Pharmacophore-based virtual screening for identification of negative modulators of GLI1 as potential anticancer agents

Article abstract of DOI:10.1021/acsmedchemlett.9b00639

Starting from known GLI1 inhibitors, a pharmacophore-based virtual screening approach was applied to databases of commercially available compounds with the aim of identifying new GLI1 modulators. As a result, three different chemical scaffolds emerged that were characterized by a significant ability to reduce the transcriptional activity of the endogenous Hedgehog-GLI pathway and GLI1 protein level in murine NIH3T3 cells. They also showed a micromolar antiproliferative activity in human melanoma (A375) and medulloblastoma (DAOY) cell lines, without cytotoxicity in non-neoplastic mammary epithelial cells.

Full text of DOI:10.1021/acsmedchemlett.9b00639

pubs.acs.org/acsmedchemlett
Letter
Pharmacophore-Based Virtual Screening for Identification of
Negative Modulators of GLI1 as Potential Anticancer Agents
Fabrizio Manetti,* Barbara Stecca,* Roberta Santini, Luisa Maresca, Giuseppe Giannini,
Maurizio Taddei, and Elena Petricci*
Cite This: https://dx.doi.org/10.1021/acsmedchemlett.9b00639
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Starting from known GLI1 inhibitors, a pharmaco-
phore-based virtual screening approach was applied to databases of
commercially available compounds with the aim of identifying new
GLI1 modulators. As a result, three different chemical scaffolds
emerged that were characterized by a significant ability to reduce
the transcriptional activity of the endogenous Hedgehog-GLI
pathway and GLI1 protein level in murine NIH3T3 cells. They also showed a micromolar antiproliferative activity in human
melanoma (A375) and medulloblastoma (DAOY) cell lines, without cytotoxicity in non-neoplastic mammary epithelial cells.
KEYWORDS: Pharmacophore-based virtual screening, Hedgehog pathway, GLI-1 modulators, anticancer agents, melanoma,
medulloblastoma
The few GLI1 modulators known so far suffer from poor
drug-like properties and activity, thus prompting the
researchers to search for novel chemical entities. For this
purpose, following a common feature pharmacophore
generation approach implemented within the Phase software,23
the two compounds were used to build a three-dimensional
model that corresponded to the arrangement of their shared
chemical portions. The resulting five-feature pharmacophore
was constituted by two hydrophobic groups, two hydrogen
bond acceptors, and one aromatic ring. They mapped the two
terminal methyl groups of the m-prenyl side chain of GlaB and
of the 7-prenyl group of vismione E, the oxygen atoms at
positions 4 and 5 of GlaB and positions 8 and 9 of vismione E,
the condensed phenyl ring of GlaB, and the central phenyl ring
of vismione E, respectively (Figure 1B).
The pharmacophore was applied as a filter to virtually screen
the Asinex and the AKos databases of commercially available
compounds. As a result, among the 41 chemical entries
prioritized by the virtual screening protocol, three structurally
different hit compounds showed an interesting biological
profile. Among them, α-mangostine (SST0673, a diprenylxan-
thone resembling GlaB in the prenyl side chains and the 5,7-
dioxo-chromen-4-one core) showed a perfect superimposition
to the pharmacophoric model. In particular, the oxygen atoms
at positions 1 and 9 mapped the hydrogen bond acceptor
edgehog (HH) signaling pathway plays a crucial role in
Hembryonic cells differentiation and development, while it
is usually deactivated in adults.1,2 An aberrant activation of HH
pathway occurs in several so-called HH-dependent human
cancers, such as basal cell carcinoma (BCC),3 medulloblasto-
ma,4 meningioma,5 as well as in breast6 and colorectal
cancers,7 and many others.8 The G protein-coupled receptor
Smoothened (SMO) and the GLI transcription factors are the
most promising targets for HH modulation. SMO remains the
most investigated target for HH modulation9 with more than
18 compounds in preclinical studies (i.e., cyclopamine,10
CAT3,11 MRT-92)12 and active or completed clinical trials
(i.e., taladegib,13 LEQ506).14 Four SMO inhibitors are already
in the market for treatment of different cancers. Vismodegib
and sonidegib are employed for treatment of advanced BCC,
glasdegib for acute myeloid leukemia patients, and patidegib as
