ORIGINAL PAPER

V. M. Veloso Æ A. J. Romanha Æ M. Lana

S. M. F. Murta Æ C. M. Carneiro Æ C. F. AlvesE. C. Borges Æ W. L. Tafuri Æ G. L. L. Machado-Coelho

E. Chiari Æ M. T. Bahia

Influence of the long-term Trypanosoma cruzi infection in vertebratehost on the genetic and biological diversity of the parasite

Received: 11 November 2004 / Accepted: 23 March 2005 / Published online: 7 June 2005� Springer-Verlag 2005

Abstract The influence of the long-term Trypanosomacruzi infection in vertebrate host on the biological andgenetic properties of the parasite was evaluated. FourT. cruzi isolates obtained from different chronic chagasicdogs infected with Berenice-78 T. cruzi strain during 2and 7 years were comparatively analyzed. The long-termT. cruzi infection has led to alterations in parasitemia,virulence and pathogenicity of Be-78 strain for mice.These biological parameters varied from low to high inrealation to the parental strain. Randomly amplifiedpolymorphic DNA and isoenzyme profiles detected twodistinct genetic groups of parasites. The first groupincluded the parental strain and two T. cruzi isolates,and the second group the two other isolates. Interest-ingly, the isolates of the second group showed areversibility of the genetic profile to the parental strainafter 25 passages in mice. No correlation between the

genetic groups and biological properties of the isolateswas observed. Our findings confirmed the populationheterogeneity of the Be-78 strain, and showed how dif-ferently it responds to the long-term infection in thesame vertebrate hosts.

Introduction

The diverse clinical outcome of human Trypanosomacruzi infection has been attributed to the genetic heter-ogeneity of parasite populations and to the host’sgenetic background. Studies using cloned and unclonedT. cruzi stocks reinforce this parasite’s biological andgenetic heterogeneity. Also they led to the hypothesisthat, in nature, T. cruzi exhibits a complex and hetero-geneous population (Morel et al. 1986; Finley andDvorak 1987; Marques de Araujo and Chiari 1988;Lauria-Pires et al. 1996; Macedo and Pena 1998). Newconcepts have been introduced to elucidate the com-plexity of T. cruzi populations dynamics. Tibayrenc andAyala (1988) proposed that T. cruzi natural populationshave a complex multiclonal structure. Thus, the char-acteristics of one strain, maintained in laboratory after along period, would be the result of the interaction ofseveral clone behaviors (Lambrecht 1965). In this case,considering that amastigotes and epimastigotes derivedfrom clonal lines, they exhibit differences in the prere-plicative lag period, amastigote doubling time andduration of the complete intracellular cycle in vitro(Engel et al. 1985). Therefore, the experimental condi-tions used to maintain T. cruzi strains may result in aselective process. It may benefit the more adaptedclones, and their expression would be perpetuated inrelation to the others subpopulation present.

In this context, the long interaction of multiclonalT. cruzi populations and vertebrate host can determinechanges in interaction patterns among the clones,especially by the action of host defense mechanisms. Inaddition, experimental evidence has demonstrated that

V. M. Veloso Æ C. F. Alves Æ W. L. Tafuri Æ M. T. Bahia (&)Departamento de Ciencias Biologicas,Instituto de Ciencias Exatas e Biologicas,Universidade Federal de Ouro Preto,Minas Gerais, BrasilE-mail: mtbahia@nupeb.ufop.brFax: +55-21-3135591680

A. J. Romanha Æ S. M. F. Murta Æ E. C. BorgesCentro de Pesquisas Rene Rachou,Fundacao Oswaldo Cruz,Belo Horizonte, Brasil

M. Lana Æ C. M. CarneiroDepartamento de Analises,Clınicas Escola de Farmacia,Universidade Federal de Ouro Preto,Minas Gerais, Brasil

G. L. L. Machado-CoelhoDepartamento de Farmacia, Escola de Farmacia,Universidade Federal de Ouro Preto,Minas Gerais, Brasil

C. F. Alves Æ E. ChiariDepartamento de Parasitologia,Instituto de Ciencias Biologicas,Universidade Federal de Minas Gerais,Minas Gerais, Brasil

Parasitol Res (2005) 96: 382–389DOI 10.1007/s00436-005-1373-z

strains isolated from patients in the acute phase of Chagasdisease had a more complex genetic composition thanthose obtained during the chronic phase (Macedo andPena 1998).

Lana and Chiari (1986), using the murine model,demonstrated a lower virulence for the Be-78 strain,isolated from the patient Berenice, in relation to theBe-62 strain, isolated from the same patient 16 yearsearlier. In addition, Veloso et al. (2001) observedalterations in the benznidazole susceptibility amongT. cruzi populations isolated from dogs infected with theBe-78 strain during the chronic phase of the disease.These observed changes in biological behavior mayreflect changes in the population dynamics of theparental strain or the genetic plasticity observed inT. cruzi. To confirm this hypothesis, in the present work,the genetic and biological characteristics of four T. cruzipopulations isolated from dogs inoculated with the sameparental strain (Be-78) and further maintained in micewere analyzed after a long term infection.

Materials and methods

Trypanosoma cruzi stocks

Be-78 strain – T. cruzi II. This T. cruzi strain was iso-lated from patient Berenice in 1978 (Lana and Chiari1986), and maintained cryopreserved in liquid nitrogenand by successive blood passages in mice. Berenice isconsidered to be the first reported human case of Chagasdisease (Chagas 1909). Four T. cruzi isolates obtainedfrom different chronic chagasic outbred dogs infectedwith the parental strain for 2 (Be-78D) and 7 (Be-78 A, Band C) years were comparatively analyzed. Table 1shows details about the origin of T. cruzi isolates.Immediately after isolation from the dog, parasites wereinoculated in mice and maintained through successiveblood passages. After the 1st and 25th passages, micewere submitted to hemoculture for parasite isolation.T. cruzi isolates were grown in LIT medium (Camargo1964) at 28�C, washed three times in PBS pH 7.4, bycentrifugation for 10 min at 1,500 · g at 4�C and theparasite pellets stored at �70�C until use to prepareenzymatic extracts and DNA.

Mice infections and parasitemia curves

Groups of six outbred male Swiss albino mice (18–21 g)were inoculated intraperitoneally with 5·103 blood-

stream trypomastigotes from the Be-78 T. cruzi parentalstrain and its isolates. Parasitemia was followed dailyfrom the 4th up to the 40th day of infection thereafterevery other day until negativation. Fresh blood wascollected from the mouse’s tail and the number ofparasites counted as described by Brener (1962). Para-sitemia curves were plotted at the 1st and 25th successiveblood passages in mice. Mortality rate was expressed asa percentage of accumulated deaths for 100 days afterparasite inoculation.

Histopathology

Groups of six mice were infected with the Be-78parental T. cruzi strain and its isolates. Three mice weresacrificed during the acute (20 days of infection) andthree during the chronic (100 days of infection) phases.Tissue fragments from brain, heart, spleen, liver andskeletal muscle were fixed in 10% formalin, embeddedin paraffin, sectioned (5 m thick), stained with hema-toxylin-eosin (HE) and examined by light microscopy.Sections were evaluated considering the presence andintensity of the inflammatory process, according to thefollowing criteria: ± = rare, + = discreet, inflam-matory cells isolated or in small foci; ++ = moderate,multifocal inflammatory cell infiltrates (6–15) or diffuseinfiltrates in small areas; +++ = accentuated, mul-tifocal inflammatory cell infiltrates (15–20) or diffuseinfiltrates in most of the tissue section; ++++ =intense, multifocal inflammatory cell infiltrates (morethan 20) or diffuse infiltrates in most of the tissue sec-tions. Tissue parasitism was examined in all micro-scopic fields (630·) and characterized as: ±= rare, +=discreet, single parasite nests; ++ = moderate, 5–15nests; +++ = accentuated, 16–50 nests; ++++ =intense, more than 50 nests.

Isoenzyme patterns

Isoenzyme patterns were obtained as described by Car-neiro et al. (1990). Five polymorphic enzymes were ana-lyzed: alanine aminotransferase (ALAT) [E.C.2.6.1.2];aspartate aminotransferase (ASAT) [E.C.2.6.1.1]; glucosephosphate isomerase (GPI) [E.C.5.3.1.9]; phosphogluco-mutase (PGM) [E.C.2.7.5.1] and glucose-6-phosphatedehydrogenase (G6PD) [E.C.1.1.1.49]. Enzymatic ex-tracts from standard T. cruzi zymodemes Z1, Z2, ZBand ZC were used as reference (Miles et al. 1977; Roma-nha et al. 1979).

