Hepatitis E virus infection and acute-on-chronic liver failure in West Africa: a case-control study from The Gambia.

BACKGROUND: In sub-Saharan Africa, it is unknown whether hepatitis E virus (HEV) infection is a common precipitating event of acute-on-chronic liver failure (ACLF). AIMS: To estimate the prevalence of HEV infection in general population and assess whether HEV is a common trigger of ACLF in cirrhotic patients in The Gambia, West Africa. METHODS: We first conducted an HEV sero-survey in healthy volunteers. We then tested cirrhotic patients with ACLF (cases) and compensated cirrhosis (controls) for anti-HEV IgG as a marker of exposure to HEV, and anti-HEV IgA and HEV RNA as a marker of recent infection. We also described the characteristics and survival of the ACLF cases and controls. RESULTS: In the healthy volunteers (n = 204), 13.7% (95% CI: 9.6-19.2) were positive for anti-HEV IgG, and none had positive HEV viraemia. After adjusting for age and sex, the following were associated with positive anti-HEV IgG: being a Christian, a farmer, drinking water from wells, handling pigs and eating pork. In 40 cases (median age: 45 years, 72.5% male) and 71 controls (39 years, 74.6% male), ≥70% were infected with hepatitis B virus. Although hepatitis B flare and sepsis were important precipitating events of ACLF, none had marker of acute HEV. ACLF cases had high (70.0%) 28-day mortality. CONCLUSIONS: Hepatitis E virus infection is endemic in The Gambia, where both faecal-oral route (contaminated water) and zoonotic transmission (pigs/pork meat) may be important. However, acute HEV was not a common cause of acute-on-chronic liver failure in The Gambia.

Single nucleotide polymorphisms of the mannose-binding lectin are associated with susceptibility to primary biliary cirrhosis.

Although the immunopathogenesis of primary biliary cirrhosis (PBC) remains unknown, familial clustering of patients with PBC suggests an important role for genetic factors. In addition, recent data support the thesis that the mucosal immune response against intraluminal pathogens may be involved with the onset of PBC. Mannose-binding lectin (MBL) is a key factor in innate mucosal defenses and has several key single nucleotide polymorphisms (SNPs). To study whether MBL gene SNPs are associated with susceptibility to PBC, we studied 65 patients with PBC and 218 controls by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) and sequence specific priming-polymerase chain reaction (SSP-PCR) to examine four polymorphic loci: two (H/L and X/Y) within the promoter region and the other two (P/Q and A/B) within exon-1. We also analyzed serum MBL concentrations. Interestingly, the prevalence of haplotype HYPA, leading to hyper-production of MBL, as well as HYPA/HYPA genotype were significantly increased in PBC compared to controls (0.53 vs. 0.44, P=0.031; 33.9%vs. 17.0%, P=0.003, respectively). Furthermore, individuals homozygous for HYPA had a significantly increased risk for PBC (odds ratio (OR)=2.51, 95% confidence interval (CI)=1.34-4.66). Our results demonstrate that the MBL genotype can be significantly associated with increased risk for PBC, and further, that increased production of MBL plays a critical role in immunopathogenesis.

Novel hepatitis B virus strain from a chimpanzee of Central Africa (Pan troglodytes troglodytes) with an unusual antigenicity of the core protein.

We and others have previously reported a hepatitis B virus (HBV)-like hepadnavirus strain which seemed to be indigenous to West African chimpanzees (Pan troglodytes verus). After that, we obtained an HBsAg-positive serum sample from a chimpanzee from Central Africa, named Bassi, belonging to another subspecies (P. troglodytes troglodytes). The full-genome nucleotide sequence of the hepadnavirus from Bassi showed a significant difference (9-26%) from those so far reported from primates including humans, chimpanzees and gorillas, suggesting a novel strain. More interestingly, however, the core antigen (HBcAg) deduced from Bassi's sequence showed only 78-82% similarity to known primate strains at the amino acid level, whereas the other strains shared more than 90% similarity. HBcAg expressed from Bassi HBV failed to react with monoclonal antibodies that were directed at an epitope borne by codons 135-145 of HBcAg of conventional hepadnaviruses. This could explain why Bassi was negative for anti-HBc in a routine test. Here we report the novel HBV strain presumably indigenous to P. troglodytes troglodytes in Central Africa.

Full-genome nucleotide sequence of a hepatitis E virus strain that may be indigenous to Japan.

We identified hepatitis E virus (HEV) RNA in serum from a Japanese patient with acute hepatitis, who had never been abroad. The full-genome nucleotide sequence of the HEV isolate (JRA1) from this patient was composed of 7227 nucleotides excepting the poly(A) tail and had ORF1 coding for 1703 amino acids (aa), ORF2 coding for 660 aa, and ORF3 coding for 122 aa. This Japanese strain showed approximately 87% nucleotide similarity to human and swine strains reported from the United States, while it had only 73-76% similarity to Asian and Mexican strains. Here we report the characteristics of the HEV-JRA1 isolate, which might be the first example of an indigenous strain(s) of HEV in Japan.

Hepatitis C virus (HCV) genotype 1b sequences from fifteen patients with hepatocellular carcinoma: the 'progression score' revisited.

The genome of hepatitis C virus (HCV) associated with hepatocellular carcinoma (HCC) may have some characteristics which would barely be found in those of HCV from asymptomatic carriers (ASC). We analyzed 15 HCC patients who were infected with HCV genotype 1b (HCV-1b) for complete nucleotide sequences of the viral genomes. Of the 15 isolates, three were sequenced up to the first nucleotide of the 5'UTR, and six were sequenced to encompass the X-tail at the 3' end: sequencing of at least three-quarters of the 5'UTR and entire polyprotein-ORF was accomplished in all 15 isolates. Analyses of these sequences together with those reported previously by Nagayama et al. [Hepatology; 31 (2000) 745] suggested that nine residues (nt 119 of 5'UTR and aa 90, 434, 938, 962, 1176, 1412, 2143, and 2774 of polyprotein) might be useful to discriminate HCC-type sequences from ASC-type ones. The 'progression score' was 1.4+/-0.9 in ASC versus 3.7+/-1.5 in HCC (P=3.87E-07) when calculated with the Nagayama et al.'s seven residues, but was 1.4+/-0.6 versus 4.6+/-1.9 (P=1.33E-09) with our nine residues: a greater difference between HCC and ASC was achieved in the latter system. Further analyses, by increasing the sample size and/or by extending the comparison to include entire 5'UTR and 3'UTR/X-tail, may thus contribute to define the 'progression score' more appropriately.

Three cases with TT virus infection and idiopathic neonatal hepatitis.

We present three cases of infants with idiopathic neonatal hepatitis showing diffuse intrahepatic fatty degeneration. Prolonged cholestasis has improved immediately upon intravenous administration of a high-dose gammaglobulin treatment in all three patients. The TT virus (TTV) genome was detectable in the serum of two patients, in the duodenal fluid of one and in the liver of all three. By analyzing sequence homology, we observed that the respective TTV isolated from serum, duodenal fluid and liver tissue were completely identical in cases 2 and 3. These findings suggest that TTV infection was one of the contributing factors for neonatal cholestasis in these patients. TTV was isolated from the serum of two out of the three mothers. The viruses were either completely or almost identical in sequence to those isolated from their respective infants, suggesting that they had been transmitted from mother to infant in these 2 cases. The patients presented here, whose livers were infected with the TTV and showed a favorable response to gammaglobulin therapy, may represent a subset of idiopathic neonatal hepatitis patients.

Genetic polymorphism of the MxA gene promoter and interferon responsiveness of hepatitis C patients: revisited by analyzing two SNP sites (-123 and -88) in vivo and in vitro.

We have previously reported a single nucleotide polymorphism (SNP) at nucleotide (nt) position -88 (G or T) within an interferon-stimulated response element-like sequence in the promoter region of the MxA gene, which correlated with responsiveness of hepatitis C patients to interferon. Upstream of it, we then identified another SNP (C or A at nt -123) and investigated whether this SNP also correlates with interferon responsiveness. The two SNPs showed a high linkage to each other: all the individuals having G at -88 had C at -123, and 73% of those having T at -88 had A at -123. As was expected from this observation, the SNP at -123 also exhibited a correlation with interferon responsiveness (C/C homozygotes were more frequent among nonresponders than among responders: 65% of 107 vs. 40% of 52, p = 0.0028). These in vivo data from patients were further supported by results from in vitro experiments. The MxA promoter sequence with A at -123 and T at -88 showed about 4-fold higher activity of upregulating the downstream reporter gene than that with C at -123 and G at -88, in a luciferase reporter assay.

Full or near full length nucleotide sequences of TT virus variants (Types SANBAN and YONBAN) and the TT virus-like mini virus.

