Simple repeat evolution includes dramatic primary sequence changes that conserve folding potential

https://doi.org/10.1016/j.bbrc.2007.01.200Get rights and content

Abstract

We previously demonstrated that many “weak-folding” simple repeats were replaced during evolution by alternative weak-folding repeats. This suggested repeat selection at the level of higher order structure potential. Here, we demonstrate similar phenomena for “strong-folding” simple repeats in non-coding DNA. The Rabgap1 gene’s 3′ UTR contained the self-complementary repeat (AT)n in Homo sapiens but, in Mus musculus, this site was occupied by the complementary repeats (GT)n and (AC)n. Similarly, primate Plag1 UTRs contained various (GT)n–(AC)n palindromes but in rodents, this site was occupied by (AT)n, preserving folding potential more than primary sequence. The Znf516, Senp1, Rock2, and other UTRs exhibited similar replacements. In the Bnc2 UTR, (AT)n was replaced by sequences that evolved with approximate symmetry about a central axis, a pattern difficult to explain without invoking selection to preserve secondary structure. These observations reflect a predictable evolutionary pattern for some common non-coding genomic sequences.

Section snippets

Materials and methods

Database searches. We searched non-redundant 5′ and 3′ UTR databases (version, 2.2.1, 4-13-01, 368,154 sequences (http://bighost.area.ba.cnr.it/BIG/UTR-Home/) for all H. sapiens UTRs containing (AT)n  7 [8]. We also searched H. sapiens genomic sequences designated as cDNAs at Ensembl[21]. Combining both databases, there were 235 total (AT)n  7-containing UTRs reported for H. sapiens. Repeat flanking sequences consisting of 100 bp were then used to search all available genomes including: Bos Taurus

Sequences with folding potential replace one another during evolution

The H. sapiens Rabgap1 3′ UTR contained an uninterrupted (AT)n repeat. In M. musculus, this was replaced by a slightly interrupted (GT)n–(AC)n sequence combination (Fig. 1). Again, the actual replacement could have happened in the human rather than the rodent lineage. By themselves, (GT)n and (AC)n are weak-folding STRs. However, they are complementary and in combination have substantial, predicted folding energy similar to that of (AT)n by itself (Fig. 1). The predicted mouse Rabgap1 stem has

Discussion

It is difficult to explain the reported replacements and variations in primary sequence without invoking selection to preserve secondary structure. Conservation of secondary structure potential has proven valuable for identifying both known and previously unsuspected fRNAs [4]. Regardless of folding parameter choices, or whether the precise structures form, the hypothesis that simple repeats were selected for their secondary structure potential accurately predicted simple repeat replacements,

Acknowledgments

This study was supported by an NIH RO1 Grant DK38955 and a Veteran Affairs Merit Review Award, Bacterial DNA in Prostate Disease.

References (25)

  • G. Pesole et al.

    UTRdb and UTRsite: specialized databases of sequences and functional elements of 5′ and 3′ untranslated regions of eukaryotic mRNAs. Update 2002

    Nucleic Acids Res.

    (2002)
  • M. Zuker

    Mfold web server for nucleic acid folding and hybridization prediction

    Nucleic Acids Res.

    (2003)
  • Cited by (0)

    View full text