Exploration of the Reactivity of Multivalent Electrophiles for Affinity Labeling: Sulfonyl Fluoride as a Highly Efficient and Selective Label

Article abstract of DOI:10.1002/anie.202104347

Here we explored the reactivity of a set of multivalent electrophiles cofunctionalized with a carbohydrate ligand on gold nanoparticles to achieve efficient affinity labeling for target protein analysis. Evaluation of the reactivity and selectivity of the electrophiles against three different cognate binding proteins identified arylsulfonyl fluoride as the most efficient protein-reactive group in this study. We demonstrated that multivalent arylsulfonyl fluoride probe 4 at 50 nm concentration achieved selective affinity labeling and enrichment of a model protein PNA in cell lysate, which was more effective than photoaffinity probe 1 with arylazide group. Labeling site analysis by LC–MS/MS revealed that the nanoparticle-immobilized arylsulfonyl fluoride group can target multiple amino acid residues around the ligand binding site of the target proteins. Our study highlights the utility of arylsulfonyl fluoride as a highly effective multivalent affinity label suitable for covalently capturing unknown target proteins.

Full text of DOI:10.1002/anie.202104347

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Exploration of the reactivity of multivalent electrophiles for affinity
labeling: sulfonyl fluoride as a highly efficient and selective label
Authors: Kaori Sakurai, Nanako Suto, Shoichi Hosoya, and Shione
Kamoshita
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202104347
Link to VoR: https://doi.org/10.1002/anie.202104347

10.1002/anie.202104347
Angewandte Chemie International Edition
RESEARCH ARTICLE
Exploration of the reactivity of multivalent electrophiles for
affinity labeling: sulfonyl fluoride as a highly efficient and
selective label
Nanako Suto,[a] Shione Kamoshita,[a] Shoichi Hosoya[b] and Kaori Sakurai*[a]
[a]
N. Suto, S. Kamoshita, Prof. K. Sakurai
Department of Bioengineering and Life Science
Tokyo University of Agriculture and Technology
4-24-16,Naka-cho, Koganei-shi, Tokyo, 184-8588, Japan
E-mail: sakuraik@cc.tuat.ac.jp
[b]
S. Hosoya
Institute of Research
Tokyo Medical and Dental University,
1-5-45, Yushima, Bunkyo-ku, Tokyo, 113-8510, Japan
Supporting information for this article is given via a link at the end of the document.
Abstract: Designing suitable affinity labels for elucidating small
molecule target proteins is an important problem in chemical probe-
based approaches to drug target discovery and functional analyses.
Here we explored the reactivity of a set of multivalent electrophiles
cofunctionalized with a carbohydrate ligand on gold-nanoparticles to
achieve efficient affinity labeling for target protein analysis. Evaluation
of the reactivity and selectivity of the electrophiles against three
different cognate binding proteins identified aryl sulfonyl fluoride as
the most efficient protein-reactive group in this study. We
demonstrated that multivalent aryl sulfonyl fluoride probe 4 at 50 nM
concentration achieved selective affinity labeling and enrichment of a
model protein PNA in cell lysate, which was more effective than
photoaffinity probe 1 with aryl azide group. Labeling site analysis by
LC-MS/MS revealed that the nanoparticle-immobilized aryl sulfonyl
fluoride group can target multiple amino acid residues around the
ligand binding site of the target proteins. Our study highlighted the
utility of aryl sulfonyl fluoride as a highly effective multivalent affinity
label suitable for covalently capturing unknown target proteins.
benzophenone, aryl azide, and diazirine are uniquely suited for
target identification based on their highly reactive photo-activated
intermediates that can nonselectively react with any proximal
amino acid residues.[2,10-12] Although photoaffinity labeling has
received a renewed attention as a promising approach, the low
labeling yields and selectivity hamper their routine use in
detecting low-abundance or low-affinity proteins.[13] Another type
of functionality involves electrophilic groups, a wide variety of
which are used in affinity labeling,[14-17] activity-based protein
profiling (ABPP),[3,18-22] reactivity-based protein profiling,[7,22,23]
ligand-directed chemistries,[5,24,25] and covalent drugs[26-30] where
a good balance of the reactivity and chemoselectivity of
electrophiles enables targeting of locally reactive nucleophilic side
chains in proteins such as enzyme active site residues. In contrast
to photoreactive groups, the availability of a diverse array of
electrophiles with a broad range of reactivity and chemoselectivity
represents an attractive resource for researchers in designing
protein-reactive probes. In most of earlier examples, these
electrophilic groups served as a warhead in the ligand-binding
site.[14-17,31,32] More recently, the Hamachi group extensively
demonstrated that it is possible to selectively label surface-
exposed nucleophilic residues at the periphery of the ligand
binding site by various electrophilic probes in a ligand-dependent
fashion[5,24,25] However, utilization of electrophiles for crosslinking
unknown proteins has been limited to cases where the presence
of a nucleophilic amino acid residue at the ligand-binding site of a
target protein is known or predictable.[33-36] This is largely due to
the inherent reactivity of electrophiles with free nucleophiles and
their selectivity toward certain nucleophilic amino acid residues.
Other factors determining the reactivity and selectivity of affinity
labeling by electrophilic groups include the availability of
accessible nucleophilic amino acid residues and their reactivity
toward the electrophiles, which is often dependent on the local
microenvironment.[5,37,38] Therefore, the general applicability of
electrophiles for covalent capture of unknown target proteins with
no structural information or prediction remains to be tested.
