Chronic hypersensitivity pneumonitis caused by Saccharopolyspora rectivirgula is not associated with a switch to a Th2 response

Animals and S. rectivirgula exposure protocol.

C57BL/6 female mice were purchased from The Jackson Laboratory (Bar Harbor, ME) at 6 wk of age. tlr2/9^−/− mice on a C57BL/6 background were bred at the University of Tennessee Health Science Center (UTHSC) and were in at least the 15th generation backcross. All animals were housed in sterile microisolator cages with sterile food and water ad libitum, and were maintained by the Division of Comparative Medicine at UTHSC according to the guidelines of the Animal Welfare Act. All animal care procedures were performed according to protocols approved by the UTHSC animal care and use committee.

S. rectivirgula (strain designation A1313-ATCC) was grown at 55°C in trypticase soy broth. The bacterial preparation was washed in endotoxin-free distilled water three times and lyophilized. The lyophilized preparation was reconstituted with endotoxin-free saline, and mice were exposed intranasally with 150 μg SR for 4 wk (three times per week for 3 wk and once in week 4) or similarly for 14–15 wk. Mice were analyzed 2–3 days after the last exposure.

BAL and lung cell isolation.

BAL was performed by intratracheal injection of 1 ml of PBS with 2 mM EDTA into the lungs with immediate vacuum aspiration. The amount of fluid recovered was routinely around 70%. Cells were recovered from the BAL fluid (BALF) by centrifugation and counted using Trypan blue dye exclusion. The BALF was frozen at −80°C until used for protein determinations or ELISA assays for cytokine and chemokine measurement.

Lungs were perfused with PBS to remove blood, and lobes were removed. Lung tissue was digested with collagenase (20 U/ml) and deoxyribonuclease I (40 μg/ml) for 45 min at 37°C. Cells were freed by disruption in a Stomacher tissue processor and then isolated by centrifugation on a discontinuous Percoll gradient. Mononuclear cells were isolated at the 40/80% interface following density gradient centrifugation and used in the following assays.

ELISA.

Cytokines present in lung homogenates were measured by ELISA according to the manufacturer's instructions (Biolegend, San Diego, CA). Cytokine standards ranging from 15.6 to 4,000 pg/ml were prepared to determine the concentration of cytokine in the samples. For data analysis, a curve fit was applied to the standards and the sample concentrations were extrapolated from the standard curve using four-parameter logistic software (SoftMax Pro, Sunnyvale, CA).

Cell surface and intracellular flow cytometry.

Cell surface flow cytometry was performed on isolated BAL or lung cells using fluorochrome-conjugated antibodies to CD11b, CD11c, Gr-1, F4/80, CD45, CD4, CD8, βTcR, CD200, CD200R, and CD69 (BD Biosciences, San Jose, CA or ebiosciences, San Diego, CA). For intracellular cytokine staining, lung cells were incubated with PMA and ionomycin in the presence of brefeldin A (GolgiPlug; BD Biosciences). The cells were incubated with antibodies to CD45, CD4, and βTcR to identify CD4⁺ T cells and then fixed and permeabilized with 1% saponin followed by incubation with IL-17A, IL-4, or isotype control antibodies. A minimum of 10,000 events/sample were collected for BAL cells and 100,000 events/sample for lung cells on an LSRII cytometer (BD Biosciences). Expression of cell surface markers and intracellular cytokines was analyzed using DIVA software.

RNA isolation and qRT-PCR.

Total RNA was extracted from the upper right lobe of lung from individual mice using TRIzol (Invitrogen). Contaminating genomic DNA was removed by treatment with DNA-free (Ambion, Austin, TX) according to the manufacturer's directions. RNA (2 μg) was reverse transcribed into cDNA with a reverse transcription system (Promega, Madison, WI). Real-time PCR was performed with gene-specific primers to IL-17A, IL-4, IL-10, TNF, IFNγ, NLRP3, and caspase-1 using a LightCycler 480 real-time PCR thermal cycler (Roche Diagnostics, Indianapolis, IN). All primers and probes were chosen using the online software Universal Probe Library. Data were normalized to the HPRT gene and plotted as fold induction over unexposed control mice.

Lung function measurements.

Measurements of lung mechanics were made using the flexiVent system (SCIREQ, Montreal, QC, Canada). Mice were anesthetized using an intraperitoneal injection of ketamine/xylazine mixture (1:1, 0.15 ml/10 g body wt) and 1% inhaled isofluorane using a precision gas mixer (PEGAS 400; Columbus Instruments, Columbus, OH) and then tracheostomized with an 18-gauge needle and placed on the flexiVent. Quasistatic lung compliance was measured by obtaining pressure-volume (PV) curves. Each set of PV curves was preceded by two inflation maneuvers to total lung capacity to ensure equal volume history. Compliance was calculated by fitting data derived from the PV curves to the Salazar Knowles equation. Only measurements with a coefficient of determination (COD) greater than 0.9 were included in analysis. Measurements were repeated until at least two PV curves and three snapshot perturbations with acceptable CODs were obtained. The average measurement was calculated for each mouse.

Histology.

The left lobe of lung was removed and fixed with neutral-buffered formalin and embedded in paraffin. Sections (8 μm) were cut and stained with hematoxylin and eosin to analyze inflammation or Masson's trichrome to analyze lung fibrosis. Slides were digitized (Scanscope XT; Aperio, Vista, CA) and analyzed using Spectrum Imagescope software (v11.2, Aperio). Quantification of collagen staining in the trichrome-stained lung sections divided into 12 1-mm² regions was performed using the positive pixel count algorithm (v9.1). The blue collagen was detected using a positive hue value of 0.65, a hue width of 0.2 (to include a range of color shades), and a color saturation threshold of 0.1, all other settings remained at default values. Data are reported as the average number of positive and strong positive pixels/lung section.

Statistical analysis.

Data from two groups were analyzed by a one-way or two-way Student's t-test using GraphPad Prism statistical software (GraphPad Software, San Diego, CA). Data from three groups were analyzed with ANOVA and a Tukey's post hoc test. Intracellular cytokine staining data were analyzed by two-way ANOVA with a Bonferonni post hoc test. Differences were considered significant at P < 0.05.

Mice with long-term exposure to SR developed a lymphocytic alveolitis that is independent of TLRs 2 and 9.

One of the defining features of patients with both acute and chronic HP is the development of lymphocytic alveolitis. To measure the extent to which T cells contribute to alveolitis during long-term SR exposure we exposed wild-type (WT) and tlr2/9^−/− mice to SR for 4 or 14 wk and performed flow cytometry on cells isolated from the BALF 2 days following the last exposure (Table 1). The results demonstrate that WT mice exposed to SR for 14 wk exhibited an increased frequency of alveolar macrophages (AMs) (CD11c⁺/F4/80⁺) and T cells (βTcR⁺) compared with WT mice exposed for 4 wk. We did not observe a predominance of CD8⁺ T cells with long-term SR exposure; the ratio of CD4⁺ to CD8⁺ T cells was close to 1.0. Within the CD4⁺ and the CD8⁺ T cell population, the percentage of activated T cells (CD69⁺/βTcR⁺/CD4⁺ or CD8⁺) was not significantly different in WT mice with long-term exposure to SR compared with 4-wk exposure. Neutrophils were recruited into the lung within 6 h following SR exposure and their frequency dropped in the days following exposure (30). Two days following long-term SR exposure, the frequency of neutrophils in the alveolitis (∼15%) did not differ from that of the 4-wk SR-exposed mice. Despite the low frequency of neutrophils in the BAL at 2 days postexposure, long-term SR exposure is associated with a strong neutrophil influx at 1 day postexposure (% polymorphonuclear cells − WT unexposed = 0.6% ± 0.4; WT/SR exposed = 65% ± 10; tlr2/9^−/−/SR = 59 ± 26).

Our previous studies demonstrated that mice deficient in TLRs 2 and 9 had alterations in immune cell recruitment and activation following SR exposure for 3 wk. Specifically, neutrophil recruitment and production of the neutrophil chemokine CXCL2 was significantly reduced in tlr2/9^−/− mice 1 day post-SR exposure (4). However, in this study examining long-term SR exposure, there was no significant difference in the frequency of neutrophils in the tlr2/9^−/− mice compared with the WT mice at 2 days following 14 wk of SR exposure. Interestingly, levels of CXCL2 in the BALF were very low at this stage of the disease, whereas another neutrophil chemokine, CXCL1, was increased (Fig. 1). We observed a decrease in CD4⁺ T cells expressing the activation marker CD69 in tlr2/9^−/− mice at 4 wk postexposure; however, by 14 wk of SR exposure there was no difference in the frequency of CD69⁺/CD4⁺ T cells. These results suggest that immune cell recruitment and activation is not dependent on TLRs 2 and 9 following long-term SR exposure, in contrast to short-term exposure.

Long-term SR exposure caused an increase in CD200R expression by CD4⁺ T cell and AM populations.

