Notch Signaling Exacerbates Pulmonary Fibrosis by Regulating the Differentiation of CD4+ Tissue-Resident Memory T Cells

TL;DR

CD4+ tissue-resident memory T cells worsen lung fibrosis by promoting inflammation, and their differentiation is regulated by the Notch signaling pathway.

Contribution

Identifies CD4+ TRM cells as pathogenic drivers in pulmonary fibrosis and reveals Notch signaling as a key regulator of their differentiation.

Findings

CD4+ TRM cells are present in higher numbers in fibrotic lungs and correlate with disease severity.

Depletion of CD4+ TRM cells reduces fibrosis in mouse models.

Notch signaling inhibition suppresses pro-fibrotic functions of CD4+ TRM cells.

Abstract

The involvement of the immune system in pulmonary fibrosis is established, the precise contributions of tissue-resident memory T (TRM) cells are still poorly defined. This study sought to define the contribution of CD4+ TRM cells to pulmonary fibrosis, their origin, and regulatory mechanisms. We combined bioinformatic analysis of human fibrotic lung single-cell RNA-sequencing data with experiments in a bleomycin-induced C57BL/6 mouse model. Flow cytometry, targeted in vivo depletion, lymphocyte trafficking blockade, cell co-culture, and pharmacological inhibition were employed. CD4+ TRM cells were observed at higher frequencies within fibrotic lung tissue. Their presence correlated with disease severity, and they were found to exhibit a pro-inflammatory and pro-fibrotic phenotype. Their specific depletion alleviated fibrosis. These cells primarily originated from recruited circulating lymphocytes, as blocking this recruitment reduced TRM accumulation and attenuated disease. Furthermore, the Notch signaling pathway was activated in fibrotic lung CD4+ TRM cells, and its inhibition suppressed their differentiation and impaired their pro-fibrotic function. We conclude that CD4+ TRM cells are pathogenic drivers in pulmonary fibrosis, originating from circulating precursors and being regulated by Notch signaling, underscoring their relevance for therapeutic intervention.

Figures

Full text

1. Introduction

Interstitial lung disease (ILD) encompasses a spectrum of disorders. Pathologically, they are defined by the dual processes of alveolar inflammation and interstitial fibrosis [1]. Idiopathic pulmonary fibrosis (IPF) represents the most common diagnosis among ILDs, with connective tissue disease-associated ILD (CTD-ILD) and hypersensitivity pneumonitis (HP) also frequently encountered [2]. Pulmonary fibrosis (PF) represents a common end-stage pathological manifestation across various ILD subtypes. Despite standard treatment, some patients may experience disease progression, eventually leading to PF, manifested by worsening respiratory symptoms, a progressive deterioration of pulmonary function, impaired daily functioning, and even death [3,4]. The pathogenesis of PF remains incompletely elucidated, and no effective therapies are currently available to halt or reverse the ensuing pulmonary dysfunction [5,6].

In the pathogenesis of PF, the immune system is of central importance. Emerging evidence points to the critical involvement of T cells in this process, with different subsets exerting distinct functions [7,8]. Early research primarily focused on elucidating the complex functions of CD4^+^ helper T cells in balancing pro- and anti-fibrotic responses, with particular attention to subsets such as Th1, Th2, Th17, and regulatory T (Treg) cells [7,9,10,11,12]. Beyond these classical helper T cell subsets, memory T cell populations are increasingly recognized for sustaining chronic inflammation and tissue remodeling [13,14,15]. Effector memory T cells directly drive both the establishment and perpetuation of inflammatory and fibrotic microenvironments in the lung [16]. The identification and characterization of tissue-resident memory T (T_RM_) cells has brought about a major conceptual change in how local immunity is understood [17,18]. These cells have thus emerged as key targets for understanding the immunopathological mechanisms underlying PF.

While conventional memory T cells undergo systemic recirculation, T_RM_ cells exhibit long-term tissue retention and lack involvement in the general circulatory pathway [13]. They commonly display the expression of CD69, CD103, CD49a, the chemokine receptors CXCR3 and CXCR6, and the transcription factors Blimp-1 and Hobit while lacking expression of CD62L and CCR7 [14,15]. T_RM_ cells can be rapidly activated locally without the need for recruitment from the periphery, thereby providing a first line of defense against mucosal infections [19]. Although T_RM_ cells were initially recognized for their protective role in combating pathogen infections [20], more recently, a growing body of research has supported their potential involvement in mediating pathological processes in disease [21,22]. At present, the exact manner in which T_RM_ cells drive the pathogenesis and progression of PF is still poorly understood.

