|Home | About | Journals | Submit | Contact Us | Français|
Candida albicans causes oropharyngeal candidiasis (OPC) but rarely disseminates to deep organs in human immunodeficiency virus (HIV) infection. Here, we used a model of OPC in CD4C/HIVMut transgenic (Tg) mice to investigate the role of polymorphonuclear leukocytes (PMNs) and CD8+ T cells in limiting candidiasis to the mucosa. Numbers of circulating PMNs and their oxidative burst were both augmented in CD4C/HIVMutA Tg mice expressing rev, env, and nef of HIV type 1 (HIV-1), while phagocytosis and killing of C. albicans were largely unimpaired compared to those in non-Tg mice. Depletion of PMNs in these Tg mice did not alter oral or gastrointestinal burdens of C. albicans or cause systemic dissemination. However, oral burdens of C. albicans were increased in CD4C/HIVMutG Tg mice expressing only the nef gene of HIV-1 and bred on a CD8 gene-deficient background (CD8−/−), compared to control or heterozygous CD8+/− CD4C/HIVMutG Tg mice. Thus, CD8+ T cells contribute to the host defense against oral candidiasis in vivo, specifically in the context of nef expression in a subset of immune cells.
Oropharyngeal candidiasis (OPC) is the most frequent opportunistic fungal infection among human immunodeficiency virus (HIV)-infected patients (46). Although the incidence of OPC in HIV infection is reduced with use of highly active antiretroviral therapy (32), it remains a common opportunistic infection. The mechanisms underlying the predisposition to OPC among HIV-infected patients have not been precisely defined. Although defective adaptive immunity is pivotal to the immunopathogenesis of OPC in HIV infection (14), completely or partly preserved compensatory host defense mechanisms most likely limit Candida albicans proliferation to the mucosa and prevent systemic dissemination.
CD8+ T cells accumulate in the basal epithelial layer of the oral mucosa in HIV-infected patients with OPC (35, 42), demonstrating that these cells are actively recruited to the mucosa in response to candidiasis. However, the role of CD8+ T cells in mucosal containment of C. albicans in HIV infection has not been determined. In addition, HIV-infected patients with the erythematous form of OPC have abundant neutrophilic microabscesses in the parakeratin layer of the epithelium (18, 40, 42). Although recruitment of polymorphonuclear leukocytes (PMNs) to the oral epithelium does not appear to be perturbed by HIV infection, investigations comparing production of reactive oxygen intermediates with phagocytosis and killing of C. albicans by PMNs from normal and HIV-infected patients have produced conflicting results (8, 17, 25, 48, 49).
The availability of CD4C/HIVMut transgenic (Tg) mice expressing gene products of HIV type 1 (HIV-1) in immune cells and developing an AIDS-like disease (27) has provided the opportunity to devise a model of OPC that closely mimics the clinical and pathological features of candidal infection in human AIDS (13). With the recognition that a cause-and-effect analysis of the immunopathogenesis of mucosal candidiasis in HIV infection can now be achieved under controlled conditions with these Tg mice, the present study was undertaken to determine the roles of PMNs and CD8+ T cells in limiting chronic oral carriage and dissemination of C. albicans to deep organs. Here we show that although PMNs from these Tg mice are quantitatively augmented and nearly intact functionally, they are nevertheless dispensable for limiting chronic oral carriage and for preventing systemic dissemination of C. albicans. We also demonstrate that augmentation of oral burdens of C. albicans in mice bred on the CD8 knockout (KO) (CD8−/−) background occurs in CD4C/HIVMutG Tg mice which express only the nef gene of HIV-1. These results provide evidence indicating that CD8+ T cells participate in the host defense against C. albicans in vivo.
The CD4C/HIVMutA Tg mice have been described elsewhere (27). CD4C/HIVMutA mutant DNA harbors mouse CD4 enhancer and human CD4 promoter elements to drive the expression of HIV-1 genes in CD4+ CD8+ and CD4+ thymocytes, in peripheral CD4+ T cells, and in macrophages and dendritic cells. Founder mouse F21388 was bred on the C3H background. Animals from this line express moderate levels of the transgene, with 50% survival at 3 months (27). Several HIV-1 genes (gag, pol, vif, vpr, tat, and vpu) are mutated in the CD4C/HIVMutA DNA, whereas env, rev, and nef are intact. The generation of CD4C/HIVMutG mice revealed that selective expression of the nef gene is required and sufficient to elicit an AIDS-like disease in these Tg mice (27). This disease is characterized by failure to thrive, wasting, severe atrophy and fibrosis of lymphoid organs, loss of CD4+ T cells, interstitial pneumonitis, and segmental glomerulosclerosis associated with tubulointerstitial nephritis (27).
