Molecular and biological characterization of Trypanosoma cruzi strains isolated from children from Jequitinhonha Valley , State of Minas Gerais , Brazil

Introduction: The biological diversity of Trypanosoma cruzi strains plays an important role in the clinical and epidemiological features of Chagas disease. Methods: Eight T. cruzi strains isolated from children living in a Chagas disease vector-controlled area of Jequitinhonha Valley, State of Minas Gerais, Brazil, were genetically and biologically characterized. Results: The characterizations demonstrated that all of the strains belonged to T. cruzi II, and showed high infectivity and a variable mean maximum peak of parasitemia. Six strains displayed low parasitemia, and two displayed moderate parasitemia. Later peaks of parasitemia and a predominance of intermediate and large trypomastigotes in all T. cruzi strains were observed. The mean prepatent period was relatively short (4.2±0.25 to 13.7±3.08 days), whereas the patent period ranged from 3.3±1.08 to 34.5±3.52 days. Mortality was observed only in animals infected with strain 806 (62.5%). Histopathological analysis of the heart showed that strains 501 and 806 caused infl ammation, but fi brosis was observed only in animals infected with strain 806. Conclusions: The results indicate the presence of an association between the biological behavior in mice and the genetic characteristics of the parasites. The study also confi rmed general data from Brazil where T. cruzi II lineage is the most prevalent in the domiciliary cycle and generally has low virulence, with some strains capable of inducing infl ammatory processes and fi brosis.

Chagas disease is caused by the protozoan parasite Trypanosoma cruzi and is an important tropical disease that affects 10 million people worldwide, mostly in Latin America where it is endemic.It is estimated that over 10,000 people die each year due to the clinical manifestations of Chagas disease, which mainly affects the heart and the gastrointestinal tract 1,2 .
It is well established that T. cruzi is a complex taxon that exhibits great genetic variability.Trypanosoma cruzi is distributed into six (I to VI) discrete taxonomic units (DTUs) 3 that show signifi cant differences with respect to their ecological and geographic distributions 4 .The clonal nature of Trypanosoma cruzi 5 predicts a linkage between its genetic variability and biological properties, such as morphology, virulence, pathogenesis, mortality, and drug susceptibility, as demonstrated under experimental conditions [6][7][8][9] .
Furthermore, geographical variations in the prevalence of the different clinical manifestations of Chagas disease have also been observed 1 , and these variations were associated with the genetic heterogeneity of T. cruzi populations 10 .Some authors have demonstrated that different T. cruzi strains cause peculiar tissue lesions due to their specifi c and predominant tropisms in different mammalian cells, such as macrophages, cardiac and skeletal muscle cells, and neurons 11,12 .
Considering the infl uence of the genetic characteristics of T. cruzi populations on their biology, epidemiology 4,8,13 , and therapeutic susceptibility/resistance in vitro 7 and in vivo 9 , the goal of this study was to characteriz e the genetics, biology, and morphology of T. cruzi samples isolated from children living in a vector-controlled area endemic for Chagas disease in Jequitinhonha Valley, State of Minas Gerais, Brazil 14 .This study may aid further studies concerning the clinical and therapeutic management of patients within the region studied.

Patients and Trypanosoma cruzi strains
Eight children were identifi ed through a serological inquiry of Chagas disease in Berilo and José Gonçalves de Minas, Jequitinhonha Valley, State of Minas Gerais (MG), Brazil, from a total of 1,412 samples analyzed 14 .Five children (8-16 years old) were from Berilo, and three (11-16 years old) were from José Gonçalves de Minas.All of the children presented with the indeterminate clinical form of the disease 14 .These municipalities are very close to each other (12km apart).One T. cruzi strain was isolated from each child, and a total of eight T. cruzi strains were isolated by hemoculture 15 .The strains were identifi ed as 1661, 501, 2405, 817, 795, 829, 806, and 855.

