Characterization Of Immune Responses During Pneumocystis Pneumonia

Clinical Center

Investigators

Linked publications, trials & patents

Abstract

We have previously examined immune responses in immunodecient mice, including those with defective innate immune responses, such as Myd88 knockout mice which are decient in TLR signals, and CD1d KO mice, which lack NKT cells. Our studies have shown that these mice are not susceptible to uncontrolled Pneumocystis infection, suggesting that neither TLR signaling, nor NKT cells are critical to controlling Pneumocystis infection. We have examined changes in expression of a variety of chemokine and chemokine receptor genes, using Taqman real-time PCR assays, and have been able to demonstrate upregulation of CCR2 and its ligands during the response to Pneumocystis infection. However, in examining the susceptibility of CCR2 knock-out mice to Pneumocystis infection, we found that they are able to clear infection similar to wild-type animals. We also examined CX3CR1 knock-out mice for susceptibility to Pneumocystis infection, since CX3CR1 and its ligand identify alternately activated macrophages and are critical to clearance of another fungus, Candida, in mouse models. These mice were again able to clear Pneumocystis infection with kinetics similar to wild-type mice. Using microarray analysis of puried CD4 cells isolated from lungs of Pneumocystis-infected mice, we have shown that CXCR6 is preferentially upregulated compared to cells from uninfected animals, and by Q-PCR that both CXCR6 and its ligand, CXCL16, are unregulated in lung tissue from immunocompetent Pneumocystis-infected mice. Using CXCR6 knock-out mice, we again found that they were able to clear infection with kinetics similar to wild-type mice. Intriguingly, inux of CD4 cells driving GFP expression through the CXCR6 promoter was similar in both heterozygous mice, with functional CXCR6, and homozygous mice, who were unable to express CXCR6. This suggests that CXCR6 is not required for recruitment of these cells to the lungs, and that there is either substantial redundancy in chemokine function of these CD4+T cells, or that these are resident memory cells that were already present in the lung. CD4+ T cells are critical in clearing Pneumocystis infection. To determine if the CD4 molecule is required for this clearance, we examined the kinetics of infection in a CD4 KO mouse model. We found that control of infection is impaired: infection was never cleared in most mice, but organism loads were lower than those found in CD40L KO mice, suggesting some mechanisms were controlling infection to a limited extent. By immunolabeling we demonstrated an influx of CD8+ T cells, and an even greater influx of double negative (i.e. CD3+, CD4- and CD8-) T cells. qPCR documented that levels of ThPOK, a transcription factor that programs T cells to a CD4 phenotype, were similar in wild type and CD4 KO mice, suggesting that some of the double negative cells were programmed to function to some extent like CD4+ T cells. We utilize flow cytometry to examine the kinetics of immune responses in the lung of healthy and immunodeficient animals exposed to Pneumocystis. We have found that interferon-gamma, and less predictably, IL-17, expression is increased in lymphocytes of the lung, primarily CD4 cells, during Pneumocystis infection and clearance, suggesting that Th1 and Th17 CD4+ T cells may play a role in clearing infection. Immunodecient CD40L KO mice did not show changes in either of these populations during Pneumocystis infection. IL-17A knockout mice are, however, able to clear Pneumocystis infection, demonstrating that IL-17 is not critical to control of Pneumocystis infection. We have extended these studies to demonstrate that Th2 and Treg cells show minimal changes in in the lung during Pneumocystis infection in healthy mice, and that an anti-interferon-gamma antibody administered to immunocompetent mice causes a shift from Th1 to Th17 responses, but does not impact control of infection, with kinetics of clearance that are similar to untreated controls. We have also examined the contribution of beta-glucans in cysts to the inflammatory response to Pneumocystis. We have demonstrated that, similar to other pathogenic fungi, glucans are to a large extent masked by surface proteins in 3 Pneumocystis species (P. murina, P. carinii, and P. jirovecii). Moreover, we have shown, by comparing infected mice treated with caspofungin, an inhibitor of beta 1,3 glucan synthase, to untreated mice, that beta-glucans are major contributors to the inammatory response to Pneumocystis. We are also examining the response of dendritic cells to the most abundant Pneumocystis antigen, the major surface glycoprotein (MSG). We found that MSG did not activate dendritic cells, as assessed by surface markers, cytokine production, and microarray analysis. We hypothesize that this is because Pneumocystis cannot produce high mannans, which are important activators of innate immune responses through binding to C-type lectins, because they have lost the enzymes needed for such high mannosylation. Of note however, Msg was able to bind to 2 relevant C-type lectins, DC-SIGN and macrophage mannose receptor. We are following up on these initial studies to better characterize the interaction between dendritic cells and MSG, specically to determine if MSG modulates the interaction of dendritic cells with beta glucans, which are present in the cyst wall. Mucosal associated invariant T cells (MAIT cells) are innate T cells that can rapidly respond to riboavin metabolites with potentially protective effector functions. Because Pneumocystis can produce riboflavin metabolites, we examined the role of MAIT cells in response to Pneumocystis infection in immunocompetent C57BL/6J mice. We found that MAIT cells do increase during Pneumocystis infection in the latter strain, and that they remain increased for many weeks after infection is cleared. However, we also demonstrated that MAIT cells are not needed for clearance of infection, since MR1 knockout mice, which lack MAIT cells, are able to clear infection with kinetics similar to C57BL/6J mice. Understanding host-Pneumocystis interactions is an important step in developing novel approaches to the management of PCP. We have undertaken studies to examine binding of Pneumocystis to host cells, to determine which host molecules are critical to such binding. While type 1 pneumocytes are the primary cells in the lung to which organisms bind, based on EM studies, they can also bind to type 2 pneumocytes. MUC1 is a surface protein of type 2 pneumocytes that has been shown to play a role in binding of other organisms. We have recently demonstrated, using transfection or knock-down methods, that MUC1 also plays a role in binding of Pneumocystis in vitro, and that MUC1 co-localized to some extent with organisms in a mouse model. Theses studies suggest that MUC1 may play a role in adherence of the organism to lung alveolar cells, perhaps as a rst step in binding to cells after inhalation. Finally, we are undertaking studies to examine immune responses to Pneumocystis infection in mice using single cell RNAseq. We have begun by examining responses in immunocompetent mice, and are examining the expression profiles of both immune (T and B lymphocytes, macrophages, and dendritic cells) and non-immune (epithelial and endothelial) cells. These studies should provide more insights into relevant immune responses as it will allow a more granular characterization of cellular responses in the lung, potentially providing a better understanding of the specic cell populations that are responsible for controlling infection.