Nerandomilast alleviates myositis-associated interstitial lung disease by modulating the non-Smad signalling pathway and activating the cAMP–PKA–RhoA pathway

Abstract

Objectives

Nerandomilast, a selective phosphodiesterase 4B (PDE4B) inhibitor, has been extensively investigated for the treatment of pulmonary fibrosis; however, its therapeutic potential and mechanisms of action in idiopathic inflammatory myopathies-associated interstitial lung disease (IIM-ILD) remain to be elucidated.

Methods

A myositis-associated ILD mouse model was treated with nerandomilast (BI 1015550), and lung pathology was assessed histologically. Human lung microvascular endothelial cells-5a were stimulated with neutrophil extracellular traps (NETs) to induce endothelial–mesenchymal transition (EndMT). Western blotting, immunofluorescence, qPCR and ELISA were employed to analyse pathways and cytokines.

Results

PDE4B was upregulated in the lungs of patients with IIM-ILD and in mice. Treatment with BI 1015550 significantly reduced lung inflammation and fibrosis scores, decreased inflammatory cytokines in bronchoalveolar lavage fluid and serum, alleviated muscle inflammation and lowered kinase activity without evident hepatorenal toxicity. Immunofluorescence and immunohistochemistry revealed diminished NET markers, reduced inflammatory cell infiltration (CD3, CD11b, F4/80), suppression of EndMT, downregulation of Ras homolog family member A (RhoA) and inhibition of key fibrotic pathways: phosphorylated phosphoinositide 3-kinase (P-PI3K), phosphorylated protein kinase B (P-AKT), phosphorylated p38 mitogen-activated protein kinase (P-P38), phosphorylated extracellular signal-regulated kinase (P-ERK), phosphorylated nuclear factor kappa-B (P-NF-κB). In vitro, BI 1015550 inhibited phorbol 12-myristate 13-acetate-induced neutrophil extracellular trap formation (NETosis) and EndMT.

Conclusions

Nerandomilast alleviates IIM-ILD by inhibiting NET formation and suppressing EndMT in lung microvascular endothelial cells, potentially through non-Smad signalling pathway modulation and cAMP–PKA–RhoA pathway activation. These findings suggest that the specific inhibition of PDE4B is a potential therapeutic approach for IIM-ILD.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Interstitial lung disease (ILD), the most common and severe complication of idiopathic inflammatory myopathy (IIM), is also the leading cause of death in these patients. However, effective treatment options are currently lacking.

• It is urgent to explore more effective and better-tolerated targeted therapeutic agents for the treatment of IIM-ILD.

WHAT THIS STUDY ADDS

• This study demonstrates for the first time that the novel selective phosphodiesterase 4B (PDE4B) inhibitor nerandomilast can alleviate lung progression in a mouse model of myositis-associated interstitial lung disease.

• Nerandomilast exerts its anti-inflammatory effects by inhibiting pulmonary inflammatory cell infiltration, neutrophil extracellular traps formation and the release of inflammatory factors.

• Concurrently, it exerts antifibrotic effects by activating the cyclic adenosine monophosphate–protein kinase A–Ras homolog family member A axis to reduce endothelial–mesenchymal transition and by inhibiting the non-Smads signalling pathway.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• This study clarifies the pathogenic role of PDE4B in IIM-ILD, offering a novel perspective for research into disease mechanisms. It also proposes a new therapeutic strategy of ‘targeting the source of inflammation while concurrently addressing antifibrosis’ for IIM-ILD, which holds promise for overcoming current treatment limitations.

• Nerandomilast shows potential as an innovative candidate drug for the treatment of IIM-ILD.

Introduction

Idiopathic inflammatory myopathies (IIMs), including dermatomyositis (DM), polymyositis (PM), immune-mediated necrotising myopathy and antisynthetase syndrome, are autoimmune disorders characterised by chronic muscle inflammation and multiorgan involvement.¹ Interstitial lung disease (ILD), the most frequent and serious extramuscular manifestation, occurs in 30%–60% of IIM patients and is a major cause of progressive pulmonary decline and respiratory failure.2,4 Current first-line treatment relies on glucocorticoids and immunosuppressants; however, their efficacy is often limited by treatment resistance and long-term adverse effects, such as infections and osteoporosis.^{1 5} We have previously demonstrated that colchicine, pirfenidone and stimulator of interferon genes inhibitors ameliorate lung inflammation and fibrosis in a mouse model of myositis-associated ILD (MAILD),6,8 supporting the feasibility of targeted therapeutic strategies. Thus, elucidating the pathogenesis of IIM-ILD and developing more effective and tolerable targeted therapies remain urgent priorities.

