Search

Monocytes are known as an alternative target for HIV/SIV infection, but the contribution of monocytes to viral spread in a host is unclear. In this study, CD14 monocytes were monitored in 6 macaques until six weeks postinfection (wpi) with SIVmac239 to evaluate their contribution to viral load. The monocyte count in blood significantly increased with peak viremia at 2 wpi and the expression level of CD14 on monocytes significantly decreased at 1-2 wpi, though the number of CD4(+) T cells was stable in these macaques. The number of CD14 monocytes and the expression level of CD14 on monocytes at 2 wpi were also significantly related to the extent of viremia in plasma. An increased number of monocytes at 2 wpi was associated with a lower postacute viral load, suggesting that monocytes have a role in suppressing the virus. The lower expression level of CD14 in monocytes at 2 wpi was associated with a higher viral load and greater degree of infection of monocytes. This correlation suggests that monocytes with a low level of CD14 may be more susceptible to SIV and may enhance viral replication. The analysis of monocytes in persistently infected macaques revealed that the expression level of CD14 was also significantly low during persistent infection compared with naïve macaques, though the monocyte count was within the normal range. Monocytes may suppress viruses, perhaps by their immune function, during acute infection. However, infection of monocytes may increase the viral load and spread viruses in a host.

We first sequenced a full genome of simian immunodeficiency virus isolated from African green monkey (SIVagm) but the clone sequenced was found not to be biologically active. We subsequently succeeded in reconstructing a full genome infectious molecular clone, named pSA212. The infectious pSA212 clone (known as the TYO-1 strain of SIVagm) has been distributed widely for research analysis of SIVagm but its genome has never been fully sequenced. Here, we report the whole genome sequence of the infectious pSA212.

Seven isolates of human T cell leukemia virus type 1 (HTLV-1) were taken in southern India and phylogenetically analyzed to gain new insights into the origin and dissemination of HTLV-1 in the subcontinent. The new Indian HTLV-1s were found to be members of subgroup A (Transcontinental subgroup) of the Cosmopolitan group. They formed three different clusters (South African/Caribbean, Middle Eastern, and East Asian clusters). These results demonstrate that Indian HTLV-1s are genetically heterogeneous and include the most divergent strain of subgroup A. On the basis of these results, we speculate that subgroup A HTLV- 1s may have been present for thousands of years in India.

To investigate the prevalence of subtypes A and C, and the existence of recombinants of both subtypes in the southeast of the Democratic Republic of Congo (DRC), blood samples were collected from 27 HIV-infected individuals in Likasi, located in an area bordering close to Zambia, and analyzed phylogenetically. Out of the 24 strains with a positive PCR profile for pol-IN and env-C2V3, 15 (62.5%) had a discordant subtype or CRF designation: one subtype A/G (pol/env), four A/U (unclassified), three G/A, one G/CRF01, three H/A, one J/C, one CRF02 (G)/A, and one U/A. Nine (37.5%) strains had a concordant subtype or CRF designation: five subtype A, two C, one D, and one CRF02/G. The remaining three samples negative for PCR with env-C2V3 primers used in this study were further analyzed with env-gp41 primers and revealed the presence of two profiles: two J/J (pol-IN/env-gp41) and one C/G. These data highlight the presence of a high proportion (16/27, 59.3%) of recombinant strains and a low prevalence (4.1 and 7.4%) of subtype C based on env-C2V3 and pol-IN analyses, respectively, in Likasi. In addition, this is the first report that CRF02_AG exists in DRC, though the epidemiological significance of the existence of CRF02_AG in DRC remains unknown.

To investigate the route of zoonotic transmission of HIV-1, we isolated three and seven HIV-1 strains from 449 Pygmy hunter gatherers and 169 neighboring Bantu, respectively, in southern Cameroon. Phylogenetic analysis based on pol-integrase and env-C2V3 sequences revealed that strains from Pygmies were 1CRF02_AG/CRF02_AG, 1 subtype G/CRF02 AG (pol/env), and 1 CRFll_cpx/CRF11_cpx, and that those from Bantu were 2 CRF02_AG/CRF02_AG, 1 CRF02_AG/CRF01_AE/A, 1 CRF02_AG/subtype A, 1 G/A, 1G/CRF02_AG, and 1 unclassified fH. CRF02_AG and CRF11_cpx have been identified in Cameroon. The results suggest that HIV-1 has been introduced into Pygmies through their neighboring Bantu rather than directly from nonhuman primates.

