Trypanosoma cruzi in the sylvatic environment: distinct transmission cycles involving two sympatric marsupials

Trypanosoma cruzi in the sylvatic environment: involving two sympatric marsupials

distinct

transmission

cycles

and Ana M. Jansenl Ana Paula Pinho’, Elisa Cupolillo*, Regina Helena Mangia ), Octavia Fernandes Depanments of ‘Protozoology, 21rnmunology, and ‘Tropical Medicine, Institute OswaMo Cruz, FIOCRUZ, RJ, Brazil Abstract Thirty-five specimens of Philanderfienata and 36 Didelphis marsupklis were captured in the same Atlantic forest area of Brazil between 1992 and 1994. Haemocultures showed that 50% of I? frenata and 60% of D. marsupialis were infected with Trypansoma cruzi. Biological, biochemical and molecular characterization of the isolates suggested 2 distinct transmission cycles of T. cruzi occurred between these 2 sympatric didelphids. The T. cruzi isolates could be distinguished according to their association with each marsupial species. Biochemical characterization (multilocus enzyme electrophoresis) revealed 15 zymodemes; more variability was observed among the I? jknata isolates than among the isolates from D. marsupialzs. The course ofnatural and experimental infection in D. marsupialis and I? fienata was different and suggested that D. marsupialis was more resistant to infection than I? fienata. In the studied area, I? frenata seems to be a more important reservoir of T. cruzi than D. marsupialis, since 40% of the characterized isolates from l? fienata belonged to the T. cruzi II group, which is associated with human infections. Keywords: Chagas disease, Trypanosomacruzi, zymodemes, Philanderfienata, LXdelphismarsupialis, Brazil Introduction Typanosoma cruzi, the aetiological agent of Chagas disease, circulates in both sylvatic and domestic environments. In the sylvatic environment, T. cruzi infects dozens of triatomine bugs (Reduviidae, Triatominae) and 7 mammalian orders. It is assumed that a domestic transmission cycle can be established as the result of the colonization or invasion of human dwellings by infected vectors or mammals (ZELED~N et al., 1970; BARRETO & RI~ErR0,1979). The population structure of T. cruzi is assumed to be basically clonal (TIBAYRENC et al., 1990, 1991), with significant inn-a-specific heterogeneity that has been evaluated by biological, biochemical and molecular methods (DVORAK, 1980; MOREL et al., 1980; ANDRADE et al.. 1983: RBAYRENC et al.. 1985. 1986. 1990,1991;‘k1~~~1&~~ & AYALA,~~~~; TIBA~RENC; 1995; SOUTO et al., 1996; FERNANDES et aZ.,1998). Although this subject is still controversial, some correlations between these markers and subpopulations of the parasite have been established. Three zymodemes, based on the electrophoretic profile of 6 enzymes, have been defined: Zl and 23, corresponding to the sylvatic transmission cycle, and 22, corresponding to the domestic cycle (MILES et al., 1977, 1978, 1980). Analysis of ribosomal ribonucleic acid (rRNA) genes and mini-exon repeats of hundreds of isolates derived from humans, triatomines and wild mammals has led to the definition of 2 distinct and phylogenetically distant lineages of T. cruzi, associated with domestic (lineage 1) and sylvatic (lineage 2) transmission cycles (SOVTO et al., 1996; ZINGALES et al., 1998). In a recent international meeting in Rio de Janeiro, Brazil, the main zymodemes and lineages were redefined as 2 groups termed T. cruzi II and T. cruzi I, corresponding to 22 (= lineage 1) and Zl (= lineage 2), respectively (ANONYMOUS, 1999). FERNANDES et al. (19991 have characterized 68 T. cruzi isolates obtained‘fiom’triatomines and 4 orders of sylvatic mammals by means of mini-exon gene and 24Sa rRNA gene tvning. All isolates derived from mimates (Leonto>ithe& ro&a) were classified as lineage 1 (T. cruziII), but all other mammals, except 2 opossums with mixed infections (lineages 1 and 2), harboured lineage 2 ( T. cruzi I). These findings indicate that different sylvatic cycles may involve different reservoir hosts (and vectors), which will influence the facility with which domestic Address for correspondence: Dr A. M. Jansen, Department of Protozoology, Institute Oswald0 Crux, Av. Brazil 4365, Manguinhos, Rio de Janeiro, RJ 2 1045-900, Brazil; fax +55 2 1 280 1589, e-mail [email protected]

