Citations

ABSTRACT Lactylation, a protein post-translational modification driven by lactate, plays crucial roles in various physiological and pathological processes. However, the function and underlying mechanisms of lactylation in infection and pathogenesis of influenza A virus (IAV) remain poorly understood. Here, we demonstrated that sirtuin 1 (SIRT1) suppressed IAV replication by mediating delactylation of viral proteins. We found that IAV infection significantly downregulated the expression of SIRT1. Functional studies revealed that depletion of SIRT1 markedly enhanced replication of IAV, whereas overexpression of SIRT1 attenuated the viral replication. Mechanistically, we observed that SIRT1 interacted with IAV NP, PA, PB1, and PB2 proteins, components of the viral ribonucleoprotein (vRNP) complex, and restrained the activity of the vRNP complex, thereby repressing transcription and replication of the IAV genome. Furthermore, we identified that viral NP, PA, PB1, and PB2 proteins underwent lactylation during IAV infection, and SIRT1 facilitated the delactylation of NP, PA, and PB2. Loss of SIRT1 enhanced the lactylation of NP, PA, and PB2, while overexpression of SIRT1 impaired lactylation of these viral proteins during IAV infection. Moreover, inhibition or depletion of lactate dehydrogenase A (LDHA) robustly reduced the lactylation of NP, PA, and PB2, thereby abolishing the promotion of IAV replication caused by SIRT1 deficiency, suggesting that SIRT1 repressed IAV replication via delactylation of viral NP, PA, and PB2 proteins. Together, these results uncover a mechanism that SIRT1 suppresses IAV replication via delactylation of the vRNP components and elicit a critical role for the lactylation of viral proteins in regulating IAV infection and pathogenesis. IMPORTANCE Lactylation plays vital roles in diverse pathological processes. However, the relationship between lactylation and influenza A virus (IAV) infection remains largely unexplored. Particularly, little information is available about the lactylation of viral proteins and their roles in the IAV pathogenesis. Here, viral proteins (NP, PA, PB1, and PB2) were identified to become lactylated during IAV infection. SIRT1 interacted with and promoted the delactylation of viral NP, PA, and PB2 proteins, thereby restraining the vRNP activity and inhibiting the viral transcription and replication. Interestingly, the expression of SIRT1 decreased following IAV infection, suggesting that IAV had evolved a mechanism to downregulate the SIRT1 expression, thereby ensuring appropriate lactylation of viral proteins and facilitating the IAV replication. Together, these findings reveal a critical role for the lactylation of viral proteins in regulating the replication of IAV and provide an important insight into the complicated interplay between the host and IAV.

Cellular and molecular heterogeneity in the liver has been increasingly recognized to drive liver fibrosis progression, but the particular events that occur initially in response to liver injury and trigger immune cell recruitment remain unclear. Here, we identify epigenetically aberrant liver sinusoidal endothelial cells (LSECs) as key players in this process. Mechanistically, the epigenetic readers like bromodomain-containing protein 4–dependent (BRD4-dependent) super enhancers (SEs) activate proinflammatory genes, including promyelocytic leukemia (PML). The PML protein, in turn, binds BRD4 and amplifies proinflammatory angiocrine signaling through phase separation–dependent SE activation via PML/BRD4 condensate formation. In mouse models, LSEC-specific depletion of the PML/BRD4 complex mitigates liver inflammation and fibrosis. Single-cell RNA-seq reveals that epigenetically aberrant LSECs exhibit a reprogrammed proinflammatory angiocrine landscape in mouse fibrotic livers. TIMP1 + LSECs promote the recruitment of CD63 + monocyte–derived macrophages (MoMFs) during liver fibrosis progression. Thereby, PML/BRD4 in LSECs governs inflammatory immune cell recruitment in liver fibrosis. Pharmacological BRD4 inhibition or epigenetic PML-SE repression alleviates liver inflammation and fibrosis. In conclusion, PML/BRD4-mediated SE activation via phase separation drives proinflammatory angiocrine signaling in LSECs, initiating the inflammatory cascade and subsequent immune cell recruitment during liver fibrosis.

Aging is a major risk factor for increased morbidity and mortality following acute respiratory virus infections. To elucidate the immune determinants underlying viral pathogenesis and delayed lung repair in the aged lung, a comprehensive time-course study was conducted. Single-cell RNA sequencing (scRNAseq) and high-dimensional flow cytometry were utilized to compare lungs from young and aged mice infected with influenza A virus (IAV). Aged hosts displayed diminished alveolar macrophage (AM) and dendritic cell (DC) but elevated monocyte-derived macrophage (MoM) and interstitial macrophage (IM) presence following infection. Additionally, enhanced accumulation of adaptive immune cells, including CD4+ tissue-resident helper (TRH) cells, CD8+ tissue-resident memory (TRM) cells, and a B cell subset resembling age-associated B cells, was observed in the memory phase. Pathway analysis revealed that elevated type I and II interferon (IFNα/γ) signaling, especially in MoM/IM subsets, distinguished the aged hosts from the young. Inhibition of IFNα/γ signaling after viral clearance improved long-term respiratory outcomes and reduced both IM and TRH populations in aged mice. These findings highlight the pivotal role of IFNα/γ signaling, likely within MoM/IM subsets, in driving the exuberant persistence of adaptive immune cells and chronic immunopathology in the aged lung following acute viral infection.

