The Infection Rates of Trypanosomes in Squirrel Monkeys at Two Sites in the Brazilian Amazon

A study was conducted to determine the prevalence of natural infections by trypanosome species in squirrel monkeys: Saimiri sciureus (Linnaeus) and Saimiri ustus (Geoffroy) caught respectively near 2 hydroelectric plants: Balbina, in the State of Amazonas, and Samuel, in the State of Rondônia, Brazil. A total of 165 squirrel monkeys were examined by thick and thin blood smears (BS), haemocultures and xenodiagnosis: 112 monkeys, 67.9%,(being 52.7% with mix infections) were positive to trypanosomes. Four species of trypanosomes were found in monkeys from the 2 areas: Trypanosoma (Tejeraia) rangeli Tejera or T. rangeli-like parasites in 58 squirrel monkeys (35.2%), Trypanosoma (Megatrypanum) minasense Chagas in 55 (33.3%), Trypanosoma (Herpetosoma) saimirii Rodhain or T. saimirii-like parasites in 53 (32.1%) and Trypanosoma (Schizotrypanum) cruzi Chagas in 17 (10.3%). As T. saimirii resembles T. minasense in blood-stream trypomastigotes and T. rangeli in cultural forms and in this survey almost all monke ys presenting trypanosomes morphologically indistinguishable from T. saimirii and/or T. minasense in BS were found through xenodiagnosis and/or haemoculture to be infected by T. rangeli, we suggest that the validity of T. saimirii needs to be evaluated.

Squirrel monkeys are unusual among laboratory animals; many are captured in forests and not bred for research. Wild-trapped Saimiri monkeys are often naturally infected with pathogens. Sullivan et al. (1993) pointed out that such naturally acquired infections in experimental animals are potential sources of accidental transmission to handlers and uninfected monkeys and may introduce confounding variables into otherwise wellplanned and well-carried out studies. This paper describes the high incidence and diversity of natural infections by trypanosomes in squirrel monkeys from the Brazilian Amazon basin.

MATERIALS AND METHODS
The squirrel monkeys were captured in 2 areas, near hydroelectric plants built on rivers on each side of the Amazon basin: Balbina, on the Uatumã River, State of Amazonas (1°55'S 59°28'W) and Samuel, on the Jamari River, State of Rondônia (8°10'S 62°29'W), Brazil. After the construction of both dams, primates were captured in the forests of the dam reservoir flooded plains. A description of these 2 sites was given by Lourenço-de-Oliveira and Deane (1995).
From June 1988 to February 1990 blood samples were obtained by venipuncture from 165 squirrel monkeys belonging to 2 species: 70 S. sciureus captured in Balbina and 95 S. ustus from Samuel. The nomenclature and systematics of the primates are according to Hershkovitz (1984). Giemsa-stained thick and thin blood smears (BS) were examined for the presence of trypanosomes. Haemocultures were made from the blood of 24 S. sciureus and 14 S. ustus in liver infusion tryptose (LIT) over Novy, McNeal and Nicolle medium (NNN) slopes supplemented with 10% bovine serum and penicillin 200 UI/ml, maintained at 27°C for 7 days. The cultures were then microscopically examined weekly for 1 month. The isolated parasites were either subcultured into the same cultural media or cryopreserved in liquid nitrogen.
Twenty two S. sciureus and 14 S. ustus, were submitted to xenodiagnosis using 6 nymphs of 3rd and 4th instar of Rhodnius prolixus which were allowed to feed to repletion on each monkey. Fresh smears of faeces and hemolymph of bugs were examined around 30 days after their blood meals. Bugs harboring parasites in their hemolymph were fed on uninfected mice. After 15 days of fasting these bugs fed up on other 20 days-old albino mice. This procedure tended the triatomid's lifespan and allowed the development of the parasites for a longer period of time in the insect, increasing the probability of detecting salivary gland infections and transmission to mice. Fresh blood samples of the mice were examined from 1 to 15 days, when haemoculture (NNN + LIT) was performed.
