Biological and virologic characteristics of primary HIV infection

BACKGROUND: The clinical events surrounding acute HIV-1 infection have been well described, but little is known about whether the virologic course of acute HIV-1 infection influences the subsequent progression of disease. OBJECTIVE: To define the virologic natural history of acute and very early HIV infection. DESIGN: Prospective, longitudinal cohort study. SETTING: University of Washington Research Clinic PARTICIPANTS: 74 adults enrolled soon after acquisition of HIV (mean, 69 days). MEASUREMENTS: Plasma HIV-1 RNA levels; quantitative cell cultures; CD4 cell counts; and detailed clinical assessments done at study entry, biweekly for 1 month, monthly for 2 months, and quarterly thereafter. RESULTS: In the first 30 days after acquisition of HIV, HIV-1 RNA levels varied greatly among participants (range, 27,200 to 1.6 x 10(6) copies per mL of plasma). Levels of HIV-1 RNA decreased by a mean of 6.5% per week for the first 120 days and then increased by a mean of 0.15% per week. CD4 cell counts decreased by a mean of 5.2 cells/mm3 per week for the first 160 days and by a mean of 1.9 cells/mm3 per week thereafter (P < 0.01). Disease progressed faster in participants who sought medical care for their acute seroconversion syndrome (P = 0.01) and those who had high plasma HIV-1 RNA levels 120 to 365 days after acquisition (P < 0.01). Peak levels in the first 120 days were not predictive of disease progression. CONCLUSIONS: The variability in viral RNA levels associated with acute HIV-1 infection is greater than previously appreciated. Within 120 days of acquisition, plasma HIV RNA levels rapidly decrease to an inflection point, after which they gradually increase. Virus-host interactions soon after acquisition seem to have a major influence on the long-term outcome of HIV-1 disease.     

Apparent enhancement of perinatal transmission of human immunodeficiency virus type 1 by high maternal anti-gp160 antibody titer

The presence of antibodies able to enhance infection in vitro in sera from human immunodeficiency virus (HIV)-1-infected patients raises the possibility that antibodies exert a deleterious activity during natural infection. The anti-HIV-1 humoral response and plasma HIV-1 RNA were measured in a cohort of 98 infected mothers, included in the French Prospective Study on Pediatric HIV Infection, 49 of whom transmitted HIV to their children. Transmission from mother to child was associated with antibody responses to the envelope gp160 (P = .009 for serum dilution of 1/400) and to a highly conserved domain of the transmembrane glycoprotein (P = .055 for serum dilution of 1/400) and with plasma HIV-1 RNA levels (P < .0001). Multivariate logistic regression indicated that a high anti-gp160 response and a high plasma virus load are independent risk factors for perinatal transmission of HIV-1 (odds ratio, 3.4; 95% confidence interval, 1.1-9.9 for anti-gp160; odds ratio, 2.8; 95% confidence interval, 1.6-5.0 for virus load).     

Differential effects of interleukin-12, interleukin-15, and interleukin-2 on human immunodeficiency virus type 1 replication in vitro

Cytokines may have clinical utility as therapeutic agents for human immunodeficiency virus type 1 (HIV-1) infection and as an adjuvant for vaccines. The effect of interleukin-12 (IL-12) and IL-15 on in vitro HIV-1 replication was investigated. IL-12 and IL-15 at doses up to 10 ng/ml had little effect on basal HIV-1 p24 antigen production by chronically HIV-infected T (ACH-2) and monocytic (U1) cell lines. For ACH-2 cells stimulated with phorbol 12-myristate 13-acetate (PMA; 50 ng/ml), IL-12 and IL-15 significantly increased p24 antigen production by 20 and 30%, respectively (n = 6). In contrast, IL-12 and IL-15 (10 ng/ml) treatment of PMA-stimulated U1 cells decreased p24 antigen production by 16 and 15%, respectively (n = 6). We next studied the effect of IL-12 and IL-15 on HIV-infected peripheral blood mononuclear cells (PBMCs). In 10 HIV-seropositive patients' PBMCs cocultured with mitogen-activated HIV-seronegative donor cells, two patterns of p24 antigen production were observed in response to IL-2: low (p24 antigen production < 10(3) pg/ml; n = 8) and high (p24 antigen production > 10(3) pg/ml; n = 2) response. For the low-response pattern, IL-12 and IL-15 increased viral replication by 97-fold and 100-fold, respectively (P = 0.05 and 0.004, respectively). For the high-response pattern, both IL-12 and IL-15 suppressed HIV replication. The effect of IL-2, IL-12, and IL-15 on acute in vitro infection by HIV-1JRCSF was also examined. IL-12 did not increase p24 antigen production above basal levels while IL-2 and IL-15 significantly enhanced p24 antigen production (by approximately 2-fold). In conclusion, IL-12 and IL-15 may have differential effects on latent and acute HIV infection, and their ability to enhance HIV production may depend on cell activation. Thus, the use of these cytokines may be dictated by the clinical state of the patient.     

