Electron-deficient p-benzoyl-l-phenylalanine derivatives increase covalent chemical capture yields for protein–protein interactions

Article abstract of DOI:10.1002/pro.3621

The photoactivatable amino acid p-benzoyl-l-phenylalanine (pBpa) has been used for the covalent capture of protein–protein interactions (PPIs) in vitro and in living cells. However, this technique often suffers from poor photocrosslinking yields due to th

Full text of DOI:10.1002/pro.3621

Tools for Protein Science
Protein Science
DOI10.1002/pro.3621
Electron-deficient p-benzoyl-L-phenylalanine derivatives
increase covalent chemical capture yields for protein-protein
interactions
Cassandra M. Joiner1,2, ‡, Meghan E. Breen2, ‡, Anna K. Mapp1,2,3*
1 Department of Chemistry, University of Michigan, 930 N. University, Ann Arbor, MI 48109
2 Life Sciences Institute, University of Michigan, 210 Washtenaw Ave., Ann Arbor, MI 48109
3 Program in Chemical Biology, University of Michigan, 210 Washtenaw Ave., Ann Arbor, MI
48109
Dedication We dedicate this work to Professor Ronald Raines on the occasion of his 60th
birthday. The use of physical organic principles to dissect complex biological problems by
Professor Raines serves an inspiration.
This article has been accepted for publication and undergone full peer review but has not been through
the copyediting, typesetting, pagination and proofreading process which may lead to differences
between this version and the Version of Record. Please cite this article as doi: 10.1002/pro.3621
© 2019 The Protein Society
Received: Feb 01, 2019;Revised: Apr 09, 2019;Accepted: Apr 10, 2019
This article is protected by copyright. All rights reserved.

ABSTRACT
The photoactivatable amino acid p-benzoyl-L-phenylalanine (pBpa) has been used for the
covalent capture of protein-protein interactions (PPIs) in vitro and in living cells. However, this
technique often suffers from poor photocrosslinking yields due to the low reactivity of the active
species. Here we demonstrate that the incorporation of halogenated pBpa analogs into proteins
leads to increased crosslinking yields for protein-protein interactions. The analogs can be
incorporated into live yeast and upon irradiation capture endogenous PPIs. Halogenated pBpas
will extend the scope of PPIs that can be captured and expand the toolbox for mapping PPIs in
their native environment.
Keywords: protein-protein interactions, covalent chemical capture, photocrosslinking, p-
benzoyl-L-phenylalanine
This article is protected by copyright. All rights reserved.

INTRODUCTION
Transient protein-protein interactions (PPIs) are critical for the precise flow of
information in cellular processes, and misregulation of these interactions have been implicated in
an array of human diseases.1-4 There is great interest in methods to successfully detect and
characterize these interactions. This is particularly true because the often short (ns-ps) lifetimes
and moderate strengths (micromolar to millimolar dissociation constants) of these PPIs makes
them difficult to study in their native environment using traditional biochemical techniques.5-9
With the advance of amber nonsense suppression, photo-activatable unnatural amino acids
(UAAs) such as p-benzoyl-L-phenylalanine (pBpa) have been incorporated into proteins and
used to capture transient, moderate affinity interactions for mechanistic purposes.10-28 For
example, our lab and others have utilized pBpa to interrogate the interactions involved in a
variety of different cellular processes, such as those between transcriptional activators and their
coactivator binding partners required for transcriptional initiation and those between chaperones
and their substrates required for proper protein folding.29-34
pBpa is activated for crosslinking through exposure to UV light (350-365 nm). Upon
irradiation, the benzophenone carbonyl undergoes an n to π* transition to form a diradical
species, and the triplet state can abstract hydrogen atoms from activated C-H bonds and
recombine to form a covalent adduct [Fig. 1(A)]. If an activated C-H bond is not available, the
triplet state will relax back to the ground state and can be reactivated.35-36 As a result, pBpa
undergoes minimal side reactions with solvent, making it particularly useful for in in vivo
This article is protected by copyright. All rights reserved.

