![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Entry tags:
Ничто на Земле не проходит бесследно..
Заболевание ковидом приводит к тому, что в иммунных клетках (и короткоживущих, и их предшественниках, и в долгоживущих клетках врожденного иммунитета) формируются и закрепляются эпигенетические изменения, которые изменяют их функциональность и активность.
Надолго.
Это, конечно, может случаться не только при ковиде, но при любом серьезном заболевании (а то и при иммунизации, например, БЦЖ), но в работе ученых из Нью Йорка исследованы подобные изменения, остающиеся после ковида.
В итоге, после острой болезни выходит, что человек как бы выздоровел, но иногда остался "нездоров" (постковид).
И возбудитель, и воспалительная реакция при ковиде в организме приводят к тому, что множественные линии иммунных клеток приобретают "про-воспалительный" профиль, и на ничтожные раздражители начинают реагировать "неадекватно".
Для тех кто переболел нетяжело, это может быть повышенная антивирусная активность- что может быть преимуществом для сезонных ОРЗшек. Для тех кого зацепило посерьезнее, и особенно для тех кто нуждался в госпитализации-реанимации, это более "воспалительный" перекос в работе иммунитета, и он может быть ответственнен за долгий ковид.
Подробности в работе.
Coronavirus disease 2019 (COVID-19), the illness caused by infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV2), is characterized by a broad range of symptoms and severity and can result in a protracted course. Delayed adaptive immune and interferon (IFN) responses together with robust innate immune cell activity feature prominently in acute severe COVID-19 (1–10). However, the long-term effects of COVID-19 on the immune system are unclear. Durable changes in the immune system following COVID-19 could influence subsequent immune responses to pathogens, vaccines, or even contribute to long-term clinical symptoms, i.e., post-acute sequelae of SARS-CoV-2 infection (PASC) (11–13) and COVID-19-associated multisystem inflammatory syndrome in adults (MIS-A) and children (MIS-C) (14–18). Despite clinical observations of long-term sequelae, the nature of persistent molecular and cellular changes following COVID-19 are poorly understood. Recent studies have established that innate immune cells and their progenitors can maintain durable epigenetic memory of previous infectious or inflammatory encounters, thereby altering innate immune equilibrium and responses to subsequent challenges (19). This innate immune memory, also termed trained immunity, has been attributed largely to persistent chromatin alterations that modify the type and scope of responsiveness of the cells that harbor them, including long-lived innate immune cells (20), epithelial stem cells (21), and self-renewing hematopoietic progenitors (22) and their mature progeny cells (23–28). While innate immune memory phenotypes have been well-studied with in vivo mouse models, the breadth, relevance, and molecular features of such phenotypes in humans have been more elusive. A paradox of innate immune memory has been that many of the innate immune cells that retain durable alterations are themselves short-lived (29). Several recent studies in mice have provided a partial explanation of this paradox, revealing that hematopoietic stem and progenitor cells (HSPC) can be durably altered and epigenetically reprogrammed to persistently convey inflammatory memory to mature progeny cells with altered phenotypes, sometimes termed central trained immunity (19, 20, 22, 26, 30). Molecular features of such phenotypes in humans have focused on innate immune and hematopoietic progenitor cell alterations following administration of the tuberculosis vaccine Bacillus Calmette-Guérin (BCG) (23–28, 31), an area of intrigue since protection to heterologous (non-tuberculous) infections post-BCG vaccination were historically observed (32–34). The complexities of studying human HSPC, especially in the context of large, cohort-level infectious disease studies, have limited our understanding of this type of hematopoietic progenitor cell-based memory. HSPC are long-lived self-renewing precursors to diverse mature immune cells. Thus, HSPC are endowed with the unique potential to serve as reservoirs of post-inflammation epigenetic memory and retain programs that drive altered hematopoiesis and phenotypes in innate immune cell progeny. Based on these characteristics and on the inflammatory features of COVID-19, we hypothesized that exposure of HSPC to inflammatory signaling events during COVID-19 may result in epigenetic memory and persisting altered phenotypes following COVID-19. This may be especially true following severe COVID-19, which is characterized by sustained fever, elevated inflammatory cytokines and chemokines, immune complement activity, and tissue damage (35–41). Understanding the long-lasting effects of COVID-19- associated inflammation on hematopoiesis and innate immune memory is relevant for the hundreds of millions of people worldwide who have recovered from COVID-19 and especially the millions that still experience COVID-19 associated symptoms. The long-term clinical sequelae of COVID-19 (“longhaulers,” “long-COVID,” post-acute sequelae of SARS-CoV-2 infection or PASC) (11, 13), particularly among those admitted to the ICU (42), suggests that persistent changes or alterations in immune activity. may play a role. Furthermore, the etiology of incomplete recovery after critical illnesses is poorly understood, and it is possible that lasting alterations in immune cells and their inflammatory response programs contribute to incomplete recovery and post-ICU syndrome (PICS) (43). In this study, we identify epigenetic innate immune memory that results from SARS-CoV-2 infection by characterizing the molecular features of the post-infection period (2 to 12 months after the start of mild and severe COVID-19). We focus on comprehensive analyses of alterations in chromatin and transcription at the single-cell level in monocytes and their HSPC progenitors. In order to study HSPC in-depth, at single-cell resolution, we developed a new workflow to enrich and profile rare hematopoietic stem and progenitor cells from peripheral blood, termed Peripheral Blood Mononuclear Cell analysis with Progenitor Input Enrichment (PBMC-PIE). PBMC-PIE addresses the challenge in access to HSPC (generally acquired through bone marrow aspirate), a limitation that has severely hindered the study of altered hematopoiesis and innate immune memory in the context of human infection and disease. We paired PBMC-PIE with single-nuclei combined RNA-sequencing (RNA-seq) and assay for transposaseaccessible chromatin (snRNA/ATAC-seq). This approach revealed a high-resolution transcriptomic and chromatin accessibility map of diverse HSPC subsets and PBMC in a unique cohort of convalescent COVID-19 study participants, including mild (non-hospitalized) and severe (S, requiring intensive care unit (ICU) admission) convalescent COVID-19 patients, following acute SARS-CoV-2 infection from months (2-4mo, early convalescence) to a year (4-12mo, late convalescence). We compared these cohorts to healthy participants and also to participants recovering from non-COVID-19 critical illness (nonCoV; patients requiring ICU admission) in order to identify features common among patients recovering from critical illness and also unique to patients recovering from severe COVID-19. To complement this deep characterization of HSPC and PBMC from convalescent COVID-19 study participants, we also profiled (i) plasma proteins, by liquid chromatography-mass spectrometry (LC-MS); (ii) SARS-CoV-2 antibody quantity and quality; (iii) circulating cytokines/chemokines; and (iv) progenitor and PBMC populations, by flow cytometry. Thus, the study was designed to understand the molecular features of recovery for a range of SARS-CoV-2 infection severity over a period of months to one year, with the inclusion of uninfected individuals and also non-COVID-19 critical illness as an important control group. We reveal the persistence of epigenetic and transcription programs in HSPC and monocytes following severe disease that are indicative of an altered innate immune responsiveness. These included durable epigenetic memory linked to inflammatory programs, neutrophil differentiation, and monocyte phenotypes. Further, following mild COVID-19, we uncover a months-long, active, interferon regulatory factor/signal transducer and activator of transcription (IRF/STAT)-driven, anti-viral signature in monocytes (with matching programs in HSPC progenitors), indicative of a state of antiviral vigilance (44). This highlights the potential for acute viral infection to drive a durable program of HSPC origin that is conveyed through to progeny monocytes to mediate a heightened anti-viral response program, with potential implications for heterologous protection and resilience to seasonal infections. We studied molecular features of immune cells from convalescent COVID-19 patients with varying disease severities (mild and severe), compared with those of i) healthy participants (n=47) and ii) study participants recovering from non-COVID-19-related critical illness (due to either infectious or noninfectious non-COVID-19 disease, “nonCoV,” n=25). Our convalescent COVID-19 cohort included both mild (“M,” non-hospitalized, WHO score 1-2; n=25) and severe (“S,” WHO score 6-7, requiring ICU. admission; n=87) study participants. To study the durability of molecular features of immune cells, we collected samples between 2 to 4 months following the onset of symptoms (S:2-4mo and M:2-4mo), termed “early convalescence,” or between 4 to 12 months post-acute (S:4-12mo and M:4-12mo), termed “late convalescence.” Healthy donors (symptom-free and seronegative) were enrolled with consideration to age, sex, and comorbidity to match COVID-19 groups. Plasma analysis included SARS-CoV-2 receptor-binding domain (RBD) spike-specific antibody levels, surrogate neutralizing antibody activities, and antibody avidity; 18-plex cytokine and chemokine quantification (Luminex); and unbiased LC/MS-based mapping of plasma proteins. These data and their analyses uncovered plasma factors that persist for months post-acute SARS-CoV-2 infection (fig. S1BC), and also expected antibody dynamics, with increased quality and durability of antibody responses in those with severe disease. No sustained alterations in plasma factors (i.e., proteins or cytokines/chemokines from LC/MS and Luminex analysis) were observed in mild post-COVID-19 (M:2- 4mo and M:4-12mo, fig. S1B-C) compared to healthy study participants. In contrast, we observed a persistent and distinctive active plasma response featuring inflammatory cytokines (IL-6, IL-8) (fig. S1C), immune complement (C1QA, C1R, C3, C4A), and vascular response factors (SAA1, ORM1, LBP, fibrinogen- FGG/FGA, LRG1) in early convalescent severe COVID-19 study participants (S:2-4mo) (fig. S1B). Levels of acute-phase proteins and platelet factors were elevated in both S:2-4mo and nonCoV, compared to healthy study participants (fig. S1B). These data establish the presence of a months-long activation of immune complement and vascular response factors that could contribute to post-COVID-19 sequelae. Importantly, in late convalescence following severe COVID-19 (S:4-12mo), there were no significant alterations in plasma factors by LC/MS compared to healthy donors, indicating a full resolution of the active plasma response and overt inflammation at the protein level (fig. S1B). This suggests that any persistent changes in cellular composition or phenotypes in late convalescence of study participants with severe COVID-19 are independent of an ongoing and active inflammatory program and hence point to epigenetic mechanisms behind prolonged effects of COVID-19. Innate immune memory in response to natural infection in humans has yet to be well characterized. Moreover, whether and how hematopoietic stem and progenitor cells are altered in humans following infection is poorly understood. This study explores if COVID-19 and its associated inflammatory responses result in innate immune memory and whether these phenotypes propagate through HSPC even after the resolution of acute infection. Enrichment of circulating CD34+ HSPC among PBMC via PBMC-PIE (PBMC analysis with progenitor input enrichment), paired with single-cell combined ATAC/RNA analysis, enabled deep characterization of these rare circulating CD34+ HSPC cells in diverse clinical cohorts. We reveal that while rare (~0.05% of PBMC), these cells accurately capture the diversity of bone marrow HSPC subsets. Therefore, PBMCPIE serves as a powerful tool to study hematopoiesis, epigenetic programming of HSPC, and the relationship between progenitor and progeny cell types in blood without directly accessing the bone marrow. We applied this approach to study the lasting effects of COVID-19 and critical illness on hematopoiesis and central trained immunity concepts (20), with a unique cohort including study participants (i) following mild COVID-19 in early (M:2-4mo) and late (M:4-12mo) convalescence; (ii) following severe COVID-19 in early (S:2-4mo) and late (S:4-12mo) convalescence; and (iii) recovering from non-COVID-19 related critical illness (nonCoV). With these participants, we assessed plasma factors, SARS-CoV-2 antibodies, PBMC immunophenotypes by flow, and epigenomic and transcriptional changes in circulating HSPC and immune cells. Notably, we show that severe disease induces long-lasting epigenetic alterations in HSPC and their short-lived progeny, circulating monocytes. Transcription factor activity, chromatin accessibility, and gene expression programs linked phenotypes of post-infection HSPC and monocytes. This indicates that inflammatory signaling in progenitor cells establishes epigenetic memory that not only persists in self-renewing stem cells but is also conveyed through differentiation to influence progeny cell phenotypes. Our results establish a precedent for central trained immunity following viral infection and suggest that recent observations of persistent monocyte epigenetic memory following influenza vaccination (44) may also derive from HSPC alterations. Beyond this epigenetic innate immune memory, we demonstrate that hematopoiesis is altered following COVID- 19, with increases in myeloid and neutrophil progenitor populations. Potential Mechanisms Contributing to Long-term Sequelae following COVID-19 Our study design and cohort size were suitable for the detection of post-COVID-19 programs that are commonly altered after COVID-19 and severe disease. Further studies that follow long-term outcome could parse the molecular features that associate with the vast range of unexplained long-term sequelae after severe (and sometimes mild) illness, including post-acute sequelae of SARS-CoV-2 infection (PASC) (12, 13, 74) and post-ICU syndrome (PICS) (75). Indeed, our findings suggest potential molecular mechanisms that may contribute to PASC and PICS, including (i) a months-long active plasma response including complement activity, acute phase proteins, and vascular risk factors, which, if deposited in tissues, could contribute to ongoing inflammation; (ii) persistently activated (CCL4, CCL5, IL7R) and tissue-migratory (CCR7) monocyte phenotypes, with underlying HSPC epigenetic programs, that may continue to fuel inflammation and fibrosis, notably in lung and upper respiratory mucosa; and (iii) altered hematopoiesis including increased myeloid and especially neutrophil progenitors whose progeny may contribute to ongoing inflammation. Based upon neutrophil stimulating CSF-3 (G-CSF) production by lung epithelium in response to SARS-CoV-2 infection (76) and persistent CSF-3R upregulation in post-COVID-19 HSPC (Fig. 1-2), we suggest that a feedforward loop of CSF-3/CSF-3R could contribute to neutrophil driven pathology in both acute and, due to the durability of this response, also in chronic COVID-19 disease. Innate Immune Memory Following Mild and Severe SARS-CoV-2 Infection One standout finding of our study is the identification of months-long altered monocyte programs following severe, but also mild, SARS-CoV-2 infection, suggesting monocytes may also contribute to chronic inflammation, either in affected tissues, via migratory and chemoattractant programs, or systemically. These epigenetic programs underlie distinct CD14+ monocyte phenotypes with variable persistence depending on disease