Investigation of hepatitis B virus mutations associated with immune escape and drug resistance in human immunodeficiency virus-infected patients

Study design and population

We applied a convenience sampling method to collect fifty archival sera in this descriptive exploratory investigation. The sample size was not determined, and we used samples that were already available. Sera were from people who underwent HIV testing at the Inkosi Albert Luthuli Central Hospital (NHLS-IALCH-NHLS) in Durban, KwaZulu-Natal Province, South Africa. Based on the sample size of 50 sera, we used the confidence level of 95% and the margin of error for our study was 14%. The 50 participants included men and women who previously tested positive for HBsAg and HIV. Participants received a written informed consent form, the information on the consent form was given in a language the patient understands, which is english and their native language (Zulu). Demographic data (age, sex, and ethnicity) were also collected. The specimen’s codes were de-linked to keep patients anonymous; with only additional data on the age and gender of the study participants provided.

Laboratory testing

Hepatitis B surface antigen (HBsAg) assay

The Monolisa HBsAg ultra confirmatory kit was used in according to the manufacturer’s instructions to perform an enzyme-linked immunosorbent assay (ELISA) on samples previously positive for HBsAg to confirm the presence of HBsAg marker (BioRad, Raymond Poincare, Marnes-la-Coquette, France). To identify and measure the presence of HBsAg, excess antibodies (anti-HBs; anti-HBs diluent: neutralization reagent) were used to neutralize 200 μL of undiluted sera specimens. At 450 nm, the optical density (OD) index of the sample was determined and compared to the cut-off value (COV) mean of a negative control. Reactive specimens for HBsAg were defined as those with an index greater than or equal to the COV.

DNA extraction of HBV

The sera obtained from patients were used to extract HBV deoxyribonucleic acid (DNA) using the QIAamp DNA Mini kit (catalog number: 51304) from (Qiagen, Hilden, Germany) following the manufacturer’s instructions. This technique allows the isolation and purification of total DNA from contaminants, inhibitors, and nucleases in the serum. An aliquot of 200 μL of the serum was added into 1.5 μL Eppendorf tube, to which 20 μL of proteinase K and 200 μL buffer AL (binding buffer mixed with poly [A] carrier RNA) was added. The mixture was pulse-vortex for 15 seconds to allow lysis of the mixture and destroy RNA, followed by a 10 minutes incubation at 56 °C. The mixture was then transferred to a QIAamp spin column to allow binding of the DNA and centrifuged for 1 minute at 8 000 rpm. The column was placed into a clean collection tube, then 500 μL of buffer AW1 was added, and it was centrifuged for 1 minute at 8 000 rpm. The solution was aspirated, 500 μL of buffer AW2 was added to purify the DNA, and it was followed by centrifugation for 3 minutes at 14 000 rpm. The QIAamp spin column was placed in a sterile 1.5 μL Eppendorf tube, and 50 μL of elution buffer (provided by the kit as buffer AE) was added directly into the column and incubated at room temperature for 5 minutes to precipitate the DNA. The DNA was eluted by centrifugation at 8 000 rpm for 1 minute and stored at -20 °C until further analyses were performed. The negative control, consisting of nuclease-free water, was included in the extraction procedure to identify contamination.

Polymerase chain reaction

First round and nested-PCR

A nested polymerase chain reaction (PCR) amplification of the overlapping surface/polymerase gene covering nucleotides 256 to 796 from EcoRI site was done as described previously (Mphahlele, 2008) with slight modification. Outer sense strand (forward primer) S1 (5′-CCT GCT GGT GGC TCC AGT TC-3′), and anti-sense strand (reverse primer) S2Na (5′-CCA CAA TTC KTTGAC ATA CTT TCC A-3′) were used. The master mix were prepared using Ampli Taq Gold DNA polymerase (ThermoFisher Scientific, Waltham, Massachusetts, United States). For each sample the following reagent volumes and concentration of the master mix were prepared as follows: 18.5 μL nuclease-free water, 2.5 μL of 1x PCR buffer with MgCl ₂, 0.5 μL (0.2 mM dNTP mix), 0.5 μL (10 μM) forward primer S1; 0.5 μL (10 μM) reverse S2Na anti-sense primer, 0.125 Taq DNA polymerase. A total of 22.1 μL of master mix was aliquoted into a 0.5 mL thin-walled PCR tube and 3 μL of DNA template was added. The PCR reaction mixtures (25.1 μL) was subjected to amplification of HBV DNA, carried out in an automated touch down thermal cycler CFX96 (Bio-Rad, Raymond Poincare, Marnes-la-Coquette, France). The HBV DNA amplification conditions were initial denaturation at 95 °C for 4 minutes, followed by 40× cycles involving denaturation at 95 °C for 4 minutes, annealing at 58 °C for 30 seconds, elongation at 72 °C for 1 minute, and final extension at 72 °C for 10 minutes.

