Mechanistic comparison between spinal and trigeminal neuropathic pain

Abstract

Chronic neuropathic pain is resistant to classical analgesics and is characterised by allodynia, hyperalgesia and spontaneous pain. Pain associated with damage to the trigeminal nerve may be particularly persistent lasting twice as long as that in the spinal nerves, which possibly reflects differences between trigeminal and spinal mechanisms of synaptic plasticity. Neuropathic sensitisation occurs at the first synapses in the dorsal horn of the spinal cord and in the trigeminal spinal complex. The NMDA glutamate receptor plays a key role in this process. It is known to bind to adapter proteins, such as the membrane-associated guanylate kinases (MAGUKs including PSD-95, SAP-102, SAP-97, and Chapsyn-110), linking the receptor to a complex of signalling, anchoring, docking, and scaffolding proteins. One of these adapter proteins, PSD-95, has previously been shown by this laboratory to be crucial in the development of neuropathic pain in the spinal cord.

For this study, we used rodent models of chronic constriction injury of sciatic or trigeminal nerve to investigate the electrophysiological responsiveness of single neurones to mechanical stimuli. This strategy allowed comparison of the degree of sensitisation in the two areas. We also examined changes in expression of NMDA receptor subunits and MAGUK proteins.

Our results show a marked facilitation of responsiveness in thermal and mechanical behavioural reflexes in both spinal and trigeminal neuropathic pain models. Electrophysiological experiments indicated an increase in responsiveness of individual neurons to mechanical stimulation in spinal neuropathic animals but this increase was not as pronounced in trigeminal neuropathic animals. Further differences in electrophysiological response characteristics to various peripheral sensory stimuli between spinal and trigeminal neurons were shown in normal animals and following nerve injury. That is, neurons from neuropathic animals show a marked post-stimulus discharge response (PSDR). The length of discharge was an average of 8333 ±1610 action potentials from spinal cord neurons and 46390 ± 16026 action potentials for trigeminal neurons, whilst the mean threshold force for eliciting a PSDR for spinal neurons was 2.1 fold that for trigeminal neurons.

Trigeminal neurons were also tested for responses to von Frey filaments both before and after a brief brush or cold stimulus applied to the face, ipsilateral to injury. Using ix this protocol for before and after brush stimulation, 14 neurons from trigeminal neuropathic animals were tested. Of these, 7 neurons showed an increased initial response to low (4g) and high (15g) forces after the brush stimulus compared to that before (4g force: 12.2 spikes per second ± 0.9 before and 24.3 spikes per second ± 2.9 after. 15g force 19.6 spikes per second ± 5.8 before and 26.8 spikes per second ± 7.0 after. Conditioning stimuli experiments were not carried out in spinal cord preparation animals due to time constraints.

Biochemical experiments revealed that changes in expression of some NMDA receptor subunits, as well as associated MAGUK proteins, differed between spinal and trigeminal neuropathic animals, and within different regions of the trigeminal complex itself. A reduction in NR1 expression in the spinal cord ipsilateral to CCI compared with the contralateral side a mean reduction of 27%, was shown, whilst no change in NR1 expression was seen in any of the trigeminal regions investigated. Differential injury-induced changes were also seen in NMDA R-interacting proteins. PSD-95 shows no change in expression in regions of the trigeminal complex following CCI, but does increase in expression ipsilateral to nerve injury in the spinal cord, a mean increase of 140% compared to the contralateral side. The rise in NR2B subunit and PSD-95 protein expression at a time concomitant with the development of neuropathic pain behaviours is consistent with previous reports showing the necessity for an intact NR2B-PSD-95 complex for the development of neuropathic behaviours in PSD-95 mutant mice (Garry et al, 2003). Furthermore, Chapsyn-110/PSD-93 which shows no demonstrable change in expression in the spinal cord exhibits a marked 40% decrease in ipsilateral expression in the trigeminal caudalis compared to the contralateral side following CCI.

We further investigated the potential role of proteins such as persyn (known to influence cytoskeletal network integrity) and a-synuclein (implicated in cell death), which may particularly influence the development, duration or recovery from neuropathic pain. We investigated the role of persyn and a-synuclein proteins in neuropathic pain, using two null-expression mutant mouse strains. However, no substantial differences were observed between the reflex behavioural responses of these mutant animals and wild type animals following nerve injury.

In conclusion, this study provides evidence for mechanistic differences in neuropathic sensitisation between trigeminal and spinal regions. These differences may lead to targets for improved therapeutic treatment of intractable pain states.