Mem Inst Oswaldo Cruz, Rio de Janeiro, 98(8) November 2003
Siphonaptera Parasites of Wild Rodents and Marsupials Trapped in Three Mountain Ranges of the Atlantic Forest in Southeastern Brazil
Departamento de Parasitologia, Instituto de Biologia, Universidade Estadual de Campinas, Caixa Postal 6109,u00a0 13083-970 Campinas, SP, Brasil
A study of the associations between small mammals and fleas was undertaken in three areas of the Atlantic Forest in Souhtheastern Brazil: Serra da Fartura, SP, Serra da Bocaina, SP, and Itatiaia, RJ. Trapping of small rodents and marsupials was done every 3 months during 2 years, from June 1999 to May 2001. A total 502 rodents (13 species) and 50 marsupials (7 species) were collected, and 185 hosts out of 552 (33.5%) captured in the traps were parasitized by 327 fleas belonging to 11 different species. New host records were determined for several flea species, and 5 significant associations between fleas and hosts were also found.
The fleas (Insecta: Siphonaptera) have approximately 3000 species worldwide with approximately 59 species and subspecies recorded from Brazil; of these, 31 were originally described from Brazilian specimens. Approximately 15% of the genera and 29% of the species are endemic. Their main hosts are mammals and they show great preference for rodents (Linardi & Guimarães 2000).
There are several studies dealing with the Brazilian flea fauna (Guimarães 1972, Botelho et al.1981, Linardi 1985, Linardi et al.1984, 1987, Botelho & Linardi 1996, Barros-Battesti & Arzua 1997, Barros-Battesti et al.1998). However, few were done in areas of the Atlantic rain forest (Linardi 1977, Guitton et al.1986, Bicho et al.1999, Carvalho et al.2001, Bossi et al.2001). This ecosystem is characterized by great biodiversity and presently is under heavy pressure caused rapid urbanization and human disturbance. This study objectived to determine the flea fauna and the flea-host associations in three mountain ranges of the Atlantic forest in Southeastern Brazil: Serra da Fartura and Parque Nacional da Serra da Bocaina (state of São Paulo), and Parque Nacional do Itatiaia (state of Rio de Janeiro).
MATERIALS AND METHODS
Collections were done every three months during two consecutive years, from June 1999 to May 2001 in three areas of the Atlantic Forest in Southeastern Brazil: Serra da Fartura (São João da Boa Vista, SP - 21°53.621' W 46°45.188'), Serra da Bocaina National Park (São José do Barreiro, SP - 22°44.125' S 44°37.007' W), and Itatiaia National Park (Itatiaia, RJ - 22°26.187' S, 44°37.511' W). Small mammal trapping was conducted in an irregular grid covering 5 ha, with 14 transects varying from 200 to 240 m, spaced 20-m from each other and 1000 to 1500 m above sea level. ShermanÒ live traps (100) baited with banana or manioc with peanut butter were set on the forest floor for four consecutive nights. Each trap site was recorded, and trapped animals were taken to a nearby station laboratory. Handling of small mammals was similar to the procedure used by Gettinger (1992). The animals were placed in plastic bags with cotton balls soaked with ethyl ether until they became unconscious, after which they were marked using an ear-code (Monteiro-Filho 1987), weighed, and sexed. Fleas were removed from the fur, ears, and tail with fine combs, toothbrushes or tweezers and stored in 70% alcohol. The hosts harboring specimens of Tunga were taken to the laboratory of the Department of Parasitology, State University of Campinas, where the fleas were removed. The hosts were subsequently released at the same trap site. A voucher collection of skins and skulls of male and female of all collected host species were kept at the Department of Ecology, State University of Rio de Janei-ro. Mammal identification was based on morphological characters and cariotype, using material from the sacrificed specimens. Fleas that fell from the animals into the plastic bag during anesthesia were also stored in 70% alcohol. Each trapping was carried out for four consecutive nights, totaling eight collections in each location, and the total trapping effort was 3200 traps/night/site. Serra da Bocaina National Park belongs to the Serra do Mar mountain complex whereas the other two locations belong to the Serra da Mantiqueira complex, and the two complexes are separated by the Paraíba do Sul river valley. All three areas are covered by the Atlantic rain forest vegetation. The fleas were clarified and mounted in glass slides and identified using the key of Linardi and Guimarães (2000). The Rhopalopsyllinae were named according to the nomenclature revision proposed by Linardi and Guimarães (1993). The mean abundance, defined as I = nr fleas/nr of collected hosts (Bush et al. 1997), was determined for the collected species of fleas and their hosts, as well as for the three sites. This index is the same as the Flea Index commonly used for the Siphonapera (Linardi & Guimarães 2000).
