Trypanosoma (Megatrypanum) lainsoni n. sp. from Mesomys hispidus (Rodentia: Echimyidae) in Brazil: trypomastigotes described from experimentally infected laboratory mice

We report the detection, isolation and description of Trypanosoma (Megatrypanum) lainsoni n. sp. from a caviomorph rodent, Mesomys hispidus (Rodentia: Echimyidae), obtained in the Rio Negro region of the state of Amazonas, in northern Brazil. Laboratory-bred white mice (Mus musculus) and rats (Rattus rattus) were inoculated with large numbers of culture forms by intraperitoneal route, and trypomastigotes appeared in their blood 3–8 days post-inoculation. One single epimastigote was also found in Mus musculus. Similar attempts to infect Rattus norvegicus, hamsters (Mesocricetus auratus), the opossum Didelphis marsupialis, the anteater Tamandua tetradactyla and triatomine bugs were unsuccessful, following six months of observations and microscopic examinations of blood films and blood cultures. As we have found no previous record of a Trypanosoma (Megatrypanum) species naturally infecting a member of the family Echimyidae, or any other caviomorph rodent, we conclude that this is the first time such an infection has been reported. The new species is unusual in the subgenus for its infectivity to laboratory mice.


Introduction
In the last few decades, the number of recorded species of trypanosomes has increased considerably and the medical and veterinary importance of these parasites has been discussed in many textbooks of protozoology, parasitology and related subjects.
The subgenus Megatrypanum is a somewhat heterogeneous group of large trypanosomes. The internal structure of members of this genus is characterised by the kinetoplast which is *Corresponding author: rdnaiff@yahoo.com.br Roberto Daibes Naiff -urn:lsid:zoobank.org:author:732C8DAA-7301-41DD-99E4-3348A69A8A98 Toby Vincent Barrett -urn www.parasite-journal.org typically placed far from the posterior end of the parasite and near the nucleus [13]. When blood forms can be found in stained blood films, two morphologically different trypomastigotes can often be seen; one is large, but slender and the other large and broad. Frequently, however, the infection in the natural host can only be detected following haemoculture. All species of Megatrypanum are normally non-pathogenic in their natural hosts [6]. Data regarding reproduction in the mammalian host, vectors and life cycles of members of the subgenus Megatrypanum are fragmentary or completely lacking for most species, and many of these trypanosomes appear to be restricted to individual species or genera of their mammalian hosts [5,11]. The great majority of these trypanosomes are strictly host-specific in their natural environment [6].
The species described here is the first species of the subgenus Megatrypanum [5] to be described in a caviomorph rodent. It was isolated, in NNN culture medium, from a spiny tree-rat obtained in the mid-Rio Negro region of Amazonas State, northern Brazil.

Material and methods
An adult male specimen of the spiny tree-rat, Mesomys hispidus, was killed, by a collector of ''piassaba'', in the crown of the palm tree Leopoldinia piassaba Wallace, in lowland rainforest near the Igarapé Japaumerí, a stream flowing into the River Padauarí, northern Amazonas State, Brazil. The rat was necropsied under field conditions after asepsis with iodated alcohol, and heart blood then inoculated into seven tubes of diphasic NNN blood-agar culture medium [10]; two thin blood films were fixed in absolute methyl alcohol and stained by Giemsa's method [2]. Fragments of skin tissue from the nose, ears and base of the tail were triturated in saline solution (0.9% NaCl) containing penicillin (200 U/mL) and streptomycin (0.312 mg), incubated at room temperature for 2 h and then used to inoculate three-month-old hamsters. Two hamsters were inoculated intradermally with the suspension into the nose and rear paws, and a similar suspension of triturated liver and spleen was inoculated intradermally and intraperitoneally into two other hamsters. Following successful isolation of a trypanosome in the NNN culture medium (first sub-passage), the culture forms were inoculated into 21-day-old laboratory mice by the intraperitoneal route. Measurements of the trypomastigotes (Table 1) follow Hoare [6]. All measurements are in lm.
Trypanosoma (Megatrypanum) lainsoni n. sp.  middle of body with a nuclear index (NI) of 0.7-1.4, but mainly 0.9-1.1; kinetoplast closer to nucleus than to posterior end with kinetoplastic index (KI) of 2.2-6.0, but mainly 0.9-1.1, and marginal. Other measurements are given in Table 1. A single bloodstream epimastigote was detected, with its nucleus in a strongly posterior position and NI of 0.4 ( Figure 1).

