Effect of Humoral Immune Responses on Controlling Viremia during Primary Infection of Rhesus Monkeys with Simian Immunodeficiency Virus

Abstract

Cellular immune responses mediated by CD8⁺ lymphocytes exert efficient control of virus replication during primary simian immunodeficiency virus (SIV) infection. However, the role that antibodies may play in the early control of virus replication remains unclear. To evaluate how antibody responses may affect virus replication during primary SIVmac infection, we depleted rhesus monkeys of B cells with anti-CD20 antibody. In normal rhesus monkeys immunized with tetanus toxoid, anti-CD20 treatment and resulting depletion of B cells inhibited the generation of antitetanus antibodies, while tetanus-specific T-cell responses were preserved. During the first 4 weeks after inoculation with SIVmac251, development of SIV-specific neutralizing antibody was delayed, and titers were significantly lower in B-cell-depleted monkeys than control-antibody-treated monkeys. Despite the lower neutralizing antibody titers, the levels of plasma SIV RNA and the linear slope of the decline seen in B-cell-depleted monkeys did not differ from that observed in monkeys treated with control antibody. However, beginning at day 28 after SIV infection, the B-cell-depleted monkeys showed a significant inverse correlation between neutralizing antibody titers and plasma virus level. These results suggest that the rapid decline of peak viremia that typically occurs during the first 3 weeks of infection was not significantly affected by SIV-specific antibodies. However, the inverse correlation between neutralizing antibodies and plasma virus level during the postacute phases of infection suggests that humoral immune responses may contribute to the control of SIV replication.

Clinical observations in human immunodeficiency virus (HIV)-infected humans and experimental studies in simian immunodeficiency virus (SIV)-infected nonhuman primates have shown that the early control of primary viremia has the potential to affect clinical outcome. Following experimental SIV inoculation, some macaques naturally limit the level of primary viremia and exhibit lower setpoint levels of plasma virus. These macaques have a slower disease course than those that fail to control primary viremia and exhibit higher plasma virus setpoint levels (15). Furthermore, prior immunization (2) or administration of antiretroviral therapy during primary infection (16, 31) lowers levels of primary viremia and improves clinical outcome. Thus, understanding which immune mechanisms can contribute to controlling early viremia will be important in understanding AIDS pathogenesis and designing interventional strategies.

The control or clearance of many viral infections follows the emergence of both cellular and humoral immune responses, suggesting that both components may contribute functionally to this process. Likewise, infection with HIV and the other primate lentiviruses results in the induction of both virus-specific antibodies and T cells (13). Clinical and experimental data have conclusively shown the importance of cellular immune responses mediated by CD8⁺ lymphocytes in controlling early replication of primate lentiviruses (8, 27). However, the potential for humoral immune responses to affect early HIV replication remains uncertain. HIV-specific antibodies, either alone or in conjunction with other components of the immune system, can act beneficially to limit virus replication (19, 22). However, virus-specific antibodies can also have undesirable effects by promoting virus accumulation and survival in lymphoid germinal centers (9, 26, 29). Furthermore, declining HIV-specific antibody titers have the potential to enhance infectivity (32).

Strong and broadly cross-reactive neutralizing antibodies do arise in AIDS virus infections but appear later than cellular immune responses and fail to reach titers observed in other viral infections. This inability to generate more effective antibody responses may be due to viral cytopathicity directed against CD4⁺ T cells required for normal antigen recognition and B-cell response (3, 5). Pathological changes in lymph nodes following HIV infection ultimately result in germinal center destruction (30). Furthermore, antigenic variation and dense carbohydrate masking of neutralizing determinants on envelope glycoproteins may also hinder an effective humoral response (6). Nevertheless, observations in the nonhuman primate AIDS models illustrate the potential for antibody-mediated responses to contribute to protection.

SIVmac-infected rhesus macaques that undergo rapid disease progression fail to develop significant antiviral antibody titers (10, 14). Furthermore, numerous studies have shown that passive administration of antiviral antibodies can blunt primary viremia or completely block infection after experimental challenge, demonstrating the potentially beneficial effect of humoral immunity (1, 4, 7, 17, 18, 21-23). In the natural course of SIV infection, low-titer virus-specific antibodies are present at the time that primary viremia clears. Thus, it would be useful to understand their potential to contribute to early control of replication.

