In our study, we used the GenomePlex® Whole Genome Amplification ChIP-chip analyses and found that HBc can bind to more than three-thousand promoters in the human genome. Among this set of gene promoters, the high CpG density promoters were the most common. By evaluating the corresponding the GO annotations and the KEGG pathways for genes affiliated with each of these promoters, we determined that HBc tended to target the regulatory regions of genes with molecular function and malignant transformation in the liver cell repertoire. These results were consistent with our previous data showing that HBc preferentially binds CpG islands within the HBV covalently closed circular DNA
]. To our knowledge, our current study is the first to provide evidence that HBc can bind to a large number of host gene promoters throughout the whole human genome. Our findings may provide new insights towards understanding the mechanisms underlying the ability of HBc to regulate host gene transcription.
The ChIP technique has proven to be a powerful tool for detecting protein-DNA interactions in living cells, and it is currently the gold standard procedure for identifying target genes of a given transcription factor of interest. Over the last several years, great strides have been made in expanding the use of ChIP from a one gene-at-a-time approach to a global type of analysis by hybridizing samples to genomic microarrays (i.e., ChIP-on-chip assay)
]. In such a study, Dere et al.
previously identified 1896 of the 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD)-responsive genes in mouse liver by using microarray analysis
]. Most regions were enriched 5.7-fold with the fold-enrichment values ranging from 1.7- to 111.4-fold
]. Using ChIP-chip technique, Koudritsky and Domany showed that the number of target genes of each studied transcription factor (TF) ranged from a few hundred to several thousand
]. Consistent with these data, 3100 HBc-immunoprecipitated host promoter regions were found to be enriched by at least 2-fold in HBc-infected tissues (p
0.05, compared with control). Moreover, 2883 of 3100 gene promoters were found in all of the four pooled samples investigated by ChIP-on-chip, which indicates that this approach has a good level of repeatability and the reliability. In our study, arrays representing 18028 promoter regions and CpG islands of the whole human genome were used in combination with HBV HBc-based ChIP to identify putative targets of this viral factor. We also identified 266 HBc-bound gene promoters that were different among the four pooled samples. Such apparently inconsistent findings may reflect background differences among the pooled ChIP samples. We found it necessary to use pooled samples for the ChIP-on-chip assays since none of our ChIP samples provided enough DNA for genomic microarray analysis. The practice of pooling ChIP samples for ChIP-on-chip is very common, but is known to produce very high background when the samples are analyzed on genomic tiling arrays
]. To verify the reliability of our ChIP-on-chip results, the method of ChIP-qPCR was used to detect the promoter sequences of randomly selected genes in samples of our output DNA. Our results confirmed that HBc can bind to the promoters of all 12 genes evaluated, including the MxA gene, which has been demonstrated as a bona fide
target of HBc and was used as a positive control in our study
The underlying mechanisms of the interaction between HBc and the human genome are poorly understood. HBc is a 183 residue protein with two domains: the N-terminal 144 amino acid domain that is sufficient for self-assembly into capsid particles, and the C-terminal arginine-rich domain, which shares a high similarity to protamine and functions as a nucleic acid-binding domain
]. Although the C-terminal arginine-rich domain has been identified as imporatnt for HBc binding to pregenome RNA or genome DNA
], no specific target sequence for the DNA-binding domain of HBc has been determined. Thus, while we know that the C-terminal arginine-rich domain can bind HBV-specific nucleotides in vivo
, we have yet to determine the mechanisms mediating this binding specificity. However, based on the fact that HBV core particles specifically bind to the HBV pregenome or genome in vivo
, we inferred that the binding specificity to the human genome might be mediated by the C-terminal arginine-rich domain, along with other regulatory regions in the N-terminal assembly domain. Previous studies have shown that a signal for nuclear transport is located near the carboxy termini of HBc, in the arginine-rich domain. This signal is comprised of a set of two direct PRRRRSQS repeats and is highly conserved among mammalian hepadnaviruses
]. Recently, one study used a novel GST-fusion protein-based gel shift method to demonstrate that the C-terminal arginine-rich domain alone is capable of binding to DNA in a sequence-independent manner
]. In addition, another study showed that HBc can enhance host NF-kB DNA-binding ability
], which suggested that HBc may bind to and interact with other host nuclear proteins to enhance or inhibit their transcriptional activator functions. The large number of potential binding targets of HBc that were identified in our genome-wide analysis of human promoters supports the notion that HBc may utilize more than one of these proposed mechanisms. However, further study is needed to clarify this issue.