Animal inoculation and culture
Following periodic examination over a period of six months none of the hamsters inoculated with tissues of Mesomys hispidus showed any evidence of infection with Leishmania at the sites of inoculation, or the presence of trypanosomes in their blood. This clearly indicated the absence of Leishmania infection in M. hispidus and that hamsters could not be infected by T. (M.) lainsoni n. sp.
One of the seven original cultures of the spiny rat's blood produced good epimastigote growth of a trypanosome; four tubes were contaminated by bacteria and fungi and although two tubes escaped such contamination they had not isolated the trypanosome. Three to eight days post-inoculation (p.i.) of the culture forms, mice (Mus musculus) showed an average of 5-6 trypomastigotes per microscope field (magnification · 40) in fresh preparations, and at five months p.i. parasitaemia was sub-patent microscopically but still demonstrable in some mice by haemoculture. One single epimastigote was also found in mice.
Inoculation of other animals with flagellates of 31-day-old culture forms gave the following results after microscopic examination of fresh blood preparations or haemoculture. One 30-day-old Rattus rattus showed 1-3 trypomastigotes per field on day 3 p.i. By haemoculture two 16-day-old Rattus norvegicus were negative, as were one juvenile Tamandua tetradactyla and two young Didelphis marsupialis.

Xenodiagnosis using triatomine bugs
Twenty Rhodnius pictipes, 20 R. robustus and 20 R. brethesi were fed on mice showing abundant trypomastigotes in their peripheral blood and examined after 3-6 weeks. All failed to become infected.

Discussion
We have been unable to find any previous record of a trypanosome of the subgenus Megatrypanum in echimyid rodents or any other caviomorphs (The Zoological Record 1978 to 2013, Scopus, PubMed, ISI).
Species of the subgenus recorded in neotropical rodents include T. phyllotis [4], from Phyllotis spp (Cricetidae) in semi-arid regions of western Peru; T. amileari [7], from Oligoryzomys eliurus [11] (Cricetidae) in Brazil (northern Goiás; now the State of Tocantins), T. rochasilvai [10] from Oryzomys sp. (Cricetidae) in the State of São Paulo, Brazil; and T. zeledoni [3], from Liomys salvini (Heteromyidae). With the exception of T. phyllotis, attempts to cultivate these trypanosomes in NNN blood-agar medium or to infect mice and rats have been unsuccessful and, apart from T. amilcari (L = 32-40) all are, on average, much larger than T. lainsoni. T. phyllotis was cultured in NNN medium and infected the sand fly Lutzomyia noguchi and laboratory rats of up to six days old. The mean length of T. phyllotis was given as 47, while that of T. lainsoni was only 33.4; the kinetoplast of the former was smaller than that of the latter and the nucleus more rounded. The kinetoplast of T. amileari was shown to be closer to the nucleus (KN 2-6, KI 5) than that of T. lainsoni; T. rochasilvai was very much larger (L = 50-73 vs. 28-37). The length of T. zeledoni was given as 36-56; the nucleus is placed more strongly anterior than that of T. lainsoni (NI = 1.4-1.5 vs. 0.67-1.36) and the kinetoplast is rounded and sub-marginal rather than marginal: finally, the flagellum is relatively short compared with that of T. lainsoni.
As is the case for many other members of the subgenus Megatrypanum, the above criteria for separating species may seem to be somewhat tenuous in the absence of direct comparison under controlled conditions [12]. Species of this subgenus, however, are in general host-restricted [5], and the fact that our isolate is from a distinct mammalian suborder (Caviomorpha) has influenced our decision to accord its specific status. We also feel that it is useful to make a name available for a parasite which is of potential use as a new trypanosome laboratory model.
Our failure to experimentally infect three species of phlebotomine sand flies does not necessarily preclude Lutzomyia species as vectors of T. (M.) lainsoni. Apart from T. phyllotis, other mammalian T. (Megatrypanum) species suspected to have a phlebotomine vector are T. leonidasdeanei [14], probably transmitted by Lutzomyia vespertilionis, and T. freitasi [9], found in a wild-caught Lutzomyia claustrei [8].
Epimastigotes thought to belong to a T. (Megatrypanum) species were observed in lymphoid tissues of the armadillo Dasypus novemcinctus [1], but our Figure 1 of T. lainsoni is probably the first record of the epimastigote stage of a T. (Megatrypanum) species, other than that of T. theileri, in mammalian blood. The elongated form of the nucleus and its pattern of staining are suggestive of an early stage of division, and the structure just anterior to the nucleus looks like a small, second kinetoplast and associated flagellum.
The potential of Aristides Herrer's Phyllotis/Lutzomyia noguchii model appears to have been neglected by later students of Megatrypanum, and we suggest that the behaviour of T. lainsoni in mice could provide a rewarding area of study of these elegant and phylogenetically ancient trypanosomes.