Immune modulation with sulfasalazine attenuates immunopathogenesis but enhances macrophage-mediated fungal clearance during Pneumocystis pneumonia

Abstract

Although T cells are critical for host defense against respiratory fungal infections, they also contribute to the immunopathogenesis of Pneumocystis pneumonia (PcP). However, the precise downstream effector mechanisms by which T cells mediate these diverse processes are undefined. In the current study the effects of immune modulation with sulfasalazine were evaluated in a mouse model of PcP-related Immune Reconstitution Inflammatory Syndrome (PcP-IRIS). Recovery of T cell-mediated immunity in Pneumocystis-infected immunodeficient mice restored host defense, but also initiated the marked pulmonary inflammation and severe pulmonary function deficits characteristic of IRIS. Sulfasalazine produced a profound attenuation of IRIS, with the unexpected consequence of accelerated fungal clearance. To determine whether macrophage phagocytosis is an effector mechanism of T cell-mediated Pneumocystis clearance and whether sulfasalazine enhances clearance by altering alveolar macrophage phagocytic activity, a novel multispectral imaging flow cytometer-based method was developed to quantify the phagocytosis of Pneumocystis in vivo. Following immune reconstitution, alveolar macrophages from PcP-IRIS mice exhibited a dramatic increase in their ability to actively phagocytose Pneumocystis. Increased phagocytosis correlated temporally with fungal clearance, and required the presence of CD4(+) T cells. Sulfasalazine accelerated the onset of the CD4(+) T cell-dependent alveolar macrophage phagocytic response in PcP-IRIS mice, resulting in enhanced fungal clearance. Furthermore, sulfasalazine promoted a TH2-polarized cytokine environment in the lung, and sulfasalazine-enhanced phagocytosis of Pneumocystis was associated with an alternatively activated alveolar macrophage phenotype. These results provide evidence that macrophage phagocytosis is an important in vivo effector mechanism for T cell-mediated Pneumocystis clearance, and that macrophage phenotype can be altered to enhance phagocytosis without exacerbating inflammation. Immune modulation can diminish pulmonary inflammation while preserving host defense, and has therapeutic potential for the treatment of PcP-related immunopathogenesis.

Figure 1. SSZ treatment reduces the severity of PcP-related IRIS.

Pc-infected SCID mice were immunologically reconstituted and then treated with PBS or SSZ beginning at day 1 post-reconstitution. (A and B) Body weight and respiratory rates were monitored non-invasively. (C, D, E) Dynamic lung compliance, lung resistance and BAL albumin content were measured at 13, 18, and 25 days post-reconstitution. Values are mean ±1 standard error measurement (SEM) (n≥15 for day 13, n≥37 for day 18, and n≥15 for day 25). ** and #, P<0.05 as compared to uninfected mice at the same time point. *, P<0.05 as compared to SSZ treated mice at the same time point. Mean represents combined data from eight independent experiments.

Figure 2. SSZ reduces pulmonary inflammation during PcP-related IRIS.

Pc-infected SCID mice were immunologically reconstituted, and then treated with PBS (Panels A–C) or SSZ (Panels D–F) beginning at day 1 post-reconstitution. Lung sections from mice at days 13 (A, D), 18 (B, E) or 25 (c, F) post-reconstitution were stained with Hematoxylin and Eosin. Representative pictures were taken by microscopy under 40× magnification. Arrows denote peribronchiolar regions in each section, while asterisks denote alveolar regions.

Figure 3. SSZ reduces inflammatory chemokine and cytokine production during PcP-related IRIS.

Pc-infected SCID mice were immunologically reconstituted, and then treated with PBS or SSZ. (A, B, C, and D) MCP-1, RANTES, TNF-α, and IFN-γ levels were measured in the BAL fluid. Values are mean ±1 SEM (n≥15/time point/group). ** and #, P<0.05 as compared to uninfected mice at the same time point. *, P<0.05 as compared to SSZ treated mice at the same time point. Mean represents combined data from three independent experiments.

Figure 4. SSZ enhances Pc clearance through a CD4-dependent mechanism.