transmission cycles becomes established. Some aspects of the sylvatic transmission cycles, such as the preferential association of the stocks of T. cruzi with different sylvatic reservoirs and the distribution and dynamics of the parasite and host populations, are still unknown. In this studv we investigated natural T. cruziinfections of LXdelphis &arsupialis &d Philander~ata (opossums; Marsuvialia, Didelvhidae). 2 svmvatric svecies cavtured in the same -sylvati, area & B&z& We characteriied T. cruzi isolates derived from these species using biological, molecular and biochemical markers. We also compared the course of experimental infections in both didelphid species. Material and Methods Marsupials Thirty-six I? frenata and 35 D. marsupialis were captured in baited TomahawkTM traps, in Caleme, Tere&polis, Rio de Janeiro State, Brazil (a secondary Atlantic rain forest area), between 1992 and 1994. No autochthonous Chagas disease has been reported in the locality. Captured marsupials were individually caged and fed on commercial dog food, fiuits and eggs. Uninfected animals were released. Natural infections Ten D. marsupialis and 10 I? fvenata, naturally infected with T. cruzi, were maintained for 4-30 months at the FIOCRUZ animal facilities and periodically bled under anaesthesia (KetamineTM, 13 mg/kg). Blood cultures in NNN medium with liver infusion overlay were prepared every 2 weeks during 2 months. Examination of the anal scent glands for parasites was performed by gentle manual squeezing, collection of the content and microscopical examination. Immunofluorescence assay was carried out as described previously (JANSEN et al., 1985). Fresh blood films were also prepared and examined. All isolates from I? fienata were numbered with the prefix C. Experimental infections and strains The following 4 T. cruzi strains were used, maintained in LIT medium and by serial passages in o&bred Swiss mice: MHOM/BR/OON~ T. cruzi III, isolated from a human ~~~~~~SILVA&NU~SEN~~~(~~~~);IOO/BR/OO/ F(T. cruzi I); MDID/BFU83/G-49 (T.cruzi I), isolated from a naturallv infected ovossum cavtured in Mirmel Pereira, Rio de-Janeiro in i983, and bIDIBRI9sC13( T. cruziI1). a recent isolate from I? frenatu cavtured in Teies6polis. ‘b. marsupialis experime&ally infected with strain G-49 developed stable, subpatent parasitaemia, with 80% positive haemocultures (HANSEN et al., 1991). Three batches of 3 recently weaned laboratory-bred I?

510

AX4

fienatu were infected subcutaneously, on the inner sur-

PAUIAPINHO

to T. cruzi II (formerly lineage 1) or T. cnczi I (formerly lineage 2), respectively.

face of the right thigh, with 200 cultured metacyclic trypomastigotes per gram of body weight of strains Y, C 13 and G-49. Similarly, 3 recently weaned laboratorybred D. marsupialis were infected with strain C- 13. The animals were examined as described above for natural infections. Infected faeces from 7 Rhodnius prokws collected in the study area (PINHO et al., 1998) were inoculated into Swiss mice. These isolates were given the prefixes BF or BPT. Positive haemocultures derived from the naturally infected D. ~a~u~i~~ (n = 10) and R j&rata (n = 10) were amplified in LIT medium and 1O5metacyclic forms from each isolate were inoculated intraperitoneally into 5 outbred Swiss mice weighing 18-20 g. The duration of prepatent and patent periods, peak parasitaemia and mortality (percentage and time of survival) were determined. The prepatent period was determined by daily fresh blood film examinations and parasitaemia was followed by parasite counts in a Neubauer chamber every 2 days.

