INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Primary infection with human immunodeficiency virus type 1 (HIV-1) can elicit an acute retroviral syndrome characterized by fever, pharyngitis, lymphadenopathy, myalgia, rash, and headache (2). Although estimates vary, as many as half of HIV-1-infected persons in the United States experience some symptoms of this acute retroviral syndrome between 2 and 6 weeks after exposure (21, 32). The variable response to HIV-1 infection may be linked to host genetics, coincident infections with viral or nonviral agents, previous immunological experience, or particulars of the infecting virus strain. At present, there is little understanding of the events during acute infection and we have no explanations for variation within a population or for how differences in the initial host-virus interactions influence the course of disease progression.

Studies of acute HIV-1 syndrome noted transiently high levels of plasma viremia that were not correlated directly with the severity of symptoms (2, 6). Elevated plasma HIV-1 RNA levels during the initial 4 months were not associated subsequently with higher rates of CD4⁺ T-cell depletion (26), although faster progression to AIDS may occur among individuals who experience more severe acute infection symptoms (23, 26). At the end of the acute infection interval, virus-specific cellular and humoral immune responses arise (21) and plasma viral RNA levels stabilize to provide a prognostic marker for the risk of AIDS (12, 26). Our studies seek to understand host-virus interactions during acute infection, to explain mechanisms accounting for variation within a population of infected individuals, and to show how events of early infection influence subsequent disease progression.

The rhesus macaque model is well suited for studies of acute virus infection. This outbred animal population retains sufficiently complex host genetics to reveal populational variation, and we can control for virus type, along with the time, dose, and route of exposure (33). Use of the pathogenic simian/human immunodeficiency virus (SHIV) strain SHIV_89.6PD (25) and intrarectal (i.r.) inoculation (29) provides a model for infection with acute CD4⁺ T-cell loss. In this system, rapid disease progression was associated with greater than 90% CD4⁺ T-cell loss during the first month after infection whereas prolonged survival was correlated with moderate CD4⁺ T-cell losses, increases in circulating B cells, and production of virus-binding antibody (29).

In the present study, we first used a cohort of six macaques that were infected with SHIV_89.6PD by the intravenous (i.v.), i.r., or intravaginal (i.vag.) route. These animals were necropsied between days 4 and 15 after infection to assess patterns and kinetics of virus dissemination. A second cohort of three macaques was infected with SHIV_89.6PD by the i.r. route for a detailed study of changes in lymphocyte phenotype and the kinetics of cell depletion, and we obtained baseline data from a number of virus-naive animals to show the range of lymphocyte counts and expression of activation markers. We observed that CD4⁺ T cells are highly activated during acute infection and are unusually susceptible to activation-induced cell death (AICD). Overall levels of lymphocyte activation marker expression were maximal between weeks 2 and 4 and coincided with increased CD4⁺ T-cell susceptibility to apoptosis and peak viremia at week 2. Substantial lymphocyte activation appeared to increase the level of SHIV replication and promote cell death by apoptosis. The combined effect of these mechanisms may be to reduce the levels of antiviral immunity and increase subsequent rates of disease progression.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Animals and virus infections. Eighteen captive-bred rhesus macaques (Macaca mulatta) were housed at the Wisconsin Regional Primate Research Center (WRPRC) and used in these studies. The WRPRC is accredited by the American Academy of Laboratory Animal Care. All animal research protocols were approved by the Institutional Animal Care and Use Committee. Animals were confirmed negative for antibodies to simian immunodeficiency virus and type D simian retroviruses and were negative by standard coculture assays prior to these studies. Macaques were restrained with ketamine hydrochloride (10 mg/kg of body weight) before all virus inoculations and blood collections. General Medical Laboratories (Madison, Wis.) performed automated complete blood counts (CBC) on all samples.