In the present study, we show that B-cell depletion at the time of inoculation with SIVmac251 delayed virus-specific humoral immunity for 2 weeks. Unlike persistent CD8⁺ lymphocyte depletion, which resulted in uncontrolled primary SIV viremia (27), the early control of high-level primary viremia was not significantly affected by the delay in SIV-specific antibodies. However, from postinoculation day 28 forward, neutralizing antibody titers were inversely correlated with levels of plasma virus, indicating that antibodies may contribute to the control of SIV replication.

MATERIALS AND METHODS

Animals, treatments, and virus inoculations.

The depletion of B cells from rhesus monkeys (Macaca mulatta) was accomplished by administering three weekly (days 0, 7, and 14) intravenous injections (20 mg/kg of body weight) of monoclonal mouse-human chimeric anti-human CD20 antibody (Rituxan; IDEC Pharmaceuticals Corp., San Diego, Calif.). As a control, monkeys received an equivalent amount of mouse-human chimeric monoclonal antibody against respiratory syncytial virus (kindly made available by MedImmune, Inc., Gaithersburg, Md.). The age of the monkeys used in these experiments ranged from 4 to 7 years.

In experiments to determine the effect of B-cell depletion on generation of cellular and humoral immune responses to tetanus toxoid, normal rhesus monkeys received a single intramuscular injection (0.5 ml) of tetanus toxoid (Pasteur Mérieux Connaught, Swiftwater, Pa.) at the time of the first monoclonal antibody injection (day 0). Blood specimens were then obtained weekly through day 35 for measuring immune responses as described below.

In experiments designed to determine the effect of delaying SIV-specific antibody generation on the control of viremia, rhesus monkeys received anti-CD20 or control antibody treatments as described above. In a first experiment, two monkeys were inoculated on day 14 following initiation of anti-CD20 treatment with SIVmac strain 251 (equivalent to 3 × 10⁶ copies of SIV RNA). Since the length of the B-cell depletion was quite heterogeneous (3 weeks and 3 months after SIV infection), we chose to inoculate another four monkeys on day 7 after initiation of the anti-CD20 treatment to ensure depletion of B cells during primary SIV infection. Animals were anesthetized with ketamine HCl for antibody administration, virus inoculation, and specimen collection. All animals were maintained in accordance with the guidelines of the Committee on Animals for the Harvard Medical School and the Guide for the Care and Use of Laboratory Animals (National Research Council, National Academic Press, Washington, D.C., 1996).

Antitetanus antibody and proliferation assays.

Tetanus-specific antibodies were measured in plasma of immunized monkeys with an enzyme-linked immunosorbent assay similar to that described previously (25). Briefly, 96-well microtiter plates were coated overnight at 4°C with tetanus toxoid (tetanus toxoid, fluid, purified; Wyeth Laboratories, Marietta, Pa.) diluted 1:1,000 in phosphate-buffered saline, and the plates were then blocked with phosphate-buffered saline-2% nonfat dry milk. Plasma specimens (100 μl) were serially diluted and added to the plates, incubated for 1 h at room temperature, and then washed. Binding of monkey antibody to the plate was detected with horseradish peroxidase-conjugated goat anti-human immunoglobulin G (Jackson ImmunoResearch, West Grove, Pa.). After washing, 100 μl of tetramethylbenzidine (Kirkegard-Perry) was added to each well, the reaction was stopped after 20 min with H₂SO₄, and absorbance was read on at 490 nm. Endpoint plasma dilution was determined as the highest dilution to exceed the mean optical density + 2 standard deviations of the preimmunization specimen.

