ABSTRACT
Pneumocystis pneumonia was first diagnosed in malnourished children and has more recently been found in children with upper respiratory symptoms. We previously reported that there is a significant delay in the immune response in newborn mice infected with Pneumocystis compared to adults (Garvy BA, Harmsen AG, Infect. Immun. 64:3987–3992, 1996, and Garvy BA, Qureshi M, J. Immunol. 165:6480–6486, 2000). This delay is characterized by the failure of neonatal lungs to upregulate proinflammatory cytokines and attract T cells into the alveoli. Here, we report that regardless of the age at which we infected the mice, they failed to mount an inflammatory response in the alveolar spaces until they were 21 days of age or older. Anti-inflammatory cytokines had some role in dampening inflammation, since interleukin-10 (IL-10)-deficient pups cleared Pneumocystis faster than wild-type pups and the neutralization of transforming growth factor beta (TGF-β) with specific antibody enhanced T cell migration into the lungs at later time points. However, the clearance kinetics were similar to those of control pups, suggesting that there is an intrinsic deficiency in the ability of innate immunity to control Pneumocystis. We found, using an adoptive transfer strategy, that the lung environment contributes to association of Pneumocystis organisms with alveolar macrophages, implying no intrinsic deficiency in the binding of Pneumocystis by neonatal macrophages. Using both in vivo and in vitro assays, we found that Pneumocystis organisms were less able to stimulate translocation of NF-κB to the nucleus of alveolar macrophages from neonatal mice. These data indicate that there is an early unresponsiveness of neonatal alveolar macrophages to Pneumocystis infection that is both intrinsic and related to the immunosuppressive environment found in neonatal lungs.
INTRODUCTION
Pneumocystis species are opportunistic fungal pathogens that cause pneumonia in immunosuppressed mammalian hosts. Outbreaks of Pneumocystis pneumonia (PCP) were first described in orphan homes among malnourished children (18). Most children encounter Pneumocystis infection by the age of 2 years, as evidenced by the presence of specific antibodies in the peripheral blood (33), and recent studies suggest that up to 30% of young children carry Pneumocystis in the lungs (50, 51). Clearance of Pneumocystis is dependent on a functional CD4+ T cell compartment as well as on B cells and alveolar macrophages (23, 28–30, 42). Infants with AIDS tend to have a more fulminate course of PCP than do adults with AIDS, possibly due to the immaturity of the immune system (8, 39).
Susceptibility of newborn infants to infection is thought to be due to the inexperience of the adaptive immune system along with functional deficits of both innate and adaptive immune responses (1, 26). We have previously published that it takes 3 weeks for mice infected at 2 days of age to mount a CD4+ T cell response into the alveolar spaces in response to mouse-specific Pneumocystis murina infection (20, 21, 37). In contrast, CD4+ T cells infiltrated the alveoli within 4 to 5 days postinfection in adult mice challenged with a comparable dose. This delay in lung T cell infiltration is associated with delayed infiltration and activation of macrophages, delayed chemokine and cytokine production, and delayed adhesion molecule upregulation (12, 13, 35–37). We also found that lung mRNA expression of transforming growth factor β2 (TGF-β2) and TGF-β3 isoforms was upregulated in the uninfected lungs of infant mice, and TGF-β1 mRNA expression was similar in adult and infant mice (37).
Since TGF-β is known to have anti-inflammatory activity, we hypothesized that TGF-β is constitutively expressed in postnatally developing lungs as a protective mechanism against overly exuberant inflammatory responses to infectious agents or foreign particles. Rapid alveolarization takes place during the first 2 weeks after birth in mice and the first 6 months in human infants (7, 25, 48, 54). Inflammation in the lungs during this time has the potential to cause significant damage that could be irreversible. To determine whether the immune response to P. murina is developmentally regulated, we infected mice at various times after birth to determine when the immune response was initiated. We also examined TGF-β protein levels in the lungs of infant and adult mice. Our data confirm that TGF-β protein levels are developmentally controlled in the lungs and that TGF-β contributes to a delayed immune response to P. murina in mice infected prior to 3 weeks of age. However, other intrinsic host factors contribute to the delayed response to P. murina in neonatal mice, since neutralization of TGF-β in interleukin-10 (IL-10)-deficient mice did not significantly affect clearance of the organisms, even though the inflammatory response was more intense. Adoptive transfer experiments suggested that the neonatal lung environment affects the interaction of alveolar macrophages with P. murina. However, there is also an intrinsic inability of neonatal alveolar macrophages to respond to P. murina, as confirmed by the failure to activate NF-κB both in vitro and in vivo.
MATERIALS AND METHODS
Mice.Eight-week-old BALB/c, B6D2F1/J, C57BL/6, B6.129P2-IL-10tm1Cgn/J (IL-10−/−), or C57BL/6-Tg(UBC-GFP)30Scha/J (GFP) mice were purchased from Charles River (Wilmington, MA), Taconic (Hudson, NY), or Jackson Laboratories (Bar Harbor, ME) and bred in our animal facilities. Mice were maintained at the Veterinary Medical Unit of the Veterans Administration Medical Center (VAMC) or University of Kentucky Department of Laboratory Animal Resources (DLAR) under specific-pathogen-free conditions. C.B-Igh-1b/ICrTac-Prkdcscid (designated SCID) or C1.29S6(B6)-Rag2tm1FwaN12 (designated RagKO) mice, originally from Taconic (Germantown, NY), were used to maintain a source of P. murina and were bred at the VAMC or DLAR in microisolator cages with sterilized food and water. The VAMC Institutional Animal Care and Use Committee (IACUC) and University of Kentucky IACUC approved all protocols regarding animal use.
