As part of our ongoing coverage of DISC1 2010, held 3-6 September 2010, in Edinburgh, the United Kingdom, we bring you a meeting missive from Elise Malavasi, a graduate student at the University of Edinburgh.
5 October 2010. The second Biochemistry and Pharmacology session of Sunday afternoon was moderated by Carsten Korth of Heinrich Heine University Düsseldorf Medical School, Germany. The first speaker, Nick Brandon from Pfizer, Inc., Groton, Connecticut, USA, briefly introduced a new ubiquitous DISC1 RNAi mouse model that his team is currently generating, but then focused his talk on the interaction between DISC1 and TNIK in the regulation of synaptic function (see coverage in SRF related news story). Brandon pointed out that a potential molecular explanation for the effects of TNIK inhibition by DISC1 peptides on synaptic physiology comes from the recent findings published in Neuron by Kawabe and colleagues (Kawabe et al., 2010). The Kawabe paper shows that TNIK is necessary for the interaction between the E3 ubiquitin ligase Nedd4-1 and its substrate Rap2, a small GTPase. Once ubiquitinated by Nedd4-1, Rap2 can no longer activate the 2 kinases MINK and TNIK, and this in turn promotes dendrite arborization. According to these findings, by acting as a bridge between Nedd4-1 and Rap2, TNIK provides a positive feedback in the control of dendrite growth. This explains the observation that TNIK knock-down results in shorter dendrites, while inhibition of TNIK kinase activity but not its linking function has the opposite effect. What could be the position of DISC1 in this picture? Brandon speculated that by simultaneously regulating the TNIK and Kal7/Rac1 pathways, which converge on the actin cytoskeleton, DISC1 could control the morphogenesis and functional maturation of dendritic spines. TNIK KO mice are currently being characterized at Pfizer to better understand the function of this kinase in the physiology of neuronal maturation.
Nick Bradshaw from the University of Edinburgh, UK, presented his work on the biochemical characterization of the DISC1/NDE1/PDE4 pathway, and provided evidence implicating this pathway in neuronal function. Bradshaw found that a combination of forskolin (adenylate cyclase activator) and rolipram (specific PDE4 inhibitor), but not forskolin alone, dramatically increases PKA phosphorylation of NDE1 in transfected cells. This indicates that PDE4 normally prevents NDE1 phosphorylation. Bradshaw went on to show that a similar effect can be obtained by overexpressing full-length DISC1, and increased further using a mutant form of DISC1 lacking most of the PDE4 binding sites, suggesting that DISC1 regulates the phosphorylation of NDE1 by modulating the activity of PDE4. This is corroborated by the observation that DISC1 overexpression prevents the increase in PDE4 activity induced by high cAMP concentration.
Bradshaw and colleagues identified threonine-131 as a novel PKA phosphorylation site on NDE1 and used 3D homology modeling to predict that phosphorylation of T131 prevents dimerization of NDE1 and affects its binding to NDEL1 and LIS1. In accordance with such predictions, Bradshaw found that a phospho-mimic mutation in position 131 of NDE1 decreased its binding to LIS1 while increasing its interaction with NDEL1. By using an antibody that specifically detects NDE1 phosphorylated at T131, Bradshaw showed that phosphorylation of T131 promotes the positioning of NDE1 at the centrosome during mitosis and phosphorylated NDE1 is enriched at the dendritic spines and axon in primary neurons. Mutant NDE1 mimicking phosphorylation also negatively affects neurite outgrowth. In conclusion, Bradshaw proposed that the modulation of the phosphorylation of NDE1, and in turn its interaction with NDEL1 and LIS1, is an additional way by which DISC1 contributes to the regulation of neuronal proliferation, maturation and functioning.
To better characterize the dynamics of the interaction of DISC1 with itself and its binding partners, Bradshaw and Dinesh Soares (University of Edinburgh, UK) are currently using biochemical methods to purify recombinant DISC1 and test its oligomerization state. This work was presented in the poster by Soares, which illustrated that the middle portion of DISC1 (containing the self-association domain) can be expressed recombinantly, purified and cross-linked to study its quaternary structure. This technique will be used by Soares and Bradshaw to reveal the oligomerization state of full-length and truncated DISC1 in isolation from its binding partners. They will then analyze the structure of DISC1 co-purified with PDE4, NDE1 and NDEL1 to identify all the domain-domain interactions and to assess the effects of risk variants of DISC1 that are predicted to affect the interaction with this binding partners.
In the last talk of the session, Stephen Haggarty from Harvard University, Cambridge, Massachusetts, described the use of small molecules as probes to dissect the function of DISC1. Haggarty highlighted the limitations of the “serendipitous” drugs currently used to treat psychiatric conditions and emphasised the need to develop targeted, knowledge-based therapies. As a tool to better understand the function of DISC1 and to identify potential novel selective drugs, Haggarty and his team probed small molecule microarrays (SMMs) containing thousands of different compounds with lysates of HEK293 cells stably expressing one of five different variants of DISC1 (long, long variant, short, extra short and truncated). The binding of a particular chemical to a specific DISC1 isoform or its interactors was detected using a fluorescently labeled antibody. By applying this method, Haggarty and colleagues identified hundreds of compounds that “trapped” at least one DISC1 variant. Some of these compounds were able to bind to all DISC1 isoforms while others interacted with one specific isoform only. Haggarty plans to extend this search to identify molecules that display differential binding to other disease-associated DISC1 variants, such as the ones generated by non-synonymous SNPs.
To test the ability of the "small interactors of DISC1" (SMIDs) identified by SMMs to modulate the function of DISC1, Haggarty and colleagues utilised a Wnt signaling reporter assay in immortalized mouse hippocampal progenitors. They found that several SMIDs are indeed able to modulate Wnt signalling, as detected by changes in TCF/LEF-mediated transcription. Haggarty’s team is currently refining the search for GSK3β-inhibiting SMIDs by performing high-throughput TCF/LEF reporter assays in patient-derived neural progenitors obtained from induced pluripotent cells (iPS cells). In the future, Haggarty plans to test the effect of SMIDs on human neurons differentiated from iPS cells. Haggarty hopes that these tools will enable us not only to dissect the function of different DISC1 isoforms, but also to develop better treatments for psychiatric disorders.—Elise Malavasi.