Tissue regeneration templates based on collagen-glycosaminoglycan copolymers

Evaluation of tubular poly(trimethylene carbonate) tissue engineering scaffolds in a circulating pulsatile flow system

Tubular scaffolds (internal diameter approximately 3 mm and wall thickness approximately 0.8 mm) with a porosity of approximately 83% and an average pore size of 116 µm were prepared from flexible poly(trimethylene carbonate) (PTMC) polymer by dip-coating and particulate leaching methods. PTMC is a flexible and biocompatible polymer that crosslinks upon irradiation; porous network structures were obtained by irradiating the specimens in vacuum at 25 kGy before leaching soluble salt particles. To assess the suitability of these scaffolds in dynamic cell culturing for cardiovascular tissue engineering, the scaffolds were coated with a thin (0.1 to 0.2 mm) non-porous PTMC layer and its performance was evaluated in a closed pulsatile flow system (PFS). For this, the PFS was operated at physiological conditions at liquid flows of 1.56 ml/s with pressures varying from 80-120 mmHg at a frequency of 70 pulsations per minute. The mechanical properties of these coated porous PTMC scaffolds were not significantly different than non-coated scaffolds. Typical tensile strengths in the radial direction were 0.15 MPa, initial stiffness values were close to 1.4 MPa. Their creep resistance in cyclic deformation experiments was excellent. In the pulsatile flow setup, the distention rates of these flexible and elastic scaffolds were approximately 0.10% per mmHg, which is comparable to that of a porcine carotid artery (0.11% per mmHg). The compliance and stiffness index values were close to those of natural arteries.?In long-term deformation studies, where the scaffolds were subjected to physiological pulsatile pressures for one week, the morphology and mechanical properties of the PTMC scaffolds did not change. This suggests their suitability for application in a dynamic cell culture bioreactor.

Binding and release characteristics of insulin-like growth factor-1 from a collagen-glycosaminoglycan scaffold

Tissue engineering is a promising technique for cartilage repair, but to optimize novel scaffolds before clinical trials, it is necessary to determine their characteristics for binding and release of growth factors. Toward this goal, a novel, porous collagen-glycosaminoglycan scaffold was loaded with a range of concentrations of insulin-like growth factor-1 (IGF-1) to evaluate its potential as a controlled delivery device. The kinetics of IGF-1 adsorption and release from the scaffold was demonstrated using radiolabeled IGF-1. Adsorption was rapid, and was approximately proportional to the loading concentration. Ionic bonding contributed to this interaction. IGF-1 release was studied over 14 days to compare the release profiles from different loading groups. Two distinct phases occurred: first, a burst release of up to 44% was noted within the first 24 h; then, a slow, sustained release (13%-16%) was observed from day 1 to 14. When the burst release was subtracted, the relative percentage of remaining IGF-1 released was similar for all loading groups and broadly followed t(½) kinetics until approximately day 6. Scaffold cross-linking using dehydrothermal treatment did not affect IGF-1 adsorption or release. Bioactivity of released IGF-1 was confirmed by seeding scaffolds (preadsorbed with unlabeled IGF-1) with human osteoarthritic chondrocytes and demonstrating increased proteoglycan production in vitro.

Porous hybrid structures based on P(DLLA-co-TMC) and collagen for tissue engineering of small-diameter blood vessels

Poly (D,L-lactide)-7co-(1,3-trimethylene carbonate) [P(DLLA-co-TMC)] (83 mol % DLLA) was used to produce matrices suitable for tissue engineering of small-diameter blood vessels. The copolymer was processed into tubular structures with a porosity of approximately 98% by melt spinning and fiber winding, thus obviating the need of organic solvents that may compromise subsequent cell culture. Unexpectedly, incubation in culture medium at 37 degrees C resulted in disconnection of the contact points between the polymer fibers. To improve the structural stability of these P(DLLA-co-TMC) scaffolds, a collagen microsponge was formed inside the pores of the synthetic matrix by dip coating and freeze drying. Hybrid structures with a porosity of 97% and an average pore size of 102 mum were obtained. Structural stability was preserved during incubation in culture medium at 37 degrees C. Smooth-muscle cells (SMCs) were seeded in these hybrid scaffolds and cultured under pulsatile flow conditions in a bioreactor (120 beats/min, 80-120 mmHg). After 7 days of culture in a dynamic environment viable SMCs were homogeneously distributed throughout the constructs, which were five times stronger and stiffer than noncultured scaffolds. Values for yield stress (2.8 +/- 0.6 MPa), stiffness (1.6 +/- 0.4 MPa), and yield strain (120% +/- 20%) were comparable to those of the human artery mesenterica.