Previous studies have shown that HBV DNA integration and/or expression of HBV proteins may have direct effects on host cellular functions
]. However, the roles of interaction between HBV proteins, especially HBc, and the human genome have not been well studied. Locarnini et al
] examined the effect of the HBV core protein on cellular gene expression in the hepatoma Huh-7 cell line by using a commercial high-density oligonucleotide array (Affymetrix Hu95A GeneChip). They found that only five genes had differential mRNA expression that was greater than three-fold at day 7 post-HBc expression. Among these five genes, four were down-regulated (by 3
15-fold) and included genes whose encoded products affect intermediary metabolism, cell surface receptors, and intracellular signaling. The fifth, a cytokine gene, was up-regulated. In contrast, our study utilized a microarray containing the genome-wide set of human gene promoters to search for potential binding sites of HBc, and identified over 3000 putative functional targets. Even though the full panel of these promoters has yet to be confirmed, the diverse spectrum of genes represented suggests that HBc may have multiple functions in regulating HBV pathogenesis and survival in the human host.
It is likely that HBc binding does not exert significant regulatory effects on every gene for which its promoter is bound. For some or many genes, HBc may merely mediate a slight modulation of the transcriptional activity. However, the cumulative effect of slight transcriptional modifications across the genome may impact overall cellular function. In our study, we found that HBc can bind to 64 gene promoters of the MAPK pathways, and 41 gene promoters of the Wnt/β-catenin signaling pathways. These two pathways are known to be critically involved in the development of HBV-related hepatocellular carcinoma. The MAPK pathways regulate a variety of normal human cellular activities, including proliferation, differentiation, survival, and death. As such, the components of MAPK signaling play a key role in several steps of tumorigenesis, including cancer cell proliferation, migration, and invasion
]. Meanwhile, the Wnt/β-catenin signaling pathway has emerged as a critical player in both the development of normal liver and as an oncogenic driver in hepatocellular carcinoma
]. As described above, the accumulation of slight effects from HBc binding to many gene promoters may produce sizeable effects on host cellular functions, possibly increasing a cell’s susceptibility to harmful factors, such as carcinogens
]. Certainly, further studies will be needed to verify this theory and define the detailed factor(s) and step(s) that HBc works through, or that work through HBc, to regulate host gene transcription.
We investigated the functional relevance of HBV core protein binding to five of the genes whose promoters were identified in the ChIP-on-chip analysis. Transcriptional activity assays indicated that SRC, IGF1R, and NTRK2 genes were up-regulated by more than 2-fold when co-transfected with HBc. The HRAS gene, however, was down-regulated in the presence of HBc. SRC is the best characterized member of the family of nine tyrosine kinases that regulates cellular responses to extracellular stimuli, and activated mutants of SRC are oncogenic
]. Among these four genes, SRC showed the greatest change in gene expression in HBc co-transfections, which may be related to the SRC gene promoter sequence that contains a potential NF-κB binding sequence (5′-GGGAAAATCC-3′). Thus, HBc might enhance the NF-κB DNA-binding ability on SRC
]. However, it should be noted that the possible roles of the HBV core protein to the transcriptional activity of only small number of gene promoters were evaluated in our study; more sensitive and accurate high-throughput DNA-protein interaction and gene expression analysis techniques should be used in further studies.