Investigating the proteome reactivity and selectivity of aryl halides

Article abstract of DOI:10.1021/ja4116204

Protein-reactive electrophiles are critical to chemical proteomic applications including activity-based protein profiling, site-selective protein modification, and covalent inhibitor development. Here, we explore the protein reactivity of a panel of aryl halides that function through a nucleophilic aromatic substitution (S_NAr) mechanism. We show that the reactivity of these electrophiles can be finely tuned by varying the substituents on the aryl ring. We identify p-chloro-and fluoronitrobenzenes and dichlorotriazines as covalent protein modifiers at low micromolar concentrations. Interestingly, investigating the site of labeling of these electrophiles within complex proteomes identified p-chloronitrobenzene as highly cysteine selective, whereas the dichlorotriazine favored reactivity with lysines. These studies illustrate the diverse reactivity and amino-acid selectivity of aryl halides and enable the future application of this class of electrophiles in chemical proteomics.

Full text of DOI:10.1021/ja4116204

Communication
pubs.acs.org/JACS
Investigating the Proteome Reactivity and Selectivity of Aryl Halides
D. Alexander Shannon,† Ranjan Banerjee,† Elizabeth R. Webster,† Daniel W. Bak,† Chu Wang,‡
and Eranthie Weerapana*,†
†Department of Chemistry, Merkert Chemistry Center, Boston College, Chestnut Hill, Massachusetts 02467, United States
‡Department of Chemical Biology, College of Chemistry and Molecular Engineering, Synthetic and Functional Biomolecules Center,
and Peking-Tsinghua Center for Life Sciences, Peking University, Beijing 100871, China
S
* Supporting Information
into cysteine-reactive peroxisome proliferator-activated receptor
(PPARγ)19 and β-tubulin-modifying compounds.20 4-Halopyr-
ABSTRACT: Protein-reactive electrophiles are critical to
chemical proteomic applications including activity-based
protein profiling, site-selective protein modification, and
covalent inhibitor development. Here, we explore the
protein reactivity of a panel of aryl halides that function
through a nucleophilic aromatic substitution (SNAr)
mechanism. We show that the reactivity of these
electrophiles can be finely tuned by varying the
substituents on the aryl ring. We identify p-chloro- and
fluoronitrobenzenes and dichlorotriazines as covalent
protein modifiers at low micromolar concentrations.
Interestingly, investigating the site of labeling of these
electrophiles within complex proteomes identified p-
chloronitrobenzene as highly cysteine selective, whereas
the dichlorotriazine favored reactivity with lysines. These
studies illustrate the diverse reactivity and amino-acid
selectivity of aryl halides and enable the future application
of this class of electrophiles in chemical proteomics.
idines have been shown to covalently modify the active-site
cysteine of dimethylarginine dimethylaminohydrolase.21,22
Lastly, perfluoroaryl groups have been used for covalent
peptide stapling.23 In all these cases, arylation of the cysteine
residue is presumed to progress via a nucleophilic aromatic
substitution (SNAr) mechanism, similar to that occurring
during the enzymatic conjugation of glutathione to activated
aryl groups by glutathione S-transferases.24 Here, we synthe-
sized a panel of alkyne-functionalized aryl halides and
systematically evaluated the reactivity and selectivity of these
electrophiles within the context of a complex proteome. The
alkyne handle facilitated the use of copper-assisted azide−
alkyne cycloaddition (CuAAC) for gel and mass spectrometry
(MS)-based analysis of protein reactivity.25,26
We initiated our studies by exploring the proteome reactivity
of a panel of chloronitrobenzenes (CNBs; Figure 1A). These
he covalent modification of proteins by small molecules
1,2
Thas widespread applications in drug design,
activity-
based protein profiling,3,4 and imaging.5 These applications rely
on the availability of an arsenal of electrophiles with tunable
reactivity and selectivity.6 Well-characterized electrophiles
include haloacetamide,7−9 maleimide,9 and α,β-unsaturated
ketones,7 which have been shown to be highly selective for
the thiol group of cysteine residues. Electrophiles that target
other amino acids include sulfonate esters (aspartate, glutamate,
and tyrosine),7 fluorophosphonates (serine),10 sulfonyl fluo-
rides (serine, tyrosine),11,12 and acyl phosphates (lysine).13 In
addition to functional group specificity, electrophiles also
demonstrate diverse reactivities. Many of the electrophiles
listed above demonstrate high reactivity and covalently modify
proteins in the absence of a binding motif. The epoxysucci-
nates,14 spiroepoxides,15 carbamates,16 acyloxymethyl ke-
tones,17 phenoxymethyl ketones,18 and β-lactams16 are milder
and often require a binding element to facilitate covalent
modification. Due to the widespread applications of covalent
protein modification, there is a constant need for novel
electrophiles with distinct and tunable reactivities and
selectivities.