Nested PCR

First round PCR product was used as a template for nested PCR. An aliquot of 3 μL of the first round PCR reaction was subjected to a nested PCR, the master mix volume and concentration were prepared as same for the first round PCR. The nested PCR conditions used were the same as first round PCR protocol except the annealing temperature at 55 °C for 45 seconds. Forward primers S6E (5′-GAGAAT TCCGAGGACTGG GGA CCC TG-3′) and reverse primer S7B (5′-CGG GAT CCT TAG GGT TTA AAT GTATAC C-3′) were used during nested PCR. The negative control consisting of nuclease-free water and a positive control were included in the PCR amplification assays.

PCR products verification

PCR amplification products were verified using 1% agarose gel (ThermoFischer, Waltham, Massachusetts) stained with 0.15 U/μL ethidium bromide (Biorad, California, USA). Aliquot of 10 μL PCR amplicon product was mixed with 2 μL 10x loading buffer (ThermoFischer, Waltham, Massachusetts). The mixtures were run on 1% gel along with a 1 Kb Invitrogen molecular weight maker (ThermoFischer, Waltham, Massachusetts) as a band size reference. The agarose gel was run at 100 V for 45 minutes. The gel was placed inside the ultraviolet (UV) transilluminator (Bio-rad, Hercules, California, United States) to visualise and image capturing.

Sequencing reaction

The PCR products and the nested PCR primers S6E and S7B were sent for bi-directional Sanger sequencing at the Inqaba (Inqaba Biotechnological Industry, Pretoria, South Africa). The amplicons were prepared for direct sequencing using the BigDye terminator v3.0 cycle sequencing ready reaction kit (catalog number: 4458687) from (ThermoFischer, Waltham, Massachusetts). Briefly, an aliquot of 50 μL of the 1:1 ratio of sodium acetate: ethanol (NaAc:EtOH) was added to the amplicons solution and centrifuged at 2000 g for 30 minutes. The well plates were inverted and centrifuged at 150 g for 1 minute. Pre-chilled 70% ethanol was added into the wells and then centrifuged at 2000 g for 5 minutes. The pellets were dried at 65 °C for 5 minutes followed by an addition of 10 μL Hi-Di formamide for 5 minutes and loaded into the sequencing machine ABI 3130XL genetic analyser (Applied Biosystems, Foster City, CA).

Sequences analysis

Nucleotide sequences of HBV chromatograms were viewed and edited by removing unwanted and mixed nucleotides character from the sequences by ChromasPro, version.1. The contiguous sequences were formed by joining overlapping DNA sequences of a gene using BioEdit. The consensus sequences were compared with the GenBank complementary genotype sequences using the basic local alignment search tool (BLAST). Representative sequences belonging to different genotypes were redeemed from GenBank to make comparisons with the study sequences. Multiple sequences alignment was performed with ClustalW within the MEGA software package version 7.0 (TomHall, North Carolina State University). Genotyping and identification of the sequence was done on Geno2pheno (Geno2pheno, 2023). Sequences were uploaded into the website where they were aligned at the sequence position 1 to 3182 (~3.18 kbp) relative to the Hepatitis B virus (taxon:10407) and a boostscan analysis of ≥ 80.0 was considered positive.

Mutations analysis

The aligned sequences were uploaded into the BioEdit and analysed for nucleotide base and amino acids changes. The sequences were further uploaded into the Geno2pheno to identify DRMs in the reverse transcriptase (RT) of polymerase and mutations in the overlapping S region.

