Phylogenetic signal analysis in the basicranium of Ursidae (Carnivora, Mammalia)

Multiple modes of inference reveal less phylogenetic signal in marsupial basicranial shape compared with the rest of the cranium

Incorporating morphological data into modern phylogenies allows integration of fossil evidence, facilitating divergence dating and macroevolutionary inferences. Improvements in the phylogenetic utility of morphological data have been sought via Procrustes-based geometric morphometrics (GMM), but with mixed success and little clarity over what anatomical areas are most suitable. Here, we assess GMM-based phylogenetic reconstructions in a heavily sampled source of discrete characters for mammalian phylogenetics-the basicranium-in 57 species of marsupial mammals, compared with the remainder of the cranium. We show less phylogenetic signal in the basicranium compared with a 'Rest of Cranium' partition, using diverse metrics of phylogenetic signal (Kmult, phylogenetically aligned principal components analysis, comparisons of UPGMA/neighbour-joining/parsimony trees and cophenetic distances to a reference phylogeny) for scaled, Procrustes-aligned landmarks and allometry-corrected residuals. Surprisingly, a similar pattern emerged from parsimony-based analyses of discrete cranial characters. The consistent results across methods suggest that easily computed metrics such as Kmult can provide good guidance on phylogenetic information in a landmarking configuration. In addition, GMM data may be less informative for intricate but conservative anatomical regions such as the basicranium, while better-but not necessarily novel-phylogenetic information can be expected for broadly characterized shapes such as entire bones. This article is part of the theme issue 'The mammalian skull: development, structure and function'.

Absence of some cytochrome P450 (CYP) and hydroxysteroid dehydrogenase (HSD) enzymes in hagfishes

Corticosteroids are synthesized from cholesterol by steroidogenic enzyme catalysts belonging to two main families: the cytochrome p450s (CYPs) and hydroxysteroid dehydrogenases (HSDs). The action of these steroidogenic enzymes allows the genesis of the terminal active corticosteroids 11-deoxycortisol (S), 1ɑ-hydroxycorticosterone (1α-OH-B), or cortisol in different fish species. However, for Cyclostomes like hagfishes, the terminal corticosteroid is still undefined. In this study, we examined the presence or absence of CYPs and HSDs as traits in fishes to gain insight about the primary corticosteroid synthesis pathways of the hagfishes. We used published cytochrome c oxidase I (COXI) amino acid sequences to construct a phylogeny of fishes and then mapped the CYPs and HSDs as morphological traits onto the tree to predict the ancestral character states through ancestral character reconstruction (ACR). There is a clear phylogenetic signal for CYP (i.e., CYP11a1, 17, 21, and 11b) and HSD (i.e., 11-βHSD and 3β-HSD) derivatives of interest throughout the more derived fishes. Using trait-based ACR, we also found that hagfishes possess genes for 3β-HSD, CYP11a1, CYP17, and CYP21. Importantly, the presence of CYP21 implies that hagfish can synthesize 11-deoxycorticosterone (11-DOC) and S. Previous research demonstrated that despite hagfish having CYP21, neither 11-DOC nor S could be detected in hagfish. This discrepancy between the presence of steroidogenic enzymes and products brings into question the expression and/or function of CYP21 in hagfishes.