Table 1 Origin of the T. cruziBerenice isolates

aInfected triatomine feces andurine For more details seeMaterial and Methods

Dog Source of thetrypomastigote

Time of infection(Years)

Clinical formof the dog

T. cruzi isolate

A Metacyclica 7 Cardiac Be-78AB Metacyclica 7 Indeterminate Be-78BC Blood 7 Indeterminate Be-78CD Blood 2 Indeterminate Be-78D

383

DNA preparation and randomly amplified polymorphicDNA patterns (RAPD)

Parasite DNA extraction, amplification and electro-phoresis were performed as described previously (Ste-indel et al. 1993). Five random primers, 3302 (5¢CTGATGCTAC 3¢), 3303 (5¢ TCACGATGCA 3¢),3304 (5¢ GCACTGTCA 3¢), 3307 (5¢ AGTGCTACGC3¢) and gt11F (5¢ GGTGGCGACGACTCCTGGAGCCCG 3¢) were used. RAPD analysis was limited toPCR-amplified products between 0.3 Kb and 2.5 Kb. Aphenetic tree based on band sharing between all pos-sible pairs in an analysis group was constructed usingthe Dice similarity coefficient (Dice 1945). Data derivedfrom this coefficient were collated into a matrix of

similarity that was then used for unweighted pair-groupmethod analysis (UPGMA) (Sneath and Sokal 1962).

Statistics

Eight biological parameters were analyzed betweendifferent experimental groups: infectivity, mortality,pre-patent period, patent period, maximum of parasit-emia, day of maximum parasitemia, parasitism and tis-sue lesions. Infectivity and mortality, determined bymouse survival rates, were compared using the Chi-square test. Scheffe’s test was used to compare mean ofprepatent, patent period, maximum of parasitemia andday of maximum parasitemia. (Snedecor and Cochran

Table 2 Means and the standard deviations of biological parameters analyzed in mice inoculated with blood trypomastigotes in the 1stand 25th passages in mice of the Berenice-78 parental T. cruzi strain and four isolates inoculated in mice

Biological parameters Be-78 parental strain Be-78 isolates

A n=12 B n=12 C n=12 D n=12

Infectivity (%) 100a 100a 100a 100a 100a

Mortality (%)d 0a 8.3a 8.3a 16.7a 0a

Prepatent period (days) 6.0±0.7b 9.5±2.1a 9.5±0.7a 5.5±0.7b 11±1.4a

Patent period (days) 26±4.6a 33±8.6a 30±9.4a 29±6.5a 18±5.5a

Maximum of parasitemia (MP)e 11.5±2.5a (1st peak) 5.7±0.3b 4.7±0.8b 10.4±1.5a (1st peak) 1.6±0.6c

12.2±5.2a (2nd peak) NA NA 17.9±4.1a (2nd peak) NADay of MP 14.0±4.29a (1st peak) 22.5±2.5b 30.5±0.5b 15±2a (1st peak) 24.0±3b

24±0a (2nd peak) NA NA 22.5±2.5a (2nd peak) NAHistopathologyf

Acute phase parasitism +++ +++ +++ ++++ ++Inflammation +++ +++ +++ ++++ ++Chronic phase parasitism ± ± ± + NDInflammation ± ± ± + ND

Prepatent and patent period, maximum and day of parasitemia wereexpressed as mean ± standard deviation of the mean (x ± sdm)NA = not appeared, ND = not detecteda, b, c Different letters indicate significant differences and the samesimilar values. P<0.05

dDetermined 100 days after infection.eNumber of trypomastigotes · 104 / 0.1 ml of bloodf++++ intense, +++ accentuated, ++ moderate, + discreetand ± rare

Fig. 1 Representativeparasitemia curves of Be-78parental T. cruzi strain and itsfour isolates (Be78A, B, C andD) at the 25th passage in mice.For details see Material andmethods

384

1989). Parasitemia levels for the animals in differentgroups were compared using Kolmogorov and Smirnovtest (Conover 1980). Differences were considered sig-nificant at P<0.05.