Results Natural infection

As previously described in other areas (HANSEN et al., 1997), we found that 50% of D. marsupials were infected, with subpatent, long-lasting parasitaemias. Twenty-nine percent of the 98 haemocultures prepared during the follow-up period were positive. Natural T. cnrzd infection in l? fienata was also subpatent, stable and long-lasting: 48% of haemocultures examined during the follow-up period were positive. No clinical sign of disease was observed, and no parasite was seen in the scent glands, of either didelphid species. Experimental infections

The 9 experimentally infected l? frenata exhibited low parasitaemias with rare positive fresh blood films. No local lesion or clinical sign of disease was observed. Although experimental infection with strain Y resulted in only subpatent parasitaemia, 94% of the haemocultures were positive during the follow-up period. Experimental infections with strains G-49 and C-13 resulted in long-lasting patent parasitaemias and positive haemocultures (Table 1). The 3 infected D. rna~~p~a~ exhibited only subpatent parasitaemia (Table 1).

Multilocus enzyme electrophoresis

Mnltilocus enzyme electrophoresis (MIEE) was performed as described by CUPOLILLO et al: (1994) _Ten T. cnazi isolates from D. rnar~~a~~, 10 from i? frelsata and 7 from R. prolixus were studied. The following enzymes were studied, each in agarose gel: glucose phosphate dehydrogenase (G6PDH; EC 1.1.1.49), malate dehydrogenase (MDH; EC 1.1.37), isocitrate dehydrogenase (IDH; EC 1.1.1.42), 2 forms of malic enzyme (ME; EC 1.1.1.40), glucose phosphate isomerase (GPI; EC 5.3.1.9), phosphoglucomutase (PGM, EC 1.4.1.9), and 2 peptidases (PEP2; EC 3.4.11.1). The isoenzyme bands produced on the gels were numbered according to mobility and the enzyme profiles submitted to numerical analysis to group the samples into zymodemes (Table 2). Similarity was calculated by Jaccard’s coefficient and clustering was determined with the aid of UPGMA algorithms (NTSYS-pcTM package, version 1.70; Exeter Software, Setauket, NY, USA). The frequencies of alleles in each population were used to calculate Nei’s genetic distance, using the PHYLIPTM package (FELSENSTEIN, 1989). Strains Y (MHOM/BB/OOIY; T. crazi II) and F (T. cruzi I ) were used as references.

Biological characterization of T. cruzi isolates

Two groups could be distinguished by the mortality rate in Swiss mice: the first group, which consisted mainly of P. frenata isolates, was more virulent for Swiss mice than was the group in which isolates from D. marsupialis predominated. Experimental infection of mice by the former isolates resulted in significantly higher mortality, the prepatent period was shorter, and the parasitaemic peak occurred later. The other characters examined did not differ significantly (Table 2). Z~odemes

The 27 stocks examined could be grouped into 15 zvmodemes (Table 2 and Fig. 1). ME1 was the or& monomorphic locus. The most polymorphic locus was PGM (5 genotypes), followed by PEP2IB (4 genotypes). Heterozygotic patterns were observed for GPI, IDH, PGM, and PEP2. None of the stocks was identical with the T. cruzi reference strains F and Y. Some zymodemes differed from others in only 1 or 2 loci. Phenetic analysis of the enzyme data revealed 2 major groups of T. cntzi strains. The first was represented by 11 isolates, grouped into 8 zymodemes, together with the T. cruzi reference strain Y. Three subgroups could be observed in this group, which contained 8 Philander isolates (prefixed C) and 3 DideZphis isolates (650, 656 and 660). The second group was represented by 16

Molecular characterization

Deoxyribonucleic acid (DNA) was extracted by column chromatography using a l?harmaciaTM DNA extraction kit following the manufacturer’s instructions. A heterogeneous spot in the mini-exon gene non-transcribed spacer was amplified by the polymerase chain reaction as previously reported (FERN~ES et al., 1998). The amplified products were examined by agarose gel electrophoresis. Parasites presenting products corresponding to 300 or 350 bp were typed as belonging

