CURRENT PROTOCOLS
in Chemical Biology
cp
Current Protocols in Chemical Biology
Online ISBN: 9780470559277 DOI: 10.1002/9780470559277 Editors & Contributors
EDITORIAL BOARD Adam P. Arkin University of California, Berkeley Berkeley, California Lara Mahal New York University New York, New York Floyd Romesberg The Scripps Research Institute La Jolla, California Kavita Shah Purdue University West Lafayette, Indiana Caroline Shamu Harvard Medical School Boston, Massachusetts Craig Thomas NIH Chemical Genomics Center Rockville, Maryland ASSOCIATE EDITORS Michael Burkart University of California, San Diego San Diego, California John Ellman Yale University New Haven, Connecticut Howard Hang The Rockefeller University New York, New York Hans Luecke National Institute of Diabetes and Digestive and Kidney Diseases, NIH Bethesda, Maryland Andreas Marx Universität Konstanz Konstanz, Germany
Michael Rape University of California, Berkeley Berkeley, California Carsten Schultz EMBL Heidelberg Heidelberg, Germany Oliver Seitz Universität zu Berlin Berlin, Germany Katherine L. Seley-Radtke University of Maryland Baltimore, Maryland Nicky Tolliday Broad Institute of Harvard and MIT Cambridge, Massachusetts Gregory A. Weiss University of California, Irvine Irvine, California CONTRIBUTORS Jasmina J. Allen Howard Hughes Medical Institute University of California, San Francisco San Francisco, California Ruben T. Almaraz The Johns Hopkins University Baltimore, Maryland Rogerio Alves de Almeida University of Manchester Manchester, United Kingdom Alma L. Burlingame Howard Hughes Medical Institute University of California, San Francisco San Francisco, California Christopher T. Campbell National Cancer Institute Frederick, Maryland Jennifer Campbell Harvard Medical School Boston, Massachusetts Yong Chi Fred Hutchinson Cancer Research Center Seattle, Washington Bruce E. Clurman Fred Hutchinson Cancer Research Center Seattle, Washington Benjamin F. Cravatt The Scripps Research Institute La Jolla, California Richard D. Cummings Emory University Atlanta, Georgia Arvin C. Dar
Howard Hughes Medical Institute University of California, San Francisco San Francisco, California Jian Du The Johns Hopkins University Baltimore, Maryland Jeremy R. Duvall The Broad Institute of MIT and Harvard Cambridge, Massachusetts Meng Fang University of Georgia Athens, Georgia Matthew Francis University of California, Berkeley Berkeley, California Jeffrey C. Gildersleeve National Cancer Institute Frederick, Maryland Christian Gloeckner University of Konstanz Konstanz, Germany Jay T. Groves Howard Hughes Medical Institute University of California, San Francisco San Francisco, California and National University of Singapore Singapore And Lawrence Berkeley National Laboratory Berkeley, California Howard C. Hang The Rockefeller University New York, New York Rami N. Hannoush Genentech South San Francisco, California Jamie Heimburg-Molinaro Emory University Atlanta, Georgia Nicholas T. Hertz Howard Hughes Medical Institute University of California, San Francisco San Francisco, California Michal Hocek Academy of Sciences of the Czech Republic Prague, Czech Republic Gregory R. Hoffman Harvard Medical School Boston, Massachusetts Eun Ryoung Jang University of Kentucky Lexington, Kentucky Sean Johnston Harvard Medical School Boston, Massachusetts
Hargun S. Khanna The Johns Hopkins University Baltimore, Maryland Kyung Bo Kim University of Kentucky Lexington, Kentucky Ramon Kranaster University of Konstanz Konstanz, Germany Robert D. Kuchta University of Colorado Boulder, Colorado Maya T. Kunkel University of California at San Diego La Jolla, California Wooin Lee University of Kentucky Lexington, Kentucky Jae-Min Lim University of Georgia Athens, Georgia Wan-Chen Lin Howard Hughes Medical Institute University of California, San Francisco San Francisco, California Hana Macickova-Cahová Academy of Sciences of the Czech Republic Prague, Czech Republic Andrew L. MacKinnon University of California San Francisco San Francisco, California Lisa A. Marcaurelle The Broad Institute of MIT and Harvard Cambridge, Massachusetts Gerard Marriott University of California, Berkeley Berkeley, California Andreas Marx University of Konstanz Konstanz, Germany Nathan J. Moerke Harvard Medical School Boston, Massachusetts Nuzhat Motlekar University of Pennsylvania Philadelphia, Pennsylvania Andrew D. Napper University of Pennsylvania Philadelphia, Pennsylvania and University of Manchester Manchester, United Kingdom and Nemours Center for Childhood Cancer Research Wilmington, Delaware
Alexandra C. Newton University of California at San Diego La Jolla, California Takeaki Ozawa The University of Tokyo and Japan Science and Technology Agency Tokyo, Japan Graham D. Pavitt University of Manchester Manchester, United Kingdom Chutima Petchprayoon University of California, Berkeley Berkeley, California Stewart Rudnicki Harvard Medical School Boston, Massachusetts Kevan M. Shokat Howard Hughes Medical Institute University of California, San Francisco San Francisco, California Sharmila Sivendran University of Pennsylvania Philadelphia, Pennsylvania and GlaxoSmithKline Collegeville, Pennsylvania David F. Smith Emory University Atlanta, Georgia Xuezheng Song Emory University Atlanta, Georgia Anna E. Speers The Scripps Research Institute La Jolla, California Elaine Tan The Johns Hopkins University Baltimore, Maryland Jack Taunton University of California San Francisco San Francisco, California Nicola Tolliday The Broad Institute of MIT and Harvard Cambridge, Massachusetts Sara Triffo Howard Hughes Medical Institute University of California, San Francisco San Francisco, California Yoshio Umezawa Musashino University Tokyo, Japan Milan Vrábel Academy of Sciences of the Czech Republic Prague, Czech Republic
Anita Vrcic The Broad Institute of MIT and Harvard Cambridge, Massachusetts Beatrice T. Wang Howard Hughes Medical Institute University of California, San Francisco San Francisco, California Lance Wells University of Georgia Athens, Georgia Leah S. Witus University of California, Berkeley Berkeley, California Kevin J. Yarema The Johns Hopkins University Baltimore, Maryland Jacob S. Yount The Rockefeller University New York, New York Cheng-Han Yu National University of Singapore Singapore Chao Zhang Howard Hughes Medical Institute University of California, San Francisco San Francisco, California Mingzi M. Zhang The Rockefeller University New York, New York Yalong Zhang National Cancer Institute Frederick, Maryland
Cross-Coupling Modification of Nucleoside Triphosphates, PEX, and PCR Construction of Base-Modified DNA Hana Macickova-Cahov´a,1 Milan Vr´abel,1 and Michal Hocek1 1
Institute of Organic Chemistry and Biochemistry, Academy of Sciences of the Czech Republic, Prague, Czech Republic
ABSTRACT A novel, efficient, two-step methodology is presented for construction of base-modified oligonucleotides or DNA, involving aqueous cross-coupling reactions of halogenated nucleoside triphosphates (dNTPs) with terminal acetylenes or arylboronic acids, followed by polymerase incorporation of the modified dNTPs either using primer extension (PEX) C 2010 by John or polymerase chain reaction (PCR). Curr. Protoc. Chem. Biol. 2:1-14 Wiley & Sons, Inc. Keywords: nucleoside triphosphates r DNA polymerase r base-modified DNA r cross-coupling
INTRODUCTION A novel, efficient, two-step methodology is presented for construction of base-modified oligonucleotides or DNA, involving aqueous cross-coupling reactions of halogenated nucleoside triphosphates (dNTPs) with terminal acetylenes or arylboronic acids, followed by DNA polymerase-mediated incorporation of the modified dNTPs either using primer extension (PEX) or polymerase chain reaction (PCR). Each protocol presents typical examples of cross-couplings, PEX experiments, and PCR experiments. Basic Protocol 1 describes the synthesis of 7-(ferrocenylethynyl)-7-deaza-2 -deoxyadenosine triphosphate (2) by Sonogashira cross-coupling of 7-iodo-7-deaza-dATP (1) with ethynylferrocene. Basic Protocol 2 describes the Suzuki cross-coupling of 5-iodo-dCTP (3) with 3-aminophenylboronic acid to give 5-(3-aminophenyl)-dCTP (4). Basic Protocol 3 describes the incorporation of 2 into DNA by PEX, while Basic Protocol 4 describes the incorporation of 4 into DNA by PCR. These are only representative examples of a wide variety of possible modifications of nucleobases that can be introduced and incorporated into DNA. Functional groups at position 5 of pyrimidine and at position 7 of 7-deazapurine bases point out of the major groove of DNA and, therefore, they usually do not distort the stability of the duplex (in many cases they even increase the stability). Such major-groove-functionalized DNA or oligonucleotides may find diverse applications in chemical biology (bioconjugations, studying of interactions with proteins, etc.), bioanalysis or diagnostics (fluorescent or redox labeling, sensors for hybridization), or materials science (DNA-encoded arrays of functional groups).
THE SONOGASHIRA CROSS-COUPLING REACTION OF 9-(2-DEOXY-β-DERYTHRO-PENTOFURANOSYL)-7-IODO-7-DEAZAADENINE 5 -O-TRIPHOSPHATE WITH ETHYNYLFERROCENE Synthesis of 7-(ferrocenylethynyl)-7-deaza-2 -deoxyadenosine triphosphate (2) by Sonogashira cross-coupling of the 7-iodo-7-deaza-dATP (1) with ethynylferrocene is described (Fig. 1). Ferrocene is a useful redox label for electrochemical detection (see Br´azdilov´a et al., 2007).
Current Protocols in Chemical Biology 2: 1-14, February 2010 Published online February 2010 in Wiley Interscience (www.interscience.wiley.com). DOI: 10.1002/9780470559277.ch090177 C 2010 John Wiley & Sons, Inc. Copyright
BASIC PROTOCOL 1
Cross-Coupling Modification of Nucleoside Triphosphates
1 Volume 2
Fe NH2
I
N O O O -O P O P O P O OOO-
N O OH 1
Fe NH2 +
N Pd(OAc)2 TAE, Cul
N O O O -O P O P O P O OOO-
N
N
O OH 2
Figure 1 Synthesis of 9-(2-deoxy-β-D-erythro-pentofuranosyl)-7-(ferrocene-1-yl-ethynyl)-7-deazaadenine 5 -O-triphosphate (2).
Materials 9-(2-deoxy-β-D-erythro-pentofuranosyl)-7-iodo-7-deazapurine 5 -O-triphosphate ˇ (1) [synthesized according to Capek et al. (2007), and commercially available from TriLink BioTechnologies; http://www.trilinkbiotech.com/] Cross-coupling reagents: Ethynylferrocene (Sigma-Aldrich) Palladium(II) acetate [Pd(OAc)2 ;Fluka] Tris(3-sulfonatophenyl)phosphine hydrate, sodium salt (TPPTS; Strem Chemicals, http://www.strem.com/) CuI (Sigma-Aldrich) Triethylamine (TEA; Fluka) Argon Acetonitrile (Sigma-Aldrich) H2 O (HPLC grade) Triethylammonium bicarbonate (TEAB, pH 7.5 at 10◦ to 12◦ C, 2M; store at 4◦ to 6◦ C) Methanol (MeOH, HPLC grade) Mobile phase solution A: 0.1 M TEAB Mobile phase solution B: 0.1 M TEAB in 50% MeOH Mobile phase solution C: MeOH
Cross-Coupling Modification of Nucleoside Triphosphates
10-ml and 50-ml flasks Rubber (PTFE/silicone) septa Magnetic stirrer with Heating MR Hei-Standard (Heidolph; http://www.heidolph.com/) 21-G needles Vacuum system (oil pump with manifold and cold trap) Single-use 1-ml syringe Oil bath Rotary evaporator (Heidolph) equipped with a vacuum system (membrane pump Vacuubrand) Waters 600 HPLC system Column packed with 10 μm C18 reversed-phase medium (Phenomenex, Luna 10 μm C18, 100A HPLC Column 250 × 21.2 mm) 17-mm nylon syringe filter, 0.2 μm (National Scientific) Lyophilizer
2 Volume 2
Current Protocols in Chemical Biology
Perform cross-coupling 1. Add 50 mg (0.05 mmol) 1, 16 mg (0.08 mmol) ethynylferrocene, and 1 mg (0.005 mmol) CuI to one 10-ml flask. 2. Add 0.6 mg (0.0025 mmol) Pd(OAc)2 and 7.5 mg (0.0125 mmol) TPPTS to another 10-ml flask. This flask contains the catalytic system.
3. Place a magnetic stir bar in each flask and seal with a PTFE/silicone septum. Place each of the flasks on a magnetic stir plate and connect to an argon/vacuum line via a 21-G needle inserted through the septum. 4. Exchange vacuum and argon three times very slowly to prevent scattering of the powders and leave each of the flasks attached to the argon line on the manifold. This purges the system and creates an argon atmosphere.
5. Add, through the septum, 1 ml of water:acetonitrile (2:1) mixture followed by 58 μl (0.4 mmol) TEA to the flask prepared in step 1. Repeat step 4. 6. Add 0.5 ml of water:acetonitrile (2:1) to the flask prepared in step 2, with the catalytic system (for faster dissolution of catalytic system, place tube in ultrasonic bath for 2 min). Repeat step 4. 7. When the catalytic system is dissolved, withdraw it by syringe and add it through the septum to the flask prepared in step 1. Repeat step 4 with that flask. 8. Place in an oil bath and heat to 70◦ C with stirring for 1 hr. 9. Cool the reaction mixture to ambient temperature. 10. Evaporate the solvent from the reaction mixture using a rotary evaporator and dissolve the residue in a maximum of 5 ml of HPLC-grade water.
Purify by reversed-phase chromatography 11. Program the gradient system to start with 90% mobile phase A (0.1 M TEAB) and 10% mobile phase B (0.1 M TEAB in 50% MeOH), then increase the percentage of mobile phase solution B (0.1 M TEAB in 50% MeOH) using a linear gradient (Table 1). Ensure that sufficient mobile phase has been installed to keep intakes covered during the chromatography run. 12. Equilibrate the HPLC system with the starting mobile phase composition until a flat baseline is achieved at the desired detection wavelength. Most nucleoside triphosphates absorb at 254 nm, but in specific cases of compounds having absorption minima at 254 nm, an appropriate wavelength should be determined using a photodiode array detector on the analytical HPLC system.
13. Filter sample through a nylon syringe filter. Inject the filtrate. If using a different chemical modification, it is recommended to first perform a pilot run with a small amount of the reaction to determine a suitable gradient.
14. Collect the desired fractions either with an automated fraction collector, or by observing the chromatogram in real time and manually collecting the eluate. 15. Concentrate the collected fraction on a rotary evaporator, and then co-evaporate at least three times with pure water until the sample is without odor of triethylammonium. 16. Freeze-dry to obtain an orange solid product.
Cross-Coupling Modification of Nucleoside Triphosphates
3 Current Protocols in Chemical Biology
Volume 2
Table 1 Gradient Program Used for Purification of 2 After Sonogashira Cross-Coupling Reaction
Elapsed time (min)
Flow (ml/min)
% Mobile phase A
% Mobile phase B
% Mobile phase C
0
10
90
10
0
60
10
0
100
0
80
10
0
0
100
90
10
0
0
100
91
0
0
0
100
17. Confirm the structure of the product by NMR, MS, and UV/Vis spectroscopy. 9-(2-Deoxy-β-D-erythro-pentofuranosyl)-7-(ferrocene-1-yl-ethynyl)-7-deazaadenine 5 O-triphosphate (2) Yield 48%. MS (ESI+ ): m/z: 721.53 (30,M+Na), 801.46 (70,M+Et3 N), 902.64 (100,M+2×Et3 N). UV/Vis (H2 O): lmax (e)=232 (109 849), 281 (89 393), 440 nm (3788) 1 H NMR (500 MHz, D2 O, refdioxane =3.75 ppm): 2.42-.60 (m, 2H; H-2 ), 4.23-.37 (m, 8H; H-4 ,5 and cp), 4.12 (mbr, 2H; H-3 ,4 ), 4.58 and 4.64 (2×m br, 2H; H-2 ,5 ), 4.70 (m br, 1H; H-3 ), 6.48 (t br, J1 ,2 =5.9 Hz, 1H; H-1 ), 7.72 (s, 1H; H-8), 8.23 ppm (s br, 1H; H-2). C NMR (125.7 MHz, D2 O, refdioxane =69.3 ppm): 42.77 (CH2 –2 ), 65.6 (C-1 ), 68.4 (d, JC,P =4 Hz, CH2 –5 ), 72.48 and 72.51 (CH-3 ,4 ), 72.8 (CH-cp), 73.7 (CH-3 ), 74.2 and 74.3 (CH-2 ,5 ), 78.4 (C=C-Fc), 86.6 (CH-1 ), 88.3 (d, JC,P =9 Hz, CH-4 ), 96.3 (C=C-Fc), 102.0 (C-7), 104.3 (C-5), 129.8 (CH-8), 146.3 (CH-2), 148.5 (C-4), 153.6 ppm (C-6); 31 P NMR (202.3 MHz, D2 O): -22.33 (t, J=23, 20 Hz, Pβ ), -10.46 (d, J= 20 Hz, Pα ), -10.10 ppm (d, J=23 Hz, Pγ ).
13
BASIC PROTOCOL 2
THE SUZUKI-MIYAURA CROSS-COUPLING REACTION OF 1-(2-DEOXY-β-D-ERYTHRO-PENTOFURANOSYL)-5-IODOCYTIDINE 5 -O-TRIPHOSPHATE WITH 3-AMINOPHENYLBORONIC ACID The Suzuki cross-coupling of 5-iodo-dCTP (3) with 3-aminophenylboronic acid to give 5-(aminophenyl)-dCTP (4) is described (Fig. 2). The aminophenyl group is a useful redox label for electrochemical detection (see Cahov´a et al., 2008b).
Materials
Cross-Coupling Modification of Nucleoside Triphosphates
1-(2-deoxy-β-D-erythro-pentofuranosyl)-5-iodocytidine 5 -O-triphosphate (3) [prepared according to Cahov´a (2008b); commercially available from TriLink BioTechnologies; http://www.trilinkbiotech.com/] Cross-coupling reagents: 3-aminophenylboronic acid hydrochloride (Sigma-Aldrich) Palladium(II) acetate [Pd(OAc)2 ; Fluka] Tris(3-sulfonatophenyl)phosphine hydrate, sodium salt (TPPTS, Strem Chemicals, http://www.strem.com/) Cesium carbonate (Cs2 CO3 , Sigma-Aldrich) Argon Acetonitrile (Sigma-Aldrich) H2 O (HPLC grade) Methanol (MeOH, HPLC grade) Triethylammonium bicarbonate (TEAB, pH 7.5 at 10◦ to 12◦ C, 2 M, stored at 4◦ to 6◦ C) Mobile phase solution A : 0.1 M TEAB Mobile phase solution B: 0.1 M TEAB in 50% MeOH Mobile phase solution C : MeOH
4 Volume 2
Current Protocols in Chemical Biology
NH2 I O O O -O P O P O P O OOO-
N N
NH2
B(OH)2 +
OH
N
NH2
O
O
NH2
Pd(OAc)2 TTPPTS Cs2CO3
O O O -O P O P O P O OOO-
3 Figure 2
N
O
O OH 4
Synthesis of 1-(2-deoxy-β-D-erythro-pentofuranosyl)-5-(3-aminophenyl)-cytidine 5 -O-triphosphate (4).
Dowex 50WX8 in Na+ cycle (Fluka) Lyophilizer 10-ml and 50-ml flasks Vacuum system (oil pump with manifold and cold trap) Rubber (PTFE/silicone) septa 21-G needles Magnetic stirrer with Heating MR Hei-Standard (Heidolph) Single-use 1-ml syringe Oil bath 17-mm nylon syringe filter 0.2 μm (National Scientific) Waters 600 HPLC system Column packed with 10 μm C18 reversed phase (Phenomenex, Luna 10 μm C18, 100A HPLC Column 250 × 21.2 mm) Nylon 66 membrane filter, 0.2 μm × 47 mm (Supelco) Rotary evaporator (Heidolph) equipped with a vacuum system (membrane pump Vacuubrand) Glass column (1.5 cm i.d. × 12.5 cm length) Perform cross-coupling 1. Add 50 mg (0.05 mmol) 3, 17.3 mg (0.1 mmol) 3-aminophenylboronic acid, and 81 mg Cs2 CO3 (0.25 mmol) to one 10-ml flask. 2. Add 1.12 mg (0.005 mmol) Pd(OAc)2 and 14.2 mg (0.025 mmol) TPPTS to another 10-ml flask. This flask contains the catalytic system.
3. Place a magnetic stir bar in each flask and seal with PTFE/silicone septum. Place the each of the flasks on a magnetic stir plate and connect to an argon/vacuum line via a 21-G needle inserted through the septum. 4. Exchange vacuum and argon three times very slowly to prevent scattering of the powders, and leave each of the the flasks attached to the argon line on the manifold. This purges the system and creates an argon atmosphere.
5. Add, through the septum, 0.5 ml of water:acetonitrile (2:1) mixture to the flask prepared in step 1 and 0.3 ml of water:acetonitrile (2:1) to the flask prepared in step 2, with the catalytic system (for faster dissolution of catalytic system, place tube in ultrasonic bath for 2 min). Repeat step 4 with both flasks.
Cross-Coupling Modification of Nucleoside Triphosphates
5 Current Protocols in Chemical Biology
Volume 2
6. When the catalytic system is dissolved, withdraw it by syringe and add it through the septum to the flask prepared in step 1. Repeat step 4 with that flask. 7. Place in an oil bath on the magnetic stir plate and heat the reaction flask to 120◦ C. Stir the reaction mixture 30 min at 120◦ C. 8. Cool the reaction mixture to ambient temperature.
Purify by reversed-phase chromatography 9. Program the gradient system to start with 90% mobile phase A (0.1 M TEAB) and 10% mobile phase B ( 0.1 M TEAB in 50% MeOH), then increase the percentage of mobile phase solution B (0.1 M TEAB in 50% MeOH) with time (Table 1). Ensure that sufficient mobile phase has been installed to keep intakes covered during time. 10. Equilibrate the HPLC system with the starting mobile phase composition until a flat baseline is achieved at the desired detection wavelength. Most nucleoside triphosphates absorb at 254 nm, but in specific cases of compounds having absorption minima at 254 nm, an appropriate wavelength should be determined using a photodiode array detector on the analytical HPLC system.
11. Filter sample through a nylon syringe filter. Inject the filtrate. If using a different chemical modification, it is recommended to first perform a pilot run with a small amount of the reaction to determine a suitable gradient.
12. Collect the desired fractions either with an automated fraction collector, or by observing the chromatogram in real time and manually collecting the eluate. 13. Concentrate the collected fraction on a rotary evaporator and then coevaporate at least three times with pure water until the sample is without odor of triethylammonium. 14. Pour Dowex 50WX8 in Na+ cycle on a small glass column and apply evaporated sample in 5 ml of water. Elute with 20 ml water. 15. Freeze-dry to obtain a white solid product. 16. Confirm the structure of the product by NMR, MS, and HR MS. 1-(2-Deoxy-β-D-erythro-pentofuranosyl)-5-(3-aminophenyl)-cytidine 5 -O-triphosphate (4) Yield 43%. MS(ESI- ): 557 (100, M-1), HRMS: for C15 H20 N4 O13 P3 calculated 557.0240 found 557.0255. NMR spectra for 4×Na+ salt at pH 7: 1 H NMR (500 MHz, D2 O, refdioxane = 3.75 ppm, pH = 7.1): 2.36 (dt, 1H, Jgem = 14.1, J2 b,1 = J2 b,3 = 6.7, H2 b); 2.46 (ddd, 1H, Jgem = 14.1, J2 a,1 = 6.2, J2 a,3 = 3.6, H-2 a); 4.15-4.20 (m, 2H, H-5 ); 4.22 (m, 1H, H-4 ); 4.62 (dt, 1H, J3 ,2 = 6.7, 3.6, J3 ,4 = 3.6, H-3 ); 6.34 (dd, 1H, J1 2 = 6.7, 6.2, H-1 ); 6.98-7.06 (m, 3H, H-2,4,6-C6 H4 NH2 ); 7.38 (t, 1H, J5,4 = J5,6 = 7.8, H-5-C6 H4 NH2 ); 7.80 (s, 1H, H-6). 13 C NMR (151 MHz, D2 O, refdioxane = 69.3 ppm): 42.61 (CH2 -2 ); 67.64 (d, JC,P = 5, CH2 -5 ); 72.65 (CH-3 ); 88.93 (d, JC,P = 9, CH4 ); 89.42 (CH-1 ); 110.61 (C-5); 126.59 (CH-4-C6 H4 NH2 ); 127.40 (CH-2-C6 H4 NH2 ); 131.71 (CH-6-C6 H4 NH2 ); 133.43 (CH-5-C6 H4 NH2 ); 134.30 (C-1-C6 H4 NH2 ); 134.67 (C-3-C6 H4 NH2 ); 145.64 (CH-6); 151.81 (C-2); 161.57 (C-4). 31 P (1 H dec.) NMR (202.3 MHz, D2 O, refH3 PO4 = 0 ppm, pH = 7.1): -22.98 (t, J = 19.3, Pβ ); -11.67 (d, J = 19.3, Pα ); -10.14 (d, J = 19.3, Pγ ). BASIC PROTOCOL 3
Cross-Coupling Modification of Nucleoside Triphosphates
PRIMER EXTENSION EXPERIMENT (PEX) WITH 7-MODIFIED 7-DEAZA-dATP Incorporation of 2 to DNA by primer extension is described here. Primer extension is useful for preparation of shorter DNA sequences containing one or several modified nucleobases.
6 Volume 2
Current Protocols in Chemical Biology
Materials PCR Ultra H2 O (Top-Bio, Czech Republic) 10 U/ml T4 polynucleotide kinase (Takara) and 10× polynucleotide kinase buffer Adenosine 5 -[γ-32 P] triphosphate, triethylammonium salt (Izotop, Institutes of Isotopes Co.; http://www.izotop.hu) Primer 5 -CATGGGCGGCATGGG-3 (VBC Biotech, http://www.vbc-biotech.at/) Dynazyme DNA polymerase and 10× buffer (Finnzymes, http://www.finnzymes.com/) Template temp2A : 5 -GCGTGGAAGTGGAGCCCATGCCGCCCATG-3 (VBC Biotech) Natural deoxynucleoside triphosphates: dATP, dCTP, dGTP, dTTP (Fermentas) 2 (Basic Protocol 1) PAGE stop solution (see recipe) Rotiphorese Sequencing Gel Concentrate (Carl Roth; http://www.carl-roth.de) 2× and 1× TBE buffer (see recipe) N,N,N ,N -tetramethylethylenediamine (TEMED, Sigma-Aldrich) Ammonium persulfate (APS, Sigma-Aldrich) 1.5-ml microcentrifuge tubes (Axygen) MicroSpin G-25 Columns (GE Healthcare) Heating blocks 250 ml Erlenmeyer flask Sequigen electrophoresis apparatus, 21 × 40 (BioRad) PowerPac HV electrophoresis power supply (BioRad) Scalpel 3 MM chromatography paper (Whatman) Gel Dryer Model 583 (BioRad) Plastic wrap (e.g., Saran Wrap) Storage phosphor screen and cassette (Amersham Biosciences) Typhoon 9410 Gel Imager (Amersham Biosciences) CAUTION: Radioactive materials require special handling. The authors recommend the use of Sekuroka radiation protection screens, protection boxes, and protection waste boxes (available from Carl Roth). Dispose of all waste material in accordance with institutional radiation safety regulations.
Radiolabel the primer (working on ice) 1. To a 1.5-ml microcentrifuge tube on ice, add in the following order: 39 μl PCR Ultra H2 O 5 μl 10× T4 polynucleotide kinase buffer 2 μl 10 μM primer 2 μl [γ-32 P]ATP 2 μl 10 U/μl T4 polynucleotide kinase. 2. Mix using vortex and then briefly microcentrifuge to collect the solution at the bottom of the tube. 3. Incubate 1 hr at 37◦ C and then 5 min at 95◦ C to inactivate T4 polynucleotide kinase. 4. Resuspend the resin in the MicroSpin Column by vortexing, loosen the cap onequarter turn and snap off the bottom closure, place the column in a supplied collection tube, and pre-spin for 1 min at 735 × g, room temperature. 5. Transfer the column to a new clean microcentrifuge tube. Apply the sample to the center of the resin bed. Spin the column 2 min at 735 × g, room temperature.
Cross-Coupling Modification of Nucleoside Triphosphates
7 Current Protocols in Chemical Biology
Volume 2
6. Withdraw the filtrate from the collection tube and transfer it to new 1.5-ml microcentrifuge tube. 7. Add 20 μl 10 μM unlabeled primer to the filtrate to obtain a 3 μM solution of radiolabeled primer. Vortex and briefly microcentrifuge to collect the solution at the bottom of the tube.
Perform PEX with compound 2 (working on ice) 8. Prepare master mix: to a 1.5-ml microcentrifuge tube, add in this order: 25.5 μl PCR Ultra H2 O 6 μl 10× Dynazyme polymerase buffer 4.5 μl 3 μM temp2A 7.5 μl 1 mM mixture of dCTP, dGTP, and dTTP 3 μl 3 μM radiolabeled primer (from step 7) 6 μl (12 U) of Dynazyme DNA polymerase. 9. Mix using vortex then microcentrifuge briefly to collect the solution at the bottom of the tube. 10. Prepare two new 1.5-ml microcentrifuge tubes and label one as positive control and one as the experiment with modified dATP. 11. Transfer 17.5 μl of master mix to each microcentrifuge tube. 12. Add 2.5 μl of 1 mM dATP to the microcentrifuge tube labeled as the positive control and 2.5 μl of 1 mM 2 to the second microcentrifuge tube. 13. Mix each sample by vortexing then microcentrifuge briefly to collect the sample at the bottom of the tube. 14. Incubate 30 min at 60◦ C. 15. Add 40 μl of PAGE stop solution to each microcentrifuge tube. 16. Vortex to mix, then microcentrifuge briefly to collect the solution at the bottom of the tube.
Perform denaturing polyacrylamide gel electrophoresis 17. Mix the following PAGE gel reagents in a 250-ml Erlenmeyer flask: 50 ml Rotiphorese Sequencing Gel Concentrate 50 ml 2× TBE buffer 40 μl TEMED 800 μl 10% APS 18. Immediately pour gel mixture in Sequigen electrophoresis apparatus. Insert comb (supplied with apparatus) between the glass plates. 19. After 1 hr remove the comb and cover gel with 1× TBE buffer. Remove any air bubbles by gently pipetting buffer into the wells. 20. Prerun gel and heat it to 50◦ C using temperature probe (supplied with apparatus) connected to the PowerPac power supply (set up 30 mA). Cross-Coupling Modification of Nucleoside Triphosphates
21. Load 2 μl of every sample in neighboring slots. First load primer (19 μl PCR Ultra H2 O, 1 μl of 3 μM radiolabeled primer, and 40 μl PAGE stop solution), followed by positive control and experimental sample.
8 Volume 2
Current Protocols in Chemical Biology
C G C A C C T T C A C C T +
Fc-A-DNA
Figure 3 12% polyacrylamide gel of PEX products [+ = positive control, i.e., DNA containing natural nucleotides; Fc-A-DNA = DNA containing 7-(ferrocene-1-yl-ethynyl)-7-deazadeoxyadenosine instead of natural deoxyadenosine; on the right-hand side the sequence of product is given].
22. Reconnect the electrophoresis apparatus to the power supply, electrophorese at 20 mA for ∼1.5 hr, and maintain the temperature of the apparatus around 50◦ C to keep strands denatured. 23. Stop electrophoresis, carefully separate the glass plates, cut the desired portion of the gel using a scalpel (DNA is in the middle between two bands of blue colors used in stop solution), place an appropriately sized piece of wet chromatography paper on the gel, gently remove the gel from the glass, and cover with plastic wrap. 24. Dry in a gel dryer 1 hr at 85◦ C. 25. Place dry gel in storage phosphor cassette with screen and cover with plastic wrap. 26. After 12 hr (depending on the specific activity of the ATP [γ-32 P]), scan on Typhoon Gel Imager. Use Image J and Adobe Photoshop software for manipulation of the image (Fig. 3).
POLYMERASE CHAIN REACTION (PCR) INCORPORATION OF 5-MODIFIED dCTP INTO DNA
BASIC PROTOCOL 4
Incorporation of 4 into DNA duplex by PCR is described here. PCR is useful for synthesis and amplification of long DNA duplexes containing multiple modifications.
Materials PCR Ultra H2 O (Top-Bio, Czech Republic) Vent(exo-) DNA polymerase and 10× buffer (New England Biolabs) Template Temp PCR (VBC Biotech; see Table 2 for sequence)
Cross-Coupling Modification of Nucleoside Triphosphates
9 Current Protocols in Chemical Biology
Volume 2
Table 2 Sequences of Template and Primers Employed in PCR
Sequences Temp PCR
5 -GCGTGTGGAGTATTTGGATGACAGAAACACTTTTCGACATA GTGTGGTGGTGCCCTATGAGCCGCCTGAGGTTGGCTCTGACTG TACCACCATCCACTACAACTACATGTGTAACAGTTCCTGCAT GGGCGGCATGAACCGGAGGCCCATCCTCACCATCATCACACTGG AAGACTCCAGTGGTAATCTACTGGGACGGAACAGCTTTGAGGTG CGTGTTTGTGCCTGTCCTGGGAGAGACCGGCGCACAGAGGA-3
Primer for
5 -ATCCAAATACTCCACACGCA-3
Primer rev
5 -AGAGACCGGCGCACAGAGGA-3
Natural deoxynucleoside triphosphates: dATP, dCTP, dGTP, dTTP (Fermentas) PCR primers: Primer for and Primer rev (VBC Biotech; see Table 2 for sequences) Tetramethylammonium chloride (TMAC, Fluka) 4 (Basic Protocol 2) Loading color buffer (see recipe) Agarose for DNA electrophoresis (Serva) 0.5× TBE buffer (see recipe) Low Molecular Weight DNA Ladder (New England Biolabs) 0.5× TBE buffer containing 0.5 μg/ml ethidium bromide 1.5-ml and 0.5-ml microcentrifuge tubes (Axygen) TGradient 96 thermal cycler (Whatman Biometra) 250-ml Erlenmeyer flask Owl B1 type electrophoresis system (Owl Separation System, Inc.) PowerPac HV (Bio-Rad) Opaque plastic box UltraCam 8gD Digital Imaging System ( Ultra.L¯um, http://www.ultralum.com/) Electronic dual-wave transilluminator (Ultra.L¯um) Perform PCR with 4 (working on ice) 1. Prepare master mix: In a 1.5-ml microcentrifuge tube on ice, add in this order: 18 μl PCR ultra H2 O 6 μl 10× Vent(exo-) polymerase buffer 12 μl 100 nM template Temp PCR 3 μl 4 mM mixture of dATP, dGTP, and dTTP 6 μl 10 μM Primer for 6 μl 10 μM Primer rev 3 μl 50 mM TMAC 3 μl (6 U) Vent(exo-) DNA polymerase. 2. Vortex to mix, then briefly microcentrifuge to collect the solution at the bottom of the tube. 3. Prepare two new 0.5-ml microcentrifuge tubes and label one as positive control and one as experiment with modified dCTP. 4. Transfer 19 μl of master mix to each tube. Cross-Coupling Modification of Nucleoside Triphosphates
5. Add 1 μl of 4 mM dCTP to the tube with the mixture labeled as positive control and 1 μl of a 4 mM solution of 4 to the second tube. 6. Vortex to mix and then briefly microcentrifuge to collect the solution at the bottom of the tube.
10 Volume 2
Current Protocols in Chemical Biology
7. Transfer tubes to TGradient thermocycler preheated to 80◦ C and perform PCR using the following program: 1 cycle: 30 cycles:
1 cycle:
3 min 1.5 min 2 min 3 min 5 min
94◦ C 94◦ C 60◦ C 72◦ C 72◦ C
(initial denaturation) (denaturation) (annealing) (extension) (final extension).
The choice of DNA polymerase and additives must be optimized, and depends on the type of modification. Temperature and time of annealing and denaturation depend on length and sequence of template and primers. Before using a different template, PCR conditions must be optimized for the primers and/or modified dNTPs used.
8. After PCR, add 10 μl of loading color buffer to each tube. 9. Mix by vortexing, then microcentrifuge briefly to collect the solution at the bottom of the tube.
Perform agarose gel electrophoresis 10. Add 2 g agarose to 100 ml of 0.5× TBE in a 250-ml Erlenmeyer flask. 11. Put the flask in microwave oven and heat at 700 W at least 1 min. When it boils, swirl the flask to mix (wearing protective gloves), put it back in oven, and let it boil; repeat this step until the solution is clear without bubbles. If there are air bubbles in the gel, remove them with a pipet tip. 12. Cool the agarose solution slightly under flowing tap water. The cooling is done to prevent breaking of electrophoresis system due to extreme, rapid temperature changes.
13. Pour gel into electrophoresis system and insert comb (supplied with electrophoresis apparatus). 14. Wait for at least 30 min until the gel solidifies. 15. Remove comb and then pour 0.5× TBE buffer on top of the gel. 16. Load 10 μl of each sample in neighboring slots. First load 20 μl Low Molecular Weight DNA Ladder in 90 μl 0.5× TBE combined with 55 μl color solution supplied with the DNA ladder), followed by the positive control and the experimental sample.
L
+
NH2-C-DNA
200 bp
1
2
3
4
5
6
7
Figure 4 2% agarose gel of PCR products post-stained by ethidium bromide [L=ladder; 1(+) = positive control (i.e., DNA containing natural nucleotides); lane 6 (NH2 -C-DNA ) = DNA containing 5-(3-aminophenyl)-deoxycytidine instead of natural deoxycytidine; lanes 2, 3, 4, 5, 7 are experiments with other modified dNTPs (not discussed here) in this order: 9-(2-Deoxy-β-D-erythro-pentofuranosyl)-7-(3-aminophenyl)-7-deazaadenine 5 -O-triphosphate, 9-(2-Deoxy-β-D-erythro-pentofuranosyl)-7-(3-nitrophenyl)-7-deazaadenine 5 -O-triphosphate, 1-(2-Deoxy-β-D-erythro-pentofuranosyl)-5-(3-aminophenyl)-uridine- 5 -O-triphosphate, 1-(2Deoxy-β-D-erythro-pentofuranosyl)-5-(3-nitrophenyl)-uridine- 5 -O-triphosphate, 1-(2-Deoxy-βD-erythro-pentofuranosyl)-5-(3-nitrophenyl)-cytidine 5 -O-triphosphate)].
Cross-Coupling Modification of Nucleoside Triphosphates
11 Current Protocols in Chemical Biology
Volume 2
16. Connect electrophoresis apparatus to PowerPac machine and electrophorese at 120 V for approximately 2 hr. 17. Transfer gel to an opaque plastic box containing 0.5× TBE with 0.5 μg/ml ethidium bromide. Leave in buffer at least 20 min. CAUTION: Ethidium bromide is a toxic and mutagenic compound. Use nitrile gloves for handling.
18. Transfer gel to an electronic dual-wave transilluminator and record images of the gel (e.g., using the UltraCam 8gD Digital Imaging System) Representative results are shown in Fig. 4.
REAGENTS AND SOLUTIONS Use Milli-Q purified water or equivalent in all recipes and protocol steps.
Loading color buffer 0.5× TBE buffer (see recipe for 5×) containing: 1 mg/ml xylene cyanol (Sigma-Aldrich) 1 mg/ml bromphenol blue (Sigma-Aldrich) Store up to 1 year at 4◦ to 6◦ C PAGE stop buffer 80% (v/v) formamide 10 mM disodium EDTA 1 mg/ml xylene cyanol 1 mg/ml bromphenol blue Store up to 1 year at 4◦ to 6◦ C TBE buffer, 2× 21.6 g Tris base 11 g boric acid 1.5 g disodium EDTA H2 O to 40 ml Store up to 1 month at 4◦ to 6◦ C TBE buffer, 5× 54 g Tris base 27.5 g boric acid 3.8 g disodium EDTA H2 O to 40 ml Store up to 1 month at 4◦ to 6◦ C COMMENTARY Background Information
Cross-Coupling Modification of Nucleoside Triphosphates
Base-functionalized nucleic acids are of great current interest due to applications in chemical biology, bioanalysis, or nanotechnology and materials science (Peracchi, 2005; Condon, 2006; Famulok et al., 2007). There are several approaches (Weisbrod and Marx, 2008) to their synthesis: chemical synthesis of oligonucleotides on solid support using modified phosphoramidites, enzymatic incorporation of modified nucleo-
side triphosphates, or post-synthetic modifications of oligonucleotides [e.g., by clickchemistry or Staudinger ligation, Weisbrod and Marx (2007); Gramlich et al. (2008)]. Particularly efficient is our two-step methodology consisting of the synthesis of modified dNTPs by aqueous cross-coupling reactions followed by polymerase incorporation (Fig. 5, Hocek and Fojta, 2008). 5-Substituted pyrimidine dNTPs are usually good substrates for DNA polymerases, while 8-substituted purine
12 Volume 2
Current Protocols in Chemical Biology
r
ke
lin
M lin
step 1:
nucleobase
nucleobase
O
aqueous-phase cross-coupling
O O O -O P O P O P O OOO-
O OH
OH
linker
polymerase incorporation
linker
step 2:
linker
O O O -O P O P O P O OOO-
r
ke
I
Figure 5 General scheme of the two-step synthesis of base-modified DNA. The first step is aqueous cross-coupling reaction to attach the modification to dNTP and the second step is polymerase incorporation of the modified nucleotide to DNA.
dNTPs have repeatedly been shown to be poor substrates which should be replaced by 7substituted 7-deazapurine dNTPs (Jager et al., 2005; Cahov´a et al., 2008a). We have recently used this novel approach for the synthesis ˇ of DNA-bearing amino acids (Capek et al., 2007), ferrocenes (Br´azdilov´a et al., 2007), amino- and nitrophenyl groups (Cahov´a et al., 2008b), or Ru/Os(bpy)3 complexes (Vr´abel et al., 2009). The high tolerance of the polymerases to the presence of substituents of different natures and sizes suggests that this methodology is a good general approach to obtaining DNA modified by almost any kind of substituent needed for wide range of applications, at least with appropriate linkers (see review in Hocek et al., 2008). The most important applications have been in construction of oligonucleotide probes labeled by redoxactive markers and their use in bioanalysis (minisequencing or multicolor electrochemical coding of DNA bases; Br´azdilov´a et al., 2007; Cahov´a et al., 2008b; Vr´abel et al., 2009). The primer extension usually works well for most types of modified dNTPs using vari-
ous DNA polymerases (Vent(exo-), Pwo, DyNAzyme, Klenow, Phusion etc.). It is a useful method for introduction of one or several modifications into one strand of DNA. On the other hand, PCR is much more demanding and only some modified dNTPs work well in it to yield clean, full-length amplified products. This method is suitable for synthesis of large DNA duplexes with a high density of modifications (i.e., when using modified dATP, all A in the DNA will bear the modifications).
Critical Parameters and Troubleshooting The most critical parameter in working with 2 -deoxyribo 5 -O-triphosphates is their instability toward hydrolysis in aqueous solutions (especially at higher temperature and in acidic conditions). Therefore the reaction time should be as short as possible and evaporation has to be performed at 40◦ C. Prolongation of the reaction time or increasing the temperature of the cross-coupling reaction (30 min at 120◦ C or 1 hr at 70◦ C) leads to higher conversion of the cross-coupling, but unfortunately also to higher amounts of
Cross-Coupling Modification of Nucleoside Triphosphates
13 Current Protocols in Chemical Biology
Volume 2
diphosphate, a product of partial hydrolysis. The prepared triphosphates should be lyophilized and stored in the freezer. The solutions of prepared triphosphates should be portioned into several aliquots, which are thawed and refrozen a maximum of five times. After that, use a new aliquot. Frequent thawing and freezing of the solutions causes partial hydrolysis of triphosphates to diphosphates and monophosphates. Diphosphates and as well inorganic pyrophosphate inhibit DNA polymerase. For these reasons, avoid using impure or partially hydrolyzed triphosphates in PEX and PCR.
Anticipated Results Using this methodology one can, in principle, attach almost any kind of substituent either via an acetylene spacer or via an aryl group at position 5 of pyrimidine or position 7 of 7-deazapurine nucleoside triphosphates. The yields (due to partial hydrolysis) vary from 10% to 70%. However, as the method is extremely rapid and straightforward, the separation is usually easy and only small amounts of modified dNTPs are needed for polymerase incorporations—even moderate yields are generally acceptable. The basemodified dNTPs then can be incorporated into DNA by the polymerase. For each modification, it is useful to test several polymerases and select the most efficient one. PEX incorporations usually work well with most types of modifications, while PCR is much more demanding and is efficient only for some modifications. PEX (in combination with magnetic separation) is suitable for synthesis of singlestrand oligonucleotides bearing one or several modifications. PCR is suitable for construction and amplification of DNA duplexes bearing manifold modifications.
Time Considerations
Cross-Coupling Modification of Nucleoside Triphosphates
The Sonogashira cross-coupling reaction with commercial acetylene derivatives is completed in 1 hr, the Suzuki-Miyaura crosscoupling reaction with commercial boronic acid in 30 min. The purification by HPLC and conversion to sodium salt take at least 3 additional hours. The freeze drying can be done overnight. Performing the PEX experiment and PAGE to obtain the final dry gel takes another 5 hr. PCR are accomplished with mentioned gradient within 3.5 hr. Analysis of the products of PCR consumes ∼3 hr.
Literature Cited Br´azdilov´a, P., Vr´abel, M., Pohl, R., Pivoˇnkov´a, H., Havran, L., Hocek, M., and Fojta, M.
2007. Ferrocenylethynyl derivatives of nucleoside triphosphates: Synthesis, incorporation, electrochemistry, and bioanalytical applications. Chem. Eur. J. 13:9527-9533. Cahov´a, H., Pohl, R., Bedn´arov´a, L., Nov´akov´a, K., Cvaˇcka, J., and Hocek, M. 2008a. Synthesis of 8-bromo-, 8-methyl- and 8-phenyl-dATP and their polymerase incorporation into DNA. Org. Biomol. Chem. 6:3657-3660. Cahov´a, H., Havran, L., Br´azdilov´a, P., Pivoˇnkov´a, H., Pohl, R., Fojta, M., and Hocek, M. 2008b. Aminophenyl- and nitrophenyl-labeled nucleoside triphosphates: Synthesis, enzymatic incorporation, and electrochemical detection. Angew. Chem. Int. Ed. 47:2059-2062. ˇ Capek, P., Cahov´a, H., Pohl, R., Hocek, M., Gloeckner, C., and Marx, A. 2007. An efficient method for the construction of functionalized DNA bearing amino acid groups through crosscoupling reactions of nucleoside triphosphates followed by primer extension or PCR. Chem. Eur. J. 13:6196-6203. Condon, A. 2006. Designed DNA molecules: Principles and applications of molecular nanotechnology. Nat. Rev. Genet. 7:565-575. Famulok, M., Hartig, J.S., and Mayer, G. 2007. Functional aptamers and aptazymes in biotechnology, diagnostics, and therapy. Chem. Rev. 107:3715-3743. Gramlich, P.M., Wirges, C.T., Gierlich, J., and Carell, T. 2008. Synthesis of modified DNA by PCR with alkyne-bearing purines followed by a click reaction. Org. Lett. 10:249-251. Hocek, M. and Fojta, M. 2008. Cross-coupling reactions of nucleoside triphosphates followed by polymerase incorporation: Construction and applications of base-functionalized nucleic acids. Org. Biomol. Chem. 6:22332241. Jager, S., Rasched, G., Kornreich-Leshem, H., Engeser, M., Thum, O., and Famulok, M. 2005. A versatile toolbox for variable DNA functionalization at high density. J. Am. Chem. Soc. 127:15071-15082. Peracchi, A. 2005. DNA catalysis: Potential, limitations, open questions. ChemBioChem 6:13161322. Vr´abel, M., Hor´akov´a, P., Pivoˇnkov´a, H., ˇ Kalachova, L., Cernock´ a, H., Cahov´a, H., Pohl, ˇ P., Havran, L., Hocek, M. and FoR., Sebest, jta, M. 2009. Base-modified DNA labeled by [Ru(bpy)(3)](2+) and [Os(bpy)(3)](2+) complexes: Construction by polymerase incorporation of modified nucleoside triphosphates, electrochemical and luminescent properties, and applications. Chem. Eur. J. 15:11441154. Weisbrod, S.H. and Marx, A. 2007. A nucleoside triphosphate for site-specific labelling of DNA by the Staudinger ligation. Chem. Commun. 1828-1830. Weisbrod, S.H. and Marx, A. 2008. Novel strategies for the site-specific covalent labelling of nucleic acids. Chem. Commun. 5675-5685.
14 Volume 2
Current Protocols in Chemical Biology
Chemical Genetic Approach for Kinase-Substrate Mapping by Covalent Capture of Thiophosphopeptides and Analysis by Mass Spectrometry Nicholas T. Hertz,1,2 Beatrice T. Wang,2 Jasmina J. Allen,1,2 Chao Zhang,2,4 Arvin C. Dar,2 Alma L. Burlingame,3 and Kevan M. Shokat2,4 1
Chemistry and Chemical Biology Graduate Program, Howard Hughes Medical Institute, University of California, San Francisco, San Francisco, California 2 Department of Cellular and Molecular Pharmacology, Howard Hughes Medical Institute, University of California, San Francisco, San Francisco, California 3 Department of Pharmaceutical Chemistry, Howard Hughes Medical Institute, University of California, San Francisco, San Francisco, California 4 Howard Hughes Medical Institute, University of California, San Francisco, San Francisco, California
ABSTRACT Mapping kinase-substrate interactions demands robust methods to rapidly and unequivocally identify substrates from complex protein mixtures. Toward this goal, we present a method in which a kinase, engineered to utilize synthetic ATPγS analogs, specifically thiophosphorylates its substrates in a complex lysate. The thiophosphate label provides a bio-orthogonal tag that can be used to affinity purify and identify labeled proteins. Following the labeling reaction, proteins are digested with trypsin; thiol-containing peptides are then covalently captured and non-thiol-containing peptides are washed from the resin. Oxidation-promoted hydrolysis, at sites of thiophosphorylation, releases phosphopeptides for analysis by tandem mass spectrometry. By incorporating two specificity gates—kinase engineering and peptide affinity purification—this method yields highconfidence substrate identifications. This method gives both the identity of the substrates and phosphorylation-site localization. With this information, investigators can analyze the biological significance of the phosphorylation mark immediately following confirmation of the kinase-substrate relationship. Here, we provide an optimized version of this technique to further enable widespread utilization of this technology. Curr. Protoc. C 2010 by John Wiley & Sons, Inc. Chem. Biol. 2:15-36 Keywords: phosphorylation r chemical genetics r analog specific kinase r kinase substrate identification r thiophosphate labeling
INTRODUCTION Members of the kinase superfamily are integral components of many signaling pathways. Kinases play pivotal roles in regulating growth and development, and are misregulated in many diseases, including cancer. Kinases transduce extracellular signals by altering the functions of substrate proteins through phosphorylating specific sites on these proteins. It is estimated that one-third of all proteins are phosphorylated, and therefore regulated by kinases; however, there are only 518 known human kinases (Cohen, 2001; Manning et al., 2002). Therefore, many kinases must phosphorylate multiple substrates. The mapping of these relationships is complicated by the shared enzymology of kinases, and, therefore, it is extremely difficult to assign a phosphorylation site to a specific kinase.
Current Protocols in Chemical Biology 2: 15-36, February 2010 Published online February 2010 in Wiley Interscience (www.interscience.wiley.com). DOI: 10.1002/9780470559277.ch090201 C 2010 John Wiley & Sons, Inc. Copyright
KinaseSubstrate Mapping
15 Volume 2
A NH 2
ATP N
ATP␥S
N
HN N
O
O
N
O
HO P O P O P O OH
OH
O
O
N
O
HS P O P O P O
OH OH
N
N
O
OH
OH
OH
O
OH
O OH
O
OH
P OH OH
N
OH
O O P SH OH
OH
NH2 N
N N
N
O
H
OH
O
O P O P
O H
kinase substrate
H H OH
OH
OH
HN
O
phosphorylated kinase substrate
O P OH OH
N
HN N O
O OH
N
O
P O P O P O OH
1
O O
NO 2
O S
H H OH
OH
O
thiophosphorylated kinase substrate
O O P SH
OH
OH
1. Optimize ATP analogs O
OH
R=
Candidate Substrate
H
1
2
3
NO 2
O
2
2. Candidate substrate approach
N
P S OH
O
O
R
N
O
OH OH
H 3C
OH
AS-kinase
HS
P SH OH
O
O P O P
O H
B
O
N
H
wt kinase
O
N
N
O
P
OH OH
+
WB α thiophosphate ester NO 2
3 O
WB: α thiophosphate ester
S
O
IP: Candidate substrate (1-3) WB: α thiophosphate ester
P OH OH
1. Substrate/ Phospho Site ID
2. List of putative substrates
O O
SH
P SH OH
Trypsin O O
P
OH OH
affinity purification by covalent capture and release SH
LC MS/MS analysis
Figure 1 Thiophosphopeptide purification scheme. (A) Wild-type (wt) kinase utilizes ATP to phosphorylate its substrates. In contrast, an engineered analog-specific (AS) kinase has a new active-site pocket that allows it to accept an unnatural bulky ATP analog. This “lock and key” allows the AS kinase to transfer the thiophospho group to its substrates. (B) Detail of the two different affinity-purification methods. The thiophosphorylated substrates are found in a background of phosphorylated and unlabeled proteins. In the first technique, the thiophosphorylated proteins are reacted with a thiol-specific alkylating agent that generates a bio-orthogonal thiophosphate ester. Labeled proteins are detected by a thiophosphate ester-specific antibody. In the second approach, the lysate is first digested to generate tryptic peptides. Thiol-containing peptides are then captured by reaction with iodoacetyl agarose beads, and all non-thiol-containing peptides are washed away. The remaining peptides are treated with Oxone, which releases thiophosphate ester-linked peptides by spontaneous hydrolysis.
The authors have developed bio-orthogonal chemical reactions that allow the assignment of a particular phospho-site to a specific kinase (Shah et al., 1997; Bishop et al., 2000). In this approach, the active site of the kinase of interest (KOI) is engineered by creating a new active-site pocket that allows the kinase to accept a bulky ATP analog (Fig. 1A). This engineered “lock and key” provides selectivity for the KOI, as the engineered kinase will accept the N6 -substituted ATP analog, whereas the vast majority of wild-type kinases cannot use these analogs.
KinaseSubstrate Mapping
In order to identify the substrates of a particular enzyme, many of the most widespread proteomic methods rely on affinity-based approaches to enrich for proteins of interest. To apply this approach to kinase-substrate mapping, a modified gamma-phosphate analog (thiophosphate) capture-and-release strategy for thiophosphate purification has been developed. The ATP analog used to identify direct kinase substrates is modified in two
16 Volume 2
Current Protocols in Chemical Biology
A
B HN N
N
N
N
O
O
O
O P O P
O H H
H H OH
OH
OH
O P SH
OH
OH
AS-kinase
NO 2
O
O O
OH
O
P
SH OH
O
H 3C
P
S
O
O
O
SH OH
O
PNBM
NH2
N
N O
O H
SH
NO 2
N
N
H
H
OH
H OH
O
O
P
O P
OH
OH
O O
P
OH
OH
SH
SH
O
wt kinase
OH
NO 2
P S OH
O
S
O O
P OH OH
P OH OH
O O
P OH OH
WB α thiophosphate ester
Labeling with A*TPγS analogs
C
O
O
O
P
P
SH OH
O
trypsin
O
O
O
I
S OH
O
O
P
O
S OH
O
H 2O
SH
O O
O
O
P OH OH
O
1. Digestion
P
OH OH
S
I
O
O
S OH O OH
4. LC-MS/MS
S
2. Capture
P OH OH
3. Release
Figure 2 (A) Labeling of a cell lysate by incubating AS kinase and N6 -substituted ATPγS analogs generates thiophosphorylated and phosphorylated proteins. (B) Alkylating thiol-containing proteins generates the bio-orthogonal thiophosphate ester. Thiophosphate ester–labeled proteins are detected by a thiophosphate ester–specific antibody (clone 51-8). (C) Thiol-containing peptides are captured by reaction with iodoacetyl beads. During washing, all other non-thiol-containing peptides are washed away. The bound peptides are treated with Oxone to convert the thiol group to a sulfoxide. The presence of an electrophilic phosphoester in the thiophosphate ester-linked peptides catalyzes the spontaneous hydrolysis of these peptides and, after concentration, these peptides are analyzed by tandem LC-MS/MS.
ways: a bulky group is added in the N6 position, providing exclusive recognition by the engineered kinase of interest, and the gamma phosphate is replaced with a thiophosphate moiety (Fig. 1). The modified analog-specific (AS) kinase will use these analogs most efficiently; therefore, the substrates of the AS kinase will be uniquely labeled with the thiophosphate affinity tag. This tag can then be used to purify and identify the tagged proteins (Fig. 1B). The authors have recently published two different approaches for the affinity purification of thiophosphorylated substrates (Allen et al., 2007; Blethrow et al., 2008). Here, updated protocols for the application of these techniques are presented, with demonstration of how they can be used in a complementary manner to rapidly identify the substrates of one kinase from a complex protein mixture. In the first technique the thiophosphorylated proteins are reacted with a thiol-specific alkylating agent that generates a bio-orthogonal thiophosphate ester. The tagged proteins are affinity purified using a thiophosphate ester-specific antibody 51-8 (Allen et al., 2005, 2007; also see Figure 2B). The antibody is able to distinguish between a thiophosphate ester and a cysteine thioether, and in this way tagged substrates can be specifically visualized and identified by immunoblotting (IB) and immunoprecipitation (IP). This approach cannot reliably allow for the identification of the site of phosphorylation because the alkylating agent can prevent normal collision-induced dissociation of the modified peptide. Given that it is currently difficult to produce the monoclonal antibody 51-8 in sufficient amounts for IP, this technique is now routinely used for two optimization steps:
KinaseSubstrate Mapping
17 Current Protocols in Chemical Biology
Volume 2
(1) to determine the optimal N6 -substituted ATPγS analog for a KOI, and (2) to optimize N6 -substituted ATPγS, ATP, and GTP concentrations to achieve the lowest background labeling in cell lysates. Once labeling conditions are optimized, a candidate protein may also be verified as a KOI substrate by immunoprecipitating the candidate substrate and immunoblotting with monoclonal antibody 51-8. After optimization of labeling conditions using the thiophosphate-specific antibody, the second technique provides unambiguously assigned sites of modification on substrates of the engineered kinase. In this technique, the modified peptides are directly affinity purified, and then subjected to tandem mass spectrometry to identify the parent protein and to assign phosphorylation sites (Blethrow et al., 2008; also see Figure 2B). To accomplish the affinity purification, the labeled protein lysate is first digested and the peptides are covalently captured by reacting the digested peptides with iodoacetyl agarose beads, thus capturing all thiol-containing peptides. Unbound peptides are washed away, and the beads are then treated with Oxone to oxidize the sulfur in the thiophosphate ester to a sulfoxide. The peptides linked by a thiophosphate ester bond to the agarose beads are then released by spontaneous hydrolysis, while those linked by a cysteine thioether are retained on the resin. The Basic Protocol provides the most updated and optimized form of these techniques and includes detailed positive and negative controls to ensure their successful implementation by any lab.
STRATEGIC PLANNING Engineering the Kinase of Interest To apply this method, the KOI must be able to catalytically transfer thiophosphate to substrate proteins and tolerate the mutation of the normally bulky gatekeeper to a smaller residue (Fig. 3). Most kinases should be able to use ATPγS as a phosphorylation cofactor; in preliminary experiments, 13 out of 15 kinases were able to use ATPγS to thiophosphorylate substrate proteins (Allen et al., 2007). To determine whether a KOI can transfer thiophosphate, the thiophosphate-ester-specific antibody is used to rapidly determine the thiophosphorylation state of a substrate protein after incubation with ATPγS and the KOI. The first step is therefore to set up an in vitro kinase reaction with the KOI, substrate protein, and ATPγS, followed by immunoblotting with the thiophosphateester-specific antibody (Support Protocol 1). The bioinformatic technique for identifying the gatekeeper as well as the generation of an AS kinase have been described previously (Buzko and Shokat, 2002; Blethrow et al., 2004; Gregan et al., 2007). As described in Buzko and Shokat (2002), the kinase database is a online tool that can be used to easily find the gatekeeper residue for a kinase (http://sequoia.ucsf.edu/ksd/). After identifying the gatekeeper residue (equivalent to I338 in v-Src), conventional site-directed mutagenesis techniques can be used to mutate the gatekeeper to either a glycine (analog sensitive–1, as1) or alanine (analog sensitive–2, as2). These space-creating mutations will allow the kinase to utilize bulky ATP analogs (Buzko and Shokat, 2002; Gregan et al., 2007).
KinaseSubstrate Mapping
Certain protein kinases lose catalytic activity upon mutation of their gatekeeper residue to Gly or Ala. In such cases, second-site mutations can be used to rescue activity of the kinase (Zhang et al., 2005). Identification of second-site suppressor mutations has been successfully executed through bioinformatic analysis or genetic selection. Mutations at a few positions were found to rescue activity in multiple kinases and thus can be considered as top candidate sites for rescue mutations (C. Zhang and K. Shokat, unpub. observ.). For example, one position is immediately N-terminal from the conserved DFG motif (DFG-1), and Ala was found beneficial for kinase activity at this position. If the natural residue in a kinase is different from Ala at DFG-1, one can change it to Ala in an attempt to rescue activity of the kinase. Another position is typically 11 residues N-terminal from the DFG motif (DFG-11), and Leu has been found to be beneficial for kinase activity in
18 Volume 2
Current Protocols in Chemical Biology
A
HN NH2
N N
N
O
OH
O
O P O P
O H H
H H OH
OH
OH
NH 2
N
N N
N
N
O
O
H H
OH
OH
O
O P O P
O
O P OH
H H OH
OH
OH
N
O O P SH
CDK
N
AS-kinase
AS-kinase
wt kinase
N N
OH
asCDK
asCDK INMPPI
B
Src
PKA
asSrc
asPKA
asSrc INAPPI
asPKA INMPPI
Aurora
ERK
asAurora
asERK
asAurora INAPPI
asERK INAPPI
Figure 3 (A) The gatekeeper residue is shown in CDK1 as a phenylalanine above the bound ATP in the CDK1 active site. After mutation to the much smaller glycine, a pocket is opened that allows for a bulky substituent to bind (pictured as asCDK1). A “bumped” (N6 -substituted) ATP analog or inhibitor (in this case 1NMPP1) can now bind to the engineered kinase by utilizing the extra space above the N7 position. (B) The active sites of SRC, PKA, aurora, and ERK are shown along with their gatekeeper residues and the successful mutation of these bulky residues to smaller ones. The mutation of the gatekeeper to a smaller residue creates a similar pocket in each of these four kinases. These kinases are shown with ATP bound or with bumped inhibitors that target these engineered kinases, but not the wild type. The similarity in the engineered pocket among disparate kinases demonstrates the wide applicability of this method.
general here. If different from Leu, the residue at DFG-11 could be mutated to Leu for rescue of activity. It should be noted that successful rescue mutations vary from kinase to kinase, and that there seems to be no universal rescue mutation. Once the AS kinase protein has been expressed and purified, the kinase assay is repeated to identify the N6 -substituted ATP analog that is preferentially utilized by the AS-KOI (Alternate Protocol 1 and Fig. 1B).
Lysate Optimization Optimization of N6 -substituted ATPγS analog concentration in each specific lysate is also required to ensure the lowest levels of background and therefore the highest confidence in the identified substrates. There are several different parameters required to optimize the labeling conditions. Although the N6 -substituted ATPγS analogs are preferentially used by the engineered AS kinase, at high enough concentrations other kinases can also utilize these analogs. Providing sufficient unmodified ATP and GTP ensures that these kinases are occupied, and will therefore decrease the nonspecific use of the N6 -substituted ATPγS analogs. In addition, each lysate will have a different N6 -substituted ATPγS analog that will produce the lowest level of background. The conditions that give the highest signalto-noise ratio should be determined empirically by varying the levels of AS kinase, N6 -substituted ATPγS analog, ATP, and GTP in the presence of AS kinase. Utilizing the
KinaseSubstrate Mapping
19 Current Protocols in Chemical Biology
Volume 2
thiophosphate-ester-specific antibody to detect background thiophosphorylated proteins is the best method for lysate optimization. The amount of kinase is critical to the efficiency of the labeling. When using a recombinant form of the AS kinase, up to 1% (by total weight) of the AS kinase can be added [e.g., 10 μg of AS kinase in 1 mg of total lysate (100 μl of 10 mg/ml lysate)]. If using a transfected cell line, then the level of kinase can be varied by using different expression vectors, but, usually, the highest level of expression is necessary to achieving the best signal-to-noise ratio. Varying the levels of ATP (50 to 200 μM), GTP (1 to 3 mM), and N6 -substituted ATPγS analog (100 to 500 μM) should allow determination of an optimal set of reaction conditions in which the lowest background and the highest levels of AS-specific labeling is observed (Support Protocol 2).
Generating Positive-Control Peptides and Protein It is useful to generate a hyper-thiophosphorylated control protein [myelin basic protein (MBP)] and control peptide (GSK3 substrate peptide CREB), which can be accomplished using commercially available materials (Support Protocol 3). The labeled protein will serve as a control for the retention of the thiophosphate modification during the digestion steps, and both controls will serve to ensure that the covalent capture, release, and mass spectrometric analysis of thiophosphorylated peptides are working properly. During the labeling procedure, two negative controls are essential—minus N6 -substituted ATPγS analog and minus AS kinase. These controls will provide information about which background substrates are detected and therefore excluded from the list of identified substrates. Using these controls to troubleshoot and optimize each step of the procedure is critical to the success of the protocol. BASIC PROTOCOL
DIGESTION AND COVALENT CAPTURE OF THIOPHOSPHORYLATED PEPTIDES In order to identify specific substrates of an engineered kinase, several steps must be taken to engineer the kinase to accept N6 -substituted ATPγS analogs, and then to optimize the lysate for AS-kinase-specific labeling. Once a labeled lysate is produced, thiophosphorylated proteins are digested to the peptide level, thiol-containing peptides are covalently captured, and thiophosphate ester–linked peptides are oxidatively hydrolyzed from iodoacetyl agarose beads. The phosphopeptides that are released from the resin are then analyzed by tandem mass spectroscopy to identify the substrate proteins, as well as the site of phosphorylation.
Materials
KinaseSubstrate Mapping
2× denaturation buffer (see recipe) 1 M tris(2-carboxyethyl)phosphine (TCEP) in H2 O; store up to 6 month at −80◦ C Urea (99% pure); store indefinitely at room temperature Lysate to be analyzed (Strategic Planning) ATPγS analog (Support Protocol 4; several N6 -substituted ATPγS analogs are also available from Biolog, http://www.biolog.de) AS kinase of interest (KOI: typically prepared in 100 mM Tris·Cl, pH 7.5/150 mM NaCl/1 mM DTT, which may be augmented with other reagents, depending on the kinase; see published literature) Controls (these controls will ensure that all the steps of this protocol are working correctly; the MBP alone and MBP plus lysate will lead to recovery of only one peptide (see Anticipated Results); the lysate controls will provide a list of nonspecific substrates): 100 pmol of hyper-thiophosphorylated MBP (see Support Protocol 3)
20 Volume 2
Current Protocols in Chemical Biology
Unlabeled lysate: same lysate used for labeling, with no kinase or ATP added (a normal starting point is to use 1 mg total protein lysate in 100 μl total volume of 100 mM Tris·Cl, pH 7.5/150 mM NaCl/1 mM DTT/10 mM MgCl2 with protease inhibitors) Unlabeled lysate plus 100 pmol hyper-thiophosphorylated MBP Lysate plus ATPγS analog minus kinase An additional covalent capture reaction control (see step 7) with CREB peptide (see Support Protocol 3) Experimental sample: lysate plus ATPγS analog plus AS kinase (the amount of AS kinase should be ∼1% with respect to the mass of protein in the lysate, e.g., 10 μg AS kinase/1 mg protein; using the thiophosphate ester–specific western blot, looking for which conditions give the least nonspecific background and best labeling can optimize the ratio; Shah et al., 1997) 50 mM NH4 HCO3 Trypsin (Promega, cat. no. V5113) 2.5% (v/v) trifluoroacetic acid (TFA) 0.1% TFA/50% acetonitrile in H2 O (store indefinitely at room temperature) 0.1% TFA in H2 O (store indefinitely at room temperature) Iodoacetyl agarose beads, 50% slurry (Pierce; store up to 6 months at 4◦ C) 200 mM HEPES, pH 7.0 (store up to 4 months at 22◦ C) 50% (v/v) acetonitrile/50% (v/v) 20 mM HEPES, pH 7.0 (store up to 4 months at 22◦ C) 5 mg/ml bovine serum albumin (BSA) Acetonitrile 50% (v/v) acetonitrile/50% (v/v) H2 O (store indefinitely at room temperature) 5 M NaCl (store indefinitely at room temperature) 5% (v/v) formic acid (store up to 3 months at room temperature) 1 M dithiothreitol (DTT; store up to 6 months at −80◦ C) Oxone (DuPont) Siliconized microcentrifuge tubes 55◦ C water bath C-18 Sep Pak column (Sep Pak Classic cartridge; total volume, 0.5 ml; Waters; store indefinitely at room temperature in desiccator); alternatively use Oasis SPE (Waters) Small disposable columns (Isolute SPE Accessories Double Fritted Column 120-1021-A and Single Fritted Res 120-1111-A; Biotage, http://www.biotage.com/) 10- and 100-μl C-18 ZipTips (Millipore) QSTAR Elite Mass spectrometer (Applied Biosystems) or other tandem LC MS/MS capable mass spectrometer (also see Carr and Annan, 1996) Mass spectrometry analysis software (Carr and Annan, 1996) Additional reagents and equipment for mass spectrometry (Carr and Annan, 1996) Digest labeled protein lysate 1. In a siliconized microcentrifuge tube, prepare 500 μl complete (1×) denaturation buffer by mixing the following: 250 μl 2× denaturation buffer 5 μl 1 M TCEP (10 mM final) 68 μl H2 O 240 mg urea (8 M final) Making the denaturation buffer fresh will limit carbamate formation and will ensure that the TCEP is fully reduced. Using a non-thiol reducing agent will reduce non-peptide side reactions with the iodoacetyl beads. Performing the digestion in siliconized tubes will ensure that minimal sample is lost to nonspecific adsorption on the tube walls. Current Protocols in Chemical Biology
KinaseSubstrate Mapping
21 Volume 2
2. Add 1× denaturation buffer to each sample or control (lysate alone, lysate plus thiophosphorylated MBP, lysate plus ATPγS minus AS kinase, and lysate plus ATPγS plus 1% by weight AS kinase) such that the final concentration is 6 M urea (20 μl of lysate should be dissolved in 60 μl of 1× buffer). Incubate the sample at 55◦ C for 1 hr, then cool the sample at room temperature for 10 min. Do not incubate the sample with an alkylating agent such as iodoacetamide. This will react with the thiophosphate group, thereby destroying the reactivity of the tag.
3. Dilute the experimental sample by adding 50 mM NH4 HCO3 such that the final concentration of urea is 2 M, then add 1 M TCEP to a final concentration of 10 mM. 4. To both samples and controls, add trypsin in a 1:50 to 1:10 ratio by weight (e.g., 1:10 would be 10 μg trypsin for 100 μg of lysate protein, to digest all proteins to the peptide level). As no iodoacetamide is used in this protocol, high concentrations of urea (2 M) and reducing agent (10 mM TCEP) are used to ensure that proteins in the lysate remain unfolded during the digestion. One easy way to optimize the amount of trypsin and the incubation time is to add BSA to the lysate and then analyze the digested peptides to determine the sequence coverage. Coverage of at least 60% to 80% of BSA or ∼30 to 35 peptides should be accomplished.
5. After incubation at 37◦ C for 6 hr to overnight, acidify the sample by adding 2.5% TFA to attain 0.1% TFA. The thiophosphate modification is very labile to acid-promoted hydrolysis, so do not acidify to 2% TFA, as is commonly done.
6. Wash a C-18 Sep Pak (or a Oasis HLB) column with 10 ml of 0.1% TFA/50% acetonitrile in water, followed by 10 ml of 0.1% TFA in water. Load the acidified sample onto the column and pass it through the column five times. Wash the column with 10 ml of 0.1% TFA in water. Elute the peptides with 1 ml of 0.1% TFA/50% acetonitrile in water. Concentrate to near dryness using a SpeedVac evaporator. Different solid-phase extraction columns can lead to differential retention of peptides. Either type of column (the more hydrophobic C-18 as well as the more hydrophilic Oasis HLB type column) can be used, however, each lysate will require optimization. Do not take the sample to dryness, as the peptides may then be difficult to resuspend. The peptides should be in ∼50 μl after evaporation so that the correct reaction volume can be attained.
Covalently capture thiophosphorylated peptides 7. Prepare the iodoacetyl agarose beads for each sample by pipetting 100 μl of the 50% resin slurry into a siliconized 0.5-ml tube. Microcentrifuge 30 sec at 10,000 × g to collect the beads at the bottom, and remove the supernatant. Wash with 200 μl of 200 mM HEPES, pH 7.0, microcentrifuge again as before, and remove supernatant. Add 150 μl of 50% acetonitrile/50% 20 mM HEPES, pH 7.0, then add 5 μl of 5 mg/ml BSA, mix by vortexing briefly, and incubate 10 min in the dark to block the beads. The beads are supplied in an acidic buffer; washing with 200 mM HEPES, pH 7.0, is critical to achieving the correct final pH. The addition of BSA blocks the agarose beads and prevents sample loss by nonspecific adsorption. Mixing the slurry for 5 min with gentle rocking before pipetting should yield a homogeneous slurry.
KinaseSubstrate Mapping
8. Optional: Set up an additional covalent capture reaction by directly adding several (10 to 50) pmol of thiophosphorylated CREB peptide to the beads prepared in 50% acetonitrile/50% 20 mM HEPES, pH 7.0/BSA as described in step 7.
22 Volume 2
Current Protocols in Chemical Biology
9. Adjust the pH of the digested labeled peptide mixture (from step 6) by adding 200 mM HEPES pH 7.0, to a final concentration of 20 mM (for example, add 15 μl of 200 mM HEPES, pH 7.0, 10 μl water, and 75 μl acetonitrile to 50 μl digested peptides, for a final volume of 150 μl). Spin the beads down again as described in step 7, remove the supernatant, and add each sample to one tube of bead mixture, followed by room temperature incubation in the dark with gentle rocking for 12 to 16 hr. Adding the additional labeled peptide control at this point will simplify troubleshooting later. To prevent light from generating radicals within the sample, wrap the tubes in aluminum foil.
10. Prepare one disposable column for washing and elution of each reaction by washing the empty column with 1 ml 50% acetonitrile/50% water, and then with 1 ml water. At this point, either of two methods can be used, a column method or a spin method. The protocol is the same in each case; however, the column method is detailed below. For the spin method, all the washes should be performed by addition of the same volume followed by a short 30-sec spin at 10,000 × g.
11. Add the entire reaction to the top of the column and let it drain into a 1.5-ml microcentrifuge tube. Rinse the reaction tube with 100 μl 50% acetonitrile/50% 20 mM HEPES, pH 7.0, being sure to resuspend all the beads, and add the residual beads to the column. 12. Wash the beads by successively adding 1 ml of each wash liquid (see below; apply in the order depicted) to the top of each column, then push the wash liquid through using a 1000-μl pipet tip. a. b. c. d.
H2 O 5 M NaCl 50% acetonitrile/50% H2 O 5% formic acid If the level of wash liquid drops below the top of the beads, the recovery of phosphopeptides will be severely reduced. In addition, the order of washing is critical to the success of the washing steps.
13. Prepare a solution of 10 mM DTT from 1 M stock, add 1 ml to the column, allow it to drain half way, and then incubate for 10 min with the solution in the column. The addition of DTT to the beads will react with any available iodoacetyl groups left on the beads and will ensure that no I2 forms during the oxidation step, which can iodinate tyrosine residues (unpub. observ.).
14. Prepare a fresh solution of 1 mg/ml Oxone, pH ∼3.5, in water (if pH is not correct, more Oxone can be added) and add 100 μl of this solution to the column, being sure to resuspend the beads. After draining the column, add an additional 100 μl of 1 mg/ml Oxone and incubate for 10 min. Be sure to resuspend the beads at this point to ensure the complete oxidation of the thiophosphate ester.
15. Immediately desalt and concentrate the phosphopeptides with 10-μl C-18 ZipTips. a. Wash the ZipTip three times, each time with 0.1% TFA/50% acetonitrile in water, then three times, each time with 0.1% TFA/water. b. Attach the ZipTip to the end of a 200-μl pipet tip and draw the sample through the tip using the pipettor at a setting of 100 μl, passing the sample through the ZipTip a total of four times.
KinaseSubstrate Mapping
23 Current Protocols in Chemical Biology
Volume 2
c. Wash twice, each time with 20 μl of 0.1% TFA in water and elute by washing three times, each time with 20 μl of 0.1% TFA in 50% acetonitrile/water. Passing the entire sample through the ZipTip multiple times ensures the retention of lowabundance peptides. The 10-μl ZipTip should fit on the end of a 200- or 1000-μl pipet tip to simplify the passage of the entire sample across the C-18 column.
16. Concentrate the sample to 10 μl on a SpeedVac evaporator and analyze 5 μl by tandem mass spectrometry (e.g., using a QSTAR ELITE that includes coupled liquid chromatography on a reversed phase C-18 column utilizing a 3% to 32% acetonitrile gradient in 0.1% formic acid). Analyze the phosphopeptides in positive-ion mode. Acquire mass spectra for 1 sec. For each mass spectrum, select the two most intense multiply charged peaks for generation of subsequent collision-induced dissociation MS. Carr and Annan (1996) describes general procedures in mass spectrometry. This analysis will yield the identity of both the substrate protein and the phosphorylation site.
17. Analyze the data by centroiding using Analyst QS software or another appropriate software. Search the MS/MS spectra against the entire UniProt database utilizing Protein Prospector or another appropriate software. Typically, we select no constant modifications and allow the following variable modifications: phospho: serine, threonine, oxidized: methionine. Protein Prospector is available free of charge online: http://prospector.ucsf.edu. SUPPORT PROTOCOL 1
KINASE REACTION WITH ATPγS FOLLOWED BY IMMUNOBLOTTING UTILIZING THIOPHOSPHATE-SPECIFIC ANTIBODY Detection of thiophosphorylated substrate proteins is a critical analysis tool to determine whether this technique will work for a KOI in a particular lysate. An in vitro kinase reaction with a KOI, substrate, and ATPγS is used here to determine if the kinase will use ATPγS to thiophosphorylate substrate proteins. If the kinase is able to use ATPγS to thiophosphorylate substrate proteins, the band corresponding to the substrate should be immunoreactive with the thiophosphate ester–specific antibody (51-8).
KinaseSubstrate Mapping
Materials 1× HEPES-buffered saline (HBS; see recipe for 10×) or other kinase reaction buffer suitable for the KOI 1 M MgCl2 Kinase substrate (or general kinase substrate, e.g., myelin basic protein or histone H1) Kinase of interest (KOI) N6 -substituted adenosine 5 -[γ-thio]triphosphate (exclusively from Biolog: http://www.biolog.de), store 10 mM stock in aliquots at –80◦ C for up to 1 year and avoid freeze-thaw cycles p-nitrobenzyl mesylate (PNBM; exclusively from Epitomics, http://www.epitomics.com); store solid for up to 1 year at 4◦ C (50 mM stock in DMSO should be prepared fresh just before use) 5× sample buffer (Gallagher, 2006) 5% skim milk in TBST (see recipe for TBST) Primary antibody: thiophosphate ester rabbit monoclonal antibody, clone 51-8 (exclusively from Epitomics, http://www.epitomics.com; store up to 1 month at 4◦ C or indefinitely at −20◦ C)
24 Volume 2
Current Protocols in Chemical Biology
Secondary antibody: anti-rabbit HRP-conjugated antibody (Epitomics, http://www.epitomics.com); store up to 1 month at 4◦ C or indefinitely at −20◦ C ECL detection system Additional reagents and equipment for SDS-PAGE (Gallagher, 2006) and immunoblotting (Gallagher et al., 2008) 1. On ice, prepare a master mix of 1× HBS (or other kinase reaction buffer suitable for the KOI), 10 mM MgCl2 (add from 1 M MgCl2 stock), substrate (or general kinase substrate at 1 mg/ml), and KOI at between 50 and 200 ng per kinase reaction. Order of addition: H2 O, HBS, MgCl2 , substrate, KOI. These general conditions will work for most kinases, but check the literature for kinase-specific reaction conditions. To determine the optimal substrate concentration, run a kinase assay at several concentrations of the substrate; 1 mg/ml is a good starting point.
2. Label four tubes A to D and aliquot 27 μl of master mix into each tube. Tubes A to C are controls for nonspecific alkylation of cysteine residues by the thiolspecific alkylating agent PNBM.
3. Add 3 μl 10 mM N6 -substituted ATPγS (final concentration of 1 mM) to two of the samples (samples C and D). Flick the tubes to mix, and then briefly centrifuge to collect the kinase reaction in the bottom of the tube.
4. Let the reaction proceed for 30 min at room temperature (or suitable temperature for the KOI). 5. Add 1.5 μl of 50 mM PNBM to samples B and D to yield a final concentration of 2.5 mM. Allow the alkylating reaction to proceed for 1 hr at room temperature. If desired, the kinase reaction may be quenched by adding EDTA to 20 mM. The 50 mM PNBM should be prepared immediately before adding to the samples by dissolving PNBM in DMSO (50 mM is equal to 12 mg/ml).
6. Add 5× sample buffer to the samples and run the samples on a SDS-PAGE gel (Gallagher, 2006), then transfer to nitrocellulose or PDVF membrane (Gallagher et al., 2008). Block the membrane for 1 hr at room temperature with 5% skim milk in TBST. 7. Incubate the blot overnight at 4◦ C with a 1:5000 dilution (in 5% skim milk/TBST) of thiophosphate ester rabbit monoclonal antibody, clone 51-8. Incubating the blot with the antibody for 1 to 3 hr at room temperature can also lead to detection of modified proteins, although with lower sensitivity.
8. Wash the blot four times with TBST, then incubate for 1 hr with anti-rabbit HRP secondary antibody. Wash four times with TBST again, and then add ECL reagent to visualize PNBM-alkylated/thiophosphorylated proteins.
IDENTIFYING OPTIMAL N6 -SUBSTITUTED ATPγS ANALOG Each analog-specific kinase will have a slightly different substrate specificity. To determine the preferred N6 -substituted ATPγS, each kinase must be tested with several ATP analogs. Following this protocol to perform a kinase assay with several N6 -substituted ATPγS analogs will provide the optimal ATP analog for further studies to identify substrates. For materials, see Support Protocol 1.
ALTERNATE PROTOCOL 1
KinaseSubstrate Mapping
25 Current Protocols in Chemical Biology
Volume 2
1. Follow Support Protocol 1, utilizing several N6 -substituted ATPγS analogs in place of ATPγS to identify which analog is used optimally by the AS KOI. The optimal analog is the one that results in the most substrate labeling as detected by thiophosphate ester rabbit monoclonal antibody, clone 51-8. Wild-type KOI should not be able to utilize this N6 -substituted ATPγS analog. SUPPORT PROTOCOL 2
THIOPHOSPHORYLATION OF A CANDIDATE KINASE SUBSTRATE IN CELL LYSATE The following method is used to determine if a candidate protein is a substrate of a KOI in a complex protein mixture. First, the AS KOI is transiently expressed in cells and activated, then cell lysate is prepared and N6 -substituted ATPγS is added to initiate the substrate labeling reaction. Thiophosphorylated proteins are alkylated, and the candidate substrate is immunoprecipitated using an antibody against that protein. If a band is detected by immunoblotting with thiophosphate ester rabbit monoclonal antibody, then the candidate is a substrate of the KOI.
Materials Appropriate mammalian cells Plasmid containing (wild-type or AS) KOI suitable for expression of kinase in mammalian cells Transfection reagents (e.g., Lipofectamine, FuGene; see manufacturer’s protocol for transfection of plasmid) Phosphate buffered saline (PBS), store indefinitely at 4◦ C 2× RIPA buffer (see recipe) Protease inhibitor cocktail without EDTA (Roche); store up to 1 year at 4◦ C Phosphatase inhibitor cocktail (Roche); store up to 1 year at 4◦ C ATP (Sigma; store solid indefinitely at −20◦ C) GTP (Sigma; store solid indefinitely at −20◦ C) N6 -substituted adenosine 5 -[γ-thio]triphosphate (exclusively from Biolog: http://www.biolog.de), store 10 mM stock in aliquots at –80◦ C for up to 1 year and avoid freeze-thaw cycles Disodium EDTA (Sigma), store solid indefinitely at room temperature p-nitrobenzyl mesylate (PNBM; exclusively from Epitomics, http://www.epitomics.com); store solid for up to 1 year at 4◦ C (50 mM stock in DMSO should be prepared fresh just before use) Protein A or G Magnetic beads (Invitrogen, Dynabeads); store up to 1 year at 4◦ C 5× sample buffer (Gallagher, 2006) 10-cm cell culture dishes Cell scrapers End-over-end rotator Magnetic stand that holds 1.5-ml microcentrifuge tubes Additional reagents and equipment for SDS-PAGE and immunoblotting (Support Protocol 1) 1. Grow appropriate mammalian cells to 60% to 80% confluence on 10-cm dishes. 2. Transfect with plasmid encoding wild-type or AS KOI using transfection reagent of choice, following the manufacturer’s protocol. 3. Allow cells to grow for 24 to 48 hr. KinaseSubstrate Mapping
4. If appropriate, stimulate cells to activate KOI. For example, mitogen-activated protein kinase ERK may be activated by epidermal growth factor (Hazzalin et al., 1997).
26 Volume 2
Current Protocols in Chemical Biology
5. Remove medium and rinse once with 5 ml cold PBS. 6. Using a cell scraper, lyse and scrape cells on ice in 500 μl 1× RIPA buffer containing 1× protease inhibitor and 1× phosphatase inhibitor. 7. Centrifuge 10 min at 10,000 × g, 4◦ C to remove cell debris. Keep supernatant. 8. Add 100 μM N6 -substituted ATPγS, 100 μM ATP, and 3 mM GTP to each sample (order does not matter). Recommended concentration ranges for optimizing labeling conditions: N6 -substituted ATPγ S: 50 to 500 μM ATP: 50 to 200 μM GTP: 1 to 3 mM.
9. Allow thiophosphorylation of kinase substrates to occur for 20 min at room temperature. 10. Quench reaction by adding 500 μl of 1× RIPA buffer containing 40 mM disodium EDTA (final EDTA concentration, 20 mM) and 5 mM PNBM (final PNBM concentration, 2.5 mM), then alkylate with PNBM for 1 hr at room temperature on an end-over-end rotator. 11. Immunoprecipitate candidate substrate according to antibody manufacturer’s recommendation. Incubate overnight at 4◦ C. Save 20 μl of the immunoprecipitation supernatant for immunoblotting to check immunoprecipitation efficiency. The immunoprecipitation was effective if no protein remains in the supernatant.
12. Before use, wash appropriate (protein A or G) magnetic beads once with 1 ml 1× RIPA buffer, then resuspend in 1× RIPA buffer to create a 50% slurry. Add 40 μl of the 50% bead slurry to each sample and incubate 3 to 4 hr at 4◦ C on an end-over-end rotator. Choose A or G beads based on the type of Ig domain present in your antibody (see http://www.invitrogen.com).
13. Collect the beads using a magnet and magnetic stand according to the manufacturer’s instructions. Save 20 μl of the supernatant for immunoblotting to check immunoprecipitation efficiency. Wash the beads five times with 1 ml of 1× RIPA buffer containing 1× protease inhibitor and 1× phosphatase inhibitor. 14. Resuspend beads in 20 μl 1× RIPA buffer containing 1× protease inhibitor, 1× phosphatase inhibitor, and 1× sample buffer for SDS-PAGE. 15. Heat samples at 95◦ C for 2.5 min. 16. Load the samples onto an SDS-PAGE gel and proceed with immunoblotting as described in steps 6 to 8 of Support Protocol 1, using thiophosphate ester rabbit monoclonal antibody.
PREPARATION OF A THIOPHOSPHORYLATED POSITIVE CONTROL PEPTIDE AND PROTEIN The two control samples, a thiophosphorylated peptide and protein, can be prepared from easily purchased materials. A kinase reaction is used to make the thiophosphorylated substrates. MALDI-MS analysis can be used to confirm whether the CREB peptide has been thiophosphorylated. For the protein reaction, after labeling, the thiophosphorylated myelin basic protein (MBP) can be run on a denaturing polyacrylamide gel followed by an immunoblot using 51-8 (Support Protocol 1). The peptide can then be used to evaluate
SUPPORT PROTOCOL 3
KinaseSubstrate Mapping
27 Current Protocols in Chemical Biology
Volume 2
whether the covalent capture reaction is working properly. Labeled MBP can be added to a lysate as a control for the digestion and covalent capture reactions.
Materials 10× HEPES-buffered saline (HBS; see recipe) 1 M MgCl2 10 mM stock in H2 O of adenosine 5 -[γ-thio]triphosphate tetralithium salt (Sigma), store up to 1 year at −80◦ C Substrates: CREB peptide (KRREILSRRPS(p)YR; 2 nmol/μl), store up to 1 year at −80◦ C Dephosphorylated myelin basic protein (2.5 mg/ml) (Millipore) Purified active KOI or plasmid containing KOI suitable for expression of kinase in E.coli Recombinant GSK3β (Millipore) Liquid N2 0.1% TFA/50% acetonitrile in H2 O (store indefinitely at room temperature) 0.1% TFA in H2 O (store indefinitely at room temperature) OMIX 100-μl ZipTips (Varian) 1. On ice prepare two kinase reactions that each contain:
6 μl 10× HBS 0.6 μl 1 M MgCl2 (to 10 mM) 6 ml 10 mM ATPγS (to 1 mM final) Either 25 μl 2.5 mg/ml MBP (5 nmol total) or 18 μl 2 nmol/μl CREB peptide (5 nmol) 500 ng KOI H2 O to a final volume of 60 μl. Order of addition: water, HBS, MgCl2 substrate, KOI.
2. Add 500 ng recombinant GSK3β and incubate reaction at room temperature for 4 hr. Incubate for longer reaction times to ensure stoichiometric thiophosphorylation of MBP.
3a. For MBP labeling reactions: Aliquot the reaction containing MBP into 0.5-ml microcentrifuge tubes at 5 μl per tube, snap freeze in liquid N2 , and store at −80◦ C until needed 3b. For CREB labeling reactions: Use a large-capacity ZipTip (OMIX 100 μl) to desalt the CREB thiophosphorylation reaction, eluting into 200 μl 0.1% TFA in 50% acetonitrile. Concentrate the sample using a SpeedVac evaporator to 60 μl. Aliquot into 0.5-ml microcentrifuge tubes (5 μl/tube) and snap freeze them in liquid N2 . Store at –80◦ C for up to 3 months. Utilize the ZipTip according to product literature: wash two times with 0.1% TFA/50% acetonitrile, then two times with 0.1% TFA in H2 O. Pass the sample through the tip five times, being sure not to introduce any air into the tip. Wash two times with 0.1% TFA in H2 O and then elute two times with 0.1% TFA in 50% acetonitrile. SUPPORT PROTOCOL 4
KinaseSubstrate Mapping
CHEMICAL SYNTHESIS OF N6 -SUBSTITUTED ATPγS The synthesis of N6 -substituted ATPγS is presented to those interested in making additional analogs (Fig. 4). The synthesis will require significant knowledge of organic chemistry synthesis procedures and a well-equipped laboratory. This protocol will allow the synthesis of larger quantities of N6 -substituted ATPγS analogs, and the synthesis of those that are not available from Biolog. This synthesis can be found in the supplementary information in (Allen et al., 2007).
28 Volume 2
Current Protocols in Chemical Biology
R
Cl
1
2
N N HO
N
O OH
N HO
S (Na + )3 5
P O
7 1. H 2 O/DMF O
O
(Na + ) 2
O P O
O S
1. POCl 3
O
O–
OH
OH
N
O OH
O Cl
NH 2
O
O–
N
N
O P O P O
2. H 3 PO 4 , DBU OH
6
N N
NH
N
–
reflux, EtOH OH
O
3
O
O
4
N
N R-NH 2
R NH
P
(OPh)2
8
O
O
O
P P S (PhO)2 H O O
1. Couple 2. NaOH Deprotection
NH 2
NH 2
Cl
9
R
NH
N O HS P O–
O
O
O P O P O O–
N
O–
N
O OH
Figure 4
N
OH
Synthetic scheme for the production of N6 -substituted ATPγS analogs.
Materials 6-chloropurine ribonucleoside (Sigma, cat. no. C8276-56) Ethanol (Sigma) Alkylamine containing desired N6 modification: e.g., phenethylamine Triethylphosphate (TEP; Sigma) POCl3 (Sigma) H3 PO4 (Sigma) 1,8-diazabicyclo[5.4.0]undec-7-en (DBU) (Sigma) 2 M TEAB (see recipe) Trisodium thiophosphate (Sigma-Aldrich) 3-chloropropionamide (Sigma-Aldrich) Dimethylformamide (DMF) DOWEX 501-X8 ion exchange resin (pyridinium form) Pyridine, dry Methanol Tri-n-octylamine (Sigma-Aldrich) Dioxane, dry Diphenyl phosphorochloridate (Sigma-Aldrich) Tri-n-butylamine (Sigma-Aldrich) Diethyl ether Petroleum ether 0.2 M NaOH β-mercaptoethanol Oil bath Reflux condenser Buchi Rotovapor Model R-200 or equivalent rotary evaporator PTFE syringe filter (0.45 μm Pall Acrodisc) HiPrep 16/10 QFF anion-exchange columns (Amersham Biosciences) Peristaltic pump ACTA FLPC system (GE Healthcare) including gradient former
KinaseSubstrate Mapping
29 Current Protocols in Chemical Biology
Volume 2
Small molecule-capable LC-MS instrument (e.g., Waters; see Carr and Annan, 1996) Lyophilizer Filter paper Buchner funnel Small plastic column Round-bottom flasks High-vacuum source and high-vacuum manifold Additional reagents and equipment for mass spectrometry (Carr and Annan, 1996) 1. Dissolve 1 g (3.5 mmol) of 6-chloropurine ribonucleoside in 21 ml of ethanol while stirring. 2. Heat to at 95◦ C then add seven equivalents of the alkylamine desired to be the bulky substituent in the N6 position [in this case phenethylamine, 2.5 g (21 mmol)] and stir under a reflux condenser for 12 hr. 3. Remove solvent under reduced pressure in Buchi Rotovapor. 4. Add 150 ml of ethanol and place at 4◦ C for 5 hr. 5. Filter the crystals and wash with cold ethanol to obtain N6 phenethyl (or additional alkyl) adenosine.
Synthesize N6 phenethyl ADP 6. Dissolve 2 mmol of the N6 phenethyl (or additional alkyl) adenosine (0.76 g) in 5 ml of triethethylphosphate (TEP). 7. Cool to 0◦ C in an ice/NaCl bath while stirring. 8. Add 0.6 g (4 mmol) of POCl3 dropwise and continue to stir at 0◦ C for 2 hr. 9. In a separate container, dissolve 8 mmol of 1,8-diazabicyclo[5.4.0]undec-7-en (DBU) in 5 ml TEP, then add 8 mmol of H3 PO4. The order of addition is critical here: first dissolve the DBU then add the H3 PO4 .
10. Add the entire H3 PO4 DBU mixture to the mixture of the N6 -substituted adenosine and POCl3 . A yellow solid should form immediately.
11. Continue to stir for 2 min, then quench the reaction by adding 30 ml of 0.1 M triethylammonium bicarbonate (TEAB) and stir for 30 min. 12. Filter the mixture through a 0.45-μm PTFE syringe filter. 13. Load the mixture onto two HiPrep 16/10 QFF anion exchange columns in series using a peristaltic pump. 14. Using the ACTA FLPC system, run an 80-min gradient from 100% 0.1 M TEAB to 50% 2 M TEAB:50% 0.1 M TEAB. Collect 15-ml fractions. This step serves to purify ADP from AMP and ATP.
15. Analyze the fractions on a small-molecule-capable LC-MS instrument and pool those fractions containing the diphosphate. Carr and Annan (1996) describes general mass spectrometry procedures. KinaseSubstrate Mapping
16. Lyophilize fractions containing the diphosphate. N6 phenylethyl ADP: calculated mass 530.10 m/z.
30 Volume 2
Current Protocols in Chemical Biology
Synthesize disodium S-2-carbamoylethyl phosphorothioate 17. Dissolve 18.6 mmol of trisodium thiophosphate in 28 ml of water. 18. Dissolve 28 mmol of 3-chloropropionamide in 5.6 ml of DMF and add mixture to the trisodium thiophosphate mixture followed by stirring for 48 hr at room temperature. 19. Filter the reaction, add 100 ml of ethanol to the filtrate, and then cool to 4◦ C and incubate at that temperature overnight (16 hr). 20. Filter the precipitate, wash with ethanol, and dry to yield disodium S-2carbamoylethyl phosphorothioate.
Convert disodium S-2-carbamoylethyl phosphorothioate and N6 phenylethyl ADP (TEA salt) to pyridinium salt 21. Swell 5 ml of DOWEX 501-X8 ion exchange resin in 50 ml 5% pyridine/95% water. 22. Add the DOWEX 501-X8 ion exchange resin to a small plastic column and let the 5% pyridine run through until the pH is greater than 10. When the pH is above 10, all of the sites on the resin are occupied with pyridine.
23. Dissolve 2.0 mmol of disodium S-2-carbamoylethyl phosphorothioate (from step 20) or 0.3 mmol N6 phenylethyl ADP (TEA salt from step 16) in 10 ml (for the former) or 5 ml (for the latter) of 1:1 MeOH:H2 O. 24. Add the mixture to the top of the DOWEX 501-X8 ion exchange resin column (pyridinium form). 25. Elute by adding another 10 ml (for the S-2-carbamoylethyl phosphorothioate or 5 ml (for the N6 phenylethyl ADP TEA salt) of 1:1 MeOH:H2 O. 26. Dry in vacuo on a Buchi rotary evaporator to yield the pyridinium salt of S-2carbamoylethyl phosphorothioate or N6 phenylethyl ADP.
Synthesize O-diphenyl phosphoro S-2-carbamoylethyl phosphorothioate 27. Add 20 ml methanol to 2.0 mmol of the pyridinium salt of S-2-carbamoylethyl phosphorothioate (from step 26). 28. Add 2 mmol of tri-n-octylamine to convert to the mono(tri-n-octylammonium) salt. 29. Remove solvent in vacuo on a Buchi rotary evaporator. Repeat evaporation (two times) after addition of 10-ml aliquots of dry DMF. 30. Dissolve the residue in 14 ml of dry dioxane, then add 2.9 mmol diphenyl phosphorochloridate followed by 3.8 mmol of tri-n-butylamine, and stir for 2 hr at room temperature. 31. Remove solvent in vacuo on a Buchi rotary evaporator. Add 20 ml diethyl ether and, after dissolution, add 40 ml warm petroleum ether. 32. Place mixture at 4◦ C for 30 min until a higher-density oily mixture separates and can be decanted into a separate round-bottom flask 33. Dry the oily residue under high vacuum on a high-vacuum manifold for 1 hr to yield O-diphenyl phosphoro S-2-carbamoylethyl phosphorothioate.
Synthesis of N6 phenylethyl ATPγ S 34. Add 12 ml methanol to the pyridinium salt of N6 phenylethyl ADP (from step 26). 35. Add 0.32 mmol tri-n-octylamine and 0.32 mmol tri-n-butylamine; incubate at room temperature until complete dissolution occurs.
KinaseSubstrate Mapping
31 Current Protocols in Chemical Biology
Volume 2
36. Remove solvent in vacuo on a Buchi rotary evaporator. Dry the residue by repeated evaporation (three times) with 5-ml aliquots of dry pyridine. 37. Dissolve residue in 3.3 ml dry pyridine and add this to the dry residue of O-diphenyl phosphoro S-2-carbamoylethyl phosphorothioate (from step 33). 38. Stir at room temperature for 2 hr. A precipitate will form during this time.
