These results support the findings of Greene et. al. that this region contains the minimal FXN promoter. In the HeLa cell line, the next longest insert containing 4,923 bp of DNA but excluding any of the identified conserved regions, produced a similar level of expression. All of the other constructs, which contained an insert longer than 5,577 bp, produced a statistically significant increase in expression compared to that with the shortest insert. The greatest level of induction was seen with the construct that contained the 5,577 bp insert, which includes only conserved region 1. This resultAMN107indicates that the presence of conserved region 1 plays a role in inducing the expression of the FXN gene, and which was only modestly modified in the presence of the other conserved regions. In the BE -M17 cell line, a greater variation was observed in the effects on FXN gene expression depending on the conserved regions present in each construct. Increased or reduced expression was observed depending on the length of the insert and presumably the presence of particular conserved regions. This could possibly be attributed to interactions between positive and negative regulatory roles of the different conserved regions in this particular cell line. Nonetheless, in accordance with the findings in the HeLa cell line, the construct that containedNutlin-3 Mdm2 inhibitor the 5,577 bp insert, which includes only conserved region 1, led to increased expression in the BE -M17 cell line, albeit to a lesser extent than in HeLa cells. As these results indicated that conserved region 1 harbors an important regulatory element that influences the expression of the FXN gene, subsequent experiments focused on refining the specific sequence within this region responsible for the observed effects. The rapid advances that have resulted from the sequencing of the human genome and those of other species, and the development of bioinformatics programs and matrices have facilitated the study of gene regulation. Prior to this study, information about the regulation of the normal human FXN gene was limited to the regions immediately flanking the first exon of the gene. There was no data on the position of any long-range, cis-acting regulatory sequences or of known transcription factors that may control FXN gene expression. In this study, in silico approaches were used in conjunction with different reporter systems to identify and better understand potential long distance cis-acting regulatory elements controlling the expression of the human FXN gene. The search of regulatory elements in this study was restricted to the intergenic region upstream of exon 1 of the FXN gene to the PRKACG gene. A detailed analysis of this 21.3 kb region revealed the presence of eight highly conserved non-coding regions.