applications and for interaction surfaces where considerable water is present. However, the low
molar absorptivity and quantum yield (εmax < 300 M-1cm-1; Φ = 0.05-0.4) of pBpa often limits
crosslinking yield. For example, significant starting pBpa-containing protein is often observed in
experiments (Fig. S1).
Early mechanistic studies of benzophenone suggest that the incorporation of electron-
withdrawing groups (EWGs) onto one or more of the aromatic rings decreases the energy barrier
for the n to π* transition and concomitantly increases the rate of hydrogen atom abstraction.35-39
Accordingly, we hypothesized that pBpa analogs substituted with electron-withdrawing groups
would increase crosslinking yields [Fig. 1(B)]. Here, we demonstrate that halogenated pBpa
analogs can be incorporated into live yeast cells using the bioorthogonal Escherichia coli tyrosyl
tRNA/tRNA synthetase (tRNATYRCUA/TyrRS) system and can capture in vivo transcriptional
activator-coactivator PPIs. Additionally, the crosslinking yields for EWG-modified pBpa is
substantively enhanced.
RESULTS AND DISCUSSION
The design of pBpa analogs containing EWGs was influenced by the crystal structure of
the E. coli tyrosyl tRNA synthetase that would be used for incorporation.40 Examination of the
structure suggested that the active site would likely accommodate substituents at the para
position of the distal ring, but that meta substituents would be more likely to experience steric
This article is protected by copyright. All rights reserved.

clashes (Fig. S2).41 Seven mono-substituted pBpa analogs containing halogen moieties (Cl, F,
Br, and CF3) at either the meta or para positions of the benzophenone ring were prepared via an
air-tolerant carbonylative Suzuki-Miyaura coupling with mono-substituted boronic acids and 4-
iodo-L-phenylalanine [Fig. 1(C)].42,43 The substituted pBpa analogs could be accessed with high
purity and excellent yields over 5 steps (see Supporting Information). The incorporation of each
analog was assessed in live yeast using the E. coli tRNATYRCUA-TyrRS system.40 We evaluated
the incorporation of each analog into the prototypical Gal4 transcriptional activation domain
(TAD) at position 849 which has been well characterized for UAA incorporation.30 Western
blotting indicated that all five of the para-substituted analogs were incorporated into LexA-Gal4
at position 849 at levels comparable to pBpa (Fig. 2). As suggested by the crystal structure,
substitution at the meta position was not well tolerated and sharp declines in incorporation were
observed with substituents larger than fluorine [Fig. 2(C), 2(D)]. Due to the failure of the
synthetase to efficiently incorporate the 3-CF3 Bpa analog, a 3-Br Bpa analog was not prepared
for evaluation. These findings are in agreement with previous work by Mehl and others that has
demonstrated bioorthogonal tRNA synthetases developed for specific UAAs can incorporate
analogs of the cognate UAA without any further mutagenesis to the active site.42,44-48
Furthermore, this suggests the flexibility of the pBpa specific synthetase could enable the use of
a variety of pBpa analogs, particularly those with substitution at the para position.
With the successful incorporation of the analogs, the modified pBpa analogs were
assessed for their function as crosslinkers in vivo by examining the well-defined PPI between
This article is protected by copyright. All rights reserved.

Gal4 and its masking protein, Gal80. In yeast, Gal4 interacts with the Gal80 repressor through its
TAD under normal glucose conditions.49-51 Our lab has previously demonstrated that pBpa
incorporated in either the Gal4 TAD or Gal80 can capture this endogenous interaction in live
yeast. Because of the high incorporation yields, the 3-F and 4-F Bpa analogs were selected for
comparison with pBpa in this model system. Live yeast expressing a LexA+Gal4 fusion
construct with the amber mutation at position 849 and a Myc epitope-tagged Gal80 were grown
in glucose in the presence of either pBpa or one of the fluorine analogs [Fig. 3(A)]. Live cells
were irradiated with UV light to capture all Gal4 binding partners and were then lysed and
immunoprecipitated for the LexA DNA binding domain (DBD) to isolate all LexA+Gal4 protein
interactions. To fully characterize the crosslinking abilities of the UAAs, a duplex western
blotting strategy was used such that both the binding profile of Gal4 (red) and its direct
interaction with Gal80 (green) could be observed on the same western blot (Fig. 3). As seen in
Figure 3(B), both fluorine-containing pBpa analogs captured Gal4’s endogenous binding
partners, including the Gal4-Gal80 interaction (yellow). It was not, however, possible to
quantitatively determine a change in crosslinking yield in this system. This is likely due to the
high affinity of the Gal4•Gal80 complex, which in turn results in high crosslinking yields with
even unmodified pBpa.49
To quantitate changes in crosslinking yield, we identified an interaction that was more
modest in affinity and that also could be easily examined in vitro, thus removing additional
variables in vivo such as changes in degradation rate that could influence any observations. The
This article is protected by copyright. All rights reserved.

