Aligned nanofibrillar collagen regulates endothelial organization and migration.

We developed a flow processing technique to generate anisotropic nanofibrillar collagen. Human ECs were cultured on aligned or on randomly oriented collagen, and their cellular alignment and cytoskeletal organization were characterized by immunofluorescence staining and time-lapse microscopy.

2 Materials and Methods 2.1 Fabrication of aligned collagen strips and grafts High concentration rat tail collagen-Type I (BD Biosciences) was initially prepared at a stock concentration of 9–10 mg/mL in 0.02N Acetic Acid, pH 3.5 and dialyzed to 30 mg/mL using a previously published method [ 14 ]. Using a 1 mL plastic syringe with a 22 gauge blunt needle, the dialyzed collagen was manually deposited as thin strips of collagen onto a glass slide ( Fig. 1A ). The syringe moved at an approximate rate of 340 mm/s, while the collagen was extruded at 3.2 mL/min. In contrast, strips of non-aligned collagen fibrils, or the randomly oriented control, were produced by moving the syringe at a significantly slower velocity, 50 mm/s. The slow syringe velocity was also coupled with a slow extrusion rate, 0.02 mL/min, to produce strips of random collagen with similar thickness and width as the aligned collagen strips. The glass slide and the needle were submerged in 37°C, 10X-phosphate buffered saline (PBS), pH 7.4, during the procedure in order to initiate fibrillogenesis immediately as the collagen was deposited. The collagen completely dried under a laminar flow hood overnight and rinsed several times with MilliQ water afterwards. The collagen strips remained adhered onto glass slides ( Fig. 1A ). This method produces randomly oriented collagen strips that are ~1.1 × 0.8 × 30 mm and aligned collagen strips that are ~1.0 × 1.2 × 30 mm. In order to produce a three-dimensional aligned collagen conduit for potential use as a vascular graft, the collagen was extruded using an aluminum syringe in a plunger motion, as described previously [ 14 ]. The collagen grafts then received a cross-linking treatment of 50 mM 1-ethyl-3-(3-dimethyl aminopropyl) carbodiimide hydrochloride, EDC, (Pierce) in 100% ethanol, agitated overnight at room temperature. The EDC solution was then aspirated and the grafts were placed in a vacuum desiccator for 2 hours or until the collagen was completely dehydrated. In preparation for cell culture or mechanical testing, the grafts were rehydrated in PBS. Scanning Electron Microscopy (SEM) Samples were processed for routine SEM, according to previous publications [ 15 ], using a Hitachi S-3400N VP scanning electron microscope (Hitachi Ltd). In vitro culture experiments Primary human dermal microvascular ECs were grown in EGM2-MV (Lonza) media. Prior to in vitro studies, the collagen substrates were sterilized in 70% ethanol and then incubated in PBS before cell seeding. ECs were plated at a seeding density of 23–28k cells/cm 2 and incubated for 24 hours (collagen strips) or 72 hours (vascular grafts).

Immunofluorescence staining

At indicated time points, samples were fixed in 4% paraformaldehyde. Samples were immunofluorescently stained using antibodies against CD31 (Dako Cytomation), Ki67 (DakoCytomation), or paxillin (BD Transduction Labs), according to previous publications [ 16 , 17 ]. In addition, phalloidin (Invitrogen) was used to visualize the F-actin assembly. All nuclei were visualized by Hoechst33342 dye and imaged on a Nikon fluorescence microscope or a Leica SP2 Acousto-Optical Beam Splitters confocal microscope.

Analysis of cellular alignment and proliferation

Cell alignment was measured, using ImageJ (NIH, v.1.42), as the angle formed by the cell’s principal axis, relative to the direction of the collagen nanofibrils, where a minimum angle of 0° denoted parallel alignment from the axis of the nanofibrils, and a maximum of 90° suggested perpendicular alignment [ 16 ]. For quantification of randomly oriented fibrils, an arbitrary axis was selected. Proliferation was defined as the percentage of total nuclei expressing the proliferation marker Ki67. The cell shape index (CSI) was equivalent to (4π*Area)/(Perimeter 2 ), where the value of 1 approximated the shape of a circle and a value of 0 depicted that of a straight line [ 18 ]. Cells cultured on aligned or random collagen were imaged in five representative fields, and 25 cells in each field were evaluated. The data are shown as mean ± standard deviation (n≥3). A Student’s t -test was compared between two groups, and significance was accepted at P