Stabilized Collagen and Elastin-Based Scaffolds for Mitral Valve Tissue Engineering

Abstract

There is a significant clinical need for new approaches to treatment of mitral valve disease. The aim of this study was to develop a tissue-engineered mitral valve scaffold possessing appropriate composition and structure to ensure ideal characteristics of mitral valves, such as large orifice, rapid opening and closure, maintenance of mitral annulus-papillary muscle continuity, in vivo biocompatibility and extended durability. An extracellular matrix-based scaffold was generated, based on the native porcine mitral valve as starting material and a technique for porcine cell removal without causing damage to the matrix components. To stabilize these structures and slow down their degradation, acellular scaffolds were treated with penta-galloyl glucose (PGG), a well-characterized polyphenol with high affinity for collagen and elastin. Biaxial mechanical testing presented similar characteristics for the PGG-treated scaffolds compared to fresh tissues. The extracellular matrix components, crucial for maintaining the valve shape and function, were well preserved in leaflets, and in chordae, as shown by their resistance to collagenase and elastin. When extracted with strong detergents, the PGG-treated scaffolds released a reduced amount of soluble matrix peptides, compared to untreated scaffolds; this correlated with diminished activation of fibroblasts seeded on scaffolds treated with PGG. Cell-seeded scaffolds conditioned for 5 weeks in a valve bioreactor showed good cell viability. Finally, rat subdermal implantation studies showed that PGG-treated mitral valve scaffolds were biocompatible, nonimmunogenic, noninflammatory, and noncalcifying. In conclusion, a biocompatible mitral valve scaffold was developed, which preserved the biochemical composition and structural integrity of the valve, essential for its highly dynamic mechanical demands, and its biologic durability.

Keywords: bioreactor; matrikines; mitral valve prolapse.

FIG. 1.

Mitral valve decellularization. (A) Macroscopic aspect of a decellularized mitral valve leaflet. (B) Ethidium bromide agarose gel electrophoresis analysis of DNA extracted from fresh tissue (F) and decellularized valve tissue (D1-3); S, DNA ladder standard. (C) Representative histology analysis of fresh and decellularized (Decell) mitral valve stained with H&E and DAPI. Bars in (C) are 100 μm. H&E, hematoxylin and eosin. Color images available online at www.liebertpub.com/tea

FIG. 2.

Histology of mitral valve scaffolds. Representative histology analysis of fresh and decellularized (Decell) mitral valve leaflets and chordae stained with Movats' Pentachrome (A, F), VVG in (B and E), and immunohistochemistry for type IV collagen and laminin (C, D). Bars are 100 μm. VVG, Voerhoff van Gieson. Color images available online at www.liebertpub.com/tea

FIG. 3.

Matrix stabilization in acellular mitral valves. (A, B) Biaxial stress-strain analysis of samples of fresh, acellular (Decelled) and PGG-treated (PGG) acellular valve leaflets tested in radial (X) and circumferential (Y) directions. Avg., average values for n = 5. (C) Elastic moduli of fresh, Decelled and PGG leaflets. (D) Differential scanning calorimetry evaluation of fresh, Decelled and PGG-treated acellular valve leaflets. T_d, thermal denaturation temperature. *p < 0.05 compared to fresh. PGG, penta-galloyl glucose. Color images available online at www.liebertpub.com/tea

FIG. 4.

Resistance to proteases. (A) Resistance to collagenase and (B) Resistance to elastase degradation of fresh, acellular (Decell) and PGG-treated (PGG) acellular mitral valve leaflets. Values are expressed as % dry weight loss after exposure to enzyme for 24 or 48 h (hr). *p < 0.05 compared to fresh. Color images available online at www.liebertpub.com/tea

FIG. 5.

Cytocompatibility of acellular mitral scaffolds. Representative images of PGG-treated scaffolds seeded with cells and analyzed 3 days after seeding. (A) DAPI nuclear staining, (B) Live/Dead staining, (C) H&E stain, (D) vimentin immunohistochemistry and (E) Movat's Pentachrome. Bars are 100 μm. Color images available online at www.liebertpub.com/tea

FIG. 6.

Bioreactor study. (A) Assembled mitral valve bioreactor composed of 1, ventricular chamber, 2, atrial chamber, 3, ventricular reservoir, 4, atrial reservoir, 5, pressure transducer, 6, flow meter, 7, sterile air filter. (B) Ventricular aspect of a representative cell-seeded, PGG-treated mitral valve scaffold during testing in culture media (red). (C) Representative image of Live/Dead (live, green, dead, red) stained cells after 3 weeks in the mitral valve bioreactor. (D) Representative immunohistochemistry images of PGG-treated mitral valve scaffolds seeded with cells and analyzed 3 weeks after bioreactor testing. SMA, α-smooth muscle cell actin, VIM, vimentin, INT, integrin β1. Bars are 100 μm. Color images available online at www.liebertpub.com/tea

FIG. 7.

Effects of PGG stabilization on fibroblast activation. Fibroblasts were seeded onto PGG-treated mitral valve scaffolds (PGG) and their activation compared to cells seeded onto non-PGG-treated scaffolds. (A) TGF secretion assessed by ELISA. (B) MMP secretion evaluated by zymography (insert shows actual gel). (C) Representative immunohistochemistry images of PGG-treated mitral valve scaffolds seeded with cells and stained for SMA. Bars are 100 μm. (D) Soluble peptide release assessed by gradient SDS-PAGE followed by silver staining. Std, molecular weight standards. MMP, matrix metalloproteinase. Color images available online at www.liebertpub.com/tea

FIG. 8.

Effects of PGG on host infiltration into acellular mitral valve scaffolds at 4 weeks postimplantation. (A) Representative histology images of leaflet (L) and chordae (C) scaffolds stained with Trichrome (collagen, blue; cells, red) and anti-CD8 antibody (positive, brown). (B) Quantitative analysis of total infiltrating cells (top) and CD8+ infiltrating cells (bottom), both normalized to scaffold cross-sectional area. Scale bars are 100 μm in Trichrome images and 20 μm in CD8 images. *p < 0.05 by ANOVA. ANOVA, analysis of variance. Color images available online at www.liebertpub.com/tea