Fresh colon cancer (n=55) and matched non-tumourous colon tissue (NT) (n=55) were collected following surgical resection at the First Affiliated Hospital of Nanjing Medical University (Nanjing, China) from 2010 to 2015. The NT tissue was obtained at the resection margin, located at a distance of >5 cm away from the primary tumour. The mean age at diagnosis was 64.00±11.91 years. Plasma was collected from patients with colon cancer (n=55) and sex- and age-matched healthy donors (n=55; mean age, 74.51±8.42 years; 25 male and 20 female). Tissues were washed in ice-cold PBS and a portion of tissues were formalin-fixed and paraffin-embedded. The rest of the tissue and plasma samples were stored at −40°C. All patients had a histological diagnosis of colon cancer and none had received therapy. All specimens were obtained with verbal consent from the patients and healthy donors, and the study was approved by the Institutional Ethics Committee of Nanjing Medical University (approval no. 2016-SR-217).

As 5 of the specimens were not large enough, 50 cases were included in the IHC assay. The tissues were fixed with 4% paraformaldehyde (PFA) at room temperature for 20 min and embedded into paraffin. Tissue paraffin blocks were serially sectioned at 4 µm. Sections were deparaffinized at 60°C for 30 min and washed in xylene, and rehydrated in a descending alcohol series. Tissue sections were placed in a repair box filled with EDTA antigen repair buffer (pH 9.0) (Wuhan Goodbio Technology, Co., Ltd.; cat. no. G12036), and antigen retrieval was performed by microwaving for 15 min, endogenous peroxidase was blocked at room temperature with 3% hydrogen peroxide in methanol for 25 min and tissue was blocked with normal rabbit serum (Gibco; Thermo Fisher Scientific, Inc.; cat. no. 16120107; 1:10) for 30 min at room temperature. CRC tissues were incubated with anti-S1PR2 primary antibody (Santa Cruz Biotechnology, Inc.; cat. no. sc-365589; 1:500) overnight at 4°C. Sections were washed with PBS and incubated with an HRP-conjugated secondary antibody (Invitrogen; Thermo Fisher Scientific, Inc.; cat. no. A16160; 1:1,000) at room temperature for 50 min. After being washed with PBS, the target protein was stained with 1% 3,3′-diaminobenzidine(DAKO; Shanghai Li Min Industrial Co., Ltd.; cat. no. K5007) at room temperature for 5 min, and cell nuclei were counterstained blue with haematoxylin at room temperature for 3 min. The slides were scanned using the Pannoramic Midi II histological scanner 1.18.1 (3DHISTECH Ltd.). Next, using the Pannoramic Viewer Program 1.15.3 (3DHISTECH Ltd.), representative areas were selected. A light microscope (IX73; Olympus) was used for imaging at ×100 and ×200 magnification. The staining score for tumour and adjacent normal tissue (0, negative; 1+, weak; 2+, moderate and 3+, strong) was recorded separately. The estimated proportion of positive tumour cells was calculated as a percentage. To assess the mean degree of staining within a tumour, three representative regions were analysed and ≥100 tumour cells were assessed. S1PR2 expression was assessed by histochemistry (H)-score system as follows: H-score=∑(I xPi), where I=intensity of staining and Pi=percentage of stained tumour cells, producing a score ranging from 0 to 300. The scoring was independently assessed by two investigators who were blinded to the patient characteristics and outcomes.

Total RNA was extracted from tissue with an RNA Extraction kit (Takara Biotechnology Co., Ltd.; cat. no. 9767) according to the manufacturer's protocol. Total RNA was quantified using a spectrophotometer (Nanodrop ND-1000; Thermo Fisher Scientific, Inc.). Only RNA with a 260/280 ratio of 1.8-2.0 was used for cDNA synthesis and reverse transcribed into cDNA using PrimeScript^™ RT Master Mix (Takara Biotechnology Co., Ltd.; cat. no. RR036A) at 37°C for 15 min. mRNA levels of target genes were detected by qPCR using SYBR premix Ex Taq^™ (Takara Biotechnology Co., Ltd.; cat. no. 638319) on a StepOnePlus^™ Real-Time PCR System (Applied Biosystems; Thermo Fisher Scientific, Inc.). The following thermocycling conditions were used for qPCR: Initial denaturation at 95°C for 30 sec, followed by 40 cycles of denaturation at 95°C for 5 sec, annealing at 60°C for 30 sec and extension at 60°C for 1 min. Following normalization using β-actin as an internal control, the data were analysed using the ΔΔCq method and expressed as target gene/internal ratio [2^{−ΔCq(target gene-internal)}] (28). The primers were as follows: S1PR1 forward, 5′-TCCTCGCCATCGCCATTG-3′ and reverse, 5′-GAGAGCAGAAGCAGAGTGAAG-3′; S1PR2 forward, 5′-CAAGGTCCAGGAACACTATA-3′and reverse, 5′-AACAGAGGATGACGATGAA-3′; S1PR3 forward, 5′-CTACGCACGCATCTACTTCC-3′ and reverse, 5′-CACGCTCACCACAATCACC-3′; S1PR4 forward, 5′-GCTGAAGACGGTGCTGATG-3′ and reverse, 5′-CTGCTGCGGAAGGAGTAG-3′; S1PR5 forward, 5′-CTTCCTGCTGCTGTTGCTC-3′ and reverse, 5′-GCCACTCGGGTCTCTGC-3′; SPHK1 forward, 5′-TGTGTAGCCTCCCAGCAG-3′ and reverse, 5′-CCCAGACGCCGATACTTC-3′; SPHK2 forward, 5′-ACTGCCCTCACCTGTCTG-3′ and reverse, 5′-TTCTGTCGTTCTGTCTGGATG-3′ and β-actin forward, 5′-TGACGTGGACATCCGCAAAG-3′ and reverse, 5′-CTGGAAGGTGGACAGCGAGG'.

