For the analysis of mRNA and protein levels of PTPRJ, 8-paired human cervical tumor and non-tumor samples were obtained from patients with stage I–III adenocarcinoma and immediately frozen in liquid nitrogen and stored at the Department of Pathology, Gansu Provincial Hospital in Lanzhou, China. Immunochemistry methods were used to characterize these tissues.

Human cervical carcinoma cell line, C33A, was purchased from the American Type Culture Collection (Rockville, MD, USA) and maintained in RPMI-1640 (Invitrogen Corp., Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (Hyclone, South Logan, UT, USA), 100 U/ml penicillin and 100 mg/ml streptomycin at 37°C in a humidified incubator with 5% CO₂. 5-FU was obtained from Sigma Chemical.

PTPRJ cDNA was PCR amplified from the whole genome of HEK293T using primers for PTPRJ: 5′-ATGAAGCCGGCGGCGCGGGA-3′ (sense) and 5′-TTAGGCGATGTAACCATTGG-3′ (antisense). The cDNAs were ligated into pSicoR after being digested by EcoRI and XhoI to create pSicoR-PTPRJ. RNA interference vectors used in the present study were constructed in pLKO.1 puro. Target sequences are as follows: NTC (AATTCTCCGAACGTGTCACGT), shPTPRJ-1 (ACGAGTCGTCATCTAACTATA), shPTPRJ-2 (CCGATACAATGCCACCGTTTA).

The indicated constructs were transfected into the HEK293T cells in combination with the lentiviral packaging vectors pRSV-Rev, pMD2.G, and pCMV-VSV-G using Lipofectamine 2000 (Invitrogen). After 60 h, the media containing the viruses were collected using a 0.22-μm filter (Millipore, Bedford, MA, USA). Viral supernatant from a 6-well plate was used to infect ~10³ C33A cells for 48 h. After a 24-h infection, the C33A cells were selected for 48 h in 2.5 μg/ml puromycin and two colonies were chosen to expand and detect the protein levels. For 5-FU treatment, the indicated stable cell lines were plated in 6-well plates and cultured for 24 h. Then the culture media were changed with fresh DMEM containing 20 μg/ml 5-FU or DMSO.

The cells were washed with PBS and lysed in ice-cold lysis buffer consisting of 50 mM Tris-HCl (pH 8.0), 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS along with phosphatase and protease inhibitors. After sonication on ice, protein lysates were obtained by centrifugation at 12,000 × g at 4°C for 10 min. The protein concentration was determined using the BCA assay. Fifty micrograms of the total protein was mixed with 4X loading buffer [360 mM TrisHCl (pH 6.8), 30% glycerol and 10% SDS]. The protein samples were then heated at 95°C for 5 min and electroseparated using 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto a PVDF membrane (Immobilon-P; Millipore). After being blocked with 0.5% skim milk, the membrane was probed with a monoclonal antibody for PTPRJ (R&D, MAB1934), or PARP (#9532), pro-caspase-3 (#9665), cleaved-caspase-3 (#9661), phospho-STAT3 (Tyr705) (#9145), phospho-JAK1 (Tyr1022/1023) (#3331), STAT3 (#9132), JAK1 (#1013) and GAPDH (#3683), which were purchased from Cell Signaling Technology. The membrane was then incubated with a horseradish peroxidase-conjugated secondary antibody and developed with the SuperSignal chemiluminescence kit (Immobilon Western Chemiluminescent HRP Substrate; EMD Millipore).

C33A cells were cultured into 96-well plates at ~5,000 cells/well. After 24 h at 37°C, the cells were transfected with the indicated plasmids and allowed another 48-h culture. Cell viability was assessed using the Dojindo Cell Counting Kit-8 (CCK-8; Dojindo Molecular Technologies, Kumamoto, Japan). Experiments were performed in octuplicate. For the colony formation assay, the stable C33A cells were cultured in a 6-well tissue culture plate at 500 cells/well. Cells were grown for 10–14 days. After being fixed by 4% paraformaldehyde for 30 min, the colonies were stained with 0.1% crystal violet for 15 min and washed. The colonies were counted by Image J software.

Stable C33A cells (200 μl) in RPMI-1640 medium without FBS were loaded into the upper chambers of BD Falcon™ Cell Culture Inserts (8-μm pore size; 1×10⁵ cells/chamber). The lower chambers were filled with 800 μl medium (RPMI-1640 plus 10% FBS). After a 24-h incubation at 37°C, the cells were fixed with paraformaldehyde and stained with 4′,6-diamidino-2-phenylindole (DAPI). The stained cells on the lower side were observed under a fluorescence microscope. Three fields per chamber were photographed and the number of migrated cells was counted by Image J software.

