Anti-PD-1 Antibody Treatment Promotes Clearance of Persistent Cryptococcal Lung Infection in Mice

Abstract

Activation of immunomodulatory pathways in response to invasive fungi can impair clearance and promote persistent infections. The programmed cell death protein-1 (PD-1) signaling pathway inhibits immune effector responses against tumors, and immune checkpoint inhibitors that block this pathway are being increasingly used as cancer therapy. The objective of this study was to investigate whether this pathway contributes to persistent fungal infection and to determine whether anti-PD-1 Ab treatment improves fungal clearance. Studies were performed using C57BL/6 mice infected with a moderately virulent strain of Cryptococcus neoformans (52D), which resulted in prolonged elevations in fungal burden and histopathologic evidence of chronic lung inflammation. Persistent infection was associated with increased and sustained expression of PD-1 on lung lymphocytes, including a mixed population of CD4⁺ T cells. In parallel, expression of the PD-1 ligands, PD-1 ligands 1 and 2, was similarly upregulated on specific subsets of resident and recruited lung dendritic cells and macrophages. Treatment of persistently infected mice for 4 wk by repetitive administration of neutralizing anti-PD-1 Ab significantly improved pulmonary fungal clearance. Treatment was well tolerated without evidence of morbidity. Immunophenotyping revealed that anti-PD-1 Ab treatment did not alter immune effector cell numbers or myeloid cell activation. Treatment did reduce gene expression of IL-5 and IL-10 by lung leukocytes and promoted sustained upregulation of OX40 by Th1 and Th17 cells. Collectively, this study demonstrates that PD-1 signaling promotes persistent cryptococcal lung infection and identifies this pathway as a potential target for novel immune-based treatments of chronic fungal disease.

Figure 1

Induction of persistent cryptococcal lung infection in C57BL/6 mice. (A - D) C57BL/6 mice were infected by the intratracheal route with C. neoformans strain 52D. At day 0 (uninfected) and 2, 3, and 4 weeks post infection (WPI), lungs were harvested for analysis. (A) Fungal lung burden was assessed using CFU assays. (B) Number of CD45⁺ lung leukocytes was assessed using flow cytometric analysis. (C, D) Representative lung sections (H&E stain) from infected mice at 4 WPI at 200 X (C) and 400 X (D) magnifications. Note the presence of: numerous intracellular cryptococci within macrophages (closed block arrows), extracellular cryptococci in the alveolar space (black arrowheads), eosinophils (open block arrows), and extracellular crystals (thin black arrows). For A, B: n=4–5 mice assayed individually per time-point; *p<0.05 by ANOVA with Fisher’s LSD post hoc test vs. Day 0 (uninfected).

Figure 2

Persistent cryptococcal lung infection promotes lymphoid cell accumulation and increased T cell PD-1 expression. Following infection of mice with C. neoformans strain 52D, lungs were collected and processed for flow cytometry analysis at 0, 2, 3, and 4 WPI. Lung single cell suspensions were subjected to 12-color flow cytometry for analysis of lymphoid cell populations and PD-1 expression (A-H). Representative gating on cells obtained at 3 WPI is shown. Following selection of viable leukocytes based on expression of CD45 and a viability dye (not shown), T and B cells were characterized based on expression of CD19 (B cells; A) or CD3ε (T cells; A). Next, subsets of T cells were characterized by expression of CD8α (CD8⁺ T cells; B), CD4 (CD4⁺ T cells; B), or TCRδ (γδ T cells; C). Representative PD-1 staining pattern for CD4⁺ T cells is shown (D). Subsets of CD4⁺ T cells were characterized based on expression of hallmark transcription factors T-bet (Th1 cells; E), RORγt (Th17 cells; F), Gata3 (Th2 cells; G), and FoxP3 (Treg cells; H). (I-L) Application of this gating scheme facilitated: (I) the enumeration of specific lymphocyte subsets; (J) PD-1 expression on lymphocytes subsets; (K) enumeration of CD4⁺ T cell subsets; and (L) PD-1 expression on CD4+ T cell subsets. For I-L: n=4–5 mice assayed individually per time-point; *=p<0.05 by ANOVA with Fisher’s LSD post hoc test vs. Day 0 (uninfected).

