EXPERIMENTAL PARASITOLOGY 75, 187-195 (1992)

Characterization of Chilean, Bolivian, and Argentinian Trypanosoma cruzi Populations by Restriction Endonuclease and lsoenzyme Analysis A. SOLARI, *J S. MuAoz,* J. VENEGAS,* A. WALLACE,? X. AGUILERA,~ W. APT,$ S. F. BREN&RE,§ AND M. TIBAYRENC§ *Departamento de Bioquimica, fDepartamento de Medicina Experimental, and fUnidad de Parasitologia, Departamento de Medicina Experimental, Fact&ad de Medicina, Universidad de Chile, Casilla 70086, Santiago 7, Chile; and lLaboratorie de Genetique des Parasites et des Vecteurs, Orstom, Montpellier, France SOLARI, A., Murjoz, S., VENEGAS, J., WALLACE, A., AGUILERA, X., APT. W., BRENI~RE, S. F., AND TIBAYRENC, M. 1992. Characterization of Chilean, Bolivian, and Argentinian Trypanosoma cruzi populations by restriction endonuclease and isoenzyme analysis. Experimentnl Parasitology 75, 187-195. Ninety-one Chilean, 15 Bolivian, and 9 Argentinian Trypanosoma cruzi stocks, isolated from various hosts and vectors, were characterized by schizodeme analysis with EcoRI and MspI endonucleases. The three major

similar pattern groups that emerged from this sample correlated with results of isoenzyme analysis. This result confirms previous work and supports the hypothesis of the clonal structure of natural populations of T. cruzi, fully defined at the level of isoenzyme analysis, quantitative kinetoplast DNA restriction fragment length polymorphism, and kinetoplast DNA hybridization analysis. In Chile, sylvatic and domestic cycles of T. cruzi transmission appear to be mainly independent: genetically different families of natural clones are specific to these cycles. Nevertheless, the possibility of overlap remains unclear. Results described here indicate that natural clones inhabiting Chilean regions appear genetically related to the natural clones identified in neighboring countries. In Chile the more frequently sampled parasite types are natural clone 39 and a genetically closely related clone NP13. In this work an evaluation of T. cruzi natural clone mixtures in T. cruzi stocks from Chile was performed for the first time by schizodeme analysis before and after serial transfer in mouse maintenance. The results indicate that six of nine stocks are composed of two or more natural clones. This observation raises the relevant question of whether specific T. cruzi natural clones generate different clinical features of Chagas’ disease. o 1992AcademicPRSS, IIIC. INDEX DESCRIPTORS: Trypanosoma cruzi, clonal structure; Schizodemes; Zymodemes; Chagas’ disease; Epidemiology; Parasite natural selection.

Trypanosoma cruzi is the etiologic agent of Chagas’ disease, which affects around 20 million people in Latin America. It is a pleomorphic entity in which different clinical manifestations can occur during the acute and chronic phases. Some evidence suggests that parasite factors can explain this variability (Brener 1973). The genetic variability of T. crud has been explored by isoenzyme studies as well as by kinetoplast DNA restriction fragment ’ To whom correspondence should be addressed.

length polymorphism (kDNA RFLP) analysis (Miles et al. 1984; Morel 1984). Population genetic studies (isoenzymatic genetic markers) on an extensive T. cruzi sample revealed a clonal structure of T. cruzi natural populations: the zymodemes described by 15 isozyme loci can be equated to natural clones or families of genetically closely related parasites (Tibayrenc et al. 1986; Tibayrenc et al. 1991). It is worth noting that the term “natural clone” is different from the classical term “clone” or “laboratory clone,” to indicate a parasitic line experimentally obtained from single cells. 187 0014-4894192$5.00 Copyright AU rights

0 1992 by Academic Press, Inc. of reproduction in any form reserved.

188

SOLAR1ETAL.

