Increased Host Resistance against Pneumocystis carinii Pneumonia in γδ T-Cell-Deficient Mice: Protective Role of Gamma Interferon and CD8⁺ T Cells

Abstract

Although a clear relationship between αβ T-cell receptor-positive (αβ-TCR⁺) CD4⁺ T cells and susceptibility to Pneumocystis carinii infection exists, the role of other T-cell subsets is less clearly defined. Previous studies have shown that γδ-TCR⁺ T cells infiltrate into the lung during P. carinii pneumonia. Therefore, the present study examined the role of γδ-TCR⁺ T cells in host defense against P. carinii pneumonia. C57BL/6 (control) and B6.129P2-Tcrd^tm1Mom (γδ-TCR⁺ T-cell-deficient) mice were inoculated intratracheally with P. carinii. At specific time points, mice were sacrificed and analyzed for P. carinii burden, T-cell subsets, and cytokine levels in lung tissue. Analysis of P. carinii burden showed a more rapid and complete resolution of infection in γδ-TCR⁺ T-cell-deficient mice than in C57BL/6 controls. This augmented resolution was associated with elevated gamma interferon (IFN-γ) levels in bronchoalveolar lavage fluid predominantly produced by CD8⁺ T cells, as well as an increased recruitment of CD8⁺ T cells in general. In separate experiments, neutralization of IFN-γ or depletion of CD8⁺ T cells early during infection abolished the augmented resolution previously observed in γδ-TCR⁺ T-cell-deficient mice. These results show that the presence of γδ-TCR⁺ T cells modulates host susceptibility to P. carinii pneumonia through interactions with pulmonary CD8⁺ T cells and tissue production of IFN-γ.

Despite the advent of highly active antiretroviral therapy, pulmonary infection with Pneumocystis carinii remains a common opportunistic infection in human immunodeficiency virus-positive (HIV⁺) individuals (12). Clinical observations have shown that individuals with systemic CD4 cell counts of less than 200 cells/μl, estimated at more than 180,000 people in the United States alone, are the most susceptible to P. carinii infections (14). Although P. carinii prophylaxis has become widespread, more than 80% of all new cases are breakthrough cases (30). Despite the clear correlation between CD4⁺ T cells and P. carinii infection (29), the role of other T-cell subsets and associated cytokines in protection against infection is less clearly defined.

The role of γδ T-cell receptor-positive (γδ-TCR⁺) T cells during P. carinii infection is not well characterized. Clinical observations have shown that γδ-TCR⁺ T cells are elevated in blood and bronchoalveolar lavage (BAL) fluid of individuals with AIDS and P. carinii pneumonia (1, 13), although the functional aspects of γδ-TCR⁺ T cells in these studies were not studied. Data from prior experimental studies have shown that γδ-TCR⁺ T-cell-deficient mice are not permissive for long-term P. carinii infection (10), while another study suggests that γδ-TCR⁺ T cells provide residual host defense functions in αβ-TCR⁺ T-cell-deficient mice (9). In other models of infection, γδ-TCR⁺ T cells have varying roles. A protective role for γδ-TCR⁺ T cells has been reported in experimental models of Toxoplasma, Nocardia, and Klebsiella infections (15, 17, 23), while other infections, such as Chlamydia and Salmonella infections, have been shown to be less severe when γδ-TCR⁺ T cells are absent (7, 31).

In the present study, we determined whether mice deficient in γδ-TCR⁺ T cells through gene deletion were susceptible to experimental P. carinii pneumonia, and we characterized the local immune response to P. carinii. Our results show that the absence of γδ-TCR⁺ T cells in experimental P. carinii infection regulates CD8⁺ T-cell recruitment and tissue production of gamma interferon (IFN-γ) during pneumonia which, in this model, results in increased host resistance to infection.

MATERIALS AND METHODS

Mice.

Male B6.129P2-Tcrd^tm1Mom (γδ-TCR⁺ T-cell-deficient) mice backcrossed to a C57BL/6 background for 12 generations and control C57BL/6 mice, 6 to 8 weeks of age, were purchased from Jackson Laboratory (Bar Harbor, Maine). A colony of C57BL/6J-Tnfsf5^tm1Imx mice (CD40 ligand knockout mice; Jackson Laboratory) backcrossed to C57BL/6 are maintained at the Louisiana State University Health Sciences Center as a reservoir of P. carinii. All animals were housed in a specific-pathogen-free facility and handled according to institutionally recommended guidelines.

