HAPLN3 inhibits apoptosis and promotes EMT of clear cell renal cell carcinoma via ERK and Bcl-2 signal pathways

Hyaluronan and proteoglycan link protein 3 (HAPLN3) is a member of the hyaluronan and proteoglycan link protein family expressed in the extracellular matrix closely associated with the development and occurrence of various malignant tumors; yet, its function in clear cell renal cell cancer (ccRCC) is still poorly understood. The following study investigated the progress and mechanism of HAPLN3 on ccRCC using bioinformatics analysis and in vitro experiments. In order to determine whether HAPLN3 is differentially expressed in ccRCC, we analyzed data from the Cancer Genome Atlas (TCGA) and GSE40435 and further validated them in the Human Protein Atlas (HPA) database. Simultaneously, the TCGA dataset was utilized to study the relationship between HAPLN3 expression and the progression of ccRCC and its prognostic value in ccRCC. Gene enrichment analysis (GSEA) was used to explore HAPLN3-related signaling pathways in ccRCC. The TIMER database investigates the link for both HAPLN3 and immune cell infiltration. Different ccRCC cell lines the role of HAPLN3 on cell biological behavior in vitro. HAPLN3 was increased in ccRCC, and its high expression was related to the patients' survival rates and clinical characteristics. GSEA showed that HAPLN3 is mainly enriched in proliferative and metastatic pathways. In addition, HAPLN3 was an independently associated significant predictor in patients with ccRCC. Functional experiments demonstrated that HAPLN3 could promote the proliferation, migration, and invasion of ccRCC cells through the ERK1/2 signaling pathway. To sum up, our data suggest that HAPLN3 may serve as a new prognostic biomarker and potential therapeutic target for ccRCC.


Introduction
Renal cell carcinoma (RCC) is a frequent malignant tumor of the urinary system that accounts for 2.2% of all cancer patients (Sung et al. 2021). In 2020, there were more than 400,000 new cases of RCC globally, with 175,000 deaths attributable to the disease. Clear cell renal cell carcinoma (ccRCC) is the most common RCC subtype, representing approximately 75% of all RCC cases (Hsieh et al. 2017). Treatments for people with ccRCC include surgery and immunotherapy, considering that traditional chemotherapy and radiation therapy are largely ineffective for these patients. Yet, despite the continuous improvement of current diagnostic techniques, one-third of patients initially diagnosed with RCC still have local or distant metastases. For ccRCC patients with distant metastases, the 5-year overall survival rate does not exceed 12% (Zhang et al. 2022). Moreover, more than 20% of ccRCC patients experience recurrence after nephrectomy (Choueiri and Motzer 2017;Suarez et al. 2018). With the increasing number of genetic technology methods, the identification and validation of biomarkers are particularly important for optimizing molecularly targeted therapeutic drug sequences (Atkins and Tannir 2018). Therefore, searching for new and effective prognostic molecular markers is critical for improving prognoses of these patients.
Hyaluronan and proteoglycan link protein 3 (HAPLN3) is an extracellular matrix linker protein, a member of the hyaluronan and proteoglycan linker protein family, involved in the binding of proteoglycans to hyaluronic acid as well as cell adhesion (Spicer et al. 2003). HAPLN3 is expressed in most tissues and is thought to be essential for creating and remitting hyaluronic acid-dependent extracellular matrix (Spicer et al. 2003;Ogawa et al. 2004). HAPLN1 has a key role in the malignant progression of many forms of cancer (Ivanova et al. 2009;Evanko et al. 2020). Relevant studies have indicated that HAPLN3 overexpression in breast cancer is linked to its occurrence and metastasis (Kuo et al. 2010). In addition, HAPLN3 methylation of circulating tumor DNA is significantly increased in metastatic prostate cancer and serves as a post-treatment risk predictor (Bjerre et al. 2020). Yet, the role of HAPLN3 in ccRCC is still poorly understood.
In the present research, we evaluated the expression and prognostic value of HAPLN3 in ccRCC using bioinformatics analysis and investigated whether HAPLN3 is involved in immune cell infiltration in ccRCC. In addition, the biological effects of HAPLN3 in ccRCC were confirmed in vitro. Our findings elucidated the critical role of HAPLN3 in ccRCC for the first time, indicating that HAPLN3 could be an independent prognostic molecule for ccRCC and a potential target for future therapeutic strategies.