The salivary glands and gut contents of bugs whose faeces and hemolymph were negative were examined 30 days after xenodiagnosis. Trypanosomes detected in faeces, salivary glands and gut contents of bugs were Giemsa-stained as well as inoculated intraperitoneally (IP) or subcutaneously (SC) into albino mice, whose blood was examined weekly (fresh and Giemsa-stained smears) 30 or 45 days after inoculation. Then, all mice were sacrificed and their blood haemocultured (NNN + LIT).
Samples of 0.6 µl of the haemoculture from 10 squirrel monkeys were injected in the hemocoel of adult R. prolixus, which were fed up on 20 daysold albino mice after 30 days and subsequently examined for the search of flagellates in their salivary glands. Blood samples (fresh and Giemsastained smears) of the mice were examined from 1 to 15 days, when haemoculture (NNN + LIT) was performed.
The identification of trypanosome species was done by both morphological analyses and biological parameters (Hoare 1972, D'Alessandro 1976, Miles et al. 1983, Steindel et al. 1991) that included comparisons with standard strains of T. cruzi and T. rangeli (stock R1625 -CDC) for development in arthropod vectors, in culture media and in experimental vertebrate hosts.
The same 4 Trypanosoma species were detected in squirrel monkeys from both areas: T. minasense, T. rangeli, T. cruzi and T. saimirii-like. In 8 animals the identification of deformed blood stream trypomastigotes was not possible.
Trypanosomes were isolated in haemoculture from 19 S. sciureus and all 14 S. ustus, which led to the detection and identification of T. rangeli and T. cruzi. T. rangeli or T. rangeli-like parasites was isolated from haemoculture from 14 S. ustus; T. cruzi was simultaneously isolated from 3 of them (Table).
T. rangeli was isolated in haemoculture from 8 S. sciureus, 1 of which had a mixed infection with T. cruzi. These T. rangeli stocks were identical in all pertinent morphological and biological parameters with standard strains. In the haemoculture of 7 other S. sciureus, flagellates morphologically indistinguishable from T. rangeli were also found, but these could not be compared biologically to standard stocks. The long and slender epimastigotes among other cultural forms, found in the haemoculture and in the gut of bugs used in xenodiagnosis from T. saimirii-like infected squirrel monkeys (as determined by BS) were found to resemble to those of T. rangeli, described by several workers (e.g. Herbig-Sandreuter 1957, Hoare 1972), but distinct from those of T. minasense (Ziccardi et al. 1996). Actually, the cultural forms of T. saimirii described by Rodhain (1937) and Deane and Damasceno (1961) also resemble those of T. rangeli.
Following the development in the bug of flagellates isolated from T. saimirii-like infected monkeys, we verified that hemolymph and salivary glands became infected in xenodiagnosis of squirrel monkeys numbered 105, 133, 137, 138, 139, 140, 147, 167 and 221. In some of these infections, the glands became packed with metatrypanosomes and the parasite was transmitted to mice through biting in half of the cases. Besides, after the injection of 0.6 µl of haemoculture from squirrel monkeys numbered 113, 119, 120, 141 and 217 in the hemocoel of adult of R. prolixus, the flagellates invaded their salivary glands and T. rangeli was transmitted to mice through their biting (except number 141).
Although the identification of parasites as T. rangeli by xenodiagnosis or haemoculture in squirrel monkeys numbered 108, 109, 111, 123, 125, 134 and 220 was not possible by biological criteria (such as, development up to metatrypanosomes in the salivary glands of bugs and the transmission to mice through biting) the flagellates developing in both the bugs and haemoculture were morphologically indistinguishable from T. rangeli.
The flagellates from faeces and/or gut contents of triatomines infected with all T. rangeli isolates were not infective to mice.

