NOTES AND COMMENTS

Identification of pumas (Puma concolor (Linnaeus, 1771)) through faeces: a comparison between morphological and molecular methods Miotto, RA.a*, Ciocheti, G.b, Rodrigues, FP.a and Galetti Jr., PM.a aDepartamento de Genética e Evolução, Universidade Federal de São Carlos – UFSCar, Rodovia Washington Luís, Km 235, CEP 13565-905, São Carlos, SP, Brazil bInstituto de Biociências, Universidade de São Paulo – USP, São Paulo, SP, Brazil *e-mail: [email protected] Received May 31, 2006 – Accepted October 23, 2006 – Distributed December 1, 2007 (With 2 figures)

Pumas (Puma concolor (Linnaeus, 1771)) occur in and a characteristic odor, aspects that differ them from low densities and possess habits that hinder observation the faeces of other large carnivores (Chame, 2003). Of (Ernest et al., 2000). Often, the presence of these animals the 32 collected faeces, 12 were dissolved by rainwa- may be confirmed through the identification of their fae- ter or extremely deteriorated by the action of insects and ces, avoiding the elevated costs of capture and its possi- were discarded from the analyses. We placed the samples ble damages (Wayne and Morin, 2004). Faecal analysis in sterile preservative-free plastic tubes without any con- allows the monitoring of populations, their movements servative solution and kept at –22 °C in the laboratory (Kohn and Wayne, 1997; Taberlet et al., 1997; Prugh until DNA extraction. As a control during the DNA am- et al., 2005), and provides diet-related information plification reactions, we used blood samples from pumas (Farrel et al., 2000). (Puma ­concolor and Puma yagouaroundi (É. Geoffroy Traditionally, the identification of the species that Saint‑Hilaire, 1803)) and ocelots (Leopardus pardalis), originally deposited the faeces is based on the deposi- species found in the studied areas (Talamoni et al. 2000). tion site, size, odor, and tracks associated with faeces, Both DNA and guard hair analyses were done independ- although these characteristics may be subjective or in- ently by two distinct researchers in such a manner that consistent (Becker and Dalponte, 1999). Body size can one did not know the results from the other. vary greatly within a species, and an individual can leave We extracted faecal DNA using a QIAmp DNA Stool scats in a broad range of sizes (Farrell et al., 2000). In ad- Mini Kit (Quiagen) according to the manufacturer’s rec- dition, puma and ocelot (Leopardus pardalis (Linnaeus, ommendations and used phenol/chloroform/isoamylic 1758)) faeces may also present similar sizes (Miotto alcohol (Sambrook et al., 1989) for DNA extraction of et al., 2007), a condition that increases the chance of blood samples. these species to be misidentified. We amplified a 146 base pairs (bp) fragment of the Complementary methods for faecal identification cytochrome b gene from the mitochondrial genome via are important tools in works based on this kind of mate- polymerase chain reaction (PCR) using primers designed rial. We compared two methods to identify the species to by Farrel et al. (2000). We compared the obtained se- which sample faeces collected in field belong: analysis quences with GenBank reference sequences from pumas of faecal DNA and analysis of guard hairs contained in and other felids that are also found in the sampled re- the faeces. As part of a self-cleaning behavior, felines gions (Miotto et al., 2007). Of the 20 analyzed samples, often extract their own hairs with the tongue and swallow 12 amplified the mitochondrial DNA fragment - ten from them (personal observation). Once isolated from faeces, pumas and two from ocelots (Figure 1). guard hair analysis represents an alternative to identify For the guard hair analysis, we determined the cu- the species that originally deposited faecal samples. ticular pattern by means of impression of the hair sur- Another possibility is by analyzing DNA content, faces and compared it to a reference collection (Quadros, considering that few grams of faeces contain DNA from 2002). Since the medular pattern (large medula with fim- thousands of intestinal mucosal cells from the individual briated edges) is similar among all species belonging to that made the deposit (Albaugh et al., 1992). In the last the Felidae group (Quadros, 2002), we carried out only decade, a large number of population studies have been the cuticular pattern analysis from the hair shaft. This based on faecal DNA analysis (Taberlet et al., 1997; guard hair microstructure analysis found the same results Wasser et al., 1997; Palomares et al., 2002; Riddle et al., obtained by DNA analysis. We submitted the remaining 2003; Deagle et al., 2005; Prugh et al., 2005). eight samples that did not amplify the mitochondrial From October 2004 to August 2005, we collected DNA only to the microstructure analysis of the guard 32 faeces of supposed pumas in two protected cerrado hair, and all were identified as belonging to pumas. areas in Southeastern Brazil: Vassununga State Park The cuticular patterns observed in Felidae are petal, (approximately 21° 41’ S and 47° 34’ W) and Jataí diamond petal, or regular waved (Quadros, 2002). For Ecological Station (approximately 21° 51’ S and 47° pumas, we found the transversally waved pattern, i.e., 82’ W). Feline faeces are segmented, with tapered ends, scales with a waved contour and disposed transversally

