Simian-human immunodeficiency virus (SHIV) containing the nef/long terminal repeat region of the highly virulent SIVsmmPBj14 causes PBj-like activation of cultured resting peripheral blood mononuclear cells, but the chimera showed No increase in virulence

SIVsmmPBj14 is a highly pathogenic lentivirus which causes acute diarrhea, rash, massive lymphocyte proliferation predominantly in the gastrointestinal tract, and death within 7 to 14 days. In cell culture, the virus has mitogenic effects on resting macaque T lymphocytes. In contrast, SIVmac239 causes AIDS in rhesus macaques, generally within 2 years after inoculation. In a previous study, replacement of amino acid residues 17 and 18 of the Nef protein of SIVmac239 with the corresponding amino acid residues of the Nef protein of SIVsmmPBj14 yielded a PBj-like virus that caused extensive activation of resting T lymphocytes in cultures and acute PBj-like disease when inoculated into pig-tailed macaques. This study suggested that nef played a major role in both processes. In this study, we replaced the nef/long terminal repeat (LTR) region of a nonpathogenic simian-human immunodeficiency virus (SHIV), SHIVPPc, with the corresponding region from SIVsmmPBj14 and examined the biological properties of the resultant virus. Like SIVsmmPBj14, SHIVPPcPBjnef caused massive stimulation of resting peripheral blood mononuclear cells (PBMC), which then produced virus in the absence of extraneous interleukin 2. However, when inoculated into macaques, the virus failed to replicate productively or cause disease. Thus, while these results confirmed that the nef/LTR region of SIVsmmPBj14 played a major role in the activation of resting PBMC, duplication of the cellular activation process in macaques may require a further interaction between nef and the envelope glycoprotein of simian immunodeficiency virus because SHIV, containing the envelope of human immunodeficiency virus type 1, failed to cause activation in vivo.

Nucleotide substitutions in the long terminal repeat are not required for development of neurovirulence by simian immunodeficiency virus strain mac

The question of whether consensus nucleotide substitutions in the long terminal repeat (LTR) region of simian immunodeficiency virus strain mac (SIVmac) are important for neurovirulence was investigated in this report. Brains and lymph nodes from two macaques that developed AIDS and encephalitis following inoculation with two strains of neurovirulent SIVmac, and from one animal with AIDS but no neurological disease after inoculation with non-neurovirulent SIVmac239 were used. The 5' LTR regions from neurovirulent SIVmacR71/17E and SIVmac7F-Lu were amplified, cloned and sequenced and these sequences were compared to the LTRs amplified from three regions of the respective encephalitic brains and lymph nodes from macaques inoculated with each virus. The SIVmac7F-Lu and SIVmacR71/17E viruses had zero and three consensus substitutions, respectively, in the U3, R and U5 regions of the LTR compared to that of SIVmac239. The only consensus substitution in the LTR-gag region of the genome was a T to C change at position 829 within the tRNA binding site. The sequences amplified from the brain and lymph nodes of the two animals with AIDS and encephalitis were identical. This single common substitution in this region of the virus genome, the T to C substitution at position 829, was also found in the LTRs isolated from the brain and lymphoid organs from the macaque inoculated with SIVmac239. The virtual identity in nucleotide sequences in the LTR of the neurovirulent and non-neurovirulent viruses and in CNS and lymph tissues of animals inoculated with the viruses suggests that the LTR has no effect on the tissue tropisms of the viruses.

Neurovirulent simian immunodeficiency virus induces calbindin-D-28K in astrocytes

Astrocyte activation has been postulated to be a major contributor to functional changes in the brain of AIDS patients. We assessed astrocyte activation in the simian immunodeficiency virus (SIV) model. Four groups of macaque brains were examined: uninoculated controls, animals inoculated with virus that did not cause disease, animals inoculated with virus that caused AIDS but did not cause encephalitis, and animals with SIV encephalitis. We examined expression of calbindin-D-28K, a calcium binding protein that is upregulated in astrocytes during excitotoxic events, as well as glial fibrillary acidic protein (GFAP). The presence of calbindin in astrocytes was confirmed by double-labeling using confocal microscopy. Increases in calbindin staining were most apparent in the white matter, but increases in GFAP staining were most apparent in middle layers of the cerebral cortex. Six of the seven animals with SIV encephalitis had calbindin immunoreactive astrocytes in the subcortical white matter, corpus callosum, internal capsule, cerebral peduncle, pontine white matter, and cerebellar white matter. Very rarely, a few, very lightly calbindin-immunoreactive astrocytes were present in the uninoculated control brains. The increase in calbindin expression by astrocytes in SIV encephalitis suggests that these cells are subject to calcium toxicity. In uninoculated control macaques, and in macaques inoculated with virus that did not cause disease, GFAP-immunoreactive astrocytes were present throughout the subcortical white matter and in layer I, but very few were found in layers III-V of the cerebral cortex. Two animals that died of AIDS without encephalitis had somewhat higher numbers of GFAP immunoreactive astrocytes in middle cortical layers. In seven animals that received passaged neurovirulent virus and developed both AIDS and encephalitis, the number of GFAP-immunoreactive astrocytes in middle cortical layers was high, indicating widespread astrocyte activation.