Data collection and processing
The RNA-Seq expression data of ccRCC patients (including 539 tumor samples and adjacent tissue samples) were downloaded from the TCGA database (https:// tcga-data. nci. nih. gov/ tcga/), and 537 patients' HAPLN3 gene expression data and clinical information were sorted for further analysis. Meanwhile, to ensure the precision of TCGA data results, we used GSE40435 in the GEO data to verify the expression level of HAPLN3. The dataset includes 101 pairs of ccRCC tissue and adjacent tissue samples. In addition, we also downloaded HAPLN3-related immunohistochemistry (IHC) data from the HPA database (http:// www. prote inatl as. org/) to compare the protein expression of HAPLN3 in normal and ccRCC tissues.

Gene set enrichment analysis (GSEA)
GSEA can be used to analyze whether a set of genes with a priori-defined differential expression between HAPLN3 (high expression) and HAPLN3 (low expression) are enriched in the MSigDB collection (Molecular Signatures Database, c2.cp.kegg.v7.5.symbols.gmt) (Subramanian et al. 2005). Using the median expression of HAPLN3 mRNA as a criteria, all ccRCC samples were classified into lowand high-expression groups, and enrichment analysis using GSEA software identified HAPLN3-related signaling pathways (with NOM P-value < 0.05 and FDR < 0.25 as cutoff criteria).

Samples of patient tissue and cell culture
Seven pairs of ccRCC tumor tissues and paracancerous tissues were collected at The First Affiliated Hospital of Nanchang University. Informed consent was obtained from all patients. In addition, this study was approved by the First Affiliated Hospital Ethics Committee of Nanchang University.
The A498, 786-O, OSRC-2, ACHN, and HK2 cell lines were purchased from The Cell Bank of The Chinese Academy of Sciences (Shanghai, China) and used for in vitro study. HK2 cells were cultured in DMEM/F12 medium (BI, China); A498 and ACHN cells were cultured in MEM medium (Gibco, China); 786-O and OSRC-2 cells were cultured in RIPM-1604 medium (Gibco, China). All of the media included 1% penicillin-streptomycin and 10% fetal bovine serum (FBS, BI, China). All cells were cultured in a humidified atmosphere containing 5%CO 2 / 95% air at 37 °C.

Western blot
Total protein was extracted using Solarbio's RIPA Lysis Buffer, after which protein samples were quantified, denatured, and all proteins were separated by SDS-PAGE and transferred to PVDF membranes (Bio-Rad, USA). After blocking with 5% nonfat milk (Solarbio, China) for 1 h, the membrane was incubated with the following primary antibodies: HAPLN3 (Novus, USA), Bcl-2, PARP, p-ERK1/2, ERK1/2, E-cadherin, N-cadherin (CST, USA), GAPDH, Vimentin (Servicebio, China), and Bax (Proteintech, China) at 4 °C overnight; all antibodies were diluted in Primary Antibody Diluent (NCM Biotech China). Then, the membrane was rinsed three times with TBST for 10 min. The samples were then incubated for 1 h at room temperature with the secondary antibody. The signal was seen using an enhanced chemiluminescence kit (Millipore, USA) and quantified on PVDF membrane by LAS-4000 Mini (Fuji Film, Japan).

Cell transfection
Small interfering RNA targeting HAPLN3 (siHAPLN3) and negative control (siNC) were developed and manufactured by GenePharma (Shanghai, China). The siRNA was transfected into 50% confluent ccRCC cells using Lipofectamine ® 2000 (Invitrogen, USA), following the manufacturer's instructions. The effectiveness of transfection was determined by qRT-PCR after 24 h and by Western blot analysis after 48 h. The siHAPLN3 sequence was as follows: 5′-GGC GCU ACC AGU UCA ACU UTT-3′.

Cell counting kit-8 assay
Forty-eight hours after cell transfection, 2.5 × 10 3 cells per well were plated in 96-well culture plates and cultured for 0, 24, 48, 72, and 96 h. After each time point, a sterile CCK-8 dye (10 µl/ well) was added to each well and incubated for another 2 h at 37 °C. The absorbance at 450 nm was determined using a microplate reader (SpectraMax Plus 384).

EdU assay
Forty-eight hours after cell transfection, 2 × 10 4 cells per well were seeded in a 96-well culture plate, and the culture was continued for 24 h. The EdU stock solution was then diluted to 50 μmol/L with a complete medium, adding 100 μL of the solution to each well. After 2 h, the medium was removed, and EDU click response solution and Hoechst 33,342 staining were added to each well, following the EDU staining kit instructions. Cell proliferation was monitored using a fluorescence microscope.

