Bradley W. Richmond

Goblet cell hyperplasia (GCH) is a hallmark of chronic obstructive pulmonary disease (COPD) and contributes to morbidity and mortality. We investigated the cellular and molecular origins of GCH in COPD using single-cell RNA sequencing (scRNA-seq), spatial transcriptomics, and in vitro models. We identified “basal-to-goblet transitional cells” (BGTC) which are transcriptionally and physically located between basal and goblet cells and are characterized by expression of clade B serpins. In vitro studies indicate that inflammatory cytokines, including IL-1β, stimulate basal-to-goblet differentiation through a SERPINB3/4 ⁺ intermediate and SERPINB3 overexpression induces differentiation of airway basal cells into IL1B -producing inflammatory goblet cells. In addition, we identified a subset of SERPINB4 ⁺ “goblet-variant basal cells” from lungs of COPD patients that express goblet cell and inflammatory genes ex vivo. Together, these results indicate clade B serpins contribute to altered epithelial differentiation and inflammatory signaling in COPD and support therapeutic targeting of clade B serpins to reduce GCH.

Idiopathic pulmonary fibrosis (IPF) is a progressive fibrotic interstitial lung disease associated with failed alveolar epithelial repair with an expansion of aberrant airway-like epithelium in the alveolar space leading to lung function decline usually resulting in death within 3-5 years of diagnosis. While single-cell and spatial transcriptomic approaches have been used to characterize disease-emergent cell populations, less is known about the regulation of transcriptional programs that drive failed alveolar epithelial cell repair in IPF. DNA methylation is a fundamental layer of gene regulation that stabilizes differentiated cell identity; however, changes in methylation in the epithelial compartment in IPF have not been studied. To identify novel epigenetic mediators of epithelial cell dysfunction, we performed high-resolution DNA methylation profiling of purified distal lung epithelial cells from 10 age-matched control and 14 IPF lungs using Oxford Nanopore Technologies (ONT) whole-genome, long-read sequencing. We identified widespread methylome remodeling in the IPF lung epithelium, with 84% of differentially methylated regions (DMRs) hypomethylated. DMRs were largely found outside of promoters, with 88% outside of ±3 kb from the transcription start site (TSS), consistent with altered distal regulatory element activity. DMRs were enriched for transcription factor (TF) binding site motifs and gene associations consistent with dysregulation of polycomb repressive complex 2 (PRC2) and increases in p63 activity. DNA methylation at DMRs associated with p63 target genes, including KRT5 , and genes implicated in failed epithelial repair, including MUC5B, MMP7, and S100A2 inversely correlated with gene expression. Consistent with the observed dysregulation of the IPF epithelial methylome, an experimental co-culture model of alveolar type II epithelial cell (AT2s) to basal-like epithelial cell transdifferentiation revealed widespread hypomethylation. Further, sites associated with NKX2.1 and FOXA1/2 binding, TFs involved in alveolar fate maintenance, were hypermethylated, suggesting loss of epigenetic regulation of alveolar identity. Together, these data implicate DNA methylation in the failed alveolar epithelial repair processes in IPF, potentially providing future therapeutic strategies by identifying putative regulatory elements associated with aberrant transcriptional programs.

Land-based military deployers to Afghanistan and/or Southwest Asia (SWA) encountered exposure to high concentrations of respirable particulate matter (PM) from multiple sources, including desert dust, burn pit smoke, and military occupations. Adverse lung health effects following deployment have been noted, including upper and lower respiratory tract symptoms, asthma, and small airway and other abnormalities on lung biopsy. The American Thoracic Society (ATS) convened a Workshop in 2018 to review studies assessing post-deployment respiratory health, describe emerging research, and highlight knowledge gaps. Progress on understanding post-deployment health prompted a second ATS Workshop to update current knowledge by 1) reviewing new studies linking exposure assessments to symptoms and/or clinical disease; 2) describing the spectrum of lung pathology reported in previously deployed personnel; 3) evaluating current knowledge of long-term health outcomes after deployment; 4) reviewing data from recent experimental models of deployment-related respiratory diseases (DRRDs); and 5) providing recommendations for future research priorities. Workshop participants agreed that there is substantial evidence linking deployment-related exposures to respiratory symptoms, pulmonary diagnoses, and lung pathology. Knowledge gaps include understanding: 1) the extent and mechanisms through which specific exposures result in impaired pulmonary function, small airways disease, and potentially future chronic pulmonary diseases; 2) the contribution of exposure-related foreign material in the lung to clinical and pathologic findings; and 3) the relationship of pathologic findings to respiratory health, especially those involving small airways.