We recently reported gold nanoparticle-based photoaffinity
probes that display small-molecule ligands and photoreactive
groups in a multivalent fashion.[39-41] The gold-nanoparticle
scaffold promotes high local concentrations of both the ligands
Introduction
Determining biological targets of small molecules is critical for a
comprehensive understanding of their cellular interactome so that
we can better predict their utility.[1,2] Protein-reactive chemical
probes have become essential tools to facilitate target
identification and engagement analysis of small molecules.[3-7]
For target identification, affinity labeling probes, in which a
protein-reactive group is derivatized on a small molecule of
interest, are employed to covalently crosslink otherwise reversible
protein-ligand complexes for downstream functional analysis.
Since there is no prior knowledge on the structural features of
target proteins, it is a significant challenge to design a suitable
protein-reactive group that can react with target proteins with high
efficiency and exhibits a broad substrate scope, while having no
or low reactivity toward non-targets.[1,8,9] However, no generally
useful protein-reactive group yet fulfills these requirements of
affinity labeling probes.
Two major types of functionalities are currently available as
protein-reactive groups. Photoreactive groups such as
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202104347
Angewandte Chemie International Edition
RESEARCH ARTICLE
(a)
(b)
Figure 1. (a) Scheme showing gold nanoparticle-based affinity labeling for identifying specific carbohydrate-binding proteins. (b) Structures of gold-nanoparticle
affinity labeling probes with a photoreactive group (1), different electrophilic groups (2-5) and a lactose-immobilized gold-nanoparticle probe (6).
and photoreactive groups, and reactions over a large surface area. utility of multivalency in maximizing protein capture efficiency,[42-
As a consequence, these probes exert enhanced protein binding
affinity and ligand-dependent reactivity. With the goal of
expanding the utility of our gold-nanoparticle probes for target
identification, we hypothesized that the low labeling efficiency of
photoreactive groups could be addressed by employing
electrophilic groups as an affinity label instead. A particular
advantage of the gold-nanoparticle system is the modular nature
of surface functionalization, which allows facile optimization of the
probe design. While polar, nucleophilic amino acid residues such
as lysine are preponderant on the protein surfaces, whether they
happen to be within the reach of a probe greatly affects the
outcome of labeling. We anticipated that the multivalent
presentation of a suitable electrophilic group should overcome
this problem by the avidity effect, allowing broad and
simultaneous screening for labeling sites in the proximal area of
the ligand binding site.
46] which is critical for target identification.
Results and Discussion
To explore the potential of gold nanoparticle-based affinity
labeling, we first employed lactose as a model low-affinity ligand.
Lactose is known to bind several lectins such as peanut agglutinin
(PNA), Erythrina cristagalli agglutinin (ECA), and Ricinus
communis agglutinin (RCA) with a binding affinity in the high
micromolar range.[47-49] For affinity labels, we employed a panel
of four different electrophilic groups, comprising a benzyl chloride
(BnCl), squaramide ester (SQ), aryl sulfonyl fluoride (SF), and
acrylamide (AA) (Fig. 1). We reasoned that they represent a
variety of protein-reactive groups with different reaction types and
different sets of target amino acid residues. BnCl potentially re-
acts with a cysteine residue via SN2 reaction,[50] whereas SQ[51,52]
and AA[28,53] serve as Michael acceptors in ABPP probes or
protein crosslinking reagents to react with lysine and cysteine
residues. SF has been found to be an excellent reactant as a part
of covalent drugs or ABPP probes to promote sulfur (IV)-fluoride
exchange (SuFEX) reactions[54,55] with several nucleophilic amino
acid residues including cysteine, lysine, tyrosine, histidine, serine,
and threonine.[56-61] We also employed the aryl azide (ArAz)
group[10,62,63] to incorporate a photoaffinity label to provide a point
of comparison with our previously developed gold nanoparticle-
based photoaffinity probes.[39,40] Several factors were furthermore
considered for the design of the nanoparticle probes that could
affect their function. The gold-nanoparticles tend to agglomerate
in aqueous solution, hence incorporating hydrophilic moiety is
often critical. Since our model ligand, lactose is a hydrophilic
molecule, it was expected to confer colloidal stability to the
nanoparticle probes, as long as the relative ratio of the
Here we designed and synthesized
a set of gold
nanoparticle-immobilized electrophiles as novel multivalent
affinity labels and systematically characterized their reactivity,
selectivity, and targetable amino acid residues. Among the four
different types of electrophilic groups evaluated, aryl sulfonyl
fluoride was identified as the most efficient and selective protein
reactive group toward three different model carbohydrate-binding
proteins. We showed that these gold nanoparticle-immobilized
electrophiles can remain unreactive at high dilution (~nM
concentration) but become highly reactive toward a protein that is
bound to a ligand. The combination of these effects offers an ideal
reactive group that can efficiently crosslink target proteins in a
pseudo-intramolecular setting while being stable toward
bimolecular nucleophilic substitution reactions with nonspecific
proteins.[16] Moreover, the multivalent sulfonyl fluoride probe
allowed labeling of target proteins at several of their surface
residues around the ligand binding site, which demonstrated the
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202104347
Angewandte Chemie International Edition
RESEARCH ARTICLE
intensity of a protein band corresponding to the labeled lectin to
that of an input lectin in an SDS-PAGE gel. Among the five probes,
SF probe 4 gave the highest labeling yields across all three lectins
(25-42%) (Fig. 3a, Table S2).[66] ArAz probe 1 showed ~16%
labeling of the three lectins. BnCl probe 2 showed little ligand-
dependent reactivity toward all lectins. SQ probe 3 and AA probe
5 were rather unreactive toward PNA and ECA but were able to
label RCA at 13-15% yield. RCA has 13 cysteine residues, of
which Cys20 and Cys39 reside in proximity to lactose binding
sites, while PNA and ECA have no cysteine residues.[67,68]
Therefore, cysteine residues may be responsible for the higher
labeling of RCA than PNA and ECA by SQ probe 3 and AA probe
5. Importantly, all probes showed low nonselective labeling of
BSA, which is not a lactose binding protein and is known to
display multiple surface-exposed lysine and cysteine residues.[69]
Time-course analysis of the labeling reaction of PNA with probes
2-5 showed that a longer reaction time (16 h) allowed SF probe 4
to achieve quantitative labeling (Fig. 3b). No further increase in
the labeling products was observed for the other electrophilic
probes (2, 3, and 5). These results underscored the unique
reactivity of the SF group, which has sufficient reactivity toward
proximal protein residues while being resistant toward
hydrolysis.[54,55] Taken together, the comparative reactivity
analysis found SF probe 4 to display high ligand-dependent
reactivity in general, which was superior to the reactivity profile of
ArAz photoaffinity probe 1.
