Glycosylation in the injured and regenerating spinal cord, and the influence of biomaterial treatment

Abstract

Traumatic spinal cord injury (SCI) results in disruption of tissue integrity leading to a loss of function. There has been minimal success in treating these injuries clinically, due to the inhibitory environment which develops over time. In this work, we study the glycosylation response to SCI in two models: Xenopus laevis is used to compare successful to failed regeneration, by comparing the response to injury pre- and post-metamorphosis, while we use a rat model to understand the pathophysiology of the injury and how this may be modified by treatment with an aligned collagen hydrogel. We hypothesise that glycosylation is altered in response to injury, that it has a role in determining the success of regeneration, and that biomaterial treatment can influence this glycosylation response. Complete transection was used to model SCI in pre- and post-metamorphic stages of Xenopus laevis and in adult rat. Collagen hydrogels with aligned or non-aligned fibres were placed at the transection site in the rat spinal cord only. Lectin histochemistry was used to investigate glycosylation changes in Xenopus laevis. HILIC-UPLC N-glycoprofiling was used to characterise the normal N-glycome of the healthy adult rat, and to investigate changes in response to SCI and collagen hydrogel treatment. Immunohistochemistry was used to investigate neuronal or glial changes in response to the injury in each species. In the healthy rat spinal cord UPLC N-glycoprofiling identified all three major categories of N-glycan, i.e. complex, hybrid and high mannose. Many of the complex structures terminated in galactose and were commonly decorated with core and outer arm fucose residues. There was a low level of sialylation. Following SCI there was a decrease in complex glycans in the rat spinal cord and a corresponding increase in high mannose and hybrid structures, with and without core fucose. There was a loss of outer arm fucosylation, while bisecting GlcNAc and sialic acid were increased. Treatment with aligned collagen hydrogel lead to small changes in the glycosylation response. Pre-metamorphic Xenopus laevis regenerated the spinal cord within 10 days, with a thin tissue bridge seen at 7 days. There was significantly more GFAP in the non-regenerating froglet spinal cord. In both tadpole and froglet Xenopus there was an early increase in sialic acid following SCI but this was far more pronounced in the non-regenerative froglet. In the rat injury also resulted in an increase in sialic acid, which was found to be associated with microglia or macrophages. Sugars carrying GlcNAc were increased during the tadpole regenerative response but were decreased in the froglet. This may relate to N-glycan complexity which was decreased following SCI in the rat. There were also differences in GalNAc decoration depending on the regenerative potential, being increased in the tadpole but decreased in the froglet. This sugar was not identified on rat N-glycans. This work suggests that glycosylation may influence the success of regeneration. In particular an early but transient increase in sialic acid may be important for repair, and the loss of complexity in rat spinal cord glycans may be a contributing factor to the failure of regeneration