Quantification of S1P from serum was performed by ELISA using a Human S1P Assay kit (Nanjing SenBeiJia Biological Technology Co., Ltd.; cat. no. SBJ-H2060-96T) according to the manufacturer's instructions.

CRC SW480 and LOVO cell lines, purchased from Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, were maintained in DMEM supplemented with 10% foetal bovine serum (Invitrogen; Thermo Fisher Scientific, Inc.), 100 U/ml penicillin G and 100 mg/ml streptomycin (Sigma-Aldrich; Merck KGaA). All cells were cultured at 37°C with 5% CO₂. Cells were grown in 100-mm dishes and incubated at 37°C in DMEM for 24 h. The small interfering RNA (siRNA) oligonucleotide duplexes (si-S1PR2: 5′-CCUUCGUAGCCAAUACCUUTT-3′; negative control, 5′-UUCUCCGAACGUGUCACGUTTACGUGACACGUUCGGAGAATT-3′; Shanghai GenePharma Co., Ltd.) were transfected into cells using Lipofectamine^® 2000 (cat. no. 11668019; Invitrogen; Thermo Fisher Scientific, Inc.). S1PR2 siRNA (120 pmol) and transfection reagent (6 µl) were mixed in 300 µl serum-free DMEM, left to stand for 5 min and then mixed again. Following incubation at room temperature for 20 min, the mix was added to serum-starved cells and incubated at 37°C for 4 h. CRC cells were plated the day before transfection at 400,000 cells/well in 6-well plates in complete growth medium (Gibco; Thermo Fisher Scientific, Inc.). S1P, JTE013 and FTY720 were purchased from Cayman Chemical Company and dissolved in dimethyl sulfoxide (DMSO; Sigma-Aldrich; Merck KGaA). Cells were pre-treated at 37°C with S1PR2 antagonist JTE013 (5 µM), S1PR1,3,4,5 antagonist FTY720 (5 µM) or DMSO (1%) for 1 h before stimulation in the presence or absence of S1P (1 µM) at 37°C for 1 h. DMSO was applied as vehicle. Cells were harvested 24 h later for migration and invasion assay.

Human CRC SW480 cells (5×10³ cells/well) were seeded into 96-well plates, allowed to adhere and serum-starved overnight at 37°C in 95% air and 5% CO₂. Next, cells were transfected with S1PR2-specific siRNA and treated with S1P (1 µM) or DMSO. Following 24 h incubation at 37°C, cells were stained with sterile MTT for 4 h at 37°C, the supernatant was collected by centrifugation at 13,225 × g for 2 min at 4°C and removed, and the crystals formed were dissolved using DMSO. The absorbance at 490 nm was examined.

SW480 or LOVO cells were plated at a density of 5×10⁵ cells/well in 6-well culture dishes and allowed to form a monolayer (70–90% confluence). Following serum-starvation for 24 h, the cells were scratched with a sterile pipette tip (200-µl), washed with PBS to remove floating and detached cells and photographed (time, 0 h) under the 10× objective of an Olympus 1X71 light microscope (Olympus Corporation). Cells were pre-treated with JTE-013 (5 µM), FTY720 (5 µM) or DMSO for 1 h and treated with S1P (1 µM) or DMSO. After 24 h, the wound area was. The assay was performed three times. ImageJ v1.8.0 software (National Institutes of Health) was used to analyse the cell-free areas in images. A total of 6–8 horizontal lines were drawn on each image and the width was calculated.