Because of the clonal structure and the absence of sexual recombination, an isoenzyme characterization restricted to a few enzymatic systems gives only a preliminary classification (Miles et al. 1984). kDNA RFLP analysis has demonstrated different schizodemes (a schizodeme consists of populations displaying similar or identical patterns). These schizodeme patterns appear to coincide with zymodeme characterizations (Carreiio et al. 1987; Morel 1984) and a statistically significant correlation has been found between a quantitative comparison of the different schizodemes and genetic distances evaluated by isoenzyme analysis (Tibayrenc and Ayala, 1987). This last result has been interpreted as a confirmation of the clonal structure of T. cruzi natural populations. To understand better the epidemiological and clinical implications of T. cruzi genetic variability, it is necessary to know the geographic distribution of the different natural clones in different transmission cycles. In this work, T. cruzi stocks were characterized by isoenzyme studies and kDNA RFLP analysis with two restriction endonucleases. Ninety-one stocks of T. cruzi from various areas of Chile where it is endemic and 24 stocks from specific areas of the neighboring countries of Bolivia and Argentina were studied. Also, the heterogeneous nature of Chilean T. cruzi stocks was evaluated by schizodeme analysis of these stocks before and after mouse infection.

tion of at least four zymodemes as previously described (Miles and Cibulskis 1986). A second analysis was performed on some strains with 11 enzyme systems corresponding to 12 loci as described earlier (Brenibre et aE. 1991)(see Table I). This method classifies the T. cruzi stocks in 43 different natural clones or genetically closely related parasites. Schizodeme analysis. Parasite kDNA was obtained according to conditions previously described (Goncalves et al. 1984). kDNA extracts were digested to completion with an excess of restriction endonuclease under the manufacturer’s buffer conditions (Boehringer and PL Biochemical laboratories). The digestion products were electrophoresed in a 4.5%10% polyacrylamide gel gradient and further stained with silver nitrate as described elsewhere (Goncalves et al. 1984). Quantification of schizodeme variability. To quantify the differences observed between two schizodeme patterns run side by side on the same gel, an index of banding homology was calculated using the formula 2N, x lOO/(N, + NJ, where N, is the number of total common bands and N, and NY are the total number of bands in stocks x and y, respectively (Tibayrenc and Ayala, 1987). This is a basic numerical taxonomy approach; it does not imply that all common bands have identical sequences but is only a statistical representation. Experimental infection of mice. Balbic, irradiated (450 rad), 20- to 22-g mice were first inoculated with 7’. cruzi metacyclic culture forms obtained by spontaneous differentiation or by culture in triatomine artificial urine (TAU 3AAG) medium as described previously (Contreras et al. 1985). Next, parasites were serially passaged by inoculation of blood trypomastigote forms collected at the peak of parasitemia for each stock (5 x lo4 to 8 x lo6 parasites/ml). At least 15 passages were made with a frequency of 7 to 35 days according to the parasitemia curve of each stock. These parasites were then isolated by hemocultures and amplified as described above for schizodeme analysis.

MATERIAL AND METHODS

The kDNA RFLP patterns obtained after digestion with both restriction endonucleases (MspI and EcoRI) permits a firm and clear classification of the studied stocks into three major groups. This grouping arises from a qualitative genotypic interpretation based on the similarities of the restriction patterns (Figs. 2, 3, and 4). Figure 1 shows the distribution of banding patterns from kDNA RFLP analysis of the three groups. The profiles of parasites belonging to group 1 and group 2 are quite

Parasites. Table I summarizes the geographic source, the host origin, and the isoenzyme characterization (see below) of the different T. cruzi stocks studied. Cultures were initiated with blood samples or triatomine intestinal contents; parasites were grown on a rabbit blood agar base and a liquid Diamond medium overlay with 5% fetal calf serum (Diamond 1986) and further cryopreserved in 7% DMSO. Zymodeme analysis. Two types of isoenzyme analyses were performed. The simplest was the electrophoretic analysis of glucose phosphate isomerase (GPI) and phosphoglucomutase (PGM), performed on all 115 stocks. This fust analysis allows the classifica-

RESULTS

CHARACTERIZATION

OF

189

Trypanosoma cruzi POPULATIONS

TABLE I Geographic and Host Origin and Genotype of Trypano~oma cruzi Stocks Studied Number of stocks 4 2 1 1 1 4 1 4 22

11

3 2 11 1 11 1 10 1 5 1 1 4 2 1 1 1 1 4 Total 115

Stock

ALD,DCC,EPP,FCV vGM,vMV4 vMV3 vTV

Locality

Country

Host

Genotype

I I I I II II III III

Chile Chile Chile Chile Chile Chile Chile Chile

Human

Human

Z, bol (GR clone NP 13;

