No abstract

Sequences from ten species of Mus (Rodentia, Muridae) of the melanocortin-1 receptor (Mc1r) gene (945 bp), which plays a key role in coat color determination, were compared with an existing Mc1r dataset (ca. 498 bp) from 12 species of Mustela and Martes (Carnivora, Mustelidae). The dN/dS ratio (ω) was estimated at 0.19 for Mus and 0.35 for the mustelids, using a likelihood-based one-ratio model with empirical codon frequencies. Running the model with equal codon frequencies gave a dramatic increase in ω for the mustelids (1.02) but not for Mus (0.31), indicating stronger codon usage bias in Mc1r among mustelids. When ω was estimated with the free-ratio model, significantly accelerated rates of amino acid replacement (nearly 1 in ω) were seen in several regions of the Mus phylogeny, such as in the ancestral subgeneric lineages, possibly associated with ecological niche shifting. Our results suggest that both functional constraints on coat color variation and selective constraints on codon usage bias have participated in structuring Mc1r gene sequences. Furthermore, they suggest that these contrasting influences have acted differentially in Mus and the mustelid lineages, and also differentially during the course of evolution within the genus Mus.

Sequence variation within the 5' flanking (about 240 bp) and exon regions (426 bp) of the melanocortin-1 receptor (MC1R) gene was examined to determine the potential role of this protein in the melanistic coat coloration of 17 mustelid species in four genera: Gulo (wolverines), Martes (martens), Mustela (weasels), and Meles (badgers). Members of the genera Mustela and Meles, together with Martes flavigula and Martes pennanti, were shown to have intact gene sequences. However, several "in frame" deletions of the MC1R gene region implicated in melanism of other species were detected within members of the genera Martes and Gulo. For instance, Gulo gulo possessed a 15 bp deletion in the second transmembrane domain coding region, while Martes americana, Martes melampus, Martes zibellina, and Martes martes shared a 45 bp deletion overlapping this area. In addition, Martes foina was found to possess a 10 bp insertion followed closely by a 28 bp deletion immediately downstream of the deletion found in other martens. Notably, none of these indels was associated with a melanistic phenotype. Phylogenetic analysis revealed that each of these nonrandomly distributed deletions arose independently during the evolution of this family. Specific indel-neighboring motifs appear to largely account for the biased and repeated occurrence of deletion events in the Martes/Gulo clade.

Mus lepidoides of central Burma (Myanmar) was described 75 years ago but has since been dismissed as a regional variant of the Indian field mouse, M. booduga. DNA sequences of multiple mitochondrial and nuclear genes from recently collected specimens, combined with a fresh morphological reassessment, reaffirm the distinctiveness of M. lepidoides from M. booduga and from all other species of Mus. Mus lepidoides is so distinct in fact that it warrants placement in its own Species Group within subgenus Mus. Molecular and morphological assessments of phylogenetic affinities converge on the exciting possibility that M. lepidoides represents the previously elusive sibling taxon to the Mus musculus Species Group. If confirmed, this relationship would provide the previously missing connection between the main radiation of subgenus Mus in Southeast and South Asia, and the radiation of the M. musculus Species Group in western Asia and Europe. We speculate that a common ancestor of M. lepidoides and the M. musculus Species Group occupied a continuous but episodic tract of xeric habitat that linked central Burma with northern India at various times during the late Pliocene and Quaternary. Further molecular and cytogenetic studies on the phylogenetic position of M. lepidoides clearly represent a high priority in mouse research.

No abstract