39. Remove solvent in vacuo on a Buchi rotary evaporator. Add 20 ml 0.2 M NaOH then heat the reaction to 100◦ C for 10 min. 40. Quench the reaction by adding 10 ml DOWEX 501-X8 ion exchange resin (pyridinium form; preswelled as in step 21) and 0.4 ml β-mercaptoethanol. 41. Filter the reaction and then purify on the two HiPrep 16/10 QFF anion exchange columns exactly as described in steps 13 to 14. 42. Analyze fractions by LC-MS (also see Carr and Annan, 1996), then pool and lyophilize fractions containing the N6 phenylethyl ATPγS. Redissolving the dried fractions in 0.1 M TEAB and another round of lyophilization can be performed to concentrate and to produce a white solid. After lyophilization, the compounds should be assumed to be in the TEA form. To obtain accurate concentrations, the absorbance at 280 nm should be compared to several standard solutions of the N6 -modified adenosine.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps.
Denaturation buffer, 2× 200 mM NH4 HCO3 4 mM EDTA Store indefinitely at 22◦ C HEPES buffered saline (HBS), 10× 200 mM HEPES pH 7.4 1.5 M NaCl Store up to 6 months at room temperature RIPA lysis buffer, 2× 100 mM Tris·Cl, pH 8.0 150 mM NaCl 1% (w/v) NP-40 0.1% (w/v) SDS Store indefinitely at 4◦ C TBST 10 mM Tris·Cl, pH 7.5 150 mM NaCl 0.05% (v/v) Tween 20 Store up to 1 month at room temperature TEAB, 2 M KinaseSubstrate Mapping
Prepare 2 M TEAB by mixing 557.5 ml triethylamine (TEA; Sigma) with 1442.5 ml water. Bubble CO2 through this mixture until the TEA is dissolved.
32 Volume 2
Current Protocols in Chemical Biology
COMMENTARY Background Information Identifying the substrates of a kinase is a critical step toward understanding its biological role. Traditional approaches for kinase substrate mapping depend on knockout experiments followed by phosphopeptide enrichment and mass spectrometry. While useful, this method fails to identify all kinase substrates, as there exists redundancy in which a single substrate may be phosphorylated by multiple kinases. Methods for phosphopeptide enrichment include metal affinity chromatography using titanium oxide columns (Trinidad et al., 2008) and identifying new phosphorylated proteins by enriching thiophosphorylated peptides (Kwon et al., 2003). A quantitative mass spectrometry approach can more precisely define the effects of stimulation or knockout of a particular kinase. In an application of this technique, a SILAC labeling approach followed by TiO2 enrichment was used to determine the effects of growth factors on global phosphorylation status (Olsen et al., 2006). All of these approaches fail to specifically assign a phosphorylation site to a particular kinase, but are useful for defining the global changes in phosphorylation and for mapping novel phosphorylation sites. The approach described here allows for the labeling of the substrates of a single kinase in a chemical genetic approach, and can therefore unambiguously assign kinasesubstrate relationships. Originally, the N6 modified ATP analogs were radioactively labeled in the gamma-phosphate position. Putative substrates were then identified by coupling this approach with affinity-tagged libraries. This approach was successfully used to identify Cdk1 and Pho 85 substrates in S. cerevisiae (Ubersax et al., 2003; Dephoure et al., 2005). Although this engineered “lock and key” provides selectivity for a single kinase, the transfer of a radioactively tagged phosphate by the N6 -modified ATP substrate only assists in visualization and traditional biochemical purification of direct kinase substrates from tagged libraries that are not available in more complex eukaryotes. Introduction of a bio-orthogonal tag onto the substrate proteins of a particular kinase simplifies the affinity purification and identification of the substrates of any protein kinase in a complex lysate. This approach relies on two separate specificity gates: the kinase is engineered to accept bulky ATP analogs, and the substrates are tagged with a bio-orthogonal
phosphate analog. The affinity purification of the tagged substrates also depends on two specificity gates: only thiol-containing peptides are pulled down, and of those peptides that are covalently bound to the affinity matrix, only those linked by thiophosphate ester bonds are cleaved off the resin. These specificity elements allow for the purification of very low-abundance substrates, while also increasing the confidence in the identification of the identified substrates.
Critical Parameters Utilization of the N6 -modified ATP analogs by other kinases can lead to identification of false kinase-substrate relationships. It is therefore crucial to fully optimize the labeling reaction. Using a wide range of ATP, GTP, and N6 -modified ATP analogs during the lysate optimization steps will provide the best conditions for these reactions. Nonspecific binding of peptides to the agarose beads can also lead to the identification of false substrate proteins. Generating several positive and negative control samples will help control for the nonspecific purification of peptides. Several controls are essential for distinguishing background from AS kinase-specific labeling. First, it is essential to demonstrate the AS kinase dependence of purification of any phosphopeptides. When labeling is observed, or phosphopeptides are affinity purified from samples that do not contain AS kinase but do contain the N6 -modified ATP analog, these proteins must be subtracted from any list of putative substrates. When nonphosphorylated peptides are identified in the absence of AS kinase, it may indicate that the washing steps are not effective. Increasing the stringency of the wash steps by increasing the volume, or incubating the washes for 10 min, should lower the background signal. Secondly, the positive control thiophosphorylated peptide and protein are very useful for diagnosing the digestion and affinity-purification steps. In typical tryptic digests, complex lysates are reduced and alkylated before digestion to ensure the complete digestion of all proteins. The alkylation step with iodoacetamide is incompatible with this technique, thereby complicating the digestion of thiophosphorylated protein lysates. Also, the thiophosphate mark is susceptible to acid-promoted hydrolysis. It is therefore important to include the hyper-thiophosphorylated MBP as a digestion control to ensure the mark is retained on
KinaseSubstrate Mapping
33 Current Protocols in Chemical Biology
Volume 2
A P OH OH
O
O
O O
O
P OH OH
O
P OH OH
O O
P SH OH m/z=1796.96
m/z=1796.96
m/z=1891.93
C
B
y12 y11y10 y9
O O
y7 y6 y5
y2
N I V T P R T (Phospho) P P P S Q G K + 3
P
OH OH
b3 b4
b7
O O
P OH OH
y12 +2
m/z=1875.95
IVT
y2
y2 y11 +2 y10-H3PO4 a6+2
b4 b7+2 b4-NH3
y5
y9 +2
y10 +2 y11-H2O y6
y7
y12-H2O
Figure 5 (A) Raw data showing an expected MALDI analysis of thiophosphorylated CREB peptide. Both the singly phosphorylated (KRREILSRRPS(p)YR) (1795.84 MH+) and the thiophosphorylated (KRREILS(thiophosphate)RRPS(p)YR) 1891.986 (MH+) peptide are shown in the first pane. Analysis of the flow-through after covalent capture with iodoacetyl agarose beads in the second pane shows the complete reaction of the thiophosphorylated (1891.986) peptide, but no reaction with the parent peptide (1795.84 MH+). Analysis of the flow-through with additional HeLa lysate after covalent capture with iodoacetyl agarose beads in the third pane shows many peptides, but the CREB peptide is not seen. (B) Following hydrolysis, the only peptide present is the double phosphorylated CREB peak at 1875.986 (MH+), demonstrating the specificity of the technique. (C) After the covalent capture reaction with digested thiophosphorylated myelin basic protein, only one peptide (NIVT(phospho)PRTPPPSQGK) is observed. MS-MS analysis of the 1587.8001 (MH+), NIVT(phospho)PRTPPPSQGK (seen as the triply phosphorylated peptide at 524.6 m/z) shows good sequence coverage of this peptide with loss of phosphate from the y10 ion.
substrate proteins. A rapid way to determine whether the covalent capture-and-release reaction is working is to include a separate covalent capture reaction with the CREB peptide, followed by MALDI-TOF analysis (Fig. 5).
Troubleshooting A troubleshooting guide is presented in Table 1.
Anticipated Results
KinaseSubstrate Mapping
The application of this technique to appropriately labeled lysates should result in the identification of phosphopeptides. After subtraction of any phosphopeptides found in the control lysates, these peptides should predominantly be the substrates of the AS kinase that was used in the labeling reaction. The location of the phosphorylation mark on the substrate proteins will also allow the investigator to produce non-phosphorylatable substrates, and thus analyze the biological significance of the phosphorylation mark immediately fol-
lowing confirmation of the kinase-substrate relationship. Thiophosphorylation of CREB (KRREILSRRPS(p)YR) (1795.84 MH+) by GSK3B should yield a predominant peak at 1891.986 (MH+). After covalent capture with iodoacetyl agarose beads followed by specific hydrolysis, the only product ion present should be a doubly phosphorylated CREB peak at 1875.986 (MH+). The hyper-thiophosphorylation of myelin basic protein followed by digestion with trypsin and analysis by LC-MS/MS will lead to the detection of a thiophosphorylated peptide 1800.8939 (MH+) NIVT(Thiophospho)PRTPPPSQGKGR or 1587.7951 NIVT(Thiophospho)PRTPPPSQGK or 796.4044 (MH+) NIVT(Thiophospho)PR. After covalent capture with iodoacetyl agarose beads followed by specific hydrolysis, the only multiply charged ions present will be the same peptide a phospho group instead of a thiophospho modification, a loss of 16 Daltons.
34 Volume 2
Current Protocols in Chemical Biology
Table 1 Troubleshooting Guide for Kinase-Substrate Mapping
Problem
Cause
No thiophosphorylation of substrate seen during in-vitro ATPγS kinase reaction
Inactive kinase or kinase does Generate sufficient quantities of pure kinase. Check not utilize ATPγS activity with ATPγ35 S kinase assay. Inactive kinase
Solution
Generate a constitutively active kinase
Singly charged polymer or detergent ions seen during mass spectrometric analysis
Concentration of detergents is Typically the lysis buffer should only contain low too high in the lysis buffer amounts of a nonionic, nondenaturing detergent like 0.1% NP-40
No phospho-CREB detected after test covalent capture reaction
pH is too low or high in covalent capture buffer
Check the pH of the flow-through after the covalent capture reaction. Ensure the pH of the agarose beads is correctly adjusted by washing the agarose beads carefully with 200 mM HEPES, pH 7.0.
Insufficient reaction time for the covalent capture reaction
Incubate the thiophosphopeptides overnight (11-16 hr, especially if using a complex lysate)
Insufficient incubation with 10 mM DTT
Incubate with 10 mM DTT for at least 120 min following the 5% formic acid wash step
Nonspecific adsorption of peptides to microcentrifuge tubes
Using siliconized tubes and BSA will reduce nonspecific adsorption
Thiophospho CREB levels are Increase thiophospho CREB to see a signal, and too low then adjust other parameters to diagnose where the sample loss is occurring No thiophospho peptide is detected after digesting the labeled MBP
Phosphopeptides detected in no AS kinase control
Levels of TCEP are too low during digestion
Make a fresh solution of 1 M TCEP before the digestion and add to 10 mM in both the reduction and the digestion steps
Levels of urea are too low during digestion
Do not dilute to 1 M urea, only dilute to 2 M as the additional denaturant will assist in unfolding proteins during the digestion
Digestion not allowed to proceed long enough
Incubate digestion for a longer time (11-16 hr)
Insufficient trypsin added
Add trypsin to 1:10 w/w
Background kinases are utilizing the ATP analogs
Follow all optimization steps to determine the correct amount of ATP, GTP, and N6 -modified ATP to use during the labeling
Background kinases are utilizing the ATP analogs
Identify which background kinases are utilizing the ATP analogs and include inhibitors for these kinases during the labeling reaction
No phosphopeptides detected Kinase activity is low in the plus AS kinase reaction
Increase the amount of AS kinase in the labeling reaction
Stoichiometry of labeling is low
Increase the effective concentration of substrates by fractionating the lysate or by isolating specific subcellular compartments prior to labeling
N6 ATP analog concentration is too low
Increase the amount of ATP analog in both the negative control and the sample with AS kinase
KinaseSubstrate Mapping
35 Current Protocols in Chemical Biology
Volume 2
1784.8939 (MH+) NIVT(phospho)PRTPPPSQGKGR, 1587.8001 (MH+), NIVT(phospho)PRTPPPSQGK, or 780.403 (MH+) NIVT(phospho)PR. In addition, incubating the labeled MBP with the lysate to be used for substrate identification and then performing the covalent capture reaction should yield these same peptides, but no background should be seen, as there should be no source of thiophosphate in the lysate for background labeling to occur.
Time Considerations A significant investment of time and resources is required for the design and production of an AS kinase. Once a kinase has been engineered, optimization of labeling conditions and identification of putative substrate proteins should be rapidly achieved in 1 to 2 weeks. Production of the labeled positive controls should take 1 to 2 days, depending on how long the digestion of labeled MBP is performed. Once labeling conditions, digestion, and preparation of covalent capture reactions are optimized, an experienced investigator should be able to prepare and process several samples in 2 to 3 days.
Dephoure, N., Howson, R.W., Blethrow, J.D., Shokat, K.M., and O’Shea, E.K. 2005. Combining chemical genetics and proteomics to identify protein kinase substrates. Proc. Natl. Acad. Sci. U.S.A. 102:17940-17945. Gallagher, S.R. 2006. One-dimensional SDS gel electrophoresis of proteins. Curr. Protoc. Mol. Biol. 75:10.2A.1-10.2A.37. Gallagher, S., Winston, S.E., Fuller, S.A., and Hurrell, J.G.R. 2008. Immunoblotting and immunodetection. Curr. Protoc. Mol. Biol. 83:10.8.1-10.8.28. Gregan, J., Zhang, C., Rumpf, C., Cipak, L., Li, Z., Uluocak, P., Nasmyth, K., and Shokat, K.M. 2007. Construction of conditional analog-sensitive kinase alleles in the fission yeast Schizosaccharomyces pombe. Nat. Protoc. 2:2996-3000. Hazzalin, C.A., Cuenda, A., Cano, E., Cohen, P., and Mahadevan, L.C. 1997. Effects of the inhibition of p38/RK MAP kinase on induction of five fos and jun genes by diverse stimuli. Oncogene 15:2321-2331.
Literature Cited
Kwon, S.W., Kim, S.C., Jaunbergs, J., Falck, J.R., and Zhao, Y. 2003. Selective enrichment of thiophosphorylated polypeptides as a tool for the analysis of protein phosphorylation. Mol. Cell Proteomics 2:242-247.
Allen, J.J., Lazerwith, S.E., and Shokat, K.M. 2005. Bio-orthogonal affinity purification of direct kinase substrates. J. Am. Chem. Soc. 127:52885289.
Manning, G., Whyte, D.B., Martinez, R., Hunter, T., and Sudarsanam, S. 2002. The protein kinase complement of the human genome. Science 298:1912-1934.
Allen, J.J., Li, M., Brinkworth, C.S., Paulson, J.L., Wang, D., Hubner, A., Chou, W.H., Davis, R.J., Burlingame, A.L., Messing, R.O., Katayama, C.D., Hedrick, S.M., and Shokat, K.M. 2007. A semisynthetic epitope for kinase substrates. Nat. Methods 4:511-516.
Olsen, J.V., Blagoev, B., Gnad, F., Macek, B., Kumar, C., Mortensen, P., and Mann, M. 2006. Global, in vivo, and site-specific phosphorylation dynamics in signaling networks. Cell 127:635-648.
Bishop, A., Buzko, O., Heyeck-Dumas, S., Jung, I., Kraybill, B., Liu, Y., Shah, K., Ulrich, S., Witucki, L., Yang, F., Zhang, C., and Shokat, K.M. 2000. Unnatural ligands for engineered proteins: New tools for chemical genetics. Annu. Rev. Biophys. Biomol. Struct. 29:577-606.
KinaseSubstrate Mapping
Cohen, P. 2001. The role of protein phosphorylation in human health and disease: The Sir Hans Krebs Medal Lecture. Eur. J. Biochem. 268:5001-5010.
Shah, K., Liu, Y., Deirmengian, C., and Shokat, K.M. 1997. Engineering unnatural nucleotide specificity for Rous sarcoma virus tyrosine kinase to uniquely label its direct substrates. Proc. Natl. Acad. Sci. U.S.A. 94:35653570.
Blethrow, J., Zhang, C., Shokat, K.M., and Weiss, E.L. 2004. Design and use of analogsensitive protein kinases. Curr. Protoc. Mol. Biol. 66:18.11.1-18.11.19.
Trinidad, J.C., Thalhammer, A., Specht, C.G., Lynn, A.J., Baker, P.R., Schoepfer, R., and Burlingame, A.L. 2008. Quantitative analysis of synaptic phosphorylation and protein expression. Mol. Cell Proteomics 7:684-696.
Blethrow, J.D., Glavy, J.S., Morgan, D.O., and Shokat, K.M. 2008. Covalent capture of kinasespecific phosphopeptides reveals Cdk1-cyclin B substrates. Proc. Natl. Acad. Sci. U.S.A. 105:1442-1447.
Ubersax, J.A., Woodbury, E.L., Quang, P.N., Paraz, M., Blethrow, J.D., Shah, K., Shokat, K.M., and Morgan, D.O. 2003. Targets of the cyclin-dependent kinase Cdk1. Nature 425:859864.
Buzko, O. and Shokat, K.M. 2002. A kinase sequence database: Sequence alignments and family assignment. Bioinformatics 18:1274-1275.
Zhang, C., Kenski, D.M., Paulson, J.L., Bonshtien, A., Sessa, G., Cross, J.V., Templeton, D.J., and Shokat, K.M. 2005. A second-site suppressor strategy for chemical genetic analysis of diverse protein kinases. Nat. Methods 2:435441.
Carr, S.A. and Annan, R.S. 1996. Overview of peptide and protein analysis by mass spectrometry. Curr. Protoc. Protein Sci. 4:16.1.1-16.1.27.
36 Volume 2
Current Protocols in Chemical Biology
Construction and Use of Glycan Microarrays Christopher T. Campbell,1 Yalong Zhang,1 and Jeffrey C. Gildersleeve1 1
National Cancer Institute, Frederick, Maryland
ABSTRACT Glycosylation is an important post-translational modification that influences many biological processes critical for development, normal physiologic function, and diseases. Unfortunately, progress toward understanding the roles of glycans in biology has been slow due to the challenges of studying glycans and the proteins that interact with them. Glycan microarrays provide a high-throughput approach for the rapid analysis of carbohydratemacromolecule interactions. Protocols detailed here are intended to help laboratories with basic familiarity of DNA or protein microarrays to begin printing and performing assays using glycan microarrays. Basic and advanced data processing are also detailed, along with strategies for improving reproducibility of data collected with glycan arrays. C 2010 by John Wiley & Sons, Inc. Curr. Protoc. Chem. Biol. 2:37-53 Keywords: glycosylation r microarray r neoglycoconjugate r carbohydrate-dependent binding r serum antibody profile
INTRODUCTION DNA microarray technology has revolutionized our ability to probe expression of thousands of genes during a single high-throughput experiment. Extension of microarray technology to proteins has similarly advanced the field of proteomics. More recently, application of microarray technology to study carbohydrate-binding molecules has impacted the field of glycobiology. The importance of glycosylation on the interactions of molecules has long been recognized. However, the tools available for studying glycobiology have lagged behind other fields. Glycan microarrays, which display many different carbohydrates or glycans on a solid support in a spatially defined arrangement, provide a versatile tool for studying carbohydrate-mediated binding that addresses many of the technical limitations that have traditionally slowed glycobiology (Liang et al., 2008; Liu et al., 2009; Oyelaran and Gildersleeve, 2009; Song and Pohl, 2009). This chapter details protocols for printing glycan microarrays, assays for studying glycan-binding properties of various samples, and key aspects of data processing related to these high-throughput experiments. The protocols focus on the construction of neoglycoprotein/glycoprotein arrays. Many of the considerations and technical challenges, however, are relevant to other glycan array formats and protein arrays.
PRODUCTION OF GLYCAN MICROARRAYS Production of high-quality arrays is critical for a successful glycan array experiment. This printing protocol describes attachment of glycoproteins and neoglycoproteins to an epoxide-coated slide. These protocols will allow a laboratory equipped for microarray processing to adapt existing protocols for DNA or protein microarrays to glycan microarrays. The protocols require a robotic array printer and fluorescence scanner, which are widely used for printing and processing DNA and/or protein microarrays. Additionally, the following protocol describes quality-control checks for evaluating array slides.
BASIC PROTOCOL 1
Glycan Microarrays Current Protocols in Chemical Biology 2: 37-53, March 2010 Published online March 2010 in Wiley Interscience (www.interscience.wiley.com). DOI: 10.1002/9780470559277.ch090228 C 2010 John Wiley & Sons, Inc. Copyright
37 Volume 2
Neoglycoproteins Acquisition of a diverse set of glycans and carbohydrates is a key step for construction of a glycan array. The focus of this protocol is on the production of neoglycoprotein arrays. Neoglycoproteins are non-naturally-occurring conjugates formed by covalently linking a glycan to a carrier protein, such as albumin. A variety of BSA and HSA conjugates are commercially available (some sources include Dextra Labs, IsoSep AB, Glycorex, and Glycotech), and a number of other neoglycoproteins can be obtained in a single step by reductive amination of BSA with oligosaccharides containing a free reducing end (Roy et al., 1984; Gildersleeve et al., 2008). Using commercial sources and simple conjugations, one can obtain approximately 50 to 100 neoglycoproteins at present.
Materials Neoglycoproteins and/or glycoproteins (see protocol introduction above regarding sources) Printing buffer (see recipe) Controls: unmodified BSA and HSA (negative controls) and BSA conjugated with Cy3 or Alexa Fluor 647 384-well V-bottom sample plates with lids (X6004, Genetix) Aluminum plate seals (07-200-683, Costar) Centrifuge with microtiter plate carrier Robotic Microarray Printer (MicroGrid II, Genomic Solutions; http://bioinformatics.genomicsolutions.com/) Microscope (dissecting or basic optical transmission microscope) Pins (Stealth Microspotting SMP3 Pins, Arrayit; http://www.arrayit.com/) Epoxide Coated Slides (SuperEpoxy 2 Premium Microarray Substrates, Arrayit; http://www.arrayit.com/) Hygrometer Prepare sample plates containing glycoproteins and neoglycoproteins 1. Dilute glycoproteins and neoglycoproteins in printing buffer to a concentration of 125 μg/ml. The concentration of albumin conjugates can be determined by measuring UV absorption (280 nm) and comparing the results to a titration curve generated from pure albumin for conjugates with functional groups that do not absorb in the same region. Alternatively, concentration can be determined using the Bradford assay. A concentration of 125 μg/ml was selected to saturate the printed surface with glycoprotein or neoglycoproteins. Concentration of protein in the printing buffers should be adjusted to achieve saturation if either the composition of the printing buffer or the array surface changes.
2. Determine the layout of the sample plates, which must be coordinated with the number and configuration of pins that will be used for printing. See Figure 1 for details. Include appropriate controls such as unmodified BSA and HSA (negative controls) and BSA conjugated with Cy3 or Alexa Fluor 647. These controls will provide standards for scanning and aid in aligning images during data processing.
3. Place 15 μl of solution into the appropriate well(s) of a 384-well V-bottom sample plate with lid. Depending on the number of samples and setup of the sample plate, multiple sample plates may be needed. Glycan Microarrays
The maximum number of spots that can be printed within a well depends on a number of parameters, including the spot size, the pitch, the location and layout of pins, and the type
38 Volume 2
Current Protocols in Chemical Biology
A
B
2 ⫻ 2 pin configuration
Figure 1 Coordination of slide layout, configuration of sample plate, and pin positioning. (A) An example is shown with four pins loaded in the pin tool. Pins are lowered into the wells of a source plate (e.g., 384-well plate), allowing the pins to fill with neoglycoprotein solution. The robot then moves the pin head to the slides to print arrays of spots. 16 complete arrays are printed on each slide. The spacing of the pins and the array layout should match the spacing of the 16-well slide module. (B) Slides are fitted with a 16-well module to form 16 separated wells. The wells have the same spacing as is normally found on 96-well plates, and a multichannel pipettor can be used to add solutions to the wells. The slide shown in the picture has a mask that subdivides the slide into 16 areas; however, this is shown for illustrative purposes, and slides without masks can also be used.
of arrayer and software. Using the listed parameters, arrayer, and a 16 well/grid format, 440 spots can be printed in each well.
4. Seal plate with an aluminum plate seal to prevent evaporation and cross-over contamination. 5. Store sealed plates at −20◦ C in a non-defrosting freezer. Minimize freeze-thawing of samples. Sample plates may be reused for different print batches provided that they are sealed with aluminum foil and frozen at −20◦ C between uses. Before reusing the sample plate, check for evaporation from the printing buffer that may occur during printing. Samples may lose 2 to 5 μl of water during printing. Add Milli-Q water to bring the sample volume up to 15 μl, and mix sample by pipetting up and down. Sample plates handled in this way have been used for up to 6 months.
6. Immediately before printing and prior to removing the aluminum seal, thaw samples, and spin plates down for 5 min at 200 × g, 20◦ C, to collect the solutions at the bottoms of the wells.
Program robotic printer The programming methods and parameters are specific to the printer and software. Some guidelines are described below based on the use of a MicroGrid II array printer and BioRobotics TAS Application Suite software. 7. Determine the configuration of multiple array copies to be printed onto the slide. Position multiple copies of the array so that the arrays are aligned with the wells of slide modules positioned on top of the slide. For instance, printing 16 arrays in an 8 × 2 configuration conveniently allows for an 8-well multichannel pipet to be used with commercially available slide modules (Fig. 1).
8. Program the arrayer to print two to four adjacent spots for each component.
Glycan Microarrays
39 Current Protocols in Chemical Biology
Volume 2
9. Program the robotic printer to print spots at an appropriate separation. The distance between the centers of adjacent spots (referred to as the pitch) is typically twice the desired diameter of microarray spots.
10. Optimize printer settings (dwell time, velocity, and number of pre-spots) to achieve uniform spots of the desired size. The amount of material transferred during printing depends on the pin’s weight, composition, and geometry, as well as the slide’s surface properties. Thus, for each combination of pins and slides, printer settings should be optimized to obtain uniform spots of the desired size. As a starting point, the authors’ preferred settings for printing are described in Table 1. The Troubleshooting section discusses strategies for refining printing parameters.
Set up and initiate robotic printer 11. Check pins under microscope for debris or damage. Follow manufacturer’s guidelines for cleaning pins by sonication and/or blowing with particle-free compressed gas (see Fig. 2, and also see Video 1 at http://www.currentprotocols.com/protocol/ch090228). 12. Load pins into tool. Number of pins and configuration of pins depends on the desired distance between arrays and the number of arrays to be printed on each slide. See Figure 1 for further explanation and example configuration.
13. Humidify printing chamber to a relative humidity of 50% to 60%. Maintain humidity within this range throughout printing. 14. Thaw for 5 min at room temperature, and centrifuge the sample plate 5 min at 200 × g, 20◦ C, to collect samples at the bottoms of the wells. Remove aluminum plate seal and discard. 15. Load pre-spotting and array slides such that the slide’s printed surface faces upward (see Video 2 at http://www.currentprotocols.com/protocol/ch090228). 16. Load the first sample plate into the arrayer. In order to minimize evaporation, program arrayer to replace the plate lid between printing samples (see Video 3 at http://www.currentprotocols.com/protocol/ch090228). 17. Start the print run. 18. During printing, periodically check humidity and ascertain that pins are moving freely within the tool. Maintain humidity with the microarrayer at 50% to 60% by adjusting the microarrayer’s humidifier. Pin sticking is an occasional problem caused by moist pins or debris on the tool. If a pin sticks persistently, use forceps to move the pin until it slides freely.
19. After the first sample plate is complete, load the second sample plate. Even if the arrayer can load several plates at once, plates should be loaded individually so that only the sample plate currently being printed is in the arrayer. Minimize the time that sample plates are in the arrayer in order to reduce evaporation. Promptly reseal plates with a new aluminum plate seal, and return them to the freezer. Alternatively, plates may be refrigerated at 4◦ C for up to 3 days if multiple slide batches are to be printed in succession.
20. After the print run completes, inspect slides using a microscope for smearing, merged spots, missing spots, and other defects (see Figure 3). Keep a record of any defects, and save an electronic image of the arrays, if possible. Glycan Microarrays
40 Volume 2
Current Protocols in Chemical Biology
Table 1 Recommended Printing Parameters
Recommended value
Step
Parameter
Description
Sample loading
Pin depth
Distance pin is submerged into source solution
Speed
Speed pin is lowered into the source solution
Dwell time
Time pin held in source solution before wiggles or dips
Number of wiggles
Times pin moved laterally while submerged
0
Number of dips
Times pin moved vertically while submerged
2
Dip height
Distance pin is further submerged
2 mm
Inter-dip delay
Time pin kept submerged between dips
1 sec
Lid handling
Specifies when lid is replaced
Number of pre-spots
Times a freshly loaded pin is spotted onto a slide before it is used to print slides that will be used for experiments
18
Delay before pre-spotting
Time pin is momentarily suspended above slide before descending to contact the slide
0.5 sec
Target height
Distance tool is lowered to bring pin into contact with slide
1.5 mm
Speed
Speed that tool is lowered to bring pin into contact with slide
Surface height
Thickness of slide
Dwell time
Time tool is held in place at its target height
Multiple strikes
Number of times the pin is tapped onto the slide to form each spot
Pitch
Distance pin moves laterally or vertically between pre-spots
Number of print slides
High-print-quality slides that will be used for experiments
Delay before spotting
Time pin is momentarily suspended above slide before contact
0 sec
Target height
Distance tool is lowered as pin contacts slide
1 mm
Speed
Speed that tool is lowered as pin contacts slide
Dwell time
Time tool is held in place at its target height
Multiple strikes
Number of times the pin is tapped onto the slide to form each spot
Number of spots/sample
Number of spots per samples
2-4 spots
Pitch
Distance pin moves laterally or vertically between spots
0.2 mm
Dip height
Distance pins submerged into recirculating baths
5 mm
Bath dwell time
Time pins are held in bath
3 sec
MWS washes
Times wash in Main Wash Station (MWS) is repeated
Fill time
Time that water is pumped into MWS
3 sec
Drain time
Time that vacuum is turned on to drain MWS
8 sec
Pre-spotting
Printing
Washes
5 mm 4 mm/sec 1 sec
Immediately
4 mm/sec 1 mm 0.250 sec 1 0.45 mm 12
4 mm/sec 0 sec 1
2
Number of wash cycles Number of times the entire wash cycle is repeated
2 Glycan Microarrays
41 Current Protocols in Chemical Biology
Volume 2
clean pin
debris in channel
debris on pin surface
Figure 2 Inspection of pins prior to printing. Magnified views of the pins are shown. The full length of the pin should be inspected for any debris that may clog the channel. A clean pin is free of debris throughout its channel and near the pin tip.
Glycan Microarrays
A
B
C
D
Figure 3 Microscope images of printed arrays. The figure shows magnified portions of arrays after printing but prior to an assay. (A) High-quality printing produces uniform spots evenly spaced on the glass surface, which is free of debris. (B) Small spots may be due to a partially clogged pin or low volume of sample in the pin due to poor pin loading or excessive pre-spotting. (C) Debris on the surface may have no or high signal, depending on its fluorescence. (D) Missing spots are typically due to pin sticking.
42 Volume 2
Current Protocols in Chemical Biology
21. Store slides at −20◦ C in a non-defrosting freezer until use. Sealing slides in a bag containing Drierite helps prevent condensation of water vapor onto the slides. Slides stored in this manner can be used for up to 6 months. Longer time frames have not been tested.
GLYCAN ARRAY PROFILING OF CARBOHYDRATE-BINDING PROPERTIES
BASIC PROTOCOL 2
Many different glycan-binding proteins (GBPs) can be studied using glycan microarrays. Some of the most commonly studied samples include lectins, monoclonal antibodies, and serum antibodies. The general steps of the array assay are similar for each of these, but specific conditions and reagents should be tailored to the particular sample. A key consideration is matching the sample with an appropriate detection method. A variety of methods have been used to detect sample binding, but the most common approach relies on either direct or indirect tagging to produce a fluorescent signal. Lectins can be labeled with a fluorophore for direct detection or can be labeled with a tag, such as biotin, followed by detection with an appropriate fluorophore-labeled secondary reagent, such as Cy3-labeled streptavidin. Monoclonal and serum antibodies are typically detected with a fluorophore-labeled secondary antibody specific for the primary antibody’s species and isotype of interest. Selection of an appropriate fluorophore is a key consideration. In addition to matching the appropriate excitation and emission properties for the scanner, degradation and photobleaching should be considered. In the authors’ experience, Cy3, Alexa555, and DyeLight fluorophores work well, but Cy5 is not recommended due to high levels of photobleaching and degradation. Technical parameters of the assay, such as blocking buffers, incubation times, number of washes, and choice of secondary antibodies should be optimized for new samples, similar to optimization of ELISA and immunoblotting.
Materials Glycan arrays (Basic Protocol 1) Blocking buffer: 3% (w/v) bovine serum albumin (BSA, Ig-free; Sigma, cat. no. A-3059) in 1× phosphate-buffered saline (PBS; see recipe for 10×) PBST array wash buffer (see recipe) Appropriate incubation buffer for sample, consisting of BSA and PBST (Table 2) Sample (see Table 2 for appropriate concentrations) Reference sample, e.g., pooled serum or biotinylated lectins known to bind glycans on the array Secondary antibody and/or Cy3-conjugated streptavidin (Jackson ImmunoResearch) and appropriate buffer (Table 3) Microscope (dissecting or basic optical transmission microscope) Slide module (ProPlate, Invitrogen) Adhesive seals for slide module (ProPlate, Invitrogen) Orbital shaker 50-ml conical centrifuge tubes Centrifuge NOTE: Prepare all solutions fresh on the day of the experiment. NOTE: All BSA and HSA must be globulin-free to avoid high levels of background binding. Albumin that has not been sufficiently purified contains globulins, which will interfere in the assay. Other sources of albumin purified by gel electrophoresis may be
Glycan Microarrays
43 Current Protocols in Chemical Biology
Volume 2
substituted after confirming that they do not contain contaminating immunoglobulins that bind to the array.
Assemble slide module and microarray 1. Remove glycan arrays from storage at −20◦ C. 2. Inspect glycan arrays for smudging and other defects under microscope. For highest-quality results, use only arrays that are free of defects. Arrays with defects can be used in certain circumstances, depending on the location and extent of the defects. For example, they are often used for optimizing assay parameters and confirming results.
3. Place the slide atop the slide module such that the printed surface faces the slide module and the wells align with the printed arrays on the slide. 4. Secure the slide module in place according to the manufacturer’s instructions. 5. Inspect the microarray under a microscope to verify that none of the array components have been smeared.
Block glycan microarray 6. Prepare a fresh solution of blocking buffer. 7. Gently pipet blocking buffer into each well of the slide module (for a 16-well slide module, add 200 μl of blocking buffer). Use care to pipet solution against the side wall of slide module, rather than dropping the liquid directly on the array spots, to prevent smearing of the printed array surface. Rapid or forceful pipetting of blocking buffer may smear the glycan microarray or cause spreading of the array spots. These can appear as “comets” or spots with long tails. Once the unprinted array surface has been blocked, the printed array features are protected from smearing.
8. Seal the slide module with an adhesive strip to prevent evaporation. 9. Incubate for 2 hr at room temperature for lectins and monoclonal antibodies, or overnight at 4◦ C for assays involving serum. 10. Remove blocking solution and wash array immediately prior to addition of samples (see step 13) so as not to dry out the arrays while preparing samples. Wash the arrays six times, each time with a sufficient volume of PBST array wash buffer to cover the bottom of the well (e.g., 200 μl/well for a 16-well module). Discard wash solution by inverting and vigorously tapping the slide module over absorbent paper towels to remove as much wash buffer as possible.
Incubate sample with glycan microarray 11. Prepare samples in the incubation buffer appropriate for the type of sample being analyzed, at the recommended sample concentration, according to Table 2. The optimal concentration depends on the particular sample. Most plant lectins and many monoclonal antibodies give strong signals in the 1 to 10 μg/ml range. Many lectins require metals, such as calcium or manganese, for full activity. Serum antibodies from certain individuals and animals can react with albumin BSA (Igfree; Sigma, cat. no. A-3059) should be included in the buffer to avoid binding of these antibody populations with albumin conjugates on the array surface.
12. Prepare a reference sample that serves as a positive control and will be used later in data analysis. Glycan Microarrays
a. Pooled serum from multiple donors can provide a reference when analyzing serum samples. To minimize freeze-thaw cycles for the reference sample, divide a single
44 Volume 2
Current Protocols in Chemical Biology
Table 2 Recommended Buffers and Starting Dilutions of Samples for Incubation on Glycan Microarray
Sample type
Incubation buffera
Concentration
Lectin
1% BSA in PBST
1-50 μg/ml
Room temperature for 2 hr
Monoclonal antibody
3% BSA in PBST
1-50 μg/ml
37◦ C while gently shaken for 2-4 hr
Serum antibodies
3% BSA in PBST
1:50 to 1:200
Conditions
37◦ C while gently shaken for 4 hr
a PBST: PBST: see recipe for PBST array wash buffer in Reagents and Solutions.
pooled sample into 100 to 200 aliquots. Use a freshly thawed sample for each experiment. b. Lectins, either individually or pooled, can provide a similar reference when analyzing lectins. A combination of the following four biotinylated lectins will provide signals for a broad range of glycans on arrays: wheat germ agglutinin (WGA, 10 μg/ml), concanavalin A (ConA, 10 μg/ml), Bauhinia purpurea lectin (BPL, 10 μg/ml), and Ricinus communis agglutinin (RCA120, 10 μg/ml). 13. Pipet enough sample to cover bottom of well. For a 16-well module, 100 μl/well of sample solution is recommended for best reproducibility. However, the volume can be reduced to 50 μl/well if required to conserve sample. For best results, analyze each sample in duplicate, utilizing wells on two different slides printed with different pins.
14. Seal wells with adhesive seals to prevent evaporation. 15. Place arrays in orbital shaker at 100 rpm during incubation. See Table 2 for recommended incubation conditions.
Incubate with fluorophore-labeled secondary reagent (if necessary) 16. In cases where samples have been fluorophore-labeled prior to incubation, skip to step 20. Otherwise, thoroughly remove unbound samples by washing with PBST array wash buffer (200 μl/well for a 16-well module), letting the solution stand for 2 min before discarding wash. Repeat for a total of three washes. 17. Dilute the secondary reagent in the appropriate buffer according to Table 3. 18. Add secondary reagent solution to the well (100 μl/well for a 16-well module). 19. Seal wells and incubate according to Table 3. Prevent photobleaching of fluorescently labeled samples by covering arrays with aluminum foil to block light during incubation.
Perform final wash prior to scanning 20. Wash wells with PBST array wash buffer (200 μl/well for a 16-well module; immediately remove and discard buffer after each wash). Repeat for a total of seven washes. 21. Gently remove the slide module from the glass slide according to the manufacturer’s instructions. Handle the array by its edges, and use caution not to smear the printed surface. 22. Submerge the glass slides in a bath of PBST array wash buffer for 5 min. Position the glass slides so that the printed surface faces up. Glycan Microarrays
45 Current Protocols in Chemical Biology
Volume 2
Table 3 Recommended Secondary Reagents and Conditions
Sample type
Secondary label
Buffer
Concentration
Conditions
Biotinylated lectin
Cy3-Streptavidin
1% BSA in PBS
2 μg/ml
Room temperature for 2 hr
Monoclonal antibody
Monoclonal antibody specific for species and isotype of primary Ab
3% BSA in PBS
2 μg/ml
37◦ C while gently shaken for 2 hr
Human serum antibodies
Monoclonal anti-human IgG, IgM, and/or IgA
3% HSA + 1% BSA in PBS
2 μg/ml
37 ◦ C while gently shaken for 2 hr
23. Dry the glass slides using centrifugation as follows: a. Place the glass slides into 50-ml conical tubes and orient the slides so that they are perpendicular to the arm of the centrifuge. b. Centrifuge 5 min at 200 × g, 20◦ C. c. Carefully decant any liquid from the centrifuge tube so as not to re-wet the array’s printed surface. d. Remove the slides from centrifuge tubes. e. Clean the non-printed back sides of the glass slides using Kimwipes. 24. Protect the slide from light by using a covered slide holder until ready for scanning. Scan promptly (Basic Protocol 3). Scanning immediately following the final wash is recommended. However, little signal loss has been reported over hours to days when the slide is stored properly in the dark at −20◦ C (Gu et al., 2006). BASIC PROTOCOL 3
SCANNING AND DATA ANALYSIS OF GLYCAN MICROARRAYS Processed glycan microarrays are imaged using a fluorescence scanner with a resolution of 10 μm or finer. Quality images show small, uniform spots of varying intensity depending on the amount of sample bound to each neoglycoconjugate or glycoprotein on the array (see Figure 4). Specialized software, such as GenePix Pro 6.0 (Molecular Devices), measures the average and/or median pixel intensity for each spot and can be used to subtract background pixel intensity. Background-corrected pixel intensities can be used to measure sample binding in most cases. Further processing of pixel intensities is needed in certain cases. The optional Alternate Protocols 1 and 2, below, include increasing the dynamic range of the array measurements and normalizing array data using a reference sample. These advanced techniques may be useful when comparing samples with largely different affinities for an array component or when analyzing samples on different arrays.
Materials Arrays (Basic Protocol 2) Fluorescence scanner (GenePix 4000A, Molecular Devices) Image processing software (GenePix Pro 6.0, Molecular Devices) Microsoft Excel 1. Set fluorescence scanner to a resolution of 10 μm or finer. Glycan Microarrays
2. Scan using a laser power and photomultiplier tube (PMT) voltage sufficiently high to maximize image quality of array features with low amounts of bound sample, without saturating high signals.
46 Volume 2
Current Protocols in Chemical Biology
A
B
excellent
small
irregular
intense periphery
100 m
Figure 4 Scans of processed arrays. Examples of images scanned using a fluorescence scanner. (A) High-quality results show circular spots of varying intensity but uniform size. (B) Under higher magnification, high-quality spots have homogeneous intensity throughout the spot, and duplicate spots have nearly identical intensity. Printing and processing problems can result in variations in intensity across individual spots, irregular spot morphology, or missing spots.
3. Use GenePix to generate a GAL file, which contains information regarding where the array features (spots) are supposed to be, based on the printing parameters. This will produce a grid of spot boundaries for each array on the slide. Using the software, adjust the positions of the grids to align with the actual spots in the image. Positive controls, such as Cy3-BSA, help confirm alignment. Additional refinement of the spot positions and resizing of the spot boundaries can be done manually for each spot on the array, or can be carried out using object-recognition algorithms in the imageprocessing software. Any bad spots (e.g., spots obscured by lint or fluorescent streaks) should be flagged using the software. Optimal intensity measurement is highly dependent on accurate identification of spots. The positioning and size of each spot boundary should be checked carefully prior to processing.
4. Use image-processing software to measure median pixel intensity (MPI) of individual array features, neighboring background pixel intensity (BPI), and MPI-BPI for each array feature. 5. Transfer or export the data to Excel. 6. Determine an appropriate minimum value of pixel intensity (floor value), and set all values that are below the floor value so that they equal the floor value. Use of a floor value is not essential, but is very helpful. In certain cases, the background can be equal to or somewhat higher than the intensity within a spot. After background correction, the median values for these spots will be negative or zero, which can disrupt many processing steps during subsequent analyses of the data. Therefore, we always set values below 1 equal to 1. Use of higher floor values can also be beneficial. Spot intensities that are only slightly above the background intensity level are highly variable. By using a higher floor value, differences resulting from experimental noise can be de-emphasized. A floor value of 0.5 to 1.0 times the background intensity is recommended.
Glycan Microarrays
47 Current Protocols in Chemical Biology
Volume 2
7. Calculate the final signal value for each array component by averaging the signal from each of the corresponding spots. 8. Compute the coefficient of variance for the spots corresponding to each array component. For components with a coefficient of variance greater than 30%, re-inspect the scanned images for smearing, lint, and other abnormalities. ALTERNATE PROTOCOL 1
ADVANCED PROCESSING AND METHODS: EXTENDING THE DYNAMIC RANGE OF PIXEL INTENSITY MEASUREMENTS If a single scan cannot be obtained that provides good signal from low intensity spots without saturating high intensity spots, two scans can be combined to increase dynamic range according to a method described by Lyng et al. (2004). This optional method is most effective when there are a number of signals on the array, such as with serum antibodies. 1. First, obtain the primary scan at a higher PMT voltage in order to obtain good signal from low-intensity spots. 2. Perform a secondary scan with a lower PMT voltage such that no pixels or spots are saturated. 3. Obtain the raw pixel intensities for these scans as described for the basic scanning protocol (steps 3 to 6 of Basic Protocol 3). 4. For the entire slide, identify all array features with intensities that are ∼30% to 50% of saturation. For each of these middle-intensity features, calculate the ratio of pixel intensity measured at the higher PMT voltage to the pixel intensity measured at the lower PMT voltage to give a ratio for each of the middle-intensity features. Next, find the median value of these ratios (this is the correction factor, CF). 5. For only the features that are saturated or nearly saturated in the primary scan, use the correction factor to calculate the adjusted intensity based on the feature’s intensity in the secondary scan (the adjusted intensity is the intensity in the secondary scan multiplied by the correction factor). For all other spots, use the values from the primary, higher PMT voltage scan. 6. Compute the average signal intensity from duplicate spots on a slide. 7. Log transform (base 2) the signal intensities to help compare signals over a large range.
ALTERNATE PROTOCOL 2
NORMALIZE SLIDE USING REFERENCE SAMPLE Some variability or drift over time can occur for the scanner and secondary reagents. When comparing data across multiple slides over longer periods of time, normalization can be used to reduce variability. If a reference sample is run on each slide, this data can be used for normalization. Continual analysis of the reference sample on different slides also provides a useful quality-control check. 1. Determine the median signal intensity of the reference sample over the entire array, excluding any control samples (such as Cy3) that do not bind sample and do not vary due to minor processing differences.
Glycan Microarrays
2. Calculate the ratio of the reference sample’s median signal calculated in the preceding step to a standard value (this ratio is the normalization factor, NF; for example, we use a value of 10,000).
48 Volume 2
Current Protocols in Chemical Biology
3. Multiply all signal intensities on the slide by the normalization factor. However, if the signals have been log transformed, normalize the data by subtracting the log transform (base 2) of the normalization factor. 4. Reset values that were previously set to the floor to an equivalent floor on the log base 2 scale. Both slides should have the same floor value after normalization.
5. Calculate the final data value as the average of data from the two slides.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps.
PBST array wash buffer 1× phosphate-buffered saline (PBS), pH 7.4 (see recipe for 10×) 0.05% (v/v) Tween 20 (Sigma-Aldrich) Store at room temperature Phosphate-buffered saline (PBS), 10× 14.4 g Na2 HPO4 2.0 g KCl 2.4 g K2 HPO4 Milli-Q water to 1 liter Store at room temperature Printing buffer 1× phosphate-buffered saline (PBS), pH 7.4 (see recipe for 10×) 2.5% (v/v) glycerol 0.006% (v/v) Triton X-100 (Sigma-Aldrich) Store at room temperature COMMENTARY Background Information Glycan microarrays are emerging as a powerful tool for studying the interactions between a broad range of glycans and carbohydratebinding molecules (Liang et al., 2008; Liu et al., 2009; Oyelaran and Gildersleeve, 2009; Song and Pohl, 2009). These arrays are beginning to provide new insights into the many biological processes influenced by glycosylation. So far, the applications of glycan microarrays include characterizing binding properties of monoclonal antibodies and lectins [for some recent examples, see Manimala et al. (2006, 2007); Blixt et al. (2008a); Schallus et al. (2008); Song et al. (2009)], defining substrate specificities of glycosyltransferases (Park and Shin, 2007; Ban and Mrksich, 2008), and measuring antibody titers against tumors (Wang et al., 2008), pathogens (Blixt et al., 2008b; Kamena et al., 2008; Parthasarathy et al., 2008), and autoantigens (Freedman et al., 2009; Seow et al., 2009). As glycan
microarrays become more widely accessible, their application will likely be extended to new problems. A key issue for the construction of a glycan microarray is obtaining a diverse collection of carbohydrates to populate the array. It has been estimated that there are about 7,000 glycan determinants in the human glycome (Cummings, 2009), but only a tiny fraction of these are readily accessible in homogenous form. At present, even the largest glycan arrays contain only several hundred glycans. The focus of this protocol is on the construction of neoglycoprotein microarrays. The neoglycoprotein approach was chosen for several reasons. First, neoglycoproteins have been used for many years to study carbohydrate-protein interaction (Wong, 1995). Second, a variety of carbohydrate-BSA and carbohydrate-HSA conjugates are commercially available. Many others can be prepared in a single step from readily accessible oligosaccharides (Roy et al.,
Glycan Microarrays
49 Current Protocols in Chemical Biology
Volume 2
1984; Gildersleeve et al., 2008). Therefore, moderately sized neoglycoprotein arrays can be produced by many laboratories. Third, neoglycoproteins are multivalent probes that can be used in a variety of other experiments such as ELISAs and western blots, and as multivalent inhibitors. This is very useful for confirming results observed on the array and for carrying out other follow-up experiments. It should be noted that a number of other glycan array formats have been developed (Culf et al., 2006). Examples of other strategies include immobilization of neoglycolipids (Liu et al., 2007), fluorous-tagged oligosaccharides (Pohl, 2008), nucleophile-linked oligosaccharides (Blixt et al., 2004; Ratner et al., 2004), and noncovalent adsorption of large polysaccharides (Wang et al., 2002). Methods for accessing larger glycan libraries and making protein conjugates and neoglycoconjugates have been reviewed (Seeberger, 2008). In addition to variations in glycan structure, the array format is well suited for varying the mode of glycan presentation. Many carbohydrate-binding proteins achieve highavidity interactions via formation of multivalent complexes. Features of presentation such as spacing and orientation of glycans affect the ability to form a multivalent interaction. Microarrays containing neoglycoconjugates can easily vary glycan density on the array. For example, glycan density can be varied by modulating the average number of glycans attached to each molecule of carrier protein (Oyelaran et al., 2009).
Critical Parameters
Glycan Microarrays
Methods and parameters used for imaging and processing data can significantly affect results. Small adjustments in the size and position of boundaries used to define array spots can have a substantial effect on results. Parameters for finding spots (such as limits on resizing features) can be easily overlooked details, yet can significantly impact results. Setting limits on the resizing features, such as no less than 70% and no more than 150% of the expected diameter, can improve automatic object recognition. Manual inspection of spot recognition is highly recommended. Although manual adjustment of objects is needed occasionally, it is important to do so with caution and consistency; the same standards for measuring pixel intensity and background should be applied to all components on the array.
Troubleshooting Printing problems Printing problems are the most likely cause of irregular spot morphology or high variability in the data. To help find printing problems early, slides should be viewed under a microscope after printing and prior to blocking. Table 4 summarizes causes and remedies for several common printing problems. The first step in evaluating printing problems is to determine if the problems are associated with a particular pin or sample. If the problems are restricted to a single pin, that pin should be cleaned or replaced. Clogged or damaged pins typically lead to missing spots, although irregular spot morphology can occur. Missing or irregular spots are also indications of sample-specific problems, such as insufficient loading of sample into pins. Small prespots and a decrease in spot size as the run progresses can occur when volume of sample in the sample plate is low. When problems appear more widespread, consider adjusting printing parameters to either increase or decrease the amount of sample deposited with each pin contact. Pre-spotting reduces the amount of sample deposited by the first pin strikes after loading. Increasing the number of pre-spots may eliminate some merged spots. Prolonging the contact time (i.e., increased dwell time) between pins and slide will make spots larger. Also check the concentration of neoglycoprotein and Triton X-100 in the print solution. Intense signal at the periphery of spots (known as “donuts” and shown in panel B of Figure 4) may result when Triton X-100 is not included in the print buffer (Deng et al., 2006). Deterioration or contamination of the sample plates are another cause of poor-quality slides. Many print problems can be corrected by replacing the sample plates. However, this is not always practical when samples must be conserved. Adding water to the samples to replace losses from evaporation and mixing samples by pipetting up and down can also improve print quality. High background High background can be caused by insufficient blocking, high sample concentration, or slide contamination. The blocking solution may contain contaminants, such as immunoglobulins, that bind to some array components. Alternatively, the blocking solution may not be effective at blocking.
50 Volume 2
Current Protocols in Chemical Biology
Table 4 Troubleshooting Guide for Glycan Microarray Printing and Processing
Problem
Possible cause
Solution
Missing spots
Pin sticking Clogged pin Damaged pin
Dry pins and tool thoroughly Sonicate pins Replace pin
Irregular spot morphology/merged spots
Inadequate pre-spotting
Change pre-spotting For “donuts”, be sure Triton X 100 is in the print buffer For small spots, increase neoglycoprotein concentration in source plate
High background
Globulins in BSA or HSA Inadequate washes Poor blocking Dirty slide module
Block only with globulin-free albumin Increase number of washes Increase concentration of BSA, HSA, or tween during blocking Replace slide module
Low or no signal
Inadequate incubation time Low concentration of GBP Photobleaching Protein is inactive or lacks carbohydrate binding activity Appropriate glycan ligand is not present on the array
Increase time or temperature of incubation Increase concentration of GBP Protect fluorescent dyes from light during incubation
High variability
Batch-to-batch printing variations
Analyze groups of samples on arrays from the same printing batch
Problems with the blocking solution can be found by analyzing binding of a secondary antibody to a blocked slide that has not been incubated with any sample. If the secondary antibody alone binds to a blocked slide, consider increasing the concentration of blocking solution, changing the solution, or adding another blocking agent. It is important to note that many common blocking agents for biological experiments, such as gelatin, casein, milk, and animal serum, contain glycoproteins and/or carbohydrates which may inhibit binding to glycans on the array. High background can also be reduced by lowering the concentration of sample, increasing washes, or using a fresh slide module. Contamination of the sample plate (in particular carry-over from insufficiently washed pins) may increase background of samples printed after high-signal samples. Low or no signal An absence of signal can be due to a variety of causes. First, the glycan array may not possess the correct carbohydrate ligand for recognition, or it may not present that ligand in an appropriate context for binding. Second, the carbohydrate binding entity being tested
may denature or lose activity during purification, during storage, or as a result of freezing and thawing prior to the assay. Third, the secondary reagent used for detection may be inactive or unable to recognize the sample of interest. Finally, one should keep in mind that not every protein evaluated on a glycan array is capable of binding to carbohydrates. When a sample shows low or no signal, troubleshooting begins by analyzing a positive control (such as monoclonal antibodies, sera, or lectins known to bind glycans on the array) to rule out technical problems with the array. Problems with fluorophore tags can produce low signal for positive controls. Since the signal from fluorophore-labeled reagents fade over time, they require periodic replacement. Adjust concentration of the secondary reagent or switch to a more stable dye such as Cy3 or an Alexa dye if the fluorophore requires frequent replacement. If positive controls produce high signal, low concentration of sample or suboptimal incubation conditions are likely causes of the sample’s low signal. Check that the sample has been properly mixed, and increase the concentration of sample. Also, check that the incubation conditions (especially temperature,
Glycan Microarrays
51 Current Protocols in Chemical Biology
Volume 2
length, and metal and salt concentrations) are appropriate. Some lectins require metals for binding.
Anticipated Results Characteristics of high-quality images Processing and scanning of array samples produces images of multiple, well defined spots, as shown in Figure 4. Spot intensity correlates with the amount of sample present. Slides should have few, if any, saturated pixels. High-quality images have few missing spots and are free of any smears or debris. Spots should be uniformly sized. Merged, irregularly shaped, large, or small spots indicate printing problems. Although spots may differ in intensity, each individual spot should be homogeneous in intensity. Spots with brighter intensity at the periphery (referred to as “donuts”) are also indications of immobilization irregularities. Expected variation in array data Variability in glycan microarray data should be carefully monitored to ensure reliable results. The first check on variability is comparing the signal from each of the spots for each array component. The median absolute difference between components’ final binding value on the two slides should be ∼20% in absolute signal, which corresponds to ± ∼0.3 on a log base 2 scale. Another check on variability is to compare array values for the same sample (e.g., a reference sample) analyzed at several different times. It is highly recommended to include a reference sample on each slide and to continually compare newly acquired data for the reference sample to previously obtained data for that sample. The median standard deviation of each array component should be less than or equal to 35%, even across different arrays and different batches of slides. Array components whose signals vary substantially more or drift over time could indicate printing or storage problems.
Time Considerations
Glycan Microarrays
Time considerations for printing Depending on the number of components on the array, printing can be completed in a few hours to 1.5 days. If printing slides will take more than a day, the arrayer should be kept on overnight with the array slides secured in place at room temperature. The vacuum must stay on to keep the slides aligned in position. The sample plates, however, should be promptly
returned to the refrigerator or freezer to minimize evaporation. When multiple batches are to be printed one after the other (for example, three batches of 12 slides printed over 5 days), the sample plates should be resealed with aluminum seals and stored at 4◦ C between printing runs. This is done to minimize the number of freeze-thaw cycles. Running more than three print cycles without replacing water is not recommended, due to evaporation loss. Time considerations for assays There are no natural pause points when performing binding assays with the arrays. Once the slides have been blocked, it is recommended that samples be processed without interruption. The assays can be performed in a single day. Depending on the number of washes and lengths of incubation periods, the assay generally takes 5 to 8 hr. If needed, the processed slides can be stored for several hours to days at −20◦ C in the dark prior to scanning (Gu et al., 2006).
Literature Cited Ban, L. and Mrksich, M. 2008. On-chip synthesis and label-free assays of oligosaccharide arrays. Angew. Chem. Int. Ed. Engl. 47:3396-3399. Blixt, O., Head, S., Mondala, T., Scanlan, C., Huflejt, M.E., Alvarez, R., Bryan, M.C., Fazio, F., Calarese, D., Stevens, J., Razi, N., Stevens, D.J., Skehel, J.J., van Die, I., Burton, D.R., Wilson, I.A., Cummings, R., Bovin, N., Wong, C.H., and Paulson, J.C. 2004. Printed covalent glycan array for ligand profiling of diverse glycan binding proteins. Proc. Natl. Acad. Sci. U.S.A. 101:17033-17038. Blixt, O., Han, S., Liao, L., Zeng, Y., Hoffmann, J., Futakawa, S., and Paulson, J.C. 2008a. Sialoside analogue arrays for rapid identification of high affinity siglec ligands. J. Am. Chem. Soc. 130:6680-6681. Blixt, O., Hoffmann, J., Svenson, S., and Norberg, T. 2008b. Pathogen specific carbohydrate antigen microarrays: A chip for detection of Salmonella O-antigen specific antibodies. Glycoconj. J. 25:27-36. Culf, A.S., Cuperlovic-Culf, M., and Ouellette, R.J. 2006. Carbohydrate microarrays: Survey of fabrication techniques. OMICS 10:289-310. Cummings, R.D. 2009. The repertoire of glycan determinants in the human glycome. Mol. Biosyst. 5:1087-1104. Deng, Y., Zhu, X.Y., Kienlen, T., and Guo, A. 2006. Transport at the air/water interface is the reason for rings in protein microarrays. J. Am. Chem. Soc. 128:2768-69. Freedman, M.S., Laks, J., Dotan, N., Altstock, R.T., Dukler, A., and Sindic, C.J. 2009. Anti-alphaglucose-based glycan IgM antibodies predict
52 Volume 2
Current Protocols in Chemical Biology
relapse activity in multiple sclerosis after the first neurological event. Mult. Scler. 15:422430. Gildersleeve, J.C., Oyelaran, O., Simpson, J.T., and Allred, B. 2008. Improved procedure for direct coupling of carbohydrates to proteins via reductive amination. Bioconjug. Chem. 19:14851490. Gu, Q., Sivanandam, T.M., and Kim, C.A. 2006. Signal stability of Cy3 and Cy5 on antibody microarrays. Proteome Sci. 4:21. Kamena, F., Tamborrini, M., Liu, X., Kwon, Y.U., Thompson, F., Pluschke, G., and Seeberger, P.H. 2008. Synthetic GPI array to study antitoxic malaria response. Nat. Chem. Biol. 4:238-240. Liang, P.H., Wu, C.Y., Greenberg, W.A., and Wong, C.H. 2008. Glycan arrays: Biological and medical applications. Curr. Opin. Chem. Biol. 12:8692.
Adamovicz, J.J., Bavari, S., and Waag, D.M. 2008. Application of carbohydrate microarray technology for the detection of Burkholderia pseudomallei, Bacillus anthracis and Francisella tularensis antibodies. Carbohydr. Res. 2008 Jun 14. [Epub ahead of print] Pohl, N.L. 2008. Fluorous tags catching on microarrays. Angew. Chem. Int. Ed. Engl. 47:38683870. Ratner, D.M., Adams, E.W., Su, J., O’Keefe, B.R., Mrksich, M., and Seeberger, P.H. 2004. Probing protein-carbohydrate interactions with microarrays of synthetic oligosaccharides. Chembiochem 5:379-382. Roy, R., Katzenellenbogen, E., and Jennings, H.J. 1984. Improved procedures for the conjugation of oligosaccharides to protein by reductive amination. Can. J. Biochem. Cell Biol. 62:270275.
Liu, Y., Feizi, T., Campanero-Rhodes, M.A., Childs, R.A., Zhang, Y., Mulloy, B., Evans, P.G., Osborn, H.M., Otto, D., Crocker, P.R., and Chai, W. 2007. Neoglycolipid probes prepared via oxime ligation for microarray analysis of oligosaccharide-protein interactions. Chem. Biol. 14:847-859.
Schallus, T., Jaeckh, C., Feher, K., Palma, A.S., Liu, Y., Simpson, J.C., Mackeen, M., Stier, G., Gibson, T.J., Feizi, T., Pieler, T., and MuhleGoll, C. 2008. Malectin: A novel carbohydratebinding protein of the endoplasmic reticulum and a candidate player in the early steps of protein N-glycosylation. Mol. Biol. Cell 19:34043414.
Liu, Y., Palma, A.S. and Feizi, T. 2009. Carbohydrate microarrays: Key developments in glycobiology. Biol. Chem. 390:647-656.
Seeberger, P.H. 2008. Automated oligosaccharide synthesis. Chem. Soc. Rev. 37:19-28.
Lyng, H., Badiee, A., Svendsrud, D.H., Hovig, E., Myklebost, O., and Stokke, T. 2004. Profound influence of microarray scanner characteristics on gene expression ratios: Analysis and procedure for correction. BMC Genomics 5:10.
Seow, C.H., Stempak, J.M., Xu, W., Lan, H., Griffiths, A.M., Greenberg, G.R., Steinhart, A.H., Dotan, N., and Silverberg, M.S. 2009. Novel anti-glycan antibodies related to inflammatory bowel disease diagnosis and phenotype. Am. J. Gastroenterol. 104:1426-1434.
Manimala, J.C., Roach, T.A., Li, Z., and Gildersleeve, J.C. 2006. High-throughput carbohydrate microarray analysis of 24 lectins. Angew. Chem. Int. Ed. Engl. 45:3607-3610.
Song, E.H. and Pohl, N.L. 2009. Carbohydrate arrays: Recent developments in fabrication and detection methods with applications. Curr. Opin. Chem. Biol. 13:626-632.
Manimala, J.C., Roach, T.A., Li, Z., and Gildersleeve, J.C. 2007. High-throughput carbohydrate microarray profiling of 27 antibodies demonstrates widespread specificity problems. Glycobiology 17:17C-23C.
Song, X., Xia, B., Stowell, S.R., Lasanajak, Y., Smith, D.F., and Cummings, R.D. 2009. Novel fluorescent glycan microarray strategy reveals ligands for galectins. Chem. Biol. 16:3647.
Oyelaran, O. and Gildersleeve, J.C. 2009. Glycan arrays: Recent advances and future challenges. Curr. Opin. Chem. Biol. 13:406-413.
Wang, C.C., Huang, Y.L., Ren, C.T., Lin, C.W., Hung, J.T., Yu, J.C., Yu, A.L., Wu, C.Y., and Wong, C.H. 2008. Glycan microarray of Globo H and related structures for quantitative analysis of breast cancer. Proc. Natl. Acad. Sci. U.S.A. 105:11661-11666.
Oyelaran, O., Li, Q., Farnsworth, D., and Gildersleeve, J.C. 2009. Microarrays with varying carbohydrate density reveal distinct subpopulations of serum antibodies. J. Proteome Res. 8:3529-3538. Park, S. and Shin, I. 2007. Carbohydrate microarrays for assaying galactosyltransferase activity. Org. Lett. 9:1675-678.
Wang, D., Liu, S., Trummer, B.J., Deng, C., and Wang, A. 2002. Carbohydrate microarrays for the recognition of cross-reactive molecular markers of microbes and host cells. Nat. Biotechnol. 20:275-281.
Parthasarathy, N., Saksena, R., Kovac, P., Deshazer, D., Peacock, S.J., Wuthiekanun, V., Heine, H.S., Friedlander, A.M., Cote, C.K., Welkos, S.L.,
Wong, S.Y. 1995. Neoglycoconjugates and their applications in glycobiology. Curr. Opin. Struct. Biol. 5:599-604.
Glycan Microarrays
53 Current Protocols in Chemical Biology
Volume 2
Quantitative Glycomics of Cultured Cells Using Isotopic Detection of Aminosugars with Glutamine (IDAWG) Meng Fang,1 Jae-Min Lim,1 and Lance Wells1 1
Complex Carbohydrate Research Center, University of Georgia, Athens, Georgia
ABSTRACT IDAWG (Isotopic Detection of Aminosugars With Glutamine) is a newly reported, in vivo, stable isotopic labeling strategy for quantitative glycomics of cultured cells. Detailed procedures are provided for glycan analysis using IDAWG including labeling, release of both N- and O-linked glycans, permethylation, and mass spectrometry analysis. The methods for data processing and calculations are also introduced here but have not yet C 2010 by John Wiley & Sons, Inc. been automated. Curr. Protoc. Chem. Biol. 2:55-69 Keywords: IDAWG r stable isotopic labeling r quantitative glycomics r cell culture r glycan analysis
INTRODUCTION Glycomics is the comprehensive study of the entire complement of complex carbohydrates, often called glycans, that are polymers of monosaccharides often found attached to proteins and lipids (Aoki-Kinoshita, 2008). As a post-translational modification, glycosylation plays critical roles in numerous physiological processes including protein folding and cellular signaling, and altered carbohydrate expression is a common feature of many types of cancers (Dube and Bertozzi, 2005). One of the important tasks for glycomics is to develop technologies capable of comparative, relative-quantitative analysis for examining global alterations of glycans between different biological samples. Following the steps of proteomics, several non-isotope-based and stable-isotope-based labeling strategies have recently been developed for quantitative glycomics by mass spectrometry (Alvarez-Manilla et al., 2007; Aoki et al., 2007; Kang et al., 2007). This unit provides detailed protocols for Isotopic Detection of Aminosugars With Glutamine, termed IDAWG, which is the first in vivo cell culture isotope-labeling strategy for glycan analysis (Orlando et al., 2009). The methods of labeling the aminosugars (GlcNAc, GalNAc, and sialic acid) with amide-15 N-Gln in cell culture is described in Basic Protocol 1, and the detection of these aminosugars in both N- and O-linked glycan structures is described in Basic Protocol 2. The mathematical calculations used to calculate under-incorporation rate and relative ratios of light/heavy species are described in the Support Protocol.
CELL CULTURE FOR IDAWG LABELING This basic protocol is for labeling any cells in culture. Due to large differences in metabolic flux, different cells must be cultured in heavy Gln for different periods of time to achieve a high degree (>95%) of labeling. In our hands, human embryonic stem cells (hESCs) that divide quickly (doubling time of ∼30 hr) and are very metabolically active label to completion in 72 hr (changing the medium every 24 hr). However, we have also labeled differentiated 3T3-L1 adipocytes that are post-mitotic, which required 6 days of labeling to achieve >90% labeling. Thus, the optimum time of labeling must be determined for each individual cell line, but since the label is relatively inexpensive, 1 week Current Protocols in Chemical Biology 2: 55-69, April 2010 Published online April 2010 in Wiley Interscience (www.interscience.wiley.com). DOI: 10.1002/9780470559277.ch090207 C 2010 John Wiley & Sons, Inc. Copyright
BASIC PROTOCOL 1
Quantitative Glycomics Using IDAWG
55 Volume 2
of labeling is a good starting point for an unknown cell type, and under-incorporation can be calculated (see Support Protocol). This 1-week labeling is a starting point that can be adjusted based on the metabolic activity and to some extent the doubling time of the cells. In terms of scale, it is recommended to begin with a minimum of 2 × 106 cells per growth condition (scaling up for making protein powder, as described in Basic Protocol 2, is straightforward).
Materials Cells Gln-free medium appropriate for cells (Invitrogen) Amide-15 N-Gln (98%) (Cambridge Isotope Laboratories, Inc.) Normal-abundance 14 N-Gln (L-glutamine; Invitrogen, cat. no. 21051024) Phosphate-buffered saline (PBS; see recipe) Appropriate tissue culture equipment Cell scrapers 1. Grow cells as normal using Gln-free medium supplemented with either amide-15 NGln or corresponding normal-abundance 14 N-Gln at 2 mM final concentration. Cells incubated with both the labeled and normal Gln are required for preparing the protein powder in Basic Protocol 2.
2. Change the medium daily and allow the cells to grow for at least 72 hr in heavy label. 3. Do not harvest cells using trypsin, as it will cleave off many of the cell surface glycoproteins. Instead harvest adherent cells by scraping in PBS, and prepare cell pellets. BASIC PROTOCOL 2
N- AND O-LINKED GLYCAN ANALYSIS WITH IDAWG This basic protocol describes the detailed procedure for analysis via mass spectrometry of both permethylated N- and O-linked glycans released from cultured cells. The cells are first lysed, then delipidated and turned into protein powders. Both N- and O-linked glycans are released from the protein, and the permethylation of glycans is performed (see Fig. 1). Permethylated glycans are then analyzed using high-resolution mass spectrometry. To compare the quantities of each glycan structures in different samples, the protein powders of normal and 15 N-labeled samples are mixed together by weight (see Basic Protocol 2) or by cell number upon culture harvesting. To calculate the underincorporation rate for each glycan, which is crucial for the calculation of the ratio of light/heavy structures, an extra analysis of pure 15 N-labeled sample with exactly the same procedure can be performed (see Support Protocol).
Materials
Quantitative Glycomics Using IDAWG
Cell pellets from cell culture (see Basic Protocol 1) Phosphate-buffered saline (PBS; see recipe) Methanol, HPLC grade, ice cold Chloroform, HPLC grade 18.2 M (Milli-Q) water Acetone, HPLC grade, ice cold Source of dry N2 40 mM NH4 HCO3 , pH 8 2 mg/ml trypsin (Sigma-Aldrich, cat. no. T8003) in 40 mM NH4 HCO3 , pH 8 (store at −20◦ C) 2 mg/ml chymotrypsin (Sigma-Aldrich, cat. no. C4129) in 40 mM NH4 HCO3 , pH 8 (store at −20◦ C)
56 Volume 2
Current Protocols in Chemical Biology
light medium
heavy medium
Amide-14N L-glutamine “GIn-14”
Amide-15N L-glutamine “Gln-15” 1 2
mixed protein powder tryptic digestion peptide mixture C18 reverse phase PNGase F digestion
-elimination
N-linked glycan mixture
O-linked glycan mixture
C18 reverse phase permethylation
deionization permethylation
permethylated glycans
permethylated glycans
100 95 90 85 80 75 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0 1030
light Gln-14
1031
heavy Gln-15
1032
1033
1034
1035
1036
Relative abundance
Relative abundance
mass spectrometry
486.23
100 95 90 85 80 75 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0
901.93
790.88
966.47
474.23
300
400
500
908.51 600
700
800
900
1000 1100 1200
m/z
m/z
Quantification by full MS
Characterization by MS/MS
1. Combine the cells before homogenization and delipidation to get mixed protein powder. 2. Separate homogenization and delipidation before mixing the protein powder.
Figure 1 Schematic workflow of IDAWG labeling strategy for comparative glycomics. It is possible either to combine equal cell numbers before homogenization and delipidation to obtain the mixed protein powder, or to perform the homogenization and delipidation before mixing the protein powder by weight.
2 M urea in 40 mM NH4 HCO3 (prepare fresh) Acetonitrile, HPLC grade 5% and 10% (v/v) acetic acid (HPLC grade) 20% (v/v) isopropanol (HPLC grade) in 5% acetic acid 40% (v/v) isopropanol (HPLC grade) in 5% acetic acid Isopropanol, HPLC grade 100 mM sodium phosphate, pH 7.5 7.5 μg/ml Peptide-N-glycosidase F (PNGase F), store at 4◦ C (ProZyme, http://www.prozyme.com/) 1 M sodium borohydride (prepare fresh) AG 50W–X8 resin stock (see recipe) 1 M HCl
Quantitative Glycomics Using IDAWG
57 Current Protocols in Chemical Biology
Volume 2
9:1 methanol:glacial acetic acid (HPLC grade) 50% w/w sodium hydroxide solution Anhydrous methanol (99.8%, Sigma) Anhydrous dimethylsulfoxide (DMSO; 99.9%, Sigma) Iodomethane (99.5%, Aldrich) Dichloromethane, HPLC grade 1 mM NaOH in 50% methanol 10-ml Dounce homogenizer 8-ml screw-top glass tubes, precleaned with methanol End-over-end rotator VWR Clinical 50 centrifuge or equivalent Pierce Reacti-Vap Evaporating Unit and Reacti-Therm Heating/Stirring Module Heating block (e.g., Fisher Isotemp 125D) Bakerbond SPE Octadecyl (C18 ) disposable extraction columns (J.T. Baker) Speed-Vac evaporator Bath sonicator (Branson Ultrasonic Cleaner; Model 1510R-MT) 500-μl microsyringe Fused-silica emitter (360 × 75 × 30 μm, SilicaTip; New Objective, http://www.newobjective.com/) LTQ-Orbitrap XL mass spectrometer with nano ESI source (ThermoFisher) or equivalent NOTE: In this protocol, we introduce the method of mixing samples by weighing protein powder (in step 15). Alternatively, cells labeled light and heavy can be combined before step 1, based on accurate equal cell numbers.
Perform cell lysis and delipidation 1. Wash cell pellet (∼2 × 106 cells) by adding 5 to 10 ml PBS, centrifuging 5 min at 1200 × g, 4◦ C, and then removing the supernatant. To the pellet, add water to 100 μl, and transfer the suspension to a 1.5-ml microcentrifuge tube. Cells grown in media with the light and heavy label, respectively, are required.
2. Add 500 μl ice-cold methanol to the tube and move the cells to a 10-ml Dounce homogenizer. 3. Disrupt the cells well on ice by Dounce homogenization (6 to 8 strokes), then transfer the mixture to a tared 8-ml glass tube. 4. Add 3.5 ml methanol, 1.5 ml water, and 2 ml chloroform to yield a final ratio of 4:8:3 chloroform:methanol:water. 5. Incubate the mixture for 3 hr at room temperature with end-over-end rotation to extract lipids. 6. Centrifuge 30 min at 3300 × g, 4◦ C. 7. Remove the supernatant, add 4 ml methanol, 1.5 ml water, and 2 ml chloroform to the insoluble materials, and incubate for 2 more hr at room temperature to extract the lipids again. The supernatant can be discarded or stored for analysis of glycolipids.
8. Centrifuge 30 min at 3300 × g, 4◦ C, and remove the supernatant. 9. Add 1 ml water to the insoluble materials in glass tube and vortex well. Quantitative Glycomics Using IDAWG
10. Add 6 ml ice-cold acetone to the glass tube and vortex well to precipitate the proteins.
58 Volume 2
Current Protocols in Chemical Biology
11. Incubate for 10 min on ice and then centrifuge 30 min at 3300 × g, 4◦ C, and remove the supernatant. 12. Repeat steps 9 to 11 two more times. 13. Dry the insoluble protein powder under a stream of nitrogen at 40◦ C using the Pierce Reacti-Vap Evaporating Unit and Reacti-Therm Heating/Stirring Module. 14. Weigh the protein powders in the tared 8-ml glass tubes. 15. Take the same amount of protein powder (3 to 5 mg each) of normal and 15 N-labeled cell populations and mix together. Prepare one sample of mixed protein powder to prepare N-linked glycans and another sample to prepare O-linked glycans. Proceed to release N-linked glycans (step 16) and O-linked glycans.
Release N-linked glycans 16. Resuspend the mixed protein powder in an 8-ml glass tube with 200 μl of 40 mM NH4 HCO3 , seal the tube, then place in a heating block with temperature set at 100◦ C for 5 min. 17. After cooling to room temperature, centrifuge briefly to collect the solution at the bottom of the tube, then add 25 μl of 2 mg/ml trypsin solution in 40 mM NH4 HCO3 and 25 μl of 2 mg/ml chymotrypsin in 40 mM NH4 HCO3 . Finally, add 250 μl of 2 M urea (in 40 mM NH4 HCO3 to yield a final concentration of 1 M urea. Seal the tube. 18. Incubate the solution for 18 hr at 37◦ C to digest the proteins. After incubation, place the tube in the 100◦ C heating block for 5 min to deactivate the enzymes. Allow to cool to room temperature. 19. While the solution is cooling, equilibrate a Sep-Pak C18 cartridge column by washing three times with 100% acetonitrile followed by three washes with 5% acetic acid. 20. Add 500 μl of 10% acetic acid to the cooled solution, then load the solution onto the equilibrated Sep-Pak C18 cartridge column. After loading all the sample, wash the column with 1 ml of 5% acetic acid three times. Next, put a glass collection tube under the column and elute the peptides stepwise, first with 1 ml of 20% isopropanol in 5% acetic acid, then with 1 ml of 40% isopropanol in 5% acetic acid, and finally with 1 ml of 100% isopropanol. Combine the elutions and dry down in a Speed-Vac evaporator to remove solvents. It is possible either to use stepwise elutions or simply to use 100% isopropanol for all three washes.
21. Resuspend the peptides with 45 μl of 100 mM sodium phosphate (pH 7.5) and then add 5 μl PNGase F stock to release the N-linked glycans from the peptides. Incubate for 20 hr at 37◦ C. 22. After PNGase F digestion, add 450 μl of 5% acetic acid to solution, then the load solution onto an equilibrated Sep-Pac C18 column (see step 19 for equilibration). Elute the N-linked glycans with three 1-ml aliquots of 5% acetic acid and collect them in a glass tube. Dry down in Speed-Vac evaporator for permethylation (proceed to step 28). If desired, peptides can be eluted with a high organic solvent percentage in the 5% acetic acid.
Release O-linked glycans 23. In an 8-ml glass tube, add 1 ml of freshly prepared 1 M sodium borohydride to the mixed protein powder made in step 15, vortex, and sonicate the sample tube quickly
Quantitative Glycomics Using IDAWG
59 Current Protocols in Chemical Biology
Volume 2
in a Branson Ultrasonic Cleaner at room temperature using the default settings. Incubate for 18 hr at 45◦ C in a heating block. 24. After incubation, cool the sample to room temperature. Add 10% acetic acid dropwise to the tube while vortexing, until bubbling stops. 25. Pack a 1-ml bed volume of AG 50W–X8 resin stock into a Pasteur pipet to make the cation-exchange column and wash the column sequentially with 2 ml methanol, 1 ml of 1 M hydrochloric acid and 2 ml 5% acetic acid. 26. Load sample onto equilibrated column, elute the O-linked glycans with 7 ml of 5% acetic acid, and collect in an 8-ml glass tube. Dry down in a Speed-Vac evaporator. If desired, peptides can be recovered with high salt or high pH wash. To recover the peptides, wash three times using 1 M NaCl in 40 mM NH4 HCO3 , pH 8.
27. Add 1.5 ml of a 9:1 methanol:glacial acetic acid mixture and then dry the sample under a stream of dry nitrogen using the Pierce Reacti-Vap Evaporating Unit and Reacti-Therm Heating/Stirring Module. Repeat this step two more times to remove the borate. Proceed to step 28 for permethylation.
Permethylate glycans 28. Prepare the dry sodium hydroxide solution for permethylation: a. In a glass tube, combine 100 μl of 50% w/w sodium hydroxide solution and 200 μl anhydrous methanol and vortex briefly. b. Add 4 ml anhydrous dimethylsulfoxide (DMSO) and vortex. c. Centrifuge the tube quickly (30 sec) at 1068 × g, room temperature, and pipet off dry DMSO, salts, and white residue, leaving clean sodium hydroxide solution in the tube. d. Repeat steps 2 and 3 four to five more times to remove all of the white residue in the tube. e. Once the tube is clean, add 2 ml dry DMSO and pipet up and down gently. 29. Add 200 μl dry DMSO to the released N- or O-linked glycan samples (from step 22 or 27) and purge the tube with dry N2 to remove air. Next, sonicate 2 min in a Branson Ultrasonic Cleaner at room temperature using the default settings and vortex quickly to dissolve sample. 30. Add 250 μl of the prepared sodium hydroxide solution (from step 28) to sample tube, purge with dry N2 , and sonicate quickly in a Branson Ultrasonic Cleaner at room temperature using the default settings. 31. Add 100 μl iodomethane with a 500-μl microsyringe to the sample, purge with dry N2 , and vortex vigorously for 5 min. 32. Add 2 ml water and bubble off iodomethane with dry N2 gently. After the solution becomes clear, add 2 ml dichloromethane and vortex. 33. Centrifuge the tube quickly (30 sec) at 1068 × g, room temperature, and then remove the top aqueous layer. 34. Add 2 ml water, vortex, and centrifuge quickly (30 sec) at 1068 × g, room temperature, and then remove the aqueous layer.
Quantitative Glycomics Using IDAWG
35. Repeat steps 33 and 34 four more times, then dry the sample under a gentle stream of dry N2 using the Pierce Reacti-Vap Evaporating Unit and Reacti-Therm Heating/Stirring Module.
60 Volume 2
Current Protocols in Chemical Biology
Analyze glycans by mass spectrometry 36. Dissolve the permethylated N- or O-linked glycans in 30 μl of 100% methanol. Place 15 μl of the solution into a 1.5-ml microcentrifuge tube, then add 35 μl of 1 mM NaOH in 50% methanol to achieve a final volume of 50 μl. 37. Infuse the solution directly into the mass spectrometer using a nanospray ion source with a fused-silica emitter (360 × 75 × 30 μm, SilicaTip) at 2.0 kV capillary voltage, 240◦ C capillary temperature, and a syringe flow rate of 0.4 μl/min. 38. Acquire the full FTMS (Fourier Transform Mass Spectrometry) spectra at 400 to 2000 m/z in positive ion and profile mode with two microscans and 1000 maximum injection time (msec). This step must be optimized for the particular instrument and can be combined with tandem mass spectrometry analysis of the analytes.
39. Calculate the ratios of the same glycan structures in each sample (normal and 15 N-labeled) according to Support Protocol 1. N- or O-linked glycan structures can be manually interpreted using GlycoWorkbench software (Ceroni et al., 2008), for example, but it is outside the scope of this article to describe this interpretation.
CALCULATING RELATIVE RATIOS OF GLYCANS The IDAWG technology is designed to compare the quantities of each glycan structure in different samples. The glycan structures can be manually interpreted before the calculations (as mentioned in the annotation under step 39 of Basic Protocol 2). Thus, the number of nitrogens can be easily determined based on the total number of GlcNAc, GalNAc, and sialic acid in the structure. Because the reagent with amide-15 N-Gln used in cell culture to label the glycans (see Basic Protocol 1) is not 100% pure, there will be under-incorporation of 15 N for each glycan structure labeled. The under-incorporation rate could be different from one structure to another, and this value is needed when calculating the relative ratios. Thus, it is useful to perform an extra procedure of glycan analysis with the 15 N-labeled sample alone, instead of the mixture of both light and heavy. This protocol describes the mathematical calculations of under-incorporation rate and relative ratios of light/heavy species after we get the N- and O-linked glycan data in separate analyses of the permethylated sugars (see Basic Protocol 1). The discussions in this section will be based on an example spectrum of a 15 N-labeled N-linked glycan structure (Fig. 2A) and a mixture of a light and heavy O-linked glycan structure (Fig. 2B). It should be noted that a software package to automate the calculations is currently under development.
SUPPORT PROTOCOL
Calculations of under-incorporation rate To calculate the under-incorporation rate, determine the ratio of the area of peaks that correspond to under-labeling to the total area of all isotopic peaks resulting from one glycan structure. For example, in Fig. 2A, the area of isotopic peak (labeled as 13 C0 15 N0 , 13 C 15 N 13 C 15 N 13 C 15 N 13 C 15 N 13 C 15 N 13 C 15 N ) is A (i = 1, 2, 3, 4, 5, 6, 0 1, 0 2, 1 2, 2 2, 3 2, 4 2 i 7), so the overall under-incorporation rate (UI) can be calculated as:
∑ UI = ∑
2 i =1 7 i =1
Ai Ai
Equation 1
Quantitative Glycomics Using IDAWG
61 Current Protocols in Chemical Biology
Volume 2
A
13C 15N 0 2
1108.53
100
13C 15N 1 2
1109.03 [M⫹2Na]2⫹⫽ 1108.53 m/z (mono)
90
Relative abundance
80
* *
70
13C 15N 2 2
1109.53
60
Amide-15N-GLN
50 40 13
C315N2 1110.04
30 13C 15N 0 1
20 C0 N0 1107.53
1108.03
1107.5
1108.0
13
10 0 1107.0
B 100
15
13C 15N 4 2
1110.54
1108.5
1109.0 m /z
1109.5
1110.0
1110.5
1111.0
[M⫹Na]⫹: 1256.636 and 1259.627 m /z (mono)
1256.637
1259.626
*
90
⫹
*
*
Relative abundance
80 70
Amide-14N-GLN Amide15N-GLN
1257.639
60
1260.628
50 40
1258.637
30 1261.631
20 10 0 1255
1262.633 1256
1257
1258
1259
1260
1261
1262
1263
1264
m /z
Figure 2 Example mass spectra of IDAWG glycans. (A) Isotopic pattern of 15 N-labeled N-linked glycan structure (Man8 GluNAc2 ). (B) Isotopic pattern of mixture of normal and 15 N-labeled O-linked glycan structure. Green circle: mannose; blue square: GlcNAc; purple diamond: Neu5Ac; yellow circle: galactose; yellow square: GalNAc.
For any glycan structure, if the number of nitrogens in the molecule is K and the number of isotopic peaks shown in the spectrum is M:
∑ UI = ∑ Quantitative Glycomics Using IDAWG
K i =1 M i =1
Ai Ai
Equation 2
The incorporation rate for each nitrogen can be calculated as:
62 Volume 2
Current Protocols in Chemical Biology
incorporation = K (1 − UI ) Equation 3
Calculations of relative ratios 1. Generate the theoretical isotopic patterns for the light structure by software called “emass” written by the Somerharju Lipid Group, University of Helsinki (http://www.helsinki.fi/science/lipids/software.html). 2. In Fig. 2B, there are seven major isotopic peaks observed in the spectrum. Calculate the area of each peak (A1 through A7 ) using peak list and intensities extracted from the spectrum. Due to the peak overlap and under-incorporation, both the light and heavy structures can contribute to the area of each peak. So, if the actual area resulting from light structure is L1 through L7 , and from heavy structure is H1 through H7 , then we have:
L1 + H1 = A1 L2 + H2 = A2 L3 + H3 = A3 L4 + H4 = A4 L5 + H5 = A5 L6 + H6 = A6 L7 + H7 = A7 3. If the ratio of theoretical isotopic pattern for the seven peaks from the light structure is P1 (=1):P2 :P3 :P4 :P5 :P6 :P7 , which can be generated using the software “emass,” we can have:
L1 = L1 × P1 (P1 = 1) L2 = L1 × P2 L3 = L1 × P3 L4 = L1 × P4 L5 = L1 × P5 L6 = L1 × P6 L7 = L1 × P7 4. By the definition of under-incorporation rate shown before:
∑ UI = ∑
3 i =1 7 i =1
Hi Hi
Equation 4
Determine the value of UI based on the spectrum for heavy labeled glycans. Given that Hi = Ai – Li , rearrange Equation 4, with a substitution of Hi into Ai and Li , according to the equations above, to yield Equation 5:
A1 + A 2 + A 3 − UI × ∑ i =1 A i 7
L1 =
P1 + P2 + P3 − UI × ∑ i =1 Pi 7
Equation 5
Quantitative Glycomics Using IDAWG
63 Current Protocols in Chemical Biology
Volume 2
5. After calculation of L1 , calculate Li (i = 2 to 7) and Hi (i = 1 to 7) based on the above equations. The equation for heavy/light ratio will be:
Amide-15 N-Gln = Amide-14 N-Gln
∑ ∑
7 i =1 7 i =1
Hi Li
Equation 6
6. For any glycan structure, if the number of nitrogens in the molecule is K and the number of isotopic peaks shown in the spectrum of light/heavy mixture is M, Equations 5 and 6 can be rewritten as Equations 7 and 8, respectively:
∑ = ∑
K
L1
i =1 K
A i − UI × ∑ i =1 A i M
P − UI × ∑ i =1 Pi M
i =1 i
Equation 7
Amide-15 N-Gln = Amide-14 N-Gln
∑ ∑
M i =1 M i =1
Hi Li
Equation 8
Example calculation Taking the structures in Fig. 2B as an example, the ratio of theoretical isotopic pattern for the seven peaks from the light structure generated by using the software “emass” is P1 (=1):P2 :P3 :P4 :P5 :P6 :P7 = 1:0.6517:0.264:0.0798:0.0198:0.0042:0.0008, and the total is:
∑
7
P = 2.0203
i =1 i
Equation 9
The peak areas can be calculated by extracting the peak list and peak intensities from raw data and using the software “OriginPro 8” (http://www.originlab.com/) and the areas are: A1 = 288182.7; A2 = 180298.2; A3 = 91459.51; A4 = 276748.8; A5 = 153348.9; A6 = 54793.07; A7 = 10981.6. The total is:
∑
7 i =1
A i = 1055813
Equation 10
UI (under-incorporation rate) of this structure is 0.17 (using the calculating method introduced before). Taking all the values into Equation 5, then we have: A1 + A 2 + A 3 − UI × ∑ i =1 A i 7
Quantitative Glycomics Using IDAWG
64 Volume 2
L1 =
P1 + P2 + P3 − UI × ∑ i =1 Pi 7
=
288182.7 + 180298.2 + 91459.51 − 0.17 × 1055813 = 241979.6 1 + 0.6517 + 0.264 − 0.17 × 2.0203 Equation 11 Current Protocols in Chemical Biology
After the calculation of L1 , based on the series of equations under step 3, we can calculate L2 through L7 . Based on the series of equations under step 1, we can then calculate H1 through H7 . The totals are:
∑
7 i =1
Li = L1 +L2 + L3 + L 4 + L5 + L6 + L7 = 488872 Equation 12
7
∑H = H i
1
+ H 2 + H 3 + H 4 + H 5 + H 6 + H 7 = 566941
i=1
Equation 13
Based on Equation 8:
Amide-15 N-Gln = Amide-14 N-Gln
∑ ∑
M i =1 M i =1
Hi Li
=
566941 = 1.16 488872
Equation 14
The value obtained (1.16) represents the relative ratio of this very glycan structure in two samples from different stages of stem cell differentiation. Upon verification by duplicated experiments, we can conclude based on the value 1.16 that the abundance of this glycan structure does not change dramatically between these two stages of stem cell differentiation, since we mix the two samples in a 1:1 ratio.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps.
AG 50W-X8 resin To prepare the resin stock, add the AG 50W-X8 resin (BioRad) to HPLC-grade methanol, stir well, and decant the methanol. Repeat this batch washing procedure twice more, suspend the resin in methanol, and incubate overnight at room temperature. After this, put the resin in a column and wash the resin successively with methanol, 1 M HCl, and 5% acetic acid by pushing the solvents through the column with air. Keep the resin in methanol at 4◦ C as the stock.
Phosphate-buffered saline (PBS) 80.0 g NaCl 2.0 g KCl 14.4 g Na2 HPO4 2.4 g KH2 PO4 Milli-Q water (18.2 M) to 1 liter COMMENTARY Background Information Glycomics comprehensively studies all the glycan structures in a given biological system. The glycans are complex carbohydrates which are oligosaccharide chains usually linked to
proteins and lipids (Aoki-Kinoshita, 2008). Glycosylation is a process of covalent attachment of glycan structures on protein backbones, which is one of the most common post-translational modifications of proteins
Quantitative Glycomics Using IDAWG
65 Current Protocols in Chemical Biology
Volume 2
Quantitative Glycomics Using IDAWG
(Morelle et al., 2009). It has been estimated that glycosylated proteins account for ∼60% to 80% of all mammalian proteins at some point during their existence and nearly 100% of all membrane and secreted proteins (Atwood et al., 2008). There are two main ways that carbohydrate chains are linked to protein backbones: N-linked glycan is linked through the side chain of an asparagine residue present in the tripeptide consensus sequence, Asn-X-Thr/Ser (where X can be any amino acid except proline); O-linked glycan is attached to the oxygen on the side chain of a serine or threonine (Morelle et al., 2009). Each of these glycosylation sites can be attached with many different glycan structures. Glycans often play critical roles in various physiological processes. The important functions of glycans include but are not limited to: modulation of biological activity, cell-cell recognition and interaction, and distribution in tissues and signal transduction (Lowe and Marth, 2003). As affected by different factors, the expression of glycans in a biological system can vary with species, tissue, developmental stage, and even the genetic and physiological state (Atwood et al., 2008). It has also been noticed that altered carbohydrate expression is a common feature of many types of cancers (Dube and Bertozzi, 2005). Moreover, particular protein glycosylations may be altered more specifically or frequently than their underlying core protein in certain disease states, which is a potential advantage of using glycans for diagnostics (Yue et al., 2009). Given all the important roles of glycans in physiological processes, considerable effort has been taken to develop technologies to identify and quantify glycan structures in various environments. It is one of the major challenges in the fields of -omics to develop relative-quantitative analysis technologies that are able to generate meaningful data to compare the expression levels of targeted molecules in different biological samples or developmental states. For proteomics, there are already some powerful tools available. Since mass spectrometry has fulfilled its role as a rapid and reliable method for proteomics studies, MS signal intensities, ion chromatograms, spectral counts, and accurate mass retention time pairs have been used as label-free methods to quantify changes in protein abundances (Wang et al., 2003; Liu et al., 2004; Radulovic et al., 2004; Silva et al., 2005). In addition to label-free methods, stable isotopic labeling methods have become more popular in recent years. Among these labeling
methods, ICAT (Isotope-Coded Affinity Tags) as an in vitro labeling strategy and SILAC (Stable Isotope Labeling with Amino acids in Cell culture) as an in vivo labeling strategy are broadly applied and have been commercialized (Gygi et al., 1999; Ong et al., 2002). Analytical technologies for glycomics are not as mature as those for proteomics. However, the field of glycomics has followed in the steps of proteomics and successfully adapted some of the quantitative proteomics tools for glycomic analysis. For example, total ion mapping, a label-free strategy, allows identification of glycan structures based on fragmentation information from tandem MS, and permits quantification of the prevalence of each glycan structure by normalizing each ion intensity on the mass spectrum to the total (Aoki et al. 2007). As for in vitro isotopic labeling methods, several groups have used heavy methyl iodide (13 CH3 I, 12 CDH2 I, 12 CHD2 I, and/or 12 CD3 I) to label glycans during the permethylation procedure, which is normally performed before MS analysis for both N- and O-linked glycans. The glycan samples containing heavy isotopes are then mixed with samples permethylated using light methyl iodide (12 CH3 I) prior to analysis (Alvarez-Manilla et al., 2007; Aoki et al., 2007; Kang et al., 2007). In 2009, IDAWG (Isotopic Detection of Aminosugars With Glutamine) was reported as the first in vivo–stable isotopic labeling strategy for quantitative glycomics (Orlando et al., 2009). The IDAWG methodology takes advantage of the hexosamine biosynthetic pathway, which uses the side chain of glutamine as the only source of nitrogen when producing aminosugars. As a result, if the cells are fed Gln-free media and glutamine with a 15 N-labeled side chain, all the aminosugars (GlcNAc, GalNAc, and sialic acids) produced in the cells will be labeled with 15 N, and thus the mass of all glycan structures will be shifted by +1 dalton per aminosugar (see Fig. 3). As an in vivo labeling strategy, IDAWG shares some advantages with SILAC over in vitro labeling methods. After the labeling, the 15 Nlabeled sample can be mixed with the normal sample immediately after the cell harvest or the cell lysis. Thus, the glycans from two cell types are subjected to the same experimental conditions for glycan release and permethylation until they are analyzed by MS, which will dramatically reduce technical variability. The IDAWG technology has been successfully used to analyze both N- and O-linked glycans released from murine embryonic stem
66 Volume 2
Current Protocols in Chemical Biology
H2N
OH
O OH
O
O
H2N
(Glc) OH O
O
OH
NH2
O O P
O
OH
O O
OH
OH
HO
O
P
OH
(GLN ) (GLC)
O
O
O
OH O
O
O
P
OH
O
OH O
O HO
OH
OH
(Glc-6-P) OH O HO
O P
P
O
O
OH
(GlcN-6-P) O
O N
O O
NH
O
O
O HO
HO
OH
OH
(Fru-6-P)
OH O
NH2
HO
OH
P
HO
O O
O
O
P
O
HO O
O
OH
NH O
NH
HO
CH3
(UDP-GlcNAc)
OH
*
*
*
NH2
HO OH
O
(GlcNAc-1-P)
CH3
(GlcN-1-P)
⫹1 Da
Figure 3 Schematic of the hexosamine biosynthetic pathway demonstrating that the side chain of glutamine is the only source for nitrogen in the production of aminosugars. If the cells are fed with Gln-free medium and glutamine with a 15 Nlabeled side chain, all the aminosugars including GlcNAc (blue square), GalNAc (yellow square), and sialic acids (purple diamond) produced in cells will be labeled with 15 N, and thus the mass of all glycan structures will be shifted by +1 dalton per aminosugar (Orlando et al., 2009).
cells, and is predicted to be useful for various comparative glycomic studies in the future (Orlando et al., 2009).
Critical Parameters The method of quantitative glycomics of cultured cells using IDAWG consists of six steps: 1. Cell culture: cells are either fed amide14 N-Gln or amide-15 N-Gln. 2. Cell lysis: differently labeled cell populations are lysed, delipidated, and then combined. 3. Release of glycans: N- and O-linked glycans are released from glycopeptides enzymatically or chemically. 4. Permethylation: released glycans are permethylated prior to mass spectrometry analysis. 5. Mass spectrometry analysis and data collection. 6. Calculations of relative ratios. For the cell culture step, labeling using amide-15 N-Gln for 3 days should be sufficient for embryonic stem cells to achieve a good de-
gree of incorporation. However, for those cells with a relatively slower metabolic rate, longer labeling times may be required to attain a high degree of labeling. A recommendation of 7 days is advised for new cell lines. For the cell lysis step, there are three parameters to consider. First, the delipidating solvent should have the ratio of chloroform:methanol:water equal to 4:8:3, or it will introduce layers into the solvent and impair delipidation. Second, washing with ice-cold acetone and water after delipidation is crucial because acetone helps precipitate proteins and water can dissolve oligomeric hexose ladders, which are common contaminants in glycan analyses. Third, if it is possible to count the cell number, mixing equal amounts of the two cell populations together immediately after cell culture based on cell number instead of by protein weight is also an option when performing IDAWG experiments. For releasing O-linked glycans, an extra cleanup via C18 columns may be added to the protocol to help remove the borate. The sample can be dissolved in 5% acetic acid,
Quantitative Glycomics Using IDAWG
67 Current Protocols in Chemical Biology
Volume 2
loaded onto equilibrated columns, and eluted with 5% acetic acid. Throughout all the procedures, when drying the sample under a stream of nitrogen gas, always keep the drying time as short as possible. Overdrying will likely reduce the yield of the experiments because some product could be volatilized or displaced from the tube by the nitrogen gas after the solution is dried.
Troubleshooting Low 15 N incorporation rate The 15 N labeling time in cell culture may be too short. Label embryonic stem cells for 3 days and label differentiated cell types for at least 7 days to obtain sufficiently high (>90%) incorporation rates. No glycans detected in mass spectrum 1. The enzymatic digestion or chemical reaction may be incomplete. Make sure the correct amounts of enzyme or chemical reagents are used and that appropriate conditions are applied (see Basic Protocol 2). Multiple methods exist, including commercial glycoprotein stains or lectin blotting of proteins, that can be used to confirm if deglycosylation has been successful. 2. Excessive contamination might have been introduced throughout the procedures. Make sure the protein powder is washed with acetone and water properly. All the microcentrifuge tubes, glass tubes, and pipets must be precleaned with methanol, or the polymer peaks can dominate the spectra.