RMS,AAM,PMC,JCh, CG,FG,CA,FS,JTG, MN,RPC,MGA,MGC,EMD, NC,SCC,ES,CP,PCC,MC, MXCH43,MXCH% MCV,CBB,NT,CEE MVB,MXCH46,MXCH53, MXCHBO,MXCHSS, MXCH89,MCH3 WT,LQ,LGN

Region IV

Chile

Human

NR clone 39) Z2 bol (MN,RMS clone 39)

Region IV

Chile

Human

Zz bra (CBB,MCH,, MXCH53 clone 33; CEE,MXCH88 clone 32

Region IV

Chile

Human

~1682.~1738 v~X,V~X,V~~X,V~~X,VS~,V~~, v98,v101,v102,v1672,v0V1, sp153 spl,spAII,sp3l,sp54,splO4, v1646,~1649,CHI22,WALLF, spCOMB1 ,spCOMBZ JBC-P HSG,CLA,RNCLA,RNGRO, SPA,CO,MF,JT,HFLA,JFV 862039 SO16,SO18,SO22, so34,so50

Region IV Region IV

Chile Chile

7. infestans 1. infestans

Z, (LQ clone NW; LGN clone NP4) Z, bra ’ Z, bol (v2X clone 39)

Region IV Region IV

Chile Chile

T. spinolai T. spinolai

Region IV Metropolitan region Yungas Potosi

Chile Chile

dog

Bolivia Bolivia

D. marsupialis T. infestans

TPKl so35 PP-B,AC-B,JR-B,GJ-B P263,JT-B 85847 862036 CSH-14 clone CA 69 3-2,49-5,10-A,3-1 1l-A,40-2,40-4

Yungas Potosi Unknown Unknown Alto Beni Yungas Cordoba Unknown Saha Saha

Bolivia Bolivia Bolivia Bolivia Bolivia Bolivia Argentina Argentina Argentina Argentina

T. infestans T. infestans

v195 vllS,vl20,vl21,SABChlO8d ~124 AP,GR,NR,JGG

Region Region Region Region Region Region Region Region

7. 7. T. T. T. T.

i&tans infestans infestans infestans infestans infestnns

Human

Human Human D. novemcinctus D. marsupialis

Z, bol Z, bol

Zl

Z, bra Z, bol Z,

Z,

Zz bra 2, (spAI clone NPS; sp31 clone NP6; sp104 clone 19) Z, Zz bol Z, (clone 6) Z, (SO16,SO22clone NP3; SOl8,SOSOclone 19; SO34clone 20) Zz bol (clone 39) Z, bol (clone 39) Z, (clone 20) Z2 bol (clone 39) Near clone 35 Clone 43

Human Human T. infestans T. infestans

New profile

Nore. T. cruzi used in this study are indicated together with the host or vector from which they were obtained and the region of their isolation. Zymodeme characterization was performed by analysis of two enzyme markers in all stocks; in parentheses is the classification obtained for some stocks with 1I enzyme systems.

different from those of group 3 when EcoRI is used, but groups 1 and 3 display different profiles with respect to group 2 when samples are digested with 44~~1. Group 1 (Figs. 2A and 2B) is composed of stocks from Bolivia and Chile. In Bolivia they are found both in the domestic (patients and Triatoma infestans vector) and sylvatic cycles (mammalian reservoirs). In

Chile they are found mainly in Triatoma spinolui, a vector implicated in the sylvatic transmission cycle of Geographic Region IV and also in T. in&tans from the highland Chilean areas (Region II). Only three stocks of this group belonging to zymodeme 1 were found in Chilean patients (Region IV). They were detected during a human survey and represent 3% of the total

190

SOLARI MSPI

ET AL.