P. murina isolation, infection, adoptive transfers, and TGF-β neutralization.Lungs were excised from P. murina-infected SCID or RagKO mice and pushed through stainless steel mesh in Hank's balanced salt solution (HBSS). Cell debris was removed by centrifugation at 100 × g for 3 min. The organisms were pelleted after lysis of red blood cells with water, resuspended in HBSS, and enumerated for inoculation for clearance studies. For purification of trophozoites for in vitro studies, red blood cells were lysed with water and organisms suspended in HBSS containing 0.5% glutathione at pH 7.3. Organisms were incubated with 200 U DNase at 37°C, and clumps were broken up by aspirating through a 26-gauge needle. Cell debris was removed by low-speed centrifugation. The supernatant was spun at 400 × g to pellet cysts. The supernatant from the 400 × g spin was centrifuged at 1,300 × g to pellet trophozoites. Trophozoites were resuspended in HBSS and filtered through 20-μm Magna nylon filters (GE Osmonics). This preparation results in more than 99% pure trophozoites. The pellet from the 400 × g spin was resuspended in HBSS and contained a mixed population of cysts and trophozoites, and this population was used as an enriched cyst population for in vitro studies. Aliquots of lung homogenates or purified organisms were spun onto glass slides, fixed with methanol, and stained with DiffQuik (Siemens Healthcare Diagnostics, Inc., Deerfield, IL). P. murina organisms were enumerated by microscopy as described previously (20, 22). Freshly isolated organisms were used for clearance experiments and routinely consisted of a ratio of 1:10 cysts to trophic forms.
Mice were anesthetized lightly with halothane or isoflurane anesthesia, and the P. murina inoculum (5 × 105 organisms/g of body weight in 10 μl for neonates and 50 μl for adults) was placed over both nares (intranasally [i.n.]). For some experiments, organisms purified as described above to the point of filtration (mixed cysts and trophozoites) were stained with CellTrace Far Red DDAO-SE (DDAO), which has an active succinimidyl ester that binds amine groups, by incubating organisms with the dye according to the manufacturer's instructions (Molecular Probes, Invitrogen, Carlsbad, CA) prior to infecting mice. The labeling of organisms was confirmed by flow cytometry and fluorescence microscopy. Control animals were given inoculations of DDAO-labeled preparations of uninfected RagKO lungs to confirm that the dye did not bind to any leftover lung debris (see Fig. S1 in the supplemental material). For the adoptive transfer of alveolar macrophages, anesthetized adult or 8-day-old mice received i.n. inoculations of 1.5 × 105 or 5 × 104 green fluorescent protein-positive (GFP+) cells, respectively. For neutralization of TGF-β, neonatal mice were given intraperitoneal injections of 10 or 50 to 200 μg/g anti-TGF-β1, TGF-β2, and TGF-β3 monoclonal antibody (TGF-β1,2,3 MAb) once or twice per week (11) (clone 1D11; R&D Systems, Minneapolis, MN; or BioXcell, West Lebanon, NH). A second set of mice was injected with an irrelevant mouse IgG1 as an isotype-matched control.
Isolation of cells from alveolar spaces, lungs, and lymph nodes.Mice were exsanguinated under deep halothane or isoflurane anesthesia, and lungs were lavaged with 5 washes of HBSS containing 3 mM EDTA. Bronchial alveolar lavage fluid (BALF) from the first wash was saved for quantification of cytokines. For isolation of alveolar macrophages, cells were pelleted from BALF and resuspended in appropriate media for flow cytometry, culture, or adoptive transfer. Right lung lobes were excised, minced, and digested in RPMI containing 3% heat-inactivated fetal calf serum, 1 mg/ml collagenase A, and 50 U/ml DNase for 1 h at 37°C. Digested lungs were pushed through 70-μm nylon mesh screens to obtain single-cell suspensions, and aliquots were taken for enumeration of P. murina. Tracheobronchial lymph nodes (TBLN) were also excised and pushed through 70-μm nylon mesh screens in HBSS. Erythrocytes were removed using a hypotonic lysing buffer. Cells were washed and counted.
Enumeration of Pneumocystis in the lungs of mice.Aliquots of lung homogenates were diluted, and 100-μl aliquots were spun onto glass slides. Slides were fixed in methanol and stained with DiffQuik (Siemens). P. murina nuclei were enumerated microscopically as previously described (20). Lung burden is expressed as log10 P. murina nuclei per right lung lobe, and the limit of detection was 3.23 log10 nuclei per lung.
Flow-cytometric analysis.Lung lavage, lung digest, and TBLN cells were washed with phosphate-buffered saline (PBS) containing 0.1% bovine serum albumin and 0.02% NaN3 and stained with appropriate concentrations of fluorochrome-conjugated antibodies specific for murine T cells (CD4, CD8, CD44, and CD62L) or macrophages (CD11c, CD11b, F4/80, and Ia). Antibodies were purchased from BD Biosciences-Pharmingen (San Diego, CA) or eBioscience (San Diego, CA). Expression of these molecules on the surface of the cells was determined by multiparameter flow cytometry using a FACSCalibur or LSRII cytofluorimeter (Becton, Dickinson, Mountain View, CA). Ten thousand to 50,000 events were routinely acquired.
Analysis of cytokine levels and P. murina-specific IgG in BALF.Cells and debris were removed from the first wash of BALF by centrifugation, and the supernatant was frozen at −80°C for later use. The quantitation of multiple cytokines in the same sample of BALF was performed using bead array kits purchased from Upstate Cell Signaling Solutions (Lake Placid, NY) or BD Biosciences-Pharmingen. Assays were performed according to the manufacturer's directions and analyzed using a Luminex 100 system (Luminex Corp., Austin, TX) or flow cytometry on a FACSCalibur. P. murina-specific IgG was measured by enzyme-linked immunosorbent assay (ELISA). A sonicate of P. murina organisms (10 μg protein/ml) was coated onto 96-well plates, and wells were blocked with 5% dry milk in HBSS containing 0.05% Tween 20. Sera were diluted and incubated on plates overnight. Plates were extensively washed, and bound IgG was detected using alkaline phosphatase-conjugated anti-mouse IgG (Sigma). Plates were washed and secondary antibodies detected using p-nitrophenylphosphate at 1 mg/ml in diethanolamine buffer. Optical density was read at 405 nm using a plate reader equipped with KC Junior software (Bio-Tek Instruments, Inc., Winnoski, VT).
