- PY/z3T D* Mem Cruz, Rio de Janeiro, 90(4): jul./aug. 1995 instDa%{ Oswaldo Vol. 443-448, 443 ig,d - OZ+& Inter-relationof Sylvak and Domestic Transmission of Trypanosoma cruzi in Areas with and without Domestic Vectorial Transmission in ,

CF L Diotaiuti, AS Pereira*, Loiola*’,**** AJ Fernandes**,JC Schofield***, JPPujardin ,JCP Dias**, E Chiari XWA - fi@cL Departamento de Parasitologia *Departamento de Bioquímica, ICB, UFMG, Caixa Postal 2486, 31270-901 , MG, Brasil **Centro de Pesquisas Ren6 Rachou-FIOCRUZ, Caixa Postal 1743, 30190-002 Belo Horizonte, MG, Brasil ***London**** Schoof of Tropical Medicine and Hygiene, WC1 E7HT, London, UK ORSTOM, BP 5445, 34032 Montpellier, France

During the period 1980-1 986, we captured triatomine bugs and mammalian reservoir hosts fim sylvatic and domestic situations in different municipalities of the State of Minas Gerais. Trypanosoma cruzi was isolated from captured bugs, mammals and patients. AJer cultivation in LIT medium, the electrophoretic enzyme projìles were determined. We obtained a total of 32 parasite isolateshm ye- gions with active domestic transmission, and 24 isolates fimareas under control. For the first arem the results suggest introduction of T. cruzi fimsylvatic habitats, through incursion of infected opos- sums andor sylvatic T. sordida, which appears to have given rise to at least one acute human infection. Ofparticular interest is thejìnding of sylvatic opossums and a T. sordida nymph infected with ZB, that could indicate return of parasites pom chronic human infections to sylvatic transmission cycles. For the areas under control we also interpret the results as interaction between sylvatic and domestic cycles ZI of transmission, here through the invasion of houses by bugs carrying the zymodeme from the syl- vatic environment. The Multivariate Correspondence Analysis gives a spatial description between the different parasite isolates and confirms the existence of a bridge in the opposite direction in the region 22 with active vectorial transmission including the exporting of through the peridomestic environment into the sylvatic cycle. For the other areas this bridge corresponds especially to Panstrongylus megistus, importing Zl into the domestic environment.

