Gene copy number variations in the leukocyte genome of hepatocellular carcinoma patients with integrated hepatitis B virus DNA

Abstract

Integration of hepatitis B virus (HBV) DNA into the human liver cell genome is believed to promote HBV-related carcinogenesis. This study aimed to quantify the integration of HBV DNA into the leukocyte genome in hepatocellular carcinoma (HCC) patients in order to identify potential biomarkers for HBV-related diseases. Whole-genome comparative genomic hybridization (CGH) chip array analyses were performed to screen gene copy number variations (CNV) in the leukocyte genome, and the results were confirmed by quantitative polymerase chain reaction (qPCR). The commonly detected regions included chromosome arms 19p, 5q, 1q and 15p, where 200 copy number gain events and 270 copy number loss events were noted. In particular, gains were observed in 5q35.3 (OR4F3) and 19p13.3 (OR4F17) in 90% of the samples. Successful homologous recombination of OR4F3 and the HBV P gene was demonstrated, and the amplification at 5q35.3 is potentially associated with the integration of HBV P gene into natural killer cells isolated from peripheral blood mononuclear cells (PBMCs). Receiver operating characteristic (ROC) curve analysis indicated that the combination of OR4F3 and OR4F17 a novel potential biomarker of HBV-related diseases.

Keywords: CGH; HBV-HCC; biomarker; integration.

Figure 1. OR4F3 and OR4F17 amplifications

A chromosome view of the CGH array results showing the terminal copy number gains on 5q35.3 and 19p13.3 including OR4F3 and OR4F17. Arrowheads indicate targets with upward-shifted log₂ ratios. Chromosome (Chr).

Figure 2. The homologous recombination of the OR4F3 5′UTR and HBV P gene

A. A schematic alignment of OR4F3 and HBV is presented. The complementary sites between the OR4F3 gene and the HBV genome P gene are indicated by red frames. B. Compared with the control (Con), the agarose gel electrophoresis results show the OR4F3 and HBV chimeric fusion genes in the HBV-HCC samples. Sequencing results show one complementary site and a short HBV P gene sequence. The M lanes contain the DNA size markers (DL2000). The Con lane is healthy control, and lanes 1 and 2 are the HBV-HCC samples. UTR (untranslated region). C. The OR4F3 and HBV P gene chimeric fusion genes in the leukocyte genome are shown. Agarose gel electrophoresis showing the PCR amplification products for the healthy control (Con), non-HBV-HCC (N), non-active HBV-HCC (NA), active HBV-HCC (A), and antiviral therapy (AT) samples as well as marker (M). The chimeric fusion DNA fragments of P gene and OR4F3 were observed in the sorted NK cells of HBV- samples and B cells of samples from patient receiving antiviral therapy. T-lymphocyte cells (T cells); B-lymphocyte cells (B cells); Natural killer cells (NK cells).

Figure 3. OR4F3 and OR4F17 as biomarkers for HBV-related diseases

A. Quantification of OR4F3 and OR4F17 levels in healthy controls (n = 20), CHB patients (n = 100), HBV-LC patients (n = 100), and HBV-HCC patients (n = 100). The copy number levels were determined using absolute qPCR. Results are presented as copies per microliter. The unpaired t-test was performed to assess significance of differences between HBV-related groups and control groups. P-values of less than 0.05 were deemed to be significant. (*p < 0.05; **p < 0.005). Control (Con); chronic hepatitis B (CHB); hepatitis B virus-liver cirrhosis (HBV-LC); hepatitis B virus- hepatocellular carcinoma (HBV-HCC); OR4F3, olfactory receptor, family 4, subfamily F, member 3; OR4F17, olfactory receptor, family 4, subfamily F, member 17. B. ROC curves for the OR4F3 and OR4F17 genes between the CHB and HBV-HCC as well as HBV-HCC and normal are presented. OR4F3 yielded AUCs (areas under the ROC curve) of 0.852 for discriminating HBV-HCC from CHB and 0.836 for discriminating HBV-HCC from normal. OR4F17 yielded AUCs of 0.754 for discriminating HBV-HCC from CHB and 0.910 for discriminating HBV-HCC from normal. The ROC curves between CHB and HBV-LC as well as HBV-HCC and HBV-LC are presented. OR4F3 yielded AUCs of 0.762 for discriminating HBV-LC from CHB and 0.637 for discriminating HBV-HCC from HBV-LC. OR4F17 yielded AUCs of 0.534 for discriminating HBV-LC from CHB and 0.736 for discriminating HBV-HCC from HBV-LC. Control (Con).

Figure 4. Clinical analysis of the samples and the threshold accuracy of OR4F3 and OR4F17

A. Correlation between clinical characteristics and the concentration of OR4F3 and OR4F17. According to the imaging and histological examination of liver, the patients were grouped into 100 CHB patients, 100 HBV-LC patients and 100 HBV-HCC patients. The concentration of OR gene did not correlate with the HBV-DNA level, genotype, case history, family history, ages, smoking, or alcoholism (P>0.05); however, the concentrations of OR4F3 and OR4F17 were varied between males and females (P = 0.0107 and P = 0.0322, respectively). Units: copies/ul. B. Threshold accuracy of OR4F3 and OR4F17. The threshold accuracy for OR4F3 and OR4F17. The cut-off value unit was copies/ul.