Analysis of TGF-β1 levels in lung homogenates with ELISA.Mouse left lung lobes were snap-frozen in liquid nitrogen at the time of euthanasia and stored at −80°C for later analysis. Five hundred μl of diluted protease inhibitor cocktail (1:100 in PBS; Sigma-Aldrich, St. Louis, MO) was added, and the lungs were homogenized using a Dounce homogenizer. The homogenates then were centrifuged at 400 × g for 10 min at 4°C. Supernatants were collected and used in a TGF-β ELISA kit (R&D Systems, Minneapolis, MN, or eBioscience). The ELISA was performed according to the manufacturer's instructions. Data were normalized to lung protein concentrations, which were determined using a Bio-Rad DC protein assay kit (Hercules, CA) according to the manufacturer's instructions.
Analysis of NF-κB translocation ex vivo and in vitro.Mice were infected with P. murina at 2 days of age or as adults as previously described. At multiple time points, lungs were removed and snap-frozen for subsequent analysis. Whole lung tissue was homogenized and nuclei extracted as previously described (53, 57, 58), and protein concentrations of nuclear extracts were determined using the DC protein assay kit from Bio-Rad Laboratories. Nuclear extracts (5 μg of protein) were incubated with electrophoretic mobility shift assay (EMSA) binding buffer and 10 fmol of 32P-labeled NF-κB probe (double-stranded oligonucleotides containing consensus κB sequence 5′-TCAGAGGGGACTTTCCGAGAGG-3′, underlining denotes the NF-κB finding site) for 20 min at room temperature. Reaction mixtures were separated in a 6% nondenaturing polyacrylamide gel by electrophoresis. The gel was transferred to blotter paper, dried, and exposed to X-ray film using intensifying screens at −70°C. Relative nuclear binding activities for NF-κB were quantified by scanning densitometry.
For examining NF-κB activation in alveolar macrophages, mice were infected with P. murina at 2 days of age or as adults as previously described. At multiple time points, lungs were lavaged and alveolar macrophages pooled from up to 15 mice. Nuclear extracts were isolated using a kit from Panomics (Fremont, CA) per the manufacturer's instructions. Nuclear extract protein levels were quantified using the microassay procedure of the DC protein assay kit from Bio-Rad Laboratories. Equal protein concentrations were used in a Chemicon NF-κBp65 nuclear translocation colorimetric plate-based assay (Millipore, Bellerica, MA) according to the manufacturer's instructions. Data are expressed as the optical density read at 450 nm in up to triplicate wells per experimental group.
For in vitro assays, alveolar macrophages pooled from adult and 10- to 14-day-old mice were placed into RPMI medium containing 5% fetal bovine serum, antibiotics, and 2-mercaptoethanol.
Cells were rested overnight at 2 × 106 cells/ml in 6-well plates or at 2 × 105 on chamber slides and then stimulated with 100 ng/ml lipopolysaccharide (LPS) (from Escherichia coli; Sigma), a 1:50 dilution of sonicated P. murina organisms (at 108 nuclei/ml), 25 mg/ml zymosan, trophozoites at 10 per 1 macrophage, or a mix of cysts and trophozoites (1:10) at 10 cysts per 1 macrophage for up to 4 h. Medium was removed and frozen. Adherent cells in 6-well plates were lysed and nuclei extracted using a Chemicon nuclear extraction kit per the manufacturer's instructions (Millipore). NF-κB translocation was determined using the plate-based Chemicon kit as described above. Adherent cells cultured on chamber slides were fixed in 10% formalin for 15 min, washed in PBS, and permeabilized in methanol for 10 min. After blocking with 5% normal goat serum in PBS plus 0.3% Triton X-100, cells were incubated overnight with anti-NF-κB p65 (Cell Signaling Technology, Beverly, MA) followed by anti-rabbit IgG-Alexa 594 for 2 h. Nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI) prior to examination using a Zeiss imager Z1 fluorescence microscope equipped with an AxioCam HRc and AxioVision Rel.4.8 software or a Leica TSP SPS inverted confocal microscope. The proportion of cells with translocated NF-κB was determined microscopically by counting a minimum of 100 cells and determining the number with colocalization of NF-κB and nuclear staining.
Statistical analysis.Data were analyzed utilizing the SigmaStat statistical software package (SPSS Inc., Chicago, IL). Analysis of variance (ANOVA) was used to determine differences between and within groups. Student Newman Keul's post hoc tests were applied to discriminate differences between groups at individual time points. Data were determined to be significantly different when the P value was less than 0.05 using a two-tailed test. For data that did not meet the assumptions of equal variance or normal distribution, nonparametric tests were used.
RESULTS
Resolution of Pneumocystis infection is age dependent.We have previously shown that clearance of P. murina is delayed in mice infected as neonates compared to clearance in adult mice (20). To determine whether P. murina organism clearance and inflammation is developmentally controlled, mice were infected intranasally at 2, 7, or 14 days of age, and their ability to resolve P. murina infection was compared to that of adult mice. As we have previously reported, organism burden had peaked by day 7 postinfection in adult mice, and they cleared P. murina by day 21 postinfection (Fig. 1A). In contrast, lung burdens of mice infected as 2-, 7-, or 14-day-old pups peaked 1 to 2 weeks later than in adult mice. Furthermore, none of the mice infected as pups were able to control their lung burden until after they reached 3 weeks of age. Mice infected at 2 weeks of age were able to control P. murina burden better than mice infected at 2 and 7 days of age, although they were still unable to clear the organisms until 4 weeks postinfection, or about 6 weeks of age (Fig. 1A). Notably, mice were infected with equivalent numbers of organisms based on body weight, which means that adults received approximately 10-fold more organisms than 2-day-old mice. Variability in the size of the mice between groups due to the age at infection and the inoculum dose, which was normalized to weight, likely contribute to the differences in lung burden between the groups at day 7 postinfection. Alternatively, there may be a developmental difference in the lungs during the first 3 postnatal weeks that is more permissive for P. murina growth at 7 days than at 14 days of age. Regardless of the initial inoculum, organism burden increased in the lungs of all pup groups through 2 weeks postinfection, indicating permissiveness of Pneumocystis growth in the lungs of young mice. These data confirm that between 14 and 21 days of age is a critical time in pup immune development in the context of P. murina clearance.