Figure 1. Evaluating the proteome reactivity of chloronitrobenzenes
(CNBs). (A) The panel of alkyne-functionalized CNBs used in this
study. (B) In-gel fluorescence evaluation of the reactivity of CNBs in
proteomes after incorporation of Rh−N3 using CuAAC. The
corresponding Coomassie-stained gel can be found in the Supporting
Information (Figure S1).
electrophiles are anticipated to react via an SNAr mechanism,
whereby the rate-limiting step is formation of the Meisen-
heimer or σ-complex.27,28 Therefore, we hypothesized that the
reactivity of these aryl halides can be tuned by varying the
position of the electron-withdrawing nitro substituent. To test
this hypothesis, we synthesized four alkyne-functionalized CNB
probes with the nitro group ortho (ERW1 and ERW2), meta
The proteome reactivity of aryl chloride-based electrophiles
has been poorly characterized, with a few scattered examples in
the literature. p-Chloronitrobenzenes have been incorporated
Received: November 19, 2013
Published: February 18, 2014
© 2014 American Chemical Society
3330
dx.doi.org/10.1021/ja4116204 | J. Am. Chem. Soc. 2014, 136, 3330−3333

Journal of the American Chemical Society
Communication
Figure 2. Investigating the proteome reactivity of aryl halides. (A) Panel of aryl halides (chloronitrobenzene, fluoronitrobenzene, chloropyridine,
chloropyrimidine, and chlorotriazine) analyzed in this study. (B) In-gel fluorescence image of probe labeling in HeLa cell lysates treated with 100 μM
of each probe, tagged with Rh-N3 using click chemistry, and separated by SDS-PAGE. (C) Comparison of probe labeling of RB7 (5 μM) with RB2
and ERW3 (20 μM). (D) Effect of iodoacetamide (IAA) treatment (1 mM) prior to labeling with RB2, ERW3, and RB7. Coomassie-stained gels can
be found in the Supporting Information (Figures S2−S4).
(RB1), and para (RB2) to the chloro substituent (Figure 1A;
Scheme S1). We incubated HeLa cell lysates with each probe at
100 μM concentrations for 1 h, after which we appended a
fluorescent rhodamine-azide (Rh-N3) using CuAAC, separated
the proteins on SDS-PAGE, and visualized protein labeling
using in-gel fluorescence (Figure 1B). As anticipated, the p-
nitro-substituted CNB was the most reactive (Figure 1B, lane
2) due to resonance stabilization of the Meisenheimer complex
by the electron-withdrawing nitro group. Shifting the nitro
substituent to the meta position (RB1) fully abrogated
reactivity (Figure 1B, lane 1), whereas the ortho-substituted
ERW1 and ERW2 showed protein reactivity (Figure 1A, lanes 3
and 4), albeit less than for RB2. These data demonstrate that
the proteome reactivity of CNBs can be finely adjusted by
modulating the electronics and sterics of the aryl ring system,
thereby providing an ideal tunable electrophile for chemical
proteomic applications.