Chimeric SHIV that causes CD4+ T cell loss and AIDS in rhesus macaques

By animal to animal passage in rhesus and pig-tailed macaques, we developed a rhesus model of HIV-1 disease in humans. Rhesus macaques infected with a cell-free stock of SHIVKU-2 developed CD4+ T cell loss, primary lentiviral encephalitis and pneumonia, and AIDS. Six of nine rhesus macaques died within eight months post-inoculation, while the remaining three are at five, five, and eight months post-inoculation, respectively. Animals infected by either mucosal or parenteral routes of infection had a similar course of infection.

Fractionator analysis shows loss of neurons in the lateral geniculate nucleus of macaques infected with neurovirulent simian immunodeficiency virus

Infection of macaques with neurovirulent strains of simian immunodeficiency virus (SIVmac) is an experimental model for the neurological manifestations of AIDS. Loss of neurons has been reported in the cerebral cortex following immunodeficiency viral infection, but thalamic structures which may contribute to electrophysiological changes and neurological deficits have not been examined. In this study, the lateral geniculate nucleus (LGN) of macaques inoculated with macrophage-tropic, neurovirulent virus SIVmac239 (R71 and 17E) was examined for neuron loss using the optical fractionator method. Estimates of the number of neurons in the P layers of the lateral geniculate nucleus of age-matched control macaques ranged from 1.0 to 1.3 x 10(6), while the number of neurons in SIV infected macaques ranged from 0.8 to 1.1 x 10(6), reflecting neuron loss of up to 28%. Neuron loss was not observed in the magnocellular layer. The total number of glia and glial density were unchanged. Loss of neurons in the lateral geniculate nucleus was correlated with the pattern of neuropathological changes. Neuron loss was most severe in animals with encephalitis concentrated in the brain stem and subcortical white matter and was less apparent in animals with diffuse encephalitis. Neuron loss in the lateral geniculate nucleus did not explain changes observed in the visual evoked potential, which was severely affected in two animals which showed a loss of 24 and 26%, while it was normal in a third animal which showed neuron loss of 28%.

Neuroinvasion by ovine lentivirus in infected sheep mediated by inflammatory cells associated with experimental allergic encephalomyelitis

Maedi Visna Virus (MVV) is a prototypic lentivirus that causes infection only in cells of macrophage lineage, unlike the primate lentiviruses which infect both CD4+ T lymphocytes and macrophages. In primates, the earliest viral invasion is associated with the ability of the virus to infect and activate T cells which convey virus to the brain. Infected monocytes in blood rarely cause CNS infection in absence of activation of CD4+ T cells. In the face of lack of infection or activation of T cells by MVV in sheep, the question arises, how does MVV gain access to the brain to cause the classical lesions of visna? In previous studies on experimental induction of visna, sheep were inoculated with virus directly in the brain. In this study, we asked whether neuroinvasion by MVV would occur if sheep were inoculated with virus in a non-neural site. Nine sheep were inoculated intratracheally and all developed systemic infection when examined 3 weeks later. At this time, five were injected intramuscularly with brain white matter homogenized in Freund's complete adjuvant to induce EAE. None of the four animals inoculated with virus alone developed CNS infection despite typical lentiviral infection in lungs, lymphoid tissues and blood-borne mononuclear cells. In contrast, all five of the sheep injected with brain homogenate developed infection in the brain. Virus was produced by macrophages associated with the EAE lesions. This study illustrated that both activated T cells specific for antigen in the CNS and infected macrophages are essential for lentivirus neuropathogenesis.