Table 1. Course of infection of Trypanosoma cruzi in the opossuxns PhiZandtw @nata marsupialis after subcutaneous inoctdation with culture-derived metacyclic trypanosomes 13 and G-49

Peak parasitaemia Inoculuma

Mean pre-patent period (d)

P. fienata Y

c-13 G-49 D.c~~z4pidis

Mean duration of patency (d)

Mean no./mm3

-

Day (mean) -

s 23

2

20 400

-

-

-

i-:

*Groupsof 3 opossums.Eachopossumreceived200 trypanosomes per grambody weight. bNo. positive cultures/totaf

no. set up.

ETA,!..

and Z3ideZphis of strains Y, C-

Haemoculturesb

Duration of follow-up (months)

18/19 10130 4116

1% 12

4124

13

1: 1

1

:

; 2 2 z

: 1 I

; 2

: nd 1

: :

; 2

ME1 1 :

nd 1

1 2

ME2 1 :

:

z

G6PDH 2 ;

Enzymesb

1,2 2 1,2

z 3

1.2 1,2 1;2 ;

IDH 192 132 132 192 L2

133 l,i,3 1,2,4

3

i:

:

:

PGM 3 3 1 1,2,3 1,2,3

grouped in 15 zymodemes and 2 reference strains, determined

2

MDH :

GPI ;

Zymodemes” Zl c2, C13, c21 22 C22 23 c4 24 660,650 Z5 656

27 c40 26 C45 21 28 C43 Z9 C48, BF3, BF5, BPTl, BPT3, BPT4 : ZlO C60,661,672,655,666 Zll BF4 : 212 648 2 213 645 1 214 659 192 Z15 BFl 132 F” YC : “2 = zymodeme;the otherlettersand numbersreferto isolates. %eetext for full names;nd = not determined. ’ Referencestrain.

cruzi

Table 2. Isoenzyme profiles of 27 isolates of Trypanosoma electrophoresis of 9 enzymic loci

1,2,:,4,5 1,2,3,4,5 3 3

1,2, :, 4,5

1,2, z, 4,5 b&3,4,5

3

3 i:

PEP2IA 1,2,3,4,5 ;

by multilocus

::s

22

;

: 2 1

iJ

PEPZlB 2 0 0 1,3 1,3

enzyme

E

g s

4 5 8 ii @ i

ANA PAULA

PINHO

ETAL.

Jaccsrd’s coefficient 0.00 I

0.25

0.50

1.00 I

0.75

r

I-,“C;C”,

C21

-645 -659

Fig. 2. Phenetic tree showing the relationships of Typanosoma cr-uziisolates from the opossums Philanderfrenatu (strain numbers prefixed C) and Lk’delphis marsupialis and the reduviid Rhodnius prolixus (strain numbers prefixed BF and BM’) in a sylvatic environment in Brazil, constructed using Jaccard’s coefficient and UGPA algorithms. Y and F are reference strains of T. cruzi.

c45

C48

645

648

BF’TI

Fig. 1. Multilocus enzyme electrophoretic profiles for peptidase 2 (A, B, C,) and isocitrate dehydrogenase (D)

isolates, grouped into 7 zymodemes, with T. cruzi reference strain F. This second group was less heterogeneous and was not subdivided; it contained all the Rhodniw (prefixed B) isolates, most (7) of the Didelphis isolates, and only 2 isolates from Philander (prefixed C) . The Rhodnius isolates, except BFl and BF4, belonged to the same zymodeme (Z9), together with 1 Philander isolate, C48 (Fig. 2). Molecular

characterization

Two D. marsupialis had mixed infections and 4 l? frettutu were infected with T.cruzi II. All the other Didelphis (8), 6 Philander and all 7 Rhodnius were infected with T. cruzi I (Table 2). Discussion Although l? j?enata and D. marsupialis are considered important reservoirs of T. cruzi (see ALIXJQUERQUE et al., 1971; BARRETO et&., 1979; ZELEDON et al., 1984), they seem to handle infection by T. cruzi with different strategies. No extracellular multiplication cycle, such as has been described in the anal scent glands of D. marsupialis, was observed in P. jkmata. The percentage of positive haemocultures during follow-up was significantly higher than in D. marsupialis, and LEGEY et al. (1999) have shown that, despite high individual variation, l? frenata responded to T. cruzi infection with higher serological titres than did D. marsupialis. Moreover, l? fienatu appeared to be a more susceptible host for T. cruzi since it supported persistent subpatent parasitaemia with the Y strain, which is known to be controlled, and at times eliminated, by very young, still

marsupium-dependent, D. marsupialis (see DEANE et al., 1984; TANSEN et al., 1991, 1997). The course of ewerimen&i infection difi‘ers significantly in parasitaemiaievel (HANSEN et al., 1991) (Table 2) and recognition of T. cruzi antigens (LEGEY, 1997). The electrophoretic data presented in this paper are in accordance with previous studies on population genetics of wild stocks of T. cruzi, revealing a marked level of heterogeneity (MILES et al.; 1978; T~AYFCENC & AYALA, 1988: LEWIcKA et al.. 1995; DE LUCA D’ORO et al., 1993). The 27 stocks examined could be grouped into 15 zymodemes. The present work revealed heterogeneity in several biological properties among the sylvatic T. cruzi isolates. It was difficult, however, to establish a correlation between biological properties and isoenzyme profiles. Biological characterization divided the isolates into 2 groups, 1 mainly associated with l? jkenata and the other with D. marsupialis and R. prolixus. l? frenata was able to support infection with a broader range of T. cruzi subpopulations. Phenetic analysis of the MLEE data also revealed 2 groups, 1 represented mainly by P. jknatu isolates and also including the Y strain and the other by D. marsuPialis and R. prolixus isolates together with the F strain. Mini-exon gene typing identified 4 isolates from l? tknata as T. cruzi II and those from D. marsumalis mainly as T. cnczi I. However, 3 isolates from D. marsupialis (650, 656, 660) were placed by MLEE in the I? ji-enata group (Fig. 1). Two of those isolates (650, 656) were typed as T. cruzi II + T. cruzi I, probably due to mixed infection (FERNANDES et al., 1999). The nicture obtained from this limited set of stocks is in agreement with the statements that T. cruzi can be subdivided into 2 maior lineages. The genetic distance between the 2 phenetic clusters, cal&lated by the frequency of alleles for the reduced number of loci examined, was 0.52, corroborating the idea that these 2 lineages are separated by a vast evolutionary distance (TIBAYRENC, 1995; SOUTO et al., 1996). Previous studies have shown a preferential association of T. cruzi II with the domestic transmission cvcle and of T. cruzi I with the sylvatic one (SOUTO et-al., 1996;

TRYPANOSOMA

CRUZI IN MAR8UPlAL5

513

of Trypanosoma cruzi in outbred Swiss mice after intraperitoneal inoculation Table 3. Course of infection with culture-derived metacyclic trypansomes of 20 isolates from LWe1phi-s marsupialis and Philander frenata and characterization of the isolates as T. crud groups I or II Peak parasitaemia Isolate”

Mean pre-patient period (d)

D. marsupialis

645 648 650 656

iz

Mean duration of patency (d)

Mean no./mn?

Day (mean)

15 21 2 4

90 220 -

-

-

-

Mortality rate (%)

Day of death (mean)

T. cruzi group

-

I, II

00

--

:

:

: I, II

660 661

; 14

ii16

900 160

:i

666 659 672 655

B 10

E8

340 170 140 310

9 E 29

1:oo 0 0

1; -

: I I

C48 :t3

s 13

23 17

550 1000 200

32 30 14

100 40 60

34 E

III

c45 C60 c40 Zl

6 5 :

2

80 2:: 20

0 0 40 40

21 z!:

:

:t 1:

34 15 19 14

Cl3 c22

: 4

18 17

4:

14

100

14

:: 40

P. j?enata

: I :: II

*Groups of 5 mice. Each mouse received 1O5trypanosomes “Group I = lineage II, group II = lineage I.

et al., 1998, 1999; ZINGALES et al., 1998), as was suggested by MILES et al. (1978, 1980). This study revealed the presence of both groups of T. cruzi, I and II, in sympatric wild mammals and the preferential linkage of subpopulations to a particular mammal. Biological, biochemical and molecular characterization clustered the T. cruzi isolates from triatomine bugs and from D. marsupialis into 1 group, and those from l? fienata into another. These data support the existence of 2 distinct transmission cycles in the studied area, 1 involving predominantly D. marsupial& and the other mainly I? fienatu. Experimental observations (data not shown) ruled out a role for Rhudnius in this distinction, since both D&&his and Philander isolates infected, and developed in, this insect similarly. Although D. marsuPialis is considered a svlvatic sneI ties, it is-well known that it often frequents and even colonizes human dwellings and may therefore act as a link between the sylvatic and domestic transmission cycles. However, infection of this species with T. cruzi II subpopulations (which are strongly associated with human infection) is rare (FERNANDES et al., 1998; ZINGALES et al., 1998). As we know that D. ma&upial& easily controls infection with T. cruzi strain Y ( T. cruziI1) (DE&E et al., 1984) and was not a good h&t for a i? cruzi stock isolated from Philander in the study area, we may consider l? frenata to be a better reservoir than Didelphis. Being a more strictly sylvatic species, P. ji-enata is never found or trapped in human dwellings. These aspects could explain the absence of human infection from the studied area. Our observations on the complexity of the sylvatic cycle of T. cruzi point to the need to take ecological factors into account in the sudy of any host-parasite interaction and epidemiology. Knowledge of factors such as host population density, habits and dynamics, as well as the peculiarities of the interaction of the parasite with its reservoir host, is a basic requirement for comprehension of the epidemiology of parasitic diseases. FERNANDES

Acknowledgements We thank-Alcidineia Ivo and Marcos Antonio dos Santos Lima for technical assistance. This study was supported by CNPq/ FAPERJIPAPESIFIOCRUZ.

References Albuquerque, R. D. R. & Barreto, M. P. (197 1). Estudos sobre reservatorios e vectores silvestres do Trypanosoma cruzi. XLXIII: InfeccHo natural do marsupial Philander fienata quica (Temrnide, 1825) pelo Typanosoma cruzi. Rev&a Brasileira a%Biologia, 3, 371-376. Andrade, V., Brodskyn, C. & Andrade, S. G. (1983). Correlation between isoenzyrne patterns and biological behaviour of different strains of Tfypanosoma cruzi. Transactions of the Royal Society of Tropical Medicine and Hygiene, 17,796-799. Anonymous (1999). Recommendations from a satellite meeting. Memdria do Znstimw OswaMo Cmz, 94, supplement 1, 429-432. Barreto, M. P. & Ribeiro, R. D. (1979). Reservatorios silvestres do Typanosoma (Schizotrypanum) cruzi, Chagas 1909. Revista do Znstituto A&&o Lutz, 39,25-36. Cupolillo, E., Grimaldi, G. & Momem, H. (1994). A general classification of New World Leishmania using numerical zymotaxonomy. American Journal of Tropical Medicine and Hygiene, SO, 296-3 11. De Luca d’Oro, G. M., Gardenal, C. N., Perret, B., Crisi, J. V. & Montamat, E. E. (1993). Genetic structure of Ttypansoma cntzi populations from Argentina estimated from enzyme polymorphism. Parasitology, 107,479-487. Deane, M. P., Jansen, A. M. & Len& H. L. (1984). Vertebrate and invertebrate cycles in the same mammal host-the opossum Didelphis marsupialis. Membrias do Znstinrto Oswala’o cruz, 79,513-515. Dvorak, J. A., Hartman, D. L. & Miles, M. A. (1980). Trypanosoma cntzi: correlation of growth kinetics to zymodeme type in clones derived from various sources. Journal of Pmtozoology, 27,472-474. Felsenstein, J. (1989). PhylipPhylogeny Interference Package (version 3.2). Cladistics, 5, 164-166. Femandes, O., Souto, R. P., Castro, J. A., Pereira, J. B., Femandes, N. C., Jtmqueira, A. C. V., Nails, R. D., Barrett, T.V., Degrave, W., Zingales, B., Campbell, D. A. & Coura, J. R. (1998). Brazilian isolates of Twanosoma cruzi from

514

humans and triatomines classified into two lineages using mini-exon and ribosomal RNA sequences. American Journal of Tropical Medicine and Hygiene, S&807-8 11. Femandes, O., Mangia, R. H., Lisboa, C. V., Pinho, A. P., Morel, C. M., Zingales, B., Campbell, D. A. & Jansen, A. M. (1999). The complexity of the sylvatic cycle of Typanosoma cmzi in Rio de Janeiro State revealed by the non-transcribed spacer of the mini-exon gene. PurusiroZ@y, 118, 16 1- 166. Jansen, A. M. (1991). 0 gambol Didelphis marsupialis e o Trypanosoma cruzi. Estudos sobre uma harmoniosa interqcio parasita-hospedeiro. Thesis, Universidade Federale do Rio de Taneiro. Brazil. Jansen, A: M., Moriearty, P. L., Castro, B. G. & Deane, M. P. (1985). Typanosoma cruzi in the opossum Didelphis marsupiak an indirect fluorescent antibody test for the diagnosis and follow-up of natural and experimental infections. Transactions of the Royal Society of Tropical Medicine and Hyn’ene, 79, _I 444-477. - Jansen, A. M., Madeira, F., Carreira, J. C., Medina-Acosta, E., & Deane, M. P. (1997). Typanosoma cnrzi in the opossum Didelphis marsupialis: a study ofthe correlations and kinetics of the systemic and scent gland infections in naturally and experimentally infected animals. Experimental Parasitology, 86,37-44. Legey, A. I’., Pinho, A. I’., Xavier, S. C. C., Leon, L. & Jansen, A. M. (1999). Humoral immune resuonse kinetics in Philander bpossum and Didelphis marsupialis infected and immunized by Ttypanosoma cmzi employing an immunofluorescence antibody test. Memdrias do Institute Oswald0 Cruz. 94.37 l-376. Lewicki, I(‘.; Brenier-Campana, S. F., Bernabe, C., Dedet, J. P. & Tibayrenc, M. (1995). An isoenzyme survey of Tmanosoma cruzi genetic variability in sylvatic cycles from French Guyana. Experimental Parasitology, 81, 20-28. Miles, M. A., Tov& P. 1.. Oswald. S. C. & Godfrev. D. G. (1957). fie ide&fi&t~on by iioenzyme pattern;.of two distinct strain-groups of Typanosoma cruzi, circulating independently in a rural area of Brazil. Transactions ofthe Royal Society of Tropical Medicine and Hygiene, 71, 217-225. Miles, M. A., Souza, A., Pov6a, M., Shaw, J. J., Lainson, R. & Toy&, P. J. (1978). Isozymic heterogeneity of Trypanosoma cruzi in the first autochtonous patients with Chagas’ disease in Amazonian Brazil. Nature, 272,819-821. Miles, M. A., Lanham, S. M., de Souza, A. A. & Pov6a, M. (1980). Further enzvmic characters of Trvbanosoma cruzi and heir ivaluation for-strain identification:.Transactions of Ihe Royal Society of Tropical Medicine and Hygiene, 74, 22 l-237. Morel, C., Chiari, E., Camargo, E. I’., Mattei, D. M., Romanha, A. J. & Simpson, L. (1980). Strains and clones of Trypanosoma cruzi can be characterized by pattern of restriction endonuclease products of kinetoplast DNA minicircles. Proceedings of the Nation&Academy of Sciences of the USA, 77, 6810-6814. Pinho, A. I’., GonGalves, T. C. M., Mangia, R. H., Russell, N. S. N. & Jansen, A. M. (1998). Occurrence of Rhodniusprolixus (Stal, 1859), naturally infected by Typanosoma cruzi in the State of Rio de Janeiro, Brazil (Hemiptera, Reduviidae,

ANA PAULA PINHO ETAL. Triatominae). Membrias do Instituw Oswaldo Cruz, 93, 141-143. Silva, L. H. I’. & Nussenzweig, V. (1953). Sobre uma cepa de Typanosoma cruzi altamente virulenta para camundongo branco. Folha Clinica e Biologica, 20, 19 l-208. Souto, R. I’., Femandes, O., Macedo, A. M., Campbell, D. A. & Zingales, B. (1996). DNA markers define two major phylogenetic lineages of Typanosoma cruzi. Molecular and Biochemical Parasiwbgy, 83, 141- 152. Tibayrenc, M. (1995). An isoenzyme survey of Typanosoma cmzigenetic variability in sylvatic cycles from French Guiana. Experimental Parasitology, 81, 20-28. Tibayrenc, M. & Ayala, F. J. (1998). Isozyme variability in Typanosoma cmzi, the agent of Chagas’ disease: genetical, taxonomical and epidemiological significance. Evolution, 42, 377-397-_-.

Tib&renc, M., Cariou, M. L., Solignac, M., Dedet, J. P., Poch, 0. & Desieux, I’. (1985). New eletroohoretic evidence of genetic vahation and diploidy in Typankoma cruzi, the agent of Chagas’disease. Genetica, 67,223-230. Tibayrenc, M., Ward, I’., Moya, A. & Ayala, F. J. (1986). Natural populations of Typanosoma cruzi, the agent of Chagas’ disease, have a complex multiclonal sticture. Proceedinns of the National Academy of Sciences of the USA, 83,115-118. Tibayrenc, M., Kjellberg, F. & Ayala, F. J. (1990). A clonal theory of parasitic protozoa: the population structures of Emamoeba, Giardia, Leishmania, Naegleria, Plasmodium, Trichomonas, and Trypanosoma and their medical and taxonomical consequences. Proceedings of the National Academy of Sciences of the USA, 87,2414-2418. Tibayrenc, M., Neubauer, K., Bamabe, C., Guenini, E., Skarecky, D. & Ayala, F. J. (1991). Genetic characterization of six parasitic protozoa: parity between random-primer DNA typing and multilocus enzyme electrophoresis. Proceedings of the Nat&al Academy of Sciences of the USA, 90, 1335- 1339. Zeledbn, R., Solano, G., Senz, G. S. & Swanwelder, J. C. V. (1970). Wild reservoir of Typanosoma cruzi with special mention of the opossum Didelphis marsupialis and its role in the epidemiology of Chagas’ disease in an endemic area of Costa Rica. Journal of I’arasitology, 56,38-42. Zeledbn, R., Bolnaos, R. & Rojas, M. (1984). Scanning electronmicroscopy of the fmal phase of the life cycle of Typanosoma cruzi in its insect vector. Acta Tropica, 41, 34-43. Zingales, B., Souto, R. P., Mangia, R. H., Lisboa, C. V., Campbell, D. A., Coura, D. J., Jansen, A. M. & Femandes, 0. (1998). Molecular epidemiology of American trypanosomiasis in Brazil based on dimorphisms of rRNA and miniexon gene sequences. hernational Journalfor Parasitology, 28, 105-112.

Received SJuly 1999; revised 15 February publication 17 Februay 2000

2000; acceptedfor