Virus isolation and plasma antigenemia. Peripheral blood was collected in heparinized blood tubes. Mononuclear cells were isolated by Ficoll-Hypaque (Pharmacia Biotech, Piscataway, N.J.) gradient centrifugation. Starting with 10⁶ PBMC, duplicate serial 10-fold dilutions of cells were stimulated overnight with 0.5 µg of phytohemagglutinin (PHA; Murex Diagnostics Inc., Dartford, United Kingdom) per ml in complete RPMI medium (Gibco Bethesda Research Laboratories, Grand Island, N.Y.) with 10% fetal bovine serum (Harlan, Indianapolis, Ind.), 2 mM L-glutamine, 100 U of penicillin per ml, and 100 µg of streptomycin (Sigma, St. Louis, Mo.) per ml. The next day, medium was removed and cells were cocultured with 2.5 × 10⁵ CEM×174 cells in 2 ml of complete RPMI medium. Cultures were split twice weekly until positive virus isolation was determined by the appearance of a cytopathic effect or the cultures were scored as virus isolation negative after 1 month. Cocultures of mononuclear cells from tissues were performed as described above, with the exception that as many as 5 × 10⁶ cells were seeded per well as the upper limit. Plasma samples were assayed for p27 by enzyme-linked immunosorbent assay ELISA (Coulter, Miami, Fla.).

Tissue collection and processing. Venous blood was collected from animals anesthetized with ketamine hydrochloride. Euthanasia was performed by i.v. administration of 0.3 ml of Beuthanasia (pentobarbital sodium and phenytoin sodium; Schering-Plough Animal Health, Kenilworth, N.J.) per kg of body weight. At necropsy, the lymph nodes (LN), spleen, thymus, ileum, cecum, and rectum were harvested. Portions of each tissue type were embedded in paraffin for histopathologic analysis and in situ hybridization. Single-cell suspensions prepared from LN, spleen, and thymus tissues were subjected to Ficoll-Hypaque gradient centrifugation. Mononuclear cells isolated from the blood, LN, spleen, and thymus were used for flow cytometry, proliferation assays, and virus isolation.

In situ hybridization for viral RNA. In situ hybridization used a pool of digoxigenin-labeled RNA probes generated by Sp6 or T7 polymerase transcription from the entire genome of SIV_mac²³⁹ and HIV-1_bh10 (Lofstrand, Gaithersburg, Md.). The method was described previously in detail (13). Formalin-fixed, paraffin-embedded tissue sections (4- to 5-µm thickness) were placed on diethylpyrocarbonate water-N-tris(hydroxymethyl)methyl-2-aminoethanesulfonic acid-coated glass slides. They were dried overnight and treated as previously described (13). Slides were prehybridized in buffer (50% formamide, 4× SSC [1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate], 1× Denhardt's solution, 4 mM NaPO₄, 0.1% sodium dodecyl sulfate, 5% dextran sulfate, 250 µg of tRNA per ml, 250 µg of salmon sperm DNA per ml in diethylpyrocarbonate water) in a preheated humidity chamber for 15 min. Slides were hybridized with 1.75 ng of riboprobes per ml at 52°C overnight, washed in 2× SSC-50% formamide solution and then in 2× SSC and incubated in an RNase solution (RNase T₁ and RNase A in 2× SSC) for 30 min at 37°C. The slides were blocked with a buffer containing 2% horse serum, 2% sheep serum, 150 mM NaCl, 100 mM Tris (pH 7.4), and 12 mg of levamisole per ml for 1 h. Slides were incubated for 1 h with sheep anti-digoxigenin-alkaline phosphatase conjugate (Boehringer Mannheim, Indianapolis, Ind.) at a 1:500 dilution, rinsed in 0.1 M Tris buffer (pH 7.4) and then in 0.1 M Tris buffer (pH 9.5), and incubated overnight at room temperature with nitroblue tetrazolium-5-bromo-4-chloro-3-indolylphosphate (Vector, Burlingame, Calif.) substrate in the dark. The stained specimens were rinsed in water, counterstained with nuclear fast red, dehydrated, and overlaid with glass coverslips. Controls included sense probes hybridized on SHIV-infected tissues and antisense probes with uninfected tissues.

