New approaches to unveil the Transcriptional landscape of dopaminergic neurons

Recent advances in studying the mammalian transcriptome arised new questions about how genes are organized and what is the function of noncoding RNAs. Furthermore, the discovery of large amounts of polyA- transcripts and antisense transcription proved that a portion of the transcriptome has still to be characterized. The complex anatomo-functional organization of the brain has prevented a comprehensive analysis of the transcriptional landscape of this tissue. New techniques must be developed to approach neuronal heterogeneity. In this study we combined Laser Capture Microdissection (LCM) and nanoCAGE, based on Cap Analysis of Gene Expression (CAGE), to describe expressed genes and map their transcription start sites (TSS) in two specific populations, A9 and A10, of mouse mesencephalic dopaminergic cells. Although sharing common dopaminergic marker genes, these two populations are part of different midbrain anatomical structures, substantia nigra (SN) for A9 and ventral tegmental area (VTA) for A10, project to relatively distinct areas, participate to distinct ascending dopaminergic pathways, exhibit different electrophysiological properties and different susceptibility to neurodegeneration in Parkinson`s disease. Specific neurons were identified by the expression of Green Fluorescent Protein driven by a celltype specific promoter in transgenic mice. High-quality RNAs were purified from 1000-2500 cells collected by LCM. We adapted the CAGE technique to analyze limiting amounts of RNAs (nanoCAGE). We took advantage of the cap-switching properties of the reverse transcriptase to specifically tag the 5`end of transcripts with a sequence containing a class III restriction site for EcoP15I. By creating 32bp 5`tags, we considerably improved the TSS mapping rate on the genome. A semi-suppressive PCR strategy was used to prevent primer dimers formation. The use of random priming in the 1st strand synthesis allowed to capture poly(A)- RNAs. 5`tags were sequenced with Illumina-Solexa platform. Here we show that this new nanoCAGE technology ensures a true high-throughput coverage of the transcriptome of a small number of identified neurons and can be used as an effective mean for gene discovery in the noncoding RNAs, to uncover putative alternative promoters associated to variants of protein coding transcripts and to detect potentially regulatory antisense transcripts. A further experimental validation by 5`RACE (Rapid Amplification of cDNA Ends) and RT-PCR on few candidate genes, have confirmed the existence in vivo of alternative TSS in the case of key regulatory genes involved in specifying and maintaining the dopaminergic phenotype of these neurons such as α-synuclein (Snca), dopamine transporter (Dat), vescicular monoamine transporter 2 (Vmat2), catechol-O-methyltransferase (Comt). Furthermore the differential expression of an antisense transcript overlapping to the polyubiquitin (Ubc) gene was detected as potentially interesting candidate gene accounting for differences in the ubiquitin-proteasome system (UPS) function in the two neuron populations. The potential implications deriving from these newly discovered alternative promoters and transcripts are discussed, considering also the potential consequences for the corresponding protein isoforms.