Innate Immunity in Human Immunodeficiency Virus Infection: Effect of Viremia on Natural Killer Cell Function

Abstract

We examined the effect of viremia on cell contact and soluble factor–mediated suppression of endogenous human immunodeficiency virus (HIV) replication in CD4⁺ T cells from HIV-1–infected individuals by autologous natural killer (NK) and CD8⁺ T cells. NK cells suppressed HIV replication as effectively as did CD8⁺ T cells. Suppression of HIV replication by NK cell culture supernatant was predominantly mediated by CC-chemokine secretion and was considerably greater in patients without viremia than in patients with viremia. Furthermore, there was an inverse correlation between the level of viremia and the ability of NK cells and NK-derived supernatants to suppress virus replication. The ability of NK cells to control HIV replication was independent of levels of interferon-γ expression and cytolytic activity. Our results demonstrate that NK-mediated suppression of HIV replication is as potent as that of CD8⁺ T cells; it is mediated predominantly by secretion of CC-chemokines, and the presence of viremia markedly impairs this NK-mediated inhibitory effect on HIV replication

NK cells are a critical component of host innate immune response to a variety of viruses, fungi, parasites, and bacteria [1, 2]. They were originally identified on a functional basis because of their ability to lyse certain tumor cells without previous stimulation. However, they do not express clonally distributed receptors for antigen, and it has been established recently that HLA class I molecules inhibit cytolytic activity of NK cells by their interaction with a series of NK cell inhibitory receptors. [3, 4]. After activation, NK cells release cytokines and chemokines that induce inflammatory responses, modulate hematopoeisis, control monocyte and granulocyte cell growth and function, and influence the type of adaptive immune responses that follow. NK cells have both antiviral and antitumor activities, and their ability to lyse targets does not depend on previous sensitization, nor is their activity major histocompatibility complex (MHC) restricted [5]. With regard to human immunodeficiency virus (HIV) infection, the nonspecificity of NK cell activity might be relevant to the maintenance of a degree of antiviral activity in the face of a high level of virus replication that negatively impacts HIV-specific cellular and humoral immune responses

It has been well established that CC-chemokines, particularly RANTES, macrophage inflammatory protein (MIP)–1α, and MIP-1β, block the entry of HIV that uses the CCR5 coreceptor by competitive inhibition [6–8 ]. NK cells from HIV-uninfected individuals have been shown to produce MIP-1α in response to interleukin (IL)–12 plus IL-15 [9]. Furthermore, studies from our laboratory [10] and others [11] have shown that NK cells from HIV-infected individuals also are capable of secreting CC-chemokines and suppressing HIV replication in vitro

Cell-mediated immune responses play an important role in host defenses against viral infections [12]. Among HIV-infected individuals, CD8⁺ T cells have been shown to play an integral role in controlling HIV viremia by MHC-restricted cytolytic [13, 14] or non–MHC-restricted noncytolytic responses [15–18 ]. We have shown elsewhere [19] that CD8⁺ T cells provide potent antiviral activity against endogenous HIV replication via autologous cell-to-cell contact among long-term nonprogressors and those who are treated with highly active antiretroviral therapy (HAART). Profound antiviral activity mediated by either cell-to-cell contact or soluble factor(s) has been demonstrated among HIV-infected individuals, for whom HAART was initiated early in HIV infection [19]. However, many of these patients demonstrated potent antiviral activity mediated by soluble factor(s) independent of CC-chemokines [19]. In this regard, the precise nature of NK cell contact and soluble factor–mediated suppression of endogenous HIV replication among HIV-infected individuals has not yet been fully delineated

In the present study, we compared the ability of NK cells with that of CD8⁺ T cells to suppress HIV replication in autologous CD4⁺ T cells in an ex vivo system. In addition, we examined the effect of viremia on these HIV inhibitory functions. Our results demonstrate that NK cells have a degree of suppression of HIV replication comparable to that of CD8⁺ T cells. However, unlike CD8⁺ T cells, which use both chemokine and nonchemokine mechanisms of soluble factor–mediated suppression of endogenous HIV replication, such suppression by NK cells is mediated almost exclusively by secretion of CC-chemokines. We also found that there is an inverse correlation between the presence and degree of viremia and the HIV inhibitory function