First steps towards tissue engineering of small-diameter blood vessels: preparation of flat scaffolds of collagen and elastin by means of freeze drying

Porous scaffolds composed of collagen or collagen and elastin were prepared by freeze drying at temperatures between -18 and -196 degrees C. All scaffolds had a porosity of 90-98% and a homogeneous distribution of pores. Freeze drying at -18 degrees C afforded collagen and collagen/elastin matrices with average pore sizes of 340 and 130 mum, respectively. After 20 successive cycles up to 10% of strain, collagen/elastin dense films had a total degree of strain recovery of 70% +/- 5%, which was higher than that of collagen films (42% +/- 6%). Crosslinking of collagen/elastin matrices either in water or ethanol/water (40% v/v) was carried out using a carbodiimide (N-(3-dimethylaminopropyl)-N'-ethylcarbodiimide hydrochloride, EDC) in combination with a succinimide (N-hydroxysuccinimide, NHS) in the presence or absence of a diamine (J230) or by reaction with butanediol diglycidylether (BDGE), followed by EDC/NHS. Crosslinking with EDC/NHS or EDC/NHS/J230 resulted in matrices with increased stiffness as compared to noncrosslinked matrices, whereas sequential crosslinking with the diglycidylether and EDC/NHS yielded very brittle scaffolds. Ethanol/water was the preferred solvent in the crosslinking process because of its ability to preserve the open porous structure during crosslinking. Smooth muscle cells were seeded on the (crosslinked) scaffolds and could be expanded during 14 days of culturing.

Physical characterization of vascular grafts cultured in a bioreactor

Tubular scaffolds of collagen and elastin (weight ratio 1:1) with interconnected pores were prepared by freeze drying and crosslinked with N-(3-dimethylaminopropyl)-N'-ethylcarbodiimide hydrochloride (EDC) and N-hydroxysuccinimide (NHS) in the presence or absence of a Jeffamine spacer (poly(propylene glycol)-bis-(2-aminopropyl ether), J230). The crosslinked and uncrosslinked matrices had porosities of 90% and average pore sizes of 131-151 microm. Smooth muscle cells (SMC) were cultured in the crosslinked and uncrosslinked tubular scaffolds under pulsatile flow conditions (mean flow rate 9.6 ml/min, 120 beats/min, pressure 80-120 mmHg). All the constructs could withstand cyclic mechanical strain in the absence of any mechanical support without cracking or suffering permanent deformation. After 7d, SMC were homogeneously distributed throughout the uncrosslinked and EDC/NHS crosslinked constructs, whereas hardly any cell was observed on the luminal side of J230/EDC/NHS crosslinked matrices. Considering the better mechanical performance of EDC/NHS crosslinked matrices compared to non-crosslinked constructs after 7d of culture, SMC were dynamically cultured in the former scaffolds for 14d. During this period, the high strain stiffness of the constructs increased more than two-fold to 38+/-2 kPa, whereas the low strain stiffness doubled to 8+/-2 kPa. The yield stress and yield strain were 30+/-10 kPa and 120+/-20%, respectively. SMC were homogeneously distributed throughout the EDC/NHS crosslinked collagen/elastin constructs and collagen fibres tended to orient in the circumferential direction.