Transwell migration and invasion assay
A mixture of Matrigel (Corning, USA) and a serum-free medium (ratio of 1:8) was evenly spread on the bottom of datasets. D Protein expression of HAPLN3 in ccRCC and normal tissue from HPA data. *P < 0.05; **P < 0.01; ***P < 0.001 a Transwell chamber and left at 37 °C for approximately 2 h to solidify. Subsequently, the cells were digested and counted, diluted with serum-free media to a certain ratio, and incubated in the Transwell chamber before being grown in a mixture containing 20% FBS. After 24 h of cell culture, we took out the Transwell chamber, washed, fixed, and stained the cells, took pictures and counted the cells in five random fields of view, and drew the corresponding histogram.
Migration and invasion experiments differed only in that Matrigel was not used, while all other steps were identical.

Statistical analysis
R 4.0.3 and GraphPad Prime 8.0 were used to conduct all statistical analyses. An independent prognostic analysis of HAPLN3 was performed using both univariate and multivariate Cox analysis. Kaplan-Meier method was used to calculate survival curves. Mann-Whitney U tests and two-tailed Student's t tests were used to compare the different groups. P < 0.05 was regarded as statistically significant. Data from three separate experiments were collected and processed as means and standard deviations.

Expression of HAPLN3 in ccRCC
The TCGA database results revealed that the expression level of HAPLN3 mRNA in ccRCC samples was significantly higher than in normal tissues (Fig. 1A, B).  Subsequently, high HAPLN3 mRNA expression in ccRCC tissues was found after analyzing GSE40435 in the GEO database (Fig. 1C).
To further examine HAPLN3 protein expression, we retrieved IHC data from the HPA database. In normal kidney tissue, the intensity of HAPLN3 staining was obviously weaker than that in ccRCC tissue (Fig. 1D).

Relationship between HAPLN3 expression and clinical characteristics
To explore the link between HAPLN3 expression and clinical characteristics, we downloaded 537 ccRCC samples with HAPLN3 gene expression and used clinical data from the TCGA database for Wilcox test analysis. The results showed that tumor grade (Grade) (P < 0.001), clinical-stage (Stage) (P < 0.001), tumor invasion depth (T) (P < 0.001), lymph node invasion (N) (P = 0.004), and distant metastases (M) (P < 0.001) were all significantly correlated with HAPLN3 gene expression ( Fig. 2A-E). Furthermore, we assessed the overall survival of HAPLN3 mRNA expression in ccRCC patients using Kaplan-Meier survival analysis. In patients with ccRCC, elevated levels of HAPLN3 mRNA were obviously associated with worse survival (P = 0.004, Fig. 2F). To sum up, this data suggests that high levels of HAPLN3 may be responsible for accelerating the malignant progression of ccRCC.

GSEA identifies signaling pathways associated with HAPLN3
To screen for potential signaling pathways that are differentially activated in ccRCC, we compared high and low HAPLN3 gene expression datasets by GSEA. GSEA indicated considerable disparities in the MSigDB collection's richness (c.2.cp.kegg.v7.0.symbols.gmt). According to the criteria of NOM P < 0.05 and FDR < 0.05, the differently enriched pathways in the high-expression phenotype of the  (Table 1). The JAK-STAT signaling pathway, MAPK signaling pathway, VEGF signaling pathway, Hedgehog signaling pathway, NOTCH signaling pathway, and pathways in cancer were significantly enriched in the HAPLN3 high-expression phenotype, according to GSEA results (Fig. 3).

The prognostic analysis of HAPLN3 in ccRCC
Using Cox regression models, we analyzed multivariate hazard ratios for different variables from ccRCC patients in the TCGA-ccRCC dataset. Age, grade, stage, T classification (P < 0.001), TNM stage, and HAPLN3 overexpression (P < 0.001) were all significantly linked to the poor prognosis of ccRCC patients in a univariate analysis. Multivariate analysis further revealed that age, grade, and HAPLN3 overexpression (P = 0.006) could be considered independent prognostic variables for ccRCC (Fig. 4B). This data suggests that HAPLN3 could serve as an independent prognostic marker in patients with ccRCC.

