Molecular evaluation of hepatitis B virus infection and predominant mutations of pre-core, basal core promoter and S regions in an Iranian population with type 2 diabetes mellitus: a case–control study

All of the patients with type 2 diabetes mellitus attending diabetic clinics of the Bushehr University of Medical Sciences located in southern Iran were included consecutively in this study. In accordance with the American Diabetes Association criteria, diabetic cases were defined as patients with fasting serum glucose levels ≥ 126 mg/dL on at least 2 different occasions accompanied by regular use of oral antidiabetic medications [18]. The demographic, laboratory, and clinical data of each patient were obtained from the patient’s medical records at the diabetic clinics. As a control group, 782 non-diabetic volunteers, who were matched in sex, age (± 3 years), and date of participation with diabetic patients, were recruited from the out-patient populations attending the hospitals of the Bushehr University of Medical Sciences for blood tests. The patients were excluded if they had markers of HCV and/or HIV infections and exhibited evidence of hepatogenous diabetes, gestational diabetes, type 1 diabetes, and autoimmune or metabolic chronic liver disease. All of the participants were informed about the purpose of the research and requested to sign a written informed consent to use their leftover serum samples for HBV detection and analysis. The study was approved by the Ethical Committee of the Bushehr University of Medical Sciences with research project number IR.BPUMS.REC.1395.54 and was funded by the Deputy Research and Affairs of the University with grant number 3254. In addition, all methods were performed under the relevant guidelines and regulations.

All of the serum samples were tested for the detection of hepatitis B surface antigen (HBsAg) and hepatitis B core antibody (HBcAb) using HBsAg one-Version ULTRA and HBc Ab ELISA kits (DIA.PRO, Milan, Italy), respectively. The sensitivity and specificity of these kits were 100%. The HBsAg and/or HBcAb seropositive samples were tested using nested PCR, targeting the S region of the HBV genome, followed by sequencing to determine the genotypic pattern of HBV infection. Briefly, HBV DNA was extracted from serum samples using the High Pure Viral Nucleic Acid kit (Roche, Mannheim, Germany) according to the manufacturer’s instructions. To determine HBV genotypes, 447 nucleotides and 416 nucleotides of the S region were amplified by nested PCR using outer primers [forward primer (244-HBS-F1): GAGTCTAGACTCGTGGTGGACTTC; reverse primer (691-HBS-R1): AAATKGCACTAGTAAACTGAGCCA] and inner primers [forward primer (255-HBS-F2): CGTGGTGGACTTCTCTCAATTTTC; reverse primer (671-HBS-R2): GCCARGAGAAACGGRCTGAGGCCC], respectively [19‐21].

To determine mutations in BCP and pre-C regions, the pre-C region was amplified in the first round of PCR using primers 1606-pre-C-F1 (GCATGGAGACCACCGTGAAC) and 2395-pre-C-R1 (AGGCGAGGGAGTTCTTCTTC). The second round of PCR was performed using primers 1653-pre-C-F2 (CATAAGAGGACTCTTGGACT) and 2393-pre-C-R2 (GCGAGGGAGTTCTTCTTC) [21‐23]. The 789 bp and 740 bp length fragments from the pre-C region were amplified in the first and second rounds of nested PCR, respectively. The sequences of primers and PCR conditions for amplification of the S and pre-C regions of the HBV genome are summarized in Table 1.

The PCR products were analyzed using agarose gel electrophoresis. Following the extraction of amplicons from the agarose gel using the Agarose Gel DNA Extraction Kit (Roche, Mannheim, Germany), the 416 bp length fragments from the S region and the 740 length fragments from the pre-C region were sequenced to determine HBV genotypes and mutations by Sanger dideoxy sequencing technology (Macrogen Co., Korea). The obtained sequences were analyzed by using GenBank Basic Local Alignment Search Tool (BLAST), compared with the reference sequences of S and pre-C regions of the standard HBV genotypes available at the nucleotide database of the NCBI and submitted to the GenBank sequence database. Then, the S and pre-C sequences isolated from the patients and the reference sequences were aligned by the ClustalW program in the Molecular Evolutionary Genetics Analysis (MEGA) software version 7.0 (Biodesign Institute, Tempe, AZ, USA) [24]. The phylogenetic tree was constructed by the neighbor-joining method using MEGA software, as described previously [25]. The reference strains representing the standard HBV genotypes were used as references.

