Lionel Ambroise Micol
Several urinary tract disorders, such as hypospadias or bladder cancer, require reconstructive surgery to preserve the patient's quality of life and at times even save their life. The current available surgical treatments, involving grafts from other tissue origin, can, however, lead to long-term complications in up to 50% of these patients. New alternatives are therefore considered among which tissue-engineered grafts are thought to be the most promising approach. Tissue-engineered grafts have so far been shown to require the presence of cells, which involves different steps prior to implantation including cell isolation and expansion, seeding on a matrix, and further culture. These steps are time-consuming, costly, and raise safety concerns regarding the in vitro cultured cells. In this thesis, we sought to further improve some tissue-engineered grafts with approaches that would either reduce or replace the cell culture steps. One of these approaches relied on the plastic compression of collagen gels. Ready cell-seeded plastic compressed collagen gels can be rapidly formed within minutes and have much better mechanical properties than regular collagen gels that are known to be fragile. In order to have a stronger gel we adapted the plastic compression concept and developed high-density collagen gels tubes (hdCGTs) for urinary tract tissue-engineering. We demonstrated that they were suitable for smooth muscle cell (SMC) proliferation and that their mechanical response to dilation and extension were compatible with ureter and urethra physiologic values. Hence, these ready-seeded hdCGTS represent promising tissue-engineered constructs that can shorten the production time of cell-seeded grafts for urinary tract regeneration. We further tested our hdCGTs in a rabbit model to investigate their regenerative capacity in vivo. Autologous rabbit SMCs were isolated from a bladder biopsy to form cellular constructs, while acellular constructs were used as control. Both types of constructs were sealed with fibrin glue at the site of a newly created urethral defect. Rabbits were evaluated by retrograde urethrography after 1 or 3 months. Despite a high rate of fistula and a relative narrowing of graft calibers, all rabbits survived and showed micturition within 1-3 days following surgery. SMC-seeded grafts showed, however, a better outcome at 3 months in accordance to literature. Since plastic compressed collagen gels are still too weak to be sutured, we decided to strengthen them by combining them with a polymer mesh. We investigated flat collagen-poly(lactic acid-co-ε-caprolactone) (PLAC) hybrid scaffolds seeded with human bladder cells for their behaviour in the subcutaneous space of nude mice. Host cell infiltration, scaffold degradation, and the presence of the seeded bladder cells were analyzed at different time points on a total period of 6 months. The hybrids showed a lower inflammatory reaction in vivo than PLAC meshes alone, and first signs of polymer degradation were visible at six months. The collagen-PLAC hybrids have potential for urinary tract tissue regeneration as they show efficient cell seeding and proliferation as well as proper mechanical properties for surgical handling. Another approach towards improving tissue-engineered grafts was to use off-the-shelf smart biomatrices to replace the cell culture steps. This involved the development of cell-controlled release of insulin-like growth factor-1 (IGF1) in an acellular collagen matrix. IGF1 induces SMC migration, proliferation, and differentiation, and can hence promote the invasion and population of the acellular matrix by cells in the surrounding tissue once implanted in vivo. In this approach, we designed and produced a recombinant collagen-binding IGF1. The IGF1 domain would then be released after cleavage from an affinity binding domain to collagen under the action of cell-secreted matrix metallo-proteases (MMP). Although our recombinant IGF1 showed the expected bioactivity on SMCs and sensitivity to MMP-2, its affinity for type I collagen (Col-I) could not be demonstrated. We further investigated covalent binding alternatives involving transglutaminase (TG) crosslinking or IGF binding protein (IGFBP)-mediated binding to collagen. TG crosslinking did not allow for sufficient amount of a TG-IGF, a recombinant form of IGF1 presenting a substrate sequence for plasma TG, to be bound to Col-I. The entrapment of IGFBP5 within plastic compressed collagen gels however suggests that collagen compression and collagen-binding means could be combined in order to allow for a more controlled-release profile of IGF1 from collagen gels. Lastly, we sought to develop a new bioengineered solution for bladder sphincter disorders, such as urinary incontinence or vesico-ureteral reflux. These disorders are treated with open surgeries susceptible of complications and extended hospital stay. Endoscopic treatment by injection of bulking agents is thus usually preferred, although they often necessitate additional interventions for a sustained effect. With the idea to suppress this need for additional intervention and thus to limit discomfort for patients as well as costs, we sought for a permanent bulking agent. We therefore developed an injectable poly(acrylonitrile) hydrogel paste injected into the submucosal space of pig bladders. The implants were harvested at days 7, 14, 21, 28, 84, and 168 and analyzed morphologically and by histology. The persistence of the implants was demonstrated, indicating that this injectable bulking agent could potentially be used as biomaterial for the treatment of urinary incontinence and vesico-ureteral reflux.
EPFL2011