Flow cytometry analysis of blood and tissue lymphocyte subsets. Mononuclear cells (2 × 10⁵) were stained for 30 min at 4°C with the fluorescein isothiocyanate (FITC)-conjugated monoclonal antibody (MAb) against CD2 (Antigenix, Franklin Square, N.Y.) and either MAb CD4-phycoerythrin (PE) or CD8-PE (Antigenix) or CD20-PE (Becton Dickinson, Mountain View, Calif.). To assess activation marker expression on lymphocyte subsets, mononuclear cells were stained with MAb CD4-PE, CD8-PE, or CD20-PE and MAb CD25-FITC (Endogen, Woburn, Mass.), HLA-DR-FITC, or CD69-FITC (Becton Dickinson). Samples stained with the appropriate isotype controls and compensation controls were included. Cells were washed and fixed with 1% paraformaldehyde in phosphate-buffered saline (PBS). Ten thousand events per sample were acquired by a FACSCalibur flow cytometer, and data were analyzed by Cell Quest software (Becton Dickinson).

Proliferation assays. Mononuclear cells (10⁵) from the blood, LN, or spleen were plated in triplicate in flat-bottom 96-well tissue culture plates in complete RPMI medium with or without 0.5 µg of PHA per ml or 5 µg of p27 per ml. The recombinant p27 was prepared as described previously (30). Plates were incubated in a humidified atmosphere of 5% CO₂ in air at 37°C. Samples containing PHA were pulsed with 1 µCi of tritiated thymidine per well on day 3 and harvested on day 4. Samples including p27 were pulsed with 1 µCi of tritiated thymidine per well on day 6 and harvested on day 7. Samples were harvested onto glass fiber filters and allowed to dry. Tritiated-thymidine incorporation was measured by liquid scintillation counting. Mean numbers of counts per minute of triplicate samples were determined and reported as such or as a stimulation index (SI). The SI was determined by dividing the mean number of counts per minute incorporated in the presence of PHA or p27 with the mean number of counts per minute incorporated in the presence of medium alone. Replicate plates without addition of tritiated thymidine were set up for live- and dead-cell counts based on eosin dye exclusion.

Apoptosis assay. The 7-amino actinomycin D (7-AAD) assay was used to determine the percentage of lymphocytes undergoing apoptosis (28). Mononuclear cells were cultured for 24 h in complete RPMI medium with or without 0.5 µg of PHA per ml. Cells were counted and washed with PBS. Cell samples were stained for surface markers with MAb CD4-PE (Antigenix), CD8-FITC (Coulter-Immunotech, Miami, Fla.), CD20-FITC (Becton Dickinson), or the relevant isotype control. Cells were washed and incubated with 20 µg of 7-AAD (Sigma) per ml in PBS for 20 min at 4°C. Cells were washed twice with PBS and then fixed with 1% paraformaldehyde in PBS containing 20 µg of actinomycin D (Sigma) per ml to block any nonspecific uptake of 7-AAD after fixation. Samples were acquired by a FACSCalibur flow cytometer, and data were analyzed by Flow Jo software (Becton Dickinson).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Virus dissemination. Six rhesus macaques were infected with SHIV_89.6PD and euthanized between 4 and 15 days later. Two animals were infected i.v., two were infected i.r., and two were infected i.vag. The i.v. infected macaques (94074 and 94089) were euthanized on days 4 and 8 after infection. For i.r. and i.vag. infected animals, we euthanized one from each group at day 8 and the second animal from each group at day 15 after inoculation. We collected blood and tissue samples that were used to prepare mononuclear cells or fixed and embedded for in situ hybridization studies.

CD4⁺ T-cell depletion and T-cell response to mitogen. At 15 days after mucosal infection (i.r. or i.vag.), we observed substantial depletion of CD4⁺ T cells from the peripheral blood, spleen, and LN (Table 2). We reported similar observations of 90% or greater CD4⁺ T-cell loss from peripheral blood of rapidly progressing SHIV_89.6PD i.r. inoculated macaques within 2 weeks after infection (29), and we reported previously (24) that uninfected macaques (n = 5) showed approximately 40% CD4⁺ T cells in the mononuclear cells from inguinal LN biopsies. The short-term serial sacrifice studies presented here demonstrated that CD4⁺ T-cell loss after infection was substantial in tissues. In 8 day i.v. infected macaque 94089, CD4⁺ T-cell depletion was extensive in the spleen and moderate in the LN. Although the percentage of 94089 CD4⁺ T cells in blood had not changed substantially compared to preinfection values, the absolute number of CD4⁺ T cells per microliter of blood fell dramatically by 8 days after i.v. infection (Table 2). Lymphocytes from the 15-day mucosally infected animals were predominantly CD8⁺ CD2⁺ T cells or CD20⁺ B cells (Table 2). These two subsets together represent the majority of lymphocytes from blood or tissue in these animals. Interestingly, the percentage of CD4⁺ CD8⁺ thymocytes remained high in the infected macaques, despite the high viral load in the thymus by day 15 in mucosally infected animals. Thus, the acute loss was restricted mainly to mature CD4⁺ T cells.

