Characterization of gag and pol sequences from long-term survivors of human immunodeficiency virus type 1 infection

We previously identified a group of 10 long-term survivors (LTS) of human immunodeficiency virus type 1 (HIV-1) infection. Extensive biological analysis revealed that some of these individuals do well, at least in part, because they possess weakened or attenuated viruses. Also, previously, to determine the genotype associated with the attenuated phenotype in vivo, we characterized nef, vif, vpr, vpu, env, and LTR in our cohort of LTS. In this study, we analyzed gag and pol genes derived from eight individuals in our cohort. For each subject multiple full-length gag and pol clones were obtained for analysis. In most cases, the sequences derived from the LTS had an intact open reading frame. At the protein level, there were no discernible differences between the sequences derived from LTS and those derived from patients with AIDS. Thus, no common defect in gag and pol was found in our cohort. One individual (subject SF), however, had only Gag-defective proviral sequences (10 of 10) in his peripheral blood mononuclear cells. Furthermore, longitudinal studies of the samples collected from SF over a 2-year period showed that all p17 gag clones sequenced (24 of 24) were defective due to G-to-A hypermutations. This viral defect in Gag may provide the molecular basis for this individual's extremely low viral load and long-term asymptomatic state. These results, together with previous findings in our LTS cohort, reinforce the notion that it is unlikely that a single common viral genetic determinant accounts for the lack of disease progression in all cases. Multiple host and viral factors undoubtedly contribute to the well-being of LTS of HIV-1 infection.     

The role of upstream U3 sequences in the pathogenesis of simian immunodeficiency virus-induced AIDS in rhesus monkeys

The nef reading frame overlaps about 70% of the U3 region of the 3' long terminal repeat (LTR) in primate lentiviruses. We investigated the functional role of these overlapping U3 sequences by analyzing the properties of three mutant forms of the pathogenic SIVmac239 clone. In mutant UScon, 90 of 275 bp in the upstream sequences (US) of U3 were changed in a conservative fashion without changing the predicted nef coding sequence. In mutant USnon, 101 of 275 bp in this region were changed in a nonconservative fashion, again without changing the predicted nef coding sequence. In mutant delta US, 275 bp in this region were deleted. Full-size, immunoreactive nef protein was synthesized in cells infected with the UScon and USnon mutants. The USnon and delta US mutants replicated with similar kinetics and to similar extents as wild-type, parental SIVmac239 in primary rhesus monkey peripheral blood mononuclear cell (PBMC) cultures. The UScon mutant replicated with slightly delayed kinetics in rhesus monkey PBMC cultures. In the CEMx174 cell line, the delta US mutant replicated similarly to the wild type, but the UScon and USnon mutants replicated with significantly delayed kinetics. Analysis of LTR-driven chloramphenicol acetyltransferase (CAT) activity and the effects of 5-azacytidine on virus replication suggested that the growth defect of the point mutants in CEMx174 cells was due in whole or in part to the introduction of multiple CG methylation sites in proviral DNA. Rhesus monkeys were experimentally infected with the UScon and USnon mutants, and the characteristics of the infection were compared with those of the parental SIVmac239. Analysis of the levels of plasma antigenemia, virus load, and CD4+ cells in PBMC revealed no decreased virulence of the mutant viruses. Analysis of lymph node biopsies taken from animals that received mutant viruses revealed histologic changes and levels of virus expression indistinguishable from those of the wild type. Furthermore, the wild-type behavior of the mutant viruses in rhesus monkeys occurred without any specific reversional events through at least 20 weeks of infection. These results, and the recent results of Kirchhoff et al. (F. Kirchoff, H. W. Kestler III, and R. C. Desrosiers, J. Virol. 68:2031-2037, 1994), suggest that these upstream sequences in U3 are primarily or exclusively nef coding sequence.     