Immune response to and clearance of P. murina is delayed in young mice infected at up to 2 weeks of age compared to adults. Mice were infected at 2, 7, or 14 days (young) or at 8 weeks (adult) of age with 5 × 105 organisms/g of body weight. At the indicated time points, lung organism (PC) burden (A), P. murina-specific serum IgG (B), percent activated alveolar macrophages in the BALF (C), percent CD4+ T cells in the BALF (D), and TNF concentrations in the BALF (E) were determined as described in Materials and Methods. Activation status of CD11c+ F4/80+ macrophages was determined by expression of CD11b and MHC-II (Ia). Data represent the means ± standard deviations (SD) from 5 mice per group and are representative of two separate experiments. For some data points, error bars are smaller than the symbols. *, P < 0.05 compared to adults at the same time point. O.D. 405 nm, optical density at 405 nm.
The immune response to Pneumocystis is age dependent.Alveolar macrophages have been shown to be the effector cells that kill P. murina, and specific antibody can have a significant role in opsonizing and targeting organisms for clearance (13, 28, 43, 47). P. murina-specific IgG is elevated in the blood of adult mice at day 14 postinfection; however, mice infected at 2, 7, or 14 days of age did not have measurable specific IgG in the blood until at least 3 weeks postinfection (Fig. 1B). Mice infected at 2 days of age did not have measurable P. murina-specific IgG until after day 21 postinfection, the time at which we terminated the experiment for these mice. We have previously published that specific IgG is not detected until at least 4 weeks postinfection when mice are infected at 2 days of age (20). In addition to the delay in antibody responses, activation of alveolar macrophages was delayed in the mice infected with P. murina at 2, 7, and 14 days of age compared to adults (Fig. 1C). Activated macrophages as defined by the upregulation of major histocompatibility complex class II (MHC-II) (Ia) and CD11b had already peaked at day 7 postinfection in adult mice, whereas the peak for the mice infected as pups lagged behind by as much as 3 weeks. This corresponded to significantly lower levels of tumor necrosis factor (TNF) in the BALF of mice infected as pups compared to mice infected as adults (Fig. 1E). TNF has been shown to induce the expression of adhesion molecules and to be required early in the host response against P. murina for resolution to occur (9, 36). Macrophages have been shown to produce TNF in response to interactions with Pneumocystis β-glucans (24), so the delay in production of TNF in the lungs of mice infected at 2 weeks of age and younger is consistent with delayed activation of alveolar macrophages.
It is well known that CD4+ T cells are required for mounting an effective host response to P. murina infection (5, 22, 43), and we have reported that there is a delay in the infiltration of lymphocytes into the alveolar spaces of the lungs in mice infected as neonates compared to adults (20). Therefore, we determined how increasing age affects infiltration of T cells into the alveolar spaces of the lung during P. murina infection. As shown in Fig. 1D, migration of activated CD4+ T cells into the alveolar spaces was delayed in pups infected at 2, 7, or 14 days of age compared to adults. The number of CD4+ T cells peaked in adult lungs by day 7 postinfection. In contrast, alveolar CD4+ T cells peaked at day 14 postinfection in pups infected at 2 weeks of age but not until after day 14 in pups infected at 2 days and 1 week of age. Notably, infiltration of CD4+ T cells into the alveoli was not detected in any group of mice until they were 3 weeks of age or older (Fig. 1D).
IL-10 is partially responsible for delaying clearance of Pneumocystis from neonatal lungs.We have previously shown that there is constitutive mRNA expression of IL-10 and TGF-β isoforms in the lungs of neonatal mice, and that these levels were comparable to or higher than those in adults (37). Furthermore, we reported that adult IL-10−/− mice clear P. murina faster and develop a more intense inflammatory response than do wild-type mice (38). Moreover, IL-10 produced by epithelial cells has been associated with controlling inflammation in adult lungs (16). To determine whether the presence of IL-10 is responsible for the delayed inflammatory response to P. murina in neonatal mice, IL-10−/− pups were infected with organisms at 2 days of age. Lung P. murina burden was cleared with slightly faster kinetics in IL-10−/− pups than in wild-type C57BL/6 pups (Fig. 2A). There were significant differences in lung organism burdens between IL-10−/− and wild-type pups by days 24 and 31 postinfection (Fig. 2A). The faster clearance kinetics in IL-10−/− mice corresponded to elevated numbers of activated CD4+ T cells in lung digests of IL-10−/− mice around 3 weeks postinfection (Fig. 2C). BALF also had elevated, though not statistically significant, numbers of activated CD4+ T cells in IL-10−/− pups (Fig. 2B). The elevation of activated T cells in IL-10−/− mice was due to increased proportions of cells in the lung digest or BALF as opposed to differences in absolute numbers of cells (data not shown). There were reduced numbers of activated CD4+ T cells in the TBLN of IL-10−/− mice, likely due to increased migration to the lungs (Fig. 2D). The elevated T cells at later time points in the IL-10−/− pups did not affect help to B cells, since there were no differences in P. murina-specific IgG in the sera (Fig. 2E). Examination of lung proinflammatory cytokines by RNase protection assay or ELISA on BALF indicated that at later time points (at 3 weeks of age or greater), infected IL-10−/− mice had elevated levels of TNF-α, IL-1β, IL-6, and gamma interferon (IFN-γ) compared to wild-type mice in some experiments, but this was not consistent among all experiments performed (5 separate experiments) (data not shown). Together, these data suggest that IL-10 has a role in controlling inflammation in neonatal lungs. However, because differences were not seen in the IL-10−/− pups until after the mice were 3 weeks of age, there are clearly other mechanisms responsible for the lack of responsiveness to P. murina in neonates compared to adults.