Molecular typing
Epimastigote forms obtained from the liver infusion tryptose (LIT) cultures were washed with phosphate-buffered solution (PBS) by centrifugation at 4°C and 3,500rpm.The parasite pellets were stored at -70°C.For molecular typing, polymorphisms of isoenzyme profi les and the triplice assay 16 that explore the combined analysis of 24Sα ribosomal DNA (rDNA) gene, the mini-exon intergenic regions (ITR) gene, and the cytochrome oxidase subunit II (COII) gene polymorphism of T. cruzi were employed.
Mini-exon intergenic regions (ITR) gene: molecular typing of the ITR spliced leader was performed 20 using the primers TcIII (5'-CTCCCCAGTGTGGCCTGGG 3') and UTCC (5'-CGTACCAATATAGTACAGAAACTG-3').This PCR strategy targeting the SL-ITR gene was devised to distinguish populations belonging to T. cruzi III and IV (amplicons of approximately 200bp) from populations of T. cruzi I, II, V, and VI, which present fragments of approximately 150bp.The reactions were performed in a total volume of 15µL containing 20mM Tris-HCl (pH 8.4), 50mM KCl, 3mM MgCl 2 , 250µM of each dNTP, 3µM of each primer, 0.5U Platinum Taq DNA polymerase (Invitrogen, Carlsbad, CA), and 1µL of T. cruzi DNA covered with mineral oil.The PCR amplifi cation cycles consisted of an initial denaturation of 3min at 94°C, annealing at 68°C for 1min, extension at 72°C for 1min and denaturation at 94°C.Every three cycles, the annealing temperature was decreased by 2 degrees to 66, 64, 62, and 60°C.At the fi nal temperature, the number of cycles was increased to 35 and was followed by a fi nal extension step at 72°C for 10min.The PCR products were analyzed by electrophoresis on a 2% agarose gel stained with ethidium bromide.DNA obtained from strains and clones that yielded approximately 200bp (231, T. cruzi III) and 150bp (CL Brener, T. cruzi VI) fragments were used as reference amplicons for the SL-ITR gene.
Polymorphism of the COII gene: for molecular characterization, parasite DNA was extracted 21 , and the mitochondrial COII gene was amplifi ed using the primers Tcmit-10 (5´-CCATATATTGTTGCATTATT-3´) and Tcmit-21 (5´-TTGTAATAGGAGTCATGTTT-3´) to produce a 375bp fragment from T. cruzi maxicircle DNA 22 .Each PCR was performed in a final volume of 15µL containing 10mM Tris-HCl (pH 8.4), 50mM KCl, 0.1% Triton X-100, 1.5mM MgCl 2 (Buffer IB, Phoneutria, MG, Brazil), 1.0U Taq DNA Polymerase (Promega), 250µM of each dNTP, 0.3µM of each primer, and 3ng of DNA template.The amplifi cation program consisted of a denaturation step of 5min at 95°C followed by 40 cycles of a 45sec denaturation step at 95°C, annealing for 45 sec at 48°C, and primer extension for 1min at 72°C.After PCR amplifi cation, the amplicons were digested with AluI (Promega) for 16h according to the manufacturer's instructions.The digested products were analyzed on 6% polyacrylamide gels and visualized by silver staining.COII restriction fragment length polymorphism DNA samples from T. cruzi populations belonging to T. cruzi I (Col1.7G2,mitochondrial haplotype A), T. cruzi II (JG, mitochondrial haplotype C), and T. cruzi VI (CL Brener, which shares the mitochondrial haplotype B with all T. cruzi DTUs) were used as references.

Animals and experimental Trypanosoma cruzi infection:
Female swiss mice (28-30 days old) obtained from the CCA (Centro de Ciência Animal) of the Universidade Federal de Ouro Preto (UFOP), MG, Brazil, were used in this study.

