Towards embryonic-like scaffolds for skin tissue engineering: identification of effector molecules and construction of scaffolds.

Autologous skin grafts are the gold standard for the treatment of burn wounds. In a number of cases, treatment with autologous tissue is not possible and skin substitutes are used. The outcome, however, is not optimal and improvements are needed. Inspired by scarless healing in early embryonic development, we here set out a strategy for the design and construction of embryonic-like scaffolds for skin tissue engineering. This strategy may serve as a general approach in the construction of embryonic-like scaffolds for other tissues/organ. As a first step, key effector molecules upregulated during embryonic and neonatal skin formation were identified using a comparative gene expressing analysis. A set of 20 effector molecules was identified, from which insulin-like growth factor 2 (IGF2) and sonic hedgehog (SHH) were selected for incorporation into a type I collagen-heparin scaffold. Porous scaffolds were constructed using purified collagen fibrils and 6% covalently bound heparin (to bind and protect the growth factors), and IGF2 and SHH were incorporated either individually (~0.7 and 0.4 µg/mg scaffolds) or in combination (combined ~1.5 µg/mg scaffolds). In addition, scaffolds containing hyaluronan (up to 20 µg/mg scaffold) were prepared, based on the up- or downregulation of genes involved in hyaluronan synthesis/degradation and its suggested role in scarless healing. In conclusion, based on a comprehensive gene expression analysis, a set of effector molecules and matrix molecules was identified and incorporated into porous scaffolds. The scaffolds thus prepared may create an 'embryonic-like' environment for cells to recapitulate embryonic events and for new tissues/organs.

Design and in vivo evaluation of a molecularly defined acellular skin construct: reduction of early contraction and increase in early blood vessel formation.

Skin substitutes are of great benefit in the treatment of patients with full thickness wounds, but there is a need for improvement with respect to wound closure with minimal contraction, early vascularisation, and elastin formation. In this study we designed and developed an acellular double-layered skin construct, using matrix molecules and growth factors to target specific biological processes. The epidermal layer was prepared using type I collagen, heparin and fibroblast growth factor 7 (FGF7), while the porous dermal layer was prepared using type I collagen, solubilised elastin, dermatan sulfate, heparin, fibroblast growth factor 2 (FGF2) and vascular endothelial growth factor (VEGF). The construct was biochemically and morphologically characterised and evaluated in vivo using a rat full thickness wound model. The results were compared with the commercial skin substitute IntegraDRT and untreated wounds. The double-layered construct was prepared according to the design specifications. The epidermal layer was about 40 µm thick, containing 9% heparin and 0.2 µg FGF7 mg per layer, localised at the periphery. The dermal layer was 2.5 mm thick, had rounded pores and contained 10% dermatan sulfate+heparin, and 0.7 µg FGF2+VEGF mg per layer. The double-layered skin construct was implanted in a skin defect and on day 7, 14, 28 and 112 the (remaining) wound area was photographed, excised and (immuno) histologically evaluated. The double-layered skin construct showed more cell influx, significantly less contraction and increased blood vessel formation at early time points in comparison with IntegraDRT and/or the untreated wound. On day 14 the double-layered skin construct also had the fewest myofibroblasts present. On day 112 the double-layered skin construct contained more elastic fibres than IntegraDRT and the untreated wound. Structures resembling hair follicles and sebaceous glands were found in the double-layered skin construct and the untreated wound, but hardly any were found in IntegraDRT. The results provide new opportunities for the application of acellular skin constructs in the treatment of surgical wounds.

Controlled fabrication of triple layered and molecularly defined collagen/elastin vascular grafts resembling the native blood vessel.

There is a consistent need for a suitable natural biomaterial to function as an arterial prosthesis in achieving arterial regeneration. Natural grafts are generally obtained by decellularization of native blood vessels, but batch to batch variations may occur and the nature/content of remaining contaminants is generally unknown. In this study we fabricated a molecularly defined natural arterial graft from scratch resembling the native three layered architecture from the fibrillar extracellular matrix components collagen and elastin. Using casting, moulding, freezing and lyophilization techniques, a triple layered construct was prepared consisting of an inner layer of elastin fibres, a middle (porous) film layer of collagen fibrils and an outer scaffold layer of collagen fibrils. The construct was carbodiimide cross-linked and heparinized. Characterization included biochemical/biophysical analyses, scanning electron microscopy, micro-computed tomography, (immuno)histology and haemocompatibility. Burst pressures were up to 400mm Hg and largely conferred by the intermediate porous collagen film layer. The highly purified type I collagen fibrils and elastin fibres used did not evoke platelet aggregation in vitro. Suturability of the graft in end to side anastomosis was successful and considered adequate for in vivo application.

Reduction of anionic sites in the glomerular basement membrane by heparanase does not lead to proteinuria.

Heparan sulfate in the glomerular basement membrane has been considered crucial for charge-selective filtration. In many proteinuric diseases, increased glomerular expression of heparanase is associated with decreased heparan sulfate. Here, we used mice overexpressing heparanase and evaluated the expression of different heparan sulfate domains in the kidney and other tissues measured with anti-heparan sulfate antibodies. Glycosaminoglycan-associated anionic sites were visualized by the cationic dye cupromeronic blue. Transgenic mice showed a differential loss of heparan sulfate domains in several tissues. An unmodified and a sulfated heparan sulfate domain resisted heparanase action in vivo and in vitro. Glycosaminoglycan-associated anionic sites were reduced about fivefold in the glomerular basement membrane of transgenic mice, whereas glomerular ultrastructure and renal function remained normal. Heparanase-resistant heparan sulfate domains may represent remnant chains or chains not susceptible to cleavage. Importantly, the strong reduction of glycosaminoglycan-associated anionic sites in the glomerular basement membrane without development of a clear renal phenotype questions the primary role of heparan sulfate in charge-selective filtration. We cannot, however, exclude that overexpression of heparanase and heparan sulfate loss in the basement membrane in glomerular diseases contributes to proteinuria.