Signaling pathway arrays were conducted by a luciferase assay. Reporters HSF1-luc, phos-phatidylinositol-4,5-bisphosphate 3-kinase (PI3K)/AKT-luc and STAT3-luc were purchased from Qiagen (Valencia, CA, USA). C33A cells were cotransfected with reporter plasmids and pSicoR-PTPRJ or shPTPRJ in 24-well plates. After 48 h, the cells were lysed in 1X reporter lysis buffer (Promega, Madison, WI, USA) and the luciferase activity was measured by a Dual-Luciferase^® Reporter assay system (Promega). Experiments were conducted in triplicates and repeated at least three times.

Total RNA was isolated using a TRIzol RNA extraction kit, and 1 μg of RNA was reverse transcribed into cDNA according to the manufacturer’s instructions (RevertAid First strand cDNA Synthesis kit; Thermo). Real-time PCR was performed to measure the expression of cyclin D1, Bcl-2, Bax, VEGF and MMP9. The following primers were used: GAPDH-F, 5′-TGCACCACCAACTGCTTAGC-3′ and GAPDH-R, 5′-GCATGGACTGTGGTCATGAG-3′; BAX-F, 5′-GAGGATGATTGCCGCCGTGGACA-3′and BAX-R, 5′-GGTGGGGGAGGAGGCTTGAGG-3′; Bcl-2-F, 5′-ATG TGTGTGGAGAGCGTCAACC-3′ and Bcl-2-R, 5′-TGAGCAG AGTCTTCAGAGACAGCC-3′; MMP9-F, 5′-CCTGGAGAC CTGAGAACCAATC-3′ and MMP9-R, 5′-CCACCCGAGT GTAACCATAGC-3′; and VEGF-F, 5′-TGCCCACTGAGGA GTCCAAC-3′ and VEGF-R, 5′-TGGTTCCCGAAACGC TGA G-3′. All samples were read in triplicate, and values were normalized to GAPDH expression.

The stable C33A cells were harvested and washed twice with PBS and fixed in 70% cold ethanol at 4°C overnight. Before the analysis, the cells were washed twice with PBS, and then resuspended with 400 μl PBS and 100 μg/ml RNaseA (Sigma) and 50 μg/ml propidium iodide (PI) (Sigma). After incubation for 30 min at 37°C, the cells were subjected to DNA content analysis using a FACS Calibur (Becton Dickinson, San Jose, CA, USA). The results were analyzed with Flowjo software.

SPSS 19 software was used for statistical analyses. Data are presented as means ± SD. Statistical significance was determined by the Student’s t-test in some experiments. Differences with P-values of <0.05 were considered significant.

PTPRJ is widely regarded as a tumor suppressor and its expression is decreased in various tumor types. In the present study, we analyzed the mRNA and protein levels of PTPRJ in 8-paired patient samples. We observed that both the mRNA and protein levels of PTPRJ were decreased in the majority of the examined tumor tissues (75%). There was a >50% decrease in 6-paired tumor and non-tumor tissues (Fig. 1A and B).

To investigate the role of PTPRJ in the C33A cell line, we used construct codings for two different shRNAs against PTPRJ or a control shRNA. After puromycin selection, we obtained the stable cell lines expressing shPTPRJ. The PTPRJ shRNAs effectively decreased the mRNA and protein levels of PTPRJ, as determined by western blotting (Fig. 2A).

To assess whether the modulation of PTPRJ level influences the tumorigenetic properties of the C33A cells, we measured the cell viability, proliferation and migration capabilities by using these stable cell lines or transfected C33A cells with various doses of the indicated plasmids. By the CCK8 assay, we observed that the shPTPRJ-transfected cells showed an increased cell viability in a dose-dependent manner (Fig. 2B). By a colony formation assay, we observed that the shPTPRJ-infected C33A cells exhibited increased proliferation (Fig. 2C). To further investigate the role of PTPRJ in cell migration, we performed an invasion assay by using a Transwell chamber under serum-starved conditions. We found that the shPTPRJ infection significantly enhanced the C33A cell invasive ability by ~2.2-fold relative to the scramble infection (Fig. 2D). Finally, the cell cycle distribution of the shPTPRJ/scramble-infected C33A cells was measured by use of propidium iodide staining. As shown in Fig. 2E, the percentage of shPTPRJ-infected cells in the G1 phase was considerably decreased with significantly increased percentages of cells in the S and G2/M phases, in comparison to the scramble-infected cells.

