Impaired Macrophage Phagocytosis of Apoptotic Neutrophils in Patients with Human Immunodeficiency Virus Type 1 Infection

Abstract

Dysfunction of neutrophils (polymorphonuclear leukocytes [PMNL]) and macrophagic cells occurs as a consequence of human immunodeficiency virus type 1 (HIV-1) infection. Macrophages contribute to the resolution of early inflammation ingesting PMNL apoptotic bodies. This study investigated macrophage ability to phagocytose PMNL apoptotic bodies in patients with HIV-1 infection in comparison with uninfected individuals and the effect of HIV Nef protein on apoptotic body phagocytosis to determine if phagocytic activity is impaired by HIV infection. Monocytes/macrophages were isolated from 10 HIV-1-infected patients and from five healthy volunteers, whereas PMNL were isolated from healthy volunteers. Macrophage phagocytosis of apoptotic PMNL was determined by staining of apoptotic bodies with fluorescein-conjugated concanavalin A or with fluorescein-labeled phalloidin. Our data show significant impairment of PMNL apoptotic body macrophage phagocytosis in subjects with HIV-1 infection presenting a concentration of CD4⁺ T lymphocytes of >200/mm³ and in particular in those with <200 CD4⁺ T lymphocyte cells/mm³. In addition, HIV-1 recombinant Nef protein is able to decrease phagocytosis of apoptotic PMNL from normal human macrophages in a dose-dependent manner. The results of our study suggest that impaired macrophage phagocytosis of PMNL apoptotic bodies may contribute to the persistence of the inflammatory state in HIV-infected subjects, especially during opportunistic infections that are often favored by defective phagocytic activity.

Neutrophil (polymorphonuclear leukocytes [PMNL]) function, including chemotaxis, phagocytosis, oxidative burst capacity, and bacterial killing, is impaired in the course of human immunodeficiency virus type 1 (HIV-1) infection, particularly in the later stages of the disease, and this abnormal function may predispose to some secondary bacterial infections and/or to opportunistic infections (4, 8, 10, 11, 12). PMNL have the shortest half-life of all circulating leukocytes and are programmed to die within 1 day. These aging leukocytes spontaneously undergo apoptosis and are recognized and phagocytosed by macrophages (15).

Pitrak et al. (13) have demonstrated that the rate of PMNL apoptosis is accelerated in AIDS patients, and this defect is intrinsic and not an effect of endogenous serum factors. Moreover, it has been proposed that the ingestion of apoptotic PMNL triggers production of anti-inflammatory mediators from macrophages (6, 9), whereas persistent PMNL-rich inflammatory infiltrates have been associated with unresolved inflammatory reactions, including adult respiratory distress syndrome and rheumatoid arthritis (16). Thus, the removal of apoptotic cells appears to be critical in the resolution of inflammation. We have previously demonstrated that macrophages from HIV-positive subjects have a reduced ability to phagocytose Candida albicans cells, and there is a significant decrease in oxidative processes for the intracellular killing. These phenomena seem to be induced, at least in part, by HIV Nef protein (14).

Since the effects of macrophage phagocytosis of apoptotic PMNL have not been completely investigated, especially in HIV-positive subjects, the purpose of this study was to evaluate phagocytosis of PMNL apoptotic bodies performed by macrophagic cells obtained from HIV-1-positive subjects and in parallel by the macrophages obtained from healthy individuals. Furthermore, we studied the effect of Nef protein on PMNL apoptotic body macrophagic phagocytosis, since this viral protein is able to depress both specific and nonspecific immune responses in HIV-infected patients, particularly microbial phagocytosis (1, 14).

MATERIALS AND METHODS

Subjects.

Ten HIV-1-infected subjects (mean age, 35.3 ± 5.8 years) were enrolled and five healthy volunteers (mean age, 37.1 ± 4.4 years), without HIV-1 risk factors, served as controls. Five of the HIV-1-infected subjects presented more than 200 CD4⁺ T lymphocytes/mm³ (mean = 517 ± 225), had a CD4/CD8 T-cell ratio of 0.5 ± 0.1, and had mean HIV-1 RNA levels in plasma of 32,628 ± 42,188 copies/ml. Five patients had less than 200 CD4⁺ T lymphocytes/mm³ (mean = 101 ± 70), a ratio of 0.2 ± 0.2, and mean HIV-1 RNA levels of 375,000 ± 246,815/ml. All these patients had a moderate anemic state and were generally studied before receiving antiretroviral therapy. In fact, a significant part of our study population included individuals who had ignored their seropositive condition for a long time and in consequence came to medical evaluation late.

