Designing Selectivity in Dirhodium Metallopeptide Catalysts for Protein Modification

Article abstract of DOI:10.1021/acs.bioconjchem.6b00716

The ability to chemically alter proteins is important for broad areas of chemical biology, biophysics, and medicine. Chemical catalysts for protein modification, and particularly rhodium(II) conjugates, represent an important new approach to protein modification that develops novel functionalization approaches while shedding light on the development of selective chemistries in complex environments. Here, we elucidate the reaction parameters that allow selective catalysis and even discrimination among highly similar proteins. Furthermore, we show that quantifying modification allows the measurement of competitive ligand affinity, permitting straightforward measurement of protein-peptide interactions and inhibitors thereof. Taken as a whole, rhodium(II) conjugates replicate many features of enzymes in an entirely chemical construct.

Full text of DOI:10.1021/acs.bioconjchem.6b00716

Article
pubs.acs.org/bc
Designing Selectivity in Dirhodium Metallopeptide Catalysts for
Protein Modification
Samuel C. Martin, Farrukh Vohidov, Haopei Wang, Sarah E. Knudsen, Alex A. Marzec,
and Zachary T. Ball*
Department of Chemistry, Rice University, Houston, Texas 77005, United States
S
* Supporting Information
ABSTRACT: The ability to chemically alter proteins is
important for broad areas of chemical biology, biophysics, and
medicine. Chemical catalysts for protein modification, and
particularly rhodium(II) conjugates, represent an important
new approach to protein modification that develops novel
functionalization approaches while shedding light on the
development of selective chemistries in complex environments.
Here, we elucidate the reaction parameters that allow selective
catalysis and even discrimination among highly similar proteins. Furthermore, we show that quantifying modification allows the
measurement of competitive ligand affinity, permitting straightforward measurement of protein−peptide interactions and
inhibitors thereof. Taken as a whole, rhodium(II) conjugates replicate many features of enzymes in an entirely chemical
construct.
tetrahydrothiophene group of biotin that explains previous
contradictory results and use this discovery to design improved
diazo reagents and improved selectivity in protein modification
reactions. The several effects are combined to demonstrate
orthogonal modification of individual proteins among highly
similar SH3 folds. In addition, we show that quantifying
modification efficiency can be a fast, efficient, and useful way to
measure the potency of inhibitors against protein−protein
interactions.
INTRODUCTION
■
Selective modification, functionalization, or conjugation of
natural proteins remains a daunting challenge. Like related
efforts to perform selective chemistry on nucleic acids1−5 or
secondary metabolites,6−9 modified proteins are broadly useful
in chemical biology, as therapeutics, and as building blocks for
biomaterials. Among other challenges, biopolymers lack unique
reactive handles, and so new selectivity concepts are required.
Ideal methods should enable the modification of specific
individual sites on a target protein and should discriminate
among even highly similar proteins. Finally, simple and versatile
reagents should allow flexible modification with various
functional handles under biologically relevant conditions.
While protein modification has often focused on engineered
substrates or residue-specific chemistry, ligand−catalyst con-
jugates are an emerging approach to site-specific protein
modification.10−15 Proximity-driven modification by transition-
metal catalysts or organocatalysts has numerous advantages,
including access to unique, bioorthogonal chemistries. In
contrast to reactive probes, a catalytic approach separates the
molecular recognition motif (“ligand”) on the catalyst from the
stoichiometric reactive group and functional tag (Figure 1a).
The concentration of ligand in the ligand−catalyst conjugate
can be kept to substoichiometric, biologically relevant levels. In
preliminary efforts, we discovered rhodium(II) conjugates that
act as efficient catalysts for modification at specific side chains
based on molecular recognition.15 Single-protein modification
was possible even in complex environments, such as cell lysate.
