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
Parallel High-Throughput Automated Assays to Measure Cell Growth and Beta-Galactosidase Reporter Gene Expression in the Yeast Saccharomyces cerevisiae Andrew D. Napper,1,3 Nuzhat Motlekar,1 Rogerio Alves de Almeida,2 and Graham D. Pavitt2 1 Penn Center for Molecular Discovery, Institute for Medicine and Engineering, and Department of Chemical and Biomolecular Engineering, University of Pennsylvania, Philadelphia, Pennsylvania 2 Faculty of Life Sciences, University of Manchester, Manchester, United Kingdom 3 Nemours Center for Childhood Cancer Research, Wilmington, Delaware
ABSTRACT Parallel high-throughput automated assays are described for the measurement of cell growth and β-galactosidase reporter gene expression from a single culture of the yeast S. cerevisiae. The dual assay measures the effect of test compounds on expression of a specific gene of interest linked to the β-galactosidase reporter gene, and simultaneously tests for compound toxicity and other effects on cell growth. Examples of assay development and validation results are used to illustrate how this protocol may be used to screen two yeast cell lines in parallel. Yeast cells are grown overnight in V-bottom polypropylene 384-well plates, after which portions of the cell suspension are transferred to clear and to white flat-bottom 384-well plates for measurement of cell growth and reporter gene expression, respectively. Cell growth is determined by measurement of absorbance at 595 nm, and β-galactosidase expression is quantified by Beta-Glo, a commercially available luminescent β-galactosidase substrate. Curr. Protoc. Chem. Biol. C 2011 by John Wiley & Sons, Inc. 3:1-14 Keywords: cell growth r yeast r reporter gene r luciferase r cell-based assay r model organism r mutant gene r β-galactosidase r luminescence
INTRODUCTION Here, we describe HTS assays to assess reporter gene activity in the yeast Saccharomyces cerevisiae. Yeasts have been used extensively to study the effect of specific genes and genetic changes on cellular phenotypes. These studies have provided valuable insight into human disease processes and novel approaches for therapeutic intervention due to the similarity of yeast cellular control mechanisms to those in mammalian cells and the relative ease with which yeast genes can be manipulated. The optimization of these assays is exemplified by their validation for a high-throughput screen designed to discover chemical modulators of stress-response pathways in yeast (Alves de Almeida et al., 2008; Motlekar et al., 2009). Both cell growth and expression of a chromosomally integrated GCN4-dependent reporter gene fusion to β-galactosidase were measured in two yeast strains, wild-type Saccharomyces cerevisiae and an eIF2B mutant. GCN4 is a central regulator of general amino acid control stress response (GAAC), many effects of which are mediated through the translation factor eIF2B. Mutant and wildtype yeast strains were screened in parallel to identify compounds acting specifically on either strain: wild type–specific compounds affecting GAAC through functional eIF2B, Current Protocols in Chemical Biology 3: 1-14, January 2011 Published online January 2011 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/9780470559277.ch100119 C 2011 John Wiley & Sons, Inc. Copyright
HighThroughput Assays of Yeast Cell Growth
1 Volume 3
and mutant-specific compounds that restore normal function to mutant eIF2B. The gene expression assay described here was optimized for high-throughput screening (HTS) by adaptation of a commercially available coupled β-galactosidase–firefly luciferase system for use with yeast cells (Alves de Almeida et al., 2008; Motlekar et al., 2009). In this assay β-galactosidase present in the yeast cells reacts with 6-O-β-galactopyranosyl-luciferin substrate to generate D-luciferin. This, in turn, reacts with ATP and firefly luciferase (provided in the commercially available assay reagent kit) in the presence of atmospheric oxygen to form oxyluciferin and emit light. In the discussion that follows the protocols, technical challenges encountered in miniaturizing to a robot-compatible 384-well format are described. The progression of protocols is as follows:
r Basic Protocol 1: Preparation of frozen yeast stocks (matched pair, e.g., wild type r r r r r r
BASIC PROTOCOL 1
and mutant gene), optimization of growth conditions, and determination of the effect of DMSO on cell growth Basic Protocol 2: Yeast end-point growth assay: absorbance at 595 nm Basic Protocol 3: Time-course luminescence measurement of β-galactosidase reporter expression Basic Protocol 4: End-point luminescence measurement of β-galactosidase expression Basic Protocol 5: Single concentration compound screening Basic Protocol 6: Data analysis and hit selection Basic Protocol 7: Dose-response testing, curve fitting, IC50 determination, and hit confirmation
PREPARATION OF FROZEN YEAST STOCKS, OPTIMIZATION OF CELL GROWTH CONDITIONS, AND DETERMINATION OF THE EFFECT OF DMSO ON YEAST CELL GROWTH Yeast cell stocks are prepared in sufficient quantity for assay validation, HTS, and anticipated follow-up experiments. These stocks are divided into single-use aliquots to ensure a consistent source of cells throughout all subsequent experiments. Experiments to determine a cell growth time course and the effect of DMSO on cell growth are carried out prior to configuring the assay for HTS.
Materials Matched pair of yeast strains YPD plates (contain YPD medium with an addition of 20 g/liter agar; supplied by Formedium) YPD medium (10 g/liter yeast extract, 20 g/liter bacto peptone, and 20 g/liter glucose; available from Sigma) Freezing medium: YPD medium with 15% (v/v) glycerol Dry ice 0% to 1.6% (v/v) final dimethyl sulfoxide (DMSO)
High-Throughput Assays of Yeast Cell Growth
2 Volume 3
30◦ C incubator 250-ml Erlenmeyer flasks Shaking incubator Freezer vials for yeast stocks: 0.2-ml snap-cap PCR tubes (from any supplier of standard laboratory accessories) Polypropylene V-bottom 384-well plates (Greiner Bio-One, cat. no. 781280) Pipetting workstation equipped with 384-tip pipetting head (e.g., JANUS from Perkin Elmer or equivalent; Rudnicki and Johnston, 2009) Reagent dispenser (e.g., Multidrop-384 reagent dispenser from Thermo Scientific; Rudnicki and Johnston, 2009) Plate reader capable of reading absorbance and luminescence in 384-well plates (e.g., EnVision multimode plate reader from Perkin Elmer) Current Protocols in Chemical Biology
Clear 384-well plates (Corning, cat. no. 3702) Breathe-Easy sealing films (Sigma, cat. no. Z380059) Prepare the frozen yeast stocks 1. Remove the strains from the –80◦ C freezer, streak onto YPD plates, and incubate the plates for 3 days at 30◦ C. For the study used as an example here, wild-type yeast strain GP4213 (MATα leu2-3 leu2-112 ura3-52::HIS4–lacZ ino1 gcd6 gcn2::hisG pAV1265[GCD6 CEN6 LEU2]) and an isogenic gcd6 mutant strain GP4198 (MATα leu2-3 leu2-112 ura3-52::HIS4–lacZ ino1 gcd6 gcn2::hisG pAV1744[gcd6-R284H CEN6 LEU2]) (Alves de Almeida et al., 2008) were used. Store at –80◦ C as single-use frozen stocks in YPD containing 15% (v/v) glycerol. Generation of yeast strains is outside the scope of this protocol. It is assumed that the reader has already obtained the desired strains.
2. Prepare cell stocks of the two yeast strains by inoculating a single colony into 50 ml of YPD medium at room temperature in a 250-ml Erlenmeyer flask and growing the cultures overnight at 30◦ C with vigorous shaking (200 rpm). 3. Measure absorbance at 595 nm (A595) after the overnight growth. The cell suspension should give an absorbance of at least 4 to 6 before proceeding to step 4. (Absorbance at 595 nm is a measure of cell density: increased cell density increases turbidity of the cell suspension, and turbidity may be measured by absorbance at 595 nm.) Note that dilution of the culture might be necessary to obtain a reading in the linear range of the plate reader. See Critical Parameters for a more detailed explanation of the use of absorbance to measure cell density.
4. Centrifuge the cell culture at 5 min 3000 × g, 4◦ C, decant the supernatant, and resuspend the pelleted cells in freezing medium to generate a cell suspension with A595 = 5.0. 5. Divide the cells into aliquots in freezer vials for single use. Freeze on dry ice and store up to 1 year at –80◦ C. A volume of 0.5 ml contains sufficient cell stock for screening of twenty plates, taking into account dead volume and wastage of the liquid dispenser.
1.2
Absorbance at 595 nm
1
0.8
0.6
0.4
0.2
0 0
0.2
0.4
0.6
0.8
1
1.2
Fraction of stock concentration
Figure 1 Dilution of a suspension of yeast cell stock used to determine the linear range of absorbance determination at 595 nm (A595 ). Previously published in Motlekar (2009). Current Protocols in Chemical Biology
HighThroughput Assays of Yeast Cell Growth
3 Volume 3
Determine linear range of absorbance values as a measure of cell density 6. Thaw the yeast cell stock frozen at –80◦ C (from step 5) and serially dilute in a clear 384-well plate in YPD medium to obtain 50 μl per well of dilutions ranging from 1:5 to 1:500. Measure A595 in the plate reader. 7. Graph A595 data versus dilution to determine the linear range of absorbance values useful for monitoring yeast growth (Fig. 1). See Critical Parameters for more detailed notes on determining the linear range of absorbance values useful for monitoring yeast growth.
Determine optimal time to incubate yeast suspensions prior to recording growth 8. Select the optimal time for end-point reading in the growth assay by determining a growth time course over 24 hr as follows: a. For each time point fill a clear polystyrene 384-well plate with 25 μl per well of YPD medium. b. Dilute the thawed wild-type and mutant yeast cells 1:500 in YPD medium. Inoculate 25 μl of each of the diluted stocks to specific wells as follows: Add 25 μl diluted wild-type cells to wells A-P in columns 2 and 3-12, add 25 μl diluted mutant cells to wells A-P in columns 13-22 and 24, and add 25 μl of YPD medium in place of yeast cells to wells designated as blanks (wells A-P in columns 1 and 23). c. Seal the plates with Breathe-Easy sealing films and incubate the yeast cultures at 30◦ C without shaking for 4, 8, 16, or 24 hr. d. After incubation, remove the Breathe-Easy membranes and measure A595 in the plate reader. The optimal incubation time prior to recording growth will be one in which the strain is still growing in exponential phase and the cell density is in the linear range of absorbance values (Fig. 2). See Critical Parameters for more details.
0.5
Absorbance at 595 nm
0.4
0.3
0.2
0.1
0.0 0
5
10
15
20
25
Time (hr)
High-Throughput Assays of Yeast Cell Growth
Figure 2 Time course of yeast growth. Growth time course of wild-type (square) and mutant (triangle) yeast determined by measurement of A595 at each time point. Blanks consisting of medium alone (diamond) were used to monitor A595 in the absence of cell growth. Previously published in Motlekar (2009).
4 Volume 3
Current Protocols in Chemical Biology
0.30
Absorbance at 595 nm
0.25
0.20
0.15
0.10
0.05
0.00 0
0.25
1
1.6
Percent DMSO
Figure 3 Effect of DMSO on yeast growth. Cell growth after 16 hr was determined by measurement of A595 of a suspension of wild-type cells (light gray), mutant cells (black), or medium blank (dark gray). Error bars represent mean ± standard deviation of 16 replicate wells. Previously published in Motlekar (2009).
Determine the effect of DMSO on yeast cell growth 9. Grow both wild-type and mutant yeast cells for 16 hr in the presence of 0% to 1.6% (v/v) final DMSO concentration. Perform the growth assay as described in step 8a-c, except mix DMSO with the medium in the assay plate prior to the addition of yeast cells. Include wells without added DMSO as controls. To calculate the percent growth inhibition caused by addition of DMSO, see Basic Protocol 6. Figure 3 shows the effect of DMSO on the growth of wild-type and mutant cells. See Critical Parameters for more details on the effects of DMSO on the growth of S. cerevisiae.
YEAST END-POINT GROWTH ASSAY: ABSORBANCE AT 595 NM If growth of two yeast cell lines (e.g., mutant and wild type) is to be tested in parallel, a control column of the other cell line should be included in each test plate, as shown in Figure 4. Only negative control wells (yeast grown in the absence of test compounds) are included in the plate layout described here. Positive control wells (containing compounds known to affect the growth of the yeast strains in the assay) could be included if desired in columns 2 or 24.
BASIC PROTOCOL 2
Materials YPD medium (10 g/liter yeast extract, 20 g/liter bacto peptone, and 20 g/liter glucose; available from Sigma) Test compounds in DMSO Yeast strains, wild type and mutant Z buffer (see recipe) Polypropylene V-bottom 384-well plates (Greiner Bio-One, cat. no. 781280) Pipetting workstation equipped with pintool consisting of 384 pins with nominal transfer volume of 100 nl (e.g., JANUS MDT from Perkin Elmer or equivalent; Rudnicki and Johnston, 2009) Pipetting workstation equipped with 384-tip pipetting head (e.g., JANUS from Perkin Elmer or equivalent; Rudnicki and Johnston, 2009)
HighThroughput Assays of Yeast Cell Growth
5 Current Protocols in Chemical Biology
Volume 3
Figure 4
Assay plate formats showing (A) the wild-type assay plate and the (B) mutant assay plate.
Vortex Breathe-Easy sealing films (Sigma, cat. no. Z380059) 30◦ C incubator White 384-well plates (Corning Life Sciences, cat. no. 3652) Clear 384-well plates (Corning Life Sciences, cat. no. 3702) Plate reader capable of reading absorbance and luminescence in 384-well plates (e.g., EnVision multimode plate reader from Perkin Elmer) 1. Fill the V-bottom polypropylene 384-well plates with 25 μl of YPD medium using a reagent dispenser, and add an additional 25 μl of YPD medium to column 1 (for blanks). See Figure 4 for plate layout. 2. Add test compounds by pintool transfer to give a final concentration not to exceed 25 μM of each compound in 0.25% DMSO. Reserve wells in columns 1, 2, 23, and 24 for controls and blanks containing 0.25% DMSO but no compound.
High-Throughput Assays of Yeast Cell Growth
6 Volume 3
In the example assay here, a 384-pin pintool was used to transfer 120 nl of a 10 mM solution in DMSO into the medium-containing V-bottom growth plates, giving a final concentration of 24 μM of each compound in 0.24% DMSO in columns 3-22 once yeast cells are added. Columns 1, 2, 23, and 24 of the library plates contain DMSO in this case, so 0.24% DMSO (final concentration), but no compound, is transferred to wells in those columns.
3. Thaw frozen stocks of wild-type and mutant yeast cells and dilute 1:500 in YPD medium. Mix by vortexing to ensure even resuspension. Current Protocols in Chemical Biology
4a. Wild-type yeast test plates: Add 25 μl of diluted wild-type stock to all columns, except 1 and 23, in V-bottom growth plates using a reagent dispenser. Add 25 μl of diluted mutant stock to column 23. 4b. Mutant strain test plates: Add 25 μl of diluted mutant stock to all columns, except 1 and 23, in V-bottom growth plates using a reagent dispenser. Add 25 μl of diluted wild-type stock to column 23. 5. Seal plates with Breathe-Easy sealing film and incubate the yeast cultures for 16 hr at 30◦ C. This incubation time was selected based on earlier determinations of growth curves in Basic Protocol 1. The optimal time may vary between different strains and needs to be determined empirically in each case.
6. Mix the cell suspension by repeated aspirate-dispense cycles using 30-μl disposable tips on the 384-well pipetting head in the pipetting workstation. Transfer 5 μl to a white 384-well plate containing 20 μl Z buffer for use in Basic Protocol 3. Transfer 30 μl from each well of the portion that remains into wells in a clear 384-well flat-bottom plate, taking care to avoid formation of bubbles in the destination plate. Record A595 in the clear flat-bottom plate using the plate reader. 7. Calculate percent inhibition of growth from the A595 data, as described in Basic Protocol 6. β-GALACTOSIDASE REPORTER EXPRESSION TIME COURSE IN YEAST This protocol measures expression of a gene of interest linked to the expression of the reporter gene β-galactosidase. Enzymatic conversion of the Beta-Glo reagent by β-galactosidase results in long-lived luminescence. To obtain consistent results, the light output should be allowed to reach a steady state prior to recording luminescence. The time to reach a steady state is determined from a luminescence time course.
BASIC PROTOCOL 3
Materials Yeast cell suspension prepared in Basic Protocol 2, step 6 Z buffer Beta-Glo assay system (Promega, cat. no. E4780) White, flat-bottomed 384-well plates Reagent dispenser (e.g., Multidrop-384 reagent dispenser from Thermo Scientific; Rudnicki and Johnston, 2009) Pipetting workstation equipped with 384-tip pipetting head (e.g., JANUS from Perkin Elmer or equivalent; Rudnicki and Johnston, 2009) Plate reader capable of reading absorbance and luminescence in 384-well plates (e.g., EnVision multimode plate reader from Perkin Elmer) 1. Add 20 μl Z buffer to each well of a white, flat-bottomed 384-well plates using a reagent dispenser. 2. Add 5 μl of yeast cell suspension by tip transfer from the V-bottom yeast growth plates using a 384-well pipetting head (see Basic Protocol 2, step 6). 3. Mix the cells and Z buffer by pipetting, using the pipets on the pipetting head on the workstation from step 2, and incubate for 20 min at room temperature to permeabilize the cells. 4. Add 25 μl of Beta-Glo reagent using a reagent dispenser. 5. Measure luminescence at 2-min intervals for 4 hr on the Envision plate reader. For data workup, see Basic Protocol 6. Figure 5 depicts sample data from a luminescence time course. Current Protocols in Chemical Biology
HighThroughput Assays of Yeast Cell Growth
7 Volume 3
800
Luminescence units (⫻ 10⫺3)
700 600 500 400 300 200 100 0 0
50
100
150
200
250
Time (min)
Figure 5 Time course of luminescence resulting from GCN4-dependent expression of the βgalactosidase reporter gene. Wild-type (circle) and eIF2B mutant (diamond) yeast strains were grown overnight and mixed with Beta-Glo reagent. Error bars represent mean ± standard deviation of 16 replicate wells. In this experiment, there was no significant difference in growth rates between the mutant and wild-type strain and the data shown have not been corrected for cell growth. Previously published in Motlekar (2009).
BASIC PROTOCOL 4
LUMINESCENCE END-POINT β-GALACTOSIDASE EXPRESSION ASSAY
BASIC PROTOCOL 5
SINGLE-CONCENTRATION COMPOUND SCREENING
BASIC PROTOCOL 6
DATA ANALYSIS AND HIT SELECTION
This protocol is identical to Basic Protocol 3, except in step 5, luminescence is read once after light output has reached steady state. Using the strain and protocol described here, that time is 90 min after Beta-Glo reagent is added.
Add test compounds to yeast growth plates as described in Basic Protocol 2. Measure the cell growth (Basic Protocol 2) and β-galactosidase reporter gene expression at a single timepoint (Basic Protocol 4).
Yeast cell growth data and β-galactosidase reporter gene expression data obtained in Basic Protocol 5 is analyzed as follows to determine the percent inhibition of growth and percent inhibition of luminescence. Test compounds are selected as hits if these values exceed a defined threshold. 1. Calculate percent inhibition of growth or percent inhibition of luminescence for each test compound using the signal in absorbance units (OD) or luminescence units (LU) readout for each well, and the mean of the plate’s negative controls (columns 2 and 24), and the mean of the plate’s blanks (column 1; see plate maps in Figure 4):
% Inhibition =
High-Throughput Assays of Yeast Cell Growth
100 × [1 - (signal-blank mean) ] ( negative control mean-blank mean) Equation 1
The reference yeast strain (column 23) is not used in the calculation.
8 Volume 3
Current Protocols in Chemical Biology
2. Correct percent inhibition of luminescence for cell growth:
Corrected % inhibition =
100 × [1 – (100 − % reporter gene inhibition)] (100 − % inhibition of growth) Equation 2
3. Equations 1 and 2 may also be used if the goal of the compound screening is to identify activators of growth or luminescence. A negative value of percent inhibition indicates activation, in which growth or luminescence in the presence of test compound is increased relative to the negative control mean. These negative percent inhibition values may be converted to percent activation as follows: Percent activation = −1 × percent inhibition Equation 3
4. Select hits based on percent inhibition (or activation) exceeding a defined threshold. For the hits to be statistically significant (i.e., not just due to random scatter in the data), the threshold should be below (or above for activators) the mean of the plate’s negative controls by a minimum of three standard deviations of the controls. The standard deviation of the plate controls in the yeast growth and luminescence assays typically ranges from 10% to 15%; thus, the minimal hit threshold would be set at 30% to 45% inhibition. For assays with a high hit rate (>1%), the threshold may be set as high as 80% inhibition to keep the number of hits to a manageable number for follow-up testing.
DOSE-RESPONSE TESTING, CURVE FITTING, IC50 DETERMINATION, AND HIT CONFIRMATION
BASIC PROTOCOL 7
This protocol is identical to single compound screening (Basic Protocol 5), except that compounds are tested at multiple concentrations obtained by serial dilution. A suitable layout is sixteen two-fold dilutions of each compound, giving an assay concentration range of 100 μM to 3 nM. In this case, compounds may be serially diluted vertically down each plate such that columns 3 to 22 each contain the dilutions of one compound. Following dose-response testing, the yeast growth and luminescence data are fit to a dose-response curve to calculate IC50 (for inhibitors) or EC50 (for activators).
Materials Compound stocks (10 mM in DMSO) Dimethyl sulfoxide (DMSO) 384-well V-bottom polypropylene plates Reagent dispenser (e.g., Multidrop-384 reagent dispenser from Thermo Scientific; Rudnicki and Johnston, 2009) Pipetting workstation equipped with pintool consisting of 384 pins with nominal transfer volume of 100 nl (e.g., JANUS MDT from Perkin Elmer or equivalent; Rudnicki and Johnston, 2009) Microsoft Excel: for calculation of percent inhibition or activation (if screening database is not available) GraphPad Prism or equivalent: for graphing data and curve fitting for IC50 and EC50 calculation (if screening database is not available) OpenHTS (CeuticalSoft), ActivityBase (IDBS), or equivalent screening database: for calculation of percent inhibition or activation, comparison of data sets and selection of hits
HighThroughput Assays of Yeast Cell Growth
9 Current Protocols in Chemical Biology
Volume 3
Perform serial dilution in DMSO 1. Add 20 μl of each compound stock (10 mM in DMSO) arranged twenty per plate in wells A3 to A22 of 384-well V-bottom polypropylene plates. 2. Add 20 μl DMSO to wells A1, A2, A23, and A24 3. Add 10 μl DMSO per well to the entire plate except row A using a reagent dispenser. 4. Serial dilute the compounds two-fold by transferring 10 μl row-by-row from row A to row P using a single row of disposable tips, then remove and discard 10 μl from row P. The resulting dose-response plates contain sixteen two-fold dilutions of each compound, ranging from 10 mM to 305 nM, arranged one compound per column in columns 3 to 22. Pintool transfer into growth plates (Basic Protocol 2) gives a final range of 100 μM to 3 nM.
5. Proceed with the screening assay (Basic Protocol 5).
Fit the data to determine IC50 or EC50 6. Calculate percent growth and β-galactosidase activity for each dilution of each compound from A595 and luminescence, respectively, and the means of the plate controls and plate blanks (see plate maps in Fig. 4): % Activity = 100 × [(signal-blank mean)/(control mean-blank mean)] Equation 4
7. If there is a dose-dependent effect on growth (as revealed by a decrease in percent growth as the test compound concentration increases), correct the luminescence assay percent activity values at each compound dose for cell growth using the following equation: Corrected % activity = % activity/% growth (relative to growth controls) Equation 5
8. Fit percent activity data to a dose-response curve using nonlinear regression. A screening database or a curve-fitting and graphing program, such as GraphFit Prism, will perform a four-parameter logistic fit and calculate IC50 or EC50 as appropriate.
REAGENTS AND SOLUTIONS Z buffer 82 mM disodium hydrogen phosphate 9 mM sodium dihydrogen phosphate 0.1% (w/v) SDS 1 mM DTT Protease inhibitor tablets (Roche Applied Science, cat. no. 1873580) The sodium phosphate/SDS buffer is stable at room temperature for at least 1 year, but DTT and protease inhibitors should be added fresh on each day of use. DTT is stored frozen at −20◦ C as a 1 M stock (1000×).
COMMENTARY Background Information High-Throughput Assays of Yeast Cell Growth
Reporter gene fusions have been widely employed in biological research for decades. They provide a simple experimental means to study the control of expression of genes.
Ease and sensitivity of measurement, and determination of whether the host system naturally expresses the reporter gene are important considerations that govern the usefulness and choice of reporter gene fusion.
10 Volume 3
Current Protocols in Chemical Biology
Expression of commonly employed reporter genes may be detected by specific substrates that produce colored chemical products or emit light, easily detected with a spectrophotometer or luminometer (Sambrook and Russell, 2001). The Escherichia coli lacZ gene is one such reporter. The β-galactosidase enzyme was first studied at least 60 years ago (Lederberg, 1950) and is encoded by lacZ. β-galactosidase naturally cleaves lactose into glucose and galactose. The development of colorimetric substrates that mimic the natural lactose substrate, including O-nitrophenylβ-D-galactopyranoside (ONPG), greatly enhanced its subsequent widespread utilization. ONPG is colorless, but when cleaved it releases O-nitrophenol, which is yellow and absorbs at A420 . Thus, in the presence of excess ONPG, the yellow color develops at a rate proportional to the amount of LacZ protein present in a cell lysate (Miller, 1972). In recent years, parallel development of instrumentation, data analysis, and substrates with enhanced sensitivity has made it possible for reporter gene studies to be performed in a variety of high-throughput screening (HTS) formats. Luminescent reagents have been especially useful in facilitating low-microliter volume “mix-and-read” assays in 384- and 1536-well plates amenable to high-throughput robotic liquid handling (Inglese et al., 2007).
sure cell density after an overnight incubation, provided the yeast cells remain in an exponential growth phase. To ensure optimal growth conditions in Basic Protocol 2, yeast growth is first studied and optimized in Basic Protocol 1. Yeast growth is determined by measurement of absorbance at 595 nm (A595 ). Points to note: i. In this assay, absorbance is an indirect measure of turbidity. Turbidity of the yeast suspension increases in proportion to cell number during growth. ii. Increase in A595 is due to light scattering by the turbid suspension and not absorbance. Measurement at a long wavelength (595 nm or above) ensures that there is no absorbance due to the yeast cells or most test compounds. iii. At low cell density, there is a direct linear relationship between A595 and yeast cell number determined by turbidity of the suspension. Above a certain cell density, the relationship between absorbance and cell number deviates from linearity, and A595 understates the true cell number (Warringer and Blomberg, 2003). iv. To ensure that A595 provides a true measure of cell number, it is necessary to obtain a standard curve of A595 against yeast cell density to determine the linear range of the growth assay (see iii above). v. DMSO is known to affect yeast cells (Murata et al., 2003), so sensitivity of yeast cell growth to DMSO should be assessed.
Critical Parameters and Troubleshooting
Yeast end-point growth assay: Absorbance at 595 nm In Basic Protocol 2, yeast cells are grown overnight, after which growth is measured and a portion of the cells are removed for testing in the β-galactosidase reporter gene assay (Basic Protocol 4). Points to note: i. Incubation time for the growth assay should be selected based on optimization of growth conditions (Basic Protocol 1, steps 6 to 8). A 16-hr incubation time was selected for the yeast cell lines described here. ii. IMPORTANT: Overnight growth in Vbottom polypropylene plates instead of flatbottom plates was necessary for consistent mixing of the suspension with pipet tips and precise cell transfer by automated liquid handling for the β-galactosidase reporter gene assay (also see Luminescence end-point βgalactosidase expression assay). iii. The concentration of the test compounds should be no more than 25 μM to avoid nonspecific toxicity. Likewise, DMSO concentration should be below the level at which effects on growth were observed (Basic Protocol 1, step 9).
The yeast growth (Basic Protocols 1 and 2) and β-galactosidase expression assays (Basic Protocols 3 and 4) require careful and extensive optimization. Given the multitude of parameters that can affect data quality and reproducibility, careful analysis of the results is required. These issues are discussed here. Preparation of frozen yeast stocks Yeast strains are grown once to generate cell suspension sufficient for all assay development, screening, and hit confirmation. Advantages include: Consistency: A portion of the same cell suspension is used for each experiment. Ease of assay setup: Avoids the need to grow yeast strains before each experiment. Cell suspension samples are stored in 0.2ml PCR tubes, as these allow for rapid freezing and thawing. Optimization of growth conditions To allow sufficient time to discern test compound effects, the yeast end-point growth assay (Basic Protocol 2) was designed to mea-
HighThroughput Assays of Yeast Cell Growth
11 Current Protocols in Chemical Biology
Volume 3
Luminescence end-point β-galactosidase expression assay Yeast cells are grown overnight in V-bottom polypropylene plates, after which a portion of the cells is removed for testing in the βgalactosidase reporter gene assay (Basic Protocol 4). Points to note: i. IMPORTANT: Mixing and transferring of the cell suspensions from the yeast growth plate to the β-galactosidase assay plate must be carefully optimized. After overnight growth, the yeast cells needed to be resuspended to ensure reproducible transfer of a small volume of cells for the luminescence assay. Mixing in a flat-bottom plate tends to concentrate the cells to the rims of the well; therefore, cells should be grown in V-bottom polypropylene plates. Efficient mixing necessitates slow aspiration of the cell suspension with each pipet tip a fraction of a millimeter above the well bottom. ii. Luminescence should be read after light output has reached a steady state.
High-Throughput Assays of Yeast Cell Growth
Data analysis and hit selection The selection of “hits” following compound screening is a critical step. It is important that the data be analyzed in such a way that compounds with the desired activity are selected. Points to note: i. Luminescence values obtained in Basic Protocol 4 will be a true measure of inhibition (or activation) of β-galactosidase gene expression only if the data are corrected to take account of each compound’s effect on cell growth in Basic Protocol 2. In the absence of this growth correction (Basic Protocol 6, step 2), luminescence values are decreased or increased in proportion to the cell density following overnight yeast growth. ii. If the primary screening data are highly scattered (standard deviation of plate controls is >15%), the correction described in (i) above may not be reliable. In this case, the primary screening should be performed in duplicate, or hits should be selected based on uncorrected percent inhibition (or activation) of luminescence, and the correction made upon doseresponse testing (Basic Protocol 7). iii. If two cell lines are screened in parallel (Basic Protocol 2), compounds selectively active against one of the cell lines may be identified after primary screening, provided the data quality is acceptable (standard deviation of plate controls is <15%). Otherwise, determination of selectivity should be made after dose-response testing. For results of a screen against a library of 27,000 test compounds and
in-depth discussion of hit selection, dose response testing, and data analysis, see Motlekar (2009). Dose-response testing Because DMSO affects yeast cell growth, it is important that compounds be serially diluted in DMSO. Transfer of each dilution into the growth plate by pintool then yields a constant percent DMSO in each well. Serial dilution of compounds from a DMSO stock into buffer or growth medium results in a different amount of DMSO in each well.
Anticipated Results Basic Protocol 1: Preparation of frozen yeast stocks, optimization of cell growth conditions, and determination of the effect of DMSO on yeast cell growth 1. Linear range of absorbance values as a measure of cell density. Figure 1 is a standard curve of serially diluted yeast stock. Absorbance values up to 0.4 are directly proportional to cell density. 2. Optimal time to incubate yeast suspensions prior to recording growth. Overnight growth curves confirmed that a 16-hr growth of a 1:1000 dilution of the frozen yeast stocks gave absorbance values within the linear range of detection (Fig. 2). 3. Effect of DMSO on growth. In Figure 3, there was no significant effect on growth at 0.25% DMSO, but a reduction of almost 25% in the presence of 1% DMSO. Therefore, DMSO should be limited to 0.25% during compound screening for these particular strains, and for most other S. cerevisiae strains as well. Note that in this case, growth of the mutant yeast was repressed more strongly than that of the wild-type yeast. Basic Protocol 3: β-galactosidase reporter expression timecourse Figure 5 shows a time course of luminescence resulting from GCN4-dependent expression of the β-galactosidase reporter gene. In this experiment, light output reached a steady state after 90 min. Parallel testing of wild-type yeast and a strain containing mutant eIF2B showed that the mutant eIF2B strain gave >2fold higher luminescence. This difference provided an assay signal window sufficient for parallel screening of both wild-type and eIF2B mutant yeast to identify test compounds that restore the function of mutant eIF2B (Motlekar et al., 2009).
12 Volume 3
Current Protocols in Chemical Biology
Figure 6 Reporter gene assay validation plate. eIF2B mutant control (square), wild-type control (triangle), blank (black circle), and test (diamond) wells to which DMSO has been added to a final concentration of 0.24% were assayed for β-galactosidase expression. Previously published in Motlekar (2009).
Basic Protocol 4: Luminescence end-point β-galactosidase expression assay Figure 6 depicts results from a reporter gene assay validation plate. The cell suspension to which DMSO had been added by pintool (no test compounds) was grown overnight in a V-bottom 384-well plate, after which 5 μl was transferred to a white 384-well plate and mixed with Z buffer and Beta-Glo. Luminescence results shown provide a measure of well-to-well variability and confirm the 2-fold difference between wild-type and eIF2B mutant expression of the β-galactosidase reporter gene. Depending on the hit threshold chosen for screening (Basic Protocol 6, step 4), several outliers in Figure 6 might be selected as hits (known as false positives since the deviation from the plate control mean is not due to compound activity). Luminescence values from wells 120, 240, and 255, in particular, deviate significantly from the remainder of the data. If outliers such as these are observed, a repeat of the experiment will reveal whether there is a problem associated with a specific pin in the pintool or a specific channel in the pipetting head. If the outliers are not replicated at the same well locations between experiments, then they may be attributed to random data scatter. If further refinement of the liquid handling steps fails to reduce the occurrence of such outliers, screening may nonetheless
proceed. False-positive outliers will be eliminated during dose-response testing (Basic Protocol 7).
Time Considerations Preparation of frozen yeast stocks (Basic Protocol 1): Total growth time is 4 days, with 1 to 2 hr of hands-on time on days 0, 3, and 4. Yeast stocks are stable at −80◦ C for at least one year, so this procedure may be required only once. Optimization of growth conditions (Basic Protocol 1): Total growth time is 1 day, with 2 to 3 hr of hands-on time on days 0 and 1. Provided the experiment works according to the protocol, this optimization will only need to be performed once. End-point growth assay: absorbance at 595 nm (Basic Protocol 2): Total growth time is 16 hr, with 2 to 3 hr of hands-on time on days 0 and 1. β-galactosidase reporter expression time course (Basic Protocol 3): Total growth time is 16 hr, followed by a 3-hr luminescence time. Allow 1 to 2 hr before and 2 to 4 hr after for set up and data analysis, respectively. Luminescence end-point β-galactosidase expression assay (Basic Protocol 4): Total growth time is 16 hr, followed by a 90-min luminescence assay. Allow 1 to 2 hr before and 2 to 4 hr after for set up and data collection, respectively.
HighThroughput Assays of Yeast Cell Growth
13 Current Protocols in Chemical Biology
Volume 3
Single concentration compound screening (Basic Protocol 5): Total growth time is 16 hr, followed by a 90-min luminescence assay. Up to twenty plates may be nested for batchwise testing. For a twenty-plate screening run, allow 4 to 5 hr before starting yeast growth, and 8 hr the following day for luminescence assay set up and data collection. Data analysis and hit selection (Basic Protocol 6): Allow 1 day per week during screening. Dose-response testing, curve fitting, IC50 determination, and hit confirmation (Basic Protocol 7): Allow 1 day to select hits and serially dilute. Testing time similar to single-concentration compound screening above. Allow 1 day for data analysis of 100 to 200 compounds.
Acknowledgements Development and validation of the assays reported here was supported by the NIH Molecular Libraries Screening Center Network (Grant U54HG003915-02) and grant ELA 2005-008C5 from the European Leukodystrophy Association.
Literature Cited Alves de Almeida, R., Burgess, D., Shema, R., Motlekar, N., Napper, A.D., Diamond, S.L., and Pavitt, G.D. 2008. A Saccharomyces cerevisiae cell-based quantitative beta-galactosidase assay compatible with robotic handling and highthroughput screening. Yeast 25:71-76.
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. Lederberg, J. 1950. The beta-D-galactosidase of Escherichia coli, strain K-12. J. Bacteriol. 60:381392. Miller, J.H. 1972. Experiments in Molecular Genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. Motlekar, N., de Almeida, R.A., Pavitt, G.D., Diamond, S.L., and Napper, A.D. 2009. Discovery of chemical modulators of a conserved translational control pathway by parallel screening in yeast. Assay Drug Dev. Technol. 7:479494. Murata, Y., Watanabe, T., Sato, M., Momose, Y., Nakahara, T., Oka, S., and Iwahashi, H. 2003. Dimethyl sulfoxide exposure facilitates phospholipid biosynthesis and cellular membrane proliferation in yeast cells. J. Biol. Chem. 278:33185-33193. Rudnicki, S. and Johnston, S. 2009. Overview of liquid handling Instrumentation for highthroughput screening applications. Curr. Protoc. Chem. Biol. 1:43-54. Sambrook, J., and Russell, D.W. 2001. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. Warringer, J. and Blomberg, A. 2003. Automated screening in environmental arrays allows analysis of quantitative phenotypic profiles in Saccharomyces cerevisiae. Yeast 20:53-67.
High-Throughput Assays of Yeast Cell Growth
14 Volume 3
Current Protocols in Chemical Biology
Development of Chemical Probes for Biochemical Detection and Cellular Imaging of Myristoylated and Palmitoylated Proteins Rami N. Hannoush1 1
Genentech, South San Francisco, California
ABSTRACT Fatty acylation of proteins regulates their spatial localization and activity in living cells. Methods to monitor fatty acylation are invaluable for studying its role in regulating protein dynamics. The protocols in this unit describe a procedure that involves metabolic labeling with ω-alkynyl fatty acids for detecting and cellular imaging of fatty-acylated proteins, namely myristoylated and palmitoylated proteins. Curr. Protoc. Chem. Biol. C 2011 by John Wiley & Sons, Inc. 3:15-26 Keywords: myristoylation r palmitoylation r alkynyl fatty acids r chemical probes r metabolic labeling r click chemistry r fatty acylation
INTRODUCTION This unit describes a general procedure for the biochemical detection and cellular imaging of myristoylated and palmitoylated proteins using ω-alkynyl fatty acid probes and click chemistry. Fatty acylation of cellular proteins regulates their spatial localization and activity in living cells. On a molecular level, fatty acylation involves the enzymecatalyzed addition of 14-carbon (myristoylation) or 16-carbon (palmitoylation) fatty acid chains to cellular proteins via amide (N-myristoylation) or thioester (S-palmitoylation) linkages, respectively. A wide variety of mammalian proteins are fatty acylated, and these include a subpopulation of kinases, GTPases, heterotrimeric G proteins, cytokines, and phosphatases (Resh, 2006; Linder and Deschenes, 2007). Methods to monitor the status of protein myristoylation or palmitoylation are invaluable for studying cellular fatty acylation and its role in regulating protein behavior (Hannoush and Sun, 2010). The method described herein utilizes fatty acid analogues modified at their termini with alkyne groups (Hannoush and Arenas-Ramirez, 2009). These probes are added to cultured cells and are incorporated into cellular proteins (Fig. 1). Proteins modified with these probes are selectively conjugated to biotin-azide or rhodamine-azide that reacts with the alkyne group via a Cu(I)-catalyzed [3+2] Huisgen cycloaddition reaction (or click reaction) (Rostovtsev et al., 2002; Tornoe et al., 2002; Wang et al., 2003). The labeled proteins are then separated by SDS-PAGE and detected by either immunoblotting or in-gel fluorescence (Basic Protocol 1 and Fig. 2). Alternatively, the user can image fattyacylated proteins by fluorescence microscopy (Fig. 1). This entails fixing cells that have been metabolically labeled with the fatty acid probes and processing them with click chemistry (Basic Protocol 2 and Fig. 3). The labeling procedure has a number of attractive features. It is nonradioactive, highly sensitive, and applicable across a wide range of cellular systems. It also requires short detection time. Furthermore, the alkynyl fatty acid reagents are portable, tunable in concentration, and can be conveniently stored in a freezer for immediate use. All of
Current Protocols in Chemical Biology 3: 15-26, February 2011 Published online February 2011 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/9780470559277.ch100143 C 2011 John Wiley & Sons, Inc. Copyright
Probes for Imaging Myristoylated, Palmitoylated Proteins
15 Volume 3
A
O
n
n=9 n = 10 n = 12 n = 14
OH
Symbol
Chemical name
Detection
Alk-C13 Alk-C14 Alk-C16 Alk-C18
12-Tridecynoic acid 13-Tetradecynoic acid 15-Hexadecynoic acid 17-Octadecynoic acid
myristoylation myristoylation palmitoylation palmitoylation
B
O
O
n
X
N3
lyse cells
X = N or S
X
= biotin or rhodamine
n n = 10, 12, 14
fix and permeabilize cells
N N
CuSO4, TCEP, TBTA
O OH
N
n
Detection: immunoblot or in-gel fluorescence
N3 cellular imaging CuSO4, TCEP, TBTA
Figure 1 Schematic depicting the method for monitoring myristoylated and palmitoylated cellular proteins. (A) ω-Alkynyl fatty acid probes used for detecting protein myristoylation and palmitoylation. (B) ω-Alkynyl fatty acids added to growth medium are metabolically incorporated into cellular fatty-acylated proteins. The cells are then either lysed and the proteome detected by immunoblotting or fixed for imaging by fluorescence microscopy. In both cases, click chemistry is used to chemoselectively conjugate a biotin-azide or rhodamineazide to fatty-acylated proteins. Alk-C13 and Alk-C14 are probes for detecting protein myristoylation, while Alk-C16 and Alk-C18 detect protein palmitoylation.
this makes this technique well suited for characterizing and studying fatty acylation of cellular proteins of interest in various contexts. A key application of the alkynyl fatty acid probes, when combined with fluorescence microscopy, is imaging the global subcellular distribution of myristoylated and palmitoylated proteins in different cell lines (Hannoush and Arenas-Ramirez, 2009). This provides important information about the dynamics and behavior of fatty-acylated proteins in living cells.
STRATEGIC PLANNING
Probes for Imaging Myristoylated, Palmitoylated Proteins
Researchers first need to determine the cell type to be used for studying fatty acylation. For detecting protein myristoylation, Alk-C13 or Alk-C14 should be used (Fig. 1). Alternatively, Alk-C16 and Alk-C18 are appropriate probes for detecting protein palmitoylation (Fig. 1). In general, these probes work across a wide range of mammalian cell lines and their uptake may vary depending on the particular cell type used. In every case, the researcher should determine the optimal concentration of the probe for obtaining the maximum signal-to-noise ratio. Furthermore, one needs to check for palmitoylation or myristoylation by verifying the type of linkage via which the individual probe is incorporated into cellular proteins (thioester versus amide linkage). This is done by measuring sensitivity to hydroxylamine (in Basic Protocol 1), a reagent that cleaves thioester- but not amide-linked acyl chains.
16 Volume 3
Current Protocols in Chemical Biology
BIOCHEMICAL DETECTION OF MYRISTOYLATED AND PALMITOYLATED CELLULAR PROTEINS BY IMMUNOBLOTTING OR IN-GEL FLUORESCENCE
BASIC PROTOCOL 1
This protocol enables researchers to monitor biochemically the myristoylation and palmitoylation of cellular proteins. It involves metabolic labeling of cells with ω-alkynyl fatty acids and click chemistry followed by immunoblotting or in-gel fluorescence. The expected outcome will reveal the status of fatty acylation of cellular proteins.
Materials Cell line of interest: Examples include: Raw 264.7 macrophages (ATCC #CCL-2278) MDCK epithelial cells (ATCC #CCL-34) PC-3 cells (ATCC #CRL-1435) Mouse L-cells (ATCC #CRL-2648) HeLa cells (ATCC #CCL-2) Jurkat T cells (ATCC #TIB-152) COS-7 cells (ATCC #CRL-1651) Appropriate cell culture growth medium (e.g., DMEM, F-12K, RPMI) Fatty acid stock solution (see recipe) Bovine serum albumin (BSA; fatty acid-free, Sigma-Aldrich) Dimethyl sulfoxide (DMSO) Ethanol Phosphate-buffered saline (PBS; see recipe) Lysis buffer (see recipe) BCA protein assay kit (Thermo Scientific) Biotin-azide or rhodamine-azide (see recipe) Tris (2-carboxyethyl)phosphine hydrochloride (TCEP; Sigma-Aldrich) Tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]amine (TBTA; Sigma-Aldrich) Copper sulfate (CuSO4 ; Sigma-Aldrich) Acetone, ice-cold Novex Tris glycine SDS sample buffer (2×) (Invitrogen) NuPAGE sample reducing agent (10×) (Invitrogen) Tris-glycine gels, precast (Invitrogen) PBS-T (see recipe) Nonfat dried milk (any grocery store) Streptavidin-horseradish peroxidase (Invitrogen) Hydroxylamine solution, 50% in water (NH2 OH; Sigma-Aldrich) Restore western blot stripping buffer (Thermo Scientific) Anti-β-tubulin HRP antibody (Invitrogen) 37◦ C, 5% CO2 humidified incubator 6-well plates Sonicator (Branson) Cell scraper (25 cm, Starstedt) Centrifuge (Eppendorf) Nanocep centrifugal ultrafiltration devices (Pall Corporation) Vortexer (VWR) Thermomixer heating block (Eppendorf) ECL immunoblotting detection kit (GE Healthcare) Amersham Hyperfilm ECL (GE Healthcare) Desktop scanner ImageJ or AdobePhotoshop Typhoon scanner
Probes for Imaging Myristoylated, Palmitoylated Proteins
17 Current Protocols in Chemical Biology
Volume 3
Additional reagents and equipment for transferring the proteins onto a nitrocellulose membrane (Gallagher et al., 2008) Culture the cells 1. Grow the preferred cells in suitable growth medium in a 5% CO2 humidified chamber at 37◦ C for at least 24 hr. Examples of cell lines in which this method has been validated include: Raw 264.7 macrophages (culture in high-glucose DMEM supplemented with 10% FBS and 2 mM GlutaMax); PC-3 cells (culture in F-12K medium supplemented with 10% FBS); Jurkat (culture in RPMI supplemented with 10% FBS); MDCK, L-cells, HeLa, and COS-7 (culture in DMEM supplemented with 10% FBS). Typically, ∼8 × 105 cells per sample are needed for an experiment.
Seed the cells 2. Seed the cells with growth medium onto 6-well plates at a density of 8 × 105 cells/well. 3. Incubate for 24 hr in a 5% CO2 humidified chamber at 37◦ C before treating with fatty acids.
Prepare the fatty acid probe-containing medium 4. Dissolve the fatty acid stock solution in serum-free growth medium supplemented with 5% BSA (fatty acid-free) to a final concentration of 100 μM just prior to the experiment. As a control, dissolve DMSO to 100 μM. Perform all dissolutions in a tissue culture hood to keep the medium sterile. 5. Sonicate the solutions in closed plastic tubes in a water bath for 15 min at room temperature and then incubate for 15 min at room temperature. Wipe the tube with ethanol to minimize contamination before transferring it back to the tissue culture hood.
Treat the cells with fatty acid probes 6. Aspirate the growth medium from the seeded cells that have been growing for 24 hr. 7. Wash the cells once with 2 ml PBS. 8. Add 2 ml of fatty acid probe-containing medium per well. 9. Incubate for 24 hr at 37◦ C, 5% CO2 .
Prepare the cell lysates 10. Remove the medium and wash the cells three times, each time with 1 ml cold PBS. 11. Add 400 μl lysis buffer, scrape the cells with a cell scraper, collect the lysate in a 1.5-ml sterile plastic tube, and rock at speed 2 for 1 hr at 4◦ C. 12. Centrifuge cell lysates 10 min at 16,000 × g, 4◦ C, and collect the supernatant. 13. Concentrate the supernatant by centrifuging 20 min at 16,000 × g, 4◦ C, with Nanocep centrifugal ultrafiltration devices. This will yield a final proteome concentration of ∼2 mg/ml, and this may vary depending on cell type and confluency.
Probes for Imaging Myristoylated, Palmitoylated Proteins
14. Measure the protein concentration by BCA protein assay kit following the manufacturer’s protocol. At this point, cell lysates can be snap-frozen in liquid nitrogen and stored for up to 4 months at −80◦ C without detectable changes in their integrity.
18 Volume 3
Current Protocols in Chemical Biology
Label the cell lysates with biotin-azide or rhodamine-azide These volumes are based on a 25-μl reaction scale. 15. Add 23 μl of cell lysate at ∼2 mg/ml into a 1.5-ml plastic tube. Protein concentration in the range 1 to 2 mg/ml would work as long as 40 to 50 μg of total cell lysate is included in the reaction.
16. Add 0.5 μl of 5 mM biotin-azide to a final concentration of 0.1 mM. 17. Add 0.5 μl of 50 mM TCEP to a final concentration of 1 mM. 18. Add 0.5 μl of 10 mM TBTA to a final concentration of 0.2 mM. 19. Vortex for 5 sec at a high setting. 20. Add 0.5 μl of 50 mM CuSO4 to a final concentration of 1 mM. 21. Vortex for 5 sec at a high setting. 22. Incubate the click reaction cocktail (steps 15 to 21) for 1 hr at room temperature in dark. The order of addition of the reagents to the protein extract is important and must be performed as described above. TCEP is a reducing agent, TBTA acts as a stabilizing ligand for Cu(I), and Cu(II)SO4 is a catalyst for the click reaction.
Precipitate the proteins and prepare the samples for gel loading 23. Add 250 μl of ice-cold acetone. 24. Vortex for 5 sec (at a high setting) and incubate for 2 hr or overnight at –20◦ C. Recovery is about the same for the two incubation periods; the overnight incubation is for convenience.
25. Centrifuge 10 min at 16,000 × g, 4◦ C. There should be a visible pellet.
26. Aspirate the supernatant. 27. Air dry the tube for 5 to 10 min. 28. Resuspend the pellet in 16 μl lysis buffer. 29. Add 15.5 μl of 2× SDS sample buffer and 3.5 μl of 10× sample reducing agent. 30. Heat the sample for 5 min at 95◦ C in a heating block. 31. Load 35 μl/lane on a 10-well gel (∼30 to 40 μg protein/lane). See Critical Parameters for suggestions regarding controls.
Detect signal by immunoblotting 32. Resolve biotin-labeled protein lysates by SDS-PAGE using 4% to 20% Tris-glycine or 4% to 12% Bis-Tris gels. 33. Transfer proteins onto a nitrocellulose membrane (e.g., Gallagher et al., 2008). 34. Block the membrane with 5 ml PBS-T and 5 ml 5% nonfat dried milk for 2 hr at room temperature or overnight at 4◦ C. 35. Wash the membrane three times with PBS-T, 5 min each time. 36. Incubate the membrane with streptavidin-linked horseradish peroxidase (1:5000 in PBS-T) for 1 hr at room temperature.
Probes for Imaging Myristoylated, Palmitoylated Proteins
19 Current Protocols in Chemical Biology
Volume 3
37. Wash the membrane three times, each time with 5 ml PBS-T, 10 min each. 38. Develop signal using ECL detection kit according to manufacturer’s protocol. Capture signal on film with 1- to 2-min exposure time.
Check for protein incorporation of fatty acid probes via amide or thioester linkages This protocol involves the use of hydroxylamine and is carried out right after protein transfer (step 33 above). 39. Incubate membranes for 65 to 72 hr at room temperature with 5 ml PBS-T and 5 ml 5% NH2 OH. 40. Block the membrane with 5 ml 5% nonfat dried milk in PBS-T for 2 hr at room temperature or overnight at 4◦ C. 41. Analyze by streptavidin blot as described above (steps 35 to 38).
Demonstrate equal levels of protein loading This can be carried out either after step 38 or 41.There are several ways to check for levels of protein loading which include: Coomassie staining (total protein levels), anti-actin, anti-tubulin, etc. Below is a procedure for detecting tubulin. 42. Incubate the streptavidin blots with Restore western blot stripping buffer for 15 min at room temperature. 43. Repeat steps 34 to 35. 44. Incubate the membrane with anti-β-tubulin HRP antibody (1:10,000 in PBS-T) for 1 hr at room temperature. 45. Repeat steps 37 to 38.
B SO
2
3
148 -
5
6
kDa
1
2
3
4
5
6
148 -
98 -
98 -
64 -
64 -
50 -
50 -
36 -
36 -
22 -
22 -
-tubulin
Probes for Imaging Myristoylated, Palmitoylated Proteins
4
M
1
D
kDa
D
M
SO
A
-tubulin
Figure 2 Biochemical detection of fatty-acylated proteins. Cell lysates were prepared from MDCK cells treated with different ω-alkynyl fatty acids (100 μM) for 24 hr. The proteome was subjected to a click reaction with biotin-azide and then separated by SDS-PAGE. The membrane was soaked in PBS-T buffer (A) or in hydroxylamine (B), and the signal was detected by streptavidin-linked horseradish peroxidase (see Basic Protocol 1). Lanes: 1, Alk-C10 (n = 6); 2, Alk-C11 (n = 7); 3, Alk-C13 (n = 9); 4, Alk-C14 (n = 10); 5, Alk-C16 (n = 12); 6, Alk-C18 (n = 14). In (A), asterisks indicate specific bands labeled by the various probes. Arrows indicate nonspecific bands labeled in both nontreated (DMSO) and treated lanes. Alk-C14 yields more labeled bands than Alk-C13.
20 Volume 3
Current Protocols in Chemical Biology
Obtain images 46. Scan films using a desktop scanner and import images into ImageJ or AdobePhotoshop. See Figure 2 for typical staining with Alk-C13, Alk-C14, Alk-C16, Alk-C18, and other fatty acid probes.
Detect signal by in-gel fluorescence In cases where fluorescence scanners are available, use of rhodamine-azide is recommended as it results in minimal background and high signal-to-noise ratio relative to biotin-azide. 47. Follow steps 15 to 32 above, except replace biotin-azide with rhodamine-azide in step 16. Minimize exposure of the gel to light while carrying out the protocol. 48. Scan gel on a Typhoon scanner using the Cy3/rhodamine fluorescence channel. 49. Export and process the image in the Typhoon scanner image-processing software.
CELLULAR IMAGING OF MYRISTOYLATED AND PLAMITOYLATED PROTEINS
BASIC PROTOCOL 2
This protocol enables imaging of the global cellular myristoylated and palmitoylated proteome by fluorescence microscopy. It allows the user to capture a view of the spatial localization and dynamics of myristoylated and palmitoylated proteins under different conditions and cellular states.
Materials Phosphate-buffered saline (PBS; see recipe) Methanol (Sigma-Aldrich), prechilled PBS/0.1% (v/v) Triton X-100 Rhodamine-azide (see recipe) Tris (2-carboxyethyl)phosphine hydrochloride (TCEP; see recipe) Copper sulfate (CuSO4 ; see recipe) Hoechst 33342 (Invitrogen, cat. no. H21492) 12-well tissue culture plates Glass coverslips, washed in ethanol and left to dry for 10 min under UV light in a tissue culture hood before use 37◦ C, 5% CO2 incubator Microscope slides Fluorescence microscope and a 40× or 63× objective Standard image analysis software: e.g., Slidebook 5.0 (Intelligent Imaging Innovation) or ImageJ (NIH) Culture the cells 1. Perform step 1 in Basic Protocol 1. Seed the cells 2. Seed the cells onto 12-well plates (4 × 105 cells/well) containing ethanol-washed and dried glass coverslips and incubate for 24 hr before treatment in a 37◦ C, 5% CO2 humidified incubator. Prepare the fatty acid probe-containing medium 3. Perform steps 4 and 5 in Basic Protocol 1. Treat the cells with fatty acid probes 4. Aspirate the growth medium from wells containing seeded cells.
Probes for Imaging Myristoylated, Palmitoylated Proteins
21 Current Protocols in Chemical Biology
Volume 3
5. Wash the cells gently once with 1 ml PBS. 6. Add 1 ml of fatty acid probe–containing medium per well and incubate for 4 hr at 37◦ C, 5% CO2 .
Fix and permeabilize the cells 7. Wash the cells four times, each time with 1 ml PBS to remove excess probe. 8. Fix the cells with 1 ml prechilled methanol at −20◦ C for 10 min. 9. Aspirate the methanol. 10. Permeabilize the cells with 1 ml PBS/0.1% (v/v) Triton X-100 for 5 min at room temperature. 11. Wash the cells extensively six times, each time with 1 ml PBS.
Label the cells with rhodamine-azide 12. Remove the coverslips from the wells and place them in a humidified chamber for subsequent steps. 13. Prepare 50 μl of click reaction cocktail as follows: a. Add 47 μl of PBS into a 1.5-ml plastic tube. b. Add 1 μl of 5 mM rhodamine-azide to a final concentration of 0.1 mM. c. Add 1 μl of 50 mM TCEP to a final concentration of 1 mM. d. Add 1 μl of 50 mM CuSO4 to a final concentration of 1 mM.
nuclei
fatty-acid labeling
merge
DMSO
Alk-C14
Alk-C16
Alk-C18
10 m
Probes for Imaging Myristoylated, Palmitoylated Proteins
Figure 3 Cellular imaging of myristoylated and palmitoylated proteins. Prostate cancer cells were treated with fatty acid probes (100 μM) for 3 hr. After fixation, the cells were processed with rhodamine-azide and imaged by epifluorescence microscopy using a 40× objective (see Basic Protocol 2). Images show the perinuclear and punctuate distribution of cellular proteins labeled with Alk-C14, Alk-C16, and Alk-C18.
22 Volume 3
Current Protocols in Chemical Biology
14. Add the click reaction cocktail (step 13 above) to the coverslips and incubate for 1 hr in the dark at room temperature in the humidified chamber. 15. Rinse the cells extensively six times, each time with 100 μl PBS. 16. Incubate the cells with 50 μl of Hoechst 33342 (1:10,000 in PBS) for 10 min at room temperature. 17. Wash the cells three times, each time with 100 μl PBS. 18. Mount the coverslips onto the microscope glass slides.
Capture images by fluorescence microscopy 19. Capture images by using a fluorescence microscope and a 40× or 63× objective. We recommend an inverted Zeiss AX10 microscope equipped with a CoolSnap CCD camera (Roper Scientific), a Piezo stage for obtaining optical Z-sections, and standard excitation filters for Hoechst/DAPI (377 nm) and rhodamine (568 nm).
20. Acquire 50 to 70 optical Z-sections per image with 0.3-μm spacing.
Analyze the images 21. Export and process the images in a standard image analysis software, such as Slidebook 5.0 (Intelligent Imaging Innovation) or ImageJ (NIH). See Figure 3 for typical staining with Alk-C14, Alk-C16, and Alk-C18. REAGENTS AND SOLUTIONS Use Milli-Q purified water or equivalent in all recipes and protocol steps.
Biotin-azide or rhodamine-azide Reagents can be obtained from our laboratory, synthesized as described earlier (Lewis et al., 2004; Speers and Cravatt, 2004; Hsu et al., 2007; Hannoush and Arenas-Ramirez, 2009), or purchased from Invitrogen. Dissolve powder in dimethyl sulofoxide (DMSO) to a concentration of 5 mM. Divide into 50-μl aliquots and store up to 1 year at −20◦ C. Protect from light.
CuSO4 Dissolve powder in phosphate-buffered saline (PBS; see recipe) to a concentration of 50 mM just prior to use in the click reaction (see Critical Parameters). Prepare fresh. Fatty acid stock solution Dissolve lyophilized fatty acid powder [obtained from our laboratory or synthesized as described earlier in Hannoush and Arenas-Ramirez (2009)] in dimethyl sulfoxide (DMSO) to a concentration of 50 mM. Divide the solution into 50-μl aliquots and store indefinitely at −80◦ C. Alk-C13 and Alk-C18 can be obtained from Otava and Sigma-Aldrich, respectively.
Lysis buffer 100 mM sodium phosphate, pH 7.5 150 mM NaCl 1% (v/v) nonidet P-40 Store up to 3 months at 4◦ C Immediately before the experiment, add fresh protease and phosphatase inhibitor cocktail (Pierce)
Probes for Imaging Myristoylated, Palmitoylated Proteins
23 Current Protocols in Chemical Biology
Volume 3
PBS-T Phosphate-buffered saline (PBS; see recipe) 0.1% (v/v) Tween-20 Store up to 6 months at room temperature Phosphate-buffered saline (PBS) 8 g NaCl 0.2 g KCl 1.13 g Na2 HPO4 (sodium phosphate, dibasic anhydrous) 0.2 g KH2 PO4 (potassium phosphate, monobasic) 1 liter of H2 O Adjust pH to 7.2 ± 0.1 with 6 N HCl Store up to 6 months at room temperature Tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]amine (TBTA) Dissolve powder in DMSO/t-butanol (1:4, v:v) solution to a concentration of 10 mM. Divide the solution into 100-μl aliquots and store up to 1 year at −20◦ C.
Tris (2-carboxyethyl)phosphine hydrochloride (TCEP) Dissolve powder in water to a concentration of 50 mM just prior to use in the click reaction (see Critical Parameters).
COMMENTARY Background Information
Probes for Imaging Myristoylated, Palmitoylated Proteins
A number of techniques have been employed to monitor cellular fatty acylation (Hannoush and Sun, 2010). Radioactivity has been the standard method for detecting myristoylation and palmitoylation of cellular proteins. Typically, [3 H]- or [125 I]-labeled myristic and palmitic acids are added to cells for metabolic incorporation into cellular proteins (Schlesinger et al., 1980; Peseckis et al., 1993). The signal is developed by autoradiography and requires lengthy film exposures. Because of the hazards and costs associated with radioactivity, this method is less than ideal for use. Recently, the acyl-biotin exchange method was introduced for detecting protein palmitoylation (Drisdel and Green, 2004). This method relies on blocking the free sulfhydryl groups of palmitoylated proteins, followed by cleavage of the palmitate moiety and subsequent tagging of the unmasked cysteine with biotin for affinity capture and enrichment. While this method is a breakthrough, it has limitations primarily associated with false positives (Roth et al., 2006), and it only detects S-palmitoylation but not myristoylation. For a detailed description of the various methods and their limitations, the reader is referred to a comprehensive review on this topic (Hannoush and Sun, 2010). Recently, nonradioactive probes based on alkyne fatty acids have been developed for
metabolic labeling of myristoylated and palmitoylated proteins (Hannoush and ArenasRamirez, 2009). We reasoned that appending an alkyne group to the terminal end of a fatty acid would not interfere with the hydrophobic nature of the fatty acid and the mechanism by which it inserts into lipid membranes. The alkynyl fatty acid probes are portable and can be stored indefinitely in the freezer for immediate use. They are metabolically incorporated into fatty-acylated proteins in human and mouse cell lines. The tagged proteins are then conjugated to biotin-azide or rhodamine azide via a click reaction and detected by immunoblotting or in-gel fluorescence, respectively. Additionally, these probes enable cellular imaging of the global subcellular distribution of fatty-acylated proteins and shed light on the dynamics and turnover of such proteins (Hannoush and Arenas-Ramirez, 2009). Based on hydroxylamine sensitivity, Alk-C14 is a probe for protein myristoylation while Alk-C16 and Alk-C18 are probes for protein palmitoylation. Other emerging applications of these probes include in vitro labeling of recombinant proteins, monitoring turnover of palmitoylated proteins, enrichment of trace proteins, proteomics studies, and immunoprecipitation of specific proteins of interest for profiling their fatty acylation status (Heal et al., 2008; Martin and Cravatt, 2009; Hannoush and Sun, 2010; Yap et al., 2010). However,
24 Volume 3
Current Protocols in Chemical Biology
this method has some limitations. The toxicity of the copper reagent makes it inappropriate for monitoring fatty acylation by live-cell imaging. Furthermore, the presence of certain endogenous proteins that chelate copper may interfere with the signal of the assay, and this varies depending on the particular cellular system being studied.
Critical Parameters Reagent preparation and concentrations Avoid multiple freeze-thaw cycles of the fatty acid probe stocks as this may affect reagent quality. Also, copper sulfate and TCEP reagents should be freshly prepared just prior to use in each experiment. Because the optimal concentration of ωalkynyl fatty acid needed to obtain the best signal-to-noise ratio may vary depending on the particular cell type used, it is recommended to first test the individual probe at various increasing concentrations (recommended starting range is 1 to 200 μM) and for different times of incubation to determine the optimal conditions for their cellular system. Experimental controls Two protein gels should be run simultaneously for each set of samples. One gel is run and analyzed as described in steps 31 to 38 (Basic Protocol 1), while the other gel is treated with hydroxylamine and analyzed as described in steps 39 to 41 (Basic Protocol 1). This allows the user to determine the type of linkage via which fatty acid probes incorporate into cellular proteins. Sometimes, detection by streptavidin-HRP results in varying degrees of background staining, which stems from streptavidin labeling of endogenously biotinylated proteins. Therefore, care should be taken to decipher nonspecifically labeled bands on the gel, and it is critical that the researcher include negative controls, such as those lacking copper sulfate or the fatty acid probes. Furthermore, for imaging by fluorescence microscopy, it is critical to ensure that the signal observed is not due to nonspecific incorporation of the fatty acid probe into biological membranes. Therefore, optimizing the time for methanol fixation and the permeabilization step for the particular cell type used is crucial to avoid nonspecific incorporation into lipid bilayers. The intensity of staining will decrease with longer detergent extraction times if there is nonspecific incorporation of the probes.
Troubleshooting While the protocols described herein detail a robust method for detecting protein myristoylation and palmitoylation, in very few cases a signal may not be observed and below are recommended troubleshooting procedures. Reagent quality and concentration Fatty acid probes were not used at the appropriate concentrations to obtain the best signal-to-noise ratio. Also, the click chemistry reaction works best only when the reagents are freshly prepared just prior to the experiment. Signal optimization Sometimes, staining is not homogeneous throughout the membrane blot. This may be due to uneven transfer of proteins from the gel. In this case, ensuring that the appropriate transfer procedure is used is critical. Also, if the intensity of bands on the membrane is high, then reducing exposure times or using a more diluted stock of streptavidin-HRP can resolve this issue. Another source of reduction or loss in signal is proteins that may interfere with the click reaction, such as those that may chelate copper. These proteins may exist in a particular cellular system, and steps need to be taken to eliminate them if feasible.
Anticipated Results Basic Protocol 1 will generate gel scans of myristoylated and palmitoylated proteins (Fig. 2). This enables researchers to analyze the myristoylation or palmitoylation status of specific proteins of interest when combined with standard immunoprecipitation protocols on cells that have been metabolically labeled with Alk-C14, Alk-C16, or Alk-C18. In Basic Protocol 2, a cellular view of the distribution of myristoylated and palmitoylated proteins will be obtained (Fig. 3). The labeling method described here is very robust and the experimental results are typically consistent and reproducible across multiple runs.
Time Considerations Once the cells are seeded in plates and all the necessary reagents are in hand, the procedures described above can be completed within 2 to 3 days.
Acknowledgement I would like to thank Natalia Arenas, JingLucy Sun, and other members of the Early Discovery Biochemistry Department who contributed directly or indirectly to this project.
Probes for Imaging Myristoylated, Palmitoylated Proteins
25 Current Protocols in Chemical Biology
Volume 3
Literature Cited Drisdel, R.C. and Green, W.N. 2004. Labeling and quantifying sites of protein palmitoylation. Biotechniques 36:276-285. Gallagher, S., Winston, S.E., Fuller, S.A., and Hurrell, J.G. 2008. Immunoblotting and Immunodetection. Curr. Protoc. Mol. Biol. 83:10.8.110.8.28. Hannoush, R.N. and Arenas-Ramirez, N. 2009. Imaging the lipidome: Omega-alkynyl fatty acids for detection and cellular visualization of lipid-modified proteins. ACS Chem. Biol. 4:581587. Hannoush, R.N. and Sun, J. 2010. The chemical toolbox for monitoring protein fatty acylation and prenylation. Nat. Chem. Biol. 6:498-506. Heal, W.P., Wickramasinghe, S.R., Bowyer, P.W., Holder, A.A., Smith, D.F., Leatherbarrow, R.J., and Tate, E.W. 2008. Site-specific N-terminal labelling of proteins in vitro and in vivo using Nmyristoyl transferase and bioorthogonal ligation chemistry. Chem. Commun. (Camb) 4:480-482. Hsu, T.L., Hanson, S.R., Kishikawa, K., Wang, S.K., Sawa, M., and Wong, C.H. 2007. Alkynyl sugar analogs for the labeling and visualization of glycoconjugates in cells. Proc. Natl. Acad. Sci. U.S.A. 104:2614-2619. Lewis, W.G., Magallon, F.G., Fokin, V.V., and Finn, M.G. 2004. Discovery and characterization of catalysts for azide−alkyne cycloaddition by fluorescence quenching. J. Am. Chem. Soc. 126:9152-9153. Linder, M.E. and Deschenes, R.J. 2007. Palmitoylation: Policing protein stability and traffic. Nat. Rev. Mol. Cell Biol. 8:74-84. Martin, B.R. and Cravatt, B.F. 2009. Large-scale profiling of protein palmitoylation in mammalian cells. Nat. Methods 6:135-138. Peseckis, S.M., Deichaite, I., and Resh, M.D. 1993. Iodinated fatty acids as probes for myristate pro-
cessing and function. Incorporation into pp60vsrc. J. Biol. Chem. 268:5107-5114. Resh, M.D. 2006. Trafficking and signaling by fatty-acylated and prenylated proteins. Nat. Chem. Biol. 2:584-590. Rostovtsev, V.V., Green, L.G., Fokin, V.V., and Sharpless, K.B. 2002. A stepwise Huisgen cycloaddition process: Copper(I)-catalyzed regioselective “ligation” of azides and terminal alkynes. Angew. Chem. Int. Ed. Engl. 41:25962599. Roth, A.F., Wan, J., Green, W.N., Yates, J.R., and Davis, N.G. 2006. Proteomic identification of palmitoylated proteins. Methods 40:135142. Schlesinger, M.J., Magee, A.I., and Schmidt, M.F.G. 1980. Fatty acid acylation of proteins in cultured cells. J. Biol. Chem. 255:1002110024. Speers, A.E. and Cravatt, B.F. 2004. Profiling enzyme activities in vivo using click chemistry methods. Chem. Biol. 11:535-546. Tornoe, C.W., Christensen, C., and Meldal, M. 2002. Peptidotriazoles on solid phase: [1,2,3]Triazoles by regiospecific copper(i)-catalyzed 1,3-dipolar cycloadditions of terminal alkynes to azides. J. Org. Chem. 67:3057-3064. Wang, Q., Chan, T.R., Hilgraf, R., Fokin, V.V., Sharpless, K.B., and Finn, M.G. 2003. Bioconjugation by copper(I)-catalyzed azide-alkyne [3 + 2] cycloaddition. J. Am. Chem. Soc. 125:3192-3193. Yap, M.C., Kostiuk, M.A., Martin, D.D.O., Perinpanayagam, M.A., Hak, P.C., Siddam, A., Majjigapu, J.R., Rajaiah, G., Keller, B.O., Prescher, J.A., Wu, P., Bertozzi, C.R., Falck, J.R., and Berthiaume, L.G. 2010. Rapid and selective detection of fatty acylated proteins using omega-alkynyl-fatty acids and click chemistry. J. Lipid Res. 51:1566-1580.
