Pulmonary innate immune remodeling in Pneumocystis pneumonia with adaptive immunodeficiency

Abstract

Background:

Pneumocystis pneumonia (PCP) is a severe fungal infection that predominantly affects individuals with compromised immune systems, specifically those with adaptive immunodeficiency. The mechanisms governing the remodeling of the innate immune system in the context of PCP, particularly in the presence of complete adaptive immunodeficiency, remain insufficiently elucidated. Here, using single-cell RNA sequencing (scRNA-seq) in a murine PCP model, we reveal how innate immune responses are remodeled in the absence of adaptive immunity, providing new mechanistic insights into PCP pathogenesis and a foundation for targeted therapy.

Methods:

An experimental model of PCP was established using male CB-17 SCID mice with adaptive immunodeficiency. Mice were weight-matched into six pairs and randomly divided into a PCP group and a control (CON) group (n = 6 per group) using a computer-generated random sequence; PCP mice received an intranasal instillation of 1 × 10⁶ cysts in 100 μL sterile phosphate-buffered saline (PBS) under isoflurane anesthesia, whereas CON mice received an equal volume of sterile PBS alone. Lung tissues were collected for scRNA-seq to profile pulmonary innate immune cells. Bioinformatic analyses were employed to characterize several subsets of monocyte-derived macrophages, and immunofluorescence and quantitative reverse transcription PCR were used to validate these findings.

Results:

scRNA-seq indicated an expansion of Ccr2⁺ monocyte-derived Nckap5⁺ macrophages in PCP mice, which was associated with anti-inflammatory features and activation of pro-proliferative pathways. Inhibition of the MAPK pathway using Adezmapimod effectively reversed the induction of Nckap5 (4.463 ± 0.125 vs. 0.991 ± 0.052, P < 0.001) and the expression of anti-inflammatory cytokines, including Il10 (44.090 ± 1.246 vs. 4.131 ± 0.151, P < 0.001), Arg1 (2.455 ± 0.135 vs. 0.918 ± 0.018, P < 0.001) and Tgfβ (1.147 ± 0.043 vs. 0.994 ± 0.031, P < 0.01).

Conclusion:

This study advances understanding of PCP pathogenesis and identifies Nckap5⁺ macrophages as a potential therapeutic target for patients with compromised immune systems.

1. Introduction

Pneumocystis, a widespread opportunistic fungal pathogen, primarily results in severe and potentially life-threatening Pneumocystis pneumonia (PCP). With over 400,000 reported cases annually, PCP poses a significant threat to global health.^[1] Hosts with normal immune function can eliminate Pneumocystis asymptomatically, whereas immunocompromised hosts, such as HIV-positive individuals, may develop PCP.^[2,3] This condition is accompanied by an inappropriate and excessive inflammatory response that results in lung injury.^[4]

Previous studies have explored the pathogenesis of PCP in glucocorticoid-induced immunosuppression models (e.g., dexamethasone). In these models, investigations into T cells and B cells during PCP have confirmed that adaptive immunity plays a central role in host defense against Pneumocystis infection.^[5,6] However, glucocorticoid-induced immunosuppression reflects a pharmacologically perturbed immune landscape, characterized by global inhibition of both innate and adaptive responses, which differs fundamentally from the loss of adaptive immunity observed in hosts with T/B cell deficiency.^[7,8]

Hosts with complete adaptive immunodeficiency rely entirely on innate immune mechanisms—primarily pulmonary macrophages, neutrophils, and natural killer (NK) cells—for defense against Pneumocystis.^[9] Macrophages, as central effectors in fungal phagocytosis, antigen presentation, and cytokine secretion, are hypothesized to play essential roles in this context.^[10] Yet, their transcriptional heterogeneity and functional dynamics during PCP in the absence of adaptive immunity remain poorly defined.

