Effect of cytokine interplay on macrophage polarization during chronic pulmonary infection with Cryptococcus neoformans

Abstract

The immune response to Cryptococcus neoformans following pulmonary infection of C57BL/6 wild-type (WT) mice results in the development of persistent infection with characteristics of allergic bronchopulmonary mycosis (ABPM). To further clarify the role of Th1/Th2 polarizing cytokines in this model, we performed kinetic analysis of cytokine responses and compared cytokine profiles, pathologies, and macrophage (Mac) polarization status in C. neoformans-infected WT, interleukin-4-deficient (IL-4(-/-)), and gamma interferon-deficient (IFN-γ(-/-)) C57BL/6 mice. Results show that cytokine expression in the infected WT mice is not permanently Th2 biased but changes dynamically over time. Using multiple Mac activation markers, we further demonstrate that IL-4 and IFN-γ regulate the polarization state of Macs in this model. A higher IL-4/IFN-γ ratio leads to the development of alternatively activated Macs (aaMacs), whereas a higher IFN-γ/IL-4 ratio leads to the generation of classically activated Macs (caMacs). WT mice that coexpress IL-4 and IFN-γ during fungal infection concurrently display both types of Mac polarization markers. Concurrent stimulation of Macs with IFN-γ and IL-4 results in an upregulation of both sets of markers within the same cells, i.e., formation of an intermediate aaMac/caMac phenotype. These cells express both inducible nitric oxide synthase (important for clearance) and arginase (associated with chronic/progressive infection). Together, our data demonstrate that the interplay between Th1 and Th2 cytokines supports chronic infection, chronic inflammation, and the development of ABPM pathology in C. neoformans-infected lungs. This cytokine interplay modulates Mac differentiation, including generation of an intermediate caMac/aaMac phenotype, which in turn may support chronic "steady-state" fungal infection and the resultant ABPM pathology.

Fig. 1.

Effect of IL-4 and IFN-γ deletion on cytokine expression in C. neoformans-infected lungs. Total lung RNA was isolated from uninfected and C. neoformans-infected mice at 2, 3, and 5 wpi. RT PCR was performed for selected cytokines, and electrophoresis of PCR products was performed in a 2% agarose gel. Photographs show bands of the predicted sizes. φ, no-RNA control; Un, uninfected WT mice. Each lane is representative of 3 to 5 mice/group/time point. Baseline expression levels of cytokines were similar in uninfected lungs from WT, IL-4^−/−, and IFN-γ^−/− mice.

Fig. 2.

Effect of IL-4 and IFN-γ deletion on the kinetics of iNOS and arginase mRNA expression during C. neoformans infection. RNA from WT, IL-4^−/−, and IFN-γ^−/− mice was isolated and analyzed as described in the legend for Fig. 1. Photographs show bands of the predicted sizes for each PCR product. φ, no-RNA control; Un, uninfected WT mice. Each lane is representative of 3 to 5 mice/group/time point.

Fig. 3.

Effect of IL-4 and IFN-γ deletion on ABPM pathology in C. neoformans-infected lungs. Photomicrographs are H&E-stained sections of C. neoformans-infected lungs from WT (A, D, and G), IL-4^−/− (B, E, and H), and IFN-γ^−/− (C, F, and I to K) mice at 5 wpi. Low-power images (A, B, and C) demonstrate differences in the extent of inflammatory responses and consolidation of lung tissue (10× objective). High-power images (D, E, and F) of the boxed areas demonstrate cellular composition and differences in Mac morphology (40× objective). Note the absence of microbe in IL-4^−/− lungs and the increased microbial presence and accumulation of eosinophilic deposits in the Mac in IFN-γ^−/− lungs. High-power images of airway sections (G, H, and I) demonstrate differences in airway epithelium. Note the presence of eosinophilic inclusions in the epithelial cells of IFN-γ^−/− mice (I). Severe pulmonary pathology in IFN-γ^−/− lungs is highlighted by the formation of cryptococcomas (J, 10× objective), the presence of eosinophilic crystals in the airway lumen (K, 40× objective, red arrowheads), and the secretion of eosinophilic inclusions in the airway lumen, or hyalinosis (K, green arrows).