gentinian stocks. Previous results of hybridization with kDNA probes of Argentinian stocks indicate the unique character of some of these isolates (Macina et al. 1987) for which we do not have isoenzyme typification results. However, schizodeme analyses with MspI (Fig. 4B) and Hinff (data not shown) suggest that they belong to this group rather than to others. I 3 2 3 Group 3 displayed a unique EcoRI reFIG 1. Diagram of banding distribution after kDNA striction pattern with few bands. This group RFLP analysis (EcoRI and MspI) among three T. cruzi is composed of Chilean stocks isolated stock groups. Molecular weight markers are indicated from humans as well as from T. infestans, in base pairs. both pertaining to the domestic transmission cycle. Exceptionally, one stock sample as previously communicated (Apt et (~~153) was isolated from the insect vector al. 1987). The isoenzyme study performed T. spinolui, which has a sylvatic behavior. The isoenzyme study of group 3 stocks with 12 loci showed some heterogeneity within group 1 (Table I; see also Breniere et demonstrates their relationship to natural al. 1989): the Chilean stocks appeared clones 32 and 33 (Veas et al. 1991). Howclosely related to natural clones 19 and 20, ever, MspI generates a large number of previously identified in a Bolivian sample bands among these parasite types and was able to discriminate between natural clones (Tibayrenc and Ayala, 1988). Moreover, quantitative analysis of kDNA RFLP 32 and 33 (Fig. 5B). Schizodeme analysis performed on the within this group confirms these results. Thus, Bolivian T. cruzi stocks analyzed different stocks after axenic culture showed with MspI or EcoRI have average homolo- reproducible patterns. To evaluate the hetgies of 44.6 and 34%, respectively. How- erogeneous character of the Chilean isoever, two stocks grouped as natural clone lates, we performed a schizodeme analysis 19 have 90% homology with both enzymes. on 9 stocks before and after mouse infecChilean natural clones NP5, NP6, and 19, tions. Figure 6 shows the MspI endonuon the other hand, have an average homol- clease profiles for 6 stocks before and after ogy of 52% with either MspI or EcoRI. The mouse maintenance. Two of three stocks other Chilean natural clones, NP4 and NP7, not shown in Fig. 6 (WT and spl) changed are less related to natural clone 19 and dis- their schizodeme pattern and one (RMS) kept its profile (not shown). In conclusion, play homology of only 34%. Group 2 found by schizodeme analysis is 6 of 9 (66%) Chilean T. cruzi stocks tested composed of T. cruzi ubiquitously found in were composed of heterogeneous populaChile, Argentina, and Bolivia (Figs. 3 and tions: schizodeme profiles before and after 4). The GPI isoenzyme marker for Bolivian mouse infection showed altered patterns. and Chilean stocks displays the character- In contrast, only 1 of 35 Chilean stocks had istic heterozygous pattern of natural clones altered zymodemes, as studied by two en39 and NP13 (Tibayrenc and Ayala, 1988). zymatic systems (GPI and PGM), when Moreover, the 67% homology with EcoRI maintained on axenic culture for long periand MspI of these parasite natural clones ods. (39 and NP13) reinforces their close relaDISCUSSION tionship. Figure 4A shows similar EcoRI To understand the clinical implications of restriction patterns among Chilean and ArECO RI

CHARACTERIZATION

clone

OF Trypanosoma

cruzi POPULATIONS

191

w m 0, m m 0 o 0 o --ssNNNN

FIG. 2. Schizodeme profiles of Chilean and Bolivian zymodeme 1 parasites. Digestion of kDNA samples was performed with EcoRI (A) or MspI (B). Bolivian samples (left side) and Chilean samples (right side). Polyacrylamide gradient gel electrophoresis and staining were performed as described under Material and Methods. Molecular weight markers are from (top to bottom) 1353, 1078, 872,603, 310, 271, and 234 bp.

the genetic variability of T. cruzi natural clones, it is necessary to analyze a large number of parasite samples from different geographic regions and from humans exhibiting different clinical symptoms. Moreover, the identification of the T. cruzi natural clones of domestic and sylvatic cycles is relevant to understanding the epidemiology of the disease and to clarifying the transmission cycles operating in nature. Very recently, the new term “clonet” was proposed to designate these clonal lineages, which have been characterized for several genetic markers (Tibayrenc et al. 1991). In this work, various genetic discrimination techniques were used to characterize T. cruzi stocks. As in previous studies, the first isoenzymatic study performed with only two enzyme systems allowed a preliminary classification which was improved by a multilocus isoenzyme study (12 loci). The

