Cell therapy for nervous tissue repair is limited by low transplant survival. We investigated the effects of a polyurethane-based reverse thermal gel, poly(ethylene glycol)-poly(serinol hexamethylene urethane) (ESHU) on bone marrow stromal cell (BMSC) transplant survival and repair using a rat model of spinal cord contusion. Transplantation of BMSCs in ESHU at three days post-contusion resulted in a 3.5-fold increase in BMSC survival at one week post-injury and a 66% increase in spared nervous tissue volume at four weeks post-injury. These improvements were accompanied by enhanced hindlimb motor and sensorimotor recovery. In vitro, we found that ESHU protected BMSCs from hydrogen peroxide-mediated death, resulting in a four-fold increase in BMSC survival with two-fold fewer BMSCs expressing the apoptosis marker, caspase 3 and the DNA oxidation marker, 8-Oxo-deoxyguanosine. We argue that ESHU protected BMSCs transplanted is a spinal cord contusion from death thereby augmenting their effects on neuroprotection leading to improved behavioral restoration. The data show that the repair effects of intraneural BMSC transplants depend on the degree of their survival and may have a widespread impact on cell-based regenerative medicine.

The ability of intraspinal bone marrow stromal cell (BMSC) transplants to elicit repair is thought to result from paracrine effects by secreted trophic factors including brain-derived neurotrophic factor (BDNF). Here we used gene therapy to increase or silence BDNF production in BMSCs to investigate the role of BDNF in BMSC-mediated neuroprotection. In a spinal cord organotypic culture, BMSC-conditioned medium significantly enhanced spinal motoneuron survival by 64% compared with culture medium only. Only conditioned medium of BDNF-hypersecreting BMSCs sustained this neuroprotective effect. In a rat model of spinal cord contusion, a BDNF-dependent neuroprotective effect was confirmed; only with a subacute transplant of BDNFhypersecreting BMSCs were significantly more spared motoneurons found at 4 weeks postinjury compared with vehicle controls. Spared nervous tissue volume was improved by 68% with both control BMSCs and BDNF-hypersecreting BMSCs. In addition, blood vessel density in the contusion with BDNF-hypersecreting BMSCs was 35% higher compared with BMSC controls and sixfold higher compared with vehicle controls. BDNF-silenced BMSCs did not survive the first week of transplantation, and no neuroprotective effect was found at 4 weeks after transplantation. Together, our data broaden our understanding of the role of BDNF in BMSC-mediated neuroprotection and successfully exploit BDNF dependency to enhance anatomical spinal cord repair.

The efficacy of protein-based therapies for treating injured nervous tissue is limited by the short half-life of free proteins in the body. Affinity-based biomaterial delivery systems provide sustained release of proteins, thereby extending the efficacy of such therapies. Here, we investigated the biocompatibility of a novel coacervate delivery system based on poly(ethylene argininylaspartate diglyceride) (PEAD) and heparin in the damaged spinal cord. We found that the presence of the [PEAD:heparin] coacervate did not affect the macrophage response, glial scarring, or nervous tissue loss, which are hallmarks of spinal cord injury. Moreover, the density of axons, including serotonergic axons, at the injury site and the recovery of motor and sensorimotor function were comparable in rats with and without the coacervate. These results revealed the biocompatibility of our delivery system and supported its potential to deliver therapeutic proteins to the injured nervous system.