Pathogenesis of ovine lentiviral encephalitis: derivation of a neurovirulent strain by in vivo passage

The lentiviruses of sheep replicate almost exclusively in macrophages and cause chronic interstitial pneumonia, arthritis, and mastitis, but only rarely encephalitis. This study was undertaken to determine whether a non-neurovirulent field strain of ovine lentivirus isolated from joint fluid that replicated productively in lung and joint macrophages could be adapted to enter and replicate in the brain and cause encephalitis. The field isolate was passed seven times sequentially by intracerebral inoculation of sheep. The neuroadapted strain of virus caused severe encephalitis typical of visna in four of four sheep inoculated intracerebrally. The virus replicated to high titers in the brains of these animals and in cultured microglia. The inflammatory response in the brain was characterized by intense infiltrates of macrophages and CD8+ and CD4+ T cells. Many of the perivascular macrophages demonstrated TNF-alpha expression and there was upregulation of MHC Class II antigen expression on both inflammatory cells and endothelium. Inoculation of this neuroadapted virus into the bone marrow of three animals resulted in persistent infection and cell-associated viremia, but not encephalitis. Virus was not detected in brains from these animals, indicating that the virus was not neuroinvasive. These data suggest that neuroinvasiveness and neurovirulence are separate pathogenic determinants, both of which are required for the development of encephalitis during natural infection.

Infected macaques that controlled replication of SIVmac or nonpathogenic SHIV developed sterilizing resistance against pathogenic SHIV(KU-1)

Twenty macaques were used to evaluate the ability of nonpathogenic SIV(mac) or nonpathogenic chimeric SIV-HIV (SHIV) to induce protection in macaques against superinfection with a pathogenic variant of SHIV (SHIV(KU-1)) originally containing the tat, rev, vpu, and env of HIV-1 (strain HXB2) in a genetic background of SIV(mac)239. Specifically, three macaques inoculated with molecularly cloned, macrophage-tropic SIV(mac)LG1 developed an early systemic infection but recovered with only traces of SIV(mac) DNA in visceral lymphoid tissues. These animals were then inoculated parenterally with pathogenic SHIV(KU-1). All three animals resisted infection with SHIV(KU-1), as indicated by lack of virus recovery and absence of SHIV-specific env and vpu sequences in the visceral lymphoid tissues and multiple regions in the CNS. We also examined the ability of five macaques that had been inoculated with nonpathogenic SHIV (NP-SHIV) to withstand challenge with the pathogenic SHIV(KU-1). Like the SIV(mac)LG1-inoculated macaques, these animals also resisted SHIV(KU-1) challenge as judged by the inability to recover infectious virus, normal CD4+ T cell counts, and the absence of SHIV(KU-1) signature sequences in the lymph node tissue. Thus, eight of eight animals that developed control over primary lentivirus infections had also developed resistance to infection with pathogenic SHIV(KU-1). Three groups of macaques were used as controls for this study. The first group consisted of six macaques inoculated with SHIV(KU-1) alone. All animals developed viremia, showed severe loss of CD4+ T cells within 4 weeks, and succumbed to AIDS within 6 months. The second group of three macaques was inoculated first with SHIV(KU-1) and inoculated later with uncloned, neurovirulent SIV(mac)7F-Lu. A third group of three macaques was inoculated with SIV(mac)7F-Lu followed by inoculation with SHIV(KU-1). PCR analyses using oligonucleotide primers specific for the SIV or HIV env revealed that macaques from the last two groups had widespread infection with both SHIV(KU-1) and SIV(mac), indicating that animals that failed to control productive replication of either SHIV(KU-1) or SIV(mac)7F-Lu could not resist superinfection with the other virus. These data indicate that sterilizing immunity against the virulent SHIV could be induced in animals that had experienced an immunizing infection. Moreover, the divergence of the envelope glycoprotein of the protective avirulent and virulent challenge virus suggests that a single vaccine could protect against infection with a virus containing a different envelope glycoprotein.

Significance of macrophage tropism of SIV in the macaque model of HIV disease

Microglia, alveolar macrophages, and Langerhans cells are representatives of cells of macrophage lineage that are susceptible to infection with HIV-1 and they play important roles in the pathogenesis of AIDS dementia, lymphoid interstitial pneumonia, and systemic viral invasion from mucosal surfaces, respectively. In contrast, elimination of CD4+ T cells with resultant development of immunosuppression and AIDS is thought to be reflective of the exclusive tropism of the virus for CD4+ T cells. Examination of these concepts in macaques infected with molecularly cloned strains of SIVmac suggested that all strains of the virus are both macrophage- and lymphocyte-tropic and that all aspects of pathogenesis including loss of CD4+ T cells are dependent on infection in both cell types. However, viral clones that caused productive lytic infection in macrophages were less virulent than those which caused persistent nonproductive infection. The former caused subclinical and even immunizing infections, whereas the latter caused activation and productive infection in CD4+ T cells, AIDS, and systemic infection, even after inoculation of the virus on mucosal surfaces. If these findings on SIVmac are relevant to HIV-1 disease, then demonstration that HIV-1 isolates are macrophage-tropic probably does not necessarily correlate with their pathogenic potential.