moderate affinity interaction between the VP16 TAD and the Med25 subunit of the Mediator
complex was used for this quantitation.52-53 It has previously been shown that a minimal peptide
sequence (441-DFDLDMLG-448) within the N-terminal TAD of VP16 is sufficient for
interaction with the Med25 activator interaction domain (AcID) and transcriptional activation.54-
56
We hypothesized that incorporating pBpa at position F442 would not abrogate binding with
Med25 AcID and would enable us to observe changes in crosslinking efficiency for the pBpa
analogs.
To determine the binding affinities, biotinylated UAA-containing VP16 peptides were
incubated with Med25 coated plates. Following incubation with streptavidin-HRP and
tetramethylbenzidine (TMB) substrate, the plates were read at 450 nm and it was found that the
binding affinities of each pBpa analog containing VP16 peptides were minimally weakened (2-3
fold decrease in affinity) compared to the parent pBpa peptide (Fig. S3).52,57 These values are in
line with the binding affinity of pBpa incorporated VP16 measured by fluorescence polarization
(Fig. S4). To evaluate the crosslinking yield, Med25 was incubated with each UAA-containing
VP16 peptide at 12.5% of its Kd value to normalize for occupancy. Following 10 minutes of UV
irradiation with 365 nm light to covalently capture the interaction, the amount of crosslinked
VP16 peptide was quantified by Western blot analysis. As seen in Figure 4, all pBpa analogs
gave increased crosslinking yields, and the largest increases were observed with the 3-CF3 (49-
fold increase), 3-Cl (30-fold increase), and 4-CF3 (23-fold increase) analogs. Similar trends in
reactivity for the pBpa analogs were also observed with 1 minute irradiation time (Fig. S5). To
This article is protected by copyright. All rights reserved.

further validate these results, crosslinking yields were also analyzed via ELISA. The biotinylated
VP16 peptides were incubated with Med25 coated plates and irradiated with 365 nm light.
Following washing and incubation with streptavidin-HRP, the covalently captured VP16 peptide
was visualized by addition of TMB substrate [Fig. S6(A)]. Similar trends in reactivity were
observed [Fig. S6(B), S6(C)], although the total increase in crosslinking yield is lower. This can
be attributed to higher background to the format of the assay. Taken together, our results support
the predicted effect of appending electron withdrawing groups onto the benzophenone
framework to decrease the excitation energy gap for the n to π* transition of the carbonyl
oxygen’s electrons, thereby increasing the reactivity of the benzophenone core and ultimately
increasing the crosslinking yield of the molecule.
CONCLUSION
Although pBpa has been shown to have lower reactivity compared to the other common
photocrosslinking groups such as the aryl azides and diazarines, its ability to be reactivated and
minimal reactivity with solvent makes it a strong tool for the covalent capture of transient PPIs in
vivo.14 Therefore, there was a need to increase the crosslinking yield of pBpa to improve the
capture of PPIs and accelerate their characterization. Based on previous studies with
benzophenones, we hypothesized that pBpa analogs substituted with electron withdrawing
groups would increase crosslinking yields. We synthesized seven mono-substituted pBpa analogs
(Fig. 1) and showed that upon irradiation of live yeast expressing UAA-containing LexA+Gal4
This article is protected by copyright. All rights reserved.

we capture Gal4’s endogenous binding partners, including its masking protein Gal80.
Furthermore, we demonstrated in vitro the halogenated pBpa analogs increased crosslinking
yields for weak transient interactions. Taken together, we recommend use of the 4-CF3 analog in
cases where increased crosslinking yield is needed as it is incorporated at similar levels to pBpa
and produced a 23-fold increase in crosslinking yield.
Through this study, we have expanded our toolbox of chemical probes for capturing
challenging PPIs in their native environment. We anticipate that these halogenated pBpa analogs
will have additional downstream applications post-crosslinking and isolation. For instance,
bromine and chlorine atoms have enabled quantitative proteomic analysis without the need for
expensive isotopic labeling systems, such as stable isotope labeling using amino acids in cell
culture (SILAC) systems.58-59 Upon MS analysis, the protein targets can be quantified between
samples using the specific chlorine or bromine isotopes to accurately identify interaction partners
with minimal bias between experimental replicates. The unique isotopic signatures of these
halogens could enable the identification of PPIs such as low abundance activator-coactivator
interactions that may be drowned out by high abundance proteins and have eluded research
efforts for years.60 Utilizing the halogenated pBpa analogs developed will more readily facilitate
the successful implementation of in vivo covalent capture for studying PPIs involved in a variety
of biological processes.
This article is protected by copyright. All rights reserved.