To compare the reactivity of CNBs to that of other aryl
halides, we expanded our panel to include a fluoronitrobenzene
(ERW3), chloropyridines (RB3 and RB4), chloropyrimidines
(RB5 and RB6), and a chlorotriazine (RB7) (Figure 2A). The
extent of resonance stabilization provided by the electron-
withdrawing nitro substituent or the nitrogen(s) present within
the ring itself will determine the reactivity of each of these aryl
halides.27,28 Since we identified RB2 as the most reactive of the
CNBs, we compared each probe to RB2 in terms of their
reactivity (Figure 2B, lane 1). Despite the presence of the
poorer fluorine leaving group, ERW3 exhibited reactivity similar
to that of RB2 (Figure 2B, lane 2). The chloropyridines (RB3
and RB4; Figure 2B, lanes 3 and 4) and the chloropyrimidines
(RB5 and RB6; Figure 2B, lanes 5 and 6) showed significantly
decreased reactivity relative to RB2. Interestingly, the
dichlorotriazine RB7 exhibited very potent labeling (Figure
2B, lane 7), suggesting that the stabilization of the SNAr
transition state due to the presence of the third ring nitrogen is
greater than that provided by the p-nitro group of RB2.
This initial screening identified RB2, ERW3, and RB7 as aryl
halides with high proteome reactivity. To better visualize the
pattern of labeling across these three probes, we reduced the
concentration of the highly reactive RB7 probe to 5 μM (Figure
2C, lane 3) and compared it to RB2 and ERW3 at 20 μM
(Figure 2C, lanes 1 and 2). This comparison revealed that the
majority of the bands for RB7 do not coincide with those of
RB2 and ERW3, suggesting that this dichlorotriazine probe was
targeting a subset of proteins distinct from the halonitroben-
zenes. We then proceeded to interrogate the cysteine selectivity
of these electrophiles by competition with a known highly
reactive, cysteine-selective electrophile, iodoacetamide (IAA).
To achieve this competition, we pretreated HeLa lysates with
IAA (0 and 2 mM), followed by labeling with RB2 (20 μM),
ERW3 (20 μM), or RB7 (1 μM) (Figure 2D). IAA treatment
almost completely abolished RB2 and ERW3 labeling,
suggesting the selective modification of cysteine residues by
these two electrophiles. In contrast, the majority of RB7-labeled
proteins remained after IAA treatment. In fact, labeling of some
bands was enhanced by IAA, likely due to partial denaturation
and exposure of previously inaccessible sites under high IAA
concentrations. These data thereby allude to the fact that RB7
predominantly targets amino acids other than cysteine within
the proteome. We also administered the less reactive probes
(RB1, RB3, RB4, RB5, and RB6) at high millimolar
concentrations and show that labeling by these probes is
similar to that of RB2 and therefore cysteine-selective (Figures
S5−S9).
To confirm the cysteine selectivity of RB2 and identify the
amino-acid targets of RB7, we employed a MS platform geared
to identify the site of probe labeling within a complex
proteome. This method, termed tandem orthogonal proteolysis
activity-based protein profiling (TOP-ABPP),29,30 allows for the
specific enrichment of probe-labeled peptides for identification
of the site of labeling. Mouse liver and HeLa lysates were
treated with RB2 or RB7 (100 μM) and analyzed by TOP-
ABPP. The resulting MS data were analyzed for probe
modifications on all nucleophilic amino acids (cysteine,
aspartate, glutamate, histidine, lysine, serine, threonine, and
tyrosine) using the SEQUEST algorithm.31 The percentage of
3331
dx.doi.org/10.1021/ja4116204 | J. Am. Chem. Soc. 2014, 136, 3330−3333

Journal of the American Chemical Society
Communication
unique peptides modified for each reactive amino acid was
plotted for both RB2 and RB7 (Figure 3A). In agreement with
further explore the targets of RB7, we performed an isoTOP-
ABPP analysis,8 whereby we compared labeling at low (10 μM)
and high (100 μM) concentrations of RB7 to rank the labeled
lysine residues in order of reactivity (Table S4). Within these
reactive lysines, we identify many that are annotated as sites of
acetylation, as well as known active sites (Glud1) or ATP-
binding (Nme1/2) sites. The acetylation sites are generally
surface exposed, whereas the active-site residues are found to be
buried within the protein core (Figures S12−S14). We
subsequently validated selected labeled peptides by manually
calculating theoretical y- and b-ions and annotating the
associated MS2 spectra for both RB2 and RB7-labeled peptides
(Figures 3C,D and S5). Additionally, we performed probe
labeling in situ with both RB2 and RB7 probes and found that
these electrophiles show similar protein labeling patterns when
administered to living cells (Figure S10).
Previous studies with carbon electrophiles such as chlor-
oacetamides and sulfonate esters demonstrated that the
solution reactivity of electrophiles is often not predictive of
reactivity observed in a proteome.7 This is likely due to the
unique protein microenvironment that serves to modulate the
pKa and reactivity of amino-acid side chains. We sought to
determine if the lysine selectivity observed for RB7 in
proteomes is reflected in solution reactivity with free amino
acids. To achieve this, we incubated RB2 and RB7 with a 20-
fold excess of C- and N-terminal protected lysine and cysteine
amino-acid derivatives and analyzed the formation of the
corresponding adducts by LC-MS. We observed formation of
the RB2-Cys adduct but could not detect the corresponding
lysine adduct (Figure 4A). In contrast, RB7 formed adducts
Figure 3. Characterizing amino-acid selectivity in proteomes. (A)
Percentage of unique peptides labeled on each nucleophilic amino acid
by RB2 and RB7 (100 μM) in mouse liver (RB2, n = 3; RB7, n = 5)
and HeLa (RB2, n = 2; RB7, n = 4) proteomes. (B) A subset of RB2-
and RB7-labeled peptides identified in the proteomic studies using
TOP-ABPP. (C) Annotated MS2 of RB2-labeled HsHSPD1 peptide.
(D) Annotated MS2 of RB7-labeled HsEEF1A1 peptide.
Figure 4. Reactivity of RB2 and RB7 with protected amino-acid
derivatives in solution (12-h incubation in PBS). (A) RB2 with 20-fold
excess cysteine or lysine derivatives. (B) RB7 with 20-fold excess
cysteine or lysine derivatives. (C). RB2 and RB7 were mixed together
and treated with 20-fold excess cysteine derivative. (D) RB7 was
concurrently treated with 10-fold excess of both cysteine and lysine
derivatives.
our IAA competition study (Figure 2C), RB2 primarily labels
cysteine residues (Figure 3A). In contrast, RB7 shows very high
reactivity with lysine, with minimal cysteine modification
observed (Figure 3A).
Closer investigation of the probe-labeled peptides identified
for RB2 and RB7 showed targeting of diverse proteins (Figure
3B; Tables S1−S3). RB2 targets known active-site nucleophiles
of aldehyde dehydrogenase (ALDH6A1) and phosphoenol
pyruvate carboxykinase (PCK1), alluding to the use of this
electrophile to target functional cysteines in the proteome. To
with both cysteine and lysine (Figure 4B). To further evaluate
the cysteine reactivity of RB2 and RB7, both probes were
simultaneously incubated with excess cysteine (Figure 4C).
This experiment showed predominantly the RB7-Cys adduct,
implying that RB7 was more reactive than RB2. Lastly, when a
limiting amount of RB7 was incubated with excess cysteine and
lysine derivatives, only the RB7-Cys adduct was observed
3332
dx.doi.org/10.1021/ja4116204 | J. Am. Chem. Soc. 2014, 136, 3330−3333

Journal of the American Chemical Society
Communication
(2) Johnson, D. S.; Weerapana, E.; Cravatt, B. F. Future Med. Chem.
2010, 2, 949.
(3) Cravatt, B. F.; Wright, A. T.; Kozarich, J. W. Annu. Rev. Biochem.