IL-10−/− pups clear P. murina faster than do wild-type mice. Two-day-old IL-10−/− and C57BL/6 (wild-type) pups were infected with P. murina or were inoculated with buffer. At the indicated time points, lung organism (PC) burden (A) and activated CD4+ T cells in the BALF (B), lung parenchyma (C), and draining lymph nodes (D) were determined as described in Materials and Methods. For flow cytometry, cells were gated on CD4+ lymphocytes and analyzed for CD44 and CD62L expression. (E) P. murina-specific IgG levels in sera were examined by ELISA. Data represent means ± SD from 4 to 8 mice per group and are representative of five separate experiments. *, P < 0.05 compared to infected wild-type mice at the same time point.
Increased levels of TGF-β may play a role in the delayed migration of T cells in pups.Since we found that there are factors other than IL-10 that must contribute to the unresponsiveness of neonatal mice to P. murina, we examined whether TGF-β is a contributing factor. TGF-β1 has long been associated with immunosuppression, and mice deficient in this cytokine die of a lymphoproliferative disease by about 4 weeks of age (45). Moreover, TGF-β is one of several growth factors elevated in the lungs of postnatally developing mice (2, 32). A TGF-β1-specific ELISA was performed on supernatants from lung homogenates of uninfected and P. murina-infected pups and adults. Lung homogenates were used because we were unable to detect TGF-β in BALF. Uninfected pups had significantly higher levels of total TGF-β1 per mg of lung protein at 9 and 16 days of age than uninfected adults (Fig. 3A). TGF-β1 concentrations dropped somewhat in the lungs of pups infected at 2 days of age with P. murina (Fig. 3B). At day 7 postinfection, TGF-β1 levels in infected pups had dropped about 30% compared to those of uninfected pups. However, the concentration of TGF-β1 in the lungs of infected pups at day 7 was still about 2-fold greater than in P. murina-infected adult lungs (Fig. 3B). The higher TGF-β1 levels in pups decreased to adult levels by day 21 in both uninfected and infected pups, suggesting that TGF-β is developmentally controlled but can be affected by an infectious agent.
TGF-β1 is constitutively expressed at high levels in neonatal lungs. Lungs were obtained from uninfected pups or from mice infected with P. murina at 2 days of age or as adults and were snap-frozen. Lungs were homogenized in protease inhibitor cocktail, and TGF-β1 concentrations in the supernatants were determined by ELISA and normalized for total protein levels. Two separate experiments show TGF-β1 concentrations in uninfected mice (A) and TGF-β1 concentrations in mice infected as neonates or adults (B). Data represent means ± SD from 3 to 5 mice per group and are representative of two separate experiments. *, P < 0.05 compared to pups at the same time point.
Neutralization of TGF-β in IL-10−/− mice results in increased inflammation and modest control of Pneumocystis growth.To determine whether TGF-β is responsible for the delayed response to P. murina in neonatal compared to adult mice, we neutralized TGF-β in vivo by injection of 10 μg/g of specific monoclonal antibody at days −1, 4, 11, and 17 postinfection. Neutralization of TGF-β had no effect on clearance of P. murina, although we did see trends toward elevated proinflammatory cytokines in the lungs (data not shown). We next neutralized TGF-β1,2 in IL-10−/− mice to determine whether reduction of two anti-inflammatory mediators would induce the neonatal immune response to control P. murina growth comparable to adult mice. Since we saw only modest reduction in lung TGF-β1 levels with our previous antibody treatments, for this experiment we injected anti-TGF-β1,2,3 MAb twice per week with doses of 50 to 200 μg/g beginning on the day of infection with P. murina. The level of growth of P. murina was not reduced in the IL-10−/− pups treated with anti-TGF-β (Fig. 4A). However, increased infiltration of CD4+ T cells into the alveoli of anti-TGF-β-treated IL-10−/− mice was observed (Fig. 4B and C). Moreover, though not statistically significant, neutralization of TGF-β consistently resulted in a modest increase in TNF-α in IL-10−/− pups at about 2 weeks postinfection (Fig. 4D). Together, these data suggest that IL-10 and TGF-β contribute to the control of inflammation during the clearance phase of P. murina infection.
Neutralization of TGF-β in IL-10−/− pups accentuates the inflammatory response to P. murina but does not affect clearance. Neonatal IL-10−/− mice were treated with intraperitoneal injections of 50 μg/g anti-TGF-β1,2 MAb or vehicle at day 0 and twice per week after infection with P. murina. At the indicated time points, lung P. murina (PC) burden (A), percent CD4+ cells in BALF (B), total CD4+ cells in BALF (C), and BALF TNF-α concentration (D) were determined. Data represent means ± SD from 3 to 4 mice per group. *, P < 0.05 compared to vehicle control at the same time point. +, P < 0.05 compared to anti-TGF-β-treated mice irrespective of time point. Data are representative of two separate experiments.
The neonatal lung environment contributes to reduced association of alveolar macrophages with P. murina.Figure 1 indicates that P. murina grows virtually unchecked in the lungs of infant mice until they are about 3 weeks of age. Neutralization of anti-inflammatory mediators did not significantly affect the growth of the organisms during these first 2 to 3 weeks (Fig. 2A and 4A). We used an adoptive transfer approach to determine whether the neonatal lung environment had a significant effect on alveolar macrophage association with P. murina. Alveolar macrophages were isolated from adult or 7-day-old mice expressing green fluorescent protein (GFP) under the control of the actin promoter and inoculated into the lungs of 8-day-old or adult mice, respectively. Adult and pup GFP+ alveolar macrophages expressed CD11c but were CD11b negative at the time of transfer (data not shown) and were considered unactivated. Mice were infected 24 h later with P. murina labeled with DDAO, and 18 h after infection lungs were lavaged and analyzed by flow cytometry. As shown in Fig. 5A, donor neonatal and host adult alveolar macrophages could be distinguished by expression of GFP. The proportion of CD11c+ CD11b− cells that had bound P. murina and were positive for DDAO was not different between the adult GFP− host adult cells and the GFP+ transferred pup cells (Fig. 5B). Moreover, the proportion of CD11c+ cells that upregulated CD11b and bound P. murina was similar in donor pups and host adults (Fig. 5B). Together, these data confirm that alveolar macrophages from neonatal mice associate with P. murina and become activated, as determined by CD11b expression in an environment that is less immunosuppressive than the neonatal lung.