Background: The widespread adoption of electronic health records has resulted in the generation of large volumes of clinical notes. Learning algorithms and large language models can be trained on these resources, but they are susceptible to noise—irrelevant or noninformative data. This sensitivity can lead to significant challenges, including performance degradation and the generation of inaccurate predictions or “hallucinations.” This study addresses a critical challenge in clinical informatics: efficiently filtering millions of documents for relevance before advanced language model processing, particularly in resource-constrained environments. Objective: We present a novel framework for determining document relevance in clinical settings using a chronic obstructive pulmonary disease (COPD) dataset. Methods: We developed a novel framework using weak supervision and domain-expert heuristics to generate “silver standard” labels for training data and gold standard expert-annotated labels, creating 2 datasets to optimize the model during the development phase and subsequent testing phase. Various text representation techniques (bag of words, term frequency–inverse document frequency, lightweight document embeddings, compression-based features, and Unified Medical Language System concept extraction) were evaluated. These representations were used to train random forest, extreme gradient boosting, and k-nearest neighbor classifiers. Models were optimized on a small expert-annotated dataset and evaluated on a held-out test set. Results: The combination of lightweight document embedding with a random forest classifier demonstrated the best performance, achieving a precision of 0.73, recall of 0.86, and F ₁ -score of 0.80 (95% CI 0.76-0.87) for identifying relevant COPD documents. This significantly outperformed baseline heuristics (precision=0.70; recall=0.38; F ₁ -score=0.50, 95% CI 0.43-0.56) and other tested methods. Conclusions: Our study presents a novel framework for identifying COPD-relevant clinical documents using lightweight embedding and machine learning. This approach effectively filters pertinent documents, enhancing information retrieval precision. The framework’s scalability and minimal annotation needs make it promising for diverse health care applications, potentially optimizing clinical outcomes through efficient document selection for data-driven decision support systems.

Background: Small airways disease is a feature of many respiratory conditions. Currently available methods of diagnosing small airways lack sensitivity and/or cannot evaluate spatial heterogeneity. New diagnostic strategies for diagnosing small airways disease are needed. Methods: We determined the regional displacement of lung tissue calculated from fluoroscopic lung images acquired at multiple angles over sequential time points as a surrogate of ventilation. We applied this technique, which we call X-ray velocimetry (XV), to patients with chronic obstructive pulmonary disease (COPD) and impaired spirometry and veterans with deployment-related constrictive bronchiolitis (DR-CB) but preserved spirometry to determine XV-derived biomarkers specific for each condition. Results: We identified disease- and stage-specific XV biomarkers for COPD patients that correlated with airflow obstruction on spirometry. Further, we identified a set of XV-derived biomarkers that could distinguish veterans with DR-CB from controls despite normal spirometry in most patients from both groups. Conclusions: XV may provide a safe and widely-available strategy for diagnosing small airways disease while preserving spatial information. Future studies are required to validate the biomarkers described here in larger patient cohorts. Trial registration: Not required for this study. However, participants enrolled at VUMC were enrolled under ClinicalTrials.gov study NCT04489758 (submitted 07/23/2020).

Pulmonary type 2 innate lymphoid cells (ILC2s) exhibit immune memory, termed “trained immunity,” which enhances their activation following exposure to an independent protease-containing allergen. The role of prostaglandin I2 (PGI2), a cyclooxygenase (COX) pathway metabolite, in modulating these trained ILC2 responses remains unclear. PGI2 acts through its G protein–coupled receptor IP. We hypothesized that IP signaling inhibits ILC2 training. To test this hypothesis, we used a mouse ILC2 training model in which we challenged wild-type (WT) and IP knockout (KO) mice with Alternaria alternata extract (Alt) to induce ILC2 activation and training. After a 33-d resting period, ILC2 responses subsided to a homeostatic level. Mice were then intranasally challenged with papain to evaluate responses to an unrelated allergen. IP KO mice displayed significantly heightened ILC2 interleukin (IL)-13 expression and with concomitant increased eosinophilia in the lungs post–papain challenge compared with WT control mice. Notably, neither WT nor IP KO mice challenged with papain only, devoid of ILC2 training, exhibited lung type 2 responses. The augmented type 2 inflammation observed in IP KO mice following both Alt and papain challenges correlated with increased numbers and percentages of IL-13-producing ILC2s and greater mean fluorescence intensity of IL-13 compared with WT mice. Furthermore, RNA sequencing of sorted ILC2s from WT and IP KO mice following Alt-papain challenges revealed heightened activation of immune response pathways and mitochondrial respiratory pathways in IP-deficient ILC2s. These findings reveal an inhibitory role of PGI2 signaling in trained ILC2 responses, emphasizing its pivotal contribution to innate immune responses and allergic inflammation.

