Colocalization Strategy Unveils an Underside Binding Site in the Transmembrane Domain of Smoothened Receptor

Article abstract of DOI:10.1021/acs.jmedchem.9b00283

We unveiled an underside binding site on smoothened receptor (SMO) by a colocalization strategy using two structurally complementary photoaffinity probes derived from a known ligand Allo-1. Docking study and structural dissection identified key interactions within the site, including hydrogen bonding, πinteractions, and hydrophobic interactions between Allo-1 and its contacting residues. Taken together, our results reveal the molecular base of Allo-1 binding and provide a basis for the design of new-generation ligands to overcome drug resistance.

Full text of DOI:10.1021/acs.jmedchem.9b00283

Subscriber access provided by Nottingham Trent University
Brief Article
Co-Localization Strategy Unveils an Underside Binding Site
in the Transmembrane Domain of Smoothened Receptor
Fang Zhou, Kang Ding, Yiqing Zhou, Yang Liu, Xiaoyan Liu, Fei Zhao, Yiran Wu, Xianjun Zhang,
Qiwen Tan, Raymond C. Stevens, Fei Xu, Wenfu Tan, Youli Xiao, Suwen Zhao, and Houchao Tao
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b00283 • Publication Date (Web): 13 Aug 2019
Downloaded from pubs.acs.org on August 14, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 8
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 8
Template for Submission of Manuscripts to American Chemical
Society Journals
1
2
3
4
Word 2010, Page Wide Abstract Version
5
6
7
8
9
This template is a guide to be used to prepare manuscripts for submission. Please consult the Instructions to Authors
or a recent issue of the journal for detailed guidelines and procedures for submission. This template is intended to
benefit to the author in that the entire manuscript (text, tables, and graphics) may be submitted in one file. Inserting
graphics and tables close to the point at which they are discussed in the text of the manuscript can also be a benefit
for the reviewer.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
When you submit a manuscript using this template, you will not actually see the page formatting that appears in
the printed journal. This will occur as part of the editorial production process. Abbreviated instructions for using the
template follow. Consult the documentation for your specific application and version for more information. Additional
instructions can be found in the readme file at the web page where you downloaded this template.
Using the template
In ACS publications there are many different components of a manuscript (i.e., title, abstract, main text, figure captions,
etc.) that are represented in the template. See the Guide, Notes, Notice, or Instructions for Authors on the journal’s
homepage to determine which parts should be included for the manuscript that you are preparing
1. If typing your manuscript directly into the template, select (highlight) the text of the template that you want to replace
and begin typing your manuscript (i.e., select the Title section for typing in your title).
2. If you have already prepared your document in a Word file, you will need to attach the template to your working
document in order to apply the Word Style tags. Further instructions can be found in the readme file at the web page
where you downloaded this template.
a. Go to the Word Style list on the formatting toolbar and you will see all the Word Styles from the template that
has now been imported into the current document. A Styles toolbar has been generated that will display the
different Styles for you to choose from. If this is not present, select View, Toolbars, and then select Styles and
it should appear. You can close this at any time and then reopen it when needed.
b. Click in the sentence or paragraph and then go to the Word Style menu on the toolbar and select the relevant
Word Style. This will apply the Word Style to the entire text (sentence or paragraph). Do this for all sections of
the manuscript.
3. To insert graphics within the text or as a figure, chart, scheme, or table, create a new line and insert the graphic
where desired. If your graphic is not visible, ensure that the Word Style is “Normal” with an automatic height
adjustment. If the size of the artwork needs to be adjusted, re-size the artwork in your graphics program and re-paste
the artwork into the template (maximum width for single-column artwork, 3.3 in. (8.5 cm); maximum width for
double-column artwork, 7 in. (17.8 cm)). NOTE: If you are submitting a Table of Contents graphic, please insert
the graphic at the end of the file.
4. Ensure that page numbers are present on all pages before submitting your manuscript.
5. Delete these instructions and any sections that are not needed.
6. Save the file with the graphics in place: select Save As (File menu) and save it as a document file (not a .dot template
file).
7. Proof the manuscript to ensure that all parts of the manuscript are present and clearly legible.
ACS Paragon Plus Environment

Page 3 of 8
Journal of Medicinal Chemistry
Co-Localization Strategy Unveils an Underside Binding Site in the
Transmembrane Domain of Smoothened Receptor
Fang Zhou1,2,5,6†, Kang Ding1,3,5,6†, Yiqing Zhou4†, Yang Liu1, Xiaoyan Liu1, Fei Zhao1, Yiran Wu1,
Xianjun Zhang1,5,6,8, Qiwen Tan1, Raymond C. Stevens1,6, Fei Xu1,6, Wenfu Tan7, Youli Xiao4,5*,
Suwen Zhao1,6*, and Houchao Tao1*
1
2
3
4
5
6
7
8
1iHuman Institute, ShanghaiTech University, Ren Building, 393 Middle Huaxia Road, Pudong New District,
Shanghai 201210, China
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2Shanghai Institute of Materia Medica, Chinese Academy of Sciences, 555 Zuchongzhi Road, Building 3, Room 426,
Shanghai 201203, China
3Key Laboratory of Computational Biology, CAS-MPG Partner Institute for Computational Biology, Shanghai
Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai 200031, China
4CAS Key Laboratory of Synthetic Biology, Institute of Plant Physiology and Ecology, Shanghai Institutes for
Biological Sciences, Chinese Academy of Sciences, Shanghai 200032, China
5University of Chinese Academy of Sciences, No. 19A, Yuquan Road, Beijing 100049, China
6School of Life Science and Technology, ShanghaiTech University, Shanghai 201210, China
7Department of Pharmacology, School of Pharmacy, Fudan University, Shanghai 201203, China
8Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of
Sciences, Shanghai 200031, China.
Supporting information
ABSTRACT: We unveiled an underside binding site on smoothened receptor (SMO) by a co-localization strategy using two
structurally-complementary photoaffinity probes, derived from a known ligand Allo-1. Docking study and structural
dissection identified key interactions within the site, including hydrogen bonding, π-π interactions, and hydrophobic
interactions between Allo-1 and its contacting residues. Taken together, our results reveal the molecular base of Allo-1
binding and provide a basis for the design of new-generation ligands to overcome drug resistance.
resolution structures of human SMOs have been resolved,
including 7-transmebrane domain (TMD) and
INTRODUCTION
Hedgehog (Hh) signal transduction is of vital importance
in the regulation of embryonic development and adult
tissue maintenance1-2. This pathway is initiated by the
binding of the Hh protein to the receptor Patched1 (Ptch1),
which releases the inhibition of Ptch1 to the smoothened
receptor (SMO) and eventually induces the expression of
target genes3. However, over-activation of the Hh pathway
is associated with the initiation of several human tumors,
suggesting SMO as a target of anti-tumor therapeutic
development4-6. Pharmaceutical efforts have resulted in
several FDA-approved drugs, including vismodegib and
sonidegib for the treatment of basal cell carcinoma, the
most common skin malignancy7, and glasdegib for
treatment of acute myeloid leukemia, the most common
type of acute leukemia in adults8. However, drug-resistant
mutation of SMO often occurs after clinical use of these
drugs9. For example, the D473H mutant, identified from
tumor relapse, showed significant attenuated binding to
vismodegib10-12. Thus far, exploring novel small-molecule
ligands with alternative mechanisms to overcome drug
resistance caused by SMO mutation remains challenging13.
extracellular domain (ECD), as well as multi-domain SMO
reported recently 15. Most of these SMO structures are
bound to small molecule ligands, including the antagonists
(cyclopamine (CP), SANT-1, LY0940280) and agonists
(SAG) in traditional TMD, and sterols in ECD (Figure S1)16-
21
.
In the past decades, thousands of allosteric modulators
targeting GPCRs have been developed. However, a limited
number of binding poses (less than 20) have been
elucidated due to low binding affinity and incomplete
structure-activity
relationship
(SAR)
information.
Structural studies have contributed substantially to
understanding of the SMO function22, but the binding sites
of several significant allosteric ligands remain unknown
(Figure S2). Allo-1 is one of the most well-known allosteric
ligands that maintain inhibitory activity against drug-
resistant mutant SMO, but few analyses of SAR have been
carried out 10, 15. According to previous studies, Allo-1 binds
to neither TMD nor ECD, where multiple known ligands
19,
bind (Table S1)23,17,
24-25. Confusingly, the effective
competition of Allo-1 with BODIPY-CP (Figure S3), a
classic fluorescent probe derived from CP, indicated that
Allo-1’s binding site probably overlaps the occupancy of the
The tremendous progress in structural biology has resulted
in a plethora of SMO research14. To date, several high-
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
BODIPY moiety but not of CP. However, this finding was
Page 4 of 8
benzene moiety (Figure 1B and Scheme S1)29. Both probes,
together with bifunctional probe 3 bearing one more
terminal alkyne tag, exhibited comparable antagonism in
the luciferase reporter assay and binding affinity in the
fluorescence-based competition test compared with the
parent compound Allo-1; this finding indicates that the
introduction of a small linear azide and a propagyl group
was largely tolerated (Figure 1B). The specificity of the
probe was further demonstrated in a two-step treatment of
probe 3 (Figure 1C) and competition test in the presence of
Allo-1 (Figure S5). In comparison, another photoactivable
group diazirine was not allowed because its introduction
to Allo-1 (probe 4, Figure S6) led to a significant loss of
1
2
3
4
5
6
7
8
not supported by our docking and simulation studies
(Figure S4). The undergoing program of co-crystallization
has not been successful yet to unmask cryptic binding. As
part of our continuous efforts in structural studies and
therapeutic development on SMO, we report the
determination of an underside binding site for Allo-1 by
using an integrated approach combining photoaffinity
labeling, computational modeling and structural
dissection study.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
RESULTS AND DISCUSSION
B
A
activity, due to size incompatibility29, 30-31
.
M
C
A
C
M
r
M
A
C
r
O
C
r

M
AM
A
CBB
Figure 1. Design and characterization of the photoaffinity
probes (1-2). (A) Definition of a new ligand binding site by
using co-localization strategy by connecting two modified
areas caused by two probes. (B) Structures of Allo-1 and
derived probes with azido groups at the two ends. The potency
of the analogs was evaluated by the Gli-luciferase reporter
assay, and binding affinity was evaluated by fluorescence-
based competition test with Bodipy-LY. IC50 and Ki values
represent the mean ± SEM of three independent experiments,
each at least in duplicate. (C) SDS-PAGE analysis and
fluorescence detection indicated the specific conjugation of
probes with the SMO. Dually purposed probe 3 incubated with
SMO protein was irradiated with UV light for 30 min, followed
by “click” conjugation to tetramethylrhodamine-azide
(TAMRA-N3).
B
Design and characterization of Allo-1-derived
photoaffinity probes. To elucidate Allo-1 binding, we
developed photoaffinity ligands26-27. Upon irradiation with
UV light, active intermediates, such as nitrene or carbene,
trap the target proteins through covalent alkylating of
adjacent amino acid residues, which could be interpreted
by proteolysis and mass spectroscopy28. Ideally, the
binding region of the ligand could be navigated around the
spherical area centered by the alkylated residues, which
often involve a time-consuming trial and error procedure.
Inevitably, nonspecific labeling often occurs, thereby
complicating the identification. Herein, we introduced a
co-localization strategy (Figure 1A) that rapidly and
accurately defines the binding area by simply connecting
the two modified areas caused by two probes. Only the
connection that matches the size of the ligand can be a
potential site for further docking study. To this end, two
probes 1 and 2 were designed by installing azido groups at
both ends of Allo-1, i.e., at the para-substituent of either
C
Figure 2. Mass spectrum-based determination of Allo-1
binding site. (A) Multiple residues, labeled with probe 1 (as
green) and probe 2 (as violet), were located in distinct regions
of SMO (PDB code: 4N4W). The Allo-1 binding site (marked
as grey ellipse) located underneath the TMD of SMO was
defined by the restriction between E518 and GTG, which were
assigned from the MS/MS analysis of probes 1 and probe 2
modified tryptic peptide (C) and (B), respectively.
ACS Paragon Plus Environment

Page 5 of 8
Journal of Medicinal Chemistry
Allo-1 binds underneath the traditional site of TMD.
After radiation with UV light, SMO labeled with probe 1 or
2 was subjected to tryptic digestion. The resulting peptides
were analyzed by LC-MS on Thermo Orbitrap Fusion
Lumos mass spectrometer. The m/z increases of 293.12 and
327.08 were consistent with the molecular weight added by
probes 1 and 2, respectively, after the loss of N2. We chose
the top 10 precursor ions in the MS1 spectra for MS2
acquisition using high-energy collision induced
dissociation (HCD) activation. Both probes labeled
multiple amino acid residues located in distinct regions of
SMO (Table S2, Figure 2A), although many of the labeled
residues located in the cytoplasmic domains were likely
from non-specific interactions and therefore excluded.
Taking advantage of the dual-probe strategy, we identified
the true binding site by measuring the distance between
the labeled residues by
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
A
B
C
Figure 3. Computational analysis of the underside binding site for Allo-1 (PDB code: 4N4W). (A) Docking of Allo-1 in the defined
site of SMO. Close-up view of the Allo-1 (green carbons) binding showed two hydrogen bonds (green dashed lines) and strong π-
cage formed with adjacent residues (blue dashed lines). (B) Molecular dynamic simulation showed accurate and stable binding of
Allo-1 (green) in SMO (blue). (C) The indicated residues showed continuous corresponding interactions with Allo-1 during 200 ns-
simulation.
both probes; as such, the distance should be comparable to
the molecular size of Allo-1 (~ 13 Å, Figure S7). After
calculating the distance between all the labeled residues
from both probes, only two pairs (Glu518 with Gly527 and
Thr528 respectively) fell within the range of Allo-1 size
(Table S2). Glu518 was assigned from the MS/MS analysis
of probe 1 modified peptide 511NRPSLLVEK519 Da according
to numerous b- and y-type ions that formed after HCD
fragmentation (Figure 2B). Gly527, Thr528, and Gly529 in a
slightly longer distance from Glu518, were continuous
residues from one same peptide alkylated by probe 2,
indicating a relatively promiscuous pattern and probably
freely swinging binding (Figure 2C). Collectively, an
underneath binding site was assigned for Allo-1 by the
restriction of the upper residue (Glu518) and lower region
(Gly527 to Gly529) (Figure 2A).
and the adjacent aromatic residues His470, Phe391, and
Trp281, which formed a “π-cage”. Apart from other ligands,
even SANT-1 with a deep binding , Allo-1 did not interact
with extracellular loops, and seemed far away from Asp473
(~3.7 Å), whose mutation to histidine will not severely
attenuate its binding23 (Figure S8).
To investigate the binding stability between Allo-1 and
SMO, we performed a 200 ns molecular dynamics (MD)
simulation (Figure 3B). Allo-1 only showed a 0.3 Å RMSD
during the simulation, implying that the binding pose was
accurate. Continuous interactions, including hydrogen
bond and π-π interaction involving residues His470,
Phe391, and Trp281, locked Allo-1 in a stable conformation
(Figure 3C). In addition to the π-cage effect, hydrophobic
interactions with residues located at the bottom of the
pocket, such as Phe274, Leu325, Val329, and Val463,
contributed to the tight binding of Allo-1.
Key interactions responsible for the binding of Allo-1
within the underside site of SMO. A docking study of
Allo-1 was carried out on the defined binding site in SMO
(PDB: 4N4W) (Figure 3A). High-density interactions
within the compact site were identified, confirming Allo-1
as a potent ligand. In this binding, two pairs of hydrogen
bonds were formed between each of carbonyl oxygen with
His470 and Trp281. A set of strongly stabilizing π-π
interactions were identified between the upper benzene
Structure-activity relationship (SAR) analysis of Allo-
1 analogs within the binding site. To validate the major
interactions in the Allo-1 binding, we employed a
dissection approach for SAR analysis (Table 1 and Scheme
S2)32. Allo-1 was dissected into three parts, each of which
was chemically mutated so as to weaken the assigned
interaction, while maintaining the molecular size.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
After replacing the benzene ring with the cyclohexyl group,
Page 6 of 8
benzene replacement (10), the transplacement of benzene
(11), and even the sulfur replacement of urea carbonyl
oxygen (12), significantly decreased the binding free
energy, which originated from the twist of the hydantoin
ring or the orientation of the right benzene that disordered
the molecular packing.
1
2
3
4
5
6
7
8
compounds 5 and 9 showed dramatic loss of activity in the
luciferase reporter assay compared with Allo-1. The right
benzene seemed even more sensitive to dearomatization,
consistent with the loss of the π-π interactions within the
π-cage. In parallel, the perfluorated benzene in compounds
6 and 10 led to reduced potency. Perfluorobenzene usually
enhances its interaction with benzene via face-to-face
packed π-π interaction33-35. However, when packed in pure
hydrophobic site, such as in the area composed of Val463,
Val329, Ile408, Leu325, and Met326, the perfluoroaryl of
compound 6 is unfavored due to electrostatic repulsion36.
The decreased potency of compound 10 indicated the
mismatched packing of perfluorobenzene with its
interacting aromatic residues. We also attempted the
transplacement of the benzenes by tuning the linker to the
hydantoin core on the left (7-8) and right (11). The right
benzene seemed highly conserved, and a single carbon
extension abolished the activity; meanwhile the left
benzene was relatively tolerable.
Allo-1 binding site might be targeted by Allo-2. Similar
to Allo-1, Allo-2 (Figure S2) also showed allosteric
antagonism activity of SMO23. Allo-2 noncompetitively
antagonizes SMO induced by SAG while competing with
Bodipy-CP (Table S1). Our docking study generated good
docking pose with a high score at -11.4 (Figure 4A) in the
Allo-1 binding site, similar to Allo-1 score (-11.4, Table S3).
In this binding mode, Allo-2 superimposed well with Allo-
1. The benzene group and indazole group of Allo-2
occupied the similar places of the two benzene groups of
Allo-1, and the pyrimidine group of Allo-2 located in the
middle of the pocket similar to hydantoin in Allo-1. A
similar π-cage, formed by His470, Phe391, and Trp281,
captured the upper benzene group of Allo-2. In addition,
two hydrogen bonds were found, namely, between His470
and pyrimidinyl amine in the middle, and between the
indazole group of Allo-2 and Thr528 at the bottom of the
molecule; these bonds significantly helped stabilize the
overall conformation. Though highly overlapped to each
other, Allo-2 seems to be longer than Allo-1, with the
trifluoromethoxy group on the top, which is probably
responsible for competition with CP, making Allo-2
different from Allo-1. The distances between CP and Allo-1
and Allo-2 are 3.8 Å and 1.6 Å, respectively (Figure 4B and
4C). Considering the van der Waals radius of O (1.40 Å) and
F (1.35 Å), Allo-2 is sufficiently close to clash with CP and
compete for binding, while leaving Allo-1 not interrupted.
Additionally, LY0940680 would partially overlap with
Allo-1 (Figure S10). Allo-2 showed similar competition
binding affinity (Ki= 30±21 nM, Figure S 11) with Allo-1 (Ki=
33±14 nM, Table 1). Meanwhile, Probe 3 specific fluorescent
labeling was gradually competed by Allo-2, providing the
evidence that Allo-2 targeted in Allo-1 binding site (Figure
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1. SAR analysis of Allo-1 analogs.
C
M
C
M
M
Compds
IC50 (nM)a
59±9
Compds
IC50 (nM)a
8006±1994
1533±262
Allo-1
9
10
11
5
6
7
8
383±109
7382±2618
547±453
92±38
8097±1921
3791±673
12
13
4552±2245
A
B
aThe potency of the analogs was evaluated by Gli-luciferase
reporter assay. IC50 values represent the mean ± SEM of three
independent experiments, each conducted in duplicate.
S12).
We also investigated the hydrogen bonds between the
carbonyls and the protein. When sulfur replaced the urea
carbonyl oxygen in compound 12, the proposed hydrogen
bonding with residue His470 was not enforced37-38. By
contrast, a significant decrease was reflected in the
moderate EC50. The gap implied that the tight binding of
Allo-1 tolerated no enlarged atom at the carbonyl oxygen39.
The removal of another carbonyl oxygen in compound 13
also reduced the activity, albeit to a lesser extent,
correlating with the disappearance of the corresponding
hydrogen bond.
C
The attenuation of the potency of the Allo-1 analogs was
also reflected in the MM-GBSA calculation and docking
study (Figure S9, Table S3)40. As indicated, the maximum
loss of binding free energy, and the lowest docking score
clearly reflected the complete disappearance of the π-cage
effect in cyclohexyl analog 9 (Figure S9E). The perfluorated
Figure 4. Allo-2 binds to the same site as Allo-1 (PDB code:
4N4W). (A) Close up view of Allo-2 binding showed two
hydrogen bonds (green dashes) of pyrimidine and indazole
with His470 and Thr528, as well as the π-cage (blue dashes)
formed with the upper benzene. The binding of Allo-2 is closer
ACS Paragon Plus Environment

Page 7 of 8
Journal of Medicinal Chemistry
to CP, as indicated by the distance measured for Allo-1 (B) and
Allo-2 (C).
(0.48 mmol) in 0.5 mL 6 N HCl was added dropwise NaNO2
(0.72 mmol, 50 mg) in 3 mL H2O at 0 °C. Then NaN3 (0.72
mmol, 47 mg) in 2 mL H2O was slowly added to the
mixture. After that, the reaction was stirred for another 40
minutes at room temperature before being quenched with
NaHCO3 solution and extracted with EtOAc three times.
The organic layer was washed with brine and dried over
Na2SO4. Finally, the reaction was concentrated and
purified with flash column chromatography (EtOAc/PE
1:3) on silica gel to obtain compounds 1 and 2.
1
2
3
4
5
6
7
8
CONCLUSION
Here, we report the identification of an underside pocket
of SMO hosting Allo-1 by a cross-disciplinary approach
combining mass spectrometry, computational biology and
medicinal chemistry. Photoaffinity labeling has been
widely used in binding site identification, but is often
accompanied by significant non-specific incorporation, as
also indicated in each single labeling in this study (Figure
2A). Specific probes are not easily accessed due to the size
limitation of the labeling group and sometimes of the
ligand itself. Our strategy intrinsically circumvents the
misleading, non-specific photolabeling and facilitates
docking using a defined orientation of the ligand.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(S)-3-(4-azidophenyl)-1-benzyl-5-
methylimidazolidine-2,4-dione (1) white solid. 1H NMR
(500 MHz, Chloroform-d), δ (ppm) 7.50-7.43 (m, 2H), 7.42-
7.29 (m, 5H), 7.14-7.08 (m, 2H), 5.07 (d, J = 15.2 Hz, 1H), 4.23
(d, J = 15.2 Hz, 1H), 3.95 (q, J = 7.0 Hz, 1H), 1.49 (d, J = 7.0
Hz, 3H); 13C NMR (126 MHz, Chloroform-d), δ (ppm) 172.4,
155.1, 139.8, 135.6, 129.2, 128.6, 128.40, 128.36, 127.4, 119.6,
54.7, 44.9, 15.5. HRMS calcd for C17H15N5O2 [M+H]+:
322.1299; found: 322.1345. HPLC: tR 11.2 min, purity >95%.
The structural insights of this binding site will open
possibilities for development of new design and virtual
screening of next-generation anti-cancer drugs against
resistant mutants of SMO. Compared with the challenging
(S)-1-(4-azidobenzyl)-3-(4-chlorophenyl)-5-
methylimidazolidine-2,4-dione (2) white solid. 1H NMR
(500 MHz, Chloroform-d), δ (ppm) 7.45-7.40 (m, 4H), 7.31
(d, J = 8.7 Hz, 2H), 7.03 (d, J = 8.7 Hz, 2H), 4.99 (d, J = 15.3
Hz, 1H), 4.24 (d, J = 15.3 Hz, 1H), 3.96 (q, J = 7.0 Hz, 1H), 1.48
(d, J = 7.0 Hz, 3H); 13C NMR (126 MHz, Chloroform-d), δ
(ppm) 172.1, 154.9, 140.3, 133.9, 132.3, 130.4, 129.9, 129.3, 127.1,
119.8, 54.8, 44.4, 15.6; HRMS calcd for C17H14ClN5O2 [M+H]+:
356.0909; found: 356.0906. HPLC: tR 11.6 min, purity>95%.
co-crystallization
attempts,
we
contributed
a
straightforward and efficient approach to the community
of structural biologists. Our methodology should also
facilitate the characterization of unknown binding sites of
enormous allosteric GPCR modulators that is essential in
the understanding of GPCR function and innovative drug
discovery.
Synthesis of Compound 3. To a solution of S9 (0.068
mmol, 23 mg) in 2 mL acetone was added K2CO3 (0.5 M)
and 3-bromopropyne (0.14 mmol, 17 mg). The reaction was
stirred at room temperature overnight. After that, the
reaction was quenched with saturated NH4Cl solution and
extracted with EtOAc three times. The combined organic
layer was washed with saturated NH4Cl solution and brine,
dried over Na2SO4. Finally, the reaction was concentrated
and purified with flash column chromatography
(EtOAc/PE 1:3) on silica gel to obtain compound 3.
EXPERIMENTAL SECTION
General. All commercial reagents and solvents were
purchased from Accela, J&K, Adamas, Bide and Sigma-
Aldrich. Reagents and solvents were used without further
purification. High-resolution mass spectra were recorded
on an Agilent 6230 mass spectrometer using ESI
(electrospray ionization). Chromatography was performed
on silica gel 200-300 mesh. NMR spectra were recorded on
a Bruker AVANCE III 500, 600 or 800 spectrometer (FT,
500/600/800 MHz for 1H NMR; 126/150/201 MHz for 13C
NMR. The purity of all biologically evaluated compounds
was determined by HPLC (Shimadzu, LC20AD) and then
confirmed with Agilent mass spectrometer (Agilent 6230,
TOF, LC/MS). Systems were run with 10%-90%
acetonitrile/water gradient with 0.05% TFA. (column:
Waters X bridge shield RP 18, 5 μm, column 4.6 × 250 mm;
temperature = 25 °C; solvent A = H2O, 0.05% TFA; solvent
B = MeCN, 0.05% TFA; flow rate = 1.0 mL/min; method:
gradient: 10% B [over 3 min], then 10% B to 90% B [over 2
min], then 90% B [over 10 min]), then 10% B [over 5 min].
