Role of membrane-associated guanylate kinases in somatosensory cortical development
In order to process information, neurons must connect together to form a neuronal circuit. The formation of neuronal circuits is dependent on synaptic activity through glutamate receptors and downstream molecules within the post-synaptic density (PSD). The pathways downstream of glutamate receptors play an important role in maintaining appropriate synapses and forming neural circuits; mutations in genes that encode PSD proteins disrupt these pathways and are associated with many forms of intellectual disability in humans (Grant 2012). The development of neuronal circuits relies on two key developmental events; neurons must first send out axons locally and to disparate brain regions and then, neurons must form connections with the dendrites of target neurons. The rodent trigeminal system is a neuronal circuit that processes somatosensory information from whiskers on the facepad via nuclei in the brainstem to the thalamus and ultimately the cerebral cortex. Brain regions that comprise the trigeminal system are organised in a manner that topographically recapitulates the whisker pattern; each whisker on the rodent facepad corresponds to a physiological and anatomical unit in the primary somatosensensory cortex (S1) called a barrel. This topographical organisation creates a pattern consisting of thalamocortical axons (TCA) clustered into distinct whisker-related bundles and layer IV spiny stellate cells which segregate around the outside of TCA bundles. Three different anatomical patterns can be identified within the mouse S1 by labelling the cell soma, axons or the extracellular matrix. This strict organisation makes the rodent S1 an excellent model for discerning the proteins involved in neural circuit formation, and by screening genetic mutants for S1 patterning defects the molecular pathway involved in setting up neuronal circuits can be elucidated. Furthermore an understanding of these pathways may provide insight into how neuronal networks are disrupted in human intellectual disability. In the first data chapter, the expression profile of three neurofilament subunits were characterised in order to develop a method of identifying anatomical defects in barrel morphology. The precise organisation of the rodent S1 can also be used as a method to identify the cellular localisation of neurofilament subunits ex vivo. Neurofilaments are polymers formed from three subunits identified by their relative molecular weight. By using the unique patterning of S1, each neurofilament subunit shows a unique spatialtemporal expression pattern in the somatosensory cortex. Two neurofilaments subunits; the medium and the heavy neurofilament subunits can be used to identify TCA which can be used as an indicator of anatomical defects in barrel patterning. In chapter 4 neurofilament labelling was used in conjunction with other histological techniques to investigate S1 organisation in mice lacking synapse associated protein 102 (SAP102). SAP102 is a PSD scaffolding molecule that binds to both NMDA receptor subunits and SynGAP, a synaptic GTPase activating protein, furthermore it is associated with X-linked mental retardation in humans (Tarpey et al., 2004; Zanni et al., 2010). Mutant mice lacking functional NMDA receptors or PSD proteins such as SynGAP show defects in S1 pattern formation (Barnett et al., 2006; Iwasato et al., 2000; Wijetunge, Till, Gillingwater, Ingham, & Kind, 2008). SAP102 null mutants (SAP102-/y) have defects in synaptic plasticity and are slow to learn on behavioural tasks (Cuthbert et al., 2007), however it is unclear how the loss of SAP102 may disrupt neural networks. SAP102-/y were found to have a reduction in brain mass compared to wild-type littermates, but cortical thickness and patterning of S1 is unchanged. SAP102-/y have fewer TCA reaching the cortex compared to littermates; furthermore SAP102-/y mutants have layer specific defects in the density of dendritic spines. These data suggest that in the absence of SAP102 connectivity in the S1 is altered by layer specific changes in synapses number and fewer axons innervating cortical layer IV. In the final experimental chapter (chapter 5) the combined role of SAP102 and Postsynaptic density protein 95 (PSD95) in S1 patterning was investigated. SAP102 and PSD95 are the main members of the membrane-associated guanylate kinase (MAGUK) family expressed during early cortical development. These proteins share a similar protein structure, perform similar functions at the synapse (Elias & Nicoll, 2007) and have been shown to compensate for each other in vitro (Elias, Elias, Apostolides, Kriegstein, & Nicoll, 2008). Genetic mutants lacking both SAP102 and PSD95 are not viable and do not survive beyond birth (Cuthbert et al., 2007; Petrie, 2008). Therefore in order to investigate the combined role of these proteins a novel approach was developed that utilises X-inactivation to produce mosaic animals containing cells that lack both proteins. The distribution of cells containing the same X-chromosome was investigated and found to be evenly distributed throughout the cortex, demonstrating that this method could be used to investigate allosomal genes. In mosaic animals where approximately half the cells only lack PSD95 and the remaining cells lack both SAP102 and PSD95, double knockout cells are viable and are equally represented in S1. These double knockout cells contribute to normally barrel formation which suggests that SAP102 and PSD95 are not required for barrel formation.