(Figure 4D), indicating that while RB7 can react with free
amines at biologically relevant pH, the thiol adduct is the
favored product. These studies serve to highlight the disparity
often observed between the solution reactivity and the
proteome reactivity of electrophiles. In this case, RB2 prefers
cysteine in solution and in proteomes, whereas RB7 appears to
preferentially react with cysteines in solution but with lysine in
the context of a proteome.
2008, 77, 383.
(4) Fonovic, M.; Bogyo, M. Exp. Rev. Proteomics 2008, 5, 721.
(5) Edgington, L. E.; Verdoes, M.; Bogyo, M. Curr. Opin. Chem. Biol.
2011, 15, 798.
(6) Haedke, U.; Kuttler, E. V.; Vosyka, O.; Yang, Y.; Verhelst, S. H.
Curr. Opin. Chem. Biol. 2013, 17, 102.
In summary, we have systematically evaluated the reactivity
and amino-acid selectivity of a panel of aryl halides that
function through an SNAr mechanism. We show that
chloronitrobenzenes are highly tunable electrophiles, where
reactivity can be matched for a desired application by
modifications to the steric and electronic properties of the
aryl ring system. As expected, due to resonance stabilization of
the SNAr intermediate, the p-chloronitrobenzenes were the
most reactive. We also evaluated the proteome reactivity of
fluoronitrobenzenes, halopyridines, halopyrimidines, and di-
chlorotriazines, which have been poorly characterized and
utilized in proteomic applications. We show that the p-
halonitrobenzenes and dichlorotriazine are reactive at low
micromolar concentrations. Identification of the sites of
modification for these aryl halides in proteomes demonstrated
that the p-chloronitrobenzene preferentially reacts with cysteine
residues whereas dichlorotriazine prefers lysine. The dichloro-
triazine provides an aromatic, synthetically tractable, and
hydrolytically stable electrophile to add to the arsenal of
lysine-reactive groups available for protein modification. Lastly,
we show that the observed amino-acid selectivity is not
universally mimicked when we expose the electrophiles to free
amino acids in solution, underscoring the importance of
studying reactivity and selectivity in the context of a proteome.
We anticipate that our studies into the proteome reactivity of
aryl halides will promote the educated use of these electrophiles
in chemical proteomics, where the reactivity and amino-acid
selectivity can be judiciously matched to the desired
application.
(7) Weerapana, E.; Simon, G. M.; Cravatt, B. F. Nat. Chem. Biol.
2008, 4, 405.
(8) Weerapana, E.; Wang, C.; Simon, G. M.; Richter, F.; Khare, S.;
Dillon, M. B.; Bachovchin, D. A.; Mowen, K.; Baker, D.; Cravatt, B. F.
Nature 2010, 468, 790.
(9) Shin, N. Y.; Liu, Q.; Stamer, S. L.; Liebler, D. C. Chem. Res.
Toxicol. 2007, 20, 859.
(10) Kidd, D.; Liu, Y.; Cravatt, B. F. Biochemistry 2001, 40, 4005.
(11) Gu, C.; Shannon, D. A.; Colby, T.; Wang, Z.; Shabab, M.;
Kumari, S.; Villamor, J. G.; McLaughlin, C. J.; Weerapana, E.; Kaiser,
M.; Cravatt, B. F.; van der Hoorn, R. A. Chem. Biol. 2013, 20, 541.
(12) Shannon, D. A.; Gu, C.; McLaughlin, C. J.; Kaiser, M.; van der
Hoorn, R. A.; Weerapana, E. Chembiochem 2012, 13, 2327.
(13) Patricelli, M. P.; Szardenings, A. K.; Liyanage, M.; Nomanbhoy,
T. K.; Wu, M.; Weissig, H.; Aban, A.; Chun, D.; Tanner, S.; Kozarich,
J. W. Biochemistry 2007, 46, 350.
(14) Greenbaum, D.; Medzihradszky, K. F.; Burlingame, A.; Bogyo,
M. Chem. Biol. 2000, 7, 569.
(15) Evans, M. J.; Morris, G. M.; Wu, J.; Olson, A. J.; Sorensen, E. J.;
Cravatt, B. F. Mol. Biosyst. 2007, 3, 495.