Reduction of diagnostic window by new fourth-generation human immunodeficiency virus screening assays

In order to reduce the diagnostic window between the time of human immunodeficiency virus (HIV) infection and laboratory diagnosis, new screening enzyme-linked immunosorbent assays (ELISAs) which permit the simultaneous detection of HIV antigen and antibody have been developed. Two fourth-generation assays, HIV DUO (Biomérieux) and HIV Combi (Boehringer Mannheim), for the combined detection of HIV antigen and antibody, were compared with a third-generation assay (HIV-1/HIV-2 3rd Generation Plus enzyme immunoassay [EIA]; Abbott) and a p24 antigen test (HIV-1 Ag monoclonal; Abbott). A total of 17 seroconversion panels, 15 cell culture supernatants infected with different HIV type 1 (HIV-1) subtypes, and 255 potentially cross-reactive serum samples were tested. Ten seroconversions were detected an average of 8.1 days earlier with HIV DUO and 7.5 days earlier with HIV Combi than with the third-generation ELISA. Overall, in the 17 seroconversion panels tested, HIV DUO detected HIV-1 infection an average of 4.8 days and HIV Combi detected infection an average of 4.4 days earlier than HIV-1/HIV-2 3rd Generation Plus EIA. HIV antigen was detected with HIV DUO and HIV Combi in all of the 15 cell culture supernatants infected with different HIV-1 subtypes, including subtype O. With fourth-generation assays, considerably fewer false-positive results (n = 4 to 6) were obtained, in comparison with the third-generation EIA (n = 18). Fourth-generation assays permit an earlier diagnosis of HIV infection than third-generation antibody screening assays through the detection of p24 antigen, which may be present in serum samples from individuals with recent HIV infection prior to seroconversion.     

Comparison of NucliSens and Amplicor monitor assays for quantification of human immunodeficiency virus type 1 (HIV-1) RNA in plasma of persons with HIV-1 subtype A infection in Abidjan, C?te d'Ivoire

We compared the sensitivity and accuracy of the NucliSens assay and those of both the standard and modified (addition of a new primer set, primer mix 1, supplied by Roche) Amplicor HIV Monitor assays to quantify human immunodeficiency virus type 1 (HIV-1) RNA in persons infected with HIV-1 subtype A in Abidjan, C?te d'Ivoire. Seventy-one plasma samples from HIV-1-seropositive persons at different stages of HIV infection and 15 samples from HIV antibody-negative persons were analyzed. The HIV-1 genetic subtype was determined either by DNA sequencing or by a restriction fragment length polymorphism assay. Of the 71 samples, 70 (98%) were subtype A and 1 was subtype G. Of the 70 subtype A samples, the proportion of RNA-positive plasma samples and mean HIV-1 RNA levels were significantly higher by the modified HIV Monitor assay (n = 67 [96%]; mean RNA levels, 5.2 log10 HIV-1 RNA copies/ml) than the NucliSens assay (n = 56 [80%]; 4.3 log10 HIV-1 RNA copies/ml) or the standard HIV Monitor assay (n = 44 [63%]; mean RNA levels, 3.8 log10 HIV-1 RNA copies/ml) (all P values were <0.05). The HIV-1 RNA levels by the modified HIV Monitor assay correlated significantly with those by the NucliSens assay (r = 0.76; P < 0.001) and the standard HIV Monitor assay (r = 0.57; P < 0.001), as did the RNA levels by the NucliSens and the standard HIV Monitor assays (r = 0.60; P < 0. 001). Lower CD4 cell counts were significantly correlated with higher HIV-1 RNA levels by all three assays (r = -0.47 for the NucliSens assay, -0.45 for the standard HIV Monitor assay, and -0.62 for the modified HIV Monitor assay). These results indicate that the modified HIV Monitor assay has the highest sensitivity and efficiency at quantifying the levels of RNA in persons infected with HIV-1 subtype A and thus constitutes a valuable tool for the monitoring of RNA levels in areas of Africa were HIV-1 subtype A is predominant.     

