{"id":14303,"date":"2026-04-19T06:00:00","date_gmt":"2026-04-19T10:00:00","guid":{"rendered":"https:\/\/cov19longhaulfoundation.org\/?p=14303"},"modified":"2026-04-02T09:17:00","modified_gmt":"2026-04-02T13:17:00","slug":"systemic-immune-dysregulation-to-pulmonary-impairment-in-long-covid","status":"publish","type":"post","link":"https:\/\/cov19longhaulfoundation.org\/?p=14303","title":{"rendered":"Systemic immune dysregulation to pulmonary Impairment in long COVID"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li class=\"has-small-font-size\">Saumya Kumar,\u00a0Chaofan Li,\u00a0Liang Zhou,\u00a0et. al.. <a href=\"https:\/\/www.nature.com\/ni\"><em>Nature Immunology<\/em><\/a>\u00a0<strong>volume\u00a027<\/strong>,\u00a0pages200\u2013212 (2026)<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"Abs1\">Abstract<\/h2>\n\n\n\n<p>The mechanisms driving immune dysregulation in long COVID disease remain elusive. Here we integrated single-cell multiome data, immunological profiling and functional assays to investigate immune alterations across multiple cohorts. A transcriptional state in circulating monocytes (LC-Mo) was enriched in individuals with mild\u2013moderate acute infection and accompanied by persistent elevations of plasma CCL2, CXCL11 and TNF. LC-Mo showed TGF\u03b2 and WNT\u2013\u03b2-catenin signaling and correlated with fatigue severity. Protein markers of LC-Mo were increased in individuals with pronounced fatigue or dyspnea, and those with severe respiratory symptoms showed higher LC-Mo expression. Epigenetically, LC-Mo exhibited AP-1- and NF-\u03baB1-driven profibrotic programs. LC-Mo-like macrophages in bronchoalveolar lavage samples from individuals with severe respiratory symptoms displayed a profibrotic profile, and individuals with a high LC-Mo transcriptional state showed impaired interferon responses after stimulation. Collectively, our findings define a pathogenic monocyte transcriptional state linking systemic immune dysfunction to persistent long COVID disease, providing mechanistic insights and potential therapeutic targets.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"Sec1\">Main<\/h2>\n\n\n\n<p>Long COVID affects 10\u201320% of individuals after severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, with symptoms ranging from mild discomfort to severe, long-lasting impairments such as fatigue, respiratory issues and neurological problems. These symptoms can persist for over 3 years (refs.&nbsp;<sup><a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1?utm_source=chatgpt.com#ref-CR1\">1<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1?utm_source=chatgpt.com#ref-CR2\">2<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1?utm_source=chatgpt.com#ref-CR3\">3<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1?utm_source=chatgpt.com#ref-CR4\">4<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR5\">5<\/a><\/sup>), representing a substantial health burden and prompting efforts to better characterize long COVID (LC), including biomarker discovery for improved diagnosis<sup><a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1?utm_source=chatgpt.com#ref-CR6\">6<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1?utm_source=chatgpt.com#ref-CR7\">7<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1?utm_source=chatgpt.com#ref-CR8\">8<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1?utm_source=chatgpt.com#ref-CR9\">9<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR10\">10<\/a><\/sup>.<\/p>\n\n\n\n<p>LC presents with diverse symptoms reflecting multiorgan system abnormalities<sup><a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1?utm_source=chatgpt.com#ref-CR11\">11<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1?utm_source=chatgpt.com#ref-CR12\">12<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR13\">13<\/a><\/sup>. The evidence suggests multiple possible causes, including persistence of viral remnants or reactivation of latent viruses<sup><a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR7\">7<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1?utm_source=chatgpt.com#ref-CR14\">14<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1?utm_source=chatgpt.com#ref-CR15\">15<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1?utm_source=chatgpt.com#ref-CR16\">16<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR17\">17<\/a><\/sup>. Yet, persistent immune dysregulation is a consistent finding in LC studies<sup><a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR10\">10<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR11\">11<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR14\">14<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1?utm_source=chatgpt.com#ref-CR16\">16<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1?utm_source=chatgpt.com#ref-CR17\">17<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1?utm_source=chatgpt.com#ref-CR18\">18<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR19\">19<\/a><\/sup>. Although most LC cases follow mild-to-moderate acute illness, many studies do not stratify individuals by acute infection (AI) severity<sup><a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1?utm_source=chatgpt.com#ref-CR6\">6<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1?utm_source=chatgpt.com#ref-CR7\">7<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR8\">8<\/a><\/sup>, which is crucial because severe cases, especially those treated in the intensive care unit, develop immune changes due to intensive medical interventions<sup><a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR20\">20<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR21\">21<\/a><\/sup>. Failing to account for these differences may confound LC-associated molecular signatures, highlighting the importance of refined patient grouping.