Nociception in the Glycine Receptor Deficient Mutant Mouse Spastic

Abstract

Glycine receptors (GlyRs) are the primary mediators of fast inhibitory transmission in the mammalian spinal cord, where they modulate sensory and motor signaling. Mutations in GlyR genes as well as some other genes underlie the hereditary disorder hyperekplexia, characterized by episodic muscle stiffness and exaggerated startle responses. Here, we have investigated pain-related behavior and GlyR expression in the spinal cord of the GlyR deficient mutant mouse spastic (spa). In spastic mice, the GlyR number is reduced due to a β subunit gene (Glrb) mutation resulting in aberrant splicing of GlyRβ transcripts. Via direct physical interaction with the GlyR anchoring protein gephyrin, this subunit is crucially involved in the postsynaptic clustering of heteromeric GlyRs. We show that the mutation differentially affects aspects of the pain-related behavior of homozygous Glrb^spa/Glrb^spa mice. While response latencies to noxious heat were unchanged, chemically induced pain-related behavior revealed a reduction of the licking time and an increase in flinching in spastic homozygotes during both phases of the formalin test. Mechanically induced nocifensive behavior was reduced in spastic mice, although hind paw inflammation (by zymosan) resulted in allodynia comparable to wild-type mice. Immunohistochemical staining of the spinal cord revealed a massive reduction of dotted GlyRα subunit immunoreactivity in both ventral and dorsal horns, suggesting a reduction of clustered receptors at synaptic sites. Transcripts for all GlyRα subunit variants, however, were not reduced throughout the dorsal horn of spastic mice. These findings suggest that the loss of functional GlyRβ subunits and hence synaptically localized GlyRs compromises sensory processing differentially, depending on stimulus modality.

Keywords: glycine receptor; glycinergic inhibition; nociception; pain; spastic; synaptic clustering.

Figure 1

Region-specific expression of GlyR mRNA at the spinal level. GlyR mRNA expression was investigated in dorsal root ganglia (drg), superficial dorsal horn (sfdh) and ventral horn of the spinal cord (vsc). For GlyRβ, multiple bands in tissues derived from Glrb^spa/spa animals correspond to truncated transcripts generated by exon skipping (413 bp, transcript lacks exons 4 and 5; 643 bp, transcript lacks exon 5); the upper band represents the full-length transcript (726 bps; Mülhardt et al., 1994). No GlyRα subunit RNA could be detected in DRGs, whereas GlyRβ was expressed. Transcripts encoding GlyRα1–4 and GlyRβ were found in the superficial dorsal horn (sfdh; sizes: α1: 831 bp; α2: 1,152 bp; α3: 1,158 bp; α4: 509 bp; β - full length: 726 bp). The predominant subunit in the ventral spinal cord (vsc) was GlyRα1 (low signal). In Glrb^spa/spa mice, the regional distribution of glycine receptor mRNAs remained unchanged.

Figure 2

Distribution of GlyR immunoreactivity in the spinal cord. (A) Confocal images of glycine receptor α-subunits (GlyRa; mAb4a; red) and AMPA receptors (GluR1; green) immunoreactivity in spinal cord sections of mice with the indicated genotype. The white line indicates the outer boundary of GluR1 fluorescence, which marks the superficial layers (layer I-III) of the sfdh. (B) Example of an immunoflourescence picture of the GlyR and GluR1 fluorescence in the sfdh that was used for quantification (Scale bar = 20 μm). (C) Enlargement of the area within the presumptive layer II of the sfdh depicted in (B). Note that all detectable immunofluorescence is localized in clusters. (D) Overview of GlyRα and GluR1 immunoreactive clusters detected by á-trous-wavelet transformation. (E) Spatial distribution of cluster densities in the confocal images presented in (B). (F) Quantification of the cluster densities in relation to the distance from the surface of the sfdh given in 10 μm binnings.

Figure 3

Motor performance and pain-related behavior. (A) Rotarod-test; mean time on the Rotarod. Motor ability was decreased markedly in mutant mice. (B,C) Heat-sensitivity assays. (B) Tail-flick assay; mean tail flick latencies. (C) Hotplate assay; mean reaction latencies. No differences were observed between the genotypes(***p < 0.001).

Figure 4

Formalin test; PRB after formalin injection. (A) Time course of licking activity (left panel; A1) and mean licking time in phase 1 (0–10 min) and 2 (right panel; A2). The licking time was reduced in spastic animals during the 60 min of the test. (B) Time course of flinches (B1) and mean flinching in phase 1 and 2 (B2). The number of flinches per minute was increased in mutant mice in both phases. Phases of both acute (0–15 min, phase 1) and delayed (>15 min phase 2) PRB could be clearly distinguished by measuring either flinching or licking-time. (C) Mean flinching frequency over the mean licking time frequency at individual minutes comparing spastic (C1) and wildtype (C2) mice. The number of flinches per minute was correlated positively with the mean licking time (*p < 0.05; **p < 0.01).

Figure 5

Mechanically induced nocifensive behavior after zymosan A injection into the right hind paw. (A) Baseline reproducibility; mean withdrawal thresholds (L, left paw; R, right paw); there were no significant differences between the withdrawal thresholds of left and right paw subgroups. WTh’s were elevated in the spastic group. (B) Time course of mean WTh’s after zymosane A injection; both, spastic and wildtype mice develop allodynia. *p < 0.05; **p < 0.01; n.s., not significant. Error bars indicating S.E.M.