Probes for Imaging Myristoylated, Palmitoylated Proteins
26 Volume 3
Current Protocols in Chemical Biology
Imaging of Endogenous RNA Using Genetically Encoded Probes Takeaki Ozawa1 and Yoshio Umezawa2 1
Department of Chemistry, Graduate School of Science, The University of Tokyo, Tokyo, and Japan Science and Technology Agency, Tokyo, Japan 2 Research Institute of Pharmaceutical Sciences, Musashino University, Tokyo, Japan
ABSTRACT Imaging of RNAs in single cells revealed their localized transcription and specific function. Such information cannot be obtained from bulk measurements. This unit contains a protocol of an imaging method capable of visualizing endogenous RNAs bound to genetically encoded fluorescent probes in single living cells. The protocol includes methods of design and construction of the probes, their characterization, and imaging a target RNA in living cells. The methods for RNA imaging are generally applicable to many kinds of RNAs and may allow for elucidating novel functions of localized RNAs and C 2011 understanding their dynamics in living cells. Curr. Protoc. Chem. Biol. 3:27-37 by John Wiley & Sons, Inc. Keywords: RNA r imaging r GFP r fluorescence r molecular beacon
INTRODUCTION Because genome sequences of many species have been documented completely over the past decade, elucidating the respective functions of gene products, such as RNAs, is extremely important. Numerous data related to RNAs and their functions are now available. Nevertheless, the information typically represents an average of a population of cells, which sometimes obscures the actual characteristics and behavior of RNAs. Consequently, the continued development of new tools and methods to probe RNAs is vital for additional elucidation of their localization and dynamics. This challenge necessitates single-cell-based analyses using fluorescent probes and microscopes. Fluorescence imaging technologies using green fluorescent protein (GFP) and spectral variants of GFP have been widely used to visualize molecules with high spatial and temporal resolution in live cells. Among the technologies, the split-GFP complementation and reconstitution methodologies have been most commonly applied toward elucidation of protein-protein interactions in living cells (Ozawa et al., 2000, 2003; Hu and Kerppola, 2003), wherein reassembly of conditional GFP fragments is facilitated by the direct interactions of two proteins. Based on this concept of GFP fragments reassembly, we have recently developed genetically encoded fluorescent probes to visualize a target mRNA in live cells (Ozawa et al., 2007). The RNA probes consist of split fragments of enhanced GFP (EGFP) or yellow fluorescent protein (Venus) (Nagai et al., 2002), each of which is connected with a sequence-specific RNA binding domain of PUMILIO (Pumilio homology domain; PUM-HD) (Fig. 1A). PUMILIO was discovered in D. melanogaster as a protein that regulates mRNA expression by binding to specific sequences in the 3’untranslated region (UTR) of target mRNAs. The PUMILIO family of proteins has been identified in many species including C. elegans, S. cerevisiae, and H. sapiens. The crystal structure of H. sapiens PUM-HD (HsPUM1-HD) in complex with its RNA revealed that it comprises eight sequence repeats (named from R1 to R8), which recognize a consensus sequence 5 -UGUAUAUA-3 (Wang et al., 2002). Each RNA base is recognized by three
Current Protocols in Chemical Biology 3: 27-37, February 2011 Published online February 2011 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/9780470559277.ch100152 C 2011 John Wiley & Sons, Inc. Copyright
Imaging RNA Using Genetically Encoded Probes
27 Volume 3
A C-terminal R8 mRNA
R7
N-terminal
R6 R5 R1
R4 R2
R3
B PUM mutant 2 (mPUM2)
PUM mutant 1 (mPUM1)
reconstituted EGFP (fluorescence)
mRNA
C-EGFP
N-EGFP ND6 mRNA (186–210) : 5 - AA UGAUGGUU GUCUU UGGAUAUA CU -3 binding proteins: mPUM1 mPUM2
C PUM1 repeat wild PUM-HD
RNA sequence
R8 Q1126 Y1123 N1122
R7 E1083 N1080 S1079
R6 Q1047 Y1044 N1043
R5 Q1031 R1008 C1007
R4 Q975 H972
R3 Q939 R936 C935
R2 Q903 Y900 N899
R1 Q867 R864 S863
5 -U1
G2
U3
A4
U5
A6
U7
A8-3
Figure 1 (A) Structure of the human PUM-HD complexed with RNA. The helical repeats are shown alternately blue and yellow, which are labeled as repeat 1(R1) to repeat 8(R8). Each repeat recognizes a specific base of RNA. (B) Basic principle of the RNA probes. Two RNA-binding domains of PUM are engineered to recognize specific sequences on a target mRNA (mPUM1 and mPUM2). In the presence of the target mRNA, mPUM1 and mPUM2 bind to their target sequences, bringing together the N- and C-terminal fragments of EGFP, resulting in functional reconstitution of the fluorescent protein. (C) RNA sequences of PUM-HD for RNA. The amino acids that interact with RNA bases are shown. The amino acids in square frames are necessary for stacking between upper and lower RNA bases. The other amino acids are for hydrogen bonds or van der Waals interactions.
Imaging RNA Using Genetically Encoded Probes
28 Volume 3
Current Protocols in Chemical Biology
conserved amino acids at specific positions in the second of three α helices in each repeat of HsPUM1-HD. Two amino acid side chains make hydrogen bonds or van der Waals interactions like the Watson-Crick edge of an RNA base, and the third amino acid side chain forms a stacking interaction with aromatic rings of the RNA bases. The recognition sequences can be tailored based on an eight-base sequence of target RNAs (Cheong and Hall, 2006); the construct is determined simply by changing the two residues in a repeat that interact with the Watson-Crick edge of the base. There is no need to change the third side chain used for stacking interactions. However, contributions of the 8-base repeat to the binding affinity are not equivalent (Wang et al., 2002). Although there is no reported systematic analysis of the affinity of PUM-HD-RNA complex, see Cheong and Hall (2006) for additional information on the design of PUM-HD mutants. The applicability of PUM-HD to technology was first demonstrated for imaging mitochondrial mRNA, specifically NADH dehydrogenase 6 (ND6) mRNA. The HsPUM1-HD (hereafter named PUM1) was designed to match the RNA sequence of ND6, which was connected with EGFP fragments (Fig. 1B). Upon interacting with RNA, PUM1 brings the fragments of EGFP into an orientation and proximity in which they are sufficiently close for the fragments to associate and fluoresce. Monitoring the fluorescence signals allows spatial and temporal analysis of mRNA localization in individual living cells. The basic concept has recently been applied for imaging of viral RNA genomes in plants (Tilsner et al., 2009). PUM1, fused to either the N-terminal or the C-terminal halves of split mCitrine, was engineered to recognize two closely adjacent eight-nucleotide sequences in the genomic RNA of tobacco mosaic virus (TMV). They demonstrated that most RNAs were sequestered in characteristic inclusion bodies known as viral replication complexes (VRCs). The PUM1-based approach can be adopted for various RNAs as structural understanding of the PUM-HD–RNA interaction increases in the future. Described herein is the basic experimental protocol of PUM1-based imaging of mitochondrial mRNA.
STRATEGIC PLANNING Design of Mutant PUM for Specific RNA Recognition The specificity and selectivity of engineered PUM1 to RNA sequences have been examined in vitro (Wang et al., 2002). PUM1 comprises eight tandem repeats (Fig. 1A). The RNA runs antiparallel to the protein such that nucleotides from U1 to A8 (5 U1 G2 U3 A4 U5 A6 U7 A8 -3 of PUM1) are recognized individually by the repeats from R8 to R1, respectively. Each repeat recognizes a single RNA base through three conserved side chains of amino acids; two chains form hydrogen bonds or van der Waals interactions with the Watson-Crick edge of an RNA base, and a third side chain stacks with the same base and/or the preceding base (Fig. 1C). Construction of a mutant PUM1 requires only a change of the two residues in a repeat; there is no need to change the third side chain used for stacking interactions. The two amino acid residues that recognize an RNA base are named B1 and B5 from the N-terminal end (e.g., B5 Q867 and B1 R864 interact with A8). When two amino acids in a repeat of PUM1 are mutated to match an RNA base (X), the following combination (X: B1, B5) is recommended: (U: Asn, Gln), (G: Ser, Glu) and (A: Cys or Ser, Gln). No information is currently available on how to mutate B1 and B5 to match the RNA base T. The contributions of the eight repeats to RNA binding affinity are not equivalent. The most important factor is the consensus sequences of PUM1 beginning with 5 -U1 G2 U3 -3 , which recognize R8-R6. The specificity is sensitive to mutations and it is therefore recommended that the consensus sequence not be changed, if possible. The R5 recognizes a purine base (A or G). The repeats from R4 to R1 bind to the sequence, 5 -U5A6U7A8-3 .
Imaging RNA Using Genetically Encoded Probes
29 Current Protocols in Chemical Biology
Volume 3
Recognition of each repeat to the base is also sensitive to mutations. In addition to the general rules, many empirical data have been reported (Cheong and Hall, 2006; Miller et al., 2008; Lu et al., 2009). The empirical data will be useful for the design of PUM1 mutants. BASIC PROTOCOL 1
CHARACTERIZATION OF RNA PROBES Before acquiring images of a target RNA with the RNA probes, it is important to examine whether the mutated PUM binds directly to the RNA in the cells of interest. For this purpose, it is recommended to use immunoprecipitation (Bonifacino et al., 2001) and reverse transcription PCR (RT-PCR; Beverly, 2001). In addition, it is recommended to examine the binding affinity of the engineered PUM to a target sequence using an electrophoretic mobility shift assay (gel-shift assay; Buratowski and Chodosh, 2001) to confirm the specificity of the RNA sequence recognition. Representative methods of immunoprecipitation and RT-PCR are described below. The experimental procedure is illustrated in Figure 2.
cultured cells transfect plasmids: GN-mPUM1, VN-mPUM1, mPUM2-GC, mPUM2-VC incubate for 48 hr cells expressing the probes lyse the cells with lysis buffer A centrifuge 3 min at 15,000
precipitate (discarded)
g
supernatant add anti-flag antibody or anti-GFP antibody incubate for 1 hr at 4°C on a rotator mix ProteinSepharose 4FF beads incubate for 1 hr at 4°C on a rotator wash the beads four times with lysis buffer B add 25 to 50 l of 2
loading buffer
boil at 95° to 100°C for 5 min centrifuge 3 min at 15,000
precipitate (discarded)
g
supernatant (including RNA) convert the extracted RNA into cDNA supernatant (including cDNA) run PCR using the cDNA as a template
Imaging RNA Using Genetically Encoded Probes
PCR products (the size analysis)
Figure 2
Flow chart of RNA-probe characterization.
30 Volume 3
Current Protocols in Chemical Biology
BamHI
Not I
Xho I
EcoRI
GN-mPUM1 MTS
BamHI
EGFP-N (1 – 157 a.a.)
Not I
mPUM1
Xho I
FLAG
EcoRI
VN-mPUM1 MTS
Venus-N (1 – 154 a.a.)
XbaI
Not I
mPUM1
Xho I
FLAG
XbaI
mPUM2-GC MTSX3
XbaI
mPUM2
Not I
EGFP-C (158 – 238 a.a.)
Xho I
XbaI
mPUM2-VC MTSX3
BamHI
mPUM2
Venus-C (155 – 238 a.a.)
Not I
Xho I
MTS-DsRed-Ex MTS
DsRed-Express (full length)
Figure 3 Constructs of the plasmids. FLAG, FLAG epitope; MTS, matrix-targeting signal derived from subunit VIII of cytochrome C oxidase. The cDNA is inserted into an expression vector.
Materials HeLa cells DMEM with 10% FBS Plasmids: GN-mPUM1, VN-mPUM1, mPUM2-GC, mPUM2-VC, and MTS-DsRed-Ex, cloned into a mammalian expression vector, pcDNA3.1 (+) (Invitrogen) (Fig. 3) Lipofectamine 2000 (Invitrogen) Lysis buffer A (see recipe) Mouse monoclonal anti-Flag antibody (Sigma) or mouse monoclonal anti-GFP antibody (Roche) Protein Sepharose 4FF beads (GE healthcare) Lysis buffer B (see recipe) 2× loading buffer (see recipe) cDNA synthesis kit (Invitrogen) Primers: ND6F (5 -ATGATGTATGCTTTGTTTCT-3 ) and ND6R (5 -CCTATTCCCCCGAGCAATCT-3 ) for mitochondrial ND6 mRNA, and ND1F (5 -ATACCCATGGCCAACCTCCT-3 ) and ND1R (5 -TTAGGTTTGAGGGGGAATGC-3 ) for controls 6-well plates 37◦ C cell culture incubator 1.5-ml tubes Rotator UV transilluminator Additional reagents and equipment for agarose gel electrophoresis (Voytas, 2001)
Imaging RNA Using Genetically Encoded Probes
31 Current Protocols in Chemical Biology
Volume 3
NOTE: The plasmids and more information on their cDNA sequences and enzyme sites may be obtained upon request from the authors’ laboratory (e-mail:
[email protected]). 1. Seed 1.0 × 106 HeLa cells in 6-well plates and culture the cells in DMEM with 10% FBS at 37◦ C for 24 hr. 2. When the cells are 90% confluent, transfect the cells with plasmids GN-mPUM1, VN-mPUM1, mPUM2-GC, or mPUM2-VC using Lipofectamine 2000 according to the manufacturer’s instructions. 3. Incubate the cells 48 hr at 37◦ C. 4. Lyse the cells by adding 1 ml of ice-cold lysis buffer A. Shake the plates gently for 3 min on ice and then collect the lysates in a 1.5-ml tube. 5. Centrifuge the cells 3 min at 15,000 × g, 4◦ C, and collect the supernatant. 6. Add anti-flag antibody (1/1000 v/v) or anti-GFP antibody (1/2000 v/v) into the solution and incubate for 1 hr at 4◦ C with a rotator. 7. To absorb the immunoprecipitates, add 10 μl of protein Sepharose 4FF beads to the solution, and then incubate for 1 hr at 4◦ C. 8. Wash the beads four times, each time with 1 ml lysis buffer B at 4◦ C. 9. To denature the protein and mRNA and separate them from the beads, add 25 to 50 μl of 2× loading buffer and boil 5 min at 95◦ C. 10. Centrifuge the solution 3 min at 15,000 × g, 4◦ C, and collect the supernatant. 11. Convert the RNAs in the supernatant into cDNA using a cDNA synthesis kit (Invitrogen) according to the manufacturer’s protocol. 12. Perform PCR using the 1 μl of cDNA as a template. Select a pair of the following primers, ND6F (5 -ATGATGTATGCTTTGTTTCT-3 ) and ND6R (5 -CCTATTCCCCCGAGCAATCT-3 ) for mitochondrial ND6 mRNA, and ND1F (5 ATACCCATGGCCAACCTCCT-3 ) and ND1R (5 -TTAGGTTTGAGGGGGAATGC-3 ) as a control experiment. In the control experiments, run PCR using cDNAs prepared from total RNA in the HeLa cells as a template. Use the following PCR conditions: 30 cycles: 30 sec 30 sec 4 min
98◦ C (denaturation) 58◦ C (annealing) 72◦ C (extension).
13. Load 10 μl of the PCR products and molecular weight marker (x) onto a 2% agarose gel, which includes ethidium bromide. Electrophorese for 10 min at 100 V (Voytas, 2001). 14. Transfer the gel onto a UV transilluminator. Measure the size of the amplified DNA at 525 bp and 957 bp, which corresponds to the ND6 mRNA and ND1 mRNA, respectively. Confirm that amplified DNAs at both 525 bp and 957 bp are obtained from the total RNAs of HeLa cells, whereas only amplified DNA at 525 bp is obtained from the immunoprecipitated RNAs. BASIC PROTOCOL 2
IMAGING ENDOGENOUS RNA USING GENETICALLY ENGINEERED FLUORESCENT PROBES Described below is a procedure for imaging mitochondrial RNA using HeLa cells. The procedure can be adapted for many other cell types, including plant cells, by replacing the transfection method with a conventional one.
32 Volume 3
Current Protocols in Chemical Biology
A
B
Mito Tracker
C
ND6 mRNA
D
DAPI
E
ND6 mRNA Mito Tracker
ND6 mRNA DAPI
Figure 4 Fluorescence images of HeLa cells expressing GN-mPUM1 and mPUM2-GC stained with MitoTracker and DAPI: (A) Localization of mitochondria, (B) reconstituted EGFP, and (C) mtDNA. Panels D and E show their merged images. Bar, 5 μm. The insets are enlarged images of the boxed region of (A) (bar, 1.2 μm). White arrows indicate colocalization of mtDNA and ND6 mRNA.
Materials HeLa cells DMEM with 10% FBS Plasmids: GN-mPUM1, mPUM2-GC, mPUM2-VC, and MTS-DsRed-Ex, cloned into a mammalian expression vector, pcDNA3.1 (+) (Invitrogen) (Fig. 3) Lipofectamine 2000 (Invitrogen) DAPI (Invitrogen) MitoTracker (Invitrogen) HBSS (Sigma) containing 5% FBS (Invitrogen) 10-cm culture dishes 37◦ C cell culture incubator 3.5-cm glass-bottom dish Inverted fluorescence microscope, IX71 (Olympus), equipped with 100×, 1.40-NA oil objective, a 100-W mercury arc lamp for illumination and 50-W xenon lamp for bleaching with a double lamp-house system EM-CCD camera (iXon, ANDOR Technology) to acquire cell images MetaMorph software
Imaging RNA Using Genetically Encoded Probes
33 Current Protocols in Chemical Biology
Volume 3
NOTE: The plasmids and more information on their cDNA sequences and enzyme sites may be obtained upon request from the authors’ laboratory (e-mail:
[email protected]). 1. Seed HeLa cells (1 × 106 cells) onto a 10-cm dish 24 hr before transfection, and incubate the cells in DMEM with 5% FBS at 37◦ C. 2. Mix three plasmids to yield a 2.5-μl solution of 1 μl GN-mPUM1, 1 μl mPUM2-GC, and 0.1 μl MTS-DsRed-Ex. 3. Transfect the three-plasmid mixture into the HeLa cells using Lipofectamine 2000 according to the manufacturer’s instructions. 4. Incubate the cells for 12 hr at 37◦ C and then seed onto a 3.5-cm glass-bottom dish. Incubate the cells in DMEM with 5% FBS for an additional 12 hr at 37◦ C. 5. To visualize DNAs and mitochondria, add 10 μM DAPI and 1 μM MitoTracker in the culture medium. Incubate the cells 5 min at 37◦ C. 6. Aspirate the medium and replace with 1 ml HBSS containing 5% FBS. 7. Place the glass-bottom dish containing the HeLa cells on an inverted fluorescence microscope equipped with a mercury arc lamp and the 100×, 1.40 numerical aperture (NA) oil objective. Adjust the excitation filter, emission filter, and a dichroic mirror as needed. Set the cooling temperature of EM-CCD camera at –50◦ C. 8. Acquire images with the CCD camera and analyze using MetaMorph software. Typical results are shown in Figure 4.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps.
Loading buffer, 2× 10% 2-mercaptoethanol 4% (w/v) SDS 250 mM Tris·Cl, pH 6.8 Store up to 6 months at 4◦ C Lysis buffer A 100 mM NaCl 1mM EDTA 10 mM NaF 2 mM sodium orthovanadate 1 mM phenylmethylsulfonyl fluoride (PMSF; add PMSF right before use) 10 μg/ml pepstatin 10 μg/ml leupeptin 10 μg/ml aprotinin 0.1% Triton X-100 50 mM Tris·Cl, pH 7.4 Store up to 1 month at 4◦ C Lysis buffer B
Imaging RNA Using Genetically Encoded Probes
10% (w/v) SDS 250 mM Tris·Cl, pH 6.8 Store up to 6 months at 4◦ C
34 Volume 3
Current Protocols in Chemical Biology
COMMENTARY Background Information The RNA imaging method using PUMbased probes described here has many advantages over other methods. The PUM-based probes are genetically encoded; therefore, the cDNA encoding the probes can be introduced into living cells or inserted into chromosomal genes without damage. In addition, the principle is based on complementation of fluorescent protein fragments, which are able to emit fluorescence only in the presence of a target RNA. High-contrast fluorescence images of localized RNA can be acquired because of low background fluorescence. In addition, the method accommodates probes designed to match the sequence of target mRNAs by simple amino acid mutations. The probes currently most commonly used are nucleic acid probes, termed molecular beacons (Tyagi and Kramer, 1996). The molecular beacons are hairpin-shaped DNAs with an internally quenched fluorophore whose fluorescence is restored when they bind to a target nucleic acid sequence. The specificity for a target RNA is defined by the complementary sequence of the nucleotides of the probes, typically using ∼15 nucleotides. Therefore, the specificity is relatively high and the design of the molecule is quite simple. However, microinjection techniques are often required for inserting the probes into living cells, which causes significant damage to the cells. It is also difficult to target the beacons to a distinct intracellular organelle. Another method for RNA imaging is the use of GFP-tagged MS2 protein (Bertrand et al., 1998). A drawback of this approach is that the expressed probes emit luminescence. Because of the high background fluorescence, it is often difficult to distinguish GFP-tagged protein bound to mRNA from unbound tagged protein. In addition, the approach requires an injection of target artificial RNAs, indicating that it is impossible to visualize endogenous RNAs in living cells. From this comparison of methods, it can be concluded that the PUMbased probes are generally and widely applicable for imaging RNAs in living cells (Tyagi, 2007).
Critical Parameters and Troubleshooting Validation of imaging data is important for confirming RNA localization in living cells. Structures of fluorescent proteins reconstituted from complementary fragments are very stable
and their fragments cannot be dissociated spontaneously. The reconstituted proteins remain fluorescent even after degradation of RNA so the fluorescent proteins themselves do not indicate precise location of target RNA. To overcome this issue, it is recommend to use VN-mPUM1 and mPUM2-VC, which encode split fragments of the Venus protein. Venus emits fluorescence as strong as EGFP and is suitable for bleaching experiments. In order to visualize only the Venus that makes a complex with RNA, the fluorescence recovery after photobleaching (FRAP) technique is used. The fluorescence of Venus is bleached by irradiation with light at 480 nm, and then fluorescence images are acquired every few minutes. Fluorescence of Venus is recovered gradually by reconstitution of the Venus fragments complexed with target RNA. If the time for data acquisition is faster than the rate of RNA decomposition, the localization of reconstituted Venus indicates accurate localization of RNA in living cells. In one study, a signal sequence of subunit VIII of cytochrome c oxidase (MSVLTPLLLRGLTGSARRLPVPRAKIHSL) was used to target probes into mitochondria (Hegeman et al., 1995). GN–mPUM1 localized exclusively to the mitochondrial matrix, whereas three repeats of MTS were required for targeting mPUM2-GC; use of only one or two MTS sequences failed to localize mPUM2-GC to the mitochondrial matrix. Short linker amino acids (AAA) were used to connect the signal sequences to the probes. Other signal sequences may be used for targeting the probes to the nucleus or chloroplasts (Emanuelsson and von Heijne, 2001). When the signal sequence is used, it is then necessary to verify localization of the probes inside the cells. The length of the linker between the split fragments of EGFP and mutant PUM-HD is variable. There is no data available on the effect of the linker length on efficiency of fluorescence recovery. When an expected fluorescence intensity is not obtained, probes with different linker lengths should be constructed and FRAP behavior examined. In the design of the probes, the exact distance between the two probes may not be critical for efficient complementation of fluorescent proteins. In the case of mitochondrial ND6 mRNA, the RNA sequence that bridges mPUM1 and mPUM2 is five bases (GUCUU). For imaging plant virus, seven bases (AGUUUUU) were used for bridging the mutant
Imaging RNA Using Genetically Encoded Probes
35 Current Protocols in Chemical Biology
Volume 3
PUM-HDs. Because a single strand of RNA is flexible, it is likely that the RNA sequence between the two PUM-HD recognition sequences does not affect the binding of RNA to PUM-HD.
Anticipated Results Localization of ND6 mRNA in mitochondria is confirmed by staining mitochondria with MitoTracker Red (CMXRos) and mtDNA with DAPI (Fig. 4). The fluorescence intensity of reconstituted EGFP (Fig. 4B) is very weak owing to the low amounts of mtRNA; therefore, the exposure time of the CCD camera may be set to >300 msec. In contrast, each image of mitochondria (Fig. 4A) and mtDNA (Fig. 4C) required <50 msec exposure. The merged image of ND6 mRNA and mitochondria (Fig. 4D) indicates that ND6 mRNA is localized in all mitochondria. In addition, several foci originating from concentrated EGFP were observed (inset in Fig. 4D). The localization is the same as that of mtDNA shown in Figure 4E, suggesting that mtRNA encoding ND6 accumulates at the mtDNAs. Thus, the multicolor labeling of organelles, DNAs, or specific proteins provides valuable information about the localization and function of the target mRNA.
Time Considerations Plasmid construction, including insertion of mutations into PUM1, can be performed in 2 weeks. For characterization of the probes by immunoprecipitation and subsequent RTPCR, 3 days are needed for plasmid transfection and protein expression in living cells. After obtaining the cell lysates, the immunoprecipitation and reverse transcription can be performed in 24 hr to yield cDNA. Once started, do not interrupt or stop the immunoprecipitation and reverse transcription reactions. The cDNAs can be stored up to 6 months at −80◦ C until needed for PCR. In the RNA imaging using the probes, the cells expressing the probes can be prepared up to 2 days in advance. The cells can be used immediately for acquisition of fluorescence images by fluorescence microscopy.
Acknowledgments
Imaging RNA Using Genetically Encoded Probes
This work was supported by grants from the Core Research for Evolutional Science and Technology (CREST) of the Japan Science and Technology (JST) and the Ministry of Education, Science, and Culture, Japan.
Literature Cited Bertrand, E., Chartrand, P., Schaefer, M., Shenoy, S.M., Singer, R.H., and Long, R.M. 1998. Localization of ASH1 mRNA particles in living yeast. Mol. Cell 2:437-445. Beverly, S.M. 2001. Enzymatic Amplification of RNA by PCR (RT-PCR). Curr. Protoc. Mol. Biol. 56:15.5.1-15.5.6. Bonifacino, J.S., Dell’Angelica, E.C., and Springer, T.A. 2001. Immunoprecipitation. Curr. Protoc. Mol. Biol. 48:10.16.1-10.16.29. Buratowski, S. and Chodosh, L. A. 2001. Mobility shift DNA-binding assay using gel electrophoresis. Curr. Protoc. Mol. Biol. 36:12.2.1-12.2.11. Cheong, C.G. and Hall, T.M. 2006. Engineering RNA sequence specificity of Pumilio repeats. Proc. Natl. Acad. Sci. U.S.A. 103:1363513639. Emanuelsson, O. and von Heijne, G. 2001. Prediction of organelle targeting signals. Biochim. Biophys. Acta 1541:114-119. Hegeman, A.D., Brown, J.S., and Lomax, M.I. 1995. Sequence of the cDNA for the heart/muscle isoform of mouse cytochrome c oxidase subunit VIII. Biochim. Biophys. Acta 1261:311-314. Hu, C.D. and Kerppola, T.K. 2003. Simultaneous visualization of multiple protein interactions in living cells using multicolor fluorescence complementation analysis. Nat. Biotechnol. 21:539545. Lu, G., Dolgner, S.J., and Hall, T.M. 2009. Understanding and engineering RNA sequence specificity of PUF proteins. Curr. Opin. Struct. Biol. 19:110-115. Miller, M.T., Higgin, J.J., and Hall, T.M. 2008. Basis of altered RNA-binding specificity by PUF proteins revealed by crystal structures of yeast Puf4p. Nat. Struct. Mol. Biol. 15:397-402. Nagai, T., Ibata, K., Park, E.S., Kubota, M., Mikoshiba, K., and Miyawaki, A. 2002. A variant of yellow fluorescent protein with fast and efficient maturation for cell-biological applications. Nat. Biotechnol. 20:87-90. Ozawa, T., Nogami, S., Sato, M., Ohya, Y., and Umezawa, Y. 2000. A fluorescent indicator for detecting protein-protein interactions in vivo based on protein splicing. Anal. Chem. 72:51515157. Ozawa, T., Sako, Y., Sato, M., Kitamura, T., and Umezawa, Y. 2003. A genetic approach to identifying mitochondrial proteins. Nat. Biotechnol. 21:287-293. Ozawa, T., Natori, Y., Sato, M., and Umezawa, Y. 2007. Imaging dynamics of endogenous mitochondrial RNA in single living cells. Nat. Methods 4:413-419. Tilsner, J., Linnik, O., Christensen, N.M., Bell, K., Roberts, I.M., Lacomme, C., and Oparka, K.J. 2009. Live-cell imaging of viral RNA genomes using a Pumilio-based reporter. Plant J. 57:758770.
36 Volume 3
Current Protocols in Chemical Biology
Tyagi, S. 2007. Splitting or stacking fluorescent proteins to visualize mRNA in living cells. Nat. Methods 4:391-392. Tyagi, S. and Kramer, F.R. 1996. Molecular beacons: probes that fluoresce upon hybridization. Nat. Biotechnol. 14:303-308. Voytas, D. 2001. Agarose gel electrophoresis. Curr. Protoc. Mol. Biol. 51:2.5A.1-2.5A.9. Wang, X., McLachlan, J., Zamore, P.D., and Hall, T.M. 2002. Modular recognition of RNA by a human pumilio-homology domain. Cell 110:501-512.
Imaging RNA Using Genetically Encoded Probes
37 Current Protocols in Chemical Biology
Volume 3
Development of In-Cell Western Assays Using Far-Red Fluorophores Nathan J. Moerke1 and Gregory R. Hoffman2 1 2
Harvard Medical School, ICCB-Longwood Screening Facility, Boston, Massachusetts Harvard Medical School, Department of Cell Biology, Boston, Massachusetts
ABSTRACT The in-cell western (ICW) technique is a cell-based immunoassay method for quantitative measurement of protein expression or phosphorylation levels that can be used for both small molecule and siRNA screening. The method involves growth of cells in microplates, fixation, permeabilization, and staining with specific antibodies and/or cell labeling dyes. ICW assays take advantage of the properties of near-infrared dyes to achieve higher signal-to-noise ratios than are possible for methods utilizing fluorophores in the visible range of the spectrum, and typically involve measurements using two fluorescent channels: one to measure levels of the target of interest, and one to measure total cell number for normalization. The ICW method is readily adaptable to high-throughput format and has been successfully used with a variety of targets and cell lines. The protocols in this unit describe an ICW procedure for quantitative measurement of rpS6phosphorylation as an endpoint for monitoring mTORC1 signaling in HeLa cells. This assay can be used for small molecule or siRNA screening, and with modification is C 2011 by adaptable to other cell lines and targets.Curr. Protoc. Chem. Biol. 3:39-52 John Wiley & Sons, Inc. Keywords: in-cell western r ICW r high-throughput screening r mTOR r rpS6 r rapamycin r phosphorylation
INTRODUCTION This unit describes a procedure for quantitative measurement of rpS6-phosphorylation as an endpoint for monitoring mTORC1 signaling using the “in-cell western” (ICW) technique. This procedure can readily be adapted to the measurement of the phosphorylation states or total levels of other proteins. The ICW is a cell-based assay format carried out in 96-well or, more commonly, 384-well microplates (Wong, 2004; Chen et al., 2005). It is similar to conventional immunofluorescence assays in that it involves growing cells in microplates, fixation, permeabilization, and probing with specific antibodies to monitor protein levels. A key feature of the ICW is that it uses specialized secondary antibodies conjugated to dyes that fluoresce in the near-IR region of the spectrum (at 700 or 800 nm). Microplates can then be read by scanners, such as the Odyssey or Aerius Infrared Imaging Systems (LI-COR Biosciences), that contain pairs of near-infrared lasers and detectors for excitation and detection of fluorescent signals at these wavelengths. The use of nearIR dyes provides high signal-to-noise ratios due to a number of advantages these have over shorter wavelength dyes that fluoresce in the visible part of the spectrum: reduced light scattering, less signal loss due to absorbance, and reduced assay interference due to autofluorescence from cellular components and microplates. Another feature of the ICW is the simultaneous use of the 700-nm and 800-nm wavelengths to either measure the relative levels of two targets of interest or, more typically, to measure total cell number for normalization of the target measurement. There are three main methods that can be used for normalization. The first involves detection of In-Cell Western Assays Current Protocols in Chemical Biology 3: 39-52, March 2011 Published online March 2011 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/9780470559277.ch100153 C 2011 John Wiley & Sons, Inc. Copyright
39 Volume 3
a generally expressed housekeeping protein, such as beta-actin, using another primary and near-IR dye-labeled secondary antibody. Alternatively, if the purpose of the assay is to measure the phosphorylated form of a protein, then a primary antibody detecting total levels of that protein can be used. The second normalization method uses a DNA-intercalating dye fluorescing in the near-IR region, such as DRAQ5 or TO-PRO3. Finally, normalization can be achieved by covalent labeling of cellular proteins with an amine-reactive dye. Dyes of this type label bulk protein by forming stable conjugates with lysine residues. An example of this type of dye, used in the procedure described here, is Alexa-680 N-hydroxysuccinimidyl (NHS) ester (see the Commentary for a more detailed discussion of the factors influencing the choice of normalization strategy). The ICW method is suitable for laboratory automation and has a number of advantages over other technologies that make it particularly suitable for high-throughput screening. One of these is the speed with which plates can be imaged using an infrared scanner such as the Aerius. Using typical settings, a single 384-well plate can be scanned and analyzed in 5 to 10 min on this instrument. In comparison, it will typically take 30 min to 2 hr to image a plate using visible fluorophores on a high-throughput fluorescence microscope. Furthermore, since ICW assay data are acquired on a per-well basis rather than a per-cell basis, analysis time is significantly shorter than with high content imaging. Although high-throughput microscopy potentially provides information on subcellular localization due to its higher resolution, this may not be informative for proteins (such as rpS6) that have a uniform cytosolic distribution. We have adopted a strategy that takes advantage of the higher throughput and lower cost of the ICW assay for running primary screens, and then collecting more detailed high content data on screening hits using focused secondary assays. Another advantage of the ICW method is that it is often possible to use lower antibody concentrations due to the high sensitivity and low background of near-IR fluorophores. For a large-scale screening project, this can result in significant cost savings. Finally, the cell-based nature of the ICW assay makes it suitable for both siRNA and small molecule-based screens.
STRATEGIC PLANNING
In-Cell Western Assays
The ICW method is well suited for the study of regulation of protein expression or signal transduction pathways, and has been used successfully for numerous protein markers in a wide range of cell lines (Olive, 2004). The first step in ICW assay development involves identification of an appropriate protein target for measurement and a suitable antibody or antibodies for detection of this target. Next, a cell number normalization method must be chosen. When beginning a new screening project, small molecule compounds that are known to inhibit the pathway of interest provide a good starting point for assay development. In addition, it is useful to test inhibitors of unrelated pathways to evaluate assay specificity. Typically, dose-response curves are determined for small molecule controls in the ICW assay format, and IC50 values calculated from these are compared to previously known literature values. For siRNA screening assay development, the next step should be testing the assay using control reagents that might downregulate (or, possibly, upregulate) the assay readout. siRNA controls are chosen to knock down known genes in the pathway of interest, as well as genes that, when knocked down, cause cytotoxicity (in order to validate the cell normalization method). We find that it is useful to include multiple assay-specific positive controls during this phase of optimization, as different cell lines might be more or less sensitive to particular positive controls. This phase of assay development may involve optimization of the transfection protocol and assay conditions (see Critical Parameters and Troubleshooting) until the controls are working reproducibly.
40 Volume 3
Current Protocols in Chemical Biology
Once the assay has been validated with appropriate controls, the next step is to evaluate its statistical robustness using a factor referred to as the Z that is given by Equation 1 (Zhang et al., 1999):
Z′ = 1−
(3
pos
+3
pos
−
neg
)
neg
Equation 1
Here, σ and μ represent the standard deviations and means of the positive and negative controls, respectively. Typically, the Z value is calculated based on a single plate that is approximately half filled with positive and half filled with negative control wells. An assay with a Z of at least 0.5 is recommended for high-throughput screening, and one with a Z of 0.7 or higher is considered an excellent assay. Multiple iterations of assay optimization may be required until the desired Z factor is achieved. Once this is done, the assay can be considered ready for high-throughput screening. The protocols in this unit describe an ICW assay designed for high-throughput screening for inhibitors of mTORC1 signaling. The mTORC1 pathway is a central regulator of mammalian cell growth (Wullschleger et al., 2006). Hyperactivation of this signaling network is found in nearly all cancers, and it plays a causative role in other human diseases, such as tuberous sclerosis complex (TSC) and lymphangioleiomyomatosis (LAM) (Inoki et al., 2005; Choo and Blenis, 2006). The ribosomal protein S6 (rpS6) is a component of the protein translation control machinery that is phosphorylated on multiple serine residues in response to mTORC1 activation. Quantitation of rpS6 phosphorylation provides a reliable method for monitoring the activation state of the pathway and is the basis for this ICW assay. The procedures are suitable for screening of both small molecule and siRNA libraries to identify inhibitors of mTORC1 signaling, and pilot screens using HeLa cells of both a known bioactive compound library and a kinase/phosphatase siRNA library have been published (Hoffman et al., 2010). Basic Protocol 1 describes a smallmolecule treatment procedure, and Basic Protocol 2 an siRNA transfection procedure. Following one of these protocols, the assay continues with fixation, permeabilization, staining, and imaging, as described in Basic Protocol 3. Assay validation data generated using this protocol for both siRNA and small molecule controls and Z determination is shown in Figures 1 and 2, respectively. This assay procedure can be adapted to the measurement of other targets using different primary antibodies and other modifications, such as changes in fixation, permeabilization, and blocking conditions. We expect that the normalization method used will be broadly applicable for many cell lines and proteins.
COMPOUND TREATMENT OF CELLS FOR IN-CELL WESTERN ASSAY This protocol describes a compound treatment method suitable for ICW assays and used in the assay for inhibitors of the mTORC1 pathway. This protocol was used for doseresponse testing of rapamycin (Fig. 1A) and LY294002 (Fig. 1B) as positive controls for inhibition of mTORC1 signaling. As a negative control, a dose-response for the MEK inhibitor UO126 was used (Fig. 1C), showing that the MAP kinase pathway does not regulate rpS6 phosphorylation under these screening conditions. A Z value for small molecule screening was determined using 20 ng/ml rapamycin as the positive control, and DMSO as the negative control (Fig. 1D). The assay was then used in a pilot screen of a known bioactives library (Hoffman et al., 2010).
BASIC PROTOCOL 1
In-Cell Western Assays
41 Current Protocols in Chemical Biology
Volume 3
A 1 Relative phospho–S6 (arbitrary units)
0.9 0.8 0.7 0.6
IC50
[Rap]
224 pM
0.5 0.4
10
9
8
Log[Rap] (M)
B
Overlay
11
Cell #
12
p-S6
0.3
C 0.9 1 Relative phospho–S6 (arbitrary units)
Relative phospho–S6 (arbitrary units)
0.8 0.7 0.6 0.5
IC50
2.5 M
0.4 0.3
0.8 0.6 0.4 0.2
0.2 7.5
7
6.5
6
5.5
5
4.5
0 10
4
9
D Not used 20 ng/ml Rapamycin
DMSO Not used
*
*
No Cells
Figure 1
No 1°
8
7
6
5
Log[UO126] (M)
Relative phospho–S6 intensity (arbitrary units)
Log[LY294002] (M)
8 7 6 5 4
Z
3
0.80
2 1 0
0
50
100
150
200
Well number
(legend appears on next page)
Materials
In-Cell Western Assays
Dimethyl sulfoxide (DMSO) Compounds being studied: e.g., Rapamycin, LY294002, and UO126 (Sigma) Dishes of HeLa cells Trypsin/EDTA (Invitrogen, cat. no. 25200-056) DMEM (CellGro, cat. no. 10-013-CV)
42 Volume 3
Current Protocols in Chemical Biology
Fetal bovine serum (FBS) (PAA Laboratories, cat. no. A15-301) Penicillin/streptomycin (CellGro, cat. no. 30-002-CI) Tissue culture hood Polypropylene 384-well compound storage plates (Thermo Scientific, cat. no. AB-1056 ) Pin transfer apparatus (V&P Scientific, http://www.vp-scientific.com) (Rudnicki and Johnston, 2009) Matrix WellMate (or other automated liquid dispenser for multiwell plates) (Rudnicki and Johnston, 2009) WellMate manifold (sterile) Black opaque 384-well tissue culture microplates (Corning, cat. no. 3712) Standard tissue culture incubator capable of maintaining an environment of 37◦ C and 5% CO2 1. Before the experiment, prepare stock plates of compounds dissolved in DMSO for testing of control small molecules in the assay, determination of the Z value for the assay, and/or screening. A sample plate map for the stock compounds is shown in Figure 1D. Compound stocks are prepared in polypropylene 384-well storage plates. The stock plate should contain compounds at 300× the desired final concentration in each well (assuming a pin transfer volume of 100 nl and an assay volume of 30 μl). Small molecule stock plates can be stored long-term at −20◦ C.
2. Prepare the HeLa cell suspension as follows:. a. Trypsinize the contents of one or more dishes of HeLa cells by first aspirating the medium from the plates and then briefly washing with sterile PBS. After the brief wash, add 2 ml 0.25% trypsin/EDTA and incubate 5 min at 37◦ C. Add 10 ml complete medium (to stop the trypsinization), centrifuge 5 min at 1000 × g, aspirate the supernatant, and resuspend the cell pellet in fresh medium. b. Resuspend the cells at a final density of 33,333 cells/ml in DMEM, containing 10% FBS and 1% penicillin/streptomycin. Figure 1 (appears on previous page) Small molecule validation data. (A) HeLa cells were treated with increasing doses of rapamycin in a 384-well format following the procedure described in Basic Protocol 1. The cells were then fixed and stained for analysis on the LI-COR Aerius instrument using the Alexa-680 succinimidyl ester as a counter stain for cell number and the phospho-rpS6 antibody followed by an IRDye-800CW secondary antibody for quantitation of rp-S6 phosphorylation, following the procedure described in Basic Protocol 3. The dose-dependent inhibition of rpS6-staining was fit with a standard four-parameter model giving an IC50 of 224 pM. Each data point is an average of four independent experiments. The raw image from the LI-COR Aerius scanner is shown on the right. (B) A dose response experiment was performed using the same procedure as in A but with the PI3-kinase inhibitor, LY294002, showing an IC50 of 2.5 μM. (C) A dose response using the MEK inhibitor UO126 was performed, showing no appreciable inhibition of rpS6-phosphorylation across a range of concentrations. (D) The statistical reproducibility of the assay for small molecule screening was determined by calculating a Z-factor for the assay comparing 20 ng/ml rapamycin to DMSO control in a 384-well format. The raw data for this assay is shown in the left panel, with the phospho-rpS6 signal (green) overlaid with the Alexa-680 succinimidyl ester cell number stain (red). The quantitation of the raw data is shown in the right panel and it was used to calculate a Z-factor (Z ) of 0.80 for this experiment. The asterisks (*) denote untreated wells that were not used in the Z-factor determination. In the assay plate used for the Z determination, the wells in rows C-H and columns 3-22 contained 20 ng/ml rapamycin as a positive control for inhibition of mTORC1 activity and rpS6-phosphorylation. The wells in rows I-N and columns 3-22 contained DMSO as a negative control. No cells were plated in column 1, as a negative control for staining in both the 700- and 800-nm channels. In column 24, primary antibody was omitted, as a control to determine the level of nonspecific staining in the 800-nm channel.
In-Cell Western Assays
43 Current Protocols in Chemical Biology
Volume 3
Cells are grown to 70% to 80% confluency in 15-cm plates (BD Falcon, cat. no. 353025). This cell suspension will give a final plating density of 1000 cells/well when 30 μl of cell suspension are delivered to the plates in step 3.
3. Add 30 μl HeLa cell suspension to each well of the black opaque tissue culture microplates using the WellMate, leaving 1 column empty for the no-cells control wells. Transfer the plates into the tissue culture incubator and incubate at 37◦ C for 24 hr to allow the cells to adhere to the plates. 4. If necessary, thaw the small molecule stock plates 1 hr prior to transferring the compounds to plates. 5. After 24 hr, remove the plates from the incubator, and pin transfer the library compounds to plate using the pin transfer robot and a 100-nl pin array. 6. Incubate the plates for 2 hr at 37◦ C in a tissue culture incubator. This incubation time is optimized for this assay system, monitoring rpS6 phosphorylation. For other assay systems, monitoring other proteins, the incubation time with control or test compounds will vary (e.g., between 2 and 24 hr) and should be optimized during assay development.
7. Proceed to Basic Protocol 3. BASIC PROTOCOL 2
siRNA TRANSFECTION OF CELLS FOR IN-CELL WESTERN ASSAYS This protocol describes an siRNA transfection method that can be used with the ICW assay to screen for inhibitors of the mTORC1 pathway. This protocol was used for testing siRNA controls to validate the assay (Fig. 2A). siRNAs directed against the genes mTOR, PDK1, S6K1, and S6K2 were used to knock down mTORC1 pathway signaling, and siRNAs directed against PLK1 were used to kill cells. A Z factor for this siRNA screening assay was determined using siRNA directed against S6K1 and S6K2 as the positive controls, and a nontargeting siRNA as the negative control (Fig. 2B). This assay was then used in a pilot screen of a human kinase/phosphatase siRNA library (Hoffman et al., 2010).
Materials Non-targeting control, PDK1, S6K1, S6K2, mTOR, and PLK1 siRNA pools (Dharmacon) Dish containing HeLa cells DMEM (CellGro, cat. no. 10-013-CV) Fetal bovine serum (FBS; PAA Laboratories, cat. no. A15-301) Oligofectamine transfection reagent (Invitrogen, cat. no. 12252-011) Opti-MEM (Invitrogen, cat. no. 11058-021) Penicillin/streptomycin (CellGro, cat. no. 30-002-CI)
In-Cell Western Assays
Tissue culture hood 384-well siRNA storage plates (Eppendorf, cat. no. 951020745) Matrix WellMate (or other automated liquid dispenser for multiwell plates) (Rudnicki and Johnston, 2009) WellMate manifold (sterile) Black opaque 384-well tissue culture microplates (Corning, cat. no. 3712) Centrifuge capable of holding multiwall plates 20◦ C incubator Liquid handling robot (e.g., Velocity 11 Bravo) (Rudnicki and Johnston, 2009) Standard tissue culture incubator capable of maintaining an environment of 37◦ C and 5% CO2
44 Volume 3
Current Protocols in Chemical Biology
1. Before the experiment, prepare siRNA stocks in a 384-well siRNA storage plate for testing control siRNAs, determination of the assay Z value, and/or screening. siRNA stock plates can be stored long-term at −20◦ C. siRNAs are dissolved according to the manufacturer’s instructions in 1× siRNA buffer. 1× siRNA buffer was prepared by dilution of 5× siRNA buffer (Dharmacon, cat. no. B-002000-UB-100) into nuclease-free water (Ambion, cat. no. AM9932). A sample siRNA control plate format is shown in Figure 2B. siRNAs are plated at 1 μM in the stock plate (Tsui et al., 2009).
2. If necessary, thaw siRNA stock plates that will be used in the transfections. 3. Prepare the HeLa cell suspension as follows: a. Trypsinize the contents of one or more dishes of HeLa cells by first aspirating the medium from the plates and then briefly washing with sterile PBS. After the brief wash, add 2 ml 0.25% trypsin/EDTA and incubate 5 min at 37◦ C. Add 10 ml complete medium (to stop the trypsinization), centrifuge 5 min at 1000 × g, aspirate the supernatant, and resuspend the cell pellet in fresh medium. b. Resuspend the cells at a final density of 40,000 cells/ml in DMEM containing 10% FBS but no penicillin or streptomycin. Cells are grown to 70% to 80% confluency in 15-cm plates (BD Falcon, cat. no. 353025). This cell suspension will give a final plating density of 800 cells/well when 20 μl of cell suspension are delivered to the plates in step 9 (see below).
4. Prepare the Oligofectamine dilution in Opti-MEM. Use a ratio of 8.86 μl Opti-MEM + 0.14 μl Oligofectamine per well, and scale up appropriately. 5. Using a Matrix WellMate liquid dispenser, aliquot 9 μl of the Oligofectamine OptiMEM mix per well into a 384-well tissue culture plate. 6. Centrifuge the plates 1 min at 1000 × g, room temperature, in a centrifuge capable of holding multiwell plates to ensure that all of the liquid is at the bottom of the wells. 7. To prepare the siRNA/lipid complexes, add siRNA from the stock plates to the Oligofectamine-Opti-MEM mix in the 384-well tissue culture plates using the Velocity 11 Bravo robot. The robot takes 3 μl of siRNA from the stock plates, transfers 1.5 μl to each well of the transfection plate, and then mixes the siRNA and lipid.
8. Incubate the transfection plates 15 min at 20◦ C to allow the siRNA/lipid complexes to form. 9. Add 20 μl of the HeLa cell suspension per well to the transfection plates using the WellMate liquid dispenser. Leave one column empty of cells to use as a no-cell control. Transfer the plates to the tissue culture incubator. 10. After incubating the cells with the transfection mix for 20 hr, prepare medium composed of DMEM + 10% FBS + 1% penicillin/streptomycin. Using the WellMate and a sterile manifold, add 55 μl of the medium to each well of the plate and return the plate to the incubator. 11. Continue to incubate for a total of 72 hr after the transfection. 12. Proceed to Basic Protocol 3.
45 Current Protocols in Chemical Biology
Volume 3
A Control
PDK1
S6K1 & 2
S6K1
Relative phospho–S6 staining (arbitrary units)
3.0 2.5 2.0 1.5 1.0 0.5
PL K1
S6 K2 m TO R
S6 K1
2 S6 K1 /
N
PD K1
TC
0.0 S6K2
siRNA 3000 Relative Alexa–680 staining (arbitrary units)
mTOR
A N R si
O
ve r
la y
80 -6 Al ex a
ph
os
ph
o-
S6
PLK1
2500 2000 1500 1000 500
R
PL K1
TO m
K2 S6
S6
K1
/2 K1 S6
K1 PD
N TC
0
B Not used S6K1 & 2 siRNA
Control siRNA Not used
*
*
No Cells
Figure 2
No 1°
Relative phospho–S6 intensity (arbitrary units)
siRNA 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0
Z
0
50
0.67
100
150
200
Well number
(legend appears on next page)
BASIC PROTOCOL 3
IN-CELL WESTERN ASSAY FOR QUANTIFICATION OF rpS6 PHOSPHORYLATION
In-Cell Western Assays
This protocol describes the main body of the procedure for measurement of rpS6 phosphorylation. Specifically, it employs a primary antibody that recognizes rpS6 phosphorylated at serines 235 and 236 (Cell Signaling Technology). This is used in combination with a secondary antibody conjugated to IRDye 800CW, a far-red fluorophore. For cell num-
46 Volume 3
Current Protocols in Chemical Biology
ber normalization, an amine reactive near-IR dye is used for bulk protein labeling: the succinimidyl ester of Alexa-680 (Invitrogen).
Materials Phosphate-buffered saline (PBS; see recipe) EM-grade formaldehyde (Polysciences, cat. no. 04018) Plates to be assayed (see Basic Protocol 1 or Basic Protocol 2) Triton washing solution (see recipe) Alexa Fluor 680 succinimidyl ester (see recipe) Blocking solution (see recipe) Anti-rpS6 phospho-S235/S236 antibody (Cell Signaling Technologies, cat. no. 2211) IRDye 800CW-conjugated goat anti-rabbit secondary antibody (LI-COR Biosciences, cat. no. 926-32211) 24-pin aspiration wand (Drummond) modified by slipping cut pieces of laboratory tubing over the ends of the wand so that the pins remain a few millimeters above the bottom of the microplate wells when aspirating Matrix WellMate manifold (nonsterile; Thermo Scientific) Matrix WellMate microplate dispenser (or other automated liquid dispenser for multiwell plates; Thermo Scientific) (Rudnicki and Johnston, 2009) Multiwell plate washer (e.g., Bio-Tek ELx405) (Rudnicki and Johnston, 2009) Dark box or cupboard or aluminum foil Parafilm Metal foil seals for microplates (Corning, cat. no. 6570) Near-infrared-capable plate scanner (e.g., Aerius, LI-COR) Spreadsheet or graphing software Curve fitting software (optional) Fix the cells 1. Immediately before fixation of the cells, prepare a fresh solution of 4% formaldehyde in PBS by dilution of the formaldehyde stock solution. 2. Remove the plates to be assayed from the incubator, and aspirate the medium from the wells using the modified 24-pin aspiration wand. When the modified wand is used to wash fixed cells in this protocol, 5 to 10 μl of liquid should remain on top of the cells because the aspirator pins will not touch the bottom of the plate. Do not allow the cells to dry out. Figure 2 (appears on previous page) siRNA validation data. (A) As described in Basic Protocol 2, HeLa cells were transfected with siRNAs directed against the indicated genes. 72 hr post-transfection, the cells were fixed and stained for analysis on the LI-COR Aerius instrument following the procedure described in Basic Protocol 3. The raw scan data from representative wells is shown in the panel on the left. Quantitation of cell number by Alexa680 succinamidyl ester staining is shown in the graph on the lower right panel. Quantitation of the normalized phospho-S6 data is shown in the graph on the upper right panel (n = 10, error bars show one standard deviation). (B) The statistical reproducibility of the ICW assay for siRNA screening was determined by calculating a Z-factor comparing S6K1/2 siRNA to nontargeting siRNA control in a 384-well format. Quantitation of the raw data is shown and used to calculate a Z-factor (Z ) of 0.67 for this experiment. The asterisks (*) denote untreated wells that were not used in the Z-factor determination. In the assay plate used for the Z determination, the wells in rows C-H and columns 3-22 are transfected with siRNA specific for S6K isoforms 1 and 2, as a positive control for inhibition (Dharmacon, cat. nos. L-003616-00 and L-004671-00). The wells in rows I-N and columns 3-22 are transfected with a nontargeting siRNA, as a negative control for inhibition (Dharmacon, cat. no. D-001810-10-20). No cells were plated in column 1, as a negative control for staining in both the 700- and 800-nm channels. In column 24, primary antibody was omitted, as a control to determine the level of nonspecific staining in the 800-nm channel.
In-Cell Western Assays
47 Current Protocols in Chemical Biology
Volume 3
3. Using a nonsterile WellMate manifold, add 75 μl of the 4% formaldehyde solution to each well. 4. Incubate the plates for 20 min at room temperature to allow complete fixation.
Permeabilize the cells 5. Aspirate the fixative solution using the modified wand (again, leaving 5 to 10 μl of liquid on the cells to prevent drying) and add 75 μl Triton washing solution to each well using the WellMate to stop fixation. 6. Using the automated plate washer, wash the plates four times, each time with 75 μl Triton washing solution per well. No incubation period with wash buffer is necessary; wash buffer can be removed immediately after it is added to each well. However, if a stack of plates is being processed, causing a delay between the times of addition and aspiration of wash buffer, it is acceptable to leave wash buffer on the cells for up to 1 hr at room temperature. If a suitable automated plate washer is not available, the wash steps can be carried out in a semi-automated manner by adding wash solutions using the automated liquid dispenser and aspirating using the wand.
Stain the cells with succinimidyl (NHS) ester 7. While washing the plates, prepare the staining solution, composed of Alexa-680 succinimidyl ester at 1:50,000 in Triton washing solution (20 ng/ml final). This solution should always be prepared fresh immediately prior to staining of plates.
8. Using the WellMate liquid dispenser, add 75 μl staining solution to each well and incubate for 15 min at room temperature. It is important to keep this incubation as close to 15 min as possible because longer incubations will lead to high staining levels that may saturate the detector during imaging.
9. Aspirate the staining solution using the wand, leaving 5 to 10 μl of liquid on top of the cells. In order to promptly remove residual Alexa-680, immediately add 75 μl Triton washing solution per well to the plates, and aspirate using the wand. 10. Wash the plates four times, each time using the plate washer, as described in step 6.
Block the cells 11. Using the WellMate, add 75 μl of blocking solution to each plate well. 12. Block for 1 hr at room temperature, keeping the plates protected from light by placing them in a dark box or cupboard, or by wrapping them in foil. 13. Aspirate the blocking solution using the wand.
Stain the cells with the primary antibody 14. Turn the plates upside down on a sheet of bench paper and allow them to drain for 5 to 10 min. This step allows for removal of the extra 10 μl of buffer left in the well to avoid unnecessary dilution of the primary antibody solution.
15. While the plates are draining, prepare the primary antibody solution by diluting the antibody 1:1000 in blocking solution. 16. Using the Wellmate, add 20 μl of primary antibody solution to each well. For a “no-primary” control, add blocking solution instead of primary antibody solution to the last column (column 24) of each plate (see the plate layouts in Figs. 1D and 2B). In-Cell Western Assays
The inclusion of a column of “no-primary” control wells on each plate will allow quantification of the amount of nonspecific background staining by the secondary antibody.
48 Volume 3
Current Protocols in Chemical Biology
17. Seal the lids of the plates with a thin strip of Parafilm, and incubate overnight at 4◦ C in the dark.
Stain with the secondary antibody 18. The next day, wash all of the plates on the plate washer four times, each time with 75 μl Triton washing solution. 19. After the final wash, dry the plates upside down as described in step 14. 20. While the plates are drying, prepare the secondary antibody dilution of 1:1000 in blocking solution. The labeled secondary antibody should be protected from direct light exposure during experiments and in storage.
21. Using the WellMate, add 20 μl of the diluted secondary antibody to each well, and incubate the plates for 1 hr at room temperature protected from light. 22. Wash on the plate washer four times, each time with 75 μl Triton washing solution. 23. After the final wash, add 30 μl PBS per well using the WellMate. 24. Seal the plates using metal foil seals, and store at 4◦ C in the dark if they are not to be scanned immediately. Aluminum foil is not recommended for sealing the plates, as it will not prevent evaporation. Plates can be stored for at least up to a week after staining without appreciable loss of signal.
Scan the plates and analyze the data 25. Scan the plates using a fluorescence-scanning instrument capable of measuring fluorescence at 700 and 800 nm (e.g., a LI-COR Aerius). To obtain optimal signal, it is important to optimize the focus offset position of the scanner or plate reader and the intensity settings for both the 700- and 800-nm channels. The details will depend on the instrument used.
26. Using the instrument software, export the raw image data into a format usable by Microsoft Excel or other suitable spreadsheet software. 27. For each well, calculate the normalized phospho-rpS6 intensity by dividing the trimmed mean pixel intensity for the phospho-S6 staining on the 800-nm channel by the trimmed mean pixel intensity of the Alexa-680 stain measured on the 700-nm channel. The trimmed mean is determined by calculating the mean pixel intensity after discarding 20% of the pixels at the high and low ends of the distribution. Using the trimmed mean improves the data quality by reducing or eliminating the effect of artifacts, such as dust particles in the wells. Wells in which the Alexa-680 staining is very low are indicative of the small molecule or siRNA in that well having cytotoxic activity.
REAGENTS AND SOLUTIONS Unless a different solvent is specified, use Milli-Q purified water or equivalent in all recipes and protocol steps.
Alexa-680 NHS ester stock solution Prepare a 1 mg/ml stock solution of Alexa-680 NHS ester in lyophilized powder form (Invitrogen, cat. no. A20108) in dry (anhydrous grade) dimethyl sulfoxide (DMSO). Do not resuspend the dye in aqueous solution or buffer, as the succinimidyl ester reactive group will be quickly hydrolyzed and inactivated by water. Store the DMSO solution protected from light indefinitely at −20◦ C.
In-Cell Western Assays
49 Current Protocols in Chemical Biology
Volume 3
Blocking solution Combine an equal volume of phosphate-buffered saline (PBS; see recipe) with Odyssey blocking buffer (LI-COR Biosciences, cat. no. 927-40000) in a 1:1 ratio. Prepare fresh.
Phosphate-buffered saline (PBS) 37 mM NaCl 2. mM KCl 10 mM sodium phosphate dibasic 2 mM potassium phosphate monobasic Adjust to pH 7.4 using NaOH and/or HCl Store indefinitely at room temperature if sterile-filtered or autoclaved Triton washing solution To PBS (see recipe), add Triton X-100 to a final concentration of 0.2%, and sodium azide (NaN3 ) to a final concentration of 1 mM. Store up to 1 month at room temperature.
COMMENTARY Background Information
In-Cell Western Assays
The rapid growth in the number and quality of antibodies specific for protein phosphorylation sites and other post-translational modifications has significantly enhanced the capability of researchers to measure the activity of signaling pathways in mammalian cells. Traditional, low-throughput techniques utilizing these reagents include western blotting, dot blotting, conventional ELISAs, flow cytometry, and immunocytochemistry. A major area of assay development has been the conversion and adaptation of these antibody-based methods to high-throughput, microplate-based formats. Carrying out such assays in the cellular environment is advantageous for a number of reasons. One is that this allows measurement of the effects of an inhibitor on an entire pathway: while a small molecule might effectively inhibit a kinase or other component of a signaling network in a biochemical assay, in living cells this may have no effect on the pathway due to alternative signaling routes and/or compensatory mechanisms. In addition, an assay in a cellular context allows measurement of the ability of small molecules to enter cells, and determination of any cytotoxic effects that compounds or siRNAs have, in addition to inhibition or activation of the pathway of interest. One such assay method is the cytoblot (Stockwell et al., 1999), in which cells are fixed, probed with a primary antibody against the target of interest, and probed with a secondary antibody conjugated to horseradish peroxidase for chemiluminescent detection of target levels. The in-cell western assay was
developed using a similar approach, but instead of using chemiluminescence, uses fluorescence detection and exploits the favorable properties of far-red fluorophores. Recently, several variants of the in-cell western method have been developed, including the on-cell western (Daigle et al., 2008), in which a protein expressed on the surface of a cell is quantified by comparison of signals in cells that are permeabilized with those that are not. One important caveat when using in-cell westerns, cytoblots, and similar techniques is that they do not resolve the target protein based on size. Thus, when using new antibodies with these methods, it is important to confirm that the measured signal is target-specific using a conventional western blotting experiment.
Critical Parameters Antibody selection As with any immunofluorescence method, the availability and selection of high-quality antibodies against targets of interest is central to the development of robust ICW assays. Generally, this technique can be used with any antibody that is already known to be suitable for immunofluorescence-based detection of the protein of interest. If the antibody to be used has not already been validated for such applications by the commercial supplier, this can be done using standard fluorescence microscopy techniques. If multiple suitable primary antibodies are available, it may be worth testing them against each other to determine which gives the best assay performance.
50 Volume 3
Current Protocols in Chemical Biology
Choice of normalization method Another key parameter in this type of assay is the cell number normalization method used. Fluorescent reagents that covalently label proteins, such as the Alexa Fluor 680 NHS ester used in this protocol, are highly cost effective and allow cell number determination over a wide linear range. Since these dyes label cellular proteins and not nuclear DNA, they are suitable for assays in which the cells in some wells may undergo apoptosis. Intercalating dyes, such as TO-PRO3, can cause aberrant staining of nuclei in which the DNA is fragmented. During the early stages of development of the assay described in this unit, it was found that the intensity of many DNA-binding dyes did not correlate well with cell number, particularly in cells transfected with siRNAs inducing apoptosis. For phospho-proteins, an alternative means of normalization is use of an additional primary antibody that detects total cellular levels of that protein. Although use of a total rpS6 antibody was found to give slightly more statistically robust normalization than the Alexa-680 succinimidyl staining for the assay described here, this method also significantly increased the cost of the assay, which may make it impractical for largescale screening efforts. We have found that direct labeling of cells with amine-reactive dyes, such as the Alexa-680 NHS-ester described in this protocol, provides an accurate and economical method for normalization in the ICW method and we have adopted this approach in our screening work. Microplate selection Proper selection of microplates is necessary to achieving good results with the ICW method. Plates used must have clear bottoms to permit scanning and should have black walls to minimize autofluorescence and well-to-well cross-talk. In addition, for the most commonly used scanners (Aerius and Odyssey) it is essential that there is no more than 4-mm distance between the plate bottom and the well bottom. For cell lines that attach poorly and are easily lost during washing steps, it can be useful to use plates with special coatings [e.g., PureCoat microplates (BD) and PDL-coated plates (Perkin Elmer)] designed to promote stronger attachment. For each type of plate used, the focus offset should be optimized on the instrument of choice to maximize the fluorescent signal-tonoise ratio. The focus offset is an instrument parameter that adjusts for the distance between
the scanner bed and the bottom of the plate, and thus needs be optimized for each plate type.
Troubleshooting Weak antibody staining If the primary antibody does not produce an adequate signal, it may be necessary to increase the intensity settings on the plate reader. Titration of the primary and secondary antibody concentrations used in staining may also be necessary to determine the dilution factors that give optimal signal-to-noise while maintaining a reasonable cost of antibody consumption. Primary antibodies can exhibit significantly reduced signal in different blocking solutions, so changing the solution conditions can dramatically improve performance. If the primary antibody has been successfully used for immunofluorescent staining in other experiments, the blocking buffer used in these should be tried. Finally, fixation and permeabilization conditions will also affect antibody staining, and if conditions for the target cell line and/or protein have already been established, these may be more appropriate than the conditions described in these protocols. High background staining If control siRNAs and/or compounds do not produce the expected loss of fluorescent signal in control wells, there may be a problem with high background staining. The level of background staining can be evaluated by inclusion of a column on each plate in which the primary antibody has been omitted, as described in Basic Protocol 3. High background can be addressed by titration of antibody dilutions and adjustment of blocking, fixation, and permeabilization conditions, similar to that for weak antibody staining. Saturated fluorescent signal If the fluorescent signal is saturated in one or both of the channels, it will be necessary to re-image plates at a lower intensity setting. In this protocol, saturation is typically seen if plates are stained for too long with the Alexa-680 dye, and/or at too high a dye concentration. Loss of cells during plate washing In some cases, there may be an unexpected loss of cells in assay wells that are not expected to show cytotoxicity. This may be caused by disruption of the cell layer by the wand used for aspiration of solutions, particularly if the loss of cells is concentrated in the center of the
In-Cell Western Assays
51 Current Protocols in Chemical Biology
Volume 3
wells. As described in Basic Protocol 3, the aspiration wand should be modified by placing small sections of laboratory tubing over each end to keep the pins from touching the bottom of the wells. This minimizes cell loss during aspiration. Cell loss might also be caused by overly harsh washing conditions. If an automated plate washer is being used, parameters such as the speed of liquid dispensing need to carefully optimized to minimize disruption of cells. In our experience, the design of an automated plate washer can have a significant effect on cell retention in an ICW assay. The washer used in this protocol has liquid dispensing pins that are angled so that dispensed fluids hit the sides rather than the bottom of wells. This resulted in better assay performance than an earlier model of the plate washer in which the pins dispensed fluid directly to the bottom. Finally, mammalian cells can vary significantly in strength of attachment to tissue culture plates, and if cell loss is a problem, it is worth testing 384-well plates with coatings specifically designed for poorly attaching cells such as PureCoat plates (BD Falcon).
Anticipated Results The procedures described above can be used to generate initial validation data for siRNA and/or small molecule controls in the ICW assay, and Z factors for the assay (as shown in Figs. 1 and 2). It can then be used for high-throughput screening of small molecule and/or siRNA libraries.
Time Considerations Once all of the necessary reagents are in hand, the procedures described above can be completed in a standard work week for an assay monitoring effects of siRNA transfection, and in two full days for the small molecule protocol. If the incubation with compounds were extended to overnight or longer, then this would take additional days. In practice, development and validation of a new assay can take from several weeks to several months depending on how much optimization and testing of different reagents is required.
Acknowledgements The authors would like to acknowledge funding support from a User Innovation Grant
from LI-COR Biosciences and NIH Grants CA46595, GM51405, and CA139980.
Literature Cited Chen, H., Kovar, J., Sissons, S., Cox, K., Matter, W., Chadwell, F., Luan, P., Vlahos, C.J., Schutz-Geschwender, A., and Olive, D.M. 2005. A cell-based immunocytochemical assay for monitoring kinase signaling pathways and drug efficacy. Anal. Biochem. 338:136142. Choo, A.Y. and Blenis, J. 2006. TORgeting oncogene addiction for cancer therapy. Cancer Cell 9:77-79. Daigle, T.L., Kearn, C.S., and Mackie, K. 2008. Rapid CB1 cannabinoid receptor desensitization defines the time course of ERK1/2 MAP kinase signaling. Neuropharmacology 54:36-44. Hoffman, G.R., Moerke, N.J., Hsia, M., Shamu, C.E., and Blenis, J. 2010. A high-throughput, cell-based screening method for siRNA and small molecule inhibitors of mTORC1 signaling using the In Cell Western technique. Assay Drug Dev. Technol. 8:186-199. Inoki, K., Corradetti, M.N., and Guan, K.L. 2005. Dysregulation of the TSC-mTOR pathway in human disease. Nat. Genet. 37:19-24. Olive, D.M. 2004. Quantitative methods for the analysis of protein phosphorylation in drug development. Exp. Rev. Proteomics 1:327341. Rudnicki, S. and Johnston, S. 2009. Overview of liquid handling instrumentation for highthroughput screening applications. Curr. Protoc. Chem. Biol. 1:43-54. Stockwell, B.R., Haggarty, S.J., and Schreiber, S.L. 1999. High-throughput screening of small molecules in miniaturized mammalian cellbased assays involving post-translational modifications. Chem. Biol. 6:71-83. Tsui, M., Xie, T., Orth, J.D., Carpenter, A.E., Rudnicki, S., Kim, S., Shamu, C.E., and Mitchison, T.J. 2009. An intermittent live cell imaging screen for siRNA enhancers and suppressors of a kinesin-5 inhibitor. PLoSOne 4:e7339. Wong, S.K. 2004. A 384-well cell-based phosphoERK assay for dopamine D2 and D3 receptors. Anal. Biochem. 333:265-272. Wullschleger, S., Loewith, R., and Hall, M.N. 2006. TOR signaling in growth and metabolism. Cell 124:471-484. 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. 4:67-73.
