MMP-sensitive PEG diacrylate hydrogels with spatial variations in matrix properties stimulate directional vascular sprout formation

Abstract

The spatial presentation of immobilized extracellular matrix (ECM) cues and matrix mechanical properties play an important role in directed and guided cell behavior and neovascularization. The goal of this work was to explore whether gradients of elastic modulus, immobilized matrix metalloproteinase (MMP)-sensitivity, and YRGDS cell adhesion ligands are capable of directing 3D vascular sprout formation in tissue engineered scaffolds. PEGDA hydrogels were engineered with mechanical and biofunctional gradients using perfusion-based frontal photopolymerization (PBFP). Bulk photopolymerized hydrogels with uniform mechanical properties, degradation, and immobilized biofunctionality served as controls. Gradient hydrogels exhibited an 80.4% decrease in elastic modulus and a 56.2% decrease in immobilized YRGDS. PBFP hydrogels also demonstrated gradients in hydrogel degradation with degradation times ranging from 10-12 hours in the more crosslinked regions to 4-6 hours in less crosslinked regions. An in vitro model of neovascularization, composed of co-culture aggregates of endothelial and smooth muscle cells, was used to evaluate the effect of these gradients on vascular sprout formation. Aggregate invasion in gradient hydrogels occurred bi-directionally with sprout alignment observed in the direction parallel to the gradient while control hydrogels with homogeneous properties resulted in uniform invasion. In PBFP gradient hydrogels, aggregate sprout length was found to be twice as long in the direction parallel to the gradient as compared to the perpendicular direction after three weeks in culture. This directionality was found to be more prominent in gradient regions of increased stiffness, crosslinked MMP-sensitive peptide presentation, and immobilized YRGDS concentration.

Figure 1. PBFP hydrogel generation and gradient evaluation.

(A) Schematic of glass frit perfusion chamber and (B) resultant hydrogel. (C) Hydrogel sectioning for gradient evaluation as well as (D) aggregate seeding in hydrogels (note: hydrogels were rotated 90° to facilitate aggregate placement in the z-direction) and (E) aggregate location by region. In all cases, the gradient runs in the y-direction with the top of the gel taken as the staring point for gradient characterization.

Figure 2. Quantification of aggregate invasion.

Schematic representation of the methods used to quantify aggregate invasion by (A) total projected area as well as directional vascular sprout formation via (B) anisotropy index and (C) sprout length by angle.

Figure 3. Quantification of hydrogel mechanical and biofunctional gradients.

Spatial variations of (A) hydrogel elastic modulus and (B) immobilized concentrations of YRGDS in MMP-sensitive PBFP gradient hydrogels as well as equivalent bulk control gels (# = p<0.05 as compared to top section (0–2 mm) of gel; n = 3). Error bars represent ± standard deviation.

Figure 4. Quantification of hydrogel degradation kinetics.

Degradation kinetics of hydrogel sections taken from different regions of (A) bulk polymerized control gels and (B) PBFP gradient hydrogels when exposed to 1 µg/mL collagenase-1A (MMP-1) (n = 3). Error bars represent ± standard deviation.

Figure 5. Coordinated cellular interactions of vascular sprout invasion within PEGDA hydrogels.

Confocal images taken using 20× magnification at (A) 24 and (B) 48 hours post-aggregate placement in hydrogels. Fluorescently labeled endothelial cells appear in red and smooth muscle cells appear in green (scalebar = 50 µm).

Figure 6. Vascular sprout invasion within PEGDA hydrogels.

Flattened 3D mosaic renderings of co-culture aggregates seeded in (A) bulk control hydrogels as well as in the (B) top, (C) middle, (D) and bottom regions of PBFP gradient hydrogels. Aggregates were fixed after 21 days in culture and stained for F-actin (Note: the gradient runs in the y-direction in B, C, and D with elastic modulus and YRGDS concentration increasing in the direction of the top of the image; scalebar = 200 µm).

Figure 7. Three-dimensional vascular sprout invasion in PEG Scaffolds.

3D fluorescent image reconstructions of vascular sprout invasion in (A) bulk control gels and (B) PBFP gradient hydrogels taken at 10× magnification (Note: the gradient runs in the y-direction in B).

Figure 8. Characterization of vascular sprout invasion.

(A) Total projected area, (B) quantile-quantile (Q–Q) plot of x- and y-direction aggregate diameters, and (C) anisotropy index of co-culture aggregates seeded within the top, middle, and bottom regions of PBFP gradient hydrogels as well as in bulk control gels over time (n = 8; * = p<0.05 (significance between time points omitted in (A) for clarity); ^# = significant difference (p<0.05) from bulk control at same time point). Error bars represent ± standard deviation.

Figure 9. Quantification of vascular sprout length.

(A-D) Average sprout length by angle of co-culture aggregates seeded within bulk control gels (A) as well as the top, middle, and bottom regions of PBFP gradient hydrogels (B-D) over three weeks (Note: gradients run from regions of high elastic modulus and YRGDS concentration at 90° to regions of lower elastic modulus and YRGDS concentration at 270°). (E) Day 21 sprout length as a function of location and gel type at 0°, 90°, 180° and 270° (n = 8; * = p<0.05; # = significant difference (p<0.05) from bulk control at same time point). Error bars represent ± standard deviation.