MATERIALS AND METHODS
UAA incorporation and expression
LS41 yeast was transformed with pLexA+Gal4 849 TAG-1X Flag or pLexA+Gal4WT-
1X Flag and pSNRtRNA-pBpaRS plasmids. Individual colonies were grown in 5 mL SC media
containing 2% raffinose, but lacking histidine, tryptophan, and uracil for selection. The cultures
were incubated at 30°C with 250 rpm agitation. Following incubation, these cultures were used
to inoculate 5 mL cultures of SC media containing 2% raffinose and 2% galactose, with or
without 1 mM pBpa/1 mM pBpa EWG analog (dissolved in 50 μL 1 M NaOH), and 50 μL 1 M
HCl. The cultures were incubated at 30°C with agitation to an OD660 of 1.0. Three ODs were
isolated, washed with sterile water, and stored at -20°C. The samples were lysed in 10 μL 4X
NuPAGE LDS Sample loading buffer (Invitrogen), 10 μL 1X lysis buffer (50 mM Tris-Acetate
pH 7.9, 100 mM potassium acetate, 20% glycerol, 0.2% Tween20, 2 mM β-mercaptoethanol, and
2 mM magnesium acetate), and 10 μL 1 M DTT by boiling at 95°C for 10 min. The samples
were run on a 3-8% Tris-acetate SDS-PAGE gel and analyzed by Western blot with the anti-Flag
(M2) antibody (Sigma Aldrich). Expression levels were quantified using ImageJ and relative
levels of LexA-Gal4 protein for each experiment were expressed as follows ((experimental/WT
LexA-Gal4)*100).
In vivo crosslinking analysis
This article is protected by copyright. All rights reserved.

A colony of LexA+Gal4 849 TAG-5X Flag:Gal80-6X Myc was grown in 5 mL SC
media containing 2% glucose, but lacking histidine, tryptophan, leucine and uracil for selection.
The cultures were incubated at 30°C with 250 rpm agitation. Following incubation, these
cultures were used to inoculate 100 mL cultures of SC media containing 2% glucose, with 1 mM
pBpa or 1 mM EWG pBpa analog (dissolved in 1 mL 1 M NaOH), and 100 mL 1 M HCl. The
cultures were incubated at 30°C with agitation to an OD660 of 1.0. For each culture, the cells
were isolated by centrifugation and washed with the SC media lacking histidine, tryptophan, and
uracil. The cell pellets were resuspended in 2 mL SC media containing 2% glucose and then
transferred to small culture dishes and subjected to UV light at 365nm (Eurosolar 15W UV
lamp) with cooling for 30 minutes. The cells were isolated by centrifugation and stored at -20°C
until lysis. The control samples were washed with 1 mL SC media containing 2% glucose,
isolated by centrifugation, and stored at -20°C until lysis.
For lysis, cells were resuspended in 600 μL lysis buffer (50 mM HEPES – KOH pH 7.5,
140 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.1% Na-Deoxycholate and 2X Complete Mini,
EDTA Free Protease Inhibitor (Roche)) and lysed using glass beads by vortexing at 4°C.
Subsequently, the lysate was pelleted and the supernatant incubated with anti-LexA antibody (sc-
1725, Santa Cruz Biotechnologies) for 2 hours at 4°C for immunoprecipitation. The proteins
bound to the antibody were isolated by incubation for 1 hour with 40 μL pre-washed
Dynabeads protein G magnetic beads (ThermoFisher) at 4°C. After immunoprecipitation, the
This article is protected by copyright. All rights reserved.

beads were washed six times with 1 mL Wash Buffer (10 mM Tris-HCl pH 8.0, 250 mM LiCl,
0.5% NP-40, 0.1% Na-Deoxycholate, and 1 mM EDTA) and stored at -20°C until elution.
The samples were eluted from the beads by heating at 95°C for 10 minutes in 10 μL
NuPAGE 4X LDS Sample Loading Buffer (Invitrogen), 10 μL water, and 10 μL 1M DTT. The
samples were run on a 4-15% Mini-PROTEAN TGX pre-cast SDS-PAGE gel (Bio-Rad).
The gel was transferred to immobilin PDVF membrane (Millapore) and blocked for 1 hour at
room temperature using SuperBlock (PBS) Blocking Buffer (ThermoFisher). The membrane
was incubated with both mouse anti-Flag (M2) (Sigma Aldrich, F1804) and rabbit anti-cMyc
(Sigma Aldrich, C3956) primary antibodies overnight at 4oC. The membrane was washed three
times for 10 minutes each with PBS-T (10 mM Na2HPO4•7H2O, 1.7 mM KH2PO4, 140 mM
NaCl, 3 mM KCl, 0.05% Tween-20, pH 7.4), followed by incubation with anti-mouse 680nm
(LI-COR, 926-68072) and anti-rabbit 800nm (LI-COR, 926-32211) secondary antibodies for 1
hour at room temperature. The membrane was washed three times for 10 minutes each with PBS-
T, 5 minutes with PBS (10 mM Na2HPO4•7H2O, 1.7 mM KH2PO4, 140 mM NaCl, 3 mM KCl,
pH 7.4), and visualized using an Azure c600 western blot imager (Azure Biosystems).
In vitro crosslinking analysis
Med25 AcID 6x-His was diluted in binding buffer (25 mM HEPES, 40 mM KCl, 8 mM
MgCl2, 100 mM NaCl, 0.01% Tween-20, pH 7.4) to 22 μM, and biotin labeled VP16 peptides
This article is protected by copyright. All rights reserved.

were diluted to 25% of their Kd values in binding buffer. Equivalent volumes of Med25 AcID
and peptide were incubated at room temperature for 30 minutes in a clear, flat bottom,
polystyrene, 96-well plates (Fisherbrand). The solutions were irradiated with UV light at 365nm
(Eurosolar 15W UV lamp) with cooling for 10 minutes. 15 μL of each sample was mixed with 5
μL NuPAGE 4X LDS Sample Loading Buffer (Invitrogen) and boiled for 5 minutes. The
samples were run on a 4-20% Mini-PROTEAN TGX pre-cast SDS-PAGE gel (Bio-Rad).
The gel was transferred to PDVF membrane (Bio-Rad) and blocked for 1 hour at room
temperature with SuperBlock (PBS) Blocking Buffer (ThermoFisher). The membrane was
incubated with 1:100 diluted anti-HIS HRP conjugated antibody (Santa Cruz sc-8036 HRP) for 1
h at room temperature. The membrane was washed three times for 10 minutes each with PBS-T
(10 mM Na2HPO4•7H2O, 1.7 mM KH2PO4, 140 mM NaCl, 3 mM KCl, 0.05% Tween-20, pH
7.4), and visualized with SuperSignal West Femto ECL substrate (ThermoFisher) using an Azure
c600 western blot imager (Azure Biosystems). The membrane was washed three times with PBS-
T then stripped by incubating in Restore western blot stripping buffer (ThermoFisher) for 45
minutes at 37oC. The membrane was washed three times with PBS-T then incubated with 1:2000
diluted HRP-conjugated streptavidin (Abcam 7403) for 1 hour at room temperature. The
membrane was washed three times for 10 minutes each with PBS-T and visualized with
SuperSignal West Femto ECL substrate (ThermoFisher) using an Azure c600 western blot
imager (Azure Biosystems). Crosslinking yields were quantified using ImageJ. The relative
amount of Med25-VP16 crosslinked product for each sample was normalized based on the
This article is protected by copyright. All rights reserved.

protein levels of Med25 in the HIS-HRP blot. The fold changes were expressed as follows
(Med25-VP16 F442UAA crosslinked product / Med25-VP16 F442Bpa crosslinked product)
Supplementary Material
Additional supplemental figures, detailed experimental methods, and compound
characterization can be found in the Supporting Information.
Acknowledgments
We are grateful for the NSF CHE 1412759 for support of this work. M. Breen is a
Howard Hughes Medical Institute Fellow of the Life Sciences Research Foundation. We thank
Dr. James Clayton and Dr. Rachel Pricer for constructive feedback and Julie Garlick and Kevon
Stanford for experimental assistance.
References
1.
2.
3.
4.
5.
Mapp AK, Pricer R, Sturlis S (2015) Targeting transcription is no longer a quixotic quest.
Nat Chem Biol 11:891-894.
Cesa LC, Mapp AK, Gestwicki JE (2015) Direct and propagated effects of small
molecules on protein–protein interaction networks. Front Bioengin Biotech 3:119.
Arkin M (2005) Protein–protein interactions and cancer: small molecules going in for the
kill. Curr Opin Chem Biol 9:317-324.
Lage K (2014) Protein-protein interactions and genetic diseases: the interactome.
Biochim Biophys Acta 1842:1971-1980.
Berggård T, Linse S, James P (2007) Methods for the detection and analysis of protein–
protein interactions. Proteomics 7:2833-2842.
This article is protected by copyright. All rights reserved.