Fig. 4.

Effects of IFN-γ and IL-4 deletion on regulation of Ym1/2 production by Mac in uninfected and C. neoformans-infected lungs. Sections of uninfected WT, IL-4^−/−, and IFN-γ^−/− mouse lungs (A, B, and C) and C. neoformans-infected lungs (D to O) were immunostained with anti-Ym1/2 antibody and developed with a peroxidase-linked detection system. Sections shown in panels A to F and J to L were photographed using a 10× objective, while those in panels G to I and M to O were photographed using a 100× objective. Note that the basal level of Ym1/2 production by Mac in the uninfected lungs was not affected by IL-4 or IFN-γ deletion; however, accumulation of Ym1/2-positive Mac and intracellular Ym1/2 crystal deposition are ablated in IL-4^−/− and augmented in IFN-γ^−/− infected lungs.

Fig. 5.

Effects of IFN-γ and IL-4 deletion on numbers of Ym1/2-expressing CD11b⁺ cells in C. neoformans-infected lungs. (A) CD11b-expressing cells were isolated from C. neoformans-infected WT, IL-4^−/−, and IFN-γ^−/− mice (3 wpi) by MACS separation and counted under a microscope. The data are means from 3 independent experiments, with ≥3 mice/group/experiment. (B) CD11b-selected cells were stained with Ym1/2 Ab to enumerate Ym1/2-positive and -negative cells. Approximately 100 cells were counted in 10 different fields on a slide. **, P < 0.005 for the IL-4^−/− group compared to the WT group; ††, P < 0.005 for the IL-4^−/− group compared to the IFN-γ^−/− group. Note that the total number of CD11b-selected cells was not affected by IL-4 or IFN-γ deletion but that there was a strong effect on the numbers of cell staining positive for Ym1/2.

Fig. 6.

Effects of IFN-γ and IL-4 deletion on the expression of selected caMac and aaMac genes by Mac from uninfected and C. neoformans-infected mice. RNA from enriched Mac isolates was analyzed at 3 wpi. (A) RT PCR mRNA analysis of RNA from iNOS, Arg1, and Fizz1 was performed as described in the legend for Fig. 1. The band photographs are representative of 3 independent experiments. (B) RNA from each sample was subjected to semiquantitative RT PCR analysis for Arg1 and Fizz1, and expression was normalized to GAPDH (glyceraldehyde-3-phosphate dehydrogenase) housekeeping gene expression within each sample. The average relative expression of each gene in IL-4^−/− cells in each independent experiment was used to calculate the relative expression ratios for WT and IFN-γ^−/− cells. The graph is representative of 3 or more experiments, where each sample was run in triplicate (n = 3 mice/group).

Fig. 7.

Effects of IFN-γ and IL-4 deletion on the expression of aaMac- and caMac-specific markers by Mac from C. neoformans-infected mice. CD11b-selected cell isolates were analyzed for mRNA expression of chemokines (A) or MGL2 and 12/15-LOX (B) at 3 wpi. Photographs show the bands of predicted sizes and are representative of 3 or more independent experiments.

Fig. 8.

Effects of IFN-γ and IL-4 on induction of iNOS2 and Arg1 in BMM. BMM from WT mice were precultured and isolated as described in Materials and Methods and transferred to 8-well glass Labtek tissue culture plates at a cell density of 1 × 10⁵ cells/well for 24 h at 37°C and 5% CO₂ in media alone or with recombinant mouse IFN-γ, IL-4, or a combination of both cytokines. (A) RNA from these cells was collected and analyzed by RT PCR. (B) Cells were fixed, permeabilized, incubated sequentially with anti-iNOS FITC-conjugated antibodies, anti-Arg1 Alexa Fluor 546 phalloidin (Imgenex, San Diego, CA), and DAPI (to stain cell nuclei), and analyzed with a Zeiss LSM510 confocal microscope. Note the red Arg1 staining in IL-4-treated BMM, the green iNOS staining in IFN-γ-treated BMM, and the ARG1/iNOS double-positive staining in BMM treated with a combination of IL-4/IFN-γ. The individual green- and red-channel pictures are presented in the left and middle columns, respectively; the merged images are presented in the right column.