Genetic population structure of spiny lobster Palinurus delagoae in the south-western Indian Ocean, and the evolutionary history of Palinurus

Gopal, Keshni (2007-03)

Thesis (MSc)--University of Stellenbosch, 2007.

Thesis

ENGLISH ABSTRACT: This study investigated the evolution of the genus Palinurus at the higher and lower taxonomic levels. The population genetics of the spiny lobster, Palinurus delagoae, was investigated by making use of a portion of the mitochondrial DNA (mtDNA) control region (547 base pairs) that was sequenced for 285 lobsters from the southeastern coast of Africa (six sites) and 49 lobsters from Walters Shoals (one site), a submerged seamount on the Madagascar Ridge. Lobsters from these two areas shared no haplotypes and differed by at least 27 mutational steps. An analysis of molecular variance showed significant genetic partitioning, and pairwise comparisons suggested that lobsters from Walters Shoals are distinct from those of other sampling areas. Along the south east African coastline there was shallow genetic partitioning between four southern sites (South Africa) and two northern (Mozambique) sites, suggesting two Management Units along the African coast. Female gene flow along the African coast may be propagated by larval dispersal in the Mozambique and Agulhas Currents and counter-current migrations by benthic juveniles along the shelf, but the mtDNA data strongly suggest that larvae at Walters Shoals have been, or are currently still retained by other oceanographic processes. The magnitude of mtDNA divergence among lobsters from the southeastern coast of Africa and those at Walters Shoals, together with the absence of any shared haplotypes between these regions, strongly suggested that these two taxa represent distinct species. The molecular data of the large subunit ribosomal RNA, 16S rRNA (481 bp), and cytochrome oxidase subunit I, COI (520 bp) were then used for a higher level phylogenetic analysis of the genus. A total of 33 individuals (five representatives from each of the six species), and two outgroups (Projasus parkeri and Palinustus unicornutus), were subjected to maximum parsimony, maximum likelihood and Bayesian inference analyses. All analyses were conducted on both the separate data sets as well as a combination of the two genes. Bootstrap analyses of the 16S rRNA data resulted in >70% support for the monophyly of all six Palinurus species but no support could be obtained for any of the interspecific associations. Likewise, individual analyses of the COI gene resulted in strong support for the monophyly of the species. The combined data (parsimony analyses) increased the resolution considerably and apart from the monophyly of all six species, good bootstrap support was also obtained for associations among species. The topology for the maximum likelihood analyses displayed a more resolved and well supported tree when the basal ingroup taxon P. elephas was used to root the tree. The combined Bayesian analyses did not result in a well resolved topology and no significant posterior probabilities could be obtained reflecting the associations among species.

AFRIKAANSE OPSOMMING: Hierdie studie het die evolusie van die genus Palinurus by hoë en laer taksonomiese vlakke ondersoek. Die bevolkingsgenetika studie op die kreef, Palinurus delagoae, is ondersoek deur gebruik te maak van 'n gedeelte van die mitokondriale (mtDNA) kontrole-area (547 basispare) waarvan die volgorde bepaal is vir 285 krewe van die suidoos-kus van Afrika (afkomstig van ses verskillende gebiede) en 49 krewe afkomstig van Walters Shoals (een gebied), 'n ondersese berg op die Madagaskar Rand. Krewe van hierdie twee areas deel geen haplotipes nie en verskil met ten minste 27 mutasiestappe. 'n Analise van die molekulêre variansie toon dat daar 'n beduidende genetiese verdeling tussen die twee groepe is en 'n gepaarde vergelyking toon dat krewe afkomstig van Walters Shoals verskil beduidend van krewe uit ander gebiede. Volgens die vlak genetiese verdeling tussen die vier suidelike (Suid-Afrika) en twee noordelike (Mosambiek) gebiede van die suidoos-kus van Afrika wil dit voorkom of daar twee bestuurseenhede langs die kuslyn van Afrika is. Vroulike geenvloei langs hierdie kuslyn kan dalk bevarder word deur larwale verspreiding in die Mosambiek- en Agulhas- Seestrome en teenstroom migrasie van jong bodemwonende krefies op die kontinentale plaat. Die mtDNA data stel egter voor dat kreeflarwes by Walters Shoals deur ander oseanografiese prosesse steeds (of tot onlangs toe) behou word. Die grootte van mtDNA divergering tussen krewe van die suidoos-kus van Afrika en die by Walters Shoals, sowel as die afwesigheid van enige gemeenskaplike haplotipes tussen die twee gebiede, toon met beduidende sekerheid aan dat hierdie twee taksa twee unieke spesies verteenwoordig. Die molekulêre data van die 16S-rRNA (481bp) van die groot ribosomale-subeenheid en die sitochroom oksidase subeenheid, COI (520bp) is gebruik om 'n hoër resolusie filogenetiese analise van die genus te bepaal. Data van 33 individue (vyf individue uit elk van die ses spesies) en twee buitegroepe (Projasnus parkeri en Palinustus uniconutus) is geanaliseer deur gebruik te maak van die maksimum-parsimonie, die maksimum-waarskynlikheid en die Bayes-inferensie metodes. Alle analises is uitgevoer op beide die afsonderlike datastelle sowel as op die gekombineerde data van die twee gene. Analise van die 16S-rRNA data deur die skoenlusmetode (steekproefhersteekproef- metode) toon meer as 70% steun vir die monofilie van al ses Palinurus spesies maar dit toon geen steun vir enige van die interspesifieke assosiasies nie. Net so toon individuele analise van die COI geen beduidende steun vir die monofilie van die spesies. Die gekombineerde data (parsimonie) het 'n aansienlike verhoging in die resolusie teweeg gebring en behalwe vir die monofilie van al ses die spesies was daar ook goeie steun deur die skoenlusmetode vir die assosiasie tussen spesies verkry. Die topologie vir die maksimum-parsimonie het 'n goed gesteunde en hoër resolusie boom vir die gekombineerde datastel (sonder die buitegroepe) getoon. Die gekombineerde Bayesanalise het nie 'n soortgelyke boom opgelewer nie en die assosiasie tussen die spesies is nie ondersteun nie aangesien geen beduidende a posteriori-waarskynlikheid verkry kon word nie.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/21777
This item appears in the following collections: