N‐acetylgalactosaminyltransferase GALNT6 is a potential therapeutic target of clear cell renal cell carcinoma progression

Abstract

High expression of truncated O‐glycans Tn antigen predicts adverse clinical outcome in patients with clear cell renal cell carcinoma (ccRCC). To understand the biosynthetic underpinnings of Tn antigen changes in ccRCC, we focused on N‐acetylgalactosaminyltransferases (GALNTs, also known as GalNAcTs) known to be involved in Tn antigen synthesis. Data from GSE15641 profile and local cohort showed that GALNT6 was significantly upregulated in ccRCC tissues. The current study aimed to determine the role of GALNT6 in ccRCC, and whether GALNT6‐mediated O‐glycosylation aggravates malignant behaviors. Gain‐ and loss‐of‐function experiments showed that overexpression of GALNT6 accelerated ccRCC cell proliferation, migration, and invasion, as well as promoted ccRCC‐derived xenograft tumor growth and lung metastasis. In line with this, silencing of GALNT6 yielded the opposite results. Mechanically, high expression of GALNT6 led to the accumulation of Tn antigen in ccRCC cells. By undertaking immunoprecipitation coupled with liquid chromatography/mass spectrometry, vicia villosa agglutinin blot, and site‐directed mutagenesis assays, we found that O‐glycosylation of prohibitin 2 (PHB2) at Ser161 was required for the GALNT6‐induced ccRCC cell proliferation, migration, and invasion. Additionally, we identified lens epithelium‐derived growth factor (LEDGF) as a key regulator of GALNT6 transcriptional induction in ccRCC growth and an upstream contributor to ccRCC aggressive behavior. Collectively, our findings indicate that GALNT6‐mediated abnormal O‐glycosylation promotes ccRCC progression, which provides a potential therapeutic target in ccRCC development.

INTRODUCTION

Renal cell carcinoma (RCC) is one of the malignancies that affect the urinary system and affects over 400,000 individuals worldwide annually.¹ Renal cell carcinoma consists of various subtypes with the most prevalent being clear cell RCC (ccRCC), comprising approximately 75% of RCC cases.² Even with considerable extension of survival in patients with ccRCC through surgical intervention, approximately 30% of patients relapse and present with metastases.³ Consequently, there exists a pressing demand for biomarkers that can effectively facilitate earlier diagnosis and improve the prognosis of ccRCC patients.

Glycans are biopolymers prevalent in nature that are attached to proteins or lipids through glycosylation to produce glycoproteins and glycolipids.⁴ Based on the linkage between glycans and the protein core regions, protein glycosylation is divided into categories, with O-type being one of the most common forms.⁵ In mammals, mucin-type O-glycosylation is one of the most abundant types. It is initiated through polypeptide N-acetylgalactosaminyltransferases (GALNTs, also known as GalNAcTs), which catalyze the transfer of N-acetylgalactosamine (GalNAc) from the uridine diphosphate sugar donor to Ser or Thr residues of the protein acceptor.⁶ The GalNAc α1-Ser/Thr (Tn antigen) is a mucin-type O-glycan and occurs frequently in tumor tissues, while infrequently in normal cells and tissues.⁷ Moreover, highly expressed Tn antigen is associated with a series of cancer activities, such as cell proliferation, cell-to-cell adhesion, and tumor metastasis.⁷ This means that O-glycosylation plays an important role in the process of tumor growth and metastasis. It has been known that GALNT1 overexpression confers Tn accumulation and triggered O-glycosylation initiation of oncoprotein mucin 1 (MUC1) and OPN, thereby promoting breast cancer metastasis.⁸ Silencing of GALNT14 inhibits hepatoma cell growth, migration, and resistance to anticancer drugs by decreasing generation of Tn antigen.⁹

Previously, NguyenHoang et al. found that the expression of Tn antigen in ccRCC tissues was much higher than that in para-carcinoma tissues, and ccRCC patients with high Tn-antigen expression had high Fuhrman grade, necrosis rate, and sarcomatoid risk.¹⁰ These data suggest that aberrant O-glycosylation could drive ccRCC progression. To understand the biosynthetic underpinnings of these changes, we initially focused on GALNTs known to be involved in Tn antigen synthesis using publicly available transcriptomic data retrieved from ccRCC Gene Expression Omnibus databases. Results of the GSE15641 dataset showed that GALNT6 was significantly upregulated in ccRCC tissues. Early studies revealed that GALNT6-mediated aberrant O-glycosylation stimulated hyperproliferation and metastasis of neoplastic cells, including breast cancer, ovarian cancer, pancreatic cancer, and lung cancer.^11–16 In breast tumors, GALNT6 overexpression increased O-glycosylation initiation of MUC1, thus promoting tumorigenicity and metastasis.¹³ Overexpression of GALNT6 also led to proliferation of breast cancer cells through O-glycosylation of fibronectin.¹⁷ In addition, GALNT6 overexpression facilitated epithelial–mesenchymal transition (EMT), migration, and invasion of lung adenocarcinoma cells through O-glycosylating and stabilizing chaperone protein GRP78.¹⁶ Currently, the role of GALNT6 in ccRCC remains unknown.

In this study, both gain- and loss-of-function of GALNT6 experiments were undertaken to investigate its role in the growth and metastasis of ccRCC cells and xenografted tumors. The underlying molecular mechanism of GALNT6 was further explored.

MATERIALS AND METHODS

Acquirement of microarray data

Gene expression profiling was acquired by a public functional genomics data repository, Gene Expression Omnibus (, accession number: GSE15641). Thirty-two ccRCC samples and 23 normal kidney samples from this dataset were used to analyze the expression of target genes in ccRCC.

Tissue samples

A total of 28 paired ccRCC tissues and the matched paracancerous tissues were collected from the First Affiliated Hospital of China Medical University. The study protocol was approved by the medical ethics committee of the First Affiliated Hospital of China Medical University (approval ID 2023117). No patients underwent radiotherapy or chemotherapy before surgery. All tissue samples were obtained with informed consent from patients. The fresh samples were snap-frozen in liquid nitrogen until RNA extraction and gene expression analysis.

Cell culture

Human ccRCC cell lines CAKI1 and OS-RC-2 were obtained from iCell Bioscience. CAKI1 cells were incubated in McCoy's 5A medium (Cat no. M9420; Solarbio Life Sciences) containing 10% FBS (Cat no. 11011–8611; Solelybio). OS-RC-2 cells were cultured in RPMI-1640 medium (Cat no. 31800; Solarbio Life Sciences) supplemented with 10% FBS. All cells were kept at 37°C in a humidified atmosphere of 5% CO₂.

Viral infection and plasmid transfection

For GALNT6 overexpression, its coding region sequence (CDS) fragments were inserted into pLVX-mCherry-N1 (Cat no. FH1891; Fenghui Bio) plasmid between XhoI and EcoRI sites. To silence GALNT6 and reduce sequence-specific off-target effects, five shRNAs targeting the CDS of GALNT6 were inserted into pLKO.1-mCherry-Puro (Cat no. KT2909; Fenghui Bio) plasmid between AgeI and EcoRI sites. Five shRNA target sequences are shown in Table S1. Lipofectamine transfection (Cat no. L3000015; Invitrogen) was undertaken in HEK293T cells to further package corresponding lentiviruses. Subsequently, CAKI1 and OS-RC-2 cells were infected with the above lentiviruses (MOI = 50) using Lipofectamine 3000 according to the manufacturer's protocol. Stably expressed ccRCC cells were sorted using a flow cytometer 48 h after lentivirus infection and subsequently expanded.

To determine whether prohibitin 2 (PHB2) was a substrate for GALNT6-mediated O-glycosylation in ccRCC cells, the CAKI1 cells were transiently transfected with pcDNA 3.1 (+) plasmids of a C-terminally MYC-tagged PHB2 (WT or Ser161Ala, Thr288Ala, Ser291Ala mutation) and a C-terminally 3× Flag-tagged GALNT6, using Lipofectamine 3000 following the manufacturer's protocol.

Functionally, pRNA-H1.1/Neo was used to silence PHB2 and explore whether PHB2 knockdown reversed the tumor-promoting role of GALNT6 overexpression in ccRCC. A PHB2-specific shRNA was inserted into BamHI/HindIII sites of the pRNA-H1.1/Neo plasmid. Using Lipofectamine 3000, the plasmid was transfected into CAKI1 cells following the manufacturer's protocol. In addition, we constructed pcDNA 3.1 (+)-LEDGF (lens epithelium-derived growth factor) plasmid and transfected into the GALNT6 stably expressed CAKI1 cells to assess whether LEDGF overexpression was responsible for GALNT6-mediated carcinogenesis.

Colony formation assay

A total of 300 cells were seeded onto a Petri dish and cultivated for 2 weeks. The colonies were then fixed with 4% paraformaldehyde for 25 min at room temperature and stained with Wright-Giemsa dye (Cat no. KGA227; KeyGEN BioTECH) for 3 min.

Enzyme-linked immunosorbent assay

The concentration of MMP-2 and MMP-9 in cell supernatant was measured following the instructions of corresponding ELISA kits (Cat no. EK1M02/EK1M09; Multisciences).

Animal studies

The institutional animal care and use committee at the animal research center of the First Affiliated Hospital of China Medical University approved in vivo studies, which were carried out in accordance with the National Research Council: Guide for the Care and Use of Laboratory Animals. Seven-week-old BALB/c nude mice were maintained in a standard environment (temperature of 22 ± 1°C, humidity of 45%–55% and a 12:12 h light : dark cycle) and had free access to food and water. After a week of adaptation, mice were inoculated subcutaneously with stable GALNT6-overexpressed or -silenced CAKI1 cells (5 × 10⁶ cells). Tumor volumes were measured using caliper measurements every 4 days, following the formula: (length × width²)/2 of tumor. The mice were killed at 28 days post cell injection and xenograft tumors were removed for immunohistochemistry examination. To evaluate the in vivo pulmonary metastasis of ccRCC cells, 7-week-old BALB/c nude mice were injected with stable CAKI1 cells (2 × 10⁶ cells) into the tail vein. After 28 days, pulmonary metastasis was monitored by bioluminescence imaging. Thereafter, mice were killed and the lungs were excised and photographed. The number of metastatic lung nodules was visually counted, and histological assessments were further undertaken after embedding the slices.

