Dear Editor Acute respiratory distress syndrome (ARDS) is a complex life-threatening syndrome. Injury mechanisms behind ARDS rapid condition change and their association with treatment efficacy remain unclear. Using multitemporal single-cell RNA sequencing (scRNA-seq) on bronchoalveolar lavage fluid (BALF), we provided the first paired human lung immune cell atlas before and after ARDS condition change. We identified IL1βhi neutrophils as dominant in severe ARDS and linked to disease worsening. Neutrophil metabolic reprogramming and interactions with macrophages amplify inflammation, suggesting IL-1 blockade potential. This study included eight severe ARDS patients, comprising four COVID-19-related ARDS (CARDS) and four non-CARDS patients (Table S1), with 23 public CARDS samples as controls. BALF samples were obtained upon admission (t1) and condition change (t2; Figure 1A). ScRNA-seq captured 87044 cells, labelled as major types1 (Figure 1B,C). Compared to GSE145926 CARDS BALF data,1 our CARDS patients exhibited significantly higher neutrophil proportions. Neutrophils were predominant in our results, followed by macrophages (Figure 1D). ΔMurray scores (MS) categorised patients into exacerbation (ΔMS ≥ +1) and remission groups, confirmed by computed tomography (CT; Figure 1E). In CARDS, after treatment, neutrophils significantly decreased and epithelial cells increased in remission groups, while no such change occurred in exacerbated patients (Figure 1F), suggesting neutrophils as important determinants of CARDS progression. In non-CARDS, we detected macrophages significantly increase in exacerbated patients, suggesting macrophages may have more impacts on non-CARDS progression (Figure 1G). To identify neutrophil subtypes related to ARDS progression, we classified neutrophils into progenitor (CD63+), mature (IL1Βhi, S100A12+CXCR2+ and CCL4+), and hybrid neutrophils (CD74+)2 (Figures 2A,B and S1A–F). Neutrophil evolution followed two trajectories: immature—S100A12+CXCR2+ neutrophils and immature—IL1βhi—S100A12+CXCR2+ neutrophils (Figures 2A and S1G,H). IL1βhi neutrophils significantly increased in exacerbated patients compared to baselines (Figure 2C,D), with upregulated inflammatory genes (IL1B, ISG15, and PTGS2) and interferon-related pathways (Figure 2E,F). These genes, like PTGS2, are identified as key genes in SARS-CoV-2 immunopathogenesis.3, 4 IL1βhi neutrophil signatures and IL1B gene can also be used as prognostic indicators for CARDS (Area under curve AUC: .8693, Figure S1I). Comparing our neutrophils with neutrophils from CARDS peripheral blood5 (Figure 2G), peripheral neutrophils rarely express IL1βhi neutrophil signatures, while cells with high expression nearly all come from BALF (Figure 2H). This demonstrates that this progression-related neutrophil group could be unique to BALF, further confirming the uniqueness of our findings. Comparing CARDS and non-CARDS, CARDS patients exhibited higher IL1B, IL1R1, and IL1RAP expression than non-CARDS (Figure 2I), while IL1βhi neutrophils remained low in both non-CARDS remission and exacerbation groups (Figure 2J). Considering non-CARDS cell-type proportions changes (Figure 1F), we think that associations between IL1βhi neutrophils and progression may be unique to CARDS. Next, we explored pro-inflammatory mechanisms of IL1βhi neutrophils. Metabolic analysis indicated IL1βhi neutrophils undergo distinct shifts from glucose to fatty acid metabolism (Figure 3A,B), with elevated lipid-metabolism enzymes (e.g., ACSL1; Figure 3C). Studies have shown the relationship of ACSL1 with IL1β release and neutrophil chemotaxis.6 Transcription factor (TF) analysis revealed upregulation of NFKB2, RELB, and STAT (Figure 3D). Thus, we hypothesise that elevated neutrophil ACSL1 may contribute to lysosomal stress via intracellular lipid accumulation, which could activate inflammatory pathways and facilitate IL1β release. Also, elevated ACSL1 promoted fatty acid oxidation, increased mitochondrial metabolism and ATP production, which also promoted neutrophil differentiation (Figure 3E). To investigate relationships between other cell types and IL1βhi neutrophils, we classified macrophages into subclusters (Figure 4A,B). Highly inflammatory M1 macrophages (IFIT+CXCL10+) increased in exacerbated patients, suggesting their association with CARDS progression (Figures 4C,D and S2A), with upregulation of interferon-related genes (IFIT3, ISG20, ISG15) and located at the end of pseudotime trajectories as a pathogen-responsive state (Figures 4E,F and S2B,C). Fibrosis signatures, which significantly upregulated in immunomodulatory profibrotic M2 and alveolar macrophages, aren't enriched in IFIT+CXCL10+ macrophages (Figure S2D). Metabolism analysis further detected increased oxidative phosphorylation in IFIT+CXCL10+ macrophages (Figure S2E). While prior studies focused on macrophage-derived