P. carinii inoculation.

The P. carinii inoculum was prepared as previously described (18, 27). Briefly, CD40 ligand knockout mice previously inoculated with P. carinii were injected with a lethal dose of pentobarbital, and the lungs were aseptically removed and frozen for 30 min at −70°C in 1 ml of phosphate-buffered saline (PBS). Frozen lungs were thereafter homogenized in 10 ml of PBS (model 80 stomacher; Tekmar Instruments, Cincinnati, Ohio), filtered through sterile gauze, and pelleted at 500 × g for 10 min at 4°C. The pellet was resuspended in PBS, and the number of P. carinii cysts was determined microscopically by using a modified Giemsa stain (Diff-Quik; Baxter, McGaw Park, Ill.). The inoculum was adjusted to 2 × 10⁶ cysts/ml. Gram stains were performed on the inoculum preparation to exclude contamination with bacteria.

BAL and lung RNA isolation.

Cells and fluid from the lower respiratory tract were obtained by BAL of mice anesthetized with intraperitoneal pentobarbital as previously described (18). Briefly, lungs were lavaged through an intratracheal catheter with prewarmed (37°C) calcium and magnesium-free PBS supplemented with 0.6 mM EDTA. A total of 11 ml was used in each mouse in 1-ml increments with a 30-s dwell time. The first milliliter was processed at 500 × g and the supernatant was stored at −70°C until use. The remaining cell pellet and the other 10 ml of lavage fluid were pooled and centrifuged at 800 × g for 10 min, and the cells were collected for flow cytometry. Thereafter, the right lung was excised and homogenized in 1 ml of TRIzol for total lung RNA isolation and stored at −70°C.

Real-time PCR analysis of P. carinii infection.

Total RNA was isolated from the right lung of infected mice by a single-step method using TRIzol reagent (Life Technologies) as per the manufacturer's instructions. Thereafter, RNA was transcribed to cDNA and real-time PCR was performed as previously described (33). This assay has a correlation coefficient of 0.98 over 8 logs of P. carinii RNA concentration and correlates with viable P. carinii, since both heat killing and exposure to trimethoprim-sulfamethoxazole eliminate the signal.

ELISA.

BAL fluid was assayed for IFN-γ, tumor necrosis factor alpha (TNF-α), interleukin-4 (IL-4), and IL-10 by enzyme-linked immunosorbent assay (ELISA) using commercially available capture and biotinylated detection antibodies (BD Pharmingen Corp., San Diego, Calif.). Standard curves were generated using the respective recombinant murine cytokines. The assays were performed in high-protein-binding EIA/A2 96-well tissue culture plates (Costar, Corning, N.Y.). Briefly, the plates were coated with 50 μl of the capture antibody (1 μg/ml) diluted in coating buffer (0.1 M NaHCO₃) and incubated overnight at 4°C. The plates were washed four times with PBS containing 0.05% Tween 80 (wash buffer). Nonspecific binding sites were blocked with PBS containing 10% fetal calf serum and 0.05% Tween 80 (blocking buffer) in a volume of 100 μl for 2 h at room temperature. After washing the plates twice with wash buffer, the standards and samples were added to each well in triplicate in a volume of 50 μl and incubated at room temperature for 2 h. After incubation, the plates were washed four times with wash buffer and the biotinylated detection antibody (1 μg/ml) was added to each well and incubated for 1 h at room temperature. After washing four times with wash buffer, a solution of horseradish peroxidase (diluted 1:250) was added for 1 h at room temperature, followed by washing an additional four times with wash buffer. Finally, the substrate o-phenylenediamine dihydrochloride (Sigma, St. Louis, Mo.) was added. The absorbance values and concentrations of each cytokine were determined using a μQuant automated microplate reader (Bio-Tek Corp., Wisnooski, Vt.) and Kineticalc software (Bio-Tek). Data were expressed as picograms of cytokine per milliliter.

IFN-γ secretion assay.

BAL cells were collected as described above. Spontaneous or stimulated cellular secretion of IFN-γ was analyzed with an IFN-γ secretion assay kit (Miltenyi Biotec, Auburn, Calif.) per the manufacturer's instructions. BAL cells were cultured for 24 h at 37°C, 5% CO₂ in medium alone (spontaneous production) or with concanavalin A (0.5 μg/ml; stimulated production). At the conclusion of the assay, BAL cells were either incubated with optimal dilutions of fluorescein isothiocyanate (FITC)-conjugated anti-CD4 (clone RM4-4) or anti-CD8 (clone 53-6.7) antibodies (both from Pharmingen) for 20 min at room temperature. Thereafter, 10,000 cells were analyzed with a FACSCalibur flow cytometer (Becton Dickinson, Franklin Lakes, N.J.). Gates were set for lymphocytes in the analyzed cell population using forward and right-angle light scatter, and the percentage of dual-positive cells was calculated. Controls included isotype-matched control antibodies conjugated to identical fluorochromes. To ensure accurate enumeration of activated lymphocytes which may have altered forward and side-scatter profiles, stained samples were back-gated against the forward and side-scatter profiles. Data are expressed as the percentage of dual-positive cells.

