A New Tile in the Biochemical Puzzle of Insulin Biology

Background: Alternative gene transcript splicing permits a single gene to produce multiple proteins with varied functions. Bioinformatic investigations have identified numerous splice variants, but whether these transcripts are translated to functional proteins and the physiological significance of these alternative proteins are largely unknown. Through direct identification of splice variants associated with disease states, we can begin to address these questions and to elucidate their roles in disease predisposition and pathophysiology. This work specifically sought to identify disease-associated alternative splicing patterns in ion channel genes by comprehensively screening affected brain tissue collected from patients with mesial temporal lobe epilepsy and Alzheimer's disease. New technology permitting the screening of alternative splice variants in microarray format was employed. Real time quantitative PCR was used to verify observed splice variant patterns. Results: This work shows for the first time that two common neurological conditions are associated with extensive changes in gene splicing, with 25% and 12% of the genes considered having significant changes in splicing patterns associated with mesial temporal lobe epilepsy and Alzheimer's disease, respectively. Furthermore, these changes were found to exhibit unique and consistent patterns within the disease groups. Conclusion: This work has identified a set of disease-associated, alternatively spliced gene products that represent high priorities for detailed functional investigations into how these changes impact the pathophysiology of mesial temporal lobe epilepsy and Alzheimer's disease. Published: 7 March 2007 Genome Biology 2007, 8:R32 (doi:10.1186/gb-2007-8-3-r32) Received: 6 November 2006 Revised: 16 February 2007 Accepted: 7 March 2007 The electronic version of this article is the complete one and can be found online at http://genomebiology.com/2007/8/3/R32 Genome Biology 2007, 8:R32 R32.2 Genome Biology 2007, Volume 8, Issue 3, Article R32 Heinzen et al. http://genomebiology.com/2007/8/3/R32 Background The complexity of the genome lies not only in the many genes comprising it, but also in the many levels of processing that influence the proteins that are produced and their abundance. One key site of regulation is the splicing of precursor RNAs to their associated mRNA transcripts. This process alone allows a single gene to have multiple different mRNA transcripts, producing proteins that may differ substantially from one another, even to the extent of having opposing effects [1]. Overall, however, little is known about the functional differences amongst the alternative proteins produced from the same gene. Because the functional characterization of proteins can be laborious, it would be useful to be able to prioritize alternative transcripts more likely to have biological significance. One direction for prioritization is on the basis of association with human disease. Alternative splicing of key genes generates alternative proteins that contribute to several prominent human diseases, for example, the spinal motor neuron protein in spinal muscular atrophy [2], cardiac troponin T, insulin receptor, myotubularin-related 1, and other proteins in myotonic dystrophies [3-5], and the tau protein in frontotemporal dementia and Alzheimer's disease [3,4] (other examples are reviewed extensively in [5]). Furthermore, alternative splicing of a sodium channel gene, SCN1A, has also recently been associated with altered response to antiepileptic medications [6]. There are potentially many more undetected examples of splicing alterations associated with disease pathophysiology and drug response variation in humans. Studies of alternative splicing have usually been restricted to a single gene or small gene family. To date, there are only a few reports of splice variation screens in human disease and none has been reported for any central nervous system disease. Recently, new technology has become available that allows for the comprehensive investigation of alternative splicing through the use of splice variant microarrays. This technology uses probes in a microarray format and screens for unique exon-exon junctions specific to a particular splicing event [7-10]. Here we applied this systematic approach to assess the relationship between alternative splicing and two common and important neurological conditions, with the aim of identifying alternative splicing patterns of potential relevance to human disease. Mesial temporal lobe epilepsy (mTLE) and Alzheimer's disease (AD) are highly complex neurological diseases characterized by aberrant neuronal excitation and neurodegeneration. While the pathological processes differ substantially, both diseases exhibit pathophysiology linked to ion channel activity. Seizure activity characteristic of epilepsy is the result of a dysregulation of inhibitory and excitatory neuronal signaling largely controlled by ion channel activity [11]. Likewise, abnormal ion channel function also has been associated extensively with AD. AD-related neurodegeneration is believed to be, in part, caused by the overactivation of N-methyl-D-aspartate receptor activation and subsequent increases in intracellular calcium, oxidative stress, and neurodegeneration [12]. Other ion channels, including glutamate receptors, nicotinic cholinergic receptors, and calcium and potassium channels, also have been implicated in AD pathophysiology [13-18]. Little information exists regarding the impact of splicing variation of ion channel genes on mTLE and AD. Our work sought to comprehensively evaluate ion channel splice variation in these two neurological diseases using a microarray format (ExonHit Therapeutics). We evaluated 1,665 known and potential splice events across 287 ion channel genes in human brain tissue samples collected from patients with AD and mTLE. In addition to identifying disease-associated splicing variation, a secondary aim of this work was to assess the reliability of the array-based identification of splicing changes through the use of real time PCR (rtPCR) to validate associations detected using the highthroughput platform. Results Mesial temporal lobe epilepsy Following our initial screen of 1,665 possible alternative splicing events, a total of 221 splicing changes were identified as statistically significantly changed in mTLE samples, with p < 0.05, when comparing splice variant ratio (SVR) values calculated using equation 1 (see Materials and methods). Selected statistically significant events representing a range of p values were chosen for rtPCR confirmation. Of 13 splice array-identified alternative splicing events with an associated p value of less than 0.05, 9 were verified using rtPCR in a larger sample size. Evaluation of discrete groups of p value ranges revealed increased success rates with lower p values (0.02 0.05) were confirmed not to be changed using rtPCR. The compiled list of rtPCR confirmed mTLE-associated alternative splicing events are included in Table 2. Our splice array studies revealed an mTLE-associated splicing change in CACNA1B (p = 0.017, variant GenBank: M94173). This particular event was randomly selected for rtPCR confirmation, and we observed a change opposite that detected with the splice array. This likely occurred due to the presence of unknown splicing events that were being detected either by the splice array probes, or possibly by the rtPCR assay probes. Due to the uncertainty linked to this event, we deemed this splicing change in our studies as an event that Genome Biology 2007, 8:R32 http://genomebiology.com/2007/8/3/R32 Genome Biology 2007, Volume 8, Issue 3, Article R32 Heinzen et al. R32.3