Role of Type I IFNs in Pulmonary Complications of Pneumocystis murina Infection
http://www.100md.com
免疫学杂志 2005年第9期
Abstract
Despite the advent of highly active antiretroviral therapy, pulmonary complications in AIDS are a common clinical problem. Pneumocystis jiroveci infection causes a life-threatening pneumonia, especially in individuals with CD4 T cell deficiencies as occurs in AIDS. Although Pneumocystis sp. is an extracellular fungal pathogen, CD8 T cells are the predominant lymphocyte recruited to the lung in CD4-deficient humans and mice during Pneumocystis pneumonia, and we have found that these CD8 T cells are responsible for subsequent lung damage in CD4 T cell-depleted mice. Comparing CD4 T cell-depleted IFN- receptor knockout (KO) mice to wild-type mice, we found that this CD8 T cell recruitment and lung damage is type I IFN (IFN-) dependent. However, in both CD4 competent, wild-type and IFN- receptor (IFNAR) KO mice, Pneumocystis infection leads to an eosinophilic granulocyte influx with bronchial epithelial changes as seen in asthma. This response is delayed in IFNAR KO mice, as is pathogen clearance. Although the inflammation is transient in wild-type animals and resolves upon Pneumocystis clearance, it is more severe and persists through day 35 postinfection in IFNAR KO mice, leading to fibrosis. In addition, IFNAR KO, but not wild-type, mice mount a Pneumocystis-specific IgE response, an indicator of allergic sensitization. Thus, in the absence of IFNAR signaling and CD4 T cells, Pneumocystis-mediated lung damage does not occur, whereas in CD4-competent animals, the absence of IFNAR signaling results in an exacerbated Th2 response, asthma-like symptoms, and fibrosis. Therefore, both CD4 T cell- and type I IFN-mediated mechanisms can determine pulmonary complications from Pneumocystis infection.
Introduction
Pneumocystis jiroveci infection rarely causes clinical disease in the general population; however, it causes a severe and life-threatening pneumonia in the immune-compromised host. It is especially prominent in patients with CD4 T cell deficiencies and is one of the most common opportunistic infections in patients suffering from AIDS (1). Although patients with CD4 T cell counts <200–300 cell/μl are very likely to develop Pneumocystis pneumonia, there is increasing evidence for an important role of the innate immune functions of the type I IFN system in the control and susceptibility to opportunistic infections. In this regard there is a subgroup of AIDS patients with relatively high CD4 T cell numbers that are unable to control the development of AIDS-related complications such as Kaposi’s sarcoma, and these patients demonstrate a loss of type I IFN (IFN-)-producing dendritic cell numbers, resulting in low IFN- production. Conversely, normal IFN- production and high numbers of type I IFN-producing cells has been associated with long term survival of AIDS patients despite low CD4 T cell counts and absence of antiretroviral treatment (2, 3, 4, 5, 6, 7).
Pneumocystis species (hereafter referred to as Pneumocystis) has been characterized as a fungus (8), and although it is a strictly extracellular pathogen, in the CD4 T cell-deficient human and animal, there is a vigorous CD8 T cell influx into the alveolar as well as the interstitial space of the lung (9). The role of this influx is controversial; in the mouse model of Pneumocystis pneumonia there are opposing reports on the relevance of CD8 T cells in resistance to and clearance of the infection (9, 10). However, there is good evidence that CD8 cells can be made to clear Pneumocystis murina by the overexpression of IFN- in mouse lung epithelium (11). Regardless of the role of CD8 T cells in resistance to Pneumocystis, there is evidence that CD8 T cells mediate lung damage in the CD4 T cell-depleted host; however, the mechanisms involved are poorly understood (9).
Although IFN- (type I IFN) were initially recognized for their innate antiviral and antitumor activities (12), their importance as regulators of AIDS has recently become apparent. In humans and mice, >10 IFN-- and one IFN- molecules have been identified. All molecules signal via one IFN- receptor (IFNAR), 3 a heterodimer comprised of two chains (IFNAR1 and IFNAR2). They induce dendritic cell maturation, resulting in the secretion of IL-12, which supports Th1 cell differentiation, and IL-15, which induces proliferation and expansion of memory and possibly naive CD8 T cells (13, 14, 15, 16). Although CD8 T cells commonly recognize endogenous Ag in the context of MHC class I (17), type I IFN are important licensing factors for dendritic cells to cross-present exogenous Ag in the context of MHC class I to CD8 T cells, and this has become a current model of CD8 T cell activation in the lymph node, leading to their clonal expansion (18, 19, 20).
By taking advantage of its regulatory influence on the induction of Th1 responses (21, 22, 23), IFN- has recently been used successfully in the treatment of steroid-resistant asthma and related diseases where an exaggerated Th2 immune response leads to eosinophilic lung disease (24, 25).
Because of its important immune regulatory function, especially its influence on Ag presentation to CD8 T cells and CD8 T cell activation, we examined the role of IFNAR signaling in the induction of CD8 T cell-mediated lung damage during Pneumocystis pneumonia (PCP) in CD4-depleted mice as well as the role of IFNAR in clearance of the infection in CD4-competent (undepleted) mice. We found in IFNAR KO mice a significant reduction of CD8 T cell recruitment into the lung during PCP. This was associated with reduced lung damage, improved arterial oxygen saturation, and reduced proinflammatory cytokine secretion into the bronchoalveolar lavage (BAL) fluid compared with CD4 T cell-depleted wild-type animals with the same level of Pneumocystis infection. These findings indicate a critical role of IFNAR signaling in the activation and recruitment of CD8 T cells in CD4-depleted animals and, thus, in the development of immune-mediated damage in PCP.
In the CD4-competent host, in both wild-type and IFNAR KO mice, a strong Th2-mediated eosinophilic inflammation occurred; however, in the absence of IFNAR, the Th2 response to Pneumocystis was exacerbated. This was characterized by more intense and persistent eosinophil accumulation, mucus cell metaphasia, and IgE production, as well as beginning lung fibrosis. Thus, signaling through the IFNAR results in exacerbation of the Th2 response to Pneumocystis when CD4 T cells are present, but results in amelioration of CD8 T cell-mediated lung damage when CD4 T cells are absent.
Materials and Methods
Mice
C.B17 SCID mice, as a source of Pneumocystis organisms, were bred and maintained at Montana State University’s Animal Resource Center. All animal experiments were approved by the institutional animal care and use committee and strictly followed the animal care and handling procedures required by the Animal Welfare Act as well as National Institutes of Health and U.S. Department of Agriculture guidelines. A breeder pair of IFNAR KO mice (animals are on a B6/129 mixed background) was provided by Prof. D. E. Levy (New York University School of Medicine, New York, NY) and were bred and maintained at Montana State University’s Animal Resource Center. Mixed background control mice were purchased from The Jackson Laboratory (B6129PF2/J Stock: 100903).
CD4 T cell depletion
For in vivo CD4 T cell depletion, animals were injected with 300 μg of purified GK1.5 rat anti-mouse CD4 Ab i.p. 2–4 days before infection with Pneumocystis and subsequently injected twice per week until the end of the experiment as previously described (9, 26, 27). The GK1.5 hybridoma cell line was obtained from American Type Culture Collection, and the cells were grown according to the instructions provided and with Abs produced in ascites.
Infection of mice with Pneumocystis
Lung homogenates from Pneumocystis-infected C.B17 SCID mice were used as a source of Pneumocystis nuclei, and CD4-depleted or CD4-undepleted (CD4 T cell-competent) recipient animals were infected with 107 Pneumocystis nuclei in a 100-μl volume as previously described via intratracheal instillation with mice under deep isoflurane gas anesthesia (26, 28).
Blood gas analysis
As a measure of lung function, arterial blood was obtained from the tail artery of each animal into capillary tubes and analyzed at Deaconess Hospital (Bozeman, MT) on an AVL Omni3 Blood gas autoanalyzer (AVL Medical Instruments) for pO2, pCO2, and pH as we have previously described (9).
Assessment of cellular infiltrates into lungs and lymphocyte differentiation using FACS analysis
BALs were performed as previously described (28). Briefly, mice were lethally injected with 90 mg/kg sodium pentobarbital solution i.p. and exsanguinated. Lung lavages were performed by intratracheal cannulation with a total volume of 5 ml of HBSS and 0.5% EDTA. A cell aliquot was spun onto glass slides and stained with Diff-Quik (Dade-Behring) for leukocyte differentiation. Cell counts of BALs were performed with a hemocytometer, and 106 cells/sample were used to assess lymphocyte subset distribution via flow cytometry. For this, a four-color immunofluorescence stain was performed using directly conjugated allophycocyanin-anti-mouse CD4 (L3T4), CyChrome-anti-mouse CD8a (Ly2), PE-anti-mouse CD43 (1B11), and FITC-anti-mouse CD44 (Ly-24, clone IM7) Abs. All Abs were obtained from BD Pharmingen. Before staining, cells were incubated with unlabeled anti-mouse FcR Abs to block FcR-mediated unspecific staining, then incubated with a mixture of the appropriate dilutions of the above Abs for 30 min. Cells were washed three times with PBS/2% FCS and analyzed on a FACSCalibur (BD Biosciences).
Histological analysis of lung tissue
Left lung lobes were inflated with 10% neutral buffered formalin through a tracheal cannula, removed en bloc, and immersed in formalin for at least 24 h, as previously described, before paraffin embedding (29). Sections were cut 5 μm thick on a Leica RM2155 microtome and stained with H&E to assess tissue morphology, with Alcin Blue and periodic acid-Schiff to detect mucus production, and with Masson Tri-Chrome to detect collagen deposition. All reagents were purchased from Richard Allen Scientific and stained according to the manufacturer’s instructions.
Enumeration of Pneumocystis nuclei
Pneumocystis burden was assessed by enumeration of its nuclei as previously described (26). Briefly, lungs of exsanguinated animals were removed, and the right lobe was transferred into 5 ml of HBSS (Invitrogen Life Technologies) and homogenized through mesh screens. A 100-μl aliquot of a 1/20 diluted homogenate was spun onto a glass slide using a Shandon Cytospin 3 centrifuge, smears wee stained with Diff-Quik (Dade-Behring), and the number of Pneumocystis nuclei in 10–50 oil immersion fields was counted. The limit of detection for this technique is log10 4.43 for right lobe homogenates.
Albumin quantification in BAL fluids
Albumin concentrations in BAL fluids were determined as readout for inflammatory exudates and a measure for the extent of lung damage. Albumin reagent (BCG) from Sigma-Aldrich and an albumin standard solution were used for this quantitative colorimetric test. Briefly, standards were prepared in 1/2 dilutions in HBSS/3 mM EDTA ranging from 0.1–2 mg/ml. One hundred microliters of standard per sample was added to the appropriate well in a 96-well plate and mixed with 50 μl of albumin reagent (BCG). Production of green-blue color was immediately read on a ThermoMax microplate reader (Molecular Devices) at 628 nm and analyzed using Softmax software (Molecular Devices).
Cytokine analysis using bead arrays in BALs and tissue culture supernatants
TNF-, IFN-, IL-5, IL-4, IL-2, IL-6, MCP-1, and IL-10 cytokine secretion into the BAL fluid was assessed using the Mouse Th1-Th2 and Mouse Inflammation Cytometric Bead Array assay kit (BD Biosciences) according to the manufacturer’s instructions. Assays were read on a FACScan flow cytometer (BD Biosciences), and data were analyzed using cytometric bead array software (BD Biosciences). In addition, IL-10, IL-13, and TGF-1 concentrations in BAL fluid were measured using a commercially available mouse specific sandwich ELISA (R&D Systems).