Immunohistochemistry

Cells or ccRCC xenografts were paraformaldehyde-fixed, dehydrated, permeabilized, paraffin-embedded, and cut into 5-μm-thick sections. The xylene-deparaffinized and rehydrated sections were subjected to heat-induced epitope retrieval. Afterward, the sections were inactivated by endogenous peroxidase using 3% H₂O₂ for 15 min, blocked by nonspecific binding using 1% BSA for 15 min at room temperature, and incubated at 4°C overnight with a rabbit anti-GALNT6 polyclonal (Cat no. Bs-16222R; Bioss) or rabbit anti-Ki-67 polyclonal (Cat no. AF0198; Affinity Biosciences) primary Ab diluted 1: 100 in PBS. Next, the sections were incubated at room temperature for 45 min with an HRP-conjugated secondary Ab (Cat no. SE134; Solarbio Life Sciences) diluted 1: 100 in PBS. Finally, the sections were developed with 3,3′-diamino-benzidine and counterstained with hematoxylin. Immunohistochemical images were taken using the Olympus BX53 microscope at ×200 magnification.

Hematoxylin and eosin staining

The ccRCC xenografts were fixed in 4% paraformaldehyde, paraffin-embedded, and cut into 5-μm-thick sections. After dewaxing and rehydration, the sections were stained with hematoxylin (Cat no. H8070; Solarbio Life Sciences) and eosin (Cat no. A600190; Sangon Biotech), followed by examination under the Olympus BX53 microscopy at ×40 magnification.

Immunofluorescence

After fixation, the cells were permeabilized with 0.1% Triton X-100 for 30 min at room temperature. Unspecific binding sites were blocked with 1% BSA for 15 min at room temperature. The cells were incubated at 4°C overnight with primary Abs or lectin diluted 1:100 in PBS. The details of primary Abs and lectin were as follows: rabbit anti-E-cadherin polyclonal (Cat no. AF0131; Affinity Biosciences), rabbit anti-N-cadherin polyclonal (Cat no. AF5239; Affinity Biosciences), rabbit anti-Ki67 polyclonal (Cat no. AF0198; Affinity Biosciences), and biotinylated vicia villosa agglutinin (VVA) lectin (specific to Tn antigen; Cat no. B-1235; Vector Laboratories). Next, the cells were incubated for 1 h at room temperature with the Cy3-labeled goat anti-rabbit IgG secondary Ab (Cat no. SA00009-2; Proteintech) or streptavidin–FITC (Cat no. 405,201; BioLegend) diluted 1:200 in PBS. Cell nuclei were counterstained with DAPI (Cat no. D106471-5 mg; Aladdin Reagent). Immunofluorescent images were captured using the Olympus BX53 microscope at ×400 magnification.

Real-time quantitative PCR

RNA was isolated from tissue samples and cells using TRIpure (Cat no. RP1001; BioTeke Bio) and processed into cDNA with BeyoRT II M-MLV reverse transcriptase (Cat no. D7160L; Beyotime Biotech) and Rnase inhibitor (Cat no. BL780A; Biosharp Life Sciences). The real-time quantitative PCR (RT-qPCR) reaction was run utilizing cDNA, SYBR Green (Cat no. SY1020; Solarbio Life Sciences), 2× Taq PCR MasterMix (Cat no. PC1150; Solarbio Life Sciences), and specific primers. The sequences of primers were as follows: GALNT6 forward, 5′-AAACCAGTCCTGCCTCC-3′; GALNT6 reverse, 5′-GCTTCTTATAGCCTTCTTCC-3′; PHB2 forward, 5′-ACAGAGCCATCTTCTTCAATC-3′; PHB2 reverse, 5′-ACTCGTTCCTCGTAGTCCAG-3′; LEDGF forward, 5′-GGGAGGAACTTTCAGACT-3′; LEDGF reverse, 5′-ATTGATGTTTCTCGCTTC-3′; β-actin forward, 5′-TCAGGGTGAGGATGCCTCTC-3′; and β-actin reverse, 5′-CTCGTCGTCGACAACGGCT-3′. The expression level of GALNT6 in tissue samples was analyzed using the $2^{-\mathrm{\Delta }{C}_{\mathrm{T}}}$ method. The expression of target genes in ccRCC cells was calculated by the $2^{-∆∆C_{\mathrm{T}}}$ method and normalized to β-actin.

Western blot analysis, immunoprecipitation, and liquid chromatography/mass spectrometry

Protein was extracted from tissue samples and cells using RIPA lysis (Cat no. PR20001; Proteintech) with 1% protease inhibitor (Cat no. PR20032; Proteintech). Protein lysates were separated on 10% or 12% SDS-PAGE and then transferred onto a PVDF membrane (Cat no. LC2005; Thermo Fisher Scientific). Unspecific binding sites were blocked with 5% nonfat dry milk in TBS containing Tween-20 (Cat no. PR20011; Proteintech) and incubated for 2 h. The membrane was incubated at 4°C overnight with a 1:500 dilution of mouse anti-GALNT6 monoclonal (Cat no. sc-100755; Santa Cruz Biotechnology Inc.), 1:5000 dilution of rabbit anti-PHB2 polyclonal (Cat no. 12295-1-AP), 1:1000 dilution of rabbit anti-LEDGF polyclonal (Cat no. 25504-1-AP), 1:5000 dilution of rabbit anti-MYC tag polyclonal (Cat no. 16286-1-AP), 1:3000 dilution of rabbit anti-Flag tag recombinant (Cat no. 80010-1-RR), 1:3000 dilution of VVA lectin (Cat no. B-1235), 1:5000 dilution of rabbit anti-cyclin D polyclonal (Cat no. 26939-1-AP), 1:1000 dilution of rabbit anti-Bcl-2 polyclonal (Cat no. 26593-1-AP), 1:5000 dilution of rabbit anti-c-myc polyclonal (Cat no. 10828-1-AP), or 1:20,000 dilution of mouse anti-β-actin monoclonal (Cat no. 66009-1-Ig) (all from Proteintech) primary Ab in 5% nonfat dry milk. Subsequently, the membrane was incubated at 37°C for 40 min with species-specific HRP-conjugated secondary Ab (Cat no. SA00001, diluted 1:10000 in 5% nonfat dry milk; Proteintech) or HRP-conjugated streptavidin (Cat no. SA00001-0, diluted 1:5000 in 5% nonfat dry milk; Proteintech). After washing, the protein bands were visualized using an enhanced chemiluminescence kit (Cat no. PK10003; Proteintech).