Monocyte and PMNL isolation and apoptotic body preparation.

Monocytes obtained from peripheral blood of healthy subjects and HIV-1-infected patients were selected as adhering cells after separation of peripheral blood mononuclear cells with a Ficoll Paque gradient (Pharmacia, Uppsala, Sweden). After repeated washes, the adherent cells were harvested by trypsinization and resuspended at a concentration of 10⁶/ml in RPMI-1640 medium (Gibco, Paisley, Scotland), supplemented with 10% fetal calf serum (Celbio, Milan, Italy). Then, macrophage colony-stimulating factor (10 ng/ml; Genzyme, Milan, Italy) was added to the medium in order to differentiate monocytic cells into activated macrophages, and the cultures were maintained at 37°C with 5% CO₂.

Cellular cultures were enriched for macrophages by adherence and repeated trypsinization and, to obtain a pure population, by phenotypic evaluation (more than 98% of cells). After 7 days of incubation in flasks (25 cm²; Nunc Kamstrup, Baltimore, Md.), the cells were harvested by trypsin treatment and transferred, at a final concentration of 5 × 10⁵/ml, to four-well chamber slides (Lab-Tek; Nunc).

Cultures of PMNL were obtained from peripheral blood of healthy subjects by separation with a Polymorphoprep gradient (Pharmacia). PMNL cells spontaneously go to apoptosis after 24 h of culture, and this method was adopted to induce apoptosis. In order to separate the apoptotic bodies of PMNL, after centrifugation at 600 rpm for 10 min (to remove nonapoptotic cells), the supernatants were centrifuged at 3,000 rpm, and the pellets containing PMNL apoptotic bodies were resuspended at the indicated concentrations.

Fluorescent microscopy for phagocytosis detection.

Two staining methods were used to detect apoptotic bodies of PMNL in macrophagic cells. In the first method, 5 × 10⁶ apoptotic bodies/ml were incubated for 20 min at room temperature with fluorescein-conjugated concanavalin A (ConA; Calbiochem Corp., La Jolla, Calif.) at a final concentration of 150 μg/ml. Then the apoptotic bodies were washed twice by centrifugation for 10 min at 3,000 rpm and resuspended in RPMI-1640 with 10% fetal calf serum at a concentration of 5 × 10⁶/ml to obtain a ratio of 5:1 per macrophagic cell (this preparation was diluted 1:2 in the wells containing macrophagic cells). This method allows uniform staining of the apoptotic body surface, making the apoptotic bodies easily detectable in macrophagic cells without having any significant influence on phagocytosing activity.

In the second method, fluorescein-labeled phalloidin (Sigma-Aldrich Chemicals, Milan, Italy) was used at a final concentration of 5 μg/ml. This product is able to stain the interior of the apoptotic bodies, binding firmly to the microtubular structures. This staining method was used like the first one.

In a subsequent set of experiments, macrophage phagocytosis of PMNL apoptotic bodies was also investigated in the presence of various concentrations of recombinant Nef protein (American Biotechnologies, Inc., Cambridge, Mass.), using macrophagic cells obtained from four healthy individuals and performing the experiments in triplicate. The challenge with stained apoptotic bodies, using both the first and second methods, was made at a ratio of 5:1.

After 2 h of incubation in the presence of apoptotic bodies, the supernatants were removed, and slides were detached from their supports and washed twice with phosphate-buffered saline (PBS). Then the slides were fixed with 10% ethyl alcohol solution, mounted in phosphate-buffered glycerol (30% PBS and 70% glycerol, vol/vol), and examined under a fluorescence microscope at ×400 by four blinded microscopists. At least five microscopic fields were observed for each sample.

Macrophage oxidative phenomenon evaluation.

We indirectly studied the oxidative phenomena of macrophages in the absence or presence of Nef protein, by evaluating the antioxidant power of the supernatants, as previously described (14). For this purpose we used the antioxidant power (PAO) kit furnished by Med. Dia S.r.l. (San Germano, Vercelli, Italy). The test is based on the detection of Cu⁺ ions produced by the reduction of a known amount of Cu²⁺; this reduction is induced by the antioxidant factors present in the culture, the activity of which decreases in an inversely proportional manner to the activity of macrophage oxidative phenomena. The concentration of Cu²⁺ ion was detected through the formation of complexes consisting of Cu⁺ and the chelating agent betacuproine disulfonate (2,9-dimethyl-4,7-diphenyl-1,10-phenanthrolinedisulfonate), and it was measured spectrophotometrically at 490 nm (Metertech spectrophotometer; Medical System, Genoa, Italy). The data are expressed as microequivalents per liter of reducing equivalents ± standard deviation, corresponding to PAO units.

