Human osteoblast response in vitro to platelet-derived growth factor and transforming growth factor-beta delivered from controlled-release polymer rods.

The purpose of this work was (1) to develop extrudable ethylene-vinyl acetate (EVA) copolymer delivery systems capable of sustained release of bioactive proteins and (2) to determine the effect of platelet-derived growth factor (PDGF-BB) and/or transforming growth factor-beta2 (TGF-beta2) on human osteoblast proliferation and differentiation. Human osteoblasts were plated in vitro and proliferation and protein synthesis assayed at 48 and 96 h. EVA-PDGF rods releasing about 34 ng per ml PDGF per day produced a dramatic early increase in osteoblast proliferation and no effect on protein synthesis. EVA-TGF-beta2 rods releasing about 23 ng per ml per day increased protein synthesis but had no effect on proliferation. PDGF and TGF-beta2 together resulted in moderate increases in proliferation and a marked increase in protein synthesis. Morphologically, PDGF-treated cells became confluent as early as 48 h, while TGF-beta2-treated cells formed into nodules. This work shows that (1) it is possible to deliver physiological levels of bioactive proteins from an extrudable EVA delivery system, and (2) bone cell response is dependent on the sequence and timing of delivery. Controlled-release delivery systems which mimic injury-induced healing cascades may be useful in evaluating the role of various molecules in osseous repair.

Influence of surface texture of polymeric sheets on peripheral nerve regeneration in a two-compartment guidance system.

Synthetic guidance channels are useful tools to study the mechanisms underlying peripheral nerve regeneration. In the present study, the lumen of silicone elastomer tubes was divided into two compartments by a polymer strip 10 mm long placed along the tube length. The influence of varying the surface texture of hydrophilic and hydrophobic polymer strips on the morphology of the regenerated neural tissue was analysed. Hydrophilic nitrocellulose (NC) and hydrophobic polyvinylidene fluoride (PVDF) films with smooth (S-NC and S-PVDF) or rough (R-NC, R-PVDF) surface texture were used. Five channels of each type were used to repair transected rat sciatic nerves and analysed after 4 wk. Tissue strips bridged the nerve stumps in all R-NC and R-PVDF tubes, in five of the S-NC and three of the S-PVDF tubes. In R-NC and R-PVDF tubes, bell-shaped tissue adhering to the polymer strip was observed, whereas in S-NC and S-PVDF tubes round, free-floating nerve cables were seen. All the cables contained myelinated and unmyelinated axons and Schwann cells grouped in microfascicles and surrounded by an epineurial layer. For both rough strips, the initial cell layer consisted of macrophages adhering to the polymer surface. The epineurial nerve tissue contacting the rough surface was significantly thinner for PVDF compared with NC strips. No difference in epineurial thickness was observed for nerves facing the silicone tube or for smooth NC and PVDF strips. S-PVDF tubes contained significantly more myelinated axons than S-NC tubes.(ABSTRACT TRUNCATED AT 250 WORDS)

Improved nerve regeneration through piezoelectric vinylidenefluoride-trifluoroethylene copolymer guidance channels.

Piezoelectric materials generating electrical charges in response to mechanical strain may be used to stimulate axonal regeneration following nerve injury. Tubular nerve guidance channels were extruded from a vinylidenefluoride-trifluoroethylene copolymer using a melt-extrusion process. Unlike vinylidenefluoride homopolymer, the copolymer does not need mechanical stretching to achieve a dipole-containing crystal structure, enabling the fabrication of complex piezoelectric devices. Selected tubes were rendered piezoelectric in a high voltage corona poling apparatus. Crystal structure changes induced by poling were evaluated with differential scanning calorimetry. In contrast to unpoled samples, poled ones displayed a sharp endothermic peak and a greater heat of transition at the Curie temperature, indicative of an increase in crystal order and size. The piezoelectric output of poled tubes was characterized using a laser-monitored deflection system interfaced with a charge amplifier and oscilloscope. Poled tubes generated significant voltages in response to slight mechanical deformations. The magnitude of electrical output was independent of the poling polarity. Unpoled tubes showed no electrical output. Positive, negative and unpoled vinylidenefluoride-trifluoroethylene copolymer tubes were used to repair a 10 mm gap in transected sciatic nerves of adult rats. Nerves regenerated in positively poled channels had a significantly greater number of myelinated axons than those regenerated in unpoled channels 4 wk post-implantation. Negatively poled channels contained an intermediate number of myelinated axons. We concluded that piezoelectrically active vinylidenefluoride-trifluoroethylene copolymer tubes significantly enhance nerve regeneration as compared to chemically identical, unpoled tubes and that the polarity of the corona poling procedure used to fabricate piezoelectric materials may play a role in determining biological responses.

Electrically charged polymeric substrates enhance nerve fibre outgrowth in vitro.

