Contribution of population genetics to investigation efforts on wildlife trafficking in Brazil: development of microssatelites and analysis of the genetic structure of Paroaria dominicana and Saltator similis(Aves: Passeriformes: Thraupidae)

Author(s)

Juliana Machado Ferreira

• Document type: Doctoral Thesis
• Press: São Paulo.
• Institution: Universidade de São Paulo (USP). Instituto de Biociências (IBIOC/SB)
• Defense date: 2012-08-13

Examining board members

Joao Stenghel Morgante; Anelisa Ferreira de Almeida Magalhães; Cristina Yumi Miyaki; Luis Fábio Silveira

• Advisor: Joao Stenghel Morgante

Abstract

Wildlife trafficking is a major threat to Brazilian biodiversity, with millions of specimens withdrawn from nature every year to supply the consumer markets. In Brazil, the pet trade generates a major demand for birds. After seizure from the illegal trade, rehabilitation and release is one way of mitigating the negative impacts on wild bird populations, but those releases could create more problems for the natural populations. Knowledge of the species\' genetic structure is crucial to help avoid the possibility of outbreeding depression and to guide responsible releases. This project´s main goal was the study of the genetic structure of four passerine species, among the most trafficked birds in Brazil: ultramarine grosbeak (Cyanoloxia brissonii), green-winged saltator (Saltator similes), red-cowled cardinal (Paroaria dominicana) and buffy-fronted seedeater (Sporophila frontalis). We obtained 522 tissue/blood samples (from individuals with known orginin as well as those seized from the ilegal trade) both in field work trips and through collaboration with various institutions. We built four species-specific genomic libraries, all of which were cross-tested in the other three sepcies. From over 700 sequenced clones, we developed eight informative microsatellite loci for P. dominicana, nine for S. similes, seven for C. brissonii and four for S. frontalis. Genetic structure analyzes were performed only for P. dominicana and S. similis due to the difficulty of obtaining representative samples for the other two species. Data was analyzed both with and without a priori definition of populations using discriminant analysis of principal components, Bayesian clustering algorithms, and fixation indices. Genetic structure of P. dominicana was inferred using 71 samples with known origin and subtle but significant (p<0.01) genetic structure was found, with the description of three clusters: (1) northeast, which encompass the sampling localities of Paraíba, Trindade e Petrolina, with mean observed heterozigosity (Ho) of 0.736 and mean expected heterozigosity (He) of 0.753; (2) southwest, which encompass the localites Parnaguá, Beira Rio e Vitória da Conquista, with Ho 0.624 and He 0.731; and (3) east, which encompass Canudos and Feira de Santana, with Ho 0.750 and He 0.768. Occurrence of outbreeding depression for this species is unlikely due to high gene flow among clusters. The presence of genetic structure for nuclear markers not detected for mtDNA (previous studies) indicates male philopatry and different dispersion patterns between male and females. No population expansion signal was detected using kg tests, which precludes us from corroborating a previously proposed hypothesis of population expansion congruent with the Caating expansion 15 thousand years ago. Two tests (Bottleneck e M-ratio) indicated the occurrence of recent population bottlenecks for P. dominicana, which can be explained by the intense occupation, habitat degradation and over-exploitation of the Caatinga in the last 200 years. We also propose a new hypothesis, to be tested in future studies: the existence of a dispersion route along the Sao Francisco river, promoting gene flow between northeast and southwest clusters. Furthermore, we performed assignment tests in 49 samples from specimens seized from the illegal trade in Sao Paulo. All of them were assigned to the same \"northeast\" cluster with probability 0.554-0.713, showing a clear trend and a relevant result which corroborates police intelligence information. For S. similis, no significant genetic structure was detected (using 66 samples), so we could not refut the pan-mixia null hypothesis. Although this species\' range is a large, and now fragmented area, the possibility of outbreeding depression resulting from mixing populations is higly unlikely, which does not preclude consideration of other characteristics prior to release into the wild. The observed heterozigosity was 0.643 while the expected was 0.757. High levels of deviations from Hardy-Weinberg equilibrium may be due to population fluctuations. No population expansion signal was detected (kg test), but two tests (Bottleneck e M-ratio) indicate the occurrence of recent population bottlenecks, which could be related to the intense occupation, habitat degradation and over-exploitation of the Cerrado and Atlantic Forest in the last 200 years. Variation in song dialects observed in this species is probably related to characteristics of sound propagation in different environments, rather than to genetic isolation. Knowledge of genetic structure is one of the most important steps towards intelligent management of rehabilitated seized animals (AU)