A total of 502 rodents and 50 marsupials belonging to 20 species were collected. The rodents were represented by 13 species: Akodon cursor (Winge), Akodon montensis (Thomas), Akodon serrensis (Thomas), Delomys dorsalis(Henzel), Delomys sublineatus (Thomas), Oligoryzomys flavescens (Watherhouse), Oligoryzomys nigripes (Olfers),Oryzomys ratticeps (Hensel), Oxymycterus sp. (Watherhouse), Thaptomys nigrita (Lichtenstein), Nectomys squamipes (Brants), Trinomys gratiosus (Günter),and an undescribed species of Juliomys. The eight collected marsupial species were: Didelphis aurita (Wied-Neuwied), Gracilinanus agilis (Burmeister), Marmosops incanus(Lund), Marmosops paulensis (Tate), Monodelphis americana (Muller), Monodelphis scalops (Thomas), Philander frenata (Olfers), and Thylamys velutinus (Wagner).
The frequency of each collected species is presented in Table I. A total of 185 specimens belonging to 17 species (11 rodents and 6 marsupials) was parasitized by fleas, and Oligoryzomys flavescens and Juliomys sp. (rodents),Marmosops paulensis, and Monodelphis scallops (marsupials) had no fleas.
Considering the five most collected rodent species, the prevalence of parasitism was greater in A. cursor (92.3%) followed by O. nigripes (66.7%), A. serrensis (42.7%), A. montensis (33.3%), and D. sublineatus (19.6%).
The total number of collected fleas was 327, and the most abundant species was Polygenis (Neopolygenis) pradoi(Wagner) with 110 specimens (33.6% of the total fleas) followed by Polygenis (Polygenis) rimatus (Jordan) (N = 53),Polygenis (N.) atopus (Jordan and Rothschild) (N = 50), Craneopsylla minerva minerva (Rothschild) (N = 35),Polygenis (P.) roberti roberti (Rothschild) (N = 29), Adoratopsylla (Adoratopsylla) antiquorum antiquorum(Rothschild) (N = 13), Polygenis (N.) dentei (Guimarães) (N = 12), Polygenis (N.) frustratus (Johnson) (N = 8),Tunga caecata (Enderlein) (N = 9), Hechtiella lakoi (Guimarães) (N = 7), and Polygenis (P.) tripus (Jordan) (N = 1).The abundance for each flea species in relation to the rodent and marsupial hosts are presented respectively in Tables II and III.
Hechtiella lakoi was collected only in Itatiaia National Park, parasitizing N. squamipes and T. gratiosus, totaling respectively 50% (4 out of 8) and 100% (3 of 3) of the fleas found in these hosts. Considering the most abundant rodents, there was a strong association between P. pradoi and the hosts A. montensis and A. serrensis, which represented respectively 55.9% e 43.5% of the fleas collected on these cricetids (Table II). The association was also significant between P. pradoi and A. montensis (c² = 8.6; p < 0.01) and between P. pradoi and A. serrensis (c² = 12.8; p < 0.001), considering all three collecting sites together. A significant association between P. pradoi and another rodent, Oxymycterus sp., was found in Araucaria coniferous forests in souheastern of the state of Paraná (Barros-Battesti et al. 1998).
Significant associations between P. r. roberti and O. nigripes (c² = 77; p < 0.001), between P. dentei and A. serrensis(c² = 28; p < 0.001) and between C. m. minerva and A. serrensis (c² = 8.76 p < 0.01) were also found, the last one considering only associations with rodents.