To evaluate the labeling selectivity, we next assessed the
efficiency of affinity labeling by gold-nanoparticle probes 1-5 with
a mixed solution of PNA and 5-fold excess BSA. Washing with a
high concentration of lactose solution[70] confirmed that SF probe
4 indeed covalently labeled PNA with high efficiency (Fig. S4, lane
8). The time-course analysis demonstrated that SF probe 4
maintained labeling selectivity toward PNA over a duration of 16
h (Fig. 3c, Fig. S5). Its labeling yields were slightly lower than
when reacted alone with PNA as shown in Fig. 3b, likely due to
sequestration by BSA. We next reacted gold-nanoparticle probes
(1-5) with HL-60 cell lysate containing PNA (0.8 w/w%). To ensure
enrichment of covalently labeled proteins from unreacted lysate
proteins, stringent washing conditions involving 2% SDS solution
were applied. Despite the presence of excess amounts of various
nonspecific proteins, all probes selectively enriched PNA (Fig. 3d,
Fig. S6a). Moreover, SF probe 4 exhibited the highest labeling
efficiency (37%) and selectivity, which were superior to those of
photoaffinity probe 1 (Fig. 3d, lane1 and 4). The ligand-
dependence of PNA labeling by SF probe 4 was verified by a
negative control experiment where excess amounts of free
lactose ligand inhibited the labeling reaction (Fig. 3d, lane 6). The
labeling of PNA was also demonstrated by reacting probes 1-5
with fluorescently-tagged, FITC-PNA instead of PNA in cell lysate,
which showed probe 4 as most efficient (Fig. S6b-c). These
results thus suggested that SF is a promising protein reactive
group suited for covalent labeling of target proteins in a complex
proteome.
Figure 2. Characterization of functionalized gold-nanoparticle probes 1-5. (a)
UV-VIS spectra of citrated-coated gold-nanoparticles and probes 1-5. (b)
Agarose gel analysis of probes 1-6. C: Citrate-coated gold-nanoparticles, B:
BSPP-stabilized gold-nanoparticles.
electrophilic group, which is usually hydrophobic, was kept
relatively low. To achieve optimal binding affinity and labeling
efficiency, we anticipated that the ratio of ligand to a reactive
group around 2:1 would be a good starting point since the high
local concentration of both a ligand moiety and an electrophilic
group should promote protein binding as well as efficient labeling
reaction. It was also expected that the slightly higher fraction of a
hydrophilic ligand to an electrophilic group should confer colloidal
stability.
Therefore, we designed and synthesized five types of gold
nanoparticle-based probes (1-5), with a lactose ligand and one of
five different groups (four electrophilic groups and one
photoreactive group) at a 2:1 ratio to evaluate which group
enables efficient crosslinking of binding proteins. The lactose
ligand and electrophiles were derivatized with lipoic acid for
immobilizing onto the surface of gold-nanoparticles with
approximately 13-nm diameter through bivalent Au-S bonds by a
two-step ligand exchange method.[39-41] Functionalized gold-
nanoparticles were characterized by transmission electron
microscopy (TEM), UV-visible spectroscopy (UV-VIS), agarose
gel electrophoresis, dynamic light scattering (DLS), and matrix-
assisted laser desorption/ionization spectrometry (MALDI-MS)
analysis (Fig. 2, Fig. S1-S2, Table S1). To verify that the presence
of multivalent ligands on a gold-nanoparticle scaffold increases
affinity,[39,64] the Kd values of lactose-functionalized gold-
nanoparticle 6 were measured with PNA, ECA, and RCA by the
previously developed pull-down assay.[39] Lactose-gold
nanoparticle 6 exhibited high affinity toward all three lectins with
Kd values of 26 nM (PNA), 8.7 nM (ECA), and 18 nM (RCA) (Fig.
