Sympatry influence in the interaction of Trypanosoma cruzi with triatomine

Introduction: Trypanosoma cruzi, the etiologic agent of Chagas disease, is widely distributed in nature, circulating between triatomine bugs and sylvatic mammals, and has large genetic diversity. Both the vector species and the genetic lineages of T. cruzi present a varied geographical distribution. This study aimed to verify the influence of sympatry in the interaction of T. cruzi with triatomines. Methods: The behavior of the strains PR2256 (T. cruzi II) and AM14 (T. cruzi IV) was studied in Triatoma sordida (TS) and Rhodnius robustus (RR). Eleven fifth-stage nymphs were fed by artificial xenodiagnosis with 5.6 × 103 blood trypomastigotes/0.1mL of each T. cruzi strain. Every 20 days, their excreta were examined for up to 100 days, and every 30 days, the intestinal content was examined for up to 120 days, by parasitological (fresh examination and differential count with Giemsa-stained smears) and molecular (PCR) methods. Rates of infectivity, metacyclogenesis and mortality, and mean number of parasites per insect and of excreted parasites were determined. Results: Sympatric groups RR+AM14 and TS+PR2256 showed higher values of the four parameters, except for mortality rate, which was higher (27.3%) in the TS+AM14 group. General infectivity was 72.7%, which was mainly proven by PCR, showing the following decreasing order: RR+AM14 (100%), TS+PR2256 (81.8%), RR+PR2256 (72.7%) and TS+AM14 (36.4%). Conclusions: Our working hypothesis was confirmed once higher infectivity and vector capacity (flagellate production and elimination of infective metacyclic forms) were recorded in the groups that contained sympatric T. cruzi lineages and triatomine species.


INTRODUCTION
Triatomines (Hemiptera: Reduviidae) are recognized as vectors of Trypanosoma cruzi, the etiologic agent of Chagas disease or American trypanosomiasis in Latin America 1,2 .This disease is considered to be the fourth most prevalent parasitic disease, with approximately 300,000 new cases registered annually.It is responsible for approximately 23,000 deaths per year 3 .Currently, it is estimated that approximately 6-7 million people are infected with T. cruzi worldwide 4 .
Under natural conditions, transmission of T. cruzi occurs through vectorial route, which involves contact of the vertebrate hosts with the excreta of vector bugs, contaminated with metacyclic trypomastigotes.This is the primary mechanism for the spread of the disease.The other mechanisms of transmission, referred to as the secondary mechanisms 5 , such as infections caused by blood transfusion, congenital transmission, and organ transplantation, in addition to the occupational and oral exposure 6 , are dependent on the primary mechanism.Although oral transmission is accidental, it has been considered to be responsible for the recent outbreaks of Chagas disease [7][8][9] .
The natural populations of T. cruzi are polyclonal and its subpopulations can be exposed to selective pressure depending on the genetic constitution of the vertebrate host 10,11 , different species of triatomines in which the parasites develop and circulate 12,13 , and genetic constitution of these parasites 14 .However, some aspects of the coevolution of this parasite and vector bugs remain poorly understood 15 .
Different studies on the biological, biochemical, and molecular characterization of T. cruzi suggest that it exhibits high genetic variability 16 .This species has a complex population structure, with isolates distributed in six discrete typing units (DTUs), namely Trypanosoma cruzi I (TcI) to T. cruzi VI (TcVI) 17 .TcI, TcII, TcV, and TcVI DTUs are the main agents of human Chagas disease in the Americas 18 .Moreover, there is a tendency for local T. cruzi strains to more effectively infect vectors from the same geographic areas 19,20 .
Sympatry is the concomitant occurrence, in space and time, of two or more different species in a given geographical area, resulting from total or partial overlap of their respective geographical distribution areas 21 .The vector species as well as the genetic lineages of T. cruzi vary according to the geographical area.Remarkably, the triatomine species that are most frequently found to be infected by T. cruzi in the State of Amazonas, Northern Brazil, belong to the genus Rhodnius whereas those in the State of Paraná, Southern region of Brazil, are Panstrongylus megistus and Triatoma sordida.In contrast, DTUs of T. cruzi found to infect humans in the Amazon region are TcI and TcIV 22 whereas, in Paraná, it is mostly TcII 23 .In this context, to test the hypothesis that hemoflagellate strains from a given geographical region are transmitted more efficiently by vectors from the same region than by those from another geographical area, this study evaluated the influence of sympatry on the parameters related to the biology of interaction between T. cruzi and triatomines.

Triatomines
Triatoma sordida, currently one of the most frequently found triatomine species in Southeastern and Southern Brazil, including the State of Paraná 24,25 , and Rhodnius robustus, which is a vector of Chagas disease in the Brazilian Amazon region, including the State of Amazonas [26][27][28] were used in this study (Figure 1).Colonies of these insects were kindly provided by Dr. José Jurberg of the National and International Reference Laboratory for Triatomines Taxonomy of Oswaldo Cruz Institute, Fiocruz-Rio de Janeiro, and have been kept in the insectarium of the Laboratory of Parasitology, Department of Basic Health Sciences, State University of Maringá (UEM).

Strains of Trypanosoma cruzi
Two strains of T. cruzi belonging to DTUs TcII (PR2256) and TcIV (AM14), were used.The PR2256 strain was isolated in Paraná from a patient in the chronic phase of the infection, from the Municipality of Virgem de Lapa, Minas Gerais, Southeast Region of Brazil (Figure 1).The genetic lineage (DTU) of this strain was previously determined as TcII by polymerase chain reaction (PCR) of ribosomal ribonucleic acid (rRNA) gene and polymerase chain reaction/restriction fragment length polymorphism (PCR/RFLP) analyses of mitochondrial cytochrome c oxidase subunit II (COII) 23 .The AM14 strain was isolated from an acute Chagas disease case that occurred during an oral transmission outbreak in the Municipality of Coarí, Amazonas, Northern Region of Brazil (Figure 1).It was genotyped previously as TcIV by PCR of the mini-exon and rRNA genes and by sequencing of the COII and glucose-phosphate isomerase genes 22 .These strains are kept cryopreserved in the Trypanosomatid Collection of the Chagas Disease Laboratory of State University of Maringá, Paraná.

Experimental groups
The strains PR2256 and AM14 were evaluated using sympatric and non-sympatric vector species, that is, in triatomines species from the same locality and a different locality (T.sordida from Paraná and R. robustus from Amazonas).Thus, the sympatric groups consisted of R. robustus and AM14 (RR+AM14) and T. sordida and PR2256 (TS+PR2256), and the non-sympatric groups consisted of R. robustus and PR2256 (RR+PR2256) and T. sordida and AM14 (TS+AM14).

Artificial xenodiagnosis
Eleven fifth-stage nymphs of each species were individually placed in glass containers with perforated plastic caps and made to fast for 15 days prior to the infective repast.Nymphs were experimentally infected with the aid of an artificial feeder containing mice blood with 5.6 × 10 3 blood trypomastigotes/0.1mL of blood.Artificial xenodiagnosis was performed with heparinized blood collected from five Swiss mice, previously inoculated through intraperitoneal route with metacyclic trypomastigotes from culture in liver infusion tryptose (LIT) medium 29 .Triatomines were weighed before and after the repast, during which time they were individually monitored until total repletion to determine the amount of blood ingested, considering that 1mg gain in weight is equivalent to 1μL of the blood ingested.Insects were kept in a biochemical oxygen demand (BOD) incubator under controlled conditions of temperature (28 ± 1°C) and relative humidity (60 ± 5%), without illumination.
Triatomines were fed, every 20 days, with blood from uninfected mice and kept alone for 4h so that spontaneous excretion could be collected.On the 30 th , 60 th , 90 th , and 120 th day of infective repast, two insects from each group were dissected to collect their intestinal content.Both the biological materials were diluted in 100μL of 0.15M PBS 30 .

Parasitological and molecular evaluation
Excreta and intestinal contents (5µL) were used to perform each of the following methods: 1) fresh examination (FE) under an optical microscope with 400X magnification 31 ; 2) global count (GC) in Neubauer chamber at a dilution of 1:100; and 3) differential count (DC) of parasitic forms in Giemsastained smears.These techniques allowed the confirmation of infection and determination of the proportion (%) of metacyclic trypomastigotes, epimastigotes, and atypical forms (all the forms not included in the previous categories) of each insect.
The excreta and intestinal contents of insects were stored in 70% ethanol prior to deoxyribonucleic acid (DNA) extraction 14 and confirmed for infection by molecular analysis through PCR.Triatomines that tested positive in at least one of four techniques used (FE, GC, DC, and PCR) were considered as infected.