DISCUSSION
In the Americas, many animals have been examined as reservoirs of microorganisms pathogenic to man. During these surveys, many species of parasites, including trypanosomes, have been found. Except for T. cruzi and T. rangeli, no other trypanosome species is considered potentially pathogenic or infective to humans. However, other trypanosomes may be as frequent as T. cruzi and T. rangeli in certain mammals, such as squirrel monkeys (Table).
Approximately 32% of the squirrel monkey from both study areas were infected with T. saimirii-like trypanosomes. This prevalence is much higher than that found by Deane and Damasceno (1961) in S. sciureus from the State of Pará, Brazil (1.38%). Lourenço-de-Oliveira (1988) and Lourenço-de-Oliveira et al. (1991) reported respectively 66.7% and 28.8% of a few S. sciureus previously examined from Balbina infected with T. saimirii-like parasites, i.e., harboring blood trypomastigotes indistinguishable from those described by Rodhain (1937) and Deane and Damasceno (1961). Lourenço-de-Oliveira et al. (1991) observed that the flagellates developing in haemocultures and triatomine bugs fed on infected squirrel monkeys were indistinguishable from T. rangeli, and that in some of the insects the hemolymph and salivary glands became infected.
T. minasense has been recorded from 12% to 23.1% of squirrel monkeys from the Amazon, Brazil (Deane et al. 1989, Lourenço-de-Oliveira et al. 1991. Dunn (1968), in a survey of hemoparasites of Saimiri from Colombia and Peru, determined that the prevalence of T. minasense and similar trypanosomes was respectively 36.8% and 47.6%. However, this author considered T. minasense a synonym for a group of simian trypanosomes, some of which are usually considered as distinct species, such as T. devei and T. mycetae (Deane & Damasceno 1961, Hoare 1972, Lourenço-de-Oliveira 1988, Deane et al. 1989, Lourenço-de-Oliveira et al. 1991. T. rangeli and T. rangeli-like parasite was the most frequent trypanosome among the squirrel monkeys from Balbina and Samuel (Table). However, if the haemoculture and xenodiagnosis were performed on all animals a higher incidence of infection by T. rangeli would certainly be found.
T. rangeli, or T. rangeli-like parasites, are the most frequently reported trypanosome in squirrel monkeys from Colombia, Peru, Panama, Bolivia and Brazil (Dunn et al. 1963, Ayala 1964, Marinkelle 1966, Baker 1972, D'Alessandro et al. 1986, Sullivan et al. 1993. Most authors usually identify the majority of the blood trypomastigotes detected in squirrel monkeys as T. rangeli and T. rangeli-like, or T. minasense, but almost never as T. saimirii, due to the fact that T. saimirii is poorly characterized. T. saimirii, as described by Rodhain (1941) and other authors (e.g. Deane & Damasceno 1961), resembles T. minasense in blood-stream trypomastigotes (although narrower, shorter, with its kinetoplast nearer to the posterior end of body that is gradually sharpened to a point, unlike T. minasense). Besides, T. saimirii develops in the gut of triatomine bugs, without invading the hemocoel and salivary glands, as T. rangeli generally does. On the other hand, T. minasense do not develop in triatomine bugs and its vector is still unknown (Rodhain 1941, Dias & Campos-Seabra 1943, Deane & Damasceno 1961. In this survey almost all squirrel monkeys presenting trypanosomes morphologically indistin-guishable from T. saimirii and/or T. minasense in BS were finally found through xenodiagnosis and/ or haemoculture to be infected by T. rangeli (or T. rangeli-like). Only 2 animals (numbered 112 and 220) had negative haemoculture and/or xenodiagnosis.
The data presented here strengthen the hypothesis of Deane and Damasceno (1961) that the flagellates developing in the gut of bugs fed on squirrel monkeys infected with T. saimirii may belong to another Trypanosoma species undetected by BS. The probability of the squirrel monkeys examined by Rodhain (1937Rodhain ( , 1941 and Deane and Damasceno (1961) being infected with mixed T. rangeli in subpatent parasitaemia must be considered and the validity of T. saimirii needs to be evaluated.