Thrombospondin-4 contributes to spinal cord injury-induced changes in nociception

Abstract

Background: Our previous data have indicated that nerve injury-induced up-regulation of thrombospondin-4 (TSP4) proteins in dorsal spinal cord plays a causal role in neuropathic pain state development in a spinal nerve ligation model. To investigate whether TSP4 proteins also contribute to the development of centrally mediated changes in nociception after spinal cord injury (SCI), we investigated whether SCI induced TSP4 dysregulation, and if so, whether this change correlated with changes in nociception in a T9 spinal cord contusion injury model.

Methods: Behavioural sensitivity to mechanical, thermal stimuli and locomotor function recovery were tested blindly in SCI or sham rats post-injury. Intrathecal antisense or mismatch control oligodeoxynucleotides were used to treat SCI rats with nociceptive hyperreflexia, and Western blots were used to measure TSP4 protein levels in dorsal spinal cord samples.

Results: SCI induced below-level hindpaw hypersensitivity to stimuli. TSP4 protein levels are up-regulated in dorsal spinal cord of SCI rats with nociceptive hyperreflexia, but not in SCI rats without nociceptive hyperreflexia. There was no significant difference in motor function recovery post-injury between SCI rats with or without nociceptive hyperreflexia. Intrathecal treatment with TSP4 antisense, but not mismatch control, oligodeoxynucleotides led to reversal of injury-induced TSP4 up-regulation and nociceptive hyperreflexia in SCI rats.

Conclusions: SCI leads to TSP4 up-regulation in lumbar spinal cord that may play a critical role in mediating centrally mediated behavioural hypersensitivity. Blocking this pathway may be helpful in management of SCI-induced changes in nociception.

Figure 1. Development of below-level hypersensitivity to mechanical stimulation in some SCI rats

The 50% paw withdrawal thresholds to von Frey filament stimulation were examined at the hind paws of SCI rats at designated time points. In sham rats, the hindpaw thresholds were between 10-15 grams, similar to the pre-surgery level of naïve rats. Behavioral hyperreflexia as indicated by reduced paw withdrawal thresholds to von Frey filament stimulation (around or below 5 g) was developed in the hindpaws of approximately half of the SCI rats. Another half of the SCI rats had hindpaw sensitivity similar to that from the sham control. Data presented are the means ± SEM from 34 rats in the pre-SCI groups, 15 rats in the sham group, 23 and 25 rats each in the hyperreflexic or non-hyperreflexic groups, respectively. **p < 0.01, ***p < 0.001 compared with the non-hyperreflexic group; ^#p < 0.05, ^###p < 0.001 compared with day 10 post injury as determined by two-way ANOVA analyses.

Fig. 2. Upregulation of TSP4 levels in L4-6 dorsal spinal cord of SCI rats with behavioral hypersensitivity

Total proteins were extracted from L4-6 dorsal spinal cord after behavioral hypersensitivity was fully developed in some SCI rats (30 - 40 days after surgery varying among different SCI groups), and subjected to Western blot analysis for TSP4 protein levels. Representative Western blots were shown on top of each summarized bar graph (means ± SEM) from three naïve rats, six each of the sham, SCI group with (hyperreflexic) or without (non-hyperreflexic) behavioral hypersensitivity. The ratio of TSP4 band density to that of β-actin within each sample was calculated for normalization of sample loading before cross-sample comparison analyses. **p < 0.01, ***p < 0.001 determined by Students’ t test.

Fig. 3. Intrathecal treatments with TSP4 antisense oligodeoxynucleotides reversed upregulation of TSP4 proteins in dorsal spinal cord and behavioral hypersensitivity in SCI rats

A. SCI rats with hyperreflexic hindpaws 40 days post SCI were treated daily for 4 days with TSP4 antisense or mismatch oligodeoxynucleotides (50 μg/rat/day) via direct intrathecal injection into the L5-6 region. Paw withdrawal thresholds (PWT) to von Frey filament stimulation were measured daily before the injection, and continued after the last injection as indicated. Data presented are the means ± SEM from at least seven rats from each group except that 3 rats were in the mismatch oligodeoxynucleotide treatment group for day five and six after treatment initiation. **p < 0.01, ***p < 0.001 compared with pre-treatment levels; and ^##p < 0.01, ^###p < 0.001 compared with the 50 μg/rat/day antisense oligodeoxynucleotide treated group as determined by two-way ANOVA analyses. B. Western blot analysis was used to measure TSP4 levels in dorsal spinal cord collected approximately 24 hrs after the last injection that correlated with the peak anti-nociceptive effects of TSP4 antisense oligodeoxynucleotide treatment. Representative Western blots were shown on top of each summarized bar graph presenting the means ± SEM from five rats in each group. The ratio of TSP4 band density to that of β-actin within each sample was calculated for normalization of sample loading before cross-sample comparison analyses. *p < 0.05 compared with the mismatch oligodeoxynucleotide control treatment as determined by Student's t test.