severity. Unlike mild post-COVID-19 interferon programming, the HSPC and monocyte programs following severe COVID-19 are complex, with individual cells bearing mixed inflammatory and interferon signatures and reduced expression of key negative feedback factors DUSP1 and NFKBIA. We show that mild (non-hospitalized) COVID-19 can result in a months-long epigenetic and transcriptional program, characterized by prominent IRF transcription factor activity and interferon-stimulated gene (ISG) expression (e.g., ISG15, MX1, MX2, IFI44L, IFI44, OAS3, AOAH, and PARP14). In contrast, CD14+ monocytes of patients recovering from severe disease (2-4 months postacute, following discharge from the ICU) feature epigenetic and transcriptional signatures of inflammation likely mediated by NFkB and AP-1 TFs. This active inflammatory CD14+ monocyte program resolves in late convalescence (4-12 months), though a distinct epigenetic monocyte phenotype persists, including increased chromatin accessibility at certain chemokines (e.g., CCL2, CCL7, CCL24), chemokine receptors (e.g., CCR1, CCR3, CCRL2), ISG (e.g., IFI6, SOC3, OASL), and inflammatory genes (e.g., IL8, caspases, S100A genes). This highlights the importance of further research to better understand functional changes in the post-COVID-19 immune system and the clinical implications of these prolonged epigenetic signatures of severe COVID-19 in HPSCs and their progeny. Further, persisting alterations in HSPC and monocytes following mild and severe SARS-CoV-2 infection suggests the possibility of months-long alterations in innate immune status as a general feature of diverse infections. We propose that this dynamic aspect of blood development and innate immune memory could have major implications for vaccine responses and design, understanding post-infectious inflammatory disease, non-genetic variance in responses to infection, and the epidemiology of seasonal infections. Based on these results, it is intriguing to speculate that acute viral infections may induce months-long anti-viral resilience programs similar to what we describe following mild SARS-CoV-2 infection. For example, as a corollary, the aberrantly low frequency of non-SARS-CoV-2 respiratory infections in the winter of 2020-2021 may have led to subsequent increased susceptibility to pathogenic viral infections and the unusual epidemic of respiratory syncytial virus and rhinovirus in the summer of 2020 (77). AP-1 and IRF Transcription Factor Programs Underlie Post-COVID-19 Phenotypes How the transcriptional and epigenetic changes we observed in HSPC and monocytes following COVID- 19 might alter cell differentiation, function, and response to stimulation is a critical open question. The molecular phenotypes of mild post-COVID-19 monocytes were very similar to those described in monocytes following adjuvanted influenza vaccine (H5N1+AS03), which provided a degree of heterologous anti-viral protection (Zika and Dengue) (44). Shared features of post-influenza vaccine and post-COVID-19 monocytes include a reduced AP-1 signature and increased IRF activity (which, in the case of COVID-19, persisted in both HSPC and monocytes). Notably, for up to one year into late convalescence following severe COVID-19, AP-1 activity returns to, or below, baseline, as a result of negative feedback regulation of AP-1 family members, while IRF factor activity remains elevated. Our single-cell analyses also reveal that both reduced AP-1 and increased IRF programs can co-exist within the same cells (Fig. 5J), raising the possibility that the persistent IRF activity following severe disease may represent a primed rather than active anti-viral program. Indeed, IRF factors interact with BAF complex (SWI/SNF) chromatin remodelers to maintain open or poised chromatin states and to drive active transcription (78). IRF1 activity, which drives active inflammation-responsive interferon-stimulated gene transcription, was reduced post-COVID-19, but several other IRF factors were increased, including IRF2 and IRF3, which have been shown to interact with the BAF complex to retain ISG in a poised state (79). We suggest that persisting IRF chromatin binding activity post-COVID-19 could result in increased poising and responsiveness of IRF target genes, in part through the maintenance of accessibility via IRFBAF complex interactions.While our study focuses on blood cells, it is important to point out that diverse other cell types have been demonstrated to harbor epigenetic memory (21, 80, 81). Particularly when they reside in affected tissues, these cells may change in their frequencies, differentiation programs, and phenotypes, and also retain epigenetic memory of anti-viral inflammation with important and enduring influence on tissue defense or sequelae (80, 82). Here, we present evidence of central trained immunity, in the form of epigenetic reprogramming in HSPC, in humans following viral infection and severe illness. Importantly, enrichment of rare circulating progenitor cells using PBMC-PIE was a critical advance enabling evaluation of hematopoietic stem and progenitor cells together with their progeny immune cells from peripheral blood samples. Extending this approach to diverse tissues (particularly those with resident stem and progenitor cells, e.g., intestinal epithelium) and disorders (hematologic disease, malignancy, inflammation, and infection) can unveil epigenetic and progenitor-based mechanisms of pathogenesis and inform therapeutic strategies and targets.
Надолго.
Это, конечно, может случаться не только при ковиде, но при любом серьезном заболевании (а то и при иммунизации, например, БЦЖ), но в работе ученых из Нью Йорка исследованы подобные изменения, остающиеся после ковида.
В итоге, после острой болезни выходит, что человек как бы выздоровел, но иногда остался "нездоров" (постковид).
И возбудитель, и воспалительная реакция при ковиде в организме приводят к тому, что множественные линии иммунных клеток приобретают "про-воспалительный" профиль, и на ничтожные раздражители начинают реагировать "неадекватно".
Для тех кто переболел нетяжело, это может быть повышенная антивирусная активность- что может быть преимуществом для сезонных ОРЗшек. Для тех кого зацепило посерьезнее, и особенно для тех кто нуждался в госпитализации-реанимации, это более "воспалительный" перекос в работе иммунитета, и он может быть ответственнен за долгий ковид.
Подробности в работе.
Coronavirus disease 2019 (COVID-19), the illness caused by infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV2), is characterized by a broad range of symptoms and severity and can result in a protracted course. Delayed adaptive immune and interferon (IFN) responses together with robust innate immune cell activity feature prominently in acute severe COVID-19 (1–10). However, the long-term effects of COVID-19 on the immune system are unclear. Durable changes in the immune system following COVID-19 could influence subsequent immune responses to pathogens, vaccines, or even contribute to long-term clinical symptoms, i.e., post-acute sequelae of SARS-CoV-2 infection (PASC) (11–13) and COVID-19-associated multisystem inflammatory syndrome in adults (MIS-A) and children (MIS-C) (14–18). Despite clinical observations of long-term sequelae, the nature of persistent molecular and cellular changes following COVID-19 are poorly understood. Recent studies have established that innate immune cells and their progenitors can maintain durable epigenetic memory of previous infectious or inflammatory encounters, thereby altering innate immune equilibrium and responses to subsequent challenges (19). This innate immune memory, also termed trained immunity, has been attributed largely to persistent chromatin alterations that modify the type and scope of responsiveness of the cells that harbor them, including long-lived innate immune cells (20), epithelial stem cells (21), and self-renewing hematopoietic progenitors (22) and their mature progeny cells (23–28). While innate immune memory phenotypes have been well-studied with in vivo mouse models, the breadth, relevance, and molecular features of such phenotypes in humans have been more elusive. A paradox of innate immune memory has been that many of the innate immune cells that retain durable alterations are themselves short-lived (29). Several recent studies in mice have provided a partial explanation of this paradox, revealing that hematopoietic stem and progenitor cells (HSPC) can be durably altered and epigenetically reprogrammed to persistently convey inflammatory memory to mature progeny cells with altered phenotypes, sometimes termed central trained immunity (19, 20, 22, 26, 30). Molecular features of such phenotypes in humans have focused on innate immune and hematopoietic progenitor cell alterations following administration of the tuberculosis vaccine Bacillus Calmette-Guérin (BCG) (23–28, 31), an area of intrigue since protection to heterologous (non-tuberculous) infections post-BCG vaccination were historically observed (32–34). The complexities of studying human HSPC, especially in the context of large, cohort-level infectious disease studies, have limited our understanding of this type of hematopoietic progenitor cell-based memory. HSPC are long-lived self-renewing precursors to diverse mature immune cells. Thus, HSPC are endowed with the unique potential to serve as reservoirs of post-inflammation epigenetic memory and retain programs that drive altered hematopoiesis and phenotypes in innate immune cell progeny. Based on these characteristics and on the inflammatory features of COVID-19, we hypothesized that exposure of HSPC to inflammatory signaling events during COVID-19 may result in epigenetic memory and persisting altered phenotypes following COVID-19. This may be especially true following severe COVID-19, which is characterized by sustained fever, elevated inflammatory cytokines and chemokines, immune complement activity, and tissue damage (35–41). Understanding the long-lasting effects of COVID-19- associated inflammation on hematopoiesis and innate immune memory is relevant for the hundreds of millions of people worldwide who have recovered from COVID-19 and especially the millions that still experience COVID-19 associated symptoms. The long-term clinical sequelae of COVID-19 (“longhaulers,” “long-COVID,” post-acute sequelae of SARS-CoV-2 infection or PASC) (11, 13), particularly among those admitted to the ICU (42), suggests that persistent changes or alterations in immune activity. may play a role. Furthermore, the etiology of incomplete recovery after critical illnesses is poorly understood, and it is possible that lasting alterations in immune cells and their inflammatory response programs contribute to incomplete recovery and post-ICU syndrome (PICS) (43). In this study, we identify epigenetic innate immune memory that results from SARS-CoV-2 infection by characterizing