Results

Biological characterization

Our results show that the Be-78 T. cruzi parental strainand its isolates presented significant differences in somebiological parameters. Figure 1 shows the representa-tive parasitemia curves generated in mice inoculatedwith the Be-78 parental strain and its four T. cruziisolates at the 25th passage in mice. Three parasitemiaprofiles were observed. The first one produced by theBe-78 parental strain and Be-78C that shows parasit-emia curves with two distinct peaks and higher para-sitemia levels in relation to the other isolates. Thesecond one produced by Be-78A and B isolates thatshows a slow increase in parasitemia, with irregular andalmost no peaks. The third one produced by theBe-78D isolate shows a rather flat curve. Comparativeanalysis between the areas under the curves and the

maximum parasitemia (Fig. 1, Table 2) showed thatmice infected with the Be-78C isolate and Be-78parental strain exhibited high levels; Be-78A andBe-78B intermediate and Be-78D low parasitemia levels(P<0.01). Considering the first parasitemia peak, theincrease of parasitemia was significantly most evident(P<0.01) between animals infected with Be-78 parentalstrain and Be-78C compared with those infected withBe-78A, Be-78B and Be-78D. Table 2 also shows thatthe parental strain and Be-78C presented similarprepatent period whereas animals infected with otherisolates exhibited a higher prepatent period (P<0.01).Concerning infectivity, mortality and patent period, nosignificant differences were observed among theparental strain and its isolates. Furthermore, no chan-ges in the biological behavioral pattern occurred during25 successive blood passages in mice carried out forapproximately 12 months. However, a discreet tendencyof parasitemia increase was observed for all isolates inmice inoculated with blood trypomastigotes during thesuccessive passages in mice (data not shown).

The histopathology, during the acute phase, showedmononuclear inflammatory infiltrates that were some-times associated with amastigote forms of T. cruzi. Bothinflammatory processes and parasites occur mainly inthe heart and skeletal muscle, being absent in brain, liverand spleen. A similar tissue tropism was observed forthe parental strain and its isolates. On the other hand,variation in the intensity of the inflammatory processand parasitism were observed in animals inoculated withdifferent isolates of T. cruzi. Infiltrate and parasitism

Fig. 2 Mononuclear inflammatory infiltrates in cardiac muscletissue during acute phase of mice infected with Be 78 T. cruziparental and its isolates: a) accentuated (+++)—Be 78 parentalstrain, b) accentuated (+++)—Be 78A, c) intense(++++)—Be 78C and d) moderate (++)—Be 78D. HE 630·

385

were more intense in animals infected with Be-78C iso-late, accentuated with the parental strain, Be-78A andBe-78B isolates, and moderated with Be-78D (Table 2,Fig. 2). These results can be related to the evaluationperiod. Mice infected with the parental strain showed amaximum of parasitemia on the 13th day of infectionwhereas the animals infected with the isolates reached amaximum around the 20th day when the histopatho-logical evaluation was performed.

During the chronic phase, rare or discreet inflam-matory process and parasitism were observed, beingmore evident among animals infected with Be-78C, fol-lowed by those inoculated with the parental strain andBe-78A and Be-78B isolates. In animals inoculated withBe-78D, inflammatory process and parasitism were notdetected (Table 2). Discreet foci of fibrosis wereobserved only in heart muscle of animals infected withthe parental strain and Be-78C isolate (data not shown).

Molecular characterization

The genetic variability of the Be-78 T. cruzi strainobtained from dogs and its four isolates after one(Be-78A1, B1, C1and D1) and 25 passages in mice(Be-78A25, B25, C25and D25) was evaluated by isoenzymeand RAPD. Figure 3 shows the results of one repre-sentative enzyme (GPI). It shows the presence of twoparasite subpopulations (zymodemes Z2 and ZC),represented by two bands. After the 1st passage in miceBe-78A1 and Be-78C1 isolates and the Be-78 parental

strain presented the same profile different from that ofBe-78B1 and Be-78D1. After 25 passages in mice, allisolates, except Be-78A25, presented the same profile asBe-78 parental strain. The results observed by GPI wereconfirmed by the other four polymorphic enzymes(Table 3). In addition to Z2 and ZC zymodemesobserved in GPI and ASAT, ZB zymodeme was alsoobserved in PGM, G6PD and ALAT. The presence andproportion of each subpopulation varied according tothe isolate and the passages in mice. Clearly observedwas the exchange of populations and their tendency inchanging towards the profile of the parental strain. Theisoenzyme profile changes were gradual and reversible,clearly showing the appearance and disappearance ofsubpopulations. In several occasions the simultaneouspresence of two zymodemes could be observed. Z2, ZBand ZC zymodemes were identified at different propor-tions in most of the isolates, suggesting the presence ofat least three T. cruzi subpopulations in Be-78 strain.

