Controversy in mechanotransduction: the role of endothelial cell-cell junctions in fluid shear stress sensing

The Authors declare no competing interests. Data availability: authors will make all data available on reasonable request. Abstract Fluid shear stress (FSS) from blood flow, sensed by the vascular endothelial cells (ECs) that line all blood vessels, regulates vascular development during embryogenesis, controls adult vascular physiology and determines the location of atherosclerotic plaque formation. While a number of papers that reported a critical role for cell-cell adhesions or adhesion receptors in these processes, a recent publication challenged this paradigm, presenting evidence that ECs can very rapidly align in fluid flow as single cells without cell-cell contacts. To address this controversy, four independent laboratories assessed EC alignment in fluid flow across a range of EC cell types. These studies demonstrate a strict requirement for cell-cell contact in shear stress


Introduction
FSS acting upon the vascular endothelium is required for critical steps during vessel morphogenesis (Garcia-Cardena and Slegtenhorst, 2016;Roman and Pekkan, 2012) and in controlling blood flow and vessel remodeling in adults physiology (Baeyens et al., 2016;Lu and Kassab, 2011).Atherosclerotic plaques form preferentially at regions of lower and multidirectional FSS, termed disturbed FSS (Chatzizisis et al., 2007;Souilhol et al., 2020).One of the earliest observed response to physiological levels of FSS is the elongation and alignment of ECs in the direction of flow, which occurs over 6-24 hours (Blackman et al., 2002;Dewey, 1984;Galbraith et al., 1998;Levesque and Nerem, 1985;van der Meer et al., 2010).By contrast, ECs fail to align in disturbed FSS, which has been identified as a causal factor in activation of the inflammatory pathways that drive atherogenesis (Wang et al., 2013).
These receptors were also found to modulate flow-dependent vascular remodeling in vivo and development of atherosclerotic plaques (Chen and Tzima, 2009;Goel et al., 2008;Harry et al., 2008;Mehta et al., 2020;Stevens et al., 2008).Indeed, nanoparticles that target these molecules are under development for therapeutic applications (Khodabandehlou et al., 2017).
However, a recent high-profile publication cast some doubts upon these findings (Mylvaganam et al., 2022).This paper described a novel mechanism mediating the well-known elongation and alignment of vascular endothelial cells in the direction of fluid flow.Specifically, they reported that in an immortalized endothelial cell line, teloHAEC, alignment in the direction of fluid flow occurred over 30-60 minutes and occurred in sparse cultures where cells lack cell-cell contacts.
Given the importance of these issues to our understanding and treatment of vascular disease, we investigated the timing and cell contact dependence of flow-induced alignment in multiple EC types.

Results and Discussion
We set out to determine the time course and contact dependence for EC alignment in the direction of flow across a wide range of endothelial cell types.We examined human umbilical vein endothelial cells (HUVECs), human coronary artery endothelial cells (HCAECs), human aortic endothelial cells (HAECs), and the same teloHAEC line from ATCC used by Mylvaganam We conclude that all EC types tested align in flow over multiple hours and require cell-cell contacts.Thus, the results of Mylvaganam et al cannot be replicated by experienced investigators in four independent laboratories.These new findings cement the current paradigm that junctional receptors play a central role in vascular responses to fluid flow, which clears the way for further progress in this area.

Methods
HUVEC were obtained from the Yale Vascular Biology and Therapeutics core facility.Each batch contains cells pooled from three deidentified donors.HUAEC, HAEC, HCAEC were obtained from promoc-cell and TeloHAEC from ATCC.HUVEC, teloHAEC, HUAEC, and teloHAEC (SG) were cultured in Endothelial Cell Growth Medium-2 with bullet kit (Lonza) and 5% FBS and were seeded on fibronectin coated glass slides at 100% vs 10% of confluent density (confluent vs Journal of Cell Science • Accepted manuscript sparse).Cells were subject to laminar shear stress at 15 dynes/cm2 using parallel plate flow chambers for the indicated and times.HAEC and HCAEC were cultured in Endothelial Cell Growth Medium MV2 + supplements from Promocell (Catalogue number C-22022; Fetal Calf Serum 0.05 ml/ml, Epidermal Growth Factor 5 ng/ml, Basic Fibroblast Growth Factor 10 ng/ml, Insulin-like Growth Factor 20 ng/ml, Vascular Endothelial Growth Factor 165 0.5 ng/ml, Ascorbic Acid 1 µg/ml, Hydrocortisone 0.2 µg/ml) + 100 μg/ml streptomycin, 100 U/ml penicillin, 2.5ug/ml Amphotericin B. Cells were seeded on gelatin coated 0.4um Ibidi® slides at 10% vs 100% confluency and an Ibidi® pump system was used for 12dyn/cm2 shear stress for indicated times.All cells were cultured at 37° C and 95% humidity.For imaging, cells were fixed in 4% PFA, permeabilized, and stained for F-actin with rhodamine-phalloidin. Alignment in the direction of flow was quantified using a custom script which bins F-actin fibers by direction in 15° increments.This script is available at: https://github.com/sudo-shaka/Actin-Alignment/.
HUVEC and teloHAEC rose plots combine counts across three independent replicates with > 100 cells per condition.Other cell line data consists of one replicate with > 50 cells per condition.
and co-workers.Cells were examined in confluent cultures and at low density using shear stress in the standard range of 12-15 dynes/cm2, similar to the 15 dynes/cm2 used in Mylvaganam et al.Confluent cells aligned over 6-24h as expected, with no detectable alignment at 1h. Single cells from the sparse cultures showed much less if any alignment over this period (Fig 1).These results thus confirm the reported rate and cell-cell contact dependence of endothelial cell alignment across multiple endothelial cell types.Dr. Grinstein, the senior author in the Mylvaganam paper, kindly sent us the specific teloHAEC line used in his study and communicated details of the conditions used for the paper.However, we still see alignment over 12-24h and no alignment in sparse cultures (Fig 1B).

For
each cell line and condition, the percentage of aligned actin fibers was calculated.Actin fibers were considered aligned if the orientation angle was < 45° from the flow direction.Each point represents a cell line at that condition (Fig 1C).Statistics were determined using two-way ANOVA comparing flow conditions and confluence; ** p<0.005; ***p< 0.005; **** p<0.001.