Alveolar macrophages from pups bind P. murina when transferred into adult lungs. Alveolar macrophages were isolated from the BALF of pup transgenic mice expressing GFP under the control of the β-actin promoter and 1.5 × 105 cells injected i.n. into adult wild-type mice. Host mice were infected with DDAO-labeled P. murina organisms, and the next morning alveolar macrophages were examined for attachment or uptake of DDAO-expressing organisms using flow cytometry. (A) Dot plot and histograms showing the gating scheme for examining host alveolar macrophage (CD11c+ GFP−) or donor alveolar macrophage (CD11c+ GFP+) cells. Graphs depicting the proportion of CD11c+ gated cells that expressed CD11b (B and C, top) and bound P. murina (PC) (B and C, bottom) are shown for transfers of pup cells to adults (B) and adult cells to pups (C). Data represent the means ± SD from 5 to 6 mice and are representative of 2 separate experiments. No statistically significant different results were found between pup and adult AMs.
To confirm whether the neonatal lung environment is suppressive for alveolar macrophage interactions with P. murina, we adoptively transferred GFP+ alveolar macrophages from adult mice into the lungs of 7-day-old mice, followed by infection with DDAO-labeled P. murina the next day. As shown in Fig. 5C, less than 3% of either host pup or transferred adult alveolar macrophages were positive for DDAO 24 h after infection, indicative of a lack of association with P. murina. This corresponded to less than 20% of the alveolar macrophages from either host or donor expressing CD11b upon infection. Together, these data demonstrate that the neonatal lung environment suppresses activation of alveolar macrophages and association with P. murina.
Neonatal macrophages fail to activate NF-κB in response to Pneumocystis.To this point, our data indicate that neither TGF-β1 nor IL-10 can completely account for the failure of neonates to respond to P. murina prior to weaning age, although there is clearly suppression taking place, as evidenced by the failure of alveolar macrophages to associate with the organisms in the neonatal lung. Recognition of P. murina in the lungs is necessary for initiating an immune response; thus, resident alveolar macrophages were examined more closely as the likely first line of defense against the pathogen. Alveolar macrophages produce TNF-α in response to Pneumocystis via activation of NF-κB (56). We have previously shown that neonatal alveolar macrophages fail to upregulate costimulatory molecules and produce TNF-α in response to P. murina infection (12, 13), and our data in Fig. 5 suggest that this is due to the neonatal lung environment. To begin to tease apart the mechanism responsible for this unresponsiveness in alveolar macrophages, we examined translocation of the transcription factor NF-κB to the nucleus of whole lung cells and alveolar macrophages after infection with P. murina. Utilizing an EMSA on whole lung cells, Fig. 6A shows that by day 1 postinfection adult lung cells already had activated NF-κB expression in the nuclei, and at day 15 mice infected as neonates still did not have NF-κB activation above background levels. Since multiple cell types could be stimulated in the lungs, we also examined NF-κB activation in alveolar macrophages isolated up to 48 h postinfection. Cells were lavaged from the lungs of mice infected at 2 days of age and as adults and pooled by group. The purity of alveolar macrophages was determined by differential counts and was 92% in pups at both time points and 86 and 91% in adults at 24 and 48 h postinfection, respectively. The rest of the cells were neutrophils with no lymphocytes detected in any of the pools of cells. Utilizing a plate-based assay for nuclear extraction and NF-κB detection, we found that by 48 h, macrophages from adult lungs had significantly elevated levels of translocated NF-κB, whereas macrophages from neonatal mice did not (Fig. 6B). To determine whether this was an intrinsic defect or due to the lung environment, we next isolated alveolar macrophages from adult and 10- to 14-day-old mice and stimulated them in vitro with either LPS or sonicated P. murina antigen. The purity of the alveolar macrophages was more than 95%. As shown in Fig. 6C, LPS and sonicated P. murina induced significant activation of NF-κB in cells from adult mice within 1 h; however, NF-κB activation in cells from pups was at or below background levels. This corresponded to failure of pup alveolar macrophages to produce TNF-α above background levels within 24 h after in vitro stimulation (data not shown). We next utilized fluorescence microscopy to determine the age at which alveolar macrophages respond similarly to adults. Alveolar macrophages were isolated from mice of ages ranging from 14 days to adult, stimulated with P. murina in vitro, and stained for p65 of NF-κB using fluorescently labeled antibody (see Fig. S2 in supplemental material). The proportion of alveolar macrophages with p65 colocalized in the nucleus (stained with DAPI) was quantitated microscopically. Figure 6D shows that the proportion of macrophages with activated NF-κB increased over time, and by about 5 weeks of age the proportion of activated macrophages was approaching that of adults. Together, these data demonstrate not only that the neonatal lung environment contributes to unresponsiveness to P. murina but also that neonatal alveolar macrophages are intrinsically unresponsive to stimulation with P. murina as well as LPS.
Activation of NF-κB is delayed in the lungs and alveolar macrophages of neonates in response to P. murina. (A and B) Neonatal and adult mice were infected with 5 × 105 P. murina organisms/g. At the indicated time points, lungs were snap-frozen and then homogenized prior to nuclear extraction (A), or they were lavaged and nuclei extracted from alveolar macrophages (B). (A) Nuclear translocation of NF-κB was determined for lungs from individual mice by EMSA, and data are expressed as means ± SD of relative densitometry units (arbitrary units) from 3 mice per group. +, P < 0.05 compared to pup lungs irrespective of time point. (B) Nuclear translocation of pooled samples was determined by a colorimetric plate-based commercial assay, and data are expressed as means ± SD of replicate wells (optical density read at 450 nm). *, P < 0.05 compared to pups at the same time point. Alveolar macrophages were isolated from uninfected 10-day-old (C) and adult mice or mice of various ages (D) and rested in culture medium overnight. (C) Alveolar macrophages were stimulated with LPS or sonicated P. murina, and 1 h later cells were lysed and nuclei extracted. NF-κB translocation was determined from in vitro-stimulated cells using the plate-based assay. Data are expressed as a percentage of the uninfected control, because the background NF-κB activation was consistently higher in neonatal macrophages than in adults. The means ± SD from replicate wells normalized to controls are shown. No statistical differences were found. (D) Alveolar macrophages from mice at the indicated ages were stimulated with P. murina and stained with DAPI and fluorescent antibody specific for p65. The proportion of alveolar macrophages with colocalized p65 and DAPI was determined by counting microscopically. Data represent the proportion of alveolar macrophages (AMs) with p65 in the nuclei per minimum of 100 cells counted. Fluorescent micrographs of alveolar macrophages can be found in Fig. S2 in the supplemental material.
