Analysis of Hepatitis B Virus-Immunoglobulin Isotype Complexes by a Novel Immuno-Capture Polymerase Chain Reaction Method
Abstract
Hepatitis B virus (HBV) can be present in the circulating blood either as free virus or as a virion-immunoglobulin (Ig) complex. Presently, it remains unclear what specific role each Ig plays in the clearance of HBV. In this study, a novel method that combined immuno-capture and polymerase chain reaction (PCR) amplification was used for detecting and distinguishing different HBV-Ig complexes. Three isotypes of Ig (IgM, IgG and IgA) bound to HBV were detected in the four clinically defined stages of HBV infection in 108 patients. The results showed that all the three isotypes of Ig could bind to HBV, and the patterns of HBV-Ig complexes varied according to disease categories. Interestingly, the frequency of HBV DNA-Ig complexes in hepatitis B e antigen (HBeAg)-positive patients was significantly lower than that in HBeAg-negative patients. All the data suggest that the three isotypes of HBV DNA-Ig circulating immune complex (CIC) may have different biological meanings. In summary, HBV bound to an antibody is a common feature of hepatitis B, and immuno-capture PCR is a valuable method for the analysis of the composition of the immune complexes. The detection of HBV-Ig complexes may provide new and valuable insights into HBV pathogenesis.
Introduction
Hepatitis B virus (HBV) is a major problem and a common cause of cirrhosis and hepatocellular carcinoma [1–3]. Immune responses to components of HBV are assumed to play an essential role not only in the elimination of virus but also in the pathogenesis of HBV-induced hepatitis [4, 5]. HBV can cause cytolytic infections in which viral replication occurs at the expense of the host cell viability [6–8]. However, the host's innate and adaptive immune defences can also be altered to eliminate the virus and terminate the infection [9, 10]. This process requires the participation of B lymphocytes and plasma cells which produce antibodies. Antibodies may bind viral proteins including intact viral envelope proteins to construct virion-immunoglobulin (Ig) complexes, leading to the clearance of the virus particles [11, 12]. Previous studies on the pathological role of circulating immune complexes (CICs) have shown that different Ig isotypes in CIC may have different functions in the pathogenesis of HBV infection [13]. However, information regarding the Ig isotype representation in the HBV-Ig complexes is currently not available because of the lack of useful methods to detect them.
Recently, we have developed a novel immuno-capture reverse transcriptase-polymerase chain reaction (iRT-PCR) method to detect three isotypes of HCV RNA-Ig CIC [14]. The study on HCV-bound isotypes of Ig CIC by the iRT-PCR showed that Ig-complexed virions and free virions might have different biological consequences, with the latter being elusive to immunological elimination, and that defective IgG in HCV-CIC might contribute to long-term viraemia [15]. Furthermore, we found that there were higher frequency of HCV-C1q CIC than that of HCV-factor B and deviation of complement from Ig in HCV-CIC by the detection of HCV-bound CIC with the use of iRT-PCR method [16]. When the principle of iRT-PCR was used in the detection of bacteria, Aeromonas hydrophila and Shigella species and Shiga toxin-producing bacteria could be identified rapidly and conveniently [17–19]. In this study, we extend the immuno-capture PCR (iPCR) technique to detect HBV DNA extracted from IgM-, IgG- or IgA-bound CIC virions captured by antihuman μ-, γ- or α-chain antibodies coupled to polystyrene 96-well plates, respectively. Our results demonstrated that HBV bound to antibody was a common feature of hepatitis B, and the iPCR for HBV-Ig CIC was a valuable method for the analysis of the composition of the immune complexes.
Materials and methods
Study population This study included 108 patients with hepatitis B – 61 males and 47 females. The age range of these patients was from 18 to 58 years (median, 31 years). Alanine aminotransferase (ALT), aspartic aminotransferase (AST), total bilirubin (TBIL) and direct reacting bilirubin (DBIL) tests were carried out in 108, 98, 99 and 99 samples, respectively. Serological tests with respect to hepatitis B surface antigen (HBsAg), anti-HBs and anti-HBcore total were performed in 108 samples, and hepatitis B e antigen (HBeAg) and anti-HBe in 76 samples. According to the International Group's Criteria, these patients were classified into acute hepatitis (AH, 20 patients), chronic hepatitis (CH, 56 patients), acute fulminant hepatitis (AFH, 19 patients) and post-hepatitis B cirrhosis (PHC, 13 patients). Forty healthy blood donors (25 men, 15 women, age range 21–45 years, median 30 years) were used as normal controls, and five patients with hepatitis C were used as disease controls in this study. Patients with hepatitis C as well as with other causes of liver diseases were eliminated, as previously described [14]. Informed consent was obtained from each subject; the protocol for this study was approved by the Institutional Review Board of Xiamen University.