Subjects, Materials, and Methods

Study subjectsSeventeen HIV-1–infected individuals were included in the study (table 1). These patients included 9 patients with viremia who were either therapy naive or not receiving HAART at the time of study and 8 patients without viremia who were receiving therapy (plasma HIV RNA level <50 copies/mL) at the time of study. Three patients were enrolled in the primary HIV infection study at the University of Washington Hospital (Seattle), in accordance with the Institutional Review Board approval. Within the group of patients without viremia, HAART was initiated for 3 patients within 4 months after the onset of acute HIV symptoms and still without viremia (plasma HIV RNA level <50 copies/mL) at the time of study. One patient was diagnosed with acute HIV infection by documented seroconversion, but remained without viremia and did not receive HAART. Four patients received HAART during the chronic phase of infection and were without viremia at the time of study. To obtain peripheral blood mononuclear cells (PBMCs), all infected individuals were subjected to an apheresis, according to a protocol that was approved by the National Institute of Allergy and Infectious Diseases Institutional Review Board

Table 1

Profile of human immunodeficiency virus type 1 (HIV-1)–infected patients

Isolation of cellsCD4⁺ T cells, CD8⁺ T cells, and NK cells were isolated from peripheral blood of HIV-1–infected individuals by use of column-based cell separation techniques (StemCell Technologies), as described elsewhere [20, 21]. The CD4⁺ T cell enrichment cocktail contained antibodies to CD8, CD14, CD16, CD19, CD56, and glycophorin A. The CD8⁺ T cell enrichment cocktail contained antibodies to CD4, CD14, CD16, CD19, CD56, and glycophorin A. The NK cell enrichment cocktail contained antibodies to CD3, CD4, CD14, CD19, CD66b and glycophorin A. The cells obtained had a purity >95%

Cell culturesAfter isolation of CD4⁺ T cells, CD8⁺ T cells, and NK cells from HIV-1–infected subjects, each population of cells (1–2×10⁶ ) was incubated in 48-well plates with complete medium consisting of RPMI 1640 medium supplemented with 10% fetal calf serum (FCS), penicillin-streptomycin, l-glutamine, IL-2 (20 IU/mL; Roche Molecular Biochemicals), and anti-CD3 antibody (OKT3) in the presence of 2×10⁶ irradiated allogeneic PBMCs isolated from healthy HIV-seronegative donors. The following culture groups were established: CD4⁺ T cells alone, coculture of CD4⁺ T cells with either CD8⁺ T cells or NK cells (1:1), and CD4⁺ T cells with supernatants obtained from cultures of either CD8⁺ T cells or NK cells (1:1) in the presence or absence of monoclonal anti–CC-chemokine antibodies (20 μg/mL each of anti–MIP-1α, MIP-1β, and RANTES; R&D systems). Cultures were incubated in a 37°C CO₂ incubator for up to 12 days, and supernatants from each culture were removed and replaced with fresh medium and anti–CC-chemokine antibodies, where indicated, on days 3, 6, 9, and 12, for the determination of HIV-1 p24 antigen levels by ELISA. Culture supernatants from CD8⁺ T cells and NK cells (cultured under similar conditions as mentioned above) also were tested for levels of RANTES, MIP-1α, and MIP-1β secretion by ELISA (R&D systems)

Determination of intracellular interferon (IFN)–γ staining in NK cellsTo determine the frequency of NK cells expressing IFN-γ after stimulation with a target cell line, K562 (erythroleukemia cell line; ATCC category no. CCL-243), purified NK cells were incubated with target cells at a fixed effector-to-target (E:T) ratio of 30:1 for 4 h with complete medium in a 37°C CO₂ incubator. After 4 h of incubation, Brefeldin A (Sigma) was added to the medium at a final concentration of 10 μg/mL, to inhibit the secretion of IFN-γ. After the 4-h incubation, the cells were washed twice and fixed with 1× fixing solution (Becton Dickinson) for 10 min at room temperature, followed by another washing. Cells were permeabilized with 1× permeabilization solution 2 (Becton Dickinson) and further incubated at room temperature for 10 min. Cells then were washed and stained with the following antibodies: phycoerythrin (PE)–conjugated anti-CD56, fluorescein isothiocyanate (FITC)–conjugated IFN-γ, and peridnine chlorophyll protein (PerCP)–conjugated anti-CD3 antibodies (Becton Dickinson). With a gate on CD3⁻ CD56⁺ lymphocytes, ∼100,000 events were collected by use of FACSCalibur (Becton Dickinson), and the frequency of CD56⁺ IFN-γ⁺ cells was analyzed by use of CELLQUEST software (Becton Dickinson)

NK cell cytotoxicity assayFreshly isolated NK cells were tested for cytotoxicity against K562 targets. K562 cells were suspended in complete medium and labeled with the green fluorescent cell linker PKH67-GL (Sigma). The stained cells were washed 3 times in complete medium, and 2×10⁶ target cells were added in a volume of 100 μL in duplicate to 12×75-mm³ round bottom polystyrene tubes (Falcon). Effectors cells in a volume of 200 μL were added to yield different E:T ratios of 30:1, 20:1, and 10:1. Control tubes containing only target cells also were assayed. Effector and target cells were mixed by gentle tapping, centrifuged at 25°C for 4 min at 25 g and incubated at 37°C in 5% CO₂ for 3 h. After incubation, 10 μL of propidium iodide (PI; Pharmingen) was added to each tube 10–15 min before data acquisition. Two thousand target cells were collected per sample (corresponding to 20,000–200,000 total events) on FACSCalibur (Becton Dickinson). Target cells were gated by side scatter (SSC) and fluorescence (FL-1). PI uptake was determined within the gated cells. The percentage of NK-specific lysis was calculated by use of the following formula: NK-specific lysis = [(percentage of PI staining of sample-percentage of PI staining of negative control) / (100-percentage of PI staining of negative control)] × 100

Phenotyping of cultured HIVVirus-containing supernatants collected at the peak of viral replication in stimulated CD4⁺ T cells were phenotyped for coreceptor usage by exposure to the glioma cell lines U87/CD4 expressing the chemokine receptors CCR5 or CXCR4 (kindly provided by D. R. Littman, New York University, New York)

Statistical analysisAssociations between the suppressive capacity of CD8⁺ T cells and NK cells against replication of HIV-1 in CD4⁺ T cells in various culture conditions and obtained from patients with various virologic and immunologic parameters were determined by Spearman’s&amp;rank correlation test. Levels of chemokine secretion by NK and CD8⁺ T cells in the 2 groups of patients were compared by Student’s t test. Adjustment of P values for multiple testing was done by the Bonferroni method

