Ectoparasites from small mammals from the Cerrado region in the Minas Gerais state, Brazil

Besides being an important component of the ecosystem and part of biodiversity, some ectoparasites of small mammals have economic and public health importance

The most common ectoparasites that infest small mammals belong mainly to the group of Ixodida (Ixodidae and Argasidae), Gamasida (Laelapidae and Macronyssidae), in the subclass of Acari; Siphonaptera (Rhopalopsyllidae) and Phthiraptera (Amblycera, Ischnocera and Hoplopleuridae) in the class of Insecta (Nieri-Bastos et al. 2004). Besides being an important component of the ecosystem and part of biodiversity, some of these ectoparasites have an extreme epidemiological importance as they act as vectors and host of various pathogenic for humans microorganisms causing world known serious diseases such as murine typhus, bubonic plague, tularemia or Lyme disease (Baker & Wharton 1952, Abel et al. 2000. Although the ectoparasites of small mammals have ecologic, economic and public health importance, there are not many studies in Brazil (Graciolli et al. 2006). Most of the researches are concentrated on the southern and southeastern regions the country in oristic parts of the cities situated close to human settlements, where these ectoparasites are linked with human diseases , Guedes et al. 2005, Barros-Batestti et al. 2006, Guglielmone et al. 2006. The studies are limited to the records of new species and taxonomical revisions, and there is a lack of researches concerning ecological relations between parasites and their hosts what has an extreme importance to understand the epidemiology of various diseases (Marshall 1981, Graciolli & Aguiar 2002, Nava et al. 2003, Dias et al. 2009, Ribeiro et al. 2010.
The objective of the present work was to use ecological parameters of small mammals ectoparasite infestation to analyze relations between parasites and their mammalian hosts in the National Park Serra do Cipó, Minas Gerais State, Brazil.

METHODOLOGY
Small mammals captures were conducted in the National Park Serra do Cipó. The Park was created in 1984 and encompass an area of about 33800ha. It is located among Jaboticatubas, Santana do Riacho, Morro do Pilar and Itambé do Mato Dentro Municipalities in the state of Minas Gerais, Southeast Brazil (central point of the Park 19º20'S, 43º38'W, elevation ranging from 900 to 1600m). The Park is located in the Cerrado biome, savannah-like ecosystem which one of the most important biodiversity hotspots in Brazil, and is characterised by an enormous range of plant and animal biodiversity (Myers et al. 2000, Eterovick & Sazima 2004. The climate is humid subtropical with mild and rainy summers (annual rainfall ca. 1,500mm, October-March) and dry winters April-September with a mean annual temperature between 17 and 18,5ºC (Galvão & Nimer 1965).
The sampled areas were selected based on anterior study (Câmara et al. 2003). Five trails of 100 m long were chosen in ve distinct areas of the Park. Small mammals were captured between April and September 2007, with 20 "live-trap" Shermann cages during four consecutive days each month, totalizing 600 hours/night of trapping e ort. The bait was made of mixture of banana, oil from canned sardine, oat our, peanut butter and maize and it was replaced every day. The captured rodents and marsupials were anesthetized with a piece of cotton soaked with sulfuric ether according to Barros-Battesti et al. (1998). We identi ed captured mammals to the species, banded, and examined them for the presence of ectoparasites, which were all collected for later identi cation. Nomenclature and identi cation of mammals follows Wilson & Reeder (2005). Thereafter, all mammals were released at the same site. The ectoparasites were collected by brushing and combing over the white basin and immediately preserved in 70% ethanol for further identi cation. Tweezers were used when necessary. In the laboratory mites and eas were mounted on slides according to the conventional techniques for Acari (Flechtmann 1975) and Siphonaptera (Linardi & Guimarães 2000). Mites were identi ed following Fonseca (1935Fonseca ( /36, 1939, eas according to Linardi and Guimarães (2000), ticks according to , lice following Werneck (1942) and Ferris (1951), Dipteran larvae according to Baird (1983), Manrique-Saide et al. (2000) and Slansky (2006).
For each host species the following parameters were calculated: mean abundance of ectoparasites (MA-total number of individuals of a particular species of parasite in a sample of a particular host species / total number of hosts of this specie), the percentage (P) of mammal individuals infested with ectoparasites (number of infested mammals / number of examined mammals x 100, for each mammal species) and the ectoparasites ratio between females and males (number of female ectoparasites of a particular species/ number of male ectoparasites of the same species) as described previously (Botelho 1990, Bush et al. 1997).
This study was licensed by the environmental agency responsible (Instituto Brasileiro de Meio Ambiente e Recursos Naturais Renováveis -IBAMA), with process number 10204-1.

