Analyses using novel Markov models of substitution support a significant role for germline methylation in male biased evolution.

Matthew J Wakefield1, Gavin A Huttley2, Alexander Isaev, Andrew Butterfield, Edward Lang & Cath Lawrence, Centre for Bioinformation Science, The Australian National University;, Centre for Bioinformation Science, The Australian National University

Male biased evolution is the increased rate of mutation and substitution in the germ line of males. Various mechanisms have been proposed for this observed difference in rate including the increased number of cell divisions increasing the number of replication related mutations, oxidative stress, and methylation at CpG dinucleotides. Huttley et al (2000) demonstrated with log linear modelling that the Y chromosome shows a significant decrease in the number of CpG dinucleotides indicative of methylation contributing to male biased evolution.

Methylated cytosines in CpG dinucleotides are produced by post-synthetic enzymatic modification of cytosines. They are inherently mutable as spontaneous deamination of the methylated cytosine creates thymine residues. This deamination can occur on the plus strand (CpG &rarr TpG) or the minus strand (CpG &rarr CpA).

To further investigate the contribution of methylation in male biased evolution we have constructed a novel Markov model of dinucleotide substitution that include parameters for methylation and strand using the EVOLVE maximum likelihood modelling toolkit (See Huttley Poster). This model was used to analyse evolution of a 38.6kb region that has been transposed to the human Y from the human X chromosome since the divergence of chimpanzee and human (Bohossian 2000). Likelihood ratio tests revealed an enormously significant influence of the methylation parameter (P < 10-79), and a significant influence for the strand parameter (P < 0.011). A Y chromosome specific methylation parameter was also significant (P < 0.04).

Our analysis shows a strongly significant effect of methylation on the evolution of this sequence and a significant effect of the strand on which the methylation driven mutation occurs. The result supports the conjecture that differential methylation in the germline elevates male mutation rates.

These results clearly demonstrate the utility of the EVOLVE toolkit in the derivation and application of new models of sequence evolution to illuminate key problems in biology.

Huttley et al 2000. Mol. Biol. Evol. 17:929-37.

Bohossian et al 2000. Nature 406:622-5.