To evaluate the determinants of HIV-1 RNA shedding in cervicovaginal secretions and the effects of antiretroviral therapy in a group of infected women. A total of 122 women from whom paired peripheral blood and cervicovaginal lavage samples were available were enrolled in the study. HIV-1 RNA was quantified in the plasma and cell-free fraction of cervicovaginal lavages by the nucleic acid sequence-based amplification assay (lower limit of detection 80 copies/ml). Seventy-one per cent of the women had detectable viral load in the cervicovaginal lavage and this appeared to be correlated to plasma viral load and to the degree of immunodeficiency as expressed by the absolute number of CD4 cells. Antiretroviral-treated patients had a lower risk of shedding the virus in the genital tract, but this association was limited to patients treated with highly active antiretroviral therapy (HAART). However, in 25% of women with undetectable plasma viral load, a genital shedding of the virus was demonstrated. Plasma viral load may fail as a marker of infectivity of genital secretions. HAART treatment seems to be more efficacious in suppressing viral shedding at the genital level. The female genital tract represents a distinct compartment for HIV-1 replication/evolution.

Neutralizing activity against primary human immunodeficiency virus type 1 (HIV-1) isolates from 17 persons who were long-term disease nonprogressors (LTNPs) and 13 persons who were fast progressors (FPs) was compared. Sera from LTNPs showed higher neutralizing activity both in titer and in host spectrum than did sera from FPs. However, LTNP sera had limited neutralizing activity against HIV-1 subtypes from different geographic areas. Sera collected 6 years earlier from both groups had limited neutralizing activity, indicating that early responses are not predictive for disease progression. LTNPs had very low virus loads, as reflected by only one positive isolation, which was an MT-2-negative phenotype. Virus was isolated from all FPs, and the isolates showed a phenotype switch from MT-2 negative to MT-2 positive. Development of high-titer, broadly cross-reactive neutralizing antibodies is associated with control of virus replication and low virus load in HIV-1-infected LTNPs.

Epidemiologic studies suggest that human herpesvirus 8 (HHV-8) may be sexually transmitted. To study the potential for HHV-8 transmission through cervicovaginal (CV) secretions, the presence of HHV-8 DNA was investigated by nested polymerase chain reaction in the cellular fraction of CV secretions from 36 human immunodeficiency virus type 1 (HIV-1)-seropositive and 29 HIV-1—seronegative women. The same patients were tested for antibodies to two defined HHV-8 antigens (latency-associated nuclear antigen and open-reading frame 65-encoded structural protein) and for HHV-8 DNA in their peripheral blood mononuclear cells (PBMC). The findings were compared with the rate of HHV-8 detection in semen samples of 20 HIV-1—infected men. HHV-8 DNA was detected in the CV samples from only 1 HHV-8-seropositive AIDS patient, in 3 PBMC samples (1/29 HIV-1—seronegative patients, 1/3 AIDS patients with Kaposi's sarcoma, and 1/19 AIDS patients), and in 1 of 20 semen samples. HHV-8 infection was more common in HIV-1—infected than uninfected women. Thus HHV-8 DNA is only rarely detectable in CV secretions and semen of HHV-8—infected individuals.

To verify the compartmentalization of HIV-1 within the central nervous system (CNS) and to define whether viral phenotype of HIV-1 isolates from cerebrospinal fluid (CSF) samples and CSF viral load correlate with the presence and type of neurological disorders. A total of 33 HIV-1-infected patients with and without neurological disorders were included in the study. HIV-1 isolation from paired CSF and peripheral blood mononuclear cell (PBMC) samples was attempted by a standard cocultivation technique; the biological phenotype of HIV-1 isolates was assessed by the MT-2 cell assay. CSF and plasma HIV-RNA levels were measured by a quantitative reverse transcripase–polymerase chain reaction. The rate of HIV-1 isolation from CSF and PBMC was 66% (22 isolates) and 85% (28 isolates), respectively. Seventeen out of 22 (77%) CSF HIV-1 isolates were characterized as non-syncytium-inducing, and 15 out of 28 (68%) isolates from PBMC were typed as syncytium-inducing (SI). The presence of SI isolates in CSF was limited to patients with HIV-1-, cytomegalovirus- or JC virus-related disorders and was often associated with high levels of HIV-1 RNA in the CSF. Our results demonstrate a correlation between high levels of HIV RNA in CSF and the presence of neurological disorders thus indicating a possible role for HIV-1 RNA in the CSF as a biological marker of neurological disease. The finding of viruses with a different phenotype in paired CSF and PBMC indicates that HIV-1 may evolve differently in the brain and in the blood. This suggests compartmentalization of HIV-1 within the CNS.