Pneumocystis-specific serum IgG and IgE analysis
Pneumocystis-specific IgG and IgE responses were determined using a sandwich ELISA technique. Briefly, 96-well, flat-bottom plates were coated with 10 μg/ml Pneumocystis Ag isolated from lungs of heavily infected source mice as previously described (30) in 0.05 M sodium carbonate buffer, pH 9.6, for 3 h. Plates were washed five times with PBS/0.05% Tween 20 (wash buffer) and blocked overnight with 5% nonfat dry milk powder (Blotto) in PBS. Sample sera were diluted 1/100 in sample buffer (1% BSA/PBS) for Pneumocystis-specific IgG and 1/25 for Pneumocystis-specific IgE, and 100 μl/well was incubated at 37°C for 2 h. Plates were washed six times; 100 μl/well of alkaline phosphatase-conjugated goat anti-mouse IgG (Sigma-Aldrich) at 1/2500 in sample buffer was added to plates for Pneumocystis-specific IgG detection, and 2 μg/ml biotinylated anti-mouse IgE (BD Biosciences; clone R35-118) was added for Pneumocystis-specific IgE detection, and incubation was performed for 1 h at room temperature, then plates were washed six times. Avidin-HRP (Vector Laboratories) was diluted 1/10,000 in sample buffer, and 100 μl/well was added to plates for IgE detection, then incubation was performed for 30 min. Plates were washed again six times, then 50 μl/well ABTS buffer (Sigma-Aldrich) was added to plates for IgE detection, and 50 μl/well p-nitrophenyl-phosphate (Sigma-Aldrich) in dietholamine buffer, pH 9.8 (Sigma-Aldrich), at 1 mg/ml was added to plates for IgG detection, and incubation proceeded for 30 min. Plates were read at 405 nm wavelength on a VersaMax plate reader (Molecular Devices). Samples were read for OD, and positive and negative control sera were included in the assay.
Total serum IgE was measured using a mouse IgE capture and detection Ab pair plus mouse IgE standard from BD Biosciences, and the assay was performed according to the manufacturer’s protocol.
Assessment of airway hyperreactivity in response to methacholine challenges using a whole-body plethysmograph
Airway hyperresponsiveness was measured by induction of the enhanced pause (penh) in the airflow upon exposure to increasing dosages of aerosolized methacholine (0, 2.5, 5, 10, and 20 mg/ml) in a whole-body plethysmograph for rodents (Buxco). Between each methacholine dosage, animals were allowed to rest for up to 15 min to return to baseline values. The lowest dosage of methacholine that induced significant differences in penh values between the groups was compared.
Statistical analysis
For statistical analysis, a one-way ANOVA was performed, and pairwise comparisons were made using the post-hoc Tukey test. Data were assumed to be significantly different at p < 0.05. Data are shown as the mean ± SEM.
Results
Type I IFN signaling is required in the induction of immune-mediated lung damage during the course of PCP in CD4-depleted mice
To assess the role of type I IFNs in the activation of CD8 T cells in the course of PCP, IFNAR KO mice were CD4 T cell-depleted and Pneumocystis-infected via intratracheal instillation of 107 nuclei of Pneumocystis. The course of disease progression and lung damage was assessed on day 28 postinfection and compared with the course of disease in CD4 T cell-depleted wild-type animals. Day 28 postinfection was chosen because we have previously shown that lung damage in Pneumocystis-infected, CD4 T cell-depleted animals is well established by this time, although the animals are not yet terminally sick (9). Each experiment included a control group consisting of uninfected wild-type animals, because previous comparative analysis of uninfected lungs from IFNAR KO mice and wild-type mice showed no differences with regard to lung structure and cellular composition of BAL cells (>95% alveolar macrophages). Each experimental group consisted of five animals, and each experiment was repeated three times. Arterial blood gases were taken from all groups, animals were then killed, and Pneumocystis lung burden, lung histopathology, and airway inflammatory responses were determined. Blood gas analysis revealed a significantly better arterial oxygenation of IFNAR KO mice on day 28 postinfection compared with that in wild-type animals (Fig. 1A), and albumin levels, as a measure for vascular leakage, were significantly reduced in IFNAR KO mice compared with the infected wild-type animals (Fig. 1B). Pneumocystis burden was not different in the two groups and cannot account for the differences in measures of lung function and damage (Fig. 1C).
FIGURE 1. CD4 T cell-depleted IFNAR KO mice show better lung function and reduced lung damage during the course of PCP than CD4-depleted wild-type animals. Arterial blood gases were taken as a measure of lung function (A), and albumin leakage into the alveolar spaces was assessed in BAL fluid as a measure of lung damage (B). The Pneumocystis burden of the lung was assessed via Pneumocystis nuclei count of lung homogenate cytospin slides (C). Significant differences between the groups are indicated: *, p < 0.05; **, p < 0.01; ***, p < 0.001. Significant differences between infected IFNAR KO and wild-type animals are indicated by an asterisk above the bar of the wild-type group; significant differences between the infected IFNAR KO group and the uninfected control group are indicated by an asterisk above the bar for the IFNAR KO group. IFNAR KO mice showed better arterial pO2 levels and less lung damage than wild-type mice during the course of PCP, which was not due to a lesser Pneumocystis burden of the lungs compared with wild-type animals.
Leukocyte populations recruited into the BAL fluids were assessed by differential cell count and revealed a similar lung accumulation of neutrophils and macrophages in wild-type and IFNAR KO mice, with a slightly reduced accumulation of lymphocytes in IFNAR KO mice (data not shown). Lymphocyte subset analysis via FACS confirmed depletion of CD4 T cells using an anti-CD4 Ab (clone L3T4) that does not compete with the GK1.5 mAb epitope. Furthermore, CD8 T cells were the predominant lymphocyte population recruited to the lung, with a similar activation profile in the two groups as assessed by the percentage of CD44- and CD43high-expressing CD8 T cells (data not shown). However, total lymphocyte numbers were significantly lower in BAL fluid of IFNAR KO mice compared with that of wild-type mice due to a reduced influx of activated CD8 T cells (Fig. 2). In agreement with these findings, cytokine analysis in the BAL fluid showed significantly reduced secretion of the proinflammatory cytokines TNF-, IFN-, IL-6, and MCP-1 (Fig. 3), because they can be secreted by activated CD8 T cells and other cells. The levels of the anti-inflammatory cytokines IL-10 and TGF- were not significantly different between the groups of mice (data not shown).
FIGURE 2. Lymphocyte numbers are reduced in BAL fluids of CD4-depleted IFNAR KO mice compared with wild-type animals during the course of PCP due to reduced influx of activated CD8 T cells. Lymphocyte numbers were assessed by performing total leukocyte count using a hemocytometer and differential cell counts of leukocytes in BAL fluid using Diff-Quik-stained cytospin slides. Lymphocyte differentiation and their activation profile were performed via FACS analysis staining for CD4, CD8, CD44, and CD43 surface markers. A, Comparative analysis of absolute lymphocyte numbers assessed in BAL fluid in IFNAR KO mice compared with wild-type mice. Total CD8 T cell (B) and activated CD44/CD43+ CD8 T cell numbers (C) were calculated based on the percentage of total lymphocytes. As no lymphocytes were detectable in BAL fluid of uninfected control animals, absolute lymphocyte numbers as well as subset values were zero. Statistical differences between infected IFNAR KO and wild-type animals are indicated above the bars of the wild-type group: *, p < 0.05; **, p < 0.01; ***, p < 0.001.
FIGURE 3. Proinflammatory cytokine levels are significantly lower in BAL fluid of IFNAR KO mice compared with wild-type animals during the course of PCP. Cytokine concentrations as a measure of inflammatory activity were assessed in BAL fluids of CD4 T cell-depleted IFNAR KO and wild-type mice suffering from PCP using the inflammation bead array (BD Biosciences). TNF- (A), IFN- (B), IL-6 (C), and MCP-1 (D) were significantly lower in IFNAR KO mice. Cytokine concentrations in the BAL fluid of uninfected control animals were between 0 and 5 pg/ml for all cytokines and may appear as undetectable in the appropriate graphs. Statistical differences between infected IFNAR KO and wild-type animals are indicated above the data bars of the wild-type group: *, p < 0.05; **, p < 0.01; ***, p < 0.001.
Lack of IFNAR signaling leads to a delayed, but complete, clearance of Pneumocystis infection in CD4-competent animals
We have previously found that in immune-competent, wild-type animals, Pneumocystis infection is cleared by days 14–21, with minimal or no residual inflammation of lung tissue by day 28 postinfection. Because of the reduced inflammation seen in CD4-depleted IFNAR KO mice in response to Pneumocystis infection, we determined whether clearance of the infection would be impaired in CD4 T cell-competent (undepleted) mice. Therefore, we infected both CD4-competent, wild-type and IFNAR KO mice with Pneumocystis intratracheally and evaluated Pneumocystis clearance, Pneumocystis-specific serum immune globulin levels, and the inflammatory response induced on days 14, 21, and 28 postinfection.
Fig. 4A shows the kinetics of Pneumocystis clearance over a period of 14, 21, and 28 days. In wild-type animals, Pneumocystis infection was barely above the microscopic detection limit on days 14 and 21 postinfection and was completely undetectable on day 28. In contrast, in IFNAR KO animals, a slight growth occurred by day 14, which was even greater on day 21. However, the infection was completely cleared by day 28. This delayed clearance pattern in IFNAR KO mice was associated with a delayed induction of a Pneumocystis-specific IgG response, with no Pneumocystis-specific Abs detectable on day 14, a subsequent increase in production by day 21, and normal Pneumocystis-specific IgG levels on day 28 compared with the IgG response in wild-type animals (Fig. 4B). BAL fluid cell counts in wild-type animals were highest on day 14 and decreased consecutively over the course of clearance, whereas in IFNAR KO mice, the cellular influx into lung airways did not peak until day 21 (Fig. 4C).
FIGURE 4. Clearance of Pneumocystis infection is delayed, but complete, in CD4-competent IFNAR KO mice compared with wild-type animals due to delayed induction of both cellular and humeral immune responses. CD4-competent IFNAR KO and wild-type animals were infected with Pneumocystis and euthanized on day 14, 21, or 28 postinfection. Pneumocystis lung burden was assessed via Pneumocystis nuclei counts in lung homogenates. Although the level of infection in both wild-type animals () and IFNAR KO mice () was low 14 days postinfection (limit of detection, 4.43 log10 nuclei) and was almost completely cleared on day 21 in wild-type animals, IFNAR KO mice showed significant Pneumocystis growth by day 21 and cleared the infection by day 28 postinoculation (A). The delayed clearance pattern is associated with the delayed induction of a Pneumocystis-specific serum IgG response (B) as well as delayed influx of cells into the lungs of infected IFNAR KO mice (C) compared with wild-type mice. Statistical differences between infected IFNAR KO- and wild-type animals are indicated above the data bars of the wild-type group: *, p < 0.05; **, p < 0.01; ***, p < 0.001.
Pneumocystis infection of CD4 T cell-sufficient mice leads to transient eosinophilic immune response in wild-type animals, resulting in pathological changes similar to those of asthma. This response is prolonged and exaggerated in IFNAR KO animals and leads to pulmonary fibrosis. BAL fluid leukocyte differentiation revealed that clearance of the pathogen in both wild-type and IFNAR KO animals was associated with a strong eosinophil granulocyte influx. This influx was again delayed in IFNAR KO mice to day 21 (Fig. 5A). Because eosinophilic airway inflammation can be associated with increased airway hyperreactivity as in allergic asthma, animals were exposed to increasing dosages of methacholine, and airway resistance was assessed using a whole-body plethysmograph for rodents. On day 14 postinfection, wild-type animals showed transiently increased penh values in response to low dose methacholine (2.5 mg/ml), whereas IFNAR KO mice did not. However, on day 28 postinfection, IFNAR KO animals demonstrated increased penh values (Fig. 5B), whereas wild-type animals had returned to baseline levels. BALB/c as well as C57BL/6 wild-type mice were tested in the same experimental setting, and similar transient airway eosinophilia as well as airway hyperreactivity were found (data not shown) (S. Swain, and A. Harmsen, manuscript in preparation) during the course of clearance of the infection, indicating a general immune phenomenon rather than a strain-specific response. To further exclude that eosinophilic influx was not due to an Ag incompatibility reaction due to the instillation of Pneumocystis-infected mouse lung homogenate derived from SCID mice on a BALB/c genetic background into animals with a mixed (B6/129) genetic background, B6/129 mixed background wild-type animals were either intratracheally instilled with lung homogenate from an uninfected allogeneic B6/129 mouse or an uninfected BALB/c SCID mouse and compared with mice receiving the instillation of Pneumocystis-infected lung homogenate from BALB/c SCID mice used in the experimental settings described above. Intratracheal instillation of uninfected lung homogenates from BALB/c SCID mice or allogeneic B6/129 wild-type mice did not lead to any detectable eosinophilic cell response in BAL fluids on day 14 postinstillation, whereas the instillation of Pneumocystis-infected lung homogenates did, indicating that the eosinophilic response was due to the presence of Pneumocystis pathogen in the lung homogenate (data not shown).