In this manuscript, we describe studies on the effects of
several reaction parameters on reaction selectivity. We
elucidate, in turn, the effects of diazo structure, additives, and
reaction pH. We discovered an unanticipated effect of the
FUNCTIONALIZED DIAZO REAGENTS
■
For the accommodation of a variety of modification and
functionalization needs, broadly effective reactivity that
tolerates a variety of functionalized reagents is desirable, and
we have developed a set of simple diazoacetate reagents for
protein modification (Figure 1, 1−5). The highly stabilized
nature of α-styryl-α-diazoacetate derivatives makes them
optimal for protein modification,16−18 in which stabilization
of the metallocarbene intermediate improves reaction efficiency
and minimizes the nonproductive O−H insertion of water, a
common pathway for rhodium metallocarbene intermediates.19
The Francis lab showed,20 and we have subsequently verified,
that some simpler diazo esters (α-phenyl-α-diazoacetates or
even ethyl α-diazoacetate) succeed to some extent in
polypeptide modification, but α-styryl-α-diazoacetates are
significantly more efficient. However, changes to the ester
−OR group are fairly broadly tolerated. Due to the hydro-
Received: December 12, 2016
Revised: December 29, 2016
Published: December 30, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.bioconjchem.6b00716
Bioconjugate Chem. XXXX, XXX, XXX−XXX

Bioconjugate Chemistry
Article
known to provide optimal reactivity and (2) that the functional
handle (alkyne) is linked via a stable amine linker rather than
via the ester linkage. As an added benefit, the newly installed
amine would be charged under physiological conditions, aiding
probe solubility. A new synthetic route was required to produce
a probe with a site of functionalization that would remain intact
after ester cleavage (Scheme 1). We assumed access to diazo 9
Scheme 1. Synthetic Approaches to Alkyne-diazo 9
Figure 1. (a) Protein modification reactions proceed via metal-
locarbene and subsequent formation of functionalized protein. (b) α-
Styryl-α-diazoacetate derivatives are most common for aqueous
protein modification reactions. (c) Preparation of the desthiobiotin−
diazo reagent 5. For full details, see the Supporting Information.
from acid 10. Initially, we planned access to acid 10 by the
Wittig olefination of aldehyde 11, which is readily available by
the alkylation of bromoaldehyde 12, but in our hands, the
subsequent Wittig transformation with the unique phospho-
nium salt 1328 was not effective. Alternatively, desymmetrizing
olefination with phosphonium 1429,30 and the dialdehyde 15
afforded an aldehyde intermediate, and the alkyne group was
subsequently introduced by reductive amination, providing
alkyne 16. However, we were unsuccessful at late-stage
oxidation to the potentially unstable β,γ-unsaturated carboxylic
acid 10 in spite of some limited precedent for similar
transformations.31−34 In the end, a new approach (Scheme 2)
produced the desired target compound by late-stage one-
carbon homologation via cyanide alkylation to alkyne 17
followed by hydrolysis to the styrylacetic acid derivative, which
obviated the need for late-stage oxidation. The route takes
advantage of a recent preparation of aldehyde 18 by DIBAL-H
reduction of cyano-ester 19.35
phobicity of the α-styryl-α-diazoacetate moiety, an oligo-
ethylene glycol linker has been included for solubility in water.
For affinity pull-down and related tasks, biotin is perhaps the
most commonly used group in biological applications.21 For
applications with rhodium(II) catalysts, biotin−diazo conjugate
3 is readily prepared from styrylacetic acid.13 However, unique
(and at times deleterious) reactivity was observed with biotin−
diazo 3, especially in more-demanding modification reactions
with full-length protein substrates (see below for details).
Limitations of the biotin conjugate, together with an interest in
developing more flexible and diverse functionalized diazo
reagents, led to the preparation and study of other diazo
reagents.13,22−24 The synthesis of desthiobiotin-containing
compound 5 avoids reactivity problems of the biotin probe 3
(see below) and also allows for improved release of bound
protein in affinity purification applications.25−27 The synthesis
of desthiobiotin−diazo 5 (Figure 1c) is illustrative of many α-
styryl-α-diazoacetate derivatives, requiring late-stage esterifica-
tion of a functionalized alcohol with styrylacetic acid, followed
by diazo transfer with p-ABSA (see the Supporting Information
for information on the preparation of analogous compounds
1−4.).