CD8a homozygous KO mice were obtained from Tak Mak (formerly from Amgen) and contained a targeted mutation of the CD8 receptor alpha chain (24). They were bred for at least six generations on the C3H background and then with the CD4C/HIVMutG Tg mice to generate homozygous (CD8−/−) or heterozygous (CD8+/−) CD4C/HIVMutG Tg mice.
Specific-pathogen-free male and female Tg mice and non-Tg littermates were housed in sterilized individual cages equipped with filter hoods and were supplied with sterile water and fed with sterile mouse chow. All animal experiments were approved by the Animal Care Committee of the University of Montreal.
Oral inoculation with C. albicans LAM-1, assessments for signs of morbidity, quantification of C. albicans in the oral cavities of individual mice, and determination of burdens of C. albicans in the gastrointestinal tract and internal organs were conducted as described previously (13).
Blood was drawn from the saphenous vein by using the Unopette microcollection system for determination of total leukocytes (Becton Dickinson Vacutainer Systems, Franklin Lakes, NJ). The differential leukocyte count was performed manually by staining a smear of blood with May-Grunwald-Giemsa stain, and the absolute leukocyte count was calculated and expressed as number of cells per cubic millimeter.
PMNs were isolated from whole peripheral blood by using a Histopaque-1119/-1077 double gradient (Sigma-Aldrich, St. Louis, MO). The purity of PMNs was >90% by May-Grunwald-Giemsa staining and flow cytometry analysis.
The PMN oxidative burst was measured by conversion of nonfluorescent dichlorofluorescein diacetate to the fluorescent compound 2′,7′-dichlorofluorescein, using the dichlorofluorescein diacetate peroxide CellProbe reagent (Beckman Coulter, Fullerton, CA). Mean fluorescence of PMNs was measured on a FACStar flow cytometer (BD Biosciences).
To assay phagocytosis, C. albicans LAM-1 was grown in Sabouraud dextrose broth (Becton, Dickinson and Company, Sparks, MD) at 30°C for 18 h with rotary agitation. The cells were collected by centrifugation and washed twice with 10 mM phosphate-buffered saline, pH 7.2. Then, 2.5 × 105 PMNs were incubated with fluorescein isothiocyanate (FITC)-labeled, heat-inactivated (90°C, 45 min) C. albicans blastoconidia in 100 μl of supplemented RPMI medium at different PMN-to-blastoconidia ratios for 30 min at 37°C. Phagocytosis was stopped by cooling the samples to 4°C. A control phagocytosis assay was conducted at 4°C (1:5 ratio) in the presence of cytochalasin B (Sigma; 5 μg/ml) (21). Ethidium bromide (10 μg/ml) was added, and the percentage of PMNs with phagocytosed C. albicans was measured by flow cytometry (47). Phagocytosis of C. albicans by PMNs was confirmed by confocal microscopy.
To assay killing of C. albicans, 2.5 × 105 PMNs were incubated with FITC-labeled, live C. albicans blastoconidia in 100 μl of supplemented RPMI medium at a 1:1 effector/target cell ratio at 37°C in independent wells. A control assay was performed at 4°C. After PMNs were lysed with 2.5% sodium desoxycholate, killed C. albicans cells were labeled with ethidium bromide and quantitated by flow cytometry.
Rat immunoglobulin G2b (IgG2b) monoclonal antibody RB6-8C5 (BD Biosciences), recognizing the Ly-6G (Gr-1) myeloid differentiation antigen on mouse granulocytes (23, 36), was used to deplete PMNs in the Tg mice. Three successive 50-μg doses of purified RB6-8C5 antibody administered intraperitoneally every 2 days resulted in a >90% depletion of circulating PMNs, which was sustained for at least 7 days. Control mice were treated with 50 μg of rat IgG2b isotype control (clone A95-1; BD Biosciences).
