S1PR2 inhibitors potently reverse 5-FU resistance by downregulating DPD expression in colorectal cancer

Article abstract of DOI:10.1016/j.phrs.2020.104717

In this study, S1PR2 was reckoned as a brand-new GPCR target for designing inhibitors to reverse 5-FU resistance. Herein a series of pyrrolidine pyrazoles as the S1PR2 inhibitors were designed, synthesized and evaluated for their activities of anti-FU-resistance. Among them, the most promising compound JTE-013, exhibited excellent inhibition on DPD expression and potent anti-FU-resistance activity in various human cancer cell lines, along with the in vivo HCT116^DPD cells xenograft model, in which the inhibition rate of 5-FU was greatly increased from 13.01percent–75.87percent. The underlying mechanism was uncovered that JTE-013 demonstrated an anti-FU-resistance activity by blocking S1PR2 internalization to the endoplasmic reticulum (ER), which inhibited the degradation of 5-FU into α-fluoro-β-alanine (FBAL) by downregulating tumoral DPD expression. Overall, JTE-013 could serve as the lead compound for the discovery of new anti-FU-resistance drugs. Significance: This study provides novel insights that S1PR2 inhibitors could sensitize 5-FU therapy in colorectal cancer.

Full text of DOI:10.1016/j.phrs.2020.104717

Journal Pre-proof
S1PR2 inhibitors potently reverse 5-FU resistance by downregulating
DPD expression in colorectal cancer
Yu-Hang Zhang, Dong-Dong Luo, Sheng-Biao Wan, Xian-Jun Qu
PII:
S1043-6618(19)32493-4
DOI:
https://doi.org/10.1016/j.phrs.2020.104717
YPHRS 104717
Reference:
To appear in:
Pharmacological Research
Received Date:
Revised Date:
Accepted Date:
6 November 2019
6 February 2020
17 February 2020
Please cite this article as: Zhang Y-Hang, Luo D-Dong, Wan S-Biao, Qu X-Jun, S1PR2
inhibitors potently reverse 5-FU resistance by downregulating DPD expression in colorectal
cancer, Pharmacological Research (2020), doi: https://doi.org/10.1016/j.phrs.2020.104717
This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal
pertain.
© 2020 Published by Elsevier.

S1PR2 inhibitors potently reverse 5-FU resistance by downregulating DPD
expression in colorectal cancer
Running title: S1PR2 inhibitors prevent 5-FU resistance by downregulating DPD
Yu-Hang Zhang1†, Dong-Dong Luo2†, Sheng-Biao Wan2*, Xian-Jun Qu1*
1Department of Pharmacology, School of Basic Medical Sciences, Capital Medical
University, Beijing, China, 100069
2Laboratory for Marine Drugs and Bioproducts of Qingdao National Laboratory for
Marine Science and Technology, Key Laboratory of Marine Drugs, Ministry of
Education, School of Medicine and Pharmacy, Ocean University of China, Qingdao,
China, 266003
†These authors contributed equally: Yu-Hang Zhang, Dong-Dong Luo
*Corresponding authors: Sheng-Biao Wan and Xian-Jun Qu
Tel/Fax: 86-010-83911516, e-mail: quxj@ccmu.edu.cn or biaowan@ouc.edu.cn
Graphical abstract
1

Highlights



S1PR2 is the potential upstream regulator of DPD expression in CRC cells
JTE-013 effectively reversed 5-FU resistance of CRC both in vivo and in vitro
JTE-013 levels off DPD-catalyzed degradation of intracellular 5-FU in CRC
Abstract
In this study, S1PR2 was reckoned as a brand-new GPCR target for designing
inhibitors to reverse 5-FU resistance. Herein a series of pyrrolidine pyrazoles as the
S1PR2 inhibitors were designed, synthesized and evaluated for their activities of anti-
2

FU-resistance. Among them, the most promising compound JTE-013, exhibited
excellent inhibition on DPD expression and potent anti-FU-resistance activity in
various human cancer cell lines, along with the in vivo HCT116DPD cells xenograft
model, in which the inhibition rate of 5-FU was greatly increased from 13.01% to
75.87%. The underlying mechanism was uncovered that JTE-013 demonstrated an
anti-FU-resistance activity by blocking S1PR2 internalization to the endoplasmic
reticulum (ER), which inhibited the degradation of 5-FU into α-fluoro-β-alanine
(FBAL) by downregulating tumoral DPD expression. Overall, JTE-013 could serve as
the lead compound for the discovery of new anti-FU-resistance drugs.
Significance: This study provides novel insights that S1PR2 inhibitors could sensitize
5-FU therapy in colorectal cancer.
Key words: Colorectal cancer (CRC); S1P receptor 2 (S1PR2); Dihydropyrimidine
dehydrogenase (DPD); 5-FU resistance.
Introduction
5-Fluorouracil (5-FU) has long been the basic drug for colorectal cancers (CRCs),
however, drug resistance remains the great obstacle to its clinical use [1, 2]. In
mammalian cells, 5-FU is converted into fluorodeoxyuridine monophosphate
3