Human T cell lymphotropic virus type I does not increase human immunodeficiency virus viral load in vivo

Human T lymphotropic virus type I (HTLV-I) can increase human immunodeficiency virus (HIV) replication in vitro, and several studies suggest that HTLV-I accelerates the progression of HIV infection. To determine whether HTLV-I enhances HIV replication in vivo, a case-control study was done of serum HIV viral load, using polymerase chain reaction, in 23 subjects with HTLV-I/HIV coinfection and 92 control subjects with HIV single infection. The geometric mean serum RNA level was 11,482 copies/mL in the coinfected group and 13,804 in the single-infection group (P = .57), a result that did not change after adjustment for zidovudine use and CD4 cell count. Among subjects with advanced HIV infection, there was a trend toward higher viral load among singly infected subjects. HTLV-I did not appear to increase HIV plasma RNA levels in subjects with coinfection. These results do not provide a biologic basis for the hypothesis that HTLV-I accelerates the course of HIV infection.     

The prevalence of herpes family virus DNA in the conjunctiva of patients positive and negative for human immunodeficiency virus using the polymerase chain reaction

OBJECTIVE: To help understand the pathogenesis of herpes family virus ocular infection among patients positive for HIV, the authors compared the rates of detection of herpes family virus DNA from the conjunctiva of patients who are positive and negative for human immunodeficiency virus (HIV) using the polymerase chain reaction (PCR). DESIGN: Cross-sectional study. PARTICIPANTS: The conjunctival scrapings of 30 patients positive for HIV and 30 patients negative for HIV were examined. INTERVENTION: PCR was used to assay for the presence of herpes simplex virus type 1 (HSV), varicella-zoster virus (VZV), cytomegalovirus (CMV), and Epstein-Barr virus (EBV) DNA (n = 240 samples). MAIN OUTCOME MEASURE: The rate of detection of virus DNA in the two groups, controlling for age, gender, and race, was measured. RESULTS: HSV and VZV DNA were not detected in any of the HIV-positive or HIV-negative samples. CMV DNA was detected in 20% (6 of 30) of patients positive for HIV and was undetected in control subjects negative for HIV (P = 0.01). EBV DNA was detected in 40% (12 of 30) of patients positive for HIV and in 47% (14 of 30) of control subjects negative for HIV (P = 0.58). CONCLUSIONS: There was no difference in the frequency of detection of HSV, VZV, or EBV DNA from the conjunctiva of patients positive or negative for HIV. Only CMV DNA was detected at a significantly higher rate in the conjunctiva of patients positive for HIV compared with control subjects negative for HIV. These different rates of peripheral virus shedding may be one possible explanation for the different rates of clinical infection among the herpes family viruses among patients positive for HIV.     

Levels of HIV-infected peripheral blood cells remain stable throughout the natural history of HIV-1 infection. Swiss HIV Cohort Study

OBJECTIVE: To clarify the relationship between the number of provirus-bearing peripheral blood mononuclear cells (PBMC) and HIV-1 disease progression during the natural history of infection. DESIGN: Twenty-four HIV-1-infected subjects with known seroconversion dates and long-term follow-up were retrospectively identified using the Swiss HIV Cohort Database. PBMC specimens from this cohort were retrieved from storage for analysis. METHODS: Infected PBMC equivalents were determined by HIV-1 DNA quantitative competitive (QC)-PCR. The results were analysed with respect to HIV-1 disease stage and compared with a mathematical model of long-term HIV-1 disease progression. RESULTS: PBMC HIV-1 DNA did not correlate with major indices of disease progression, including time following primary infection, time before reaching a CD4 cell count less than 200 x 10(6)/l, and time before death. The number of PBMC harbouring HIV-1 provirus was relatively constant throughout the clinical stages of HIV-1 infection, consistent with simulated data from a mathematical model of long-term HIV-1 infection. We also showed that a biased interpretation of the QC-PCR data may arise when the values are expressed as HIV-1 DNA copies per PBMC or per CD4 cell. CONCLUSIONS: This analysis suggests that levels of provirus-bearing PBMC remain constant during the natural course of HIV-1 infection, whereas plasma virus load typically increases logarithmically during the same period. The hypothesis that plasma virus levels are directly related to the number of infected cells may deserve reconsideration.     