Blood collection and HBV DNA isolation Blood samples collected from all individuals in this study were kept at −70 °C until use. HBV DNA isolation procedures were followed as described [14]. Briefly, 200 µl of the sera was precipitated by the addition of polyethylene glycol (PEG-6000) to a final concentration of 3.5% at 4 °C overnight. Following centrifugation at 3000 × g at 4 °C for 20 min, precipitates were suspended in 200 µl of 0.1 m boric acid buffer (pH 8.4) and precipitated again with PEG overnight, and then resuspended in 200 µl of 0.01 m, pH 7.4, phosphate-buffered saline (PBS). 7.5 µg/ml of antihuman μ-, γ- and α-chain antibodies (Institute of Nanzhou Bioproducts, Nanzhou, China) was separately coupled to polystyrene 96-well plates using 0.1 m carbonic acid buffer (pH 9.6) at 4 °C for 18 h. Plates were then washed with PBS containing 0.05% Tween-20 and incubated with 40 µl of the PBS dilution of precipitates at 37 °C for 1 h. Following washing, the wells were incubated with 40 µl of sterile distilled water at 95 °C for 20 min to draw HBV DNA template for the iPCR.
PCR procedure Primers were designed on the basis of the sequence of pADR-1 of HBVadr subtype; the forward primer was (1861–1878) 5′-TGTTCAAGCCTCCAAGCT-3′ and the reverse was (2268–2284) 5′-CTCACACCTAAGCGTGA-3′ and were synthesized by Sagon Technologies (Shanghai, China). PCR was carried out in a 50 µl volume of a reaction mixture containing 5 µl of 10× PCR buffer, 4 µl of MgCl2 (25 mmol/l), 1 µl of dNTPs (10 mmol/ml for each), 15 pmol of each primer, 1 U of Taq polymerase and 30 µl of HBV DNA from HBV-Ig complexes isolated by the preceding procedure for 35 cycles, using a thermoprofile of 94 °C for 1 min (denaturation), 55 °C for 30 s (annealing) and 72 °C for 1 min (extension), with a final extension of 10 min at 72 °C. The amplified DNA products were separated by electrophoresis on 3% agarose gels. HBV DNA in whole serum were detected by both heat-denatured PCR (40 µl of sera was denatured at 100 °C for 15 min and PCR carried out as described as above) and commercial PCR kit for the comparison of sensitivity. Commercial PCR kit was obtained from Tailun (Xiamen, China) and used according to the manufacturer's instructions.
Specificity check of the iPCR Blocking, neutralizing, replacing and repeating tests were performed to check the specificity of the immuno-capture method. For the blocking test, human IgM, IgG or IgA was put into wells in a blocking group and PBS in a control well before samples were added. In the neutralizing test, samples were mixed with antihuman μ-, γ- and α-chain antibodies in a neutralizing group or with PBS in a control at 37 °C for 1 h and were then added into wells. In the replacing test, five samples of HCV infection instead of patients with hepatitis B were detected, and the results were negative. In addition, a group with three positive and three negative samples was tested by this method 15 times; all of the 45 positive samples were positive and 45 negative samples were negative.
Statistical analysis Differences between groups were tested for significance using χ2-test, with Yates's continuity correction for small numbers, and three significance levels (0.05, 0.01 and 0.001) were used.
Results
High specificity of the iPCR procedure
Specificity primers were used to amplify HBV DNA; the fragment of PCR product was about 424 bp. It was consistent with the expected result. To further check the fidelity of PCR, Sau3A I was used to digest PCR products. Their restriction mappings were equal to the expected fragments. No PCR product was seen with the control samples. The results from the blocking test and the neutralizing test showed that the iPCR was specific (Fig. 1).
Restriction mapping and electrophoresis analysis of polymerase chain reaction (PCR) products. M, marker; lane 1, PCR product (424 bp); lane 2, the restriction mapping (Sau3A I) of PCR product; lane 3, neutralizing test for the specificity of this immuno-capture method by the detection of samples mixed goat antihuman μ-, γ- and α-chain antibody for 1 h at 37 °C; lane 4, blocking test for specificity of this immuno-capture method by adding healthy serum to the wells for 1 h at 37 °C before samples.