FIGURE 5. Eosinophils are the major leukocyte population recruited to the lung during the course of Pneumocystis clearance, and acute airway hyperresponsiveness is associated with this transient eosinophil influx. Leukocyte differentiation of cells in BAL fluid of both IFNAR KO mice () and wild-type animals () revealed that the major population recruited to the lung during the course of clearance was eosinophils. The influx of these cells was again delayed in IFNAR KO animals (A). Airway hyper-responsiveness was assessed in response to methacholine and was measured as the penh by whole-body plethysmography. Increased airway hyperresponsiveness was detected at 2.5 mg/ml methacholine on day 14 in wild-type animals and was delayed until day 28 in IFNAR KO mice (B). Statistical differences between infected IFNAR KO and wild-type animals are indicated: *, p < 0.05; **, p < 0.01; ***, p < 0.001.
Histological analysis of formalin-fixed lung tissue (Fig. 6) in wild-type mice on day 14 revealed an intense Th2-like immune response and confirmed recruitment of eosinophils and lymphocytes into the lung airways and interstitium as well as giant cell formation and epithelial and goblet cell metaplasia of the bronchi resembling an asthmatic inflammatory response. In contrast, only mild inflammation was detected on day 14 postinfection in lungs of IFNAR KO mice, with no eosinophil accumulation detectable, but by this time, some lymphocyte recruitment had occurred as well as the formation of some giant cells (Fig. 6, top row). On day 21, however, lungs of IFNAR KO mice showed an even more severe inflammation in response to the infection than that on day 14 in wild-type animals with massive eosinophil and lymphocyte infiltrate, giant cell formation, and mucus cell hyperplasia with almost complete consolidation of the lung tissue. In contrast, wild-type animals were already resolving the response, with only minor peribronchial eosinophil and lymphocyte infiltrate remaining, and the bronchial epithelia height had nearly returned normal (Fig. 6, second row). On day 28 postinfection, wild-type animals showed only minimal residual inflammation, whereas IFNAR KO mice still showed severe asthmatic inflammation similar to that on day 21 (Fig. 6, third row). Because the intense Th2-like response in IFNAR mice persisted, we repeated the experiments and chose day 35 instead of day 28 as our latest time point. The time course of the inflammation in the second experiment mirrored the previous findings and showed a still vigorous eosinophilic inflammation of the lung present through day 35 postinfection, with prominent thickening of the bronchial epithelium, goblet cell hyperplasia, and peribronchial as well as alveolar cellular infiltrate (Fig. 6, bottom row). In addition to the H&E stains, we performed Alcin Blue/periodic acid-Schiff stains, which stained mucus glycoprotein and confirmed mucus cell hyperplasia as a hallmark of asthma throughout day 35 in both wild-type and IFNAR KO animals (Fig. 7, rows 1 and 2). Chronic inflammation of the lung can lead to irreversible changes in the form of tissue fibrosis with loss of tissue elasticity and reduced gas exchange. Masson Tri-Chrome staining of formalin-fixed lung tissue revealed prominent interstitial as well as peribronchial collagen deposition as a sign of lung fibrosis in IFNAR KO mice on day 35 postinfection, whereas lung tissue of wild-type animals at the same time point looked like that in uninfected controls (Fig. 7, row 4).
FIGURE 6. Eosinophilic inflammation was apparent in both CD4-competent IFNAR KO and wild-type animals, with lung changes similar to those seen in asthma. However, the response was delayed, but exacerbated and prolonged, in IFNAR KO mice. Shown are photomicrographs of formalin-fixed and H&E-stained lung sections taken at x10 and x200 magnifications. Eosinophilic lung inflammation with bronchial cell hypertrophy (*) and giant cell formation (<) can be seen in all sections. Delayed onset of inflammation in IFNAR KO mice on day 14 (first row) was noticed and was more severe on day 21 in IFNAR KO mice compared with that on day 14 in wild-type animals. Although the inflammation in wild-type animals waned to almost complete resolution on days 28 and 35 (rows 3 and 4), inflammation remained almost unchanged in IFNAR KO mice through day 35 despite clearance of the pathogen by day 28 (rows 2–4).
FIGURE 7. PAS/Alcin Blue stain confirmed mucus secretion from hypertrophic goblet cells in both CD4-competent, wild-type and IFNAR KO animals in response to the Pneumocystis infection, and Masson Tri-Chrome stain revealed fibrotic lung changes in IFNAR KO mice in response to the prolonged ongoing inflammation. One characteristic feature of asthma is increased mucus cell secretion, which can lead to mucus plugging of the airways. PAS/Alcin Blue stained mucus glycoproteins dark blue, was induced in both wild-type and IFNAR KO animals, and remained throughout day 35 postinfection (rows 1 and 2). Uninfected control animals did not show positive mucus staining of their airways (row 3, left photograph). Fibrosis of tissue can occur in response to chronic inflammation and is a concern in chronic asthma (remodeling of the lung). The Masson Tri-Chrome procedure stained collagen depositions in the tissue blue and showed peribronchial and interstitial changes on day 35 postinfection only in IFNAR KO mice (row 4, *).
Pneumocystis exposure in CD4-competent mice leads to Pneumocystis-specific allergic sensitization
Because of the transient histological changes in wild-type animals and the prolonged inflammation in IFNAR KO animals resembling allergic inflammation, we determined whether Pneumocystis could induce a specific IgE response as found during allergic sensitization. Total IgE was detectable in both wild-type and IFNAR KO animals (Fig. 8A), but only in IFNAR KO mice did total IgE increase over 28 days, although this was not statistically significant compared with the wild-type group. However, a dramatic Pneumocytis-specific IgE response was only induced in the IFNAR KO group beginning on day 21 postinfection, with highest readings on day 28 (Fig. 8B). Therefore, a delayed clearance of the pathogen in IFNAR KO mice may have been responsible for the induction of the Pneumocystis-specific IgE response changing Pneumocystis from an immunological Ag to an allergen.
FIGURE 8. Pneumocystis-specific serum IgE is induced in CD4-competent IFNAR KO, but not in wild-type, animals during the course of Pneumocystis clearance. Total serum IgE (A) as well as Pneumocystis-specific serum IgE (B) were measured using a sandwich ELISA technique. Total IgE increased in both wild-type and IFNAR KO mice, but Pneumocystis-specific IgE was only induced in IFNAR KO mice. Statistical differences between infected IFNAR KO and wild-type animals are indicated: *, p < 0.05; **, p < 0.01; ***, p < 0.001.
Cytokine profile analysis of BAL fluids from both groups over the above time course revealed a Th2 pattern (Fig. 9, A–C) with increased production of IL-5, IL-13, and TGF-. Highest cytokine concentrations were measured on day 14 in wild-type animals and on day 21 in IFNAR KO animals, and the induction of these cytokines was comparable between the two groups at their individual peak time points. IFN- concentrations were unchanged during the time course of clearance in both groups (Fig. 9D). A second, independent experiment showed similar results. However, in this experiment, IL-10 was also measured, because it has been identified to down-regulate specific immune responses. Wild-type mice produced significant amounts of IL-10 on day 14 postinfection, but no IL-10 was produced on day 14, and little was produced on day 21 postinfection in IFNAR KO mice (Fig. 9E).
FIGURE 9. Cytokine analysis of BAL fluid from both CD4-competent, wild-type, and IFNAR KO animals showed a Th2-like cytokine profile. Cytokine analysis of BAL fluid from both CD4-competent IFNAR KO and wild-type animals was performed using a Th1/Th2 cytokine bead array assay (BD Biosciences) as well as a sandwich ELISA for IL-13, TGF-, and IL-10 (R&D Systems). Levels of IL-5, IL-13, and TGF- (A–C) were highest on day 14 in wild-type mice and were comparable on day 21 in IFNAR KO animals, confirming the delayed, but then similar, cytokine response in IFNAR KO mice. IFN- secretion did not vary over the course of clearance (D). IL-10 secretion was significantly detectable on day 14 in wild-type animals; however, it was only weakly induced on day 21 in IFNAR KO mice (E). Statistical differences between infected IFNAR KO and wild-type animals are indicated: *, p < 0.05; **, p < 0.01; ***, p < 0.001.
Discussion
In humans and mouse models of PCP, it has been shown that in the absence of CD4 T cells, a vigorous CD8 T cell influx into the infected lung occurs during the course of the disease, which mediates the occurring lung damage (9).
We found that CD4-depleted IFNAR KO mice with disrupted signaling for both IFN- and IFN- showed significantly better lung function as well as less lung damage compared with wild-type animals, although the Pneumocystis burden of both groups was identical. This was associated with significantly reduced influx of activated CD8 T cells into the lungs of infected IFNAR KO mice. The mechanisms of CD8 T cell-mediated lung damage during the course of PCP are not clear, but CD8 effector cells have been reported to secrete proinflammatory cytokines/chemokines such as IFN-, TNF-, and RANTES (31, 32), which are involved in effector functions and can induce target cell death. Wright et al. (33) suggested a role for TNF- receptor signaling in the mechanisms of CD8 T cell-mediated lung damage during the course of PCP by influencing the secretion of other proinflammatory cytokines and chemokines, such as TNF-, keratinocyte-derived cytokine, RANTES, and MCP-1. We found in IFNAR KO mice, compared with wild-type mice, significantly reduced concentrations of proinflammatory cytokines, including TNF-, IFN-, MCP-1, and IL-6, all of which could be produced by activated CD8 T cells. This occurred despite vigorous Pneumocystis growth in the lungs of both groups. These findings indicate that the induction of these cytokines and recruitment of CD8 T cells to the lung require type I IFN-dependent mechanisms during the course of PCP.