For the immunoprecipitation (IP) assay, cells were lysed using RIPA lysis with 1% protease inhibitor, followed by IP with AminoLink Plus Coupling Resin (Cat no. 26149; Thermo Fisher Scientific) conjugated with IP-specific Abs or VVA lectin (Cat no. B-1235). The details of IP-specific Abs were as follows: anti-GALNT6 (Cat no. sc-100755), anti-PHB2 (Cat no. 12295-1-AP), anti-MYC tag (Cat no. 16286-1-AP), and anti-Flag tag (Cat no. 80010-1-RR). The immunoprecipitated proteins were then analyzed by western blot assay as described above.

Immunoprecipitation followed by liquid chromatography/mass spectrometry (IP-LC/MS) technology was used to identify other endogenous proteins that bind with GALNT6. The GALNT6-interacting complexes were collected by IP assay as described above. Subsequently, LC/MS analysis and protein identification were undertaken by Qinglian Baiao BioTECH.

Dual-luciferase reporter assay

The JASPAR 2024 online database () revealed putative LEDGF binding sites in the promoter region of GALNT6. To confirm the transcriptional regulation of LEDGF on GALNT6, a dual-luciferase reporter assay was performed in CAKI1 cells. Briefly, the cells were seeded onto 12-well plates and incubated to reach 70% confluence. The cells were then cotransfected with LEDGF overexpression or pcDNA 3.1 (+) plasmid, pGL3-basic reporter vector carrying GALNT6 promoter fragment (−2000 ~ +15 bp) or empty reporter vector, and pRL-TK vector. Forty-eight hours after transfection, luciferase activity was measured using a Dual-Luciferase Reporter Assay kit (Cat no. KGAF040; KeyGEN BioTECH) according to the manufacturer's protocol. Luciferase activity was normalized to Renilla luciferase activity.

Cell proliferation assay

Cells were seeded onto 96-well plates at a density of 5 × 10³ per well and incubated for an indicated time at 37°C. Next, 50 μL MTT solution (Cat no. KGA311; KeyGEN BioTECH) was added to each well and incubated for 4 h. The produced formazan crystals were dissolved in 150 μL DMSO. The absorbance at 490 nm was measured by a microplate reader (800TS; BioTek Instruments). Higher formazan absorbance values indicated stronger cell proliferation ability.

Transwell migration and invasion assays

For migration assay, 8-μm-pore Transwell plastic inserts (Cat no. 14341; Labselect) were used to create a two-chamber system. Cells (5000 per well) suspended in 200 μL serum-free culture medium were put into the upper chamber, while 800 μL complete growth medium containing 10% FBS attractant was added to the lower chamber in a 24-well plate. For the invasion assay, Transwell inserts precoated with Matrigel (Cat no. 356234; Corning Inc.) were utilized to mimic the ECM (3 × 10⁵ per well). After 48 h, cells that migrated or invaded to the opposite side of the inserts were fixed in 4% paraformaldehyde for 25 min and stained with 0.5% crystal violet for 5 min. Migratory and Invasive cells were counted under an inverted microscope (BX53; Olympus) at ×200 magnification.

Statistical analysis

Statistical analysis was CARRIED OUT using GraphPad Prism 8.2.0. Data had passed the Shapiro–Wilk normality test and homogeneity test of variance. An unpaired two-tailed Student's t-test was used for comparisons between two groups. For multiple group comparisons, one-way or two-way ANOVA was applied with a Tukey post-test. Kaplan–Meier analysis followed by a log-rank test was used to analyze the effect of GALNT6 on overall survival in ccRCC patients with advanced TNM stages. Results are expressed as the mean ± SD. The results were considered significantly different if p values ≤0.05.

RESULTS

GALNT6 is highly expressed in ccRCC tissues and its overexpression accelerates proliferation of ccRCC cells

To investigate the role of GALNT6 in ccRCC, we first analyzed 32 ccRCC tissues and 23 normal kidney tissues using publicly available transcriptomic data retrieved from the Gene Expression Omnibus database (accession: GSE15641). We identified GALNT6 as upregulated in ccRCC tissues (Figure 1A). To validate this result, we detected the mRNA expression level of GALNT6 in 28 paired ccRCC tissues and para-carcinoma tissues in our hospital. Consistently, GALNT6 was an upregulated gene in tumor tissues (Figure 1B). Western blotting indicated the upregulation of GALNT6 protein in three pairs of ccRCC tissues and para-carcinoma tissues, which was also consistent with transcriptional expression (Figure 1C). GALNT6 stable overexpression or silencing ccRCC lines were constructed using lentivirus RNA interference or overexpression systems. The expression of GALNT6 was confirmed using RT-qPCR and western blot analyses (Figure S1). In vitro gain-of-function and loss-of-function experiments were undertaken to explore the effect of GALNT6 on ccRCC cell growth. As revealed in Figure 1D, GALNT6 overexpression facilitated the proliferation of ccRCC cells. GALNT6 overexpression also enhanced the capacity of colony formation in ccRCC cells (Figure 1E). In addition, increased Ki-67 fluorescent intensity in ccRCC cells was observed after GALNT6 overexpression (Figure 1F). In line with this, silencing of GALNT6 significantly decreased proliferation, colony formation, and Ki-67 intensity in ccRCC cells (Figure 1D–F).

[IMAGE OMITTED. SEE PDF]

GALNT6 overexpression facilitates migration and invasion of ccRCC cells

GALNT6 has been implicated in various aspects of tumor biology including cell migration and invasion.^11,18 Therefore, in vitro Transwell assays were carried out to assess the role of GALNT6 in these processes. As shown in Figure 2A, high GALNT6 expression enhanced the migratory ability and invasiveness of ccRCC cells. Silencing of GALNT6 resulted in opposite alterations in ccRCC cells. Furthermore, results of ELISA revealed that MMP-2 and MMP-9 levels in ccRCC cells were increased after GALNT6 overexpression, yet decreased after GALNT6 silencing (Figure 2B). Based on the above findings, we hypothesized that GALNT6 might regulate ccRCC cell migration and invasion through EMT progress, a well-known mechanism by which cancer cells become more motile and able to invade by acquiring mesenchymal characteristics.¹⁹ GALNT6 overexpression downregulated the expression of E-cadherin (an epithelial marker) and upregulated the expression of N-cadherin (a mesenchymal marker) in ccRCC cells (Figure 2C). In line with this, the opposite effects were observed following GALNT6 silencing (Figure 2C).

[IMAGE OMITTED. SEE PDF]

In vivo ccRCC xenograft model confirms pro-tumorigenic and pro-metastatic roles of GALNT6 overexpression

Mice were subcutaneously injected with GALNT6 stably overexpressed or silenced CAKI1 cells. Consistent with our in vitro findings, the tumors of subcutaneously inoculating GALNT6-overexpressed cells had a faster growth rate in comparison to inoculating vector cells (Figure 3A). Furthermore, the tumor volume of inoculating GALNT6-overexpressed CAKI1 cells was larger than vector tumors on day 28 post injection (Figure 3B). More GALNT6- and Ki-67-positive cells in tumors formed by GALNT6-overexpressed CAKI1 cells also confirmed that high GALNT6 expression resulted in the enhancement of proliferative capability (Figure 3C,D). GALNT6-silenced CAKI1 cells yielded the opposite results in vivo (Figure 3A–D). Furthermore, fluorescence intensity and number of metastatic lung nodules were increased following GALNT6 overexpression, yet decreased following GALNT6 silencing (Figure 3E,F). Results of histological examination of H&E staining showed that GALNT6 overexpression resulted in destruction of lung parenchyma within the metastases (Figure 3G). Consistently, the normal lung parenchyma was preserved after GALNT6 silencing (Figure 3G).

[IMAGE OMITTED. SEE PDF]

PHB2 is a substrate for GALNT6-mediated O-glycosylation in ccRCC cells

Immunofluorescence analysis was carried out using VVA lectins, which specifically recognized Tn antigens that frequently occurred in ccRCC progression.¹⁰ As shown in Figure 4A, Tn antigen in ccRCC cells was accumulated following GALNT6 overexpression, but reduced following GALNT6 silencing. Using IP-LC/MS analysis, a total of 1012 specific binding proteins were identified to be potential GALNT6-interacting partners in CAKI1 cells (Figure 4B). Previous studies reported that PHB2 was a tumor promoter in ccRCC progression and regulated by O-glycosylation.^9,20 Therefore, PHB2 was chosen as a protein of interest to us. The binding of PHB2 to GALNT6 was validated in two ccRCC cells by co-IP (Figure 4C). To confirm whether GALNT6 modified O-glycosylation of PHB2, cell lysates were immunoprecipitated for PHB2 in CAKI1 cells with GALNT6 overexpression or silencing. As shown in Figure 4D, immunoprecipitated PHB2 manifested strengthened Tn antigen in GALNT6-overexpressed CAKI1 cells, and presented reduced Tn antigen in GALNT6-silenced CAKI1 cells. Early evidence showed three potential O-glycosylation sites (Ser161, Thr288, and Ser291) in PHB2, and demonstrated that O-glycosylation at Ser161 of PHB2 was required for the GALNT14-mediated growth-promoting phenotype in hepatoma cells.⁹ Here, we determined that Ser161 was a GALNT6-mediated O-glycosylation site in ccRCC cells by site-directed mutagenesis experiments. In GALNT6-overexpressed CAKI1 cells, just PHB2 with the Ser161Ala point mutation weakened its ability to bind with VVA, not the Thr288Ala or Ser291Ala (Figure 4E).

[IMAGE OMITTED. SEE PDF]

Silencing of PHB2 inhibits GALNT6 overexpression-induced proliferation, migration, and invasion of ccRCC cells

Next, we investigated whether PHB2 mediated the role of GALNT6 in ccRCC. Both RT-qPCR and western blot analyses confirmed an efficient knockdown of PHB2 in CAKI1 cells after transfection of PHB2 shRNA (Figure S2A). The following functional experiments revealed that silencing of PHB2 inhibited GALNT6 overexpression-induced viability and motility of CAKI1 cells (Figure 5).