In-Cell Western Assays
52 Volume 3
Current Protocols in Chemical Biology
Derivatization of Free Natural Glycans for Incorporation onto Glycan Arrays: Derivatizing Glycans on the Microscale for Microarray and Other Applications Xuezheng Song,1 Jamie Heimburg-Molinaro,1 David F. Smith,1 and Richard D. Cummings1 1
Emory University, Atlanta, Georgia
ABSTRACT Nature possesses an unlimited number and source of biologically relevant natural glycans, many of which are too complicated to synthesize in the laboratory. To capitalize on the naturally occurring plethora of glycans, a method is presented here to fluorescently tag isolated free glycans while maintaining the closed-ring structure. After purification of the labeled glycans, they can be printed on a glass surface to create a natural glycan microarray, suitable for interrogation with potential glycan-binding proteins. The derivatization of these natural glycans has vastly expanded the number of glycans available for funcC 2011 by John Wiley & Sons, Inc. tional studies. Curr. Protoc. Chem. Biol. 3:53-63 Keywords: fluorescence r reductive amination r glycan microarray r conjugation
INTRODUCTION Glycan microarrays have become an important hypothesis-generating platform in studies of glycan-protein interactions in signal transduction, cellular adhesion, and host-pathogen interaction (Feizi and Chai, 2004; Alvarez and Blixt, 2006; de Paz and Seeberger, 2006; Paulson et al., 2006; Stevens et al., 2006; Horlacher and Seeberger, 2008). With hundreds of defined glycans presented and available simultaneously on the same microarray for interrogation, the specific structures or structural motifs recognized by glycan-binding proteins (GBPs) or microorganisms can be rapidly discovered. Understanding these specificities and motifs provides important clues to glycan function. For example, the realization that certain galectins had high affinity for specific human blood-group glycan epitopes led to the discovery that galectins function as innate immune proteins by binding to blood group–like glycans on microorganisms (Stowell et al., 2010). The power of glycan microarrays to detect specificities and motifs of GBPs is directly related to the availability of diverse structures, which have been generated largely by chemical and chemo/enzymatic synthesis (Blixt et al., 2004). The current version of the Glycan Microarray that is publicly available through the NIGMS-funded Consortium for Functional Glycomics (CFG; http://www.functionalglycomics.org) comprises over 500 defined glycans; however, the human glycome is estimated to possess in excess of 7000 glycan determinants (Cummings, 2009). Expansion of the defined array is fraught with difficulties associated with the chemical synthesis of large and complex oligosaccharides and, although it is an approachable task, other methods to enhance the diversity of these platforms are being explored. An obvious source of glycans for this expansion is nature itself, which offers an inexhaustible supply of complex glycan structures and, since they are natural products made by complex genetic/biochemical pathways, they may be presumed to be biologically relevant. Mining these structures from nature, however, presents significant challenges, and Current Protocols in Chemical Biology 3: 53-63, April 2011 Published online April 2011 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/9780470559277.ch100194 C 2011 John Wiley & Sons, Inc. Copyright
Derivatizing Glycans on the Microscale
53 Volume 3
to this end, chemical methods have been developed to acquire naturally occurring glycans on a microscale. Since these glycans are generally invisible to nondestructive detection methods and often available in only nanomole quantities, it is necessary to introduce fluorescent linkers at their reducing ends to provide sensitive detection. In addition, they must be chemically functionalized to permit immobilization on array surfaces (Xia et al., 2005; Song et al., 2009a,b). The resulting derivatives can then be quantified and printed in the same way as synthetic glycans with a primary amino group. The bifunctional fluorescent labeling of free glycans, as a complementary method to synthetic approaches, has shown great potential in increasing the size of the glycan library and facilitating the study of GBPs. In addition to their use for expanding the population of defined glycans on glycan microarrays, the fluorescent glycan derivatives that also possess an alkyl primary amine function can be used for linking defined glycans to any type of derivatized surface for use in exploring glycan recognition. For example, fluorescent glycan derivatives can be coupled to biotin using commercially available NHS–biotin, and the resulting reagent can be used for coupling glycan to any surface coated with streptavidin. Alternatively, the glycans can be covalently coupled to epoxide or NHS-derivatized surfaces or to carboxylated microspheres. In this article, we discuss the preparation of closed-ring glycan conjugates from free reducing glycans using a derivative developed in our laboratory (Basic Protocol). Alternatively, we provide a method for preparing glycan conjugates, which can include fluorescently labeled neoglycoproteins, using commercially available chemicals (Alternate Protocol). These conjugates are useful in microarray technologies and other applications. BASIC PROTOCOL
PREPARATION OF CLOSED-RING GLYCAN-AEAB FROM FREE REDUCING GLYCANS Even with efficient extraction and isolation methods, naturally occurring glycans are often only available in submilligram quantities. To utilize such quantities of glycans in studies of their interactions with GBPs, the glycan amounts need to be accurately quantified and immobilized to solid phases such as microarray chips. Bifunctional fluorescent tags such as 2,6-diaminopyridine (DAP) and 2-amino(N-aminoethyl) benzamide (AEAB) can be coupled to reducing glycans through reductive amination, and the resulting glycan conjugates are fluorescent and possess a primary amino group (Xia et al., 2005; Song et al., 2009a,b,c). While the fluorescence greatly facilitates its separation and quantification, the
R4
R4 O
R3
OH R2
NH4HCO3
R3
NH2 R2
R1
free reducing glycan
R4
acryloyl chloride
O
O R3 R2
R1
glycosylamine intermediate
O NH O R2
Derivatizing Glycans on the Microscale
Figure 1
closed-ring O
R4 R3
R1 O
glycosylamide R4
ozone
NH
R1 O
AEAB NaCNBH3
R3
NH
fluorescence tag functional O group HN NH2 fluorescently labeled, reactive glycan R2
R1 O
NH
The closed-ring fluorescent conjugation of reducing glycans.
54 Volume 3
Current Protocols in Chemical Biology
amino group is used to immobilize the conjugates to solid surfaces for further functional study with GBPs. However, direct reductive amination of free reducing glycans destroys the ring structure at the reducing end, which is desired in many situations. Therefore, we have developed a microscale process combining glycosylamide formation, ozonolysis, and AEAB reductive amination (summarized in Fig. 1) to prepare a glycan derivative retaining its ring structure (Song et al., 2009a,b,c). NOTE: All chemicals used in this protocol are analytical grade and are used without further purification.
Materials 0.05 to 1 mg free reducing glycan, lyophilized (e.g., Sigma-Aldrich, V-labs, Carbosynth) Milli-Q purified water (Millipore), or equivalent Ammonium bicarbonate 50% (v/v) acetonitrile, HPLC grade (Fisher Scientific)/10 mM ammonium bicarbonate 10 mM ammonium bicarbonate Sodium bicarbonate Saturated sodium bicarbonate solution, ice-cold Acryloyl chloride (e.g., Sigma-Aldrich) Sodium borohydride (e.g., Sigma-Aldrich), optional Acetic acid, ACS grade (Fisher Scientific), optional 50% (v/v) acetonitrile/0.1% (v/v) trifluoroacetic acid (TFA), HPLC grade (Fisher Scientific) Methanol Ethanol Ozone Nitrogen gas Methyl sulfide 7:3 (v/v) dimethyl sulfoxide (DMSO), ACS grade (Fisher Scientific)/acetic acid Sodium cyanoborohydride 2-(N-aminoethyl)-amino benzamide (AEAB) hydrochloride (Song et al., 2009c) Acetonitrile 1% (v/v) trifluoroacetic acid (TFA), HPLC grade (Fisher Scientific) 1.5-ml screw-cap polypropylene centrifuge tubes 55◦ C and 65◦ C heating block or water bath 150 mg, 300 mg, and 1 g carbograph solid phase extraction (SPE) columns (Alltech) Rotary evaporator (e.g., SpeedVac, Thermo Scientific) Lyophilizer 15-ml conical polypropylene centrifuge tubes Porous graphitized carbon (PGC) analytical HPLC column (Thermo Scientific) High-performance liquid chromatography (HPLC) system Prepare glycosylamine 1. Dissolve 0.05 mg to 1 mg free reducing glycan in 50 μl Milli-Q water in a 1.5-ml screw-cap tube. Add 100 mg ammonium bicarbonate. Incubate the mixture 1.5 hr at 55◦ C, and then cool to room temperature. Normal polypropylene centrifuge tubes can be used as the reaction vessel throughout the protocol. Glassware is not necessary. For the carbograph desalting, the glycan size needs to be bigger than trisaccharide to ensure adsorption on carbograph.
Derivatizing Glycans on the Microscale
55 Current Protocols in Chemical Biology
Volume 3
2. Condition a small (150-mg) carbograph solid phase extraction (SPE) column using 1 column volume (∼3 ml) of 50% acetonitrile containing 10 mM ammonium bicarbonate, followed by 3 column volumes of 10 mM aqueous ammonium bicarbonate. 3. Use 1 ml of 10 mM ammonium bicarbonate to transfer the glycan sample to the preconditioned carbograph SPE column. 4. Wash the column with 2 column volumes of 10 mM ammonium bicarbonate to remove salts. Elute the carbograph with 2 column volumes of 50% acetonitrile/ 10 mM ammonium bicarbonate. For larger amounts of glycans, the reaction can easily be scaled up. For example, for 10 mg free reducing glycan, 1 ml water, and 2 g ammonium bicarbonate are used. The reaction should be incubated in a 15-ml conical tube, and a 1-g carbograph column should be used in the desalting procedures.
5. Evaporate the glycosylamine eluate in 50% acetonitrile/10 mM ammonium bicarbonate in a rotary evaporator for 2 hr to remove acetonitrile, and remove residual water and ammonium bicarbonate by lyophilizing in a 15-ml conical tube.
Prepare and purify acryloylated glycan 6. To the lyophilized powder in a 15-ml conical tube (sitting on ice), add 50 mg sodium bicarbonate and 0.5 ml ice-cold saturated sodium bicarbonate solution. Add 20 μl acryloyl chloride, immediately cap the mixture, and agitate by vortexing for 5 min. 7. Slightly open the cap to release the pressure, and close it again. Then shake the mixture at room temperature for another 60 min, by vortexing on the lowest setting. 8. (Optional) When starting with a mixture of glycans in step 1 (e.g., N-glycans released from a glycoprotein by PNGase F), carry out a sodium borohydride reduction after the acryloylation reaction as follows: a. For x mg glycan starting material, directly add x mg sodium borohydride (dissolved in water) while cooling on ice. The concentration of glycan solutions should be kept to 1 to10 mg/ml in this step.
b. Shake the reaction with frequent cooling for 10 min. c. Add, while cooling, a volume of acetic acid equal in μl to two times the number of mg of glycan used in step 8a, and incubate for 10 min. Due to the instability of the glycosylamine, a percentage of reducing glycan often occurs in the mixture after acryloylation. This portion of reducing glycan is also conjugated with AEAB in subsequent steps to form the corresponding open-ring conjugate. While this is not a problem for a conjugation started from a single glycan, the purification by HPLC of a glycan mixture is somewhat more complicated. Thus, it is important to introduce the step of sodium borohydride reduction before the AEAB conjugation to solve this problem. This minor fraction of reduced glycans is not reactive with AEAB or DAP.
9. Condition a small (150 mg) carbograph SPE column using 1 column volume (∼3 ml) of 50% acetonitrile/0.1% trifluoroacetic acid (TFA), followed by 3 column volumes of water. 10. Dissolve the acryloylated glycan from step 7 (or optional step 8) in 3 ml water and apply to the carbograph column. 11. Wash the column with 6 column volumes of water, and elute the glycosylamide with 2 column volumes of 50% acetonitrile/0.1% TFA. Derivatizing Glycans on the Microscale
12. Evaporate the eluted solution using a rotary concentrator for 2 hr to remove acetonitrile, and remove water by lyophilizing.
56 Volume 3
Current Protocols in Chemical Biology
13. Dissolve the lyophilized material in 1 ml methanol, and chill to −78◦ C using dry ice/ethanol. 14. Bubble ozone through the solution for ∼1 min, until it turns blue and the blue color remains. Then turn off the ozone generator, and bubble nitrogen through the solution for ∼1 min until the blue color disappears. 15. Add 50 μl methyl sulfide, and let the mixture warm to room temperature and stand for 1 hr. 16. Dry the solution under a stream of nitrogen.
Prepare AEAB conjugates 17. While the solution is drying under nitrogen, prepare an AEAB solution by adding 1 ml of 7:3 (v/v) DMSO/acetic acid to 88 mg AEAB hydrochloride and shake at room temperature for 10 min. 18. Centrifuge 1 min at 1000 × g, room temperature, and collect the supernatant (for use in step 20). 19. Prepare a solution of sodium cyanoborohydride by adding 1 ml of 7:3 DMSO/acetic acid to 64 mg sodium cyanoborohydride. 20. To the dried residue from step 16, add 10 to 50 μl of freshly prepared 0.35 M AEAB solution (from step 18) and an equal volume of 1 M sodium cyanoborohydride solution (from step 19). A precipitate forms briefly after mixing.
21. Heat the suspension 2 hr at 65◦ C. 22. Purify the glycan-AEAB conjugates or the closed-ring conjugates from the reaction mixture by adding 10 volumes acetonitrile, which causes the conjugates to precipitate. Transfer the mixture to a 1.5-ml microcentrifuge tube and chill to −20◦ C. 23. Centrifuge the precipitated glycan conjugates 3 min at 10,000 × g, room temperature. 24. Remove and discard the supernatant, and dissolve the pellet in 100 μl water. The sample is now ready for MALDI-TOF analysis and HPLC purification (step 25). If closed-ring conjugates are not required, direct conjugation of AEAB starting with lyophilized or dry, free reducing glycans can be easily conducted by simply following steps 17 to 24, starting from the preparation of AEAB solution.
Purify AEAB derivatives 25. Purify both closed-ring and open-ring glycan AEAB derivatives, using a Hypercarb (PGC) analytical HPLC column coupled with a Javelin guard cartridge, and the following HPLC program: Solvents Acetonitrile (A) 1% TFA (B) Milli-Q purified water (C) Flow rate: 1 ml/min Linear gradient 0 min: 15% A, 10% B, and 75% C 30 min, 45% A, 10% B, and 45% C 30.1 min: 15% A, 10% B, and 75% C 40 min: stop Fluorescence detection: 330 nm (exitation)/420 nm (emission) UV detection: 330 nm.
Derivatizing Glycans on the Microscale
57 Current Protocols in Chemical Biology
Volume 3
The column has sufficient capacity and resolving power to purify conjugates at a scale of 1 μg to 1 mg. The HPLC purification procedure for open-ring and closed-ring AEAB conjugates is essentially identical, with the latter showing relatively longer retention times.
26. Collect peaks containing the AEAB derivatives (retention time typically 10 to 30 min), evaporate in a rotary evaporator for 2 hr, and lyophilize to remove water. 27. Based on its fluorescence and UV absorption in the HPLC procedure, reconstitute each glycan or glycan fraction to 200 μM in water. Both the closed-ring and open-ring AEAB derivatives can be stored up to 6 months at −20◦ C without noticeable changes. This solution of closed-ring AEAB derivative is suitable for printing in the appropriate buffer; however, the printing of microarray is beyond the scope of this article and can be found elsewhere (Smith et al., 2010). The quantification and reconstitution of the glycan-AEAB conjugates are based on standard curves generated from accurately weighed standard lactose-AEAB conjugate, prepared in the lab using steps 17 to 24. Lactose-AEAB from 10 pmol to 1 nmol can be detected by fluorescence linearly, and lactose-AEAB from 500 nmol can be detected by UV absorbance at 330 nm linearly. ALTERNATE PROTOCOL
PREPARATION OF AEAB CONJUGATES USING COMMERCIAL CHEMICALS The microscale preparation of open-ring and closed-ring AEAB conjugates from free reducing glycans provides a quick and potentially high-throughput way to obtain natural glycan derivatives that can be immobilized on microarrays for study of interactions with GBPs. The AEAB conjugates have a primary amino group as an active nucleophile. To expand the scope and application of glycan derivatives, we have developed another bifunctional tag, p-nitrophenyl anthranilate (PNPA; Luyai et al., 2009). The preparation of glycan conjugates with reductive amination, as described above in the Basic Protocol, can be readily applied to the conjugation of PNPA. However, the PNPA derivatives have a p-nitrophenyl ester, which is an active ester and can react easily with diamines to generate a free amino group. Because PNPA is a commercially available chemical (AEAB is not), we have developed methods for its use in conjugation with glycans. Here we describe a protocol for PNPA conjugation and the conversion of PNPA derivatives to AEAB conjugates (Fig. 2).
Materials 7:3 (v/v) dimethyl sulfoxide (DMSO)/acetic acid solution p-nitrophenyl anthranilate, 98% (PNPA, Fisher Scientific) Sodium cyanoborohydride, 95% (Sigma-Aldrich) 0.05 to 1 mg free reducing glycan, lyophilized (e.g., Sigma-Aldrich, V-labs, Carbosynth) Acetonitrile 1% (v/v) trifluoroacetic acid (TFA) Milli-Q purified water, or equivalent Ethylenediamine solution: dissolve 100 μl ethylenediamine in 1 ml DMSO 10% (v/v) acetic acid
Derivatizing Glycans on the Microscale
58 Volume 3
65◦ C heating block or water bath C18 analytical column High-performance liquid chromatography (HPLC) system Rotary evaporator (e.g., SpeedVac, Thermo Scientific) Lyophilizer Hypercarb (PGC) HPLC column Current Protocols in Chemical Biology
R4 O R3
OH R2
free reducing glycans or R4 O R3 NH O R2
NO2
NH2 O
R1
reductive amination
O
+
p-nitrophenyl anthranilate (PNPA) R4
R1 O
OH R3
closed-ring aldehyde
OH R3
n yle eth
O NH
R2
am
edi
R4 O
UV active, separable reactive ester
pro
R2
R1
NH NH2
glycan-AEAB
NO2
R1
ine
O NH
tein
s
fluorescent neoglycoproteins
Figure 2 The preparation of AEAB conjugates or fluorescent neoglycoproteins by the reaction of PNPA derivatives with ethylenediamine or proteins, respectively.
Prepare PNPA conjugates 1. Prepare a solution of PNPA by adding 1 ml of 7:3 (v/v) DMSO/acetic solution to 90 mg PNPA. 2. Prepare a solution of sodium cyanoborohydride by adding 1 ml of 7:3 (v/v) DMSO/acetic acid solution to 64 mg sodium cyanoborohydride. 3. To 0.05 to 1 mg free reducing glycan, add 10 to 50 μl of freshly prepared 0.35 M PNPA solution (from step 1) and an equal volume of 1 M sodium cyanoborohydride solution (from step 2). Heat the mixture 2 hr at 65◦ C. While the protocol uses free reducing glycans, it can also be used for the ozone-treated, closed-ring glycan derivatives. This would result in closed-ring PNPA conjugates, which would be converted to closed-ring AEAB conjugates in steps 4 to 10.
4. Precipitate the conjugates by addition of 10 volumes of acetonitrile. Transfer the mixture to a 1.5-ml microcentrifuge tube, and cool to −20◦ C. 5. Centrifuge 3 min at 10,000 × g, room temperature. Remove and discard the supernatant. Dissolve the pellet in 100 μl water. The sample is now ready for MALDI-TOF analysis, HPLC purification (step 6), or direct conjugation to proteins.
Purify PNPA conjugates 6. Use a C18 HPLC column and the following HPLC program: The HPLC program for the purification of the PNPA conjugates: Solvents Acetonitrile (A) 1% TFA (B) Water (C) Flow rate: 1 ml/min
Derivatizing Glycans on the Microscale
59 Current Protocols in Chemical Biology
Volume 3
Linear gradient 0 min: 1% A, 10% B, and 89% C 30 min, 90% A, 10% B, and 0% C 30.1 min: 1% A, 10% B, and 89% C 40 min: stop UV detection: 330 nm. The PNPA conjugates can be purified on a C18-HPLC column prior to the reaction with ethylenediamine, due to the greater hydrophobicity of the PNPA tag. A normal C18 analytical column (4.6 mm × 250 mm) has sufficient capacity and resolving power to purify conjugates at a scale of 1 μg to 1 mg.
7. Collect peaks (the PNPA derivative is usually found at retention time 10 to 20 min), evaporate in a rotary evaporator for 2 hr, and remove the residual water by lyophilizing. Quantify the conjugates based on UV absorption compared to a standard glycan-PNPA derivative such as lactose-PNPA of known concentration. A lactose-PNPA standard can be prepared in the lab using steps 17 to 24 in the Basic Protocol. PNPA conjugates can be used directly in certain applications, e.g., in the preparation of neoglycoproteins, or converted to AEAB conjugates in the following steps.
Convert PNPA conjugates to AEAB conjugates and purify 8. To convert PNPA conjugates to AEAB conjugates, add 25 μl ethylenediamine solution to the pellet until the pellet is dissolved. The reaction is completed immediately during this process.
9. Add 75 μl 10% acetic acid and proceed to HPLC purification. 10. Purify the AEAB conjugates, using a Hypercarb (PGC) HPLC column as described in the Basic Protocol, steps 25 to 27.
COMMENTARY Background Information
Derivatizing Glycans on the Microscale
Due to the lack of suitable chromophores for spectrometric detection of carbohydrates, their analysis had historically relied on chemical colorimetric methods (e.g., the phenolsulfuric acid assay; Dubois et al., 1951), which are destructive and relatively insensitive. The derivatization of free reducing glycans with fluorescent small molecules by reductive amination has greatly facilitated the HPLC profiling of glycans (Bigge et al., 1995). Other than the alteration at the reducing end, these reaction products retain the structures of the derivatized glycans. However, without a functional group, these conjugates are of limited practical value on a small scale, other than their use in structural analyses by mass spectrometry. On the other hand, generating conjugates having, in addition to fluorescent properties, a functional group that does not severely alter chromatographic properties, provides a derivative that can be quantified with high sensitivity and coupled to derivatized surfaces. Such an approach has an obvious applica-
tion for derivatizing larger, biologically relevant glycans from natural glycans for constructing libraries to extend the diversity of defined, synthetic glycan microarrays. For this reason, we have investigated a variety of methods for preparing bifunctional fluorescent labeling reagents (Xia et al., 2005; Song et al., 2009a,b,c). This article describes the utility of AEAB, which possesses fluorescence properties and has two functional amino groups. Due to the different reactivities of the alkyl amine and the aryl amine of AEAB, the resulting glycanAEAB conjugate can be prepared as a pure primary alkyl amine, suitable for immobilization on both epoxy- and NHS-coated glass slides. Since some protein-carbohydrate interactions may require a natural acetyl or ring structure, we have developed a more sophisticated approach by introducing glycosylamine formation to generate the natural ring structure at the reducing end. Several high-yielding chemical reactions, including glycosylamine formation, amidation, ozonolysis, and reductive
60 Volume 3
Current Protocols in Chemical Biology
amination have been elegantly combined to provide a microscale preparation of fluorescent and immobilizable glycan conjugates (Manger et al., 1992; Song et al., 2009c, 2011).
by derivatizing a mixture of glycans and chromatographically separating the glycan-AEAB to derive the principal components. Purification of either of the AEAB derivatives is accomplished by two-dimensional HPLC separations, and this is greatly facilitated by their fluorescent properties. It is recommended that a normal phase separation be used for the first dimension and the PGC be used as the second dimension. PGC is a convenient step in the second dimension because elution is with acetonitrile/water, but without a nonvolatile counter ion, so that the resulting peaks may be collected and dried to obtain salt-free fractions for MALDI analysis or for direct coupling or other reactions. The retention of glycan structures on the carbograph SPE column used for desalting varies greatly among different glycans, with larger and charged glycans generally being adsorbed much more strongly. Therefore, this procedure is not suitable for preparation of mono- and disaccharide conjugates. Furthermore, highly charged, especially highly sulfated, oligosaccharides such as heparin
Critical Parameters and Troubleshooting Commonly used polypropylene tubes, including microcentrifuge tubes, can be used throughout the protocol, with no need for glassware. AEAB is prepared as HCl salts; it reacts with free aldehydes through the aryl amine function, making the more reactive alkyl amine available for subsequent reactions or covalent coupling to activated surfaces. It is important to remove all free AEAB from the glycan-AEAB derivative to ensure its accurate quantification and efficient coupling of the glycan derivative to an activated surface. For defined glycan arrays, it is imperative that the glycan-AEAB conjugate be composed of a single, defined glycan, or at least be >95% free of contaminating structures. This can be accomplished by starting the process with a purified, structurally defined glycan or
40,000 20,000 0
blood group H antibody (10 g/ml) a2
a2
b4 a3 b3
b4 a3
20,000 blood group B antibody (10 g/ml) 15,000 RFU
RFU
60,000
b4
5,000 0
1 4 7 10 13 16 19 22 25 Glycan number
30,000 25,000
a3 b4 a3
SLex antibody 2 (50 g/ml)
15,000 10,000
4,000 0
1 4 7 10 13 16 19 22 25 Glycan number
20,000
12,000
RFU
RFU
20,000 SLex antibody 1 (50 g/ml) 16,000
8,000
a3 a2 b4 a3
10,000
a3 b4 a3
5,000 1 4 7 10 13 16 19 22 25 Glycan number open-ring Glc
Gal
GlcNAc
0
1 4 7 10 13 16 19 22 25 Glycan number closed-ring Fuc
Neu5Ac
Figure 3 Comparison of the binding of four different antibodies to open-ring and closed-ring AEAB conjugates on a glycan array. The arrays were printed using a piezo printer (Perkin-Elmer) with open- and closed-ring AEAB derivatives at 300 μM on NHS-derivatized slides. Antibodies were applied to the glycan array at the concentrations indicated in the figure, and detected with appropriate fluorescently labeled secondary antibodies (Song et al., 2009b). The x axis represents different glycans on the array by number, and the y axis represents the relative fluorescence units (RFU) detected on the microarray.
Derivatizing Glycans on the Microscale
61 Current Protocols in Chemical Biology
Volume 3
oligosaccharides are not compatible with either carbograph SPE or PGC HPLC, presumably due to extremely strong retention. The capacity of carbograph SPE columns is ∼100 mg carbon/1 mg glycan.
Anticipated Results The purified closed-ring and open-ring conjugates can be treated as any other synthetic material with a primary alkyl amine group. Their reactivity toward both epoxy- and NHSderivatized surfaces is high. Although the printing, assay, and data analysis of glycan microarray is not within the scope of this article, Figure 3 shows several histograms that demonstrate the comparison of four antibodies binding to a glycan microarray printed with 26 different open-ring glycan-AEAB derivatives and 26 corresponding closed-ring glycanAEAB derivatives (Song et al., 2009b). The structures of the bound glycans are indicated using symbols in each panel. In the case of the human anti-H and anti-B blood group antibodies, the binding is apparently directed more toward the nonreducing end of the oligosaccharide, as binding is observed with both the natural closed-ring and open-ring structure. However, in the case of the two anti-sialylLewis x (SLex) antibodies, the open-ring GlcNAc of the AEAB derivative destroys the glycan epitope. Thus, if the binding motif of a GBP is at the nonreducing end of the glycan, which is generally the case, both types of conjugates show similar binding. However, if the binding motif is close to or includes the reducing end, the open-ring AEAB conjugates may not be appropriate for binding assays.
Time Considerations With the available reagents, the procedure for preparing 1 to 20 glycans can be completed in 3 to 4 days. This includes scheduling the freeze-drying overnight between steps. The procedure can be shortened significantly if the drying steps are concluded more quickly. In addition, multiple parallel syntheses can be carried out at the same time, which will greatly increase the efficiency. Testing the glycan array with GBPs, although not covered in this article, takes 3 to 4 hr, depending on the number of samples and number of steps for detecting the GBPs (Smith et al., 2010). Derivatizing Glycans on the Microscale
Literature Cited Alvarez, R.A. and Blixt, O. 2006. Identification of ligand specificities for glycan-binding pro-
teins using glycan arrays. Methods Enzymol. 415:292-310. Bigge, J.C., Patel, T.P., Bruce, J.A., Goulding, P.N., Charles, S.M., and Parekh, R.B. 1995. Nonselective and efficient fluorescent labeling of glycans using 2-amino benzamide and anthranilic acid. Anal. Biochem. 230:229-238. 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. Cummings, R.D. 2009. The repertoire of glycan determinants in the human glycome. Mol. Biosyst. 5:1087-1104. de Paz, J.L. and Seeberger, P.H. 2006. Recent advances in carbohydrate microarrays. QSAR Combin. Sci. 25:1027-1032. Dubois, M., Gilles, K., Hamilton, J.K., Rebers, P.A., and Smith, F. 1951. A colorimetric method for the determination of sugars. Nature 168:167. Feizi, T. and Chai, W. 2004. Oligosaccharide microarrays to decipher the glyco code. Nat. Rev. Mol. Cell Biol. 5:582-588. Horlacher, T. and Seeberger, P.H. 2008. Carbohydrate arrays as tools for research and diagnostics. Chem. Soc. Rev. 37:1414-1422. Luyai, A., Lasanajak, Y., Smith, D.F., Cummings, R.D., and Song, X. 2009. Facile preparation of fluorescent neoglycoproteins using pnitrophenyl anthranilate as a heterobifunctional linker. Bioconjug. Chem. 20:1618-1624. Manger, I.D., Rademacher, T.W., and Dwek, R.A. 1992. 1-N-glycyl beta-oligosaccharide derivatives as stable intermediates for the formation of glycoconjugate probes. Biochemistry 31:1072410732. Paulson, J.C., Blixt, O., and Collins, B.E. 2006. Sweet spots in functional glycomics. Nat. Chem. Biol. 2:238-248. Smith, D.F., Song, X., and Cummings, R.D. 2010. Use of glycan microarrays to explore specificity of glycan-binding proteins. Methods Enzymol. 480:417-444. Song, X., Lasanajak, Y., Rivera-Marrero, C., Luyai, A., Willard, M., Smith, D.F., and Cummings, R.D. 2009a. Generation of a natural glycan microarray using 9-fluorenylmethyl chloroformate (FmocCl) as a cleavable fluorescent tag. Anal. Biochem. 395:151-160. Song, X., Lasanajak, Y., Xia, B., Smith, D.F., and Cummings, R.D. 2009b. Fluorescent glycosylamides produced by microscale derivatization of free glycans for natural glycan microarrays. ACS Chem. Biol. 4:741-750. Song, X., Xia, B., Stowell, S.R., Lasanajak, Y., Smith, D.F., and Cummings, R.D. 2009c. Novel fluorescent glycan microarray strategy reveals ligands for galectins. Chem. Biol. 16:36-47.
62 Volume 3
Current Protocols in Chemical Biology
Song, X., Lasanajak, Y., Xia, B., HeimburgMolinaro, J., Rhea, J.M., Ju, H., Zhao, C., Molinaro, R.J., Cummings, R.D., and Smith, D.F. 2011. Shotgun glycomics: A microarray strategy for functional glycomics. Nat. Methods 8:85-90. Stevens, J., Blixt, O., Paulson, J.C., and Wilson, I.A. 2006. Glycan microarray technologies: Tools to survey host specificity of influenza viruses. Nat. Rev. Microbiol. 4:857-864. Stowell, S.R., Arthur, C.M., Dias-Baruffi, M., Rodrigues, L.C., Gourdine, J.P., HeimburgMolinaro, J., Ju, T., Molinaro, R.J., RiveraMarrero, C., Xia, B., Smith, D.F., and Cummings, R.D. 2010. Innate immune lectins kill bacteria expressing blood group antigen. Nat. Med. 16:295-301. Xia, B., Kawar, Z.S., Ju, T., Alvarez, R.A., Sachdev, G.P., and Cummings, R.D. 2005. Versatile fluorescent derivatization of glycans for glycomic analysis. Nat. Methods 2:845-850.
Derivatizing Glycans on the Microscale
63 Current Protocols in Chemical Biology
Volume 3
Visualization and Identification of Fatty Acylated Proteins Using Chemical Reporters Jacob S. Yount,1 Mingzi M. Zhang,1 and Howard C. Hang1 1
The Rockefeller University, New York, New York
ABSTRACT Protein fatty acylation, the covalent addition of a lipid chain at specific amino acids, changes the inherent hydrophobicity of a protein, often targeting it to cellular membrane compartments and regulating protein activity, stability, and interactions. Fatty acylation can be analyzed using chemical reporters that mimic natural lipids and contain bioorthogonal chemical handles, allowing them to be reacted with detection tags such as fluorophores or affinity tags. Our laboratory has used alkynyl-chemical reporters of protein myristoylation, S-palmitoylation, prenylation, and acetylation to provide robust, nonradioactive methods for examining the acylation states of full cellular proteomes and individual proteins of interest by (1) metabolically incorporating these chemical reporters into proteins in living cells; (2) selectively reacting the labeled proteins in cell lysates with azido-rhodamine via click chemistry, and globally visualizing them with fluorescence gel scanning; (3) analyzing protein acylation on individual candidate proteins using immunoprecipitation, click chemistry, and fluorescence gel scanning; and (4) identifying novel fatty acylated proteins by reacting chemical reporter–labeled proteins with azido-biotin via click chemistry and selective retrieval using streptavidin beads. This is particularly valuable for the examining protein S-palmitoylation, which does not occur in readily predicted consensus amino acid motifs. Curr. Protoc. Chem. C 2011 by John Wiley & Sons, Inc. Biol. 3:65-79 Keywords: fatty acylation r S-palmitoylation r click chemistry
INTRODUCTION Protein fatty acylation is the covalent addition of a lipid chain at specific amino acids. This modification changes the inherent hydrophobicity of a protein, often targeting it to cellular membrane compartments (Linder and Deschenes, 2007). Acylation may also regulate protein activity, stability, and protein-protein interactions (Linder and Deschenes, 2007). Fatty acylation can be analyzed using chemical reporters that mimic natural lipids and contain bioorthogonal chemical handles that allow them to be reacted with secondary detection tags such as fluorophores or affinity tags (Charron et al., 2009b). For example, our laboratory has successfully utilized alkynyl-chemical reporters of protein myristoylation (Charron et al., 2009b), S-palmitoylation (Charron et al., 2009b; Yount et al., 2010; Zhang et al., 2010; Fig. 1), prenylation (Fig. 1; Charron et al., 2010), and acetylation (Yang et al., 2010a). Basic Protocol 1 describes metabolic incorporation of these chemical reporters onto proteins in living cells. Basic Protocol 2 describes the global visualization of reporter-labeled proteins by selectively reacting alkyne-containing chemical reporter–labeled proteins in cell lysates with azido-rhodamine via click chemistry and fluorescence gel scanning. Basic Protocol 3 describes analysis of protein acylation on individual candidate proteins using immunoprecipitation, click chemistry, and fluorescence gel scanning. Finally, Basic Protocol 4 allows identification of novel fatty acylated proteins by reacting chemical reporter–labeled proteins with azido-biotin via click chemistry and selective retrieval using streptavidin beads. This may be particularly Current Protocols in Chemical Biology 3: 65-79, May 2011 Published online May 2011 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/9780470559277.ch100225 C 2011 John Wiley & Sons, Inc. Copyright
Visualization of Fatty Acylated Proteins
65 Volume 3
A
O O
S
H N
N H
COOH
H2N
N H
O
N-myristoylation
B
H2N
COOH
N H
O
S-palmitoylation
H N
COOH
O
S-prenylation
O
O OH
O
OH
alk-12
C
S
H N
alk-16
OH
alk-FOH
O N3
5 N
O S
N O
Et2N
HN O Cl
NEt2 ⫺
O N H
O
O
O
N H
NH
N
N
N3
HO
O
azido-azo-biotin
azido-rhodamine
Figure 1 Chemical tools for studying protein fatty acylation. (A) Chemical structures for protein myristoylation of Nterminal glycines, palmitoylation of cysteine residues, and prenylation of C-terminal CaaX motifs (cysteine, two aliphatic amino acids, and a variable amino acid). (B) Alkyne chemical reporters used for studying protein myristoylation (alk-12), palmitoylation (alk-16), and prenylation (alk-FOH). (C) Secondary detection reagents for visualization (azido-rhodamine) or affinity enrichment (azido-azo-biotin) of alkyne chemical reporter–labeled proteins.
valuable for the examination of S-palmitoylomes in different cell types or activation states, as these modifications do not occur on readily predicted consensus amino acid motifs. Overall, these techniques provide robust, nonradioactive methods for examining the acylation states of full cellular proteomes and individual proteins of interest (Fig. 2). BASIC PROTOCOL 1
METABOLIC INCORPORATION OF CHEMICAL REPORTERS OF PROTEIN FATTY ACYLATION IN LIVING CELLS Cellular growth medium is supplemented with alkyne-bearing chemical reporters of fatty acylation, allowing cellular uptake and metabolic incorporation of the reporters onto proteins at sites of acylation. NOTE: The volume of cell culture that should be grown depends upon the type of cell used, amount of protein of interest produced by the cells, and type of analysis being performed. These parameters should be determined before beginning labeling studies. The numbers of HeLa cells required for the analyses described in Basic Protocols 2, 3, and 4 are given in those protocols.
Materials
Visualization of Fatty Acylated Proteins
Cultured cells of interest, growing in appropriate complete cell culture medium Serum-free cell culture medium appropriate for growing the cell type of interest, supplemented with 2% (v/v) charcoal/dextran-stripped fetal bovine serum (FBS; e.g., Invitrogen), 37◦ C DMSO (USB grade) 50 mM alkyne-fatty acid chemical reporter stock in DMSO, e.g., Alk-12 (Charron et al., 2009b; for studying myristoylation) Alk-16 (Sigma-Aldrich cat no. O8382; also see Charron et al., 2009b; for studying palmitoylation) Alk-FOH (Charron et al., 2010; for studying prenylation)
66 Volume 3
Current Protocols in Chemical Biology
visualize alk-labeled protein (Basic Protocol 3)
alk-16: ⫺
⫹
⫺
⫹
rho -N3 fluorescence immunoblot
immunoprecipitation
click chemistry alkyne-labeled protein
alk-16: ⫺
⫹
⫺
⫹
visualize alk-labeled proteome (Basic Protocol 2) rho -N3
cell
chemical reporter labeling (Basic Protocol 1)
click chemistry alkyne-labeled cell lysate
fluorescence coomassie
alk-16: ⫺
⫹
biotin -N3
1) click chemistry
biotin N
N
2) SM removal 3) Streptavidin beads
N
Na2S2O4 elution
identify alk-labeled proteins (Basic Protocol 4)
coomassie
proteomics
Figure 2 Workflow in protocols used for studying protein fatty acylation. Basic Protocol 1 describes labeling of proteins in cells with alkyne chemical reporters of fatty acylation. Basic Protocols 2 to 4 utilize the labeled cells for visualization of labeled proteomes by bioorthogonal ligation with azido-rhodamine (rho, Basic Protocol 2), visualization of individual candidate proteins using immunoprecipitation and bioorthogonal ligation with azido-rhodamine (Basic Protocol 3), and/or affinity enrichment of labeled proteins by bioorthogonal ligation with azido-azo-biotin (biotin) and selective elution from streptavidin beads (Basic Protocol 4). Data shown for Basic Protocols 2 and 4 is from DC2.4 cells labeled with 50 μM alk-16 for 1 hr. Data shown for Basic Protocol 3 is from HeLa cells transfected with HA-tagged IFITM3 protein and labeled with 50 μM alk-16 for 1 hr, followed by immunoprecipitation, click chemistry, and immunoblotting with anti-HA antibodies. Abbrevation: SM, small molecule.
Phosphate-buffered saline (PBS; see recipe), ice-cold Liquid nitrogen or dry-ice/ethanol bath 37◦ C, 5% CO2 cell culture incubator Cell scrapers Refrigerated centrifuge 1. Replace the serum-containing complete cell culture medium with 37◦ C, serum-free cell culture medium, supplemented with 2% charcoal/dextran-filtered FBS, and either DMSO as a solvent control or 20 to 100 μM (final concentration) alkyne-fatty acid chemical reporter. We label our HeLa cells when they are confluent. For other studies, cells may need to be labeled at other stages in the cell cycle when the proteins of interest are being produced. Medium at 37◦ C allows solubilization of fatty acid chemical reporters. Charcoal/dextran-filtered FBS removes lipids present in serum, allowing cellular uptake of lipid chemical reporters without competition from serum lipids.
Visualization of Fatty Acylated Proteins
67 Current Protocols in Chemical Biology
Volume 3
Fatty acid chemical reporters are typically effective and show minimal toxicity when used at concentrations of 20 to100 μM. A concentration of 20 μM can be used as a starting point for most experiments and can be increased if enhanced signal strength is desired.
2. Incubate the cells with chemical reporters 1 hr at 37◦ C. Longer incubation times may be necessary to label protein sites with minimal turnover and to label sites that are modified upon de novo protein synthesis. This should be titrated for each protein of interest, although a 1-hr labeling time is sufficient for most proteins.
3. Harvest the cells by scraping them into suspension and centrifuging 3 min at 1000 × g, 4◦ C. Use of trypsin to harvest the cells may degrade cell surface proteins.
4. Remove the supernatant and wash the cells by suspending them in a large volume of ice-cold PBS (e.g., the volume of the container with the cell pellet). Centrifuge 3 min at 1000 × g, 4◦ C. This removes the serum proteins.
5. Remove the supernatant, and repeat the wash. 6. Remove the supernatant, freeze the washed cell pellets in liquid nitrogen or in a dry ice/ethanol bath, and store up to 6 months at −80◦ C, or continue to Basic Protocol 2, 3, or 4. Cell pellets have been stored for up to 6 months without apparent loss of signal. BASIC PROTOCOL 2
GLOBAL FLUORESCENT PROFILING OF FATTY ACYLATED PROTEINS IN WHOLE-CELL LYSATES Chemical reporters mimic natural lipids but contain bioorthogonal groups allowing a selective chemical reaction for appending detection tags to the labeled proteins. Proteins labeled with alkyne-bearing chemical reporters of fatty acylation are reacted with the azide-bearing detection tag azido-rhodamine via click chemistry, allowing fluorescent visualization of protein bands after SDS-PAGE.
Materials
Visualization of Fatty Acylated Proteins
Chemical reporter–labeled cells (Basic Protocol 1), e.g., ∼1 × 106 HeLa cells/pellet 4% (w/v) sodium dodecyl sulfate (SDS) with EDTA-free protease inhibitors (see recipe) 250 U/μl Benzonase (Sigma-Aldrich, ultrapure) BCA assay reagents (Pierce Protein Research) 5 mM azido-rhodamine(Charron et al., 2009b) in dimethyl sulfoxide (DMSO) or tetramethylrhodamine-5-carbonyl azide (Invitrogen; see Martin and Cravatt, 2009) in DMSO 50 mM tris(2-carboxyethyl)phosphine (TCEP): prepare fresh 2 mM tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl] amine (TBTA) in 1:4 (v/v) DMSO/butanol 50 mM CuSO4 : prepare fresh Methanol, ice-cold Chloroform, ice-cold Water, ice-cold 4× loading buffer (see recipe) 2-mercaptoethanol (2-ME) 18-well 4% to 20% Tris·Cl protein gels (Bio-Rad)
68 Volume 3
Current Protocols in Chemical Biology
Fluorescent protein molecular weight ladder (15 to 250 kDa; Bio-Rad), diluted as necessary Fluorescence gel destaining solution: 40% (v/v) methanol/50% (v/v) acetic acid/10% (v/v) water Coomassie blue staining reagents (Pierce Protein Research) 1.5-ml microcentrifuge tubes Bath sonicator (optional) 95◦ C heating block Fluorescence gel scanner with a 532-nm excitation and 580-nm detection filters and 30-nm band-pass (e.g., Typhoon 9400, Amersham Biosciences) Additional reagents and equipment for quantifying protein concentration using BCA (Olson and Markwell, 2007), carrying out SDS-PAGE (Gallagher, 2006), and staining proteins (e.g., see Sasse and Gallagher, 2009) Lyse cells 1. Lyse chemical reporter–labeled cells (e.g., ∼1 × 106 HeLa cells) by adding 50 μl of 4% SDS buffer with EDTA-free protease inhibitors and 1 μl (250 U) of Benzonase to the cell pellet and vortexing. Lysing cells with SDS-containing buffer maximally solubilizes the proteins. Protease inhibitors containing EDTA are not compatible with Benzonase or click chemistry reactions.
2. Quantify the protein concentrations using a standard BCA assay (Olson and Markwell, 2007). Protein concentrations obtained for cell lines are generally 1 to 10 mg/ml, depending on the cell type.
3. Dispense aliquots of equal amounts of protein (∼50 μg) for each sample into 1.5-ml microcentrifuge tubes, and bring the volumes to 44.5 μl with 4% SDS buffer with EDTA-free protease inhibitors. Generally, 50 μg total protein is adequate for visualization of fatty acylation with chemical reporters. Using less protein makes precipitation more difficult in later steps.
Ligate labeled proteins to azido-rhodamine 4. Prepare a click chemistry master mix, by combining the following volumes of reagents per sample: 1 μl of 5 mM azido-rhodamine in DMSO 1 μl of 50 mM TCEP 2.5 μl of 2 mM TBTA in 1:4 (v/v) DMSO/butanol 1 μl of 50 mM CuSO4 . We synthesize the azido-rhodamine used in this step (synthesis described in Charron et al., 2009b). Tetramethyl rhodamine-5-carbonyl azide has also been used in click chemistry reactions (Martin and Cravatt, 2009) and is commercially available. Solutions of TCEP and CuSO4 should be prepared fresh for each experiment. A master mix ensures that equal amounts of each reagent are added to each sample.
5. Add 5.5 μl of the click chemistry master mix to each protein sample for a final volume of 50 μl, and vortex to mix. 6. Incubate 1 hr at room temperature. Visualization of Fatty Acylated Proteins
69 Current Protocols in Chemical Biology
Volume 3
Remove unreacted molecules 7. Perform a chloroform/methanol precipitation of protein to remove unreacted azidorhodamine by adding the following ice-cold reagents to each tube: 200 μl methanol 75 μl chloroform 150 μl water. Vortex, and centrifuge 15 min at 20,000 × g, at 4◦ C. Precipitation reagents should be ice-cold.
8. Remove and discard the upper aqueous phase, leaving the lower organic phase and the white layer of protein between the two layers. The upper aqueous layer should be clear, and the lower organic layer contains the pink-colored rhodamine. A white layer of protein should have formed between the two layers.
9. Add 1 ml of ice-cold methanol to each sample and mix gently, causing the protein pellet to sink to the bottom of the tube. Centrifuge 10 min at 20,000 × g, 4◦ C. The protein pellet should be a light pink color.
10. Remove all of the liquid by pipetting, being careful not to disturb the pellet. Wash the protein pellet by adding 1 ml of ice-cold methanol and inverting the tube. Centrifuge 10 min at 20,000 × g, 4◦ C. 11. Carefully remove the methanol leaving behind the protein pellet. Allow the remaining methanol to evaporate by leaving the sample tubes open on the bench for 20 min at room temperature. The dried pellet will be a gray/white color.
Resolubilize proteins for SDS-PAGE 12. Add 50 μl of 4% SDS buffer with EDTA-free protease inhibitors to dissolve the protein pellets. A bath sonicator may be used at this step to speed the solubilization process. Pellets are generally dissolved within 5 sec using a sonicator.
13. Add 17.5 μl of 4× loading buffer and 3.5 μl of 2-ME to each sample (5% 2-ME final concentration), and vortex the samples. DTT has been used successfully as an alternative to 2-mercaptoethanol.
14. Denature the proteins for 5 min at 95◦ C, using a heating block. 15. Vortex, and centrifuge 1 min at 5000 × g. room temperature. 16. Load 20 μl of samples and diluted fluorescent protein ladder onto an 18-well 4% to 20% Tris·Cl gel, and electrophorese 1 hr at 200 V. A fluorescent protein ladder should be used as a molecular weight standard. Commercially available protein ladders may require as much as a 1:10,000 dilution with 4% SDS buffer and appropriate addition of 4× loading buffer to avoid saturation and bleed-through of the standard into other lanes of the gel.
Destain gel and scan 17. Destain the gel by rocking for 1 hr at room temperature in fluoresecence gel destaining solution to remove remaining traces of loading buffer and unreacted rhodamine. Visualization of Fatty Acylated Proteins
70 Volume 3
Current Protocols in Chemical Biology
18. Scan the gel using a fluorescence gel scanner. For the azido-rhodamine derivative shown in Figure 1, an Amersham Biosciences Typhoon 9400 scanner has been successfully used with 532-nm excitation and 580-nm detection filters with 30-nm band-pass. For more information about azido-rhodamine visualization, see Charron et al. (2009b).
19. Stain the gel with Coomassie blue (e.g., see Sasse and Gallagher, 2009) to demonstrate equal protein loading (see Fig. 2).
FLUORESCENT VISUALIZATION OF FATTY ACYLATION ON CANDIDATE PROTEINS
BASIC PROTOCOL 3
Proteins labeled with chemical reporters of fatty acylation are immunoprecipitated prior to click chemistry ligation with azido-rhodamine and SDS-PAGE.
Materials Chemical reporter–labeled cells (Basic Protocol 1), e.g., ∼2 × 106 HeLa cells/pellet 1% (w/v) Brij97 buffer with EDTA-free protease inhibitors (see recipe) BCA assay reagents (Pierce Protein Research) 4% (w/v) SDS buffer with EDTA-free protease inhibitors (see recipe) 5 mM azido-rhodamine (Charron et al., 2009b) in DMSO or tetramethylrhodamine-5-carbonyl azide (Invitrogen; see Martin and Cravatt, 2009) in DMSO 50 mM tris(2-carboxyethyl)phosphine (TCEP): prepare fresh 2 mM tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl] amine (TBTA) in 1:4 (v/v) dimethyl sulfoxide (DMSO)/butanol 50 mM CuSO4 : prepare fresh 4× loading buffer (see recipe) 2-mercaptoethanol (2-ME) 25% (w/v) hydroxylamine (NH2 OH), pH 7.0 (optional) Fluorescent protein molecular weight ladder (10 to 250 kDa, dual color; Bio-Rad), diluted as necessary 18-well 4% to 20% Tris·Cl protein gels (Bio-Rad) Fluorescence gel destaining solution: 50% (v/v) methanol/10% (v/v) acetic acid/40% (v/v) H2 O 1.5-ml microcentrifuge tubes 95◦ C heating block Fluorescence gel scanner with a 532-nm excitation and 580 nm detection filter and 30 nm band-pass (e.g., Typhoon Amersham Biosciences 9400, scanner) Additional reagents and equipment for quantifying protein concentration using BCA (Olson and Markwell, 2007), performing immunoprecipitation techniques (Bonifacino et al., 2006), carrying out SDS-PAGE (Gallagher, 2006), staining proteins on gels (Sasse and Gallagher, 2009), and performing immunoblotting (Gallagher et al., 2008) 1. Lyse chemical reporter–labeled cells (e.g., ∼2 × 106 HeLa cells) by adding 100 μl of 1% Brij97 buffer with EDTA-free protease inhibitors to the cell pellet. Centrifuge 5 min at 1000 × g, 4◦ C, to remove cellular debris. Other lysis buffers compatible with immunoprecipitation may be substituted for Brij97 buffer. Protease inhibitors containing EDTA are not compatible with click chemistry.
Visualization of Fatty Acylated Proteins
71 Current Protocols in Chemical Biology
Volume 3
2. Quantify protein concentrations using a standard BCA assay (e.g., see Olson and Markwell, 2007). 3. Dispense 500 μg of protein for each sample into 1.5-ml microcentrifuge tubes. Bring the volumes to 100 μl with 1% Brij97 buffer with EDTA-free protease inhibitors. Increased quantities of proteins may be needed, particularly when examining proteins of low cellular abundance. Generally, 500 μg is adequate for the study of overexpressed proteins and most native proteins.
4. Carry out standard immunoprecipitation techniques for the protein of interest (e.g., see Bonifacino et al., 2006), and wash the agarose beads with the antibody-linked protein at least three times with lysis buffer. 5. Add 22.25 μl of 4% SDS buffer with EDTA-free protease inhibitors to the pelleted agarose beads used for the immunoprecipitations. The addition of 4% SDS buffer will dissociate antibody-protein complexes. Thus, proteins will no longer be associated with beads and additional washes subsequent to this step should not be performed.
6. Prepare a click chemistry master mix, by combining the following volumes of components per sample:
0.5 μl of 5 mM azido-rhodamine in DMSO 0.5 μl of 50 mM TCEP 1.25 μl of 2 mM TBTA in 1:4 (v/v) DMSO/butanol 0.5 μl of 50 mM CuSO4 . Tetramethyl rhodamine-5-carbonyl azide has also been used instead of the azidorhodamine in click chemistry reactions (Martin and Cravatt, 2009) and is commercially available.
7. Add 2.75 μl of the click chemistry reagent master mix to each sample (containing the bead pellets and the dissociated protein) for a final volume of 25 μl, and vortex to mix. 8. Incubate 1 hr at room temperature. 9. Add 9 μl of 4× loading buffer and 1.8 μl of 2-ME to each sample (5% 2-ME final concentration), and vortex. 10. Optional: Split the sample into two microcentrifuge tubes, and add 2 μl of 25% neutral NH2 OH to one tube (2.5% final concentration) and 2 μl of water to the second tube, and mix by vortexing. Continue to step 11. NH2 OH is known to cleave thioester bonds characteristic of protein palmitoylation, and thus fluorescence signal should be lost or greatly reduced for proteins that are labeled with alkyne-palmitate reporter on cysteine residues.
11. Denature the proteins 5 min at 95◦ C, using a heating block. 12. Vortex, and centrifuge 1 min at 9000 × g, room temperature, to pellet agarose beads used earlier for immunoprecipitation. 13. Load 15 μl of the samples (avoiding the pelleted beads) and a diluted fluorescent protein ladder onto duplicate 18-well 4% to 20% Tris·Cl gels, and electrophorese 1 hr at 200 V (e.g., see Gallagher, 2006). Gel 1 will be used for fluorescent gel scanning, and gel 2 will be used in immunoblotting for the protein of interest as a loading control. Visualization of Fatty Acylated Proteins
If samples were split for NH2 OH treatment, additional lysis buffer and 4× loading buffer can be added before loading onto gels. Alternatively, lower volumes can be loaded.
72 Volume 3
Current Protocols in Chemical Biology
A fluorescent protein ladder should be used as a molecular weight standard. Commercially available protein ladders may require as much as a 1:10,000 dilution with 4% SDS-buffer and appropriate addition of 4× loading buffer for the fluorescent gel to avoid saturation and bleed-through of the standard into other lanes of the gel.
14. To visualize fluorescence: Destain gel 1 with fluorescence gel destaining solution, and perform fluorescence gel scanning as described in Basic Protocol 2, steps 17 and 18. 15. To perform an immunoblot: Transfer proteins from gel 2 onto a nitrocellulose membrane, and probe for the protein of interest using standard immunoblotting techniques (e.g., see Gallagher et al., 2008).
PROTEOMIC IDENTIFICATION OF PROTEINS LABELED WITH CHEMICAL REPORTERS OF FATTY ACYLATION
BASIC PROTOCOL 4
Proteins labeled with alkyne-bearing chemical reporters of fatty acylation are reacted with the azide-bearing detection tag azido-azo-biotin via click chemistry, allowing selective retrieval of the labeled proteins using streptavidin beads. Na2 S2 O4 cleavage of the azo group allows specific elution of labeled proteins from beads, without boiling.
Materials Chemical reporter–labeled cells (Basic Protocol 1), e.g., ∼20 × 106 HeLa cells/pellet 4% (w/v) SDS buffer with EDTA-free protease inhibitors (see recipe) 250 U/μl Benzonase BCA assay reagents (Pierce Protein Research) 5 mM azido-azo-biotin (Yang et al., 2010b) or azide-biotin (Invitrogen; also see Martin and Cravatt, 2009) 50 mM and 200 mM tris(2-carboxyethyl)phosphine (TCEP): prepare fresh 2 mM tris[(1-benzyl-1H -1,2,3-triazol-4-yl)methyl] amine (TBTA) in 1:4 (v/v) dimethyl sulfoxide (DMSO)/butanol 50 mM CuSO4 : prepare fresh Methanol, ice-cold Chloroform, ice-cold Water, ice-cold 0.5 M EDTA 1% Brij97 buffer with EDTA-free protease inhibitors (see recipe) Streptavidin agarose beads (Thermo Scientific) PBS, ice cold, pH 7.4 (see recipe) PBS/0.2% (w/v) SDS, pH 7.4 (ice-cold) ABC buffer (250 mM ammonium bicarbonate), ice-cold 8 M urea 400 mM iodoacetamide: prepare fresh 1% (w/v) SDS Na2 S2 O4 elution buffer (see recipe) NuPAGE 4× LDS buffer (Invitrogen): dilute to 1× before use) 2-mercaptoethanol (2-ME) Fluorescent protein molecular weight ladder (15 to 250 kDa; Bio-Rad) 18-well 4% to 20% Tris·Cl protein gels (Bio-Rad) Coomassie blue staining reagents (Pierce Protein Research) 15-ml conical, polypropylene tubes Nutating mixer 2-ml dolphin microcentrifuge tubes Green Centricon devices (YM-10 membranes with 10 kDa molecular weight cutoff; Millipore)
Visualization of Fatty Acylated Proteins
73 Current Protocols in Chemical Biology
Volume 3
Rotary evaporater (e.g., SpeedVac) 95◦ C heating block Additional reagents and equipment for quantifying protein concentration using BCA (Olson and Markwell, 2007), carrying out SDS-PAGE (Gallagher, 2006), and staining gels for protein (Sasse and Gallagher, 2009) 1. Lyse chemical reporter–labeled cells (e.g., ∼20 × 106 HeLa cells) by adding 1 ml of 4% SDS buffer with EDTA-free protease inhibitors and 1 μl (250 units) of Benzonase to the cell pellet. Protease inhibitors containing EDTA are not compatible with Benzonase or click chemistry reactions.
2. Quantify protein concentrations using a standard BCA assay. 3. Dispense equal amounts (>10 mg) of protein for each sample into 15-ml conical, polypropylene tubes and bring the volume to 4.45 ml with 4% SDS buffer with EDTA-free protease inhibitors. A minimum of 10 mg total protein is recommended.
Ligate chemical reporter–labeled proteins with azido-azo-biotin 4. Prepare a click chemistry master mix by combining the following volumes of components per sample: 100 μl of 5 mM azido-azo-biotin 100 μl of 50 mM TCEP 250 μl of 2 mM TBTA in1:4 DMSO/butanol 100 μl of 50 mM CuSO4 . Solutions of TCEP and CuSO4 should be prepared fresh for each experiment. Azide-biotin is available from Invitrogen and has also been successfully used in click chemistry reactions and proteomic studies (Martin and Cravatt, 2009).
5. Add 550 μl of the click chemistry master mix to each protein sample (final reaction volume of 5 ml), and vortex to mix. 6. Incubate the click chemistry reaction 1.5 hr at room temperature. 7. Perform a methanol/chloroform precipitation of the protein as in Basic Protocol 2, steps 7 to 11, scaling up the volumes proportionately (100×), and washing the protein pellet twice with 50 ml ice cold methanol, dissociating the protein pellet by pipeting. 8. Resuspend the protein pellet in 1 ml of 4% SDS buffer with or without EDTA-free protease inhibitors, supplemented with 20 μl of 0.5 M EDTA solution. EDTA will chelate remaining copper and prevent any further click chemistry reaction.
9. Quantify the resuspended protein by BCA assay (e.g., see Olson and Markwell, 2007). Generally, 50% to 80% of proteins are recovered after precipitation.
Bind reporter-labeled and biotin-ligated proteins to streptavidin beads 10. Add 5 mg protein in 1 ml of 4% SDS buffer with EDTA-free protease inhibitors to individual 15-ml Falcon tubes. Visualization of Fatty Acylated Proteins
11. Add 2 ml of 1% Brij97 buffer with EDTA-free protease inhibitors to dilute the SDS concentration to a level compatible with biotin/streptavidin binding (i.e., less than 1.5% SDS).
74 Volume 3
Current Protocols in Chemical Biology
12. Wash 100 μl of streptavidin agarose beads per 5 mg protein with 50 ml PBS three times, centrifuging the beads for 2 min at 4000 × g, room temperature, to remove the supernatant after each wash. 13. Resuspend the washed streptavidin agarose beads in 100 μl of 1% Brij97 buffer with EDTA-free protease inhibitors per sample, and add 100 μl of the bead suspension to each protein sample. 14. Rock the protein/bead mixture on a nutating mixture 1 hr at room temperature.
Wash streptavidin beads 15. Wash beads by inverting several times in 10 ml PBS/0.2% SDS, and centrifuge 2 min at 4000 × g, 4◦ C. 16. Discard the supernatant, and wash the beads three times with 10 ml PBS, centrifuging 2 min at 4000 × g, 4◦ C, between washes. 17. Discard the supernatant, and wash the beads twice with 10 ml of 250 mM ABC buffer, centrifuging 2 min at 4000 × g, 4◦ C, between washes. Washing minimizes nonspecific pulldown of proteins.
Cap cysteines for mass spectrometry analysis 18. Remove the supernatant, and add the following: 500 μl 8 M urea 25 μl 200 mM TCEP 25 μl 400 mM iodoacetamide. Mix by pipetting, and incubate 30 min at room temperature. This reaction alkylates cysteines, thereby preventing formation of high molecular weight aggregates by random disulfide bonding.
19. Add 10 ml ABC buffer and centrifuge 2 min at 4000 × g, 4◦ C. 20. Remove the supernatant, resuspend the beads in 1 ml ABC buffer, and transfer the suspension to 2-ml dolphin tubes. 21. Centrifuge 2 min at 4000 × g, 4◦ C.
Elute labeled proteins from streptavidin beads 22. Remove the supernatant. add 250 μl Na2 S2 O4 elution buffer to the beads, resuspend the beads by pipetting, and incubate 30 min at room temperature. 23. Centrifuge 2 min at 4000 × g, 4◦ C, and save the supernatant containing the eluted proteins. 24. Add an additional 250 μl Na2 S2 O4 elution buffer to the beads and incubate 30 min at room temperature. 25. Centrifuge 2 min at 4000 × g, 4◦ C, collect the eluent, and combine with the previously eluted proteins.
Concentrate eluted proteins 26. Wash a green Centricon device with 500 μl of 1% SDS and centrifuge 30 min at 9000 × g, room temperature. 27. Wash the Centricon with 500 μl water and centrifuge 30 min at 9000 × g, room temperature.
Visualization of Fatty Acylated Proteins
75 Current Protocols in Chemical Biology
Volume 3
28. Add 500 μl of eluted proteins to the Centricon, and centrifuge 30 min at 9000 × g, room temperature. 29. Turn the Centricon upside down and centrifuge into a collection tube 3 min at 1000 × g, room temperature. 30. Add 50 μl of 1% SDS supplemented with 75 mM of 2-ME (5.3 μl 2-ME in 1 ml SDS solution)to the Centricon reservoir, pipet up and down several times, and centrifuge the upside-down Centricon into the same collection tube 3 min at 1000 × g, room temperature. 31. Repeat step 30. The final volume will be approximately 125 μl.
32. Lyophilize the samples to a solid white powder using a rotary evaporator.
Perform SDS-PAGE of eluted proteins 33. Resuspend the white powder in 25 μl of NuPAGE 1× LDS containing 5% (v/v) of 2-ME, and vortex vigorously. 34. Incubate the samples for 5 min on a 95◦ C heating block. 35. Centrifuge 1 min at 1000 × g, room temperature. 36. Load 20 μl protein mixture for each sample onto a 4% to 20% Tris·Cl gel. For gel slices to be cut for proteomic analysis, blank lanes should be placed between samples. This will aid in preventing contamination during gel slicing for gel-based proteomics. Blanks should contain LDS buffer with ∼20% SDS and 5% 2-mercaptoethanol. If SDS is not added to the blank lanes, the sample proteins may expand horizontally in the gel during electrophoresis. A fraction of the retrieved proteins can be saved for immunoblotting to confirm that the proteins identified by mass spectrometry (step 38) are indeed selectively retrieved.
37. Electrophorese the gel 1 hr at 200 V (e.g., see Gallagher, 2006), and stain with Coomassie blue (e.g., see Sasse and Gallagher, 2009). 38. Cut gel slices for proteomic analysis according to standard procedures (see Jim´enez et al., 1998).
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps.
Brij97 buffer (1%) with EDTA-free protease inhibitors, pH 7.4 1% (w/v) Brij97 150 mM NaCl 50 mM triethanolamine (dilute 1 M triethanolamine stock, pH 7.4) Sterilize by passing through a 0.22-μm filter, and store up to 1 year at room temperature. Just before use, add EDTA-free protease inhibitor cocktail tablets (Roche) according to the manufacturer’s directions. Loading buffer, 4×
Visualization of Fatty Acylated Proteins
40% (v/v) glycerol 240 mM Tris·Cl, pH 6.8 8% (w/v) sodium dodecyl sulfate (SDS) 0.04% (w/v) bromphenol blue Store up to 1 year at room temperature
76 Volume 3
Current Protocols in Chemical Biology
Na2 S2 O4 elution buffer 250 mM ammonium bicarbonate 0.1% (w/v) sodium dodecyl sulfate (SDS) 25 mM Na2 S2 O4 Prepare fresh for each experiment Phosphate-buffered saline (PBS), pH 7.4 144 mg/liter KH2 PO4 9 g/liter NaCl 95 mg/liter Na2 HPO4 ·7H2 O Sterilize by passing through a 0.22-μm filter, and store up to 1 year at 4◦ C SDS buffer (4%) with EDTA-free protease inhibitors, pH 7.4 4% (w/v) sodium dodecyl sulfate (SDS) 150 mM NaCl 50 mM triethanolamine (dilute 1 M triethanolamine stock, pH 7.4) Sterilize by passing through a 0.22-μm filter, and store up to 1 yr at room temperature. Just before use, add EDTA-free protease inhibitor cocktail tablets (Roche) according to the manufacturer’s directions. COMMENTARY Background Information The use of chemical reporters and bioorthogonal ligation chemistries has a rich history in the field of protein glycosylation (Prescher and Bertozzi, 2006) and has recently been extended to the study of protein myristoylation, S-palmitoylation (Charron et al., 2009b; Yount et al., 2010; Zhang et al., 2010), prenylation (Charron et al., 2010; Fig. 1A,B), and acetylation (Yang et al., 2010a). Protein acylation has classically been studied using radiolabeled lipids. However, this method is hazardous and provides low sensitivity, often requiring film exposure times of weeks to months. For example, the bacterial effector protein SifA was demonstrated to be prenylated using radiolabeled lipids in an in vitro translation system, but radiolabeled lipids did not provide the sensitivity needed to detect SifA prenylation in living cells (Reinicke et al., 2005). In contrast, study of protein acylation with alkyne-bearing chemical reporters and bioorthogonal ligation with azido-rhodamine provides a robust and sensitive way to visualize these modifications (Charron et al., 2009a). For example, SifA prenylation in living cells was recently confirmed using a prenylation reporter, allowing identification of classes of cellular transferases responsible for SifA prenylation (Charron et al., 2010). The increased sensitivity of chemical reporters is also accompanied by an increase in efficiency of experimental workflow, as a typical cellular labeling and visualization procedure can be
completed in a single workday without long exposure times. Chemical reporters coupled with biotin and selective elution of labeled proteins allows the identification of novel acylated proteins (Martin and Cravatt, 2009; Yount et al., 2010). This method has proved fruitful in the identification of new palmitoylated proteins, which has traditionally been difficult, as this modification does not occur on a predictable motif. The interferon-induced transmembrane protein 3 (IFITM3) was discovered in this way to be palmitoylated and, using subsequent labeling and visualization of IFITM3 mutants, the sites of palmitoylation were mapped (Yount et al., 2010). Palmitoylation was found to be essential for full antiviral activity of IFITM3 against influenza virus, demonstrating the unique and interesting biological findings that can be elucidated with a chemical reporter approach (Yount et al., 2010). Other methods for identifying novel palmitoylated proteins have been described e.g., acyl-biotin exchange (ABE) chemistry, which has been important in expanding our knowledge of the multitude of specific proteins on which palmitoylation can occur (Roth et al., 2006). ABE chemistry and chemical reporters afford complementary tools for analyzing S-fatty acylation. Chemical reporters of acylation are likely to be especially useful for identifying novel acylated proteins during different cellular states, e.g., differentiation, apoptosis, or cellular
Visualization of Fatty Acylated Proteins
77 Current Protocols in Chemical Biology
Volume 3
infection. For example, chemical reporters of myristoylation, which occurs predictably on N-terminal glycines, could allow identification of new myristoylation sites that are revealed upon caspase cleavage of proteins during apoptosis. Likewise, C-terminal CaaX motifs may also be identified after protein cleavage events or on uncharacterized protein isoforms with the use of chemical reporters and proteomic methods.
Critical Parameters and Troubleshooting
Visualization of Fatty Acylated Proteins
Cellular labeling of proteins with reporters of fatty acylation requires complete solubilization of the reporter, as precipitates are toxic to most cell types. To this end, chemical reporter should always be added to media at 37◦ C and mixed thoroughly. Washes of labeled cells should be done quickly and with ice-cold PBS to avoid turnover of some types of acylation, thus maximizing the detected lipidation signal. The choice of detergents for lysing cells is critically important to the success of the experiment, particularly when performing immunoprecipitations. While SDS-containing buffer maximally solubilizes proteins and is especially useful for acylated proteome profiling, Brij97-containing buffer also solubilizes many lipidated membrane proteins and is compatible with immunoprecipitations. When performing the click chemistry reaction, it is imperative that fresh solutions of TCEP and CuSO4 be used. TCEP may become oxidized over time, and Cu(I) is required for correct coordination of the reaction. Upon completion of SDS-PAGE, destaining the gel destined for fluorescence gel scanning greatly improves signal to noise, as residual azidorhodamine in the lanes may cause background signal. High quality reagents should be used to prepare the destain solution to avoid any contaminants that may produce a fluorescent signal. Likewise, the gel should be handled only on the edges, as physical manipulations of the gel can often appear as smudges or fingerprints in fluorescence gel scans even when gloves are worn. The fluorescence gel molecular weight standard should also be titrated so that bleedthrough into other lanes does not occur. Wiping the pipet tip before loading and starting the current immediately upon addition may also help if bleedthrough of the standard is observed. For selective retrieval of acylated proteins, protein input of 5 to 10 mg is a good starting point. Smaller amounts of protein may limit
detection to only the most abundant proteins. In the precipitation step following click chemistry with azido-azo-biotin, it is important that the protein pellet is washed twice with icecold methanol so that unreacted biotin does not block streptavidin binding sites on the beads used for pulldown of labeled proteins. Following the methanol washes, using 4% SDS buffer and sonication is an effective way to completely resolubilize the protein pellet. Because 4% SDS is not compatible with the biotinstreptavidin binding, the SDS concentration must be diluted for this step. It has also been reported that two elutions using Na2 S2 O4 are more effective than one longer elution (Verhelst et al., 2007), and improved protein recovery from the streptavidin beads is observed when performing two elutions. Concentration of the eluted protein with a Centricon is necessary for buffer exchange prior to SDS-PAGE.
Anticipated Results Using alkyne-bearing chemical reporters and azido-modified detection reagents, negligible signal should be seen in the DMSO-treated control lanes. An experiment is deemed successful if signal is detected above these control lanes for both rhodamine visualization and Coomassie staining. The magnitude of the signal can be increased or decreased as necessary by changing the concentration of chemical reporter used, changing the labeling time, or by modifying the amount of protein added to each lane of the gel. If signal is undetectable, the investigator should label Jurkat T cells with the alk-16 reporter as a positive control, as these cells are efficiently labeled with this reporter at 20 μM for 1 hr. Likewise, an immunoprecipitation for endogenous palmitoylated Lck from alk-16 labeled Jurkat cell lysate can also serve as a positive control using a mouse anti-Lck monoclonal antibody (Clone 3A5 from Invitrogen; Zhang et al., 2010). For proteomic experiments, the authors have routinely identified more than 100 selectively retrieved lipidated proteins when starting with 5 mg or more of total protein in cell lysate. When using lipidation reporters of different chain lengths, alk-12 preferentially labels myristoylated proteins while alk-16 preferentially labels palmitoylated proteins. Minimal, but observable, cross-talk has been observed for both reporters (Charron et al., 2009b; Martin and Cravatt, 2009; Wilson et al., 2010; Yount et al., 2010), likely due to substrate promiscuity of the respective acyltransferases.