RESULTS
The study was approved by the Ethics Committee in Animal Experimentation of UFOP, MG, Brazil (Process 2009/10).The animals were maintained according to the guidelines of the Colégio Brasileiro de Experimentação Animal (COBEA).
Metacyclic trypomastigotes obtained from LIT cultures were intraperitoneally inoculated into the mice.For each T. cruzi strain, 8 mice were infected with 1.0 x 10 4 blood trypomastigotes from previously infected animals.The following parameters were evaluated: Infectivity and mortality: infectivity (%INF), defi ned as the percentage of mice that presented a positive fresh blood examination (FBE) and/or hemoculture (Hm); and mortality (%MOR), which was registered daily after inoculation and expressed as the cumulative percentage of death throughout the infection period.
Fresh blood examination: fi ve microliters of blood collected from the tail vein were examined on alternate days, and parasitemia was evaluated 23 .The examination began four days after inoculation and continued for fi ve consecutive days after negativation.The results were expressed as the percentage of mice with positive FBE (%+FBE).The following parameters were also assessed: pre-patent period (PPP), patent period (PP), maximum peak of parasitemia (MPP), and day of maximum peak of parasitemia (DMPP).
Morphology of blood trypomastigotes: parasite morphology was evaluated daily by microscopic examination of fresh blood obtained from the mouse tail vein during the acute phase of infection.The percentages of slender, intermediate, and large parasites were recorded 24 .
Hemoculture: thirty days after infection, the animals that had negative FBE results were subjected to hemoculture 25 .Each tube of culture was examined for the presence of parasites after 30, 60, 90, and 120 days, and the results were expressed as the percentage of mice with a positive Hm (%+Hm).
Histopathology: for each T. cruzi strain, three infected mice were euthanized during the chronic phase of the infection (290 days after inoculation).The heart was fixed in 10% buffered formalin (pH 7.2) and embedded in paraffi n.Sections (5µm thick) were mounted on glass slides and stained with hematoxylin-eosin.Morphometric studies of infl ammation involved the analysis of 15 randomly selected fi elds (total area, 1.1 x 10 6 µm 2 ) of tissue sections on a single slide per animal.Infl ammatory infi ltration in the heart was quantifi ed by counting the cellular nuclei.Animals that were not infected showed an average of 200 cellular nuclei in their cardiac muscular tissue.Thus, presence of infl ammation was considered when the heart of a mouse presented more than 200 cellular nuclei.Images registered with a 40× objective were analyzed with Leica QWin software (Leica Microsystems, Wetzlar, Germany).

Statistical analysis
The data for the biological parameters (PP, PM, DPMP, and area under the parasitemia curve) were analyzed using the program Prism 5 for Windows, version 5.0.The Kolmogorov-Smirnov test of normality was used for data corresponding to all parameters.Data with normal distributions were evaluated by analysis of variance (ANOVA) followed by the Newman-Keuls post-test.For the data that were not normally distributed, the non-parametric Mann-Whitney test was employed.The analyses of mortality and infectivity were performed using the chi-square test of homogeneity.Statistical signifi cance was defi ned as p < 0.05.

Ethical considerations
The inclusion of patients in the study and the blood collections were performed after the patients signed a consent form approved by the Ethics Committee for Research in Humans from René Rachou Research Center (CPqRR), of the Oswaldo Cruz Foundation (FIOCRUZ), Belo Horizonte, MG (Process Number 007/02)

Molecular typing
The molecular characterization revealed that all of the T. cruzi strains analyzed had genetic profiles indicative of T. cruzi II 3 .Figures 1A and B show the typical electrophoretic profiles for GPI and G6PD, respectively.
Analysis of the 24Sα rDNA gene revealed the presence of the 125bp fragment (rDNA type 1) in all strains, compatible with T. cruzi II or T. cruzi VI (Figure 2A).The analysis of the amplifi ed products of ITS revealed the presence of 150bp fragments that were compatible with T. cruzi II or T. cruzi VI for all strains (Figure 2B).The restriction fragment length polymorphism (RFLP ) profi les for the COII gene showed that all of the strains had the 81 and 212bp fragments related to mitochondrial haplotype C that are characteristic of T. cruzi II (Figure 2C).

Biological characterization
Infectivity and mortality: the percentage of infectivity (%INF) observed in mice inoculated with six T. cruzi strains (795, 806, 817, 829, 1661, and 855) isolated from children was 100%, as demonstrated by a positive FBE and Hm for all of these strains (Table 1).It was necessary to carry out Hm to confirm infection in three animals.One animal infected with strain 829 was positive by Hm, but the animals infected with strains 501 and 2405 were negative (Table 1).The experimental infections with strains 501 and 2405 both had a %INF of 87.5% (Table 1).No significant differences were observed between strains with respect to this parameter.
No mortality was observed in the majority of the animals infected with the eight T. cruzi strains during the acute and chronic phases of infection; the exception was strain 806, which was associated with a mortality rate of 62.5% (5/8) in the first month of infection (Table 1).Regarding this parameter, this strain was significantly different from the other strains studied.
Samples 806 and 1661 showed moderate parasitemia and presented higher numbers of parasites in the peripheral blood (p≤0.05)than the other strains, but no significant differences were observed between them (Figure 3 and Table 1).Significant differences were also detected between the areas under the parasitemia curves of strains 795 and 829.No differences were detected when the other pairs of strains were compared.
Strains 806 and 1661 were more virulent (higher parasitemia) than the other strains and exhibited the lowest mean PPP (6.0 days for strain 806 and 4.2±0.25 days for strain 1,661, Table 1).These strains showed an earlier mean DMPP, which was 15.75±2.15days for the 806 strain and 17.7±0.79days for the 1,661 strain (Table 1).
Strain 829 exhibited the lowest number of parasites in the peripheral blood during the PP (5.500 trypomastigotes/0.1mL of blood), a longer PPP (13.7±3.08), and a shorter PP (3.3±1.08)compared with the other strains (Table 1).This strain was also the least virulent (Figure 3).
In general, the mean PPPs were relatively short and oscillated between 4.2±0.25 and 13.7±3.08days, whereas the PP was between 3.3±1.08and 34.5±3.52 days.The MPP values oscillated between 5.5±1.45 and 840.9±364.43x 103 trypomastigotes/0.1mL of blood.A DMPP was also observed, with values between 15.75±2.15and 26.25±4.86days.The biological parameters of all of the tested strains of T. cruzi are shown in Table 1.
Morphological analysis of blood trypomastigotes revealed a predominance of intermediate and large forms for all strains.Strain 806 exhibited the highest percentage of slender trypomastigotes, which disappeared as the infection progressed.
Histopathology: histopathological analyses of the hearts of mice infected with the T. cruzi strains showed mononuclear infi ltrate cells during the chronic phase of the infection in the animals infected with strains 501 and 806 (Figure 4), and fi brosis was detected only in animals infected with strain 806 (data not shown).Infl ammation and fi brosis were absent during the chronic phase of the infection in the animals infected with the other strains.

DISCUSSION
Variations in the clinical forms of Chagas disease and in the susceptibility/resistance of T. cruzi to specifi c treatment have been observed during evaluations of patients from distinct endemic regions 1,26 .These differences may be attributed to T. cruzi and host genetic diversity 10 in addition to environmental conditions, which were the same for all experimental groups in this study and were consequently not considered in our analyses 27 .
These observations have prompted several research groups to isolate and characterize different samples of the parasite from distinct hosts and vectors species from domestic, peridomestic, and sylvatic environments [28][29][30][31] .Differences in T. cruzi strains or clones related to biological behavior in vertebrate and invertebrate hosts, resistance to drugs, induction of immune response, and geographical distribution were also demonstrated 4,8,9,11,25,32 .