Recognition of unopsonized P. murina by alveolar macrophages occurs through pattern recognition receptors, including Dectin-1, which recognize β-glucan in the cyst cell wall (47), or mannose receptors, which recognize mannosylated glycoproteins expressed on the cell surface of both cyst and trophozoite forms of the organisms (15). Dectin-1 transduces a signal through an ITAM-like motif that results in activation of NF-κB, while mannose receptors are largely phagocytic receptors. We have previously demonstrated that Dectin-1 expression is not different between neonatal and adult alveolar macrophages (12). P. murina isolated from RagKO mice, which were used as the source of organisms for in vitro experiments, consists of only 10% of the cyst life form, so we used zymosan, a yeast cell wall extract rich in β-glucan, to provide a strong β-glucan stimulus of alveolar macrophages from adult mice and 2-week-old pups. Fluorescence microscopy was used to visualize NF-κB translocation to the nucleus. Figure 7C, D, and I show that when stimulated with zymosan, alveolar macrophages from both pups and adults were able to translocate NF-κB to the nucleus, whereas unstimulated cells were not (Fig. 7A and B). A pure population of P. murina trophozoites did not stimulate activation of NF-κB (Fig. 7G, H, and I); however, a population enriched for cysts (10:1 cysts to macrophages) consistently stimulated NF-κB translocation in adult alveolar macrophages but not in pup macrophages (Fig. 7E, F, and I). Figure 7I shows the proportion of double-positive DAPI-staining and NF-κB-expressing macrophages for each group. To determine whether increasing the organism-to-macrophage ratio would result in activation of NF-κB in alveolar macrophages from pups, we stimulated alveolar macrophages with a cyst-to-macrophage ratio of 35:1 and found that the pup alveolar macrophages were able to translocate NF-κB similarly to adult macrophages (see Fig. S3 in the supplemental material). Increasing the ratio of trophic forms by up to 20-fold had no effect on NF-κB translocation in either pup or adult alveolar macrophages (Fig. S3C and F in the supplemental material). Together, these data demonstrate that alveolar macrophages from pups of up to 2 weeks of age do not respond to physiological levels of the cyst form of P. murina through β-glucan receptors.
P. murina cysts stimulate NF-κB translocation in adult alveolar macrophages but not in macrophages from pups. Alveolar macrophages were isolated from adult and 14-day-old mice and rested in culture overnight. Cells were unstimulated (A and B) or stimulated with zymosan (C and D), an enriched population of P. murina cysts containing trophic forms (PC mix) (E and F), or trophic forms of P. murina (PC trophs) (G and H) for 4 h, followed by fixation and staining with antibody to NF-κB p65. Nuclei were stained with DAPI. Data are shown at ×100 magnification and are representative of 5 separate experiments. Red, p65; blue, DAPI; pink, colocalization. (I) The proportion of cells with colocalization of DAPI and NF-κB was determined by microscopic counting.
DISCUSSION
We have previously reported that there is a significant delay in the ability of neonatal mice to mount an inflammatory response to P. murina compared to adult mice. The data presented herein demonstrate that the immune response to P. murina is developmentally regulated. We had hypothesized that anti-inflammatory cytokines, including TGF-β1 and IL-10, play a role in this regulation. However, the presence of these cytokines does not explain the unresponsiveness to P. murina during the first 3 weeks of life. We found that in addition to a seemingly immunosuppressive lung environment, unresponsiveness in neonatal lungs is also due to failure of alveolar macrophages to activate NF-κB in response to P. murina cysts. These data demonstrate that the neonatal lung environment is a complex mix of anti-inflammatory mediators and immature immune cells that sets the threshold of activation in response to infection higher than that in adult lungs.
It has long been known that CD4+ T cells are required for host defense against P. murina and that this is most likely due to the requirement for proinflammatory cytokines, such as TNF-α and IFN-γ, to stimulate alveolar macrophages to phagocytose and kill the organisms (3, 4, 9, 22, 28, 42). We previously demonstrated that infiltration of CD4+ T cells into the airways precedes clearance of P. murina in neonates (20). Here, we found that CD4+ T cells failed to infiltrate the alveolar spaces of mice infected at 2, 7, or 14 days of age prior to their reaching 3 weeks of age. This could be due to low numbers of specific T cells in neonates. However, we have previously shown that T cells isolated from the draining lymph nodes of 10- to 14-day-old P. murina-infected pups can effect clearance of P. murina when transferred to adult SCID mice (21), suggesting that sufficient numbers of antigen-specific T cells are present within 10 days but that they are unable to enter neonatal lungs. In this regard, we have also found that lung expression levels of ICAM-1 and VCAM-1 are significantly reduced in neonatal compared to adult lungs infected with P. murina (36). Moreover, neonatal T cells do not migrate efficiently across endothelial cells in an in vitro transwell assay (21, 36). Together, our data suggest that there are some aspects of neonatal T cell function that are different from adult T cells, but the neonatal lung environment contributes to the overall lack of responsiveness of neonatal T cells. Our current data suggest that the lung environmental factors are developmentally regulated, since no matter when we infected the pups they still did not respond until 3 weeks of age.
