http://jbx.sagepub.com/Journal of Biomolecular Screening

http://jbx.sagepub.com/content/14/9/1129The online version of this article can be found at:

 DOI: 10.1177/1087057109345528

2009 14: 1129 originally published online 4 September 2009J Biomol ScreenArunkumar Dhayalan, Emilia Dimitrova, Philipp Rathert and Albert Jeltsch

ScreeningA Continuous Protein Methyltransferase (G9a) Assay for Enzyme Activity Measurement and Inhibitor

  

Published by:

http://www.sagepublications.com

On behalf of: 

  Journal of Biomolecular Screening

can be found at:Journal of Biomolecular ScreeningAdditional services and information for    

  http://jbx.sagepub.com/cgi/alertsEmail Alerts:

 

http://jbx.sagepub.com/subscriptionsSubscriptions:  

http://www.sagepub.com/journalsReprints.navReprints:  

http://www.sagepub.com/journalsPermissions.navPermissions:  

http://jbx.sagepub.com/content/14/9/1129.refs.htmlCitations:  

What is This? 

- Sep 4, 2009 OnlineFirst Version of Record 

- Nov 12, 2009Version of Record >>

by guest on October 11, 2013jbx.sagepub.comDownloaded from by guest on October 11, 2013jbx.sagepub.comDownloaded from by guest on October 11, 2013jbx.sagepub.comDownloaded from by guest on October 11, 2013jbx.sagepub.comDownloaded from by guest on October 11, 2013jbx.sagepub.comDownloaded from by guest on October 11, 2013jbx.sagepub.comDownloaded from

© 2009 Society for Biomolecular Sciences www.sbsonline.org 1129

INTRODUCTION

Posttranslational modifications of histone tails such as acetylation, methylation, phosphorylation, ubiqui-

tylation, and sumoylation play important roles in epigenetic regulation.1 Among these various modifications, methylation of lysine residues in histone tails is quite unique, because depend-ing on the position of the lysine, which is getting methylated, the functional outcome may vary. For example, H3K9 methylation is associated with a condensed state of chromatin and repression of transcription, whereas H3K4 methylation is marking an active state of chromatin. Moreover, lysine can be monomethylated, dimethylated, or trimethylated, and these levels of methylation add another layer of complexity in regulation.2 Histone lysine methyltransferases use S-adenosyl-L-methionine (AdoMet) as a cofactor. Most of them harbor a SET domain of approximately 130 amino acids in size as catalytic domain.2

The p53 tumor suppressor protein was the 1st nonhistone protein, shown to be methylated by SET7/9, a protein lysine

methyltransferase (PKMT) that was originally described as H3K4 methyltransferase.3 Later, other PKMTs were shown to methylate p53 as well, and it was discovered that the functional outcome of p53 lysine methylation depends on the methylation site. For example, methylation of p53 at a specific residue by SET7/9 leads to activation of p53, whereas methylation at 2 other lysine residues by SMYD2 or Set8 leads to its inactiva-tion.4 In addition, many more nonhistone proteins are methy-lated by PKMTs, and the list is growing rapidly.4,5

Different methods exist to study histone lysine methyltrans-ferases by using synthetic peptides. Methylation can be detected by mass spectrometry6 or by following the transfer of the radio-actively labeled methyl group from the cofactor S-adenosyl-L-methionine (AdoMet) to the peptide substrate.7-9 In addition, fluorescent assays were developed for the study of histone lysine methyltransferases in which the turnover of cofactor is followed10 or the generation of the methylated peptide is detected by antibody binding.11 The latter assay had been used to develop an inhibitor of the G9a histone methyltransferase (BIX-01294).

Synthetic peptide substrates are well suited for studying methylation of histone tails because the N-terminal tails of the histones are unstructured. In contrast, they are not sufficient for studying protein methylation because proteins exhibit a com-plex 3-dimensional fold, which may protect some peptide regions against enzymatic modification and create 3-dimensional sur-face epitopes that could be recognized by protein methyltrans-ferases. Currently, few methods are available for the analysis of protein methylation. Antibodies against different methylation states of lysine in the histone tails can be used to study histone methylation.12 For nonhistone targets, the caveat is that often

1Biochemistry Laboratory, School of Engineering and Science, Jacobs University Bremen, Bremen, Germany.

2Biochemistry and Cell Biology Program, School of Engineering and Science, Jacobs University Bremen, Bremen, Germany.

3Current affiliation: Active Motif Europe, Rixensart, Belgium.