an orphan drug for the treatment of basal cell nevus syndrome,
as well as a topical gel formulation for sporadic nodular BCC.15
However, development of early resistance due to SMO
mutations has been observed in many patients treated with
vismodegib and sonidegib, thus limiting the efficacy of these
drugs.16 On the other hand, the downstream effector GLI1 can
be activated beyond SMO, overcoming its inhibition, and
probably represents the most promising target to treat HH-
dependent tumors avoiding resistance phenomena.17 The
naturally occurring compounds vismione E18 and glabrescione
B (GlaB)18−20 (Figure 1A) were recently described as
inhibitors of GLI1 transcriptional activity, acting to a similar
extent as GANT-61.21 However, a recent paper showed that
the binding modes of GlaB and GANT-61 are significantly
different.22
Special Issue: In Memory of Maurizio Botta: His
Vision of Medicinal Chemistry
Received: December 21, 2019
Accepted: March 25, 2020
Published: March 25, 2020
https://dx.doi.org/10.1021/acsmedchemlett.9b00639
© XXXX American Chemical Society
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX
A

ACS Medicinal Chemistry Letters
pubs.acs.org/acsmedchemlett
Letter
Figure 1. (A) Chemical structure of GlaB and vismione E used to generate the pharmacophore model for GLI1 inhibitors. (B) Graphical
representation of the five-feature pharmacophoric model for compounds affecting GLI1 protein level. GlaB (ball and stick notation) and vismione E
(stick notation) are superimposed. Green spheres are hydrophobic regions (HY); red spheres are hydrogen bond acceptor groups (HBA); the red
ring is an aromatic ring feature (RA).
Figure 2. Graphical representation of the best-ranked superposition of α-mangostine (left), SST0683 (middle), and SST0704 (right) to the
pharmacophoric model. Green spheres are hydrophobic regions (HY); red spheres are hydrogen bond acceptor groups (HBA); the red ring is an
aromatic ring feature (RA).
Figure 3. Effects of compounds on endogenous GLI1 protein level in NIH3T3 cells. Western blot analysis of GLI1 in cells treated with 100 nM
SAG and DMSO, GANT-61, GlaB, or our putative GLI inhibitors (5 μM) for 48 h. HSP90 was used as loading control. GLI1 protein
quantification is indicated in italic.
features (Figure 2, left), the methyl groups of the 8-prenyl
appendage were the hydrophobic portions, and ring A fitted
the aromatic ring feature.
at the thiophene represented the chemical features required to
map the pharmacophoric model (Figure 2, middle). All these
molecular portions were confirmed by molecular docking
simulations as very important for the binding of SST0682 to
GLI1 (Figure 5). Replacement of the phenyl ring at the amide
terminus with an ethyl chain as in SST0683 maintained a
micromolar antiproliferative activity (4.1 and 6.0 μM), a
reduction of GLI1 protein level, and 42% inhibition in the
luciferase assay. Further size reduction of the terminal
substituent to a methyl group led to SST0681 that did not
affect GLI1 protein level, while it retained a micromolar
antiproliferative activity (5.1 and 2.8 μM). Moreover, changing
the phenyl ring to the 2-furyl side chain led to SST0784, which
strongly reduced the GLI1 protein level maintaining micro-
molar antiproliferative activity on cancer cells, while enlarge-
ment to a naphthyl moiety resulted in the inactive SST0781
(Table 1). Decoration of the phenyl ring with the electron-
donating methyl group resulted in inactive compounds
(SST0783, SST0767, and SST0768) independently from the
substitution pattern. On the contrary, electron-withdrawing
substituents, such as nitro (SST0780) and halogens (SST0785
The biological evaluation of SST0673 showed a significant
reduction of GLI1 protein level in murine NIH3T3 cells
treated with SAG (100 nM) (Figure 3). Moreover, about 43%
inhibition was found in the luciferase assay in the same cell
line, in comparison to SAG (100% luciferase activity, Table 1,
Figure 4). Finally, this compound reduced cell proliferation at
micromolar concentrations (2.7 and 1.9 μM) in melanoma
(A375) and medulloblastoma (DAOY) cell lines. Interestingly,
the recent literature reports that α-mangostine may act as GLI
inhibitor.24,25 This result provides experimental evidence that
could support the reliability of the virtual screening protocol
for the identification of putative GLI1 inhibitors.
The second class of prioritized compounds was populated by
thiophene derivatives (Table 1). Among them, SST0682
showed a reduction of GLI1 protein level and an
antiproliferative activity in medulloblastoma cells significantly
stronger (0.9 μM) than in melanoma cells (12 μM). The
trichloromethyl group, the ester function, and the phenyl ring
B
https://dx.doi.org/10.1021/acsmedchemlett.9b00639
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
pubs.acs.org/acsmedchemlett
Letter
Table 1. Structure and Antiproliferative Activity of Compounds That Affect GLI1
a
Expressed as IC50 values (micromolar concentrations) calculated using GraphPad prism, version 6, from triplicate experiments in melanoma
b
(A375) and medulloblastoma (DAOY) cells. ND: not determined. No activ.: no effects on proliferative activity. GLI1 protein level was
determined by Western blotting (see Figure 3 for details) in murine NIH3T3 cells treated with SAG (100 nM) and each compound (5 μM) for 48
c
h. NA: not affected. Luciferase assay in murine NIH3T3 cells. Data are represented as percentage of inhibition. Cells treated with SAG were
equated to 100% (see Figure 4 for details). ND: not determined.
C
https://dx.doi.org/10.1021/acsmedchemlett.9b00639
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
pubs.acs.org/acsmedchemlett
Letter
and reduced GLI1 protein level. As a representative example,
SST0780 led to a >70% reduction of GLI1 protein level in
comparison to SAG-treated cells (Figure 3).
Within the third class of hit compounds (Table 1), the
pyrazolo[1,5-a]pyrimidine analogue SST0704 greatly de-
creased GLI1 protein level, inhibited the luciferase assay (by
40%), and showed a micromolar antiproliferative activity (2.2
and 2.3 μM) in A375 and DAOY cell lines, respectively. In the
attempt to rule SAR considerations on this class of
compounds, several derivatives of SST0704 were synthesized
or purchased from available commercial sources. The
analogues SST0894−897 were synthesized by cyclization of
the commercial 1-cyclopropyl-4,4-difluorobutane-1,3-dione 1
with pyrazole 2, thus producing the ester 3. It was converted,
after saponification, into the corresponding pyrazolo[1,5-
a]pyrimidines SST0894−897 in good yields and purity, by
treatment with the appropriate amines in the presence of
HATU as a coupling agent (Scheme 1). These analogs did not
affect GLI1 protein level (Supporting Information Figure 1A)
and, therefore, were not further characterized.
Figure 4. Effects of compounds on the transcriptional activity of the
endogenous Hedgehog pathway. Quantification of GLI-dependent
luciferase reporter assay in HH-responsive NIH3T3 cells treated with
100 nM SAG and the GLI inhibitors GANT-61, GlaB, or compounds
(5 μM) for 48 h. Relative luciferase activity was GLI-dependent
reporter firefly/renilla control ratios, with cells treated with SAG
equated to 100%. *, p < 0.05; **, p < 0.01.
Replacement of the C5 cyclopropyl ring with a methyl group
(SST0703) or removal of the gem-dimethyl substituent on the
dihydro-chromene moiety (SST0778) led to inactive com-
pounds that affected neither cell proliferation nor GLI1 protein
level, thus suggesting an important role for both the C5
substituent and the hydrophobic dimethyl moiety on the
chromene ring. Accordingly, docking simulations showed that
the dimethyl assemblage, which was required to satisfy the
hydrophobic features of the pharmacophoric model (Figure 2,
right), pointed toward the polymethylenic side chain of Arg223
(an essential amino acid for the interaction with DNA) and
Thr243 (Figure 6). On the other hand, the cyclopropyl ring
was directed toward the zinc binding site, possibly affecting the
coordination of this ion. Simplification of the chromene core to
a tetrahydro-naphthalene ring led to SST0788 that retained a
micromolar antiproliferative activity (5.0 and 7.5 μM) and
showed a significant reduction of GLI1 protein and inhibition
of the luciferase assay (54%). Although SST0788 lacked the
dimethyl substitution on the chromene moiety, it gained a
methylene fragment in place of the oxygen atom, which
maintained the hydrophobicity that seemed required on this
part of the molecule. Further manipulation of the appendage at
the amide side chain led to inactive compounds (SST0894-
SST0897). At the opposite molecular edge, insertion of small
aromatic rings at C5 led to the most active pyrazolo-pyrimidine
derivatives. In fact, SST0789 showed a micromolar anti-
proliferative activity (0.6 and 5.7 μM) and a significant
reduction of GLI1 protein level and luciferase activity (66%).
Similarly, the corresponding thiophenyl analogue SST0790
showed an antiproliferative activity (1.2 and 4.4 μM)
Figure 5. Best-ranked docking pose of the thiophene derivative
SST0682. The phenyl rings at the amide spacer and at the thiophene
nucleus are involved in π−cation interactions with the guanidino
moiety of Arg223 and the ammonium group of Lys209, respectively.
Moreover, the carbonyl oxygen of the ester side chain makes a
hydrogen bond with His220 of the zinc chelating architecture. The
trichloromethyl side chain is accommodated within the region
occupied by the dimethyl moiety of SST0704 (Figure 6). The
structure of SST0682 is represented by thick sticks, amino acid side
chains by thin sticks, the zinc ion as a blue sphere, hydrogen bonds as
dotted yellow lines, and π−cation interaction as dotted green lines.
and SST0779), restored a micromolar antiproliferative activity,
strongly affected luciferase assay (inhibition higher than 74%),
Scheme 1. Synthesis of Pyrazolo[1,5-a]pyrimidine Derivatives SST0894−897
D
https://dx.doi.org/10.1021/acsmedchemlett.9b00639
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
pubs.acs.org/acsmedchemlett
Letter
structure with those reported in previous literature gave further
support to the reliability of the computational docking
protocol. The latter was then applied also to thiophene and
pyrazolo-pyrimidine hit compounds in the attempt to predict
their binding mode and significant interactions with GLI1.
The thiophene derivative SST0682 was accommodated in
such a way that the phenyl ring at the amide portion showed a
cation−π interaction with the side chain of Arg223 (Figure 5),
thus accounting for the importance of a phenyl ring at this
position. Moreover, a similar interaction occurred between the
phenyl ring on the thiophene cycle (corresponding to the
aromatic ring feature of the pharmacophore) and the side
chain of Lys209. Further stabilization came by a hydrogen
bond between the ester side chain (that corresponds to an
HBA feature of the model) and the His220 ring. The
trichloromethyl group was embedded in a pocket comprised
within the side chains of Thr224 and Thr243, where also the
gem-dimethyl group of SST0704 was located (Figure 6).
The same binding pocket occupied by both GlaB and
SST0682 also accommodated the pyrazolo-pyrimidine deriv-
ative SST0704 (Figure 6). In particular, the pyrazole ring gave
a cation−π interaction with the terminal ammonium group of
Lys209, and the amide oxygen of the ligand (that corresponds
to one of the HBA features of the pharmacophore) interacted
by hydrogen bond with the His220 heterocycle. Hydrophobic
contacts involving both the chromene and the cyclopropyl
rings further stabilized the complex. In particular, the latter
pointed toward Cys207 that constituted the metal binding
pocket for zinc ion of the zinc finger 4.
To further confirm the effects of these compounds on the
GLI transcription factors, two pyrazolo-pyrimidines and two
thiophene derivatives were selected to perform a gene
expression analysis after treatment for 48 h at 5 μM.
Quantitative real-time PCR (qPCR) showed that all four
compounds are able to reduce the expression of GLI1, PTCH1,
HIP1, GLI2, and Cyclin D1 mRNA (Figure 7). Similar results
were obtained using GlaB. These results reinforce the effects of
our compounds as GLI inhibitors.
To confirm the specificity of our compounds for GLI
transcription factors, the effects of SST0683 and SST0790 on
melanoma cell viability in GLI1- or GLI2-silenced A375 cells
were also tested. This assay showed that depletion of GLI1 or
GLI2 greatly attenuated the effects of our putative GLI
inhibitors on cell viability, suggesting the specificity of our
compounds for GLI1 and GLI2 (Figure 8). Finally, to exclude
nonspecific cytotoxic effects, we demonstrated that the four
selected compounds had no effect on cell viability in a non-
neoplastic mammary epithelial cell line (MCF10A, Supporting
Figure 6. Best-ranked docking pose of SST0704 within GLI1 protein.
Important interaction sites are represented by the pyrazole ring that
gave a π−cation interaction with Lys209, the carbonyl group that is at
hydrogen bond distance from His220 of the zinc binding pocket, the
cyclopropyl ring that points toward the zinc binding pocket, and the
gem-dimethyl moiety that is allocated between the side chains of
Arg223 and Thr243. The structure of SST0704 is represented by
thick sticks, amino acid side chains in thin sticks, the zinc ion as a blue
sphere, hydrogen bonds as dotted yellow lines, and π−cation
interaction as dotted green lines.
associated to an almost complete inhibition of GLI1 protein
expression (Figure 3).
The hydrophobic character of pyrazolo-pyrimidines ap-
peared to be a crucial parameter that affected compound
activity. LogP values (calculated by MarvinSketch software,
www.chemaxon.com) lower than 3.5 were associated to
inactive compounds, while activity occurred in compounds
with higher values, with the only exception SST0896 that was
inactive with a logP value of 4.84.
The ability of such compounds to affect GLI1 protein level
and inhibit luciferase assay suggested that they might interact
with GLI1 itself. To support this hypothesis, molecular
docking simulations (Glide software)26 were performed on
the X-ray structure of the five-finger GLI1/DNA complex (2gli
within the protein data bank).27 Best scored poses of GlaB
showed that it accommodates in the region between zinc finger
4 and 5. Moreover, Lys209, Lys219, and Arg223 (we adopt
amino acid numbering of the protein data bank even if a list to
convert it into the full length numbering is reported in
Supporting Information Table 1) make crucial interactions
with GLI1 (Supporting Information Figure 2), as previously
described.18 The agreement between the binding pose of
docked GlaB and its major interactions within the GLI1
Figure 7. Gene expression analysis of Hedgehog pathway components upon treatment with compounds and GlaB. Quantitative PCR (qPCR)
analysis of GLI1, PTCH1, HIP1, GLI2, and Cyclin D1 in A375 cells treated with DMSO, SST0682, SST0683, SST0704, and SST0790 or GlaB (5
μM) for 48 h. The y axis represents expression ratio of gene/(HPRT and EIF2α average), with DMSO control equated to 1. Data shown as mean
SD.
E
https://dx.doi.org/10.1021/acsmedchemlett.9b00639
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
pubs.acs.org/acsmedchemlett
Letter
Figure 8. Silencing of GLI1 and GLI2 mitigates the effect of SST0683 and SST0790 on melanoma cell viability. (A) Western blot analysis of GLI1
and GLI2 in A375 cells transduced with LV-c, LV-shGLI1, or LV-shGLI2. HSP90 was used as loading control. (B−D) Effect of SST0683 (B) and
SST0790 (C) on viability of A375 melanoma cells transduced as indicated. GlaB was used as a reference compound (D). Cells were treated with
DMSO and compounds at the indicated doses for 72 h. Data are shown as mean SD of at least three independent experiments.
Author Contributions
Information Figure 1B), which expresses very low levels of
GLI1 and of other HH pathway components and does not
respond to HH pathway inhibition.28
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
In conclusion, a pharmacophore-based virtual screening
protocol led to discovery of three different chemical classes of
compounds that reduced the transcriptional activity of the
hedgehog-GLI pathway, affected GLI1 protein level, and
showed antiproliferative activity toward both human melano-
ma and medulloblastoma cancer cells. They represent a good
starting point for further steps toward a hit-to-lead process and,
finally, a lead optimization phase up to the development of a
new GLI1 negative modulator drug candidate.
Funding
This work was supported in part by a research grant from the
AIRC foundation (grant IG 2017 Id 20758) and by the project
“Development and application of QM/MM technologies for
the design of light responsive proteins or protein-mimics based
on rhodopsin architecture” within the program “Dipartimenti
di Eccellenza -2018−2022 financed by MIUR (Roma, Italy).
We are grateful for funding obtained from Istituto per lo
Studio, la Prevenzione e la Rete Oncologica (ISPRO). Luisa
Maresca is supported by a postdoctoral fellowship from Italian
Association for Cancer Research (AIRC, project n. 22644).
Additional financial support was provided by Leadiant
Biosciences S.A. (Mendrisio, CH) (formerly Sigma-Tau
Research Switzerland, S.A.). The MIUR research project
2015LZE994 (Insights into the functions of DNA damage
processing and repair factors to design novel selective
anticancer drugs), 2017SA5837 are also acknowledged.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acsmedchemlett.9b00639.
■
sı
Details of the computational protocols, synthetic
procedures, analytical data, and conditions for biological
assays. (PDF)
AUTHOR INFORMATION
Corresponding Authors
■
Notes
The authors declare the following competing financial
interest(s): G. Giannini participated in the project as Project
Leader, thanks to the agreement signed between Leadiant
Bioscience and Sigma-Tau IFR SpA (now Alfasigma SpA). Part
of the results described herein have been the subject of a
patent application: "GLI inhibitors and uses thereof"
Application No EP 17166194.5 April 12, 2017.
EP3388419A1. Giannini, G.; Taddei, M.; Manetti, F.; Petricci,
E.; Stecca, B.
Fabrizio Manetti − Dipartimento di Biotecnologie Chimica e
̀
Farmacia, Universita di Siena, I-53100 Siena, Italy; Lead
Discovery Siena, I-53100 Siena, Italy; orcid.org/0000-
0002-9598-2339; Phone: +39 0577 234330;
Email: fabrizio.manetti@unisi.it
Barbara Stecca − Istituto per lo Studio, la Prevenzione e la Rete
Oncologica (ISPRO), I-50139 Firenze, Italy; Phone: +39 055
7944567; Email: b.stecca@ispro.toscana.it
Elena Petricci − Dipartimento di Biotecnologie Chimica e
̀
Farmacia, Universita di Siena, I-53100 Siena, Italy;
Phone: +39 0577 234276; Email: elena.petricci@unisi.it
ACKNOWLEDGMENTS
■
Authors
The authors would like to thank G. Battistuzzi (R&D
Alfasigma S.p.A., formerly Sigma-Tau IFR S.p.A.) for assistance
with a few aspects of this project and A. Noseda (Leadiant
Biosciences SA) for facilitating this study. Glabrescione B was a
gift from B. Botta and L. Di Marcotullio, University of Rome
La Sapienza.
Roberta Santini − Istituto per lo Studio, la Prevenzione e la Rete
Oncologica (ISPRO), I-50139 Firenze, Italy
Luisa Maresca − Istituto per lo Studio, la Prevenzione e la Rete
Oncologica (ISPRO), I-50139 Firenze, Italy
Giuseppe Giannini − R&D, Alfasigma SpA, I-00071 Pomezia
(Roma), Italy; orcid.org/0000-0002-7127-985X
Maurizio Taddei − Dipartimento di Biotecnologie Chimica e
ABBREVIATIONS
̀
Farmacia, Universita di Siena, I-53100 Siena, Italy; Lead
■
Discovery Siena, I-53100 Siena, Italy
HH, Hedgehog; SMO, Smoothened; BCC, basal cell
carcinoma; HATU, hexafluorophosphate azabenzotriazole
tetramethyl uranium; GlaB, glabrescione B.
Complete contact information is available at:
https://pubs.acs.org/10.1021/acsmedchemlett.9b00639
F
https://dx.doi.org/10.1021/acsmedchemlett.9b00639
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
pubs.acs.org/acsmedchemlett
Letter
D’Amico, D.; Canettieri, G.; De Smaele, E.; Ferretti, E.; Screpanti, I.;
Uccello Barretta, G.; Botta, M.; Botta, B.; Gulino, A.; Di Marcotullio,
L. Gli1/DNA interaction is a druggable target for Hedgehog-
dependent tumors. EMBO J. 2015, 34, 200−217.
(19) Ingallina, C.; Costa, P. M.; Ghirga, F.; Klippstein, R.; Wang, J.
T.; Berardozzi, S.; Hodgins, N.; Infante, P.; Pollard, S. M.; Botta, B.;
Al-Jamal, K. T. Polymeric glabrescione B nanocapsules for passive
targeting of Hedgehog-dependent tumor therapy in vitro. Nano-
medicine 2017, 12, 711−728.
(20) D’Alessandro, G.; Quaglio, D.; Monaco, L.; Lauro, C.; Ghirga,
F.; Ingallina, C.; De Martino, M.; Fucile, S.; Porzia, A.; Di Castro, M.
A.; Bellato, F.; Mastrotto, F.; Mori, M.; Infante, P.; Turano, P.;
Salmaso, S.; Caliceti, P.; Di Marcotullio, L.; Botta, B.; Ghini, V.;
Limatola, C. 1H NMR metabolomics reveals the Glabrescione B
exacerbation of glycolytic metabolism beside the cell growth
inhibitory effect in glioma. Cell Commun. Signaling 2019, 17, 108.
REFERENCES
■
(1) Ingham, P. W.; McMahon, A. P. Hedgehog signaling in animal
development: paradigms and principles. Genes Dev. 2001, 15, 3059−
3087.
(2) Lum, L.; Beachy, P. A. The Hedgehog response network:
sensors, switches, and routers. Science 2004, 304, 1755−1759.
(3) Johnson, R. L.; Rothman, A. L.; Xie, J.; Goodrich, L. V.; Bare, J.
W.; Bonifas, J. M.; Quinn, A. G.; Myers, R. M.; Cox, D. R.; Epstein, E.
H., Jr.; Scott, M. P. Human homolog of patched, a candidate gene for
the basal cell nevus syndrome. Science 1996, 272, 1668−1671.
(4) De Souza, R. M.; Jones, B. R.; Lowis, S. P.; Kurian, K. M.
Pediatric medulloblastoma - update on molecular classification driving
targeted therapies. Front. Oncol. 2014, 4, 176.
(5) Aavikko, M.; Li, S. P.; Saarinen, S.; Alhopuro, P.; Kaasinen, E.;
Morgunova, E.; Li, Y.; Vesanen, K.; Smith, M. J.; Evans, D. G.;
̈
̈
Poyhonen, M.; Kiuru, A.; Auvinen, A.; Aaltonen, L. A.; Taipale, J.;
Vahteristo, P. Loss of SUFU function in familial multiple meningioma.
Am. J. Hum. Genet. 2012, 91, 520−526.
̈
(21) Lauth, M.; Bergstrom, Å.; Shimokawa, T.; Toftgård, R.
Inhibition of GLI-mediated transcription and tumor cell growth by
small-molecule antagonists. Proc. Natl. Acad. Sci. U. S. A. 2007, 104,
8455−8460.
(6) Bhateja, P.; Cherian, M.; Majumder, S.; Ramaswamy, B. The
Hedgehog signaling pathway: a viable target in breast cancer? Cancers
2019, 11, No. E1126.
(22) Calcaterra, A.; Iovine, V.; Botta, B.; Quaglio, D.; D’Acquarica,
I.; Ciogli, A.; Iazzetti, A.; Alfonsi, R.; Lospinoso Severini, L.; Infante,
P.; Di Marcotullio, L.; Mori, M.; Ghirga, F. Chemical, computational
and functional insights into the chemical stability of the Hedgehog
pathway inhibitor GANT61. J. Enzyme Inhib. Med. Chem. 2018, 33,
349−358.
(7) Gupta, R.; Bhatt, L. K.; Johnston, T. P.; Prabhavalkar, K. S.
Colon cancer stem cells: potential target for the treatment of
colorectal cancer. Cancer Biol. Ther. 2019, 20, 1068−1082.
(8) Lospinoso Severini, L.; Quaglio, D.; Basili, I.; Ghirga, F.;
Bufalieri, F.; Caimano, M.; Balducci, S.; Moretti, M.; Romeo, I.;
Loricchio, E.; Maroder, M.; Botta, B.; Mori, M.; Infante, P.; Di
Marcotullio, L. A Smo/Gli multitarget hedgehog pathway inhibitor
impairs tumor growth. Cancers 2019, 11, 1518.
(9) Bariwal, J.; Kumar, V.; Dong, Y.; Mahato, R. I. Design of
Hedgehog pathway inhibitors for cancer treatment. Med. Res. Rev.
2019, 39, 1137−1204.
(10) Taipale, J.; Chen, J. K.; Cooper, M. K.; Wang, B.; Mann, R. K.;
Milenkovic, L.; Scott, M. P.; Beachy, P. A. Effects of oncogenic
mutations in Smoothened and Patched can be reversed by
cyclopamine. Nature 2000, 406, 1005−1009.
(11) Chen, J.; Lv, H.; Hu, J.; Ji, M.; Xue, N.; Li, C.; Ma, S.; Zhou,
Q.; Lin, B.; Li, Y.; Yu, S.; Chen, X. CAT3, a novel agent for
medulloblastoma and glioblastoma treatment, inhibits tumor growth
by disrupting the Hedgehog signaling pathway. Cancer Lett. 2016,
381, 391−403.
(12) Vesci, L.; Milazzo, F. M.; Stasi, M. A.; Pace, S.; Manera, F.;
Tallarico, C.; Cini, E.; Petricci, E.; Manetti, F.; De Santis, R.;
Giannini, G. Hedgehog pathway inhibitors of the acylthiourea and
acylguanidine class show antitumor activity on colon cancer in vitro
and in vivo. Eur. J. Med. Chem. 2018, 157, 368−379.
(23) Dixon, S. L.; Smondyrev, A. M.; Rao, S. N. PHASE: a novel
approach to pharmacophore modeling and 3D database searching.
Chem. Biol. Drug Des. 2006, 67, 370−372.
(24) Verma, R. K.; Yu, W.; Shrivastava, A.; Shankar, S.; Srivastava, R.
K. α-Mangostin-encapsulated PLGA nanoparticles inhibit pancreatic
carcinogenesis by targeting cancer stem cells in human, and transgenic
(KrasG12D, and KrasG12D/tp53R270H) mice. Sci. Rep. 2016, 6, 32743.
(25) Ma, Y.; Yu, W.; Shrivastava, A.; Srivastava, R. K.; Shankar, S.
Inhibition of pancreatic cancer stem cell characteristics by α-
mangostin: molecular mechanisms involving Sonic Hedgehog and
Nanog. J. Cell. Mol. Med. 2019, 23, 2719−2730.
(26) Friesner, R. A.; Banks, J. L.; Murphy, R. B.; Halgren, T. A.;
Klicic, J. J.; Mainz, D. T.; Repasky, M. P.; Knoll, E. H.; Shaw, D. E.;
Shelley, M.; Perry, J. K.; Francis, P.; Shenkin, P. S. Glide: a new
approach for rapid, accurate docking and scoring. 1. Method and
assessment of docking accuracy. J. Med. Chem. 2004, 47, 1739−1749.
(27) Pavletich, N. P.; Pabo, C. O. Crystal structure of a five-finger
GLI-DNA complex: new perspectives on zinc fingers. Science 1993,
261, 1701−1707.
(28) Zhang, X.; Harrington, N.; Moraes, R. C.; Wu, M. F.;
Hilsenbeck, S. G.; Lewis, M. T. Cyclopamine inhibition of human
breast cancer cell growth independent of Smoothened (Smo). Breast
Cancer Res. Treat. 2009, 115, 505−521.
(13) Bendell, J.; Andre, V.; Ho, A.; Kudchadkar, R.; Migden, M.;
Infante, J.; Tiu, R. V.; Pitou, C.; Tucker, T.; Brail, L.; Von Hoff, D.
Phase I study of LY2940680, a Smo antagonist, in patients with
advanced cancer including treatment-native and previously treated
basal cell carcinoma. Clin. Cancer Res. 2018, 24, 2082−2091.
(14) Peukert, S.; He, F.; Dai, M.; Zhang, R.; Sun, Y.; Miller-Moslin,
K.; McEwan, M.; Lagu, B.; Wang, K.; Yusuff, N.; Bourret, A.;
Ramamurthy, A.; Maniara, W.; Amaral, A.; Vattay, A.; Wang, A.; Guo,
R.; Yuan, J.; Green, J.; Williams, J.; Buonamici, S.; Kelleher, J. F.;
Dorsch, M. Discovery of NVP-LEQ506, a second-generation inhibitor
of smoothened. ChemMedChem 2013, 8, 1261−1265.
(15) Peer, E.; Tesanovic, S.; Aberger, F. Next-Generation Hedge-
hog/GLI pathway inhibitors for cancer therapy. Cancers 2019, 11,
538.
(16) Pricl, S.; Cortelazzi, B.; Dal Col, V.; Marson, D.; Laurini, E.;
Fermeglia, M.; Licitra, L.; Pilotti, S.; Bossi, P.; Perrone, F.
Smoothened (SMO) receptor mutations dictate resistance to
vismodegib in basal cell carcinoma. Mol. Oncol. 2015, 9, 389−397.
(17) Pietrobono, S.; Gagliardi, S.; Stecca, B. Non-canonical
Hedgehog signaling pathway in cancer: activation of GLI transcription
factors beyond Smoothened. Front. Genet. 2019, 10, 556.
(18) Infante, P.; Mori, M.; Alfonsi, R.; Ghirga, F.; Aiello, F.;
Toscano, S.; Ingallina, C.; Siler, M.; Cucchi, D.; Po, A.; Miele, E.;
G
https://dx.doi.org/10.1021/acsmedchemlett.9b00639
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX