Promoter hypermethylation of LGALS4 correlates with poor prognosis in patients with urothelial carcinoma

Galectine-4 (gal-4), encoded by the LGALS4 gene, was recently shown to exhibit a tumor suppressive effect in colorectal carcinoma and pancreatic adenocarcinoma, although how the expression of this gene is regulated remains unknown. No reports describe the significance of gal-4 in the malignant potential of urothelial tumors. Thus, we analyzed LGALS4 methylation and gene expression and their clinical relevance and biological function in urothelial carcinoma (UC). LGALS4 methylation was initially identified as a progression biomarker for UC patients through genome-wide DNA methylation profiling of 16 tumor samples. Bisulfite sequencing PCR and immunohistochemistry were performed to validate the promoter methylation and expression of LGALS4. We used quantitative methylation-specific PCR to determine the methylation levels of LGALS4 normalized to ACTB in the tumor samples of 79 UC patients and compared the levels between patients with different clinicopathological characteristics. The association with survival probability was analyzed with the Kaplan-Meier method and Cox regression analysis. The ectopic expression of gal-4 in cancer cell lines was used to address its biological function in UC in vitro. The promoter hypermethylation of LGALS4 (>2.51, log10 scale) revealed a positive correlation with high levels of both histological grade and tumor T category and with lymph node metastasis (all P≤0.001). In addition, LGALS4 hypermethylation was an independent predictor of inferior survival in UC patients (P<0.05). The ectopic expression studies demonstrated that gal-4 suppressed urothelial cancer cell growth, migration, and invasion. Thus, LGALS4 may function as a tumor suppressor gene in UC progression. Our findings provide evidence that methylation-mediated LGALS4 gene repression may be involved in urothelial tumor progression.


DNA extraction
UC frozen tissues were lysed with 1.0 mL DNAzol reagent (Molecular Research Center, Cincinnati, OH, USA) and 200 μg/mL proteinase K (Sigma, St. Louis, MO, USA) by gently pipetting. A 0.5-mL aliquot of chloroform was added to the cell lysate. This mixture was shaken vigorously and then centrifuged at 12,000 rpm for 5 min. After centrifugation, the supernatant was collected and 1 mL 75% ethanol was added. The tube was agitated vigorously until the DNA pellet was completely dispersed. The resulting mixture was centrifuged at 12,000 rpm for 5 min at room temperature. Ethanol wash was decanted carefully and the tubes were stored vertically for approximately 1 min. The DNA pellet was dissolved in 50 μL 8 mM NaOH and incubated at room temperature for 3-5 min, followed by repetitive pipetting. DNA was quantified with NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA), with 1 μL of sample at 260 nm wavelength for determination of double-stranded DNA.

Immunohistochemistry
Protein expression of LGALS4 in FFPE specimens was determined using an immunohistochemical assay based on the avidin-biotin-peroxidase method. Paraffin sections, 5-μm thick on microscope glass slides, were deparaffinized in xylene and rehydrated in graded ethanol and water. Endogenous peroxidase activity was then blocked by incubation with peroxidase blocking reagent containing 3% H 2 O 2 (DAKO Corp, Carpinteria, CA, USA) according to the manufacturer's protocol. After heat-induced antigen retrieval, non-specific protein-binding sites and any sites with affinity for biotin were saturated. The sections were incubated with gal-4 primary antibody solution (Santa Cruz Biotechnology, Santa Cruz, CA, USA) at a final dilution of 1:100 for overnight at 4°C. The slides were washed and incubated with biotinylated secondary antibodies, and the bound markers were visualized using avidin-biotin-peroxidase complex kit reagents with 3, 3'-diaminobenzidine (DAB) in 0.03% H 2 O 2 solution as the chromogenic substarte (DAKO). Slides were counterstained with hematoxylin, mounted, and coverslipped. Control sections were treated with PBS to exclude antigenindependent reactions.

RNA extraction and reverse transcriptase-PCR (RT-PCR)
Total RNA from cultured cells was extracted using TRI reagent (Molecular Research Center), according to the manufacturer's protocol. For RT-PCR analysis, total RNA was first incubated with RNase-free DNaseI at 37°C for 30 min to digest residual genomic DNA (Roche Diagnostics, Foster City, CA, USA), and 2 μg of resulting RNA was then reverse transcribed (RT) to synthesize cDNA for PCR amplification. RT reaction was carried out for 50 min at 50°C in a mixture of RT buffer (50 mM Tris/HCl, pH8.3, 75 mM KCl, 3 mM MgCl 2 ), 10 mM DTT, 10 mM each dNTPs, 0.5 μg random hexamer primers and 20 U SSRTII, and terminated by an incubation for 15 min at 70°C. Subsequently, PCR amplification was carried out in 20 μL reaction mixture, containing 2 μL cDNA, 0.5 U Tag polymerase, 50 mM KCl, 10 mM Tris/HCl (pH8.8), 1.5 mM MgCl 2 , 50 μM each dNTPs, and 0.1 μM each paired primers. Nucleotide sequences for primer set were GCT CAA CGT GGG AAT GTC TGT (sense), and GAG CCC ACC TTG AAG TTG ATA (antisense). PCR condition was run as heating the samples to 94°C for 4 min, followed by 36 cycles of 94°C for 1 min, 60°C for 1 min, and 72°C for 2 min, and ending with an extension period of 10 min at 72°C. The PCR products were separated in 2% TAE (or 1% agarose) gels, stained by ethidium bromide, and visualized under UV light. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was amplified and included in each gel electrophoresis run as an internal control.

5-aza-2'-deoxycytidine (5-aza-dC) and trichostatin A (TSA) treatment
We examined gene expression of LGALS4 in several human urothelial cell lines exposed to 5-aza-dC (Sigma-Aldrich), a DNA methyltransferase inhibitor, and/or TSA (Sigma-Aldrich), a histone deacetylase inhibitor, to examine the potential involvement of epigenetic silencing in LGALS4 repression. Cells were grown to a 25% confluency in 60-mm Petri dish on day 1. Cells were then cultured for another 72 h (day 2 ~ day 4) while replacing with fresh medium containing 1 μM 5-aza-dC or DMSO (as a negative control) every 24 h. TSA was added on day 3 at a concentration of 0.1 μM and cultured for 24 h. On day 4 after treatments, the cells were harvested for the extraction of RNA being used in subsequent RT-PCR analysis.

Western blotting analysis
Western blotting analysis was performed as described previously [1][2]. In brief, cells were harvested and lysed at 0°C in RIPA lysis buffer containing protease inhibitors (Thermo Fisher Scientific). After quantitation with the Bradford method (Bio-Rad Laboratory, Hercules, CA, USA), equal amounts of cell lysates were separated by SDS-PAGE on 10% polyacrylamide gel, and the proteins were transferred to PVDF membranes (GE Healthcare, Munich, Germany). After 1 h blocking, the membranes were washed and incubate overnight at 4°C with primary antibodies against gal-4 (Santa Cruz Biotechnology), and β-actin (Abcam, Cambridge, MA, USA) as a loading control. After incubation, secondary horseradish peroxidase-conjugated antibodies (Abcam) were used in PBST/1% BSA for an incubation of 60 min at room temperature. Bound antibodies were visualized using the SuperSignal West Pico Chemiluminescent Substrate (Pierce Biotechnology, Rockford, IL, USA). A total of 15 CpG sites within the promoter of LGALS4 spanning -252 to +184 nt. were determined. The first two CpG dinucleotides were analyzed in qMSP assay including the significant loci -194 nt. identified from analysis results by the Infinium Methylation 27K BeadChip assay. PBMC: peripheral blood mononuclear cells.