Sex identification of South American parrots (Psittacidae, aves) using the human minisatellite probe 33.15

CRISTINA Y MIYAKI, 1 J. MAURICIO B. DUARTE,2 RENATO CAPARROZ,1, 2 ADAUTO L. V NUNES, 3 AND ANITA WAJNTAL 1, 4

Many species of South American parrots are endangered, and captive breeding has become a standard procedure for species conservation. Because most South American parrots are not sexually dimorphic, an efficient means of determining the sex of individuals is an important tool in establishing and maintaining a viable breeding population in captivity.

DNA fingerprinting (Jeffreys et al. 1985) has been applied in a variety of wild species, including birds (Burke and Bruford 1987, Wetton et al. 1987). It was used to monitor genetic variability in captive Puerto Rican Parrots (Amazona vittata; Brock and White 1992) and to establish paternity in endangered species (Mathe et al. 1993). Recently, we used DNA fingerprinting to identify the sex of Peach-fronted (Aratinga aurea) and Golden (Guaruba [Aratinga] guarouba) parakeets (Miyaki et al. 1992) and suggested that fingerprinting also could be used to determine sex in other psittacines (Miyaki et al. 1993, 1995). Here, we present results on sex determination using the human minisatellite probe 33.15 (Jeffreys et al. 1985) in 36 species belonging to 13 genera of South American parrots.

Methods.-Blood samples were collected from birds belonging to aviculturists and official establishments in Brazil. For some individuals, sex was determined by karyotyping, laparoscopy, or breeding behavior Whenever possible, growing feathers (ca. 25 days old) were collected from birds whose sex was unknown and processed for karyotype analysis. Chromosome preparation and analysis followed Duarte and Caparroz (1995). The species and number of individuals studied are shown in Table 1.

The protocols used to obtain multilocus fingerprints followed Bruford et al. (1992). For each bird, 5 (mu)g of genomic DNA were digested with the restriction enzyme MboI (for Amazona) or HaeIII (for the other genera). Fragments were separated by electrophoresis through a 30-cm long 1% horizontal agarose gel. Electrophoresis was stopped when the 2-kilobase (kb) marker band had migrated to the bottom of the gel. The fractionated DNA fragments were transferred onto a nylon membrane by standard capillary Southern blotting (Sambrook et al. 1989).

The membrane was hybridized with minisatellite 33.15 probe (Jeffreys et al. 1985), which was labeled with [a-32P] dCTP or [a?zP] dATE Pre-hybridization in 0.263 M Na2 HPO4, 1 mM EDTA, 7% SDS, and 1% BSA at 65C lasted for 4 h, and the probe was added and left overnight at 65C. The membrane was washed in 0.25 M Na2HPO4, 1% SDS, 2xSSC, 0.1% SDS and in 1 x SSC, 0.1% SDS at 65oC. The filter was then autoradiographed at -70C using Kodak RX film with one or two intensifying screens.

Results.-A pattern of two to four intense sex-specific bands was detected with the 33.15 probe in all females studied in the genera Ara and Aratinga and also in females of Anodorhynchus hyacinthinus, A. leari, Cyanopsitta spixii, Guaruba guarouba, and Nandayus nenday (see Fig. 1). These intense bands also were present in some previously unsexed individuals of Aratinga solstitialis auricapilla, A. s. jandaya, and A. mitrata (Table 1). Each species showed unique patterns of female-linked bands according to their molecular size (Table 2); these species-specific profiles are a potential tool for species identification. Female-linked bands also were observed in long-tailed species even when other restriction enzymes were used (data not shown). Sex-linked bands were absent in known females of eight species of Amazona and of Pionus menstraus, Deroptyus accipitrinus, Pionites leucogaster, Pionopsitta pileata, Pyrrhura egregia, Pfrontalis, P picta, and Triclaria malachitacea. We used MboI for members of Amazona because this enzyme produced more polymorphic DNA fingerprint profiles; even when we used HaeIII, no sex-linked fragments were observed (data not shown).

Discussion.-Until recently, only three methods of sex identification were available for birds: fecal hormone assay, chromosome analysis, and laparoscopy. Hormone analyses vary with reproductive condition and thus are age and season dependent. Chromosome analysis is reliable but time consuming, and in many occasions appropriate metaphase cells cannot be found for analysis without repeated attempts. Birds have to be handled twice for karyotyping, once for removing a sample of feathers and a second time for collecting the growing feather pulp. Because laparoscopy is a surgical procedure, birds are exposed to anesthesia and to the risks of surgery.

Recently, several DNA techniques have been developed for sex identification in birds (de Kloet and de Kloet 1992, Griffiths and Tiwari 1993, May et al. 1993). Griffiths and Tiwari (1995) determined the sex of the last Spix's Macaw (Cyanopsitta spix) from shed feathers that were collected in the wild and used as a source of DNA for PCR amplification. In our application of DNA fingerprinting in South American parrots, we identified female-specific intense bands in some species using the human minisatellite probe 33.15 (Miyaki et al. 1992, 1993). Other authors also have detected sex-specific bands in DNA fingerprinting of birds using probe 33.15 (Rabenold et al. 1991, Longmire et al. 1992) and probe 33.6 (Graves et al. 1993). Although probe 33.15 could not be used to identify the sex of all the species we studied, it was invaluable for sexing macaws (Ara, Anodorhynchus, and Cyanopsitta) and conures (Aratinga and Nandayus). These genera belong to the "long, point-tailed" group and are considered to be closely related. Our method could not be used to sex the short-tailed parrots in the genus Amazona, and our data suggest that it also was not adequate for sexing other short-tailed species such as Pionus menstruus, Pionites leucogaster, and Pionopsitta pileata (which has sexually dimorphic plumage), as well as the "long, wide-tailed" Deroptyus accipitrinus and Triclaria malachitacea (the latter has sexually dimorphic plumage). The absence of female-linked bands was not due to the fact that a different restriction enzyme was applied. No sex-specific pattern was detected in the three species of Pyrrhura we studied, each of which has a long, sharp-pointed tail.