RESULTS
A total of 95 mammals were captured, representing nine species. Six of these species belong to the order Rodentia: Cerradomys sub avus (Weksler 2006)  Due to lack of speci c literature and identi cation keys based on morphology not all collected ectoparasites were identi ed to the level of species thus ectoparasites Macronyssidae were identi ed only to the level of family. The ectoparasites from the family Oestridae and Ixodida were identi ed only to the level of genera Cuterebra sp. and Amblyomma sp., respectively. The detailed results of ectoparasites found during this work are shown in the Table 1 and 2.
The sex ratio was calculated for mites form the family Laelapidae, the eas (Siphonaptera) and for lice (Phthiraptera). Others ectoparasites species were or collected in the immature stages or identi ed only to the level of family or genera hence were not included into analyses. The proportion of sex ratio of females to males was varying among species being higher for mites than for lice and eas. In the case of mites, there was small numbers of males found or simply females were observed (Table 3).

DISCUSSION
The present study reports ectoparasites species parasitizing wild small mammals in a region of the Cerrado, Minas Gerais, southeast Brazil. All presented here records of ectoparasites are novel for the Park. The highest numbers of ectoparasites collected during this work belong to the order Gamasida what is in agreement with previous researches , Botelho 1978, Barros-Battesti et al. 1998, Bossi et al. 2002, Nava et al. 2003, Nieri-Bastos et al. 2004. Especially the mites from the family Laelapidae are commonly found on small mammals. L. paulistanensis was found on all examined R. mastacalis and O. nigripes and on 2% of C. sub avus. According to literature L. paulistanensis parasitize commonly rodents from the genera Oligoryzomys and Oryzomys (Barros-Battesti et al. 1998, Botelho 1990) and have been already observed parasitizing all mentioned above species , Nava et al. 2003. L. differens was found on 83,3% of C. sub avus and on 33,3% of T. apereoides. C. sub avus was already observed parasitized by this mite by Botelho (1990), but T. apereoides is shown for the rst time to serve as a host for this species. L. manguinhosi was found on 82% of N. squamipes and on 25% of N. lasiurus. This mite is frequently found on small semiaquatic rodents form the genera Nectomys and Holochilus (Fonseca 1936, Martins-Hatano et al. 2002, and is occasionally found on other species of rodents and birds (Fonseca & Trindade 1958, Gettinger 1992. The species T. latreventralis was found on all examined T. apereoides, and on two species of marsupials G. agilis and D. albiventris with the prevalence 33,3 and 20% respectively what is in concordance with the previous results . In the present study G. goyanensis was found exclusively on N. squamipes (91%). This result is similar with the studies of Fonseca (1939) and Botelho (1978), and the latter de ned N. squamipes as the main host of this parasite. Esbérard et al. (2005), Gettinger (1987 and Martins-Hatano et al. (2002) found high levels of infestation of N. squamipes as well, however, they found additionally Oxymycterus dasytrichus (Schinz 1821) and Trinomys dimidiatus (Günther 1877) being parasitized by G. goyanensis. G. vitzthumi was found exclusively on C. sub avus (89,6%). Its narrow preference to C. sub avus, de ned by Botelho (1990) as the main host for this parasite, was also found in other resarches , Gettinger 1987, Botelho 1990). In the contrast, the mite A. fahrenholzi showed the low level of preference to host; it was found on ve from all nine species captured (C.   (Nava et al. 2003). To our knowledge, this is the rst time that N. squamipes and T. apereoides were found parasitized by this mite.
Various larvae of ticks form the genera Amblyomma were found on C. sub avus (23%), N. squamipes (18,2%), T. apereoides (44,4%), R. mastacalis (14%) and on D. albiventris (40%). Small rodents form these genera are rarely found infested by the ticks Amblyomma (Figueiredo et al. 1999, Brum et al. 2003, Muller et al. 2005. On the other hand, there are records of frequent parasitism of ticks form the genera Ixodes (Barros-Battesti et al. 2000;Evans et al. 2000). Because most of the ≈30 Amblyomma species currently known to occur in Brazil have their larval stages still not described (Guglielmone et al. 2003); consequently, there is no su cient literature for a proper identi cation of Brazilian Amblyomma larvae and records are limited to genus level (Amblyomma sp.).