Complex RAPD profiles were obtained with fivedifferent primers. Figure 4 shows RAPD profiles of fourT. cruzi isolates and Be-78 parental strain, amplifiedwith primer 3303. Eighty-two bands obtained with thefive primers were used for band sharing analysis. Thebands chosen were based on their resolution, repro-ducibility in separate amplification reactions and inten-sity. A matrix of Dice similarity coefficients, based on apairwise band sharing was calculated for the samplesallowing the construction of a UPGMA dendrogram.The dendrogram shows two distinct groups (Fig. 5).Group I includes the Be-78 parental strain and Be-78A1and25, Be-78 C1and25, Be-78 B25 and Be-78 D25 iso-lates. Group II comprises Be-78 B1 and Be-78 D1 iso-lates. Interestingly, the 25 passages in mice, reverted theisoenzyme and RAPD profiles of Be-78B1 and Be-78D1

isolates to that of the parental strain. A correlationbetween RAPD and isoenzyme analysis was observed.

Discussion

The existence of biological and genetic variability inT. cruzi populations has been clearly demonstrated(Marques de Araujo and Chiari 1988; Murta andRomanha 1999; Toledo et al. 2002). Several studies have

Fig. 3 Zymogram of the enzyme glucose phosphate isomerase(GPI) of the Be 78 T. cruzi parental strain and its four isolates afterone (A1, B1, C1 and D1) and 25 (A25, B25, C25 and D25) passages inmice. Z1, ZB and ZC are reference zymodemes

Table 3 Zymodemes of the Be-78 parental T. cruzi strain and the four isolates after one (A1, B1, C1 and D1) and 25 (A25, B25, C25 and D25)passages in mice

Enzyme Be-78 parental strain Be-78 isolates

A1 A25 B1 B25 C1 C25 D1 D25

GPI ZC z2a ZC z2 ZC Z2 zc Z2 ZC ZC ZC Z2 zc ZC

PGM ZB ZB ZB Z2 zB ZB ZB ZB Z2 zB ZBG6PD ZB ZB Z2 zB Z2 zB ZB ZB ZB Z2 zB ZBALAT ZB ZB Z2 zB Z2 ZB ZB ZB Z2 ZBASAT ZC ZC ZC Z2 zC ZC z2 ZC ZC Z2 zC ZC

a Mean that parasites from two zymodemes were present. In that case, ZC is present in a higher proportion than Z2

386

attempted to establish a relationship between T. cruzistrain and the clinical manifestations of Chagas disease(Oliveira et al. 1997; Espinoza et al. 1998). Despite thesestudies, no clear correlation between a parasite strain,defined by molecular markers, and a particular featureof the disease, has been demonstrated (Carneiro et al.1991; Lauria-Pires and Teixeira 1996). On the otherhand, studies using cloned T. cruzi populations havereported a correlation between genetic characteristics ofthe parasites and biological behaviors (Revollo et al.1998; Lana et al. 1998; Toledo et al. 2002). Fernandeset al. (1998) demonstrated that areas with high mor-bidity present the circulation of T. cruzi II while T. cruzi

I is present in areas where Chagas disease is infrequent,and morbidity, evaluated by the level of abnormalelectrocardiograms, is low.

In the present study, the long-term T. cruzi infec-tions led to alterations in the biological and geneticpattern of the Be-78 strain. Our results demonstratedsignificant differences between the Be-78 parental strainand its isolates in relation to parasitemia, virulence andpathogenicity for mice. Differences were constant inmost of the parameters examined: (a) Be-78C showedthe highest virulence, similar to the parental strain; (b)Be-78A and Be-78B, a moderate virulence; (c) Be-78D,a low virulence. In addition, mice infected withBe-78B, Be-78C and Be-78D isolates displayed differ-ent spectra of susceptibility to benznidazole, with curerates of 100%, 50% and 70%, respectively (Velosoet al. 2001). Indeed, such heterogeneity either onT. cruzi Berenice strain (Lana and Chiari 1986) orother strains (Lauria-Pires et al. 1996; Marques deAraujo and Chiari 1988, Marques de Brito et al. 2003)have already been described.