The inflammation–fibrosis vicious cycle is a central pathological mechanism in ILD progression. Persistent pulmonary inflammation damages the alveolar epithelium, resulting in the release of cytokines, including transforming growth factor-beta (TGF-β) and tumour necrosis factor-alpha (TNF-α), which activate fibroblast proliferation and differentiation. Activated fibroblasts amplify inflammation by secreting chemokines that recruit additional immune cells, establishing a self-sustaining inflammatory–fibrotic feedback loop.^{9 10} Intracellular cyclic adenosine monophosphate (cAMP) plays a key regulatory role in this process, exerting anti-inflammatory and antifibrotic effects by suppressing immune cell activation and inhibiting fibroblast collagen production.¹¹ PDE4, particularly the phosphodiesterase 4B (PDE4B) subtype, which is highly expressed in lung macrophages, Th17 cells and fibroblasts,¹² plays a critical role in regulating cAMP levels. PDE4B is overactivated in ILD, leading to accelerated cAMP degradation and loss of its protective effects, thereby promoting fibrotic progression.13,17 Nerandomilast, a novel selective PDE4B inhibitor, has demonstrated antifibrotic potential in preclinical studies. Recent results from the FIBRONEER-ILD trial presented at TS 2025 showed that nerandomilast significantly slowed the decline in forced vital capacity in patients with progressive pulmonary fibrosis.¹⁸ Nevertheless, the therapeutic utility and mechanism of PDE4 inhibition in IIM-ILD remain uncharacterised.

Nerandomilast is an investigational agent for idiopathic pulmonary fibrosis and other progressive fibrosing ILDs. Thus, we investigated PDE4B’s role in IIM-ILD pathogenesis to validate it as a novel therapeutic target. Furthermore, we evaluated the regulatory effects of nerandomilast on profibrotic pathways in IIM-ILD. Clinical samples and an MAILD mouse model were employed to elucidate the dual effects of nerandomilast on suppressing lung inflammation and retarding fibrosis, thereby providing a translational foundation for precisely targeted IIM-ILD therapies.

Materials and methods

Human lung tissue pathological study

This study included IIM patients diagnosed per the 2017 European League Against Rheumatism/American College of Rheumatology criteria and with ILD confirmed after excluding infection and malignancy.¹⁹ Lung biopsies from two patients with DM and one normal paracarcinoma control were analysed via immunohistochemistry.

Construction of the MAILD mouse model and intervention

Female BALB/c mice (6–8 weeks old, 18–22 g) and Sprague Dawley rats (8 weeks, 200 g) were obtained from the Lanzhou Veterinary Research Institute and maintained under SPF conditions. The mice were randomly divided into three groups (n=8 per group): control, MAILD and nerandomilast (BI 1015550). The MAILD model was induced as previously described by immunising mice on days 0, 7, 14 and 21 with a subcutaneous injection of rat skeletal muscle homogenate (30 mg/mL) emulsified in Complete Freund’s Adjuvant (0.25 mL per side), combined with an intraperitoneal injection of 10 µg ODN2395 (TLR9 agonist).²⁰ Based on the previous literature, this study selected two doses, 3 mg/kg and 12 mg/kg, for in vivo efficacy evaluation, referencing the intervention doses used in the bleomycin-induced pulmonary fibrosis mouse model.²¹ Starting on day 3, the BI 1015550 group received oral administration of the inhibitor (3 or 12 mg/kg, two times per day), whereas the control group received an equal volume of hydroxypropyl cellulose sodium. All mice were euthanised on day 28. After anaesthesia with 0.3% sodium pentobarbital, blood was collected via cardiac puncture and centrifuged to obtain serum. Bronchoalveolar lavage fluid (BALF) was collected from three mice per group using 1 mL PBS. The left lung was processed for histopathology, the right lung for western blotting and quantitative real-time PCR, and thigh muscles for histological analysis.

Histopathological staining

Following 24-hour fixation in 4% neutral formalin, the left lung and thigh muscles were paraffin-embedded and sectioned at 4 µm for histological staining (H&E for inflammation; Masson’s trichrome for fibrosis). The right lung was cryopreserved for molecular analyses. Digitised whole-slide images (TissueFAXS Plus) were used for blinded pathological scoring: lung lesions were quantified by affected area and muscle inflammation was assessed using a Likert scale.²²

Immunohistochemical staining

Following antigen retrieval, the paraffin-embedded lung sections were incubated overnight at 4°C with primary antibodies, followed by incubation with horseradish peroxidase(HRP)-conjugated secondary antibodies. The signal was developed using 3,3'-diaminobenzidine, and slides were mounted for microscopic visualisation.

Measurement of mouse serum biochemical indicators

Mouse serum was diluted fivefold in normal saline, and the levels of aspartate aminotransferase, alanine aminotransferase, creatinine, creatine kinase and lactate dehydrogenase in mouse serum were measured using a URIT fully automated small animal biochemical analyser (CA-200Vet).

Measurement of serum NET levels and inflammatory cytokines in mice

Circulating free DNA (cfDNA) was evaluated in the serum as an indirect measure of neutrophil extracellular trap (NET) formation using the Quant-iT PicoGreen dsDNA Assay Kit (Invitrogen), with triplicate samples quantified against a standard curve. Concurrently, concentrations of interleukin (IL)-1β, IL-6 and TNF-α in both the serum and BALF were determined using specific ELISA kits (Jianglai, Shanghai, China) per the manufacturer’s instructions.

Isolation of neutrophils and extraction of NETs

Neutrophils were isolated from whole blood collected from healthy volunteers in EDTA tubes. Within 30 min, the blood was layered over a discontinuous gradient of Histopaque-1119 and Histopaque-1077 (Sigma) at a ratio of 5:2:7 and centrifuged at ×750g for 30 min at room temperature. The neutrophil layer was collected, washed with phosphate-buffered saline (PBS) and centrifuged at ×450g for 10 min. The cells were then resuspended in Roswell Park Memorial Institute 1640 medium (Gibco) supplemented with 10% fetal bovine serum and counted.

For NET extraction, neutrophils (1×10⁶ cells/well) were seeded in six-well plates and incubated for 30 min at 37°C with 5% CO₂. The cells were stimulated with 100 nM phorbol 12-myristate 13-acetate (PMA) (Solarbio) for 4 hours. After washing with calcium-free and magnesium-free PBS and centrifugation at ×450g for 10 min, the cell-free supernatant containing NETs was collected. NETs were quantified using the PicoGreen dsDNA assay and stored at −80°C.

Cell culture and treatment

We cultured human lung microvascular endothelial cells (HULEC-5a) (Meisen CTCC; passages 2–3) under three conditions. (1) Control: MCDB131 medium; (2) NETs: 100 ng/mL NETs for 24 hours; (3) NETs+inhibitor: 30 min pretreatment with BI 1015550 (TargetMol, T72031) followed by NET co-culture.

RNA sequencing and differential gene analysis

Total RNA was extracted from control and NET-treated HULEC-5a cells (as above) using TRIzol reagent (Invitrogen) and stored at −80°C. Shanghai OE Biotech performed transcriptomic sequencing of three biological replicates per group. The bioinformatic analysis included quality control, read alignment, gene quantification and identification of differentially expressed genes (DEGs), according to the criteria of ∣log₂FC∣>1 and false discovery rate <0.05. Functional enrichment analyses of the DEGs were conducted using the Gene Ontology and Kyoto Encyclopedia of Genes and Genomes (KEGG) databases.

Cell and tissue immunofluorescence

HULEC-5a cells were plated, NET-treated (100 ng/mL, 24 hours), fixed, permeabilised and stained with primary/secondary antibodies and 4',6-diamidino-2-phenylindole (DAPI). Images were acquired via two-photon confocal microscopy.

Lung sections (4 µm) underwent antigen retrieval, incubation with primary/secondary antibodies and DAPI mounting. Expression was visualised using a panoramic quantitative microscope.

Western blotting

Protein extracts from HULEC-5a cells and lung tissues in radioimmunoprecipitation assay buffer containing protease inhibitors (Solarbio, R0010) were quantified using bicinchoninic acid assay, denatured and separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis. After transfer to polyvinylidene difluoride membranes, the membranes were blocked with 5% skim milk. The membranes were then probed with primary antibodies (1:1000–1:2000, 4°C, overnight) and HRP-conjugated secondary antibodies (room temperature, 1 hour), and signals were detected with an enhanced chemiluminescence system.

Quantitative real-time PCR

Total RNA from mouse lung tissues was extracted with TRIzol (AgBio), quantified spectrophotometrically and reverse-transcribed into cDNA using the RevertAid kit. qPCR was conducted in 20 µL reactions containing SYBR Green Master Mix, primers (0.8 µL, 10 µM) and cDNA (2 µL) under the following cycling conditions: 95°C for 30 s, then 40 cycles of 95°C for 5 s and 60°C for 30 s (table 1). GAPDH served as the reference gene.

Table 1. Primer sequences for quantitative real-time PCR assay.

Gene	Forward primer (5′ to 3′)	Reverse primer (5′ to 3′)
GAPDH	AGGTCGGTGTGAACGGATTTG	GGGGTCGTTGATGGC AACA
PDE4A	GAACCGGGAACTCACACACC	GTACTCTGAGACCTGGTTCCT
PDE4B	CGCAGGGAGTCGTTCCTCTA	CTCCTGTGGTCGCACACTTG
PDE4C	TCCGAGAGCCAGTGGATTCT	CCTTGAGTTCCAATCGTGAAGAC
PDE4D	TTTTGCCAGTGCAATACATGATG	CAGAGCGAGTTCCGAGTTTGT

GAPDH, glyceraldehyde-3-phosphate dehydrogenase; PDE4A, phosphodiesterase 4A; PDE4B, phosphodiesterase 4B; PDE4C, phosphodiesterase 4C; PDE4D, phosphodiesterase 4D.

High-content live-cell imaging

We divided neutrophils into the control (PBS), PMA (100 nM) and PMA+BI 1015550 groups. The inhibitor group was pretreated with BI 1015550 (1 or 10 µM) for 1 hour before PMA stimulation. After 2 hours of incubation, the cells were stained with SYTOX Green and Hoechst and imaged using a high-content imaging system for fluorescence analysis.

Data analysis

All quantitative data represent three independent experiments and are expressed as the mean±SD. Normality was tested with the Shapiro-Wilk test. Subsequent analyses used t-tests/analysis of variance for parametric data or the Kruskal-Wallis test for non-parametric data. Analyses were performed using GraphPad Prism V.10.0. A p value <0.05 was considered to indicate significance.

Results

PDE4B expression is upregulated in the lung tissues of patients with DM-ILD and MAILD mice

Immunohistochemistry revealed a significant upregulation of PDE4B expression in lung tissues of patients with DM-ILD (figure 1A), suggesting a potential role for the PDE4B gene in IIM-ILD pathogenesis. Similarly, immunohistochemical staining confirmed markedly elevated PDE4B expression in the lung tissues of both experimental autoimmune myositis (EAM) and MAILD mice (figure 1B). Western blotting further validated increased PDE4B protein levels in the lung tissues of MAILD mice (figure 1C). To further characterise the expression profile of PDE4 isozymes in lung tissue, we measured the mRNA levels of PDE4A, PDE4B, PDE4C and PDE4D in MAILD mouse lungs. PDE4B expression was significantly upregulated, whereas Pde4A and PDE4D expression was significantly lower in model versus control mice (figure 1D). Thus, PDE4B is the predominant subtype expressed in the lung tissues of IIM-ILD, establishing its pathological basis as a therapeutic target.