(FdUMP), which forms a stable complex with thymidylate synthase (TS or TYMS)
and causes cytotoxicity by inhibiting TS activity [3, 4]. Dihydropyrimidine
dehydrogenase (DPD or DPYD)-mediated conversion from 5-FU to
dihydrofluorouracil (DHFU) is a rate-limiting step of 5-FU catabolism [5, 6].
Therefore, high levels of TS and DPD are widely taken as two major causes of 5-FU
resistance [7]. Currently, many reports have been issued about the modulatory
mechanisms of TS expression, yet the knowledge of DPD is still limited to its single
nucleotide polymorphisms (SNPs). Nonetheless, known SNPs only account for
approximately 30% cases of severe resistance to 5-FU [8, 9] and the majority of
patients who developed 5-FU resistance do not carry known toxicity-associated SNPs
for DPYD [10]. Most importantly, 5-FU itself is the ideal inhibitor of TS and
leucovorin is used to stabilize FdUMP-TS complex [11, 12], whereas current DPD
inhibitors (i.e. CDHP in S-1 regimen) suppress catalytic activity of DPD only in liver
but not in cancer cells [13, 14]. Moreover, it was reported that these reversible
competitive enzyme inhibitors of DPD augmented the liver toxicity of 5-FU [15].
Therefore, it is of paramount importance to elucidate the regulatory target of tumoral
DPD for the sake of optimizing 5-FU regimens.
Even though it was reported that increased H3K27me3 at the DPYD promoter
could downregulate DPD expression [16], it was still difficult for an inhibitor to
directly enter into the nucleus and modify the methylation of H3K27 in cancer cells. If
a transmembrane receptor could be identified as the upstream regulator of
4

H3K27me3, the inhibitors of this transmembrane receptor would effectively
downregulate the expression of DPD in cancer cells. S1PR2, one of five G protein-
coupled receptors (GPCRs) of sphingosine-1-phosphate (S1P), has been ignored in the
development of drug discovery [17]. The immunomodulatory drug fingolimod
(FTY720, GilenyaR) was approved for treatment of relapsing-remitting multiple
sclerosis, due to its impressive efficacy and good tolerability [18-20]. As fingolimod
resembles in chemical structure of sphingosine, it serves as the phosphorylated
substrate of sphingosine kinase 2 (SphK2) [21, 22]. The metabolite fingolimod-P
showed similar affinity as S1P by binding with S1PR1, S1PR3, S1PR4 and S1PR5
[23, 24]. However, it was verified that Fingolimod-P is not a ligand for S1PR2 [25].
Several inhibitors of S1PR1,3-5, such as Ponesimod and Etrasimod, have currently
entered into the stage III clinical trials [26, 27]. However, S1PR2 is still left as the
vacant GPCR target for further investigation.
In this study, we found that S1PR2 inhibitors could downregulate DPD
expression to suppress the degradation of intracellular 5-FU into inactive α-fluoro-β-
alanine (FBAL). A series of novel pyrrolidine pyrazole inhibitors referenced to JTE-
013 were synthesized and evaluated for their activity of overcoming 5-FU resistance
through downregulating tumoral DPD. We thus suggest that JTE-013, the most
superior S1PR2 inhibitor, could be developed as a potential reversal agent of 5-FU
resistance for clinical applications.
5

Results
Identification of S1PR2 as a potential upstream regulator of DPYD expression in
TCGA database and cell lines. To investigate the potential role of S1PR2 in
regulating 5-FU resistance, we firstly screened the expression data from The Cancer
Genome Atlas (TCGA). Overall, S1PR2 expression was significantly increased in a
variety of tumoral samples as compared to the paracancerous tissues, including
colorectal cancer, skin cutaneous melanoma and brain lower grade Glioma (Fig. 1A
and 1B). Among these cancer types, CRC was the only cancer type given 5-FU-based
adjuvant chemotherapy as the first-line treatment. Since DPD (encoded by DPYD or
dpyd), TS (encoded by TYMS) and thymidine phosphorylase (encoded by TYMP) are
all considered to be involved in the development of 5-FU resistance, we further
screened TCGA database to identify which gene might be upregulated by S1PR2high
DPYD expression showed positive correlation with S1PR2 in CRCs (R = 0.75) but
.
not in paracancerous tissues (R = 0.31). Neither TYMS nor TYMP were significantly
correlated to S1PR2 in CRCs or paracancerous tissues (Fig. 1C and D).
Further, we determined the expressions of S1PR2 and DPD in various naïve cell
lines, including HCT116, HT-29, SW620, LOVO, NCM460, MCF-7, BT474, Hela,
Huh7 and cccHEL-1 cells. It showed a significant difference in the expressions of
S1PR2 and DPD among these cell lines but a satisfying consistence between high
S1PR2 and elevated DPD. HCT116, SW620, Huh7 and cccHEL-1 cells were found
with both higher S1PR2 and elevated DPD as compared with other cell lines (Fig.
6

1E). We further performed the RT-qPCR assay to analyze the mRNA levels of s1pr2
and dpyd. Higher levels of s1pr2 and dpyd were also seen in HCT116, SW620, Huh7
and cccHEL-1 cells than other cell lines (Fig. 1F). Also, S1PR2 shRNA was
transfected into HCT116, SW620, Huh7 and cccHEL-1 cells to determine their DPD
levels. It showed that these four cell lines transfected with S1PR2 shRNA all
exhibited lower DPD expression as compared with their vector controls (Fig. 1G).
Finally, we treated the above shS1PR2 cells with dose gradient of 5-FU with the
concentrations from 6.25 to 100 μM. It showed that S1PR2 shRNA significantly
increased the sensitivity of 5-FU treatment in HCT116, SW620, Huh7 cells but not in
cccHEL-1 cells (Fig. 1H). Overall, these results suggest that S1PR2 might be a novel
GPCR target for designing reversal agents of 5-FU resistance by downregulating DPD
expression.
Synthetic chemistry of S1PR2 inhibitors. Since shS1PR2 could effectively enhance
the sensitivity of cancer cells to 5-FU treatment by downregulating DPD expression,
we sought to design, synthesize and evaluate a series of compounds referenced to
JTE-013. The syntheses of S1PR2 inhibitors were performed as shown in Scheme 1.
All S1PR2 inhibitors were synthesized from hydrazine intermediates (6a-6c, 13a-13k
or 22) by condensation with the corresponding acid azide. The reflux of 5-amino-1,3-
dimethylpyrazole in acetic acid with esters afforded the correspondent compounds 4a-
4c. The reaction yield was low and the separation of the product was difficult through
7