Key words: Trypanosoma cruzi - isoenzyme - Chagas’ disease

Enzymatic variation in Trypanosoma cruzi gions of the country, where the 22 zymodeme is seems largely due to clonal expansion of the para- episodic in the State of Ceara (Alencar 1987) and site under different conditions of selection (Tiba- absent in Amazonia and Venezuela wies 1985). yrenc & Ayala 1988). However, the electrophoretic Mixture areas in the State of Bahia (Barrett et al. isoenzyme patterns can be grouped into major 1980) are probably explained by the close distri- zymodemes that are each associated with particu- bution of T. infestans with the strictly domestic lar epidemiological situations mies 1983). In Panstrongylus megistus. Brazil for example, zymodemes Z1 and 23 are ’ Studying chagasic patients from Bambuí, principally associated with sylvatic transmission Minas Gerais, Brazil, Romanha (1982) classified (21 mainly with opossums, and 23 mainly with the parasite isolates in four zymodemes, ZA, ZB, armadillos) (Miles 1985), while 22 has been ZC, and ZD. ZA is equivalent to 22 of Miles et mainly isolated from patients in the chronic phase al. (1977), while ZB and ZC are distinct (Bogliolo of Chagas’ disease, particularly in the Southern et al. 1986) and ZD represented a mi’uture of pa- and Central regions (Romanha 1982, Miles 1983, rasites with patterns ZA and ZC. Romanha’s study Luquetti et al. 1986), which correspond to the prin- showed a predominance of ZA (or 22) amongst cipal Triatoma infestans distribution area in the the isolates from humans. Different papers con- c~mtry.The association I: infestansLZ2 zymodeme firm the great homogeneity of the parasites in the becomes weaker in the North and Northern re- sylvatic environment, showing prevalently Z 1 isozyme profiles @Mes et al. 1981, Póvoa et al. Received 1-3 September 1994 1984, Fernandes et al. 1991, Steindel et al. 1993). Accepted 18 April>f995 The similarity of banding patterns between para- \ ORSTOM Fonds Documentaire EX Cote: (3 % 6% : A T: 444 Sylvatic and Domes& Transmission of cruzi L tDiotaiuti et al. sites from a fatal acute case from the north of the where the vectorial transmission of Chaga dis- State of Minas Gerais and two isolates from opos- ease is still occurring, and regions where this sums led Romanha (1982) to suggest that T cruzi transmission was interrupted. íì-om sylvatic transmission cycles had been recently MATERIALS AND METHODS introduced into the domestic situation. Use of During the period 1980-1986, we captured isozyme profiles as epidemiological markers is a helpful tool for examining the relationship be- triatomine bugs and mammalian reservoir hosts tween sylvatic and domestic transmission cycles from sylvatic and domestic situations in Werent of T. cruzi. The technique was developed after T municipalities of the State of Minas Gerais, East infestuns had become established as the principle region of Brazil. This period covered the start and domestic vector in Brazil (Schofield 1988), but conclusion of a major national control campaign before major control interventions against this vec- against í". infestuns, the principal domestic vector tor had been carried out over the greater part of of T. cruzi in the region pias 1987). the country (Dias 1987). a result, the initial T. cruzi was isolated from captured bugs by As inoculation of bugs faeces into mice. Xenodiagno- interpretations of association between 21 zymo- deme with the sylvatic cycle, and 22 zymdeme sis using laboratory-bred T infestuns nymphs was with human disease, although representing con- carried out on the captured mammals, and faeces siderable simplification, for the southern and cen- from these bugs were also inoculated into mice. tral Brazil probably does