Differential expression of GABAB receptor subtypes in the cerebellum and their involvement in motor function

Charlotte Mann1, Joshua Foster1, Bernhard Bettler2, Ian Kitchen1, Ying Chen1. 1Faculty of Health and Medical Sciences, University of Surrey, Guildford, GU2 7XH, United Kingdom, 2Department of Biomedicine, Institute of Physiology, Pharmazentrum, University of Basel, Basel, CH-4056, Switzerland.

GABAB receptors are the G-protein coupled receptors of the inhibitory neurotransmitter GABA. Functional GABAB receptors require the co-assembly of the GABAB1 and GABAB2 subunits1, and the two main isoforms of the GABAB1 subunit, GABAB1a and GABAB1b, form receptor subtypes. The GABAB receptors are highly expressed in the cerebellar cortex, with the predominant expression of the GABAB1a isoform in granule cells, and GABAB1b in Purkinje cells2. This reflects their respective roles in producing pre- and postsynaptic inhibition at the granule cell-Purkinje cell synapse. Although it is known that GABAB receptor agonists can cause ataxia3, it is not known whether the two receptor subtypes are differentially involved in cerebellar motor function and co-ordination.

The expression of GABAB receptor subunits in the cerebellum was examined in GABAB1a-/- and GABAB1b-/- mice using immunoperoxidase labelling of the GABAB1 and GABAB2 subunits. We confirmed the predominant expression of GABAB1b in Purkinje cells and GABAB1a in granule cells, and found that the GABAB2 immunostaining was primarily confined to the molecular layer, producing a striking array of parasagittal stripes in both wild-type and GABAB1a-/- mice that closely resemble the characteristic pattern of anti-zebrin II immunoreactivity. In contrast, the GABAB1b-/- cerebellum produced diffuse immunopositive staining throughout the molecular layer. As the GABAB2 subunit is essential for cell surface expression of GABAB1 and G-protein coupling, the expression pattern of GABAB2 may represent that of functional heteromeric receptors.

To investigate whether GABAB receptor subtypes have differential roles in motor function, a battery of behavioural tests were carried out in GABAB1a-/- and GABAB1b-/- mice. Footprint analysis4 showed that the average stride length of 6.7 ± 0.2 cm was not different between wild-type, GABAB1a-/- mice and GABAB1b-/- mice. However, GABAB1a-/- mice walked with their hind feet further apart than those of WT mice (p<0.05, One way ANOVA). In contrast, the GABAB1b-/- mice showed a significantly greater hind base to front base ratio when compared to both WT (p<0.05) and GABAB1a-/- mice (p<0.05, Bonferonni post hoc test). In addition, in Hang wire test5, only 50% of the GABAB1a-/- mice passed (n = 14) in comparison to 92% of the wild-type (n = 13) and 83% of the GABAB1b-/- mice (n = 12). This deficit was not influenced by body weight, as GABAB1a-/- mice did not show significant muscle weakness in a grip strength test5 (188.1 ± 6.0 gf, n = 14; compared to 190.6 ± 5.4 gf, n = 13 for the wild-type and 203.9 ± 6.7 gf, n = 12 for the GABAB1b-/- mice).

In conclusion, both GABAB1a-/- mice and GABAB1b-/- mice show deficits in gait and the vestibular control function of GABAB1a-/- mice may be impaired. Differential roles of GABAB receptor subtypes in motor function were demonstrated, and the differential localisation of GABAB subtypes at the granule cell-Purkinje cell synapse may be responsible.

1. Jones et al., (1998) Nature. 396; 674-9

2. Billinton et al., (1999) Br. J. Pharmacol. 126; 1387-92

3. Carter et al., (2005) J. Pharmacol. Exp. Ther. 313; 1314-23

4. Carter et al., (1999) J. Neurosci. 19; 3248-57

5. Glynn et al., (2005) Hum. Mol. Genet. 14; 2369-85