Received May 19, 2009, and in revised form Jul 15, 2009. Accepted for publi-cation Jul 18, 2009.

Journal of Biomolecular Screening 14(9); 2009 DOI:10.1177/1087057109345528

A Continuous Protein Methyltransferase (G9a) Assay for Enzyme Activity Measurement and Inhibitor Screening

ARUNKUMAR DHAYALAN,1 EMILIA DIMITROVA,2 PHILIPP RATHERT,1,3 and ALBERT JELTSCH1

The authors describe a continuous protein methylation assay using the G9a protein lysine methyltransferase and its substrate protein WIZ (widely interspaced zinc finger motifs). The assay is based on the coupling of the biotinylated substrate protein to streptavidin-coated FlashPlates and the transfer of radioactive methyl groups from the S-adenosyl-L-methionine to the substrate. The reaction progress is monitored continuously by proximity scintillation counting. The assay is very accurate, convenient, well suited for automation, and highly reproducible with standard errors in the range of 5%. Because of few pipetting steps and continuous data readout, it is ideal for high-throughput applications such as screening of inhibitors, test-ing many enzyme variants, or analyzing differences in methylation rates of different substrates under various conditions. By using this new assay, the IC50 of AdoHcy and the G9a inhibitor BIX-01294 were determined for methylation of the G9a nonhistone substrate WIZ. (Journal of Biomolecular Screening 2009:1129-1133)

Key words: protein modifications, enzyme inhibitor, protein methyltransferase, inhibitor screening, enzyme catalysis

Application Note

Dhayalan et al.

1130 www.sbsonline.org Journal of Biomolecular Screening 14(9); 2009

these antibodies have to be raised in a modification-specific form for each individual target (although in some cases, good pan-specific lysine methyl antibodies might be available).8 Protein lysine methylation can also be studied by radioactive assays employing radioactive methylation of the protein, gel electrophoresis, and autoradiography.8 However, this approach is not suited for high-throughput applications of many samples in a quantitative manner.

EXPERIMENTAL PROCEDURES

Materials

[methyl-H3] S-adenosyl-L-methionine was purchased from PerkinElmer (Waltham, MA). S-adenosyl-L-homocysteine and BIX01294 were purchased from Sigma (St. Louis, MO).

Protein purification

The sequence-encoding C-terminal 280 residues of human G9a was cloned pET28a vector (Novagen, San Diego, CA) in frame with N-terminal His6-tag. For expression, Escherichia coli BL21 DE3 cells (Novagen) carrying the G9a expression construct were grown in Luria-Bertani medium supplemented with Kanamycin and 50 µM ZnSO4 at 37 °C to OD600 of 0.6, then shifted to 16 °C for 20 min and induced overnight with 0.8 mM IPTG. The cells were collected by centrifugation, resus-pended in lysis buffer (50 mM potassium phosphate buffer [pH 8], 500 mM NaCl, 0.5 mM DTT, 10 mM imidazole, and 5% glycerol) and disrupted by sonication. The supernatant was passed through an Ni-NTA resin (Qiagen, Venlo, the Netherlands) and washed with lysis buffer, and the bound protein was eluted with lysis buffer containing 250 mM imidazole and dialyzed against 20 mM Tris (pH 7.4), 100 mM KCl, 0.5 mM DTT, 10% glyc-erol for 4 h then overnight against the same buffer with 70% glycerol (Fig. 1B). The GST-tagged substrate protein domain WIZ was expressed and purified as described.8

Biotinylation of the WIZ target protein domain

WIZ was biotinylated by using sulfhydryl-reactive biotiny-lation reagent (Maleimide-PEG2-Biotin; Pierce, Rockford, IL) according to the manufacturer’s instructions. Briefly, the puri-fied protein was dialyzed in dialysis buffer (20 mM HEPES pH 7.0, 100 mM KCl, and 10% glycerol) to remove the DTT pres-ent in the protein storage buffer. The dialyzed protein was incubated with 20-fold molar excess of Maleimide-PEG2-Biotin for 8 h at 4 °C. Afterward, the protein samples were dialyzed in the dialysis buffer to remove the unbound Maleimide-PEG2-Biotin, and the concentration of the biotiny-lated protein was determined by ultraviolet spectroscopy. The efficiency of biotinylation was >50% as determined by the yield of binding of the biotinylated protein to avidin plates.