second method, which detects greater variability, was correlated with the schizodeme analysis as in a previous study (Tibayrenc and Ayala 1987). It is important to recall that schizodeme analysis takes advantage of the heterogeneous nature of the miniand maxicircles of the kDNA of each T. cruzi population. In this study, we have been able to cluster all stocks analyzed into three groups of similar RFLP patterns, which confirms the correlation previously mentioned. Within these three groups we obtained great schizodeme variability, particularly between stocks from different countries. The observation of very similar schizodeme patterns (90% homology) in Bolivian stocks SO18 and SO50, both belonging to isoenzymatically classified natural clone 19, favors the hypothesis of clonal propagation of T. cruzi rather than sexual reproduction. However, slight differences

192

SOLAR1 ET AL.

39

NPl3

FIG. 3. Schizodeme profiles of Chilean zymodeme 2 bol parasites. Digestion of kDNA samples was performed with EcoRI (A) and MspI (B). Conditions are as described in legend to Fig. 2.

in kDNA are frequently observed in stocks belonging to the same zymodeme defined by 12 loci as previously reported (Tibayrent and Ayala 1988). These differences can be explained by the presence of heterogeneous populations in the natural stocks which could probably have a mixture of different natural clones or by variability not detected with 12 isoenzyme loci. Important epidemiological data were elucidated by this study. T. cruzi natural clones found in the Chilean sylvatic transmission cycle (stocks isolated from T. spinolui) are radically different from those identified in the domestic cycle (T. infesruns and human subjects). The exception is found in the Chilean highland close to Bolivia (Region II) where these same stocks have been found in T. infestam. They have also been found in T. infestans and humans of Bolivia and Southern Peru as reported previously (Brenitre et al. 1989). The main stocks found in the Chilean domestic transmission cycle belong to two different

schizodeme groups. A more complete isoenzyme study of theses groups with 12 loci classified each into two types: zymodeme 2 bol (natural clones 39 and NP13) and zymodeme 2 bra (natural clones 32 and 33). Both zymodemes were found in 81 and 16%, respectively (Apt et al. 1987). The last parasite type exists in Brazil as zymodeme A and displays typical GPI electrophoretic mobility and kDNA RFLP with EcoRI as described (Morel 1984; Carneiro et al. 1990). Our schizodeme data suggest that some Argentinian stocks are also related to natural clone 39. However, a multilocus characterization is necessary to ascertain their taxonomic status. These results confirm previous hybridization studies that suggest the presence in Argentina of stocks related to natural clone 39 (4&t), although there are others clearly different (3-2) (Solar-i et al. 1991). Experimental studies of mouse infections showed the high virulence of stocks belonging to natural clones 19 and 20 (first schizo-

CHARACTERIZATION

OF Ttypanosoma

cruzi POPULATIONS

193

le FIG. 4. Schizodeme profiles of Argentinian and Chilean zymodeme 2 bol parasites. Digestions were

performed with EcoRI (A) on kDNA samples from Argentinian provinces of C6rdova and Salta (lanes 1 to 8). Chilean kDNA samples of this type are shown for comparison (lanes 9 to 14). Digestion with MspI (B) of Argentinian samples (left panel), Bolivian samples (center panel), and Chilean samples (right panel). Conditions are as described in legend to Fig. 2.

deme group), in contrast to stocks of natural clone 39 or closely related stocks that are notoriously less virulent (Sanchez et al. 1990). This last group of stocks comprises the main T. cruzi populations infecting human subjects in Chile, where the disease is more benign because infected people are mainly asymptomatic. Nevertheless, the different clinical manifestations among humans infected by natural clones 39, 32, and 33 in Chile are not sufficiently documented and a previous study did not reach a definite conclusion (Apt et al. 1987). The clinical consequences of T. cruzi genetic variability are difficult to evaluate in natural infections because of the frequent mixed infections in humans (two or more different zymodemes in one host) as is the case in Bolivia (Breniere et al. 1989). The present methods of T. cruzi characterization require the isolation of the stocks and their amplification by axenic culture. It

appears that axenic culture may produce a selection of populations. Mixed isoenzymatic patterns generally change during long-term maintenance in axenic culture, and later they remain unchanged. Few mixed infections have been identified in Chile by isoenzyme studies. These results report for the first time the high frequency of mixed infections in Chile. Inoculation of Chilean natural stocks in mice after long axenic culture allowed the detection of new profiles from stocks characterized before mouse inoculation. As previously described in Brazilian stocks (Morel 1984; Carneiro et al. 1990), the new ecologic medium (mammalian host) should act as a new environmental selection pressure and also allow the development of other natural clones that could have been maintained at very low frequencies in axenic culture. Present characterization of T. cruzi stocks probably involves important biases due to

SOLAR1 ETAL.

clone

32

33

32 32

FIG. 5. Schizodeme profdes of Chilean zymodeme 2 bra parasites. Digestion of kDNA samples was performed with EcoRI (A) and MspI (B). Conditions are as described in legend to Fig. 2.

culture and xenodiagnostic selection. Recently, we have generated kDNA-specific probes of natural clones identified by multilocus isoenzyme studies (12 loci), obtained by PCR amplification of the hypervariable regions of the minicircles of the kDNA (Veas et al. 1990). This technique should allow direct detection of these different natural clones in mammalian blood and triatomine bug feces without culture amplification. In conclusion, to understand better the epidemiologic features of Chagas’ disease in different areas, it is important to develop appropriate methods of genetic characterization of T. cruzi clones. Multilocus isoenzyme studies provide a good understanding of the genetic variability of the different natural stocks and identify the clonal units useful for following epidemiologic studies. Schizodemes can be developed for the detection of residual variability within these

FIG. 6. Schizodeme profiles of Chilean T. cruzi after long-term maintenance in mice. Digestions were performed with MspI as described in Fig. 2. From left to right, patterns of each isolate before and after maintenance in mice. Conditions of mouse inoculation and passages are described under Material and Methods.

units defined by isoenzymes which at the moment are not known to have clinical or biological properties. Due to the dynamics of the transmission as well as to the geographic distribution of these natural clones, techniques for their direct detection in biological samples (mammalian blood, triatomine bug feces) must be developed to avoid biases by culture and isolation methods. ACKNOWLEDGMENTS This research was supported by the UNDPWorld Bank/WHO Special Programme for Research and Training in Tropical Diseases to AS. and by FONDE(XT-Chile. We thank Dr. Carlos Frasch for providing the Argentinian kDNA samples. REFERENCES AFT., W., AGUILERA, X., ARIUBADA, A., GOMEZ, L., MILES, M., AND WIDMER, G. 1987. Epidemiology of

CHARACTERIZATION

OF

Ttypanosoma Cruzi POPULATIONS

Chagas’ disease in Northern Chile: Isozyme profdes in Trypanosoma cruzi from domestic and sylvatic transmission cycles and their association with cardiopathy. American Journal of Tropical Medicine and Hygiene 37, 302-307. BRENER, 2. 1973. Biology of Trypanosoma cruzi. Annual Review of Microbiology 27, 347-383. BRENIBRE, S. F., CARRASCO, R., REVOLLO, S., APARKIO, G., DESJEUX, P., AND TIBAYRENC, M. 1989.

Chagas’ disease in Bolivia: Clinical and epidemiological features and zymodeme variability in Trypanosoma cruzi strains isolated from patients. American Journal

of Tropical

Medicine

and Hygiene

41,

521-529. CARNEIRO, M., CHIARI, E., GON~ALVES, A. M., DA SILVA, A. A., PEREIRA, C. M., MOREL, C. M., AND ROMANHA, A. J. 1990. Changes in the isoenzyme and kinetoplast DNA patterns of Trypanosoma cruzi strains induced by maintenance in mice. Acta Tropica 47, 35-45. CARREP~O, H., ROJAS, C., AGUILERA, X., APT, W., MILES, M. A., AND SOLARI, A. 1987. Schizodeme analysis of Trypanosoma cruzi zymodemes from Chile. Experimental Parasitology 64, 252-260. CONTRERAS, V., SALLES, J. M., THOMAS, N., MOREL, C., AND GOLDENBERG, S. 1985. In vitro differentiation of Trypanosoma cruzi under chemically defined condition. Molecular and Biochemical Parasitology 16, 315-327. DIAMOND, L. S. 1968. Improved method for the monoaxenic cultivation of Entamoeba histolytica Schaudin (1903) and E. hisrolytica-like, amoebae with Trypanosomatids. Journal of Parasitology 54,

715-719. N. S., AND MOREL, C. M. 1984. Trypanosomatid characterization by schizodeme analysis. In “Genes and Antigens of Parasites: A Laboratory Manual” (C. Morel, Ed.), pp. 102-l 15. Fundacao Oswald0 Cruz, Rio de Janeiro. MACINA, R. A., ARAUZO, S., REYES, M. B., GONCALVES, A. M., NEHME,

SANCHEZ, D. O., BASOMBRIO, M. A., MONTAMAT, E. E., SOLARI, A., AND FRASCH, A. C. C. 1987. Trypanosoma cruzi isolates from Argentina and Chile grouped with the aid of DNA probes. Molecular and Biochemical Parasitology 25, 45-53. MILES, M. A., APT, W., WIDMER, G., POVOA, M. M., AND SCHOFIELD, C. J. 1984. Isoenzyme heterogeneity and numerical taxonomy of Trypanosoma cruzi stocks from Chile. Transactions of the Royal Society of Tropical Medicine and Hygiene 78, 526

535. MILES, M. A., AND CIBULSKIS, R. E. 1986. Zymodeme characterization of Trypanosoma cruzi. Parasitology Today 2, 9698. MOREL, C. M. 1984. Schizodeme characterization of

195

natural and artificial populations of Trypanosoma cruzi as a tool in the study of Chagas’ disease. In “New Approaches to the Identification of Parasites and Their Vectors” (B. Newton and F. Minchal, Eds.), pp. 253-275. Schwabe and Co, Basel. MOREL, C., CHIARI, E., PLESSMAN CAMARGO, E., MATTEI, D. D., ROMANHA, A. J., AND SIMPSON, L. 1980. Strains and clones of Trypanosoma cruzi can

be characterized by patterns of restriction endonuclease products of kinetoplast DNA minicircles. Proceedings of the National Academy of Sciences USA 77, 6810-6814. SANCHEZ, G., WALLACE, A., OLIVARES, M., DIAZ, N., AGUILERA, X., APT, W., AND SOLARI, A. 1990. Biological characterization of Trypanosoma cruzi zymodemes: In vitro differentiation of epimastigotes

and infectivity of culture metacyclic trypomastigotes to mice. Experimental Parasitology 71, 125133. SOLARI, A., VENEGAS, J., GONZALES, E., AND VASQUEZ, D. 1991. Detection and classification of Trypanosoma cruzi by DNA hybridization with non38, radioactive probes. Journal of Protozoology

559-565. TIBAYRENC, M., WARD, P., MOYA, A., AND AYALA, F. J. 1986. Natural populations of Trypanosoma cruzi, the agent of Chagas’ disease, have a complex multiclonal structure. Proceedings of the National Academy of Science USA 83, 115-l 19. TIBAYRENC, M., AND AYALA, F. J. 1987. Forte corre-

lation entre classification isoenzymatique et variabilite de I’ADN kinetoplastique chez Trypanosoma cruzi. Comptes Rendus des Seances de l’dcademie des Sciences 304, 89-92. TIBAYRENC, M., AND AYALA, F. 1988. Isozyme variability of Trypanosoma cruzi, the agent of Chagas’

disease: Genetical, taxonomical and epidemiological significance. Evolution 42(2), 277-292. TIBAYRENC, M., KJELLBERG, F., AND AYALA, F. 1991. The clonal theory of parasitic protozoa: A taxonomic proposal applicable to other clonal organisms. Bioscience 41, 767-774. VEAS, F., CUNY, G., BRENIERE, S. F., AND TIBAYRENC, M. 1990. Subspecitic kDNA probes for major clones of Trypanosoma cruzi. Acta Tropica 48, 7%

82. VEAS, F., BRENI~RE, S. F., CUNY, G., BRENGUES, C., SOLARI, A., AND TIBAYRENC, M. 1991. General pro-

cedure to construct highly specific kDNA probes for clones of Trypanosoma cruzi for sensitive detection by polymerase chain reaction. Cellular and Molecular Biology

37, 73-84.

Received 30 January 1992; accepted with revision 4 June 1992