The Student’s t-test was used to analyze and compare quantitative variables between HBV-positive and HBV-negative diabetic patients. Categorical variables were compared by χ² test or Fisher’s exact test. Logistic regression analysis was used to determine the risk factors associated with the prevalence of HBV infection among diabetic patients, and the odds ratio with 95% confidence intervals was calculated. SPSS 17 package program (SPSS Inc., Chicago, IL, USA) was used for data analyses, and P values < 0.05 were considered significant.

Serum samples were obtained from 733 patients with type 2 diabetes mellitus, including 256 males and 477 females. The mean age ± SD of diabetic patients was 55.2 ± 11.7 years with a range of 26–96 years. The majority of diabetic patients were in the age groups 51–60 years (38.7%) and 61–70 years (20.1%), respectively. Moreover, 782 non-diabetic volunteers, including 304 males and 478 females, were enrolled in the study as the control group. The mean age ± SD of the non-diabetic controls was 53.37 ± 13.36 years with a range of 20–93 years.

Of 733 patients with type 2 diabetes mellitus, 94 cases (12.82%, 95% CI 10.60–15.44%) had HBcAb, and 28 cases (3.82%, 95% CI 2.66–5.47%) were positive for HBsAg. The seroprevalence of HBsAg and HBcAb in the non-diabetic controls were 1.15% (95% CI 0.61–2.17%) and 10.74% (95% CI 8.76–13.11%), respectively. Diabetic patients had a significantly higher seroprevalence of HBsAg than the non-diabetic controls (P = 0.01). Although diabetic patients had a higher seroprevalence of HBcAb than the non-diabetic controls, the difference was not statistically significant (P = 0.23) (Additional file 1: Table S1).

The highest rate of HBV seroprevalence was observed in the age group 61–70 years (6.1%) for HBsAg and in the age group > 71 years (30.3%) for HBcAb, whereas the lowest HBcAb seroprevalence was found in the age group 26–30 years (6.25%), and those aged < 40 years did not show HBsAg seropositivity (Additional file 1: Tables S2 and S3). Overall, HBV seropositivity increased with age so that HBsAg seropositive diabetic patients had a significantly higher mean age (59.92 ± 13.2) compared to HBsAg seronegative diabetic patients (55.00 ± 11.66) (P = 0.03). Besides, HBcAb seropositive diabetic patients showed significantly higher mean age (60.31 ± 12.57) compared to HBcAb seronegative diabetic patients (54.44 ± 11.44) (P = 0.001). HBsAg and HBcAb seropositivity were more prevalent in female diabetic patients than male patients, although the differences were not statistically significant (Additional file 1: Tables S2 and S3). HBsAg seropositive diabetic patients had higher AST and ALT levels but lower TCH, TG, and FBS levels than HBsAg seronegative patients. Nevertheless, the seroprevalence of HBsAg among diabetic patients was not statistically associated with the serum levels of liver enzymes, TCH, TG, and FBS (Table 2).

HBsAg seroprevalence was higher among the non-diabetic controls aged 70 to 79 years (4.5%), whereas those aged 50–59 years had the lowest rate of HBsAg seroprevalence (0.7%) (Additional file 1: Table S4). HBcAb seroprevalence increased with age so that the non-diabetics over 80 years old had a significantly higher HBcAb seroprevalence (36.7%) compared to the other age groups (P = 0.0001) (Additional file 1: Table S5). Besides, HBsAg and HBcAb seroprevalence was more prevalent in male non-diabetics than female non-diabetics, although the differences were not statistically significant (Additional file 1: Tables S4 and S5).