the molecular features of the post-infection period (2 to 12 months after the start of mild and severe COVID-19). We focus on comprehensive analyses of alterations in chromatin and transcription at the single-cell level in monocytes and their HSPC progenitors. In order to study HSPC in-depth, at single-cell resolution, we developed a new workflow to enrich and profile rare hematopoietic stem and progenitor cells from peripheral blood, termed Peripheral Blood Mononuclear Cell analysis with Progenitor Input Enrichment (PBMC-PIE). PBMC-PIE addresses the challenge in access to HSPC (generally acquired through bone marrow aspirate), a limitation that has severely hindered the study of altered hematopoiesis and innate immune memory in the context of human infection and disease. We paired PBMC-PIE with single-nuclei combined RNA-sequencing (RNA-seq) and assay for transposaseaccessible chromatin (snRNA/ATAC-seq). This approach revealed a high-resolution transcriptomic and chromatin accessibility map of diverse HSPC subsets and PBMC in a unique cohort of convalescent COVID-19 study participants, including mild (non-hospitalized) and severe (S, requiring intensive care unit (ICU) admission) convalescent COVID-19 patients, following acute SARS-CoV-2 infection from months (2-4mo, early convalescence) to a year (4-12mo, late convalescence). We compared these cohorts to healthy participants and also to participants recovering from non-COVID-19 critical illness (nonCoV; patients requiring ICU admission) in order to identify features common among patients recovering from critical illness and also unique to patients recovering from severe COVID-19. To complement this deep characterization of HSPC and PBMC from convalescent COVID-19 study participants, we also profiled (i) plasma proteins, by liquid chromatography-mass spectrometry (LC-MS); (ii) SARS-CoV-2 antibody quantity and quality; (iii) circulating cytokines/chemokines; and (iv) progenitor and PBMC populations, by flow cytometry. Thus, the study was designed to understand the molecular features of recovery for a range of SARS-CoV-2 infection severity over a period of months to one year, with the inclusion of uninfected individuals and also non-COVID-19 critical illness as an important control group. We reveal the persistence of epigenetic and transcription programs in HSPC and monocytes following severe disease that are indicative of an altered innate immune responsiveness. These included durable epigenetic memory linked to inflammatory programs, neutrophil differentiation, and monocyte phenotypes. Further, following mild COVID-19, we uncover a months-long, active, interferon regulatory factor/signal transducer and activator of transcription (IRF/STAT)-driven, anti-viral signature in monocytes (with matching programs in HSPC progenitors), indicative of a state of antiviral vigilance (44). This highlights the potential for acute viral infection to drive a durable program of HSPC origin that is conveyed through to progeny monocytes to mediate a heightened anti-viral response program, with potential implications for heterologous protection and resilience to seasonal infections. We studied molecular features of immune cells from convalescent COVID-19 patients with varying disease severities (mild and severe), compared with those of i) healthy participants (n=47) and ii) study participants recovering from non-COVID-19-related critical illness (due to either infectious or noninfectious non-COVID-19 disease, “nonCoV,” n=25). Our convalescent COVID-19 cohort included both mild (“M,” non-hospitalized, WHO score 1-2; n=25) and severe (“S,” WHO score 6-7, requiring ICU. admission; n=87) study participants. To study the durability of molecular features of immune cells, we collected samples between 2 to 4 months following the onset of symptoms (S:2-4mo and M:2-4mo), termed “early convalescence,” or between 4 to 12 months post-acute (S:4-12mo and M:4-12mo), termed “late convalescence.” Healthy donors (symptom-free and seronegative) were enrolled with consideration to age, sex, and comorbidity to match COVID-19 groups. Plasma analysis included SARS-CoV-2 receptor-binding domain (RBD) spike-specific antibody levels, surrogate neutralizing antibody activities, and antibody avidity; 18-plex cytokine and chemokine quantification (Luminex); and unbiased LC/MS-based mapping of plasma proteins. These data and their analyses uncovered plasma factors that persist for months post-acute SARS-CoV-2 infection (fig. S1BC), and also expected antibody dynamics, with increased quality and durability of antibody responses in those with severe disease. No sustained alterations in plasma factors (i.e., proteins or cytokines/chemokines from LC/MS and Luminex analysis) were observed in mild post-COVID-19 (M:2- 4mo and M:4-12mo, fig. S1B-C) compared to healthy study participants. In contrast, we observed a persistent and distinctive active plasma response featuring inflammatory cytokines (IL-6, IL-8) (fig. S1C), immune complement (C1QA, C1R, C3, C4A), and vascular response factors (SAA1, ORM1, LBP, fibrinogen- FGG/FGA, LRG1) in early convalescent severe COVID-19 study participants (S:2-4mo) (fig. S1B). Levels of acute-phase proteins and platelet factors were elevated in both S:2-4mo and nonCoV, compared to healthy study participants (fig. S1B). These data establish the presence of a months-long activation of immune complement