Blood samples (n=544) from two different populations (Pygmies and Bantus) in Cameroon, West Africa, were analysed. Serological tests indicated that the anti-hepatitis C virus (HCV) prevalence in Bantus (20?3 %) was higher than that in Pygmies (2?3 %, P<0?0001), whereas the distribution of hepatitis B virus (HBV) serological markers was equally high in both populations: in total, 9?4, 17?3 and 86?8 % for HBsAg, anti-HBs and anti-HBc, respectively. HBV genotype A (HBV/A) and HBV/E were predominant (43?5 % each) in both populations, and HBV/D was found in a minority (13 %). The preS/S region was sequenced in nine cases (five HBV/A and four HBV/E) and the complete genome in six cases (four HBV/A and two HBV/E). Subsequent phylogenetic analysis revealed that the HBV/A strains were distinct from the subtypes (subgenotypes) described previously, Ae (A2) and Aa (A1), and in the preS/S region they clustered with previously reported sequences from Cameroon. Based on the nucleotide difference from Aa (A1) and Ae (A2), more than 4 % in the complete genome, the Cameroonian strains were suggested to represent a new subtype (subgenotype), designated HBV/Ac (A3). A high (3?9 %) nucleotide divergence in HBV/Ac (A3) strains suggested that the subtype (subgenotype) has a long natural history in the population of Cameroon. One of the HBV/Ac (A3) strains was found to be a recombinant with an HBV/E-specific sequence in the polymerase reverse transcriptase domain. Further cohort studies will be required to assess detailed epidemiological, virological and clinical characteristics of HBV/Ac (A3), as well as its recombinant form.

Some wild African green monkeys are known to be naturally infected with a retrovirus related to human immunodeficiency virus (HIV) without having any apparent symptoms of an AIDS-like disease. This simian immunodeficiency virus, designated SIVAGM, may be helpful in clarifying the evolution and pathogenicity of HIV. Some virus strains that were previously reported to be isolated from African green monkeys were shown to be laboratory contaminations of SIVMAC (SIV from a rhesus macaque) Here we report the complete DNA sequence of authentic SIVAGM, which was isolated from a naturally infected African green monkey of Kenyan origin. Comparison of the genome of SIVAGM with those of known HIV/SIVs indicates that the virus is a new simian lentivirus that is approximately equally distantly related to HIV-1 and HIV-2 in contrast to SIVMAC, which is much closer to HIV-2 than to HIV-1 (refs 5, 9).

Isolates of human T-lymphotropic virus type I (HTLV-I) were phylogenetically analyzed from native inhabitants in India and South America (Colombia and Chile) and from Ainu (regarded as pure Japanese descendants from the preagricultural "Jomon" period). Their genomes were partially sequenced together with isolates from Gabon in central Africa and from Ghana in West Africa. The phylogenetic tree was constructed from the sequence data obtained and those of previously reported HTLV-I isolates and simian T-lymphotropic virus type I (STLV-I) isolates. The heterogeneity of HTLV-I was recently recognized, and one major type, generally called the "cosmopolitan" type, contained Japanese, Caribbean, and West African isolates. The phylogenetic tree constructed in the present study has shown that this cosmopolitan type can be further grouped into three lineages (subtypes A, B, and C). Subtype A consists of some Caribbean, two South American, and some Japanese isolates, including that from the Ainu, in addition to an Indian isolate, and subtype B consists of other Japanese isolates in addition to another Indian isolate, suggesting that there might be at least two ancestral lineages of the Japanese HTLV-I. Subtype A implies a close connection of the Caribbean and South American natives with the Japanese and thereby a possible migration of the lineage to the American continent via Beringia in the Paleolithic era. Subtype C consists of the West African and other Caribbean isolates, indicating that not all but part of the Caribbean strains directly originated from West Africa probably during the period of slave trade. The tree also has shown that the HTLV-I isolate from Gabon in central Africa forms a cluster with STLV-I from a chimpanzee, suggesting a possible interspecies transmission between man and the chimpanzee in the past. No specific clustering was observed in the tree in relation to manifestations of the disease such as adult T-cell leukemia and HTLV-I-related neurological disorders. Thus, the topology of the phylogenetic tree reflects the movement of people carrying the virus in the past.

Since the isolation of an HIV-2-related virus from captive macaques (SIVMAC), the origin of human immunodeficiency viruses, a much debated subject, has been attributed to monkeys. The sequence of SIVAGM, which is derived from a naturally infected African green monkey, shows equal relatedness to HIV-1 and HIV-2, suggesting that the derivation of these viruses from SIVAGM is unlikely. Recent sequence analysis of SIV from a captive sooty mangabey (SIVMAC), however, shows its close relatedness to HIV-2 and SIVMAC, indicating a possible origin of HIV-2 and SIVMAC from SIVSM (refs 4, 7, 9). We report here the sequence of a novel simian lentivirus, SIVMND, isolated from a wild-caught mandrill in Africa. It is distinct from the three other main groups, HIV-1, HIV-2/SIVMAC/SIVSM and SIVAGM, and therefore represents a fourth main group of primate lentiviruses. Phylogenetic analysis indicates that these four main virus groups might have diverged from a common ancestor at about the same time, long before the spread of AIDS in humans.