Results

Suppression of viral replication by autologous NK and CD8⁺ T cells in patients with viremiaTo determine the degree of suppression of viral replication in stimulated CD4⁺ T cells by autologous NK and CD8⁺ T cells from patients with viremia who were not receiving HAART, we performed the following cultures: CD4⁺ T cells alone, CD4⁺ and NK or CD8⁺ T cell coculture, CD4⁺ T cells with supernatant from NK or CD8⁺ T cells, and CD4⁺ T cells with supernatant from NK and CD8⁺ T cells in the presence or absence of anti–CC-chemokine antibodies. The patients with viremia had variable levels of plasma viremia. After stimulation of purified CD4⁺ T cells with irradiated allogeneic PBMCs from HIV-uninfected donors, IL-2 and anti-CD3 antibodies, viral replication was detected in cultures of all 9 patients in each culture condition and at all time points examined (3, 6, 9, and 12 days). The peak p24 values from day 12 cultures are shown in figure 1A. When CD4⁺ T cells were cocultured with either NK or CD8⁺ T cells at 1:1 ratios, varying degrees of suppression was observed, ranging from 1.46 log to 2.76 log (mean log reduction, 2.25 log) for NK cells and from 2.08 to 3.97 log (mean log reduction, 2.70 log) for CD8⁺ T cells, compared with viral replication in the CD4⁺ T cells alone. However, when supernatants from cultures of NK cells or CD8⁺ T cells were added to CD4⁺ T cells, relatively low degrees of virus suppression were observed for both NK cells, ranging from 0.43 log to 2.51 log (mean log reduction, 1.42 log), and CD8⁺ T cells, ranging from 1.17 log to 3.96 log (mean log reduction, 2.10 log) (figure 1A). The presence of anti–CC-chemokine antibodies in cultures, in which supernatant from NK or CD8⁺ T cells were added to stimulated CD4⁺ T cells, had minimal effects on the levels of virus suppression (NK cells: range, 0.12–1.76 log; mean log reduction, 0.51; CD8⁺ T cells: range, 0.28–1.94; mean log reduction, 0.87; figure 1A). We used equal amounts of mouse IgG antibodies as a control and have not observed any detectable levels of HIV suppression. Taken together, these data demonstrate that modest antiviral activity by CD8⁺ T cells and NK cells is mediated by cellular contact and that soluble factors, including CC-chemokines, induce only a minor degree of suppression of viral replication in CD4⁺ T cells from chronically infected individuals who were not receiving antiretroviral therapy. Levels of MIP-1α and RANTES secreted by NK and CD8⁺ T cells from this group of patients were relatively low (<2 ng/mL). Supernatants from cultured NK cells did not produce detectable levels of p24 antigen on day 3, 6, 9, or 12

Figure 1

Role of NK cells and CD8⁺ T cells in the suppression of human immunodeficiency virus (HIV) replication in autologous CD4⁺ T cells in an ex vivo system. A Patients with viremia who were not receiving highly active antiretroviral therapy (HAART) at the time of study; B patients without viremia who were receiving HAART (B). Enriched NK, CD4⁺, and CD8⁺ T cells were stimulated with irradiated peripheral blood mononuclear cells (PBMCs) from HIV-seronegative donors and anti-CD3 antibodies in 4 different culture conditions, each as described in Subjects, Materials, and Methods. Cultures were maintained for 12 days with supernatants being removed every 3 days and replaced with fresh complete medium and anti–CC-chemokine antibodies where indicated. Supernatants were assayed for p24 on days 3, 6, 9, and 12. Results of p24 ELISA on day 12 are shown. p24 values <15 pg/mL were plotted under the dotted line.

Suppression of viral replication by autologous NK and CD8⁺ T cells in HIV-1–infected patients without viremiaIt has been demonstrated elsewhere that CD8⁺ T cell–mediated suppressive activity against HIV declines with disease progression [22–25 ]. However, studies addressing the role of NK cell–mediated suppressive factors in the inhibition of endogenous HIV replication in patients without viremia have been lacking. To further examine the role of NK cell–mediated suppression of endogenous HIV replication, cells from patients receiving HAART in whom aviremia was achieved were subjected to the same culture conditions as described above. Profound suppression of endogenous HIV replication (NK cells: range, 2.71–4.05; mean, 3.25 log; CD8⁺ T cells: range, 2.19–3.68 log; mean, 2.92 log) was noted (figure 1B). Furthermore, there was substantial suppression of HIV replication in CD4⁺ T cells when supernatants from either NK cells (range, 1.87–3.68 log; mean, 2.90 log) or CD8⁺ T cells (range, 2.58–3.68 log; mean, 3.05 log) were added (figure 1B). Of note, in the presence of excess anti–CC-chemokine antibodies, the suppressive effect of soluble NK cell-mediated factors was almost completely abrogated (range, 0.26–0.6 log; mean, 0.41 log), relative to replication of HIV in CD4⁺ T cells (figure 1B), which indicates that CC-chemokines played an important role in NK cell–mediated suppression of endogenous replication of HIV in this ex vivo system. However, addition of anti–CC-chemokine antibodies in excess only partially reversed the suppression mediated by soluble CD8⁺ T cell–mediated factors (range, 1.59–3.3 log; mean, 2.29 log; figure 1B), which suggests that CC-chemokines play only a partial role in suppressing endogenous HIV replication in this system. These findings suggest that, in HIV-1–infected patients who initiate HAART and in whom aviremia is achieved, cell contact and soluble factor-mediated suppression of HIV by both CD8⁺ T cells and NK cells is preserved. Furthermore, our data suggest that the suppression induced by NK cells in this system is almost exclusively mediated by CC-chemokines, whereas that induced by CD8⁺ T cells is mediated predominantly by non–CC-chemokine factors. Supernatants from cultured NK cells did not produce detectable levels of p24 production on day 3, 6, 9, or 12

