: Acheta domesticus. J. exp. Biol. 43, 139-153.
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
345
Faber, A. (1953). Laut- und Gebardensprache bei Insekten: Orthoptera (Geradflugler). Teil I. Mittl. Mus. Nutursk. Stuttgart. Fentress, J. C., Ed. (1976). “Simpler networks and behavior.” Sinauer Ass. Inc., Sunderland, Mass. Friedman, M. N. (1972a). A light and electron microscopic study of sensory organs and associated structures in the foreleg tibia of the cricket, Gryllus ussimilis. J. Morphol. 138, 263-328. Friedman, M. N. (1972b). An electron microscopic study of the tympanal organ and associated structures in the foreleg tibia of the cricket, Gryllus assimilis. J. Morphol. 138,329-348. Frings, M. and Frings, H. (1960). Sound production and sound reception by insects. A bibliography. Perms. State Univ. Press. Fulton, B. B. (1933). Inheritance of song in hybrids of two subspecies of Nemobius fusciatus. Ann. Entom. SOC.Am. 26, 368. Gettrup, E. (1963). Phasic stimulation of a thoracic stretch receptor in locusts. J. exp. Biol. 40, 323-333. Ghiradella, H. (1971). Fine structure of the noctuid moth ear. I. The transducer area and connections to the tympanic membrane in Feltiu subgothicu Hawthorn. J. Morphol. 134,2146. Gnatzy, W. and Schmidt, K. (1971). Die Feinstruktur der Sinneshaare auf den Cerci von Gtyllus bimuculutus Deg. (Saltatoria, Gryllidae). I. Faden- und Keulenhaare. Z . Zellforsch. 122, 190-209. Gray, E. G. (1960). The fine structure of the insect ear. Phil. Trans. Roy. SOC.B243,75-94. Halfmann, K. (1977). Adulte und larvale Stridulationsmuster bei Feldheuschrecken der Gattung Chorthippus. Diplomarbeit, Univ. Koln. Hartley, J. C. and Robinson, D. J. (1 976). Acoustic behaviour of both sexes of the speckled bush cricket Leptophyes punctutissimu. Physiol. Entomol. 1. Haskell, P. T. (1958). Stridulation and associated behaviour in certain Orthoptera. 2. Stridulation of females and their behaviour with males. Anim. Behuu. 6 , 2 7 4 2 . Haskell, P. T. (1960). Stridulation and associated behaviour in certain Orthoptera. The influence of the gonads. Anim. Behau. 8 , 7 6 8 1 . Haskell, P. T. (196 1). “Insect Sounds.” Witherby, London. Haskell, P. T. (1964). Sound production. In “The Physiology of Insecta” (Ed. M. Rockstein), I, 563-668. Academic Press, New York, London. Heath, J. E. and Josephson, R. K. (1970). Body temperature and singing in the katydid Neoconocephalus robustus (Orthoptera, Tettigonioideae). Biol. Bull. 138, 272-285. Heiligenberg, W. (1966). The stimulation of territorial singing in house crickets (Achetu domesticus). 2. vergl. Physiol. 53, 114-129. Heiligenberg, W. (1969). The effect of stimulus chirps on a crickets chirping (Achetu domesticus). Z . vergl. Physiol. 65, 70-97. Helversen, D. von (1972). Gesang des Miinnchens und Lautschema des Weibchens bei der Feldheuschrecke Chorthippus biguttulus (Orthoptera, Acrididae). J. comp. Physiol. 81, 381422. Helversen, D. von and Elsner, N. (1975). Untersuchungen zur Neurogenetik des Heuschreckengesangs. Verh. Dtsch. Zool. Ges. Karlsruhe, 1975, p. 113. Helversen, 0. von and Elsner, H. (1977). The stridulatory movements of acridid grasshoppers recorded with an opto-electronic device. J. comp. Physiol. A (in press). Helversen, D. von and Helversen, 0. von (1975a). Verhaltensgenetische Untersuchungen am akustischen Kommunikationssystem der Feldheuschrecken 1. Der Gesang von Artbastarden zwischen Chorthippus biguttulus und Chorthippus mollis. J. comp. Physiol. 104,273-299. Helversen, D. von and Helversen, 0. von (1975b). Verhaltensgenetische Untersuchungen am akustischen Kommunikationssystem der Feldheuschrecken. 11. Das Lautschema von
346
NORBERT ELSNER AND ANDREJ V POPOV
Artbastarden zwischen Chorthippus biguttulus und Chorthippus mollis. J. comp. Physiol. 104,301-323. Helversen, D. von and Helversen, 0. von (197%). Das Lautschema von Bastarden zwischen zwei Feldheuschreckenarten. Verh. Dtsch. Zool. Ges. Karlsruhe, 1975, p. 114. Helversen, D. von and Helversen, 0. von (1977). Korrespondenz zwischen Gesang und auslosendem Schema bei Feldheuschrecken. NOUQACIQLeopoldina, (in press). Hill, K. G. (1974). Carrier frequency as a factor in phonotactic behaviour of female crickets (Teleogryllus commodus). J. comp. Physiol. 93, 7-18. Hill, K. G. (1975). Acoustic communication in the Australian field crickets Teleogryllus commodus and T. oceanicus (Orthoptera: Gryllidae). Doct. Thesis, University of Melbourne. Hormann-Heck, S. von (1957). Untersuchungen uber den Erbgang einiger Verhaltensweisen bei Grillenbastarden (Gryllus campestris L. und Gryllus bimQcu/atur de Geer). Z . Tierpsychol. 14,137-183. Horridge, G. A. (1960). Pitch discrimination in Orthoptera (Insecta) demonstrated by responses of central auditory neurons. Nature, Lond. 185,623-624. Horridge, G. A. (1961). Pitch discrimination in locusts. Proc. Roy. SOC.B155, 218-231. Horridge, G. A., Scholes, J. H., Shaw, S. and Tunstall, J. (1964). Extracellular recordings from single neurones in the optic lobe and brain of the locust. In “The Physiology of the Insect Central Nervous System” (Eds J. E. Treherne and J. W. L. Beament), pp. 165-202. Academic Press, London, New York. Hoy, R. R. (1974). Genetic control of acoustic behavior in crickets. Amer. 2001.14, 10671080. Hoy, R. R., Hahn, J. and Paul R. C. (1977). Hybrid cricket auditory behavior: Evidence for genetic coupling in animal communication. Science 195,82-84. Hoy, R. R. and Paul, R. C. (1973). Genetic control of song specfity in crickets. Science, 180, 82-83. Hoyle, G. (1957). Nervous control of insect muscles. In “Recent Advances in Invertebrate physiology” (Ed. Scheer), pp. 73-98. University of Oregon Press. Hoyle, G. (1964). Exploration of neuronal mechanisms underlying behaviour in insects. In “Neural Theory and Modeling” (Ed. R. F. Reiss), pp. 346-376, Stanford University Press. Hoyle, G. (1970). Cellular mechanisms underlying behavior-neuroethology. In “Advances in Insect Physiology”, Vol. 7, 3 4 9 4 4 4 . Academic Press, New York and London. Huber, F. (1952). Verhaltensstudien am Mannchen der Feldgrille (Gryllus campestris L.) nach Eingriffen in das Zentralnervensystem. Zool. Anz. 46,138-149. Huber, F. (1955). Sitz und Bedeutung nervoser Zentren fur Instinkthandlungen beim Mannchen von Gryllus campestris L. Z . Tierpsychol. 12, 12-48. Huber, F. (1 960). Untersuchungen uber die Funktion des Zentralnervensystems und insbesondere des Gehirns bei der Fortbewegung und der Lauterzeugung der Grillen. Z. uergl. Physiol. 44,60-132. Huber, F. (1962). Central nervous control of sound production in crickets and some speculations on its evolution. Evolution, 14,429-442. Huber, F. (1963). The role of the central nervous system in Orthoptera during coordination and control of stridulation. I n “Acoustic Behavior of Animals” (Ed. R. G. Busnel), pp. 440-488. Elsevier, Amsterdam. Huber, F. (1965). Aktuelle Probleme der Physiologie des Nervensystems der Insekten. Naturw. Rundschau, 18,143-156. Huber, F. (1967). Central control of movements and behaviour of invertebrates. In “Invertebrate Nervous System” (Ed. C. A. G. Wiersma), pp. 333-35 1. University of Chicago Press. Huber, F. (1970). Nervose Grundlagen der akustischen Kommunikation bei Insekten. Rhein. Westf. Akad. Wiss. Westdeutscher Verlag Opladen.
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
347
Huber, F. (1974). Neural integration (central nervous system). In “The Physiology of Insecta” (Ed. M. Rockstein), 2nd Edn., Vol. IV, 3-100. Academic Press, London, New York. Huber, F. (1975a). Principles of motor co-ordination in cyclically recurring behaviour in insects. In “Simple Nervous Systems” (Eds. P. N. R. Usherwood and D. R. Newth), pp. 381-413. Arnold, London. Huber, F. (1975b). Sensory and neuronal mechanisms underlying acoustic communication in Orthopteran insects. In “Sensory Physiology and Behavior” (Eds R. Galun, P. Hillman, I. Parnas and R. Werman), pp. 55-97. Plenum, New York, London. Innenmoser, J. (1 974). Stabilitat des motorischen Musters der Gesange von Gryllus bimaculatus de Geer vor und nach subletaler Applikation des insektiziden Phosphorsaureesters Parathion. Diss. Univ. Koln. Isely, F. B. (1936). Flight stridulation in American acridians (Orthoptera: Acrididae). Ent. News, 47, 199-205. Jacobs, W. (1953). Verhaltensbiologische Studien an Feldheuschrecken. Beihdt I zur Z . Tierpsychol. 10. Jones, M. D. R. (1966a). The acoustic behaviour of the brush cricket Pholidoptera griseoaptera. I. Alternation, synchronism and rivalry between males. J.exp. Biol. 45, 15-30. Jones, M. D. R. (1966b). The acoustic behaviour of the bush cricket Pholidoptera griseoaptera. 11. Interaction with artificial sound signals. J. exp. Biol. 45, 3 1 4 4 . Jones, M. D. R. (1974). The effect of acoustic signals on the chirp rhythm in the bush cricket Pholidoptera griseoaptera. J. exp. Biol. 61,345-355. Jones, M. D. R. and Dambach, M. (1973). Response to sound in crickets without tympana1 organs (Gryllus campestris L.).J. comp. Physiol. 87, 89-98. Josephson, R. K. (1973). Contraction kinetics of the fast muscles used in singing by a katydid J. exp. Biol. 59,781-801. Josephson, R. K. and Halverson, R. C. (1971). High frequency muscles used in sound production by a katydid. I. Organization ofthe motor system. Biol. Bull. 141,413-433. Josephson, R. K., Stokes, D. R. arid Chen, V. (1975). The neural control of contraction in a fast insect muscle. J. exp. 2001.193,281-300. Kalmring, K. (1 97 1). Akustische Neuronen im Unterschlundganglion der Wanderheuschrecke Locusta migratoria. Z. vergl. Physiol. 72,95-110. Kalmring, K. (1975a). The afferent auditory pathway in the ventral cord of Locusta migratoria (Acrididae). I. Synaptic connectivity and information processing among the auditory neurons of the ventral cord J. comp. Physiol. 104, 103-141. Kalmring, K. (1975b). The afferent auditory pathway in the ventral cord ofLocusta migratoria (Acrididae). 11. Responses of the auditory ventral cord neurons to natural sounds. J . comp. Physiol. 104, 143-159. Kalmring, K., Rheinlaender, J. and Rehbein, H. (1972a). Akustische Neuronen im Bauchmark der Wanderheuschrecke Locusta migratoria. Z. vergl. Physiol. 76,3 14-332. Kalmring, K., Rheinlaender, J. and Romer, H. (1972b). Akustische Neuronen im Bauchmark von Locusta migratoria. Der EinfluR der Schallrichtung auf die Antwortmuster. J. comp. Physiol. 80,325-352. Katsuki, Y. and Suga, N. (1958). Electrophysiological studies on hearing in common insects of Japan. Proc. Jap. Acad. 34,633-638. Katsuki, Y . and Suga, N. (1960). Neural mechanisms of hearing in insects. J. exp. Biol. 37,279290. Kay, R. E. (1969). Acoustic signalling and its possible relationship to assembling and navigation in the moth, Heliothis zea. J. Ins. Physiol. 15,989-1001. Kevan, D. K. Mc.E. (1955). Methodes inhabituelles de production de son chez les Orthopteres. In “Colloque sur I’Acoustique des Orthopteres” (Ed. R. G. Busnel), pp. 103-141. Fasc. Hors. Ser. Ann. Epiphyties, Paris.