To investigate the changes of neutralizing antibodies and viruses during simian and human immunodeficiency virus (SHIV) infection, we examined the cross-neutralizing ability of sequential sera from three macaques infected with SHIV, NM-3rN, and analyzed the sensitivity of the reisolate to neutralizing antibodies. Neutralizing activities of macaques' sera against the parental HIV-1 showed a persistent increase. Neutralizing activities were highly strain specific, but the spectrum of the neutralizing activity expanded against various clades of primary HIV-1s at 3 years after infection in one of the three macaques. The reisolate from an NM3-rN-infected macaque at 56 wpi, designated as R4356, was neutralized by sera from this macaque at a much lower titer than NM-3rN, even by the sera collected 2 years after the reisolation. Sera from macaques that were newly infected with R4356 also did not neutralize R4356 despite neutralizing NM-3rN strongly. These results suggested that long-term persistent infection with SHIV induced neutralizing antibodies with a broad spectrum. However, a virus resistant to the neutralizing antibodies emerged in the persistently infected macaque.

To investigate the HIV-1 subtypes prevalent in the Republic of Congo, we isolated 28 HIV-1 strains from Congolese AIDS patients in 1996 and 1997, and analyzed them phylogenetically. Phylogenetic analysis based on part of the 5' tat-env (vpu) and env sequences revealed that only 13 (46.4%) of the 28 isolates belonged to the same subtype in the vpu tree as in the env tree; the remaining 15 (53.6%) strains showed discordant subtypes between vpu and env with 6 different profiles; that is, 1 A/A (vpu/env), 1 D/D, 5 G/G, 4 H/H, 2 unclassified (U)/U, 9 G/A, 2 G/H, 1 G/J, 1 H/G, 1 U/A, and 1 U/J. Thus, 9 of the 15 discordant HIV-1s were of the G/A (vpu/env) type, and did not form any subcluster within the subtype G lineage in the vpu-based phylogenetic tree. In addition, CRF02_AG (IbNG), which is a G/A (vpu/env) type, was not found in the Republic of Congo. These data suggest that the majority of HIV-1 subtypes circulating in the Republic of Congo have mosaic structures and may have been derived from independent recombinational events.

We investigated apoptosis induced by in vitro infection with the chimeric virus of simian immunodeficiency virus and human immunodeficiency virus (SHIV). Macaque and human peripheral blood mononuclear cells (PBMCs) were infected with pathogenic SHIV-89.6p (89.6p) or nonpathogenic SHIV-NM-3rN (NM-3rN). In macaque PBMCs, the extent of virus production and apoptosis induction in CD4(+) cells was much greater in 89.6p infection than in NM-3rN infection. The result was consistent with our previous study of in vivo SHIV infection. In human PBMCs, 89.6p replicated and induced apoptosis more extensively than did NM-3rN, when the cells were infected with the same infectious doses of the viruses. However, in cells infected with a high dose of NM-3rN, the levels of virus production and apoptosis induction were comparable to those in 89.6p infection. There was no significant difference in the extent of apoptosis induction between 89.6p and NM-3rN infection when growth curves of the two viruses matched. Thus, apoptosis induction by SHIV might depend quantitatively on the amount of virus production rather than on the strains of the virus. Moreover, the correlation between the extent of apoptosis induction and virus pathogenicity in macaque PBMCs has also been found in SHIV-infected macaques. This suggests that the profiles of SHIV infection in vitro reflect the in vivo phenomena. Therefore, the in vitro evaluation of apoptosis induction by SHIV could be useful as a safety test for the development of live-attenuated vaccines.

We found a novel primate lentivirus in mandrill (Mandrillus sphinx). To clarify the evolutionary relationships and transmission patterns of human/simian immunodeficiency virus (HIV/SIV), we screened blood samples from 30 wild-born healthy Cameroonian mandrills. Five (16.7%) of them were seropositive for SIV. Three SIV strains were isolated from the five seropositive mandrills by cocultivation of their peripheral blood mononuclear cells (PBMCs) with PBMCs of rhesus macaques, a human T cell line (M8166), and/or a cynomolgus macaque T cell line (HSC-F). One of the newly isolated SIV strains was intravenously inoculated into two rhesus macaques and resulted in chronic infection. In the SIV-infected macaques at 45 weeks after inoculation, we observed a mild decline in the number of peripheral CD4(+) lymphocytes, lymphadenopathy, and blastic follicular dendritic cells with mild follicular hyperplasia in the peripheral lymph nodes. A phylogenetic analysis based on the pol sequence showed that the newly found SIVs from Cameroonian mandrills did not cluster with SIVmndGB1, which is the former representative strain of SIVmnd. The SIVmnds from Cameroon formed a new, independent lineage that branched before the root of the HIV-1/SIVcpz lineage with 996 of 1000 bootstrap replications. They clustered host specifically, and exhibited about 16.9% diversity at the level of nucleotide sequence among Cameroonian SIVmnd strains. These results indicate that the SIVmnds isolated in Cameroon are a novel type of SIVmnd and have infected Cameroonian mandrills for a long time. We therefore designated the Cameroonian SIVmnd as SIVmnd type 2 and redesignated SIVmndGB1 as SIVmnd type 1. To date, M. sphinx is the only primate species other than humans that is naturally infected with two different types of SIV.

Human T cell leukemia virus type 1 (HTLV-1) is endemic among three ethnically distinguishable populations in Japan (the Ainu, Ryukyuans, and Wajin), which, together, account for most of the population in Japan. While much is known about the phylogeny of the Ryukyuan and Wajin strains of HTLV-1, only one Ainu strain has been phylogenetically analyzed. We report here a new HTLV-1 strain from the Ainu. The new isolate (U8306), as well as the previously reported isolate, are members of the Cosmopolitan group and further belong to the Transcontinental subgroup. This subgroup also predominates among the Ryukyuans, whereas the Japanese subgroup is the major subgroup among the Wajin. The predominance of subgroup A in the Ainu and Ryukyuans suggests that they share a common origin of HTLV-1.

To assess the molecular epidemiology of HIV-1 in Republic of Congo (Congo), we investigated 29 HIV-1s obtained from 82 Congolese AIDS and ARC patients in 1996 and 1997. Part of the env region including the V3 loop was phylogenetically analyzed. The genotypes observed were varied: of 29 specimens, 12 (41 %) were subtype A, 1 (3%) was subtype D, 6 (21%) were subtype G, 6 (21%) were subtype H, 2 (7%) were subtype J, and 2 (7%) could not be classified as any known subtypes (U, unclassified). The heterogeneous profile of HIV-1 infection was different from the profiles of neighboring Central African countries. These data show that subtypes G and H as well as subtype A were circulating with high prevalence. The fact that new genetic subtypes (J and U) are circulating indicates a need for a greater surveillance for these subtypes both in Congo as well as in other parts of the world.

The switch from a Th1- to a Th2-type cytokine response is reported to be involved in human immunodeficiency virus (HIV) disease progression. To study the effect of IL-6, one of the Th2-type cytokines, on AIDS pathogenesis, we constructed an SIV/HIV-1 chimeric virus (SHIV) having the human IL-6 gene (SHIV-IL6) SHIV-IL6 could replicate in M8166, a human T cell line, as well as in monkey and human peripheral blood mononuclear cells (PBMCs). Along with the SHIV-IL6 replication, IL-6 was detected in the culture supernatant by ELISA. The maximum level of IL-6 was 35, 15, and 8 ng/ml in M8166, human PBMCs, and monkey PBMCs, respectively. The expressed IL-6 was biologically active as shown by the proliferation of IL-6-dependent murine hybridoma (MH-60) cells. The inserted IL-6 gene was stable for at least four passages (45 days after the initial infection) in M8166 cells, suggesting the ability to achieve stable expression of IL-6 in long-term experiments. Therefore, we successfully established an SHIV system expressing IL-6, and this is the first report of an SHIV expressing a Th2-type cytokine. With this system, IL-6 should be expressed in the regions where the virus replicates, and therefore the inoculation of macaque monkeys with SHIV-IL6 is expected to provide further information on the etiology of AIDS.

Chimeric simian and human immunodeficiency viruses (SHIVs) are useful for evaluating vaccine candidates against HIV-1 and for investigating the pathogenesis of HIV-1 in vivo. In addition, SHIVs are candidates for a vaccine against HIV-1 because attenuated SHIVs can induce long-lasting anti-HIV-1 Env humoral and cell-mediated immunity in monkeys without AIDS-like diseases. In this study, we inserted IL-12 genes in a nef-deleted SHIV to increase the ability of the SHIV to induce cell-mediated immunity against HIV-1. The SHIV vector was constructed by deleting the nef gene and replacing it with restriction enzyme sites. Since IL-12 consists of two subunit genes, p35 and p40, SHIVs with one or both of these genes were constructed. SHIVs with either one of the subunit genes could replicate without a deletion of the inserted gene, but SHIVs with two subunit genes replicated poorly and the inserted genes were rapidly deleted. Production of IL-12 was detected when both of the single-subunit SHIVs were coinfected. The production of IL-12 by the coinfection reached 800 pg/ml, and IL-12 was detected after serial passage in cell cultures, although this amount of IL-12 heterodimer was 150-1500 times less than that of the p40 subunits. These IL-12-producing SHIVs are candidates for a live-attenuated vaccine to induce effective cellular immunity against HIV-1.

To investigate the role of apoptosis in the progressive loss of CD4+ lymphocytes in HIV infection, we have used macaques infected with SHIV, a hybrid virus of HIV and simian immunodeficiency virus (SIV). In the present study, we sequentially analyzed apoptosis induction in the acute phase of SHIV infection. Four macaques infected with a pathogenic SHIV, SHIV89.6P, and four macaques infected with a nonpathogenic SHIV, NM-3rN, were analyzed during the first 2 or 4 weeks postinfection. In the 89.6P-infected macaques the number of peripheral CD4+ cells sharply decreased at 2 weeks postinfection and was maintained below 50/microl until 4 weeks postinfection, while in the NM-3rN-infected macques the number of the CD4+ cells did not change significantly. Plasma viral loads peaked at 2 and 2-3 weeks postinfection, and the peak values were about 1 x 10(9) and 10(6)-10(7) copies/ml in the 89.6P- and the NM-3rN-infected macaques, respectively. In the 89.6P-infected macaques, Fas antigen expression and the extent of apoptosis in PBMCs and peripheral lymph nodes increased at 1-2 weeks postinfection. A high number of apoptotic cells was also observed in thymus sections 2 and 4 weeks postinfection. On the other hand, apoptosis was scarcely induced in the NM-3rN-infected macaques. These results suggest that the extent of apoptosis induction is closely correlated with the pathogenicity of SHIV and that the apoptosis induction in peripheral lymphoid tissues and thymus, where T cell maturation occurs, may play an important role in the depletion of CD4+ lymphocytes in 89.6P infection.

To confirm the geographic and ethnic segregation of HTLV-I and HTLV-II carriers in native populations in South America, we have conducted a seroepidemiological study of native populations in South America, including HTLV-I carriers distributed among seven ethnic groups in the Andes highlands of Colombia, Peru, Bolivia, Argentina, and Chile, and two ethnic groups on Chiloe Island and Easter Island; and HTLV-II carriers distributed among seven ethnic groups of the lowlands along the Atlantic coast of Colombia, Orinoco, Amazon, and Patagonia, and one ethnic group on Chiloe Island. The incidence rate of HTLV-I and HTLV-II carriers varied among the ethnic groups, ranging from 0.8 to 6.8% for HTLV-I seropositivity and from 1.4 to 57.9% for HTLV-II seropositivity. A new HTLV-I focus was found among the Peruvian Aymara (1.6%), the Bolivian Aymara (5.3%) and Quechua (4.5%), the Argentine Puna (2.3%), and the Chilean Atacama (4.1%), while on HTLV-II focus was found among the Brazilian Kayapo (57.9%), the Paraguayan Chaco (16.4%), and the Chilean Alacalf (34.8%) and Yahgan (9.1%). The distribution of HTLV-I/II foci showed a geographic clustering of HTLV-I foci in the Andes highlands and of HTLV-II foci in the lowlands of South America. It was thus suggested that South American natives might be divided into two major ethnic groups by HTLV-I and HTLV-II carrier state.