<\/p>\n\n\n\n<p>To address this gap, we stratified individuals with LC by acute COVID-19 severity to better resolve immune heterogeneity and identify molecular features underlying chronic symptoms. We applied single-cell multiomics profiling of peripheral blood mononuclear cells (PBMCs) and measured plasma cytokines from individuals with LC with fatigue and respiratory symptoms using longitudinal and cross-sectional samples. We identified a distinct circulating CD14\u207a monocyte state associated with LC (\u2018LC-Mo\u2019), which was enriched in individuals with mild-to-moderate AI. This state coincided with persistent elevation of circulating cytokines, indicating systemic inflammation. In two independent cohorts of individuals with LC with severe respiratory symptoms and abnormal lung function, LC-Mo expression was increased in circulating CD14\u207a monocyte subsets. In bronchoalveolar lavage (BAL) myeloid cells from individuals with severe respiratory symptoms, LC-Mo-like macrophages showed a profibrotic gene expression profile. Functionally, CD14\u207a monocytes from individuals with LC-Mo enrichment showed dysregulated responses to ex vivo stimulation, indicating impaired immune regulation. Together, these findings provide systemic insight into the cellular and molecular basis of LC and highlight potential therapeutic targets.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"Sec2\">Results<\/h2>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"Sec3\">LC has a distinct transcriptome after mild or moderate disease<\/h3>\n\n\n\n<p>Individuals presenting with headache, dyspnea or fatigue to the pneumology outpatient clinic at Hannover Medical School (MHH) were recruited according to the German S1 guidelines<sup><a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR22\">22<\/a><\/sup>&nbsp;and the Delphi Consensus Criteria<sup><a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR21\">21<\/a><\/sup>&nbsp;for LC (4\u201312 weeks) and post-COVID-19 syndrome (&gt;12 weeks). These criteria included symptoms persisting beyond the acute phase of SARS-CoV-2 infection or its treatment, new symptoms emerging after recovery and attributed to prior infection or worsening of pre-existing conditions. Because heterogeneity in LC molecular profiles may be shaped by acute disease severity and treatment, we stratified individuals with acute SARS-CoV-2 infection (AI) and LC into those with mild-to-moderate (WHO score of 1\u20135) AI (AI<sup>M<\/sup>&nbsp;and LC<sup>AM<\/sup>) and those with severe (WHO scores 6\u20139) AI (AI<sup>S<\/sup>&nbsp;and LC<sup>AS<\/sup>).<\/p>\n\n\n\n<p>Cohort 1 included 45 individuals recruited between April 2020 and August 2021 at MHH, of which 9 gave longitudinal samples and 36 gave cross-sectional samples (<em>n<\/em>\u2009=\u200978 total samples), including 11 donors with AI categorized as AI<sup>M<\/sup>&nbsp;(<em>n<\/em>\u2009=\u20097 donors, 42.8% women, median age\u2009=\u200952, range 23\u201366 years of age, WHO score range 1\u20135) and AI<sup>S<\/sup>&nbsp;(<em>n<\/em>\u2009=\u20094 donors, 50% women, median age\u2009=\u200937, range 32\u201354, WHO score range 6\u20139), 37 donors with LC categorized as LC<sup>AM<\/sup>&nbsp;(<em>n<\/em>\u2009=\u200929 donors, 8 longitudinal donors with two to three time points and 21 single-time-point donors, 58% women, median age\u2009=\u200949, range 31\u201384 years) and LC<sup>AS<\/sup>&nbsp;(<em>n<\/em>\u2009=\u20098 donors, 3 with two to four time points, 5 single-time-point donors, 25% women, median age\u2009=\u200946, range 19\u201375) and 8 donors who had recovered after 4\u20138 months of LC (R<sup>LC<\/sup>; 1 longitudinal donor with two time points and 7 single-time-point donors, 37.5% women, median age\u2009=\u200938, range 19\u201365), in addition to 6 prepandemic noninfected control individuals (NI; 50% women, median age\u2009=\u200940, range 24\u201361). LC and R<sup>LC<\/sup>&nbsp;samples were collected 1.7\u201310.2 months after infection. Cohort 2 included 117 LC<sup>AM<\/sup>&nbsp;donors (24 donors with two to four time points, 93 single-time-point donors, 58.9% women, median age\u2009=\u200948, range 19\u201383) and 25 LC<sup>AS<\/sup>&nbsp;donors (12 longitudinal donors, 13 single-time-point donors, 20% women, median age\u2009=\u200953, range 18\u201381), recruited between May 2020 and August 2021 at MHH, along with 33 prepandemic NI samples (48.4% women, median age\u2009=\u200940, range 25\u201365). Cohort 3 included only LC<sup>AM<\/sup>&nbsp;donors (<em>n<\/em>\u2009=\u20098 donors, 62.5% women, median age\u2009=\u200945, range 21\u201363), all with respiratory postacute sequelae of SARS-CoV-2 infection (PASC) recruited between October and November 2023 at the Pulmonary Rehabilitation Clinic in Sch\u00f6nau am K\u00f6nigssee, Germany. Cohort 4 included LC<sup>AM<\/sup>&nbsp;donors (<em>n<\/em>\u2009=\u200929, 58.6% women, median age\u2009=\u200949, range 33\u201372), LC<sup>AS<\/sup>&nbsp;donors (<em>n<\/em>\u2009=\u200911 donors, 18% female, median age\u2009=\u200957, range 35\u201381), 8 donors recovered from AI (R<sup>A<\/sup>) and 2 NI donors (60% women, median age\u2009=\u200941, range 29\u201367) recruited between August 2020 and June 2022 at MHH. Cohort 5 included LC donors (<em>n<\/em>\u2009=\u20099 donors, 44.4% women, median age\u2009=\u200964, range 62\u201383, including 5 with respiratory PASC) and NI donors (<em>n<\/em>\u2009=\u20092 donors, 50% women, median age\u2009=\u200977, range 73\u201377), recruited between October 2020 and November 2021 at Mayo Clinic, a previously published study<sup><a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR23\">23<\/a><\/sup>&nbsp;(Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig1\">1a<\/a>&nbsp;and&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Sec11\">Methods<\/a>).