Cellular proliferative responses to tetanus antigen were quantified by culturing monkey peripheral blood mononuclear cells obtained by density gradient centrifugation in RPMI 1640 supplemented with l-glutamine, penicillin, streptomycin, and 5% autologous preimmunization serum. Triplicate wells of 10⁵ cells were incubated with tetanus antigen (tetanus toxoid, plain; Pasteur Mérieux Connaught; final dilution, 1:1,000) at 37°C in 5% CO₂ for 6 days. During the last 6 h, cells were pulse labeled with [³H]thymidine, and incorporation into DNA was determined by routine harvesting of cells and liquid scintillography. The stimulation index was determined by dividing the median counts per minute from the tetanus toxoid-supplemented cultures by the median counts per minute from the control cultures.

Immunophenotyping and tetramer staining of peripheral blood lymphocytes.

Lymphocytes in peripheral blood were immunophenotyped with EDTA-anticoagulated blood specimens in a whole-blood lysis technique. Lymph node lymphocyte suspensions were obtained by gently teasing peripheral lymph node specimens that had been obtained by excisional biopsy. Cells in the lymph node cell suspension were adjusted to 10⁶/ml in RPMI 1640-10% fetal bovine serum. Fluorochrome-conjugated antibodies were incubated with 100 μl of whole blood or lymph node cell suspension for 20 min at room temperature. Antibodies used were anti-CD3(FN18) conjugated to allophycocyanin, anti-CD4 conjugated to phycoerythrin (19Thy5D7), anti-CD8 conjugated to fluorescein isothiocyanate (7PT-3F9), anti-CD8αβ conjugated to phycoerythrin-Texas Red (2ST8.5H7; Beckman Coulter), anti-CD19 conjugated to phycoerythrin (J4.119; Beckman Coulter), and anti-CD20 conjugated to phycoerythrin-Texas Red (B1; Beckman Coulter).

Specimens from monkeys expressing the major histocompatibility complex class I allele Mamu-A∗01 were further analyzed for binding of Mamu-A∗01/SIV Gag pllC tetrameric complexes as described previously (11). Erythrocytes were lysed with an ImmunoPrep reagent system and a TQ-Prep workstation (Beckman Coulter). To reduce the background level of staining, lysed samples were washed with 1 ml of phosphate-buffered saline, centrifuged for 3 min at 300 × g, and fixed in phosphate-buffered saline-1% formalin. Specimens were routinely analyzed for four-color immunofluorescence with a manually determined light-scatter gate for gating of lymphocytes. Absolute lymphocyte counts on blood specimens were obtained with a T540 hematology analyzer (Beckman Coulter).

Immunohistochemistry.

Lymph node biopsy specimens were cut into two portions. One portion was snap-frozen and kept at −70°C until immunohistochemistry was performed. The second portion was used for preparation of single cells for flow cytometric evaluation of lymphocyte subsets as described above.

Immunohistochemical identification of lymphocytes in lymph nodes was performed on snap-frozen biopsy specimens that were sectioned (6 μm) on a cryostat and fixed in 2% paraformaldehyde at room temperature for 15 min. Sections were rinsed in phosphate-buffered saline and incubated with monoclonal antibodies to CD20 (1:50; clone L26), Ki67 antigen (1:100; clone MIB-1), or follicular dendritic cells (1:200; clone R4/23) (all from Dakopatts, Copenhagen, Denmark). To detect CD4⁺ or CD8⁺ T cells, the sections were incubated either with a cocktail of anti-CD4 antibodies (Leu3a and M-T477 from BD-PharMingen, San Diego, Calif.) or with Leu2a (1:70; BD-PharMingen). Binding of primary antibodies was visualized by the alkaline phosphatase-anti-alkaline phosphatase technique with New Fuchsin as the chromogen. Sections were counterstained with hematoxylin and mounted.

Plasma viral RNA levels.

Virion-associated SIV RNA was measured in plasma samples with a real-time reverse transcription-PCR assay with a threshold sensitivity of 100 copy equivalents/ml and intra-assay coefficient of variation of <25% (16).

Neutralizing antibody titers to SIVmac251.