Statistical analysis.

The results were expressed as mean and standard deviation. The differences among the experimental and control groups were statistically evaluated using Student's t test. Statistical significance was defined as P ≤ 0.05. Moreover, a linear correlation test was employed for the correlation study reported in the results.

RESULTS

In the first set of experiments, we studied the phagocytosis of PMNL apoptotic bodies stained with fluoresceinated ConA from macrophages of patients with HIV-1 infection and from those of uninfected healthy individuals.

A significant decrease in the percentage of phagocytosing cells was observed in HIV-1-positive subjects presenting a CD4⁺ T-cell concentration of >200 cells/mm³ (24.2% ± 12.3%; P = 0.002) and in those with <200 CD4⁺ T cells/mm³ (23.7% ± 6.9%; P < 0.001), compared to healthy control subjects (50.1% ± 3.7%) (Table 1). We point out that by analyzing all the HIV-positive subjects, the mean number of phagocytosing cells was 23.9% ± 10% in 52% of the controls (P < 0.001). Since fluoresceinated ConA only stains the surface of apoptotic bodies, we performed further experiments with fluoresceinated phalloidin, which is endowed with a specific ability to bind to microtubular structures. We also observed a reduction of about 50% in phagocytosing cells in HIV-positive subjects compared to the controls (unreported data).

TABLE 1.

Percentage of macrophages phagocytosing PMNL apoptotic bodies stained with fluoresceinated ConA: ex vivo test of macrophages from uninfected and HIV-1-infected subjects^a

Subjects	No.	Mean % phagocytosing macrophages ± SD
Controls	5	50.1 ± 3.7
HIV seropositive, >200 CD4+ T cells/μl	5	24.2 ± 12.3*
HIV seropositives, <200 CD4+ T cells/μl	5	23.7 ± 6.9**

^a*, P = 0.002 compared with controls; **, P < 0.001 compared with controls; P < 0.001 for controls versus all HIV-seropositive subjects (23.9% ± 10%).

As reported in Table 2, a significant decrease in the number of PMNL apoptotic bodies phagocytosed per macrophagic cell was observed in HIV-1 subjects with >200 CD4 T cells/mm³ (0.6 ± 0.4; P = 0.001) and in those with <200 CD4 T cells/mm³ (0.3 ± 0.1; P < 0.001) compared to the controls (2.4 ± 0.7).

TABLE 2.

Number of PMNL apoptotic bodies phagocytosed per macrophage^a

Subjects	No.	Mean no. of apoptotic PMNL/macrophage ± SD
Controls	5	2.4 ± 0.7
HIV-positive subjects with >200 CD4+ T cells/μl	5	0.6 ± 0.4*
HIV-positive subjects with <200 CD4+ T cells/μl	5	0.3 ± 0.1**

^a*, P = 0.001 compared with controls; **, P < 0.001 compared with controls; P < 0.001 for controls versus all HIV-positive subjects (0.45% ± 0.3%).

Evaluating all the HIV-positive subjects, the mean number of apoptotic bodies phagocytosed per cell was 0.45 ± 0.3 (P < 0.001). As can be seen in Table 2, the reduction of the mean number of PMNL apoptotic bodies phagocytosed per cell in HIV-positive subjects compared to the controls is fivefold, and the percent reduction in phagocytosing cells in infected patients is twofold. This observation suggests a decreased ability of single phagocytosing cells to swallow apoptotic bodies.

In a further set of experiments, we studied the phagocytic activity of PMNL apoptotic bodies in normal human macrophages preincubated with recombinant Nef protein. The results obtained demonstrate that this protein is able to inhibit the phagocytosis of PMNL apoptotic bodies. In particular, preincubation with Nef (2 h before the challenge with apoptotic bodies) inhibited macrophage phagocytosis of PMNL apoptotic bodies by about 50% (P = 0.01) at the concentration of 1 μg/ml and 65% at the concentration of 2 μg/ml (P < 0.001). When the preincubation time was prolonged to overnight, inhibition of macrophage phagocytosis was 80% at a Nef protein concentration of 2 μg/ml.

Table 3 reports the mean number of PMNL apoptotic bodies stained with fluoresceinated phalloidin phagocytosed per cell in the absence or presence of Nef protein. As shown, the reduction was about 34%, 90%, and 92% with Nef at 0.5, 1, and 2 μg/ml, respectively. Consequently, in these experiments the reductions induced by Nef were also more evident when evaluating the number of PMNL apoptotic bodies phagocytosed per cell than the percentage of phagocytosing cells.

TABLE 3.

Effect of Nef viral protein on phagocytosis of PMNL apoptotic bodies^a

Nef concn (μg/ml)	Mean no. of apoptotic bodies phagocytosed/cell ± SD
0	3.8 ± 2.7
0.5	2.5 ± 2.0
1.0	0.4 ± 0.6*
2.0	0.3 ± 0.5*

^a*, P < 0.001 compared with controls.

Moreover, we observed a negative and significant correlation between the Nef protein concentration and the number of PMNL apoptotic bodies phagocytosed per cell (r = −0.65; P = 0.0221). The mean number of apoptotic bodies phagocytosed related to 10⁶ macrophages was 1.18 × 10⁶ in the absence of HIV infection (23% of the total) and 48,000 with a viral load in plasma of 8 × 10⁵ HIV-1 RNA copies/ml (about 1%). A negative correlation was found between the HIV RNA load and the number of apoptotic bodies phagocytosed (r = −0.623; P < 0.05).

Finally, we evaluated the antioxidant power (PAO) of the macrophage supernatants in the absence or presence of various concentrations of Nef, and we found a significant increase in PAO that meant a reduction of oxidative phenomena in the presence of the viral protein compared to the control values, especially at a concentration of 2 μg/ml (PAO units = 164.1 ± 8.0 with Nef at 2 μg/ml versus 105.6 ± 12.9 without Nef; P < 0.001).

DISCUSSION

The results of this study show that phagocytosis of PMNL apoptotic bodies by macrophages of HIV-1-infected patients is impaired and that Nef, a regulatory viral protein of HIV, is able to decrease phagocytosis of PMNL apoptotic bodies by human normal macrophages. In addition, macrophagic oxidative phenomena are depressed by Nef protein. This is relevant because these effects are involved in intracellular killing processes and consequently in the destruction of phagocytosed particles (14). The decrease in oxidative processes and the inhibition of macrophagic function seem to occur in parallel.

Aging PMNL spontaneously undergo apoptosis and are recognized and phagocytosed by monocytes and macrophages (15). PMNL function is impaired in all stages of HIV-1 infection and especially in the terminal stage of the disease (17). Pitrak et al. (13) have shown that abnormalities of PMNL function observed in HIV-positive subjects might partly depend on the accelerated apoptosis induced by HIV infection. In fact, during all stages of HIV-1 infection, there is an increased number of apoptotic PMNL which are unable to function as host defenders (2, 17). The removal of apoptotic cells appears to be central to the resolution of inflammation. In fact, the clearance of apoptotic PMNL not only prevents the release of toxic and immunogenic intracellular contents, but also stimulates the macrophages to produce inflammatory mediators, including transforming growth factor β1, prostaglandin E₂, and platelet-activating factor, and inhibits the production of tumor necrosis factor, interleukin-1β, and interleukin-8 (6).

Dysregulation of PMNL function, along with that of monocytic macrophagic cells, in HIV-1-infected patients is reflected in the increased incidence of some microbial infections among these patients (5, 7). During several microbial infections, PMNL migrate and accumulate at the inflammatory sites, followed by removal of inflammatory cells. This occurs mainly by apoptosis and by phagocytosis of apoptotic bodies. In parallel, some antimicrobial pathogens can be phagocytosed and killed by macrophages. These phenomena appears to be critical to the resolution of inflammation and infection. However, the decreased phagocytosis of apoptotic PMNL by macrophages in HIV-1-infected patients and the accelerated apoptosis of PMNL lead to accumulation of apoptotic inflammatory PMNL and a decrease in their clearance. The persistence of apoptotic PMNL and their apoptotic bodies at the inflammatory site may maintain the inflammatory state through persistent stimulation of proinflammatory cytokines (6). This can explain some of the pathological conditions in the gastrointestinal tract or bronchoalveolar tract that have been reported (3).