Down-regulation of SFRP1 as a putative tumor suppressor gene can contribute to human hepatocellular carcinoma

Abstract

Background: Hepatocellular carcinoma (HCC) is one of the most common cancers in the world. SFRP1 (the secreted frizzled-related protein 1), a putative tumor suppressor gene mapped onto chromosome 8p12-p11.1, the frequent loss of heterozygosity (LOH) region in human HCC, encodes a Wingless-type (Wnt) signaling antagonist and is frequently inactivated by promoter methylation in many human cancers. However, whether the down-regulation of SFRP1 can contribute to hepatocarcinogenesis still remains unclear.

Methods: We investigated the expression of SFRP1 through real time RT-PCR and immunohistochemistry staining. The cell growth and colony formation were observed as the overexpression and knockdown of SFRP1. The DNA methylation status within SFRP1 promoter was analyzed through methylation-specific PCR or bisulphate-treated DNA sequencing assays. Loss of heterozygosity was here detected with microsatellite markers.

Results: SFRP1 was significantly down-regulated in 76.1% (35/46) HCC specimens at mRNA level and in 30% (30/100) HCCs indicated by immunohistochemistry staining, as compared to adjacent non-cancerous livers. The overexpression of SFRP1 can significantly inhibit the cell growth and colony formation of YY-8103, SMMC7721, and Hep3B cells. The RNA interference against the constitutional SFRP1 in the offspring SMMC7721 cells, which were stably transfected by ectopic SFRP1, can markedly promote cell growth of these cells. LOH of both microsatellite markers D8S532 and D8SAC016868 flanking the gene locus was found in 13% (6 of 46 HCCs) and 6.5% (3 of 46 HCCs) of the informative cases, respectively, where 5 of 8 HCC specimens with LOH showed the down-regulation of SFRP1. DNA hypermethylation within SFRP1 promoter was identified in two of three HCC specimens without SFRP1 expression. Moreover, the DNA methylation of SFRP1 promoter was significantly reduced, along with the re-expression of the gene, in those HCC cell lines, Bel7404, QGY7701, and MHCC-H, as treated by DAC.

Conclusion: Our data suggested that the down-regulation of SFRP1 as a candidate tumor suppressor gene, triggered by the epigenetic and/or genetic events, could contribute to the oncogenesis of HCC.

Figure 1

Expression pattern of SFRP1 in HCC specimens by RT-PCR and immunohistochemical staining. (A) Representative results of semi-quantitative RT-PCR of SFRP1 from 24 pairs of HCCs (C) and corresponding adjacent non-cancerous livers (N), where β-actin was employed as an internal control. Each PCR was generally performed in 32 thermal cycles and PCR products were visualized after electrophoresis through 2% agarose. The length of PCR product of SFRP1 and β-actin are 328 bp and 295 bp, respectively. (B) To confirm the absence of influences of genomic DNA contamination, 6 Paired HCCs and non-HCCs were selected randomly to detect the SFRP1 and beta-actin expression by RT-PCR. Experiments were performed by using RT (RT+) or no RT (RT-) in each sample. (C) Real time RT-PCR analysis of SFRP1 was carried out on 46 paired HCCs and adjacent non-cancerous livers. For each sample, the relative mRNA level of SFRP1 was normalized based on that of β-actin. The line within each box represents the median -ΔCt value; the upper and lower edges of each box represent the 75th and 25th percentile, respectively; the upper and lower bars indicate the highest and lowest values determined, respectively. * indicates p value <0.001. (D) Representative immunohistochemical staining of a pair of HCC specimen and corresponding non-cancerous liver with anti-SFRP1 antibody. The nuclei were countered stained with hematoxylin. (E) Expression pattern of SFRP1 was evaluated in HCC cell lines, fetal and adult normal livers through RT-PCR, where β-actin was used as a loading control.