<\/p>\n\n\n\n<figure class=\"wp-block-image\"><a class=\"c-article-section__figure-link\" href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1\/figures\/1\"><img decoding=\"async\" src=\"https:\/\/media.springernature.com\/lw685\/springer-static\/image\/art%3A10.1038%2Fs41590-025-02387-1\/MediaObjects\/41590_2025_2387_Fig1_HTML.png\" alt=\"figure 1\"\/><\/a><figcaption class=\"wp-element-caption\"><strong>Fig. 1: Transcriptomes of circulating immune cells show heterogeneity in individuals with LC.<\/strong><\/figcaption><\/figure>\n\n\n\n<p>Clinical assessment included blood gas analysis, pulmonary function tests and standardized participant-reported outcome measures: the fatigue assessment scale (FAS), validated in chronic fatigue<sup><a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1?utm_source=chatgpt.com#ref-CR24\">24<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1?utm_source=chatgpt.com#ref-CR25\">25<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR26\">26<\/a><\/sup>&nbsp;and LC, and the modified medical Research Council (mMRC) dyspnea scale (0\u20134, where 0 indicates no breathlessness, 1 indicates breathlessness on exertion, 2 indicates breathlessness when hurrying or walking uphill, 3 indicates stopping for breath after ~100\u2009m or a few minutes, and 4 indicates too breathless to leave the house or when dressing), along with quality-of-life metrics<sup><a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR27\">27<\/a><\/sup>. All clinical assessment data were systematically collected at each participant visit for cohorts 1\u20134 (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig1\">1b<\/a>&nbsp;and Supplementary Tables&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#MOESM1\">1<\/a>\u2013<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#MOESM1\">5<\/a>).<\/p>\n\n\n\n<p>To study molecular signatures of disease progression, we stratified samples in cohorts 1 and 2 by months since AI (months 1.5\/1.7\u20132.9, 3\u20135.9, 6\u20138.9 and 9\u201311; Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig1\">1c<\/a>&nbsp;and&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Sec11\">Methods<\/a>). For cohort 1, we generated single-nucleus RNA-sequencing (snRNA-seq) and single-nucleus assay for transposase-accessible chromatin with sequencing (snATAC-seq) data from 78 PBMC samples from NI, R<sup>LC<\/sup>, AI<sup>M<\/sup>, AI<sup>S<\/sup>, LC<sup>AM<\/sup>&nbsp;and LC<sup>AS<\/sup>&nbsp;donors across all time points. In cohort 2 we measured the concentrations of 14 cytokines in plasma samples from LC<sup>AM<\/sup>&nbsp;or LC<sup>AS<\/sup>&nbsp;and NI donors across all time points. Validation was performed using single-cell RNA-sequencing (scRNA-seq; cohort 3), flow cytometry (cohort 4) and a published PBMC\/BAL single-cell dataset<sup><a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR23\">23<\/a><\/sup>&nbsp;(cohort 5). All samples, except those from participants with AI, were PCR negative at collection. We used an integrative multistep analysis to identify cell-type-specific immune dysregulation and link and assess relevance in LC (Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig8\">1a<\/a>).<\/p>\n\n\n\n<p>Analysis of single-cell data from cohort 1 PBMCs yielded ~118,000 high-quality cells (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig1\">1d<\/a>). snRNA-seq data showed distinct patterns in LC<sup>AM<\/sup>&nbsp;and LC<sup>AS<\/sup>&nbsp;compared to R<sup>LC<\/sup>&nbsp;and AI across major PBMCs (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig1\">1e<\/a>). LC<sup>AM<\/sup>&nbsp;showed downregulated AI genes by months 6\u20138.9, whereas LC<sup>AS<\/sup>&nbsp;retained an acute COVID-19-like transcriptomic profile, indicating heterogeneity based on AI history (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig1\">1e<\/a>). Differential gene expression (DGE) analysis identified 1,737 upregulated genes in CD14<sup>+<\/sup>&nbsp;monocytes from LC<sup>AM<\/sup>&nbsp;donors compared to those from AI and R<sup>LC<\/sup>&nbsp;(Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig1\">1e<\/a>), with upregulation over 1.7\u20138.9 months, and showed participant-specific heterogeneity (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig1\">1e<\/a>&nbsp;and Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig8\">1b<\/a>). LC<sup>AM<\/sup>&nbsp;CD14<sup>+<\/sup>&nbsp;monocytes showed persistent upregulation of proinflammatory (<em>CSF1<\/em>,&nbsp;<em>IRF8<\/em>,&nbsp;<em>RELA<\/em>&nbsp;and&nbsp;<em>NOTCH1<\/em>) and anti-inflammatory (<em>TGFB1<\/em>,&nbsp;<em>SMADs<\/em>,&nbsp;<em>ENG<\/em>&nbsp;and&nbsp;<em>SERPINE1<\/em>) markers (Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig8\">1c<\/a>) at all time points, whereas other signature genes showed increased expression from 3 to 8.9 months (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig1\">1e<\/a>). This signature diminished during months 9\u201311, possibly due to lower cell numbers (Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig8\">1d<\/a>), but showed upregulation of a subset of acute-phase genes, including&nbsp;<em>IL1B<\/em>,&nbsp;<em>S100A4<\/em>,&nbsp;<em>PDIA3<\/em>&nbsp;and&nbsp;<em>MTRNR2L1<\/em>. LC<sup>AM<\/sup>&nbsp;natural killer (NK) cells also showed distinct increased expression of&nbsp;<em>SREBF1<\/em>,&nbsp;<em>TAGLN2<\/em>,&nbsp;<em>TNIP1<\/em>,&nbsp;<em>NFKBIA<\/em>&nbsp;and&nbsp;<em>CD83<\/em>&nbsp;among others compared to R<sup>LC<\/sup>&nbsp;and AI NK cells (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig1\">1e<\/a>). Collectively, transcriptional profiles in individuals with LC reflected differences based on AI severity, with notable molecular changes in LC<sup>AM<\/sup>&nbsp;monocytes and NK cells, whereas LC<sup>AS<\/sup>&nbsp;displayed persistent but milder expression of acute-phase genes.