Substrate regulation of vascular endothelial cell morphology and alignment

1

Keywords:

• Endothelial cells
• morphology
• alignment
• atherosclerosis
• patterned surfaces

1. Introduction

In many cell types, function is strongly tied to shape. This is particularly true for vascular endothelial cells (ECs) where polygonal (round) shapes have been correlated with a pro-inflammatory phenotype that is susceptible to atherosclerosis whereas an elongated cellular morphology and alignment in the direction of blood flow correspond to an anti-inflammatory profile that remains largely protected from the disease. Therefore, understanding how EC shape and orientation are regulated is of importance.

EC shape and orientation have long been known to be regulated by the flow field to which the cells are subjected; however, recent in vitro studies have demonstrated that EC shape can also be regulated by the substrate on which the cells are cultured. For instance, significant EC elongation can be induced by culturing the cells on planar patterned surfaces with selectively-defined motifs of adhesive and non-adhesive zones. Cell elongation can also be induced by plating the cells on topographic surfaces consisting of series of nano- or micro-scale gratings (ridges and grooves). It remains unclear, however, how ECs perceive these different types of patterned substrates and if the effects of the planar bio-adhesive and the topographic substrates on ECs occur via similar mechanisms. The goal of the present study is to quantitatively characterize EC elongation and alignment on both planar and topographic patterned surfaces of similar dimensions and to shed light onto the underlying mechanisms.

2. Methods

Planar micropatterned (µP) substrates containing alternating 5 µm-wide adhesive and non-adhesive stripes were produced on PDMS using the deep UV light method (Azioune et al. 2010.). Topographic micrograting (µG) substrates were fabricated by replica molding of PDMS on a silicon master containing straight channels with a groove/ridge width of 5 µm and a ridge height of 1 µm. In some experiments, other ridge/groove dimensions and larger ridge heights (up to 7 µm) were also investigated. Unpatterned PDMS substrates served as controls. Prior to cell seeding, all substrates were incubated with a 50 µg/mL fibronectin solution in PBS for 1 hr. In other experiments, fibronectin was adsorbed selectively on the ridges (but not the grooves) of the topographic micrograting substrates by using microcontact printing to produce a topographic surface (µG-FnR) with similar adhesive regions as the µP substrate. Bovine aortic ECs (BAECs) were seeded on the different substrates, fixed and immunostained for vinculin (FAs) and actin either 2 or 24 hrs after seeding. Quantification of FA distribution and of cell alignment and elongation was performed using Fiji software.

3. Results and discussion

On the µG substrates, FAs were distributed uniformly on both the ridges and grooves. In contrast, on the µP and µG-FnR substrates vinculin concentration exhibited sharp peaks along the edges of the adhesive zones (adhesive lines on µP and functionalized ridges on µG-FnR), suggesting FA clustering in those regions. To understand how the FA distribution affected cell morphology and orientation, cell elongation and alignment relative to the pattern direction were quantified. ECs on all patterned substrates were aligned in the pattern direction while cells on unpatterned surfaces expectedly showed a random orientation. Interestingly, ECs on the µP substrates were significantly more elongated than cells on the unpatterned substrates as expected but surprisingly also then cells on the µG substrates. ECs on the µG-FnR substrate were even more elongated than those on the µP substrate, suggesting that FA organization is the primary determinant of EC elongation.

FAs organize actin stress fibers in cells; therefore, we wondered if FA clustering along pattern edges led to a particular stress fiber organization. To address this issue, we used confocal microscopy to visualize the spatial distribution of actin filaments in ECs cultured on all the different surfaces. In ECs cultured on unpatterned substrates, stress fibers were randomly oriented with dense peripheral actin microfilament bundles, typical of ECs cultured under static (no flow) conditions (Prasain and Stevens 2009). On the µP substrates, bundles of actin fibers at the EC basal plane were present in between adjacent fibronectin stripes, bridging FAs located at the borders of neighboring adhesive areas. In ECs cultured on µG surfaces, confocal z-stack imaging revealed two distinct stress fiber arrangements: stress fibers on the ridges had no clear spatial organization, whereas stress fibers in the grooves formed packed bundles oriented in the pattern direction. These bundles were associated with long and well aligned FAs detected inside the groove, suggesting that the groove surface provided directional guidance for the spatial organization of stress fibers. When ECs were cultured on µG-FnR surfaces in which the groove was no longer accessible for adhesion, the actin network exhibited similarities to that in cells on µP substrates, most notably suspended thick stress fiber bundles that connected FAs located at the boundaries of neighboring adhesive ridges and thus presumably formed suspended bridges between adjacent ridges.

A key question is how FA clustering and the resulting stress fiber organization promote cellular elongation. One possibility is that the contractile stress fiber cables that run laterally over the non-adhesive stripes on the µP and µG-FnR substrates reduce the capacity of ECs to extend orthogonal to the pattern in a manner somewhat similar to the mechanism proposed by Thery et al. (2006). A second possibility relates to the dynamic nature of FAs, which can glide on surfaces due to traction forces in a treadmilling-like manner (Wolfenson et al. 2009). In the case of the µP and µG-FnR substrates, FAs are separated by the non- adhesive stripes or grooves. The resulting inhibition of FA movement acts to resist stress fiber contraction, ultimately stabilizing FA-stress fiber assembly. The validity of either or both of these potential mechanisms of cellular elongation remains to be investigated.

4. Conclusions

The present findings indicate that on patterned substrates, FA clustering regulates EC morphology through an effect on cytoskeletal organization. These results provide new insight into how substrate topography and patterning regulate EC shape and orientation and promise to inform strategies of substrate engineering to target specific EC functionality.

Additional information

Funding

Work supported in part by an endowment in Cardiovascular Bioengineering from the AXA Research Fund and a research grant from the Fondation Lefoulon Delalande.