348
NORBERT ELSNER AND ANDREJ V. POPOV
Knjazev, A. N. (1976). Structure and function of the cercal system of crickets Gryllus bimaculafus (in Russian). Diss. Thesis Sechenov Institute of evolutionary Physiol. Bioch., Leningrad. Knjazev, A. N. (1977). Reaction of single cercal mechanoreceptors of a cricket to mechanical stimulation. J. evol. Biochem. Physiol. 13 (4), (in press). Knjazev, A. N. and Popov, A. V. (1977). Reaction of single cercal mechanoreceptors of a cricket to sound and sinusoidal mechanical stimulation. Reports Acad. Sci. USSR, 232 (5) 12111214. Kohsen, R. ( 1 976). Der Gesang der Feldheuschreckenweibchen und seine neuromuskularen Grundlagen. Staatsarbeit, UNv. Koln. Koppers, S. (1977). Die neuromuskularen Grundlagen des Gesangs der Feldheuschrecke Stauroderus scalaris. Staatsarbeit Univ. Koln. Kiippers, J. (1974). Measurements on the ionic milieu of the receptor terminal in mechanoreceptive sensilla of insects. In “Mechanoreception” (Ed. J. Schwartzkopff), pp. 387-394. Westdeutscher Verlag, Opladen. Kutsch, W. (1969). Neuromuskulare Aktivitat bei verschiedenen Verhaltensweisen von drei Grillenarten. Z . vergl. Physiol. 63,335-378. Kutsch, W. and Huber, F. (1970). Zentrale versus periphere Kontrolle des Gesangs von Grillen (Gryllus campestris). 2. vergl. Physiol. 67, 140-159. Kutsch, W. and Otto, D. (1972). Evidence for song production independent of head ganglia in Gryllus campestris. J. comp. Physiol. 81, 1 15- 1 19. Lamparter, H. E., Steiger, U., Sandri, C. and Akert, K. (1969). Zum Feinbau der Synapsen im Zentralnervensystem der Insekten. Z . Zellforsch. 99,432-442. Leroy, Y. (1966). Signaux acoustiques comportement et systematique de quelque especes de Gryllides (Orthopteres, Ensiferes). Bull. Biol. Fr. Belg. 100, 63-1 34. Lewis, D. B., Seymour, C. and Broughton, W. B. (1975). The responses characteristics of the tympana1 organs of two species of brush cricket and some studies of the problem of sound transmission. J. comp. Physiol. 104, 325-35 1. Lewis, D. B. (1974a). The physiology of the Tettigoniid ear. I. The implications of the anatomy of the ear to its function in sound reception. J. exp. Biol. 60, 821-837. Lewis, D. B. (1974b). The physiology of the Tettigoniid ear. 11. Response characteristics of the ear to differential inputs: lesion and blocking experiments. J. exp. Biol. 60,839-85 1. Lewis, D. B. ( 1 9 7 4 ~ )The . physiology of the Tettigoniid ear. 111. The response characteristics of the intact ear and some biophysical considerations. J. exp. Biol. 60, 853-859. Lewis, D. B. (1974d). The physiology of the Tettigoniid ear. IV. A new hypothesis for acoustic orientation behaviour. J. exp. Biol. 60,86 1-869. Lindberg, D. (1974). Die Bistabilitat des Stridulationsmusters der Heuschrecke Chorthippus biguttulus mit besonderer Beriicksichtigung der sensorischen Kontrolle. Staatsarbeit, Univ. Koln. Lindberg, D. and Elsner, N. (1977). Sensory influence upon grasshopper stridulation. Naturw. 64, 342. Loher. W. ( 1 966). Die Steuerung sexueller Verhaltensweisen und der Oocytenentwicklung bei Gomphocerus rufus L. Z . vergl. Physiol. 53,277-3 16. Loher, W. and Chandrashekaran, M. K. (1972). Communicative behaviour of the grasshopper Syrbula fuscoviffafa(Thomas) (Gomphocerinae) with particular consideration of the male courtship. Z . Tierphyschol. 31, 78-97. Loher, W. and Huber, F. (1964). Experimentelle Untersuchungen uber das Sexualverhalten des Weibchens der Heuschrecke Gomphocerus yfus L. J. Insect Physiol. 10, 13-36. Loher, W. and Huber, F. (1966). Nervous and endocrine control of sexual behaviour in a grasshopper (Gomphocerus rufus L.). Symp. Soc. exp. Biol. 20,381-400. Mancini, G. and Frontali, N. (1967). Fine structure of the mushroom body neuropile of the brain of the cockroach, Periplaneta americana. Z . Zellforsch. 83,334-343.
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
349
Mandelshtam, J. E. (1974). Structural peculiarities of synapses in the locust, Locustu migrutoriu, J. evol. Bioch. Physiol. (Russian) 10,36-39. Markl, H. (1972). Neue Entwicklungen in der Bioakustik der wirbellosen Tiere. J. Ornith. 113, 91-104.
Markl, H. (1973). Leistungen des Vibrationssinns bei wirbellosen Tieren. Fortschr. Zool. 21, 100-120.
Markovich, A. M. (1976). Functional properties of peripheral and prothoracic auditory neurons of the cricket, Gryllus bimuculutus. Diss. Thesis, Sechenov Institute of evolutionary Physiol. Bioch. Leningrad. McHugh, R. (1972). Aspects of acoustic interaction in the bushcricket genus Metrioptera (Orth., Tettigonioidae). Ph. D.Thesis, University of London. McKay, J. M. (1969). The auditory system of Homorocoryphus (Tettigonioidea, Orthoptera). J. exp. Biol. 51, 787-802. McKay, J. M. (1970). Central control of an insect sensory interneurone. J. exp. B i d . 53, 137145.
Mebes, H.-D. (1974). The biophysics of sound production of leafhoppers. Fomu et Functio, 7, 95-1 18.
Michel, K. (1974). Das Tympanalorgan von Gryllus bimuculutus Degeer (Saltatoria, Gryllidae). Z . Morph. Tiere, 77,285-315. Michel, K. (1975). Das Tympanalorgan von Cicada orni L. (Cicadina, Homoptera). Erne Lichtund Elektronenmikroskopische Untersuchung. Zoomorphologie, 82,63-78. Michelsen, A. (1966). Pitch discrimination in the locust ear: observation on single sense cells. J. Insect Physiol. 12, 11 19-1 131. Michelsen, A. (1968). Frequency discrimination in the locust ear by means of four groups of receptor cells. Nature, Lond. 220,585-586. Michelsen, A. (1971). The physiology of the locust ear. I. Frequency sensitivity of single cells in the isolated ear. 11. Frequency discrimination based upon resonances in the tympanum. 111. Acoustical properties of the intact ear. Z. vergl. Physiol. 71,49-128. Michelsen, A. (1974). Hearing in invertebrates. In “Handbook of Sensory Physiology” (Eds H. Autrum, R. Jung, W. R. Lowenstein, D. M. McKay and H. L. Teuber), Vol. V/I, pp. 389422. Springer-Verlag, Berlin, Heidelberg, New York. Michelsen, A. and Nocke, H. (1974). Biophysical aspects of sound communication in insects. In “Advances of Insect Physiology” (Eds J. E. Treherne, M. J. Berridge and V.B. Wigglesworth), 10,247-296. Academic Press, London, New York. Miller, L. A. (1974). Directional hearing in the locust, Schistocercu greguriu. Abstract from SOC. Neurosci. 4th annual meeting, Program and Abstracts, October 1974, p. 339. Mrahl, B. and Miller, L. A. (1976). Ultrasonic clicks produced by the peacock butterfly: a possible bat-repellent mechanism. J. exp. B i d . 64, 639-644. Morris, G. K. (1970). Sound analysis of Metriopteru sphugnorum (Orthoptera: Tettigoniidae). Can. entom. 102,363-368. Morris, G. K., Kerr, G. E. and Gwynne, D. (1975). Calling song function in the bog katydid, Metriopteru sphugnorum (F. Walker) (Orthoptera, Tettigoniidae): Female phonotaxis to normal and altered song. Z . Tierpsychol. 37,502-5 14. Morris, G. K. and Pipher, R. E. (1967). Tegminal amplifiers and spectrum consistencies in Conocephulus nigropleurum (Brunner), Tettigoniidae. J. Insect Physiol. 13, 1075-1085. Morris, G. K. and Pipher, R. E. (1972). The relation of song structure to tegminal movement in Metriopteru sphugnorum (Orthoptera: Tettigoniidae). Can. Ent. 104,977-985. Moss, D. (1967). Proprioceptoren im Thorax und Abdomen von Grillen (Orthoptera, Gryllidae). Zool. Anz. Suppl. Bd. 31, 123-136. Moss, D. (197 I). Sinnesorgane im Bereich des Flugels der Feldgrille (Gryllus cumpestris L.) und ihre Bedeutung fur die Kontrolle per Singbewegung und die Einstellung der Flugellage. Z . vergl. Physiol. 73, 53-83.
3 50
NORBERT ELSNER AND ANDREJ V POPOV
Nicklaus, R. (1965). Die Erregung einzelner Fadenhaare von Periplunetu americunu in Abhangigkeit von der GroBe und Richtung der Auslenkung. Z. Vergl. Physiol. 40,33 1-362. Nocke, H. (1971). Biophysik der Schallerzeugung durch die Vorderflugel der Grillen. Z. vergl. Physiol. 74,272-3 14. Nocke, H. (1972). Physiological aspects of sound communication in crickets (Gryllus campestris L.). J. comp. Physiol. 80, 141-162. Nocke, H. (1974). The tympanal trachea as an integral part of the ear in Acripezu reticulafa Guerin (Orthoptera, Tettigoniidae). Z. Naturforsch. 2!k, 652-654. Nocke, H. (1975). Physical and physiological properties of the tettigoniid (“grasshopper”) ear. J. comp. Physiol. A100, 25-57. Otte, D. (1970). A comparative study of communicative behavior in grasshoppers. Misc. Publ. Mus.Zool. Univ. Mich. 141, 1-168. Otte, D. (1972). Simple versus elaborate behaviour in grasshoppers. An analysis of communication in the genus Syrbula. Behaviour, 42,29 1-322. Otto, D. (1967). Untersuchungen zur nervosen Kontrolle des Gril!engesangs. Zool. Anz. 31, SUPPI.Bd. 585-592. Otto, D. (1969). Hirnreizinduzierte komplexe Verhaltensfolgen bei Grillen. Zool. Anz. 33, Suppl. Bd. 472-477. Otto, D. (197 1). Untersuchungen zur zentralnervosen Kontrolle der Lauterzeugung von Grillen. Z. vergl. Physiol. 74,227-27 1. Pabst, H. (1965). Elektrophysiologische Untersuchung des Streckrezeptors am Flugelgelenk der Wanderheuschrecke Locusfa migraforia. Z. vergl. Physiol. 50,498-54 1. Pasquinelly, F. and Busnel, M. C. (1955). etudes priliminaires sur les mecanismes de la production des sons par les Orthopteres. Ann. des Epiphyties. I.N.R.A., 145. Paul, D. H. (1973). Central projections of the tympanic fibres in noctuid moths. J. Insect Physiol. 19, 1785-1792. Paul, D. H. (1974). Responses to acoustic stimulation of thoracic neurons in noctuid moths. J. Insect Physiol. 20,2205-2218. Payne, R. S., Roeder, R. D. and Wallman, J. (1966). Directional sensitivity of the ears of noctuid moths. J. exp. Biol. 44, 17-31. Pearson, K. G. (1972). Central programming and reflex control of walking in the cockroach. J . exp. Biol. 56, 173-193. Perdeck, A. C. (1958). The isolating value of specific song patterns in two sibling species of grasshoppers (Chorfhippus brunneus Thunb. and Chorfhippusbiguttulus L.). Behaviour, 12, 1-75. Petrovskaya, E. D. (1969). On frequency selectivity of cercal receptors in the cricket, Gryllus domesticus. J. evol. Bioch. Physiol. (Russian), 5,337-338. Petrovskaya, E. D., Rojkova, G. I. and Tokareva, V. S. (1970). Single cercal receptor characteristics in the cricket (Gryllus domesticus). Biophysika (Russian), 15, 1112-1 119. Pipher, R. E. and Morris, G. K. (1974). Frequency modulation in Conocephalus nigropleurum, the black-sided meadow katydid (Orthoptera: Tettigoniidae). Can. Ent. 106, 9971001. Pitman, R., Tweedle, C. D. and Cohen, M. J. (1972). Branching of central neurons: Intracellular cobalt injection for light and electron microscopy. Science, 176,4 12-414. Popov, A. V. (1965). Electrophysiological studies on peripheral auditory neurons in the locust (in Russian). J. evo!. Bioch. Physiol. I, 239-250. Popov, A. V. (1968). Electrophysiological study of functional properties of peripheral and central neurons in the auditory system of the locust (in Russian). Diss. Thesis Pavlov Institute of Physiol. Leningrad. Popov, A. V. (1967a). Synaptic transmission at the level of the first synapses of the auditory system in Locusta migraforia (in Russian). In “Evolutionary Neurophysiology and Neurochemistry” (Ed. E. M. Krebs), pp. 54-67. Nauka, Leningrad.