6.
Surade S, Blundell TL (2012) Structural biology and drug discovery of difficult targets:
the limits of ligandability. Chem Biol 19:42-50.
7.
Perkins JR, Diboun I, Dessailly BH, Lees JG, Orengo C (2010) Transient protein-protein
interactions: structural, functional, and network properties. Structure 18:1233-1243.
Hayes S, Malacrida B, Kiely M, Kiely PA (2016) Studying protein–protein interactions:
progress, pitfalls and solutions. Biochem Soc Trans 44:994.
8.
9.
Legrain P, Rain JC (2014) Twenty years of protein interaction studies for biological
function deciphering. J Proteomics 107: 93-97.
10.
11.
12.
Liu CC, Schultz PG (2010) Adding new chemistries to the genetic code. Ann Rev
Biochem 79:413-444.
Liu Z, Myers LC (2012) Med5(Nut1) and Med17(Srb4) are direct targets of mediator
histone H4 tail interactions. PLoS ONE 7:e38416.
Yu D, Wowor AJ, Cole JL, Kendall DA (2013) Defining the Escherichia coli SecA dimer
interface residues through in vivo site-specific photo-cross-linking. J Bacteriol 195:2817-
2825.
13.
14.
Berg M, Michalowski A, Palzer S, Rupp S, Sohn K (2014) An in vivo photo-cross-
linking approach reveals a homodimerization domain of Aha1 in S. cerevisiae. PLoS
ONE 9:e89436.
Tanaka Y, Bond MR, Kohler JJ (2008) Photocrosslinkers illuminate interactions in living
cells. Mol BioSyst 4:473.
15.
16.
Chin JW (2017) Expanding and reprogramming the genetic code. Nature 550:53-60.
Young DD, Schultz PG (2018) Playing with the molecules of life. ACS Chem Biol
13:854-870.
17.
18.
Italia JS, Zheng Y, Kelemen RE, Erickson SB, Addy PS, Chatterjee A (2017) Expanding
the genetic code of mammalian cells. Biochem Soc Trans 45:555-562.
Italia JS, Addy PS, Wrobel CJJ, Crawford LA, Lajoie MJ, Zheng Y, Chatterjee A (2017)
An orthogonalized platform for genetic code expansion in both bacteria and eukaryotes.
Nat Chem Biol 13:446-450.
19.
20.
21.
22.
23.
Coin I (2018) Application of non-canonical crosslinking amino acids to study protein–
protein interactions in live cells. Curr Opin Chem Biol 46:156-163.
Nguyen TA, Cigler M, Lang K (2018) Expanding the genetic code to study protein–
protein interactions. Angew Chem Intl Ed 57:14350-14361.
Yang Y, Song H, Chen PR (2016) Genetically encoded photocrosslinkers for identifying
and mapping protein-protein interactions in living cells. IUBMB Life 68:879-886.
Neumann-Staubitz P, Neumann H (2016) The use of unnatural amino acids to study and
engineer protein function. Curr Opin Struct Biol 38:119-128.
Miyazaki R, Myougo N, Mori H, Akiyama Y (2018) A photo-cross-linking approach to
monitor folding and assembly of newly synthesized proteins in a living cell. J Biol Chem
293:677-686.
This article is protected by copyright. All rights reserved.

24.
Yeh FL, Tung L, Chang TH (2016) Detection of protein–protein interaction within an
RNA–protein complex via unnatural-amino-acid-mediated photochemical crosslinking.
In RNA-Protein Complexes and Interactions: Methods and Protocols, Springer, pp 175-
189.
25.
26.
Mohibullah N, Hahn S (2008) Site-specific cross-linking of TBP in vivo and in vitro
reveals a direct functional interaction with the SAGA subunit Spt3. Genes Develop
22:2994-3006.
Wu CC, Lin YC, Chen HT (2011) The TFIIF-like Rpc37/53 dimer lies at the center of a
protein network to connect TFIIIC, Bdp1, and the RNA polymerase III active center. Mol
Cell Biol31:2715-2728.
27.
28.
Chen HT, Warfield L, Hahn S (2007) The positions of TFIIF and TFIIE in the RNA
polymerase II transcription preinitiation complex. Nat Struct Mol Biol14:696-703.
Hino N, Okazaki Y, Kobayashi T, Hayashi A, Sakamoto K, Yokoyama S (2005) Protein
photo-cross-linking in mammalian cells by site-specific incorporation of a photoreactive
amino acid. Nat Meth 2:201-206.
29.
Majmudar CY, Wang B, Lum JK, Håkansson K, Mapp AK (2009) A high-resolution
interaction map of three transcriptional activation domains with a key coactivator from
photo-cross-linking and multiplexed mass spectrometry. Angew Chem Intl Ed 48:7021-
7024.
30.
31.
32.
Majmudar CY, Lee LW, Lancia JK, Nwokoye A, Wang Q, Wangs AM, Wang L, Mapp
AK (2009) Impact of nonnatural amino acid mutagensis on the in vivo function and
binding modes of a transcriptional activator. J Am Chem Soc 131:14240-14242.
Krishnamurthy M, Dugan A, Nwokoye A, Fung YH, Lancia JK, Majmudar CY, Mapp
AK (2011) Caught in the act: covalent cross-linking captures activator–coactivator
interactions in vivo. ACS Chem Biol 6:1321-1326.
Dugan A, Majmudar CY, Pricer R, Niessen S, Lancia JK, Fung HY, Cravatt BF, Mapp
AK (2016) Discovery of enzymatic targets of transcriptional activators via in vivo
covalent chemical capture. J Am Chem Soc 138:12629-12635.
33.
34.
Jaya N, Garcia V, Vierling E (2009) Substrate binding site flexibility of the small heat
shock protein molecular chaperones. Proc Natl Acad Sci USA 106:15604-15609.
Raju M, Santhoshkumar P, Sharma KK (2012) αA-crystallin–derived mini-chaperone
modulates stability and function of cataract causing αAG98R-crystallin. PLoS ONE
7:e44077.
35.
36.
Dorman G, Prestwich GD (1994) Benzophenone photophores in biochemistry.
Biochemistry 33:5661-5673.
Dormán G, Nakamura H, Pulsipher A, Prestwich GD (2016) The life of pi star: exploring
the exciting and forbidden worlds of the benzophenone photophore. Chem Rev
116:15284-15398.
37.
Boscá F, Miranda MA (1998) Photosensitizing drugs containing the benzophenone
chromophore. J Photochem Photobiol B Biol 43:1-26.
This article is protected by copyright. All rights reserved.