78 Volume 3
Current Protocols in Chemical Biology
Time Considerations Basic Protocol 1 can typically be completed in 2 hr, and cell pellets can be frozen, or cells can be utilized in Basic Protocols 2 or 3 in the same day. Both protocols typically require ∼6 hr before fluorescence gel scans are obtained. Loading controls for these experiments, i.e., Coomassie gel staining and immunoblotting, respectively, can also be completed the same day. Alternatively, membranes for blotting can be blocked overnight and probed the following day. Likewise, the Coomassie stain can be applied overnight, or the destain procedure can be performed overnight at 4◦ C. Throughout Basic Protocols 2 and 3 there are at least three 1-hr incubation periods during which the experimenter can focus on other tasks. Basic Protocol 4 is a more lengthy procedure, and requires approximately 15 hr to prepare samples for proteomic analysis, although this can easily be spread over 2 days, as there are several stopping points within the protocol. There are also several 1-hr or 30-min incubation periods. If gelbased proteomics is desired, then another 2 days of sample preparation time should be planned.
Literature Cited Bonifacino, J.S., Dell’Angelica, E.C., and Springer, T.A., 2006. Immuunoprecipitation. Curr. Protoc. Neurosci. 5.24.1-5.24.28. Charron, G., Wilson, J., and Hang, H.C. 2009a. Chemical tools for understanding protein lipidation in eukaryotes. Curr. Opin. Chem. Biol. 13:382-391. Charron, G., Zhang, M.M., Yount, J.S., Wilson, J., Raghavan, A.S., Shamir, E., and Hang, H.C. 2009b. Robust fluorescent detection of protein fatty-acylation with chemical reporters. J. Am. Chem. Soc. 131:4967-4975. Charron, G., Tsou, L.K., Maguire, W., Yount, J.S., and Hang, H.C. 2010. Alkynyl-farnesol reporters for detection of protein S-prenylation in cells. Mol. Biosyst. 7:67-73. Gallagher, S.R. 2006. One-dimensional SDS gel electrophoresis of proteins. Curr. Protoc. Mol. Biol. 75:10.2A.1-10.2A.37. Gallagher, S.R, Winston, S.E., Fuller, S.A., and Hurrell, J.G.R. 2008. Immunoblotting and immunodetection. Curr. Protoc. Mol. Biol. 10.8.110.8.28. Jim´enez, C.R., Huang, L., Qiu, Y., and Burlingame, A.L. 1998. In-gel digestion of proteins for
MALDI-MS fingerprint mapping. Curr. Protoc. Protein Sci. 14:16.4.1-16.4.5. Linder, M.E. and Deschenes, R.J. 2007. Palmitoylation: Policing protein stability and traffic. Nat. Rev. Mol. Cell Biol. 8:74-84. Martin, B.R. and Cravatt, B.F. 2009. Large-scale profiling of protein palmitoylation in mammalian cells. Nat. Methods 6:135-138. Olson, J.S.C. and Markwell, J. 2007. Assays for determination of protein concentration. Curr. Protoc. Protein Sci. 3.4.1-3.4.29. Prescher, J.A. and Bertozzi, C.R. 2006. Chemical technologies for probing glycans. Cell 126:851854. Reinicke, A.T., Hutchinson, J.L., Magee, A.I., Mastroeni, P., Trowsdale, J., and Kelly, A.P. 2005. A Salmonella typhimurium effector protein SifA is modified by host cell prenylation and S-acylation machinery. J. Biol. Chem. 280:14620-14627. Roth, A.F., Wan, J., Bailey, A.O., Sun, B., Kuchar, J.A., Green, W.N., Phinney, B.S., Yates, J.R., 3rd, and Davis, N.G. 2006. Global analysis of protein palmitoylation in yeast. Cell 125:10031013. Sasse, J. and Gallagher, S.R. 2009. Staining proteins in gels. Curr. Protoc. Molec. Biol. 10.6.110.6.27. Verhelst, S.H., Fonovic, M., and Bogyo, M. 2007. A mild chemically cleavable linker system for functional proteomic applications. Angew. Chem. Int. Ed. Engl. 46:1284-1286. Wilson, J.P., Raghavan , A.S., Yang, Y.Y., Charron, G., and Hang, H.C. 2010. Proteomic analysis of fatty-acylated proteins in mammalian cells with chemical reporters reveals S-acylation of histone H3 variants. Mol. Cell Proteomics 10:M110.001198. Yang, Y.Y., Ascano, J.M., and Hang, H.C. 2010a. Bioorthogonal chemical reporters for monitoring protein acetylation. J. Am. Chem. Soc. 132:3640-3641. Yang, Y.Y., Grammel, M., Raghavan, A.S., Charron, G., and Hang, H.C. 2010b. Comparative analysis of cleavable azobenzene-based affinity tags for bioorthogonal chemical proteomics. Chem. Biol. 17:1212-1222. Yount, J.S., Moltedo, B., Yang, Y.Y., Charron, G., Moran, T.M., Lopez, C.B., and Hang, H.C. 2010. Palmitoylome profiling reveals Spalmitoylation-dependent antiviral activity of IFITM3. Nat. Chem. Biol. 6:610-614. Zhang, M.M., Tsou, L.K., Charron, G., Raghavan, A.S., and Hang, H.C. 2010. Tandem fluorescence imaging of dynamic S-acylation and protein turnover. Proc. Natl. Acad. Sci. U.S.A. 107:8627-8632.
Visualization of Fatty Acylated Proteins
79 Current Protocols in Chemical Biology
Volume 3
Miniaturized High-Throughput Fluorescent Assay for Conversion of NAD(P)H to NAD(P) Andrew D. Napper1,2 and Sharmila Sivendran1,3 1
Penn Center for Molecular Discovery, Institute for Medicine and Engineering and Department of Chemical and Biomolecular Engineering, University of Pennsylvania, Philadelphia, Pennsylvania 2 Nemours Center for Childhood Cancer Research, Wilmington, Delaware 3 GlaxoSmithKline, Collegeville, Pennsylvania
ABSTRACT This unit describes a miniaturized fluorescence assay that monitors the conversion of NADPH to NADP+ . The same assay format may also be used to measure NADH to NAD+ conversion. Examples of assay development and validation results are presented to illustrate the use of this protocol to screen an enzyme that consumes NADPH as a cofactor during conversion of substrate to a reduced product. Enzymatic assays are carried out in low-volume 384-well plates, in which the turnover of NADPH is monitored by the decrease in fluorescence emission at 460 nm between an initial measurement and a second reading after 90 min. A follow-up assay is used to rule out false-positive artifacts C arising from compounds that fluoresce at 460 nm. Curr. Protoc. Chem. Biol. 3:81-97 2011 by John Wiley & Sons, Inc. Keywords: NADPH detection r NADH detection r fluorescence r high-throughput screening r oxidoreductase r enzyme assay r fluorescent artifact
BASIC PROTOCOL We have optimized a miniaturized high-throughput screening assay that monitors the decrease in fluorescence at 460 nm upon conversion of NADPH to NADP+ . To correct for the intrinsic fluorescence at 460 nm of some test compounds, two readings of each plate are obtained: one immediately after compound and reagent addition, and the second after a 90-min incubation with the target enzyme. Nevertheless, a slight increase in compound fluorescence readings over 90 min is sufficient to give rise to a number of false positives (compounds that appear to be inhibitors of the target enzyme but in fact are not). A counterscreen is used to identify these false positives and exclude them from further study. Two assays are described here: Primary screening assay: NADPH fluorescence assay: Enzyme activity is monitored by the decrease in NADPH fluorescence over 90 min. Counterscreen assay: Fluorescence change in the absence of enzyme and substrate: Fluorescence change due to test compounds is monitored in assay buffer alone. The optimization of these assays and their use to confirm screening hits is exemplified by assay validation and HTS to discover inhibitors of two enzymes, RmlC and RmlD, essential for cell wall biosynthesis in Mycobacterium tuberculosis (Sivendran et al., 2010). The two enzymes are assayed together as a mixture: RmlC epimerizes the keto sugar nucleotide, dTDP-4-keto-6-deoxy-D-xylo-hexulose, and RmlD then uses the cofactor NADPH to reduce the epimerized keto sugar nucleotide to dTDP-rhamnose (Fig. 1).
Current Protocols in Chemical Biology 3: 81-97, June 2011 Published online June 2011 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/9780470559277.ch100155 C 2011 John Wiley & Sons, Inc. Copyright
HTS Fluorescent Assay for NAD(P)H to NAD(P)
81 Volume 3
O O HO
CH3
O
HN
O OH O O P O P O O
O
TDP-KDX
OH
O RmlC
H
CH3
HN
N
O
O
O CH3
RmlD
H3C HO HO
O
O
NADPH
P O P O O
OH
O
O
N
O
O OH
H
NADP
Figure 1 RmlC/D enzyme assay. RmlC epimerizes the keto sugar nucleotide dTDP-4-keto-6-deoxy-D-xylo-hexulose (TDP-KDX), and RmlD then uses the cofactor NADPH to reduce the epimerized keto sugar nucleotide to dTDP-rhamnose. Enzyme activity is determined by monitoring the decrease in NADPH fluorescence.
To simplify and to provide a clear example, the protocol describes the method used for the Rml enzymes assay. The Critical Parameters and Troubleshooting section describes how the Rml protocol may be modified and broadened to be generally applicable to a wider range of NADPH- or NADH-utilizing enzymes.
Materials NADPH (see recipe) Assay buffer (see recipe) Target enzyme(s): For the study used as an example here, the M. tb rhamnosyl biosynthetic enzymes dTDP-6-deoxy-D-xylo-4-hexulose 3,5-epimerase (RmlC) and dTDP-6-deoxy-L-lyxo-4-hexulose reductase (RmlD) were cloned and expressed in E. coli and purified by Michael McNeil and co-workers (Ma et al., 2001); these enzymes are not commercially available Substrates: TDP-6-deoxy-D-xylo-hexopyranosid-4-ulose (TDP-KDX) was synthesized enzymatically and stored frozen at a concentration of 1 mg/ml in 50 mM MOPS buffer (Sigma), pH 7.4, at −80◦ C as previously described (Sivendran et al., 2010); this substrate is not commercially available Positive control inhibitor: thymidine diphosphate (TDP; Sigma) Analytical-grade dimethyl sulfoxide (DMSO; Fisher or VWR), anhydrous Test compounds: Store dissolved in DMSO (e.g., at 10 mM stock concentration) in 384-well polypropylene plates (room temperature storage is recommended if the compounds are to be reused within a period of less than one week; for longer term storage, compounds stocks should be frozen at <−20◦ C)
HTS Fluorescent Assay for NAD(P)H to NAD(P)
Pipetting workstation equipped with 384-tip MDT pipetting head and with pintool consisting of 384 pins with nominal transfer volume of 100 nl, e.g., a JANUS from Perkin Elmer or equivalent (earlier model known as Evolution EP3) (Rudnicki and Johnston, 2009) Compound dilution plates: Polypropylene V-bottom 384-well plates (Greiner Bio-One, cat. no. 781280) Assay plates: Black low-volume 384-well plate (Corning, cat. no. 3676) Plate reader capable of reading fluorescence in 384-well plates (e.g., an EnVision multimode plate reader from Perkin Elmer) GraphPad Prism (or equivalent): for graphing data and curve fitting for IC50 calculation Multichannel pipettor 25◦ C incubator Reagent dispenser (e.g., a Multidrop-384 reagent dispenser from Thermo Scientific; see Rudnicki and Johnston, 2009)
82 Volume 3
Current Protocols in Chemical Biology
500-μl microcentrifuge tubes Microsoft Excel, OpenHTS (CeuticalSoft), or ActivityBase (IDBS) (or equivalent): for percent inhibition calculations, and evaluation of datasets and selection of hits Generate NADPH standard curve 1. Dissolve NADPH in water to a concentration of 1 mM. 2. Perform NADPH dilution as follows: a. Serially dilute NADPH in one column of a 384-well polypropylene plate by 1.5-fold dilutions in assay buffer (1×). b. Add 105 μl of 80 μM NADPH in assay buffer to the first well and 35 μl of assay buffer to nine of the remaining wells in the column. c. Serially transfer 70 μl from the first well to the second, then from the second to the third, and so on until 70 μl is added to and then discarded from the ninth well to yield nine concentrations of NADPH ranging from 3 to 80 μM and one well containing buffer alone. 3. Assay start: Transfer 10 μl per well of serially diluted NADPH (from step 2) to three columns of a 384-well assay plate to give triplicate wells at each concentration of NADPH. 4. Read the fluorescence (excitation 340 nm, emission 460 nm). 5. Standard curve: Plot the fluorescence against the NADPH concentration. Fit to a straight line by linear regression in GraphPad Prism or similar curve-fitting and graphing program.
Perform enzyme titration and timecourse 6. Enzyme dilution: Make dilutions of enzymes (2× final concentration in assay) in assay buffer (1×). In the case of RmlC and RmlD, a useful starting point was provided by an earlier 96well absorbance assay, in which optimal concentrations were determined to be 1.75 × 10-4 μg/μl RmlC and 4.35 × 10-4 μg/μl RmlD (Ma et al., 2001). To confirm that these concentrations are also appropriate for the 384-well fluorescent assay, prepare solutions to give final enzyme concentrations of 0.5, 1.0, 1.5, 2.0, and 3.0 times the above values in the 10-μl fluorescence assay.
7. Prepare the substrate and cofactor stock as follows: a. Mix the substrate and NADPH cofactor (2× final concentration in assay) in assay buffer (1×). Based on prior assay results (Ma et al., 2001), use final assay concentrations of 400 μM TDP-KDX and 50 μM NADPH. Thus, a solution containing 800 μM TDP-KDX and 100 μM NADPH should be prepared in 1× assay buffer.
b. Add 5 μl per well of this TDP-KDX/NADPH solution to a block of 15 wells (5 down by 3 across) in a 384-well assay plate using a multichannel pipettor. Include an adjacent block of 15 blank wells containing NADPH cofactor without substrate. 8. Assay start: Using a multichannel pipettor, transfer 5 μl of the enzyme dilutions to the substrate and cofactor mix and blanks to give triplicates of each enzyme concentration in both the substrate and cofactor mix and the blank wells (final assay volume 10 μl).
HTS Fluorescent Assay for NAD(P)H to NAD(P)
83 Current Protocols in Chemical Biology
Volume 3
9. Timecourse: Read the fluorescence (excitation 340 nm, emission 460 nm) immediately and at 3-min intervals thereafter for 180 min. Incubate the plate between reads at 25◦ C. Provided the assay plate temperature does not exceed 25◦ C, it may be left uncovered for the duration of the assay; some evaporation from the assay wells will occur, but this should not be enough to adversely affect the results.
Analyze the data 10. For each concentration of enzyme(s), plot fluorescence at each timepoint (0 to 180 min). In addition, plot fluorescence of blanks to ensure that change over timecourse is minimal. For screening, select the concentration of enzyme(s) giving the largest total change in fluorescence that also displays a linear change in fluorescence over 90 min. 11. From the NADPH standard curve (step 5), convert the change in fluorescence to the amount of NADPH converted. This allows determination of the total amount of NADPH converted over the 3-hr timecourse.
Determine Km values The goal of the Km determination studies is to select appropriate concentrations of substrate and NADPH cofactor for compound screening. To maximize the sensitivity of the assay to enzyme inhibitors, substrate and cofactor should be used close to their respective Km values. Substrate Km is determined by varying substrate concentration at a fixed concentration of cofactor, and likewise cofactor Km is determined by varying cofactor concentration at a fixed concentration of substrate. For further explanation and discussion of this methodology, see Km determinations under Critical Parameters and Troubleshooting. Useful literature references include: Assay Guidance Manual, Enzyme General 2010, How to Measure Km (listed under Internet resources) and Fersht (1985). Determine substrate Km 12. Serially dilute TDP-KDX in one column of a 384-well polypropylene compound dilution plate by 1.5-fold dilutions in assay buffer (1×) containing 30 μM NADPH cofactor to yield 10 concentrations of substrate ranging from 0 to 2000 μM in the presence of a fixed concentration of NADPH cofactor. 13. Add 5 μl of each TDP-KDX dilution per well to two columns of a 384-well assay plate using a multichannel pipettor, giving duplicate wells of each TDP-KDX dilution. 14. Assay start: Add 5 μl per well of a mixture of 3.5 × 10−4 μg/μl RmlC and 8.7 × 10−4 μg/μl RmlD in assay buffer (1×). Thus, each 10-μl assay has 15 μM NADPH, 1.75 × 10−4 μg/μl RmlC, 4.35 × 10−4 μg/μl RmlD, and 0 to 1000 μM TDP-KDX substrate. 15. Read the fluorescence (excitation 340 nm, emission 460 nm) immediately and after 90 min at 25◦ C. 16. Proceed to step 21 for data analysis.
HTS Fluorescent Assay for NAD(P)H to NAD(P)
Determine NADPH cofactor Km 17. Serially dilute NADPH in one column of a 384-well polypropylene compound dilution plate by 1.5-fold dilutions in assay buffer (1×) containing 400 μM TDP-KDX to give twelve concentrations of cofactor ranging from 0 to 200 μM in the presence of a fixed concentration of TDP-KDX substrate.
84 Volume 3
Current Protocols in Chemical Biology
18. Add 5 μl per well of each NADPH dilution to two columns of a 384-well assay plate using a multichannel pipettor, yielding duplicate wells of each NADPH dilution. 19. Assay start: Add 5 μl per well of a mixture of 3.5 × 10−4 μg/μl RmlC and 8.7 × 10−4 μg/μl RmlD in assay buffer (1×). Thus, each 10-μl assay has 200 μM substrate, 1.75 × 10−4 μg/μl RmlC, 4.35 × 10−4 μg/μl RmlD, and 0 to 100 μM NADPH. 20. Read the fluorescence (excitation 340 nm, emission 460 nm) immediately and after 90 min at 25◦ C.
Data analysis (for both Km determinations) 21. For each well, subtract the fluorescence after 90 min from the fluorescence at the start of the assay to determine change in fluorescence. 22. From the NADPH standard curve (step 5), convert the change in fluorescence to the amount of NADPH converted over 90 min. This may be expressed as a rate (e.g., μM NADPH/min), allowing determination of Vmax (maximal velocity of the enzymatic conversion at saturating concentration of substrate).
23. Plot the rate of NADPH conversion against concentration (of substrate or cofactor, respectively). 24. Fit the rate of NADPH conversion vs. concentration plot to sigmoidal curve using GraphPad Prism or a similar curve-fitting and graphing program. The curve fit will approach a maximum rate at high concentrations of the x-value (substrate or cofactor concentration). This maximum rate represents Vmax , the maximal velocity of the enzymatic conversion at saturating concentration of the varied reagent (substrate or cofactor). The Km value determined from the curve fit is defined as the concentration of varied reagent (substrate or cofactor) that gives a rate equal to half the Vmax .
Determine the positive control inhibitor dose-response and IC50 value The aim of this procedure is to determine the IC50 value (inhibitory concentration giving half maximal response) for TDP, a known inhibitor of the RmlC/D enzymes (Sivendran et al., 2010). The measured IC50 is then compared with previously determined values to ensure that the enzymes show the expected sensitivity to inhibition. TDP dose-response 25. Dissolve TDP in water to a concentration of 75 mM. Divide into 100-μl aliquots and store in 500-μl microcentrifuge tubes frozen at −20◦ C.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
1
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64
65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80
81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96
97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112
113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128
129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144
145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160
161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176
177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192
193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208
209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224
225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240
241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256
257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272
273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288
289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304
305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320
Blanks (no TDP-KDX substrate) Negative controls (no test compounds) Test wells (compound number)
Figure 2 Assay plate format. The layout of blanks and negative controls shown here is also used for the QC plate and the dose-response plates.
85 Current Protocols in Chemical Biology
Volume 3
26. Serially dilute TDP in one column of a 384-well polypropylene compound dilution plate by 2-fold dilutions in water from a top concentration of 75 mM to yield 16 concentrations of positive control inhibitor ranging from 2.3 μM to 75 mM (note that this will be 2.5× the final concentration in assay). 27. Add 4 μl of water per well to columns 1-2 and 23-24 of a 384-well assay plate using a multichannel pipettor. Columns 1 and 23 will be used as blanks, and columns 2 and 24 as no-inhibitor controls. See Figure 2: location of blanks is labeled as such, and no-inhibitor controls are labeled negative controls.
28. Add 4 μl per well of serially diluted TDP (from step 26) to columns 3, 4, and 5 of the assay plate using a multichannel pipettor, giving triplicate columns of TDP dilutions. 29. Add 5 μl per well of assay buffer (1.67×) containing 5.26 × 10−4 μg/μl RmlC, 13.06 × 10−4 μg/μl RmlD, and 50 μM NADPH to columns 1-5 and 23-24 using a multichannel pipettor or reagent dispenser (columns 6-22 are empty for this experiment). Final reagent concentrations: 2.63 × 10−4 μg/μl Rml C, 6.53 × 10−4 μg/μl Rml D, and 25 μM NADPH in 1× assay buffer.
30. Add 1 μl per well of assay buffer (1.67×) into columns 1 and 23 (blank) using a reagent dispenser. 31. Assay start: Add 1 μl of 2 mM TDP-KDX substrate (diluted in 1.67× assay buffer) per well to columns 2-5 and 24 using a reagent dispenser. Final concentration of TDP-KDX is 200 μM.
32. Read the fluorescence (excitation 340 nm, emission 460 nm) immediately and after 90 min at 25◦ C.
Determine IC50 values 33. For each well, subtract the fluorescence after 90 min from the fluorescence at the start of the assay to determine the change in fluorescence. 34. Use Equation 1 to calculate percent activity using the fluorescence change at each TDP concentration and the mean fluorescence change of the plate blanks (columns 1 and 23), and the mean fluorescence change of the plate controls (columns 2 and 24) (see plate map in Fig. 2):
⎡ (fluorescence change − blank mean) ⎤ Percent activity = 100 × ⎢ ⎥ (control mean − blank mean) ⎦ ⎣ Equation 1
35. Plot the percent activity against the concentration of the inhibitor [use log scale for inhibitor concentration (x-axis) and linear scale for percent activity (y-axis)].
HTS Fluorescent Assay for NAD(P)H to NAD(P)
36. Fit the percent activity vs. log of concentration to a sigmoidal dose-response curve. Use the four-parameter logistic equation entitled “log(inhibitor) vs. response— variable slope” in GraphPad Prism or an equivalent curve-fitting equation in another graphing program. Report IC50 , Hill slope (defined as slope at mid-point of curve), and top and bottom values.
86 Volume 3
Current Protocols in Chemical Biology
For further explanation and discussion of IC50 determination, see Assay Guidance Manual, Enzyme General 2010, IC50 determination (listed under Internet resources). For a very extensive discussion of nonlinear regression, with numerous examples including IC50 curves, see the GraphPad Prism Regression Guide (listed under Internet Resources).
Quality control (QC) validation plate The IC50 value determined for the positive control inhibitor TDP (steps 33 to 36) is used to set the concentration for a QC plate to monitor the consistency of the assay between plates and between experiments performed on different days. QC plate set up 37. Dissolve TDP in DMSO to a concentration of 40.3 mM. At least 8 ml of TDP in DMSO is required for step 38. The remainder may be stored frozen up to 6 months at −86◦ C.
38. Add 20 μl TDP in DMSO to each well in columns 3-22 of a polypropylene V-bottom 384-well plate (TDP storage plate). Add 20 μl DMSO alone to each well in columns 1-2 and 23-24 for blanks and negative controls (see plate map in Fig. 2). Plate may be used immediately or sealed with an adhesive plate seal and stored at room temperature for up to 2 weeks before use.
Testing of QC plate 39. Dispense 4 μl of water per well to a 384-well assay plate using a reagent dispenser. 40. Pintool transfer TDP in DMSO from the TDP storage plate (120 nl per well). Final concentrations of TDP and DMSO are 500 μM and 1.2%, respectively.
41. Add 5 μl assay buffer (1.67×) containing 5.26 × 10−4 μg/μl RmlC, 13.06 × 10−4 μg/μl RmlD, and 50 μM NADPH in each well. Final reagent concentrations are 2.63 × 10−4 μg/μl Rml C, 6.53 × 10−4 μg/μl Rml D, and 25 μM NADPH.
42. Add 1 μl assay buffer (1.67×) into each well in columns 1 and 23 (blank) using a reagent dispenser. 43. Assay start: Add 1 μl of 2 mM TDP-KDX substrate in 1.67× assay buffer to each well in columns 2-22 and 24 using a reagent dispenser. Final concentration of the TDP-KDX is 200 μM.
44. Read the fluorescence (excitation 340 nm, emission 460 nm) immediately and after 90 min at 25◦ C.
Calculation of percent inhibition 45. For each well, subtract the fluorescence after 90 min from the fluorescence at the start of the assay to determine the change in fluorescence. 46. Use Equation 2 to calculate percent inhibition using the fluorescence change in each TDP test well (columns 3-22), the mean fluorescence change of the plate blanks (columns 1 and 23), and the mean fluorescence change of the plate negative controls (columns 2 and 24) (see plate map in Fig. 2):
⎧ ⎡ (fluorescence change − blank mean) ⎤ ⎫ Percent inhibition = 100 × ⎨1 − ⎢ ⎥ (control mean − blank mean) ⎦ ⎬ ⎩ ⎣ ⎭
HTS Fluorescent Assay for NAD(P)H to NAD(P)
Equation 2
87 Current Protocols in Chemical Biology
Volume 3
Conduct a primary screening assay (single-concentration compound screening) 47. Follow steps 39 to 44, except that pintool transfer (step 40) into each assay plate is from a compound storage plate containing test compounds in DMSO in columns 3-22 and DMSO alone in columns 1-2 and 23-24. The final test compound concentration should not exceed 25 μM (e.g., transfer of 120 nl of 10 mM compound in DMSO gives 12 μM compound in 1.2% DMSO in the final assay volume). Reserve wells in columns 1-2 and 23-24 for negative controls and blanks treated with DMSO but no compound (as shown in Fig. 2). Each screening run should include testing of at least one QC plate (steps 37-46) per 20 test compound plates.
Analyze the data and select hits 48. For data analysis, follow steps 45 to 46. 49. For hit selection, select hits based on percent inhibition exceeding a defined threshold. For the hits to be statistically significant and not just due to random scatter in the data, the threshold should be below the mean of the plate controls by a minimum of 3× the standard deviation of the negative controls. The standard deviation of the negative controls in enzyme inhibition assays typically ranges from 5% to 10%; thus, the minimal hit threshold may be set at 20% to 30% inhibition. To select for only higher potency hits, the threshold may be set at 50% inhibition.
Perform dose-response testing This protocol is identical to single compound screening (step 47), except that compounds are tested at multiple concentrations obtained by serial dilution. A suitable layout is 16 two-fold dilutions of each compound, giving an assay concentration range of 50 μM to 1.5 nM. In this case, compounds may be serially diluted vertically down each plate such that columns 3-22 each contain the dilutions of a different compound. The protocol for serial dilution in DMSO is as described in steps 50 to 53. The resulting dose-response plates may be used immediately or sealed with an adhesive plate seal and stored up to 1 week at room temperature before use. 50. Add 20 μl of each compound stock (4.2 mM in DMSO) arranged 20 per plate in wells A3 to A22 of 384-well V-bottom polypropylene plates. 51. Add 20 μl DMSO to wells A1, A2, A23, and A24. 52. Add 10 μl DMSO to each well in the entire plate, except row A, using a reagent, dispenser. 53. Two-fold serially dilute compounds by transferring 10 μl row-by-row from row A to row P using a single row of disposable tips, then discarding 10 μl from row P. The resulting dose-response plates contain 16 two-fold dilutions of each compound, ranging from 4.2 mM to 125 nM, arranged one compound per column in columns 3-22. Pintool transfer from dose-response plates into assay plates, as described in step 47, gives a final compound dilution range of 50 μM to 1.5 nM.
Perform curve fitting, IC50 determination, and hit confirmation 54. Following dose-response testing (steps 50 to 53), fit the data and determine IC50 values as described in steps 33 to 36. Compounds giving an IC50 below the top concentration tested (50 μM) are deemed to be confirmed hits.
HTS Fluorescent Assay for NAD(P)H to NAD(P)
Counterscreen assay (elimination of false positives) Confirmed hits from the primary screening assay (steps 47 to 54) must be tested in the counterscreen assay to eliminate false positives. These false positives are compounds that erroneously appear to be enzyme inhibitors due to compound fluorescence that increases between the 0 min and 90 min reads during the assay.
88 Volume 3
Current Protocols in Chemical Biology
Dose-response plate testing 55. Dispense 4 μl of water per well in a 384-well assay plate using a reagent dispenser. 56. Pintool transfer the compound in DMSO (120 nl per well) from the dose-response plate (step 53) into the assay plate to give a final compound dilution range of 50 μM to 1.5 nM. 57. Assay start: Add 6 μl per well of assay buffer (1.67×) using a reagent dispenser. 58. Read the fluorescence (excitation 340 nm, emission 460 nm) immediately and after 90 min at 25◦ C.
EC50 determination 59. For each well, subtract the fluorescence at the start of the assay from the fluorescence after 90 min to determine the fluorescence increase (if any). 60. For each compound, plot the fluorescence increase against the compound concentration [use log scale for compound concentration (x-axis) and linear scale for fluorescence increase (y-axis)]. 61. Fit the fluorescence increase vs. log of concentration to a sigmoidal dose-response curve. Use a four-parameter logistic equation in GraphPad Prism or a similar curvefitting and graphing program. Report EC50 , Hill slope (defined as slope at mid-point of curve), and top and bottom values. The compound concentration at the midpoint of the curve is reported as EC50 (effective concentration giving half maximal response), rather than IC50 as reported in step 36, as the compound concentration gives rise to an increase in assay signal rather than the decrease observed in step 36. For an extensive discussion of nonlinear regression, with numerous examples including EC50 curves, see the GraphPad Prism Regression Guide (listed under Internet resources).
Elimination of false positives Compounds giving a measurable increase in fluorescence over 90 min and an EC50 <3× IC50 (determined in step 54) are deemed to be false-positive artifacts and should be eliminated from further study. A compound EC50 value less than 3-fold greater than its IC50 determined in step 54 strongly suggests that the compound “activity” in the primary screening assay was due to its intrinsic fluorescence and not to inhibition of either of the enzymes RmlC and RmlD. Conversely, compounds that give no measurable EC50 in the counterscreen assay, or an EC50 much higher than the corresponding IC50 obtained in the primary screening assay, are unlikely to be false positives due to intrinsic fluorescence. These compounds should therefore be selected for further study as inhibitors of RmlC and RmlD. This is explained in more detail in Critical Parameters and Troubleshooting under the Counterscreen assay section. REAGENTS AND SOLUTIONS Use Milli-Q purified water or equivalent in all recipes and protocol steps.
Assay buffer, 1× 50 mM MOPS (Sigma), pH 7.4, containing 1 mM MgCl2 , 0.01% Triton X-100 (Sigma), and 10% (w/v) glycerol. Buffer without added glycerol may be stored as a 5× stock after filtration through a sterile 0.45-μm filter. Glycerol is added upon dilution of the buffer to the final 1× concentration.
NADPH NADPH was purchased from Roche, dissolved in water to a concentration of 1 mM, and stored frozen at −20◦ C.
HTS Fluorescent Assay for NAD(P)H to NAD(P)
89 Current Protocols in Chemical Biology
Volume 3
COMMENTARY Background Information Monitoring of the conversion of NADH to NAD+ (or NADPH to NADP+ ) has been used for many years to assay the activity of enzymes catalyzing oxidation-reduction reactions mediated by NADH/NAD as cofactor. An assay for lactate dehydrogenase (LDH) that monitors the decrease in absorbance at 340 nm upon conversion of NADH to NAD was published in 1963 (Bergmeyer and Bernt, 1963). Two characteristics of absorbancebased methods such as this are not wellsuited to high-throughput microplate-based compound screening: 1. A relatively small change in absorbance limits the sensitivity of the assay and requires that it is monitored as a timecourse. 2. Absorbance measurements in microplates require a flat well bottom and a liquid depth of several millimeters. The first characteristic limits throughput because each plate has to be monitored for at least 15 min. By contrast, an assay with a single endpoint read allows for the processing of multiple plates in this time. The second characteristic requires an assay volume of at least 50 μl in 384-well plates. Absorbance assays in 1536-well plates are not practical. This limitation leads to increased reagent consumption and cost compared with lower volume formats of 384- and 1536-well assays, which may be run in volumes of 10 μl or less. Several approaches to NADH monitoring have been taken to avoid the limitations of absorbance assays. NADH fluoresces at 460 nm (excitation at 340 nm), and decrease in fluorescence at this wavelength may be used to monitor NADH to NAD conversion (Moran and Schnellmann, 1996). To increase detection sensitivity, NADH may be quantified by coupling enzymatic conversion to NAD with the generation of resorufin, a highly fluorescent molecule with an emission maximum of 590 nm (excitation 530 nm). One assay uses NADH oxidase to generate hydrogen peroxide (H2 O2 ) from NADH, followed by horseradish peroxidase to convert Amplex Red to resorufin in the presence of H2 O2 (Batchelor and Zhou, 2002). Another method uses the enzyme diaphorase to convert resazurin to resorufin in the presence of NADPH (Batchelor and Zhou, 2004).
Critical Parameters and Troubleshooting The protocol above describes exactly the method used for the Rml enzymes assay (Sivendran et al., 2010) and the Anticipated Results section presents results obtained using this assay. Below is a discussion of how the Rml protocol may be modified and broadened to be generally applicable to a wider range of NADPH- or NADH-utilizing enzymes. Enzyme titration and timecourse The purpose of the enzyme titration and timecourse (steps 6 to 11) is to determine an enzyme concentration that produces a linear decrease in NADPH fluorescence over 90 min while maximizing the total change in fluorescence over this time. Longer reaction times result in evaporation from low-volume 384well assay plates, leading to more scatter in the data. Shorter times will require more enzyme in order to produce a detectable amount of product. Deviation from linearity beyond a certain point in the timecourse generally indicates depletion of the substrate. It is important to adjust enzyme concentration and time to achieve linearity; a nonlinear timecourse reduces the sensitivity of the assay to enzyme inhibition. Factors to consider in applying this protocol to other enzymes a. Enzyme dilution. Serial dilution of the target enzyme(s) in a single column in a polypropylene 384-well V-bottom plate provides enzyme dilutions in the appropriate format for direct transfer to an assay plate. The concentrations and dilution range used depends on the extent of enzyme activity information already available. In the absence of prior data suggesting a suitable enzyme concentration, seven semi-log dilutions ranging from a 1/10 dilution of the enzyme stock to 1/10,000 will cover a broad range of enzyme activities. Typically, the target enzyme has already been assayed for activity during purification and characterization. Although established assays may not be suitable for HTS, they may help to guide the choice of enzyme concentrations to test. In the case of RmlC and RmlD, the activities had already been determined in an NADPH absorbance assay (see Introduction), so the enzyme
HTS Fluorescent Assay for NAD(P)H to NAD(P)
90 Volume 3
Current Protocols in Chemical Biology
A
B Change in NADPH fluorescence
Change in NADPH fluorescence
320,000 280,000 240,000 200,000 160,000 120,000 80,000 40,000 0
300
600 TDP-KDX (M)
900
200,000 160,000 120,000 80,000 40,000
0
12
24 NADPH (M)
36
Figure 3 Km determinations for TDP-KDX (A) and NADPH (B). The protocol was based on that described under steps 12 to 24, Km determinations. Data points represent the mean of triplicate determinations; error bars depict standard deviation. (A) The concentration of TDP-KDX was varied from 0 to 1000 μM in the presence of a fixed concentration of NADPH (25 μM). The fluorescence change in the presence of a mixture of 2.63 × 10−4 μg/μl Rml C, 6.53 × 10−4 μg/μl Rml D in 1× assay buffer was monitored over 90 min at 25◦ C. Data were fit to a sigmoidal plot using GraphPad Prism, and the Km of TDP-KDX was calculated to be 200 μM. (B) The concentration of NADPH was varied from 0 to 36 μM in the presence of a fixed concentration of TDP-KDX (200 μM). The fluorescence change in the presence of a mixture of 2.63 × 10−4 μg/μl Rml C, 6.53 × 10−4 μg/μl Rml D in 1× assay buffer was monitored over 90 min at 25◦ C. Data were fit to a sigmoidal plot using GraphPad Prism, and the Km of NADPH was calculated to be 12.5 μM.
concentrations were varied over a small range (8.75 × 10−5 to 5.25 × 10−4 μg/μl RmlC and 2.175 × 10−4 to 1.31 × 10−3 μg/μl RmlD). b. Substrate and cofactor stock. In the absence of results from a prior enzyme activity assay, several combinations of substrate and NAD(P)H cofactor may need to be tested ranging in concentration from 1 to 500 μM. An enzyme assay will rarely require concentrations outside this range. Typically, the enzyme will have already been tested for activity, so the conditions of this earlier assay will guide the choice of suitable substrate and cofactor concentrations. In the case of the 384-well Rml enzyme titration and timecourse experiments, 400 μM TDP-KDX and 50 μM NADPH were used based on previous results obtained in a 96-well absorbance assay. Subsequently, these values were adjusted for the HTS assay based on Km determinations carried out in the 384-well assay (steps 12 to 24; and see below under Km determinations for more information about using Km values to select reaction conditions). c. Timecourse (step 9 in the Basic Protocol). It is not useful to extend the timecourse of a 10 μl assay beyond 180 min at 25◦ C because evaporation causes significant variability in the data. If the enzyme requires incubation at 37◦ C, acceptable data cannot be obtained beyond 1 hr.
Km determinations The goal of the Km determination studies is to select appropriate concentrations of substrate and NAD(P)H cofactor for compound screening. To maximize the sensitivity of the assay to enzyme inhibitors, regardless of their mechanism of inhibition, substrate and cofactor should be used close to their respective Km values. In the case of the Rml assay used to exemplify this protocol, TDP-KDX substrate was used at its Km of 200 μM, but NADPH was used at twice its Km of 12.5 μM. Use of NADPH at a concentration two times higher than its Km gave a more robust and linear change in fluorescence over the 90-min timecourse of the assay. The relationship of inhibitor IC50 to substrate concentration and how this relates to the mechanism of inhibition is explained in the Assay Guidance Manual, Section 13, Performing MOA Studies (listed under Internet Resources). Detailed discussion of this topic may also be found in Enzyme Kinetics: Behavior and Analysis of Rapid Equilibrium and Steady-State Enzyme Systems (Segel, 1975). Factors to consider in applying this protocol to other enzymes a. Ultimately, the Km of the substrate should be determined in the presence of cofactor fixed at its Km , and vice versa. This may require several iterations as the Km values for each are
HTS Fluorescent Assay for NAD(P)H to NAD(P)
91 Current Protocols in Chemical Biology
Volume 3
NADPH fluorescence
250,000 200,000
DMSO control (100% activity)
150,000 TDP: 66% inhibition @ 500 M
100,000 50,000
Blanks (no TDP-KDX; 0% activity)
0 0
40 80 120 160 290 240 280 320 360 Well number
Figure 4 Representative QC plate data. Each point represents the change in NADPH fluorescence over 90 min at 25◦ C in a single well in a 384-well assay plate. This part of the protocol is described under steps 37 to 45, Quality control (QC) validation plate. TDP was added to test wells (columns 3-22) at a final concentration of 500 μM.
refined. In the absence of prior results, Km determinations may need to be carried out initially with the fixed component at a high concentration, close to saturation. The analysis becomes simpler if it appears that the Km of one component is not affected by the concentration of the other. b. Data analysis. Typically, Km data show simple saturation-binding kinetics, and may be fit accordingly. In the case of the Rml enzymes, the Km curves were sigmoidal (Fig. 3). This may be indicative of enzyme cooperativity or allosteric modulation by substrate or cofactor, but can also arise for other reasons (Segel, 1975). For the purposes of assay development, the Km values obtained from the sigmoidal curves may be used without further analysis to determine the cause of the sigmoidal behavior. Positive control inhibitor dose-response and IC50 determination The aim of this procedure is to determine IC50 value for TDP, a known inhibitor of the RmlC/D enzymes (Sivendran et al., 2010). The measured IC50 is then compared with previously determined values to ensure that the enzymes show the expected sensitivity to inhibition. The IC50 value is used to set the concentration for a QC plate to monitor the consistency of the assay over time.
HTS Fluorescent Assay for NAD(P)H to NAD(P)
Factors to consider in applying this protocol to other enzymes A known inhibitor of an enzyme target to be subjected to HTS is a valuable positive control to use as a benchmark to track the sensitivity of the enzyme toward inhibition during assay development and over the course of the screening campaign. A suitable inhibitor does not need to be potent provided it is soluble at a concentration that gives 30% to 70% inhibi-
tion. It should ideally be a reversible, substratecompetitive inhibitor that does not covalently modify or denature the protein. Quality control (QC) validation plate The aim of this procedure is to deliver TDP (an inhibitor) by pintool transfer to an entire 384-well assay plate (320 wells, columns 3-22) at a concentration slightly over its IC50 value. The calculated percent inhibition (expected to be ∼65%) is used to monitor well-to-well variability and consistency of the assay over time. Factors to consider in applying this protocol to other enzymes Figure 4 reveals three outliers (wells 48, 78, and 215) that appear to deviate significantly from the remainder of the TDP-inhibited data. When outliers such as these are observed, a repeat of the experiment will reveal whether there is a problem associated with a specific pin in the pintool or a specific channel in the reagent dispenser. If the outliers are not replicated at the same well locations between experiments, then they may be attributed to random data scatter. If further refinement of the liquid handling steps fails to reduce the occurrence of such outliers, screening may nonetheless proceed. False-positive outliers will be eliminated during dose-response testing (steps 50 to 53) and other follow-up assays. The occasional false negative (a true inhibitor missed during HTS) may also be deemed acceptable, provided the incidence is very low. Dose-response testing Factors to consider in applying this protocol to other enzymes It is important that compounds be serially diluted in DMSO. Transfer of each dilution
92 Volume 3
Current Protocols in Chemical Biology
A
B
Low compound concentration
Low compound concentration
Fluorescence
Fluorescence
50% increase in fluorescence 85% “inhibition”
Compound fluorescence
50% increase in fluorescence
Baseline (no compound fluorescence)
15% “inhibition” 0
90 Time (min)
0
90 Time (min)
Figure 5 The effect on the NADPH assay of change in compound fluorescence over time increases with compound concentration. The percent increase in compound fluorescence is independent of compound concentration; thus, the absolute fluorescence change increases with increasing concentration. The concentration of NADPH is fixed, so the fluorescence decrease due to enzymatic conversion of NADPH does not change from well to well. Therefore, as the concentration of a fluorescent compound is increased, the magnitude of its fluorescence “drift” relative to the decrease in NADPH fluorescence becomes larger, resulting in a dose-dependent increase in “inhibition”. (A and B) The purple shaded area represents compound fluorescence, which increases the total fluorescence above the baseline fluorescence in the absence of added compound as shown. The green line represents the decrease in NADPH fluorescence over time. The fluorescence measured in the assay (shown as an orange line) is the sum of the fluorescence due to NADPH and the compound fluorescence. As an illustrative example, compound fluorescence is shown increasing by 50% over the course of the 90 min assay. (A) At low compound concentration, a 50% increase in compound fluorescence partially offsets the decrease in NADPH fluorescence, resulting in apparent 15% inhibition of NADPH conversion. (B) At high compound concentration, a 50% increase in compound fluorescence almost entirely offsets the decrease in NADPH fluorescence, resulting in apparent 85% inhibition of NADPH conversion.
into the assay plate by pintool then gives a constant percent DMSO in each well, avoiding the possibility of well-to-well differences caused by the effect of DMSO alone on the enzyme activity. Serial dilution of compounds from a DMSO stock into buffer also may result in compound precipitation. Counterscreen assay (elimination of false positives) The purpose of the counterscreen assay is to test hits from the primary screening assay (steps 47 to 54) to eliminate false positives due to compound fluorescence that increases between the 0 min and 90 min reads in the assay. One cause of the increase in compound fluorescence over time might be slow evaporation, which would tend to concentrate the fluorescent compound in the center of the round-bottom plate well. An explanation for how the change in compound fluorescence gives rise to a dose-
response curve that appears similar to that of a bona fide enzyme inhibitor is presented in Figures 5 and 6. The percent increase in compound fluorescence is independent of compound concentration; thus, the absolute fluorescence change increases with increasing concentration. The concentration of NADPH is fixed, so the fluorescence decrease due to enzymatic conversion of NADPH does not change from well to well. Therefore, as the concentration of a fluorescent compound is increased, the magnitude of its fluorescence “drift” relative to the decrease in NADPH fluorescence becomes larger, resulting in a dosedependent increase in “inhibition” (Fig. 5). In the case of highly fluorescent compounds, the increase in fluorescence at high concentrations is so large that the overall fluorescence in the assay increases over the 90 min assay timecourse, giving negative percent-activity values (Fig. 6). In primary HTS, negative percent activity corresponds to percent inhibition >100%
HTS Fluorescent Assay for NAD(P)H to NAD(P)
93 Current Protocols in Chemical Biology
Volume 3
A Moderately fluorescent
% activity
100
C 70% activity 0
B Highly fluorescent
Fluorescence
15% activity
15% activity
% activity
100
100% activity
0
0
90 Time (min)
70% activity
Figure 6 (A) A moderately fluorescent compound gives a dose-response curve similar to that of a bona fide enzyme inhibitor. (B) A highly fluorescent compound gives rise to an increase in fluorescence larger than the decrease in NADPH fluorescence, giving apparent negative percent activity at a high concentration of compound. (C) Decrease in NADPH fluorescence over time is shown by a green line. In the presence of a high concentration of the moderately fluorescent compound giving the dose-response curve shown in panel A, increase in compound fluorescence over the course of the assay offsets the decrease in NADPH fluorescence to the extent that the amount of NADPH conversion over 90 min is reduced to 15% of the conversion in the absence of added compound (see purple line labeled 15% activity). (See Fig. 5 for further explanation of this effect.) In the presence of the highly fluorescent compound giving the dose-response curve shown in panel B, increase in compound fluorescence over the course of the assay is greater in magnitude than the decrease in NADPH fluorescence, giving an overall increase in fluorescence over the course of the assay. Calculation of percent activity from a fluorescence increase using Equation 1 gives a negative value, in this case apparent −70% activity at a high compound concentration (see red line labeled −70% activity). Negative percent activity values (and the corresponding percent inhibition values >100 calculated using Eq. 2) are not meaningful in the context of bona fide enzyme inhibition. Therefore, the appearance of a dose-response curve such as the one shown in panel B immediately suggests that the compound “activity” is an artifact of its intrinsic fluorescence.
(see Eq. 2). Examples of highly fluorescent compounds (e.g., coumarins) giving percent inhibition values up to 1500% can be found in PubChem assay IDs 1532 and 1533 (see Internet Resources). Factors to consider in applying this protocol to other enzymes A significant number of compounds in screening libraries are fluorescent under the detection conditions used here (excitation 340 nm, emission 460). Therefore, it is important to follow HTS based on NAD(P)H fluorescence with the counterscreen assay.
Anticipated Results HTS Fluorescent Assay for NAD(P)H to NAD(P)
Enzyme titration and timecourse Figure 7 shows an enzyme timecourse in which the selected dilution of Rml enzymes mixture gives a linear decrease in NADPH
fluorescence over 180 min. The amount of NADPH converted to NADP+ may be calculated from the fluorescence change using the NADPH standard curve (steps 1 to 5). Km determinations Figure 3 shows Km determinations for TDPKDX and NADPH using the Rml enzymes. Note that the sigmoidal curves, as seen in this figure, are fairly unusual (see also Km determinations under Critical Parameters and Troubleshooting). Positive control inhibitor dose-response and IC50 determination Figure 8 shows a dose-response curve for TDP inhibition of RmlC/D. Nonlinear regression analysis (steps 33 to 36) gave the curve fit shown on the figure and an IC50 of 388 μM.
94 Volume 3
Current Protocols in Chemical Biology
NADPH fluorescence
900,000 800,000 700,000 600,000 500,000 Blank Control (Rxn) assay
400,000 300,000 0
30
60
90 120 Time (min)
150
180
Figure 7 Enzyme activity timecourse. Example of dilution of Rml enzymes mixture that gives linear timecourse over 180 min. The protocol was based on that described under steps 6 to 10, Enzyme titration and timecourse. Data points represent the mean of triplicate determinations; error bars depict standard deviation. Control assay: The fluorescence of a mixture of 2.63 × 10−4 μg/μl Rml C, 6.53 × 10−4 μg/μl Rml D, 25 μM NADPH, and 200 μM TDP-KDX in 1× assay buffer was monitored over 180 min at 25◦ C. Blank: The Control mixture without TDP-KDX was used to determine the change in NADPH fluorescence in the absence of enzymatic activity. (The Rml enzymes do not convert NADPH to NADP in the absence of TDP-KDX.)
Change in NADPH fluorescence
240,000 200,000 160,000 120,000 80,000 40,000 0 10
100
1000
10,000
TDP ( M)
Figure 8 Dose-response curve for TDP inhibition of enzymes RmlC/D. Data points represent a single well at each concentration. The protocol is described under steps 25 to 36, Positive control inhibitor dose-response and IC50 determination. TDP was tested at concentrations ranging from 1 μM to 30 mM. From these data, the IC50 for TDP inhibition was determined to be 388 μM.
Quality control (QC) validation plate Figure 4 shows representative QC plate data. TDP at a concentration of 500 μM gave 66% inhibition, essentially identical to the potency predicted based on the IC50 value of 388 μM determined from the dose-response curve shown in Figure 8. Summary of RmlC/D screening results This section provides a brief summary of the outcome of the RmlC/D screening, doseresponse testing, and counterscreening. A to-
tal of 201,368 unique compounds (265,000 including duplicates) from the Molecular Libraries Small Molecule Repository (MLSMR) were screened at a concentration of 5.5 μM as mixtures of four per well in 384-well plates. A total of 2328 mixtures that showed >30% inhibition were selected as hits. (The standard deviation of the mean of the negative controls in each plate was <10% of the mean, so the 30% inhibition threshold for hit selection was at least three standard deviations below the mean of the negative controls.)
HTS Fluorescent Assay for NAD(P)H to NAD(P)
95 Current Protocols in Chemical Biology
Volume 3
Subject to availability, all four individual compounds in each mixture selected as a hit were reordered from the MLSMR and retested one compound per well in duplicate at a concentration of 5.5 μM. Out of 5266 compounds retested individually, 470 gave >30% inhibition. The very low retest rate is to be expected given the likelihood that the activity of each mixture stemmed from only one compound in that mixture. Only 388 out of the 470 compounds active at 5.5 μM were available for dose-response testing; of these, 372 gave an IC50 <50 μM. The counterscreen assay revealed 358 of these compounds to be false positives due to intrinsic compound fluorescence. The remaining 14 were selected as confirmed hits for further study. For more details, see the following Assay IDs in PubChem (see Internet Resources): Primary screening assay (single concentration compound screening): Assay IDs 1532 and 1533. Data analysis and hit selection: Assay IDs 1532 and 1533. Dose-response testing: Assay ID 1695. Curve fitting, IC50 determination, and hit confirmation: Assay ID 1695. Counterscreen assay (elimination of false positives): Assay ID 1696.
Time Considerations NADPH standard curve Total assay time is 5 min. Allow 1 to 2 hr to plan the experiment and set up the assay. Allow 1 to 2 hr after the assay to analyze the data. Enzyme titration and timecourse Total assay time is 90 min. Allow 4 to 5 hr to plan the experiment and set up the assay. Allow 4 to 5 hr after the assay to analyze the data. Km determinations Total assay time is 90 min. Allow 4 to 5 hr to plan the experiment and set up the assay. Allow 4 to 5 hr after the assay to analyze the data. Positive control inhibitor dose-response and IC50 determination Total assay time is 90 min. Allow 2 to 3 hr to plan the experiment and set up the assay. Allow 2 to 3 hr after the assay to analyze the data.
HTS Fluorescent Assay for NAD(P)H to NAD(P)
Quality control (QC) validation plate Allow 2 to 3 hr to make up the positive control compound storage plate. This may be reused for multiple assays. Total assay time is 90 min. Allow 1 to 2 hr before and after for set up and data collection, respectively.
Primary screening assay (single-concentration compound screening) Total assay time is 90 min. Allow 1 to 2 hr before and after for set up and data collection, respectively. Up to 30 plates may be tested batchwise, provided the fluorescence of each plate is read immediately after reagent addition and exactly 90 min later. For a 30-plate screening run, allow 3 to 4 hr before starting for assay set up, and 1 to 2 hr following the assay for clean up and data collection. Data analysis and hit selection Allow 1 day per week during screening. Dose-response testing Allow 1 day to select hits and for serial dilutions. Testing time is similar to Positive control inhibitor dose-response and IC50 determination above. Curve fitting, IC50 determination, and hit confirmation Allow 1 day for data analysis per 100 to 200 compounds. Counterscreen assay (elimination of false positives) Dose-response plates may be reused. Allow 1 day for assay and data analysis per 100 compounds.
Acknowledgements Assay development and validation and high-throughput screening was supported by the NIH Molecular Libraries Screening Center Network (Grant U54H G003915-02).
Literature Cited Batchelor, R.H. and Zhou, M. 2002. A resorufinbased fluorescent assay for quantifying NADH. Anal. Biochem. 305:118-119. Batchelor, R.H. and Zhou, M. 2004. Use of cellular glucose-6-phosphate dehydrogenase for cell quantitation: Applications in cytotoxicity and apoptosis assays. Anal. Biochem. 329:3542. Bergmeyer, H.U. and Bernt, E. 1963. Lactate dehydrogenase. In Methods of Enzymatic Analysis. (H.U. Bergmeyer, ed.) pp. 574-579. Academic Press, London. Fersht, A. 1985. Enzyme Structure and Mechanism, 2nd edition. Freeman, New York. Ma, Y., Stern, R.J., Scherman, M.S., Vissa, V.D., Yan, W., Jones, V.C., Zhang, F., Franzblau, S.G., Lewis, W.H., and McNeil, M.R. 2001. Drug targeting Mycobacterium tuberculosis cell wall synthesis: Genetics of dTDP-rhamnose synthetic enzymes and development of a microtiter plate-based screen for inhibitors of conversion of dTDP-glucose to dTDP-rhamnose. Antimicrob. Agents Chemother. 45:1407-1416.
96 Volume 3
Current Protocols in Chemical Biology
Moran, J.H. and Schnellmann, R.G. 1996. A rapid beta-NADH-linked fluorescence assay for lactate dehydrogenase in cellular death. J. Pharmacol. Toxicol. Methods 36:41-44. Rudnicki S. and Johnston, S. 2009. Overview of liquid handling instrumentation for highthroughput screening applications. Curr. Protoc.Chem. Biol. 1:43-54. Segel, I. 1975. Enzyme Kinetics: Behavior and Analysis of Rapid Equilibrium and Steady-State Enzyme Systems. Wiley, New York. Sivendran, S., Jones, V., Sun, D., Wang, Y., Grzegorzewicz, A.E., Scherman, M.S., Napper, A.D., McCammon, J.A., Lee, R.E., Diamond, S.L., and McNeil, M. 2010. Identification of triazinoindol-benzimidazolones as nanomolar inhibitors of the Mycobacterium tuberculosis enzyme TDP-6-deoxy-d-xylo-4hexopyranosid-4-ulose 3,5-epimerase (RmlC). Bioorg. Med. Chem. 18:896-908.
Internet Resources http://pubchem.ncbi.nlm.nih.gov/assay/assay .cgi?aid=1532&loc=ea ras http://pubchem.ncbi.nlm.nih.gov/assay/assay .cgi?aid=1533&loc=ea ras Results and analysis from screening 265,000 compounds against RmlC/D enzymes.
http://pubchem.ncbi.nlm.nih.gov/assay/ assay.cgi?aid=1695&loc=ea ras Dose response testing following RmlC/D HTS and hit confirmation by curve fitting and IC50 determination. http://pubchem.ncbi.nlm.nih.gov/assay/ assay.cgi?aid=1696&loc=ea ras Dose response testing to eliminate fluorescent compounds as false positives following RmlC/D HTS. http://spotlite.nih.gov/assay/index.php/Enzyme General 2010#How to Measure Km Section of the Assay Guidance Manual that describes how to measure Km . http://spotlite.nih.gov/assay/index.php/Enzyme General 2010#IC50 Determination Section of the Assay Guidance Manual that describes how to determine IC50 values. http://spotlite.nih.gov/assay/index.php/ Section13:Performing MOA Studies Section of the Assay Guidance Manual that describes the relationship of enzyme inhibitor IC50 to substrate concentration and how this relates to the mechanism of inhibition. http://www.graphpad.com/downloads/docs/ Prism5Regression.pdf GraphPad Prism Regression Guide, which provides a highly detailed discussion and many examples of nonlinear regression.
HTS Fluorescent Assay for NAD(P)H to NAD(P)
97 Current Protocols in Chemical Biology
Volume 3
Chemical Synthesis of Hydrocarbon-Stapled Peptides for Protein Interaction Research and Therapeutic Targeting Gregory H. Bird,1 W. Christian Crannell,1 and Loren D. Walensky1 1
Dana-Farber Cancer Institute and Children’s Hospital Boston, Harvard Medical School, Boston, Massachusetts
ABSTRACT The peptide α-helix represents one of nature’s most featured protein shapes and is employed in a diversity of protein architectures, from the cytoskeletal infrastructure to the most intimate contact points between crucial signaling proteins. By installing an allhydrocarbon crosslink into native sequences, the shape and biological activity of natural peptide α-helices can be recapitulated, yielding a chemical toolbox that can be used both to interrogate the protein interactome and to modulate interaction networks for potential therapeutic benefit. Here, current methodology for synthesizing stabilized αhelices (SAH) corresponding to key protein interaction domains is described. A stepwise approach is taken for the production of crosslinking non-natural amino acids, incorporation of the residues into peptide templates, and application of ruthenium-catalyzed ring-closing metathesis to generate hydrocarbon-stapled peptides. Through facile derivatization and functionalization steps, SAHs can be tailored for a broad range of applications in biochemical, structural, proteomic, cellular, and in vivo studies. Curr. Protoc. Chem. C 2011 by John Wiley & Sons, Inc. Biol. 3:99-117 Keywords: α-helix r peptide r hydrocarbon stapling r olefin metathesis r photoreactive r protein interaction r targeting
INTRODUCTION Why staple a peptide? The simplest answer is to restore and stabilize the natural α-helical structure of a peptide that otherwise unfolds when taken out of context from the full-length protein. By installing a non-natural all-hydrocarbon restraint within the peptide at one or more locations, we have consistently observed by circular dichroism that unfolded helical peptides can be zipped back into shape (Walensky et al., 2004, 2006; Bird et al., 2010; Fig. 1A). As a characteristic example, a 21-amino-acid synthetic peptide corresponding to the BCL-2 homology domain 3 (BH3) of the anti-apoptotic protein MCL-1 exhibited 18% α-helicity in aqueous solution by circular dichroism, whereas single i, i+4 stapling at five distinct positions along the peptide sequence yielded a panel of hydrocarbonstapled peptides displaying 62% to 100% α-helical character (Stewart et al., 2010). The crystal structure of one of these stapled BH3 peptides in complex with its protein target confirmed that the chemically stabilized peptide recapitulated the α-helical structure of the native domain (Stewart et al., 2010). The staple-induced structural restoration turns out to have several beneficial side-effects. First, the proteolytic resistance of the peptide is dramatically enhanced (Walensky et al., 2004; Bird et al., 2010; Fig. 1B). Whereas linear amide bonds are efficiently hydrolyzed by proteases, amide bonds engaged in the hydrogen-bonding network of a structured peptide helix are poor enzymatic substrates. In fact, there are two tiers of staple-induced protease resistance, including (1) slowed kinetics of proteolysis through helical induction Current Protocols in Chemical Biology 3: 99-117, September 2011 Published online September 2011 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/9780470559277.ch110042 C 2011 John Wiley & Sons, Inc. Copyright
Synthesis of Stapled Peptides
99 Volume 3
A
B
␣-Helical structure
Protease resistance
T649v Singly stapled T649v
⫺20,000 ⫺30,000
Fraction intact
[θ]
60,000 50,000 40,000 30,000 20,000 10,000 0 ⫺10,000190 200 210 220 230 240 250 260
Enfuvirtide Singly stapled Enfuvirtide Doubly stapled Enfuvirtide 1.0
0.5
0.0 0
200
Wavelength (nm)
D
Cellular uptake
4 hr FITC-BID SAHBA
TRITC-Dextran
800
Binding activity
24 hr 200
FITC-BID SAHBA
MitoTracker
Polarization units
C
400 600 Time (min)
BIM BH3 BIM SAHBA
175 150 125 100 75 50 25 3
4
5 6 Log pM BCL-X L
7
3
4
5 6 Log pM BAX
7
Merge
Merge
Polarization units
150 125 100 75 50 25
Figure 1 Characteristic features of hydrocarbon-stapled peptides. (A) Circular dichroism demonstrates that hydrocarbon stapling can transform an unfolded peptide into a sturdy α-helix (Bird et al., 2010). (B) Protease resistance of hydrocarbon-stapled peptides. In this example, unmodified, singly stapled, and doubly stapled peptides were exposed to chymotrypsin in vitro and the persistence of full-length peptide was monitored over time by LC/MS (Bird et al., 2010). (C) Fluorescently-labeled stapled peptides are taken up by cells via the pinosomal pathway and gradually released into the cytosol for distribution to target binding sites, as shown here for a BCL-2 family α-helix (BID SAHB) that tracks to mitochondria (Walensky et al., 2004). (D) Stapled helices can exhibit enhanced binding affinity for their target protein compared to the corresponding unmodified peptide. In this example, a BCL-2 family α-helix (BIM SAHB) demonstrated improved binding affinity for an anti-apoptotic target (top) and a previously undetected binding interaction with a pro-apoptotic target (bottom) (Walensky et al., 2006).
distal to the staple and (2) complete blockade of proteolysis in the constrained region immediately adjacent to and within the boundaries of the staple itself (Bird et al., 2010).
Synthesis of Stapled Peptides
Second, stapled peptides gain entry into cells through a constitutive, ATP-dependent, vesicular import pathway called pinocytosis (Walensky et al., 2004). Confocal imaging of live cells treated with fluorescently labeled stapled peptides enables tracking of peptides from the pinosomal import stage (e.g., colocalization with dextran) to ultimate
100 Volume 3
Current Protocols in Chemical Biology
release into the cytosol for distribution to the site(s) of target engagement (e.g., mitochondria; Fig. 1C). Third, target binding affinity is enhanced due to prefolding and structural reinforcement, which respectively establish and maintain the proper orientation of interacting residues. We have observed (1) leftward shifts in binding isotherms of stapled peptides compared to the corresponding unmodified templates (Fig. 1D, top) and (2) sequence-specific binding activity of stapled peptide helices when there is no detectable binding for the corresponding unstructured peptides (Fig. 1D, bottom; Walensky et al., 2004, 2006; Bernal et al., 2010; Bird et al., 2010). In addition, we recently observed that an appropriately positioned staple can itself enhance binding activity even further and without compromising selectivity by engaging hydrophobic contacts at the perimeter of an interaction site (Stewart et al., 2010). Thus, hydrocarbon stapling can produce a structurally stabilized α-helical peptide with the capacity to resist proteolysis and gain cellular entry, overcoming key shortcomings of peptides as research tools and prototype therapeutics. What can you do with stapled peptides? Since we began applying hydrocarbon stapling to generate bioactive helices ten years ago, we have experimented with a variety of applications for their use in target binding analyses, structure determination, proteomic discovery, signal transduction research, cellular analyses, imaging, and in vivo bioactivity studies (Walensky et al., 2004, 2006; Bernal et al., 2010; Danial et al., 2008; Gavathiotis et al., 2008, 2010; Bird et al., 2010; Braun et al., 2010; Stewart et al., 2010). We have explored protein targeting at the plasma membrane, within the cytosol and nucleus, and at specific organelles. We believe that the track record of stapled peptides in uncovering new protein interactions and targeting known interactions for therapeutic benefit in preclinical models speaks to their dual capacity to serve as effective research tools and as promising drug prototypes. As with all new technologies, we continue to learn the rules, make improvements, and discover new applications. Below we present our most current methods for synthesizing and derivatizing stapled peptides for protein interaction research and therapeutic targeting.
STRATEGIC PLANNING The first step when preparing to generate stapled peptides is stockpiling the non-natural amino acids used for olefin metathesis-based crosslinking, whether they be (S)-2-{[(9Hfluoren-9-yl)methoxy]carbonylamino}-2-methyl-hept-6-enoic acid (“S5”) used for i, i+4 staples, or the additional (R)-2-{[(9H-fluoren-9-yl)methoxy]carbonylamino}-2-methyldec-9-enoic acid (“R8”) paired with S5 to produce i, i+7 staples. We have tested and validated two methods for the production of these crosslinking amino acids. The first approach is based on the method developed by Williams and colleagues (Williams et al., 1988, 2003; Williams and Im, 1991), applied by Verdine and colleagues (Schafmeister et al., 2000; Walensky et al., 2004), and used extensively by our group, as described (Bird et al., 2008). As no significant changes have been made in this synthetic route since our last report (Bird et al., 2008), this unit describes an alternative, facile approach that we have recently tested and adapted for generating the non-natural amino acids based on the methods of Hruby, Ryzhov, and colleagues (Belokon et al., 1998; Qiu et al., 2000). With stapling amino acids in hand, the next step is to generate designs for your first panel of peptides. The more structural data available for your template peptide and its interactor(s) the better, since the initial goal is to place the hydrocarbon staple(s) at the non-interacting face of the helix to avoid disruption of key helix-target interactions. If there are no structural data available but secondary structure prediction software suggests the peptide template is likely an α-helix, a panel of constructs with differentially localized
Synthesis of Stapled Peptides
101 Current Protocols in Chemical Biology
Volume 3
staples can be generated at the outset to determine optimal staple placement. Once an effective binder is identified, subsequent panels can be geared toward generating mutant controls, alanine and/or staple scans, and any further iterations to optimize binding activity, protease resistance, cellular penetrance, and/or biological activity. Depending on the experimental application, the N terminus, for example, can be capped with acetyl, FITC, biotin, spin label, radiolabel, or other functionalities, so it is best to consider during the peptide design stage which derivatizations will be best suited for your work. Finally, in preparation for peptide synthesis, it is best to identify a research group or core facility with longstanding experience in automated peptide synthesis, derivatization, purification, and quantitation. Importantly, the group must be willing to modify the standard Fmoc conditions used for generic peptide synthesis to accommodate the extended coupling times and other necessary protocol adjustments for optimal stapled peptide synthesis, as delineated below. To date, stapled peptides have been successfully generated in high yield and purity on the following automated synthesizers: ABI 433A (Applied Biosystems), Apex 396 (AAPPTec), Tetras (CreoSalus), and Liberty-12 (CEM). Finally, once the stapled peptides are made and purified, we recommend quantitation by amino acid analysis, which will require identification of a core facility experienced in providing this service. BASIC PROTOCOL 1
ASYMMETRIC SYNTHESIS OF STAPLING AMINO ACIDS VIA BENZYLPROLYLAMINOBENZOPHENONE The following method for generating α,α-disubstituted non-natural amino acids bearing olefin tethers was adapted from Belokon et al. (1998) and Qiu et al. (2000) and employs a benzylprolylaminobenzophenone (BPB)−based chiral auxiliary. The advantages of this synthetic approach include its simplicity, the alkylation efficiency, the high stereochemical purity of the product, and the ability to recycle the chiral auxiliary. The synthetic steps are outlined in Figure 2.
Materials
Synthesis of Stapled Peptides
Nitrogen source Potassium hydroxide pellets (KOH) Isopropanol, HPLC grade D-Proline Benzyl chloride, anhydrous Hydrochloric acid (HCl) Chloroform, HPLC grade (CHCl3 ) Magnesium sulfate powder, anhydrous (MgSO4 ) Acetone, HPLC grade Dichloromethane, reagent grade and anhydrous (DCM) 2 M thionyl chloride (SOCl2 , anhydrous) in DCM o-Aminobenzophenone Saturated sodium carbonate solution Brine Racemic alanine Nickel (II) nitrate hexahydrate (Ni(NO3 )2 ·6H2 O) Methanol, reagent grade (MeOH) Acetic acid (AcOH) Sodium iodide (NaI) 8-Bromo-1-octene Ethyl acetate, reagent grade (EtOAc) Hexanes, reagent grade Dimethylformamide, anhydrous (DMF)
102 Volume 3
Current Protocols in Chemical Biology
Celite Acetonitrile (ACN) Trifluoroacetic acid (TFA) 9-Fluorenylmethoxycarbonyl-N-hydroxysuccinimide (Fmoc-OSu) 500-ml three-neck round-bottom flasks Glass stoppers Addition funnels Medium porosity Buchner funnels Syringes 50-, 250-, 500-, and 1000-ml separatory funnels 250-ml two-neck round-bottom flasks Reflux condensers 500- and 1000-ml Erlenmeyer flasks Rubber septa 250-ml round-bottom flask 43-g reversed-phase (C18) MPLC column Additional reagents and equipment for column chromatography Synthesize (R)-N-benzylproline (BP) 1. In a 500-ml three-neck round-bottom flask fitted with a glass stopper, addition funnel, and nitrogen line, dissolve freshly ground KOH (3.8 equiv) in isopropanol (110 ml). 2. Add D-proline (15 g, 1 equiv), stir the solution at 40◦ C until transparent, and then add benzyl chloride (1.5 equiv) via the addition funnel at 40◦ C. Stir the cloudy solution at 40◦ C for 6 hr. 3. Cool the reaction to room temperature and then quench by gradually adding concentrated HCl (∼20 ml total) until pH 5-6 is achieved. 4. To purify the product, add CHCl3 (45 ml) and allow the solution to sit overnight at room temperature. 5. Vacuum filter the resulting precipitate using a medium porosity Buchner funnel and wash with three 20-ml portions of CHCl3 . 6. Combine the CHCl3 filtrates, dry over MgSO4 , and concentrate in vacuo to afford a wet-appearing orange solid. 7. Treat the solid with ∼100 ml HPLC-grade acetone to dissolve impurities and filter the resulting white solid, (R)-N-benzylproline (15.21 g, 74.1 mmol, 57% isolated yield).
Synthesize (R)-2-[N-(N -benzylprolyl)amino]benzophenone (BPB) 8. To a 500-ml three-neck round-bottom flask equipped with an addition funnel, nitrogen line, and glass stopper, dissolve BP (15.21 g, 1 equiv) in 34 ml anhydrous DCM and stir at −20◦ C for 20 min. 9. Slowly add thionyl chloride (1.25 equiv, 2 M in DCM) via syringe at −20◦ C and stir the solution at −15◦ C for 20 min. 10. Add a solution of o-aminobenzophenone (1 equiv) in DCM (60 ml) dropwise via addition funnel at −30◦ C in a dry ice/acetone bath. Slowly bring the reaction to room temperature by allowing the cooling bath to warm naturally, and stir overnight.. If a precipitate forms, dissolve the solid by quenching the reaction with a saturated sodium carbonate solution (2 equiv in 60 ml H2 O).
Synthesis of Stapled Peptides
103 Current Protocols in Chemical Biology
Volume 3
Step 1
Cl
Ph
O
H N
N
OH
Ph O
O N
OH
2. HClaq
Step 2 Ph
O
1. KOH/i PrOH, 40°C
NH2
O
SOCl2
N
NH
CH2Cl2, ⫺20°C
OH
O
Step 3 Ph
O N
O H H3C
Ala (racemic), Ni(NO3)2 6H2O, KOH NH MeOH, reflux
O Ni N NH
N
O
O
Step 4 Nal
Br
I
acetone, reflux
Step 5
Major product O H3C
O N
O H H3C
Ni N NH O
8-iodo-1-octene, KOH
O Ni N N NH
DMF, 4°C
+ Minor product O
O
O
H3C
N
Ni N NH O
Step 6 O H3C
O O
CH3
HO
Ni N N NH
3 M HCl/MeOH, reflux
NH3Cl
O
Ph
+
O N
NH O
Step 7
O
O CH3
HO NH3Cl
Fmoc-OSu, Na2CO3
CH3
HO NHFmoc
1:1 H2O/acetone, RT
Figure 2 Synthetic steps for generating α,α-disubstituted non-natural amino acids bearing olefin tethers using a BPB-based chiral auxiliary.
11. Transfer the solution to a 500-ml separatory funnel and extract the aqueous solution with three 50-ml portions of DCM. 12. Combine the organic layers and wash with 100 ml brine. 13. Dry the organic layer over MgSO4 and concentrate in vacuo to yield a brown-orange solid (BPB), which is used without further purification. Synthesis of Stapled Peptides
If desired, a short silica column using an EtOAc/hexanes gradient (0% to 20%, Rf = 0.29) can be used to purify the product.
104 Volume 3
Current Protocols in Chemical Biology
Synthesize chiral auxiliary, BPB-Ni(II)-Ala 14. Combine BPB (9.16 g, 1 equiv), racemic alanine (2 equiv), and Ni(NO3 )2 ·6H2 O (2 equiv) under nitrogen in a 250-ml two-neck round-bottom flask equipped with a glass stopper and condenser. 15. Add methanol (85 ml) and stir the solution for 10 min at 60◦ C. 16. Add a solution of KOH (7 equiv) in MeOH (36 ml). Stir the reaction at reflux for 2 hr and then cool to room temperature with stirring. Upon addition, the reaction mixture will turn from green to brick red.
17. Quench the reaction by adding 9.6 ml AcOH (7 equiv), transfer to a 500-ml Erlenmeyer flask, and dilute with 268 ml H2 O. Allow the flask to sit undisturbed overnight to crystallize BPB-Ni(II)-Ala. 18. Vacuum filter the product using a medium porosity Buchner funnel and wash with ∼100 ml H2 O to yield 9.5 g of pure chiral auxiliary (18.5 mmol, 80% yield; Rf = 0.31 in 5% MeOH/DCM).
Synthesize 8-iodo-1-octene by Finkelstein reaction 19. Dissolve NaI (35.7 g, 2 equiv) in acetone (180 ml) in a 250-ml two-neck roundbottom flask equipped with a rubber septum and condenser. 20. Add 8-bromo-1-octene (20 ml, 1 equiv) via syringe (the solution becomes cloudy) and reflux the reaction at 60◦ C with stirring for 2 hr. 21. Gradually cool the reaction to room temperature by removing from heat and then transfer to a 250-ml separatory funnel. Extract the organic mixture with three 50-ml portions of hexanes and then wash with three 20-ml portions of water. 22. Dry the organic layer over MgSO4 and concentrate in vacuo to afford clear liquid 8-iodo-1-octene in near quantitative yield. The pure product can be stored at 4◦ C in the dark until ready for use. Depending on their reactivity, alkenyl bromides may be used directly in the subsequent alkylation step, obviating the need for halide exchange.
Perform alkylation to yield (R)-BPB-Ni(II)-2-methyldec-9-enoate 23. Dissolve BPB-Ni(II)-Ala (9.5 g, 1 equiv) in 95 ml DMF under nitrogen in a 250-ml round-bottom flask submerged in an ice bath. 24. Add freshly ground KOH (10 equiv) and stir for 20 min at 4◦ C. 25. Add 8-iodo-1-octene (1.2 equiv) as a solution in DMF (8.4 ml) via syringe at 4◦ C. Stir the reaction, allowing it to warm to room temperature, and monitor for generation of the reaction product (Rf = 0.41, 5% MeOH/DCM), which migrates just above the starting material upon TLC. 26. Quench after 1.5 hr by adding the reaction mixture to 422 ml chilled 5% AcOH in a 1-liter Erlenmeyer flask. 27. Transfer the solution to a 1-liter separatory funnel, extract the organic layer with three 100-ml portions of DCM, and wash with three 50-ml portions of water. 28. Dry the organic layer over MgSO4 and concentrate in vacuo to afford a deep-red oil. 29. Adsorb the crude material onto celite, dry load a silica column, and elute with a gradient of 0% to 70 % EtOAc/hexanes. Synthesis of Stapled Peptides
105 Current Protocols in Chemical Biology
Volume 3
30. Combine only the exquisitely clean fractions of product to avoid trace contamination by the unwanted diastereomer, which elutes after the desired product, (R)-BPBNi(II)-2-methyldec-9-enoate (77% yield, 8.7 g, 14.0 mmol). (R)-BPB-Ni(II)-2-methyldec-9-enoate can be stored at 4◦ C until ready for use.
Perform decomposition to yield unprotected (R)-2-amino-2-methyldec-9-enoic acid 31. Dissolve (R)-BPB-Ni(II)-2-methyldec-9-enoate (3.5 g, 1 equiv) in 23 ml MeOH and then add the solution to a 1:1 mixture of 3 M HCl/MeOH (38 ml, 10 equiv HCl) in a 250-ml two-neck round-bottom flask equipped with a condenser and rubber septum. 32. Reflux the reaction at 70◦ C for 30 min and then cool to room temperature. 33. Quench the sea green−colored reaction by concentrating under high vacuum. 34. Add 20 ml H2 O at room temperature to promptly precipitate BPB as a white solid. 35. Vacuum filter the solid using a medium porosity Buchner funnel and wash with water to yield 2.15 g BPB (5.61 mmol, 100%), which can be stored at room temperature for reuse. 36. Transfer the remaining aqueous filtrate to a 50-ml separatory funnel and wash with five 10-ml portions of DCM to remove any residual BPB. 37. Concentrate the aqueous layer in vacuo to yield crude (R)-2-amino-2-methyldec-9enoic acid, then dry by lyophilization. 38. To purify the amino acid, dissolve the crude material in a minimum amount of 50:50 (v/v) H2 O/ACN and load onto a 43-g reversed-phase (C18) MPLC column. Elute with a gradient of 5% to 95% ACN/H2 O (+ 0.1% TFA) to yield 1.11 g of pure (R)-2-amino-2-methyldec-9-enoic acid (5.57 mmol, 99% yield). The blue-green nickel complex will elute first in approximately four 25-ml fractions, followed by the desired product, which elutes in four to six 25-ml fractions. The reaction products are monitored by LC/MS due to the lack of a fluorophore for TLC monitoring and functional groups for staining.
Perform Fmoc protection to yield (R)-2-{[(9H-fluoren-9-yl)methoxy]carbonylamino} -2-methyl-dec-9-enoic acid 39. Dissolve (R)-2-amino-2-methyldec-9-enoic acid (2.56 g, 1 equiv) in a 1:1 (v/v) H2 O/acetone (150 ml) in a 250-ml round-bottom flask equipped with a septum and nitrogen line. 40. Add Fmoc-OSu (4.52 g, 1.10 equiv) followed by Na2 CO3 (5.43 g, 4 equiv) and stir overnight at room temperature. 41. Add 150 ml of 2:1 H2 O/hexanes, transfer the mixture to a 500-ml separatory funnel, and pour off the organic layer. 42. Wash the aqueous layer with an additional 50 ml hexanes and then back-extract twice with 30 ml of 2:1 hexanes/EtOAc. 43. Adjust the aqueous layer to pH 2-3 with 6 M HCl, then extract the organic layer with three 30-ml portions of DCM. 44. Combine and dry the DCM layers over MgSO4 , and then concentrate in vacuo to yield a pale yellow oil. 45. To purify, load the crude material in a minimum amount of DCM onto a 43-g MPLC column, and elute with a gradient of 0% to 10% MeOH/CHCl3 to isolate Synthesis of Stapled Peptides
106 Volume 3
Current Protocols in Chemical Biology
(R)-2-{[(9H-fluoren-9-yl)methoxy]carbonylamino}-2-methyl-dec-9-enoic acid (3.06 g, 7.3 mmol, 57% yield; Rf = 0.33 in 5% MeOH/CHCl3 + 0.5% AcOH). The pure product is characterized by LC/MS and NMR, and stored at 4◦ C until ready for use.
SYNTHESIS AND DERIVATIZATION OF HYDROCARBON-STAPLED PEPTIDES
BASIC PROTOCOL 2
Automated solid-phase peptide synthesis using Fmoc chemistry is an efficient and reliable method for generating hydrocarbon-stapled peptides. We employ a standard Fmoc protocol that is adjusted to lengthen and/or repeat deprotection and coupling reactions as needed to optimize incorporation of both the stapling amino acids and the natural residues that immediately follow a non-natural amino acid. With these minor modifications, as outlined below, the yields of singly- and doubly-stapled peptides can readily match those of corresponding unmodified peptides. If the peptide has natural α-helical propensity, olefin metathesis proceeds smoothly at room temperature within 2 hr. Sequences with low α-helical propensity may require extended reaction times at higher temperatures to achieve appreciable metathesis, whereas wholly unstructured templates may fail to react, even after refluxing for days. Facile derivatization of peptides affords a variety of functionalized stapled peptides tailored for diverse experimental applications. The synthesis is outlined in Figure 3.
Materials Fmoc-4-benzoyl-L-phenylalanine (Fmoc-Bpa, Advanced ChemTech, optional) Rink amide AM resin (200-400 mesh, EMD Biosciences) Fmoc-protected amino acids: stapling amino acids (see Basic Protocol 1) and natural amino acids (Advanced ChemTech, EMD Biosciences) N-Methylpyrrolidinone (NMP, Aldrich; anhydrous, 99.5% for reagent solutions; ReagentPlus, 99%, for washing) Deprotection solution: 20% (v/v) piperidine in NMP N,N-Dimethylformamide (DMF, Aldrich; HPLC grade for washing) 2-(6-Chloro-1H-benzotriazole-1-yl)-1,1,3,3-tetramethylaminium hexafluorophosphate (HCTU, Peptides International) Diisopropylethylamine (DIPEA) Trifluoroacetic acid (TFA) Triisopropylsilane (TIS) Diethyl ether (ACS grade) Hexanes (technical grade) Acetonitrile (HPLC grade) Grubbs catalyst, first generation: benzylidene-bis(tricyclohexylphosphine) dichlororuthenium (Sigma-Aldrich/Fluka) 1,2-Dichloroethane (DCE) Fluorescein isothiocyanate isomer I (FITC, Sigma, ≥90%, optional) D-Biotin-OSu (optional) β-Alanine or short PEG linker (e.g., Fmoc-NH-(PEG)n -COOH, n = 1-5; optional) Ethanedithiol Formic acid Methanol (MeOH, reagent grade, 99%) Acetic anhydride (Ac2 O) S-(2,2,5,5-Tetramethyl-2,5-dihydro-1H-pyrrol-3-yl)methyl methanesulfonothioate (MTSL, optional) Dimethyl sulfoxide (DMSO, optional) Synthesis of Stapled Peptides
107 Current Protocols in Chemical Biology
Volume 3
Automated solid-phase peptide synthesizer High-performance liquid chromatography/mass spectrometer (LC/MS) Tabletop centrifuge Lyophilizer 1. Design the stapled peptide(s). When the structure of the helical binding interface is available, place hydrocarbon staples on the non-interacting face of the helix to avoid disruption of target binding. To properly space the non-natural amino acids, place S5 residues at i and i+4 positions for a staple that spans one helical turn, and place R8 at i and S5 at i+7 positions, respectively, for a staple that spans two helical turns. 2. Replace amino acids sparingly. One of the goals of peptide stapling is to maximally preserve the natural bioactive peptide sequence. However, we typically replace methionine, whose thioether can decrease the efficiency of ruthenium catalysis, with
H N
Fmoc
R Fmoc
HCTU/DIPEA in NMP
O
R N H
O
H N
R N H
O
n'
R
R
H N O
1. 20% piperidine in NMP 2. Fmoc
Fmoc
N H
H N O
n
6
OH
N H
O
H N
N H
1. 20% piperidine in NMP O
2. Ac2O, DIPEA, NMP
n 1. 20% piperidine in NMP HCTU/DIPEA in NMP
2. Fmoc
H N
O (S)
R
OH Ac
O
H N
Fmoc
R N H
N H
H N
O
R N H
O
R N H
O
n'
H N
O
H N
H N O
n
6
O
n 1. 20% piperidine in NMP HCTU/DIPEA in NMP
R Fmoc
N H
O
H N
R N H
O
1. Cl
R
2.
Fmoc
OH
N H
Cl
O
n R
2. Fmoc
H N
O (R)
OH
HCTU/DIPEA in NMP
O
R N H
PCy3
O
1. 20% piperidine in NMP
H N
, DCE
2. TFA:TIS:H2O (95:2.5:2.5)
H N
6
Fmoc
PCy3 Ru
H N O
O
R N H
Ac
N H
H N O
n'
O
R N H
H N O
6
O
R N H
NH2 O
n
H N O
n
6
1. 20% piperidine in NMP
HCTU/DIPEA in NMP
Synthesis of Stapled Peptides
2.
R Fmoc
N H
OH O
Figure 3 Synthetic steps for the automated production of hydrocarbon-stapled peptides using Fmoc chemistry.
108 Volume 3
Current Protocols in Chemical Biology
an isosteric amino acid such as norleucine to preserve chain length and aliphatic character. An important advantage of stapled peptides for protein interaction research is the ability to carefully vet specificity of action through mutagenesis that intentionally disrupts bioactivity. Whether by alanine scanning to sequentially substitute each natural residue with Ala or by staple scanning to iteratively sample all staple positions along the helical surfaces, the development of negative control mutants is a valuable feature of employing stapled peptides for biological and therapeutic targeting studies. In addition, we have recently employed Fmoc-Bpa to install a photoreactive benzophenone moiety at a sampling of locations throughout the peptide (including, in particular, at the binding interface and its perimeter), in order to crosslink stapled peptide helices to their protein interactors for target discovery and binding site identification using proteomic methods (Braun et al., 2010).
3. Create the peptide synthesis methods file. Modify the synthesizer’s standard Fmoc settings to define all non-natural amino acids using the available letters B, J, O, U, X, and Z. For example, we choose B for norleucine, U for Bpa, X for S5, and Z for R8. Adjust the method file for incorporation of stapling non-natural amino acids. Because the amine in α,α-disubstituted amino acids is linked to a quaternary carbon center, the Fmoc removal and acylation steps of the standard Fmoc protocol require modification to achieve optimal reaction yields. For example, whereas 2 × 10 min Fmoc deprotections are standard for natural Fmoc amino acids, 4 × 10 min deprotections are used for the stapling amino acids. For acylation, coupling frequency and incubation times are typically 2 × 30 min for standard residues, 2 × 45 min for the stapling amino acids, and 3 × 45 min for the residue following a stapling amino acid.
4. Prepare the peptide synthesis reagents and solutions. a. For a 50-μmol-scale peptide synthesis, weigh out ∼70 mg of Rink amide AM resin (loading level 0.4-0.6 mmol/g resin) per reaction chamber and dissolve the calculated amount of each amino acid in anhydrous NMP. Although more expensive than DMF, we employ NMP because of its greater stability at room temperature, as synthetic runs can last ∼1 week for a typical 24-member library of ∼30 residues per peptide. The amount of each amino acid to weigh out, and the corresponding volume of NMP to dissolve it in, is readily calculated by the automated peptide synthesizer’s software based on user-defined parameters, such as equivalents of amino acid per coupling reaction (e.g., 4 equiv) and number of couplings per residue position (e.g., double-coupling).
b. Prepare the deprotection solution and stock sufficient NMP for the repeated washes. Deprotection and wash solutions can alternatively contain DMF to mitigate costs. Of note, it is critical that all piperidine be removed from the beads during the washing step. Although one approach is to perform at least four DMF washes, we prefer to remove piperidine by swelling and shrinking the resin with two cycles of alternating methanol (or isopropranol) and DMF washes.
c. Prepare a 0.5 M solution of the coupling reagent. We choose HCTU over other aminium-based coupling reagents, such as HBTU or HATU, because HCTU is more active than HBTU and more stable than HATU over a 1-week period. Typically, 4 equiv of Fmoc-amino acid, 3.9 equiv of coupling reagent, and 8 equiv of DIPEA are used, and double-coupling is performed. Synthesis of Stapled Peptides
109 Current Protocols in Chemical Biology
Volume 3
5. Execute the peptide synthesis. A summary of the reaction steps for Fmoc-based, solid-phase, stapled peptide synthesis is presented in Figure 3. It is essential to monitor the consumption of reagents and solutions throughout the synthesis to avoid depletion of ingredients during the run.
6. Gauge the success of peptide synthesis. Once the synthesis is complete, perform a test cleavage to determine if full-length peptide was generated as the predominant reaction product. Touch a glass pipet to the resin and transfer approximately 1/8-in.worth of sample to a 0.5-ml solution of 95:2.5:2.5 (v/v) TFA/TIS/water. Stir for 1.5 hr. After removing the resin by filtration, precipitate the peptide by adding the solution dropwise to 1.5 ml of 1:1 (v/v) diethyl ether/hexanes. Incubate on ice for 10 min and then centrifuge at maximum speed in a tabletop centrifuge for 2 min to pellet the precipitated peptide. Dissolve the sample in 150 μl of 1:1 water/acetonitrile and inject 30 μl for LC/MS analysis. Monitor for the desired mass and analyze truncated products for missing residues, which could reflect difficult couplings (e.g., residues following the stapling amino acid). A common M+44 impurity derives from incomplete hydrolysis of the tryptophan carbamic acid, which ultimately hydrolyzes to Trp upon exposure to aqueous solution for >3 hr in the autosampler vial.
7. Apportion resin for the desired functionalizations. The steps taken to finalize stapled peptide synthesis depend on the selected N-terminal or other functionalization. If no derivatization is desired, the N-terminal Fmoc is removed with 20% piperidine/NMP and the peptide is capped with an acetyl group [4:1:0.1 (v/v) NMP/Ac2 O/DIPEA] prior to metathesis. Having a free N-terminus will otherwise result in a sluggish metathesis reaction. For functionalizations that contain sulfur, such as FITC or biotin, the N-terminal Fmoc of an appended β-Ala is retained and metathesis is performed prior to N-terminal capping to avoid interference with ruthenium catalysis.
8. Perform olefin metathesis to generate the staple(s). Staple the olefin-containing peptide using Grubbs generation 1 catalyst. For a 50-μmol-scale reaction, dissolve 8 mg catalyst in 2 ml of DCE and stir for 1.5-3 hr. Then, remove the reaction solution, wash the resin with DCE, and repeat the reaction procedure once or twice more using freshly prepared catalyst. A freshly prepared solution of catalyst should be purple and the color will gradually turn brown during metathesis. Of note, we have observed that the reaction does not strictly require an inert atmosphere of nitrogen or argon. More importantly, because olefin metathesis is an equilibrium reaction, allowing for the gaseous ethylene byproduct to escape will drive the reaction to completion. It is desirable to use DCE because of its higher vapor pressure compared to DCM, though DCM can be substituted (and replenished during the reaction) if DCE is not available. Because separating the stapled from unstapled peptide by HPLC can be challenging, it is important to confirm by LC/MS that metathesis has gone to completion. If necessary, repeat the stapling reaction for longer intervals (even overnight) and/or with a higher reaction temperature. In general, peptides with a sequence-based predisposition to assume an α-helical configuration undergo facile metathesis; therefore, if the structure of the peptide is unknown and the metathesis reaction is unsuccessful or incomplete, this may be an indication that the peptide is not naturally shaped as an α-helix. Synthesis of Stapled Peptides
9. Acylate with sulfur-containing moieties post-metathesis. Remove the Fmoc group by shaking in 2 ml of 20% piperidine/DMF for 30 min. For FITC derivatization, a preceding non-α-amino acid such as β-alanine or 6-aminohexanoic acid is required to
110 Volume 3
Current Protocols in Chemical Biology
avoid an Edman-type elimination reaction. Then, capping with FITC is accomplished by stirring the resin overnight in a 2-ml solution of 0.1 M fluorescein isothiocyanate (isomer I)/0.2 M DIPEA. For biotinylation, a spacer may also be desired, and we typically use β-alanine or a short PEG linker (e.g., Fmoc-NH-(PEG)n -COOH; n = 1-5). The biotinylation reaction employs a 2 ml solution of 0.1 M biotin-OSu/0.4 M DIPEA, stirring overnight. At this stage, a test cleavage is again performed to ensure that the desired reaction has gone to completion. 10. Perform a full-scale cleavage reaction to liberate the unprotected peptide. Vigorously stir the resin in 2 ml of 95:2.5:2.5 (v/v/v) TFA/TIS/water for 2 hr. If the peptide sequence contains cysteine, include ethanedithiol (2%) in the cleavage mixture to prevent oxidation. Filter off the resin and apply the cleavage solution dropwise into 35 ml of 1:1 (v/v) ether/hexanes, mix by inversion, release the pressure, and allow to incubate on dry ice for 20 min. Isolate the precipitate by centrifugation at 1500 × g for 20 min at 4◦ C, remove the reaction solution, and dry the peptide pellet at room temperature overnight. 11. Purify the desired stapled peptide product by LC/MS. Dissolve the peptide precipitate in 1.5 ml of 1:1 (v/v) water/acetonitrile, 3% formic acid, and centrifuge to remove insoluble material. Purify the desired product by HPLC (typically using a C18 column) or, if available, by LC/MS with mass-triggered fraction collection. Reinject the fractions containing the desired full-length peptide, pool all of the pure fractions, and lyophilize, making sure that the acetonitrile concentration is <40% so that the solution remains frozen. Resuspend the pure peptide powder in a known volume of 30% acetonitrile/water, take 1:10 and 1:20 dilutions for amino acid analysis, and then aliquot and lyophilize the dried powder for storage at −20◦ C. Use the amino acid analysis results to calculate the amount of peptide in each aliquot. The shelf life of lyophilized stapled peptides stored at −20◦ C is typically several years, but may vary based on sequence composition.
12. Perform off-resin derivatizations. For certain applications, off-resin derivatizations of the stapled peptide may be desirable. For example, we have generated paramagnetically labeled stapled peptides for paramagnetic relaxation enhancement (PRE) NMR by derivatization with MTSL. Dissolve MTSL in 10 mM DMSO and dissolve the purified stapled peptide (containing an installed cysteine) in 1 mM anhydrous DMF. Mix the two solutions together (1:1), monitor the reaction for completion by LC/MS, and purify the desired product as described in step 9.
COMMENTARY Background Information Current drug therapy is based on two major classes of pharmaceuticals that recognize protein targets in and on the cell: small molecules penetrate intact cells to bind focal, greasy cavities within target proteins/enzymes, and antibodies specialize in binding to large and complex cell surface protein topographies. Along with these major strengths, each class has its accompanying weakness: (1) small molecules are often too small to blanket the large, flat, and complex intracellular protein interaction landscape required for disrupting pathologic protein engagement, and (2) antibodies, ideally suited for binding to complex and extended protein surfaces, cannot access intracellular targets. To fill the drug gap, a class
of “compromise” compounds are envisioned that are sufficiently small to maintain cell penetrance but large enough to achieve a greater breadth of binding terrain to modulate intracellular protein interactions for therapeutic benefit. In the last decade, there has been an explosion of renewed interest in peptides (among other natural motifs such as RNA) to populate this middle ground of drug development. Because protein interactions are typically mediated by protein subdomains with defined structure, chemists, chemical biologists, and drug developers have explored a battery of new approaches to emulate nature’s solution to protein targeting and modulation. Important goals of these efforts include recapitulating and stabilizing native bioactive structure, overcoming
Synthesis of Stapled Peptides
111 Current Protocols in Chemical Biology
Volume 3
vulnerability to proteolysis, and achieving efficient cellular uptake. Within the realm of α-helical peptide mimicry, a variety of exciting new strategies have emerged, including peptide stabilization chemistries and alternative non-natural scaffolds for presenting key amino acid moieties (Henchey et al., 2008). An important byproduct of these approaches is a wealth of new reagents for fundamental biological investigation, which in turn can lead to the discovery of novel targets and protein interactions that can serve as the substrates for next-generation therapeutics. Here, we describe the production of hydrocarbon-stapled peptides and their derivatives based on the incorporation of olefinic non-natural amino acids that, when crosslinked by ruthenium-catalyzed olefin metathesis, yield structurally stabilized, protease-resistant, and cell-permeable αhelices for protein interaction research and therapeutic targeting (Fig. 4). By deploying our stapled peptide approach, we have developed a new therapeutic strategy for targeting intracellular protein interactions (Walensky et al., 2004), uncovered an unanticipated function for a death protein in metabolism (Danial et al., 2008), structurally defined the elusive activation site on an essential executioner protein of the cell death pathway (Gavathiotis et al., 2008), identified a natural α-helical peptide that can function as an exclusive inhibitor of a formidable anti-apoptotic protein linked to cancer (Stewart et al., 2010), remedied the proteolytic instability of a lengthy peptide therapeutic (Bird et al., 2010), defined the key conformational changes that transform an inactive death protein into a toxic mitochondrial oligomer (Gavathiotis et al., 2010), and recapitulated the essential features of a key transcription factor motif to restore the tumor suppressor pathway in cancer cells (Bernal et al., 2010). In describing our latest synthetic approaches, we hope to further expand accessibility to stapled peptides and broaden their utility in advancing protein interaction research and novel therapeutic strategies.
Critical Parameters
Synthesis of Stapled Peptides
Chemical synthesis of the stapling amino acids The most challenging synthetic steps include (1) isolating the correct diastereomer and (2) isolating the free amino acid following cleavage from the chiral auxiliary. To purify
the correct diastereomer, be certain to first resolve the product by TLC, co-spotting against BPB-Ni(II)-Ala using a 5% MeOH/DCM solvent system. For example, the desired R8 product (Rf = 0.41) travels just ahead of the starting material, whereas the undesired diastereomer migrates just below the starting material. Determining the TLC migration profile of the desired product will ensure pooling of the correct, clean fractions following silica gel purification. Following acid-catalyzed cleavage of (R)BPB-Ni(II)-2-methyldec-9-enoate, the free amino acid must be isolated from the aqueous solvent by vigorously drying the material using strong vacuum techniques including lyophilization. Once dry, the amino acid can be purified using a C18 reversed-phase silica gel column. Because the product has no UV activity, analyzing the fractions by LC/MS is recommended to delineate the pure amino acid−containing fractions. Typically, we find that the sea-green fractions containing free nickel elute quickly, with (R)-2-amino2-methyldec-9-enoic acid eluting in the subsequent four to six 25-ml fractions. Once the fractions containing pure amino acid are pooled, the material is again concentrated in vacuo and thoroughly dried by lyophilization. Stapled peptide synthesis A variety of automated peptide synthesizers and Fmoc protocols have been applied to successfully generate stapled peptides. If you are performing peptide synthesis for the first time, optimize the equipment and method using the unmodified template peptide first to ensure that the standard protocol is generating the desired product in high yield and purity before advancing to stapled peptide synthesis. If an automated synthesizer is not available, the entire synthesis can be performed manually. Although laborious, we routinely made stapled peptides in this fashion before adapting the procedure for automation. Once a first panel of stapled peptides is designed and synthesized, a series of critical characterization studies must be performed prior to application of the reagents to biological investigation: 1. What is the solubility profile of the stapled peptide? It is essential to determine the optimal approach to solubilizing your material. Some stapled peptides are highly soluble in water alone and others need to be dissolved in 100% DMSO prior to dilution into aqueous buffers. The HPLC elution profile is an excellent guide, since late-eluting peptides are more
112 Volume 3
Current Protocols in Chemical Biology
Figure 4 Production and derivatization of hydrocarbon-stapled peptide α-helices for a diversity of experimental and therapeutic applications.
hydrophobic and potentially more challenging to solubilize. We suggest that you experiment with dissolving your peptide in a variety of aqueous buffers, varying the pH and salt concentration. Whether dissolving the peptide powder in water or diluting it from a DMSO stock into aqueous buffers, always be sure to verify that the peptide is actually in solution by performing a tabletop centrifugation at maximum speed and then checking for the presence
of a pellet, which indicates incomplete solubility. If inadequately dissolved, rigorous evaluation of stapled peptide activity will be compromised regardless of the assay employed. Iterative dilution of the DMSO stock into increasingly larger volumes of aqueous buffer until the goal concentration is reached can be an effective strategy for achieving solubilization. In the extreme circumstance that you are unable to identify conditions to solubilize your
Synthesis of Stapled Peptides
113 Current Protocols in Chemical Biology
Volume 3
Synthesis of Stapled Peptides
construct, redesign your peptide to decrease overall hydrophobicity. For example, eliminating one or more non-essential hydrophobic residues at the N or C terminus, or extending the peptide to incorporate native hydrophilic or charged residues at the N or C terminus, can often yield a more soluble stapled peptide. 2. What is the structure of the stapled peptide in solution? For rapid assessment of αhelical structural stabilization, perform circular dichroism spectroscopy (Bird et al., 2008) to compare the unmodified peptide to the stapled construct, ensuring that both peptides are completely solubilized. Varying the location of the staple may yield particular constructs with optimal α-helicity. 3. What is the behavior of the stapled peptide in solution? Like many chemical compounds, peptides can aggregate, depending upon their composition and concentration. For most applications, stapled peptides are employed in the nanomolar and low micromolar range, concentrations at which selfassociation is rarely observed. Nevertheless, the self-association propensity of the peptide can be rapidly assessed by evaluating samples at various concentrations using native gel electrophoresis and/or gel filtration chromatography. If aggregation is observed at a particular concentration, employ the stapled peptide below this concentration in biological studies, experiment with alternate solubilization buffers, or redesign the peptide to remedy its propensity to self-associate (e.g., decrease hydrophobicity, as described above). 4. Is the stapled peptide cell permeable? Cellular uptake of stapled peptides occurs in a time-, temperature-, and ATP-dependent manner, consistent with a pinocytotic mechanism (Walensky et al., 2004). In the event that a stapled peptide binds to serum proteins such as albumin, initiate the analysis of cellular uptake using a serum-free medium (e.g., Opti-MEM) or by treating cells in the absence of serum for 1-4 hr followed by serum replacement. We evaluate cellular uptake of FITC-stapled peptides by live confocal microscopy, FACS analysis of treated cells, and fluorescence scan of electrophoresed lysates from treated cells. Prior to analysis, cells are washed to remove stapled peptide−containing medium. For FACS and cell lysate evaluation, the cells are further treated with trypsin to digest surface proteins, thereby eliminating any non-specifically bound peptide. Based on our evaluation of many series of stapled peptides, we have observed that their propensity to be
taken up by cells derives from a combination of factors, including charge, hydrophobicity, and α-helical structure, with negatively charged and less structured constructs typically requiring modification to achieve cell penetrance. Successful interventions include (1) substituting Gln for Glu and/or Asn for Asp, or appending native charged residues at the N or C terminus to adjust the overall charge to 0 to +2, and (2) producing constructs with greater α-helical content through differential staple placement. 5. Does the stapled peptide exhibit ontarget activity? A key benefit of working with peptides in biological systems is the ability to track binding activity and selectivity by using negative control point mutants or scrambled peptide constructs. In addition, protein targeting in cells or cellular lysates can be carefully examined with stapled peptide pull-down assays that employ FITC-tagged, biotinylated, and/or photoreactive constructs, followed by protein detection by western blotting and/or proteomic analyses (Walensky et al., 2006; Bernal et al., 2010; Braun et al., 2010). It is important to be aware that certain peptides can disrupt membranes as a result of their amino acid composition (e.g., cationic antimicrobial peptides; Bechinger, 1997). Therefore, if you are planning to treat cells or purified organelles with your stapled peptide, be sure to perform a maximally tolerated dose titration to screen for constructs that perturb membranes based on composition or dose range. Simple studies such as monitoring cells by light microscopy (e.g., trypan blue exclusion) immediately after treatment can flag disruptive peptides. Stapled peptides should be used at tolerated doses only, and, if necessary, redesigned to eliminate unwanted biophysical properties so that ontarget, sequence-dependent biological activity is achieved. Anticipate that generating the optimal stapled peptide for a particular scientific application — whether for in vitro binding studies, cellular localization analyses, structure determination, cellular signal transduction studies, or in vivo activity analyses — is likely to require several rounds of synthetic iteration to achieve optimal solubility, structural stability, cell permeability, and biological activity. A commitment to taking a rigorous, stepwise, and iterative approach to the production, optimization, and application of stapled peptides is the best formula for success in developing these reagents for protein interaction research and therapeutic targeting.
114 Volume 3
Current Protocols in Chemical Biology
Troubleshooting Chemical synthesis of the stapling amino acids The most frequent challenges associated with non-natural amino acid synthesis, and suggested solutions, are the following: If (R)-N-benzylproline cannot be isolated from the reaction mixture, be sure that the reaction was not acidified below pH 5 upon quenching (see Basic Protocol 1, step 3). If BPB-Ni(II)-Ala does not crystallize (step 17), scour the Erlenmeyer flask with a Pasteur pipet. If a clear liquid product is not obtained from the Finkelstein reaction, the alkenyl iodide likely evaporated due to its low vapor pressure. Use gentle vacuum to evaporate the organic solvent only (step 22). If the (R)-BPB-Ni(II)-2-methyldec-9enoate reaction product cannot be resolved, use a larger silica column and broader EtOAc gradient (step 29). If the (R)-2-amino-2-methyldec-9-enoic acid reaction product cannot be adequately dried by vacuum pump (step 37), flash freeze the product in liquid nitrogen and lyophilize. If Fmoc protection fails, make certain that the starting material, (R)-2-amino-2methyldec-9-enoic acid, is free of nickel and other impurities. If necessary, repurify the product by column chromatography (step 38). If you do not recover the reaction product, (R)-2-{[(9H-fluoren-9-yl)methoxy ]carbonylamino}-2-methyl-dec-9-enoic acid, from the DCM layer (steps 44-45), subject each wash and extraction to thin layer chromatography (TLC), as the Fmoc-protected amino acid can enter the aqueous layer if the correct pH (2-3) is not achieved or maintained. Stapled peptide synthesis The most frequent complication of peptide synthesis is failure to generate a fulllength construct due to difficult amino acid couplings. Common examples include coupling (1) β-branched amino acids (e.g., Thr, Ile, and Val), (2) Arg residues, (3) the nonnatural amino acids, and (4) residues following the non-natural amino acids. To overcome difficult couplings, extend the deprotection and coupling times and perform multiple rounds of coupling using fresh reagent, as described above. Anticipating difficult sequences at the outset and adjusting the method accordingly will improve the success rate and yield. Additional complications can include cross-reactions and progressive inaccessibility
of the N terminus due to on-resin aggregation. The dipeptide motif Asp-Gly is the most likely amino acid pair to undergo a reaction known as aspartimide formation, which occurs when the -NH- of Asp, upon repeated exposure to piperidine, attacks its ester-protected side chain, displacing t-BuOH to form a five-membered ring. Subsequent attack and ring opening by water or piperidine results in production of a peptide containing racemized Asp or a piperamide, respectively. This side reaction can be completely suppressed by employing the commercially available side-chain-protected dipeptide pair Fmoc-Asp(OtBu)-(Dmb)Gly-OH (EMD Biosciences). Exposure of the reactive N terminus can be hindered by aggregation of the growing peptide chains as β-sheets on the polymeric bead. Interventions that prevent aggregation include incorporation of the stapling amino acids themselves, which promote αhelical conformation, and the use of pseudoproline Ser and Thr dipeptides (EMD Biosciences), which can be substituted at X-Ser and X-Thr positions, producing a kink that disrupts β-sheet formation.
Anticipated Results The overall synthetic yield for generating the non-natural amino acids using the above method is ∼20%. For optimized stapled peptide synthesis, one can expect to achieve the purity and yield of the corresponding unmodified peptide. A purity of 90% for the postcleavage crude material is common and can be improved to >95% after HPLC, with overall yields of 30% routinely obtained. Absent any unanticipated coupling challenges, side reactions, or on-bead aggregation, the majority of stapled peptides can be generated successfully on the first attempt.
Time Considerations An experienced chemist can anticipate completing the synthesis of the non-natural amino acid(s) used for all-hydrocarbon peptide stapling in 3-4 weeks. For stapled peptide synthesis, a time frame of 2-3 weeks is standard, but will depend on the type of synthesizer employed and variables such as the length and complexity of the peptide (i.e., need for multiple coupling reactions per residue) and the planned derivatizations. An automated XYZ probe-type synthesizer (e.g., Apex 396, AAPPTec) excels at generating many peptides at once, with only marginal increases in time for expanding
Synthesis of Stapled Peptides
115 Current Protocols in Chemical Biology
Volume 3
Synthesis of Stapled Peptides
the size of the library. The duration of a synthetic run will vary depending on whether you choose to double-couple with fewer equivalents of amino acid or single-couple with larger amino acid quantities. We typically doublecouple at 4:1 equivalents of amino acid to resin for 40 min each, followed by two NMP washes, 2-15 min for Fmoc deprotection using 20% piperidine/NMP, and repeat methanol and NMP washes. Accounting for all robotic arm movements, reagent delivery, and probe washings, one complete cycle of amino acid incorporation for 24 peptides takes ∼3-5 hr, depending on the variety of amino acids at each position. For 24 peptides of 22-30 residues in length, an automated synthesis run is completed in 1 week, which nicely parallels the window of stability for key reagents, such as HCTU. By comparison, a microwave-based peptide synthesizer (Liberty, CEM) that generates one peptide at a time requires only ∼12 hr to complete the synthesis of a 22-mer peptide. N-terminal derivatization and olefin metathesis can take 1-3 days to complete. N-terminal acetylation with 5 ml of 4:1:0.1 NMP/Ac2 O/DIPEA is accomplished in ∼30 min, whereas Fmoc-β-Ala or -PEG can be coupled using HCTU/DIPEA in ∼1 hr. Olefin metathesis using Grubbs I catalyst is performed in 3× 2 hr, after which a small-scale cleavage is recommended to confirm that the reaction has gone to completion. Depending on the number of peptides being processed, small-scale test cleavage and LC/MS analysis require 3-15 hr. With completion of metathesis confirmed, removal of the N-terminal Fmoc from β-Ala or PEG takes 30 min, followed by coupling of FITC (14 hr) or biotin (4 hr). A test cleavage and LC/MS analysis is also recommended at this stage to confirm successful and complete derivatization. To finish the production process, largescale cleavage takes 2 hr, followed by drying the precipitated peptide in a hood overnight. HPLC or LC/MS purification can require 2 hr per peptide and analysis of the fractions for the desired product can take hours to days, depending on the number of peptides being generated and the number of fractions per peptide. Lyophilization of the purified and pooled peptide fractions can take 1-3 days, depending on the number and volume of the fractions. Finally, amino acid analysis of the peptide will take at least 2 days in order to conduct the acid hydrolysis, derivatization, and analysis steps.
Acknowledgments The authors thank E. Smith for figure design and editorial assistance, and Walensky laboratory members past and present, including F. Bernal, C. Braun, E. Gavathiotis, M. Stewart, S. Katz, and J. LaBelle, for their scientific contributions to the advancement of stapled peptide research. We are tremendously grateful for the federal, foundation, and professional society support of our protein interaction and therapeutic targeting research using stapled peptides, including NIH grants 5R01CA50239, 1R01OD005851, 1R01AI084102, and 5P01CA92625, a Leukemia and Lymphoma Society Specialized Center of Research Award, a Stand Up To Cancer Innovative Research Award, and a grant from the Wolpoff Family Foundation.
Literature Cited Bechinger, B. 1997. Structure and functions of channel-forming peptides: Magainins, cecropins, melittin and alamethicin. J. Membr. Biol. 156:197-211. Belokon, Y., Tararov, V., Maleev, V., Savel’eva, T., and Ryzhov, M. 1998. Improved procedures for the synthesis of (S)-2-[N-(N benzylprolyl)amino]benzophenone (BPB) and Ni(II) complexes of Schiff’s bases derived from BPB and amino acids. Tetrahedron-Asymmetry 9:4249-4252. Bernal, F., Wade, M., Godes, M., Davis, T.N., Whitehead, D.G., Kung, A.L., Wahl, G.M., and Walensky, L.D. 2010. A stapled p53 helix overcomes HDMX-mediated suppression of p53. Cancer Cell 18:411-422. Bird, G.H., Bernal, F., Pitter, K., and Walensky, L.D. 2008. Synthesis and biophysical characterization of stabilized alpha-helices of BCL-2 domains. Methods Enzymol. 446:369-386. Bird, G.H., Madani, N., Perry, A.F., Princiotto, A.M., Supko, J.G., He, X., Gavathiotis, E., Sodroski, J.G., and Walensky, L.D. 2010. Hydrocarbon double-stapling remedies the proteolytic instability of a lengthy peptide therapeutic. Proc. Natl. Acad. Sci. U.S.A. 107:14093-14098. Braun, C.R., Mintseris, J., Gavathiotis, E., Bird, G.H., Gygi, S.P., and Walensky, L.D. 2010. Photoreactive stapled BH3 peptides to dissect the BCL-2 family interactome. Chem. Biol. 17:1325-1333. Danial, N.N., Walensky, L.D., Zhang, C.Y., Choi, C.S., Fisher, J.K., Molina, A.J., Datta, S.R., Pitter, K.L., Bird, G.H., Wikstrom, J.D., Deeney, J.T., Robertson, K., Morash, J., Kulkarni, A., Neschen, S., Kim, S., Greenberg, M.E., Corkey, B.E., Shirihai, O.S., Shulman, G.I., Lowell, B.B., and Korsmeyer, S.J. 2008. Dual role of proapoptotic BAD in insulin secretion and beta cell survival. Nat. Med. 14:144-153.
116 Volume 3
Current Protocols in Chemical Biology
Gavathiotis, E., Suzuki, M., Davis, M.L., Pitter, K., Bird, G.H., Katz, S.G., Tu, H.C., Kim, H., Cheng, E.H., Tjandra, N., and Walensky, L.D. 2008. BAX activation is initiated at a novel interaction site. Nature 455:1076-1081.
Walensky, L.D., Kung, A.L., Escher, I., Malia, T.J., Barbuto, S., Wright, R.D., Wagner, G., Verdine, G.L., and Korsmeyer, S.J. 2004. Activation of apoptosis in vivo by a hydrocarbon-stapled BH3 helix. Science 305:1466-1470.
Gavathiotis, E., Reyna, D.E., Davis, M.L., Bird, G.H., and Walensky, L.D. 2010. BH3-triggered structural reorganization drives the activation of proapoptotic BAX. Mol. Cell 40:481-492.
Walensky, L.D., Pitter, K., Morash, J., Oh, K.J., Barbuto, S., Fisher, J., Smith, E., Verdine, G.L., and Korsmeyer, S.J. 2006. A stapled BID BH3 helix directly binds and activates BAX. Mol. Cell 24:199-210.
Henchey, L.K., Jochim, A.L., and Arora, P.S. 2008. Contemporary strategies for the stabilization of peptides in the alpha-helical conformation. Curr. Opin. Chem. Biol. 12:692-697. Qiu, W., Soloshonok, V., Cai, C., Tang, X., and Hruby, V. 2000. Convenient, large-scale asymmetric synthesis of enantiomerically pure transcinnamylglycine and -α-alanine. Tetrahedron 56:2577-2582. Schafmeister, C., Po, J., and Verdine, G. 2000. An all-hydrocarbon cross-linking system for enhancing the helicity and metabolic stability of peptides. J. Am. Chem. Soc. 122:5891-5892. Stewart, M.L., Fire, E., Keating, A.E., and Walensky, L.D. 2010. The MCL-1 BH3 helix is an exclusive MCL-1 inhibitor and apoptosis sensitizer. Nat. Chem. Biol. 6:595-601.
Williams, R.M. and Im, M.N. 1991. Asymmetric synthesis of monosubstituted and alpha, alphadisubstituted amino acids via diastereoselective glycine enolate alkylations. J. Am. Chem. Soc. 113:9276-9286. Williams, R.M., Sinclair, P.J., Zhai, D., and Chen, D. 1988. Practical asymmetric syntheses of alpha-amino acids through carbon-carbon bond constructions on electrophilic glycine templates. J. Am. Chem. Soc. 110:1547-1557. Williams, R.M., Sinclair, P.J., DeMong, D.E., Chen, D., and Zhai, D. 2003. Asymmetric synthesis of N-tert-butoxycarbonyl alpha-amino acids: Synthesis of (5S, 6R)-4-tert-butoxycarbonyl-5,6diphenylmorpholin-2-one. Org. Synth. 80:1830.
Synthesis of Stapled Peptides
117 Current Protocols in Chemical Biology
Volume 3
Triple-Addition Assay Protocols for Detecting and Characterizing Modulators of Seven-Transmembrane Receptors C. David Weaver1 1
Vanderbilt University School of Medicine, Nashville, Tennessee
ABSTRACT The detection and characterization of seven-transmembrane-receptor modulators (orthosteric binding site agonists, antagonists, and more recently allosteric modulators) is an area of intense interest for both drug discovery and basic research. Traditionally, assays used to detect and characterize these different modes of modulation have been executed as separate, discrete protocols focused on a particular mode of action (e.g., agonism). In recent years, investigators have begun to combine aspects of these separate protocols to produce methods that detect multiple modes of modulation simultaneously. The power of such approaches is revealed not only in conservation of time and resources, but more importantly in a superior ability to discover and characterize novel modulators of the targets of interest. The protocols in this article describe a general procedure for developing, validating, and utilizing triple-addition assays to enable the simultaneous detection and characterization of multiple modes of seven-transmembrane-receptor modulation. Curr. C 2011 by John Wiley & Sons, Inc. Protoc. Chem. Biol. 3:119-140 Keywords: seven-transmembrane receptor r G-protein-coupled receptor r agonist r potentiator r antagonist r allosteric r fluorescence r high-throughput screening
INTRODUCTION In recent years there has been a dramatic increase in interest for allosteric potentiators of seven-transmembrane receptors (Christopoulos and Kenakin, 2002; Marino and Conn, 2006; Hoare, 2007; Leach et al., 2007; May et al., 2007; Rees et al., 2008; Conn et al., 2009; Wang et al., 2009; De Amici et al., 2010; Cleva and Olive, 2011; Urwyler, 2011). Previously, most research efforts had focused on the detection and characterization of agonists and antagonists that bind to the receptor’s orthosteric binding site. This shift in emphasis has driven investigators to modify the way they approach both screening for and characterization of seven-transmembrane-receptor modulators. One way that investigators have approached this challenge is to retool classic seven-transmembranereceptor screening protocols like intracellular calcium flux by including additional steps. One such approach is a triple-addition assay approach. The triple-addition approach typically includes the addition of test compound followed by addition of a known agonist at EC20 and then EC80 concentrations (i.e., the concentrations of agonist that produce 20% and 80% of the agonist’s maximal effect in the assay; Jacoby et al., 2006; Niswender et al., 2008; Rodriguez et al., 2010). The present article describes general procedures for the development of fluorescencebased, kinetic triple-addition calcium flux assays that can be used to detect multiple modes of modulation and complex pharmacology of seven-transmembrane receptors. The basic principle of a triple-addition protocol is to enable the near-simultaneous detection of multiple modes of receptor modulation. As shown in Figure 1, the sequential addition of test compound and two concentrations of agonist produce a rich set of data
Current Protocols in Chemical Biology 3: 119-140, September 2011 Published online September 2011 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/9780470559277.ch110060 C 2011 John Wiley & Sons, Inc. Copyright
Triple-Addition Protocols for 7-TM Receptors
119 Volume 3
A
B
C
%Emax
80
0 0
240
0
Time (sec)
240 Time (sec)
0
240 Time (sec)
Figure 1 Data from a typical triple-addition assay depicting some of the types of signal modulation that can be observed. All signal amplitudes are scaled relative to a maximally effective concentration of an agonist. Solid traces are from wells containing (A) agonist, (B) potentiator, and (C) antagonist. Dashed lines are from vehicle control wells. Arrows show points at which test compound or vehicle (0.2% DMSO in assay buffer) was added (first arrow), an EC20 concentration of agonist was added (second arrow), or an EC80 concentration of an agonist was added (third arrow).
enabling investigators to observe agonism, potentiation, antagonism, and some forms of desensitization all in the same assay. This approach saves time and resources compared to running agonist, potentiator, and antagonist assays separately as part of a high-throughput screen designed to discover seven-transmembrane-receptor modulators. Triple-addition assays also provide a mechanism to monitor how the structure of small molecules can affect transitions from agonism to potentiation to inhibition when the assays are used to generate concentration-response relationships as part of hits-to-leads and lead optimization efforts.
STRATEGIC PLANNING Triple-addition assays can be useful for a wide variety of targets in diverse cellular backgrounds with a multitude of different assay technologies. For the purposes of this article, the focus will be on assays of seven-transmembrane receptors coupled to modulation of intracellular calcium concentration detected using fluorescent calcium indicator dyes on a kinetic plate reader. In preparation for developing and using triple-addition assays, one must obtain a cell line expressing the target of interest or a clone of the target of interest in a suitable eukaryotic expression vector. Clones of most seven-transmembrane receptors are now available through a variety of vendors such as Origene and the Missouri S&T cDNA Resource Center. Alternatively, seven-transmembrane receptors may be synthesized by a variety of vendors such as Origene, GeneScript, and GeneArt. If the target receptor is known to couple to robust changes in intracellular calcium in the proposed cellular context, a major hurdle has already been overcome. However, if the target of interest is known not to couple to changes in intracellular calcium in the proposed cellular context, work may be required to establish this coupling. Typically, this is achieved through the use of chimeric G-proteins (Conklin et al., 1993; Coward et al., 1999) or promiscuous G-proteins (Offermanns and Simon, 1995; Stables et al., 1997), but may also be accomplished by co-expression of certain members of the TrpC family of ion channels (Miller et al., 2011). Extensive testing may be required to achieve the best coupling of the target of interest to changes in intracellular calcium.
Triple-Addition Protocols for 7-TM Receptors
Another consideration is whether to use a stably expressing cell line, transient transfection, or viral transduction. Much has been written recently about the successful use of all of these modes for small-scale and large-scale high-throughput screening (Digan et al., 2005; Kunapuli et al., 2005; Chen et al., 2007; Kost et al., 2007). The protocols described
120 Volume 3
Current Protocols in Chemical Biology
herein presume the availability of cells that express the target of interest under conditions where it couples to robust changes in intracellular calcium. Although non-adherent suspension cultures may be used for triple-addition assays, the Basic Protocols in this article are written with adherent cells in mind. The availability of necessary quantities of agonist and control compounds (potentiators, antagonists, etc.) must also be considered. If the desired agonist is not commercially available, there is a growing number of vendors that provide custom synthesis (Vanderbilt Chemical Synthesis Core, J-Star Research, GVK Bioscience, WuXi AppTec, ChemPartner, Aurora Fine Chemicals, Chempacific, Biofine International). When the choice of more than one agonist exists, availability and cost should certainly be considered, but so should any potential bias of certain agonists on downstream signaling (Leach et al., 2007; Drake et al., 2008; Vaidehi and Kenakin, 2010; Evans et al., 2011). Such biased agonism may affect the outcome of a high-throughput screen or the interpretation of data generated in support of establishing structure-activity relationships. The protocols in this article describe general methods for establishing proof-of-concept for a triple-addition assay in 384-well plates (Basic Protocol 1), validating that assay for use in high-throughput screening to discover modulators of a seven-transmembrane receptor (Basic Protocol 2), executing a high-throughput screen (Basic Protocol 3), and using the assay to generate concentration-response relationships (Basic Protocol 4). A specific example is given using mGluR5 expressed in HEK-293 cells. Positive and negative modulators of mGluR5 are highly interesting as targets for a number of therapeutic indications (Gasparini et al., 2008; Lindsley and Emmitte, 2009; Bird and Lawrence, 2009; Cleva and Olive, 2011).
MEASUREMENT OF SEVEN-TRANSMEMBRANE RECEPTOR ACTIVITY USING A TRIPLE-ADDITION ASSAY
BASIC PROTOCOL 1
Once a suitable target-expressing cell system capable of robust coupling to changes in intracellular calcium has been established (see Strategic Planning), the first step in the development of a triple-addition assay is to determine the appropriate cell number and the timing, volumes, and concentrations for agonist addition. This will help to determine whether a triple-addition assay will be feasible with the target and cell background of interest. Investigators should use these instructions as a general framework to guide initial development of triple-addition assays. The suggested addition volumes, time intervals, and concentrations should serve as beginning values that may be varied based on experimental data with individual targets and cell lines. The choice of a 2× concentration for the first addition is intended to present the test compounds in as low a concentration as possible relative to their final intended assay concentration in an attempt to maximize their solubility. The choice of an EC20 concentration for the first agonist addition is intended to provide enough agonism to ensure that some level of activity is observed in every well, accounting for normal well-to-well variation, while keeping agonism low enough to help ensure a large window to observe potentiation. The choice of an EC80 concentration for the second agonist addition is intended to provide a large window to observe a signal decrease in the presence of antagonists without using a saturating concentration that could negatively impact the sensitivity of the assay for certain types of antagonists.
Materials Cells expressing the seven-transmembrane receptor of interest (e.g., mGlu5-expressing HEK-293 cells), e.g., in a T175 tissue culture flask Cell plating medium (see recipe) Dye loading solution (see recipe)
Triple-Addition Protocols for 7-TM Receptors
121 Current Protocols in Chemical Biology
Volume 3
Seven-transmembrane-receptor agonist (e.g., l-glutamic acid, Tocris, cat. no. 0218; see recipe) Assay buffer (e.g., 20 mM HEPES-buffered HBSS, pH 7.3; see recipe) Dimethyl sulfoxide (DMSO) 10-ml Pasteur pipets Black-walled, clear-bottom, poly-D-lysine-coated 384-well plates (e.g., Becton Dickinson, cat. no. 356936) Multichannel electronic pipettor (e.g., Biohit, cat. no. 73036X; optional) Humidified, 37◦ C, 5% CO2 cell culture incubator 384-well polypropylene compound plates (e.g., Greiner, cat. no. 781280) Kinetic imaging plate reader (e.g., Hamamatsu FDSS 6000) Spreadsheet software (e.g., Excel 2010, Microsoft, with XLfit add-in, IDBS) Additional reagents and equipment for counting cells with a hemacytometer and trypan blue exclusion (e.g., Phelan et al., 2006) NOTE: All steps are performed at room temperature with medium that has been prewarmed to room temperature.
Plate cells 1. Remove target-expressing cells from a T175 tissue culture flask using standard protocols, taking care to produce a cell suspension free of clumps. 2. Pellet cells by centrifuging 5 min at 500 × g, room temperature. Remove supernatant. 3. Resuspend cell pellet in 10 ml cell plating medium by gently triturating with a 10-ml Pasteur pipet. 4. Determine cell concentration using a hemacytometer and trypan blue exclusion (e.g., Phelan, 2006). 5. Dilute the cell suspension to 1000 cells/μl in cell plating medium. 6. Plate 20 μl/well (2 × 104 cells/well) in one or more black-walled, clear-bottom, poly-D-lysine-coated 384-well plates. The most appropriate plating volume and cell concentration will vary depending on cell type, target, and assay technology. These factors should be explored as an initial step in the development of a triple-addition assay. Typical cell concentrations range from 250 to 5000 cells/μl and typical plating volumes are from 10 to 50 μl/well in standard-volume 384-well plates. Cells may be plated with an electronic multichannel pipettor or, if many plates are required, a device such as a Multidrop Combi Reagent Dispenser (Thermo Fisher).
7. Cover the cell plates and incubate overnight in a humidified, 37◦ C, 5% CO2 cell culture incubator. Some investigators report that allowing cells to settle and attach to the bottom of the plates before placing them in the incubator promotes more uniform cell coverage in the wells.
Load dye and prepare ligand plate 8. Remove cell plating medium and replace with 20 μl/well dye loading solution.
Triple-Addition Protocols for 7-TM Receptors
Cell plating medium may be removed by holding the plate over a sink and slinging most of the medium out with a flick of the wrist, followed by firm but gentle patting of the plate on a stack of paper towels. Alternatively, a cell washer (e.g., ELx405 CW, BioTek) may be used to remove cell plating medium. Dye loading solution may be added with a multichannel electronic pipettor or Multidrop Combi (or equivalent).
122 Volume 3
Current Protocols in Chemical Biology
Some investigators choose to use so-called “no wash” dye kits. While these kits may cost substantially more, contain unnecessary reagents, and affect/complicate target pharmacology, they may be desirable under certain circumstances, such as with poorly adherent cells. When kits are used, investigators should refer to the manufacturer’s instructions.
9. Cover cell plates and incubate for 1 hr at room temperature. Loading times and temperatures may vary with cell type and assay reagents. The most appropriate conditions should be determined empirically as part of the development of any specific triple-addition assay protocol.
10. Prepare the first addition plate (2× agonist) by adding 40 μl/well of serially diluted agonist in assay buffer containing 0.2% (v/v) DMSO in columns 1-8 of a 384-well polypropylene plate. Place the highest concentration of agonist in wells A1-A8, the next highest in wells B1-B8, and so on through O1-O8. Add 40 μl/well of assay buffer containing 0.2% (v/v) DMSO to the remaining wells of the plate. Preparation of agonist stocks, serial dilution of those stocks, and agonist concentration ranges will vary with agonist, target (and target expression level), and assay technology. Typically a concentration series ranging from a final assay concentration 100-fold above a known maximally effective concentration for the target of interest using a similar assay technology, followed by fourteen 3-fold serial dilution steps, results in an appropriate concentration-response series from which to make decisions regarding the two agonist concentrations required for a triple-addition assay. For each target, DMSO tolerance tests should be conducted. In general, DMSO concentrations should be kept as low as possible, consistent with what is practical for the concentration and stability of compounds being tested. Typically, final DMSO concentrations in triple-addition assays range from 0.1% to a maximum of 1%.
11. Prepare the second addition plate (5× agonist) by adding 40 μl/well of serially diluted agonist in assay buffer in columns 1-16 of a 384-well polypropylene plate. Place the highest concentration of agonist in wells A1-A16, the next highest in wells B9-B16, and so on through O1-O16. Add 40 μl/well of assay buffer to the remaining wells of the plate. 12. Prepare the third addition plate (5× agonist) by adding 40 μl/well of serially diluted agonist in assay buffer in columns 1-24 of a 384-well polypropylene plate. Place the highest concentration of agonist in wells A1-A24, the next highest in wells B1-B24, and so on through O1-O24. Add 40 μl/well of assay buffer to the remaining wells of the plate. 13. Remove dye loading solution from the plate containing cells and replace with 20 μl/well assay buffer. Solutions may be removed and added as described above. This step may be omitted if a “no wash” kit is used (refer to kit’s instructions). Optimal timing between the dye removal step and assay execution may vary from target to target. The optimal interval may be evaluated by performing Basic Protocol 1 with intervals ranging from 1 to 10 min. Care should be taken to observe whether timing affects apparent agonist potency or well-to-well variability.
Obtain measurements 14. Load the agonist addition plates in the appropriate stage/nest positions on the kinetic imaging plate reader. 15. Transfer a cell plate to the kinetic imaging plate reader. 16. Execute triple-addition assay using the following conditions:
Excitation 480 nm (40-nm bandpass) Emission 530 nm (40-nm bandpass)
Triple-Addition Protocols for 7-TM Receptors
123 Current Protocols in Chemical Biology
Volume 3
Image acquisition rate 1 frame/sec First addition of 20 μl/well at 10 sec (dispense rate 20 μl/sec, dispense height 1.7 mm) Second addition of 10 μl/well at 4 min 10 sec Third addition of 12 μl/well at 5 min 10 sec Additional data collection 1 min. The exact excitation and emission wavelengths, acquisition rate, addition timing, dispense volumes, and other reader settings may vary with cell type, molecular target, and assay technology. These factors should be explored as an early part of establishing any tripleaddition assay.
Analyze data 17. Export the data as a text file and open it in a spreadsheet application such as Excel. 18. Optional: Correct fluorescence measurements for background contributions. Subtract the values from each well of a plate containing only assay buffer from each data point in each of the corresponding wells of the test plate(s). Background subtraction is usually only necessary if the background represents >10% of the total signal. Background measurements should be collected at the same instrument settings (exposure length, camera gain, illumination intensity, and filter settings) used for the data being analyzed. Some kinetic imaging plate readers can be set to subtract background automatically. Investigators can choose this option, but should make sure that the background correction corresponds to the appropriate plate type and reader settings. Refer to the instrument’s user manual for specific instructions.
19. Optional: Normalize the response in each well based on that well’s initial values. To do this, divide each value for each well by the first value obtained for that well. If the initial value for each well accurately represents the dye-loaded cell number, this ratio operation will normalize the readings from each well to the dye-loaded cell number. Doing this will help remove some errors due to uneven cell plating and uneven optical properties of the reader. Some kinetic imaging plate readers can be set to perform this or a similar operation automatically. Care should be taken when performing this normalization to make sure that the initial counts are actually dominated by the dye-loaded cell number and not background contributions, particularly if those contributions vary from well to well. Refer to the instrument’s user manual for specific instructions.
20. Optional: Subtract a representative negative control waveform from the waveform from each well. To do this, subtract the values from one or more waves representing a user-defined negative control condition (e.g., dye-loaded cells treated with vehicle/assay buffer) from the corresponding values for each wave. Careful choice of the appropriate negative control wells can be a powerful mechanism to reveal subtle/small signals that may not be obvious in the raw data traces. Some kinetic imaging plate readers can perform this function in their software. Refer to the instrument’s user manual for specific instructions.
21. Extract three data points from each well’s fluorescence versus time waveform to yield reduced data. To do this, define time windows bounding the three sample addition steps of the triple-addition experiment. Within each window, perform a data reduction function. Triple-Addition Protocols for 7-TM Receptors
124 Volume 3
Current Protocols in Chemical Biology
A
B
%Emax
100
50
0 0 Time (sec)
240
Time (sec)
240
EC80
100 Fmax–Fmin
0
50 EC20
0 Log[agonist]
Log[agonist]
Figure 2 Data from two experiments executed as part of the development of a triple-addition assay using Basic Protocol 1. (A) Data traces resulting from adding vehicle (0.2% DMSO in assay buffer) in the first addition followed by a concentration series of an agonist for the target of interest in the second addition, and vehicle again in the third addition. (B) Similar set of data where the first addition was vehicle, the second addition was ∼EC20 of the agonist, and the third addition was a concentration series of the agonist. Arrows show timing of the first, second, and third additions. Gray areas bounded by dotted lines show the time range where the maximum and minimum fluorescence values were obtained and used to generate concentration-response curves. EC20 and EC80 values are obtained from those curves, and these values were used in Basic Protocols 2-4 for the second and third additions, respectively.
Most commonly for calcium flux data generated as part of a Gq-coupled seventransmembrane receptor assay, a maximum change in fluorescence amplitude is calculated by subtracting the minimum value within the specified time window from the maximum value in that window. A variety of data reduction functions may be performed depending on the specifics of the target and assay type. These may include, but are not limited to, slope, average amplitude over a specific time range, or integrated signal over a specific time range. Some kinetic imaging plate readers can perform this or a similar operation automatically. Refer to the instrument’s user manual for specific instructions.
22. Once reduced data have been obtained for each of the three additions in the tripleaddition experiment, fit the data to produce concentration-response curves (Fig. 2). Much has been written about fitting data to produce and interpret concentration response curves (e.g., Motulsky and Christopoulos, 2004).
SIGNAL UNIFORMITY VALIDATION FOR HIGH-THROUGHPUT SCREENING FOR MODULATORS OF SEVEN-TRANSMEMBRANE RECEPTORS USING A TRIPLE-ADDITION ASSAY Once general cell conditions have been established for a triple-addition assay, the next step in developing the assay for use in high-throughput screening is to evaluate the Current Protocols in Chemical Biology
BASIC PROTOCOL 2 Triple-Addition Protocols for 7-TM Receptors
125 Volume 3
90 EC80
%Emax
70
50
Z⬘
30 EC20 10 VHL ⫺10 0
100
200
300
Well number
Figure 3 Example of a plot useful for viewing and interpreting data from a triple-addition assay as part of validation for high-throughput screening (Basic Protocol 2). Points represent values obtained from each of the three additions for an entire 384-well plate, where each well is treated with identical conditions. The first addition is vehicle (VHL, 0.2% DMSO in assay buffer). The second and third additions are EC20 and EC80 concentrations of agonist, respectively, determined from the concentration-response curves in Basic Protocol 1. Solid horizontal lines represent the means of each population; dashed lines represent the three standard deviation boundaries for each population of data. The distance between the standard deviation boundaries of any two of the populations (e.g., EC20 and EC80 ) as a fraction of the distance between the population’s means corresponds to Z .
well-to-well uniformity and stability of the assay. This protocol describes a general method for assessing well-to-well uniformity of responses and quantifying them using coefficient of variance, as well as assessing the separation of activity levels in the assay, EC20 and EC80 , using the popular Z statistic (Zhang et al., 1999). The plot shown in Figure 3 is a useful way to visualize patterns in data that may be caused by edge effects or other systematic errors. When such effects are observed, they may often be reduced or eliminated through experimentation with assay conditions or instrument maintenance. Time spent reducing systematic errors, leading to uniform separation of control populations across screening plates, is time well spent towards ensuring a high-quality triple-addition assay. This protocol should be repeated on at least three days with cells and reagents prepared as they would be to support a high-throughput screen. For some targets and some cell lines, large batches of reagents and cells can be prepared in one session, validated for screening, aliquoted, stored, and used throughout the screen. For other targets and cell lines this is not the case. This procedure provides a framework for evaluating this and other variables that will affect screening logistics and screen quality.
Plate cells 1. Plate a minimum of three cell plates as described in Basic Protocol 1. Triple-Addition Protocols for 7-TM Receptors
Load dye and prepare ligand plates 2. Load cell plates with dye as described in Basic Protocol 1. 3. Prepare three 384-well polypropylene ligand plates.
126 Volume 3
Current Protocols in Chemical Biology
a. Into each well of the first plate place 70 μl/well assay buffer containing 0.2% (v/v) DMSO. b. Into each well of the second plate place 70 μl/well assay buffer containing a concentration of agonist 5-fold over the EC20 determined from the curve fits obtained from the second addition window in Basic Protocol 1. c. Into each well of the third plate place 70 μl/well of assay buffer containing a concentration of agonist 5-fold over the EC80 determined from the curve fits obtained from the third addition window in Basic Protocol 1. Many cell lines and targets will display stable EC20 and EC80 values over many experimental days. In this case, Basic Protocol 1 and Basic Protocol 2 may be executed on separate days. Other cell lines and targets require that the protocols be run on the same day to ensure accurate determination of EC20 and EC80 concentrations for use in Basic Protocol 2.
Obtain measurements 4. Execute the same kinetic plate reader protocol used in Basic Protocol 1. Run this protocol three times on three cell plates with the set of ligand plates described above. Care should be taken to wash tips well between ligand additions and between cell plates to minimize ligand carryover from plate to plate. Some kinetic imaging plate readers provide a mechanism to change tips between experiments. Investigators may choose this option if available when ligand carryover is a serious concern. Refer to the instrument’s user manual for specific instructions on setting up washing and/or tip changing protocols.
Analyze data 5. Analyze data using the methods described in Basic Protocol 1 up to the point of fitting concentration-response curves (i.e., through step 21). 6. Instead of fitting concentration-response curves, average the data points from each of the three time windows associated with each of the addition steps and calculate standard deviations of the mean. Use these averages and standard deviations to calculate Z values for the second addition (EC20 ) and third addition (EC80 ) (Fig. 3) and percent coefficients of variation (%CV) for each of the three time windows. Z = 1 − (3σ EC80 + 3σ EC20 )/|μEC80 -μEC20 |, where σ and μ are the standard deviation and means of the EC20 and EC80 populations, respectively. %CV = 100 × σ /μ As a rule of thumb, a Z of 0.5 measured for relevant data populations is considered suitable for high-throughput screening. However, the better resolved relevant populations are across entire screening plates, the better the screen will be for detecting subtle hits.
HIGH-THROUGHPUT SCREENING FOR MODULATORS OF SEVEN-TRANSMEMBRANE RECEPTORS USING A TRIPLE-ADDITION ASSAY
BASIC PROTOCOL 3
After conditions that provide a high degree of well-to-well uniformity (typically %CV < 10%) and good separation of EC20 and EC80 populations (typically Z > 0.5) have been achieved, the use of the triple-addition assay to screen a collection of compounds can be initiated. This protocol provides a framework for conducting a high-throughput screen and selecting hits. Before beginning large-scale screening it is advisable to perform a small-scale pilot screen. The purpose of pilot screening is to test how well the triple-addition protocol performs under actual screening conditions. When available, it is advisable to include some compounds of known activity (e.g., agonists, antagonists, and potentiators) seeded into the set of pilot screening plates in a blinded fashion as controls. In addition, small Current Protocols in Chemical Biology
Triple-Addition Protocols for 7-TM Receptors
127 Volume 3
collections (∼2000 samples) containing some compounds with known activities are attractive for pilot screening. Collections such as the Spectrum Collection (MicroSource) and LOPAC (Sigma-Aldrich) are examples. Use of such a collection can often provide a set of compounds that might be expected to be active agonists or antagonists for the target of interest, since these collections are often enriched in seven-transmembrane modulators. An examination of the expected hits from the pilot screening collection or the control compounds seeded into the pilot screening collection will provide a basis for estimating the screen’s false-negative rate; in some cases, the pilot screen may provide unanticipated hits that may provide proof-of-concept for the screening approach and a potentially useful control for additional screening if no such control was previously available. Occasionally, the pilot screen may provide sufficiently novel findings to preclude the immediate need for larger-scale screening. In other instances the pilot screen may reveal unforeseen complications with the screening protocol, such as poor reproducibility of hits, large amounts of interfering compounds, or a high false-positive rate. Such findings, while disappointing, may lead to a retooling of the screen and a substantial time and cost savings compared to running a large-scale high-throughput screen that was compromised from the beginning. Immediately following the high-throughput screen, it is advisable to pick the compounds identified as hits and retest them in duplicate to confirm that they are hits. It is further advisable to test the hits in a cell line that does not possess the target of interest but is largely identical to the cell background that the target was screened in. It is desirable to have the parental cell line if the target is heterologously expressed (e.g., the parental HEK-293 cell line if the target receptor is expressed in HEK-293 cells), or to be able to knock down or inhibit the expression of the target in the target-expressing cell line using an siRNA or inducible expression system, as appropriate. Compounds that show activity only in the cell line expressing the target but not in the cells that do not express the target should be pursued by investigating the concentration-dependence of their activity (see Basic Protocol 4).
Additional Materials (also see Basic Protocol 1) Control compounds, if available (e.g., mGluR5 potentiator VU 1545, Tocris, cat. no. 3325; mGluR5 antagonist MPEP hydrochloride, Tocris, cat. no. 1212) Collection of compounds to be screened Means of transferring test compounds from source plates to daughter plates: e.g., a contact-based liquid handler (e.g., Bravo, Agilent) equipped with a pin tool (VP Scientific) or an acoustic liquid transfer device (e.g., Echo 550, Labcyte) Plate cells 1. Plate cell plates as described in Basic Protocol 1. Load dye and prepare test compound plates 2. Load cell plates with dye as described in Basic Protocol 1. 3. Prepare three sets of 384-well polypropylene plates.
Triple-Addition Protocols for 7-TM Receptors
a. Into wells A1-P2, A23-P24 of the first plate place 40 μl/well assay buffer containing 0.2% (v/v) DMSO. b. Into the remaining wells of the first plate place 40 μl/well assay buffer containing 20 μM test compound and 0.2% DMSO. c. Into each well of the second plate, place 25 μl/well assay buffer containing a concentration of agonist 5-fold over the EC20 determined from the curve fits obtained from the second addition window in Basic Protocol 1.
128 Volume 3
Current Protocols in Chemical Biology
d. Into each well of the third plate place 25 μl/well of assay buffer containing a concentration of agonist 5-fold over the EC80 determined from the curve fits obtained from the third addition window in Basic Protocol 1. Many cell lines and targets will display stable EC20 and EC80 values over many experimental days. In this case, Basic Protocol 1 and Basic Protocol 3 may be executed on separate days. Other cell lines and targets require that the protocols be run on the same day to ensure accurate determination of EC20 and EC80 for use in Basic Protocol 3. If available, control compounds can be added to every third well of columns 1, 2, 23, and 24 on each screening plate. Prior to full-scale high-throughput screening, available control compounds should be added to pilot screening plates in a blinded fashion as a way of testing the screen’s ability to detect compounds of known activity. The 10 μM test compound concentration is a standard concentration used in highthroughput screening, but is by no means the only concentration that could be used for testing. As discussed in Basic Protocol 1, it is important to choose a DMSO concentration that will not interfere with the assay. A DMSO tolerance experiment should be performed as an early part of assay development. Investigators may choose to provide additional amounts of agonist-containing solution in the second and third plates so that they may be used for multiple test compound plates. If this approach is used, particularly on an instrument with a single pipetting head that does not change tips between transfer from plates 1 and 2, great care should be taken to ensure adequate tip washing between plates to prevent contamination of plate 2 with test compounds. Refer to the instrument’s user manual for specific instructions.
Obtain measurements 4. Execute the same kinetic plate reader protocol used in Basic Protocol 1. Run this protocol for every cell plate. Care should be taken to wash tips well between test compound plates to minimize test compound carryover from cell plate to cell plate. Some kinetic imaging plate readers provide a mechanism to change tips between experiments. Investigators may choose this option if available when test compound or ligand carryover is a serious concern. Refer to the instrument’s user manual for specific instructions.
Analyze data 5. Analyze data using the same methods as in Basic Protocol 1 up to the point of fitting concentration-response curves (i.e., through step 21). 6. Instead of fitting concentration-response curves, average the data points for all three time windows for the vehicle control wells and positive control wells, if applicable, and calculate standard deviations. 7. Use these averages and standard deviations to calculate Z and coefficients of variation as in Basic Protocol 2 as a means to assess plate-to-plate quality. 8. Average the data points for the test compounds from each of the three time windows associated with each of the addition steps and calculate standard deviations of the mean. 9. From these numbers, establish a z-score threshold for hit picking. A commonly used hit-picking cutoff is z >3 for agonists and potentiators and z < −3 for antagonists (Fig. 4). Alternate or additional hit picking approaches may also be used, including B scores (Brideau et al., 2003) and other approaches (Posner et al., 2009). In addition, when practical for smaller test sets, direct inspection of the raw or normalized waveforms can be quite useful for identifying subtle effects and eliminating false positives.
Triple-Addition Protocols for 7-TM Receptors
129 Current Protocols in Chemical Biology
Volume 3
A
B
C
90 %Emax
70 50 30 10 ⫺10 0
100
200
Well number
300
0
100
200 Well number
300
0
100
200
300
Well number
Figure 4 Example of a single 384-well screening plate tested using a triple-addition assay as described in Basic Protocol 3. (A) Values derived from the first (agonist) window of a triple-addition assay. Points surrounded by circles are putative agonist hits based on a hit threshold of z > 3. (B) Results for the second (potentiator) window. The point surrounded by a circle is the same as one selected as a hit from the agonist window. Points surrounded by diamonds are putative potentiator hits based on an a hit threshold of z > 3. Points surrounded by squares are putative antagonist hits based on a hit threshold of z < −3. (C) Results for the third (antagonist) window. The point surrounded by a circle is the same as one selected as a hit from the agonist window. The point surrounded by a diamond is the same as one selected as a hit from the potentiator window. Points surrounded by squares are putative antagonist hits based on a hit threshold of z < −3. In all cases the hit thresholds based on the z-score are shown with dashed lines, and points at the far right and left represent positive controls.
BASIC PROTOCOL 4
MEASUREMENT OF CONCENTRATION-RESPONSE RELATIONSHIPS FOR MODULATORS OF SEVEN-TRANSMEMBRANE RECEPTORS USING A TRIPLE-ADDITION ASSAY This protocol provides an example of how the triple-addition assay can be applied to concentration series experiments. It can be useful as stand-alone assay for evaluating the potency of efficacy of compounds. The triple-addition approach is particularly useful for evaluating sets of related compounds when one wishes to observe how changes in compound structure translate to changes in activity and transitions between modes of activity (e.g., potentiation to antagonism). Such transitions are not uncommon for some series of seven-transmembrane-receptor modulators (Rodriguez et al. 2005; Sharma et al., 2008, 2009). This protocol can also be used as part of a progression from hits to leads, as alluded to in Basic Protocol 3. Following hit retest and screening against an appropriate control cell line that does not express the target of interest, compounds should be tested using this protocol to determine whether they produce concentration-dependent efficacy. Compounds that produce concentration-dependent effects may move forward for further characterization. Care should be taken when analyzing concentration-response relationships, particularly when characterizing hits from a high-throughput screen. Some compounds that produce concentration-dependent effects may do this via nonspecific mechanisms such as compound aggregation (Shoichet, 2006). Appropriate controls, such as cell lines that do not contain the target of interest, are useful in identifying “promiscuous” hits. Use of the triple-addition approach offers an opportunity to easily observe how compounds behave when added alone and in the presence of two concentrations of agonist. These data may provide a basis to categorize activity profiles that may in turn provide valuable insights about the compounds.
Triple-Addition Protocols for 7-TM Receptors
The example provided in Figure 5 shows three separate eleven-point serial dilutions positioned at different regions on a plate. The curve fits are performed on the three individual data points extracted from the raw data waves. The specific approach outlined in this protocol is not the only way to construct a dilution series, lay out an assay plate, or
130 Volume 3
Current Protocols in Chemical Biology
A
17
18
19
20
21
22
23
24
P S0 S1 S2 S3 S4 S5 S6 S7 S8 S9 S0 S1 S2 S3 S4
16
P S0 S1 S2 S3 S4 S5 S6 S7 S8 S9 S0 S1 S2 S3 S4
15
8
P S0 S1 S2 S3 S4 S5 S6 S7 S8 S9 S0 S1 S2 S3 S4
14
7
P S0 S1 S2 S3 S4 S5 S6 S7 S8 S9 S0 S1 S2 S3 S4
13
6
P S0 S1 S2 S3 S4 S5 S6 S7 S8 S9 S0 S1 S2 S3 S4
12
5
P S0 S1 S2 S3 S4 S5 S6 S7 S8 S9 S0 S1 S2 S3 S4
11
4
P S0 S1 S2 S3 S4 S5 S6 S7 S8 S9 S0 S1 S2 S3 S4
9
3
V V V V V V V V V V V V V V V V
10
2
A B C D E F G H I J K L M N O P
[Sample]
1
[Sample]
P S0 S1 S2 S3 S4 S5 S6 S7 S8 S9 S0 S1 S2 S3 S4
P S0 S1 S2 S3 S4 S5 S6 S7 S8 S9 S0 S1 S2 S3 S4
P S0 S1 S2 S3 S4 S5 S6 S7 S8 S9 S0 S1 S2 S3 S4
P S0 S1 S2 S3 S4 S5 S6 S7 S8 S9 S0 S1 S2 S3 S4
S5 S6 S7 S8 S9 S0 S1 S2 S3 S4 S5 S6 S7 S8 S9 P
S5 S6 S7 S8 S9 S0 S1 S2 S3 S4 S5 S6 S7 S8 S9 P
S5 S6 S7 S8 S9 S0 S1 S2 S3 S4 S5 S6 S7 S8 S9 P
S5 S6 S7 S8 S9 S0 S1 S2 S3 S4 S5 S6 S7 S8 S9 P
S5 S6 S7 S8 S9 S0 S1 S2 S3 S4 S5 S6 S7 S8 S9 P
S5 S6 S7 S8 S9 S0 S1 S2 S3 S4 S5 S6 S7 S8 S9 P
S5 S6 S7 S8 S9 S0 S1 S2 S3 S4 S5 S6 S7 S8 S9 P
S5 S6 S7 S8 S9 S0 S1 S2 S3 S4 S5 S6 S7 S8 S9 P
S5 S6 S7 S8 S9 S0 S1 S2 S3 S4 S5 S6 S7 S8 S9 P
S5 S6 S7 S8 S9 S0 S1 S2 S3 S4 S5 S6 S7 S8 S9 P
S5 S6 S7 S8 S9 S0 S1 S2 S3 S4 S5 S6 S7 S8 S9 P
V V V V V V V V V V V V V V V V
B
%Emax
80
0 0
240 Time (sec)
C
D
E
%Emax
100
0 Log [test sample]
Log [test sample]
Log [test sample]
Figure 5 Example of a plate layout and data obtained from a triple-addition assay performed as described in Basic Protocol 4. (A) A 384-well test compound plate layout where wells in columns 1 and 24 contain vehicle (V, 0.2% DMSO in assay buffer). Wells A2-A12 and P13-24 contain a known positive control (P, e.g., allosteric potentiator). The remaining wells contain test sample compounds (S0-S9). Black wedges at the top of the figure indicate that the positive control and samples are plated at different concentrations beginning with highest concentrations in columns 2 and 13 and declining to lowest concentrations in columns 12 and 23. Each concentration of each sample is represented in triplicate and each concentration of positive control is in duplicate. Locations for sample S0 are highlighted in gray. (B) Data traces from a concentration series for one set of wells of a representative test compound. Areas bounded by dotted lines show the time range where maximum and minimum fluorescence values were obtained. These values were used to generate concentrationresponse curves shown below in panels C-E. (C-E) Fits to triplicate data points for the representative test compound for the agonist, potentiator, and antagonist windows, respectively.
Triple-Addition Protocols for 7-TM Receptors
131 Current Protocols in Chemical Biology
Volume 3
analyze the data. It is intended to generate fits over a broad range of concentrations and account for errors that may be contributed by the serial dilution process or by specific locations on the plate.
Plate cells 1. Plate cell plates as described in Basic Protocol 1. Load dye and prepare test compound plate 2. Load cell plates with dye as described in Basis Protocol 1. 3. Prepare three sets of 384-well polypropylene ligand plates. a. For the first plate, into each well of columns 1 and 24 place 40 μl/well assay buffer with 0.2% DMSO. b. Into the remaining wells place 40 μl/well of serially diluted test compounds in assay buffer with 0.2% DMSO. c. Set up the second and third plates as described in Basic Protocol 3. An example of a plate layout is provided in Figure 5. As a starting place for compounds identified as active using 10 μM as a screening concentration in a high-throughput screen according to Basic Protocol 3, a top final concentration for the concentration series should be 30 μM declining in 3-fold steps for a total of eleven concentrations ranging from 30 μM to 0.5 nM. When a series of compounds is known to be considerably more or less potent, the starting point for the concentration series can be adjusted. Similarly, the dilution factor between steps can be adjusted as appropriate to produce high-quality fits. For most compounds, serial dilutions should be performed in DMSO with tip changes or extensive washing between dilution steps. The resulting serially diluted compounds are ready for low-volume plate-to-plate transfer as described in Basic Protocol 3. When available, control compounds representing the type(s) of activity (e.g., potentiator, antagonist) being studied should be included as concentration series references during each screening run. In a plate layout such as the one depicted in Figure 5, control concentration series may be placed in wells A2-12 and P13-23.
Obtain measurements 4. Execute the same reader kinetic plate reader protocol used in Basic Protocol 1. Run this protocol for every cell plate. Care should be taken to wash tips well between test compound plates to minimize test compound carryover from cell plate to cell plate. This is critically important for this protocol, since it is reasonable to believe that the compounds tested may be very active and thus liable to contaminate subsequent plates. When affordable, changing tips between each compound test plate is preferred.
Analyze data 5. Analyze data using the same methods as in Basic Protocol 1. Good practices for fitting and interpreting concentration-response curves can be found in Motulsky and Christopoulos (2004). Examples of fits for each of three time windows for a compound tested in triplicate are shown in Figure 5. This compound apparently displays a degree of agonism in the first time window (C) as well as potentiation in the second time window (D). Receptor desensitization may be apparent in the third time window (E). Triple-Addition Protocols for 7-TM Receptors
132 Volume 3
Current Protocols in Chemical Biology
ALTERNATE PROTOCOLS The basic protocols described in this article may be modified in a variety of ways to suit the specific needs of the scientific question at hand. For example, for discovery or characterization of compounds that modify the activity of an allosteric potentiator (either negatively or positively), the assay may be reconfigured so that the test compound is added first, an EC30 of an orthosteric agonist is added second to determine whether the test compound appears to have modulatory activity of its own, and finally an EC70 concentration of a known allosteric potentiator is added third to assess whether the unknown compound has the capacity to potentiate or inhibit the activity of the known allosteric potentiator. It is becoming increasingly evident that a single assay technology is not sufficient to describe the activity of a molecule at a target. Therefore, whether the triple-addition assay is used to support hit identification as part of a high-throughput screen or is used to support structure-activity relationship generation or lead optimization, it should be used as a part of a screening tier. Other assays designed to complement the triple-addition assay such as phosphoinositide hydrolysis, radioligand binding, β-arrestin recruitment, and others should be considered. Modification of the concepts outlined in these triple-addition assay protocols may be applied to a wide variety of target classes in addition to seven-transmembrane receptors including, but not limited to, ion channels and transporters. Furthermore, triple-addition assay protocols may be used with a variety of assay types/detection modalities in addition to calcium flux including, but not limited to, thallium flux, voltage-sensitive dyes, the Promega GloSensor cAMP assay, and label-free technologies such as MDS Cell Key and Corning Epic (now available in the Perkin Elmer EnSpire multimode plate reader).
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. Use freshly opened or appropriately stored DMSO to ensure that it has not absorbed a significant amount of water.
Agonist and antagonist solutions Agonist and antagonist solutions should be prepared as concentrated stocks in DMSO (or another appropriate solvent) and diluted in assay buffer to make solutions that are 5-fold higher than the final targeted assay concentration. Alternatively, agonist and antagonist solutions may be prepared directly in assay buffer if they are soluble in aqueous solution. For peptide ligands, the addition of fatty-acid-free bovine serum albumin (0.1% w/v) to the assay buffer may be advantageous to decrease ligand interaction with plastics from pipet tips and plates.
Assay buffer A typical assay buffer for a triple-addition assay includes HBSS (see recipe) plus 20 mM HEPES, pH 7.3. Some cell lines, e.g., Chinese hamster ovary (CHO) cells, possess a large capacity to transport organic anions such as the fluorescent indicator dyes used in the assay. For such cell lines, the addition of a transport inhibitor like sulfinpyrazone or probenecid may be required for good dye loading and retention (Di Virgilio et al., 1988). Assay buffer without transport inhibitors can be stored at room temperature or 4◦ C for at least 3 months. Assay buffer containing transport inhibitors should be used the same day. Triple-Addition Protocols for 7-TM Receptors
133 Current Protocols in Chemical Biology
Volume 3
Cell plating medium DMEM (e.g., Invitrogen, cat no. 11960-069) 1 mM sodium pyruvate 10% (v/v) dialyzed fetal bovine serum 20 mM HEPES buffer, pH 7.3 Dissolve components in DMEM. Filter through a sterile 0.2-μm filter into a sterile container and store at 4◦ C for up to 1 month. Many different types of cells may be amenable to triple-addition assays. This cell plating medium is one example of a medium appropriate for HEK-293 cells. Investigators may add antibiotics (e.g., G418) as required to maintain selection for their target of interest when stable cell lines are used for screening.
Dye loading solution Dye stock solution: Dissolve 1 mg (e.g., Fluo-2 MA [AM], Teflabs) or other appropriate dye in 400 μl DMSO. Dispense in 10- to 50-μl aliquots in 0.5-ml microcentrifuge tubes and store up to 3 months at −20◦ C. Pluronic F-127 solution: Add 20% (v/w) Pluronic F-127 (Sigma-Aldrich, cat. no. P2443) to DMSO and stir to dissolve. Store up to 1 year at room temperature in a desiccator. Dye loading solution (for one 384-well assay plate): Combine 10 μl dye stock solution with 5 μl Pluronic F-127 solution and mix by pipetting. Add to 10 ml assay buffer and mix well. Store up to 4 hr protected from direct light.
Hanks’ balanced salt solution (HBSS) 138 mM NaCl 5.33 mM KCl 4.17 mM NaHCO3 1.26 mM CaCl2 0.49 mM MgCl2 0.44 mM KH2 PO4 0.41 mM MgSO4 0.34 mM Na2 HPO4 5.56 mM D-glucose HBSS can be stored at room temperature or 4◦ C for at least 3 months.
Test compound solutions Stock solutions: Although there is no universal solvent system for all compounds to be tested for activity using the triple-addition assay, unless known to be inappropriate, compounds should be dissolved at a concentration of 10 mM in DMSO. Depending on their structure, compounds may be stored for many months in a dry atmosphere at room temperature, 4◦ C, or −20◦ C. Working solutions: Test compound stock solutions should be diluted in assay buffer to 2-fold above the concentration at which they are to be tested. Typically this means that a test compound would be diluted from a 10 mM stock in DMSO to 20 μM in assay buffer with vigorous mixing. Test compound solutions should be used immediately (preferred) or stored for up to 4 hr at room temperature. Triple-Addition Protocols for 7-TM Receptors
134 Volume 3
Current Protocols in Chemical Biology
COMMENTARY Background Information Seven-transmembrane receptors represent one of the largest and most successfully developed classes of drug target. However, not all seven-transmembrane receptors have proven to be easy targets for therapeutic intervention. Many of the difficulties stem from the high degree of conservation of the orthosteric binding sites within certain receptor families (e.g., muscarinic acetylcholine receptors). This conservation has made the development of orthosteric ligands with adequate selectivity extremely difficult. In some cases, drug candidates that did not possess the appropriate degree of selectivity have shown severe doselimiting side effects. Another issue has been the difficulty in discovering and developing ligands for large-peptide receptors (e.g., GLP1 receptor). The apparent requirement for large orthosteric ligand size has generally proven incompatible with small-molecule discovery programs. In recent years allosteric modulators of seven-transmembrane receptors have been discovered and put forward in an attempt to address these problems. Allosteric modulators may also provide better levels of subtlety and control compared to orthosteric ligands, producing more temporally and physiologically appropriate effects by acting only in the presence of the naturally occurring orthosteric ligand. For recent reviews of allosteric modulators, readers are referred to a number of excellent publications (Christopoulos and Kenakin, 2002; Marino and Conn, 2006; Hoare, 2007; Leach et al., 2007; May et al., 2007; Rees et al., 2008; Conn et al., 2009; Wang et al., 2009; De Amici et al., 2010; Cleva and Olive, 2011; Urwyler, 2011). This new focus on allosteric modulation has prompted the need for new screening methodologies designed to more easily enable the discovery and characterization of allosteric modulators. Traditional functional approaches to orthosteric ligand discovery typically include screening for agonists with a single addition of test compound. For antagonists, a common screening mode includes adding test compounds to plates “off-line” and then executing an experiment, such as a calcium flux measurement, with a single addition of an EC80 concentration of ligand. Such screening approaches were often followed by radioligand displacement assays designed to select compounds that display competitive displacement of a known orthosteric ligand. These
approaches tend to bias screens that favor molecules with binding sites that overlap with or are the same as the orthosteric binding site. Early approaches to discovery of allosteric modulators either (1) looked for changes in binding of orthosteric ligands that weren’t consistent with competitive mechanisms (Im et al., 2003) or (2) replaced the EC80 agonist in single-addition experiments with an EC20 concentration (Marino et al., 2003; O’Brien et al., 2003). Eventually, protocols were modified to include three additions: test compounds, EC20 , and EC80 (Jacoby et al., 2006; Niswender et al., 2008; Rodriguez et al., 2010). Such approaches are now widely used for simultaneous screening of orthosteric and allosteric modulators. Using a triple-addition approach may increase the chance of detecting allosteric modulator scaffolds. It is evident from recent publications that small changes in the structure of allosteric modulators may lead to qualitative changes in efficacy, i.e., from inhibitor to potentiator, potentiator to inhibitor, potentiator to neutral/silent (Rodriguez et al., 2005; Sharma et al., 2008, 2009). Tripleaddition protocols provide a better opportunity to identify such phenomena as compared to single-addition approaches. For seven-transmembrane receptors, the ability to perform a triple-addition assay is enabled by the liquid handling capabilities of modern plate readers and by signal detection technologies that support continuous, realtime evaluation of the target’s activation state. The ability to add a concentration of ligand that modestly stimulates the receptor’s signaling pathway followed later in the same experiment by a ligand concentration that more fully stimulates the receptor’s signaling pathway provides for a broad dynamic range in the signal evoked by compounds that act as agonists, potentiators, and inhibitors of the assay signal. Triple-addition assays are most easily performed in systems where continuous, realtime monitoring of receptor activation is possible and where the signal monitored during the assay is under tight temporal regulation. The tight regulation and temporal control of intracellular calcium concentrations proves excellent for triple-addition assays, since addition of low concentrations of agonists often results in temporally discrete signals that return to baseline prior to the next addition, over timeframes consistent with high-throughput screening. It is important for investigators to realize that not everything that produces a signal
Triple-Addition Protocols for 7-TM Receptors
135 Current Protocols in Chemical Biology
Volume 3
in a triple-addition assay is a direct modulator of the target of interest, and the apparent mode of efficacy may be an artifact of the specific experimental protocol/signal detection modality. Assays that measure intracellular calcium also monitor many factors that are involved in intracellular calcium homeostasis, including but not limited to the target of interest. Other undeclared targets may include G-protein subunits, GTPase-activating proteins (GAPs), phospholipases, phosphoinositide kinases, IP3 receptors, and calcium pumps. Care should be taken in triple-addition assays to choose appropriate controls that help elucidate the mechanism of activation, potentiation, or inhibition of a signal. Similarly, the mode of efficacy observed in a tripleaddition assay may only partially describe the pharmacology/mechanism of the compound that yielded the effect. For instance, a compound that preferentially activates or potentiates β-arrestin recruitment and receptor internalization might appear as inactive or as an inhibitor in a calcium flux assay. Thus, though triple-addition assays can be quite valuable in producing much richer and more informative data sets than single-addition assays alone, they are not sufficient to completely characterize the actions of a compound on a target of interest.
Critical Parameters
Triple-Addition Protocols for 7-TM Receptors
Calcium indicator Calcium indicator dyes are readily available from a number of sources including Teflabs and Invitrogen. The specific protocols detailed in this article use Teflabs Fluo-2 MA (AM) or the closely related Fluo-4 AM from Invitrogen. A number of calcium indicator dyes are sold as “no wash” kits that contain the organic anion transporter probenecid and/or a cell-impermeant dye intended to quench extracellular fluorescence contributed by extracellular indicator dye or other extracellular sources of fluorescence (e.g., the test compound). Probenecid is intended to inhibit indicator dye unloading or compartmentalization, which is a problem in some cell lines. Some investigators may choose to use a fluorescent calcium indicator dye with a higher or lower affinity for calcium, or a dye with different optical properties, such as Fura-2 or Asante Red. In choosing an indicator dye, investigators should consider many factors, including cost, ease of use, and risks associated with the variety of available indicators and kits. Such risks include shifts in the apparent potency
of ligands and modulators. These shifts may be caused by differences in the dyes’ affinity for calcium as well as the ability of some extracellular “quench” dyes found in assay kits to dramatically affect the apparent potency of certain compounds for certain targets. Another risk is the degree of optical interference from test compounds. The degree of interference may depend both on the choice of dye and the optical properties of the compounds being tested. Multiwell plates Many cell types (e.g., CHO cells) will readily adhere to tissue culture (TC)−treated plastic plates. Others (e.g., HEK-293 cells) perform much better on plates coated with poly-D-lysine, and still others prefer other surfaces such as collagen, fibronectin, and laminin. Other factors, including the type of plastic used to make the bottom of the plate and how it is attached to the walls of the wells, can affect the performance of some dyes and the adherence of some cell types. These factors, as well as availability and cost, should all be considered when developing a triple-addition assay for high-throughput screening. Plate reader Although some triple-addition assays could be performed manually on slides or multiwell plates using a microscope, a plate reader capable of making three reagent additions is strongly preferred. Plate readers of this type include but are not limited to Hamamatsu FDSS, Hamamatsu μCell, Molecular Devices FLIPR, and Molecular Devices FlexStation.
Troubleshooting A large number of factors can affect the successful development of a triple-addition protocol. Many are not unique to triple-addition assays, but the complex nature of these assays may exacerbate them. Two of the most vexing factors that can affect the use of the tripleaddition approach are discussed below. Inconsistent EC20 and EC80 values The success of triple-addition protocols depends critically on an ability to achieve consistent results from well to well, plate to plate, and day to day. One issue that can cause problems is inconsistent responses to additions of compounds at EC20 and EC80 concentrations. A number of factors can affect EC20 and EC80 responses, including: whether the cell plating medium naturally contains a ligand for the receptor of interest;
136 Volume 3
Current Protocols in Chemical Biology
whether the cells expressing the receptor of interest produce a ligand for the receptor; cell plating density; time from plating until the experiment is executed; time from dye loading and removal until assay is executed; time between test compound, first ligand addition, and second ligand addition; and efficacy of tip washing between additions and plates (discussed below). It is important to be systematic and consistent with cell culture procedures including plating density, time after plating until the experiment is executed, and cell passage number. If the ligand for the receptor is known to be or may be present in a component of the plating medium, efforts can be made to eliminate that component from the medium. Many HEK-293 cell lines perform well when grown in serum-free medium overnight, for example. If the component cannot be removed without compromising cell health, then an effective mechanism to control the ligand concentration prior to the experiment is important. This is often best accomplished by thorough removal of plating medium prior to the experiment, typically after indicator dye loading. Both manual and automated removal of solutions from plates can be effective, but consistency in the process is very important. Variability in solution exchange can often be reduced by carefully designing and using an automated approach on a plate washing device such as the ELx405 plate washer (BioTek). Finally, the timing between the final solution exchange prior to starting the experiment (step 13 of Basic Protocol 1) and execution of the experiment can produce substantial effects on measured ligand potencies for some receptors. Systematic evaluation of timing effects and consistency in timing of protocol steps may be critical to achieving plate-to-plate and dayto-day success. When available, automation of these protocols may be a very valuable and effective way to increase reproducibility. Evaluation of the effects of timing between the final solution exchange and the execution of the experiment can be performed as noted in Basic Protocol 1. Carryover effects Some ligands, particularly peptide ligands, can be especially difficult to remove from pipet tips. Incomplete removal of such ligands or failure to isolate them from other assay components may introduce severe complications for triple-addition protocols when running multiple compound plates in succession. A number
of strategies may be used to address ligand carryover. First, if the plate reader has the ability to change pipetting heads or tips during an experiment, it is often valuable to isolate the highest ligand concentration (often EC80 ) on its own head/set of tips. In a configuration like this, the test compound and the lowest concentration of ligand (often EC20 ) are applied by one head/set of tips and the higher concentration of ligand is applied by the second head/set of tips. In such a scenario, care still must be taken to wash the pipet tips thoroughly between addition of test compound and the lower ligand concentration and again after the addition of the low ligand concentration, but the hope is that the lower ligand concentration will be low enough to enable more thorough elimination between plates. A valuable test to determine the effectiveness and limitations of this approach includes the addition of a potent and effective potentiator as a control. Failure to completely eliminate the ligand from the tips may cause the potentiator to appear as though it were an agonist. In this case, the experimental parameters may be adjusted by altering the washing conditions, incorporating three heads/tip sets (if possible), using fresh tips for every plate (if possible or economically feasible), or by simply understanding that “hits” identified in the first (agonist) window may not be real agonists and their true nature may be revealed through additional experimentation. Second, if it is not possible to isolate ligand addition steps with separate heads/tips sets, aggressive washing conditions may be pursued. If two or more tip wash stations are available, multiple solvents may be used in an attempt to adequately wash tips. A dualwash scenario employs a first wash station with DMSO, ethanol, or other organic solvent or solvent mixture, followed by a second wash station with water to further wash the tips and prepare them for the next experiment. A highly effective solvent system utilizes a first wash station containing 2%-4% (v/v) RBS35 detergent (Thermo Fisher) in 80:20 (v/v) water:ethanol, followed by water in the second wash station.
Anticipated Results Basic Protocol 1 will generate as set of concentration-response curves for each of the three additions in the triple-addition assay. Two of the curves will be used to determine EC20 and EC80 concentrations of agonist that will be used for Basic Protocols 2-4.
Triple-Addition Protocols for 7-TM Receptors
137 Current Protocols in Chemical Biology
Volume 3
Basic Protocol 1 will also yield information regarding the optimal number or cells/well, as well as proper choice of assay plates, calcium indicators, and whether or not plate washing/buffer exchange will be tolerated. Basic Protocol 2 will be used to demonstrate that the triple-addition assay can produce a high degree of well-to-well uniformity and good separation of activity levels in the assay (e.g., EC20 from EC80 ). These qualities will be assessed using a coefficient of variation (Z ) and by inspection of data using plots like the one shown in Figure 3. Achievement of good well-to-well performance and separation of activity levels will pave the way for using the triple-addition assay in Basic Protocols 3 and 4. Failure to achieve good results in Basic Protocol 2 will require troubleshooting to improve Basic Protocol 2 results. Basic Protocol 3 will generate a set of hits resulting from testing a set of compounds in a high-throughput screen. These hits may be immediately useful as tools or may form the basis for hits-to-leads studies resulting in an understanding of structure-activity relationships (SAR). Results from Basic Protocol 3 may also provide information about false-positive and false-negative rates in the triple-addition assay, which may provide the basis for further assay refinement. Basic Protocol 4 will generate concentration-response relationships for each of the additions in the triple-addition assay for each compound tested. From these concentrationresponse relationships, when appropriate based on whether high-quality data fits are achieved, potency and efficacy values for the compounds may be generated. In addition, qualitative assessments can be made to describe a compound’s behavior in each of the segments of the triple-addition experiment. As noted above, care must be taken not to over-interpret the data from triple-addition assays, but instead to use it as part of a suite of assays in a screening tier.
Time Considerations
Triple-Addition Protocols for 7-TM Receptors
Once the target constructs/cell lines have been constructed and confirmed to robustly couple to changes in intracellular calcium, and all of the required reagents and instruments are in hand, Basic Protocol 1 can usually be accomplished in 2 days including cell plating. Basic Protocol 2, when repeated on 3 separate days, can in theory be completed in as few as 4 days including cell plating. In practice, however, the refinement of all conditions re-
quired to produce high-throughput screeningready data usually requires an average of 1 month or more. The time required for Basic Protocol 3 depends on the scale of the high-throughput screen and how well the triple-addition assay, developed and validated using Basic Protocol 2, performs in actual screening. At least 1 month should be allowed for pilot-scale screening including follow-up tests (e.g., testing hits in a cell line that does not express the target and concentration-response assays as outlined in Basic Protocol 4). Although it is common for modern high-throughput screening labs to screen >>10,000 samples/day using the triple-addition protocol, at least 3 months should be budgeted for a screen of 100,000 samples including the follow-up assay noted above. Once Basic Protocols 1 and 2 have been successfully reduced to practice, Basic Protocol 4 can be accomplished in as little as 2 days including cell plating, but it would be wise to budget at least 1 week for the first attempt.
Literature Cited Bird, M.K. and Lawrence, A.J. 2009. The promiscuous mGlu5 receptor—a range of partners for therapeutic possibilities? Trends Pharmacol. Sci. 30:617-623. 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. Chen, J., Lake, M.R., Sabet, R.S., Niforatos, W., Pratt, S.D., Cassar, S.C., Xu, J., Gopalakrishnan, S., Pereda-Lopez, A., Gopalakrishnan, M., Holzman, T.F., Moreland, R.B., Walter, K.A., Faltynek, C.R., Warrior, U., and Scott, V.E. 2007. Utility of large-scale transiently transfected cells for cell-based high-throughput screens to identify transient receptor potential channel A1 (TRPA1) antagonists. J. Biomol. Screen. 12:61-69. Christopoulos, A. and Kenakin, T. 2002. G proteincoupled receptor allosterism and complexing. Pharmacol. Rev. 54:323-374. Cleva, R.M. and Olive, M.F. 2011. Positive allosteric modulators of type 5 metabotropic glutamate receptors (mGluR5) and their therapeutic potential for the treatment of CNS disorders. Molecules 16:2097-2106. Conklin, B.R., Farfel, Z., Lustig, K.D., Julius, D., and Bourne, H.R. 1993. Substitution of three amino acids switches receptor specificity of Gq alpha to that of Gi alpha. Nature 363:274-276. Conn, P.J., Christopoulos, A., and Lindsley, C.W. 2009. Allosteric modulators of GPCRs: A novel approach for the treatment of CNS disorders. Nat. Rev. Drug Discov. 8:41-54.
138 Volume 3
Current Protocols in Chemical Biology
Coward, P., Chan, S.D., Wada, H.G., Humphries, G.M., and Conklin, B.R. 1999. Chimeric G proteins allow a high-throughput signaling assay of Gi-coupled receptors. Anal. Biochem. 270:242248.
ulators of metabotropic glutamate receptors. Curr. Opin. Pharmacol. 6:98-102.
De Amici, M., Dallanoce, C., Holzgrabe, U., Tr¨ankle, C., and Mohr, K. 2010. Allosteric ligands for G protein-coupled receptors: A novel strategy with attractive therapeutic opportunities. Med. Res. Rev. 30:463-549.
Marino, M.J., Williams, D.L. Jr., O’Brien, J.A., Valenti, O., McDonald, T.P., Clements, M.K., Wang, R., DiLella, A.G., Hess, J.F., Kinney, G.G., and Conn, P.J. 2003. Allosteric modulation of group III metabotropic glutamate receptor 4: A potential approach to Parkinson’s disease treatment. Proc. Natl. Acad. Sci. U.S.A. 100:13668-13673.
Digan, M.E., Pou, C., Niu, H., and Zhang, J.H. 2005. Evaluation of division-arrested cells for cell-based high-throughput screening and profiling. J. Biomol. Screen. 10:615-623.
May, L.T., Leach, K., Sexton, P.M., and Christopoulos, A. 2007. Allosteric modulation of G proteincoupled receptors. Annu. Rev. Pharmacol. Toxicol. 47:1-51.
Di Virgilio, F., Fasolato, C., and Steinberg, T.H. 1988. Inhibitors of membrane transport system for organic anions block fura-2 excretion from PC12 and N2A cells. Biochem. J. 256:959-963.
Miller, M., Wu, M., Xu, J., Weaver, D., Li, M., and Zhu, M.X. 2011. High-throughput screening of TRPC channel ligands using cell-based assays. In TRP Channels (Methods in Signal Transduction Series, M.X. Zhu, ed.) pp. 2-18. Taylor and Francis, Boca Raton, Fla.
Drake, M.T., Violin, J.D., Whalen, E.J., Wisler, J.W., Shenoy, S.K., and Lefkowitz, R.J. 2008. Beta-arrestin-biased agonism at the beta2adrenergic receptor. J. Biol. Chem. 283:56695676. Evans, B.A., Broxton, N., Merlin, J., Sato, M., Hutchinson, D.S., Christopoulos, A., and Summers, R.J. 2011. Quantification of functional selectivity at the human α(1A)-adrenoceptor. Mol. Pharmacol. 79:298-307. Gasparini, F., Bilbe, G., Gomez-Mancilla, B., and Spooren, W. 2008. mGluR5 antagonists: Discovery, characterization and drug development. Curr. Opin. Drug Discov. Devel. 11:655-665. Hoare, S.R. 2007. Allosteric modulators of class B G-protein-coupled receptors. Curr. Neuropharmacol. 5:168-179. Im, W.B., Chio, C.L., Alberts, G.L., and Dinh, D.M. 2003. Positive allosteric modulator of the human 5-HT2C receptor. Mol. Pharmacol. 64:78-84. Jacoby, E., Bouhelal, R., Gerspacher, M., and Seuwen, K. 2006. The 7 TM G-protein-coupled receptor target family. ChemMedChem 8:761782. Kost, T.A., Condreay, J.P., Ames, R.S., Rees, S., and Romanos, M.A. 2007. Implementation of BacMam virus gene delivery technology in a drug discovery setting. Drug Discov. Today 12:396-403. Kunapuli, P., Zheng, W., Weber, M., Solly, K., Mull, R., Platchek, M., Cong, M., Zhong, Z., and Strulovici, B. 2005. Application of division arrest technology to cell-based HTS: Comparison with frozen and fresh cells. Assay Drug Dev. Technol. 3:17-26. Leach, K., Sexton, P.M., and Christopoulos, A. 2007. Allosteric GPCR modulators: Taking advantage of permissive receptor pharmacology. Trends Pharmacol. Sci. 28:382-389. Lindsley, C.W. and Emmitte, K.A. 2009. Recent progress in the discovery and development of negative allosteric modulators of mGluR5. Curr. Opin. Drug Discov. Devel. 12:446-457. Marino, M.J. and Conn, P.J. 2006. Glutamatebased therapeutic approaches: Allosteric mod-
Motulsky, H. and Christopoulos, A. 2004. Fitting Models to Biological Data Using Linear and Non-Linear Regression: A Practical Guide to Curve Fitting. Oxford University Press, New York. Niswender, C.M., Johnson, K.A., Weaver, C.D., Jones, C.K., Xiang, Z., Luo, Q., Rodriguez, A.L., Marlo, J.E., de Paulis, T., Thompson, A.D., Days, E.L., Nalywajko, T., Austin, C.A., Williams, M.B., Ayala, J.E., Williams, R., Lindsley, C.W., and Conn, P.J. 2008. Discovery, characterization, and antiparkinsonian effect of novel positive allosteric modulators of metabotropic glutamate receptor 4. Mol. Pharmacol. 74:1345-1358. O’Brien, J.A., Lemaire, W., Chen, T.B., Chang, R.S., Jacobson, M.A., Ha, S.N., Lindsley, C.W., Schaffhauser, H.J., Sur, C., Pettibone, D.J., Conn, P.J., and Williams, D.L. Jr. 2003. A family of highly selective allosteric modulators of the metabotropic glutamate receptor subtype 5. Mol. Pharmacol. 64:731-740. Offermanns, S. and Simon, M.I. 1995. G alpha 15 and G alpha 16 couple a wide variety of receptors to phospholipase C. J. Biol. Chem. 270:1517515180. Phelan, M.C. 2006. Techniques for mammalian cell tissue culture. Curr. Protoc. Mol. Biol. 74:A.3F.1-A.3F.18. Posner, B.A., Xi, H., and Mills, J.E. 2009. Enhanced HTS hit selection via a local hit rate analysis. J. Chem. Inf. Model. 49:2202-2210. Rees, S., Morrow, D., and Kenakin, T. 2008. GPCR drug discovery through the exploitation of allosteric drug binding sites. Receptors Channels 8:261-268. Rodriguez, A.L., Nong, Y., Sekaran, N.K., Alagille, D., Tamagnan, G.D., and Conn, P.J. 2005. A close structural analog of 2-methyl6-(phenylethynyl)-pyridine acts as a neutral allosteric site ligand on metabotropic glutamate receptor subtype 5 and blocks the effects of multiple allosteric modulators. Mol. Pharmacol. 68:1793-1802.
Triple-Addition Protocols for 7-TM Receptors
139 Current Protocols in Chemical Biology
Volume 3
Rodriguez, A.L., Grier, M.D., Jones, C.K., Herman, E.J., Kane, A.S., Smith, R.L., Williams, R., Zhou, Y., Marlo, J.E., Days, E.L., Blatt, T.N., Jadhav, S., Menon, U.N., Vinson, P.N., Rook, J.M., Stauffer, S.R., Niswender, C.M., Lindsley, C.W., Weaver, C.D., and Conn, P.J. 2010. Discovery of novel allosteric modulators of metabotropic glutamate receptor subtype 5 reveals chemical and functional diversity and in vivo activity in rat behavioral models of anxiolytic and antipsychotic activity. Mol. Pharmacol. 78:1105-1123. Sharma, S., Rodriguez, A.L., Conn, P.J., and Lindsley, C.W. 2008. Synthesis and SAR of a mGluR5 allosteric partial antagonist lead: Unexpected modulation of pharmacology with slight structural modifications to a 5-(phenylethynyl)pyrimidine scaffold. Bioorg. Med. Chem. Lett. 18:4098-4101. Sharma, S., Kedrowski, J., Rook, J.M., Smith, R.L., Jones, C.K., Rodriguez, A.L., Conn, P.J., and Lindsley, C.W. 2009. Discovery of molecular switches that modulate modes of metabotropic glutamate receptor subtype 5 (mGlu5) pharmacology in vitro and in vivo within a series of functionalized, regioisomeric 2- and 5-(phenylethynyl)pyrimidines. J. Med. Chem. 52:4103-4106. Stables, J., Green, A., Marshall, F., Fraser, N., Knight, E., Sautel, M., Milligan, G., Lee, M., and Rees, S. 1997. A bioluminescent assay for
agonist activity at potentially any G-proteincoupled receptor. Anal. Biochem. 252:115-126. Shoichet, B.K. 2006. Screening in a spirit haunted world. Drug Discov. Today 11:607-615. Urwyler, S. 2011. Allosteric modulation of family C G-protein-coupled receptors: From molecular insights to therapeutic perspectives. Pharmacol. Rev. 63:59-126. Vaidehi, N. and Kenakin, T. 2010. The role of conformational ensembles of seven transmembrane receptors in functional selectivity. Curr. Opin. Pharmacol. 10:775-781. Wang, L., Martin, B., Brenneman, R., Luttrell, L.M., and Maudsley, S. 2009. Allosteric modulators of g protein-coupled receptors: Future therapeutics for complex physiological disorders. J. Pharmacol. Exp. Ther. 331:340-348. 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.
Internet Resources http://www.ncgc.nih.gov/guidance/manual toc. html NIH Chemical Genomics Center Assay Guidance Manual. This manual provides an overview of a wide variety of high-throughput screening methods.
Triple-Addition Protocols for 7-TM Receptors
140 Volume 3
Current Protocols in Chemical Biology
Compound Management: Guidelines for Compound Storage, Provision, and Quality Control Sue Holland-Crimmin,1 Paul Gosnell,1 and Chad Quinn1 1
Platform Technology and Sciences, GlaxoSmithKline, Collegeville, Pennsylvania
ABSTRACT The scientific discipline of compound management has developed significantly over the last decade, as witnessed by the large number of conferences dedicated to this topic. The key elements of compound management include (1) the management, storage, and processing of both solids and liquids; (2) compound delivery and interface with key customers; (3) performance of instruments and automation that support these operations; (4) analytical techniques used for quality assurance; and (5) sample informatics, including registration, routing, and compound quality data. This article incorporates guidelines, best practices, and experimental protocols for these key aspects C 2011 by John Wiley & Sons, of compound management. Curr. Protoc. Chem. Biol. 3:141-152 Inc. Keywords: adaptive focused acoustics r AFA r compound management r CM r high throughput screening r HTS r quality assurance r QA r volatile solvent transfer r VST
INTRODUCTION When starting a compound management operation, it is crucial to focus attention on the basic functions that need to be supported while scaling operations and investments to meet the size of the compound collection as well as the throughput and demand capacity. A compound management operation in its most basic description, regardless of the size of the collection or the size of the budget, is a logistical operation—a supply chain. As a supply chain, its primary purpose is to accelerate movement of inventory and information (goods and services) from a point of synthesis or acquisition (manufacturer) to a point of use (the screening facility or the investigator at the bench). The basic functions of compound management include: receipt and sorting solids handling liquids handling solubilization mixing (sonication and centrifugation) creation of compound storage containers (plates and/or tubes) plating capabilities for serial dilution and replication barcode technology (labeling and scanning)
plate sealing storage (manual, automated, plates, tubes, vials, room temperature and freezer) compound delivery and pickup. Informatics is a key element of compound management. Most large pharmaceutical organizations have invested significantly in information technology (IT) systems to support their compound management operations, simply because, until recently, there were no suitable external applications. A full review of sample management informatics is beyond the scope of this article. For a startup organization, however, there are a number of relatively inexpensive systems that can be used. Several of these are linked to liquid-handling systems, including Hamilton (http://www.hamiltonrobotics.com/) and Tecan (http://www.tecan.com). A critical element for many operations is barcode tracking of samples. At the simplest level, a lowtemperature refrigerator and a freezer, combined with an inventory and barcode tracking system, and linked to a suitable liquid-handing robot, can suffice. Simple commercial software packages such as Excel and Access can be used to manage compound collection, as can more customized systems, such as Titian Mosaic applications (http://www.titian.co.uk).
Current Protocols in Chemical Biology 3: 141-152, September 2011 Published online September 2011 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/9780470559277.ch110095 C 2011 John Wiley & Sons, Inc. Copyright
Compound Management Guidelines
141 Volume 3
SOLID COMPOUND MANAGEMENT Typically, most sample management groups receive compounds from a variety of different sources including in-house chemists, contract research organizations (CROs), and commercial compound vendors. Many organizations utilize an electronic laboratory notebook (ELN) system, which allows chemists to register the structure with appropriate information. Previously, large pharmaceutical companies tended to develop proprietary software for this task, but now it is more common for organizations to utilize commercially available software, either with or without customization, for example software from CambridgeSoft (http://www.cambridgesoft.com) or IDBS (http://www/idbs.com).
Solid Compound Receipt Storage and Processing
Compound Management Guidelines
The solid receiving station should have adequate benchtop area to sort and inspect the typical daily workload of compounds (inputs). Basic equipment should include a computer with barcode scanner, printer, and barcodelabel printer (if compounds are not arriving pre-bar-coded). If compound processing is to be performed in batch and not as the compounds arrive, consider adding suitable storage for the materials: racks for solids and/or a freezer for solubilized compound submissions. Powder compounds are usually stored in glass vials or polypropylene tubes under ambient conditions, although some compounds are more reactive with water and are stored in desiccated chambers. Storage at a lower temperature such as 4◦ C can extend the shelf life of some compounds, but care must be taken to ensure that when the containers are moved into ambient conditions, they are first allowed to warm in a non-condensing environment to minimize the chance that water will come into contact with the compounds. Solid compounds that require low-temperature storage are not candidates for inclusion in a typical screening deck, as they are prone to significant degradation under conditions used for most highthroughput screening (HTS) operations. Options for the storage of solid materials range from manual and semi-manual (e.g., Kardex systems; http://www.kardex.com), to fully automated bottle stores. For manual-based systems, errors in pick and place can occur even with barcode tracking and an electronic inventory. The decision on which option to select is primarily governed by the library size, throughput requirements, and budget.
In a typical sample workflow, compounds are weighed out for testing in a biological assay or for inclusion in a compound collection. Compound weighing is generally done manually for several reasons. First, until recently, the availability of commercial automated platforms was relatively limited, and second, the proportion of compounds suitable for automated weighing can be limited. In many cases the cost benefit of automated weighing versus a manual operation does not justify the automated option. The typical weighing station should include a ventilated enclosure approved for handling particularly hazardous substances (PHS), a computer with scanner, a balance with shields to weigh compounds accurately to 0.01 mg, a barcode printer to apply labels to output containers, spatulas for compound transfer, and ethanol and wipes for cleaning between transfers. Depending upon the daily throughput expected for weighing, this station may become a high-use workstation. If so, it is important to pay particular attention to the station’s ergonomics. The operator should be provided with ergonomic seating, a workstation that provides sufficient space while placing necessary supplies within easy reach, and adequate leg room so that the operator can remain close to the work and maintain proper posture. Several factors, including air drafts, temperature changes, electrical and electromagnetic interference, magnetism, levelness, vibration, and static electricity will all affect the accuracy of the weighing process. In terms of minimizing static, some simple precautions that can be very helpful are relatively inexpensive to implement: ensuring correct packaging of glassware and the use of antistatic gloves. More costly solutions include antistatic bars.
Volatile Solvent Transfer (VST) In order to overcome some of the problems associated with dispensing non-powderlike compounds, some organizations utilize a process called volatile solvent transfer or VST. This process originated from combinatorial chemistry and utilizes liquid dispensing for non-weighable solids such as films, gums, or oils. The process involves addition of a known volume of a solvent and mixing to ensure dissolution of the compound, followed by transfer of an aliquot of the solution into a vial or plate. The solvent is then removed via a process of evaporation. There are several methods that can be employed to remove solvent. Vacuum centrifugal concentration is one of the preferred and practical approaches. The
142 Volume 3
Current Protocols in Chemical Biology
combination of vacuum and heat results in solvent evaporation. In principle, centrifugation minimizes compound “bumping” or splashing to adjacent vessels. The application of a vacuum reduces the pressure in the chamber and lowers the boiling point of the solvent. Heat is applied to the samples to counteract evaporative cooling of the solvent and speeds the evaporation process. In the solvent evaporation process, it is essential to prevent compound overheating to avoid degradation of the sample. Details of conditions for typical removal of DMSO are listed below.
Experimental Conditions for Removal of DMSO by Vacuum Centrifugation To remove dimethylsulfoxide (DMSO) using a ThermoSavant Explorer SpeedVac with a sample volume of 4 ml per vial, heat time is 2.5 hr with the chamber temperature at 75◦ C and the block temperature at 45◦ C. The vacuum is set at maximum (<0.7 torr) and the ramp is set to full (Matson et al., 2009).
COMPOUND SOLUBILIZATION Following synthesis and purification, compound solubilization is probably the most critical step in establishing and maintaining sample quality in a collection. In the past, large collections of small molecules were routinely produced by combinatorial methods (Geysen et al., 2003). These processes produced large arrays of compounds in rapid succession, but this rapid expansion of the library came at a cost. In studies conducted at Pfizer (Lipinski, 2004), it was determined that as much as 40% of a 70,000-member collection screened at their site in Groton had poor aqueous solubility. Half of the problem was due to molecular size and lipophilicity, while the remaining half was due to crystal packing (compounds that were low-energy crystalline polymorphs, as opposed to high-energy amorphous forms). It is in this latter group that the design and validation of mixing methods become most critical. To put it simply, the greater the energy introduced into the sample during solubilization, the greater the success in achieving the target concentration.
Solubilization and Mixing A number of procedures are used routinely in laboratories for solubilization and mixing of compounds. These include heat treatment, mechanical agitation such as vortexing and orbital shaking, sonication, and ultrasonics. Compound solubility is governed by two fac-
tors: thermodynamics, which dictates the maximum concentration of the solution, and kinetics, which is how quickly a compound can attain that concentration. Techniques to enhance solubilization can speed the process of dissolution (the kinetics) but cannot affect the thermodynamics unless the compound is changed to a high-energy, more soluble form. Techniques such as sonication have been shown to drive precipitated compounds back into solution. Compound precipitation and degradation occur with DMSO stock solutions following the introduction of water. Freeze–thawing of these hydrated DMSO solutions can exacerbate precipitation. The introduction of miniaturized formats for HTS, such as 384- and 1536-well microplates, was a significant breakthrough for drug discovery but brought up several technical issues in terms of mixing the contents of the compound and assay plates. Conventional mixing techniques were insufficient to overcome the reduced surface area to volume ratio (and thus increased surface tension and capillary forces) and ensure adequate mixing. Several technologies have emerged to support solubilization and mixing of compounds in low-volume miniaturized formats. These include the SonicMan (Matrical Inc.; http:// www.matrical.com), the Hendrix Lateral Ultrasonic Thrust (LUT) TM technology (http://www.microsonix.com), and Adaptive Focused Acoustics (AFA; Covaris; http://www.covarisinc.com). The solubilization of compounds in large volumes in vials (e.g., 1 ml) can be achieved using mechanical agitation (e.g., vortexing). However, the use of acoustic mixing via the Covaris AFA system speeds dissolution significantly. Figure 1 illustrates the dissolution of chloroquine in DMSO subjected to AFA versus vortexing or sonication. For compounds such as chloroquine that are difficult to solubilize, AFA significantly shortens dissolution time.
SOLUTION LIBRARIES AND DMSO For several decades, dimethylsulfoxide (DMSO) has been the primary solvent used for dissolution of small molecule collections for screening. As a solvent, DMSO has been demonstrated to be very reliable for a wide range of compounds, and is considered to be a universal solvent. Compound stocks for screening are typically stored in analytical-grade DMSO (e.g., Sigma-Aldrich 34869-4X4L) at a stock concentration of 5 to 20 mM. Many organizations store samples at
Compound Management Guidelines
143 Current Protocols in Chemical Biology
Volume 3
Concentration (mM)
2.40 2.10 1.80 1.50 1.20
2.10
0.90 0.38
0.60 0.30 0.00
0.26
0.92
1.27
60 AFA
80 AFA
0.11
0.50 20 AFA
40 AFA
100 AFA
100 Vortex
100 Sonication
Time (sec)
Figure 1 Comparison of dissolution of chloroquine at 10 mM nominal concentration by vortex treatment, sonication, and AFA (Nixon et al., 2009).
multiple concentrations for different purposes—typically at 10 mM for doseresponse studies and 1 mM for primary HTS. Recently, there has been a trend of storing small-molecular-weight drug fragments at higher concentrations ≥50 mM. Maintenance of DMSO-based collections requires an understanding of the properties of DMSO and how to manage some of the more common issues that occur with its use: freezing, water uptake, and sample evaporation.
Freezing Point Storing samples frozen is very important to maintaining the quality of the collection, as small molecules are generally more stable when stored at lower temperatures. In its anhydrous form, DMSO has a freezing point of 18◦ C. However, the introduction of water rapidly decreases the freezing point such that a 72%/28% DMSO/H2 O solution will not freeze at −20◦ C, the working temperature for a standard laboratory freezer. As water uptake continues, the binding of water to the DMSO molecules suppresses freezing to a point such that neither the DMSO nor water will freeze solid (Fig. 2; Rasmussen and MacKenzie, 1968.)
Water Uptake and Sample Evaporation
Compound Management Guidelines
Probably the single most important issue for storage of a DMSO-based collection is control of water uptake. In its pure anhydrous form, DMSO is extremely hydrophilic. Early studies conducted at Sphinx Pharmaceuticals in 1998 (Nie et al., 1996) demonstrated that
DMSO samples rapidly absorbed water, even when visually they appear frozen (Fig. 3). It was demonstrated that samples stored under average room temperatures and lowhumidity conditions (24◦ C, 10% relative humidity) could rapidly evaporate (Fig. 4; Nie et al., 1996). However, these studies also showed that standard adhesive and heatsealing plate seals are an effective deterrent to water uptake. Thus, with little to no engineering, a small-scale, plate-based compound collection can be easily maintained by simply limiting compound exposure time during preparation/use and by sealing plates during storage.
Freeze-Thaw Cycles Freeze-thawing of samples under conditions that prevent water uptake has been shown to have little effect on the integrity of the sample (Cheng et al., 2003; Blaxill et al., 2009), particularly at ≤10 mM concentration. Figure 5 shows the results from a study in which 50 compounds held in septum-sealed tubes were subjected to multiple freeze-thaw cycles at three different locations in identical tube stores. Ten compounds were identical and 40 were different across the locations. The samples subjected to freeze-thawing were compared to control tubes. The results shown on Figure 5 are the average values of purity, concentration, and water content of the samples. The data indicate no significant change in purity, concentration, or water content correlating with numbers of freeze-thaw cycles. Furthermore, for two compounds that did degrade over the course of the study, the level of degradation was identical between the control
144 Volume 3
Current Protocols in Chemical Biology
40
20
liquid
Temperature (C)
0
20
liquid + DMSO2.5H2O
ice + liquid
liquid + DMSO3H2O
40
60
liquid + DMSO 1 H2O 2
80
Figure 2
DMSO2.5H2O + DMSO 1 H2O 2
ice + DMSO3H2O DMSO3H2O + DMSO 2.5H2O
100
120
liquid + DMSO (solid)
0
10
20
30
40 50 60 Weight percent DMSO
70
80
DMSO 1H O 2 2 + DMSO (solid) 90
100
Phase diagram for water-dimethylsulfoxide system.
210 200 190 sealed lid stacked open
Weight (mg)
180 170 160 150 140 130 120 110 100 0
10
20
30 40 50 Time (days)
60
70
80
Figure 3 Sample weight change of 5-μl samples stored in 384-well plates at −20◦ C under different conditions (water absorption and evaporation).
samples and the samples experiencing multiple freeze-thaw cycles.
Storage and Maintenance of Solutions Stability studies of compounds dissolved in DMSO and maintained under various storage conditions have been published which show that, for a typical collection used in drug discovery, room temperature storage reduces the shelf life of a significant percentage of the
compounds to just a few months (MacArthur et al., 2009). Storage at −20◦ C will extend the shelf life to a decade or longer, provided that water uptake is prevented (Blaxill et al., 2009). It has also been shown that exposure to oxygen is far less of a concern than exposure to water, so all reasonable means of controlling the moisture when working with the solvated compounds during storage and preparation should be employed, but storing under nitrogen or argon is usually not required.
Compound Management Guidelines
145 Current Protocols in Chemical Biology
Volume 3
110 Weight (% of initial weight)
100 90 80 70 60 50 40
sealed lid stacked open
30 20 10 0 10
0
20
30
40 50 Time (days)
60
70
80
99 98 97 96 95 94 93 92 91 90
20 18 16 14 12 10 8 6 4 2 0 D1
D2
D5
D10 D15 D20 D25 Dispense (F/T cycle)
concentration (mM)
water content (%)
D30
D35
Purity (%)
Water content (%) Concentration (mM)
Figure 4 Sample weight of 5-μl samples stored in 384-well plates at 24◦ C, 10% relative humidity, under different conditions (water absorption and evaporation).
D40
purity (%)
Figure 5 Analysis of the purity, concentration and water content of a range of samples after different numbers of freeze-thaw cycles.
Compound Management Guidelines
It is generally accepted that storage in a well controlled, low-temperature environment will prolong the shelf life and integrity of solution libraries. Most laboratories, either large or small, prefer to use sealed tubes for longterm storage to facilitate access to individual compounds. Many suppliers of laboratory consumables manufacture high-quality tubes, e.g., Matrix (http://www.matrixtechcorp. com/storage-systems), Abgene (http://www. abgene.com), and Thermo Scientific (http:// www.thermoscientific.com). Such tubes have a dot-matrix barcode which enables an audit trail and tracking of compounds. Tubes are stored in 96-well racks or in custom trays for automated tube-storage systems. How the tubes are stored depends on the scale of the operation. For small-scale operations, a simple commercial freezer is adequate, and this can be coupled with a relatively inexpensive manual software package (e.g., Titian Software) to guide the user through the
process of manually retrieving items required by orders, and putting them away afterwards. There are a wide range of relatively affordable small to mid-size off-the-shelf automated tube storage systems that can be used as stand-alone instruments with a manual interface (e.g., TTP; http://www.ttplabtech.com), or, alternatively, these can be linked to a liquid-handling platform (e.g., Hamilton). For larger-scale systems, more suitable for storage of collections of >500,000 compounds, multiple vendors offer systems, including RTS Life Sciences (http://www.rts-group.com), TAP (http://www. tapbiosystems.com), and Matrical.
Solution Storage for Small-Scale Operations A simple and inexpensive approach is required for laboratories with smaller compound collections in which throughput is low and, consequently, significant automation is not
146 Volume 3
Current Protocols in Chemical Biology
Scatter plot
100 75
Stamped Nov. 2007
50 25 0 25 50 75 100 100
75
50
25
0
25
50
75
100
Stamped May 2008
Figure 6 Correlation of activity in an ion channel FLIPR screen: comparison of plates freshly prepared as compared to those stored at −20◦ C sealed for 6 months.
required. In these cases, laboratories should consider tube-based storage for long-term storage at −20◦ C with a barcode and manual inventory tracking system. The tubes can be used to provide material for IC50 determinations. In order to facilitate compound plating for primary screening, a plate-based storage approach is practical. It is advisable to make several copies of the library from master stock plates which are maintained at −20◦ C. One copy can be considered as a working stock, used for 3 to 6 months, and then discarded and replaced with a fresh copy. This allows maintenance of the working stock at a higher temperature—some laboratories hold the working stock sealed at room temperature for several months. Data generated on the stability of DMSO solutions at room temperature have indicated that the probability of observing a compound was 92% after 3 months of storage at room temperature, 83% after 6 months, and 52% after 1 year (Kozikowski et al., 2003).
Storage of Assay-Ready Compound Plates In many organizations, assay-ready compound plates are prepared ahead of time and stored prior to screening. This is because “Just in time” coupling of compound preparation and screening can produce a bottle-
neck in high-throughput operations that screen >500,000 assay points per day. The most common use of assay-ready plates is as single-shot concentration plates for primary HTS in which the compounds are added to the assay plate at a concentration typically 10-fold below the master stock concentration, e.g., 1 mM. The plate is sealed and stored at low temperature and then, at the time of assay, the plates are defrosted and unsealed prior to the addition of the screening reagents. Some organizations create assay-ready plates for dose-response studies, but the requirement for this is less frequent due to the lower throughput. In this case, the storage times tend to be less. Figure 6 shows the correlation of samples for an ion channel FLIPR assay prepared freshly as compared to a set stored for 6 months at −20◦ C. There is excellent correlation between the freshly prepared samples and the plates stored sealed at −20◦ C for 6 months. Table 1 provides general guidelines for appropriate assay-ready plate storage. Sealing is key for long-term storage, and heat sealing is preferable, as it avoids issues with adhesive residues. In order of preference, heat sealing is better than adhesive, which is better than foil. Different plate types will need different conditions for heat sealing, depending on plate composition (polypropylene versus polystyrene plates) and capacity (high-volume, >500 nl, versus low-volume,
Compound Management Guidelines
147 Current Protocols in Chemical Biology
Volume 3
Table 1 Best Practices Guidelines for Storage of Assay-Ready Plates
Sealing condition
Room temperature
With seal
24 hr for low volume 1 work week for high volume only
4◦ C
With lid or stacked
1 work week
With seal
2 months
With lid or stacked
2 months
With seal
6 months
◦
−20 C
<500 nl). The Velocity 11 Thermo Sealer provides tight control of temperature and has an adapter for different plate types. Seal removal is commonly a manual process, but a recent development by Nexus offers an automated de-sealer. It is recommended that all assay-ready compound storage plates be spun for 1 min at 140 × g prior to unsealing to gather the contents at the bottoms of the wells. Centrifugation of compound plates can be critical in 1536-well plate assays where reagent addition volumes are low (e.g., <5 μl).
Sample Dispensing
Compound Management Guidelines
Maximum stay (discard plates after this point)
Temperature
Depending upon the daily expectations for throughput and capacity, as well as assay formats that need to be supported, the liquidhandling functions for compound management can be managed by as few as one liquidhandling robot that is very generic in its implementation or by several robots that are very specialized for specific tasks. Solubilization normally occurs in vials, making the standard, variable-span 8-tipped robotic platform well suited to handle this task. For preparation of plate copies, liquid handlers with 96- or 384well capabilities are deployed. Liquid-handling devices can be classified into two categories: either contact or noncontact. In contact dispensing, the tip touches the target surface (e.g., the bottom of the microplate or vial) and includes air displacement, positive displacement, and pin tool devices. Non-contact dispensers include peristaltic, thermal inkjet, syringe-solenoid, and the more recent acoustic dispensing devices offered by Labcyte (http://www.labcyte.com) and EDC (http://www.edcbiosystems.com). In order to accurately dispense samples, the acoustic process requires the derivation of parameters about the sample, which themselves provide valuable quality information, including the hydration level of samples and volume monitoring within the well. Compound-
management organizations have exploited this information to improve the quality and integrity of their processes. For example, the hydration levels of compound source plates can be routinely tracked to gauge when plates should be retired.
THE IMPORTANCE OF QUALITY The primary work of compound management in most organizations is in smallmolecule discovery spanning gene to candidate selection. All aspects of quality are critical to achieving a successful gene-to-candidate process. As a starting point, the structural quality of the compound collection dictates the sort of leads identified. Compound diversity geared to lead-like properties provides excellent starting points for new therapeutic programs. Once the molecules of choice are assembled in a collection, maintaining and ensuring their analytical quality is a key requirement and, as the targets are advanced through to candidate selection, running that process to avoid errors is critical. All three aspects are critical to the delivery of high-quality biological results. In terms of process, one cannot inspect quality into a process. Inspection only detects defects and enables those defects to be removed. It is better to design a high-quality process than to fix the errors downstream. Process quality can be developed using simple approaches, a concept pioneered by the Japanese called Poka–yoke, or mistake-proofing. For example, in a compound-management laboratory, a simple process such as using a corner cut-out of a plate can ascertain correct orientation and prevent compounds from being misdispensed. Quality approaches can be as simple as using a control series of tubes with fixed volumes to compare by eye with dispensed tubes to check volume accuracy, or more sophisticated, such as the use of tube auditors which measure volume and precipitate with camera systems.
148 Volume 3
Current Protocols in Chemical Biology
needs to be invested in setting up the protocols and the data handling and analysis packages to monitor and interpret the data (Taylor et al., 2002). A commercial alternative is available from Artel MVS (http://www.artel-usa.com), which uses a dual-dye, dual-wavelength volume measurement system based on ratiometric photometry. The technology relies on two colorimetric dyes with distinct absorbance maxima at 520 nm (red dye) and 730 nm (blue dye). The system is used to quantify volume from each dispensing channel by measuring the absorbance of these dyes. If using fixed-tip instruments to transfer compounds, it is important to monitor the potential for compound carryover between transfers. Transfer methods must include wash steps between source plates to eliminate or minimize the percentage of carryover from tips between compound plates. QA carryover studies should be performed regularly on liquid handlers with a goal of minimizing carryover while using the smallest volumes of wash reagents possible. Typically, in setting up instruments, carryover targets of ≤0.01% are sought using dyes with a range of physicochemical properties to mimic a broad range of compounds in a collection (see Table 3 and Sanfiz-Pinto et al., 2009).
Liquid-Handling Quality Control The correct performance of liquid-handling equipment is a critical process in a compoundmanagement operation. The consequences of poor performance and liquid-handling errors can be severe, ranging from failed experiments to releasing inaccurate results. A good standard practice is to establish the performance capabilities of instruments, and then to monitor that performance to establish any deviation from the expected control limits. For most instruments, this will include precision and accuracy measurements, an acceptable number of outliers, fold dilution, and, for fixed-tip instruments, carryover. The thresholds need to be both low enough to be acceptable from a biological potency variability perspective and high enough to allow the instruments to pass the quality assurance (QA) analysis on a regular basis. Table 2 indicates typical generic quality criteria for liquid-handling instruments based on the volume and the type of dispensements performed. Many laboratories establish their own protocols and standard operating procedures (SOPs), based on single dyes, to measure absorbance or fluorescence. Once established, such protocols tend to be easy to implement and low in cost; however, considerable effort
Table 2 Quality Criteria for Liquid Handling Instruments
Accuracy
≤5% for >5 μl dispense ≤10% for <5 μl dispense
Precision
≤20% CV for system-liquid instruments (e.g., Tecan) ≤10% CV for air-displacement instruments (e.g., Biomek FX)
Outliers
Serial dilution: <5% random and <2 per series Single shot: <4 random, 0 systematic
Fold dilution
Fold dilution ± 10%
Carryover
≤0.01%
Tip dilution
≤5%
Table 3 Comparison of Dye Properties for Carryover Studiesa
Lipinski
Tartrazine RhGr110 DDAF
No. H-bond donors
≤5
1
3
2
No. H-bond acceptors
≤10
12
5
7
Mol. wt.
<500
534.35
366.8
527.6
<5
−2.49
2.922
5.5
clogP
a The three dyes are described according to the four Lipinski attributes, i.e.,
hydrogen bond donor and acceptor, mol. wt., and clogP.
Compound Management Guidelines
149 Current Protocols in Chemical Biology
Volume 3
dye solution
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
D?
DMSO
96-tube rack a wash cycle is performed after every dispensing cycle
DMSO
Figure 7
Schematic for performing a carryover study on a Tecan Genesis.
Experiments should be designed to determine the number of wash cycles and the volume of wash reagents sufficient to attain a target of ≤0.01% carryover, thereby eliminating carryover as a concern for all compounds with an IC50 of ≥1 nM. Figure 7 lists a schematic for the design of a carryover test from compounds held in a 96-tube rack to mimic a cherry pick process from tubes using a Tecan 8-tip span Genesis. A 96-tube rack is filled with 1 ml of dye and DMSO in alternating columns. The procedure involves aspirating 100 μl of solution from each of the tubes in the tube rack and dispensing into a Nunc 96well clear flat-bottom UV-readable plate. The volume of wash fluid is varied to determine a minimum carryover. The Tecan has two wash stations or reservoirs, and, for most scenarios, 1.8 ml of DMSO per tip from wash station 1 and 1 ml from wash reservoir 2 (cleaner) are appropriate conditions. A standard curve is prepared using a range of concentrations of dye. This is used for the calculation of carryover after washing the Tecan tips in DMSO. The optical density of each well on this plate is then measured using a UV reader (e.g., Tecan SpectraFluor).
Analyzing Compound Quality
Compound Management Guidelines
One of the key requirements of a compound management operation is having access to analytical systems to assess compound purity and confirm compound integrity. Analytical evaluation needs to be applied as compounds enter the collection and as compounds are processed through and delivered to biological assays. Many large pharmaceutical organizations have invested significant resources to clean up their compound collections by analyzing their whole collections (Lane et al., 2006).