Texture analysis of cerebral white matter in SIV-infected macaque monkeys

Image texture analysis is used in a wide variety of applications in medical research. Neurovirulent simian immunodeficiency virus (SIV) infection in monkeys is considered a good model for HIV-1 infection in humans and causes neuropathological changes in white matter which can include diffuse myelin pallor, subtle white matter astrocytosis, perivascular macrophage infiltrates, and microglial nodules with multinucleated giant cells. The ability of image texture analysis to quantify these changes was evaluated. Sections of thionin-stained brain tissue from eight male rhesus macaques ranging in age from 42-59 months were used. Four animals served as controls and four animals were infected with neurovirulent SIVmac239/17E-R71 by bone marrow inoculation. Images of cerebral white matter were captured and analyzed by calculating 13 textural features based on statistical analysis of spatial co-occurrence matrices. Statistical analysis of the results included multiple comparisons using the Newman-Keuls multiple range test. The effect of variation in background illumination used at image acquisition was also evaluated. Ten of the 13 textural features used in this study successfully discriminated between tissue from control and SIV-infected animals and were consistent with independent neuropathological assessment. Three textural features were highly sensitive to variation in background illumination and found not useful in this application.

Characterization of the pathogenic KU-SHIV model of acquired immunodeficiency syndrome in macaques

By animal-to-animal passage in macaques we derived a pathogenic chimeric simian-human immunodeficiency virus (SHIV) that caused CD4+ T cell loss and AIDS in pigtail macaques and used it to inoculate 20 rhesus and pigtail macaques by the intravaginal and intravenous routes. On the basis of the outcome of infection and patterns of CD4+ T cell loss and viral load, disease was classified into four patterns: acute, subacute, chronic, and nonprogressive infection. During the study period, 15 of the 20 animals developed fatal disease, including AIDS, encephalitis, pneumonia, and severe anemia. Opportunistic pathogens identified in these animals included Pneumocystis, cytomegalovirus, Cryptosporidium, Toxoplasma, and Candida. No single parameter by itself predicted outcome, although a combination of low CD4+ T cell counts in blood, high plasma virus levels, and presence of autoantibodies to red blood cells reliably predicted a fatal outcome. Five animals (25%) died within 3 months of inoculation and constituted the group with acute disease, whereas the nine animals (45%) with subacute disease died between 3 and 8 months postinoculation. This 70% mortality within 8 months is significantly shorter than in HIV-1-infected human beings, of whom 70% develop fatal disease a decade after infection. SHIV infection in macaques provides a useful model with which to evaluate antiviral strategies, combining all the advantages of the SIVmac system, yet using a virus bearing the envelope gene of HIV-1.

A cell-free stock of simian-human immunodeficiency virus that causes AIDS in pig-tailed macaques has a limited number of amino acid substitutions in both SIVmac and HIV-1 regions of the genome and has offered cytotropism

We have examined both the sequence changes in the LTR, gag, vif, vpr, vpx, tat, rev, vpu, env, and nef genes and the cell tropism of a cell-free stock of chimeric simian-human immunodeficiency virus (SHIV) isolated from the cerebrospinal fluid of a pig-tailed macaque (PNb) that developed AIDS. This virus (SHIVKU-1) is highly pathogenic when inoculated into other macaques. DNA sequence analysis of PCR-amplified products revealed a total of 5 nucleotide changes in the LTR while vif had 2 consensus amino acid changes. The gag, vif, and vpx had no consensus amino acid substitutions, whereas vpr had 1 consensus substitution. The tat and rev genes of the HXB2 region of SHIVKU-1 had 2 and 1 consensus amino acid changes, respectively. The vpu gene of the HXB2 region of SHIV, which originally had an ACG at the beginning of the gene, reverted to an initiation ATG codon and in addition contained a consensus amino acid substitution at position 69 of this protein. As expected, the majority of the nucleotide substitutions were found in the env and nef genes. Thirteen and 5 amino acid changes were predicted for the corresponding Env and Nef proteins, respectively. In addition, one-third of the env gene clones isolated from the SHIVKU-1 stock had a 5-amino-acid deletion in the V4 region. Using three independent assays, we determined that the changes in the SHIVKU-1 were associated with an increase in the efficiency of replication in macrophages. The strikingly few consensus changes in the virus suggest that conversion of this virus to one capable of causing AIDS in pig-tailed macaques was associated with relatively few changes in the viral envelope and/or accessory genes. These results will provide the basis for the development of a pathogenic, molecular clone of SHIV capable of causing AIDS in pig-tailed macaques.

