Hepatitis B Splice-Generated Protein Antibodies in Syrian Chronic Hepatitis B Patients: Incidence and Significance

Background: Previous studies have suggested hepatitis B splice-generated protein (HBSP), when expressed, is involved in the pathogenesis of HBV infection. Objectives: We aimed to evaluate anti-HBSP incidence and association with several HBV infection parameters in a group of Syrian chronic hepatitis B patients. Patients and Methods: Eighty treatment-naïve HBsAg-positive adult chronic hepatitis B patients' sera were included in our prospective targeted study. Liver function, virological and histological tests results were obtained from patients’ medical files. Three variants of a 20-mer HBSP-derived peptide were designed based on HBV genome sequences obtained from Syrian patients' sera (GenBank Accession No. JN257148-JN257217). Microtiter plate wells were coated with the synthetic peptides and used to detect anti-HBSP antibodies by an optimized indirect enzyme-linked immunosorbent assay (ELISA). Samples were considered positive when showed optical density (OD) values higher than the cut-off value for at least one peptide variant. Results: Seven out of eighty (9%) CHB patients were positive for anti-HBSP antibodies. Mean OD values were not significantly different between HBeAg-positive and -negative patients (P > 0.05). OD values showed weak positive correlation with ALT and AST values (P < 0.05), and weak to moderate positive correlation with liver biopsy staging ranks (P < 0.05). No significant correlation was revealed with viral load values or liver biopsy grading ranks (P > 0.05). Conclusions: We introduced an anti-HBSP antibodies ELISA, designed for locally circulating HBV strains. Correlation observed of Anti-HBSP with liver fibrosis staging regardless of viral replication and liver inflammation suggests anti-HBSP antibodies as possible indicator for HBV-associated liver fibrosis.

HBV infection is a serious global health issue. More than 240 million chronic hepatitis B (CHB) patients worldwide are at high risk of death due to cirrhosis and hepatocellular carcinoma (HCC) (1). In Syria, HBV infection is intermediately endemic (5-7%) and genotype D is predominant (2). Hepatitis B virus (HBV) is a DNA retro-transcribing virus including a circular 2.3 kb-length partially doublestranded DNA (dsDNA) genome with four overlapping open reading frames (ORFs) (3,4). Splicing events in the viral mRNAs that might be subsequently encapsidated and retro-transcribed giving rise to defective viral particles have been reported in chronic hepatitis B (CHB) infection (5)(6)(7)(8)(9). Consequently, splice-generated viral proteins might be produced. A viral 111 aa-length protein generated by a fusion of HBV polymerase N-terminal to a new open reading frame, and encoded by a singly spliced mRNA has been reported (10,11). This immunogenic hepatitis B splice-generated protein (HBSP) has been detected in the liver biopsies of patients with active chronic hepatitis (10,12) and its involvement in the liver disease pathogenesis has been suggested (13). Antibodies to HBSP have been found in CHB patients sera and anti-HBSP detection has been proposed as a marker of HBV-related disease (12).

Objectives
The present study aimed at designing a semi-quantitative enzyme-linked immunosorbant assay (ELISA) to detect antibodies to hepatitis B spliced protein, and evaluate anti-HBSP incidence and association with HBV infection parameters in a group of Syrian chronic hepatitis B patients.

Specimens
Our prospective targeted study recruited eighty treatment-naive HBsAg-positive adult patients diagnosed with chronic HBV infection by credentialed gastroenterologists. None of the CHB patients manifested co-infection with HCV, HDV or HIV (anti-HCV-negative, anti-HDV-negative and anti-HIV-negative), or were alcohol-consuming or immuno-suppressed. Liver function tests (ALT and AST), virological markers (HBeAg and HBV DNA) and histological analysis, which was assessed according to Scheuer's classification for grading and staging of chronic hepatitis (14), were performed within maximally 4-week period around our study serum sampling. All aforementioned tests results were obtained from patients' medical files. Forty-six HBsAg-negative, anti-HCV-negative healthy adults were also enrolled to obtain control sera. After the ethical committee's approval, written informed consents were obtained and peripheral blood specimens were drawn from all patients and healthy individuals. All sera were kept in -80°C.

Antigen Coating
High-binding polystyrene microtiter plate wells (R&D Systems, Minneapolis, USA ) were coated with 1 µg of a synthetic peptide using carbonate buffer (pH = 9.6) at 4°C overnight. Separate wells were dedicated for each synthetic peptide.