Continuous methylation assay

The biotinylated WIZ protein was bound to the streptavidin surface of the 96-well FlashPlate® (PerkinElmer, Waltham, MA) by adding 40 µL of 0.75 µM biotinylated WIZ in binding buffer (20 mM HEPES pH 7.5, 100 mM KCl, 10% glycerol, 1 mM EDTA, and 0.1 mM DTT) in each well of the streptavidin-coated plate. After 8 h incubation at 4 °C, the wells were washed 5 times with PBST (500 mM NaCl, 2.7 mM KCl, 4.3 mM Na2HPO4, 1.4 mM K2HPO4, 0.05% v/v Tween 20, pH 7.2). For the methylation reaction, 60 µL reaction mixture (50 mM glycine, pH 9.8, 2 mM DTT, 25 µg/mL bovine serum albumin, 10% glycerol) containing G9a enzyme (1.1 µM) and 0.76 µM tritium-labeled AdoMet (specific activity: 2.7 TBq/mmol; PerkinElmer) was added to each well. For readout, a radioac-tive microplate reader (Topcount NXT; PerkinElmer) was used. The radioactive signal was detected and averaged for 10 s, and after each reading, the counts and absolute time were written in the report file. From that file, data were extracted and rear-ranged using an in-house program.

RESULTS AND DISCUSSION

In the present study, we have developed a high-throughput assay to analyze the activity of PKMTs with protein substrates using the protein lysine methyltransferase G9a and its substrate protein domain Wiz as model system.8 The assay is an adaptation of a peptide methylation assay developed in our laboratory.9 It uses streptavidin-coated FlashPlate 96-well microplates.13 In these plates, each well is permanently coated with a thin layer of polystyrene-based scintillant followed by covalent binding of streptavidin molecules. The substrate protein for methylation is biotinylated by using sulfhydryl-reactive bioti-nylation reagent. The maleimide group in this biotinylation reagent reacts specifically and efficiently with reduced thiols to form stable thioether bonds. Then, the biotinylated target pro-tein is bound to the wells of a FlashPlate. After washing away the unbound substrate protein, the methylation reaction mixture is added, which contains enzyme and AdoMet with the radio-actively labeled methyl group in a suitable reaction buffer. In solution, the β-particles emitted from the radioactively labeled AdoMet will not give a strong scintillation signal because most β-particles are quenched by solvent before they reach the bot-tom of the well, where the scintillant is localized. However, when the labeled methyl groups are enzymatically transferred to the substrate protein, this leads to a strong scintillation signal because the substrate is also attached to the bottom of the well. Therefore, the reaction progress can be continuously monitored by the increase in scintillation signal without any manual pipetting and washing (Fig. 1A).

For the proof of concept of the methylation assay, the pro-tein lysine methyltransferase G9a was used to methylate one of

Continuous Protein Methyltransferase Assay

Journal of Biomolecular Screening 14(9); 2009 www.sbsonline.org 1131

its protein substrates, WIZ.8 After the addition of the enzyme to the wells, which were previously coated with biotinylated pro-tein substrate WIZ, a rapid and very strong increase in scintilla-tion signal was observed (Fig. 1C). There was no signal change in the absence of enzyme or radioactively labeled AdoMet (Fig. 1C). However, there was a very weak signal observed in the absence of biotinylated protein (data not shown). Although this background signal was weak and therefore did not affect our methylation assay, we tested several conditions to minimize it. The background methylation could be futher reduced by over-night preincubation of the FlashPlate with G9a enzyme and unlabelled AdoMet at 4 °C and washing before commencing the actual experiment (Fig. 1C). This result suggests that the back-ground is due to some weak methylation activity of G9a toward the streptavidin itself, which is possible because the specificity profile of G9a is loose.8

To investigate the reproducibility of our assay, 5 identical methylation reactions containing the same amount of enzyme and radioactively labeled AdoMet were done in 5 independent wells. The primary data of all 5 reactions were readily superimposable

(Fig. 1C). We also determined the rates of methylation for all 5 individual reactions by fitting the data to an exponential curve. The fit showed that the rates were reproduced with a standard error of less than 6%, indicating the excellent reproducibility of the assay. The signal-to-background ratio was 10.2 as calcu-lated by comparing the average initial slopes of the enzymatic reactions with the slopes of the background reaction without substrate, which showed the highest background activity (see above). The signal-to-noise ratio was 12.5, as calculated by dividing the average of the initial slopes by its standard error. The Z-score of this measurement (as defined by Seiler and oth-ers14) was 5.6.

