| Abstract |
In the first part of this thesis, we constructed a high resolution, sequence-ready, physical map of the gene-rich regions q23.3-q25 and q26.2-qter of human chromosome 10. These regions comprise about 35 Mb, i.e. about ? of the length of the chromosome. For the map construction, we initially screened a grided human BAC library, using about 500 distinct sequence tagged sites (STSs) from the above regions as probes, divided in groups, and isolated 3350 distinct, chromosome 10 specific BACs. Our clones, together with those similarly isolated by our Sanger Centre collaborators, formed the human chromosome 10 specific library, consisting of ~18.000 BACs. Screening of this library with each one of the above STSs, led us to the detailed and highly confident correlation of each one of the BACs with a group of STSs. We used these data, in combination with those derived from the BAC fingerprinting analysis (Sanger Center UK, Washington University St. Louis US) for the construction of the preliminary contigs (groups of overlapping clones) mapped along the above chromosomal regions. For the extension and progressive merging of the contigs, we determined by chromosome walking, about 80 new STS-markers, derived from the end-sequences of the most distal clones of each contig. This task gradually resulted in the construction of a unique, high resolution physical map of 10q23.3-q25, with a length of approximately 28 Mb. Contig depth, as defined by the number of overlapping clones in every map position, confidently supports the relative topography of each genomic site. In our map, mapping of each position was supported by at least 8 overlapping, well-characterized, BACs. Similarly, in terms of strategy and confidence, the physical map of the subtelomeric region was constructed, consisting of two contigs, 4.6 Mb and 2 Mb in length. Our mapping data were exploited by the collaborating laboratory of automated DNA sequencing (Sanger Center, UK), initially for the construction of a minimal tiling path of about 300 BACs mapped with the highest confidency, and subsequently, for the complete sequencing of the above regions. The corresponding finished sequence, currently available freely through the network, has a total size of about 35 Mb, consists of 272 out of 816 chromosome 10 protein coding genes and is associated with 40 disease-related genetic loci. With the aim to identify disease-related genes, in the second part of this thesis, we exploited both physical mapping and sequencing data to construct in silico, the genic map of a 10q23.3 region, 3 Mb in length. Among others, this area is associated with i) the genetic locus of the autosomal dominant lateral temporal lobe epilepsy (ADLTE), and ii) the expression of the cytogenetic fragile site FRA10A. In collaboration with other labs, altogether forming the European Consortium of ADLTE, we characterized several ADLTE candidate genes. Among those, LGI1 was determined to be directly associated with the disease, since it was found mutated in patients of ADLTE families. LGI1 a gene of unknown function, belongs to a novel gene family consisting of four members (LGI1-4) dispersed in the genome. All LGI genes are active in the brain displaying a partially overlapping pattern of expression; the encoded proteins are characterized by a variable number of tandemly linked EPT repeats possibly involved in protein-protein interactions. It is reasonable to hypothesize that the characteristic genetic heterogeneity of ADLTE or other neurological diseases may be related with members of the LGI family. Finally, by in silico screening of the nucleotide sequence of chromosome 10 for (CGG)n repeats, we identified a novel gene, FRA10AC1, carrying a polymorphic (CGG)n repeats in its 5 UTR and showed its direct involvement with the manifestation of FRA10A fragile site. Indeed, FRA10A carriers display expansion of the (CGG)n repeat by approximately 200-fold. This expansion, as shown by a collaborating group, results in hypermethylation of the region and the consequent transcriptional silencing of the gene. FRA10AC1 transcripts are detected ubiquitously in human tissues, being more abundant in organs with highly transcriptional activity. The presence of FRA10AC1 alternative transcripts in gonads, suggested the existence of five protein isoforms displaying structural heterogeneity at their carboxy-terminus. FRA10AC1 protein is accumulated in the nucleoplasm and remains well-conserved in multi-cellular eukaryotes, implying participation of the FRA10AC1 in a fundamental cellular function. In order to investigate its biochemical role, using the yeast two-hybrid system, we showed that FRA10AC1 interacts with two proteins, SAP145 and DGCR14 and verified our finding by in vitro and cell culture assays. According to the recent literature, both interacting proteins are involved in the process of pre-mRNA splicing. Indeed, SAP145 is a subunit of the SF3b splicing factor participating in the formation of U2 snRNP, a component of the major spliseosome. Furthermore, DGCR14 has been repeatedly isolated by biochemical approaches, as a component of the same complex. Therefore, it is reasonable to believe that our data support a functional role of FRA10AC1 in the pre-mRNA splicing machinery or, its involvement in functionally linked procedures, i.e. synthesis, processing and nucleo-cytoplasmic transport of mRNA.
|
Collections