Statistical analyses

Microsoft Excel and the data science statistical program STATA were used for data analyses (version 15). Excel (.csv.) file was imported into STATA and was used to calculate the frequency of age as numeric values and the frequency of HBsAg and mutations as categorical and numeric variables. The proportions of data reported as medians with interquartile ranges (IQR) for continuous variables (age) as frequencies and percentages for categorical variables (sex, ethnicity and mutations).

Ethical approval

The study ethics certificate was provided by the North-West University research ethics regulatory committee (ref. NWU-00068-15-A9).

Baseline demographic data

The baseline demographics of the study showed that all 50 HIV positive samples included were female at 64% (N=32/50) and 36% (N=18/50) male as shown in Table 1. The median age was 33 years [IQR: 18-55] and there was no statistical difference on age between the female and male black Africans at p=0.8.

Laboratory testing

HBsAg assay

The confirmatory screening of HBsAg reported a prevalence of 86% (N=43/50) with 12% (N=6/50) being negative and 2% (N=1/50) missing data ( Table 1). Majority of the female’s participant were HBsAg positive at 58% (N=29/50) when compared to the males at 28% (N=14/50) as shown in Table 1.

PCR amplification and genotyping

The PCR amplification of HBV DNA amplicons was successful in 82% (N=41/50) of the samples and consisting of overlapping surface/polymerase region. PCR amplification could not be obtained for 18% (N=09/50) samples. Only 92% (N=38/41) sequence products could be obtained by Sanger sequencing with only 26 sequences used to perform genotyping and 12 sequences were excluded due to the poor quality of the sequence reads. The genotypes of the sequences were confirmed by depositing nucleotides sequences into the Genotype2pheno database which showed that most of the nucleotide sequences had homology similarity to genotype A at 73% (N=19/26) and 12% (N=3/26) as genotype G and 15% (N=4/26) being genotype C ( Table 2). The study sequences with the highest similarity identity to genotype A were: (Q4P7 and Q7P7 to KX520697.1|South Africa; Q9P7 to U87728.1|South Africa; Q27P7 to AY233274.1|South Africa. Reference sequence of AY233277.1|South Africa had between 97%-99% similarity to fifteen sequences (Q5P7, Q8P7, Q10P7, Q13P7, Q18P7, Q19P7, Q20P7, Q21P7, Q22P7, Q23P7, Q24P7, Q26P7, Q27P7, Q29P7 and Q39P7). The sequences Q6P7, Q17P7, and Q42P7 had similarity to genotype G (EU694179.1|South Africa and sequences Q11P7, Q43P7, Q44P7, Q45P7 showed similarity to genotype C AB562444.1|Vietnam ( Table 2). Sub-genotypes of the sequences were confirmed by depositing nucleotides sequences into the Genotype2pheno database. The results retrieved from the Geno2Pheno database had a 96.85%-99.0% percentage of similarity to sub-genotype A1 for the genotype A sequences only.

Mutations within the surface region

The prevalence of mutations in the surface gene was 47% (N=18/38) and mutations were found in the “α”, “β”, “T” and outside the “α” epitope as shown in Table 3. The most common mutations on the surface region and their frequencies were S207N at 71% (27/38), followed by L216V and A194V at 23%, and the least prevalent being S204R, S117N, T143S, G145R, Y206H, P127T, Y200T, F129T and K122R all at 3% as shown in Table 3.

Mutations within the polymerase region

The prevalence of mutations in the RT of polymerase was reported at 36% (N=14/38). Mutations showed resistance to lamivudine (LMV) at (35% [5/14]), telbivudine (LdT) at (29% [4/14]), (14% [2/14]) for entecavir (ETV) and (21% [3/14]) for adefovir (ADV). Mutations causing resistance to LMV and LdT were M204V, L180M, V163I, and S202K. The S202K mutation causes resistance to ETV in addition to V204I and S169T. The ADV resistance mutations were I253Y, A236T and M250I in addition to complementary resistant mutation (L80I) not shown in the table. Multiple DRMs within a single sample were identified in one sample containing L180M, M204V, S202K and M250I mutations ( Table 3).