Figure 3

Persistent cryptococcal lung infection causes accumulation of numerous myeloid cell populations and increased dendritic cell and macrophage PD-L1 and PD-L2 expression. Single cell suspensions of lung cells were subjected to 12-color flow cytometry for analysis of myeloid cell populations (A-F). Representative gating on cells obtained at 3 WPI is shown. Following selection of viable leukocytes (not shown), T and B cells were excluded by expression of CD19 or TCRβ (not shown). Next, neutrophils were characterized based on expression of CD11b and Ly6G (A). Non-neutrophils were then separated into Siglec F⁺ and Siglec F⁻ populations (B). Within the Siglec F⁺ population, alveolar macrophages were identified as CD11b⁻CD11c⁺, whereas eosinophils were defined as CD11b⁺CD11c⁻ (C). Siglec F⁻ cells were gated on CD11c⁺ cells (D). The Siglec F⁻CD11c⁺ population was divided into exudate macrophages (ExMs) and dendritic cells (DCs); exudate macrophages were characterized as autofluorescence (AF)⁺ whereas dendritic cells were AF⁻ (E). Dendritic cells were divided into three subgroups based on expression of CD24 and CD11b (F): CD11b⁻ cDCs (CD24⁺CD11b⁻; i.e., CD103⁺ DCs), CD11b⁺ cDCs (CD24⁺CD11b⁺), and monocyte-derived DCs (moDCs; CD24⁻CD11b⁺). (G-H) Application of this gating scheme facilitated the enumeration of: (G) granulocyte subsets and (H) myeloid cell subsets including CD11b⁻ and CD11b⁺ conventional DCs (cDCs), monocyte-derived DCs (moDCs), alveolar macrophages (AMs), and exudate macrophages (ExMs). Expression of PD-L1 and PD-L2 on myeloid cells was assessed by flow cytometry. Relative expression of PD-L1 (I) and PD-L2 (J) on dendritic cells and macrophages is shown. For G-J: n=4–5 mice assayed individually per time-point; *=p<0.05 by ANOVA with Fisher’s LSD post hoc test vs. Day 0 (uninfected).

Figure 4

Anti-PD1 antibody treatment promotes fungal clearance in mice with cryptococcal lung infection. (A) C57BL/6 mice were infected I.T. with C. neoformans strain 52D. Beginning at 3 WPI, mice were administered 200 μg of either neutralizing anti-PD-1 antibody (RMP1-14) or isotype-matched control antibody (2A3) twice per week for 2–4 weeks of treatment (WOT). (B) Weight change in cohorts of treated mice. (C) Fungal burden in lungs, spleens, and brains of treated cohorts of mice. For B, C: n=4–9/cohort assayed individually in two separate experiments; *=p<0.05 by unpaired Student t test.

Figure 5

Effect of anti-PD-1 antibody treatment on lung histopathology. Representative lung sections obtained from mice treated (at 3 WPI) with control (A,B) and anti-PD-1 antibody (C,D) for 4 weeks. Sections were H&E stained and examined by light microscopy at 200x (A,C) or 400x (B,D) magnification. Sections at low magnification (A,C) show similar overall immune cell infiltration between treatments. Sections obtained from the isotype-treated group at high magnification (B) show numerous cryptococci located both intracellularly (closed block arrows) and extracellularly (black arrowheads). Sections obtained from the anti-PD-1 treated group at high magnification (D) reveal fewer cryptococci and a diminished number of foamy macrophages (relative to control treated mice).

Figure 6

Anti-PD-1 antibody treatment does not substantially alter lung myeloid cell accumulation in the lungs of mice with cryptococcal lung infection. Infected C57BL/6 mice treated with either neutralizing anti-PD-1 antibody or control antibody were evaluated (by flow cytometric analysis) at 2 WOT (top panels) or 4 WOT (bottom panels) for numbers of (A) total lung CD45⁺ leukocytes, and (B) subsets of lung DC and macrophages. For A, B: n=4–9/cohort assayed individually in two separate experiments; *p<0.05 by unpaired Student t test.

Figure 7

Anti-PD-1 antibody treatment does not substantially alter lung myeloid cell activation in the lungs of mice with cryptococcal lung infection. Infected C57BL/6 mice treated with either neutralizing anti-PD-1 or control antibody were evaluated for iNOS and arginase expression by (A) qPCR and (B) flow cytometric analysis using lung leukocytes obtained at 2 WOT (top panels) or 4 WOT (bottom panels). For A, B: n=4–9/cohort assayed individually in two separate experiments; *p<0.05 by unpaired Student t test.

Figure 8

Anti-PD-1 antibody treatment decreases IL-5 and IL-10 gene expression in mice with cryptococcal lung infection. Infected C57BL/6 mice treated with either neutralizing anti-PD-1 or control antibody were evaluated for (A) IL-5 and (B) IL-10 gene expression by qPCR analysis using lung leukocytes obtained at 2 WOT (top panels) or 4 WOT (bottom panels). n=4-9/cohort assayed individually in two separate experiments; *=p<0.05 by unpaired Student t test.

Figure 9

Anti-PD-1 antibody treatment promotes sustained expression of ICOS and OX40 on Th1 and Th17 cells. Infected C57BL/6 mice treated with either neutralizing anti-PD-1 or control antibody were evaluated (by flow cytometric analysis) at 2 WOT (top panels) or 4 WOT (bottom panels) for: (A) total numbers of CD4⁺ T cells and CD4⁺ T cell subsets and their expression of T cell activation markers including (B) ICOS, and (C) OX40. For A-C: n=4–9/cohort assayed individually in two separate experiments; *=p<0.05 by unpaired Student t test.