Xanthenylacetic Acid Derivatives Effectively Target Lysophosphatidic Acid Receptor 6 to Inhibit Hepatocellular Carcinoma Cell Growth

Article abstract of DOI:10.1002/cmdc.202100032

Despite the increasing incidence of hepatocellular carcinoma (HCC) worldwide, current pharmacological treatments are still unsatisfactory. We have previously shown that lysophosphatidic acid receptor 6 (LPAR6) supports HCC growth and that 9-xanthenylacetic acid (XAA) acts as an LPAR6 antagonist inhibiting HCC growth without toxicity. Here, we synthesized four novel XAA derivatives, (±)-2-(9H-xanthen-9-yl)propanoic acid (compound 4 – MC9), (±)-2-(9H-xanthen-9-yl)butanoic acid (compound 5 – MC6), (±)-2-(9H-xanthen-9-yl)hexanoic acid (compound 7 – MC11), and (±)-2-(9H-xanthen-9-yl)octanoic acid (compound 8 – MC12, sodium salt) by introducing alkyl groups of increasing length at the acetic α-carbon atom. Two of these compounds were characterized by X-ray powder diffraction and quantum mechanical calculations, while molecular docking simulations suggested their enantioselectivity for LPAR6. Biological data showed anti-HCC activity for all XAA derivatives, with the maximum effect observed for MC11. Our findings support the view that increasing the length of the alkyl group improves the inhibitory action of XAA and that enantioselectivity can be exploited for designing novel and more effective XAA-based LPAR6 antagonists.

Full text of DOI:10.1002/cmdc.202100032

Accepted Article
Title: Xanthenylacetic Acid Derivatives Effectively Target
Lysophosphatidic Acid Receptor 6 to Inhibit Hepatocellular
Carcinoma Cell Growth
Authors: Davide Gnocchi, Maria M. Cavalluzzi, Giuseppe F.
Mangiatordi, Rosanna Rizzi, Cosimo Tortorella, Mauro
Spennacchio, Giovanni Lentini, Angela Altomare, Carlo
Sabbà, and Antonio Mazzocca
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: ChemMedChem 10.1002/cmdc.202100032
Link to VoR: https://doi.org/10.1002/cmdc.202100032
01/2020

10.1002/cmdc.202100032
ChemMedChem
FULL PAPER
Xanthenylacetic Acid Derivatives Effectively Target
Lysophosphatidic Acid Receptor 6 to Inhibit Hepatocellular
Carcinoma Cell Growth
Davide Gnocchi,[a] Maria M. Cavalluzzi,[b] Giuseppe F. Mangiatordi,[c] Rosanna Rizzi,[c] Cosimo
Tortorella,[a] Mauro Spennacchio,[b],[c] Giovanni Lentini,[b] Angela Altomare,[c] Carlo Sabbà,[a] and
Antonio Mazzocca *[a]
[a]
Dr. D. Gnocchi, Dr. C. Tortorella, Dr. C. Sabbà, Dr. A. Mazzocca
Interdisciplinary Department of Medicine, University of Bari School of Medicine
Piazza G. Cesare, 11 - 70124 Bari, Italy
E-mail: antonio.mazzocca@uniba.it
[b]
Dr. M.M. Cavalluzzi, M. Spennacchio, Dr. Giovanni Lentini
Department of Pharmacy - Drug Sciences, University of Bari Aldo Moro
via Orabona, 4 - 70125 Bari, Italy
[c]
Dr. G.F. Mangiatordi, Dr. R. Rizzi, Dr. A. Altomare, M. Spennacchio.
Institute of Crystallography – CNR
Via Amendola 122/o – 70126 Bari, Italy
Supporting information for this article is given via a link at the end of the document.
Abstract: Despite the increasing incidence of Hepatocellular
carcinoma (HCC) worldwide, current pharmacological treatments are
still unsatisfactory. We have previously shown that lysophosphatidic
that 9-xanthenylacetic acid (XAA) inhibits HCC growth without
toxic effects by acting as an LPAR6 antagonist.[10]
It has been reported that chirality as well as the increase in
the degree of saturation [i.e., sp3 hybridized carbon atom fraction
(Fsp3)] are positively related to the potential clinical efficiency of
new drug candidates.[11] Here, we aimed to explore the effect of
the introduction of an alkyl group at the α-position of the XAA
carboxyl function on the interaction with LPAR6 and if such
modification could result in an improvement in the inhibition of
HCC cell growth. Hence, (±)-2-(9H-xanthen-9-yl)propanoic acid
acid receptor
6 (LPAR6) supports HCC growth and that 9-
xanthenylacetic acid (XAA) acts as an LPAR6 antagonist inhibiting
HCC growth without toxicity. Here, we synthesized four novel XAA
derivatives, (±)-2-(9H-xanthen-9-yl)propanoic acid (compound 4 –
MC9), (±)-2-(9H-xanthen-9-yl)butanoic acid (compound 5 – MC6),
(±)-2-(9H-xanthen-9-yl)hexanoic acid (compound 7 – MC11), and
(±)-2-(9H-xanthen-9-yl)octanoic acid (compound 8 – MC12, sodium
salt) by introducing alkyl groups of increasing length at the acetic α-
carbon atom. Two of these compounds were characterized by X-ray
powder diffraction and quantum mechanical calculations, while
molecular docking simulations suggested their enantioselectivity for
LPAR6. Biological data showed anti-HCC activity for all XAA
derivatives, with the maximum effect observed for MC11. Our
findings support the view that increasing the length of the alkyl group
improves the inhibitory action of XAA and that enantioselectivity can
be exploited for designing novel and more effective XAA-based
LPAR6 antagonists.
(compound 4
– MC9), (±)-2-(9H-xanthen-9-yl)butanoic acid
(compound 5 – MC6), (±)-2-(9H-xanthen-9-yl)hexanoic acid
(compound 7 – MC11), and (±)-2-(9H-xanthen-9-yl)octanoic acid
(compound 8 – MC12, sodium salt) were synthesized, and
structurally characterized by X-ray powder diffraction, submitted
to quantum mechanical calculations, and evaluated for their
biological activity in inhibiting HCC growth in two different cell
lines. Moreover, molecular docking simulations were performed
to obtain a reliable working hypothesis about which interactions
are critical for molecular recognition in the LPAR6 binding site.
These simulations gave interesting insights into the
characteristics of the LPAR6 binding site and provided a
theoretical basis for further optimization of these compounds.
Introduction
Today, despite the decline in the incidence of chronic hepatitis
infection, hepatocellular carcinoma (HCC) is the sixth leading
cause of cancer-related deaths in the world and it is expected to
become the third leading cause of cancer-related deaths in
Western countries by 2030.[1] The increase in the incidence of
metabolic diseases explains this trend. Indeed, recent evidence
suggests a correlation between metabolic syndrome, diabetes,
obesity, and HCC.[2] The pharmacological treatment of HCC is
mainly based on tyrosine kinase inhibitors (i.e., sorafenib,
regorafenib, and lenvatinib[2]), alone or combined with
immunotherapeutic drugs, such as pembrolizumab and
nivolumab.[3] Nevertheless, such approaches are characterized
by several unfavorable effects, making them not well tolerated
by patients in the long term.[4],[5] Thus, there is an extreme need
for new efficient and better-tolerated therapeutics for HCC. The
autotaxin-lysophosphatidic acid (ATX-LPA) axis signaling
pathway is particularly relevant to HCC development and
progression.[6],[7] We previously demonstrated that LPA triggers
the trans-differentiation of peritumoral tissue fibroblasts (PTFs)
in carcinoma-associated fibroblasts (CAF)[8] and that LPA
Results and Discussion
Synthesis of XAA derivatives
O
O
COOH
iii
+
R
COOH
2a-d
OH
1
R
COOH
5a (MC9)
5b (MC6)
5c (MC11)
5d (MC12)
ii
COOEt
a
b
c
R = Me
R = Et
n-
R = Bu
i
EtOOC
COOEt
R
COOEt
4c,d
d
n-
R = hexyl
3
Scheme 1: Reagents and conditions: i) suitable alkyl iodide, NaH, THF/DMF,
reflux, 24 h; ii) NaOH, H2O, reflux, 4 h; iii) KOH, pyrrolidine, i-PrOH, toluene,
reflux, 24 h.
receptor
6
(LPAR6) promotes HCC tumorigenicity, also
worsening clinical consequences in patients.[9] We later showed
1
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202100032
ChemMedChem
FULL PAPER
A
B
C
(±)-2-(9H-Xanthen-9-yl)alkanoic acids 5a–d were prepared
starting from 9-hydroxyxanthene (1) and suitable α-alkyl malonic
acids (2a–d). Since the previously reported procedure[10] gave
very low yield when applied to the synthesis of 5a, the synthesis
of 5a–d was carried out through alternative experimental
conditions that allowed us to ensure higher yields, while avoiding
the use of the toxic pyridine as a solvent. Therefore, 9-
hydroxyxanthene (1) was reacted with 2a–d in toluene at reflux
for 24 h, and pyrrolidine was used as an organocatalyst for the
decarboxylative reaction (Scheme 1).[12] In particular, methyl and
ethylmalonic acids (2a and 2b, respectively) were commercially
available, while n-butyl and n-hexylmalonic acids (2c and 2d,
7.35 7.30 7.25 7.20 7.15 7.10 7.05
f1 (ppm)
Figure 1. 1H NMR signals in the aromatic region of the spectra of XAA (A),
7.35 7.30 7.25 7.20 7.15 7.10 7.05 7.35 7.30 7.25 7.20 7.15 7.10 7.05
f1 (ppm) f1 (ppm)
MC9 (B), and MC6 (C).
respectively)
were
synthesized
modifying
literature
procedures.[12],[13] Diethyl malonate (3) was submitted to an
alkylation reaction at the alpha carbon with the suitable alkyl
iodide[13] to give the intermediates 4c and 4d which were in turn
hydrolyzed to the corresponding α-alkyl malonic acids[14] (2c and
2d).
Quantum mechanical calculations
The two higher homologs of XAA, namely MC9 and MC6,
displayed an increased reluctance to solubilization (data not
shown) and this was in line with the corresponding higher
lipophilic character (cLog P 2.5, 3.0, and 3.5 for XAA, MC9, and
MC6, respectively; Spartan'16). The pharmacological activity is
expected to increase with lipophilicity in the congeneric series.[11]
However, the higher lipophilicity per se might not be the only
factor affecting any observed hierarchy of activities. Indeed, the
alkyl substituents introduced onto the side chain α-position might
contribute to the interaction with specific lipophilic residues in the
binding site or forcing the orientation of the carboxylic group, a
putative pharmacophoric element, to favorable directions. The
latter possibility seems to be supported by the characteristic of
Figure 2. (A, B) Global minimum conformer (DFT B3LYP/6-31G*//DFT
B3LYP/6-31G*) of MC6; (C, D) folded conformer of MC6. The views in panels
(B) and (D) were obtained observing the models along the Cα-C9 bond
direction. For the sake of clarity, all the hydrogen atoms were removed. Color
legend: oxygen (red); carbon (grey).
X-ray powder diffraction
MC6, MC9, and XAA were characterized by X-ray powder
diffraction data analysis, and the structure determination process,
from pattern indexation to Rietveld refinement,[16] was performed
via the EXPO software.[17] The structure solution was obtained in
the reciprocal space[18] with Direct Methods for XAA and MC9,
and in the direct space[19] by Simulated Annealing algorithm for
MC6. The refined molecular structures of XAA, MC9, and MC6
are shown in Figure 3A, B, and C, respectively, resulting in good
agreement with the conformer distributions obtained with
quantum mechanical calculations (Figure 2). The powder X-ray
data collection information, the crystal structure analysis, and
the refinement statistics are described in the Experimental
Section.
1
the H NMR signals in the aromatic region (Figure 1), where an
increased rigidity of the rotatable bonds might be envisaged as
the cause of the progressive alteration of the signal definition
and up-field shift, observed when passing from XAA (Figure 1A)
to MC6 (Figure 1C) through MC9 (Figure 1B). Indeed, the
putative increase in rigidity may also be related to the relatively
low solubility observed in MC9.[15]
To gain an in-depth understanding of the effects of α-
substitution on the conformational freedom, the analysis of the
conformer distribution of the studied compounds was performed
together with an optimization of the geometric structure of the
corresponding conformers through quantum mechanical
calculations (DFT B3LYP/6-31G*//DFT B3LYP/6-31G*, gaseous
phase). The study was conducted on both separated
enantiomers and, as expected, similar results were obtained for
the enantiomers of both MC9 and MC6. For the sake of
simplicity, only the S enantiomers of MC9 and MC6 will be
discussed. Relatively to the benzo ring, the most stable
conformers of the three compounds have the same synclinal
and antiperiplanar orientation (Figure 2A and B for MC6).
However, a folded, less stable conformer (Figure 2C and D for
MC6) was found in the population of each analog (E = 2.0−3.2
kcal/mol, higher than the global minimum conformational isomer
energy). Based on the relative Boltzmann weights, this
conformer was increasingly more abundant when passing from
XAA to MC6 through MC9. Similar results were obtained when
running the above geometry optimization on the corresponding
ionized structures (XAA, MC9, and MC6 carboxylate anions)
applying the conductor-like polarizable continuum model (C-
PCM; Spartan'16), to allow for aqueous solvating effect
consideration.
A
B
C
Figure 3. Refined molecular structures in the asymmetric unit: XAA (A), MC9
(B), and MC6 (C). For the MC6 compound, the intermolecular hydrogen bonds
are shown. The hydrogen atoms are omitted for clarity. Color legend: oxygen
(red); carbon (grey).
Molecular docking
To gain insight into the molecular interactions established by
XAA, MC9, MC6, and MC11 in the LPAR6 binding site, we
conducted molecular docking simulations. It should be noted
that thanks to the recently released first X-ray structure of the
zebrafish LPAR6 (PDB code: 5XSZ),[20] reliable docking
simulations are now possible. Both enantiomers of MC9, MC6,
2
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202100032
ChemMedChem
FULL PAPER
and MC11 were considered during the performed simulations
and Figure 4 shows the obtained top-scored docking poses.
It is worth noting that docking simulations were also performed
on both MC12 enantiomers returning a different binding mode,
and a lower docking score in comparison with the other XAA
derivatives (data not shown). Therefore, the ΔGbind of this
compound was not calculated.
Table 1. ΔGbind computed for all the simulated compounds using MM-GBSA
calculations.
Compound
XAA
ΔGbind (kcal/mol)
-46.48
(S)-MC9
(R)-MC9
(S)-MC6
(R)-MC6
(S)-MC11
(R)-MC11
-47.76
-40.70
-50.65
-33.89
-50.54
-48.85
Biological findings
Effect of XAA derivatives on HCC cell growth
We tested the effect of XAA, MC9, MC6, MC11, and MC12 in
inhibiting HCC cell growth in two different cell lines, Huh7 and
HepG2. We took advantage of two different methods: end-point
cell counting by Trypan Blue exclusion test and crystal violet
staining. Results reported in Figure 5A and B show the following
efficacy trend in inhibiting HCC cell growth: MC11>
MC6>XAA≥ MC9>MC12, which is consistent with the docking
simulation data. This trend is corroborated also by experiments
on cell cycle phase distribution reported in Figure 5C, which
evidences a cytostatic action due to an arrest in the G1 phase of
the cell cycle. Figure 5D reports the IC50 calculated based on
the inhibitory activity of XAA derivatives on the growth of HCC
cells. We also tested the effect of the two optically active forms
of MC9, demonstrating the presence of enantioselectivity and
supporting docking simulation studies (Figure 5B).
Figure 4. Top-scored docking poses of XAA, (S)-MC9, (R)-MC9, (S)-MC6,
(R)-MC6, (S)-MC11 and (R)-MC11. Important residues are represented as
sticks, while LPAR6 protein structure is represented as a cartoon. Salt-bridge
and cation-pi interactions are depicted by black and cyan dotted lines,
respectively.
Since human cancers preferentially grow in acidic conditions,
which negatively affect the effectiveness of many anticancer
drugs[21],[22],[23], we tested the effect of XAA derivatives in such
conditions. We employed an acidic cell culture medium as
previously reported.[24] We found that acidic pH did not affect the
anti-proliferative effect of XAA derivatives in Huh7 cells, while in
HepG2 cells, this effect was reduced (Figure S2).
In particular, the same binding mode is predicted for XAA, (S)-
MC9, (S)-MC6, and (S)-MC11: the negatively charged carboxyl
group establishes two salt-bridge interactions with the positively
charged side chains of K26 and R281 while the xanthenyl group
interacts with R83 via cation-pi interaction. Noteworthy,
Taniguchi et al.[20] showed that K26, R281, and R83 are residues
particularly important for the recognition of the negatively
charged phosphate head group of LPA, based on mutagenesis
analyses. The authors also performed docking simulations of
LPA and hypothesized two possible binding modes, one of
which was characterized by two salt bridges involving the
phosphate group and both K26 and R281. These lines of
evidence strongly support the robustness of the predicted
binding mode for XAA, (S)-MC9, (S)-MC6, and (S)-MC11.
Different binding modes are instead predicted for (R)-MC9, (R)-
MC6, and (R)-MC11, thus suggesting that the binding site of
LPAR6 is enantioselective. Such a hypothesis is supported by
the obtained binding free energies (ΔGbind) computed through
MM-GBSA calculations (Table 1). Indeed, (S)-MC9 (-47.76
kcal/mol), (S)-MC6 (-50.65 kcal/mol) and (S)-MC11 (-50.54
kcal/mol) gave better values of ΔGbind with respect to (R)-MC9 (-
40.70 kcal/mol), (R)-MC6 (-33.89 kcal/mol) and (R)-MC11 (-
48.86 kcal/mol) respectively. Last but not least, both (S)-MC11
and (R)-MC11 return better ΔGbind than XAA (-46.48 kcal/mol),
suggesting that this compound has a higher affinity for LPAR6.
HepG2
A
B
Huh7
HepG2
** *** ***
*
Huh7
*
*
**
**
*
*
***
***
*
10
5
4
3
2
1
0
**
4000
3000
2000
1000
0
4000
3000
2000
1000
0
Total Count
Live Cells
***
*
*
**** **
*
** *** **** *
5
0
e
e
cl
i
e
0
e
0
M]
M]
M]
M]
M]
6
M]
M]
M]
µ
cl
M]
M]
i
M] M] M] M] M] M] M]
M] M] M] M] M] M] M]
µ µ µ µ µ µ µ
µ
µ
µ
µ
µ
µ
µ
cl
i
h
e
cl
i
h
[
µ
h
µ
h
µ
µ
µ
µ
µ
µ
µ
0
0
0
0
e
V
0
0
0
0
2
[
2
e
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
[
2
2
2
e
V
2
2
2
2
V
2
[
[
[
[
2
[
[
[
2
2
2
2
2
2
2
2
2
2
2
2
2
[
2
1
[
[
V
[
[
[
[
[
[
[
[
[
[
[
9
6
1
2
9
1
9
6
1
2
9
6
1
1
1
1
1
(-)
1
(-)
MC
MC MC
1
1
(+)
(+)
MC
MC
XAA
MC
MC
XAA
9
9
XAA
MC
XAA
MC
MC
9
MC
9
MC
MC
MC
MC
MC
MC
MC
MC
MC
C
D
HepG2
Huh7
100
100
100
50
0
IC50 (µM)
18.0
G0/G1
G2/M
S
XAA
80
60
40
20
0
22.0
MC9
MC6
MC11
MC12
XAA
16.0
14.0
26.0
MC9
MC6
MC11
MC12
50
0
e
M]
µ
M]
M]
µ
0
e
0
20
40
[Antagonist] [µM]
60
80
M]
9
M]
M]
M]
µ
cl
µ
M]
i
cl
i
µ
µ
µ
µ
0
2
[
h
0
0
h
0
0
0
0
2
[
e
2
2
2
[
1
e
2
[
[
2
2
V
[
[
[
V
6
9
6
1
1
1
MC
MC
XAA
MC
MC
MC
XAA
MC
Figure 5. Effect of XAA, MC9, MC6, MC11 and MC12 on HCC cell growth. (A)
Effect of XAA, MC9, MC6, MC11 and MC12 on HCC cell growth assessed by
end-point cell counting by Trypan Blue exclusion test. *p < 0.05, **p < 0.01,
***p < 0.001 as determined by Two Way ANOVA followed by Dunnett's post-
hoc test. Data represent three independent biological replicates performed in
3
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202100032
ChemMedChem
FULL PAPER
duplicate. (B) Effect of XAA, MC9, MC6, MC11 and MC12 on HCC cell growth
assessed by crystal violet staining. *p < 0.05, **p < 0.01, ****p < 0.0001 as
determined One-Way ANOVA followed by Dunnett's post-hoc test. Data
represent three independent biological replicates performed in duplicate. (C)
Effect of XAA, MC9, MC6, and MC11 on cell cycle phase distribution assessed
by a cytofluorimetric assay. Data are representative of two independent
biological replicates. (D) The IC50 of the pharmacological inhibitory effect of
XAA, MC9, MC6, MC11, and MC12 on the growth of HCC cell lines.
≥ MC9. Therefore, extending the length of the alkyl group on the
α-carbon of acetic acid improves the antagonistic activity of XAA
on LPAR6, thereby enhancing the effect of inhibiting the growth
of HCC cells. Nevertheless, the activity of MC9 is comparable to
XAA and this can be explained with the enantioselectivity of
1
LPAR6, as shown by docking simulation studies. The H-NMR
spectra recorded on the studied compounds point to possible
steric hindrance effects exerted by the α-substituent on the
rotation of the Cα–C9 bond. Quantum mechanical calculations
indicated two common possible conformations as the most
stable both in the gaseous and aqueous phases. The most
stable one presents the carboxyl group antiperiplanar to one
benzene ring and synclinal to the other. The second
conformation resulted to be folded so that the carboxyl group is
suspended on the planar moiety of the compounds. Interestingly,
the two most stable conformations are unambiguously confirmed
by the crystal structures as obtained by X-ray powder diffraction
data analysis (Figure 3), also supporting as the compounds
under investigation assume the same conformation in the solid-
state as in the gaseous and aqueous phase. The folded
conformer represents only a small fraction in the conformer
distribution (less than 2%). However, this relatively unstable
conformation was two times more probable in MC6 distribution
than in the distribution of MC9 and XAA. This observation may
suggest that increasing the size of the alkyl substituent in the
alpha-position of the side chain results in increased
pharmacological activity if the increased size of the substituents
would further increase the probability of the existence of the
putatively “right” conformation in the population of the so-
obtained congeners. Of course, an increase in the size of the
substituent should increase lipophilicity and contribute positively
to potency. Finally, it should not be overlooked the possibility
that the so-obtained congeners may better fit the binding site by
interacting with further lipophilic residues therein. It should be
noted that the efficacy of MC12 is lower than that of other
derivatives, which indicates that further increasing the length of
the alkyl group does not lead to an increase in inhibitory activity.
This effect may be due to steric hindrance.
Effect of the XAA derivatives on cell toxicity and apoptosis
We then examined the effect of XAA, MC9, MC6, MC11, and
MC12 on cell toxicity using two different tests: SRB and MTT.
Results reported in Figure 6 A and B show that all three
compounds under evaluation exert a similar effect, even if a
trend MC11 > MC6 > XAA ≥ MC9 can still be noticed. Also, we
evaluated the effect of XAA, MC9, MC6, MC11, and MC12 on
cell toxicity and apoptosis using a cytofluorimetric approach,
which allows us to discriminate viable, apoptotic, and dead cells,
and to determine the number of viable cells. Results reported in
Figure 6 C and D illustrate the same trend (MC11 > MC6 > XAA
≥ MC9 > MC12) in the number of viable cells. Moreover, MC6
and MC9 show a similar effect in inducing apoptosis, even if
some differences are observable between the two cell lines.
HepG2
A
B
Huh7
n.s.
n.s. n.s.
n.s. n.s.
*
Huh7
*** *** *
HepG2
*
2000
1500
1000
500
0
2000
1500
1000
500
0
4000
3000
2000
1000
0
4000
3000
2000
1000
0
n.s. n.s. n.s. n.s.
n.s.n.s.
***
*
*
**** ** **** *** ***
e
0
e
0
e
2
e
2
M] M] M] M] M] M] M]
M] M] M] M] M] M] M]
µ µ µ µ µ µ
M] M] M] M] M]
cl
cl
i
M] M] M] M] M]
cl
cl
i
h
e
i
i
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
0
µ
µ
µ
µ
µ
0
h
h
h
0
0
0
0
0
0
0
0
0
0
0
0
2
[
e
e
0
0
0
0
0
0
0
0
e
2
2
2
2
2
2
2
2
2
2
2
2
2
[
2
1
2
2
2
2
[
V
V
2
2
2
2
[
[
[
[
[
[
[
[
[
[
[
[
[
V
V
[
[
[
[
[
[
[
[
9
6
1
2
9
6
1
9
6
1
2
9
6
1
2
(-)
1
(-)
MC
1
1
1
1
(+)
(+)
1
1
9
9
XAA
MC
XAA
MC
MC
MC
9
MC
9
XAA
XAA
MC MC
MC
MC
MC MC
MC
MC
MC
MC
MC
MC
MC
MC
Huh7
** *** *** *** **
Huh7
HepG2
HepG2
C
D
4
100
50
0
100
50
0
4
3
2
1
0
*
*
*
*
*
Viable
Apoptotic
Dead
3
2
1
0
e
0
e
e
cl
i
e
0
M] M] M] M]
M] M] M] M]
M] M] M] M]
M] M] M] M]
cl
i
h
e
cl
i
M]
µ
cl
M]
M]
µ
M]
i
µ
µ
µ
µ
0
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
0
µ
h
µ
0
h
e
h
0
0
0
0
0
0
0
e
V
0
0
0
0
2
[
0
0
0
e
0
2
2
2
2
[
2
2
2
2
2
2
2
2
2
2
2
2
[
2
V
V
2
[
[
[
V
2
[
[
[
2
[
[
[
[
2
[
[
[
[
[
[
[
9
6
1
9
6
1
2
9
6
1
1
9
6
1
2
1
1
1
1
1
1
1
MC MC
XAA
MC MC
MC MC
XAA
MC MC
XAA
XAA
MC
MC
MC
MC
MC
MC
MC
MC
The picture that emerged from molecular docking
simulations is consistent with the experimental findings and
suggests two salt-bridge interactions with the positively charged
side chains of K26 and R281 and a cation-pi interaction
involving the xanthenyl group and R83 as crucial for the binding
of our derivatives towards LPAR6. It is worth mentioning here
that all these residues are fully conserved among different
species.[20] Moreover, it must be pointed out that for MC6, MC9,
and MC11, the top MM-GBSA scored pose was that returned by
the S-enantiomer, thus indicating that the interactions taking
place at the binding pocket could be stereoselective. Based on
these data, important clues for designing higher affinity ligands
can be derived. In particular, we can here hypothesize that
molecular recognition benefits from the presence of polar or
charged substituents on the xanthenyl group, which can interact
with R83 through H bonds or salt bridge interactions.
Figure 6. Effect of XAA, MC9, MC6, MC11 and MC12 on HCC cell toxicity. (A)
Effect of XAA, MC9, MC6, MC11 and MC12 on HCC cell toxicity assessed by
SRB staining test. *p < 0.05 and ***p < 0.001 as determined by One Way
ANOVA followed by Dunnett's post-hoc test. n.s. not significative. Data
represent three independent biological replicates performed in duplicate. (B)
Effect of XAA, MC9, MC6, MC11 and MC12 on HCC cell toxicity assessed by
MTT staining test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 as
determined One-Way ANOVA followed by Dunnett's post-hoc test. Data
represent three independent biological replicates performed in duplicate. (C)
and (D) Effect of XAA, MC9, MC6, MC11 and MC12 on cell toxicity and cell
death as assessed by cytofluorimetric approach. *p < 0.05, **p < 0.01, ***p <
0.001 as determined One-Way ANOVA followed by Dunnett's post-hoc test.
Data represent two independent biological replicates.
Owing to the high incidence of HCC and the lack of
adequate pharmacological approaches, the identification of
novel molecules capable of effectively inhibit LPAR6-driven HCC
growth with fewer side effects compared to the currently
available therapeutic approaches would be of high relevance. In
the search for novel LPAR6 antagonists, we recently identified
9-xanthenylacetic acid (XAA), which showed significant anti-
proliferative effects both in vitro and in vivo at therapeutic doses
without significant toxicity.[10] In this context, our work aimed to
synthesize, characterize, and study the biological effects of four
new XAA derivatives, namely (±)-2-(9H-xanthen-9-yl)propionic
acid (compound 4 - MC9), (±)-2- (9H-xanthen-9-yl)butyric acid
Conclusion
Hepatocellular carcinoma (HCC) is an emerging and severe
disease with a worldwide epidemiological diffusion. So far, the
pharmacological approaches available against HCC show
limited efficacy and evident side effects. Here, we presented four
novel xanthenylacetic acid (XAA) derivatives, MC9 (compound
4), MC6 (compound 5), MC11 (compound 7) and MC12
(compound 8), obtained by increasing the length of the alkyl
group at the acetic α-carbon atom. By using docking simulations,
we speculated that the binding of MC9, MC6, MC11, and MC12
can be enantioselective. The biological data presented in this
(compound
5 - MC6), (±)-2-(9H-xanthen-9-yl)hexanoic acid
(compound 7 – MC11), and (±)-2-(9H-xanthen-9-yl)octanoic acid
(compound 8 – MC12, sodium salt) as LPAR6 antagonists with
antitumor activity in HCC. In all the biological tests conducted in
this study, we obtained data that highlighted a trend in the
biological activity of these antagonists, namely MC6 > XAA
4
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202100032
ChemMedChem
FULL PAPER
GC-MS (70 eV) m/z (%) 268 (M+, <1), 181 (100). Anal.
(C17H16O3 0.2H2O) C, H.
study on MC9, MC6, MC11, and MC12 are consistent with
docking simulation data, thus corroborating their robustness.
Moreover, we showed that the inhibitory activity is not
maximized in MC12, indicating that further increasing the length
of the alkyl group will not lead to an improvement of the
inhibitory activity. Therefore, our results provide novel insights
for the design of novel LPAR6 antagonists with specific
translational potential. Also, our observations lay the foundation
and open the way for further improvements in the design of new
and more efficient LPAR6 antagonists. Indeed, our research is
now aimed to design and test novel XAA derivatives with
different groups. We provided evidence on the chemical and
physical properties of novel LPAR6 antagonists, thus paving the
way for the development of new pharmacological agents for
HCC. This will help expanding the therapeutic arsenal available
against HCC.
.
(±)-2-(9H-Xanthen-9-yl)hexanoic acid (5c, MC11)
Prepared as reported for 5a starting from 1 and butylmalonic acid (2c).
After purification of the crude by column chromatography on silica gel
(EtOAc/hexane, 2:8), the desired product (5c) was obtained as a pale
brown solid (85%): mp 131–133 °C; 1H NMR (500 MHz): δ 0.77 (t, J = 7.2
Hz, 3H, CH3), 1.02–1.35 (m, 5H, 2 × CH2 + CHHCH), 1.48–1.56 (m, 1H,
CHHCH), 2.55–2.60 (m, 1H, CHCH2), 4.31 (d, J = 7.1 Hz, 1H, CHAr),
7.09 (apparent t, 2H, ArH), 7.15 (d, J = 8.2 Hz, 2H, ArH), 7.20 (dd, J =
7.5, 1.4 Hz, 1H, ArH), 7.25–7.30 (m, 3H, ArH); 13C NMR (125 MHz): δ
13.9 (1C), 22.6 (1C), 27.9 (1C), 29.8 (1C), 42.5 (1C), 54.2 (1C), 116.7
(1C), 116.8 (1C), 122.7 (1C), 123.2 (1C), 123.6 (1C), 124.3 (1C), 128.3
(1C), 128.4 (1C), 128.6 (1C), 129.5 (1C), 153.19 (1C), 153.20 (1C),
180.4 (1C); HRMS (ESITOF) m/z [M – H]– calcd for C19H19O3: 295.1340;
.
found 295.1338. Anal. (C19H20O3 0.25H2O) C, H.
(±)-2-(9H-Xanthen-9-yl)octanoic acid (5d, MC12)
Prepared as reported for 5a starting from 1 and hexylmalonic acid (2d).
The desired product (5d) was obtained as a brown oil (47%) which was
converted into its sodium salt (pale brown waxy solid): 1H NMR (500 MHz,
CD3OD): δ 0.80 (t, J = 7.3 Hz, 3H, CH3), 1.05–1.10 (m, 4H, 2 × CH2),
1.16–1.20 (m, 2H, CH2), 1.30–1.32 (m, 3H, CH2 + CHHCH), 1.40–1.46 (m,
1H, CHHCH), 2.26–2.30 (ddd, J = 11.1, 7.8, 3.4 Hz, 1H, CHCH2), 4.23 (d,
J = 7.7 Hz, 1H, CHAr), 7.02–7.06 (m, 4H, ArH), 7.15–7.22 (m, 2H, ArH),
7.27 (dd, J = 7.6, 1.0 Hz , 1H, ArH), 7.43 (dd, J = 7.6, 1.0 Hz, 1H, ArH);
13C NMR (125 MHz, CD3OD): δ 14.4 (1C), 23.6 (1C), 29.1 (1C), 30.1
(1C), 30.4 (1C), 32.8 (1C), 44.1 (1C), 59.0 (1C), 116.9 (1C), 117.0 (1C),
123.6 (1C), 124.3 (1C), 125.6 (1C), 128.0 (1C), 128.3 (1C), 128.5 (1C),
130.1 (1C), 131.2 (1C), 154.3 (1C), 154.5 (1C), 182.1 (1C); HRMS
(ESITOF) m/z [M – H]– calcd for C21H24O3: 323.1653; found 323.1648.
Experimental Section
Chemistry part. All chemicals were purchased from Sigma-Aldrich in the
highest quality commercially available. Solvents were RP grade unless
otherwise indicated. Yields refer to purified products and were not
optimized. The structures of the compounds were confirmed by routine
.
Anal. (C21H23NaO3 1.33H2O) C, H.
spectrometric analyses. Melting points were determined on
a
Gallenkamp melting point apparatus in open glass capillary tubes and
are uncorrected. 1H NMR and 13C NMR spectra were recorded on a
Varian Mercury-VX spectrometer operating at 300 and 75 MHz for 1H and
13C, respectively, or an AGILENT 500 MHz operating at 500 and 125
MHz for 1H and 13C, respectively, using CDCl3 as the solvent, unless
otherwise indicated. Chemical shifts are reported in parts per million
(ppm) relative to solvent resonance: CDCl3, δ 7.26 (1H NMR), and δ 77.3
General procedure for the synthesis of n-alkylmalonic acids (4c,d)
The method adopted for the synthesis of n-butylmalonic acid (4c) is
described. NaH (600 mg, 25.0 mmol) was suspended in a mixture of THF
and DMF (3:1, 40 mL). Diethyl malonate 3 (2.0 g, 1.9 mL, 12.5 mmol)
was added dropwise and the solution was stirred for 15 min at room
temperature. 1-Iodobutane (2.53 g, 1.6 mL, 13.7 mmol) was added and
the mixture heated to reflux for 24 h. Then, the mixture was allowed to
cool and the solvent was removed under reduced pressure. The residue
was poured into water (20 mL) and extracted with diethyl ether (3 × 25
mL). The organic layer was dried over anhydrous Na2SO4 and the
solvent was removed under reduced pressure affording a yellow oil which
was used for the next step without further purification.
(
13C NMR). JJ values are given in Hz. EIMS spectra were recorded on a
Hewlett-Packard
6890-5973
MSD
gas
chromatograph/mass
spectrometer at low resolution. Elemental analyses were performed on a
Eurovector Euro EA 3000 analyzer. TLC analyses were performed on
precoated silica gel on aluminum sheets (Kieselgel 60 F254, Merck).
Diethyl butylmalonate 4c (2.46 g, 11.38 mmol) was added to a stirred
solution of sodium hydroxide (2.46 g, 61.5 mmol) in water (7 mL). The
reaction mixture was heated to reflux for 4 h, then cooled to room
temperature. 2 M HCl was added and the aqueous layer was extracted
with ethyl acetate (3 × 25 mL). The organic layer was washed with brine,
dried over anhydrous Na2SO4, and concentrated under reduced pressure.
Hexane (30 mL) was added to the residue and the solution was stirred
for 3 h. The product was isolated by filtration and dried under reduced
pressure affording 4c (1.43 g, 71%) as a white solid: m.p. 102–104 °C;
1H NMR (500 MHz): δ 0.92 (t, J = 7.1 Hz, 3H, CH3), 1.36–1.39 (m, 4H, 2
× CH2), 1.92–1.98 (m, 2H, CH2CH), 3.44 (t, J = 7.4 Hz, 1H, CH); HRMS
General procedure for the synthesis of (±)-2-(9H-xanthen-9-
yl)alkanoic acids (5a–d)
The method adopted for the synthesis of (±)-2-(9H-xanthen-9-
yl)propanoic acid (5a, MC9) is described. Methylmalonic acid (2a) (595
mg, 5.04 mmol), KOH (10 mg, 0.16 mmol), pyrrolidine (54 mg, 0.76
mmol), and isopropyl alcohol (30 mg, 0.50 mmol) were added to a stirred
solution of 9H-xanthen-9-ol (1) (500 mg, 2.52 mmol) in toluene (8 mL).
The reaction mixture was heated to reflux for 24 h and then cooled to
room temperature. The mixture was extracted with a saturated solution of
sodium carbonate (3 × 10 mL). After adjusting pH to 1–2 with 6 M HCl,
the aqueous phase was extracted with ethyl acetate (3 × 30 mL). The
combined organic phases were dried over anhydrous Na2SO4 and the
solvent removed under reduced pressure to give the desired compound
(5a) as a brown solid (0.634 g, 99%) which was recrystallized from
EtOAc/hexane (49%): mp 124–125 °C; 1H NMR (300 MHz): δ 0.92 (d, J =
7.0 Hz, 3H, CH3), 2.65–2.80 (m, 1H, CHCH3), 4.60 (d, J = 5.3 Hz, 1H,
CHAr), 7.04–7.15(m, 4H, Ar), 7.20–7.35 (m, 4H, Ar); 13C NMR (75 MHz):
δ 11.5 (1C), 41.6 (1C), 48.4 (1C), 116.5 (2C), 121.4 (1C), 123.2
(1C),123.5 (1C), 123.8 (1C), 128.1 (1C), 128.3 (1C), 128.5 (1C), 129.0
(1C), 152.9 (1C), 153.1 (1C), 179.3 (1C); GC-MS (70 eV) m/z (%) 254
+
(ESITOF) m/z [M + 2Na + H]+ calcd for C7H11Na2O4 205.0447; found
205.0448.
n-Hexylmalonic acid (4d)
Prepared as reported for 4c starting from 3 and 1-iodohexane as a gray
solid: m.p. 106–108 °C; 1H NMR (300 MHz): δ 0.88 (t, J = 6.6 Hz, 3H,
CH3), 1.25–1.35 (m, 8H, 4 × CH2), 1.92–1.98 (m, 2H, CH2CH), 3.44 (d, J
–
= 7.3 Hz , 1H, CH); HRMS (ESITOF) m/z [M – H]– calcd for C9H15O4
187.0976; found 187.0968.
.
(M+, <1), 181 (100). Anal. (C16H14O3 0.2H2O) C, H.
(±)-2-(9H-Xanthen-9-yl)butanoic acid (5b, MC6)
X-ray powder diffraction: data collection, structure determination,
and refinement. To complete the study and validate the reliability of
quantum mechanical analysis results, crystallography studies of
structural characterization of XAA, MC9, and MC6 compounds were
performed by using X-ray powder diffraction data. All the patterns were
collected at room temperature by using an automated Rigaku RINT2500
laboratory diffractometer (50 kV, 200 mA) equipped with the silicon strip
Rigaku D/teX Ultra detector. An asymmetric Johansson Ge[111] crystal
was used to select the monochromatic Cu Kα1 radiation (λ=1.54056 Å).
The angular range 6°-95° (2θ) for MC6 and 7°–80° (2θ) for XAA and
Prepared as reported for 5a starting from 1 and ethylmalonic acid (2b).
The desired product (5b) was obtained as a pale green solid (80%) which
was recrystallized from toluene/hexane (30%): mp 115–116 °C; 1H NMR
(300 MHz): δ 0.81 (t, J = 7.4 Hz, 3H, CH3), 1.30–1.44 (m, 1H, CHHCH3),
1.45–1.58 (m, 1H, CHHCH3), 2.46–2.54 (m, 1H, CHCH2), 4.30 (d, J = 7.0
Hz, 1H, CHAr), 7.05–7.20 (m, 4H, ArH), 7.22–7.35 (m, 4H, ArH); 13C
NMR (75 MHz): δ 12.0 (1C), 21.3 (1C), 42.2 (1C), 55.7 (1C), 112.5 (1C),
116.6 (1C), 122.6 (1C), 123.0 (1C), 123.4 (1C), 124.2 (1C), 128.1 (1C),
128.2 (1C), 128.4 (1C), 129.3 (1C), 153.0 (1C), 153.1 (1C), 178.6 (1C);
5
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202100032
ChemMedChem
FULL PAPER
MC9 was scanned with a step size of 0.02° (2θ) and counting time of 6
s/step. The measurements were executed in transmission mode, by
introducing the sample in a special glass capillary with a 0.5 mm
diameter and mounted on the axis of the goniometer. To reduce the
possible preferred orientation effects, the capillary was rotated during the
measurement to improve the randomization of the orientations of the
individual crystallites. The main acquisition parameters are reported in
Table S1 of the Supplementary Materials.
CQC] and treatments were performed using volumes not exceeding 1%
volume of the cell culture media. DMSO was used as vehicle control at
0.5% or 1% volume of cell culture media.
Cell proliferation assays. End-point proliferation was assayed by
Crystal Violet staining 72 hours after drug incubation. Crystal Violet
[Sigma-Aldrich cat. #C3886] was diluted in EtOH/H2O 10% v/v to obtain a
1
mg/mL solution, which was added to cells after fixation in 4%
The collected data were investigated via the EXPO software[17] in a
paraformaldehyde. The color was eluted with 10% acetic acid and
absorbance was read at λ=595 nm using an iMark™ plate reader [Bio-
Rad cat. #168-1135].
completely automatic way, performing the pathway of
a typical
polycrystalline structure determination process. It consists of the
calculation of the unit cell parameters, the identification of the space
group, the structure solution, and the structure model refinement via the
Rietveld method.[16] Each diffraction pattern was indexed using N-
TREOR09,[25] software integrated into EXPO, by selecting and fitting the
first 25 low-angle well-defined peaks of the powder pattern. The unit cell
parameters with the largest figures of merit [M(20 = 68, 45, and 34, for
XAA, MC9, and MC6, respectively] indexed all the peaks with an
orthorhombic cell. Systematic absences suggested the space group
P212121 for XAA, Pbcn for MC9, and Pna21 for MC6, as the most
probable ones.
Cell cycle analysis. Cell cycle analysis was performed using a Guava
EasyCyte benchtop flow cytometer [Merck cat. #0500-5009] employing
the "Guava Cell Cycle Assay" kit [Merck cat. #4500-0220]. Samples were
prepared following the producers' instructions. Briefly, cells were serum-
starved for 24 h when ≅ 30%-35% confluent and therefore treated with
the specified stimuli for the reported times. After culture media were
harvested, cells were washed twice with PBS, which was then collected
to have the whole cell population. Cells were then detached from
culturing support using trypsin and added to the collected media and
PBS. After centrifugation, cells were resuspended in PBS+2%FBS to
carefully remove culturing media, centrifuged again, and resuspended in
200 µL PBS+2%FBS. This cell suspension was then poured dropwise
into ice-cold 70% ethanol for fixation and permeabilization. After a
minimum of 24 h fixation at 4° C, ethanol was removed by centrifugation
and, after PBS washings, the reagent was added. Data were acquired
after 30 min incubation in the dark.
By using EXPO, the crystal structure determination process was
approached in the reciprocal space[18] by Direct Methods (DM) for XAA
and MC9 samples and in the direct space[19] via Simulated Annealing
(SA) global optimization algorithm for MC6.
DM are probabilistic approaches, based on two main steps: 1) the
decomposition of the experimental profile for extracting the integrated
intensities; 2) the solution of the phase problem. For the compounds
under investigation, the molecular structure determination process
required no default runs of EXPO, and all the obtained final DM models,
being approximate, were submitted to model optimization procedures.
They are integrated into EXPO as non-standard strategies and are aimed
at improving the structural model recovering the missed atoms, removing
occurred false positions, and locating atom peaks in better
Toxicity assays. Sulforhodamine B (SRB) assay was performed using a
commercial kit following the producer's directions of use [Canvax cat.
#CA050]. For MTT [SIGMA cat. #M2128] assay, a 5 mg/mL solution in
DPBS was added to the culture media (10% v/v). After 4 h incubation at
37°C, the color was eluted with an acidified isopropanol solution with 1%
Triton X-100 (0.1 M HCl in isopropanol 100%+Triton X-100 [Sigma-
Aldrich cat. #T9284]). Absorbance was measured using an iMark plate
reader at λ=570 nm.
positions.[26] [27]
,
About MC6, the structure solution was performed in the Pna21 space
group, implying the presence of two crystallographically independent
molecules in the asymmetric unit. Due to a large number of atoms (40
non-hydrogen atoms), DM are not able to find a correct and complete
final model and the structure was solved using the SA algorithm as
implemented in the EXPO software.
The method is based on the minimization of the difference between
observed and calculated intensities moving, within the unit cell, an
expected molecular model by varying its position, orientation, and
conformation. The angular range 6° < 2θ < 45.30° (2.0 Å of resolution)
was used. The starting expected model of MC6 was assembled using the
All of the toxicity assays were performed 48 h after adding treatments.
Determination of cell number, cell viability, and apoptosis. The
determination of cell number, cell viability, and apoptosis was performed
using a Guava EasyCyte benchtop flow cytometer [Merck cat. #0500-
5009] using the Guava ViaCount reagent [Merck cat. #4000-0040]
following producers' directions of use. This reagent allows a quantitative
evaluation of cell number, viability, and apoptosis by discriminating viable
and non-viable cells exploiting differential permeability properties of two
DNA-binding dyes. The nuclear dye stains only nucleated cells, whereas
the viability of dye stains dying cells. This permits us to distinguish viable,
apoptotic, and dead cells. Debris is excluded based on negative staining
with the nuclear dye. Briefly, cells were treated with the indicated stimuli
for the indicated times. Cells were then detached from culturing support
by using trypsin, and 50 µL of cell suspension was added to 450 µL of
reagent. After 10 minutes of incubation in the dark, data were acquired.
Assays were performed 48 h after adding treatments.
sketching
facilities
of
ACD/ChemSketch
(ACD/ChemSketch:
and the
http://www.acdlabs.com/resources/freeware/chemsketch/)
geometry was optimized by using MOPAC (J.J.P. Stewart, MOPAC2016.
Stewart Computational Chemistry (Colorado, CO, USA, Springs), 2016.
http://OpenMOPAC.net). A total of 17 parameters were optimized by
EXPO during the minimization process: five coordinates to describe the
position of the centers of mass, six angles describing the orientation, and
six torsion angles to describe the conformation. The algorithm was run 20
times and the best solution with the lowest cost function Rwp = 5.79 was
selected. The criterion to accept the solution was based also on the
soundness of the crystal packing. Direct Methods also confirmed the
solution achieved by the real space method.[28]
The obtained crystal structure solutions were supplied as a starting
model to the Rietveld refinement. All H atoms attached to C atoms were
treated as riding, with Uiso(H) = 1.2Uiso(C) of the carrier atoms. The
peak shapes were modeled using the Pearson VII function. The atomic
displacement parameters were refined isotropically and constrained to
have the same value for atoms of the same chemical element.
In Supplementary Materials, the detailed crystallographic results together
with the crystal structure refinement data are reported in Table S1 while
in Figure S1, and only for MC6 structure, the final Rietveld plot is
displayed.
Statistical analyses. In all other experiments, One-Way ANOVA
followed by Dunnett's post-hoc test determined statistical significance.
Normality was preliminary checked with D'Agostino-Pearson's Omnibus
K2 test.
The normality of data was preliminary verified with D'Agostino-Pearson's
Omnibus K2 test. One-Way ANOVA followed by Dunnett's post-hoc test
was used to determine statistical significance when data were normally
distributed. Otherwise, statistical significance was determined by the
Kruskal-Wallis test followed by Dunn's multiple comparisons test.
Statistical analyses and graphs were performed with Graphpad Prism 9
software.
Docking simulations. Compounds XAA, MC9, MC6 and MC11 were
docked on the recently published crystal structure of zebrafish LPAR6
(PDB code: 5XSZ).[20] Both enantiomers of MC9, MC6 and MC11 were
considered during the performed simulations. The retrieved .pdb file was
prepared using Protein Preparation Wizard, available in the Schrödinger
Suite 2019-3 for adding missing hydrogen atoms, reconstructing
incomplete side chains, and assigning favorable protonation states at
physiological pH [Schrödinger Release 2029-3: Protein Preparation
Wizard; Epik, Schrödinger, LLC, New York, NY, 2020; Impact,
Schrödinger, LLC, New York, NY, 2020; Prime, Schrödinger, LLC, New
York, NY, 2019]. To generate all the possible tautomers and ionization
states at a pH value of 7.0 ± 2.0, the ligands were prepared using
LigPrep, available in the Schrödinger Suite 2019-3 [Schrödinger Release
Further details of the crystal structure investigations may be obtained
from the joint CCDC/FIZ Karlsruhe online deposition service:
https://www.ccdc.cam.ac.uk/structures/ by quoting the deposition number
CCDC-1996157 for XAA, CCDC-1996155 for MC9, and CCDC-1996154
for MC6.
Biological studies
Cell lines and cell culture. HepG2 and Huh7 cell lines were purchased
from JCRB Cell Bank. HepG2 and Huh7 were grown in DMEM [Corning
cat. # 10-014-CVR] supplemented with 10% FBS [Corning cat. # 35-079-
CV]. Compounds used were diluted in DMSO [Corning cat. # 25-950-
6
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202100032
ChemMedChem
FULL PAPER
2019-3: LigPrep, Schrödinger, LLC, New York, NY, 2019]. The obtained
files were used for docking simulations performed by Grid-based ligand
docking with energetics (GLIDE) [Schrödinger Release 2019-3: Glide,
Schrödinger, LLC, New York, NY, 2019]. During the docking process, full
flexibility was allowed for the ligands while the protein was held fixed.
The default Force Field OPLS_2005[29] and the standard precision (SP)
protocol were employed. A cubic having an edge of 12 Å for the inner
box and 32 Å for the outer box and centered on the residues K26, R83,
R267 and R281 were employed.
[7]
S. Nakagawa, L. Wei, W. M. Song, T. Higashi, S. Ghoshal, R.
S. Kim, C. B. Bian, S. Yamada, X. Sun, A. Venkatesh, N. Goossens,
G. Bain, G. Y. Lauwers, A. P. Koh, M. El-Abtah, N. B. Ahmad, H.
Hoshida, D. J. Erstad, G. Gunasekaran, Y. Lee, M. L. Yu, W. L.
Chuang, C. Y. Dai, M. Kobayashi, H. Kumada, T. Beppu, H. Baba, M.
Mahajan, V. D. Nair, M. Lanuti, A. Villanueva, A. Sangiovanni, M.
Iavarone, M. Colombo, J. M. Llovet, A. Subramanian, A. M. Tager, S.
L. Friedman, T. F. Baumert, M. E. Schwarz, R. T. Chung, K. K.
Tanabe, B. Zhang, B. C. Fuchs, Y. Hoshida, C. Precision Liver
Cancer Prevention, Cancer Cell 2016, 30(6), 879-890.
[8]
Giannelli, Hepatology 2011, 54(3), 920-930.
[9] A. Mazzocca, F. Dituri, F. De Santis, A. Filannino, C. Lopane,
R. C. Betz, Y. Y. Li, N. Mukaida, P. Winter, C. Tortorella, G. Giannelli,
C. Sabba, Cancer Res 2015, 75(3), 532-543.
[10] D. Gnocchi, S. Kapoor, P. Nitti, M. M. Cavalluzzi, G. Lentini, N.
Denora, C. Sabba, A. Mazzocca, J Mol Med (Berl) 2019.
[11] M. M. Cavalluzzi, G. F. Mangiatordi, O. Nicolotti, G. Lentini,
Expert Opin Drug Discov 2017, 12(11), 1087-1104.
[12] S. J. Pridmore, J. M. J. Williams, Tetrahedron Lett 2008,
49(52), 7413-7415.
[13] C. Bruno, A. Catalano, J. F. Desaphy, M. M. Cavalluzzi, A.
Carocci, A. Dipalma, C. Franchini, G. Lentini, D. C. Camerino, V.
Tortorella, Heterocycles 2007, 71(9), 2011-2026.
[14] S. Yang, W. Shi, D. Xing, Z. Zhao, F. Lv, L. Yang, Y. Yang, W.
Hu, Eur J Med Chem 2014, 86, 133-152.
[15] J. Huuskonen, D. J. Livingstone, D. T. Manallack, Sar and
Qsar in Environmental Research 2008, 19(3-4), 191-212.
[16] H. M. Rietveld, Phys Scripta 2014, 89(9).
[17] A. Altomare, C. Cuocci, C. Giacovazzo, A. Moliterni, R. Rizzi,
N. Corriero, A. Falcicchio, J Appl Crystallogr 2013, 46, 1231-1235.
[18] R. Rizzi, A. Altomare, Crystals 2020, 10(12).
[19] W. I. F. David, K. Shankland, Acta Crystallogr A 2008, 64, 52-
64.
MM-GBSA calculations. The binding free energies (∆Gbind) between
protein and ligands were computed by applying the Molecular
Mechanics/Generalized Born Surface Area (MM-GBSA)[30] calculations
on the obtained top-scored docking poses. More specifically, we
employed Prime software, available in the Schrodinger Suite 2019-3
[Schrödinger Release 2020-3: Prime, Schrödinger, LLC, New York, NY,
2020], applying the following Equation:
A. Mazzocca, F. Dituri, L. Lupo, M. Quaranta, S. Antonaci, G.
ΔGbind = ΔEMM + ΔGsolv + ΔGSA
where ΔEMM, ΔGsolv, and ΔGSA represent the difference between the
contribution made by the ligand-protein complex and the sum of those
made by the ligand and the protein taken alone, in terms of minimized
energy, solvation energy, and surface area energy, respectively.
Flexibility was allowed to all the residues having at least one atom within
a distance of 5 Å for the ligand.
Quantum mechanical calculations. Calculations were performed
according to procedures we previously developed.[31] Briefly, the models
of the undissociated compounds XAA, MC9, and MC6 were generated
from the atomic fragments incorporated into Spartan'16 (Wavefunction
Inc., Irvine, CA) inner fragment library and assuming the suggested
default starting geometries. The generated geometries were optimized by
the molecular mechanics MMFF routine offered by the software[32] and
then submitted to a systematic conformational distribution analysis using
the default step sizes. All conformers in a window of 10 Kcal/mol above
the global minimum conformer were retained. When two conformers
differed by dihedral values lower than 10°, the less stable conformer was
left out. Conformers were then classified according to their ab initio gas-
phase energy content calculated at the RHF/6-31G* level. All conformers
falling within a window of 5 kcal/mol above the global minimum were
retained and submitted to RHF/6-31G* geometry optimization. After
removal of redundant conformers (i.e., each conformer differing from a
more stable one by less than 5° in their corresponding dihedral values),
the so-obtained set of conformers underwent geometry optimization by
density functional theory (DFT) implemented in Spartan'16 with B3LYP
functional[33] and the 6-31G* basis setw[34] in the gas phase. The
optimized structures were confirmed as real minima by IR frequency
calculation (DFT B3LYP/6-31G*//DFT B3LYP/6-31G*). The above
geometry optimization was performed also on XAA, MC9, and MC6
carboxylate anions applying the conductor-like polarizable continuum
model (C-PCM; Spartan'16) to allow for aqueous solvating effect
consideration.[35],[36]
[20] R. Taniguchi, A. Inoue, M. Sayama, A. Uwamizu, K.
Yamashita, K. Hirata, M. Yoshida, Y. Tanaka, H. E. Kato, Y.
Nakada-Nakura, Y. Otani, T. Nishizawa, T. Doi, T. Ohwada, R.
Ishitani, J. Aoki, O. Nureki, Nature 2017, 548(7667), 356-360.
[21] S. R. Pillai, M. Damaghi, Y. Marunaka, E. P. Spugnini, S. Fais,
R. J. Gillies, Cancer Metastasis Rev 2019, 38(1-2), 205-222.
[22] R. J. Gillies, C. Pilot, Y. Marunaka, S. Fais, Biochim Biophys
Acta Rev Cancer 2019, 1871(2), 273-280.
[23] E. P. Spugnini, S. Fais, Expert Opin Ther Pat 2020, 30(1), 15-
25.
[24] D. Gnocchi, L. Del Coco, C. R. Girelli, F. Castellaneta, G.
Cesari, C. Sabba, F. P. Fanizzi, A. Mazzocca, Sci Rep 2021, 11(1),
1259.
[25] A. Altomare, G. Campi, C. Cuocci, L. Eriksson, C. Giacovazzo,
A. Moliterni, R. Rizzi, P. E. Werner, J Appl Crystallogr 2009, 42, 768-
775.
Acknowledgements
[26] A. Altomare, C. Cuocci, C. Giacovazzo, G. S. Kamel, A.
Moliterni, R. Rizzi, Acta Crystallogr A 2008, 64, 326-336.
[27] A. Altomare, C. Cuocci, C. Giacovazzo, A. Moliterni, R. Rizzi, J
Appl Crystallogr 2012, 45, 789-797.
[28] A. Altomare, N. Corriero, C. Cuocci, A. Falcicchio, R. Rizzi,
Crystals 2020, 10(1).
We are grateful to Ms. Loredana Angela Acquaro and Dr. Filippo
Capodiferro for their technical support.
[29] J. L. Banks, H. S. Beard, Y. X. Cao, A. E. Cho, W. Damm, R.
Farid, A. K. Felts, T. A. Halgren, D. T. Mainz, J. R. Maple, R. Murphy,
D. M. Philipp, M. P. Repasky, L. Y. Zhang, B. J. Berne, R. A.
Friesner, E. Gallicchio, R. M. Levy, J Comput Chem 2005, 26(16),
1752-1780.
Keywords: drug design • enantioselectivity • hepatocellular
carcinoma • lysophosphatidic acid receptor 6 antagonists •
therapeutic tools
[30] S. Genheden, U. Ryde, Expert Opin Drug Dis 2015, 10(5),
449-461.
[31] A. Carocci, M. Roselli, R. Budriesi, M. Micucci, J. F. Desaphy,
C. Altamura, M. M. Cavalluzzi, M. Toma, G. I. Passeri, G. Milani, A.
Lovece, A. Catalano, C. Bruno, A. De Palma, F. Corbo, C. Franchini,
S. Habtemariam, G. Lentini, Chemmedchem 2021, 16(3), 578-588.
[32] T. A. Halgren, J Comput Chem 1996, 17(5-6), 520-552.
[33] A. D. Becke, Phys Rev A 1988, 38(6), 3098-3100.
[34] E. R. Davidson, D. Feller, Chem Rev 1986, 86(4), 681-696.
[35] A. W. Lange, J. M. Herbert, J Chem Phys 2010, 133(24).
[36] A. W. Lange, J. M. Herbert, Chem Phys Lett 2011, 509(1-3),
77-87.
[1]
C. R. de Lope, S. Tremosini, A. Forner, M. Reig, J. Bruix, J
Hepatol 2012, 56 Suppl 1, S75-87.
[2]
[3]
[4]
[5]
A. Forner, M. Reig, J. Bruix, Lancet 2018.
M. Kudo, Cancers (Basel) 2018, 10(11).
O. Waidmann, Expert Opin Biol Ther 2018, 18(8), 905-910.
Z. Wu, L. Lai, M. Li, L. Zhang, W. Zhang, Medicine (Baltimore)
2017, 96(51), e9431.
[6]
Harokopos, P. Foka, A. Dimitriadis, K. Evangelou, A. N. Moulas, U.
Georgopoulou, V. G. Gorgoulis, G. N. Dalekos, V. Aidinis,
Hepatology 2017, 65(4), 1369-1383.
E. Kaffe, A. Katsifa, N. Xylourgidis, I. Ninou, M. Zannikou, V.
7
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202100032
ChemMedChem
FULL PAPER
Entry for the Table of Contents
Increased activity
O
O
O
O
XAA
MC9
MC6
MC11
OH
CHR3
COOH
CH3-CH2
R
COOH
CH3-CH2CH2-CH2 COOH
R
Enantioselectivity
on LPAR6
Growth inhibition
Tumor cell
Xanthenylacetic acid (XAA) derivatives effectively inhibit hepatocellular carcinoma cell growth by targeting lysophosphatidic acid
receptor 6 (LPAR6) in an enantioselective manner. Our results bring original insights on the characteristics of the LPAR6 binding site
and on the chemico-physical properties of the novel LPAR6 antagonists. This will drive the design of more efficient LPAR6
antagonists expanding the therapeutic tools against HCC.
8
This article is protected by copyright. All rights reserved.