Animal model of mucosally transmitted human immunodeficiency virus type 1 disease: intravaginal and oral deposition of simian/human immunodeficiency virus in macaques results in systemic infection, elimination of CD4+ T cells, and AIDS

Chimeric simian/human immunodeficiency virus (SHIV) consists of the env, vpu, tat, and rev genes of human immunodeficiency virus type 1 (HIV-1) on a background of simian immunodeficiency virus (SIV). We derived a SHIV that caused CD4+ cell loss and AIDS in pig-tailed macaques (S. V. Joag, Z. Li, L. Foresman, E. B. Stephens, L. J. Zhao, I. Adany, D. M. Pinson, H. M. McClure, and O. Narayan, J. Virol. 70:3189-3197, 1996) and used a cell-free stock of this virus (SHIV(KU-1)) to inoculate macaques by the intravaginal route. Macaques developed high virus burdens and severe loss of CD4+ cells within 1 month, even when inoculated with only a single animal infectious dose of the virus by the intravaginal route. The infection was characterized by a burst of virus replication that peaked during the first week following intravenous inoculation and a week later in the intravaginally inoculated animals. Intravaginally inoculated animals died within 6 months, with CD4+ counts of <30/microl in peripheral blood, anemia, weight loss, and opportunistic infections (malaria, toxoplasmosis, cryptosporidiosis, and Pneumocystis carinii pneumonia). To evaluate the kinetics of virus spread, we inoculated macaques intravaginally and euthanized them after 2, 4, 7, and 15 days postinoculation. In situ hybridization and immunocytochemistry revealed cells expressing viral RNA and protein in the vagina, uterus, and pelvic and mesenteric lymph nodes in the macaque euthanized on day 2. By day 4, virus-infected cells had disseminated to the spleen and thymus, and by day 15, global elimination of CD4+ T cells was in full progress. Kinetics of viral replication and CD4+ loss were similar in an animal inoculated with pathogenic SHIV orally. This provides a sexual-transmission model of human AIDS that can be used to study the pathogenesis of mucosal infection and to evaluate the efficacy of vaccines and drugs directed against HIV-1.

Early treatment with 9-(2-phosphonylmethoxyethyl)adenine reduces virus burdens for a prolonged period in SIV-infected rhesus macaques

We evaluated the effects of a reverse transcriptase inhibitor, 9-(2-phosphonylmethoxyethyl)adenine (PMEA), on simian immunodeficiency virus (SIV) infection in rhesus macaques (Macaca mulatta). Four macaques were given PMEA (20 mg/kg) subcutaneously on days 1 and 2 and inoculated with virus on day 2. Drug treatment was continued for 30 consecutive days, after which the virus burdens and course of infection were monitored for a further 6 months. Four control animals that did not receive PMEA all developed high virus burdens and two of the four developed clinical disease. In contrast, virus burdens remained low in three of the four macaques treated with PMEA and all four remained healthy. Our results show that suppression of virus replication early in infection can result in reduced virus burdens for a much longer period.

Plasmas from lymphocyte- and macrophage-tropic SIVmac-infected macaques have antibodies with a broader spectrum of virus neutralization activity in macrophage versus lymphocyte cultures