To analyze the influence of the efficiency of biotinylation on our assay, we carried out 2 independent biotinylation reactions and studied the methylation. Although the sets showed some variation in absolute methylation levels, after fitting to expo-nential reaction progress curves, the rate constants of the methyl group transfer were similar with a standard error of ±2.5%. This indicates that the assay is appropriate for screening as long as matching control reactions are included.

FIG. 1. Development of a continuous protein methylation assay. (A) Schematic diagram explaining the principle of the assay. In the 1st step, the substrate protein is biotinlyated and bound to the wells of a FlashPlate (large, light grey circles indicate the target protein and dark grey dots indicate the biotinylation). After the addition of protein methyltransferase (dark grey ellipses) and radioactively labeled AdoMet (black dots), the transfer of radioactively labeled methyl group to the protein substrate at the bottom of the well will result in a strong scintillation signal. (B) Coomassie-stained gel of purified G9a enzyme. (C) Reproducibility of the methylation assay. Five identical methylation reactions containing the same amount of enzyme and radioactively labeled AdoMet were carried out in 5 independent wells, and the results were overlaid without any normalization. Fitting to an exponential reaction progress curve showed that the rate constants of the methyl group transfer of the 5 reactions were similar, with a standard error of ±6%.

Dhayalan et al.

1132 www.sbsonline.org Journal of Biomolecular Screening 14(9); 2009

In the next phase, we tested the applicability of our method for the screening of inhibitors. For this purpose, we carried out methylation reactions in the absence and presence of different concentrations of S-adenosyl-L-homocysteine (AdoHcy), which is the product of the methylation reaction and acts as a general inhibitor of DNA and protein methyltransferases (Fig. 2). As expected, we observed a decrease in the methylation signal with increasing concentrations of AdoHcy. We carried out 2 independent experiments and determined the IC50 of AdoHcy on G9a by plotting the enzymatic activity against AdoHcy con-centration. The IC50 values were reproducible with a standard error of ±6% (Fig. 2), indicating the suitability of the method for characterization of enzyme inhibition.

Recently, it was shown that the compound BIX-01294 (a diazepin-quinazolin-amine derivative) could inhibit G9a enzyme activity toward methylation of the histone 3 tail.11 However, it is not yet clear whether the same compound will inhibit G9a methylation activity toward nonhistone targets as well. Therefore, we have studied the inhibition effect of BIX-01294 on G9a toward methylation of its nonhistone target protein WIZ by using our new assay. For this purpose, we performed 2 inde-pendent methylation reactions in the absence and the presence of different concentrations of BIX-01294 (Fig. 3). We observed a decrease in the methylation signal with increasing concentra-tions of BIX-01294 in a dose-dependent manner. We have determined the IC50 value of BIX-01294 on G9a toward the

WIZ substrate as 20 ± 1 µM (Fig. 3). It was recently shown that BIX-01294 inhibits G9a with a half-maximal inhibitory concentration value of 1.9 µM for the histone 3 substrate,15 indicating that BIX-01294 inhibits G9a activity stronger with the histone 3 tail peptide than with the WIZ protein. This obser-vation agrees with the finding that BIX-01294 binds into the peptide-binding pocket of G9a.15 It also indicates that it is pos-sible to inhibit the G9a enzyme activity selectively toward some of its substrates.

One limitation of this assay is that the enzymatic reaction takes place after binding the protein substrate to the solid sup-port. Therefore, the effective protein concentration is not known and it also cannot be varied in a systematic manner. As a consequence of this, the assay design is more restricted than in solution assays and, for example, Km values for the protein substrate cannot be determined.

CONCLUSIONS

A detailed characterization of protein methylation is essen-tial because many of the originally described histone lysine

0

5000

10000

15000

20000

25000

0 2000 4000 6000 8000 10000

Time [s]

rel.

acti

vity

[C

PM

]

0 uM

20 uM

50 uM

100 uM

200 uM

500 uM

750 uM

1 mM

0

0.5

1

1.5

2

2.5

3

3.5

4

0 200 400 600 800 1000

rel.

acti

vity

[C

PM

/s]

A

B

cAdoHcy

cAdoHcy[µM]

FIG. 2. Application of the continuous protein methylation assay for screening of enzyme inhibitors. (A) Inhibition of protein methylation with increasing concentrations of AdoHcy. (B) Determination of the IC50 of AdoHcy on G9a by plotting enzymatic activity against AdoHcy concentration (IC50 = 25.7 [±1.5] µM).