CD8 T cell-mediated mechanisms of damage have been implicated in other lung diseases, such as extrinsic allergic alveolitis or hypersensitivity pneumonitis (34). The immune mechanisms by which CD8 T cells are activated and what Ag they detect in these diseases is unclear; however, similar pathological sequelae can be seen in interstitial lung diseases, that occur in the context of autoimmune diseases such as rheumatoid arthritis (35). Furthermore, type I IFN-dependent mechanisms have been implicated in the pathogenesis of other autoimmune diseases, such as systemic lupus erythematosus and diabetes mellitus type I, and treatment of these patients with anti-IFN- mAbs has been suggested (36, 37). That lung damage in PCP in the absence of CD4 T cells is dependent on IFNAR signaling suggests that a therapeutic approach of neutralizing IFN- signaling of classic PCP could alleviate lung damage in this disease. Thus, we determined whether a deficiency in IFNAR signaling would also affect host resistance to Pneumocystis in a CD4-competent animal. Previous pathogenesis studies revealed that IFNAR KO mice are highly susceptible to infections by a broad array of different viruses (e.g., Poxviridae, Arenaviridae, Rhabdoviridae, and Togaviridae), yet have normal resistance to the microbial pathogen Listeria monocytogenes, and it has been established that the type I IFN system (IFN-) and the type II IFN system (IFN-) are nonredundant (12, 38). In a study using influenza virus infection, the inflammation that occurred in both wild-type animals and IFN- KO mice was predominantly lymphocytic, whereas in IFNAR KO mice, it was predominantly eosinophilic (39). Interestingly, when we compared Pneumocystis clearance in CD4-competent IFNAR KO mice to that in wild-type mice, we found that both animal groups responded with a severe Th2-like immune response, with massive recruitment of eosinophils into the lung tissue as well as bronchial epithelia cell hyperplasia with mucus production and giant cell formation. However, IFNAR KO mice showed a delayed onset of inflammation (day 21 vs day 14 in wild-type animals), followed by a prolonged and augmented Th2-like immune response, resulting in exacerbated lung pathology. This was associated with delayed clearance of the pathogen due to delayed induction of a Pneumocystis-specific IgG Ab response (complete clearance in wild-type animals by day 14; in IFNAR KO mice by day 28). These data indicate that a type I IFN response is not necessary for pathogen clearance, yet it is necessary for the timing and modulation of induction of the protective immune response. Why the Pneumocystis-specific IgG response is delayed in IFNAR KO mice is not clear; however, the absence of IFNAR signaling could lead to delayed maturation of professional APCs and, therefore, specific Th cells. Furthermore, the importance of IFN- in conjunction with IL-6 in the induction of plasma cell differentiation and immune globulin secretion has recently been described (40), and the loss of IFNAR signaling could directly influence and delay plasma cell differentiation and Ab secretion in response to Pneumocystis infection.
Pneumocystis infection in IFNAR KO mice, compared with that in wild-type mice, resulted in exacerbated, asthma-like responses. Interestingly, despite concerns about side effects, IFN- treatment has been shown to be a successful option in the treatment of steroid-resistant asthma, and its mechanisms of actions are thought to involve the promotion of a Th1, rather than a Th2, immune response, thus disrupting the cycle of eosinophilic inflammation and airway hyperreactivity (24, 25). Furthermore, it was shown that histamine, released during allergic inflammation, inhibits the secretion of IFN- from plasmacytoid dendritic cells, which could promote a strong Th2 response, as in allergic inflammation. This might be an explanation for the decreased type I IFN levels found in atopic patients compared with healthy individuals (41). These findings strongly suggest a role of type I IFN signaling in modulating allergic inflammation in humans. The importance of type I IFN in allergic sensitization has previously been demonstrated in a mouse model of allergic sensitization to OVA comparing wild-type animals to animals lacking the IFN- gene (42). In the absence of IFN-, animals showed a more severe pulmonary eosinophilic inflammation, with goblet cell hypertrophy and increased production of Th2 cytokines as well as OVA-specific IgE in response to the sensitization protocol, compared with wild-type or heterozygous littermates. Also, increased OVA-specific T cell proliferation indicated an important role for type I IFN in the modulation and inhibition of allergic sensitization. Although both wild-type and IFNAR KO mice responded with a strong Th2-mediated immune response to Pneumocystis infection with changes resembling allergic (type I hypersensitivity) reactions, this was only transient in wild-type animals and waned fast. However, in animals with a defect in the IFNAR signaling pathway, a prolonged asthma-like inflammatory response of the lungs was noted, with signs of beginning fibrotic tissue remodeling, as seen in chronic asthma (43); only in this group was Pneumocystis-specific IgE production detectable. This can be regarded as evidence for allergic sensitization to the pathogen, which could, in turn, promote an IgE-mediated allergic inflammatory response upon re-exposure to the pathogen in IFNAR KO mice. Why allergic sensitization occurred in IFNAR KO mice and not in wild-type mice is unknown. However, significant secretion of IL-10 was only seen in wild-type mice before resolution of the inflammation. IL-10 has been shown to be secreted by specific regulatory T cells, which have been described to play a key role in the specific down-regulation of Th2 responses and have been considered important in the down-regulation of allergic inflammations (44, 45, 46). Therefore, the absence of IFNAR signaling may have affected the induction of regulatory T cells because of impaired maturation of regulatory dendritic cells (47). However, additional characterization of T cell and APC subsets in the lung during the course of Pneumocystis infection needs to be performed.
In summary, we found that during the course of PCP, in the absence of CD4 T cells, a type I IFN-dependent, vigorous CD8 T cell response is induced. This suggests that type I IFN could be a therapeutic target to modulate the CD4 T cell-independent inflammatory response during the course of PCP and possibly other extrinsic allergic pneumonitis-type responses. However, in a CD4-competent host, the immune response induced during the course of clearance of Pneumocystis resembles histological lung changes that occur during a classical asthmatic inflammation. Although IFNAR signaling does not appear to be critical for the clearance of the infection, it modulates and down-regulates this inflammation upon clearance of the pathogen, leading to complete recovery of the lung without residual inflammation. However, the absence of IFNAR signaling in a CD4-competent animal leads to a prolonged, Th2-mediated, eosinophilic lung pathology as well as an allergic sensitization to the pathogen with Pneumocystis-specific IgE production, possibly due to the loss of IL-10-producing regulatory T cells. Because mechanisms of allergic sensitization are still poorly understood, yet genetic as well as environmental factors are thought to contribute to the onset of the disease (48, 49), we propose that exposure to Pneumocystis can induce allergic sensitization to the pathogen in the presence of impaired type I IFN signaling, which may induce a continuous aggravation during re-exposure. Furthermore, based on the induction of a strong Th2-mediated immune response during the natural course of clearance of Pneumocystis in the lung, exposure of an individual with pre-existing asthma to Pneumocystis may lead to severe exacerbation of asthma or other obstructive lung diseases. Interestingly, a study by Morris et al. (50) found an association of chronic obstructive pulmonary lung disease severity and Pneumocystis colonization of the lung in patients with chronic obstructive pulmonary lung disease. Furthermore, there are increasing reports of pulmonary complications in AIDS patients, such as emphysema, pulmonary hypertension, noninfectious inflammatory disease, and asthma, as well as the clinical presentation of PCP as asthma (51, 52, 53). It is likely that depending on the immune status of a patient infected with HIV, the clinical manifestation of Pneumocystis infection is variable, and chronic colonization (that may or may not be detectable) in a partially immune-reconstituted patient may lead to pulmonary complications other than classical PCP.
Acknowledgments
We thank Soo Han and Pete Degel for excellent technical assistance.
Disclosures
The authors have no financial conflict of interest.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by National Institutes of Health Grant P20RR16455-01 from the BRIN Program of the National Center for Research Resources as well as National Institutes of Health Grants RO1HL55002, P01HL71659, and P20RR020185.
2 Address correspondence and reprint requests to Dr. Nicole N. Meissner, Veterinary Molecular Biology Department, Montana State University, Molecular Bioscience Building, 960 Technology Boulevard, Bozeman, MT 59718. E-mail address: nicolem@montana.edu
3 Abbreviations used in this paper: IFNAR, IFN- receptor; BAL, bronchoalveolar lavage; KO, knockout; PCP, Pneumocystis pneumonia; penh, enhanced pause.
Received for publication October 27, 2004. Accepted for publication February 7, 2005.
References
Sepkowitz, K. A.. 2002. Opportunistic infections in patients with and patients without acquired immunodeficiency syndrome. Clin. Infect. Dis. 34:1098.
Siegal, F. P., P. Fitzgerald-Bocarsly, B. K. Holland, M. Shodell. 2001. Interferon- generation and immune reconstitution during antiretroviral therapy for human immunodeficiency virus infection. AIDS 15:1603.
Siegal, F.. 2003. Interferon-producing plasmacytoid dendritic cells and the pathogenesis of AIDS. Res. Initiat. Treat. Action 8:10.
Soumelis, V., I. Scott, F. Gheyas, D. Bouhour, G. Cozon, L. Cotte, L. Huang, J. A. Levy, Y. J. Liu. 2001. Depletion of circulating natural type 1 interferon-producing cells in HIV-infected AIDS patients. Blood 98:906.
Pacanowski, J., S. Kahi, M. Baillet, P. Lebon, C. Deveau, C. Goujard, L. Meyer, E. Oksenhendler, M. Sinet, A. Hosmalin. 2001. Reduced blood CD123+ (lymphoid) and CD11c+ (myeloid) dendritic cell numbers in primary HIV-1 infection. Blood 98:3016.
Soumelis, V., I. Scott, Y. J. Liu, J. Levy. 2002. Natural type 1 interferon producing cells in HIV infection. Hum. Immunol. 63:1206.
Levy, J. A.. 2001. The importance of the innate immune system in controlling HIV infection and disease. Trends Immunol. 22:312.
Stringer, J. R.. 1996. Pneumocystis carinii: what is it, exactly?. Clin. Microbiol. Rev. 9:489
Wright, T. W., F. Gigliotti, J. N. Finkelstein, J. T. McBride, C. L. An, A. G. Harmsen. 1999. Immune-mediated inflammation directly impairs pulmonary function, contributing to the pathogenesis of Pneumocystis carinii pneumonia. J. Clin. Invest. 104:1307.
Beck, J. M., R. L. Newbury, B. E. Palmer, M. L. Warnock, P. K. Byrd, H. B. Kaltreider. 1996. Role of CD8+ lymphocytes in host defense against Pneumocystis carinii in mice. J. Lab. Clin. Med. 128:477.
Kolls, J. K., S. Habetz, M. K. Shean, C. Vazquez, J. A. Brown, D. Lei, P. Schwarzenberger, P. Ye, S. Nelson, W. R. Summer, et al 1999. IFN- and CD8+ T cells restore host defenses against Pneumocystis carinii in mice depleted of CD4+ T cells. J. Immunol. 162:2890.
Samuel, C. E.. 2001. Antiviral actions of interferons. Clin. Microbiol. Rev. 14:778.
Levy, D. E., I. Marie, A. Prakash. 2003. Ringing the interferon alarm: differential regulation of gene expression at the interface between innate and adaptive immunity. Curr. Opin. Immunol. 15:52.
Taki, S.. 2002. Type I interferons and autoimmunity: lessons from the clinic and from IRF-2-deficient mice. Cytokine Growth Factor Rev. 13:379.
Zhang, X., S. Sun, I. Hwang, D. F. Tough, J. Sprent. 1998. Potent and selective stimulation of memory-phenotype CD8+ T cells in vivo by IL-15. Immunity 8:591
Schluns, K. S., K. Williams, A. Ma, X. X. Zheng, L. Lefrancois. 2002. Cutting edge: requirement for IL-15 in the generation of primary and memory antigen-specific CD8 T cells. J. Immunol. 168:4827.
Wong, P., E. G. Pamer. 2003. CD8 T cell responses to infectious pathogens. Annu. Rev. Immunol. 21:29.
Norbury, C. C., L. J. Sigal. 2003. Cross priming or direct priming: is that really the question?. Curr. Opin. Immunol. 15:82
Le Bon, A., N. Etchart, C. Rossmann, M. Ashton, S. Hou, D. Gewert, P. Borrow, D. F. Tough. 2003. Cross-priming of CD8+ T cells stimulated by virus-induced type I interferon. Nat. Immunol. 4:1009.
Beignon, A. S., M. Skoberne, N. Bhardwaj. 2003. Type I interferons promote cross-priming: more functions for old cytokines. Nat. Immunol. 4:939.
Taki, S., T. Sato, K. Ogasawara, T. Fukuda, M. Sato, S. Hida, G. Suzuki, M. Mitsuyama, E. H. Shin, S. Kojima, et al 1997. Multistage regulation of Th1-type immune responses by the transcription factor IRF-1. Immunity 6:673.