In 1995, 53 blood samples from Muslim patients with AIDS, or who were thought to have AIDS, were collected in the main hospitals of Adamaoua Province, in the northern part of Cameroon. The variable env C2V3 region of HIV-1 was amplified by nested PCR and phylogenetically analyzed. The results indicated that of 15 amplified samples, 1 belonged to HIV-1 group O, 1 to HIV-1 subtype D, 1 to subtype G, 2 to subtype H, and 10 to subtype A. Furthermore, the northern Cameroonian subtype A could be divided into at least two subclusters as shown by the env tree as well as by two remarkably conserved hexameric amino acid sequences in the apex of V3 (GPGQAF in one subcluster and GPGQTF in the other). This distinction suggests that the HIV-1 subtype A circulating in northern Cameroon evolved from two main sources. More recently, three HIV-1 strains from Nigeria (IBNG) and Djibouti (DJ263 and DJ264), previously reported on the basis of their env C2V3 sequences as subtype A, were found to have a similar A/G mosaic structure alongside their full-length sequence and were tentatively designated as members of a new subtype called "IBNG." Interestingly, within the northern Cameroonian subtype A described, the isolates of the second subcluster clustered distinctly with these A/G mosaic strains, strongly suggesting that they may be members of the IBNG subtype.

To investigate the role of apoptosis in the early phase of HIV infection, we used macaques infected with simian immunodeficiency virus strain mac (SIVmac) as a primate model and examined sequentially the characteristics of apoptosis of lymphocytes in peripheral blood mononuclear cells (PBMCs) and lymph nodes in the early phase of SIVmac infection. Five macaques infected with a pathogenic strain of SIV, SIVmac239, were analyzed during the first 4 weeks after infection. Peripheral CD4+ and CD8+ cells transiently decreased at 1 week postinfection. The percentage of apoptotic cells in cultured PBMCs increased from about 2 weeks postinfection. The number of apoptotic cells in lymph node sections was higher on days 13 and 28 postinfection than before infection and on day 5 postinfection. Fas antigen expression on peripheral lymphocytes was upregulated from day 8 postinfection. These results indicate that apoptosis is induced about 2 weeks after SIVmac239 infection, following the upregulation of Fas antigen expression on lymphocytes. Since apoptosis was induced about 1 week after the decrease in peripheral CD4+ and CD8+ cell counts, it appears that the apoptosis induction does not play an important role in the transient lymphopenia in the early phase of SIVmac infection. In macaques infected with a nonpathogenic derivative of SIVmac239, SIVmac delta nef, apoptosis of lymphocytes was induced as it was in SIVmac239-infected macaques, but to a lesser degree, suggesting a correlation between the extent of apoptosis induction in lymphocytes in the early phase of SIVmac infection and the pathogenicity of SIVmac.

To clarify the physiological function of two zinc finger motifs in the nucleocapsid (NC) domain of the Gag protein of human immunodeficiency virus type 1 (HIV-1), we changed cysteine to serine in either of the two motifs or both by site-directed mutagenesis. Viral infectivity was lost by any of the mutations, but their effects appeared differently in the respective mutants. Northern blot analysis showed that the first finger mutant was far less efficient (approximately 10% of the wild type) in genomic RNA encapsidation and that the dual mutant of both fingers completely failed to encapsidate the RNA. In contrast, the second finger mutant retained its ability for RNA encapsidation with an efficiency similar to that of the wild type. Immunoblot analysis of the lysates of CD4-positive M8166 cells transfected with the mutant proviral DNAs showed that the processing of Gag precursors was delayed in two mutant viruses having alterations in the first finger sequence, whereas the processing of the second finger mutant appeared to be normal. On the other hand, immunoblot analysis of the virus particles showed that the second finger mutant particles contained some proteins that were thought to be degradation products of p24CA. Electron microscopic observation showed that all particles of these mutant viruses were morphologically alike except that they had a slightly larger diameter than that of the wild type. These results indicate that these finger motifs of HIV-1 NC protein do not function equivalently. Namely, the first finger is primarily responsible for RNA encapsidation and the second is required for stabilization of virus particles.

An HTLV-II was isolated from a Ghanaian female patient with ARC and anti-HTLV-II antibody by cocultivation of PBMC with Molt-4 cells. A part of the 5' LTR of this virus was sequenced. Sequence homology between this virus and the corresponding sequence of Mo, the prototype of HTLV-II, is 97.4%. Furthermore, this virus is a member of the HTLV-IIa group, which was proposed by Hall et al. This is the first report of an HTLV-II sequence from Africa confirmed not only by serological evidence, but also by genomic analysis. This finding will provide useful information for considering the spread of HTLV-II in the world and the evolution of the HTLV group.

BTEB, a GC box-binding transcription factor, was tested for its ability to activate the human immunodeficiency virus type 1 long terminal repeat (HIV-1 LTR). An electrophoretic mobility shift assay demonstrated specific binding of BTEB to GC boxes of the HIV-1 LTR. When a BTEB expression vector was cotransfected into A3.01 cells with a fusion gene of HIV-1 LTR and chloramphenicol acetyltransferase (CAT) structural gene, the CAT activity was increased. This increase was accompanied by an increase in the content of CAT mRNA. Transcriptional activity of the HIV-1 LTR, stimulated by Tat, was further enhanced by the expression of BTEB. BTEB also activated the LTR activity in cooperation with phorbol 12-myristate 13-acetate. Northern blot analysis showed that various T cell and macrophage/monocyte cell lines expressed the BTEB mRNA to a level comparable with that of Sp1, another GC box-binding transcription factor. These results suggest that BTEB, like Sp1, is involved in transcriptional activation of the HIV-1 LTR.

We established seven hybridoma clones producing monoclonal antibodies (MAbs) against the envelope transmembrane protein (TMP) of a Ghanian isolate of human immunodeficiency virus type 2 (HIV-2[GH-1]) from mice immunized with the detergent-disrupted purified whole virus. The MAbs were found to react with 35 kilodalton (kD) TMP of the HIV-2[GH-1] virus in a Western blot assay (WB). Two of these MAbs recognized 135 kD proteins in addition to TMP in the lysate of HIV-2[GH-1]-infected cells. Two other MAbs cross-reacted with viral components corresponding to TMPs of HIV-2ROD and SIVMAC isolates in a Western blot. Results of competitive binding assay suggest that there are at least three epitopes on a TMP molecule of the HIV-2[GH-1] isolate. The MAbs did not inhibit syncytium formation between HIV-2[GH-1]-infected MOLT-4 cells and MOLT-4 clone 8 cells, nor virus infection to MOLT-4 clone 8 cells.

Thirty-two hybridoma clones producing monoclonal antibody (MAb) against HIV-2[GH-1] were established from mice immunized with NP-40-disrupted purified whole virus of a Ghanaian isolate of human immunodeficiency virus type 2 (HIV-2), strain HIV-2[GH-1]. Of these 32 MAbs, 20 reacted with p26 and the other MAbs recognized p15 of the HIV-2[GH-1] isolate. From the results of serological characterization by these MAbs, p26 and p15 were identified as capsid proteins and matrix protein, respectively, of HIV-2[GH-1] gag products. In addition to two gag proteins, a 55-kD protein corresponding to the primary translational product of gag gene and 39-kD protein corresponding to an intermediate product of cleavage of p55 were recognized by these MAbs in the lysate of HIV-2[GH-1]-infected cells. Moreover, these MAbs were used to analyze the number of antigenic epitopes on p26 and p15 of HIV-2[GH-1] isolate. The results of cross-reactivity with different HIV-1, HIV-2, and simian immunodeficiency virus (SIV) isolates and competitive binding assay suggest that there are at least four and five antigenic epitopes in p26 and p15, respectively, of the HIV-2[GH-1] isolate. The biological activity of MAbs was studied by performing syncytium inhibition assay and infection inhibition assays. However, our MAbs could not inhibit syncytium formation and infection by cell-free virus.

Genetic variability in human immunodeficiency virus type 1 (HIV-1) has been studied extensively, but the total nucleotide sequence of the HIV-2 genome has been reported only in two strains. For phylogenetic analyses of HIV, the genetic variability of HIV-2 should be investigated. This paper reports the complete nucleotide sequence of an HIV-2 isolate from Ghana, HIV-2[GH-1]. This virus showed approximately 85% homology in overall nucleotide sequence with HIV-2ROD. The amino acid sequence of the gag and pol proteins of HIV-2[GH-1] showed 90% homology with those of HIV-2ROD, but its env gene and central regions were highly variable (more than 20% divergence in amino acids), indicating the presence of extensive genetic heterogeneity in HIV-2. However, the sequences with specific functions were relatively well conserved in these HIV-2 isolates.