For certain substances, optimization of the HPLC gradient
was performed. All final compounds showed a purity of
>95%. All phenyl isocyanate compounds were obtained
from commercial sources or synthesized according to the
literature41.
(S)-1-(4-azidobenzyl)-5-methyl-3-(4-(prop-2-yn-1-
yloxy) phenyl)imidazolidine-2,4-dione (3) yellow solid.
1H NMR (500 MHz, Chloroform-d), δ (ppm) 7.36-7.30 (m,
4H), 7.06-7.03 (m, 4H), 4.99 (d, J = 15.3 Hz, 1H), 4.71 (d, J =
2.4 Hz, 2H), 4.24 (d, J = 15.3 Hz, 1H), 3.93 (q, J = 7.0 Hz, 1H),
2.53 (t, J = 2.4 Hz, 1H), 1.47 (d, J = 7.0 Hz, 3H). 13C NMR (126
MHz, Chloroform-d), δ (ppm) 172.5, 157.2, 155.5, 140.3, 132.5,
129.9, 127.5, 125.4, 119.7, 115.5, 78.3, 76.0, 56.2, 54.9, 44.5, 15.6.
HRMS calcd for C20H17N5O3 [M+H]+: 376.1410; found:
376.1409. HPLC: tR 10.7 min, purity >95%.
ASSOCIATED CONTENT
Supporting Information. This material is available free of
charge via the Internet at http://pubs.acs.org.
SMILES strings for the small molecules of Figure 1B
and Table 1 (CSV)
Please refer to SI for detailed procedures for the synthesis
of all intermediates and characterization of all compounds.
AUTHOR INFORMATION
Corresponding Author
General procedure for synthesis of azide compounds
(1, 2). To a solution of amino analogs of Allo-1 (S6 and S7)
*For Y.X., Email, ylxiao@sibs.ac.cn
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 8
9.
Pan, S.; Wu, X.; Jiang, J.; Gao, W.; Wan, Y.; Cheng, D.;
*For S.Z., Email, zhaosw@shanghaitech.edu.cn
*For H.T., Email, taohch@shanghaitech.edu.cn
ORCID
Han, D.; Liu, J.; Englund, N. P.; Wang, Y.; Peukert, S.; Miller-
Moslin, K.; Yuan, J.; Guo, R.; Matsumoto, M.; Vattay, A.; Jiang, Y.;
Tsao, J.; Sun, F.; Pferdekamper, A. C.; Dodd, S.; Tuntland, T.;
Maniara, W.; Kelleher, J. F., III; Yao, Y.-m.; Warmuth, M.;
Williams, J.; Dorsch, M., Discovery of NVP-LDE225, a potent and
selective Smoothened antagonist. ACS Med. Chem. Lett. 2010, 1,
130-134.
1
2
3
4
5
6
Houchao Tao: 0000-0001-7598-0275
Author Contributions
†F.Z, K.D. and Y.Z. contributed equally.
7
8
9
10.
Sharpe, H. J.; Wang, W.; Hannoush, R. N.; de Sauvage,
Notes
F. J., Regulation of the oncoprotein Smoothened by small
molecules. Nat. Chem. Biol. 2015, 11, 246-255.
11.
The authors declare no competing financial interest.
Yauch, R. L.; Dijkgraaf, G. J. P.; Alicke, B.; Januario, T.;
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACKNOWLEDGMENT
Ahn, C. P.; Holcomb, T.; Pujara, K.; Stinson, J.; Callahan, C. A.;
Tang, T.; Bazan, J. F.; Kan, Z.; Seshagiri, S.; Hann, C. L.; Gould, S.
E.; Low, J. A.; Rudin, C. M.; de Sauvage, F. J., Smoothened
mutation confers resistance to a Hedgehog pathway inhibitor in
medulloblastoma. Science 2009, 326, 572-574.
12.
Nat. Rev. Drug Discovery 2012, 11, 437-438.
13. Atwood, S. X.; Sarin, K. Y.; Whitson, R. J.; Li, J. R.; Kim,
The authors (FX, SZ and HT) are thankful to Shanghai
Municipal Government, ShanghaiTech University for startup
financial support, and National Key R&D Program of China
(2018YFA0507000) for partial support. This work was also
supported by NSFC (21502206, 21572243, YX). We thank J. Liu,
Q. Shi, Y. Xu and L. Yang for assistance in the membrane
preparation, assay performance, plasmid preparation and
NMR data collection respectively at the Core Facility of
iHuman Institute. We thank Y. Shan for MS assistance in the
core facility center of the Institute of Plant Physiology and
Ecology. We thank Y. Zhao for the generous donation of NIH-
3T3 cell line for the assay. We thank Z.-J. Liu, G. Song, G.
Zhong, D. Wu, D. Liu, J. Cheng and C. Zhang for their
insightful discussion and assistance in manuscript
preparation.
Dlugosz, A.; Agrawal, S.; Kirkpatrick, P., Vismodegib.
G.; Rezaee, M.; Ally, M. S.; Kim, J.; Yao, C.; Chang, A. L. S.; Oro, A.
E.; Tang, J. Y., Smoothened variants explain the majority of drug
resistance in basal cell carcinoma. Cancer Cell 2015, 27, 342-353.
14.
Dijkgraaf, G. J. P.; Alicke, B.; Weinmann, L.; Januario, T.;
West, K.; Modrusan, Z.; Burdick, D.; Goldsmith, R.; Robarge, K.;
Sutherlin, D.; Scales, S. J.; Gould, S. E.; Yauch, R. L.; de Sauvage, F.
J., Small molecule inhibition of GDC-0449 refractory smoothened
mutants and downstream mechanisms of drug resistance. Cancer
Res. 2011, 71, 435-444.
15.
Byrne, E. F. X.; Luchetti, G.; Rohatgi, R.; Siebold, C.,
ABBREVIATIONS
Multiple ligand binding sites regulate the Hedgehog signal
transducer Smoothened in vertebrates. Curr. Opin. Cell Biol. 2018,
51, 81-88.
GPCR, G protein-coupled receptor; BODIPY, Difluoro{2-[1-
(3,5-dimethyl-2H-pyrrol-2-ylidene-N)ethyl]-3,5-dimethyl-1H-
pyrrolato-N}boron.
16.
Wang, C.; Wu, H.; Katritch, V.; Han, G. W.; Huang, X.-
P.; Liu, W.; Siu, F. Y.; Roth, B. L.; Cherezov, V.; Stevens, R. C.,
Structure of the human smoothened receptor bound to an
antitumor agent. Nature 2013, 497, 338-343.
REFERENCES
1.
Hedgehog signaling and its roles in development and disease. Nat.
Rev. Mol. Cell. Biol. 2013, 14, 416-429.
2.
smoothened activity by inducing a conformational switch. Nature
2007, 450, 252-258.
3.
animal development: paradigms and principles. Genes Dev 2001,
15, 3059-3087.
4.
pathway in cancer. Nat. Rev. Drug Discovery 2006, 5, 1026-1033.
5. di Magliano, M. P.; Hebrok, M., Hedgehog signaling in
cancer formation and maintenance. Nat. Rev. Cancer 2003, 3, 903-
911.
6.
small molecule inhibitors of Hh signaling as anti-cancer
chemotherapeutics. Curr. Med. Chem. 2015, 22, 4033-4057.
7.
Dina, M. S.; Goldsmith, R.; Gould, S. E.; Guichert, O.; Gunzner, J.
L.; Halladay, J.; Jia, W.; Khojasteh, C.; Koehler, M. F. T.; Kotkow,
K.; La, H.; La Londe, R. L.; Lau, K.; Lee, L.; Marshall, D.; Marsters,
J. C.; Murray, L. J.; Qian, C.; Rubin, L. L.; Salphati, L.; Stanley, M.
S.; Stibbard, J. H. A.; Sutherlin, D. P.; Ubhayaker, S.; Wang, S.;
Wong, S.; Xie, M., GDC-0449-A potent inhibitor of the hedgehog
pathway. Bioorg. Med. Chem. Lett. 2009, 19, 5576-5581.
Briscoe, J.; Therond, P. P., The mechanisms of
17.
Wang, C.; Wu, H.; Evron, T.; Vardy, E.; Han, G. W.;
Huang, X.-P.; Hufeisen, S. J.; Mangano, T. J.; Urban, D. J.; Katritch,
V.; Cherezov, V.; Caron, M. G.; Roth, B. L.; Stevens, R. C.,
Structural basis for Smoothened receptor modulation and
chemoresistance to anticancer drugs. Nat. Commun. 2014, 5, 4355-
4365.
Zhao, Y.; Tong, C.; Jiang, J., Hedgehog regulates
Ingham, P. W.; McMahon, A. P., Hedgehog signaling in
18.
Byrne, E. F. X.; Sircar, R.; Miller, P. S.; Hedger, G.;
Luchetti, G.; Nachtergaele, S.; Tully, M. D.; Mydock-McGrane, L.;
Covey, D. F.; Rambo, R. P.; Sansom, M. S. P.; Newstead, S.;
Rohatgi, R.; Siebold, C., Structural basis of Smoothened regulation
by its extracellular domains. Nature 2016, 535, 517-522.
Rubin, L. L.; de Sauvage, F. J., Targeting the hedgehog
19.
Huang, P.; Nedelcu, D.; Watanabe, M.; Jao, C.; Kim, Y.;
Liu, J.; Salic, A., Cellular cholesterol directly activates Smoothened
in Hedgehog signaling. Cell 2016, 166, 1176-1187.
20.
S.; Ishchenko, A.; Ye, L.; Lin, X.; Ding, K.; Dharmarajan, V.; Griffin,
P. R.; Gati, C.; Nelson, G.; Hunter, M. S.; Hanson, M. A.; Cherezov,
V.; Stevens, R. C.; Tan, W.; Tao, H.; Xu, F., Crystal structure of a
multi-domain human smoothened receptor in complex with a
super stabilizing ligand. Nat. Commun. 2017, 8, 15383-15392.
Maschinot, C. A.; Pace, J. R.; Hadden, M. K., Synthetic
Zhang, X.; Zhao, F.; Wu, Y.; Yang, J.; Han, G. W.; Zhao,
Robarge, K. D.; Brunton, S. A.; Castanedo, G. M.; Cui, Y.;
21.
Huang, P.; Zheng, S.; Wierbowski, B. M.; Kim, Y.;
Nedelcu, D.; Aravena, L.; Liu, J.; Kruse, A. C.; Salic, A., Structural
basis of Smoothened activation in Hedgehog signaling. Cell 2018,
174, 312–324.
22.
Dong, X.; Wang, C.; Chen, Z.; Zhao, W., Overcoming the
8.
Munchhof, M. J.; Li, Q.; Shavnya, A.; Borzillo, G. V.;
resistance mechanisms of Smoothened inhibitors. Drug Disc.
Today 2018, 23, 704-710.
23.
Wang, Y.; Ramamurthy, A.; Schultz, P. G.; Dorsch, M.; Kelleher,
J.; Wu, X., Small molecule antagonists in distinct binding modes
Boyden, T. L.; Jones, C. S.; LaGreca, S. D.; Martinez-Alsina, L.;
Patel, N.; Pelletier, K.; Reiter, L. A.; Robbins, M. D.; Tkalcevic, G.
T., Discovery of PF-04449913, a potent and orally bioavailable
inhibitor of Smoothened. ACS Med. Chem. Lett. 2012, 3, 106-111.
Tao, H.; Jin, Q.; Koo, D.-I.; Liao, X.; Englund, N. P.;
ACS Paragon Plus Environment

Page 9 of 8
Journal of Medicinal Chemistry
inhibit drug-resistant mutant of Smoothened. Chem. Biol. 2011, 18,
interactions: a new strategy for supermolecule construction.
Angew. Chem., Int. Ed. 1997, 36, 248-251.
432-437.
24.
1
Chen, J. K.; Taipale, J.; Young, K. E.; Maiti, T.; Beachy, P.
34.
Pace, C. J.; Gao, J., Exploring and exploiting polar-π
2
3
4
5
6
7
8
9
26.
Leitner, A., Cross-linking and other structural
36.
Kim, C.-Y.; Chang, J. S.; Doyon, J. B.; Baird, T. T., Jr.;
proteomics techniques: how chemistry is enabling mass
spectrometry applications in structural biology. Chem. Sci. 2016,
7, 4792-4803.
Fierke, C. A.; Jain, A.; Christianson, D. W., Contribution of
fluorine to protein-ligand affinity in the binding of fluoroaromatic
inhibitors to carbonic anhydrase II. J. Am. Chem. Soc. 2000, 122,
12125-12134.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
27.
Couvineau, A.; Tan, Y.-V.; Ceraudo, E.; Laburthe, M.,
Strategies for studying the ligand binding site of GPCRs:
photoaffinity labeling of the VPAC1 receptor, a prototype of class
B GPCRs. Methods Enzymol. 2013, 520 (G Protein Coupled
Receptors), 219-239.
37.
Crane, C. M.; Boger, D. L., Total synthesis of
[ψ[C(=S)NH]Tpg4]Vancomycin Aglycon,
[ψ[C(=NH)NH]Tpg4]Vancomycin Aglycon, and related key
compounds: reengineering vancomycin for dual D-Ala-D-Ala and
D-Ala-D-Lac binding. J. Am. Chem. Soc. 2012, 134, 1284-1297.
Xie, J.; Okano, A.; Pierce, J. G.; James, R. C.; Stamm, S.;
28.
Gilchrist, T. L.; Rees, C. W., Carbenes, Nitrenes, and
Arynes (Studies in Modern Chemistry). Nelson (London): 1969, pp
131.
29.
38.
Xie, J.; Pierce, J. G.; James, R. C.; Okano, A.; Boger, D. L.,
Bayley, H.; Knowles, J. R., Photogenerated reagents for
A redesigned vancomycin engineered for dual D-Ala-D-Ala and
D-Ala-D-Lac binding exhibits potent antimicrobial activity
against vancomycin-resistant bacteria. J. Am. Chem. Soc. 2011, 133,
13946-13949.
membrane labeling. 2. Phenylcarbene and adamantylidene
formed within the lipid bilayer. Biochemistry 1978, 17, 2420-2423.
30.
Dong, M.; Miller, L. J., Use of photoaffinity labeling to
understand the molecular basis of ligand binding to the secretin
receptor. Ann. N. Y. Acad. Sci. 2006, 1070 (VIP, PACAP, and
Related Peptides), 248-264.
39.
Lee, H.-J.; Choi, Y.-S.; Lee, K.-B.; Park, J.; Yoon, C.-J.,
Hydrogen bonding abilities of thioamide. J. Phys. Chem. A 2002,
106, 7010-7017.
31.
Grunbeck, A.; Sakmar, T. P., Probing G protein-coupled
40.
Huang, N.; Kalyanaraman, C.; Bernacki, K.; Jacobson, M.
receptor-ligand interactions with targeted photoactivatable cross-
linkers. Biochemistry 2013, 52, 8625-8632.
P., Molecular mechanics methods for predicting protein-ligand
binding. Phys. Chem. Chem. Phys. 2006, 8, 5166-5177.
32.
Ye, L.; Ding, K.; Zhao, F.; Liu, X.; Wu, Y.; Liu, Y.; Xue, D.;
41.
Liu, D. Z.; Tian, Z.; Yan, Z. H.; Wu, L. X.; Ma, Y.; Wang,
Zhou, F.; Zhang, X.; Stevens, R. C.; Xu, F.; Zhao, S.; Tao, H., A
structurally guided dissection-then-evolution strategy for ligand
optimization of smoothened receptor. MedChemComm 2017, 8,
1332-1336.
Q.; Liu, W.; Zhou, H. G.; Yang, C., Design, synthesis and
evaluation of 1,2-benzisothiazol-3-one derivatives as potent
caspase-3 inhibitors. Bioorg. Med. Chem. 2013, 21, 2960-2967.
33.
Coates, G. W.; Dunn, A. R.; Henling, L. M.; Dougherty,
D. A.; Grubbs, R. H., Phenyl-perfluorophenyl stacking
ACS Paragon Plus Environment