Hida, S., K. Ogasawara, K. Sato, M. Abe, H. Takayanagi, T. Yokochi, T. Sato, S. Hirose, T. Shirai, S. Taki, et al 2000. CD8+ T cell-mediated skin disease in mice lacking IRF-2, the transcriptional attenuator of interferon-/ signaling. Immunity 13:643.(Nicole N. Meissner, Steve)
Despite the advent of highly active antiretroviral therapy, pulmonary complications in AIDS are a common clinical problem. Pneumocystis jiroveci infection causes a life-threatening pneumonia, especially in individuals with CD4 T cell deficiencies as occurs in AIDS. Although Pneumocystis sp. is an extracellular fungal pathogen, CD8 T cells are the predominant lymphocyte recruited to the lung in CD4-deficient humans and mice during Pneumocystis pneumonia, and we have found that these CD8 T cells are responsible for subsequent lung damage in CD4 T cell-depleted mice. Comparing CD4 T cell-depleted IFN- receptor knockout (KO) mice to wild-type mice, we found that this CD8 T cell recruitment and lung damage is type I IFN (IFN-) dependent. However, in both CD4 competent, wild-type and IFN- receptor (IFNAR) KO mice, Pneumocystis infection leads to an eosinophilic granulocyte influx with bronchial epithelial changes as seen in asthma. This response is delayed in IFNAR KO mice, as is pathogen clearance. Although the inflammation is transient in wild-type animals and resolves upon Pneumocystis clearance, it is more severe and persists through day 35 postinfection in IFNAR KO mice, leading to fibrosis. In addition, IFNAR KO, but not wild-type, mice mount a Pneumocystis-specific IgE response, an indicator of allergic sensitization. Thus, in the absence of IFNAR signaling and CD4 T cells, Pneumocystis-mediated lung damage does not occur, whereas in CD4-competent animals, the absence of IFNAR signaling results in an exacerbated Th2 response, asthma-like symptoms, and fibrosis. Therefore, both CD4 T cell- and type I IFN-mediated mechanisms can determine pulmonary complications from Pneumocystis infection.
Introduction
Pneumocystis jiroveci infection rarely causes clinical disease in the general population; however, it causes a severe and life-threatening pneumonia in the immune-compromised host. It is especially prominent in patients with CD4 T cell deficiencies and is one of the most common opportunistic infections in patients suffering from AIDS (1). Although patients with CD4 T cell counts <200–300 cell/μl are very likely to develop Pneumocystis pneumonia, there is increasing evidence for an important role of the innate immune functions of the type I IFN system in the control and susceptibility to opportunistic infections. In this regard there is a subgroup of AIDS patients with relatively high CD4 T cell numbers that are unable to control the development of AIDS-related complications such as Kaposi’s sarcoma, and these patients demonstrate a loss of type I IFN (IFN-)-producing dendritic cell numbers, resulting in low IFN- production. Conversely, normal IFN- production and high numbers of type I IFN-producing cells has been associated with long term survival of AIDS patients despite low CD4 T cell counts and absence of antiretroviral treatment (2, 3, 4, 5, 6, 7).
Pneumocystis species (hereafter referred to as Pneumocystis) has been characterized as a fungus (8), and although it is a strictly extracellular pathogen, in the CD4 T cell-deficient human and animal, there is a vigorous CD8 T cell influx into the alveolar as well as the interstitial space of the lung (9). The role of this influx is controversial; in the mouse model of Pneumocystis pneumonia there are opposing reports on the relevance of CD8 T cells in resistance to and clearance of the infection (9, 10). However, there is good evidence that CD8 cells can be made to clear Pneumocystis murina by the overexpression of IFN- in mouse lung epithelium (11). Regardless of the role of CD8 T cells in resistance to Pneumocystis, there is evidence that CD8 T cells mediate lung damage in the CD4 T cell-depleted host; however, the mechanisms involved are poorly understood (9).
Although IFN- (type I IFN) were initially recognized for their innate antiviral and antitumor activities (12), their importance as regulators of AIDS has recently become apparent. In humans and mice, >10 IFN-- and one IFN- molecules have been identified. All molecules signal via one IFN- receptor (IFNAR), 3 a heterodimer comprised of two chains (IFNAR1 and IFNAR2). They induce dendritic cell maturation, resulting in the secretion of IL-12, which supports Th1 cell differentiation, and IL-15, which induces proliferation and expansion of memory and possibly naive CD8 T cells (13, 14, 15, 16). Although CD8 T cells commonly recognize endogenous Ag in the context of MHC class I (17), type I IFN are important licensing factors for dendritic cells to cross-present exogenous Ag in the context of MHC class I to CD8 T cells, and this has become a current model of CD8 T cell activation in the lymph node, leading to their clonal expansion (18, 19, 20).
By taking advantage of its regulatory influence on the induction of Th1 responses (21, 22, 23), IFN- has recently been used successfully in the treatment of steroid-resistant asthma and related diseases where an exaggerated Th2 immune response leads to eosinophilic lung disease (24, 25).
Because of its important immune regulatory function, especially its influence on Ag presentation to CD8 T cells and CD8 T cell activation, we examined the role of IFNAR signaling in the induction of CD8 T cell-mediated lung damage during Pneumocystis pneumonia (PCP) in CD4-depleted mice as well as the role of IFNAR in clearance of the infection in CD4-competent (undepleted) mice. We found in IFNAR KO mice a significant reduction of CD8 T cell recruitment into the lung during PCP. This was associated with reduced lung damage, improved arterial oxygen saturation, and reduced proinflammatory cytokine secretion into the bronchoalveolar lavage (BAL) fluid compared with CD4 T cell-depleted wild-type animals with the same level of Pneumocystis infection. These findings indicate a critical role of IFNAR signaling in the activation and recruitment of CD8 T cells in CD4-depleted animals and, thus, in the development of immune-mediated damage in PCP.
In the CD4-competent host, in both wild-type and IFNAR KO mice, a strong Th2-mediated eosinophilic inflammation occurred; however, in the absence of IFNAR, the Th2 response to Pneumocystis was exacerbated. This was characterized by more intense and persistent eosinophil accumulation, mucus cell metaphasia, and IgE production, as well as beginning lung fibrosis. Thus, signaling through the IFNAR results in exacerbation of the Th2 response to Pneumocystis when CD4 T cells are present, but results in amelioration of CD8 T cell-mediated lung damage when CD4 T cells are absent.
Materials and Methods
Mice
C.B17 SCID mice, as a source of Pneumocystis organisms, were bred and maintained at Montana State University’s Animal Resource Center. All animal experiments were approved by the institutional animal care and use committee and strictly followed the animal care and handling procedures required by the Animal Welfare Act as well as National Institutes of Health and U.S. Department of Agriculture guidelines. A breeder pair of IFNAR KO mice (animals are on a B6/129 mixed background) was provided by Prof. D. E. Levy (New York University School of Medicine, New York, NY) and were bred and maintained at Montana State University’s Animal Resource Center. Mixed background control mice were purchased from The Jackson Laboratory (B6129PF2/J Stock: 100903).
CD4 T cell depletion
For in vivo CD4 T cell depletion, animals were injected with 300 μg of purified GK1.5 rat anti-mouse CD4 Ab i.p. 2–4 days before infection with Pneumocystis and subsequently injected twice per week until the end of the experiment as previously described (9, 26, 27). The GK1.5 hybridoma cell line was obtained from American Type Culture Collection, and the cells were grown according to the instructions provided and with Abs produced in ascites.
Infection of mice with Pneumocystis
Lung homogenates from Pneumocystis-infected C.B17 SCID mice were used as a source of Pneumocystis nuclei, and CD4-depleted or CD4-undepleted (CD4 T cell-competent) recipient animals were infected with 107 Pneumocystis nuclei in a 100-μl volume as previously described via intratracheal instillation with mice under deep isoflurane gas anesthesia (26, 28).
Blood gas analysis
As a measure of lung function, arterial blood was obtained from the tail artery of each animal into capillary tubes and analyzed at Deaconess Hospital (Bozeman, MT) on an AVL Omni3 Blood gas autoanalyzer (AVL Medical Instruments) for pO2, pCO2, and pH as we have previously described (9).
Assessment of cellular infiltrates into lungs and lymphocyte differentiation using FACS analysis
BALs were performed as previously described (28). Briefly, mice were lethally injected with 90 mg/kg sodium pentobarbital solution i.p. and exsanguinated. Lung lavages were performed by intratracheal cannulation with a total volume of 5 ml of HBSS and 0.5% EDTA. A cell aliquot was spun onto glass slides and stained with Diff-Quik (Dade-Behring) for leukocyte differentiation. Cell counts of BALs were performed with a hemocytometer, and 106 cells/sample were used to assess lymphocyte subset distribution via flow cytometry. For this, a four-color immunofluorescence stain was performed using directly conjugated allophycocyanin-anti-mouse CD4 (L3T4), CyChrome-anti-mouse CD8a (Ly2), PE-anti-mouse CD43 (1B11), and FITC-anti-mouse CD44 (Ly-24, clone IM7) Abs. All Abs were obtained from BD Pharmingen. Before staining, cells were incubated with unlabeled anti-mouse FcR Abs to block FcR-mediated unspecific staining, then incubated with a mixture of the appropriate dilutions of the above Abs for 30 min. Cells were washed three times with PBS/2% FCS and analyzed on a FACSCalibur (BD Biosciences).
Histological analysis of lung tissue
Left lung lobes were inflated with 10% neutral buffered formalin through a tracheal cannula, removed en bloc, and immersed in formalin for at least 24 h, as previously described, before paraffin embedding (29). Sections were cut 5 μm thick on a Leica RM2155 microtome and stained with H&E to assess tissue morphology, with Alcin Blue and periodic acid-Schiff to detect mucus production, and with Masson Tri-Chrome to detect collagen deposition. All reagents were purchased from Richard Allen Scientific and stained according to the manufacturer’s instructions.
Enumeration of Pneumocystis nuclei
Pneumocystis burden was assessed by enumeration of its nuclei as previously described (26). Briefly, lungs of exsanguinated animals were removed, and the right lobe was transferred into 5 ml of HBSS (Invitrogen Life Technologies) and homogenized through mesh screens. A 100-μl aliquot of a 1/20 diluted homogenate was spun onto a glass slide using a Shandon Cytospin 3 centrifuge, smears wee stained with Diff-Quik (Dade-Behring), and the number of Pneumocystis nuclei in 10–50 oil immersion fields was counted. The limit of detection for this technique is log10 4.43 for right lobe homogenates.
Albumin quantification in BAL fluids
Albumin concentrations in BAL fluids were determined as readout for inflammatory exudates and a measure for the extent of lung damage. Albumin reagent (BCG) from Sigma-Aldrich and an albumin standard solution were used for this quantitative colorimetric test. Briefly, standards were prepared in 1/2 dilutions in HBSS/3 mM EDTA ranging from 0.1–2 mg/ml. One hundred microliters of standard per sample was added to the appropriate well in a 96-well plate and mixed with 50 μl of albumin reagent (BCG). Production of green-blue color was immediately read on a ThermoMax microplate reader (Molecular Devices) at 628 nm and analyzed using Softmax software (Molecular Devices).
Cytokine analysis using bead arrays in BALs and tissue culture supernatants
TNF-, IFN-, IL-5, IL-4, IL-2, IL-6, MCP-1, and IL-10 cytokine secretion into the BAL fluid was assessed using the Mouse Th1-Th2 and Mouse Inflammation Cytometric Bead Array assay kit (BD Biosciences) according to the manufacturer’s instructions. Assays were read on a FACScan flow cytometer (BD Biosciences), and data were analyzed using cytometric bead array software (BD Biosciences). In addition, IL-10, IL-13, and TGF-1 concentrations in BAL fluid were measured using a commercially available mouse specific sandwich ELISA (R&D Systems).