Increased sensitivity of iPCR than conventional PCR
Heat-denatured PCR, PCR commercial kit and the iPCR method detected HBV DNA in 23 HBV-infected patients. The frequency of HBV DNA was detected to be 52.5, 56.5 and 87.0%, respectively. The last value was significantly higher than the first two values (Table 1).
| Frequencies (positive numbers of HBV DNA-Ig/total numbers tested) | ||
|---|---|---|
| (1) | Heat-denatured PCR for whole serum | 52.5% (12/23) |
| (2) | Commercial kit PCR for whole serum | 56.5% (13/23) |
| (3) | Immuno-capture PCR for CIC* | 87.0% (20/23) |
- (1) versus (2), χ2=0.088, P> 0.05; (1) versus (3), χ2=5.03, P< 0.05; (2) versus (3), χ2=3.84, P< 0.05. Ig, immunoglobulin; CIC, circulating immune complex.
- * At least one positive for HBV DNA-Ig CIC.
Identification of three isotypes of HBV DNA-Ig CIC in four clinical types of hepatitis B
Three isotypes of HBV DNA-Ig CIC were detected in four clinical types of hepatitis B by iPCR. Of the 108 samples tested, 101 patients were positive by at least one of the HBV-Ig tests. The frequencies of HBV DNA-IgM CIC, HBV DNA-IgG CIC and HBV DNA-IgA CIC were 72.3% (73/101), 58.4% (59/101) and 68.3% (69/101), respectively. No significant difference was found between the three CICs (Table 2).
| Categories | Cases | (1) IgM CIC (%) | (2) IgG CIC (%) | (3) IgA CIC (%) |
|---|---|---|---|---|
| AH* | 18 | 10 (55.6) | 11 (61.1) | 17 (94.4) |
| CH | 52 | 38 (73.1) | 30 (57.7) | 34 (65.4) |
| AFH† | 18 | 16 (88.9) | 9 (50.0) | 10 (55.6) |
| PHC | 13 | 9 (69.2) | 9 (69.2) | 8 (61.5) |
| Total | 101 | 73 (72.3) | 59 (58.4) | 69 (68.3) |
- AH, acute hepatitis; CH, chronic hepatitis; AFH, acute fulminant hepatitis; PHC, post-hepatitis B cirrhosis.
- * (1) versus (3), P=0.020; (2) versus (3), P=0.046.
- † (1) versus (2), P=0.032.
Relation between difference in the patterns of HBV DNA-Ig CIC and clinical categories
Comparison between the three isotypes of HBV DNA-Ig CIC in four clinical types of hepatitis B showed that the frequency of HBV DNA-IgA CIC was significantly higher than that of HBV DNA-IgM CIC or HBV DNA-IgG CIC in AH, and the frequency of HBV DNA-IgM CIC was significantly higher than that of HBV DNA-IgG CIC or HBV DNA-IgA CIC in AFH. There were no significant differences in other clinical groups (Table 2). Furthermore, the frequencies of single positive of HBV DNA-Ig CIC were characterized in this study. There was only HBV DNA-IgA in AH and HBV DNA-IgM in AFH, but the three isotypes of single positive of HBV DNA-Ig CIC were detected in CH and PHC (Table 3). Comparison of the frequencies between one and more than one positive for HBV DNA-Ig CIC indicated that the frequency of the former group (35.6%) was significantly lower than that of the latter group (64.4%) in 101 samples, with at least one positive for HBV DNA-Ig CIC. Moreover, the frequencies of one positive for HBV DNA-Ig CIC were 27.8% (5/18), 40.4% (21/52), 27.8% (5/18) and 38.5% (5/13) in AH, CH, AFH and PHC, respectively, while the frequencies of the more than one positive for HBV DNA-Ig CIC were 72.2% (13/18), 59.6% (31/52), 72.2% (13/18) and 61.5% (8/13) in AH, CH, AFH and PHC, respectively. There were significant differences between them in AH, CH and AFH (Table 4).
| Categories | Cases | (1) IgM CIC (%) | (2) IgG CIC (%) | (3) IgA CIC (%) |
|---|---|---|---|---|
| AH* | 18 | 0 | 0 | 5 (27.8) |
| CH | 52 | 10 (19.2) | 5 (9.6) | 6 (11.5) |
| AFH† | 18 | 5 (27.8) | 0 | 0 |
| PHC | 13 | 1 (7.7) | 2 (15.4) | 2 (15.4) |
| Total‡ | 101 | 16 (15.8) | 7 (6.9) | 13 (12.9) |
- AH, acute hepatitis; CH, chronic hepatitis; AFH, acute fulminant hepatitis; PHC, post-hepatitis B cirrhosis.
- * (1) versus (2), P=0.045; (2) versus (3), P=0.045.
- † (1) versus (2), P= 0.045; (1) versus (3), P=0.045.
- ‡ (1) versus (2), χ2=3.97, P< 0.05.
| Categories | Cases | One positive | More than one positive | P |
|---|---|---|---|---|
| AH | 18 | 5 (27.8) | 13 (72.2) | < 0.05 |
| CH | 52 | 21 (40.4) | 31 (59.6) | < 0.05 |
| AFH | 18 | 5 (27.8) | 13 (72.2) | < 0.05 |
| PHC | 13 | 5 (38.5) | 8 (61.5) | > 0.05 |
| Total | 101 | 36 (35.6) | 65 (64.4) | < 0.001 |
- AH, acute hepatitis; CH, chronic hepatitis; AFH, acute fulminant hepatitis; PHC, post-hepatitis B cirrhosis.
In the samples positive for more than one of HBV DNA-Ig CIC, there are three types of combinations in two positive for HBV DNA-Ig CIC (HBV-IgM with HBV-IgG, HBV-IgM with HBV-IgA and HBV-IgA with HBV-IgG) and one type of combination in the three positive for HBV DNA-Ig CIC (HBV-IgM with HBV-IgG and HBV-IgA). Comparisons from the four types of combinations showed the frequencies of the three combinations were significantly higher than those of the two combinations in AH, CH and PHC (Table 5).
| Categories | Cases | (1) IgM+IgG | (2) IgM+IgA | (3) IgG+IgA | (4) IgM+IgG+IgA |
|---|---|---|---|---|---|
| AH* | 18 | 1 | 2 | 3 | 7 |
| CH† | 52 | 3 | 6 | 3 | 19 |
| AFH | 18 | 3 | 4 | 2 | 4 |
| PHC‡ | 13 | 2 | 1 | 0 | 5 |
| Total§ | 101 | 9 | 13 | 8 | 35 |
- AH, acute hepatitis; CH, chronic hepatitis; AFH, acute fulminant hepatitis; PHC, post-hepatitis B cirrhosis.
- * (1) versus (4), P=0.0455.
- † (1) versus (4), χ2=12.97, P< 0.001; (2) versus (4), χ2=8.89, P< 0.001; (3) versus (4) χ2=12.97, P< 0.001.
- ‡ (3) versus (4), P=0.039.
- § (1) versus (4), χ2=19.64, P< 0.001; (2) versus (4), χ2=13.23, P< 0.001; (3) versus (4), χ2=21.18, P< 0.001.
Significant higher frequency of HBV DNA-Ig CIC in HBeAg-negative patients than that in HBeAg-positive patients
The distributions of the three isotypes of HBV DNA-Ig CIC were analysed between ALT, AST, TBIL, DBIL normal and abnormal groups, and HBsAg, anti-HBs, HBeAg, anti-HBe and anti-HBcore total positive and negative groups. The distributions of these complexes are in concordance with the distributions of HBV DNA in whole serum except HBeAg group (data not shown). There were 17 samples with positive HBeAg and six samples with negative HBeAg in 23 patients detected by heat-denatured PCR and commercial PCR kit. The frequencies of HBV DNA were 64.7% (11/17) and 70.6% (12/17) in HBeAg-positive group, and 16.7% (1/6) and 16.7% (1/6) in HBeAg-negative group. There were significant differences between them. Of the 108 patients studied in this paper, three isotypes of HBV DNA-Ig CIC were analysed in 76 patients, with 40 HbeAg-positive and 36 HbeAg-negative. The frequencies of three isotypes of HBV DNA-Ig CIC were all significantly lower in the group with positive HBeAg than that in the group with negative HBeAg (Table 6).
| Categories | IgM-CIC (%) | IgG-CIC (%) | IgA-CIC (%) | Heat-denatured PCR (%) | Commercial kit PCR (%) |
|---|---|---|---|---|---|
| HBeAg (+) | 20/40 (50.0)*† | 15/40 (37.5)† | 21/40 (52.5)† | 11/17 (64.7)† | 12/17 (70.6)† |
| HBeAg (–) | 28/36 (77.8) | 24/36 (66.7) | 29/36 (80.6) | 1/6 (16.7) | 1/6 (16.7) |
- PCR, polymerase chain reaction.
- * Positive numbers of HBV DNA/total numbers tested (frequencies).
- † P< 0.05 (comparison between group with positive HBeAg and group with negative group).
Difference in the detection of HBV DNA between Ig CIC and whole serum
HBV DNA was detected in Ig CIC and whole serum simultaneously. The results showed that 12 were positive and two negative in both of the samples, and eight negative in whole serum were detected to be positive in Ig CIC (Table 7). There was a significant difference between Ig CIC and whole serum (P < 0.05).
| HBV DNA-Ig CIC | |||
|---|---|---|---|
| HBV DNA in whole serum | Positive number | Negative number | Total |
| Positive number | 12 | 1 | 13 |
| Negative number | 8 | 2 | 10 |
| Total | 20 | 3 | 23 |
- P< 0.05. Ig, immunoglobulin.
Discussion
In the study, by using goat IgG to human μ-, γ- and α-chain as capture antibodies to capture the isotypes of IgM, IgG and IgA to anti-Dane particles and using the strongly amplified and prime-specific PCR method to detect HBV DNA in captured substances, an iPCR to detect three isotypes of HBV DNA-Ig CIC was established. The results from blocking, neutralizing, replacing and repeating tests showed that this technique was highly specific, sensitive and reproducible. Three isotypes of HBV DNA-Ig CIC (HBV DNA-IgM CIC, HBV DNA-IgG CIC and HBV DNA-IgA CIC) were characterized in four clinical types of hepatitis B by the iPCR. In 108 patients with hepatitis B detected by this method, the frequencies of HBV DNA-IgM, -IgG and -IgA CIC were 67.6, 54.6 and 63.9%, respectively. Although there were no significant differences between the three isotypes of HBV DNA-Ig CIC, we interestingly found the association of the frequencies of HBV DNA-Ig CIC with clinical categories. These results suggest that the different characters of Igs may have a different role in complex virus.
In the normal situation, CIC formed by virus and antibody was the main pattern for the body to clear virus. In the present study, frequencies between being positive for one and more than one of the three isotypes of HBV-Ig CIC indicated that the frequencies of the former group were significantly lower than that of the latter group in AH and AFH (25.0% versus 65.0% and 26.3% versus 68.4%, respectively), but there was no significant difference between them in CH. Analysis of single positive of HBV DNA-Ig CIC in four clinical diseases found that there was only one single Ig positive of HBV DNA-Ig CIC in AH and AFH, whereas, interestingly, there were all of three single positive isotypes of HBV DNA-Ig CIC in CH and in PHC. These data suggest that HBV DNA-Ig CIC is related to clinical categories of hepatitis B, which may contribute to the chronicity and recovery from this disease.
Many studies reported that HBeAg, a marker of replicating HBV, is correlated with the frequency of HBV DNA in patients with HBV infection [20, 21]. In the present study, we also found that the frequency of HBV DNA, detected by heat-denatured PCR and commercial PCR kit, was significantly higher in the group with positive HBeAg than in the group with negative HBeAg. However, the frequencies of three isotypes of HBV DNA-Ig CIC, detected by the iPCR, were significantly lower in HBeAg-positive group than in HBeAg-negative group. These data suggest that the detection of HBV DNA in CIC might be of unique character. Indeed, our data show that there was a significant difference between detection of HBV DNA in Ig CIC and whole serum.
In summary, our results suggest that iPCR, a sensitive, specific and reproducible new method, can be applied in the analysis of HBV-Ig complexes. Moreover, blood samples from HBV infections contain different Ig isotypes with HBV in their CIC. And, HBV DNA-Ig CIC may have different biological meanings other than antigen/antibody CIC in patients with HBV.
Acknowledgments
This work was sponsored by grants from the Foundation of Chinese Education Ministry for Excellent Young Teachers, the State Foundation of China for Natural Science (No. 39660075) and the Foundation of Fujian Province for Natural Science (No. C97016).
