Optical Control of the GTP Affinity of K-Ras(G12C) by a Photoswitchable Inhibitor

Article abstract of DOI:10.1002/cbic.201900342

Photocontrol of protein activity is an emerging field in biomedicine. For optical control of a mutant small GTPase K-Ras(G12C), we developed small-molecule inhibitors with photoswitchable efficacy, where one configuration binds the target protein and exer

Full text of DOI:10.1002/cbic.201900342

Accepted Article
Title: Optical Control of GTP Affinity of K-Ras(G12C) by
Photoswitchable Inhibitor
Authors: Zhihua Ge, Zhuojin Yang, Jingshi Liang, Duoling Dong, and
Mingyan Zhu
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: ChemBioChem 10.1002/cbic.201900342
Link to VoR: http://dx.doi.org/10.1002/cbic.201900342
A Journal of
www.chembiochem.org

10.1002/cbic.201900342
ChemBioChem
COMMUNICATION
Optical Control of GTP Affinity of K-Ras(G12C) by
Photoswitchable Inhibitor
Zhihua Ge, Zhuojin Yang, Jingshi Liang, Duoling Dong, and Mingyan Zhu*[a]
Dedication
Abstract: Photocontrol of protein activity is an emerging field in
biomedicine. For optical control of a mutant small GTPase K-
Ras(G12C), we developed small-molecule inhibitors with
photoswitchable efficacy that one configuration binds the target
protein and exert different pharmacological effects upon light
irradiation. Compound design was based on the structure feature of a
previously identified allosteric pocket of K-Ras(G12C) and chemical
structure of covalent inhibitors, and resulted in the synthesis and
characterization of two representative azobenzene containing
compounds. Nucleotide exchange assay demonstrated the different
efficacy to control GTP affinity by photoswitching of one potent
compound PS-C2, which would be useful tool to probe the
conformation of mutational K-Ras. Our study demonstrated the
feasibility of designing photoswitchable modulator from allosteric
covalent inhibitor of small GTPase.
change in the G domain and signal transduction through
interactions with different effectors.[10] In normal condition the
cycle between GTP-bound active state and GDP-bound inactive
state of Ras protein is tightly regulated by guanine nucleotide
exchange factors (GEFs) and GTPase-activating proteins
(GAPs).[11] Mutation of Ras alters the nucleotide affinity which
favors GTP over GDP and results in abrrant protein activation. It
has once been considered very difficult to target the GTP binding
of Ras due to the picomolar affinity of Ras to nucleotides. The
discovery of an allosteric binding pocket and covalent inhibitors
prooved efficient to alter the GTP binding for inactivation of
Ras.[12]
Inspired by the discovery of allosteric binding pocket and
covalent inhibitors of K-Ras (G12C) (Figure 1A), we designed two
compounds as photoswitchable inhibitors of K-Ras(G12C). In the
crystal strcuture of protein-inhibitor complex the reported
inhibitors were covalently linked to Cys 12 and the benzene
moieties of the inhibitors were shown to point and insert to a
hydrophobic pocket located in the swtich II region (named as SIIP)
of K-Ras(G12C). For photoswitchable inhibitors we retained the
benzene moiety and replaced the six-membered saturated
piperidine or piperazine ring with another benzene for the purpose
of azobenzene structure incorporation in order to control the
hydrophobic interaction of terminal benzene moiety with the
allosteric pocket through photoisomerization. Due to the
hydrophobic nature of the allosteric binding pocket SIIP and the
structure similarity of reported inhibitors, para-chloride
substitution was reserved and another chloride on ortho-position
was added on the terminal benzene moiety of the photoswitciable
compounds (Figure 1B). The resulting compounds PS-C1 and
PS-C2 comprised azobenzene scaffold and chloroacetyl warhead.
We hypothesized that after conjugation to protein by the reactive
chloroacetyl group, the photoswitchable compounds would hang
on the protein surface when exist in trans form, after isomerization
to the cis form the hydrophobic terminal aromatic moiety would
point to the SIIP pocket which result in compound binding to the
pocket and prevention of protein GTP binding and locking the
protein in the inactive conformation, the substitutions on the
terminal aromatic ring of PS-C1 and PS-C2 may affect the binding
ability of the compounds with K-Ras(G12C) protein.
Optical control of the activity of biomolecules provides a
higher dimension of modulation of biological process with spatial
and temportal precision.[1] Photoswitchable probes could find
indispensable applications in function dissection for biomolecules
in dynamic biology system[2] and increasement of conditional and
regional selectivity of drugs through photopharmacology.[3]
Photoswitchable compounds have been extensively utilized in
regulating peptide and nucleic acid structure and activity, signal
transduction and protein-protein interactions.[4] Azobenzenes
(ABs) are widely used photoswitchable scaffolds due to their
robusty in switchability between the trans and cis isomers and
feasibility of bioisosterism, a large array of the applications of
azobenzenes have been reported.[5] Photoswitchable compounds
could be generated by means of “azologization” or “azoextension”
which is the well-designed incorporation of azobenzene structure
into the compound scaffold for aquirement of photoswitchablility.[6]
Photoisomerization proterties of azobenzene such as the
irradiation wavelength and half life of cis isomers can be adjusted
by substitutions on the aromatic rings.[7]
Ras proteins are among the most frequently mutation
proteins in cancer, they are small GTPases which function as
biomolecular switches[8] in cellular signal pathways related with
cell proliferation, survival and differentiation.[9] Ras proteins bind
and hydrolysis GTP to GDP, which result in conformational
[a]
Z. Ge, Z. Yang, J. Liang, D. Dong, Dr. M. Zhu
School of Pharmacy
Shanghai Jiao Tong University
800 Dongchuan Road, Minhang District, Shanghai 200240, China
E-mail: myzhu@sjtu.edu.cn
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/cbic.201900342
ChemBioChem
COMMUNICATION
transition. The trans isomer of PS-C1 exhibits maximal absorption
(max) at 364 nm (Figure 2B, Figure S1A and Figure S2A), and
trans isomer of PS-C2 exhibits maximal absorption (max) at 374
nm (Figure 2D, Figure S1B and Figure S2B). Two wavelengths
365 nm and 395 nm were tested for photoisomerization of the
compounds. Upon irradiation with both wavelengths, absorption
peak of trans isomer disappeared with time and new peak with
absorption at about 450 nm appeared (Figure S1 and Figure S2),
indicating the formation of cis isomer of the compounds.
Irradiation for a period of 30−60 s is sufficient to convert trans form
to cis, and 395 nm light seemed more effective on both
compounds since isomerization under 395 nm was faster than
365 nm (Figure S1 and Figure S2), thus 395 nm was set as
irradiation wavelength for photoisomerization of the compounds.
The spectrum of the cis isomer was estimated based on the trans
spectrum and cis-rich photostationary states (PSS)[14] (Figure 2B
and Figure 2D).
Figure 1. Photoswitchable compounds designed on the structure of allosteric
inhibitors. A) Structures of the covalent inhibitors of Ras(G12C) allosterically
impair the GTP binding and activation of Ras (The names of compounds are
retained from reference[12]). B) Structures of the designed photoswitchable
inhibitors for K-Ras(G12C) protein.
We also investigated the thermal stability of the cis isomer, in
the dark the change in absorbance with time at max of compounds
was measured after irradiation. The half-lives (t1/2) of cis isomers
were calculated (t1/2 of PS-C1 was 94.8h, and t1/2 of PS-C2 was
52.2 h, Figure 2C and 2E), which indicated moderate thermal
stability of the compounds in the dark. The cis compounds were
also observed to isomerize to trans by illumination with visible light.
The photochemical properties would be helpful in the following
conjugation to K-Ras(G12C) protein and biochemical evaluation
K-Ras nucleotide affinity experiments where in the dark the
compounds are kept as one isomer of either cis or trans structures.