After being starved overnight, SW480 cells (6×10³ cells/well) were seeded in a Matrigel-coated upper polycarbonate chamber (8-µm pore size; BD Biocoat Matrigel Invasion Chamber). The chambers were precoated with 20% Matrigel for 30 min at 37°C. The upper chamber was filled with serum-free DMEM and the lower chamber was filled with medium containing 10% FBS (Gibco-BRL, Rockville, MD, USA). Cells were pre-treated with JTE-013 (5 µM), FTY720 or vehicle DMSO for 1 h, then treated with S1P (1 µM) or DMSO and incubated at 37°C in 95% air and 5% CO₂. After 24 h, non-invasive cells were washed off twice with PBS. The cells that had migrated to the lower surface were fixed for 30 min with 3.7% paraformaldehyde at 25°C and stained for 20 min using 0.1% crystal violet solution at 25°C, and then observed and images captured at ×200 magnification on a light microscope. For each replicate (n=3), migrated cells were manually counted and expressed as the average cell number from 4 random fields. The cell number indicated the cell migration capability and invasiveness.

Tissues samples were harvested and lysed in ice-cold lysis buffer (1% Nonidet P-40, 0.5% sodium cholate, 1 mM EDTA, 1 mM EGTA, 150 mM NaCl, 20 mM HEPES, 3 mM MgCl, 1 mM PMSF, 20 mM b-glycerophosphate, 1 mM NaF and 1 mM sodium orthovanadate; pH 7.4) and sonicated (20 kHz; 0°C; 30 sec). Following centrifugation at 13,225 × g for 20 min at 4°C, protein concentrations were assayed using Bradford protein assay reagent (Bio-Rad Laboratories, Inc.) with BSA as a standard. Total protein (30 µg) was separated on a 10% SDS-PAGE gel and transferred to a PVDF membrane (Millipore). The membranes were blocked with 5% non-fat milk in TBST [10 mM Tris (pH 7.4), 100 mM NaCl and 0.5% Tween 20] for 1 h at room temperature and incubated with the primary antibody S1PR2 (Santa Cruz Biotechnology, Inc.; cat. no. sc-365589; 1:1,000). Protein bands were normalized to GAPDH and detected with anti-GAPDH (Santa Cruz Biotechnology, Inc.; cat. no. sc-47724; 1:1,000). Immunoreactive bands were detected using rabbit anti-mouse IgG HRP-conjugated secondary antibody (Santa Cruz Biotechnology, Inc.; cat. no. sc-358914; 1:5,000) and SuperSignal Chemiluminescent Substrate (Thermo Fisher Scientific, Inc.). The densities of bands were analysed using Image Lab software.

Experiments were repeated ≥3 times and data are presented as the mean and standard error of the mean (SEM). Statistical analysis for multiple comparisons was performed by one-way ANOVA followed by post hoc Bonferroni's correction. The differences in levels of S1P between the CRC group and healthy donor group were analysed using an unpaired Student's t-test. S1PR2 expression in tumour and matched normal tissues were compared using Wilcoxon's signed-rank test. P<0.05 was considered to indicate a statistically significant difference. All statistical analysis was performed using GraphPad Prism 8.0 (GraphPad Software, Inc.) software.

To test whether S1PR2 serves a role in human CRC, IHC analysis of S1PR2 in human colon tissue was performed. As 5 of the specimens were not large enough, 50 cases were included in the IHC assay. CRC specimens (n=50) had notably lower levels of S1PR2 than normal human colon tissue (n=50), which exhibited strong S1PR2-positive staining in colon cells (Fig. 1A). IHC staining was performed to evaluate expression of S1PR2 in tumour and adjacent normal tissue of patients with CRC using H-score as a continuous variable. Mean H-score was higher in normal compared with tumour tissue (33.11 vs. 12.23; P<0.001). S1PR2 expression in human CRC tissue was significantly lower in comparison with that in paired normal colon tissue (n=50; Fig. 1B). Next, expression levels of S1PR2 mRNA in tumour and adjacent non-tumourous tissue of patients with CRC were assessed using RT-qPCR (n=55). S1PR2 was significantly downregulated in colon cancer compared with non-tumourous tissue, with a log2 expression ratio of −3.46 (tumour vs. normal; Fig. 1C). The protein level of S1PR2 in colon tissue was determined using western blotting (Fig. 1D). Among patients with CRC (n=55), downregulated S1PR2 was observed in 32 (58%) cases; however; this was not significantly different (P=0.45; Wilcoxon's signed-rank test). The mean ratio of S1PR2/GAPDH was slightly higher in tumour compared with normal tissue (1.15 vs. 0.96) but this was not significantly different. Taken together, these data showed that S1PR2 was downregulated in CRC cells and low S1PR2 expression may serve a role in the development of colon cancer.

