Poster U20 - Dysfunctional mechanotransduction regulates progression of vascular malformations

Introduction: Vascular malformations (VMs) are abnormalities of the venous, capillary, or lymphatic vessels that are characterized by complex lesions of dilated and dense vascular networks. VMs without arterial involvement, known as slow-flow malformations, harbor significant morbidity, and can be life-threatening. These slow-flow malformations are sporadic and mosaic, and treatment is largely palliative and the most effective treatments are limited to superficial lesions. Systemic treatment and dosing remain a challenge due to toxicity arising from the cellular mosaicism, in which only a fraction of cells in the lesion express the causal gene variants, particularly in pediatric populations. Thus, there exists a significant unmet need for improved understanding of how small populations of mutated cells co-opt vascular beds to result in tissue-scale malformations. Improved understanding of cell-autonomous and non-cell-autonomous processes in lesion development will inform more targeted treatments to address the clinical need for effective interventions with tolerable toxicity. To address this unmet need, we developed a 3D microfluidic model of vascular malformations, which phenocopies clinical presentation of VMs including irregular and dilated microvascular structures. Here, we address the role of the unique biophysical microenvironment in the development of slow-flow malformations, and we identify hemodynamic stress as a key regulator of malformation progression. We further identified dysfunctional Rac1 signaling, a key signaling nexus in the mechanotransduction of substrate and apical fluid mechanical forces, as a driving and targetable mechanism. We discuss efforts to leverage this mechanism for improved treatment of VMs, and leveraging the microfluidic model for patient-specific drug screening.

Materials and

Methods: The majority of slow-flow malformations are caused by activating mutations in PIK3CA, including gain of function single amino acid substitutions in the helical domain of PIK3CA (PIK3CAE542K). Therefore, we used lentiviral transduction to express wildtype PIK3CA (PIK3CAWT) or PIK3CA harboring an activating mutation (PIK3CAE542K) in human umbilical cord endothelial cells (HUVEC). Cells were subject to varying magnitudes of physiologically relevant hemodynamic shear stress (4-11 dyn/cm2), and changes in cell elongation, alignment, and barrier function were quantified. To relate the observed defective fluid mechanotransduction by mutant endothelial cells in 2D culture to the pathophysiological VM phenotype, we next generated engineered microvessels with mutant and control cells embedded in compliant collagen I hydrogels. Microvessels were perfused to induce luminal shear stress, and the interstitial and luminal fluid pressure was independently modulated to induce transmural flow in source and sink configurations. To relate dysregulated flow responses to cellular mechanics, we performed nanoindentation to measure changes in cellular mechanical properties in response to flow for mutant and control cell populations. We further performed immunostaining and colocalization analysis of vinculin, mechanosensitive component of the focal adhesion and adherens junction complexes in control and mutant endothelial cells. We demonstrated that defective cellular stiffening and junctional recruitment of vinculin as consequences of uncontrolled PI3K activation in mutant ECs.

Results, Conclusions, and Discussions: When exposed to laminar flow, PIK3CAWT HUVECs aligned to the local flow direction at higher magnitudes of wall shear stress (~11 dyn/cm2), while PIK3CAE542K HUVECs remained relatively isometric at all magnitudes of shear stress. Interestingly, mutant cells demonstrated a dichotomous relationship between shear stress magnitude and barrier function, with increased permeability as compared to wild-type cells at high magnitudes of shear stress and reduced permeability at lower shear stress magnitudes. We next investigated the functional consequences of the low-flow environment that characterizes microvascular VMs using engineered microvessels (Fig. 1). At baseline under flow conditions, mutant vessels formed dilated and highly irregular vessels, consistent with the clinical phenotype (Fig. 1A). When flow was interrupted and microvessels were left in static conditions after preconditioning with flow, wild-type cells delaminated from the vessel wall, resulting in macroscopic gaps in the vessel (Fig. 1). In contrast, mutant cells were insensitive to reduced perfusion and maintained the tortuous and dilated phenotype observed in perfused culture. In addition to insensitivity to reduced perfusion, we observed increased EC invasion reminiscent of angiogenic sprouting into the subluminal collagen I hydrogel upon application of transmural flow in the basal-to-apical direction (akin to a draining vessel) in mutant microvessels as compared to wild-type. To determine whether the defective fluid mechanotransduction and increased flow-driven angiogenesis was associated with alterations in cellular mechanical properties, we performed nanoindentation to measure the elastic mechanical properties of mutant and wild-type ECs exposed to flow. Interestingly, mutant cells are characterized by a significantly reduced Young’s modulus as compared to wild-type cells. Immunostaining and colocalization analysis revealed that mutant cells demonstrated reduced recruitment of vinculin at intercellular adherens junctions as compared to wild-type cells. Together, these results suggest that hyperactivation of PIK3CA, driven by VM causal mutations, perturbs tensional homeostasis and results in defective mechanotransduction of fluid stresses, and thus suggest that these cells are conditioned for improved survival and growth in the native slow-flow environment. The results further suggest that targeting defective mechanotransduction, particularly via manipulation of Rac1 and effector signaling, could represent an avenue to reduce or regress lesion progression.

Acknowledgements (Optional): We would like to acknowledge funding support from NIH, AHA, and the Lymphatic Malformation Institute.