Collection of heparinized blood and cell surface marker analysis of peripheral blood lymphocytes were performed as reported previously (13).
Cervical lymph nodes (CLNs) were removed, mechanically disrupted by being pressed through a nylon mesh (pore size, 70 μm), and deposited in 25-mm-diameter dishes containing 2 ml of Hanks balanced salt solution (Gibco, Grand Island, NY). Cell suspensions were twice washed in Hanks balanced salt solution and resuspended in complete tissue culture medium consisting of RPMI 1640 medium (Gibco) supplemented with 10% heat-inactivated fetal bovine serum (Gibco), 20 mM HEPES buffer, 2 mM l-glutamine, 5 × 10−5 M β-mercaptoethanol, 100 U/ml penicillin and streptomycin, 0.25 μg/ml amphotericin B, and 50 μg/ml of gentamicin. Cells were filtered through a sterile nylon mesh (pore size, 70 μm) to obtain a homogeneous suspension. Cells were resuspended in complete medium and adjusted to 1 × 106 cells/ml, and cell viability was >90% by trypan blue exclusion. CLN cells were surface stained with anti-CD8-FITC, anti-CD44-biotin, and anti-CD62L-phycoerythrin antibodies (BD Biosciences), washed twice with PBS, and stained with streptavidin-allophycocyanin (BD Biosciences). After washing, flow cytometry analysis was performed by gating on CD8+ cells to quantitate the proportions of total, naive, and memory-like CD8+ T cells on a FACSCalibur flow cytometer (BD Biosciences) equipped with CellQuest software.
Oral burdens of C. albicans were compared using PROC MIXED software (SAS Institute, Cary, NC). Repeated-measurement analysis of variance with Tukey-Kramer contrast analysis was conducted with two factors, one between (group) and one within (time). Significant interactions (P < 0.05) were further analyzed, and significant differences (P < 0.05) between group means at fixed times were determined by use of the two-sample, two-tailed Student's t test for independent samples. Student's t test was used to determine differences in peripheral blood and CLN CD8+ T cells. All other data were analyzed with SPSS version 11.5 software (SPSS, Chicago, IL), using analysis of variance followed by Tukey contrast analysis. Differences were considered to be significant at a P value of <0.05.
Absolute total leukocyte counts did not differ among the four groups of mice over the course of AIDS-like disease (P> 0.05) (Fig. (Fig.1).1). However, the Tg mice were lymphopenic compared to the non-Tg mice (P < 0.05) irrespective of infection with C. albicans (P > 0.05), but without progression over time (P > 0.05) (Fig. (Fig.1).1). In the non-Tg mice, absolute lymphocyte counts were lower (P < 0.05) after oral infection with C. albicans than in uninfected controls. Absolute PMN counts were enhanced in Tg compared to non-Tg uninfected animals throughout the course of AIDS-like disease (P< 0.01) (Fig. (Fig.1).1). Oral infection with C. albicans further enhanced PMN counts in both the Tg and non-Tg mice (P < 0.05), resulting in greater absolute numbers of PMNs in infected Tg compared to infected non-Tg mice (P < 0.001) (Fig. (Fig.1).1). Circulating monocytes were significantly increased (P < 0.05) at day 70 compared to day 7 in both infected and uninfected Tg mice, indicating that HIV-1 transgene expression alone augments this cell population irrespective of infection with C. albicans. Taken together, these findings demonstrated that while lymphopenia is present in these Tg mice, chronic carriage of C. albicans not only does not result from a quantitative defect in circulating PMNs but in fact stimulates a greater neutrophil response in the Tg than in non-Tg mice.
In comparison to that in uninfected non-Tg control mice, the PMN oxidative burst was not significantly (P > 0.05) increased in infected non-Tg and uninfected Tg mice (Fig. (Fig.2).2). However, the oxidative burst of PMNs from the infected Tg mice was enhanced (P < 0.05) at both 7 and 70 days after infection compared to that in the uninfected non-Tg control animals, indicating that transgene expression and C. albicans infection together increase the PMN oxidative burst. Accordingly, chronic oral carriage of C. albicans in these CD4C/HIVMutA Tg mice was not associated with a perturbation in the oxidative capacity of PMNs.
Phagocytosis of C. albicans by PMNs from infected or uninfected Tg and non-Tg mice did not differ (P > 0.05) 7 days after infection with C. albicans (Fig. (Fig.3).3). However, at 70 days after infection, phagocytosis was decreased (P < 0.01) in the Tg compared to the non-Tg mice, but only after 30 min of incubation and at a high effector/target cell ratio (2:1). In contrast, phagocytosis was enhanced (P < 0.05) in the Tg mice at a low effector/target cell ratio (1:5) at this same time of incubation. The mean fluorescence of PMNs with phagocytosed C. albicans did not differ (P > 0.05) in these infected or uninfected Tg and non-Tg mice, indicating that numbers of endocytosed C. albicans organisms per phagocytosing PMN were comparable in the four groups of mice (data not shown).
Killing of C. albicans by PMNs from the Tg mice was modestly reduced (P < 0.05) in comparison to that by PMNs from the non-Tg mice at 7 days after infection. At 70 days after infection, however, diminished killing by PMNs from the Tg mice did not reach statistical significance (P = 0.11).
Depletion of circulating PMNs beginning on day 20, 45, or 63 after oral infection with C. albicans did not alter the level of chronic oral carriage of C. albicans in these Tg mice over the resulting 7-day period of profound neutropenia (P > 0.05) (Fig. (Fig.4).4). In addition, depletion of PMNs in the non-Tg mice did not produce a relapse of oral carriage of C. albicans, which was cleared from the oral cavities of these control mice within 10 days after oral inoculation ((13) (Fig. (Fig.4).4). Mice assessed after 7 days of neutropenia showed an absence of systemic dissemination of C. albicans, except for minimal spread to lungs and liver in two of eight Tg mice with advanced AIDS-like disease assessed 70 days after oral infection, and this depletion of PMNs did not modify the burdens of C. albicans in the gastrointestinal tracts of the Tg mice (Table (Table1).1). All mice survived until final assessment 7 days after induction of neutropenia. These results demonstrated that PMNs are dispensable for limiting C. albicans proliferation in the oral mucosa and for preventing systemic dissemination of the fungus in these Tg mice.
Throughout the chronic carrier phase (days 27 to 114 after infection), oral burdens were significantly elevated in CD4C/HIVMutG CD8−/− compared to control CD4C/HIVMutG mice (P = 0.01) (Fig. (Fig.5).5). In contrast, C. albicans was rapidly cleared from oral cavities of both CD8−/− and control non-Tg mice immediately after primary infection (P < 0.05 compared to control CD4C/HIVMutG and CD4C/HIVMutG CD8−/− mice). Augmentation of oral burdens of C. albicans in CD8−/− mice therefore occurred in animals which also express the CD4C/HIVMutG transgene but not in control mice which do not express the transgene. Sustained enhancement of infection in these CD4C/HIVMutG mice was comparable to our previous observations in CD4C/HIVMutA mice (13), indicating that the nef gene is necessary and sufficient for persistent candidal infection.
In addition to the oral cavity, burdens of C. albicans were augmented (P < 0.05) in the stomachs of CD8−/− CD4C/HIVMutG mice compared to control CD4C/HIVMutG Tg mice at final assessment of these animals (Table (Table2).2). The limited systemic dissemination of C. albicans in these MutG Tg mice was similar to that previously observed during the later stage of infection in CD4C/HIVMutA Tg mice (13) (Table (Table2).2). Taken together, these results indicated that CD8+ T cells directly or indirectly limit chronic oral and gastric carriage of C. albicans in these Tg mice.
Flow cytometry analysis confirmed the absence of CD8+ T cells in the peripheral blood and CLNs of CD8−/− CD4C/HIVMutG Tg and CD8−/− non-Tg mice and demonstrated a depletion of this cell population in the peripheral blood (P < 0.05) but its augmentation in CLNs (P < 0.05) of CD4C/HIVMutG compared to non-Tg mice (Table (Table3).3). Heterozygous CD8+/− CD4C/HIVMutG Tg mice had unaltered CD8+ T-cell frequencies in peripheral blood (mean ± standard deviation [SD], 5.1 ± 1.1 [five mice]) and CLNs (33.8 ± 7.8) and had no augmentation of oral burdens of C. albicans in comparison to control CD4C/HIVMutG mice. The proportions of CLN naive (CD44−/lo CD62L+) and memory-like (CD44hi CD62lo/−) CD8+ T cells were comparable in these CD4C/HIVMutG Tg mice and non-Tg animals orally infected with C. albicans (data not shown).
The ability of C. albicans to infect the oral and esophageal epithelia but yet to rarely disseminate to deep organs of HIV-infected patients suggests that partially or completely preserved host defense mechanisms effectively restrict proliferation of the fungus to the superficial mucosa. Studies conducted with congenitally immunodeficient mice have firmly established that a defect in the protective Th1 CD4+ T-cell response to C. albicans results in mucosal candidiasis but that an added perturbation of PMNs and/or macrophages is required for systemic dissemination (1, 2, 6, 7, 9, 29, 30, 43). The role of PMNs in host defense was therefore investigated in Tg mice expressing gene products of HIV-1 and which develop an AIDS-like disease (13, 26, 27). The absolute numbers and the oxidative burst of circulating PMNs were augmented in the Tg mice and further amplified by chronic carriage of C. albicans, indicating that these properties of the PMN response to candidal infection are quantitatively and functionally unimpaired in these Tg mice. Despite a striking reduction of CD4+ T cells in the peripheral blood and CLNs of the Tg mice (13, 27), profound depletion of PMNs sustained over 7 days did not quantitatively alter chronic oral carriage or lead to systemic dissemination of C. albicans, except for minimal spread in Tg mice with advanced AIDS-like disease. Accordingly, PMNs not only are essentially unimpaired but also are dispensable for control of mucosal and systemic candidiasis in these Tg mice, despite their established anti-Candida properties in vivo and in vitro (12, 20, 29, 30). In previous studies, depletion of PMNs in both normal (20) and SCID (29, 30) mice augmented the severity of oral infection with C. albicans and resulted in disseminated candidiasis in the SCID mice (29, 30). However, depletion of PMNs in the normal mice was initiated prior to self-limited primary infection with C. albicans (20), characterized by a marked transient influx of neutrophils to the oral mucosa (10, 19), and these studies did not examine the role of these cells during chronic carriage in a T-cell-defective host. The T- and B-cell-deficient SCID mice differ significantly from the Tg mice by the absence of a humoral response and were monoassociated with C. albicans in a germfree environment (29, 30), and they were therefore devoid of a protective bacterial flora (31). At the time of euthanasia, the PMN-depleted Tg mice were heavily and uniformly colonized by C. albicans throughout the digestive tract, demonstrating that these mice were indeed uniformly challenged by C. albicans at this portal of entry for systemic dissemination. Interestingly, a nonprotective role of PMNs has also been surmised for women who received an intravaginal Candida challenge, in whom protection against infection was noninflammatory while symptomatic infection correlated with a vaginal infiltration of PMNs and a high vaginal C. albicans burden (22).
In several investigations producing conflicting results, the oxidative burst of PMNs from HIV-infected patients has been found to be unchanged (49), increased (4, 16, 45), or decreased (11, 34, 37, 48) compared to that of PMNs from patients uninfected with HIV. Likewise, growth inhibition of C. albicans by PMNs was found to be preserved in HIV infection (8), while phagocytosis and killing of C. albicans by PMNs were determined to be either unchanged, impaired, or increased (4, 17, 33, 34, 45, 49). The present findings in the Tg mice demonstrate an augmentation in the oxidative burst of PMNs. Because the HIV-1 transgene is not expressed in PMNs (27), this increase may have been caused by altered expression of cytokines with activating (gamma interferon/tumor necrosis factor alpha/interleukin-2 [IL-2]) or deactivating (IL-4/IL-10) signals to effector phagocytes (44). The decreased phagocytosis of C. albicans by PMNs from the Tg mice at 70 days after infection and after prolonged incubation at a high E/T cell ratio may have resulted from the suppressive effect of IL-10 (41). Production of this cytokine by CD4+ T cells is enhanced in these Tg mice (D. Lewandowski, M. Marquis, F. Aumont, A.-C. Lussier-Morin, M. Raymond, S. Sénéchal, Z. Hanna, P. Jolicoeur, and L. de Repentigny, submitted for publication). However, PMNs from the Tg mice maintained a normal capacity to kill C. albicans at this late stage of AIDS-like disease, demonstrating that chronic oral candidiasis in these Tg mice cannot be explained by a functionally significant defect of PMNs against C. albicans.
Several lines of evidence have suggested a protective role of CD8+ T cells against oral candidiasis in HIV-infection. IL-2-activated CD8+ T cells exert direct growth inhibition against the hyphal form of C. albicans in vitro (5). Progressive depletion of CD8+ T cells in HIV infection results from apoptosis mediated by macrophages through interaction of HIV gp120 with chemokine receptor CXCR4 (28). In addition, Nef induces caspase-8-mediated apoptosis of CD8+ T cells by upregulating dendritic cell expression of tumor necrosis factor alpha and FasL (38). Despite their diminution in absolute numbers, remaining CD8+ T cells nevertheless successfully accumulate in the basal epithelial layer of the oral mucosa of HIV-infected patients with OPC (35, 42). However, CD8+ T cells may not be in proximity to C. albicans hyphae, which are usually confined to the upper layers of the epithelium (18, 39). Accordingly, the precise role of CD8+ T cells in mucosal containment of C. albicans in HIV infection, either by direct growth inhibition of Candida or, more likely, by production of cytokines which enhance the antimicrobial activity of macrophages and PMNs against C. albicans, has so far remained unclear. Immunohistochemical analysis has revealed an influx of CD8+ T cells in the basal layer of the oral epithelium of Tg mice orally infected with C. albicans, both immediately after recovery from primary infection (M. Marquis and L. de Repentigny, unpublished data) and late in the chronic carrier phase (13). In contrast, CD8+ T cells were not detected in uninfected Tg or non-Tg mice (13), consistent with their known exceedingly low frequency in normal murine oral mucosa (10, 15). In addition, flow cytometry analysis demonstrated a significant augmentation of oral mucosal CD8+ T cells in infected Tg compared to uninfected non-Tg animals both early and late in AIDS-like disease, again showing that the Tg mice recruit this cell population to the oral mucosa in response to candidal infection (Lewandowski et al., submitted). This increase of CD8+ T cells was also demonstrated in the CLNs of the Tg compared to the non-Tg mice orally infected with C. albicans. A role for CD8+ T cells in limiting oral proliferation of C. albicans was demonstrated by enhanced oral burdens in CD4C/HIVMutG CD8−/− mice compared to control CD4C/HIVMutG Tg mice throughout the chronic carrier state, in striking contrast to the rapid clearance of C. albicans from the oral cavities of both CD8−/− and control non-Tg mice immediately after primary infection. These results suggest that CD8+ T cells become critical to the host defense against C. albicans only when CD4+ and/or antigen-presenting cells are perturbed by expression of the HIV nef gene in CD4C/HIVMutG Tg mice (27). In previous investigations, depletion of CD8+ T cells in immunocompetent BALB/C or CBA/CaH mice did not alter the clearance of primary oral infection with C. albicans (20), and adoptive transfer of naive CD8+ T cells to BALB/C nu/nu mice resulted in only a modest and transient decrease in oral colonization (19). Likewise, β2-microglobulin knockout mice, which lack major histocompatibility complex class I expression and are deficient in CD8+ T cells, were susceptible to systemic candidiasis of endogenous origin but showed only superficial and transient infection of tongues and esophagi after monoassociation with C. albicans (3). The present results thus clearly demonstrate for the first time that CD8+ T cells participate in the host defense against oral candidiasis in vivo, specifically in the context of nef expression in a subset of immune cells.
This work was supported by the Canadian Institutes of Health Research HIV/AIDS Research Program (grant HOP-41544). Miriam Marquis and Daniel Lewandowski are recipients of studentship awards from the Medical Mycology Research Fund of the University of Montreal and the University of Montreal, respectively.
We thank Marie-Andrée Laniel for support in maintaining the transgenic mouse colony; Miguel Chagnon, Michel Lamoureux, and Yves Lepage for statistical analysis; and Sylvie Julien for manuscript preparation.
Editor: A. Casadevall