[IMAGE OMITTED. SEE PDF]

LEDGF is responsible for GALNT6-mediated ccRCC progression

Early evidence indicated that high LEDGF expression in cancer was closely linked to tumor growth, invasion, and poor prognosis,²¹ but the role of LEDGF in ccRCC progression remains unknown. The GSE15641 dataset revealed that LEDGF was highly expressed in ccRCC tissues, compared with that in normal tissues (Figure 6A). Therefore, we speculated that LEDGF might act as an oncogenic factor in ccRCC. To test the hypothesis, LEDGF overexpression was carried out in CAKI1 cells by transient transfection of pcDNA 3.1 (+)-LEDGF plasmid (Figure 6B). By the follow-up functional experiment, we observed the pro-proliferative, pro-migratory, and pro-invasive roles of LEDGF overexpression in CAKI1 cells (Figure 6C,D). The dual-luciferase reporter assay confirmed that LEDGF overexpression enhanced the luciferase activity of GALNT6 in CAKI1 cells (Figure 6E). A subsequent rescue assay was used to examine whether LEDGF was responsible for GALNT6-mediated ccRCC development. As shown in Figure 6F,G, LEDGF overexpression significantly reversed antiproliferative, antimigratory, and anti-invasive roles of GALNT6 silencing in CAKI1 cells.

[IMAGE OMITTED. SEE PDF]

DISCUSSION

In the current study, we found that GALNT6 was upregulated in ccRCC tissues of patients with advanced TNM stage. High GALNT6 expression was associated with poor prognosis in ccRCC patients with advanced TNM stage. Functionally, GALNT6 overexpression accelerated ccRCC cell proliferation, migration, and invasion. GALNT6 overexpression promoted ccRCC-derived xenograft tumor growth and lung metastasis. In line with this, silencing of GALNT6 showed the opposite results. Mechanically, GALNT6 overexpression induced ccRCC malignant phenotypes and increased Tn antigen expression by promoting aberrant glycosylation and stabilization of PHB2. In addition, the transcription factor LEDGF was identified to function as an oncogene linked to GALNT6-mediated ccRCC progression.

Aberrant expression of O-glycans (such as Tn antigen) yielded tumor-associated glycosylated proteins with the aid of GALNTs/GalNAcTs, which could affect differentiation, adhesiveness, and invasion of various cancer cells.²² As a glycosylated protein, MUC1 is upregulated in ccRCC. Furthermore, overexpression of MUC1 enhances the invasive and migratory properties of ccRCC.²³ It has been known that overexpression of GALNT6 can induce glycosylation and stabilization of MUC1, thereby contributing to mammary carcinogenesis.²⁴ Thus, GALNT6 could be a marker for aberrant O-linked glycans in ccRCC carcinogenesis. The in vivo and in vitro experiments indicated that overexpression of GALNT6 facilitates ccRCC cell proliferation, migration, and invasion, and accelerates xenograft tumor growth and metastasis. However, we noted that another glycosyltransferase, GALNT3, promotes glycosylation of E-cadherin and suppresses the mesenchymal state in ccRCC.²⁵ Interestingly, GALNT3 overexpression contributes to colorectal cancer progression by regulating O-glycosylated MUC1.²⁶ As GALNT6 shows high homology in DNA and amino-acid sequence to GALNT3 throughout the coding region,^27,28 the reason for the functional difference between both GALNTs in ccRCC is worth pursuing further.

Prohibitin 2, also called BCAP37 (B-cell receptor associate protein 37) or REA (repressor of estrogen receptor activity), is a highly conserved and ubiquitous protein that is located in several subcellular components, such as nucleus, mitochondrion, plasma membrane, and cytosol. PHB2 has been shown to play a vital role in modulating signal transduction, cell survival, cell cycle, cell apoptosis, and mitochondrial structural integrity.²⁹ Increasing evidence reveals that PHB2 is commonly upregulated in certain solid tumors, including esophageal squamous cell carcinoma,³⁰ rhabdomyosarcoma,³¹ non-small-cell lung cancer,³² hepatocellular carcinoma (HCC),³³ and colorectal cancer.³⁴ Consistent with previous studies, high levels of PHB2 were recently found in RCC tissues, and this overexpression is associated with a worse overall survival for RCC patients.²⁰ Moreover, PHB2 depletion leads to RCC cell proliferation inhibition, cycle arrest, and metastatic suppression by blunting eIF4E-mediated translation of oncogenic proteins, including c-myc, cyclin D1, and Bcl-2.²⁰ A recent study suggests that O-glycosylation at Ser161 of PHB2 mediated by GALNT14 contributes to HCC cell growth, migration, and resistance to anticancer drugs,⁹ indicating that aberrant O-glycosylation of PHB2 could participate in tumor malignancy progression. In this study, we determined that high GALNT6 expression accelerated the malignant progression of ccRCC through O-glycosylation at Ser161 of PHB2. Moreover, we noted that silencing of PHB2 reversed GALNT6 overexpression-caused upregulation of c-myc, cyclin D1, and Bcl-2 (Figure S3A). Such evidence suggests that the tumor-promoting effects of GALNT6 were dependent on PHB2.

