New Enzyme-Linked Immunosorbent Assay for Detection of Antibodies against Hepatitis Delta Virus Using a Hepatitis Delta Antigen Derived from a Taiwanese Clone and Comparison to the Abbott Radioimmunoassay

Abstract

An anti-hepatitis delta (HD) enzyme-linked immunosorbent assay (ELISA) using a specific recombinant hepatitis delta antigen derived from a local dominant hepatitis delta virus (hepatitis D virus; HDV) strain in Taiwan has been established. The detection efficiency of this assay was comparable to that of the commercially available Abbott anti-HD radioimmunoassay (RIA) and could be useful in routine laboratory diagnoses of HDV infection.

TEXT

Areas where hepatitis delta virus (hepatitis D virus; HDV) is endemic have been reported in Asia (23). In Taiwan, the prevalence of HDV infection in patients with chronic liver disease is between 5 and 12% (3). Analysis of the HDV nucleotide sequence reveals at least eight genotypes with unexplained variations and specific geographical distributions (5, 10, 23). The composition of HDV genotypes is particularly complex in Taiwan compared to that in other afflicted areas (25). However, the predominant genotype found in Taiwan is genotype II, which accounts for approximately 40% of the HDV infections in Taiwan (2, 13, 20, 22, 25), and its complete nucleotide sequence and amino acid sequence are approximately 69 to 78% related to those of type I and type III HDV (12).

Considering that concurrent infection by hepatitis B virus (HBV) and HDV not only causes more severe liver disease than infection with HBV alone but also influences an individual's response to therapy, every individual who is hepatitis B surface antigen (HBsAg) positive should be tested for HDV infection at least once (7, 10, 16, 23). To date, various serological enzyme-linked immunosorbent assays (ELISAs) have been developed for the diagnosis of HDV infection, in which HD antigen (HDAg) mainly came from liver tissues of HDV-infected animals or serum of HDV-infected individuals, and this can pose barriers to effective quality control (6, 9). In the present study, we have established an ELISA for the detection of anti-HD in human serum utilizing a specific recombinant HDAg (rHDAg) protein cloned from the local dominant HDV strain, and we have evaluated this new assay by comparing it with a commercially available radioimmunoassay (RIA).

The HDAg gene fragment of 971 bp, which included the coding region of the small-form HDAg of HDV, comprised of amino acids 1 to 127, was isolated using primers 5′-CGCCTAGCATATGATGAGCCAATCCGAGTCGAG-3′ and 5′-CCGGATCCCTACGGGAATCCCTGGTTTCC-3′ (12). Escherichia coli [BL21(DE3)plysS, Novagen] harboring the expression plasmid pET15b-SMII was grown in Luria-Bertani (LB) medium, and expression of the rHDAg was induced with isopropyl β-d-1-thiogalactopyranoside (0.4 mM). As shown in Fig. 1, the expected 23-kDa rHDAg protein was purified from soluble E. coli lysates using a HiTrap chelating HP column (Pharmacia, Sunnyvale, CA). Coats of purified rHDAg proteins (0.05 μg) were then applied to each well of a microtiter plate (Costar, Corning, NY), and the free binding sites were blocked with blocking buffer (10 mM potassium phosphate buffer [pH 7.2] containing 2.5% [wt/vol] bovine serum albumin). To test the presence of anti-HD antibodies, each serum sample was diluted (1:20) with blocking buffer and incubated in the antigen-coated wells at 37°C for 30 min. After 4 washes with washing buffer (phosphate-buffered saline [PBS] containing 0.05% Tween 20), the testing wells were then incubated at 37°C for 20 min with 1:20,000 diluted horseradish peroxidase-conjugated mouse monoclonal antibody against human IgG (Jackson ImmunoResearch Laboratories, West Grove, PA). After washing, the reaction was developed by adding 3,3′,5,5′-tetramethylbenzidine (KPL, Gaithersburg, MD), and the optical density at 450 nm was determined using a VERSAmax microplate reader (Molecular Devices, Sunnyvale, CA).

Fig 1.

