Exceptionally long 5' UTR short tandem repeats specifically linked to primates.

We have previously reported genome-scale short tandem repeats (STRs) in the core promoter interval (i.e. -120 to +1 to the transcription start site) of protein-coding genes that have evolved identically in primates vs. non-primates. Those STRs may function as evolutionary switch codes for primate speciation. In the current study, we used the Ensembl database to analyze the 5' untranslated region (5' UTR) between +1 and +60 of the transcription start site of the entire human protein-coding genes annotated in the GeneCards database, in order to identify "exceptionally long" STRs (≥5-repeats), which may be of selective/adaptive advantage. The importance of this critical interval is its function as core promoter, and its effect on transcription and translation. In order to minimize ascertainment bias, we analyzed the evolutionary status of the human 5' UTR STRs of ≥5-repeats in several species encompassing six major orders and superorders across mammals, including primates, rodents, Scandentia, Laurasiatheria, Afrotheria, and Xenarthra. We introduce primate-specific STRs, and STRs which have expanded from mouse to primates. Identical co-occurrence of the identified STRs of rare average frequency between 0.006 and 0.0001 in primates supports a role for those motifs in processes that diverged primates from other mammals, such as neuronal differentiation (e.g. APOD and FGF4), and craniofacial development (e.g. FILIP1L). A number of the identified STRs of ≥5-repeats may be human-specific (e.g. ZMYM3 and DAZAP1). Future work is warranted to examine the importance of the listed genes in primate/human evolution, development, and disease.

Polymorphic core promoter GA-repeats alter gene expression of the early embryonic developmental genes.

Protein complexes that bind to 'GAGA' DNA elements are necessary to replace nucleosomes to create a local chromatin environment that facilitates a variety of site-specific regulatory responses. Three to four elements are required for the disruption of a preassembled nucleosome. We have previously identified human protein-coding gene core promoters that are composed of exceptionally long GA-repeats. The functional implication of those GA-repeats is beginning to emerge in the core promoter of the human SOX5 gene, which is involved in multiple developmental processes. In the current study, we analyze the functional implication of GA-repeats in the core promoter of two additional genes, MECOM and GABRA3, whose expression is largely limited to embryogenesis. We report a significant difference in gene expression as a result of different alleles across those core promoters in the HEK-293 cell line. Across-species homology check for the GABRA3 GA-repeats revealed that those repeats are evolutionary conserved in mouse and primates (p<1 × 10(-8)). The MECOM core promoter GA-repeats are also conserved in numerous species, of which human has the longest repeat and complexity. We propose a novel role for GA-repeat core promoters to regulate gene expression in the genes involved in development and evolution.

Evolutionary trend of exceptionally long human core promoter short tandem repeats.

Short tandem repeats (STRs) are variable elements that play a significant role in genome evolution by creating and maintaining quantitative genetic variation. Because of their proximity to the +1 transcription start site (TSS) and polymorphic nature, core promoter STRs may be considered a novel source of variation across species. In a genome-scale analysis of the entire human protein-coding genes annotated in the GeneCards database (19,927), we analyze the prevalence and repeat numbers of different classes of core promoter STRs in the interval between -120 and +1 to the TSS. We also analyze the evolutionary trend of exceptionally long core promoter STRs of ≥6-repeats. 133 genes (~2%) had core promoter STRs of ≥6-repeats. In the majority of those genes, the STR motifs were found to be conserved across evolution. Di-nucleotide repeats had the highest representation in the human core promoter long STRs (72 genes). Tri- (52 genes), tetra-, penta-, and hexa-nucleotide STRs (9 genes) were also present in the descending prevalence. The majority of those genes (84 genes) revealed directional expansion of core promoter STRs from mouse to human. However, in a number of genes, the difference in average allele size across species was sufficiently small that there might be a constraint on the evolution of average allele size. Random drift of STRs from mouse to human was also observed in a minority of genes. Future work on the genes listed in the current study may further our knowledge into the potential importance of core promoter STRs in human evolution.

Core promoter STRs: novel mechanism for inter-individual variation in gene expression in humans.

In a genome-scale analysis of the composition of core promoter sequences, we have recently shown that approximately 25% of the human protein-coding genes have at least one short tandem repeat (STR) of 3-repeats in their core promoters (i.e. the interval between -120 to +1). Through their nucleosome processing effect, GA-repeats play a crucial role in the regulation of gene transcription. In this study, we chose the human SRY (sex determining region Y)-box 5 (SOX5) gene as a prototype of the GA-rich core promoters to investigate the role of core promoter GA-STRs in gene expression. The human SOX5 gene is indispensable for diverse embryonic developmental processes, ranging from oligodendrocyte development and corticogenesis to chondrogenesis, and regulation of the cell cycle. Whereas the absolute ratio of 99% of the genes range between 0.2 and 2, the composition of the core promoter of the two most ubiquitously expressed mRNAs of the human SOX5 gene (transcripts ID: ENST00000451604 and ENST00000309359) is exceptionally rich in purine nucleotides (purine/pyrimidine ratio: 61.5). Indeed, this core promoter is an island of four tandem GA-STRs, and lacks the known TATA and TATA-less elements for gene transcription. Evolutionary conservation of this region between human and mouse (75% homology) supports important functional role for this promoter. In this study, we show that this nucleotide composition is indeed a potent promoter (p<1×10(-10)), and different haplotypes across the region result in significant difference in gene expression (p<1×10(-6)). To our knowledge, this is the first report of functional STRs in a human gene core promoter. Based on our search on the core promoters of the entire human protein-coding genes annotated in the GeneCards database (19,927genes) for the presence of pure GA-STRs, 429 genes contain at least one GA(3)-repeat in their core promoter. Core promoters with pure GA-STRs of GA(4) and above were observed in 61 genes. Our data unravel a novel mechanism for inter-individual variation in gene expression and complex traits/phenotypes through core promoter GA-STRs.