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
351
Popov, A. V. (1967b). Characteristics of activity of the central neurons in the auditory system of the locust (in Russian). I n “Problems of Physiological Acoustics: Mechanisms of Hearing” (Ed. G. V. Gersuni), 6, 108-121. Nauka, Leningrad. Popov, A. V. (1969). Comparative analysis of sound signals and some principles of auditory system organization in cicadas and Orthoptera (in Russian). I n “Modern Problems of Structure and Function of the Nervous System of Insects” (Ed. A. K. Voskresenskaja), pp. 182-221. Nauka, Leningrad. Popov, A. V. (1971a). Sound-producing apparatus and structure of the calling song in the domestic cricket, Achetu domesticus. J. evol. Bioch. Physiol. 7,87-95. Popov, A. V. (1971b). Synaptic transformation in the auditory system of insects. In “Sensory Processes at the Neuronal and Behavioral Levels” (Ed. G. V. Gersuni), pp. 301-320. Academic Press, New York. Popov, A. V. (1 972). Sounds of crickets from southern regions of the european part of the USSR (in Russian). Rev. Entom. USSR, 51, 17-36. Popov, A. V. (1973). Frequency selectivity of the reaction of auditory neurons in the first thoracic ganglion of the cricket Gryllus bimuculutus (in Russian). J . evol. Bioch. Physiol. 9, 265-277. Popov, A. V. (1975). The structure of tymbals and characteristics of sound signals of singing cicadas (Homoptera, Cicadidae) from the southern regions of the USSR. (In Russian). Rev. Entom. USSR, 54,258-290. Popov, A. V. and Shuvalov, V. F. (1974). Spectrum intensity and directional characteristics of the calling song emission in the cricket Gryllus campestris under natural conditions (in Russian). J. evol. Bioch. Physiol. 10, 72-80. Popoov, A. V. and Shuvalov, V. F. (1976). Phonotactic behaviour of crickets. J. comp. Physiol. 119, 111-126. Popov, A. V. and Svetlogorskaya, I. D. (1971). Ultrastructural organization of the auditory nerve in the locust Locustu migrutoriu (on the problem of interaction between the receptors, (in Russian)). J. moi. Bioch. Physiol. 7 , 5 16-521. Popov, A. V., Shuvalov, V. F., Knjazev, A. N. and Clar-Spasovskaja, N. A. (1974a). Communication calling songs of crickets (Orthoptera, Gryllidae) from south-western Tadjikistan (in Russian). Rev. Entom. USSR, 53, 258-279. Popov, A. V., Shuvalov, V. F., Svetlogorskaja, I. D. and Markovich, A. M. (1974b). Acoustic behaviour and auditory system in insects. In “Mechanoreception” (Ed. J. Schwartzkopn), pp. 28 1-306. Westdeutscher Verlag, Opladen. Popov, A. V., Shuvalov, V. F. and Markovich; A. M. (1975). Spectrum of the calling songs, phonotaxis and the auditory system in the cricket Gryllus bimuculutus. J. evol. Bioch. Physiol. 11,453-460. Popov, A. V., Markovich, A. M. and Andjan, A. S. (1976a). Auditory interneurons in the prothoracic ganglion of the cricket Gryllus bimuculutus I. Large segmental neuron (LSN). (In prep.). Popov, A. V., Markovich, A. M. and Andjan, A. S. (1976b). Auditory interneurons in the prothoracic ganglion of the cricket Gryllus birnuculutus. 11. High-frequency ascending neuron (HFAN). (In prep.). Popov, A. V., Markovich, A. M. and Andjan, A. S. (1976~).Auditory interneurons in the prothoracic ganglion of the cricket Gryllus bimuculutus. 111. Low-frequency ascending neuron (LFAN). (In prep.). Potente, A. (1975). Untersuchungen zur Morphologie der cercalen Riesenfasern im Bauchmark von Lucusta migratoriu. Diplomarbeit, Lehrstuhl Gr allg. Zool., Ruhr-Univ., Bochum. Pumphrey, R. J. and Rawdon-Smith, A. F. (1936). Hearing in insects. The nature of the response of certain receptors to auditory stimuli. Proc. Roy. SOC.Ser. B121, 18-27. Pumphrey, R. J. and Rawdon-Smith, A. F. (1939). Frequency discrimination in insects: a new theory. Nature, Lond. 143, 806-807.
3 52
NORBERT ELSNER AND ANDREJ V. POPOV
Regen, I. (1913). Uber die Anlockung des Weibchens von Gryllus cumpestris L. durch telephonisch iibertragende Stridulationslaute des Mannchens. Pj7ug. Arch. ges. Physiol. 155, 193-200.
Regen, I. (1923). Uber die Onentierung des Weibchens von Lyogryllus cumpestris L. nach dem Stridulationsschall des Mannchens. S.B. Akud. Wiss. Wien 132, 81-88. Rehbein, H.-G. (1973). Experimentell-anatomische Untersuchungen iiber den Verlauf der Tympanalnervenfasern im Bauchmark von Feldheuschrecken, Laubheuschrecken und Grillen. Verh. dt. Zool. Ges. 66, 184-189. Rehbein, H.-G. (1975). Untersuchungen zur funktionellen Anatomie der Horbahn von Locustu migrutoriu (Acrididae, Orthoptera). Diss. Thesis, Ruhr-Univ., Bochum. Rehbein, H.-G., Kalmring, K. and Romer, H. (1974). Structure and function of acoustic neurons in the thoracic ventral nerve cord of Locustu migrutoriu (Acrididae). J. comp. Physiol. 95, 263-280.
Reid, K. H. (1971). Periodical cicada: mechanism of sound production. Science, 172,949-95 1. Renner, M. (1952). Analyse der Kopulationsbereitschaft des Weibchens der Feldheuschrecke Euthystiru bruchypteru Ocsk. in ihrer Abhangigkeit vom Zustand des Geschlechtsapparates. Z . Tierpsychol. 9, 122-154. Rheinlaender, J. (1975). Transmission of acoustic information at three neuronal levels in the auditory system of Decticus uerruciuorus (Tettigoniidae, Orthoptera). J. comp. Physiol, 97, 1-53.
Reinlaender, J. and Kalmring, K. (1973). Die afferente Horbahn im Bereich des Zentralnervensystems von Decticus uerrucivorus (Tettigonudae). J. comp. Physiol. 85, 36 1 - 4 10. Rheinlaender, J., Kalmring, K. and Romer, H. (1972). Akustische Neurone mit T-struktur irn Bauchmark von Tettigoniiden. J. comp. Physiol. 77,208-224. Rheinlaender, J., Kalmring, K., Popov, A. V. and Rehbein, H.-G. (1976). Brain projections and information processing of biologically significant sounds by two large ventral-cord neurons of Gryllus bimaculatus de Geer (Orthoptera, GryUidae). J. comp. Physiol. 110,25 1-269. Roeder, R. D. (1964). Aspects of the noctuid tympanic nerve response having significancein the avoidance of bats. J. Insect Physiol. 10, 529-546. Roeder, R. D. (1966). Acoustic sensitivity of the noctuid tympanic organ and its range for the cries of bats. J. Insect Physiol. 12, 843-859. Roeder, R. D. (1967). “Nerve Cells and Insect Behavior”, 2nd Edn. Harvard University Press, Cambridge, Mass. Roeder, R. D. (1969a). Acoustic interneurons in the brain of noctuid moths. J. Insect Physiol. 15,825-838.
Roeder, R. D. (1969b). Brain interneurons in noctuid moths: differential suppression by high sound intensities. J. Insect Physiol. 15, 1713-1718. Roeder, R. D. (1970). Episodes in insect brains. Amer. Sci. 58,378-389. Roeder, R. D. (1972). Acoustic and mechanical sensitivity of the distal lobe of the pilifer in choerocampine hawkmoths. J. Insect Physiol. 18, 1249-1 264. Roeder, R. D. (1973). Brain interneurons in noctuid moths: binaural excitation and slow potentials. J. Insect Physiol. 19, 1591-1601. Roeder, K. D. and Treat, A. E. (1957). Ultrasonic reception by the tympanic organ of noctuid moths. J. exp. Zool. 134, 127-157. Roeder, K. D. and Treat, A. E. (1970). An acoustic sense in some hawkmoths (Choerocampinae). J. Insect Physiol. 16, 1069-1086. Romer, H. and Schwartzkopff, J. (1975). Frequenz- und richtungsspezifische Rezeptoren der Wanderheuschrecke. Naturwiss. 62, 58 1. Rowell, F. and McKay, J. M. (1969). An acridid auditory interneurone. 11. Habituation, variation in response level and central control. J. exp. Biol. 5 1, 247-260.
NEUROETHOLOGY OF ACOUSTIC COMMUNICATION
353
Samways, M. J. (1975). Acoustic behaviour and song modification in the bush cricket Platycleis infermedia (Serv.). Ph.D. Thesis, London University. Samways, M. J. (1976a). Song modification in the Orthoptera. I. Proclamation songs of Plufycleis spp. (Tettigoniidae). Physiol. Ent. 1, 131-149. Samways, M. J. (1976b). Song modification in the Orthoptera. 111. Microsyllabic echemes in Platycleis spp. (Tettigoniidae). Physiol. Ent. (in press). Samways, M. J. (1977). Bush cricket interspecific acoustic interactions in the field. J. nut. Hist. (in press). Samways, M. J. and Broughton, W. B. (1976). Song modification in the Orthoptera. 11. Types of acoustic interaction between Platycleis intermedia and other species of the genus (Tettigoniidae). Physiol. Ent. (in press). Schumacher, R. (1973a). Morphologische Untersuchungen der tibialen Tympanalorgane von neun einheimischen Laubheuschreckenarten (Orthoptera, Tettigoniidae). Z . Morph. Tiere, 75, 267-282. Schumacher, R. (1973b). Beitrag zur Kenntnis des tibialen Tympanalorgans von Tettigoniu viridissimu L. (Orthoptera, Tettigoniidae). Mikroskopie, 29, 8-19. Schumacher, R. (1975). Scanning-electron microscope description of the tibial tympana1 organ of the Tettigonioidea (Orthoptera, Ensifera). Z . Morph. Tiere, 81, 209-219. Schiirmann, F. W. (1970). Uber die Struktur der Pilzkorper des Insektenhirns. I. Synapsen im Pedunculus. Z . Zellforsch. 103, 365-38 1. Schwabe, J. ( 1906). Beitrage zur Morphologie und Histologie der tympanalen Sinnesapparate der Orthopteren. Zoologica, Stuttgurt, 50, 1-154. Schwartzkopff, J. (1964). Mechanoreception. In “The Physiology of Insecta” (Ed. M. Rockstein), Vol. 1, pp. 509-561. Academic Press, New York, London. Selverston, A. J. (1974). Structural and functional basis of motor pattern generation in the stomatogastric ganglion of the lobster. Am. 2001.14,957-972. Shuvalov, V. F. and Popov, A. V. (1971). The reaction of the females of the domestic cricket Achefa domestica to sound signals and its changes in ontogenesis. J. evol. Bioch. Physiol. 7 , 6 12-6 16. Shuvalov, V. F. and Popov, A. V. (1973a). Significance of some of the parameters of the calling songs of male crickets Gryllus bimuculatus for phonotaxis of females J. evol. Bioch. Physiol. 9, 177-182. Shuvalov, V. F. and Popov, A. V. (1973b). The importance of the calling song rhythmic pattern of males of the genus Gryllus for phonotaxis of females. Zool. J. 52, 1179-1 185. Shuvalov, V. F. and Popov, A. V. (1976). Mechanism of recognition of male calling songs by conspecific females in crickets (genus Gryllus). In “Comparative Neurophysiology and Neurochemistry” (Ed. E. M. Krebs) (in Russian), pp. 37-43. Nauka, Leningrad. Smith, D. S. (1965). Synapses in the insect nervuus system. In “The Physiology of the Insect Ce’ntral Nervous System” (Eds. J. E. Treherne and J. W. L. Beament), pp. 39-57. Academic Press, New York, London. Spooner, J. D. (1964). The Texas bush katydid-its sounds and their significance. Anim. Behav. 12,235-244. Suga, N. (1960). Peripheral mechanism of hearing in locust. Jag. J. Physiol. 10,533-546. Suga, N. (1963). Central mechanism of hearing and sound localization in insects. J. Insect Physiol. 9,867-873. Suga, N. and Katsuki, Y. (1961). Central mechanism of hearing in insects. J. exp. Biol. 38,545558. Svetlogorskaya, I. D. and Popov, A. V. (1975). Ultrastructural organization of the anterior sensory neuropile in metathoracic ganglion of the locust Locusta rnigratoria (in Russian). J. evol. Bioch. Physiol. 11,419-424.