Expression and purification of recombinant HDAg. Lane 1, crude extract from E. coli [BL21(DE3)pLysS] transfected with the expression plasmid (pET15b-SMII); lane 2, purified rHDAg protein. The arrow marks the position of the rHDAg protein fragment (23 kDa). M, molecular size ladder.

Patients who had evidence of other liver pathologies, such as hepatitis A or hepatitis C virus infection or alcohol-induced liver damage, were excluded from this study. In total, 220 HBsAg-positive serum specimens from the serum bank in the Liver Research Unit of Chang Gung Medical Center were assayed, and the performance of the anti-HD ELISA was evaluated using the Abbott anti-HD RIA (Abbott Laboratories, Chicago, IL) as the reference assay. Each sample was tested in triplicate, and the cutoff value was determined according to the Youden index. The sensitivity, specificity, positive predictive value, and negative predictive value of the anti-HD ELISA were 97.3%, 100%, 100% and 97.3%, respectively (Table 1). The kappa (κ) value was 0.973 (P < 0.001; 95% confidence interval, 0.942, 1.004).

Table 1.

Determination of the sensitivity and specificity of the in-house ELISA for anti-HD by two-by-two analysis with commercially available anti-HD RIA (Abbott) as the reference assay^a

Anti-HD ELISA result	No. of anti-HD RIA results		Total no.
	Positive	Negative
Positive	110	0	110
Negative	3	107	110
Total no.	113	107	220

^aThe sensitivity and specificity of the anti-HD ELISA were 97.3% and 100%, respectively.

To further confirm the results, the three serum specimens that were positive by the Abbott anti-HD RIA but negative by anti-HD ELISA were sent for HDV RNA detection using reverse transcription-PCR (RT-PCR). In brief, 200 μl of each serum sample was used for the extraction of HDV RNA, and the PCR primers used were 5′-CATGGTCCCAGCCTCCTCGCTGGC-3′ (nucleotides 695 to 718) and 5′-GAAGGAAGGCCCTCGAGAACAAGA-3′ (nucleotides 1264 to 1287). The PCR program was as follows: 5 min at 95°C; 35 cycles of 1 min at 95°C, 1 min at 64°C, and 1 min at 72°C; and 10 min at 72°C. The expected size of the PCR product was 592 bp (11). No HDV RNA was detected in any of the three specimens.

Because HDV-infected individuals develop a specific humoral response to HDAg, which is the only viral capsid protein known to be encoded by the HDV genome (21), detection of anti-HDAg in serum is a practical approach to confirm HDV infection and is of particular diagnostic value given that viremia lasts only a few weeks (6). Since nucleotide sequence analysis has revealed that the RNA sequence of the HDV genome is highly variable (10), it would be particularly useful to derive the viral assay antigen from the prevalent local strain. The rHDAg gene used in this study was cloned from the locally prevalent HDV genotype II (known as Taiwan-3), and its nucleic acid and amino acid sequences have 93.8% and 89.3% homology, respectively, to an HDV genotype II (Japan-1) isolated from Japan (12).

From a practical point of view, a sufficient supply of reliable and harmless viral protein for use as a screening antigen is also important. For anti-HD antibody detection, various RIAs and enzyme immunoassays (EIAs) have been developed (4, 8, 9, 17, 18). One major drawback of these tests is that they use HDAg derived from either infected liver or infectious serum of humans or animals (e.g., chimpanzees and woodchucks) (15). Thus, the potentially biohazardous nature and limited supply of these materials have been major limitations in clinical use (14, 17, 18). The rHDAg used in this study was expressed and harvested from E. coli, allowing for a theoretically unlimited supply of HDAg free from the potential hazards of extracting antigen from infectious materials.

It is worth noting that false-positive results have also been reported previously using the Abbott anti-HD RIA, which uses HDAg extracted from an infected woodchuck, for anti-HD antibody detection (1, 6, 19). Intriguingly, a false-positive reaction was also observed in three samples using the Abbott anti-HD RIA but not in our anti-HD ELISA, and these results have been confirmed by RT-PCR in the present study. The reason for these conflicting results is unclear. It is possible that because the HDAg used in our in-house ELISA was derived from a local viral strain, it might therefore have allowed for better discrimination between positive and negative sera. To clarify this hypothesis, a better knowledge of HDAg epitope specificities is badly needed.