TGF-β and IL-10 are known to play prominent roles in downregulating immune responses, and both can be secreted by a number of cell types (27, 34). In addition, TGF-β is involved in the intricate orchestration of lung branching morphogenesis during development (25, 31). Both TGF-β and IL-10 are known to be constitutively expressed in adult lungs, presumably to regulate cell turnover, host defense, and matrix protein metabolism (10, 16). We previously found that adult IL-10−/− mice clear P. murina slightly faster than wild-type mice, and this correlated with a more intense inflammatory response (38). This is consistent with other studies that demonstrate that IL-10 negatively regulates macrophage function (6, 16, 17, 38, 40, 46, 55). As with our studies of adult IL-10−/− mice, we observed improvement in clearance of P. murina in IL-10−/− pups, arguing that IL-10 contributes to the immunosuppressive environment of neonatal lungs but is not the only factor involved. Although we can conclude that IL-10 contributes to the overall suppressive environment of neonatal lungs, deficiency of IL-10 only had an effect on mice after they reached 3 weeks of age, suggesting alternate mechanisms of immune suppression in the lungs of very young mice.
Our studies did not address which cell types in neonatal mouse lungs produce TGF-β1 and/or IL-10, although we are quite interested in this. Others have shown that lung epithelial cells can produce these cytokines as well as monocytes, alveolar macrophages, and other cell types (10, 16, 49). We favor the hypothesis that lung structural cells or macrophages are the source of these cytokines rather than regulatory T cells, because we see very few CD4+ T cells in the airways of neonatal mice prior to 3 weeks of age. In P. murina-infected pups, CD4+ T cells accumulate in the draining lymph nodes and lung parenchyma in very low numbers around days 10 to 14 postinfection (data not shown). We have examined these cells for expression of CD25 as a cursory look for regulatory T cells but have not found them in great enough numbers to be confident of the results. Experiments under way in our laboratory will determine the source of both TGF-β1 and IL-10 in the lungs of infant mice. We were surprised that neutralization of TGF-β and IL-10 in concert did not result in appreciable changes in alveolar inflammation or control of P. murina prior to mice reaching 3 weeks of age. In spite of this, we found that adult alveolar macrophages adoptively transferred into neonatal lungs did not associate with fluorescently labeled P. murina, although neonatal alveolar macrophages transferred to adult lungs were able to bind organisms. This is consistent with an immunosuppressive lung environment in neonatal mice that promotes growth of P. murina and inhibits nonopsonic uptake of organisms. Together, these data could be interpreted to mean that anti-inflammatory cytokines are important in inhibiting binding and uptake of P. murina early after infection in neonates but have a more dramatic effect on induction of adaptive immunity.
Alveolar macrophages have been demonstrated to be the effector cells most responsible for killing of Pneumocystis (28, 47), and expression of the β-glucan receptor Dectin-1 has been shown to be an important pathway for macrophage recognition and killing of the organisms (44, 47). Here, we show that the cyst form of the organisms stimulates NF-κB translocation in adult alveolar macrophages but not in macrophages from neonates. We have previously demonstrated that the expression levels of Dectin-1 are comparable on neonatal and adult alveolar macrophages, which may mean that signaling through the Dectin-1 receptor is impaired in neonatal macrophages (12). Consistent with this, the data presented herein indicate that alveolar macrophages from neonates do not activate NF-κB in the same time frame as adult macrophages. This is an age-dependent response, as alveolar macrophages increasingly responded to P. murina in vitro from 2 weeks of age through 5 weeks of age. In contrast, zymosan, a yeast cell wall extract, provided a potent signal for NF-κB translocation in neonatal alveolar macrophages, and PC was able to stimulate activation in adults, albeit to a lesser extent than zymosan. This difference between zymosan and PC in the ability to stimulate neonatal alveolar macrophages could have to do with strength of signal, since increasing the P. murina-to-macrophage ratio resulted in translocation of NF-κB p65. Alternatively, the signaling pathways differ in response to zymosan and PC. We have some gene expression data suggesting that some signaling intermediates in the NF-κB pathways are differentially expressed depending on the stimulus. We are quite interested in these genes and are in the process of performing experiments to understand these mechanisms.
Pneumocystis β-glucans have been shown to signal NF-κB activation in alveolar epithelial cells (14, 52). However, a recent study with Dectin-1−/− mice indicated that Dectin-1 was important for recognizing and killing of P. murina by macrophages but not for production of cytokines, such as TNF-α (44). Signaling through MyD88 stimulated cytokine production in Dectin-1−/− macrophages (44). There is some evidence that Dectin-1 can interact with TLR-2 (19), and it was recently demonstrated that Pneumocystis induced NF-κB activation and cytokine production through TLR-2 ligation on alveolar macrophages (56). We have also shown that TLR-2 expression levels are comparable on adult and neonatal alveolar macrophages (12). We hypothesize that there is a dysregulation of the signaling pathways of Dectin-1 and TLR-2 in neonatal alveolar macrophages that is responsible for the inherent unresponsiveness to P. murina. We can envision several mechanisms for this, including overexpression of control proteins such as A20 or epigenetic regulation, as has been shown for neonatal T cells (41). Preliminary data from our laboratory indicate that several signaling intermediates in the NF-κB pathways are significantly downregulated in neonatal alveolar macrophages compared to adults. We are in the process of determining the mechanism of this downregulation.
In summary, we have shown that the response to P. murina is developmentally controlled. This likely is not due solely to the presence of anti-inflammatory cytokines present in the postnatally developing lungs but also to factors intrinsic to immune cells, such as alveolar macrophages. Since approximately 80% of alveolar space is developed postnatally, one can envision that it is critical to strike a balance between unresponsiveness to dangerous pathogens and an overexuberant immune response that could damage sensitive lungs.
ACKNOWLEDGMENTS
This work was funded by Public Health Service grants HL062053 and HL088989 from the National Heart, Lung, and Blood Institute to B.A.G., fellowship grant RT-051-N from the American Lung Association to M.Q., VA Merit Review to S.A.B., and resources provided by the Veteran's Affairs Medical Center, Lexington, KY.
FOOTNOTES
- Received 28 July 2011.
- Returned for modification 20 August 2011.
- Accepted 23 May 2012.
- Accepted manuscript posted online 4 June 2012.
Supplemental material for this article may be found at http://dx.doi.org/10.1128/IAI.05707-11.
- Copyright © 2012, American Society for Microbiology. All Rights Reserved.