and vascular response factors that could contribute to post-COVID-19 sequelae. Importantly, in late convalescence following severe COVID-19 (S:4-12mo), there were no significant alterations in plasma factors by LC/MS compared to healthy donors, indicating a full resolution of the active plasma response and overt inflammation at the protein level (fig. S1B). This suggests that any persistent changes in cellular composition or phenotypes in late convalescence of study participants with severe COVID-19 are independent of an ongoing and active inflammatory program and hence point to epigenetic mechanisms behind prolonged effects of COVID-19. Innate immune memory in response to natural infection in humans has yet to be well characterized. Moreover, whether and how hematopoietic stem and progenitor cells are altered in humans following infection is poorly understood. This study explores if COVID-19 and its associated inflammatory responses result in innate immune memory and whether these phenotypes propagate through HSPC even after the resolution of acute infection. Enrichment of circulating CD34+ HSPC among PBMC via PBMC-PIE (PBMC analysis with progenitor input enrichment), paired with single-cell combined ATAC/RNA analysis, enabled deep characterization of these rare circulating CD34+ HSPC cells in diverse clinical cohorts. We reveal that while rare (~0.05% of PBMC), these cells accurately capture the diversity of bone marrow HSPC subsets. Therefore, PBMCPIE serves as a powerful tool to study hematopoiesis, epigenetic programming of HSPC, and the relationship between progenitor and progeny cell types in blood without directly accessing the bone marrow. We applied this approach to study the lasting effects of COVID-19 and critical illness on hematopoiesis and central trained immunity concepts (20), with a unique cohort including study participants (i) following mild COVID-19 in early (M:2-4mo) and late (M:4-12mo) convalescence; (ii) following severe COVID-19 in early (S:2-4mo) and late (S:4-12mo) convalescence; and (iii) recovering from non-COVID-19 related critical illness (nonCoV). With these participants, we assessed plasma factors, SARS-CoV-2 antibodies, PBMC immunophenotypes by flow, and epigenomic and transcriptional changes in circulating HSPC and immune cells. Notably, we show that severe disease induces long-lasting epigenetic alterations in HSPC and their short-lived progeny, circulating monocytes. Transcription factor activity, chromatin accessibility, and gene expression programs linked phenotypes of post-infection HSPC and monocytes. This indicates that inflammatory signaling in progenitor cells establishes epigenetic memory that not only persists in self-renewing stem cells but is also conveyed through differentiation to influence progeny cell phenotypes. Our results establish a precedent for central trained immunity following viral infection and suggest that recent observations of persistent monocyte epigenetic memory following influenza vaccination (44) may also derive from HSPC alterations. Beyond this epigenetic innate immune memory, we demonstrate that hematopoiesis is altered following COVID- 19, with increases in myeloid and neutrophil progenitor populations. Potential Mechanisms Contributing to Long-term Sequelae following COVID-19 Our study design and cohort size were suitable for the detection of post-COVID-19 programs that are commonly altered after COVID-19 and severe disease. Further studies that follow long-term outcome could parse the molecular features that associate with the vast range of unexplained long-term sequelae after severe (and sometimes mild) illness, including post-acute sequelae of SARS-CoV-2 infection (PASC) (12, 13, 74) and post-ICU syndrome (PICS) (75). Indeed, our findings suggest potential molecular mechanisms that may contribute to PASC and PICS, including (i) a months-long active plasma response including complement activity, acute phase proteins, and vascular risk factors, which, if deposited in tissues, could contribute to ongoing inflammation; (ii) persistently activated (CCL4, CCL5, IL7R) and tissue-migratory (CCR7) monocyte phenotypes, with underlying HSPC epigenetic programs, that may continue to fuel inflammation and fibrosis, notably in lung and upper respiratory mucosa; and (iii) altered hematopoiesis including increased myeloid and especially neutrophil progenitors whose progeny may contribute to ongoing inflammation. Based upon neutrophil stimulating CSF-3 (G-CSF) production by lung epithelium in response to SARS-CoV-2 infection (76) and persistent CSF-3R upregulation in post-COVID-19 HSPC (Fig. 1-2), we suggest that a feedforward loop of CSF-3/CSF-3R could contribute to neutrophil driven pathology in both acute and, due to the durability of this response, also in chronic COVID-19 disease. Innate Immune Memory Following Mild and Severe SARS-CoV-2 Infection One standout finding of our study is the identification of months-long altered monocyte programs following severe, but also mild, SARS-CoV-2 infection, suggesting monocytes may also contribute to chronic inflammation, either in affected tissues, via migratory and chemoattractant programs, or systemically. These epigenetic programs underlie distinct CD14+ monocyte phenotypes with variable persistence depending on disease severity. Unlike mild post-COVID-19 interferon programming, the HSPC and monocyte programs following