CD63 negatively regulates hepatocellular carcinoma development through suppression of inflammatory cytokine‐induced STAT3 activation

Tetraspanin CD63 has been widely implicated in tumour progression of human malignancies. However, its role in the tumorigenesis and metastasis of hepatocellular carcinoma (HCC) remains unclear yet. In the present study, we aimed to investigate the specific function and underlying mechanisms of CD63 in HCC progression. CD63 expression in HCC tissues was detected using immunohistochemistry and quantitative real‐time PCR analyses; effects of CD63 on HCC cell proliferation and migration were investigated by CCK‐8 assay, colony formation assay, transwell assay and a xenograft model of nude mice. RNA‐sequencing, bioinformatics analysis, dual‐luciferase reporter assay and Western blot analysis were performed to explore the underlying molecular mechanisms. Results of our experiments showed that CD63 expression was frequently reduced in HCC tissues compared with adjacent normal tissues, and decreased CD63 expression was significantly associated with larger tumour size, distant site metastasis and higher tumour stages of HCC. Overexpression of CD63 inhibited HCC cell proliferation and migration, whereas knockdown of CD63 promoted these phenotypes. IL‐6, IL‐27 and STAT3 activity was regulated by CD63, and blockade of STAT3 activation impaired the promotive effects of CD63 knockdown on HCC cell growth and migration. Our findings identified a novel CD63‐IL‐6/IL‐27‐STAT3 axis in the development of HCC and provided a potential target for the diagnosis and treatment of this disease.

chemoembolization (TACE) and multikinase inhibitor sorafenib, the long-term survival rates of HCC remain dismal due to compensatory revascularization and drug resistance. 4,5 Therefore, it is necessary to explore the underlying mechanism of HCC pathogenesis to uncover new therapeutic strategies against HCC.
Tetraspanins are a group of cell surface-associated membrane proteins expressed in a wide variety of cell types, and they are characterized by four transmembrane domains. 6,7 It is well established that tetraspanin proteins mediate cellular signal transduction and play important roles in the regulation of many physiological properties including cell motility, adhesion, proliferation, invasion and differentiation, 8 and some tetraspanin molecules have been associated with progression of various human malignancies, such as melanoma, nonsmall cell lung cancer (NSCLC), breast cancer and pancreatic cancer. 9,10 For example, tetraspanin CD82 is frequently down-regulated in advanced stages of cancer, and overexpression of CD82 inhibits tumour migration and invasion via regulation of several signal pathways such as hepatocyte growth factor receptor (HGFR) pathway. 11,12 Tetraspanin CD9 has been reported to inhibit the proliferation and tumorigenicity of colon carcinoma cells, 13 and ectopic expression of CD9 in NSCLC suppresses cell motility and promotes apoptotic cell death by regulating AKT phosphorylation and MMP2 secretion. 14 The cluster of differentiation 63 (CD63), also known as melanoma-associated antigen ME491 or MLA1, is the first identified member of the tetraspanin family. 15 CD63 is ubiquitously expressed and mainly localized within the endosomal system and at the cell surface. 16 Previous studies have implicated a role of CD63 in cellular transport of other proteins. [16][17][18] For instance, it has been found in the complex with MHC II, which suggested that CD63 might chaperone MHCII molecules through the endosomal system and plays a role in antigen presentation via MHC II. 19 More importantly, emerging evidences have implicated a correlation between abnormal CD63 expression and tumour progression. Although CD63 was first identified to be expressed in early-stage melanoma cells, it is reduced while tumour cells become more invasive, implying a negative role of CD63 in tumour invasiveness. 20 In other tumours including lung cancer and breast cancer, low CD63 expression is correlated with poor prognosis of patients. 21 These findings suggested that CD63 is related to tumour cell motility and metastasis. However, its expression and functions in HCC remain to be elucidated.
In the present study, we resorted to investigate the role of CD63 in the development of HCC and explore the underlying molecular mechanisms using molecular and cell biology experiments in vitro and in vivo.

| Western blot analysis
Total proteins were isolated from HCC cells with RIPA Lysis Buffer

| Dual-luciferase reporter assay
Briefly, HCC cells were seeded into a 24-well plate and incubated for 24 hours before transfection, then 250 ng of STAT3 reporter plasmid (Genomeditech) and 10 ng of SV40, a Renilla luciferase expression plasmid as an internal control were cotransfected into the prepared cells. 24 hours after transfection, cells were harvested and dual-luciferase reporter assay was carried out with a Dual-Glo Luciferase kit according to the manufacturer's specifications (Promega). A GloMax ® 96 Microplate Luminometer (Promega) was used to quantify both firefly and Renilla luciferase activities. All experiments were done in triplicate and repeated at least 3 times.

| RNA-sequencing and bioinformatics analysis
Briefly, Huh7 cells were seeded into a 10-cm dish at a confluence of

| Colony formation assay
For colony formation assays, stably infected HCC cells were cultured in a 6-well plate at a density of 1000 cells per well and incubated at 37°C for 1-2 weeks; then, the colonies were harvested. After fixation in 4% paraformaldehyde, the colonies were stained with crystal violet (Sangon Biotech) for 10 minutes. Images of the colonies were captured, and the number of cells was counted using Image J software.