The physical, chemical and electrical properties of synthetic guidance devices are known to influence nerve regeneration in vivo. In the present study, neurons were cultured directly on electrically charged polymer growth substrates to determine if local electrical charges enhance nerve fibre outgrowth in vitro. Piezoelectric polymers such as polyvinylidene fluoride (PVDF) generate transient surface charges under minute mechanical strain. Mouse neuroblastoma (Nb2a) cells were cultured directly on electrically poled (i.e. piezoelectric) and unpoled (i.e. nonpiezoelectric) PVDF substrates in serum-free and serum-containing media. Nerve fibre outgrowth was analysed 24, 48, 72 and 96 h after plating. Piezoelectric PVDF substrates generated 2-3 mV at 1200 Hz when placed on standard incubator shelves and unpoled PVDF substrates showed no output. Nb2a cells grown on piezoelectric substrates exhibited significantly greater levels of process outgrowth and neurite lengths at all time periods for both media conditions. Detailed surface characterization of PVDF substrates using electron spectroscopy for chemical analysis (ESCA) and a comprehensive wettability profile revealed that poled and unpoled PVDF was chemically indistinguishable and showed similar surface wettabilities and adhesive properties. Therefore, we conclude that enhanced process outgrowth was induced by the film's piezoelectric output, making poled PVDF a unique biomaterial for which cell/polymer interactions are mediated predominantly through bulk electrical properties rather than surface properties.

Controlled release of platelet-derived growth factor using ethylene vinyl acetate copolymer (EVAc) coated on stainless-steel wires.

Controlled delivery of bioactive molecules to modulate or control biological processes has a number of clinical applications. The present study reports a delivery system which was designed to deliver growth factors locally to a fracture site. Ethylene vinyl acetate copolymer (EVAc), bovine serum albumin (BSA) and platelet-derived growth factor (PDGF-BB) were combined and coated onto a stainless-steel Kirschner wire (K-wire). A K-wire can be used as an intramedullary nail in small animal fractures, such as the rat. PDGF-BB stimulates thymidine uptake in human bone cell cultures when released from the K-wire delivery system. BSA release was modified by providing a final coating on the K-wire of 10% pure EVAc at the end of the fabrication. Electron microscopic examination of the surface of the rods revealed different surface pores on the K-wires coated with pure EVAc. Differences in porosity and tortuosity may account, in part, for the different release kinetics observed.

RGD-coated titanium implants stimulate increased bone formation in vivo.

Numerous studies have demonstrated that peptide modified surfaces influence short- and long-term cell responses such as attachment, shape and function in vitro. These responses are mediated via cell receptors known as integrins which bind specifically to short peptide sequences from larger proteins. Integrins transduce information to the nucleus through several cytoplasmic signalling pathways. Little is known, however, about the ability of peptide-coated surfaces to influence cell responses in vivo. The present study was designed to evaluate the quality and quantity of the new bone formed in response to titanium rods surface-coated with the peptide sequence Arg-Gly-Asp-Cys (RGDC) using gold-thiol chemistry and implanted in rat femurs. Histomorphometric analysis of cross-sections perpendicular to the implant long axis showed a significantly thicker shell of new bone formed around RGD-modified versus plain implants at 2 weeks (26.2 +/- 1.9 vs. 20.5 +/- 2.9 microm; P < 0.01). A significant increase in bone thickness for RGD implants was also observed at 4 weeks while bone surrounding controls did not change significantly in thickness (32.7 +/- 4.6 vs. 22.6 +/- 4.0 microm; P < 0.02). Mechanical pull-out testing conducted at 4 weeks revealed the average interfacial shear strength of peptide modified rods was 38% greater than control rods although this difference was not statistically significant. These pilot data suggest that an RGDC peptide coating may enhance titanium rod osseointegration in the rat femur. Long-term studies and evaluation of other peptides in larger animal models are warranted.

The morphology of regenerating peripheral nerves is modulated by the surface microgeometry of polymeric guidance channels.

The present study was designed to evaluate the influence of synthetic guidance channel surface microgeometry on morphological patterns of neural regeneration. Tubes with smooth (S), rough (R), or alternating smooth-rough (S/R) or rough-smooth (R/S) inner surfaces but with identical chemical composition and permeability characteristics were used to bridge a 4-mm nerve gap in a transected mouse sciatic nerve. Animals received S and R channels for 1, 2 and 4 weeks and both S/R and R/S channels for 2 and 4 weeks. At 1 week, the S tubes contained a longitudinally oriented fibrin matrix not contacting the channel's smooth inner wall, whereas R tubes featured an unorganized fibrin matrix which, together with fibroblasts and macrophages, had invaded the channel's rough trabecular network. After 4 weeks, S tubes contained discrete, free-floating nerve cables with numerous myelinated and unmyelinated axons surrounded by a thin, continuous epineurial-like layer, whereas R tubes were completely filled with a loose connective tissue stroma with only a few axons. In combined S/R or R/S channels, the general morphological patterns in individual S or R segments were similar to those observed in pure S or R channels, regardless of whether the tube segment was positioned at the proximal or distal nerve end. Proximal smooth channel segments contained discrete cables which abruptly fanned out to completely fill the lumen in distal rough segments. The opposite pattern was observed with proximal rough and distal smooth segments. At 4 weeks, myelinated axons were observed along the entire length of S/R and R/S tubes. These results suggest that the surface microgeometry of guidance channels influences the outcome of peripheral nerve regeneration, potentially by affecting the early arrangement of the fibrin matrix and/or inducing different cellular responses.

Polymer electret guidance channels enhance peripheral nerve regeneration in mice.

Polytetrafluoroethylene (PTFE) tubes were prepared as electrets displaying a quasi-permanent surface charge due to the presence of trapped monopolar charge carriers. PTFE tubes containing either positive or negative charges and electrically neutral PTFE tubes were used as nerve guidance channels for the repair of a 4 mm nerve gap in the sciatic nerve of mice. After 4 weeks of implantation, positively and negatively charged PTFE electrets contained regenerated nerves with significantly more myelinated axons than nerves regenerated in uncharged PTFE tubes. This observation suggests that peripheral nerve regeneration can be enhanced by electrically charged nerve guidance channels.

Piezoelectric guidance channels enhance regeneration in the mouse sciatic nerve after axotomy.

Piezoelectric nerve guidance channels made of polyvinylidene fluoride (PVDF) were evaluated in a transected mouse sciatic nerve model. Poled PVDF channels were compared to unpoled PVDF channels after 4 and 12 weeks of implantation. In all animals, the proximal and distal nerve stumps were bridged by a continuous nerve cable. Nerves regenerated in poled channels contained a higher number of myelinated axons than those regenerated in unpoled channels at both time periods. We conclude that piezoelectric nerve guidance channels enhance peripheral nerve regeneration and provide a tool to investigate the influence of electrical activity on nerve regeneration.

Blind-ended semipermeable guidance channels support peripheral nerve regeneration in the absence of a distal nerve stump.

The presence of a distal nerve segment is considered to be essential for peripheral nerve regeneration through impermeable synthetic guidance channels. The use of a perm-selective material may provide a more appropriate regenerating environment by allowing solute exchange across the wall of the channel. We compared perm-selective acrylic copolymer (AC) channels with impermeable silicone elastomer (SE) channels in terms of regeneration in the absence of a distal nerve stump. Cohorts of 6 animals received AC and SE channels for either 4 or 8 weeks, with the distal end of the polymer tube left open in half of the animals, and plugged with the same polymer ('capped') in the other half. Capped and uncapped AC channels contained regenerated nerve cables which extended fully to the distal end of the channel, whereas capped SE channels contained only 1 mm long granulomatous tissue cables, and uncapped SE channels showed small cables with only a few myelinated axons. The nerve cables regenerated in uncapped AC channels were smaller and contained fewer myelinated axons than those observed in capped AC channels. Capped AC channels sleeved with a tight-fitting silicone tube to render them impermeable, showed no regenerated tissue within their lumen. The use of a perm-selective channel may have allowed the influx of nutrients and growth factors from the external environment while concentrating factors released by the proximal nerve stump.

Fabrication and implantation of gel-filled nerve guidance channels.

In human adults, the peripheral nervous system (PNS) is capable of healing and regeneration. In order to reestablish function, nerve tissue must heal by true regeneration of a functional structure, since healing by simple scar will not reestablish electrical connectivity. Nerve guidance systems have been used experimentally to enhance regeneration through the use of functionalized gels, the delivery of growth-promoting molecules, and the use of neuronal support cells or genetically engineered cells. The objectives of this chapter are to overview the methods used to construct gels for nerve stimulating regeneration and to outline the surgical techniques to implant nerve guidance systems.

Effects of radiofrequency glow discharge and oligopeptides on the attachment of human endothelial cells to polyurethane.

Activation of the surface of small diameter polyurethane vascular grafts using radiofrequency glow discharge (RFGD) and covalent linkage of cell attachment oligopeptides may improve graft patency. The effects of RFGD treatment were investigated on a polyurethane-polydimethylsiloxane (PU-PDMS) copolymer (Cardiothane-51; Kontron Cardiovascular, Inc., Everett, MA) membrane fabricated using the spray, phase-inversion technique. RFGD using H2O vapor was used to functionalize the membrane surface, reacted with 1',1 carbonyldiimidazole and covalently bound with oligopeptides (RGDS, RGES). Membranes not subjected to RFGD, either unmodified or with adsorbed fibronectin (Fn), vitro-nectin (Vn), RGDS, RGDV, and RGES, were used as controls. RFGD treated membrane surfaces were evaluated using electron spectroscopy for chemical analysis, which demonstrated a qualitative increase in nitrogen and silicon compared to unmodified PU-PDMS. Indirect confirmation of surface hydroxylation was provided by metallization with palladium and nickel, demonstrating uniform metallization of RFGD treated PU-PDMS surfaces. Human umbilical vein endothelial cells (HUVEC) were seeded at a density of 10(4) cells/cm2 and cell attachment assessed at 3 hr, 24 hr, and 7 days. Untreated PU-PDMS and membrane with adsorbed oligopeptides demonstrated poor HUVEC attachment at all intervals. Adsorbed Fn and Vn had significantly better early cell attachment and growth (p < 0.01). RFGD improved initial attachment and growth over non-RFGD treated controls (p < 0.01) for RGDS bound membranes, which performed as well as Fn controls (N.S.). PU-PDMS membranes can be modified using RFGD to covalently link oligopeptides. RFGD treatment alone, or with covalent linkage of cell attachment oligopeptides, improves HUVEC attachment and growth in a static environment.

Magnitude and polarity of a fluoroethylene propylene electret substrate charge influences neurite outgrowth in vitro.

Positively charged coating materials such as polylysine improve neuronal attachment in vitro. Due to the structural complexity of these charged molecules, it is unclear whether neuronal effects are due to charge or to physicochemical effects, or both. Polymeric materials with charge storage capabilities and defined surface properties may provide a model in which electrical charge and surface property effects can be separated. Fluorinated ethylenepropylene (FEP) films can store negative or positive charges injected through a corona charging process, thus generating a negative or positive external electrostatic field. In the present study, mouse neuroblastoma (Nb2a) cells were cultured on positive, negative, and uncharged FEP substrates, in both serum-containing and serum-free media. Cell attachment, differentiation, and neurite outgrowth were assessed 24, 48, 72 and 96 h after plating. Electron spectroscopy for chemical analysis (ESCA), contact angle analysis, and scanning electron microscopy (SEM) revealed no differences in surface chemistry and topography between positive, negative, and uncharged FEP. No significant differences in the levels of cell attachment on positive, negative, and uncharged substrates were observed. Significantly higher levels of neurite outgrowth, however, were observed with positive substrates as compared to negative and uncharged substrates, in both media conditions. Substrates charged to +1000 V showed greater levels of outgrowth compared to +500 and +3000 V, suggesting the presence of an optimal range of charge for neurite outgrowth. These results show that cell/charge interactions mediate cell effects on electrically charged substrates with identical surface chemistry, topography and adhesivity.

Patterned neuronal attachment and outgrowth on surface modified, electrically charged fluoropolymer substrates.

Fluorinated ethylenepropylene copolymer (FEP) and polyvinylidene fluoride (PVDF) can generate static and transient electrical charges, respectively, after bulk molecular rearrangements induced by electrical charging techniques. Neurons cultured on electrically active FEP and PVDF show increased levels of nerve fiber outgrowth compared to electrically neutral material. The purpose of the present study was to determine if the addition of charged surface groups to the surfaces of FEP and PVDF would modify the influence of bulk electrical charges on cultured neurons. Mouse neuroblastoma (Nb2a) cells were cultured on electrically charged and uncharged FEP and PVDF substrates with covalently modified surfaces containing hydroxyl (OH) and amine (NH2) groups. Surface chemical modification was performed on the entire surface or in discrete striped regions. Nb2a cells cultured on electrically active FEP and PVDF showed greater levels of differentiation than cells on electrically neutral substrates. The presence of NH2 groups attenuated these responses in serum-containing media. Cells attached to NH2 rich surfaces generally displayed a flatter morphology and tended to remain attached for longer time periods. Cells cultured on stripe-modified substrates in serum-containing media showed a strong preferential attachment to modified regions, especially on NH2 stripes. In summary, bulk electrical charges are more important than surface charges in stimulating Nb2a cell differentiation. Surface groups serve to modulate neuronal morphology and confer specific attachment promoting properties in serum-containing media. The development of an optimal neuronal regeneration template may require the incorporation of specific bulk and surface properties.

Collagen- and laminin-containing gels impede peripheral nerve regeneration through semipermeable nerve guidance channels.

Semipermeable guidance channels were filled with saline, collagen-, or laminin-containing gels and used to repair a 4-mm sciatic nerve gap in mice. After 12 weeks, nerve cables regenerated in gel-filled channels displayed fewer myelinated axons than saline-filled channels. Remnants of the exogenous substrates were still in evidence, in amounts related to the initial collagen or laminin gel concentration. The impairment of nerve regeneration by collagen or laminin-containing gels suggests that the regenerative environment created within semipermeable channels is not improved by the addition of growth substrates in a gel form.