Pneumocystis-specific serum IgG and IgE analysis
Pneumocystis-specific IgG and IgE responses were determined using a sandwich ELISA technique. Briefly, 96-well, flat-bottom plates were coated with 10 μg/ml Pneumocystis Ag isolated from lungs of heavily infected source mice as previously described (30) in 0.05 M sodium carbonate buffer, pH 9.6, for 3 h. Plates were washed five times with PBS/0.05% Tween 20 (wash buffer) and blocked overnight with 5% nonfat dry milk powder (Blotto) in PBS. Sample sera were diluted 1/100 in sample buffer (1% BSA/PBS) for Pneumocystis-specific IgG and 1/25 for Pneumocystis-specific IgE, and 100 μl/well was incubated at 37°C for 2 h. Plates were washed six times; 100 μl/well of alkaline phosphatase-conjugated goat anti-mouse IgG (Sigma-Aldrich) at 1/2500 in sample buffer was added to plates for Pneumocystis-specific IgG detection, and 2 μg/ml biotinylated anti-mouse IgE (BD Biosciences; clone R35-118) was added for Pneumocystis-specific IgE detection, and incubation was performed for 1 h at room temperature, then plates were washed six times. Avidin-HRP (Vector Laboratories) was diluted 1/10,000 in sample buffer, and 100 μl/well was added to plates for IgE detection, then incubation was performed for 30 min. Plates were washed again six times, then 50 μl/well ABTS buffer (Sigma-Aldrich) was added to plates for IgE detection, and 50 μl/well p-nitrophenyl-phosphate (Sigma-Aldrich) in dietholamine buffer, pH 9.8 (Sigma-Aldrich), at 1 mg/ml was added to plates for IgG detection, and incubation proceeded for 30 min. Plates were read at 405 nm wavelength on a VersaMax plate reader (Molecular Devices). Samples were read for OD, and positive and negative control sera were included in the assay.
Total serum IgE was measured using a mouse IgE capture and detection Ab pair plus mouse IgE standard from BD Biosciences, and the assay was performed according to the manufacturer’s protocol.
Assessment of airway hyperreactivity in response to methacholine challenges using a whole-body plethysmograph
Airway hyperresponsiveness was measured by induction of the enhanced pause (penh) in the airflow upon exposure to increasing dosages of aerosolized methacholine (0, 2.5, 5, 10, and 20 mg/ml) in a whole-body plethysmograph for rodents (Buxco). Between each methacholine dosage, animals were allowed to rest for up to 15 min to return to baseline values. The lowest dosage of methacholine that induced significant differences in penh values between the groups was compared.
Statistical analysis
For statistical analysis, a one-way ANOVA was performed, and pairwise comparisons were made using the post-hoc Tukey test. Data were assumed to be significantly different at p < 0.05. Data are shown as the mean ± SEM.
Results
Type I IFN signaling is required in the induction of immune-mediated lung damage during the course of PCP in CD4-depleted mice
To assess the role of type I IFNs in the activation of CD8 T cells in the course of PCP, IFNAR KO mice were CD4 T cell-depleted and Pneumocystis-infected via intratracheal instillation of 107 nuclei of Pneumocystis. The course of disease progression and lung damage was assessed on day 28 postinfection and compared with the course of disease in CD4 T cell-depleted wild-type animals. Day 28 postinfection was chosen because we have previously shown that lung damage in Pneumocystis-infected, CD4 T cell-depleted animals is well established by this time, although the animals are not yet terminally sick (9). Each experiment included a control group consisting of uninfected wild-type animals, because previous comparative analysis of uninfected lungs from IFNAR KO mice and wild-type mice showed no differences with regard to lung structure and cellular composition of BAL cells (>95% alveolar macrophages). Each experimental group consisted of five animals, and each experiment was repeated three times. Arterial blood gases were taken from all groups, animals were then killed, and Pneumocystis lung burden, lung histopathology, and airway inflammatory responses were determined. Blood gas analysis revealed a significantly better arterial oxygenation of IFNAR KO mice on day 28 postinfection compared with that in wild-type animals (Fig. 1A), and albumin levels, as a measure for vascular leakage, were significantly reduced in IFNAR KO mice compared with the infected wild-type animals (Fig. 1B). Pneumocystis burden was not different in the two groups and cannot account for the differences in measures of lung function and damage (Fig. 1C).
FIGURE 1. CD4 T cell-depleted IFNAR KO mice show better lung function and reduced lung damage during the course of PCP than CD4-depleted wild-type animals. Arterial blood gases were taken as a measure of lung function (A), and albumin leakage into the alveolar spaces was assessed in BAL fluid as a measure of lung damage (B). The Pneumocystis burden of the lung was assessed via Pneumocystis nuclei count of lung homogenate cytospin slides (C). Significant differences between the groups are indicated: *, p < 0.05; **, p < 0.01; ***, p < 0.001. Significant differences between infected IFNAR KO and wild-type animals are indicated by an asterisk above the bar of the wild-type group; significant differences between the infected IFNAR KO group and the uninfected control group are indicated by an asterisk above the bar for the IFNAR KO group. IFNAR KO mice showed better arterial pO2 levels and less lung damage than wild-type mice during the course of PCP, which was not due to a lesser Pneumocystis burden of the lungs compared with wild-type animals.
Leukocyte populations recruited into the BAL fluids were assessed by differential cell count and revealed a similar lung accumulation of neutrophils and macrophages in wild-type and IFNAR KO mice, with a slightly reduced accumulation of lymphocytes in IFNAR KO mice (data not shown). Lymphocyte subset analysis via FACS confirmed depletion of CD4 T cells using an anti-CD4 Ab (clone L3T4) that does not compete with the GK1.5 mAb epitope. Furthermore, CD8 T cells were the predominant lymphocyte population recruited to the lung, with a similar activation profile in the two groups as assessed by the percentage of CD44- and CD43high-expressing CD8 T cells (data not shown). However, total lymphocyte numbers were significantly lower in BAL fluid of IFNAR KO mice compared with that of wild-type mice due to a reduced influx of activated CD8 T cells (Fig. 2). In agreement with these findings, cytokine analysis in the BAL fluid showed significantly reduced secretion of the proinflammatory cytokines TNF-, IFN-, IL-6, and MCP-1 (Fig. 3), because they can be secreted by activated CD8 T cells and other cells. The levels of the anti-inflammatory cytokines IL-10 and TGF- were not significantly different between the groups of mice (data not shown).
FIGURE 2. Lymphocyte numbers are reduced in BAL fluids of CD4-depleted IFNAR KO mice compared with wild-type animals during the course of PCP due to reduced influx of activated CD8 T cells. Lymphocyte numbers were assessed by performing total leukocyte count using a hemocytometer and differential cell counts of leukocytes in BAL fluid using Diff-Quik-stained cytospin slides. Lymphocyte differentiation and their activation profile were performed via FACS analysis staining for CD4, CD8, CD44, and CD43 surface markers. A, Comparative analysis of absolute lymphocyte numbers assessed in BAL fluid in IFNAR KO mice compared with wild-type mice. Total CD8 T cell (B) and activated CD44/CD43+ CD8 T cell numbers (C) were calculated based on the percentage of total lymphocytes. As no lymphocytes were detectable in BAL fluid of uninfected control animals, absolute lymphocyte numbers as well as subset values were zero. Statistical differences between infected IFNAR KO and wild-type animals are indicated above the bars of the wild-type group: *, p < 0.05; **, p < 0.01; ***, p < 0.001.
FIGURE 3. Proinflammatory cytokine levels are significantly lower in BAL fluid of IFNAR KO mice compared with wild-type animals during the course of PCP. Cytokine concentrations as a measure of inflammatory activity were assessed in BAL fluids of CD4 T cell-depleted IFNAR KO and wild-type mice suffering from PCP using the inflammation bead array (BD Biosciences). TNF- (A), IFN- (B), IL-6 (C), and MCP-1 (D) were significantly lower in IFNAR KO mice. Cytokine concentrations in the BAL fluid of uninfected control animals were between 0 and 5 pg/ml for all cytokines and may appear as undetectable in the appropriate graphs. Statistical differences between infected IFNAR KO and wild-type animals are indicated above the data bars of the wild-type group: *, p < 0.05; **, p < 0.01; ***, p < 0.001.
Lack of IFNAR signaling leads to a delayed, but complete, clearance of Pneumocystis infection in CD4-competent animals
We have previously found that in immune-competent, wild-type animals, Pneumocystis infection is cleared by days 14–21, with minimal or no residual inflammation of lung tissue by day 28 postinfection. Because of the reduced inflammation seen in CD4-depleted IFNAR KO mice in response to Pneumocystis infection, we determined whether clearance of the infection would be impaired in CD4 T cell-competent (undepleted) mice. Therefore, we infected both CD4-competent, wild-type and IFNAR KO mice with Pneumocystis intratracheally and evaluated Pneumocystis clearance, Pneumocystis-specific serum immune globulin levels, and the inflammatory response induced on days 14, 21, and 28 postinfection.
Fig. 4A shows the kinetics of Pneumocystis clearance over a period of 14, 21, and 28 days. In wild-type animals, Pneumocystis infection was barely above the microscopic detection limit on days 14 and 21 postinfection and was completely undetectable on day 28. In contrast, in IFNAR KO animals, a slight growth occurred by day 14, which was even greater on day 21. However, the infection was completely cleared by day 28. This delayed clearance pattern in IFNAR KO mice was associated with a delayed induction of a Pneumocystis-specific IgG response, with no Pneumocystis-specific Abs detectable on day 14, a subsequent increase in production by day 21, and normal Pneumocystis-specific IgG levels on day 28 compared with the IgG response in wild-type animals (Fig. 4B). BAL fluid cell counts in wild-type animals were highest on day 14 and decreased consecutively over the course of clearance, whereas in IFNAR KO mice, the cellular influx into lung airways did not peak until day 21 (Fig. 4C).
FIGURE 4. Clearance of Pneumocystis infection is delayed, but complete, in CD4-competent IFNAR KO mice compared with wild-type animals due to delayed induction of both cellular and humeral immune responses. CD4-competent IFNAR KO and wild-type animals were infected with Pneumocystis and euthanized on day 14, 21, or 28 postinfection. Pneumocystis lung burden was assessed via Pneumocystis nuclei counts in lung homogenates. Although the level of infection in both wild-type animals () and IFNAR KO mice () was low 14 days postinfection (limit of detection, 4.43 log10 nuclei) and was almost completely cleared on day 21 in wild-type animals, IFNAR KO mice showed significant Pneumocystis growth by day 21 and cleared the infection by day 28 postinoculation (A). The delayed clearance pattern is associated with the delayed induction of a Pneumocystis-specific serum IgG response (B) as well as delayed influx of cells into the lungs of infected IFNAR KO mice (C) compared with wild-type mice. Statistical differences between infected IFNAR KO- and wild-type animals are indicated above the data bars of the wild-type group: *, p < 0.05; **, p < 0.01; ***, p < 0.001.
Pneumocystis infection of CD4 T cell-sufficient mice leads to transient eosinophilic immune response in wild-type animals, resulting in pathological changes similar to those of asthma. This response is prolonged and exaggerated in IFNAR KO animals and leads to pulmonary fibrosis. BAL fluid leukocyte differentiation revealed that clearance of the pathogen in both wild-type and IFNAR KO animals was associated with a strong eosinophil granulocyte influx. This influx was again delayed in IFNAR KO mice to day 21 (Fig. 5A). Because eosinophilic airway inflammation can be associated with increased airway hyperreactivity as in allergic asthma, animals were exposed to increasing dosages of methacholine, and airway resistance was assessed using a whole-body plethysmograph for rodents. On day 14 postinfection, wild-type animals showed transiently increased penh values in response to low dose methacholine (2.5 mg/ml), whereas IFNAR KO mice did not. However, on day 28 postinfection, IFNAR KO animals demonstrated increased penh values (Fig. 5B), whereas wild-type animals had returned to baseline levels. BALB/c as well as C57BL/6 wild-type mice were tested in the same experimental setting, and similar transient airway eosinophilia as well as airway hyperreactivity were found (data not shown) (S. Swain, and A. Harmsen, manuscript in preparation) during the course of clearance of the infection, indicating a general immune phenomenon rather than a strain-specific response. To further exclude that eosinophilic influx was not due to an Ag incompatibility reaction due to the instillation of Pneumocystis-infected mouse lung homogenate derived from SCID mice on a BALB/c genetic background into animals with a mixed (B6/129) genetic background, B6/129 mixed background wild-type animals were either intratracheally instilled with lung homogenate from an uninfected allogeneic B6/129 mouse or an uninfected BALB/c SCID mouse and compared with mice receiving the instillation of Pneumocystis-infected lung homogenate from BALB/c SCID mice used in the experimental settings described above. Intratracheal instillation of uninfected lung homogenates from BALB/c SCID mice or allogeneic B6/129 wild-type mice did not lead to any detectable eosinophilic cell response in BAL fluids on day 14 postinstillation, whereas the instillation of Pneumocystis-infected lung homogenates did, indicating that the eosinophilic response was due to the presence of Pneumocystis pathogen in the lung homogenate (data not shown).