Braz. J. Biol., 67(4, Suppl.): 963-965, 2007 963 Miotto, RA. et al.

P. concolor 10 (FS) P. concolor (BS) a b P. concolor 8 (FS) P. concolor (GB) P. concolor 16 (FS) 67 P. concolor 9 (FS) P. concolor 2 (FS) P. concolor 11 (FS) P. concolor 5 (FS) 81 P. concolor 20 (FS) P. concolor 18 (FS) P. concolor 7 (FS) 100 P. yagouaroundi (GB) 97 P. yagouaroundi (BS) L. pardalis (GB) L. pardalis 1 (FS) 82 Figure 2. Photomicrography of the cuticular patterns from L. pardalis 17 (FS) hair shaft of a) Puma concolor and b) Leopardus pardalis 46 L. pardalis (BS) obtained from faeces collected in field (200x increase). C. thous (GB)

Figure 1. Cytochrome b neighbor-joining tree of the refer- Acknowledgments — The authors thank CENAP/IBAMA and ence sequences and sequences from the analyzed faecal sam- the Associação Mata Ciliar for supplying the blood samples ples. Cerdocyon thous is an outgroup; Puma ­yagouaroundi and the Conselho Nacional de Desenvolvimento Científico e and Leopardus pardalis are other felids present in the study Tecnológico (CNPq) for financial support. area. The bootstrap values, based on 1,000 randomizations, are shown on the branches. The FS and BS nomenclature re- References fers to faecal and blood samples, respectively; GB refers to the other sequences obtained at GenBank (access numbers ALBAUGH, GP., IYENGAR, V. and LOHANI, A., 1992. DQ469940 - DQ469953, DQ790891). Isolation of exfoliated colonic epithelial cells, a novel, non- invasive approach to the study of cellular markers. Int. J. Cancer, vol. 52, p. 347-350. to the hair shaft (Figure 2 a). For ocelots, we found hairs BECKER, M. and DALPONTE, JC., 1999. Rastros de mamíferos with the intermediary petal cuticular pattern (Quadros, silvestres brasileiros. Brasília, DF, Editora UnB, 180p. 2002), with imbricated and wider scales (Figure 2 b). CHAME, M., 2003. Terrestrial mammal feces: a morphometric The confirmation of the presence of two ocelot fae- summary and description. Mem. I. Oswaldo Cruz, vol. 98, ces among those supposedly belonging to pumas indi- p. 71-94. cated that only field experience is not enough to tell them DEAGLE, BE., TOLLIT, DL., JARMAN, SN., HINDEL, MA., apart, pointing out the need of complementary methods TRITES, AW. and GALES, NJ., 2005. Molecular scatology as that may aid their identification. a tool to study diet: analysis of prey DNA in scats from captive The guard hair microstructure study proved to be a Steller sea lions. Mol. Ecol., vol. 14, p. 1831-1842. useful tool in the identification of faecal samples. The ERNEST, HB., PENEDO, MCT., MAY, BP., SYVANEN, MS. guard hairs are long hairs that stick out in the fur with a and BOYCE, WM., 2000. Molecular tracking of mountain lions mechanoreceptor function, or even the hairs whose in- in the Yosemite Valley region in California: genetic analysis using microsatellites and faecal DNA. Mol. Ecol., vol. 9, dividual coloring produces the general color pattern of p. 433-441. the pelage (Quadros, 2002). Thus, we emphasize that the simple observation of characteristics such as hair color FARREL, LE., ROMAN, J. and SUNQUIST, ME., 2000. Dietary separation of sympatric carnivores identified by molecular and size could generate incorrect results, due to a large analysis of scats. Mol. Ecol., vol. 9, p. 1583-1590. fur variation according to the geographical distribution and season of the year (personal observation). KOHN, MH. and WAYNE, RK., 1997. Facts from feces revisited. Trends Ecol. Evol., vol. 12, p. 223-227. In contrast, the genetic identification of field-col- lected faeces becomes practical when there are doubts MIOTTO, RA., CIOCHETI, G., RODRIGUES, FP. and regarding identification of the guard hairs, when they are GALETTI JR, PM., 2007. Determination of the minimum population size of pumas (Puma concolor) through faecal scarce, or when there is not a good reference collection. DNA analysis in two protected cerrado areas in the Brazilian It is mostly efficient when individual identification is southeast. Biotropica, vol. 39, p. 647-654. aimed (Miotto et al., 2007). PALOMARES, F., GODOI, JA., PIRIZ, A., O’BRIEN, SJ. and Non-invasive analyses represent a new possibility to JOHNSON, WE., 2002. Fecal genetic analysis to determinate populational study and monitoring of large carnivores the presence and distribution of elusive carnivores: design and with elusive habits. Our results indicate that the analyses feasibility for the Iberian lynx. Mol. Ecol., vol. 11, p. 2171-2182. of the DNA and guard hair found in the faeces are ef- PRUGH, LR., RITLAND, CE., ARTHUR, MA. and KREBS, ficient and, when used together, may provide higher ac- CJ., 2005. Monitoring coyote population dynamics by curacy in identification of faeces collected in the field. genotyping faeces. Mol. Ecol., vol. 14, p. 1585-1596.