Epstein-Barr virus nuclear antigen 2 transactivates the long terminal repeat of human immunodeficiency virus type 1

Human immunodeficiency virus type 1 (HIV-1)-infected subjects show a high incidence of Epstein-Barr virus (EBV) infection. This suggests that EBV may function as a cofactor that affects HIV-1 activation and may play a major role in the progression of AIDS. To test this hypothesis, we generated two EBV-negative human B-cell lines that stably express the EBNA2 gene of EBV. These EBNA2-positive cell lines were transiently transfected with plasmids that carry either the wild type or deletion mutants of the HIV-1 long terminal repeat (LTR) fused to the chloramphenicol acetyltransferase (CAT) gene. There was a consistently higher HIV-1 LTR activation in EBNA2-expressing cells than in control cells, which suggested that EBNA2 proteins could activate the HIV-1 promoter, possibly by inducing nuclear factors binding to HIV-1 cis-regulatory sequences. To test this possibility, we used CAT-based plasmids carrying deletions of the NF-kappa B (pNFA-CAT), Sp1 (pSpA-CAT), or TAR (pTAR-CAT) region of the HIV-1 LTR and retardation assays in which nuclear proteins from EBNA2-expressing cells were challenged with oligonucleotides encompassing the NF-kappa B or Sp1 region of the HIV-1 LTR. We found that both the NF-kappa B and the Sp1 sites of the HIV-1 LTR are necessary for EBNA2 transactivation and that increased expression resulted from the induction of NF-kappa B-like factors. Moreover, experiments with the TAR-deleted pTAR-CAT and with the tat-expressing pAR-TAT plasmids indicated that endogenous Tat-like proteins could participate in EBNA2-mediated activation of the HIV-1 LTR and that EBNA2 proteins can synergize with the viral tat transactivator. Transfection experiments with plasmids expressing the EBNA1, EBNA3, and EBNALP genes did not cause a significant HIV-1 LTR activation. Thus, it appears that among the latent EBV genes tested, EBNA2 was the only EBV gene active on the HIV-1 LTR. The transactivation function of EBNA2 was also observed in the HeLa epithelial cell line, which suggests that EBV and HIV-1 infection of non-B cells may result in HIV-1 promoter activation. Therefore, a specific gene product of EBV, EBNA2, can transactivate HIV-1 and possibly contribute to the clinical progression of AIDS.     

Virus load as a marker of disease progression in HIV-infected children

The relationship of virus load to clinical disease progression in HIV-infected children remains to be elucidated. In this study, HIV-1 proviral DNA load was determined in peripheral blood mononuclear cells (PBMCs) by the quantitative competitive DNA polymerase chain reaction assay (QC-PCR) in 47 HIV-infected children subdivided by age (group I, < or = 2 years; group II, > or = 5 years), who were further categorized to include 12 rapid progressors (RP, age < or = 2 years, Centers for Disease Control [CDC] defined clinical category C and/or immune category 3, or death before age 2 years) and slow progressors (SP, age > or = 5 years, excluding CDC categories C and/or immune category 3). Significantly higher mean proviral copies/10(3) PBMCs were detected in group I versus group II (75.4 +/- 104.3 and 13.0 +/- 17.8 respectively, p < 0.0001) and in RP (158.0 +/- 118.2) as compared to either SP (11.8 +/- 18.8, p < 0.0001) or other age-matched infected children (20.3 +/- 38.8, p < 0.0001). Thus HIV-infected children appear to have a higher cell-associated virus load early in life, especially in association with rapid disease progression.     

Selective amplification of simian immunodeficiency virus genotypes after intrarectal inoculation of rhesus monkeys

Animal models for sexual transmission of human immunodeficiency virus can define the influences of virus type, dose, and route of inoculation on infection and clinical outcome. We used an uncloned simian immunodeficiency virus stock (SIVmac) to inoculate cells in vitro and to inoculate rhesus monkeys by intravenous and intrarectal routes. The distribution of virus genotypes present in each of these infection examples was characterized by DNA sequence analysis of viral long terminal repeats (LTRs). Our analysis of LTR sequences from in vitro and in vivo infections revealed three main genotypes: one genotype was observed only for in vitro infection, and two other genotypes were recovered only from infected animals. By comparing animals inoculated with high intrarectal doses of SIVmac and those inoculated with low doses, we demonstrated that unique subsets of the stock were selected after intrarectal infection. Our findings indicate that minor genotypes present in the stock cross the rectal mucosa and are amplified selectively to become prominent in peripheral blood mononuclear cells from acutely infected animals. Studies with a molecular recombinant of SIV and human immunodeficiency virus type 1 sequences, SHIV, showed that viral LTR sequences do not undergo especially rapid sequence variation or rearrangement after intrarectal inoculation. The mucosal barrier exerts a significant influence on infection and disease progression by reducing the efficiency of SIVmac infection and by permitting distinct, pathogenic genotypes to become established in the host.