Aging is a major risk factor for increased morbidity and mortality following acute respiratory virus infections. To elucidate the immune determinants underlying viral pathogenesis and delayed lung repair in the aged lung, a comprehensive time-course study was conducted. Single-cell RNA sequencing (scRNAseq) and high-dimensional flow cytometry were utilized to compare lungs from young and aged mice infected with influenza A virus (IAV). Aged hosts displayed diminished alveolar macrophage (AM) and dendritic cell (DC) but elevated monocyte-derived macrophage (MoM) and interstitial macrophage (IM) presence following infection. Additionally, enhanced accumulation of adaptive immune cells, including CD4+ tissue-resident helper (TRH) cells, CD8+ tissue-resident memory (TRM) cells, and a B cell subset resembling age-associated B cells, was observed in the memory phase. Pathway analysis revealed that elevated type I and II interferon (IFNα/γ) signaling, especially in MoM/IM subsets, distinguished the aged hosts from the young. Inhibition of IFNα/γ signaling after viral clearance improved long-term respiratory outcomes and reduced both IM and TRH populations in aged mice. These findings highlight the pivotal role of IFNα/γ signaling, likely within MoM/IM subsets, in driving the exuberant persistence of adaptive immune cells and chronic immunopathology in the aged lung following acute viral infection.

Artificial intelligence (AI) and machine learning have emerged as transformative analytic methods capable of identifying patterns and proposing treatment strategies from complex, heterogeneous data generated in patient care settings. However, translating reports of AI tools into clinical decision-making requires careful interpretation and contextualization. Here, we present a structured framework (ABCDEFG framework) for clinicians to critically appraise studies reporting AI tools using a high-profile reinforcement learning analysis of vasopressin initiation in septic shock as a case example. We highlight key methodological strengths including adherence to reporting standards, alignment with clinical outcomes, and external validation. We also identify limitations related to causal inference, reward structure design, assumptions about state representation, and implementation readiness. Our framework emphasizes the importance of rigorous, clinically grounded critique before integrating AI tools into practice, and provides a roadmap for clinicians to assess the reliability and relevance of emerging AI-based decision support.

ABSTRACT Imaging mass spectrometry has made significant advancements in recent years owing to its ability to study the spatial localization and abundance of proteins, peptides, lipids, glycans, metabolites, and drugs. In this study, we employed an imaging mass spectrometry‐based workflow called MALDI‐IHC, which uses antibodies conjugated with photocleavable mass‐tags, to investigate the detection of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) along with pan‐cytokeratin in formalin‐fixed paraffin‐embedded (FFPE) sections from cases with breast cancer. The expression of ER, PR, and HER2 was consistent with the clinical diagnosis based on conventional immunohistochemistry, lacking signal in triple‐negative cases and concordant signals in the cases that were classified as ER+/PR+/HER2+, ER+/PR+/HER2−, and ER−/PR−/HER2+ groups, respectively. In addition, the potential utility of a seconday antibody‐based MALDI‐IHC approach for HER2 expression was evaluated. In agreement with immunohistochemistry results, HER2 expression co‐localized within tumor regions in a HER2+ breast cancer case, while no expression was observed in a HER2− case. Collectively, these results highlight the potential of imaging mass spectrometry‐based MALDI‐IHC workflow for multiplexed detection of proteins in clinically relevant tissue sections.

Background Predictive models employing machine learning algorithms are increasingly being used in clinical decision making, and improperly calibrated models can result in systematic harm. We sought to investigate the impact of class imbalance correction, a commonly applied preprocessing step in machine learning model development, on calibration and modelled clinical decision making in a large real-world context. Methods A histogram boosted gradient classifier was trained on a highly imbalanced national dataset of >1.8 million patients undergoing surgery, to predict the risk of 90-day mortality and complications after surgery. Class imbalance correction strategies including random oversampling, synthetic minority oversampling technique, random under-sampling, and cost-sensitive learning were compared to the natural distribution ('natural'). Models were tested and compared with classification metrics, calibration plots, decision curve analysis, and simulated clinical impact analysis. Results The natural model demonstrated high performance (AUROC 0.94, 95% CI 0.94--0.95 for mortality; 0.84, 95% CI 0.84--0.85 for complications) and calibration (log loss 0.05, 95% CI 0.04--0.05 for mortality; 0.23, 95% CI 0.23--0.24 for complications). Class imbalance mitigation (CSL, ROS, RUS, and SMOTE) did not improve AUROC or AUPRC but increased recall and F1 scores at the expense of precision and accuracy. However, these methods severely compromised model calibration, leading to significant over-prediction of risks (up to a 62.8 % increase) as further evidenced by increased log loss across all mitigation techniques. Decision curve analysis and clinical scenario testing confirmed that the natural model provided the highest net benefit. Conclusion Class imbalance correction methods result in significant miscalibration, leading to possible harm whenused for clinical decision making.