Anticipated Results
Quantitative Glycomics Using IDAWG
Example spectra are shown in Figure 2. The relative ratio of any glycan structure with a similarly complete isotopic pattern (typically with more than two isotopic peaks showing up) can be calculated using IDAWG. For example, more than 125 N-linked glycans and 35 O-linked glycans have been identified by the authors’ group for mouse embryonic stem cells using these protocols. Among these glycan structures, those that could be reliably relatively quantified by IDAWG were approximately half (about 65 structures for N-linked glycans and 20 for O-linked glycans; these glycans have more than two isotopic peaks in the spectra, which make the calculation reliable). Of course, some of these quantified glycans are in fact isobaric mixtures, and this should be taken into account when interpreting results. IDAWG is not able to separately quantify the isobaric structures, and thus a ratio is
calculated based on changes in a set of isobaric structures.
Time Considerations At least 72 hr (and sometimes longer than a week) are needed to label the cells. The cell lysis and delipidation require incubation of 5 hr to overnight. The release of N-linked glycans requires two overnight incubations for digestion, and the release of O-linked glycans requires one overnight incubation for the chemical reaction. The time for drying samples in the Speed-Vac evaporator is variable depending on the efficiency of the Speed-Vac and the solvent volume used. It takes less than 5 hr to perform permethylation and mass spectrometry analysis. Thus, the entire procedure, excluding the initial labeling and data interpretation, takes 3 to 4 days. The authors of this unit, working with others in the field, are currently attempting to automate the calculations for under-incorporation and relative ratios. Furthermore, the introduction of spiked standard glycans is being explored so that IDAWG labeling can be used to follow glycan turnover, remodeling, and synthesis.
Acknowledgements We would like to thank all members of the Wells, Orlando, Tiemeyer, Moremen, York, Dalton, and Pierce laboratories for helpful discussions. This work is supported in part by grants from NIH/NCRR 5P41RR018502 (L.W., senior investigator), and NIH/NIDDK 1R01DK075069 (L.W.).
Literature Cited Alvarez-Manilla, G., Warren, N.L., Abney, T., Atwood, J. 3rd, Azadi, P., York, W.S., Pierce, M., and Orlando, R. 2007. Tools for glycomics: Relative quantitation of glycans by isotopic permethylation using 13CH3I. Glycobiology 17:677687. Aoki-Kinoshita, K.F. 2008. An introduction to bioinformatics for glycomics research. PLoS Comput. Biol. 4:e1000075. Aoki, K., Perlman, M., Lim, J.M., Cantu, R., Wells, L., and Tiemeyer, M. 2007. Dynamic developmental elaboration of N-linked glycan complexity in the Drosophila melanogaster embryo. J. Biol. Chem. 282:9127-9142. Atwood, J.A. 3rd, Cheng, L., Alvarez-Manilla, G., Warren, N.L., York, W.S., and Orlando, R. 2008. Quantitation by isobaric labeling: Applications to glycomics. J. Proteome Res. 7:367-374. Ceroni, A., Maass, K., Geyer, H., Geyer, R., Dell, A., and Haslam, S.M. 2008. GlycoWorkbench: A tool for the computer-assisted annotation of mass spectra of glycans. J. Proteome Res. 7:1650-1659.
68 Volume 2
Current Protocols in Chemical Biology
Dube, D.H. and Bertozzi, C.R. 2005. Glycans in cancer and inflammation: Potential for therapeutics and diagnostics. Nat. Rev. Drug Discov. 4:477-488.
Pierce, M., Dalton, S., and Wells, L. 2009. IDAWG: Metabolic incorporation of stable isotope labels for quantitative glycomics of cultured cells. J. Proteome Res. 8:3816-3823.
Gygi, S.P., Rist, B., Gerber, S.A., Turecek, F., Gelb, M.H., and Aebersold, R. 1999. Quantitative analysis of complex protein mixtures using isotope-coded affinity tags. Nat. Biotechnol. 17:994-999.
Radulovic, D., Jelveh, S., Ryu, S., Hamilton, T.G., Foss, E., Mao, Y., and Emili, A. 2004. Informatics platform for global proteomic profiling and biomarker discovery using liquid chromatography-tandem mass spectrometry. Mol. Cell. Proteomics 3:984-997.
Kang, P., Mechref, Y., Kyselova, Z., Goetz, J.A., and Novotny, M.V. 2007. Comparative glycomic mapping through quantitative permethylation and stable-isotope labeling. Anal. Chem. 79:6064-6073. Liu, H., Sadygov, R.G., and Yates, J.R. 3rd. 2004. A model for random sampling and estimation of relative protein abundance in shotgun proteomics. Anal. Chem. 76:4193-4201. Lowe, J.B. and Marth, J.D. 2003. A genetic approach to mammalian glycan function. Annu. Rev. Biochem. 72:643-691. Morelle, W., Faid, V., Chirat, F., and Michalski, J.C. 2009. Analysis of N- and O-linked glycans from glycoproteins using MALDI-TOF mass spectrometry. Methods Mol. Biol. 534:5-21. Ong, S.E., Blagoev, B., Kratchmarova, I., Kristensen, D.B., Steen, H., Pandey, A., and Mann, M. 2002. Stable isotope labeling by amino acids in cell culture, SILAC, as a simple and accurate approach to expression proteomics. Mol. Cell. Proteomics 1:376-386. Orlando, R., Lim, J.M., Atwood, J.A. 3rd, Aangel, P.M., Fang, M., Aoki, K., Alvarez-Manilla, G., Moremen, K.W., York, W.S., Tiemeyer, M.,
Silva, J.C., Denny, R., Dorschel, C.A., Gorenstein, M., Kass, I.J., Li, G.Z., McKenna, T., Nold, M.J., Richardson, K., Young, P., and Geromanos, S. 2005. Quantitative proteomic analysis by accurate mass retention time pairs. Anal. Chem. 77:2187-2200. Wang, W., Zhou, H., Lin, H., Roy, S., Shaler, T.A., Hill, L.R., Norton, S., Kumar, P., Anderle, M., and Becker, C.H. 2003. Quantification of proteins and metabolites by mass spectrometry without isotopic labeling or spiked standards. Anal. Chem. 75:4818-4826. Yue, T., Goldstein, I.J., Hollingsworth, M.A., Kaul, K., Brand, R.E., and Haab, B.B. 2009. The prevalence and nature of glycan alterations on specific proteins in pancreatic cancer patients revealed using antibody-lectin sandwich arrays. Mol. Cell. Proteomics 8:1697-1707.
Key Reference Orlando et al., 2009. See above. Initial description of the IDAWG labeling strategy and showed proof-of-principle utility using murine embryonic stem cells.
Quantitative Glycomics Using IDAWG
69 Current Protocols in Chemical Biology
Volume 2
Targeted Degradation of Proteins by PROTACs Eun Ryoung Jang,1 Wooin Lee,1 and Kyung Bo Kim1 1
University of Kentucky, Lexington, Kentucky
ABSTRACT In recent years, small interference RNAs (siRNAs) have greatly enhanced our understanding of protein functions by allowing knockdown of targeted proteins at the mRNA level. Similarly, in an effort to achieve degradation of targeted proteins at the posttranslational level, chimeric small molecules called “PROTACs” (PROteolysis TArgeting Chimeric molecules) have been developed. The PROTAC approach utilizes chimeric small molecules which recruit targeted proteins to the ubiquitin-proteasome pathway, a major intracellular protein degradation system. Unlike conventional small molecules that bind to protein and inhibit its function, the PROTAC approach induces destruction of target protein via the ubiquitin-proteasome system. This article presents a typical strategy for PROTAC design and preparation and biological characterization. Curr. Protoc. Chem. C 2010 by John Wiley & Sons, Inc. Biol. 2:71-87 Keywords: PROTAC r ubiquitin r proteasome r small molecule
INTRODUCTION Small-Molecule Approaches to Protein Functions Classical gene-knockout approaches have served as powerful tools in investigating protein functions by examining the phenotypes associated with the selective loss of targeted proteins through genetic manipulations. While this genetic approach has provided useful insights into protein functions, the lack of spatial and temporal control and the possibility of indirect compensatory responses are still major concerns in dissecting complex cellular processes. In recent years, small interference RNA (siRNA)–based knockdown strategies have provided useful tools for the elucidation of protein functions. However, they also have similar limitations as well as difficulties associated with cellular delivery of siRNA. In recent years, small molecules have been increasingly used as molecular probes in the investigation of cellular processes and protein functions (Mitchison, 1994; Schreiber, 1998; Crews and Splittgerber, 1999; Crews and Mohan, 2000). Typically, small molecules directly bind to target proteins and perturb the functions of target proteins at the posttranslational level (Kim et al., 1999; Meng et al., 1999a,b; Yeh et al., 2000; Lin and Cornish, 2001). In most cases, the effects of small molecules are reversible and can be readily regulated, providing flexible spatial and temporal control, which is critical for dissecting complex protein functions (Corson et al., 2008). As compared to the gene-knockout approach, small-molecule-induced knockdown of target proteins can be particularly useful in investigating proteins that play crucial roles in embryonic development. In addition, small molecules that induce protein knockdown may offer therapeutic potential by making possible the controlled elimination of disease-related or diseasepromoting proteins.
Ubiquitin-Proteasome System (UPS) Controlled degradation of proteins is crucial for maintaining cellular protein homeostasis and regulating numerous cellular processes, such as inflammatory responses, gene Current Protocols in Chemical Biology 2: 71-87, April 2010 Published online April 2010 in Wiley Interscience (www.interscience.wiley.com). DOI: 10.1002/9780470559277.ch090242 C 2010 John Wiley & Sons, Inc. Copyright
Targeted Degradation of Proteins by PROTACs
71 Volume 2
transcription, DNA repair, cell-cycle control, and apoptosis (Ciechanover, 2005; Orlowski and Kuhn, 2008). In cells, the ubiquitin-proteasome system (UPS) plays a major role in regulating degradation of proteins. The UPS-dependent proteolysis typically involves the attachment of a polyubiquitin chain to proteins and their subsequent degradation by the 26S proteasome (Hershko and Ciechanover, 1998). Ubiquitin is a highly conserved polypeptide of 76 amino acids and is attached to target proteins as a result of an enzymatic cascade which typically involves three steps. Ubiquitin is first activated at its C-terminus by adenylation and formation of a thioester bond with the ubiquitin-activating enzymes (E1). Activated ubiquitin is subsequently transferred from E1 to a cysteine residue of the ubiquitin-conjugating enzymes (E2). Finally, ubiquitin is transferred from E2 to a lysine residue of a target protein, either directly or with the assistance of the ubiquitin ligases (E3). The E3 ubiquitin ligases appear to control the substrate specificity of the ubiquitination reaction as they directly bind to their target proteins (Deshaies, 1999). Conjugation of additional ubiquitin molecules to previously attached ubiquitin units on the targeted protein results in a polyubiquitin chain that is recognized by the 26S proteasome, ultimately leading to degradation of the target protein.
Proteolysis Targeting Chimeric Molecules (PROTACs): Use of E3 Ubiquitin Ligase-Substrate Interactions PROTACs (PROteolysis TArgeting Chimeric molecules) are a class of small molecules that target proteins of interest for degradation via the UPS. PROTACs are composed of an E3 ubiquitin ligase recognition motif on one end and a small-molecule ligand for a target protein on the other end, thereby recruiting a targeted protein to an E3 ligase complex for ubiquitination and subsequent degradation by the proteasome (Fig. 1). Numerous PROTACs have been successfully developed as summarized in Figure 2.
PROTAC E3 ligase recognition domain ligand for target protein target protein
linker
E3 ligase E2
PROTAC
E3 ligase
target protein
Ub
Ub Ub Ub Ub Ub
E2 Ub
target protein degradation proteasome
Targeted Degradation of Proteins by PROTACs
Figure 1 Overview of the PROTAC system. PROTACs act as a “bridging molecule” that brings together the E3 ubiquitin ligase with a target protein, leading to polyubiquitination and subsequent degradation of the protein by the proteasome.
72 Volume 2
Current Protocols in Chemical Biology
Target protein & E3 ligase
Ligand for target protein
PROTAC O
MetAP-2, SCFβ-TRCP (Sakamoto et al., 2001)
CH3 O OMe O
Fumagillin O
H N
N H
O
lκ Bα phosphopeptide O O
ER, SCFβ-TRCP (Sakamoto et al., 2003)
O
H N
O
NH O
Estradiol (E2)
lκ Bα phosphopeptide
HO O
GFP-AR, SCFβ-TRCP (Sakamoto et al., 2003)
DHT
O
H N
O
NH O
H H
H O
H O
DHT
NH O
H
ALA PYIP – (D-A[g])8
H
H O
O
H N
O
GFP-AR, pVHL (Schneekloth et al., 2004)
lκ Bα phosphopeptide
H
MeO
O
O
O
O
N H
O O
EGFP-FKBP12, pVHL (Schneekloth et al., 2004)
O
H N
MeO
HN (ALAPYIP)
AP21998
N
(D-Ang)5CONH2
MeO
OMe OMe O
ER, pVHL (Bargagna-Mohan et al., 2005; Zhang et al., 2004a)
O
H N
N HF-1α pentapeptide H
O O
Estradiol (E2) HO
O
O
AHR, pVHL (Lee et al., 2007; Puppala et al., 2008)
O
NH
O
Apigenin
HIF-1α pentapeptide OH
O Cl
O2N F3C
AR,MD M2 (Schneekloth et al., 2008)
Flutamide
H N
O N H
O O
O
O
H N
Cl
O
O
N
OH O
N
N
N O
O
Nutlin O
Figure 2
Examples of PROTACs.
The following sections describe the detailed procedure for PROTAC development, starting with the general strategy for PROTAC design and synthesis (see Basic Protocols 1 to 3), followed by the activity assay of the PROTAC molecule (see Basic Protocol 4). Figure 3 provides a flowchart of the entire process.
Targeted Degradation of Proteins by PROTACs
73 Current Protocols in Chemical Biology
Volume 2
selected ligand
introduction of handle (e.g., free amine or acid groups)
1. ligand binding assay 2. coupling of E3 ligase recognition motif
cell-based E3 ligase verification assay
PROTAC
control PROTAC (inactive)
cell-based protein degradation assay
verification of E3 ligasedependent degradation
assay for downstream effects of target protein degradation
Figure 3
A flowchart of PROTAC synthesis and assay.
PROTAC design and synthesis General strategy for PROTAC synthesis: A flowchart of PROTAC design and assay procedure is shown in Figure 3. Typically, chimeric PROTAC molecules are prepared through a convergent approach, first synthesizing two fragments separately and then coupling them to produce a fully assembled PROTAC. Several PROTACs have been prepared using this convergent approach. Here, the synthesis of the one fragment (E3 recognition motif) is first described, followed by the preparation of the second fragment (small-molecule ligand) with a handle that is used for the final coupling. Finally, a representative example of a two-fragment coupling experiment is described. BASIC PROTOCOL 1
SYNTHESIS OF E3 LIGASE RECOGNITION MOTIF (H2 N-Leu-Ala-ProOH -Tyr-Ile-OBzl) As described above, two E3 ubiquitin ligase recognition motifs have been used in the preparation of PROTACs: a phosphopeptide motif derived from IκBα, which is recognized by E3 ligase SCFβ-TRCP , and a pentapeptide motif derived from HIF-1α, which is recognized by E3 ligase pVHL. Overall, the syntheses of these peptide motifs are accomplished by following conventional peptide synthesis approaches. While both peptide motifs have been shown to interact and initiate ubiquitination of target proteins, the pentapeptide motif is gaining more popularity due to its superior cell permeability and bioavailability compared to the phosphopeptide motif. Therefore, the synthesis of pVHLinteracting pentapeptide is described below. Although the synthesis of the pentapeptide motif can be accomplished through solid-phase peptide chemistry, it is recommended to carry out solution-phase chemistry, due to its ability to produce the pure pentapeptide on a gram scales.
Materials Targeted Degradation of Proteins by PROTACs
NH2 -Isoleucine-O-Bzl (Advanced ChemTech; http://www.advancedchemtech.com/) FmocHN-Tyr(O-t-Bu)-CO2 H (Advanced ChemTech)
74 Volume 2
Current Protocols in Chemical Biology
HBTU (2-(1H-Benzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate) (Advanced ChemTech) HOBt (N-Hydroxybenzotriazole; Advanced ChemTech) Methylene chloride (CH2 Cl2 ; Fluka; dried over calcium hydride) DIPEA (N,N-diisopropylethylamine; Sigma-Aldrich) Flash silica gel 60 (particle size 0.040-0.063 mm; Merck) Ethyl acetate (Fluka) n-Hexane (Fluka) Piperidine (Sigma-Aldrich) Anhydrous dimethylformamide (DMF; Sigma-Aldrich) FmocHN- OH Pro-CO2 H (Advanced ChemTech) FmocHN-Leu-CO2 H (Advanced ChemTech) FmocHN-Ala-CO2 H (Advanced ChemTech) TBAF (t-butyl ammonium fluoride; Sigma-Aldrich) TFA (trifluoroacetic acid, Sigma-Aldrich) Anydrous nitrogen or argon gas Round-bottom flasks High-vacuum source 60 F254 precoated silica gel TLC plates (Merck) Rotary evaporator (Buchi) Chromatography columns 1. Place NH2 -isoleucine-O-Bzl (1 equivalent) and FmocHN-Tyr(O-t-Bu)-CO2 H (1 equivalent) in a round-bottom flask. 2. Add HBTU (2 equivalents) and HOBt (2 equivalent). 3. Add 40 ml CH2 Cl2 per gram of amino acid, and then DIPEA (N,Ndiisopropylethylamine; 5 equivalents). At this stage, amino acids and coupling reagents are not completely dissolved in CH2 Cl2 (it may look like an emulsion). As the reaction proceeds, the emulsion becomes clearer or transparent, eventually becoming a clear solution.
4. Stir the reaction mixture at room temperature overnight, using a magnetic bar and stirrer. Typically, the progress of the reaction is monitored by TLC (thin-layer chromatography). After spotting the reaction mixture using a capillary tube, the TLC plate is placed in a TLC chamber and eluted using a solvent mixture (typically, 1:1 ethyl acetate:n-hexane, v/v).
5. Concentrate the reaction solvent using a rotary evaporator to yield an oily crude product. 6. Purify the resulting FmocHN-Tyr(O-t-Bu)-Ile-OBzl by flash column chromatography using Flash silica gel 60. It is important to completely dissolve the resulting oily product using a minimum amount of CH2 Cl2 (or a mixture of CH2 Cl2 and methanol) before loading onto silica gel chromatography. Then, the product containing CH2 Cl2 (or a mixture of CH2 Cl2 and methanol) is loaded onto Flash silica gel 60 for purification. The dipeptide product is eluted using a mixture of ethyl acetate and hexane (1:1, v/v).
7. Treat the purified dipeptide (FmocHN-Tyr(O-t-Bu)-Ile-OBzl) with 20% piperidine in DMF (excess) for 5 min to remove the Fmoc group and subsequently purify the resulting product, H2 N-Tyr(O-t-Bu)-Ile-OBzl, by flash column chromatography using Flash silica gel 60.
Targeted Degradation of Proteins by PROTACs
75 Current Protocols in Chemical Biology
Volume 2
The use of a minimum amount of Flash silica gel 60 is recommended for purification by flash column chromatography. The Fmoc-deprotected product is not normally eluted by typical hexane-ethyl acetate solvent system, but most of Fmoc adducts formed during the reaction are readily eluted from the column by the solvent system. After loading crude product onto silica, the silica is washed using ethyl acetate-hexane system (1:1, v/v) to remove other adducts, and then polar solvent [such as a mixture of CH2 Cl2 -methanol (9:1, v/v)] is used to elute the Fmoc-deprotected product.
8. Place H2 N-Tyr(O-t-Bu)-Ile-OBzl (from step 7; 1 equivalent) and FmocHN-HO ProCO2 H (1 equivalent) in a round-bottom flask. 9. Add HBTU (2 equivalents) and HOBt (2 equivalents). 10. Add 40 ml CH2 Cl2 per gram of dipeptide and DIPEA (5 equivalents) and stir the reaction mixture at room temperature overnight. 11. Concentrate the reaction mixture using a rotary evaporator and purify FmocHNHO Pro-Tyr(O-t-Bu)-Ile-OBzl by a flash column chromatography as described in previous steps. 12. Repeat the process (using the techniques described in steps 1 through 7) to obtain the pentapeptide by sequentially attaching the remaining Fmoc-protected amino acids (FmocHN-Leu-CO2 H, FmocHN-Ala-CO2 H; substitute for the Fmoc-protected compound in step 1) to give H2 N-Leu-Ala-HO Pro-Tyr(O-t-Bu)-Ile-OBzl. 13. Dissolve the final product in CH2 Cl2 in a round-bottom flask. 14. Add 2 equivalents TBAF (t-butyl ammonium fluoride) and stir the reaction mixture at room temperature for 5 min. Subsequently, add TFA (10 equivalents) and CH2 Cl2 (40 ml/g starting material). 15. Remove TFA and CH2 Cl2 using a rotary evaporator to yield the pentapeptide. It is important to completely remove TFA from the crude product by repeatedly dissolving and evaporating using CH2 Cl2 , and store under anhydrous nitrogen or argon gas at −20◦ C until the final coupling with a small-molecule ligand for the targeted protein.
16. Dry the t-Bu-deprotected pentapeptide under high vacuum and characterize the pentapeptide with 1 H NMR and mass spectrometry. Store the pentapeptide at −20◦ C under anhydrous nitrogen atmosphere until later use. The pentapeptide is stable at −20◦ C for many years. BASIC PROTOCOL 2
ATTACHMENT OF AN AMINE (OR FREE ACID) HANDLE TO A SMALL-MOLECULE LIGAND IN PREPARATION FOR SYNTHESIS OF A PROTAC MOLECULE Linking a free acid or amine handle to a small-molecule ligand is critical for the preparation of a PROTAC molecule. The strategy for handle introduction into a small-molecule ligand must be determined individually for each case. If a ligand already has functional groups such as free acid or primary or secondary amine, then those groups can be used to attach a linker with a functional group that is then used to couple the ligand to an E3 ligase recognition motif. Otherwise, it requires organic manipulation to introduce a handle containing particular functional groups (Fig. 4).
Targeted Degradation of Proteins by PROTACs
It is critical to make sure that a handle-attached small-molecule ligand still maintains the same binding affinity as that of the parent ligand for target protein, using biological assays or direct binding assays.
76 Volume 2
Current Protocols in Chemical Biology
small-molecule ligand
CO2H
C O NH
NH2
NH2
NH
NH
CO2H
C O
C
O
C
O
NH
C
O
C
O
C
linker C
O
O
O O
N
O
O
N
O O
O
N
O
O O
O
N
O
E3 ligase recognition motif (pentapeptide) PROTAC
Figure 4 Introduction of a functional group (free acid or amine) into a small-molecule ligand. The introduced functional group is activated to couple with an E3 ligase recognition motif containing a primary amine.
After the preparation of E3 ligase recognition motif (the pentapeptide containing hydroxyproline in Basic Protocol 1), the synthesis of small-molecule ligands with a “handle” is the next step. PROTACs can then be easily synthesized through coupling those two motifs using conventional peptide chemistry. This protocol describes a representative example in which an activated free amine group is introduced into dihydrotestosterone (DHT), a natural agonist of the androgen receptor (AR), and used for the synthesis of PROTAC that targets the AR.
Materials FmocHN-Gly-CO2 H (Advanced ChemTech) Methylene chloride (CH2 Cl2 ; Fluka; dried over calcium hydride) Anhydrous dimethylformamide (DMF; Sigma-Aldrich) Oxalyl chloride (ClC(=O)C(=O)Cl; Sigma-Aldrich) Anhydrous nitrogen Dihydrotestosterone (DHT, Sigma-Aldrich) Dimethylaminopyridine (DMAP, Aldrich) Flash silica gel 60 (particle size 0.040 to 0.063 mm; Merck) Ethyl acetate (Fluka) n-Hexane (Fluka) TBAF (t-butyl ammonium fluoride) (Sigma-Aldrich) Disuccinimidyl suberate (Pierce)
Targeted Degradation of Proteins by PROTACs
77 Current Protocols in Chemical Biology
Volume 2
Round-bottom flasks High-vacuum source Rotary evaporator (Buchi) Chromatography columns 1. Place FmocHN-Gly-CO2 H (4 equivalents) in a round-bottom flask. 2. Add 40 ml CH2 Cl2 per gram of amino acid and then 1 drop of DMF and oxalyl chloride (8 equivalents) at 0◦ C. 3. Stir the reaction mixture for 3 hr at room temperature. 4. Remove CH2 Cl2 . It is important to evaporate CH2 Cl2 under anhydrous condition such as a stream of nitrogen gas, since the reaction intermediate FmocHN-Gly-C(=O)-Cl is highly reactive to water. A rotary evaporator may be used to remove CH2 Cl2 under an anhydrous atmosphere. After that, further drying under high vacuum is recommended to remove DMF and any unreacted oxalyl chloride. It is critical to completely remove any unreacted oxalyl chloride before proceeding to next step.
5. Redissolve the resulting yellowish solid in CH2 Cl2 and transfer to a round-bottom flask containing DHT (1 equivalent) and DMAP (4 equivalents) at 0◦ C. 6. Stir the reaction mixture overnight at room temperature. 7. Remove CH2 Cl2 using a rotary evaporator and purify the resulting DHT-GlyNHFmoc by flash column chromatography (with Flash silica gel 60), using a mixture of ethyl acetate and hexane system (1:1, v/v). 8. Place DHT-Gly-NHFmoc (1 equivalent) in a round-bottom flask containing CH2 Cl2 . 9. Treat DHT-Gly-NHFmoc in CH2 Cl2 with TBAF (2 equivalents) at room temperature for 20 min to remove Fmoc group. 10. Remove CH2 Cl2 using a rotary evaporator and dry under high vacuum. 11. Dissolve DHT-Gly-NH2 (from step 7; 1 equivalent) in anhydrous DMF. 12. Add disuccinimidyl suberate (5 equivalents) and stir overnight at room temperature. At this step, the linker length can be adjusted by using a variety of disuccinimidyl derivatives. It will be necessary to determine empirically the optimum linker length that will confer optimum ubiquitin transfer from the E3 ligase complex to the target protein.
13. Remove DMF under high vacuum and purify the resulting crude product, DHT-Glymonosuccinimidyl suberate, by flash chromatography and dry under high vacuum. Even under high vacuum, it may take some time to completely remove DMF. It is recommended not store the crude product for a long time, since the product is highly reactive to aqueous moiety.
14. Dry DHT-Gly-monosuccinimidyl suberate under high vacuum and characterize using 1 H NMR and mass spectrometry.
Targeted Degradation of Proteins by PROTACs
It is recommended to purge the vial containing DHT-Gly-monosuccinimidyl substrate using anhydrous nitrogen or argon gas and seal it with parafilm to maintain an anhydrous atmosphere and prevent unwanted reaction between water and DHT-Gly-monosuccinimidyl substrate. Use the succinimidyl derivative as soon as possible, due to its high reactivity with water.
78 Volume 2
Current Protocols in Chemical Biology
ASSEMBLY OF PROTACs: LINKING DHT-NH2 TO THE pVHL RECOGNITION MOTIF (PENTAPEPTIDE)
BASIC PROTOCOL 3
The final assembly between a E3 recognition motif and a ligand specific for a target protein is typically performed in DMSO or DMF due to the poor solubility of E3 recognition peptide motif in common organic solvents such as CH2 Cl2 . All PROTACs previously developed have been synthesized by this convergent approach (Fig. 5). Compared to a multistep linear synthesis of PROTACs, the convergent synthesis provides the fully assembled PROTACs with high efficiency. It is important to note that intermediates containing an E3 recognition peptide motif are also insoluble in solvents such as CH2 Cl2 or THF (tetrahydrofuran), widely used for organic reactions. Therefore, it is recommended to prepare a target protein ligand motif and an E3 recognition motif independently, and to couple them in DMSO or DMF.
small-molecule ligand
E3 ligase (pVHL) recognition motif O H2N
IIe
C
functional group introduction
O
Bzl
O
Bzl
O H2N
Tyr
IIe
C
OH
O
NH2 (or CO2H) H2N
functional group activation
Pr o
Tyr
IIe
OH H2N
Ala
O
C
O
Pro
Tyr
IIe
C
Leu
Ala
Pro
O Tyr
IIe
OH C
H N
Leu
Ala
Pro
Bzl
O
OH H2N
Bzl
O
C
Bzl
O Tyr
IIe
C
O
Bzl
O fully assembled PROTAC
Figure 5 A convergent approach to PROTAC synthesis. Two fragments are synthesized separately and subsequently coupled to yield a fully assembled PROTAC.
Targeted Degradation of Proteins by PROTACs
79 Current Protocols in Chemical Biology
Volume 2
Materials DHT-Gly-monosuccinimidyl suberate (Basic Protocol 2) Anhydrous dimethylformamide (DMF; Sigma-Aldrich) pVHL-recognizing pentapeptide (Basic Protocol 1) Ninhydrin Phenol Potassium cyanide Ethanol Pyridine Methylene chloride (CH2 Cl2 ; Fluka; dried over calcium hydride) Methanol 60 F254 precoated silica gel TLC plates (Merck) Glass vial High-vacuum source Additional reagents and equipment for mass spectrometry (Carr and Annan, 1996) and HPLC (Boysen and Hearn, 2001) 1. Dissolve DHT-Gly-monosuccinimidyl suberate (1.5 equivalents) in anhydrous DMF. 2. Add DMF solution containing the pVHL-recognizing pentapeptide (1 equivalent), which has a N-terminus primary amine and DMAP (1 equivalent). 3. Stir the reaction mixture for 20 min at room temperature. 4. Perform Kaiser test to confirm the completion of coupling reaction (the Kaiser test detects unreacted primary amine of the pentapeptide). a. Prepare test solutions: i. Test solution A: 5% (w/v) ninhydrin in ethanol ii. Test solution B: 4:1 (w/v) phenol:ethanol iii. Test solution C: 2% (v/v) of 1 mM aqueous potassium cyanide in pyridine b. Spot the reaction solution on a TLC plate or paper using a capillary tube and allow to dry. c. Place the TLC plate in a glass vial and add 4 drops of solution A, 2 drops of solution B, and 2 drops of solution C to the vial (make sure the spotted area is wet with the mixture of solutions A, B, and C. d. Heat the TLC plate at 100◦ C for 5 min. A blue coloration in the spotted area indicates unreacted primary amines and signals that the coupling reaction is not complete.
5. If the result from Kaiser test is negative, remove DMF under high vacuum to yield a yellowish crude product, which is a fully assembled PROTAC that targets the AR for degradation. 6. Place the product in a 10-ml vial and dry under high vacuum. Add a mixture of CH2 Cl2 /MeOH (95:5, v/v) and gently stir to dissolve unreacted starting materials and reagents. Decant off the CH2 Cl2 /MeOH mixture. Repeat this procedure three times and dry the remaining insoluble solid under high vacuum to give a solid product.
Targeted Degradation of Proteins by PROTACs
7. Confirm the identity of the fully assembled PROTAC using electrospray ionization mass spectrometry (ESI-MS; see, e.g., Carr and Annan, 1996). ESI-MS works best for PROTAC identification.
80 Volume 2
Current Protocols in Chemical Biology
8. Once the formation of the fully assembled PROTAC is confirmed, purify it using a normal-phase (silica-based column) HPLC (see, e.g., Boysen and Hearn, 2001). A gradient solvent system (hexane–isopropyl alcohol) for HPLC is recommended for best results. For PROTACs containing highly polar motifs, a reversed-phase HPLC (wateracetonitrile-TFA) is recommended.
ASSAYS FOR PROTAC-INDUCED TARGETED PROTEIN DEGRADATION: IMMUNOBLOT ANALYSES
BASIC PROTOCOL 4
The ability of PROTACs to target proteins for UPS degradation can easily be tested by measuring the cellular level of targeted proteins after treatment of PROTACs, via immunoblot analysis. It is also recommended to use PROTAC analogs modified at the E3 recognition motif to demonstrate that the action of PROTACs is dependent on the designated E3 ligase. For example, for pVHL recognition motif–based PROTACs, its chemical mutant can be designed to include norleucine instead of hydroxyl proline within the E3 ligase recognition motif, resulting in the loss of interactions with pVHL E3 ligase and thus no ubiquitination or degradation. In addition, accumulation of targeted proteins by treatment of proteasome inhibitors in PROTAC-treated cells provides evidence that the activity of PROTAC may be proteasome-dependent. The following immunoblotting protocol describes the detection of AR degradation after the PROTAC treatment.
Materials LNCaP cells (ATCC, cat. no. CRL-1740) Cell culture medium (see recipe) PROTAC (Basic Protocol 3) Negative PROTAC control (OH Pro→norLeu; synthesized following Basic Protocol 1, except using norleucine instead of hydroxyproline) Vehicle (DMSO) Lysis buffer (see recipe) Protein Assay Dye Reagent Concentrate (BioRad, cat. no. 500-0006) Laemmli Sample Buffer 2× Concentrate (Sigma-Aldrich, cat. no. S3401) 10% to 12% SDS-PAGE gel (Gallagher, 1995) Prestained SDS-PAGE Standards, Low Range (BioRad, cat. no. 161-0305) Blocking solution: 5% (w/v) blocking grade nonfat dry milk (BioRad, cat. no. 170-6404) in PBS (see recipe for PBS) Anti-AR antibody (Upstate Biotechnology Inc., cat. no. AB561) Phosphate-buffered saline (PBS; see recipe) containing 0.05% (v/v) Tween 20 Anti-IgG horseradish peroxidase (Zymed, cat. no. 81-6100) Anti-β-actin antibody (Sigma-Aldrich, cat. no. A2066) Amersham ECL Western Blotting Detection Reagents (GE Healthcare Life Sciences, cat. no. RPN2209) 24-well tissue culture plates Refrigerated centrifuge Heat block Platform rotator Immuno-Blot PVDF Membrane (BioRad, cat. no. 162-0177) Kodak BioMax XAR films (Sigma-Aldrich, cat. no. F5513) Additional reagents and equipment for SDS-PAGE (Gallagher, 1995) electrotransfer of proteins from gels to membranes (Ursitti et al., 1995) 1. Maintain LNCaP cells in cell culture medium. 2. Plate cells onto 24-well plates at least 24 hr prior to PROTAC treatment at 2 × cells/ml in 1 ml LNCaP cell medium.
105
Targeted Degradation of Proteins by PROTACs
81 Current Protocols in Chemical Biology
Volume 2
3. Once cells reach ∼70% confluence, treat them with varying concentrations of PROTAC, negative PROTAC control (OH Pro→norLeu), and vehicle (DMSO), and incubate cells under standard conditions of temperature (37◦ C), humidity (95%), and carbon dioxide (5%) for 12 to 48 hr. 4. Lyse cells using lysis buffer with 1% protease inhibitor cocktail and phosphatase inhibitor cocktail. Let cell lysates sit on ice for 30 min. 5. Centrifuge cell lysates 20 min at 14,000 × g, 4◦ C, and collect the supernatants. 6. Quantify protein in supernatants using Protein Assay Dye Reagent Concentrate according to manufacturer’s instructions. 7. Add an equal volume of 2× Laemmli sample buffer to the lyates for a final concentration of 1× and heat for 10 min at 95◦ C. 8. Load the same amount of protein (typically 10 to 25 μg) onto each lane of a 10% to 12% SDS-PAGE gel. Also load prestained SDS-PAGE standards. Basic SDS-PAGE protocols are provided in Gallagher (1995).
9. Electrophorese and then transfer proteins onto PVDF membranes. Basic electrotransfer protocols are provided in Ursitti et al. (1995). The electrotransfer is typically performed by running 100 V for 1 hr, but the conditions need to be optimized for different protein sizes.
10. Block membranes with blocking solution (5% nonfat dry milk) for 1 hr at room temperature by gently shaking on a platform rotator. 11. Decant blocking solution. Incubate the membranes with anti-AR antibody (1:500 dilution in blocking solution) overnight at 4◦ C. 12. Wash the membranes with phosphate-buffered saline (PBS) containing 0.05% Tween 20 for 10 min. Repeat the washing three to five times. Incubate the membranes with anti-IgG horseradish peroxidase (1:1000 to 1:5000 dilution in PBS) for 1 hr at room temperature. The decreased intensities of AR bands in lysates obtained from PROTAC-treated cells, compared to those obtained from vehicle-treated cells, will confirm successful degradation of the PROTAC-induced target.
13. Visualize AR on Kodak BioMax XAR films using Amersham ECL Western Blotting Detection Reagents. 14. Incubate blot 1.5 hr with anti-β-actin antibody (1:1000 dilution in blocking solution), wash once for 2 min with shaking in blocking solution, then incubate 1 hr at room temperature with anti-IgG horseradish peroxidase (1:5000 dilution in PBS). Visualize β-actin on Kodak BioMax XAR films using Amersham ECL Western Blotting Detection Reagents. The intensities of β-actin bands are used as gel-loading controls. ALTERNATE PROTOCOL
Targeted Degradation of Proteins by PROTACs
VALIDATION OF PROTAC’S MODE OF ACTION USING CELL LINES To verify that specific PROTACs direct degradation of a targeted protein as designed, the use of two cell lines (786-O and 786-O/VHL) is recommended. Specifically, the requirement of the E3 ligase pVHL for PROTAC-induced degradation of target proteins can be tested using the renal carcinoma cell line 786-O and 786-O/VHL cell lines (Baba et al., 2003). 786-O cells do not produce the VHL protein and thus lack a functional VBC-Cul2 E3 ligase complex. In contrast, 786-O/VHL cells express wild-type VHL via
82 Volume 2
Current Protocols in Chemical Biology
stable transfection of VHL. Therefore, PROTACs should be functional in 786-O/VHL cells but not in 786-O cells.
Additional Materials (also see Basic Protocol 4) 786-O and 786-O/VHL cell lines (Baba et al., 2003) 1. Maintain 786-O and 786-O/VHL cells in cell culture medium. 2. Repeat procedure for immunoblot analyses as described in Basic Protocol 4 (steps 2 to 13).
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps.
Cell culture medium Phenol-red-free RPMI 1640 medium (Invitrogen, cat. no. 11835-030) supplemented with: 10% charcoal dextran–treated FBS (ctFBS; Hyclone, cat. no. SH30068.03; ctFBS signifies serum without steroids) 100 U/ml penicillin 100 μg/ml streptomycin Add antibiotics in the form of penicillin-streptomycin (P/S; Invitrogen, cat. no. 15140-122) Lysis buffer 25 mM Tris·Cl, pH 7.6 150 mM NaCl 1% (v/v) NP-40 1% (w/v) sodium deoxycholate 0.1% (w/v) sodium dodecyl sulfate (SDS) 1% (v/v) protease inhibitor cocktail (Sigma-Aldrich, cat. no. P8340) 1% (v/v) phosphatase inhibitor cocktail (Sigma-Aldrich, cat. no. P2850) Phosphate-buffered saline (PBS) 137 mM NaCl 2.7 mM KCl 4.3 mM Na2 HPO4 1.47 mM KH2 PO4 Adjust to the final pH of 7.4 COMMENTARY Background Information Genetic and molecular tools that modulate cellular protein levels have made a profound impact on our understanding of protein functions in cells. For example, classical gene knockout and siRNA-induced protein knockdown approaches have served as major strategies in elucidating protein functions (Natt, 2007; Li and Shen, 2009). However, the lack of tight spatial and temporal control are the major limitations of these approaches in dissecting functions of individual proteins in complex signaling pathways.
In recent years, increasing attention has been directed to small-molecule-based approaches, which provide tight spatial and temporal control. In particular, the small molecule–based PROTAC approach is designed to target proteins for degradation, offering a novel class of molecular probes that mimics the effect of siRNA. A PROTAC molecule is comprised of a small-molecule ligand specific for a protein of interest and an E3 ligase recognition motif. The first proof-of-concept example was a PROTAC targeting methionine
Targeted Degradation of Proteins by PROTACs
83 Current Protocols in Chemical Biology
Volume 2
Targeted Degradation of Proteins by PROTACs
aminopeptidase-2 (MetAP-2). The MetAP2-targeting PROTAC was designed using an anti-angiogenic natural product, fumagillin, a specific ligand for MetAP-2, and an IκBα-derived motif recognized by Skp1-Cullin-F box (SCFβ-TRCP ) E3 ubiquitin ligase complex (Sakamoto et al., 2001). The SCFβ-TRCP complexes are conserved from yeast to mammals and known to promote ubiquitination and subsequent degradation of IκBα, a negative regulator of NF-κB. The recruitment of IκBα to SCFβ-TRCP is mediated by a 10-amino-acid phosphopeptide sequence of DRHD32 S∗GLD36 S∗M (∗ denotes phosphorylation) within the phosphorylated IκBα. Thus, the MetAP-2-targeting PROTAC is designed to include fumagillin at one end and the 10-amino acid phosphopeptide at the other end (Sakamoto et al., 2001). Using the same IκBα-derived 10-aminoacid phosphopeptide sequence, additional PROTACs were developed to target a number of proteins such as estrogen receptor (ER) and androgen receptor (AR) (Sakamoto et al., 2003; Schneekloth et al., 2004). Similar to the structure of MatAP-2-targeting PROTAC, these PROTACs are composed of the IκBα phosphopeptide on one end and estradiol (E2) or dihydrotestosterone (DHT) on the other end. These PROTACs were shown to successfully promote degradation of target receptors. Following these initial proof-of-concept studies, the PROTACs have been further modified for improved cell permeability and stability. For example, the E3 ligase-binding motif was modified to include an additional polyarginine chain (mimicking the HIV Tat protein) (Wender et al., 2000; Kirschberg et al., 2003) to improve cell permeability (Rodriguez-Gonzalez et al., 2008). In addition, interactions between another E3 ligase (pVHL, von Hippel-Lindau tumor suppressor protein)–substrate (HIF-1α) pair have been exploited for the design of PROTACs. This approach was designed with the goal that the pVHL recognition motif would improve PROTACs’ bioavailability compared to the phosphopeptide derived from IκBα. It has been shown that, under normoxic conditions, pVHL recruits HIF-1α (hypoxia inducing factor-1α) for ubiquitination and subsequent degradation by the proteasome (Ivan et al., 2001). Mechanistic studies revealed that a short peptide sequence containing hydroxyproline within HIF-1α is responsible for the recognition of pVHL (Ivan et al., 2001). Therefore, hydroxyproline-containing
heptapeptide (MLAPOH YIP) or octapeptide (MLAPOH YIPM) fragments derived from HIF-1α were used as a pVHL recognition motif in the synthesis of PROTACs. These PROTACs were shown to be highly effective and cell-permeable and have been successfully applied to target FK506 binding protein (FKBP12), ER, and AR for degradation (Schneekloth et al., 2004; Tang et al., 2009; Zhang et al., 2004a,b) (Fig. 2). Subsequently, the pVHL recognition sequence was further simplified to a pentapeptide sequence, which resulted in improved cell permeability (Zhang et al., 2004b). Thus far, the pentapeptide-based PROTACs have been successfully developed to target the ER and aryl hydrocarbon receptor (AHR) for degradation (Zhang et al., 2004b; Lee et al., 2007; Puppala et al., 2008). Besides using these nonphosphorylated peptides in the design of PROTACs, there have been additional efforts to develop non-peptidic E3 ligase recognition motifs to further improve cell permeability and instability associated with the peptide residues. For example, Schneekloth et al. (2008) utilized nutlin as an E3 ligase recognition motif for the design of PROTACs. Nutlin, an imidazoline derivative, is a synthetic small-molecule ligand for MDM2, an E3 ubiquitin ligase that acts as a negative p53 regulator. To this end, they designed a PROTAC composed of a non-steroidal androgen receptor ligand (SARM) on one end and nutlin on the other end, connected by a PEG-based linker. This non-peptidic PROTAC was shown to promote degradation of AR (Schneekloth et al., 2008). It is important to note that both agonistic and antagonistic small-molecule ligands can be used for synthesizing PROTACs, since both can interact with a targeted protein and recruit target proteins to an E3 ubiquitin ligase complex for ubiquitination and subsequent degradation. Therefore, unlike any other antagonist development approach, the PROTAC technology provides a unique way to develop antagonists. While the PROTAC strategy confers the abovementioned advantages, there are several points to consider in the design of a PROTAC experiment. First, prior to the PROTAC design, a small-molecule ligand specific for a targeted protein needs to be identified. Given the difficulty in developing small-molecule ligands de novo, it is recommended to use small-molecule ligands that are already identified. Secondly, the introduction of functional groups in a small-molecule ligand must be readily achievable. It is strongly suggested not
84 Volume 2
Current Protocols in Chemical Biology
to spend too much time on functional group introduction. Thirdly, the derivatized smallmolecule ligands and PROTAC must maintain the binding affinity for targeted proteins. This can be tested through a direct binding study or appropriate biological assays. A number of recent reviews on the application of PROTAC approach are available (Crews, 2003; Sakamoto, 2005; Schneekloth and Crews, 2005). In summary, the PROTAC approach is a novel strategy to achieve controlled degradation of target proteins by exploiting the intracellular protein-degradation system. The PROTAC approach may be applied to target any protein, even those existing in a multisubunit complex, using any small molecule that exhibits high binding affinity for targeted proteins (Sakamoto et al., 2003). In addition to their utility as protein-knockdown tools with tighter temporal and spatial control, PROTACs may prove useful as a new type of therapeutics by removing disease-associated or disease-promoting proteins. As compared to conventional therapeutic agents, PROTACbased therapeutic approaches may offer advantages, especially when the effectiveness of antagonistic therapeutic agents is compromised due to development of drug resistance, which often occurs through mutations or posttranslational modifications. The PROTAC action is less likely to be affected by such changes and remain effective in eliminating targeted proteins.
Impact of linker length in the activity of PROTACs Given that PROTACs act as a bridging molecule and catalyze ubiquitin transfer from the E3 ligase complex to a targeted protein, the distance between two partner proteins, which is controlled (or determined) by a linker connecting two motifs within PROTACs, may be important for the optimum activity of PROTACs. Currently, no optimum distance between two partner proteins for ubiquitin transfer has been reported. The linker was rather randomly selected and used for the preparation of PROTACs. Therefore, it is recommended to prepare PROTACs with varying length of linkers and determine which yields the best results.
Critical Parameters
Time Considerations
Use of small-molecule ligands that induce degradation of target protein in the construction of PROTACs As described above, a major limitation of the PROTAC strategy is the requirement for a pre-existing ligand that can be used for the design of PROTAC. Unlike typical antagonists, which bind to a target protein and inhibit its functions, PROTACs destroy target proteins. Therefore, it is important to include appropriate controls in all biological assays. Controls include a PROTAC analog (“negative control”) in which hydroxyproline is replaced with norleucine in the pVHL E3 ligase recognition pentapeptide of the functional PROTAC (“positive control”). The lack of target protein degradation with the PROTAC analog would provide further evidence that target degradation occurs via the pVHL E3 ligase-ubiquitinationproteasome system.
Troubleshooting Table 1 describes some problems that may be encountered with PROTAC experiments, along with descriptions of possible causes and suggestions to overcome or avoid these problems.
Anticipated Results By following Basic Protocols 1 to 3, PROTACs that target proteins of interest for degradation can typically be readily synthesized. However, the experimental conditions for the protein degradation assay (e.g., PROTAC concentration and incubation time) need to be optimized for individual proteins (details are described in Basic Protocol 4).
Typically, the most time-consuming step is the preparation of an activated small-molecule ligand for a target protein. It may take 1 to 3 weeks to prepare the activated ligand and verify its binding affinity for a target protein. The coupling between an activated small-molecule ligand and an E3 ligase recognition motif often takes 1 hr to overnight at room temperature. DMSO evaporation and purification may take additional 2 days. If a small-molecule ligand already contains a molecular handle, such as free acid or primary amine or alcohol, for linker attachment, the activated ligand preparation step is greatly facilitated. Overall, once a small-molecule ligand with a handle is identified, PROTAC synthesis and assay can be achieved within one month.
Acknowledgements Funding sources: Kentucky Lung Cancer Research Program; Department of Defense.
Targeted Degradation of Proteins by PROTACs
85 Current Protocols in Chemical Biology
Volume 2
Table 1 Troubleshooting PROTACs Experiments
Problem
Possible cause(s)
Loss of Fmoc-deprotected Detection problem using products (Basic Protocols 1 staining solution and retention and 2) of high-polarity products on silica column
Solution After loading crude product on silica column, use two different solvent systems: first, hexane:ethyl acetate (1:1, v/v) to elute away byproducts and then more polar solvent such as CH2 Cl2 :methanol (95:5→80:20, v/v) to elute the product
Failed detection of a product (fully assembled PROTAC) during the final coupling step
Interference by reaction solvent After overnight coupling reaction, completely (DMSO or DMF) with detection evaporate DMSO (or DMF) under high vacuum (it may Low quantity of PROTAC take 1-2 days to remove DMSO). Redissolve in a mixture of CH2 Cl2 :methanol (4:1, v/v) and use TLC to locate the coupled product.
Low yield in PROTAC assembly step (Basic Protocol 3)
Decomposition of activated N-succinimidyl ester residue Lack of basic catalyst
Use freshly opened bottles of anhydrous DMSO or DMF. Add excess E3 ligase recognition motif (for example, HIF-1α pentapeptide) compared to activated protein ligands.
No degradation of target protein
Mutation or deficiency of E3 ligase
Before protein degradation assay with PROTACs, make sure that cells express all the components of E3 ligase complex by using a positive control (a known PROTAC)
No degradation of target protein
Poor solubility in aqueous conditions
Prepare PROTAC stock solution in DMSO. Dilute the DMSO stock solution in cell culture buffer in a tube and carefully observe if there is any precipitate or solution is transparent. If PROTAC has poor solubility, add multiarginine to the PROTAC (RodriguezGonzalez et al., 2008). Use a longer linker, which connects protein ligand and an E3 ligase recognition motif
Short linker length Nonspecific degradation of proteins
Non-specific binding of PROTAC
When a small-molecule ligand for a target protein is derivatized for coupling with an E3 ligase recognition motif, verify that the derivative still maintains the binding affinity of the parent molecule for the target protein.
Literature Cited Baba, M., Hirai, S., Yamada-Okabe, H., Hamada, K., Tabuchi, H., Kobayashi, K., Kondo, K., Yoshida, M., Yamashita, A., Kishida, T., Nakaigawa, N., Nagashima, Y., Kubota, Y., Yao, M., and Ohno, S. 2003. Loss of von HippelLindau protein causes cell density dependent deregulation of CyclinD1 expression through hypoxia-inducible factor. Oncogene 22:27282738. Bargagna-Mohan, P., Baek, S.H., Lee, H., Kim, K., and Mohan, R. 2005. Use of PROTACS as molecular probes of angiogenesis. Bioorg. Med. Chem. Lett. 15:2724-2727. Boysen, R.I. and Hearn, M.T.W. 2001. HPLC of peptides and proteins. Curr. Protoc. Protein Sci. 23:8.7.1-8.7.40. Targeted Degradation of Proteins by PROTACs
Carr, S.A. and Annan, R.S. 1996. Overview of peptide and protein analysis by mass spectrometry. Curr. Protoc. Protein Sci. 4:16.1.116.1.27.
Ciechanover, A. 2005. Proteolysis: From the lysosome to ubiquitin and the proteasome. Nat. Rev. Mol. Cell Biol. 6:79-87. Corson, T.W., Aberle, N., and Crews, C.M. 2008. Design and applications of bifunctional small molecules: Why two heads are better than one. ACS Chem. Biol. 3:677-692. Crews, C.M. 2003. Feeding the machine: Mechanisms of proteasome-catalyzed degradation of ubiquitinated proteins. Curr. Opin. Chem. Biol. 7:534-539. Crews, C.M. and Splittgerber, U. 1999. Chemical genetics: Exploring and controlling cellular processes with chemical probes. Trends Biochem. Sci. 24:317-320. Crews, C.M. and Mohan, R. 2000. Small-molecule inhibitors of the cell cycle. Curr. Opin. Chem. Biol. 4:47-53. Deshaies, R.J. 1999. SCF and Cullin/Ring H2based ubiquitin ligases. Annu. Rev. Cell. Dev. Biol. 15:435-467.
86 Volume 2
Current Protocols in Chemical Biology
Gallagher 1995. One-dimensional SDS gel electrophoresis of proteins. Curr. Protoc. Protein Sci. 3:10.1.1-10.1.34.
Sakamoto, K.M. 2005. Chimeric molecules to target proteins for ubiquitination and degradation. Methods Enzymol. 399:833-847.
Hershko, A. and Ciechanover, A. 1998. The ubiquitin system. Annu. Rev. Biochem. 67:425-479.
Sakamoto, K.M., Kim, K.B., Kumagai, A., Mercurio, F., Crews, C.M., and Deshaies, R.J. 2001. Protacs: Chimeric molecules that target proteins to the Skp1-Cullin-F box complex for ubiquitination and degradation. Proc. Natl. Acad. Sci. U.S.A. 98:8554-8559.
Ivan, M., Kondo, K., Yang, H., Kim, W., Valiando, J., Ohh, M., Salic, A., Asara, J.M., Lane, W.S., and Kaelin, W.G. Jr. 2001. HIFalpha targeted for VHL-mediated destruction by proline hydroxylation: Implications for O2 sensing. Science 292:464-468. Kim, K.B., Myung, J., Sin, N., and Crews, C.M. 1999. Proteasome inhibition by the natural products epoxomicin and dihydroeponemycin: Insights into specificity and potency. Bioorg. Med. Chem. Lett. 9:3335-3340. Kirschberg, T.A., VanDeusen, C.L., Rothbard, J.B., Yang, M., and Wender, P.A. 2003. Argininebased molecular transporters: The synthesis and chemical evaluation of releasable taxoltransporter conjugates. Org. Lett. 5:34593462. Lee, H., Puppala, D., Choi, E.Y., Swanson, H., and Kim, K.B. 2007. Targeted degradation of the aryl hydrocarbon receptor by the PROTAC approach: A useful chemical genetic tool. Chembiochem 8:2058-2062. Li, L. and Shen, Y. 2009. Overcoming obstacles to develop effective and safe siRNA therapeutics. Expert Opin. Biol. Ther. 9:609-619. Lin, H. and Cornish, V.W. 2001. In vivo proteinprotein interaction assays: Beyond proteins. Angew. Chem. Int. Ed. Engl. 40:871-875. Meng, L., Kwok, B.H., Sin, N., and Crews, C.M. 1999a. Eponemycin exerts its antitumor effect through the inhibition of proteasome function. Cancer Res. 59:2798-2801. Meng, L., Mohan, R., Kwok, B.H., Elofsson, M., Sin, N., and Crews, C.M. 1999b. Epoxomicin, a potent and selective proteasome inhibitor, exhibits in vivo antiinflammatory activity. Proc. Natl. Acad. Sci. U.S.A. 96:10403-10408. Mitchison, T.J. 1994. Towards a pharmacological genetics. Chem. Biol. 1:3-6. Natt, F. 2007. siRNAs in drug discovery: Target validation and beyond. Curr. Opin. Mol. Ther. 9:242-247. Orlowski, R.Z. and Kuhn, D.J. 2008. Proteasome inhibitors in cancer therapy: Lessons from the first decade. Clin. Cancer Res. 14:1649-1657. Puppala, D., Lee, H., Kim, K.B., and Swanson, H.I. 2008. Development of an aryl hydrocarbon receptor antagonist using the proteolysistargeting chimeric molecules approach: A potential tool for chemoprevention. Mol. Pharmacol. 73:1064-1071. Rodriguez-Gonzalez, A., Cyrus, K., Salcius, M., Kim, K., Crews, C.M., Deshaies, R.J., and Sakamoto, K.M. 2008. Targeting steroid hormone receptors for ubiquitination and degradation in breast and prostate cancer. Oncogene 27:7201-7211.
Sakamoto, K.M., Kim, K.B., Verma, R., Ransick, A., Stein, B., Crews, C.M., and Deshaies, R.J. 2003. Development of Protacs to target cancerpromoting proteins for ubiquitination and degradation. Mol. Cell. Proteomics 2:1350-1358. Schneekloth, J.S. Jr. and Crews, C.M. 2005. Chemical approaches to controlling intracellular protein degradation. Chembiochem 6:40-46. Schneekloth, J.S. Jr., Fonseca, F.N., Koldobskiy, M., Mandal, A., Deshaies, R., Sakamoto, K., and Crews, C.M. 2004. Chemical genetic control of protein levels: Selective in vivo targeted degradation. J. Am. Chem. Soc. 126:37483754. Schneekloth, A.R., Pucheault, M., Tae, H.S., and Crews, C.M. 2008. Targeted intracellular protein degradation induced by a small molecule: En route to chemical proteomics. Bioorg. Med. Chem. Lett. 18:5904-5908. Schreiber, S.L. 1998. Chemical genetics resulting from a passion for synthetic organic chemistry. Bioorg. Med. Chem. 6:1127-1152. Tang, Y.Q., Han, B.M., Yao, X.Q., Hong, Y., Wang, Y., Zhao, F.J., Yu, S.Q., Sun, X.W., and Xia, S.J. 2009. Chimeric molecules facilitate the degradation of androgen receptors and repress the growth of LNCaP cells. Asian J. Androl. 11:119126. Ursitti, J.A., Mozdzanowski, J., and Speicher, D.W. 1995. Electroblotting from polyacrylamide gels. Curr. Protoc. Protein Sci. 00:10.7.1-10.7.14. Wender, P.A., Mitchell, D.J., Pattabiraman, K., Pelkey, E.T., Steinman, L., and Rothbard, J.B. 2000. The design, synthesis, and evaluation of molecules that enable or enhance cellular uptake: Peptoid molecular transporters. Proc. Natl. Acad. Sci. U.S.A. 97:13003-13008. Yeh, J.R., Mohan, R., and Crews, C.M. 2000. The antiangiogenic agent TNP-470 requires p53 and p21CIP/WAF for endothelial cell growth arrest. Proc. Natl. Acad. Sci. U.S.A. 97:12782-12787. Zhang, D., Baek, S.H., Ho, A., and Kim, K. 2004a. Degradation of target protein in living cells by small-molecule proteolysis inducer. Bioorg. Med. Chem. Lett. 14:645-648. Zhang, D., Baek, S.H., Ho, A., Lee, H., Jeong, Y.S., and Kim, K. 2004b. Targeted degradation of proteins by small molecules: A novel tool for functional proteomics. Comb. Chem. High T. Scr. 7:691-699. Targeted Degradation of Proteins by PROTACs
87 Current Protocols in Chemical Biology
Volume 2
Directed Evolution of DNA Polymerases: Construction and Screening of DNA Polymerase Mutant Libraries Christian Gloeckner,1 Ramon Kranaster,1 and Andreas Marx1 1
Department of Chemistry, University of Konstanz, Konstanz, Germany
ABSTRACT The protocols in this article describe the construction of a mutant DNA polymerase library using error-prone PCR (epPCR) as a method for gene randomization, followed by screening of the library using two different approaches. The examples described use an N-terminally truncated form of the thermostable DNA polymerase I of Thermus aquaticus (Taq DNA polymerase), namely Klentaq (KTQ), and protocols are included for the identiÞcation of variants with (1) increased DNA lesion-bypass ability and (2) enhanced selectivity for DNA match/mismatch recognition. The screening assays are based on double-stranded DNA detection (using SYBR Green I) which can be carried out using standard laboratory equipment. The described assays are designed for use in a 384-well plate format to increase screening throughput and reduce material costs. For improved accuracy and ease of liquid handling, the use of an automated liquid handling C 2010 by John Wiley & device is recommended. Curr. Protoc. Chem. Biol. 2:89-109 Sons, Inc. Keywords: DNA polymerase r directed evolution r screening r PCR r primer extension
INTRODUCTION DNA polymerases are key enzymes and workhorses in modern biotechnology applications such as genomic sequencing and diagnostics, and Þnd wide application in techniques such as RT-PCR and Q-PCR, cloning, and modiÞcation of DNA using artiÞcial substrates. Since new technologies and applications arise rapidly and often, there is an increasing need for improved DNA polymerase variants with enhanced efÞciency, accuracy, and processivity, or improved tolerance for artiÞcial substrates and resistance to inhibitory substances—e.g., for use with blood or soil samples in forensic applications. Methods of directed evolution have been shown to be promising for the engineering of nucleic acid polymerases with altered properties (d’Abbadie et al., 2007; Gloeckner et al., 2007; Loh et al., 2007; Kranaster and Marx, 2009; Loakes et al., 2009). Method design is very diverse and ranges from selection methods such as phage display and emulsion techniques to screening assays such as those employed here. Although selection methods deal with library sizes of up to 109 to 1010 members, which are not accessible by screening methods, they are not applicable to a large variety of scientiÞc questions. In contrast, screening assays are extremely versatile and robust, and assay design is more ßexible than in selection schemes, although at the cost of throughput. High-throughput screens (HTS) can be undertaken by applying automated liquid handling devices, and library sizes of 105 members can be examined in only a few days. Here, an approach to identify improved DNA polymerase variants is described which involves automated screening of DNA polymerase mutant libraries. Basic Protocol 1 describes the construction of a randomized DNA polymerase library using epPCR to introduce mutations. These mutations occur randomly by the inaccurate nature of the
Current Protocols in Chemical Biology 2: 89-109, May 2010 Published online May 2010 in Wiley Interscience (www.interscience.wiley.com). DOI: 10.1002/9780470559277.ch090183 C 2010 John Wiley & Sons, Inc. Copyright
Directed Evolution of DNA Polymerases
89 Volume 2
applied Taq DNA polymerase during ampliÞcation of the parental gene, although variations in Mn2+ concentrations yield different mutation rates, and the use of unbalanced dNTP mixes can shift the mutation spectrum from transition (A:T→ G:C) to transversion (A:T → C:G or T:A) mutations. A primary library is established in 384-well deep-well plates consisting of randomly mutated DNA polymerase variants that either retain or lose their general enzymatic activity and thermostability through mutations. Evaluation of enzymatic activity as well as thermostability is addressed in Basic Protocol 3. Basic Protocol 2 describes the recombinant expression and puriÞcation of a thermostable DNA polymerase library in a 96-well format. Deep-well plates with a volume of 2.2 ml per well are used for expression in E. coli BL21 (DE3) Gold cells. PuriÞcation is simple, since host proteins are not thermostable, so a heat-inactivation step followed by centrifugation leads to fairly pure lysates that are directly applied for screening purposes. Basic Protocol 3 distinguishes between PCR-active and -inactive variants in the primary library. PCR ampliÞcation using a short artiÞcial DNA template (90-mer) is performed using puriÞed lysates from Basic Protocol 2, and double-stranded DNA (dsDNA) product formation is detected by SYBR Green I ßuorescence read-out. PCR-active variants from the primary 384-well plate library are then pooled into new 384-well plates, yielding the Þnal library. The Þnal library is then employed for directed evolution purposes in Basic Protocol 4 and Basic Protocol 5. Basic Protocol 4 describes a setup to screen for enhanced match/mismatch discrimination by comparing two real-time PCR (rtPCR) reactions per screened DNA polymerase variant. One reaction is performed using a template (F90A) that pairs canonically with a primer ending with dT at the 3 end. A second rtPCR reaction is run with the same primer, but employing a template with dG (F90G) opposite the primer’s 3 end, thus forming a mismatch. Monitoring the progress of both reactions and subsequent analysis of the product formation gives information about the discrimination properties of the examined DNA polymerase variant, as well as its overall enzymatic activity. Basic Protocol 5 identiÞes DNA polymerase variants with increased DNA lesion-bypass ability. Screening is based upon dsDNA formation in a primer-extension experiment in the presence of only three dNTPs, although the template calls for all four dNTPs during extension of the primer to full length. Thus, a DNA polymerase variant extending the primer has to misincorporate a nucleotide and extend the resulting mismatch in the face of steric constraints that mimic common DNA lesions. Both of the abovementioned screening approaches have been used to yield DNA polymerase variants with the desired properties, and details can be found in Strerath et al. (2007) and Gloeckner et al. (2007). These examples show that the described approach of randomization using epPCR and subsequent screening is able to generate DNA polymerases with new properties.
STRATEGIC PLANNING
Directed Evolution of DNA Polymerases
In the procedure reported below, a strategy is described to create and screen DNA polymerase mutant libraries for several new enzymatic functions and properties. We demonstrate and describe the construction of a randomized DNA polymerase library using the N-terminally truncated form of Thermus aquaticus DNA polymerase, namely Klentaq (KTQ; Barnes, 1992), with subsequent screening efforts directed towards two different properties—increased DNA lesion bypass and enhanced single-nucleotide discrimination. All time-consuming procedures should be performed by automated liquid-handling
90 Volume 2
Current Protocols in Chemical Biology
Basic Protocol 1 epPCR in presence of Mn2⫹ introduction of mutations
final library 384-well plate sections
A B C D
cloning in expression plasmid transformation of E. coli
384-well plate sections
A B C D
Basic Protocol 2 96-well plates protein expression in 96-deep well plates, cell lysis, and purification
A
B
C
D
A
B
C
D
protein expression in 96-deep well plates, cell lysis, and purification
Basic Protocol 4
Relative fluorescence
automated or manual transfer of single colonies to primary library in 384-well plates
pooling of PCR-active variants to yield final library
96-well plates
primary library
match mismatch
PCR cycles
Basic Protocol 3
Basic Protocol 5 Relative fluorescence
screening for PCR activity
screening for enhanced match/mismatch discrimination
X
X
screening for improved DNA lesion bypass
X
SYBR SYBR
PCR cycles
Figure 1 Flow diagram showing the process for construction of a DNA polymerase mutant library followed by screening for polymerase chain reaction (PCR) activity. Afterwards, the PCR-active mutants may be screened for match/mismatch sensitivity via real-time PCR, or for lesion-bypass activities.
equipment, thus enabling high throughput if required. Figure 1 illustrates the general strategy employed. First, a randomized DNA polymerase library is created by epPCR, yielding a primary library that will include active and inactive variants. In a Þrst screening approach, this library is screened for PCR activity in order to create a Þnal library that consists only of PCR-active DNA polymerase variants. This library can then be employed to address different evolutionary goals.
ERROR-PRONE PCR LIBRARY CONSTRUCTION In this protocol, error-prone PCR (epPCR) is used to generate libraries of KTQ. Errorprone PCR has proven to be a powerful tool to quickly introduce mutations and randomize genes during the ampliÞcation process by enhancing the intrinsic error rate of the employed Taq DNA polymerase (Cadwell and Joyce, 1992). The epPCR is carried out in presence of Mn2+ ions to increase the error rate of Taq polymerase. Mn2+ concentrations have to be optimized to yield the desired mutation rate
BASIC PROTOCOL 1
Directed Evolution of DNA Polymerases
91 Current Protocols in Chemical Biology
Volume 2
and vary typically between 20 and 200 μM. The mutation rate depends on the length of the gene to be randomized and its base-pair composition. It is recommended to test several Mn2+ concentrations and to analyze the mutation rate by sequencing or testing the activity of the randomized protein. In the example chosen, a Þnal Mn2+ concentration of 50 μM is used.
Materials 5 U/μl Taq DNA polymerase (Roche) 10× Taq buffer (see recipe) 25 mM MgCl2 1 mM MnCl2 10 μM forward cloning primer KTQ fw (5 -ATG GTA CGT CTC AGC GCG CCC TGG AGG AGG CCC CCT-3 ) 10 μM reverse cloning primer KTQ rev (5 -ATG GTA CGT CTC ATA TCA CTC CTT GGC GGA GAG CCA GTC-3 ) 400 pM template (pASK-IBA37plus-KTQ-wt) (pASK-IBA37plus from IBA GmbH) 1× loading buffer for agarose gel electrophoresis (Voytas, 2000) 0.8% agarose gel (Voytas, 2000) 1-kb DNA ladder (New England Biolabs) 2 mM dNTP mix (2 mM dATP, dCTP, dGTP, dTTP) Gel elution kit (Qiagen) PCR reaction clean-up kit (Qiagen) Restriction endonucleases (New England Biolabs): 20 U/μl DpnI 10 U/μl BsmBI 10 U/μl BsaI 10× reaction buffer in which both DpnI and BsmBI are active (e.g., Buffer 4; New England Biolabs) 1 U/μl T4 DNA ligase and 10× ligase buffer pASK-IBA37-plus (BsaI-digested and dephosphorylated; see recipe) SOC medium (see recipe) Electrocompetent E. coli BL21 (DE3) Gold (Stratagene) LB liquid medium and plates containing 100 μg/ml carbenicillin (see recipe) Plasmid miniprep kit (e.g., Qiagen, Invitrogen) Sequencing primers: Forward: 5 -GAG TTA TTT TAC CAC TCC CT- 3 Reverse: 5 -CGC AGT AGC GGT AAA CG-3 ) 60% (v/v) glycerol in LB medium Thermal cycler Nanodrop ND-1000 UV/Vis spectrophotometer (http://www.nanodrop.com/) Cuvettes for electroporation, 1 mm electrode distance (BioRad) Electroporator Gene Pulser Xcell (BioRad) Incubation shaker, Titramax 1000 (Heidolph) Autoclaved wooden toothpicks 384-well deep-well plates (ABgene) Gas-permeable adhesive seals (ABgene) Self-adhesive aluminum/paper seals (ABgene) Adhesive tape (Tesa or Scotch) Directed Evolution of DNA Polymerases
Additional reagents and equipment for agarose gel electrophoresis (Voytas, 2000) and replica plating (Elbing and Brent, 2002)
92 Volume 2
Current Protocols in Chemical Biology
Prepare epPCR 1. Set up an error-prone PCR reaction as follows (100 μl total): 1 μl 5 U/μl Taq DNA polymerase (0.05 U/μl Þnal) 10 μl 10× Taq buffer (1× Þnal) 6 μl 25 mM MgCl2 (1.5 mM Þnal) 5 μl 1 mM MnCl2 (50 μM Þnal) 5 μl 10 μM primer KTQ fw (500 nM Þnal) 5 μl 10 μM primer KTQ rev (500 nM Þnal) 5 μl 400 pM template pASK-IBA37plus-KTQ-wt (20 pM Þnal) 10 μl 2 mM dNTP mix (200 μM each dNTP Þnal) 53 μl H2 O. 2. Run the following PCR thermal cycling program: 1 cycle: 14 cycles:
1 cycle:
2 min 1 min 1 min 2 min indeÞnite
94◦ C 94◦ C 65◦ C 72◦ C 4◦ C
(initial denaturation) (denaturation) (annealing) (extension) (hold).
3. Add 0.5 to 1 μl of the PCR product to 10 μl of 1× loading buffer and load onto a 0.8% agarose gel (prepared in 1× TAE buffer) along with a 1-kb DNA ladder to verify that the PCR was successful. Voytas (2000) provides detailed protocols for agarose gel electrophoresis. The PCR product should have a size of 1655 bp.
4. Purify the PCR product using a commercially available PCR reaction clean-up kit (e.g., Qiagen) following the manufacturer’s instructions. 5. Determine the DNA concentration by applying 1.5 μl of the eluate to a NanoDrop spectrophotometer (1 OD260 dsDNA = 50 μg/ml), or load 0.5, 1, and 2 μl onto a 0.8% agarose gel (prepared 1× TAE buffer; Voytas, 2000) and run, in parallel, four different amounts of a standard (5, 10, 20, 40 μl of 1-kb DNA ladder) and make an estimate of the intensity of the lanes to the samples by eye. Refer to the manufacturer’s instructions for the DNA ladder to check for the amount of DNA in each band.
Clone library into expression vector 6. Set up a restriction digest using BsmBI (type IIS restriction endonuclease) according to the manufacturer’s instructions and digest the puriÞed PCR product. To reduce the possibility of wild-type background in the library, digest the templating plasmid using the methylation-sensitive restriction endonuclease DpnI prior to BsmBI digestion: 1 μg randomized PCR product (see steps 4 and 5) 0.5 μl 20 U/μl DpnI 5 μl 10× reaction buffer H2 O to bring total reaction volume to 50 μl. Incubate 1 hr at 37◦ C, then add 0.5 μl of 10 U/μl BsmBI and increase temperature to 55◦ C for 3 hr. Heat-inactivate restriction enzymes by shifting the temperature to 80◦ C for 20 min. Be sure to choose a buffer in which both enzymes are active (e.g., NEB Buffer 4).
Directed Evolution of DNA Polymerases
93 Current Protocols in Chemical Biology
Volume 2
7. Prepare a 0.8% agarose gel in 1× TAE buffer and run the entire digested sample through the gel for puriÞcation. Use a commercially available gel elution kit (e.g., Qiagen) to elute the DNA. CAUTION: Minimize the exposure to UV light while cutting the agarose gel since it leads to additional mutations and DNA damage. Always wear gloves, long-sleeved lab coat, and eye protection when working with UV light.
8. Determine the concentration of the eluate as described in step 5. 9. Set up the following 20-μl ligation reaction and incubate at 22◦ C for 1 hr, followed by a 10-min heat inactivation at 65◦ C.
2 μl 1 U/μl T4 DNA ligase 5 μl 10× ligase buffer ∼100 ng (47.4 fmol) pASK-IBA37plus (BsaI-digested and dephosphorylated; see Reagents and Solutions) ∼500 ng (474 fmol) PCR product (BsmBI-digested; from step 7, concentration determined in step 8) H2 O to bring total reaction volume to 20 μl. 10. Analyze 5 μl of the ligation reaction on a 0.8% agarose gel (Voytas, 2000) and run ∼150 ng of the digested plasmid and the digested PCR product in parallel. In addition to the unligated plasmid backbone and PCR product, other bands should also be visible. If no additional bands are visible it is very likely that the ligation was not successful. In this case make sure to use high-quality agarose for gel puriÞcation, and to verify the amount of DNA used in the ligation reaction.
11. Purify the remaining solution by a commercially available reaction clean-up kit according to the manufacturer’s protocol.
Transform electrocompetent E. coli 12. Prechill cuvettes for electroporation (1 mm electrode distance) on ice and warm SOC medium to 37◦ C. 13. Thaw 100 μl of electrocompetent E. coli BL21 (DE3) Gold on ice. When thawed, add 1 μl of the puriÞed ligation reaction (from step 11) and incubate for 1 min on ice. 14. Transfer the cells into the cuvette and pulse with 1.8 kV. Then add 1 ml of prewarmed SOC medium, transfer the cells to a culture vessel, and incubate for 1 hr at 37◦ C with shaking. 15. Spread 100 μl of transformed cells onto LB plates containing 100 μg/ml carbenicillin and incubate overnight at 37◦ C. Make ten plates per transformation. Electroporation yields the most transformants per μg DNA and is thus the most efÞcient way to establish a large library of variants. Nevertheless, if only a small library is desired, chemically competent cells and heat-transformation protocols may be used.
Analyze diversity 16. Pick ten colonies from the plates and inoculate individually into 4-ml LB medium cultures containing 100 μg/ml carbenicillin. Grow overnight at 37◦ C with shaking in an incubator. Create a replicate LB plate with the picked colonies (see Elbing and Brent, 2002).
Directed Evolution of DNA Polymerases
17. Isolate the plasmid using a commercially available miniprep kit according to the manufacturer’s instructions and send an aliquot of each sample for DNA sequencing along with the forward and reverse sequencing primers listed in the materials list above.
94 Volume 2
Current Protocols in Chemical Biology
18. Align the results from the different sequencing reactions, and include the KTQ wt sequence using appropriate software. By comparing the DNA sequences of the picked clones to the KTQ wt sequence, you are able to determine the number of mutations introduced during epPCR into the KTQ ORF (open reading frame). KTQ wt is an N-terminal shortened form of the Taq DNA polymerase, omitting the 5 -3 exonuclease activity of the parent enzyme and harboring the polymerase activity only. The nucleotide sequence for full-length Taq DNA polymerase can be found using accession number J04639.1 in the nucleotide search at NCBI (http://www.ncbi.nlm.nih.gov/). KTQ wt consists of nucleotides 877 to 2496, yielding an enzyme of 540 amino acids. Free software to view and export the sequencing data is available on the Internet (e.g., Chromas Lite, http://www.technelysium.com.au/chromas lite.html), and multiple sequence alignments can be done applying CLUSTALW, which can be found for free at SDSC Biology Workbench (http://workbench.sdsc.edu/). Different software packages are commercially available, such as Chromas Pro (Technelysium), Vector NTI (Invitrogen), or Geneious (Biomatters).
Establish a 384-well plate library 19. Dispense 150 μl LB medium containing 100 μg/ml carbenicillin per well in a 384-well deep-well plate. 20. Pick single colonies from the plates generated in step 15, using autoclaved wooden toothpicks, and inoculate each well with one colony. Let the toothpicks sit in the wells until Þnished picking a whole plate. 21. Divide the 384-well plate in four 96-well segments (rows 1 to 6, 7 to 12, 13 to 18, and 19 to 24) and inoculate four wells in the top left corner of each segment with BL21 (DE3) Gold expressing KTQ wt as a control (A1, A2, B1, B2, etc.). pASK-IBA37plus-KTQ wt is obtained by cloning the KTQ wt gene into pASK-IBA37plus using BsaI restriction sites. Please refer to Sauter and Marx (2006) for details. BL21 (DE3) Gold are transformed as described in steps 12 to 15 above, to yield control colonies expressing the wt enzyme. Expression of the library will be performed in a 96-well format. Thus, every 96-well plate will contain four wt-expressing colonies and 92 variants. Two wells with wt expressions will later serve as positive controls and two as negative controls.
22. Apply a gas-permeable self-adhesive membrane on the 384-well deep-well plate and shake the plates overnight at 37◦ C. 23. Remove the membrane and add 100 μl of sterile 60% (v/v) glycerol in LB medium to each well. Seal the plates using aluminum self-adhesive foil and secure the foil using Scotch tape. Give the plates a brief shake in the incubator (5 min) to allow for mixing of the bacterial culture and the glycerol. 24. Freeze the library at −80◦ C until expression. The library is stable for several years if kept at −80◦ C.
EXPRESSION OF DNA POLYMERASE VARIANTS AND LYSATE PREPARATION
BASIC PROTOCOL 2
This protocol describes the expression of the KTQ DNA polymerase variants in deep-well plates and subsequent preparation of lysates for screening purposes. E.coli cells harboring library plasmids are inoculated from 384-well-plates (see Basic Protocol 1) into four separate 96-well deep-well plates and grown to exponential phase. After induction of recombinant protein expression using anhydrotetracycline (AHT),
Directed Evolution of DNA Polymerases
95 Current Protocols in Chemical Biology
Volume 2
cells are harvested by centrifugation. Subsequent cell lysis using detergent and lysozyme is followed by a heat-inactivation step, denaturing E.coli host proteins and leaving active, thermostable DNA polymerase variants in the supernatant. The lysates are then ready for screening and can be used for several days if kept at 4◦ C.
Materials 384-well plate containing E coli glycerol cultures (Basic Protocol 1) LB liquid medium containing 100 μg/ml carbenicillin (see recipe) 4 mg/liter anhydrotetracycline (AHT; IBA GmbH) in LB medium 1× KTQ lysis buffer (with lysozyme; see recipe) Automated liquid handling device (Hamilton Microlab Star) 96-well deep-well plates (2.2 ml well volume; VWR) Thermostatic incubators and shakers Gas-permeable adhesive seal (ABgene) Self-adhesive aluminum/paper seal (ABgene) Spectrophotometer Refrigerated ßoor-model centrifuge with multiwell-plate carrier 75◦ C water bath 1. Program the liquid handling device such that 5 μl of each well from a 384-well-plate is distributed to a certain position on one of four 96- well deep-well plates. Alternatively, this step may be performed manually using a multichannel pipet if no automated liquid handling device is accessible. Make certain that each clone position from a 96-well plate can be correlated to the 384-well glycerol stock plate.
2. Defrost one 384-well plate containing the respective E.coli glycerol cultures in a thermostatic shaker at 20◦ C. Incubate the glycerol cultures only for as short a time as possible, until all cultures are defrosted, which can take up to 2 hr. During defrosting, step 3 can be accomplished.
3. Fill each well of four 96-well deep-well plates with 0.945 ml LB medium (including respective antibiotic, in this case, carbenicillin at 100 μg/ml Þnal concentration) using a reagent dispenser or a multichannel pipettor at room temperature. 4. Place four 96-well deep-well plates in the liquid handling device. Aspirate 5 μl of glycerol stock for each clone from the 384-well-plate and dispense into the designated position of the 96-well deep-well plates. Be sure to adjust aspiration and dispensing properties of the liquid handling device to compensate for the high viscosity of the glycerol stocks and check for positioning along the z-axis to assure submersing of the tip in the culture solution.
5. Freeze 384-well plates at −80◦ C. Cover the four 96-well deep-well plates with a gas-permeable adhesive foil and let the bacteria grow for several hours at 37◦ C in a thermostatic shaker. Monitor the growth by measuring the absorbance at 600 nm. 6. Induce expression of the mutant enzymes, once the absorbance at 600 nm reaches 0.8 to 1.0, by addition of 50 μl 4 mg/liter AHT in LB medium (Þnal concentration of AHT, 200 μg/liter). Incubate for an additional 3 to 4 hr at 37◦ C in a thermostatic shaker.
Directed Evolution of DNA Polymerases
7. Subsequently, pellet the cells by centrifuging 20 min at 3000 × g, 4◦ C, using a ßoor centrifuge with an adaptor for multiwell plates. Remove the seal and discard the supernatant completely by turning the plate upside down (the pellets will stick to the bottoms of the wells). Cover the multi-well plates with sealing Þlm until further use.
96 Volume 2
Current Protocols in Chemical Biology
60 kDa 50 kDa
1
2
3
4
5
6
7
8
Figure 2 Representative SDS-PAGE showing bacterial lysates with dominant expression band of KTQ enzyme. Lane 1 = protein standard, Lane 2 = KTQ wt (positive control), Lane 3 = empty plasmid (negative control), Lanes 4 to 8 = randomly chosen KTQ variants.
Pellets can be stored at −20◦ C for at least several days without signiÞcant loss of active enzyme yields in the cell lysates, but once frozen, pellets are much more difÞcult to resuspend (see next step).
8. Add 800 μl 1× KTQ lysis buffer (including 0.1 mg/ml lysozyme as described in Reagents and Solutions) to each well and seal plates carefully with a self-adhesive aluminum/paper foil. Vortex plates. Observe the bottom of the plate to make sure that every pellet is resuspended in lysis buffer. 9. Incubate the plates in a water bath at 37◦ C for 20 min and then transfer to a second water bath at 75◦ C and incubate for 40 min to denature E.coli host proteins. Make sure all wells are sealed properly with the self-adhesive aluminum foil to avoid evaporation of the buffer. Adhesive tape can be used to support the self-adhesive foil.
10. Centrifuge all plates 45 min at 4000 × g, 4◦ C. The airtight-sealed lysates can be safely stored at 4◦ C up to 2 weeks without signiÞcant loss of PCR activity. Expression level and quality may be checked by SDS-PAGE (Fig. 2). A speciÞc protein band should be present in the positive control (wt) and absent in the negative control (empty plasmid).
DISTINGUISHING BETWEEN PCR-ACTIVE AND -INACTIVE VARIANTS BY SCREENING DNA POLYMERASE MUTANT LIBRARIES IN A PCR-BASED ASSAY This protocol describes screening for PCR activity to distinguish between active and inactive variants. IdentiÞed PCR-active variants from the primary library are then pooled to yield the Þnal DNA polymerase library that is employed for further screening. This step is not essential if the library size is small (fewer than ∼1000 to 2000 clones) or PCR activity is not required.
BASIC PROTOCOL 3
Directed Evolution of DNA Polymerases
97 Current Protocols in Chemical Biology
Volume 2
Real-time PCR is performed using the bacterial lysates in the presence of SYBR Green I, enabling quantiÞcation of synthesized double-stranded DNA during each cycling step (Wilhelm and Pingout, 2003). Subsequent melting-temperature measurements are performed to conÞrm the presence of the ampliÞcation product. For this approach, a standard real-time PCR cycler for multiwell plates is sufÞcient. Alternatively, if real-time cycler equipment is unavailable, the PCR can be performed in an appropriate cycler and endpoint ßuorescence values determined with a standard multiwell plate reader.
Materials 10× KTQ reaction buffer (see recipe) 10 mM dNTP mix (10 mM each dATP, dCTP, dGTP, dTTP) 0.1 μM DNA template F90A: 5 -d(CCG TCA GCT GTG CCG TCG CGC AGC ACG CGC CGC CGT GGA CAG AGG ACT GCA GAA AAT CAA CCT A TC CTC CTT CAG GAC CAA CGT ACA GAG)-3 (custom synthesis) 10,000× SYBR Green I (Invitrogen, cat. no. S-7563) 100 μM forward primer F20+ : 5 -d(CGT TGG TCC TGA AGG AGG AT)-3 (custom synthesis) 100 μM reverse primer F20− : 5 -d(CGC GCA GCA CGC GCC GCC GT)-3 (custom synthesis) Cell lysates in 96-well plates (Basic Protocol 2) 96-well PCR plate (VWR) Automated liquid handling device (Hamilton Microlab Star) Reagent dispenser (Multidrop, Thermo ScientiÞc) Optically clear heat-sealing Þlm (ABgene) Refrigerated ßoor-model centrifuge with multiwell-plate carrier Thermo-sealer or electric iron Real-time PCR thermal cycler 384-well deep-well plates (ABgene) Autoclaved wooden toothpicks 1. Prepare PCR master mix for PCR activity screening (see Table 1). CAUTION: The master mix contains SYBR Green I; be aware of its possible toxicity!
2. Fill each well of a 96-well PCR plate with 10 μl of PCR master mix using a reagent dispenser or a suitable multichannel pipet. Table 1 PCR Master Mix for PCR Activity Screening
Solution
Stock concentration
Final concentration
Milli-Q water
Pure
—
968.16 μl
10× KTQ reaction buffer
10×
1×
120 μl
dATP, dGTP, dCTP, TTP mix
10 mM each dNTP
250 μM
60 μl
DNA template F90A
0.1 μM
60 pM
100× dilution in H2 Oa 0.6×
SYBR Green I
Volume
1.44 μl 14.4 μl
+
100 μM
0.75 μM
18 μl
−
100 μM
0.75 μM
18 μl
Forward primer (F20 ) Reverse primer (F20 ) b
Total volume
1.2 ml
a Prepared from 10,000× stock, as purchased from Invitrogen. b This amount is calculated for 120 reactions, which is enough to Þll a 96-well PCR plate. The calculated volumes
Directed Evolution of DNA Polymerases
can be easily extended to less or more required reaction just by dividing the volumes by 120 and multiplying by the desired number of reactions.
98 Volume 2
Current Protocols in Chemical Biology
3. Add 10 μl of each cell lysate to the corresponding wells containing PCR master mix using a liquid handling device or a suitable multichannel pipet. Cool the PCR plate to 4◦ C during this and the following steps. Make certain that each clone position on the PCR plate can be correlated to a position on the lysate plate.
4. Centrifuge the PCR plate at 1000 × g for a few seconds to ensure that all liquids are at the bottoms of the wells. Seal plate with an optically clear heat-sealing Þlm using a thermo-sealer or an electric iron. Alternatively, a self-adhesive PCR sealing Þlm can be used.
5. Shake or vortex the sealed PCR plate for a few seconds. This step guarantees the homogeneity of the PCR mix.
6. Centrifuge the PCR plate again at 1000 × g for a few seconds to ensure that all liquids are at the bottom of the wells. The prepared PCR plate is now ready for real-time PCR and may be stored for a few hours at 4◦ C before performing the PCR.
7. Set up the real-time PCR thermal cycler using the following parameters: 1 cycle: 30 cycles:
1 cycle:
3 min 30 sec 35 sec 40 sec ramp
95◦ C 95◦ C 55◦ C 72◦ C 40◦ -94◦ C
(initial denaturation) (denaturation) (annealing; ßuorescence readout) (extension) (melting curve; ßuorescence readout each 0.5◦ C)
8. Mount the 96-well PCR plate into a real-time PCR cycler and start the cycler. 9. Analyze the real-time PCR curves by comparing the curves of the mutants with the controls (wt and empty plasmid; Fig. 3).
350 300
Fluorescence intensity
250 200 150 100 50 0 ⫺50 0
5
10
15
20
PCR cycles
Figure 3
Representative real-time PCR curves of KTQ variants.
25
30
Directed Evolution of DNA Polymerases
99 Current Protocols in Chemical Biology
Volume 2
In the case of the negative control (empty plasmid), no active enzyme is present, and thus no ampliÞcation should be visible. In the case of the positive control (plasmid with wt polymerase), a clear ampliÞcation curve should occur during the Þrst 10 cycles. The presence of PCR amplicons can be further checked during the melting curve analysis. In this case, the speciÞc ampliÞcates are 67-mers and melt at 87◦ to 88◦ C.
10. If desirable, determine the threshold-crossing point (Ct ) as a measure of ampliÞcation efÞciency. This parameter is deÞned as the point at which the reporter’s ßuorescence exceeds the background ßuorescence signiÞcantly and crosses a threshold. EfÞcient DNA polymerase variants will show a low Ct and yield a strong signal after a few PCR cycles, whereas less efÞcient variants will have a higher Ct . Inactive variants do not show any ampliÞcation and no Ct can be determined.
11. Select all clones able to yield an amplicon (or ampliÞed product) from the glycerol stocks and pool them together on a new 384-well deep-well plate using sterile wooden toothpicks or the automated liquid handling device for large libraries. Repeat all steps from Basic Protocol 1 (establishing a 384-well plate library), step 1 to step 6, to obtain a glycerol mutant library with PCR-active clones. BASIC PROTOCOL 4
SCREENING DNA POLYMERASE MUTANTS FOR ENHANCED SINGLE-NUCLEOTIDE DISCRIMINATION The lysates prepared in Basic Protocol 2 can be screened for enhanced discrimination between matched and mismatched primer-template complexes. Each mutant is tested in two reactions for its ability to discriminate between single-nucleotide mismatches. In one reaction, the primer-template complex is fully matched, whereas in the other reaction, the primer-template complex is incorrectly paired at the 3 end of the primer (see Fig. 4). Real-time PCRs are performed in the presence of SYBR Green I, thus enabling quantiÞcation of synthesized dsDNA during each cycling step (Fig. 5; Wilhelm and Pingoud, 2003). Subsequent melting-temperature measurements are performed to conÞrm the desired ampliÞcation product from those derived by false ampliÞcation. For this approach, a standard real-time PCR cycler for multiwell plates is sufÞcient.
Materials 10× KTQ reaction buffer (see recipe) 10 mM dNTP mix (10 mM each dATP, dCTP, dGTP, dTTP) 10,000× SYBR Green I (Invitrogen, cat. no. S-7563) 100 μM forward primer F20+ : 5 -d(CGT TGG TCC TGA AGG AGG AT)-3 (custom synthesis) 100 μM reverse primer F20− : 5 -d(CGC GCA GCA CGC GCC GCC GT)-3 (custom synthesis) 0.1 μM DNA template F90A (match): 5 -d(CCG TCA GCT GTG CCG TCG CGC AGC ACG CGC CGC CGT GGA CAG AGG ACT GCA GAA AAT CAA CCT A TC CTC CTT CAG GAC CAA CGT ACA GAG)-3 (custom synthesis) 0.1 μM DNA template F90G (mismatch): 5 -d(CCG TCA GCT GTG CCG TCG CGC AGC ACG CGC CGC CGT GGA CAG AGG ACT GCA GAA AAT CAA CCT G TC CTC CTT CAG GAC CAA CGT ACA GAG)-3 (custom synthesis) Cell lysates in 96-well plates (Basic Protocol 2) Directed Evolution of DNA Polymerases
96-well PCR plate, semi skirted (VWR) Automated liquid handling device (Hamilton Microlab Star) Reagent dispenser (Multidrop, Thermo ScientiÞc)
100 Volume 2
Current Protocols in Chemical Biology
match
DNA polymerase
T A mismatch
T A
DNA polymerase
T G
T G
Figure 4 In principle, the DNA polymerase catalyzes DNA synthesis in the matched case (upper half). Due to incorrect hybridization at the primer end, DNA synthesis should be suppressed in the mismatch case (lower half). Black bars represent primer/template duplexes.
350
Fluorescence intensity
300 250 matched
200 150
mismatched 100 50 0 0
5
10
15
20
25
30
PCR cycles
Figure 5 Representative real-time PCR curves of a KTQ variant amplifying a matched (blue curve) and a mismatched (red curve) primer template complex.
Refrigerated ßoor-model centrifuge with multiwell-plate carrier Optically clear heat-sealing Þlm (ABgene) Thermo-sealer or electric iron Real-time PCR thermal cycler 1. Prepare the master mix as described in Table 1, omitting DNA template F90A. Split the solution into two equal volumes of 600 μl and add, to each separate portion, 0.72 μl of the respective 0.1 μM template solution—F90A (match) and F90G (mismatch)—then mix. In order to screen lysates from one 96-well plate, two 96-well PCR plates have to be prepared, since each lysate is screened for matched and mismatched cases.
2. Fill every other row (A, C, E, G) of a 96-well PCR plate with 10 μl of the respective PCR master mix including F90A (matched case) and the remaining rows (B, D, F, H) with 10 μl of PCR master mix including F90G (mismatched case), using a reagent dispenser or a suitable multichannel pipet. CAUTION: The master mix contains SYBR Green I; be aware of the possible toxicity!
Directed Evolution of DNA Polymerases
101 Current Protocols in Chemical Biology
Volume 2
3. Distribute 10 μl of each cell lysate into each of two reaction wells (one containing F90A master mix and one containing F90G master mix) of the PCR plate using a liquid handling device or a suitable multichannel pipet. Matched and mismatched reactions of one mutant should ideally be directly neighbored; for example cell lysate with mutant 1 is present in A1 (matched case) and B1 (mismatched case), cell lysate with mutant 2 is present in A2 (matched case) and B2 (mismatched case) and so on. Cool the PCR plate during all steps to 4◦ C. Be sure to be able to correlate each clone position from the 96-well deep-well expression plate to the 96-well PCR plate.
4. Centrifuge the PCR plate at 1000 × g for a few seconds to ensure that all liquids are at the bottom of the wells. Seal plate with an optically clear heat-sealing Þlm using a thermo-sealer or an electric iron. Alternatively, a self-adhesive PCR sealing Þlm can be used.
5. Shake or vortex the sealed PCR plate for a few seconds. This step guarantees the homogeneity of the PCR mix.
6. Centrifuge the PCR plate again at 1000 × g for a few seconds to ensure that all liquids are at the bottom of the wells. The prepared PCR plate is now ready for real-time PCR and may be stored for a few hours at 4◦ C before performing the PCR.
7. Set up the real-time PCR thermal cycler using the following parameters: 1 cycle: 30 cycles:
1 cycle:
3 min 30 sec 35 sec 40 sec ramp
95◦ C 95◦ C 55◦ C 72◦ C 40◦ -94◦ C
(initial denaturation) (denaturation) (annealing; ßuorescence readout) (extension) (melting curve; ßuorescence readout each 0.5◦ C)
8. Mount the 96-well PCR plate into the real-time PCR thermal cycler and start the program. In this way, 48 mutants can be screened simultaneously in one PCR experiment. The screening throughput strongly depends on the capacity of the real-time cycler system and may be extended using, for example, a 384-well plate real-time PCR cycler.
9. Analyze the real-time PCR curves by comparing the threshold-crossing points (Ct ) of the canonical (match) versus noncanonical (mismatch) primer-template ampliÞcation. The Ct value of the match ampliÞcation is a measure for ampliÞcation efÞciency of the respective enzyme. The cycle difference (Ct) of match ampliÞcation to mismatch ampliÞcation is a good marker for the enzyme selectivity. Polymerase mutants yielding twice or higher the average value of the wt Ct should be tested again and further characterized. Control reactions should be included to compare the discrimination with wt (positive control) and to check for contamination (negative control). The presence of PCR amplicons can be further checked during the melting curve analysis. In this case, the speciÞc amplicons are 67-mers and melt at 87◦ to 88◦ C. Directed Evolution of DNA Polymerases
10. Verify interesting DNA polymerase mutants with increased ability to discriminate between the canonical (match) and noncanonical (mismatch) primer-template by
102 Volume 2
Current Protocols in Chemical Biology
repeating the expression and the above-described real-time PCR steps (see Basic Protocol 2). All veriÞed and interesting enzyme mutants should then be expressed and puriÞed according to individual protocols and further characterized by enzyme and template dilution series and kinetic measurements. All clones of interest should be sequenced.
SCREENING MUTANT DNA POLYMERASES FOR LESION-BYPASS ABILITY
BASIC PROTOCOL 5
In order to identify DNA polymerase variants that are able to synthesize DNA across lesions found in the templating DNA strand, it is assumed that a DNA lesion can be easily mimicked by misincorporation of a noncanonical base and extension of the nascent base pair. A primer-extension assay is presented below in which only three dNTPs are included, although extension of the template calls for all four dNTPs. This will lead either to an arrest of the DNA polymerase or to a misincorporation of a dNTP followed by extension of the mismatch. The careful selection of the template sequence and choice of the omitted dNTP are critical steps. By varying these parameters, the number of misincorporation events and the stringency of the screening can be directed.
Materials 10 μM primer F20+ : 5 -d(CGT TGG TCC TGA AGG AGG AT)-3 (custom synthesis) 1 μM DNA template F90A: 5 -d(CCG TCA GCT GTG CCG TCG CGC AGC ACG CGC CGC CGT GGA CAG AGG ACT GCA GAA AAT CAA CCT A TC CTC CTT CAG GAC CAA CGT ACA GAG)-3 (custom synthesis) 10× KTQ buffer (see recipe) 10 mM dNTP mix (only three dNTPs; 10 mM each dCTP, dGTP, and dTTP) 10 mM dNTP mix (all four dNTPs; 10 mM dATP, dCTP, dGTP, and dTTP) 500 mM Tris·Cl, pH 9.2 500 mM EDTA 20× SYBR Green I (prepare from 10,000× stock, as purchased from Invitrogen, cat. no. S-7563) Heat-inactivated E. coli lysates (see Basic Protocol 2) 384-well plates, black, ßat bottom, square shaped wells (VWR) Automated liquid handling device (Hamilton Microlab Star) Reagent dispenser (Multidrop, Thermo ScientiÞc) Self-adhesive aluminum/paper seal (ABgene) Thermal cycler Floor-model centrifuge with multiwell-plate carrier Plate reader for 384-well plates, heatable (POLARStar optima, BMG Labtech; http://www.bmglabtech.com) Prepare solutions 1. Prepare 420× reaction mix (without dNTPs): 141.75 μl 10 μM primer F20+ (225 nM Þnal) 945 μl 1 μM template F90A (150 nM Þnal) 630 μl 10× KTQ buffer (1× Þnal) 2357.25 μl H2 O. Mix primer and template with buffer and water in a 15-ml tube. Split the mix into several 1.5-ml tubes and heat to 94◦ C for 5 min in a heating block, then allow to cool to room temperature (this is the annealing step). After annealing, pool the reaction
Directed Evolution of DNA Polymerases
103 Current Protocols in Chemical Biology
Volume 2
mixtures in a 15-ml tube, place on ice, and add 126 μl of 10 mM dNTPs without dATP (10 mM each dCTP, dGTP, and dTTP). The 420-fold reaction mix will be sufÞcient to screen one 384-well plate.
2. Prepare 10× reaction mix for control reaction in the same way as above, but use all four dNTPs:
3.4 μl 10 μM primer F20+ (225 nM Þnal) 22.5 μl 1 μM template F90A (150 nM Þnal) 15 μl 10× KTQ buffer (1× Þnal) 56 μl H2 O 3 μl 10 mM dNTPs (10 mM dATP, dCTP, dGTP, and dTTP; 200 μM each dNTP Þnal; add last, after heating, as described in step 1) Keep on ice. Each 384-well plate contains eight control reactions using all four dNTPs.
3. Prepare stop solution (14 ml):
1.4 ml 500 mM Tris·Cl, pH 9.2 28 μl 500 mM EDTA 3.5 ml 20× SYBR Green I (3.3× Þnal) 9.1 ml H2 O. CAUTION: The stop solution contains SYBR Green I; be aware of the possible toxicity!
Set up reactions in 384-well plate 4. Place 10 μl of control reaction mix including all four dNTPs into wells A1, A2, A7, A8, A13, A14, A19, and A20 of a 384-well plate. Distribute 10 μl of the reaction mix with three dNTPs into all remaining wells using a multichannel pipettor or liquid handling device. Keep the plate chilled at all times. 5. Add 5 μl of heat-denatured E. coli lysate to each well using a multichannel pipet or liquid handling device. The 384-well plate will consist of four sections, each having 96 wells (A1 to P6, A7 to P12, A13 to P18, and A19 to P24). Each section has its own control reactions (wild-type enzyme with all four dNTPs: A1, A2; wild-type enzyme with only three dNTPs: B1, B2). One section corresponds to one 96-well deep-well expression plate.
6. Seal 384-well plate with aluminum adhesive foil and incubate for 15 min at 72◦ C in a thermal cycler. 7. Cool 384-well plate on ice, centrifuge brießy to recover any condensate and remove foil carefully. Add 30 μl of stop solution to each well and read ßuorescence in a plate reader using 485 nm excitation and 520 nm emission wavelength. Depending on the plate reader it may be necessary to adjust the gain to one of the positive controls (four dNTPs reaction using wild-type enzyme). Increasing the temperature at which the plate is read may decrease background signal due to melting smaller DNA fragments that result from primer/template pairs or only short extension products.
Directed Evolution of DNA Polymerases
Analyze data 8. Analyze the data from the 384-well plate in four sections according to the 96-well expression plates. For every section, calculate the mean ßuorescence value of the two positive controls (wild-type enzyme with four dNTPs, Fwt/4dNTPs ) and the two
104 Volume 2
Current Protocols in Chemical Biology
negative controls (wild-type enzyme with three dNTPs, Fwt/3dNTPs ). Calculate the ßuorescence threshold value Ft = Fwt/3dNTPs + F × 0.5. Microsoft Excel is sufÞcient for making these calculations. Reactions that reach a ßuorescence value equal or higher than the Ft are regarded as hit and need to be further analyzed. Note that The Ft was set to 0.5 (or 50%) of F. Depending on the screening criteria and the stringency this value can be adjusted to identify more or less positive variants.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps.
KTQ lysis buffer, 1× 10 ml 10× KTQ reaction buffer (see recipe) Add H2 O to 100 ml and mix Add 1 ml 10 mg/ml lysozyme (Sigma; cat. no. L7651; ∼0.1 mg/ml Þnal) and mix again Prepare fresh just before use and keep at 4◦ C The overall amount of 1× KTQ lysis buffer to be prepared must be adjusted according to the number of 96-well plates to be lysed: i.e., 800 μl of 1× KTQ lysis buffer per well; see Basic Protocol 2, step 8.
KTQ reaction buffer 10× stock: 60.57 g Tris base (1 M Þnal) Add 700 ml H2 O, and adjust to pH 9.2 with HCl Add: 21.14 g (NH4 )2 SO4 (160 mM Þnal) 5.08 g MgCl2 ·6H2 O (25 mM Þnal), 1 ml Tween 20 (1% Þnal) Add H2 O to 1 liter; Þlter sterilize (0.22 μm Þlter) Store up to 1 month at 4◦ C protected from light 1× KTQ reaction buffer: 1 ml of 10× KTQ reaction buffer Add H2 O to 10 ml and mix. Store up to 1 month at 4◦ C protected from light LB medium and plates 20 g LB medium, Lennox (BD Difco) 15 g agar (for plates; otherwise omit) Add H2 O to 1 liter and autoclave Cool to ∼50◦ C Add 1 ml of 100 mg/ml carbenicillin (if called for in protocol; 0.1 mg/ml Þnal) If preparing agar plates, pour 90-mm plates Store at room temperature until use pASK-IBA37plus, BsaI digested and dephosphorylated Prepare the following reaction mix: 2 μl 1 μg/μl pASK-IBA37plus (IBA GmbH) 1 μl 10 U/μl BsaI (New England Biolabs) 0.5 μl 10 mg/ml BSA continued
Directed Evolution of DNA Polymerases
105 Current Protocols in Chemical Biology
Volume 2
5 μl NEB Buffer 4 (New England Biolabs) 41.5 μl H2 O Incubate for 1 hr at 37◦ C, then heat-inactivate enzyme for 20 min at 65◦ C. Add 1 μl of 10 U/μl calf intestine phosphatase (New England Biolabs) to the reaction, mix by pipetting up and down, and incubate 1 hr at 37◦ C. Purify digested and dephosphorylated vector by agarose gel electrophoresis, as described in Basic Protocol 1, step 7.
SOC medium 20 g tryptone 5 g yeast extract 0.5 g NaCl Add distilled water to 1 liter and autoclave. When cooled to room temperature add 20 ml of a sterile-Þltered 1 M glucose solution to yield 20 mM glucose as Þnal concentration. Store at room temperature until use.
Taq buffer, 10× 100 mM Tris·Cl, pH 8.8 500 mM KCl 0.8% (v/v) Nonidet P-40 Store up to 4 weeks at 4◦ C COMMENTARY Background Information
Directed Evolution of DNA Polymerases
Methods for directed evolution of proteins or enzymes can be divided into two major subgroups, i.e., screening- or selection-based methods. Both methods are powerful and have yielded new variants of enzymes with properties that were enhanced by some orders of magnitude compared to the parent enzyme or with properties that were newly established in their class. Examples include DNA and RNA polymerases, P450, enantioselective enzymes, [catalytic] antibodies, binding proteins, or DNA methyltransferases, to name just a few (Glieder et al., 2002; Lee et al., 2002; Brogan et al., 2007; Reetz and Wu, 2009). Screening and selection methods differ and are limited by the way phenotype and genotype are linked to each other and how the single variants are separated from each other. In screening-based methods, mutants are physically separated from each other either in different wells on a microtiter plate or as plaques on a nourishing surface (e.g., agar plate). Each position has to be addressed either manually by the experimenter or by an automated handling system which renders the method timeand cost-intensive relative to selection, and requires a lot of storage space. For these reasons, screens in an academic environment are limited to a library size of <106 variants and seem to be less suited for directed evolution. The
chances of successfully evolving a protein in a speciÞc way are small compared to the mutational space possible in evolutionary experiments. The beauty and advantage of screening approaches lies in their versatile experimental design; nearly any evolutionary question can be approached since variants are physically separated from each other, and expressed proteins can be tested in separate vessels under experimental conditions that are independent from their host’s needs. Selection-based methods rely on the physical connection between genotype and phenotype—i.e., a successful mutant enzyme has to convey certain properties to its host in such a way that it can survive or be selected among the unsuccessful candidates. There is no physical separation between the different mutant-expressing hosts in separated vessels. Libraries can consist of 109 to 1010 members, which is substantially larger than what screening-based methods can handle, and a greater mutational space can be covered. Yet, the weakness of this approach lies in the limited possibilities for experimental design. The enzymatic substrate has to be covalently connected to the genotype-carrying host (e.g., phage) to separate successful variants from unsuccessful ones, or it must speciÞcally ensure viability of the host to increase its survival chances in the presence of an
106 Volume 2
Current Protocols in Chemical Biology
otherwise deadly inßuence/compound. Experimental conditions such as temperature or pH have to be in the range of the expression system, and thus are difÞcult to achieve to meet the needs of extremophile enzyme classes (e.g., thermostable DNA polymerases).
Critical Parameters and Troubleshooting Low efÞciency of ligation Use high-quality agarose for the gelpuriÞcation step. Different grades are available; choose the ultra-pure grade, since cheaper grades might contain impurities that interfere with ligation. Use low-melting point agarose. Taq DNA polymerase from the epPCR is known to bind to the PCR product and thus can interfere with the ligation if not removed thoroughly (Cirino et al., 2003). Choose a good and reliable reaction clean-up kit for this step. Low yields after error-prone PCR Try different concentrations of Mn2+ in epPCR. Increase template concentration. Use fresh and high-quality dNTPs. Increase MgCl2 concentrations to stabilize non-complementary base pairs. No product formation in positive control reactions Make sure that KTQ wt DNA polymerase was expressed by checking aliquots of the expressed and heat-denatured lysates by SDS-PAGE. Low signal-to-background ratio Increase the temperature at which the extension product is read (above the melting point of the primer/ template pair) to avoid primer/template complex formation. Increase the length of the template to yield more dsDNA by extension of the primer. Be sure primer is present in excess over template in the annealing reaction to ensure all templates are loaded with primers and thus available for extension by the DNA polymerase. Real-time PCR experiments are not reproducible Carefully adjust the concentrations of DNA templates F90A and F90G as monitored by absorption at 260 nm. Make sure that both template stocks are at equal concentrations.
Make sure that no contamination occurred between different wells during the inoculation step and cell lysate preparation. Perform SDSPAGE to check the integrity of the lysates.
Anticipated Results The initial library after error-prone PCR and cloning of the randomized yielded around 40% PCR-active clones (Sauter and Marx, 2006). These clones were pooled to yield the Þnal library that was employed for further screening approaches. DNA sequencing of several variants revealed the introduction of one to eight mutations per clone. Considering an amplicon of 1655 bp during epPCR, this results in a mutation rate of ∼0.06% to 0.5% per nucleotide, which seems to be moderate. Introduction of more mutations by increasing the Mn2+ concentration would lead to loss of PCR activity for most variants, as can be concluded from the data above. The wt enzyme is not able to discriminate between matched and mismatched primer termini and thus shows a Ct value of 0. It is possible to obtain polymerase variants that are able to discriminate matched versus mismatched primer termini for ∼10 to 13 rtPCR cycles (Strerath et al., 2007). Several DNA lesions (e.g., abasic sites) present a severe block for replicating DNA polymerases and Klentaq is not able to synthesize DNA across an abasic site. Employing the above described directed evolution approach and screening for primer extension in the presence of only three natural dNTPs yields polymerase variants that are able to handle various DNA damages or even amplify UV-irradiated DNA (Gloeckner et al., 2007).
Time Considerations Basic Protocol 1: Error-prone PCR library construction Basic Protocol 1 is the most time-consuming of all the protocols in this unit. Establishing the epPCR, cloning, and analyzing the diversity of the library by DNA sequencing can take 2 to 3 weeks, since going back to the initial PCR reaction might be necessary in order to adjust the Mn2+ to achieve the desired mutation rate. Picking of single colonies and transferring to 384-well plates can be done manually, and one person can pick several 384well plates per day, although an automated system is recommended. Depending on the desired size of the library, this step can consume 1 week. A total time minimum of at least
Directed Evolution of DNA Polymerases
107 Current Protocols in Chemical Biology
Volume 2
4 weeks should be considered to establish the initial library. Basic Protocol 2: Expression of DNA polymerase variants and lysate preparation Basic Protocol 2 requires 2 hr for the inoculation of the defrosted library into 96well deep-well plates and the rest of the day for monitoring their growth and induction of overexpression of the enzymes. Depending on E. coli culture growth, inoculation can be performed in the evening hours, followed by induction early the next morning. Preparation of the lysates for screening requires approximately another half-day. Basic Protocol 3: Screening for PCR activity Time considerations for Basic Protocol 3 strongly depend on the throughput capacity of the real-time cycler and/or plate reader system. PCR mixes and PCR plates have to be prepared, which takes approximately 1 to 2 hr. Transfer of the lysates from the expression plate to the PCR plate should be performed with a liquid handling device or a suitable multichannel pipet, and takes another hour. Once the PCR protocol is set up, one plate can be cycled within 2 hr, and data analysis takes 1 additional hour. Expression in 96-well deep-well plates is described in this unit because higher expression levels are typically observed compared to 384-well deep-well plates. Screening for increased selectivity–single nucleotide discrimination Basic Protocol 4 can be performed within 1 working day, starting with the preparation of the reaction mix, followed by real-time PCR and data analysis. Because the airtightsealed lysates can be safely stored at 4◦ C up to 2 weeks without signiÞcant loss of PCR activity, they can be prepared as stocks and afterwards screened by real-time PCR. Note that this is not the case for primer-extension screening systems (see Basic Protocol 5).
Directed Evolution of DNA Polymerases
Basic Protocol 5: Screening for lesion-bypass ability Basic Protocol 5 is a fast and straightforward screening approach which can be completed within 1 to 2 days. Setting up solutions and running test primer extensions to adjust temperatures can be done in half a day; screening one 384-well plate takes about 2 hr. Depending on the library size (i.e., number of 384-well plates) to be screened, time considerations may have to be adjusted. The subsequent analysis using appropriate software
(e.g., Excel) will require another half a day to 1 day. Finally, all identiÞed mutants should be sequenced (requiring 3 to 5 days, starting from plasmid preparation to obtaining sequencing results) and corresponding proteins puriÞed (1 day for overexpression, 1 day for puriÞcation by Ni-NTA afÞnity chromatography, and 1 additional day, at least, for purity veriÞcation by SDS-PAGE and concentration adjustment). Further characterization of the enzymes by primer-extension reactions or kinetic measurements will require several additional days. In general, timing for the screening steps strongly depends on the library size and the throughput capacity of the real-time cycler and/or plate reader system.
Literature Cited d’Abbadie, M., Hofreiter, M., Vaisman, A., Loakes, D., Gasparutto, D., Cadet, J., Woodgate, R., P¨aa¨ bo, S., and Holliger, P. 2007. Molecular breeding of polymerases for ampliÞcation of ancient DNA. Nat. Biotechnol. 25:939-943. Barnes, W.M. 1992. The Þdelity of Taq polymerase catalyzing PCR is improved by an N-terminal deletion. Gene 112:29-35. Brogan, A.P., Eubanks, L.M., Dickerson, T.J., and Janda, Antibody-catalyzed oxidation tetrahydrocannabinol. J. Am. 129:3698-3702.
Koob, G.F., K.D. 2007. of delta(9)Chem. Soc.
Cadwell, R.C. and Joyce, G.F. 1992. Randomization of genes by PCR mutagenesis. PCR Methods Appl. 2:28-33. Cirino, P.C., Mayer, K.M., and Umeno, D. 2003. Generating mutant libraries using error-prone PCR Methods Mol. Biol. 231:3-9. Elbing, K. and Brent, R. 2002. Growth on solid media. Curr. Protoc. Mol. Biol. 59:1.3.1-1.3.6. Glieder, A., Farinas, E.T., and Arnold, F.H. 2002. Laboratory evolution of a soluble, selfsufÞcient, highly active alkane hydroxylase. Nat. Biotechnol. 20:1135-1139. Kranaster, R. and Marx, A. 2009. Taking Þngerprints of DNA polymerases: Multiplex enzyme proÞling on DNA arrays. Angew. Chem. Int. Ed. Engl. 48:4625-4628. Lee, Y.-F., TawÞk, D.S., and GrifÞths, A.D. 2002. Investigating the target recognition of DNA cytosine-5 methyltransferase HhaI by library selection using in vitro compartmentalisation. Nucleic Acids Res. 30:4937-4944. Loakes, D., Gallego, J., Pinheiro, V.B., Kool, E.T., and Holliger, P. 2009. Evolving a polymerase for hydrophobic base analogues. J. Am. Chem. Soc. 131:14827-14837. Loh, E., Choe, J., and Loeb, L.A. 2007. Highly tolerated amino acid substitutions increase the Þdelity of Escherichia coli DNA polymerase I. J. Biol. Chem. 282:12201-12209.
108 Volume 2
Current Protocols in Chemical Biology
Reetz, M.T. and Wu, S. 2009. Laboratory evolution of robust and enantioselective Baeyer-Villiger monooxygenases for asymmetric catalysis. J. Am. Chem. Soc. 131:15424-15432. Sauter, K.B.M. and Marx, A. 2006. Evolving thermostable reverse transcriptase activity in a DNA polymerase scaffold. Angew. Chem. Int. Ed. 45:7633-7635. Voytas, D. 2000. Agarose gel electrophoresis. Curr. Protoc. Mol. Biol. 2.5A.1-2.5A.9. Wilhelm, J. and Pingoud, A. 2003. Real-time polymerase chain reaction. ChemBioChem 4:11201128.
Key References Sauter and Marx, 2000. See above. Construction of the DNA polymerase library (KTQ) by epPCR as described in Basic Protocol 1.
Strerath, M., Gloeckner, C., Liu, D., Schnur, A., and Marx, A. 2007. Mutations in motif C on the mismatch-extension selectivity of Thermus aquaticus DNA polymerase. ChemBioChem 8:395-401. Screening for PCR activity and increased selectivity–single nucleotide discrimination as described in Basic Protocols 3 and 4. Summerer, D., Rudinger, N.Z., Detmer, I., and Marx, A. 2005. Enhanced Þdelity in mismatch extension by DNA polymerase through directed combinatorial enzyme design. Angew. Chem. Int. Ed. 44:4712-4715. Screening for increased selectivity–single nucleotide discrimination as described in Basic Protocol 4.
Gloeckner, C., Sauter, K.B.M., and Marx, A. 2007. Evolving a thermostable DNA polymerase that ampliÞes from highly damaged templates. Angew. Chem. Int. Ed. 46:3115-3117. Screening for lesion-bypass ability as described in Basic Protocol 5.
Directed Evolution of DNA Polymerases
109 Current Protocols in Chemical Biology
Volume 2
Nucleotide Analogues as Probes for DNA and RNA Polymerases Robert D. Kuchta1 1
University of Colorado, Boulder, Colorado
ABSTRACT Nucleotide analogues represent a major class of anti-cancer and anti-viral drugs, and provide an extremely powerful tool for dissecting the mechanisms of DNA and RNA polymerases. While the basic assays themselves are relatively straightforward, a key issue is to appropriately design the studies to answer the mechanistic question of interest. This unit addresses the major issues involved in designing these studies, and some of the potential difficulties that arise in interpreting the data. Examples are given for the type of analogues typically used, the experimental approaches with different polymerases, and issues with data interpretation. Curr. Protoc. Chem. Biol. 2:111C 2010 by John Wiley & Sons, Inc. 124 Keywords: polymerase r nucleotide r DNA r RNA r kinetics
INTRODUCTION Nucleotide analogues play two important roles biochemically. First, DNA and RNA polymerases represent one of the most important classes of enzymes targeted by both anticancer and anti-viral chemotherapeutics. Over a dozen clinically useful nucleotide analogues target viral RNA and DNA polymerases and at least 8 clinically useful anticancer agents exert their antiproliferative effects via their interactions with cellular DNA polymerases (Fig. 1) (Villarreal, 2001; Vivet-Boudou et al., 2006; Burton and Everson, 2009; Parker, 2009). Almost all of these analogues consist of a normal or slightly modified base conjugated to a modified sugar. Second, nucleotide analogues have provided a means of obtaining critical mechanistic insights into the functioning of DNA and RNA polymerases (Thompson and Kuchta, 1995; Morales and Kool, 1998; Morales and Kool, 2000; Ogawa et al., 2000; Kool, 2002; Matsuda et al., 2003; Washington et al., 2003; Henry et al., 2004; Meyer et al., 2004; Ramirez-Aguilar and Kuchta, 2004; Kim et al., 2005; Kincaid et al., 2005; Zhang et al., 2005; Sintim and Kool, 2006; Berdis and McCutcheon, 2007; Hwang and Romesberg, 2008; Leconte et al., 2008; Lee et al., 2008; Cavanaugh et al., 2009; Lee and Berdis, 2009; Patro et al., 2009). This includes, for example, how these enzymes differentiate ribose and 2 deoxyribose, and how the enzymes discriminate between right and wrong (d)NTPs. This review will focus on using nucleotide analogues to better understand the mechanism
of DNA/RNA polymerases. Two different classes of nucleotide analogues are routinely used to study polymerases—sugar analogues and base analogues—and both will be discussed. The results of this type of study potentially have two other very practical applications. As alluded to above, since multiple nucleoside-based chemotherapeutics obtain their efficacy due to effects on DNA and RNA polymerases, the results of these studies might eventually lead to novel chemotherapeutics. Additionally, a long-standing goal of molecular genetics researchers has been the expansion of the genetic code via the development of an additional base pair(s) that an organism can maintain and replicate with high efficiency and fidelity. While progress has been made in this direction (Hirao et al., 2006, 2007; Hwang and Romesberg, 2008; Leconte et al., 2008), substantial improvements in current novel base pairs must be made.
SUGAR ANALOGUES As suggested in Figure 1, the synthesis of nucleotides containing sugar analogues and their interactions with polymerases has primarily been driven by the overriding goal of developing new anti-cancer or anti-viral chemotherapeutics. Reported modifications include: (1) replacing the heteroatoms at the 2 and 3 carbons with hydrogen, another heteroatom or an alkyl group (R2 , R2 , and R3 in Fig. 2); (2) replacing the Hs at the 2 carbon with a heteroatom or alkyl group; (3) replacing
Current Protocols in Chemical Biology 2: 111-124, June 2010 Published online June 2010 in Wiley Interscience (www.interscience.wiley.com). DOI: 10.1002/9780470559277.ch090203 C 2010 John Wiley & Sons, Inc. Copyright
Nucleotide Analogues as Probes for Polymerases
111 Volume 2
A NH2
NH2
NH2
O
F N N
NH O HO
S
HO
O
N
O
N
N HO
N
O
N
O S
HO O
O
5-Fluoro-2⬘,3⬘dideoxy-3⬘-thiacytidine
2⬘,3⬘-Dideoxy-didehydro thymidine
2⬘,3⬘-Dideoxycytidine
2⬘,3⬘-Dideoxy-3⬘ -thiacytidine
O
NH2 NH
HO
N
O N
O
N
O
HO
NH HO
O
N
2⬘,3⬘-Dideoxyinosine NH2 N
N O
N
N
P
HO
O
OH
OH Tenofovir
O N
NH2
N
HO O
Acyclovir
N
NH
NH N
O
O
N N
N
O
Adefovir
O
N
O
N
P
OH
Ribavirin NH2
N
HO
N
O
HO
3⬘-Azidothymidine
O
N
N3
HO
O
N
Ganciclovir
NH NH2
HO
N
N
NH2
Penciclovir
HO
Figure 1
(continued)
Nucleotide Analogues as Probes for Polymerases
the 2 and 3 carbons with a heteroatom (Y and Z), most commonly S or O; (4) removing the 2 and/or 3 carbons to generate acyclic sugars; (5) replacing the 4 -OH with N, S, or an alkyl group (X); (6) adding alkyl groups to the 4 -carbon (R4 ); (7) replacing the 5 -hydroxyl with N or a phosphonate (R1 ), and; (8) interconversion of both the sugar stereochemistry (D vs. L) and anomeric configuration (α vs. β) (Lee and Berdis, 2009; Parker, 2009). Thus, a huge number of nucleotides containing modified sugars exist for probing how a polymerase interacts with the sugar moiety of a nucleotide.
After synthesis of the sugar, the nucleosides are typically generated via relatively standard synthetic methods. In some cases, the sugar modification is installed after generation of the glycosidic bond (e.g., AZT; Horwitz et al., 1964; Lin and Mancini, 1983). Conversion to the corresponding nucleoside triphosphate or phosphoramidite also typically involves established procedures, although it should be noted that triphosphorylation is still a less-thanoptimal reaction even though researchers have used it for more than 30 years (Ludwig, 1981). In some cases, however, phosphorylation may require some modification of the procedures
112 Volume 2
Current Protocols in Chemical Biology
B
NH2
NH2 N
N HO
O
N
HO
O
O
NH2 N
N N
HO
O
O
HO
HO
HO
Cytarabine
Decitabine
NH2
F
Gencitabine NH2
NH2
N
N N
O F
OH
HO
N
N N
N
N F
N
HO
N
Cl
N
HO
N N
Cl
F
OH HO
HO
HO Fludarabine
Cladribine
Clofarabine
Figure 1 (continued) Examples of clinically useful anticancer and antiviral nucleosides that involve DNA polymerase activity. (A) Antiviral agents. (B) Cancer chemotherapeutics.
R1
base X R2⬘ R4
Figure 2
Y
Z
R3
R2
Reported modifications of the sugar of a nucleoside.
used with (deoxy)ribose due to the presence of additional heteroatoms in the sugar. For some modified nucleosides, chemoenzymatic syntheses have been developed to increase the yield of triphosphate. A number of reviews discuss nucleoside triphosphorylation in detail and can provide the interested reader with more detailed information (Burgess and Cook, 2000; Wu et al., 2004).
BASE ANALOGUES In contrast to the wide use of sugarmodified nucleosides as chemotherapeutics, very few nucleosides bearing highly modified bases are used as chemotherapeutics (Fig. 1). This appears to largely reflect the frequent inability of cellular enzymes to convert the basemodified nucleoside into a nucleoside triphosphate, the actual inhibitor of DNA or RNA
polymerases. The base analogues found in clinically useful therapeutics typically contain relatively subtle modifications, addition of a halogen at C-2 of adenine or at C-5 of a pyrimidine, replacement of N6 of adenine with oxygen, and replacement of O6 of guanine with S. Ribavirin, however, is an exception to this trend. Rather, developing an expanded genetic code and determining how polymerases identify the incoming (d)NTP as right or wrong has driven the synthesis and testing of most base-modified nucleotides. Whereas development of anti-cancer and anti-viral nucleosides has typically involved relatively minor modification of the base, studies to examine expanding the genetic code and to elucidate how polymerases differentiate right from wrong (d)NTPs have involved both small and large base modifications (Thompson
Nucleotide Analogues as Probes for Polymerases
113 Current Protocols in Chemical Biology
Volume 2
CH3 N
CH3
N
N
N
N
R
N
R
CF3
NO2
N N
R
NH2
N
N
N
N R
N CF3 R
N R N
N
N
N
R
R
N
NH N F
R
CH3 N
N
N R F
N
N R
N
N N F
N
N
N
N
R
NH2
N R Cl
R
N
NH2
N
NH2 Cl
F
N
O
N
R
R
O
NH2
N
R
R Cl
Br
Cl R
NH2
N
N
Cl O
N
R
N
N
N
N
N
R NH
N
N
N
N
NO2 O
N
Cl
O
NH2 N
R
N R
CF3
N
N
N R
N
N
N
Cl
R
R
R
R
R
S N N
N
R
N O
R
O
N
N
R
R
N
O
O
N
R
N
R
R
R
N
HN N N
N R
R
N O
N R
R
R
NH
S
N N
N
N
N
O
R
R
R
S
R
Cl
F
Cl N
R
N
R
R
R
N
N R
R
S
N R
R
S
N
O
R
Figure 3 Examples of modified bases that have been used to probe DNA and/or RNA polymerases. R = ribose or 2 -deoxyribose. This list is not meant to be comprehensive, and my apologies to those researchers and their compounds that were not included.
Nucleotide Analogues as Probes for Polymerases
and Kuchta, 1995; Morales and Kool, 1998, 2000; Ogawa et al., 2000; Kool 2002; Matsuda et al., 2003; Washington et al., 2003; Henry et al., 2004; Meyer et al., 2004; RamirezAguilar and Kuchta, 2004; Kim et al., 2005; Kincaid et al., 2005; Zhang et al., 2005; Sintim and Kool, 2006; Berdis and McCutcheon, 2007; Hwang and Romesberg, 2008; Leconte et al., 2008; Lee et al., 2008; Cavanaugh et al., 2009; Lee and Berdis, 2009; Patro et al., 2009). Figure 3 provides a sampling of the modified bases that have been converted into nucleoside triphosphates and templating nucleotides, and the ability of DNA and/or RNA polymerases to replicate them probed. It should be noted that most of these base analogues have been used with DNA polymerases and not RNA polymerases. As one might expect from the large number and varied structures of the base ana-
logues, a wide variety of synthetic schemes have evolved for both base synthesis and formation of the glycosidic bond, especially for C- versus N-glycosidic linkages. Typically, the reaction sequence involves synthesis of the base followed by glycosylation, and then either triphosphorylation or generation of the phosphoramidite for incorporation into DNA or RNA. In some cases, however, one generates the glycosidic bond and then either builds the entire base or just slightly modifies the base (Piccirilli et al., 1990).
Interaction of Modified Nucleotides with DNA and RNA Polymerases This review will consider the interaction of modified nucleotides with polymerases from two different aspects—what are the general assays that one uses to ask specific questions, and what are the limitations and precautions
114 Volume 2
Current Protocols in Chemical Biology
Figure 4
DNAn ⫹ [␣-32P]dNTP
[32P]DNAn ⫹ 1 ⫹ PPi
[32P]DNAn ⫹ dNTP
[32P]DNAn ⫹ 1 ⫹ PPi
General methods to analyze dNTP polymerization onto a primer-template.
that one must consider when interpreting the results of kinetic assays. For simplicity, the reactions will generally be described in terms of DNA polymerases and dNTPs, although many of the concerns and approaches also apply to RNA polymerases. At the end of this section, some special considerations with RNA polymerases are considered. Potentially, one could use [α-32 P]dNTP (analogues) to measure incorporation as shown in Figure 4. While feasible for the natural dNTPs since the radiolabeled forms are commercially available, this is not really convenient for the analogues since it would require synthesis of [α-32 P]dNTP analogues. This approach is also disadvantageous in terms of accuracy. After performing the assay, the amount of [32 P] incorporated into DNA requires separating unincorporated [α-32 P]dNTP from [32 P]-labeled DNA, typically either by gel electrophoresis or filter binding. Thus, this method requires accurate loading of either the filter or the gel. A much more accurate approach is to use 5 -[32 P]-primer-template ([32 P]DNAn ) and perform the assay using unlabeled dNTP (analogue), and then separate the starting [32 P]DNAn from the product [32 P]DNAn+1 by gel electrophoresis (Fig. 4). Now, the amount of product is given by the ratio of [32 P]DNAn+1 /([32 P]DNAn +[32 P]DNAn+1 ). Importantly, determining the amount of product ratiometrically means that errors in loading the gel do not affect the accuracy of the measurement (Ogawa et al., 2000; Chiaramonte et al., 2003; Urban et al., 2009). Depending upon the length of products one wants to analyze, different percentage acrylamide gels will suffice. For short products around 2 to 9 nucleotides long, 20% gels work well, while for longer products lower percentage gels work well (e.g., 15% acrylamide for separating 15- and 16-mers). The lower percentage gels are easier and faster to run, although this comes at the expense of resolution. The identity of the base at the 3 -terminus will alter the electrophoretic mobility of the DNA, especially shorter DNA (less than around 20 nucleotides, depending on the base composition of the DNA). Thus, one can often dif-
ferentiate between different nucleotides at the 3 -terminus of otherwise identical sequences (Kuchta and Willhelm, 1991; Thompson and Kuchta, 1995; Ramirez-Aguilar et al., 2005; Urban et al., 2009). Often, however, this will require the higher resolution obtained with higher percentage acrylamide gels. Alternatively, one may find that the polymerase does not incorporate the dNTP analogue. In this case, one can only determine how tightly the dNTP analogue binds to the enzyme by measuring its ability to inhibit incorporation of a normal dNTP. Thus, one would measure the ability of different analogue dNTP concentrations to inhibit normal dNTP incorporation. This will give an IC50 for inhibition. If one assumes, and is generally the case, that the dNTP analogue inhibits activity competitively with respect to dNTP, one can obtain the Ki for the dNTP analogue from the KM for the dNTP and the Michaelis-Menten expression for competitive inhibition. Alternatively, one can measure inhibition at multiple substrate (dNTP) and multiple analogue concentrations to directly obtain the Ki (Segel, 1975). Steady-state versus pre-steady-state kinetics. Two general methodologies can be used— steady-state and pre-steady-state. In general, steady-state analysis is easier and faster, while pre-steady-state analysis can provide more detailed information (Kuchta et al., 1987; Donlin et al., 1991; Patel et al., 1991; Wong et al., 1991; Hsieh et al., 1993). The latter typically requires the experimenter to use rapid quench methods and consumes much larger quantities of enzyme. For studies using nucleotide analogues, one most commonly wants to determine how modifying the structure of the incoming dNTP or templating nucleotide affects the efficiency of dNTP incorporation. Both approaches are equally accurate for obtaining this information. Steady-state methods provide a kcat /KM whereas pre-steady-state methods provide a kcat /KD (Apparent) . While the specificity constants for different substrates can be interpreted and compared to those of other substrates in a straightforward manner, extreme care must be taken in interpreting the individual kinetic parameters that make up the second-order rate constant.
Nucleotide Analogues as Probes for Polymerases
115 Current Protocols in Chemical Biology
Volume 2
For processive polymerases, even mildly processive enzymes such as pol α or Klenow Fragment, the rate of DNA dissociation likely limits the rate of correct dNTP polymerization such that kcat (Vmax ) reflects the rate of DNA dissociation (Kuchta et al., 1988). For analogues, it is impossible to say a priori what step(s) it represents. Thus, by comparing Vmax ’s for a normal dNTP and a dNTP analogue, one may end up comparing two different steps in the catalytic cycle, a relatively meaningless exercise. Likewise, the KM for the dNTP is equally hard to interpret. For both a normal dNTP and a dNTP analogue, it may include contributions from any and/or all steps of the reaction cycle, including the rate of DNA dissociation. In pre-steady-state studies where one examines a single turnover of the enzyme, kcat typically reflects all steps through dNTP polymerization. For many but probably not all polymerases, a conformational change prior to chemistry limits the overall rate (Dahlberg and Benkovic, 1991; Patel et al., 1991; Showalter and Tsai, 2002; Tsai and Johnson, 2006). For incorporation of an analogue, however, the kcat might, or might not, be the rate of the conformational change step— this can only be ascertained through more extensive analysis. The KD (Apparent) might represent the true KD for dNTP binding (i.e., formation of the E-dNTP collision complex), but caution should be taken when interpreting it this way since there may also be fast, kinetically invisible steps involved as well. In addition to measuring incorporation of a dNTP (analogue), one generally wants to know how fast the polymerase adds additional nucleotides onto this just-incorporated nucleotide (analogue). Two different approaches can be used. For example, let us suppose that one wants to measure how fast a polymerase polymerizes additional dNTPs onto a just incorporated nucleotide analogue (i.e., polymerization of dYTP in Fig. 5). First, one can chemically synthesize the appropriate primer-template, and then measure dNTP polymerization using either steady-state or pre-steady-state approaches. This approach is advantageous in that one can use either steady-state or pre-steady-state methods to measure polymerization of additional dNTPs,
Nucleotide Analogues as Probes for Polymerases
and one can measure how incorporation of the analogue affects other properties of the DNA (e.g., how the analogue affects DNA binding). However, when the analogue is at the 3 -terminus of the primer strand, this approach is disadvantageous since one must synthesize the appropriate analogue precursor and load it onto the activated solid support. If one only wants to synthesize a small amount of the primer containing the analogue at the 3 terminus, one can use an enzyme to incorporate the dNTP analogue into the DNA and then purify the desired product (Beckman et al., 2007). Alternatively, one can use the running start methodology developed by Goodman and coworkers (Boosalis et al., 1987; Cai et al., 1993). Now, one starts with the original primer-template (DNAn in Scheme 2), and then includes the dNTP analogue (dXTP) along with varying concentrations of the next correct dNTP (dYTP). Initial polymerization of the dNTP analogue generates DNAn+1 , which then has a choice of dissociating from the enzyme or being further elongated to DNAn+2 . By measuring the relative amounts of DNAn+2 and DNAn+1 as a function of the dNTP concentration, one can obtain the efficiency with which the polymerase added the next correct dNTP onto the analogue. This latter approach is limited, however, to only providing information on the efficiency of elongation. Typically, it is useful to examine polymerization of the next several nucleotides after a base pair containing an analogue since the analogue can affect polymerization events significantly downstream of the analogue. For example, incorporating 3-deazaadenine or 3-deazaguanine very strongly inhibits polymerization of the dNTP two nucleotides downstream by B family polymerases (Hendrickson et al., 2004; Beckman et al., 2007; Cavanaugh et al., 2009), but has very mild effects on polymerization of the next correct dNTP. Similarly, incorporation of gemcitabine has only small effects on polymerization of the next correct dNTP, but potently inhibits polymerization of the following dNTP (Huang et al., 1991; Plunkett et al., 1996). Finally, incorporation of the anti-hepatitis B drug entecavir (Fig. 6) by HIV reverse transcriptase
dXTP [32P]DNAn Figure 5
dYTP [32P]DNAn ⫹ X
[32P]DNAn ⫹ X⫹ Y
Polymerization of two consecutive dNTPs by a DNA polymerase.
116 Volume 2
Current Protocols in Chemical Biology
O NH
N N
N
HO
Figure 6
NH2
OH
Structure of entecavir.
ACVTP DNAn
dNTP DNAn⫹ACV
DNAn⫹ACV•dNTP
exonuclease
DNAn
Figure 7
X DNAn•dNTP
Inhibition of herpes DNA polymerase by acyclovir triphosphate (ACVTP).
results in strong chain termination 3 nucleotides later (Tchesnokov et al., 2008). For those nucleotide analogues whose incorporation results in chain termination, an important question becomes whether or not the next correct dNTP can still bind. If it can bind, this result raises the possibility that this analogue may be a particularly potent inhibitor of the polymerase. The classic example of this is acyclovir triphosphate and herpes polymerase (Reardon, 1989, 1990). If one only adds acyclovir triphosphate and an appropriate primertemplate to herpes polymerase, the enzyme incorporates the acyclovir and then the exonuclease associated with the polymerase removes it. Under these conditions, acyclovir triphosphate is a very poor inhibitor of the polymerase. However, if one also includes the next correct dNTP (Fig. 7), herpes polymerase incorporates the acyclovir to generate DNAn+ACV , and then the next correct dNTP binds to generate E-DNAn+ACV -dNTP locked in the polymerase active site. Under these conditions, acyclovir triphosphate very potently inhibits herpes polymerase, which is also the observed result in vivo. As implied by this discussion, one can identify dNTP analogues with this property via “induced inhibition.” If you assay your enzyme in the presence of just the dNTP analogue and a competing normal dNTP, the dNTP analogue weakly inhibits your polymerase. However, if you also include
the second correct dNTP, now the dNTP analogue potently inhibits the enzyme.
EFFECTS OF DNA SEQUENCE The sequence of the DNA used for assaying polymerase activity can have remarkably large effects on the properties of the enzyme. The rates of misincorporation along a template vary dramatically, even for the same misincorporation event (e.g., misincorporation of dGTP opposite different templating Ts) (Kunkel, 1985; Kunkel and Alexander, 1986; Lai and Beattie, 1988; Bell et al., 1997). The molecular mechanism by which DNA sequence affects polymerase fidelity remains unclear. While the effect of varying the DNA sequence on polymerization of nucleotide analogues has not been examined, the data for misincorporation of natural dNTPs suggests that polymerases will incorporate dNTP analogues at different rates in different sequence contexts. Thus, it is essential to use the same DNA sequence for all of the dNTP (analogues) tested.
3 -5 EXONUCLEASE ACTIVITY Many DNA polymerases contain a proofreading 3 -5 exonuclease that can remove the 3 -terminal nucleotide from both singlestranded and double-stranded DNA. If one wants to just study the incorporation of a dNTP analogue, the exonuclease activity can
Nucleotide Analogues as Probes for Polymerases
117 Current Protocols in Chemical Biology
Volume 2
seriously complicate the analysis—for incorporation of a single nucleotide, the true incorporation rate will be faster than the measured rate because the exonuclease will have hydrolyzed some of the incorporation product (DNAn+1 ) back to the starting material (DNAn ). The most common solution to this problem is to use a form of the polymerase that lacks exonuclease activity (exo- Klenow Fragment, exo- herpes polymerase, etc.) generated by mutating the exonuclease active site. Alternatively, one may want to know how fast the exonuclease hydrolyzes the 3 -terminal nucleotide on the primer strand due to the presence of a nucleotide analogue in either the template or primer strand. Assays are performed analogously to the polymerase assay, except now one starts with [32 P]DNAn+1 and monitors the production of [32 P]DNAn by denaturing gel electrophoresis.
CHALLENGES OF RNA POLYMERASES Depending upon the RNA polymerase used, these enzymes can present a special challenge to examining incorporation of nucleotide analogues. In some cases, one can use a RNA primer-template, directly analogous to the assays with DNA polymerases (Castro et al., 2005; Anand and Patel, 2006; Te Velthuis et al., 2009). In other cases, however, this is not so easily accomplished. Many RNA polymerases produce single-stranded RNA as the product, and one cannot easily set up a RNA primer-template system to look at addition of single nucleotide. In this case, it may still be possible to obtain detailed kinetic data using a partitioning analysis (Kuchta et al., 1992; Choi et al., 1996; Ramirez-Aguilar et al., 2005;
Urban et al., 2009). As depicted in Figure 8, the RNA polymerase initiates synthesis de novo and polymerizes NTPs until it reaches the site X, thereby generating RNAn . How one measures the polymerization efficiency of the NTP (analogue) opposite X will depend on what happens when the RNA polymerase encounters X. If the polymerase always terminates RNA synthesis upon producing RNAn , one can simply include varying concentrations of the analogue and measure the fraction of RNAn converted to RNAn+1 (Polymerization of YTP in Fig. 8), analogous to the running start methodology described above. Alternatively, the RNA polymerase may polymerize one of the NTPs in the assay opposite X. Provided that one can differentiate the products due to incorporation of YTP from those due to incorporation of the NTP, one can measure the frequency with which the RNA polymerase incorporates the analogue as a function of analogue concentration. Comparing different NTP (analogues) under otherwise identical experimental conditions then allows one to quantify the effects of making a specific change (Urban et al., 2009). Particularly for short RNA products up to around 6 nucleotides long, high percentage acrylamide gels (30% to 40%) will generally differentiate whether the enzyme incorporated a natural or analogue nucleotide (Moore et al., 2004; Ramirez-Aguilar et al., 2005). The high acrylamide percentage precludes including urea in the gel. However, because the products are so short and the gelloading buffer contains EDTA to chelate any divalent metals, the products will not remain bound to the template. It should also be noted that these high-percentage gels require much longer times to run, often up to 15 hr.
further NTP polymerization
NTPs Promoter promoter ...CCTCTGGTCTCCGGXCCCG
GGAGACCAG YTP AGGCC ...CCTCTGGTCTCCGGXCCCG
dissociation
GGAGACCAGAGGCC
Figure 8
GGAGACCAG AGGCCY ...CCTCTGGTCTCCGGXCCCG
dissociation
GGAGACCAGAGGCCY
Partitioning analysis of NTP polymerization by a RNA polymerase.
118 Volume 2
Current Protocols in Chemical Biology
A second complication with many RNA polymerases, especially those involved in transcription, is that they have two modes of synthesis. During polymerization of the first few NTPs, the polymerases typically have low processivity and frequently abort synthesis (Hieb et al., 2006; Wang et al., 2007; Hatoum and Roberts, 2008). However, once they undergo “promoter escape,” synthesis becomes highly processive. Thus, it may be important to consider the effects of the nucleotide analogue during both phases of transcription.
DATA INTERPRETATION Perhaps the biggest issue facing the use of analogues is that of data interpretation. A number of factors have conspired to make data interpretation particularly problematic, and each will be discussed in turn. Kinetic parameters. As noted earlier, the kinetic parameter of greatest interest is the specificity constant, Vmax /KM or kpol /KD (Apparent) . Comparing two dNTP (analogues) shows how a specific modification affects the efficiency of polymerization. The individual components of the specificity parameter, kcat or Vmax , and KD or KM, are notoriously difficult to rigorously interpret without experiments showing what step(s) each constant represents. Different polymerases. Extending conclusions from one polymerase to other polymerases is extremely dangerous, especially if they are members of different evolutionary families. Polymerases form a number of evolutionarily families, including the A, B, C, X, Y, reverse transcriptase, bacterial primase, and eukaryotic/archael primase families (Burgers et al., 2001; Kuchta and Stengel, 2009). Importantly, within these families, several distinct mechanisms by which the enzymes interact with the templating base and incoming (d)NTP exist, and different families do not use the same mechanisms for deciding whether or not to polymerize a (d)NTP. For example, A family DNA polymerases generally efficiently and accurately polymerize 2,4-difluorotoluene dNTP opposite a templating A (Moran et al., 1997; Morales and Kool, 1998, 2000). In contrast, B family polymerases generally do not efficiently incorporate 2,4-difluorotoluene dNTP (Morales and Kool, 2000). Thus, it is completely inappropriate to conclude that what one observes with, for example, the A family enzyme Klenow Fragment will also be true for the B family enzyme T4 DNA polymerase and vice versa. In some cases, even different members of the same evolutionary family may have
different requirements. For example, most B family polymerase very efficiently add the next correct dNTP onto a primer containing either 3-deaza-dA or 3-deaza-dG at the primer terminus (DNAn + dNTP → DNAn+1 ), but only extremely poorly add the second correct dNTP (DNAn+1 + dNTP → DNAn+2 (Hendrickson et al., 2004; Beckman et al., 2007; Cavanaugh et al., 2009). In contrast, the B family polymerase Tli pol efficiently adds both the first and second dNTPs (Hendrickson et al., 2004). Identifying critical chemical features in the nucleotide. One of the most daunting tasks of using nucleotide analogues is deconvoluting the data to identify the critical parameters. A simple inspection of a base reveals a number of potential parameters that a polymerase might recognize, including hydrophobicity, dipole moment, size, Watson-Crick hydrogen bonding capacity, hydrogen bonding capacity of other groups in the major and minor grooves (Major groove: N7 and N6 /O6 of a purine, N4 of a pyrimidine; minor groove, N-3 of a purine and O2 of a pyridine), and π electron density of the rings. Importantly, the electronic features of all of the atoms are interconnected due to the aromaticity of the bases. Thus, changing any one atom will result in electronic effects that reverberate throughout the structure. For example, adding a methyl group versus a halogen at C-2 of a purine will have very different effects since the methyl group is slightly electron donating whereas the halogens are strongly electron withdrawing. These changes in electronic character can have major effects on the ability of the various heteroatoms in the ring system to participate in hydrogen bonds. Thus, for example, if one converts dATP into 2-chloro-dATP and the change affects the efficiency of incorporation, the effects can be extraordinarily hard to interpret. Minimally, any effects of this simple change could be due to: (1) steric effects, since Cl is much larger than H; (2) altered Watson-Crick hydrogen bonding, since the electron withdrawing effect of Cl will interfere with the ability of N-1 to form a Watson-Crick hydrogen bond with N-3 of uridine, and; (3) altered minor groove hydrogen bonding due to the electron withdrawing effects of the Cl on N-3. How, then, does one try to isolate the important parameters affected by the chlorine that are sensed by the polymerase? Ultimately, it requires looking at multiple analogues that test the same parameter, as well as looking at the same modification within the context of multiple, similar bases. For example, if one
Nucleotide Analogues as Probes for Polymerases
119 Current Protocols in Chemical Biology
Volume 2
suspects that the effects of the chlorine resulted from electronic effects on N-1 and consequent weaker hydrogen bonding to NH-3 of a template thymine, one could first examine how the enzyme interacts with 2-fluoro-dATP and 2-methyl-dATP. Concomitant with this, one should examine a series of analogues that explicitly test the role of Watson-Crick hydrogen bonds during dNTP incorporation (e.g., 1-deaza-dATP opposite T). Unfortunately, even after examining extensive modifications, one may be left with a less than clear picture of exactly how a polymerase functions (see below). Different chemical contexts. Especially essential is to examine the same modification in different chemical contexts. The most extensive data on the effects of substrate modifications exists for Klenow Fragment (an A family enzyme), herpes, and human DNA primase, and three B family polymerases (Pol α, herpes DNA pol, and T4 DNA pol) (Thompson and Kuchta, 1995; Morales and Kool, 1998, 2000; Ogawa et al., 2000; Kool, 2002; Matsuda et al., 2003; Washington et al., 2003; Henry et al., 2004; Meyer et al., 2004; RamirezAguilar and Kuchta, 2004; Kim et al., 2005; Kincaid et al., 2005; Zhang et al., 2005; Sintim and Kool, 2006; Berdis and McCutcheon, 2007; Hwang and Romesberg, 2008; Leconte et al., 2008; Lee et al., 2008; Cavanaugh et al., 2009; Lee and Berdis, 2009; Patro et al., 2009). From these data, it is clear that the structural context of a mutation can dramatically alter the effects of modifying the base of a dNTP or NTP in a polymerase-dependent manner. For the B family enzymes, similar modifications tend to have the similar effects within differ-
ent structural contexts (Beckman et al., 2007; Cavanaugh et al., 2009). The A family enzyme Klenow Fragment typically shows similar effects, although notable exceptions exist. For example, Romesberg and coworkers examined the effects of converting furo[2,3-c]pyridin7(6H)-one dNTP into furo[2,3-c]pyridine-7thiol dNTP and converting furo[3,2-c]pyridin4(5H)-one dNTP into furo[3,2-c]pyridine4-thiol dNTP (Fig. 9). Even though both conversions involve changing an oxygen into sulfur, the effect of this change varied by up to >3000-fold in the case of polymerization opposite a template dC (Henry et al., 2003). The primases, however, often show radically different effects of the identical change in the base but within two different contexts (Urban et al., 2009). For example, whereas removing N-3 of a templating A has minor effects on polymerization of UTP, removing O2 of a templating T or C severely impairs polymerization of ATP and GTP, respectively. Similarly, adding back N-1 to 1-deazapurine NTP greatly enhances incorporation of the resulting purine NTP, but adding back the chemically equivalent N-1 to 2-pyridone NTP greatly impairs polymerization of the resulting zubularine triphosphate. Thus, one clearly must exercise caution when interpreting the effects of a modification and extending the modification from one structural context to another. In addition to structural context, whether the modification resides within the templating base or the incoming dNTP can dramatically affect the results. Indeed, this asymmetry has greatly impeded the development of novel base pairs. Very frequently, a polymerase will generate a novel base pair by efficiently
O
N
O
Furo[2,3-c]pyridin7(6H)-one O
N
Nucleotide Analogues as Probes for Polymerases
O
N
S
Furo[2,3-c]pyridine7-thiol O
O
Furo[2,3-c]pyridin4(5H)-one
N
S
Furo[2,3-c]pyridine4-thiol
Figure 9 Two chemically similar interconversions that have dramatically different effects on incorporation of the resulting dNTP by Klenow fragment.
120 Volume 2
Current Protocols in Chemical Biology
polymerizing dXTP opposite a templating Y, but does not efficiently polymerize dYTP opposite a templating X (for examples, see Piccirilli et al. 1990; Lutz et al., 1996; Chiaramonte et al., 2003; and Henry et al., 2003).
Crick hydrogen bonds in order to polymerize a NTP. However, examination of a more extensive series of base analogues showed that while primase usually only efficiently incorporates NTPs whose bases can form Watson-Crick hydrogen bonds, several remarkable exceptions occur (Urban et al., 2009). Most notably, herpes primase polymerizes 2-pyridone NTP and 4-methyl-2-pyridone NTP very efficiently across from the natural template bases, especially G, thereby indicating that primase does not absolutely require formation of WatsonCrick hydrogen bonds between the incoming NTP and templating base to efficiently polymerize the NTP. Klenow fragment. Klenow fragment, an A family DNA polymerase, has probably been probed with more nucleotide analogues than any other enzyme. A series of studies using 2,4-dihalotoluene dNTPs and 2,4difluorotoluene templating bases, a group of analogues isosteric with T, showed that Klenow fragment very efficiently generated base-pairs between these analogues and A but did not generate base-pairs with the other
EXAMPLES OF DIFFICULTY INTERPRETING HOW A POLYMERASE INTERACTS WITH A NUCLEOTIDE As suggested in this discussion, extreme care must be used when interpreting data as to how a polymerase interacts with a nucleotide. Two examples of this difficulty are discussed—the role of Watson-Crick hydrogen bonds with herpes primase and the role of shape with Klenow Fragment. Herpes primase. From initial studies using 15 purine NTP analogues, herpes primase only efficiently incorporated those that could form Watson-Crick hydrogen bonds involving the purine N-1 and N6 /O6 and the pyrimidine N-1 and N4 /O4 , respectively (Ramirez-Aguilar et al., 2005). This led to the hypothesis that herpes primase requires the formation of Watson-
O O
N
H2N
N
N
5-O-Methylbenzimidazole
N
NH2
O N⫹
N
N
HN
O⫺
N
N
N
N
N
5-Nitrobenzimidazole O
O N
HN N
H2N
S
N
N
Furo[2,3-c]pyridine7-thiol O O
N
HN N S
H2N
N
N
Furo[2,3-c]pyridine4-thiol
Figure 10
Unusually shaped base pairs that Klenow fragment generates with high efficiency.
Nucleotide Analogues as Probes for Polymerases
121 Current Protocols in Chemical Biology
Volume 2
3 natural bases (Moran et al., 1997; Morales and Kool, 1998, 2000; Sintim and Kool, 2006). Furthermore, altering the shape of the 2,4dihalotoluene so that it no longer resembled T eliminated the preference for generation of base-pairs with A. These data led to the idea that Klenow fragment discriminates between right and wrong dNTPs based on the shape of the base pair between the incoming dNTP and templating base. However, Klenow fragment will also very efficiently generate some very mis-shapen base pairs (e.g., 5-O-Methylbenzimidazole dNTP:G, 5Nitroindole dNTP:A, Furo[2,3-c]pyridine-7thiol dNTP:G, Furo[3,2-c]pyridine-4-thiol; Fig. 10) (Chiaramonte et al., 2003; Henry et al., 2003; Kincaid et al., 2005). These data raise several important questions relating to polymerase mechanisms and using nucleotide analogues. How many analogues should one examine to feel confident in ones conclusions? While the bulk of the data with both herpes primase and Klenow fragment clearly support key roles for Watson-Crick hydrogen bonds and shape, respectively, as key parameters for efficient (d)NTP incorporation by each enzyme, any model posited should ideally accommodate all of the data. How does one account for these outliers within the context of these models and the other analogues? Do these results reflect an unexplored property of the outliers, or are these results idiosyncratic to the polymerase of interest? Regardless, of the answers to these questions, it is clear that data should be interpreted cautiously.
ACKNOWLEDGEMENTS This work was supported by NIH grants GM54194 and AI59764 to R.D.K.
LITERATURE CITED Anand, V.S. and Patel, S.S. 2006. Transient state kinetics of transcription elongation by T7 RNA polymerase. J. Biol. Chem. 281:35677-35685. Beckman, J., Kincaid, K., Hocek, M., Spratt, T., Engels, J., Cosstick, R., and Kuchta, R.D. 2007. Human DNA polymerase alpha uses a combination of positive and negative selectivity to polymerize purine dNTPs with high fidelity. Biochemistry 46:448-460.
Nucleotide Analogues as Probes for Polymerases
Bell, J.B., Eckert, K.A., Joyce, C.M., and Kunkel, T.A. 1997. Base miscoding and strand misalignment errors by mutator Klenow polymerases with amino acid substitutions at tyrosine 766 in the O-helix of the Fingers subdomain. J. Biol. Chem. 272:7345-7351. Berdis, A.J. and McCutcheon, D. 2007. The use of non-natural nucleotides to probe template-
independent DNA synthesis. Chembiochem 8:1399-1408. Boosalis, M.S., Petruska, J., and Goodman, M.F. 1987. DNA polymerase insertion fidelity: Gel assay for site-specific kinetics. J. Biol. Chem. 262:14689-14696. Burgers, P.M., Koonin, E.V., Bruford, E., Blanco, L., Burtis, K.C., Christman, M.F., Copeland, W.C., Friedberg, E.C., Hanaoka, F., Hinkle, D.C., Lawrence, C.W., Nakanishi, M., Ohmori, H., Prakash, L., Prakash, S., Reynaud, C.A., Sugino, A., Todo, T., Wang, Z., Weill, J.C., and Woodgate, R. 2001. Eukaryotic DNA polymerases: Proposal for a revised nomenclature. J. Biol. Chem. 276:43487-43490. Burgess, K. and Cook, D. 2000. Syntheses of nucleoside triphosphates. Chem. Rev. 100: 20472060. Burton, J.R. and Everson, G.T. 2009. HCV NS5B polymerase inhibitors. Curr. Liver Dis. 13:453465. Cai, H., Bloom, L.B., Eritja, R., and Goodman, M.F. 1993. Kinetics of deoxyribonucleotide insertion and extension at abasic template lesions in different sequence contexts using HIV-1 reverse transcriptase. J. Biol. Chem. 268:23567-23572. Castro, C., Arnold, J.J., and Cameron, C.E. 2005. Incorporation fidelity of the viral RNAdependent RNA polymerase: A kinetic, thermodynamic and structural perspective. Virus Res. 107:141-149. Cavanaugh, N.A., Urban, M., Beckman, J., Spratt, T., and Kuchta, R.D. 2009. Identifying the features of purine dNTPs that allow accurate and efficient DNA synthesis by herpes simplex virus 1 DNA polymerase. Biochemistry 48:35543564. Chiaramonte, M., Moore, C.L., Kincaid, K., and Kuchta, R.D. 2003. Facile polymerization of dNTPs bearing unnatural base analogues by DNA polymerase alpha and Klenow fragment (DNA polymerase I). Biochemistry 42:1047210481. Choi, D.J., Roth, R.B., Liu, T., Geacintov, N.E., and Scicchitano, D.A. 1996. Incorrect base insertion and prematurely terminated transcripts during T7 RNA polymerase Transcription elongation past benzo[a]pyrenediol epoxide-modified DNA. J. Mol. Biol. 264:123-219. Dahlberg, M.E. and Benkovic, S.J. 1991. Kinetic mechanism of DNA polymerase I (Klenow fragment): Identification of a second conformational change and evaluation of the internal equilibrium constant. Biochemistry 30:4835-4843. Donlin, M.J., Patell, S.S., and Johnson, K.A. 1991. Kinetic partitioning between the exonuclease and polymerase sites in DNA error correction. Biochemistry 30:538-547. Hatoum, A. and Roberts, J. 2008. Prevalence of RNA polymerase stalling at Escherichia coli promoters after open complex formation. Mol. Microbiol. 68:17-28. Hendrickson, C.L., Devine, K.G., and Benner, S.A. 2004. Probing minor groove recognition
122 Volume 2
Current Protocols in Chemical Biology
contacts by DNA polymerases and reverse transcriptases using 3-deaza-2’-deoxyadenosine. Nucl. Acids Res. 32:2241-2250. Henry, A.A., Yu, C., and Romesberg, F.E. 2003. Determinants of unnatural nucleobase stability and polymerase recognition. J. Am. Chem. Soc. 125:9638-9646. Henry, A.A., Olsen, A.G., Matsuda, S., Yu, C., Geierstanger, B.H., and Romesberg, F.E. 2004. Efforts to expand the genetic alphabet: Identification of a replicable unnatural DNA self-pair. J. Am. Chem. Soc. 126:6923-6931. Hieb, A.R., Baran, S., Goodrich, J.A., and Kugel, J.F. 2006. An 8 Nt. RNA triggers a rate-limiting shift of RNA polymerase II complexes into elongation. EMBO J. 25:3100-3109. Hirao, I., Kimoto, M., Mitsui, T., Fujiwara, T., Kawai, R., Sato, A., Harada, Y., and Yokoyama, S. 2006. An unnatural hydrophobic base pair system: Site-specific incorporation of nucleotide analogs into DNA and RNA. Nat. Methods 3:729-735. Hirao, I., Mitsui, T., Kimoto, M., and Yokoyama, S. 2007. An efficient unnatural base pair for PCR amplification. J. Am. Chem. Soc. 129:1554915555. Horwitz, J.P., Chua, J., Noel, M., and DaRooge, M.A. 1964. Nucleosides. IV. 1-(2-deoxy-betaD-lyxofuranosyl)-5-iodouracil. J. Med. Chem. 7:385-386. Hsieh, J.C., Zinnen, S., and Modrich, P. 1993. Kinetic mechanism of the DNA-dependent DNA polymerase activity of human immunodeficiency virus reverse transcriptase. J. Biol. Chem. 268:24607-24613. Huang, P., Chubb, S., Hertel, L.W., Grindey, G.B., and Plunkett, W. 1991. Action of 2 ,2 difluorocytidine on DNA synthesis. Canc. Res. 51:6110-6117. Hwang, G.T. and Romesberg, F.E. 2008. Unnatural substrate repertoire of A, B, and X family DNA polymerases. J. Am. Chem. Soc. 130:1487214882. Kim, T.W., Delaney, J.C., Essigmann, J.M., and Kool, E.T. 2005. Probing the active site tightness of DNA polymerase in subangstrom increments. Proc. Nat. Acad. Sci. U.S.A. 102:1580315808. Kincaid, K., Beckman, J., Zivkovic, A., Halcomb, R.L., Engels, J., and Kuchta, R.D. 2005. Exploration of factors driving incorporation of unnatural dNTPs into DNA by Klenow fragment (DNA polymerase I) and DNA polymerase alpha. Nucl. Acids Res. 33:2620-2628. Kool, E.T. 2002. Active site tightness and substrate fit in DNA replication. Ann. Rev. Biochem. 71:191-219. Kuchta, R.D. and Willhelm, L. 1991. Inhibition of DNA primase by 9-beta-Darabinofuranosyladenosine triphosphate. Biochemistry 30:797-803. Kuchta, R.D., and Stengel, G. 2009. Mechanism and evolution of DNA primases. Biochim. Biophys. Acta. In press.
Kuchta, R.D., Mizrahi, V., Benkovic, P.A., Johnson, K.A., and Benkovic, S.J. 1987. Kinetic mechanism of DNA polymerase I (Klenow). Biochemistry 26:8410-8417. Kuchta, R.D., Benkovic, P., and Benkovic, S.J. 1988. Kinetic mechanism whereby DNA polymerase I (Klenow) replicates DNA with high fidelity. Biochemistry 27:6716-6725. Kuchta, R.D., Ilsley, D., Kravig, K.D., Schubert, S., and Harris, B. 1992. Inhibition of DNA primase and polymerase alpha by arabinofuranosylnucleoside triphosphates and related compounds. Biochemistry 31:4720-4728. Kunkel, T.A. 1985. The mutational specificity of DNA polymerase beta during in vitro DNA synthesis. Production of frameshift, base substitution, and deletion mutations. J. Biol. Chem. 260:5787-5796. Kunkel, T.A. and Alexander, P.S. 1986. The base substitution fidelity of eukaryotic DNA polymerases. Mispairing frequencies, site preferences, insertion preferences, and base substitution by dislocation. J. Biol. Chem. 261:160166. Lai, M.D. and Beattie, K.L. 1988. Influence of DNA sequence on the nature of mispairing during DNA synthesis. Biochemistry 27:1722-1728. Leconte, A.M., Hwang, G.T., Matsuda, S., Capek, P., Hari, Y., and Romesberg, F.E. 2008. Discovery, characterization, and optimization of an unnatural base pair for expansion of the genetic alphabet. J. Am. Chem. Soc. 130:2336-2343. Lee, I. and Berdis, A.J. 2009. Non-natural nucleotides as probes for the mechanism and fidelity of DNA polymerases. Biochim. Biophys. Acta. Epub ahead of print. Lee, J.R., Helquist, S.A., Kool, E.T., and Johnson, K.A. 2008. Importance of hydrogen bonding for efficiency and specificity of the human mitochondrial DNA polymerase. J. Biol. Chem. 283:14402-14410. Lin, T.S. and Mancini, W.R. 1983. Synthesis and antineoplastic activity of 3 -azido and 3 -amino analogs of pyrimidine deoxyribonuclosides. J. Med. Chem. 26:544-588. Ludwig, J. 1981. A new route to nucleoside 5 triphosphates. Acta Biochim. Biophys. Acad. Sci. Hung. 16:131-135. Lutz, M.J., Held, H.A., Hottinger, M., Hubscher, U., and Benner, S.A. 1996. Differential discrimination of DNA polymerases for variants of the nonstandard nucleobase pair between xanthosine and 2,4-diaminopyrimidine, two components of an expanded genetic alphabet. Nucl. Acids Res. 24:1308-1313. Matsuda, S., Henry, A.A., Schultz, S.S., and Romesberg, F.E. 2003. The effects of minorgroove hydrogen-bond acceptors and donors on the stability and replication of four unnatural base pairs. J. Am. Chem. Soc. 125:6135-6139. Meyer, A.S., Blandino, M., and Spratt, T. 2004. E. coli DNA polymerase I (Klenow fragment) uses a hydrogen-bonding fork from Arg668 to the primer terminus and incoming dNTP to catalyze
Nucleotide Analogues as Probes for Polymerases
123 Current Protocols in Chemical Biology
Volume 2
DNA replication. J. Biol. Chem. 279:3304333046. Moore, C.L., Zivkovic, A., Engels, J., and Kuchta, R.D. 2004. Human DNA primase uses Watson-Crick hydrogen bonding groups to distinguish between correct and incorrect NTPs. Biochemistry 43:12367-12374. Morales, J.C. and Kool, E.T. 1998. Efficient replication between non-hydrogen bonded nucleoside shape analogs. Nat. Struct. Biol. 5:950-954. Morales, J.C. and Kool, E.T. 2000. Varied molecular interactions at the active sites of several DNA polymerases: Nonpolar nucleoside isosteres as probes. J. Am. Chem. Soc. 122:1001-1007. Moran, S., Ren, R.X.-F., Rumney, S., and Kool, E.T. 1997. Difluorotoluene, a nonpolar isostere of thymine, codes specifically and efficiently for adenine in DNA replication. J. Am. Chem. Soc. 119:2056-2057. Ogawa, A.K., Wu, Y., McMinn, D.L., Liu, J., Schultz, P.G., and Romesberg, F.E. 2000. Efforts toward the expansion of the genetic alphabet: Information storage and replication with unnatural hydrophobic base pairs. J. Am. Chem. Soc. 122:3274-3287. Parker, W. 2009. Enzymology of purine and pyrimidine antimetabolites in the treatment of cancer. Chem. Rev. 109:2880-2893. Patel, S.S., Wong, I., and Johnson, K.A. 1991. Presteady state kinetic analysis of processive DNA replication including complete characterization of an exonuclease deficient mutant. Biochemistry 30:511. Patro, J.N., Urban, M., and Kuchta, R.D. 2009. Role of the 2-amino group of purines during dNTP polymerization by human DNA polymerase alpha. Biochemistry 48:180-189. Piccirilli, J.A., Krauch, T., Moroney, S.E., and Benner, S.A. 1990. Enzymatic incorporation of a new base pair into DNA and RNA extends the genetic alphabet. Nature 343:33-37. Plunkett, W., Huang, P., Searcy, C.E., and Gandhi, V. 1996. Gemcitabine: Preclinical pharmacology and mechanisms of action. Semin. Oncol. 56:3-15. Ramirez-Aguilar, K.A. and Kuchta, R.D. 2004. Herpes simplex virus 1 DNA primase: A polymerase with extraordinarily low fidelity. Biochemistry 43:9084-9091. Ramirez-Aguilar, K.A., Moore, C.L., and Kuchta, R.D. 2005. Herpes simplex virus I primase employs Watson-Crick hydrogen bonding to identify cognate NTPs. Biochemistry 44:1558515593. Reardon, J.E. 1989. Herpes simplex virus type 1 DNA polymerase. Mechanism of inhibition by acyclovir triphosphate. J. Biol. Chem. 264:74057411. Nucleotide Analogues as Probes for Polymerases
Reardon, J.E. 1990. Herpes simplex virus type I DNA polymerase: Mechanism-based affinity chromatography. J. Biol. Chem. 265:7112. Segel, I.H. 1975. Enzyme Kinetics. John Wiley and Sons, New York.
Showalter, A.K. and Tsai, M.-D. 2002. A reexamination of the nucleotide incorporation fidelity of DNA polymerases. Biochemistry 41:1057110576. Sintim, H.O. and Kool, E.T. 2006. Remarkable sensitivity to DNA base shape in the DNA polymerase active site. Angew. Chem. Int. Ed. 45:1974-1979. Tchesnokov, E.P., Obikhod, A., Schinazi, R.B., and Gotte, M. 2008. Delayed chain termination protects the anti-hepatitis B virus drug entecavir from excision by HIV-1 reverse transcriptase. J. Biol. Chem. 283:34218-34228. Te Velthuis, A.J., Arnold, J.J., Cameron, C.E., van de Worm, S.J., and Snijder, E.J. 2010. The RNA polymerase activity of SARS-coronavirus nsp12 is primer dependent. Nucleic Acids Res. 38:203214. Thompson, H. and Kuchta, R.D. 1995. Arabinofuranosyl nucleotides are not chain-terminators during initiation of new strands of DNA by DNA polymerase alpha-primase. Biochemistry 34:11198-11203. Tsai, Y.-C. and Johnson, K.A. 2006. A new paradigm for DNA polymerase specificity. Biochemistry 45:9675-9687. Urban, M., Joubert, N., Hocek, M., Alexander, R.E., and Kuchta, R.D. 2009. Herpes simplex virus1 DNA primase: A remarkably inaccurate yet selective polymerase. Biochemistry 48:1086610881. Villarreal, E.C. 2001. Current and potential therapies for the treatment of herpes virus infections. Progr. Drug Res. 185-228. Vivet-Boudou, V., Didierjean, J., Isel, C., and Marquet, R. 2006. Nucleoside and nucleotide inhibitors of HIV-1 replication. Cell. Mol. Life Sci. 63:163-186. Wang, Q., Tullius, T.D., and Levin, J.R. 2007. Effects of discontinuities in the DNA template on abortive initiation and promoter escape by Escherichia coli RNA polymerase. J. Biol. Chem. 282:26917-26927. Washington, M.T., Helquist, S.A., Kool, E.T., Prakash, L., and Prakash, S. 2003. Requirement of Watson-Crick hydrogen bonding for DNA synthesis by yeast DNA polymerase eta. Mol. Cell. Biol. 23:5107-5112. Wong, I., Patel, S.S., and Johnson, K.A. 1991. An induced fit kinetic mechanism for DNA replication fidelity: Direct measurement by single turnover kinetics. Biochemistry 30:526537. Wu, W., Bergstrom, D.E., and Jo Davisson, V. 2004. Chemoenzymatic preparation of nucleoside triphoshates. Curr. Protoc. Nucleic Acid Chem. 16: 13.2.1-13.2.19. Zhang, X., Lee, I., and Berdis, A.J. 2005. The use of nonnatural nucleotides to probe the contribution of shape complementarity and pi electron surface area during DNA polymerization. Biochemistry 44:13101-13110.
124 Volume 2
Current Protocols in Chemical Biology
Site-Specific Protein Bioconjugation via a Pyridoxal 5 -Phosphate-Mediated N-Terminal Transamination Reaction Leah S. Witus1 and Matthew Francis1 1
University of California, Berkeley, Department of Chemistry, Berkeley, California
ABSTRACT The covalent attachment of chemical groups to proteins is a critically important tool for the study of protein function and the creation of protein-based materials. Methods of site-specific protein modification are necessary for the generation of well defined bioconjugates possessing a new functional group in a single position in the amino acid sequence. This article describes a pyridoxal 5 -phosphate (PLP)–mediated transamination reaction that is specific for the N-terminus of a protein. The reaction oxidizes the Nterminal amine to a ketone or an aldehyde, which can form a stable oxime linkage with an alkoxyamine reagent of choice. Screening studies have identified the most reactive Nterminal residues, facilitating the use of site-directed mutagenesis to achieve high levels of conversion. Additionally, this reaction has been shown to be effective for a number of targets that are not easily accessed through heterologous expression, such as monoclonal C 2010 by John Wiley & Sons, Inc. antibodies. Curr. Protoc. Chem. Biol. 2:125-134 Keywords: bioconjugation r N-terminus r pyridoxal phosphate r PLP r oximation
INTRODUCTION The generation of a bioconjugate that is modified in a single predicted site on the surface of a protein is a difficult task, but one that is often crucial for the success of a given application. Because it is frequently difficult to target a unique instance of a particular amino acid side chain, this protocol instead describes an N-terminal transamination reaction effected by the common biological cofactor pyridoxal-5 -phosphate (PLP; Gilmore et al., 2006). As shown in Figure 1, incubation with PLP converts the N-terminus of a protein into a ketone or an aldehyde group. This newly installed functionality can be further derivatized through the formation of a stable oxime bond using a diverse array of alkoxyamine probes. This bioconjugation scheme has been used for the site-specific modification of proteins with fluorophores (Scheck and Francis, 2007), polymer chains (Esser-Kahn and Francis, 2008), polymer initiators (Heredia and Maynard, 2007; Gao et al., 2009), and surfaces (Christman et al., 2007; Lempens et al., 2009). Due to its specificity for the N-terminal amino group, the PLP-mediated transamination reaction can be used in conjunction with other bioconjugation techniques, such as cysteine modification or expressed protein ligation, for dual protein modification (Esser-Kahn and Francis, 2008; Dedeo et al., 2010). PLP-mediated bioconjugation proceeds at mild pH and temperature, and is tolerant of a number of buffers and conditions. Peptide screening studies have shown that this reaction proceeds with high yield for a number of N-terminal residues (Ala, Gly, Asn, Glu, and Asp, in particular), and to a lesser degree for others (Scheck et al., 2008). As such, the reaction has often proven successful for native protein sequences and, in cases where it is not, standard molecular biology techniques can typically be used to change the N-terminal sequence to obtain increased levels of conversion (Scheck et al., 2008). PLP-Mediated Site-Specific Protein Bioconjugation Current Protocols in Chemical Biology 2: 125-134, June 2010 Published online June 2010 in Wiley Interscience (www.interscience.wiley.com). DOI: 10.1002/9780470559277.ch100018 C 2010 John Wiley & Sons, Inc. Copyright
125 Volume 2
A
B O
H
HO
R
H N
H2 N O
⫹
N H pH 6.5 37⬚C 2-24 hr
OPO3H⫺
O
O
H N
R
O NH2
N
H N
R O
O
“keto-protein”
oxime-linked protein bioconjugate
Figure 1 Reaction scheme. (A) In the first step, the protein is incubated with PLP, which transaminates the N-terminus to form a ketone or an aldehyde. (B) In the second step, the keto-protein reacts with an alkoxyamine probe to form an oxime-linked protein bioconjugate. The amine groups of lysine residues may reversibly form imines with PLP, but they do not proceed through the transamination process. Thus, the protein is modified in a single, specific location using this procedure.
STRATEGIC PLANNING A wide variety of proteins with both known and unknown sequences have been successfully modified using PLP-mediated bioconjugation. Therefore, a new target should initially be evaluated empirically. If the protein is recombinantly expressed, its suitability for N-terminal modification can often be improved using the methodology described here. First, the identity and accessibility of the N-terminal residues should be taken into consideration. In general, buried and/or highly hydrophobic N-terminal sequences react poorly, most likely because they do not participate in imine formation with PLP. In these cases, installation of one to three “spacer” residues can lead to enhanced reactivity (for a specific example, see Scheck et al., 2008). The reaction is also not appropriate for proteins (such as actin) that have been acylated at the N-terminal amino group through post-translational modifications.
PLP-Mediated Site-Specific Protein Bioconjugation
The influence of the identity of the N-terminal residue on the reaction outcome has been illuminated by recent studies (Scheck et al., 2008). As shown in Figure 2, a library of tetrapeptides including all 20 natural amino acids as the N-terminal residue was exposed to the transamination/oxime formation sequence under identical reaction conditions (10 mM PLP, 18 hr, room temperature). The resulting products included the desired oxime product, keto-peptides that formed but did not participate in subsequent oxime formation, and covalent adducts of PLP to the N-terminal group. As shown in the figure, the N-terminal residues can be grouped by yield into the categories of high conversion, intermediate conversion, or byproduct formation. The high conversion category includes alanine, glycine, aspartate, glutamate, and asparagine, which were found to result in the highest yields of oxime product. Cysteine, arginine, threonine, tyrosine, leucine, serine, methionine, phenylalanine, and valine all experienced intermediate levels of conversion. The byproduct formation category includes various species of byproducts arising from N-terminal glutamine, histidine, tryptophan, lysine, and proline residues. Glutamine was found to transaminate to produce a ketone, but this structure was resistant to further oxime formation. Tryptophan and histidine undergo a previously reported Pictet-Spengler reaction (Li et al., 2000). Lysine and proline each form unique types of PLP adducts, and in these cases the formation of the covalent PLP adducts does not preclude oxime formation (although they have been found to be slightly less reactive than the keto groups formed on the N-termini). The lysine-terminal PLP adduct presumably
126 Volume 2
Current Protocols in Chemical Biology
R
10 mM PLP 18 hr room temp
H N
H2N
peptide
O
H N
R
O
O N
250 mM BnONH2 peptide
H N
peptide O oxime product
R
O keto-protein
100
% oxime product % keto-protein % PLP adduct
80
60
40
20
0 XKWA
A
G
D
E
high conversion
N
C
R
T
Y
L
S
M
intermediate conversion
F
V
I
Q
H
K
W
P
byproduct formation
Figure 2 Peptide studies on tetrapeptide XKWA, where X includes all twenty natural amino acids, show the role of the N-terminal residue on reactivity toward the PLP-mediated transamination reaction. The products include the desired benzyloxime product (shown in blue), the species that has undergone the transamination reaction to the keto-peptide species but did not continue to form an oxime (shown in green), and various species resulting from covalent PLP addition to the N-terminus (all shown as orange). The N-terminal residues are grouped into the categories of high conversion, intermediate conversion, and byproduct formation.
forms from transamination followed by the loss of water to form a cyclic imine. The nucleophilic enamine tautomer of this structure undergoes aldol addition to additional molecules of PLP, to form a covalent adduct. The proline-terminal adduct observed involves the covalent attachment of PLP through an as-yet unknown pathway. The intended application and empirical analysis of new target proteins will dictate the need for mutagenesis. In many cases, sufficient reactivity can be achieved using the native terminus, and in some cases lower levels of conversion are tolerable for the application (e.g., surface attachment or radiolabeling). In other cases, standard molecular biology protocols can be used to extend and change the N-terminus of proteins that terminate with intermediate conversion or byproduct formation residues. As noted above, extending the N-terminus by one to three residues at the same time as converting the N-terminal residue to one of high reactivity may improve the outcome by increasing the accessibility of the N-terminus. Similarly, if a protein that already terminates in a high conversion residue is found to have low yield, mutagenesis to extend the N-terminus and increase accessibility may help to improve the reactivity. If the location of the modification on the protein is crucial, circular permutation has been used in one case to change the location of the N-terminus (Dedeo et al., 2010). In this technique, the gene is reorganized such that new N- and C-termini are created in a desired location (Regan, 1999). It has been found that the second and third residues can also influence the reactivity of the N-terminal residue, and studies are underway to clarify these synergistic effects. PLP-Mediated Site-Specific Protein Bioconjugation
127 Current Protocols in Chemical Biology
Volume 2
BASIC PROTOCOL
PROTEIN LABELING VIA A SITE-SPECIFIC N-TERMINAL TRANSAMINATION REACTION This protocol divides the process of conjugating a chemical group of interest to the Nterminus of a protein into two stages. The first step is the PLP-mediated transamination, which installs a reactive keto group that can be used for further derivitization. The second step of the protocol describes typical conditions for the reaction of this site with a smallmolecule alkoxyamine to form an oxime linkage, which is stable under physiological conditions. A similar strategy may be used to attach hydrazide-functionalized compounds to form a hydrazone linkage, although these groups are more labile (Kalia and Raines, 2008). The analysis technique used to verify the modification will vary according to the chemical moiety being conjugated to the protein: the attachment of small molecules can be confirmed using mass spectrometry, while the attachment of fluorescent or highmolecular-weight compounds can be detected by UV-vis spectroscopy, HPLC, or SDSPAGE. If the bioconjugation probe is in short supply or is difficult to detect using these analytical techniques, a proxy alkoxyamine reagent (such as commercially available benzyloxyamine or aminooxyacetic acid) can be used during the optimization of the transamination segment of the protocol.
Materials Target protein stock solution (see recipe) PLP stock solution (see recipe) Alkoxyamine solution (see recipe) Additional reagents and equipment for protein separations and analysis of protein modification, including ESI-MS or MALDI-TOF MS methods (Coligan et al., 2010) PLP-mediated N-terminal transamination 1. Combine the protein solution and the PLP stock solution in a 1.5-ml microcentrifuge tube to give a final concentration of 10 mM PLP and 10 to 500 μM protein at pH 6.5. Carefully check the pH of this solution. If the pH of the PLP stock was not adjusted (see Reagents and Solutions) this solution may be overly acidic, leading to protein precipitation or suboptimal levels of conversion.
2. Incubate at 37◦ C for 4 to 20 hr. These conditions are standard for most proteins, and can be used as a starting point for optimization with each new protein target. Factors that may be varied include the concentration of PLP, the reaction temperature, the incubation time, and the pH. The reaction may be complete in as little as an hour, typically when a higher concentration of PLP (up to 100 mM) is used. The stability of the protein may dictate the range of acceptable temperature and pH conditions. Consult the Troubleshooting section for details on how these parameters may affect reaction outcome.
3. Remove PLP using one of a variety of size-exclusion methods (such as dialysis, buffer exchange, or gel filtration; see Coligan et al., 2010), capitalizing on the difference in the mass of the protein relative to PLP. If the target protein or peptide has a low molecular weight, reversed-phase separation (also described in Coligan et al., 2010) can be used. If excess PLP is not removed, the aldehyde of the PLP molecule will quench the ketoreactive group in the next step, significantly lowering its effective concentration. PLP-Mediated Site-Specific Protein Bioconjugation
Derivatization via oxime-formation 4. Using a freshly prepared solution of keto-protein from step 3, add the alkoxyamine solution such that the alkoxyamine is in 10- to 1000-fold molar excess of the protein.
128 Volume 2
Current Protocols in Chemical Biology
Since oxime formation reactions occur at an optimal pH of 4.5 (Jencks, 1959; Carey and Sundberg, 1977), prepare the alkoxyamine solution such that the final pH will be as acidic as possible within the stability requirements of the protein. For larger proteins, it is common to carry out this reaction step between pH 5.5 and 6.5. If the protein is not stable at acidic pH, or if it is not possible to use a large excess of alkoxyamine, then aniline or anisidine catalysis of oxime formation may be used to increase the reaction rate (Dirksen et al., 2006; Dirksen and Dawson, 2008).
5. Incubate at room temperature for 18 to 24 hr. 6. Analyze conversion and remove excess alkoxyamine, if necessary. If analyzing by mass spectrometry, small amounts of PLP adduct species may be observed. For identification of common product masses, including those with PLP, see Troubleshooting.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps.
Alkoxyamine solution The alkoxyamine should be prepared as a concentrated stock solution. The storage conditions will vary with each individual reagent, but exposure to adventitious carbonyl groups, such as acetone, should be strictly avoided. Commercially available alkoxyamines include small molecules and dyes, and the nomenclature may include the names “hydroxylamine,” “aminooxy,” or “alkoxyamine.” Smallmolecule alkoxyamines that have been used to determine conversion of a protein by mass spectrometry include benzylalkoxyamine (O-benzylhydroxylamine hydrochloride) and (aminooxy)acetic acid (O-(carboxymethyl)hydroxylamine hemihydrochloride). Poly(ethylene glycol) alkoxyamines can be synthesized (Schlick et al., 2005) for measurement of protein conversion by SDS-PAGE shifts.
PLP stock solution The PLP stock solution should be prepared immediately before use, as this compound has been reported to degrade in aqueous solution (Ball, 2006). The PLP solution should be made in the same buffer in which the transamination reaction will be run, typically 10 to 50 mM phosphate buffer at pH 6.5. After the addition of PLP to the buffer, it is important to check and adjust the pH, as the phosphate group of PLP may significantly alter the pH of the buffer solution. It is often convenient to make the PLP stock solution at two times the desired concentration for the reaction, assuming that it will be added to the protein solution in a 1:1 volume ratio. For a transamination reaction run with 10 mM PLP, the following guidelines may be followed to make 1 ml of a 2× (20 mM) PLP stock solution. (1) Add 5.3 mg of pyridoxal 5 -phosphate monohydrate (Sigma) to 1 ml of 25 mM phosphate buffer, pH 6.5, followed by the addition of 24 μl of 1 M NaOH and brief sonication and vortexing. (2) Check the pH with pH paper or a microelectrode and adjust to 6.5 if necessary.
Protein stock solution The protein stock solution should be prepared in the buffer in which the transamination reaction will be run. The standard buffer is 10 to 50 mM potassium phosphate at pH 6.5; however, other buffers, such as Tris and HEPES, will also work (Gilmore et al., 2006). The presence of glycerol and other polyalcohols should be avoided, as they may form acetals with the PLP aldehyde group and thus lower its effective concentration. Note that many spin concentrators are packed in glycerol solutions, and can serve as unexpected sources of this impurity.
PLP-Mediated Site-Specific Protein Bioconjugation
129 Current Protocols in Chemical Biology
Volume 2
COMMENTARY Background Information
PLP-Mediated Site-Specific Protein Bioconjugation
Protein bioconjugation is used to study proteins in their biological context through fluorophore attachment (Griffin et al., 1998; O’Hare et al., 2007; Cravatt et al., 2008), improve their efficacy as therapeutic agents by conjugation to polymer chains (Zalipsky, 1995; Baker et al., 2006), and enable the construction of new types of protein-based materials (Wang et al., 2002; Christman et al., 2007; Esser-Kahn and Francis, 2008; Abedin et al., 2009). A number of reactions exist to modify the functional groups on amino acid side chains (Hermanson, 1996); however, the product mixtures that result from the indiscriminate modification of multiple sites on the protein surface are unsuitable for many applications. To create well defined protein bioconjugates with a better chance of preserving protein stability and function, reactions that can modify a protein a single time at a specific site are essential. Site-specific protein modification presents a considerable challenge, considering the hundreds of polar spectator groups that are present on any given protein surface. Selectivity must be achieved without the use of protective groups, and the reaction must proceed in aqueous solution under mild pH and temperature conditions. Amino acids that have a low natural occurrence on protein surfaces offer better opportunities for site-specific protein bioconjugation, as a single copy can often be introduced through genetic manipulation. Cysteine is widely used for this purpose, and can be modified using many commercially available maleimide and iodoacetamide reagents (Hermanson, 1996). In cases where a uniquely reactive cysteine cannot be introduced, reactions targeting tryptophan and tyrosine residues can serve as useful alternatives (Joshi et al., 2004; Tilley and Francis, 2006; Antos et al., 2009; Ban et al., 2010). The C-terminus of proteins can also serve as a site for selective modification through native chemical ligation. In this technique, proteins expressed or synthesized with a C-terminal thioester react chemoselectively with cysteine derivatives (Dawson et al., 1994; Muir, 2003). Additionally, a wide range of bioorthogonal chemistry is accessible by incorporating nonnatural amino acids bearing alkynes and azides for Huisgen cycloaddition chemistry (Xie and Schultz, 2006; Strable et al., 2008), or aniline amino acid derivatives for oxidative coupling
reactions (Hooker et al., 2006; Carrico et al., 2008; Tong et al., 2009). To target a single protein in a complex mixture, enzymatic labeling reactions targeting specific recognition sequences can be highly useful. This can be achieved using biotin ligase and formylglycine-generating enzymes, which have been used to install ketones and aldehydes, respectively, on their recognition sequences for subsequent reaction with hydrazine or alkoxyamine probes (Chen et al., 2005; Carrico et al., 2007). Another strategy is the direct conjugation of functionalized substrates to the recognition sequence, as seen with bacterial sortases that capitalize on the tolerance of the enzymes for non-natural oligoglycine substrates (Mao et al., 2004). Guanosine derivatives can also be used to modify specific domains of fusion proteins using the “SNAP” tagging strategy (Keppler et al., 2003). These reactions provide just a few examples of the rapidly growing set of enzymatic labeling techniques that are now available (O’Hare et al., 2007; Sletten and Bertozzi, 2009). While the above strategies apply to proteins that can be genetically manipulated and heterologously expressed, chemical reactions that target the N terminus can, in principle, be used on proteins from virtually any source. A number of chemoselective reactions exist for the modification of specific N-terminal residues, such as the sodium periodate cleavage of Nterminal serine or threonine residues, followed by derivatization of the resulting aldehyde group (Geoghegan and Stroh, 1992) or the condensation of aldehyde reagents with Nterminal tryptophan residues to form PictetSpengler products (Li et al., 2000). As a complement to these strategies for specific Ntermini, N-terminal transamination reactions can successfully convert a number of different residues into reactive keto groups. This can be accomplished using copper(II) ions and glyoxylic acid (Dixon and Fields, 1972; Dixon, 1984), or by using the PLP-mediated transamination reaction described in this protocol. The latter reaction proceeds under particularly mild reaction conditions and does not require the use of metal ions. It has been successfully used to modify a number of protein targets, including GFP (Gilmore et al., 2006), monoclonal antibodies (Scheck and Francis, 2007), viral coat proteins (Scheck et al., 2008), metallothioneins (Esser-Kahn et al., 2008), and others.
130 Volume 2
Current Protocols in Chemical Biology
eters (including the addition of alkoxyamine) are held constant. During the PLP-mediated transamination reaction, PLP aldol addition products can form from the reversible keto-enol tautomerization of the keto-protein product, as seen in Figure 3. Both species have been observed to form oximes upon exposure to the alkoxyamine reagent in the next step; however, the presence of both intermediates creates a complex product mixture. If the goal is simply ligation of the alkoxyamine reagent to the protein, the presence of the PLP adduct species may be tolerable, but if the PLP adduct is detrimental, the following changes in the reaction conditions can often be used to minimize the amount that is formed. As would be expected for a bimolecular aldol reaction, the yield of the PLP addition product increases with higher PLP concentration and longer reaction times—two parameters that also influence the transamination yield. Thus, a balance between these reaction pathways must be struck. Higher concentrations of PLP (30 to 100 mM) with shorter reaction times (1 to 6 hr) or lower concentrations of PLP (5 to 10 mM) with longer reaction times (8 to 20 hr) are conditions that will
Critical Parameters It is important that the PLP solution be prepared according to the guidelines described in the Reagents and Solutions section. Using a fresh solution of PLP, checking and adjusting the pH, and storing the solid PLP at 4◦ C until use will yield optimal results. As noted above, glycerol and other polyhydroxylated compounds should be removed before the transamination step. It should be noted that many commercially available spin concentrators are packed in glycerol solutions. Adventitious ketones (such as acetone) should be avoided when handling the alkoxyamine compounds.
Troubleshooting Regardless of N-terminal residue identity, it is possible to vary the reaction conditions for each protein target to optimize conversion. A proxy alkoxymine that is commercially available and/or readily analyzed can be used while the optimal conditions are being screened for a new target protein. See Reagents and Solutions for examples. To confirm the compatibility of a protein to the overall reaction conditions, it is often beneficial to run a control reaction that lacks PLP, but in which all of the other param-
O
H
HO H N
H 2N
R
OPO3H⫺ O
N
H N
protein
O
N
O H N
R–ONH2 protein
O
protein
O
desired keto-product additional reaction with PLP (high [PLP] and/or long reaction time) OH
OH
O
N
H N
R OH
protein
O OPO3H⫺
OH
O
N
H N
R–ONH2 N
protein
O OPO3H⫺
PLP aldol adduct
Figure 3 In addition to the desired keto-protein product, the PLP-mediated transamination reaction can also result in the formation of an N-terminal adduct with PLP. Shown here are the observed product structures for an alanine-terminal protein. Both intermediates can form oximes upon exposure to the alkoxyamine reagent in the second step. However, this results in a more complex product mixture. To maximize keto-protein yield while minimizing the aldol adduct, the recommended strategies are to use a high concentration of PLP for a short time, or a low concentration of PLP for a longer reaction time.
131 Current Protocols in Chemical Biology
Volume 2
Table 1 Troubleshooting PLP-Mediated Bioconjugation
Problem
Possible cause
Solution
Poor conversion to oxime product
Low yield of transamination reaction
Increase the concentration of PLP, screening from 10 mM to 100 mM Increase the incubation time or the temperature of the transamination reaction
Low yield of oxime formation reaction
Increase the equivalents of alkoxyamine Lower the pH during oxime-formation step Use aniline catalysis, as reported by Dirksen et al. (2006) Check quality of alkoxyamine reagent by NMR
High conversion to undesired PLP adduct
Byproduct formation
Use lower concentrations of PLP Use shorter reaction times To identify common byproducts, see the table of common mass adducts Consider mutagenesis of N-terminal residue
Nominal reactivity
Inaccessible or unreactive Consider mutagenesis N-terminal residues
Protein precipitation
PLP solution is too acidic
Check and adjust pH of PLP solution before mixing with protein
Protein insolubility
The transamination reaction can be run under buffer, pH, and temperature conditions that vary from those listed in the Basic Protocol. Alter these parameters as dictated by the stability of the target protein.
help maximize the yield of the desired oxime product while keeping PLP addition products to a minimum. The optimal strategy will depend on the specific protein, so exploring multiple sets of conditions is recommended during screening. Table 1 lists suggested strategies for overcoming potential issues encountered during PLP-mediated bioconjugation. When analyzing the mass spectrum of a PLP-modified protein, the information presented in Table 2 can be used to identify the observed species. In some cases, a species may produce more than one of these mass changes (e.g., a PLP aldol addition product may also form an oxime).
mild reaction conditions are not expected to denature most proteins, and in most cases they retain their desired activity after modification.
Anticipated Results PLP-Mediated Site-Specific Protein Bioconjugation
For many protein targets, optimization of the basic protocol can result in 50% to 90% conversion to the desired oxime product. Sitedirected mutagenesis may improve the conversion of initially unreactive target proteins. The
Time Considerations N-terminal protein modification using PLP-mediated transamination and subsequent oxime formation can be completed in 3 days, with minimal hands-on time. On the first day, the PLP transamination reaction is started and typically allowed to proceed overnight. On day 2, the excess small molecule is removed and the oxime-forming reaction is started. After overnight incubation, the conversion can be analyzed and the protein conjugate is ready for use. The incubation times of both the PLP transamination and the oxime formation may be reduced in some cases. Heating may reduce the conversion time for the transamination reaction, and an excess of alkoxyamine will increase the rate of the oxime-formation step. Considering the small amount of active time, it is quite feasible to modify multiple proteins
132 Volume 2
Current Protocols in Chemical Biology
Table 2 Identifying Observed Species in Mass Spectrometry of PLP-Modified Proteinsa
Observed mass
N-terminus
Possible productb
M−1
Any
Keto-protein
M + Alk − 19
Any
Oxime
M − 44
Asp
Decarboxylation
M − 16
Ser
Beta elimination
M − 32
Cys
Beta elimination
M + 247
Any
PLP aldol addition
M + 229
Any
PLP aldol addition with dehydration
M + 229
His, Trp
M − 19
Lys
Cyclic enamine
M + 293
Pro
Ring opening with PLP addition
Pictet-Spengler addition of PLP
a M = unmodified protein mass. Alk = alkoxyamine mass. b Note that many of these species can also form oximes.
in parallel or to screen multiple reaction conditions at once. A gradient program on a PCR thermocycler is convenient for screening conditions.
Literature Cited Abedin, M.J., Liepold, L., Suci, P., Young, M., and Douglas, T. 2009. Synthesis of a cross-linked branched polymer network in the interior of a protein cage. J. Am. Chem. Soc. 131:43464354. Antos, J.M., McFarland, J.M., Iavarone, A.T., and Francis, M.B. 2009. Chemoselective tryptophan labeling with rhodium carbenoids at mild pH. J. Am. Chem. Soc. 131:6301-6308. Baker, D.P., Lin, E.Y., Lin, K., Pellegrini, M., Petter, R.C., Chen, L.L., Arduini, R.M., Brickelmaier, M., Wen, D., Hess, D.M., Chen, L., Grant, D., Whitty, A., Gill, A., Lindner, D.J., and Pepinsky, R.B. 2006. N-terminally PEGylated human interferon-β-1a with improved pharmacokinetic properties and in vivo efficacy in a melanoma angiogenesis model. Bioconjugate Chem. 17:179-188. Ball, G.F.M. 2006. Vitamins in foods. CRC Press. New York. Ban, H., Gavrilyuk, J., and Barbas, C.F. 2010. Tyrosine bioconjugation through aqueous enetype reactions: A click-like reaction for tyrosine. J. Am. Chem. Soc. 132:1523-1525. Carey, F.A. and Sundberg, R.J. 1977. Advanced Organic Chemistry. Plenum, New York. Carrico, I.S., Carlson, B.L., and Bertozzi, C.R. 2007. Introducing genetically encoded aldehydes into proteins. Nat. Chem. Biol. 3:321322. Carrico, Z. M., Romanini, D.W., Mehl, R.A., and Francis, M.B. 2008. Oxidative coupling of peptides to a virus capsid containing unnatural amino acids. Chem. Commun. 10:1205-1207.
Chen, I., Howarth, M., Lin, W., and Ting, A.Y. 2005. Site-specific labeling of cell surface proteins with biophysical probes using biotin ligase. Nat. Methods 2:99-104. Christman, K.L., Broyer, R.M., Tolstyka, Z.P., and Maynard, H.D. 2007. Site-specific protein immobilization through N-terminal oxime linkages. J. Mater. Chem. 17:2021-2027. Coligan, J.E., Dunn, B.M., Speicher, D.W., and Wingfield, P.T. 2010. Current Protocols in Protein Science. John Wiley & Sons, Hoboken, N.J. Cravatt, B.F., Wright, A.T., and Kozarich, J.W. 2008. Activity-based protein profiling: From enzyme chemistry to proteomic chemistry. Annu. Rev. Biochem. 77:383-414. Dawson, P.E., Muir, T.W., Clark-Lewis, I., and Kent, S.B. 1994. Synthesis of proteins by native chemical ligation. Science 266:776779. Dedeo, M.T., Duderstadt, K.E., Berger, J.M., and Francis, M.B. 2010. Nanoscale protein assemblies from a circular permutant of the tobacco mosaic virus. Nano Lett. 10:181-186. Dirksen, A. and Dawson, P.E. 2008. Rapid oxime and hydrazone ligations with aromatic aldehydes for biomolecular labeling. Bioconjugate Chem. 19:2543-2548. Dirksen, A., Hackeng, T.M., and Dawson, P.E. 2006. Nucleophilic catalysis of oxime ligation. Angew. Chem. Int. Ed. 45:7581-7584. Dixon, H.B.F. 1984. N-terminal modification of proteins—A review. J. Prot. Chem. 3:99-108. Dixon, H.B.F. and Fields, R. 1972. Specific modification of NH2 -terminal residues by transamination. Methods Enzymol. 25:409-419. Esser-Kahn, A.P. and Francis, M.B. 2008. Proteincross-linked polymeric materials through siteselective bioconjugation. Angew. Chem. Int. Ed. 120:3811-3814.
PLP-Mediated Site-Specific Protein Bioconjugation
133 Current Protocols in Chemical Biology
Volume 2
Esser-Kahn, A.P., Iavarone, A.T., and Francis, M.B. 2008. Metallothionein-cross-linked hydrogels for the selective removal of heavy metals from water. J. Am. Chem. Soc. 130:1582015822. Gao, W., Liu, W., Mackay, J.A, Zalutsky, M.R., Toone, E.J., and Chilkoti, A. 2009. In situ growth of a stoichiometric PEG-like conjugate at a protein’s N-terminus with significantly improved pharmacokinetics. Proc. Natl. Acad. Sci. U.S.A. 106:15231-15236.
Li, X., Zhang, L., Hall, S.E., and Tam, J.P. 2000. A new ligation method for N-terminal tryptophancontaining peptides using the Pictet-Spengler reaction. Tetrahedron Lett. 41:4069-4073. Mao, H., Hart, S.A., Schink, A., and Pollok, B.A. 2004. Sortase-mediated protein ligation: A new method for protein engineering. J. Am. Chem. Soc. 126:2670-2671. Muir, T.W. 2003. Semisynthesis of proteins by expressed protein ligation. Annu. Rev. Biochem. 72:249-289.
Geoghegan, K.F. and Stroh, J.G. 1992. Site-directed conjugation of nonpeptide groups to peptides and proteins via periodate oxidation of a 2amino alcohol: Application to modification at N-terminal serine. Bioconjugate Chem. 3:138146.
O’Hare, H.M., Johnsson, K., and Gautier, A. 2007. Chemical probes shed light on protein function. Curr. Opin. Struct. Bio. 17:488-494.
Gilmore, J.M., Scheck, R.A., Esser-Kahn, A.P., Joshi, N.S., and Francis, M.B. 2006. N-terminal protein modification through a biomimetic transamination reaction. Angew. Chem. Int. Ed. 45:5307-5311.
Scheck, R.A. and Francis, M.B. 2007. Regioselective labeling of antibodies through N-terminal transamination. ACS Chem. Biol. 2:247-251.
Griffin, B.A., Adams, S.R., and Tsien, R.Y. 1998. Specific covalent labeling of recombinant protein molecules inside live cells. Science 281:269-272. Heredia, K.L. and Maynard, H.D. 2007. Synthesis of protein-polymer conjugates. Org. Biomol. Chem. 5:45-53. Hermanson, G.T. 1996. Bioconjugate Techniques, 1st ed. Academic Press, San Diego, Calif. Hooker, J.M., Esser-Kahn, A.P., and Francis, M.B. 2006. Modification of aniline containing proteins using an oxidative coupling strategy. J. Am. Chem. Soc. 128:15558-15559. Jencks, W.P. 1959. Studies on the mechanism of oxime and semicarbazone formation. J. Am. Chem. Soc. 81:475-481. Joshi, N.S., Whitaker, L.R., and Francis, M.B. 2004. A three-component Mannich-type reaction for selective tyrosine bioconjugation. J. Am. Chem. Soc. 126:15942-15943. Kalia, J. and Raines, R.T. 2008. Hydrolytic stability of hydrazones and oximes. Angew. Chem. Int. Ed. 47:7523-7526. Keppler, A., Gendreizig, S., Gronemeyer, T., Pick, H., Vogel, H., and Johnsson, K. 2003. A general method for the covalent labeling of fusion proteins with small molecules in vivo. Nat. Biotech. 21:86-89. Lempens, E.H.M., Helms, B.A., Merkx, M., and Meijer, E.W. 2009. Efficient and chemoselective surface immobilization of proteins by using aniline-catalyzed oxime chemistry. ChemBioChem 10:658-662.
Regan, L. 1999. Protein redesign. Curr. Opin. Struct. Biol. 9:494-499.
Scheck, R.A., Dedeo, M.T., Iavarone, A.T., and Francis, M.B. 2008. Optimization of a biomimetic transamination reaction. J. Am. Chem. Soc. 130:11762-11770. Schlick, T.L., Ding, Z., Kovacs, E.W., and Francis, M.B. 2005. Dual-surface modification of the tobacco mosaic yirus. J. Am. Chem. Soc. 127:3718-3723. Sletten, E.M. and Bertozzi, C.R. 2009. Bioorthogonal chemistry: Fishing for selectivity in a sea of functionality. Angew. Chem. Int. Ed. 48:69746998. Strable, E., Prasuhn, D.E., Udit, A.K., Brown, S, Link, A.J., Ngo, J.T., Lander, G., Quispe, J., Potter, C.S., Carragher, B., Tirrell, D.A., and Finn, M.G. 2008. Unnatural amino acid incorporation into virus-like particles. Bioconjugate Chem. 19:866-875. Tilley, S.D. and Francis, M.B. 2006. Tyrosineselective protein alkylation using πallylpalladium complexes. J. Am. Chem. Soc. 128:1080-1081. Tong, G.J., Hsiao, S.C., Carrico, Z.M., and Francis, M.B. 2009. Viral capsid DNA aptamer conjugates as multivalent cell-targeting vehicles. J. Am. Chem. Soc. 131:11174-11178. Wang, Q., Lin, T., Tang, L., Johnson, J.E., and Finn, M.G. 2002. Icosahedral virus particles as addressable nanoscale building blocks. Angew. Chem. Int. Ed. 41:459-462. Xie, J. and Schultz, P.G. 2006. A chemical toolkit for proteins—An expanded genetic code. Nat. Rev. Mol. Cell. Biol. 7:775-782. Zalipsky, S. 1995. Functionalized poly(ethylene glycols) for preparation of biologically relevant conjugates. Bioconjugate Chem. 6:150-165.
PLP-Mediated Site-Specific Protein Bioconjugation
134 Volume 2
Current Protocols in Chemical Biology
Small-Molecule Library Synthesis on Silicon-Functionalized SynPhase Lanterns Jeremy R. Duvall,1 Anita Vrcic,1 and Lisa A. Marcaurelle1 1
Chemical Biology Platform, The Broad Institute of MIT and Harvard, Cambridge, Massachusetts
ABSTRACT Silicon-functionalized SynPhase Lanterns are useful for the combinatorial synthesis of small-molecule libraries. Lanterns bearing an alkyl-tethered diisopropylarylsilane are Þrst activated with trißic acid to afford the corresponding diisopropylsilyl trißate, which is then reacted with a library scaffold bearing a free alcohol. Once the scaffold has been loaded onto the solid phase, a variety of transformations can be run, including amine cappings, cross-coupling reactions, and amide bond formation. These reactions can yield a variety of products when run sequentially using split-pool synthesis strategies. Upon completion of the solid-phase transformations, the small molecules are released from the Lanterns using HF/pyridine. Using the techniques described here, libraries can be made ranging from a few compounds to >10,000 members in a highly efÞcient manner. Curr. C 2010 by John Wiley & Sons, Inc. Protoc. Chem. Biol. 2:135-151 Keywords: solid-phase r combinatorial r synthesis r diversity r Lanterns r silicon
INTRODUCTION Combinatorial chemistry is a powerful tool in the synthesis of small-molecule libraries for the development of biological probes and novel therapeutics (Crooks and Charles, 2000; Dolle et al., 2009). The use of combinatorial strategies, such as diversity-oriented synthesis (DOS; Nielsen and Schreiber, 2008), can afford highly diverse and structurally complex small molecules with great synthetic efÞciency. Combinatorial libraries can be synthesized in either solution- or solid-phase formats, both of which have their merits. Solution-phase parallel synthesis has the beneÞt of being compatible with a broad spectrum of reactions and allows for easy reaction monitoring. Meanwhile, solid-phase synthesis has the advantage of employing simple “split-pool” techniques, allowing for rapid generation of large compound libraries without need for automation. Assuming a successful synthesis, products of solid-phase synthesis are generally of sufÞcient purity for biological testing, while libraries produced by solution-phase synthesis often require puriÞcation, typically by HPLC. This need for puriÞcation can be reduced or avoided through the judicious selection of solid-phase scavengers and reagents (Ley et al., 2000; Ley and Baxendale, 2002; Weinbrenner and Tzschucke, 2006). Advances in solid-phase combinatorial synthesis have provided a variety of methods for compound immobilization, including different options for solid supports and a wide array of linkers (Scott, 2009). SynPhase Lanterns (http://www.mimotopes.com) are a practical alternative to conventional resins for library synthesis due to favorable reaction kinetics, easy handling, and simple washing procedures. Figure 1 shows a schematic of the composition of SynPhase Lanterns, as well as the different sizes offered by Mimotopes. SynPhase Lanterns are cylindrical in shape, containing a rigid polypropylene base coated with either a polystyrene (PS) or polyamide (PA) surface. The PS-Lanterns are most suitable for general organic synthesis in nonpolar solvents, while PA-Lanterns are useful for conducting reactions in hydrophilic or aqueous conditions. Many different linkers
Current Protocols in Chemical Biology 2: 135-151, July 2010 Published online July 2010 in Wiley Interscience (www.interscience.wiley.com). DOI: 10.1002/9780470559277.ch100038 C 2010 John Wiley & Sons, Inc. Copyright
Library Synthesis on SynPhase Lanterns
135 Volume 2
A grafted surface polymer
base polymer
B A-series
L-series
D-series
Figure 1 (A) Composition of SynPhase Lanterns: an unreactive base polymer and an outer polymer graft consisting of polystyrene or polyamide. (B) Lanterns are available in three sizes to accommodate different loading needs. A-series: 75 μmol/Lantern (5 mm × 5 mm). D-series: 35 μmol/Lantern (12.5 mm × 5 mm). L-series: 15 μmol/Lantern (17 mm × 6 mm). Figure reproduced with permission from Mimotopes.
are available, such as the Rink Amide (Verdi´e et al., 2008; Brucoli et al., 2009) and Backbone Amide (Zajdel et al., 2009) linkers. Herein, we focus on the L-series siliconfunctionalized SynPhase PS-Lanterns (Ryba et al., 2009), due to the compatibility of this linker with a wide array of transformations and proven robustness for library synthesis (Tallarico et al., 2001; Taylor et al., 2004; Marcaurelle et al., 2009).
Library Synthesis on SynPhase Lanterns
As depicted in Figure 2, the following protocol is divided into three sections: (1) activation of Lanterns and loading of the library scaffold, (2) solid-phase transformations of the immobilized scaffold via split-pool synthesis, and (3) cleavage of the small molecule from the Lantern. Activation of the Lantern is achieved via treatment with TfOH to form the intermediate diisopropyl silyl trißate. This reactive intermediate is then immediately treated with the library scaffold bearing an alcohol in the presence of excess
136 Volume 2
Current Protocols in Chemical Biology
i-Pr
i-Pr
TfOH, CH2Cl2
Si OMe
i-Pr
i-Pr Si
activation
loading R ⫽ library scaffold
⫽ L-Series SynPhase Lantern (15 mol)
i-Pr
i-Pr Si
OR
2,6-lutidine, ROH
OTf
see Figure 3
15% HF/Pyr, THF, 2 hr; TMSOMe, 10 min
split-pool synthesis
cleavage
R1-OH R2-OH
R3-OH R4-OH
library products immobilized scaffold
Figure 2 Use of silicon-functionalized Lanterns for library synthesis: activation, loading, split-pool synthesis, and cleavage.
2,6-lutidine to form a silyl ether. A variety of solid-phase transformations can then be carried out depending on the functional groups present on the library scaffold. Representative reactions compatible with the silyl ether linker are shown in Figure 3, and detailed procedures are provided in Basic Protocol 2. A number of nitrogen protecting groups are compatible with the silicon linker including Fmoc, Alloc, and Nosyl. Capping of the resulting amines can be achieved with sulfonyl chlorides, isocyanates, acids, and aldehydes to afford the corresponding sulfonamides, ureas, amides, and tertiary amines, respectively. An azide can also serve as a masked amine, as reduction can be performed with PBu3 in aqueous THF. Alternatively, azides can be converted to triazoles via a Huisgen 1,3-dipolar cycloaddition with alkynes. Esters can be hydrolyzed under mild conditions (using KOTMS) to provide an acid for coupling with amines. Lastly, aryl halides can undergo cross-coupling reactions such as Suzuki and Sonogashira reactions, with boronic acid and alkynes, respectively. Cleavage of library products (Fig. 2) from the Lantern can be achieved via treatment with HF/pyridine in THF. Quenching of the reaction with TMSOMe provides volatile by-products TMSF and MeOH, which can be removed by evaporation.
STRATEGIC PLANNING As mentioned above, solid-phase synthesis is a proven method for developing collections of small molecules, whether the objective is a small focused library for medicinal chemistry purposes or a large discovery library for initial screening efforts. Regardless of the type of library being synthesized, it is important to be mindful of the physicochemical properties (e.g., molecular weight, logP, polar surface area) of the library products at the outset of the synthesis (Blake, 2004). In order to prioritize library synthesis, cheminformatics methods such as principal component analysis (PCA; Feher and Schmidt, 2003), multi-fusion similarity (MFS) maps (Medina-Franco et al., 2007), ChemGPS (Oprea and Gottfries, 2001), and principal moments of intertia (PMI) plots (Sauer and Schwarz, 2003) can be utilized to analyze chemical diversity. Having designed a set of small molecules for synthesis, another critical element is the tracking of compounds during the library production. There are several options for tagging Lanterns, depending on the library design. If the library size is large (>100 compounds),
Library Synthesis on SynPhase Lanterns
137 Current Protocols in Chemical Biology
Volume 2
Protecting group removal: Fmoc: 20% Pip/DMF Alloc: Pd(PPh3)4, DMBA Nosyl: PhSH, K2CO3, DMF
PG N
R
Amine capping (RSO2Cl, RNCO, RCO2H, RCHO):
NH2 H N
PBu3, THF, H2O, rt, o/n
CO2Me
H N
O
H N
O
R
CO2H NH R
Suzuki coupling:
N
Amide coupling: O
X
CO2H
Sonogashira coupling:
X
O
HN NH2 NH2
KOTMS, THF, MeOH, H2O, rt, o/n
O S
R NH2
N3
H N
R N
CuAAC:
N3
N N N
Figure 3 Representative solid-phase transformations useful for split-pool library synthesis on silicon-functionalized Lanterns.
radiofrequency (RF) stems can be used for their ease in sorting and identiÞcation of the immobilized small molecule. The RF Transtems (Fig. 4A) Þt conveniently onto the Lanterns, and a Transort RF reader (Fig. 4C) can be used to sort the library at each synthesis step. RF-directed sorting requires the enumeration of the library and encoding of compound identities onto the Transtems. Several tools are available for library enumeration, including Pipeline Pilot, Chemaxon, Cambridgesoft, and Daylight. Enumeration and encoding for smaller libraries can be avoided by tracking members with colored spindles and cogs (Fig. 4B). Record keeping is essential when using this method of identifying compounds, especially for sorting Lanterns, to ensure that the intended compounds are synthesized.
Library Synthesis on SynPhase Lanterns
138 Volume 2
To ensure that a library of high purity is synthesized, quality control (QC) Lanterns should be included at each synthesis step for LCMS (or NMR) analysis. Colored RF Transtems are commercially available for this purpose to allow easy identiÞcation from a reaction ßask. For smaller libraries, the color-coded cogs and spindles can be used. If yield determination is required, then a full Lantern should be cleaved; otherwise Lanterns can be cut into quarters for cleavage and analysis to preserve material. When cleaving full Lanterns, the inclusion of multiple QC copies is recommended in case resubjection of the reaction is necessary. In general it is a good idea to determine yield at each step of the synthesis to ensure that premature cleavage from the Lantern is not occurring under the reaction conditions. The number of QC Lanterns per reaction ßask will depend on the size of the library and the number of combinatorial steps. It is recommended to sample a variety of compounds per reaction, as the success of the reaction may depend on the building block used in the previous step. Current Protocols in Chemical Biology
A
B
C
Figure 4 L-Series SynPhase Lanterns equipped with (A) radio frequency (RF) tags and (B) color-coded spindles and cogs. (C) Work station for RF-directed sorting.
LOADING OF A LIBRARY SCAFFOLD ONTO A SILICON-FUNCTIONALIZED LANTERN
BASIC PROTOCOL 1
Once a library has been designed and Lanterns have been suitably tagged and encoded (if necessary), production of the solid-phase library can begin. The following protocol describes the loading of a library scaffold bearing a free alcohol onto a silicon-functionalized L-series Lantern (∼15 μmol/Lantern). After the loading has been completed and before subsequent reactions are run, QC analysis is carried out to determine the success of the loading. Generally, this is done via cleavage and recovery of the scaffold (see Basic Protocol 3), but Fmoc quantitation (Gude et al., 2002) can also be used, if applicable.
Materials L-series alkyl tethered diisopropylarylsilane Lanterns (Mimotopes, cat. no. MIL10431000;http://www.mimotopes.com; also see Ryba et al., 2009) Transtems (stems with enclosed RF transponder, Mimotopes, cat. no. MIT10260010) Standard color tagging kit (colored cogs and spindles, Mimotopes, cat. no. MIT10430001) 3% trißuoromethanesulfonic acid solution in dichloromethane (TfOH in DCM; see recipe) Dichloromethane (DCM, anhydrous for reactions; HPLC grade for washings) Nitrogen source 2,6-lutidine (anhydrous; Aldrich, cat. no. 336106)
Library Synthesis on SynPhase Lanterns
139 Current Protocols in Chemical Biology
Volume 2
Library scaffold containing primary or secondary alcohol (coevaporated from benzene or toluene) N,N-dimethylformamide (DMF, HPLC grade) Tetrahydrofuran (THF, containing BHT as inhibitor) Isopropanol Oven-dried reaction vessel with screw top: e.g., ChemGlass, cat. no. CG-1880-42; (http://www.chemglass.com/) or Mimotopes, cat. no. MIA10140006 (http://www.mimotopes.com) Incubator shaker (New Brunswick ScientiÞc, model M1353-0004 or similar for large libraries) Rubber septa (Sigma-Aldrich, cat. no. Z512222 or similar) Washing apparatus (ceramic Buchner funnel and waste container) Lyophilizer or high-vacuum manifold SynPhase work station (cleavage tray, Lantern tray, stem ejector, SynPhase press and stem tray; Mimotopes, cat. no. MIA10910001) Transort RF reader and software (Mimotopes, cat. no. MIT10520001) CAUTION: TfOH is highly corrosive and personal contact can result in injury. Extreme caution should be exercised. NOTE: Transtems can be attached to Lanterns using the SynPhase press (see SynPhase workstation) to enable RF sorting with the Transort reader and software.
Activate Lanterns with TfOH 1. To an oven-dried shaker vessel containing the Lanterns equipped with RF or colorcoded tags (e.g., Transtems; see Fig. 4A), add a 3% TfOH solution in DCM under a ßow of nitrogen. Add enough TfOH solution to cover all Lanterns. Cap vessel and shake at room temperature for 10 min. In general, for large-scale reactions (>100 Lanterns), the TfOH solution can be poured directly into the reaction vessel under a ßow of nitrogen. For smaller reactions, a syringe can be used for transferring this reagent to a ßask equipped with a rubber septum. The Lanterns should turn an orange or deep red color upon addition of TfOH.
Carry out loading 2. After 10 min, exchange the cap for a rubber septum and remove the TfOH solution with a syringe (or cannula) while maintaining an inert atmosphere (e.g., nitrogen) inside the ßask. 3. Add anhydrous 2,6-lutidine (12.0 eq. with respect to Si). Purge with nitrogen and shake Lanterns for several minutes. The number of equivalents of reagent added are based on the equivalents of Si per Lantern. For L-series Lanterns, an average of 15 μmol/Lantern is observed. The Lanterns should turn white after several minutes. Add additional 2,6-lutidine as needed until Lanterns turn white.
4. Add the library scaffold (alcohol, 1.2 eq. with respect to Si) in anhydrous DCM via syringe or cannula and purge with nitrogen. Exchange rubber septum for a screw cap, and transfer reaction vessel to incubator shaker. Shake at room temperature overnight.
Library Synthesis on SynPhase Lanterns
140 Volume 2
The scaffold should be coevaporated with either benzene or toluene (benzene is recommended) prior to the loading to remove any residual water, and then placed on high vacuum overnight. DCM should be used conservatively during the loading to maintain as high a concentration of the scaffold as possible, while still covering all the Lanterns. A slight excess of solvent should be used to account for swelling of Lanterns. Current Protocols in Chemical Biology
For scaffolds with low reactivity or limited solubility in DCM, longer reaction times (up to 4 days) can be used to facilitate loading. In general, higher loading levels are achieved with primary alcohols than secondary alcohols.
Wash and dry Lanterns 5. After shaking overnight, remove the reaction mixture and set aside. Wash Lanterns with DCM for 5 min in the reaction vessel. The reaction mixture is kept until the analysis for the loading has been completed, to ensure that loading was successful. If the loading was unsuccessful, the scaffold can be salvaged from this mixture via acidic aqueous wash to remove excess 2,6-lutidine, and puriÞed by silica gel puriÞcation. For washing Lanterns, the washing solvent and Lanterns are Þltered through a ceramic B¨uchner funnel via gravity Þltration, with the wash going into a waste container and the Lanterns being returned to the reaction ßask. Alternatively, a drilled cap with Þlter holes (Mimotopes, cat. no. MIA10070012) can be used to drain the solvent from the reaction ßask.
6. Wash the Lanterns (in the reaction vessel) using the following solvents for 30 min each (∼0.8 ml/Lantern): DMF (two washes), THF/H2 O (3:1), THF/isopropanol (3:1), THF, and DCM. THF containing BHT as an inhibitor should be used to avoid possible oxidation of amines.
7. Place the washed Lanterns on a lyophilizer or high vacuum overnight before proceeding to the next reaction.
Determine loading level 8. Follow cleavage protocol (see Basic Protocol 3) with three Lanterns for loading analysis. At least three Lanterns should be used to measure consistency of loading across all Lanterns. Expected loading amounts are calculated based on recovered amounts of compound and typically average ∼15 μmol/Lantern. Where applicable, Fmoc quantitation (Gude et al., 2002) can also be used to measure success of loading. QC analysis ensures there was no decomposition during the loading step.
9. Sort the Lanterns using Transort RF reader and software for upcoming reactions, if necessary. If color-coded stems and cogs were used rather than RF tags, refer back to the master sheet to sort for the upcoming reactions.
SOLID-PHASE TRANSFORMATIONS OF SMALL-MOLECULE LIBRARIES ON LANTERNS Once the loading of a library scaffold has been achieved, a split-pool combinatorial approach is applied for the introduction of appendage diversity. Below, a series of solidphase transformations is presented that have been found to be compatible with siliconfunctionalized Lanterns. The protocol includes conditions for the removal of a variety of protecting groups (e.g., Fmoc, Alloc, Nosyl), amine capping, amide coupling, azidealkyne cycloaddition, and cross-coupling reactions. The reaction conditions provided are intended to be a starting point for use with a wide variety of building blocks (e.g., sulfonyl chloride, isocyanates, acids, etc.). Some optimization may be required. Once a solid-phase transformation has been carried out, LC-MS analysis can be utilized to determine the success of the reaction. If a reaction is deemed incomplete, the Lanterns can simply be re-subjected to the reaction conditions until the desired results are observed. Between synthesis steps, Lanterns can be sorted using either the color-coded spindles/cogs or RF tags (using the Transort reader). The next reaction can then be run, and the QC process repeated. This process is continued until the synthesis is complete.
BASIC PROTOCOL 2
Library Synthesis on SynPhase Lanterns
141 Current Protocols in Chemical Biology
Volume 2
The following set of procedures are not intended for sequential use, but rather implemented as desired to Þt the design of the library. The success of each reaction may be dependent on the library scaffold used; therefore, some optimization may be required.
Materials Lanterns, loaded (see Basic Protocol 1) Tetrahydrofuran (THF), anhydrous, containing BHT inhibitor Tetrakis(triphenylphosphine)palladium(0) (Pd(PPh3 )4 ; Aldrich, cat. no. 216666) 1,3-dimethylbarbituric acid (DMBA) N,N-dimethylformamide (DMF, HPLC grade) 0.1 M sodium cyanide in a 1:1 THF/H2 O solution (prepare fresh) Isopropanol Dichloromethane (DCM, anhydrous for reactions; HPLC grade for washings) 20% (v/v) piperidine in DMF (store up to 1 month at 25◦ C) Thiophenol Potassium carbonate Methanol (MeOH) Potassium trimethylsilanolate (KOTMS; Aldrich, cat. no. 324868) Tributylphosphine (PBu3 ) 2,6-lutidine (anhydrous) Sulfonyl chlorides Isocyanates Triethylamine (EtN3 , anhydrous) Carboxylic acids (Benzotriazol-1-yloxy)tripyrrolidinophosphonium hexaßuorophosphate (PyBOP; Aldrich, cat. no. 377848) Sodium triacetoxyborohydride [Na(OAc)3 BH] 2% (v/v) acetic acid in DMF Aldehydes Ethanol (EtOH) Boronic acids Bis(triphenylphosphine)palladium(II) dichloride (Pd(PPh3 )2 Cl2 ; Aldrich, cat. no. 412720) Nitrogen source Alkynes N,N-diisopropylethylamine (DIEA) Copper (I) iodide (CuI) Amines Oven-dried reaction vessel with screw top (e.g., ChemGlass, cat. no. CG-1880-42, Mimotopes, cat. no. MIA10140006) Ceramic B¨uchner funnel Incubator shaker (New Brunswick ScientiÞc, model M1353-0004 or similar for large libraries) Additional reagents and equipment for quality control analysis (Basic Protocol 3) CAUTION: Some washes require the use of a solution of NaCN to remove residual metals (e.g., Pd, Cu). Exercise great caution when performing these washes as NaCN is highly toxic.
Library Synthesis on SynPhase Lanterns
For Alloc removal 1a. To a reaction vessel containing Lanterns, add THF (0.8 ml/Lantern), followed by Pd(PPh3 )4 (1 eq. with respect to Si), and 1,3-dimethylbarbituric acid (30 eq. with respect to Si). Seal the vessel and shake at room temperature overnight.
142 Volume 2
Current Protocols in Chemical Biology
In general, 1,3-dimethylbarbituric acid is superior π -allyl scavenger as compared to other reagents (e.g., phenysilane, morpholine) for preventing N-allylation.
2a. Remove the reaction mixture and wash the Lanterns in the reaction vessel with DMF, followed by three washes (30 min each) with 0.1 M NaCN in 1:1 THF/H2 O, removing solvent after each wash by gravity Þltration through a ceramic B¨uchner funnel. 3a. Wash the Lanterns (in the reaction vessel) using the following solvents for 30 min each (∼0.8 ml/Lantern): DMF (two washes), THF/H2 O (3:1), THF/isopropanol (3:1), THF, and DCM. Follow Basic Protocol 3 for QC analysis. If reaction is incomplete or N-allyation is observed, repeat starting at step 1a.
Fmoc removal 1b. To a reaction vessel containing Lanterns, add a solution of 20% piperidine in DMF (0.8 ml/Lantern). Seal the vessel and shake for 30 min. 2b. Remove the piperidine solution. Wash the Lanterns (in the reaction vessel) using the following solvents for 30 min each (∼0.8 ml/Lantern): DMF (two washes), THF/H2 O (3:1), THF/isopropanol (3:1), THF, and DCM. Follow Basic Protocol 3 for QC analysis. If reaction is incomplete, repeat starting at step 1b.
Nosyl removal 1c. To a reaction vessel containing Lanterns, add DMF (0.8 ml/Lantern) followed by thiophenol (20 eq. with respect to Si) and potassium carbonate (30 eq. with respect to Si). Seal the vessel and shake at room temperature overnight. 2c. Remove the reaction mixture. Wash the Lanterns (in the reaction vessel) using the following solvents for 30 min each (∼0.8 ml/Lantern): DMF (two washes), THF/H2 O (3:1), THF/isopropanol (3:1), THF, and DCM. Follow Basic Protocol 3 for QC analysis. If reaction is incomplete, repeat starting at step 1c.
Ester hydrolysis 1d. To a reaction vessel containing Lanterns, add THF/MeOH/H2 O (5:2:2) (0.8 ml/Lantern) followed by KOTMS (10 eq. with respect to Si). Seal the vessel and shake at room temperature overnight. 2d. Remove the reaction mixture. Wash the Lanterns (in the reaction vessel) using the following solvents for 30 min each (∼0.8 ml/Lantern): DMF (two washes), THF/H2 O (3:1), THF/isopropanol (3:1), THF, and DCM. Follow Basic Protocol 3 for QC analysis. If reaction is incomplete, repeat starting at step 1d.
Azide reduction 1e. To a reaction vessel containing Lanterns, add 9:1 THF/H2 O (0.8 ml/Lantern) followed by PBu3 (10 eq. with respect to Si). Seal the vessel and shake at room temperature overnight. 2e. Remove the reaction mixture, then add THF/H2 O (3:1). Seal the vessel and shake at room temperature for 6 hr. 3e. Wash the Lanterns (in the reaction vessel) using the following solvents for 30 min each (∼0.8 ml/Lantern): DMF (two washes), THF/H2 O (3:1), THF/isopropanol (3:1), THF, and DCM. Follow Basic Protocol 3 for QC analysis. If reaction is incomplete, repeat starting at step 1e.
Amine capping: sulfonyl chlorides 1f. To a reaction vessel containing Lanterns, add DCM (0.8 ml/Lantern) followed by 2,6-lutidine (25 eq. with respect to Si) and a selected sulfonyl chloride (20 eq. with respect to Si). Seal the vessel and shake at room temperature overnight.
Library Synthesis on SynPhase Lanterns
Current Protocols in Chemical Biology
Volume 2
143
2f. Remove the reaction mixture, and wash Lanterns with DCM in B¨uchner funnel. Continue to Wash the Lanterns (in the reaction vessel) using the following solvents for 30 min each (∼0.8 ml/Lantern): DMF (two washes), THF/H2 O (3:1), THF/isopropanol (3:1), THF, and DCM. Follow Basic Protocol 3 for QC analysis. If reaction is incomplete, repeat starting at step 1f.
Amine capping: isocyanates 1g. To a reaction vessel containing Lanterns, add DCM (0.8 ml/Lantern) followed by a selected isocyanate (15 eq. with respect to Si). Seal the vessel and shake at room temperature overnight. 2g. Remove the reaction mixture and wash Lanterns with DCM in the reaction vessel. Continue to wash the Lanterns (in the reaction vessel) using the following solvents for 30 min each (∼0.8 ml/Lantern): DMF (two washes), THF/H2 O (3:1), THF/isopropanol (3:1), THF, and DCM. Follow Basic Protocol 3 for QC analysis. If reaction is incomplete, repeat starting at step 1g.
Amine capping: carboxylic acids 1h. To a reaction vessel containing Lanterns, add DCM (0.8 ml/Lantern), followed by Et3 N (30 eq. with respect to Si), a selected carboxylic acid (20 eq. with respect to Si), and PyBOP (20 eq. with respect to Si). Seal the vessel and shake at room temperature overnight. 2h. Remove the reaction mixture and wash with DCM in the reaction vessel. Continue to wash the Lanterns (in the reaction vessel) using the following solvents for 30 min each (∼0.8 ml/Lantern): DMF (two washes), THF/H2 O (3:1), THF/isopropanol (3:1), THF, and DCM. Follow Basic Protocol 3 for QC analysis. If reaction is incomplete, repeat starting at step 1h. Occasionally, using the above protocol with a primary amine, bis-capping can result in imide formation. Treatment with a 3:1 THF/pyrrolidine solution at room temperature overnight typically will provide the desired amide.
Amine capping: aldehydes 1i. To a reaction vessel containing Lanterns, add DMF containing 2% acetic acid (0.8 ml/Lantern) followed by a selected aldehyde (20 eq. with respect to Si). Seal the vessel and shake at room temperature for 1 hr. 2i. Remove cap and add Na(OAc)3 BH (22 eq. with respect to Si). Seal the vessel and shake at room temperature for 3 days. 3i. Remove the reaction mixture and wash the Lanterns (in the reaction vessel) using the following solvents for 30 min each (∼0.8 ml/Lantern): DMF (two washes), THF/H2 O (3:1), THF/isopropanol (3:1), THF, and DCM. Follow Basic Protocol 3 for QC analysis. If reaction is incomplete, repeat starting at step 1i. The amine capping with aldehydes is ideally suited for secondary amines (or the dialkylation of primary amines), as it is difÞcult to selectively mono-alkylate primary amines using this protocol.
Cross coupling: boronic acids 1j. To a reaction vessel containing Lanterns add EtOH (0.8 ml/Lantern), followed by a selected boronic acid (20 eq. with respect to Si), Et3 N (40 eq. with respect to Si), and Pd(PPh3 )2 Cl2 (1 eq. with respect to Si). Purge vessel with nitrogen, cap, and shake at 60◦ C for 5 days. Library Synthesis on SynPhase Lanterns
2j. Remove the reaction mixture and wash Lanterns with DMF several times in the reaction vessel. Next, wash Lanterns three times, each time for 30 min, with a solution of NaCN (0.1 M in 1:1 THF/H2 O solution).
144 Volume 2
Current Protocols in Chemical Biology
3j. Continue to wash the Lanterns (in the reaction vessel) using the following solvents for 30 min each (∼0.8 ml/Lantern): DMF (two washes), THF/H2 O (3:1), THF/isopropanol (3:1), THF, and DCM. Follow Basic Protocol 3 for QC analysis. If reaction is incomplete, repeat starting at step 1j.
Cross coupling: alkynes 1k. To a reaction vessel containing Lanterns add DMF (0.8 ml/Lantern), followed by a selected alkyne (20 eq. with respect to Si), DIEA (30 eq. with respect to Si), CuI (30 eq. with respect to Si), and Pd(PPh3 )2 Cl2 (2.0 eq. with respect to Si). Purge vessel with nitrogen, cap and shake at 60◦ C overnight. 2k. Remove the reaction mixture and wash Lanterns with DMF several times in the reaction vessel. Next, wash Lanterns three times with a solution of NaCN (0.1 M in 1:1 THF/H2 O). 3k. Remove the reaction mixture and wash Lanterns with DMF several times in the reaction vessel. Next, wash the Lanterns (in the reaction vessel) using the following solvents for 30 min each (∼0.8 ml/Lantern): DMF (two washes), THF/H2 O (3:1), THF/isopropanol (3:1), THF, and DCM. Follow Basic Protocol 3 for QC analysis. If reaction is incomplete, repeat starting at step 1k.
Cu-catalyzed azide/alkyne cycloaddition (CuAAC) 1l. To a reaction vessel containing Lanterns add THF (0.8 ml/Lantern) followed by a selected alkyne (30 eq. with respect to Si), CuI (10 eq. with respect to Si), and DIEA (30 eq. with respect to Si). Seal the vessel and shake at 60◦ C for 24 hr. Although the example here is for solid-support azides, in theory, solid-supported alkynes could be used with a variety of azides to yield similar triazoles.
2l. Remove the reaction mixture and wash Lanterns with DMF several times in a B¨uchner funnel. Next, wash Lanterns three times, each time for 30 min, with a solution of NaCN (0.1 M in 1:1 THF/H2 O). 3l. Continue washing the Lanterns (in the reaction vessel) using the following solvents for 30 min each (∼0.8 ml/Lantern): DMF (two washes), THF/H2 O (3:1), THF/isopropanol (3:1), THF, and DCM. Follow Basic Protocol 3 for QC analysis. If reaction is incomplete, repeat starting at step 1l.
Amidation: PyBOP coupling 1m. To a reaction vessel containing Lanterns add a 3:1 DCM/DMF solution (0.8 ml/Lantern), followed by a selected amine (20 eq. with respect to Si), DIEA (10 eq. with respect to Si), and PyBOP (10 eq. with respect to Si). Seal the vessel and shake at room temperature overnight. 2m. Wash the Lanterns (in the reaction vessel) using the following solvents for 30 min each (∼0.8 ml/Lantern): DMF (two washes), THF/H2 O (3:1), THF/isopropanol (3:1), THF, and DCM. Follow Basic Protocol 3 for QC analysis. If reaction is incomplete, repeat starting at step 1m.
CLEAVAGE OF FUNCTIONALIZED SCAFFOLD FROM SOLID SUPPORT Once the desired solid-phase transformations have been performed, the product is removed from the Lantern via treatment with HF/pyridine. Cleavage may be carried out during the course of a library synthesis to determine yield after loading (see Basic Protocol 1) or to monitor the success of solid-phase reactions (see Basic Protocol 2). Here, a standardized cleavage protocol is presented for use with silicon-functionalized Lanterns, which can be carried out in a deep 96-well plate or microcentrifuge tube. When processing multiple samples at once, it is preferable to use a 96-well plate as this allows for
BASIC PROTOCOL 3
Library Synthesis on SynPhase Lanterns
145 Current Protocols in Chemical Biology
Volume 2
the use of a multichannel pipet. The following protocol describes cleavage of up to 96 Lanterns in a single 96-well plate using a multichannel pipet. If producing a large-scale library, the use of an automated liquid handler is preferred.
Materials Lanterns, loaded, subjected to desired solid-phase transformations (see Basic Protocol 2) Cleavage solution (see recipe) Methoxytrimethylsilane (TMSOMe; 99%, Aldrich, cat. no. 253006-250g) Methanol (MeOH) Dichloroethane (DCE) Dimethylsulfoxide (DMSO) Labeled deep 96-well plate (Seahorse Biosciences, cat. no. XPO128; http://www.seahorsebio.com/) Other 96-well plates to use as covers Spreadsheet software Multichannel pipettors and polypropylene reservoir Labeled tube rack with preweighed 2-D barcoded glass mini-tubes (Tradewinds, 1.2-ml hi-recovery mini-tube, cat. no. 063227-0301, http://www.twdtradewinds.com/, or similar) Centrifugal evaporator (e.g., Genevac HT12 or HT24, SP Industries) Automated weighing station (e.g., FlexiWeigh, Mettler Toledo) 2-D barcode reader (e.g., VisionMatePlus, Thermo Fisher ScientiÞc) Additional reagents and equipment for liquid chromatography/mass spectrometry (LC-MS) CAUTION: Use proper personal protection equipment when handling HF/pyridine (safety glasses, nitrile gloves, and lab coat). Labware that comes into contact with HF/pyridine should be quenched with methoxy- or ethoxytrimethylsilane. Prepare a solution of 25% methoxytrimethylsilane in THF and keep it nearby at all times. NOTE: Remove any identiÞers (Transtems or colored spindles) prior to cleavage. For removing Transtems, the use of a Stem Recycler Module (SynPhase stem recycler module for L- and D-series Lanterns; Mimotopes, cat. no. MIA10880001) is required.
Cleave Lanterns with HF/pyridine 1. Put one Lantern per well in a deep 96-well plate. Note the position of each Lantern in spreadsheet software. For testing at an intermediate stage of the synthesis, it is not necessary to cleave a whole Lantern if yield determination is not required. Lanterns can be cut into quarters using a razor blade. The amount of cleavage solution and TMSOMe should be reduced accordingly. For 1/4 Lantern, 200 μl of HF/pyridine solution is sufÞcient (see step 2). After extensive testing of different deep 96-well plates, Seahorse Biosciences cleavage plates were selected for use, as they released the least amount of detectable plasticizers.
2. Using multichannel pipet add 350 μl of cleavage solution to each well containing a Lantern. The Lantern should be fully submerged in cleavage solution. When using the multichannel pipet, the cleavage solution should be transferred from a polypropylene reservoir, due to the reactive nature of HF/pyridine and THF. Library Synthesis on SynPhase Lanterns
3. Cover cleavage plate with another 96-well plate and let sit for 3 hr at room temperature.
146 Volume 2
Current Protocols in Chemical Biology
4. Quench reaction by adding two volumes (700 μl) of methoxytrimethylsilane. Addition of methoxytrimethylsilane should be slow, as the reaction is exothermic.
Transfer cleavage mixture to glass tubes 5. After 10 to 15 min, use a multichannel pipettor to mix the contents of the wells and transfer quenched solution from each well into its own preweighed 2-D barcoded glass tube (or other appropriate glass tube). Glass tubes should be used to avoid release of plasticizers into compound solution.
6. Add 200 μl of MeOH to each well containing Lanterns. 7. Using a multichannel pipettor, mix the contents of the wells to wash the Lanterns, and transfer the wash into its respective tube. 8. Dry contents of the tubes using centrifugal evaporator. To ensure proper drying conditions, program the evaporator to remove most of the THF Þrst (∼2 hr at 23 mbar; use REMP feature to prevent solvent bumping), then apply full vacuum for at least 12 hr to remove pyridine.
Determine yield and purity 9. Weigh tubes using automated weighing station. 10. Calculate yield for each compound by subtracting tare tube weight from Þlled tube weight. Determine loading level if required. 11. Dissolve product in 1:1 DCE/MeOH to ∼10 mg/ml. 12. Remove a 5-μl aliquot and add it to 45 μl of DMSO. Mix well and analyze by LC-MS. Analysis can be done in LC-MS vial or from a 96- or 384-well microplate depending on the number of samples. At times the formation of a TMS adduct may be observed by LC-MS (M+72). This impurity can be reduced by resuspending the compound in MeOH, soaking for 4 hr at room temperature, and drying in a centrifugal evaporator.
REAGENTS AND SOLUTIONS Use deionized water in all recipes and protocol steps.
Cleavage solution To prepare 100 ml of cleavage solution, add 15 ml of 30% (v/v) hydroßuoric acid in pyridine (Aldrich, cat. no. 184225-100g) to 85 ml of anhydrous stabilized tetrahydrofuran (THF) containing BHT as an inhibitor (Aldrich, cat. no. 4017571L). Use a polypropylene bottle to store the solution. The cleavage solution can be made in advance and stored in a freezer up to 3 months at −20◦ C for future use.
TfOH in DCM, 3% Remove the Sure-Seal cap of a 100-ml Sure-Seal bottle of dichloromethane (SigmaAldrich; do not discard Sure-Seal cap) and pour 5 g of trißic acid (from a 5-g ampule; Aldrich, cat. no. 347817) directly into the dichloromethane bottle. Replace the Sure-Seal cap, along with the screw cap, and shake well. Use the solution soon after preparing. Library Synthesis on SynPhase Lanterns
147 Current Protocols in Chemical Biology
Volume 2
COMMENTARY Background Information
Library Synthesis on SynPhase Lanterns
SynPhase PS-Lanterns produced by Mimotopes are solid supports containing a rigid polypropylene base coated with a polystyrene (PS) mobile surface. The unique “Lantern” shape allows for the attachment of snap-Þtting tags for encoding purposes, and allows for the free ßow of reactants and rapid drainage of washing solvents. PS-Lanterns can be functionalized with a variety of different linkers allowing for ßexibility in loading strategies. For example, amines can be immobilized via reductive alkylation using Backbone Amide (BAL) Lanterns while carboxylic acids and phenols can be loaded through Hydroxymethylphenoxy (HMP) Lanterns. Meanwhile, primary and secondary alcohols can be loaded using a variety of linkers (Nam et al., 2003), including silicon-functionalized Lanterns (Ryba et al., 2009), now commercially available from Mimotopes. This latter technology was chosen as the focus of this article due to the compatibility and robustness of the linker for library synthesis applications (Tallarico et al., 2001; Taylor et al., 2004; Marcaurelle et al., 2009). Primary and secondary alcohols can also be used as a handle for afÞnity chromatography, smallmolecule microarrays, and the attachment of biasing elements during subsequent follow-up in biological assays (Wong et al., 2003; Radner et al., 2006; Wang et al., 2008). The mobile surface of the siliconfunctionalized Lantern contains a pmethoxyphenyl (PMP)–protected diisopropylsilyl linker. Upon activation with trißic acid, the PMP group is removed and the subsequent silyl trißate is formed. Due to the highly reactive nature of silyl trißates, inert reaction conditions are required to prevent hydrolysis of the activated Lanterns, resulting in unsatisfactory loading. Under these inert conditions, in the presence of base, a scaffold containing a free alcohol can be easily immobilized onto the Lantern. The silicon-functionalized linker can withstand a variety of reaction conditions, but does have certain limitations. Although the linker is stable to a wide range of basic conditions, including reductions and metal-mediated reactions, the use of strongly acidic conditions, as well as sources of ßuoride, should be avoided, as this can result in premature cleavage of compound from the Lantern. In addition, while a library scaffold can be recovered if the loading is unsuccessful, the Lanterns cannot be reused
or resubjected to the loading reaction. Thus, great care should be taken during the loading step.
Critical Parameters The success of a library synthesis depends largely on the extent to which solid-phase feasibility studies are carried out prior to production. For example, loading levels may vary from one scaffold to another; thus, it is highly recommended that loading be tested on a minimum of 5 Lanterns prior to library synthesis. While the procedures provided in Basic Protocol 2 serve as a good starting point for solidphase diversiÞcation, some optimization may be required, as the success of each reaction may be scaffold-dependent. It is recommended that a representative set of building blocks with varying reactivity (e.g., alkyl, aryl, electronwithdrawing, electron-donating, hindered) be tested under the suggested reaction conditions prior to library production. The cross-coupling reactions, in particular, may require the screening of multiple building blocks. Varying the recommended reaction conditions (e.g., temperature, solvent, Pd source) may also be necessary. In general, scouting reaction conditions in solution phase is not necessary, or even worthwhile, as the conditions often do not translate to the solid phase. It is important to consider the sequence of steps to be employed in a library synthesis, particularly with respect to protecting groups. When the reaction sequence is not obvious, multiple options may be pursued in parallel during feasibility studies. In general, Fmoc is quite labile and should be removed early in the library synthesis, while Alloc is compatible with most solid-phase transformations with the exception of cross-coupling procedures. In order to prevent premature cleavage of products from the Lantern, protecting groups that require the use of harsh acid or base for removal should be avoided. Other critical parameters for a successful library production include the use of reagents and scaffolds of high purity to ensure clean reactions, and, ultimately, clean products directly from the library production. In general, it is recommended that 3% to 5% of the library be sampled for QC to ensure good results overall. The synthesis of a pilot library (100 to 500 compounds) can also be beneÞcial prior to engaging in a large-scale (>1000 compounds) library production.
148 Volume 2
Current Protocols in Chemical Biology
Troubleshooting Poor loading The easiest problem to diagnose when there is low loading of a library scaffold is poor solubility. If a scaffold has low solubility in DCM, the amount of solvent employed in the loading can be increased, and reaction time can be extended to up to 5 days to try improving loading levels. Sonication of the scaffold in DCM prior to addition to Lanterns may also be beneÞcial. If results are still suboptimal, then changes to the library scaffold may be required to improve solubility. For example, a simple change in protecting groups may facilitate dissolution in DCM. Heating the loading reaction generally does not lead to improved results, and cosolvents such as THF, DMF, or MeOH are incompatible with the loading reaction. An unsuccessful loading reaction can also be due to moisture. Although difÞcult to detect, trace amounts of water in the trißic acid, 2,6-lutidine, or DCM can lead to poor loading results. The TfOH solution should always be prepared fresh prior to use, and the 2,6lutidine obtained from an unopened Sure-Seal bottle (Sigma-Aldrich), or freshly distilled. The Lanterns should also be washed with anhydrous DCM and dried under high vacuum prior to use, and the library scaffold coevaporated from benzene or toluene. The use of an oven- or ßame-dried reaction vessel is also recommended. If the library scaffold is highly soluble and great care has been taken to exclude moisture from the loading reaction, and loading levels are still low (<10 μmol), the loading site may be inaccessible. In general, higher loading levels are obtained with primary alcohols than secondary alcohols; however, reasonable loading levels (>10 μmol) can be achieved for secondary alcohols if the loading site is not sterically hindered. Phenols can also be utilized as a handle for loading, but the aryl silyl ether linkage is slightly more sensitive to basic conditions. If modiÞcations to the loading site are not feasible (or desirable), then the use of an alternate linker strategy may be needed. Low conversion Depending on the type of reaction or building block employed, low conversion may be observed that is not overcome by simply repeating the recommended reaction conditions. This typically can be remedied by employing harsher reaction conditions, assuming the required conditions are compatible with the silyl
linker. For example, when capping an amine with a sterically hindered isocyanate, heating the reaction in toluene may be required to drive the reaction to completion. Meanwhile, sluggish reaction when capping an amine with a sulfonyl chloride can often be overcome by switching from 2,6-lutidine to pyridine. Reaction optimization will vary from substrate to substrate, and feasibility testing can prevent low conversion from happening during the actual library production. Low purity Reagent quality can have the greatest impact on the purity of Þnal products. It is recommended that detailed and precise records be maintained during production of a library in case undesirable results are obtained for a particular building block. These records should also include reagent supplier information. In some cases, impurities may be traced back to a reagent. In other instances, a “combinatorial effect” may be observed where certain building-block combinations yield products of low purity. For the most part, this problem cannot be avoided even with thorough solid-phase feasibility studies. A common impurity that may be observed upon HF cleavage is a TMS adduct (M+72), resulting from quenching of the reaction mixture with TMSOMe. Fortunately, this impurity can be reduced by treatment with MeOH at room temperature. On rare occasions, unexpected rearrangements may occur under the HF/pyridine cleavage conditions. In such instances, the use of buffered HF/pyridine (e.g., 15% HF/pyridine in 1:1 THF/pyridine) may prevent or minimize undesired rearrangement. If the product yield is lower than expected using buffered conditions, then the cleavage can be repeated. For the cleavage of highly acid-sensitive products, the use of TBAF may be preferred. In this case, HPLC puriÞcation of the Þnal products will be needed to remove excess TBAF. For products containing tertiary amines, there is a risk of N-oxide formation if unstabilized THF is used during HF-cleavage or washing steps. This problem is easily detectible by LCMS (M+16). For this reason, THF containing BHT as an inhibitor of oxidation should be used for all washing and cleavage steps. The use of DCM/MeOH (or DCE/MeOH) is recommended for all volatile transfers post-cleavage, as the continued use of stabilized THF leads to an increase in BHT impurities.
Library Synthesis on SynPhase Lanterns
149 Current Protocols in Chemical Biology
Volume 2
Low yield If a low yield of Þnal product is obtained after treatment with HF/pyridine, a second round of cleavage should be carried out to ensure all compound has been removed from the Lantern. Assuming initial loading levels were satisfactory, if additional product is not recovered after a second cleavage event, this may indicate premature release of compound from the Lantern.
Anticipated Results The loading step (Basic Protocol 1) typically yields ∼15 μmol of core/Lantern for Lseries Lanterns. During the subsequent steps, this number should remain constant and the Þnal product obtained in similar amounts. The solid-phase manipulations are usually clean, but the conversions may not always be complete. Re-subjecting the Lanterns to the reaction conditions can be expected at some point during the library production. Driving the reaction to completion is always recommended unless byproducts begin to appear and counter the progress of increasing the conversion. Purity levels can vary based on the choice of reactions, building blocks, and number of steps. Feasibility studies and optimization of reactions should be carried out to ensure that a majority of the library is above the desired purity threshold. Typically, with the reactions described in Basic Protocol 2 and simple building blocks, >85% of the compounds can be expected to be >80% purity for a 4- to 6-step sequence.
Time Considerations
Library Synthesis on SynPhase Lanterns
The amount of time needed to produce a library of small molecules using siliconfunctionalized Lanterns depends on several factors and can range from a few weeks to several months. This timeframe does not include time required for solid-phase feasibility studies, which depends highly on the scaffold and associated chemistry. Very little time, relative to the whole, is spent on the library setup. The main factors in contributing to the length of time are size of the library, number of steps, and difÞculty of the reaction sequence. The size of the library does not increase the reaction time, but does add more time during the sorting and cleavage steps, as well as LC-MS analysis. A very difÞcult reaction can result in poor conversion and the need for repeating a reaction step. Some reactions require longer reaction times (such as the Suzuki reaction), which may add to the overall production time
line. For a medium- to large-sized library, typically 1 week should be set aside for each step in the library synthesis to account for reaction setup, reaction time (assuming overnight reaction), washing, drying, quality control, and sorting. For example, the solid-phase synthesis of 1000-membered library involving six synthesis steps, including loading, protecting group removal, amine capping, ester hydrolysis, amide coupling, and cleavage, would take approximately 6 weeks, assuming no resubjections are required.
Acknowledgments This work was supported in part by CMLD funding from the NIH (Broad Institute CMLD; P50 GM069721). The authors would like to thank Dr. Lakshmi Akella for helpful reading of this Protocol.
Literature Cited Blake, J.F. 2004. Integrating cheminformatic analysis in combinatorial chemistry. Curr. Opin. Chem. Bio. 8:407-411. Brucoli, F., Howard, P.W., and Thurston, D.E. 2009. EfÞcient solid-phase synthesis of a library of distamycin analogs containing novel biaryl motifs on SynPhase Lanterns. J. Comb. Chem. 11:576586. Crooks, S.L. and Charles, L.J. 2000. Overview of combinatorial chemistry. Curr. Protoc. Pharmacol. 10:9.3.1-9.3.16. Dolle, R.E., Le Bourdonnec, B., Goodman, A.J., Morales, G.A., Thomas, C.J., and Zhang, W. 2009. Comprehensive survey of chemical libraries for drug discovery and chemical biology: 2008. J. Comb. Chem. 11:739-790. Feher, M. and Schmidt, J.M. 2003. Property distributions: Differences between drugs, natural products, and molecules from combinatorial chemistry. J. Chem. Inf. Comput. Sci. 43:218227. Gude, M., Ryf, J., and White, P.D. 2002. An accurate method for the quantitation of Fmocderivatized solid phase supports. Lett. Pept. Sci. 9:203-206. Ley, S.V. and Baxendale, I.R. 2002. New tools and concepts for modern organic synthesis. Nat. Rev. Drug Discov. 1:573-586. Ley, S.V., Baxendale, I.R., Bream, R.N., Jackson, P.S., Leach, A.G., Longbottom, D.A., Nesi, M., Scott, J.S., Storer, R.I., and Taylor, S.J. 2000. Multi-step organic synthesis using solid-supported reagents and scavengers: A new paradigm in chemical library generation. J. Chem. Soc., Perkin Trans. 1 38154195. Marcaurelle, L.A., Johannes, C., Yohannes, D., Tillotson, B.P., and Mann, D. 2009. Diversityoriented synthesis of a cytisine-inspired pyridone library leading to the discovery of novel
150 Volume 2
Current Protocols in Chemical Biology
inhibitors of Bcl-2. Bioorg. Med. Chem. Lett. 19:2500-2503. Medina-Franco, J.L., Maggiora, G.M., Giulianotti, M.A., Pinilla, C., and Houghten, R.A. 2007. A similarity-based data-fusion approach to the visual characterization and comparison of compound databases. Chem. Biol. Drug Des. 70:393-412. Nam, N.-H., Sardari, S., and Parang, K. 2003. Reactions of solid-supported reagents and solid supports with alcohols and phenols through their hydroxyl functional group. J. Comb. Chem. 5:479-546. Nielsen, T.E. and Schreiber, S.L. 2008. Towards the optimal screening collection: A synthesis strategy. Angew. Chem. Int. Ed. 47:48-56. Oprea, T.I. and Gottfries, J. 2001. Chemography: The art of navigating in chemical space. J. Comb. Chem. 3:157-166. Radner, J.E., McPherson, O.M., Mazitschek, R., Barnes-Seeman, D., Shen, J.P., Dhaliwal, J., Stevenson, K.E., Duffner, J.L., Park, S.B., Neuberg, D.S., Nghiem, P., Schreiber, S.L., and Koehler, A.N. 2006. A robust small-molecule microarray platform for screening cell lysates. Chem. Biol. 13:493-504. Ryba, T.D., Depew, K.M., and Marcaurelle, L.A. 2009. Large scale preparation of siliconfunctionalized SynPhase polystyrene Lanterns for solid-phase synthesis. J. Comb. Chem. 11:110-116. Sauer, W.H.B. and Schwarz, M.K. 2003. Molecular shape diversity of combinatorial libraries: A prerequisite for broad bioactivity. J. Chem. Inf. Comput. Sci. 43:987-1003. Scott, P.J.H. 2009. Linker strategies in solid-phase organic synthesis. Wiley, Chichester, U.K. Tallarico, J.A., Depew, K.M., Pelish, H.E., Westwood, N.J., Lindsley, C.W., Shair, M.D., Schreiber, S.L., and Foley, M.A. 2001. An
alkylsilyl-tethered, high-capacity solid support amenable to diversity-oriented synthesis for one-bead, one-stock solution chemical genetics. J. Comb. Chem. 3:312-318. Taylor, S.J., Taylor, A.M., and Schreiber, S.L. 2004. Synthetic strategy toward skeletal diversity via solid-supported, otherwise unstable reactive intermediates. Angew. Chem. Int. Ed. 43:16811685. Verdi´e, P., Subra, G., Averland-Petit, M.-C., Amblard, M., and Martinez, J. 2008. Solidphase synthesis of 4-methylcarboxy-1,4benzodiazepine-2,5-diones. J. Comb. Chem. 10:869-874. Wang, X., Imber, B.S., and Schreiber, S.L. 2008. Small-molecule reagents for cellular pull-down experiments. Bioconjugate Chem. 19:585-587. Weinbrenner, S. and Tzschucke, C.C. 2006. PuriÞcation principles in high-speed solution-phase synthesis. In Methods and Principles in Medicinal Chemistry, Vol. 26: Combinatorial Chemistry (W. Bannwarth, B. Hinzen, eds.) pp. 1-31. Wiley-VCH, Weinheim, Germany. Wong, J.C., Hong, R., and Schreiber, S.L. 2003. Structural biasing elements for in-cell histone deacetylase paralog selectivity. J. Am. Chem. Soc. 125:5586-5587. Zajdel, P., Subra, G., Verdie, P., Gabzdyl, E., Bojarski, A.J., Duszy´nska, B., Martinez, J., and Pawłowski, M. 2009. Sulfonamides with the N-alkyl-N -dialkylguanidine moiety as 5HT7 receptor ligands. Bioorg. Med. Chem. Lett. 19:4827-4831.
Internet Resources http://www.mimotopes.com/knowledgeBase. asp?cid=26,34 An introduction to SynPhase Lanterns is provided on the Mimotopes Web site, as well as a variety of protocols for Lanterns with alternate linkers.
Library Synthesis on SynPhase Lanterns
151 Current Protocols in Chemical Biology
Volume 2
Preparation, Characterization, and Application of Optical Switch Probes Chutima Petchprayoon1 and Gerard Marriott1 1
Department of Bioengineering, University of California, Berkeley, California
ABSTRACT Optical switches represent a new class of molecular probe with applications in highcontrast imaging and optical manipulation of protein interactions. Small molecule, organic optical switches based on nitrospirobenzopyran (NitroBIPS) and their reactive derivatives and conjugates undergo efÞcient, rapid and reversible, orthogonal opticallydriven transitions between a colorless spiro (SP)-state and a colored merocyanine (MC)state. The excited MC-state also emits ßuorescence, which serves as the readout of the state of the switch. DeÞned optical perturbations of SP and MC generate a deÞned waveform of MC-ßuorescence that can be isolated against unmodulated background signals by using a digital optical lock-in detection approach or to control speciÞc dipolar interactions on proteins. The protocols describe general procedures for the synthesis and spectroscopic characterization of NitroBIPS and speciÞcally labeled conjugates along with methods for the manipulation of dipolar interactions on proteins and imaging of the C 2010 MC-state of NitroBIPS within living cells. Curr. Protoc. Chem. Biol. 2:153-169 by John Wiley & Sons, Inc. Keywords: optical switch r nitrospiropyran r imaging
INTRODUCTION Fluorescence microscopy is widely used to image changes in the distribution of speciÞc proteins and their complexes within living cells. However, ßuorescence imaging in living cells and tissue is challenging when imaging low levels of ßuorescently tagged proteins. In particular, the signal from endogenous ßuorescent molecules is often so high that the ßuorescent protein of interest must be present at such a high level that it leads to either complete inhibition or super-activation of the protein activity. Moreover, autoßuorescence signals, which amount to between 1 and 10,000 ßuorescein equivalents (Aubin, 1979), rule out conventional ßuorescence imaging of individual probe molecules in living cells. Various approaches have been used to reduce autoßuorescence, but these methods often involve chemical Þxation or cause cell stress, and there is currently no efÞcient method to reduce autoßuorescence signals within living, healthy cells. The authors have recently introduced a new imaging modality that allows for the isolation of speciÞc modulated ßuorescence signals from a new class of ßuorophore against unmodulated signals arising from endogenous ßuorophores (Marriott et al., 2008). The key to this approach is a probe or optical switch whose quantum yield for ßuorescence emission can be modulated between two different values. Optical switches are characterized by their ability to undergo rapid and reversible, high-Þdelity, orthogonal, optically-driven transitions between two states that have distinct structural or environmental and spectroscopic properties (Fig. 1; Sakata et al., 2005a,b). Synthetic optical switches based on nitrospirobenzopyran (NitroBIPS), a well-known photochrome (Fischer and Hirschberg, 1952), exist either as a pink colored, highly polarized and ßuorescent merocyanine (MC)-state or in a colorless and poorly polarized spiro (SP)-state (Fig. 1). The colorless SP-state NitroBIPS changes to a colored MC-state by illumination with UV light and this MC-state can be reverted to the colorless SP-state either thermally or by exposure to Current Protocols in Chemical Biology 2: 153-169, August 2010 Published online August 2010 in Wiley Interscience (www.interscience.wiley.com). DOI: 10.1002/9780470559277.ch100054 C 2010 John Wiley & Sons, Inc. Copyright
Optical Switch Probes
153 Volume 2
NO2 365 nm N
O
NO2
546 nm ⌬
spiropyran (SP) state
⫹
N
⫺
O
merocyanine (MC) state red fluorescence
Figure 1 Photochromic properties of benzospiropyran. The merocyanine (MC) state was generated by irradiation the spiropyran (SP) state with UV light (365 nm). While the excitation of the MC-state with visible light (546 nm) generated the SP-state.
visible light. The excitation of colored MC-state results in decay of its excited state by either photo-chemistry to the UV-absorbing state, or by returning to the MC-ground state with emission of red photon. This process can be repeated many times through orthogonal, optical control of transitions between the two states and occurs without the release of a photoproduct (Inouye, 1994; Sakata et al., 2005b). Recent studies have shown that the exposure of cells to short (50 to 100 msec) pulses of 365-nm light does not lead to any obvious deleterious effects on cell health (Marriott et al., 2008; Sakata et al., 2008). The quantum yields for optical switching of NitroBIPS depends on general and speciÞc solvent effects, with higher yields achieved in apolar solvents and lower-efÞciency transitions occurring in bulk water. Attaching NitroBIPS to proteins increases the quantum yield for switching compared to that attainable in bulk water and moreover, high-Þdelity transitions are most likely when a single NitroBIPS probe is attached to a unique site on the protein rather than random labeling, e.g., of lysine residues (Patolsky et al., 1998; Medintz et al., 2004). Highly selective labeling of probes on proteins is achieved through coupling of the probe to the thiol group of cysteine. Fortunately, most proteins contain few, if any, free thiol groups, and they can be introduced using site-directed mutagenesis. An alternative approach for speciÞc labeling of proteins with optical switches is through reactions with suicide substrates, and in special cases through reaction of amino groups on the N-terminus or ε-amino group of lysine residues. Here, protocols for the synthesis of reactive NitroBIPS probes and subsequent spectroscopic and photophysical characterization of the optical switching properties and ßuorescence emission of the protein conjugates are described. The protocols also show how NitroBIPS-derived probes and protein conjugates can be employed in conjunction with optical lock-in detection image microscopy to achieve high-contrast imaging of speciÞc proteins or cellular structures within living cells.
STRATEGIC PLANNING
Optical Switch Probes
Design of Optical Switch Probes The NitroBIPS probes used in these protocols are synthesized as described in Basic Protocols 1 and 2. The nitro group, introduced at position 6 in the BIPS molecule is used to shift the ground state equilibrium of the photochromic molecule toward the MC-state (Hirshberg and Fischer, 1954; Flennery, 1968; Song et al., 1995). NitroBIPS probes harboring an amino or thiol reactive functional group are synthesized with the aim of coupling to amino groups on the surface of micron-scale latex beads or to proteins (Sakata et al., 2005a). SpeciÞcally, the amino reactive optical switch, 1 ,3 ,3 -trimethyl-5 -(N-succinimidyloxycarbonylpropyl)-6-nitro-spiro(2Hbenzopyran-2,2 ,3H-indole) (5 -NHS ester of NitroBIPS, structure 5 in Fig. 2), is synthesized in several steps from commercially available 4-(4-aminophenyl)butanoic
154 Volume 2
Current Protocols in Chemical Biology
HO
O
a) NaNO2, SnCl2•2H2O, HCl O
NaOH O
NH2 b) 3-methyl-2-butanone,H2SO4, ethanol
N
HO
methanol/water
O
N
1
O
2
HO CH3l
HO
NO2 O
CH2Cl2
⫺N
ethanol, NEt3
3
HO
N -hydroxysuccinimide, DCC O
N O
NO2
THF
4
O N
O O
O
N O
NO2
5
Figure 2 indole).
Synthesis
of
1 ,3 ,3 -trimethyl-5 -(N-succinimidyloxycarbonylpropyl)-6-nitro-spiro(2H-benzopyran-2,2 ,3H-
acid and 5-nitrosalicylaldehyde, as described in Basic Protocol 1. On the other hand, the thiol-reactive optical switch, 1 ,3 ,3 -trimethyl-6-nitro-8-iodomethylspiro(2Hbenzopyran-2,2 ,3H-indole) (8-Iodo-NitroBIPS, structure 7 in Fig. 3), is readily synthesized in two steps from commercially available 1,3,3-trimethyl-2-methyleneindoline and 3-chloromethyl-5-nitrosalicylaldehyde, as described in Basic Protocol 2.
SYNTHESIS OF 1 ,3 ,3 -TRIMETHYL-5 -(N-SUCCINIMIDYLOXYCARBONYLPROPYL)-6-NITRO-SPIRO(2H-BENZOPYRAN-2,2 ,3H-INDOLE) (5 -NHS ESTER OF NitroBIPS)
BASIC PROTOCOL 1
1 ,3 ,3 -Trimethyl-5 -(N-succinimidyloxycarbonylpropyl)-6-nitro-spiro(2H-benzopyran-2,2 ,3H-indole) (5) is synthesized according to the scheme presented in Figure 2. First, the 2,3,3-trimethyl-5-(ethyloxycarbonylpropyl)-3H-indole (1) is prepared by diazotization of 4-(4-aminophenyl)butanoic acid with sodium nitrite in the presence of hydrochloric acid, followed by reduction with stannous chloride dihydrate to yield the hydrazine derivative. This product is subsequently condensed with 3-methyl-2butanone in the presence of sulfuric acid in ethanol. The hydroxylation of 1 with aqueous sodium hydroxide yields 2,3,3-trimethyl-5-(carboxypropyl)-3H-indole (2), which is subsequently subject to methylation with methyl iodide to yield 1,2,3,3tetramethyl-5-(carboxypropyl)-3H-indoleninium iodide (3). Condensation of 3 with 5-nitrosalicylaldehyde in ethanol in the presence of triethylamine yields 1 ,3 ,3 trimethyl-5 -(carboxypropyl)-6-nitro-spiro(2H-benzopyran-2,2 ,3H-indole) (4). Finally, treatment of 4 with N-hydroxysuccinimide in the presence of N, N -dicyclohexylcarbodiimide (DCC) affords the amino reactive N-hydroxysuccinimide ester 5.
Materials 4-(4-Aminophenyl)butanoic acid Concentrated hydrochloric acid (HCl) Sodium nitrite (NaNO2 ) Stannous chloride dihydrate (SnCl2 ·2H2 O) Anhydrous ethanol
Optical Switch Probes
155 Current Protocols in Chemical Biology
Volume 2
3-Methyl-2-butanone Sulfuric acid (H2 SO4 ) Saturated aqueous sodium carbonate Diethyl ether Anhydrous magnesium sulfate Methanol 1 M aqueous sodium hydroxide 1 M aqueous hydrochloric acid (HCl) Ethyl acetate Dichloromethane (DCM) Methyl iodide 5-Nitrosalicylaldehyde Triethylamine Silica gel 70 to 230 mesh Hexane N,N -Dicyclohexylcarbodiimide (DCC) N-Hydroxysuccinimide (NHS) Anhydrous tetrahydrofuran (THF) Nitrogen gas 10- and 25-ml round-bottomed ßasks Magnetic stir bar Stirring hotplate 1.5-ml microcentrifuge tubes 4-ml glass vials Filter paper B¨uchner funnel Vacuum pump Suction ßask Oil bath Condenser Rotary evaporator 0.5-in. diameter chromatography column with frit disc ◦ Silica gel thin layer chromatography (250 μm, 60 A with ßuorescent indicator UV254 ) 254/365-nm hand-held UV lamp Synthesis of 2,3,3-trimethyl-5-(carboxypropyl)-3H-indole (2) 1. Suspend 4-(4-aminophenyl)butanoic acid (100 mg, 0.558 mmol) in a 1:1 mixture of concentrated HCl and water (2 ml) in a 25-ml round-bottomed ßask. Stir 1 hr at 0◦ C in an ice bath using a stirring plate. 2. Weigh 42 mg of sodium nitrite (NaNO2 , 0.609 mmol) and place in a 1.5-ml microcentrifuge tube. Add 500 μl of water and pipet up and down or sonicate until a clear solution is obtained. Cool the solution on ice 30 min before proceeding to step 3. 3. Add, drop-wise, the sodium nitrite solution into the suspension obtained from step 1. Stir the mixture 1 hr at 0◦ C. The suspension will become a clear yellow solution and will generate gas bubbles.
4. While waiting for the reaction in step 3, weigh 377 mg of stannous chloride dihydrate (SnCl2 ·2H2 O, 1.670 mmol) and place into a 4-ml glass vial. Add 500 μl concentrated HCl and pipet up and down or sonicate until the solution becomes clear. Optical Switch Probes
156 Volume 2
Current Protocols in Chemical Biology
5. Add, drop-wise, the stannous chloride dihydrate solution into the reaction mixture from step 3. Stir 1 hr at 0◦ C and then Þlter through Þlter paper using a B¨uchner funnel and vacuum pump to yield a white, milk colored, solid product. When adding the stannous chloride dihydrate solution a white solid and gas bubbles are generated. The suspension should be cooled to precipitate the product. The suspension may be left for 2 hr at 4◦ C before Þltration.
6. Dissolve the crude product from step 5 in 5 ml of ethanol in a 25-ml round-bottomed ßask. Add 729 μl of 3-methyl-2-butanone (1.125 mmol) and 56 μl of concentrated sulfuric acid (H2 SO4 , 0.560 mmol). Stir under reßux 12 hr in an 80◦ C oil bath using a water-cooled condenser. 7. Concentrate the reaction mixture by evaporation under reduced pressure using a rotary evaporator. Add 10 ml of water, neutralize with saturated aqueous sodium carbonate to pH 6 to 7, and extract with 10 ml diethyl ether three times. Combine the organic extracts, dry over anhydrous magnesium sulfate, and then evaporate under reduced pressure to dryness. During neutralization, the product will precipitate to yield a yellow suspension.
8. Dissolve the crude product in 1 ml methanol in a 25-ml round-bottomed ßask. Add 5 ml of 1 M aqueous sodium hydroxide and stir 1 hr at room temperature. 9. Add 1 M aqueous HCl to the reaction mixture until a pH of ∼1 to 2 is reached. Extract the product with 10 ml ethyl acetate three times, combine the organic layers, and dry over anhydrous magnesium sulfate for 5 min. Evaporate under reduced pressure using a rotary evaporator yielding 98 mg of 2,3,3-trimethyl-5-(carboxypropyl)-3H-indole (2, 0.4 mmol).
Synthesize 1,2,3,3-tetramethyl-5-(carboxypropyl)-3H-indoleninium iodide (3) 10. Place the 98 mg of 2 (0.4 mmol) from step 9 into a 25-ml round-bottomed ßask and add 5 ml of DCM. Add 200 μl of methyl iodide (3.2 mmol) then stir and reßux 12 hr at 60◦ C in an oil bath using a water-cooled condenser apparatus. 11. Cool the reaction mixture to room temperature and evaporate to dryness in a rotary evaporator to yield a crude product 3, which is further dried using a vacuum pump.
Synthesize 1 ,3 ,3 -trimethyl-5 -(carboxypropyl)-6-nitro-spiro(2H-benzopyran2,2 ,3H-indole)(5 -carboxy-NitroBIPS, 4) 12. Dissolve the crude product 3 in 8 ml of anhydrous ethanol in a 25-ml round-bottomed ßask. 13. Add 67 mg of 5-nitrosalicylaldehyde (0.4 mmol) and 56 μl of triethylamine (0.4 mmol) into the solution from step 12. Then stir and reßux 2 hr in an 80◦ C oil bath under a water-cooled condensation apparatus. During reßux, the solution turns to a purplish red color.
14. Cool the reaction to room temperature and evaporate under reduced pressure using a rotary evaporator to dryness. 15. Dissolve the crude product in a small volume of DCM and load onto the top of a 0.5-in. diameter silica gel ßash column equilibrated in hexane. 16. Elute the column with a 6:4 mixture of hexane and ethyl acetate. Combine the pure 5-carboxy-NitroBIPS-containing fractions using TLC as a guide and evaporate the product to dryness.
Optical Switch Probes
157 Current Protocols in Chemical Biology
Volume 2
Synthesize 1 ,3 ,3 -trimethyl-5 -(N-succinimidyloxycarbonylpropyl)-6-nitrospiro(2H-benzopyran-2,2 ,3H-indole) (5-NHS ester of NitroBIPS, 5) 17. Place 122 mg of 4 (0.3 mmol), 93 mg of DCC (0.45 mmol), and 52 mg of NHS (0.45 mmol) in a 10-ml round-bottomed ßask. Glassware and starting material should be dried for a more efÞcient reaction.
18. Add 5 ml of anhydrous THF and stir the mixture overnight under N2 gas at room temperature. 19. Filter off the precipitate using a B¨uchner funnel and Þlter paper. Evaporate the Þltrate to dryness to yield 123 mg of 5 -NHS ester of NitroBIPS (5; 0.243 mmol).
SYNTHESIS OF 1 ,3 ,3 -TRIMETHYL-6-NITRO-8-IODOMETHYLSPIRO(2HBENZOPYRAN-2,2 ,3H-INDOLE) (8-IODO-NitroBIPS, 7)
BASIC PROTOCOL 2
This compound is prepared by a halogenation exchange reaction of 1 ,3 ,3 -trimethyl6-nitro-8-chloromethylspiro(2H-benzopyran-2,2 ,3H-indole), which is synthesized by condensation of commercially available 1,3,3-trimethyl-2-methyleneindoline and 3chloromethyl-5-nitrosalicylaldehyde. The synthetic scheme for structure 7 is presented in Figure 3.
Materials 1,3,3-Trimethyl-2-methyleneindoline 3-Chloromethyl-5-nitrosalicylaldehyde Anhydrous tetrahydrofuran (THF) Acetone Potassium iodide (KI) Nitrogen gas Dichloromethane (DCM) Silica gel 70 to 230 mesh Hexane Ethyl acetate 25-ml round-bottomed ßasks Magnetic bar Stirring hotplate Fume hood Oil bath (70◦ C) Rotary evaporator
O HO NO2 Cl N
Kl THF
6
Figure 3
NO2
N O
Cl
acetone
NO2
N O
7
l
Synthesis of 1 ,3 ,3 -trimethyl-6-nitro-8-iodomethylspiro(2H-benzopyran-2,2 ,3H-indole).
Optical Switch Probes
158 Volume 2
Current Protocols in Chemical Biology
Condenser Filter paper B¨uchner funnel Suction ßask Vacuum pump 0.5-in. diameter chromatography column with frit disc ◦ Silica gel thin-layer chromatography (250 μm, 60 A with ßuorescent indicator UV254 ) 254/365-nm hand-held UV lamp Synthesize 1 ,3 ,3 -trimethyl-6-nitro-8-chloromethylspiro(2H-benzopyran-2,2 , 3H-indole) (8-cloro-NitroBIPS, 6) 1. Place 177 μl of 1,3,3-trimethyl-2-methyleneindoline (1 mmol) and 216 mg of 3chloromethyl-5-nitrosalicylaldehyde (1 mmol) into a 25-ml round-bottomed ßask. 2. Add 10 ml of anhydrous THF and stir 2 hr under reßux in a 70◦ C oil bath. 3. Cool the reaction to room temperature (∼20 to 30 min) and evaporate under reduced pressure using a rotary evaporator to dryness to give a crude product 6.
Synthesize 1 ,3 ,3 -trimethyl-6-nitro-8-iodomethylspiro(2H-benzopyran-2,2 , 3H-indole) (8-iodo-NitroBIPS, 7) 4. Dissolve the crude product 6 in 10 ml of acetone in a 25-ml round-bottomed ßask. 5. Add 664 mg of anhydrous potassium iodide (4 mmol) to the solution and stir under N2 gas overnight at room temperature. 6. Filter off the precipitate using Þlter paper and a B¨uchner funnel. Evaporate the Þltrate to dryness. 7. Dissolve the crude product from step 6 in a small volume of DCM and load on top of a 0.5-in. diameter silica gel ßash column equilibrated with hexane. 8. Elute the column with a 10:1 mixture of hexane and ethyl acetate. Combine the pure 8-iodo-NitroBIPS-containing fractions using TLC as a guide and evaporate the product to dryness under reduced pressure using a rotary evaporator to give 370 mg of 7 (0.8 mmol).
ABSORPTION CHARACTERIZATION OF NitroBIPS The optical switch NitroBIPS undergoes rapid and reversible, optically-driven transitions between the SP- and MC-states. These two states have different UV-visible absorption spectra. Only the MC-state of NitroBIPS has an absorption band in the visible region. The SP to MC transition is generated upon irradiation of the SP-state with near ultraviolet light (365 nm), while the MC- to SP-state is generated upon excitation of the MC-state with visible light (546 nm). The following protocol describes a method to measure the absorption spectra of NitroBIPS in the SP- and MC-states.
BASIC PROTOCOL 3
Materials 5 mM 1 ,3 -Dihydro-1 ,3 ,3 -trimethyl-6-nitrospiro(2H-1-benzopyran-2,2 -(2H)indole) (NitroBIPS) stock solution in methanol (Sigma-Aldrich, see recipe) Methanol 1.5-ml microcentrifuge tubes 2- or 4-window quartz cuvettes (Hellma) Spectrophotometer (Shimadzu PC1601) Hand-held UV lamp (365 nm) 530-nm output of an LED or 546-nm output of a 100-Watt mercury arc lamp
Optical Switch Probes
159 Current Protocols in Chemical Biology
Volume 2
Prepare NitroBIPS solution for absorption measurement 1. Prepare 1 ml of 50 μM NitroBIPS solution in methanol by adding 10 μl of 5 mM NitroBIPS stock solution to 990 μl of methanol in a 1.5-ml microcentrifuge tube. Mix well by repeated pipetting. The 1-ml volume is suitable for a semimicro-cuvette. In case of standard or macro cuvette (1 × 1 × 4–cm), 3 ml of NitroBIPS solution will be required.
Absorption measurement 2. Warm up deuterium and tungsten lamps for ∼15 min. 3. Set the wavelength range from 250 to 750 nm in fast-scan mode. 4. Fill both cuvettes (sample and reference) with methanol and record a reference spectrum over the same wavelength range (250 to 750 nm). 5. Fill the sample cuvette with a freshly prepared NitroBIPS solution. The thermodynamically stable form of NitroBIPS in methanol is the SP-state. Fill the reference cuvette with methanol. Measure the absorption spectrum of the SP-state of NitroBIPS between 250 and 750 nm. If the absorption peaks of interest are >2 absorbance units, dilute the sample and measure again. Ideally, the maximum absorption should be ∼0.3.
6. Irradiate the sample cuvette with 365-nm light for 1 min using a handheld UV-lamp. This will convert the SP-state of NitroBIPS to the MC-state.
7. Immediately record the absorption spectrum of the MC-state of NitroBIPS. 8. Irradiate the sample from step 7 with visible light (530 nm or 546 nm) for 1 min to convert the MC-state of NitroBIPS to the SP-state. 9. Record the absorption spectrum of the photogenerated SP-state of NitroBIPS. 10. Repeat steps 6 to 9 to demonstrate high-Þdelity switch between the SP- and MCstates of NitroBIPS in methanol. Figure 4 shows the absorption spectra of NitroBIPS in the SP- and MC-states in methanol during a single cycle of optical switching.
0.7 0.6
MC-state SP-state
Absorbance
0.5 0.4 0.3 0.2 0.1 0 250
350
450 550 Wavelength (nm)
650
750
Figure 4 UV absorption spectra of NitroBIPS in the SP- and MC-states in methanol during a single cycle of optical switching. Optical Switch Probes
160 Volume 2
Current Protocols in Chemical Biology
FLUORESCENCE CHARACTERIZATION OF NitroBIPS NitroBIPS has two distinct and optically interconvertible states, the colorless SP-state and the colored MC-state. Excitation of MC with 546-nm light generates the MC-excited state, which decays by one of two independent processes. The Þrst process is the excitedstate intramolecular reaction that reforms the SP-state and proceeds with a quantum yield as high as 0.8 in apolar solvents (Hobley et al., 2002). The competing process involves the return of the excited MC to the same MC-ground state with emission of a red photon at ∼618 nm. This reaction has a low quantum yield of between 0.05 and 0.1 (Chibisov and G¨orner, 1997). The following protocol describes the method used to record the ßuorescence emission spectrum of MC-NitroBIPS.
BASIC PROTOCOL 4
Materials 5 mM 1 ,3 -Dihydro-1 ,3 ,3 -trimethyl-6-nitrospiro(2H-1-benzopyran-2,2 -(2H)indole) (NitroBIPS) stock solution in methanol (Sigma-Aldrich, see recipe) Methanol 4-Window quartz cuvettes (Hellma) Fluorescence spectrophotometer (SLM-AB2) Hand-held UV lamp (365 nm) 530-nm output of an LED or 546-nm output of a 100-Watt mercury arc lamp Prepare NitroBIPS solution for ßuorescence spectrum measurement 1. Prepare a sample solution as detailed in Basic Protocol 2, step 1. Only the MC-state of NitroBIPS has a red ßuorescence emission.
2. Warm up the xenon arc lamp in the ßuorescence spectrophotometer for ∼15 min.
Create emission spectrum 3. Find the maximum visible absorption wavelength of NitroBIPS using Basic Protocol 3, steps 2 to 7 (close to 530 nm). 4. Fix the excitation wavelength at maximum MC-absorption as indicated in step 3 (530 nm) and scan the ßuorescence emission between 400 and 750 nm to obtain a better idea of the maximum emission wavelength. 5. Measure the ßuorescence spectrum of freshly prepared MC-state of NitroBIPS. To avoid the so-called inner Þlter effect, it will be necessary to decrease the absorption of the MC-state at 530 nm to an absorption value that is <0.2. Irradiation of the NitroBIPS solution with 365 nm light for ∼1 min will generate the MC-state of NitroBIPS. Record the ßuorescence intensity immediately. If the spectrum is off-scale, adjust the sensitivity (voltage) and measure the MC-ßuorescence intensity again.
6. Identify the emission wavelength that gives the maximum ßuorescence intensity ∼610 to 620 nm.
Obtain excitation spectrum 7. Fix the emission wavelength at ∼620 nm, or as determined from step 6. Record the excitation spectrum of the MC-state of NitroBIPS between 400 and 600 nm. 8. Identify the maximum excitation wavelength from the spectrum, ∼530 nm. In some cases, the excitation spectrum may need to be corrected to account for variation in the energy of the mercury arc lamp over the range of the excitation wavelengths. Optical Switch Probes
161 Current Protocols in Chemical Biology
Volume 2
90
Fluorescence intensity
80
MC-state SP-state
70 60 50 40 30 20 10 0 550
Figure 5
600
650 Wavelength (nm)
700
750
Fluorescence spectra of NitroBIPS in methanol in the SP- and MC-states.
Obtain ßuorescence emission spectrum 9. Fix the excitation and emission wavelengths from the previous experiments (excitation at 535 nm and emission at 630 nm). 10. Collect ßuorescence spectrum of NitroBIPS between 550 and 750 nm. Figure 5 shows the ßuorescence emission spectrum of NitroBIPS in methanol. BASIC PROTOCOL 5
MICROSCOPIC CHARACTERIZATION OF NitroBIPS The 5 -NHS ester of NitroBIPS (5) was designed to covalently link NitroBIPS to aminoderivatized latex beads, and proteins through their lysine residues, for microscope-based imaging for the characterization of optical switching between the SP- and MC-states. The red MC-ßuorescence is used to image the NitroBIPS-labeled bead and to record the kinetics of transitions between the SP- and MC-states.
Materials 3-μm Polybead amino microspheres (Polysciences) 1× phosphate buffer saline (PBS; see recipe) 5 mM 5 -NHS ester of NitroBIPS stock solution in DMF (see recipe) Immersion oil 1.5-ml microcentrifuge tube Glass-bottom petri dish: 35-mm diameter incorporating a 14-mm glass coverslip, no. 0 thickness (Mat Tek) 100-Watt mercury arc lamps EMCCD camera (Hamamatsu) Inverted ßuorescence microscope (Zeiss) adapted for pulse probe microscopy (see Fig. 6) Computer-controlled mechanical shutter (Melles Griot) Interference Þlter for excitation of Cy3 emission UV-pass Þlter (UG11, Schott glass) Prepare NitroBIPS-labeled latex beads 1. Stir a suspension of amino-derivatized latex beads and add 1 drop into a 1.5-ml microcentrifuge tube. Add 200 μl of PBS and suspend by pipetting up and down. Optical Switch Probes
2. Add 10 μl of 5 mM 5 -NHS ester of NitroBIPS stock solution to suspension of beads. Mix well by pipetting up and down or vortexing. 3. Stir the sample 2 hr at 4◦ C.
162 Volume 2
Current Protocols in Chemical Biology
EM-CCD
visible light source (546 nm) 365-nm flash lamp
Figure 6 Schematic represent the inverted fluorescence microscope (Zeiss) adapted for pulseprobe microscopy.
4. Centrifuge 5 min at 10,000 × g, room temperature. Discard supernatant and resuspend pellet in 500 μl of PBS. This step is used to remove free, unbound 5 -NHS ester of NitroBIPS.
5. Repeat step 4 four additional times. Resuspend the Þnal pellet with 200 μl of PBS. 6. Place 2 μl of the NitroBIPS-labeled beads suspension on the coverslip glass petri dish.
Microscope-based optical switch NitroBIPS-labeled latex beads 7. Turn on the microscope lamps and EMCCD camera 15 min before use. 8. With reference to Figure 6, use an in-plane 100-W Hg-arc lamp with a closely associated 546-nm interference Þlter for continuous excitation of the sample with 546-nm light. Remove the Cy3 excitation Þlter from the Þlter cube set. 9. Use a second Hg-arc lamp with integrated UG11 Þlter and shutter, orthogonally aligned to the 546-nm excitation light path, to deliver pulses of 365-nm light to the sample. Direct the 365-nm light into the excitation optical path using a dichroic mirror from a ßuorescein Þlter set aligned 45◦ with respect the excitation light path. Set the shutter controls to deliver pulses of 365-nm light to the sample with 100-msec duration every 1 sec. 10. Use a 100× oil immersion objective and immersion oil to image the MC-ßuorescence from beads. 11. Adjust the coarse focusing knob to view beads bound to the surface of the coverslip and use the Þne focus knob to generate a sharper image of the immobilized beads. During focusing, both UV light and green light must be turned on to drive the SP to MC transition and to elicit emission of red ßuorescence from the MC-state.
12. Record the MC-ßuorescence intensity from individual beads on the EMCCD camera at video rate and store the images as a single movie Þle. 13. Initiate recording of the MC-ßuorescence movie in concert with activating the mechanical shutter over at least Þve cycles. Set the camera frame rate at 33 frames/sec or as fast as possible.
Analyze data 14. Select the area of interest from beads in the movie Þle recorded from step 13. Use the integrated software on the camera or a dedicated program to measure the average intensity of MC-ßuorescence within selected areas for each frame in the movie. Current Protocols in Chemical Biology
Optical Switch Probes
163 Volume 2
B Fluorescence intensity
A
7000 MC-state
6500 6000 5500 5000
SP-state
4500 4000 0
1
2
3
4
5 6 Time (sec)
7
8
9
10
Figure 7 (A) Fluorescence intensity image of NitroBIPS-labeled beads. (B) The MC-fluorescence change corresponding to optical switch cycles of NitroBIPS in individual bead (red circle in A) over eight cycles of optical switching.
15. Export the measured data, plot and analyze the average MC-ßuorescence intensity as a function of time using an Excel spreadsheet or other graphic software. When using a 100-msec UV pulse, the Þrst few frames collected at 33 frames/sec that follow the 365-nm pulse may be buried within the signal arising from the tail of the 365-nm pulse. These overlap frames can usually be deleted without affecting the analysis. Figure 7 shows the ßuorescence intensity changes corresponding to optical switch cycles of NitroBIPS in individual beads over eight cycles of optical switching. BASIC PROTOCOL 6
ACTIN-LABELED 8-IODO-NitroBIPS The thiol-reactive optical switch, 8-iodo-NitroBIPS (7), was designed for coupling NitroBIPS to proteins via their cysteine residues. In this protocol, 8-iodoNitroBIPS is conjugated to G-actin via alkylation of cysteine-374. The conjugate is puriÞed from free probe using size-exclusion chromatography. The process for coupling and puriÞcation of the NitroBIPS conjugate of G-actin is described below.
Materials PuriÞed lyophilized actin (Cytoskeleton) G-buffer (see recipe) G-buffer with 1 mM DTT (see recipe) 5 mM 8-iodo-NitroBIPS stock solution in DMF (see recipe) 1.5-ml microcentrifuge tubes Vortexer PD-10 column (GE Healthcare) Spectrophotometer Prepare G-actin conjugated 8-iodo-NitroBIPS 1. Dissolve 1 mg of actin powder in 500 μl of G-buffer and transfer to a 1.5-ml microcentrifuge tube. Add 660 μl of G-buffer and mix well by gentle pipetting to generate a 20 μM solution of G-actin. Optical Switch Probes
2. Pipet 500 μl of the G-actin solution from step 1 to a new 1.5-ml microcentrifuge tube, add 4 μl of 5 mM 8-iodo-NitroBIPS stock solution and mix rapidly and thoroughly
164 Volume 2
Current Protocols in Chemical Biology
0.5 0.45
MC-state SP-state
0.4 Absorbance
0.35 0.3 0.25 0.2 0.15 0.1 0.05 0 280
Figure 8
330
380
430 480 530 Wavelength (nm)
580
630
680
UV absorption spectra of NitroBIPS-G-actin conjugate in the SP- and MC-state.
by vortexing for a few seconds. Incubate the reaction 1 hr at room temperature in the dark. The reaction can also proceed overnight at 4◦ C in the dark.
Purify NitroBIPS-labeled G-actin 3. Gently apply the solution (500 μl) from step 2 to the top of a PD-10 column equilibrated in G-buffer containing 1 mM DTT. 4. Elute the labeled conjugate with G-buffer containing 1 mM DTT. Collect 1-ml fractions and use absorption at 350 nm to identify the protein conjugate. Pool the fractions containing the conjugate. For a PD-10 column, the protein will usually elute after 2 ml and is complete by 5 ml. The pool of eluted conjugate is usually diluted about two-fold compared to the applied volume.
Create absorption spectroscopy of the G-actin conjugate 5. Follow the same procedure detailed in Basic Protocol 3, steps 2 to 9. Figure 8 shows the absorption spectrum of NitroBIPS-G-actin conjugate in the SP- and MC-state.
LIVE-CELL IMAGING OF NitroBIPS Measurements of MC-ßuorescence intensity may be used to quantify the state of the NitroBIPS optical switch, i.e., the percentage of SP- and MC-state in the sample at any time, while providing a sensitive signal to image the distribution of the MC-state of NitroBIPS in cells. This protocol describes how to record and image the MC-ßuorescence of NitroBIPS in living cells.
BASIC PROTOCOL 7
Materials HEK-293 cells Cell culture medium (see recipe) 5 mM 8-iodo-NitroBIPS stock solution in DMF (see recipe) Glass-bottom petri dish: 35-mm diameter incorporating a 14-mm glass coverslip, no. 0 thickness (Mat Tek Corporation) 2-ml microcentrifuge tubes 37◦ C, 5% CO2 humidiÞed incubator Tissue culture hood
Optical Switch Probes
165 Current Protocols in Chemical Biology
Volume 2
Inverted ßuorescence microscope (Zeiss) adapted for pulse probe microscopy (see Fig. 6) 100-Watt mercury arc lamps or laser power source Computer-controlled mechanical shutter (Melles Griot) Interference Þlter for excitation of Cy3 emission UV-pass Þlter (UG11, Schott glass) EMCCD camera (Hamamatsu) Culture cells 1. Grow HEK-293 cells in a glass-bottom petri dish with a Þnal cell culture medium volume of 2 ml. Check the cell density every day until 60% to 80% conßuent. Prepare NitroBIPS sample 2. Pipet 2 ml of cell culture medium into a sterile 2-ml microcentrifuge tube. 3. Pipet 4 μl of 5 mM 8-iodo-NitroBIPS stock solution into the 2-ml microcentrifuge tube from step 2 (the Þnal concentration is 10 μM). Mix well by pipetting up and down.
Load NitroBIPS into living HEK 293 cell 4. Remove the old cell culture medium by aspiration or pipetting. 5. Rapidly add the new cell culture medium containing 8-iodo-NitroBIPS from step 3. 6. Incubate cells 1 hr in a 37◦ C, 5% CO2 humidiÞed incubator. 7. Remove the old cell culture medium and add 2 ml of fresh cell culture medium. 8. Repeat step 7 three additional times to remove any free 8-iodo-NitroBIPS.
Image living cell 9. Follow the procedure detailed in Basic Protocol 5, steps 7 to 13. Analyze data 10. Follow the procedure detailed in Basic Protocol 5, steps 14 and 15. Figure 9 shows the MC-ßuorescence intensity change in living HEK-293 cells during several cycles of optical switching.
A
B Fluorescence intensity
5400 MC-state
5300 5200 5100 5000 4900 4800
SP-state
4700 0
1
2
3
4 5 Time (sec)
6
7
8
Figure 9 (A) Fluorescence intensity image of NitroBIPS in live HEK-293 cell. (B) The MC-fluorescence change of NitroBIPS in a region of interest (red circle in A) in live HEK-293 cell during optical switch cycles. Optical Switch Probes
166 Volume 2
Current Protocols in Chemical Biology
REAGENTS AND SOLUTIONS Use Milli-Q water in all recipes and protocol steps.
Cell culture medium 10% fetal bovine serum 100 U/ml penicillin G sodium 100 μg/ml streptomycin 4 mM L-glutamine Add to 500 ml Dulbecco’s modiÞed Eagle’s medium (DMEM) Store up to 2 months at 4◦ C and warm to 37◦ C before use G-buffer 5 mM Tris base 0.2 mM CaCl2 0.2 mM ATP Dissolve in 1000 ml water and adjust pH to 8.0 Store up to 1 month at 4◦ C G-buffer with DTT 1 mM DTT Dissolve in G-buffer (see recipe) Adjust pH to 8.0 with 1 M NaOH or 1 M HCl Store up to 1 month at 4◦ C 8-Iodo-NitroBIPS stock solution, 5 mM 1 mg 8-iodo-NitroBIPS Dissolve in 432 μl DMF Store up to 1 month at −20◦ C 5 -NHS ester of NitroBIPS stock solution, 5 mM 1 mg 5 -NHS ester of NitroBIPS Dissolve in 395 μl DMF Store up to 1 month at −20◦ C NitroBIPS stock solution, 5 mM 1 mg NitroBIPS Dissolve in 620 μl methanol Store up to 1 month at −20◦ C Phosphate buffered saline, 1× 8 g sodium chloride 0.2 g potassium chloride 1.44 g sodium phosphate dibasic (Na2 HPO4 ) 0.24 g potassium phosphate monobasic (KH2 PO4 ) Dissolve in 800 ml water Adjust pH to 7.4 with HCl Add water to 1000 ml Store up to 1 year at room temperature
Optical Switch Probes
167 Current Protocols in Chemical Biology
Volume 2
COMMENTARY
Optical Switch Probes
Background Information
Critical Parameters
A major challenge in cell biology is to elucidate molecular mechanisms that underlie the spatio-temporal control of cell processes and behavior. Since signaling pathways are often initiated by a small number of receptor proteins and conÞned to speciÞc loci in the cell, these studies require microscope techniques and development of new probes that can generate high-contrast and high-resolution images of changes in the distribution and interactions of speciÞc proteins in a cell (Yan and Marriott, 2003; Betzig et al., 2006). Some of these imaging techniques employ optical switch reagents. Nitrospirobenzopyran (NitroBIPS) is an optical switch that undergoes rapid and reversible, optically driven deterministic transitions between an SP- and MC-state (Seto, 1990; Bertelson, 1999). The SP to MC transition is triggered by exciting the SP-state with near-ultraviolet light from 320 to 420 nm and has a high two-photon absorption crosssection, which peaks at 720 nm. The MC-state, on the other hand, has an absorption maximum that is sensitive to its molecular environment and ranges from 485 nm to 585 nm (Sakata et al., 2005a,b; Mao et al., 2008). Excitation of the MC-state of NitroBIPS with 546-nm light generates the MC-excited state that can decay via two independent processes, each with a unique quantum yield. The Þrst process is an excited state intramolecular reaction that reforms the SP-state and has a high quantum yield (Hobley et al., 2002), whereas the second process involves a return to the ground state with the emission of a red photon with a low quantum yield between 0.05 and 0.1 (Chibisov and G¨orner, 1997). However, even though weak, the MC-ßuorescence of NitroBIPS is useful for imaging and analysis of the progress of optical switching in cells and tissue (Mao et al., 2008; Marriott et al., 2008; Sakata et al., 2008). Other optical switches, such as spironaphthoxazine (NISO), have MCstate absorption spectra that allow for orthogonal control of two switch probes (Sakata et al., 2005a). The modulated signal from the optical switch is truly unique and can be isolated from all other unmodulated signals such as autoßuorescence using an optical lock-in detection (OLID) approach developed in Marriott et al. (2008). The authors are currently improving the ßuorescence quantum yield of optical switch probes to increase the sensitivity of OLID microscopy to the detection level of single molecules.
The puriÞcation process using silica gel column chromatography should be done as quickly as possible to avoid complications from the acidity of silica gel, light, and the solvent. In particular, these effects can generate the MC-state of NitroBIPS, which is retained in the column and decreases the yield. To ensure a high yield of product, organic solvents should be anhydrous grade while performing the reaction under an atmosphere of nitrogen gas to reduce oxygen and moisture. All glassware used for these reactions should be ovendried. For microscope-based characterization of optical switching of NitroBIPS, it is best to use a sub-saturating level of illumination. A 100-msec UV pulse in this experiment is used to convert a sub-population of NitroBIPS from the SP- to MC-state; however, complete conversion of molecules in the image Þeld can be realized by using a longer irradiation time and higher power. The proÞle of the MCßuorescence intensity decay depends on the energy of the 546-nm illumination energy. The higher the illumination energy, the faster the MC-ßuorescence will decay.
Troubleshooting When characterizing the optical switching efÞciency of NitroBIPS using MCßuorescence, some attention should be given to improving the ßuorescence signal from MC through control of the illumination energy of the 365-nm and 546-nm light. In general, one wants to generate a maximum amount of MC using 365-nm illumination in the shortest period of time. On the other hand, it is necessary to record several data points to quantify the decay of MC-ßuorescence using the 546-nm light. If the energy of the 546-nm light is too high, then the MC-state will rapidly convert to the SP-state, i.e., faster than the video capture rate. In this case it is possible to reduce the energy of the 546-nm light either by using the light level control switch on the lamp or by inserting a neutral density Þlter in the 546-nm light path.
Anticipated Results For Basic Protocol 1, step 5, if the suspension is not cooled, then one can expect a lower yield of product. If the puriÞcation of the product using silica gel is rapid, then one can expect yields for 5-carboxy-NitroBIPS (4) and 8-iodo-NitroBIPS of ∼80%. If the glassware,
168 Volume 2
Current Protocols in Chemical Biology
solvent, and starting material are dry, the yield of 5 -NHS ester of NitroBIPS (5) is usually close to 100%.
Flennery, J.B. Jr. 1968. The photo- and thermochromic transients from substituted 1 , 3 , 3 -trimethylindolinobenzospiropyrans. J. Am. Chem. Soc. 90:5660-5671.
Time Considerations
Hirshberg, Y. and Fischer, E. 1954. Photochromism and reversible multiple internal transitions in some spiropyrans at low temperatures. Part II. J. Chem. Soc. 297:3129-3137.
The synthesis of 5 -NHS ester of NitroBIPS requires ∼4 days, with one overnight reaction for preparation of indole, an overnight reaction for the methylation, and an overnight reaction for the preparation of the NHS derivative. The synthesis of 8-iodo-NitroBIPS requires 1 day with an overnight reaction for the halogenation exchange. The coupling of 5 -NHS ester of NitroBIPS to amino-latex beads requires 2 hr at room temperature, but this could be lengthened to increase the labeling ratio. Characterization of the optical switch in solution can be completed in 1 day. Characterization of the optical switch in the microscope including optimization of light levels can also be completed within 1 day. Labeling of proteins with 8-iodo-NitroBIPS and puriÞcation of the conjugate can be accomplished within 1 day. Growing HEK-293 cells to ∼60% to 80% conßuence requires 1 to 3 days. The procedure to load and image 8-iodo-NitroBIPS within cells can be carried out within 1 day.
Acknowledgements The authors would like to acknowledge National Institutes of Health for supporting this work (grant nos. NIH R01EB005217 and R01GM086233).
Literature Cited
Hobley, J., Pfeifer-Fukumura, U., Bletz, M., Asahi, T., Masuhara, H., and Fukumura, H. 2002. Ultrafast photo-dynamics of a reversible photochromic spiropyran. J. Phys. Chem. A 106:2265-2270. Inouye, M. 1994. Spiropyran derivatives as multifunctional artiÞcial receptors for biologically important species. Mol. Cryst. Liq. Cryst. Sci. Tech. 246:169-172. Mao, S., Benninger, R.K.P., Yan, Y., Petchprayoon, C., Jackson, D., Easley, C.J., Piston, D.W., and Marriott, G. 2008. Optical lock-in detection of FRET using synthetic and genetically encoded optical switches. Biophys. J. 99:45154524. Marriott, G., Mao, S., Sakata, T., Ran, J., Jackson, D.K., Petchprayoon, C., Gomez, T.J., Erica, W., Tulyathan, O., Aaron, H.L., Isacoff, E.Y., and Yan, Y. 2008. Optical lock-in detection imaging microscopy for contrast-enhanced imaging in living cells. Proc. Natl. Acad. Sci. U.S.A. 105:17789-17794. Medintz, I.L., Trammell, S.A., Mattoussi, H., and Mauro, J.M. 2004. Reversible modulation of quantum dot photoluminescence using a protein-bound photochromic ßuorescence resonance energy transfer acceptor. J. Am. Chem. Soc. 126:30-31. Patolsky, F., Filanovsky, B., Katz, E., and Willner, I. 1998. Photoswitchable antigen-antibody interactions studied by impedance spectroscopy. J. Phys. Chem. B 102:10359-10367.
Aubin, J.E. 1979. Autoßuorescence of viable cultured mammalian cells. J. Histochem. Cytochem. 27:36-43.
Sakata, T., Yan, Y., and Marriott, G. 2005a. Family of site-selective molecular optical switches. J. Org. Chem. 70:2009-2013.
Bertelson, R.C. 1999. Spiropyran. In Organic Photochromic and Thermochromic Compounds, Vol. 1: Photochromic Families (J.C. Crano and R.J. Guglielmetti, eds.) pp. 85-109. Plenum Press, New York.
Sakata, T., Yan, Y., and Marriott, G. 2005b. Optical switching of dipolar interactions on proteins. Proc. Natl. Acad. Sci. U.S.A. 102:4759-4764.
Betzig, E., Patterson, G.H., Sougrat, R., Lindwasser, O.W., Olenych, S., Bonifacino, J.S., Davidson, M.W., Lippincott-Schwartz, J., and Hess, H.F. 2006. Imaging intracellular ßuorescent proteins at nanometer resolution. Science 15:1642-1645.
Sakata, T., Jackson, D.K., Mao, S., and Marriott, G. 2008. Optically switchable chelates: Optical control and sensing of metal ions. J. Org. Chem. 73:227-233. Seto, J. 1990. Photochromic dyes. In Infrared Absorbing Dyes (M. Matsuoka, ed.) pp. 71-88. Plenum Press, New York.
Chibisov, A.K. and G¨orner, H. 1997. Singlet versus triplet photoprocesses in indodicarbocyanine dyes and spiropyran-derived rnerocyanines. J. Photochem. Photobiol. A 105:261-267.
Song, X., Zhou, J., Li, Y., and Tang, Y. 1995. Correlations between solvatochromism, Lewis acid-base equilibrium and photochromism of an indoline spiropyran. J. Photochem. Photobiol. A 92:99-103.
Fischer, E. and Hirshberg, Y. 1952. Formation of coloured forms of spirans by low-temperature irradiation. J. Chem. Soc. 74:4522.
Yan, Y. and Marriott, G. 2003. Analysis of protein interactions using ßuorescence technologies. Curr. Opin. Chem. Biol. 7:635-640. Optical Switch Probes
169 Current Protocols in Chemical Biology
Volume 2
Experimental Design Considerations for In Vitro Non-Natural Glycan Display via Metabolic Oligosaccharide Engineering Elaine Tan,1 Ruben T. Almaraz,1 Hargun S. Khanna,1 Jian Du,1 and Kevin J. Yarema1 1
Department of Biomedical Engineering, The Johns Hopkins University, Baltimore, Maryland
ABSTRACT Metabolic oligosaccharide engineering (MOE) refers to a technique where non-natural monosaccharide analogs are introduced into living biological systems. Once inside a cell, these compounds intercept a targeted biosynthetic glycosylation pathway and in turn are metabolically incorporated into cell-surface-displayed oligosaccharides where they can modulate a host of biological activities or be exploited as “tags” for bio-orthogonal and chemoselective ligation reactions. Undertaking a MOE experiment can be a daunting task based on the growing repertoire of analogs now available and the ever increasing number of metabolic pathways that can be targeted; therefore, a major emphasis of this article is to describe a general approach for analog design and selection and then provide protocols to ensure safe and efficacious analog usage by cells. Once cell-surface glycans have been successfully remodeled by MOE methodology, the stage is set for probing changes to the myriad cellular responses modulated by these versatile molecules. Curr. C 2010 by John Wiley & Sons, Inc. Protoc. Chem. Biol. 2:171-194 Keywords: sialic acid glycoengineering r glycosylation r ManNAc analogs r selectin-based adhesion
INTRODUCTION Metabolic oligosaccharide engineering—MOE; also referred to as “metabolic glycoengineering” (Du et al., 2009)—is a technology developed in the early 1990s by Werner Reutter’s group, who discovered that certain structural variants of ManNAc (Kayser et al., 1992) can be taken up by the sialic acid biosynthetic pathway and utilized for oligosaccharide biosynthesis (Fig. 1). As a result, the cell-surface carbohydrates can be endowed with novel structural and chemical features. In theory, the ability to alter the repertoire of sialic acids on the cell surface provides a foundation to manipulate virtually any cellular function modulated by these versatile and multifaceted molecules (Campbell et al., 2007; Aich and Yarema, 2008; Wang et al., 2009; Du and Yarema, 2010). Importantly, the incorporation of bio-orthogonal chemical functionalities such as the ketone (Mahal et al., 1997), azide (Saxon and Bertozzi, 2000), or thiol (Sampathkumar et al., 2006c) groups into the glycocalyx opened the door to the conjugation with complementary probes for applications such as drug and gene delivery (Lee et al., 1999; Mahal et al., 1999). In its first decade, MOE was primarily utilized to modify the cell surface without directly or explicitly altering the biology of the host cell; instead, responses were elicited indirectly (e.g., by changing viral binding; Keppler et al., 2001) or by directing drugs to the modified glycans (Mahal et al., 1997; Nauman and Bertozzi, 2001). Later, as it became evident that MOE can modulate biological responses directly (Schmidt et al., 1998; B¨uttner et al., 2002; Horstkorte et al., 2004), efforts aimed at using this technology to control cell behavior have intensified. For example, the interrelated abilities of MOE to control Non-Natural Glycan Display Current Protocols in Chemical Biology 2: 171-194, September 2010 Published online September 2010 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/9780470559277.ch100059 C 2010 John Wiley & Sons, Inc. Copyright
171 Volume 2
natural cell surface glycans
R1 “R1”-modified glycan ManNAc analog –
glycosylation pathways
A
R1
B
D C
–
R1
Neu5Ac analog
– Neu5Ac analog
R1
CMP-Neu5Ac analog
Figure 1 Overview of metabolic oligosaccharide engineering (MOE). MOE was pioneered for the sialic acid biosynthetic pathway, which is one of the glycosylation pathways of a cell that can accommodate non-natural analogs. Modifications in this pathway can be introduced as “R1 ”modified forms of (A) ManNAc, (B) Neu5Ac, or (C) CMP-Neu5Ac. These metabolic intermediates intercept the sialic acid pathway at various points, are utilized in place of the natural pathway components, and ultimately appear on the cell surface as chemically altered form of Neu5Ac, the most common form of sialic acid found in human cells. Examples of “R1 ” modifications are shown in Figure 3.
cell-substrate binding (Villavicencio-Lorini et al., 2002) and alter gene expression (Kontou et al., 2008; Elmouelhi et al., 2009) are opening new horizons for MOE. Another fairly recent development of MOE is the extension of this technology beyond the sialic acid pathway to include pathways that process GlcNAc (Khidekel et al., 2004), GalNAc (Hang and Bertozzi, 2001) and fucose (Sawa et al., 2006). As a result, an increased number of glycans found within the glycocalyx can be endowed with novel structural and chemical features. In this article, we focus on ManNAc analogs that target the sialic acid pathway, and describe in detail parameters that require optimizing when adapting this technology for other analogs (as reviewed in detail elsewhere; Aich and Yarema, 2008; Campbell et al., 2007; Du et al., 2009), pathways, and cell types. This set of protocols is intended to serve as a foundation to broadly adapt MOE for any biological application, which theoretically includes any of the vast array of biological processes modulated by cell-surface carbohydrates.
STRATEGIC PLANNING
Non-Natural Glycan Display
This article describes the planning process and general procedures required for characterization of cellular metabolism and measurement of biosynthetic incorporation of novel sugar analogs used in MOE experiments. Because of the many analogs now available (see Aich and Yarema, 2008, for a perspective), numerous options exist for MOE experiments, and it is necessary to tailor published protocols for previously untested analogs or cell lines. Accordingly, a perspective of various steps involved in setting up an MOE experiment is provided here with a brief discussion of alternative approaches and common pitfalls to avoid. Overall, as outlined in the flow chart shown in Figure 2, the process begins with the incubation of cells with non-natural analogs (Basic
172 Volume 2
Current Protocols in Chemical Biology
cell line selection Strategic Planning
pathway selection analog design Strategic Planning
monosaccharide modification delivery/efficiency options
proliferation rate initial evaluation of cell viability
viability cytotoxicity
Basic Protocol 1 Basic Protocol 2
cellular uptake colorimetric assays fluorescent labeling
characterization of analog metabolism and glycan modification growth inhibition
Basic Protocol 3 Basic Protocol 4
chromatography mass spectroscopy lectin arrays
altered gene expression signaling pathways control of differentiation
analysis of changes in cell function and behavior
Figure 2 Flow diagram depicting various steps involved in a typical MOE experiment. Procedures that are not described in this unit are denoted with a star; additional information on these steps is provided in the references cited in the main text.
Protocol 1), followed by protocols to assess cell viability by quantifying growth inhibition and cytotoxicity (Basic Protocol 2). Next, the periodate-resorcinol assay, which provides a readout of successful incorporation of analog into the sialic acid pathway, is described (Basic Protocol 3). Protocols are then provided for the quantitative estimation of analog incorporation into cell-surface glycans by using bio-orthogonal ligation reactions and flow cytometry assays (Basic Protocol 4).
Cell Line Selection In the early years of MOE, most experiments were performed with human cancer or rodent cell lines, which readily accommodated sugar analog incorporation. More recently, as efforts have turned to applying MOE to healthy cells, it is becoming clear that metabolic incorporation of analogs can vary significantly depending on the glycosylation pathway targeted, the analog used, and even the species involved. Therefore, although MOE technology can be used for virtually any mammalian cell type, the Jurkat line makes an excellent positive control to run in parallel with cell lines of unknown MOE capabilities to ensure the fidelity of reagents and to troubleshoot procedures. Jurkat cells have served as the “workhorse” cell type for MOE (Du et al., 2009) because of several factors. First, they have a relatively modestly sized glycocalyx (∼8 nm thick; Freitas, 1999), which ensures that analog uptake will not be inhibited by the thick (up to microns; Cohen et al., 2004) glycocalyx found in some cell types. Conversely, the installed glycans will be easy to detect because they will not be buried under a thick layer of hyaluronic acid or other polysaccharides that comprise a steric barrier to protein-sized probes in some cell types (Weinbaum et al., 2007). Second, Jurkat cells grow in suspension under normal culture conditions, making them trivial to harvest and amenable to time-course evaluations. Suspension culture also avoids potential problems with damage to surface glycoconjugates during the detachment steps required for adherent cell lines. An important caveat for Jurkat cells, however, is that they express truncated O-glycans (Piller et al., 1990), making this line best suited for analyzing analog incorporation into N-glycans.
Non-Natural Glycan Display
173 Current Protocols in Chemical Biology
Volume 2
Analog Selection Analog selection requires several decisions; the first is to decide what glycan biosynthetic pathway to target. Currently, there are three main monosaccharide targets for surface display (sialic acid, GalNAc, and fucose) and one option for nucleocytoplasmic modifications (O-GlcNAc). These sugars can be experimentally accessed with ManNAc, GalNAc, fucose, and GlcNAc “core” structures shown in Figure 3A. Sialic acid has been the preferred pathway for MOE experiments because there are multiple points to intercept the pathway (e.g., by using ManNAc, Neu5Ac, and CMP- Neu5Ac analogs; see Fig. 1) and the location of this sugar at the outer termini of mammalian glycoconjugates makes it readily accessible for interactions with the microenvironment or exogenously delivered labeling reagents. In practice, ManNAc analogs are preferred over Neu5Ac or CMP-Neu5Ac analogs for MOE experiments largely because of ease of synthesis and the poor cellular uptake properties of CMP-Neu5Ac (a detailed discussion of the relative benefits of ManNAc versus Neu5Ac analogs is provided in Aich and Yarema, 2008 and in Du et al., 2009).
A
E
“R2” groups
(Ac) ManNAc analog scaffold
Fucose analog scaffold
(Bu) GalNAc analog scaffold
B
“R1” groups
GlcNAc analog scaffold
C
D
Figure 3 Analog design features. (A) The “core” structures of ManNAc, GalNAc, fucose, and GlcNAc analogs used in MOE that are further derivatized with R1 and R2 groups. Categories of R1 groups, which are carried through on to the cell surface and displayed as a modify glycan, include those with (B) enlarged hydrocarbon groups, (C) heteroatom (or other) containing functional groups, and (D) direct fluorophores. (E) Examples of R2 groups include a proton (i.e., the -H found on hydroxyl groups) and 2- to 5-carbon short chain fatty acids that are ester-linked to the hydroxyl groups of the monosaccharide analog. Ac and Bu denote acetyl and n-butanoyl substitutions shown in Figure 4 and referenced throughout the main text.
Non-Natural Glycan Display
174 Volume 2
Current Protocols in Chemical Biology
The next decision involves selecting a chemical modification to install into cell-surface glycans; options fall into three main categories: enlarged alkyl hydrocarbon (e.g., alkyl) moieties (Fig. 3B), chemical functional groups (Fig. 3C), or fluorescent tags (Fig. 3D). The first category is exemplified by the initial analogs bearing elongated alkyl chains utilized in the Reutter laboratory’s landmark MOE experiments (Kayser et al., 1992). This group of compounds now includes branched and even ring structures and is ideal for high-flux applications (Kim et al. 2004). One drawback is that the surface incorporation of these analogs cannot be readily quantified. The second category consists of chemical functional groups, many of which are not found in sugars, thereby opening the door to highly versatile bio-orthogonal and chemoselective ligation options for glycan labeling (Lemieux and Bertozzi, 1998). Finally, the third group of fluorophore-bearing analogs is ideal for easy, one-step labeling of surface glycans but, because fluorophores tend to be unwieldy and bulky, they are only incorporated into glycans when introduced as CMP-Neu5Ac analogs (Brossmer and Gross, 1994). A final design consideration involves the use of short chain fatty acid (SCFA) protecting groups that increase the lipophilicity and cellular uptake of the sugar analog (examples are the “R2 ” groups shown in Fig. 3E). A benefit of appending ester-linked SCFA to mask the hydrophilic hydroxyl groups of a monosaccharide is that the concentration of analog required to evoke a biological response is dramatically reduced, ranging from ∼600fold for acetyl groups (Jones et al., 2004) to more than 2100-fold for n-butyrate (Kim et al., 2004). This feature is particularly important if experiments will be extended from small-scale cell culture experiments of the type described in this article to animal models or larger-scale biotechnology applications such as the use of analog in recombinant glycoprotein production (Viswanathan et al., 2003; Luchansky et al., 2004). As a potential pitfall, the inclusion of SCFA protecting groups in analog design can substantially alter biological activity (Aich et al., 2008) and, of practical importance, lead to dose-limiting cytotoxcity that depends on the structure of both the “R1 ” and “R2 ” groups shown in Figure 3 (Kim et al., 2004). Consequently, as outlined in Basic Protocol 2, empirical determination of the effects of analog on cell viability is required.
Initial Evaluation of Analog Effects on Cell Viability Depending on the cell line and the analog under evaluation, biological responses can be realized at concentrations as low as 5 μM (Elmouelhi et al., 2009) for SCFA-derivatized analogs, with the dynamic range reaching 75 mM or higher for free-hydroxyl compounds (Yarema et al. 1998). The upper limit for SCFA-derivatized analogs is usually determined by the dose-limiting toxicity (DLT) and can vary significantly (from 25 to >700 μM) depending on cell line (Yarema et al., 1998), cell density (Jones et al., 2004), rate of delivery, and N-acyl substituent (Kim et al., 2004). Therefore, to narrow the concentration range before undertaking highly complex procedures, such as the flow cytometry-based quantification of surface glycan display outlined in Basic Protocol 4, one (or both) of two rapid screening type assays are recommended to measure cell viability (Basic Protocol 2) and/or metabolic flux (Basic Protocol 3). In general, the effects of novel analogs on cell viability should be evaluated over a range of concentrations (typically 0 to 250 to 500 μM for SCFA-derivatized analogs) as a first step in any MOE experiment. Detailed Characterization of Analog Metabolism and Glycan Modification Metabolic flux can be confirmed by colorimetric assays Analogs sometimes have no deleterious effects during pilot in vitro testing and do not noticeably affect cell viability endpoints tested in Basic Protocol 2. In this case, the analog may lack cytotoxicity or, alternatively, simply may not have been successfully taken up by the cell. Because monosaccharide analogs have chemical properties similar to many components of sugar-rich cell culture media, efforts to probe cellular uptake by
Non-Natural Glycan Display
175 Current Protocols in Chemical Biology
Volume 2
measuring residual analog levels in the medium have proved cumbersome (Jones et al., 2004). Therefore, in the absence of radiolabeled metabolic precursors (which are usable but prohibitively expensive; Jacobs et al., 2001), colorimetric techniques can provide a ready means for quantitative estimation of analog flux into intracellular glycosylation pathways. Colorimetric methods exist for quantifying flux into several glycosylation pathways (e.g., glucose, King and Garner, 1947, for various monosaccharides; Walborg and Christensson, 1965), but the periodate-resorcinol assay is particularly valuable because it selectively detects sialic acids while avoiding competing signal from other monosaccharides (Jourdian et al., 1971). Our laboratory typically uses a modified version of the periodate-resorcinol method (described in Basic Protocol 3) that involves scaling the reaction from the initial 4.0-ml sample volumes to 100-μl volumes amenable to quantification in 96-well plates. The periodate-resorcinol assay involves two major steps: (1) periodate oxidation in which a periodate ion reacts with two hydroxyl groups of vicinal diols, breaking the C-C bond to form two aldehyde groups (Tiziani et al., 2003), and (2) heating the sample with resorcinol to improve sensitivity (up to a few micrograms of sialic acid; Sussich and Cesaro, 2000). It is particularly valuable to gauge flux of analogs—such as ManNProp or ManNBut—that do not have bio-orthogonal functional groups that can be exploited ready determination of the surface display of non-natural glycans (as described for ketones and thiols in Basic Protocol 4). Certain analogs, however (e.g., ManNHex or ManNLev), do not measurably increase intracellular metabolite levels (Kim et al., 2004), although they do successfully transit the pathway and modify surface glycans (Yarema et al., 1998). For these “low-flux” analogs, colorimetric assays will not be informative (comparison with panels of “high-flux” analogs appropriate for the periodate assay are provided in previous publications, e.g., Aich et al., 2008; Kim et al., 2004). Despite these limitations, the combination of cell viability and flux assays reliably provide a narrowed concentration range for select analogs (e.g., 75 to 150 μM instead of 5 to 1500 μM) to use in subsequent assays.
Quantitative estimation of non-natural analog incorporation into surface glycans Another method of determining analog uptake and metabolism is to directly characterize non-natural glycan display on the cell surface. This approach is more thorough than the colorimetric assay described in Basic Protocol 3 because it measures flux all the way through a metabolic pathway to the cell surface rather than just quantifying pathway intermediates. One method to accomplish this objective is to stain non-natural sugars on the cell surface with complementary antibodies (Lemieux and Bertozzi, 2001; Chefalo et al., 2004); this approach is limited, however, by a lack of antibodies (or lectins) with specificity for most non-natural monosaccharides used in MOE. Therefore, in Basic Protocol 4, a quantitation technique is described that uses chemoselective ligation reactions to detect the subset of analogs that install unique chemical functional groups (some of which are shown in Fig. 3C) into cell-surface-displayed glycans.
Non-Natural Glycan Display
There is an ever increasing repertoire of bio-orthogonal reactions—based on the incorporation of chemical functional groups not naturally found in the glycocalyx—that fit into this category. Here, the discussion is limited to cells incubated with either a ketone- (Ac4 ManNLev, peracetylated N-levulinoyl mannosamine; Chang et al., 2007) or thiol-bearing analog (Ac5 ManNTGc; Sampathkumar et al., 2006c), but the general methodology can be readily adapted to additional functional groups that include such as azides (Saxon and Bertozzi, 2000), alkynes (Sawa et al., 2006), diazirines (Tanaka and Kohler, 2008; Bond et al., 2009), and aryl azides (Han et al., 2005). ManNLev and ManNTGc were selected for description here because they have several properties worth highlighting. For example, both analogs transit the sialic acid biosynthetic
176 Volume 2
Current Protocols in Chemical Biology
pathway and install high levels of modified glycans (estimated at up to 20 million) on the cell surface but do not lead to large increases in intermediary metabolites that are easily detected by the periodate-resorcinol assay described in Basic Protocol 3. Also, in both cases, the new functional groups displayed in the form of modified glycans can be targeted using chemoselective ligands conjugated to biotin that can then be conjugated to FITC-avidin for quantification using flow cytometry; specifically, cells displaying ketone groups can be chemoselectively ligated to hydrazides to form an acyl hydrazine (Yarema et al., 1998), whereas cells bearing thiols can be chemoselectively ligated to maleimide to form thio-ether groups (Sampathkumar et al., 2006c). Due to their chemical properties and the slightly oxidizing conditions in the extracellular milieu in normal tissue culture conditions, an additional nuance exists for glycan-displayed thiols that is worth noting insofar as a majority of cell-surface thiols form cis-disulfides (Sampathkumar et al., 2006b). Therefore, to better estimate the total level of surface sugar analog display, this protocol introduces the mild reducing agent triscarboxyethyl)phosphine (TCEP), which reduces disulfide bonds before chemoselective ligation of thiols is performed.
Additional options for characterization of surface glycans in MOE cells In cases where analog incorporation does not result in the display of novel bioorthogonal functional groups on the cell surface, and in the absence of an appropriate antibody or lectin for detection of the non-natural glycan, high-performance liquid chromatography (HPLC) can be used to estimate metabolic incorporation of analog provided that an authentic sample of the surface glycan is available (e.g., Sia5Prop for cells treated with ManNProp; Gagiannis et al., 2007). For the more rigorous and definitive characterization, mass spectrometry can be used to identify metabolically glycoengineered oligosaccharides (Schilling et al., 2001). Finally, the MOE process has the potential to enact unanticipated changes to surface glycans, which can be probed via emergent lectin array technologies (e.g., see Pilobello et al. 2005). NOTE: A limited number of analogs used in MOE are commercially available; these include ManNAc (from many suppliers), Ac4 ManNAc (New Zealand Pharmaceutics), Ac4 ManNLev (Invitrogen), Ac4 ManNAz, Ac4 GalNAz, and Ac4 GlcNAc; each of these three azido analogs is available from both Invitrogen and Sigma-Aldrich. Other analogs can be prepared following published protocols (e.g., see Sampathkumar et al., 2006d) that contain recipe-type instructions for synthesizing Ac5 ManNGc and Ac5 ManNTGc. NOTE: Monosaccharide analogs stored in dried powder form at –20◦ C are stable for extended time periods (at least several months up to a few years). Stock solutions are commonly prepared by dissolving the SCFA-derivatized analogs (e.g., 1, 3, 4, or 5, Fig. 4) in ethanol (EtOH) or DMSO at a concentration of 10 mM and storing at 4◦ C; although sugar analogs are stable in this form over a several-month period, it is recommended to prepare modest-sized “stock” solutions that can be used within a few (6 to 8) weeks. Analogs are typically stored in ethanol for two reasons—the first is that SCFA-derivatized monosaccharides have limited solubility (often 1.0 mM or less, which does not pose a problem for biological assays usually performed at <500 μM)—and the second is to maintain sterility over extended storage periods; DMSO is an alternative solvent vehicle, but is used more often for in vivo studies than in cell culture. Typically, when the solvent level is kept below 1.0% (v/v), neither ethanol nor DMSO has measurable effects on endpoints commonly probed in MOE experiments. Nevertheless, the preferred protocol is to avoid any solvent exposure by allowing evaporation before cells are added or, when this is not possible, to ensure that the same volume is added to each sample.
Non-Natural Glycan Display
177 Current Protocols in Chemical Biology
Volume 2
BASIC PROTOCOL 1
INCUBATION OF CELLS WITH SUGAR ANALOGS The first step in MOE is the incubation of cells with sugar analogs. The protocol described herein is for the incubation of Jurkat cells with the thiol-bearing ManNAc analog Ac5 ManNTGc (1 in Fig. 4). Although an exhaustive bevy of analog controls are not always required, in initial experiments it is recommended to compare the effects of a target analog (in this case Ac5 ManNTGc) with appropriate control analogs. For example, ManNAc (2) serves as a control for increased flux through the sialic acid pathway and Ac4 ManNAc (3) for increased flux plus any “off-target” responses engendered by the acetyl protecting groups, either from their scaffold-dependent effects (Elmouelhi et al., 2009) or from SCFA-specific responses (e.g., HDACi activity; Sampathkumar et al. 2006a) after removal by cellular esterases. Taking a step towards increased complexity, Ac4 ManNProp (4) and/or Ac5 ManNGc (5) would be appropriate controls for an elongated N-acyl side chain reported to enhance signaling propensities (Kontou et al., 2008) or receptor binding properties (Collins et al., 2000) of these molecules, but without the unique chemical properties of the thiol installed by Ac5 ManNTGc at this position.
Materials Stock solutions (10 mM in ethanol or DMSO) of sugar analogs shown in Figure 4 [Ac5 ManNTGc (1), Ac4 ManNAc (3), Ac4 ManNProp (4), and Ac4 ManNGc (5)], or other analogs of choice; if a free hydroxyl monosaccharide (e.g., ManNAc (2) or ManNLev; Mahal et al., 1997) is used in any experiment, a 500 mM stock solution should be prepared in PBS or serum-free culture medium and filter sterilized Jurkat clone E6-1 (ATCC, cat. no. TIB-152; see Support Protocol 1 for culture and harvesting) Complete RPMI 1640 medium (see recipe) Ethanol (200 proof; Pharmco) 6-, 12-, or 24-well tissue culture plates 15- or 50-ml polypropylene centrifuge tubes Z2 Coulter particle count and size analyzer (Beckman Coulter) or hemacytometer for counting cells Additional reagents and equipment for harvesting Jurkat cells (Support Protocol 1) 1. Label sterile 24-well tissue culture plates, e.g., with the concentrations of SCFAderivatized analogs shown in Table 1 or higher millimolar concentrations of free hydroxyl analogs reported elsewhere (Yarema et al., 1998), in triplicate for each
Ac5ManNTGc (1)
Ac4ManNProp (4)
Non-Natural Glycan Display
Figure 4
ManNAc (2)
Ac4ManNAc (3)
Ac5ManNGc (5)
Chemical structures of ManNAc analogs used in this unit.
178 Volume 2
Current Protocols in Chemical Biology
Table 1 Final Concentrations of Analogs in Culture Medium after 0.5 ml of Cells are Added
Final concentration of analog (1, 3, 4, 5; μM)
0
5.0
10
20
40
80
160
320
Stock solution of 10 mM 1, 3, 4, or 5 added (μl)
0
0.25
0.5
1.0
2.0
4.0
8.0
16
Ethanol added (μl) (optional, see text)
16
15.75
15.5
15
14
12
8.0
0
concentration. Add the appropriate volume of a 10 mM stock solution of 1 to give the desired final concentrations when 0.5 ml of cells are added later (in step 4). For volumes of ≤1 or 2 μl, even if a pipettor capable of measuring 0.2 to 2 μl is available, it is advisable to use 10× volumes of a 1.0 mM “stock” solution of analog to avoid pipetting errors, because the surface tension of ethanol differs from aqueous solutions sufficiently that it is difficult to add such small volumes accurately and reproducibly. When adherent cells are being tested, the same protocol described for Jurkat cells in steps 1 to 5 can be followed, where the analog is added to the culture plate before the cells. Alternatively, the appropriate amount of analog can be added to the cell culture medium after the cells are plated (steps 3 to 4); this procedure minimizes the deleterious effects of cytotoxic analogs. In this case, the total amount of ethanol added to each well should be kept constant (e.g., by separately adding the appropriate amount shown in the “optional” line of Table 1).
2. Leave tissue culture plates in sterile hood with their lids open for ∼15 min to allow the ethanol to evaporate. 3. While ethanol is evaporating, harvest cells as described in Support Protocol 1 for Jurkat cells, or using usual methods for an alternate cell line of choice. 4. Adjust Jurkat cell density to 5.0 × 105 cells per ml complete RPMI medium for a 48-hr experiment (or 2.5 × 105 cells per ml for a 72-hr experiment) and mix thoroughly to obtain a homogeneous single cell suspension. Add 0.5 ml of cell suspension to each well and incubate cells at 37◦ C. When using cell lines that deviate significantly from the growth rate of Jurkat cells, which have a doubling time of ∼1 day (∼26 hr), the initial cell density should be adjusted to ensure that approximately the same number of cells are present at the end of the experiment for each condition tested. Generally, as long as cells are in a range that supports robust growth (e.g., from 2.0 × 105 to 2.0 ×106 cells per ml for the Jurkat line), cell density does not significantly influence analog metabolism (Jones et al., 2004) but as a cautionary note, very low cell densities (e.g., <1.0 × 105 cells/ ml) exacerbate analog toxicity and should be avoided.
5. After 48 or 72 hr (or other intervals for time-dependency experiments; see step 6), thoroughly mix to obtain a single-cell suspension by gently pipetting up and down several times with a 1.0-ml pipet. Count an aliquot of the culture to determine cell density. Cells can then be used for toxicity, labeling, or functional assays. Typically, the surface display of modified glycans reaches maximal levels after 2 or 3 days of incubation with SCFA-derivatized analogs (e.g., Ac4 ManNLev), whereas free hydroxyl monosaccharides (e.g., ManNLev) may require up to 5 days to achieve saturating surface display. The shorter time period is advantageous because cells do not need to be passaged (which they do for a 4-day or longer incubation) and it is also generally long enough to observe growth inhibition and early indications of analog -induced cytotoxicity and apoptosis. Non-Natural Glycan Display
179 Current Protocols in Chemical Biology
Volume 2
6. Study analog incorporation as a function of time. Several options are available. For short time intervals (e.g., <48 hr), we suggest preparing separate plates as described in step 1 and simply harvesting each one at a specified time. For time points <24 hr, a higher initial cell density (e.g., 1.0 × 106 cell/ml) may be required to ensure a sufficient number of cells for analysis. For multi-day time points, because Jurkat and other human cancer cell lines double in ∼1 day, one option is to prepare larger volume cultures (e.g., 1.0 ml) and remove 0.5 ml each day for analysis. The volume is replenished with 0.5 ml of fresh medium and the appropriate volume of analog can be added to maintain the initial final concentration; in this case, when the analog is added directly to the culture rather than allowed to evaporate as described in step 2, the “optional” amount of ethanol (e.g., as shown in line 3 of Table 1) should be added to maintain the same amount of solvent in each well. SUPPORT PROTOCOL 1
ROUTINE GROWTH AND MAINTENANCE OF JURKAT CELLS This protocol describes routine methods to maintain Jurkat cells, a lymphocyte line derived from human acute T cell leukemia (clone E6-1 are often referred to as “wildtype” Jurkat cells). In general, no special procedures are needed for preparation of cells for MOE experiments; therefore, the usual methods used to culture the cell line of choice can be followed to prepare cells for use in Basic Protocol 1, above.
Materials Jurkat clone E6-1 (ATCC, cat. no. TIB-152) Complete RPMI 1640 medium (see recipe) Z2 Coulter particle count and size analyzer (Beckman Coulter Inc) or hemacytometer for counting cells 75-cm2 tissue culture flask (Sarstedt, cat. no. 83.1813) 37◦ C, 5% CO2 , humidified incubator 15- or 50-ml polypropylene centrifuge tubes 1. Add Jurkat cells to the appropriate amount of complete RPMI 1640 medium to attain a cell density of ∼1.0-2.5 × 105 cells/ml (e.g., 1.5 × 106 cells in 15 ml in a 75-cm2 flask) and place the flask or culture dish in incubator. The volume of initial culture medium and/or the number of cells used should be adjusted depending on the number of cells required in step 5 (for use in Basic Protocol 1).
2. Incubate up to 4 days. The cells should reach a density of ≤2.0 × 106 cells/ml (if the cells in step 1 are freshly thawed, cell growth to saturating cell density may take up to 1 week). Jurkat cells are suspension cells and tend to form small, loosely associated clusters while growing in suspension; the cells should be dissociated, which can be accomplished by gently pipetting them up and down several times, before counting to ensure accuracy. In cases where the pH indicator in the culture medium has not changed color indicating accumulation of metabolic waste products, and where the cells will be diluted significantly in subsequent use, steps 3 and 4 may not be necessary.
3. Transfer the cell suspension into a 15- or 50-ml centrifuge tube and centrifuge 3.5 min at 1300 × g, room temperature. Carefully discard supernatant and resuspend cells in 10 ml of prewarmed (37◦ C) complete RPMI medium. 4. Count the cell density using either an electronic cell counter or a hemacytometer; again, be sure to disperse cell clumps by gently pipetting up and down several times before counting. Non-Natural Glycan Display
180 Volume 2
Current Protocols in Chemical Biology
5. Replate an aliquot of the cells as described in step 1 and place the stock flask in incubator; the remainder of the harvested cells (typically 85% to 95% of the total number of cells initially present) can be used in MOE protocols.
CELL VIABILITY ASSAYS SCFA-monosaccharide analogs often have a negative impact on cell viability; these deleterious effects vary greatly, ranging from transient growth inhibition to—at high enough concentrations—the rapid onset of apoptosis. Cytotoxicity depends on analog structure. For example, Jurkat cells incubated with acetylated ManNAc analogs experience cytotoxicity not observed with free hydroxyl monosaccharides, and the degree of toxicity is influenced by the N-acyl modification (Kim et al., 2004); thus, both the “R1 ” and “R2 ” modifications shown in Figure 3 affect cell viability. Cell viability can be evaluated with various levels of sophistication and thoroughness. At the simplest, cell counts obtained in Basic Protocol 1 identify analogs that inhibit cell growth (which can be quantified as an IC50 value that gives an inhibitory concentration that reduces cell counts to 50% of control values) after short-term analog exposure (e.g., 2, 3, or 5 day). Because actual cell death through analog exposure is slow over concentration ranges used in most MOE experiments, additional analyses must be done to determine the viability of cells in samples that contain fewer cells than the negative control (i.e., without analog). Reduced cell numbers could be due to transient growth inhibition from which the cells can recover and resume robust growth, or the remaining cells could be fated to die over the next several days (Sampathkumar et al., 2006a). If this issue needs resolution, trypan blue (or other live/dead assays) can be used to detect if dead cells are already present; standard assays for apoptosis can be run to determine if programmed cell death has been set in motion, or the MTT assay can be used to evaluate the metabolic state of the cells. A straightforward and reliable way to determine lethality, however, is to continue to monitor cell counts over a 2-week period (Sampathkumar et al., 2006a); in general, IC50 data obtained from 2-, 3-, or 5-day cell counts or LD50 data from 15-day experiments are adequate to determine safe levels of analog to use for subsequent metabolic uptake (e.g., Basic Protocol 3) and labeling (Basic Protocol 4) experiments.
BASIC PROTOCOL 2
Materials Jurkat clone E6-1 (ATCC, cat. no. TIB-152; see Support Protocol 1 for culture and harvesting) Complete RPMI 1640 medium (see recipe) Phosphate-buffered saline (PBS), pH 7.4 (Invitrogen, cat. no. 10010-049) 10 mM stock solutions of analog: e.g., Ac4 ManNAc (3) plus any additional analogs required for user-specific applications) Ethanol (EtOH) (200 proof) for controls 24-well tissue culture plates Z2 Coulter particle count and size analyzer (Beckman Coulter) or hemacytometer for counting cells Growth inhibition (IC50 ) and cytotoxicity (LD50 ) assays 1. Incubate Jurkat cells (1.0 × 105 cells in 0.5 ml) with analog (e.g., 2, 0 to 320 μM), using a 24-well plate as described in Basic Protocol 1, in a 37◦ C, 5% CO2 incubator for 3 days. If adherent cells are used, two plates (times 3 replicates) for each condition will be needed because the cells will be trypsinized, counted, and discarded at each of the two time points (days 3 and 5; step 2). Steps 3 to 5 are typically not performed for adherent cells because without trypsinization and passaging they tend to become overconfluent, and with these procedures the manipulations introduce errors that lead to inaccurate results.
Non-Natural Glycan Display
181 Current Protocols in Chemical Biology
Volume 2
2. On day 3, gently pipet Jurkat cells up and down using a 1-ml pipet to dislodge cell clumps. Take two 100-μl aliquots of cell suspension and add each to 10 ml of PBS in a cuvette. Measure and record the cell density using a cell counter. If a cell counter is unavailable, perform a manual cell count using a hemacytometer. Add 1.0 ml of medium to each well and place cells in incubator. Repeat this step on day 5. 3. On day 7, gently pipet the Jurkat cells to dislodge cell clumps. Remove 1.0 ml of cell suspension and add 1.0 ml of fresh medium. Repeat this step for days 9, 11, and 13. 4. On day 15, gently pipet the Jurkat cells to dislodge cell clumps. Take a 1.0-ml aliquot and add it to 10 ml of PBS in a cuvette. Measure and record the cell density using a cell counter (or hemacytometer). 5. Calculate the relative cell number for each analog concentration with respect to the EtOH control for day 3, 5, and 15. Plot the relative cell number with respect to analog concentration for days 2 or 3, 5, and 15, respectively, to determine IC50 or LD50 values, which are the concentrations where relative cell numbers are 50% of untreated control samples. BASIC PROTOCOL 3
PERIODATE-RESORCINOL ASSAY TO MEASURE ANALOG UPTAKE BY A CELL AND INCORPORATION INTO METABOLIC PATHWAYS Once cell viability has been analyzed to determine the maximum concentration of glycan analog that can be used without deleteriously affecting cell viability, the next step is to quantify the extent of analog incorporation. The ideal way to measure analog uptake and metabolism is to quantify non-natural glycan display on the cell surface as described in Basic Protocol 4; this approach is preferred to the periodate-resorcinol assay described here in Basic Protocol 3 because it measures flux all the way through a metabolic pathway to the cell surface. However, the detection of non-natural glycans on the cell surface is not trivial and almost always requires more highly sophisticated instrumentation than used for the periodate-resorcinol assay. In brief, the periodate-resorcinol assay is a colorimetric technique that quantifies total and glycoconjugate-bound sialic acids that can provide a rapid quantitative estimation of ManNAc analog incorporation into the sialic acid biosynthetic pathway. It is attractive because it provides a low-cost and technically simple option to monitor cellular uptake and subsequent successful incorporation of “high-flux” analogs into the sialic acid biosynthetic pathway.
Materials Jurkat clone E6-1 (ATCC, cat. no. TIB-152; see Support Protocol 1 for culture and harvesting) Complete RPMI 1640 medium (see recipe) 10 mM Ac4 ManNAc (3) stock solution plus additional sugar analogs of choice Ethanol (EtOH) (200 proof; Pharmco) for controls Phosphate-buffered saline (PBS) pH 7.4 (Invitrogen, cat. no. 10010-049) 0.4 M periodic acid stock solution (see recipe) 6% (w/v) resorcinol (Sigma, cat. no. 108-46-3; store at –20◦ C) t-butyl alcohol (2-methyl-propan-2-ol; Sigma, cat. no. 3972-25-6) 10 mM sialic acid stock solution (see recipe) 2.5 mM CuSO4 (see recipe) Concentrated HCl
Non-Natural Glycan Display
Z2 Coulter particle count and size analyzer (Beckman Coulter) or hemacytometer for counting cells 25-cm2 tissue culture flasks Heat block
182 Volume 2
Current Protocols in Chemical Biology
96-well microtiter plates Microplate reader (μQuant, Bio-Tek Instruments) 1. Incubate Jurkat cells (or other line being tested) with sugar analog(s) as described in Basic Protocol 1 but use 25-cm2 tissue culture flasks instead of 24-well plates and seed ∼1.0 × 107 cells in 10 ml. For Jurkat cells, a volume of 10 ml in a 100-mm culture dish will provide a sufficient number of cells to perform the resorcinol assay in triplicate; in some cases a culture size of 2 ml in a 6-well plate or 35-mm culture dish will yield sufficient cells. At a minimum, ∼1.0 × 106 cells are recommended for each replicate to obtain reliable results. For adherent cells, a 75-cm2 tissue culture flask will provide enough cells for triplicates. We suggest using a high-flux, low-toxicity analog such as 3 in all experiments as positive control. Over a concentration range of 0 to 250 μM, these compounds are expected to increase levels of total sialic acid by at least 10-fold in most cell lines.
2. Harvest the cells after 2 to 3 days. For Jurkat cells, gently pipet cell suspension up and down using a 1-ml pipet to dislodge cell clumps. If adherent cells are being tested, detach the cells from the culture flask using trypsin. Place the cells in a 15-ml centrifuge tube and centrifuge the cell suspension 5.0 min at 1000 × g, and discard supernatant. Gently resuspend cells in 10 ml PBS.
3. Gently pipet cell suspension up and down using to dislodge cell clumps. 4. Centrifuge cell suspensions in 15-ml centrifuge tubes 3.5 min at 1300 × g, room temperature, and discard the supernatant. Resuspend cells in 5.0 ml of PBS by gentle pipetting. Repeat this step an additional two times to wash the cells. FBS used in culture media is rich in glycoconjugate-bound sialic acids and must be thoroughly removed to ensure accuracy and reproducibility.
5. Remove two 100-μl aliquots of cells to count the cell density using a cell counter or hemacytometer. Based on the total number of cells in each condition, add the appropriate amount of PBS to obtain cell density of 1.0 × 106 cells/ml. Pipet the cell suspension up and down gently to obtain an homogenous cell suspension. 6. Aliquot 1.0 ml of resuspended cells into three 1.5-ml microcentrifuge tubes (1.0 × 106 cells/tube) to yield three replicates per condition. In general, at least 1.0 × 106 cells are needed to provide a robust colorimetric signal; an exception is provided by “high-flux” analogs (such as 3 or 4) that increase sialic acid levels by 10-fold or more. In this case, a correspondingly lower number of cells can be used for each replicate.
7. Pellet the cells by centrifugation (as described in step 4) and resuspend in 300 μl of PBS. 8. Freeze the tubes by placing in a −80◦ C freezer for ∼1 hr (or if available, the tubes can be placed in pulverized dry ice, in which case they will freeze in 5 to 10 min). Thaw the tubes in a 37◦ C water bath. Repeat for two more freeze/thaw cycles (three times in total). 9. While tubes are undergoing freeze-thaw cycles, prepare the other reagents. a. Remove periodic acid stock solution and 6% (w/v) resorcinol solution from storage in a −20◦ C freezer to thaw at room temperature. b. Warm t-butyl alcohol to 25◦ C (or just above room temperature). c. Prewarm heating block to 100◦ C. d. Prepare 300 μl sialic acid standards in 1.5-ml microcentrifuge tubes over the concentration ranges shown in Table 2.
Non-Natural Glycan Display
183 Current Protocols in Chemical Biology
Volume 2
Table 2 Preparation of Sialic Acid Standardsa
Volume of 1.0 mM sialic acid stock solution
Volume of PBS
Final sialic acid concentration
0 μl
300 μl
0 μM
3 μl
297 μl
10 μM
6 μl
294 μl
20 μM
9 μl
291 μl
30 μM
12 μl
288 μl
40 μM
15 μl
285 μl
50 μM
a Prepare a fresh 1.0 mM stock solution by diluting 50 μl of 10 mM sialic
acid (see recipe) in 450 μl of PBS.
Sialic acid standards should be prepared using the same buffer as the samples, which are usually in PBS.
10. Add 5.0 μl 0.4 M periodic acid to each sample (including standards). Vortex and place on ice for 15 min. 11. Incubate samples for glycoconjugate-bound measurements at 37◦ C for 80 min. This step dissociates free sialic acid–periodate complexes such that only “bound” sialic acids (sialic acids that have been converted by cells into glycoconjugates) are detected in the last step. Sialic acids that have only been taken up by the cells but not incorporated into the cells’ metabolic pathways are not detected. To only measure the total sialic acid content (free and bound), omit this step.
12. Prepare mixture of resorcinol working reagent as follows:
1.5 ml 6.0% resorcinol 1.5 ml 2.5 mM CuSO4 6.6 ml deionized H2 O 5.4 ml concentrated HCl. Add 500 μl of resorcinol working reagent into each tube, vortex and incubate at 100◦ C for 10 min using the heating block. To ensure more uniform heating, fill heat block with water after placing the microcentrifuge tubes in it. When heating, place an empty heat block on top of the microcentrifuge tubes to prevent them from opening.
13. Remove the tubes from the heat block and cool rapidly in room temperature water. 14. Add 500 μl of t-butyl alcohol to each tube. Mix the tube contents by vortexing. Centrifuge the tubes at maximum speed in a benchtop microcentrifuge for 2 min. 15. Aliquot samples into 96-well plates (two wells for each sample). Measure absorbance at 630 nm.
Non-Natural Glycan Display
Absorbance can also be measured by placing samples in a cuvette and measuring using a spectrophotometer. Measurement using a microplate reader allows for smaller-volume samples to be used and provides a semi–high throughput alternative to cuvette-based methods. One disadvantage to the 96-well format is that, unlike a cuvette where path length is always constant, the depth of solution in a well changes the path length and thereby changes the spectroscopic reading. Therefore, it is critical to aliquot precisely the same volume (e.g., 100 μl) into each well; this feature also provides flexibility insofar that if the sample is too dilute to obtain a reading, 200 or 300 μl can be added to the well (conversely, if the sample is too concentrated, 25 or 50 μl can be used).
184 Volume 2
Current Protocols in Chemical Biology
QUANTITATION OF CELL-SURFACE GLYCOCONJUGATES The display of non-natural glycans on the cell surface is an inherently inefficient process, especially at steps where analogs enter a cell, as well as bottlenecks within intracellular glycosylation pathways themselves (Viswanathan et al., 2003). Therefore, analog incorporation into glycans is not a linear function of medium concentration (or even intracellular metabolite levels as determined by the periodate-resorcinol assay), but is better estimated through direct quantification of non-natural analogs displayed on the cell surface. The development of non-natural analogs with orthogonal chemistries such as ketones (Yarema et al., 1998), azides (Saxon and Bertozzi, 2000), and thiols (Sampathkumar et al., 2006b) that do not naturally occur in the glycocalyx has greatly facilitated detection and quantification of surface-displayed analogs, as they can be conjugated with fluorescent labels using chemoselective ligation reactions. Commonly, the cell-surface functional groups are reacted with biotin-conjugated chemoselective ligands and subsequently stained using fluorescein-conjugated avidin or streptavidin; quantitative estimation of surface display is then performed using flow cytometry. In this protocol, an approach for the detection of glycan-displayed thiol and ketone functional groups is described.
BASIC PROTOCOL 4
Materials Jurkat clone E6-1 (ATCC, cat. no. TIB-152; see Support Protocol 1 for culture and harvesting) Complete RPMI culture medium (see recipe) 10 mM Ac5 ManNTGc (1) stock solution for thiol labeling experiments and/or Ac4 ManNLev (Kim et al., 2004) [or 1,3,4-O-Bu3 ManNLev (Aich et al., 2008)) stock solution for ketone labeling] Ethanol (EtOH; 200 proof) Phosphate-buffered saline (PBS) pH 7.4 (Invitrogen, cat. no. 10010-049) Tris(2,carboxyethyl)phosphine hydrochloride (TCEP; Sigma, cat. no. C4706) MB solution (see recipe), freshly prepared Biotin buffer (see recipe), freshly prepared 5.0 mM biotin hydrazide stock solution (see recipe) Avidin staining buffer (ASB; see recipe) Fluorescein isothiocyanate (FITC)-labeled avidin stock solution (see recipe) Z2 Coulter particle count and size analyzer (Beckman Coulter Inc) or hemacytometer for counting cells 6-well tissue culture plates Centrifuge Tubes for flow cytometer (5-ml polystyrene round-bottom tubes, 12 × 75–mm; BD Falcon, BD Biosciences, cat. no. 352054) Flow cytometer equipped with a 488-nm argon laser (Becton Dickinson FACSCalibur, BD Biosciences) Incubate cells with sugar analogs 1. Incubate Jurkat cells with sugar analog (e.g., 1 for thiol expression or Ac4 ManNLev or1,3,4-O-Bu3 ManNLev ) as described in Basic Protocol 1, but use 6-well tissueculture plates (or 35-mm dishes) instead of 24-well plates, and seed ∼1.0 × 106 cells in 2.0 ml for each sample to be analyzed. This volume should provide sufficient number of cells at the end of 2 or 3 days of incubation (i.e., ∼ 4.0 × 106 cells in total) to perform the quantitative estimation of surface display in triplicate. When cell availability is limiting, as few as 2 × 105 cells per replicate can be used, or the number of replicates can be reduced to two.
Non-Natural Glycan Display
185 Current Protocols in Chemical Biology
Volume 2
Jurkat cell density should be maintained between 0.5 to 2 × 106 cells/ml. Below this density, the cells are more sensitive to analog toxicity, and above this level, their growth rates slow, which may deleteriously affect analog metabolism (Jones et al., 2004).
Prepare cells for labeling of surface functional groups 2. Gently pipet cell suspension up and down using a 5.0-ml pipet to dislodge cell clumps. Count an aliquot of cells to determine the cell density. The cell density—and the total number of cells per sample—is not critically important in these labeling steps. However, if there are too many cells (e.g., >2.0 × 106 per sample), a large pellet is formed that traps reagents during wash steps. By contrast, if there are too few cells (e.g., <0.5 × 106 per sample), the pellet may not be large enough to see during the wash steps and can potentially be lost.
3. Centrifuge cell suspensions in 15-ml tubes 3.5 min at 1300 × g, room temperature, and discard supernatant. 4. Add 3.0 ml of PBS to the cells and resuspend the cells by gentle pipetting. Avoid vortexing, to minimize damage of cells. 5. Aliquot 1.0 ml of resuspended cells to three 1.5-ml microcentrifuge tube (there should be a maximum of 2.0 × 106 cells/tube). This will provide three replicates per condition.
To label for thiol expression 6a. Wash the cells twice with PBS. For each wash cycle, centrifuge the cells 3.5 min at 3000 × g, room temperature, discard the supernatant, and gently resuspend cells in 1.0 ml PBS. 7a. Prepare a 10 mM solution of TCEP by combining 2.87 mg TCEP with 1.0 ml PBS. Centrifuge the cell suspension 3.5 min at 3000 × g, room temperature, and discard supernatant. Resuspend the cell pellets in 360 μl of PBS and add 40 μl of 10 mM TCEP solution to the samples to yield a final TCEP concentration of 1.0 mM. Lay the samples on the benchtop at room temperature for 1.0 hr. Gently invert the tubes every 15 min to ensure uniform labeling. It is crucial to mix the reagent immediately upon its addition into the cell suspension, to avoid cell aggregation or cross-linking. The time and concentration required for TCEP treatment has been optimized as a compromise between complete reduction of disulfides (which is estimated to be only 90% to 95% complete under the given conditions) and a loss of cell viability that occurs at longer time periods or higher concentrations of TCEP. To ensure reproducibility between experiments performed on different days, it is recommended to maintain TCEP concentration, time of incubation, and temperature as uniform as possible (note that if room temperature varies from 18◦ to 25◦ C over time, this change could measurably affect TCEP reduction of disulfides as well as the labeling efficiency in step 8a). If TCEP treatment for disulfide bond detection is not intended, or to compare TCEPtreated and untreated cells to estimate the degree of disulfide formation under normal culture conditions, add 400 μl of PBS to the cell aliquot(s) and directly proceed to step 8a.
8a. Add 100 μl of freshly prepared MB solution to the cell suspension and mix immediately with gentle pipetting. 9a. Lay the tubes flat on the table at room temperature for 1.0 hr. Flip the tubes every 15 min to ensure uniform labeling. Non-Natural Glycan Display
186 Volume 2
Current Protocols in Chemical Biology
To label for ketone expression 6b. Wash the cells twice with freshly prepared biotin buffer, then twice with PBS. For each wash, centrifuge the cells 2.5 min at 3000 × g, room temperature, discard the supernatant, and gently resuspend cells in 1.0 ml biotin buffer. 7b. Centrifuge cells 3.5 min at 3000 × g, room temperature, and resuspend gently in 250 μl biotin buffer. 8b. Add 250 μl of 5.0 mM biotin hydrazide stock and mix immediately by gently pipetting up and down. It is crucial that the cells be properly resuspended before adding biotin hydrazide. Mix the reagent immediately upon its addition into the cell suspension to avoid cell aggregation and cross-linking (note that other precautions described for thiol detection in steps 6a to 9a also apply for ketone labeling here in steps 6b to 9b).
9b. Lay the tubes flat on the table at room temperature for 2.0 hr. Mix the tubes every 15 min to ensure uniform labeling.
Stain with FITC-avidin 10. While the cells are being incubated in the MB (or biotin hydrazide) solution, cool 50 ml of ASB and 50 ml of PBS in an ice bath. 11. Centrifuge the tubes from step 9a or b 3.5 min at 3000 × g, room temperature, and discard supernatant. Add 1.0 ml of ice cold ASB to each tube and resuspend gently. Repeat this step two more times. During the third wash, resuspend cells in 100 μl ASB. 12. Prepare FITC-avidin working solution by diluting FITC-avidin stock solution 1:250 in PBS. Prepare slightly more than the required amount (200 μl per tube) to avoid insufficient solution for the last tube. Mix well and keep in dark on ice (not more than 12 hr). 13. Add 100 μl of FITC-avidin working solution to each tube and mix immediately by gentle pipetting. Incubate tubes on ice out of direct light for 15 min. It is important to properly resuspend the cells before addition of FITC-avidin. Also, mix the solution immediately after addition to obtain uniform labeling and avoid cell clumping.
14. Centrifuge the tubes 3.5 min at 3000 × g, room temperature, discard supernatant, and resuspend cells gently in 100 μl of ASB. Repeat step 13. Inconsistent results may be caused by incomplete removal of unreacted biotin reagents from the cell suspension before staining with FITC-avidin, as FITC-avidin reacts preferentially with the “free” biotin reagents instead of the cell surface. In this protocol, the first addition of FITC-avidin functions to remove any remaining reagents, allowing cells to be more reproducibly stained the second time.
15. During incubation with FITC-avidin, collect untreated Jurkat cells and place ∼1.0 × 106 cells into each of two 1.5-ml microcentrifuge tubes. Wash these cells twice, resuspend in 500 μl PBS, and transfer cell suspension into 5-ml polystyrene tubes. Keep the tubes on ice. These will be used to adjust flow cytometer settings and define the healthy cell population on the forward scatter–side scatter plot.
16. Centrifuge the tubes 3.5 min at 3000 × g, room temperature, discard supernatant, and resuspend cells gently in 1.0 ml of ice cold ASB. Wash the cells twice to ensure complete removal of unbound FITC-avidin. For the third wash, resuspend cells in 500 μl PBS and transfer to 5-ml polystyrene tubes and keep on ice out of direct light. Samples may be kept for up to 3 hr before analysis by flow cytometry.
Non-Natural Glycan Display
187 Current Protocols in Chemical Biology
Volume 2
Analyze by flow cytometry 17. Analyze the samples by flow cytometry. It is advisable to collect at least 10,000 events for each run for more accurate results. Run analysis twice for each sample.
18. Measure the relative fluorescence content (and hence relative surface display) of cells using the geometric mean in arbitrary units.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps.
Avidin staining buffer (ASB) 25 ml FBS (HyClone, cat. no. SH30071.03) 5.0 ml 10% (w/v) sodium azide (EM Science, cat. no. SX0299-1) 500 ml sterile phosphate-buffered saline (PBS; Invitrogen, cat. no. 10010-049) Store up to 6 to 8 weeks at 4◦ C This buffer is used while staining cells for surface display.
Biotin buffer 95 ml PBS (Adjusted to pH 6.5 with concentrated hydrochloric acid) 5.0 ml FBS (HyClone, cat. no. SH30071.03) Filter sterilize Prepare fresh This buffer is used to prepare biotin hydrazide solution for labeling surface ketone groups.
Biotin hydrazide stock solution, 5.0 mM 25.834 mg biotin hydrazide (Sigma, cat. no. B-7639) 20 ml PBS (adjusted to pH 6.5 with concentrated hydrochloric acid) Store up to 6 to 8 weeks at 4◦ C This solution is used for labeling surface ketone groups. Biotin hydrazide may take up to 1 hr to dissolve. If precipitate forms during storage, do not use.
Complete RPMI 1640 medium 470 ml RPMI 1640 medium (Invitrogen, cat. no. 11875-119) 25 ml fetal bovine serum (FBS; HyClone, cat. no. SH30071.03) 5.0 ml 100× penicillin-streptomycin solution (P/S; Sigma, cat. no. P-0781) Store up to one 1 month at 4◦ C This medium is used for culturing Jurkat cells.
CuSO4 solution, 2.5 mM 1.0 ml 40 mM copper sulfate (Sigma, cat. no. 7758-98-7) 15 ml deionized water Store up to 6 to 8 weeks at 4◦ C FITC-avidin stock solution 2.0 mg FITC-labeled avidin (Sigma, cat. no. A-2901; see recipe) 714 μl PBS (Invitrogen, cat. no. 10010-049) Store in dark at 2◦ C for up to 6 months Dissolve FITC-labeled avidin completely in PBS. Non-Natural Glycan Display
Dilute stock solution 1:250 in PBS for staining MB labeled cells.
188 Volume 2
Current Protocols in Chemical Biology
MB solution, 5.0 mM 3.9 mg EZ link maleimide PEO2-biotin (MB; Pierce Biotechnology, cat. no. 21901) 1.5 ml PBS (Invitrogen, cat. no. 10010-049) This solution is used to label surface thiol groups. Dissolve MB in PBS immediately before use.
Periodic acid, 0.4 M 910 mg periodic acid (Sigma, cat. no. 10450-60-9) 10 ml deionized water Aliquot into ten 1.5-ml microcentrifuge tubes (1.0 ml each) Store at −20◦ C This solution is for use in the periodate-resorcinol assay.
Sialic acid stock solution, 10 mM 10 mg N-acetylneuraminic acid (Sigma, cat. no. A-9646) 3.233 ml PBS (Invitrogen, cat. no. 10010-049) Store up to 2 weeks at 4◦ C This stock solution is used to establish sialic acid standards in the periodate-resorcinol assay.
COMMENTARY Background Information Oligosaccharides on mammalian cell surfaces play critical biological functions, as they contribute to most interactions between the host cell and its environment. These functional roles include numerous biological recognition processes such as viral and bacterial infection, innate and adaptive immunity, and cancer metastasis. Additionally, oligosaccharides, which are commonly conjugated to proteins or lipids to form glycoconjugates, play an essential role in determining the glycoconjugates’ biological activities (Varki, 1993). Accordingly, MOE technology holds great potential for manipulating basic cellular functions for medical applications such as tissue engineering, cancer therapy, and carbohydratebased medicines. Here, the general steps and protocols for characterization of novel sugar analogs used for MOE are described. Critical parameters for consideration and common problems related to these experiments are discussed below.
Critical Parameters and Troubleshooting Incubation of cells with sugar analogs For consistent results and optimal surface labeling, experiments should be performed using healthy and robustly growing cells harvested at subconfluency. Of course, if the goal of the experiment is to probe effects
of anomalous metabolic states—such as hypoxia, toxic waste products from overgrowth of cells (such as ammonia, which affects sialic acid metabolism; Zanghi et al. 1998a,b), or pharmacological agents—on glycan production, this caveat does not apply. The current set of protocols can be adapted to other sugar analogs and cell lines with the conditions (e.g., analog concentrations and incubation time) described herein serving as a reasonable starting point. Variations in the optimal conditions for different cell types and analogs should be expected, but can usually be covered by evaluating cellular responses in the 0 to 250 to 500 μM concentration range for SCFA-derivatized analogs. The current protocol also describes 0.5-ml cultures, which typically provide an ample number of cells for flow cytometry characterization, toxicity and viability assays, or functional characterization. If multiple endpoints are to be measured from a single batch of cells, this protocol can readily be scaled to 1.0 ml in 12-well plates or to 2.0 ml in 6-well plates. Another potential pitfall is that certain analogs (e.g., ManNProp or ManNAz) experience approximately the same rate of flux in human and rodent cells, whereas others (e.g., ManNLev) supported substantially lower incorporation (e.g., ∼20-fold) in rodent cells (Yarema et al., 1998). Finally, to date, the only human cells that we have tested that are relatively refractory to analog metabolism are
Non-Natural Glycan Display
189 Current Protocols in Chemical Biology
Volume 2
mesenchymal stem cells and chondrocytes, suggesting that factors such as extracellular matrix secretion by the test cells (both of these lines are copious secretors), which potentially sequester the analog and thereby hinder uptake by a cell, can confound experimental outcomes. Periodate-resorcinol colorimetric assay The periodate-resorcinol assay measures glycoconjugate-bound sialic acid when the periodate-oxidation step is performed at 37◦ C and total sialic acid when the oxidation step is performed on ice (Basic Protocol 3, step 11). Because the oxidation step is highly dependent on temperature, all temperatures stated in the protocol should be carefully controlled to avoid inaccurate results. Another important consideration is that, although conducting the oxidation step at 37◦ C is generally described as a way to measure “glycoconjugatebound” sialic acids (Jourdian et al., 1971), we have found that CMP-sialic acid(s) also provides colorimetric signal under these conditions. As a consequence, an increase in “glycoconjugate-bound” signal does not guarantee that cell-surface sialylation has increased (lectin binding analysis can be used to address this issue in more depth), merely that sialic acid has been converted to the nucleotide sugar form within the cell. Accurate cell enumeration is another critical parameter. Conducting measurements in triplicate minimizes errors and yields results accurate to ±2% (generally expressed as molecules of sialic acid per cell) when cell counting is performed with an electronic cell counter. Moreover, a cell counter can provide cell size information that allows moleculesper-cell data to be easily converted to molar concentrations within a cell. If an electronic cell counter is not available and cell counts are performed with a hemacytometer, our suggestion is to reserve a portion of the cell sample for standard protein assays and express the amount of sialic acid as a function of protein abundance.
Non-Natural Glycan Display
Quantification of cell-surface glycoconjugates Accurate quantitative labeling of surface thiols and ketone groups (Basic Protocol 4) depends on several variables, which include optimal reagent concentrations, incubation times, and reaction temperatures. It is recommended that reagents—e.g., the working solutions of TCEP, MB, biotin hydrazide, and FITCavidin—be prepared in small amounts on the day of the assay and stored at 4◦ C or on ice
until immediately prior to use. Further optimization may be required when a previously untested cell line is used. For example, certain cell types may be less tolerant to the reaction conditions, thereby necessitating shorter labeling periods to maintain cell viability. Similarly, while the basic methodology described for thiols and ketones can be readily adapted for additional functional groups now available in MOE analogs, customization of reagent concentrations and reaction times is required for each analog, chemoselective ligation partner (and other chemical reagents used), and cell type under test. When analog cytotoxicity reduces cell growth, the resulting low number of cells at the conclusion of the incubation period can be compensated by adjusting culture conditions appropriately. In general, increased starting densities of cells ameliorate growth inhibition and help maintain cell viability (Jones et al., 2004). When adherent cells are used, trypsinization and re-plating often result in a growth lag and increased sensitivity to sugar analog toxicity, which can significantly reduce the final cell count. One solution is to plate the cells and allow them to resume growth for 12 to 24 hr before introducing sugar analog into the culture medium. This approach may reduce analog uptake efficiency, and higher analog concentrations will be required to achieve targeted levels of surface display (Sampathkumar et al., 2006b). A common problem, which can be easily detected when samples are run in triplicate, is inconsistent measurement of cellsurface glycan levels. As noted in Basic Protocol 4, step 14, and discussed elsewhere in detail (Yarema 2002), erratic and lowerthan-expected signal may be due to the incomplete removal of unreacted biotin reagents from the cell suspension before staining with FITC-avidin. To avoid this problem, samples should be washed at least three times to remove unreacted biotin reagents before FITC-avidin staining is performed. Additionally, staining can be performed twice to aid in the removal of excess biotin reagents. The authors have encountered situations when surface labeling is much higher than expected when adherent cells were used. This outcome can result from detachment reagents (e.g., trypsin), used while harvesting the cells, that damage the membrane and allow nonspecific uptake of labeling reagents. It is therefore recommended that alternate detachment reagents (e.g., nonenzymatic buffers instead of trypsin) or a cell scraper be used for detachment of cells.
190 Volume 2
Current Protocols in Chemical Biology
However, for some cell lines, cell scraping causes mechanical damage to cells, so again, procedures for each cell line may need to be optimized.
Anticipated Results Cell viability assays Typically, monosaccharide analogs used in metabolic labeling have either negligible or relatively minor cytotoxicity that does not deleteriously impact cell viability at concentrations required to label cellular glycans. In cases when analogs, such as peracetylated ManNLev, do exhibit noticeable toxicity (Kim et al., 2004), the worst of these effects can be avoided by straightforward procedures such as controlling the rate of analog delivery (Aich et al., 2010) or avoiding very low cell densities (Jones et al., 2004). Periodate-resorcinol colorimetric assays The periodate-resorcinol assay provides a quantitative measure of the total sialic acid content and the glycoconjugate-bound sialic acid content of cells. When compared to untreated cells, it is common that the total sialic acid content of cells treated with “high-flux” analogs when a natural “core” sugar is used (such as ManNAc), or when one with a modest structural alteration (e.g., ManNProp) is used, can increase by several fold, often 10-fold or more (Kim et al., 2004). Although the assay does not explicitly provide levels of the free monosaccharide form of sialic acid, this endpoint can be easily estimated by subtracting glycoconjugate-bound sialic acids from total levels. Generally, the level of free sialic acid cannot be measured accurately in untreated cells by this assay because it is below the detection limit, which is roughly ∼5 μM intracellular levels. Upon analog supplementation, the free sialic acid levels increase dramatically to as high as the millimolar level. A large increase in intracellular levels of free sialic acid usually does not translate into a corresponding increase in surface display. For example, a 100-fold increase may only result in a 10% increase in sialoglycoconjugate display because of severe bottlenecks in converting the free sugar into the nucleotide sugar donor, transport into the Golgi lumen, and “empty/available” site on nascent glycans. By contrast, as noted earlier, certain analogs do not measurably change total levels of sialic acid in a cell but do transit the pathway as evidenced by appearance on the cell surface
when assayed as described in Basic Protocol 4. Overall, it is important to be aware that there is not a linear correspondence between analog uptake and incorporation into a metabolic pathway (measured in Basic Protocol 3) and subsequent surface display (measured in Basic Protocol 4). Quantitation of cell-surface glycoconjugates Multiple factors can affect the amount of sugar analogs incorporated and displayed on the cell surface, including the cell type, cell density, incubation time, sugar analog, and metabolic status of the cell (Jones et al., 2004). Also, it should be noted that although the procedure described is highly reproducible, the exact fluorescence intensity value (geometric mean value) obtained from the flow cytometer will vary based on instrument settings. For standard conditions described in this unit, where actively growing Jurkat cells are incubated with 40 μM of 1 for 48 hr, an increase in fluorescence intensity of up to about 10-fold is typically observed.
Time Considerations Incubation of cells with sugar analogs The time required to set up Basic Protocol 1 should be less than 2 hr once the sugar analog stock solution has been obtained. Cells are typically incubated with analogs for 2 to 5 days depending on subsequent experiments performed. Note that if adherent cells are used, it is advisable to plate the cells and allow them to recover and grow for 12 to 24 hr before addition of sugar analogs into the culture medium. Cell viability assays After initial incubation with sugar analogs (Basic Protocol 1), cell viability assays typically take an additional 2 to 5 days, although, if long-term toxicity is being evaluated, this endpoint requires up to 15 days to complete. Cell counting for suspension cells is rather straightforward and should not take longer than an hour for each time point (day 3, 5, and 15). Dispersion of cell clumps and changing of medium on day 7, 9, 11, and 13 is also straightforward and can be completed in half an hour per time point. If adherent cells are used, growth inhibition assays typically take 5 days to complete, with cell counting being performed on days 3 and 5. Although the cells will require trypsinization and resuspension before counting, this procedure can be completed within an hour per time point.
Non-Natural Glycan Display
191 Current Protocols in Chemical Biology
Volume 2
Periodate-resorcinol assay In general, after initial incubation with sugar analogs (Basic Protocol 1), the complete assay can be performed in about 6 to 7 hours for 36 to 48 samples. The resuspension, counting, and initial washes followed by three freeze-thaw cycles of the cells should take about 3.5 hr. Reagent preparation can be done concurrently with the freeze-thaw cycles, and therefore does not require additional time. After freeze-thaw cycles are completed, periodic acid oxidation on ice requires about 30 min (to determine total sialic acid content) or about 2 hr (at 37◦ C) to determine conjugate-bound sialic acid. Finally, incubation with resorcinol working reagent, followed by addition of t-butyl alcohol and measurement of absorbance, should take an additional hour. The total time for a set of periodate-resorcinol assays is 8 to 12 hr. Quantitation of cell-surface glycoconjugates After initial incubation with sugar analogs (Basic Protocol 1), the resuspension of cells and conjugation with biotin reagents (TCEP treatment and biotin maleimide for thiol groups or biotin hydrazide for ketone groups) should take about 3 hr for 24 to 36 samples. After these conjugation reactions are complete, washing of cells to remove unreacted biotinylation reagents, followed by staining of cells with FITC-avidin and additional washes to remove unbound reagent, takes another 90 min. Finally, flow cytometry will take up to 2 hr depending on the number of samples. Overall, the entire process can be completed in about 6 to 8 hr.
Acknowledgements The authors would like to thank the National Institutes of Health (NCI, CA112314-05 and NIBIB, EB005692-04) for financial support.
Literature Cited Aich, U. and Yarema, K.J. 2008. Metabolic oligosaccharide engineering: Perspectives, applications, and future directions. In Glycosciences. 2nd ed. (B. Fraser-Reid, K. Tatsuta, and J. Thiem, eds.) pp. 2136-2190. SpringerVerlag, Berlin, Heidelberg. Aich, U., Campbell, C.T., Elmouelhi, N., Weier, C.A., Sampathkumar, S.-G., Choi, S.S., and Yarema, K.J. 2008. Regioisomeric SCFA attachment to hexosamines separates metabolic flux from cytotoxcity and MUC1 suppression. ACS Chem. Biol. 3:230-240. Non-Natural Glycan Display
Aich, U., Meledeo, M.A., Sampathkumar, S.-G., Fu, J., Jones, M.B., Weier, C.A., Chung, S.Y., Tang, B.C., Yang, M., Hanes, J., and Yarema,
K.J. 2010. Development of delivery methods for carbohydrate-based drugs: controlled release of biologically-active short chain fatty acidhexosamine analogs. Glycoconjug. J. 27:445459. Bond, M.R., Zhang, H., Vu, P.D., and Kohler, J.J. 2009. Photocrosslinking of glycoconjugates using metabolically incorporated diazirinecontaining sugars. Nat. Protoc. 4:1044-1063. Brossmer, R. and Gross, H.J. 1994. Fluorescent and photoactivatable sialic acids. Meth. Enzymol. 247:177-193. B¨uttner, B., Kannicht, C., Schmidt, C., L¨oster, K., Reutter, W., Lee, H.-Y., N¨ohring, S., and Horstkorte, R. 2002. Biochemical engineering of cell surface sialic acids stimulates axonal growth. J. Neurosci. 22:8869-8875. Campbell, C.T., Sampathkumar, S.-G., Weier, C., and Yarema, K.J. 2007. Metabolic oligosaccharide engineering: perspectives, applications, and future directions. Mol. Biosys. 3:187-194. Chang, P.V., Prescher, J.A., Hangauer, M.J., and Bertozzi, C.R. 2007. Imaging cell surface glycans with bioorthogonal chemical reporters. J. Am. Chem. Soc. 129:8400-8401. Chefalo, P., Pan, Y.-B., Nagy, N., Harding, C., and Guo, Z.-W. 2004. Preparation and immunological studies of protein conjugates of N-acylneuraminic acids. Glycoconjug. J. 20:407-414. Cohen, M., Joester, D., Geiger, B., and Addadi, L. 2004. Spatial and temporal sequence of events in cell adhesion: From molecular recognition to focal adhesion assembly. ChemBioChem 5:13931399. Collins, B.E., Fralich, T.J., Itonori, S., Ichikawa, Y., and Schnaar, R.L. 2000. Conversion of cellular sialic acid expression from N-acetyl- to Nglycolylneuraminic acid using a synthetic precursor, N-glycolylmannosamine pentaacetate: Inhibition of myelin-associated glycoprotein binding to neural cells. Glycobiology 10:1120. Du, J. and Yarema, K.J. 2010. Carbohydrate engineered cells for regenerative medicine. Adv. Drug Deliv. Rev. 62:671-682. Du, J., Meledeo, M.A., Wang, Z., Khanna, H.S., Paruchuri, V.D., and Yarema, K.J. 2009. Metabolic glycoengineering: Sialic acid and beyond. Glycobiology 19:1382-1401. Elmouelhi, N., Aich, U., Paruchuri, V.D.P., Meledeo, M.A., Campbell, C.T., Wang, J.J., Srinivas, R., Khanna, H.S., and Yarema, K.J. 2009. Hexosamine template: A platform for modulating gene expression and for sugar-based drug discovery. J. Med. Chem. 52:2515-2530. Freitas, R.A. Jr. 1999. Nanomedicine, Volume I: Basic Capabilities. Landes Bioscience, Georgetown, Texas. Gagiannis, D., Gossrau, R., Reutter, W., Zimmermann-Kordmann, M., and Horstkorte, R. 2007. Engineering the sialic acid in organs of mice using N-propanoylmannosamine. Biochim. Biophys. Acta 1770:297-306.
192 Volume 2
Current Protocols in Chemical Biology
Han, S., Collins, B.E., Bengtson, P., and Paulson, J.C. 2005. Homo-multimeric complexes of CD22 revealed by in situ photoaffinity proteinglycan crosslinking. Nat. Chem. Biol. 1:93-97.
Lemieux, G.A. and Bertozzi, C.R. 1998. Chemoselective ligation reactions with proteins, oligosaccharides and cells. Trends Biotechnol. 16:506513.
Hang, H.C. and Bertozzi, C.R. 2001. Ketone isosteres of 2-N-acetamidosugars as substrates for metabolic cell surface engineering. J. Am. Chem. Soc. 123:1242-1243.
Lemieux, G.A. and Bertozzi, C.R. 2001. Modulating cell surface immunoreactivity by metabolic induction of unnatural carbohydrate antigens. Chem. Biol. 8:265-275.
Horstkorte, R., Rau, K., Laabs, S., Danker, K., and Reutter, W. 2004. Biochemical engineering of the N-acyl side chain of sialic acid leads to increased calcium influx from intracellular compartments and promotes differentiation of HL60 cells. FEBS Lett. 571:99-102.
Luchansky, S.J., Argade, S., Hayes, B.K., and Bertozzi, C.R. 2004. Metabolic functionalization of recombinant glycoproteins. Biochemistry 43:12358-12366.
Jacobs, C.L., Goon, S., Yarema, K.J., Hinderlich, S., Hang, H.C., Chai, D.H., and Bertozzi, C.R. 2001. Substrate specificity of the sialic acid biosynthetic pathway. Biochemistry 40:1286412874. Jones, M.B., Teng, H., Rhee, J.K., Baskaran, G., Lahar, N., and Yarema, K.J. 2004. Characterization of the cellular uptake and metabolic conversion of acetylated N-acetylmannosamine (ManNAc) analogs to sialic acids. Biotechnol. Bioeng. 85:394-405. Jourdian, G.W., Dean, L., and Roseman, S. 1971. The sialic acids. XI. A periodate-resorcinol method for the quantitative estimation of free sialic acids and their glycosides. J. Biol. Chem. 246:430-435. Kayser, H., Zeitler, R., Kannicht, C., Grunow, D., Nuck, R., and Reutter, W. 1992. Biosynthesis of a nonphysiological sialic acid in different rat organs, using N-propanoyl-D-hexosamines as precursors. J. Biol. Chem. 267:16934-16938. Keppler, O.T., Horstkorte, R., Pawlita, M., Schmidt, C., and Reutter, W. 2001. Biochemical engineering of the N-acyl side chain of sialic acid: Biological implications. Glycobiology 11:11R18R. Khidekel, N., Ficarro, S.B., Peters, E.C., and HsiehWilson, L.C. 2004. Exploring the O-GlcNAc proteome: Direct identification of O-GlcNAcmodified proteins from the brain. Proc. Natl. Acad. Sci. U.S.A. 101:13132-13137. Kim, E.J., Sampathkumar, S.-G., Jones, M.B., Rhee, J.K., Baskaran, G., and Yarema, K.J. 2004. Characterization of the metabolic flux and apoptotic effects of O-hydroxyl- and N-acetylmannosamine (ManNAc) analogs in Jurkat (human T-lymphoma-derived) cells. J. Biol. Chem. 279:18342-18352. King, E.J. and Garner, R.J. 1947. The colorimetric determination of glucose. J. Clin. Pathol. 1:3033. Kontou, M., Bauer, C., Reutter, W., and Horstkorte, R. 2008. Sialic acid metabolism is involved in the regulation of gene expression during neuronal differentiation of PC12 cells. Glycoconjug. J. 25:237-244. Lee, J.H., Baker, T.J., Mahal, L.K., Zabner, J., Bertozzi, C.R., Wiemar, D.F., and Welsh, M.J. 1999. Engineering novel cell surface receptors for virus-mediated gene transfer. J. Biol. Chem. 274:21878-21884.
Mahal, L.K., Yarema, K.J., and Bertozzi, C.R. 1997. Engineering chemical reactivity on cell surfaces through oligosaccharide biosynthesis. Science 276:1125-1128. Mahal, L.K., Yarema, K.J., Lemieux, G.A., and Bertozzi, C.R. 1999. Chemical approaches to glycobiology: Engineering cell surface sialic acids for tumor targeting. In Sialobiology and Other Novel Forms of GlycosylationY. Inoue, Y.C. Lee, and F.A. Troy II, eds.) pp. 273-280. Gakushin Publishing Company, Osaka, Japan. Nauman, D.A. and Bertozzi, C.R. 2001. Kinetic parameters for small-molecule drug delivery by covalent cell surface targeting. Biochim. Biophys. Acta 1568:147-154. Piller, V., Piller, F., and Fukuda, M. 1990. Biosynthesis of truncated O-glycans in the T cell line Jurkat. Localization of O-glycan initiation. J. Biol. Chem. 265:9264-9271. Pilobello, K.T., Krishnamoorthy, L., Slawek, D., and Mahal, L.K. 2005. Development of a lectin microarray for the rapid analysis of protein glycopatterns. ChemBioChem 6:985-989. Sampathkumar, S.-G., Jones, M.B., Meledeo, M.A., Campbell, C.T., Choi, S.S., Hida, K., Gomutputra, P., Sheh, A., Gilmartin, T., Head, S.R., and Yarema, K.J. 2006a. Targeting glycosylation pathways and the cell cycle: Sugardependent activity of butyrate-carbohydrate cancer prodrugs. Chem. Biol. 13:1265-1275. Sampathkumar, S.-G., Jones, M.B., and Yarema, K.J. 2006b. Metabolic expression of thiolderivatized sialic acids on the cell surface and their quantitative estimation by flow cytometry. Nat. Protoc. 1:1840-1851. Sampathkumar, S.-G., Li, A.V., Jones, M.B., Sun, Z., and Yarema, K.J. 2006c. Metabolic installation of thiols into sialic acid modulates adhesion and stem cell biology. Nat. Chem. Biol. 2:149152. Sampathkumar, S.-G., Li, A.V., and Yarema, K.J. 2006d. Synthesis of non-natural ManNAc analogs for the expression of thiols on cell surface sialic acids. Nat. Protoc. 1:2377-2385. Sawa, M., Hsu, T.-L., Itoh, T., Sugiyama, M., Hanson, S.R., Vogt, P.K., and Wong, C.-H. 2006. Glycoproteomic probes for fluorescent imaging of fucosylated glycans in vivo. Proc. Natl. Acad. Sci. U.S.A. 103:12371-12376. Saxon, E. and Bertozzi, C.R. 2000. Cell surface engineering by a modified Staudinger reaction. Science 287:2007-2010.
Non-Natural Glycan Display
193 Current Protocols in Chemical Biology
Volume 2
Schilling, B., Goon, S., Samuels, N.M., Gaucher, S.P., Leary, J.A., Bertozzi, C.R., and Gibson, B.W. 2001. Biosynthesis of sialylated lipooligosaccharides in Haemophilus ducreyi is dependent on exogenous sialic acid and not mannosamine: Incorporation studies using N-acylmannosamine analogs, N-glycolylneuraminic acid, and 13C-labeled Nacetylneuraminic acid. Biochemistry 40:1266612677. Schmidt, C., Stehling, P., Schnitzer, J., Reutter, W., and Horstkorte, R. 1998. Biochemical engineering of neural cell surfaces by the synthetic Npropanoyl-substituted neuraminic acid precursor. J. Biol. Chem. 273:19146-19152. Sussich, F. and Cesaro, A. 2000. The kinetics of periodate oxidation of carbohydrates: A calorimetric approach. Carbohydr. Res. 329:87-95. Tanaka, Y. and Kohler, J.J. 2008. Photoactivatable crosslinking sugars for capturing glycoprotein interactions. J. Am. Chem. Soc. 130:3278-3279. Tiziani, S., Sussich, F., and Ces`aro, A. 2003. The kinetics of periodate oxidation of carbohydrates 2. Polymeric substrates. Carbohydr. Res. 338:1083-1095. Varki, A. 1993. Biological roles of oligosaccharides: All of the theories are correct. Glycobiology 3:97-130. Villavicencio-Lorini, P., Laabs, S., Danker, K., Reutter, W., and Horstkorte, R. 2002. Biochemical engineering of the acyl side chain of sialic acids stimulates integrin-dependent adhesion of HL60 cells to fibronectin. J. Mol. Med. 80:671677. Viswanathan, K., Lawrence, S., Hinderlich, S., Yarema, K.J., Lee, Y.C., and Betenbaugh, M. 2003. Engineering sialic acid synthetic ability
into insect cells: Identifying metabolic bottlenecks and devising strategies to overcome them. Biochemistry 42:15215-15225. Walborg, E.F. Jr. and Christensson, L. 1965. A colorimetric method for the quantitative determination of monosaccharides. Anal. Biochem. 13:186-193. Wang, Z., Du, J., Che, P.-L., Meledeo, M.A., and Yarema, K.J. 2009. Hexosamine analogs: From metabolic glycoengineering to drug discovery. Curr. Opin. Chem. Biol. 13:565-572. Weinbaum, S., Tarbell, J.M., and Damiano, E.R. 2007. The structure and function of the endothelial glycocalyx layer. Annu. Rev. Biomed. Eng. 9:121-167. Yarema, K.J. 2002. A metabolic substrate-based approach to engineering new chemical reactivity into cellular sialoglycoconjugates. In Cell Engineering 3. Glycosylation (M. Al-Rubeai, ed.) pp. 171-196. Kluwer Academic Publishers, Dordrecht, The Netherlands. Yarema, K.J., Mahal, L.K., Bruehl, R.E., Rodriguez, E.C., and Bertozzi, C.R. 1998. Metabolic delivery of ketone groups to sialic acid residues: Application to cell surface glycoform engineering. J. Biol. Chem. 273:3116831179. Zanghi, J.A., Mendoza, T.P., Knop, R.H., and Miller, W.M. 1998a. Ammonia decreases NCAM polysialylation in Chinese hamster ovary and small cell lung cancer cells. J. Cell. Physiol. 177:248-263. Zanghi, J.A., Mendoza, T.P., Schmelzer, A.E., Knop, R.H., and Miller, W.M. 1998b. Role of nucleotide sugar pools in the inhibition of NCAM polysialylation by ammonia. Biotechnol. Prog. 14:834-844.
Non-Natural Glycan Display
194 Volume 2
Current Protocols in Chemical Biology
High-Throughput Assessment of Bacterial Growth Inhibition by Optical Density Measurements Jennifer Campbell1 1
Department of Microbiology and Molecular Genetics, Harvard Medical School, Boston, Massachusetts
ABSTRACT The increasing incidence of antibiotic-resistant bacterial infections both in hospitals and in the community intensifies the need for new antibacterial strategies and targets. Although high-throughput screening against live bacteria allows rapid discovery of compounds with growth-inhibitory activities, these efforts have failed to fill the pipeline with the anticipated antibacterial compounds because target identification is often onerous. Recently, a strategy was reported that employs a bacterial growth inhibition assay readout using optical density measurements on paired strains—both a wild-type strain and a pathway-null mutant—to find inhibitors of wild-type bacterial growth that specifically target conditionally essential enzymes in the pathway of interest. Protocols are provided here for determining the robustness of an assay, screening in a high-throughput format, and setting up dose-response curves in paired Staphylococcus aureus strains. However, the protocols can be used to screen for growth-inhibitory compounds in any bacterial strain C 2010 by John Wiley & Sons, Inc. of interest. Curr. Protoc. Chem. Biol. 2:195-208 Keywords: high-throughput screening r Staphylococcus aureus r optical density r cell-based assay r pathway specific screen r bacterial growth inhibition
INTRODUCTION The increasing incidence of antibiotic-resistant bacterial infections in both hospitals and the community has intensified the need for new antibacterial strategies and targets (Walsh, 2003; Grundmann et al., 2006). A relatively new idea for targeting bacteria is to inhibit virulence factors (Marra, 2004; Clatworthy et al., 2007). Broadly defined, virulence factors are required for a pathogenic bacterium to establish an infection in a host, but are not required for its survival in vitro. Although inhibiting virulence factors is a promising antibacterial strategy, discovery of inhibitors that block their synthesis has been difficult. Several challenges include reconstituting the enzymatic machinery, obtaining lipid-linked substrates, and developing appropriate high-throughput assays (Inglese et al., 2007). The protocols in this article describe a newly reported screening strategy that circumvents these challenges and allows for the discovery of small molecules that specifically target downstream enzymes within certain virulence-factor biosynthetic pathways in a cellular context. Many pathogenic bacteria synthesize surface-displayed, polysaccharide-based polymers on bactoprenol carrier lipids that function as virulence factors, including exopolysaccharides, O-antigens, capsular polysaccharides, and wall teichoic acids. Interestingly, many of these virulence-factor biosynthetic pathways share a common feature that lends itself to a general screening strategy: they contain conditionally essential genes. That is to say, once flux into the biosynthetic pathway has started, blocking it is detrimental to the organism. The fact that the polymers are nonessential to the bacteria means that the initiating genes are expendable. The downstream genes, however, are essential unless
Current Protocols in Chemical Biology 2: 195-208, October 2010 Published online October 2010 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/9780470559277.ch100115 C 2010 John Wiley & Sons, Inc. Copyright
HTS for Bacterial Growth Inhibitors
195 Volume 2
Table 1 Concept Behind Paired Strain Screening: Small Molecules that Inhibit the Growth of the Wild-Type but not the Pathway-Null Mutant Should Target the Conditionally Essential Enzymes in the Nonessential Biosynthetic Pathway of Interest
Wild-type
Mutanta
Growth
Growth
No growth
No growth
No growth
Growth
Compound does not block any cellular process Compound blocks an essential cellular process b
Compound blocks a conditionally essential enzyme
a The first enzyme in a nonessential pathway of interest is inactive. b Growth will be uninhibited in the mutant strain if the compound hits a conditionally essential enzyme within
the deleted pathway.
peptidoglycan exopolysaccharides enterobacterial common antigen capsular polysaccharides wall teichoic acids
peptidoglycan biosynthesis
bactoprenol
peptidoglycan attachment
Gtf1 D
Gtf2 D
D
D
Figure 1 Schematic showing the mixed dispensability present in the biosynthetic pathways of many Gram-positive cell-wall surface polymers that are assembled on a bactoprenol carrier lipid and act as virulence factors. The first two nonessential genes are typically glycosyltransferases (Gtfs), which are depicted here as light gray arrows; downstream genes shown as dark gray arrows are conditionally essential, that is, they cannot be deleted except in strains that lack one of the initiating Gtfs.
polymer initiation is prevented (see Fig. 1). Hence, stably maintained deletions of conditionally essential genes are typically accompanied by suppressor mutations that prevent committed flux into these biosynthetic pathways (Yuasa et al., 1969; Katzen et al., 1998; Burrows and Lam, 1999; D’Elia et al., 2006; Xayarath and Yother, 2007). The mixed gene dispensability pattern in these bactoprenol-dependent biosynthetic pathways can be exploited in a high-throughput screen to discover inhibitors of the downstream, conditionally essential enzymes. Chemicals are tested for growth-inhibitory activities against paired strains—the wild-type strain and a mutant in which flux through the pathway has been prevented (and the downstream enzymes are no longer essential). Those molecules that stop wild-type growth, but do not inhibit the growth of the pathway-null mutant, should be selective for the conditionally essential enzymes in the pathway of interest (see Table 1). Screening against paired strains combines advantages of both in vitro and in vivo screening by limiting the targets to the conditionally essential enzymes within the desired pathway and by identifying active compounds in a cellular context. The utility of this screening strategy has been demonstrated by the discovery of an inhibitor of wall teichoic acid biosynthesis in the clinically relevant pathogen S. aureus (see Background Information and Swoboda et al., 2009, 2010). HTS for Bacterial Growth Inhibitors
The protocols outlined here allow one to design, optimize, and carry out a high-throughput screen against paired bacterial strains to find growth-inhibitory compounds that target
196 Volume 2
Current Protocols in Chemical Biology
certain enzymes within the virulence-factor pathway of interest. A simple assay using an optical density readout to measure the extent of bacterial growth is used. Calculation of Z values for the assay (Basic Protocol 1) allows determination of whether the chosen incubation conditions allow adequate and reproducible growth. The more robust the assay, the easier it becomes to identify screening positives with confidence. Basic Protocol 2, “Screening for growth-inhibitory compounds,” describes the steps involved in setting up a high-throughput screen. It is assumed that the screener has access to a chemical library and a pin transfer robot. Once the data have been analyzed and screening positives have been identified, Basic Protocol 3 outlines the setup of 6-point dose-response curves for secondary screening. Using this protocol, it is possible to determine the minimum inhibitory concentrations (MICs) of 48 compounds in a single 384-well plate.
STRATEGIC PLANNING Acquiring Bacterial Strains Mixed gene dispensability is frequently observed in biosynthetic pathways that have two things in common: (1) they make cell surface polymers and (2) they assemble them on a bactoprenol carrier. Many of these cell surface polymers function as virulence factors and share key similarities in their biosynthetic pathways: the first two genes can be knocked out with the strain remaining viable, but downstream genes cannot be knocked out unless suppressor mutations that inactivate one of the first two enzymes are present (Fig. 1). Theoretically, the screening strategy described here should be applicable to all of these biosynthetic pathways. In order to screen paired bacterial strains for inhibitors of downstream enzymes within virulence-factor biosynthetic pathways, it is first necessary to confirm that these enzymes are conditionally essential. To do this, the first gene in the pathway is placed under an inducible promoter and then, in the absence of inducer, knockouts of the downstream genes are made. Growth curves of the resulting strains are then acquired in the absence and presence of inducer. Knockout strains of conditionally essential enzymes will be able to grow in the absence of inducer, but not in its presence (D’Elia et al., 2006; Swoboda et al., 2009). Screening is then performed against a wild-type strain and a mutant in which the first enzyme in the targeted biosynthetic pathway is inactive. NOTE: If the goal is to screen small molecules for growth-inhibitory activities against only a single bacterial strain, the protocols outlined below can easily be adapted for this purpose.
Screening in Duplicate The majority of high-throughput screens show a high degree of intrinsic variability and error. It is therefore recommended that these assays be run in duplicate in a second set of assay plates. This assay duplication is built into the high-throughput screening protocol described here. Acquiring duplicate data points for each library compound can reduce the rates of false positives by as much as 50%. Biosafety Considerations If possible, it is best to screen against bacterial strains that have a Biosafety Level 1 (BSL-1) rating. These strains can be handled on the benchtop, and no additional safety measures need to be taken. However, if screening against a BSL-2-rated pathogen, one must handle this strain in a biosafety cabinet at all times unless it is in a sealed container. When carrying out a screen using a bacterial strain that requires biocontainment, initial medium dispensing and pin transfer of compounds can be performed on the benchtop, but once the plates have been inoculated with the bacteria, they must be transported and incubated in leak-proof containers (e.g., Nalgene, cat. no. 7135). Because these
HTS for Bacterial Growth Inhibitors
197 Current Protocols in Chemical Biology
Volume 2
containers are airtight, only ∼10 assay plates can be incubated in each one. Otherwise, because of limited oxygen, the negative controls would not grow sufficiently for optimal assay conditions. To obtain assay data when growing a BSL-2 strain, a plate reader would need to be used inside a biosafety cabinet and decontaminated before removal. Assay plates then need to be disposed of properly. These extra precautions can greatly reduce the throughput of an assay.
Selective Antibiotics Contamination can be introduced into an assay when liquid dispensing and pin transfer steps occur on the benchtop. If the contaminating bacteria are sensitive to the positive control compound, it is almost impossible to detect their presence through standard optical density measurements. Hence, it is suggested that selective antibiotics be used throughout the screen. They should be present in the medium at all times (during both pin transfer and bacterial inoculation). Plasmid-borne selective markers have been successfully used to avoid contamination during screening (see Background Information and Swoboda et al., 2009). BASIC PROTOCOL 1
Z DETERMINATION TO ASSESS SCREENING ASSAY ROBUSTNESS For high-throughput screening, it is necessary to ensure that the readout window be sufficiently large and reproducible to enable the identification of screening positives with confidence. A standard parameter used to assess the quality and robustness of an assay is the Z value (Zhang et al., 1999).
Z′ =1−
3σneg + 3σpos μ neg − μ pos
Equation 1
This parameter requires a mean (μ) and standard deviation (σ ) for both the positive and negative controls. Increasing the sample size improves the statistical significance, so half of a 384-well plate should be used for each control condition. An assay should have a Z of at least 0.5 to be considered robust enough for high-throughput screening, while an assay with a Z greater than 0.7 is considered excellent.
Materials Sterile tryptic soy broth (TSB; 30 g/l; sterile filtered or autoclaved; BD Difco, cat. no. 211823) plus selective antibiotic, if appropriate Plated bacterial strain(s) of interest (e.g., S. aureus) Positive control compound that inhibits growth of bacterial strain (e.g., for S. aureus, erythromycin at 10 mg/ml in ethanol) 30◦ C shaking incubator Multichannel pipettor (volume range from 20 to 100 μl) with aerosol-barrier pipet tips (VWR, cat. no. 53510-106) and sterile solvent reservoirs (VWR, cat. no. 89094-680) 384-well clear-bottom plate (Corning, cat. no. 3702) Low-evaporation lid (Corning, cat. no. 3009) Plate reader capable of measuring optical density at 600 nm (OD600 ) in a 384-well format (e.g., EnVision from PerkinElmer)
HTS for Bacterial Growth Inhibitors
198 Volume 2
1. Inoculate 2 ml of TSB medium with a single colony of bacteria and grow to saturation by shaking at 30◦ C overnight. For paired strain screening, it is necessary to establish a Z for each strain (wild-type and mutant). Make sure to include selective antibiotics if necessary. Current Protocols in Chemical Biology
9 20 7 18 1 5 16 1 3 14 1 1 2 1 10 11 8 9 6 7 4 5 3 1 2
A B C D E F G H I
21 22
23 24
positive control wells negative control wells
J K L M N O P
Figure 2 Schematic showing a 384-well plate layout for Z determination. In a growth-inhibitory assay, the positive control wells will most likely contain bacteria in medium containing erythromycin (10 μg/ml) or another appropriate antibiotic, while the negative control wells will be filled with medium and bacteria only.
2. Dilute the bacteria to ∼1 × 106 cfu/ml in 40 ml of TSB medium. For S. aureus, this is typically a 1/1000 dilution from a saturated culture; however, the proper dilution for the bacterial strain of interest can be calculated by following the Support Protocol.
3. Split the 40-ml culture in half and add the appropriate concentration of positive control compound to one of the 20-ml cultures. Retain the other 20 ml for the negative control wells. For S. aureus, erythromycin at 10 μg/ml final concentration can be used as the positive control (1/1000 dilution of stock).
4. Using a multichannel pipettor and solvent reservoir, dispense 80 μl of positive control culture into all of the wells in the top half of a 384-well plate (see Fig. 2). Use one 384-well plate per strain. Be careful to avoid dispensing air bubbles at this point. They will not pop overnight, and will affect the optical density and, ultimately, the Z value of the assay. Alternatively, use the liquid handling system (plate filler) that will be employed during screening, as this will equalize volumes and minimize the likelihood of bubbles forming in the bottoms of the wells. If a plate filler is used, scale up the medium to account for the dead volume in the tubing (∼10 ml).
5. Dispense 80 μl of negative control culture into all of the wells in the bottom half of the 384-well plate. 6. Cover the plate with a low-evaporation lid and incubate overnight without shaking (typically 16 to 24 hr) at 30◦ C. The optimal incubation time needs to be determined experimentally. Take OD600 readings of the plate each hour starting at 16 hr (continue incubating at 30◦ C between each reading) and calculate the Z for each time point. Choose an incubation length that yields a maximal Z value (minimal growth of the positive controls and maximal growth of the negative controls).
7. Using a plate reader, measure the optical density of each well at 600 nm and calculate the Z value using Equation 1. It is recommended that this protocol be repeated on three separate occasions to confirm that the growth conditions and controls yield a robust and reproducible assay.
HTS for Bacterial Growth Inhibitors
199 Current Protocols in Chemical Biology
Volume 2
BASIC PROTOCOL 2
SCREENING FOR GROWTH-INHIBITORY COMPOUNDS To find growth-inhibitory compounds of a wild-type strain, compound library plates should be screened in duplicate (see Strategic Planning). To discover inhibitors of conditionally essential enzymes in a nonessential biosynthetic pathway, one should also screen the compound library plates in duplicate against a mutant strain that lacks the first enzyme in the pathway of interest. Those compounds that inhibit growth of the wild-type strain but not the mutant should target the conditionally essential enzymes within the desired pathway (see Table 1). This protocol is designed for one screening session of 40 compound library plates in duplicate against a single bacterial strain of interest, or for 20 compound library plates in duplicate against both a wild-type and a mutant strain (four daughter plates per library plate). For this protocol, it is assumed that columns 23 and 24 of each compound library plate are empty and thus those columns can be used for negative (column 23) and positive (column 24) controls for the assay (Fig. 3). The location of control wells can be changed as appropriate for screens using compound library plates that are formatted differently.
Materials Sterile tryptic soy broth (TSB; 30 g/l; sterile filtered or autoclaved; BD Difco, cat. no. 211823) plus selective antibiotic, if appropriate Plated bacterial strain(s) of interest (e.g., S. aureus) 20 ml TSB medium containing positive control compound at 2× final concentration (typically erythromycin at 20 μg/ml) Compound library plates, with compounds diluted in DMSO at 5 mg/ml or 10 mM stock concentration 1-liter flasks (sterile) 30◦ C shaking incubator Multichannel pipettor (volume range from 20 to 100 μl) with aerosol-barrier pipet tips (VWR, cat. no. 53510-106) and sterile solvent reservoirs (VWR, cat. no. 89094-680) 384-well clear-bottom plates (80 per screening session; Corning, cat. no. 3702) Microplate dispenser (Matrix WellMate or comparable liquid handler; see Rudnicki and Johnston, 2009) Pin transfer robot (Rudnicki and Johnston, 2009) Low-evaporation plate lids (16–20; Corning, cat. no. 3009) Plate reader capable of measuring optical density at 600 nm (OD600 ) in a 384-well format (e.g., EnVision from Perkin Elmer) 1. Inoculate 2 ml of TSB medium with a single colony of the bacteria and grow to saturation by shaking at 30◦ C overnight. Make sure to include selective antibiotics if necessary.
2. Dilute both the wild-type and mutant bacteria to ∼2 × 106 cfu/ml in separate flasks containing 600 ml of TSB medium. For S. aureus, this is typically a 1/500 dilution from a saturated culture; however, the proper dilution for the bacterial strain of interest can be calculated by following the Support Protocol.
Screen the library 3. Label the sides of the 384-well plates with numbers that correspond to compound library plates and with letters that designate replicate plates (e.g., wild-type: A and B; mutant: C and D). HTS for Bacterial Growth Inhibitors
4. Using a plate filler, dispense 40 μl of TSB into columns 1 to 23 of each 384-well plate.
200 Volume 2
Current Protocols in Chemical Biology
positive control wells negative control wells
12 10 11 8 9 6 7 5 3 4 1 2
19 17 18 15 16 13 14
20 21
24 22 23
A B C D E F G H I J K L M N O P
Figure 3 Schematic showing a 384-well plate layout for high-throughput screening. This format assumes that the library compounds are arrayed in columns 1 to 22, allowing for a column each of negative and positive controls. Columns 1 to 23 are filled with 40 μl of medium by the plate filler, while positive control medium (40 μl medium plus erythromycin for a 10 μg/ml final screening concentration) is added manually to column 24 using a multichannel pipettor. Following the pin transfer of small molecules from compound library plates, all wells are filled with 40 μl of medium containing the bacteria of interest.
