Dissecting the major African snake radiation: a molecular phylogeny of the Lamprophiidae Fitzinger (Serpentes, Caenophidia)

Abstract

The Elapoidea includes the Elapidae and a large (~60 genera, 280 sp.) and mostly African (including Madagascar) radiation termed Lamprophiidae by Vidal et al. (2007), that includes at least four major groups: the psammophiines, atractaspidines, lamprophiines and pseudoxyrhophiines. In this work, we reviewed the recent taxonomic history of the lamprophiids, and built a data set including two nuclear protein-coding genes (c-mos and RAG2), two mitochondrial rRNA genes (12S and 16S rRNA) and two mitochondrial protein-coding genes (cytochrome b and ND4) for 85 species belonging to 45 genera (thus representing about 75% of the generic diversity and 30% of the specific diversity of the radiation), in order to clarify the phylogenetic relationships of this large and neglected group at the subfamilial and generic levels. To this aim, 480 new sequences were produced. The vast majority of the investigated genera fall into four main monophyletic clusters, that correspond to the four subfamilies mentioned above, although the content of atractaspidines, lamprophiines and pseudoxyrhophiines is revised. We confirm the polyphyly of the genus Stenophis, and the relegation of the genus name Dromophis to the synonymy of the genus name Psammophis. Gonionotophis brussauxi is nested within Mehelya. The genus Lamprophis Fitzinger, 1843 is paraphyletic with respect to Lycodonomorphus Fitzinger, 1843. Lamprophis swazicus is the sister-group to Hormonotus modestus, and may warrant generic recognition. Molecular data do not support the traditional placement of Micrelaps within the Atractaspidinae, but its phylogenetic position, along with that of Oxyrhabdium (previously considered to belong to the Xenodermatidae), requires additional molecular data and they are both treated as Elapoidea incertae sedis. The interrelationships of Psammophiinae, Atractaspidinae, Lamprophiinae, Pseudoxyrhophiinae, Prosymna (13 sp.), Pseudaspis (1 sp.) and Pythonodipsas (1 sp.), Buhoma (2 species), and Psammodynastes (1 sp.) remain unresolved. Finally, the genus Lycognathophis, endemic to the Seychelles, does not belong to the African radiation, but to the Natricidae.

Molecular Evidence for the Nonmonophyly of the Asian Natricid GenusXenochrophis(Serpentes, Colubroidea) as Inferred from Mitochondrial and Nuclear Genes

“…The constituents of new subfamily Carphophiinae (Natriciteres + Rhabdophis + Xenochrophis), as proposed by Zaher et al (2009), could not be well supported by our analyses. The placement of Lycognathophis seychellensis in Natricinae (Vidal et al, 2008) was confirmed by our c-mos data with strong nodal support. Interestingly, our c-mos results indicates Lycognathophis to be related closely to the African natricids.…”

Phylogenetic relationships among the Stiletto Snakes (genusAtractaspis) based on external morphology

“…Heise et al (1995) and Pinou et al (2004) found Atractaspis to be most closely related to the Elapidae (but did not include any other atractaspidid genera in their analysis). Vidal et al (2007Vidal et al ( , 2008 placed Atractaspis (along with the genera Lamprophis, Pythonodipsas, Mehelya, Psammophylax, Pseudaspis, Buhoma, Psammodynastes, Prosymna and Leioheterodon) within the newly erected family Elapoidea, the sister group to Elapidae and the new family Lamprophiidae (corresponding closely to the Boodontinae of Lawson et al 2005). Kelly et al (2008) found that Atractaspis and several other atractaspidid genera represented a basal African radiation of colubroids, closely associated with Buhoma, Psammodynastes, and the Lamprophiidae (e.g.…”

“…However, the only molecular phylogenetic studies to include a significant representation of atractaspidid species (Nagy et al 2005;Vidal et al 2008) found Homoroselaps lacteus to be the sister group to Atractaspis, with Macrelaps more distantly related. Homoroselaps is a problematic genus, considered on the basis of morphology to be most closely affiliated with the Elapidae by some researchers (e.g.…”

“…Those molecular phylogenies in which it has been included find it to be either 'unplaced' (Kelly et al 2003) or more closely affiliated with the Atractaspididae than with the Elapidae (e.g. Lawson et al 2005;Vidal et al 2008;Kelly et al 2009;Zaher et al 2009). However, these studies included very few atractaspidids in their analyses.…”

“…In the absence of many other atractaspidid species for comparison, evidence that Homoroselaps is affiliated more closely with Atractaspis than with the Elapidae does not necessarily indicate that it is the sister group of the former. However, Nagy et al (2005) and Vidal et al (2008) each included six atractaspidid genera, including Macrelaps, in their analysis which found Homoroselaps to be the sister group to Atractaspis. We therefore decided to perform two analyses of relationships within Atractaspis for comparison using two different outgroups: Macrelaps microlepidotus and Homoroselaps lacteus.…”