FIGURE 5. Eosinophils are the major leukocyte population recruited to the lung during the course of Pneumocystis clearance, and acute airway hyperresponsiveness is associated with this transient eosinophil influx. Leukocyte differentiation of cells in BAL fluid of both IFNAR KO mice () and wild-type animals () revealed that the major population recruited to the lung during the course of clearance was eosinophils. The influx of these cells was again delayed in IFNAR KO animals (A). Airway hyper-responsiveness was assessed in response to methacholine and was measured as the penh by whole-body plethysmography. Increased airway hyperresponsiveness was detected at 2.5 mg/ml methacholine on day 14 in wild-type animals and was delayed until day 28 in IFNAR KO mice (B). Statistical differences between infected IFNAR KO and wild-type animals are indicated: *, p < 0.05; **, p < 0.01; ***, p < 0.001.
Histological analysis of formalin-fixed lung tissue (Fig. 6) in wild-type mice on day 14 revealed an intense Th2-like immune response and confirmed recruitment of eosinophils and lymphocytes into the lung airways and interstitium as well as giant cell formation and epithelial and goblet cell metaplasia of the bronchi resembling an asthmatic inflammatory response. In contrast, only mild inflammation was detected on day 14 postinfection in lungs of IFNAR KO mice, with no eosinophil accumulation detectable, but by this time, some lymphocyte recruitment had occurred as well as the formation of some giant cells (Fig. 6, top row). On day 21, however, lungs of IFNAR KO mice showed an even more severe inflammation in response to the infection than that on day 14 in wild-type animals with massive eosinophil and lymphocyte infiltrate, giant cell formation, and mucus cell hyperplasia with almost complete consolidation of the lung tissue. In contrast, wild-type animals were already resolving the response, with only minor peribronchial eosinophil and lymphocyte infiltrate remaining, and the bronchial epithelia height had nearly returned normal (Fig. 6, second row). On day 28 postinfection, wild-type animals showed only minimal residual inflammation, whereas IFNAR KO mice still showed severe asthmatic inflammation similar to that on day 21 (Fig. 6, third row). Because the intense Th2-like response in IFNAR mice persisted, we repeated the experiments and chose day 35 instead of day 28 as our latest time point. The time course of the inflammation in the second experiment mirrored the previous findings and showed a still vigorous eosinophilic inflammation of the lung present through day 35 postinfection, with prominent thickening of the bronchial epithelium, goblet cell hyperplasia, and peribronchial as well as alveolar cellular infiltrate (Fig. 6, bottom row). In addition to the H&E stains, we performed Alcin Blue/periodic acid-Schiff stains, which stained mucus glycoprotein and confirmed mucus cell hyperplasia as a hallmark of asthma throughout day 35 in both wild-type and IFNAR KO animals (Fig. 7, rows 1 and 2). Chronic inflammation of the lung can lead to irreversible changes in the form of tissue fibrosis with loss of tissue elasticity and reduced gas exchange. Masson Tri-Chrome staining of formalin-fixed lung tissue revealed prominent interstitial as well as peribronchial collagen deposition as a sign of lung fibrosis in IFNAR KO mice on day 35 postinfection, whereas lung tissue of wild-type animals at the same time point looked like that in uninfected controls (Fig. 7, row 4).
FIGURE 6. Eosinophilic inflammation was apparent in both CD4-competent IFNAR KO and wild-type animals, with lung changes similar to those seen in asthma. However, the response was delayed, but exacerbated and prolonged, in IFNAR KO mice. Shown are photomicrographs of formalin-fixed and H&E-stained lung sections taken at x10 and x200 magnifications. Eosinophilic lung inflammation with bronchial cell hypertrophy (*) and giant cell formation (<) can be seen in all sections. Delayed onset of inflammation in IFNAR KO mice on day 14 (first row) was noticed and was more severe on day 21 in IFNAR KO mice compared with that on day 14 in wild-type animals. Although the inflammation in wild-type animals waned to almost complete resolution on days 28 and 35 (rows 3 and 4), inflammation remained almost unchanged in IFNAR KO mice through day 35 despite clearance of the pathogen by day 28 (rows 2–4).
FIGURE 7. PAS/Alcin Blue stain confirmed mucus secretion from hypertrophic goblet cells in both CD4-competent, wild-type and IFNAR KO animals in response to the Pneumocystis infection, and Masson Tri-Chrome stain revealed fibrotic lung changes in IFNAR KO mice in response to the prolonged ongoing inflammation. One characteristic feature of asthma is increased mucus cell secretion, which can lead to mucus plugging of the airways. PAS/Alcin Blue stained mucus glycoproteins dark blue, was induced in both wild-type and IFNAR KO animals, and remained throughout day 35 postinfection (rows 1 and 2). Uninfected control animals did not show positive mucus staining of their airways (row 3, left photograph). Fibrosis of tissue can occur in response to chronic inflammation and is a concern in chronic asthma (remodeling of the lung). The Masson Tri-Chrome procedure stained collagen depositions in the tissue blue and showed peribronchial and interstitial changes on day 35 postinfection only in IFNAR KO mice (row 4, *).
Pneumocystis exposure in CD4-competent mice leads to Pneumocystis-specific allergic sensitization
Because of the transient histological changes in wild-type animals and the prolonged inflammation in IFNAR KO animals resembling allergic inflammation, we determined whether Pneumocystis could induce a specific IgE response as found during allergic sensitization. Total IgE was detectable in both wild-type and IFNAR KO animals (Fig. 8A), but only in IFNAR KO mice did total IgE increase over 28 days, although this was not statistically significant compared with the wild-type group. However, a dramatic Pneumocytis-specific IgE response was only induced in the IFNAR KO group beginning on day 21 postinfection, with highest readings on day 28 (Fig. 8B). Therefore, a delayed clearance of the pathogen in IFNAR KO mice may have been responsible for the induction of the Pneumocystis-specific IgE response changing Pneumocystis from an immunological Ag to an allergen.
FIGURE 8. Pneumocystis-specific serum IgE is induced in CD4-competent IFNAR KO, but not in wild-type, animals during the course of Pneumocystis clearance. Total serum IgE (A) as well as Pneumocystis-specific serum IgE (B) were measured using a sandwich ELISA technique. Total IgE increased in both wild-type and IFNAR KO mice, but Pneumocystis-specific IgE was only induced in IFNAR KO mice. Statistical differences between infected IFNAR KO and wild-type animals are indicated: *, p < 0.05; **, p < 0.01; ***, p < 0.001.
Cytokine profile analysis of BAL fluids from both groups over the above time course revealed a Th2 pattern (Fig. 9, A–C) with increased production of IL-5, IL-13, and TGF-. Highest cytokine concentrations were measured on day 14 in wild-type animals and on day 21 in IFNAR KO animals, and the induction of these cytokines was comparable between the two groups at their individual peak time points. IFN- concentrations were unchanged during the time course of clearance in both groups (Fig. 9D). A second, independent experiment showed similar results. However, in this experiment, IL-10 was also measured, because it has been identified to down-regulate specific immune responses. Wild-type mice produced significant amounts of IL-10 on day 14 postinfection, but no IL-10 was produced on day 14, and little was produced on day 21 postinfection in IFNAR KO mice (Fig. 9E).
FIGURE 9. Cytokine analysis of BAL fluid from both CD4-competent, wild-type, and IFNAR KO animals showed a Th2-like cytokine profile. Cytokine analysis of BAL fluid from both CD4-competent IFNAR KO and wild-type animals was performed using a Th1/Th2 cytokine bead array assay (BD Biosciences) as well as a sandwich ELISA for IL-13, TGF-, and IL-10 (R&D Systems). Levels of IL-5, IL-13, and TGF- (A–C) were highest on day 14 in wild-type mice and were comparable on day 21 in IFNAR KO animals, confirming the delayed, but then similar, cytokine response in IFNAR KO mice. IFN- secretion did not vary over the course of clearance (D). IL-10 secretion was significantly detectable on day 14 in wild-type animals; however, it was only weakly induced on day 21 in IFNAR KO mice (E). Statistical differences between infected IFNAR KO and wild-type animals are indicated: *, p < 0.05; **, p < 0.01; ***, p < 0.001.
Discussion
In humans and mouse models of PCP, it has been shown that in the absence of CD4 T cells, a vigorous CD8 T cell influx into the infected lung occurs during the course of the disease, which mediates the occurring lung damage (9).
We found that CD4-depleted IFNAR KO mice with disrupted signaling for both IFN- and IFN- showed significantly better lung function as well as less lung damage compared with wild-type animals, although the Pneumocystis burden of both groups was identical. This was associated with significantly reduced influx of activated CD8 T cells into the lungs of infected IFNAR KO mice. The mechanisms of CD8 T cell-mediated lung damage during the course of PCP are not clear, but CD8 effector cells have been reported to secrete proinflammatory cytokines/chemokines such as IFN-, TNF-, and RANTES (31, 32), which are involved in effector functions and can induce target cell death. Wright et al. (33) suggested a role for TNF- receptor signaling in the mechanisms of CD8 T cell-mediated lung damage during the course of PCP by influencing the secretion of other proinflammatory cytokines and chemokines, such as TNF-, keratinocyte-derived cytokine, RANTES, and MCP-1. We found in IFNAR KO mice, compared with wild-type mice, significantly reduced concentrations of proinflammatory cytokines, including TNF-, IFN-, MCP-1, and IL-6, all of which could be produced by activated CD8 T cells. This occurred despite vigorous Pneumocystis growth in the lungs of both groups. These findings indicate that the induction of these cytokines and recruitment of CD8 T cells to the lung require type I IFN-dependent mechanisms during the course of PCP.