Figure 2

The overexpression of SFRP1 can significantly inhibit the cell growth and colony formation of HCC cells. (A) Plasmid pcDNA3.0-SFRP1 was transiently transfected into YY-8103 cells, confirmed by immunoblotting assay, where empty vector was used as control and β-actin was used as an internal reference. (B) Cell growth curve of YY-8103 cells with the exogenous SFRP1, which cultured in RPMI 1640 with 10% FBS. The cells transfected with the empty vector pcDNA3.0 were served as control. The experiments were repeated at least three times. The result represents the average value of triplicate wells, with standard deviation. T-test was performed to determine the statistical significance between both vector and SFRP1 experiments using SPSS software, and p<0.05. (C) Plasmid pcDNA3.0-SFRP1 was also transiently transfected into Hep3B cells, where the overexpression of SFRP1 was confirmed by immunoblotting assay, as compared to the control transfected by empty vector. (D) The colony formation of Hep3B cells was markedly inhibited as transfected with exogenous SFRP1, where the empty vector pcDNA3.0 was served as control (left). Here, after transfection for 24 h, the cells were striped and plated on 100 mm-dishes and then cultured by G418 (600 mg/ml) for 3 weeks. The dishes were stained with crystal violet solution and the number of colonies was counted from three independent experiments. The right histogram showed the colony formation efficiency, where the numbers represented the average value of three independent experiments, with standard deviation (p < 0.01, as compared with that of vector control).

Figure 3

The overexpression and RNA interference of SFRP1 can affect the cell growth of SMMC7721 cells. (A) SMMC7721 cells were stably transfected by plasmid containing SFRP1, where both offspring subclones, SMMC7721Z and SMMC7721Y, were validated to be overexpressed through western blotting assay, whereas no expression of the gene was confirmed in SMMC7721-mock although transfected with the same vector. (B) The overexpression of SFRP1 can suppress the cell growth of SMMC7721 cells as compared with the mock. The result represents the average value of triplicate wells, with standard deviation, p < 0.01. (C) Analysis of SFRP1 expression level of SMMC7721Z and normal human liver through RT-PCR, where β-actin was used as a loading control. (D) RNA interference using both siRNA_888 and siRNA_1094 was employed to knockdown the constitutive SFRP1 in SMMC7721Z cells, where both these siRNAs were effective as compared to siRNA_NC as control, demonstrated by western blotting assay. (E) Cell growth of SMMC7721Z cells was promoted by both siRNA_888 and siRNA_1094, not siRNA_NC. The result represented the average value of triplicate wells, with standard deviation, p < 0.05.

Figure 4

Loss of heterozygosity (LOH) analysis on HCC samples. (A) Schematic representation of the microsatellite markers of D8S532 and D8SAC016868 located on chromosome 8p11.2 flanking the SFRP1 locus. (B)The microsatellite markers, D8S532 and D8SAC016868, flanking the SFRP1 locus were employed to analyze the LOH on 46 pairs of HCC samples. Here, we showed a representative result of LOH from primary HCC (below) and corresponding adjacent non-cancer liver (upper), where the arrows indicated the deleted alleles in the tumor DNA.

Figure 5

The correlation between the expression of SFRP1 gene and DNA methylation status of SFRP1 promoter. (A) DNA methylation status of SFRP1 promoter was assessed in some HCC cell lines with or without the expression of endogenetic SFRP1, as well as fetal and adult normal livers, through MSP assay with specific primers. The peripheral blood lymphocyte (PBL) DNA treated with SssI Methylase (New England Biolabs, Beverly, MA) was used as a positive control for methylation (IVD), and water was used as a negative control (H₂O). (B) Bel7402, QGY7701 and MHCC-H cells without the expression of endogenous SFRP1 were treated with 5-aza-2'-deoxycytidine (DAC) and trichostatin A (TSA) alone or in combination. The expression of SFRP1 was then evaluated by RT-PCR. Untreated HCC cells were employed as control (lane 1). (C) The DNA methylation status of SFRP1 promoter in these cells treated by DAC and TSA were further evaluated by MSP assay and the same primers, where bisulfite-treated genomic DNA from these specimens was used as template. M, methylated; U, unmethylated.

Figure 6

DNA hypermethylation of SFRP1 promoter in primary HCC samples. (A) Transcriptional expression of SFRP1 was analyzed by RT-PCR in two pairs of HCCs and corresponding non-cancerous livers, where β-actin was used as an internal control. (B) The bisulfite-treated DNA sequencing was preformed on the genomic region (-160 ~ + 200) around TSS of SFRP1, where the extent of methylation of the promoter was evaluated through DNA sequencing on 8 random clones inserted by PCR products. There were 56 CpG dinucleotides (CpGs), represented by circles, located on the region. Black and white circles represented the methylated and unmethylated CpG dinucleotides, respectively, where the methylation status was determined by bisulfite-treated DNA sequencing on the corresponding clones. C, HCCs; N, non-cancerous livers.