Anti-HBSP Antibodies Detection
Indirect enzyme-linked immunosorbent assay (ELISA) was performed for all patients and control sera in duplicate for the three peptide variants. Fifty microliters of serum diluted 1:100 in PBS, 0.05% Tween, and 1% PVP (pH 7.4) were added to each well and incubated for 1 hour at 37°C. After washing with PBS and 0.05% Tween, 50 µL of anti-human IgG HRP conjugate (1:2500) (Promega, Minneapolis, USA) were added, incubated for 30 minutes at 37°C and washed. Fifty microliters of freshly prepared OPD substrate solution (0.4 mg/mL o-phenylenediamin (OPD), 0.5 μl/ml 30% H 2 O 2 in 0.1 M citric acid adjusted to pH 5.0 by NaOH) were added and incubated for 15 minutes. Optical density (OD) was measured by a spectrophotometer at 490 nm. The cut-off value was determined for each peptide variant as the mean OD of all control sera plus 3-fold the standard deviation (SD). The grey zone ranged between the mean OD of all control sera plus 2-fold the standard deviation and the cut-off value. Samples were considered positive when showing OD values higher than the cut-off value for at least one peptide variant.

Statistical Analysis
Mean OD values were compared using Student's t test and difference was considered significant when P < 0.05. Correlation of OD values with other serological, virological and histological markers was assessed using Pearson's and Spearman's correlation coefficients. All analyses were performed using Microsoft Excel 2010 and IBM SPSS Statistics 17.0 software (International Business Machines Corp., New York, USA).

Discussion
Our anti-HBSP antibodies semi-quantitative ELISA showed significant discrimination between chronic hepatitis B patients and the control sera (P < 0.001). However, a relatively low incidence rate (9%) of anti-HBSP antibodies among Syrian CHB patients enrolled in our study compared to previous studies (12) might be attributed to different technical conditions adopted including synthetic peptide variants and ELISA procedure. Furthermore, inability to detect the antibodies due to low antibody levels, antigen-antibody complex formation (21), or HBSP production down-regulation (20) might be speculated. Thus, anti-HBSP antibodies incidence rate might have increased if follow-up and retesting were accomplished for CHB patients showing negative or grey-zone results.
In the present study, no association was seen between anti-HBSP antibody detection and viral replication manifested by the viral loads and HBeAg status. HBSP hypothesized mechanism of action does not seem to impact viral replication (12). HBeAg production might be disabled by precore and basal core promoter mutations. Hence, predominant HBeAg-negative status in Syria does not necessarily indicate better prognosis or less replication (2).
Conversely, Anti-HBSP antibodies status correlated moderately with fibrosis severity indicated by the liver biopsy staging ranks. This finding might be due to HBSP-induced apoptosis in the liver hepatocellular cells (22,23), which might diminish the immune neutralization due to the viral particles spread (24). Accordingly, lack of/weak correlation of anti-HBSP antibodies status with necroinflammation grades/aminotransferase levels, respectively, is rationally explained. Furthermore, hepatic steatosis was found in three CHB patients' liver biopsies with fibrosis staging S1 (one patient) or S2 (two patients). However, this was overlooked since the association of liver steatosis with fibrosis severity in CHB infection is controversial (25)(26)(27). Our ELISA results using the three synthetic peptide variants were partially consistent. In particular, the peptide variant two could not detect all anti-HBSP positive sera. Moreover, the results of the peptide variant three showed the highest correlation coefficient with fibrosis severity. This might suggest variant three (LLLKE-PLCIPPVAVQNLRTE) to be efficient for detecting anti-HBSP antibodies.
Albeit semi-quantitative, our assay paved the way to a tentative fibrosis assessment. Hence, developing the anti-HBSP ELISA described herein into a quantitative assay and evaluating its sensitivity and specificity empirically might be advisable to determine anti-HBSP antibodies levels discriminating different liver biopsy staging ranks. In conclusion, we introduced an anti-HBSP antibody ELISA, purposely designed based on genomic sequences obtained from HBV strains circulating in the Syrian population. Although anti-HBSP test did not show any significant association with parameters of viral replication and liver inflammation, it delivered partial indication on liver fibrosis associated with chronic hepatitis B.