This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
Different HBV genotypes have effect on the progression and outcome of chronic hepatitis B and on the response to antiviral therapy. Accurate and sensitive genotyping methods are needed in clinical practice.
Methods In this multicenter retrospective study, 504 HBV DNA positive serum or plasma samples of CHB patients or asymptomatic carriers were genotyped by INNO-LiPA HBV genotyping and direct sequencing. A part of the consistent and discrepant results were confirmed by clonal sequencing.
Results Except two samples with indeterminate results by INNO-LiPA, 463 had the same genotype results by INNO-LiPA and direct sequencing with a concordance of 92.23% (463/502). However 24 had completely inconsistent results by the two methods, seven of which were confirmed by clonal sequencing as the same genotype as by direct sequencing. Fifteen samples were determined as B/C mixed genotypes by INNO-LiPA but singe B or C genotype by direct sequencing, five of which were confirmed by clonal sequencing as B/C mixed infection. Further studies showed INNO-LiPA could detect 6% of genotype B and 5% genotype C at HBV DNA â‰¥3Ã-103 IU/ml.
Conclusions INNO-LiPA is accurate and sensitive in detecting B, C, D and B/C mixed genotype infection in Chinese CHB patients and asymptomatic carriers.
Keywords: Hepatitis B virus; Genotype; Line probe assay
Hepatitis B virus is endemic and remains as a major public health problem in China. About three hundred thousand Chinese patients with chronic hepatitis B (CHB) die of either complications of cirrhosis or hepatocellular carcinoma each year. Unlike in America or Europe, in China genotype B and C are predominant accounting for 41% and 53% in chronic HBV carriers, respectively, and genotype D is distributed in the west and northwest [1, 2]. The clinical significance of HBV genotype has been reported in the past few years. Firstly, genotype has an effect on the progression of CHB. For instance, genotype C is an independent risk factor for the development of hepatocellular carcinoma in addition to liver cirrhosis compared with genotype B . Secondly, different genotypes have a very important impact on response to antiviral therapy. For instance, genotype C is characterized by a less favorable sustained response to alfa-interferon therapy in patients with HBeAg positive hepatitis, and is also correlated to a lower rate of HBeAg and HBsAg seroconversion than genotype B . Thirdly, the mixed genotype infection has been increasingly recognized recently to have a poorer outcome than single genotype infection [5-7]. For these reasons, HBV genotyping may be used to identify patients at high risk for disease progression and predict the outcome of antiviral therapy in clinical laboratories.
Based on nucleotide sequence divergence of more than 8% in the whole genome, HBV can be classified to eight genotypes termed from A to H [8-11]. Up to date, several methods have been developed for the determination of HBV genotype including sequencing based methods (clonal sequencing and PCR product direct sequencing) [8-11], restriction fragment length polymorphism (RFLP) , PCR sequence specific primer , and hybridization based method such as microarray and INNO-LiPA genotyping[13-15]. Clonal sequencing for the entire genome remains as the gold standard for HBV genotyping; whereas it's time consuming and labor intensive. Both microarray and RFLP often suffer from the lack of quality control. In clinical laboratories the INNO-LiPA genotyping and direct sequencing are more practical because they are relatively simple, rapid and reliable. LiPA, a line probe assay, is based on reverse hybridization principle such that biotinylated amplicons hybridize to specific oligonucleotide probes that are immobilized as parallel lines on membrane-based strips . It has been reported to be an easy and quick tool for genotyping with high sensitivity and specificity [13-15]. However, there have been few data on the performance of INNO-LiPA genotyping in Chinese population, which has a markedly different pattern of HBV genotype distribution. Therefore, we carried out a multicenter research on 504 serum or plasma samples of CHB patients or asymptomatic carriers from China mainland to evaluate the test performance of INNO-LiPA genotyping.