Scheme 2. Preparation of Alkyne-diazo 9
A significant weakness of first-generation α-styryl-α-diazo-
acetate derivatives is the ester linkage in reagents 1−5, which is
susceptible to chemical or enzymatic hydrolysis. We encoun-
tered some applications and conditions for which this was a
substantial limitation. Thus, probe 9 was designed to abrogate
these concerns. The most important design features of this
probe are (1) that it preserves the α-styryl-α-diazoester core
B
DOI: 10.1021/acs.bioconjchem.6b00716
Bioconjugate Chem. XXXX, XXX, XXX−XXX

Bioconjugate Chemistry
Article
The new linkage in diazo 9 does prevent hydrolysis-based
loss of the functional handle (Figure 2a). For example, under
Table 1. Influence of Biotin Additive on Protein
Modification Efficiency
a
conversion
(%)
entry
protein
conditions
residue
1
2
3
4
5
6
7
Fyn SH3 R5DRh
Trp119
Trp119
100
0
Figure 2. (a) Protein−mod construct formed by transesterification
with TBHA buffer or by hydrolysis. In either case, any functionality
incorporated into the diazo reagent through an ester linkage is
removed, preventing further manipulation. (b,c) Reaction conditions:
Fyn SH3 (2 μM), R5DRh (1 μM), diazo (100 mM), N-tert-
butylhydroxylamine (TBHA) buffer (100 mM, pH = 6.2), and room
temperature for 4 h. Matrix-assisted laser desorption/ionization−mass
spectrometry (MALDI−MS) data demonstrates the tendency of the
ester linkage to transesterify. Protein modified by diazo 2 no longer
features the alkyne handle; protein modified by diazo 6 retains its
incorporated alkyne. R5DRh: Ac−VSLADRhRPLPPLPP−NH2.15
Fyn SH3 Rh2(OAc)4
Fyn SH3 Rh2(OAc)4, saturated biotin Trp119
Lck SH3 R5DRh
Lck SH3 R5DRh, saturated biotin
Hck SH3 R6ERh
Hck SH3 R6ERh, saturated biotin
25
2
Trp97
Trp97
His93
His93
50
90
20
a
Reaction conditions: protein (10 μM), dirhodium catalyst (5 μM),
diazo 2 (500 μM), and room temperature for 5 h. For Trp mod:
TBHA buffer (100 mM, pH of 6.2; for His mod: Tris buffer (100 mM,
pH of 7.4). Conversion values determined by MALDI−MS
integration. R5DRh: Ac−VSLADRhRPLPPLPP−NH2.15 R6ERh: Ac−
VSLARERhPLPPLPP−NH2.24 In the generic reaction scheme at top,
the native peptide VSL12 is shown bound to the Lck SH3. Residue
numbering is based on the full Lck protein. Lck Trp97 is analogous to
Fyn Trp119; Lck His76 is analogous to Hck His93. PDB files: 2IIM
and 1QWF.
some modification and purification conditions, we see extensive
transesterification of the modification product by the buffer N-
tert-butylhydroxylamine (TBHA), which is often optimal for
protein modification.16,17 In the case of diazo 2, this results in
the loss of the alkyne handle, while transesterification of the
product derived from diazo 9 retains the alkyne handle (Figure
2b,c).
modified product (Figure 3, entry 1). However, further
investigation has revealed that a biotin moiety is essential for
effective reactivity. The alkyne-diazo 2 gives no conversion at all
(entry 2), which is in line with results from proximity-driven
modification, in which low background reactivity was observed.
Simple unfunctionalized diazo compounds such as 1 are
similarly unreactive.
ADDITIVES AND BIOTIN
■
Additives, in the form of coordinating buffers, can significantly
affect protein modification reactions with rhodium(II)
catalysts.16,17 In the course of our investigations, we recognized
that biotin also had significant effects. Typical of our experience
with SH3 domain modification, the Trp119 residue of the Fyn
SH3 domain (see Table 1) could be uniquely modified upon
treatment with a suitable Fyn-binding metallopeptide catalyst
(R5DRh),15 while using the negative control, Rh2(OAc)4, gave
no product (Table 1, entries 1 and 2). In the whole-cell lysate
reaction, we sometimes experienced nonselective background
reaction with biotin−diazo 3 but not with other functionalized
diazo compounds.15 To better understand this result, we added
biotin to protein-modification reactions and observed Fyn
modification, even in negative-control experiments with
Rh2(OAc)4, albeit in decreased yield (entry 3). The increased
tryptophan modification in the presence of biotin is general.
Indeed, including biotin as an additive could be employed to
facilitate modification of a specific tryptophan residue of Lck
(Trp97) that was quite sluggish under normal conditions
(entries 4 and 5).
Figure 3. Protein modification at cysteine using proceeds efficiently
using even Rh2(OAc)4 if a biotin moiety (within the diazo reagent or
exogenous) is present in the reaction mixture. Reaction conditions:
E3g2C peptide (100 μM), Rh2(OAc)4 (10 μM), diazo (2 mM), TBHA
buffer (10 mM, pH of 6.2), and room temperature for 5 h. Conversion
values determined by MALDI−MS integration. E3g2C: EISALEKCI-
SALEQEISALEK.23
The unique effects of biotin on the efficiency of rhodium(II)-
catalyzed diazo reactions extends to a completely different
amino-acid substrate case. We examined cysteine alkylation
with diazo compounds that were catalyzed by Rh2(OAc)4. In
that study, the biotin−diazo reagent, 3, was used to facilitate
analysis. Reagent 3 delivered high conversion to the cysteine-
C
DOI: 10.1021/acs.bioconjchem.6b00716
Bioconjugate Chem. XXXX, XXX, XXX−XXX

Bioconjugate Chemistry
Article
The addition of external biotin could recover some cysteine-
modification activity. Biotin itself resulted in some minimal
observed reactivity, presumably due to poor solubility under the
reaction conditions (Figure 3, entry 3). However, the biotin−
PEG 22, produced by thermal decomposition of the biotin−
diazo reagent, 3,36,37has improved aqueous solubility and
resulted in significantly increased cysteine modification when
used as an additive in reactions with alkyne-diazo 2 (entry 4).