0

5000

10000

15000

20000

25000

0 2000 4000 6000 8000 10000

Time (S)

rel.

acti

vity

(C

PM

) 0 uM

20 uM

50 uM

200 uM

500 uM

750 uM

1 mM

A

B

0

0.5

1

1.5

2

2.5

3

3.5

4

0 200 400 600 800 1000re

l. ac

tivi

ty [

CP

M/s

]

cBIX-01294

cBIX-01294[µM]

FIG. 3. Application of the continuous protein methylation assay for the study of BIX-01294 inhibition of G9a enzyme toward its substrate WIZ. (A) Inhibition of protein methylation with increasing concentra-tions of BIX-01294. (B) Determination of the IC50 of BIX-01294 on G9a by plotting enzymatic activity against BIX-01294 concentration (IC50 = 20 [±1] µM).

Continuous Protein Methyltransferase Assay

Journal of Biomolecular Screening 14(9); 2009 www.sbsonline.org 1133

methyltransferases also act on nonhistone substrate proteins, and the methylation of these target protein has important bio-logical roles.4,5 Here we present a reliable, quantitative, and continuous assay for the study of protein lysine methyltrans-ferases that can be used to screen inhibitors for protein lysine methyltransferases in a high-throughput manner.

ACKNOWLEDGMENTS

This work has been supported by DFG Je252/7-1 and NIH DK08267.

REFERENCES

1. Kouzarides T: Chromatin modifications and their function. Cell 2007;128:693-705.

2. Cheng X, Collins RE, Zhang X: Structural and sequence motifs of protein (histone) methylation enzymes. Annu Rev Biophys Biomol Struct 2005;34: 267-294.

3. Chuikov S, Kurash JK, Wilson JR, Xiao B, Justin N, Ivanov GS, et al.: Regulation of p53 activity through lysine methylation. Nature 2004;432: 353-360.

4. Huang J, Berger SL: The emerging field of dynamic lysine methylation of non-histone proteins. Curr Opin Genet Dev 2008;18:152-158.

5. Rathert P, Dhayalan A, Ma H, Jeltsch A: Specificity of protein lysine methyltransferases and methods for detection of lysine methylation of non-histone proteins. Mol Biosyst 2008;4:1186-1190.

6. Bonaldi T, Regula JT, Imhof A: The use of mass spectrometry for the analysis of histone modifications. Methods Enzymol 2004;377: 111-30.

7. Zhang X, Tamaru H, Khan SI, Horton JR, Keefe LJ, Selker EU, et al.: Structure of the Neurospora SET domain protein DIM-5, a histone H3 lysine methyltransferase. Cell 2002;111:117-127.

8. Rathert P, Dhayalan A, Murakami M, Zhang X, Tamas R, Jurkowska R, et al.: Protein lysine methyltransferase G9a acts on non-histone targets. Nat Chem Biol 2008;4:344-346.

9. Rathert P, Cheng X, Jeltsch A: Continuous enzymatic assay for histone lysine methyltransferases. Biotechniques 2007;43:602, 604, 606.

10. Collazo E, Couture JF, Bulfer S, Trievel RC: A coupled fluorescent assay for histone methyltransferases. Anal Biochem 2005;342:86-92.

11. Kubicek S, O’Sullivan RJ, August EM, Hickey ER, Zhang Q, Teodoro ML, et al.: Reversal of H3K9me2 by a small-molecule inhibitor for the G9a histone methyltransferase. Mol Cell 2007;25:473-481.

12. Sarma K, Nishioka K, Reinberg D: Tips in analyzing antibodies directed against specific histone tail modifications. Methods Enzymol 2004;376:255-269.

13. Brown BA, Cain M, Broadbent J, Tompkins S, Henrich G, Joseph R, et al.: FlashPlate technology. In Devlin JP (eds): High Throughput Screening. New York: Dekker, 1997:317-328.

14. Seiler KP, George GA, Happ MP, Bodycombe NE, Carrinski HA, Norton S, et al.: ChemBank: a small-molecule screening and cheminformatics resource database. Nucleic Acids Res 2008;36:D351-D359.

15. Chang Y, Zhang X, Horton JR, Upadhyay AK, Spannhoff A, Liu J, et al.: Structural basis for G9a-like protein lysine methyltransferase inhibition by BIX-01294. Nat Struct Mol Biol 2009;16:312-317.

Address correspondence to:Prof. Dr. Albert Jeltsch

Biochemistry LaboratorySchool of Engineering and Science

Jacobs UniversityBremen, Campus Ring 1

28759 Bremen, Germany

E-mail: a.jeltsch@jacobs-university.deWeb: http://www.jacobs-university.de/ses/ajeltsch/