Composition of intraperitoneally implanted electrospun conduits modulates cellular elastic matrix generation

Abstract

Improving elastic matrix generation is critical to developing functional tissue engineered vascular grafts. Therefore, this study pursued a strategy to grow autologous tissue in vivo by recruiting potentially more elastogenic cells to conduits implanted within the peritoneal cavity. The goal was to determine the impacts of electrospun conduit composition and hyaluronan oligomer (HA-o) modification on the recruitment of peritoneal cells, and their phenotype and ability to synthesize elastic matrix. These responses were assessed as a function of conduit intra-peritoneal implantation time. This study showed that the blending of collagen with poly(ε-caprolactone) (PCL) promotes a faster wound healing response, as assessed by trends in expression of macrophage and smooth muscle cell (SMC) contractile markers and in matrix deposition, compared to the more chronic response for PCL alone. This result, along with the increase in elastic matrix production, demonstrates the benefits of incorporating as little as 25% w/w collagen into the conduit. In addition, PCR analysis demonstrated the challenges in differentiating between a myofibroblast and an SMC using traditional phenotypic markers. Finally, the impact of the tethered HA-o is limited within the inflammatory environment, unlike the significant response found previously in vitro. In conclusion, these results demonstrate the importance of both careful control of implanted scaffold composition and the development of appropriate delivery methods for HA-o.

Keywords: Collagen; Elastin; Electrospinning; Peritoneal cavity; Vascular grafts.

Fig. 1

Electrospun conduits were produced from a synthetic polymer and a natural/synthetic blend. Shown are SEM images of the outside surface of conduits electrospun from PCL (A) and a 25% w/w collagen/PCL blend (B). An EDS spectrum shows the presence of nitrogen in blend conduits after electrospinning (C). The fibrous surfaces of meshes with immobilized HA-o (F–G) exhibited significantly more intense HABP stainin g than those without (D–E). This occurred with both PCL (D–F) and blend (EG) conduits. Contrast and brightness were adjusted evenly for images with / without HA-o modification with each scaffold material. However, the PCL and blend meshes were adjusted separately because of their different thicknesses, which resulted in differing levels of background.

Fig. 2

Quantitative gene expression profiles for constructs composed of PCL (A, C) and a 25% w/w collagen/PCL blend (B, D). Data shows differences in expression of macrophage (A, B) and contractile SMC (C, D) phenotypic markers. This data shows changes from 4 to 6 weeks of intra-peritoneal implantation of conduits with and without tethered HA-o (indicated by + or −, respectively). The error bars represent the standard error.

Fig. 3

Temporal changes in phenotype of peritoneal cells recruited to conduits of different compositions. Shown are immunohistochemical images of myofibroblast, SMC, and macrophage markers – i.e. αSMA (A) and CD68 (B) – on HA-o modified surfaces. Cross-sections of PCL and blend conduits are shown at both 4 and 6 weeks implantation time. The outer surface of the fibrous capsule and implanted conduit is on the left edge in these images. Overlay channels also include DAPI stained nuclei (blue) and autofluorescent background (green).

Fig. 4

Quantitative expression profiles of matrix genes by cells within PCL (A) and 25% w/w collagen/PCL blend (B) constructs. This data shows temporal changes in gene expression between 4 and 6 weeks of intra-peritoneal implantation of constructs with/without tethered HA-o (indicated by + or −, respectively). The error bars represent the standard error. Individual conditions were not statistically different. However, there were statistical differences between experimental parameters, as described in the text.

Fig. 5

Matrix synthesis within implanted conduits as a function of conduit composition, HA-o modification, and implantation time. Shown is normalized total tissue mass in the constructs. * indicates statistical significance from non-implanted constructs and # indicates significance from HA-o-modified PCL conduits.

Fig. 6

Hematoxylin and Eosin (H & E) staining of tissue generated within implanted conduits. Shown are representative images that demonstrate significant differences in cell and ECM synthesis responses to PCL (A–E) and blend (F–J) conduits that are HA-o modified. The images show constructs after 2 (A, F), 4 (B–C, G–H), and 6 week (D–E, I–J) implantation times. Cell nuclei stain blue and cytoplasm of cells and ECM stain pink. The outer surface of the conduits is on the right side of the images. Representative cell nuclei are circled in the high magnification images (C, H). Histology images of tissues generated within conduits without HA-o modification are not shown since there were no noticeable differences in tissue generation versus for the corresponding HA-o modified conduits.

Fig. 7

Matrix generation within intraperitoneal tissue constructs is influenced by conduit composition, the presence of HA-o, and duration of implantation. Shown within the constructs is collagen content via the hydroxy-Pro assay (A) and matrix elastin, via the Fastin^® assay (B). Statistical significance is indicated by * from the HA-o modified blend conduits, # from PCL conduits for all time points, and @ from 2 weeks of conduit implantation. Immunohistochemical images of elastin on HA-o modified surfaces (C). Cross-sections of PCL and blend conduits are shown at both 4 and 6 weeks implantation time. Inter-section variations in elastin content (red) were observed, and the included images represent areas that stained most intensely for elastin. The outer surface of the fibrous capsule and implanted conduit is on the left edge in these images. Overlay channels also include DAPI stained nuclei (blue) and autofluorescent background (green).