Figure 1. PDE4B expression is upregulated in the lung tissues of patients with DM-ILD and MAILD mice. (A) Immunohistochemical staining assessing PDE4B expression levels in the lung tissues of patients with DM-ILD. Scale bars: 50 µm, 20 µm. (B) Immunohistochemical staining assessing PDE4B expression levels in the lung tissues of EAM and MAILD mice. Scale bars: 50 µm, 20 µm. (C) Western blotting assessing PDE4B expression levels in the lung tissues of MAILD mice. (D) Quantitative PCR analysis of the mRNA levels of PDE4A, PDE4B, PDE4C and PDE4D in the MAILD mouse model. *p<0.05, **p<0.01, ***p<0.001 and ****p<0.0001. DM-ILD, dermatomyositis-associated interstitial lung disease; EAM, experimental autoimmune myositis; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; MAILD, myositis-associated interstitial lung disease; ns, not significant; PDE4B, phosphodiesterase 4B.

The PDE4B-selective inhibitor nerandomilast ameliorates ILD progression in mice

Nerandomilast (BI 1015550), a selective PDE4B inhibitor, was administered to MAILD model mice. Histological analysis of lung tissue via H&E and Masson staining revealed that both the low-dose and high-dose nerandomilast treatment groups showed significant alleviation of lung inflammation and fibrosis compared with that observed in the model group (figure 2A,B). However, H&E staining of the mouse thigh muscle revealed that, compared with that in the model group, inflammatory cell infiltration in the intervention group was decreased (figure 2C), and serum kinase levels were significantly reduced; no significant increase was observed in liver or kidney function indicators (figure 2D). Our results demonstrate that nerandomilast exerts significant pharmacological effects, alleviating interstitial lung disease in MAILD mice. The treatment exerts potential protective effects on muscle and shows no liver or renal toxicity, thereby confirming its therapeutic safety.

Figure 2. The phosphodiesterase 4B-selective inhibitor nerandomilast ameliorates interstitial lung disease progression in mice. (A) H&E staining of lung tissues and corresponding lung inflammation scores for mice in different groups. Scale bar: 50 µm. (B) Masson’s trichrome staining of lung tissues and corresponding fibrosis scores for mice in different groups. Scale bar: 50 µm. (C) H&E staining of muscle tissues and corresponding muscle inflammation scores for mice in different groups. Scale bar: 50 µm. (D) Serum levels of ALT, AST, CRE, CK and LDH in mice from different groups. L Inhibitor: 3 mg/kg BI 1015550; H Inhibitor: 12 mg/kg BI 1015550. *p<0.05, **p<0.01, ***p<0.001 and ****p<0.0001. ALT, aspartate aminotransferase; AST, aspartate aminotransferase; CK, creatine kinase; CRE, creatinine; LDH, lactate dehydrogenase; MAILD, myositis-associated interstitial lung disease; ns, not significant.

Nerandomilast alleviates the progression of MAILD by inhibiting NET formation

Building on our previous findings that NETs are a potential therapeutic target in MAILD mice, we evaluated the effect of nerandomilast on NET formation in vivo. Immunohistochemistry demonstrated that, compared with that in the model group, the expression of NET markers (neutrophil elastase (NE), myeloperoxidase (MPO) and citrullinated histone H3 (CitH3)) was significantly reduced in the lung tissues of nerandomilast-treated mice (figure 3A). Immunofluorescence colocalisation staining further revealed a marked decrease in the colocalisation signal of CitH3 and MPO in the treatment group (figure 3B). Furthermore, the measurement of serum NET-related cfDNA levels using the Picogreen assay revealed a significant decrease in the nerandomilast-treated versus model group (figure 3C). In vitro high-content live-cell imaging of PMA-induced NET formation in neutrophils with fluorescence intensity quantification revealed that nerandomilast significantly inhibited PMA-induced NET formation (figure 3D). Thus, nerandomilast likely alleviates the progression of MAILD by inhibiting the formation of NETs.

Figure 3. Nerandomilast alleviates the progression of myositis-associated interstitial lung disease by inhibiting neutrophil extracellular trap (NET) formation. (A) Immunohistochemical staining assessing the expression levels of NE, MPO and CitH3 in the lung tissues of mice from different groups. Scale bar: 50 µm. (B) Immunofluorescence analysis of CitH3 and MPO coexpression in the lung tissues of mice from different groups. Scale bar: 20 µm. (C) Serum cfDNA levels in mice from different groups. (D) Levels of NET formation by neutrophils from different treatment groups. Scale bar: 10 µm. In vivo treatments (A-C): L Inhibitor: 3 mg/kg BI 1015550; H Inhibitor: 12 mg/kg BI 1015550. In vitro treatments (D): L Inhibitor: 1 µM BI 1015550; H Inhibitor: 10 µM BI 1015550. *p<0.05, **p<0.01, ***p<0.001 and ****p<0.0001. CitH3, citrullinated histone H3; DAPI, 4',6-diamidino-2-phenylindole; MAILD, myositis-associated interstitial lung disease; MPO, myeloperoxidase; NC, negative control; NE, neutrophil elastase; ns, not significant; PMA, phorbol 12-myristate 13-acetate.