GALNT6 expression was regulated by circ-hnRNPU at the transcriptional level in gastric cancer cells.³⁵ The current study indicated that GALNT6 expression was controlled by the transcription factor LEDGF in ccRCC cells. Moreover, LEDGF was responsible for GALNT6-mediated ccRCC cell proliferation, migration, and invasion. LEDGF/p75, a leukemia-specific target, has been identified to be essential for the development of solid tumors. For example, silencing of LEDGF induced a caspase-independent lysosomal cell death pathway in cancer cells.³⁶ Silencing of LEDGF inhibited prostate cancer cell proliferation, invasiveness, and migration, and induced apoptosis by downregulating heat-shock protein 27 through interaction with stress response elements (STREs) (nA/TGGGGA/Tn).²¹ In H1299 human lung cancer cell- and C6 rat glioma cell-derived mouse tumor xenografts, LEDGF bound to a conserved STRE located in the promoter of vascular endothelial growth factor C and stimulated its expression, thereby augmenting angiogenesis and lymphangiogenesis.³⁷ Another study reported that LEDGF tethers Myc-interacting protein JPO2 to chromatin, affecting transcription in an STRE-independent manner.³⁸ In this study, we provide evidence that LEDGF overexpression enhanced the transactivation of GALNT6 promoter and increased the migration and invasiveness of CAKI1 cells. Knockdown of GALNT6 reversed LEDGF overexpression-caused carcinogenesis.

In conclusion, we clarified that the underlying mechanism involved in silencing of GALNT6 curbed carcinogenesis by promoting aberrant glycosylation and stabilization of PHB2 in human ccRCC cells. Certainly, more work is needed for a full understanding of the role of GALNT6 in ccRCC. It is necessary to evaluate clinicopathologic correlations between GALNT6 expression and ccRCC patients in further studies to support our conclusion. Here, our study identifies that GALNT6 is a potential therapeutic target of ccRCC progression.

AUTHOR CONTRIBUTIONS

Luhaoran Sun: Conceptualization; investigation; writing – original draft. Zeyu Li: Investigation. Peng Shu: Investigation. Min Lu: Funding acquisition; writing – review and editing.

ACKNOWLEDGMENTS

None.

FUNDING INFORMATION

This study was funded by the High-quality Development Fund Project of Science and Technology of China Medical University (Grant No. 2023JH2/20200023).

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest.

ETHICS STATEMENTS

Approval of the research protocol by an institutional review board: The study protocol was approved by the medical ethics committee of the First Affiliated Hospital of China Medical University (approval ID 2023117).

Informed consent: All tissue samples were obtained with informed consent from patients.

Registry and the registration no. of the study/trial: N/A.

Animal studies: The institutional animal care and use committee at the animal research center of the First Affiliated Hospital of China Medical University approved in vivo studies (approval ID cmu20240022), which were performed in accordance with the National Research Council: Guide for the Care and Use of Laboratory Animals.

References

Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2018;68(6):394‐424.

Rini BI, Campbell SC, Escudier B. Renal cell carcinoma. Lancet. 2009;373(9669):1119‐1132.

Jonasch E, Gao J, Rathmell WK. Renal cell carcinoma. BMJ. 2014;349: [eLocator: g4797].

Pang X, Li H, Guan F, Li X. Multiple roles of Glycans in hematological malignancies. Front Oncol. 2018;8:364.

Zhu X, Al‐Danakh A, Zhang L, et al. Glycosylation in renal cell carcinoma: mechanisms and clinical implications. Cells. 2022;11(16):2598.

Bennett EP, Mandel U, Clausen H, Gerken TA, Fritz TA, Tabak LA. Control of mucin‐type O‐glycosylation: a classification of the polypeptide GalNAc‐transferase gene family. Glycobiology. 2012;22(6):736‐756.

Fu C, Zhao H, Wang Y, et al. Tumor‐associated antigens: Tn antigen, sTn antigen, and T antigen. HLA. 2016;88(6):275‐286.

Fang R, Xu F, Shi H, et al. LAMTOR5 raises abnormal initiation of O‐glycosylation in breast cancer metastasis via modulating GALNT1 activity. Oncogene. 2020;39(11):2290‐2304.

Chu YD, Fan TC, Lai MW, Yeh CT. GALNT14‐mediated O‐glycosylation on PHB2 serine‐161 enhances cell growth, migration and drug resistance by activating IGF1R cascade in hepatoma cells. Cell Death Dis. 2022;13(11):956.

NguyenHoang S, Liu Y, Xu L, et al. High truncated‐O‐glycan score predicts adverse clinical outcome in patients with localized clear‐cell renal cell carcinoma after surgery. Oncotarget. 2017;8(45):80083‐80092.

Liu C, Li Z, Xu L, et al. GALNT6 promotes breast cancer metastasis by increasing mucin‐type O‐glycosylation of alpha2M. Aging (Albany NY). 2020;12(12):11794‐11811.

Kimura R, Yoshimaru T, Matsushita Y, et al. The GALNT6‐LGALS3BP axis promotes breast cancer cell growth. Int J Oncol. 2020;56(2):581‐595.

Mao Y, Zhang Y, Fan S, et al. GALNT6 promotes Tumorigenicity and metastasis of breast cancer cell via beta‐catenin/MUC1‐C signaling pathway. Int J Biol Sci. 2019;15(1):169‐182.

Lin TC, Chen ST, Huang MC, et al. GALNT6 expression enhances aggressive phenotypes of ovarian cancer cells by regulating EGFR activity. Oncotarget. 2017;8(26):42588‐42601.

Tarhan YE, Kato T, Jang M, et al. Morphological changes, cadherin switching, and growth suppression in pancreatic cancer by GALNT6 knockdown. Neoplasia. 2016;18(5):265‐272.

Song J, Liu W, Wang J, et al. GALNT6 promotes invasion and metastasis of human lung adenocarcinoma cells through O‐glycosylating chaperone protein GRP78. Cell Death Dis. 2020;11(5):352.

