Discovery of a new human T-cell lymphotropic virus (HTLV-3) in Central Africa

Abstract

Human T-cell Leukemia virus type 1 (HTLV-1) and type 2 (HTLV-2) are pathogenic retroviruses that infect humans and cause severe hematological and neurological diseases. Both viruses have simian counterparts (STLV-1 and STLV-2). STLV-3 belongs to a third group of lymphotropic viruses which infect numerous African monkeys species. Among 240 Cameroonian plasma tested for the presence of HTLV-1 and/or HTLV-2 antibodies, 48 scored positive by immunofluorescence. Among those, 27 had indeterminate western-blot pattern. PCR amplification of pol and tax regions, using HTLV-1, -2 and STLV-3 highly conserved primers, demonstrated the presence of a new human retrovirus in one DNA sample. tax (180 bp) and pol (318 bp) phylogenetic analyses demonstrated the strong relationships between the novel human strain (Pyl43) and STLV-3 isolates from Cameroon. The virus, that we tentatively named HTLV-3, originated from a 62 years old Bakola Pygmy living in a remote settlement in the rain forest of Southern Cameroon. The plasma was reactive on MT2 cells but was negative on C19 cells. The HTLV 2.4 western-blot exhibited a strong reactivity to p19 and a faint one to MTA-1. On the INNO-LIA strip, it reacted faintly with the generic p19 (I/II), but strongly to the generic gp46 (I/II) and to the specific HTLV-2 gp46. The molecular relationships between Pyl43 and STLV-3 are thus not paralleled by the serological results, as most of the STLV-3 infected monkeys have an "HTLV-2 like" WB pattern. In the context of the multiple interspecies transmissions which occurred in the past, and led to the present-day distribution of the PTLV-1, it is thus very tempting to speculate that this newly discovered human retrovirus HTLV-3 might be widespread, at least in the African continent.

Figure 1

HTLV-3 is closely related to STLV-3. Unrooted phylogenetic tree generated with the Neighbor-joining method, performed in the PAUP program (v4.0b10), on a 180 bp fragment of the tax gene using all full length PTLV-1/2 available sequences and all published STLV-3 tax sequences. The PTLV-1/2/3 strains, including (in bold), the novel sequence generated in this work (Pyl43), were aligned with the DAMBE program (version 4.2.13). The final alignment was submitted to the Modeltest program (version 3.6) to select, according to the Akaike Information Criterion (AIC), the best model to apply to phylogenetic analyses. The selected model was the TrN+G. Bootstrap support (1,000 replicates) is noted on the branches of the tree. The branch lengths are drawn to scale, with the bar indicating 0.1 nucleotide replacement per site.

Figure 2

Serological pattern of the person infected by the HTLV-3 Pyl43 strain. (A) Western Blot from Genelabs Diagnostics (HTLV BLOT 2.4 version) and (B) a line immunoassay (INNO-LIA HTLV confirmation Immunogenetics). The HTLV 2.4 western blot kit is based on strips incorporating HTLV-1/2 native viral antigens (originating from HTLV-1 infected cells) to which HTLV-1 (MTA-1) or HTLV-2 (K55) gp46s or HTLV-1 and HTLV-2 (GD21) gp21 recombinant proteins have been added [11]. The INNO LIA kit uses only recombinant antigens and synthetic peptides derived from both HTLV-1 and HTLV-2 proteins sequences. Whereas gag p19 I/II corresponds both to a recombinant protein and synthetic peptides being recognized by anti HTLV-1 and HTLV-2 immune sera, env gp46 I/II corresponds only to synthetic peptides recognized by anti HTLV-1 and HTLV-2 immune sera. env gp46 II corresponds to synthetic peptides specific of HTLV-2 [12]. (A, B) Lane 1: HTLV-1 positive control; lane 2: HTLV-2 positive control; lane 3: STLV-3 positive control (STLV-3₆₀₄ strain); lane 4: STLV-3 positive control (STLV-3_F3); lane 5: HTLV-1/2 negative control; lane 6: plasma from the person infected by HTLV-3 (Pyl43 strain).