| Cell migration assay
Cell migration assays were performed with a 24-well transwell cham- Microsystems) was used to photograph and count the number of migrated cells in five randomly selected fields at a magnification of ×100.

| Animal experiments
To establish a xenograft model of HCC, 4-to 6-week-old male BALB/c nude mice were purchased from Sippr-BK laboratory animal corporation, Shanghai, China. Equal amount of stably infected HCC cells (2 × 10 6 cells) were injected subcutaneously into each flank of armpit area of nude mice, respectively (n = 5). After 4 weeks, the mice were euthanized,

| Enzyme-linked immunosorbent assay (ELISA)
To evaluate the effects of CD63 on IL-6 and IL-27 protein levels, the indicated HCC cells were treated with LPS (1 μg/mL, Beyotime) for 6 hours to induce the production of inflammatory cytokines and the culture supernatants were collected. IL-6 and IL-27 protein levels were measured using ELISA kits (#KE00007 and #KE00089, Proteintech) according to the manufacturer's instructions.

| Statistical analysis
Quantitative data were represented as the means ± SD, and GraphPad Prism software 7.0 was used to perform statistical analyses. The chi-square test was used to explore the relationships between CD63 expression and clinicopathological parameters of HCC patients. Statistical significance was determined by one-way analysis of variance (one-way ANOVA), and Dunnett's multiple comparison test was used as a post hoc test or the Student t test.
A P value less than 0.05 was considered as statistically significant.

| CD63 expression is reduced in HCC tissues and associated with clinicopathological parameters of HCC patients
The protein expression of CD63 in a tissue microarray (TMA) consisting of 75 HCC tumour tissues and paired adjacent normal tissues was analysed by immunohistochemistry. The results showed that CD63 was frequently down-regulated in tumour tissues compared with adjacent F I G U R E 2 CD63 acts as a tumour suppressor gene in HCC cell proliferation in vitro and in vivo. A, 48 h after cell transfection, the efficacies of CD63-expressing plasmids (pCD63) and siRNAs targeting CD63 (siCD63) were confirmed by Western blot analysis, respectively. Empty vector (pVEC) and non-specific control siRNA (siNC) were used as control group. B, proliferative abilities of HCC cells were evaluated using CCK-8 assays. C, CD63 expression in stably infected HCC cells were examined by Western blot analysis. D, colony formation assays were performed using stably infected HCC cells. E, Huh7 and MHCC-LM3 cells infected with LV-CD63/LV-shCD63 were subcutaneously inoculated into nude mice, respectively. Cells infected with LV-VEC/LV-shNC were used as control group. (n = 5 per group). Tumours were resected and weighted after 4 wk. *P < .05, **P < .01 normal tissues at protein level ( Figure 1A,B). Next, we analysed the relationship between CD63 expression and clinicopathological parameters in the TMA. Lower CD63 expression was significantly associated with larger tumour size (P < .05), distant site metastasis (P < .01) and higher TNM stage (P < .01) ( Table 1 and Figure 1C). Furthermore, the mRNA level of CD63 in additional 35 paired HCC tissues was determined by qPCR analysis. Consistently, CD63 mRNA expression in tumour tissues was remarkably lower than in adjacent normal tissues ( Figure 1D), and it exhibited a <0.5-fold decrease in 21/35 (60%) paired tissues ( Figure 1E).
Taken together, these results suggested that CD63 expression is reduced in HCC tissues and directly associated with the development of HCC.

| CD63 inhibits HCC cell proliferation in vitro and tumorigenicity in animal models
To explore the impacts of CD63 on cell proliferation of HCC, we intervened its expression by transfecting CD63-expressing plasmids Therefore, these results demonstrated that CD63 plays a critical role in suppressing HCC cell proliferation in vitro and in vivo.

| CD63 negatively regulates HCC cell migration in vitro
In previous studies, CD63 has been associated with tumour metastasis in other tumours including melanoma and breast cancer. 20,23 Therefore, we next examined whether CD63 could influence HCC cell migration ability in vitro. After intervening CD63 expression in the indicated HCC cell lines, transwell migration assays were performed.
As presented in Figure 3A,B, overexpression of CD63 significantly re-

| CD63 inhibits inflammation-related oncogenic signalling pathways in HCC cells
To explore the molecular mechanisms through which CD63 inhibited HCC cell proliferation and migration, RNA-sequencing was applied using Huh7 cells with ectopic CD63 expression, and global  Figure S1). The results mentioned above suggested that CD63 may suppress HCC cell proliferation and migration possibly by inhibiting the expression of cytokines IL-6 and IL27, which in turn inactivate inflammation-related oncogenic signalling pathways.