(16) Bachovchin, D. A.; Ji, T.; Li, W.; Simon, G. M.; Blankman, J. L.;
Adibekian, A.; Hoover, H.; Niessen, S.; Cravatt, B. F. Proc. Natl. Acad.
Sci. U.S.A. 2010, 107, 20941.
(17) Kato, D.; Boatright, K. M.; Berger, A. B.; Nazif, T.; Blum, G.;
Ryan, C.; Chehade, K. A.; Salvesen, G. S.; Bogyo, M. Nat. Chem. Biol.
2005, 1, 33.
(18) Verdoes, M.; Oresic Bender, K.; Segal, E.; van der Linden, W.
A.; Syed, S.; Withana, N. P.; Sanman, L. E.; Bogyo, M. J. Am. Chem.
Soc. 2013, 135, 14726.
(19) Leesnitzer, L. M.; Parks, D. J.; Bledsoe, R. K.; Cobb, J. E.;
Collins, J. L.; Consler, T. G.; Davis, R. G.; Hull-Ryde, E. A.; Lenhard, J.
M.; Patel, L.; Plunket, K. D.; Shenk, J. L.; Stimmel, J. B.; Therapontos,
C.; Willson, T. M.; Blanchard, S. G. Biochemistry 2002, 41, 6640.
(20) Banerjee, R.; Pace, N. J.; Brown, D. R.; Weerapana, E. J. Am.
Chem. Soc. 2013, 135, 2497.
(21) Johnson, C. M.; Monzingo, A. F.; Ke, Z.; Yoon, D. W.; Linsky,
T. W.; Guo, H.; Robertus, J. D.; Fast, W. J. Am. Chem. Soc. 2011, 133,
10951.
(22) Johnson, C. M.; Linsky, T. W.; Yoon, D. W.; Person, M. D.;
Fast, W. J. Am. Chem. Soc. 2011, 133, 1553.
(23) Spokoyny, A. M.; Zou, Y.; Ling, J. J.; Yu, H.; Lin, Y. S.;
Pentelute, B. L. J. Am. Chem. Soc. 2013, 135, 5946.
(24) Ji, X.; Armstrong, R. N.; Gilliland, G. L. Biochemistry 1993, 32,
12949.
ASSOCIATED CONTENT
* Supporting Information
■
S
Detailed synthetic and experimental procedures, NMR and HR-
MS characterization of probes, supplementary figures, and
tables containing complete results of TOP-ABPP experiments.
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
eranthie@bc.edu
■
(25) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B.
Angew. Chem., Int. Ed. 2002, 41, 2596.
(26) Speers, A. E.; Cravatt, B. F. Chem. Biol. 2004, 11, 535.
(27) Miller, J. J. Am. Chem. Soc. 1963, 85, 1628.
(28) Bernasconi, C. F. Chimia 1980, 34, 1.
Notes
The authors declare no competing financial interest.
(29) Speers, A. E.; Cravatt, B. F. J. Am. Chem. Soc. 2005, 127, 10018.
(30) Weerapana, E.; Speers, A. E.; Cravatt, B. F. Nat. Protoc. 2007, 2,
1414.
(31) Eng, J. K.; Mccormack, A. L.; Yates, J. R. J. Am. Soc. Mass
Spectrom. 1994, 5, 976.
ACKNOWLEDGMENTS
■
E.W. is a Damon Runyon-Rachleff Innovator (DRR-18-12). We
are grateful for financial support from the Smith Family
Foundation and Boston College. We thank members of the
Weerapana laboratory for comments and critical reading of the
manuscript and Bo Li for acquiring crystallography data.
REFERENCES
■
(1) Singh, J.; Petter, R. C.; Baillie, T. A.; Whitty, A. Nat. Rev. Drug
Discov. 2011, 10, 307.
3333
dx.doi.org/10.1021/ja4116204 | J. Am. Chem. Soc. 2014, 136, 3330−3333