(A) Pc burden in immune reconstituted SCID mice that received SSZ- (∇) or PBS (▪). Values are mean ±1 SEM (n≥15 for day 13, n≥37 for day 18, and n≥15 for day 25). ** and #, P<0.05 as compared to uninfected mice at the same time point. *, P<0.05 as compared to SSZ treated mice at the same time point. Mean represents combined data from eight independent experiments. (B) Pc burden in non-reconstituted SCID mice after 13 and 20 days of SSZ (∇), PBS (▪), or TMP-SMX (◊) treatment. Values are mean ±1 SEM (n≥11 for day 13, and n≥17 for day 20). *, P<0.05 as compared to SSZ and PBS treated mice at the same time point. Mean represents combined data from three independent experiments. (C) Pc burden in immune reconstituted, CD4-depleted SCID mice receiving SSZ (∇) or PBS (▪). Values are mean ±1 SEM (n≥8 for day 18 and n≥14 for day 23). Mean represents combined data from four independent experiments.

Figure 5. SSZ enhances CD4+ T cell-dependent AM phagocytosis of Pc.

BAL cells were collected from PBS- and SSZ- treated mice at day 13, 17 or 18 (17/18) and 21 post-reconstitution. Cells were stained with antibodies specific for CD11c (green) and Pc (red). (A) Imaging flow cytometry was used to quantify Pc internalization by AMs. Representative images of brightfield (BF), CD11c, Pc, and merged CD11c/Pc are shown for SSZ or PBS-treated mice following immune reconstitution. The no internalization control is a representative CD11c⁺ cell without internalized Pc. ImageStream IDEAS software was used to quantify (B) percentile and (C) number of AMs with Internalized Pc in immune reconstituted SCID mice treated with SSZ (∇) or PBS (•). Values are mean ±1 SEM (n≥5 for day 13, n≥9 for day 17/18, and n≥9 for day 21). *, P<0.05 as compared to PBS-treated mice at the same time point. **, P<0.05 as compared to PBS-treated mice at day 17/18. Mean represents combined data from three independent experiments. (D) Quantification of in vivo Pc phagocytosis in the presence or absence of CD4⁺ T cells. Reconstituted SCID mice were treated with PBS- (•), PBS + anti-CD4 (∇), SSZ (▪), or SSZ + anti-CD4 (⋄). Phagocytosis was measured at 17/18 or 21 days post-reconstitution. Values are mean +1 SEM (n≥5 for day 13, n≥9 for day 17/18, and n≥8 for day 21). *, P<0.05 as compared to PBS treated, PBS + anti-CD4-treated, and SSZ + anti-CD4-treated mice at the same time point. **, P<0.05 as compared to PBS + anti-CD4-treated mice at the same time point. Mean represents combined data from three independent experiments. Note that values for PBS treated, PBS + anti-CD4-treated, and SSZ + anti-CD4-treated mice at day 17/18 are all indistinguishable, and the symbols overlap.

Figure 6. Co-localization of internalized Pc with the lysosomal protein LAMP-1.

BAL cells were collected from PBS-treated mice at day 21 and SSZ-treated mice at day 17 post-reconstitution. Cells were stained with antibodies specific for CD11c (gray), Pc (green) and LAMP-1 (red). Nuclei were stained with DAPI (blue). Control image compares two AM in the same field; one with internalized Pc and one without. The scale bar represents 10 µM. Images are representative of two independent experiments.

Figure 7. SSZ promotes a TH2 cytokine environment during PcP-related IRIS.

Pc-infected SCID mice were immunologically reconstituted, and then treated with PBS or SSZ. (A) IL-4 levels and (B) IL-4/IFN-γ ratios were measured in the lung homogenates of experimental mice at indicated time points. Values are mean ±1 SEM (n≥10/time point/group). *, P<0.05 as compared to PBS treated mice at the same time point. Mean represents combined data from five independent experiments.

Figure 8. SSZ promotes an alternatively activated macrophage phenotype during PcP-related IRIS.

BAL cells were collected from PBS- and SSZ- treated mice at day 13 post-reconstitution. (A) Cells were stained with antibodies specific for CD11c (red) and iNOS (green). (B) Cells were stained with antibodies specific for CD11c (green) and Arginase (red). Nuclei were stained with DAPI (blue). Images shown are 40× magnification. ImageJ software was used to quantify the mean fluorescent intensity of iNOS (C) and Arginase (D) staining of CD11c positive cells. Values are mean ±1 SEM (Panel A, n = 23 for PBS and n = 58 for SSZ; Panel B, n = 44 for PBS and n = 105 for SSZ). *, P<0.05 as compared to PBS treated mice at the same time point. Mean represents combined data from two independent experiments.