Development of a biomimetic spiral artery remodeling model

Introduction: : Preeclampsia (PE) is the leading cause of maternal/fetal morbidity/mortality in the United States, accounting for more than 70,000 maternal and 500,000 fetal deaths every year. The leading mechanistic hypothesis for pathogenesis is the dysregulation of spiral artery remodeling (SAR)—where invasive placental cells (extravillous trophoblasts, EVT) remodel maternal vasculature to meet metabolic needs of the growing fetus in the 2nd and 3rd trimesters. Although there have been great efforts in developing a mechanistic understanding for PE, we have yet to define clinically translatable mechanisms for disease presentation and treatment. This is largely attributed to our inability to replicate the spatiotemporal microenvironmental intricacy required throughout SAR in current models.

Herein, we adapt of a new generation of polystyrene (PS) microfluidic chips (Fig.1a) to generate biomimetic vascular networks using a human endometrial microvascular endothelial cell (HEMEC): normal human lung fibroblasts (NHLF) self-assembly. These chips consist of 28 individual wells each with 3 separate compartments (Fig.1b) that can be used to pattern different configurations of cells and matrix to robustly self-assemble vascular networks in as little as 5 days of culture.

Materials and

Methods: : Cell Culture
HEMEC (ScienCell P3-P5) were cultured in EGM-2MV BullitKit (Lonza) or on 2ug/cm2 fibronectin and ScienCell media. NHLF were cultured per vendor instructions (Lonza; P2-P9).

Vascular Self Assembly
PS chips were treated with plasma (high RF, 1.5-3 min) to pattern individual compartments via differential surface tension introduced by PS beam heights (Fig. 1b). Fibrin/Collagen-I matrix (2.5 mg/mL Fibrinogen, 0.12 mg/mL Collagen-I, 0.15 U/mL Aprotinin, 0.5 U/mL Thrombin) containing different HEMEC concentrations (1x107- 2x107 cell/mL) and NHLFs (1x107 cell/mL) in charcoal-stripped FBS media was patterned in the central channel at room temperature for 10 min. HEMEC ( 5x106 cell/mL) side walls were patterned by pipetting 3 uL of cells in media at a differential location along the well side wall. Chips were placed at 90 degrees for 10 min at 37C on each side to allow gravity to pull cells down towards the central gel. 100 uL of charcoal-stripped media was added to both reservoirs (R) and changed daily for 5 days. Hydrostatic pressure was introduced (R=100 uL and 20 uL) after first day of coculture to perfuse chips with 500nM S1P; media was changed daily.

Immunofluorescence Imaging
Cells were fixed with 4% paraformaldehyde for 15 min, blocked with 3% BSA solution for 1 day, and stained with UEA-I Dylight 649 (1:200; Vector Labs) overnight at 4C. DAPI (1:1000) staining was done for 30 min at room temperature. Confocal z-stacks (Olympus FV3000) were taken (10x magnification), background subtracted (FIJI), and z-projected (maximum intensity).

Results, Conclusions, and Discussions:: Lumenized endometrial vascular networks were formed using all cell concentrations and media conditions after 5 days (Fig. 1c-f). However, fibrin gels containing ScienCell experienced a greater degree of background noise compared to gels composed of EGM-2MV prf media (Fig. 1c-d). We hypothesize that this is due to differential gel properties invoked by ScienCell basal media salt concentrations. We were able to validate this macroscopically using a 24 well gel—ScienCell gels did not form until after overnight incubation and resulted in a more tightly packed gel compared to the EGM-2MVprf fibrin gels (data not shown). EGM-2MV will be used as coculture and gel media in future experiments, given that the high degree of background noise observed in the ScienCell conditions will inhibit our ability to quantitate vascular structure using an image analysis (Angiogenesis Analyzer; FIJI) approach in future studies. Differential gelation kinetics may also explain significant differences between vascular networks observed in ScienCell (globular) and EGM-2MV (interconnected) conditions (Fig. 1c-d). We also observed a plateau in vascular network interconnectedness with respect to HEMEC densities (Fig. 1d-e); future studies will be done to validate this phenomenon and identify optimal seeding densities. 500 nM SIP treatment increased interconnectedness of vascular networks (Fig.1e-f). VEGFa and bFGF will also be added to 500nM S1P in order increase endometrial vascular network formation. Given that endometrial vascularization is largely dictated by sex hormones (estrogen and progesterone), we aim to validate whether sex hormones can be used to generate lumenized interconnected vascular networks in vitro—similarly to generically used proangiogenic factors (S1P, VEGFa, and bFGF). These studies will serve as the foundation for the development of a biomimetic spiral artery remodeling (SAR) model capable of hierarchically characterizing vascular tissue structure-functions during normal and PE-related SAR, particularly advancing our understanding of cell-cell and cell-ECM regulation of EVT-mediated remodeling.

Acknowledgements and/or References (Optional):: This project was supported by the National Center for Advancing Translational Sciences, National Institutes of Health, through BU-CTSI Grant Number 1TL1TR001410. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH.

This research was supported in part by the National Institutes of Health training grant at Boston University, T32 EB006359. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Research reported in this publication was supported by the Boston University Micro and Nano Imaging Facility and the Office of the Director, National Institutes of Health of the National Institutes of Health under award Number S10OD024993. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of Health.