HUVEC were obtained from PromoCell (ref. C-12200) and cultured in M199 supplemented with 20% foetal bovine serum FBS (Gibco), 1% antibiotics (penicillin/streptomycin, Gibco), 1% L-glutamine (2 mM Gibco), 1% heparin (500umh/mL, Gibco) at 37 °C in a humidified incubator with 5% CO₂. The medium was replaced every 2–3 days and HUVECs were grown to near confluence and passaged when required. For cell passage, the media was aspirated and replaced with 4 mL of trypsin. After a 10-minute incubation, the cell suspension was neutralised with 2 mL FBS and centrifuged at 200 g for 5 minutes. The pellet was resuspended in 1 mL of media and transferred into 75 cm² culture flasks at a resulting density of 10,000 cells/cm².

Collagen (Coll, purified bovine type I collagen, ref CBPE2, Symatese) and fibronectin (Fn, human plasma fibronectin, Ref. 11051407, Roche Diagnostics) were used in this study as single- and double-layered ECM coatings. Each protein was diluted according to the manufacturer's guidelines. Fn was used at a concentration of 2 μg/mL in phosphate buffer saline (PBS) (120 ng/cm²), Coll was used at 1 mg/mL in distilled water (60 μg/cm²). This coating protocol has been described by Nagel et al. [20] and a Coll coating at this concentration has been shown to enable fibril formation in vitro [21].

ECM protein coated polystyrene culture plates were used for studies with and without an ECM coating. The Fn solution was allowed to adsorb on polystyrene for 45 minutes, at 37 °C and washed with PBS 3 times, for 10 minutes. The Coll was left to adsorb for 10 minutes at room temperature and then washed twice in PBS for 10 minutes. The protocol for the Coll + Fn coating was as described for the single coating, with the Coll coated first, followed by the Fn. Petri dishes were freshly coated prior to each application.

Cells were seeded at a density of 10,000 cells/cm² on freshly coated polystyrene culture plates. After 24 and 48 hours in culture, the samples were rinsed in PBS to remove non-adherent cells. The cells were then trypsinized, and counted with trypan blue (Sigma-Aldrich). The proliferation rate was calculated by dividing the number of living cells by the number of cells seeded. The experiment was conducted in triplicate and repeated three times (n = 9).

The Oris™ cell migration assay (Platypus Technologies) was used to assess the cell migration. Micro-titration plates (96-well) with a black border were coated with Coll, Fn and Coll + Fn and migration stoppers were inserted into each well. HUVEC were seeded around the perimeter of the stoppers at a density of 20,000 cells/cm². After a 24-hour incubation, the stoppers were removed and cells were cultured for a further 24 hours to enable migration into the detection zone, then the cells were fixed and labelled with Fluorescein Isothiocyanate (FITC, 2 mg/mL for 2 minutes, Sigma-Aldrich). Each well was thoroughly washed with PBS to remove any FITC excess. Migration was observed by quantifying FITC emission in the detection zone by using a BioTek Synergy™ HT Multi-Mode Microplate Reader at 525 nm. Data represent the mean ± SE from 9 wells for each condition.

FITC labeled Fn and Rhodamine Isothiocyanate (Rhod) labeled Coll, were prepared as followed, 1 mg of Fn or 1 mg of Coll, was dissolved in 1 mL of NaHCO₃ 0.1 M pH 9, and incubated for 1 hour at RT with 10 μL FITC (F351, Sigma-Aldrich) or 10 μL of Rhod (R1755, Sigma-Aldrich^©) both 10 mg/mL in dimethyl sulfoxide (DMSO). The unbound fluorochromes were separated on PD-10 desalting columns (GE Healthcare). Fn-FITC and Coll-Rhod were coated on polystyrene culture plates as the non-labeled ECM proteins. Cells were seeded at 20,000 cells/cm² and incubated for 2 hours at 37 °C. After the incubation period, the culture medium was eliminated and the polystyrene culture plates were observed under fluorescence microscope (DMI600, Leica System).

The HUVEC morphology was examined by different microscopic methods. Cells were seeded at 10,000 cells/cm² and cultured on ECM coated surfaces. Fluorescence microscopy was used to determine the cytoskeleton arrangements of HUVEC. Twenty-four hours post-seeding, HUVEC were fixed with 4% formaldehyde overnight, permeabilized in 0.1% Triton X-10 and incubated in 2% bovine serum albumin (BSA) for 1 hour at RT. The cells were then incubated in the dark for 1 hour at RT with a cytoskeleton stain containing phalloidin (25 μL/mL) (phalloidin-X5, Fluoprobes) and DAPI (1 μg/mL, Sigma-Aldrich) diluted in PBS. Cells were mounted in Mowiol (PolySciences) and observed with a fluorescence microscope (DMI600, Leica System).

The cells were observed with a Scanning Electron Microscope (SEM). Forty-eight hours post-seeding cells were fixed with 4% formaldehyde for 24 hours. The HUVEC were then covered with a layer of gold (about 50 nm) for 3 minutes under 25 mA, using a SEM coating unit E5100 (Polaron Equipment Ltd) and viewed under the SEM (ESEM FEG XL30 Philips).

