Proviral HTLV-I and HTLV-II in the Efe Pygmies of Northeastern Zaire

To the Editor: We showed that HTLV-II type-specific antibodies can be found among the Efe pygmies who belong to the Mbuti group of northeastern Zaire (1). This was confirmed on two samplings taken 20 years apart (2). Similarly, HTLV-II seropositivity could be found alongside HTLV-I seropositive in pygmies of Cameroon at the western end of pygmy distribution, but was conspicuously absent in pygmies of the Central African Republic (2-4). Recently, an HTLV-II isolate from a Cameroonian pygmy was characterized (5). The presence of HTLV-II in pygmies represents a challenge to the views on the evolution of HTLV-II, which was previously considered a virus that had spread from American Indians. In the present investigation, we confirm the presence of HTLV-II in Efe pygmies by demonstration of the proviral genome with the polymerase chain reaction (PCR). Twenty-three Efe pygmies presenting at the health center of Nduye (Ituri Forest, Zaire) in January and February 1995 for various ailments agreed to provide a blood sample. Four sports of capillary blood were taken on a filter paper (Whatman 2), and 1 ml of venous blood was mixed with an equal quantity of ethanol. After drying of the filter paper sample, the samples were kept in a refrigerator until shipment to Leuven. One spot of each sample was eluted in phosphate-buffered saline and screened for HTLV antibodies with a particle agglutination assay (Serodia, Fujirebio). Confirmatory assays included a Western blot, which also includes group and type-specific recombinant antigens (HTLV blot 2.3, Genelabs, Singapore), and ELISAs with type-specific synthetic antigens (Select HTLV, IAF/Biochem, Montreal). Three sera were positive on screening. Two of them were HTLV-II with both confirmatory assays. One gave an indeterminate pattern in Western blot (faint reactivity with p24 and rgp21, clear reactivity with p19, and absence of reactivity with the type-specific recombinants) and reacted with an HTLV-I-specific, p19 derived, synthetic peptide in ELISA. Among the 20 screening negative samples, nine had an indeterminate Western blot with no consistent pattern. DNA was extracted from one blood spot of each of the 23 samples and from 21 ethanol-fixed samples. The methods had been optimized using similarly treated blood samples spiked with different quantities of HTLV-infected cells. Briefly, each spot (≈1 cm in diameter) was cut in four and pretreated with methanol, followed by an overnight proteinase K digestion of each quarter of a spot. DNA was then extracted from the regrouped digestion buffer using the QIAmp Blood kit (Qiagen, Hilden, Germany). Similarly, the fixed whole blood was proteinase K digested after evaporation of the ethanol, and the DNA was extracted with the QIAmp Blood kit. An HTLV generic nested PCR amplifying part of the tax/rex region was performed on all samples using the TR 101-104 nested primer set (6) and the conditions described in reference (7). An amplicon of expected length (159 bp) was obtained in the ethanol-fixed sample of one of the HTLV-II seropositives and of the screening positive with an indeterminate pattern. In addition, the filter paper spot of the same HTLV-II seropositive was also PCR positive. Direct cycle sequencing of the amplicons in both directions showed that the HTLV-II seropositive sample indeed contains a genuine HTLV-II proviral sequence (pygmy2TR in Fig. 1), and the seroindeterminate sample contains HTLV-I (pygmy1TR in Fig. 1). These results were confirmed on material from a repeat PCR. The HTLV-II sequence found in a sample from a pygmy is truly HTLV-II. This region of the genome is too conserved to be able to establish relationships between this and the two established subtypes of HTLV-II (a and b). The pygmy HTLV-II differs by two nucletodies from HTLV-IIa (prototype Mo strain) and by one nucleoptide from HTLV-IIb (prototype NRA strain), while HTLV-IIa and -b differ one nucleotide from each another in this stretch. Preliminary sequencing results of the LTR region of the genome indicate that it differs from both subtypes, a and b. This should be further confirmed on this and other genomic regions to establish the true evolutionary relationships. In the highly conserved region analyzed here, even the very divergent Melanesian HTLV-I (Mel5 strain) differs only one nucleotide from the cosmopolitan type strain (ATK1). Still, the pygmy HTLV-I differs from the known HTLV-I subtypes (including the central African represented here by the Zairean ITIS strain) by one to three nucleotides. The presence of HTLV-I in one of the pygmies could indicate a separate reservoir of HTLV-I in pygmies. In a study of 102 Efe pygmies of Zaire sampled in 1970, one HTLV-I seropositive and four untypable HTLV positives were found. Similarly, in Cameroonian pygmies, 2 of 214 were HTLV-I positive (2). In the Central African Republic, Gessain and collaborators found five HTLV-I infected pygmies out of 410 (4). It is unlikely that HTLV-I would represent a recent importation into these groups from other human populations. As for HTLVL-II, the most logical flow is from pygmies to Bantus following the genetic flow, not the opposite (1,8). The lack of potentially infectious contacts with others is further shown by the absence in pygmies of HIV (1,9), a much more easily transmitted retrovirus with high prevalences in many parts of Africa. The two HTLV-I subtypes present in Africa (cosmopolitan and central African) probably arose from separate acquisitions from simians (10,11). The Efe pygmies are hunters with much contact with monkey blood. Both the observed nucleotide sequence and the indeterminate Western blot pattern in serology are compatible with a divergent HTLV-I strain in pygmies. It would not be surprising that HTLV-I in the pygmies represents a separate acquisition and independent evolutionary line of the virus. This is supported by data on pygmies from Cameroon and from the Central African Republic presented recently at the seventh international conference on human retrovirology in Paris (12,13). Acknowledgment: We thank Mr. Lou Michels and Dr. Tuanzebe Epasa for their valued help with this study in Zaire, and Mr. Martin Reynders and Ms. Christel Van Laethem for technical assistance in Belgium. This study was supported in part by the Belgian Fonds Voor Geneeskundig Wetenschappelijk Onderzoek (krediet 3.0098.94) and the Collen Research Foundation. Patrick Goubau; Anne-Mieke Vandamme; Kurt Beuselinck; Jan Desmyter Rega Institute and University Hospitals; Katholieke Universiteit Leuven; Leuven, BelgiumFIG. 1: . Nucleotide sequence of amplicons obtained by nested polymerase chain reaction (PCR) with tax/rex inner primers TR103 and TR104 (6) from blood samples of two pygmies (Pygmy1Tr, Pygmy2Tr) and compared to different HTLV/STLV strains. HTLV-la ATK1, HTLV-lb ITIS, HTLV-lc Mel5 are representatives of the cosmopolitan, the central African, and the Melanesian subtypes of HTLV-I, respectively. MTLV-IIa Mo and HTLV-IIb NRA are the type strains of the two known subtypes of HTLV-II. STLV PP1664 is a divergent isolate from a bonobo chimpanzee (Pan paniscus) (14). A similar STLV isolated from a bonobo by another group (15) differs three nucleotides from STLV PP1664 in this stretch (not included). STLV PH969 is the type strain of PTLV-L isolated from a hamadryas baboon (Papio hamiadryas) (16).