Facile synthesis and biological evaluation of a cell-permeable probe to detect redox-regulated proteins

Article abstract of DOI:10.1016/j.bmcl.2008.11.073

We have developed an improved synthesis for the cell-permeable, sulfenic acid probe DAz-1. Using DAz-1, we detect sulfenic acid modifications in the cell-cycle regulatory phosphatase Cdc25A. In addition, we show that DAz-1 has superior potency in cells co

Full text of DOI:10.1016/j.bmcl.2008.11.073

Bioorganic & Medicinal Chemistry Letters 19 (2009) 356–359
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Facile synthesis and biological evaluation of a cell-permeable probe
to detect redox-regulated proteins
Young Ho Seo a, Kate S. Carroll a,b,
*
a Life Sciences Institute, University of Michigan, 210 Washtenaw Ave. #4401, Ann Arbor, MI 48109-2216, USA
b Department of Chemistry, University of Michigan, Ann Arbor, MI 48109-2216, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
We have developed an improved synthesis for the cell-permeable, sulfenic acid probe DAz-1. Using
DAz-1, we detect sulfenic acid modifications in the cell-cycle regulatory phosphatase Cdc25A. In addition,
we show that DAz-1 has superior potency in cells compared to a biotinylated derivative. Collectively,
these findings set the stage for the development of activity-based inhibitors of Cdc25 cell-cycle phospha-
tases, which are sensitive to the redox state of the active-site cysteine and demonstrate the advantage of
bioorthogonal conjugation methods to detect protein sulfenic acids in cells.
Received 15 October 2008
Revised 19 November 2008
Accepted 20 November 2008
Available online 24 November 2008
Keywords:
Ó 2008 Elsevier Ltd. All rights reserved.
Redox regulation
Cysteine oxidation
Sulfenic acids
Tyrosine phosphatase
Chemical probes
Staudinger ligation
Click chemistry
The reversible oxidation of cysteine is a widespread mechanism
in the regulation of protein function.1 In mammalian cells, activa-
tion of cell surface receptors can trigger enzymatic production of
reactive oxygen species (ROS). These oxidants function as second
messengers and modify signaling proteins through cysteine oxida-
tion.2,3 For example, oxidation of a critical active-site cysteine
inhibits protein tyrosine phosphatases (PTPs) and results in in-
creased phosphorylation levels.4 Redox-based signal transduction
plays an important role in many normal biological events, includ-
ing cell proliferation, differentiation, migration and programed cell
death.3 In addition, abnormal ROS production correlates with can-
cer and aging-associated degenerative diseases.2,3 Because of the
central role that cysteine oxidation plays in cell signaling and hu-
man pathology, there has been considerable interest in developing
small-molecule probes to detect these post-translational modifica-
tions directly in cells.
lates, including 5,5-dimethyl-1,3-cyclohexadione (1; dimedone,
Figure 1). Under aqueous conditions, dimedone reacts selectively
with cysteine sulfenic acids and not with other protein functional
groups.5–7 This chemoselective reaction has been exploited to
detect sulfenic acid modifications via mass analysis; fluorescent-
dimedone conjugates have also been reported.3 However, these
reagents are not suitable for proteomic studies since they lack an
affinity handle for isolating tagged proteins. To extend the utility
of the dimedone scaffold as a probe for protein sulfenic acid mod-
ifications we recently developed DAz-1 (2; Figure 1).7 DAz-1 cova-
lently modifies sulfenic acid-modified proteins directly in cells and
Sulfenic acid (RSOH) is the simplest cysteine oxoform and is
formed by reaction of a thiolate anion (RSÀ) with cellular oxidants
such as hydrogen peroxide (H2O2). Sulfenic acids can oxidize fur-
ther to sulfinic (RSO2H) and sulfonic (RSO3H) acid, condense with
a neighboring thiolate to form a disulfide, or be stabilized by the
protein microenvironment. The electrophilic sulfur atom in sulfe-
nic acid reacts with carbon nucleophiles such as alkenes and eno-
* Corresponding author. Tel.: +1 734 615 2739.
E-mail address: katesc@umich.edu (K.S. Carroll).
Figure 1. Structures of small-molecule probes for sulfenic acids.
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.11.073

Y. H. Seo, K. S. Carroll / Bioorg. Med. Chem. Lett. 19 (2009) 356–359
357
has an azide chemical handle, which can be functionalized with a
wide variety of phosphine or alkyne-based reporter tags for
detection and isolation.8 In this report, we describe an improved
synthesis of DAz-1 and investigate sulfenic acid formation in the
cell-cycle regulatory phosphatase Cdc25A. In addition, we compare
the activity of DAz-1 to a biotinylated derivative in vitro and in
HeLa cells.
transition and Cdc25B/Cdc25C involved in the G2/M.12 Like all tyro-
sine phosphatases, members of the Cdc family contain an active-
site cysteine involved in formation of a phosphocysteine interme-
diate. The pKa of this active-site residue is significantly perturbed
from the typical 8.5 to ꢀ6.13 Cdc25 phosphatases also possess a
second conserved cysteine located ꢀ5 Å from the active-site resi-
due. This additional cysteine is not required for activity, but can
form an intramolecular disulfide with the catalytic cysteine under
mild oxidizing conditions.14 The proposed function of this disulfide
is to prevent overoxidation of the catalytic cysteine to irreversible
sulfinic and sulfonic acid forms.
Biochemical studies indicate that the active-site cysteine in
Cdc25B and Cdc25C is sensitive to oxidants and that sulfenic acid
formation inhibits phosphatase activity.15 However, it is not
known whether Cdc25A—the only essential Cdc25 isoform—is sus-
ceptible to oxidation. Therefore, we treated the recombinant solu-
ble catalytic domain of Cdc25A C384S16 with DAz-1 and
conjugated it to biotin reporter tags 9 or 10 using the Staudinger
ligation or click chemistry, respectively (Scheme 2). Western blot
analysis shows DAz-1-dependent labeling of Cdc25A C384S
(Fig. 2). Pre-treatment of the phosphatase with the reducing agent
dithiothreitol (DTT) significantly reduced labeling, as expected
(Fig. S1a). Trapping the sulfenic acid modification also blocked for-
mation of disulfide-linked Cdc25A homodimers (Fig. S1b). Taken
together, these data indicate that the active-site cysteine in Cdc25A
can oxidize to a sulfenic acid, which can be trapped by DAz-1. Since
Cdc25A expression is elevated in a wide variety of cancers,12 small-
molecules inhibitors, which target the phosphatase active-site and
are sensitive to the redox state of the catalytic cysteine, may inhibit
proliferation of transformed cells that are associated with high lev-
els of ROS.
In recent work, Charles and colleagues reported a biotinylated
dimedone derivative to monitor protein sulfenic acid formation
in peroxide-treated rat ventricular myocytes.17 In their studies,
Charles et al. observed protein labeling only when primary cells
were treated with hydrogen peroxide.17 One interpretation of
these data is that the basal level of cellular sulfenic acids in pri-
mary myocytes is below the threshold of detection. Alternatively,
since oxidants stimulate programed cell death and necrosis in cul-
tured cells,18,19 it is possible that treatment compromised mem-
brane integrity and allowed the dimedone-biotin derivative to
enter the dying cell. Consistent with the latter proposal, several re-
cent reports demonstrate that direct conjugation of a reporter tag
such as biotin or a fluorophore to an inhibitor reduces potency
and prevents passive diffusion across cell membranes.20,21 For
In a previous report, we pursued the synthesis of DAz-1 via
route A (Scheme 1, a ? b ? c ? f).7 Ni-catalyzed hydrogenation
of 3,5-dihydroxybenzoic acid (7) gave diketone 6 in 85% yield.9 Ini-
tial attempts to couple the carboxylic acid functional group on 6
with 3-azidopropylamine (8)10 using standard peptide coupling re-
agents such as PyBop/TEA, DIC/4-DMAP or TBTU/DIEA were not
successful and produced nucleophilic substitution products at the
carbonyl carbon of compound 6. To prevent these undesired reac-
tions we protected the ketone 6 as the methyl ether. PTSA cata-
lyzed protection was achieved in MeOH at room temperature to
give compound 5 in 97% yield. The amide coupling reaction of car-
boxylic acid 5 with amine 8 was carried out with TBTU and DIEA in
DMF to provide amide 3 in 99% yield. The synthesis of DAz-1 was
completed through deprotection of methyl vinyl ether with 2 N
HCl in THF. The overall yield of DAz-1 through route A was 64%.
In small-scale reactions amide coupling with TBTU, an HOBt-
based aminium salt, afforded DAz-1 in excellent yield. However,
when the reaction was scaled up compound DAz-1 and HOBt were
not completely resolved by silica gel chromatography. Although re-
verse phase HPLC could be used to purify DAz-1 the procedure is
time consuming and reduced the overall yield to 20%. For this rea-
son, we explored an alternate synthetic route B (Scheme 1,
a ? b ? d ? e ? f). Pentafluorophenyl trifluoroacetate (TFAPfp)
can activate carboxylic acids as Pfp esters in good yields and vola-
tile side-products TFA and pentafluorophenol are easily removed
during workup.11 Therefore, we used TFAPfp as the activating re-
agent for the carboxylic acid functional group on 5. Esterification
of carboxylic acid 5 was achieved with TFAPfp and DIEA in DMF
in quantitative yield. Pfp ester 4 was then coupled with 3-azido-
propylamine (8) to furnish compound 3 in high purity. For the final
step of the synthesis, we used cerium (IV) ammonium nitrate
(CAN) to deprotect 3. Non-acidic deprotection using catalytic
amount of CAN provided DAz-1 in 96%. Route B did not require
HPLC purification and furnished DAz-1 over five steps in 52% yield.
Having established an efficient synthesis for DAz-1, we next
investigated sulfenic acid formation in the cell-cycle regulatory
phosphatase Cdc25A (Scheme 2). In humans, three different
Cdc25 family members exist with Cdc25A required for the G1/S
Scheme 1. Synthesis of DAz-1 (1). Reagents and conditions: (a) T1 Raney nickel, NaOH, 750 psi, 70 °C, 85%; (b) PTSA, MeOH, rt, 10 min, 97%; (c) 3-azidopropylamine (8), TBTU,
DIEA, DMF, rt, 15 min, 99%; (d) TFAPfp, TEA, DMF, rt, 2 h, 100%; (e) 3-azidopropylamine (8), DIEA, DMF, rt, 3 min, 89%; (f) 2 N HCl, THF, rt, 1 h, 78% or CAN, H2O-MeCN, reflux,
1 h, 96%.

358
Y. H. Seo, K. S. Carroll / Bioorg. Med. Chem. Lett. 19 (2009) 356–359
Scheme 2. Strategy to detect sulfenic acid-modified Cdc25A with biotinylated reporter tags and the Staudinger ligation (top) or click chemistry (bottom) methods.
Figure 2. DAz-1 detects sulfenic acid modifications in the cell-cycle regulatory
phosphatase Cdc25A. Cdc25A C384S was reacted with 10 mM DAz-1, separated
from excess probe and conjugated to 9 or 10 using standard procedures.8 Samples
were analyzed by Western blot using HRP-conjugated streptavidin (top). Equivalent
protein loading for the Staudinger and click reactions was verified by SDS–PAGE
(bottom).
Figure 3. Immunofluorescence microscopy and Western blot showing sulfenic
acid-modified proteins labeled by DAz-1 in HeLa cells. (a) HeLa cells were incubated
these reasons, we hypothesized that direct conjugation of the bio-
in media containing 0.1 mM DAz-1 (top) or DAz-1-biotin (bottom). After fixation,
tin reporter tag could decrease cellular activity.
DAz-1-modified proteins were conjugated to p-biotin. Biotinylated proteins from
To test this hypothesis, we synthesized DAz-1-biotin 11 (Fig. 1
DAz-1 or DAz-1-biotin treated cells were detected by streptavidin-Alexa Fluor 555
and Scheme S1) and conducted a side-by-side comparison of sulfe-
nic acid labeling in vitro and in cells. To generate DAz-1-biotin, Pfp
ester 5 was coupled to biotin-PEO3-amine (13). Subsequently,
methyl vinyl ether 12 was deprotected with 1 N HCl in THF. The
DAz-1-biotin 11 probe differs from the reagent reported by Charles
et al. only in the absence of a methyl substitutent at the C-5 posi-
tion of the 1,3-cyclohexadione scaffold. In biochemical experi-
ments, DAz-1 and DAz-1-biotin covalently tagged oxidized
glyceraldehyde-3-phosphate-dehydrogenase (GAPDH), as demon-
strated by Western blot detection (Fig. S2a). In competition exper-
(red). Nuclei were counterstained with DAPI (blue). (b) Protein was isolated from
cells incubated in media containing 10 mM DAz-1 or DAz-1-biotin. Protein lysate
from DAz-1 treated cells was conjugated to p-biotin. Biotinylated proteins from
DAz-1 or DAz-1-biotin treated cells were resolved by SDS–PAGE and detected by
Western blot analysis using HRP-conjugated streptavidin (top). Equal protein
loading was verified by reprobing the blot with an antibody against GAPDH
(bottom).
bodies we confirmed the identities of several DAz-1 labeled pro-
teins as GAPDH, actin and peroxiredoxin (Fig. 3b), which are well
known to contain sulfenic acid modifications.1,3 Furthermore, no
significant differences in cell viability were observed among DMSO,
DAz-1-biotin and DAz-1-treated cells (Fig. S3). These data demon-
strate the advantage of DAz-1 and bioorthogonal conjugation
methods for detecting protein sulfenic acids directly in living cells.
In summary, we have reported an improved synthesis for the
sulfenic acid probe DAz-1. This probe was successfully applied to
detect sulfenic acid modifications in the cell-cycle regulatory phos-
phatase Cdc25A. These findings set the stage for development of
activity-based tyrosine phosphatase inhibitors that are sensitive
to the redox state of the active-site cysteine. Furthermore, we show
that DAz-1 and bioorthogonal conjugation with reporter tags is the
method of choice for detecting protein sulfenic acids under physi-
iments, 250 lM DAz-1 was sufficient to block 90% of protein
labeling by 1 mM DAz-1-biotin (Fig. S2). Differences in the extent
and pattern of protein labeling by DAz-1-biotin and DAz-1 were
also observed in HeLa cell lysate (Fig. S2c). In particular, more indi-
vidual sulfenic acid-modified proteins were visualized in DAz-1-
treated samples. Since the specificity of DAz-1 and dimedone
derivatives has been rigorously established in prior work5–7 the
simplest explanation for the observed differences is that the large,
hydrophobic reporter tag on DAz-1-biotin interfered with protein
binding. When tested in intact HeLa cells, DAz-1 was significantly
more potent than DAz-1-biotin as shown by immunofluorescence
microscopy (Fig. 3a) and Western blot (Fig. 3b). Using specific anti-

Y. H. Seo, K. S. Carroll / Bioorg. Med. Chem. Lett. 19 (2009) 356–359
359
3. Poole, L. B.; Nelson, K. J. Curr. Opin. Chem. Biol. 2008, 12, 18. and references
therein.
4. Tonks, N. K. Nat. Rev. 2006, 7, 833. and references therein.
5. Benitez, L. V.; Allison, W. S. J. Biol. Chem. 1974, 249, 6234.
6. Poole, L. B.; Zeng, B. B.; Knaggs, S. A.; Yakubu, M.; King, S. B. Bioconjug. Chem.
2005, 16, 1624.
ological conditions in cells. Bi-functional probes such as DAz-1
should facilitate determination of the specific roles played by sul-
fenic acid modifications in redox-based cell signaling and in regu-
latory mechanisms that involve oxidation of cysteine residues.
7. Reddie, K. G.; Seo, Y. H.; Muse, W. B.; Leonard, S. E.; Carroll, K. S. Mol. Biosyst.
2008, 4, 521.
Acknowledgments
8. Agard, N. J.; Baskin, J. M.; Prescher, J. A.; Lo, A.; Bertozzi, C. R. ACS Chem. Biol.
2006, 1, 644.
9. Van Tamelen, E. E.; Hildahl, G. T. J. Am. Chem. Soc. 1956, 78, 4405.
10. Hatzakis, S. N.; Engelkamp, H.; Velonia, K.; Hofkens, J.; Christianen, P. C. M.;
Svendsen, A.; Patkar, S. A.; Vind, J.; Maan, J. C.; Rowan, A. E.; Nolte, R. J. M. Chem.
Commun. (Camb) 2006, 2012.
11. Green, M.; Berman, J. Tet. Lett. 1990, 31, 5851.
12. Boutros, R.; Dozier, C.; Ducommun, B. Curr. Opin. Cell Biol. 2006, 18, 185. and
references therein.
We thank Prof. Mark Saper for the generous gift of Cdc25A
C384S and Prof. Howard Hang for alkyne-biotin reporter 10. We
also thank the Life Sciences Institute and the Leukemia & Lym-
phoma Society Special Fellows Award #3100-07 and the American
Heart Association Scientist Development Grant #0835419N to
K.S.C. for support of this work.
13. Rudolph, J. Antioxid. Redox Signal. 2005, 7, 761.
14. Rudolph, J. Biochemistry 2007, 46, 3595.
Supplementary data
15. Sohn, J.; Rudolph, J. Biochemistry 2003, 42, 10060.
16. The soluble catalytic domain of Cdc25A consists of residues 336–523; the
back-door cysteine, which is not required for phosphatase activity, has been
changed to serine (Cdc25A C384S).
17. Charles, R. L.; Schroder, E.; May, G.; Free, P.; Gaffney, P. R.; Wait, R.; Begum, S.;
Heads, R. J.; Eaton, P. Mol. Cell. Proteomics 2007, 6, 1473.
18. Kwon, S. H.; Pimentel, D. R.; Remondino, A.; Sawyer, D. B.; Colucci, W. S. J. Mol.
Cell. Cardiol. 2003, 35, 615.
19. Troyano, A.; Sancho, P.; Fernandez, C.; de Blas, E.; Bernardi, P.; Aller, P. Cell
Death Differ. 2003, 10, 889.
20. Cohen, M. S.; Hadjivassiliou, H.; Taunton, J. Nat. Chem. Biol. 2007, 3, 156.
21. Hang, H. C.; Loureiro, J.; Spooner, E.; van der Velden, A. W.; Kim, Y. M.;
Pollington, A. M.; Maehr, R.; Starnbach, M. N.; Ploegh, H. L. ACS Chem. Biol.
2006, 1, 713.
Synthetic procedures, analysis data, and protocols for biochem-
ical studies. Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.bmcl.2008.11.073.
References and notes
1. Reddie, K. G.; Carroll, K. S. Curr. Opin. Chem. Biol. 2008. doi:10.1016/
j.cbpa.2008.07.028. and references therein.
2. Miller, E. W.; Chang, C. J. Curr. Opin. Chem. Biol. 2007, 11, 620. and references
therein.