Mixed isotope photoaffinity reagents for identification of small-molecule targets by mass spectrometry

Article abstract of DOI:10.1002/anie.200600743

(Figure Presented) Twin peaks: A facile and economic route has been developed for the incorporation of deuterated benzophenones (a widely used photoaffinity tag) into probes. Attachment of these mixed-isotope labels to the immunosuppressive drug CsA facil

Full text of DOI:10.1002/anie.200600743

Angewandte
Chemie
Chemical Genetics
DOI: 10.1002/anie.200600743
Mixed Isotope Photoaffinity Reagents for
Identification of Small-Molecule Targets by Mass
Spectrometry**
Shane M. Lamos, Casey J. Krusemark,
Christopher J. McGee, Mark Scalf, Lloyd M. Smith, and
Peter J. Belshaw*
Identifying the protein targets of biologically active small
molecules is often the rate-determining step in the process of
elucidating protein function through chemical genetics.[1]
A
[*] Prof. P. J. Belshaw
Departments of Chemistry and Biochemistry
University of Wisconsin, Madison
Madison, WI 53706 (USA)
Fax: (+1)608-265-4534
E-mail: belshaw@chem.wisc.edu
S. M. Lamos,[+] Dr. M. Scalf, Prof. L. M. Smith
Department of Chemistry
University of Wisconsin, Madison
Madison, WI 53706 (USA)
C. J. Krusemark[+]
Department of Biochemistry
University of Wisconsin, Madison
Madison, WI 53706 (USA)
C. J. McGee
Department of Chemistry
University of California-Irvine
Irvine, CA 92697 (USA)
[+] S.M.L. and C.J.K. contributed equally to this work.
[**] We thank M. Shortreed for assistance with the MS studies, and also
M. Cancilla and A. Hegeman for insightful discussions. C.J.K. was
supported by a Biotechnology Training Grant fellowship. This work
was supported by generous financial support from the NIH
(GM065406) and NHLBI (N01-HV-28182).
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2006, 45, 4329 –4333
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4329