Typically, this has taken the form of LCMS analysis of solution libraries. In general, if resources are available, analysis should be performed on all compounds that enter a collection. Pragmatically, analysis is usually performed only at key points of the screening process, to validate hits following HTS or during important stages of lead optimization. Guidelines for the QA criteria for entry into a screening collection are typically “confirmed structural identity (typically by mol. wt.) and a minimum relative purity of 80%.” For purity failures, if purity is less than the target level, then screening wells can be deleted electronically and all master stocks discarded. This is because master stocks are typically individual tubes and are therefore easy to discard, whereas screening wells are distributed from pre-dispensed plates. For solids, if purity is 50% to 80%, the solids can be retained and annotated. If purity is <50%, then the solids should be discarded.
Integrity and purity measurement by liquid chromatography and mass spectrometry When choosing an analytical technique to support the large-volume demands of compound management, consideration needs to be given to the sensitivity, reproducibility, speed, cost, ability to automate, and facile use of the instrument. Also, the technique needs to be able to unambiguously evaluate compound purity and integrity. Compound data need to be validated and reported in a timely manner, and stored in a database for future access and data mining. Due to the separation power of liquid chromatography (LC) combined with the detector selectivity of mass spectrometry
150 Volume 3
Current Protocols in Chemical Biology
(MS), LC-MS is the recommended analytical technique. All compounds entering a storage system might undergo QA testing to confirm identity and determine relative purity. Analysis by LC-MS with UV diode array detection is used to determine relative purity. MS is used to confirm compound molecular weight and integrity of the compound, to confirm that it has not degraded. Evaporative light scattering detection (ELSD) is an additional parallel method of detection that can be coupled to the LC-MS system and be used in the absence of a chromophore to determine purity of the compound.
Reversed-phase HPLC LC is a technique for separating a sample into various fractions and then monitoring the contents of the fractions. The majority of compounds in a collection are either chemically or thermally labile and usually require LC to determine the purity of the sample. Reversedphase HPLC uses a nonpolar stationary phase and an aqueous mobile phase, and is the most common form of chromatography used in LCMS applications. To elute the compounds, a solvent with greater solvent strength, such as methanol or acetonitrile, is then introduced at increasing concentrations to displace the analyte from the stationary phase and elute it from the column. This type of separation is referred to as a gradient elution. The output of the mobile phase is directed into the ionization source of the mass spectrometer to confirm the molecular weight of the compound.
Mass spectrometry Mass spectrometry can be considered the most powerful detector for chromatography because of its high sensitivity and robustness, and because it provides both qualitative and quantitative information about compounds eluted from the HPLC system. Mass spectrometers come in a variety of combinations of ion sources and mass analyzers. The most common mass spectrometry system for LC-MS is the electrospray ionization source and a quadrupole mass analyzer. Electrospray ionization most often provides only molecular weight information, with limited structural information, simplifying data interpretation and providing confirmation of a desired compound’s molecular weight. Therefore, this combination is ideal for coupling a chromatographic system such as liquid chromatography to a mass spectrometer.
Use of NMR Traditionally, chemists have used NMR in conjunction with MS to characterize the compounds they synthesize. The NMR data provides information on the structural integrity of the compound, while MS supports the proposed molecular weight. Unfortunately, NMR data analysis is completely manual and becomes a time-consuming activity when working with a large number of samples. Therefore, NMR is used as a follow-up analytical technique in the process to elucidate the structure of compounds identified as incorrect and pure or unconfirmed by LC-MS. The NMR analysis can then help to correct structural errors in the compound registry while limiting the proliferation of discrepant data. Also, NMR techniques can be used to provide additional information such as the concentration of the compound in solution and/or the water content of the deuterated DMSO solutions. But again, these are time-consuming analyses and would not necessarily work in an automated highthroughput process to support the volume of samples from compound management.
CONCLUSION The past decade in compound management has been a period of development and growth. As an integral part of numerous discovery organizations, compound management has demonstrated itself to be essential to the success of those strategies and efforts. The key to this success has been focusing on the compound users and scaling the operation to meet their requirements. Compound management has emerged and established itself as a scientific logistical service, a supply chain that accelerates the flow of data and material to the scientist that is served. Sample management needs to be flexible to adapt to new needs. The compound management team today has a distinct advantage over teams a decade ago. A wealth of data about processes and best practices has been well documented, as presented within this review, to be followed and improved.
LITERATURE CITED Blaxill, Z., Holland-Crimmin, S., and Lifely, R. 2009. Stability through the ages: The GSK experience. J. Biomol. Screen. 14:547. Cheng, X., Hochlowski, J., Tang, H., Hepp, D., Beckner, C., Kantor, S., and Schmitt, R. 2003. Studies on repository compound stability in DMSO under various conditions. J. Biomol. Screen. 8:292-304.
Compound Management Guidelines
151 Current Protocols in Chemical Biology
Volume 3
Geysen, M.H., Schoenen, F., Wagner, D., and Wagner, R. 2003. Combinatorial compound libraries for drug discovery an ongoing challenge. Nat. Rev. Drug Discov. 2:222230. Kozikowski, B.A., Tirey, D.A., Williams, L.E., Kuzmak, B., Stanton, D.T., Morand, K.L., and Nelson, S.L. 2003. The effect of roomtemperature storage on the stability of compounds in DMSO. J. Biomol. Screen. 8:205209. Lane, S.J., Eggleston, D.S., Brinded, K.A., Hollerton, J.C., Taylor, N.L., and Readshaw, S.A. 2006. Defining and maintaining a high quality screening collection: the GSK experience. Drug Discov. Today 11:267-272. Lipinski, C. 2004. Compound solubility and HTS screening. Laboratory Robotics Interest Group (LRIG) Mid Atlantic Meeting, January 2004. LRIG, Martinsville, N.J. MacArthur, R., Leister, W., Veith, H., Shinn, P, Southall, N., Austin, C.P., Inglese, J., and Auld, D.S. 2009. Monitoring compound integrity with cytochrome P450 assays and qHTS. J. Biomol. Screen. 14:538-546. Matson, S.L., Moneesh, C., Stock, D.A., Leet, J.E., Dumas, E.A., Ferrante, C.D., Monahan, W.E., Cook, L.S., Watson, J., Cloutier, N.J., Ferrante, M.A., Houston, J.G., and Banks, M.N. 2009.
Best practices in compound management for preserving compound integrity and accurately providing samples for assays. J. Biomol. Screen. 14:476-484. Nie, D., Hilton, A., and Gosnell, P. 1996. DMSO in HTS: The effect of water absorption and evaporation. Society for Biomolecular Screening, 3rd Annual Conference, Baltimore, Md. Nixon, E., Holland-Crimmin, S., Lupotsky, B., Chan, J., Curtis, J., Dobbs, K., and Blaxill, Z. 2009. Applications of adaptive focused acoustics to compound management. J. Biomol. Screen. 14:460-467. Rasmussen, D.H. and MacKenzie, A.D. 1968. Phase diagram for the system waterdimethylsulfoxide. Nature 220:1315-1317. Sanfiz Pinto, B., Gannon, B., and Holland, S. 2009. Determining optimal minimum carryover with fix tipped liquid handlers. Poster presented at the Society for Biomolecular Screening, Lille, France, April 2009. Taylor, P.B., Ashman, S., Baddeley, S., Bartram, S.L., Battle, C.L., Bond, B.C., Clements, Y.M., Gaul, N.J., McAllister, E.W., Mostacero, J.A., Ramon, F., Wilson, J.M., Hertzberg, R.P., Pope, A.J., and Macarron, R. 2002. A standard operating procedure for assessing liquid handler performance in high-throughput screening. J. Biomol. Screen. 7:554-569.
Compound Management Guidelines
152 Volume 3
Current Protocols in Chemical Biology
Copper-Catalyzed Azide–Alkyne Click Chemistry for Bioconjugation Stanislav I. Presolski,1 Vu Phong Hong,1 and M.G. Finn1 1
The Scripps Research Institute, La Jolla, California
ABSTRACT The copper-catalyzed azide-alkyne cycloaddition reaction is widely used for the connection of molecular entities of all sizes. A protocol is provided here for the process with biomolecules. Ascorbate is used as reducing agent to maintain the required cuprous oxidation state. Since these convenient conditions produce reactive oxygen species, five equivalents of a copper-binding ligand are used with respect to metal. The ligand both accelerates the reaction and serves as a sacrificial reductant, protecting the biomolecules from oxidation. A procedure is also described for testing the efficiency of the reaction under desired conditions for purposes of optimization, before expensive biological reagents C 2011 by John Wiley & Sons, Inc. are used. Curr. Protoc. Chem. Biol. 3:153-162 Keywords: click chemistry r azides r alkynes r bioconjugation r proteins r nucleic acids r copper
INTRODUCTION A reliable set of procedures for the copper-catalyzed azide-alkyne cycloaddition (CuAAC) reaction (Fig. 1) is described here, based on investigations of the reaction mechanism and optimization using accelerating ligands for bioconjugative applications (Chan et al., 2004; Lewis et al., 2004; Himo et al., 2005; Rodionov et al., 2005; Sen Gupta et al., 2005; Chan and Fokin, 2007; Rodionov et al., 2007a,b; Hong et al., 2008, 2009; Buckley et al., 2010; Presolski et al., 2010). In practical terms, “bioconjugation” is usually used to signify making covalent bonds to biological molecules (principally oligopeptides, proteins, oligonucleotides, nucleic acids, and fatty acids), assemblies of biological molecules (such as lipid bilayers and vesicles, virus particles, and protein aggregates of many kinds), and living systems (usually cells in culture). However, these procedures are also applicable to polymers, having been particularly important for the attachment of polyethylene glycol to many compounds, or to any molecules that are handled in aqueous solvents at low concentrations. High concentrations and nonaqueous reaction conditions require a different catalyst formulation for demanding situations, as has been discussed elsewhere (Presolski et al., 2010). The basic CuAAC process requires only copper ions in the +1 oxidation state. These may be supplied by a discrete CuI complex (Rostovtsev et al., 2002; Tornøe et al., 2002), by metallic copper (Rostovtsev et al., 2002), or by copper-impregnated materials (Lipshutz et al., 2006) that expose cuprous ions to the reaction solution, or, most conveniently, by a mixture of a CuII salt and a reducing agent, sodium ascorbate being by far the most popular (Rostovtsev et al., 2002). The development of accelerating ligands for the reaction is primarily driven by the need to maintain a sufficient concentration of CuI in solution, since copper ions can undergo fast and debilitating redox and disproportionation reactions if not properly bound by chelating ligands. CopperCatalyzed Azide-Alkyne Click Chemistry for Bioconjugation Current Protocols in Chemical Biology 3: 153-162, December 2011 Published online December 2011 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/9780470559277.ch110148 C 2011 John Wiley & Sons, Inc. Copyright
153 Volume 3
HO R1 – N 3
R2
H
OH
N N N N
N Cu
R1
I
ligand
R2
N N
N
N
N
H2N
N N N
H2N
THPTA (1)
NH2 2
N OH
Figure 1 The general CuAAC reaction, and structures of accelerating ligand THPTA (1) and aminoguanidine additive (2).
BASIC PROTOCOL
COPPER-CATALYZED AZIDE-ALKYNE CYCLOADDITION FOR COUPLING OF CARGO-AZIDE TO BIOMOLECULE-ALKYNE The basic procedure below describes the ligation of a functional (“cargo”) azide to a biomolecule-alkyne. It can be used equally well in the reversed sense (biomoleculeazide + cargo-alkyne), and the cargo can also be a biomolecule. Aminoguanidine is recommended when side reactions between dehydroascorbate and protein side chains (principally arginine) are to be suppressed, but is not otherwise helpful and should be omitted, if possible. Tris(3-hydroxypropyltriazolyl-methyl)amine] (THPTA, structure 1, Fig. 1) serves the dual purpose of protecting biomolecules from hydrolysis by Cu(II) byproducts, and sacrificially intercepting the radicals and/or peroxides derived from O2 /Cu/ascorbate that oxidize histidine and other residues. An excess of this ligand does not dramatically slow the reaction, so more than five equivalents can be used, if necessary. Note that other water-soluble versions of the tris(triazolylmethyl)amine motif have been reported (Hong et al., 2008; Soriano del Amo et al., 2010), and are also suitable. NOTE: All of the materials are commercially available from standard suppliers, except for the ligand (THPTA). THPTA can be prepared by the published procedure (Hong et al., 2009) or is available from the authors in small quantities until it becomes commercially available.
Materials Biomolecule-alkyne of interest 100 mM potassium phosphate buffer, pH 7 (see Critical Parameters regarding buffers) 5 mM cargo-azide 20 mM CuSO4 in water 50 mM ligand THPTA (structure 1 in Fig. 1; available in small quantities from the authors,
[email protected]), in water 100 mM sodium ascorbate: prepare fresh just before use by adding 1 ml of water to 20 mg ascorbate 100 mM aminoguanidine hydrochloride (structure 2 in Fig. 1): add 1 ml of water to 11 mg aminoguanidine) CopperCatalyzed Azide-Alkyne Click Chemistry for Bioconjugation
2-ml microcentrifuge tubes End-over-end rotator
154 Volume 3
Current Protocols in Chemical Biology
1. In a 2-ml microcentrifuge tube, combine the following reagents in the order indicated: a. Biomolecule-alkyne + buffer to make 432.5 μl of solution that is 57.8 μM in alkyne. The procedure has been successfully done with 2 μM and higher concentrations of biomolecule-alkyne.
b. 10 μl of 5 mM cargo-azide. The amount of cargo-azide added should be in ∼2-fold excess with respect to alkyne groups on the biomolecule, down to 20 μM (in other words, if the alkyne concentration is very low, more than two equivalents of azide are needed for fast reaction).
c. A premixed solution of 2.5 μl of 20 mM CuSO4 (2.5 μl) and 5.0 μl of 50 mM ligand THPTA (structure 1). Final concentrations in the microcentrifuge tube: CuSO4 , 0.10 mM (note that this can be adjusted as desired between 0.05 and 0.25 mM); ligand 1, 0.50 mM (ligand to copper ratio, 5:1).
d. 25 μl of 100 mM aminoguanidine. Final concentration of aminoguanidine in the tube, 5 mM.
e. 25 μl of 100 mM sodium ascorbate. Final concentration of sodium ascorbate in the tube, 5 mM.
2. Close the tube (to slow the introduction of oxygen), mix by inverting the tube several times, or attach to a slow end-over-end rotator (∼30 rotations per min). Allow the reaction to proceed for 1 hr. Workup will depend on the particular application. Copper ions can be removed by washing or performing dialysis with solutions of buffered ethylenediamine tetraacetic acid (EDTA). The addition of an excess of EDTA relative to Cu also serves to stop the reaction when it is not desirable to simply expose the mixture to air and allow any remaining reducing agent to be used up (thus generating reactive oxygen species in the process, as described above). Copper-adsorbing resins such as Cuprisorb (http://www.seachem.com) are useful in cases of preparative organic synthesis, but tend to bind biomolecules and thus be of lesser value for bioconjugation. In such cases the conjugates are purified directly after the reaction in such a way as to leave small molecules behind.
DETERMINING THE EFFICIENCY OF BIOCONJUGATION CuAAC WITH A FLUOROGENIC PROBE It is often helpful to test the reactivity of a biomolecule-alkyne in a way that allows an easy readout. For this purpose, the fluorogenic coumarin azide of Wang and coworkers is usually employed (structure 3 in Fig. 2; Sivakumar et al., 2004). A convenient assessment can be made of CuAAC efficiency under a particular set of conditions by first reacting 3 with an excess of a small-molecule “model” alkyne such as propargyl alcohol or phenylacetylene to ensure completion of the click reaction. The resulting solution is then diluted to the same concentration at which the biomolecule alkyne will be used. For a better control, also include the biomolecule without its alkyne appendage to provide a good mimic of the bioconjugation environment. The fluorescence of that solution can be used to define 100% reaction. The CuAAC reaction on the desired biomoleculealkyne can then be performed, and the fluorescence intensity used directly to estimate the progress of the reaction. The assumption that the fluorescence wavelength and intensity of triazole 4 (Fig. 2) will not be much changed when attached to the biomolecule is usually a reasonable one, as has been demonstrated previously (Hong et al., 2009). The example here describes a representative procedure for optimizing CuAAC conditions for a protein (“X”) that has been decorated with an aliphatic terminal alkyne group,
SUPPORT PROTOCOL
CopperCatalyzed Azide-Alkyne Click Chemistry for Bioconjugation
155 Current Protocols in Chemical Biology
Volume 3
O
HO HO
O
CuSO4 ligand
O H
N
R
N N
Na ascorbate
N3 3 not fluorescent
O
4
R CH2OH, CH2OR , CH2NHCOR , Ar, etc., including biomolecule
ex ex
404 nm 477 nm
R
Figure 2 CuAAC reaction with Wang’s fluorogenic azide (3), used to investigate reaction conditions before using the expensive biomolecule or cargo reagent(s).
designated “X-alkyne,” to be used at a concentration of 25 μM (1 mg/ml of a 40-kDa protein bearing one alkyne). Compound 3 serves as a surrogate for the cargo-azide, to be used at a concentration of 50 μM (two equivalents with respect to alkyne). The buffer used here should be the same buffer that will be used in the desired conjugation process.
Additional Materials (also see Basic Protocol) The materials used here are identical to those used in Basic Protocol 2, except for the “cargo-azide,” which is compound 3 in Figure 2; this molecule is available from Glen Research (http://www.glenresearch.com, cat. no. 50-2004-92), or it can also be prepared by following the published procedure (Sivakumar et al., 2004); the solid or DMSO stock solution of this compound should be stored in a refrigerator and protected from light Protein X (see above) or surrogate protein, e.g., bovine serum albumin (BSA) Fluorometer Perform the CuAAC reaction 1. In a 2-ml microcentrifuge tube, combine the following reagents in the order indicated: a. Propargyl alcohol + buffer to make 446.2 μl of solution that is 560 μM in alkyne. Propargyl alcohol is used as the model alkyne.
b. 10 μl of 5 mM azide 3 stock solution The final concentration of azide 3 in the tube will be 100 μM.
c. A premixed solution of 6.3 μl of 20 mM CuSO4 and 12.5 μl of 50 mM ligand THPTA (structure 1). The final concentration of copper in the tube will be 0.25 mM and the final concentration of ligand 1 in the tube will be 1.25 mM (ligand to copper ratio, 5:1).
d. 25 μl of 100 mM sodium ascorbate. The final concentration of sodium ascorbate in the tube will be 5 mM.
2. Close the tube (to prevent more oxygen from diffusing in), mix by inverting the tube several times, or attach to an slow end-over-end rotator (∼30 rotations per min). Allow the reaction to proceed for 1 hr.
CopperCatalyzed Azide-Alkyne Click Chemistry for Bioconjugation
3. Dilute the reaction mixture by a factor of 4 with buffer to obtain a solution approximately 25 μM in triazole 4. In this dilution step, include protein X, if possible (or a surrogate protein such as bovine serum albumin), so that the final concentration of protein in the diluted solution is also 25 μM. Read the fluorescence intensity at 477 nm (excitation wavelength 404 nm), which should be substantially greater than that of a 25 μM solution of 3. Routine users of the CuAAC bioconjugation reaction may wish to prepare a supply of triazole 4 using propargyl alcohol and phenylacetylene to act as standards for
156 Volume 3
Current Protocols in Chemical Biology
reactions involving aliphatic and aromatic biomolecule-alkynes, respectively. These pure compounds can then be diluted to the appropriate concentration in the buffer/protein mixtures of interest to establish an approximate expected fluorescence intensity readout of a complete click reaction under desired conditions.
Determine CuAAC efficiency 4. In a 2-ml microcentrifuge tube, combine the following reagents in the order indicated: a. Combine biomolecule-alkyne + buffer to make 437.5 μl of solution that is 28.6 μM in alkyne. b. 5 μl of 5 mM azide 3. The final concentration of azide 3 in the tube will be 50 μM.
c. A premixed solution of 2.5 μl of 20 mM CuSO4 and 5.0 μl of 50 mM ligand THPTA (structure 1). Final concentrations in the microcentrifuge tube: CuSO4 , 0.10 mM (note that this can be adjusted as desired between 50 and 250 μM); ligand 1, 0.50 mM (ligand to copper ratio, 5:1).
d. 25 μl of 100 mM aminoguanidine. The final concentration of aminoguanidine in the tube will be 5 mM.
e. 25 μl of 100 mM sodium ascorbate. The final concentration of sodium ascorbate in the tube will be 5 mM.
5. Close the tube (to prevent more oxygen from diffusing in), mix by inverting the tube several times, or attach to a slow end-over-end rotator (∼30 rotations per min). Allow the reaction to proceed for 1 hr. 6. Read the fluorescence intensity of the mixture at 477 nm (excitation wavelength 404 nm) and compare to the intensity observed for the model reaction as described above. The observation of substantially lower fluorescence suggests the need for adjustment of the reaction conditions.
COMMENTARY Background Information The marriage of azides and alkynes in biological molecules has emerged from an appreciation of the insights made possible by the application of earlier bioorthogonal reactions in chemical biology (Lemieux and Bertozzi, 1998; Prescher and Bertozzi, 2005). A bioorthogonal reaction is performed in the presence of, or with the participation of, biological molecules, in which the molecular components of the native structures do not participate. For example, Bertozzi’s use of the Staudinger ligation to attach carbohydrate molecules to proteins helped popularize the use of azides in this approach, the application being notable for the fact that efficient connections were achieved without engaging the many hydroxyl, carboxylic acid, amine, indole, imidazole, thioether, and even thiol groups present on one or both of the participating partners (Hang and Bertozzi, 2001).
Azides and alkynes are particularly useful because they are small and chemically unobtrusive. Lacking the ability to engage in strong hydrogen bonding, acid-base, hydrophobic, coulombic, dipolar, or π -stacking interactions, they are unlikely to perturb the biological molecules to which they are attached as long as their density of presentation on such scaffolds is not overwhelmingly great. An outstanding example of their invisible nature is provided by the growing number of examples in which azide- or alkynederivatized nutrients or cofactors are taken up and incorporated into biological molecules by living cells (Kiick et al., 2002; Prescher et al., 2004; Laughlin et al., 2006; Baskin and Bertozzi, 2007; Gierlich et al., 2007; Hsu et al., 2007; Stabler et al., 2007; Wirges et al., 2007; Ning et al., 2008; Chang et al., 2009; P.S. Banerjee et al., 2010; Breidenbach et al., 2010; Rangan et al., 2010). While exquisitely
CopperCatalyzed Azide-Alkyne Click Chemistry for Bioconjugation
157 Current Protocols in Chemical Biology
Volume 3
selective, the uncatalyzed 1,3-dipolar cycloaddition reaction of standard azides and alkynes is quite slow unless the alkyne is rendered sufficiently electron-deficient to open up conjugate addition pathways and thus compromise its bioorthogonality (Huisgen, 1962, 1989). Copper catalysis of the reaction between azides and terminal alkynes represents one solution to this problem, first described independently in 2002 by the groups of Meldal in Denmark (in the context of solid-phase peptide modification; Tornøe et al., 2002) and Sharpless/Fokin in the U.S. (in the context of solution-phase reactivity; Rostovtsev et al., 2002). This reaction has come to represent “click chemistry” for many, although other reactions are also included in this concept (Kolb et al., 2001). Another general solution to the azide-alkyne rate problem is to make the alkyne highly strained in a ring structure, and is sometimes referred to as “copper-free click chemistry.” This ligation method is not discussed here, but the reader should be aware that recent developments from the Bertozzi laboratory (Jewett et al., 2010) have made this process fast enough to be applicable to most bioconjugation conditions. In such situations, the major practical differences between the two triazole-forming reactions are the size of the participating alkyne group, favoring the CuAAC process which uses only a small and easily installed terminal alkyne group, versus the need for a catalyst, which of course only the CuAAC reaction requires.
Critical Parameters
CopperCatalyzed Azide-Alkyne Click Chemistry for Bioconjugation
The authors have found, in reviewing the literature and discussing click chemistry with others, that several practices of questionable value are sometimes employed in the execution of the CuAAC reaction. Several are listed below in approximate order of frequency. Note that these practices are not necessarily fatal; the reaction usually still works, just not at optimal rates. 1. The use of strong base in nonaqueous conditions. H¨unig’s base (iPr2 NEt) is often used and yet is unnecessary and likely to diminish cycloaddition rates. The CuAAC reaction involves both the deprotonation of alkyne and reprotonation of Cu-triazolyl intermediate. While alkynes are not very acidic, there is almost never a need to accelerate Cu-acetylide formation with base, since this is already a very fast process. 2. Failure to use an appropriate accelerating ligand. One of the reasons for the popularity of the CuAAC reaction is its permissive
nature: a wide variety of functional groups in substrates and solvates are tolerated, and ligands are often not required. However, when higher temperatures cannot be used, or rate accelerations beyond those that can be achieved by heating are required, accelerating ligands can solve may problems, since the CuAAC reaction is strongly aided by the correct coordination environment. 3. Failure to use rudimentary methods to protect reactions from oxygen. Even when excess sodium ascorbate is present to maintain a sufficient concentration of cuprous ions, it is useful to at least cap reactions to minimize oxygen exposure. Otherwise, copper will catalyze the oxidation of ascorbate, eventually depleting the reducing agent, killing the CuAAC catalyst, promoting CuII -mediated alkyne-alkyne (Glaser) coupling, and generating larger amounts of reactive oxygen species than is necessary. While many CuAAC reactions are fast enough to withstand these challenges, it is recommended that exposure to oxygen be limited by all convenient means, especially for reactions involving small amounts or low concentrations of azide and alkyne reagents, as are present in most bioconjugation situations. 4. Unrealistic expectations. Even a reaction as fast and reliable as CuAAC cannot connect two reactants to each other in a few hours if both are present in low nanomolar concentrations. In such situations, either the concentration of one of the reaction partners must be increased, or the two partners must be engineered to interact with each other in a preorganized fashion, such as via the binding of two complementary oligonucleotide chains. Also note that in this protocol, and in most CuAAC bioconjugation reactions, the copper complex is not used in catalytic amounts, but rather is present in stoichiometric or excess amounts relative to azide and alkyne. This is because the rate is dependent on copper concentration in a non-obvious way: for most catalysts of the type described here, a threshold behavior has been observed, such that little reactivity occurs below 50 μM in Cu, and maximal activity is reached at approximately 250 μM Cu (Rodionov et al., 2005; Presolski et al., 2010). 5. The use of cuprous iodide as a copper source. Iodide ions are good ligands for CuI and can therefore interfere with the CuAAC reaction under most bioconjugation conditions. In other cases, iodotriazoles can be formed via formation of intermediate iodoalkynes (Hein et al., 2009).
158 Volume 3
Current Protocols in Chemical Biology
6. The use of TCEP as reducing agent. While the use of TCEP was reported early on (Wang et al., 2003), it was later found that the Cu-binding and azide-reducing properties of phosphines can interfere with the CuAAC reaction. When ascorbate cannot be used, hydroxylamine can function as a reducing agent for CuII , or Cu wire can be used to protect a preformed CuI complex if the reaction mixture is not exposed to excessive amounts of oxygen. 7. Use of nonoptimal buffers or solvent mixtures. With the catalyst described here, the reaction has been shown to proceed well over a broad pH range (Presolski et al., 2010); however, a pH around 7 is recommended for most cases. Buffers that contain very high concentrations (greater than approximately 0.2 M) of chloride ion are to be avoided because chloride can compete for Cu at these concentrations. Tris buffers can also slow CuAAC reactions because of Cu binding by the Tris molecule. Cu-phosphate complexes are often insoluble, but if the Cu source is premixed with the ligand, such complexes do not form (or do not precipitate) even in phosphate-based buffers, and reaction rates are high. Phosphate, acetate, HEPES, or MOPS buffers are commonly employed, but others are almost certainly suitable as long as they do not contain Cu-binding species. Similarly, the role of DMSO and other coordinating co-solvents is an important one to appreciate. See Presolski et al. (2010) for a complete account, the overall lesson of which is that when high concentrations of such cosolvents are required (greater than approximately 30% to 50% of the solvent volume), a different ligand is suggested. 8. Failure to turn up the heat. The CuAAC process benefits greatly from higher temperatures; even modest increases that are tolerated by some biological molecules can produce good results. We speculate that many cases in which CuAAC bioconjugation does not work well suffer from the sequestering of the metal by competing coordinating species in solution, such as donor solvent molecules or donor groups in protein or other species present (Presolski et al., 2010). In such situations, the great kinetic lability of CuI centers may be compromised, and a little heat may be all that is needed to free the Cu ions. 9. Safety notes. Hazards associated with azide decomposition do not exist in bioconjugation situations for which the amounts of azides used tend to be small and the
molecules to which they are attached are large. Still, hazardous practices in the synthesis of organic azides should be mentioned here. Of greatest concern is the potential for the generation of hydroazidoic acid (HN3 ) in preparative-scale reactions using an electrophile and sodium azide. When working up such reactions that contain excess inorganic azide, exposure to acid will generate HN3 , which is volatile, highly toxic, and explosive. In general, therefore, such workups should avoid acid; in large amounts, azide ion should be quenched by nitration (Clusius and Effenberger, 1955; Stedman, 1960) as follows: In a fume hood, treat a stirred aqueous solution containing no more than 5% azide ion with sodium nitrite (∼7 ml of 20% aqueous NaNO2 per gram NaN3 , representing a 40% excess). Slowly add aqueous sulfuric acid (20% solution) with stirring until the reaction mixture is acidic. Note that it is important for sodium nitrite to be introduced first. When the evolution of nitrogen oxides (toxic—keep this reaction in the hood) ceases, test the acidic solution with starch iodide paper. If it turns blue, excess nitrite is present, indicating complete azide decomposition. Although not generally used with biomolecules, chlorinated solvents (particularly CH2 Cl2 and CHCl3 ) should be avoided with azide ion since it is possible to generate CH2 (N3 )2 and CH(N3 )3 , which can be highly explosive when concentrated, such as in the trap of a vacuum line. Similarly, small molecule-azides should never be isolated away from solvent in significant quantities, such as by distillation, precipitation, or recrystallization.
Troubleshooting Troubleshooting suggestions for common problems encountered with the CuAAC reaction are presented in Table 1, along with explanations of possible causes and recommended solutions for overcoming or avoiding these problems.
Anticipated Results This CuAAC protocol has been used successfully for the attachment of a wide variety of ligands [small molecules (Astronomo et al., 2010; Hong et al., 2010; Pokorski et al., 2011) proteins (D. Banerjee et al., 2010) DNA (Cigler et al., 2010), and organic polymers (Manzenrieder et al., 2011)] to a wide variety of biological molecules, such as the surfaces of
CopperCatalyzed Azide-Alkyne Click Chemistry for Bioconjugation
159 Current Protocols in Chemical Biology
Volume 3
Table 1 Troubleshooting Guide for Bioconjugation by the CuAAC Click Reaction
Problem
Cause
Solution
A dextran-alkyne at 25 mM would not click in water, even when accelerating ligands were used
Collapse of hydrophobic regions, burying the alkynes and making them inaccessible; similar challenges can occur with proteins and oligonucleotides
Perform the reaction in denaturing or solvating conditions such as the use of lots of DMSO, along with appropriate ligand (ligand 7 in Presolski et al., 2010)
Low yields in complex biological systems, such as with the product resulting from enzymatic incorporation of an alkyne into RNA
Potential failure of one or more Perform a test reaction with coumarin steps preceding the CuAAC step azide 3. If little or no reaction is observed, add propargyl alcohol. Continued failure of the reaction suggests that the biological substrate is sequestering the copper catalyst, in which case excess Cu can be added. It is also possible to release active Cu by the addition of Zn2+ .
Substrates are incompatible with ascorbate (such as showing excessive sensitivity to reactive oxygen species or dehydroascorbate byproduct formation, not solved by the recommended use of excess ligand and aminoguanidine)
Oxidation, binding, etc.?
(a) Use hydroxylamine as reducing agent for CuII (typically 10 mM). (b) Use a CuI complex such as CuBr, CuOAc, or CuOTf, and protect the reaction from air. (c) Generate CuI electrochemically.(Hong et al., 2008).
Invitrogen Click-It kit doesn’t work
Various
Use the procedure outlined in this unit
Failure of this protocol involving Sequestration of Cu away from protein, DNA, RNA, gold nanoparticles, azide and alkyne reactants or other species that can bind Cu ions
(a) Use excess Cu and ligand. (b) Add ZnII as a sacrificial metal to whatever is removing Cu from the reaction. (c) Use alternative ligand as previously described (Presolski et al., 2010) in the proper ligand:Cu ratio of 1:1 or 2:1.
CuAAC doesn’t work in fresh cell lysate Strong Cu-thiolate binding ties or under other conditions that may up the metal (CuAAC with contain free thiols THPTA usually tolerates glutathione up to 1 mM, but not more than that)
Using an accelerating ligand will help, and excess Cu, ZnII or NiII can often occupy the thiols and leave some CuI free to mediate the CuAAC reaction
Failure of His6 -tagged proteins to engage in CuAAC ligation
Use excess copper or sacrificial metals, such as ZnII or NiII ; may also switch to a FLAG or other peptide tag
The His-tag binds copper
DNA damage observed, sometimes more Oxidation from excess reactive Use CuSO4 + ascorbate, but minimize seriously when the reaction mix is oxygen species (and, in one case, agitation of the solution and keep it capped as much as possible. In extremely vortexed the use of cupric nitrate) sensitive cases, CuOAc with Cu wire provides the mildest conditions. Reactions are slow with CuBr, CuI, TCEP
CopperCatalyzed Azide-Alkyne Click Chemistry for Bioconjugation
Low solubility and bad reducing agent
virus-like particles. In many cases, the attachments result in polyvalent displays of triazoles on the biomolecular scaffold, and so yield can be difficult to quantify. However, when yields can be measured, CuAAC reactions usually
Use CuSO4 , sodium ascorbate, and DMSO, DMF, or NMP as co-solvent (up to 10%). These solvents appear to be biocompatible and help to dissolve most small molecules of interest.
give quantitative or near-quantitative yields, with excellent recovery of the desired conjugates, showing that the reaction conditions do not induce substantial cleavage of biological molecules.
160 Volume 3
Current Protocols in Chemical Biology
Time Considerations In general, the CuAAC reaction is easy to set up and perform. When the reactants are present in sufficient concentration (greater than 10 μM each), the reaction is performed properly, and no unexpected sequestration of copper ions takes place, the CuAAC reaction can be expected to provide quantitative yields of triazoles within an hour or two at room temperature. The time required for workup and purification varies from minutes for simple precipitation or molecular weight cutoff filtration to separate biomolecules from the catalyst and small-molecule reagents, to hours for chromatographic techniques.
Acknowledgements This work was supported by The Skaggs Institute for Chemical Biology and the NIH (RR021886). We are especially grateful to the discoverer of the solution-phase CuAAC reaction, Prof. Valery V. Fokin, for his insights and contributions, and we thank the many coworkers and colleagues who have contributed to our understanding of the reaction. These include Prof. K. Barry Sharpless, Dr. Valentin Rodionov, Dr. Reshma Jagasia, Dr. Warren Lewis, Dr. Andrew Udit, Dr. Yeon-Hee Lim, Dr. SoHye Cho, Dr. David D´ıaz, and Dr. Sayam Sen Gupta. Stanislav I. Presolski and Vu Phong Hong contributed equally to this work.
Literature Cited Astronomo, R.D., Kaltgrad, E., Udit, A.K., Wang, S.-K., Doores, K.J., Huang, C.-Y., Pantophlet, R., Paulson, J.C., Wong, C.-H., Finn, M.G., and Burton, D.R. 2010. Defining criteria for oligomannose immunogens for HIV using icosahedral virus capsid scaffolds. Chem. Biol. 17:357370. Banerjee, D., Liu, A., Voss, N., Schmid, S., and Finn, M.G. 2010. Multivalent display and receptor-mediated endocytosis of transferrin on virus-like particles. ChemBioChem 11:12731279. Banerjee, P.S., Ostapchuk, P., Hearing, P., and Carrico, I. 2010. Chemoselective attachment of small molecule effector functionality to human adenoviruses facilitates gene delivery to cancer cells. J. Am. Chem. Soc. 132:13,615-13,617.
of dinuclear alkynylcopper(I) complexes in alkyne-azide chemistry. Chem. Eur. J. 16:62786284. Chan, T.R. and Fokin, V.V. 2007. Polymersupported copper(I) catalysts for the experimentally simplified azide-alkyne cycloaddition. QSAR Comb. Sci. 26:1274-1279. Chan, T.R., Hilgraf, R., Sharpless, K.B., and Fokin, V.V. 2004. Polytriazoles as copper(I)-stabilizing ligands in catalysis. Org. Lett. 6:2853-2855. Chang, P., Chen, X., Smyrniotis, C., Hu, T., Bertozzi, C.R., and Wu, P. 2009. Metabolic labeling of sialic acids in living animals with alkynyl sugars. Angew. Chem. Int. Ed. 48:40304033. Cigler, P., Lytton-Jean, A.K.R., Anderson, D.G., Finn, M.G., and Park, S.Y. 2010. DNAcontrolled assembly of a NaTl lattice structure from gold nanoparticles and protein nanoparticles. Nat. Mater. 9:918-922. Clusius, K. and Effenberger, E. 1955. Reaktionen mit N-15. 21. Die Einwirkung von Nitrit auf Stickstoffwasserstoffsaure. Helv. Chim. Acta 38:1843-1847. Gierlich, J., Gutsmiedl, K., Gramlich, P.M.E., Schmidt, A., Burley, G.A., and Carell, T. 2007. Synthesis of highly modified DNA by a combination of PCR with alkyne-bearing triphosphates and click chemistry. Chem. Eur. J. 13:9486-9494. Hang, H.C. and Bertozzi, C.R. 2001. Chemoselective approaches to glycoprotein assembly. Accounts Chem. Res. 34:727-736. Hein, J.E., Tripp, J.C., Krasnova, L.B., Sharpless, K.B., and Fokin, V.V. 2009. Copper(I)catalyzed cycloaddition of organic azides and 1-iodoalkynes. Angew. Chem. Int. Ed. 48:80188021. Himo, F., Lovell, T., Hilgraf, R., Rostovtsev, V.V., Fokin, V.V., Noodleman, L., and Sharpless, K.B. 2005. Copper(I)-catalyzed synthesis of azoles. DFT predicts unprecedented reactivity and intermediates. J. Am. Chem. Soc. 127:210-216. Hong, V., Udit, A.K., Evans, R.A., and Finn, M.G. 2008. Electrochemically protected copper(I)catalyzed azide-alkyne cycloaddition. ChemBioChem 9:1481-1486. Hong, V., Presolski, S.I., Ma, C., and Finn, M.G. 2009. Analysis and optimization of coppercatalyzed azide-alkyne cycloaddition for bioconjugation. Angew. Chem. Int. Ed. 48:98799883.
Baskin, J.M. and Bertozzi, C.R. 2007. Bioorthogonal click chemistry: Covalent labeling in living systems. QSAR Comb. Sci. 26:1211-1219.
Hong, V., Steinmetz, N.F., Manchester, M., and Finn, M.G. 2010. Labeling live cells by coppercatalyzed alkyne-azide click chemistry. Bioconjugate Chem. 21:1912-1916.
Breidenbach, M.A., Gallagher, J.E.G., King, D.S., Smart, B.P., Wu, P., and Bertozzi, C.R. 2010. Targeted metabolic labeling of yeast N-glycans with unnatural sugars. Proc. Natl. Acad. Sci. U.S.A. 107:3988-3993.
Hsu, T.-L., Hanson, S.R., Kishikawa, K., Wang, S.-K., Sawa, M., and Wong, C.-H. 2007. Alkynyl sugar analogs for the labeling and visualization of glycoconjugates in cells. Proc. Natl. Acad. Sci. U.S.A 104:2614-2619.
Buckley, B.R., Dann, S.E., and Heaney, H. 2010. Experimental evidence for the involvement
Huisgen, R. 1962. 1,3-Dipolar cycloadditions past and future. Angew. Chem. Int. Ed. 2:565-632.
CopperCatalyzed Azide-Alkyne Click Chemistry for Bioconjugation
161 Current Protocols in Chemical Biology
Volume 3
Huisgen, R. 1989. Kinetics and reaction mechanisms: Selected examples from the experience of forty years. Pure Appl. Chem. 61:613-628.
profiling of bacterial lipoproteins with chemical reporters. J. Am. Chem. Soc. 132:10,62810,629.
Jewett, J.C., Sletten, E.M., and Bertozzi, C.R. 2010. Rapid Cu-free click chemistry with readily synthesized biarylazacyclooctynones. J. Am. Chem. Soc. 132:3688-3690.
Rodionov, V.O., Fokin, V.V., and Finn, M.G. 2005. Mechanism of the ligand-free CuI -catalyzed azide-alkyne cycloaddition reaction. Angew. Chem. Int. Ed. 44:2210-2215.
Kiick, K.L., Saxon, E., Tirrell, D.A., and Bertozzi, C.R. 2002. Incorporation of azides into recombinant proteins for chemoselective modification by the Staudinger ligation. Proc. Natl. Acad. Sci. U.S.A. 99:19-24.
Rodionov, V.O., Presolski, S., D´ıaz, D.D., Fokin, V.V., and Finn, M.G. 2007a. Ligandaccelerated Cu-catalyzed azide-alkyne cycloaddition: A mechanistic report. J. Am. Chem. Soc. 129:12705-12712.
Kolb, H.C., Finn, M.G., and Sharpless, K.B. 2001. Click chemistry: Diverse chemical function from a few good reactions. Angew. Chem. Int. Ed. 40:2004-2021.
Rodionov, V.O., Presolski, S., Gardinier, S., Lim, Y.-H., and Finn, M.G. 2007b. Benzimidazole and related ligands for Cu-catalyzed azide-alkyne cycloaddition. J. Am. Chem. Soc. 129:12,696-12,704.
Laughlin, S.T., Agard, N.J., Baskin, J.M., Carrico, I.S., Chang, P.V., Ganguli, A.S., Hangauer, M.J., Lo, A., Prescher, J.A., and Bertozzi, C.R. 2006. Metabolic labeling of glycans with azido sugars for visualization and glycoproteomics. Glycobiology 415:230-250. Lemieux, G.A. and Bertozzi, C.R. 1998. Chemoselective ligation reactions with proteins, oligosaccharides and cells. Trends Biotechnol. 16:506513. Lewis, W.G., Magallon, F.G., Fokin, V.V., and Finn, M.G. 2004. Discovery and characterization of catalysts for azide-alkyne cycloaddition by fluorescence quenching. J. Am. Chem. Soc. 126:9152-9153. Lipshutz, B.H., Frieman, B.A., and Tomaso, J.A.E. 2006. Copper-in-charcoal (Cu/C): Heterogeneous, copper-catalyzed asymmetric hydrosilylations. Angew. Chem. Int. Ed. 45:1259-1264.
Sen Gupta, S., Kuzelka, J., Singh, P., Lewis, W.G., Manchester, M., and Finn, M.G. 2005. Accelerated bioorthogonal conjugation: A practical method for the ligation of diverse functional molecules to a polyvalent virus scaffold. Bioconjugate Chem. 16:1572-1579. Sivakumar, K., Xie, F., Cash, B.M., Long, S., Barnhill, H.N., and Wang, Q. 2004. A fluorogenic 1,3-dipolar cycloaddition reaction of 3-azidocoumarins and acetylenes. Org. Lett. 6:4603-4606.
Manzenrieder, F., Luxenhofer, R., Retzlaff, M., Jordan, R., and Finn, M.G. 2011. Stabilization of virus-like particles with poly(2-oxazoline)s. Angew. Chem. Int. Ed. 50:2601-2605.
Soriano del Amo, D., Wang, W., Jiang, H., Besanceney, C., Yan, A., Levy, M., Liu, Y., Marlow, F.L., and Wu, P. 2010. Biocompatible copper(I) catalysts for in vivo imaging of glycans. J. Am. Chem. Soc. 132:16,893-16,899.
Ning, X., Guo, J., Wolfert, M.A., and Boons, G.-J. 2008. Visualizing metabolically labeled glycoconjugates of living cells by copper-free and fast Huisgen cycloadditions. Angew. Chem. Int. Ed. 47:2253-2255.
Stabler, C.L., Sun, X.-L., Cui, X., Wilson, J.T., Haller, C.A., and Chaikof, E.L. 2007. Chemoand bioorthogonal surface re-engineering of pancreatic islets. Bioconjugate Chem. 18:17131715.
Pokorski, J.K., Breitenkamp, K., Liepold, L.O., Qazi, S., and Finn, M.G. 2011. Functional virusbased polymer-protein nanoparticles by atom transfer radical polymerization. J. Am. Chem. Soc. 133:9242-9245.
Stedman, G. 1960. Mechanism of the azide-nitrite reaction. Part IV. J. Chem. Soc. 1960:17021709.
Prescher, J.A. and Bertozzi, C.R. 2005. Chemistry in living systems. Nat. Chem. Biol. 1:13-21. Prescher, J.A., Dube, D.H., and Bertozzi, C.R. 2004. Chemical remodelling of cell surfaces in living animals. Nature 430:873-877.
CopperCatalyzed Azide-Alkyne Click Chemistry for Bioconjugation
Rostovtsev, V.V., Green, L.G., Fokin, V.V., and Sharpless, K.B. 2002. A stepwise Huisgen cycloaddition process: Copper(I)-catalyzed regioselective ligation of azides and terminal alkynes. Angew. Chem. Int. Ed. 41:25962599.
Presolski, S.I., Hong, V., Cho, S.-H., and Finn, M.G. 2010. Tailored ligand acceleration of the Cucatalyzed azide-alkyne cycloaddition reaction: Practical and mechanistic implications. J. Am. Chem. Soc. 132:14,570-14,576. Rangan, K.J., Yang, Y.Y., Charron, G., and Hang, H.C. 2010. Rapid visualization and large-scale
Tornøe, C.W., Christensen, C., and Meldal, M. 2002. Peptidotriazoles on solid phase: [1,2,3]Triazoles by regiospecific copper(I)-catalyzed 1,3-dipolar cycloadditions of terminal alkynes to azides. J. Org. Chem. 67:3057-3064. Wang, Q., Chan, T.R., Hilgraf, R., Fokin, V.V., Sharpless, K.B., and Finn, M.G. 2003. Bioconjugation by copper(I)-catalyzed azide-alkyne [3+2] cycloaddition. J. Am. Chem. Soc. 125:3192-3193. Wirges, C.T., Gramlich, P.M.E., Gutsmiedl, K., Gierlich, J., Burley, G.A., and Carell, T. 2007. Pronounced effect of DNA hybridization on click reaction efficiency. QSAR Comb. Sci. 26:1159-1164.
162 Volume 3
Current Protocols in Chemical Biology
Target Identification Using Drug Affinity Responsive Target Stability (DARTS) Brett Lomenick,1 Gwanghyun Jung,1 James A. Wohlschlegel,2,3 and Jing Huang1,3 1
Department of Molecular and Medical Pharmacology, David Geffen School of Medicine, University of California Los Angeles, Los Angeles, California 2 Department of Biological Chemistry, David Geffen School of Medicine, University of California Los Angeles, Los Angeles, California 3 Molecular Biology Institute, University of California Los Angeles, Los Angeles, California
ABSTRACT Drug affinity responsive target stability (DARTS) is a general methodology for identifying and studying protein-ligand interactions. The technique is based on the principle that when a small molecule compound binds to a protein, the interaction stabilizes the target protein’s structure such that it becomes resistant to proteases. DARTS is particularly useful for the initial identification of the protein targets of small molecules, but can also be used to validate potential protein-ligand interactions predicted or identified by other means and to estimate the affinity of interactions. The approach is simple and advantageous because it can be performed using crude cell lysates and other complex protein mixtures (without requiring purified proteins), and it uses native, unmodified small molecules. The protocols in this unit describe the general approach for performing DARTS experiments, which can be easily modified and scaled to fit project-specific C 2011 by John Wiley & Sons, Inc. criteria. Curr. Protoc. Chem. Biol. 3:163-180 Keywords: target ID r DARTS r proteins r ligands r binding r proteomics r mass spectrometry
INTRODUCTION This unit describes general procedures for performing drug affinity responsive target stability (DARTS) assays to identify protein targets of small molecule ligands. The basis for DARTS is that a protein becomes stabilized upon binding to a small molecule compound or other ligand, which leads to decreased susceptibility of the target protein to degradation by proteases (Lomenick et al., 2009, 2011). This decreased proteolysis is specific to the target protein(s) and occurs for both high- and low-affinity compounds. Moreover, DARTS works especially well using extremely complex protein samples, such as whole-cell lysates, where nonspecific protein-ligand interactions are minimized due to the large number and variety of proteins in the mixture. DARTS is advantageous because any small molecule can be used in its native form, meaning no structure-activity relationship (SAR) studies or chemical modifications to the ligand are necessary for target identification. In contrast, affinity chromatography and most other affinity-based methods for target identification require each ligand to be chemically modified, which presents a range of complications [see Lomenick et al. (2011) for further discussion]. In addition to de novo target identification, DARTS has also proven useful for validating binding of small molecules to proposed target proteins identified through other means (Aghajan et al., 2010; Chen et al., 2011). In DARTS, protein samples are incubated with the compound of interest and an inactive analog or solvent control, followed by digestion with proteases. The samples are then compared to look for proteins whose proteolysis is decreased when the compound is Current Protocols in Chemical Biology 3: 163-180, December 2011 Published online December 2011 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/9780470559277.ch110180 C 2011 John Wiley & Sons, Inc. Copyright
Target Identification Using DARTS
163 Volume 3
present. Basic Protocol 1 describes how to perform DARTS using mammalian cell lysates and the protease mixture Pronase. Basic Protocol 2 is very similar to Basic Protocol 1, but uses yeast cell lysates and the protease thermolysin. Both protocols are flexible and can be modified for use with any protein source. DARTS experiments performed using both basic protocols are typically analyzed by immunoblotting for proteins of interest or by SDSPAGE and gel staining to search for differentially proteolysed proteins in an unbiased manner. Support Protocols 1 and 2 describe how to perform shotgun mass spectrometry (MudPIT) and Gelfree fractionation of DARTS samples, which are more sensitive than the gel-based approaches and may facilitate identification of lower abundance protein targets.
STRATEGIC PLANNING There are a few general key choices that must be made for each DARTS experiment, but overall the procedure is relatively standardized. First, the source of protein must be chosen. Generally, any cell type that is sensitive to the biological effects of the small molecule can be used. We have successfully performed DARTS with lysates from a range of mammalian cells, including HEK293, HeLa, Jurkat, A549, MEFs, and Raw264.7, as well as from the yeast S. cerevisiae. Second, any small molecule believed to bind proteins should be suitable for DARTS (including drug-like small molecules that are not susceptible to proteolysis, and peptide ligands that are themselves susceptible to proteolysis but whose binding to target proteins would protect them from proteolysis), but the concentration range of small molecule to use is also an important consideration. Given that the targets of most small molecules being used for DARTS are unknown, the binding affinities will also not be known. Therefore, one can only estimate the binding affinity to the most relevant target(s) based upon the EC50 of the compound, although this correlation may not be valid for all compounds and biological systems. Since the EC50 gives only a rough estimate of binding affinity, we suggest initially using a concentration of the compound that is 10-fold higher than the EC50 . Using a concentration of compound that is significantly higher than the KD will help ensure maximal protection of the target protein from proteolysis by saturating the protein with ligand. There are many different proteases available that could be used with DARTS. We have primarily used only two: thermolysin and pronase. Thermolysin is a metalloendopeptidase from a thermophilic bacterium with low cleavage specificity. Its primary attribute is that it can only efficiently digest proteins that are unfolded (Arnold et al., 1996). Pronase, on the other hand, is a mixture of multiple proteases that can digest both folded and unfolded proteins. For proteins that can be digested by thermolysin under standard conditions, the magnitude of protection from proteolysis afforded by small molecule ligands is typically high. However, if the target protein is unable to be digested with thermolysin (e.g., GAPDH), then pronase must be used. When DARTS is used as a tool for novel target identification, pronase is the preferred choice of enzyme given the limited substrate pool of thermolysin.
Target Identification Using DARTS
In order to identify novel targets of small molecules by DARTS, one or more proteomics techniques must be used subsequent to the proteolysis. In our initial DARTS study, we used the simplest method possible, SDS-PAGE and gel staining, to identify the known target of didemnin B and a novel target of resveratrol (Lomenick et al., 2009). However, the majority of proteins in most cells and tissues are not present in high abundance and cannot be observed using unbiased gel-based proteomics. Therefore, more sensitive, quantitative proteomics technologies must be employed [see Lomenick et al. (2011) for a complete discussion]. Our laboratory is currently combining many of the newest and most sensitive proteomics methods available with DARTS to identify the less abundant protein targets of small molecules. For the purpose of this protocol, we are focusing on providing
164 Volume 3
Current Protocols in Chemical Biology
as much information as possible for performing the initial part of a DARTS experiment. It will be up to the individual investigator to determine which proteomics method to use in conjunction with DARTS, which includes but is not limited to immunoblotting, SDS-PAGE, 2D-PAGE, and gel-free mass spectrometry-based proteomics [see Lomenick et al. (2011) and Support Protocols 1 and 2 for further discussion].
DARTS EXPERIMENT FOR ANALYZING DRUG BINDING TO PROTEIN TARGETS IN MAMMALIAN CELL LYSATES WITH PRONASE
BASIC PROTOCOL 1
Once the details of cell type and ligand concentration are decided, the protocol provided here can be used in conjunction with any quantitative analytical platform, such as immunoblotting and gel-based or gel-free MS-based proteomics. The amounts of reagents used in this procedure are optimized for performing DARTS with the purpose of analyzing the results by immunoblotting or SDS-PAGE and gel staining, but can easily be scaled up or down as necessary (see Support Protocol 2 for gel-free proteomics). In this procedure cells are collected and lysed, lysates are prepared and quantified, and the lysates are then treated with the ligand/compound of interest and a negative control (e.g., solvent or inactive analog), and proteolysis is then performed (see Fig. 1). When looking at specific proteins by immunoblotting, it is recommended to initially use a range of pronase concentrations to determine the optimal amount needed to partially digest the proteins of interest. Based on our observations that nearly all proteins will be partially to completely digested within the range of 1:100 to 1:10,000 pronase to protein ratio (a 1:100 ratio will use 1 μg pronase for every 100 μg protein), we begin by digesting aliquots of the samples with 1:100, 1:300, 1:1000, 1:3000, and 1:10,000 ratios. Here, we use the
cell lysates 600 l
add 66 l of 10 TNC and measure the protein concentration DMSO
drug
add 3 l DMSO or drug (100 ) and incubate 1 hr at room temperature for drug binding
binding reactions 300 l
make 50 l aliquots, add 2 l diluted pronase solutions (0, 1:100, 1:300, 1:1000, 1:3000, 1:10,000) and incubate for 30 min
pronase drug target protein
immunoblotting
nontarget protein
Figure 1
Experimental scheme of DARTS using mammalian cell lysates.
165 Current Protocols in Chemical Biology
Volume 3
mass ratio of pronase to protein, since protein mass directly correlates with the number of peptide bonds, which are the actual substrate of proteases. Once the optimal amount of pronase is determined for a protein of interest (one in which approximately 50% to 80% of the protein is digested), subsequent experiments to perform concentration-effect curves (or time courses) or to test multiple small molecules can be done using this single pronase concentration. On the other hand, when mass spectrometry identification will be used to look for unknown protein targets, we typically achieve the best results with high concentrations of pronase between 1:1000 and 1:100. Lower amounts of pronase result in too many proteins not being digested at all; this limits the effectiveness of the gel-based or gel-free proteomics approaches. Furthermore, we have found that we can detect significant amounts of specific proteins using mass spectrometry in proteolysed samples even when the protein can no longer be detected by immunoblotting. We believe this is due to the fact that the proteins have been cut into multiple small peptides that are no longer efficiently recognized by antibodies, although other explanations may exist.
Materials M-PER lysis buffer (Pierce, cat. no. 78501) Phosphatase inhibitor solutions (100 mM β-pyrophosphate, 50 mM sodium pyrophosphate, 200 mM sodium vanadate, and 1 M sodium fluoride) 20× Protease inhibitor solution (see recipe) Phosphate-buffered saline [PBS; 1× PBS was diluted from 10× PBS (MPbio, cat. no. PBS10X02) using filter-sterilized distilled water] 10-cm dish cell culture in logarithmic growth phase (up to 70% to 80% confluent) 10× TNC buffer (see recipe) BCA protein concentration assay (Pierce, cat. no. 23225 or similar assay) Dimethyl sulfoxide (DMSO; or other corresponding vehicle) 100× stock solution of small molecule(s) in DMSO (DMSO can be replaced with other vehicle of choice) Pronase stock solution (see recipe) 1× TNC buffer (dilute 10× TNC buffer 10-fold in dH2 O) Cell scraper (e.g., Fisher, cat. no. 08-100-241) 1.5-ml tubes Refrigerated centrifuge capable of 18,000 × g (e.g., Beckman Microcentrifuge 22R) Collect and lyse the cells 1. Prepare 1 ml of M-PER lysis buffer by mixing 883 μl M-PER, 10 μl 100 mM βpyrophosphate, 50 μl 50 mM sodium pyrophosphate, 5 μl 200 mM sodium vanadate, 2 μl 1 M sodium fluoride, and 50 μl 20× protease inhibitor solution on ice. For DARTS experiments with mammalian cells, we typically lyse using the commercial M-PER buffer, which has worked well for a variety of cell lines and small molecule-protein pairs. However, it is not necessary to use M-PER, as DARTS can be successfully performed with a variety of gentle, nondenaturing lysis buffers (e.g., 0.2% to 1% Triton X-100, NP-40). However, harsh, denaturing lysis buffers like RIPA (radio-immuno-precipitation assay) buffer should be avoided for DARTS, as most proteins must retain their native complexes and/or structures in order to bind ligands. The exact protease and phosphatase inhibitors used are typically not critical, and numerous companies sell inhibitor cocktails that are similar in composition and could be used instead of the individual ones listed here. These inhibitors will not significantly affect pronase (nor thermolysin or subtilisin), which are added in excess during the ensuing DARTS reactions. Whether or not adding phosphatase inhibitors is crucial for DARTS has not been tested. Target Identification Using DARTS
2. Remove the growth medium from a single 10-cm dish of cells and wash the cells once with 1 ml cold PBS.
166 Volume 3
Current Protocols in Chemical Biology
If cells that grow in suspension are to be used, simply pellet the cells by centrifuging 10 min at 1000 × g, 4◦ C. After centrifuging, remove the growth medium, resuspend the cells in M-PER lysis buffer, and then proceed to step 4 of this protocol. For Jurkat cells, 3 × 107 cells lysed in 600 μl M-PER will typically yield soluble lysates of ∼5 μg protein/μl.
3. Apply 600 μl cold M-PER lysis buffer (with added protease and phosphatase inhibitors) onto the cells and collect the cells with a cell scraper by holding the plate at an angle and gently scraping from top to bottom. 4. Transfer the M-PER/lysing cells into a 1.5-ml tube prechilled on ice, mix it well by pipetting up and down several times, and incubate the tube on ice for 10 min. Upon completion of the incubation the cells will be completely lysed and soluble proteins will be extracted into the M-PER solution. A white, cloudy material, which is primarily DNA, should be visible floating within the solution. This material must first be separated from the soluble proteins before performing DARTS.
Prepare cell lysates for DARTS experiment 5. Centrifuge the tube 10 min at 18,000 × g, 4◦ C (e.g., Beckman Coulter microcentrifuge 22R). 6. Transfer 600 μl of the supernatant into a new 1.5-ml tube and discard the pellet. The pellet contains the DNA and some other insoluble materials from the cell. The supernatant contains the soluble proteins from the cytoplasm, nucleus, mitochondria, and other cellular compartments.
7. Add 66.7 μl 10× TNC buffer to the protein lysates and mix well. DARTS with pronase works with or without adding the TNC buffer (although we have not done a side-by-side comparison). In our earlier DARTS experiments with thermolysin the TNC buffer was added to ensure that there was enough Ca2+ for thermolysin to be fully active, given that the exact composition of M-PER is unknown and that most common lysis buffers do not contain Ca2+ .
8. Measure the protein concentration of the lysates using the BCA, Bradford, or other assay, according to the manufacturer’s instructions. We routinely use the BCA protein concentration assay, which is not affected by the M-PER or TNC buffers. The exact protein concentration is not critical, but DARTS may not work as well if the protein is too dilute. We find that an optimal protein concentration for DARTS may be between 4 to 6 μg/μl, though we have also used 2 μg/μl with good results (lower has not been tested extensively). The amount of lysis buffer and cells used in this protocol should result in lysates of ∼5 μg/μl. Although the exact protein concentration does not appear to be critical, the use of a fixed protein concentration in this protocol is primarily to standardize the protease amount. We initially optimized the amount of proteolysis for specific proteins (wherein 50% to 80% of the protein of interest is digested, see above) based on the ratio of protease to protein. However, the amount of proteolysis in a sample depends on a multitude of factors, including the concentration of the protein sample, the concentration of the protease, the buffer components, and the time of proteolysis. Therefore, two samples with different protein concentrations but digested with the same protease:protein ratio will not be digested equally because both the protein and the protease concentrations will be different.
9. Split the lysates into two samples by transferring 297 μl into each of two 1.5-ml tubes. At this point, the lysates should be allowed to warm to room temperature. Some small molecules may have low solubility in aqueous solution and will be more prone to precipitate at lower temperatures. Moreover, hydrophobic interactions are destabilized at low temperatures, and binding kinetics should be faster at the higher temperature. The rest of the experiment could also be performed at physiological temperature, but this is not necessary. Current Protocols in Chemical Biology
Target Identification Using DARTS
167 Volume 3
Incubate the protein lysates with the small molecule 10. Add 3 μl DMSO into one tube, and into the other tube add 3 μl small molecule at 100× the final concentration for a 1:100 dilution. The small molecule should be dissolved in 100% DMSO. If another solvent, such as ethanol, must be used, then substitute this solvent for DMSO in this step. Alternatively, if an inactive analog of the small molecule is available, it could be used as the control. The small molecule stock solution does not have to be exactly 100×, but we advise that the final DMSO concentration in the lysates be kept low, around 1% to 2%, as higher concentrations may interfere with binding and proteolysis. We also prefer to not pipet volumes less than 1 μl, as the chance of significant pipetting error becomes greater. Therefore, do not use a small molecule stock solution that is greater than 300× directly (dilute it first), unless the volume of lysates is also larger.
11. Mix the samples immediately by gently flicking the tubes with a finger several times, briefly microcentrifuge to bring all the liquid to the bottom of the tubes, and allow them to incubate at room temperature for 1 hr (or longer if determined necessary from pilot studies). It is critical that the samples be mixed thoroughly but gently to ensure that the small molecule/protein solution becomes homogeneous and binding equilibrium is reached. Furthermore, it is essential that you do not vortex or use any other harsh mixing method on any sample that contains protein! This will cause many proteins to become denatured and/or aggregated.
Prepare for proteolysis 12. Thaw one aliquot of 10 mg/ml pronase quickly and place it on ice. 13. Dilute pronase to 1.25 mg/ml by mixing 12.5 μl pronase with 87.5 μl cold 1× TNC buffer, which will serve as the 1:100 pronase stock solution. It is essential to keep all protease solutions on ice at all times to prevent them from starting to digest themselves. The 1.25 mg/ml pronase solution will be the highest concentration used in this experiment. The pronase stock solutions prepared in this and the following step are calculated for lysates of 5 μg protein/μl, and will need to be modified if the protein concentration is significantly different.
14. Dilute the 1:100 pronase solution serially by mixing with 1× TNC to create 1:300, 1:1000, 1:3000, and 1:10,000 pronase stock solutions.
Perform proteolysis 15. Prepare five aliquots from both protein samples, each with 50 μl, and save the remaining 50 μl of each sample as a nondigested control sample. One aliquot of the compound-treated sample and one of the DMSO control sample will be digested with each of the five pronase stock solutions.
16. Start a timer and immediately add 2 μl 1:100 pronase solution to one aliquot of compound-treated sample, mix well, and incubate at room temperature. 17. At exactly 1 min after starting the first digest, add 2 μl 1:100 pronase solution to one aliquot of the DMSO sample, mix well, and incubate at room temperature. 18. Continue to begin digests of the eight remaining aliquots in 1-min intervals by adding 2 μl of the corresponding pronase stock solution into an aliquot of each protein sample. 19. After 30 min, stop the digestion of the first aliquot started by adding 3 μl cold 20× protease inhibitor solution, mixing well, and placing on ice. Target Identification Using DARTS
168 Volume 3
Current Protocols in Chemical Biology
Alternatively, if the samples will be run on SDS-PAGE gels, the digestions can be stopped by directly adding SDS loading buffer into each sample and heating to 70◦ C for 10 min or 95◦ C for 5 min immediately, without needing to add protease inhibitors. After heating, the samples may be loaded onto gels right away or stored at −20◦ C.
20. Stop all the remaining digestions in the order they were started in 1-min intervals by adding 3 μl cold 20× protease inhibitor solution to each (or SDS loading buffer and heating immediately as described above), mixing well, and placing on ice. After stopping the proteolysis, the DARTS experiment is finished. Now the samples can be analyzed by SDS-PAGE or immunoblotting, using standard equipment and reagents. When performing immunoblotting to identify differentially proteolysed target proteins, it is critical to blot for additional control proteins to ensure that the differential proteolysis is not widespread, as would occur if the drug used affects the activity of the DARTS protease(s). We routinely blot for GAPDH, actin, and tubulin as control proteins, but other proteins could also be used. It is important to not be reliant on just one control protein, especially as these proteins are known targets of a variety of small molecules.
THERMOLYSIN DIGESTION DARTS EXPERIMENT FOR ANALYZING BINDING OF SMALL MOLECULES TO PROTEIN TARGETS IN YEAST CELL LYSATES
BASIC PROTOCOL 2
For the same reason that yeast has been a favored organism for genetics/genomics-based target identification approaches [reviewed in Lomenick et al. (2009)], the smaller size and lesser complexity of the yeast proteome (e.g., compared to the human proteome) means that de novo DARTS target identification using yeast cell lysates is also likely advantageous—when the target of a small molecule is conserved in yeast. However, unlike phenotype-based target identification, DARTS may still be useful even when there is no conserved yeast target; for instance, because DARTS can detect very low affinity (mid to high-micromolar) binding interactions, a non-optimal binding target identified in yeast could give clues to homologous mammalian target proteins or domains that may be higher affinity targets. This protocol provides instructions for performing DARTS using S. cerevisiae cell lysates and the protease thermolysin. Most of the protocol is very similar to Basic Protocol 1, with the main difference being that yeast cells require different lysis conditions due to their cell wall. Both pronase and thermolysin can be used with lysates from any cell type, but typically much higher amounts of thermolysin are required to achieve significant digestion of the proteome. Despite the limited digestive capabilities of thermolysin, as discussed above, it still proves useful for some proteins. This protocol, like the first one, can be scaled up or down as necessary. The volumes of reagents given provide sufficient protein for immunoblotting and SDS-PAGE. Although high concentrations of thermolysin are required to visualize significant digestion on stained gels, some individual proteins are still quite sensitive to its proteolysis activity. Therefore, this protocol provides instructions for digesting with thermolysin to protein ratios of 1:10, 1:50, 1:250, 1:1250, and 1:6250. If the DARTS experiment will be analyzed using proteomics methods, such as gel staining or MudPIT, the lower amounts of thermolysin may be omitted since most proteins will not be digested. However, it is recommended to use the lower amounts of thermolysin when performing immunoblotting in order to avoid over-digestion of target proteins. DARTS-immunoblot analysis in yeast is largely facilitated by the availability of genomewide yeast epitope-tagged collections. We have made use of the commercialized library of TAP-tagged strains (Ghaemmaghami et al., 2003) in many DARTS experiments [Lomenick et al. (2009), and B. Lomenick et al., unpub. observ.], and the HA-tagged and GST-tagged strains may also be useful for this purpose (Ross-Macdonald et al., 1999; Martzen et al., 1999; Zhu et al., 2001).
Target Identification Using DARTS
169 Current Protocols in Chemical Biology
Volume 3
Materials 2× Triton X-100 lysis buffer (see recipe) 4× Phosphatase inhibitor solution (see recipe) 20× Protease inhibitor solution: (see recipe) Wild-type strain of S. cerevisiae grown to mid-log phase (∼2 × 107 cells/ml) BCA protein concentration assay (Pierce, cat. no. 23225 or similar assay) 1× TNC Buffer (diluted from 10× TNC buffer; see recipe for 10× TNC buffer) Dimethyl sulfoxide (DMSO) 100× small molecule stock solution in DMSO Thermolysin stock solution (see recipe) 0.5 M EDTA (pH 8.0) Refrigerated centrifuge capable of 18,000 × g (e.g., Beckman microcentrifuge 22R) 1.5-ml tubes 0.5-mm glass beads (e.g., BioSpec Products, cat. no. 11079105) Vortex mixer 21-G needles Benchtop microcentrifuge Collect and lyse the cells 1. Prepare 300 μl Triton X-100 lysis buffer by mixing on ice 60 μl dH2 O, 150 μl 2× Triton X-100 lysis buffer, 75 μl 4× phosphatase inhibitor solution, and 15 μl 20× protease inhibitor solution. 2. Collect 25 ml yeast cells grown to mid-log phase (∼2 × 107 cells/ml) by centrifuging 5 min at 1000 × g, 4◦ C. 3. Remove the medium completely and put the cell pellet on ice. 4. Resuspend the cells in 300 μl cold Triton X-100 lysis buffer and transfer into a 1.5-ml tube containing ∼600 μl of 0.5-mm glass beads. 5. Vortex the tube of cells 1 min at maximum speed, room temperature. 6. Transfer the tube of cells to ice and allow them to rest for 1 min on ice. 7. Repeat this cycle of 1 min vortexing and 1 min resting four times. 8. Invert the tube and poke a single hole in the bottom using a 21-G needle, then place the tube into another 1.5-ml tube and microcentrifuge for 15 sec at 3000 rpm, room temperature, in a standard benchtop microcentrifuge to separate the lysates from the beads. The 0.5-mm glass beads will remain in the original tube and can be discarded.
Prepare cell lysates for DARTS experiment 9. Microcentrifuge the new tube with the lysates 10 min at 18,000 × g (or maximum speed), 4◦ C. 10. Transfer the supernatant into a new 1.5-ml tube on ice and discard the pellet. The pellet contains the DNA and other insoluble materials from the cell, while most soluble proteins will be in the supernatant.
11. Measure the protein concentration of the lysates using the BCA or similar assay, according to the manufacturer’s instructions. Target Identification Using DARTS
The amount of cells and lysis buffer used should yield around 10 mg/ml protein.
12. Dilute lysates to 5 mg/ml by mixing 300 μl lysates with 300 μl 1× TNC buffer.
170 Volume 3
Current Protocols in Chemical Biology
Incubate protein lysates with the drug 13. Transfer 297 μl lysates into two different 1.5-ml tubes. 14. Add 3 μl DMSO into one tube and 3 μl 100× compound into the other, mix them well, and incubate 1 hr at room temperature.