38.
39.
40.
41.
Demeter A, Horváth K, Böőr K, Molnár L, Soós T, Lendvay G (2013) Substituent effect
on the photoreduction kinetics of benzophenone. J Phys Chem A 117:10196-10210.
Klán P, Wirz J (2009) Introduction in photochemistry of organic compounds. John Wiley
& Sons, Ltd, pp 1-23.
Wang Q, Wang L (2008) New methods enabling efficient incorporation of unnatural
amino acids in yeast. J Am Chem Soc 130:6066-6067.
Liu W, Alfonta L, Mack AV, Schultz PG (2007) Structural basis for the recognition of
para-benzoyl-L-phenylalanine by evolved aminoacyl-tRNA synthetases. Angew Chem
Intl Ed 46:6073-6075.
42.
43.
44.
45.
Joiner CM, Breen ME, Clayton J, Mapp AK (2017) A bifunctional amino acid enables
both covalent chemical capture and isolation of in vivo protein–protein interactions.
ChemBioChem 18:181-184.
Ahlburg A, Lindhardt AT, Taaning RH, Modvig AE, Skrydstrup T (2013) An air-tolerant
approach to the carbonylative suzuki–miyaura coupling: applications in isotope labeling.
J Organ Chem 78:10310-10318.
Stokes AL, Miyake-Stoner SJ, Peeler JC, Nguyen DP, Hammer RP, Mehl RA (2009)
Enhancing the utility of unnatural amino acid synthetases by manipulating broad
substrate specificity. Mol BioSyst 5:1032.
Miyake-Stoner SJ, Refakis CA, Hammill JT, Lusic H, Hazen JL, Deiters A, Mehl RA
(2010) Generating permissive site-specific unnatural aminoacyl-tRNA synthetases.
Biochemistry 49:1667-1677.
46.
47.
Young DD, Jockush S, Turro NJ, Schultz PG (2011) Synthetase polyspecificity as a tool
to modulate protein function. Bioorgan Med Chem Lett 21:7502-7504.
Tuley A, Wang YS, Fang X, Kurra Y, Rezenom YH, Liu WR (2014) The genetic
incorporation of thirteen novel non-canonical amino acids. Chem Commun 50:2673-
2675.
48.
49.
Wang YS, Fang X, Wallace AL, Wu B, Liu WR (2012) A rationally designed pyrrolysyl-
tRNA synthetase mutant with a broad substrate spectrum. J Am Chem Soc 134:2950-
2953.
Thoden JB, Ryan LA, Reece RJ, Holden HM (2008) The interaction between an acidic
transcriptional activator and its inhibitor: the molecular basis of Gal4p recognition by
Gal80p. J Biol Chem 283:30266-30272.
50.
51.
Ansari AZ, Reece RJ, Ptashne M (1998) A transcriptional activating region with two
contrasting modes of protein interaction. Proc Natl Acad Sci USA 95:13543-13548.
Hashimoto H, Kikuchi Y, Nogi Y, Fukasawa T (1983) Regulation of expression of the
galactose gene cluster in Saccharomyces: isolation and characterization of the regulatory
gene Gal4. Mol General Genet 191:31-38.
52.
Milbradt AG, Kulkarni M, Yi T, Takeuchi K, Sun ZYJ, Luna RE, Selenko P, Näär AM,
Wagner G (2011) Structure of the VP16 transactivator target in the Mediator. Nat Struct
Mol Biol 18:410-415.
This article is protected by copyright. All rights reserved.

