A model for the correlation of mutation rate with GC content and the origin of GC-rich isochores

Abstract

Based on the biochemical kinetics of DNA replication and mutagenesis, including misincorporation and correction, a model has been developed for studying the relationships among the mutation rate (u), the G+C content of the sequence (f), and the G+C proportion in the nucleotide precursor pool (N). Also a measure for the next-nucleotide effect, called the maximum capacity of the next-nucleotide effect (MC), has been proposed. Under the normal physiological conditions of mammalian germ cells, our results indicate: (1) the equilibrium G+C content in a sequence is approximately equal to the G+C proportion in the nucleotide precursor pool, i.e., f approximately N, which is independent of the next-nucleotide effect; (2) an inverted-V-shaped distribution of mutation rates with respect to G+C contents is predicted, when the next-nucleotide effect is week, i.e., MC approximately 1; (3) the distribution becomes flatter (i.e., inverted-U-shaped) as MC increases, but the peak at 50% GC is still observed when MC < 2; and (4) the peak disappears when MC > 2.8, that is, when the next-nucleotide effect becomes strong. Our results suggest that changes in the relative concentrations of nucleotide precursors can cause variations among genes both in mutation rate and in G+C content and that compositional isochores (DNA segments with a homogeneous G+C content) can arise in a genome due to differences in replication times of DNA segments.

Covariation in Frequencies of Substitution, Deletion, Transposition, and Recombination During Eutherian Evolution

“…and differences in the timing of replication in S-phase (Wolfe 1991;Gu and Li 1994) are potentially quite complex and cannot be resolved with this type of data. However, future examination of intranuclear localization, including proximity to matrix attachment sites, pericentromeric heterochromatin, the nuclear membrane, and sites of chiasmata during meiosis may show significant correlations with divergence.…”

Impact of replication timing on non-CpG and CpG substitution rates in mammalian genomes

“…Pioneering studies suggested that changes in nucleotide pools during replication could be responsible for mutation rate fluctuations (Wolfe et al 1989;Gu and Li 1994), because different genomic regions replicate at different times during the S phase (MacAlpine et al 2004;Woodfine et al 2004;Karnani et al 2007;Hiratani et al 2008). Studies of limited portions of the human genome confirmed that mutation rates increase in late-replicating regions (Watanabe et al 2002;Stamatoyannopoulos et al 2009).…”

Analysis of DNA sequences using methods of statistical physics

“…They found the existence of strong long-range correlations in noncoding regimes, and weaker long-range correlations in coding regimes in excellent agreement with [49]. More recently, they found that the correlations in coding DNA are restricted to the third degenerate nucleotide of the codon [67], which may be related to the presence of the isochores, i.e., long regions of high or low C+G content [68].…”

“…[64,65]). These patches may represent di erent structural elements of 3D chromosome organization, e.g., the DNA double helix with period 10.5 bp [66] nucleosomes about 200 bp long, 30 nm ÿber, looped domains of about 10 5 bp, and chromatin bands [67,68] or isochores that may consist of several million nucleotides. Such hierarchical structure of several length scales may produce e ective long-range power-law correlations.…”

“…The longest correlations, on the length scales of isochores may originate due to base-substitution mutations during replication (see Ref. [68]). Indeed, it is known that di erent parts of chromosomes replicate at di erent stages of cell division.…”