Population genetic structure of the endangered tropical tree species Caryocar brasiliense, based on variability at microsatellite loci

Authors

  • Rosane G. Collevatti,

    Corresponding author
    1. Pós-Graduação em Biotecnologia e Recursos Genéticos, Universidade Católica de Brasília, Brasília, SGAN 916, Mod. B, Asa Norte, Brasília, D.F. Brasil, 70790–160,
    2. Departamento de Ecologia, Universidade de Brasília, Brasília, D.F. Brasil, 70910–900
      Rosane Garcia Collevatti. Fax: + 55 61 340–5550; E-mail:rosanegc@pos.ucb.br
    Search for more papers by this author
  • Dario Grattapaglia,

    1. Pós-Graduação em Biotecnologia e Recursos Genéticos, Universidade Católica de Brasília, Brasília, SGAN 916, Mod. B, Asa Norte, Brasília, D.F. Brasil, 70790–160,
    2. Laboratório de Genética de Plantas, Recursos Genéticos e Biotecnologia–EMBRAPA, C.P.02372, Brasília, D.F. Brasil. 70770–900,
    Search for more papers by this author
  • John D. Hay

    1. Departamento de Ecologia, Universidade de Brasília, Brasília, D.F. Brasil, 70910–900
    Search for more papers by this author

Rosane Garcia Collevatti. Fax: + 55 61 340–5550; E-mail:rosanegc@pos.ucb.br

Abstract

We report the population genetic structure of the endangered tropical tree species Caryocar brasiliense, based on variability at 10 microsatellite loci. Additionally, we compare heterozygosity and inbreeding estimates for continuous and fragmented populations and discuss the consequences for conservation. For a total of 314 individuals over 10 populations, the number of alleles per locus ranged from 20 to 27 and expected and observed heterozygosity varied from 0.129 to 0.924 and 0.067 to 1.000, respectively. Significant values of θ and RST showed important genetic differentiation among populations. θ was much lower than RST, suggesting that identity by state and identity by descent have diverged in these populations. Although a significant amount of inbreeding was found under the identity by descent model (f = 0.11), an estimate of inbreeding for microsatellite markers based on a more adequate stepwise mutation model showed no evidence of nonrandom mating (RIS = 0.04). Differentiation (pairwise FST) was positively correlated with geographical distance, as expected under the isolation by distance model. No effect of fragmentation on heterozygosity or inbreeding could be detected. This is most likely due to the fact that Cerrado fragmentation is a relatively recent event (~60 years) compared to the species life cycle. Also, the populations surveyed from both fragmented and disturbed areas were composed mainly of adult individuals, already present prior to ecosystem fragmentation. Adequate hypothesis testing of the effect of habitat fragmentation will require the recurrent analysis of juveniles across generations in both fragmented and nonfragmented areas.

Ancillary