Biological characterisation of vascular grafts cultured in a bioreactor

In this study, the development is described of a tissue-engineered construct mimicking the structure of a natural blood vessel. Smooth muscle cells (SMC) were cultured under pulsatile flow conditions in porous tubular scaffolds composed of crosslinked type I insoluble collagen and insoluble elastin. Under these dynamic culture conditions, average wall shear rate, systolic and diastolic pressures and pressure wave-forms comparable to conditions in the human carotid artery were obtained. Culturing of SMC in tubular scaffolds under dynamic conditions resulted in enhanced tissue formation compared to static conditions. Higher SMC numbers, a more homogeneous distribution of SMC throughout the scaffolds and higher collagen mRNA expression levels were found when cells were cultured under dynamic compared to static conditions. mRNA expression levels of markers of proliferation and apoptosis showed that the higher cell numbers in the scaffolds cultured under dynamic conditions can be explained by increased cell proliferation but not by decreased apoptosis. Glucose consumption and lactate formation by the cells showed that cell metabolism was more aerobic under dynamic compared to static conditions. Lining of the dynamically cultured constructs with a luminal monolayer of endothelial cells might result in vessels suitable for in vivo applications.

Evidence for sequential utilization of fibronectin, vitronectin, and collagen during fibroblast-mediated collagen contraction

Contraction plays a major role in wound healing and is inevitably mediated through the mechanical interaction of fibroblast cytoskeleton and integrins with their extracellular matrix ligands. Cell-matrix attachment is critical for such events. In human dermal fibroblasts most such interactions are mediated by the beta1-type integrins. This study investigated the role played by key components in this system, notably fibronectin, vitronectin, and integrin subcomponents alpha2 and alpha5, which recognize collagen and fibronectin. Inhibition of adhesion through these ligands was studied either by antibody blocking or with fibronectin and/or vitronectin depletion. Functional effects of inhibition were monitored as force generation in collagen-glycosaminoglycan (IntegraTM) sponges, over 20 hours using a culture force monitor. Dose and time-course inhibition studies indicated that initial attachment and force generation (approx. 0-5 hours postseeding) was through fibronectin receptors and this was followed by vitronectin ligand and receptor utilization (4 hours onward). Utilization of the collagen integrin subcomponent alpha2 appeared to be increasingly important between 6 and 16 hours and dominant thereafter. Additionally, there was evidence for functional interdependence between the three ligand systems fibronectin, vitronectin, and collagen. We propose that there is a short cascade of sequential integrin-ligand interactions as cells attach to, extend through, and eventually contract their matrix. (WOUND REP REG 2002;10:-408)

Contraction-mediated pinocytosis of RGD-peptide by dermal fibroblasts: inhibition of matrix attachment blocks contraction and disrupts microfilament organisation

Force generation in collagen and matrix contraction are basic functions of fibroblasts and important elements of tissue repair. Cell-matrix attachment is critical to this contraction, involving RGD-binding integrins. We have investigated how this process operates, in terms of force generation (in the Culture Force Monitor) and cytoskeletal structure, using a synthetic RGD-decapeptide. The RGD-peptide blocked force generation over the first 6 h, followed by near complete recovery by 20 h. However, dose response was complex indicating multiple processes were operating. Analysis of cytoskeletal structure after treatment with RGD-peptide indicated major disruption with condensed aggregates of actin and microtubular fragmentation. Fluorescent labeling and tracking of the RGD-peptide demonstrated intracellular uptake into discrete cytoplasmic aggregates. Critically, these RGD-peptide pools co-localised with the condensed actin microfilament aggregates. It is concluded that RGD-peptide uptake was by a form of contraction-mediated pinocytosis, resulting from mechanical tension applied to the untethered RGD-peptide-integrin, as contractile microfilament were assembled. These findings emphasize the importance of sound mechanical attachment of ligand-occupied integrins (e.g., to extracellular matrix) for normal cytoskeletal function. Conversely, this aspect of unrestrained cytoskeletal contraction may have important pathogenic and therapeutic applications.

Peripheral nerve injury: a review and approach to tissue engineered constructs

Eleven thousand Americans each year are affected by paralysis, a devastating injury that possesses associated annual costs of $7 billion (American Paralysis Association, 1997). Currently, there is no effective treatment for damage to the central nervous system (CNS), and acute spinal cord injury has been extraordinarily resistant to treatment. Compared to spinal cord injury, damage to peripheral nerves is considerably more common. In 1995, there were in excess of 50,000 peripheral nerve repair procedures performed. (National Center for Health Statistics based on Classification of Diseases, 9th Revision, Clinical Modification for the following categories: ICD-9 CM Code: 04.3, 04.5, 04.6, 04.7). These data, however, probably underestimate the number of nerve injuries appreciated, as not all surgical or traumatic lesions can be repaired. Further, intraabodominal procedures may add to the number of neurologic injuries by damage to the autonomic system through tumor resection. For example, studies assessing the outcome of impotency following radical prostatectomy demonstrated 212 of 503 previously potent men (42%) suffered impotency when partial or complete resection of one or both cavernosal nerve(s). This impotency rate decreased to 24% when the nerves were left intact (Quinlan et al., J. Urol. 1991;145:380-383; J. Urol. 1991;145:998-1002).

Primary neural precursor cell expansion, differentiation and cytosolic Ca(2+) response in three-dimensional collagen gel

To investigate the ability to culture neural precursor cells in a three-dimensional (3D) collagen gel, neuroepithelial cells were isolated from embryonic day 13 rat cortex, dispersed within type I collagen and maintained for up to 30 days in vitro. Cultured in Neuorobasal medium supplemented with B27 containing basic fibroblast growth factor, the collagen-entrapped precursor cells actively expanded and formed clone-like clusters. Many cells in the center of the cluster were proliferating as revealed by 5-bromo-2'-deoxyuridine uptake. Some cells began to migrate away from the center at 5 days and were labeled by either neuronal marker neuron-specific beta-tubulin (TuJ1) or astrocytic marker glial fibrillary acidic protein. The differentiated neurons (TuJ1(+)) exhibited characteristic cytosolic Ca(2+) oscillations in response to excitatory neurotransmitter glutamate. These findings suggest the suitability of the 3D culture system for the proliferation and differentiation of neural precursor cells.

Challenges to nerve regeneration

Peripheral nerve injuries can result from mechanical, thermal, chemical, congenital, or pathological etiologies. Failure to restore these damaged nerves can lead to the loss of muscle function, impaired sensation, and painful neuropathies. Current surgical strategies for the repair of critical nerves involve the transfer of normal donor nerve from an uninjured body location. However, these "gold standard" methods for tissue restoration frequently are limited by tissue availability, risk of disease spread, secondary deformities, and potential differences in tissue structure and size. One possible alternative to autogenous tissue replacement is the development of engineered constructs to replace those elements necessary for axonal proliferation, including a scaffold, support cells, induction factors, and extracellular matrices. Despite advances and contributions in the field of tissue engineering, results to date with nerve conduits have failed to equal the nerve regeneration achieved with autogenous grafts for large distances. We review the current challenges to tissue-engineered constructs. Each of the four components is reviewed and approaches are outlined. Semin. Surg. Oncol. 19:312-318, 2000.

Tensional homeostasis in dermal fibroblasts: mechanical responses to mechanical loading in three-dimensional substrates

Many soft connective tissues are under endogenous tension, and their resident cells generate considerable contractile forces on the extracellular matrix. The present work was aimed to determine quantitatively how fibroblasts, grown within three-dimensional collagen lattices, respond mechanically to precisely defined tensional loads. Forces generated in response to changes in applied load were measured using a tensional culture force monitor. In a number of variant systems, resident cells consistently reacted to modify the endogenous matrix tension in the opposite direction to externally applied loads. That is, increased external loading was followed immediately by a reduction in cell-mediated contraction whilst decreased external loading elicited increased contraction. Responses were cell-mediated and not a result of material properties of the matrices. This is the first detailed characterisation of a tensional homeostatic response in cells. The maintained force, after 8 h in culture, was typically around 40-60 dynes/million cells). Maintenance of an active tensional homeostasis has widespread implications for cells in culture and for whole tissue function.