Relationship between suppression of viral replication by NK cells and levels of virologic and immunologic parameters To address the relationship between the capacity of NK cells to suppress viral replication in CD4⁺ T cells and various virologic and immunologic parameters of HIV-infected patients, we obtained HIV RNA plasma levels and CD4⁺ and CD8⁺ T cell counts at the time of study for all the patients. We performed statistical tests of associations between the degrees of suppression of HIV replication by NK cells, either by cell-to-cell contact or by soluble factors and levels of plasma viremia at the time of study. As shown in figure 2, the ability of NK cells to suppress endogenous HIV replication either by cell-to-cell contact (figure 2A) or by soluble factors (figure 2B) inversely correlated with plasma viremia at the time of study (r=-0.90 and r=-0.84, respectively; P⩽.001). These data suggest that the higher the plasma virus load, the lower the ability of NK cells to suppress endogenous HIV replication in CD4⁺ T cells

Figure 2

Relationship between the ability of NK cells to suppress endogenous human immunodeficiency virus (HIV) replication and plasma HIV viremia at the time of study. Data are relationship between the level of plasma viremia and the suppression of HIV replication in culture conditions of cell-to-cell contact (A) and soluble factors (B). There was a strong inverse correlation between the ability of NK cells to suppress HIV replication, either by cell-to-cell contact or by soluble factors, and the level of plasma viremia

In addition, statistical comparisons were made between the log suppression of viral replication in both culture conditions (cell-to-cell contact and supernatant) and CD4⁺ and CD8⁺ T cell counts at the time of study. There was no statistically significant correlation between NK cell contact or soluble factor–mediated suppression of HIV replication and CD4⁺ T cell counts (P>.5) or CD8⁺ T cell counts (P=.07 with cell-to-cell contact–mediated and P=.16 with supernatant-mediated suppression)

Relationship between levels of CC-chemokines secreted by NK cells and plasma viremiaSince the ability of NK cells to suppress endogenous HIV replication in CD4⁺ T cells is predominantly mediated by secretion of CC-chemokines and inversely correlates with plasma viremia (see above), we next investigated whether plasma viremia had any effect on the levels of CC-chemokines secreted by NK cells. Supernatants from purified NK cells and CD8⁺ T cells were tested for levels of RANTES, MIP-1α, and MIP-1β, as measured by ELISA. Peak levels of CC-chemokine secretion were seen as early as day 3. The levels of CC-chemokines secreted by NK cells and CD8⁺ T cells on day 3 were higher among patients without viremia and receiving HAART, compared with those patients with viremia at the time of study (figure 3). The levels of both RANTES and MIP-1α secreted by NK cells inversely correlated with the levels of plasma viremia (r=-0.91 and r=-0.547, respectively; P=.05; figure 4A and 4B)