IL1β,7, 8 our findings suggest IL1β-related neutrophils may have stronger associations with progression. Our analysis revealed strong inflammatory macrophage–neutrophil interactions, with macrophage-derived ligands (particularly IL-6 from IFIT+CXCL10+ macrophages) prominently activating IL1βhi neutrophils (Figure 4G,H). Olink confirmed IL-6 and other proteins related to neutrophil activation were significantly elevated in BALF supernatant and peripheral blood plasma of exacerbated patients (Figures 4I and S2F). Studies have shown that neutrophils can activate STAT3 and NFKB under IL6, thereby mediating high intracellular inflammation.9 Our analysis also suggests these two TFs activate in IL1βhi neutrophils. Confirming IL1B and IL6 cellular sources, we speculate that IFIT+CXCL10+ macrophages act on neutrophils by releasing IL6, causing high inflammation (Figures 4J and S4G,H). Also, considering IL1βhi neutrophils as upstream cells, we found that IL1β/TNF from IL1βhi neutrophils may potentially induce IFIT+CXCL10+ macrophages (Figure S2I,J). IL1β receptors are highly expressed in IFIT+CXCL10+ macrophages (Figure S2J), further suggesting pro-inflammatory positive feedback loops with IL1βhi neutrophils and IFIT+CXCL10+ macrophages, causing lung-injury exacerbation. As all these patients had baseline immunosuppression, we investigated neutrophil–T interactions. Identifying T-cell subgroups (Figure S3A,B) and comparing IL1βhi neutrophil–T and other neutrophil–T interactions, we found IL1βhi neutrophil–T with stronger PDL1/PD1 signalling (Figure S3C). We further confirmed PDL1 mainly expressed in IL1βhi neutrophils and PD1 mainly in Treg (Figure S3D). These indicate that in immunosuppressed CARDS patients, neutrophil-induced immunosuppression potentially accelerates cellular immune decline, warranting further attention. In conclusion, we provided the first multitemporal human ARDS lung immune cell atlas, to our knowledge. Neutrophils dominate BALF samples, with alveolar-specific IL1βhi neutrophils increasing during exacerbation, suggesting links to ARDS progression. These neutrophils express inflammation-related genes and exhibit metabolic signatures suggestive of shifts from glucose to fatty acid metabolism, potentially associated with pro-inflammatory phenotypes. Bioinformatic analysis indicated they may also act as intercellular communication hubs, forming signalling axes with macrophages and T cells. BALF proteomics confirmed the IL-6 signals. IL-1 blockers are found with ability to control inflammation in various diseases, which effect on ARDS also was supported by retrospective cohort studies and mouse models.10 Our findings highlight IL-1 signalling and alveolar inflammatory loops as potential therapeutic targets in severe ARDS, although further validation is required. While our single-cell transcriptomic data suggest associations between alveolar immune changes and CARDS progression, direct experimental validation remains necessary. Future experiments are warranted to assess whether IL-6 or fatty acid exposure can induce IL1βhi neutrophil phenotypes in vitro, and evaluate the therapeutic impact of neutrophil depletion, IL-1 blockade, or metabolic inhibition in murine ARDS models. Yingying Yang, Haochen Li, Peng Liu, Jinmeng Jia, Lei Wei, Xiangyu Chen, Ziqi Tan, Hantian Li, Qianlin Wang, Siyi Yuan, Liangyu Mi, Yuechuan Xue, Yi Chi, Shuaishuai Yang, Yanjie Zhao: Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; Yingying Yang, Haochen Li, Peng Liu, Jinmeng Jia, Lei Wei, Longxiang Su, Yun Long: Drafting the work or reviewing it critically for important intellectual content; Longxiang Su, Yun Long: Final approval of the version to be published; Longxiang Su, Yun Long: Agreement to be accountable for all aspects of the work. We would like to thank the enrolled patients and their families for participating in this research and permitting us to use their specimens for this research. This research was supported by CAMS Innovation Fund for Medical Sciences (CIFMS) from Chinese Academy of Medical Sciences (2023-I2M-2-002-2024-B1) and National High-Level Hospital Clinical Research Funding (2022-PUMCH-D-005, 2025-PUMCH-C-036). We also thank the Clinical Biobank (ISO 20387) at Peking Union Medical College Hospital, Chinese Academy of Medical Sciences, for their assistance in sample collection. The authors declare no conflicts of interest. All patients signed an informed consent for study enrolment. The study was supported by the Ethics Committee of Peking Union Medical College Hospital (ZS-3391). The accession number for the raw and processed sequencing data reported in this paper is Gene Expression Omnibus (GEO): GSE240580. Public data were also collected from GSE145926, GSE157789, and GSE136831. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
Yang et al. (Thu,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: