Genetic diversity and structure of wild Tunisian Myrtus communis L. (Myrtaceae) populations

Myrtus communis L. (Myrtaceae), in Tunisia, is closely associated with Quercus suber L. forest which stretched continuously from the North to parts of Cap Bon and Tunisian Dorsal. The destruction of the primary oak forests associated to an over-exploitation of Myrtus for its essential oil quality had led to discontinuous populations exhibiting various levels of degradation. Using starch gel electrophoresis, we analyzed the polymorphism of nine isozymes in order to assess genetic diversity and structuring of 17 natural populations prospected in the three geographical regions and coinciding with subhumid, humid inferior and semi-arid superior bioclimates. The analysis of the level and the distribution of the genetic diversity in this species might help in its conservation. Out of the 18 loci detected for all populations and isozymes analyzed, 12 loci were polymorphic. Allelic frequencies differed according to populations and particular alleles characterized ecological groups. A high level of genetic variation within populations was observed. The mean number of alleles per polymorphic locus was Ap = 1.67, the percentage of polymorphic loci was P = 60.3% and the observed (Ho) and the expected (He) heterozygosities were respectively 0.144 and 0.215. Populations belonging to subhumid (Cap Bon) and semi-arid superior (Tunisian Dorsal) climates, located in degraded sites exhibited the highest level of inbreeding (0.425 < F-IS < 0.450). A high level of differentiation (F-ST = 0.396) and a low gene flow (Nm = 0.337) among populations, as a result of habitat intermediate destruction, were revealed. The differentiation of populations within the same bioclimate (or geographic) group was substantial and relatively higher for semi-arid superior populations (F-ST = 0.262), which were more distant. The three ecological groups exhibited a high level of structuring (F-ST = 0.401). These differentiations might be due to geographic distances and to the variations of ecological factors between sites, including human activities and environmental factors. Nei's (1972) genetic distances calculated between pairs of populations were globally low (0.006 < D < 0.367) with a mean of 0.15. They indicate a high level of similarity between populations. UPGMA clustering, established through Nei's genetic distances, showed three population aggregates according to their geographic/bioclimatic appartenances. The high differentiation between populations and the low level of their genetic divergence indicated their recent isolation under anthropic pressures. The species conservation (in situ or ex situ) strategies should take into account the genetic diversity level within populations and its variation between geographic groups.