Rationale Vascular calcification is a regulated process that involves osteoprogenitor cells and frequently complicates common vascular disease such as atherosclerosis and diabetic vasculopathy. However, it is not clear if the vascular endothelium has a role in contributing osteoprogenitor cells to the calcific lesions. Objective To determine if the vascular endothelium contributes osteoprogenitor cells to vascular calcification. Methods and Results In this study, we use two mouse models of vascular calcification, mice with gene deletion of matrix Gla protein (MGP), a BMP-inhibitor, and Ins2Akita/+ mice, a diabetes model. We show that enhanced bone morphogenetic protein (BMP) signaling in both types of mice stimulates the vascular endothelium to contribute osteoprogenitor cells to the vascular calcification. The enhanced BMP signaling results in endothelial-mesenchymal transitions and the emergence of multipotent cells, followed by osteoinduction. Endothelial markers co-localize with multipotent and osteogenic markers in calcified arteries by immunostaining and fluorescence-activated cell sorting. Lineage tracing using Tie2-Gfp transgenic mice supports an endothelial origin of the osteogenic cells. Enhancement of MGP expression in Ins2Akita/+ mice, as mediated by an Mgp transgene limits the generation of multipotent cells. Moreover, MGP-depleted human aortic endothelial cells in vitro acquire multipotency rendering the cells susceptible to osteoinduction by BMP and high glucose. Conclusions Our data suggest that the endothelium is a source of osteoprogenitor cells in vascular calcification that occurs in disorders with high BMP activation such as deficiency of BMP inhibitors and diabetes.
Significance Cerebral arteriovenous malformations (AVMs) are common vascular abnormalities that may lead to strokes. Signaling by bone morphogenetic proteins (BMPs) and Notch play important roles in the formation of cerebral AVMs, but the cross-talk between the pathways is poorly understood. We report that gene deletion of matrix Gla protein (MGP), a BMP inhibitor, causes cerebral AVMs in mice by activating activin receptor-like kinase 1, a BMP receptor. This activation enhances Notch activity and disrupts endothelial cell differentiation by inducing the Notch ligands Jagged 1 and 2. Reducing Jagged 1 and 2 expression prevents the disruption in differentiation and AVM formation. The findings suggest that MGP maintains the balance between BMP and Notch signaling and promotes a normal brain vasculature.
The importance of morphogenetic proteins (BMPs) and their antagonists in vascular development is increasingly being recognized. BMP-4 is essential for angiogenesis and is antagonized by matrix Gla protein (MGP) and crossveinless 2 (CV2), both induced by the activin receptor like-kinase 1 (ALK1) when stimulated by BMP-9. In this study, however, we show that CV2 preferentially binds and inhibits BMP-9 thereby providing strong feedback inhibition for BMP-9/ALK1 signaling rather than for BMP-4/ALK2 signaling. CV2 disrupts complex formation involving ALK2, ALK1, BMP-4, and BMP-9 required for the induction of both BMP antagonists. It also limits VEGF expression, proliferation, and tube formation in ALK1-expressing endothelial cells.
Rationale: Newer approaches to asthma research include electronic health record (EHR) big data analyses and clinical decision support. Accurate identification of asthma cases from the EHR is challenging as its diagnoses is not always straightforward. Although different algorithm rules (i.e. computable phenotypes) exist, there is not a unique solution and methods have not been validated outside their region of development. We studied two asthma computable phenotypes with reported high sensitivity, specificity and positive predictive values (PPV): the CAPriCORN Asthma Cohort Committee (Chicago) and the Phenotype KnowledgeBase PheKB.org (Philadelphia) algorithms. Our objective in this study was to compare the two algorithms on a large cohort of pediatric patients in the Los Angeles area, with particular focus on the algorithms' performance on pre-existing, physician-diagnosed asthma cases. Methods: This IRB-approved study used data from the EHR (Epic Systems) clinical data warehouse within the University of California Los Angeles (UCLA) healthcare system for patients ages 5-17 years with encounters from March 1, 2013 to present (n=100,869). We first applied the CAPriCORN and PheKB algorithm rules to all pediatric patients in the data warehouse. Among those with at least one asthma diagnosis (n=10,803) we randomly selected a subset labeled as asthma and 100 patients who were not for manual review. Two clinicians reviewed the cases to determine the true asthma status using a guideline-based criteria key. Discrepancies were adjudicated by a clinical domain expert (pediatric pulmonologist). Results: The CAPriCORN and PheKB algorithms agreed upon the asthma label between 132/200 (66%) cases. The estimated PPV of the CAPriCORN algorithm was 86% and PheKB was 70%. Asthma patients missed by both algorithms were due to human error (miscoded diagnoses that were exclusion criteria, i.e., COPD) or missing data (confirmed asthma based on scanned documentation). The CAPriCORN algorithm was able to identify cases missed by PheKB using medication only rather than medication-plus-asthma ICD code criteria for subsequent encounters. PheKB was able to identify cases missed by CAPriCORN when the confirming information was found in free-text. However, the use of text may have led to more false positives. Both algorithms reported false positives due non-excluded confounding diagnoses and multiple visits in a short timeframe not considered a recurrent episode. Conclusions: The CAPriCORN and PheKB algorithms have strengths and weaknesses to identify pediatric asthma cases from the EHR. We found the algorithms to have complimentary features and next plan to combine these to reinforce a robust asthma computable phenotype algorithm.
Objectives Asthma is a heterogenous condition with significant diagnostic complexity, including variations in symptoms and temporal criteria. The disease can be difficult for clinicians to diagnose accurately. Properly identifying asthma patients from the electronic health record is consequently challenging as current algorithms (computable phenotypes) rely on diagnostic codes (e.g., International Classification of Disease, ICD) in addition to other criteria (e.g., inhaler medications)—but presume an accurate diagnosis. As such, there is no universally accepted or rigorously tested computable phenotype for asthma. Methods We compared two established asthma computable phenotypes: the Chicago Area Patient-Outcomes Research Network (CAPriCORN) and Phenotype KnowledgeBase (PheKB). We established a large-scale, consensus gold standard (n = 1,365) from the University of California, Los Angeles Health System's clinical data warehouse for patients 5 to 17 years old. Results were manually reviewed and predictive performance (positive predictive value [PPV], sensitivity/specificity, F1-score) determined. We then examined the classification errors to gain insight for future algorithm optimizations. Results As applied to our final cohort of 1,365 expert-defined gold standard patients, the CAPriCORN algorithms performed with a balanced PPV = 95.8% (95% CI: 94.4–97.2%), sensitivity = 85.7% (95% CI: 83.9–87.5%), and harmonized F1 = 90.4% (95% CI: 89.2–91.7%). The PheKB algorithm was performed with a balanced PPV = 83.1% (95% CI: 80.5–85.7%), sensitivity = 69.4% (95% CI: 66.3–72.5%), and F1 = 75.4% (95% CI: 73.1–77.8%). Four categories of errors were identified related to method limitations, disease definition, human error, and design implementation. Conclusion The performance of the CAPriCORN and PheKB algorithms was lower than previously reported as applied to pediatric data (PPV = 97.7 and 96%, respectively). There is room to improve the performance of current methods, including targeted use of natural language processing and clinical feature engineering.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.