reflect a generalised dis- Using the same procedure, we also obtained para- site isolates from two chronic-phase patients from tinction between domestic transmission involving T. and sylvatic transmission involving a region where no Chagas' disease transmission infestuns had been reported for many years, and one other species of Triatominae. is from T. infestans acute-phase patient a region where vector thought to have invaded Brazil since the turn of from transmission is still active. this century, mainly by passive dispersal in asso- ciation with human migrations (Schofield 1988). Parasites from the inoculated mice were culti- vated in LIT medium (Camargo 1964), and their Its origins seem to have been in the Cochabamba electrophoretic enzyme profiles determined using region of Bolivia (Dujardin et al. 1987), and it the methods of Romanha (1982). Each enzyme test- may have been responsible for the spread of the 2 plate included control samples representing 1, 22 zymodeme 22 of T cruzi from that region, intro- and 23 et al. 1977), and control samples of ducing it into Brazil where the predominant mies ZA, ZB, ZC andZD (Romanha 1982). Six enzymes zymodeme had previously been 21. The associa- were developed: LAT, (EC 2.6.1.2), ASAT (EC tion between the zymodeme 22 and P. megistus in 2.6.1.1), GPI 5.3.1.9), PGM (EC 2.7.5.1), the State of Bahia reinforces the correlation of (EC (EC thrs G6PD (EC 1.1.1.49) andME 1.1.40). pattern with the domestic cycle of T. cruzi whete Data were considered by area, with or without this triatomine accomplishes the epidemiological actual transmission (Tables I, II), and were ana- behaviour ofí? infestuns in Recôncavo Bahiano ai. lyzed by a Multiple Correspondence Analysis (Barrett et 1980). In at least one case, intro- (Thioulouse 1990). The factor maps &om this duction of í? into northeastern Brazil was infestuns I: analysis are presented for the two first factor coor- associated with an outbreak of cmzi transmis- dinates (Figs 1, 2), corresponding to Tables I and sion involving zymodeme 22 (Barrett et al. 1980). II, respectively. In Brazil, there has been considerable success in interrupting domestic cycles of ?: cruzi transmis- RESULTS AND DISCUSSION sion by eliminating domestic species of the We obtained a total of 32 parasite isolates from triatominebug vectors (Dias 1987). However, there regions with actual active domestic transmission, is an ever present risk of reinvasion of treated and 24 isolates from areas under control. The ZB houses by sylvatic species of Triatominae, which zymodeme profiles all corresponded to and ZC may lead to important changes in the epidemiol- of Romanha (1982), or 21 and 22 of Miles et al. ogy of Chagas' disease in the treated regions. For (1977) (Tables I, 11). example, after elimination of I: infestuns as the In areas with actual active domestic transmis- major domestic vector in parts of the State of sion we found Z1,22 and ZB zymodemes of í? two ZB, Minas Gerais, other triatomine species - I: cruzi (Table I). Most í? infestans had 22 or sordida and I! megistus - have become more fre- which are considered characteristic of human in- quently reported from the mal houses, represent- fections in southern and central Brazil (Miles et ing reinvasion from sylvatic ecotopes (Diotaiuti al. 1977, Romanha 1982, Luquetti et al. 1986). et al. 1993). The aim of the present study is there- However, some of the I: infestuns, most í? sordida fore to compare the epidemiological patterns of í? and most opossums showed 21 zymodeme which cruzi transmission in regions of Minas Gerais is generally considered characteristic of sylvatic -e. -. "3 I>; - I - . ,,+-*I+ '-1 11 I ï-t ? RJ Mem hst Oswa\do Cruz, Rio delaneiro, Vol. 90(4),jul./aug. 1995 445

