Genomic imprinting is disrupted in interspecific Peromyscus hybrids.

Genomic imprinting, the unequal expression of gene alleles on the basis of parent of origin, is a major exception to mendelian laws of inheritance. By maintaining one allele of a gene in a silent state, imprinted genes discard the advantages of diploidy, and for this reason the rationale for the evolution of imprinting has been debated. One explanation is the parent-offspring conflict model, which proposes that imprinting arose in polyandrous mammals as the result of a parental conflict over the allocation of maternal resources to embryos. This theory predicts that there should be no selection for imprinting in a monogamous species. Crosses between the monogamous rodent species Peromyscus polionotus and the polyandrous Peromyscus maniculatus yield progeny with parent-of-origin growth defects that could be explained if imprinting was absent in the monogamous species. We find, however, that imprinting is maintained in P. polionotus, but there is widespread disruption of imprinting in the hybrids. We suggest that the signals governing genomic imprinting are rapidly evolving and that disruptions in the process may contribute to mammalian speciation.

Genetic and epigenetic incompatibilities underlie hybrid dysgenesis in Peromyscus.

Crosses between the two North American rodent species Peromyscus polionotus (PO) and Peromyscus maniculatus (BW) yield parent-of-origin effects on both embryonic and placental growth. The two species are approximately the same size, but a female BW crossed with a male PO produces offspring that are smaller than either parent. In the reciprocal cross, the offspring are oversized and typically die before birth. Rare survivors are exclusively female, consistent with Haldane's rule, which states that in instances of hybrid sterility or inviability, the heterogametic sex tends to be more severely affected. To understand these sex- and parent-of-origin-specific patterns of overgrowth, we analysed reciprocal backcrosses. Our studies reveal that hybrid inviability is partially due to a maternally expressed X-linked PO locus and an imprinted paternally expressed autosomal BW locus. In addition, the hybrids display skewing of X-chromosome inactivation in favour of the expression of the BW X chromosome. The most severe overgrowth is accompanied by widespread relaxation of imprinting of mostly paternally expressed genes. Both genetic and epigenetic mechanisms underlie hybrid inviability in Peromyscus and hence have a role in the establishment and maintenance of reproductive isolation barriers in mammals.

Molecular evolution and phylogenetic utility of the polyubiquitin locus in mammals and higher vertebrates.

The product of ubiquitin genes is a small protein involved in intracellular sorting of other proteins. The locus consists of tandemly arrayed, uninterrupted copies of the gene. As several studies have noted, the Polyubiquitin locus is a model system for studying concerted evolution. While the protein is among the most conserved known, individual copies within an organism show variation in nucleotide sequence despite clear evidence of concerted evolution. When treated as individuals, repeats from a given locus form a monophyletic group. Furthermore adjacent copies often cluster, suggestive of the mechanism of concerted evolution. Due to this concerted evolution of repeats (and loci in organisms with multiple polyubiquitins), sequencing of heterogeneous PCR products consisting of all the repeats in a given organism may yield phylogenetic signal, as with other multicopy genes. We test this possibility through 22 original sequences using primers designed so that only tandem copies are amplified. Using these and previously published data, we further explore these phenomena in higher vertebrates and mammals in particular. We suggest that multiple locus duplications have occurred within mammals. Positional codon bias is strongly evident. We also find substitutional bias with regard to codon type. GC content of the locus appears to be generally high across vertebrates. Intraorganismal variation is tallied as an indication of frequency of change in codon position and transition/transversion ratios to further elucidate the tempo and mode of molecular evolution. Using these data, a weighting scheme for ubiquitin is also presented. Despite the gene's high GC content, transitional changes still appear more frequent. While the phylogenetic utility of ubiquitin does not appear great, its mechanistic insights seem far from exhausted.

Allelic expression of IGF2 in marsupials and birds.

Genomic imprinting, the parent-of-origin- specific expression of genes, has been observed in a variety of eutherian mammals. One gene that has been shown to be imprinted in all eutherians examined is the IGF2 gene. This gene encodes a potent fetal-specific growth factor that is expressed almost exclusively from the paternal chromosome. Several other imprinted genes in the IGF2 pathway are imprinted as well, suggesting that IGF2 is a focal point for the selective pressure leading to imprinted gene expression. This observation is in keeping with a proposal that imprinting arose as the result of a genetic conflict between parents over the allocation of maternal resources to the embryo. One prediction of this model is that imprinting exists in species in which there is at least some contribution of maternal resources to the embryo, and in which polyandry is observed. To test this prediction the allelic expression of the IGF2 gene was examined in two noneutherian species. The IGF2 gene was shown to be expressed in a paternal-specific manner identical to that in eutherians in Monodelphis domestica, a placental South American opossum. In contrast, the IGF2 gene is biallelic in expression in chickens, which are oviparous, and make no postfertilization contribution of maternal resources to the offspring.

Higher level relationships of the arctoid Carnivora based on sequence data and "total evidence".

The relationships of the lesser or red panda, Ailurus, have remained elusive even as any doubts about the identity of the giant panda as a bear have been erased. While usually classified as a member of the Procyonidae (raccoons), recent anatomical studies have suggested that the red panda may not fall in any of the arctoid carnivore families but instead may reflect an early offshoot of the lineage leading to ursids (bears) and pinnipeds (seals, sea lions, and walruses). Sequence data from the cytochrome b and 12S genes for multiple representatives of all relevant families support this hypothesis. Such a systematic position makes this threatened species particularly worthy of conservation. Sequence data alone, as well as a combined analysis of the sequence and anatomical data, strongly support a single origin of pinnipeds and their aquatic adaptations, lending some resolution to the general disagreement about familial relationships in this group. These molecular data also support canids as the basal members of this caniform clade, but are unresolved with respect to whether mustelids or procyonids constitute the sister group to the (ursid, pinniped, Ailurus) clade. There is support for the notion that skunks are a genetically divergent and possibly nonmustelid lineage.

Genomic imprinting of a placental lactogen gene in Peromyscus.

The mammalian genome contains over 30 genes whose expression is dependent upon their parent-of-origin. Of these imprinted genes the majority are involved in regulating the rate of fetal growth. In this report we show that in the deer mouse Peromyscusthe placental lactogen-1-variant ( pPl1-v) gene is paternally expressed throughout fetal development, whereas the linked and closely related pPl1gene is expressed in a biallelic manner. Neither the more distantly related pPl2Agene, nor the Mus Pl1gene displays any preferential expression of the paternal allele, suggesting that the acquisition of imprinting of pPl1-v is a relatively recent event in evolution. Although pPl1 expression is temporally mis-regulated in the dysplastic placentae of hybrids between two Peromyscus species, its over-expression cannot account for the aberrant phenotypes of these placentae. We argue that the species-specific imprinting of pPl1-v, encoding a growth factor that regulates nutrient transfer from mothers to their offspring, is consistent with the parent-offspring conflict model that has been proposed to explain the evolution of genomic imprinting.