Thus, our initial success with cysteine alkylation was a
serendipitous discovery that relied on the unique reactivity of
the biotin−diazo functionality, which we chose for convenience
in the initial study. We have since verified that the
tetrahydrothiophene ring is necessary for this anomalous
biotin-mediated reactivity. The use of desthiobiotin as an
affinity handle produces diazo compounds such as 5 that react
normally. Furthermore, simple water-soluble thioether com-
pounds, such as 2-methylthioethanol, do not increase diazo
reactivity as does the biotin tetrahydrothiophene functionality.
The reasons for altered reactivity in the presence of biotin−
diazo 3 and the pyrazole control 22 are challenging to assess.
Biotin and other thioether moieties bind RhII axial sites with
modest affinity (Kd ≈ 100 μM in aqueous media).38−40 In spite
of early evidence to the contrary, there is now convincing
evidence that axial ligands can influence the reactivity and
selectivity of rhodium metallocarbene intermediates.41−44
Whatever the exact nature of the effect, biotin additives provide
a convenient way to improve reaction conversion in preparative
applications.
SELECTIVITY
■
Taken together, the effects of catalyst design, of reaction pH,
and of a biotin additive provide impressive and exquisite
control over the design of orthogonal protein reactivity. We
examined a representative collection of Src-family SH3
domains: Hck, Yes, and Lck. These three domains exhibit
significant homology and have 77% sequence similarity,
providing a stringent test for the design of orthogonal catalysts.
Gratifyingly, sequence and structure analysis and optimization
of reaction conditions according to the discoveries outlined
above did indeed allow highly selective single-domain
modification (see Figures 5 and 6). Structural analysis (Figure
5) shows the presence of reactive residues (histidine and
In sharp contrast, modification at histidine residues in the
SH3 fold differs significantly from modification at tryptophan
or cysteine. Biotin strongly inhibits histidine modification, in
contrast to the opposite effect observed with tryptophan or
cysteine (Table 1, entries 6−7). However, histidine mod-
ification is strongly correlated with increasing reaction pH
(Figure 4). For the illustrative example of Lck modification (at
Figure 5. Schematic of the SH3 domain of Lck with the parent peptide
VSL12 bound. Multiple reactive Trp and His residues are conserved
among Hck, Yes, and Lck. The arrows point to the site of attachment
of the catalytic dirhodium core in the various metallopeptides. R5DRh:
Ac−VSLADRhRPLPPLPP−NH2.15 R6ERh: Ac−VSLARERhPLPPLPP−
NH2.24 11LRh: Ac−VSLARRPLPPLRh.15 PDB files: 1BU1, 2HDA,
2IIM, and 1QWF.
tryptophan) on the ligand-binding surface of the SH3 domains,
and labels indicate potential substrate residues the three SH3
domains (Hck, Yes, and Lck). The Yes SH3 domain, uniquely
among the three proteins, has no histidine residues at the
ligand-binding interface, the presence of which we previously
demonstrated leads to slower modification.45 Thus, the catalyst
R5DRh cleanly modifies Yes SH3 at this tryptophan only.
Importantly, this tryptophan residue is conserved across all Src
family SH3 domains, but modification there is not observed in
Hck or Lck (Figure 6, middle row). Because Hck and Lck
possess a conserved histidine residue (His93 and His 76,
respectively), which is close in space to the reactive tryptophan,
neither protein is modified by R5DRh. However, the Lck
domain contains a unique histidine residue, His70, which is
located at the bottom of the binding face. Thus, the catalyst
11LRh, which moves the dirhodium core in the vicinity of this
histidine side chain, produces only single modification at the
His70 of Lck and does not modify either Hck or Yes (Figure 6,
bottom row).
Figure 4. Increasing reaction pH increases efficiency of protein
modification at His residues. The metallopeptide 11LRh was used to
modify Lck SH3. Reaction conditions: Lck SH3 (10 μM), 11LRh (5
μM), diazo 1 (500 μM), Tris buffer (20 mM), and room temperature
for 5 h. 11LRh: Ac−VSLARRPLPPLRh.15
His70) with the 11LRh catalyst,15 histidine modification could
be suppressed below pH 6, and conversion reached quantitative
levels at pH 7.4. In contrast, tryptophan modification displays
slightly decreasing reaction conversion with increasing pH,
consistent with previous studies.16,17 The pH effect with
histidine is consistent with increased imidazole protonation to
an imidazolium at lower pH, thus decreasing the concentration
of available nucleophilic imidazole groups.