Prepare for proteolysis 15. Thaw one aliquot of 10 mg/ml thermolysin and place on ice immediately. 16. Dilute the 10 mg/ml thermolysin serially by mixing with 1× TNC to create 2 mg/ml, 0.4 mg/ml, 80 μg/ml, and 16 μg/ml thermolysin stock solutions. The final ratios of the thermolysin solutions will be 1:50, 1:250, 1:1250, and 1:6250, along with the original 10 mg/ml solution for 1:10.
17. Prepare five 50-μl aliquots from both protein samples and save the remaining 50 μl of each sample as a nondigested control sample. One aliquot of the compound-treated sample and one of the DMSO control samples will be digested with each of the five thermolysin stock solutions.
Perform proteolysis 18. Start a timer and immediately add 2.5 μl 10 mg/ml thermolysin solution to one aliquot of compound-treated sample, mix well, and incubate at room temperature. 19. At exactly 1 min after starting the first digest, add 2.5 μl 10 mg/ml thermolysin solution to one aliquot of the DMSO sample, mix well, and incubate at room temperature. 20. Continue to begin digests of the eight remaining aliquots in 1-min intervals by adding 2.5 μl of the corresponding thermolysin solution into an aliquot of each protein sample. 21. After 15 min, stop the digestion of the first aliquot started by adding 5 μl of 0.5 M EDTA, mixing well, and placing on ice. 22. Stop all the remaining digestions in the order they were started in 1-min intervals by adding 5 μl of 0.5 M EDTA to each, mixing well, and placing on ice. After stopping the proteolysis, the samples can be analyzed by SDS-PAGE or immunoblotting, using standard equipment and reagents, as described in Basic Protocol 1.
PREPARATION OF DARTS SAMPLES FOR MudPIT ANALYSIS The following is a general protocol for preparing DARTS samples for analysis by shotgun mass spectrometry, as recently described (Lomenick et al., 2011). Filtering away small digested peptides from DARTS samples is not necessary for most other analyses, such as SDS-PAGE, immunoblotting, or gel-free fractionation (Support Protocol 2). DARTS should be performed as described in Basic Protocol 1 or Basic Protocol 2 first, and then the samples processed using this protocol prior to multidimensional protein identification technology (MudPIT) analysis (Washburn et al., 2001; Wolters et al., 2001). The purpose of this procedure is to filter away the small peptides generated by the DARTS proteolysis so that the sample only contains whole proteins and large protein fragments (≥10 kDa). The filtered protein sample can then be analyzed by shotgun mass spectrometry (MudPIT) to identify protein targets. Dialysis of the DARTS samples can be performed instead of filtration, but this requires much more time (overnight) and both procedures yield similar results.
SUPPORT PROTOCOL 1
Target Identification Using DARTS
171 Current Protocols in Chemical Biology
Volume 3
Additional Materials (also see Basic Protocol 1 or Basic Protocol 2) TE buffer (see recipe) Protein samples (see Basic Protocol 1 or Basic Protocol 2) 100% (w/v) trichloroacetic acid (TCA; Sigma, cat. no. T0699) HPLC-grade acetone (Fisher, cat. no. AC26831) 8 M urea, 100 mM Tris·Cl, pH 8.5 (prepared fresh) 200 mM Tris(2-carboxylethyl)-phosphine hydrochloride (TCEP) 500 mM Iodoacetamide (prepared fresh) Sequencing-grade endoproteinase Lys-C (Princeton Separations, cat. no. EN-130) 100 mM Tris, pH 8.5 100 mM CaCl2 Sequencing-grade trypsin 90% formic acid (HPLC grade; e.g.,Fisher, cat. no. PI-28905 diluted with HPLC-grade water to 90% final concentration) Vivacon 500 10K MWCO Spin Columns (Sartorius Stedim, cat. no. VN01H01) 1. Perform DARTS for the compound of interest using pronase ratios of 1:300 and 1:100 and/or a thermolysin ratio of 1:10, as described in Basic Protocols 1 and 2. Other protease amounts may be used, but lower protease amounts are unlikely to provide sufficient digestion of most proteins. On the other hand, higher amounts of protease or longer digestion times may allow for the identification of target proteins that are especially stable and resistant to proteolysis.
Remove digested peptides by filtration 2. Add 400 μl cold TE buffer into one Vivacon 500 10K spin column per sample. For the pronase-digested samples, include protease inhibitors in the TE buffer. This is not necessary for thermolysin-digested samples.
3. Load 100 μg protein per DARTS sample into each Vivacon column. If SILAC or another stable isotope labeling method will be used for quantitative comparison, combine 50 μg labeled protein from the compound-treated sample and 50 μg labeled protein from the control sample into one Vivacon column and process together.
4. Centrifuge the columns at 12,000 × g at 4◦ C until only about 20 μl remains in the top of the column. Try not to spin the columns dry, as protein recovery may be affected due to precipitation and protein sticking to the membrane. If the column is spun dry, resuspend the protein in TE buffer by pipetting up and down several times.
5. Discard the flow through, add 400 μl cold TE buffer, and repeat the spin. Repeat this step. 6. Collect the washed samples from the top of the filter columns and transfer into new 1.5-ml tubes. 7. Wash the top of the filter columns with 30 μl cold TE buffer to recover any remaining protein and combine this with the rest of each sample.
Precipitate the protein samples 8. Add 100% TCA to a final concentration of 20% and mix. 9. Incubate the samples on ice for at least 30 min (can be overnight). 10. Centrifuge the samples 15 min at 18,000 × g, 4◦ C. Target Identification Using DARTS
172 Volume 3
11. Remove the supernatant without disturbing the white protein pellet. The pellet contains all the protein from the sample. It is acceptable to leave a few microliters of the supernatant if necessary to avoid disturbing the pellet. Current Protocols in Chemical Biology
12. Wash the pellet with 500 μl ice-cold acetone. 13. Centrifuge the samples 15 min at 18,000 × g, 4◦ C. 14. Remove the supernatant as above without disturbing the protein pellet. 15. Repeat steps 12 to 14 for a second wash. 16. Allow the samples to air-dry by placing into a chemical hood for several minutes with the lids open until all the remaining acetone has evaporated. At this point, the tube can be sealed with Parafilm and the protein pellet can be safely stored long-term (up to 1 year) at −20◦ C.
Prepare the protein pellets for Lys-C/trypsin digestion 17. Dissolve each pellet in 50 μl 8 M urea, 100 mM Tris·Cl, pH 8.5. 18. Add 1.25 μl 200 mM TCEP to a final concentration of 5 mM. 19. Mix and incubate the sample 20 min at room temperature. 20. Add 1 μl 500 mM iodoacetamide to a final concentration of 10 mM. 21. Mix and incubate the sample 20 min at room temperature in the dark.
Digest the samples with Lys-C and trypsin 22. Add 1 μg Lys-C into the sample and incubate 4 hr at 37◦ C in the dark. 23. Add 150 μl of 100 mM Tris, pH 8.5, for a final volume of 200 μl and a final urea concentration of 2 M. 24. Add 2 μl of 100 mM CaCl2 for a final concentration of 1 mM. 25. Add 2 μg trypsin and incubate 4 hr at 37◦ C in the dark. This step can be performed overnight.
26. Stop the digestion by adding 11.5 μl of 90% formic acid for a final concentration of 5%. The sample can be stored up to 1 year at −80◦ C. For additional information concerning preparing protein samples for mass spectrometry, see Washburn (2008).
Analyze the samples by MudPIT 27. Analyze the samples by MudPIT. For additional details, see Washburn (2008). ANALYSIS OF DARTS SAMPLES BY IN-SOLUTION MOLECULAR WEIGHT–BASED FRACTIONATION AND MASS SPECTROMETRY
SUPPORT PROTOCOL 2
In addition to the shotgun approach to DARTS described in Support Protocol 1, molecular weight–based fractionation of DARTS samples, followed by mass spectrometry analysis of the fractions, has proven to be an even more powerful approach. In order for the shotgun approach to identify a target protein, a significant fraction of the protein must be completely digested into pieces small enough to be filtered away, resulting in an overall enrichment of the protein level in the compound-treated sample versus the control sample (Fig. 2, Scenario A). However, in many DARTS experiments the target protein is only partially digested, in which case protein fragments that are too large to be filtered away remain in the sample after proteolysis (Fig. 2, Scenario B; also see Fig. 3). Although the target protein is protected from proteolysis (indicated by the depletion of the full-length protein in the control sample relative to the compound-treated sample), the total amount of the protein remains unchanged. In this scenario (B), it would be more difficult to identify the target protein using the “whole-sample” shotgun approach since that would
Target Identification Using DARTS
173 Current Protocols in Chemical Biology
Volume 3
protease
protease
protease drug
mol. wt. : High
target protein bands
Low
Scenario A
Scenario B
total amount of target protein
a
b
c
a
b
c
a
b
c
drug Gelfree fractionation drug Low
High
Low
High
Low
High : mol. wt.
Figure 2 DARTS target identification by Gelfree fractionation vs. whole-sample shotgun MS. The starting amount of target protein is equivalent in aliquots of lysate containing and lacking the small molecule. Compound binding causes differential proteolysis of the target protein in DARTS. In Scenario A, the target protein is nearly completely digested in the absence of compound, whereas most of the target protein is undigested in the presence of compound. Both Gelfree fractionation and whole sample shotgun MS analysis are able to identify the target protein here due to its depletion from the control sample. In Scenario B, however, the target protein is only partially digested in the control sample. Here the total amount of target protein is not significantly different between the compound-treated and control samples despite clear protection from proteolysis by the compound. While whole sample shotgun MS is unable to identify the target protein in this case, the Gelfree fractionation approach that takes molecular weight into consideration can identify the target protein based upon its differently sized fragments in the two samples.
require the identification and quantification of the small number of peptides derived from the proteolytically protected fragment in the background of all of the peptides from the entire protein. To avoid failing to identify target proteins in such cases of partial proteolysis (false negatives), we hypothesized that a proteomics approach that analyzes the size of the proteins in addition to the amount present in a sample would be better able to identify target proteins with DARTS. We therefore implemented an in-solution molecular weight–based fractionation system called Gelfree (Protein Discovery) that allows for precise separation of complex protein mixtures into multiple fractions of discrete molecular weight ranges. Combining the fractionation approach with DARTS and mass spectrometry analysis allows for the identification of proteins that have been proteolysed to a different extent between samples, even when the total amount of protein in the sample remains unchanged (Fig. 2, bottom row). This protocol provides instructions for preparing DARTS samples for Gelfree fractionation.
Additional Materials (also see Basic Protocol 1 or Basic Protocol 2) Target Identification Using DARTS
1 M dithiothreitol (DTT) 5× sample buffer (provided with the cartridge kit)
174 Volume 3
Current Protocols in Chemical Biology
Zeba Spin Desalting Columns, 7K MWCO, 0.5 ml (Thermo Scientific, cat. no. 89882) 10% Tris-acetate cartridge kit (Protein Discovery, cat. no. 42105) Gelfree 8100 Fractionation Station (Protein Discovery) Additional reagents and equipment for filter-aided sample preparation (FASP; Wisniewski et al., 2009) 1. Perform DARTS for the small molecule of interest using pronase:protein ratios of 1:300 and 1:100. 2. Desalt 200 μg aliquots from each DARTS sample with Zeba spin desalting columns according to the manufacturer’s instructions. If SILAC or another stable isotope labeling method will be used for quantitative comparison, combine 100 μg labeled protein from the compound-treated sample and 100 μg labeled protein from the control sample into one and process as a single sample throughout this protocol.
3. Prepare the samples for fractionation by bringing their total volumes up to 112 μl with dH2 O, adding 8 μl 1M DTT and 30 μl 5× sample buffer (provided in cartridge kit), mixing, and heating to 50◦ C for 10 min. 4. Allow the samples to cool to room temperature and load each one into an individual lane of a 10% Tris-acetate cartridge. 5. Separate the samples into a desired number of fractions according to the instructions provided by Protein Discovery.
00 12 0 1: 0 60 0 1: 30 0 1:
24 1:
1:
0
pronase:
48
00
The Gelfree system allows complete user control over how many fractions can be collected and to what molecular weight ranges those fractions correspond. We recommend initially using the general fractionation method provided with the cartridge kit, which will yield twelve evenly spread fractions of proteins ranging in size from about 5 to 100 kDa. If desired, variations from this method can be tested on a single DARTS sample and the fractions analyzed by traditional SDS-PAGE and silver staining to determine suitability prior to mass spectrometry analysis.
DB: 50 35
IB: eEF-1A
25 50
IB: beta-tubulin 35 50
IB: alpha-tubulin 35 35
IB: GAPDH
25
Figure 3 Representative DARTS results for pronase-digested Jurkat cell lysates. Jurkat cell lysates (6.2 μg/μl) were prepared as described in Basic Protocol 1, incubated with 10 μM didemnin B or DMSO for 1 hr, followed by digestion with pronase to protein ratios of 1:4800, 1:2400, 1:1200, 1:600, and 1:300 for 30 min. Immunoblotting shows protection of the known target protein eEF-1A, whereas digestion of the nontarget proteins GAPDH, alpha-tubulin, and beta-tubulin is unchanged by didemnin B. Notice that GAPDH is more resistant to proteolysis than the other three proteins, requiring a higher pronase to protein ratio for digestion. Current Protocols in Chemical Biology
Target Identification Using DARTS
175 Volume 3
6. Prepare the fractions for mass spectrometry analysis by alkylation and trypsin digestion using the filter-aided sample preparation (FASP) protocol (Wisniewski et al., 2009). FASP is recommended because it is a robust, efficient method for preparing SDScontaining samples for mass spectrometry. Alternatively, the fractions can be precipitated using TCA or acetone and in-solution reduced, alkylated, and digested, but it is critical that the preparation method used removes the SDS from the samples prior to mass spectrometry analysis.
7. Perform LC-MS or MudPIT analysis of the fractions. If LC-MS analysis will be used, it is recommended that the HPLC gradient used for each fraction be at least 90 min long, as most fractions will contain a very large number of peptides from dozens or even hundreds of proteins.
REAGENTS AND SOLUTIONS Use filter-sterilized glass-distilled water or Milli-Q water in all recipes and protocol steps.
Phosphatase inhibitor solution, 4× 40 mM sodium pyrophosphate 200 mM NaF 0.4 mM sodium orthovanadate (Na3 VO4 ; see recipe) Divide into 75-μl aliquots Store up to at least 1 year at −20◦ C The sodium orthovanadate should be diluted from the 200 mM stock solution (see recipe).
Pronase stock solution Weigh out a desired amount of Pronase (Roche, cat. no. 10165921001) and dissolve it in dH2 O to a final concentration of 10 mg/ml. Prepare a desired number of 20- to 50-μl aliquots, and store up to 1 year at −20◦ C. Use a new aliquot for each DARTS experiment.
20× Protease inhibitor solution Dissolve one Complete Mini Tablet (Roche, cat. no. 11836153001) in 0.5 ml dH2 O. This solution can be stored frozen at −20◦ C and reused for 3 months.
Sodium orthovanadate (Na3 VO4 ), 200 mM Dissolve sodium orthovanadate to a final concentration of 200 mM in sterile dH2 O. Adjust the pH to 10 using HCl and then boil the solution until it turns translucent. Cool the solution and readjust the pH to 10 using HCl. Divide the solution into aliquots and store frozen up to at least 1 year at −20◦ C (see Gordon, 1991)
TE buffer 10 mM Tris·Cl, pH 8.0 1 mM disodium EDTA Store up to 6 months at 4◦ C
Thermolysin stock solution
Target Identification Using DARTS
Weigh out a desired amount of thermolysin (Sigma, cat. no. P1512) and dissolve it in 1× TNC buffer at a final concentration of 10 mg/ml. Prepare aliquots just as for pronase and use a new aliquot for each experiment.
176 Volume 3
Current Protocols in Chemical Biology
TNC (Tris, NaCl, CaCl2 ) buffer, 10× 500 mM Tris·Cl (pH 8.0) 500 mM NaCl 100 mM CaCl2 Divide into 1-ml aliquots and store up to 1 year at −20◦ C for long-term use. We find that this and other Tris-based buffers are labile over time, and should not be used for more than 1 month after thawing.
Triton X-100 lysis buffer, 2× 0.4% Triton X-100 400 mM NaCl 100 mM Tris·Cl, pH 7.5 20% (v/v) glycerol Store up to 6 months at −20◦ C
COMMENTARY Background Information DARTS was developed as an alternative to the other common affinity-based target identification methods that necessitate modification of the small molecule [reviewed in Lomenick et al. (2011)]. Its primary advantage over these methods is that the native, unmodified small molecule is used, thereby allowing target identification studies without medicinal chemistry (Lomenick et al., 2009). A detailed review of the development of DARTS, the theory on how it works, and comparison to other target identification methods was recently published and is not discussed here (Lomenick et al., 2011).
Critical Parameters and Troubleshooting Preparation of protein lysates The exact lysis buffer and method used to extract proteins for use with DARTS is unlikely to be critical. In most cases, however, it should not be overly harsh, so that the proteins remain in their “native” (binding-competent) state. In general, (1) proteins should be kept cold, (2) lysis buffers should include protease and phosphatase inhibitors, and (3) mixing should be gentle (proteins and detergents should not be vortexed). The temperature should be kept cold and protease and phosphatase inhibitors must be included to prevent premature degradation and other alterations from occurring to the proteins before the DARTS experiment is performed. The sample is warmed to room temperature just before addition of test compounds and room temperature is maintained during the proteolysis to help prevent precipitation of the small molecule and to facilitate binding and enzyme
activity. Although many small molecules will be completely soluble and even capable of binding their target proteins at 4◦ C, others are poorly soluble in aqueous solution and may need fairly high concentrations to reach saturation binding of their targets. Increasing the temperature to a physiological 30◦ C or 37◦ C could also be done to help increase solubility and binding. An exception to the “no vortexing” rule is during the bead beating procedure used to lyse yeast or bacterial cells. This procedure has been compatible with DARTS for the proteins we have studied to date, but alternative lysis buffers for yeast and bacteria that do not require a harsh vortexing step (such as Pierce’s Y-PER and B-PER) could be used, if necessary. Small molecule treatment In most cases, compounds are added directly to the protein lysates of untreated cells. We generally prefer this “in vitro” DARTS approach in order to prevent any small moleculeinduced changes in protein levels (expression or turnover), post-translational modifications (such as those that may signal to the proteasome, aggresome, or caspases and other endogenous proteases), etc. that may occur during the cellular response to the compound. Such changes would not be solely caused by the compound binding to its target protein and may confound the DARTS assay. On the other hand, identification of such changes may still provide valuable information about the compound’s cellular mechanism of action. At least in the cases of didemnin B, rapamycin, and SMER3, the cells could also
Target Identification Using DARTS
177 Current Protocols in Chemical Biology
Volume 3
be treated with the compounds (without readdition of the compounds during or after cell lysis), and protection of the target proteins was equally as effective as when the compounds were added directly to cell lysates of untreated cells (Lomenick et al., 2009). For some weaker binding compounds, however, it may be worthwhile adding the small molecule both to the cells and into the lysate to help ensure saturation binding of the target proteins. Another potential advantage of in situ compound treatment would be to identify target proteins of metabolites of the small molecule. Some compounds may require chemical modifications inside the cell for their biological activity, and such metabolism of the parent molecules may not occur in the cell lysates.
Target Identification Using DARTS
Proteolysis Despite the presence of protease inhibitors in the lysate, which are there solely to prevent the activity of endogenous proteases, the amount of exogenous proteases added into the samples for DARTS is in large excess and overcomes inhibition by the protease inhibitors. At this point in the experiment, even if the in vivo proteases become re-activated (due to potential competitive binding for the protease inhibitors by the DARTS pronase), it should not confound the DARTS assay because the small molecules are already present to protect the target proteins from proteolysis. Whether the proteolysis occurs via pronase or endogenous proteases is not critical even though the majority of protease activity in the sample is due to the exogenous protease. The purpose of the protease inhibitors is to prevent degradation of the proteins during or after cell lysis but before the compounds are added to the lysates. If the target proteins were to become degraded before compound addition, then DARTS would never work. During the proteolysis it is critical that each sample be treated exactly the same to ensure that the amount and time of proteolysis is identical. This is facilitated in the protocol by ensuring that every sample being compared has the same total volume, concentration of protein, salts and other buffer components, and same concentration of DMSO or other solvent. All samples, including the protease, should be thoroughly mixed just before pipetting. Furthermore, the amount of protease added and time of proteolysis for each sample must be identical. This is normally accomplished by spacing the proteolysis of individual samples in 1-min intervals. Alternatively, it should be possible to proteolyse many samples simulta-
neously using a multi-channel pipet (or higher throughput equipment) in a 96-well format. Analysis of results Regardless of the method used to analyze the DARTS results, it is essential to verify that proteolysis of the compound-treated and control samples were comparable. When immunoblotting is used, at least one, but preferably multiple, control proteins should be included in the blot to demonstrate that their extent of digestion is identical. Traditional loading control proteins like tubulin, actin, and GAPDH can be used (except when a “control” itself is targeted by the small molecule of interest, of course), but it is not necessary to be limited to these “housekeeping” proteins. Better controls may include proteins from similar protein families or with similar protease sensitivities to the target proteins. The best control proteins will depend on the amount of protease used. When very high amounts of protease are used, such as pronase to protein ratios of 1:100 to 1:1000, more stable, protease-resistant proteins, such as GAPDH, Hsp90, and pyruvate kinase, can be used as digestion controls. When lower amounts of protease are used, such as pronase to protein ratios of less than 1:1000, these stable proteins will not be digested much or at all and therefore can only serve as loading controls. Actin, tubulin, and elongation factor-1A are less protease-resistant and can usually serve as digestion controls at this range. In addition to blotting for control proteins, digestion levels can be estimated by staining the membrane with Ponceau S or SYPRO Ruby protein blot stains (staining will not interfere with, and therefore can also be performed prior to, immunoblotting). The majority of bands should look identical between samples digested with the same amount of protease. There should also be a clear difference compared to the nondigested control samples, with either less overall protein present in the digested samples or possibly less protein in the higher molecular weight bands and slightly increased amounts of protein at low molecular weights corresponding to partially digested protein fragments. If any proteomics method besides immunoblotting will be used, whether it is gel-based or gelfree, it should be sufficiently quantitative such that the target proteins can be distinguished as enriched versus the majority of nontarget proteins that should be present in equal quantities in the samples (Lomenick et al., 2011).
178 Volume 3
Current Protocols in Chemical Biology
Anticipated Results Basic Protocol 1 will result in a range of proteolysis by pronase within which most proteins will be partially digested at lower pronase amounts, and completely digested at higher pronase amounts (see Fig. 3 for example results). Basic Protocol 2 will likewise result in a range of proteolysis by thermolysin under which many proteins will be partially to completely digested. However, some proteins are insensitive to thermolysin and will not be digested even with very high amounts of the protease. These proteins, such as GAPDH, can serve as loading controls for immunoblotting. To determine if small molecules bind these proteins, pronase or other proteases must be used. The magnitude of protection of a compound’s target protein from proteolysis varies. We have observed from a 2-fold to a 20-fold enrichment of a target protein after proteolysis, but three- to five-fold enrichment is most common. Support Protocols 1 and 2 should yield positive identifications for at least several hundred, but up to several thousand proteins, depending on the protein source, the digestion conditions, and the sensitivity of the mass spectrometer. The majority of the identified proteins should be present in equal quantities in the compound-treated and solvent control samples. Any proteins that are significantly enriched in the compound-treated sample (at least 2-fold increase) should be considered potential targets of the small molecule. Validation of small-molecule binding to these putative targets can then be performed by repeating the DARTS experiment and immunoblotting for the proteins of interest if specific antibodies are available, or by using cDNA constructs of the tagged (e.g., FLAG or HIS tag) protein expressed in vitro or in vivo as a protein source for DARTS, followed by immunoblotting with antibodies that recognize the tag. For Support Protocol 2, it is critical that the results of each fraction be analyzed separately rather than summed together, as this is necessary to identify target proteins whose digestion pattern changes, but which are not completely depleted from the control sample (see Fig. 2).
Time Considerations Once all the reagents are prepared and divided into aliquots, a single DARTS experiment, from preparation of protein lysates to stopping the proteolysis, will usually take 3 to 4 hr. Additional compounds, doses of compounds, or protein lysates could be included
into a single experiment, which would add anywhere from a few minutes to a couple hours to the procedure for an experienced investigator. The time necessary for the remaining analysis, whether it be by SDS-PAGE, immunoblotting, or alternative proteomics techniques, will require a few hours to several days, depending on which method is chosen and the exact protocols used.
Acknowledgements We thank Simon Diep and Melody Pai for critical reading of the manuscript, Gregory Weiss for insightful comments, and our colleagues around the world for their interest and helpful discussions. DARTS has been developed with funding support from the National Institutes of Health (R01 CA124974 and R21 CA149774) and the American Cancer Society (RSG-07-035-01-CCG). B.L. was a trainee of the National Institutes of Health UCLA Chemistry-Biology Interface Predoctoral Training Program (T32 GM008496).
Literature Cited Aghajan, M., Jonai, N., Flick, K., Fu, F., Luo, M., Cai, X., Ouni, I., Pierce, N., Tang, X., Lomenick, B., Damoiseaux, R., Hao, R., Del Moral, P.M., Verma, R., Li, Y., Li, C., Houk, K.N., Jung, M.E., Zheng, N., Huang, L., Deshaies, R.J., Kaiser, P., and Huang, J. 2010. Chemical genetics screen for enhancers of rapamycin identifies a specific inhibitor of an SCF family E3 ubiquitin ligase. Nat. Biotechnol. 28:738-742. Arnold, U., Rucknagel, K.P., Schierhorn, A., and Ulbrich-Hofmann, R. 1996. Thermal unfolding and proteolytic susceptibility of ribonuclease A. Eur. J. Biochem. 237:862-869. Chen, T., Ozel, D., Qiao, Y., Harbinski, F., Chen, L., Denoyelle, S., He, X., Zvereva, N., Supko, J.G., Chorev, M., Halperin, J.A., and Aktas, B.H. 2011. Chemical genetics identify eIF2alpha kinase heme-regulated inhibitor as an anticancer target. Nat. Chem. Biol. 7:610-616. Ghaemmaghami, S., Huh, W.K., Bower, K., Howson, R.W., Belle, A., Dephoure, N., O’Shea, E.K., and Weissman, J.S. 2003. Global analysis of protein expression in yeast. Nature 425:737741. Gordon, J. 1991. Use of vanadate as proteinphosphotyrosine phosphatase inhibitor. Methods Enzymol. 201:477-482. Lomenick, B., Hao, R., Jonai, N., Chin, R.M., Aghajan, M., Warburton, S., Wang, J., Wu, R.P., Gomez, F., Loo, J.A., Wohlschlegel, J.A., Vondriska, T.M., Pelletier, J., Herschman, H.R., Clardy, J., Clarke, C.F., and Huang, J. 2009. Target identification using drug affinity responsive target stability (DARTS). Proc. Natl. Acad. Sci. U.S.A. 106:21984-21989.
Target Identification Using DARTS
179 Current Protocols in Chemical Biology
Volume 3
Lomenick, B., Olsen, R.W., and Huang, J. 2011. Identification of direct protein targets of small molecules. ACS Chem. Biol. 6:34-46. Martzen, M.R., McCraith, S.M., Spinelli, S.L., Torres, F.M., Fields, S., Grayhack, E.J. and Phizicky, E.M. 1999. A biochemical genomics approach for identifying genes by the activity of their products. Science 286:11531155. Ross-Macdonald, P., Coelho, P.S., Roemer, T., Agarwal, S., Kumar, A., Jansen, R., Cheung, K.H., Sheehan, A., Symoniatis, D., Umansky, L., Heidtman, M., Nelson, F.K., Iwasaki, H., Hager, K., Gerstein, M., Miller, P., Roeder, G.S., and Snyder, M. 1999. Large-scale analysis of the yeast genome by transposon tagging and gene disruption. Nature 402:413-418. Washburn, M.P. 2008. Sample preparation and in-solution protease digestion of proteins for chromatography-based proteomic analysis. Curr. Protoc. Prot. Sci. 53:23.6.1-23.6.11. Washburn, M.P., Wolters, D., and Yates, J.R. 3rd. 2001. Large-scale analysis of the yeast proteome
by multidimensional protein identification technology. Nat. Biotechnol. 19:242-247. Wisniewski, J.R., Zougman, A., Nagaraj, N., and Mann, M. 2009. Universal sample preparation method for proteome analysis. Nat. Methods 6:359-362. Wolters, D.A., Washburn, M.P., and Yates, J.R. 3rd. 2001. An automated multidimensional protein identification technology for shotgun proteomics. Anal. Chem. 73:5683-5690. Zhu, H., Bilgin, M., Bangham, R., Hall, D., Casamayor, A., Bertone, P., Lan, N., Jansen, R., Bidlingmaier, S., Houfek, T., Mitchell, T., Miller, P., Dean, R.A., Gerstein, M., and Snyder, M. 2001. Global analysis of protein activities using proteome chips. Science 293:2101-2105.
Internet Resources http://labs.pharmacology.ucla.edu/huanglab/ DARTS faq.html DARTS Help Forum. This provides real-time updated answers and discussions for the frequently asked questions from the community.
Target Identification Using DARTS
180 Volume 3
Current Protocols in Chemical Biology
Analyzing In Vivo Metabolite-Protein Interactions by Large-Scale Systematic Analyses Xiyan Li1 and Michael Snyder1 1
Department of Genetics, Stanford University, Stanford, California
ABSTRACT Metabolites interact with proteins in vivo in various ways other than enzymatic reactions. Profiling of such interactions may help disclose unknown molecular mechanisms that regulate protein functions, and provide potential targets for disease treatment. Here a procedure is described for systematic analyses of metabolite-protein interactions in vivo. This procedure couples protein affinity purification and mass spectrometry to identify metabolite-protein interactions. The primary effort can be completed within 1 day and scaled to process hundreds of samples in a batch. Originally developed in yeast, the same principles and protocol can be adapted to other organisms. Curr. Protoc. Chem. Biol. C 2011 by John Wiley & Sons, Inc. 3:181-196 Keywords: metabolite-protein interaction mass spectrometry r LC-MS r metabolite
r liquid chromatography r r protein affinity purification r yeast
INTRODUCTION In the cell, metabolites are a group of small compounds whose presence and levels are often dynamically regulated during the process of growth and differentiation, as well as in a diversity of biological responses. Although typically regarded primarily as simple building blocks and energy sources within a cell, metabolites may also interact with proteins in various ways other than enzymatic reactions, many of which can regulate protein functions, such as modulation of enzyme activity and regulation of transcription, signaling, and neural transmission. Profiling of such in vivo interactions provides important clues to the molecular mechanisms that modulate protein functions through physical interaction. A systematic procedure has been developed to study in vivo metaboliteprotein interactions in yeast, which may be adapted to other organisms, as long as certain criteria are met (see Commentary; also see Li et al., 2010). No prior knowledge of the protein of interest is required for this assay. The whole procedure involves protein expression, protein affinity purification, metabolite extraction, liquid chromatography-coupled mass spectrometry (LC-MS), and extensive data analysis. Depending on the throughput of mass spectrometry, this procedure can be conveniently scaled up to process several hundred samples at once.
STRATEGIC PLANNING A general strategy for studying metabolite-protein interactions is shown in Figure 1. The basic procedure involves purification of a protein of interest, elution of the bound small molecules, and separation and identification of small molecules using LC-MS. Several factors are crucial for success. First, an appropriate protein expression system is crucial to produce biologically meaningful and reliable data in this experiment. Since there can be potential metabolomic variation between species and experimental conditions, proteins of interest should be produced in their natural hosting cells whenever possible. Second, rapid purification of the protein using either an epitope tag or capture agent is
Current Protocols in Chemical Biology 3: 181-196, December 2011 Published online December 2011 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/9780470559277.ch110193 C 2011 John Wiley & Sons, Inc. Copyright
Analyzing In Vivo MetaboliteProtein Interactions
181 Volume 3
sample
control UPLC-APCI-MS
bead
LC
TIC
lgG beads
time %
lysis
MS
* *
*
bead m/z 100-1500 wash
SDSPAGE
1) methanol extraction
lgG
2) SDS extraction
Figure 1 Flowchart for the identification of small metabolites bound to proteins. Yeast proteins tagged with an IgG-binding protein domain are isolated from lysates using magnetic beads coated with IgG. After washing, the small metabolites are extracted in organic solvent and analyzed using liquid chromatography-coupled mass spectrometry (LC-MS). The purified proteins adsorbed to magnetic beads are later extracted with SDS sample buffer and analyzed by SDS-PAGE. A yeast strain lacking the fusion protein is used as the negative control in parallel experiments. The metabolites significantly enriched in the fusion protein sample relative to the negative control are scored as protein-bound metabolites. Reprinted from Li et al. (2010) with permission.
Protein Expression and Purification Design Sufficient protein production Efficient protein purification Negative control
Selection of LC-MS Methods Metabolite extraction solvent LC columns LC mobile phases MS probe MS acquisition mode
Sample Collection, Storage and Processing Cell harvesting Storage temperature and duration Processing batch
Figure 2
Strategic considerations for studying metabolite-protein interactions.
Analyzing In Vivo MetaboliteProtein Interactions
182 Volume 3
Current Protocols in Chemical Biology
A
100 80
UPLC phenyl NL: 1.99E7
60 40 20
Relative abundance (%)
0 100 80
UPLC T3 low pH NL: 1.70E7
60 40 20 0 100 80
UPLC C18 NL: 1.77E7
60 40 20 0 100 80
UPLC amide NL: 1.42E7
60 40 20 0 0
2
4
6
8
10
12
14
16
Time (min)
B
100
APCI Low pH Positive mode NL: 1.31E7
50
0 100
ESI High pH Negative mode NL: 1.44E7
Relative abundance (%)
50
0 100
ESI High pH Positive mode NL: 2.06E7
50
0 100
ESI Low pH Negative mode NL: 2.38E7
50
0 100
ESI Low pH Positive mode NL: 1.74E7
50
0 0
2
4
6
8
10
12
14
16
Time (min)
Figure 3 The influence of different LC-MS methods on mass spectral patterns. The same yeast metabolite extract was used for all conditions at a scan range of 85 to 1200 m/z. NL indicates the maximum intensity. All plots are base peak intensity (BPI) mass spectra. (A) The influence of UPLC columns on mass spectra. All mass spectra were acquired in positive mode with mobile phases at pH 4.25. (B) The influence of MS probes, the pH of mobile phases, and polarity on mass spectra. APCI or ESI probes, high pH (9.75), or low pH (4.25), and polarity are indicated. Analyzing In Vivo MetaboliteProtein Interactions
183 Current Protocols in Chemical Biology
Volume 3
valuable. The former is typically used for high-throughput analyses, and we often tag proteins using the ZZ domain of protein A which binds IgG beads with very high affinity (Lowenadler et al., 1987; Nilsson et al., 1987). The ZZ domain consists of 116 amino acid residues, and the whole tag is about 19 kDa (Gelperin et al., 2005). However, caution must be exercised when using ZZ-tagged proteins in immunoassays, because the ZZ domain interacts strongly with most primary antibodies raised against specific proteins. A summary of additional important factors in experimental design is described in Figure 2 and also discussed in detail below. Another important parameter is the choice of an appropriate LC-MS method, the most challenging part of this experiment. Because metabolites have enormous chemical diversity, it is not possible to analyze them using one general LC-MS method. Several different methods must be used to analyze as many metabolites as possible (as described in Fig. 3). Two simple methods are described here for the general analysis of hydrophobic and hydrophilic metabolites, respectively. However, more sensitive methods are possible by focusing on analyzing a particular group of metabolites at the same time (e.g., hydrophobic or hydrophilic molecules). BASIC PROTOCOL 1
AFFINITY PURIFICATION OF YEAST PROTEIN AND EXTRACTION OF PROTEIN-INTERACTING METABOLITES Protein A (ZZ domain)–tagged protein is adsorbed on rabbit IgG–conjugated magnetic beads (Dynabeads) in a solution compatible with mass spectrometry. The protein-bound metabolites are then extracted for LC-MS analyses. The protein yield is also assessed by gel staining afterwards. A graphic summary this procedure is described in Figure 1.
Materials Rabbit IgG–conjugated Dynabeads (see Support Protocol 1) Lysis buffer (see recipe) with and without DTT and protease/phosphatase inhibitors Zirconia silica beads (Biospec) Cell pellet (stored at −80◦ C, see Support Protocol 2) Wash buffer 1 (see recipe) Wash buffer 2 (see recipe) 75% (v/v) ethanol (mass spectrometry grade) 15-well 4-12% NuPAGE Bis-Tris gel (Invitrogen) 2× Laemmli SDS sample buffer (for SDS-PAGE; see recipe) Page Ruler Plus prestained protein ladder (Fermentas) 20× NuPAGE MOPS SDS running buffer (Invitrogen, cat. no. NP0001-02) ProtoBlue safe colloidal Coomassie staining solution (National Diagnostics, cat. no. EC-722) Gel drying solution (see recipe)
Analyzing In Vivo MetaboliteProtein Interactions
Refrigerated centrifuge Hula mixer (Invitrogen) or similar product Magnetic stand for 1.5/2.0 ml tubes (Invitrogen) Eppendorf protein LoBind tubes (2.0 ml and 1.5 ml) FastPrep cell lyser with an adapter for 2-ml tubes (Qbiogene) Heat blocks maintained at 42◦ C and 95◦ C Glass mass spectrometry vials with inserts Gel-drying frame Cellophane membrane for gel drying Gel scanner Additional reagents and equipment for SDS-PAGE (e.g., Gallagher, 2006)
184 Volume 3
Current Protocols in Chemical Biology
NOTE: Steps 1 to 8 should be carried out in a cold room at 4◦ C. NOTE: Non-filtered pipet tips should be used to avoid introducing polymers that are often found in filters. NOTE: Nitrile gloves are preferred to create a cleaner background on LC-MS. 1. Wash an appropriate amount of rabbit IgG–conjugated Dynabeads three times in lysis buffer without DTT or protease/phosphatase inhibitors. To do this, each time separate beads from the supernatant on a magnetic stand, discard supernatant, add lysis buffer, and then mix by inversion on a Hula mixer at 20 rpm for 3 min at 4◦ C. Briefly microcentrifuge for 5 to 10 sec before separation on the magnet to help recover beads. After the last wash, resuspend the beads in lysis buffer and aliquot into clean 2-ml LoBind tubes for step 3. It is convenient to use 50 μl (∼1.25 mg Dynabeads) of IgG Dynabeads per sample. This will yield a sufficient amount of protein for visualization by Coomassie blue staining. More Dynabeads can be used to obtain higher amounts of protein.
2. Add an equal volume of 0.5 mm zirconia/silica beads (stored at –20◦ C) to a yeast cell pellet prepared from 150 ml culture as described in Support Protocol 2 (grown to OD600 = 1.0). Add 950 μl lysis buffer. Lyse on FastPrep for 3 × 40 sec at 6.0 m/sec, placing the sample on ice for 2 min between homogenization runs. Cell pellets can be stored at −80◦ C for 6 to 12 months with no apparent loss of yield. Glass beads (1.0 mm) or ceramic beads (1.4 mm, if milder lysis condition is desired) may be used with comparable performance.
3. Centrifuge at 20,000 × g for 10 min at 4◦ C, and transfer the supernatant (lysate) to a 2.0-ml LoBind tube. Store at 4◦ C. Be careful not to transfer cell debris that sometimes layers above zirconia/silica beads.
4. Add 950 μl lysis buffer to the cell pellet and perform lysis again as in step 2. 5. Repeat step 3 and combine and transfer the lysate into the 2-ml LoBind tube containing IgG Dynabeads. Be careful not to transfer cell debris that sometimes layers above zirconia/silica beads.
6. Mix by inversion for 30 to 45 min at 4◦ C on a Hula mixer at 20 rpm. The mixing speed should be fast enough to prevent Dynabeads from settling to the side. A brownish suspension is expected in this step.
7. Separate and remove the lysate in the supernatant on a magnetic stand. Briefly microcentrifuge 5 to 10 sec before separation on magnet to help recover beads. Add 950 μl wash buffer 1 to the beads, mix on a mixer for 5 min at 4◦ C. This step washes off most nonspecific proteins in the high-molecular-weight range.
8. Separate and remove wash buffer 1 in the supernatant on a magnetic stand. Briefly microcentrifuge 5 to 10 sec before separation on magnet to help recover beads. 9. Add 950 μl wash buffer 2, transfer bead suspension to a fresh 1.5-ml LoBind tube. Mix by inversion on a Hula mixer for 5 min at 4◦ C. A fresh tube is used to eliminate nonspecific adsorption of metabolites to the old tube.
10. Separate and remove wash buffer 2 in the supernatant on a magnetic stand. Add 60 μl 75% ethanol (MS grade) to the beads, mix by pipetting up and down, and heat at 42◦ C on a heat block for 15 min.
Analyzing In Vivo MetaboliteProtein Interactions
185 Current Protocols in Chemical Biology
Volume 3
The high organic composition of 75% ethanol is used to disrupt the metabolite-protein interaction and dissolve the protein-bound metabolites while leaving behind proteins and other unwanted polymers. 75% ethanol serves as a general solvent that is suited for most metabolites. Ethanol can be replaced with pure methanol for most hydrophobic metabolites. Other solvents are also possible but yet have not been tested. The volume of extraction solvent can be modified according to the requirements for the mass spectrometer. Each sample should have at least three technical replicates for each LC-MS method to allow for statistical analyses.
11. Briefly spin the tubes for 10 sec. Separate the beads on a magnetic stand, then transfer the metabolite extract in the supernatant into a MS vial with glass insert. The extract is ready for LC-MS analysis. The authors use 150-μl glass inserts in 12 × 32–mm brown glass vials with preslit PTFE/silicon septa screw caps for metabolite extract and mass spectrometry. LC-MS analysis of the samples should ideally be performed within 1 to 2 days after sample preparation. On some occasions, the samples can be stored at −20◦ C or −80◦ C for up to 1 month without losing certain metabolites.
12. Add 30 μl of 2× Laemmli SDS sample buffer to the beads, flick the tube to bring the beads down into sample buffer, then heat for 15 min on a heat block at >95◦ C. In this step, the protein of interest is stripped off the beads for abundance assessment. Random leakage of IgG heavy chain and light chain into the sample buffer is expected if no BS3 cross-linking is used after IgG Dynabeads conjugation (see Basic Protocol 2), because β-mercaptoethanol disrupts disulfide bonds that are formed between the heavy chains and light chains. Alternatively, nonreducing elution solutions or BS3-cross-linking Dynabeads may be used to avoid this leakage problem.
13. Briefly centrifuge the tubes for 5 sec after cooling the samples at room temperature. Separate on a magnetic stand, load 15 μl per well onto a 15-well 4% to 12% NuPAGE gel, and electrophorese 0.5 hr at 120 V and 1 hr at 150 V using NuPAGE MOPS SDS running buffer and Page Ruler Plus prestained protein ladder. See Gallagher (2006) for general SDS-PAGE protocols.
14. Stain the gel with ProtoBlue Safe Colloidal Coomassie staining solution overnight at room temperature and destain in water according to manufacturer’s instruction. 15. Scan the gel with a minimum resolution of 600 dpi to assess protein yield. A dry gel can be prepared for long-term preservation: soak the gel in gel drying solution for 15 to 30 min. Make a dry gel using two layers of cellophane for each gel. Scan the gel to assess protein abundance. SUPPORT PROTOCOL 1
Analyzing In Vivo MetaboliteProtein Interactions
CONJUGATION OF IMMUNOGLOBULIN TO DYNABEADS Rabbit IgG is conjugated to epoxy Dynabeads via covalent interaction between the epoxy group on the beads and the amine or thiol group on the proteins. In some cases the conjugated beads can be further cross-linked to avoid IgG leakage into the eluate (Fig. 4). The conjugated beads are later used for affinity purification of TAP-tagged proteins that have the ZZ domain of protein A. The performance of Dynabeads is superior to agarose beads because of their high surface area and low background binding.
Materials 10 mg (11 mg/ml) rabbit IgG (ChromaPure) in 0.01 mM sodium phosphate buffer, pH 7.6/0.25 M NaCl (Jackson ImmunoResearch, cat. no. 011-000-003) Bradford protein assay kit and bovine gamma globulin as standard
186 Volume 3
Current Protocols in Chemical Biology
Epoxy Dynabeads (Invitrogen, 300-mg size, cat. no. 143.02D) Dynabeads antibody coupling kit (Invitrogen, cat. no. 143.11D) containing C1, C2, LB, HB, SB solutions 143 mM (∼50 mg/ml) n-dodecyl glucoside (nDG, Sigma, cat. no. D8035-1g) in methanol for downstream mass spectrometry application 0.05% (v/v) Tween-20 Sodium azide BS3 (Bis(sulfosuccinimidyl) suberate (Thermo Scientific, cat. no. PI-21580, 21585, or 21586; optional) BS3 conjugation buffer: 20 mM HEPES or 20 mM NaPO4 /0.15 M NaCl (pH 7.5) BS3 quenching buffer: 1 M Tris·Cl (pH 7.5) Refrigerated microcentrifuge Magnetic stand for 1.5/2.0 ml tubes (Invitrogen) Eppendorf protein LoBind tubes (2.0 ml and 1.5 ml) 15-ml conical centrifuge tubes (e.g., BD Falcon) Hula mixer (Invitrogen) or similar product
1
2
3
1*
2*
3*
50
75
110
50
75
110
185 115 80
65 50
30 25
15 10 kDa:
Figure 4 Proteins of different sizes can be purified using rabbit IgG-conjugated Dynabeads. The expected size of each protein is indicated at the bottom of each lane. Molecular size markers are on the left. The gels were stained with colloidal Coomassie Blue G250. Lanes 1 to 3, proteins purified on regular IgG Dynabeads. Lanes 1* to 3*, the same proteins purified on BS3-cross-linked IgG Dynabeads. Note the IgG leakage at 55 kDa and 25 kDa are absent in these lanes. Equal amounts of yeast cells were used for each protein purification.
Analyzing In Vivo MetaboliteProtein Interactions
187 Current Protocols in Chemical Biology
Volume 3
Day 1 1. Microcentrifuge the IgG solution 10 to 15 min at 20,000 × g, 4◦ C. Only use the supernatant for the following steps. Determine protein concentration by Bradford assay using bovine gamma globulin (BGG) as standards. Make sure that the IgG stock solution does not contain any reagents, such as Tris, that have amine or thiol groups. Centrifugation helps remove IgG aggregates that may interfere with the protein purification. BGG provides a better estimation of the IgG concentration than does BSA.
2. Resuspend the epoxy Dynabeads (300 mg) in 3 ml C1 solution from the coupling kit. Split and transfer to two 1.5- or 2.0-ml LoBind tubes. Briefly microcentrifuge, then separate on a magnetic stand. Remove the supernatant. 3. Wash the beads with another 1.5 ml C1 solution in each tube for 2 min, separate on a magnetic stand, briefly centrifuge, and remove the supernatant. 4. Resuspend the beads in each tube with 2 ml C1 solution and transfer the beads from each tube into a new 15-ml conical centrifuge tube. 5. To each tube, add 450 μg rabbit IgG and appropriate volume of C1 (a total of 2 ml of IgG and C1). Put on a Hula mixer at 20 rpm for 1 min. 6. Add 4 ml C2 to each 15-ml conical tube, then mix by inversion on a Hula mixer at 20 rpm for 1 min. 7. Wrap the tube with aluminum foil. Incubate on a mixer (20 rpm) at 37◦ C for 16 to 24 hr. The cross-link efficiency typically varies between 75% and 85%.
8. Add 25 μl of 143 mM nDG or 0.05% Tween-20 to 25 ml SB, then mix on a Hula mixer at 4◦ C overnight. nDG does not suppress the ionization of mass analytes on ESI mass spectrometry, but is a weaker detergent to remove noncovalent binding in following steps, and is more difficult to handle due to its poorer solubility (compared to Tween-20) in water.
Day 2 9. Split the suspension from the two 15-ml conical tubes into six 2-ml LoBind tubes. Remove the supernatant on a magnetic stand. Save 1 ml of supernatant in each of steps 9 to 14 to calculate conjugation efficiency. 10. Add 1.6 ml HB solution (from coupling kit) to each tube. Mix for 1 min (or until homogeneous) on the Hula mixer, separate on a magnetic stand, and remove the supernatant. 11. Add 1.6 ml LB solution (from coupling kit) to each tube. Mix for 1 min, or until homogeneous, on the Hula mixer, separate on a magnetic stand, and remove the supernatant. 12. Add 1.6 ml SB with nDG (from step 8) to each tube. Mix for 1 min or until homogeneous on the Hula mixer, separate on a magnetic stand, and remove the supernatant. Analyzing In Vivo MetaboliteProtein Interactions
13. Add 1.6 ml SB with nDG (from step 8) to each tube. Mix 15 min on the Hula mixer, separate on a magnetic stand, and remove the supernatant. 14. Repeat step 13.
188 Volume 3
Current Protocols in Chemical Biology
15. Resuspend the beads in SB to a total volume of 2 ml for each tube of beads. Split into 2 tubes, each with 1 ml. Microcentrifuge 3 sec to pellet the beads. Store at 4◦ C until use, for up to a month. Sodium azide may be added to 0.02% for long-term storage (up to 1 year).
16. Determine conjugation efficiency by measuring IgG concentration with Bradford assay, using BGG as standards. Desalting may be required for other protein assays. A typical dilution is as follows: Loading IgG, 1000× dilution (>10 mg/ml). Flowthrough, HB, LB, and SB wash, 5× dilution. Standard bovine gamma globulin range uses 1.25to 20 μg/ml.
Optional: cross-linking conjugated beads This procedure is useful when IgG leakage happens during protein purification and interferes with other applications. It can be performed at a later time after bead conjugation. Sodium azide must be removed by washing or not used in previous steps. 17. Wash each tube of 1 ml IgG-coupled Dynabeads three times, each time on a Hula mixer (20 rpm) for 5 min in 1.8 ml conjugation buffer. Separate on the magnetic stand for 30 sec, then remove supernatant. 18. Dissolve 50 mg BS3 in 18 ml freshly prepared BS3 conjugation buffer to make 5 mM solution. Prepare fresh BS3 solutions every time. Discard unused solution.
19. Add 1.5 ml 5 mM BS3 solution to the washed Dynabeads in 2-ml LoBind tubes. 20. Incubate at room temperature for 30 min with tilting/rotation on Hula mixer (20 rpm). 21. Quench the cross-linking reaction by adding 75 μl BS3 quenching buffer. 22. Incubate at room temperature for 15 min with tilting/rotation. 23. Wash the cross-linked Dynabeads three times with 1.5 ml SB each. Add equal start volume of SB for storage at 4◦ C before use. The beads are ready to use, and little IgG leakage into the eluate should be observed afterwards.
YEAST GROWTH AND CELL COLLECTION The authors have typically used MORF yeast strains, each hosting a galactose-inducible protein expression construct (Gelperin et al., 2005). Healthy cells at mid-logarithmic growth stage in raffinose medium are subject to galactose induction of target gene expression. Cells are washed and stored at −80◦ C until analysis.
SUPPORT PROTOCOL 2
Materials Yeast strain SC-URA solid medium plates with glucose (see recipe) SC-URA liquid medium with glucose or raffinose (see recipe) 3× YP/Gal (see recipe) 30◦ C incubator 1-liter flask Platform shaker capable of accommodating 1-liter flasks
Analyzing In Vivo MetaboliteProtein Interactions
189 Current Protocols in Chemical Biology
Volume 3
Spectrophotometer 500-ml centrifuge bottles Centrifuge with JA-10 or SLA-3000 rotor 15-ml conical centrifuge tubes 2-ml thick-wall screw-cap tubes 1. Streak yeast cells from glycerol stocks on fresh SC-URA plates, and incubate 2 to 3 days at 30◦ C. 2. Pick fresh single colonies and inoculate into 3 ml liquid medium (SC-URA/glucose) and grow overnight (this is conveniently done after 3 pm) at 30◦ C with shaking. If the plates have ever been stored in cold room, restreak single colonies on fresh plates and wait for another 1 to 2 days for yeast colonies to form.
3. Using a spectrophotometer, check the optical density at 600 nm (OD600 ) of the yeast culture. Use fresh medium as blank, and make a 1:10 dilution of the starter culture.
4. Calculate the volume to be added to 200 ml SC-URA/raff (in 1 liter): V(ml) = 2.75/OD600 V is usually <0.5 ml.
5. Prepare the 200-ml culture in a 1-liter flask and shake for roughly 15.5 hr at 30◦ C (critical for mid-log phase control). 6. Measure OD600 using 0.8 ml culture. At this point, absorbance in mid-log phase should be 0.5 to 0.9.
7. Add 100 ml 3× YP/Gal (pre-warmed at 30◦ C) to each flask, and shake for another 5.5 to 7 hr. 8. Collect cells in 500 ml centrifuge bottle (for JA-10 or SLA-3000 rotor) by centrifugation for 5 min at 1500 × g, 4◦ C. 9. Decant supernatant, resuspend cells in 10 ml ice-cold sterile distilled deionized water, and transfer to a 15-ml conical centrifuge tube. 10. Pellet cells by centrifugation at 1500 × g for 5 min, 4◦ C. 11. Remove supernatant. Wash cells again by adding ice-cold sterile distilled deionized water to a total volume of 3.8 ml, mix and, transfer 1.9 ml each into two 2-ml thick-wall screw-cap tubes. 12. Centrifuge 5 min at maximum speed, 4◦ C. Remove supernatant. 13. Quickly freeze cell pellets on dry ice and store up to 1 year in −80◦ C freezer. BASIC PROTOCOL 2
Analyzing In Vivo MetaboliteProtein Interactions
LC-MS OF PROTEIN-BOUND METABOLITES Due to the rapid development and technical diversification of mass spectrometer models, only a general method is described here. It is recommended to consult a mass spectrometry expert for the best solution available. Because of the explorative nature of this assay, a full scan of 80 to 1600 amu is sufficient for initial profiling. Quantitative multiple reaction monitoring (MRM) can be introduced at a later stage for verification and validation.
190 Volume 3
Current Protocols in Chemical Biology
Materials Metabolite samples (ideally freshly prepared within 1 to 2 days; see protocols above) Mobile phase solutions: gradient elution of LC often uses 10% and 90% acetonitrile in water for ESI, and 10% to 100% methanol for APCI; buffer reagents, such as 10 mM ammonium acetate, can be added to improve LC peak shape stability—pH can be adjusted with acetic acid or ammonium hydroxide as needed to help ionization of analytes Strong wash, weak wash, and needle wash solutions: weak and strong wash solutions are the same as the start and end mobile phases; for example, 10% and 90% acetonitrile in water can be used as weak and strong wash solutions for a reversed-phase gradient—needle wash is similar to weak wash; do not use buffer or salt reagents in wash solutions Reversed-phase UPLC columns for low polarity to very hydrophobic metabolites: UPLC C18, hexyl/phenyl, or C8 columns from Waters Normal-phase or HILIC columns for polar hydrophilic metabolites: UPLC Amide, T3, or other columns from Waters Waters Acquity UPLC-coupled Thermo Exactive Orbitrap mass spectrometer, equipped with an electrospray ionization (ESI) probe or an atmospheric pressure chemical ionization (APCI) probe NOTE: ESI and APCI complement each other to expand the detection scope. APCI is especially suitable to detect thermostable nonpolar hydrophobic molecules with a molecular weight below 1000 Da. APCI is preferred when it can detect the metabolites of interest. ESI has broader detection coverage in chemical properties and molecular weight than APCI, yet it requires optimization of many parameters for sound performance. 1. Calibrate the mass spectrometer with standard calibrants and prime the LC system with LC solvents. It may take at least 1 to 2 hr before the performance of mass spectrometer is stable.
2. Install the UPLC column, and run at least three to five gradients with blank injections of weak wash solvent. This step stabilizes column performance.
3. Run at least three replicates for each sample. Uninterrupted running is strongly advised for batch processing of all samples in one study. The running time for each injection depends on the flow rate and column size. A typical gradient usually includes a wash of 10 column bed volumes, followed by 5 to 10 column bed volumes of weak wash to stabilize the column.
4. After acquisition, wash and store the column with solvents without buffer or salt reagents. Inclusion of organic solvent at 10% or higher helps prevent microbial growth.
LC-MS DATA PROCESSING USING XCMS In addition to proprietary software packages from mass spectrometer vendors, many analysis programs are freely available in academia. The authors routinely use XCMS for data processing. XCMS is a free R package that has been developed to streamline LCMS-based metabolomics analyses (Smith et al., 2006). Raw LC-MS data are converted to centroid CDF or mzXML format before XCMS analysis. Technical replicates are placed into one folder for each sample. The following is a script running in R, a freely available
SUPPORT PROTOCOL 3
Analyzing In Vivo MetaboliteProtein Interactions
191 Current Protocols in Chemical Biology
Volume 3
statistics software package. It works under R 1.13.0 with XCMS build 1.26.0. For more information see (http://metlin.scripps.edu/xcms/).
## install XCMS update.packages() source("http://bioconductor.org/biocLite.R") biocLite("xcms") rm(list=ls(all=TRUE)) ## load xcms, set up sample list and file path library(xcms) ## change the following file path to the folder where all those CDF files are saved. mzXMLpath="X:/MassSpecData" setwd="X:/MassSpecData" list.files("mzXML",path=mzXMLpath, recursive=TRUE, full.names=TRUE) mzXMLfiles<-list.files("mzXML",path=mzXMLpath, recursive=TRUE, full.names=TRUE) ## extract mass features in positive mode from the raw data # type "?findPeaks.centWave" in R for parameter settings for centWave xset<-xcmsSet(mzXMLfiles, method=’centWave’, ppm=2.5, peakwidth=c(15,90), snthresh=3, prefilter=c(3,260), integrate=2, mzdiff=0.00005, noise=260, verbose. columns=TRUE,fitgauss=TRUE,mzCenterFun="wMean", polarity="positive", nSlaves=4) xset<-group(xset, bw=10, minfrac=0.6) xset ## Align the retention time of mass features detected among samples #xset2<-retcor(xset, method="obiwarp", plottype="deviation") xset2 ## Fill blanks in samples without a particular mass feature xset2<-group(xset2, bw=10) xset3<-fillPeaks(xset2) xset3 ## Write out the peak table for all samples pt <- peakTable(xset3) write.table(pt,file="peaktable.txt",sep="\t", row.names=F) ## Statistic comparison of two samples, and write out the table reporting P value and fold enrichment reporttab1<-diffreport(xset3,"name of control", "name of sample", "name of sample.txt") The accurate mass reported in the “mz” column for each mass feature can be used to search for possible hits on metabolite databases such as METLIN (http://metlin.scripps.edu) and HMDB (http://www.hmdb.ca). Analyzing In Vivo MetaboliteProtein Interactions
192 Volume 3
Current Protocols in Chemical Biology
REAGENTS AND SOLUTIONS Use double-deionized, distilled water in all recipes and protocol steps.
Gel drying solution 25% (v/v) ethanol 5% (v/v) glycerol Store up to 2 years at room temperature 2× Laemmli sample buffer Add 50 μl β-mercaptoethanol to 950 μl 2× Laemmli sample buffer (Bio-Rad, cat. no. 161-0737). Vortex to mix. Store up to 1 week at room temperature.
Lysis buffer Prepare the following in mass spectrometry–grade H2 O: 200 mM ammonium acetate (prepare from 5 M stock; Sigma, cat. no. 73594) 5 mM EGTA (prepare from 500 mM stock) 5 mM EDTA (prepare from 500 mM stock) 1× nDG (prepare from 1000× stock: 50 mg/ml n-dodecyl glucoside; Sigma, cat. no. D8035) 1 mM DTT (prepare from 1 M stock; omit for washing IgG Dynabeads) 1× Halt protease and phosphatase inhibitor (Pierce; omit for washing IgG Dynabeads) Prepare fresh on day of use SC-URA liquid medium and plates 6.7 g/liter yeast nitrogen base without amino acids (BD Difco) 1.92 g/liter yeast synthetic drop-out medium (Sigma, cat. no. Y1501) 20 g/liter agar (for solid medium) Autoclave Add 2% (w/v) carbon source (glucose or raffinose) Store up to 6 months at 4◦ C Carbon source stock solution (20% w/v) is sterilized and added after medium autoclaving. Add 20 g/liter agar before autoclaving for solid medium.
Wash buffer 1 Prepare the following in mass spectrometry–grade H2 O: 500 mM ammonium acetate (prepare from 5 M stock; Sigma, cat. no. 73594) 1× nDG (prepare from 1000× stock: 50 mg/ml n-dodecyl glucoside; Sigma, cat. no. D8035) Prepare fresh on day of use Wash buffer 2 Prepare the following in mass spectrometry–grade H2 O: 50 mM ammonium acetate (prepare from 5 M stock; Sigma, cat. no. 73594) 1× nDG (prepare from 1000× stock: 50 mg/ml n-dodecyl glucoside; Sigma, cat. no. D8035) Prepare fresh on day of use 3× YP/Gal 3% w/v yeast extract 6% w/v peptone 6% w/v galactose For 1 liter, autoclave 700 ml of 3× YP, then add 300 ml filter-sterilized 20% galactose.
Analyzing In Vivo MetaboliteProtein Interactions
193 Current Protocols in Chemical Biology
Volume 3
COMMENTARY Background Information Metabolite-protein interaction represents another layer of complexity in biological interaction networks. Many have been reported adventitiously in the past, including some classical examples such as cofactors in enzymology, nutrient-sensing operons and sterolsensing mechanisms in the regulation of transcription, second messengers in signaling, and neurotransmitters in synapses. A systematic approach was not possible until recent advances in genomics and proteomics that allow the prediction and production of virtually all proteins in one organism. Several studies have since disclosed extensive and degenerate metabolite-protein interactions in vivo and in vitro (Morozov et al., 2003; Gallego et al., 2010; Li et al., 2010). These degenerate patterns may bridge gaps between other types of interaction networks and provide novel insights into how cells function as a system (Li and Snyder, 2011). Compared to proteins, the small size of metabolites often preclude their use as baits, as immobilized metabolites would likely sterically hinder many metabolite-protein interactions. Proteins are thus used as baits in the aforementioned studies. Assays to assess metabolite-protein interaction, which require incubation of purified protein and metabolite mixture in vitro, using protein affinity purification (Tagore et al., 2008) or protein microarrays (Zhu et al., 2001; Morozov et al., 2003; Gallego et al., 2010), have been described. These studies have high throughput only when combined with protein microarrays and preselected metabolite targets. However, they might suffer from undesired artifacts, since it can be very challenging to mimic the exact in vivo binding environment and physiological ranges of metabolite composition. Furthermore, unexpected interactions cannot be detected because the detection scope is limited by the metabolites used. Here, protein purification is directly coupled with downstream mass spectrometry for fast and scalable experimental practice and unbiased metabolite profiling. This method has been successfully used to profile both hydrophobic and hydrophilic molecules.
Critical Parameters and Troubleshooting Analyzing In Vivo MetaboliteProtein Interactions
Sufficient protein input is necessary to detect stoichiometrically protein-interacting metabolites. Because picomoles of metabolites can be reliably detected by mass spec-
trometry, the biomaterials for each experiment should be sufficient to purify at least 1 picomole of the protein of interest (around 100 ng for a 100-kDa protein). To minimize variation in the performance of LC-MS, it is advised that the entire batch of samples be analyzed with the same LC-MS method and in the same run. Frequent switching of LC-MS setups may result in unwanted migration in retention time and altered MS sensitivity. It is important not to inadvertently introduce small-molecule contamination during experiments. Handle samples with nitrile gloves and in ultra-clean tubes. Always use the highest grade reagents available. Avoid unnecessary use of polyethylglycol polymer–based detergents and membrane filters. Analyzing LC-MS data can be daunting despite the many software tools available for metabolomic profiling. While different peak-calling algorithms, peak-alignment algorithms, and statistical filters may apply, it is advisable to keep in mind that, in this assay, an all-or-none or infinite enrichment of metabolites in test samples versus the negative control is much more desired when compared with genuine metabolome-profiling studies. The parameters in data processing have to be adjusted according to real data for best performance.
Anticipated Results Protein-interacting metabolites can be scored by the presence and enrichment of a reproducible LC-MS feature with retention time and accurate mass measurement within several parts-per-million (ppm) of the theoretical values. The monoisotopic mass peak pattern should also match the theoretical values within a range of 20%. These true positive mass features are often not obvious to visual inspection because their abundance is much lower than most matrix mass features arising from solution ingredients. Follow-up experiments Several strategies can be used for validation after general profiling of protein-associating metabolites. The first two can be systematically carried out without prior knowledge, while the third requires expertise in related fields. First, a metabolite standard, if available, can be analyzed under identical LC-MS conditions for matching retention time and monoisotopic and adduct patterns with the analytes of
194 Volume 3
Current Protocols in Chemical Biology
interest. This is the first and most straightforward step one can take to verify the identities of protein-binding metabolites. To minimize the matrix effect on retention time (such as salt, solvent, etc.), it would be better to prepare standard in similar solutions used for the real experiment. Additionally, fragmentation patterns of the target metabolites can be acquired by LC-MS operating in multiple reaction monitoring (MRM) in repeated biological experiments, and used to search metabolite databases for matching fragment distribution. Second, in vitro binding assays between purified protein and metabolite standards can be used to characterize binding affinity and stoichiometry. A saturable binding curve is expected for specific binding while linear binding often indicates the binding is nonspecific. An affinity constant within the physiological range of the binding metabolite and low metabolite-to-protein binding ratio often suggest that the interaction is consequential. A desalting column that removes small molecules (<7000 Da) can be used to remove non-binding metabolites in this assay, as described in Li et al. (2010) and Li and Snyder (2011). Caution must be taken during protein purification to strip off endogenously bound metabolites by extensive washing to make the binding sites available for this in vitro binding assay. This assay may not work if the metabolite-protein binding requires energy molecules (e.g., ATP) or other chaperones. Third, the biological significance of metabolite-protein binding can be explored by determining changes in the function of target proteins after genetic or pharmacological alteration of target metabolite levels in cells. For example, target metabolite synthesis can be eliminated by knocking out the enzymeencoding gene(s) or reduced by introducing specific inhibitors of the enzymes. Conversely, the levels of the target metabolite can be compared between cells expressing its interacting protein at normal, excessive, and reduced (or absent) levels. Furthermore, it is also possible to confirm predicted metabolite-protein interactions by introducing a chemically modified metabolite or its precursor into the biological system (e.g., cell culture) for subsequent interaction analysis. The predicted binding event, as supported by this type of method, could be a consequence of indirect association. Even in such cases, the results could still be biologically meaningful, as long as manipulation of either proteins or metabolites can be shown to have significant effects on the biology of interest.
Time Considerations Cell collection (Support Protocol 2) can take weeks to months for a large batch of samples, which is the most time-consuming step. We have found that the cells can be stored at −80◦ C up to 2 years without noticeable difference in performance. However, it is recommended to use samples within 1 to 3 months of collection. The actual number of samples processed per day depends on the throughput of LCMS. We typically run three technical replicates per sample. Each run takes 20 min, meaning 24 samples is the maximum that can be processed each day. Basic Protocol 1 takes ∼8 hr from cell lysis to staining of the NuPAGE gels, assuming 24 samples in one batch. The bead conjugation takes a maximum of two days. 300 mg Dynabeads is sufficient for several hundred samples. LC-MS data analyses can take several days to weeks of computing time for large sample size, and the manual checking raw data can be even longer.
Acknowledgements The authors thank Dr. Peichuan Zhang for critical review of this manuscript. This research was supported by a grant from the NIH.
Literature Cited Gallagher, S. 2006. One-dimensional SDS gel electrophoresis of proteins. Curr. Protoc. Mol. Biol. 75:10.2A.1-10.2A.37. Gallego, O., Betts, M.J., Gvozdenovic-Jeremic, J., Maeda, K., Matetzki, C., Aguilar-Gurrieri, C., Beltran-Alvarez, P., Bonn, S., FernandezTornero, C., Jensen, L.J., Kuhn, M., Trott, J., Rybin, V., M¨uller, C.W., Bork, P., Kaksonen, M., Russell, R.B., and Gavin, A.C. 2010. A systematic screen for protein-lipid interactions in Saccharomyces cerevisiae. Mol. Syst. Biol. 6:430. Gelperin, D.M., White, M.A., Wilkinson, M.L., Kon, Y., Kung, L.A., Wise, K.J., Lopez-Hoyo, N., Jiang, L., Piccirillo, S., Yu, H., Gerstein, M., Dumont, M.E., Phizicky, E.M., Snyder, M., and Grayhack, E.J. 2005. Biochemical and genetic analysis of the yeast proteome with a movable ORF collection. Genes Dev. 19:2816-2826 . Li, X. and Snyder, M. 2011. Metabolites as global regulators: a new view of protein regulation: Systematic investigation of metabolite-protein interactions may help bridge the gap between genome-wide association studies and small molecule screening studies. Bioessays 33:485489. Li, X., Gianoulis, T.A., Yip, K.Y., Gerstein, M., and Snyder,, M. 2010. Extensive in vivo metaboliteprotein interactions revealed by large-scale systematic analyses. Cell 143:639-650.
Analyzing In Vivo MetaboliteProtein Interactions
195 Current Protocols in Chemical Biology
Volume 3
Lowenadler, B., Jansson, B., Paleus, S., Holmgren, E., Nilsson, B., Moks, T., Palm, G., Josephson, S., Philipson, L., and Uhlen, M. 1987. A gene fusion system for generating antibodies against short peptides. Gene 58:87-97.
Smith, C.A., Want, E.J., O’Maille, G., Abagyan, R., and Siuzdak, G. 2006. XCMS: Processing mass spectrometry data for metabolite profiling using nonlinear peak alignment, matching, and identification. Anal. Chem. 78:779-787.
Morozov, V.N., Morozova, T.Y., Johnson, K.L., and Naylor, S. 2003. Parallel determination of multiple protein metabolite interactions using cell extract, protein microarrays and mass spectrometric detection. Rapid Commun. Mass Sp. 17:2430-2438.
Tagore, R., Thomas, H.R., Homan, E.A., Munawar, A., and Saghatelian, A. 2008. A global metabolite profiling approach to identify proteinmetabolite interactions. J. Am. Chem. Soc. 130:14111-14113.
Nilsson, B., Moks, T., Jansson, B., Abrahmsen, L., Elmblad, A., Holmgren, E., Henrichson, C., Jones, T.A., and Uhlen, M. 1987. A synthetic IgG-binding domain based on staphylococcal protein A. Protein Eng. 1:107-113.
Zhu, H., Bilgin, M., Bangham, R., Hall, D., Casamayor, A., Bertone, P., Lan, N., Jansen, R., Bidlingmaier, S., Houfek, T., Mitchell, T., Miller, P., Dean, R.A., Gerstein, M., and Snyder, M. 2001. Global analysis of protein activities using proteome chips. Science 293:2101-2105.
Analyzing In Vivo MetaboliteProtein Interactions
196 Volume 3
Current Protocols in Chemical Biology