Lymphocyte activation markers. We chose to explore the possibility that AICD was responsible for loss of in vitro proliferative responses. Initially, we examined lymphocyte surface markers for evidence of polyclonal activation in vivo. We began this study by infecting three additional macaques, AQ93, AQ80, and AR08, by the i.r. route with SHIV_89.6PD. A fourth animal, AQ73, received the same inoculum but was negative for infection by all laboratory tests and did not even show transient viremia. By establishing a new infected-animal cohort, we were able to obtain fresh PBMC samples that improved our ability to study activation markers and cellular apoptosis. Infected macaques had detectable p27 antigenemia at 2 weeks after inoculation, and virus could be isolated from as few as 1,000 PBMC. The number of infected cells in the PBMC decreased to 1 in 10,000 or 1 in 100,000 by 4 weeks after infection. These animals experienced significant losses of peripheral blood CD4⁺ T cells between 2 and 4 weeks after infection (Fig. 1A). AQ80 and AQ93 produced virus-specific antibodies by 4 weeks after infection, and AR08 seroconverted at 8 weeks after infection. All animals developed moderate to high levels of virus-binding antibodies (29) that were assayed at a serum dilution of 1:500 (A₄₀₅, >0.2) by using a commercial ELISA kit (Sanofi/Pasteur, Chaska, Minn.).

View larger version (29K): [in this window] [in a new window]   FIG. 1.   CD4+ T-cell numbers and activation marker expression after SHIV89.6PD infection. Rhesus macaques AQ73, AQ80, AQ93, and AR08 were inoculated i.r. with SHIV89.6PD. All except AQ73 were confirmed to be infected on the basis of at least two positive virus isolations, seroconversion, and PCR assay for viral gag sequences as described previously (33). Blood was drawn on weeks 0, 1, 2, 4, 8, and 13 after inoculation. The absolute number of CD4+ T cells per microliter of blood (A) was determined from the percentage of lymphocytes double positive for CD2 and CD4 and the total number of lymphocytes per microliter of blood as determined by CBC. The percentage of CD4+ T cells positive for HLA-DR (B), CD25 (C), or CD69 (D) was determined by flow cytometry.

View larger version (29K): [in this window] [in a new window]   FIG. 2.   CD8+ T-cell number and activation marker expression after SHIV89.6PD infection. Rhesus macaques AQ73, AQ80, AQ93, and AR08 were inoculated with SHIV89.6PD i.r. All except AQ73 were infected. Blood was drawn on weeks 0, 1, 2, 4, 8, and 13 after inoculation. The absolute number of CD8+ T cells per microliter of blood (A) was determined from the percentage of lymphocytes double positive for CD2 and CD8 and the total number of lymphocytes per microliter of blood as determined by CBC. The percentage of CD8+ T cells positive for HLA-DR (B), CD25 (C), or CD69 (D) was determined by flow cytometry.