According to the molecular evaluation, 19 diabetic patients (2.59%, 95% CI 1.67–4.01%) had HBV-DNA (Additional file 1: Figs. S1 and S2). The mean age of HBV DNA-positive diabetic patients (60.0 ± 14.3) was higher than that of HBV DNA-negative diabetic patients (55.07 ± 11.66), but this difference was statistically insignificant (P = 0.07). Overall, HBV-DNA positivity in diabetic patients was not significantly associated with age, gender distribution, and the serum levels of AST, TCH, TG, and FBS (Table 3). In contrast, HBV-DNA positivity was associated with ALT levels so that those patients with ALT levels of ≥ 81 IU/L (13.0%) had a significantly higher prevalence of HBV-DNA positivity than those patients with ALT levels of ≤ 24 IU/L (1.9%) (OR: 9.60; 95% CI 1.51–61.07; P = 0.02). Therefore, an ALT level of ≥ 81 IU/L was the only significant predictive variable for HBV-PCR positivity in diabetic patients (Table 3). An occult HBV infection (OBI) was found in one of the HBV-DNA positive samples, which was positive for HBcAb but negative for HBsAg. The prevalence of HBV-DNA positivity among diabetic patients (2.59%) was significantly higher than HBV-PCR positivity among the non-diabetic controls (0.77%, 95% CI 0.36–1.66%) (P = 0.007). None of the non-diabetic controls had occult HBV infection. The HBV sequences isolated from patients with type 2 diabetes mellitus were identified as genotype D, sub-genotype D1/D3, and subtype ayw2 (Fig. 1). The same genotypic pattern was found among the non-diabetic controls. Subtype ayw2 was characterized by Arg (R) at positions 122, Pro (P) at position 127, and Lys (K) at position 160 of HBsAg, but not Phe (F) and Ala (A) at positions 134 and 159 of HBsAg, respectively (Fig. S3). The HBV sequence isolated from diabetic patient with OBI was genotype D, sub-genotype D1/D3, and subtype ayw2.

Frequent point substitution mutations were detected in the S, pre-C, and BCP regions of the HBV genome isolated from the patients. Of those mutations in the S region, the conversion of Pro → Thr at amino acid 120 (P120T) of HBsAg was detected in five samples with HBsAg OD near the cut off of the ELISA kit, and conversion of Gly → Arg at amino acid 145 (G145R) of HBsAg was detected in one sample (Table 4 and Additional file 1: Fig. S3). The S G145R/P120T double mutation was observed in one HBsAg-negative sample. This sample was an OBI case (Accession No. MF419215). In the samples with P120T and/or G145R mutations, HBV-DNA was detectable in the second round of nested PCR. Of those mutations in the pre-C region, a G to A substitution at nucleotide position 1896 (G1896A) was detected in four samples, whereas G1899A mutation was found in eight samples. G1899A mutation was detected in the pre-C region of the occult HBV strain (Accession No. OK382075). Six samples had 5 nucleotide substitutions in the BCP region, including T1753C, A1762T, G1764T, C1766G, and C1773T. These 5 nucleotide substitutions were detected in the BCP region of the occult HBV strain (Accession No. OK382075). There were two samples with pre-C G1896A/G1899A double mutation, while six samples had BCP A1762T/G1764T double mutation. The serum samples with pre-C G1896A mutation exhibited no concomitant BCP A1762T and G1764T mutations; however, six out of the eight samples with pre-C G1899A mutation also exhibited A1762T and G1764T mutations (Table 5 and Additional file 1: Fig. S4). The A1762T and G1764T mutations in the BCP region caused the conversion of Lys → Ile at amino acid 130 (K130I) and Val → Leu at amino acid 131 (V131L) of the X protein (Table 5 and Additional file 1: Fig. S5). I127T, K130I, V131L were detected in the X protein of the occult HBV strain (Accession No. OK382075). The mutation analysis of the OBI case (patient no. 63) has been reported in tables 4 and 5. Moreover, the possible mutations in the S, pre-C, and BCP regions of the HBV genome are shown in Additional file 1: Table S6.