Lysis of CD4+ T cells expressing HIV-1 gag peptides by gag-specific CD8+ cytotoxic T cells

The vast majority of in vitro experiments testing the cytotoxic T lymphocytes (CTL) activity in HIV infection has been performed with target cells consisting of autologous EBV-transformed B lymphoblastoid cell lines (B-LCLs) expressing Human immunodeficiency virus type I (HIV-1) proteins. However data concerning the lysis of primary CD4+ T lymphocytes expressing HIV-1 antigens by CTLs is still lacking. To study the CTL activity against such primary targets, we used a system involving PBMCs of an HIV+ asymptomatic patient (PT) as effector cells and the CD4+ lymphocytes or B-LCLs of his healthy HLA-identical twin brother (HTW) as target cells. These syngeneic targets were either infected with recombinant vaccinia virus containing HIV-1 gag gene (gag-vac), or coated with HIV-1 gag peptides. We demonstrate in this study that PT CTLs (which were CD3+, CD4-, CD8+, TCRalphabeta+, TCRgammadelta-, CD56-) specifically lysed both types of syngeneic target cells expressing gag-vac; however, CD4+ T cells expressing HIV gag proteins were lysed less efficiently than B-LCLs expressing the same HIV epitopes. On the other hand, no specific lysis was detected when the target cells were uninfected or infected by wild-type vaccinia virus.     

HIV-1 genetic subtype A/B recombinant strain causing an explosive epidemic in injecting drug users in Kaliningrad

OBJECTIVES: To investigate the molecular epidemiology and genetic structure of the virus strain(s) causing an outbreak of HIV-1 infection in the Kaliningrad province of the Russian Federation and to investigate the relationship of this outbreak to some other emerging HIV-1 epidemics in the countries of the former Soviet Union. DESIGN: A molecular epidemiological investigation was conducted in the city of Kaliningrad amongst individuals recently diagnosed as HIV-1-positive. Samples were also collected from neighbouring Lithuania and from the Ukraine. METHODS: Incident and population data was collected from official health statistics in Kaliningrad. A standardized questionnaire was administered to newly diagnosed individuals to assess risk factors for HIV-1 infection. For genotyping, two regions of the virus (env C2-V3 and gag NCp7) were directly sequenced. RESULTS: The number of newly diagnosed individuals testing seropositive for HIV-1 infection in Kaliningrad rose from less than one per month to more than 100 per month during the period of July-October 1996. A total of 1335 new infections were identified between 1 July 1996 and 30 June 1997. The main reported risk factor for HIV-1 infection (80%) was injecting drug use, in particular with a locally produced opiate. Sequence analysis of patient viruses in Kaliningrad (n = 50) showed that the epidemic was caused by a highly homogenous HIV-1 strain, recombinant between the genetic subtypes A and B. Comparison with subtype A strains prevalent amongst injecting drug users (IDU) in the Ukraine showed that one of these strains was the direct subtype A parent of the epidemic A/B recombinant strain in Kaliningrad. CONCLUSIONS: The HIV-1 epidemic in Kaliningrad probably started from a single source, with rapid spread of the virus through the IDU population. The origin of the epidemic strain is a recombination event occurring between the subtype A strain virus prevalent among IDU in some southern CIS countries, and a subtype B strain of unknown origin.     

Cross-clade human immunodeficiency virus (HIV)-specific cytotoxic T-lymphocyte responses in HIV-infected Zambians

We have examined cross-clade HIV-specific cytotoxic T-lymphocyte (CTL) activity in peripheral blood of eight Zambian individuals infected with non-B-clade human immunodeficiency virus type 1 (HIV-1). Heteroduplex mobility assay and partial sequence analysis of env and gag genes strongly suggests that all the HIV-infected subjects were infected with clade C HIV-1. Six of eight C-clade HIV-infected individuals elicited CTL activity specific for recombinant vaccinia virus-infected autologous targets expressing HIV gag-pol-env derived from B-clade HIV-1 (IIIB). Recognition of individual recombinant HIV-1 B-clade vaccinia virus-infected targets expressing gag, pol, or env was variable among the patients tested, indicating that cross-clade CTL activity is not limited to a single HIV protein. These data demonstrate that HIV clade C-infected individuals can mount vigorous HIV clade B-reactive CTL responses.     

Role of nitric oxide in the regulation of lymphocyte apoptosis and HIV-1 replication

Nitric oxide (NO) has been implicated in certain immunopathogenetic mechanisms during the course of infection with human immunodeficiency virus (HIV). We have evaluated the levels of NO release and lymphocyte apoptosis in peripheral blood mononuclear cell (PBMC) cultures from HIV-1 infected subjects and healthy controls. We have also examined these 2 parameters in parallel cultures maintained under conditions where either NO synthesis was inhibited or high level of NO was present. Nitrite contents in culture supernatants were measured as the stable end products of the released NO. Levels of spontaneous apoptosis and activation-induced cell death (AICD) by anti-CD3 or by phytohemagglutinin were evaluated using flow cytometry. Additional experiments were also aimed at addressing a potential link between NO synthesis and HIV-1 replication in human monocyte-derived macrophages (MDMs). Acutely infected MDMs with HIV-1Bal were maintained in culture, without any additional activation signal, for a period of 14 days. Nitrites in the supernatants and mRNA accumulation of the inducible NO synthase (iNOS) in infected cells were assessed over the whole culture period. In addition, the effect of blocking NO synthesis during and after infection of MDMs, using an inhibitor of NO, was evaluated on the level of viral replication as measured by the presence of P24 antigen in the supernatants. Similarly, the effect on HIV replication of high NO levels in MDM cultures, supplied by a donor of NO during the 24 h period of infection, was also studied. We conclude that no elevation in NO release could be detected in PBMC cultures from HIV-1 infected subjects and that modulation of NO content may slightly regulate the level of spontaneous lymphocyte apoptosis but not that of AICD. Infection of MDMs with HIV-1 does not seem to induce detectable NO release or iNOS mRNA accumulation. Similarly, neither inhibition of NO synthesis nor the presence of high NO levels during the infection period could modify the outcome of virus replication in macrophages.     

Cultured blood dendritic cells retain HIV-1 antigen-presenting capacity for memory CTL during progressive HIV-1 infection

Dendritic cells (DC) are potent APC that may be involved in the pathogenesis of HIV-1 infection. We studied the APC function of DC from HIV-1-infected subjects that were derived from monocyte-depleted PBMC by culture in human IL-4 and human granulocyte-macrophage CSF. The cultured cells from the HIV-1-infected subjects had similar morphology and phenotype of mature DC (CD80 = 41 +/- 8%, CD86 = 77 +/- 5%, CD40 = 87 +/- 6%, CD1a = 1 +/- 1%) to DC cultured from seronegative subjects. The yield of these DC was lower than from HIV-1-seronegative subjects (4 +/- 0% vs 11 +/- 2%, p < 0.01), and the lower DC yields correlated with lower numbers of blood CD4+ T cells (r = 0.60, p < 0.01) and higher plasma viral load (r = -0.49, p < 0.01). DC from HIV-1-infected subjects were infected with recombinant vaccinia virus vectors expressing Gag, Pol, and Env and were able to stimulate equal or higher levels of MHC class I-restricted, anti-HIV-1 memory CTL (CTLm) than were similarly treated, autologous B lymphocyte cell lines. DC pulsed with peptides representing HIV-1 CTL epitopes stimulated higher levels of anti-HIV-1 CTLm responses than did DC infected with the vaccinia virus-HIV-1 constructs. Allogeneic, MHC class I-matched DC also stimulated anti-HIV-1 CTLm activity in cells from HIV-1-infected subjects. DC from early and late stages of HIV-1 infection had a similar ability to activate CTLm specific for targets expressing either HIV-1 genes via vaccinia virus vectors or HIV-1 immunodominant synthetic peptides. However, DC from either early or late stages of HIV-1 infection could not overcome the defect in anti-HIV-1 CTLm response in advanced infection.     