Correlation between HAPLN3 expression and immune markers
The relationship between HAPLN3 and six key immune cells, including neutrophils, B lymphocytes, CD4+ T lymphocytes, CD8+ T lymphocytes, macrophages, and dendritic cells, was analyzed by the TIMER database. In ccRCC, HAPLN3 expression was significantly associated with CD8+ T lymphocytes, CD4+ T lymphocytes, macrophages, neutrophils, and dendritic cells (all P < 0.05). Among them, CD8+ T lymphocytes, neutrophils, and dendritic cells were significantly positively correlated with HAPLN3 expression (Fig. 5). These results suggest HAPLN3 may influence the occurrence and progression of ccRCC through immune cells.

The role of HAPLN3 in immune escape of ccRCC
CTLA-4, PD-1, and PD-L1 are the three main immune checkpoints associated with immune escape in cancer (Rappold et al. 2021). To investigate the role of HAPLN3 in the occurrence and progression of ccRCC, we explored the correlation between HAPLN3 and the CTLA-4, PD-1, PD-L1 corresponding markers CTLA4, PDCD1, CD274 through the TIMER database. As illustrated in Fig. 6, the level of HAPLN3 expression was significantly positively correlated with CTLA4 and PDCD1 in ccRCC after adjustment for purity. Based on these findings, HAPLN3 may affect the development of ccRCC by influencing tumor immune escape.

Validation of the expression of HAPLN3 in ccRCC in vitro
To validate the expression results of HAPLN3 in ccRCC, we used RT-qPCR and Western blotting. In ccRCC tissues, the protein expression level of HAPLN3 was significantly higher than that in the corresponding adjacent tissues (Fig. 7A). Moreover, HAPLN3 expression in the ccRCC cell line was significantly higher than in the normal cortex/proximal tubule epithelial cell line (HK2) (Fig. 7B). To further evaluate the biological function of HAPLN3 in ccRCC, we employed small interfering RNA technology to Fig. 6 Correlation of HAPLN3 expression with CTLA-4, PD-1, and PD-L1 expression in ccRCC. A Relationship between HAPLN3 and CTLA-4 expression in ccRCC adjusted for purity. B Relationship between HAPLN3 and PD-1 expression in ccRCC adjusted for purity. C Relationship between HAPLN3 and PD-L1 expression in ccRCC adjusted for purity knock down HAPLN3 expression in 786-O and A498 cells. According to the results, the level of HAPLN3 expression was significantly down-regulated after transfection (Fig. 7C).

Knockdown of HAPLN3 inhibits the proliferation, migration, and invasion of ccRCC cells
To determine the effect of HAPLN3 on ccRCC cell proliferation, we used CCK-8 and EdU cell proliferation experiments for verification (experimental grouping: the negative control group, i.e., siNC, and experimental group, i.e., siHAPLN3 group). The CCK-8 experiment revealed that when HAPLN3 was knocked down (siHAPLN3 group), the proliferation ability of 786-O and A498 cells was significantly reduced compared with siNC (Fig. 8A). The EdU experiment results further demonstrated that the siHAPLN3 group significantly inhibited the proliferation ability of ccRCC cell lines 786-O and A498 compared to the siNC group (Fig. 8B). In addition, HAPLN3 knockdown significantly reduced the ability of 786-O and A498 cells to migrate and invade (Fig. 9A, B). This data suggests that HAPLN3 may function as an oncogene in ccRCC, promoting tumor progression.

HAPLN3 regulates tumor growth via ERK1/2 signaling
Based on the GSEA results, the group with high HAPLN3 expression was primarily involved in the proliferation and metastasis pathways. Further defining the underlying molecular mechanisms of HAPLN3 regulation of ccRCC cell growth, by using Western blot assay, we discovered that knocking down HAPLN3 in ccRCCs reduces the expression of the p-ERK1/2 protein (Fig. 10A). In addition, after knocking down the expression of HAPLN3, the expression of apoptosis-related proteins, Bax, was up-regulated, and the expressions of Bcl-2 and PARP were down-regulated. Also, the expression of E-cadherin was significantly up-regulated, and the expressions of N-cadherin and Vimentin were reduced (Fig. 10B, C). Therefore, we speculate that HAPLN3 promotes ccRCC proliferation and metastasis via the ERK1/2 pathway.