We examined plasma from macaques infected with three different phenotypes of SIVmac for their ability to neutralize the infectivity of homologous and heterologous virus in lymphocyte (CEMx174 cells or normal rhesus macaque peripheral blood lymphocytes) or normal rhesus macaque macrophage (Mphi) cultures. Similar to previous findings, we observed that some plasmas failed to neutralize or poorly neutralized the infectivity of SIVmac239 and SIVmac251(<1:20 plasma dilution) in lymphocyte cultures. In contrast, when primary rhesus Mphi cultures were used as the indicator cells, the same plasmas neutralized both viruses at high dilutions (1:200 to 1:20,000). Neutralization of virus infectivity by the various plasmas was confirmed by SIV core antigen capture assays. We excluded the possibility that this differential neutralization in Mphi was related to the differences in the ability of the virus strain to replicate in these two cell types by demonstrating that the replication efficiency of SIVmac251 in CEMx174 cells, PBMC, and Mphi cultures was very similar. The role of Fc receptors on the Mphi surface in the clearance of the virus-antibody complexes was also excluded since similar neutralizing results were obtained using whole plasmas, purified IgG antibodies, and purified Fab fragments derived from the IgG fraction of these plasmas. The mechanism of virus neutralization in Mphi does not appear to involve blocking of virus entry into the cells since radiolabeled virus reacted with anti-SIV antibodies was taken up by rhesus Mphi as efficiently as virus reacted with normal antibody. DNA of the neutralized virus was identified in the Mphi cultures, but virus replication, as evidenced by accumulation of viral protein products, was not detectable so long as the antibodies were present in the medium. Removal of the antibodies resulted in a resumption of virus replication in the Mphi. These results indicate that virus infectivity can be efficiently neutralized by antibodies in Mphi cultures by a mechanism that is fundamentally different from that in lymphocyte cultures.

Antibody to human MHC class I inhibits SIVsmmPBj1.9-induced proliferation of pigtailed macaque lymphocytes

Previously we have shown that the simian immunodeficiency virus SIVsmmPBj1.9, a molecular clone of SIVsmmPBj14, induces proliferation of human peripheral blood mononuclear cells (PBMC). We have extended this observation to show that SIVsmmPBj1.9 induces proliferation of PBMC from pigtailed macaques. This proliferative response was markedly inhibited by mAbs against human class I MHC, class II MHC and CD4 antigens, and partially inhibited by mAbs against integrin beta 2 subunit (CD18) and LFA-1 (CD11a). However, these antibodies differed in their ability to inhibit in vitro viral infectivity of PBMC. While anti-CD4, MHC class II, and LFA-1 strongly inhibited viral infectivity, antibodies to MHC class I demonstrated little effect on viral infectivity. A control antibody (PLM2) against porcine CD18 inhibited neither virus-induced proliferation nor viral infectivity. Based on these results, we suggest that SIVsmmPBj1.9-induced proliferation requires the participation of class I MHC, class II MHC and CD4 molecules. In addition, the observation that anti-class I MHC Ab inhibited proliferation of macaque PBMC induced by mitogen (PHA) and bacterial superantigens, such as Staphylococcus enterotoxin A and toxin shock syndrome toxin-1, suggests that SIVsmmPBj1.9 also contains a viral superantigen similar to that previously demonstrated in SIVsmmPBj14.

Genetic characterization of two phenotypically distinct North American ovine lentiviruses and their possible origin from caprine arthritis-encephalitis virus

Ovine and caprine lentiviruses are closely related genetically and antigenically although the diseases that these viruses cause in their respective host animals can vary greatly. In sheep, syndromes consist primarily of interstitial pneumonia with rare occurrences of arthritis and encephalitis, whereas in goats, the disease expresses mainly as arthritis in adult animals with rare cases of encephalitis in newborns. Experimentally, viruses from either sheep or goats can infect animals of the reciprocal species and many field strains of ovine lentivirus have biological properties similar to those of caprine viruses. However, a molecular correlation for the phenotypic differences between ovine and caprine lentivirus strains is unknown. To investigate this, we examined genetic characteristics of two phenotypically distinct North American ovine lentiviruses. Nucleotide sequence analysis of the envelope regions from virus strains 85/34 and 84/28 showed that despite significant biological differences, these viruses are closely related to each other and are genotypically more homologous to caprine arthritis-encephalitis virus (CAEV) than to visna virus of sheep. Furthermore, analysis of the nucleotide substitutions in their env regions indicated that when differences between the two ovine viruses and CAEV were found, the changes often resulted in nucleotides homologous with visna virus. These results suggest that the two field strains of ovine lentivirus may have originated from a cross-species infection of sheep by a CAEV-like virus and, evolution of their genomes toward that of ovine lentivirus may be reflective of adaptation of these viruses to the new ovine host.

