Abstracts

Rationale: The molecular events contributing to seizures in Tuberous Sclerosis Complex (TSC) patients are poorly understood. Using tissue resected for treatment of seizures, we previously identified four microRNAs (miRs-23a, 34a, 34b*, and 532-5p) overexpressed in epileptogenic tubers compared to patient-matched non-tuber tissue and used computational methods to predict transcripts targeted by two or more of the four microRNAs. We found the predicted target set was enriched for genes involved in synaptic signal transmission. In this work, we used multiplexed quantitative LC-MS/MS proteomic profiling of epileptogenic tuber tissue to test whether there was a decrease in protein abundance for predicted combinatorial target genes of the four overexpressed microRNAs. Methods: Tandem Mass Tags (TMT) labeling and MuDPIT LC-MS/MS analysis were used to measure relative protein abundances in tuber and adjacent tissue. We analyzed eight tissue samples in total: four epileptogenic tubers and four perituberal tissues from patients with TSC1 or TSC2 mutations. Tissues resected during epilepsy surgery were solubilized, digested, and labeled individually with TMT10plex reagents then pooled. Peptides were separated by 2D liquid chromatography then sequenced on an Orbitrap Fusion MS system. Peptide assignments of spectra were made using Proteome Discoverer, and those identified with high confidence (1% false discovery rate) were imported into GeneSpring for quantitative analysis. Statistical significance for changes in protein level between tuber and perituberal samples was determined using a two-sided student's t-test. Statistical significance for ontology enrichment was performed using DAVID (http://david.abcc.ncifcrf.gov/) and Fisher's exact test. Results: Quantifiable mass spectra were obtained for 5749 proteins. Comparing epileptogenic tubers and non-tuber tissues, 1527 proteins had a significant difference in expression (P ≤ 0.05 and ≥ 1.5-fold). Of these, 72 proteins were predicted as combinatorial targets of the four overexpressed microRNAs. Strikingly, 71% of predicted target proteins were decreased in epileptogenic tubers (P=0.0007), providing evidence of repression by the elevated microRNAs. Ontology analysis of the repressed proteins revealed enrichment for genes/proteins involved in synaptic or nerve impulse transmission, including known epilepsy risk genes TSC1, synapsin 2 and GABRB3 and GABBR2 receptor subunits. Conclusions: Using quantitative proteomics we have confirmed the repression of protein levels for a set of transcripts predicted to be combinatorial targets of several microRNAs overexpressed in epileptogenic tubers. The set of transcripts is enriched for genes involved in synaptic signal transmission, a number of which are known epilepsy risk genes. Furthermore, TSC1 protein (hamartin) levels were reduced in epileptogenic tubers regardless of the TSC genotype, suggesting that aberrant microRNA expression in TSC tubers may exacerbate the effect of TSC gene mutations on the mTOR pathway.