In the present investigation, both CD4^+^ and CD8^+^ T_RM_ cell populations were detected in lung tissues of individuals diagnosed with PF. Utilizing a murine model of pulmonary fibrosis elicited by bleomycin (BLM), we observed a positive association between the abundance of CD4^+^ T_RM_ cells and the severity of disease. Furthermore, this subset exhibited pro-inflammatory and pro-fibrotic functions during disease progression. Pathway analysis at the mechanistic level showed that CD4^+^ T_RM_ cells within fibrotic pulmonary tissues display active Notch signaling. Blockade of the Notch signaling led to a significant decrease in CD4^+^ T_RM_ cell burden and effectively mitigated pulmonary fibrosis.

2. Materials and Methods

2.1. Mice

Six- to eight-week-old male C57BL/6J mice were utilized for all experiments in this study. These mice were acquired from GemPharmatech Co., Ltd. (Jiangsu, China) and maintained in an SPF environment at the Animal Experiment Center of Sun Yat-sen University North Campus. Mice were maintained under standardized environmental conditions, with a temperature of 20–24 °C, relative humidity of 40–50%, and a 12 h light/dark photoperiod, to guarantee stable physiological homeostasis. All animal experiments in this study were carried out in strict accordance with relevant guidelines for laboratory animal care and were approved by the Institutional Ethics Committee of Sun Yat-sen University.

2.2. BLM-Induced PF

The PF model induced by bleomycin (BLM) was established using procedures adapted from previously reported methodologies [23]. Briefly, mice were anesthetized using 1% sodium pentobarbital administered intraperitoneally at a dose of 50 mg/kg. A midline cervical incision was made to expose the trachea, followed by intratracheal injection of BLM (MCE, Monmouth Junction, NJ, USA, Cat#: HY-17565A, 2 mg/kg) or an equal volume of PBS as a control.

To investigate the origin of lung T_RM_ cells in PF, FTY720 (MCE, Monmouth Junction, NJ, USA, Cat#: HY-12005, 1 mg/kg) or vehicle was administered via intraperitoneal injection starting three days before BLM exposure and continued daily thereafter.

In order to evaluate the contribution of T_RM_ cells to PF, NAD^+^ (MCE, Monmouth Junction, NJ, USA, Cat#: HY-B0445, 60 mg/mouse) or PBS was delivered through tail vein injection beginning one day prior to BLM administration, with a second dose administered on day 7.

To examine the regulatory effect of Notch signaling on T_RM_ cells, DAPT (MCE, Monmouth Junction, NJ, USA, Cat#: HY-13027, 10 mg/kg) or vehicle was injected intraperitoneally starting one hour before BLM treatment and continued once daily.

2.3. Histopathological Analysis

Paraffin-embedded sections of 5 μm in thickness were prepared from dehydrated left lung tissue samples. The paraffin-embedded sections were subsequently deparaffinized and rehydrated for further experimental procedures. Hematoxylin and eosin (H&E) staining was performed on the lung sections, and Ashcroft scoring was conducted according to established criteria to evaluate the degree of fibrosis. Furthermore, Masson’s trichrome staining was applied to assess collagen deposition [24]. Masson’s trichrome staining was additionally performed to evaluate the extent of collagen deposition in lung tissues. For immunohistochemical analysis, the sections first underwent antigen retrieval. Then, endogenous peroxidase activity was blocked by incubation with 3% hydrogen peroxide at room temperature for 1 h, followed by blocking with 5% BSA for 1 h. Overnight incubation at 4 °C was performed on tissue sections with primary antibody against α-smooth muscle actin (α-SMA) obtained from Abcam (Cambridge, UK, Cat# ab5694, 1:400). After being rinsed with PBS the next day, the sections underwent incubation with HRP-conjugated anti-rabbit secondary antibody (Servicebio, Wuhan, China, Cat# GB23303, 1:200) at room temperature for 60 min, and then subjected to DAB staining. Image acquisition and analysis were conducted using the ImageJ software package (v1.53t, National Institutes of Health, Bethesda, MD, USA).

