Genetic variation of the IL-28B promoter affecting gene expression

Abstract

The current standard of care for the treatment of chronic hepatitis C is pegylated interferon-α (PEG-IFNα) and ribavirin (RBV). The treatment achieves a sustained viral clearance in only approximately 50% of patients. Recent whole genome association studies revealed that single nucleotide polymorphisms (SNPs) around IL-28B have been associated with response to the standard therapy and could predict treatment responses at approximately 80%. However, it is not clear which SNP is most informative because the genomic region containing significant SNPs shows strong linkage disequilibrium. We focused on SNPs in close proximity to the IL-28B gene to evaluate the function of each and identify the SNP affecting the IL-28B expression level most. The structures of IL-28A/B from 5' to 3'-UTR were determined by complete cDNA cloning. Both IL-28A and 28B genes consisted of 6 exons, differing from the CCDS data of NCBI. Two intron SNPs and a nonsynonymous SNP did not affect IL-28B gene function and expression levels but a SNP located in the proximal promoter region influenced gene expression. A (TA) dinucleotide repeat, rs72258881, located in the promoter region was discovered by our functional studies of the proximal SNPs upstream of IL-28B; the transcriptional activity of the promoter increased gradually in a (TA)(n) length-dependent manner following IFN-α and lipopolysaccharide stimulation. Healthy Japanese donors exhibited a broad range of (TA) dinucleotide repeat numbers from 10 to 18 and the most prevalent genotype was 12/12 (75%), differing from the database (13/13). However, genetic variation of IL-28A corresponding to that of IL-28B was not detected in these Japanese donors. These findings suggest that the dinucleotide repeat could be associated with the transcriptional activity of IL-28B as well as being a marker to improve the prediction of the response to interferon-based hepatitis C virus treatment.

Figure 1. The position of significant SNPs and IL-28A/B in chromosome 19, retrieved from the database.

(A) The IL-28A/B genes located in chromosome 19q13 are described in the genome map of the UCSC genome browser. The significant proximal SNPs around IL-28B associated with response to PEG-IFN/RBV therapy are shown in the map . SNPs of (TA)_n variation at the regulatory region of IL-28A are displayed in the position corresponding to that of IL-28B, which is not associated with anti-HCV therapy. (B) The schematic of IL-28A/B gene structure is described based NCBI CCDS data. Arrows show five significant SNPs examined in this study (see Table 1).

Figure 2. The determination of IL-28B gene structure and UTR region.

IL-28A/B cDNA was isolated using a complete cDNA cloning method and the entire sequences were determined using HeLa, MT-2, and Raji call lines and PBMC from healthy volunteers. (A) 5′- and 3′-RACE analyses were used to determine the complete sequence of IL-28A/B mRNA after LPS stimulation (3 µg/mL) for 4 h following IFN-α treatment (100 U/mL) for 16 h. A representative example of agarose gel electrophoresis is shown for the non-stimulated control (NC). PCR products were inserted into the cloning vector and 6 clones of 5′- and 3′-RACE were analyzed by sequencing. (B) mRNA sequences of the 5′ terminal region were aligned using CCDS retrieved from NCBI and RACE data of IL-28A/B. The upper two sequences are reference sequences from the NCBI CCDS and the lower two are representative sequences of IL-28A and 28B obtained from 5′-RACE. The underlined triplet indicates the start codon of each gene and arrow shows the splice junctions. (C) mRNA sequences of the 3′ terminal region were aligned using CCDS retrieved from NCBI and RACE data from IL-28A/B. The double-underlined triplet indicates the stop codon of each gene and arrows show the splice junctions. The polyA signal and representative site of polyadenylation also are shown. (D) The derived gene structure of the IL-28B is shown with the significant SNPs. The location of SNP No. 3 was changed from the regulatory to an intron region. The transcription start site (TSS) is found behind SNP No. 2.

Figure 3. Transcriptional activity of the IL-28B promoter region compared between major and minor alleles.