To confirm the role of the S1P axis in colon cancer, mRNA expression levels of S1PR1, S1PR3, S1PR4 and S1PR5 were measured in the tumour and adjacent non-tumourous tissue of patients with CRC using RT-qPCR (Fig. 2A-D). S1PR1 and S1PR3 were highly expressed S1PR subtypes in human colon tissue, whereas S1P4 and S1P5 were barely detectable. Transcript levels for S1PR1 gene were significantly higher in tumour than in normal tissue, while those of S1PR4 gene were lower (P<0.05). Transcripts of S1PR3 gene were higher in tumour tissue but there was no significant difference (P=0.0748). The results were consistent with previous studies that suggested that S1PR1 and S1PR3 are upregulated in numerous types of cancer (27). mRNA expression levels of SPHK1 and SPHK2 in colon tissue were measured (Fig. 2E and F). SPHK1 gene expression was significantly upregulated in tumour tissue, which is consistent with previous data (29). These results demonstrated variation in expression of S1PR and sphingolipid metabolism enzymes in colon cancer.

Activation of SPHK1 leads to synthesis and secretion of S1P, which has been shown to influence cellular function via interactions with specific receptors, including S1PR2 (30). Expression of S1P in the peripheral blood of patients with CRC may be altered via the SPHK/S1P/S1PR signalling pathway. Therefore, S1P levels were measured in serum from patients with CRC (n=55) and healthy individuals (n=55) using ELISA. S1P level were higher in patients with CRC (Fig. 3).

S1P has been shown to regulate cellular physiological functions, including proliferation and migration, in different types of cancer (4). To determine the effect of exogenous S1P on the colon cancer cell lines SW480 and LOVO in vitro, wound healing assay was performed to investigate the role of S1P in colon cancer cell migration. LOVO cells treated with 1 µM S1P showed a significant increase in migration distance, whereas that of SW480 cells treated with 1 µM S1P was significantly inhibited (Fig. 4A). JTE013, a specific S1PR2 antagonist, further promoted migration of SW480 and LOVO cells stimulated by S1P (Fig. 4A and C). To confirm the role of S1PR2 in cell migration, siRNA was used to specifically silence S1PR2 expression. Relative to the vehicle control, knockdown of S1PR2 by siRNA significantly promoted migration of SW480 and LOVO cells, which is consistent with the aforementioned results (Fig. 4B and D). Inhibition of S1PR2 by JTE013 or S1PR2-siRNA significantly promoted migration of SW480 and LOVO cells. These data indicated that expression of S1PR2 may inhibit the migration of colon cancer cells. Expression levels of S1PR in SW480 and LOVO cells were determined. Transcripts for S1PR genes were expressed at varying levels in colon cancer cells (Fig. 5). The mRNA expression levels of S1PR2 and S1PR3 were high in SW480 cells (Fig. 5A), whereas S1PR2 was expressed only at low levels in LOVO cells (Fig. 5B). Therefore, the effect of S1P on migration of different colon cancer cell lines may depend on S1PR subtypes expressed on the cell surface.

To determine the role of S1PR2 in the migration and invasion of SW480 cells, wound healing and Transwell assays were performed using SW480 cells pre-treated in the presence or absence of JTE-013 and FTY720 for 1 h before S1P stimulation. There was no significant difference between the S1P and the control group with regard to the wound healing of SW480 cells (Fig. 6A). Treatment with 5 µM JTE-013 increased SW480 cell migration but this was not significant (Fig. 6A and C). Quantification of cell migration width confirmed a significant increase in cell migration following treatment with JTE013 + S1P compared with JTE013-alone (Fig. 6A; P<0.05). Co-treatment with S1P and 5 µM FTY720, an inhibitor of other S1PRs (S1PR1, 3, 4 and 5), significantly attenuated migration compared with the JTE013 + S1P group (Fig. 6A; P<0.01). S1P significantly enhanced invasion of SW480 cells in Transwell assay compared with the control group (Fig. 6B; P<0.01). Pre-treatment with JTE013 further promoted SW480 cell invasion, and invasion was higher in the FTY720 + S1P group than the Con + S1P group (Fig. 6B; P<0.01). These results suggested that S1PR2 inhibited migration of cancer. Following transfection with S1PR2-specific siRNA, SW480 cell proliferation was significantly increased (Fig. 6E; P<0.05 and P<0.001, Con vs. si-S1PR and Con + S1P vs. si-S1PR + S1P, respectively), while there was no significant difference in SW480 cell viability following S1P stimulation.