53.
Vojnic E, Mourão A, Seizl M, Simon M, Wenzeck L, Larivière L, Baumli S, Baumgart
K, Meisterernst M, Sattler M, Cramer P (2011) Structure and VP16 binding of the
Mediator Med25 activator interaction domain. Nat Struct Mol Biol 18:404-409.
Ansari AZ, Mapp A, Nguyen DH, Dervan PB, Ptashne M (2001) Towards a minimal
motif for artificial transcriptional activators. Chem Biol 8:583-592.
Seipel K, Georgiev O, Schaffner WA (1994) minimal transcription activation domain
consisting of a specific array of aspartic acid and leucine residues. Biol Chem 375:663-
470.
54.
55.
56.
57.
Cress W, Triezenberg S (1991) Critical structural elements of the VP16 transcriptional
activation domain. Science 251:87-90.
Mittler G, Stühler T, Santolin L, Uhlmann T, Kremmer E, Lottspeich F, Berti L,
Meisterernst M (2003) A novel docking site on Mediator is critical for activation by
VP16 in mammalian cells. EMBO J 22:6494-6504.
58.
59.
Chen X, Wei S, Ji Y, Guo X, Yang F (2015) Quantitative proteomics using SILAC:
principles, applications, and developments. Proteomics 15:3175-3192.
Kaake RM, Wang X, Huang L (2010) Profiling of protein interaction networks of protein
complexes using affinity purification and quantitative mass spectrometry. Mol Cell
Proteomics 9:1650-1665.
60.
Palaniappan KK, Pitcher AA, Smart BP, Spiciarich DR, Iavarone AT, Bertozzi CR
(2011) Isotopic signature transfer and mass pattern prediction (IsoStamp): an enabling
technique for chemically-directed proteomics. ACS Chem Biol 6:829-836.
This article is protected by copyright. All rights reserved.

Figure Legends
Figure 1. (A) Covalent chemical capture mechanism of pBpa. Upon irradiation with 365 nm UV
light, pBpa reversibly forms a diradical species. If an activated C-H bond is available, hydrogen
atom abstraction followed by radical recombination will yield the crosslinked adduct. (B) A
series of pBpa analogs with halogen substituents either meta or para to the carbonyl group were
prepared. (C) Synthesis of halogenated pBpa analogs. Full experimental details and compound
characterization can be found in the Supporting Information.
Figure 2. In vivo incorporation of Halo-Bpa analogs. (A) Position 849 was mutated to the amber
stop codon (TAG) for UAA incorporation within the LexA+Gal4 chimeric transcriptional
activator. A C-terminal Flag tag was appended for visualization. (B-E) Incorporation of the
Halo-Bpa analogs into LexA+Gal4 849TAG in the presence or absence of 1 mM UAA. Loading
control is an endogenous yeast protein that consistently comes out with Flag visualization. (F)
Relative expression levels of the Halo-Bpa analogs compared to pBpa. Expression levels of
LexA+Gal4 849UAA mutants relative to LexA+Gal4 WT were quantified by using ImageJ.
Figure 3. LexA+Gal4 849UAA in vivo crosslinking. (A) Experimental scheme of in vivo
covalent chemical capture of Gal4 binding partners. (B) 3-F and 4-F Bpa capture Gal4’s
endogenous binding partners (red) and the Gal4-Gal80 interaction (yellow) when irradiated with
UV light.
Figure 4. VP16 (441-448) F442UAA in vitro crosslinking. Each Halo-Bpa analog had increased
crosslinking yield for the Med25-VP16 adduct compared to pBpa upon UV irradiation. (A)
Western blot visualization of 6XHis+Med25:Biotin+VP16 F442UAA in vitro crosslinking.
Crosslinked adducts visualized with Streptavidin-HRP and Med25 loading visualized with anti-
HIS-HRP antibody. (B) Relative fold change in Halo-Bpa crosslinking yields compared to pBpa
from 6XHis+Med25:Biotin+VP16 F442UAA western blot. The yields are an average of two
replicates. Crosslinking yields were quantified using ImageJ. The relative amount of Med25-
VP16 crosslinked adducts for each sample was normalized based on the protein levels of Med25
in the HIS-HRP blot. The fold changes were expressed as follows (Med25-VP16 F442UAA
crosslinked product / Med25-VP16 F442Bpa crosslinked product).
This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.