r Q TABLE I

Local of capture, origin and electrophoretic isoenzyme patterns of Trypanosonla cmzi isolates from human patients, triatomines and reservoirs in areas with active vectorial transmission of Chagas’ disease, in the State of Minas Gerais, Brazil - ~~~~ Isolate number Municipality Local of capture Origin Zymodeme

1 Itacarambi domestic T. infestans B ” 2 Itacarambi domestic T. infestans 21 3 Itacarambi domestic T. infestans 21 4 Itacarambi domestic T. infestans 21 5 Itacarambi - Humadacute z1 6 Itacarambi domestic T. infistans 21 7 Itacarambi peridomestic T. sordida 21 8 Manga domestic T. infistans B 9 Manga domestic T. infistans B 10 Manga domestic T. infestans B 11 Montalvânia peridomestic T. sordida B 12 Montalvânia domestic T. sordida z1 13 Mato Verde peridomestic T. sordida 21 14 Mato Verde domestic T. injèstans B 15 Mato Verde sylvatic T. sordida 21 16 Mato Verde sylvatic D. albiventris B+Z1 17 Mato Verde sylvatic D. albiventris B+Z1 18 Mato Verde sylvatic T. sordida B 19 Mato Verde sylvatic D. albiventris 21 20 Mato Verde sylvatic D. albiventris 21 21 Mato Verde domestic í7 infestans 22 22 Mato Verde domestic T. infestans B 23 Mato Verde sylvatic T. sordida 21 24 Mato Verde peridomestic D. albiventris 21 25 Mato Verde sylvatic D. albiventris 21 26 Mato Verde sylvatic D. albiventris z1 27 Mato Verde sylvatic D. albiventris z1 28 Mato Verde sylvatic D. albiventris 21 29 Mato Verde sylvatic D. albiventris 21 30 Mato Verde sylvatic D. albiventris z1 31 Mato Verde sylvatic D. albiventris 21 32 Mato Verde sylvatic D. albiventris 21