Apoptosis of lymphocytes from SHIV_89.6PD-infected macaques. The lack of response to PHA was associated with cell death in vitro. We next wanted to determine whether apoptosis was occurring in these stimulated lymphocyte cultures, and we employed a flow cytometry method utilizing 7-AAD staining in conjunction with cell surface staining for lymphocyte differentiation markers (28). This method gave results similar to those of annexin-V assays and was preferred for obtaining data that could be combined with cell surface marker phenotype studies (15). Moreover, 7-AAD staining easily allowed the identification and exclusion of debris or apoptotic bodies. Cells staining dimly with 7-AAD are entering apoptosis, whereas those which stain brightly are late in apoptosis or already dead. PBMC proliferation in response to PHA was diminished substantially at 2 weeks after infection for AQ80 and AQ93 and at 4 weeks after infection for AR08. SIs for uninfected animal AQ73 remained above 150 at all time points (Fig. 3). We utilized the 7-AAD assay to determine the percentage of cells undergoing apoptosis after 24 h in culture with PHA (Table 6 and Fig. 4). By gating on dimly 7-AAD-staining lymphocytes, we were able to determine the percentage of apoptotic cells. Substantial increases in the percentages of lymphocytes undergoing apoptosis after PHA stimulation were observed in blood from infected animals AQ93 and AR08 by 2 weeks after infection (Table 6). Sufficient PBMC were not available from macaque AQ80 to perform the 7-AAD assay at week 2. However, the percentage of PBMC undergoing apoptosis was increased in this infected animal by 4 weeks after SHIV infection compared to the earlier time points (Table 6). Apoptotic cells from AQ93 and AR08 at 2 weeks were predominantly CD4⁺ T cells. CD8⁺ T cells and CD20⁺ B cells together accounted for less than 8% of the apoptotic cells (Fig. 4). It is possible that some CD8⁺ T cells and CD20⁺ B cells made up a proportion of the late-apoptotic or dead-cell population at the time of assay. However, as dead cells often bind surface antibodies nonspecifically, we chose not to analyze the brightly 7-AAD-staining population. The percentage of apoptotic lymphocytes after PHA stimulation remained between 4 and 6% for all blood draws from uninfected macaque AQ73 (Table 6 and Fig. 4). Additional data for the percentage of apoptotic lymphocytes after PHA stimulation from virus-naive macaques were similarly low (Table 6). Consistent with these results, we observed higher numbers of eosin-staining dead cells from cultures with high levels of brightly 7-AAD-staining populations. Overnight culture of PBMC in medium alone did not induce apoptosis in lymphocytes from either infected or uninfected macaques (data not shown).

View larger version (18K): [in this window] [in a new window]   FIG. 3.   Proliferation of rhesus macaque lymphocytes in response to PHA. Rhesus macaques AQ73, AQ80, AQ93, and AR08 were inoculated with SHIV89.6PD i.r. All except AQ73 were infected. PBMC isolated from the animals on weeks 0, 1, 2, 4, 8, and 13 after inoculation were tested for the ability to proliferate in response to PHA or medium alone. Data are reported as SIs.

View larger version (35K): [in this window] [in a new window]   FIG. 4.   Apoptosis of CD4+ T cells from macaques. Data are results from one uninfected macaque, AQ73, and two SHIV89.6PD-infected macaques, AQ93 and AR08. PBMC isolated from the animals 2 weeks after inoculation were stimulated by incubation with PHA in complete RPMI medium for 24 h. Cells were counted, and samples were stained with MAbs against surface markers (CD4-PE, CD8-FITC, or CD20-FITC) or appropriate isotype control MAbs. Cells were then stained with 7-AAD as described in Materials and Methods. Data were analyzed by the Flow Jo analysis program. A lymphocyte gate was first set by using forward-side scatter (FSC) on the x axis and side scatter on the y axis. From the defined lymphocyte gate, dot plots with FSC on the x axis and 7-AAD brightness on the y axis are shown. Lymphocytes reacting dimly with 7-AAD in the central region of these plots are undergoing apoptosis. Cells in this apoptotic gate were analyzed for surface marker expression as shown by the histograms.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

We examined events during acute SHIV_89.6PD infection of macaques to characterize changes in host-virus interactions that might dictate disease progression rates. These studies tested the hypothesis that a wave of lymphocyte activation occurs very soon after primary infection and serves to accelerate disease progression by increasing direct (virus infection) and indirect (apoptosis) CD4⁺ T-cell depletion and limits the magnitude of the subsequent immune responses to virus.