CD8 T cell-mediated mechanisms of damage have been implicated in other lung diseases, such as extrinsic allergic alveolitis or hypersensitivity pneumonitis (34). The immune mechanisms by which CD8 T cells are activated and what Ag they detect in these diseases is unclear; however, similar pathological sequelae can be seen in interstitial lung diseases, that occur in the context of autoimmune diseases such as rheumatoid arthritis (35). Furthermore, type I IFN-dependent mechanisms have been implicated in the pathogenesis of other autoimmune diseases, such as systemic lupus erythematosus and diabetes mellitus type I, and treatment of these patients with anti-IFN- mAbs has been suggested (36, 37). That lung damage in PCP in the absence of CD4 T cells is dependent on IFNAR signaling suggests that a therapeutic approach of neutralizing IFN- signaling of classic PCP could alleviate lung damage in this disease. Thus, we determined whether a deficiency in IFNAR signaling would also affect host resistance to Pneumocystis in a CD4-competent animal. Previous pathogenesis studies revealed that IFNAR KO mice are highly susceptible to infections by a broad array of different viruses (e.g., Poxviridae, Arenaviridae, Rhabdoviridae, and Togaviridae), yet have normal resistance to the microbial pathogen Listeria monocytogenes, and it has been established that the type I IFN system (IFN-) and the type II IFN system (IFN-) are nonredundant (12, 38). In a study using influenza virus infection, the inflammation that occurred in both wild-type animals and IFN- KO mice was predominantly lymphocytic, whereas in IFNAR KO mice, it was predominantly eosinophilic (39). Interestingly, when we compared Pneumocystis clearance in CD4-competent IFNAR KO mice to that in wild-type mice, we found that both animal groups responded with a severe Th2-like immune response, with massive recruitment of eosinophils into the lung tissue as well as bronchial epithelia cell hyperplasia with mucus production and giant cell formation. However, IFNAR KO mice showed a delayed onset of inflammation (day 21 vs day 14 in wild-type animals), followed by a prolonged and augmented Th2-like immune response, resulting in exacerbated lung pathology. This was associated with delayed clearance of the pathogen due to delayed induction of a Pneumocystis-specific IgG Ab response (complete clearance in wild-type animals by day 14; in IFNAR KO mice by day 28). These data indicate that a type I IFN response is not necessary for pathogen clearance, yet it is necessary for the timing and modulation of induction of the protective immune response. Why the Pneumocystis-specific IgG response is delayed in IFNAR KO mice is not clear; however, the absence of IFNAR signaling could lead to delayed maturation of professional APCs and, therefore, specific Th cells. Furthermore, the importance of IFN- in conjunction with IL-6 in the induction of plasma cell differentiation and immune globulin secretion has recently been described (40), and the loss of IFNAR signaling could directly influence and delay plasma cell differentiation and Ab secretion in response to Pneumocystis infection.
Pneumocystis infection in IFNAR KO mice, compared with that in wild-type mice, resulted in exacerbated, asthma-like responses. Interestingly, despite concerns about side effects, IFN- treatment has been shown to be a successful option in the treatment of steroid-resistant asthma, and its mechanisms of actions are thought to involve the promotion of a Th1, rather than a Th2, immune response, thus disrupting the cycle of eosinophilic inflammation and airway hyperreactivity (24, 25). Furthermore, it was shown that histamine, released during allergic inflammation, inhibits the secretion of IFN- from plasmacytoid dendritic cells, which could promote a strong Th2 response, as in allergic inflammation. This might be an explanation for the decreased type I IFN levels found in atopic patients compared with healthy individuals (41). These findings strongly suggest a role of type I IFN signaling in modulating allergic inflammation in humans. The importance of type I IFN in allergic sensitization has previously been demonstrated in a mouse model of allergic sensitization to OVA comparing wild-type animals to animals lacking the IFN- gene (42). In the absence of IFN-, animals showed a more severe pulmonary eosinophilic inflammation, with goblet cell hypertrophy and increased production of Th2 cytokines as well as OVA-specific IgE in response to the sensitization protocol, compared with wild-type or heterozygous littermates. Also, increased OVA-specific T cell proliferation indicated an important role for type I IFN in the modulation and inhibition of allergic sensitization. Although both wild-type and IFNAR KO mice responded with a strong Th2-mediated immune response to Pneumocystis infection with changes resembling allergic (type I hypersensitivity) reactions, this was only transient in wild-type animals and waned fast. However, in animals with a defect in the IFNAR signaling pathway, a prolonged asthma-like inflammatory response of the lungs was noted, with signs of beginning fibrotic tissue remodeling, as seen in chronic asthma (43); only in this group was Pneumocystis-specific IgE production detectable. This can be regarded as evidence for allergic sensitization to the pathogen, which could, in turn, promote an IgE-mediated allergic inflammatory response upon re-exposure to the pathogen in IFNAR KO mice. Why allergic sensitization occurred in IFNAR KO mice and not in wild-type mice is unknown. However, significant secretion of IL-10 was only seen in wild-type mice before resolution of the inflammation. IL-10 has been shown to be secreted by specific regulatory T cells, which have been described to play a key role in the specific down-regulation of Th2 responses and have been considered important in the down-regulation of allergic inflammations (44, 45, 46). Therefore, the absence of IFNAR signaling may have affected the induction of regulatory T cells because of impaired maturation of regulatory dendritic cells (47). However, additional characterization of T cell and APC subsets in the lung during the course of Pneumocystis infection needs to be performed.
In summary, we found that during the course of PCP, in the absence of CD4 T cells, a type I IFN-dependent, vigorous CD8 T cell response is induced. This suggests that type I IFN could be a therapeutic target to modulate the CD4 T cell-independent inflammatory response during the course of PCP and possibly other extrinsic allergic pneumonitis-type responses. However, in a CD4-competent host, the immune response induced during the course of clearance of Pneumocystis resembles histological lung changes that occur during a classical asthmatic inflammation. Although IFNAR signaling does not appear to be critical for the clearance of the infection, it modulates and down-regulates this inflammation upon clearance of the pathogen, leading to complete recovery of the lung without residual inflammation. However, the absence of IFNAR signaling in a CD4-competent animal leads to a prolonged, Th2-mediated, eosinophilic lung pathology as well as an allergic sensitization to the pathogen with Pneumocystis-specific IgE production, possibly due to the loss of IL-10-producing regulatory T cells. Because mechanisms of allergic sensitization are still poorly understood, yet genetic as well as environmental factors are thought to contribute to the onset of the disease (48, 49), we propose that exposure to Pneumocystis can induce allergic sensitization to the pathogen in the presence of impaired type I IFN signaling, which may induce a continuous aggravation during re-exposure. Furthermore, based on the induction of a strong Th2-mediated immune response during the natural course of clearance of Pneumocystis in the lung, exposure of an individual with pre-existing asthma to Pneumocystis may lead to severe exacerbation of asthma or other obstructive lung diseases. Interestingly, a study by Morris et al. (50) found an association of chronic obstructive pulmonary lung disease severity and Pneumocystis colonization of the lung in patients with chronic obstructive pulmonary lung disease. Furthermore, there are increasing reports of pulmonary complications in AIDS patients, such as emphysema, pulmonary hypertension, noninfectious inflammatory disease, and asthma, as well as the clinical presentation of PCP as asthma (51, 52, 53). It is likely that depending on the immune status of a patient infected with HIV, the clinical manifestation of Pneumocystis infection is variable, and chronic colonization (that may or may not be detectable) in a partially immune-reconstituted patient may lead to pulmonary complications other than classical PCP.
Acknowledgments
We thank Soo Han and Pete Degel for excellent technical assistance.
Disclosures
The authors have no financial conflict of interest.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by National Institutes of Health Grant P20RR16455-01 from the BRIN Program of the National Center for Research Resources as well as National Institutes of Health Grants RO1HL55002, P01HL71659, and P20RR020185.
2 Address correspondence and reprint requests to Dr. Nicole N. Meissner, Veterinary Molecular Biology Department, Montana State University, Molecular Bioscience Building, 960 Technology Boulevard, Bozeman, MT 59718. E-mail address: nicolem@montana.edu
3 Abbreviations used in this paper: IFNAR, IFN- receptor; BAL, bronchoalveolar lavage; KO, knockout; PCP, Pneumocystis pneumonia; penh, enhanced pause.
Received for publication October 27, 2004. Accepted for publication February 7, 2005.
References
Sepkowitz, K. A.. 2002. Opportunistic infections in patients with and patients without acquired immunodeficiency syndrome. Clin. Infect. Dis. 34:1098.
Siegal, F. P., P. Fitzgerald-Bocarsly, B. K. Holland, M. Shodell. 2001. Interferon- generation and immune reconstitution during antiretroviral therapy for human immunodeficiency virus infection. AIDS 15:1603.
Siegal, F.. 2003. Interferon-producing plasmacytoid dendritic cells and the pathogenesis of AIDS. Res. Initiat. Treat. Action 8:10.
Soumelis, V., I. Scott, F. Gheyas, D. Bouhour, G. Cozon, L. Cotte, L. Huang, J. A. Levy, Y. J. Liu. 2001. Depletion of circulating natural type 1 interferon-producing cells in HIV-infected AIDS patients. Blood 98:906.
Pacanowski, J., S. Kahi, M. Baillet, P. Lebon, C. Deveau, C. Goujard, L. Meyer, E. Oksenhendler, M. Sinet, A. Hosmalin. 2001. Reduced blood CD123+ (lymphoid) and CD11c+ (myeloid) dendritic cell numbers in primary HIV-1 infection. Blood 98:3016.
Soumelis, V., I. Scott, Y. J. Liu, J. Levy. 2002. Natural type 1 interferon producing cells in HIV infection. Hum. Immunol. 63:1206.
Levy, J. A.. 2001. The importance of the innate immune system in controlling HIV infection and disease. Trends Immunol. 22:312.
Stringer, J. R.. 1996. Pneumocystis carinii: what is it, exactly?. Clin. Microbiol. Rev. 9:489
Wright, T. W., F. Gigliotti, J. N. Finkelstein, J. T. McBride, C. L. An, A. G. Harmsen. 1999. Immune-mediated inflammation directly impairs pulmonary function, contributing to the pathogenesis of Pneumocystis carinii pneumonia. J. Clin. Invest. 104:1307.
Beck, J. M., R. L. Newbury, B. E. Palmer, M. L. Warnock, P. K. Byrd, H. B. Kaltreider. 1996. Role of CD8+ lymphocytes in host defense against Pneumocystis carinii in mice. J. Lab. Clin. Med. 128:477.
Kolls, J. K., S. Habetz, M. K. Shean, C. Vazquez, J. A. Brown, D. Lei, P. Schwarzenberger, P. Ye, S. Nelson, W. R. Summer, et al 1999. IFN- and CD8+ T cells restore host defenses against Pneumocystis carinii in mice depleted of CD4+ T cells. J. Immunol. 162:2890.
Samuel, C. E.. 2001. Antiviral actions of interferons. Clin. Microbiol. Rev. 14:778.
Levy, D. E., I. Marie, A. Prakash. 2003. Ringing the interferon alarm: differential regulation of gene expression at the interface between innate and adaptive immunity. Curr. Opin. Immunol. 15:52.
Taki, S.. 2002. Type I interferons and autoimmunity: lessons from the clinic and from IRF-2-deficient mice. Cytokine Growth Factor Rev. 13:379.
Zhang, X., S. Sun, I. Hwang, D. F. Tough, J. Sprent. 1998. Potent and selective stimulation of memory-phenotype CD8+ T cells in vivo by IL-15. Immunity 8:591
Schluns, K. S., K. Williams, A. Ma, X. X. Zheng, L. Lefrancois. 2002. Cutting edge: requirement for IL-15 in the generation of primary and memory antigen-specific CD8 T cells. J. Immunol. 168:4827.
Wong, P., E. G. Pamer. 2003. CD8 T cell responses to infectious pathogens. Annu. Rev. Immunol. 21:29.
Norbury, C. C., L. J. Sigal. 2003. Cross priming or direct priming: is that really the question?. Curr. Opin. Immunol. 15:82
Le Bon, A., N. Etchart, C. Rossmann, M. Ashton, S. Hou, D. Gewert, P. Borrow, D. F. Tough. 2003. Cross-priming of CD8+ T cells stimulated by virus-induced type I interferon. Nat. Immunol. 4:1009.
Beignon, A. S., M. Skoberne, N. Bhardwaj. 2003. Type I interferons promote cross-priming: more functions for old cytokines. Nat. Immunol. 4:939.
Taki, S., T. Sato, K. Ogasawara, T. Fukuda, M. Sato, S. Hida, G. Suzuki, M. Mitsuyama, E. H. Shin, S. Kojima, et al 1997. Multistage regulation of Th1-type immune responses by the transcription factor IRF-1. Immunity 6:673.
Hida, S., K. Ogasawara, K. Sato, M. Abe, H. Takayanagi, T. Yokochi, T. Sato, S. Hirose, T. Shirai, S. Taki, et al 2000. CD8+ T cell-mediated skin disease in mice lacking IRF-2, the transcriptional attenuator of interferon-/ signaling. Immunity 13:643.(Nicole N. Meissner, Steve)