I: cmzi. This strongly suggests introduction of ir: ertheless, some autochthonous species of Triato- cmzi from sylvatic habitats, through incursion of minae are found in sylvatic ecotopes and, occa- infected opossums andor sylvatic I: sordida, and sionally, inside houses. In these regions (Table II) appears to have given rise to at least one acute (ie. we verified the presence of 21 in sylvatic Triato- recent) human infection (Table I). Of particular minae and opossums, and in most of the bugs (l? interest here, however, is the finding of sylvatic megistus) captured in and around houses. Two opossums and a I: sordida nymph infected with patients with chronic infections showed 22 isoen- ZB Zymodeme of I: cmzi. These were captured in zyme profiles, and four of the I? megistus showed a wood about 50m from the nearest house (where ZB or 22 profiles whic3 are more usually associ- at least one inhabitant was seropositive for I: cruzi) ated with chronic human disease. Again therefore, and could possibly indicate return of parasites from we interpret these results to indicate interaction chronic human infections to sylvatic transmission between sylvatic and domestic cycles of transmis- cycles. sion. Here, it would appear that invasion of houses The areas under control representI: regions by otherwise sylvatic bugs can not only introduce where vectorial transmission ,of cruzi was pre- Z1into the domestic environment, but these bugs viously high, but now do not record the presence can also reconstitute domestic transmission of 22. of triatomine colonizationwithin the houses. Nev- However, Tibayrenc et al. (1986, 1990) have pre- 9-

of 446 Sylvatic and Dom&tic Transmission I cruzi L Diotaiuti et al.

TABLE II

Local of capture, origin and electrophoretic is”yme pattems of Trypansoma cruzi isolates from human patients, triatomines and reservoirs in areas without active vectorial transmission of Chagas’ disease, in the state of Minas Gerais, Brazil

~~ -~ Isolate number Municipality Local of capture Origin Zymodeme

1 Caeté sylvatic D. albiventris 21 2 domestic I? megistus 22 3 Patos de Minas - Humádcronic 22 4 Patos de Minas domestic F! megistus z1 5 Patos de Minas domestic E! megistus z1 6 Patos de Minas peridomestic I? megistus 21 22 7 domestic E! megistus 8 Urgem da Lapa domestic T vitticeps C 9 - Humadcronic 22 10 domestic I? megistus z1 11 peridomestic I? megistus z1 12 Betim . peridomestic E! megistus z1 13 Ibirité domestic I? megistus B 14 domestic I? megistus B 15 Guaranésia domestic F! megistus B 16 Guaranésia peridomestic E! megistus z1 17 peridomestic I? megistus z1 peridomestic I? megistus 21 18 C. 19 Mato Dentro domestic I? megiktus 22 20 Itapagipe domestic I? megistus 22 21 Itapagipe sylvatic R. neglectus 21 22 Itapagipe sylvatic I? megistus 21 23 Itapagipe sylvatic D. albiventris 21 24 Itapagipe sylvatic D. albiventris 21

Viously noted the presence of mixed zymodemes the remaining groups which correspond to isolates in infections of individual humans and triatomines, from T sordida from sylvatic, peridomestic and while Finlay and Dvorak (1987) have modelled domestic habitats, together with the single the rapid overgrowth of fast-growing I: cruzi clones peridomestic isolate from Didelphis. The implica- during the passage of new isolates. We cannot tion of this correspondence is that T. sordida is therefore exclude the possibility of loss of some acting as an effective brigde between sylvatic and biological clones during our passage of isolates domestic habitats, importing parasites of through cultures and mice, such that 22 zymodeme zymodeme Z1 from the sylvatic cycle to the do- for example, may have been present among some mestic environment. However, the position of of the primary isolates subsequently identified as groups d and h could imply also a bridge in the 21 zymodeme. opposite direction, exporting 22 zymodeme of I: The Multivariate Correspondence Analysis cruzi through the peridomestic environment into helps to illustrate the apparent interactions between the sylvatic cycle. qhitic and domestic transmission in the two ar- For the under transmission control, the eas, areas by giving a spatial description of correspon- same analysis drew fewer groups, but again illus- dence between the different parasite isolates. For trated a correspondence between the parasite iso- the areas with actual active transmission, the analy- lates in accordance with their domestic, perido- sis show eight grouping of parasite isolates mestic Or sylvatic origins (Fig. 2). Groups a and b (Fig.1). Groups and b correspond to the do- ZB, a I: all correspond to domesticI: isolates showing Z1 mestic isolates from infestam and from a hu- and 22 zymodemes of cmzi, group c conesponds man acute case, while groups c and d correspond to peridomestic Z1 zymodeme isolates, and group to all the sylvatic isolates from Didelphis. These d corresponds to sylvatic 21 zymodeme isolates. groups are clearly distinguished í?om each.other Ag& the distribution of parasite isolates within y by the factor axis, and by the x factor axis from these groups implies a bridge, this time by P. Mem /n~toswa/do cruz, Rio debeiro, Vol. 90(4),jul./aug. 1995 447