3 54
NORBERT ELSNER AND ANDREJ V POPOV
Swihart, S. L. (1967). Hearing in butterflies (Nymphalidae: Heliconius, Ageronia). J. Insect Physiol. 13,469-476. Thurm, U. ( 1970). Untersuchungen zur funktionellen Organisation sensorischer Zellverbiinde. Verhandl. Deutsch. 2001. Ges. 64,79-88. Thurm, U. (1974). Basics of the generation of receptor potentials in epidermal mechanoreceptors of insects. I n “Mechanoreception” (Ed. J. Schwartzkopff), pp. 355-386. Westdeutscher Verlag, Opladen. Tuxen, S. L. (1966). “Insektenstimmen.” Springer-Verlag, Berlin, Heidelberg. Ulagaraj, S. M. (1976). Sound production in mole crickets (Orthoptera: Gryllotalpidae: Scapteriscus). Ann. Ent. SOC.Am. 69,299-306. Ulagaraj, .S. M. and Walker, T. J. (1975). Response of flying mole crickets to three parameters of synthetic songs broadcast outdoors. Nature, 253,530-532. Usherwood, P. N. R. and Newth, D. R., Eds. (1975). “Simple nervous systems.” Arnold, London. Usherwood, P. N. R., Runion, H. I. and Campbell, J. I. (1968). Structure and function of a chordotonal organ in the locust leg. J. exp. Biol. 48,305-323. Uvarov, B. (1966). “Grasshoppers and locusts.” Vol. I. Cambridge University Press. Vogel, R. (1923). fher ein tympanales Sinnesorgan, das mutmaaliche Hororgan der Singzikaden. Z . ges. Anat. I . Z . Anat. u. Entw. Gesch. 67, 190-231. Walker, T. J. (1957). Specificity in the response of female tree crickets (Orthoptera, Gryllidae, Oecanthinae) to calling songs of the males. Ann. Entom. SOC.Amer. 50, 626-636. Walker, T. J. (1964). Cryptic species among sound-producing Ensiferan Orthoptera (Gryllidae and Tettigoniidae). Quart. Rev. Biol. 39, 345-355. Walker, T. J. (1969a). Systematics and acoustic behavior of United States crickets of the genus Orocharis (Orthoptera, Gryllidae). Ann. Ent. Soc. Amer. 62,752-762. Walker, T. J. (1969b). Systematics and acoustic behavior of United States crickets of the genus Cyrtoxipha (Orthoptera, Gryllidae). Ann. Ent. Soc. Amer. 62,945-952. Walker, T. J. (1973). Systematics and acoustic behavior of United States and Carribean shorttailed crickets (Orthoptera: Gryllidae: Anurogryllus). Ann. Ent. Soc. Amer. 66, 12691277.
Walker, T. J. ( 1975). Stridulatory movements in eight species of Neoconocephales (Tettigoniidae). J. Insect Physiol. 21, 595-603. Walker, T. J. and Carlysle, T. C. (1975). Stridulatory file teeth in crickets: taxonomic and acoustic implications (Orthoptera, Gryllidae). Int. J. Insect Morphol. Embryol. 4, 15 1-158. Walker, T. J., Brandt, J. F. and Dew, D. (1972). Sound-synchronized, ultra high-speed photography: A method for studying stridulation in crickets and katydids (Orthoptera). Ann. Ent. Soc. Amer. 63,9 10-9 12. Walker, T. J. and Gurney, A. B. (1972). Systematics and acoustic behavior of Borinquenula, a new genus of Brachypterous coneheaded katydids endemic to Puerto Rico (Orthoptera, Tettigoniidae, Copiphorinae). Ann. Ent. SOC.Amer. 65,460-474. Walker, T. J., Whitesell, J. J. and Alexander, R. D. (1973). The robust conehead: two widespread sibling species (Orthoptera: Tettigoniidae: Neoconocephalus “Robustus”). Ohio J. Sci. 13,321-330. Weber, Th. (1972). Stabilisierung des Flugrhythmus durch Erfahrung bei Gryllus campestris L. Natunvissenschaflen, 59,366. Weber, Th. (1974). Elektrophysiologische Untersuchungen zur Entwicklung und zum Verlauf von Verhaltensweisen bei Gryllus campestris L. Diss. Univ. Koln. Weih, A. S. (1951). Untersuchungen iiber das Wechselsingen (Anaphonie) und iiber das angeborene Lautschema einiger Feldheuschreckenarten. 2.Tierpsychol. 8, 1-4 1. Weiss, P. (1941). Self-differentiation of the basic patterns of co-ordination. Comp. Psych. Monogr. 17, 1-96.
NE U R 0 ET H0 LOGY 0 F ACOUSTIC CO M MU N ICAT I0N
355
Wilson, D. M. (1961). The central nervous control offlight in a locust. J. exp. Biol. 38,471-490. Wilson, D. M. (1962). Bifunctional muscles in the thorax of grasshoppers. J. exp. Biol. 39,669677.
Wilson, D. M. (1964). The origin of the flight-motor command in grasshoppers. I n “Neural Theory and Modelling” (Ed. R. F. Reiss), 33 1-345. Stanford University Press. Wilson, D. M. (1968). The nervous control of insect flight and related behaviour. In “Advances in Insect Physiology”, Vol. 5,289-338. Academic Press, New York and London. Wilson, D. M. and Weis-Fogh, T. (1962). Patterned activity of co-ordinated motor units studied in flying locusts. J. exp. Biol. 39, 643-667. Wohlers, D. W. (1974). Synaptic acoustical responses related to oscillatory activity in the prothoracic ganglion of Achetu domesticus. Thesis, Master of Arts, Andrews University, USA. Worden, F. G. and Galambos, R. (1972). Auditory processing of biologically significance sounds. Neurosciences Res. Progrumm Bull. 10. Wuchenich, J. D. (1974). Synaptic coding of the temporal parameter of patterned acoustic stimuli by Achetu domesticus. Thesis, Master of Arts, Andrews University, USA. Yanagisawa, K., Hashimoto, T. and Katsuki, Y.(1967). Frequency discrimination in the central nerve cords of locusts. J. Insect Physiol. 13,635443. Young, D. (1972). Neuromuscular mechanisms of sound production in australian cicadas. J. comp. Physiol. 19, 343-362. Young, D. (1973). Fine structure of the sensory cilium of an insect auditory receptor. J . Neurocytol. 247-58. Young, D. (1975). Chordotonal organs associated with the sound producing apparatus of cicadas (Insecta, Homoptera). Z . Morphol. Tiere, 81, 1 1 1-135. Young, D. and Ball, E. (1974). Structure and development of the auditory system in the prothoracic leg of the cricket Teleogryllus commodus (Walker). 1. Adult structure. Z . Zelvorsch. 141,293-3 12 Zaretsky, M. D. (1972). Specificity of the calling song and short term changes in the phonotactic response by female crickets Scupsipedus marginatus (Gryllidae). J. comp. Physiol. 79, 153172.
Zhantiev, R. D. (1969). Ultrastructural organization of mechanoreceptor sensillae of insects. J. Gen. Biol. (Russian), 30,224-231. Zhantiev, R. D. (1971). Frequency characteristics of the tympanal organs of grasshoppers (Orthoptera, Tettigoniidae). Zool. J. (Russian), 50,507-5 14. Zhantiev, R. D. and Tshukanov, V. S. (1972). Frequency characteristics of tympanal organs of the cricket Gryllus birnuculutus Deg. (Orthoptera, Gryllidae). Vestn. Mosk. Univ. 2, 3-8. (in Russian). Zhantiev, R. D., Kalinkina, I. N. and Tshukanov, V. S. (1975). Directional characteristics of tympanal organs of the cricket GryNus bimuculutus Deg. (Orthoptera, Gryllidae). Reo. Entom. USSR, 54,249-257.
This Page Intentionally Left Blank
Subject Index
Abdominal pumping in hemolymph circulation, 179 A bsyrtus luteus, wingbeat frequency, temperature and, 139 Acanthomyops clauiger, alkanes in, function, 24 biological activity of alkanes and alkenes in, 22 Acheta domesticus, alkane biosynthesis in, 20 alkenes in, 2 flight metabolism, development, 198 hair sensilla, 283 methylalkanes in, 4, 9, 1 1 muscle activity, co-ordination, sound production and, 245 pulses per chirp, phonotactic response, 273 song patterns, evolution, 332 sound production, proprioceptive control, 258 stridulatory patterns, modification by external stimuli, 252 tympanal organs, 294 Achefa simplex, methylalkanes in, 1 1 Acoustic behaviour, female innate releasing mechanisms, 279-28 1 Acoustic communication, development, 3 16320 evolution, 329-338 genetics, 320-329 neuroethology, 229-355 Acoustic feedback, sound production and, 254 Acrididae, amplitude modulation, 3 14 auditory neurons, 306 contralateral co-ordination, sound production and, 249 female, phonotactic reaction, 275-277 innate releasing mechanisms, interspecific hybrids, 326-329
postembryonic development, 3 19 song patterns, genetics, 324-326 sound patterns, stridulatory movements and, 236 tympanal organs, 285,286-288,294,296 Acridid grasshoppers, song patterns, 239-241 Acridoidea, stridulatory mechanisms, 23 1 Acvrthosiphon pisum, trimethylalkanes in, 17 Adaptation in insect visual pigments, 57-60 Adipokinetic hormone release, 177, 178 Adult diapause, flight metabolism and, 206, 207 Aedes, flight muscle temperature, 196 Aedes aegypti, rhodopsin and metarhodopsin, 146 sex peptides, 9 1 Aedes Jauescens, oxygen consumption, flight and, 135 Aedes nearcficus, metabolic rate, mass, wingloading wingbeat frequency and, 140 Aeschna grandis, oxygen consumption, flight and, 135 Age, flight metabolism and, 210 Aggregation pheromone, alkanes, 24 Aging, flight muscles, 208 Aglais, dark regeneration, 52 Aglais urticae, rhodopsin and metarhodopsin, 46 Agreement song, 268 ala-arg-thr-bradykinin, 1 18 PAlanyltyrosine. See Sarcophagine Alarm-defence substances, alkanes and, 24 Algae. methylalkanes in, 7, 13 polyolefins in, 3 Alkanes, 2 , 3 biosynthesis, 17 dimethyl, 13-16 internally branched methyl, analysis, 4-6 methyl, 6- 13
3 57
358
SUBJECT INDEX
Alkanes (conr.) 2-methyl,3,4 3-methy1, 3,4 trimethyl, 16, 17 Alkenes, 2, 3 Allinum porrum, alkanes in, function, 25 Allonemobius fasciatus, alkane biosynthesis, 18 2-methylalkanes in, 4 Amplitude modulation, innate releasing mechanisms and, 268-277 pattern, 3 10-314 Anabrus simplex, dimethylalkanes in, 13, 14, 15, 16
methylalkanes in, 9 3-methylalkanes in, 4 Ants, stridulatory organs, 320 Anrheraea pernyi, flight metabolism, development and senescence, 201 oxygen consumption, flight and, 135 Anfherae po!vphemus, flight metabolism, development and senescence, 201 flight muscles, 157 motor patterns, development, 202, 203 Anrhonomus grandis, alkenes in, 2 methylalkanes in, 7, 11, 12 Apamin. from bee venom, 113 pharmacological activity, 114, 115 structure, 113, 114 function relationships, 1 14 synthesis, 114 Aphids, flight metabolism, polymorphism and, 207 Apis, flight speed, metabolic rate and, 145 phylogenetic relationships, I 1 1, I12 Apis cerana, phylogenetic relationships, 1 1 1 Apis dorsata, phylogenetic relationships, 1 1 1 ApisJlorea. phylogenetic relationships, 1 1 1 Apis mellifera, alkenes in. 2 fibrillar muscles, 203 flight fuel, 165 flight speed, metabolic rate and, 145 hyperglycaemic hormone, 101 isolation of visual pigments from, 39 metabolic rate, mass, wing-loading wingbeat frequency and, 140 methylalkanes in, 8 oxygen consumption during flight, 142 oxygen consumption, flight and, 136 phylogenetic relationships, 11 1
power output, control mechanisms. 153 pre-flight warm-up, 187, 188 rhodopsin and metarhodopsin, 46 substrate-cycling, 195 wingbeat frequency temperature and, I39 Apis mellifera carnica, melettin in, 106 Archilochus colubris, metabolic rate and, 146 Ascending auditory neurons, 302-3 14 Ascalaphus, metarhodopsin, 49 opsin, molecular weight, 47 retinous, phospholipids, 6 1 rhodopsin, 62 chromophore, 48 visual sensitivity and, 58 ultraviolet sensitive rhodopsin, 54 Ascalaphus macaronius, rhodopsin and merarhodopsin, 46 ultraviolet sensitive visual pigment, 50 visual pigment, 44,45 Aspergillus cvanea, methylalkanes in, 7 Aspergillus nidulans, methylalkanes in, 7 Aspergillus variabilis, methylalkanes in, 7 ATP in flight muscle, 161 Atta cephalotes isthmicola, trimethylalkanes in, 16 Atta colombica, trimethylalkanes in, 16 Atta sexdens, trimethylalkanes in, 16 Auditory mechanism, evolution, 338 Auditory neurons, information processing by, 296-3 16 Bacillus subrilis, alkane biosynthesis, 18 Bacteria, alkenes in, 3 dimethylalkenes in, 4 Beeswax, methylalkanes in, 1 1 Bee venom. peptides, 106-116 y-BHC, glutathione S-aryltransterase conjugation with, 83-85 Biosynthesis, alkanes, 17-21 Birds, metabolic rate during flight, 137 Bitumen, trimethylalkanes in, 17 Bituminous shales, dimethylalkanes in, 16 Blaberus craniifer, alkanes in, function, 24 biological activity of alkanes and alkenes in, 23 Blaberus discoidalis, hyperglycaemic hormone, 101 Blarra germanica, alkanes in, function, 24 Blarra orientalis, methylalkanes in, 9, 1 1 3-methylalkanes in, 4