Figure 3

Levels of CC-chemokine production by the cultures of purified NK (A) and CD8⁺ T (B) cells from patients with or without viremia. Supernatants from both enriched NK and CD8⁺ T cells were collected daily for 3 days, and levels of RANTES, macrophage inflammatory protein (MIP)–1α, and MIP-1β were measured by ELISA. Levels of RANTES, MIP-1α, and MIP-1β from day 3 supernatants are shown. Levels of RANTES secreted by both NK and CD8⁺ T cells were significantly higher in patients without viremia than in patients with viremia (P<.001). There was no statistical significance of the levels of MIP-1α and MIP-1β secreted by both NK and CD8⁺ T cells between the 2 patient groups

Figure 4

Relationship between RANTES and macrophage inflammatory protein (MIP)–1α secretion by NK cells and plasma viremia. Relationship between plasma viremia of all the participating patients and levels of RANTES (A) and MIP-1α (B) secretion is shown. There is a strong inverse correlation between RANTES and plasma viremia. Levels of RANTES and MIP-1α measured by ELISA on day 3 are shown

Relationship between NK cell cytotoxicity and intracellular IFN-γ production and levels of plasma viremiaNK cells exhibit spontaneous cytotoxicity against various susceptible tumor cell lines. K562 is an erythroleukemia cell line highly susceptible to cytotoxicity by fresh NK cells. We used K562 as targets to test the cytotoxic capability of freshly isolated NK cells. Statistical associations by use of Spearman’s&amp;rank correlation method were made between the percentage of cytotoxicity at a fixed E:T ratio of 30:1 and plasma viremia at the time of study (figure 5A). Intracellular IFN-γ secretion by freshly isolated NK cells also were measured using K562 cells as targets. There was a strong correlation between the percentage of cytotoxicity against K562 cells and the frequency of NK cells expressing intracellular IFN-γ using K562 cells as targets (r=-0.79; P⩽.001). Correlations were determined between the percentage of NK cells staining for intracellular IFN-γ at the same fixed E:T ratio of 30:1 and plasma viremia. Neither NK cell cytotoxicity nor intracellular IFN-γ staining correlated with plasma viremia (figure 5). These results indicate that neither NK cell cytotoxicity against K562 nor intracellular IFN-γ staining using K562 as targets were affected by the level of plasma viremia and suggest, by extrapolation, that they were not responsible for the observed suppression of HIV replication by NK cells in both culture conditions

Figure 5

Relationship between NK cell–mediated cytotoxicity and intracellular interferon (IFN)–γ secretion and plasma viremia at the time of study. Freshly isolated NK cells were tested for cytotoxicity and intracellular IFN-γ staining using the K562 cell line as targets ex vivo, as described in Subjects, Materials, and Methods. Relationship between plasma viremia and percentage of NK cell cytotoxicity (NKCC) at an effector-to-target (E:T) ratio of 30:1 (A) and percentage of NK cells staining for intracellular IFN-γ at an E:T ratio of 30:1 (B) is shown. There was no correlation between cytotoxicity or percentage of NK cells staining for intracellular IFN-γ and plasma viremia

Discussion

The present study demonstrates that NK cells can mediate suppression of endogenous HIV replication in autologous CD4⁺ T cells of HIV-1–infected individuals to a level comparable with that of CD8⁺ T cells. The suppression was mediated in a culture system of cell-to-cell contact, as well as by soluble culture supernatant. Of note is the fact that the degree of suppression by both NK and CD8⁺ T cells was much more profound in patients without viremia, compared with that in patients with viremia, in both systems (cell-to-cell contact and supernatant) and that there was an inverse correlation between the level of plasma viremia and the ability to suppress HIV replication. Furthermore, the suppression of HIV replication by soluble factors was virtually exclusively mediated by CC-chemokines in NK cells, whereas nonchemokine factors were involved predominantly in CD8⁺ T cells. In this regard, anti–CC-chemokine antibodies almost completely abrogated the suppression of HIV replication mediated by NK–derived soluble factors, whereas such antibodies had only a modest effect on the suppression mediated by CD8⁺ T cell–derived soluble factors. This latter observation is consistent with that reported in previous studies [15–17 ], in which, as yet unidentified, non–CC-chemokine factors were shown to have substantial anti-HIV activity. Of note, a recent study has suggested that α-defensins–1, -2, and -3 may be responsible for at least a part of the anti-HIV activity not attributable to CC-chemokines [26]. In addition, this observation confirms the results of our previous study [19] that used a similar ex vivo system, in which we demonstrated that, in patients treated during the acute stage of HIV infection, supernatants derived from CD8⁺ T cells mediated profound suppression of HIV replication by non–CC-chemokine factors. Although it has been shown in previous studies that CD8⁺ T cells are capable of suppressing endogenous HIV replication both by cell-to-cell contact and by secreting soluble factors [16, 19], the suppressive activity of autologous NK cells against HIV replication in endogenously infected CD4⁺ T cells has not been studied previously, nor has the effect of viremia on NK versus CD8⁺ T cell–mediated suppression of HIV replication. The supernatants from activated NK cells used in these experiments did not contain any detectable level of p24 antigen as detected by ELISA, which indicates that the viral production was restricted entirely to endogenous CD4⁺ T cells

