Shear stress unveils patient-specific transcriptional signatures in PAH: Towards personalized molecular diagnostics.

Abstract

Rationale: Pulmonary arterial hypertension (PAH) is a life-threatening disorder characterized by increased pulmonary blood pressures and regional inhomogeneities in flows, with diagnostic and treatment challenges arising from diverse underlying pathogenic mechanisms. Conventional in vitro models often obscure the mechanistic nuances of PAH by failing to replicate the dynamic mechanical environment of the diseased lung, limiting the identification of specific molecular patterns. To address this, we employed an in vitro shear stress model simulating physiological or pathological conditions to explore the transcriptional heterogeneity of human pulmonary microvascular endothelial cells (hPMECs) from PAH patients and healthy controls within their respective biomechanical context. Methods & Results: hPMECs from PAH patients and controls were exposed to static, low shear stress (LSS), and high shear stress (HSS) conditions, followed by bulk RNA-sequencing. While increasing shear stress resulted in a greater number of differentially expressed genes, traditional grouped analysis showed minimal overall transcriptional differences. Further, pathway enrichment analysis indicated common shear-induced responses in both groups, suggesting that standard analysis methods may mask meaningful disease-specific changes. Crucially, detailed dimensionality reduction analyses revealed pronounced inter-patient variability among PAH donors in response to increasing shear stress, facilitating the identification of 398 genes driving this transcriptional heterogeneity. Unsupervised clustering of these high-variability genes enabled the sub-classification of patients based on their unique transcriptomic profiles, each linked to specific combinations of PAH associated pathogenic pathways such as mesenchymal transition, inflammation, metabolism, extracellular matrix remodeling, and cell cycle/DNA damage signaling. Importantly, re-analysis of published peripheral blood mononuclear cell (PBMC) omics data from PAH patients confirmed the clinical feasibility to utilize these high-variability genes as a non-invasive, accessible approach for molecular patient stratification. Conclusion: Our study uncovers patient-specific transcriptomic patterns in PAH, providing a novel molecular sub-classification strategy. These findings represent a significant step toward personalized molecular diagnostics in PAH and eventual therapeutic interventions for clinically well-defined PAH patients, with potential applications in clinically accessible cell populations such as PBMCs.