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"Sec4\">TNF and TNF signaling genes are upregulated in LC<sup>AM<\/sup><\/h3>\n\n\n\n<p>We next performed gene set enrichment analysis (GSEA) using pseudobulk counts for each cell subset in LC<sup>AM<\/sup>&nbsp;or LC<sup>AS<\/sup>&nbsp;samples across all time points, comparing them to the AI and R<sup>LC<\/sup>&nbsp;cell samples. LC<sup>AM<\/sup>&nbsp;showed persistent upregulation of the TNF signaling pathway and persistent downregulation of interferon (IFN) signaling and response pathways across all major cell subsets (CD4<sup>+<\/sup>&nbsp;and CD8<sup>+<\/sup>&nbsp;T cells, B cells and CD14<sup>+<\/sup>&nbsp;and CD16<sup>+<\/sup>&nbsp;monocytes) compared to AI, up to month 8.9 (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig2\">2a<\/a>&nbsp;and Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig9\">2a<\/a>). CD8<sup>+<\/sup>&nbsp;T cells and NK cells from LC<sup>AM<\/sup>&nbsp;samples exhibited increased activation of the \u2018TLR signaling cascades\u2019 pathway relative to R<sup>LC<\/sup>&nbsp;samples at months 3\u20138.9 (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig2\">2a<\/a>). In LC<sup>AM<\/sup>&nbsp;CD14<sup>+<\/sup>&nbsp;monocytes, the TNF signaling pathway was transiently upregulated at months 1.7\u20135.9 and downregulated at months 6\u20138.9, whereas pathways including transforming growth factor-\u03b2 (TGF\u03b2), WNT\u2013\u03b2-catenin and Notch signaling were upregulated at months 3\u20138.9 compared to in AI and R<sup>LC<\/sup>&nbsp;CD14<sup>+<\/sup>&nbsp;monocytes (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig2\">2a<\/a>). In LC<sup>AS<\/sup>, the TNF signaling pathway was sparsely activated in CD14<sup>+<\/sup>&nbsp;monocytes and CD8<sup>+<\/sup>&nbsp;T cells up to 5.9 months (Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig9\">2b<\/a>). LC<sup>AS<\/sup>&nbsp;CD14<sup>+<\/sup>&nbsp;monocytes upregulated PD-1 signaling and MHC class II antigen presentation pathways compared to AI, but not R<sup>LC<\/sup>&nbsp;(Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig9\">2b<\/a>, top). CD8<sup>+<\/sup>&nbsp;and CD4<sup>+<\/sup>&nbsp;T cells and NK cells from LC<sup>AS<\/sup>&nbsp;samples displayed increased activation of IFN response pathways compared to CD8<sup>+<\/sup>&nbsp;and CD4<sup>+<\/sup>&nbsp;T cells and NK cells from R<sup>LC<\/sup>&nbsp;samples (Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig9\">2b<\/a>).<\/p>\n\n\n\n<figure class=\"wp-block-image\"><a class=\"c-article-section__figure-link\" href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1\/figures\/2\"><img decoding=\"async\" src=\"https:\/\/media.springernature.com\/lw685\/springer-static\/image\/art%3A10.1038%2Fs41590-025-02387-1\/MediaObjects\/41590_2025_2387_Fig2_HTML.png\" alt=\"figure 2\"\/><\/a><figcaption class=\"wp-element-caption\"><strong>Fig. 2: TNF and inflammatory pathways in circulating immune cells indicate systemic inflammation in LC.<\/strong><\/figcaption><\/figure>\n\n\n\n<p>We also profiled 14 proinflammatory cytokines in cohort 2 plasma using a multiplex bead-based assay (Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig9\">2c<\/a>), excluding interleukin-4 (IL-4) and IL-5 due to low detection. CXCL11, CCL2 and TNF were persistently elevated in individuals with LC compared to in NI donors up to month 9 (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig2\">2b<\/a>).&nbsp;<em>TNF<\/em>&nbsp;mRNA was also persistently upregulated in individuals with LC<sup>AM<\/sup>&nbsp;across most immune cell types and time points (Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig9\">2d<\/a>). TNF protein exhibited a statistically significant negative correlation with arterial oxygenation (pO<sub>2<\/sub>) in individuals with LC (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig2\">2c<\/a>), which remained statistically significant in LC<sup>AM<\/sup>, but not in LC<sup>AS<\/sup>, up to month 8.9 (Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig10\">3a<\/a>). No other cytokines showed consistent correlations across all time points (Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig10\">3b,c<\/a>).<\/p>\n\n\n\n<p>Correlation analysis between key pathways upregulated in CD8<sup>+<\/sup>&nbsp;T cells, NK cells and CD14<sup>+<\/sup>&nbsp;monocytes and FAS scores indicated that TGF\u03b2 and WNT\u2013\u03b2-catenin signaling in CD14<sup>+<\/sup>&nbsp;monocytes showed modest positive correlations with FAS scores in LC alone and stronger correlations when LC and R<sup>LC<\/sup>&nbsp;were combined (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig2\">2d<\/a>). IFN\u03b1\/IFN\u03b2 induction pathways positively correlated with FAS scores in CD8<sup>+<\/sup>&nbsp;T cells and NK cells in both LC only or LC\u2009+\u2009R<sup>LC<\/sup>&nbsp;combined analyses (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig2\">2d<\/a>). WNT\u2013\u03b2-catenin signaling in CD8<sup>+<\/sup>&nbsp;T cells and Toll-like receptor (TLR) signaling cascades in NK cells, but not TNF signaling in these cells, also correlated with FAS scores (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig2\">2d<\/a>). These results indicate that persistent upregulation of inflammatory pathways and cytokines in LC<sup>AM<\/sup>&nbsp;immune cells might contribute to the clinical symptoms in LC.