964 Braz. J. Biol., 67(4, Suppl.): 963-965, 2007 Identification of puma faeces

QUADROS, J., 2002. Identificação microscópica de pêlos de endangered Pyrenean brown bear population. Mol. Ecol., mamíferos brasileiros e sua aplicação no estudo da dieta de vol. 6, p. 869-876. carnívoros. 127 p. (Tese de Doutorado) – Universidade Federal do Paraná, Curitiba, PR. TALAMONI, SA., MOTTA JUNIOR, JC. and DIAS, MM., 2000. Fauna de mamíferos da Estação Ecológica de Jataí e RIDDLE AE., PILGRIM, KL., MILLS, LS., MCKELVEY, da Estação Experimental de Luiz Antônio. In SANTOS, JE. KS. and RUGGIERO, LF., 2003. Identification of mustelids and PIRES, JSR. (eds.). Estudos integrados em ecossistemas, using mitochondrial DNA and non-invasive sampling. Conserv. Estação Ecológica de Jataí. vol. 1. São Carlos, Editora Rima, Genet., vol. 4, p. 241-243. p. 317-329. SAMBROOK, J., FRITSCH, EF. and MANIATIS, T., 1989. WASSER, SK., HOUSTON, GM., CADD, GG. and FAIN, R., Molecular Cloning: A Laboratory Manual. Cold Springs 1997. Techniques for application of fecal DNA methods to field Harbor, NY, Cold Springs Harbor Laboratory Press. studies of Ursids. Mol. Ecol., vol. 6, p. 1091-1097. TABERLET, P., CAMARRA, JJ., GRIFFIN, S., HANOTTE, WAYNE, RK. and MORIN, PA., 2004. Conservation genetics in O., WAITS, LP., DUBOI-PAGANON, C., BURKE, T. and the new molecular age. Front. Ecol. Environ., vol. 2, p. 89-97. BOUVET, J., 1997. Noninvasive genetic tracking of the

Braz. J. Biol., 67(4, Suppl.): 963-965, 2007 965