Congenitally, SCID mice exhibit a complete deficiency in adaptive immunity while maintaining intact innate immune mechanisms.^[11] Therefore, they are highly susceptible to Pneumocystis infection without requiring chemical immunosuppressive such as glucocorticoids, rendering them an ideal animal model for studying Pneumocystis-associated diseases. In this study, we established a PCP infection model in SCID mice and used scRNA-seq to systematically map the landscape of innate immune cells in the lungs during infection. Subsequent data analysis revealed a correlation between Nckap5⁺ macrophages and anti-inflammatory regulation, suggesting a potential role in the progression of Pneumocystis infection. Overall, our findings elucidate the remodeling mechanisms of innate immune responses in the absence of adaptive immunity, provide novel insights into the pathogenesis of clinical PCP, and offer experimental evidence for the development of targeted therapeutic strategies.

2. Materials and methods

2.1. Ethical approval

All animal procedures were performed in accordance with humane principles, approved by the Animal Care and Use Committee of Capital Medical University (approval number 25-2007), and adhered to ARRIVE 2.0 guidelines for reporting animal research.^[12]

2.2. Mice

Male CB-17 SCID mice, aged 6–8 weeks and weighing 18–22 g, were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. All animals were housed in specific pathogen-free conditions and maintained in a 12-hour light: 12-hour dark cycle at temperatures of 22–24 °C and 30%–70% humidity, with free food and water. Pneumocystis murina (P. murina), sourced from the American Type Culture Collection, Manassas, VA, was propagated in CB-17 SCID mice. Mice were weight-matched into six pairs and randomly divided into a PCP group and a control (CON) group (n = 6 per group) using a computer-generated random sequence. For infection establishment, PCP mice received an intranasal instillation of 1 × 10⁶ cysts in 100 μL sterile phosphate-buffered saline (PBS) under isoflurane anesthesia, whereas CON mice received an equal volume of sterile PBS alone as described.^[13] Meloxicam was administered after instillation for analgesia. At specified time points, the mice were euthanized under deep isoflurane anesthesia. The right upper lobe was fixed in 4% paraformaldehyde for histological assessment by hematoxylin and eosin (H&E) staining and Gomori’s methenamine silver (GMS) staining, and the left lung was reserved for scRNA-seq analysis and quantitative reverse transcription PCR (qRT-PCR).

2.3. Sample collection and tissue processing

The harvested lungs were minced and washed twice with 1 × PBS. Tissue fragments were digested at 37°C in an enzyme mixture containing collagenase and DNase I. The digested product was filtered through a 40 μm strainer (MACS® SmartStrainer, Miltenyi Biotec, Germany) to isolate single cells. Cells were pelleted by centrifugation at 300 × g for 5 minutes. After removing the supernatant, the pellet was resuspended in 1 mL of 1 × PBS with 0.04% BSA. Final cell counts and viability were measured using an automated cell counter.

2.4. scRNA-seq and data analysis

Lung tissues were collected 2 weeks post-infection for scRNA-seq analysis. In both the PCP and CON groups (n = 6 mice per group), two mice were randomly selected using a computer-generated random sequence, and half of the left lung from each mouse was taken and pooled to create one biological replicate. This pooling approach resulted in three independent biological replicates per group, which were subsequently prepared for scRNA-seq.

Single-cell suspensions were processed using the 10x Genomics Chromium X platform to partition individual cells into droplets. Library preparation was performed following the manufacturer’s protocol for the Chromium Single Cell 3′ Reagent Kits v4. Following reverse transcription, cDNA was amplified, and gene expression libraries were constructed. The resulting libraries were pooled and sequenced on an Illumina Novaseq X-25B platform with a paired-end 150 bp configuration. All sequencing procedures were conducted using CapitalBio Technology (Beijing).

Following alignment to the GRCM38 reference genome and gene counting via Cell Ranger (v9.0.1), the resulting UMI (Unique Molecular Identifier) count matrix was demultiplexed using the Seurat R package (v4.2.0). The dataset was first curated by eliminating doublets (DoubletFinder, v2.0.3) and applying quality-control thresholds: Cells were retained if they expressed at least 200 genes and exhibited mitochondrial gene content below 25%, while genes were kept if detected in a minimum of 3 cells. After normalization and scaling, the 2,000 most variable genes were selected for dimensionality reduction via principal component analysis (PCA). A shared nearest-neighbor graph was constructed based on the first 30 principal components, and cell clusters were delineated using a graph-based clustering algorithm at a resolution of 0.6.