It is reasonable to speculate that parasite subpopu-lations with different genotypes are present in lowpercentages in the Be-78 T. cruzi strain, and that sub-population could be selected after a long-term vertebratehost T. cruzi interaction, indicating the polyclonality ofthe parental strain. The polyclonality is clearly demon-strated by the simultaneous presence of two zymodemesin the same isolate. This result shows that the transitionfrom one subpopulation to the other is gradual and that,eventually one subpopulation becomes dominant inrelation to the other. Isoenzyme and RAPD analysesshowed a reversibility of the genetic profile of twoT. cruzi isolates to the parental strain after 25 passagesin mice. Alves et al. (1993) have also observed reversible

Fig. 4 RAPD profiles of the Be 78 T.cruzi parental strain and itsfour isolates after one (A1, B1, C1 and D1) and 25 (A25, B25, C25 andD25) passages in mice.The primer used was 3303. Gel is at 6%acrylamide and silver stained. The numbers on the left are the sizemarkers in base pairs (bp)

Fig. 5 UPGMA dendrogram ofBe 78 T.cruzi parental strainand its four isolates after one(A1, B1, C1 and D1) and 25(A25, B25, C25 and D25) passagesin mice. The numbers on thehorizontal scale were derivedfrom the Dice similaritycoefficient

387

changes in the isoenzyme profile of T. cruzi Y strainafter maintenance in medium of different composition orafter passage in newborn mice.

The biological diversity observed among the isolatescould be attributed to the genetic potential of each ver-tebrate-host, in this case dogs and mice, in selectingparasite clones or subpopulations that are more adaptedto the new environment. This phenomenon seems to behost-dependent, because although each dog had beeninoculated with the same T. cruzi strain, they displayeddifferent clinical forms of the disease. As example,Be-78A isolate obtained from a dog with chronicchagasic cardiopathy was significantly less virulent formice than Be-78C, obtained from a dog with the inde-terminate form of the disease. These results stronglysuggest the influence of host factors in determining thedisease outcome. Similar results were obtained byLauria-Pires et al. (1996, 1997) with parasites obtainedfrom chagasic patients. They demonstrated a great bio-logical and genetic diversity among clones isolated fromchagasic patients with cardiopathy and megaesopha-gous. The authors admit that each chagasic patient wasinfected with several T. cruzi subpopulations and thatclonal lineages derived from each stock can lead to bio-logical characteristics different from the original strain.

Correlation between biological and phylogeneticdivergence of T. cruzi using in vitro and in vivoparameters have been reported (Revollo et al. 1998;Lana et al. 1998; Toledo et al. 2002). Their findingsindicate that T. cruzi populations, circulating in naturecan be composed of clones with different biologicalbehaviors, sharing different virulence factors or strivingfor selective advantages such as distinct tropisms andpathogenicity. Consequently, genetic characteristics ofsuch clones are relevant to explain the pathogenesis ofChagas disease. In addition, when Andrade et al. (1999),inoculated mice simultaneously with two differentT. cruzi populations they observed a clear differentialtissue distribution for the two populations. This suggestsa significant influence of genetic polymorphism ofinfecting T. cruzi populations in the pathogenesis ofchronic Chagas disease. Gomes et al. (1998) argued thatthe human host might select specific varieties of T. cruziclones from presumably mixed parasite populations.Our data agree with this hypothesis, demonstrating thatT. cruzi populations isolated from different dogschronically infected with the same T. cruzi strain andfurther maintained in mice, showed distinct biologicaland genetic characteristics.

Interestingly, in spite of the differences in virulence andpathogenicity observed among the Be-78 parental T.cruzistrain and its isolates, a major genetic profile was detectedafter 25 passages in mice. Different selective pressurespossibly induce parasite subpopulations change duringthe chronic phase of the infection in dogs and/or duringthe acute infection in mice. The genetic diversity observedin these T. cruzi populations is possibly related to theemergence of minor parasite subpopulations, as aconsequence of selection by the immune system of the

vertebrate host. In our study, however, no correlationbetween the genetic groups detected by RAPD andisoenzymes and biological properties of the T. cruzi iso-lates was observed. The biological and genetic variabili-ties observed in T. cruzi populations isolated fromdifferent dogs infected with the T. cruzi Be-78 parentalstrain are probably a host-dependent phenomenon. Themechanism that generates parasite biological and geneticvariabilities during a long-term infection in vertebrate-host requires further investigation.

Acknowledgements This work was supported by grants fromFAPEMIG (Fundacao de Amparo a Pesquisa do Estado de MinasGerais), CNPq (Conselho Nacional de Desenvolvimento Cientıficoe Tecnologico), UFOP (Universidade Federal de Ouro Preto) andFIOCRUZ (Fundacao Oswaldo Cruz) Brasil.

References

Alves AMB, Tanuri A, De Almeida DF, Von Kruger W (1993)Reversible changes in the isoenzyme electrophoretic mobilitypattern and infectivity in clones of Trypanosoma cruzi. ExpParasitol 77:246–253

Andrade LO, Machado CR, Chiari E, Pena SD, Macedo AM(1999) Differential tissue distribution of diverse clones of Try-panosoma cruzi in infected mice. Mol Biochem Parasitol100:163–172

Brener Z (1962) Therapeutic activity and criterion of cure on miceexperimentally infected with Trypanosoma cruzi. Rev Inst MedTrop Sao Paulo 4:389–396

Camargo EP (1964) Growth and differentiation in Trypanosomacruzi. I—Origin of metacyclic trypanosomes in liquid media.Rev Inst Med Trop Sao Paulo 6:93–100

Carneiro M, Chiari E, Goncalves AM, Pereira AAS, Morel CM,Romanha AJ (1990) Changes in the isoenzyme and kinetoplastDNA patterns of Trypanosoma cruzi strains induced by main-tenance in mice. Acta Trop 47:35–45

Carneiro M, Romanha AJ, Chiari E (1991) Biological character-ization of Trypanosoma cruzi strains from different zymodemesand schizodemes. Mem Inst Oswaldo Cruz 86:387–393

Chagas C (1909) Nova tripanozomiase humana. Estudos sobre amorfolojia e o ciclo evolutivo do Schizotrypanum cruzi n gen, nsp, ajente etiolojico de nova entidade morbida do homem. MemInst Oswaldo Cruz 1:159–218

Conover WJ (1980) Practical non parametric statistics, 2nd edn.Wiley, New York

Dice LR (1945) Measures of the amount of ecological associationbetween species. Ecology 26:297–302

Engel JC, Doyle PS, Dvorak JA (1985) Trypanosoma cruzi: bio-logical characterization of clones derived from chronic chagasicpatients. II. Quantitative analysis of the intracellular cycle.J Protozool 32:80–83

Espinoza B, Vera-Cruz JM, Gonzales H, Ortega E, Hernandez R(1998)Genotype and virulence correlationwithinMexican stocksofTrypanosoma cruzi isolated from patients. Acta Trop 70:63–72

Fernandes O, Souto RP, Castro JA, Pereira JB, Fernandes NC,Junqueira AC, Naiff RD, Barrett TV, Degrave W, Zingales B,Campbell DA, Coura JR (1998) Brazilian isolates of Trypano-soma cruzi from humans and triatomines classified into twolineages using mini-exon and ribosomal RNA sequences. Am JTrop Med Hyg 58:807–811

Finley RW, Dvorak JA (1987) Trypanosoma cruzi: analysis of thepopulation dynamics of heterogeneous mixtures. J Protozool34:409–415

Gomes ML, Macedo AM, Pena SD, Chiari E (1998) Geneticrelationships between Trypanosoma cruzi strains isolated fromchronic chagasic patients in souther Brazil as revealed byRAPD and SSP-PCR analysis. Acta Trop 69:99–109

388

Lambrecht FL (1965) Biological variations in trypanosomes andtheir relation to the epidemiology of Chagas’ disease. Rev InstMed Trop Sao Paulo 7:346–352

Lana M, Chiari CA (1986) Caracterizacao biologica comparativadas cepas Berenice e Berenice-78 de Trypanosoma cruzi isoladasda mesma paciente em diferentes perıodos. Mem Inst OswaldoCruz 81:247–253