Park JH, Katagiri T, Chung S, Kijima K, Nakamura Y. Polypeptide N‐acetylgalactosaminyltransferase 6 disrupts mammary acinar morphogenesis through O‐glycosylation of fibronectin. Neoplasia. 2011;13(4):320‐326.

Ding M, Liu J, Lv H, et al. Knocking down GALNT6 promotes pyroptosis of pancreatic ductal adenocarcinoma cells through NF‐kappaB/NLRP3/GSDMD and GSDME signaling pathway. Front Oncol. 2023;13: [eLocator: 1097772].

Guarino M, Rubino B, Ballabio G. The role of epithelial‐mesenchymal transition in cancer pathology. Pathology. 2007;39(3):305‐318.

Yang J, Li G, Huang Y, Liu Y. Decreasing expression of Prohibitin‐2 lowers the oncogenicity of renal cell carcinoma cells by suppressing eIF4E‐mediated oncogene translation via MNK inhibition. Toxicol Appl Pharmacol. 2023;466: [eLocator: 116458].

Bhargavan B, Fatma N, Chhunchha B, Singh V, Kubo E, Singh DP. LEDGF gene silencing impairs the tumorigenicity of prostate cancer DU145 cells by abating the expression of Hsp27 and activation of the Akt/ERK signaling pathway. Cell Death Dis. 2012;3(5): [eLocator: e316].

Andergassen U, Liesche F, Kolbl AC, et al. Glycosyltransferases as markers for early tumorigenesis. Biomed Res Int. 2015;2015: [eLocator: 792672].

Aubert S, Fauquette V, Hemon B, et al. MUC1, a new hypoxia inducible factor target gene, is an actor in clear renal cell carcinoma tumor progression. Cancer Res. 2009;69(14):5707‐5715.

Park JH, Nishidate T, Kijima K, et al. Critical roles of mucin 1 glycosylation by transactivated polypeptide N‐acetylgalactosaminyltransferase 6 in mammary carcinogenesis. Cancer Res. 2010;70(7):2759‐2769.

Shu G, Lu X, Pan Y, et al. Exosomal circSPIRE1 mediates glycosylation of E‐cadherin to suppress metastasis of renal cell carcinoma. Oncogene. 2023;42(22):1802‐1820.

Liu B, Pan S, Xiao Y, Liu Q, Xu J, Jia L. LINC01296/miR‐26a/GALNT3 axis contributes to colorectal cancer progression by regulating O‐glycosylated MUC1 via PI3K/AKT pathway. J Exp Clin Cancer Res. 2018;37(1):316.

Schwientek T, Bennett EP, Flores C, et al. Functional conservation of subfamilies of putative UDP‐N‐acetylgalactosamine:polypeptide N‐acetylgalactosaminyltransferases in drosophila, Caenorhabditis elegans, and mammals. One subfamily composed of l(2)35Aa is essential in drosophila. J Biol Chem. 2002;277(25):22623‐22638.

Ten Hagen KG, Fritz TA, Tabak LA. All in the family: the UDP‐GalNAc:polypeptide N‐acetylgalactosaminyltransferases. Glycobiology. 2003;13(1):1R‐16R.

Bavelloni A, Piazzi M, Raffini M, Faenza I, Blalock WL. Prohibitin 2: at a communications crossroads. IUBMB Life. 2015;67(4):239‐254.

Cai XW, Yu WW, Yu W, et al. Tissue‐based quantitative proteomics to screen and identify the potential biomarkers for early recurrence/metastasis of esophageal squamous cell carcinoma. Cancer Med. 2018;7(6):2504‐2517.

Fu P, Yang Z, Bach LA. Prohibitin‐2 binding modulates insulin‐like growth factor‐binding protein‐6 (IGFBP‐6)‐induced rhabdomyosarcoma cell migration. J Biol Chem. 2013;288(41):29890‐29900.

Wu B, Chang N, Xi H, et al. PHB2 promotes tumorigenesis via RACK1 in non‐small cell lung cancer. Theranostics. 2021;11(7):3150‐3166.

Cheng J, Gao F, Chen X, et al. Prohibitin‐2 promotes hepatocellular carcinoma malignancy progression in hypoxia based on a label‐free quantitative proteomics strategy. Mol Carcinog. 2014;53(10):820‐832.

Ren L, Meng L, Gao J, et al. PHB2 promotes colorectal cancer cell proliferation and tumorigenesis through NDUFS1‐mediated oxidative phosphorylation. Cell Death Dis. 2023;14(1):44.

Li H, Jiao W, Song J, et al. Circ‐hnRNPU inhibits NONO‐mediated c‐Myc transactivation and mRNA stabilization essential for glycosylation and cancer progression. J Exp Clin Cancer Res. 2023;42(1):313.

Daugaard M, Kirkegaard‐Sorensen T, Ostenfeld MS, et al. Lens epithelium‐derived growth factor is an Hsp70‐2 regulated guardian of lysosomal stability in human cancer. Cancer Res. 2007;67(6):2559‐2567.

Cohen B, Addadi Y, Sapoznik S, et al. Transcriptional regulation of vascular endothelial growth factor C by oxidative and thermal stress is mediated by lens epithelium‐derived growth factor/p75. Neoplasia. 2009;11(9):921‐933.

Maertens GN, Cherepanov P, Engelman A. Transcriptional co‐activator p75 binds and tethers the Myc‐interacting protein JPO2 to chromatin. J Cell Sci. 2006;119(Pt 12):2563‐2571.