Selective Hck modification represents perhaps the most
daunting challenge of these three SH3 domains. The Hck
D
DOI: 10.1021/acs.bioconjchem.6b00716
Bioconjugate Chem. XXXX, XXX, XXX−XXX

Bioconjugate Chemistry
Article
Figure 6. Orthogonal catalysts for the modification of Hck, Yes, and
Lck SH3 domains, demonstrated by MALDI−MS analysis. Reaction
conditions: protein (10 μM), metallopeptide (5 μM), diazo 2 (1 mM),
and room temperature for 5 h. For Trp mod: TBHA buffer (100 mM,
pH of 6.2; for His mod: Tris buffer (100 mM, pH of 7.4). PDB file:
1KIK.
sequence contains two adjacent histidine residues in the ligand-
binding pocket, His93 and His94. Although both are potential
targets for modification, proximate histidine residues can block
modification by simultaneous coordination to both reactive
rhodium atoms.24 Additionally His93 is conserved in Lck, so
selective reaction there is unlikely. Nevertheless, we found that
the R6ERh catalyst24 delivered selective modification of His93
on Hck but did not target the structurally similar His76 on the
Lck domain (Figure 6, top row). Elevated pH (7.4) also
enforces histidine modification and suppresses modification of
Hck, Yes, or Lck at tryptophan. Thus, three different
metallopeptide catalysts allow the individual and selective
modification of three SH3 domains (Hck, Yes, and Lck)
without appreciable cross-reactivity.
Figure 7. (a) MALDI−MS spectra of modification of the Lck SH3
domain in the presence of increasing amounts of S2E peptide,46 which
competes with the 11LRh catalyst15 for the binding site. Reaction
conditions: Lck SH3 (4 μM), 11LRh (2.5 μM), diazo 2 (2 mM), Tris
buffer (100 mM, pH of 7.4), and room temperature for 2 h. (b) A
general scheme of the interactions taking place during the assay;
inhibitor (blue) and rhodium(II) catalyst (green) compete for the
same binding site, with modification blocking further inhibitor-binding
events. (c) Structures of metallopeptide catalyst 11LRh and peptide
inhibitor S2E. (d) Data from (a) plotted with a logistic fit yielding an
IC50 of 15 μM, which is in line with the reported value for VSL12-
derived peptides and Src-family SH3 domains.
QUANTIFICATION
■
Conceptually, proximity-driven protein modification correlates
a noncovalent protein−ligand assembly with a covalently
modified protein product. This correlation can be used to
identify competitive ligand-binding behavior in the presence of
another ligand.14 Extending this concept, quantitative informa-
tion about protein−peptide interactions can be gleaned from
modification reactions in the presence of varying inhibitor
levels (Figure 7). Varied concentrations of a simple peptide
inhibitor of Src-family SH3 interactions (S2E46) displayed
characteristic sigmoidal plots of modification versus log-
[inhibitor]. The IC50 values from modification reactions,
stopped at short reaction times to approximate initial rates,
correlated well with Kd values derived from isothermal titration
calorimetry.
quantification of blot image intensity. Thus, proximity-driven
modification provides a simple way to assess the binding affinity
of ligands for protein−protein interactions.
CONCLUSIONS
■
The rhodium(II) conjugates described exhibit many properties
of natural enzymatic systems despite being completely
designed, nonbiological constructs. The system allows mod-
ification of residues, such as histidine or tryptophan, that are
largely inaccessible to other chemical methods and provides a
general strategy for tag-free chemical modification.14 The
dose−response behavior in the presence of competing protein
ligands provides a fundamentally new assay of inhibition
potency, in analogy to enzyme-based assays. This assay format
succeeds with small amounts of material (∼5 nmol/well) and is
A similar analysis can be conducted to probe the affinity of a
small-molecule inhibitor of Src-family SH3 interactions (Figure
8). As with the peptide inhibitors, quantification of
modification by matrix-assisted laser desorption/ionization−
mass spectrometry (MALDI−MS) as a function of inhibitor
concentration allows a measurement of competitive inhibition.
Relative modification levels are also sufficient to extract affinity
measurements. Additionally, chemical blot visualization of
modified proteins enabled us to extract IC50 values after
E
DOI: 10.1021/acs.bioconjchem.6b00716
Bioconjugate Chem. XXXX, XXX, XXX−XXX

Bioconjugate Chemistry
Article
REFERENCES
■
(1) Tishinov, K., Schmidt, K., Haussinger, D., and Gillingham, D. G.