2.5. Differentially expressed genes and pathway enrichment

Functional profiling for the top 50 marker genes of each cluster was performed using the clusterProfiler R package (v4.4.4). This analysis included Gene Ontology (GO) and the Kyoto Encyclopedia of Genes and Genomes (KEGG). Significantly enriched terms were identified using the Benjamini–Hochberg correction for multiple testing, and the results were visualized using R-based plotting functions. Gene Set Enrichment Analysis (GSEA) was conducted with GSEA software (v4.2.3) and predefined gene sets from the Molecular Signatures Database (MSigDB v7.5). The analysis used the full complement of genes detected across all cells within a given sample. For the comparison, a gene expression dataset was constructed in which the expression value for each gene was defined as its mean UMI count in a specific cluster versus the mean UMI count across all other clusters combined. Gene set size filters were applied, excluding sets containing fewer than 15 or more than 500 genes. Cell cluster identities were determined by integrating the expression patterns of classical marker genes with evidence-based annotation.

2.6. Single-cell trajectory analysis

To reconstruct cellular differentiation dynamics, pseudotemporal ordering was performed with the Monocle package (v2.4.0). This involved dimensionality reduction via the DDRTree method, followed by trajectory inference using the OrderCells function.

2.7. qRT-PCR

Total RNA was extracted from lung tissues. Pneumocystis burden was quantified using the HiScript® III U⁺ One Step qRT-PCR Probe Kit (Vazyme, Q225) with specific primers and a TaqMan probe referenced from previous studies.^[14] For analysis of inflammatory cytokine and chemokine mRNA expression, RNA was subjected to reverse transcription using the PrimeScript RT reagent kit (Takara, RR037A). qPCR assays were carried out with PowerUp SYBR Green Master Mix (Applied Biosystems, A25742) on an ABI StepOnePlus. Primer and probe sequences are provided in online supplement table S1, http://links.lww.com/IDI/A81.

2.8. Cell culture and experimental treatments

Mononuclear cells were isolated from the hind legs of CB-17 SCID mice and differentiated into macrophages using 25 ng/mL recombinant mouse M-CSF and maintained in Dulbecco’s modified Eagle medium containing 10% fetal bovine serum, 2 mM L-glutamine, and 1% penicillin -streptomycin. Experiments were repeated three times using cells from independent mice (n = 3 independent experiments). During the three-day differentiation period, the cells were treated with or without the MAPK inhibitor Adezmapimod (10 μM) for 24 hours. Following this, cells were exposed to 2.5 P. murina per cell or left unstimulated for an additional three hours. After all treatments were completed, the cells were collected for subsequent qRT-PCR analyses and immunofluorescence staining.

2.9. Immunofluorescence staining

Following treatment, cells were fixed with 4% paraformaldehyde for 20 minutes at room temperature and then permeabilized using 0.3% Triton X-100 for 10 minutes. To prevent nonspecific binding, 5% BSA was applied for 1 hour at room temperature. Cells were incubated overnight at 4°C with anti-Nckap5 (Proteintech, 31199-1-AP, 1:200) and anti-F4/80 (Abcam, ab6640, 1:200) antibodies. Subsequently, Alexa Fluor 488- and 546-conjugated secondary antibodies were applied for 1 hour, followed by Hoechst counterstaining. Images were captured using a ZEISS LSM 900 confocal microscope.

2.10. Statistical analysis

GraphPad Prism 10.0 was used for statistical analysis, and ImageJ software was used for quantification. The qRT-PCR data conformed to the assumptions of normality and were subsequently analyzed for statistical differences using a two-tailed unpaired Student’s t test or one-way analysis of variance (ANOVA) with the Tukey’s multiple-comparison test. Results are expressed as mean ± SD, with significance set at P < 0.05.

3. Results

3.1. Fungal burden and lung pathology in Pneumocystis-infected SCID mice

To elucidate the pulmonary immune landscape during Pneumocystis infection in hosts with adaptive immunodeficiency, we developed a murine model of PCP (Figure 1A). Lung tissues were collected from both the PCP group and the CON group two weeks post-infection. qRT-PCR demonstrated a mean quantification cycle (Cq) value of 26, confirming the presence of Pneumocystis in the mice of PCP group. In contrast, no pathogen was detected in the CON group. These results were further substantiated by GMS staining (Figure 1B, C). H&E staining revealed no significant increase in overall inflammatory responses; however, a distinct pathological feature—alveolar septal thickening—was observed (Figure 1D). Further qRT-PCR analysis of inflammatory mediators, including Tnfα and Il6, indicated no statistically significant differences in transcript levels between CON and PCP groups (Figure 1E).

Figure 1.

Evaluating fungal burden and lung pathology in Pneumocystis-infected SCID mice. (A) Diagrammatic representation of the experimental timeline and design. CB-17 SCID mice were administered with PBS or Pneumocystis cysts and evaluated after two weeks using scRNA-seq, H&E staining, GMS staining and qRT-PCR. (B) qRT-PCR analysis at the two-week time point indicated the presence of Pneumocystis cysts in infected mice, with no pathogen burden detected (ND) in CON mice (n = 3). (C) GMS staining of lung sections. Representative cysts are circled in black; no cysts were observed in CON mice. Magnification, 280×. Scale bar, 10 μm. (D) H&E staining at two weeks post-infection highlighted the pathological features in both CON and PCP mice. Magnification, 25×. Scale bar, 100 μm. (E) qRT-PCR analysis of Il6 and Tnfα in CON and PCP mice. qRT-PCR, quantitative reverse transcription PCR; NS, not significant (two tailed Student’s t test). Data are shown as mean ± SD of n = 6 mice. CON, Control; PCP, Pneumocystis pneumonia; PBS, Phosphate-buffered saline; Cq value, Quantification cycle value; H&E, Hematoxylin and eosin; GMS, Gomori’s methenamine silver.

To evaluate the progression of infection and its associated immunopathology, we extended our analysis to a later time point. Lung tissues from control and PCP mice were collected at 6 weeks post-infection for comparative histopathological assessment. GMS staining identified numerous P. murina cysts, appearing as dark circular structures, within the alveolar spaces of PCP mice, confirming persistent infection at this chronic stage. No fungal elements were observed in the control lungs (Figure S1A, http://links.lww.com/IDI/A81). H&E staining demonstrated significant and sustained pulmonary pathology in PCP mice at 6 weeks, whereas the control lungs exhibited normal architecture (Figure S1B, http://links.lww.com/IDI/A81).

3.2. Characterization of the pulmonary single-cell transcriptome identifies myeloid cells as the major expanded immune subset

To elucidate the cellular composition and immune landscape associated with the distinct early-phase pathological phenotype of PCP, we conducted scRNA-seq on lung tissues obtained from both infected and control mice at the two-week post-infection time point. Following quality control, filtering, and correction for batch effects, a total of 123,441 cells were retained for further analysis. Through PCA, dimensionality reduction, and clustering, we identified seven major cell clusters encompassing both immune and non-immune cell types (Figure 2A). Each cluster was annotated based on the expression profiles of classical cell type-specific marker genes, categorizing the cells into myeloid cells (Lyz2, Mrc1, Mertk, Adgre1, Itgax), neutrophils (S100a9, S100a8, Csf3r, Retnlg), NK cells (Nkg7, Gzma, Klrb1c), innate lymphoid cells (ILCs) (Il1rl1, Gata3, Il7r), fibroblasts (Col1a2, Mgp, Col3a1, Col1a1, Pdgfra, Gsn, Itga8), endothelial cells (Cdh5, Cyyr1, Adgrf5, Calcrl, Cldn5, Ptprb), and epithelial cells (Sftpc, Sftpa1, Ager, Tppp3, Hopx, Sftpb) (Figures 2B and S2A, http://links.lww.com/IDI/A81). Compared with the CON group, the PCP group exhibited altered proportions of various immune cell lineages within lung tissues, notably myeloid cells and neutrophils (Figures 2C, D and S2B, http://links.lww.com/IDI/A81). A more detailed analysis of immune cell subset compositions between the two groups indicated that myeloid cells demonstrated the most pronounced increase in the PCP group (Figures 2C, D and S2B, http://links.lww.com/IDI/A81), suggesting a potentially essential role for myeloid cells in mitigating Pneumocystis infection. Accordingly, subsequent analyses concentrated on myeloid cells to identify functionally significant subpopulations.

