Spinal cord injury affects nearly twenty five thousand Americans, with an additional ten thousand new injuries per year. These injuries are usually life altering due to death of nervous tissue in the spinal cord creating a void where the electrochemical signal can no longer pass. The current treatment options for such injuries are limited due to the body's inability to regenerate neurons and the difficulty in stabilizing the defect created by the injury. This project proposes the use of a non-biodegradable hydrogel as a mechanical support to fill the injury site and provide a protected environment for the regenerating neurons. The hydrogel can be designed to match the mechanical stiffness of the local tissue to reduce the amount of stress shearing around the implant. The use of a thermally responsive, injectable polymer, PNIPAAm-PEG branched copolymer, is proposed as a minimally invasive surgical technique. Below its LCST, typically around 29-320C, the polymer forms a miscible solution with water, but above its LCST, it becomes hydrophobic, separating from water and forming a semi porous gel. The aqueous polymer solution can be created with cell culture media and seeded with bone marrow stromal cells in order to physically entrap cells into the scaffold when it is injected into the defect. Bone marrow stromal cells (MSC) are progenitor cells that when promoted with neurotrophic factors differentiate into neuronal cells. In order to enhance the neuronal cell proliferation, neurotrophic factors can be released in the local defect via biodegradable microparticles. Polymer microparticles allow the trophic factors to release in an initial burst, which promotes cell growth into the scaffold, and then lowers to a sustained release, which maintains the cells in the scaffold. Swelling studies have been done to characterize the changes in the hydrogel with different swelling media as well as with the addition of polymer microparticles. Preliminary work has shown that changing the aqueous solution from phosphate buffer to cell culture media with fetal bovine serum changes the physical properties of the hydrogel increasing its ability to retain volume. In order to further increase the scaffolds volume retention, so as not to shrink post injection, the copolymer was blended with poly(vinyl propylene)(PVP). The addition of PVP to the copolymer resulted in almost 100% volume retention post injection. The mechanical properties of the scaffolds were evaluated by unconfined compression as well as tensile testing at 370C in their respective swelling media. The compressive modulus of the scaffold (found from the unconfined compression testing), is a critical design parameter. The compressive modulus of the local spinal cord tissue is around 3-5 kPa1. In order to meet this requirement, the percent of copolymer dissolved into solution can be changed as well as the amount of PEG relative to PNIPAAm, and the length of the PEG chain. The tensile data was used in combination with the swelling data to find the molecular weight between crosslinks in the hydrogel. Stress relaxation studies were also done in order to determine the relaxation time constant of the different scaffolds. The morphological effect of swelling media and microparticles on the hydrogel pore structure was analyzed using environmental scanning electron microscopy (ESEM) as well as liquid extrusion porosimetery. It is important that the scaffold have an interconnected pore structure so that the cells can grow into these pores. Preliminary studies have shown that increasing the PEG content relative to PNIPAAm as well as increasing the length of the PEG chain increases the amount of pores seen via ESEM in the scaffold. In order to asses the cell viability with the polymer, hydrogels with rat marrow stromal cells entrapped were incubated at 370C, 5% CO2. At the designated time point, the gel was stained with a dual fluorescence stain, which allowed a qualitative assessment of live versus dead cells. Preliminary studies have shown that a significant number of live cells remain in the scaffold after 4 days of incubation in vitro. Further studies will be done in vivo in rats with hemi sectional spinal cord injuries. Injections of the copolymer solution with cells will be done to evaluate the ability of the scaffold to fill the injury site as well as keep the MSC in the local injury area. Injections of cells alone have shown that the cells do not stay in the local injury site, and therefore the number of cells required for injection is much larger than needed for treatment2. Upon optimization of the copolymer plus cell injections, studies will then include microparticles in the injections. Microparticles were synthesized using either a water-in-oil-in-water (W/O/W) double emulsion solvent extraction method or a solid-in-oil-in-water (S/0/W) double emulsion solvent extraction. In order to optimize the drug loading, the amount of polymer and drug were varied. It was found that the optimal method was using dichloromethane (DCM) however; DCM has been shown to destroy about 40% of the native structure of the protein loaded in the particle. Therefore, moving to a less hazardous solvent, ethyl acetate, gives lower encapsulation in the W/O/W method, but reasonable values using S/O/W. Preliminary release studies done in sink conditions in PBS, at 370C have shown that particles made from poly (lactic acid) PLA release for 2-4 days. In order to extend the release, studies using different polymers are being evaluated.

1 Ozawa, et al, J Neurosurg (Spine 1) 1: I22-I27 (2004). 2 Neuhuber et al, Brain Res. 2005 Feb 21;1035(1):73-85.