̈
(2012) Structure-selective catalytic alkylation of DNA and RNA.
Angew. Chem., Int. Ed. 51, 12000−12004.
(2) Song, C.-X., and He, C. (2011) Bioorthogonal labeling of 5-
hydroxymethylcytosine in genomic DNA and diazirine-based DNA
photo-cross-linking probes. Acc. Chem. Res. 44, 709−717.
(3) Crisalli, P., Hernandez, A. R., and Kool, E. T. (2012)
́
Fluorescence quenchers for hydrazone and oxime orthogonal
bioconjugation. Bioconjugate Chem. 23, 1969−1980.
(4) Di Antonio, M., Doria, F., Richter, S. N., Bertipaglia, C., Mella,
M., Sissi, C., Palumbo, M., and Freccero, M. (2009) Quinone methides
tethered to naphthalene diimides as selective G-quadruplex alkylating
agents. J. Am. Chem. Soc. 131, 13132−13141.
(5) Chow, C. S., Mahto, S. K., and Lamichhane, T. N. (2008)
Combined approaches to site-specific modification of RNA. ACS
Chem. Biol. 3, 30−37.
(6) Prescher, J. A., and Bertozzi, C. R. (2005) Chemistry in living
systems. Nat. Chem. Biol. 1, 13−21.
(7) Mahal, L. K. (1997) Engineering chemical reactivity on cell
surfaces through oligosaccharide biosynthesis. Science 276, 1125−1128.
(8) Lewis, C. A., and Miller, S. J. (2006) Site-selective derivatization
and remodeling of erythromycin A by using simple peptide-based
chiral catalysts. Angew. Chem., Int. Ed. 45, 5616−5619.
(9) Lewis, C. A., Merkel, J., and Miller, S. J. (2008) Catalytic site-
selective synthesis and evaluation of a series of erythromycin analogs.
Bioorg. Med. Chem. Lett. 18, 6007−6011.
(10) Wang, H., Koshi, Y., Minato, D., Nonaka, H., Kiyonaka, S., Mori,
Y., Tsukiji, S., and Hamachi, I. (2011) Chemical cell-surface receptor
engineering using affinity-guided, multivalent organocatalysts. J. Am.
Chem. Soc. 133, 12220−12228.
Figure 8. (a) Reported SH3 domain inhibitor 23.47 (b) Addition of
inhibitor 23 slows modification (red arrow) of Yes SH3−MBP fusion
protein by metallopeptide R5DRh. Reaction conditions: protein (4
μM), R5DRh (2.5 μM), diazo 2 (2 mM), TBHA buffer (100 mM, pH
of 6.2), and room temperature of 2 h. (c) Quantification of Fyn SH3
modification after 2 h by MALDI−MS. The observed IC50 value is
consistent with the Kd measured independently by isothermal titration
calorimetry (Kd = 43 μM). (d) As an alternative analytical method,
modification reactions from (b) can be visualized with a fluorogenic
probe (via chemical blotting protocol).48 R5DRh
VSLADRhRPLPPLPP−NH2.15
= Ac−
(11) Tamura, T., and Hamachi, I. (2015) Labeling proteins by
affinity-guided DMAP chemistry. In Site-Specific Protein Labeling
(Gautier, A., and Hinner, M. J., Eds.), pp 229−242, Springer, New
York, NY.
(12) Song, Z., Takaoka, Y., Kioi, Y., Komatsu, K., Tamura, T., Miki,
T., and Hamachi, I. (2015) Extended affinity-guided DMAP chemistry
with a finely tuned acyl donor for intracellular FKBP12 labeling. Chem.
Lett. 44, 333−335.
experimentally simple. As such, it provides an alternative to
more time-consuming and material-intensive assays that are
sometimes required to assess protein−peptide binding.
(13) Chen, Z., Popp, B. V., Bovet, C. L., and Ball, Z. T. (2011) Site-
specific protein modification with a dirhodium metallopeptide catalyst.
ACS Chem. Biol. 6, 920−925.
(14) Minus, M. B., Liu, W., Vohidov, F., Kasembeli, M. M., Long, X.,
Krueger, M., Stevens, A., Kolosov, M. I., Tweardy, D. J., Redell, M. S.,
et al. (2015) Rhodium(II) proximity-labeling identifies a novel target
site on STAT3 for inhibitors with potent anti-leukemia activity. Angew.
Chem., Int. Ed. 54, 13085−13089.
(15) Vohidov, F., Coughlin, J. M., and Ball, Z. T. (2015)
Rhodium(II) metallopeptide catalyst design enables fine control in
selective functionalization of natural SH3 domains. Angew. Chem., Int.
Ed. 54, 4587−4591.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.bioconj-
chem.6b00716.
General experimental details, methods for protein
modification and product analysis, and characterization
of new compounds. (PDF)
(16) Antos, J. M., McFarland, J. M., Iavarone, A. T., and Francis, M.
B. (2009) Chemoselective tryptophan labeling with rhodium
carbenoids at mild pH. J. Am. Chem. Soc. 131, 6301−6308.
(17) Antos, J. M., and Francis, M. B. (2004) Selective tryptophan
modification with rhodium carbenoids in aqueous solution. J. Am.
Chem. Soc. 126, 10256−10257.
(18) Popp, B. V., and Ball, Z. T. (2010) Structure-selective
modification of aromatic side chains with dirhodium metallopeptide
catalysts. J. Am. Chem. Soc. 132, 6660−6662.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: zb1@rice.edu.
ORCID
Zachary T. Ball: 0000-0002-8681-0789
Notes
The authors declare no competing financial interest.
(19) Miller, D. J., and Moody, C. J. (1995) Synthetic applications of
the O−H insertion reactions of carbenes and carbenoids derived from
diazocarbonyl and related diazo compounds. Tetrahedron 51, 10811−
10843.
(20) Antos, J. M. (2001) Selective protein modification with rhodium
carbenoids. Ph.D. Dissertation, University of California at Berkeley,
Berkeley, CA.
ACKNOWLEDGMENTS
■
This work was supported by Welch Foundation under award
no. C-1680 and by the National Science Foundation under
award nos. CHE-1609654, CHE-1055569 and Graduate
Research Fellowship no. 1450681.
F
DOI: 10.1021/acs.bioconjchem.6b00716
Bioconjugate Chem. XXXX, XXX, XXX−XXX

Bioconjugate Chemistry
Article
(21) Bayer, E. A., and Wilchek, M. (1990) Application of avidin
biotin technology to affinity-based separations. J. Chromatogr. A 510,
3−11.
asymmetric Rh(II) catalysis highlights unique features of on-bead
catalyst development. Chem. Sci. 5, 1401−1407.
(42) Zaykov, A. N., and Ball, Z. T. (2011) Kinetic and
stereoselectivity effects of phosphite ligands in dirhodium catalysis.
Tetrahedron 67, 4397−4401.
(22) Sambasivan, R., and Ball, Z. T. (2010) Metallopeptides for
asymmetric dirhodium catalysis. J. Am. Chem. Soc. 132, 9289−9291.
(23) Popp, B. V., and Ball, Z. T. (2011) Proximity-driven
metallopeptide catalysis: remarkable side-chain scope enables mod-
ification of the Fos bZip domain. Chem. Sci. 2, 690−695.
(24) Vohidov, F., Knudsen, S. E., Leonard, P. G., Ohata, J., Wheadon,
M. J., Popp, B. V., Ladbury, J. E., and Ball, Z. T. (2015) Potent and
selective inhibition of SH3 domains with dirhodium metalloinhibitors.
Chem. Sci. 6, 4778−4783.
(43) Pirrung, M. C., Liu, H., and Morehead, A. T. (2002) Rhodium
chemzymes: Michaelis−Menten kinetics in dirhodium(II) carboxylate-
catalyzed carbenoid reactions. J. Am. Chem. Soc. 124, 1014−1023.
(44) Nelson, T. D., Song, Z. J., Thompson, A. S., Zhao, M., DeMarco,
A., Reamer, R. A., Huntington, M. F., Grabowski, E. J. J., and Reider, P.
J. (2000) Rhodium-carbenoid-mediated intermolecular O−H insertion
reactions: a dramatic additive effect. Application in the synthesis of an
ascomycin derivative. Tetrahedron Lett. 41, 1877−1881.
(45) Popp, B. V., Chen, Z., and Ball, Z. T. (2012) Sequence-specific
inhibition of a designed metallopeptide catalyst. Chem. Commun. 48,
7492−7494.
(46) Chen, Z., Vohidov, F., Coughlin, J. M., Stagg, L. J., Arold, S. T.,
Ladbury, J. E., and Ball, Z. T. (2012) Catalytic protein modification
with dirhodium metallopeptides: specificity in designed and natural
systems. J. Am. Chem. Soc. 134, 10138−10145.
(47) Smith, J. A., Jones, R. K., Booker, G. W., and Pyke, S. M. (2008)
Sequential and selective Buchwald−Hartwig amination reactions for
the controlled functionalization of 6-bromo-2-chloroquinoline: syn-
thesis of ligands for the Tec Src Homology 3 domain. J. Org. Chem. 73,
8880−8892.