S3). Thus the multivalent probe achieved a large affinity
enhancement of approximately 15000-fold compared to
monomeric lactose (Kd = 400 µM for PNA, 323 µM for ECA, and
37 µM for RCA).[47-49,65] In addition, agglutination of nanoprobes
by lectins may contribute to the overall affinity enhancement by
forming large complexes composed of three-dimensional
networks of lectins and probes. Such complexes are likely to be
kinetically stable.[44-46]
The ligand-dependent reactivity of gold-nanoparticle probes 1-
5 was investigated upon binding with PNA, ECA, or RCA. Probes
2-5 at a concentration of 50 nM were incubated with a lectin (0.4
µM as a monomer) at 4 °C for 2 h and the labeled and enriched
proteins were analyzed by sodium dodecyl sulfate polyacrylamide
gel electrophoresis (SDS-PAGE) and fluorescence imaging. The
control photoaffinity probe 1 was irradiated with UV at 365 nm at
4 °C for 30 min according to our previous study.[39] The labeling
efficiency was determined based on the relative fluorescence
To characterize the amino acid residues targetable by the
multivalent affinity labeling probes, we chose the SF probe 4,
which displayed the highest labeling efficiency for all three lectins
and low nonspecific labeling, to determine its labeling sites. Since
our gold-nanoparticle probes were designed to display both ligand
7 and protein reactive group 8c in a multivalent fashion, it was
anticipated that multiple sites in the proximity of the ligand binding
site could be labeled. Using the optimal labeling conditions found
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202104347
Angewandte Chemie International Edition
RESEARCH ARTICLE
Figure 3. (a) Top: SDS-PAGE analysis of the affinity labeling of PNA, ECA, RCA, and BSA (monomer: 0.4 μM) by probes 1-5 (50 nM). CTL: protein used for a given
reaction; Wash: washing three times with 0.5 M lactose-1% n-octyl-β-D-glucoside-6.5% PEG8000 in PBS. Bottom: labeling efficiency of each protein by probes 1-
5. The % yields were calculated based on the amount of protein used for a given reaction. Error bars show standard errors (n=3). (b) Top: SDS-PAGE time-course
analysis of affinity labeling of PNA (0.1 μM) by probes 2-5 (50 nM). CTL: PNA used for a given reaction. Bottom: labeling efficiency of PNA by probes 2-5. The %
yields were calculated based on the amounts of PNA used for a given reaction. (c) Time course analysis of affinity labeling of PNA (100 nM) by probe 4 (50 nM) in
the presence of excess BSA (1.5 µM). (d) SDS-PAGE analysis of affinity labeling of PNA (0.4 µg, 50 nM) spiked in cell lysate (5 µg) using probes 1-5 (50 nM). M:
molecular weight marker; L: lysate 5 µg (1/10); P: 0.4 µg PNA; Lac+: 20 mM lactose.
in Fig. 3c, 10 µg of PNA or RCA was reacted with probe 4 (46
pmol) at 4 °C for 16 h, and the labeled proteins were purified by a
few cycles of centrifugation and stringent washing with excess
lactose followed by SDS-PAGE. The gel bands were excised and
then subjected to in-gel trypsin digestion and liquid
chromatography-tandem mass spectrometry (LC-MS/MS)
analysis.[71] Representative MS/MS spectra for a digested peptide
fragment carrying a protein-reacted moiety are shown in Fig. 4a.
Several peptides were detected with mass values increased by
the expected mass values of the protein-reacted moiety 17 or its
iodoacetamide derivatives 17a and 17b, its fragments 17c and
17d, or its oxidized derivative 17e (Fig. S8; 17: 547.16, 17a:
663.22, 17b: 665.19, 17c: 359.13, 17d: 184.99, 17e: 579.15). The
labeled peptides were identified from triplicate experiments, which
revealed a remarkable scope of the labeling reaction by
multivalent SF probe 4. Multiple labeled sites close to the ligand
binding site in PNA and RCA were detected at several types of
amino acid residues (Fig. 4b, c, Fig. S9-10, Table S3). Among the
labeled amino acid residues, tyrosine notably exhibited the
highest occurrence (three residues for each lectin), which is
consistent with previous studies using SF-containing covalent
drugs or ABPP probes.[56-61] The SF group has been found to
preferentially react with tyrosine and lysine and to a lesser degree
with histidine, cysteine, serine, and threonine. Other labeled
residues included lysine, aspartic acid, and asparagine,
suggesting that SF probe 4 can potentially target diverse types of
amino acid residues. These results supported our hypothesis that
when the target proteins are specifically bound to ligands on the
nanoparticle, multivalent SF groups are positioned in such a way
that some can reach out and readily react with matching
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202104347
Angewandte Chemie International Edition
RESEARCH ARTICLE
Figure 4. (a) LC-MS/MS spectrum of the peptide fragment labeled with 17, supporting the labeling of PNA Y79 by SF probe 4. The inferred labeled amino acid
residues mapped onto the X-ray structure of (b) PNA (PDB: 2PEL) and (c) RCA (PDB: 1RZO) are visualized in magenta. Bound lactose (PNA) and galactose (RCA)
in the crystal structures are shown as green sticks. The chain A residues for RCA are indicated by ' in the figure.
nucleophilic residues on the protein surface.
N-acetylated monomeric amino acids at 100 mM in
hydroxyethylpiperazine ethanesulfonic acid (HEPES) buffer at pH
7.3 and analyzed the time-dcourse of the formation of the
crosslinked products by LC/MS. As shown in Fig. 5a, the
formation of an adduct (23a) was observed only with N-acetyl-
lysine 19 after prolonged reaction time. In contrast, N-acetyl-
tyrosine 20, N-acetyl-aspartic acid 21, or N-acetyl-asparagine 22
did not show reactivity toward a small-molecule sulfonyl fluoride
reagent 18 even after 48 hours (Fig. S11). When the reactivity of
these N-acetylated monomeric amino acids at 100 mM toward
probe 4 at 50 nM in HEPES buffer (pH 7.3) was evaluated, none
of the tested amino acids yielded an adduct after 48 hours (Fig.
5b, Fig. S12). The SF groups on probe 4 hydrolyzed only slowly
over 48 hours. Our finding is consistent with the literature that aryl
sulfonyl fluoride group is relatively stable in neutral aqueous
solution and yet can be readily activated toward nucleophilic
amino acid residues of the bound proteins.[38, 55] According to the
mechanism of affinity labeling,[16] the different reactivity toward
probe 4 observed for monomeric amino acid residues and the
amino acid residues on the protein may be explained by the
increased effective molarity of the latter relative to the SF groups.