Discussion
In this research, we combined public databases and found that HAPLN3 was overexpressed in ccRCC, and its high expression was negatively correlated with poor patient prognosis. High expression of HAPLN3 was positively Fig. 7 HAPLN3 is up-regulated in ccRCC tissues and cells. A The protein expression level of HAPLN3 is higher in ccRCC tissues than in normal tissues ("N" means normal and "T" means tumor). B HAPLN3 mRNA and protein expression levels are increased in ccRCC cell lines compared with the normal cortex/proximal tubule epithelial cell line (HK2). C The knockdown efficiency of siHAPLN3 is confirmed by qRT-PCR and Western blotting in 786-O and A498 cells. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001 associated with clinical characteristics (grade, stage, and TNM stage). In addition, HAPLN3 was a significant prognostic factor affecting survival in patients with ccRCC in univariate and multivariate Cox regression analysis. These results indicate that HAPLN3 could be a new marker for ccRCC.
To explore more about the function of HAPLN3 in ccRCC, we performed GSEA. To explore more about the function of HAPLN3 in ccRCC, we performed GSEA. According to the GSEA findings, the high-expression group of HAPLN3 in ccRCC was primarily enriched in the JAK-STAT signaling pathways, MAPK signaling pathways, VEGF signaling pathways, Hedgehog signaling pathways, NOTCH signaling pathways, and pathways in cancer; all these signaling pathways are highly correlated with tumor proliferation and metastasis (Huang et al. 2020;Zhang et al. 2018;Sohn et al. 2018;Skoda et al. 2018;Fasoulakis et al. 2021). Therefore, we speculate that HAPLN3 may act as a tumor-promoting gene in ccRCC to promote tumor progression by regulating the above tumorrelated signaling pathways.
Many investigations have shown that tumor immune cell infiltration can affect cancer patients' radiation, chemotherapy, and immunotherapy, leading to a worse prognosis (Gaudreau et al. 2021;McLaughlin et al. 2020;Zhang and Zhang 2020). In ccRCC, we discovered that HAPLN3 expression was substantially linked with CD8+ T lymphocytes, neutrophils, and dendritic cells. Furthermore, the expression of HAPLN3 was found to be positively correlated with the major immune checkpoints in ccRCC. This Fig. 8 The knockdown of HAPLN3 inhibits proliferation in ccRCC cells. A The CCK-8 assay was used to measure cell proliferation at different time points. B Edu experiment was performed to analyze the number of cell proliferation between siNC group and the siHAPLN3 group. *P < 0.05; **P < 0.01; ***P < 0.001 suggests that HAPLN3 may not only influence the progression of ccRCC through immune infiltration but can also be utilized as a target for ccRCC immunotherapy.
To validate the results of the bioinformatics study, we conducted a series of in vitro functional studies. The findings revealed that HAPLN3 was highly elevated in ccRCC. Additionally, knockdown of HAPLN3 inhibited the proliferation, migration, and invasive abilities of ccRCC cells. Apoptosis dysregulation is a cancer marker and is linked to tumor growth and progression, and the Bcl-2 protein family has a key role in the intrinsic apoptotic pathway (Pistritto et al. 2016). Furthermore, EMT is one of the hallmarks of tumor metastasis (Mittal 2018). Cadherin and Vimentin, as important molecules in the occurrence of EMT, are closely related to tumor invasion and metastasis (Kaszak et al. 2020;Tang et al. 2017). In this study, we found that … which suggests that HAPLN3 acts as an oncogene in ccRCC cells.
Furthermore, the MAPK signaling pathway was found to be significantly enriched in the HAPLN3 high-expression phenotype. Cell proliferation, apoptosis, migration, and invasion are all regulated by ERK1/2, an important pathway involved in MAPK family signaling (Jiang et al. 2020;Sun et al. 2015;Zhao et al. 2018). In this Fig. 9 The knockdown of HAPLN3 inhibits ccRCC cell migration and invasion. A Transwell assay showing a significant decrease in the migration ability of cells after HAPLN3-siRNA transfection. B Tanswell assay showing a significant decrease in the invasion ability of cells after HAPLN3-siRNA transfection. *P < 0.05; **P < 0.01; ***P < 0.001 study, we explored key proteins of the ERK1/2 signaling pathway. Our results demonstrated that knockdown of HAPLN3 inhibited the expression of p-ERK1/2. Therefore, HAPLN3 may promote cell proliferation, migration, and invasion by activating the ERK1/2 signaling pathway.
The present study has limitations. The animal experiments should be performed to further explore the role of HAPLN3 in vivo.

Conclusions
Our results demonstrate that HAPLN3 is overexpressed in ccRCC and could be used as a new prognostic biomarker. HAPLN3 can promote ccRCC cell proliferation, migration, and invasion via the ERK1/2 signaling pathway. This is the first study to report the differential expression and biological function of HAPLN3 in ccRCC using a combination of bioinformatics analysis and experiments. These findings could be used to identify potential therapeutic targets for ccRCC.