Among the order Phthiraptera three species of lice were identi ed: C. cercomydis and E. lenti lenti were found exclusively on T. apereoides with the prevalence 55,5 and 66,6% respectively and H. imparata only on the N. lasiurus (50%). All species of lice were collected on their host-type, according to Ferris (1951), Werneck (1942) and Linardi et al. (1984).
Two species of Siphonaptera were found in the present study; T. penetrans was found on D. albiventris (1 adult) and P. tripus were found on C. sub avus (1 adult) and on N. lasiurus (7 adults). Such a low number of eas can be related with the fact that these insects parasite their host in the adult stage and most of the time they spent hiding rather in the nest of animals (Linardi et al. 1997). Beside this fact the eas are know to leave its host in the stressful situation what happens during animals captures (Botelho 1990). According to literature, P. tripus is known as a frequent parasite of Akodon sp. (Botelho 1978, Barros-Battesti et al. 1998 and Necromys sp. (Botelho 1990). However, N. lasiurus and C. sub avus have been also previously found parasitized by this species (Botelho 1990). The ea T. penetrans is found on various types of mammals such as carnivores, ungulates and xenarthrans (Linardi & Guimarães 2000). The opossums D. albiventris has been already observed being parasitized by this specie and the fact that only one individual was found might be due to low number of captured hosts.
Apart from discussed ectoparasites, the larvae of Cuterebra sp. (Diptera: Oestridae) were found on 14,6% of C. sub avus. This y is endemic to the New World and is and opportunist endoparasite found mainly on mammals (Guimarães & Papavero 1999). Its larvae leave in the subcutaneous tissues of their hosts for few weeks feeding on body breathing through the hole in the skin of their hosts (Slansky 2006). The larvae from the genus Cuterebra have been previously found on rats and on rabbits (lagomorfos) (Manrique-Saide et al. 2000), however here we relate the C. sub avus serving as host for this parasite for the rst time.
The observed di erences in the number of females and males among mites are noteworthy; the females were more abundant and in some cases no male was seen what is in concordance with other studies (Fonseca 1939, Botelho 1978, Linardi et al. 1985, Lopes 1989, Botelho 1990). Numerous factors can explain this phenomenon. One of them is the fact that the mites have capacity to reproduce by parthenogenesis (Baker & Wharton 1952), the mite males live also shorter than females (Willians & Kershaw 1977). In various species from the family Laelapidae males and immature forms live inside the nest of host parasitizing him only for a short time, while females need to feed with higher frequency to produce eggs after blood meal, thus they accompany their host most of the time (Flechtmann 1975).
Among eas the ratio of females to males was observed similar to other studies (Cerqueira 1975, Linardi 1977, Cerqueira & Linardi 1977, Lopes 1989, Botelho 1990). Following the theory of Linardi & Nagem (1972), slight variation between the number of females and males is a result of the di erence of longevity reproductive cycles.
All presented here records of mammals ectoparasites are novel for the Park. Several species of ectoparasites have great ecological importance as part of the community structure of small mammals, which can reduce competitive ability and increase vulnerability of hosts to predators; evolutionary importance, o ering great information related to the dynamics of past populations, which currently are extinct but preserve features closely related due to the co-evolution of host-parasite, and also environmental importance, since some species can serve as e ective indicators of environmental quality (Komplen et al. 1996, Morand et al. 2006).
Additionally, ticks, mites and eas are among the most important vectors of pathogens that cause diseases in human and domestic and wild animals (Jongejan & Uilenberg 2004). For example, ticks Amblyomma are principal vectors of bacterium Rickettsia rickettsii, the agent of Brazilian Spotted Fever in the South America (Labruna 2009) and mites Macronyssidae are suspected to be involved in the circulation of the bacteria in the natural environment (Reeves et al. 2006(Reeves et al. , 2007. Therefore, further studies concerning the ectoparasites and pathogens that they may transmit are urgently needed.