| CD63 has a negative impact on STAT3 activation in HCC
Considering that both IL-6 and IL-27 contribute to tumour progression through JAK/STAT3 signalling pathway, 24,25 we next elucidated whether CD63 regulated the activation of signal transducer and activator of transcription-3 (STAT3) by Western blot analysis. As shown in Figure 5A, overexpression of CD63 significantly decreased the level of phosphorylated STAT3 (Y705), while total protein levels of STAT3 showed no obvious difference. In contrast, silencing CD63 expression led to opposite results. To verify this regulation of STAT3 by CD63, a dual-luciferase reporter assay was performed using STAT3 response element reporter plasmid. As expected, overexpression of CD63 inhibited the transcriptional activity of STAT3 in L02 cells, while knockdown of CD63 remarkably increased the transcriptional activity of STAT3 in MHCC-LM3 cells ( Figure 5B).
These mechanistic studies demonstrated that CD63 could inactivate STAT3 in HCC cells.

| STAT3 inhibition blocks CD63 knockdowninduced cell growth and migration
To assess the effects of the relationship between CD63 and STAT3 on cell growth and migration in HCC, a STAT3-specific inhibitor F I G U R E 4 CD63 inhibits activation of inflammation-related oncogenic pathways in HCC cells. A, volcano plot showed differentially expressed genes between Huh7-CD63 cells and control Huh7 cells. Red dots indicated up-regulated genes in Huh7-CD63 cells, and blue dots indicated down-regulated genes in Huh7-CD63 cells. Fold change ≥2 or ≤−2 and q value <0.05 were considered as effective candidates. B, 539 down-regulated genes in Huh7-CD63 cells were used to perform KEGG pathway analysis. Gene numbers enriched in each pathway were exhibited as different size of circles. C, results of GSEA analysis showed that IL-17 signalling pathway, cytokine-cytokine receptor interaction pathway, NOD-like receptor signalling pathway and TNF signalling pathway were enriched in control Huh7 cells comparing with Huh7-CD63 cells. D, heat map was used to show the top 30 down-regulated genes in huh7-CD63 cells. E, after transfection with pCD63 or siCD63, qPCR analysis was performed to detect the mRNA levels of cytokines IL-6 and IL-27 in HCC cells F I G U R E 5 CD63 suppresses STAT3 activation in HCC cells. A, Western blot analysis were performed to detect the expression of p-STAT3 (Y705) and total STAT3 in HCC cells stably infected with LV-CD63 or LV-shCD63. LV-VEC and LV-shNC were used as control groups, respectively. B, 250 ng of the STAT3 response element reporter plasmid was transfected into the indicated HCC cells, respectively. After incubation for 24 h, dual-luciferase reporter assays were performed to determine the transcriptional activity of STAT3 BP-1-102 was used to block STAT3 activation 27 ( Figure 6A). Results

of CCK-8 and colony formation assays showed that treatment with
BP-1-102 significantly abolished CD63 knockdown-induced cell growth and colony formation of MHCC-LM3 cells ( Figure 6B,C).
Similarly, the data from transwell migration assays also demonstrated that BP-1-102 treatment dramatically attenuated the migratory ability of CD63 silenced MHCC-LM3 cells ( Figure 6D). These findings collectively hinted us that STAT3 signalling is required for the negative effects of CD63 on cell proliferation and migration in HCC. B and C, the indicated HCC cells were cultured in medium supplemented with BP-1-102 (10 μmol/L) or DMSO, and CCK-8 assays and colony formation assays were carried out to assess the influence of BP-1-102 on the effects of CD63 silencing in proliferation ability of MHCC-LM3 cells. D, after treating the above HCC cells with BP-1-102 (10 μmol/L) or DMSO, transwell migration assays were performed to confirm whether CD63 inhibited HCC cell migration via inactivating STAT3 of osteosarcoma also reported that knockdown of CD63 resulted in decreased pY705-STAT3 expression, 35 which supports our findings in HCC. Therefore, these data establish a CD63-IL-6/IL-27-STAT3

| D ISCUSS I ON
axis involved in the regulation of HCC cell proliferation and migration. However, it is still unknown how CD63 regulates IL-6 and

IL-27 expression in HCC cells.
It is well established that CD63 is an important biomarker of exosomes, which are microvesicles secreted by most types of cells and participate in intercellular communication via transmitting its cargo including proteins and nucleic acids into receptor cells. 36,37 Intriguingly, Cheng et al reported that a specific type of exosomes (p120ctn) inhibited HCC cell progression via STAT3 pathway, hinting us that CD63 possibly regulates inflammation-related oncogenic pathways by influencing biogenesis and secretion of exosomes. 38 However, more studies need to be done in order to verify this hypothesis.
In conclusion, these clinical and mechanistic findings identified a novel CD63-IL-6/IL-27-STAT3 axis in HCC tumorigenesis and development, and targeting this axis is a potential therapeutic strategy against HCC.

ACK N OWLED G EM ENTS
This study was partially supported by grants from the National PWYgy2018-02).

CO N FLI C T O F I NTE R E S T
The authors declare there is no conflict of interest.

E TH I C A L A PPROVA L
This study has been approved by the ethics committee of Shanghai East Hospital, Tongji University School of Medicine, China.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data used to support the findings of this study are available from the corresponding author upon request.