Characterization of a leucine-zipper-like domain in Vpr protein of human immunodeficiency virus type 1

Human immunodeficiency virus type 1 (HIV-1) replicates productively in vitro in CD4(+)-T cells and/or macrophages. In the host, however, HIV-1 replication may be restricted by the quiescence of susceptible cells. Vpr is a 15-kDa late viral gene product, which is assembled in the virion and suspected to enhance HIV-1 replication in the infected host. We demonstrated previously that Vpr interacted specifically with the cellular transcription factor Sp1, and activated transcription from the HIV-1 long-terminal-repeat. Both Vpr-Sp1 interaction and trans-activation by Vpr required a central Leu/Ile-rich domain (LR domain, aa 60-81) in Vpr. This domain of Vpr was also found critical for Vpr interaction with another cellular protein of 180 kDa. We now provide biochemical evidence that the Vpr LR-domain has a leucine-zipper-like structure. The leucine-zipper structure has been found in a variety of cellular transcription factors, which use the leucine-zipper domain to form a specific dimer before they can bind to DNA through an upstream basic domain. The LR domain of HIV-1 Vpr, when fused to the basic domain of the cellular transcription factor CREB, was capable of supporting specific DNA binding by the CREB basic domain. Point mutational analysis of the Leu/Ile residues in the LR domain suggested that multiple Leu/Ile residues may be involved in maintaining the leucine-zipper-like structure. Mutagenesis in the context of the full-length Vpr also helped identify Leu/Ile residues may be involved in maintaining the leucine-zipper-like structure. Mutagenesis in the context of the full-length Vpr also helped identify Leu/Ile residues critical for Vpr interaction with the cellular 180-kDa protein. These results suggested that the leucine-zipper-like domain may be an important functional determinant for HIV-1 Vpr.

Variations in lentiviral gene expression in monocyte-derived macrophages from naturally infected sheep

Seventy-nine 1-year-old lambs from three individual farms and a feedlot were examined for natural lentivirus infection. We used three different methods to detect infection and to identify the stage of the ovine lentivirus life cycle in blood-derived macrophages. Cytopathic infectious virus was obtained from 14/14 Border Leicester animals obtained from a naturally infected flock. Neither virus particles, virus proteins, virus specific antibodies nor viral DNA were detected in samples from 34 lambs from two South Kansas City farms. However, among 31 feedlot lambs, we identified 11 infected animals. Specific viral proteins were immunoprecipitated from macrophages of one animal, but no infectious cytopathic virus was isolated from these cells. Cells from ten of the other feedlot animals harboured viral DNA but neither viral particles nor proteins could be detected by our techniques. Thus, in these naturally infected animals, the virus life cycle either proceeded to completion, subject to differentiation of infected precursor cells in blood, or remained arrested at the DNA stage despite maturation of monocytes to macrophages. Sequence analysis of the env gene of viral genomes from two of the ten feedlot sheep showed sequences distinct from those of known ovine and caprine lentiviruses. Surprisingly, these sequences have a higher identity (of nucleotide and derived amino acid sequences) to caprine arthritis-encephalitis virus than to the ovine prototype, maedi-visna virus. These data suggest that the ovine and caprine lentiviruses found in North American sheep may have a common ancestral genotype that is closely related to the caprine virus.

Neutralization of SIVmac239/17E in lymphocyte cultures involves virus strain-specific linear and conformational epitopes encoded by different regions of the env gene including the "V3" domain

SIVmac251 and its closely related derivatives SIVmac239 and SIVmac239/17E vary greatly in their susceptibility to neutralization with homologous and heterologous antisera. Whereas SIVmac251 induces homologous neutralizing antibodies, the antibodies induced by SIVmac239 rarely neutralize infectivity of this virus in lymphocyte cultures. In contrast, SIVmac239/17E is remarkably susceptible to neutralization with homologous and heterologous antisera induced by other strains of SIVmac. In this study, we studied the molecular basis for the neutralization of SIVmac239/17E. Using chimeric viruses in which different regions of the env gene of both SIVmac239 and SIVmac239/17E were inserted into a background of either of the parental genomes, we showed that the newly acquired neutralization properties of SIVmac239/17E were attributable to amino acid substitutions between the V2 and V4 regions of gp 120. Site-directed mutagenesis of the env gene in this region showed that the arginine substitutions at positions 334 and/or 340 within the "V3" domain were fundamental to virus neutralization but other substitutions in the V2-V4 region added to the ease of its neutralization since it became neutralizable with much higher dilutions of serum. The molecular determinants for neutralization of this virus are distinct from those reported as responsible for neutralization of SIVmac251 and both are distinct from SIVmac239.