severe COVID-19 are complex, with individual cells bearing mixed inflammatory and interferon signatures and reduced expression of key negative feedback factors DUSP1 and NFKBIA. We show that mild (non-hospitalized) COVID-19 can result in a months-long epigenetic and transcriptional program, characterized by prominent IRF transcription factor activity and interferon-stimulated gene (ISG) expression (e.g., ISG15, MX1, MX2, IFI44L, IFI44, OAS3, AOAH, and PARP14). In contrast, CD14+ monocytes of patients recovering from severe disease (2-4 months postacute, following discharge from the ICU) feature epigenetic and transcriptional signatures of inflammation likely mediated by NFkB and AP-1 TFs. This active inflammatory CD14+ monocyte program resolves in late convalescence (4-12 months), though a distinct epigenetic monocyte phenotype persists, including increased chromatin accessibility at certain chemokines (e.g., CCL2, CCL7, CCL24), chemokine receptors (e.g., CCR1, CCR3, CCRL2), ISG (e.g., IFI6, SOC3, OASL), and inflammatory genes (e.g., IL8, caspases, S100A genes). This highlights the importance of further research to better understand functional changes in the post-COVID-19 immune system and the clinical implications of these prolonged epigenetic signatures of severe COVID-19 in HPSCs and their progeny. Further, persisting alterations in HSPC and monocytes following mild and severe SARS-CoV-2 infection suggests the possibility of months-long alterations in innate immune status as a general feature of diverse infections. We propose that this dynamic aspect of blood development and innate immune memory could have major implications for vaccine responses and design, understanding post-infectious inflammatory disease, non-genetic variance in responses to infection, and the epidemiology of seasonal infections. Based on these results, it is intriguing to speculate that acute viral infections may induce months-long anti-viral resilience programs similar to what we describe following mild SARS-CoV-2 infection. For example, as a corollary, the aberrantly low frequency of non-SARS-CoV-2 respiratory infections in the winter of 2020-2021 may have led to subsequent increased susceptibility to pathogenic viral infections and the unusual epidemic of respiratory syncytial virus and rhinovirus in the summer of 2020 (77). AP-1 and IRF Transcription Factor Programs Underlie Post-COVID-19 Phenotypes How the transcriptional and epigenetic changes we observed in HSPC and monocytes following COVID- 19 might alter cell differentiation, function, and response to stimulation is a critical open question. The molecular phenotypes of mild post-COVID-19 monocytes were very similar to those described in monocytes following adjuvanted influenza vaccine (H5N1+AS03), which provided a degree of heterologous anti-viral protection (Zika and Dengue) (44). Shared features of post-influenza vaccine and post-COVID-19 monocytes include a reduced AP-1 signature and increased IRF activity (which, in the case of COVID-19, persisted in both HSPC and monocytes). Notably, for up to one year into late convalescence following severe COVID-19, AP-1 activity returns to, or below, baseline, as a result of negative feedback regulation of AP-1 family members, while IRF factor activity remains elevated. Our single-cell analyses also reveal that both reduced AP-1 and increased IRF programs can co-exist within the same cells (Fig. 5J), raising the possibility that the persistent IRF activity following severe disease may represent a primed rather than active anti-viral program. Indeed, IRF factors interact with BAF complex (SWI/SNF) chromatin remodelers to maintain open or poised chromatin states and to drive active transcription (78). IRF1 activity, which drives active inflammation-responsive interferon-stimulated gene transcription, was reduced post-COVID-19, but several other IRF factors were increased, including IRF2 and IRF3, which have been shown to interact with the BAF complex to retain ISG in a poised state (79). We suggest that persisting IRF chromatin binding activity post-COVID-19 could result in increased poising and responsiveness of IRF target genes, in part through the maintenance of accessibility via IRFBAF complex interactions.While our study focuses on blood cells, it is important to point out that diverse other cell types have been demonstrated to harbor epigenetic memory (21, 80, 81). Particularly when they reside in affected tissues, these cells may change in their frequencies, differentiation programs, and phenotypes, and also retain epigenetic memory of anti-viral inflammation with important and enduring influence on tissue defense or sequelae (80, 82). Here, we present evidence of central trained immunity, in the form of epigenetic reprogramming in HSPC, in humans following viral infection and severe illness. Importantly, enrichment of rare circulating progenitor cells using PBMC-PIE was a critical advance enabling evaluation of hematopoietic stem and progenitor cells together with their progeny immune cells from peripheral blood samples. Extending this approach to diverse tissues (particularly those with resident stem and progenitor cells, e.g., intestinal epithelium) and disorders (hematologic disease, malignancy, inflammation, and infection) can unveil epigenetic and progenitor-based mechanisms of pathogenesis and inform therapeutic strategies and targets.