Flow cytometry.

BAL cells were enumerated on a hemacytometer using trypan blue dye exclusion. BAL cells (10⁵ to 10⁶/ml) were incubated for 20 min at room temperature with optimal dilutions of anti-murine Fc receptor (Pharmingen) to block nonspecific binding. BAL cells were thereafter incubated with optimal dilutions of FITC-conjugated anti-CD3 (clone 145-2C11) and either phycoerythrin-conjugated anti-CD4 (clone RM4-4) or anti-CD8 (clone 53-6.7) antibodies (all from Pharmingen) for 45 min at room temperature. γδ⁺ T cells were analyzed by staining with FITC-conjugated GL3 (Pharmingen). After washing, cells were fixed with 1% paraformaldehyde, and 10,000 cells were analyzed with a FACSCalibur flow cytometer (Becton Dickinson). Gates were set for lymphocytes in the analyzed cell population using forward and right-angle light scatter, and the percentage of dual-positive cells was calculated. Controls included isotype-matched control antibodies conjugated to identical fluorochromes. To ensure accurate enumeration of activated lymphocytes which may have altered forward and side-scatter profiles, stained samples were back-gated against the forward and side-scatter profiles. Data are expressed as total CD4⁺ and CD8⁺ T cells recovered.

Immunofluorescent staining.

Lungs were excised and fixed in 4% paraformaldehyde and thereafter placed in a 30% sucrose solution overnight at 4°C. Lungs were then placed in optimal cutting temperature medium (OCT; Sakrura Finetek, Torrance, Calif.), and 7-μm lung sections were cut using a cryostat. Sections were stained with purified anti-murine CD4 or CD8 antibodies (R & D Systems, Minneapolis, Minn.) followed by Alexa-fluor 488-conjugated goat anti-mouse immunoglobulin G (Molecular Probes, Eugene, Oreg.). Control sections employed isotype-matched antibodies. Sections were counterstained with 4′,6-diamidino-2-phenylindole, dihydrochloride (DAPI; Molecular Probes) nucleic acid stain (0.4 μg/ml; 2 min at room temperature) followed by application of Prolong mounting medium (Molecular Probes) and a coverslip. Sections were analyzed on a Leica DMRXA automated upright epifluorescence microscope equipped with Slidebook Deconvolution software (3I, Denver, Colo.).

In vivo CD8 depletion and IFN-γ neutralization.

Depletion of CD8⁺ T cells was performed by weekly intraperitoneal injections of an anti-murine CD8 depleting antibody (0.3 mg/ml) (clone 2.43; American Type Culture Collection, Rockville, Md.) or isotype control antibodies beginning 3 days prior to P. carinii inoculation. Depletion was confirmed through flow cytometric analysis of splenic CD8⁺ T cells from control and depleted mice. IFN-γ was neutralized by administration of immune rabbit serum containing antibodies against murine IFN-γ (1 mg/ml; kindly provided by Nick Lukacs, Department of Pathology, University of Michigan School of Medicine) every other day for 1 week beginning 2 days postinoculation with P. carinii (11). Control mice received naïve rabbit serum. Verification of IFN-γ neutralization was evaluated through spiking recombinant murine IFN-γ in BAL fluid from treated animals and determining the amount of excess neutralization through an ELISA.

Statistical analysis.

Data were analyzed using StatView statistical software (Brainpower Inc., Calabasas, Calif.). Comparisons between groups where data were normally distributed were made with Student's t test, and comparisons among multiple groups or nonparametric data were made with analyses of variance and appropriate follow-up testing. The Mann-Whitney test or the Wilcoxon paired-sample test was employed to make ordinal comparisons. Significance was accepted at a P value of <0.05.

RESULTS

Experimental P. carinii pneumonia in γδ-TCR⁺ T-cell-deficient mice.

Our first set of studies characterized recruitment of γδ-TCR⁺ T cells into the lungs during P. carinii infection. For these studies, C57BL/6 control or γδ-TCR⁺ T-cell-deficient mice were challenged intratracheally with P. carinii. At specific time points, mice were sacrificed and γδ-TCR⁺ T cells were enumerated in BAL fluid by flow cytometry. γδ-TCR⁺ T cells were not detected in uninfected control mice but increased progressively through 14 days postinoculation (Fig. 1A) and persisted up to 28 days postinoculation in P. carinii-infected animals. No γδ-TCR⁺ cells were observed in BAL fluid from infected γδ-TCR⁺ T-cell-deficient mice (data not shown). Next, γδ-TCR⁺ T-cell-deficient mice and control mice were challenged intratracheally with P. carinii and at specific time points they were sacrificed and analyzed for P. carinii lung burden by real-time PCR. To ensure that control and γδ-TCR⁺ T-cell-deficient mice received similar levels of P. carinii, we sacrificed mice immediately following intratracheal instillation of P. carinii organisms and analyzed P. carinii lung burden. As illustrated in Fig. 1B, both control and γδ-TCR⁺ T-cell-deficient mice received identical amounts of P. carinii organisms (day 0). Initially, approximately 80 and 50% of the P. carinii inoculum was cleared from the lungs by day 3 in control and γδ-TCR⁺ T-cell-deficient mice, respectively. Thereafter, control mice demonstrated similar kinetics of infection as previously reported in BALB/c mice (4), with increasing organism burdens at 3 and 7 days postinoculation, reaching a maximum at 14 days postinoculation followed by near resolution by 28 days postinoculation. In contrast, analysis of P. carinii organism burden in γδ-TCR⁺ T-cell-deficient mice demonstrated similar kinetics of infection at days 3 and 7 postinoculation compared to control mice but had significantly lower levels of infection at 14 and 28 days postinoculation, consistent with increased resistance. (27). By 28 days postinoculation, γδ-TCR⁺ T-cell-deficient mice had a P. carinii lung burden 2 logs lower than that of control mice (7.8 × 10⁵ ± 3.5 × 10⁵ copies [mean ± standard error of the mean] and 6.3 × 10³ ± 3.4 × 10³ copies for control and γδ-TCR⁺ T-cell-deficient mice, respectively; P < 0.05). Moreover, only 25% of γδ-TCR⁺ T-cell-deficient mice had detectable P. carinii organisms by real-time PCR, whereas over 70% of control mice were still infected above a threshold value of 10⁵ rRNA copies, the level required to observe organisms histologically using traditional Gomori-methanamine-silver staining. Thus, γδ-TCR⁺ T-cell-deficient mice have a more rapid and more complete resolution of experimental P. carinii pneumonia than control mice.

FIG. 1.

Experimental P. carinii pneumonia in γδ-TCR⁺ T-cell-deficient mice. Male B6.129P2-Tcrd^tm1Mom (Tcr-d KO) and C57BL/6 mice, 6 to 8 weeks of age, were intratracheally inoculated with 2 × 10⁵ P. carinii cysts. At specific time points, mice were sacrificed and analyzed for P. carinii lung burden by real-time PCR. The figure illustrates representative data from three separate experiments (n = 5 to 7 animals/time point) for γδ-TCR⁺ T-cell recruitment at 7, 14, and 28 days postinoculation (A) and P. carinii lung burden at 3, 7, 14, and 28 days postinoculation (B). Asterisks represent significant differences between control and γδ-TCR⁺ T-cell-deficient mice (P < 0.05). Data are expressed as values of pooled lavage samples from individual mice (A) or as the mean ± standard error of the mean (B).

Local T-cell recruitment in γδ-TCR⁺ T-cell-deficient mice during experimental P. carinii pneumonia.

Since γδ-TCR⁺ T-cell-deficient mice demonstrated augmented clearance of P. carinii, studies were performed to investigate whether the presence or absence of γδ-TCR⁺ T cells influenced CD4⁺ and CD8⁺ T-cell recruitment into BAL fluid in P. carinii-infected mice. Immunocompetent control mice recruited both CD4⁺ and CD8⁺ T cells by 7 days postinoculation, and both subsets continued to increase throughout the duration of the infection (Fig. 2A). In contrast, γδ-TCR⁺ T-cell-deficient mice recruited approximately 25% more CD4⁺ T cells and approximately 58% more CD8⁺ T cells at 7 days postinoculation (Fig. 2A) (P < 0.05). Both T-cell subsets in γδ-TCR⁺ T-cell-deficient mice either reached a plateau (CD4) or peaked (CD8) at 7 days postinoculation and then decreased throughout the remainder of the infection. Interestingly, levels of CD4⁺ or CD8⁺ T cells in control mice did not reach the levels observed in γδ-TCR⁺ T-cell-deficient mice until 14 days postinoculation (CD4) or not at all (CD8). The rapid recruitment of CD4⁺ and CD8⁺ T cells early during infection in γδ-TCR⁺ T-cell-deficient mice correlated with a lower peak of P. carinii burden and enhanced clearance of P. carinii from the lungs (Fig. 1B). In contrast, the delayed recruitment of CD4⁺ and CD8⁺ T cells in control mice correlated with a higher peak of organism burden and decreased pulmonary clearance of P. carinii (Fig. 1B).

FIG. 2.

Local T-cell recruitment in γδ-TCR⁺ T-cell-deficient mice during experimental P. carinii pneumonia. Male B6.129P2-Tcrd^tm1Mom (Tcr-d KO) and C57BL/6 mice, 6 to 8 weeks of age, were intratracheally inoculated with 2 × 10⁵ P. carinii cysts. At specific time points, mice were sacrificed and the lungs were lavaged. CD4⁺/CD3⁺ and CD8⁺/CD3⁺ BAL cells were analyzed and enumerated by flow cytometry. The figure illustrates representative data from three separate experiments (n = 5 to 7 animals/time point) for 7, 14, and 28 days postinoculation (A) Data are expressed as values of pooled lavage samples from individual mice. (B) Representative lung sections from control and γδ-TCR⁺ T-cell-deficient mice stained with anti-murine CD4 or CD8 (green) and the nuclear counterstain DAPI (blue).

To investigate whether increased numbers of CD4⁺ and CD8⁺ T cells in the BAL fluid in γδ-TCR⁺ T-cell-deficient mice correlated with CD8⁺ T cells in lung tissue, lungs were isolated at 7 days postinoculation from control and γδ-TCR⁺ T-cell-deficient mice and frozen sections were immunostained for CD4⁺ and CD8⁺ T cells. The data in Fig. 2B are from a representative lung section and show significantly more CD8⁺ T cells in the lung tissue from γδ-TCR⁺ T-cell-deficient mice than in control mice. There was no visible difference in CD4⁺ T-cell levels in lung tissue between control and γδ-TCR⁺ T-cell-deficient mice. Thus, augmented clearance of P. carinii was more closely associated with augmented CD8⁺ T-cell recruitment in γδ-TCR⁺ T-cell-deficient mice and to a lesser extent with CD4⁺ T-cell recruitment.

Local cytokine concentrations in γδ-TCR⁺ T-cell-deficient mice during experimental P. carinii pneumonia.

Although IFN-γ is not essential for host defense against P. carinii in normal mice, it has a therapeutic effect in CD4-depleted mice (2, 18) and its role in P. carinii-infected γδ-TCR⁺ T-cell-deficient mice has not been elucidated. γδ-TCR⁺ T-cell-deficient mice had significantly elevated levels of IFN-γ in lung lavage fluid early during infection compared to immunocompetent control mice (Fig. 3A) (P < 0.01). This increase in IFN-γ concentration correlated with augmented CD8⁺ T-cell recruitment (Fig. 2) and enhanced resolution of infection in γδ-TCR⁺ T-cell-deficient mice (Fig. 1B). To determine whether additional cytokines which have been shown to be critical for host defense against P. carinii, such as TNF-α (5, 19), were elevated, we determined cytokine concentrations in the BAL fluid of γδ-TCR⁺ T-cell-deficient or control mice 7 days postinoculation. The proinflammatory cytokine TNF-α and the T-helper type 2 (Th2) cytokines IL-4 and IL-10 were evaluated. In contrast to IFN-γ, there were no differences in these cytokine concentrations in BAL fluid between control and γδ-TCR⁺ T-cell-deficient mice (TNF-α, 22.3 ± 3.6 versus 18.1 ± 2.5 pg/ml in control and γδ-TCR⁺ T-cell-deficient mice, respectively; IL-4, 26.9 ± 9.2 vs. 28.0 ± 3.7 pg/ml for control and γδ-TCR⁺ T-cell-deficient mice, respectively; IL-10, 46.5 ± 33.1 vs. 48.7 ± 9.8 pg/ml for control and γδ-TCR⁺ T-cell-deficient mice, respectively).

FIG. 3.

Local cytokine concentrations in γδ-TCR⁺ T-cell-deficient mice during experimental P. carinii pneumonia. Male B6.129P2-Tcrd^tm1Mom (Tcr-d KO) and C57BL/6 mice, 6 to 8 weeks of age, were intratracheally inoculated with 2 × 10⁵ P. carinii cysts. At specific time points, mice were sacrificed and the lungs were lavaged. Concentrations of IFN-γ (A) were measured in BAL fluid by using ELISA. The results are representative data from three separate experiments (n = 5 to 7 animals/time point). (B) Spontaneous (medium alone) and concanavalin A-stimulated IFN-γ production by CD4⁺ and CD8⁺ T cells in BAL as determined by an IFN-γ secretion assay. The data are representative of two separate experiments (n = 3 for each experimental group). Asterisks represent significant differences between control and γδ-TCR⁺ T-cell-deficient mice (P < 0.05). Data are expressed as the mean ± standard error of the mean (A) or as values from pooled lavage samples (B).

Studies were thereafter conducted to determine the cellular source of IFN-γ in γδ-TCR⁺ T-cell-deficient animals. An IFN-γ secretion assay (see Materials and Methods) was performed on BAL cells collected from control and γδ-TCR⁺ T-cell-deficient mice at 7 days postinoculation. Results showed that CD8⁺ T cells from γδ-TCR⁺ T-cell-deficient mice were the predominant cellular source of both spontaneous and induced IFN-γ (Fig. 3B).

Neutralization of IFN-γ.

Based on the observation that γδ-TCR⁺ T-cell-deficient mice had elevated concentrations of IFN-γ early during P. carinii infection, we investigated the role of IFN-γ in vivo in the immune response to P. carinii. Immunocompetent and γδ-TCR⁺ T-cell-deficient mice were challenged intratracheally with P. carinii and thereafter administered naïve or immune rabbit serum containing a polyclonal antibody against murine IFN-γ every other day beginning 2 days postinoculation (three times total). On days 7 and 14 postinoculation, mice were sacrificed and analyzed for P. carinii lung burden and CD4⁺ and CD8⁺ T-cell subsets. Neutralization of IFN-γ early during P. carinii infection resulted in inhibited resolution of infection in γδ-TCR⁺ T-cell-deficient mice (Fig. 4A) (P = 0.002 between control and γδ-TCR⁺ T-cell-deficient mice that received naïve sera, and P = 0.022 between γδ-TCR⁺ T-cell-deficient mice that received naïve and immune sera). IFN-γ neutralization in control mice had no effect on P. carinii burden, consistent with prior data published on IFN-γ gene-disrupted mice (8). Excess neutralizing antibody was documented in BAL fluid of treated animals at 7 and 14 days postinoculation by spiking recombinant murine IFN-γ in BAL fluid followed by ELISA (data not shown). Analysis of CD4⁺ and CD8⁺ T-cell subsets at 7 days postinoculation showed that neutralization of IFN-γ reduced both subsets of T cells in control and deficient mice (Fig. 4B). However, inhibition of augmented clearance of P. carinii in anti-murine IFN-γ-treated γδ-TCR⁺ T-cell-deficient mice was associated with a greater than 98% reduction in recruitment of CD8⁺ T cells (Fig. 4B).

FIG. 4.

Effects of IFN-γ neutralization on host defense against P. carinii in γδ-TCR⁺ T-cell-deficient mice. Male B6.129P2-Tcrd^tm1Mom (Tcr-d KO) and C57BL/6 mice, 6 to 8 weeks of age, were intratracheally inoculated with 2 × 10⁵ P. carinii cysts. IFN-γ was neutralized through administration of immune rabbit serum containing antibodies against murine IFN-γ every other day for 1 week beginning 2 days postinoculation with P. carinii. Control mice received naïve rabbit serum. At specific time points, mice were sacrificed and analyzed for P. carinii lung burden and T-cell subsets. The figure illustrates representative data from three separate experiments (n = 5 to 7 animals/time point) for P. carinii burden (A) and for CD4⁺/CD3⁺ and CD8⁺/CD3⁺ BAL cell subsets (B). A single asterisk represents significant differences between control and γδ-TCR⁺ T-cell-deficient mice that received naïve rabbit serum, and double asterisks represent significant differences between γδ-TCR⁺ T-cell-deficient mice that received naïve and immune serum (P < 0.05). Data are expressed as the mean ± the standard error of the mean (A) and as values of pooled lavage samples from individual mice (B).

Effects of CD8⁺ T-cell depletion on experimental P. carinii pneumonia in γδ-TCR⁺ T-cell-deficient mice.

Although there is a clear relationship between CD4⁺ T cells and the ability to clear P. carinii (27), the role of CD8⁺ T cells is less clear but appears to play some role in host defense (3). Moreover, in the setting of high levels of IFN-γ in the alveolar space, CD8⁺ T cells can affect clearance of P. carinii in the absence of CD4⁺ T cells (18). Based on these data, the observations that CD8⁺ T-cell recruitment is significantly augmented in γδ-TCR⁺ T-cell-deficient mice (Fig. 2) and that neutralization of IFN-γ virtually eliminated both CD8⁺ T-cell recruitment and the resistance phenotype in γδ-TCR⁺ T-cell-deficient mice (Fig. 4B), we investigated the role of CD8⁺ T cells in γδ-TCR⁺ T-cell-deficient mice during experimental P. carinii infection. Control mice and γδ-TCR⁺ T-cell-deficient mice were administered an anti-CD8 depleting antibody 3 days prior to intratracheal challenge with P. carinii, with weekly injections thereafter. Organism burden was assessed at 14 days, which is the time point when γδ-TCR⁺ T-cell-deficient mice are resolving the infection and control mice have their peak intensity of infection (Fig. 1B). Depletion of CD8⁺ T cells resulted in significant inhibition of the augmented clearance of P. carinii infection in γδ-TCR⁺ T-cell-deficient mice (Fig. 5) (P < 0.0001 between CD8-depleted γδ-TCR⁺ T-cell-deficient mice and nondepleted γδ-TCR⁺ T-cell-deficient mice). Depletion of CD8⁺ T cells was verified through analysis of splenocytes from isotype and anti-CD8-treated animals (7 days postinoculation, 11% and <1% for isotype-matched and anti-CD8 antibodies, respectively; 14 days postinoculation, 12% and 0.2%, respectively).

FIG. 5.

Effects of CD8⁺ T-cell depletion on experimental P. carinii pneumonia in γδ-TCR⁺ T-cell-deficient mice. Three days prior to P. carinii inoculation, male B6.129P2-Tcrd^tm1Mom (Tcr-d KO) and C57BL/6 mice, 6 to 8 weeks of age, were intraperitoneally administered isotype-matched or anti-murine CD8 depleting antibodies (weekly thereafter). Mice were then intratracheally inoculated with 2 × 10⁵ P. carinii cysts and at specific time points were sacrificed and analyzed for P. carinii burden and T-cell subsets. The figure illustrates representative data from three separate experiments (n = 5 to 7 animals/time point) for P. carinii lung burden (A) and CD4⁺/CD3⁺ and CD8⁺/CD3⁺ BAL cell subsets (B). A single asterisk represents significant differences between control and nondepleted γδ-TCR⁺ T-cell-deficient mice, and double asterisks represent significant differences between depleted and nondepleted γδ-TCR⁺ T-cell-deficient mice (P < 0.05). Data are expressed as the mean ± the standard error of the mean.

DISCUSSION

Clinical and experimental studies show that CD4⁺ and CD8⁺ T cells are part of the local inflammatory response to P. carinii (4). In CD4-depleted mice, although high numbers of CD8⁺ T cells infiltrate into the lungs, mice still develop progressive infection (4). However, the infection is more severe when both T-cell subsets are absent, suggesting a role for CD8⁺ T cells in host defense (3). Our results described here suggest that γδ-TCR⁺ T cells are recruited to the lung during P. carinii infection and serve to downregulate CD8⁺ T-cell recruitment. We observed that γδ-TCR⁺ T-cell-deficient mice had demonstrable augmented resolution with a significantly lower organism burden at 14 days postinoculation and almost complete resolution of infection by 28 days postinoculation. In fact, by 28 days postinoculation, organisms could not be detected histologically in lung sections from γδ-TCR⁺ T-cell-deficient mice, whereas small amounts of residual organisms were observed in lung sections from control mice (data not shown). Analysis of T-cell infiltration during infection in γδ-TCR⁺ T-cell-deficient mice revealed a more rapid recruitment of both CD4⁺ and CD8⁺ T cells at 7 days postinoculation which paralleled clearance of P. carinii.

In contrast, P. carinii burden in control mice did not begin to resolve until after 14 days postinoculation. The delay in clearance of P. carinii in control mice was associated with a less rapid recruitment of both T-cell subsets compared to that in γδ-TCR⁺ T-cell-deficient mice. In fact, the levels of CD4⁺ T cells did not reach those observed in γδ-TCR⁺ T-cell-deficient mice until 14 days postinoculation, and the levels of CD8⁺ T cells remained below that observed in γδ-TCR⁺ T-cell-deficient mice throughout the course of the infection. Taken together, these results suggest that the presence of γδ-TCR⁺ T cells in control mice regulates the recruitment of T-cell subsets, particularly CD8⁺ T cells, into the lung during P. carinii infection. However, we also cannot exclude the possibility that slightly increased recruitment of CD4⁺ T cells in γδ-TCR⁺ T-cell-deficient mice may play a role in increasing the recruitment of CD8⁺ T cells and more rapid resolution of P. carinii infection in γδ-TCR⁺ T-cell-deficient mice. Observations in other studies have shown that γδ-TCR⁺ T cells also modulate effector functions of αβ-TCR⁺ T cells, such as IL-2 production (CD4⁺ T cells) and cytotoxic activity (CD8⁺ T cells) (16). Thus, it is possible that γδ-TCR⁺ T cells modulate not only recruitment but also function of CD4⁺ and CD8⁺ T cells during P. carinii pneumonia. This is the subject of ongoing investigations in our laboratory.

Although IFN-γ is not essential for clearance of P. carinii in normal mice up to 8 weeks after infection, early clearance of P. carinii is impaired in IFN-γ gene-disrupted mice (8). Moreover, these studies assessed infection by using P. carinii nuclei counting, which has a significantly lower dynamic range of infection intensity compared to the Taqman real-time PCR method employed in our studies, which is quantitative over 8 logs of P. carinii rRNA concentration (33). In addition, augmenting local levels of IFN-γ is associated with more rapid resolution of infection (2, 18). All of these previous studies, however, were performed in mice with intact γδ-TCR⁺ T cells. Thus, the role of IFN-γ in the absence of γδ-TCR⁺ T cells could not be addressed. We observed significantly higher IFN-γ levels in BAL fluid from γδ-TCR⁺ T-cell-deficient mice than in control mice at 7 days postinoculation. The elevated levels of IFN-γ in γδ-TCR⁺ T-cell-deficient mice correlated with increased CD8⁺ T-cell infiltration, suggesting that CD8⁺ T cells may be a potential source of IFN-γ in this model. We confirmed that CD8⁺ T cells were in fact the primary source of IFN-γ in γδ-TCR⁺ T-cell-deficient mice through analysis of cytokine secretion by BAL cells collected from control and γδ-TCR⁺ T-cell-deficient mice. CD8⁺ T cells have previously been shown to express cytokine patterns similar to T-helper type 1 cells (Tc1 CD8⁺ T cells) and Th2 CD4⁺ T cells (Tc2 CD8⁺ T cells) (21). Moreover, studies from our lab investigating IFN-γ as an adjuvant form of therapy in CD4-depleted mice show that Tc1 CD8⁺ T cells are associated with clearance mediated by IFN-γ (18, 20). Neutralization of IFN-γ early during infection or CD8⁺ T-cell depletion significantly inhibited the augmented clearance of P. carinii from the lungs previously observed in γδ-TCR⁺ T-cell-deficient mice.

Taken together, these results suggest that γδ-TCR⁺ T cells influence CD8⁺ and to a lesser extent CD4⁺ T-cell recruitment and may impair tissue IFN-γ production during P. carinii infection. An alternative possibility is that γδ-TCR⁺ T cells dampen functional activities of CD4⁺ and CD8⁺ T cells, cells known to be critical for host defense against P. carinii (26, 27). Studies are currently under way to evaluate the phenotype of CD4⁺ and CD8⁺ T cells from both control and γδ-TCR⁺ T-cell-deficient mice as well as that of γδ-TCR⁺ T cells to determine the Th1 and Th2 cytokine profiles in these T-cell subsets. Another possibility is that lower tissue IFN-γ levels, which play a known role in activation of macrophage phagocytosis (24), in the lungs of control mice adversely affect early macrophage-mediated clearance of P. carinii (22).

Although our studies suggest a prominent role for both IFN-γ and CD8⁺ T cells in early clearance of P. carinii from γδ-TCR⁺ T-cell-deficient mice, it remains unclear what the role of γδ-TCR⁺ T cells is in the setting of CD4⁺ T-cell deficiency or P. carinii pneumonia during HIV infection. It is known that γδ-TCR⁺ T cells are increased in both the circulation and BAL fluid of HIV-infected patients with P. carinii pneumonia (1, 25). However, their roles in host defense and tissue inflammation remain unclear. Studies are under way to investigate long-term infection in CD4- and CD8-depleted γδ-TCR⁺ T-cell-deficient mice. Based on our studies, mice with intact γδ-TCR⁺ T cells have a delayed clearance and adaptive immune response to P. carinii. However, γδ-TCR⁺ T cells could play a beneficial role in HIV-infected patients by dampening excessive tissue inflammation and lung injury during the responses to P. carinii. Towards this end, it has been shown that γδ-TCR⁺ T cells regulate inflammatory responses to intracellular pathogens such as Listeria monocytogenes (28) and Mycobacterium tuberculosis (6) as well as extracellular pathogens such as Klebsiella pneumoniae (23). Thus, it will be of interest to further examine the role of γδ-TCR⁺ T cells in the absence of CD4⁺ T cells as well in a splenic reconstitution model in P. carinii-infected scid mice, which results in excessive lung inflammation during P. carinii pneumonia (32).