(48) Ohata, J., Vohidov, F., and Ball, Z. T. (2015) Convenient
analysis of protein modification by chemical blotting with fluorogenic
“click” reagents. Mol. BioSyst. 11, 2846−2849.
(25) Hirsch, J. D., Eslamizar, L., Filanoski, B. J., Malekzadeh, N.,
Haugland, R. P., Beechem, J. M., and Haugland, R. P. (2002) Easily
reversible desthiobiotin binding to streptavidin, avidin, and other
biotin-binding proteins: uses for protein labeling, detection, and
isolation. Anal. Biochem. 308, 343−357.
(26) Hu, B.-H., Jones, M. R., and Messersmith, P. B. (2007) Method
for screening and MALDI−TOF MS sequencing of encoded
combinatorial libraries. Anal. Chem. 79, 7275−7285.
(27) Zhou, Z., Xiang, Y., Tong, A., and Lu, Y. (2014) Simple and
efficient method to purify DNA−protein conjugates and its sensing
applications. Anal. Chem. 86, 3869−3875.
(28) Narayanan, K. S., and Berlin, K. D. (1980) Novel synthesis of ω-
(diphenylphosphinyl)alkylcarboxylic acids from triphenyl-ω-carbox-
yalkylphosphonium salts. J. Org. Chem. 45, 2240−2243.
(29) Wurtenberger, I., Angermaier, B., Kircher, B., and Gust, R.
̈
(2013) Synthesis and in vitro pharmacological behavior of platinum-
(II) complexes containing 1,2-diamino-1-(4-fluorophenyl)-2-alkanol
ligands. J. Med. Chem. 56, 7951−7964.
(30) Ceruti, M., Balliano, G., Viola, F., Grosa, G., Rocco, F., and
Cattel, L. (1992) 2,3-epoxy-10-aza-10,11-dihydrosqualene, a high-
energy intermediate analog inhibitor of 2,3-oxidosqualene cyclase. J.
Med. Chem. 35, 3050−3058.
(31) Baumgartner, H., O’Sullivan, A. C., and Schneider, J. (1997)
The synthesis of oxa-analogs of the kainoid family. Heterocycles 45,
1537−1550.
(32) Frantz, M.-C., Pierce, J. G., Pierce, J. M., Kangying, L., Qingwei,
W., Johnson, M., and Wipf, P. (2011) Large-scale asymmetric synthesis
of the bioprotective agent JP4−039 and analogs. Org. Lett. 13, 2318−
2321.
(33) Andersson, I. E., Dzhambazov, B., Holmdahl, R., Linusson, A.,
and Kihlberg, J. (2007) Probing molecular interactions within class II
MHC A(q)/glycopeptide/T-cell receptor complexes associated with
collagen-induced arthritis. J. Med. Chem. 50, 5627−5643.
(34) Wang, J., Chen, J., Kee, C. W., and Tan, C.-H. (2012)
Enantiodivergent and γ-selective asymmetric allylic amination. Angew.
Chem., Int. Ed. 51, 2382−2386.
(35) Hu, D. X., Shibuya, G. M., and Burns, N. Z. (2013) Catalytic
enantioselective dibromination of allylic alcohols. J. Am. Chem. Soc.
135, 12960−12963.
(36) Brewbaker, J. L., and Hart, H. (1969) Cyclization of 3-
diazoalkenes to pyrazoles. J. Am. Chem. Soc. 91, 711−715.
(37) O’Connor, N. R., Bolgar, P., and Stoltz, B. M. (2016)
Development of a simple system for the oxidation of electron-rich
diazo compounds to ketones. Tetrahedron Lett. 57, 849−851.
(38) Dennis, A. M., Howard, R. A., and Bear, J. L. (1982) The
reactivity of tetra-μ-acetatodirhodium(II) with selected di- and
tripeptides, substituted pyridines and imidazole ligands. Inorg. Chim.
Acta 66, L31−L34.
(39) Das, K., and Bear, J. L. (1976) Complexation of tetra-μ-
carboxylato-dirhodium(II) with imidazole. Inorg. Chem. 15, 2093−
2095.
(40) Drago, R. S., Long, J. R., and Cosmano, R. (1981) Metal
synergism in the coordination chemistry of a metal−metal bonded
system: tetrakis(μ-butyrato)-dirhodium(II). Inorg. Chem. 20, 2920−
2927.
(41) Sambasivan, R., Zheng, W., Burya, S. J., Popp, B. V., Turro, C.,
Clementi, C., and Ball, Z. T. (2014) A tripodal peptide ligand for
G
DOI: 10.1021/acs.bioconjchem.6b00716
Bioconjugate Chem. XXXX, XXX, XXX−XXX