1

26 !2 31 21 30 Y I ,\- I

v Y I I J -2’ d Fig. 1: factor map of the multiple correspondence analysis per- two Fig. 2: factor map of the multiple correspondence analysis per- formed on Table I. The first (out of 9) factor coordinates formed on Table II. The two first (out of 8) Eictor coordinates x y are the and axis, respectively, and represent 43% of the are the x and y axis, respectively, and represent 49% of the total total variability. Domestic, peridomestic and sylvatic transmis- variability. Domestic, peridomestic and sylvatic transmissions, sions are separated mainly by the first axis (r = O.SI), while as well as the daerent vectors, are separated mainly by the first the vectors and zymodemes were more correlated to the sec- 0.88, axis (r = 0.87 and 0.93, respectively), while the zymodemes ond axis (r = 0.97 and respectively). r is the correlation were more correlated to the second axis (r = 0.60). r is the cor- between the variable and the factor. relation between the variable and the factor. megistus, importing 21 zymodeme into the domes- Barrett TV, Hoff RH, Mott KE, Miles MA, Godfrey DG, tic environment. Teixeira R, Souza JAA, Sherlock I 1980. Epide- In a few areas of Brazil - especially in the mu- miological aspects of three TIypanosoma cruzi nicipality of Bambuí, Minas Gerais control of T zymodemes in Bahia State, Brazil. Trans R Soc Trop - bled Hyg 74 84-90. infestans been effected for many years. In much has Bogliolo AR, Chiari E, Silva-Pereira RO, Silva-Pereira of the country?however, control of this vector has A 1986. Comparative study of Typanosoma cruzi only been effective since 1983 (Dias 1987). It enzyme polymorphirms in South America. Braz J would appear that control of this domestic vector bled Bio1 Research 19: 673-683. can significantly reduce the transmission ofí? cruzi Camargo EP 1964. Growth and differenciation in 22 zymodeme. However, the reinvasion of houses TTpanosoma cruzi. I. Origin of metacyclic trypa- by other species of Triatominae from sylvatic en- nosomes in liquid media. Rev Inst bled Trop de Sü0 vironments can initiate domestic transmission of Paulo 6 : 93-100. 21 zymodeme and may reconstitute domestic Dias JCP 1987. Control of Chagas disease in Brazil. transmission of 22 zymodeme. Opossums, which Parasito1 Today 3: 336-341. are often infected with T. cruzì, appear to play an Diotaiuti L, Loiola CF, Falcão PL, Dias JCP 1993. The ecology of Triatoma sordida in natural environ- important role in introducing 21 pattern into do- two mestic and peridomestic environments.This fact ments in different regions of the State of Minas emphasises the need for continued vigilance of Gerais, Brazil. Rev Inst Med Trop São Paulo 35: controlled areas in order to avoid colonization of 237-246. Dujardin JP, Tibayrenc M, Venegas E, Maldonado L, houses by the bugs and opossums which are the Desjeux P, Ayala J 1987. Isozyme evidence of lack main components of the sylvatic (21 zymodeme) of speciation between wild and domestic Triatoma transmission cycle ofí? cruzì. infestans (Heteroptera: Reduviidae) in Bolívia. J REFERENCES Med Ent 24: 140. Alencar JE 1987. História natural da doença de Chagas Fernandes AJ, Chiari E, Rodrigues RR, Dias JCP, no Estado do Ceará. Imprensa Universitária da UFC, Romanha AT 1991. The importance of the opossum Fortaleza, 341 pp. (Didelphis albiventris) as a reservoir for Trypa- P I -* 448 of T: Sylvatic and Domestic Transmission cruzi L Diotaiuti et al. # c, N - nosoma cruzì in Bambuí, Minas Gerais State. Mem Romanha 1982. Hetemgen,eidade isoenzìmótica em Ins! Oswaldo Cruz 86: 81-85. Trypanosoma cui. PHD Thesis. Universidade Fe- Finlay RW, Dvorak JA 1987. Trypan.osoma cruzi: analy- deral de Minas Gerais, Belo Horizonte, Brazil, I10 sis ofthe population dynamics of heterogenous mix- PP. tures. JProtoz 34 : 409-415. Schofield CJ 1988. The bioystematics of Triatominae. LuquetA AO, Miles MA, Rassi A, Rezende JM,Souza In BìoJystematìcs of haematophagous Insects. p. AA, Póvoa MM, Rodrigues I 1986. Tvpanosoma 284-312 MW Service, Systematic Association Spe- cruzi: zymodemes associated with acute and chronic cial Vol 37. Chagas’ disease in Central Brazil. Trans R Soc Trop Steindel M, Dias Neto E, Menezes CLP, Romanha AJ, Med Hyg 80: 462470. Simpson AJG 1993. Random amplified polymorphic Miles MA 1983. The epidemiology of South American DNA analysis of Trypanosoma cruzì strains. Mol Trypanosomiasis - biochemical and immunological Bioch Paras 60 : 71-80; approaches and their relevance to control. Trans R Thiouslouse J 1990. MacMul and GraphMu: two ma- Soc Trop Med Hyg 7: 5-23. cintosh programs for the display and analysis of Miles MA 1985. Isoenzyme characterization. Rev Soc Mltiwariate data. Computers and Geoscience I& BrasìIMed Trop 18: 53-59. 1235-1240. Miles MA, Toye PJ, Oswald SC, Godfrey DG 1977. The Tibayrenc M, Ayala FJ 1988. Isoenzyme variability in identification by isoenzyme patterns of two distinct Trypanosoma cruzì the agent of Chagas disease: strain-groups of Trypanosoma cruzì circulating in- genetical, taxonomical and epidemiological signifi- dependently in a rural area of Brazil. Trans R Soc cance. Evolution 42 : 277-292. Trop Med Hyg 71:217-225. Tibayrenc M, Hoffmann A, Pach O, Echalar L, Le Pont Miles MA, Povoa MM, Souza AA, Lainson R, Shaw JJ, F, Lemesre L, Desjeux P, Ayala FJ 1986. Additional Ketteridge DS 1981. Chagas disease in the Amazon data on Trypanosoma cruzì isoenzyme strains en- Basis. II. The distribution of Trypanosoma cruzi countered in Bolivian domestic transmission cycles. zymodeme 1and 3 in Pará state, North Brazil. Trans Trans R Soc Trop Med Hyg 80: 442-447. R Soc Trop Med Hyg 75 667674. Tibayrenc M, Kjellberg F, Ayala FJ 1990. A clonal theory Póvoa MM, Souza AA, Naiff RD, Arias JR, Naiff MF, of parasitic proton: the population strutures of En- Biancardi CB, Miles MA 1984. Chagas disease in tamoeba, Giardia, Leishmania, Naegleria, Plasmo- the Amazon Basis. N.Host records of Trypanosoma dium, Trichonomas, and Trypanosoma and their cruzì zymodemes in the State of Amazonas and Ron- medical and taxonomical consequences. Evolution dônia, Brazil. Ann Trop Med Parasit 78: 479-487. 87 2414-2418.