2.4. Cell Isolation and Culture

For the purification of murine CD4^+^ T cells, a single-cell suspension was prepared from mouse spleens using a standardized protocol. Briefly, splenic tissue was minced and gently homogenized with a syringe plunger, then passed through a 70 μm cell strainer to obtain a single-cell suspension. CD4^+^ T cells were isolated via magnetic bead-based negative selection (STEMCELL Technologies, Vancouver, BC, Canada, Cat# 19852 and 19765) and subsequently cultured in complete Advanced RPMI 1640 medium (Thermo Fisher Scientific, Waltham, MA, USA, Cat# 12633012) supplemented with 10% FBS (Procell, Wuhan, China, Cat# 164210), 2 mM L-glutamine (MCE, Monmouth Junction, NJ, USA, Cat# HY-N0390), 55 mM 2-mercaptoethanol (β-ME, Gibco, Grand Island, NE, USA, Cat# 21985023), and 1% penicillin–streptomycin (Gibco, Grand Island, NE, USA, Cat# 15140122). To achieve CD4^+^ T cell activation, cultures were exposed to 5 μg/mL anti-CD3 (BioLegend, San Diego, CA, USA, Cat# 100359) and 2 μg/mL anti-CD28 (BioLegend, San Diego, CA, USA, Cat# 102121). CD4^+^ T_RM_ cell differentiation was induced by culturing activated cells in complete medium with 10 ng/mL TGF-β (Sino Biological, Beijing, China, Cat# 80116-RNAH-5) over a 3-day period, following previously reported methods [25].

As previously described, primary mouse lung fibroblasts were isolated [26]. Briefly, mice were deeply anesthetized and sterilized with alcohol. After surgical opening of the thoracic cavity, PBS was used to perfuse the lungs, which were then excised for subsequent analysis. To remove blood components, isolated lung specimens were washed with PBS, fragmented into small pieces, and cultured in 10 cm dishes with DMEM complete medium containing 10% FBS and 1% penicillin–streptomycin over 6–7 days. Finally, the adherent fibroblast layer was digested using trypsin, with subsequent filtration to remove residual tissue pieces, followed by routine cell passaging. Cells between passages 2 and 4 were used for further analysis. For co-culture with CD4^+^ T cells, fibroblasts were preincubated with BLM and then pre-induced CD4^+^ T_RM_ cells in the Trans-well co-culture systems for 24 h. To inhibit Notch signaling, 10 μM DAPT was added to the co-culture system.

2.5. Lung Coefficient

To assess pulmonary edema, the lung coefficient was determined as the ratio of lung weight (g) to body weight (kg) [27].

2.6. Lung Single Cell Isolation

Mice were euthanized by anesthetic overdose. The thoracic cavity was opened, and lung tissues were collected following perfusion with a large volume (typically 20–30 mL) of cold PBS until the lungs appeared visually cleared of blood (turning uniformly pale). For each mouse, two lung lobes were dissected and minced into 1 mm^3^ pieces. The harvested lung tissues were resuspended in PBS solution, followed by mechanical disruption with a GentleMACS tissue dissociator [28]. Finally, the cell suspension was filtered through a 70 μm cell strainer (BD, Franklin Lakes, NJ, USA), and single lung cells were obtained by centrifugation at 1500 rpm for 5 min.

2.7. Real-Time PCR

Total RNA was isolated with TRIzol reagent (Invitrogen, Carlsbad, CA, USA, Cat# 15596018), and cDNA was generated by reverse transcription with the Evo M-MLV RT Premix kit (Accurate Biology, Guangzhou, China, Cat#: AG11706). Quantitative real-time PCR (qPCR) was performed using SYBR Green chemistry (Accurate Biotechnology, Changsha, China, Cat# AG11718), with detailed thermal cycling parameters outlined below: after an initial denaturation at 95 °C for 30 s, 35 cycles were performed (95 °C for 5 s, 60 °C for 30 s), and the program concluded with a final elongation stage. For normalization of transcript levels, Gapdh was used as the housekeeping gene; the corresponding qPCR primers are summarized in Table S1.

2.8. Flow Cytometry

For cell surface staining, cells were stained with antibodies against mouse CD45, CD3, CD4, CD8, CD69, and CD103 and incubated at 4 °C in the dark for 30 min. For intracellular cytokine staining, cells were treated with a mixture containing 50 ng/mL PMA (Sigma, St. Louis, MO, USA, Cat# P8139), 500 ng/mL ionomycin (Sigma, St. Louis, MO, USA, Cat# I0634), and 5 μg/mL brefeldin A (Sigma, St. Louis, MO, USA, Cat# B7651) at 37 °C for 4.5 h. After surface staining, cells were fixed and permeabilized using a fixation/permeabilization solution (BD Biosciences, San Jose, CA, USA, Cat# 554722), followed by incubation with antibodies against mouse IL-17A and Granzyme B (GZMB) at 4 °C in the dark for 30 min. Finally, the cell suspension was analyzed by flow cytometry (LSR Fortessa, BD Biosciences, San Jose, CA, USA), and the cytometry data were analyzed using FlowJo software (v10.8.1). The antibodies used are listed in Table S2.

2.9. Single-Cell RNA Sequencing (scRNA-seq)

The publicly available scRNA-seq dataset of human PF used in this study was obtained from the Gene Expression Omnibus (GSE122960). This dataset comprises scRNA-seq data from 16 samples, corresponding to lung tissue from 8 organ donors without interstitial lung disease or lung injury (labeled as “Donor”), 4 patients with IPF, 1 patient with hypersensitivity pneumonitis (HP), and 3 patients with CTD-ILD. The baseline data were supplemented in Table S3 according to the characteristics shown in the previous study [29]. High-quality cells were filtered based on the following criteria: (i) the number of expressed genes between 200 and 3000; (ii) total transcripts >800; and (iii) mitochondrial gene percentage <20%. The Seurat v4 package was utilized for all data analysis procedures. Briefly, highly variable genes were characterized on the basis of normalized transcriptomic data. Principal component analysis was conducted, and the top 20 principal components were selected for cell clustering. Nonlinear dimensionality reduction methods, such as t-SNE or UMAP, were employed for visualization of the clustering results.

2.10. Statistics

Statistical analysis of the scRNA-seq data was performed using R (version 4.1.0). Additional statistical analyses were conducted with GraphPad Prism 8.0. Statistical significance of differences between the two experimental groups was determined via unpaired Student’s t-test. When comparing more than two groups, statistical analysis was performed using one-way ANOVA accompanied by appropriate post hoc multiple comparison tests. For correlation assessments, Spearman’s correlation coefficient was utilized as applicable. All data are expressed as the mean ± standard error of the mean (SEM). A two-tailed p-value of less than 0.05 was regarded as statistically significant.

3. Results

3.1. Increased Proportions of CD4+ TRM Cells in Fibrotic Lungs Correlate with Disease Severity

Our analysis commenced with the use of publicly shared scRNA-seq data obtained from human fibrotic lung tissue samples. Subsequent to quality filtering, 19 clusters were defined and subsequently assigned to 14 primary cell populations on the basis of well-established marker gene expression in lung tissue (Figure S1A–C). The NK/T cells population were extracted, processed with Seurat, and classified into CD4^+^ T cells, CD8^+^ T cells, NK cells, and an undefined population (Figure S1D,E).

To determine the presence of T_RM_ cells in PF, we subclustered CD4^+^ T cells and subdivided them into six distinct subsets (Figure 1A,B). Among these, Cluster 1 uniquely exhibited a phenotype consistent with CD4^+^ T_RM_ cells. This identification was supported by two complementary visualizations: a dot plot demonstrating this cluster’s high co-expression of canonical T_RM_ markers (CD69, CD49a/ITGA1, CD103/ITGAE, CXCR3, CXCR6) and low expression of the egress marker CD62L/SELL (Figure 1C). Feature plots confirmed the specific spatial localization of these key markers within the Cluster 1 region of the t-SNE (Figure 1D). Comparative analysis demonstrated a significantly increased fraction of T_RM_ cells within CD4^+^ T cells in fibrotic lung tissues from PF patients vs. donors (Figure 1E). Applying the same strategy to CD8^+^ T cells (Figure S1F,G), we identified cluster 4 as CD8^+^ T_RM_ cells based on high expression of CD69, CD49a, CD103, and Hobit (ZNF683) (Figure S1H,I). We also observed a higher proportion of CD8^+^ T_RM_ cells within the lungs of PF patients (Figure S1J).

We next validated these findings in the BLM-induced mouse model of PF using flow cytometry. Both the frequency and absolute number of the lung CD4^+^CD69^+^CD103^+^ and CD8^+^CD69^+^CD103^+ ^T_RM_ cells were significantly increased in BLM-treated mice compared to controls (Figure 2A–F), corroborating the human data and suggesting a potential role for T_RM_ cells in PF pathogenesis. Furthermore, in BLM-induced mice, the number of CD4^+^ T_RM_ cells in lung tissues showed a significant positive correlation with fibrosis severity metrics, including Ashcroft score and collagen volume fraction (Figure 2G,H). In contrast, no such correlation was observed for lung CD8^+^ T_RM_ cells (Figure 2I,J), implicating a more critical role for CD4^+^ T_RM_ cells in PF progression.

3.2. Enhanced Pro-Inflammatory and Pro-Fibrotic Function of Lung CD4+ TRM Cells in PF

In order to define the functional status of pulmonary CD4^+^ T_RM_ cells in PF, gene set enrichment analysis (GSEA) was performed between CD4^+^ T_RM_ and CD4^+^ non-T_RM_ cells, as identified in Figure 1. Marked enrichment of gene sets involved in T cell activation and cytokine secretion was observed in the lung CD4^+^ T_RM_ subset (Figure 3A). As determined by flow cytometry, the expression levels of IL-17A and GZMB were significantly elevated in CD4^+^CD69^+^CD103^+^ T_RM_ cells in comparison with CD4^+^CD69^−^CD103^−^ and CD4^+^CD69^+^CD103^−^ T cell populations (Figure 3B–E). Moreover, the expression of both IL-17A and GZMB was further elevated in CD4^+^ T_RM_ cells from BLM-induced fibrotic lungs vs. controls (Figure 3F–I).

Since the fibroblast-to-myofibroblast transition plays a key part in fibrogenesis, we investigated the direct influence exerted by CD4^+^ T_RM_ cells on this process. BLM pre-activated mouse lung fibroblasts were co-cultured with in vitro-induced CD4^+^ T_RM_ cells (Figure 3J). qPCR analysis demonstrated that co-culture with CD4^+^ T_RM_ cells, but not with CD4^+^ non-T_RM_ cells, significantly upregulated fibroblast expression of the myofibroblast markers Acta2, Col1a1, and Fn1 (Figure 3K–M), indicating that the CD4^+^ T_RM_ cells possess a pro-fibrotic function.

3.3. Depletion of Lung CD4+ TRM Cells Attenuated BLM-Induced PF

To define the causal function of lung CD4^+^ T_RM_ cells in the pathogenesis of PF in vivo, we exploited their characteristic high expression of the purinergic receptor P2RX7 [30]. Exogenous administration of high-dose NAD^+^ was used to selectively deplete P2RX7-expressing T_RM_ cells in BLM-challenged mice (Figure 4A). Flow cytometry confirmed that NAD^+^ treatment significantly reduced both the frequency and absolute number of lung CD4^+^CD69^+^CD103^+^ T_RM_ cells without affecting CD4^+^ non-T_RM_ cells (Figure 4B–E). This finding indicated that exogenous NAD^+^ can selectively deplete CD4^+^ T_RM_ cells in the lung tissues of mice with PF.

This specific depletion of CD4^+^ T_RM_ cells resulted in a marked amelioration of PF. NAD^+^-treated mice exhibited less weight loss and a lower lung coefficient (Figure 4F–G), along with improved gross lung morphology with reduced hemorrhage and lesions (Figure 4H). Histologically, H&E staining showed attenuated inflammatory infiltration, reduced alveolar septal thickening, and better alveolar architecture preservation (Figure 4I,J) in NAD^+^-treated mice. Masson’s staining and immunohistochemical analysis revealed significantly reduced collagen deposition (Figure 4K,L) and α-SMA expression (Figure 4M,N), respectively. Consistent with the above results, qPCR detection showed that key genes involved in fibrosis (Acta2, Col1a1, Col3a1, Mmp2, and Timp1) were downregulated in the group administered NAD^+^ (Figure 4O).

3.4. Recruitment of Circulating Lymphocytes Is a Major Source of Lung CD4+ TRM Cells in BLM-Induced PF

T_RM_ cells in local tissues can undergo self-renewal through in situ proliferation or recruitment of circulating lymphocytes [31,32,33]. To investigate the source of lung CD4^+^ T_RM_ cells in PF, we blocked lymphocyte egress from lymph nodes using the S1PR1 antagonist FTY720, starting three days before BLM administration (Figure 5A). Flow cytometry demonstrated that FTY720 successfully eliminated CD4^+^ T and CD8^+^ T cells from the peripheral circulation one day preceding BLM challenge (Figure 5B–D). This depletion persisted on day 14 post-BLM (Figure 5E–G).

Lung tissue analysis at the experimental endpoint indicated that FTY720 treatment led to a significant decline in total CD4^+^ and CD8^+^ T cells, alongside a prominent reduction in CD4^+^ non-T_RM_ subsets (Figure 5H–K). Importantly, we observed a substantial decrease in the population of lung CD4^+^ T_RM_ cells in FTY720-treated mice (Figure 5L), suggesting that recruitment from the circulating pool is a major source of CD4^+^ T_RM_ cell accumulation in fibrotic lungs.

3.5. Inhibition of Circulating Lymphocyte Recruitment Alleviated BLM-Induced PF

To evaluate the effect of FTY720 treatment on the progression of PF, we first monitored changes in mouse body weight. FTY720-treated mice showed attenuated weight loss from day 5 onward (Figure 6A,B), improved lung gross morphology (Figure 6C), and a reduced lung coefficient (Figure 6D). Histopathological assessment via H&E staining demonstrated that FTY720 treatment mitigated inflammatory cell infiltration and interstitial fibrosis in mouse lungs, while the alveolar architecture was largely preserved (Figure 6E,F). Furthermore, Masson’s staining indicated decreased collagen deposition (Figure 6G,H), and α-SMA expression (Figure 6I,J). Furthermore, qPCR results confirmed downregulation of fibrosis-related markers, including Acta2, Col1a1, Col3a1, and Timp1 in lung tissue (Figure 6K). These results demonstrate that FTY720 treatment significantly ameliorates BLM-induced PF in mice.

3.6. Activation of Notch Signaling in CD4+ TRM Cells from Fibrotic Lungs

Given the circulating origin of lung T_RM_ cells, we investigated whether CD4^+^ T cells from PF hosts exhibit an enhanced propensity for tissue residency. Emerging evidence indicates that local tissue TGF-β signaling is indispensable for the induction and formation of T_RM_ cells [25]. Therefore, we isolated splenic lymphocytes from BLM-treated mice and control mice, and stimulated with TGF-β in vitro. We observed that splenic CD4^+^ and CD8^+^ T cells from BLM-treated mice possessed a markedly increased capacity for T_RM_ cell differentiation in comparison with the control group (Figure S2A–D), suggesting an intrinsic mechanism priming for tissue residency of T cells.

Existing evidence indicates that Notch signaling can modulate the directional migration and chemotactic behavior of lymphocytes [34,35,36,37,38], and sustained Notch activation can subsequently foster a pro-fibrotic microenvironment by enhancing the secretion of fibrogenic cytokines and facilitating detrimental immune-stromal interactions [39,40,41]. With the aim of determining if Notch signaling is implicated in the pro-fibrotic CD4^+^ T_RM_ cells, we assessed the activity of this pathway in CD4^+^ T_RM_ cells. GSEA indicated significant enrichment of Notch signaling gene sets in human lung CD4^+^ T_RM_ vs. non-T_RM_ cells (Figure 7A). In addition, significant upregulation of Notch signaling molecules and their downstream transcription factors was observed in lung CD4^+^ T_RM_ cells from PF patients compared with control subjects (Figure 7B). Flow cytometry analysis revealed higher protein expression of Notch1 in in vitro induced CD4^+^ T_RM_ cells from BLM mice vs. controls (Figure 7C,D). Notably, we found that Notch1 expression was markedly higher in CD4^+^CD69^+^CD103^+^ T_RM_ cells than in either the CD4^+^CD69^−^CD103^−^ or CD4^+^CD69^+^CD103^−^ T cell populations (Figure 7E–H). Furthermore, a significant upregulation of Notch1 was observed in CD4^+^CD69^+^CD103^+^ T_RM_ cells, as compared with other CD4^+^ T cell subsets, in lung tissues of BLM-induced PF mice (Figure 7I,J). Taken together, these results suggest that Notch signaling could be involved in the regulation of CD4^+^ T cell tissue residency and pro-fibrotic function.

3.7. Inhibition of Notch Signaling Suppresses CD4+ TRM Cell Differentiation and Ameliorates PF

To further investigate the regulatory role of the Notch signaling in CD4^+^ T_RM_ cells and PF, we added the Notch signaling inhibitor DAPT to an in vitro induction system. As shown by flow cytometry, DAPT administration led to a significant reduction in the percentage of CD69^+^CD103^+^ T_RM_ cells (Figure S3A,B). Moreover, the pro-fibrotic capacity of these cells was impaired, as co-culture with DAPT-treated T_RM_ cells failed to upregulate myofibroblast markers in fibroblasts (Figure S3C–F), suggesting that blockade of the Notch signaling attenuates the pro-fibrotic function of CD4^+^ T_RM_ cells.

In in vivo experiments, DAPT was administered to a BLM-induced PF mice model (Figure 8A). Compared with the control group, the DAPT-treated group exhibited a significant reduction in both the proportion and total number of pulmonary CD4^+^CD69^+^CD103^+^ T_RM_ cells (Figure 8B–D). This was associated with substantial therapeutic benefits: alleviated weight loss (Figure 8E), improved lung morphology (Figure 8F), and significantly decreased the lung coefficient (Figure 8G). Histopathological analysis revealed reduced inflammatory cell infiltration, attenuated alveolar septal thickening, lower Ashcroft scores (Figure 8H,I), collagen deposition (Figure 8J,K), and α-SMA expression (Figure 8L,M) in the DAPT-treated group. qPCR further confirmed significant downregulation of mRNA expression of fibrosis-related molecules (Acta2, Col1a1, Col3a1, Mmp2, and Timp1) in the DAPT group (Figure 8N). These results demonstrate that Notch signaling activation promotes the tissue residency and pro-fibrotic function of CD4^+^ T_RM_ cells, representing Notch inhibition as a viable strategy to mitigate pulmonary fibrosis.

4. Discussion

In the present study, we demonstrate that CD4^+^ T_RM_ cells are significantly enriched in fibrotic lungs and their abundance correlates with disease severity. Functionally, these cells exhibit an enhanced pro-inflammatory and pro-fibrotic phenotype. Their specific depletion attenuates fibrosis, while in vitro, they directly promote myofibroblast transformation. Importantly, we delineate their origin by showing that the pulmonary CD4^+^ T_RM_ pool is largely derived from the recruitment of circulating T cells, and blocking this recruitment ameliorates the disease. Mechanistically, we identify the activation of the Notch signaling pathway within these cells in fibrotic tissue and show that its inhibition suppresses CD4^+^ T_RM_ differentiation, impairs their pro-fibrotic function, and alleviates fibrotic progression.

T_RM_ cells are a subset of long-lived memory T cells that reside in epithelial and mucosal tissues, independent of recirculation [42]. Accumulating evidence indicates a significant expansion of T_RM_ cells in the intestinal tissues of patients with inflammatory bowel disease, which display a pro-inflammatory profile characterized by high levels of IFN-γ, IL-17A, and TNF [43]. Furthermore, T_RM_ cells residing in the airways of asthma patients demonstrate cytotoxic-associated features, with higher expression of GZMA, GZMB, GZMH, and FASLG [44]. Consistent with these observations, our data indicate that within the pulmonary parenchyma of PF mouse models, CD4^+^ T_RM_ cells exhibited significantly higher expression of IL-17A and GZMB in comparison with those from control mice, and significantly exceeded levels in both CD4^+^CD69^−^CD103^−^ T cell and CD4^+^CD69^+^CD103^−^ T cell subsets. It has been well documented that IL-17A is capable of driving fibrotic responses in various organs including the heart, kidneys and liver [45,46,47], while GZMB also promotes fibrosis in the liver and retina [48,49]. Collectively, our data indicate that CD4^+^ T_RM_ cells may induce lung tissue injury and accelerate fibrosis development via the release of inflammatory cytokines.

P2RX7, a purinergic receptor functioning as a danger-sensing ion channel, is abundantly expressed on T_RM_ cells in organs such as the small intestine, liver, lungs, kidneys, and salivary glands, but shows lower expression levels on circulating memory T cells [30]. High concentrations of NAD^+^ can specifically recognize and hyperactivate the highly expressed P2RX7 receptors on T_RM_ cells, leading to the opening of ion channels and pore formation. This process triggers massive efflux of intracellular K^+^ ions, ultimately resulting in specific death of T_RM_ cells [30,43]. With the aim of investigating the function of CD4^+^ T_RM_ cells in PF pathogenesis, we treated BLM-induced PF mouse models with NAD^+^. As shown by the results, NAD^+^ administration significantly reduced the abundance of CD4^+^ T_RM_ cells in lung tissue while also attenuating pulmonary inflammatory responses and fibrotic damage. Moreover, our data revealed that CD4^+^ T_RM_ cells promote the transformation of fibroblasts into myofibroblasts in vitro. The core pathological feature of PF lies in the abnormal activation and phenotypic transformation of fibroblasts. Under persistent pathological stimulation, quiescent fibroblasts can differentiate into highly secretory myofibroblasts, which promote excessive deposition of extracellular matrix (ECM) through the secretion of large amounts of inflammatory mediators, thereby driving the fibrotic process [50,51]. Collectively, these results suggest that CD4^+^ T_RM_ cells contribute to the initiation and progression of PF.

The increase in T_RM_ cells within the local microenvironment may arise from two mechanisms: local expansion of T_RM_ cells or the conversion of circulating T cells into T_RM_ cells [52,53]. To investigate the source of CD4^+^ T_RM_ cells in the context of PF, we treated BLM-induced mice with FTY720, a functional antagonist of S1PR1. By blocking the S1PR1 signaling, this agent effectively inhibits lymphocyte egress from lymph nodes [54,55]. We observed that FTY720 treatment led to a significant reduction in the abundance of CD4^+^ T_RM_ cells within pulmonary tissue, suggesting that recruitment of circulating lymphocytes is an important source of this cell population. Concurrently, we observed that FTY720 treatment alleviated the severity of PF in mice. However, because FTY720 inhibits the generation of CD4^+^ T_RM_ cells while also causing a significant reduction in circulating lymphocyte counts, this off-target effect makes it difficult to directly distinguish the respective contributions of CD4^+^ T_RM_ cells and circulating lymphocytes to the progression of PF.

The Notch family comprises highly conserved transmembrane proteins. When ligands such as DLL-1, DLL-3, DLL-4, Jagged1 and Jagged2 bind to their cognate receptors (Notch1–4), the Notch signaling cascade is initiated. This activation triggers proteolytic cleavage at the transmembrane region by γ-secretase, generating the Notch intracellular domain (NICD), which subsequently activates the downstream transcription factor Hes1 and ultimately induces corresponding biological effects [56]. Notch signaling is critically involved in regulating the proliferation and differentiation of immature T lymphocytes, in addition to controlling the function of mature T cells [57,58]. In this study, we observed that the Notch signaling pathway is activated in CD4^+^ T_RM_ cells in lung tissues of patients with PF. DAPT, a γ-secretase inhibitor, effectively blocks the generation of NICD, thereby inhibiting Notch signaling [59]. In vitro experiments demonstrated that DAPT treatment suppresses the differentiation and pro-fibrotic function of CD4^+^ T_RM_ cells. In animal models, pharmacologic inhibition of Notch signaling significantly reduced the abundance of CD4^+^ T_RM_ cells in pulmonary tissue while also ameliorating fibrotic lesions associated with PF. These results suggest that the Notch signaling could regulate the differentiation and pro-fibrotic function of CD4^+^ T_RM_ cells.

In conclusion, we observed a significant increase in CD4^+^ T_RM_ cells in fibrotic lung tissues, with recruitment of circulating lymphocytes serving as a major source of this cell population. CD4^+^ T_RM_ cells exhibit a pronounced pro-fibrotic effect, and their differentiation depends on the activation of the Notch signaling. Blockade of this signaling effectively inhibits CD4^+^ T_RM_ cell differentiation and alleviates the progression of PF. Our study uncovers a novel regulatory mechanism underlying PF, and targeting the Notch signaling pathway to interfere with CD4^+^ T_RM_ cell differentiation may provide a new therapeutic approach to PF.

5. Conclusions

Our study establishes CD4^+^ tissue-resident memory T (T_RM_) cells as critical drivers of pulmonary fibrosis pathogenesis. Through integrated analysis of human fibrotic lung data and a BLM-induced mouse model, we demonstrated that these cells accumulate in fibrotic tissue, where they adopt a pro-inflammatory and pro-fibrotic phenotype, and their specific depletion significantly ameliorates disease severity. We further elucidated that pathogenic CD4^+^ T_RM_ cells predominantly originate from circulating lymphocytes recruited to the lung, rather than from local proliferation. Mechanistically, the Notch signaling pathway was identified as a key regulator of their differentiation and pro-fibrotic function. Collectively, our findings provide new insights into the immune mechanisms of fibrosis by identifying CD4^+^ T_RM_ cells as key pathogenic players and uncover their dependence on recruitment and Notch-mediated regulation. This positions CD4^+^ T_RM_ cells and the Notch pathway as promising targets for therapeutic intervention in pulmonary fibrosis.