TT virus (TTV) is a common virus and consists of many genotypes and variants. In addition, there exists a virus which both differs greatly from and retains a considerable resemblance to TTV, such as the TTV-like mini virus (TLMV) as we reported previously. Here we report the near full length genomic sequences of 4 isolates of a new variant of TTV (designated YONBAN) along with the full length sequences of 2 isolates of the TTV-SANBAN lineage and 7 isolates of the TLMV species derived from human sera. The TTV-YONBAN sequences showed only about 50% identity at the nucleotide level to those of the prototype TTV (TA278) and to SANBAN, and even less to TLMV. Moreover, the ORF1 of YONBAN lacked the ATG initiation codon which is shared by all the TTV and TLMV isolates so far identified in humans; instead, YONBAN had a Kozak's rule-compatible ACG codon as the candidate initiation site for the ORF1 translation. Nevertheless, the overall genetic structure and the conserved amino acid motifs within the ORF1 and the ORF2 were well shared among the prototype TTV strains, the SANBAN and YONBAN variants, and TLMV. The most conserved nucleotide sequence was found in the noncoding region just upstream from the ORF2, allowing construction of a phylogenetic tree which implied that the TTV genotypes and variants, the TLMV, and chicken anemia virus could be coclassified under a superfamily for which we proposed the name of 'Paracircoviridae' in our previous report.

Identification of a single nucleotide polymorphism in the MxA gene promoter (G/T at nt -88) correlated with the response of hepatitis C patients to interferon.

The interferon (IFN)-inducible MxA protein is known to play an important role in the host defense against certain viruses. We aimed to see if any genetic polymorphism in the promoter region of the MxA gene is associated with the IFN responsiveness of hepatitis C virus (HCV)-infected patients. Initially we sequenced the promoter region of the MxA gene in 12 subjects and found a polymorphic site. We then constructed a specific PCR-RFLP system for this site and subjected 63 samples from chronic hepatitis C patients who were nonresponders (NR) to IFN therapy to it, 52 with sustained response (SR), and 42 healthy controls. Subjects were all Japanese, and unrelated. A single nucleotide polymorphism (SNP) was identified in the MxA promoter region: G/T alleles at nt position -88. Interestingly, this SNP was involved in a genetic element highly homologous to the IFN-stimulated response element consensus sequence, and the G-to-T change there makes this homology a little greater. The rate of G.G homozygosity was 31% in the SR patients, significantly lower than in the NR patients (62%, p = 0.0009), while that of healthy controls was between the two groups (48%). Differences in HCV genotypes did not influence this result. Based on these findings, we propose that the SNP of the MxA promoter at nt -88 identified in this study affects the expression of MxA protein, and may thus be associated with the response of HCV patients to IFN.

Identification of a new human DNA virus (TTV-like mini virus, TLMV) intermediately related to TT virus and chicken anemia virus.

TT virus (TTV) is the only known human virus with single-stranded circular DNA, with a possible but yet unclear relationship to chicken anemia virus (CAV) of the family Circoviridae. Here we report a new human virus resembling TTV and CAV, designated TTV-like mini virus (TLMV). This non-enveloped virus was smaller (< 30 nm) but had a similar density (1.31-1.34 g/ml in CsCl) to TTV, when a TLMV/TTV-coinfected plasma was analyzed. Full-length sequencing revealed that the TLMV genome was a circular DNA comprising 2860 nt (isolate CBD231); significantly shorter than TTV (TA278, 3852 nt) but longer than CAV (CAECUX1, 2319 nt). A strand-specific hybridization assay using oligonucleotide-coated beads suggested TLMV was negative-stranded, like TTV and CAV. In genomic organization, TLMV was similar to both TTV and CAV. The untranslated region of TLMV resembled CAV in that both had direct repeats, whereas the sequence homology was more evident between TLMV and TTV. The predicted ORF1 protein of TLMV was rich in R/W/F residues at its amino terminus; the richness in W was shared by TTV, F by CAV, and R by both. ORF2 proteins of the three viruses had a common motif, WX7HX3CXCX5H. Thus, TLMV is an intermediate between the remotely related TTV and CAV. Since CAV differs much from other circoviruses, it may better be classified together with TTV and TLMV under a new family: we would coin the Paracircoviridae.

Full-genome nucleotide sequence of a hepatitis C virus variant (isolate name VAT96) representing a new subtype within the genotype 2 (arbitrarily 2k).

Hepatitis C virus (HCV), a single-stranded RNA virus of the family Flaviviridae, has a wide range of genetic heterogeneity: 6-11 genotypes (or 6 clades) have been known and each genotype comprises multiple subtypes. Here we report the entire nucleotide sequence of an HCV isolate from a patient in Moldova with chronic hepatitis (isolate name VAT96). The genetic organization of VAT96 was, from 5' to 3' ends, 5'UTR (341 nt), polyprotein ORF (9099 nt), 3'UTR (38 nt except for the poly-U and poly-pyrimidine stretch), and X-tail (98 nt). Comparison of the polyprotein amino acid sequence of VAT96 with those of known full-genome isolates assigned VAT96 to the genotype 2 (or clade 2), and further phylogenetic analysis based on a 447-nt sequence that covers part of the C and El regions suggested that VAT96 represents a new subtype within the genotype 2, arbitrarily designated "2k" VAT96 was unique in that it possessed a U residue prior to GCC at the 5' end of its genome while all the other full-genome HCV sequences start with GCC or ACC. In addition, the polyprotein ORF of HCV-VAT96, like HCV-BEBE1 of 2c, encoded several additional amino acids in excess, compared to 2a and 2b sequences. Despite these characteristics that may be unique to VAT96, the 98-nt sequence of the X-tail of VAT96 was highly homologous to those of other isolates with different genotypes so far reported.

CYP2D6 polymorphism and the presence of anti-LKM-1 in patients with chronic hepatitis C.

Anti-LKM-1 autoantibodies are directed mostly at cytochrome P450 2D6 (CYP2D6) autoantigen, whose activity ranges from "complete deficiency" to "extensive metabolism" due to genetic polymorphism. We aimed to find any relevance of CYP2D6 alleles to the presence/absence of anti-LKM-1 in Japanese patients with chronic hepatitis C. The frequency of an extensive metabolizer-type allele (CYP2D6*1) in anti-LKM-1-positive patients was higher than that in anti-LKM-1-negative patients (0.800 vs 0.431; P = 0.0035), while the CYP2D6*10 allele with moderately reduced activity was less frequent in the former than the latter (0.050 vs 0.389; P = 0.0069). Moreover, the rate of homozygosity for CYP2D6*1 showed a striking difference between the two groups (70% vs 19%; P = 0.0021). These findings suggest that a genetic predisposition to produce the enzyme CYP2D6 of extensive metabolizer-type is associated with the induction of anti-LKM-1 in chronic hepatitis C patients.

Properties of hepatitis B virus genome recovered from Vietnamese patients with fulminant hepatitis in comparison with those of acute hepatitis.

Among the many mutations found in the hepatitis B virus (HBV) genome, some have been associated with fulminant hepatitis, as exemplified by precore-defective mutations. The aim of this study was to determine whether such mutations also are found in Vietnamese cases of fulminant hepatitis B. The full-genome nucleotide sequence of HBV in three patients with fulminant hepatitis (F-2, F-3, and F-6) and one with acute hepatitis (A-3), who were admitted to Cho Ray Hospital, Ho Chi Minh City, Vietnam was ascertained. Additionally, two patients with fulminant hepatitis (F-1 and F-7) and three with acute hepatitis (A-1, A-2, and A-5) were examined only for the precore/core region of HBV. Remarkably, the nonsense mutation at precore codon 28 (Trp82Stop) was found in four of the five patients with fulminant hepatitis, while all the acute hepatitis patients harbored wild type (one had a mixture of wild and mutant types). The missense mutations within the core region, Ile97Leu and Pro130Ile/Thr/Ser, were also remarkable in fulminant hepatitis. Only F-2 was free from these precore/core mutations, but F-2 was unique in that it possessed a chimeric genotype: it could be classified into genotype C as a whole, but its X region was of genotype B, like the other four fulminant hepatitis isolates (F-1, F-3, F-6, and F-7). The codon 41 of the X protein was Pro in all three fulminant hepatitis cases examined for this region, while it was Ser in the wild-type isolates of genotype B. Of note as negative data, the mutations C1653T and T1753M of the enhancer II (Enh II) and A1762T and G1764A of the precore/core promoter regions, once reported to be relevant to severe or fulminant hepatitis, were not found in the present cases. The results with the Vietnamese cases of fulminant hepatitis corroborated results of previous studies with respect to the mutations Trp28Stop of precore and Ile97Leu and Pro130Ile/Thr/Ser of core, but not for the mutations within Enh II and precore/core promoter region. Whether the Ser41Pro mutation in the X region of genotype B HBV is Vietnam-specific or disease-specific deserves further investigation.

Indirect evidence of TTV replication in bone marrow cells, but not in hepatocytes, of a subacute hepatitis/aplastic anemia patient.

The presence of a new DNA virus (TTV) has been reported in sera from patients with posttransfusion hepatitis of unknown etiology. The precise replication site of TTV, however, has not been established. In this study, the presence of TTV in liver autopsy material, and in bone marrow biopsy and autopsy samples taken from a subacute hepatitis/aplastic anemia patient was determined by PCR and Southern blot analyses. Liver cells were found to contain only TTV DNA and not mRNA. Bone marrow material, especially that taken at biopsy, contained high levels of TTV DNA. It is suggested that the TTV replication site was in the bone marrow rather than in the liver, and that TTV infection was the cause of this patient's aplastic anemia. The precise etiological association of TTV with hepatitis remains to be established.

TT virus (TTV) genotypes in native and non-native prostitutes of Irian Jaya, Indonesia: implication for non-occupational transmission.

We investigated prostitutes in Irian Jaya, an Indonesian territory of New Guinea, to know whether TT virus (TTV) is sexually transmitted and what genotypes of TTV exist there. An ORF2 region of the TTV genome was analyzed for 44 isolates from prostitutes (19 were Irian Jaya natives but 25 were immigrants from Java or other islands of Indonesia) and 26 isolates from women of child-bearing age in Surabaya as a control. The WX(7)HX(3)CX(1)CX(5)H motif sequence of the ORF2 was compared across the 70 Indonesian isolates with a reference of 47 Japanese isolates (filed in databases) and the prototype TTV isolate TA278. A total of 77 different sequences were generated from the comparison, but a phylogenetic analysis suggested that they could be divided into three categories: group A, group B and others (the third group was highly diverse). Interestingly, most of the Indonesian isolates belonged to group B (74%): this rate was considerably higher than that observed previously in Japanese isolates. Group B isolates were further compared for the N-terminal 95 amino acids of the ORF2, with a result that the natives of Irian Jaya had a different pattern of genotype distribution from other groups. In particular, 9 out of 19 isolates from the Irian Jaya-natives were co-classified under a distinct branch, to which none of the other Indonesian and Japanese isolates belonged. Our data indicate that TTV genotypes reflect the birth place of the infected prostitutes rather than their work environment, and thus suggest that an infection is more likely during the early period of life than through sexual transmission. In addition, the presence of an Irian Jaya-specific genotype is intriguing from an anthropological and viral evolutionary point of view, because Irian Jaya has been isolated from contact with other areas for a long time.

Full-genome sequence analyses of hepatitis B virus (HBV) strains recovered from chimpanzees infected in the wild: implications for an origin of HBV.

Hepatitis B virus (HBV) belongs to the genus Orthohepadnavirus of the family Hepadnaviridae. Having been found in various animals (duck, heron, woodchuck, ground squirrel, and primates), hepadnaviruses must have undergone a long history of evolution and may comprise more members than currently recognized. Chimpanzees may also have their own hepadnavirus, even if it might be very close to HBV. We analyzed HBV-like sequences from three chimpanzees (Pan troglodytes) that were most likely infected during their life in Africa in the wild. Two chimpanzees (Ch256 and Ch258) possessed a viral genome of 3182 nt in length with a 33-nt deletion in the preS1 region, which could not be classified into any of the six genotypes (A-F) of human HBV but was very homologous to a previously reported isolate from a London Zoo chimpanzee. Phylogenetically distinct from the HBV-like sequences from gibbons, orangutans, and a gorilla so far reported, the Ch256 and Ch258 isolates would represent an indigenous chimpanzee HBV (tentatively ChHBV). A third chimpanzee (Ch195) had a 3212-nt genome, classifiable into the genotype E of HBV. Because HBV-E has been found mostly in Africans, Ch195 may have been infected from a human source in Africa. However, an inverse scenario is also possible: a spread of HBV-E might have occurred from chimpanzees to humans a long time ago in Africa. Analysis of the arginine-rich C-terminal region of the core protein, which is well conserved among mammalian hepadnaviruses, indicated that HBV-E/F and nonhuman primate hepadnaviruses are much closer than HBV-A/B/C/D to the hepadnaviruses of woodchuck and ground squirrel. Our results support an "ex-nonhuman primate" hypothesis for the origin of HBV.