Communications
classical approach for target
identification is chromatography
with immobilized ligands. Affin-
ity chromatography approaches
are often unsuccessful for iden-
tifying binding proteins with low
affinity and low abundance
because of contaminating non-
specifically purified proteins.
These limitations have spurred
the development of new meth-
ods for the discovery of target
ligands.
Photoaffinity labeling (PAL)
is an attractive alternative to
affinity chromatography for the
discovery of target ligands[2]
since matrix and avidity effects
of immobilized ligands are
avoided. PAL methods using
soluble probes are more amena-
ble for the identification of low-
affinity targets since the ligand
concentration can be controlled.
Photoaffinity probes also have
the potential for use inside live
Figure 1. Strategy for determining the identity of small-molecule binding proteins.
cells, thus enabling affinity-based profiling of the entire
proteome in its native state. However, biochemical purifica-
tion and identification of labeled proteins, often facilitated by
the addition of biotin or radioisotopes, remains a significant
challenge in PAL. Here we describe the synthesis and
application of target identification probe (TIP) reagents for
addressing these issues. These reagents contain affinity,
photoaffinity, and mixed isotope labels to facilitate the
identification of binding proteins by mass spectrometry (MS).
Stable mixed isotopes have been applied as selective
identification tags in mass spectrometry. For example, drug
metabolites can be easily identified, even in very complex
mixtures, if isotopes are incorporated to give a unique isotopic
pattern.[3] This approach has been used successfully in the
identification of modification sites on proteins from cova-
lently modifying inhibitors[4] and protein–protein cross-link-
ing reagents.[5] The incorporation of mixed isotopes and
photoaffinity labels into peptide ligands has been shown to
aid the identification of receptor binding pockets[6] and has
been suggested as an approach to aid the identification of
small-molecule PAL targets.[7] To generalize this approach we
employ mixed isotopes within a modular photoaffinity-label-
ing reagent to clearly identify labeled proteins in the presence
of contaminants commonly observed in affinity purification
and MS.
established. Conclusive determination of labeled proteins is
made possible by the presence of peptides bearing the unique
isotopic signature (M, M + 11) of the probe.
This strategy for characterizing the labeled and unlabeled
peptides of chemically modified proteins is advantageous for
identification. Relying on a probe-labeled peptide alone to
identify a protein can be problematic,[8] particularly for photo-
cross-linked peptides.[9] Photo-cross-linking reactions gener-
ally modify peptides with little site specificity, thereby
producing isobaric peptides representing a number of chem-
ical species.[10] Indeed, the MS fragmentation of the probe
itself often results in MS/MS data that are difficult to
interpret.[11] This current approach obviates the need to
sequence labeled peptides, while providing high sequence
coverage and confidence in identifying labeled proteins even
in the presence of contaminants.
The synthesis of the probe began with the preparation of
deuterated benzophenone [D12]-1 by a Friedel–Crafts acyl-
ation using commercially available [D8]toluene and
[D5]benzoyl chloride (Scheme 1). In this way, a large isotopic
mass difference ([D0] versus [D11]) can be incorporated in a
compact and economical manner. Radical bromination of
[D12]-1 with NBS in the presence of a catalytic amount of
AIBN yielded the benzophenone [D11]-2. The commercially
available benzophenone [D0]-2 was used to prepare the
unlabeled series of compounds. Primary amine 3 was alkyl-
ated with [D11]-2 or [D0]-2 to produce secondary amines [D11]-
4 and [D0]-4, respectively. These were further alkylated with
methyl bromoacetate to afford [D11]-5 and [D0]-5. These
compounds were converted into probes [D11]-6 and [D0]-6 by
sequential cleavage of the Boc protecting groups, coupling to
biotin (DIC/HOBt/DIPEA), and amidation with neat ethyl-
ene diamine. The resulting compounds [D11]-6 and [D0]-6
Our approach is outlined in Figure 1. After suitable
attachment of the TIP to a bioactive small molecule, the
conjugate is incubated in a protein mixture and covalently
photoincorporated into target proteins. Avidin affinity chro-
matography is used to purify intact labeled proteins, which are
proteolyzed and analyzed by mass spectrometry. Sequence
information from MS/MS analysis of purified unlabeled
peptides enables a list of candidate binding proteins to be
4330
www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 4329 –4333

Angewandte
Chemie
were used as a 1:1 mixture for subsequent
labeling experiments.
The photo-cross-linking specificity of the
CsA conjugate was assessed within a protein
mixture containing CypA. Upon UV irradia-
tion, the reagent predominantly labeled CypA,
as indicated by Western blot analysis (Fig-
ure 2b, lane 4). The addition of the photolabel
produced an expected shift of the protein band
to about 22 kDa. Photolabeling in the pres-
ence of excess CsA drastically reduced CypA
labeling and slightly increased the labeling of
bystander proteins (Figure 2b, lane 5). The
results indicate that the photo-cross-linking is
highly specific for CypA and is dependent on
the ligand-binding interaction.
To demonstrate the applicability of the
approach, labeled protein was purified from
the protein mixture by using avidin resin
(Figure 3, lane 2). Tryptic digests of eluted
protein were analyzed directly by MS. A
SEQUEST search of the database against the
4 protein sequences with MS/MS data identi-
fied 11 peptides from CypA corresponding to
59% sequence coverage. Single peptides from
both FKBP and ovalbumin were also identi-
fied. A search of the full human database
similarly identified CypA and a number of
spurious proteins common to such searches,
including ubiquitin and keratin. Only two
peptides containing the unique isotopic signa-
ture were found in the mass spectra, and these
were easily identified among many nonlabeled
peptides (Figure 4). These corresponded in
Scheme 1. a) [D8]Toluene, [D5]benzoyl chloride, AlCl3; b) NBS, AIBN, CCl4, reflux 3 h;
c) 3, K2CO3, CH3CN/CH2Cl2 (5:2); d) methyl bromoacetate, DIPEA, CH2Cl2; e) 40% TFA
in CH2Cl2; f) biotin, DIC, HOBt, DIPEA, DMF; g) neat ethylenediamine. Unlabeled
compounds are derived from commercially available bromomethylbenzophenone ([D0]-
2). Boc=tert-butoxycarbonyl, NBS=N-bromosuccinimide, AIBN=azobisisobutyroni-
trile, DIPEA=N,N-diisopropylethylamine, DIC=diisopropylcarbodiimide, HOBt=1-hy-
droxy-1H-benzotriazole, TFA=trifluoroacetic acid.
represent our modular multifunctional scaffold for preparing
photoaffinity reagents.
mass to labeled CypA peptides 92–118 and 56–82 (Figure 4c)
and did not match with the theoretical labeled peptides from
any of the other proteins identified. Both of the identified
peptides are surface exposed and in proximity to the MeBmt
side chain of the CypA–CsA complex.[14] Our results demon-
strate that the presence of mixed-label peptides allows
nonspecific proteins from both the affinity chromatography
and MS analysis to be ignored, thereby allowing the target to
be confidently determined. The incorporation of a large 11-
Da mass difference permitted easy visual recognition of
labeled peptides in the mass
For a proof of principle experiment, we chose the
immunosuppressive drug CsA, whose biological activity is
mediated through binding to cyclophilin A (CypA).[12] We
derivatized CsA by olefin metathesis to its N-methyl-(4R)-4-
[(E)-2-butenyl)-4-methyl]-l-threonine (MeBmt) residue, as
previously described.[13] The modified CsA was appended
onto [D11]-6 and [D0]-6 to give the heavy and light CsA–TIP
conjugates [D11]-7 and [D0]-7, respectively (Scheme 2). These
spectra. This mass difference
enables the resolution of dou-
blets for large peptides and
even
whole
proteins
(ca. 40 kDa), despite the pres-
ence of isotope manifolds
(from natural abundance
heavy isotopes) and thus
could find utility in top-down
proteomics applications. Fur-
ther investigations with these
reagents will include applica-
tions within live cells. In the
event that the reagents are not
Scheme 2. Cyclosporin (CsA)/TIP conjugate.
Angew. Chem. Int. Ed. 2006, 45, 4329 –4333
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4331

Communications
Figure 2. Detection of CsA binding proteins in a protein mixture. A
protein mixture consisting of ovalbumin (OVA), carbonic anhydrase
(CA), CypA, and FK binding protein (FKBP; 70 mgmLÀ1 each) in
phosphate-buffered saline (PBS) was incubated with CsA–TIP reagent
(3.5 mm) and irradiated. Excess CsA (50 mm) was added where
indicated. Portions of 15 mL were subjected to sodium dodecylsulfate/
polyacrylamide gel electrophoresis (SDS-PAGE) analysis. a) Coomassie
Blue stained gel. b) Duplicate Western blot visualized with streptavi-
din-alkaline phosphatase.
Figure 4. Identification of the two peptides containing the unique
isotopic signature in the mass spectra. a) Chromatogram showing the
total ion current; b,c) ms data.
proteins that combines unique isotope signatures with broad
sequence coverage of purified, unlabeled peptide fragments.
These approaches should be broadly applicable to MS
analysis of chemically modified proteins.
Received: February 27, 2006
Published online: May 30, 2006
Keywords: chemical genetics · isotopic labeling ·
mass spectrometry · photoaffinity labeling · proteomics
Figure 3. Purification of the CsA binding protein from the protein
mixture. A protein mixture consisting of ovalbumin (OVA), carbonic
anhydrase (CA), CypA, and FKBP (70 mgmLÀ1 each) in PBS was
incubated with CsA–TIP reagent (3.5 mm) and irradiated. Labeled
species were purified with monomeric avidin resin (500 mL) and
analyzed by SDS-PAGE. Lane 1: 15 mL of protein mixture, lane 2:
elution from avidin stationary phase.
.
[1] a) L. Burdine, T. Kodadek, Chem. Biol. 2004, 11, 593; b) G.
Dorman, F. Darvas, Chemical Genomics (Eds.: F. Darvas, A.
Guttman, G. Dorman), Marcel Dekker, New York, 2004, p. 137.
[2] a) G. Dorman, Bioorganic Chemistry of Biological Signal Trans-
duction, Vol. 211 (Ed.: H. Waldman), Springer, Berlin, 2000;
b) G. Dorman, G. D. Prestwich, Trends Biotechnol. 2000, 18, 64;
c) Y. Hatanaka, Y. Sadakane, Curr. Top. Med. Chem. 2002, 2,
271.
cell permeable, our modular design should readily enable
modifications to increase cell permeability such as replace-
ment of the purification tag and/or linkage to the bioactive
molecule.
[3] T. A. Baillie, Pharmacol. Rev. 1981, 33, 81.
In summary, we have prepared a multifunctional reagent
for the identification of small-molecule binding proteins
through photoaffinity labeling and mass spectrometry. A
facile and economic route for the incorporation of deuterated
benzophenones (a widely used photoaffinity tag) into probes
has been developed. Conjugates to the ligand CsA were
prepared and used to identify a known target of CsA from a
mixture of proteins. We have described a novel strategy for
the high-confidence identification of photoaffinity-labeled
[4] N. L. Kelleher, R. B. Nicewonger, T. P. Begley, F. W. McLafferty,
J. Biol. Chem. 1997, 272, 32215.
[5] a) D. R. Muller, P. Schindler, H. Towbi, U. Wirth, H. Voshol, S.
Hoving, M. O. Steinmetz, Anal. Chem. 2001, 73, 1927; b) J. W.
Back, V. Notenboom, L. J. de Koning, A. O. Muijsers, T. K.
Sixma, C. G de Koster, L. de Jong, Anal. Chem. 2002, 74, 4417.
[6] E. Sachon, O. Tasseau, S. Lavielle, S. Sagan, G. Bolbach, Anal.
Chem. 2003, 75, 6536.
[7] a) M. Daghish, L. Hennig, M. Findeisen, S. Giesa, F. Schumer, H.
Hennig, A. G. Beck-Sickinger, P. Welzel, Angew. Chem. 2002,
4332
www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 4329 –4333

Angewandte
Chemie
114, 2404; Angew. Chem. Int. Ed. 2002, 41, 2293; b) M.
Hashimoto, Y. Hatanaka, Chem. Pharm. Bull. 2004, 52, 1385.
[8] A. E. Speers, B. F. Cravatt, J. Am. Chem. Soc. 2005, 127, 10018.
[9] K. Fang, M. Hashimoto, S. Jockusch, N. J. Turro, K. Nakanishi, J.
Am. Chem. Soc. 1998, 120, 8543.
[10] G. Dorman, G. D. Prestwich, Biochemistry 1994, 33, 5661.
[11] O. V. Borisov, M. B. Goshe, T. P. Conrads, V. S. Rakov, T. D.
Veenstra, R. D. Smith, Anal. Chem. 2002, 74, 2284.
[12] R. E. Handschumacher, M. W. Harding, J. Rice, R. J. Drugge,
D. W. Speicher, Science 1984, 226, 544.
[13] J. A. Smulik, S. T. Diver, F. Pan, J. O. Liu, Org. Lett. 2002, 4,
2051.
[14] C. Weber, G. Wider, B. Von Freyberg, R. Traber, W. Braun, H.
Widmer, K. Wuthrich, Biochemistry 1991, 30, 6563.
Angew. Chem. Int. Ed. 2006, 45, 4329 –4333
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4333