Antibody-mediated neutralization of a T-cell-line-adapted stock of SIVmac251 was assessed in a CEMx174 cell-killing assay as previously described (20). Briefly, 50 μl of cell-free virus containing 500 50% tissue culture infectious doses was added to multiple dilutions of test serum in 100 μl of growth medium (RPMI 1640 with 12% fetal calf serum and 50 μg of gentamicin) in triplicate in 96-well culture plates. The mixtures were incubated for 1 h at 37°C, followed by the addition of CEMx174 cells (5 × 10⁴ cells in 100 μl) to each well. Infection led to extensive syncytium formation and virus-induced cell killing in approximately 4 to 6 days in the absence of antibodies. Neutralization was measured by staining viable cells with Finter's neutral red in poly-l-lysine-coated plates. Percent protection was determined by calculating the difference in absorption (A₅₄₀) between test wells (cells plus serum sample plus virus) and virus control wells (cells plus virus), dividing this result by the difference in absorption between cell control wells (cells only) and virus control wells and multiplying by 100.

Neutralization was measured at a time when virus-induced cell killing in virus control wells was greater than 70% but less than 100%. Neutralizing titers are given as the reciprocal dilution required to protect 50% of cells from virus-induced killing. The cell-free stock of SIVmac251 was prepared in H9 cells.

Statistical analysis and data presentation.

Assay results at a particular follow-up time (e.g., neutralizing antibody titer or plasma SIV RNA) were compared between controls and CD20-depleted animals with the exact Wilcoxon rank sum test. For each animal, the slope over time of plasma SIV RNA (and log₁₀ plasma SIV RNA) was computed by ordinary least squares, and the resultant slopes were compared between controls and CD20-depleted animals with the exact Wilcoxon rank sum test. All significance levels were two-sided. The test of correlation of neutralizing antibodies and plasma SIV RNA used the Spearman rank correlation test. Plots were done with Microsoft PowerPoint 2000. When points were plotted on a log scale, the medians were calculated on the log scale.

RESULTS

Administration of monoclonal anti-CD20 antibody to rhesus monkeys depleted of blood and lymph node B cells.

The preclinical evaluation of a mouse-human chimeric anti-CD20 antibody indicated that this antibody efficiently depleted B cells from the blood and lymphoid tissues of macaque monkeys (24). We wished to further characterize the cell depletion and immunologic consequences that occurred following administration to rhesus monkeys. Two normal rhesus monkeys received 20-mg/kg intravenous injections of anti-CD20 antibody on days 0, 7, and 14. As controls, two monkeys received the same treatments with a monoclonal antibody with an irrelevant specificity. Following anti-CD20 treatment, B cells disappeared from the circulation when peripheral blood mononuclear cells were examined by flow cytometry with either anti-CD20 or anti-CD19 antibodies. Depletion from blood persisted for 4 weeks in one monkey and for greater than 8 weeks in the second monkey (Fig. 1). Similarly, profound B-cell depletion in peripheral lymph nodes also occurred. We confirmed lymph node depletion of B cells immunohistochemically.

FIG. 1.

Administration of anti-CD20 antibody depleted rhesus macaques of peripheral blood B cells. Two normal rhesus macaques received anti-CD20 (20 mg/kg of body weight) or an isotype-matched control antibody on days 0, 7, and 14. Control antibody had no effect on the number of CD20⁺ B cells in peripheral blood (A). However, administration of anti-CD20 antibody resulted in rapid loss of CD20⁺ B cells (B).

Figure 2 shows an absence of CD20⁺ lymphocytes in the follicles. Follicular dendritic cells showed a positive reaction with antibody to CD20, consistent with follicular dendritic cells trapping of the injected chimeric antibody. Immunostaining with monoclonal antibody R4/23 against follicular dendritic cells revealed that the network of these antigen-trapping cells was not damaged by the treatment. The usual interwoven concentric pattern of the follicular dendritic cell network was preserved (data not shown). Nevertheless, the reduced proliferation of B lymphocytes was identified by diminished staining of germinal centers with anti-Ki67, indicating a disruption of germinal center function. As expected, immunophenotypic analysis of the extrafollicular regions, including the T-cell-dependent zone of the lymph nodes, showed no changes in the number and distribution of CD4⁺ or CD8⁺ T cells (data not shown). B-cell numbers in both peripheral blood and lymph node of control-antibody-treated monkeys were unaffected (Fig. 1 and 2).

FIG. 2.

Anti-CD20 antibody depleted B cells from lymph nodes. Prior to anti-CD20 treatment, morphologically normal germinal centers (arrows) contained a normal distribution of B cells (A) and abundant anti-Ki67 staining cells in the germinal center dark zone, indicating many proliferating cells (B). At 14 and 28 days after anti-CD20 treatment, a marked reduction in the number of CD20⁺ B cells is evident; the follicular dendritic cell network is stained with this antibody, indicating that a proportion of injected chimeric anti-CD20 antibody is captured by these cells (C and E). Parallel with this, the decrease in the number of Ki67⁺ proliferating B cells within the germinal center and fewer Ki67⁺ proliferating cells were evident (D and F).

B-cell-depleted monkeys have impaired ability to generate humoral immune responses but retain ability to generate proliferative cellular responses.

To ascertain the effect of B-cell depletion on the generation of antibodies to new antigens, the four monkeys were immunized with tetanus toxoid after the first antibody injection. By 2 weeks after immunization, both control-antibody-treated monkeys developed high-titer antitetanus antibody responses (range, 1:4,096 to 1:16,384; Fig. 3A). However, in the two B-cell-depleted monkeys, antitetanus antibody titers did not exceed 1:100 through 5 weeks postimmunization.

FIG. 3.

Anti-CD20-mediated B-cell depletion inhibited humoral but not cellular immune responses. (A) Plasma titers of anti-tetanus toxoid antibody were substantially lower in anti-CD20-treated rhesus macaques than antibody-treated monkeys when immunized during antibody treatment. (B) Control antibody- and anti-CD20 antibody-treated rhesus macaques generated equivalent tetanus toxoid-specific cell proliferative responses in vitro.

To confirm that cellular immune responses were not affected by B-cell depletion, we measured proliferative responses of peripheral blood mononuclear cells to tetanus antigen. Both control-antibody-treated and anti-CD20-treated monkeys generated equivalent proliferative responses to tetanus (Fig. 3B). Thus, the cellular immune component appeared to remain intact following B-cell depletion.

B-cell depletion delayed appearance of SIVmac-specific antibody responses but did not affect clearance of primary viremia.

We used this model of B-cell depletion to assess whether anti-SIV antibodies might play a contributory role in reducing primary viremia. Six rhesus monkeys received three weekly injections of anti-CD20 or control antibody with the regimen described above. All were inoculated with SIVmac251 on day 7 or day 14 of treatment. B cells were depleted from the blood for 1 to 3 months (median, 6 weeks) in anti-CD20-treated monkeys (data not shown), resulting in a reduction of B cells during the first 4 weeks of SIV infection. The appearance of neutralizing antibodies against SIVmac251 occurred significantly later in the CD20-depleted animals (P = 0.03) Furthermore, the median anti-SIV neutralizing titer was at least one log lower in B-cell-depleted monkeys compared to controls at both 21 days (P = 0.02) and 28 days (P = 0.04) (Fig. 4A and B).

FIG. 4.

B-cell depletion resulted in a delay in appearance of SIV-specific neutralizing antibodies but had no effect on control of primary SIVmac251 viremia. Prior to inoculation with SIVmac251, rhesus macaques were depleted of B cells by receiving three anti-CD20 antibody treatments (20 mg/kg of body weight) or received an equivalent amount of a control antibody. The appearance of neutralizing antibodies was delayed and the median titer was significantly lower at day 28 in B-cell-depleted macaques (A) than in control-antibody-treated macaques (B). Despite the inhibition in neutralizing antibodies, the control of primary viremia (plasma virus level at days 10 to 28, linear slope of plasma virus decline between peak and day 28) was no different in B-cell-depleted macaques (C) than in control-antibody-treated macaques (D).