Wells in column 23 will serve as negative controls (medium plus bacteria only) for each plate.
5. Using a multichannel pipettor and a solvent reservoir, manually add 40 μl of medium containing 2× positive control compound to the wells of column 24 (see Fig. 3). 6. Stack the plates in order according to their labels and load them into the pin transfer robot. 7. Pin transfer 300 nl per well from each compound library plate into four daughter plates (A to D if screening against paired strains; if only one strain is being screened, pin transfer the compound plates into duplicate plates). Assuming the library compounds are stored as 10 mM stocks, a 300-nl pin transfer into a final volume of 80 μl will give a screening concentration of 38 μM.
8. Using a plate filler, add 40 μl of 2 × 106 cfu/ml wild-type bacteria to all wells in plates A and B, and 40 μl of 2 × 106 cfu/ml mutant bacteria to the wells of plates C and D. If paired strains are not being screened, simply make 1.2 liters of 2 × 106 cfu/ml culture and inoculate all plates with the single bacterial strain.
9. Stack the plates no more than five high and cover the top plate with a low-evaporation lid. 10. Incubate the plates at 30◦ C overnight without shaking (16 to 24 hr). Optimize this time parameter during Z determination (Basic Protocol 1).
11. Using a plate reader, measure the optical density of each well at 600 nm. The use of plate stackers during this stage of the protocol will greatly increase the throughput of the assay and will minimize human error.
HTS for Bacterial Growth Inhibitors
201 Current Protocols in Chemical Biology
Volume 2
Analyze the data 12. Calculate the Z value for each plate using Equation 1. 13. Normalize the wells of each plate for percent survival by scaling them to their respective negative and positive controls using Equation 2.
percent survival =
( OD ( OD
sample neg
− OD pos )
− ODpos )
×100
Equation 2
It is important to normalize the optical densities of the sample wells to the controls on that plate, especially if the pathway-null mutant has different growth characteristics than the wild-type. See Critical Parameters for more details.
14. Graph replicate data with sample wells from plate A as the x coordinates and sample wells from plate B as the y coordinates. Those data points that do not fall near a trendline with a slope equal to 1 did not show equal growth-inhibitory activity in duplicate, and caution should be taken in following up on these compounds.
15. Define criteria for a screening “positive” (a typical HTS “hit rate” should be less than 0.5%) and request compounds for secondary screening. For paired screening, “hits” could be defined as any compound that gives <10% survival of the wild-type while giving >50% survival of the pathway-null mutant. BASIC PROTOCOL 3
DOSE-RESPONSE ASSAYS FOR SECONDARY SCREENING OF POTENTIAL BACTERIAL GROWTH INHIBITORS Once screening “positives” have been identified, it is necessary to confirm their activity through secondary screening. Not all active compounds from the primary screen will “re-hit” at this stage. This procedure allows determination of the minimum inhibitory concentrations (MICs) of 48 compounds between 3 and 100 μM in a single 384-well plate, so that they can be ranked in terms of activity for future follow up. If paired strains are being screened, it is advised that the MICs of all cherry-picked compounds be determined against the wild-type strain first. The compounds that reproducibly inhibit growth of the wild-type strain can then be screened against the pathway-null mutant using the same procedure to test for pathway specificity.
Materials Sterile tryptic soy broth (TSB; 30 g/l; sterile filtered or autoclaved; BD Difco, cat. no. 211823) plus selective antibiotic, if appropriate Plated bacterial strain(s) of interest (e.g., S. aureus) Sterile TSB containing 2% (v/v) DMSO (plus selective antibiotic, if appropriate) Sterile TSB containing positive control compound at 2× final concentration
HTS for Bacterial Growth Inhibitors
384-well clear-bottom plates (one per 48 compounds; Corning, cat. no. 3702) Single-channel pipettor (volume range of 0.5 to 2.0 μl) Sterile solvent reservoir (VWR, cat. no. 89094-680) Multi-channel pipettor (volume range of 15 to 30 μl); alternatively use plate filler Aerosol barrier tips that fit the multichannel pipettor (VWR, cat. no. 53510-014) Low-evaporation lids (one per plate; Corning, cat. no. 3009) 30◦ C shaking incubator Plate reader capable of measuring optical density at 600 nm (OD600 ) in a 384-well format (EnVision from PerkinElmer)
202 Volume 2
Current Protocols in Chemical Biology
1. Inoculate 2 ml of TSB medium with a single colony of the bacteria and grow to saturation by shaking at 30◦ C overnight. Make sure to include selective antibiotics if necessary.
2. Dilute the bacteria to ∼2 × 106 cfu/ml in 15 ml of TSB medium and set aside. For S. aureus, this is typically a 1/500 dilution from a saturated culture; however, the proper dilution for the bacterial strain of interest can be calculated by following the Support Protocol.
3. Add 30 μl of TSB to the wells in rows A and I of a labeled 384-well plate. Using a single-channel pipettor, mix a different test compound into each well. Typically 0.6 μl of a 10 mM solution of the compound of interest is mixed into 30 μl of medium to give a 200 μM stock in each well of rows A and I.
4. Using a multichannel pipettor and solvent reservoir, aliquot 15 μl of TSB containing 2% DMSO (v/v) to the wells of rows B to G and J to O. 5. Make a series of five two-fold dilutions down the plate using the multichannel pipettor by sequentially mixing 15 μl of each compound solution into the well directly beneath it (see Fig. 4). For example, 15 μl of the 200 μM stock in well A1 is mixed with the 15 μl of TSB in well B1 to give 30 μl of a 100 μM solution, 15 μl of which is then mixed with the 15 μl of TSB in well C1 to give 30 μl of a 50 μM solution, and so forth. Tips need not be changed as dilutions are made down the plate. The extra 15 μl in well F1 will need to be discarded (along with the tips). This dilution procedure is repeated to dilute all of the wells of row A and row I).
6. Using a multichannel pipettor, add 15 μl medium containing 2× the final concentration of the positive control compound to the wells of rows H and P. Typically 20 μg/ml erythromycin is used for this step when screening against S. aureus.
7. Using a multichannel pipettor or a plate filler, add 15 μl of the bacterial culture made in step 2 to all of the wells in the 384-well plate. Final compound concentrations are shown in Figure 4.
21 22 9 20 7 18 1 5 16 1 1 4 1 13 11 12 9 10 7 8 5 6 4 3 1 2
A B C D E F G H I
23 24
stock 1 :1 1:1 1:1 1:1 1:1 tive nega ive posit 100 50
25 13 6 3 NC PC
J K L M N O P
Figure 4 Schematic showing a 384-well plate layout for dose-response screening of selected compounds. This procedure allows one to determine the MICs of 48 compounds between 3 and 100 μM in a single 384-well plate. Final screening concentrations for the compounds are shown in the schematic to the right of rows I through M. NC and PC are negative control and positive control rows, respectively.
HTS for Bacterial Growth Inhibitors
Current Protocols in Chemical Biology
Volume 2
203
8. Cover the plate with a low-evaporation lid and incubate without shaking at 30◦ C for 16 to 24 hr. 9. Read the optical densities of each well, normalize them to the positive and negative controls using Equation 2, and determine the overall Z value using Equation 1. 10. Plot MIC curves for the active compounds as percent survival versus compound concentration. Open source software is available for batch processing: http://ncgc.nih.gov/pub/openhts/ curvefit/. SUPPORT PROTOCOL
DETERMINING CORRECT BACTERIAL CONCENTRATION FOR A SCREENING ASSAY Antibacterial assays are typically run at bacterial concentrations of 105 or 106 colony forming units (cfu) per milliliter. To achieve the proper dilution, it is necessary to generate a curve that relates bacterial density (OD600 ) to bacteria concentration (cfu/ml). To do this, a bacterial culture is grown until it reaches OD600 ∼1. It is then diluted to four or five different concentrations (to be accurate, they must absorb UV light above baseline), and the bacteria are enumerated on nonselective agar plates. Once calculated, cfu/ml is graphed versus OD600 (see Fig. 5). Once this curve is generated, it is possible to determine the dilution of an overnight culture that will provide the correct bacterial concentration for a screening assay.
Additional Materials (also see Basic Protocols 1, 2, and 3) TSB plates containing 1.5% (w/v) agar Spreaders Sterile 2-ml tubes UV-vis spectrophotometer Cuvettes Graphing software 1. Inoculate 2 ml of TSB medium with a single colony of bacteria and shake at 30◦ C overnight. Make sure to include selective antibiotics if necessary.
2. Dilute the overnight culture of bacteria 1/200 into fresh TSB and shake at 30◦ C for about 3 hr (until OD600 ∼1.0).
cfu/ml
Do not overgrow the bacteria. The turbidity of the culture should be within the linear range of the UV-vis instrument.
cfu/ml ⫽ slope ⫻ OD600
OD600
HTS for Bacterial Growth Inhibitors
Figure 5 Graph depicting bacterial concentration (cfu/ml) versus optical density (OD600 ). Once determined, the slope of the trendline can be used to determine the cellular density of a bacterial culture through a simple optical density measurement.
204 Volume 2
Current Protocols in Chemical Biology
3. Dilute the freshly grown culture 4:1, 3:2, 2:3, 1:4, 1:9 (culture:TSB) into fresh TSB and measure the OD600 of each sample, including the original sample where OD ∼1. 4. Dilute each of the six samples from step 3 to 10–2 by vortexing 10 μl of sample with 990 μl of TSB in a 2-ml tube. Further dilute the samples to 10–4 by vortexing 10 μl of the 10–2 dilutions with 990 μl of TSB. Plate 100 μl of the 10–4 dilution of each sample onto a labeled TSB agar plate. 5. Dilute the solutions from step 4 further by vortexing 100 μl of the 10–4 dilutions with 900 μl TSB (to give 10–5 ); repeat this procedure by vortexing 100 μl of the 10–5 dilutions with 900 μl TSB (to give 10–6 ); plate 100 μl of the resulting 10–5 and 10–6 dilutions onto TSB agar plates. 6. Incubate the plates overnight at 30◦ C and count the resulting colonies. Assuming six samples are plated at three different dilutions (10–4 , 10–5 , and 10–6 ), there should be 18 plates total.
7. Calculate the cfu/ml for each of the original six samples using Equation 3.
cfu/ml =
# colonies on plate dilution × volume plated Equation 3
For example, 44 colonies from 100 μl of a 10−4 dilution would correspond to 4.4 × 106 cfu/ml. Colony counts are significant if there are between 30 and 300 colonies on a plate.
8. Graph data with cfu/ml on the y axis and OD600 on the x axis, setting the intercept equal to zero; add a linear trendline and record the resulting equation (see Fig. 5).
Determining the concentration of a bacterial culture 9. Dilute an overnight culture 1/10 into fresh TSB (to bring it into the linear range of the UV-vis), and measure its OD600 . Multiply this value by the slope of the trendline to calculate cfu/ml present in the 1/10 dilution. Multiply this value by 10 to obtain the cellular density of the overnight culture. Dilute the overnight culture appropriately to obtain 1×106 cfu/ml as the starting inoculum in each assay well for the bacterial growth screening assay. COMMENTARY Background Information The protocols described in this article have been successfully employed to identify inhibitors of the wall teichoic acid (WTA) biosynthetic pathway in Staphylococcus aureus (Swoboda et al., 2009). WTAs are phosphate-rich, carbohydrate-based polymers displayed on the surfaces of Gram-positive bacteria that affect both their physiology and pathogenicity. These anionic polymers are considered virulence factors because they are not essential for growth in vitro but they are required for the establishment of a robust infection. Strains lacking the first enzyme in the pathway have been made and studied, and it has been found that WTA-null S. aureus mutants are unable to cause infections
in certain animal models (Weidenmaier et al., 2004, 2005; Weidenmaier and Peschel, 2008). WTAs are synthesized intracellularly on a bactoprenol carrier lipid and are then transported to the outside of the cell, where they are covalently attached to the cell wall. Similar to other virulence-factor biosynthetic pathways, WTAs contain conditionally essential enzymes. The mixed gene dispensability of this pathway has been described previously (D’Elia et al., 2006). Researchers were unable to knock out the downstream genes in a wild-type background; however, when the first gene of the pathway, tarO, was inactive, the downstream genes were easily removed. The results of these genetic experiments suggested that once flux through the WTA biosynthetic
HTS for Bacterial Growth Inhibitors
205 Current Protocols in Chemical Biology
Volume 2
pathway had started, blocking it was detrimental to the organism. A screening strategy was devised to find growth-inhibitory compounds against wild-type S. aureus that would specifically target the conditionally essential enzymes in the WTA biosynthetic pathway. This was done by running a counter-selection in tandem with the wild-type growth-inhibitory screen. A WTA-null (tarO) mutant was screened alongside the wild-type strain to distinguish between compounds that blocked processes unrelated to WTA biosynthesis and those that inhibited conditionally essential enzymes within the WTA biosynthetic pathway (see Table 1). Because the essentiality of the downstream enzymes is relieved in the tarO mutant, compounds that selectively hit one of these targets will not affect the growth of the mutant strain. Z values were determined for both the wildtype and mutant strains using erythromycin at 10 μg/ml as a positive control and incubating the bacteria at 30◦ C for 18 hr (Z = 0.85). The screen was then performed according to Basic Protocol 2 with a wild-type strain of S. aureus RN4220 that contained a plasmid encoding constitutively expressed green fluorescent protein (GFP) and a tarO mutant that carried a plasmid encoding constitutively expressed mCherry fluorescent protein. Throughout the screen, both plasmids were maintained with 10 μg/ml chloramphenicol, and library compounds were screened at 38 μM. Following overnight incubation, plates were read for both optical density and fluorescence at the appropriate wavelengths. Although each strain was marked with a unique fluorescent marker to provide an alternative readout of bacterial growth, optical density proved to be a more robust measure of growth and was used for hit determination. A total of 55,000 compounds were screened in duplicate against the paired strains. A positive hit from the screen was defined as a compound that resulted in less than 10% survival of the wild-type with greater than 50% survival of the tarO mutant. Forty compounds were picked for dose-response determination (see Basic Protocol 3) and three of these compounds “re-hit” as being both active and selective (Swoboda et al., 2009).
from plates that do not have an acceptable Z value (e.g., <0.5) should be scrutinized further. It may be that some error occurred during screening and the data need to be discarded. However, oftentimes, one or two bad control wells can be enough to greatly reduce the Z . If these improper controls happen randomly, no action need be taken. Ideally, greater confidence can be placed in data obtained from plates with higher Z values. Normalizing wells to the plate median If sufficient positive and negative controls cannot be accommodated on the assay plates because of the format of the library compounds to be screened, it will not be possible to normalize the data according to Equation 2. In this case, the samples should be normalized to themselves. It is assumed that most of the compounds are inactive and essentially act as vehicle controls. Because of this, the sample wells can be normalized to the median of the plate (the middle value when the samples are ranked in order). The median is used instead of the average because high-throughput screening attempts to identify the outliers on a plate. Normalizing to the plate average would thus diminish the apparent activity of the outlying data points or screening “positives” (Brideau et al., 2003). Z scores—determining the validity of a “hit” Up to this point, screening actives have been described as those wells that have the greatest relative activity. However, in some cases it may be useful to also rank “hits” using a statistical scoring method. Use of Z scores can help a screener detect data points that look different from the bulk of the samples (Malo et al., 2006). Identifying a “hit” with both potent activity and a high Z score can increase confidence that it is genuine. Z scores are calculated according to Equation 4, where μplate is the average of the sample wells (no controls) and σ plate is the standard deviation of these values (Brideau et al., 2003). This statistical measure calculates the number of standard deviations away from the mean for each well, and can be used to determine the probability that a screening positive is not due to background noise.
Critical Parameters
HTS for Bacterial Growth Inhibitors
Z values—assessing the success of an assay Obtaining a Z value for each plate during a high-throughput screen can help the screener identify plates that encountered suboptimal assay conditions. The data obtained
Z score =
( OD
sample
− μplate )
σplate
Equation 4
More rigorous statistical methods that can salvage data from an assay with positional
206 Volume 2
Current Protocols in Chemical Biology
effects and other systematic variations have been described (Brideau et al., 2003). However, it is not anticipated that these B scores would need to be calculated for a simple bacterial growth assay.
Troubleshooting Optimal growth conditions Basic Protocol 1 is used to determine whether the assay conditions are suitable for high-throughput screening. With bacterial growth assays, the parameters that can be altered are incubation time and temperature. Too short an incubation time could affect the assay window by not allowing the negative control to fully grow. On the other hand, too long an incubation time could lead to evaporation or to the growth of resistant mutants. Incubation time should be experimentally optimized to maximize the assay window. That being said, the dynamic range of optical density measurements is relatively small (i.e., S. aureus cultures that give optical densities above background while not saturating the detector range from 106 to 107 cfu/ml). If a bacterial strain does not grow to a sufficient density (e.g., >OD600 = 0.6), the assay window will likely be too narrow and the Z value will be negatively affected. One way to avoid this problem is to utilize a more sensitive readout. The advantages and disadvantages of two alternative readouts are briefly described below. Fluorescent proteins Fluorescence can be used to quantify the number of bacteria present in a microtiter plate well because the amount of fluorescent protein expressed in properly engineered bacteria should be proportional to the concentration of bacteria. Plasmids containing constitutively expressed fluorescent protein genes can easily be introduced into the bacterial strain of interest (see above in Background Information). Fluorescent proteins form internal chromophores without the need of additional cofactors, enzymes, or substrates other than molecular oxygen, making them amenable to high-throughput screening. However, because chemical libraries are most often composed of flat aromatic molecules, many of the compounds autofluoresce, causing interference with fluorescence output. This interference can increase the number of false negatives encountered in a high-throughput screen. Luminescence Luminescence is produced by a chemical reaction in which light is generated by an
enzyme, luciferase, acting on its substrate, luciferin. Plasmid-borne genes encoding both the enzyme and the proteins required for substrate synthesis can be introduced into bacterial strains. However, if an appropriate plasmid is not available, the reagent BacTiter Glo (Promega, cat. no. G8230) can be used. This reagent measures the levels of ATP in each well, which should be proportional to the number of bacteria present. The reagent contains the purified enzyme and necessary substrates to produce the reaction. Its use would add a dispensing step to the assay protocol (following overnight incubation) and would also significantly increase costs. Although luminescence gives a weaker signal than fluorescent proteins, its background is negligible, making it as much as 1000 times more sensitive than fluorescent readouts (Inglese et al., 2007). Also, because no excitation energy is needed to generate this signal, compound interference is greatly reduced. However, if present in the chemical library, luciferase inhibitors and luciferin analogs can interfere with the reaction.
Anticipated Results Basic Protocol 1 allows the screener to determine whether the incubation conditions are suitable for adequate growth and minimal variability. Following optimization, this protocol should be repeated on three independent occasions to confirm reproducibility. The assay conditions can then be applied to a high-throughput screen as outlined in Basic Protocol 2. Depending in part on the chemical diversity of the screened library and the bacterial strains used, a “hit” rate of approximately 0.5% should be obtained. Note that the outer membranes of Gram-negative bacteria act as a barrier to decrease the permeability of small molecules into the cells. Because of this added defense, the “hit” rates for Gramnegative screens tend to be much lower than for Gram-positive screens. Basic Protocol 3 describes a method to determine the minimal inhibitory concentrations of screening “positives” identified in Basic Protocol 2. Those compounds that retest as both active and selective (i.e., they inhibit the growth of the wildtype strain but not the pathway-null mutant) can be advanced for further evaluation.
Time Considerations
Z determination can take as little as 24 hr, but should be repeated three times on at least two separate days. If compound library plates are tested against a single strain, 40 compound library plates (14,000 small molecules)
HTS for Bacterial Growth Inhibitors
207 Current Protocols in Chemical Biology
Volume 2
can be screened in duplicate in one day. If paired strains are screened in duplicate (four assay plates/library plate), 20 compound library plates can be screened in a 24-hr period (7000 small molecules). Data analysis can take several days. Once selected compounds are received, dose-response setup for one 384-well plate (48 compounds) takes 2 to 3 hr. MICs can then be determined within 24 hr.
Acknowledgments J.C. is currently funded by Award Number F32AI084316 from the National Institute of Allergy and Infectious Diseases. The author would like to thank Dr. Jonathan G. Swoboda for teaching her the ins and outs of HTS, and the NSRB and ICCB–Longwood staff for tireless technical assistance and support.
Literature Cited Brideau, C., Gunter, B., Pikounis, B., and Liaw, A. 2003. Improved statistical methods for hit selection in high-throughput screening. J. Biomol. Screen. 8:634-647. Burrows, L.L. and Lam, J.S. 1999. Effect of wzx (rfbX) mutations on A-band and B-band lipopolysaccharide biosynthesis in Pseudomonas aeruginosa O5. J. Bacteriol. 181:973980. Clatworthy, A.E., Pierson, E., and Hung, D.T. 2007. Targeting virulence: A new paradigm for antimicrobial therapy. Nat. Chem. Biol. 3:541-548. D’Elia, M.A., Pereira, M.P., Chung, Y.S., Zhao, W., Chau, A., Kenney, T.J., Sulavik, M.C., Black, T.A., and Brown, E.D. 2006. Lesions in teichoic acid biosynthesis in Staphylococcus aureus lead to a lethal gain of function in the otherwise dispensable pathway. J. Bacteriol. 188:41834189. Grundmann, H., Aires-de-Sousa, M., Boyce, J., and Tiemersma, E. 2006. Emergence and resurgence of methicillin-resistant Staphylococcus aureus as a public-health threat. Lancet 368:874885.
Marra, A. 2004. Can virulence factors be viable antibacterial targets? Expert Rev. Anti Infect. Ther. 2:61-72. Rudnicki, S. and Johnston, S. 2009. Overview of liquid handling instrumentation for highthroughput screening applications. Curr. Protoc. Chem. Biol. 1:43-54. Swoboda, J.G., Meredith, T.C., Campbell, J., Brown, S., Suzuki, T., Bollenbach, T., Malhowski, A.J., Kishony, R., Gilmore, M.S., and Walker, S. 2009. Discovery of a small molecule that blocks wall teichoic acid biosynthesis in Staphylococcus aureus. ACS Chem. Biol. 4:875-883. Swoboda, J.G., Campbell, J., Meredith, T.C., and Walker, S. 2010. Wall teichoic acid function, biosynthesis, and inhibition. ChemBioChem 11:35-45. Walsh, C. 2003. Where will new antibiotics come from? Nat. Rev. Microbiol. 1:65-70. Weidenmaier, C. and Peschel, A. 2008. Teichoic acids and related cell-wall glycopolymers in gram-positive physiology and host interactions. Nat. Rev. Microbiol. 6:276-287. Weidenmaier, C., Kokai-Kun, J., Kristian, S., Chanturiya, T., Kalbacher, H., Gross, M., Nicholson, G., Neumeister, B., Mond, J., and Peschel, A. 2004. Role of teichoic acids in Staphylococcus aureus nasal colonization, a major risk factor in nosocomial infections. Nat. Med. 10:243-245. Weidenmaier, C., Peschel, A., Xiong, Y.Q., Kristian, S.A., Dietz, K., Yeaman, M.R., and Bayer, A.S. 2005. Lack of wall teichoic acids in Staphylococcus aureus leads to reduced interactions with endothelial cells and to attenuated virulence in a rabbit model of endocarditis. J. Infect. Dis. 191:1771-1777. Xayarath, B. and Yother, J. 2007. Mutations blocking side chain assembly, polymerization, or transport of a wzy-dependent Streptococcus pneumoniae capsule are lethal in the absence of suppressor mutations and can affect polymer transfer to the cell wall. J. Bacteriol. 189:33693381.
Inglese, J., Johnson, R.L., Simeonov, A., Xia, M., Zheng, W., Austin, C.P., and Auld, D.S. 2007. High-throughput screening assays for the identification of chemical probes. Nat. Chem. Biol. 3:466-479.
Yuasa, R., Levinthal, M., and Nikaido, H. 1969. Biosynthesis of cell wall lipopolysaccharide in mutants of Salmonella. V. A mutant of Salmonella typhimurium defective in the synthesis of cytidine diphosphoabequose. J. Bacteriol. 100:433-444.
Katzen, F., Ferreiro, D.U., Oddo, C.G., Ielmini, M.V., Becker, A., Puhler, A., and Ielpi, L. 1998. Xanthomonas campestris pv. Campestris gum mutants: Effects on xanthan biosynthesis and plant virulence. J. Bacteriol. 180:1607-1617.
Zhang, J.H., Chung, T.D., and Oldenburg, K.R. 1999. A simple statistical parameter for use in evaluation and validation of high throughput screening assays. J. Biomol. Screen. 4:6773.
Malo, N., Hanley, J.A., Cerquozzi, S., Pelletier, J., and Nadon, R. 2006. Statistical practice in high-throughput screening data analysis. Nat. Biotechnol. 24:167-175.
Internet Resources http://ncgc.nih.gov/pub/openhts/curvefit/ Dose-response curve fitting.
HTS for Bacterial Growth Inhibitors
208 Volume 2
Current Protocols in Chemical Biology
High-Throughput Assessment of Mammalian Cell Viability by Determination of Adenosine Triphosphate Levels Nicola Tolliday1 1
Chemical Biology Platform, The Broad Institute of Harvard University and MIT, Cambridge, Massachusetts
ABSTRACT There are many assays available for high-throughput assessment of mammalian cell viability or cytotoxicity. Approaches include measurement of metabolic capacity, intracellular adenosine triphosphate (ATP) levels, induction of apoptosis, intracellular esterase and protease activity, and cellular membrane integrity. This unit provides an in-depth protocol for measurement of cellular ATP levels as a readout of mammalian cell viability, using the CellTiter-Glo assay from Promega Corporation. A comparison of the key parameters (sensitivity, speed, and cost) for this and other common high-throughput viability assays is C 2010 by John Wiley & Sons, Inc. also presented. Curr. Protoc. Chem. Biol. 2:153-161 Keywords: CellTiter-Glo r cell viability assay r cytotoxicity assay r high-throughput screening
INTRODUCTION Viable cultured cells are metabolically active and maintain high levels of adenosine triphosphate (ATP). Upon loss of membrane integrity, cells rapidly (within minutes) lose the ability to synthesize ATP, endogenous ATPases destroy any remaining ATP, and ATP levels drop significantly. Therefore, in general (but see Critical Parameters) the amount of ATP is directly proportional to the number of viable cells present in culture (Crouch et al., 1993). Described herein is a commercially available, homogeneous assay that monitors ATP levels as a readout for metabolically active, viable cells. Detection is based on the luciferase reaction in which mono-oxygenation of luciferin is catalyzed by luciferase in the presence of Mg2+ , ATP, and molecular oxygen (Fig. 1). The simple “add, mix, and read” nature of this assay, coupled with a robust and stable signal, makes it amenable to high-throughput uses (Barbie et al., 2009; Gupta et al., 2009). Protocols are provided to execute the assay in high-throughput (Basic Protocol) and for the plating of mammalian cells and treatment with test compounds (Support Protocol). NOTE: Perform all assays at least in duplicate.
beetle luciferin HO
S
N
N
S
+ATP+O2
oxyluciferin COOH
O
luciferase, Mg2+
S
N
N
S
O
+AMP+PPi+CO2 +LIGHT
Figure 1 The luciferase reaction. Mono-oxygenation of luciferin is catalyzed by luciferase in the presence of Mg2+ , molecular oxygen, and ATP, resulting in the production of light (adapted from the Promega CellTiter-Glo manual). Current Protocols in Chemical Biology 2: 153-161, July 2010 Published online July 2010 in Wiley Interscience (www.interscience.wiley.com). DOI: 10.1002/9780470559277.ch100045 C 2010 John Wiley & Sons, Inc. Copyright
High Throughput Mammalian Cell Viability Assays
153 Volume 2
BASIC PROTOCOL
CellTiter-Glo LUMINESCENT CELL VIABILITY ASSAY The CellTiter-Glo luminescent cell viability assay (Promega) is a widely used method to determine ATP levels as a surrogate readout for mammalian cell viability. Upon addition to cells, the CellTiter-Glo reagent does three things: 1. Lyses cell membranes to release ATP 2. Inhibits cellular ATPases to maintain ATP levels at or close to the levels at the time of reagent addition 3. Provides luciferin substrate, a stabilized “glow-type” luciferase and other reagents necessary to measure ATP using a bioluminescent reaction. The assay can be performed in 96-, 384-, and 1536-well microtiter plates, with the choice of format dependent upon both the end-user’s application and availability of a suitable luminometer to detect luminescence from multi-well plates. This protocol uses 384-well plates, but the assay can be scaled up or down for 96- or 1536-well plates, respectively. The assay is applicable to most cultured mammalian cells (for example, see Firestein et al., 2008; Wagner et al., 2008; Scholl et al., 2009); this protocol uses NIH/3T3 fibroblasts. The assay parameters (seeding density, time of exposure to test compounds/perturbagens, time of incubation with assay reagent etc.) will require optimization prior to execution in high-throughput mode (see Critical Parameters).
Materials CellTiter-Glo luminescent cell viability assay (Promega) containing: CellTiter-Glo buffer CellTiter-Glo substrate NIH/3T3 cells seeded at 2500 cells/well in white opaque 384-well plates and treated with test compounds/perturbagens (for cell plating and compound treatment see the Support Protocol) Vortex Multichannel pipet or automated bulk dispenser/pipettor for reagent delivery to 384-well plates (see Rudnicki and Johnston, 2009) Microtiter plate shaker/mixer Luminometer or CCD imaging device capable of reading 384-well microtiter plates (for large numbers of plates, integration with a stacker or a robotic arm is helpful) NOTE: Volumes given are for a 384-well plate. For a 96-well plate, scale up 4-fold. For a 1536-well plate, scale down 5-fold. 1. Thaw the CellTiter-Glo buffer, and equilibrate to room temperature. Use a room temperature water bath to thaw the buffer—do not expose to higher temperatures. Once thawed, the buffer is stable for 48 hr at room temperature.
2. Equilibrate the CellTiter-Glo substrate to room temperature. 3. Create the CellTiter-Glo reagent by adding the buffer to the lyophilized substrate. Transfer the contents of the buffer bottle directly into the amber bottle containing the substrate. Mix well by inverting the contents, gently vortexing, or swirling.
High Throughput Mammalian Cell Viability Assays
The CellTiter-Glo reagent should dissolve easily within 1 min. The reagent is stable for 48 hr at 4◦ C and for 6 months at −20◦ C. If freezing excess reagent for future use, prepare several working aliquots e.g., ten 10-ml aliquots from a 100-ml bottle, and thaw only the number of aliquots required.
4. Remove the 384-well microtiter plates containing the cells from the tissue culture incubator and equilibrate to room temperature for 30 min.
154 Volume 2
Current Protocols in Chemical Biology
Equilibrate the cells and the CellTiter-Glo reagent to room temperature prior to use to mitigate temperature effects on the luciferase reaction.
5. Add 30 μl/well of CellTiter-Glo reagent to each 384-well plate using an automated bulk dispenser or multichannel pipet. As discussed in Critical Parameters, it should be determined during assay development whether the reagent can be diluted—e.g., in phosphate-buffered saline (PBS)—and thus can be used for more assays. If this is feasible, the volume added should be the same as the culture volume in the wells (30 μl in this example). Consider any liquid handling “dead volume” requirements (see Rudnicki and Johnston, 2009) when calculating the total reagent required.
6. Lyse the cells by shaking the plates using a microtiter plate shaker/mixer for 2 min. Determine whether this step is necessary during assay development.
7. Incubate the plates for 10 min at room temperature. This allows stabilization of the luminescent signal. If processing many plates concurrently, establish the stability of the signal over time during assay development to determine processing batch size.
8. Read plates using a standard luminescence protocol. Protocol settings will vary with instrument. In general, an integration time of 0.1 to 1 sec per well is a good starting point. When first working with a particular luminometer it is recommended to determine the background signal of the instrument. Wells containing medium only (no cells) together with CellTiter-Glo reagent should be used to determine background luminescence, and this value should be subtracted from each of the experimental readings obtained from cells, for maximal accuracy. Additionally, the linear signal range of an instrument can be determined using a standard curve of known ATP concentrations (see the CellTiter-Glo manual, Promega). Only work within the linear range of the instrument. Typically, when testing compounds for effects on cell viability, a positive control (e.g., a cytotoxic compound) that represents 100% inhibition of growth and a neutral control (vehicle only, e.g., DMSO) that represents 0% inhibition of growth are used. The effect of an individual compound is then normalized and scaled to convert the raw luminescence values into percent inhibition. A standard process for normalization and scaling is given in Equation 1.
% Inhibition =
experimental value − mean neutral control value ×1000% mean positive control value − mean neutral control value Equation 1
PLATING OF NIH/3T3 CELLS AND ADDITION OF TEST COMPOUNDS BY PIN TRANSFER
SUPPORT PROTOCOL
The CellTiter-Glo viability assay can be used in many model assay systems. A representative example of one such system is given here, in which the effect of a panel of test compounds on the viability of NIH/3T3 cells is evaluated. This protocol describes the plating of cells into 384-well plates and the addition of compounds via pin transfer. As mentioned above, the assay parameters (seeding density, incubation time before adding test compounds, concentration of test compounds, and time of exposure to test compounds) will require optimization prior to execution in high-throughput mode (also see Critical Parameters).
Materials NIH/3T3 fibroblasts at ∼70% to 80% confluence, grown in standard tissue culture flasks or plates using Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with bovine calf serum to a final concentration of 10%
High Throughput Mammalian Cell Viability Assays
155 Current Protocols in Chemical Biology
Volume 2
Phosphate-buffered saline (PBS; tissue culture grade) 0.5 mg/ml trypsin, 0.2 mg/ml ethylenediaminetetraacetic acid (EDTA) in phosphate-buffered saline (PBS) DMEM supplemented with bovine calf serum to a final concentration of 10% Test compounds dissolved in DMSO (for example at 10 mM stock concentration) in 384-well polypropylene plates Hemacytometer Multichannel pipet or automated bulk dispenser/pipettor for reagent delivery to 384-well plates (see Rudnicki and Johnston, 2009) White opaque 384-well microtiter plates (e.g., Corning, cat. no. 8867BC or similar) 37◦ C, 5% CO2 incubator Pin tool for delivery of test compounds into assay plates: both automated and manual options are available [see V&P Scientific link under Internet Resources for more information on both options; also see Rudnicki and Johnston (2009)] Additional reagents and equipment for measuring cell viability (Basic Protocol) 1. Harvest cells with trypsin using standard aseptic cell culture techniques: a. When cells are 70% to 80% confluent, aspirate the culture medium, wash once with 10 ml PBS warmed to 37◦ C, and add trypsin. Use a volume of trypsin that is appropriate for the culture flask e.g., 0.5, 1, or 2.5 ml trypsin for a 25-cm2 , -75-cm2 , or 175-cm2 flasks, respectively. b. Incubate 5 min at room temperature. c. Add 10 ml serum-supplemented DMEM and centrifuge cells for 5 min at 1000 × g, room temperature (for a comprehensive approach to cell culture, see Freshney, 2005). Do not let NIH/3T3 cells reach 100% confluency. Washing with PBS prior to adding trypsin is critical as serum components present in the medium can inactivate trypsin. Note that the incubation conditions for trypsinization can vary by cell line (e.g., a shorter or longer incubation, either at room temperature or at 37◦ C).
2. Resuspend cells in serum-supplemented DMEM to a final concentration of 83,333 cells/ml. Use a hemacytometer or automated cell counter for accurate determination of cell number. For cells that tend to form clumps, additional pipetting to disrupt aggregates and/or use of a sterile 40-μm cell strainer to generate a single-cell suspension will help ensure accurate counting.
3. Add 30 μl cells to each well of an opaque 384-well plate using a multichannel pipet or automated bulk dispenser (final cell density is 2500/well). Consider any “dead volume” requirements of the pipettor or bulk dispenser (e.g., see Rudnicki and Johnston, 2009) when calculating the total volume of cells required. When filling large numbers of plates, swirl the flask of suspended cells regularly to prevent cell settling and uneven distribution into plates.
4. Incubate plates 16 to 24 hr (overnight) at 37◦ C, 5% CO2 . When handling large numbers of plates, consider using cassettes/racks to stack the plates, allowing uniform air movement and gas exchange around each plate. This step will mitigate differential temperature/gas exchange between plates throughout a stack.
5. Add test compounds from compound stock plates to assay plates by pin transfer. High Throughput Mammalian Cell Viability Assays
The exact volume added will depend upon the stock concentration of the compound and the final desired assay concentration. Follow recommended washing protocol for manual or automated pin transfer device between assay plates to prevent carry-over of compounds between assay plates. For single concentration screening, compounds are typically tested
156 Volume 2
Current Protocols in Chemical Biology
at a final assay concentration of ∼10 μM. Additional information on the effect of a compound can be obtained through evaluation of a range of concentrations and calculation of the half maximal effective concentration (EC50 ). For example, concentrations spanning the range of 20 μM to 160 nM can be achieved with an 8-point, two-fold dilution series and tested to determine EC50 values for each compound. Comparisons of EC50 values and the shape of each concentration-response curve can provide additional information for compound prioritization.
6. Incubate cells 24 hr at 37◦ C, 5% CO2 . See annotation to step 4 above regarding cassettes/racks for stacking assay plates.
7. Measure cell viability using the Basic Protocol.
COMMENTARY Background Information Assessment of cell viability or cytotoxicity is a routine procedure in many laboratories. In some situations this may be the primary assay of interest [e.g., when seeking to identify essential genes in a systematic, genome-wide fashion (Boutros et al., 2004)], whereas for assays that are sensitive to the number of viable cells it can serve as an important counterscreen to identify false positives [e.g., for screening using gene-expression assays (Hanh et al., 2008)]. There are many assays available to measure cell viability or cytotoxicity in a highthroughput manner. Selection of the most appropriate assay requires an understanding of what each assay measures as an endpoint, how that measurement correlates with cell viability, and the limitations of the particular assay chemistries. Additional considerations include ease of execution, assay sensitivity, assay robustness, total cost, requirements for specific detection instruments, and compatibility with automation. The CellTiter-Glo luminescent cell viability assay measures cellular ATP levels as a readout for viable cells. It is a sensitive assay that can detect as few as 15 viable cells in a 384-well format, and uses a simple “add, mix, and read” format that is amenable to automation. Additionally, the signal is robust, with a half-life of greater than 5 hr, depending upon cell type. Alternative approaches to assess cell viability or cytotoxicity include assays that measure metabolic capacity (e.g., resazurin or tetrazolium-based assays), induction of apoptosis, intracellular esterase and protease activity [e.g., calcein acetoxylmethyl ester- (Calcein-AM), glycyl-phenylalanylamino-fluorocoumarin (GF-AFC)-based assays], and membrane integrity [e.g., propidium iodide (PI) staining, lactate dehydrogenase (LDH) release]. A comparison of these ap-
proaches is provided in Table 1. Ultimately, the choice of assay will be intimately tied to the unique requirements of the biological system under investigation, with an emphasis on maintaining physiological relevance wherever possible.
Critical Parameters Assay development and optimization Successful high-throughput execution of the CellTiter-Glo luminescent cell viability assay requires advance optimization of a number of critical assay parameters. Specifically, thorough development of the assay includes optimization of the following: 1. Cell number per well to identify conditions within the linear range of the assay 2. The equilibration time between seeding cells into assay plates and adding test compound/perturbagen 3. Concentration of test compound/ perturbagen 4. The exposure time with test compound/ perturbagen 5. Concentration of CellTiter-Glo reagent (e.g., 1×, 0.5×, 0.25×, etc.) added to cells to identify the optimal balance between a robust stable signal and conservation of reagent 6. The incubation time with CellTiter-Glo reagent, and determination of whether mixing/shaking is required for efficient cell lysis 7. Stability of luminescence signal over time to determine batch size when handling a large number of plates. Optimization of each of these parameters is performed following the Basic Protocol, varying the appropriate parameters independently. Note that successful protocol optimization is an iterative process. Thus, multiple rounds of experiments, varying different critical parameters, are usually necessary to produce a robust assay. In all cases, the end goal is to identify conditions that produce a sufficient signal
High Throughput Mammalian Cell Viability Assays
157 Current Protocols in Chemical Biology
Volume 2
Table 1 Comparison of Commonly Used High-Throughput Cell Viability and Cytotoxicity Assays
Assay
Parameter measured
Sensitivitya
Incubation time
Costa
Referenceb
Comments
CellTiterGlo
ATP
High (15 cells)
Short (10 min)
Mid (9 cents/ Not suitable when assay Gupta et al. well) conditions affect cellular (2009) ATP levels.
CaspaseGlo
Effector (3/7) or initiator (8,9) caspase action
High (20 cells)
Moderate (30 min-2 hr)
High (45 cents/ well)
Caspase activity is transient and may only be detectable within a limited time window.
Geiger et al. (2006)
Alamar blue
Resazurin reduction
Mid (50 cells)
Moderate (1-4 hr)
Low (1 cent/well)
Substrate is incubated with cells for at least 1 hr prior to detection. In some cases, this may affect cell physiology and create assay artifacts.
Piantadosi and Suliman (2006)
MTSc
Tetrazolium salt reduction
Low (200 cells)
Moderate (1-4 hr)
Low (1 cent/well)
Wagner et al. (2008)
CalceinAMc
Intracellular esterase activity
Low (200 cells)
Moderate (4 hr)
Low (<1 cent/well)
Wagner et al. (2008)
LDHc activity
LDH release following cell lysis
Low (>200 cells)
Short (30 min)
Mid Measures nonviable Lightfield et al. (6 cents/well) cells. Sensitivity can be (2008) limited by LDH present in serum.
a Sensitivity and cost values are for 384-well format. b There are many references for each technology and a single example reference was selected for conciseness, with apologies to those authors whose
work has not been included. c Abbreviations: MTS, (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium); Calcein-AM, calcein acetoxyl-
methyl ester; LDH, lactate dehydrogenase.
window (typically at least 3-fold increase of signal over background), with minimal variation (coefficient of variation or %CV less than 10%, where %CV = standard deviation/mean × 100). Temperature As noted above, temperature can affect the rate of the luciferase reaction and thus change the intensity and rate of decay of the luminescent signal from the CellTiter-Glo assay. To ensure consistent results it is critical to equilibrate all reagents (assay plates, CellTiter-Glo reagent) to room temperature prior to use in the assay. When working with large numbers of plates, consider splitting plates into small stacks after removal from the tissue culture incubator to ensure uniform equilibration. High Throughput Mammalian Cell Viability Assays
Chemicals Solvents for test compounds and chemicals in culture media and sera may affect the light output in this assay. To test this, control wells
containing medium without cells and an appropriate concentration of compound solvent [usually dimethyl sulfoxide (DMSO)] can be assayed. The luminescence output from these controls provides the background signal for the assay. DMSO has been tested at final concentrations of up to 2% with minimal effects on light output. A 5% increase in light output is observed when using media lacking phenol red compared to media with phenol red (see CellTiter-Glo technical manual, Promega). Cellular ATP content Factors that affect the cellular levels of ATP may affect the relationship between cell number and luminescence in this assay. For example, depletion of oxygen has been demonstrated to decrease cellular ATP levels (Crouch et al., 1993). Additionally, anchoragedependent cells that undergo contact inhibition may show a change in ATP levels at high densities. For these and other situations where ATP levels are not correlated with the number of
158 Volume 2
Current Protocols in Chemical Biology
viable cells, use an alternative assay to measure cell viability (see Table 1). Plate selection White opaque multi-well plates for luminescence are recommended for use in this assay. Opaque-walled, clear-bottom plates that enable visual inspection of the cells under a microscope can also be used. However, it is recommended that an opaque sticker/tape be applied to the bottom surface of the plate prior to reading luminescence, as this will reduce cross-talk between the wells and prevent loss of signal that would otherwise be observed with a clear bottom plate.
Troubleshooting The majority of problems encountered with this assay derive from the cells and biology, and not from the assay chemistry. Table 2 highlights some of the more common issues, descriptions of possible causes, and recommended solutions.
Anticipated Results Figure 2 shows typical results for assessment of viability of mammalian cells following treatment with the nonspecific protein kinase inhibitor Staurosporine, which is cytotoxic. As the concentration of inhibitor
increases from 0.2 nM to 0.3 μM, the signal from the CellTiter-Glo assay decreases, reflecting a decrease in cell viability. The EC50 is ∼14 nM, as determined using the curve fitting function within GraphPad Prism. Note that the absolute luminescence values will vary with both cell line and detection device.
Time Considerations Execution of the CellTiter-Glo viability assay on a batch of ten 384-well plates can be accomplished easily using an automated bulk dispenser within 1 hr from start (removal of assay plates from the incubator, thawing, and equilibration of assay reagents) to finish (reading luminescence on a plate reader). If many plates are to be processed, the time required will scale accordingly. An additional consideration is the time required to prepare cells for the assay. This includes plating cells, equilibration in plates, and addition of and incubation with test compound/perturbagen, and this can vary greatly depending upon biological context. In the example protocol provided here, the total duration of the assay, including all incubation periods, is 3 days.
Acknowledgements I would like to thank Patrick Faloon, Leigh Carmody, and Bridget Wagner for sharing their
Table 2 Potential Problems Encountered with the CellTiter-Glo Assay
Problem
Possible cause
Solution
Signal is not linear with cell number
Signal falls outside dynamic range of assay or plate reader
Evaluate additional cell seeding densities to determine optimal conditions
ATP content is decoupled from cell Use alternative assay number in biological setting of interest High degree of signal variation: With low signal
With high signal
Too few cells
Increase cell number
Incomplete cell lysis
Increase duration and/or speed of shaking during incubation with CellTiter-Glo reagent
Edge/position effects
Optimize growth conditions to minimize differences in gas-exchange, temperature gradients, liquid evaporation etc. within/between plates (Lundholt et al., 2003; see also A. Tanner, Internet Resources)
Non-uniform cell dispensing e.g., of “clumpy” cells
Generate a single-cell suspension by passing cells through a sterile 40-μm filter after cell harvesting to ensure a uniform suspension of cells
“Cross-talk” between high signal and low signal wells
Confirm that an opaque-bottom assay plate is in use. Reduce light going to the detector e.g., use a 1536 aperture with 384-well plates.
159 Current Protocols in Chemical Biology
Volume 2
2,000,000 1,800,000 1,600,000 1,400,000 1,200,000 RLU 1,000,000 800,000 600,000 400,000 200,000 0.0001
0.001
0.01 Concentration ( M)
0.1
1
Figure 2 Representative data from the CellTiterGlo viability assay. Mammalian cells were incubated in 384-well plates for 3 days with Staurosporine ranging from 0.2 nM to 0.3 μM (16 replicates for each concentration). The effects on viability were determined using the CellTiterGlo assay, luminescence was read on an Envision Multilabel plate reader (PerkinElmer) and the data were analyzed using the curve fitting function within GraphPad Prism.
experiences with the CellTiter-Glo assay, and Bridget Wagner and Josh Syken for critical review of the manuscript.
Geiger, G.A., Parker, S.E., Beothy, A.P., Tucker, J.A., Mullins, M.C., and Kao, G.D. 2006. Zebrafish as a “biosensor”? Effects of ionizing radiation and amifostine on embryonic viability and development. Cancer Res. 66:8172-8181.
Literature Cited
Gupta, P.B., Onder, T.T., Jiang, G., Tao, K., Kuperwasser, C., Weinberg, R.A., and Lander, E.S. 2009. Identification of selective inhibitors of cancer stem cells by high-throughput screening. Cell 138:645-659.
Barbie, D.A., Tamayo, P., Boehm, J.S., Kim, S.Y., Moody, S.E., Dunn, I.F., Schinzel, A.C., Sandy, P., Meylan, E., Scholl, C., Fr¨ohling, S., Chan, E.M., Sos, M.L., Michel, K., Mermel, C., Silver, S.J., Weir, B.A., Reiling, J.H., Sheng, Q., Gupta, P.B., Wadlow, R.C., Le, H., Hoersch, S., Wittner, B.S., Ramaswamy, S., Livingston, D.M., Sabatini, D.M., Meyerson, M., Thomas, R.K., Lander, E.S., Mesirov, J.P., Root, D.E., Gilliland, D.G., Jacks, T., and Hahn, W.C. 2009. Systematic RNA interference reveals that oncogenic KRAS-driven cancers require TBK1. Nature 462:108-112. Boutros, M., Kiger, A.A., Armknecht, S., Kerr, K., Hild, M., Koch, B., Haas, S.A., Paro, R., Perrimon, N.; and Heidelberg Fly Array Consortium. 2004. Genome-wide RNAi analysis of growth and viability in Drosophila cells. Science 303:832-835. Crouch, S.P., Kozlowski, R., Slater, K.J., and Fletcher, J. 1993. The use of ATP bioluminescence as a measure of cell proliferation and cytotoxicity. J. Immunol. Methods 160:8188. Firestein, R., Bass, A.J., Kim, S.Y., Dunn, I.F., Silver, S.J., Guney, I., Freed, E., Ligon, A.H., Vena, N., Ogino, S., Chheda, M.G., Tamayo, P., Finn, S., Shrestha, Y., Boehm, J.S., Jain, S., Bojarksi, E., Mermel, C., Barretina, J., Chan, J.A., Baselga, J., Tabernero, J., Root, D.E., Fuchs, C.S., Loda, M., Shivdasani, R.A., Meyerson, M., and Hahn, W.C. 2008. CDK8 is a colorectal cancer oncogene that regulates betacatenin activity. Nature 455:547-551. High Throughput Mammalian Cell Viability Assays
Freshney, I.R. 2005. Culture of Animal Cells: A Manual of Basic Technique, 5th ed. Wiley-Liss, Hoboken, N.J.
Hahn, C.K., Ross, K.N., Warrington, I.M., Mazitschek, R., Kanegai, C.M., Wright, R.D., Kung, A.L., Golub, T.R., and Stegmaier, K. 2008. Expression-based screening identifies the combination of histone deacetylase inhibitors and retinoids for neuroblastoma differentiation. Proc. Natl. Acad. Sci. U.S.A. 105:9751-9756. Lightfield, K.L., Persson, J., Brubaker, S.W., Witte, C.E., von Moltke, J., Dunipace, E.A., Henry, T., Sun, Y., CadoD., Dietrich, W.F., Monack, D.M., Tsolis, R.M., and Vance, R.E. 2008. Critical function for Naip5 in inflammasome activation by a conserved carboxy-terminal domain of flagellin. Nat. Immunol. 9:1171-1178. Lundholt, B.K., Scudder, K.M., and Pagliaro, L. 2003. A simple technique for reducing edge effect in cell-based assays. J. Biomol. Screen. 8:566-570. Piantadosi, C.A. and Suliman, H.B. 2006. Mitochondrial transcription factor A induction by redox activation of nuclear respiratory factor 1. J. Biol. Chem. 281:324-333. Rudnicki, S. and Johnston, S. 2009. Overview of liquid handling instrumentation for highthroughput screening applications. Curr. Protoc. Chem. Biol. 1:43-54. Scholl, C., Frohling, S., Dunn, I.F., Schinzel, A.C., Barbie, D.A., Kim, S.Y., Silver, S.J., Tamayo, P., Wadlow, R.C., Ramaswamy, S., D¨ohner, K., Bullinger, L., Sandy, P., Boehm, J.S., Root, D.E., Jacks, T., Hahn, W.C., and Gilliland, D.G. 2009. Synthetic lethal interaction between oncogenic
160 Volume 2
Current Protocols in Chemical Biology
KRAS dependency and STK33 suppression in human cancer cells. Cell 137:821-834. Wagner, B.K., Kitami, T., Gilbert, T.A., Peck, D., Ramanathan, A., Schreiber, S.L., Golub, T., and Mootha, V. 2008. Large-scale chemical dissection of mitochondrial function. Nat. Biotechnol. 26:343-51.
Internet Resources http://www.promega.com/paguide/chap4.htm A protocols and application guide from Promega that provides an overview to cell viability and cytotoxicity assays. http://www.promega.com/catalog/catalogproducts. aspx?categoryname=productleaf 1505 The CellTiter-Glo technical manual from Promega. http://www.vp-scientific.com/pin tools.htm Information on manual and automated pin tools for delivery of compounds into assay plates. http://www.labautopedia.org/mw/index.php/ User:S.d.hamilton/Helpful Hints to Manage Edge Effect of Cultured Cells for High Throughput Screening Helpful Hints to Manage Edge Effect of Cultured Cells for High Throughput Screening: A Corning HTS/Assay Systems Application Note that addresses common causes of edge effects observed with cellbased assays, authored by Allison Tanner.
High Throughput Mammalian Cell Viability Assays
161 Current Protocols in Chemical Biology
Volume 2
Mass Spectrometry-Based Identification of Protein Kinase Substrates Utilizing Engineered Kinases and Thiophosphate Labeling Yong Chi1 and Bruce E. Clurman1 1
Divisions of Clinical Research and Human Biology, Fred Hutchinson Cancer Research Center, Seattle, Washington
ABSTRACT Protein kinases constitute a large enzyme family with key roles in cellular signal transduction. One way to elucidate the functions of protein kinases is to systematically identify their downstream targets. Presented here is a simple and effective method to identify direct protein kinase substrates in native cell lysates. First, the activity of the kinase of interest is isolated by engineering the normal kinase to utilize bulky ATP analogs that cannot be used by normal cellular kinases. This allows specific labeling of substrates with thiophosphate tags by performing kinase reactions in cell lysates that also include bulky ATP-γ-S analogs. After digesting the proteins in the reaction mixture, thiophosphopeptides are isolated using a single-step capture-and-release protocol and identified by mass spectrometry. This technique is easy to use and generally applicable. Curr. Protoc. Chem. C 2010 by John Wiley & Sons, Inc. Biol. 2:219-234 Keywords: chemical genetics r analog-sensitive kinase r thiophosphorylation r phosphopeptides
INTRODUCTION Identifying the physiological targets of protein kinases is critical to understanding their functions, and many approaches have been developed to systematically map kinasesubstrate relationships. Notably, the chemical genetics approach developed by the Kevan Shokat laboratory (Shah et al., 1997; Liu et al., 1998) is an attractive way to isolate the activities of any kinase of interest from the myriad of cellular kinases. This kinase engineering scheme modifies the conserved ATP-binding pocket of a kinase, enabling it to utilize unnatural ATP analogs (Shah et al., 1997; Liu et al., 1998). A key feature of this strategy is that only the engineered (or analog-sensitive) kinase, but not cellular kinases, can use the ATP analog (Fig. 1). These analog-sensitive kinases can then be used to specifically label their substrates in kinase reactions using cell lysates (Shah and Shokat, 2003). A variety of kinase substrate identification methods using this strategy have been described (Koch and Hauf, 2010), including several recent reports (Blethrow et al., 2008; Chi et al., 2008; Holt et al., 2009). Described here is an in vitro approach to identify the direct targets of a protein kinase in native cell lysates. This article provides updated protocols based on our recent study to identify human CDK2 substrates (Chi et al., 2008). The Basic Protocol includes the two key steps of the method: (1) the specific labeling of the substrates in cell lysates, and (2) and the subsequent identification of the labeled proteins. For the substrate labeling, a kinase reaction is carried out in the presence of analog-sensitive kinase and an ATP-γ-S analog (Fig. 2A). Thiophosphate labeling allows stable incorporation of thiophosphate into substrate proteins and provides an easy chemical tag for isolating the
Current Protocols in Chemical Biology 2: 219-234, November 2010 Published online November 2010 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/9780470559277.ch100151 C 2010 John Wiley & Sons, Inc. Copyright
Category: Protein Modifications, Engineering and Applications
219 Volume 2
wild-type kinase
ATP
analog-sensitive kinase
analog
analog
ATP
Figure 1 Kinase engineering strategy. Mutation of a conserved gatekeeper residue in the ATPbinding domain of a kinase allows the binding of ATP analogs. Although most analog-sensitive kinases can still use ATP, only they, but not the wild-type kinases, can use the ATP analogs.
A
analog-sensitive kinase ATP- -S analog
B
trypsin
disulfide beads
NaOH
digestion
binding
elution
thiophosphate-labeled and other proteins
All peptides
Cysteine-containing and thiophospho peptides
LC-MS/MS
candidate identification substrates phosphopeptides
Figure 2 General scheme of the substrate identification method. (A) Labeling of candidate substrates via kinase reactions using cell lysate, analog-sensitive kinase, and ATP-γ-S analog. (B) Identification of the labeled substrates. Proteins in the kinase reaction are digested with trypsin. The resulting peptides are incubated with disulfide beads, which capture both thiophosphopeptides and cysteine-containing peptides. The beads are then treated with basic solution to selectively release only the thiophosphopeptides, which are converted to normal phosphopeptides via hydrolysis. The peptide sample is subjected to liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS) for identification of the candidate substrates.
Engineered Kinases and Thiophosphate Labeling
thiophosphopeptides following digestion of the protein mixture. This protocol is similar to standard kinase assays that use unmodified kinase and ATP, but specifically labels substrates with thiophosphate. In the second step (Fig. 2B), the protein mixture in the reaction is digested with trypsin, and the resulting peptide mixture is incubated with a disulfide resin that binds both cysteine-containing peptides and thiophosphopeptides. Washing and subsequent elution of the disulfide beads with sodium hydroxide selectively releases the thiophosphopeptides (as normal phosphopeptides via hydrolysis) while leaving the cysteine-containing peptides behind. Finally, the eluted peptides are identified by
220 Volume 2
Current Protocols in Chemical Biology
mass spectrometry (MS) and database searching. This technique is broadly applicable, since it is based on a simple capture-and-release chemical enrichment procedure and only requires widely available mass spectrometry instrumentation and basic computational tools. For users’ comparison, an alternative method has also been described in this series (Hertz et al., 2010).
STRATEGIC PLANNING Design, Characterization, and Production of Analog-Sensitive Kinases The first step toward applying the method is to design analog-sensitive alleles of the kinase of interest. Since the ATP binding sites in protein kinases are conserved, analog-sensitive alleles of a kinase can usually be constructed by replacing a bulky or hydrophobic residue (called a “gatekeeper”) in the ATP-binding domain with a glycine or alanine. Descriptions of the procedures and tools have been published previously (Buzko and Shokat, 2002; Blethrow et al., 2004). Since ATP-γ-S analogs are a critical feature of this method, the authors advise that the wild-type kinase of interest be tested to see if it can utilize ATP-γ-S efficiently before designing analog-sensitive alleles. Once candidate alleles of the analogsensitive kinases have been constructed, they need to be characterized so that at least one allele meets the expected criteria. To do this, it is first necessary to express and purify the analog-sensitive kinase and carry out kinase assays using ATP-γ-S analogs. These test kinase assays are generally done in parallel with wild-type kinase using either a known substrate (Chi et al., 2008) or cell extracts (Blethrow et al., 2008). These assays must demonstrate that the analog-sensitive kinase, but not the wild-type version, can use the ATP-γ-S analog efficiently. Detection of thiophosphorylation by the analog-sensitive kinase in the presence of ATP-γ-S analog can be achieved by monitoring the electrophoretic mobility shift of a known substrate, by immunoblot of the protein using a phosphositespecific antibody (Chi et al., 2008), or by autoradiography using radiolabeled ATP-γ-S analog (Blethrow et al., 2008). Once the analog-sensitive kinase has shown both activity and specificity toward the ATP-γ-S analog, it is ready to be produced in sufficient quantity to be used in the lysate experiments. Methods for expression and purification of active analog-sensitive kinases are up to the user; the kinases can be produced in bacteria or human cells (Chi et al., 2008) or baculovirus-infected insect cells (Blethrow et al., 2008). Preparation of ATP-γ-S Analog and Cell Lysates During the characterization of the analog-sensitive kinase of interest, an appropriate ATP analog should be chosen. Two commonly used ATP analogs are the N6 -substituted ATP analogs: N6 -(Benzyl)-ATP and N6 -(2-Phenylethyl)-ATP. Both the ATP and ATP-γ-S form of these ATP analogs can be chemically synthesized (Allen et al., 2007; Hertz et al., 2010), purchased (e.g., BIOLOG Life Science Institute), or custom synthesized (e.g., N6 -(2-Phenylethyl)-ATP-γ-S from TriLink BioTechnologies). Kinase assays can be performed using whole-cell lysates. However, the enormous complexity and dynamic range of the components within whole-cell lysates limits the extent of the substrate labeling and protein identifications. It is therefore desirable to fractionate the whole-cell lysates to reduce the sample complexity, as well as to enrich the candidate substrates. Fractionation thus leads to more efficient and robust substrate labeling, as well as more protein identifications in the mass spectrometer. The protein-fractionation techniques and the extent of fractionation are flexible and depend on one’s preference and instrument availability. Typically, the procedure involves chromatography techniques, such as ion exchange (Chi et al., 2008) or precipitation techniques (Blethrow et al., 2008). As an example, a simple fractionation protocol for HEK293 cell lysate using ion-exchange chromatography is provided in Support Protocol 1. Detailed descriptions on protein fractionation techniques can be found elsewhere in the literature, such as a protocol for ion exchange chromatography (e.g., Williams and Frasca, 2001).
Category: Protein Modifications, Engineering and Applications
221 Current Protocols in Chemical Biology
Volume 2
Positive and Negative Controls A positive control is desirable for monitoring the success of the protocols. Typically, a known and well-phosphorylated substrate for the kinase of interest is chosen. It is recommended that a known substrate, prior to cell lysates, be used in the kinase reaction and subsequent thiophosphopeptide purification scheme to see if these protocols can be followed successfully. An example using the GST-Rb (Glutathione S-transferase fused to the carboxyl-terminal 156 amino acids of human retinoblastoma protein) as a standard substrate for CDK2 is provided in Support Protocol 2. Once the method can be successfully carried out using the known substrate, this substrate can then be spiked into the kinase reactions containing cell lysates as a positive control (see Basic Protocol). Although the thiophosphate labeling is expected to be highly specific to the analogsensitive kinase, background labeling can still occur due to the activities of cellular kinases that can use the ATP-γ-S analogs to a limited extent (Chi et al., 2008). To control for nonspecific labeling, a negative control reaction without the analog-sensitive kinase should be carried out in parallel. The thiophosphopeptide identifications from this reaction can then be subtracted from the list of identifications from the reaction with the analog-sensitive kinase present. BASIC PROTOCOL
SUBSTRATE LABELING AND PURIFICATION OF THE THIOPHOSPHOPEPTIDES Described here are the substrate identification procedures summarized in Figure 2. Once the necessary reagents (described above) are obtained, kinase reactions can then be carried out using the engineered kinase, ATP-γ-S analog, and cell lysates. Following tryptic digestion of the reaction mixture, the thiophosphopeptides are captured by a disulfide resin and then specifically eluted via base hydrolysis. Tandem mass spectrometry is used to identify the phosphopeptides and the substrate proteins.
Materials 5× kinase reaction buffer (see recipe) Cell lysate (buffer composition similar to kinase assay condition; see Support Protocol 1 and Strategic Planning) Purified control substrate (protein known to be a substrate of the kinase of interest) N6 -(2-Phenylethyl)-ATP-γ-S (TriLink BioTechnologies) Kinase of interest 0.5 M EDTA, pH 8 (MediaTech) Acetonitrile (VWR) Trypsin, sequencing grade (Promega) 10% (v/v) formic acid (store in glass bottle at room temperature indefinitely) ColorpHast pH-indicator strips, pH 4.0 to 7.0 (Fisher Scientific) Disulfide beads: thiopropyl-Sepharose 6B (GE Healthcare) Micro Bio-Spin chromatography columns, 0.8 ml (Bio-Rad Laboratories) Washing solution 1 (see recipe) Washing solution 2 (see recipe) 20 mM sodium hydroxide (store at room temperature up to 3 months) 1% (v/v) formic acid (store in glass bottle at room temperature indefinitely)
Engineered Kinases and Thiophosphate Labeling
0.6- and 1.7-ml microcentrifuge tubes Microcentrifuge, and mini-microcentrifuge (e.g., VWR) for smaller tubes 1-ml syringes 26-G, 1/2-in. needles Labquake rotator (Fisher Scientific)
222 Volume 2
Current Protocols in Chemical Biology
Iron stand and clamps (Fisher Scientific) Dropper bulbs for 3-ml Pasteur pipets (Fisher Scientific) Tandem mass spectrometer (ThermoFisher Scientific) Database search and data filtering software Statistical software (see Deutsch, 2010) Label kinase substrates in a kinase reaction 1. Mix the following in the order indicated in a 1.7-ml microcentrifuge tube for a 200-μl reaction: 40 μl 5× kinase reaction buffer: 40 μl Deionized H2 O needed to make up 200-μl final reaction volume. 200 to 300 μg cell lysate 100 to 200 ng control (known) substrate 5 μl 10 mM N6 -(2-Phenylethyl)-ATP-γ-S (250 μM final) Analog-sensitive kinase: up to 2 to 3 μg purified kinase protein in solution or on beads. If a larger reaction volume is needed, simply scale up all reagents proportionally. For reactions using a control substrate only, omit the cell lysate and reduce the reaction volume as needed (see Support Protocol 2). The analog-sensitive kinase can be added in the form of purified recombinant protein, eluted or bead-bound immunoprecipitate, or even a small amount of crude cell lysate in which the analog-sensitive kinase is highly overexpressed. Ideally, the amount of kinase added should approximate the physiological amount of kinase activity in the cell lysate, but supraphysiologic kinase activity may be needed to label sufficient amounts of substrates to allow their identification. However, excessive amounts of kinase may label low-affinity targets and therefore yield more falsepositive identifications. Phosphatase inhibitors are not necessary for the kinase reaction because the thiophosphates are resistant to phosphatases. Avoid using reducing agents (such as dithiothreitol, DTT) in the reaction, as they may interfere with disulfide bead binding later. Low concentrations of nonionic detergents can be included if necessary.
2. Incubate the kinase reaction at an appropriate temperature in a water bath for 1 hr. Reaction temperature is typically 25◦ C for yeast samples and 30◦ C for mammalian samples. Reaction time can range from 30 min to several hours, but shorter incubation times are preferred when possible to minimize nonspecific labeling. Since kinases utilize ATP-γ -S at a much slower rate than ATP, it might be necessary to incubate reactions for longer time periods to achieve sufficient labeling of proteins.
3. Stop the reaction by adding 0.5 M EDTA (pH 8) to a final concentration of 20 mM. 4. Proceed to the next step or store the reaction overnight at −20◦ C.
Digest the protein mixture 5. Add acetonitrile to the reaction to a final concentration of 10% 15% (v/v). Trypsin is resistant to mild denaturing conditions such as 10% of acetonitrile (Bond, 1989). Acetonitrile helps denature the lysate proteins and facilitates more complete proteolysis by trypsin. It also does not expand the sample volume appreciably. Other commonly used denaturants such as urea would require several-fold dilution of the sample volume before trypsin digestion. Binding to disulfide beads is a chemical reaction that is sensitive to volume expansion, and urea might also interfere with the disulfide-bead binding. If urea is used as denaturant, samples should be desalted and concentrated after trypsin digestion. Avoid using reducing agents (such as DTT) at this step, as they may interfere with the disulfide bead binding in the subsequent step.
6. Add trypsin to a final protease:protein ratio of 1:50 to 1:20 (w/w).
Category: Protein Modifications, Engineering and Applications
223 Current Protocols in Chemical Biology
Volume 2
7. Incubate the tube at 37◦ C for at least 3 hr. It is acceptable to allow the reaction to proceed overnight if it is near the end of the day.
8. Proceed to the next step or store the sample overnight at −20◦ C. It is a common practice to treat the peptide samples with iodoacetamide to block the cysteines following the trypsin digest. This procedure should not be performed here, as iodoacetamide reacts readily with thiophosphates.
Bind to disulfide beads 9. Microcentrifuge the sample 1 min at 10,000 × g, room temperature. This removes any precipitates that may be produced during digestion. Typically, there may be minor precipitations if large amounts of protein or crude lysates are used.
10. Transfer the supernatant to a new microcentrifuge tube. 11. Adjust the pH of the solution to a pH of ∼5 using 10% formic acid. Add 1 to 2 μl at a time using a pipet and check the pH of the sample by spotting 1 μl onto a pH strip (pH 4.0 to 7.0). The working pH for robust binding is between 4 and 7.5, and lower pH may slightly favor thiophosphate binding. Prepare the 10% formic acid solution using 99% formic acid stock and deionized water. Handle the concentrated formic acid in a fume hood. Use glass pipets or Hamilton syringes to transfer formic acid and store the 10% formic acid stock in a glass bottle.
12. Prepare a 50% (v/v) slurry of disulfide beads (thiopropyl-Sepharose 6B) according to the manufacturer’s instructions. Beads may be prepared in advance by soaking and washing the dry beads in deionized water and storing the swollen beads at 4◦ C in 20% ethanol (long-term) or deionized water (up to 2 weeks).
13. Mix disulfide beads stock and transfer 40 μl into a 0.6-ml microcentrifuge tube using a pipet tip. Cut the end of the pipet tip if necessary to allow smooth transfer of the beads. The amount of beads used depends on the amount of sulfhydryl (–SH) content in the reaction mixture and should have sufficient capacity to theoretically capture all sulfhydryl-containing molecules. Using excess beads may encourage nonspecific binding of molecules to the disulfide beads.
14. Centrifuge the tube in a mini-microcentrifuge at maximum speed for a few seconds to settle the beads. Remove the excess liquid using a 1-ml disposable syringe attached to a 26-G, 1/2-in. needle. 15. Add the peptide sample to the beads. Mix by inverting the tube several times. The bead suspension should flow smoothly upon inverting. If not, add another 10% to 20% acetonitrile to help reduce the viscosity.
16. Incubate the tube on a Labquake rotator with continuous rotation at room temperature overnight. The minimal incubation time depends on the sample complexity and volume. A few hours might be sufficient if timing is important.
Wash and elute the disulfide beads 17. Fix a 0.8-ml Micro Bio-Spin chromatography column onto an iron stand. Engineered Kinases and Thiophosphate Labeling
18. Centrifuge the tube containing the bead sample for 10 sec in a mini-microcentrifuge at maximum speed. Remove most of the supernatant containing unbound peptides.
224 Volume 2
Current Protocols in Chemical Biology
19. Resuspend the beads with 200 μl of water and load the beads mixture onto the column using a blunt-ended 200-μl pipet tip. 20. Use a rubber dropper bulb to apply air pressure to from the top of the column to elute the liquid phase. 21. Wash the beads sequentially twice with 0.5 ml of water, then three times with 0.5 ml of washing solution 1, twice again with 0.5 ml of water, three times with 0.5 ml of washing solution 2, and finally 2 × 0.5 ml water. At each washing step, mix the beads by pipetting up and down a few times using a blunt-ended 200-μl pipet tip and elute the liquid phase using the rubber bulb. 22. Collect all beads by resuspending them in 100 μl of water and pipetting them out of the column into a 1.7-ml microcentrifuge tube. Do this several times to ensure complete transfer of the beads. 23. Let the beads settle by standing for a few minutes or spin them down by a brief centrifugation at maximum speed in a mini-microcentrifuge. Carefully remove most of the liquid phase with a pipet. Remove the last bit of liquid using a 1-ml syringe. 24. Add 20 to 30 μl of 20 mM NaOH. Mix and incubate on a Labquake rotator at room temperature for 30 min. The beads should be in suspension and viscosity should keep the solution from flowing around the tube. The time of incubation is typically 30 min to 1 hr. Although the elutionagent volume is flexible, a minimum of one bead bed volume is typically used. Use low elution volume if more concentrated peptide samples are desired.
25. Spin down the beads by a brief centrifugation at maximum speed in a minimicrocentrifuge. Carefully collect the supernatant with a pipet tip without disturbing the beads. 26. Optional: Rinse the beads by mixing it with 5 μl of water. Spin down the beads and transfer the supernatant to the eluate above. 27. Acidify the eluate by adding 4 to 5 μl of 1% formic acid. The pH should be at 3 or lower so that the sample is compatible with MS analysis. A desalting step could be included here to further concentrate or purify the peptide sample before the MS analysis, but it is not necessary.
28. Proceed to the next step or store the sample up to 1 week at −20◦ C.
Analyze the peptide samples by mass spectrometry 29. Microcentrifuge the sample 1 min at 10,000 × g, room temperature. This step removes any particulate material in the sample that may interfere with the HPLC system or MS instrument.
30. Transfer the necessary volume of the supernatant to a new vial for loading. The loading volume depends on user preference and the maximum loading volume allowed in the MS instrument setup.
31. Analyze the peptide sample using a tandem mass spectrometer. The choice of MS instrument depends on the user and equipment availability. Typically, a mass spectrometer that can perform tandem mass spectrometry (MS/MS) is required. An ion-trap instrument, such the Thermo Scientific LCQ, LTQ, or LTQ Orbitrap, is suitable. Advanced instrumentation is preferred if available. As an example, the operation on the Thermo Finnigan LCQ DECA XP has been described (Yi et al., 2003).
Category: Protein Modifications, Engineering and Applications
225 Current Protocols in Chemical Biology
Volume 2
32. Search the MS/MS spectra against the appropriate protein databases using a suitable search algorithm. Differential mass modification of the serine, threonine, or tyrosine residues by the phosphate group should be specified in the search parameters. Various protein sequence databases, such as the latest version of International Protein Index (IPI), and search tools, such as SEQUEST and Mascot, can be used. An overview of the current database search algorithms has been described (Kapp and Schutz, 2007).
33. Perform statistical analysis on the search results. Probability-based statistical analysis software should be used to help filter the search results, especially on large datasets. Typically, peptides with high probability score (0.9 or higher) and low false discovery rate (0.1% or lower) are reported. One of these types of software tools has been described (Deutsch et al., 2010). If possible, all MS/MS spectra for the identified phosphopeptides should also be validated by manual inspection. SUPPORT PROTOCOL 1
FRACTIONATION OF HEK293 CELL LYSATE Because many kinase targets are low-abundance proteins, it is necessary to enrich these proteins to obtain sufficient material for identification. One way to achieve this is to fractionate the whole-cell lysate to reduce the complexity of the cell lysate proteins and enrich the candidate substrates. The following protocol describes a method for HEK293 lysate preparation and subsequent fractionation using low-pressure ion exchange chromatography and ammonium sulfate precipitation.
Materials HEK293 cells (ATCC, cat. no. CRL-11268) Phosphate-buffered saline (PBS; MediaTech, cat. no. 21-040-CV) Hypotonic lysis buffer (see recipe) 5 M NaCl BioRad protein assay dye reagent (BioRad Laboratories) SP Sepharose Fast Flow resin (GE Healthcare) Q Sepharose Fast Flow resin (GE Healthcare) Column loading buffer (see recipe) 30 mM Tris·Cl, pH 7.5 Column loading buffer (see recipe) containing 100 mM, 200 mM, 300 mM, 400 mM, and 600 mM NaCl Ammonium sulfate (Fisher Scientific) Dialysis buffer (see recipe) 15-cm cell culture dishes (Becton Dickinson) 15-ml and 50-ml conical centrifuge tubes (e.g., BD Falcon, Fisher Scientific) Tabletop centrifuge with swinging-bucket rotor Branson Sonifier 250 (Branson Ultrasonics) 1.0 × 10 cm, glass chromatography columns (Bio-Rad Laboratories) Iron stands and clamps for chromatography columns Accumet AB30 conductivity meter (Fisher Scientific)Refrigerated centrifuge Amicon Ultra-4 Centrifugal Filter, 10,000 MWCO (Millipore) SnakeSkin pleated dialysis tubing, 7000 MWCO (Pierce) Additional reagents and equipment for basic cell culture techniques including trypsinization (Phelan, 2007)
Engineered Kinases and Thiophosphate Labeling
226 Volume 2
Prepare HEK293 whole cell lysate 1. Grow HEK293 cells on four 15-cm plates to 80% to 90% confluence (also see Phelan, 2007). Typically, four to eight 15-cm plates yield sufficient proteins for a small-scale fractionation. The starting material and the chromatography column sizes can be scaled proportionally. Current Protocols in Chemical Biology
2. Wash cells with 10 ml PBS. Trypsinize cells (Phelan, 2007) and harvest them into a 15-ml conical tube. Centrifuge in a tabletop centrifuge with a swinging-bucket rotor 5 min at 250 × g, room temperature, and remove the supernatant by aspiration. Wash cells once with 10 ml PBS, centrifuge again as before, and remove supernatant by aspiration. Process the cell pellet immediately or quick freeze it in liquid nitrogen and store up to 1 month at −80◦ C.
3. Resuspend the cell pellet in 4 ml hypotonic lysis buffer. Incubate tube 30 min at 4◦ C. 4. Add 5 M NaCl to a final concentration of 150 mM and sonicate the sample using a Branson Sonifier 250. Sonicate twice for 30 sec at 30% duty cycle and output setting 3. Incubate sample on ice for 2 min between each sonication. 5. Transfer the sample to microcentrifuge tubes and pellet the cell debris by centrifugation for 10 min at 20,000 × g, 4◦ C. 6. Collect the supernatant and store the whole-cell lysate on ice. Measure the protein concentration by BioRad protein assay. This protocol typically yields 20 to 25 mg of total protein.
Fractionate HEK293 whole cell lysate 7. Prepare a 2-ml SP Sepharose column and a 2-ml Q Sepharose column by packing each of the resins into a 1.0 × 10 cm glass column fixed on an iron stand. Equilibrate the columns by gravity flow using at least 10 ml of cold column loading buffer. Carry out the entire fractionation procedure in a cold room. Alternatively, set up the columns at room temperature but keep all buffers and reagents on ice throughout the fractionation procedures. Instead of gravity flow, a peristaltic pump can be used to control the flow rate.
8. Dilute the whole-cell lysate ∼6 fold with 30 mM Tris·Cl, pH 7.5, in a 50-ml conical tube such that the final salt concentration is equivalent to the column loading buffer. Confirm by measuring its conductivity using a conductivity meter. 9. Load the diluted whole-cell lysate onto the SP column by gravity flow using a pipet. Collect the flowthrough in a 50-ml conical tube. 10. Wash the column with 5 column volumes of column loading buffer. 11. Elute the column sequentially with 3 column volumes each of column loading buffer containing 100 mM, 200 mM, 300 mM, 400 mM, and 600 mM NaCl. Collect the eluate at each salt-concentration step. 12. Load the SP flowthrough onto the Q Sepharose column and collect the flowthrough. 13. Repeat steps 10 to 11 on the Q Sepharose column. 14. To the Q Sepharose flowthrough, add ammonium sulfate powder gradually to 60% (∼360 mg/ml). Keep the solution on ice and mix constantly by vortexing the tube. 15. Pellet the proteins by centrifuging 10 min at 16,000 × g, 4◦ C. Remove the supernatant and resuspend the pellet in 0.5 ml of column loading buffer. 16. Concentrate all 10 column eluates in Amicon Ultra-4 centrifugal filter tubes by centrifugation (at 4000 × g maximum) in a tabletop centrifuge with swinging-bucket rotor. Concentrate until the sample volume is between 0.2 to 1 ml or the protein concentration is at least 2 mg/ml.
Category: Protein Modifications, Engineering and Applications
227 Current Protocols in Chemical Biology
Volume 2
17. Dialyze all lysate fractions against 1 liter of dialysis buffer for at least 3 hr at 4◦ C using MWCO 7000 dialysis tubing. Collect the fractions, quick freeze them in liquid nitrogen, and store up to 6 months at −80◦ C. Avoid using lysate volumes higher than 1/5 of the total reaction volume in a kinase assay as high glycerol concentrations may inhibit the kinase reaction. If the protein concentration of a lysate fraction is low, increase the total reaction volume to compensate for higher volume of the lysate. SUPPORT PROTOCOL 2
APPLYING THE KINASE SUBSTRATE IDENTIFICATION METHOD TO A CONTROL SUBSTRATE A simple kinase reaction using purified kinase and a known substrate is a quick way to test the method. Initial thiophosphate labeling can be achieved using wild-type kinase and ATP-γ-S, especially if the kinase is easy to produce or available commercially. The same assay can then be repeated with purified analog-sensitive kinase and an ATP-γ-S analog. The following protocol demonstrates the feasibility of the method using commercially available cyclin A-CDK2, ATP-γ-S, and purified GST-Rb. The results can be monitored quickly by mass fingerprint using matrix-assisted laser desorption/ionization time-offlight (MALDI-TOF) analysis (Henzel and Stults, 2001). The protocol is similar to the Basic Protocol with some following modifications. For more details, refer to the Basic Protocol.
Materials GST-Rb (expressed in E. coli and purified; Qin et al., 1992; contact Dr. Bruce Clurman,
[email protected]) ATP-γ-S (EMD Biosciences) Cyclin A-CDK2 (New England Biolabs) 0.5 M EDTA, pH 8 (MediaTech) Acetonitrile (VWR) 0.5 μg/μl trypsin (sequencing grade, Promega) 0.1% (v/v) formic acid (store in glass bottle at room temperature indefinitely) Disulfide beads: thiopropyl-Sepharose 6B (GE Healthcare) 20 mM NaOH 0.1% (v/v) formic acid/50% (v/v) acetonitrile (store in glass bottle at room temperature indefinitely) α-cyano-4-hydroxycinnamic acid (Agilent Technologies; also see Jim´enez et al., 1998) 1.7-ml microcentrifuge tubes 0.6 μl C18 ZipTips (Millipore) 4700 Proteomics Analyzer (Applied Biosystems) Additional reagents and equipment for preparing samples for MALDI mass spectrometry (Jim´enez et al., 1998) and MALDI-TOF mass spectrometry (Henzel and Stults, 1996) 1. Set up a 60-μl reaction containing 2 μg GST-Rb, 250 μM ATP-γ-S, and 200 ng cyclin A-CDK2. Incubate in 30◦ C water bath for 1 hr. 2. Add 2 μl 0.5 M EDTA, pH 8, 9 μl acetonitrile, and then 0.5 μg (1 μl) trypsin. Incubate at 37◦ C overnight. 3. Remove and save 1/3 (23 μl) of the total volume. Label as sample “A.” Engineered Kinases and Thiophosphate Labeling
228 Volume 2
4. Acidify the remaining sample (∼47 μl) with 1 μl 1% formic acid (to pH ∼5). Mix with 10 μl of settled disulfide beads in a 1.7-ml microcentrifuge tube. Incubate on a rotator at room temperature for 4 hr. Beads should be in suspension during the incubation. Tap the tube occasionally as needed. Current Protocols in Chemical Biology
5. Wash the beads (see Basic Protocol, step 21) and then collect them in a 1.7-ml microcentrifuge tube. 6. Elute the beads (similar to step 24 of the Basic Protocol) by adding 20 μl of 20 mM NaOH and mixing at room temperature for 30 min. 7. Collect the eluate (sample “B”) and acidify it with 1% formic acid (∼3 μl) to pH 3 or lower. Also, acidify sample A with 1% formic acid and then dilute the volume two-fold using 0.1% formic acid. Samples are acidified for the C18 desalting step below. Sample A needs to be diluted to reduce the concentration of acetonitrile.
8. Attach a 0.6-μl C18 ZipTip to a 20-μl pipet and desalt samples A and B following the manufacturer’s instructions. Briefly: a. Wet the ZipTip by pipetting up and down several times in 100% acetonitrile. b. Equilibrate the ZipTip by pipetting up and down several times in 0.1% formic acid. c. Pass the sample through the ZipTip by pipetting up and down 10 times.
A
% Intensity
6.3 x 104
Mass (m/z)
B 1358.56
% Intensity
3213
1122.46
1687.64
Mass (m /z)
Figure 3 MALDI-TOF analysis of purified phosphopeptides from thiophosphorylated GST-Rb. MALDI MS spectra with mass (m/z) shown on the x axis and % intensity shown on the y axis. (A) Peptide sample before thiophosphopeptide isolation. All expected Rb thiophosphopeptides are near background signal intensity. (B) After thiophosphopeptide isolation, three expected Rb phosphopeptides are enriched and clearly detectable: singly phosphorylated ISEGLPTPTK, m/z = 1122.46; singly phosphorylated SPYKFPSSPLR, m/z = 1358.56; and doubly phosphorylated ISEGLPTPTKMTPR, m/z = 1687.64.
229 Current Protocols in Chemical Biology
Volume 2
d. Wash the ZipTip by pipetting up and down several times in 0.1% formic acid. e. Elute the peptides by pipetting up and down several times in 3 μl of 0.1% formic acid/50% acetonitrile. Optional: Because of its larger volume, sample A can be loaded by pipetting the sample twice into another microcentrifuge tube using a C18 ZipTip and then transferring back and forth between the two tubes using the same C18 ZipTip.
9. Spot 0.5 μl of each sample (mixed with the matrix α-cyano-4-hyrdroxycinnamic acid; see Jim´enez et al., 1998) onto a MALDI plate and perform scans in MS mode using 4700 Proteomics Analyzer. Typically, apply 1000 laser shots using the appropriate laser intensity. Essential protocols for MALDI-TOF mass spectrometry are provided in Henzel and Stults (1998).
10. Compare the MS spectra of sample A and sample B. Identify enrichment of phosphopeptides of GST-Rb by the correct m/z values (Fig. 3). Phosphopeptide signal intensities depend on the degree of phosphorylation, purification yield, and their ionization efficiencies in the mass spectrometer.
REAGENTS AND SOLUTIONS Use Milli-Q purified water or equivalent in all recipes and protocol steps.
Column loading buffer 30 mM Tris·Cl, pH 7.5 25 mM NaCl 1 mM dithiothreitol (DTT) Prepare fresh in cold water and use one time Dialysis buffer 10% (v/v) glycerol 30 mM Tris·Cl, pH 7.5 100 mM NaCl Prepare fresh in cold water and use one time Hypotonic lysis buffer 50 mM Tris·Cl, pH 7.5 1 mM dithiothreitol (DTT) 1 mM MgCl2 0.1% (v/v) Triton X-100 25 U/ml Benzonase (Novagen) Protease inhibitor cocktail (Sigma-Aldrich) to 1× final Make fresh in cold water and use one time Kinase reaction buffer, 5× 200 mM Tris·Cl, pH 7.5 50 mM MgCl2 50 mM NaCl Store at room temperature indefinitely Washing solution 1 Engineered Kinases and Thiophosphate Labeling
0.1% (v/v) formic acid 50% (v/v) acetonitrile (VWR) Store room temperature up to 6 months
230 Volume 2
Current Protocols in Chemical Biology
Washing solution 2 2 M NaCl in water Store room temperature indefinitely COMMENTARY Background Information Protein phosphorylation is perhaps the most common signaling event in the cell and plays important roles in virtually all cellular process. Protein kinases mediate cellular signaltransduction events through the phosphorylation of their downstream targets. A major challenge when studying the function of kinases has been the identification of their direct targets. The large number of cellular kinases and their substrates, as well as the lack of suitable methods, have made it difficult to assign a particular protein phosphorylation to a specific kinase. Traditional and newer mass spectrometric approaches have generated ever-increasing lists of phosphorylation identifications. However, the direct kinase substrates of individual kinases remain elusive in most cases, and various strategies of substrate identification have been applied with limited success. The chemical genetic approach using analog-sensitive kinases and ATP analogs isolates the activities of a kinase of interest and allows specific and direct labeling of their targets in cell extracts. Earlier substrate-identification methods using this strategy had limited number of substrate identifications or depended on tagged substrate libraries (Koch and Hauf, 2010). The method described here labels direct kinase targets with thiophosphate tags, which allow specific capture of thiophosphopeptides (as well as cysteine-containing peptides) via a sulfhydryl (-SH) reactive resin following the digestion of the labeled proteins. Elution of the resin with a base specifically releases the thiophosphopeptides while leaving behind the vast majority of cysteine-containing peptides. Although cysteine-containing thiophosphopeptides are lost with this method, a large majority of the labeled peptides are recoverable. Identifying the labeled peptides directly allows more identifications to be achieved, and the sites of phosphorylation can also be mapped, providing further confidence for the identified substrates. A similar approach has been described in the literature (Blethrow et al., 2008), and detailed protocols have been provided (Hertz et al., 2010). Users have the option to try both methods and choose the one that suits them. It is noteworthy that the method described here
is based on in vitro reaction, and therefore are subject to false-positive identifications due to various reasons, such as the lack of normal cellular context and the use of excessive amount of kinases. Nevertheless, this direct approach can identify new physiological kinase targets, and it complements the current indirect strategies (Koch and Hauf, 2010).
Critical Parameters Effective thiophosphate labeling is an important first step toward successful substrate identification. Thus, the specificity and activity of the analog-sensitive kinase toward an ATP-γ-S analog should be optimized using a known substrate protein before moving to lysate-based identifications. The ability of a kinase to utilize ATP-γ-S can sometimes be enhanced by a different divalent cation cofactor (Parker et al., 2005). A small amount of control substrate “spike-in” should be included in kinase reactions using cell lysates to monitor the sensitivity and success of the procedures. Kinase reaction efficiency can be affected by the amount of kinase, cell lysate, or ATP-γ-S analog added, as well as the reaction time. Users can start with a reaction using more kinase, less cell lysate, and longer reaction time, and then optimize the reaction conditions using less kinase and shorter incubation time without compromising the results significantly. Although small amounts of salt and nonionic detergent are common in kinase reactions and harmless to these methods, reducing agents or sulfhydryl compounds (such as DTT) should be minimized, as they will react with the disulfide resin and reduce its capacity in the subsequent step. Large amounts of reducing agent (1 mM or higher) may interfere with the binding of thiophosphopeptides to the resin. Reducing agents used in cell lysate preparations should be significantly diluted or removed (e.g., by dialysis) prior to the kinase reaction. Similarly, sulfhydryl-reactive compounds (such as iodoacetamide) should be avoided throughout the procedure, as they will likely react with thiophosphopeptides. Efficient binding of thiophosphopeptides to disulfide beads is another critical step. Since the bead binding involves covalent
Category: Protein Modifications, Engineering and Applications
231 Current Protocols in Chemical Biology
Volume 2
Table 1 Troubleshooting Guide for Mass Spectrometry–Based Identification of Protein Kinase Substrates Utilizing Engineered Kinases and Thiophosphate Labeling
Problem
Possible cause
Solution
No thiophosphate labeling of a known substrate using ATP-γ-S
Not enough active kinase, or kinase uses ATP-γ-S poorly
Use more kinase. Increase the kinase reaction time. Try adding other divalent cations, such as Mn2+ .
No phosphopeptide identifications after the purification procedure on a known substrate
Insufficient labeling
Use more kinase. Increase the kinase reaction time. Try adding other divalent cations, such as Mn2+ .
Inefficient bead binding
Try smaller sample volume and longer incubation time (e.g., overnight)
Expected tryptic phosphopeptides are Try a different substrate or protease too large or small No thiophosphate labeling using cell Insufficient active kinase, or kinase lysates and ATP-γ-S analog uses ATP-γ-S analog poorly
No phosphopeptide identifications after the purification procedure on a lysate reaction
Use more kinase. Increase the kinase reaction time. Try adding other divalent cations, such as Mn2+ .
Insufficient ATP-γ-S analog
Increase ATP-γ-S analog concentration (up to 0.5 mM)
Inhibitory effect by the cell lysate
Use less cell lysate. Use fractionated cell lysates.
Insufficient labeling
Include a known substrate as control. Add more kinase and increase reaction time.
Insufficient digestion
Use more trypsin. Digest for longer time. Use denaturant during digestion followed by desalting of the peptide sample.
Inefficient bead binding
Reduce sample volume and increase incubation time with beads Avoid high concentration of reducing agents and denaturants (e.g., urea). Desalt the peptide sample prior to bead binding.
Phosphopeptides identified in “no-kinase” control
Low stoichiometry of substrates
Enrich the concentration of substrates by fractionating the cell lysates or cell compartments
Background labeling due to other kinases using the ATP-γ-S analog
Subtract out these peptide identifications from the list of “+ kinase” reaction Include background kinase specific inhibitors in the reaction (e.g., casein kinase II inhibitors)
High background signals in mass spectrometry analysis
Small molecules from the ATP-γ-S analog reagent that co-purify with phosphopeptides
Use purer ATP-γ-S analog. Desalt the final peptide sample using mixed cation exchange (MCX) prior to mass spectrometry.
Engineered Kinases and Thiophosphate Labeling
232 Volume 2
Current Protocols in Chemical Biology
attachments via a chemical reaction, the ratio of sample volume to bead volume is important. It is desirable to keep the peptide sample volume as small as possible to allow efficient chemical reaction. Beads should be constantly mixed with the peptide sample; the incubation time is typically overnight for complex samples. The amount of beads used is based on the sample complexity and can be adjusted accordingly so that the bead capacity is always several-fold in excess of the total sulfhydryl content of the kinase reaction. However, excess bead capacity may also increase the nonspecific binding and lead to higher background in the mass spectrometry analysis.
Troubleshooting Table 1 describes some of the potential problems that may occur during the procedure along with descriptions of possible causes and recommendations to overcome or avoid these problems.
Anticipated Results Generally, when the method is applied to simple reactions, such as the control kinase reaction described in Support Protocol 2, it is expected to be more efficient than reactions using complex cell lysates. For a known substrate, multiple phosphopeptides (if present) should be readily recovered and easily detectable by MALDI or electrospray mass spectrometers. For complex reactions in lysates, the phosphopeptide yield may be lower due to lower labeling efficiency, more competitive binding to the disulfide beads, and lower elution efficiency. Also, the sensitivity of the method may be lower due to higher background in the mass spectrometry analysis. When a control substrate is included in the lysate reactions, recovery of phosphopeptide(s) from the control substrate is expected. The more distinct peptides recovered from the control substrate, the better the overall reaction efficiency and yield of the phosphopeptides. Failure to recover any control phosphopeptides indicates problems during the procedure.
Time Considerations The most time-consuming steps are the preparatory procedures described in Strategic Planning. Once all necessary reagents are ready, and the activity and specificity of the engineered kinase has been optimized, an experiment using the standard substrate can be completed within 2 days, and an experiment involving multiple lysate reaction samples can also be processed in 2 to 3 days.
Literature Cited Allen, J.J., Li, M., Brinkworth, C.S., Paulson, J.L., Wang, D., Hubner, A., Chou, W.H., Davis, R.J., Burlingame, A.L., Messing, R.O., Katayama, C.D., Hedrick, S.M., and Shokat, K.M. 2007. A semisynthetic epitope for kinase substrates. Nat. Methods 4:511-516. Blethrow, J., Zhang, C., Shokat, K.M., and Weiss, E.L. 2004. Design and use of analogsensitive protein kinases. Curr. Protoc. Mol. Biol. 66:18.11.11-18.11.19. Blethrow, J.D., Glavy, J.S., Morgan, D.O., and Shokat, K.M. 2008. Covalent capture of kinasespecific phosphopeptides reveals Cdk1-cyclin B substrates. Proc. Natl. Acad. Sci. U.S.A. 105:1442-1447. Bond, J.S. 1989. Commercially available proteases. Appendix II in Proteolytic Enzymes, A Practical Protein Chemistry (R.J. Beynon and J.S. Bond, eds.) p. 240. IRL Press, Oxford. Buzko, O. and Shokat, K.M. 2002. A kinase sequence database: Sequence alignments and family assignment. Bioinformatics 18:1274-1275. Chi, Y., Welcker, M., Hizli, A.A., Posakony, J.J., Aebersold, R., and Clurman, B.E. 2008. Identification of CDK2 substrates in human cell lysates. Genome Biol. 9:R149. Deutsch, E.W., Mendoza, L., Shteynberg, D., Farrah, T., Lam, H., Tasman, N., Sun, Z., Nilsson, E., Pratt, B., Prazen, B., Eng, J.K., Martin, D.B., Nesvizhskii, A.I., and Aebersold, R. 2010. A guided tour of the Trans-Proteomic Pipeline. Proteomics 10:1150-1159. Henzel, W.J. and Stults, J.T. 1996. Matrix-assisted laser desorption/ionization time-of-flight mass analysis of peptides. Curr. Protoc. Protein Sci. 4:16.2.1-16.2.11. Hertz, N.T., Wang, B.T., Allen, J.J., Zhang, C., Dar, A.C., Burlingame, A.L., and Shokat, K.M. 2010. Chemical genetic approach for kinase-substrate mapping by covalent capture of thiophosphopeptides and analysis by mass spectrometry. Curr. Protoc. Chem. Biol. 2:15-36. Holt, L.J., Tuch, B.B., Villen, J., Johnson, A.D., Gygi, S.P., and Morgan, D.O. 2009. Global analysis of Cdk1 substrate phosphorylation sites provides insights into evolution. Science 325:16821686. Jim´enez, C., Huang, L., Qiu, Y. and Burlingame, A. 1998. Sample preparation for MALDI mass analysis of peptides and proteins. Curr. Protoc. Protein Sci.. 14:16.3.1–16.3.6. Kapp, E. and Schutz, F. 2007. Overview of tandem mass spectrometry (MS/MS) database search algorithms. Curr. Protoc. Protein Sci. 49:25.22.21-25.22.19. Koch, A. and Hauf, S. 2010. Strategies for the identification of kinase substrates using analogsensitive kinases. Eur. J. Cell Biol. 89:184-193. Liu, Y., Shah, K., Yang, F., Witucki, L., and Shokat, K.M. 1998. A molecular gate which controls unnatural ATP analogue recognition by the tyrosine kinase v-Src. Bioorg. Med. Chem. 6:12191226.
Category: Protein Modifications, Engineering and Applications
233 Current Protocols in Chemical Biology
Volume 2
Parker, L.L., Schilling, A.B., Kron, S.J., and Kent, S.B. 2005. Optimizing thiophosphorylation in the presence of competing phosphorylation with MALDI-TOF-MS detection. J. Proteome Res. 4:1863-1866.
Shah, K., Liu, Y., Deirmengian, C., and Shokat, K.M. 1997. Engineering unnatural nucleotide specificity for Rous sarcoma virus tyrosine kinase to uniquely label its direct substrates. Proc. Natl. Acad. Sci. U.S.A. 94:3565-3570.
Phelan, M.C. 2007. Basic techniques in mammalian cell tissue culture. Curr. Protoc. Cell Biol. 36:1.1.1–1.1.18.
Williams, A. and Frasca, V. 2001. Ion-exchange chromatography. Curr. Protoc. Protein Sci. 15:8.2.1-8.2.30.
Qin, X.Q., Chittenden, T., Livingston, D.M., and Kaelin, W.G. Jr. 1992. Identification of a growth suppression domain within the retinoblastoma gene product. Genes Dev. 6:953-964.
Yi, E.C., Lee, H., Aebersold, R., and Goodlett, D.R. 2003. A microcapillary trap cartridgemicrocapillary high-performance liquid chromatography electrospray ionization emitter device capable of peptide tandem mass spectrometry at the attomole level on an ion trap mass spectrometer with automated routine operation. Rapid Commun. Mass Spectrom. 17:2093-2098.
Shah, K. and Shokat, K.M. 2003. A chemical genetic approach for the identification of direct substrates of protein kinases. Methods Mol. Biol. 233:253-271.