The primary antibodies used to determine protein domain accessibility were polyclonal Fn (pFn; A0245, Dako), polyclonal Coll (pColl; AB749, Millipore) and monoclonal RGD (mRGD, MAB 1926, Millipore). The freshly coated 6-well polystyrene culture plates were saturated with 0.1% Tween20 + 0.25% BSA in PBS for 30 minutes at 37 °C. PBS + 0.1% Tween20 was used as the wash solution following each incubation. Primary antibodies were diluted 1/5000 in PBS and incubated for 1 hour at 37 °C. After washing, the horseradish peroxidase-conjugated secondary antibodies (goat anti-Rabbit and anti-mouse, Interchim) were diluted at 1:10,000, incubated at 37 °C for 1 hour on the corresponding antibodies. The wells were then incubated for 2 minutes in the substrate, composed of ortho-phenylene-diamine (0.5 mg/ml, Sigma-Aldrich) and H₂O₂ (0.4 μL/mL) diluted in citrate buffer. The reaction was blocked with 100 μL of 1 M HCl. The absorbance was obtained at 490 nm in a microplate reader (Biorad Laboratories). The results were represented as the absorbance of the coated wells, subtracted by the control well (all components of the assay, except the primary antibody). The results were expressed as the mean absorbance of two samples per experiment, over three individual experiments ± SE (n = 6).

Analysis of the data was conducted using the statistical program GraphPad Instat (version 3.01). Unless otherwise stated, the data were represented as the mean ± SE of 9 measurements, corresponding to 3 samples of coated polystyrene for each ECM protein repeated 3 times. Comparisons across the 4 experimental groups were determined by a one-way ANOVA followed by the Newman-Keuls Multiple Comparison post hoc test. A value of P < 0.05 was considered significant.

Cultured HUVEC were seeded on different extracellular matrix proteins and the proliferation was measured at 24 and 48 hours post-seeding. As shown in Fig. 1, the proliferation of HUVEC at 24 hours was similar across all four conditions; there was no significant difference in proliferation at 24 hours. However, a significant difference (P < 0.05, one-way ANOVA) became visible after 48 hours. Fig. 1 shows that endothelial cell proliferation was significantly higher on a double coating (3.53 ± 0.62) compared to polystyrene (2.66 ± 0.45) and both collagen and fibronectin single coatings (Coll: 2.66 ± 0.31; Fn: 2.73 ± 0.17). These results show that HUVEC prefer to grow on a double-coated surface of collagen + fibronectin.

Fig. 2 shows that a significant increase (P < 0.001; one-way ANOVA) in HUVEC migration was observed on both a single collagen coating and the collagen + fibronectin coating compared to all other conditions. This result suggests that collagen plays a role in supporting HUVEC migration across a surface. Fig. 2 also indicates that the migration property observed in a collagen + fibronectin coating is similar to that seen in the collagen-only coating, suggesting collagen can somehow interact with the HUVEC even though it is coated underneath the fibronectin.

To determine if a uniform layer of extracellular matrix proteins was adsorbed when the polystyrene was coated, the samples were analysed by fluorescence microscopy. Prior to coating, the markers FITC and Rhodamin were chemically linked to fibronectin and collagen respectively. After 2 hours in culture, the coatings were observed, and the only protein visible was the one coated and not the one produced by the cells. As shown in Fig. 3a and b, Rhodamin and FITC were visible on their respective extracellular matrix molecules as collagen and fibronectin molecules were evenly distributed across the surface of the polystyrene, visible by the uniform coverage of the fluorochrome. Fig. 3c and d demonstrate that on the double coating, collagen and fibronectin were co-distributed across the surface and could be found at identical locations (Fig. 3e) on the polystyrene. The results of Fig. 3 also show that the concentration of collagen and fibronectin was higher around the perimeter of the HUVEC, in both the single and double coatings.

To further explore the behaviour of HUVEC in the presence and absence of an extracellular coating, phalloidin-staining of the actin cytoskeleton was examined. Because the HUVEC proliferation rate at 24 hours was not significantly affected by an extracellular matrix coating, it was important to determine whether cell morphology was affected. As shown in Fig. 4, an extracellular matrix coating stimulated the HUVEC to rearrange their cytoskeleton and form filopodia. The reorganization of the HUVECs cytoskeleton caused the actin filaments to protrude from their lamellipodium frontier, to form filopodia. The presence and quantity of filopodia did not differ depending on the extracellular matrix coating (Fig. 4b, c and d). However, in the absence of coating there was no filopodia formed, only lamellipodia were visible on the HUVEC (Fig. 4a).

To determine whether the protein binding interaction caused changes in cell morphology, HUVEC were seeded on the different proteins and examined using SEM. As displayed in Fig. 5, HUVEC shape was affected by the extracellular matrix coating. In Fig. 5a and c, it is seen that the morphology is similar for non-coated and fibronectin-coated polystyrene. The cells were flat, round with a large appearance. However, HUVEC grown on a collagen or collagen + fibronectin surface had an elongated shape and a raised appearance (Fig. 5b and d). This suggests that collagen plays some role in directing cell morphology. These differences on the collagen + fibronectin coating compared to the other coatings correspond to cord-like structures forming, that could be precursors of angiogenesis [22].

The accessibility of fibronectin RGD domain (arginin-glycin-aspartic acid sequence), when coated alone or combined with collagen, was determined by ELISA. Our results show that there was a significant decrease in adsorbance with both monoclonal and polyclonal fibronectin antibodies on the double coating compared to the single coating. The decrease of adsorbance read from polyclonal fibronectin (pFn) and monoclonal RGD (mRGD) was 26.5% (P < 0.001) and 82.4% (P < 0.001), respectively. There was no significant difference in the adsorbance read from polyclonal collagen (pColl) on a double or single coating (Table 1). These results suggest that collagen and fibronectin interact and/or bind in a RGD dependent manner.