To determine the role of SIV-specific antibodies in clearing primary viremia, we measured plasma virus in B-cell-depleted and control antibody-treated monkeys. Despite the delay in neutralizing antibodies and lower overall titer in B-cell-depleted monkeys, there was no difference in the slope of the decline in plasma virus (Fig. 4C and D). A comparison of the viral load slopes between the control-antibody-treated animals and the anti-CD20 antibody-treated animals did not show a statistically significant difference (Wilcoxon rank sum test, P = 0.49). Similarly, we were unable to detect statistically significant differences in median plasma levels of SIV between anti-CD20 and control-antibody-treated groups (days 10 to 11, P = 0.64; day 14, P = 0.37; day 21, P = 0.64; and day 28, P = 0.96).

Higher neutralizing antibody titers correlated with lower levels of plasma virus during the postacute phase of infection.

Neutralizing antibody titers were quite similar among the control-antibody-treated monkeys by day 28 (Fig. 4A). However, the anti-CD20-treated monkeys showed more heterogenous titers of neutralizing antibodies (Fig. 4B). We therefore determined whether neutralizing antibody titers correlated with the level of plasma virus among the B-cell-depleted monkeys. As shown in Fig. 5 and Table 1, neutralizing antibody titers and the level of plasma virus did show a significant inverse correlation on postinoculation days 28 and 56. The same correlation was close to significance on day 21. A similar correlation was not evident among the control-antibody-treated group because of a relative lack of variability in neutralizing antibody titers and the smaller group size. Although animals with normal and delayed SIV-specific humoral immune responses showed equivalent abilities to reduce levels of primary viremia; this observation supports a role for neutralizing antibodies during the postacute phase of SIV infection.

FIG. 5.

Relationship between neutralizing antibody titer and plasma virus level among B-cell-depleted and control-antibody-treated monkeys. On postinoculation days 28 and 56, the six B-cell-depleted monkeys showed an inverse correlation between neutralizing antibody titer and plasma virus. Significant correlations were not seen in B-cell-depleted monkeys on day 21 or among control-antibody-treated monkeys at any of these time points.

TABLE 1.

Correlation of virus levels in plasma with neutralizing antibody titers

Animals	Day	Spearman correlation	P
CD20 depleted	21	−0.75	0.08
	28	−0.60	0.04
	56	−0.83	0.04
Controls	21	−0.60	0.40
	28	0.60	0.40
	56	−0.40	0.60

Clearance of primary viremia corresponded with appearance of SIV-specific CD8⁺ T cells.

We next wished to confirm that SIV-specific CD8⁺ T cells were elicited in B-cell-depleted monkeys and that their appearance corresponded with the control of primary viremia. Four monkeys infected with SIV (three B-cell-depleted monkeys and one control-antibody-treated monkey) expressed the rhesus major histocompatibility complex class I allele Mamu-A∗01. Peripheral blood lymphocytes were stained with Mamu-A∗01/SIV Gag p11C tetrameric complexes to quantitate SIV-specific CD8⁺ T cells. In all SIV-infected monkeys depleted of B cells, tetramer-positive cells were first evident on day 11 following inoculation. Tetramer staining peaked at day 14 postinoculation, when 4.5 to 4.9% of CD8⁺ T cells stained positively with tetramer. In the one control-antibody-treated monkey studied, tetramer staining was also evident at day 11 and peaked at day 18 (9.1%) post-SIV inoculation. The kinetics and magnitude of the SIV-specific CD8⁺ T-cell response in B-cell-depleted monkeys were identical to those observed previously in unmodified SIV infection (12). A representative figure demonstrating the rise of the cellular immune response in a control-antibody-treated and a B-cell-depleted monkey is shown in Fig. 6.

FIG. 6.

In B-cell-depleted macaques, control of viremia coincided with appearance of SIV-specific cellular immune responses. The appearance of SIV-specific CD8 T-cell immune responses detected by using Mamu-A∗01/SIV Gag tetramers appeared at the same time as the decline in plasma virus in Mamu-A∗01⁺ rhesus macaques infected with SIV was observed. (A) Control-antibody-treated monkey. (B) Representative anti-CD20 antibody-treated monkey.