(A) The pGL4 reporter plasmid was constructed by subcloning the IL-28B promoter subfragment (nt −858 to +30). The combinations of two regulatory SNPs (rs72258881 or rs4803219) were introduced into the pGL4 vector (pGL4/WW, MW, WM, and MM). (B) Raji cells were co-transfected with pGL4 plasmids (0.05 g), and pGL4.74 control plasmid (0.05 g), and tested for firefly as well as renilla luciferase after LPS stimulation (3 µg/mL) for 4 h following IFN-α treatment (100 U/mL) for 16 h. These cells were seeded in a 96-well plate at 10⁴ cells/well. The luciferase activities were normalized with renilla activities and data are presented as fold induction from activation of control vector. Bars indicate the means ± SD of triplicate determinations and the results are from one of three experiments. Statistical analyses are shown in table S2 to avoid complication. (C) For real-time PCR, the combinations of two regulatory SNPs (rs72258881 or rs4803219) were introduced into the pdCMV vector harboring a FLAG sequence (pdCMV/WW, MW, WM, and MM). (D) Jurkat cells were co-transfected with pdCMV plasmids (0.05 µg) and secreted alkaline phosphatase (SEAP) control plasmid (0.05 µg) and the expression levels were quantified using specific primer after LPS and IFN-α stimulation. The FLAG expression levels were normalized with SEAP activities and GAPDH as described in method section. Data are presented as fold induction from expression levels of pdCMV/WW. Bars indicate the means ± SD of triplicate determinations and the results are from one of three experiments. Statistical analyses are shown in table S3 to avoid complication.

Figure 4. The determination of intron SNPs located near the branch site of splicing.

(A) The expression plasmid of WT, d-iSNPs, or antisense (AS) derived from the CMV promoter was transfected into HeLa cells. Schematic of the WT, d-iSNPs, or AS used in the transfection experiments. PCR primers were designed to amplify products between exons. The effect of No. 3 and 5 SNPs (rs28416813 or rs11881222) on splicing were examined by amplicons x and y, respectively. The amplicon z was used for a splicing control. (B) Isolated RNAs were amplified by RT-PCR. The amplified products were checked by 2% agarose gel electrophoresis. The bands from the AS plasmid transcribing antisense represented abnormal splicing of mRNA as a control. Results shown are representative of three independent experiments.

Figure 5. The purification and the activity of recombinant IL-28B with or without nsSNP.

(A) The 6×His-tagged expression plasmid of wild type, ns-mut, or AS controlled by the CMV promoter was transfected into 293F cells. Schematics are the wild type, ns-mut and AS used in the transfection experiments. The procedure for recombinant protein purification is described in the materials and methods section. (B) The purified products were confirmed by immunoblotting using anti-IL28B antibody and the secondary antibody. The prepared proteins were loaded onto a 12% polyacrylamide gel. Bands corresponding to the expected molecular weight of IL-28B were observed in the wild type and ns-mut lanes. (C) For luciferase assay, HeLa cells were seeded into a 96-well plate at 10⁴ cells/well and transfected with pISRE-Luc and pGL4.74 control vector before 16 h of IFN-α or IL-28B stimulation. Five ng/mL of IL-28B wild or ns-mut was added to the culture medium. Flow-though liquid from AS expression was used as a negative control. IFN-α (100 U/mL) was added for positive control of ISRE activity. The luciferase activities were normalized with Renilla activities and data are presented as fold induction from the basal promoter activation of the wild type. Bars indicate the means ± SD of triplicate determinations and the results are from one of three experiments.

Figure 6. Luciferase assay of (TA)_n number.

(A) IL-28B promoter subfragment (nt −858 to +30) modifying (TA)_n number from 10 to 18 was constructed in the pGL4 vector. (B) Raji cells were co-transfected with pGL4 plasmids (0.05 g), and pGL4.74 control plasmid (0.05 g), and tested for firefly as well as renilla luciferase after LPS stimulation (3 µg/mL) for 4 h following IFN-α treatment (100 U/mL) for 16 h. These cells were seeded into a 96-well plate at 10⁴ cells/well. The luc activities were normalized with renilla activities and data are presented as fold induction from the activation of the control vector. Bars, the means ± SD of triplicate determinations and the results are from one of three experiments. Statistical analyses are shown in table S4 to avoid complication.