Figure 2.

Characterization of the pulmonary single-cell transcriptome identifies myeloid cells as the major expanded immune subset. (A) UMAP visualization shows the spatial distribution of single cells under CON and PCP conditions. (B) Dot plot depicting the average expression of highly expressed genes across different subtypes, with data color-coded according to expression intensity. (C) The percentage distribution of major lung cell populations between CON and PCP mice. (D) Profiling of pulmonary immune cells. Comparative analysis of lung cell populations from CON and PCP mice, showing a relative increase in myeloid cells following infection. For each group (n = 6), half of the left lung from two mice was pooled to produce one biological replicate (n = 3 replicates per group). CON, Control; PCP, Pneumocystis pneumonia.

3.3. Macrophages are the predominantly upregulated myeloid subset in Pneumocystis infection

To characterize the immune profile occurring within infected tissues, we conducted an in-depth investigation into the composition and heterogeneity of myeloid cells in infected lungs utilizing scRNA-seq. Following isolation of myeloid cells, we employed PCA, dimensionality reduction, clustering, and subpopulation analysis. Based on marker gene expression, myeloid cells were categorized into seven distinct subpopulations: macrophages, Mmp12⁺ inflammatory dendritic cells (infDC), Xcr1⁺ conventional dendritic cells (cDC), Ly6c2⁺ infDC, Fscn1⁺ cDC, Ccr2⁺ monocytes, and alveolar macrophages (Figure 3A, B). Comparative analysis with a healthy control group revealed that Pneumocystis infection significantly altered the composition of pulmonary myeloid cells. Further examination of immune subset proportions between the two groups demonstrated that macrophages exhibited the most pronounced increase in the PCP group (Figure 3C, D), suggesting a potentially central role for macrophages in controlling Pneumocystis infection.

Figure 3.

Macrophages represent the predominantly upregulated myeloid subset in Pneumocystis infection. (A) UMAP projection illustrating myeloid cell subtypes in the lungs of CON and PCP mice. (B) Dot plot of canonical marker expression. Color intensity reflects the magnitude of gene expression within each cell population. (C) UMAP visualization of lung myeloid cells from CON and PCP mice. Distinct spatial distributions highlight infection-induced alterations in myeloid cell composition and transcriptional states. (D) The stacked bar plot illustrates the relative abundance of myeloid cell subsets in CON and PCP mice, highlighting the predominant contribution of macrophages. CON, Control; PCP, Pneumocystis pneumonia.