DISCUSSION

In previous studies, we modified the immune responses of rhesus monkeys to elucidate the relative contribution made by different components of the immune system to the control of SIV infection. Studies performed in rhesus monkey depleted of hemolytic complement at the time of infection indicated that this component of innate immunity contributed marginally to the decline in early viremia (28). The importance of cell-mediated immunity was demonstrated by showing that monkeys transiently depleted of CD8⁺ lymphocytes showed a delay in control of primary viremia, while monkeys persistently depleted of CD8⁺ lymphocytes maintained peak levels of viremia (27). However, the role that antibodies may play in the early virus clearance is uncertain. In the present study, we used in vivo depletion of B cells to delay antibody generation to evaluate the involvement of humoral immune responses in controlling primary SIV infection.

Both binding and neutralizing antibodies emerge in association with the decline in primary viremia during SIV infection of normal rhesus monkeys. Admittedly, the antibody titer and the breadth of response are quite limited at these early postinoculation time points. However, it remains conceivable that antibodies elicited early may contribute to viral clearance. We demonstrated that depletion of CD20⁺ B cells significantly inhibited antibody responses to tetanus toxoid in normal monkeys and delayed the humoral immune responses in monkeys infected with SIV. This delay in SIV-specific antibody had no measurable effect on the clearance of primary viremia through day 28 when assessed by differences in plasma virus level or the median slope of decline in plasma virus. However, admittedly low-titer antibody responses had developed in some animals by day 21. In contrast, inhibition of cellular immune responses by depletion of CD8⁺ lymphocytes during the same period resulted in persistence of peak levels of virus replication (27). Resolution of primary viremia over the first 4 weeks of infection did correspond with the appearance of SIV Gag-specific CD8⁺ T cells in the three Mamu-A∗0⁺ monkeys embedded in the B-cell-depleted group and in the one control-antibody-treated monkey tested. Thus, we found that primary viremia declined during the first 4 weeks of infection despite a delay in humoral immune responses.

More variability in neutralizing antibody titers occurred in B-cell-depleted monkeys than in the control-antibody-treated group. In fact, antibody responses were never detected in one B-cell-depleted monkey that subsequently showed a disease course typical of a “rapid progressor.” Since this pattern of disease progression occurs in about one-fourth of all SIVmac251 infections, it is unclear whether B-cell depletion played a role in the clinical course of this monkey. Because the B-cell-depleted group exhibited a wider variation in neutralizing antibody titers than the control group, we were able to show an inverse correlation between neutralizing antibody titers and level of plasma virus. This finding supports a contributing role for virus-specific antibody in controlling SIV replication.

It must be noted, however, that the role of antibody in the immunopathophysiology of “rapid progressor” animals remains unclear. The high level of viremia and accelerated disease progression may, indeed, result directly from a failure to develop neutralizing antibody responses. Alternatively, the absence of a significant humoral response may represent a surrogate marker for a rapidly progressing disease course, wherein robust virus replication and acute immunodeficiency prevent the generation of antibody responses. Unfortunately, these studies do not help to elucidate the dynamic relationship between virus replication and humoral immune responses in this unique subset of animals with an accelerated disease course.

Clinicopathologic observations and experimental studies strongly support a role for anti-SIV antibody responses. First, it is well established that neutralizing antibodies in SIV-infected macaques broaden in specificity over time. Macaques that fail to develop this neutralizing antibody response progress rapidly to AIDS. Second, several experimental studies have demonstrated that either immunoglobulin from SIV-infected macaques or anti-SIV-HIV monoclonal antibodies can have potent effects in either blocking or modifying the course of infection (1, 4, 7, 17, 18, 21, 23).

Thus, it would appear that antibodies against the primate lentiviruses can exert control over infection if present in sufficiently broad specificity and high titer. Our studies support mechanisms other than antibodies in the control of primary SIV viremia. Taken together, these studies suggest that early control of SIV replication in the naïve host is accomplished primarily by CD8 effector T-cell-mediated responses. However, these results show the potential for antibodies to act in concert with cellular immune responses to control virus replication.