Lana M, Silveira Pinto A, Barnabe C, Quesney V, Noel S,Tibayrenc M, (1998) Trypanosoma cruzi: compared vectorialtransmissibility of three major clonal genotypes by Triatomainfestans. Exp Parasitol 90:20–25

Lauria-Pires L, Teixeira ARL (1996) Virulence and pathogenicityassociated with diversity of Trypanosoma cruzi stocks andclones derived from Chagas disease patients. Am J Trop MedHyg 55:304–310

Lauria-Pires L, Bogliolo AR, Teixeira ARL (1996) Diversity ofTrypanosoma cruzi stocks and clones derived from Chagasdisease patients II—Isozyme and RFLP characterizations. ExpParasitol 82:182–190

Lauria-Pires L, Santana JM, Tavares FS, Teixeira ARL (1997)Diversity of Trypanosoma cruzi stocks and clones derived fromChagas disease patients: I—Behavioral characterization invitro. Rev Soc Bras Med Trop 30:187–192

Macedo AM, Pena SDJ (1998) Genetic variability of Trypanosomacruzi: implications for the pathogenesis of Chagas disease.Parasitol Today 14:119–123

Marques de Araujo S, Chiari E (1988) Caracterizacao biologica declones das cepas Y, CL e MR de Trypanosoma cruzi em camun-dongos C3H isogenicos. Mem Inst Oswaldo Cruz 83:175–181

Marques de Brito CM, Pires MQ, Pacheco RS (2003) Chagasdisease and HIV co-infection: genetic analyses of two Trypan-osoma cruzi strains under experimental immunosuppression.Kinetoplastid Biol Dis 2:7–23

Miles MA, Toye PJ, Oswald SC, Godfrey DG (1977) The identi-fication by isoenzyme patterns of two distinct strain-groups ofTrypanosoma cruzi, circulating independently in a rural area ofBrasil. Trans R Soc Trop Med Hyg 71:217–225

Morel C, Deane MP, Goncalves AM (1986) The complexity ofTrypanosoma cruzi populations reveled by schizodeme analysis.Parasitol Today 2:97–101

Murta SM, Romanha AJ (1999) Characterization of Trypanosomacruzi. Mem Inst Oswaldo Cruz 94:177–180

Oliveira EC, Stefani MM, Campos DE, Andrade AL, Stefani MM,Campos DE, Andrade AL, Silva AS, Rassi A, Luquetti AO(1997) Trypanosoma cruzi stocks isolated from acute Chagasdisease patients lead to lethal murine infection. Trans R SocTrop Med Hyg 91:25–27

Revollo S, Oury B, Laurent JP, Barnabe C, Quesney V, Carriere V,Noel S, Tybayrenc M (1998) Trypanosoma cruzi: impact ofclonal evolution of the parasite on its biological and medicalproperties. Exp Parasitol 89:30–39

Romanha AJ, Pereira AAS, Chiari E, Kilgour V (1979) Isoenzymepatterns of cultured Trypanosoma cruzi: changes after pro-longed subculture. Comp Biochem Physiol 62:139–142

Sneath PHA, Sokal RR (1962) Numerical taxonomy. Nature193:853–860

Snedecor GW, Cochran WG (1989) Statistical methods, 8th edn.Iowa State University Press, Ames

Steindel M, Neto ED, Menezes CLP, Romanha AJ, Simpson AJG(1993) Random amplified polymorphic DNA analysis of Try-panosoma cruzi strains. Mol Biochem Parasitol 60:71–80

Tibayrenc M, Ayala FJ (1988) Isoenzyme variability in Trypan-osoma cruzi, the agent of Chagas disease: genetical, taxo-nomical and epidemiological significance. Evolution 42:277–292

Toledo MJO, Lana M, Carneiro CM, Bahia MT, Machado-CoelhoGLL, Veloso VM, Barnabe C, Tafuri WL (2002) Impact ofTrypanosoma cruzi clonal evolution on its biological propertiesin mice. Exp Parasitol 100:161–172

Veloso VM, Carneiro CM, Toledo MJO, Lana M, Chiari E, TafuriWL, Bahia MT (2001) Variation in susceptibility to benzni-dazole in isolates derived from Trypanosoma cruzi parentalstrains. Mem Inst Oswaldo Cruz 96:1005–1011

389