SUBJECT INDEX
Bleaching of rhodopsin, 37 Bodymass, wing-loading and, 139- 143 Body weight, Hyles lineafa, oxygen consumption and, 14 1 metabolic rate and, 140, 141 Bombilius spp., metabolic rate, 146 Bombus spp., Right fuel, 165 flight muscle, phosphofructokinase in, 172 trehalase in, 164 metabolic rate, 146 mass, wing-loading wingbeat frequency and, 140 oxygen consumption during flight, 136, 142 thermogenesis, 192 Bombus agrorum, wingbeat frequency, temperature and, 139 Bombus edwardsii, oxygen consumption body weight and, 143 Bombus horforum, hexokinase activity, 192 Bombus fernarius, substrate cycling, 195 Bombus vosnesenskii, oxygen consumption, Right and, 136 pre-flight warm-up, 187, 188 Bombyx mori, dimethylalkanes in, 16 methylalkanes in, 12 2-methylalkanes in, 4 3-methylalkanes in, 4 Bombyx sonorus, pre-Right warm-up, 185 Boophilus, glutathione S-aryltransferase in, 8 1 Bradykinin, 1 16, 1 17 structures, 116 Brain auditory neurons, 3 14-316 Brassica oleracea var. gemmifera, alkanes in, function, 25 Breathing, sound production and, 265
Caelfera, stridulatory mechanism, 23 1 Calling song, crickets, 237 Calliphora, dark regeneration, 52 eyes, isolation of intact rhabdomeres from, 39 fibrillar muscles, 203, 204 hemolymph trehalose, 177 power output, neural control, 153 pre-Right warm-up, 189 rhodopsin orientation in, 6 1 visual sensitivity and, 58 visual pigments, 55, 56 microspectrophotometry, 44
359
Calliphora erythrocephala, corpora cardiaca hormones, 175 fibrillar muscles, 205 hemolymph circulation, 179 hormone release, neural control, 176 hyperglycaemic hormone, 101. 104, 174 peptide pools, metabolic aspects, 89 rhodopsin and, metarhodopsin. 46 visual pigment, 45 Callosobruchus maculafus.flight fuel, 165 Camponofus intrepidus, methylalkanes in, 6, 8 Canfharis sp., wingbeat frequency, temperature and, 139 Carausius morosus, hyperglycaemic hormone, 101 Carbohydrates in Right muscle, 161-164 Cardiochiles nigriceps. alkanes in, function, 24 Carnitine palmitoyl-transferase in Right muscle metabolism, 173 Celerin in Celerio euphorbiae, 74 Celerio euphorbiae, celerin in, 74 Celereo lineafa,pre-flight warm-up, 185 Central nervous system, sound production and, 260-267 Cepaea nemoralis, methylalkanes in, 4 trimethylalkanes in, 17 Chirping, Grvllidae amplitude modulation, 269 Chordate, methylalkanes in, 12 Chorfhippus spp., female, innate releasing mechanism, 276 nymphs, sound production, 3 17 song pattern, evolution, 336 Chorfhippus biguftulus, electromyograms, 243 female, innate releasing mechanism, 276, 277 phonotatic reaction, 275 muscle activity, co-ordination, sound production and, 245 nymphs, sound production, 3 17 song patterns. 24 1 evolution, 334, 336 genetics, 324 sound production, proprioceptive control, 257,258,259,260 stridulation development, 3 18 Chorfhippus brunneus, female, innate releasing mechanism, 277
3 60 Chorfhippus brunneus (cont.) song pattern, evolution, 334,336 genetics, 324 stridulation development, 3 18 Chorrhippus longicornis, female, innate releasing mechanism, 277 Chorfhippus mollis, female, innate releasing mechanism, 277 motor scores, sound production and, 247 muscle activity, co-ordination, sound production and, 245 nymphs, sound production, 3 17 song patterns, 241 evolution, 334, 336 genetics, 324 sound production, proprioceptive control, 257 stridulation development, 3 I8 Chorthippus monfanus, female, innate releasing mechanism, 277 song pattern, evolution, 334 Chorthippus vagans, sound patterns, 240 Chortoicetes ferminfera, flight metabolism, development, 199,200 Chromophores, energy transfer from opsin to, 51 of insect visual pigments, 47-5 1 orientation in rhabdomeres, 6 1 Chrysopa oculafa,alkanes in, function, 24 alkenes in, 3 Cicadidae, amplitude modulation, 3 14 frequency of sounds, 235 sound patterns, stridulatory movements and, 236 sound reception, sensory mechanisms, 28 1 stridulatory organs, 230 tympanal organs, 285,286,294 Cockroach. See Peripianeta spp. Coleoptera, biological activity of alkanes and alkenes in, 22 dimethylalkanes in, 14 methylalkanes in, 7, 1 1 Colour vision, insects, 53, 54 Conocephalinae, non-resonant sound emissions, 233 Copulation, alkanes and, 24 phonotaxis and, 28 1 Corixidae, flight metabolism, development and, 207 Corpus cardiacum, physioldgically active peptides from, 9 6 , 9 7
SUBJECT INDEX
Corynebacterium diphtheriae, alkane biosynthesis in, 17 Costelytra zealandica, glutathione S-aryltransferase in, 8 I Cost of transport, metabolic rate and, 146 Courtship song, crickets, 237 Cows. trimethylalkanes in, 17 Crickets, song patterns, 237 Crude oil, dimethylalkanes in, 16 Crustacea, surfaces waxes, trimethylalkanes in, 17 Cuiicidae spp., wingbeat frequency, temperature and, 139 Curculio caryae, alkenes in, 3 methylalkanes in, 7, 1 1 Cycloalkanes, 3 Cyclorraphous flies, photopigment system, 51 visual pigments, 55, 56 Danaus plexippus, flight fuel, 165 wingbeat frequency, temperature and, 183 Dark regeneration of insect visual systems, 52,58 DDT-dehydrochlorinase, glutathione as specific co-factor, 80 Deafferentation, sound production and, 258 Decticinae, non-resonant sound emissions, 233 Decticus, Crisfa acousfica, 288 Decticus verrucivorus, auditory neurons, 306, 308 tympanal organ, 289,290,29 1 Deilephila, phospholipids, 6 1 rhodopsin, visual sensitivity and, 58 Deilephila elpenor, colour vision, 53 oxygen consumption, flight and, 135 rhodopsin and metarhodopsin, 46 visual pigment, 45 Deilephila nerii, wingbeat frequency, temperature and, 183 Dendrocranus pseudotsugae, flight lipids, 164 flight muscle development, hormonal control, 209 Detoxication mechanisms, glutathione in, 8088 Dictyoptera, oxygen consumption, flight and, 135 Digitonin for extraction of insect rhodopsins, 38 Dipeptides, 70-75
SUBJECT INDEX
Diptera, biological activity of alkanes and alkenes in, 22 dimethylalkanes in, 14 Right fuels, mobilization, 170 flight speed, metabolic rate and, 145 methylalkanes in, 8, 1 1 oxygen consumption, Right and, 135 proline as flight fuel, 165 sarcophagine in, 72, 73 sound patterns, stridulatory movements and, 236 stridulatory organs, 230 Dopamine 3-0-sulphate in Periplaneta americana, 74 Drepanoxiphus modestus, resonant sound emissions, 232,233 Drosophila, dark regeneration, 52 dipeptides in, 74 eye, fast electrical response from, 40 fibrillar muscles, 203 flight muscle, a-glycerophosphate cycle, 163 temperature, 196 metabolic rate, body weight and, 141 during flight, 136 metabolic rate, mass, wing-loading wingbeat frequency and, 140 mutants with impaired transduction, 59 peptide pools, metabolic aspects, 89 power output, neural control, 153 sex peptides from, 91-94 visual pigments, 56 Drosophila funebris, oxygen consumption, flight and, 135 peptides in, 70 sex peptides from, 93 wingbeat frequency temperature and, 139 Drosophila gibberosa, oxygen consumption, flight and, 135 Drosophila hydei, oxygen consumption, Right and, 135 Drosophila melanogaster, Right muscles, maturation, 208 metabolic rate, mass, wing-loading, wingbeat frequency and, 140 oxygen consumption, during flight, 135, 142 rhodopsin and metarhodopsin, 46 sex peptides, 92 wingbeat frequency, temperature and, 139 Drosophila repleta, oxygen consumption, flight and, 135 wingbeat frequency temperature and, 139
36 1 Drosophila virilis, power output, neural control, 151 Dvsdercus, Right metabolism, development and. 207 D-vsdercus intermedius, flight muscle development, hormonal control, 209
Ecdysone, Right muscle development and, 209 Ecology, metabolic rate and, 146, 147 Electro-myography, sound production and, 24 1,242 techniques, 242,243 Emerogryllinae, stridulatory mechanisms, 232 Ensifera, contralateral co-ordination, sound production and, 249 song patterns, evolution, 332 stridulatory mechanisms, 23 1 Enzymes in Right muscle metabolism, 17 1 Ephippigeridae, non-resonant sound emissions, 233 Erisralis tenax, metabolic rate, mass, wingloading wingbeat frequency and, 140 oxygen consumption, Right and, 136 Euchloron magaera, wingbeat frequency, temperature and, 183 Euconocephalus nasutus, motor coordination, sound production and, 249 Evolution, acoustic communication, 329-338 Factor B, 98,99 Factor C, 97,98 Fannia canicularis, alkanes in, function, 2 1 alkenes in, 2 biological activity of alkanes and alkenes in, 22 oxygen consumption, Right and, 136 Fats as Right fuel, 165 Fatty acids as Right fuels, mobilization, 170 Fibrillar muscles, in holometabolous insects, 203-206 Fish, trimethylalkanes in, 17 Flight, metabolic rate in 134-147 Flight fuel, 164, 165 mobilization, 169-1 71 Flight metabolism, 133-228 age and, 2 10 control, 156-180 development and senescence, 197-2 10 Flight motor, system, development, 198 temperature and, 181-184
362
Flight muscles, 156, 157 biochemical processes, 16 1- 169 metabolism, 17 I - 173 temperature, 195-197 Flight speed, metabolic rate and, 143-146 substrate availability and, 179. 180 Formica po!vcfena, methylalkanes in, 6. 8 Formica nigricans, methylalkanes in, 6. 8 Formica rufa. methylalkanes in, 6, 8 Frequency, innate releasing mechanisms in. 277-279 orthopteran sound, 235, 236 Frequency spectrum, auditory neurons and, 308,309 Fructose diphosphatase. in Bombus flight muscle, 192 in thermogenesis, 19 I Formational morphology, sound production and, 248,249
Galleria, dark regeneration, 52 Galleria mellonella. rhodopsin and metarhodopsin, 46 Gampsocleis buergeri, prothoracic ganglion, 288 sound source localization, 309 Genetics, acoustic communication, 320-309 Gerris lacustris, flight metabolism, development and. 207 Glossina,flight fuel, mobilization, 170 flight muscles, maturation, 208 proline as flight fuel, 167 wingbeat frequency, substrate availability and, 179 Glossha morsitans, alkanes in, function, 24 dimethylalkanes in, 13, 14 flight fuel, 165 methylalkanes in, 1 1 oxygen consumption flight and, 136 sex peptides. 94 trimethylalkanes in, 16 Glossina pallidipes, flight, amino acid concentration, 167 sex peptides, 94 Glucagon, porcine, amino acid sequence, 102 Glucose as flight fuel, 164 mobilization, 169 y-Glutamyl-cysteine synthetase in Musca dornestica, 79, 80 y-Glutamyl cycle, 75, 76
SUBJECT INDEX