Engineered Kinases and Thiophosphate Labeling
234 Volume 2
Current Protocols in Chemical Biology
Supported Membrane Formation, Characterization, Functionalization, and Patterning for Application in Biological Science and Technology Wan-Chen Lin,1 Cheng-Han Yu,2 Sara Triffo,1 and Jay T. Groves1,2,3,4 1
Howard Hughes Medical Institute, Department of Chemistry, University of California, Berkeley, California 2 Research Center of Excellence in Mechanobiology, National University of Singapore, Singapore 3 Physical Biosciences Division, Lawrence Berkeley National Laboratory, Berkeley, California 4 Materials Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, California
ABSTRACT Supported membranes, formed as a single continuous lipid bilayer on a solid substrate, such as silica, have been used extensively as a model for protein-protein and cell-cell interaction, to study the molecular interactions at interfaces and the heterogeneities of plasma membranes. The advantages of a supported membrane system include the ability to control membrane composition and the compatibility it has with various surface-sensitive microscopic and spectroscopic techniques. Recent advances in micro- and nanotechnology have greatly extended the use of supported membranes to address key questions in cell biology. Although supported membranes can be easily made by vesicle fusion, the samples need careful preparation for this process to be efficient. The protocols in this unit comprehensively describe procedures to prepare, functionalize, and characterize C 2010 by John Wiley & supported membranes. Curr. Protoc. Chem. Biol. 2:235-269 Sons, Inc. Keywords: supported membrane r supported lipid bilayer r small unilamellar vesicle (SUV) r membrane functionalization r fluorescence recovery after photobleaching (FRAP) r quantitative fluorescence measurement r photolithography
INTRODUCTION For more than 25 years, phospholipid bilayers deposited onto solid substrates have been one of the most commonly used experimental cell-surface models, and they have allowed researchers to gain insights into the heterogeneity and functions of cell membranes. Supported membranes are of practical and scientific interest because they can easily be prepared by depositing lipids onto large areas of solid substrates (in the order of cm2 ), which provides excellent mechanical stability, while maintaining good fluidity. The combination of planar geometry and solid supports of a supported membrane offers distinct advantages and allows experiments that are difficult or impossible with other model systems like freestanding black lipid membranes or spherical lipid vesicles. It is also easy to make supported membranes functional by incorporating membrane-associated proteins, thus enabling inorganic substrates to be biofunctionalized (Fig. 1). In this unit, procedures to prepare and characterize functionalized supported membranes on glass substrates, based on vesicle fusion, are described in detail. Lipid suspensions for vesicle fusion can be produced by different methods: extrusion, sonication, and freeze-thawing (Basic Protocol 1, Alternate Protocols 1 and 2, respectively). As the cleanliness of the glass substrates is of paramount importance in the formation of supported membranes, Volume 2 Current Protocols in Chemical Biology 2: 235-269, December 2010 Published online December 2010 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/9780470559277.ch100131 C 2010 John Wiley & Sons, Inc. Copyright
235 Volume 2
Figure 1 Examples of functionalized supported membranes. (A) Functionalization through polyhistidine and lipids with Ni2+ NTA headgroup. (B) Functionalization through cysteine and lipids with maleimide headgroup. The art work in this figure is kindly provided by Dr. Lars Iversen.
four different methods for cleaning substrates are provided (Basic Protocol 2, Alternate Protocols 3, 4, and 5). Standard procedures to prepare glass substrates, containing nanostructures that can guide and modulate the spatial organization of the molecule of interest, are also included (Support Protocols 1 and 2). Basic Protocol 3 describes the formation and functionalization of supported membranes. Specific instructions for several different methods for functionalization are provided in this protocol. In Support Protocol 3, universally applicable steps in depositing lipid bilayers onto micron-size particles are described, while a detailed procedure for making supported intermembrane junctions is presented in Support Protocol 4. Finally, methods for fluorescence recovery after photobleaching (FRAP), a commonly used method for measuring the diffusivity of supported membranes, and for quantitative fluorescence measurements to quantify the surface density of fluorescent molecules on the membrane are described in Basic Protocol 4.
STRATEGIC PLANNING
Supported Membrane Formation, Characterization, Functionalization and Patterning
Choice of Lipids Table 1 lists the most commonly used lipids for supported membranes. Unsaturated lipids, such as 1,2-dioleoylphosphotidylcholine (DOPC) and 1-palmitoyl-2-oleoylphosphotidylcholine (POPC), are commonly used as the main lipid species for most applications in biological sciences. Bilayers comprising these two lipids are fluid at room temperature, but these unsaturated lipids are easily oxidized and the bilayer can contain products of oxidation like lipid fragments and lyso-lipids (see Commentary for explanation). For experiments that are sensitive to the composition and mixing behavior of the lipid mixture, it is recommended that saturated lipids be used as the main component. Lipids like 1,2-dimyristoylphosphocholine (DMPC) and 1,2-dilauroylphosphocholine (DLPC) are often preferred for a fluid membrane composed of saturated lipids. Note that
236 Volume 2
Current Protocols in Chemical Biology
Table 1 Commonly Used Lipids
Table 2 Commonly Used Fluorescent Lipids
Fluorescent lipid type
Excitation/emission maxima
Comments
Source
Texas-Red DHPE
∼595/615 nm
Bright and stable (negatively charged)
Invitrogen
BODIPY lipids
Several options range from green to red.
Bright, stable, and has a high quantum yield with a narrow spectral bandwidth. Can be headgroup labeled (negatively charged) or fatty acid labeled (neutral).
Invitrogen
NBD lipids (fatty acid labeled)
∼464/531 nm
Commonly used green probe. Easily Avanti polar lipids photobleached and hence good for FRAP, but not ideal for measurements using laser excitation, such as fluorescence correlation spectroscopy.
Marina Blue DHPE
∼365/460 nm
Low quantum yield and is easily Invitrogen photobleached, but the UV excitation makes it ideal when using multiple fluorescent labels. Volume 2
237 Current Protocols in Chemical Biology
Volume 2
Table 3 Summary of Linkages for Protein Functionalization Lipid (concentration)
Linkage type
Biotin-streptavidin N-cap-biotinylDOPE (0.01 to 1 mol%)a
Protein modification
Advantages
Disadvantages
Biotinylation
• Preparation of biotinylated proteins is straightforward with a kit
• Requires two steps: the supported membrane is first functionalized with streptavidin and then the biotinylated protein is allowed to bind to it. • Poor control over protein orientation (depends on the site of biotinylation in the protein). • Streptavidin may introduce artificial cross-linking, as it has four binding sites. • Streptavidin aggregates at high surface density
Ni2+ NTAPolyhistidine
Ni2+ -NTA DOGS His 10 or greater (1 to 2 mol%)a
• Protein orientation can be specified • Proteins can easily be expressed with a polyhistidine tag
• Noncovalent linkage • Sample is sensitive to temperature change • Limited protein surface density. The upper limit for a sample containing 2 mol% Ni-NTA-lipid is ∼5000/μm2 .
Maleimidecysteine
Maleimide-DHPE (1 to 7 mol%)a
Free cysteine residue
• Covalent linkage • Protein orientation can be specified • Suitable for a wide range of surface densities
• Requires an exposed cysteine • Additional cysteins may have to be substituted if only one binding site is desired
Lipidated protein
No need for special lipids
Lipid anchor (such as • Protein orientation can GPI or alkyl chains) be specified
Proteoliposome
Proteoliposomes No extra can be directly functionalization step mixed with regular required SUV suspension to make supported membranes.
• Requires significant knowledge of synthetic pathways in organic chemistry • Samples may be sensitive to temperature change • Lipidated protein may form aggregates in solution • The insertion efficiency varies in different molecules • Protein orientation cannot be specified • Less control over protein surface density • Proteoliposomes may not rupture completely to form supported membrane. As a result, liposomes can stick to the surface and create defects in the membrane, affect the fluidity of the supported membrane, and bias quantitative measurements. • Preparation is laborious. Subject to degradation depending on additional treatment of proteins
a The concentrations listed are recommended. Supported membranes often contain immobile fractions if the concentration of the functional lipids are
higher than listed.
238 Volume 2
Current Protocols in Chemical Biology
DMPC has a transition temperature near room temperature, which can lead to complex behaviors (Yamazaki et al., 2005; Forstner et al., 2006). DLPC has short fatty acid chains and therefore it forms a relatively thin lipid bilayer, ∼0.4-nm thinner than a DOPC bilayer, which is about 4.5-nm thick. A small amount (<0.5 mol%) of fluorescent lipids can be included to visualize the supported membrane (Maier et al., 2002). If fluorescent proteins are to be used for functionalization, it is important to avoid spectral overlap between fluorescent lipids and the fluorescently labeled protein of interest. Table 2 summarizes several commonly used and commercially available fluorescent lipids and their properties. Lipids with a charged headgroup can be used in different concentrations in the membranes to vary the surface charge density of supported membranes. Examples include the negatively charged phosphatidylserine (PS), phosphatidylglycerol (PG), phosphatidic acid (PA), and the cationic lipid trimethylammonium-propane (TAP). However, it is important to select lipids that have unsaturated fatty acid chains similar to the main lipid in the membrane. For example, choose lipids that have di-oleoyl fatty acid chains, such as DOPS and DOTAP, if DOPC is the main lipid. Examples of lipid mixtures, along with detailed measurements of the resulting surface charge densities in the supported membrane, can be found in Gomez et al. (2009). It is worth noting that the substrate generally contributes a small negative charge so that neutral lipids do not make neutral supported membranes. Lipids containing a functional headgroup, such as biotin, nickel chelate of nitrilotriacetic acid (Ni-NTA), and maleimide can be included in the lipid mixtures for protein functionalization. Each linkage has different working conditions and special requirements, and a selection guide for protein functionalization is summarized in Table 3.
Methods for Preparing SUV Suspensions The basic steps in the preparation of a small unilamellar vesicle (SUV) suspension are solvent evaporation, rehydration, formation of multilamellar vesicles, and dispersion into SUVs (Fig. 2). Three different methods are provided in this unit for dispersion of SUVs—extrusion, sonication, and freeze-thawing. In general, extrusion allows better control over the size of SUVs, which is determined in part by the pore size of the
sonication extrusion freeze-thaw
rehydration
dried lipid films
multilamellar vesicles
small unilamellar vesicles
Figure 2 Steps in the formation of SUV suspension. As lipids are dried, a thin film is formed in a round-bottomed flask. Upon addition of water or buffer, the lipids self-assemble into bilayers, which lift off the glass as spherical vesicles of various diameters. These multilamellar vesicles are then treated with sonication, extrusion, or freeze-thawing to yield single-layered vesicles of roughly uniform diameter. Volume 2
239 Current Protocols in Chemical Biology
Volume 2
polycarbonate membrane used. In addition, compared to sonication (see below), lipid suspensions made by extrusion are subjected to minimal lipid oxidation and maintain the intended lipid composition. However, cross-contamination between samples can easily occur if the extruder is not properly cleaned. Furthermore, certain lipids, exemplified by PG and phosphatidylinositolphosphate (PIP), stick to the polycarbonate membrane and may clog the pores during extrusion, or alter the final lipid composition. The probe sonication method utilizes high ultrasonic energy to breakdown multilamellar vesicles into SUVs. During sonication, the energy delivered to the lipid suspension causes strong cavitation, which breaks vesicles and increases the temperature of the solution. For the same reason, lipid oxidation and fragmentation can occur more often in this method. Nevertheless, for all lipid mixtures listed above, probe sonication delivers satisfying results for supported membrane formation, when it is performed properly (see Alternate Protocol 1). The biggest advantage of this method is the ease at which the equipment, a titanium probe in this case, can be cleaned. Hence, cross-contamination between samples is usually not a problem. The last method, freeze-thawing, is the simplest among the three. However, it is suitable for certain lipid mixtures only (see Alternate Protocol 2 for details). Importantly, these methods can be employed in combination to prepare SUV suspensions. For example, freeze-thawing the suspension can be followed by extrusion or sonication.
Methods for Cleaning Substrates A key component in preserving the two-dimensional mobility of supported lipid bilayer membranes is the surface chemistry of supporting substrates. Ultraclean and hydrophilic silicon dioxide (SiO2 ) surfaces (such as borosilicate, fused silica, and native oxide on silicon wafer) have been widely used as suitable substrates for lipid bilayer deposition. Various cleaning techniques, adapted from research in interfacial and solid-state physics, can be used to clean substrates. These include piranha etching (solution-based) and base etching (solution-based), and air/O2 plasma (gas-like) and UV/ozone (gas-based) treatments. Piranha solution, a mixture of sulfuric acid and hydrogen peroxide, provides strong cleaning and effective removal of organic contamination. Since the piranha solution is a strong oxidizer, a high number of hydrogen-bonded vicinal silanol (Si-OH) groups become available on the surface of SiO2 substrates, resulting in a strongly hydrophilic surface. Gas-based cleaning techniques, such as air/O2 plasma and UV/ozone treatment, can also reduce organic contamination and promote hydrophilicity over SiO2 surfaces by generating oxidative radicals in the gas phase.
Supported Membrane Formation, Characterization, Functionalization and Patterning
Among the three methods described above, piranha etching generally provides the highest cleaning strength in a rather prompt manner. However, piranha solution is extremely corrosive. Thus, proper personal protective equipment is required (see Basic Protocol 2 for details). Base etching is a common method to clean glass surfaces in chemistry laboratories. A base bath is usually used multiple times to reduce the hazardous chemical waste. Hence, the chance of cross-contamination is higher using this method to clean the substrate. Piranha and base solutions need to be handled with caution and they require disposal as hazardous chemical wastes. Air/O2 plasma and UV/ozone cleaning may take longer, but the relatively moderate reaction may better preserve prefabricated microstructures on SiO2 surfaces. When employing plasma or UV cleaning, it is absolutely necessary to avoid exposure to UV light and to have proper ventilation for ozone. While chemical cleaning methods provide near complete removal of carbohydrates on glass surfaces, most of the inorganic contamination will not be purged. A precleaning procedure, such as intense bath sonication, is generally helpful to physically remove dust particles from surfaces.
240 Volume 2
Current Protocols in Chemical Biology
PREPARING SMALL UNILAMELLAR VESICLES (SUVs) BY EXTRUSION This protocol describes the preparation of an SUV suspension for supported membrane formation. The lipid mixture is first dried on a glass flask to form layers of lipid films, followed by rehydration in an aqueous solution. The resulting suspension containing large multilamellar vesicles is then pushed through a polycarbonate membrane of defined pore size (30 to 200 nm). During this process, large vesicles rupture into small unilamellar vesicles, the size of which is dictated by the pore size of the polycarbonate membrane (Hope et al., 1985; Patty and Frisken, 2003; Kunding et al., 2008). This method allows users to control the size distribution of resulting SUVs. In addition, the concentration of lipids is presumably the same before and after extrusion, so this is a superior method to use if a defined concentration of vesicles is required. However, this may not apply to certain lipid species that adhere to the polycarbonate membrane, such as PG and PIP.
BASIC PROTOCOL 1
Materials Stock lipid solutions in chloroform or a 2:1 (v/v) chloroform:methanol mixture, at a concentration of 0.1 to 10 mg/ml Chloroform (ACS grade and above) Nitrogen gas (industrial grade or better; if from a central supply, use a hydrophobic filter to remove oils from the gas) 1 to 4 ml rehydration solution [deionized (DI) water or TBS or PBS; see recipes; see Commentary for selection] Deionized water (resistivity ≥18 M and apparent pH 5.5) Argon 25- to 50-ml glass round-bottomed flask (cleaned by piranha or base etching; see Basic Protocol 2 and Alternate Protocol 3) Positive-displacement pipet with capillary piston made of pure polypropylene (Gilson), or Hamilton syringes Rotary evaporator attached to a vacuum pump 40◦ to 50◦ C water bath Benchtop vortex mixer Extruder: Lipex extruder (Northern Lipids) or Avanti Mini-Extruder (Avanti Polar Lipids) NOTE: All stock lipid solutions should be stored in glass vials with Teflon caps or Teflon septa at −20◦ C or lower. Chloroform and methanol should be ACS grade or above.
Form dry lipid film and rehydrate 1. Remove the stock lipid solutions from the freezer and bring to room temperature before use. 2. Rinse a round-bottomed flask and Hamilton syringes (if used) with chloroform several times. 3. Use a positive-displacement pipet or Hamilton syringes to measure the appropriate volumes of lipid stock solutions and add to the round-bottomed flask. The total amount of lipid usually ranges from 0.2 mg to 8 mg. Quickly cap the stock solution to ensure consistency of stock lipid concentration, as chloroform evaporates rapidly. Use Teflon tape to help seal the glass vial. If fluorescent lipids and/or maleimide-lipids are used, protect the samples from exposure to light by covering the samples with aluminum foil.
4. Attach the round-bottomed flask to a rotary evaporator. Volume 2
241 Current Protocols in Chemical Biology
Volume 2
5. Turn on the vacuum pump and evaporate the chloroform while rotating the flask in a warm water bath at a temperature that is higher than the phase-transition temperature of every constituent lipid. Evaporate until the solution is dry by eye, usually ∼5 min under ∼700 mmHg negative pressure. Use 40◦ C to 50◦ C for lipid mixtures that are fluid at room temperature. There should be a uniform and semi-transparent layer of dry lipid film at the bottom of the flask.
6. Detach the round-bottomed flask from the evaporator and place it under a strong stream of N2 for ∼5 min to remove excess chloroform. Alternatively, place the flask under vacuum for 2 hr.
7. Add 1 to 4 ml of rehydration solution to the flask to bring the lipid suspension to the desired final concentration (0.05 to 2 mg/ml). If either TBS or PBS is used for rehydration, keep the final concentration below 0.5 mg/ml.
8. Shake or vortex the flask for 1 min, or until there are no more dried lipids attached to the bottom of the flask. Maintaining the temperature above the phase-transition temperature of every lipid species helps in complete rehydration. The resulting suspension usually appears milky.
Perform extrusion 9. Clean and assemble the extruder by following the manufacturer’s instructions. Place a polycarbonate membrane of defined pore size in the extruder (30- to 200-nm diameter). Extruders are notorious for causing cross-contamination between samples. If possible, it is better to separate the extruder parts and store them immersed in organic solvents, such as isopropanol or methanol.
10. Wet the polycarbonate membrane by pushing 5 ml of deionized water through the extruder. 11. Extrude the lipid suspension at a temperature that is higher than the phase-transition temperature of every constituent lipid fifteen times or until the suspension reaches clarity. It is often difficult to extrude lipid suspensions through membranes with a small pore size, e.g., 30 nm. It is possible to use membranes with a larger pore size for extrusion initially and then proceed to using a smaller pore size.
Collect and store SUVs 12. Collect SUVs in a microcentrifuge tube. 13. Store SUVs at 4◦ C. Seal the microcentrifuge tube under argon for long-term storage. It is best to use the SUVs within a week after preparation. ALTERNATE PROTOCOL 1
Supported Membrane Formation, Characterization, Functionalization and Patterning
242 Volume 2
PREPARING SUVs BY PROBE SONICATION Besides extrusion, probe sonication is another method commonly used to prepare SUV suspensions (Huang, 1969; Tenchov et al., 1985). During sonication, high-energy ultrasonic waves, delivered through a titanium probe, produce strong cavitation and acoustic power for mixing and particle dispersion. This process not only breaks large vesicles into SUVs but also heats the suspension. Therefore, it is best to perform the procedures with the sample immersed in an ice bath and in a nitrogen atmosphere, to minimize lipid oxidation. This process results in vesicles of ∼30-nm mean diameter as measured by electron microscopy (Tenchov et al., 1985). Centrifugation or ultracentrifugation is commonly used after sonication to remove unbroken large vesicles and small fragments of the metal probe. The supernatant alone is used for supported membrane formation; Current Protocols in Chemical Biology
hence, the lipid concentration and composition of the SUV suspension may be altered. In addition, by running the SUVs through a gel filtration column, such as a Sepharose CL-4B column, different sizes of SUVs can be separated into different fractions. Note that it is necessary to equilibrate the column with lipid vesicles first to minimize lipid adsorption during filtration.
Materials Isopropanol (ACS grade and above) Cleaning solution: 1:1 (v/v) isopropanol/water Deionized (DI) water (resistivity ≥18 M and apparent pH 5.5) 1 to 4 ml of lipid suspension (see Basic Protocol 1, steps 1 to 8) Nitrogen gas (industrial grade or better; if from a central supply, use a hydrophobic filter to remove oils from the gas) Argon Emery sheet (3/0 grit or finer) Ultrasonic processor equipped with a double-stepped microtip (e.g., VCX750, Sonics & Materials) in a sound-abating enclosure Ice bath Centrifuge that can reach 16,000 × g Microcentrifuge tubes 1. Use an emery sheet to polish the end of the microtip until the metal surface is shiny and free of any pits. 2. Rinse the microtip with 100 ml of isopropanol. 3. Sonicate the microtip in the cleaning solution at 40% amplitude for 2 min, and then rinse with 200 ml deionized water. 4. Sonicate the microtip in deionized water for 2 min. Adjust the amplitude during the sonication to achieve strong cavitation. The amplitude that results in good cavitation varies with sample temperature and the length of the microtip. The sonicator can be tuned to generate good cavitation by adjusting the amplitude while listening to the sound. A dramatic rise in the pitch of the sound indicates good cavitation. Do not exceed the maximum amplitude allowed (usually 40%) or the microtip will snap.
5. Rinse the microtip with 200 ml deionized water. 6. Place the lipid suspension in the sonicator enclosure so that the microtip is submerged in the solution without touching the tube itself and fill the enclosure with nitrogen gas. 7. Sonicate the lipid suspension, at the amplitude producing good cavitation in a nitrogen atmosphere and in an ice bath, twice, for 30 to 40 sec with a 10-sec pause in between. The use of nitrogen gas and an ice bath prevents lipid oxidation during sonication. The lipid suspension should reach clarity after two short sonications. If the suspension still appears cloudy, perform another 30- to 40-sec sonication. Alternatively, sonicate for 1.5 to 2 min with a 1-sec pause every 5 sec.
8. Centrifuge the lipid suspension 20 min at 16,000 × g, 4◦ C. Alternatively, ultracentrifuge the lipid suspension 1 hr at 166,000 ×g, 4◦ C, for an extra clean SUV suspension.
9. Transfer the supernatant liquid (containing SUVs) into a microcentrifuge tube. 10. Store SUVs at 4◦ C. Seal the tube under argon gas for long-term storage. It is recommended to use SUVs within a week after preparation. Current Protocols in Chemical Biology
Volume 2
243 Volume 2
ALTERNATE PROTOCOL 2
PREPARING SUVS BY FREEZE-THAWING This protocol uses repetitive freeze-thawing of lipid suspensions to produce SUVs. During freezing and thawing, multilamellar vesicles progressively fragment into small vesicles. This is a relatively simple method for preparing SUVs; however, it is not suitable for every lipid mixture. It works well only with unsaturated lipids. The fragmentation is most likely due to osmotic effects and requires the presence of salt (e.g., 0.1 M of NaCl or CaCl2 ); it does not occur in pure water (Mayer et al., 1985; Macdonald et al., 1994; Traikia et al., 2000).
Materials 1 to 4 ml of lipid suspension (see Basic Protocol 1, steps 1 to 8) Argon Dry ice-ethanol bath (or liquid nitrogen) 50◦ C water bath Microcentrifuge tubes 1. Immerse the lipid suspension in the dry ice-ethanol bath for 1 min. 2. Thaw the lipid suspension in a warm water bath for 1 min. 3. Repeat the freeze-thaw cycle until the lipid suspension becomes clear. SUVs are usually formed after 10 to 20 freeze-thaw cycles.
4. Transfer the SUVs into a microcentrifuge tube. 5. Store SUVs at 4◦ C. Seal the microcentrifuge tube under argon gas for long-term storage. It is recommended to use SUVs within a week after preparation. BASIC PROTOCOL 2
Supported Membrane Formation, Characterization, Functionalization and Patterning
PREPARING MEMBRANE SUPPORTS BY PIRANHA ETCHING This protocol uses a piranha solution, which is prepared by adding hydrogen peroxide to sulfuric acid, to clean glass surfaces for supported membrane formation. Due to the strong dehydrating power of sulfuric acid, addition of hydrogen peroxide is highly exothermic. In general, the temperature of freshly prepared piranha solution can reach 100◦ C in <5 min (Reinhardt and Kern, 2008). Atomic oxygen, an extremely reactive oxygen free radical, is generated by dehydration of hydrogen peroxide. The intense heat and strong oxidative reagents, thus generated, rapidly and thoroughly oxidize the organic compounds on the glass surface. Furthermore, atomic oxygen increases the number of silanol groups (SiOH) on the glass surface (McIntire et al., 2006; Reinhardt and Kern, 2008). Polar silanol groups form hydrogen bonds with vicinal water molecules and promote surface hydrophilicity (Ashley et al., 2003; Reinhardt and Kern, 2008). At neutral pH, silanol groups deprotonate (SiO− + H+ ) and the glass surface becomes negatively charged (Reinhardt and Kern, 2008). Extended piranha cleaning may result in a small increase in the roughness of glass surfaces (Williams et al., 2003; Seu et al., 2007). Piranha solution is extremely corrosive and extreme caution must be exercised while using it. Only mix the solution in a flow hood with the sash pulled down. Personal protective equipment, such as a full face shield, heavy duty rubber gloves (regular Nitrile gloves will not provide sufficient protection), as well as an acid apron to wear on top of the laboratory coat are required. Containers used during the experiment must be very clearly labeled and a warning sign, visible by any user working under the flow hood, must be posted at all times to indicate that the solution contains piranha mixture. Always consult laboratory safety personnel for instructions for handling and disposal of piranha solution.
244 Volume 2
Current Protocols in Chemical Biology
Materials 1:1 (v/v) isopropanol/water Deionized water (resistivity ≥18 M and apparent pH 5.5) Sulfuric acid (H2 SO4 ; ACS grade) 30% hydrogen peroxide (H2 O2 ; ACS grade) Glass substrates Teflon or glass rack/holder Bath sonicator 1. Place the substrates on a Teflon or glass rack/holder and immerse them in a beaker of isopropanol/water mixture. Sonicate in a bath sonicator for at least 30 min. The purpose of this step is to preclean the substrates. Substrates can be left in the isopropanol/water mixture overnight after sonication.
2. Remove the substrates from the isopropanol/water mixture and rinse with plenty of deionized water. 3. Place the substrates into a beaker of deionized water. Sonicate in a bath sonicator for 15 min. 4. Rinse again with plenty of DI water, and repeat step 3. CAUTION: Piranha acid reacts violently with organic solvent and may cause explosions. It is important to remove all traces of isopropanol on the substrate before proceeding to the next step.
5. In an acid hood, prepare the piranha solution by adding 1 part 30% hydrogen peroxide to 3 parts sulfuric acid. CAUTION: When preparing the piranha solution, always add the peroxide to the acid. The reaction is extremely exothermic and can cause the solution to boil. Do not make more than 100 ml at a time.
6. Immediately immerse the substrates in the piranha solution for 3 min, and then move the substrates into a beaker filled with of deionized water. It is important to keep the etching time consistent, as the surface charge density of the substrate varies as a function of etching time. When using patterned substrates (see Support Protocol 1), the etching time should be decreased to 1 min, as the piranha solution can etch away metal lines on the substrate.
7. Quickly move the substrates to a continuous stream of deionized water and rinse for 3 min. While rinsing, pass the substrate frequently through the air-water interface to help remove acid molecules. 8. Store the substrates in deionized water until use. The substrates should be used within 5 days.
9. Leave the hot piranha solution in an open container until cool. Consult laboratory safety personnel for instructions for disposal of the piranha solution.
CLEANING SUBSTRATES BY BASE ETCHING This protocol describes the widely used alkaline etching method to clean glass surfaces. Basic solutions (pH >9), such as potassium or sodium hydroxide, react with surface SiO2 and produce soluble silicate (SiO2 + 4OH− → SiO4 −4 + 2H2 O; Williams et al., 2003; Doering and Nishi, 2008). While base etching directly removes interfacial SiO2 layers and absorbed particles, the supplementation with isopropanol (up to 70% v/v) in the etching solution further increases the cleaning strength by dissolving organic matter
ALTERNATE PROTOCOL 3
Volume 2
245 Current Protocols in Chemical Biology
Volume 2
present on glass surfaces (Aegerter and Mennig, 2004). Newly exposed siloxane groups (Si-O) are then hydrolyzed to silanol groups, which increases surface hydrophilicity. Due to the corrosive nature of this alkaline-based cleaning method, prolonged etching often results in rough surfaces and should be avoided. This method is usually used to clean glass substrates when piranha etching is not suitable.
Materials 1 M sodium hydroxide (NaOH; ACS grade) Deionized water (resistivity ≥18 M and apparent pH 5.5) Glass substrates 1. Incubate the substrates with 1 M NaOH for 1 hr at room temperature. 2. Move the substrates to a continuous stream of DI water and rinse for 3 min. 3. Check hydrophilicity of the glass surface by adding a drop of deionized water on the surface and observing the degree of wetting. Water should spread out and form a thin layer on a sufficiently hydrophilic surface.
4. Store the substrates in deionized water until use. The substrates should be used within 5 days. ALTERNATE PROTOCOL 4
CLEANING SUBSTRATES WITH AIR/OXYGEN PLASMA This protocol uses a gas-based cleaning method and effectively produces hydrophilic glass surfaces (Tiller et al., 1978; DeRosa et al., 2003). Ionized plasma consists of ions and electrons that are generated by a strong electrical field and then accelerated by external radio frequency (RF) waves. Bombarding glass surfaces with plasma increases the amount of hydrogen-bonded vicinal silanol (Si-OH) groups and promotes surface hydrophilicity (Tiller et al., 1978; Amirfeiz et al., 2000; DeRosa et al., 2003; Michel et al., 2009). Organic contaminations over the glass are also chemically removed by oxygen radicals generated by plasma (Kondoh et al., 1998; Goldman et al., 2009; Michel et al., 2009). Oxygen plasma is purple in color and emits intense ultraviolet light; hence, direct exposure to eyes must be avoided. Oxygen plasma in a low-pressure environment, typically 200 mTorr in a capacitatively coupled plasma generator, often provides higher cleaning power, but lowpressure air plasma may also be sufficient for cleaning. Intense plasma exposure could result in increased surface roughness (Amirfeiz et al., 2000; Michel et al., 2009; Tiller et al., 1978), especially on doped silica, such as borosilicate glass. While piranha and base etching described above are very effective in cleaning glass and maintaining surface hydrophilicity, plasma-based cleaning methods provide remarkable cleaning strength with minimal chemical waste. However, general precleaning procedures, such as bath sonication, are strongly recommended before plasma treatment.
Materials Nitrogen gas (industrial grade or higher; if from a central supply, use a hydrophobic filter to remove oils from the gas) Deionized water (resistivity ≥18 M and apparent pH 5.5)
Supported Membrane Formation, Characterization, Functionalization and Patterning
Plasma generator (SPI Plasma-Prep II, SPI Supplies/Structure Probe) Mechanical vacuum pump with oil filters (Leybold Vacuum Pumps, SPI Supplies/Structure Probe) Oxygen gas with regulator, optional Additional reagents and equipment for precleaning the glass substrates (Basic Protocol 2, steps 1 to 4)
246 Volume 2
Current Protocols in Chemical Biology
1. Perform steps 1 to 4 of Basic Protocol 2. 2. Dry the glass substrates with nitrogen gas. 3. Place the precleaned glass substrates inside the plasma chamber. Use the attached mechanical pump and venting switch to decrease air pressure to 200 mTorr in the chamber. If plasma from pure oxygen gas is preferred, use a gas regulator to set the chamber pressure to 200 mTorr.
4. Switch on the RF power of the plasma generator (typically 100 W). When precleaned glass substrates are used, a 15-min plasma treatment is typically sufficient for supported membrane deposition. The purple-colored plasma contains intense ultraviolet light, so avoid direct exposure to eyes.
5. Switch off the plasma generator and pump, and then vent the chamber through an appropriate gas exhaust, such as a chemical hood. Rinse the glass substrates with 200 ml deionized water twice and dry with nitrogen gas. Glass substrates will remain clean and hydrophilic for 3 to 5 days when stored in dehumidified cabinets.
CLEANING SUBSTRATES WITH ULTRAVIOLET LIGHT/OZONE This protocol describes another useful gas-phase cleaning method. Deep ultraviolet light sources, typically a low-pressure mercury vapor grid lamp, provide strong emission at both 254 nm and 185 nm (avoid direct exposure to eyes). UV light at 254-nm can directly degrade organic contaminations on glass surfaces (Vig, 1987; Kasi and Liehr, 1990; Moon et al., 1999). Moreover, 185-nm UV light reacts with oxygen in ambient air and generates strong oxidizing agents—ozone and atomic oxygen—which also remove organic contaminations by cleaving hydrocarbon bonds (Vig, 1985; Baumgartner et al., 1987). Improvement in hydrophilicity is achieved by diminishing organic carbon residues on the glass surface and by increasing the number of silanol (Si-OH) groups via oxidation (Pietsch et al., 1994; Michel et al., 2009). Additionally, water molecules and ozone can produce hydroxyl radicals in an aqueous environment and enhance oxidization (Vig, 1985; Legrini et al., 1993; Yu et al., 2005). Contact with or inhalation of ozone should be avoided by venting it through a proper exhaust, such as a chemical hood. Similar to cleaning with air/oxygen plasma, precleaning procedures are usually required prior to UV/ozone treatment. Overexposure to UV/ozone also results in increased surface roughness (Richter and Brisson, 2003).
ALTERNATE PROTOCOL 5
Materials Nitrogen gas (industrial grade or higher; if from a central supply, use a hydrophobic filter to remove oils from the gas) Deionized water (resistivity ≥18 M and apparent pH 5.5) UV/ozone cleaner (UV/Ozone Procleaner Plus, BioForce Nanosciences) UV protective safety glasses Additional reagents and equipment for precleaning the glass substrates (Basic Protocol 2, steps 1 to 4) 1. Perform steps 1 to 4 of Basic Protocol 2. 2. Dry the glass substrates with nitrogen gas. Volume 2
247 Current Protocols in Chemical Biology
Volume 2
3. Place the UV/ozone cleaner at a proper location in which the steady flow of gas is vented safely. Then, place the precleaned glass substrates inside the UV/ozone chamber. Avoid contact with ozone gas and oxidative radicals.
4. Switch on the UV/ozone cleaner. In most cases, when using precleaned glass substrates, a 30-min UV/ozone treatment is typically sufficient to clean surfaces for supported membrane deposition. UV protective safety glasses should be used to avoid exposure to UV light.
5. Switch off the UV/ozone cleaner and let ozone exhaust through a chemical hood. Rinse the glass substrates with 200 ml deionized water twice and dry with nitrogen gas. Glass substrates will remain clean and hydrophilic for 3 to 5 days when stored in dehumidified cabinets. SUPPORT PROTOCOL 1
PREPARING SUBSTRATES WITH DIFFUSION BARRIERS (GRIDDED SUBSTRATES) This protocol lays out the general principles for the preparation of planar glass substrates with micro-fabricated diffusion barriers, such as micron-scale metal lines. Thin metal lines (typically 5 nm in height) on the glass surface can serve as physical barriers to restrict membrane components from diffusing while the rest of the lipid membranes maintain two-dimensional fluidity (Groves et al., 1997). Photolithography and thin film depositions, adapted from integrated circuit chip fabrication processes, are employed to create micron-size patterns on glass substrates (Jaeger, 2002; see Fig. 3). Various photoresist materials can be substituted to achieve similar results. Refer to documentation from individual manufacturers for standard operating procedures when using such materials. Other lithography techniques that are not described in this protocol can also be used for more advanced applications. For example, electron-beam (E-beam) lithography and scanning-probe lithography provide a line width resolution in the range of 50 to 100 nm. Nano-imprint lithography, based on direct-printing of prefabricated imprint masters, is generally used to fabricate nano-patterned substrates in a high-throughput manner. It is required to perform procedures described in this protocol in a cleanroom that is class 100 or above. Always consult safety personnel for proper personal protective equipments and operating procedures.
Materials Hexamethyldisilazane (HMDS; ACS grade or higher) Photoresist (S1805 positive g-line photoresist, Microchem) Photoresist developer (MicroDev; Microchem) Photoresist stripper (acetone, ACS grade or higher) Deionized water (DI water; resistivity ≥18 M and apparent pH 5.5) Nitrogen gas Isopropanol (ACS grade or higher)
Supported Membrane Formation, Characterization, Functionalization and Patterning
General cleanroom (class 100 or above) Spin coater (Laurell Technologies) 90◦ C hot plate Mask aligner (NXQ 4006, Neutronix-Quintel) Photomask (design of the mask, usually 1- to 3-μm features, can be generated by L-edit Pro; Tanner EDA; quartz masks with the designed feature, such as parallel-line grid patterns, are then manufactured by a mask-making vendor) Metal target (99.99% chromium, Alfa Aesar)
248 Volume 2
Current Protocols in Chemical Biology
Figure 3 Schematic depicting the steps in the preparation of glass substrates with microfabricated thin metal lines (left) and curvatures (right). Bottom-most two images are taken using a scanning electron microscope. Image on the left shows a substrate with parallel metal lines (700-nm wide). Image on the right shows the cross-section of a curved substrate. Scale bars are 6 μm.
Thin film evaporator (Edwards EB3 electron beam evaporator; Edwards) Additional reagents and equipment for cleaning substrates using piranha etching (Basic Protocol 2) and cleaning the patterned substrates using air/oxygen plasma cleaning or UV/ozone cleaning (Alternate Protocol 4 and 5) Perform substrate fabrication in a cleanroom (class 100 and above) 1. Clean bare SiO2 substrates using piranha etching (see Basic Protocol 2). 2. Apply the adhesion promoter HMDS by incubating the cleaned SiO2 substrates in the enclosed chamber with HMDS vapor for 10 min (vapor coating). Avoid inhalation of HMDS.
3. Use a spin coater (4500 rpm, 30 sec) to deposit a 500-nm-thick layer of S1805 photoresist. Use a hot plate (90◦ C, 10 min) to dry the resist solvent. Photoresist is sensitive to UV light. Keep it and resist-coated substrates in the dark or yellow/red light cleanroom.
Volume 2
249 Current Protocols in Chemical Biology
Volume 2
4. Use a mask aligner and shine UV (27.3 mJ/cm2 of 365-nm UV light) light onto resist-coated substrates through a photomask by contact printing. UV exposure photochemically breaks chemical bonds of positive photoresist and results in transferring mask patterns onto resist-coated substrates.
5. Develop the exposed substrates by immersing in a 1:1 mixture of MicroDev and deionized water for 35 sec. Then, rinse three times, each time with 200 ml deionized water and dry the substrates with nitrogen gas. MicroDev, an alkaline-based developer, selectively dissolves exposed photoresist, and the unexposed resist remains on the SiO2 substrate as a protecting mask for metal deposition in the following steps.
6. Deposit a layer of chromium film using the Edwards EB3. Monitor the thickness of the metal film by quartz crystal microbalance. In most instances, 5- to 10-nm-thick metal films are sufficient. Deposition of thin layers of metal film requires slow evaporation rates (around 0.5 nm per sec) and good vacuum conditions (10−6 Torr).
7. Lift off the metal-covered resist by immersing in 100 ml photoresist stripper; then rinse once with 100 ml isopropanol and then twice with 500 ml deionized water. Dry the substrates with nitrogen gas. 8. Clean the patterned substrates using air/oxygen plasma cleaning or UV/ozone cleaning (Alternate Protocols 4 and 5). Using Piranha or base etching to clean patterned substrates sometimes results in destruction of micro-fabricated patterns. Adjust the etching time if a solution-based cleaning method is preferred. Patterned substrates may be reused as long as the integrity of the patterns is preserved. Used substrates may also be cleaned by overnight incubation in 10% sodium dodecyl sulfate (SDS). SUPPORT PROTOCOL 2
PREPARING SUBSTRATES WITH CURVATURE MODULATION This protocol describes a method to engineer glass substrates with three-dimensional (3-D) surface topography by the use of microfabrication techniques. Geometrical perturbation of supported lipid membranes and adhered plasma membranes in intermembrane junctions can be directly controlled by prefabricated 3-D structures on glass substrates. Various curvatures on glass surfaces can be produced by a combination of anisotropic and isotropic SiO2 etching (Fig. 3). It is required to perform procedures described in this protocol in a cleanroom that is class 100 or higher. Always consult safety personnel for proper personnel protective equipment and operating procedures.
Materials
Supported Membrane Formation, Characterization, Functionalization and Patterning
Hexamethyldisilazane (HMDS; ACS grade or higher) Photoresist (S1805 positive g-line photoresist; Microchip) Photoresist developer (MicroDev, Microchem) Distilled water (DI water; resistivity ≥18 M and apparent pH 5.5) Nitrogen gas Piranha solution (see Basic Protocol 2) 5:1 buffered hydrofluoric acid (buffered oxide etch, 5:1 CMOS; Mallinckrodt Baker) Photoresist stripper (Acetone, ACS grade or higher) General cleanroom facility (class 100 or higher) Spin coater Hotplate
250 Volume 2
Current Protocols in Chemical Biology
Mask aligner (NXQ 4006, Neutronix-Quintel) Photomask (design of the mask, usually 1 to 3 micron features, can be generated by L-edit Pro; Tanner EDA; quartz masks with the designed feature, such as parallel lines, are then manufactured by a mask-making vendor) Plasma etcher for SiO2 anisotropic etching (AutoEtch 590, Lam Research) Additional reagents and equipment for cleaning the substrates using piranha etching and preparing piranha solution (Basic Protocol 2) Perform substrate fabrication in a cleanroom 1. Clean the bare SiO2 substrates using piranha etching (Basic Protocol 2). 2. Apply the adhesion promoter HMDS by incubating the cleaned SiO2 substrates in the enclosed chamber with HMDS vapor for 10 min (vapor coating). Avoid inhalation of HMDS.
3. Use a spin coater (4500 rpm, 30 sec) to deposit a 500 nm-thick S1805 photoresist. Use a hotplate (90◦ C, 10 min) to dry the resist solvent. Photoresist is sensitive to UV light. Keep it and the resist-coated substrates in the dark or yellow/red light cleanroom.
4. Use a mask aligner and shine UV (27.3 mJ/cm2 of 436 nm UV light) light onto the resist-coated substrates through the photomask by contact printing. UV exposure photochemically breaks chemical bonds of positive photoresist and results in transferring mask patterns onto resist-coated substrates.
5. Develop the exposed substrates by immersing in a 1:1 mixture of MicroDev and deionized water for 35 sec. Then, rinse three times, each time with 100 ml deionized water and dry the substrates with nitrogen gas. MicroDev, an alkaline-based developer, selectively dissolves exposed photoresist, and the unexposed resist remains on the SiO2 substrate as a protecting mask for anisotropic etching in the following steps.
6. Using a hotplate, bake (120◦ C, 30 min) the resist-coated substrates. 7. Perform anisotropic dry etching of SiO2 using the AutoEtch Plasma Etch System. It takes ∼30 sec to etch 300-nm deep SiO2 . Anisotropic dry etching selectively etches the areas without photoresist protection and creates 90◦ -like trenches over SiO2 substrates. AutoEtch 590 is operated under 2.8 Torr gas pressure, high RF power (850 W), and small electrode gap (0.38 cm). Etch conditions: CH4 90 SCCM, CHF4 30 SCCM, helium 120 SCCM (SCCM: standard cubic centimeters per minute). Etching rate is ∼600 nm/min. A longer etching time will result in serious photoresist damage. Avoid exposure to plasmagenerated UV light.
8. Remove the photoresist using freshly prepared piranha for 10 min. Use piranha with extreme caution. Please refer to Basic Protocol 2 for piranha cleaning and safety recommendations.
9. Use isotropic wet etching of SiO2 by using 5:1 buffered hydrofluoric acid to shape 90◦ trenches into trenches with rounded edges. Typically, a 3-min wet etching of a 1-μm-wide trench created by dry etching conditions above results in rounded edges with a radius of curvature ∼250 nm. Buffered hydrofluoric acid is made by mixing 5 parts 40% ammonium fluoride and 1 part 49% hydrofluoric acid, by volume. Handle hydrofluoric acid with extreme caution. In case of contact with skin, apply calcium gluconate immediately to affected areas to prevent serious damage and call for emergency medical help.
Volume 2
251 Current Protocols in Chemical Biology
Volume 2
10. Rinse three times, each time with 100 ml deionized water. Dry the substrates with nitrogen gas. 11. Clean the patterned SiO2 substrates with either piranha cleaning, air/oxygen plasma cleaning, or UV/ozone cleaning (Basic Protocol 2, Alternate Protocols 4 and 5).
BASIC PROTOCOL 3
Supported Membrane Formation, Characterization, Functionalization and Patterning
FORMATION AND FUNCTIONALIZATION OF SUPPORTED MEMBRANES This protocol describes basic steps in the formation of supported membranes by vesicle fusion, followed by functionalization of the membrane. Vesicle fusion and membrane formation occurs spontaneously when a suitable substrate is exposed to an SUV suspension. The current viewpoint is that supported membrane formation involves the following steps: vesicles first adsorb to the surface of the substrate. Subsequently, the intact vesicles may rupture or fuse with other vesicles before they rupture, but either way, lipid bilayer disks are formed (Fig. 4). Lipid disks then enlarge by merging with one another to form a continuous membrane supported by the underlying solid substrate. It is suggested that vesicle rupture can be induced by high vesicle density (critical vesicular coverage), active bilayer edges, and by the presence of cations (Cremer and Boxer, 1999; Keller et al., 2000; Reviakine and Brisson, 2000; Johnson et al., 2002; Richter et al., 2003; Anderson et al., 2009; Seantier and Kasemo, 2009; Weirich et al., 2010). Note that vesicles can rupture and expose either the inner or the outer leaflets of their membranes. Experiments using proteoliposomes containing transmembrane proteins indicate that during vesicle fusion, the outer leaflet of the vesicles faces towards the solution and the inner leaflet towards the substrate (Contino et al., 1994; Salafsky et al., 1996). On the other hand, studies using SUVs without proteins suggest that the inner leaflets of SUVs will become the distal leaflet of the supported membrane, which faces the solution, after vesicle fusion (Jass et al., 2000; Blanchette et al., 2006). Although it is expected that molecular interactions between lipids and the solid support play a critical role in the process of vesicle rupture and lipid deposition on solid supports (Richter et al., 2006), fundamental mechanisms involved in the formation of supported membrane remain to be explored. Functionalization of supported membranes is achieved by proper incubation of the protein-of-interest after supported membrane formation, and several different methods of functionalization are listed here for the user’s selection.
Figure 4 Diagrammatic representation of the steps involved in supported membrane formation through vesicle fusion. (A) Initial vesicle adsorption to surface of substrates. (B) Fusion of adsorbed vesicles to form larger vesicles (if the initial vesicle is not large enough to rupture by itself) and flattening of adsorbed vesicles. (C) Rupture of vesicles resulting in bilayer discs on the surface, and finally coalescence of bilayer discs to form a continuous two-dimensional supported membrane.
252 Volume 2
Current Protocols in Chemical Biology
Table 4 Incubation Times and Instructions for Protein Functionalization
Protein concentration
Incubation time Special instructions and notes (at 24◦ C)
17 nM (1 μg/ml) streptavidin
45 min
• This method requires two incubation steps; the first incubation links streptavidin to the supported membrane and the second incubation links the target protein to streptavidin on the membrane surface.
50 nM biotinylated protein
45 min
• Rinsing is necessary after the first incubation to remove excess streptavidin. The surface density of the target protein can be controlled by varying the concentration of biotinylated lipid (0.01% -1%) used to prepare the supported membrane.
Ni2+ NTA lipidHis10 (or more)
150 nM-1 μM
5 min to 2 hr
• Rinse away excess protein after the incubation and let the sample rest for 20 min. • Perform a second rinse to remove proteins that are desorbed from the surface. • Samples can be brought to a higher temperature (e.g., from room temperature to 37◦ C) after the second rinse. • Avoid a temperature drop, as it results in desorption of the protein. • The final surface density of the target protein can be controlled by varying the concentration of the protein solution and incubation time.
Maleimide-cysteine residue
10-100 μM
10 min to 1 hr
• SUV suspension containing maleimide lipids needs to be prepared on the same day as protein functionalization because the functional group of maleimide hydrolyzes in an aqueous environment and loses its activity. • Protect samples containing maleimide lipids from light, as UV light promotes photochemical reactions in its functional group, α,β-unsaturated carbonyl, causing it to lose activity. • Usually, proteins containing free cysteines are stored with a reducing agent, such as 2-mercaptoethanol, dithiothreitol (DTT) or tris(2-carboxyethyl)phosphine (TCEP), to prevent the formation of disulfide bonds between proteins. Such protein solutions must be desalted immediately prior to the functionalization of the supported membrane. • The unreacted or excess maleimide can be terminated with a free thiol, such as 5 mM 2-mercaptoethanol (2-ME). A 10-min incubation is sufficient. • The surface density of the target proteins can be controlled by varying the incubation time.
Lipidated protein
1-20 μg/ml
12-16 hr
• The final surface density of the target proteins can be controlled by varying the concentration of the protein solution and incubation time.
Linkage type Biotinylated lipid– streptavidinbiotinylated protein
Volume 2
253 Current Protocols in Chemical Biology
Volume 2
Materials Nitrogen gas (industrial grade or higher; if from a central supply, use a hydrophobic filter to remove oils from the gas) Clean substrates (Basic Protocol 2, Alternate Protocols 3, 4, and 5) SUV suspension (Basic Protocol 1, Alternate Protocols 1 and 2) Spreading buffer (2× PBS or 2× TBS; see recipes; see Commentary for selection) Working buffer: deionized water or TBS (see recipe) or phosphate-buffered saline (PBS; see recipe) Blocking solution: 5 mg/ml casein in PBS (see recipe for PBS) or 0.01% (w/v) bovine serum albumin (BSA) in PBS (see recipe for PBS) Aluminum foil, optional 1. Use a stream of nitrogen gas to dry the substrate, if it has been stored in deionized water. 2. If SUVs were made in deionized water, prepare the spreading solution by combining equal volumes of the SUV suspension and the spreading buffer. Otherwise, if SUV suspensions were prepared in buffer, the suspension is sufficient to act as a spreading solution in the next step.
3. Quickly bring the substrates in contact with the SUV-containing spreading solution and incubate 10 to 15 min at room temperature. This can be achieved by flowing-in the spreading solution using a flow-cell or by inverting the substrate over a small volume (tens of microliters) of the spreading solution and letting the capillary action spread the solution over the substrate surface. Alternatively, simply add the spreading solution on the substrate (for example, on one well of a multi-well glassbottom plate). The supported membrane will form spontaneously during the incubation. Care should be taken to ensure that the substrate remains hydrated at all times from this point on, as supported membranes will be destroyed when they come into contact with air.
4. Rinse away excess vesicles with 20 ml working buffer. The working buffer is usually the same as the rehydration solution used to prepare the SUV suspension. Buffer exchange can be done after this rinsing step; however, it should be carried out slowly and gently. Avoid sudden osmotic shock.
5. Incubate the supported membrane with blocking solution for 20 min at room temperature. After incubation, rinse the blocking solution away with 20 ml working buffer. Casein and BSA are commonly used to block defects in supported membranes; thus, target proteins will not bind to the surface nonspecifically during functionalization. Protein concentration of the blocking solution should be decreased if these proteins cause defects in the supported membrane.
6. Flow-in (or add) the target protein solution to the supported membrane. Consult Table 4 for incubation times and specific instructions. If fluorescent molecules and/or maleimide-lipids are used, protect the samples from exposure to light by covering the samples with aluminum foil. 7. Keep fluorescent samples in the dark. The storage condition depends on the protein of interest. In some cases, a functionalized supported membrane can be stored up to 2 to 3 days at 4◦ C. SUPPORT PROTOCOL 3
FORMATION OF SUPPORTED MEMBRANES ON SILICA BEADS This protocol uses micron-sized silica beads as an example to describe the formation of supported lipid bilayers on micro-particles, rather than a planar substrate. The protocol can be applied to any particle that has a silicon oxide (SiO2 ) surface, e.g., 200-nm
254 Volume 2
Current Protocols in Chemical Biology
silica-coated silver nano-cubes. This type of supported membrane has been used for studying a wide range of scientific problems including multi-particle interactions (Gomez et al., 2009), protein oligomerization (Dean et al., 2003), and cell membrane interactions (Tanaka et al., 2009). Compared to planar-supported membranes, particle-supported membranes provide larger surface areas and can significantly increase detection efficiency when used in spectroscopic techniques, such as nuclear magnetic resonance (NMR; Bayerl and Bloom, 1990) and surface plasmon resonance (SPR; Galush et al., 2009).
Materials Spreading buffer: 2× TBS or 2× PBS (see recipes) SUV suspension (Basic Protocol 1, Alternate Protocols 1 and 2) Silica beads at 10 wt% solids in deionized water (Bangs Laboratories) Working buffer: deionized water or TBS (see recipe) or phosphate-buffered saline (PBS; see recipe) 1.5-ml microcentrifuge tubes Benchtop vortex mixer Mini centrifuge (max g force ∼2000 × g) 1. If SUVs were made in deionized water, prepare the spreading solution by combining equal volumes of the SUV suspension and the spreading buffer. Otherwise, if SUV suspensions were prepared in buffer, the suspension is sufficient to act as a spreading solution in the next step.
2. Combine equal volumes of spreading solution with bead stock solution in a 1.5-ml microcentrifuge tube. 3. Pulse vortex the mixture several times and then let the mixture rest at room temperature for 10 min. 4. Use a mini centrifuge to pellet the particles in the mixture by centrifuging briefly at ∼2000 × g, room temperature. Remove the supernatant. The purpose of this step is to remove excess vesicles. The centrifugation time should be as short as possible to avoid particle aggregation and destruction of supported membranes. In general, larger particles require shorter spins. For example, 5 sec of centrifugation is sufficient for 5-μm silica beads while 30 sec of spinning is required for 200-nm beads.
5. Resuspend the pellet in 1 ml working buffer by vortexing in brief pulses. The working buffer is the same as the solution used to make the SUV suspension.
6. Repeat steps 4 and 5 two more times to ensure removal of excess vesicles. Lipid-membrane-coated particles are now ready for use.
7. Use steps 3, 4, and 5 for buffer exchange and surface functionalization for protein of interest. Consult Table 4 for specific instructions on surface functionalization.
FORMATION OF SUPPORTED INTERMEMBRANE JUNCTIONS This protocol details procedures for the formation of supported intermembrane junctions (Fig. 5) by rupture of giant unilamellar vesicles (GUVs) onto preformed supported membranes. GUVs are prepared based on the protocol described in Akashi et al. (1996), but with significant simplifications. See Manley and Gordon (2008) for a more sophisticated protocol that has a higher yield of GUVs. GUVs, formed in a sucrose solution, when deposited onto supported membranes initially sink because of the higher density of the sucrose solution. Strong electrostatic and van der Waals interactions then further drive the vesicles toward the supported membrane, often causing them to rupture and deposit an upper bilayer patch tens of micrometers in size. This system has emerged as a very good
SUPPORT PROTOCOL 4
Volume 2
255 Current Protocols in Chemical Biology
Volume 2
Figure 5 The bottom portion of the figure shows an intermembrane junction (schematic, not to scale). Two lipid membranes, one of which is supported on a glass substrate, adhere. Adhesion drives mobile proteins (antibodies) into patterns of high- and low-density zones. The upper portion of the figure shows a fluorescence image of proteins (20 μm by 5 μm); top view relative to the side view of the schematic). Adapted from Parthasarathy et al., 2005.
model system to perform precise analysis of processes occurring at the intermembrane junctions (Wong and Groves, 2002; Kaizuka and Groves, 2004, 2006; Groves, 2005). It has also been used to study the mechanical and electrostatic aspects of protein organization at membrane junctions (Parthasarathy and Groves, 2004; Parthasarathy et al., 2005). In addition, it has been shown that on a glass-supported system, the double membrane scheme is required for studies of natural miscibility phase separation dynamics (Kaizuka and Groves, 2004; Parthasarathy et al., 2006).
Materials 0.5 M sucrose, warmed to 50◦ C Dry lipid film in a round-bottomed flask (Basic Protocol 1, steps 1 to 6) Supported membrane (see Basic Protocol 3) Working buffer: DI water or TBS (see recipe) or PBS (see recipe) Oven with temperature controls or a large water bath at 50◦ C 10-ml syringes Microcentrifuge tubes Prepare the GUVs 1. Slowly and gently add 50◦ C 0.5 M sucrose solution into the flask containing the dry lipid film to a final lipid concentration of 0.5 to 1 mg/ml. Make sure that the liquid level is high enough to cover the entire lipid film. The lipid mixture should include trace amounts of charged lipids, like 0.1 mol% of Texas Red-DHPE or DOPS, if higher yields are desired.
2. Immediately, place the flask into the oven or the water bath at 50◦ C. Incubate for 5 min then let the flask cool to room temperature slowly over the period of ∼10 hr. Avoid disturbance during the process.
Supported Membrane Formation, Characterization, Functionalization and Patterning
During the incubation, the lipid film will gradually lift off the glass surface. The charged species in the lipid mixture helps to increase the yield of unilamellar vesicles by introducing electrostatic repulsion between lipid layers. In a successful GUV preparation, an almost transparent and bulky white cloud will float in the middle of the solution.
3. Collect the GUV-containing cloud from the center of the flask using a 10-ml syringe. Store the contents in a microcentrifuge tube.
256 Volume 2
Current Protocols in Chemical Biology
Form the second supported membrane 4. Pipet a few microliters of the GUV-sucrose suspension directly onto the supported membrane, which is kept under working buffer. Because of the difference in densities, GUVs will settle toward the supported membrane and rupture on contact. It is possible to further enhance rupturing by adding more salt to the solution (in ∼20-mM increments).
5. Rinse away excess GUVs gently with 10 ml of working buffer. GUVs can be stored at room temperature for up to 3 weeks.
CHARACTERIZING SUPPORTED MEMBRANES The first half of this protocol describes a general method to verify the fluidity of supported membranes by fluorescence recovery after photobleaching (FRAP). FRAP, first introduced by Axelrod et al. (1976), is the method of choice to characterize two-dimensional lateral diffusion in a thin film containing fluorescent probes. In FRAP, a small spot of the fluorescent sample is quickly photobleached by a short exposure to intense light. The fluorescence recovery, due to the replenishment of bleached probes by the surrounding fluorophores, is monitored, and the recovery curve can be fitted to an analytical equation to determine the diffusion coefficient of the fluorophore (Axelrod et al., 1976; Soumpasis, 1983). Since many factors can affect the fluidity of functionalized supported membranes, FRAP should be used to ensure the quality of the samples prior to their use in experiments. The method described here applies to conventional fluorescence microscopes. Most confocal microscopes have fairly automated FRAP procedures. Users should refer to documentation from individual manufacturers when using such equipment.
BASIC PROTOCOL 4
The second half of the protocol details the steps to quantify surface density of fluorescent molecules on supported membranes (Galush et al., 2008). Supported membrane standards (DOPC/BODIPY-DHPE or DOPC/Texas Red-DHPE) are first used to obtain a calibration curve that relates observed intensity to fluorophore surface density. A scaling factor is then used to convert the fluorescence intensity of the molecule of interest to the corresponding fluorescence intensity of the standard bilayer. This converted intensity can readily be applied to the calibration curve to obtain the corresponding surface density. This method assumes a linear relationship between the fluorescence intensity and surface density and hence should only be used in the absence of self-quenching and fluorescence resonance energy transfer (FRET). Fluorescence correlation spectroscopy (FCS) can be used as an alternative for quantitatively characterizing the supported membrane. The technique requires a basic confocal microscope setup and the technical details can be found elsewhere (Chiantia et al., 2009; Vamosi et al., 2009).
Materials Fluorescent samples (i.e., supported membrane with or without protein functionalization; see Basic Protocol 3) Standard supported membranes with known fluorophore concentrations (e.g., DOPC bilayers containing either BODIPY-FL-DHPE or Texas Red-DHPE) Fluorescence microscope equipped with: Filter sets that match the excitation and emission spectrum of the fluorophore used Light source (typically a mercury lamp or a xenon lamp) Adjustable field diaphragm 60× or higher objective lens CCD camera
Volume 2
257 Current Protocols in Chemical Biology
Volume 2
Figure 6 Fluorescence recovery after photobleaching (FRAP). (A) An image of a supported membrane prior to photobleaching. The field diaphragm is adjusted to the size of the bleached spot. (B) Image of the supported membrane immediately after 10 sec of photobleaching. The field diaphragm can be seen fully opened for image acquisition. (C) Image of the supported membrane after recovery. (D) The recovery curve based on intensity measurements within the bleached spot. Fluorescent intensity at each time point is normalized to the initial intensity after background subtraction. Both initial and background intensities were obtained from (A). The half-time of the recovery is 52 sec and the radius of the bleached spot is 27 μm; thus, the diffusion coefficient of the sample was estimated to be 3.1 μm2 /sec.
Characterize the membrane fluidity using FRAP 1. Mount the fluorescent sample onto the microscope stage and, with proper illumination (i.e., correct filter sets), bring the surface to focus. Photobleaching is likely to occur during this process, hence move to a new area during actual measurement.
2. Adjust the field diaphragm to the desired spot size, which will be the size of the bleached spot. Attenuate the light source with neutral density filters and take a picture. The average intensity in the area enclosed by the diaphragm will be used as initial intensity while the intensity outside of the diaphragm will be used as background (Fig. 6A). The size of the bleached spot can range from 20- to 200-μm in diameter. Typically though, aim for the minimum size that can be achieved. Attenuate the light source, such that bleaching from the observation light is <10% of the total recovery values during the course of the recording. To determine the effect of the observation light, perform a recording without the photobleaching step.
3. Remove all neutral density filters from the light path. Expose the sample to the un-attenuated light source for 5 to 15 sec to bleach the spot. For accurate measurements, the bleaching time should be <10% of the half-time of full recovery.
4. Quickly attenuate the light source with the same set of neutral density filters, open the field diaphragm, and start to record the recovery (Fig. 6B,C). 5. Analyze the recorded images and obtain the average intensity of the bleached spot at different time points. 6. Subtract the background from each average intensity value and then normalize it to the initial intensity. Supported Membrane Formation, Characterization, Functionalization and Patterning
7. Plot the normalized intensities as a function of time. Fit the time trace (i.e., the recovery curve) with a single exponential function to obtain the half-time of the recovery t1/2 (Fig. 6D). The diffusion coefficient can be estimated using D = 0.224 R2 /t1/2 , where R is the radius of the bleach spot (Axelrod et al., 1976).
258 Volume 2
Current Protocols in Chemical Biology
Table 5 Surface Densities of Fluorescent Lipids in Standard Lipid Bilayers
Conc. of fluorescent lipid (mol%) 2
Theoretical surface density (per μm )
0.01%
0.05%
0.1%
0.25%
0.5%
278
1359
2778
6944
13889
Alternatively, a Matlab code written by Dr. Raghuveer Parthasarathy at the University of Oregon can be used to estimate the diffusion coefficient. The program analyzes a pair of images with known time step and applies Fick’s law to obtain the diffusion coefficient. This code can be found at http://www.currentprotocols.com/protocol/ch100131.
Perform quantitative fluorescence measurement 8. Acquire a fluorescent image of the sample under proper illumination settings (i.e., correct filter sets, neutral density filter, and exposure time), making sure the image is not saturated. 9. Use the same illumination settings to acquire images from each standard supported membrane. In general, use BODIPY-FL-DHPE if the molecule of interest emits in the green range and Texas Red-DHPE if it emits in the red range. See Table 5 for the surface densities of various standard membrane (assuming that the footprint of a DOPC lipid is 0.72 nm2 ).
10. Calculate the average fluorescent intensity from each image of standard supported membrane. Subtract the background and plot the net intensity as a function of surface density. Fit the data to a linear function, which will serve as the calibration curve for the experiment. As described in Galush et al. (2008), the intensity of standard is linearly proportional to the fluorophore surface density as long as the fluorescent lipid concentration is below 0.5 mol%.
11. Subtract background from the image of the sample then divide the net intensity by the scaling factor F (see Support Protocol 5). Use the calibration curve obtained from the previous step to translate these calibrated intensities into surface densities. It is recommended to only use this method if the calibrated intensities from the samples are within the linear range of the standards. Since self-quenching and/or FRET is likely to occur at high surface density, the fluorescence intensity may not scale linearly to the surface density beyond the linear range of the standards.
MEASURING THE SCALING FACTOR This protocol lists the basic steps for determining the scaling factor F to correlate between the fluorophore of interest and the standard fluorophore. This factor accounts for the differences in the optical properties of the instrument and the spectral difference between the molecule of interest and the standard fluorophore. It has been shown that the same fluorophore can behave very differently when conjugated to different chemical moieties (Galush et al., 2008). For example, AlexaFluor488-antibody was found to be three times brighter than AlexaFluor488-streptavidin. Thus, it is essential to measure the scaling factor for each sample-standard pair during quantitative fluorescence measurement.
SUPPORT PROTOCOL 5
Materials Standard SUV suspensions with known fluorescent lipid molarity (see Basic Protocol 1 for SUV suspensions and Basic Protocol 4 for selection of standard fluorophore) Fluorescent protein solutions with known protein concentrations Buffer (same as that used to make the fluorescent protein solution) Volume 2
259 Current Protocols in Chemical Biology
Volume 2
Fluorescence microscope equipped with: Filter sets that match the excitation and emission spectrum of the fluorophore used Light source (typically a mercury lamp or a xenon lamp) Adjustable field diaphragm 20× objective lens CCD camera 96-well plate (non-glass bottom) 1. Dispense all standard and protein solutions into individual wells of the 96-well plate. Avoid glass surfaces as supported lipid bilayers may form from the SUV suspensions and alter the concentration of the solutions.
2. Adjust the field diaphragm to the minimum size and bring the focus deep into the solution. The intensity should be maximal and remain constant even when the vertical focus is adjusted.
3. Acquire fluorescence images for every well under identical illumination settings. 4. For each image, measure the average intensity of a small region at the center of the field. The intensity will be uneven across the field of view due to fluorescence entering the objective from outside the focal plane, hence only the intensity from the center should be measured. As a rule, the region should not be larger than half the size of the field diaphragm.
5. Subtract background (intensity of buffer) and plot the net intensities as a function of molarity. The result should be two straight lines that intercept through the origin.
6. Fit the values to two straight lines and obtain slopes for each line. The scaling factor F = (slope of fluorescent protein)/(slope of standard SUV).
REAGENTS AND SOLUTIONS Use Milli-Q purified water or equivalent in all recipes and protocol steps.
Phosphate-buffered saline (PBS), pH 7.4 137 mM NaCl 2.7 mM KCl 10 mM Na2 HPO4 2 mM KH2 PO4 Tris-buffered saline (TBS) 19.98 mM Tris·Cl, pH 7.4 136 mM NaCl COMMENTARY Background Information Supported Membrane Formation, Characterization, Functionalization and Patterning
Two-dimensional assemblies of phospholipid bilayer membranes supported on solid substrates are referred to as supported membranes. They were first described nearly 25 years ago (Tamm and McConnell, 1985) and have been explored as bio-mimetic analogues of cell membranes (Sackmann, 1996).
They provide a controllable and mechanically stable substrate for experimentation using optical microscopy, surface-sensitive spectroscopy (Groves et al., 2008), and high resolution scanning probe techniques such as atomic force microscopy (Goksu et al., 2009). They are also highly compatible with common environmental control chambers and micro-fluidic
260 Volume 2
Current Protocols in Chemical Biology
systems (Castellana and Cremer, 2006). Importantly, they can be modified to present proteins on their surfaces while maintaining the hallmark two-dimensional fluidity of cellular membranes. Thus, supported membranes serve as powerful experimental platforms for quantitatively dissecting the way the plasma membrane environment is coupled to its functions (Mossman et al., 2005; Hartman et al., 2009; Salaita et al., 2010). Despite the ability to easily manipulate the physical characteristics of supported membranes, the number of studies that use these platforms has been limited. This is primarily due to two major challenges. First, it is not trivial to prepare good-quality supported membranes with long-term stability, especially under physiological conditions. The stability and fluidity of supported membranes are governed by a fine balance maintained among the attractive van der Waals force, the repulsive hydration force, and electrostatic interactions between the lipids and the solid support (Keller et al., 2000; Johnson et al., 2002; Richter et al., 2003). Thus, the optimal conditions for a goodquality supported membrane strongly depends on lipid composition, ionic strength of solution, surface charge density of the support, and the temperature of formation. Second, incorporating recombinant proteins into supported membranes with a specific orientation and a desired density, while maintaining lateral fluidity of the supported membrane, is difficult. For example, the traditional way of using detergents to incorporate amphipathic proteins into lipid bilayers suffers from an inability to control protein orientation and a lack of reproducible incorporation efficiency (Rigaud and Levy, 2003). One of the most successful experiments utilized glycosylphosphatidylinositol (GPI) anchors to incorporate protein into proteoliposomes. Supported membranes were then formed by vesicle fusion using a mixture of the proteoliposomes and unmodified vesicles (Grakoui et al., 1999). However, proteoliposomes generally do not form supported membranes, as well as unmodified vesicles, which is likely due to the proteins interfering with the formation of supported membranes. The resulting supported membranes usually suffer from high defect density, as well as many intact unruptured vesicles. More recently, methods that decouple supported membrane formation and protein functionalization have been developed. Highquality functionalized supported membranes containing proteins with desired orientation and surface density can now routinely be pre-
pared. The common strategy in these methods is inclusion of lipids containing functional groups in the formation of supported membranes and then subsequently conjugating it to a soluble protein without the need for detergents. For example, bilayers containing biotinylated lipids bind strongly to avidin and its modified forms, streptavidin and neutravidin (Chaiet and Wolf, 1964; Hendrickson et al., 1989). Due to their tetrameric nature (i.e., each avidin has four biotin-binding sites), the surface-bound avidin can then be used to attach biotinylated protein on top of the supported membrane. However, using such tetrameric architecture has the inherent risk of introducing artificial clusters. Another useful method takes advantage of the simple and widely used polyhistidine/Ni2+ -NTA interaction to tether proteins to bilayers containing nickel-chelating lipids. This binding is reversible, but it has been reported that stable protein-functionalized surfaces (for tens of hours) with a wide range of protein densities can be obtained with a single concentration of chelator lipid by optimizing incubation conditions (Nye and Groves, 2008). Maleimideheadgroup lipids have also been used to covalently attach proteins onto supported membranes (Gureasko et al., 2008). This method utilizes the same basic chemistry commonly used for the fluorescent modification of proteins, i.e., formation of a carbon-sulfur bond between maleimide and cysteine, and requires only an accessible unoxidized cysteine residue on the protein of interest. Besides using lipids with functional headgroups, it is possible, by simple incubation, to insert molecules containing hydrophobic anchors into supported membranes (Pfeiffer and Hook, 2004; Grogan et al., 2005; Paulick et al., 2007; Giocondi et al., 2008). However, designing proper hydrophobic anchors usually involves significant knowledge of synthetic pathways in organic chemistry and may not be trivial for non-chemists. It is well known that lipids are not able to form continuous bilayer structures over unmodified metal surfaces. This property has been utilized to partition supported membranes into micrometer-scale arrays of isolated fluid membrane corrals by forming supported membranes on glass substrates that are patterned with metal grid lines (Groves et al., 1997). Patterned supported membranes have been proven to be very useful and provide a unique experimental platform to investigate a wide range of biophysical and cellular biology questions (Mossman et al., 2005; Groves, 2006; DeMond et al., 2008;
Volume 2
261 Current Protocols in Chemical Biology
Volume 2
Hartman et al., 2009; Manz and Groves, 2010; Salaita et al., 2010; Yu and Groves, 2010). In addition to spatial patterning, the ability to control membrane geometry by prefabricated 3-D structures on glass substrates provides new possibilities to investigate curvature-dependent processes in molecular sorting, such as cholesterol-dependent phase separation (Parthasarathy et al., 2006). Imposing designated curvature onto plasma membrane in live cells also creates a unique mechanical perturbation and allows mechanobiological studies (Groves, 2007; Yu and Groves, 2010).
Critical Parameters
Supported Membrane Formation, Characterization, Functionalization and Patterning
Temperature and freshness of lipids It is well known that lipid bilayers undergo various phase transitions as a function of temperature. The most important of these transitions is the so-called main or melting temperature, Tm . Bilayers made from one type of lipid exist in fluid-phase when the temperature is above the Tm and in gel phase, where lipids exhibit little translational freedom, when the temperature is below the Tm . For a lipid mixture, the possibility of phase separation exists if the temperature is between Tm s of individual lipid species. In addition, formation of supported membrane only occurs when vesicles are in fluid state. Therefore, it is important to ensure that the working temperature is above Tm s of all lipid species during the formation of supported membranes. Two unsaturated lipids, DOPC (Tm = −20◦ C) and POPC (Tm = −2◦ C), and two saturated lipids, DLPC (Tm = −1◦ C) and DMPC (Tm = 23◦ C), are commonly used as the basic building blocks for supported membranes. Working temperatures for mixtures using these lipids as main building blocks often range from 40◦ C to 50◦ C. Another important factor in supported membrane formation is the freshness of the lipids. Oxidation and hydrolysis happen more easily with unsaturated lipids, and the resulting lyso-lipids (lipids with single fatty acid chain) and lipid fragments may alter the mixing behavior of the lipid mixture. They also have a tendency to create and stabilize defects on supported membranes. Hence, it is recommended to make fresh vesicles before each experiment and to keep the time the lipid is in direct contact with air/oxygen to a minimum to avoid oxidation. Commercially available lipids are usually shipped and stored in glass containers layered with argon or nitrogen at less than −20◦ C. As a general rule, saturated lipids can be stored
under such conditions for 1 year. For unsaturated lipids, the shelf life decreases by half for each C=C. For example, DOPC keeps well for only ∼3 months, since it has two carboncarbon double bonds. Importantly, once the lipid stock is exposed to air, it should be used within 1 month for best results. Glass substrate: Cleanliness and surface charge density One of the key parameters for successful supported membrane formation is the hydrophilicity of the glass substrate, which is directly related to the cleanliness of the substrate. It is strongly recommended to use substrates that can be cleaned independently from the imaging chamber, such as a flow cell that can be assembled over a cleaned cover glass. Forming supported membranes in commercially available glass-bottom multi-well plates is less successful than on a cover glass, which may be due to hard-to-clean adhesives used to attach the glass to the plate. The surface-charge density of the substrate is also a critical parameter in the formation and stability of supported membranes, since electrostatic interaction is one of the main forces between the substrate and the lipid bilayer. All the cleaning methods provided here make glass surfaces more negatively charged; therefore, it is important to keep the reaction/cleaning time constant for consistent results. Salt and buffers The interactions between the lipid bilayer and the glass substrate can be altered dramatically by the presence of salts, such as NaCl and KCl, in the solution. This is because these charged electrolytes screen the surface charge of both membrane and substrate and, as a result, alter the electrostatic interactions between the two surfaces. An excellent resource for bilayer/substrate interactions is Israelachvili (1992). As a general rule, for bilayers that contain <5 mol% negatively charged lipids, use deionized water to prepare the SUV suspension and then add the spreading buffer to reach ∼150 mM final salt concentration prior to vesicle fusion. For lipid mixtures that have a high negative charge, it is necessary to prepare the SUV suspension and perform vesicle fusion in the presence of salt. The ionic strength of common buffers, such as PBS (which contains ∼150 mM NaCl), is usually sufficient for the formation of stable supported membranes containing <15 mol% negatively charged lipids. The presence of large defects on the supported
262 Volume 2
Current Protocols in Chemical Biology
Figure 7 Effects of salt concentration on supported membrane formation. (A) A fluorescent image of supported membrane containing 8% DOPG and formed under DI water showing large defects (black holes). (B) Image showing vesicle budding (bright spots) in membrane formed with a salt solution (300 mM NaCl) of high ionic strength. (C) Ca2+ induces defects on a supported membrane containing 25% POPA. Defects appear upon addition of 0.5 mM CaCl2 and gradually increase in size with time. (D) and (E) Examples of diffusion coefficients of supported membranes containing various DOPG concentrations as measured by FRAP. The membranes were prepared in deionized water. For the samples containing large defects, as shown in (A), FRAP was selectively measured in the regions that were defect-free. (F) Diffusion coefficients of supported membranes containing various DOPG concentrations as measured by FCS. The membranes were formed in the presence of 150 mM NaCl. All supported membranes contained a small amount of Texas Red-DHPE for detection. Scale bars are 50 μm, unless specified otherwise.
membrane is usually an indication of low ionic strength during the formation of supported membranes (Fig. 7A). On the other hand, a high salt environment can result in vesicle budding (Fig. 7B). It is important to remember that the diffusivity and miscibility of lipids in a highly charged supported membrane may be affected by the lipid composition and cations in the solution (see Fig. 7 for examples). Tris(hydroxymethyl)aminomethane and phosphate are compatible with the formation of supported membranes on glass surfaces. Hence, typical TBS or PBS (Trisor phosphate-buffered saline, respectively) can be used as the rehydration solution, spreading buffer, or working buffer when required in the protocol. However, supported membranes formed using HEPES buffer tend to contain more defects and are less stable, often becoming non-fluid during the protein functionalization steps. For experiments that require HEPES buffer, it is better to use Tris or phosphate buffer for supported membrane formation and subsequently replace with HEPES by buffer exchange.
Troubleshooting Common problems are encountered during the course of these experiments. Their causes and solutions can be found in Table 6.
Anticipated Results Basic Protocol 1 and Alternate Protocols 1 and 2 will generate SUV suspensions for supported membrane formation. The suspension should be clear and free of any aggregates. Basic Protocol 2 and Alternate Protocols 3, 4, and 5 are used to prepare clean glass substrates. Clean substrates are extremely hydrophilic, so that water can easily spread and wet the whole surface. Support Protocols 1 and 2 can be used to generate substrates that are patterned with metal grid lines and 3-D structures, respectively. Basic Protocol 3 and Support Protocols 3 and 4 are helpful in making functionalized supported membranes. Basic Protocol 4 will be useful in making quantitative measurements of the fluidity and the surface density of fluorescent molecules on supported membranes. For a DOPC-based supported membrane, the diffusion coefficient
Volume 2
263 Current Protocols in Chemical Biology
Volume 2
Table 6 Troubleshooting
Problem
Possible causes
Solutions
SUV suspension does not reach clarity
Chloroform is not completely removed
Use N2 stream or vacuum to remove excess chloroform.
Salt concentration is too high in the Decrease salt concentration or use suspension, causing SUVs to fuse and form deionized water to rehydrate lipids. large vesicles
Poor fluidity prior to protein incubation
Large defects (holes)
Poor fluidity after protein incubation
Supported Membrane Formation, Characterization, Functionalization and Patterning
Lipids are too old
Use fresh lipids.
Old sonicator microtip
Polish the microtip or replace the microtip if the length of the microtip is too short (due to frequent polishing).
Glass substrate is not hydrophilic enough Clean the substrate better Use a different for SUVs to rupture and form a continuous method of cleaning. bilayer Low ionic strength may result in un-ruptured vesicles
Increase the salt concentration in the spreading buffer (can be as high as 4× TBS or 4× PBS) or use a salt solution to rehydrate lipids.
Lipid oxidation and lipid fragments may create and stabilize small defects (below optical resolution) on supported membranes
Use fresh lipids to make SUV suspension and use it immediately.
Protein in buffer
Ensure the buffer contains no proteins.
Salt concentration of rehydration solution is too low
Increase salt concentration of the solution.
Unnecessary or prolonged incubation of dried lipid film in the desiccator
Traditionally, this step was necessary to ensure complete removal of solvent. However, the high vacuum provided by the Rotavap should be sufficient. This step will only expose the lipids to oxidation.
Change in salt concentration during buffer exchange is too much
Avoid osmotic shock. Buffer exchange should be done gently and gradually.
Too many small defects in the supported Use solutions from previous problem, i.e., membrane. Proteins will stick to the defects poor fluidity prior to protein incubation, for and decrease the fluidity of the membrane troubleshooting. if the density of the defects is too high. Proteins stick to defects on the supported membranes. (Note: Lipid bilayer may still be fluid)
Use blocking solution prior to protein functionalization (Note: Keep the concentration of blocking solution low to avoid defect formation).
Proteins may be over-labeled with fluorophores. Most fluorophores are slightly hydrophobic hence, over-labeled proteins will have a higher tendency to disrupt the lipid bilayer by inserting themselves into the bilayer.
Decrease incubation times for protein labeling or incubate at 4◦ C.
continued
264 Volume 2
Current Protocols in Chemical Biology
Table 6 Troubleshooting, continued
Problem
Nonspecific binding of the protein of interest
Large, bright fluorescent spots
Fluorescent impurities (for single molecular experiments)
Possible causes
Solutions
Casein and BSA may cause and enlarge defects in supported membranes.
Decrease the concentration of proteins in blocking solution or avoid using it.
Proteins may be over-labeled with fluorophores.
Decrease incubation times for protein labeling or incubate at 4◦ C.
Proteins stick to defects on the supported membranes
Use blocking solution prior to protein functionalization (Note: Keep the concentration of blocking solution low to avoid defect formation).
SUVs used are too old thus do not rupture. Use fresh lipids to make SUV suspension Proteins will bind to these intact vesicles and use it immediately. during the functionalization. Fluorescent protein is over-labeled and/or is aggregating
Avoid over-labeling and increase salt concentration.
Fluorescent materials leached from plastic labware and clothes (possibly detergent or softener)
Avoid using plastic products in all steps. Wear clean laboratory coats without significant detergent residues.
Figure 8 Functionalized supported membranes on micro-fabricated substrates. (A) FRAP images of a functionalized supported membrane made of 5 mol% maleimide-DHPE, 10 mol% DOPS, and 85 mol% DOPC on micro-patterned, gridded substrate. Modified H-Ras is linked to the supported membrane through its C-terminal cysteine and then loaded with BODIPY-GTP. The surface density of BODIPY-GTP-loaded Ras is ∼2000/μm2 . FRAP images clearly show that the diffusion of proteins in corrals made of chromium lines are restricted (center of the upper right quadrant), while proteins in areas lacking grids are free to diffuse. (B) Fluorescent image of a supported membrane on a glass substrate with micro-fabricated curvatures showing phase-separation. The lipid composition, in mol %, of the bilayer is DOPC/DPPC/eggSM/cholesterol/DOTAP/TexasRed-DPPE = 48.6/19.4/0/30/1.5/0.5. The fluorescent lipid TexasRedDPPE, which partitions strongly to the disordered phase (gray areas), is used to observe the effect of curvature on phase behavior of lipid mixtures. (C) Bright-field image of the underlying substrate shown in (B). Scale bars are 10 μm.
D of lipids should be ∼4 μm2 /sec at room temperature. Note that the diffusion coefficients of lipids in supported membranes can be varied by lipid composition and cations present in the solution, as shown in Figure 7. The diffusion coefficient of proteins on supported membranes depends on the linkage used to tether the protein to the membrane. The biotinstreptavidin-biotin linkage usually creates an immobile fraction (∼50%) on the supported membrane and the diffusion coefficient of the mobile fraction is about ∼1 μm2 /sec. The dif-
fusion coefficient of proteins that are linked to the membrane through polyHistidine-Ni NTA is also ∼1 to 2 μm2 /sec, but samples prepared by this method should be free of immobile fractions. The slower diffusion of the proteins, compared to the lipids, is a result of polyvalent interactions between the poly-histidine in proteins and multiple Ni-NTA lipids. For membranes functionalized by cysteine-maleimide linkages and by direct insertion of lipidated proteins, the diffusivity of the proteins should be the same as that of lipids in the supported
Volume 2
265 Current Protocols in Chemical Biology
Volume 2
membranes, i.e., ∼4 μm2 /sec at room temperature, and the samples should be free of immobile fractions, when prepared properly. Figure 8 shows examples of functionalized supported membranes on micro-fabricated glass substrates.
Time Considerations All procedures described above can be completed within a day or two. Preparation of SUV suspensions requires ∼1 hr. Cleaning the glass substrates usually takes <1 hr (longer for the base etching method). Formation and functionalization of the supported membranes may require several hours and depends on the method chosen. Characterization of the samples can be completed within an hour or two if all the reagents are available.
Literature Cited Aegerter, M.A. and Mennig, M. 2004. Sol-gel technologies for glass producers and users. Kluwer Academic Publishers, Boston. Akashi, K., Miyata, H., Itoh, H., and Kinosita, K. Jr. 1996. Preparation of giant liposomes in physiological conditions and their characterization under an optical microscope. Biophysical J. 71:3242-3250. Amirfeiz, P., Bengtsson, S., Bergh, M., Zanghellini, E., and Borjesson, L. 2000. Formation of silicon structures by plasma-activated wafer bonding. J. Electrochem. Soc. 147:2693-2698. Anderson, T.H., Min, Y.J., Weirich, K.L., Zeng, H.B., Fygenson, D., and Israelachvili, J.N. 2009. Formation of supported bilayers on silica substrates. Langmuir 25:6997-7005. Ashley, K.M., Meredith, J.C., Amis, E., Raghavan, D., and Karim, A. 2003. Combinatorial investigation of dewetting: Polystyrene thin films on gradient hydrophilic surfaces. Polymer 44:769772. Axelrod, D., Koppel, D.E., Schlessinger, J., Elson, E., and Webb, W.W. 1976. Mobility measurement by analysis of fluorescence photobleaching recovery kinetics. Biophysical J. 16:10551069. Baumgartner, H., Fuenzalida, V., and Eisele, I. 1987. Ozone Cleaning of the Si-SiO2 System. Appl. Physics A Mater. Sci. Proc. 43:223226. Bayerl, T.M. and Bloom, M. 1990. Physical properties of single phospholipid bilayers adsorbed to micro glass beads. A new vesicular model system studied by 2 H-nuclear magnetic resonance. Biophysical J. 58:357-362. Supported Membrane Formation, Characterization, Functionalization and Patterning
Blanchette, C.D., Lin, W.-C., Ratto, T.V., and Longo, M.L. 2006. Galactosylceramide domain microstructure: Impact of cholesterol and nucleation/growth conditions. Biophysical J. 90:4466-4478.
Castellana, E.T. and Cremer, P.S. 2006. Solid supported lipid bilayers: From biophysical studies to sensor design. Surface Sci. Reports 61:429444. Chaiet, L. and Wolf, F.J. 1964. Properties of streptavidin biotin-binding protein produced by Streptomycetes. Arch. Biochem. Biophys. 106:1-5. Chiantia, S., Ries, J., and Schwille, P. 2009. Fluorescence correlation spectroscopy in membrane structure elucidation. Biochim. Biophys. Acta 1788:225-233. Contino, P.B., Hasselbacher, C.A., Ross, J.B.A., and Nemerson, Y. 1994. Use of an oriented transmembrane protein to probe the assembly of a supported phospholipid-bilayer. Biophysical J. 67:1113-1116. Cremer, P.S. and Boxer, S.G. 1999. Formation and spreading of lipid bilayers on planar glass supports. J. Phys. Chem. B 103:2554-2559. Dean, C., Scholl, F.G., Choih, J., DeMaria, S., Berger, J., Isacoff, E., and Scheiffele, P. 2003. Neurexin mediates the assembly of presynaptic terminals. Nat. Neurosci. 6:708-716. DeMond, A.L., Mossman, K.D., Starr, T., Dustin, M.L., and Groves, J.T. 2008. T cell receptor microcluster transport through molecular mazes reveals mechanism of translocation. Biophysical J. 94:3286-3292. DeRosa, R.L., Schader, P.A., and Shelby, J.E. 2003. Hydrophilic nature of silicate glass surfaces as a function of exposure condition. J. Non-Cryst. Solids 331:32-40. Doering, R. and Nishi, Y. 2008. Handbook of semiconductor manufacturing technology, 2nd ed. CRC Press, Boca Raton, Fla. Forstner, M.B., Yee, C.K., Parikh, A.N., and Groves, J.T. 2006. Lipid lateral mobility and membrane phase structure modulation by protein binding. J. Am. Chem. Soc. 128:1522115227. Galush, W.J., Nye, J.A., and Groves, J.T. 2008. Quantitative fluorescence microscopy using supported lipid bilayer standards. Biophysical J. 95:2512-2519. Galush, W.J., Shelby, S.A., Mulvihill, M.J., Tao, A., Yang, P., and Groves, J.T. 2009. A nanocube plasmonic sensor for molecular binding on membrane surfaces. Nano. Lett. 9:20772082. Giocondi, M.C., Seantier, B., Dosset, P., Milhiet, P.E., and Le Grimellec, C. 2008. Characterizing the interactions between GPI-anchored alkaline phosphatases and membrane domains by AFM. Pflug. Arch. Eur. J. Physiol. 456:179-188. Goksu, E.I., Vanegas, J.M., Blanchette, C.D., Lin, W.C., and Longo, M.L. 2009. AFM for structure and dynamics of biomembranes. Biochim. Biophys. Acta 1788:254-266. Goldman, M.A., Graves, D.B., Antonelli, G.A., Behera, S.P., and Kelber, J.A. 2009. Oxygen radical and plasma damage of low-k organosilicate glass materials: Diffusion-controlled
266 Volume 2
Current Protocols in Chemical Biology
mechanism for carbon depletion. J. Appl. Phys. 106:013311. Gomez, E.W., Clack, N.G., Wu, H.J., and Groves, J.T. 2009. Like-charge interactions between colloidal particles are asymmetric with respect to sign. Soft Matter 5:1931-1936. Grakoui, A., Bromley, S.K., Sumen, C., Davis, M.M., Shaw, A.S., Allen, P.M., and Dustin, M.L. 1999. The immunological synapse: a molecular machine controlling T cell activation. Science 285:221-227. Grogan, M.J., Kaizuka, Y., Conrad, R.M., Groves, J.T., and Bertozzi, C.R. 2005. Synthesis of lipidated green fluorescent protein and its incorporation in supported lipid bilayers. J. Am. Chem. Soc. 127:14383-14387.
Jass, J., Tjarnhage, T., and Puu, G. 2000. From liposomes to supported, planar bilayer structures on hydrophilic and hydrophobic surfaces: An atomic force microscopy study. Biophysical J. 79:3153-3163. Johnson, J.M., Ha, T., Chu, S., and Boxer, S.G. 2002. Early steps of supported bilayer formation probed by single vesicle fluorescence assays. Biophysical J. 83:3371-3379. Kaizuka, Y. and Groves, J.T. 2004. Structure and dynamics of supported intermembrane junctions. Biophysical J. 86:905-912. Kaizuka, Y. and Groves, J.T. 2006. Hydrodynamic damping of membrane thermal fluctuations near surfaces imaged by fluorescence interference microscopy. Phys. Rev. Lett. 96:118101.
Groves, J.T. 2005. Molecular organization and signal transduction at intermembrane junctions. Angew. Chem. Int. Ed. Engl. 44:3524-3538.
Kasi, S.R. and Liehr, M. 1990. Vapor-phase hydrocarbon removal for Si processing. Appl. Phys. Lett. 57:2095-2097.
Groves, J.T. 2006. Spatial mutation of the T cell immunological synapse. Curr. Opin. Chem. Biol. 10:544-550.
Keller, C.A., Glasmastar, K., Zhdanov, V.P., and Kasemo, B. 2000. Formation of supported membranes from vesicles. Phys. Rev. Lett. 84:54435446.
Groves, J.T. 2007. Bending mechanics and molecular organization in biological membranes. Annu. Rev. Phys. Chem. 58:697-717. Groves, J.T., Ulman, N., and Boxer, S.G. 1997. Micropatterning fluid lipid bilayers on solid supports. Science 275:651-653. Groves, J.T., Parthasarathy, R., and Forstner, M.B. 2008. Fluorescence imaging of membrane dynamics. Annu. Rev. Biomed. Eng. 10:311-338. Gureasko, J., Galush, W.J., Boykevisch, S., Sondermann, H., Bar-Sagi, D., Groves, J.T., and Kuriyan, J. 2008. Membrane-dependent signal integration by the Ras activator Son of sevenless. Nat. Struct. Mol. Biol. 15:452-461. Hartman, N.C., Nye, J.A., and Groves, J.T. 2009. Cluster size regulates protein sorting in the immunological synapse. Proc. Natl. Acad. Sci. U.S.A. 106:12729-12734. Hendrickson, W.A., Pahler, A., Smith, J.L., Satow, Y., Merritt, E.A., and Phizackerley, R.P. 1989. Crystal-structure of core streptavidin determined from multiwavelength anomalous diffraction of synchrotron radiation. Proc. Natl. Acad. Sci. U.S.A. 86:2190-2194. Hope, M.J., Bally, M.B., Webb, G., and Cullis, P.R. 1985. Production of large unilamellar vesicles by a rapid extrusion procedure— Characterization of size distribution, trapped volume and ability to maintain a membranepotential. Biochim. Biophys. Acta 812:5565. Huang, C.H. 1969. Studies on phosphatidylcholine vesicles . Formation and physical characteristics. Biochemistry 8:344-352. Israelachvili, J.N. 1992. Intermolecular and surface forces, Second Edition. Academic Press, New York. Jaeger, R.C. 2002. Introduction to microelectronic fabrication, 2nd edition. Modular Series on Solid State Devices. Prentice Hall, Upper Saddle River, N.J.
Kondoh, E., Baklanov, M.R., Bender, H., and Maex, K. 1998. Structural change in porous silica thin film after plasma treatment. Electrochem. Solid State Lett. 1:224-226. Kunding, A.H., Mortensen, M.W., Christensen, S.M., and Stamou, D. 2008. A fluorescencebased technique to construct size distributions from single-object measurements: Application to the extrusion of lipid vesicles. Biophysical J. 95:1176-1188. Legrini, O., Oliveros, E., and Braun, A.M. 1993. Photochemical processes for water-treatment. Chem. Rev. 93:671-698. Macdonald, R.C., Jones, F.D., and Qiu, R.Z. 1994. Fragmentation into small vesicles of dioleoylphosphatidylcholine bilayers during freezing and thawing. Biochim. Biophys. Acta Biomembr. 1191:362-370. Maier, O., Oberle, V., and Hoekstra, D. 2002. Fluorescent lipid probes: Some properties and applications (a review). Chem. Phys. Lipids 116:318. Manley, S. and Gordon, V.D. 2008. Making giant unilamellar vesicles via hydration of a lipid film. Curr. Protoc. Cell Biol. 40:24.3.1-24.3.13. Manz, B.N. and Groves, J.T. 2010. Spatial organization and signal transduction at intercellular junctions. Nat. Rev. Mol. Cell Biol. 11:342-352. Mayer, L.D., Hope, M.J., Cullis, P.R., and Janoff, A.S. 1985. Solute distributions and trapping efficiencies observed in freeze-thawed multilamellar vesicles. Biochim. Biophys. Acta 817:193196. McIntire, T.M., Smalley, S.R., Newberg, J.T., Lea, A.S., Hemminger, J.C., and Finlayson-Pitts, B.J. 2006. Substrate changes associated with the chemistry of self-assembled monolayers on silicon. Langmuir 22:5617-5624. Michel, B., Giza, M., Krumrey, M., Eichler, M., Grundmeier, G., and Klages, C.P. 2009. Effects
Volume 2
267 Current Protocols in Chemical Biology
Volume 2
of dielectric barrier discharges on silicon surfaces: Surface roughness, cleaning, and oxidation. J. Appl. Phys. 105:073302.
Rigaud, J.L. and Levy, D. 2003. Reconstitution of membrane proteins into liposomes. Methods Enzymol. 372:65-86.
Moon, D.W., Kurokawa, A., Ichimura, S., Lee, H.W., and Jeon, I.C. 1999. Ultraviolet-ozone jet cleaning process of organic surface contamination layers. J. Vacuum Sci. Technol. A 17:150154.
Sackmann, E. 1996. Supported membranes: Scientific and practical applications. Science 271:4348.
Mossman, K.D., Campi, G., Groves, J.T., and Dustin, M.L. 2005. Altered TCR signaling from geometrically repatterned immunological synapses. Science 310:1191-1193. Nye, J.A. and Groves, J.T. 2008. Kinetic control of histidine-tagged protein surface density on supported lipid bilayers. Langmuir 24:41454149. Parthasarathy, R. and Groves, J.T. 2004. Protein patterns at lipid bilayer junctions. Proc. Natl. Acad. Sci. U.S.A. 101:12798-12803. Parthasarathy, R., Cripe, P.A., and Groves, J.T. 2005. Electrostatically driven spatial patterns in lipid membrane composition. Phys. Rev. Lett. 95:048101. Parthasarathy, R., Yu, C.H., and Groves, J.T. 2006. Curvature-modulated phase separation in lipid bilayer membranes. Langmuir 22:5095-5099. Patty, P.J. and Frisken, B.J. 2003. The pressuredependence of the size of extruded vesicles. Biophysical J. 85:996-1004. Paulick, M.G., Wise, A.R., Forstner, M.B., Groves, J.T., and Bertozzi, C.R. 2007. Synthetic analogues of glycosylphosphatidylinositolanchored proteins and their behavior in supported lipid bilayers. J. Am. Chem. Soc. 129:11543-11550. Pfeiffer, I. and Hook, F. 2004. Bivalent cholesterolbased coupling of oligonucletides to lipid membrane assemblies. J. Am. Chem. Soc. 126:1022410225. Pietsch, G.J., Kohler, U., and Henzler, M. 1994. Chemistry of silicon surfaces after wet chemical preparation—A thermodesorption spectroscopy study. J. Vacuum Sci. Technol. B 12:78-87. Reinhardt, K.A. and Kern, W. 2008. Handbook of Silicon Wafer Cleaning Technology, 2nd edition. Materials Science & Process Technology Series. William Andrew, Norwich, New York. Reviakine, I. and Brisson, A. 2000. Formation of supported phospholipid bilayers from unilamellar vesicles investigated by atomic force microscopy. Langmuir 16:1806-1815. Richter, R., Mukhopadhyay, A., and Brisson, A. 2003. Pathways of lipid vesicle deposition on solid surfaces: A combined QCM-D and AFM study. Biophysical J. 85:3035-3047.
Supported Membrane Formation, Characterization, Functionalization and Patterning
Richter, R.P. and Brisson, A. 2003. Characterization of lipid bilayers and protein assemblies supported on rough surfaces by atomic force microscopy. Langmuir 19:1632-1640. Richter, R.P., Berat, R., and Brisson, A.R. 2006. Formation of solid-supported lipid bilayers: An integrated view. Langmuir 22:3497-3505.
Salafsky, J., Groves, J.T., and Boxer, S.G. 1996. Architecture and function of membrane proteins in planar supported bilayers: A study with photosynthetic reaction centers. Biochemistry 35:14773-14781. Salaita, K., Nair, P.M., Petit, R.S., Neve, R.M., Das, D., Gray, J.W., and Groves, J.T. 2010. Restriction of receptor movement alters cellular response: Physical force sensing by EphA2. Science 327:1380-1385. Seantier, B. and Kasemo, B. 2009. Influence of mono- And divalent ions on the formation of supported phospholipid bilayers via vesicle adsorption. Langmuir 25:5767-5772. Seu, K.J., Pandey, A.P., Haque, F., Proctor, E.A., Ribbe, A.E., and Hovis, J.S. 2007. Effect of surface treatment on diffusion and domain formation in supported lipid bilayers. Biophysical J. 92:2445-2450. Soumpasis, D.M. 1983. Theoretical analysis of fluorescence photobleaching recovery experiments. Biophysical J. 41:95-97. Tamm, L.K. and McConnell, H.M. 1985. Supported phospholipid bilayers. Biophysical J. 47:105113. Tanaka, M., Tutus, M., Kaufmann, S., Rossetti, F.F., Schneck, E., and Weiss, I.M. 2009. Native supported membranes on planar polymer supports and micro-particle supports. J. Struct. Biol. 168:137-142. Tenchov, B.G., Yanev, T.K., Tihova, M.G., and Koynova, R.D. 1985. A probability concept about size distributions of sonicated lipid vesicles. Biochim. Biophys. Acta 816:122-130. Tiller, H.J., Grosch, M., Dumke, K., and Sokoll, R. 1978. Influence of plasma treatment on surface properties and coating behavior of silicon dioxide and glass. Vacuum 28:125-134. Traikia, M., Warschawski, D.E., Recouvreur, M., Cartaud, J., and Devaux, P.F. 2000. Formation of unilamellar vesicles by repetitive freeze-thaw cycles: Characterization by electron microscopy and 31P-nuclear magnetic resonance. Eur. Biophys. J. 29:184-195. Vamosi, G., Damjanovich, S., Szollosi, J., and Vereb, G. 2009. Measurement of molecular mobility with fluorescence correlation spectroscopy. Curr. Protoc. Cytom. 50:2.15.12.15.19. Vig, J.R. 1985. UV ozone cleaning of surfaces. J. Vacuum Sci. Technol. A 3:1027-1034. Vig, J.R. 1987. UV/ozone cleaning of surfaces. In Treatise on Clean Surface Technology, Vol. 1 pp.1-26. Plenum Press, New York. Weirich, K.L., Israelachvili, J.N., and Fygenson, D.K. 2010. Bilayer edges catalyze supported
268 Volume 2
Current Protocols in Chemical Biology
lipid bilayer formation. Biophysical J. 98:8592. Williams, K.R., Gupta, K., and Wasilik, M. 2003. Etch rates for micromachining processing—Part II. J. Microelectromechan. Syst. 12:761-778. Wong, A.P. and Groves, J.T. 2002. Molecular topography imaging by intermembrane fluorescence resonance energy transfer. Proc. Natl. Acad. Sci. U.S.A. 99:14147-14152. Yamazaki, V., Sirenko, O., Schafer, R.J., and Groves, J.T. 2005. Lipid mobility and molecular binding in fluid lipid membranes. J. Am. Chem. Soc. 127:2826-2827. Yu, C.H. and Groves, J.T. 2010. Engineering supported membranes for cell biology. Med. Biol. Eng. Comput. 48:955-963. Yu, C.H., Parikh, A.N., and Groves, J.T. 2005. Direct patterning of membrane-derivatized colloids using in-situ UV-ozone photolithography. Adv. Materials 17:1477-1480.
Volume 2
269 Current Protocols in Chemical Biology
Volume 2