ANALYSES OF THE FUNCTION OF THE PALMITOYL TRANSFERASES GODZ AND SERZ-BETA IN KNOCK-OUT MICE

Open Access
Author:
Murakami, Shoko
Graduate Program:
Neuroscience
Degree:
Doctor of Philosophy
Document Type:
Dissertation
Date of Defense:
September 26, 2008
Committee Members:
  • Bernhard Luscher, Dissertation Advisor
  • Bernhard Luscher, Committee Chair
  • Richard W Ordway, Committee Member
  • Robert Paulson, Committee Member
  • Gong Chen, Committee Member
  • Erin Elizabeth Sheets, Committee Member
Keywords:
  • knock out
  • trafficking
  • Neuroligin 2
  • GABAA receptor
  • palmitoylation
Abstract:
Type A GABA receptors (GABAARs) are the major sites of fast synaptic inhibition in the central nervous system (CNS) and mediate the therapeutic effects of many clinically important drugs. The efficacy of synaptic inhibition is critically dependent on the postsynaptic accumulation of GABAARs. Golgi-specific DHHC zinc finger protein (GODZ; aka, DHHC3) and SERZ-beta (aka, DHHC7) have been identified as palmitoyl acyl-transferases (PATs) of the gamma2 subunit of GABAARs. In addition, the cell adhesion molecule neuroligin 2 (NL2) is localized at GABAergic synapses and implicated in GABAergic synapse formation. Previous analyses of GODZ by RNAi in cultured neurons had suggested that they are important for postsynaptic accumulation of GABAARs at inhibitory synapses. In addition these studies pointed to a novel possible role of GABAARs in the assembly of GABAergic inhibitory synapses. A first objective of this doctoral thesis was to elucidate the function of GODZ and SERZ-beta in vivo. In order to achieve this purpose, global and conditional knock-out (KO) mice were generated for both GODZ and SERZ-beta. Genetic deletion of GODZ and SERZ-beta was confirmed by genomic Southern blotting, RT-PCR and immunofluorescence analyses. Global GODZ and SERZ-beta KO mice were found to be viable, whereas GODZ and SERZ-beta double knock-out (DKO) mice showed a partially penetrant perinatally lethal phenotype. Male, but not female, GODZ KO mice exhibited a small but significant reduction in body weight compared to wild type (WT) littermate controls. A more dramatic reduction in body weight was evident in DKO mice, as quantified in females but also overtly evident in males. Compared to WT littermate controls, GODZ KO mice showed complex behavioral alterations characterized by excessive jumping in response to handling, hindleg clenching upon suspension by their tail, hyperlocomotion in an Open Field test, as well as enhanced Prepulse Inhibition of acoustic startle responses. However, GODZ KOs showed normal anxiety related behavior in Free Choice and Elevated Plus Maze tests. Immunofluorescence analyses of cortical neuron cultures prepared from GODZ KO or GODZ/SERZ-beta DKO mice revealed unaltered numbers of postsynaptic clusters of gamma2 subunit containing GABAARs and the inhibitory postsynaptic scaffold protein gephyrin. Similarly, the number of clusters of the corresponding presynaptic marker glutamate decarboxylase (GAD) and vesicular inhibitory amino acid transporter (VIATT) were also unchanged. Consistent with unaltered clustering in pure DKO cultures, the surface expression levels of the gamma2 subunit and NL2 assessed by biotinylation was unaltered in DKO compared to WT cultures. However, when mutant neurons were co-cultured with an excess of WT neurons, punctate staining for postsynaptic GABAAR clusters and presynaptic inhibitory synaptic markers of GODZ KO and DKO neurons was dramatically reduced compared to neurons of pure WT cultures. Similarly, gamma2 subunit heterozygous neurons co-cultured with an excess of WT neurons showed a significant reduction in inhibitory synapse formation, suggesting that gamma2 subunit-containing GABAARs contribute to normal GABAergic innervation. In addition, electrophysiological analyses of DKO neurons co-cultured with WT neurons revealed unaltered GABA-induced whole cell current but significant deficits in the frequency and amplitude of miniature inhibitory postsynaptic currents (mIPSCs) compared to WT neurons. These results suggest that GODZ and/or SERZ-bata are dispensable for normal surface expression of GABAARs in pure cultures but necessary for postsynaptic clustering of GABAARs and normal formation and function of inhibitory synapses under competitive conditions. As a second objective I further analyzed the role of GABAARs in GABAergic synapse formation. Preliminary immunofluorescence analyses of GABAAR gamma2 subunit KO neurons had revealed a dramatic loss of postsynaptic NL2 at inhibitory synapses, indicating that perhaps the gamma2 subunit was required for normal expression of a synaptic adhesion molecule, which in turn was required for GABAergic inhibitory synapse formation. To test this idea and to analyze the surface expression of GABAARs in gamma2 subunit KO neurons I performed a series of biotinylation assays, comparing the surface expression of NL2, alpha and beta subunits of GABAARs in WT and gamma2 subunit KO cultures. Although the total amount of NL2 and beta2/3 subunit expressed in gamma2 KO cultures was unaltered, the surface expression level of NL2, alpha1 and beta2/3 subunits was greatly reduced in gamma2 KO compared to WT neurons. This indicated that the gamma2 subunit is necessary for the surface expression of both NL2 and GABAARs. Previously, co-aggregation assay in transfected HEK293T cells provided evidence that NL2 interacts with alpha and/or beta subunits, but not the gamma2 subunit, in the plasma membrane. These results together suggest a mechanism in which GABAAR-dependent presentation of NL2 at the cell surface promotes interaction of the postsynaptic apparatus with GABAergic presynaptic terminals, thereby ensuring apposition of functionally matching pre- and postsynaptic elements.