2. Materials and Methods
This was a multicenter retrospective study and 504 serum or plasma samples were from Peking University Hepatology Institute (n=235), the 302 Hospital of People's Liberation Army (n=59) and National Institute for the Control of Pharmaceutical and Biological Products (n=210), and stored below -20 â„ƒ. Among them 294 were from CHB patients receiving lamivudine or adefovir dipivoxil treatment, and 210 were from blood donors from six provinces of China including a northwest province Qinhai screened with positive HBV DNA. Four hundred samples were tested by Cobas Taqman 48 (Roche, Branchburg, NJ) or by a local real-time PCR assay (PG Biotech, Shenzhen, China) for HBV DNA quantification. This study was approved by the Ethical Committees of the three centers.
2.2 HBV DNA isolation
HBV DNA was extracted from 200-Î¼l aliquot of serum or plasma with the QIAamp blood kit (Qiagen, Hilden, Germany) and DNA were eluted in 50Î¼l double-distilled water (ddH2O). An aliquot of 10Î¼l DNA solution was used in each PCR for INNO-LiPA or sequencing.
2.3 PCR and quality control
The quality of PCR products was essential for INNO-LiPA and sequencing. In order to prevent the potential carryover contamination, PCR was performed in specific PCR laboratories approved by the National Center for Clinical Laboratory (NCCL) with separated rooms and dedicated equipments for the extraction of DNA, PCR and PCR product electrophoresis and with a directional flow from the beginning of the procedure to the end. For each PCR run, two negative controls and a blank control were included. Negative controls included an HBsAg, anti-HBs and anti-HBc negative serum and a ddH2O control throughout the whole PCR procedure from DNA extraction, and a ddH2O blank control was included from the first round of PCR. The presence of amplified product was checked by adding 5Î¼l PCR product to 2% agarose gels. The amplicon was visualized with ethidium bromide under the ultraviolet light.
Extracted HBV DNA was amplified with biotinylated primers and PCR buffer provided in the INNO-LiPA genotyping kit (Innogenetics, Ghent, Belgium, Lot Number: 178037 and 180077) according to manufacturer's instruction. The LiPA procedure was completed using the AUTO-LiPA instrument (Innogenetics, Ghent, Belgium), on which the amplicon denaturation, hybridization, washing and color development of the INNO-LiPA genotyping strips were automatically handled.
2.5 Direct sequencing
Direct sequencing was performed in the three research centers with an in-house nested PCR using primers specific for HBV S region, 5'-AGGACCCCTGCTCGTGTTAC-3' and 5'-ACATACTTTCCAATCAATAG- 3' for the outer sense and antisense primer, and 5'-GATGTGTCTGCGGCGTTTT ATCAT-3' and 5'-ACATATCCCATGAAGTTAAG-3' for the inner sense and antisense primer, respectively. For the first round of PCR, an aliquot of 10 Î¼l DNA or ddH2O was added to the PCR reaction system with a final volume of 50Î¼l, which contained 200 mM of each dNTP (Invitrogen, Carlsbad, CA), 1.5 mM MgCl2, 20 mM Tris-HCl (pH 8.4), 50 mM KCl, 2 unit of Platinum Taq DNA Polymerase (Invitrogen, Carlsbad, CA). For the nested PCR, 1Î¼l first-round-PCR product was added to the reaction mixture with the same final volume and component concentrations. The reaction for the first round PCR and the nested PCR was started at 95Â°C for 5 min, followed by 40 cycles at 95Â°C for 30 sec, 55Â°C for 30 sec and 72Â°C for 45 sec, then 72Â°C for 10 min and cooled to 4Â°C. PCR products were purified and sequenced in three local biotechnology companies by dideoxynucleotide sequencing method. Then these sequences were aligned with reference sequences with know genotypes from the GenBank. Genetic distances were estimated by Kimura two-parameter analysis, and phylogenetic trees were constructed by the neighbor-joining method with 1000 bootstrap replicates with MEGA version 4 .
2.6 Clonal sequencing
Cloning was performed collectively in Peking University Hepatology Institute following the classical protocol . Briefly the HBV DNA was amplified with the same primers and same PCR conditions as used in direct sequencing. Then the PCR products were ligated into pGEM-T easy vector (Promega, Madison, WI, USA), then the ligated products were transformed into E. coli DH5Î± (Clontech, Basingstoke, UK) under ampicillin selection and IPTG/X-gal (Takara, Dalian, China) blue-white screening. White-colored colonies were cultured, restriction enzyme digestion analyzed and sequenced with the inner sense primer. Twelve or more clones were sequenced for each sample.
2.7 Preparation of B and C genotype mixtures to assess the sensitivity of INNO-LiPA in detecting minor genotype population
Two sera with known genotype of B and C, respectively, were used to evaluate the sensitivity of INNO-LiPA in detecting minor genotype population. Their genotypes were confirmed by clonal sequencing to ensure that no mixed genotype existed. Serum HBV DNA was quantified on the Cobas Taqman 48(Roche, Branchburg, NJ). Both samples were adjusted exactly to the concentration of 3.0Ã-105 IU/ml then mixed in different proportions. The proportion of genotype B (or C) in the mixture was 4%, 5%, 6%, 8% and 10%, respectively. The 10 mixed sera were genotyped by INNO-LiPA. Furthermore the mixed sera were serially diluted to the final DNA levels ranging from 3Ã-104 to 3Ã-101 IU/ml then genotyped by INNO-LiPA to assess the sensitivity of INNO-LiPA at lower HBV DNA levels.
3.1 HBV DNA levels of the samples
Four hundred samples had quantitative HBV DNA results ranging from 2.71 to 9.00 log10 IU/ml with a median of 4.65 log10 IU/ml. The rest of 104 samples were not quantified due to the lack of serum or plasma.
3.2 Interpretation of the INNO-LiPA results
On the INNO-LiPA strip, the uppermost red line is the marker line, allowing correct orientation of the strip. A line is considered positive when a clear purple or brown band appears (see Fig.1). Below the marker line, line 1(Conj. control) controls for the color developing and should always be positive and have the same intensity on each strip in the same run. Line 2(Amp. control) controls for the amplification of HBV DNA.
If single or multiple reactive lines shows for a single genotype, the result should be reported as the corresponding genotype (Fig 1, strip 3, 4, 5 and 6). For mixed genotype infection, if multiple genotype specific lines show for different genotypes, a mixed genotype should be reported (strip 2 with B/C mixed genotype). If multiple genotype specific lines showed for one genotype, together with a single positive line for a second genotype, this should be reported as a single genotype with the multiple positive lines (strip 7 with C genotype). And indeterminate results should be reported when single lines react for more than one genotype in conjunction with a positive amplification control line (strip 8), or when all lines are positive, or when all genotype specific lines are absent in conjunction with a positive amplification control line (strip 9). These samples should be repeatedly tested before report.
3.3 Comparison of results by INNO-LiPA genotyping and direct sequencing
Direct sequencing genotyped all the 504 samples. Two samples, however, failed to be genotyped by INNO-LiPA for the reason that the results were both "indeterminate" (Fig. 1, strip 8 and 9). One had a high viral load of 3.27Ã-108 IU/ml and genotyped by direct sequencing as genotype B. The other had no quantitative DNA result due to the lack of serum. It yielded abundant PCR product after the first round of PCR and was genotyped by direct sequencing as genotype D.
In the other 502 samples, 463 had the same genotyping results by INNO-LiPA and direct sequencing. The concordance of two methods was 92.23% (463/502). Twenty four had completely inconsistent results by the two methods, among which 15 were B genotype by INNO-LiPA but C by direct sequencing, seven C by INNO-LiPA but B by direct sequencing, and two D by INNO-LiPA but B or C by direct sequencing. And the other 15 had B/C mixed genotypes by INNO-LiPA but B or C single genotype by direct sequencing, as shown in Table 1.
3.4 Confirmation of the consistent INNO-LiPA and direct-sequencing results by means of clonal sequencing
Five samples with the same INNO-LiPA and direct-sequencing genotyping results (two B genotype and three C genotype) were randomly chosen to be clonally sequenced. The results by INNO-LiPA and direct-sequencing were all consistent with those by clonal sequencing (data not shown).
3.5 Confirmation of discrepant INNO-LiPA and direct-sequencing results by means of clonal sequencing
Twelve samples with discrepant genotyping results were subsequently clonally sequenced. The other samples with discrepant results failed to be retested by clonal sequencing due to the lack of sera or plasma. Among the samples, seven samples with single genotype were completely discrepant by INNO-LiPA and direct sequencing (Table 2, Sample ID: S240, A130, A145, CZH, ZJM, FGD and S9). The results by clonal sequencing were all identical to those by direct sequencing. The other five samples were genotype B and C co-infection determined by INNO-LiPA while singe B or C genotype by direct sequencing (Table 2, Sample ID: A30, 29, 191, WXL and A185). Clonal sequencing confirmed the B/C mixed genotypes in all the samples with the minor genotype proportions from 15% to 40%, as shown in Table 2.
3.6 Sensitivity of INNO-LiPA genotyping in detecting mixed genotype infection
In five B/C mixed genotype samples(Table 2, Sample ID: A30, 29, 191, WXL and A185), the proportion of the minor HBV subpopulation was 15%, 30%, 25%, 40% and 15%, respectively, which indicated that the INNO-LiPA could detect 15% to 40% of minor HBV genotype subpopulation.
The sensitivity of INNO-LiPA in detecting mixed genotype was further evaluated with mixed sera containing different proportions of B and C genotypes. When the HBV DNA level was 3.0Ã-105 IU/ml, INNO-LiPA could detect 6% of genotype B and 5% of genotype C, respectively, as shown in Figure 2.
In addition, the mixed sera were serially diluted to the final DNA levels ranging from 3Ã-104 to 3Ã-101 IU/ml then genotyped by INNO-LiPA. The results indicated that the sensitivity of INNO-LiPA in detecting minor genotype was related to the DNA level. When HBV DNA was above 3Ã-103 IU/ml, INNO-LiPA could still detect 6% of genotype B and 5% of genotype C, respectively. However, when the DNA was 3Ã-102 IU/ml, it could only detect 8% of genotype B or C. When the DNA was 60 IU/ml~100IU/ml it could only detect 10% of genotype B or C. When the DNA level was 30 IU/ml, the PCR product couldn't be amplified and no hybridization signal was visualized on the INNO-LiPA strip, so the limit of detection (LOD) of INNO-LiPA genotyping was approximately 60 IU/ml of serum HBV DNA. The results were shown in Table 3 and 4, respectively.
From our experience the quality of PCR product is of vital importance for the subsequent analysis. Even when low contamination occurs the results would be invalidated considering that INNO-LiPA and clonal sequencing are both capable of detecting minor mixed genotype infection. So in our multicenter research all the PCR procedures were required to be performed by the well trained technicians in the specific PCR laboratories approved by the NCCL. Once the negative or blank control band was positive this run of PCR would be cancelled and all possible contamination sources should be eliminated before a second PCR to guarantee the quality of PCR products.
Our study showed that INNO-LiPA genotyping was highly consistent with direct sequencing in detecting HBV genotype in Chinese HBV patients as well as asymptomatic carriers with a concordance of 92.23%, which was higher than the result of Osiowy (81%), and lower than those of Hussain (96%) and Pas (99%) [13-15]. Five out of 463 consistent samples were randomly chosen to be retested by clonal sequencing and the results of clonal sequencing confirmed the accuracy of both INNO-LiPA and direct sequencing.
Our study also showed that INNO-LiPA could detect B/C mixed genotype infection sensitively compared with direct sequencing. Fifteen samples were determined as B or C single genotype infection by direct sequencing while B/C genotype co-infection by INNO-LiPA, and five of them were all confirmed as B/C co-infection by clonal sequencing. Osiowy and Hussain et al reported that INNO-LiPA was sensitive in detecting A/B, A/C and A/G genotype infection compared with direct sequencing [13, 15]. Our results showed that INNO-LiPA was also sensitive in detecting B/C genotype co-infection, and 2.99% (15/502) of samples could be tested as B/C mixed genotype infection. Clonal sequencing showed that 15% of minor subpopulation could be detected by INNO-LiPA. Furthermore we used mixed and serial diluted samples to evaluate its sensitivity in detecting mixed genotype infection. The result showed that 6% of genotype B and 5% of minor genotype C could be detected by INNO-LiPA at HBV DNA levels above 3Ã-103 IU/ml. As we know, due to the limitations in sequencing technique and analytical software, direct sequencing can hardly distinguish the mixed genotype infection of HBV [13, 14, 19]. In contrast, INNO-LiPA genotyping is based on reverse hybridization and able to detect minor HBV subpopulation, which is one of its advantages. Mixed genotype infection has been reported frequent in Asia, Africa and Europe and its clinical significance is not yet very clear due to limited reports. Recently genotype mixtures have been shown to be associated with high viral loads, increased in vitro HBV replication, altered pathogenesis and deteriorated prognosis . In China B/C mixed genotype infection accounts for approximately 10% in CHB patients and it has been reported more difficult to achieve a sustained response than single genotype infection after an 18-month lamivudine monotherapy . Moreover, B/C superinfection is accompanied by acute exacerbations in CHB patients . The sensitiveness of INNO-LiPA genotyping in detecting B/C mixed genotype infection can effectively support the further study of the relationship between HBV multi-genotype infection and clinical outcome in China. However, it should be noted that the sensitivity of INNO-LiPA in detecting minor genotype was HBV DNA level dependent. INNO-LiPA could detect 6% of genotype B and 5% of genotype C, respectively at HBV DNA levels above 3Ã-103 IU/ml. But when the DNA was 3Ã-102 IU/ml, it could only detect 8% of genotype B or C. When the DNA was 60 IU/ml~100IU/ml it could only detect 10% of genotype B or C. When the DNA level was 30 IU/ml, the PCR product couldn't be visualized and no hybridization signal was visualized on the INNO-LiPA strip. So the LOD of INNO-LiPA was approximately 60 IU/ml of serum HBV DNA. With the decline of serum HBV DNA levels the minor genotype population was more difficult to be amplified and hybridized by the INNO-LiPA probes, which could account for this phenomenon.
At the same time, the limitation of INNO-LiPA should also be mentioned on the basis of our results. It can only detect the known viral nucleotide polymorphism or mutation that can be hybridized by the INNO-LiPA probes, which is, we think, one of the important reasons that indeterminate results as well as discordant results are present occasionally [0.4%(2/504) and 4.76%(24/504), respectively]. For the two samples with indeterminate INNO-LiPA results, one had a high viral load (3.27Ã-108 IU/ml) and the other had no quantitative DNA result but had abundant PCR product for INNO-LiPA hybridization. So the indeterminate results were not probably caused by the low levels of HBV DNA but the failure in hybridization with INNO-LiPA probes, as seen in other INNO-LiPA assays such as INNO-LiPA HBV Precore Research Version  and INNO-LiPA HBV drug resistance assay . We failed to find the polymorphism not yet covered by INNO-LiPA HBV genotyping because the probe sequences were not available. We suggest that the INNO-LiPA be repeated in the indeterminate samples, and if the second INNO-LiPA is still indeterminate the sample should be subject to direct sequencing or clonal sequencing for genotype.
In conclusion, INNO-LiPA genotyping is highly consistent with direct sequencing and clonal sequencing in detecting HBV genotype, and furthermore it's sensitive in detecting B/C mixed genotype infection in Chinese CHB patients and asymptomatic carriers. The accuracy and sensitivity of INNO-LiPA make it suitable in clinical laboratories to further study the relationship between HBV genotype and the progression of disease, disease outcome and the antiviral therapy response.