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"Sec5\">The LC<sup>AM<\/sup>&nbsp;monocyte signature characterizes a transcriptional state<\/h3>\n\n\n\n<p>Next, we performed a reclustering analysis of CD8<sup>+<\/sup>&nbsp;T cells, NK cells and CD14<sup>+<\/sup>&nbsp;monocytes from all donor samples. CD8<sup>+<\/sup>&nbsp;T cells and NK cells each resolved into five clusters (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig3\">3a,b<\/a>). Differential neighborhood abundance analysis (DA) comparing LC groups at each time point to AI and R<sup>LC<\/sup>&nbsp;was performed. A neighborhood defines a small local group of cells with similar gene expression profiles, representing transitional states. LC<sup>AM<\/sup>&nbsp;samples exhibited statistically significant increased abundance of neighborhoods in&nbsp;<em>CD69<\/em><sup>hi<\/sup><em>CD27<\/em><sup>hi<\/sup>&nbsp;CD8<sup>+<\/sup>&nbsp;T cells (C3),&nbsp;<em>GZMB<\/em><sup>+<\/sup><em>KLRF1<\/em><sup>+<\/sup>&nbsp;NK cells (C1) and&nbsp;<em>CD69<\/em><sup>+<\/sup><em>TGFB1<\/em><sup>+<\/sup>&nbsp;NK cells (C2; Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig11\">4a,b<\/a>), whereas LC<sup>AS<\/sup>&nbsp;samples showed increased abundance of neighborhoods in C2 NK cells at months 6\u20138.9 (Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig11\">4c,d<\/a>). C3 CD8<sup>+<\/sup>&nbsp;T cells and C2 NK cells showed&nbsp;<em>GZMK<\/em><sup>+<\/sup><em>GZMB<\/em><sup>lo<\/sup>&nbsp;signatures (Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig11\">4e,f<\/a>), reported to accumulate after SARS-CoV-2 infection and in aging<sup><a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1?utm_source=chatgpt.com#ref-CR28\">28<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1?utm_source=chatgpt.com#ref-CR29\">29<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR30\">30<\/a><\/sup>. These clusters showed higher expression of TNF and TLR signaling genes (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig3\">3a,b<\/a>), suggesting the contribution of persistent TNF signaling in the expansion of&nbsp;<em>CD69<\/em><sup>hi<\/sup><em>CD27<\/em><sup>hi<\/sup><em>GZMK<\/em><sup>+<\/sup>&nbsp;CD8<sup>+<\/sup>&nbsp;T cells and&nbsp;<em>CD69<\/em><sup>+<\/sup><em>TGFB1<\/em><sup>+<\/sup><em>GZMK<\/em><sup>+<\/sup>&nbsp;NK cells in individuals with LC<sup>AM<\/sup>.<\/p>\n\n\n\n<figure class=\"wp-block-image\"><a class=\"c-article-section__figure-link\" href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1\/figures\/3\"><img decoding=\"async\" src=\"https:\/\/media.springernature.com\/lw685\/springer-static\/image\/art%3A10.1038%2Fs41590-025-02387-1\/MediaObjects\/41590_2025_2387_Fig3_HTML.png\" alt=\"figure 3\"\/><\/a><figcaption class=\"wp-element-caption\"><strong>Fig. 3: Distinct cell subclusters drive LC signatures in NK cells, CD8<sup>+<\/sup>&nbsp;T cells and CD14<sup>+<\/sup>&nbsp;monocytes.<\/strong><\/figcaption><\/figure>\n\n\n\n<p>Within CD14<sup>+<\/sup>&nbsp;monocytes, four primary clusters (MC1\u2013MC4) were identified (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig3\">3c<\/a>). MC1 showed high expression of MHC class II molecules,&nbsp;<em>IL1B<\/em>&nbsp;and&nbsp;<em>NFKB1<\/em>; MC2 showed elevated&nbsp;<em>NFKB1<\/em>&nbsp;and&nbsp;<em>S100A4<\/em>; MC3 showed increased expression of&nbsp;<em>FCN1<\/em>, IFN-stimulated genes (<em>IFI44<\/em>,&nbsp;<em>IFI16<\/em>&nbsp;and&nbsp;<em>IFI30<\/em>) and alarmins&nbsp;<em>S100A8<\/em>&nbsp;and&nbsp;<em>S100A9<\/em>; and MC4 displayed higher levels of&nbsp;<em>IRF1<\/em>,&nbsp;<em>IRF8<\/em>,&nbsp;<em>TGFB1<\/em>,&nbsp;<em>CTNNB1<\/em>,&nbsp;<em>ENG<\/em>&nbsp;and&nbsp;<em>NOTCH1<\/em>, among others (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig3\">3d<\/a>). DA comparing LC samples with AI and R<sup>LC<\/sup>&nbsp;samples across all time points showed a consistent significant increase in MC4 neighborhoods in LC<sup>AM<\/sup>&nbsp;in both men and women (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig3\">3e<\/a>&nbsp;and Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig11\">4g<\/a>), with this becoming prominent from month 3 onward (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig3\">3e<\/a>&nbsp;and Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig12\">5a<\/a>). By contrast, MC1 neighborhoods showed a marked increase, primarily at months 1.7\u20132.9, and \u2018tapering off\u2019 by month 11 (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig3\">3e<\/a>). LC<sup>AS<\/sup>&nbsp;did not exhibit consistent changes in MC4, except for a small number of neighborhoods at months 6\u20138.9 attributable to one participant (Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig12\">5b<\/a>). Further, area under the curve (AUC) scores of pathways (calculated per cell from all donors) revealed that MC4, which was uniquely abundant in LC<sup>AM<\/sup>, showed significantly higher expression of TGF\u03b2 and WNT\u2013\u03b2-catenin signaling genes than MC1, MC2 and MC3 (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig3\">3f<\/a>). MC1 showed higher expression of the TNF signaling genes (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig3\">3f<\/a>), whereas MC1 and MC3 showed higher IFN\u03b3 response gene expression (Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig12\">5c<\/a>). We further performed trajectory analysis (unstratified by disease category or groups) that revealed that lineage 3 overlapped closely with the MC4 immune program (Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig12\">5d,e<\/a>), indicating that MC4 cells in LC<sup>AM<\/sup>&nbsp;have a distinct transcriptional profile compared to MC1\u2013MC3. We next assessed the correlation between the frequency of MC4 within CD14<sup>+<\/sup>&nbsp;monocytes for all LC and R<sup>LC<\/sup>&nbsp;samples from all time points with clinical parameters. A modest but statistically significant positive correlation was found between MC4 proportion and FAS score, whereas the correlation with pO<sub>2<\/sub>&nbsp;was negative (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig3\">3g<\/a>). By contrast, a higher MC1 proportion was negatively correlated with FAS score (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig3\">3g<\/a>). The modest MC4\u2013FAS correlation likely reflected participant heterogeneity (Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig12\">5f<\/a>). Individuals with LC with a high proportion of MC4 (MC4<sup>hi<\/sup>) exhibited significantly greater fatigue than those with LC with a low proportion of MC4 (MC4<sup>lo<\/sup>) or R<sup>LC<\/sup>&nbsp;(Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig3\">3h<\/a>). These findings indicate that increased MC4 abundance (referred to hereafter as LC monocyte transcriptional state (LC-Mo state)) is associated with LC, as demonstrated by its correlation with both FAS scores and pO<sub>2<\/sub>&nbsp;levels.