Additionally, the avidity effect of multivalent SF groups may
contribute to a selective increase in the rate of affinity labeling by
keeping the probe concentration at nanomolar. It thus suggested
that the unique reactivity feature of aryl sulfonyl fluoride, with
combined effects of the increased effective molarity and avidity
effect promoted highly efficient affinity labeling by probe 4. It is
also possible that some of the labeled amino acids identified in
PNA and RCA represent hyperactivated residues[79,80] by
When mapped to the X-ray structures of ligand-bound PNA[72]
and RCA,[73] the detected labeled sites are all located within
approximately 15 Å and 18 Å of their ligand binding sites. The
most prominent labeled sites in PNA reside within a loop region
harboring Asp75-Lys112, which is in close proximity (~14 Å) to
the lactose binding pocket. In RCA, which is a dimer of an α/β
heterodimer, major labeled sites were found in the loop region
composed of Asp220-Arg236 close to one of the three galactose
binding sites in chain B.[67,68] These flexible loop regions may
provide readily accessible nucleophilic residues for reacting with
the SF groups on the gold-nanoparticle probes. Lysine is known
to be one of the most abundant amino acid residues on protein
surfaces.[74] Lysine, tyrosine, and aspartic acid, together comprise
approximately 24% of the amino acid residues at the surfaces of
proteins.[75] If we estimate the contact surface area between
probe 4 and its target protein to be an area of a circle with a
diameter of ~30 Å, then it seems possible for multivalent SF
groups in a nanoparticle probe to find and react with one or more
lysine and tyrosine residues in the proximity of the ligand binding
site.
It has been shown that the reactivity of electrophiles toward
monomeric amino acids is not necessarily reflected by their
reactivity toward proteins.[76-78] We wondered whether the labeled
amino acid residues in PNA and RCA could be generated by
reaction with the SF group due to their inherent reactivity or due
to a specific molecular setting in which the protein and the SF
group are juxtaposed on a gold-nanoparticle surface. To test this
notion, we reacted 4-(fluorosulfonyl) benzoic acid 18 at 1 mM with
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202104347
Angewandte Chemie International Edition
RESEARCH ARTICLE
Figure 5. (a) HPLC traces monitoring the solution reactivity of 18 toward N-acetyl-lysine 19 at λ = 254 nm. Reaction mixtures after 0, 4, 16, 24, and 48 h were
analyzed. 23a was confirmed by ESI-MS. 23a: LRMS (ESI-TOF) calculated for C15H21N2O7S [M+H]+: 373.11; found 373.11. (b) Extracted ion chromatograms for
m/z=567.1669±0.0500 (8c: C23H36FN2O7S3 [M+H]+) and 565.1712±0.0500 (8c’: C23H37N2O8S3 [M+H]+) present in the reaction mixture generated by incubating SF
probe 4 and 19 for 48 h.
proximal amino acid residues, which display inherently high
reactivity toward electrophilic reagents.[81]
effect by providing a large reactive surface area toward a target
protein. Further analysis with monomeric amino acid derivatives
validated that the aryl sulfonyl fluoride groups on gold-
nanoparticles exhibit high reactivity toward various amino acids
only when the proteins are bound to the nanoparticle surface.
Notably, that SF group did not show any cross-reactivity with the
lactose, a polyol moiety despite the proximity.[82] The cross-
reactivity between the ligand moiety and electrophilic groups co-
functionalized on the nanoparticles is undesirable as it would
result in deactivated affinity labeling probes before use. When
ligands contain nucleophilic functionalities such as phenols or
amines, they are expected to display cross-reactivity with the
electrophilic moiety cofunctionalized on the nanoparticles. In such
cases, it may be more beneficial to pursue photoaffinity labeling
for target analysis with photoreactive groups instead of
electrophilic groups as labeling groups. Alternatively, the cross-
reactivity between a ligand and an electrophilic group may be
circumvented by optimizing linker length so that they would not be
positioned to react with each other. Taken together, these results
highlighted the unique reactivity of aryl sulfonyl fluoride group
Conclusion
In summary, we developed a novel gold nanoparticle-based
multivalent approach to covalently capture target proteins by
electrophilic groups. Evaluating four different electrophilic groups
against three different lectins identified aryl sulfonyl fluoride group
as the most efficient and selective affinity label. Significantly, we
showed that multivalent aryl sulfonyl fluoride probe 4 at 50 nM
concentration achieved selective affinity labeling of PNA in cell
lysate, which was more effective than photoaffinity probe 1 with
aryl azide group. It thus demonstrated a promising utility of
multivalent aryl sulfonyl fluoride affinity probe as a complementary
target identification tool. LC-MS/MS analysis revealed that the
gold nanoparticle-immobilized aryl sulfonyl fluoride groups can
label several amino acid residues around the ligand binding site.
It supported the view that multivalent display of electrophilic
groups should lead to efficient affinity labeling due to the avidity
suited as
a multivalent affinity label. Our findings have
6
This article is protected by copyright. All rights reserved.

10.1002/anie.202104347
Angewandte Chemie International Edition
RESEARCH ARTICLE
Owens, E. Verner, K. A. Brameld, R. J. Hill, D. M. Goldstein, J. Taunton,
Nat. Chem. Biol. 2015, 11, 525-531.
implications for the new design of chemical probes for target
protein analysis.
[28] T. A. Baillie, Angew. Chem. Int. Ed. 2016, 55, 13408-13421, Angew.
Chem. 2016, 128, 13606-13619.
[29] E. A. Grossman, C. C. Ward, J. N. Spradlin, L. A. Bateman, T. R.
Huffman, D. K. Miyamoto, J. I. Kleinman, D. K. Nomura, Cell Chem. Biol.
2017, 24, 1368-1376.e4.
Acknowledgments
[30] N. Shindo, H. Fuchida, M. Sato, K. Watari, T. Shibata, K. Kuwata, C.
Miura, K. Okamoto, Y. Hatsuyama, K. Tokunaga, S. Sakamoto, S.
Morimoto, Y. Abe, M. Shiroishi, J. M. M. Caaveiro, T. Ueda, T. Tamura,
N. Matsunaga, T. Nakao, S. Koyanagi, S. Ohdo, Y. Yamaguchi, I.
Hamachi, M. Ono, A. Ojida, Nat. Chem. Biol. 2019, 15, 250-258.
[31] J. L. Counihan, B. Ford, D. K. Nomura, Curr. Opin. Chem. Biol. 2016, 30,
68-76.
The authors thank Prof. Keiichi Noguchi for technical assistance
with LC/MS analysis and TEM analysis, Prof. Naoyuki Sugiyama,
and Dr. Seketsu Fukuzawa for helpful discussion on the LC-
MS/MS analysis. This study was supported by KAKENHI
18K05331 and 17H06110 by JSPS. The authors are grateful for
supports from JSPS A3 Foresight Program, ACBI and GIR
programs.
[32] M. J. C. Long, Y. Aye, Cell Chem. Biol. 2017, 24, 787-800.
[33] B. G. Hoffstrom, A. Kaplan, R. Letso, R. S. Schmid, G. J. Turmel, D. C.
Lo, B. R. Stockwell, Nat. Chem. Biol. 2010, 6, 900-906.
[34] P. Y. Yang, K. Liu, M. H. Ngai, M. J. Lear, R. Wenk, S. Q. Yao, J. Am.
Chem. Soc. 2010, 132, 656-666.
Keywords: Affinity labeling • protein modifications •
carbohydrates • nanoparticles • multivalency
[35] T. Wirth, K. Schmuck, L. F. Tietze, S. A. Sieber, Angew. Chem. Int. Ed.
2012, 51, 2874-2877, Angew. Chem. 2012, 124, 2928-2931.
[36] J. A. Clulow, E. M. Storck, T. Lanyon-Hogg, K. A. Kalesh, L. H. Jones, E.
W. Tate, Chem. Commun. 2017, 53, 5182-5185.
[1]
[2]
[3]
[4]
[5]
M. Schenone, V. Dančík, B. K. Wagner, P. A. Clemons, Nat. Chem. Biol.
2013, 9, 232-240.
S. Ziegler, V. Pries, C. Hedberg, H. Waldmann, Angew. Chem. Int. Ed.
2013, 52, 2744-2792, Angew. Chem. 2013, 125, 2808-2859.
D. Greenbaum, A. Baruch, L. Hayrapetian, Z. Darula, A. Burlingame, K.
F. Medzihradszky, M. Bogyo, Mol. Cell. Proteomics 2002, 1, 60-68.
B. F. Cravatt, A. T. Wright, J. W. Kozarich, Annu. Rev. Biochem. 2008,
77, 383-414.
[37] U. Haedke, E. V. Küttler, O. Vosyka, Y. Yang, S. H. L. Verhelst, Curr.
Opin. Chem. Biol. 2013, 17, 102-109.
[38] O. O. Fadeyi, L. R. Hoth, C. Choi, X. Feng, A. Gopalsamy, E. C. Hett, R.
E., Jr. Kyne, R. P. Robinson, L. H. Jones, ACS Chem. Biol. 2017, 12,
2015-2020.
[39] K. Sakurai, Y. Hatai, A. Okada, Chem. Sci. 2016, 7, 702-706.
[40] K. Sakurai, A. Kato, K. Adachi, Bioorg. Med. Chem. 2018, 28, 3227-3230.
[41] K. Mori, K. Sakurai, Org. Biomol. Chem. 2021, 19, 1268-1273.
[42] L. L. Kiessling, J. E. Gestwicki, L. E. Strong, Angew. Chem. Int. Ed. 2006,
45, 2348-2368, Angew. Chem. 2006,118, 2408-2429.
K. Shiraiwa, R. Cheng, H. Nonaka, T. Tamura, I. Hamachi, Cell Chem.
Biol. 2020, 27, 970-985.
[6]
[7]
M. H. Wright, S. A. Sieber, Nat. Prod. Rep. 2016, 33, 681-708.
A. M. Roberts, C. C. Ward, D. K. Nomura, Curr. Opin. Biotechnol. 2017,
43, 25-33.
[43] M. Mammen, S. K. Choi, G. M. Whitesides, Angew. Chem. Int. Ed. 1998,
37, 2754- 2794, Angew. Chem. 1998, 110, 2908-2953.
[8]
[9]
A. Blencowe, W. Hayes, Soft Matter, 2005, 1, 178-205.
W. Zheng, G. Li, X. Li, Arch. Pharm. Res. 2015, 38, 1661-1685.
[44] P. I. Kitov, D. R. Bundle, J. Am. Chem. Soc. 2003, 125, 52, 16271-16284.
[45] J. J. Lundquist, E. J. Toone, Chem. Rev. 2002, 102, 555-578.
[46] R. J. Pieters, Org. Biomol. Chem. 2009, 7, 2013-2025.
[47] A. J. Cagnoni, O. Varela, M. L. Uhrig, J. Kovensky, Eur. J. Org. Chem.
2013, 5, 972-983.
[10] F. Kotzyba-Hibert, I. Kapfer, M. Goeldner, Angew. Chem. Int. Ed. 1995,
34, 1296-1312, Angew. Chem. 1995, 107, 1391-1408.
[11] S. S. Ge, B. Chen, Y. Y. Wu, Q. S. Long, Y. L. Zhao, P. Y. Wang, S.
Yang, RSC Adv. 2018, 8, 29428-29454.
[12] J. P. Holland, M. Gut, S. Klingler, R. Fay, A. Guillou, Chem. Eur. J. 2020,
26, 33-48.
[48] C. Svensson, S. Teneberg, C. L. Nilsson, A. Kjellberg, F. P. Schwarz, N.
Sharon, U. Krengel, J. Mol. Biol. 2002, 321, 69-83.
[13] M. Hashimoto, Y. Hatanaka, Eur. J. Org. Chem. 2008, 15, 2513-2523.
[14] L. Wofsy, H. Metzger, S. J. Singer, Biochemistry 1962, 1, 1031-1039.
[15] F. Wold, Methods Enzymol. 1977, 46, 3-14.
[49] S. Sharma, S. Bharadwaj, A. Surolia, S. K. Podder, Biochem. J. 1998,
333, 539-542.
[50] G. Chen, A. Heim, D. Riether, D. Yee, Y. Milgrom, M. A. Gawinowicz, D.
Sames, J. Am. Chem. Soc. 2003, 125, 27, 8130-8133.
[16] D. Givol, M. Wilchek, in Encyclopedia of Immunology 2nd Ed., Vol. 2
(Eds.: P. Delve, I. Roitt), Academic Press, United States, 1998, pp 50-52.
[17] T. Hayashi, I. Hamachi, Acc. Chem. Res. 2012, 45, 1460-1469.
[18] M. Uttamchandani, J. Li, H. Sun, S. Q. Yao, S. Q. ChemBioChem 2008,
9, 667-675.
[51] C. Dingels, F. Wurm, M. Wagner, H. A. Klok, H. Frey, Chem. Eur. J. 2012,
18, 16828-16835.
[52] M. Fonvielle, N. Sakkas, L. Iannazzo, C. L. Fournis, D. Patin, D. Mengin-
Lecreulx, A. El-Sagheer, E. Braud, S. Cardon, T. Brown, M. Arthur, M.
Etheve-Quelquejeu, Angew. Chem. Int. Ed. 2016, 55, 13553-13557,
Angew. Chem. 2016,128, 13751-13755.
[19] M. J. Niphakis, B. F. Cravatt, Annu. Rev. Biochem. 2014, 83, 341-377.
[20] P. Yang, K. Liu, ChemBioChem 2015, 16, 712-724.
[21] B. P. Kok, S. Ghimire, W. Kim, S. Chatterjee, T. Johns, S. Kitamura, J.
Eberhardt, D. Ogasawara, J. Xu, A. Sukiasyan, S. M. Kim, C. Godio, J.
M. Bittencourt, M. Cameron, A. Galmozzi, S. Forli, D. W. Wolan, B. F.
Cravatt, D. L. Boger, E. Saez, Nat. Chem. Biol. 2020, 16, 997-1005.
[22] H. J. Benns, C. J. Wincott, E. W. Tate, M. A. Child, Curr. Opin. Chem.
Biol. 2021, 60, 20-29.
[53] K. M. Backus, B. E. Correia, K. M. Lum, S. Forli, B. D. Horning, G. E.
González-Páez, S. Chatterjee, B. R. Lanning, J. R. Teijaro, A. J. Olson,
D. W. Wolan, B. F. Cravatt, Nature, 2016, 534, 570-574.
[54] J. Dong, L. Krasnova, M. G. Finn, K. B. Sharpless, Angew. Chem. Int.
Ed. 2014, 53, 9430-9448, Angew. Chem. 2014,126, 9545-9545
[55] A. S. Barrow, C. J. Smedley, Q. Zheng, S. Li, J. Dong, J. E. Moses, Chem.
Soc. Rev. 2019, 48, 4731-4758.
[23] C. C. Ward, J. I. Kleinman, S. M. Brittain, P. S. Lee, C. Y. S. Chung, K.
Kim, Y. Petri, J. R. Thomas, J. A. Tallarico, J. M. McKenna, M. Schirle,
D. K. Nomura, ACS Chem. Biol. 2019, 14, 2430-2440.
[56] R. F. Colman, FASEB J. 1997, 11, 217-226.
[57] E. C. Hett, H. Xu, K. F. Geoghegan, A. Gopalsamy, R. E. Kyne, C. A.
Menard, A. Narayanan, M. D. Parikh, S. Liu, L. Roberts, R. P. Robinson,
M. A. Tones, L. H. Jones, ACS Chem. Biol. 2015, 10, 1094-1098.
[58] A. Narayanan L. H. Jones, Chem. Sci. 2015, 6, 2650-2659.
[59] Q. Zhao, X. Ouyang, X. Wan, K. S. Gajiwala, J. C. Kath, L. H. Jones, A.
L. Burlingame, J. Taunton, J. Am. Chem. Soc. 2017, 139, 2, 680-685.
[60] B. Yang, H. Wu, P. D. Schnier, Y. Liu, J. Liu, N. Wang, W. F. DeGrado,
L. Wang, Proc. Natl. Acad. Sci. U. S. A. 2018, 115, 11162-11167.
[61] L. H. Jones, J. W. Kelly, RSC Med. Chem. 2020, 11, 10-17.
[24] S. Tsukiji, I. Hamachi, Curr. Opin. Chem. Biol. 2014, 21, 136-143.
[25] T. Tamura, T. Ueda, T. Goto, T. Tsukidate, Y. Shapira, Y. Nishikawa, A.
Fujisawa, I. Hamachi, Nat. Commun. 2018, 9, 1870.
[26] J. Singh, R. C. Petter, T. A. Baillie, A. Whitty, Nat. Rev. Drug Discov.
2011, 10, 307-317.
[27] J. M. Bradshaw, J. M. McFarland, V. O. Paavilainen, A. Bisconte, D. Tam,
V. T. Phan, S. Romanov, D. Finkle, J. Shu, V. Patel, T. Ton, X. Li, D. G.
Loughhead, P. A. Nunn, D. E. Karr, M. E. Gerritsen, J. O. Funk, T. D.