To identify the effects of PTPRJ re-expression on the cell proliferation of C33A cells, we generated a stable cell line with high endogenous PTPRJ levels (Fig. 3A). C33A cells with an increased PTPRJ expression had a growth disadvantage, as assessed by the CCK-8 assay (Fig. 3B). Consistent with the previous studies in other cervical cancer cells such as HeLa (16), cell proliferation in the PTPRJ-overexpressing C33A cells was obviously reduced as compared to the corresponding control cells, confirming the antiproliferative activity of PTPRJ in the cervical cancer cell line C33A (Fig. 3C). We further used the PTPRJ-expressing C33A cells to perform a cell migration assay. Our results revealed that PTPRJ upregulation effectively reduced cell migration toward the serum starvation (Fig. 3D). By cell cycle analysis, we observed that the PTPRJ-overexpressing cells had a clearly decreased S-phase proportion and increased G2/M arrest (Fig. 3E).

5-FU is an important and widely used anticancer drug, including cervical cancer. However, tumor cell resistance to 5-FU-induced cell death often occurs during clinical treatment (17). The PTPRJ agonist has been shown to induce cell apoptosis (18). To investigate the role of PTPRJ in 5-FU-induced cell apoptosis, we conducted western blotting to detect the expression of caspase-3 and PARP, the effective markers for cell apoptosis. Our results revealed that cleaved-caspase-3 and cleaved-PARP were upregulated after treating the wild-type C33A cells with 50 μg/ml 5-FU, while the levels of pro-caspase-3 and full-length PARP were decreased. In the stable C33A cell line that overexpressed PTPRJ, the levels of pro-caspase-3 and full-length PARP were significantly lower, and the levels of cleaved-caspase-3 and cleaved-PARP were higher than the levels in the control group, after treatment with an equal dose of 5-FU (50 μg/ml). In contrast, pro-caspase-3 and full-length PARP were increased, while cleaved-caspase-3 and cleaved-PARP were decreased in the stable cell line expressing shPTPRJ (Fig. 4A).

Next, using the CCK-8 assay, we measured the cell viability of the wild-type C33A cells and PTPRJ/shPTPRJ stable cell lines after treatment with 5-FU. As expected, when given the 5-FU treatment, the viability of the stable cell line expressing shPTPRJ was obviously higher than the wild-type cells, while cells overexpressing PTPRJ showed a decreased cell viability, when compared with the wild-type cells (Fig. 4B).

In addition, the stable transfectants were treated with DMSO or 5-FU and subjected to FACS assays. At the basal state, the apoptotic rates of the cells transfected with the indicated genes remained unchanged. When the cells were exposed to 5-FU, ectopic overexpression of PTPRJ obviously increased the percentage of apoptotic cells, while inhibition of PTPRJ by shPTPRJ-2 attenuated the 5-FU-induced cell apoptosis, compared with the cells transfected with pSicoR-control or scramble, respectively (Fig. 4C). All things considered, the inhibition of PTPRJ decreased the sensitivity of C33A to 5-FU-induced cell apoptosis.

PTPRJ is a receptor-like protein-tyrosine phosphatase that directly regulates the dephosphorylation of several receptor tyrosine kinases, including FLT3, ERK, PDGFR. Omerovic et al (19) reported that PTPRJ negatively regulated serine/threonine protein kinase B (AKT) phosphorylation in Ras-mutated cancer cells. In accordance with previous studies, during the pathway screening assay, we observed that the luciferase activity of the PI3K/AKT pathway was decreased in the C33A cells transfected with PTPRJ, while inhibition of PTPRJ by shRNA led to an apparent increase in the activity of the PI3K/AKT pathway (Fig. 5A). Interestingly, we observed a similar change of heat shock transcription factor 1 (HSF1) and STAT3 pathway (Fig. 5A). STAT3 is crucial for cell proliferation, differentiation and apoptosis (20–22). STAT3 is regulated by JAK1 and other stimuli or proteins. In cervical cancer, STAT3 was reported to be correlated with clinical stage and positive-pSTAT3 patients have a favorable prognosis (23,24). Given the dephosphorylation activity of PTPRJ, we analyzed the phosphorylated levels of JAK1 and STAT3. The western blotting showed that the levels of pJAK1 and pSTAT3 were decreased while the total protein levels of JAK1 and STAT3 were constant when the cells were transfected with PTPRJ. On the contrary, shPTPRJ transfection caused the upregulation of the levels of pJAK1 and pSTAT3 (Fig. 5B).

To confirm the effect of PTPRJ on the JAK1/STAT3 pathway, the mRNA levels of cyclin D1, Bax, Bcl-2, VEGF and MMP9, downstream effectors of STAT3, were measured by real-time PCR. The levels of cyclin D1, Bcl-2, VEGF and MMP9 were decreased, and the expression of Bax was increased in the cells transfected with PTPRJ (Fig. 5C). We obtained the opposite results in the cells transfected with shPTPRJ (Fig. 5D). Collectively, these results revealed a negative regulation of PTPRJ on the JAK/STAT3 signaling pathway.