Approximately 2 weeks elapsed between the time of mucosal inoculation and the onset of abundant virus replication in secondary lymphoid tissues. The early spread of virus was accompanied by polyclonal cell activation that was observed as increased expression of specific cell surface markers and increased susceptibility to AICD upon exposure to mitogen in vitro. During the first week after infection, increased expression of early T-cell activation markers on CD4⁺ T cells was observed in the SHIV_89.6PD-infected macaques. Overall levels of CD4⁺ T-cell activation marker expression were maximal between weeks 2 and 4 after infection. By week 2 after mucosal infection, systemic virus replication reached high levels which correlated with the period of rapid CD4⁺ T-cell loss from the blood and tissue compartments. During this early interval, remaining CD4⁺ T cells were highly susceptible to AICD when they were stimulated in vitro with polyclonal mitogen. However, the extent of activation and subsequent AICD greatly exceeded the frequency of infected CD4⁺ T cells. The wave of polyclonal lymphocyte activation likely contributed to the high levels of SHIV_89.6PD replication. Accelerated growth of virus within the population of activated lymphocytes furthers cell depletion and establishes a condition favorable to persistent infection with progressing disease. Declining virus replication was coincident with the onset of host immune responses, and the interval of acute infection ended by 8 weeks after infection.

During the crucial first week after exposure, very low levels of virus initiated widespread changes in uninfected CD4⁺ T cells and the cells appeared to be at risk for AICD and virus infection. These events are amplified and accelerated in the case of a highly virulent virus such as SHIV_89.6PD, and this model provides a convenient system for evaluation of their contributions to pathogenesis. We have used SHIV_89.6PD infection because it exaggerates the events of acute infection. Even though this model may only represent the most aggressive HIV infections, we hope to uncover general mechanisms that can be tested in a broad range of infection examples.

T-cell unresponsiveness to mitogenic stimulation was associated with cell death and preceded substantial CD4⁺ T-cell losses in vivo. These observations suggest strongly that early immune dysfunction among CD4⁺ T cells is due to inappropriately high activation of lymphocytes during the first weeks of infection. Moreover, the kinetics of lymphocyte activation indicate the action of extracellular factors originating from the virus, the host, or both. Viral proteins, including the envelope protein and Tat, are known to induce apoptosis of uninfected CD4⁺ T cells (1, 8, 16, 34). It is also possible that specific cytokines induced early in infection, such as gamma interferon and tumor necrosis factor alpha (10, 14), promote apoptosis.

There are numerous examples of susceptibility to AICD during asymptomatic infection of HIV in humans (11, 19) or simian immunodeficiency virus in macaques (7, 9). Among HIV-1-infected persons, the extent of apoptosis in LN lymphocytes was associated directly with cell activation in these tissues (20). In a previous study, we observed loss of proliferative responses to p27 and PHA as early as 1 week after i.r. SHIV_89.6PD infection in p27-immunized rhesus macaques; depressed mitogen responses were transient and returned to normal levels by 8 weeks after infection, while proliferative responses to Gag antigen did not recover (31). Similar observations of transient lymphocyte unresponsiveness were reported in HIV-1⁺ persons during acute retroviral syndrome (5, 22).

Our analysis of acute infection seeks to identify the earliest events in viral pathogenesis and to assess their impact on subsequent disease progression. The present studies show that lymphocyte activation, triggered as a nearly immediate consequence of virus infection, begins the process of CD4⁺ T-cell depletion and renders the remaining lymphocytes more permissive for virus replication. The extent of acute CD4⁺ T-cell depletion likely limits the remaining immune repertoire and establishes a host set point, that we define as a measure of immune competence for controlling virus replication and modulating subsequent disease progression. The foundation of immune competence is the breadth and strength of an intact, virus-specific, CD4⁺ T-cell repertoire. CD4⁺ T cells provide crucial cytokine help for antiviral cytotoxic T lymphocytes and for the generation of T-cell-dependent virus-neutralizing antibodies that are associated with long-term nonprogression (3, 24). Lesions in the CD4⁺ T-cell repertoire are not repaired rapidly by antiviral therapy, despite increased CD4⁺ T-cell numbers (4). Strategies for managing acute infection should include efforts to control virus replication but may also need to account for lymphocyte activation and AICD via apoptosis that also promote destruction of the helper repertoire.

The host set point describes the capacity for immune control of disease progression. One part of this control is the ability to modulate virus replication and to establish a viral set point that provides useful quantitative values for prognosis and evaluation of therapy (17). However, to better understand AIDS pathogenesis, we need to direct our investigations closer to the time of initial exposure to uncover the host mechanisms affecting and being affected by early events in virus infection. Studies with the macaque model show that the earliest events of acute infection remodel the CD4⁺ T-cell repertoire, establish the host set point, and dictate the subsequent levels of virus replication and rates of disease progression to AIDS.