Azobenzene scaffold was synthesized by Mills reaction,[13] which is
the reaction of aromatic nitroso derivatives with anilines in glacial
acetic acid (Scheme 1 and Supporting Information). Synthesis of PS-
C1 started from commercially available N-(4-aminophenyl) acetamide,
through amine oxidation by Oxone the amine transferred to nitroso
compound N-(4-nitrosophenyl)acetamide 1, the nitroso intermediate
was reacted with p-chloroaniline and formed azobenzene
intermediate 2. Synthesis of PS-C2 started from amino group
oxidation of 2,4-dichloroaniline with Oxone which formed 2,4-dichloro-
1-nitrosobenzene 4, the nitroso intermediate reacted with 4′-amino
acetanilide and formed azobenzene intermediate 5. Removal of acetyl
group by hydrolysis of acetylamides under acidic condition yielded
aminoazobenzenes 3 and 6. Finally acylation by chloroacetyl chloride
afforded PS-C1and PS-C2.
A)
B)
C)
3.0
2.5
2.0
1.5
1.0
0.5
0.0
t1/2=94.8 h
0.0
500000.0
1000000.0
1500000.0
Time（s）
D)
E)
3.0
2.5
2.0
1.5
1.0
0.5
0.0
Scheme 1. Synthesis of photoswitchable compounds PS-C1 and PS-C2.
t1/2=52.2 h
After synthesis of the compounds, we investigated the
isomerization and thermal stability of compounds by UV-visible
spectrometry (Figure 2, Figure S1 and Figure S2). In the dark both
compounds showed characteristic absorption spectra of trans
azobenzene structures, which have strong absorption for *
0.0
500000.0
1000000.0
1500000.0
Time（s）
This article is protected by copyright. All rights reserved.

10.1002/cbic.201900342
ChemBioChem
COMMUNICATION
Figure 2. Photochemical properties of compounds. A) Structures of PS-C1 and
PS-C2 and their trans and cis isomers. B) UV-visible spectra of PS-C1 in
thermal equilibrium state and PSS by 395 nm light irradiation, the spectrum of
cis isomer was estimated for 16.7% trans in the PSS. C) Thermal isomerization
of PS-C1 in DMSO at 25 °C, the absorbance was measured at =364 nm. An
covalent inhibitor was evaluated in nucleotide exchange assay as
positive control and DMSO was vehicle. We first confirmed that
light irradiation had no effect on protein activity. As seen in Figure
3A and 3B, the curves of nucleotide titration were not changed
upon UV light irradiation, indicating the nucleotide affinity was not
affected by UV, the IC50 of GDP and GTP calculated from the
curves of positive control and vehicle were similar as in the
literature.[12] Then compounds were evaluated for the different
ability to change the nucleotide affinity of K-Ras(G12C),
nucleotide IC50 values for GDP and GTP on each protein-
compound conjugate were obtained in the dark and with UV light
irradiation, the relative nucleotide affinity were calculated by the
ratio of IC50 values.
exponential function was used to calculate thermal half-life of PS-C1-cis (t1/2
=
94.8 h). D) UV-visible spectra of PS-C2 in thermal equilibrium state and PSS by
395 nm light irradiation, the spectrum of cis isomer was estimated for 18.0%
trans in the PSS. E) Thermal isomerization of PS-C2 in DMSO at 25 °C, the
absorbance was measured at =374 nm. An exponential function was used to
calculate thermal half-life of PS-C2-cis (t1/2 = 52.2 h).
Having established the photoisomerization conditions, we
turned to the evaluation of compounds on GTP binding affinity of
K-Ras(G12C). The compounds were first conjugated to K-
Ras(G12C) by incubation of protein with compounds which have
the cystein-reactive chloroacetamide group. We considered the
trans and cis isomers might have different binding affinity with
protein thus result in different extent of covalent modification,
protein was incubated with the trans and cis isomers separately.
Protein mass spectrometry was applied for detection of covalent
linkage of compounds to the protein. Both PS-C1 and PS-C2 were
detected to covalently linked to protein, MS/MS analysis showed
the site of covalent linkage was Cys12 (Figure S3-S6). The extent
for covalent modification was calculated by the ratio of labeled
peptide with total sum of labeled and unlabeled peptide. The cis
isomers showed significant better covalent linkage than the trans
forms (Table 1). PS-C2 when incubated with protein in its cis form
The photoswitchable efficacy of compounds on the nucleotide
IC50 of K-Ras(G12C) was compared with positive control
compound and free K-Ras(G12C). When treated with vehicle, the
free K-Ras(G12C) showed preference for GTP over GDP (relative
affinity 0.79 ± 0.09) (Figure 3A-3B and Table 2) which is a
characteristic feature of activation mutant of K-Ras. The positive
control (compound 12) significantly decreased the binding ability
of GTP to K-Ras (G12C) without effect on GDP affinity, leading a
preference for GDP over GTP (relative affinity 2.56 ± 0.53,
P=0.002, t-test) (Figure 3A-3B and Table 2). In the dark, when the
compounds were in the trans form, K-Ras(G12C) shows a slight
preference for GTP (relative affinity 0.59 ± 0.07 (PS-C1) and 0.88
± 0.05 (PS-C2)), which is similar as free K-Ras(G12C). After light
irradiation, the compounds changed from trans to cis, the GTP
affinity showed distinct difference and the GDP affinity was not
changed leading to nucleotide affinity preference change for GDP
over GTP (relative affinity 2.14 ± 0.10 (PS-C2), Table 2), which is
comparable with the positive control (compound 12 Figure 3 and
Table 2). Thus compound PS-C2 showed light-controlled effect
on protein, the inhibitory activity could be switched off when in the
dark which would left the K-Ras(G12C) as free protein and
switched on after UV light irradiation to alter the GTP affinity for
GDP which could inhibit the activation of Ras(G12C).
(PS-C2-cis) fully modified the protein on Cys 12. As
a
consequence, it can be assumed that compounds in cis form may
bind the protein better than trans forms. Substituents on the
terminal benzene ring render the compounds with different
binding affinity to Ras protein, as PS-C2 which has o,p-dichloride
substitutions on the terminal benzene ring showed better ability of
covalent linkage to K-Ras(G12C).
Table 1. Covalent modification ratio of K-Ras (G12C) protein
incubated with photoswitchable compounds (PS-C1, PS-C2) in trans
and cis form.
Compound[a]
PS-C1-trans
PS-C1-cis
Modification (%)
16
30
PS-C2-trans
PS-C2-cis
50
100
[a] Here the cis and trans means the isoform of compound when
incubated with protein for covalent modification.
Finally, we evaluated the ability of compounds on the
photocontrol of K-Ras(G12C) nucleotide affinity by nucleotide
exchange assay, which is an EDTA catalyzed off exchange
reaction with 2’-deoxy-3’-O-(N-methylanthraniloyl) (mant-dGDP)
while titrating unlabeled GDP or GTP. In the assay UV light was
applied on the protein-compound conjugate before titration for
isomerization of the compounds PS-C1 and PS-C2 in the
conjugate. Compound 12 which is a reported K-Ras(G12C)
Figure 3. Nucleotide affinity change upon light irradiation in EDTA-mediated
competition of mant-dGDP on K-Ras(G12C) with free GDP and GTP. Result
from a representative experiment is shown as a sigmoidal curve for each protein.
A) GDP affinity of K-Ras(G12C) with compound 12 (squares) or K-Ras(G12C)
with vehicle DMSO (circular) in the dark (black) and after UV irradiation (purple).
B) GTP affinity of K-Ras(G12C) with compound 12 or vehicle in the dark and
after UV irradiation. C) GDP affinity of K-Ras(G12C) with PS-C2 which was
conjugated to protein in cis conformation (PS-C2-cis downward triangles) in the
This article is protected by copyright. All rights reserved.

10.1002/cbic.201900342
ChemBioChem
COMMUNICATION
dark (black) and after UV irradiation (purple). D) GTP affinity of K-Ras(G12C)
with PS-C2-cis (downward triangles) in the dark (black) and after UV irradiation
(purple).
important in the binding affinity of the compounds to protein and
the efficacy to alter nucleotide affinity of the protein. Both the
photoswtichable compounds when they were incubated with
protein as trans isomers were less effective in photoswitchable
activity to alter the GTP affinity of K-Ras(G12C) than when they
were incubated as cis isomers, this was consistent with the extent
of covalent conjugation shown in Table 1. This indicated that for
covalently tethered photoswitchable compounds, the isomeric
form would affect the modification ratio of target protein which
should be considered as a factor for efficacy of compound activity
and switchablity.