Plasma levels of monocyte chemoattractant protein-1 but not those of macrophage inhibitory protein-1alpha and RANTES correlate with virus load in human immunodeficiency virus infection

Plasma levels of proinflammatory cytokines, cytokine inhibitors, and the beta chemokines RANTES, macrophage inhibitory protein (MIP)-1alpha, and monocyte chemoattractant protein (MCP)-1 were studied in relationship with virus load in 40 patients exhibiting plasma levels of HIV RNA ranging between undetectable and levels >10(6) copies/mL. Mean plasma levels of MCP-1 were increased in patients with high virus load compared with HIV-seropositive subjects with undetectable plasma viral RNA and healthy controls. MCP-1 levels were directly correlated with plasma levels of HIV RNA. No correlation was observed between virus load and plasma concentrations of MIP-1alpha and RANTES. The results suggest that low rates of viral replication in vivo are not dependent on increased production of the suppressive chemokines RANTES and MIP-1alpha. Since MCP-1 upregulates viral replication in vitro, the results may suggest a role for MCP-1 in triggering viral replication in HIV disease.     

Evaluation of human immunodeficiency virus (HIV) type 1 RNA levels in cerebrospinal fluid and viral resistance to zidovudine in children with HIV encephalopathy

The amount of human immunodeficiency virus (HIV) type 1 RNA and the presence of a codon 215 mutation indicative of zidovudine resistance were evaluated in cerebrospinal fluid (CSF) and plasma obtained from HIV-1-infected children. The level of HIV-1 RNA in CSF was highest in children with severe encephalopathy (n = 25; median, 430 copies/mL; range, 0-2.2 x 10(5) copies/mL) followed by the moderately encephalopathic (n = 7; median, 330; range, 0-1130) and nonencephalopathic groups (n = 9; median, 0; range, 0-566) (P = .007). There was no correlation between CSF and plasma HIV-1 RNA levels. Five of 7 children with the codon 215 mutation in CSF had a progression of encephalopathy, while all 8 children with wild type codon 215 had improved or stable disease during zidovudine treatment (P = .007). These findings suggest that increased viral replication and emergence of drug-resistant HIV-1 variants within the central nervous system may play a role in progression of HIV encephalopathy.     

Quantitation of human immunodeficiency virus type 2 DNA in peripheral blood mononuclear cells by using a quantitative-competitive PCR assay

A new quantitative-competitive PCR-based human immunodeficiency virus type 2 (HIV-2) proviral DNA assay (QC-PCR) was developed and used to determine the proviral load in HIV-2-infected individuals. Proviral load varied considerably, with means of 1,831 copies per 10(6) peripheral blood mononuclear cells for asymptomatic subjects (n = 19) and 2,587 for AIDS patients (n = 2). HIV-2 viral and proviral loads also varied significantly over time in asymptomatic patients. These data suggest that a high level of virus replication occurs throughout the asymptomatic phase of HIV-2 infection.     

Virus burden in lymph nodes and blood of subjects with primary human immunodeficiency virus type 1 infection on bitherapy

At present, it is not known whether undetectable plasma viremia corresponds to an absence of human immunodeficiency virus type 1 (HIV-1) replication in lymphoid tissues. This issue has been explored in 11 subjects with primary HIV-1 infection treated with zidovudine plus didanosine by evaluating virologic markers in blood and lymphoid tissues 9-18 months after initiation of treatment. These markers include plasma viremia, measured with a sensitive assay with a detection limit of 20 HIV-1 RNA copies/mL, infectious virus titers and proviral DNA in lymph node mononuclear cells, and HIV-1 RNA in lymphoid tissue. Five subjects had plasma viremia <20 copies/mL and showed no evidence of viral replication in lymphoid tissue. Six subjects had both detectable plasma viremia and evidence of HIV-1 RNA in lymphoid tissue. The results indicate that absence of detectable HIV RNA in lymphoid tissue is associated with viremia levels of HIV-1 RNA <20 copies/mL.