7
This article is protected by copyright. All rights reserved.

10.1002/anie.202104347
Angewandte Chemie International Edition
RESEARCH ARTICLE
[62] E. Leyva, M. S. Platz, G. Persy, J. Wirz, J. Am. Chem. Soc. 1986, 108,
13, 3783-3790.
[63] ArAz group serves as a photoaffinity label by generating nitrene upon UV
irradiation. It is also known to further isomerize from nitrene into highly
electrophilic dehydroazepine via a ring-expansion, which can react with
lysine residues.
[64] D. Budhadev, E. Poole, I. Nehlmeier, Y. Liu, J. Hooper, E. Kalverda, U.
S. Akshath, N. Hondow, W. B. Turnbull, S. Pöhlmann, Y. Guo, D. Zhou,
J. Am. Chem. Soc. 2020, 142, 42, 18022-18034.
[65] The Kd values calculated based on the concentration of multivalent probe
6 are compared with those of monomeric lactose because the apparent
binding affinity per probe is important for the apparent labeling efficiency
of the multivalent probes.
[66] The band tailing observed is likely due to the presence of the lipoate
moiety, which co-eluted with the labeled proteins upon cleavage of the
functionalities from the gold-nanoparticle surface.
[67] R. Banerjee, K. Das, R. Ravishankar, K. Suguna, A. Surolia, M. Vijayan,
J. Mol. Biol. 1996, 259, 281-296. (PDB: 2PEL)
[68] A. G. Gabdoulkhakov, Y. Savochkina, N. Konareva, R. Krauspenhaar, S.
Stoeva, S. V. Nikonov, W. Voelter, C. Betzel, A. M. Mikhailov,
DOI:10.2210/pdb1RZO/pdb. The Protein Data Bank, 2004. (PDB:1RZO)
[69] K. A. Majorek, P. J. Porebski, A. Dayal, M. D. Zimmerman, K. Jablonska,
A. J. Stewart, M. Chruszcz, W. Minor, Mol. Immunol. 2012, 52, 174-182.
[70] H. Otsuka, Y. Akiyama, Y. Nagasaki, K. Kataoka, J. Am. Chem. Soc.
2001, 123, 34, 8226-8230.
[71] V. L. Mendoza, R. W. Vachet, Mass Spectrom. Rev. 2009, 28, 785-815.
[72] N. V. Konareva, A. G. Gabdulkhakov, S. Eschenburg, S. Stoeva, A. N.
Popov, R. Krauspenhaar, M. E. Andrianova, Y. Savochkina, I. I. Agapov,
A. G. Tonevitskii, A. N. Kornev, V. V. Kornilov, V. N. Zaitsev, W. Voelter,
C. Betzel, S. V. Nikonov, B. K. Vainshtein, A. M. Mikhailov, Crystallogr.
Rep. 2001, 46, 792-800.
[73] RCA, which is a dimer of an α/β heterodimer, has one galactose binding
site in chain A and three in chain B, of which that of chain B involving
Glu35, Lys40, and Asp46 is considered a primary binding site.
[74] T. E. Creighton, Proteins: structures and molecular properties, WH
Freeman, New York, 1993, p2-4.
[75] A. D. White, A. K. Nowinski, W. Huang, A. J. Keefe, F. Sun, S. Jiang,
Chem. Sci. 2012, 3, 3488-3494.
[76] E. Weerapana, G. M. Simon, B. F. Cravatt, Nat. Chem. Biol. 2008, 4,
405-407.
[77] L. M. McGregor, M. L. Jenkins, C. Kerwin, J. E. Burke, K. M. Shokat,
Biochemistry 2017, 56, 3178-3183.
[78] L. Petri, P. Ábrányi-Balogh, I. Tímea, G. Pálfy, A. Perczel, D. Knez, M.
Hrast, M. Gobec, I. Sosič, K. Nyíri, B. G. Vértessy, N. Jänsch, C. Desczyk,
F. J. Meyer-Almes, I. Ogris, S. G. Grdadolnik, L. G. Iacovino, C. Binda,
S. Gobec, G. M. Keserű, ChemBioChem 2020, 22, 743-753.
[79] E. Weerapana, C. Wang, G. M. Simon, F. Richter, S. Khare, M. B. D.
Dillon, D. A. Bachovchin, K. Mowen, D. Baker, B. F. Cravatt, Nature 2010,
468, 790-795.
[80] S. M. Hacker, K. M. Backus, M. R. Lazear, S. Forli, B. E. Correia, B. F.
Cravatt, Nat. Chem. 2017, 9, 1181-1190.
[81] An electrostatic representation of PNA suggested that the protein surface
around the labeled sites were negatively charged.
[82] A mass data search of the LC/MS data in Fig. 5(b) did not provide a mass
peak corresponding to a crosslinked product that could be formed
between lactose-PEG-lipoate 7 and sulfonyl fluoride-PEG lipoate 8c,
indicating that the lactose ligand moiety of 7 did not cross-react with aryl
sulfonyl fluoride of 8c on the same nanoparticle.
8
This article is protected by copyright. All rights reserved.

10.1002/anie.202104347
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents
Chemical probes were prepared by multivalently presenting electrophilic groups and ligands on gold-nanoparticles and their reactivities
as novel affinity labeling probes were explored. Compared to a photoaffinity probe bearing aryl azide group, aryl sulfonyl fluoride probe
at a nanomolar concentration achieved more efficient and selective labeling of model proteins in cell lysate. Labeling site analysis
revealed that the nanoparticle-immobilized aryl sulfonyl fluoride groups can target multiple amino acid residues around the ligand
binding site of the target proteins.
Institute and/or researcher Twitter usernames: @TUAT_all
9
This article is protected by copyright. All rights reserved.