Importance of primer selection for the detection of hepatitis C virus RNA with the polymerase chain reaction assay

Importance of the polymerase chain reaction in the study of hepatitis C virus infection

Recently, the principal etiological agent of parenterally transmitted non-A, non-B hepatitis was molecularly cloned from the plasma of an experimentally infected chimpanzee and has been named hepatitis C virus. Determination of the complete nucleotide sequence of the hepatitis C virus genome was a crucial step in preparing the way for future study of this medically important human pathogen. Due to the very low concentration of virus in serum, amplification of viral RNA sequences by reverse transcription and polymerase chain reaction is the only practical method currently available for demonstrating viremia in patients with hepatitis C virus infection. This review examines the pivotal role of the polymerase chain reaction in understanding the biology of hepatitis C virus.

Virological features of hepatitis C virus infection in hemodialysis patients

The clinical and epidemiological relevance of circulating antibodies to hepatitis C virus (HCV) in hemodialysis patients is uncertain, since clinical signs of infection are often mild or absent, with alanine aminotransferase (ALT) values that are virtually always normal, and liver biopsies are only rarely performed. Determination of HCV RNA in serum is therefore critical for distinguishing chronic HCV infection from previous exposure to the virus. We studied HCV viremia by reverse transcription polymerase chain reaction (RT-PCR) in the 5'-noncoding region of the viral genome in 77 dialysis patients who were screened for anti-HCV by a second-generation enzyme-linked immunosorbent assay (the enzyme immunoassay II; Ortho HCV, 2nd generation, Ortho Diagnostic Systems Raritan, N.J.) and a second-generation recombinant immunoblot assay (Chiron Corporation and Ortho Diagnostic Systems) and prospectively evaluated for ALT elevations over a period of 5 years. Of 77 patients tested, 29 (38%) had active infection as shown by a positive PCR assay result, and of these, 26 were anti-HCV positive. Although a good correlation was found between circulating anti-HCV and HCV RNA in serum, 10 (28%) of 36 anti-HCV-positive patients were HCV RNA negative by PCR, suggesting either low levels of viremia or past exposure to HCV and subsequent recovery. On the other hand, 3 (7.3%) of 41 anti-HCV-negative patients had HCV RNA in their sera, indicating seronegative HCV infection. The ALT level had no predictive value for HCV infection, because it was repeatedly normal in 18 (62%) of 29 viremic patients. HCV genotyping was also performed and indicated that all four known genotypes of HCV were present in our group. In conclusion, serological assays are reliable for detecting exposure to HCV in hemodialysis patients; however, direct identification of the viral genome is required to document current infection.

Effect of concurrent acute infection with hepatitis C virus on acute hepatitis B virus infection

OBJECTIVE

To investigate the possible interference with acute hepatitis B virus infection by co-infection with hepatitis C virus.

DESIGN

Analysis of stored sera collected for transfusion transmitted viruses study in 1970s.

SETTING

Four major medical centres in the United States.

PATIENTS

12 recipients of blood infected with hepatitis B virus.

MAIN OUTCOME MEASURES

In 1970s, presence of antibodies in hepatitis B virus and raised serum alanine aminotransferase concentration; detection of antibodies to hepatitis C virus with new enzyme linked immunoassays.

RESULTS

Five of the 12 patients were coinfected with hepatitis C virus. Hepatitis B surface antigen was first detected at day 59 in patients infected with hepatitis B virus alone and at day 97 in those coinfected with hepatitis C virus (p = 0.01); median durations of antigenaemia were 83 and 21 days respectively (p = 0.05), and the antigen concentration was lower in the coinfected patients. Alanine aminotransferase patterns were uniphasic when hepatitis B virus infection occurred alone (range 479-2465 IU/l) and biphasic in patients with combined acute infection (no value > 380 IU/l; p = 0.0025). Four coinfected recipients developed chronic hepatitis C virus infection. The fifth patient was followed for only four months.

CONCLUSIONS

Acute coinfection with hepatitis C virus and hepatitis B virus inhibits hepatitis B virus infection in humans, and onset of hepatitis B may reduce the severity of hepatitis C virus infection but not frequency of chronicity. Alanine aminotransferase concentration showed a biphasic pattern in dual infection.

Genotype dependence of hepatitis C virus antibodies detectable by the first-generation enzyme-linked immunosorbent assay with C100-3 protein

Hepatitis C virus (HCV) samples in 155 sera, from patients with chronic non-A, non-B liver disease and blood donors, were grouped into four genotypes (I, II, III, and IV) by amplification of core-gene sequences by polymerase chain reaction with type-specific primers. HCV genotypes were compared with various HCV-associated antibodies detectable by the first-generation ELISA (ELISA-1) with C100-3 protein and a second-generation immunoblot assay with four recombinant HCV proteins. Antibodies to C100-3 protein and those to its subsequence (5-1-1) were detected in 13 (93%) and 12 (86%), respectively, of 14 sera with genotype I HCV; 56 (79%) and 58 (82%) of 71 sera with genotype II; 13 (34%) and 6 (16%) of 38 sera with genotype III; and 11 (34%) and 4 (13%) of 32 sera with genotype IV. Amino acid sequences of C100-3 of genotype I HCV are conserved by approximately 90% in genotype II, but only by approximately 75% in genotypes III and IV. The sensitivity of ELISA-1, therefore, would be influenced by heterogeneity in C100-3 sequences of different genotypes.

At least 12 genotypes of hepatitis C virus predicted by sequence analysis of the putative E1 gene of isolates collected worldwide

In a previous study we sequenced the 5' noncoding (NC) region of 44 isolates of hepatitis C virus (HCV) and identified heterogeneous domains that provided evidence for additional genetic groups of HCV not previously recognized. In this study we have determined the complete nucleotide sequence of the putative envelope 1 (E1) gene in 51 HCV isolates from around the world and found that they could be grouped into at least 12 distinct genotypes. The E1 gene sequence of 8 of these genotypes has not been reported previously. Although the genetic relatedness of HCV isolates determined by the previous analysis of the 5' NC region predicted the relationships observed in the E1 gene, analysis of the 5' NC sequence alone did not accurately predict all HCV genotypes. The nucleotide and amino acid sequence identities of the E1 gene among HCV isolates of the same genotype were in the range of 88.0-99.1% and 89.1-98.4%, respectively, whereas those of HCV isolates of different genotypes were in the range of 53.5-78.6% and 49.0-82.8%, respectively. The latter differences are similar to those found when comparing the envelope gene sequences of the various serotypes of the related flaviviruses as well as other RNA viruses. We found that some genotypes of HCV were widely distributed around the world, whereas others were identified only in discreet geographical regions. Four genotypes were identified exclusively in Africa and comprised the majority of HCV isolates on that continent. The E1 gene was exactly 576 nucleotides in length in all 51 HCV isolates with no in-frame stop codons. Analysis of the predicted E1 protein identified several conserved domains that may be important for maintaining its biological function: (i) eight invariant cysteine residues, (ii) three potential N-linked glycosylation sites, (iii) a domain of nine amino acids (GHRMAWDMM), and (iv) an amino acid doublet (GV) near the putative cleavage site at the C terminus of the protein. In conclusion, the discovery of at least 12 genotypes of HCV has important implications for HCV diagnosis and vaccine development.

Detection of hepatitis C virus genome in human serum by multi-targeted polymerase chain reaction

A multi-targeted "hemi-nested" PCR (M-PCR) assay in which the primer pairs derived from the 5' non-coding (5'NC) and the nonstructural protein 3 (NS3) regions of HCV genome were concurrently used for amplification in order to compare the sensitivity and specificity of polymerase chain reaction (PCR) with different primer pairs in detecting hepatitis C virus (HCV) genome. Sera from patients with virus-associated liver diseases were examined for the presence of HCV RNA by the M-PCR method following reverse transcription to cDNA. The amplified products derived from both the 5'NC and the NS3 regions were detected in 28 (70%) of the 40 HCV RNA-positive samples. However, 12 samples (30%) were devoid of the signal of NS3-derived product. Sensitivity tests using serial dilutions of HCV RNA revealed that the 5'NC-derived band was still detectable in the 10(5)-fold diluted sample by the M-PCR method, yet the NS3-derived band could hardly be detected in the 10(4)-fold diluted sample. Thus, as previously demonstrated by a single-targeted "nested" PCR assay, the present study using the M-PCR assay has clearly shown that the 5'NC-derived primers are more sensitive and specific than the NS3-derived primers in detecting HCV RNA.

Prevalence, infectivity, and risk factor analysis of hepatitis C virus infection in prostitutes

A population of 622 prostitutes in Taiwan was tested for antibodies to the hepatitis C virus (anti-HCV) using a synthetic peptide assay composed of representative determinants from capsid and nonstructural (NS) viral proteins. Seventy-four (12%) were positive and the remaining 548 (88%) were negative. Seventy-nine samples were retested by a second-generation anti-HCV assay composed of recombinant capsid and NS proteins of HCV. Both assays had a nearly perfect agreement (Kappa value = 0.91). Of the positive cases, 31% were positive for reactivity to capsid only. Most (60/74, 81%) of the cases positive for synthetic peptide assay were HCV RNA positive, indicating potential infectivity. On the basis of the results of synthetic peptide assay, univariate analysis showed that history of paid sex for longer than 6 months, blood transfusion, acupuncture, intravenous drug abuse, and age over 20 years were significant risk factors of HCV infection (P < 0.01). Elevated alanine aminotransferase levels (> 40 U/L) were positively associated with anti-HCV, while the presence of serum hepatitis B surface antigen was a negatively associated factor. Multivariate analysis revealed that history of paid sex for longer than 6 months and blood transfusion were positively associated with anti-HCV (P < 0.001). The latter only accounted for less than one fifth of the HCV-infected prostitutes. This study indicates strongly that sexual transmission is an important route for HCV infection in prostitutes. This risk group may spread HCV to other populations as a sexually transmitted disease.

Hepatitis C virus RNA in southern African blacks with hepatocellular carcinoma

We used a sensitive and specific cDNA polymerase chain reaction assay to detect hepatitis C virus (HCV) RNA in serum samples from 128 unselected southern African Blacks with hepatocellular carcinoma (HCC) and found HCV RNA in 26 (20.3%) of these patients. Antibodies to HCV (anti-HCV) were detected in 59 patients (46.1%) with a first-generation ELISA test, and only 19 of these patients were HCV RNA-positive. Anti-HCV was detected in 25 patients (19.5%) with a second-generation ELISA test, and 17 of these patients were HCV RNA-positive. Among the second-generation ELISA- and HCV RNA-positive patients, 14 were anti-HCV positive, 2 were indeterminate, and 1 was anti-HCV negative in a second-generation recombinant immunoblot assay, whereas all patients who were second-generation ELISA positive, but HCV RNA negative, were anti-HCV negative in this immunoblot assay. A total of 5 patients were negative in both ELISA tests but were HCV RNA positive. Seventy-one patients (55.5%) had evidence of a current HBV infection, 50 (39.1%) had evidence of a previous infection, and 7 (5.5%) had no evidence of a current or previous HBV infection. The prevalence of current HBV infection was significantly lower in patients who were positive for HCV RNA than in those who were negative (P = 0.001). This difference was not dependent on sex, age, or geographical location of the patients. The mean age of HCC patients positive for HCV RNA (52.3 years) was significantly higher (P < 0.001) than that of negative patients (40.3 years), and the difference was not dependent on HBV status or geographical location. Patients positive for HCV RNA were more likely to be urban than were negative patients. Thus, a significant number of southern African Blacks with HCC had a current HCV infection but not a current HBV infection, further suggesting that infection with HCV plays a role, albeit minor, in the development of HCC in this population.

Genotypic subtyping of hepatitis C virus

Four subtypes of hepatitis C virus (HCV), Pt(I), K1(II), K2a(III) and K2b(IV), have been suggested based on the nucleotide sequences of the non-structural (NS) 5 region. A fifth subtype from Japanese patients, Tr(V), which shows a less than 68% homology in nucleotide sequence when compared with other subtypes has been identified. A one-step method which enables a quick determination of subtype using polymerase chain reaction with a mixed primer set deduced from the sequence of each subtype has been developed. Using this technique, the subtypes of 418 out of 478 Japanese patients (87.4%) were determined. The incidence of each subtype in Japan was as follows: K1(II), 307 (73.4%); K2a(III), 74 (17.7%); K2b(IV), 28 (6.7%); and Tr(V), 3 (0.7%). This one-step subtyping technique should be useful for studying the epidemiology or biology of the HCV.

Detection of hepatitis C virus RNA: comparison of one-stage polymerase chain reaction (PCR) with nested-set PCR

We evaluated a new hepatitis C virus RNA assay based on one-stage PCR followed by liquid hybridization with an oligonucleotide probe and compared it with nested-set PCR. The one-stage and nested-set PCR assays had identical sensitivities in analytical experiments and showed 100% concordance when clinical specimens were used. One-stage PCR may be less prone to contamination than nested-set PCR.

Detection of hepatitis C virus infection with recombinant immunoblot assay, synthetic immunoblot assay, and polymerase chain reaction

The newly developed immunoblot assay, RIBA SIA (recombinant and synthetic polypeptide immunoblot assay), Chiron, Calif., was compared with the commercially available second generation recombinant immunoblot assay (RIBA-2) for the detection of antibody to hepatitis C virus (anti-HCV). The two immunoblot tests were also compared with the polymerase chain reaction (PCR) for the detection of HCV RNA. Ninety-one percent of samples reactive by RIBA-2 were positive for anti-HCV by RIBA SIA. A total of 31% of RIBA-2 indeterminate samples became reactive by RIBA SIA, 24% became non-reactive, and 45% remained the same. Samples reactive by RIBA-2 or SIA from different risk groups, were mostly positive (67-100%) by PCR for HCV RNA. All indeterminate samples from hemophiliacs and intravenous drug users were PCR positive. RIBA SIA is more sensitive and specific than RIBA-2 and correlates well with PCR results.

Polymerase chain reaction assay for hepatitis C virus RNA using a single tube for reverse transcription and serial rounds of amplification with nested primer pairs

A procedure is described for the detection of hepatitis C virus (HCV) RNA in blood by means of the polymerase chain reaction (PCR) in which the reverse transcription step and two rounds of amplification are carried out in a single tube. This results in fewer manipulations, reduced risk of contamination, and economy of time. The procedures are generally applicable to other assays based on the PCR. We describe the preparation (from 100 microL serum) of test samples that remain stable for at least 6 days under specified conditions and an assay that employs nested primer pairs homologous to conserved sequences in the 5' noncoding region. The method was tested on 107 sera from the United States and Japan. Correlation with first-generation anti-HCV was 77%. Two sets of nested primer pairs homologous to sequences in the 5' noncoding region and one set based on structural region sequences showed differences in their reactivities with serum HCV RNA. The recommended single tube procedure specified a primer for reverse transcription that was conserved in all reported HCV genomes but absent from pestivirus genomic sequences. The effects of preanalytical factors on the detection of HCV RNA were studied. Qualitatively, there was no change in the HCV RNA-positivity of sera that were exposed to room temperature for 24 hours. Quantitative studies showed a decrease in titer in some specimens. Three cycles of freeze-thawing had no detectable effects on the titers of HCV RNA.

Lack of protective immunity against reinfection with hepatitis C virus

Some individuals infected with hepatitis C virus (HCV) experience multiple episodes of acute hepatitis. It is unclear whether these episodes are due to reinfection with HCV or to reactivation of the original virus infection. Markers of viral replication and host immunity were studied in five chimpanzees sequentially inoculated over a period of 3 years with different HCV strains of proven infectivity. Each rechallenge of a convalescent chimpanzee with the same or a different HCV strain resulted in the reappearance of viremia, which was due to infection with the subsequent challenge virus. The evidence indicates that HCV infection does not elicit protective immunity against reinfection with homologous or heterologous strains, which raises concerns for the development of effective vaccines against HCV.

Sequence analysis of the 5' noncoding region of hepatitis C virus

We have determined the nucleotide sequence of the 5' noncoding (NC) region of the hepatitis C virus (HCV) genome in 44 isolates from around the world. We have identified several HCV isolates with significantly greater sequence heterogeneity than reported previously within the 5' NC region. The most distantly related isolates were only 90.1% identical. Nucleotide insertions were seen in three isolates. Analysis of the nucleotide sequence from 44 HCV isolates in this study combined with that of 37 isolates reported in the literature reveals that the 5' NC region of HCV consists of highly conserved domains interspersed with variable domains. The consensus sequence was identical to the prototype HCV sequence. Nucleotide variations were found in 45 (16%) of the 282 nucleotide positions analyzed and were primarily located in three domains of significant heterogeneity (positions -239 to -222, -167 to -118, and -100 to -72). Conversely, there were three highly conserved domains consisting of 18, 22, and 63 completely invariant nucleotides (positions -263 to -246, -199 to -178, and -65 to -3, respectively). Two nucleotide domains within the 5' NC region, conserved among all HCV isolates studied to date, shared statistically significant similarity with pestivirus 5' NC sequences, providing further evidence for a close evolutionary relationship between these two groups of viruses. Additional analysis revealed the presence of short open reading frames in all HCV isolates. Our sequence analysis of the 5' NC region of the HCV genome provides additional information about conserved elements within this region and suggests a possible functional role for the region in viral replication or gene expression. These data also have implications for selection of optimal primer sequences for the detection of HCV RNA by the PCR assay.