The CD200R is upregulated on memory T cells and it has been suggested that its expression correlates with persistent antigen exposure (7, 34). To determine whether long-term exposure to SR led to an increase in CD200R expression or its ligand CD200, we performed flow cytometric analysis on cells isolated from the BALF of mice (Table 2). We measured an increase in CD200R expression by CD4⁺ T cells with 14 wk of SR exposure compared with 4 wk of SR exposure; the CD200R⁺ cells also expressed CD69, suggesting they had been activated. Expression of CD200 by CD4⁺ T cells was very low and did not change with increasing duration of exposure. There was no increase in CD200R or CD200 expression by CD8⁺ T cells. Expression of CD200R by CD4⁺ T cells was partially controlled by TLRs 2 and 9 because the tlr2/9^−/− mice had a reduced expression of CD200R on CD4⁺ T cells compared with the WT mice.

The inhibitory receptor:ligand pair CD200R:CD200 is also expressed on myeloid cells and is important in downregulating macrophage activation (22). Additionally, expression of CD200 by macrophages is believed to play a supporting role in fusion or multinucleation of macrophages (11). We measured expression of CD200 and CD200R on AMs by flow cytometry using CD11c⁺/F4/80⁺/autofluorescence⁺ as our criteria for identifying AMs in the BALF. Surprisingly, flow cytometric analysis revealed that this population could be divided into two subpopulations based on their expression of Gr1. Approximately 30% of AMs from WT mice exposed to SR for 4 or 14 wk were Gr1^high, whereas the remaining AM population was Gr1^low (Table 2). We considered the CD11c⁺/F4/80⁺/Gr1^low subpopulation to be AMs, and analysis of CD200R/CD200 expression revealed an increase in AMs expressing either CD200R or CD200 at 14 wk compared with 4 wk of SR exposure. The increase in expression of both CD200R and CD200 was partially dependent on TLRs 2 and 9 because there was a significant reduction in their expression in the tlr2/9^−/− mice compared with the WT mice with 14 wk of SR exposure. These results suggest that long-term exposure to SR correlates with an increase in the inhibitory receptor ligand pair CD200-CD200R, and that its regulation on AMs is partially controlled by TLRs 2 and 9.

Long-term SR exposure caused increased lung dysfunction.

To determine the extent to which long-term exposure to SR results in lung pathology we measured several parameters of disease severity. An increase in protein in BALF is a measure of decreased barrier integrity, and we detected an increase in protein in the BALF of WT mice with long-term exposure to SR compared with unexposed mice (Fig. 2A). Barrier dysfunction was also detected after long-term exposure in the mice with deficiency of TLRs 2 and 9. The alarmin IL-1α is present in epithelial cells lining the lung and is released by those cells when they are damaged or dying, initiating an inflammatory response (8, 11). We measured the level of IL-1α in the lungs of WT and tlr2/9^−/− mice with long-term exposure to SR (Fig. 2B). The results demonstrate an increase in IL-1α in the lungs of WT mice compared with the tlr2/9^−/− SR-exposed mice and unexposed mice. There was an increase in the level of IL-1α in tlr2/9^−/− SR-exposed mice compared with unexposed mice but not to the level of the SR-exposed WT mice. To determine the level of inflammation following long-term SR exposure, we examined hematoxylin and eosin-stained lung sections from WT and mutant mice (Fig. 2C). The sections revealed an increase in neutrophils in the interstitial lung tissue surrounding the areas of granulomatous inflammation (Fig. 2C, arrow). In addition the appearance of multinucleate cells was observed in WT and tlr2/9^−/− mice exposed to SR, a feature also common to patients with chronic HP. There were no differences in the inflammation observed in WT vs. tlr2/9^−/− mice.

To measure the level of collagen deposition in the lungs of mice with long-term exposure to SR, lung sections were stained with Masson trichrome (Fig. 3). Compared with unexposed mice, lung sections from SR-exposed WT mice exhibited an increase in blue staining representing collagen deposition around the bronchi, within the granulomatous areas, and in the alveolar walls (Fig. 3, C and D). An increase in collagen deposition was also observed in lung sections from SR-exposed tlr2/9^−/− mice (Fig. 3, E and F). To quantify the amount of collagen in the lung, the stained sections were analyzed using the positive pixel algorithm in the Spectrum Imagescope software (Fig. 3G). The results demonstrate a significant increase in the level of collagen deposition in lung sections from WT mice with long-term exposure to SR compared with unexposed WT mice. However, SR-exposed tlr2/9^−/− mice did not have a significant increase in collagen deposition compared with WT-unexposed mice.

To determine whether long-term SR exposure led to impaired lung function, we measured quasistatic compliance and total lung capacity in SR-exposed WT and tlr2/9^−/− mice. Long-term exposure of both WT and tlr2/9^−/− mice to SR resulted in a significant decrease in quasistatic compliance compared with unexposed mice (Fig. 4). Additionally, total lung capacity in SR-exposed WT mice were significantly decreased compared with that of unexposed mice (Fig. 4). Altogether, these results demonstrate that long-term exposure to SR caused barrier dysfunction and fibrosis resulting in a significant decrease in lung function.

Long-term exposure to SR by WT mice was associated with an IL-17 response but not a shift to an IL-4 response.

There is evidence in chronic pigeon breeders disease that the development of fibrosis correlates with a switch from a Th1-type response to a Th2-type immune response. However, studies with earlier time points in the SR model of FLD have demonstrated that IL-17 is associated with increased disease severity. To determine whether mice exposed to SR for a long time switch from a Th17-type response to the more profibrotic Th2-type response, we performed quantitative RT-PCR (qRT-PCR) and measured the expression of cytokine mRNA in the lungs of exposed and control mice. SR exposure resulted in a strong upregulation of IL-17A expression, which was further increased in tlr2/9^−/− mice compared with unexposed mice (Fig. 5A). We measured the level of IL-17 in the BALF 1 day after SR exposure and tlr2/9^−/− mice had a significant increase in IL-17 compared with both WT and unexposed controls (Fig. 5C). There was no increase in IL-17 in the BALF of exposed WT mice. Expression of IL-4 was not detected by qRT-PCR and no bands were observed when the PCR product was run on a gel (Fig. 5C). However, IL-13 mRNA was detected in WT SR-exposed mice but tlr2/9^−/− mice demonstrated a significant increase in IL-13 mRNA compared with both WT exposed and unexposed mice. Expression of the neutrophil chemokine CXCL2, IL-10, and IFNγ mRNA were also increased in mice following long-term SR exposure compared with unexposed mice, but there were no differences between WT and tlr2/9^−/− mice exposed to SR.

We performed intracellular cytokine staining on cells isolated from the lungs of SR-exposed (15 wk) mice to measure levels of Th17 and Th2 cells. The cells were stimulated ex vivo with PMA and ionomycin and the results demonstrate an increase in IL-17⁺/CD4⁺ Th cells in WT mice and a significant increase in tlr2/9^−/− mice compared with unexposed mice (Fig. 5, D and E). There were no detectable IL-4⁺/CD4⁺ Th cells at this time point in either WT or tlr2/9^−/− mice exposed to SR, suggesting that there was no switch to a Th2 response. These results suggest that in WT mice there was no shift to a Th2 response with long-term SR exposure and the disease remains one that is predominantly IL-17-mediated. However, the increase in IL-13 mRNA in tlr2/9^−/− mice suggests that a Th2 environment may be developing in those mice.

Long-term exposure to SR caused an increase in IL-1β.

IL-1β plays key roles in IL-17-mediated responses; IL-1β and IL-23 induce IL-17 production by γδ T cells (41). IL-1β can further amplify the response by increasing IL-23 production, which in turn, stimulates IL-17 production (27). IL-1β also contributes to Th17 cell differentiation, in part through the production of IL-6 and is closely associated with IL-17-mediated pathology (9, 10). We measured the level of IL-1β in lung homogenates from mice with long-term exposure to SR and found that was increased in both WT and tlr2/9^−/− mice (Fig. 6A). The canonical pathway for IL-1β production is activation of the inflammasome resulting in activation of caspase-1 and cleavage of pro-IL-1β to its mature form. We used qRT-PCR to measure the level of inflammasome components in the lungs of long-term SR-exposed mice. The results demonstrate an increase in mRNA for NLRP3, one of the NLR family members capable of initiating inflammasome formation, and caspase-1 in WT and tlr2/9^−/− mice (Fig. 6B). Stimulation of WT bone marrow-derived macrophages (BMDMs) with SR plus ATP resulted in release of IL-1β into the culture supernatant suggesting that SR can induce pro-IL-1β production and in the presence of an NLRP3 activator (ATP) result in IL-1β release (Fig. 6C). However, pro-IL-1β production is dependent on TLRs 2 and 9 because BMDMs derived from tlr2/9^−/− mice did not produce IL-1β following SR-plus-ATP stimulation. These results suggest an association between amplification of the IL-17 response and IL-1β. In addition, activation of PRRs that control inflammasome formation may play a role in lung pathology associated with long-term exposure to SR.