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"Sec6\">LC monocytes exhibit increased LC-Mo protein marker expression<\/h3>\n\n\n\n<p>To validate the LC-Mo state, we generated and analyzed scRNA-seq data from PBMCs from cohort 3, comprising eight individuals with LC<sup>AM<\/sup>&nbsp;with LC symptoms reported for 8\u201342 months at the time of sampling (Supplementary Table&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#MOESM1\">3<\/a>&nbsp;and&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Sec11\">Methods<\/a>). All individuals with LC reported fatigue and dyspnea (classified as respiratory PASC (\u2018Resp-PASC\u2019),&nbsp;<em>n<\/em>\u2009=\u20095), and three exhibited bronchial hyper-responsiveness (BHR)<sup><a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR31\">31<\/a><\/sup>, termed \u2018Resp-PASC-BHR\u2019 (<em>n<\/em>\u2009=\u20093). Three clusters (Clust0\u2013Clust2) were identified within CD14<sup>+<\/sup>&nbsp;monocytes (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig4\">4a<\/a>). Clust1 showed significantly elevated AUC scores for the LC-Mo signature (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig4\">4b<\/a>). Individuals with Resp-PASC-BHR showed significantly higher expression of the LC-Mo signature in Clust1 than those with Resp-PASC (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig4\">4c<\/a>), providing independent validation of the LC-Mo state in LC<sup>AM<\/sup>&nbsp;and suggesting a link with progression to severe respiratory PASC.<\/p>\n\n\n\n<figure class=\"wp-block-image\"><a class=\"c-article-section__figure-link\" href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1\/figures\/4\"><img decoding=\"async\" src=\"https:\/\/media.springernature.com\/lw685\/springer-static\/image\/art%3A10.1038%2Fs41590-025-02387-1\/MediaObjects\/41590_2025_2387_Fig4_HTML.png\" alt=\"figure 4\"\/><\/a><figcaption class=\"wp-element-caption\"><strong>Fig. 4: LC-Mo-specific proteins show elevated expression in LC CD14<sup>+<\/sup>&nbsp;monocytes.<\/strong><\/figcaption><\/figure>\n\n\n\n<p>We next performed flow cytometry analysis on PBMCs from donors in cohort 4, which included 40 LC samples 3\u201314 months after acute COVID-19 (Supplementary Table&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#MOESM1\">4<\/a>&nbsp;and&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Sec11\">Methods<\/a>) and 10 R<sup>A<\/sup>&nbsp;or NI donors. LC showed a significant increase in CD14<sup>+<\/sup>&nbsp;monocyte percentages compared to R<sup>A<\/sup>\u2009+\u2009NI (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig4\">4d<\/a>), independent of acute COVID-19 severity. We assessed the expression of 11 proteins (HLA-DR, HLA-DQ, CD105, CD51, TGF\u03b21, CD99, CD120b, CALR, IRF8, IFNGR1 and CD163) corresponding to LC-Mo transcripts in total CD14<sup>+<\/sup>&nbsp;monocytes in samples from individuals with LC and R<sup>A<\/sup>\u2009+\u2009NI (Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig13\">6a<\/a>). Median fluorescence intensity (MFI) of HLA-DQ, CD120b, CALR, CD99 and TGF\u03b21 was significantly higher in LC compared to in R<sup>A<\/sup>\u2009+\u2009NI (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig4\">4e<\/a>&nbsp;and Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig13\">6b<\/a>), whereas HLA-DR, CD51, CD105, IRF8, IFNGR1 and CD163 showed no significant difference (Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig13\">6c<\/a>). Stratification by fatigue scores and dyspnea (range 1\u20133) revealed consistently higher MFI of CALR, CD120b, HLA-DQ and TGF\u03b21 in those with more severe LC symptoms (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig4\">4f,g<\/a>), and TGF\u03b21 MFI inversely correlated with pO<sub>2<\/sub>&nbsp;(Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig13\">6d<\/a>, top). MFI of both TGF\u03b21 and IRF8 positively correlated with each other (Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig13\">6d<\/a>, bottom). Thus, protein markers specific to LC-Mo were elevated in LC, supporting an association between the LC-Mo signature and LC pathology.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"Sec7\">Chromatin profiling reveals AP-1\/NF-\u03baB1 activity in LC-Mo<\/h3>\n\n\n\n<p>We next investigated epigenetic regulation using snATAC-seq data from individuals with LC in cohort 1. Examination of motif signals in the chromatin landscapes of CD14<sup>+<\/sup>&nbsp;monocytes, CD8<sup>+<\/sup>&nbsp;T cells and NK cells from individuals with LC<sup>AM<\/sup>&nbsp;compared to those with R<sup>LC<\/sup>&nbsp;at multiple time points identified a persistent positive signal for AP-1 family activity in CD8<sup>+<\/sup>&nbsp;T cells and NK cells (Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig14\">7a<\/a>). In CD14<sup>+<\/sup>&nbsp;monocytes, AP-1 motif accessibility was elevated up to month 5.9, after which motif enrichment shifted toward transcription factors involved in downstream TGF\u03b2 signaling, notably SP1 and KLF family of transcription factors at months 3\u20138.9 (Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig14\">7b<\/a>). MC4 showed the highest number of differentially accessible regions (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig5\">5a<\/a>). The open chromatin landscape of MC4 showed highest enrichment for motifs for ETS family transcription factors, including GABPA, ETV1, ETV4, SPI1 and SPIC (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig5\">5b<\/a>). Correlating open chromatin regions with gene expression revealed significant positive associations for proangiogenic