Nerandomilast alleviates the progression of MAILD by inhibiting lymphocyte infiltration, macrophage infiltration and polarisation and the release of inflammatory factors

After confirming the significant therapeutic effect of nerandomilast on interstitial lung disease in MAILD mice, we next investigated the underlying mechanisms involved. Immunohistochemical analysis revealed that nerandomilast intervention significantly reduced the infiltration of CD3⁺ T lymphocytes, CD11b⁺ monocyte-derived macrophages and F4/80⁺ macrophages (figure 4A). M1 and M2 macrophage polarisation play important roles in ILD progression; immunofluorescence staining demonstrated that both M1 (CD80) and M2 (CD163) macrophage polarisation levels were decreased in the lung tissues of the treatment group compared with those in the control group (figure 4B). Concurrently, we evaluated the expression levels of inflammatory cytokines in nerandomilast-treated MAILD mice. ELISA demonstrated that the levels of the inflammatory cytokines IL-1β, IL-6 and TNF-α in the serum and BALF of the treatment group were significantly reduced in comparison with those in the model group (figure 4C,D), indicating that nerandomilast exerts anti-inflammatory effects. In summary, our findings demonstrate that in MAILD mouse lung tissues, nerandomilast alleviates ILD development by reducing lymphocyte and macrophage infiltration and polarisation and inhibiting the release of related inflammatory cytokines.

Figure 4. Nerandomilast alleviates the progression of myositis-associated interstitial lung disease by inhibiting lymphocyte infiltration, macrophage infiltration and polarisation and the release of inflammatory factors. (A) Immunohistochemical staining assessing the expression levels of CD3, CD11b and F4/80 in the lung tissues of mice from different groups. Scale bar: 50 µm. (B) Immunofluorescence staining assessing the expression levels of CD80 and CD163 in the lung tissues of mice from different groups. Scale bar: 20 µm. (C) Levels of interleukin (IL)-1β, IL-6 and TNF-α in the bronchoalveolar lavage fluid (BALF) of mice from different groups. (D) Serum levels of IL-1β, IL-6 and TNF-α in mice from different groups. L Inhibitor: 3 mg/kg BI 1015550; H Inhibitor: 12 mg/kg BI 1015550. *p<0.05, **p<0.01, ***p<0.001 and ****p<0.0001. DAPI, 4',6-diamidino-2-phenylindole; MAILD, myositis-associated interstitial lung disease; ns, indicates not significant; TNF-α, tumour necrosis factor-alpha.

RNA sequencing suggests that NETs promote the high expression of PDE4B and lung fibrosis-related genes

The pathogenesis of ILD is highly complex, involving multiple mechanisms, including vascular injury, altered immune responses and inflammatory activation. Endothelial cells line the entire vascular system and maintain cellular and tissue homeostasis. We have previously demonstrated that NETs stimulate endothelial cell pyroptosis, promoting ILD progression. We performed RNA transcriptomic sequencing of HULEC-5a cells treated with NETs to further elucidate the effects of NETs on endothelial cells and their relationship with PDE4. As demonstrated in the volcano plot (figure 5A), transcriptome sequencing detected changes in 1351 genes 24 hours after NET stimulation, of which 825 were upregulated and 526 were downregulated. NET stimulation of HULEC-5a cells triggered the significant upregulation of pathways related to cytokine–cytokine receptor interaction, inflammatory response, immune response, pulmonary fibrosis and proinflammatory/profibrotic signalling (figure 5B–G). Furthermore, among the four PDE4 subtypes, PDE4B expression was markedly increased in NET-stimulated HULEC-5a cells; furthermore, elevated expression of the endothelial–mesenchymal transition (EndMT) marker α-SMA gene and inflammatory cytokines was observed (figure 5H). Thus, we hypothesise that NETs may activate the PDE4B gene, promoting EndMT in HULEC-5a cells and contributing to ILD development.

Figure 5. RNA sequencing suggests that NETs promote the high expression of PDE4B and lung fibrosis-related genes. (A) RNA-seq analysis of human lung microvascular endothelial cells (HULEC-5a) cells stimulated with NETs for 24 hours, showing 825 significantly downregulated and 526 significantly upregulated genes. (B) Top 30 significantly upregulated pathways in the GO enrichment analysis. The pathways related to inflammation and inflammatory cytokines are highlighted in red. (C) Top 20 significantly upregulated pathways in the KEGG functional enrichment analysis. NET stimulation of HULEC-5a cells led to significant upregulation of fibrosis-related signalling pathways, including the TNF, NF-κB, JAK-STAT and PI3K-AKT pathways, highlighted in red. (D–H) Gene set enrichment analysis revealed significant enrichment in the inflammatory response (GO:0006954) and immune response (GO:0006955) in the GO analysis, as well as in lung fibrosis (WP3624) and proinflammatory/fibrotic pathways (WP5095) in the Wikipathways analysis. All enrichments were significant (p<0.05). (H) FPKM values showing the expression levels of PDE4A, PDE4B, PDE4C, PDE4D, ACTA2 (α-SMA), IL1B and TNF in HULEC-5a cells after NET stimulation. AGE, advanced glycation end products; ERK, extracellular signal-regulated kinase; FDR, false discovery rate; FPKM, fragments per kilobase of transcript per million mapped reads; GO, Gene Ontology; IL, interleukin; JAK, Janus kinase; KEGG, Kyoto Encyclopedia of Genes and Genomes; NC, negative control; NES, normalized enrichment score; NETs, neutrophil extracellular traps; RAGE, receptor for advanced glycation end products; STAT, signal transducer and activator of transcription; TNF, tumour necrosis factor; WP, WikiPathways.

Nerandomilast alleviates the progression of MAILD by inhibiting endothelial–mesenchymal transition via the cAMP–PKA–RhoA pathway

Immunohistochemistry, immunofluorescence and Western blotting demonstrated that lung tissues from MAILD mice exhibited decreased expression of the endothelial marker CD31 and increased expression of the mesenchymal marker α-SMA, suggesting the potential involvement of EndMT in ILD development. Concurrently, compared with the model group, the nerandomilast intervention group showed increased CD31 expression and decreased α-SMA expression (figure 6A–C). Thus, nerandomilast may alleviate ILD by inhibiting EndMT. Further analysis revealed significantly elevated expression of Ras homolog family member A (RhoA) in the lung tissues of MAILD mice, which was suppressed following nerandomilast intervention (figure 6A–C), suggesting that nerandomilast might interfere with the EndMT process by modulating the cAMP–PKA–RhoA signalling axis, thereby delaying the progression of MAILD. To validate this mechanism in vitro, we stimulated HULEC-5a endothelial cells with NETs to induce EndMT and concurrently applied nerandomilast treatment. Consistent results from western blotting and immunofluorescence demonstrated that, compared with that in the NET-stimulated group, the nerandomilast intervention group exhibited increased CD31 expression and decreased expression of both α-SMA and RhoA (figure 6D,E). These results collectively indicate that nerandomilast likely alleviates the progression of MAILD by inhibiting EndMT via the cAMP–PKA–RhoA pathway.

Figure 6. Nerandomilast alleviates the progression of myositis-associated interstitial lung disease by inhibiting endothelial–mesenchymal transition via the cAMP–PKA–RhoA pathway. (A) Immunohistochemical staining assessing the expression levels of CD31, α-SMA and RhoA in the lung tissues of mice from different groups. Scale bar: 50 µm. (B) Immunofluorescence staining assessing the expression levels of CD31, α-SMA and RhoA in the lung tissues of mice from different groups. Scale bar: 20 µm. (C) Western blotting of CD31, α-SMA and RhoA expression levels in the lung tissues of mice from different groups. (D) Western blotting of CD31, α-SMA and RhoA expression levels in human lung microvascular endothelial cells (HULEC-5a) endothelial cells from different treatment groups. (E) Immunofluorescence staining assessing the expression levels of CD31, α-SMA and RhoA in HULEC-5a endothelial cells from different treatment groups. In vivo treatments (A–C): L Inhibitor: 3 mg/kg BI 1015550; H Inhibitor: 12 mg/kg BI 1015550. In vitro treatments (D,E): L Inhibitor: 1 µM BI 1015550; H Inhibitor: 10 µM BI 1015550. *p<0.05, **p<0.01, ***p<0.001 and ****p<0.0001. cAMP, cyclic adenosine monophosphate; DAPI, 4',6-diamidino-2-phenylindole; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; MAILD, myositis-associated interstitial lung disease; MERGE, merged image; NC, negative control; NETs, neutrophil extracellular traps; ns, not significant; PKA, protein kinase A; RhoA, Ras homolog family member A.

Nerandomilast ameliorates the progression of MAILD by inhibiting PDE4B signalling and the non-Smad fibrotic signalling pathway

To further elucidate the specific molecular mechanisms by which nerandomilast alleviates IIM-ILD, we first examined its effect on PDE4B expression through western blotting. Nerandomilast specifically downregulated PDE4B protein expression in lung tissue (figure 7B). The non-Smad signalling pathway is a classical pathway involved in the development and progression of fibrotic diseases. Thus, we assessed the expression of cAMP-mediated CREB pathway and non-Smad pathway proteins in lung tissue through western blotting and immunohistochemistry. Nerandomilast enhanced CREB phosphorylation and disrupted the activation of profibrotic signalling pathways, such as PI3K/AKT and mitogen-activated protein kinase (MAPK), as well as the proinflammatory NF-κB signalling pathway (figure 7A,B).

Figure 7. Nerandomilast ameliorates the progression of myositis-associated interstitial lung disease by inhibiting PDE4B signalling and the non-Smad fibrotic signalling pathway. (A) Immunohistochemical staining assessing the expression levels of PDE4B, ERK, AKT, PI3K, p38, P-CREB and NF-κB in the lung tissues of mice from different groups. Scale bar: 50 µm. (B) Western blotting of PDE4B, P-ERK, P-AKT, P-PI3K, P-P38, P-CREB and P-NF-κB expression levels in the lung tissues of mice from different groups. L Inhibitor: 3 mg/kg BI 1015550; H Inhibitor: 12 mg/kg BI 1015550. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. AKT, protein kinase B (PKB); ERK, extracellular signal-regulated kinase; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; MAILD, myositis-associated interstitial lung disease; NF-κB, nuclear factor kappa-B; ns, not significant, PDE4B, phosphodiesterase 4B; p38, p38 mitogen-activated protein kinase; P-CREB, phosphorylated cAMP response element-binding protein; PI3K, phosphoinositide 3-kinase.

To the best of our knowledge, this study is the first to delineate the multitarget regulatory mechanism of nerandomilast in alleviating ILD in an MAILD mouse model: the drug specifically downregulates PDE4B expression, thereby reducing cAMP degradation and activating the cAMP/CREB pathway, inhibiting pulmonary fibrosis progression. Concurrently, nerandomilast suppresses the activation of the non-Smad (PI3K/AKT, MAPK) and NF-κB pathways, collectively alleviating pulmonary inflammation and interstitial remodelling, ultimately exerting an anti-ILD effect.

Discussion

Our findings confirm that PDE4B expression is significantly upregulated in lung tissues from myositis patients and MAILD model mice. Intervention with the PDE4B-selective inhibitor nerandomilast in MAILD mice markedly alleviated lung inflammation and fibrosis. Mechanistically, nerandomilast was found to reduce neutrophil recruitment, inhibit NET formation, decrease monocyte, macrophage and lymphocyte infiltration, reduce macrophage polarisation and attenuate the inflammatory response both in vitro and in vivo. Furthermore, nerandomilast was found to ameliorate pulmonary fibrosis by inhibiting the non-Smad fibrotic signalling pathway and improving EndMT in pulmonary microvascular endothelial cells via the cAMP–PKA–RhoA pathway. Nerandomilast has recently been launched in China, and clinical experience with its use remains insufficient. This study is the first to confirm the efficacy of nerandomilast in an MAILD model, suggesting its potential as a novel therapeutic option for this condition.

PDE4B is widely distributed with a tissue-specific pattern, showing high expression primarily in the lungs, immune cells and brain. Except for the brain-specific PDE4B5 subtype, all PDE4B subtypes are found in the lungs.¹¹ This study revealed distinct expression patterns of PDE4 subtypes in IIM-ILD lung tissue (PDE4B upregulation, PDE4A/D downregulation), indicating their functional heterogeneity in pulmonary inflammation and fibrosis. PDE4B, a negative regulator of cAMP signalling, may promote proinflammatory mediator release on upregulation, whereas PDE4D isoforms are predominantly expressed in the brain and associated with neurological functions.²³ This complex regulatory network suggests that targeting specific subtypes such as PDE4B, rather than pan-PDE4 inhibition, may offer a more promising therapeutic strategy. Studies have shown that in mouse models of pulmonary fibrosis, PDE4 inhibitor administration exhibits antifibrotic activity comparable to that of the antifibrotic drugs pirfenidone and nintedanib.²⁴ In canine models, unlike non-selective PDE4 inhibitors such as roflumilast, mangosteen-derived PDE4 inhibitors demonstrated antifibrotic effects similar to those of pirfenidone in a bleomycin-induced rat pulmonary fibrosis model, without causing vomiting.²⁵ Furthermore, AA6216 was also confirmed to reduce the fibrosis score, collagen-stained area and TGF-β levels in the BALF in a bleomycin-induced idiopathic pulmonary fibrosis mouse model.²⁶ This study confirms high PDE4B expression in the lung tissues of patients with IIM-ILD, EAM mice and MAILD mice. In the MAILD mouse model, H&E and Masson staining showed that the nerandomilast intervention group had reduced lung inflammation and fibrosis scores, as well as a decreased collagen-stained area, indicating that nerandomilast may alleviate the fibrotic process in MAILD.

Excessive NET formation contributes to alveolar damage in IIM-ILD. In patients with DM-ILD, elevated low-density granulocytes drive aberrant NET generation, while impaired DNase I-mediated clearance leads to NET accumulation.^{27 28} PDE4 inhibitors prevent cAMP degradation, attenuating inflammation and immune cell activation, including neutrophils.²⁹ Here, nerandomilast was found to significantly reduce NET markers (CitH3, NE and MPO) in lung tissue and decrease serum cfDNA levels. Furthermore, nerandomilast directly inhibited PMA-induced NET formation in neutrophils in vitro. These results demonstrate that nerandomilast alleviates MAILD progression by suppressing NET formation.

Multiple innate immune cells participate in the pathogenesis of idiopathic inflammatory disease-associated ILD.¹ Our previous work demonstrated that NETs exacerbate MAILD by promoting macrophage polarisation and pyroptosis through the TLR9–NF-κB/AIM2 pathway.³⁰ In addition, T-cell differentiation and activation, which are key processes underlying the secretion of inflammatory cytokines, such as IFN-γ, IL-4 and IL-17, are strongly implicated in IIM-ILD pathogenesis.^{31 32} PDE4 inhibitors, including apremilast, modulate anti-inflammatory responses in T lymphocytes and macrophages.³³ In the present study, nerandomilast significantly reduced CD3⁺ T cell, F4/80⁺ macrophage and CD11b⁺ monocyte-derived macrophage infiltration in inflamed and fibrotic lung areas and suppressed both M1 and M2 macrophage polarisation. Furthermore, nerandomilast markedly lowered TNF-α, IL-6 and IL-1β levels in BALF and serum. These findings suggest that nerandomilast ameliorates IIM-ILD by modulating T-cell and macrophage responses and attenuating inflammatory cytokine release.

Our previous work demonstrated that NETs drive IIM-ILD progression by activating the NLRP3 inflammasome and inducing pyroptosis in pulmonary microvascular endothelial cells.³⁴ Endothelial dysfunction, an early pathogenic event in multiple diseases,³⁵ is modulated by phosphodiesterase activity. While dual PDE3/PDE4 inhibition attenuates vascular remodelling in pulmonary hypertension,³⁶ selective PDE4 inhibition results in milder fibrosis in murine models. Notably, endothelial PDE4B upregulation has been shown to suppress the PKA–CREB–BMPRII axis and promote EndMT in hypoxic pulmonary hypertension.³⁷ Consistent with these findings, our sequencing data revealed significant PDE4B expression upregulation in endothelial cells, suggesting its therapeutic relevance. Consistently, nerandomilast treatment increased CD31 expression and decreased α-SMA levels, demonstrating its ability to attenuate EndMT and providing mechanistic insight into PDE4B inhibition in IIM-ILD. It should be noted that although this study observed that NET stimulation may upregulate the expression of PDE4B and that the PDE4 inhibitor nerandomilast can alleviate the progression of EndMT, the specific molecular mechanisms by which NETs regulate the transcription or translation of PDE4B have not yet been fully elucidated. The current findings support a key role for PDE4B in NET-induced EndMT; however, whether NETs directly activate the transcription of PDE4B through specific signalling pathways (such as the TLRs/NF-κB pathway) or upregulate its expression through mechanisms affecting mRNA stability requires further in-depth investigation.

In the endothelium, cAMP cooperates with Rho signalling to maintain barrier integrity, regulate permeability and modulate vascular tone.³⁸ RhoA, a key Rho GTPase, regulates cytoskeletal dynamics and influences cell morphology, migration and adhesion.³⁹ on activation by TGF-β, inflammatory cytokines or mechanical stress, RhoA promotes actin-myosin reorganisation, driving EndMT and fibrosis progression.^{40 41} Under normal conditions, cAMP-activated PKA suppresses RhoA activity, preserving endothelial homeostasis.⁴² In this study, nerandomilast reduced RhoA expression in both lung tissues and NET-stimulated endothelial cells. These findings suggest that nerandomilast may elevate cAMP to activate PKA, thereby inhibiting RhoA and its downstream NF-κB signalling, ultimately attenuating EndMT. Targeting the RhoA pathway thus represents a promising strategy for antifibrotic therapy.

Recent evidence indicates that PDE4B inhibitors ameliorate skin and lung fibrosis in a bleomycin-induced SSc-ILD mouse model via modulation of the cAMP and TGF-β1 Smad/non-Smad pathways.²¹ Nerandomilast, a novel PDE4B-selective inhibitor, potently targets the cAMP/PKA axis. By elevating intracellular cAMP levels, nerandomilast activates the cAMP/CREB pathway, thereby suppressing lung fibroblast proliferation and collagen deposition.⁴³ In the present study, nerandomilast downregulated PDE4B expression and reduced phosphorylation of NF-κB, PI3K, AKT and ERK, indicating enhanced CREB activation and concurrent inhibition of key profibrotic signalling cascades. These multi-pathway effects—spanning NF-κB, PI3K/AKT and MAPK—collectively underpin the antifibrotic activity of nerandomilast. Further studies are warranted to delineate the cell-specific actions of nerandomilast and its immunomodulatory role within the inflammatory–fibrotic niche.

A key limitation of our study is the small number of clinical lung tissue samples, as only two patients with IIM-ILD were included. Local sampling complicates the assessment of the extent and severity of lesions throughout the entire lung. Different myositis patients exhibit different pathological patterns in lung tissues, and the same patient can show different patterns at different disease stages. Due to these constraints, clinical tissue samples are scarce. Therefore, the specific expression patterns of PDE4A, PDE4B, PDE4C and PDE4D, and whether RhoA is activated in vivo, have not been fully confirmed within this study. Future research should expand the sample size to validate the findings of this study and enhance the generalisability of the results. Despite confirmed efficacy in an MAILD mouse model, extrapolating PDE4B inhibitors from IPF to IIM-ILD presents distinct challenges. First, disease heterogeneity across myositis subtypes necessitates refined patient stratification in future trials. Second, complex extrapulmonary comorbidities may overlap with gastrointestinal adverse effects, potentially impacting tolerability. Finally, the lack of long-term safety data in IIM-ILD, coupled with frequent immunosuppressant use, requires vigilant monitoring for drug interactions and infection risk.

Conclusion

Nerandomilast, a novel oral and specific PDE4B inhibitor, exerts anti-inflammatory and antifibrotic effects in the MAILD mouse model by activating the cAMP–PKA–RhoA pathway, thereby alleviating EndMT, inhibiting NET formation and suppressing the non-Smad signalling pathway. These findings suggest that nerandomilast holds promise as a therapeutic agent for IIM-ILD. Further exploration of the mechanisms of action and therapeutic potential of PDE4B inhibitors in ILD and pulmonary fibrosis will provide new molecular insights into the pathophysiology of ILD and pave the way for the clinical development of safe and effective treatments for ILD.

Data availability statement

Data are available upon reasonable request.