The trans form and the cis form of PS-C2 in the protein
conjugate showed significant difference on the ability to affect
GTP affinity (Figure 3D) of K-Ras and did not affect GDP affinity
(Figure 3C), which is consistent with the mechanism of reported
allosteric inhibitors of K-Ras(G12C). PS-C1 showed weaker effect
(Figure S7) on nucleotide affinity of K-Ras(G12C) than PS-C2,
which indicates the substitution on terminal benzene ring is
Table 2. Nucleotide IC50 value of the GDP and GTP titrations before and after UV irradiation for DMSO, compound 12, PS-C1 and
PS-C2.
Nucleotide IC50 (M) [b]
Compound[a]
GDP
GTP
GTP/GDP[c]
0.787±0.092
GDP+UV
GTP+UV
GTP+UV/GDP+UV[c]
0.786±0.059
0.866±0.110
0.675±0.064
0.965±0.361
0.737±0.212
DMSO
1.157±0.269
1.012±0.052
0.955±0.175
0.799±0.146
0.679±0.091
0.610±0.093
0.597±0.041
0.728±0.126
0.699±0.092
1.724±0.399
0.556±0.139
0.593±0.067
0.763±0.012
0.884±0.049
2.560±0.528
0.714±0.194
0.667±0.081
0.582±0.131
0.587±0.119
0.746±0.065
0.845±0.158
0.949±0.040
1.018±0.265
1.263±0.297
1.923±0.225
1.218±0.132
1.447±0.204
1.738±0.180
2.140±0.096
2.585±0.268
PS-C1-trans
PS-C1-cis
PS-C2-trans
PS-C2-cis
Compound 12
[a] Here the cis and trans means the isoform of compound when incubated with protein for covalent modification, which is the same
as in Table 1. [b] IC50 obtained from sigmoidal curves of nucleotide exchange of each protein modified with compounds, all the
experiments were repeated three times. [c] Calculated as the average of the IC50 ratios obtained from individual single nucleotide
exchange experiments.
In summary, we have developed photoswitchable inhibitors of K-
Ras(G12C) by incorporation of azobenzene strucutre into allosteric
Acknowledgements
inhbitor for photoregulation of the nucleotide binding and activtaion of
K-Ras protein. The compounds undergo trans to cis isomerization
This work was supported by the National Science Foundation of
China (21402118) and the Science Foundation of Shanghai
under UV light irradiation and the cis structure kept for relatively long
time in the dark. K-Ras (G12C) was incubated with compounds for
(17ZR1449000). We thank Professor Kevan Shokat (UCSF) for
covalent modification, the mass spectrometric analysis confirmed the
generous provision of pJ411 K-Ras plasmids.
extent for covalent modification of compound to Cys12 mutation site
of K-Ras (G12C) protein, the cis form of the compounds showed most
Conflict of interest
extent of covalent modification. Nucleotide exchange experiment
evaluated the activity difference between cis and trans structure of the
The authors declare no conflict of interest.
compounds on the nucleotide affinity of K-Ras (G12C) protein. The
compound PS-C2 showed different inhibitory activity on GTP binding
Keywords: azobenzene • photoswitch • K-Ras(G12C) •
of K-Ras(G12C) between UV light irradiation and in the dark.
allosterism • covalent inhibitor
Photoswitchable compounds could be applied to regulate the
conformation and activity of the protein with light in a definite
spatiotemporal fashion. Our work demonstrated the feasibility of
azologization through structure hopping of covalent allosteric
inhibitors of mutant K-Ras and photoregulation of the nucleotide
affinity of small GTPase allosterically.
[1]
M. W. H. Hoorens, W. Szymanski, Trends Biochem. Sci. 2018, 43, 567-
575.
[2]
[3]
L. Kowalik, J. K. Chen, Nat. Chem. Biol. 2017, 13, 587-598.
a) K. Hüll, J. Morstein, D. Trauner, Chem. Rev. 2018, 118, 10710-10747;
b) M. M. Lerch, M. J. Hansen, G. M. van Dam, W. Szymanski, B. L.
Feringa, Angew. Chem. Int. Ed. 2016, 55, 10978-10999; c) W. A. Velema,
W. Szymanski, B. L. Feringa, J. Am. Chem. Soc. 2014, 136, 2178-2191.
This article is protected by copyright. All rights reserved.

10.1002/cbic.201900342
ChemBioChem
COMMUNICATION
[4]
a) T. Fehrentz, M. Schonberger, D. Trauner, Angew. Chem. Int. Ed. 2011,
50, 12156-12182; b) J. Broichhagen, J. A. Frank, D. Trauner, Acc. Chem.
Res. 2015, 48, 1947-1960; c) A. S. Lubbe, W. Szymanski, B. L. Feringa,
Chem. Soc. Rev. 2017, 46, 1052-1079.
[5]
[6]
[7]
a) R. J. Mart, R. K. Allemann, Chem. Commun.(Cambridge, U. K.) 2016,
52, 12262-12277; b) A. A. Beharry, G. A. Woolley, Chem. Soc. Rev. 2011,
40, 4422-4437.
a) N. Ankenbruck, T. Courtney, Y. Naro, A. Deiters, Angew. Chem. Int.
Ed. 2018, 57, 2768-2798; b) M. Schoenberger, A. Damijonaitis, Z. Zhang,
D. Nagel, D. Trauner, ACS Chem. Neurosci. 2014, 5, 514-518.
M. Dong, A. Babalhavaeji, S. Samanta, A. A. Beharry, G. A. Woolley,
Acc. Chem. Res. 2015, 48, 2662-2670.
[8]
[9]
A. A. Adjei, J. Natl. Cancer Inst. 2001, 93, 1062-1074.
a) J. F. Hancock, Nat. Rev. Mol. Cell Biol. 2003, 4, 373-384; b) F. Chang,
L. S. Steelman, J. T. Lee, J. G. Shelton, P. M. Navolanic, W. L. Blalock,
R. A. Franklin, J. A. McCubrey, Leukemia 2003, 17, 1263-1293.
[10] a) K. Rajalingam, R. Schreck, U. R. Rapp, S. Albert, Biochim. Biophys.
Acta 2007, 1773, 1177-1195; b) J. A. McCubrey, L. S. Steelman, W. H.
Chappell, S. L. Abrams, E. W. Wong, F. Chang, B. Lehmann, D. M.
Terrian, M. Milella, A. Tafuri, F. Stivala, M. Libra, J. Basecke, C.
Evangelisti, A. M. Martelli, R. A. Franklin, Biochim. Biophys. Acta 2007,
1773, 1263-1284.
[11] A. Young, J. Lyons, A. L. Miller, V. T. Phan, I. R. Alarcón, F. McCormick,
Adv. Cancer Res. 2009, 102, 1-17.
[12] J. M. Ostrem, U. Peters, M. L. Sos, J. A. Wells, K. M. Shokat, Nature
2013, 503, 548-551.
[13] E. Merino, Chem. Soc. Rev. 2011, 40, 3835-3853.
[14] J. Calbo, C. E. Weston, A. J. White, H. S. Rzepa, J. Contreras-García,
M. J. Fuchter, J. Am. Chem. Soc. 2017, 139, 1261-1274.
This article is protected by copyright. All rights reserved.

10.1002/cbic.201900342
ChemBioChem
COMMUNICATION
COMMUNICATION
Photoswitchable inhibitor of K-
Ras(G12C): Compounds were
designed by azologazation of
Author(s), Corresponding Author(s)*
Page No. – Page No.
allosteric inhibitor of K-Ras(G12C).
They were able to modify the protein
on mutation cysteine. The
compounds were able to inhibit GTP
binding of the K-Ras with light
control. Which would be a useful tool
for the study of K-Ras protein.
Optical Control of GTP Affinity of K-
Ras(G12C) by Photoswitchable
Inhibitor
This article is protected by copyright. All rights reserved.