Genetic variability of wild Phaseolus lunatus L. populations from the central valley of Costa Rica and its implications in the development of an in situ conservation strategy | Variabilite genetique des populations sauvages de Phaseolus lunatus L. dans la valle centrale du Costa Rica et ses implications dans la mise au point d'une strategie de conservation in situ

Suitable electrophoretic separation methods for 34 isozymes from 17 enzymatic systems and a study of the genetic control of certain loci allowed us to investigate the genetic variability and the main factors contributing to its maintenance for wild Phaseolus lunatus L. populations originated from the central valley of Costa Rica. An analysis of the combination of seed and plant numbers to sample per population showed that ten to 80 plants and one to two seeds per plant are necessary to collect the highest genetic diversity when we consider polymorphism at 5 per cent level. When any polymorphism level is fixed, one to four seeds per plant and 200 to 300 plants per population will be required to collect all alleles at six enzyme loci. Using the designed sampling scheme and 29 populations, we quantified the genetic diversity at 22 loci. The estimates of the percentage of polymorphic loci (P), the mean number of alleles per locus (A), and the mean effective number of alleles (Ae) were 10.32 per cent, 1.10 and 1.05 respectively. The genotypic composition of the analysed populations showed an deviation from the expected Hardy-Wemberg proportions (Fit =0.878). This disequilibrium is due to either genetic differentiation between populations (Fst= 0.444) and nonrandom mating within populations (Fis = 0.777). The estimates of the total heterozygosity (Ht), the intrapopulation genetic diversity (Hs), and the interpopulation genetic diversity (Dst) were 0.193, 0.082 and 0.111 respectively. About 52 per cent of the total genetic diversity is due to interpopulation genetic variation (Gst =0.519), 48 per cent being relative to intrapopulation genetic variability. No relationships has been highlighted between the intrapopulation genetic variability and the populations size for the studied material. The maintenance of the genetic variability in populations with small size has been attributed to the soil seed bank that reduce the genetic drift effect by increasing the population effective size. An estimation of the mating system parameters and a detailed analysis of factors that influence these parameters allowed to confirm that wild P. lunatus can be classified as mixed mating with predominantly selfing breeding system (t = 0.072). Analysing the spatial structure, we demonstrated that wild Lima bean populations are subdivided into sub-populations. A study of the micro-geographical pattern of the genetic variability for 95 populations highlighted a structuration in the spatial distribution of allele frequencies. From all the obtained results, we suggested recommendations for P. lunatus genetic resources in situ conservation.