Among the most abundant hosts the greatest mean abundance was found for O. nigripes (I = 1.38) and D. sublineatus(I = 1.20) in Serra da Fartura, followed by A. serrensis (I = 1.14) and A. cursor (I = 1.00)in Itatiaia, and A. serrensis(I = 0.63)and A. montensis (I = 0.57) in Serra da Bocaina (Table I). Therefore, the three areas had different mean abundances for each rodent as well as variations in the most parasitized species. In addition, Itatiaia presented a higher mean abundance (I = 0.77) (Table I). The following records of flea-host associations are new: A. a. antiquorum on A. montensis and P. frenata; C. m. minerva on D. sublineatus, P. frenata and T. velutinus; H. lakoi onN. squamipes and T. gratiosus; P. atopus on A. montensis, A. serrensis, and D. sublineatus; P. dentei on A. serrensis;P frustratus on A. serrensis; P. pradoi on T. velutinus; P. rimatus on D. sublineatus and P. frenata; P. r. roberti on A. montensis, D. sublineatus, G. agilis, and Oligoryzomys sp.; P. tripus on A. montensis; T. caecata in A. montensis, D. dorsalis, D. sublineatus, and M. americana.
The species P. pradoi was the most frequently flea with 110 specimens collected. Similar to the findings of Linardi (1985), approximately 93% of the specimens of this flea were found parasitizing rodents of the genus Akodon. In addition, several specimens of P. frustratus had morphological characteristics intermediate between this species andP. pradoi, fact already described by Linardi (1979), who considered this group to belong to the "pradoi complex". As already found by other authors (Johnson 1957, Gomes 1969, Linardi 1977, Barros et al. 1993), P. frustratus was always collected in the same sites as P. pradoi, and in many cases on the same host. Contrary to Linardi (1985), who found that Akodon showed a prevalence of 80% of parasitism by P. frustratus, in this workA. serrensis andOxymycterus sp. presented similar infestation rates. P. rimatus was the flea with the widest host range, being found on eight species of rodents (Table II) and one of marsupials (Table III). However, this flea showed preference for hosts of the genus Akodon, since approximately 70 % of the fleas of this species were collected on this host. Linardi (1985) also found a preference of P. rimatus for these cricetids, but with an infestation rate of only 25%.
The species P. r. roberti showed a strong association with O. nigripes, with 52.9% of all fleas of this species being collected on this host. This result agrees with the findings of Linardi (1985). In Southeastern Brasil, P. tripus is more widely distributed in savannas and open areas (Linardi & Guimarães 2000). In this study this species was collected only at Serra da Fartura, an area of Atlantic Forest but close to an area of "cerrado", a savanna-like vegetation.
As reported by Linardi (1977), P. dentei was collected 63.6% in double or triple infestations together with P. pradoi,P. rimatus, and/or P. atopus, always on Akodon spp.
Approximately 50% of the occurrences of P. atopus were on Akodon spp., contrary to the findings of Linardi (1985), which reported a preference of this flea for Oryzomys. In approximately 44% of the occurrences, this species was also found in double or triple associations with P. r. roberti, P. pradoi, P. rimatus, C. m. minerva, and/or P. dentei.
The mean abundances found in this work were smaller than those found in Angra dos Reis, state of Rio de Janeiro (I = 3.6) (Guitton et al.1986) and in Serra dos Órgãos National Park, Rio de Janeiro (I = 1.54) (Carvalho et al.2001), but higher than those found in Guaraqueçaba, Para-ná (I = 0.3) (Bicho et al.1999). In addition, a new site record was determined for T. caecata,which is now the 24th flea species collected in the state of Rio de Janeiro.
All specimens of Tunga collected at Serra da Fartura were found at the base of the ears of their hosts, whereas in Itatiaia they were found at the base of the tail, a new site for T. caecata. All specimens of Tunga were determined to belong to the "caecata group" based on the characteristics of the neosoma (globose with invaginated head), but the identification of some distended gravid females to species was not possible.
The rodent A. montensis was the host with the largest number of flea species collected. It is possible that the large number of rodents collected influenced this result.
To Ibama for providing the collecting permits. Dr Helena de Godoy Bergallo (Uerj) helped the identification of the hosts; Dr Pedro Marcos Linardi (UFMG) helped the identification of the fleas; Tatiana TL Ribeiro, Renato Mangolim (Uerj), Adriano J Dall'Olio, Adriano Ronconi, and Marcos Ivo (Fundação de Ensino Otávio Bastos) helped during the field work.
Barros-Battesti DM, Arzua M 1997. Geographical distribution by biomes of some marsupial Siphonaptera from the state of Paraná. Mem Inst Oswaldo Cruz 87: 291-296.
Barros DM, Linardi PM, Botelho JR 1993. Ectoparasites of some wild rodents from Paraná state, Brazil. J Med Entomol 30: 1068-1070.
Barros-Battesti DM, Arzua M, Linardi PM, Botelho JR, Sbalqueiro IJ 1998. Interrelationship between ectoparasi-tes and wild rodents from Tijucas do Sul, state of Paraná, Brazil. Mem Inst Oswaldo Cruz 93: 719-725.
Bicho CL, Ribeiro PB, Nascimento AP, Sbalqueiro IJ 1999. Siphonaptera de roedores silvestres em Guaraqueçaba, Paraná, Brasil. Entomol y Vect 6: 53-62.
Botelho JR, Linardi PM 1996. Interrelações entre ectoparasitos e roedores em ambiente silvestre e urbano de Belo Horizonte, Minas Gerais, Brasil. Revta Bras Entomol 40: 425-430.
Botelho JR, Linardi PM, Williams P, Nagem RL 1981. Alguns hospedeiros reais de ectoparasitos do município de Caratinga, Minas Gerais, Brasil. Mem Inst Oswaldo Cruz 76: 57-59.
Bossi DEP, Linhares AX, Bergallo HG 2001. Parasitic arthropods of some wild rodents from Juréia-Itatins Ecological Station, state of São Paulo, Brazil. Mem Inst Oswaldo Cruz 97: 959-963.
Bush AO, Lafferty KD, Lotz JM, Shostak AW 1997. Parasitology meets ecology on its own terms: Margolis et al. revisited. J Parasitol 83: 575-583.
Carvalho RW, Serra-Freire NM, Linardi PM, Almeida AB, Costa JN 2001. Small rodents fleas from de bubonic plague focus located in the Serra dos Órgãos mountain range, state of Rio de Janeiro, Brazil. Mem Inst Oswaldo Cruz 96: 603-609.
Gettinger D 1992. Host specificity of Laelaps (Acari: Laelapidae) in Central Brazil. J Med Entomol 29: 71-77.
Gomes AC 1969. Pulgas colhidas em residências e sobre pequenos animais de algumas áreas do Brasil. Revta Bras Malariol Doen Trop 21: 775-780.
Guimarães LR 1972. Contribuição à epidemiologia da peste endêmica no Nordeste do Brasil e estado da Bahia. Estudo das pulgas encontradas nessa região. Revta Bras Malariol Doen Trop 22:177-181.
Guitton N, Araújo Filho NA, Sherlock IA 1986. Ectoparasitos de roedores e marsupiais no ambiente silvestre da Ilha Grande, estado do Rio de Janeiro, Brasil. Mem Inst Oswaldo Cruz 81: 233-234.
Johnson PT 1957. A classification of the Siphonaptera of South America with descriptions of new species. Mem Ent Soc Wash 5: 1-290.
Linardi PM 1977. Relações pulgas/roedores observadas nos municípios de Salesópolis e Itapetininga, SP. Bol Museu Hist Nat UFMG Zool 23: 1-25.
Linardi PM 1979. Sobre algumas espécies de Rhopalopsyllidae (Siphonaptera) sulamericanas integrantes do "complexo pradoi". Rev Bras Ent 23: 99-106.
Linardi PM 1985. Dados complementares sobre hospedeiros de sifonápteros ropalopsilinos. Revta Bras Biol 45: 73-78.
Linardi PM, Guimarães LR 1993. Systematic review of genera and subgenera of Rhopalopsyllinae (Siphonaptera: Rho-palopsyllidae) by phenetic and cladistic methods. J Med Entomol 30: 161-170.
Linardi PM, Guimarães LR 2000. Sifonápteros do Brasil, Museu de Zoologia, USP/Fapesp, São Paulo, 291 pp.
Linardi PM, Botelho JR, Neves DP, Cunha HC 1984. Sobre alguns ectoparasitos de roedores silvestres de Belo Hori-zonte. Rev Bras Biol 44: 215-219.
Linardi PM, Teixeira VP, Botelho JR, Ribeiro LS 1987. Ectoparasitos de roedores em ambientes silvestres do município de Juiz de Fora. Minas Gerais. Mem Inst Oswal-do Cruz 82: 137-139.
Monteiro Filho, ELA 1987. Biologia Reprodutiva e Espaço Domiciliar de Didelphis albiventris em uma Área Perturbada na Região de Campinas, Estado de São Paulo, MSc Thesis, State University of Campinas, Brazil.