silica gel chromatography due to the acylated side product from the reaction of 5-
amino-1,3-dimethylpyrazole with solvent acetic acid with close polarity as the
product. Fortunately, ethyl acetate allowed the separation of the product from the
acylated side product. Compounds 5a-5c afforded from bromination of 4a-4c,
followed by reaction with 80% hydrazine hydrate in ethanol to produce the key
intermediates 6a-6c. The 2,6-dichloropyridine moiety 8 was derived from the
commercially available 2,6-dichloroisonicotinic acid 7 in straightforward steps.
Reflux of compound 8 in toluene produced the isocyanate 9. Compounds 10a-10c
were obtained through subsequent adding of the solutions of hydrazine 6a-6c in
tetrahydrofuran (THF). Iodination of compound 4a to produce compound 11, and the
replacements of iodine atom by different substitutions were carried out through
Suzuki coupling reaction to obtain 12a-12k. Subsequently, the same procedures were
performed to produce 14a-14k. The different substituted isonicotinic acid 15a-15d
were used to gain acid azide 16a-16d and subsequent combination with 6a to obtain
18a-18d. Starting from 2,4-dichloro-7H-pyrrolo[2,3-d] pyrimidine 19, compound 21
was obtained through substitution and methylation. Compound 23 was then prepared
following a similar two-step sequence as shown in Scheme 1 (for more details, please
see the Supplementary Methods).
SAR of S1PR2 inhibition. In vitro anti-5-FU resistance activities against human
colorectal cell lines, including HCT116, HT-29 and NCM460, were analyzed by 3-
8

(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. The
inhibition rates of these compounds at the concentrations of 6.25, 12.5, 25, 50, 100,
200 μM and the IC50 values were listed in Table 1 (left panel). All these compounds
were then combined with 10 μM 5-FU for further analysis of the EC50 values. The
results were listed in Table 1 (right panel). JTE-013, the reported S1PR2 inhibitor,
was resynthesized and identified as a hit compound with the EC50 of 18.28 μM in
HCT116 cells. The design of inhibitors could not be aided by crystallographic
methods because crystal structure of S1PR2 has still not been elucidated. Therefore,
the SAR of S1PR2 inhibition is mainly illuminated by introducing various
modifications to different regions of the hit molecule.
SAR exploration initiated with the investigation of various substitutions at the
pyrrolopyridine structure of S1PR2 inhibitors (Table 1). Methyl and cyclopropyl
substitutions on the pyridine ring show no improvement (10a and 10b vs JTE-013),
whereas phenyl substitution exhibits better S1PR2 potencies (10c vs JTE-013). These
results indicate that the inhibitory effect of the compounds on S1PR2 is enhanced by
the increase of volume of the hydrophobic group. Replacements of hydrogen on
pyrrole ring of 10a with bulk group are also beneficial (14a-14k), in particular, as are
pyridine (14a) and fluorine benzene (14d). The effect of the chloric substitutions at
2,6-dichloropyridine structure on JTE-013 was subsequently investigated. The mono-
substituted pyridines showed no improvement (18a-18d vs JTE-013), whereas non-
substituted pyridine is a weaker inhibitor than the substituted pyridines. The above
9

results demonstrate that moderate bis-substitutions are more suitable. The impact of
pyrrolopyridine structure was subsequently examined and replacements of
pyrrolopyridine moiety on hit 1 with pyrrolopyrimidine are not beneficial (23 vs JTE-
013). In conclusion, the data of SAR illustrate that the attachment of alkyl or aryl
groups to the fused ring greatly improves S1PR2 potency and the inhibitory effect is
further enhanced by the addition of size of hydrophobic group.
According to the EC50 and IC50 values, we subsequently selected 5 most potent
and selective S1PR2 inhibitors to analyze their activity of inhibiting DPD expressions
by western blotting assay. It showed that compounds 10c and JTE-013 demonstrated
strong activity in preventing the expression of DPD (Fig. 2A). Moreover, we
previously ascertained that 5-FU could induce S1PR2 to internalize to the ER. Herein,
we found that compounds 10c and JTE-013 effectively restrained the 5-FU-induced
S1PR2 internalization (Fig. 2B). Enlightened from our previous study, we further
validated the ER stress signaling under the dose gradient of JTE-013 with the
concentrations from 1.3 to 100 μM. The reported JMJD3-H3K27me3-DPD pathway
was also determined to be gradually suppressed when cancer cells were treated with
higher dose of JTE-013 (Fig. S1). These in vitro results suggest that compounds 10c
and JTE-013 could be developed as the potential 5-FU resistance inhibitors through
the mechanism of preventing tumoral DPD.
10

Binding models for compounds 10c and JTE-013 in the S1PR2-binding pocket.
To explore the interaction between receptor residues and ligands, a homology model
of S1PR2 was established and the binding site was deduced based on the known
crystal complex of S1PR1 and ligand (Fig. 3A). Compounds 10c and JTE-013 (Fig.
3B) were well docked into the ligand-binding cavity. Comparing the binding modes of
compound 10c to JTE-013 (Fig. 3C and D), both 10c and JTE-013 adopt a similar
“U-shape” conformation. For JTE-013, however, five hydrogen bonds are formed:
semicarbazide to Tyr18, Arg108 and Val182 and nitrogen (pyridine) of the pyrazol-
pyridine ring to Glu109. In contrast, compound 10c formed one more hydrogen bond
between semicarbazide and Glu109, which might interpret why compound 10c has
stronger in vitro activity. These interactions, especially interaction with
semicarbazide, were identified as the crucial driving force for the S1PR2 binding.
Furthermore, the pyrazol-pyridine ring, the isopropyl in JTE-013 and the phenyl in
10c occupied the hydrophobic channel, formed by six residues Ala36, Ile40, Phe86,
Thr278, Phe274, and Ala275. Especially, one more hydrogen bond between 10c and
S1PR2 might be caused by the differences of hydrophobicity between isopropyl and
phenyl, which leads to slight differences in the conformations of both molecules. In
addition, the pyrrole-pyridine ring in JTE-013 and 10c could interact with the phenyl
groups of Phe274 through π−π stacking. The other parts of the structures,
dichloropyridine rings, also contributed to the interactions between S1PR2 and their
inhibitors through the hydrophobicity of chlorine atoms.
11

JTE-013 and compound 10c reversed 5-FU resistance in HCT116DPD cell
xenograft model. To compare the anti-5-FU resistance properties of compounds 10c
and JTE-013 in vivo, HCT116 cells with high DPD (HCT116DPD) xenografted
athymic mice model was established. In this model, we treated the mice with 20
mg/kg 5-FU combined with 1 mg/kg compound 10c or 1 mg/kg JTE-013 by tail vein
injection for consecutive 24 days. As shown in Fig. 4A, body weights of 5-FU+JTE-
013-treated mice rose again after the fall, whereas body weights of control groups and
5-FU+10c group continuously leveled off. Tumor growth was strongly inhibited by
JTE-013 plus 5-FU. The inhibition rate was greatly increased from 13.01% by single
use of 5-FU to 75.87% by JTE-013 plus 5-FU. In contrast, anti-5-FU resistance
activity of compound 10c was not as potent as JTE-013 and the inhibition rate of
compound 10c plus 5-FU was 46.52% in HCT116DPD xenografted athymic mice (Fig.
4B-D). In addition, JTE-013 monotherapy hardly affected the tumor proliferation with
the inhibition rate of 3.41% in the HCT116DPD model (Fig. S2).
To determine the activity of compounds 10c and JTE-013 in preventing DPD
expression in vivo, we analyzed DPD levels in colons, tumors and livers of mice
bearing HCT116DPD xenograft. Western blotting assay showed that DPD was highly
expressed in the livers, moderately expressed in the tumors and weakly expressed in
normal colons (Fig. 5A). Compounds 10c and JTE-013 strongly inhibited DPD
expression in both colonic tumors and livers (Fig. 5B). Immunohistochemistry with
12

DPD was also examined in the paraffin embedded tumor sections. As compared to the
positive staining in a majority of tumoral, colonic and liver cells in control and 5-FU
group, compounds 10c and JTE-013 strongly prevented DPD expressions (Fig. 5C).
As for 5-FU-induced S1PR2 internalization, tumoral tissues were all stained with
S1PR2 by both immunohistochemical and immunofluorescent assays. As shown in
Fig. 5D, S1PR2 merged with the ER marker Calnexin in cancer cells of 5-FU-treated
mice, while S1PR2 maintained membranous expression in naïve mice. In contrast,
there was no significant 5-FU-induced S1PR2 internalization in tumoral tissues after
the treatment of compounds 10c and JTE-013. These results indicated that
compounds 10c and JTE-013 inhibited DPD expression through restraining the 5-FU-
induced S1PR2 internalization.
JTE-013 reversed 5-FU resistance by reducing DPD-catalyzed degradation of
intracellular 5-FU. Now that JTE-013 significantly reversed 5-FU resistance by
decreasing DPD expression in vivo, we sought to verify the role of JTE-013 among
various cell lines. As shown in Fig. 6A, DPD expression demonstrated the tendency
of dose-dependent JTE-013 treatment in HCT116, SW620 and cccHEL-1 cells but
not in NCM460, Huh7 and Hela cells. Then we determined the inhibition rates of dose
gradient of 5-FU combined with 5, 10 and 20 μM JTE-013 in HCT116 and cccHEL-1
cells. Higher concentrations of JTE-013 more effectively elevated the sensitivity of 5-
FU treatment in HCT116 and cccHEL-1 cells (Fig. 6B). Since CDHP was taken as the
13

inhibitor of liver DPD [28], we subsequently compared the intracellular 5-FU levels
between cccHEL-1 cells and HCT116 cells exposed to 5-FU (25 mg/L) which were
respectively combined with CDHP (9 mg/L) or JTE-013 (25 mg/L) for 6 h.
Impressively, on the one hand, HPLC-UV assay showed that the degradation of
intracellular 5-FU was inhibited by both CDHP and JTE-013 in cccHEL-1 cells. It
showed that the concentration of intracellular 5-FU was only 0.068 mg/L in cccHEL-1
cells, while CDHP and JTE-013 could increase the intracellular 5-FU to 14.003 mg/L
and 16.769 mg/L, respectively (Fig. 6C and Fig. S3A-S3C). On the other hand, the
degradation of intracellular 5-FU was only inhibited by JTE-013 but not CDHP in
HCT116 cells. It was determined that the intracellular 5-FU was 0.336 mg/L in
HCT116 cells. CDHP did not significantly elevate the intracellular 5-FU, but JTE-
013 could increase the intracellular 5-FU to 5.763 mg/L (Fig. 6C and Fig. S3D-S3F).
The tendency of corresponding FBAL levels was in accordance with the degradation
of intracellular 5-FU (Fig. 6C). In addition, even though it was reported that JTE-013
was designed as the inhibitor of S1PR2 [29], we also treated HCT116shS1PR2 and
cccHEL-1shS1PR2 cells with 20 μM JTE-013 combined with dose gradient of 5-FU for
further analyzing their inhibition rates. It depicts that JTE-013 had no anti-5-FU
resistance activity in the above HCT116shS1PR2 and cccHEL-1shS1PR2 cells, as compared
to HCT116vector and cccHEL-1vector control cells (Fig. 6D). It thus further suggested the
anti-5-FU resistance activity of JTE-013 was due to targeting at S1PR2 but not
through other targets. Overall, these results manifested that JTE-013 reversed 5-FU
14

resistance by inhibiting DPD expression, which blocked the degradation of
intracellular 5-FU into inactive FBAL in both liver and tumoral cells.
Discussion
This study was designed to develop a novel therapeutic strategy for overcoming 5-FU
resistance. The major findings are as follows. First, our results demonstrate that
S1PR2 is closely associated with the expression of DPD in various cancer cells.
Second, by use of the S1PR2 inhibitor JTE-013 as molecular template, lead
optimization was carried out in an attempt to improve potency and drug-like
molecular properties by modifications of pyrropyrimidine structure of JTE-013. In
particular, successful conversion of the isopropyl and methyl moiety of JTE-013 into
a phenyl and hydrogen moiety without loss of agonist potency led to a highly potent
compound 10c, which acted as the lowest EC50 in vitro. Compound 10c appears to be
a promising agent that merits further investigation. In addition, in silico homology-
docking suggested that Tyr18, Arg108 and Val182 and Glu109 may be critical
residues to interact with antagonists, which contribute to a better understanding of
pharmacological behaviors caused by the interaction with S1PR2. Third, compound
10c exhibited weaker activity to reverse 5-FU resistance as compared with JTE-013 in
mouse xenograft model, limiting its further investigation. We speculate that it may
result from differences of metabolic stability. Para-hydrogen atom in the lipophilic
phenyl ring is easily oxidized into 4-hydroxyl product by P450 enzyme in the liver,
15

which was rapidly metabolized. Overall, this study demonstrated that the promising
compound, JTE-013, exhibited excellent inhibition on 5-FU resistance through
suppressing DPD expression both in vitro and in vivo. Based on the above results,
JTE-013 has been selected as a lead candidate and is currently undergoing further
characterization of the preclinical profile in our follow-up studies. In order to find
better compounds to reverse 5-FU resistance, we will also further modify the leading
compound to improve pharmacodynamics and pharmacokinetic properties of
compound 10c.
In summary, the present study identified S1PR2 as a potential upstream regulator
of DPD expression, developed an approach to quickly obtain potent S1PR2 inhibitors
and envisioned that an aromatic ring replacement would be suitable. In addition, we
found a lead compound 10c with good activity in vitro and a promising drug
candidate JTE-013 with excellent in vivo activity for further optimization and study.
Materials and Methods
Human cells lines. Human colorectal cancer cell lines HCT116 (ATCC® CCL-
247™), SW620 (ATCC® CCL-227™), HT-29 (ATCC® HTB-38™) and LOVO
(ATCC® CCL-229). Human breast cancer cell line MCF-7 (MCF7 ATCC® HTB-
22™), human breast cancer cell line BT-474 (ATCC® HTB-20), cervical cancer cell
line HeLa (ATCC® CCL-2) and hepatocellular cancer cell line Huh7 (ATCC® PTA-
4583). Human liver healthy cell line cccHEL-1 and human colon healthy cell line
16

NCM460 (Incell Corp, San Antonio) were purchased from Shanghai Cell Bank,
Chinese Academy of Science (Shanghai, China). Cells were maintained in RPMI-
1640 or DMEM (Gibco) supplemented with 10% FBS (Gibco) at 37°C with 5% CO2.
Stable transfection of shS1PR2 in cell lines. HCT116, SW620, Huh7 and cccHEL-1
cells were seeded in the 6-well plates containing 2 mL medium with cell density of
40-60% in each well. Remove media from wells and then add 2 mL media containing
polybrene (final concentration 5 µg/ml) to each well. Gently swirl the plate to mix and
remove the media containing lentiviral particles from wells. Fresh culture was
substituted after 48 h and then 500 μg/ml of puromycin was added to the culture.
After 7 to 10 days, the growth of puromycin-resistant cells was observed and selected
clones were further grown in the medium containing 200 μg/ml of puromycin for 4
weeks. Puromycin-resistant clones named shS1PR2 cells were further propagated in
the medium containing 10% FBS in the absence of puromycin and determined for
S1PR2 expression by Western blotting assay. The transfection efficiency was
affirmed by fluorescence microscopy. The Scramble_shRNA of the lentivirus is LV-
U6>Scramble-shRNA-PGK>EGFP/Puro. The control sequence is CCTAAGGTTA
AGTCGCCCTCG and the target sequence of S1PR2 is AGGAACAGCAA
GTTCTACTCA.
17

TCGA dataset. RNA-seq data for individual TCGA cancer types were processed
with a modified version of CrossHub, a tool for the multi-way analysis of TCGA
transcriptomic and genomic data. Then they were normalized by TMM (the trimmed
mean of M values) method and recalculated for 1 million library size. The derived
TPM (transcripts per million) values were used to measure mRNA level of a gene for
further analysis of expression stability. Gene expression values were transformed as
X′ = log2(X + 1), where X represents the normalized fragments per kilobase transcript
per million mapped reads values.
Subcutaneous tumor implantation in nude mice. Nude mice were caged under
controlled temperature, humidity and light, fed with standard mouse chow. Under
standard sterile conditions in the operating room, the xenografts of HCT116DPD were
established by inoculating 5 × 106 cells subcutaneously in mice. When the tumor
reached a volume of 80-120 mm3, the mice were divided into four groups (n = 6) and
administered by tail vein injection for consecutive 24 days. 5-FU, compounds 10c and
JTE-013 were injected at doses of 20 mg/kg, 1 mg/kg and 1 mg/kg, respectively
(dissolved in 30% PEG300, 5%Tween 80, 2% DMSO). Tumor volumes were
assessed by bilateral Vernier caliper measurement every 4 days and calculated
according to the following equation: [tumor volume = X × (Y2/2)], where X
represents the longer and Y represents the shorter of the two dimensions. Body weight
was also measured every 4 days, and clinical symptoms were observed daily. The
18

animals were sacrificed on day 24, and the tumors were removed and frozen in liquid
nitrogen or fixed in 10% neutral-buffered formalin. The assays of western blotting,
immunohistochemical and immunofluorescent staining were performed to test the
proteins related to S1PR2 and DPD. All experiments in mice were approved by
Animal Welfare Committee of Capital Medical University (permit no. AEEI-2016-
043), in accordance with the animal care and use guidelines.
Quantitative RT-PCR. Total RNA was isolated by using Rneasy Mini Kit (74106,
Qiagen). For cDNA synthesis, total RNA was transcribed by using PrimeScript
(DRR047A, Takara, China). The levels of specific RNAs were measured by using
ABI 7900 real-time PCR machine and Fast SybrGreen PCR mastermix according to
the manufacturer’s instructions. All samples, including the template controls, were
assayed in triplicate. The relative number of target transcripts was normalized to
GAPDH expression in the same sample. The relative quantification of target gene
expression was performed with the standard curve or comparative cycle threshold
(CT) method. Fold-induction was calculated by using the CT method as follows:
ΔΔCT = (CT target gene - CT housekeeping gene) treatment - (CT target gene - CT housekeeping gene)
nontreatment. Forward (5'-3') primer of dpyd is CCATCGCCATCGAGAGACAAG, and
reward (5'-3') primer of dpyd is CACGTAGTGCTTAGCATAGAGAG. Forward (5'-
3') primer of Gapdh is ACTCCAAGGCCACTTATCACC, and reward (5'-3') primer
of Gapdh is ATTGTTACCAACTGGGACGAA.
19

Immunohistochemistry. Following de-paraffinization in xylene and rehydration, the
slides were subject to high pH antigen retrieval (10 mmol/L citrate buffer; pH 9.0),
followed by 3% hydrogen peroxide, and blocking in 1.5% serum. Slides were then
incubated with the selected antibody at 4°C overnight. The information of primary
antibodies is presented in Table S1. The slides were developed with an EnVisionTM
method (DAKO, Capinteria, CA), visualized using diaminobenzidine solution, and
then lightly counterstained with hematoxylin (H9627, Sigma). The IHC imaging was
scanned by KF-PRO-OO5 slide viewer. Additionally, some slides were incubated
with IgG1 isotype controls.
Western blotting analysis. Proteins were extracted by using RIPA Lysis Buffer
(P0013, Beyotime, China) and quantified by using a BCA kit (P0009, Beyotime,
China). Twenty micrograms of each protein sample were separated by 8, 10 or 15%
SDS-PAGE and transferred to a polyvinylidene difluoride membrane. The membranes
were blocked with 5% BSA and incubated with primary antibodies for 10 h at 4°C.
The membranes were rinsed five times with PBS containing 0.1% Tween 20 and
incubated for 1 h with appropriate horseradish peroxidase-conjugated secondary
antibody at 37°C. Membranes were extensively washed with PBS containing 0.1%
Tween 20 for three times. The signals were stimulated with enhanced
chemiluminescence substrate (NEL105001 EA, PerkinElmer) for 1 min and detected
20

with a Bio-Rad ChemiDoc MP System (170-8280). The information of primary
antibodies is presented in Table S1. The primary images (Fig. S4) were cropped for
presentation.
MTT assay. Cell viability was assessed employing in vitro Toxicology Assay Kit-
MTT Based, Stock No.TOX-1, Sigma. The key component is 3-[4,5-dimethylthiazol-
2-yl]-2,5-diphenyl tetrazolium bromide (MTT). Cells paused at the logarithmic
growth phase were inoculated into the 96-well plates at a concentration of 3×103
cells/ml. The cells of control group were incubated with 20 μM 5-FU and tested
compounds. The final concentrations of tested compounds were 0, 2.5, 5, 10, 20, 40,
and 80 μM, respectively. Each cell mass with final concentration were set into three
holes. Each hole was routinely cultured for 48 h, and then added into 5 g/L×20 μL
MTT solution to incubate for 4 h. The supernatant in each hole was abandoned, and
each hole was added with 150 μL DMSO. After agitation for 5-10 min, the plates
were sent to enzyme-labelled meter, and the A490 was detected at the 490 nm wave.
The growth inhibition rate of each group was calculated as follows: Inhibition rate =
1- (ODDrug - ODBlank) / (ODControl - ODBlank). The values of EC50 (tested compound
plus 20 μM 5-FU) and IC50 (tested compound only) were determined according to
their growth inhibition rates.
21

Immunofluorescent assay. CRC cells (2.5×104) were seeded onto the 12-mm glass
coverslips in 6-well plates. Cells were fixed in 4.5% paraformaldehyde for 15 min,
washed in PBS, permeabilized with 1×Perm/Wash Buffer (BD Biosciences) for 10
min, washed in PBS, blocked for nonspecific antibody reactions by incubating in the
solution containing 5% BSA for 30 min, and incubated with anti-Calnexin or anti-
S1PR2 antibodies. Secondary antibodies included donkey anti-rabbit IgG (Alexa
Fluor® 488, Invitrogen) and labeled goat anti-mouse IgG (Alexa Fluor® 648, CST).
Coverslips were mounted with Vectashield® antifade mounting medium containing
DAPI (Sigma-Aldrich). Confocal laser microscopy was performed on a Leica TCS
SP5 AOBS apparatus, using excitation spectral laser lines at 488 nm. Image
acquisition and processing were conducted using Leica Confocal Software (Leica
Microsystems).
HPLC analysis of intracellular 5-FU and FBAL. 5-FU and FBAL ranged from 0.2
to 50 mg/L for preparation of the standard curves. Cells were inoculated in 6-well
plate and then treated with 5-FU (25 mg/L) for 6 h. After cell counting, cells were
resuspended in 150 μl PBS and lysed by sonication in ice-water bath for 30 min. The
procedure of solid-phase extraction was utilized with styrene-divinylbenzene resin
SPE columns as shown before [30]. Then the samples were centrifuged at 10,000 g
for 15 min. The fresh supernatants were filtered and injected into HPLC (Waters
22

QDA) for analysis. The concentrations of 5-FU and FBAL were calculated from the
respective calibration curves.
Molecular Docking. The structures of compounds 10c and JTE-013 were built using
ChemDraw14.0, followed by MM2 energy minimization. The molecular docking
study was performed using LeDock Tools in combination with PyMOL software. The
homology model of S1PR2 based on the high-resolution crystal structure of the
S1PR1 (PDB: 3V2Y) was used for the docking studies performed here. The
interaction of protein and ligands in the binding pocket was defined in LeDock. Each
docking experiment was performed 20 times, yielding 20 docked conformations. All
of other parameters used in the docking process were the default values of the system.
The best binding modes were picked based on the best stabilization energy.
Statistical analysis
Statistical analysis was conducted by GraphPad Prism 5. After checking data for
normal distribution and variance homogeneity, continuous data were compared using
the Mann-Whitney U test, multiple Student t tests or two-way ANOVA. All p values
are two-tailed, and p values < 0.05 were considered significant (*p < 0.05, **p < 0.01
and ***p < 0.001). The data are represented as mean ± S.E.M. or the median with 10
and 90 percentiles. All statistical tests were two-sided.
23

Author’s contribution
Xian-Jun Qu and Sheng-Biao Wan conceived and designed the study. Yu-Hang
Zhang performed animals and cellular studies experiments. Dong-Dong Luo designed,
synthesized and analyzed all of the compounds. Yu-Hang Zhang and Dong-Dong Luo
wrote the manuscript, which was edited by all authors.
Conflict of interest: The authors declare no potential conflicts of interest.
Acknowledgements: This work was supported by National Natural Science
Foundation of China (91629303/81673449/81872884/81973350) and Beijing Natural
Science Foundation and Scientific Research Program of Municipal Commission of
Education (KZ201710025020/KZ201810025033).
24

References:
[1] B. Gustavsson, G. Carlsson, D. Machover, N. Petrelli, A. Roth, H.J. Schmoll,
K.M. Tveit, F. Gibson, A review of the evolution of systemic chemotherapy in the
management of colorectal cancer, Clin Colorectal Cancer 14 (2015) 1-10.
[2] J.A. Meyerhardt, R.J. Mayer, Systemic therapy for colorectal cancer, N Engl J
Med 352 (2005) 476-487.
[3] P. Noordhuis, U. Holwerda, C.L. Van der Wilt, C.J. Van Groeningen, K. Smid, S.
Meijer, H.M. Pinedo, G.J. Peters, 5-Fluorouracil incorporation into RNA and DNA in
relation to thymidylate synthase inhibition of human colorectal cancers, Ann Oncol 15
(2004) 1025-1032.
[4] D.B. Longley, D.P. Harkin, P.G. Johnston, 5-fluorouracil: mechanisms of action
and clinical strategies, Nat Rev Cancer 3 (2003) 330-338.
[5] F. Innocenti, R. Danesi, G. Bocci, G. Natale, T.M. Del, 5-Fluorouracil catabolism
to 5-fluoro-5,6-dihydrouracil is reduced by acute liver impairment in mice, Toxicol
Appl Pharmacol 203 (2005) 106-113.
[6] R.B. Diasio, B.E. Harris, Clinical pharmacology of 5-fluorouracil, Clin
Pharmacokinet 16 (1989) 215-237.
[7] M. Kornmann, W. Schwabe, S. Sander, M. Kron, J. Strater, S. Polat, E. Kettner,
H.F. Weiser, W. Baumann, H. Schramm, P. Hausler, K. Ott, D. Behnke, L. Staib,
H.G. Beger, K.H. Link, Thymidylate synthase and dihydropyrimidine dehydrogenase
25

mRNA expression levels: predictors for survival in colorectal cancer patients
receiving adjuvant 5-fluorouracil, Clin Cancer Res 9 (2003) 4116-4124.
[8] A.M. Lee, Q. Shi, S.R. Alberts, D.J. Sargent, F.A. Sinicrope, J.L. Berenberg, A.
Grothey, B. Polite, E. Chan, S. Gill, M.S. Kahlenberg, S.G. Nair, A.F. Shields, R.M.
Goldberg, R.B. Diasio, Association between DPYD c.1129-5923 C&gt;G/hapB3 and
severe toxicity to 5-fluorouracil-based chemotherapy in stage III colon cancer
patients: NCCTG N0147 (Alliance), Pharmacogenet Genomics 26 (2016) 133-137.
[9] D. Rosmarin, C. Palles, A. Pagnamenta, K. Kaur, G. Pita, M. Martin, E.
Domingo, A. Jones, K. Howarth, L. Freeman-Mills, E. Johnstone, H. Wang, S. Love,
C. Scudder, P. Julier, C. Fernandez-Rozadilla, C. Ruiz-Ponte, A. Carracedo, S.
Castellvi-Bel, A. Castells, A. Gonzalez-Neira, J. Taylor, R. Kerr, D. Kerr, I.
Tomlinson, A candidate gene study of capecitabine-related toxicity in colorectal
cancer identifies new toxicity variants at DPYD and a putative role for ENOSF1
rather than TYMS, Gut 64 (2015) 111-120.
[10]K.E. Caudle, C.F. Thorn, T.E. Klein, J.J. Swen, H.L. McLeod, R.B. Diasio, M.
Schwab, Clinical Pharmacogenetics Implementation Consortium guidelines for
dihydropyrimidine dehydrogenase genotype and fluoropyrimidine dosing, Clin
Pharmacol Ther 94 (2013) 640-645.
[11]W.H. Gmeiner, Novel chemical strategies for thymidylate synthase inhibition,
Curr Med Chem 12 (2005) 191-202.
26

[12]G.J. Peters, C.L. van der Wilt, C.J. van Groeningen, K. Smid, S. Meijer, H.M.
Pinedo, Thymidylate synthase inhibition after administration of fluorouracil with or
without leucovorin in colon cancer patients: implications for treatment with
fluorouracil, J Clin Oncol 12 (1994) 2035-2042.
[13]K. Tatsumi, M. Fukushima, T. Shirasaka, S. Fujii, Inhibitory effects of
pyrimidine, barbituric acid and pyridine derivatives on 5-fluorouracil degradation in
rat liver extracts, Jpn J Cancer Res 78 (1987) 748-755.
[14]K. Hirata, N. Horikoshi, K. Aiba, M. Okazaki, R. Denno, K. Sasaki, Y. Nakano,
H. Ishizuka, Y. Yamada, S. Uno, T. Taguchi, T. Shirasaka, Pharmacokinetic study of
S-1, a novel oral fluorouracil antitumor drug, Clin Cancer Res 5 (1999) 2000-2005.
[15]M. Nakamura, H. Nagano, S. Marubashi, A. Miyamoto, Y. Takeda, S. Kobayashi,
H. Wada, T. Noda, K. Dono, K. Umeshita, M. Monden, Pilot study of combination
chemotherapy of S-1, a novel oral DPD inhibitor, and interferon-alpha for advanced
hepatocellular carcinoma with extrahepatic metastasis, Cancer-Am Cancer Soc 112
(2008) 1765-1771.
[16]R. Wu, Q. Nie, E.E. Tapper, C.R. Jerde, G.S. Dunlap, S. Shrestha, T.A. Elraiyah,
S.M. Offer, R.B. Diasio, Histone H3K27 Trimethylation Modulates 5-Fluorouracil
Resistance by Inhibiting PU.1 Binding to the DPYD Promoter, Cancer Res 76 (2016)
6362-6373.
27

[17]A. Vestri, F. Pierucci, A. Frati, L. Monaco, E. Meacci, Sphingosine 1-Phosphate
Receptors: Do They Have a Therapeutic Potential in Cardiac Fibrosis? Front
Pharmacol 8 (2017) 296.
[18]L. Kappos, J. Antel, G. Comi, X. Montalban, P. O'Connor, C.H. Polman, T. Haas,
A.A. Korn, G. Karlsson, E.W. Radue, Oral fingolimod (FTY720) for relapsing
multiple sclerosis, N Engl J Med 355 (2006) 1124-1140.
[19]J.A. Cohen, F. Barkhof, G. Comi, H.P. Hartung, B.O. Khatri, X. Montalban, J.
Pelletier, R. Capra, P. Gallo, G. Izquierdo, K. Tiel-Wilck, A. de Vera, J. Jin, T. Stites,
S. Wu, S. Aradhye, L. Kappos, Oral fingolimod or intramuscular interferon for
relapsing multiple sclerosis, N Engl J Med 362 (2010) 402-415.
[20]S.P. Cramer, H.J. Simonsen, A. Varatharaj, I. Galea, J.L. Frederiksen, H. Larsson,
Permeability of the blood-brain barrier predicts no evidence of disease activity at 2
years after natalizumab or fingolimod treatment in relapsing-remitting multiple
sclerosis, Ann Neurol 83 (2018) 902-914.
[21]N.C. Hait, L.E. Wise, J.C. Allegood, M. O'Brien, D. Avni, T.M. Reeves, P.E.
Knapp, J. Lu, C. Luo, M.F. Miles, S. Milstien, A.H. Lichtman, S. Spiegel, Active,
phosphorylated fingolimod inhibits histone deacetylases and facilitates fear extinction
memory, Nat Neurosci 17 (2014) 971-980.
[22]B. Zemann, B. Kinzel, M. Muller, R. Reuschel, D. Mechtcheriakova, N. Urtz, F.
Bornancin, T. Baumruker, A. Billich, Sphingosine kinase type 2 is essential for
28

lymphopenia induced by the immunomodulatory drug FTY720, Blood 107 (2006)
1454-1458.
[23]J.A. Cohen, J. Chun, Mechanisms of fingolimod's efficacy and adverse effects in
multiple sclerosis, Ann Neurol 69 (2011) 759-777.
[24]V. Rothhammer, J.E. Kenison, E. Tjon, M.C. Takenaka, K.A. de Lima, D.M.
Borucki, C.C. Chao, A. Wilz, M. Blain, L. Healy, J. Antel, F.J. Quintana, Sphingosine
1-phosphate receptor modulation suppresses pathogenic astrocyte activation and
chronic progressive CNS inflammation, Proc Natl Acad Sci U S A 114 (2017) 2012-
2017.
[25]L. Zhang, H.D. Wang, X.J. Ji, Z.X. Cong, J.H. Zhu, Y. Zhou, FTY720 for cancer
therapy (Review), Oncol Rep 30 (2013) 2571-2578.
[26]L. Pouzol, L. Piali, C.C. Bernard, M.M. Martinic, B. Steiner, M. Clozel,
Therapeutic Potential of Ponesimod Alone and in Combination with Dimethyl
Fumarate in Experimental Models of Multiple Sclerosis, Innov Clin Neurosci 16
(2019) 22-30.
[27]L. Peyrin-Biroulet, R. Christopher, D. Behan, C. Lassen, Modulation of
sphingosine-1-phosphate in inflammatory bowel disease, Autoimmun Rev 16 (2017)
495-503.
[28]M.W. Saif, L.S. Rosen, K. Saito, C. Zergebel, L. Ravage-Mass, D.S. Mendelson,
A phase I study evaluating the effect of CDHP as a component of S-1 on the
pharmacokinetics of 5-fluorouracil, Anticancer Res 31 (2011) 625-632.
29