Background The widespread adoption of electronic health records (EHRs) has resulted in the generation of large volumes of clinical notes. Learning algorithms and large language models (LLMs) train from these resources but are susceptible to noise—irrelevant or non-informative data from them. This sensitivity can lead to significant challenges, including performance degradation and the generation of inaccurate predictions or “hallucinations.” This study addresses a critical challenge in clinical informatics: efficiently filtering millions of documents for relevance before advanced language model processing, particularly in resource-constrained environments. We present a novel framework for determining document relevance in clinical settings, utilizing a chronic obstructive pulmonary disease (COPD) dataset. Methods We developed a novel framework using weak supervision and domain-expert heuristics to generate “silver standard” labels for training data and expert annotated labels (gold stand),creating two datasets to optimize the model during the development phase and subsequent testing phase. Various text representation techniques, including Bag-of-Words, TF-IDF, lightweight document embeddings, compression-based features, and UMLS concept extraction, were evaluated. These representations were used to train Random Forest, XGBoost, and K-Nearest Neighbors classifiers. Models were optimized on a small expert-annotated dataset and evaluated on a held-out test set. Results The combination of lightweight document embedding with a Random Forest classifier demonstrated the best performance, achieving a precision of 0.75, recall of 0.89, and F1-score of 0.81 (95% CI: 0.76-0.87) for identifying relevant COPD documents. This significantly outperformed baseline heuristics (precision: 0.70, recall: 0.38, F1-score: 0.50, 95% CI: 0.43-0.56) and other tested methods. Conclusion Our study presents a novel framework for identifying COPD-relevant clinical documents using lightweight embedding and machine learning. This approach effectively filters pertinent documents, enhancing information retrieval precision. The framework’s scalability and minimal annotation needs make it promising for diverse healthcare applications, potentially optimizing clinical outcomes through efficient document selection for data-driven decision support systems.

We developed a mice model of “post-deployment respiratory syndrome” (PDRS), a condition in Veterans with unexplained exertional dyspnea. Our model successfully recapitulates many of the pathological and physiological features of the syndrome, revealing involvement of the ROS-isoLGs-Sirt3-SOD2 pathway in pulmonary vasculature pathology. Our study provides additional knowledge about effects and long-term consequences of sulfur dioxide exposure on the respiratory system, serving as a valuable tool for future PDRS research.

Rationale: Multiciliated cell (MCC) loss and/or dysfunction is common in the small airways of patients with chronic obstructive pulmonary disease (COPD), but it is unclear if this contributes to COPD lung pathology. Objectives: To determine if loss of p73 causes a COPD-like phenotype in mice and explore whether smoking or COPD impact p73 expression. Methods: p73floxE7–E9 mice were crossed with Shh-Cre mice to generate mice lacking MCCs in the airway epithelium. The resulting p73Δairway mice were analyzed using electron microscopy, flow cytometry, morphometry, forced oscillation technique, and single-cell RNA sequencing. Furthermore, the effects of cigarette smoke on p73 transcript and protein expression were examined using in vitro and in vivo models and in studies including airway epithelium from smokers and patients with COPD. Measurements and Main Results: Loss of functional p73 in the respiratory epithelium resulted in a near-complete absence of MCCs in p73Δairway mice. In adulthood, these mice spontaneously developed neutrophilic inflammation and emphysema-like lung remodeling and had progressive loss of secretory cells. Exposure of normal airway epithelium cells to cigarette smoke rapidly and durably suppressed p73 expression in vitro and in vivo. Furthermore, tumor protein 73 mRNA expression was reduced in the airways of current smokers (n = 82) compared with former smokers (n = 69), and p73-expressing MCCs were reduced in the small airways of patients with COPD (n = 11) compared with control subjects without COPD (n = 12). Conclusions: Loss of functional p73 in murine airway epithelium results in the absence of MCCs and promotes COPD-like lung pathology. In smokers and patients with COPD, loss of p73 may contribute to MCC loss or dysfunction.