A B
Because parasite genetics and biology may be associated with the clinical aspects of the disease 33,34 , the goal of the present work was to determine the characteristics of T. cruzi strains isolated from children in the Jequitinhonha Valley, MG, Brazil, an endemic area where the clinical aspects of Chagas disease have not yet been well characterized.Additionally, the correlation between a parasite's genotype and biology was explored.
The molecular typing revealed that all of the strains studied were T. cruzi II, similar to other endemic, central regions of Brazil 16,[35][36][37][38] and other countries of the South Cone in Latin America [39][40][41] .Other researchers have studied samples of T. cruzi isolated from patients residing in Virgem da Lapa, a municipality near the studied region 42 , and they detected the presence of the same genetic group found in our study.However, it is important to note that the author of the Virgem da Lapa 42 study used an isoenzymatic characterization method that is unable to differentiate T. cruzi II from hybrid groups.
The biological characterization showed that the majority of the strains of T. cruzi isolated from children were infective for Swiss mice, presented low virulence because six of eight strains showed low parasitemia, later peaks of parasitemia, had a predominance of large blood trypomastigotes throughout the acute phase, and showed low mortality rates, characteristics well-matched with T. cruzi strains belonging to T. cruzi II 8,11,12 . .For strains 806 and 1661, moderate parasitemia was observed, particularly for strain 806, which displayed a mortality rate of 62.5%.However, during fi ve successive blood passages (data not shown) in animals infected with this same strain, no mortality was observed, what confi rms the low virulence characteristic of T. cruzi II strains.These results corroborate previous studies that demonstrated that the majority of the T. cruzi strains isolated from humans living in different states of central and southern Brazil 11,[43][44][45] , including Virgem da Lapa 42 and Berilo 29 cities of Minas Gerais, presented low virulence and mortality in mice.
The higher parasitemia of strain 806 may be associated with a high percentage of slender blood trypomastigotes early in the infection; this observation is consistent with the results of other studies 11,28,29,43,46 .The PPP and the DMPP for strains 806 and 1,661, the most virulent of the strains investigated, were signifi cantly different and precocious compared with those of the other strains.The parasitemia was signifi cantly higher compared with that of the other strains.However, strain 829 presented a longer PPP, a later DMPP, a shorter PP, and a lower MPP than the other strains studied.The long period for parasitemia detection by fresh blood examination of mice and the lower and later peak of parasitemia observed in animals infected with this strain are characteristics of low virulence compared with the other strains studied.The DMPP was later for the majority of the strains, a common characteristic of strains of low virulence 11,28,29,42 .This observation may be associated with the predominance of large blood trypomastigotes during the acute phase of the infection in mice, as observed in samples from patients in Bahia and Minas Gerais, Brazil 11,28,29,42,43 .
The presence of large trypomastigote forms is related to a later increase in parasitemia and a later DMPP because this form of the parasite is less infective, slower to invade host cells, and, consequently, slower to undergo cellular division.Thus, its growth is delayed compared with the strains with a predominance of slender trypomastigotes, which circulate for a shorter period of time after inoculation 47 .
Histopathological analysis of the hearts of infected animals demonstrated that strains 501 and 806 were the only strains that resulted in an infl ammatory process.Fibrosis was detected only in mice infected with strain 806 and has been observed in several T. cruzi strains isolated from Minas Gerais and Bahia, Brazil 45,48 .The histopathological results observed in this study may be associated with the parasitemia curve or the virulence of the strains because both strains 501 and 806 presented higher parasitemia, similar to strain 1661.We demonstrated a correlation between parasitemia and infl ammation in mice, which is in accordance with the results of other authors 29,42 who have studied cardiotropic T. cruzi strains isolated from patients in endemic areas of Minas Gerais, including the Berilo municipality.
As observed in the present study, other authors 46,47 have demonstrated that strains that multiply slowly present a later peak of parasitemia, are predominantly myotropic, and show a predominance of large blood trypomastigotes during the acute phase of infection.Thus, the present study demonstrated for the fi rst time that T. cruzi strains isolated from children living in the Berilo and José Gonçalves de Minas municipalities in Jequitinhonha Valley, MG, which were genetically characterized as T. cruzi II with appropriate molecular markers 16 , showed similar biological characteristics in mice, as in other experimental studies 7,8,13,49 .

Silva JCVO et al -Characterization of Trypanosoma cruzi strains
Finally, the results obtained in this study reinforce the idea that both the genetic and biological characterization of T. cruzi strains may be important because these methods are the most appropriate to determine whether there is a correlation between these characteristics and the clinical-pathological manifestations of Chagas disease, a polemic and complex theme in Chagas disease 33,34 .Therefore, we estimate that this work may contribute to a better understanding of the state of Chagas disease in the Jequitinhonha Valley.This knowledge may help our team provide clinical assistance and etiological treatment for the patients in the studied region, which is considered one of the most important endemic areas of this disease in our country 50 .

Silva
JCVO et al -Characterization of Trypanosoma cruzi strains

TABLE 1 -
Means of the biological parameters evaluated in Swiss mice inoculated with the Trypanosoma cruzi strains isolated from children living in Berilo and José Gonçalves de Minas, Jequitinhonha Valley, State of Minas Gerais, Brazil.PP: patent period; MPP: maximum peak of parasitemia; DMPP: day of maximum peak of parasitemia.Data are expressed as the mean ± standard error, NP: not performed; %+FBE: percentage of mice with a positive fresh blood examination; %+Hm: percentage of mice with a positive hemoculture; %INF: percentage of infectivity (percentage of mice that presented positive FBE and/or Hm); % MOR: percentage of mortality.