Restrictive type of replication of ovine/caprine lentiviruses in ovine fibroblast cell cultures

Caprine arthritis-encephalitis virus (CAEV) is a natural lentivirus pathogen of goats. CAEV, like all members of the ovine/ caprine lentivirus family, has an in vivo tropism for cells of the monocyte/macrophage cell lineage and activation of viral gene expression is observed only following differentiation of monocytes to macrophages. In addition to cells of the monocyte/ macrophage lineage, CAEV and the closely related maedi visna virus of sheep (MVV) can also replicate productively in fibro-epithelial cells derived from synovial membrane of goats (GSM). However, these viruses varied greatly in their ability to replicate in fibroblasts. We studied the biological and biochemical properties of CAEV and maedi-visna virus (MVV) of sheep following inoculation into the three ovine/caprine cell types. Our data showed no substantial differences in virus titers, viral protein biosynthesis, or processing of the viral proteins between CAEV and MVV following inoculation into primary macrophages and GSM cells. However, unlike MVV, CAEV failed to replicate productively in ovine fibroblasts (sheep choroid plexus cells). This correlated with a specific but abnormal proteolytic cleavage of the envelope glycoprotein of the virus. This abnormal proteolytic cleavage represents a novel type of host cell restriction of lentivirus replication.

Antigenic variation of SIV: mutations in V4 alter the neutralization profile

Antigenic variation is a characteristic feature of lentiviral infection. The SIV/macaque model of AIDS provides an ideal system in which to investigate the molecular basis of antigenic variation. The purpose of this study was to genetically map the nucleotide changes in env that alter the neutralization phenotype of SIV. Serum taken from an SIVmac239-infected macaque (2D) at 30 weeks postinoculation was found to neutralize the input virus (SIVmac239) and an isolate, P9, obtained at 10 weeks p.i., but did not neutralize two other isolates, P13 and P23, obtained at 20 and 52 weeks, respectively. Sequence analysis of these virus variants revealed clustered amino acid changes in V1 and single base pair changes in V2-V4 of P13 and P23. Infectious recombinant viruses in which the V1 and V1-V3 sequences of SIVmac239 were replaced with those of P13 or P23 retained the neutralization profile of SIVmac239; both were neutralized by macaque 2D serum. Recombinants containing the entire surface glycoprotein (gp120) (V1-V5) and the 5' portion of gp41 of P13 and P23 and those containing gp120 sequences from V4 through the 5' portion of the transmembrane glycoprotein (gp41) were not neutralized by 2D serum. Using a panel of monoclonal antibodies in radioimmunoprecipitation assays, P23 and recombinants containing V4 and V5 of P23 were shown to be antigenically distinct from P13 and SIVmac239. The majority of the amino acid changes in the antigenically distinct viruses were clustered in V4 (amino acids 413-418) and these changes created new potential N-linked glycosylation sites. This study demonstrates that a small number of specific amino acid changes (amino acids 412 to 418 in the env gene) in the V4 region of the SIV envelope glycoprotein can alter antibody recognition and neutralization and that these phenotypic changes may be associated with altered glycosylation of the envelope.

Prolonged infection in rhesus macaques with simian immunodeficiency virus (SIVmac239) results in animal-specific and rarely tissue-specific selection of nef variants

We analyzed the sequence of nef genes from different tissues of three rhesus macaques that had been infected with molecularly cloned SIVmac239 for 88 to 92 weeks. Comparison of the predicted amino acid sequences revealed that each macaque had selected out specific amino acid substitutions and that most of this variation (70%) was confined to four regions, amino acids 39 to 75, 90 to 105, 153 to 167, and 191 to 217, comprising 36% of the protein. The nef genes in these animals underwent extensive genetic variation with average nucleotide and amino acid substitution rates varying from 0.86 to 2.84% and 2.47 to 6.27%, respectively, although tissue-specific selection of nef variants occurred in only 1 of 14 tissues examined in this study. Comparison of the rate of nucleotide and amino acid substitutions in the nef genes to those previously reported in the env in the central nervous system (CNS) and lymph node (LN) revealed that the predicted amino acid substitution rates for Nef were much higher than for the gp120 region of env in the CNS and LN tissues for one macaque. In the two other macaques, the predicted amino acid substitution rates were similar between these two proteins in LN tissues, but the amino acid substitution rates in Nef were significantly higher than in the gp120 from the CNS. Comparison of the nucleotide substitutions in the region of overlap between the env and the nef revealed that approximately 83% of the nucleotide substitutions in this area resulted in a Nef amino acid sequence change, 26% of the nucleotide substitutions resulted in a gp41 amino acid change, and 9.5% of nucleotide substitutions resulted in amino acid sequence changes in both proteins, suggesting a preference for the selection of amino acid substitutions in the Nef in these animals. Our results indicate that in animals infected with SIVmac239 for prolonged periods, variation in the nef occurs at rates similar to or exceeding that observed for the env gene.