3.4. Nckap5⁺ macrophages display an immunosuppressive phenotype distinct from C1qb⁺ macrophages

Using scRNA-seq, we investigated macrophage heterogeneity in PCP. Based on marker gene expression, macrophages were categorized into three transcriptionally distinct subpopulations: C1qb⁺ macrophages, Nckap5⁺ macrophages, and Pclaf⁺ macrophages (Figures 4A, B and S3A, http://links.lww.com/IDI/A81). Comparative analysis of the proportions of these macrophage subsets between the CON and PCP groups revealed a significant increase in Nckap5⁺ macrophages in the PCP group (CON 34%, PCP 73%) (Figures 4C and S3C, http://links.lww.com/IDI/A81), suggesting a potentially unique role for this subpopulation in the host response to Pneumocystis infection. To further elucidate the functional characteristics underlying the expansion associated with infection, we conducted a comparative transcriptomic analysis of Nckap5⁺ and C1qb⁺ macrophages to delineate their pathway-level distinctions. KEGG pathway analysis indicated a significant downregulation of important pro-inflammatory pathways in Nckap5⁺ macrophages relative to C1qb⁺ macrophages. These pathways included the complement and coagulation cascades, chemokine signaling pathway, and cytokine–cytokine receptor interaction (Figure 4D). Furthermore, GO enrichment analysis revealed that Nckap5⁺ macrophages showed a broad suppression of immune-related biological processes compared with C1qb⁺ macrophages. Notably, downregulated terms included humoral immune response, complement activation, eosinophil chemotaxis, and chemokine-mediated signaling pathway (Figure S3D, http://links.lww.com/IDI/A81). Transcriptional profiling further characterized Nckap5⁺ macrophages by marked suppression of cytokines and chemokines important for initiating and sustaining immune responses, including the pro‑inflammatory cytokine Tnf, neutrophil chemoattractants (Cxcl1, Cxcl2, Cxcl5), monocyte‑recruiting chemokines (Ccl2, Ccl7), NK‑cell‑recruiting ligands (Cxcl10, Ccl3, Ccl4), and interferon receptor Ifnar1 (Figure 4E, F). These findings reveal that Nckap5⁺ macrophages display an anti-inflammatory transcriptional profile with impaired immune activation during PCP.

Figure 4.

Nckap5⁺ macrophages display an immunosuppressive phenotype that is distinct from C1qb⁺ macrophages. (A) UMAP projection delineates distinct macrophage subsets within the lungs of CON and PCP groups. (B) Dot plot illustrates the average expression levels of highly expressed genes for each macrophage subtype. Data are visually represented according to expression intensity. (C) Bar chart provides a comparative analysis of the percentage distribution of macrophage populations between CON and PCP mice. (D) KEGG pathway enrichment analysis contrasts differentially expressed genes between the Nckap5⁺ and C1qb⁺ macrophage clusters. (E) Volcano plot illustrates differential expression analysis of inflammatory factors between Nckap5⁺ and C1qb⁺ macrophages. (F) Expression levels of inflammatory factor genes in Nckap5⁺ versus C1qb⁺ macrophages. CON, Control; PCP, Pneumocystis pneumonia.

3.5. Expansion of Nckap5⁺ macrophages via MAPK signaling mediates immunosuppression in Pneumocystis infection

To elucidate the regulatory mechanisms governing the development and functional specification of this immunosuppressive macrophage subset, we employed pseudotime trajectory analysis. This analysis revealed a distinct differentiation continuum originating from Ccr2⁺ monocytes and culminating in Nckap5⁺ macrophages (Figure 5A, B). To explain the functional relevance of this infection-associated macrophage subset, we performed an in-depth analysis of the transcriptional profile and biological functions of Nckap5⁺ macrophages. KEGG analysis revealed significant downregulation of key pro-inflammatory pathways, including cytokine–cytokine receptor interaction, IL-17 signaling, chemokine signaling, and C-type lectin receptor signaling (Figure 5C). Concurrently, we observed marked upregulation of pro-proliferative MAPK signaling pathway in the PCP group (Figure 5D). This coordinated shift suggests a transition toward an immunosuppressive state in Nckap5⁺ macrophages. To validate the involvement of MAPK signaling in this process, cells were stimulated by P. murina and treated with or without Adezmapimod, a selective MAPK inhibitor. Immunofluorescence staining demonstrated that the infection-induced upregulation of Nckap5 in lung macrophages was abolished by MAPK inhibition, with quantitative fluorescence intensity analysis corroborating this outcome (9.430 ± 0.479 vs. 6.530 ± 0.553, P < 0.01) (Figure 5E). qRT-PCR analysis further showed that Nckap5 mRNA levels were significantly elevated in mice infected with Pneumocystis (1.000 ± 0.032 vs. 4.463 ± 0.125, P < 0.001), and this induction was completely reversed by treatment with Adezmapimod (4.463 ± 0.125 vs. 0.991 ± 0.052, P < 0.001) (Figure 5F). Simultaneously, MAPK inhibition significantly reduced the expression of the anti-inflammatory cytokines Il10 (44.090 ± 1.246 vs. 4.131 ± 0.151, P < 0.001), Arg1 (2.455 ± 0.135 vs. 0.918 ± 0.018, P < 0.001) and Tgfβ (1.147 ± 0.043 vs. 0.994 ± 0.031, P < 0.01) (Figure 5G). Collectively, these findings indicate that MAPK signaling is essential for the induction of Nckap5 expression and the associated immunosuppressive program in Nckap5⁺ macrophages during PCP. This pathogenic reprogramming undermines the host’s innate antifungal defenses and facilitates persistent fungal colonization, pointing to Nckap5⁺ macrophages as a primary driver of PCP progression in the context of adaptive immunodeficiency and as a potential therapeutic target for restoring anti-infective immunity.

Figure 5.

Expansion of Nckap5⁺ macrophages via MAPK signaling mediates immunosuppression in Pneumocystis infection. (A, B) Pseudotime trajectory analysis of Nckap5⁺ macrophage differentiation. The differentiation trajectory of Nckap5⁺ macrophages was conducted using Monocle2, incorporating all monocytes and macrophages isolated from the PCP group. Cells are colored by pseudotime in panel (A) and by cell type in panel (B). (C) Enriched KEGG pathways downregulated in Nckap5⁺ macrophages. (D) Enriched KEGG pathways upregulated in Nckap5⁺ macrophages. (E) Macrophages were subjected to a 24-hour pretreatment with or without the MAPK inhibitor Adezmapimod (10 μM), followed by a 3-hour exposure to Pneumocystis or maintenance in an unstimulated state. Subsequently, the cells were collected for qRT-PCR analysis and immunofluorescence staining. Representative immunofluorescence images of Nckap5 (red) and F4/80 (green) in macrophages. Scale bar, 10 μm. The bar graph shows quantification of mean fluorescence intensity. (F) qRT-PCR analysis of relative Nckap5 mRNA expression in the indicated groups. (G) Relative mRNA expression of anti-inflammatory cytokines Il10, Arg1, and Tgfβ in the indicated groups. ^**P < 0.01, ^***P < 0.001 (one-way ANOVA with the Tukey’s multiple-comparison test). Data are shown as mean ± SD of n = 3 independent experiments. CON, Control; PCP, Pneumocystis pneumonia; KEGG, Kyoto Encyclopedia of Genes and Genomes.

4. Discussion

Myeloid cells are essential components of tissue leukocytes, comprising a variety of cell types, including granulocytes, monocytes, dendritic cells (DCs), and macrophages, each performing distinct yet complementary functions during both homeostasis and disease.^[15] Among them, macrophages serve as primary components of innate immunity and pulmonary homeostasis, with their population diversity increasing considerably during pulmonary infections and other inflammatory conditions.^[16] Despite advances essential in the field, the precise immunological functions of myeloid subsets during PCP remain insufficiently characterized.

In this study, a PCP model was developed using congenitally SCID mice, and scRNA-seq was employed to characterize the remodeling of pulmonary innate immunity, with a particular focus on the functional roles of macrophage subsets. The findings revealed substantial alterations in the pulmonary innate immune cell landscape in PCP-infected SCID mice. Among the seven identified myeloid cell subpopulations, macrophages exhibited the most pronounced expansion, with Nckap5⁺ macrophages—originating from Ccr2⁺ monocytes—emerging as the predominantly enriched subpopulation. In contrast to C1qb⁺ pro-inflammatory macrophages, which also originate from Ccr2⁺ monocytes, Nckap5⁺ macrophages exhibited significantly reduced expression of cytokines and chemokines essential for inflammatory responses, including the pro-inflammatory cytokine Tnf,^[17] neutrophil chemoattractants (Cxcl1, Cxcl2, Cxcl5),^[18] monocyte-recruiting chemokines (Ccl2, Ccl7),^[19] and NK cell-recruiting ligands (Cxcl10, Ccl3, Ccl4).^[20] Transcriptomic analysis further indicated that Nckap5⁺ macrophages possess a dual functional phenotype, characterized by the downregulation of pro-inflammatory pathways, including cytokine−cytokine receptor interaction,^[21] IL-17 signaling,^[22] chemokine signaling,^[23] and C-type lectin receptor signaling,^[24] alongside marked upregulation of pro-proliferative pathways, such as MAPK, PI3K-Akt, Ras, and VEGF, facilitating cellular expansion.^[25,26,27] To verify the regulatory mechanism, MAPK inhibitor intervention assays were conducted. The findings revealed that the upregulation of Nckap5 induced by Pneumocystis infection was completely abolished by Adezmapimod. Furthermore, infection significantly increased the mRNA levels of the anti-inflammatory mediators Arg1, Tgfβ, and Il10, and these elevations were substantially reduced following Adezmapimod treatment. Collectively, these data indicate that Pneumocystis infection promotes the differentiation of Ccr2⁺ monocytes into a distinct Nckap5⁺ macrophage subset with reduced pro-inflammatory activity, a process regulated by the MAPK pathway. These results identify a new macrophage subpopulation implicated in the pathogenesis of PCP and suggest a potential therapeutic target for combating Pneumocystis infection.

Compared with our group’s previous research utilizing a glucocorticoid-induced immunosuppression model—which demonstrated that glucocorticoids inhibit the differentiation of Mmp12⁺ macrophages (a protective subset) and suppress pro-inflammatory cytokine secretion, thereby impairing fungal clearance^[28]—the current study reveals that the aberrant enrichment and functional reprogramming of Nckap5⁺ macrophages contribute to the progression of PCP. Together, these findings indicate that various immunodeficiency states converge on the dysregulation of macrophage subsets, exacerbating PCP and offering a fuller understanding of PCP pathogenesis across different immunocompromised conditions.

However, this study is limited by its evaluation at a single time point—specifically two weeks post-infection—and by the absence of comprehensive multiomics analyses, including proteomics and metabolomics. Translational applicability is further constrained by the use of SCID mice, which have congenital athymia and severe global T-cell deficiency, resulting in significant impairment of adaptive immunity. This contrasts with the majority of clinical PCP cases, which typically occur in individuals with partial or selective immunosuppression and are primarily characterized by reduced CD4⁺ T cell counts.^[3] Consequently, this model does not fully replicate the immune status observed in most human PCP patients, and caution is warranted when extrapolating these findings to clinical scenarios.

In summary, this study identifies Nckap5⁺ macrophages as a potentially critical cell subset in PCP progression. The expansion and inflammation-suppressive features of these cells may play a role in facilitating Pneumocystis proliferation. These findings raise the possibility that Nckap5⁺ macrophages could be explored as a therapeutic target. Strategies that focus on modulating their proliferation or pro-inflammatory capacity might offer avenues for improving antifungal immunity and outcomes in PCP patients with adaptive immunodeficiency.

Author Contributions

Lin Gu and Wan Peng conceived and designed the study, performed the experimental work, analyzed the data, and drafted the initial manuscript. Hanyujie Kang and Yuling Xu participated in experimental implementation and assisted with relevant data analysis. Jieqiong Li participated in data collection and assisted with data sorting and analysis. Zhaohui Tong and Nan Song supervised the study, provided critical scientific guidance, and revised the manuscript for important intellectual content. All authors contributed to data interpretation, manuscript revision, and approved the final submitted version.

Conflicts of Interest

None.

Data Availability Statement

The datasets analyzed during the current study are available from the corresponding authors on reasonable request.

Funding Statement

This work was supported by the National Natural Science Foundation of China (82470059, 82270009), Beijing Chao-Yang Hospital Golden Seeds Foundation (CYJZ202443).