The exact antiviral role played by NK cells has not yet been fully delineated. Several studies have correlated high frequency of NK cells or NK cell activity with reduced susceptibility of certain individuals to HIV infection [25] and with the control of initial plasma viremia in the simian immunodeficiency virus model [27]. Loss of NK cell activity has been correlated with HIV disease progression, particularly in individuals with opportunistic infections [28–30 ]. Since there was no correlation between NK-mediated cytotoxicity and plasma viremia, it appears that autologous NK cells exerted cell-to-cell contact–mediated antiviral activity against CD4⁺ T cells independently of cytolytic activity. It could be speculated that NK cells could have exerted cell membrane–bound molecules that may inhibit HIV replication in CD4⁺ T cells. With regard to interpatient variability of NK-mediated HIV suppression, we have found that there is some degree of variability among patients, as would be expected. Even patients who we monitored longitudinally have exhibited varying degrees of suppression. Since all these patients were HIV infected, changes in levels of HIV viremia probably is the predominant factor influencing this variability

Our data clearly demonstrate that the ability of NK cell–supernatant–mediated suppression of endogenous HIV replication is related to their ability to secrete CC-chemokines. The ability of NK cells to secrete CC-chemokines was significantly depressed among patients with viremia versus patients without viremia. This indicates that HIV-induced inhibition of NK cell function could involve mechanisms that lead to depressed secretion of CC-chemokines

Natural cytotoxicity and intracellular IFN-γ secretion by freshly isolated NK cells tested against K562 did not show a correlation with plasma viremia. This result probably indicates that cytotoxicity against K562 may not be the best marker for HIV suppressive activity of NK cells. Previously reported larger longitudinal studies have shown that cytotoxicity against K562 did not show a correlation with disease progression [31]. In addition, studies with CD8⁺ T cells failed to show a correlation between HIV-specific cytotoxic CD8⁺ T cells and the degree of suppression of CD8⁺ T cells of endogenous HIV replication [19]. Thus, it is necessary to address the role of cytolytic functions by NK cells in greater detail by use of a wider variety of susceptible targets

Finally, this study has potentially important implications for understanding the full scope of host defense mechanism against HIV infection, including that of the innate immune system. NK cells probably play an important role in protection against both infection and progression of HIV disease. In addition, this study provides yet another example of the multifaceted deleterious effects that active virus replication has on immune function [19, 32, 33]. Further studies on the precise mechanism of the effects of NK cell and of innate immune mechanisms on HIV replication will be important not only for furthering our understanding of HIV pathogenesis but also for widening the potential for immune-based interventions in HIV disease

Acknowledgments

We thank the National Institute of Allergy and Infectious Diseases study co-coordinators and case managers for recruiting patients. We also thank the patients for their participation. We would also like to thank Claire Stevenson and Janine Maenza for their help in recruitment of primary infection patients

Presented in part: 9th conference on Retroviruses and Opportunistic Infections, Seattle, 24–28 February 2002 (abstract 25)