Degeneration of Aortic Valves in a Bioreactor System with Pulsatile Flow

Abstract

Calcific aortic valve disease is the most common valvular heart disease in industrialized countries. Pulsatile pressure, sheer and bending stress promote initiation and progression of aortic valve degeneration. The aim of this work is to establish an ex vivo model to study the therein involved processes. Ovine aortic roots bearing aortic valve leaflets were cultivated in an elaborated bioreactor system with pulsatile flow, physiological temperature, and controlled pressure and pH values. Standard and pro-degenerative treatment were studied regarding the impact on morphology, calcification, and gene expression. In particular, differentiation, matrix remodeling, and degeneration were also compared to a static cultivation model. Bioreactor cultivation led to shrinking and thickening of the valve leaflets compared to native leaflets while gross morphology and the presence of valvular interstitial cells were preserved. Degenerative conditions induced considerable leaflet calcification. In comparison to static cultivation, collagen gene expression was stable under bioreactor cultivation, whereas expression of hypoxia-related markers was increased. Osteopontin gene expression was differentially altered compared to protein expression, indicating an enhanced protein turnover. The present ex vivo model is an adequate and effective system to analyze aortic valve degeneration under controlled physiological conditions without the need of additional growth factors.

Keywords: ECM remodeling; bioreactor system; calcific aortic valve disease; degeneration; tissue cultivation.

Figure 1

Computer-controlled bioreactor system with pulsatile flow. (A,B) The bioreactor system consists of computer-controlled regulatory elements and a tube system. (C) Schematic overview of the bioreactor system. Gas exchange occurs by surface aeration in the oxygenation camber (1). pH level (2), oxygen saturation (3), and pressure are measured and maintained via quantity and composition of gas inflow. Temperature (4) is measured and regulated by heating sleeve (5). Perfusion medium is piped through the heart valve by a pulsatile roller pump (6).

Figure 2

Macroscopic and histological analysis of ovine aortic valve conduits. Valve bearing native aortic roots were cultivated in a bioreactor system with pulsatile flow for one week. (A) Macroscopic analysis after bioreactor cultivation shows no remarkable differences between aortic roots cultivated in standard (BR-std) or pro-degenerative medium containing β-glycerophosphate and calcium chloride (BR-deg), but a thickening of leaflets compared to native valve bearing aortic roots (Native). Bars: 1 cm. (B) Overall leaflet structure (H&E staining, upper panel) and three-layer structure of extracellular matrix (Movat pentachrome staining, lower panel) remained after bioreactor cultivation. Representative images of eight replicates are presented. Bars: 400 µm.

Figure 3

Ex vivo cultivation in pro-degenerative conditions led to biomineralization. Cross-sections of native aortic valve leaflets (Native) and leaflets cultivated under standard (BR-std) or pro-degenerative conditions (BR-deg) were stained with Alizarin red (upper panel) and von Kossa (lower panel). Alizarin red staining (red) and von Kossa staining (black) indicating biomineralization, were detected after cultivation in BR-deg, but not in BR-std or native control. Representative pictures of eight replicates were chosen. Bars: 100 µm.

Figure 4

Cellular composition of the aortic valve after bioreactor cultivation. Immune staining with vimentin (green) and vWF (red) specific antibodies was performed on aortic valve cross-sections. Vimentin positive cells were detected in native tissue (first panel) as well as after one week of bioreactor cultivation in standard (BR-std, second panel) pro-degenerative medium (BR-deg, third panel). vWF staining is visible in native aortic valve leaflets (Native, first panel), but not after bioreactor cultivation (second and third panel). Nuclei were counterstained with DAPI. Blood vessels in myocardial specimens served as positive staining control (last panel). Negative control was performed by staining without primary antibodies (right column). Representative pictures of five replicates were chosen. Bars: 50 µm.

Figure 5

Characterization of the gene expression in aortic valve after bioreactor cultivation compared to static conditions. Valve bearing native aortic roots were cultivated in bioreactor with pulsatile flow (BR) or in a cell culture incubator (CC) for one week using standard medium (BR-std and CC-std) or medium with pro-degenerative conditions (BR-deg and CC-deg). Ovine aortic valve samples were examined using qRT-PCR with specific primers against the housekeeper RLP29, the pro-degenerative markers osteopontin (OPN), collagen type I (COL1A1), alkaline phosphatase (ALPL), and TGFβ as well as HIF1α and VEGF. Values are shown as means and standard error of means, dots indicate eight biological replicates. For statistical analysis Mann–Whitney test was performed to compare standard and pro-degenerative conditions and Kruskal–Wallis test was used to compare native, BR and CC cultivation. Stars indicate significant differences in Dunn’s multiple comparison post-hoc test (* p < 0.05; ** p < 0.01; *** p < 0.001).

Figure 6

Bioreactor cultivation led to altered OPN protein levels. Lysates were analyzed by Western blot with specific antibodies for (A,B) αSMA, (C,D) osteopontin (OPN), and Tubulin as loading control. Positions of molecular weight markers are indicated on the left. (B,D) Graphic representation of band intensities is displayed on the right, using means and standard error of means (n = 5). For OPN protein bands at approximately 55 kDa and 32 kDa were subjected to intensity analysis. Kruskal–Wallis test was performed and significance of Dunn’s post-test is indicated with stars (* p < 0.05). Control: native leaflets, BR: bioreactor cultivation, CC: static cultivation in cell culture, std: using standard medium, deg: using pro-degenerative medium.