and cell adhesion genes (<em>VEGFA<\/em>,&nbsp;<em>ENG<\/em>,&nbsp;<em>TGFB1<\/em>,&nbsp;<em>RXRA<\/em>,&nbsp;<em>ICAM1<\/em>&nbsp;and&nbsp;<em>ITGA5<\/em>) and genes implicated in inflammatory\/metabolic diseases (<em>TTC7A<\/em>,&nbsp;<em>LMNA<\/em>&nbsp;and&nbsp;<em>IER3<\/em>) among others (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig5\">5c<\/a>). AP-1 family, SMADs, NF-\u03baB1 and RELA transcription factor motifs showed a marked increase within the accessible chromatin regions of these genes (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig5\">5d<\/a>). Within MC4, correlation of transcription factor transcripts and target gene transcripts with accessible motifs enabled pinpointing of noncoding regulatory regions associated with gene expression, such as those for&nbsp;<em>IER3<\/em>&nbsp;and&nbsp;<em>LMNA<\/em>, and establishment of gene\u2013transcription factor relationships (such as NF-\u03baB1 and AP-1 family likely regulators of&nbsp;<em>LMNA<\/em>; Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig5\">5e\u2013g<\/a>&nbsp;and Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig14\">7c<\/a>). In summary, these findings indicate that LC-Mo is driven by ETS, AP-1 and NF-\u03baB1 transcription factors.<\/p>\n\n\n\n<figure class=\"wp-block-image\"><a class=\"c-article-section__figure-link\" href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1\/figures\/5\"><img decoding=\"async\" src=\"https:\/\/media.springernature.com\/lw685\/springer-static\/image\/art%3A10.1038%2Fs41590-025-02387-1\/MediaObjects\/41590_2025_2387_Fig5_HTML.png\" alt=\"figure 5\"\/><\/a><figcaption class=\"wp-element-caption\"><strong>Fig. 5: AP-1 and NF-\u03baB1 transcription factors regulate LC-Mo in CD14<sup>+<\/sup>&nbsp;monocytes from individuals with LC<sup>AM<\/sup><em>.<\/em><\/strong><\/figcaption><\/figure>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"Sec8\">BAL myeloid cells show LC-Mo and profibrotic programs<\/h3>\n\n\n\n<p>Circulating monocytes contribute to PASC pathogenesis, particularly pulmonary fibrosis<sup><a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR23\">23<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR32\">32<\/a><\/sup>. To assess whether LC-Mo participates in fibrotic lung remodeling, we analyzed paired PBMC and BAL fluid samples from a public dataset<sup><a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR23\">23<\/a><\/sup>&nbsp;(cohort 5) consisting of nine individuals with LC of unknown severity during AI (LC<sup>UN<\/sup>), classified based on lung function as Resp-PASC (<em>n<\/em>\u2009=\u20095) or nonResp-PASC (<em>n<\/em>\u2009=\u20094), and PBMCs from NI donors (<em>n<\/em>\u2009=\u20092; Supplementary Table&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#MOESM1\">5<\/a>). Circulating CD14<sup>+<\/sup>&nbsp;myeloid cells were reclustered to identify CD14<sup>+<\/sup>CD16<sup>\u2212<\/sup>&nbsp;monocytes (Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig15\">8a<\/a>), leading to six clusters (CL0\u2013CL5; Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig6\">6a<\/a>). CL5 showed the highest enrichment of LC-Mo signature AUC scores (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig6\">6b<\/a>&nbsp;and Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig15\">8b<\/a>). Within cluster 5, Resp-PASC exhibited significantly higher LC-Mo expression than nonResp-PASC or NI (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig6\">6c<\/a>).<\/p>\n\n\n\n<figure class=\"wp-block-image\"><a class=\"c-article-section__figure-link\" href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1\/figures\/6\"><img decoding=\"async\" src=\"https:\/\/media.springernature.com\/lw685\/springer-static\/image\/art%3A10.1038%2Fs41590-025-02387-1\/MediaObjects\/41590_2025_2387_Fig6_HTML.png\" alt=\"figure 6\"\/><\/a><figcaption class=\"wp-element-caption\"><strong>Fig. 6: The LC-Mo cluster is enriched in profibrotic monocyte-derived alveolar macrophages in BAL fluid from individuals with LC<em>.<\/em><\/strong><\/figcaption><\/figure>\n\n\n\n<p>We next integrated CD14<sup>+<\/sup>&nbsp;monocytes from PBMCs and CD163<sup>+<\/sup>&nbsp;or CD14<sup>+<\/sup>&nbsp;myeloid cells from BAL fluid. This integrated dataset identified CI1 with &gt;75% cells from BAL fluid and expressing&nbsp;<em>MARCO<\/em><sup>+<\/sup><em>FABP4<\/em><sup>+<\/sup>, markers for tissue-resident alveolar macrophages, two clusters (CI2 and CI3) with &gt;65% of cells from PBMCs and expressing&nbsp;<em>LYZ<\/em><sup>+<\/sup>CD14<sup>+<\/sup>, markers for circulating monocytes, and three mixed clusters (CI4\u2013CI6) with comparable proportion of cells from both PBMCs and BAL (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig6\">6d<\/a>&nbsp;and Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig15\">8c,d<\/a>). PBMC monocytes in CL5 primarily localized to clusters CI4\u2013CI6 (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig6\">6e<\/a>), suggesting a macrophage-polarized phenotype. Among these, cluster CI4 had the highest LC-Mo signature enrichment and higher expression of a profibrotic gene set defined in prior COVID-19 BAL studies<sup><a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR33\">33<\/a><\/sup>&nbsp;and including&nbsp;<em>TREM2<\/em>,&nbsp;<em>CALM1<\/em>,&nbsp;<em>LGMN<\/em>&nbsp;and&nbsp;<em>APOE<\/em>&nbsp;(Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig6\">6f<\/a>&nbsp;and Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig15\">8e<\/a>). Individuals with resp-PASC showed a higher proportion of CI4 cells and higher CI4\/CI5 and CI4\/CI6 ratios than individuals without resp-PASC (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig6\">6g<\/a>&nbsp;and Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig15\">8f<\/a>). Differential expression analysis revealed that CI4 cells upregulated the expression of&nbsp;<em>SPP1<\/em>,&nbsp;<em>CCL13<\/em>,&nbsp;<em>CCL2<\/em>&nbsp;and&nbsp;<em>FOLR2<\/em>&nbsp;compared to CI5 or CI6 cells from both individuals with resp-PASC and non-resp-PASC (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig6\">6h<\/a>). These results indicate LC-Mo enrichment in Resp-PASC PBMCs and its association with a profibrotic transcriptional profile in lung myeloid cells.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"Sec9\">Individuals with LC-Mo exhibit dysregulated monocyte function<\/h3>\n\n\n\n<p>To assess the functional implications of LC-Mo during immune challenge, we stimulated PBMC samples from cohort 1 (months 1.7\u20132.9 and 6\u20138.9) with heat-inactivated&nbsp;<em>Pseudomonas aeruginosa<\/em>&nbsp;for 4\u2009h and performed single-cell multiome profiling in samples from individuals with LC<sup>AM<\/sup>&nbsp;(<em>n<\/em>\u2009=\u20097), LC<sup>AS<\/sup>&nbsp;(<em>n<\/em>\u2009=\u20095) and R<sup>LC<\/sup>&nbsp;(<em>n<\/em>\u2009=\u20096; Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig16\">9a,b<\/a>). Stimulation resulted in a reduction in the numbers of CD14<sup>+<\/sup>&nbsp;and CD16<sup>+<\/sup>&nbsp;monocytes compared to unstimulated samples (Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig16\">9c,d<\/a>), consistent with prior reports<sup><a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR34\">34<\/a>,<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#ref-CR35\">35<\/a><\/sup>. Joint analysis of stimulated and unstimulated samples showed that stimulated LC<sup>AM<\/sup>&nbsp;CD14<sup>+<\/sup>&nbsp;monocytes exhibited significant downregulation of the inflammatory response, IFN\u03b3 signaling, IL-10 signaling, cytokine signaling and IL-6\u2013JAK\u2013STAT3 signaling pathways relative to stimulated R<sup>LC<\/sup>&nbsp;CD14<sup>+<\/sup>&nbsp;monocytes (Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig16\">9e<\/a>). Next, we classified donors as LC-Mo<sup>hi<\/sup>&nbsp;(&gt;10% of CD14<sup>+<\/sup>&nbsp;monocytes exhibiting the LC-Mo state) or LC-Mo<sup>lo<\/sup>&nbsp;(&lt;10%); all R<sup>LC<\/sup>&nbsp;and LC<sup>AS<\/sup>&nbsp;samples were LC-Mo<sup>lo<\/sup>&nbsp;(Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig12\">5f<\/a>). Comparison of stimulated LC-Mo<sup>hi<\/sup>&nbsp;and LC-Mo<sup>lo<\/sup>&nbsp;identified&nbsp;<em>DHFR<\/em>,&nbsp;<em>HMOX1<\/em>,&nbsp;<em>EREG<\/em>&nbsp;and&nbsp;<em>GCLC<\/em>&nbsp;among the top significantly upregulated DEGs (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig7\">7a<\/a>). Pathways related to \u2018IFN\u03b1 response\u2019 and \u2018cytokine signaling\u2019 were significantly decreased in expression (Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig7\">7b<\/a>) in stimulated LC-Mo<sup>hi<\/sup>&nbsp;compared to stimulated LC-Mo<sup>lo<\/sup>. At the gene level, stimulation induced cytokine and chemokine gene expression (<em>CCL3<\/em>,&nbsp;<em>CCL4<\/em>,&nbsp;<em>CXCL3<\/em>&nbsp;and&nbsp;<em>IL6<\/em>) in both stimulated LC-Mo<sup>hi<\/sup>&nbsp;and stimulated LC-Mo<sup>lo<\/sup>, whereas IFN response genes (<em>IRF9<\/em>,&nbsp;<em>ASCC3<\/em>,&nbsp;<em>XAF1<\/em>,&nbsp;<em>SAMD9L<\/em>,&nbsp;<em>LILRB4<\/em>&nbsp;and&nbsp;<em>CGAS<\/em>) were downregulated in LC-Mo<sup>hi<\/sup>&nbsp;(Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig7\">7c<\/a>). Motif accessibility analysis of stimulated LC-Mo<sup>hi<\/sup>&nbsp;and stimulated LC-Mo<sup>lo<\/sup>&nbsp;showed that FOXO and TCF (especially TCF7L2) and ZIC motifs were more accessible in stimulated LC-Mo<sup>hi<\/sup>, whereas stimulated LC-Mo<sup>lo<\/sup>&nbsp;showed increased ETS and AP-1 motif accessibility compared to stimulated LC-Mo<sup>hi<\/sup>&nbsp;(Extended Data Fig.&nbsp;<a href=\"https:\/\/www.nature.com\/articles\/s41590-025-02387-1#Fig16\">9f<\/a>). Together, these data suggest that LC-Mo might contribute to the functional immune dysregulation observed in individuals with LC.<\/p>\n\n\n\n<p><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Abstract The mechanisms driving immune dysregulation in long COVID disease remain elusive. Here we integrated single-cell multiome data, immunological profiling and functional assays to investigate immune alterations across multiple cohorts. [&hellip;]<\/p>\n","protected":false},"author":2,"featured_media":14627,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[150,836,292,293],"tags":[],"class_list":["post-14303","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-epigenetic","category-immune-system","category-lung","category-lung-disease"],"_links":{"self":[{"href":"https:\/\/cov19longhaulfoundation.org\/index.php?rest_route=\/wp\/v2\/posts\/14303","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/cov19longhaulfoundation.org\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/cov19longhaulfoundation.org\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/cov19longhaulfoundation.org\/index.php?rest_route=\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/cov19longhaulfoundation.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=14303"}],"version-history":[{"count":5,"href":"https:\/\/cov19longhaulfoundation.org\/index.php?rest_route=\/wp\/v2\/posts\/14303\/revisions"}],"predecessor-version":[{"id":14626,"href":"https:\/\/cov19longhaulfoundation.org\/index.php?rest_route=\/wp\/v2\/posts\/14303\/revisions\/14626"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/cov19longhaulfoundation.org\/index.php?rest_route=\/wp\/v2\/media\/14627"}],"wp:attachment":[{"href":"https:\/\/cov19longhaulfoundation.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=14303"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/cov19longhaulfoundation.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=14303"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/cov19longhaulfoundation.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=14303"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}