Polymerase chain reaction
DNA extraction: Samples of the triatomine excreta and intestinal content were centrifuged at 2,500rpm for 20 min for removing the 70% ethanol and DNA was extracted using a standard phenol chloroform method 14,32 .Briefly, 500μL of lysis buffer (80mM NaCl/45mM ethylenediaminetetraacetic acid (EDTA), pH 8.0/1% sodium dodecyl sulfate) supplemented with 5μL of 10mg/mL proteinase K (Invitrogen, U.S.A.) was added and the resulting mixture was incubated at 37°C overnight.Following phenol extraction and ethanol precipitation (two volumes), the DNA was resuspended in 10mM Tris-HCl/1mM EDTA, pH 8.0 (TE) and digested with 10mg/mL ribonuclease A (Invitrogen) at 37°C for 2h.After another round of phenol extraction and ethanol precipitation, the DNA was again resuspended in TE buffer (10mM Tris-HCI, pH 8.0/1mM EDTA, pH 8.0) and stored at -20°C until use.

Amplification of the 330-base pair (bp) fragment of the minicircle:
PCR was performed according to Gomes et al. 33 .Primers 121 (5′-AAATAATGTACGGG (T/G)GAGATGCATGA-3′) and 122 (5′-GGTTCGATTGGGGTTGGTGTAATATA-3′), described by Wincker et al. 34 were used to amplify a 330-bp fragment of the kinetoplast DNA minicircle [kinetoplast deoxyribonucleic acid (kDNA)].The PCR was processed by mixing 2µL of the DNA solution from each sample, 10mM Tris-HCl (pH 9.0), 0.1% Triton X-100, 75mM KCl, 3.5mM MgCl2, 0.2mM of each of the deoxynucleotides (dATP, dCTP, dGTP, dTTP; Sigma Company Ltd.), 1U of Taq DNA polymerase (Invitrogen), and 10pmol of each primer for a 10-µL reaction, following the protocol of Gomes et al. 33 .The reaction mixture was subjected to 35 amplification cycles consisting of denaturation at 95°C for 1 min, annealing of primers at 65°C for 1 min, and extension at 72°C for 1 min, in a Techne TC-512 thermocycler.As a control for contamination during the process, for each 8 samples, 2 negative (excreta and intestinal contents of non-infected triatomines) and 2 positive (infected triatomines) original controls from the extraction step, and one negative and one positive-control for the PCR were added.Electrophoresis was performed on 4.5% polyacrylamide gels that were silverstained for visualizing the amplified DNA.The amplification of a 330-bp band suggested the presence of T. cruzi.

Parameters evaluated
The results obtained as means, using different techniques were used as parameters for the four experimental groups (RR+AM14, RR+PR2256, TS+AM14, and TS+PR2256) to determine the following: 1) Percentage of insects with positive fresh examination (%+FE); 2) percentage of positive insects by global count (%+GC); 3) percentage of positive insects by differential count (%+DC); and 4) percentage of positive insects by PCR (%+PCR).Infectivity rate was calculated considering a positive result in at least one of the four methods used.Metacyclogenesis rate was calculated by determination of the proportion of metacyclic forms in relation to the total number of parasitic forms recorded by insect.Mean parasite number in the intestinal content as well as mean the parasite number excreted per insect was determined considering the results of the DC technique.Mortality rate was determined considering the number of dead insects after infective repast, for up to 120 days.

Statistical analyses
Data were statistically analyzed by Bioestat® version 5.3 (Belém, Pará, Brazil).Normality was verified by the Shapiro Wilk test.Differences in proportions and comparison of means were verified by the Z test.The Mann-Whitney test was used to verify the nymph weight differences before and after the infective blood meal.Statistical comparisons were performed between values obtained for each triatomine species, DTU of T. cruzi, and experimental groups (RR+AM14, RR+PR2256, TS+AM14, and TS+PR2256).A statistical significance of 5% was adopted for the tests.

Ethical considerations
Use of human-derived T. cruzi strains was approved by the Ethical Committees of Dr. Heitor Vieira Dourado Tropical Medicine Foundation (process n o 360/07) and of UEM (Process n o 100/04 and 375/07).Experimental animals handling, maintenance, and care were in accordance with the guidelines of the National Council for the Control of Animal Experimentation (CONCEA) and approved by the Ethics Committee on the Use of Animals under Experimentation at UEM (Process n o 023/2014).

RESULTS
Mean weight of triatomines before and after the infective blood meal, amount of blood ingested, and mean number of blood trypomastigotes ingested per experimental group is shown in Table 1.Triatomines presented differences in weight before and after infective repast in all the groups (p < 0.0001).In addition, the volume of blood ingested by R. robustus was higher than the volume of blood ingested by T. sordida (p < 0.0005).

Groups (n = 11)
Initial weight (mg)  Mean and standard deviation of insect weight before and after the blood meal, amount of blood ingested, and mean number of BT ingested per experimental group.Artificial xenodiagnosis with mouse blood infected with TcII (PR2256 strain) or TcIV (AM14 strain) containing 5.6 × 10 3 BT/0.1mL.
BT: blood trypomastigotes; RR: Rhodnius robustus; TS: Triatoma sordida; TcII: Trypanosoma cruzi II; TcIV: Trypanosoma cruzi IV. a There was difference in the weights before and after the blood meal (p < 0.0001).b There was difference in the amount of blood ingested between R. robustus and T. sordida (p ≤ 0.0005).c There was difference in the estimated number of parasites ingested between R. robustus and T. sordida (p ≤ 0.0006); Mann-Whitney test, significance level 5%.
Mean number of ingested parasites was 10,202 and 8,334 parasites for the groups formed by R. robustus + T. cruzi II (RR+PR2256) and R. robustus + T. cruzi IV (RR+AM14), respectively.In the groups involving T. sordida, values were lower.In T. sordida + T. cruzi II (TS+PR2256), insects ingested an average of 2,464 parasites, a higher number when compared to that of T. sordida + T. cruzi IV (TS+AM14) which was 1,182 parasites, with a high standard deviation in the first group.The estimated number of parasites ingested by R. robustus was also higher when compared to T. sordida in both the groups (p ≤ 0.0006) (Table 1).

Positive results in the techniques used and infectivity rate
The ability to detect T. cruzi in triatomines, independently of the vector species and parasite strain, varied significantly (p ≤ 0.0005) with the technique used, and decreased in the order: PCR (positive in 75.1% cases) > DC (34.1%) > GC (27.3%) > FE (25%).The positive results in PCR showed significant differences in 3/4 of the experimental groups when compared to the other techniques used (p ≤ 0.005), which showed similar results (Table 2).The sympatric group RR+AM14 showed the highest rates for all the variables indicative of triatomine infection.
The percentages of triatomines infected with PR2256 strain of T. cruzi II and AM14 strain of T. cruzi IV over 120 days after blood repast, considering the results of excreta and intestinal contents separately, are shown in Figure 2. The %INF was also higher in sympatric pairs, RR+AM14 and TS+PR2256, than in the non-sympatric pairs, RR+PR2256 and TS+AM14.

Mean number of parasites per insect
Considering both intestinal content and excreta, the mean number of parasites per insect was higher in group RR+AM14 (4,545.5 parasites/mL), followed by TS+PR2256 (3,927.3parasites/mL) and RR+PR2256 (54.6 parasites/mL).Parasite count of insects in the non-sympatric group, TS+AM14, was null (data not shown).
Intestinal content: mean number of parasite/mL in the intestinal content was higher in sympatric groups, RR+AM14 and TS+PR2256 (Figure 3A and Figure 3B).In group RR+AM14, approximately 3,900 parasites/mL were recorded in 3/4 days evaluated, except on the 60 th day when the triatomines presented 2,100 parasites/mL.In the same group, on the 30 th day, a greater number of epimastigotes [(EP); 3,500 forms/mL] was observed, and on the 60 th day there was a predominance of atypical forms (AT; 1,700 forms/mL) (Figure 3C).The presence and the highest number of metacyclic trypomastigotes (MT) in the intestinal content (2,000 forms/mL) were visualized from the 90 th day onwards.The sympatric group, TS+PR2256, presented 21,600 parasites/mL only on the 30 th day and was negative on subsequent evaluations (Figure 3B).There was predominance of EP forms (17,500 forms/mL), followed by AT (2,800 forms/ mL) and MT (1,300 forms/mL) forms (Figure 3D).Nonsympatric groups had lower concentrations of parasites in the intestinal content: RR+PR2256 with 100 parasites/mL on the 30 th and 60 th day or null concentrations (TS+AM14) (Figure 3A and Figure 3B).
Excreta: Trypanosoma cruzi replication rate was only characterized in the sympatric group RR+AM14 because the parasitic forms were not observed in excreta of the other groups under the experimental conditions used.
In the RR+AM14 excreta, the highest number of parasites was observed on the 40 th day, coinciding with the highest mean number of MT (1,320 forms/mL), indicating a higher rate of metacyclogenesis (95.1%).On the 60 th day, there was a decline in the mean number of MT (200 forms/mL) and increase in the AT forms (150 forms/mL), with no EP forms visualized.The number of parasites increased again on the 100 th day, with predominance of AT forms (660 forms/mL) compared to MT (400 forms/mL) and EP (226 forms/mL) forms (Figure 4).

Mortality rates
After the artificial xenodiagnosis, insect mortality rates for each experimental group, observed through the first 40 days after the blood meal, were in the following decreasing order: TS+AM14 (27.4%),TS+PR2256 and RR+PR2256 groups (18.9%), and RR+AM14 (0%), without any significant difference among them.

DISCUSSION
The Trypanosoma cruzi-triatomine interaction has been studied by several researchers over the years and although there are several reports 14,[35][36][37][38][39] , many gaps in the knowledge about the parasite cycle in invertebrate hosts exist.However, it is clear that Chagas disease depends highly on the degree of interaction between the vector and parasite 15 .
The study of sympatry in triatomines is an important parameter to evaluate the ability of T. cruzi to replicate and differentiate in the insect gut, because a good interaction between parasite and vector can accelerate the parasite diffusion in nature 40 .Results obtained in the present study demonstrate strong interaction between R. robustus and AM14 strain as well as between T. sordida and PR2256 strain, demonstrating that sympatry may favor both infectivity and transmission capacity of the parasite by the insect vector, because in the groups of allopatric species, which are geographically isolated (R. robustus and PR2256 strain and T. sordida and AM14 strain), the infection was less efficient.
The PR2259 strain used in this study was isolated from a chronic patient residing in Paraná (PR), but the probable place of infection was in Virgem da Lapa, State of Minas Gerais (MG).In Southern and southeastern Brazil, where PR and MG states are respectively located, as well as in the Southern Cone countries of South America, TcII has been the DTU most frequently isolated from patients with chronic Chagas disease 41 and currently T. sordida is one of the most captured triatomine species.
In contrast, the AM14 strain was not isolated from the same locality as R. robustus triatomines used in the study, which would further favor evaluation of the sympatry influence on insect susceptibility to infection and on its vector capacity (i.e., production and elimination of infective forms).However, the AM14 strain was obtained from one acute case during an oral Chagas disease outbreak in Coarí/AM, a State where R. robustus species is implicated as the vector of Chagas disease 26,27 .
Our results are consistent with those of other authors who propose that strains of T. cruzi are biologically adapted to triatomine populations of the same geographic areas 19,20,42 .In addition, the data suggest that the infectivity of triatomine and its vector capacity might be the result of interaction of host genetics with parasite genetics, because group RR+AM14, which although ingested fewer parasites when compared to RR+PR2256, presented higher infectivity rate.This corroborates with the results of other authors who have shown that the percentage of infection is not correlated simply with the amount of the infected blood ingested 43,44 .
However, experimental groups that involved R. robustus had a significantly higher blood volume when compared to the T. sordida groups and, in the case of RR+AM14, presented higher infectivity rate.Analysis of this parameter suggests that this species would be a more efficient vector in transmitting T. cruzi, with respect to the greater blood intake, the shorter time between the end of repast and first defecation, and the greater possibility of infection of a new host by the parasite 45.In this study, the spontaneous release method was used, although not all triatomines defecated during the first 4h after repast.Silva et al. 46 demonstrated that xenodiagnostic reading by this method was more efficient than abdominal compression.A reduction in mortality by approximately three times was also observed using the spontaneous deferral method 47 .
In group RR+PR2256, only epimastigotes and atypical forms were observed in the intestinal content.In studies with another species of the same genus, R. prolixus, the findings indicated that some T. cruzi strains did not develop within its intestinal tract to produce metacyclic trypomastigotes, but it was able to maintain parasites in its rectal lumen 39,48,49 .
In many cases, under our experimental conditions, infection of the insect was only proven by PCR, demonstrating the greater capacity of detection of this technique when compared to the others.Its sensitivity can be influenced by parasite genetics, because strains belonging to different T. cruzi DTUs could have dissimilar DNA content and gene dosage 50 .Moreover, we believe that the DNA detected in the excreta and intestinal contents of the insects comes from intact, extracellular or recently destroyed parasites, indicating the persistence of parasites rather than the persistence of kDNA, as already observed in mammalian hosts 51 .However, the epidemiological importance of optical microscopy is evident, because it can be used to carry out differentiation of developmental stages and consequently evaluation of the rate of metacyclogenesis, an important parameter related to the capacity of species to disseminate in nature.Thus, the combination of more sensitive techniques, such as PCR with other techniques like, FE, GC, and DC, increases the precision of the epidemiological investigation, avoiding false-negative results.
In the present study, it was not possible to characterize the T. cruzi replication rate in 3/4 of the experimental groups.Maintenance of colonies of insects under constant temperature and humidity conditions, and the supply of feed at regular periods, changes the natural conditions, where the insects find climatic variations that influence their metabolism, trophic necessities, and consequently, the biological cycle 52 .These can be the cause for the disparities in the results found by several authors.
In fecal samples of R. robustus infected with AM14 strain, there was a predominance of trypomastigote forms, confirming occurrence of the metacyclogenesis process in this group.Metacyclic trypomastigote forms detach more easily from the intestinal wall by urine action when compared to the epimastigote forms, and are easily drawn to complete the cycle 53 .In the intestinal content of this group, the epimastigote and atypical forms were predominant because the tests were carried out during the periods when they were replicating and suffering differentiation in the intestine of the insect vector.Our data showed that AM14 strain interaction with R. robustus resulted in both the proliferative and infective forms, corroborating other result of our group (AP Abreu: personal communication).
During the course of the triatomine infection, a positive oscillation was observed in the non-sympatric groups (RR+PR2256 and TS+AM14), and on the last day of evaluation (120 th day), all the insects examined were positive.This suggests that the longer the incubation time, the greater was the probability of detecting T. cruzi.Therefore, a broader investigation is important in obtaining more detailed results for the understanding of what happens in nature.
It is also worth mentioning that, in the present study, a considerable mortality was observed in 3/4 groups and the mortality rate was higher in T. sordida (22.7%, 5/22) than in R. robustus (9.1%, 2/22).A study that evaluated the population parameters for T. sordida registered high mortality, suggesting that approximately 50% of the insect population reaches the reproductive age in this species 54 .The low number of insects used for the infections (11 nymphs per group) and the examination of the intestinal contents using only two nymphs of the whole group infected is the major limitation of this study, even when considering that it was performed under similar conditions.Other authors observed a great variability when the same parameters were evaluated using different strains of T. cruzi associated with the same triatomine species.
In conclusion, under controlled temperature and humidity conditions, the experimental groups involving triatomine species and T. cruzi genetic lineages from the same geographic areas (or sympatric) present higher values for the mean number of parasites per insect (both in the intestinal content and excreta) and for the infectivity and metacyclogenesis rates than the groups involving geographically isolated (allopatric or nonsympatric) species.These results indicate a higher susceptibility to infection and greater vector capacity for sympatric groups association (T.cruzi × triatomine vector).

FIGURE 1 -
FIGURE 1 -Map of Brazil showing the geographic distribution of the triatomine species used in the study and the location where Trypanosoma cruzi infection occurs.TcII: Trypanosoma cruzi II; TcIV: Trypanosoma cruzi IV.Source: Adapted from Jurberg 28 .

FIGURE 4 -
FIGURE 4 -Mean number of parasitic forms per milliliter of Trypanosoma cruziIV (AM14 strain) present in Rhodnius robustus excreta until the 100 th day after the blood meal in artificial feeder containing infected mouse blood.

TABLE 2
Positive results of the different techniques used and rate of infectivity for Rhodnius robustus and Triatoma sordida after artificial xenodiagnosis with mouse blood infected with TcII (PR2256 strain) or TcIV (AM14 strain).There was difference between the rates of positive results of the techniques in the same line (p = 0.0036); Values with different symbols (* and #) in the same column show significant differences (p = 0.04); Z test, significance level 5%.