Cell Type Diversity in Hepatitis B Virus RNA Splicing and Its Regulation

Abstract

Although RNA splicing of hepatitis B virus (HBV) is a commonly observed in livers of hepatitis B patients as well as in the cultured cells replicating the viral genome, its biological significance in the HBV life cycle and the detailed regulatory mechanisms are still largely unclear. In this study, we found cell-type dependency of HBV splicing of the 3.5 kb pregenomic RNA, which is efficiently spliced in human hepatoma cells but not in cells derived from human hepatic stellate, mouse hepatoma and human non-hepatic cells. It may be likely that RNA splicing is one of the determinants of host range restriction of HBV. Given the finding indicating the difference in cell-type dependency of the splicing efficiency between HBV and simian virus 40, we carried out intron-swapping experiments. The results suggest the presence of putative exonic splicing enhancer that possibly works in the cell-type dependent fashion. Together with further mutational analyses, a novel 50-nt intronic splicing silencer, whose secondary structure is well conserved among the HBV strains, was identified. It appears that this intronic silencer functions effectively independent of cell backgrounds.

Keywords: alternative splicing; cis-acting element; exonic enhancer; hepatitis B virus; intronic silencer.

FIGURE 1

Cell-type dependency of HBV 3.5 kb RNA splicing. (A) pUC-HB-Ce was transfected into HepG2 and HEK293 cells. Two days after the transfection, total RNA was extracted from cells and separated on an agarose gel. HBV 3.5 kb RNA and spliced RNA(s) were detected by northern blotting using HBV RNA probe (nt 1998–2447) (Supplementary Figure S1A). Band intensities of 3.5 kb- and SP1 RNAs on the blot were determined by ImageJ software and the ratios of SP1 RNA to 3.5 kb RNA calculated were indicated. Ethidium bromide-stained agarose gel electrophoresis of 18S ribosomal RNA (rRNA) was also shown (bottom). (B) HepG2 and HEK293 cells transfected with pUC-HB-Ae (GT-A), -Bj56 (GT-B), -Ce (GT-C) and an empty vector (EV) were analyzed by immunoblotting to detect HBcAg and glyceraldehyde-3-phosphate dehydrogenase (GAPDH). (C) qRT-PCR analysis was performed to determine the levels of 3.5 kb RNA and spliced RNA in 13 kinds of cells transfected with pUC-HB-Ae (GT-A), -Bj56 (GT-B), and -Ce (GT-C). The quantity ratios of the spliced RNAs to total 3.5 kb RNA derived RNA species were calculated and those in HepG2 cells were set to 1. Values shown represent means ± SD obtained from three independent samples. (D) Total RNAs obtained from the cell samples as shown in (C) were used with semi-quantitative RT-PCR. cDNA bands corresponding to unspliced 3.5 kb RNA and its spliced forms (Sp RNA) were detected by agarose gel electrophoresis. Unspliced 3.5 kb RNA, SP9, and SP1 were identified by nucleotide sequencing. SP14 was estimated from the size of PCR product obtained. Cell lines used were indicated at the bottom.

FIGURE 2

Identification of cis-regulatory elements involved in HBV 3.5-kb RNA splicing in cell-type dependent and -independent manners. (A) A schematic representation of constructs and PCR primers used in the intron-swapping experiment is indicated. (B) HBV WT and HBV/int-SV40 were expressed in HepG2 and HEK293 cells. PCR products were analyzed by agarose gel electrophoresis (left). Schematic of unspliced and spliced forms detected is shown (right). (C) SV40-L WT and SV40/int-HBV were expressed in the cells and PCR products were analyzed as shown in (B).

FIGURE 3

Characterization of the intronic sequence important for silencing of the 3.5 kb RNA splicing. (A) A schematic representation of mutated HBV genomes derived from GT-C with deletions and/or substitutions within the major intron region is shown. Putative secondary structures predicted by the CentroidFold (http://rtools.cbrc.jp/centroidfold/) in nt 2858–2983 and 3197-48 regions are indicated. Each predicted base pair is colored with the heat color gradation from blue to red corresponding to the base-pairing probability. Bold lines are drawn along the sequences targeting to introduce substitution mutations S1 and S2. (B) RT-PCR analysis of HBV RNAs expressed from WT and a series of deletion-mutated HBV genomes (D1–D5) in HepG2 and HuH-7 cells. The quantity ratio of the spliced RNAs to the unspliced RNA (Sp ratio) was determined. ND, not determined. Open arrows, unspliced forms. ^∗cDNA products with aberrant sizes. (C) As described in (B) but HBV mutated genomes S1/D1 and S1/D1/S2 in addition to WT and D1 were used for RNA expression. (D) As in (C) but S1 mutant and WT were expressed in the cells. Bands corresponding to unspliced 3.5 kb RNA and its spliced forms (Sp RNA) were indicated.

FIGURE 4

Effect of deletions in intron region of the HBV genome on the splicing efficiency in human HCC and non-hepatic cells. RT-PCR analyses of HBV RNAs expressed from WT and deletion-mutants corresponding to D1 and D4 derived from GT-A (deleted regions; nt 2990–3202 in D1, nt 2864-48 in D4), -C (nt 2984–3196 in D1, nt 2858-48 in D4) and -D (nt 2951–3163 in D1, nt 2858-48 in D4) in human HCC (HepG2, HuH-7 and PLC/PRF/5) cells (A) and in non-hepatic (HEK293, A549, and HeLa) cells (B). Sp ratio; the quantity ratio of the spliced RNAs to the unspliced RNA. Open arrows, unspliced forms.

FIGURE 5

Diversity in the splicing efficiency among HBV genotypes/strains. (A) pUC-HB-Ae (GT-A), -Bj56 (GT-B), -Ce (GT-C) or -D_Ind60 (GT-D) was transfected into HepG2 and HuH-7 cells. Two days after the transfection, total RNA was extracted from cells and analyzed by qRT-PCR to determine the levels of 3.5 kb RNA and spliced RNAs. The quantity ratios of the spliced RNAs to total 3.5 kb RNA derived RNA species were calculated and that of GT-C was set to 1. (B) A schematic representation of a series of mutated HBV genomes replacing parts of the GT-C sequence with corresponding parts of GT-A sequence. Positions of donor and acceptor sites of SP1 RNA are indicated. (C) qRT-PCR analysis was performed to determine 3.5 kb RNA and spliced RNA levels in HepG2 and HuH-7 cells transfected with pUC-HB- Ce (GT-C), -Ae (GT-A), HBV Ce/Ae1, /Ae2, /Ae3 or /Ae4. The quantity ratios of the spliced RNAs to total 3.5 kb RNA derived RNA species were calculated and those in GT-C-expressing cells were set to 1 (upper). Representative pattern for RT-PCR result indicating expression of unspliced and spliced (Sp) RNAs derived from 3.5 kb RNA in the transfected HepG2 (middle) and HuH-7 cells (lower). (D) As in (C) but mutated genomes HBV Ce/Ae5, /Ae6, /Ae7 as well as GT-C, -A and HBV Ce/Ae2 were used for expression of HBV RNAs. Values represent mean ± SD (HepG2; n = 2, HuH-7; n = 3). ^∗p < 0.05, ^∗∗p < 0.01, by one-way ANOVA followed by Dunnett’s test compared to GT-C.