y-Glutamyl cyclotransferase in Musca dornestica, 78, 79 y-Glytamyl-phenylalanine, 73, 74 metabolic fate and function, 73, 74 y-Gutamyl transpeptidase, 75 in Musca domestica, 77, 78 Glutathione, 75-88 in detoxication mechanisms, 80-88 Glutathione synthetase in Musca dornestica, 79,80 Glutathione S-alkyltransferase in Musca dornestica, 87, 88 Glutathiom S-aryltransferase in detoxication mechanisms, 8 1 Glutathione S-transferases in detoxication mechanisms, 8 1 Glycerol kinase in locust muscle, metabolism, 172 Glycerol 3-phosphate in flight muscle, 163 Glycerol 3-phosphate dehydrogenase in flight muscle, 163 a-Glycerophosphate cycle in flight muscle, 163 a-Glycerophosphate dehydrogenase in flight muscle, 208 Glycolytic pathway in flight muscle, 163 Glycogen, as flight fuel, 164 mobilization, 170 in flight muscles, hormonal control, 174 Glycogenolysis, hyperglycaemic hormone and, 105 Gomphocerinae, non-resonant sound emissions, 233 song patterns, 239 evolution, 333 stridulatory mechanisms, 232 Gomphocerippus rufus, contralateral coordination, sound production and, 250, 25 1 muscle activity, co-ordination, sound production and, 245 song pattern, evolution, 334 sound production, central nervous system and, 261,262 proprioceptive control, 255,257 Gomphocerus rufus, female, innate releasing mechanism, 28 1 Gomphocerus sibiricus, female, phonotaxis, 279 song pattern, evolution, 334 sound production, motor co-ordination, 246
SUBJECT INDEX
Graphidostreptus tumuliporus, alkenes in, 3 cycloalkanes in, 3 methylalkanes in, 4,6 biosynthesis, 20 trimethylalkanes in, 17 Gtyllidae, amplitude modulation, innate releasing mechanism and, 268 auditory mechanism, evolution, 338 auditory neurons, 306 innate releasing mechanisms, interspecific hybrids, 326 song patterns, genetics, 321 sound patterns, stridulatory movements and, 236 tympanal organs, 285,29 1-296 Gryllodinus kerkennensis, amplitude modulation, innate releasing mechanism and, 269 calling songs, 3 1 1 female phonotactic response, 270 Gryllodinus odicus, song patterns, 237,238 Grvlloidea, stridulatory mechanisms, 23 1 Gryllotalpa gryllotalpa, resonant sound emissions, 233 Gryllotalpa vinea, resonant sound emissions, 233 Gryllus, sound production, central nervous system and, 263 Gryllus bimaculatus, amplitude modulation, innate releasing mechanism and, 269 270 amplitude modulation pattern, 3 10 auditory mechanism, evolution, 338 auditory neurons, 3 15 female, phonotaxis, 278 phonotactic response, 271,272 frequency, innate releasing mechanism and, 278 hair sensilla, 282, 284 HF-unit, 3 13 LF-unit, 3 14 muscle activity, co-ordination, sound production and, 245 phonotaxis, innate releasing mechanism, 280 pulses per chirp, phonotactic response in, 272,273 segmental auditory neurons, 302 song patterns, 238 evolution, 332 sound patterns, stridulatory movements and neuromuscular activity, 246
363 threshold curves, 3 1 1 tympanal nerve projections, 299 tympanal organs, 292,293 Grvllus campestris. amplitude modulation, innate releasing mechanism and, 269 amplitude modulation pattern, 3 10 auditory mechanism, evolution, 338 calling song, 268 contralateral co-ordination, sound production and, 250 female, phonotactic response, 271, 272 flight metabolism. development, 200 flight muscles, maturation, 208 hair sensilla, 283 muscle activity, co-ordination, sound production and. 245 non-resonant sound emissions, 234 nymphs, sound production, 3 17 phonotaxis, innate releasing mechanism, 280 pulses per chirp, phonotactic response, 273 resonant sound emissions, 232 song patterns, 237, 238 evolution, 332, 333 sound production, central nervous system a d , 264,265,267 sound production, proprioceptive control, 256.257 stridulatory patterns, modification by external stimuli, 252 tympanal organs, 292,293,296 Grvllus pennsvlvanicus, alkane biosynthesis, 18 2-methylalkanes in, 4
Hair sensilla in sound research, 282-285 Hall-generators, stridulatory movements and, 236 Hearing, postembryonic development, 3 19, 320 Heart accelerating peptides, 97- 101 amino acid compositions, 98.99 isolation, 98,99 origin, 99, 100 physiological function, 100, 101 release, 99, 100 site of synthesis, 99, 100 Heliocorpus sp., metabolic rate, bodyweight and, 141 Heliothis nigricans, biological activity of alkanes and alkenes in, 23
364
Heliofhis virescens, alkanes in, function, 24 biological activity of alkanes and alkenes in, 23 flight metabolism, development and senescence, 203 methylalkanes in, 9, 1 1 , 12 Heliofhis zea. biological activity of alkanes and alkenes in, 23 methylalkanes in, 1 1 , 12 Hemaris spp., metabolic rate, 146 Hemimetabolous insects, flight metabolism, development, 198-200 Hemolymph circulation, 178, 179 Hemolymph lipid, hormonal control, 175, 176 Hemolymph sugar, hormonal control, 173175 Hexokinase, in Bombus hortorum, 192 in flight muscle metabolism, 172, 173 High-speed photography, stridulatory movements and, 236 Holomelina opella nigricans, alkanes in, function, 24 2-methylalkanes in, 4 Holometabolous insects, fibrillar muscles, 203-206 Homorocarvphvs nifidulus,flight fuel, 165 resonant sound emissions, 232, 233 Homorocorvphus nifidulus vicinus, tympana1 organs, 296 Homorocoryphus subvitlatus, flight fuel, 165 Honeybees. See Apis melliphora Hormonal control, flight muscle development, 209,210 Hormonal control mechanisms, flight muscle metabolism, 173 Hormones, release, neural control, 176-178 Hornets, kinins from, 116-1 18 House flies. See Musca domesficus Hovering flight, metabolic rate, 146 Humans, trimethylalkanes in, I7 Hummingbird, metabolic rate during flight, 136, 137 Hvalophora cecropia. hyperglycaemic hormone. 104 phosphorylase activity. 105 Hydrocarbons, long-chain methyl branched, 1-33 Hvles euphorbia, oxygen consumption, flight and, 135 Hvles lineata, oxygen consumption, bodyweight and, 14 I
SUBJECT INDEX
during flight, 135, 142 power output, neural control, 155 Hymenoptera, biological activity of alkanes and alkenes in, 22 dimethylalkanes in, 14 flight fuels, mobilization, 170 methylalkanes in, 6, 18, 1 1 oxygen consumption, flight and, 136 Hyperglycaemic hormone, 101-105, 174 neural control, I77 physiological activity, 103, 104 physiological function, 105 release in flight, I77 site of synthesis, 103 structure, 101, 102 Innate releasing mechanisms, 268-281, 3 19, 3 20 acoustic behaviour and, 279-281 song patterns and, evolution, 329-332 song specific, interspecific hybrids, 326-329 Insecticide resistance, glutathione S-alkyltransferase activity and, 88 Insulation, flight muscle temperature and, 197 Intersegmental networks, sound production and, 260,265-267 Iridamvrmef humilis, alkenes in, 2 methylalkanes in, 6 , 8 Iphiclides podalirius, thoracic temperature, stabilization during flight, 190, 191 Ips confus, flight metabolism, development and. 207 Juglans regia, methylalkanes in, 7 Juvenile hormone, flight muscle development and, 209 Kallidin, 1 16, 1 17 Kinins from wasps and hornets, 116-1 18 Lactobacillus arabinosus, methylalkane biosynthesis, 19 Larva, stimulation of sound production, 3 16318 Lasius alienus, alkanes in, function, 24 biological activity of alkanes and alkenes in, 22 Lepidoptera, age, flight metabolism and, 210 amplitude modulation, 3 14 auditory neurons, 3 16
SUBJECT INDEX
biological activity of alkanes and alkenes in, 23 dimethylalkanes in, 14 dipeptides in, 75 flight fuel, 164 mobilization, 170 flight metabolism, development and senescence, 200-203 methylalkanes in, 9 oxygen consumption, flight and, 135 power output, neural control, 150 pre-flight warm-up, 185 sound reception, sensory mechanisms, 28 1 stridulatory organs, 230 tympanal organs, 285,286,294,296 Leptinofarsu decemlineatu, flight fuel, 165 flight metabolism, maturation and, 206 flight muscle development, hormonal control, 209 proline as flight fuel, 165 Leucophaea maderae, corpora cardiaca extracts, 174 hyperglycaemic hormone, 101 methylalkanes in, 9, 1 1 3-methylalkanes in, 4 proctolin in, 95 Lichen, methylalkanes in, 7 Lipids, as flight fuel, 164 metabolism, 169 mobilization, 170 in flight muscles, hormonal control, 174 Lisfrocuelinae,non-resonant sound emissions, 233 Locusta, tympanal organs, 286 receptors, 298 Locusta migraforia, auditory neurons, 304, 307 blood lipids, 175 corpora cardiaca, extracts, 174 flight fuel. 164 flight metabolism, development, 200 flight muscle metabolism, 173 flight speed, substrate availability, 180 heart-accelerating peptides, 97,98 hyperglycaemic hormone, 101, 104 innate releasing mechanism, postembryonic development, 320 neurogenic rhythms, 148 power output, neural control, 150 tympanal organs, 287,289,294 ventral cord neurons, 297
365
Locusta migratoria migrutorioides, Corpus cardiacum, peptides from, 96 Locusts, cerebral neurosecretory cells, 176 lipids, mobilization, 177 neurogenic rhythms, 148 Long-horn beetles, stridulatory organs, 230 Luciliu sericata, oxygen consumption, flight and, 135 Lumirhodopsin, 49 Lycoriellu mali, alkanes in, function, 2 1 biological activity of alkanes and alkenes in, 22 lysylbradykinin. See Kallidin Mammals, metabolic rate, 137 Manduca, rhodopsin, visual sensitivity and, 58 Manduca sextu, age flight metabolism and, 210 alkane biosynthesis in, 2 1 alkanes in, function, 25 colour vision, 53 dimethylalkanes in, 13, 14, 16 flight fuels, mobilization, 170 flight metabolism, development and senescence, 20 1,203 flight motor, temperature and, 18 1 flight muscles, 157, 158, 159 temperature, 196 hemolymph circulation, 179 hypoglycaemic factor, 101, 174 metabolic rate, body weight and, 141 during flight, temperature and, 138 mass, wing-loading wingbeat frequency and, 140 methylalkanes in, 9, 11, 12 oxygen consumption during flight, 135, 142 power output, neural control, 151, 155 pre-flight warm-up, 186, 187 rhodopsin and metarhodopsin, 46 thoracic temperature, stabilization during flight, 191 trimethylalkanes in, 16 wingbeat frequency, temperature and, 183 Marrubium vulgure, 3-methylalkanes in, 4 Mass spectra, methylalkanes, 5 Mast cell degranulating peptide from bee venom, 112, 113 Maturation, flight muscles, use and disuse, 208,209
366
SUBJECT INDEX
Melanogryllus desertus, amplitude modulation, innate releasing mechanism and, 273
female, innate releasing mechanism, acoustic behaviour and, 279 phonotactic reaction, 274 frequency, innate releasing mechanism and, 278
song patterns, 238 sound production, proprioceptive control, 258
Melanoplus packardii, dimethylalkanes in, 13, 14, 15, 16 methylalkanes in, 9, 11, 12 Melanoplus sanguinipes, dimethylalkanes in, 13, 14, 15, 16
methylalkanes in, 9, 1 1 Melettin. 106 biosynthesis. 108-1 11 during bee maturation, 110 structure-activity relationships, 106 Melettin F, 1 15 Melolontha melolontha, proline as flight fuel, 165. 166
Metabolic rate, bodyweight and, 140 ecology and, 146, 147 flight muscle temperature and, 180-197 in flight, 134-147 Metarhodopsin, 40-47 as pH indicators, 49 insect visual sensitivity and. 58 thermostability, 40,4 1 transduction in insect visual pigments and, 59 1 I -cis Metarhodopsin, 50
Meteorites, methylalkanes in, 3,6. 7, 13 Methionine. methylalkane biosynthesis and, 19 Methionyllysylbradykinin, 116, 117 Metopsilus procellus, flight muscles, oxygen supply. 160 oxygen consumption, flight and, 135 Metrioptera sphagnorum. female, phonotaxis, 278 frequency of sound, 235 Mevalonic acid in methylalkane biosynthesis. 20 Microplitus croceipes, alkanes in, function, 24 methylalkanes in. 12
Microspectrophotometry, Calliphora visual pigment, 44 for measurement of insect visual pigments, 39
Mimas tiliae, oxygen consumption, flight and, 135
Minimine in bee venom, 1 15 Mobilization, flight fuels, 169- 17 1 Modulation, impulse-rate, orthopteran sounds, 235,236 Molecular sieves in analysis of internally branched methylalkanes, 4 Mosquitoes, carotenoid deficiency in, 48 dark regeneration, 52 larval ocellus. metarhodopsiri thermostability, 41 Motoneural activity, sound production and, 241,242
Motor activity, co-ordination, sound production and, 243 Musca autumnalis, alkanes in, function, 21 alkenes in, 3 biological activity of alkanes and alkenes in, 22
Musca domestica, alkanes in function, 21 alkenes in, 2 biological activity of alkanes and alkenes in, 22
cycloalkanes in, 3 .IL dipeptides. 7 1 fibrillar muscles, 205 flight muscles, maturation, 208 y-glutamyl cycle enzymes in, 77-80 y-glutamyl phenylalanine in, 73 rhodopsin and metarhodopsin, 46 sex peptides, 9 1 visual threshold, 48 Musca uomitoria. oxygen consumption, flight and, 135 Muscles, motor co-ordination, sound production and. 244 Mvcobacterium smegmatis, methylalkane biosynthesis in, 19 Myogenic rhythms, 151-154 Mvrmecia gulosa, alkenes in, 2 methylalkanes in, 1 1 Mvrmeleoteltix maculatus, sound patterns, 240
Neoconocephalus robustus. flight motor, temperature and, 182
SUBJECT INDEX
367
-0ncopeltus fascialus, flight metabolism, development, 199 flight muscles development. hormonal control, 209 Opsin, energy transfer to chromophores, 5 1 Orconerles limosus. flight motor, temperature and, 182 Organophosphorus insecticides, conjugation with glutathione S-aryltransferase. 85-87 Orlhoptera, age, flight metabolism and, 210 biological activity of alkanes and alkenes in, 23 dimethylalkanes in, 14 female. phonoresponse, 268 methylalkanes in, 9, 1 1 oxygen consumption, flight and. 135 song patterns, evolution, 332 sounds, physical parameters, 232-236 reception, sensory mechanisms, 281 sound source localization, 309 stridulatory organs, 230 tympanal nerve fibres, 296 tympanal organs, 294 5-Oxoprolinase in Musca domesrica. 79 Ocypode ceratophthalma, flight motor, temOxygen consumption. flight and, 135, 136 perature and, 182 Hyles lineata, body weight and, 141 Odonata, oxygen consumption, flight and, 135 in flight. flight muscle differences and, 142 Oecanlhus niueus, female phonotactic resin flight muscle metabolism, 171 ponse, 27 1 Oecanthus pellucens, female, phonotactic res- Oxygen supply to flight muscles, 157-161 ponse, 272 Paraffin wax. cycloalkanes in, 3 resonant sound emissions, 232 methylalkanes in, 10, 13 song patterns, 237 Peak I, 97,98 Oecanthus turanicus, calling songs, 3 11 Peak 2 . 9 8 . 9 9 Oedipodinae, song patterns, evolution, 333 Peptidases in blowfly, 90 stridulatory mechanisms, 232 Peptidases P,, 97.98 Oil shale. trimethylalkanes in, 17 Peptide P,, 98. 99 Olive oil, trimethylalkanes in, 17 Omocestus viridulus, contralateral co- Peptides. pools, metabolic aspects, 88-9 1 structure and function, 69- I32 ordination, sound production and, 25 1 feriplaneta spp., flight metabolism, developfemale, phonotaxis, 279 frequency of sounds, 235 ment, 200 motor co-ordination, sound production and, proctolin in, 70 wingbeat frequency, temperature and, 139 248 feriplaneta americana, alkanes in. biosynnon-resonant sound emission, 234 thesis, 2 1 song patterns, 239 function, 24 sound patterns, 240 sound production, motor co-ordination, 246 alkenes in. 2. 3 blood lipids. 176 proprioceptive control, 257,258 Oncopellus. flight muscles development and, Corpus cardiacum, pgptides from, 96 dopamine 3-0-sulphate in. 74 207
motor co-ordination, sound production and, 249 muscle activity, co-ordination, sound production and, 245 resonant sound emissions, 232, 233 song patterns, 237 Neotenin. See Juvenile hormone Neural control, hormone release, 176-178 power output in flight, 147-156 Neuroethology, acoustic communication, 229-335 Neurogenic rhythms, 147-15 1 Neurohormone C, 98,99 Neurohormone D, 97,98 Neurospora crassa, methylalkane biosynthesis in, 19 Neurotransmitter in insect visceral muscles, 94-96 Noctuidmoths, tympanal nerve fibres, 297 Nonesuch seep oil, cycloalkanes in, 3 Nostoc sp., methylalkanes in, 7 Nostoc muscorum, methylalkanes in, 7
368
Periplanta americana (cont.) flight fuel, 164 mobilization, 169 glycogenolysis, 105 heart-accelerating peptides, 97,98 hyperglycaemic hormone, 101, 104 3-methylalkanes in, 4 oxygen consumption, flight and, 135 proctolin in, 94 rhodopsin and metarhodopsin, 46 Periplaneta australasiae, alkenes in, 2 methylalkanes in, 9, 1 1 3-methylalkanes in, 4 Periplaneta brunnea, alkenes in, 2 methylalkanes in, 9, 11 3-methylalkanes in, 4 Periplanetafuliginosa, alkenes in, 2 methylalkanes in, 9, 11 biosynthesis. 19 3-methylalkanes in, 4 Periplanetajaponica, alkenes in, 2 dimethylalkanes in, 14, 16 methylalkanes in, 7,9, 11 3-methylalkanes in, 4 Petroleum, methylalkanes in, 4, 10 Phaneropterinae, amplitude modulation, innate releasing mechanism and, 273,274 Philosamia Cynthia, flight metabolism, development 2nd senescence, 203 Phonoresponse, female Orthopterans, 268 Phonotaxis, 268 Phormia, flight muscles, trehalase in, 164 hair sensilla, 282 hyperglycaemic hormone, neural control, 177 power output neural control, 152 proline as flight fuel, 167 Phormia regina, corpora cardiaca, hyperglycaemic activity, 173 dipeptides in, 70 fibrillar mucles, 205 flight fuel, 164, 165 mobilization, 169 flight muscle, ATP, I6 1 hyperglycaemic hormone, 101 power output, neural control, 15 1 wingbeat frequency, trehalose and, 179 Phosphofructokinase in Bombus flight muscle, 172 in flight muscle, 161
SUBJECT INDEX
in thermogenesis, 19 1 Phospholipase A in bee venom, 112 Photochemistry of insect visual pigments, 4751
Photoreceptor membranes, insect, 60-62 Pieris brassicae, dipeptides in, 75 Platycleis afJinis, stridulatory patterns, modification by external stimuli, 252254 Platycleis intermedia, frequency of sounds, 235 stridulatory patterns, modification by external stimuli, 252-254 Pogonomyrmex barbatus, dimethylalkanes in, 14 methylalkanes in, 8 Pogonomyrmex barbatus rugosus, methylalkanes in, 6 Pogonomyrmex rugosus, dimethylalkanes in, 14.16 methylalkanes in, 8 Pogonomyrmex rugosus var. fuscatus, methylalkanes in, 6 Polarized light, insect rhodopsins and, 60 Polistes rothneyi iwatai, kinins from, 117 Polisteskinin, 11 7, 118 Polymorphism, flight metabolism and, 206, 207 Popillia japonica, dimethylalkanes in, 16 methylalkanes in, 12 proline as flight fuel, 167 Porphyropsins, 54 Power output in flight, neural control, 147156 Precambrian sediment, trimethylalkanes in, 17 Pre-flight warm-up, 184- 190 Procamine in bee venom, 1 15 Proctolin, 70 as neurotransmitter, 94-96 Prodenia eridania, flight lipids, 164 lipids as flight fuel, 169 Proline, as flight fuel, 165 metabolism, 165-168, 170 Promelettin, 108, 109 biosynthesis during bee maturation, 110 Propionic acid, methyl branching in biosynthesis of alkanes and, 19 Proprioceptive control, sound production, 254-260 Proteins as flight fuel, 164
SUBJECT INDEX
Proroparce sexta, hyperglycaemic hormone, 104 PS- 1, from Drosophila funebris, 93 structure, 93 PS-2, from Drosophilafunebris, 93 Pseudoclaris postica, wingbeat frequency, temperature and, 183 Psithyrus spp., substrate-cycling, 195 Psirhvrus ashtoni, substrate-cycling, 195 Pferonarcys calfornica, alkenes in, 2 methylalkanes in, 10, 12 song patterns, 237 evolution, 332 Pulse rate, innate releasing mechanism and, 268 Pulses, terminology, 233
Rats, trimethylalkanes in, 17 Reflectance photometry for measurement of insect visual pigments, 40 Regeneration, flight metabolism and, 206,207 in insect visual systems, 51-53 Retinal, 36 in insect visual pigment, 47 Retinaldehyde. See Retinal Rhabdom, 38 Rhabdomere, 38 chromophore orientation in, 6 1 Rhodopsin, 40-47 in Ascalaphus, 62 mobility, 6 1 vertebrate, 36 Rivalry song, crickets, 237 Rosmarinus oficinalis, trimethylalkanes in, 17 Samia Cynthia, pre-flight warm-up, 185 Sarcina lufea, alkane biosynthesis in, 17, 18 dimethylalkane biosynthesis in, 20 Sarcophaga spp., flight muscle, trehalase in, 164 sarcophagine in, 72,73 Sarcophaga bullata, alkanes in biosynthesis, 21 function, 25 alkenes in, 2 dipeptides, 7 1 flight muscle, carbohydrate, 16 I methylalkanes in, 8, 1 1 3-methylalkanes in, 4
369
peptides in, 70 rhodopsin and metarhodopsin, 46 Sarcophagine, 70-73 hormonal control, 72 in Diptera, 72,73 metabolic fate and function, 71, 72 Saturnia paoonia, oxygen consumption, flight and, 135 Scapsipeous marginatus, amplitude modulation, innate releasing mechanism and, 273 Scapteriscus acletus, amplitude modulation, innate releasing mechanism and, 269 resonant sound emissions, 233 Scapteriscus vicinus, resonant sound emission, 233 Scenedesmus guadricauda, trimethylalkanes in, 17 Schistocerca, brain hormones, 176 flight metabolism, development, 199 flight muscle development, hormonal control, 209 flight speed, 180 metabolic rate, body weight and, 141 tympana1 organs, 286,294 Schisrocerca greguria, adipokinetic hormone, 178 alkane biosynthesis in, 2 1 blood lipids, 175 Corpus cardiacum, peptides from, 96 detoxication mechanisms, 83 flight fuel, 165 mobilization, 169, 170 flight metabolism, development, 200 flight motor, temperature and, 181 flight muscle, metabolism, 172 oxygen supply, 160 flight speed, metabolic rate and, 145 heart-accelerating peptides, 97,98 hyperglycaemic hormones, 174 metabolic rate, during flying, temperature and, 138 mass, wing-loading wingbeat frequency and, 140 neurogenic rhythms, 147 oxygen consumption during flight, 135, 142 power output, neural control, 149, 155 sound production, proprioceptive control, 255 temperature during flight, 137
370
Schistocerca gregaria (con?.) tympana1 organs, 287 Schisrocerca uaga, dimethylalkanes in, 13, 14, 15 methylalkanes in, 9, 11, 12, 16 trimethylalkanes in, 16 Scudderia texensis, amplitude modulation, innate releasing mechanism and, 273, 274 Secapin, 1 15 Seep oil, trimethylalkanes in, 17 Segmental auditory neurons, 301. 302 Sex attractants, alkanes as, 21 Sex peptides, from Drosophila, 9 1-94 synthesis, genetic control, 92, 93 Sex pheromones, alkanes as, 2 1 Shale, trimethylalkanes in, 17 Sharks, trimethylalkanes in, 17 Shivering, thermogenesis, I9 1- 195 Silk worm. See Bombyx mori Siphula ceratites, methylalkanes in, 6 , 7 Simulium, flight muscle temperature, I96 Simulium ornatum, fibrillar muscles, 203 Simulum venustum, metabolic rate, mass, wing-loading, wingbeat frequency and, 140 oxygen consumption, Right and, 135 Simulium vittatum, metabolic rate, mass, wing-loading, wingbeat frequency and, I40 oxygen consumption, Right and, 136 Solenopsis invicta, dimethylalkanes in, 16 methylalkanes in, 1 1 3-methylalkanes in, 4 Solenopsis richteri, dimethylalkanes in, 16 methylalkanes in, 8, 11 3-methylalkanes in, 4 Song patterns, acridid grasshoppers, 239-24 1 crickets, 237 ,evolution, 332-338 genetics, 321-326 innate releasing mechanism and, evolution, 329-332 Tertigonioids, 237-239 Sound emissions, non-resonant, 233-235 resonant, 232-233 Sound patterns, stridulatory movements and, 236-241 Sound production, 339 central nervous system and, 260-267 central vs. peripheral control, 25 1-260
SUBJECT INDEX
neuromuscular basis, 24 1-25I neuronal basis, 231-267 postembryonic development, 3.16-3 19 proprioceptive control, 254-260 Sound reception, 340 neuronal basis, 268-3 16 sensory mechanisms, 281-296 Sound recognition, 340 neuronal basis, 268-3 16 Sound source localization, 309 Spodoptera frugiperh. Right fuels, mobilization, 170 Right lipids, 164 oxygen consumption, Right and, 135 Squid, rhodopsin, 40Stauroderus scalaris, motor co-ordination, sound production and, 248 non-resonant sound emissions, 234 song patterns, 239 evolution, 334 sound patterns, 240 stridulatory movements and neuromuscular activity, 246 stridulatory mechanisms, 232 Stenobothrus lineatus, motor scores, sound production and, 247, 248 muscle activity. co-ordination, sound production and, 245 song patterns, 239 sound patterns, 240 sound production, motor co-ordination, 246 Srenobothrus rubicundus, song pattern, evolution, 334,335 stridulatory mechanisms, 232 Stomoxvs, proline as flight fuel, 165 Stomoxvs calcitrans, alkanes in; function, 2 1 alkenes in, 3 biological activity of alkanes and alkenes in, 22 dimethylalkanes in, 13, 14, 16 y-glutamylphenylalanine in, 73 methylalkanes in, 6, 8, 1 1 Stridulatory mechanisms, 23 I, 232 Stridulatory movements, sound patterns and, 236-24 I Stridulatory organs, 230 Stridulatory patterns, modification by external stimuli, 252-254 Surface lipids, hydrocarbons in, 1 Surface waxes, hydrocarbons in, 1 Syrehus spp., metabolic rate, 146
37 1
SUBJECT INDEX
mass, wing-loading wingbeat frequency and, 140. pre-flight warm-up, 189 Tabanus aflnis, flight speed, metabolic rate and, 145 metabolic rate, mass, wing-loading, wingbeat frequency and, 140 oxygen consumption, flight and, 135 Tabanus septenfrionalis, metabolic rate, mass, wing-loading, wingbeat frequency and, 140 oxygen consumption, flight and, 135 Tarfaroglyllusbucharicus, calling songs, 3 11 song patterns, 238 evolution, 333 Tarfarogryllus burdigalensis, song patterns, 238 Tarfarogryllusfarfarus,calling songs, 3 1 1 song patterns, 237 Teleogryllus, sound production, central nervous system and, 263 Teleogryllus commodus, amplitude modulation, innate releasing mechanism and, 273 female, phonotaxis, 278 flight metabolism, development, 200 innate releasing mechanism, postembryonic development, 3 19 nymphs, sound production, 3 16,3 17 song patterns, genetics, 32 1 stridulation development, 3 18 Teleogryllus oceanicus, flight metabolism, development, 198,200 song patterns, genetics, 32 1 Temperature, flight motor and, 18 1-184 flight muscle, metabolic rate and, 180-197 in flight, metabolic rate and, 137-139 thoracic, stabilization during flight, 190, 19 1 wingbeat frequency and, 183 Tenebrio molifor,alkanes in, 2 blood lipids, 175 Tertiapin, 1 15 Teffigonia,tympanal organ, 288 Teffigoniaviridissima,tympanal organs, 296 Teffigoniidae, amplitude modulation, innate releasing mechanism and, 273,274 non-resonant sound emissionl233 Teftigonioidea, auditory mechanism, evolution, 338
auditory neurons, 306 innate releasing mechanism, postembryonic development, 3 19 song patterns, 237-239 sound patterns, stridulatory movements and, 236 stridulatory mechanisms, 23 1 tympanal organs, 285,288-291,294,296 Thermogenesis, shivering and, nonshivering, 191-195 Thr6-bradykinin, 1 18 Tobacco, cycloalkanes in, 3 Tocopherol in vertebrate photoreceptor membranes, 54 Transduction in insect visual pigments, 57-60 Trehalose, availability, wingbeat frequency and, 179 in Calliphora hemolymph, 177 as flight fuel, 164 mobilization, 169 in flight muscle, 164 synthesis, control by hyperglycaemic hormone, 104 Tribolium confusum, alkanes in, function, 24 alkenes, 3 biological activity of alkanes and alkenes in, 22 Trichogramma evanescens, alkanes in, function, 24 Tricoptera, methylalkanes in, 10 Trilling, Gryllidae, amplitude modulation, 268 Triphaena pronuba. oxygen consumption, flight and, 135 Triticum aestivum, methylalkanes in, 7, 1 1 Truxalinae, song patterns, evolution, 333 stridulatory mechanisms, 232 Tsetse flies. See Glossina T-shaping auditory neurons, 302-3 I4 Tympanal nerve fibres, central projections, 296-300 Tympanal organs, sound reception and, 285296 Tyrosine-0-phosphate in Drosophila, 74 Ultraviolet sensitivity, insects, 53,54 Vanessaio, oxygen consumption, flight and, 135 Venoms, insects, peptides in, 105-1 18 Ventral cord auditory neurons, 300-314
3 72 Vespa crabo, bradykinin-like peptides, 1 17 oxygen consumption, flight and, 136 Vespu oulguris, bradykinin-like peptides, 1 I 7 wingbeat frequency temperature and, 139 Vespulakinin 1, 118 Vespulakinin 2, I 18 Vespula rnaculifrons, kinins from, 1 17 Visual pigments, 35-67 insect, extraction and measurement, 38-40 Voria ruralis, sarcophagine in, 73
Wasps, kinins from, 116-118 Wasp venom, bradykinin-like peptides from, 116 Waterproofing, alkanes and, 25 Wheat. See Triticum aestivum Wingbeat frequency, ambient temperature and, 139
SUBJECT INDEX
metabolic rate and, 140 neural control, 154 power output and, 153 thoracic temperature and, 183 trehalose concentration and, 179 Wing-loading, bodymass and, 139-143 metabolic rate and, 140 Wing rotation, power output, neural control, 155
Wing stretch receptor, sound production and, 255,256 Wing-twisting, flight control mechanism, 148 power output and, 153 Wool wax, cycloalkanes in, 3 methylalkanes in, 10, 12 trimethylalkanes in, 17 Xenobiotics, conjugation with glutathione Saryltransferase, 8 1
Cumulative List of Authors Numbers in boldface indicate the volume numbers of the series
Aidley. 0.J., 4, 1 Andersen, Sven Olav, 2, 1 Ashini, E., 6, 1 Ashburner, Michael, 7, 1 Baccetti, Baccio, 9, 3 15 Barton Browne, L., 11, 1 Beament, J. W. L., 2,67 Berridge, Michael J., 9, I Bodnaryk, Robert P., 13,69 Boistel. J.. 5. I Brady, John. 10. 1 Bridges, R. G., 9,51 Burkhardt, Dietrich, 2, 13 1 Bursell, E., 4, 33 Burtt. E. T., 3, 1 Carlson, A. 0.. 6, 5 1 Catton, W. T.. 3. I Chen, P. S., 3. 53 Calhoun, E. H.. I, 1 Cottrell, C. B., 2, 175 Crossley, A. Clive 11, 1 17 Dadd, R. H., I, 47 Dagan. D., 8.96 Davey, K. G., 2,219 Edwards. John S.. 6,97 Elsner, Norbert, 13,229 Eisenstein. E. M., 9, 11 1 Fraser Rowell, C. H.. 8, 146 Gilbert. Lawrence I., 4. 69 Goodman. Lesley, 7.97 Harmsen, Rudolf, 6, 139 Harvey, W. R., 3, 133 Haskell. J. A,. 3. 133 Heinrich, Bernd. 13, 133 Hinton. H. E.. 5,65 Hoyle. Graham, 7,349 Kafatos, Fotis C.. 12, 1 Kammer, Ann E., 13. 133 Kilby, B. A.. 1. 1 1 1
Lawrence, Peter A., 7, 197 Lees, A. D., 3, 207 Linzen, Bernt, 10, 1 17 Maddrell, S. H. P., 8, 200 Michelsen, Axel, 10,247 Miles, P. W., 9, 183 Miller, P. L., 3, 279 Morgan, E. David, 12, 17 Narahashi, Toshio, 1, 175; 8, 1 Nelson, Dennis R., 13, I Neville, A. C., 4, 213 Nocke, Harold, 10,247 Parnas, I., 8, 96 Pichon, Y..9,257 Poole, Colin F., 12, 17 Popov, Andrej V., 13,229 Prince, William T.. 9, I Pringle, J. W. S., 5, 163 Riddiford, Lynn M., 10, 297 Rowell, Hugh Fraser, 12,63 Rudall. K. M.. 1. 257 Sacktor, Bertram, 7,268 Sander. Klaus, 12, 125 Shaw. J.. 1,3 15 Smith, D. S., 1,401 Steele. J. E., 12, 239 Stobbart. R. H., 1.315 Telfer. William H., 11, 223 Thomson. John A., 11,321 Treherne, J. E., 1.401; 9,257 Truman, James W., 10,297 Usherwood. P. N. R., 6,205 Waldbauer. G. P.. 5,229 Weis-Fogh. Torkel. 2, 1 White, Richard H., 13, 35 Wigglesworth. V. B., 2. 247 Wilson, Donald M., 5. 289 Wyatt. G . R.. 4. 287 Ziegler. Irmgard. 6, 139 373
This Page Intentionally Left Blank
Cumulative List of Chapter Titles Numbers in boldface indicate the i d u m e number of the series
Active Transport and Passive Movement of Water in Insects, 2 , 6 7 Amino Acid and Protein Metabolism in Insect Development, 3 , 5 3 Biochemistry of Sugars and Polysaccharides in Insects, 4, 287 Biochemistry of the Insect Fat Body, 1, I 1 1 Biology of Pteridines in Insects, 6. 139 Biophysical Aspects of Sound Communication in Insects, 10,247 Cells of the Insect Neurosecretory System: Constancy, Variability. and the Concept of the Unique Identifiable Neuron. 12.63 Cellular Mechanisms Underlying Behaviour-Neuroethology, 7, 349 Chitin Orientation in Cuticle and its Control 4, 2 13 Chitin/Protein Complexes of Insect Cuticles. 1. 257 Choline Metabolism in Insects. 9. 5 1 Colour Discrimination in Insects, 2, I3 1 Comparative Physiology of the Flight Motor. 5. 163 Consumption and Utilization of Food by Insects. 5,229 Control of Polymorphism in Aphids. 3, 207 Control of Visceral Muscles in Insects. 2, 219 Cytophysiology of Insect Blood, 11, 1 17 Development and Physiology of Oocyte-Nurse Cell Syncytium, 1I, 223 Effects of Insecticides in Excitable Tissues, 8. I Electrochemistry of Insect Muscle. 6.205 Excitation of Insect Skeletal Muscles, 4. 1 Extraction and Determination of Ecdysones in Arthropods, 12. 17 Excretion o f Nitrogen in Insects. 4. 33 Feeding Behaviour and Nutrition in Grasshoppers and Locusts, I, 47 Frost Resistance in Insects, 6, 1 Function and Structure of Polytene Chromosomes During Insect Development, 7, 1 Functional Aspects of the Organization of the Insect Nervous System. 1.401 Functional Organizations of Giant Axons in the Central Nervous Systems of Insects: New Aspects, 8. 96 Hormonal Control of Metabolism in Insects, 12.239 Hormonal Mechanisms Underlying Insect Behaviour, 10.297 Hormonal Regulation of Growth and Reproduction in Insects, 2, 247 Image Formation and Sensory Transmission in the Compound Eye, 3, 1 Insect Blood-Brain Barrier. 9. 257 Insect Ecdysis with Particular Emphasis on Cuticular Hardening and Darkening, 2, 175 Insect Flight Metabolism. 13, 133 Insect Sperm Cells. 9 , 3 15 37 5
376
CUMULATIVE LIST OF CHAPTER TITLES
Insect Visual Pigments, 13.35 Learning and Memory in Isolated Insect Ganglia, 9, 11 I Lipid Metabolism and Function in Insects, 4,69 Major Patterns of Gene Activity During Development in Holometabolous Insects, 11,321 Mechanisms of Insect Excretory Systems, 8,200 Metabolic Control Mechanisms in Insects, 3, 133 Long-Chain Methyl-Branched Hydrocarbons: Occurrence, Biosynthesis, and Function, 13, 1 Nervous Control of Insect Flight and Related Behaviour, 5,289 Neural Control of Firefly Luminescence, 6,5 1 Neuroethology of Acoustic Communication, 13,229 Osmotic and Ionic Regulation in Insects, 1, 3 15 Physiology of Insect Circadian Rhythms, 10, 1 Physiological Significance of Acetylcholine in Insects and Observations upon other Pharmacologically Active Substances, 1. 1 Polarity and Patterns in the Postembryonic Development of Insects, 7, 197 Postembryonic Development and Regeneration of the Insect Nervous System, 6,97 Properties of Insect Axons, 1, 175 Regulation of Breathing in Insects, 3,279 Regulation of Intermediary Metabolism, with Special Reference to the Control Mechanisms in Insect Muscle, 7, 268 Regulatory Mechanisms in Insect Feeding, 11, 1 Resilin. A Rubberlike Protein in Arthropod Cuticle, 2, 1 Role of Cyclic AMP and Calcium in Hormone Action, 9, 1 Saliva of Hemiptera, 9, 183 Sequential Cell Polymorphism: A Fundamental Concept in Developmental Biology, 12, 1 Specification of the Basic Body Pattern in Insect Embryogenesis, 12, 125 Spiracular Gills, 5, 65 Structure and Function of the Insect Dorsal Ocellus, 7,97 Structure and Function of Insect Peptides, 13,69 Tryptophan --t Ommochrome Pathway in Insects, 10, 117 Variable Coloration of the Acridoid Grasshoppers, 8, 146