Evolutionary analysis of genes involved in cement gland development in anurans

The cement gland (CG) is a transient organ, found in most anuran embryos and early larvae and located normally on the front of the head. Its sticky secretion allows newly hatched larvae to attach to the egg jelly or to another support later and remain hidden and stationary until feeding starts. Our ultrastructural studies showed that prominence structure of the CG in some species exists, but is lacking in some others. Previous work has shown that a large number of genes have a role in CG development in Xenopus laevis. The aim of the present study is to find out whether the loss of cement gland formation for those species studied here occurs because of missing genes or for other reasons. In order to test whether some of these genes are present in other anuran species, especially in those where the CG does not form, genomic DNAs were examined for sequence similarity by low stringency hybridization. Sequences from three different genes with a role in controlling CG development in Xenopus (otx-2, xcg-1 and xag-1) were individually hybridised with genomic DNA of four species of anurans (X. laevies, Leptodactylus fuscus, Phyllomedusa trinitatis and Physalaemus pustulosus) and one species of rodent (Muss musculus domesticus strain C57Bl/ 6). The results showed that Xenopus probes can detect the presence of potential homologues of all three genes in the different species. For the two genes most specifically involved in CG development, xcg-1 and xag-1, both are clearly present, even in the two species which lack CG development, though in one of these, P. trinitatis, xag-1 shows considerable difference from the other species. At this stage, we can conclude that the missing cement gland for those species studied here is not due to a lack of the genes responsible for the gland development. In order to carry this work further, in situ hybridisation should be used to determine the actual expression patterns of these genes.