Monitoring Trypanosoma cruzi infection in triatomines using PCR in Mato Grosso do Sul , Brazil

Introduction: The aim of the present study was to assess the polymerase chain reaction (PCR) as a method for detecting Trypanosoma cruzi infection in triatomines that had been previously determined by microscopic examination in the State of Mato Grosso do Sul, Brazil. Methods: In total, 515 specimens were collected. Material from the digestive tract of each triatomine was analyzed for the presence of T. cruzi by microscopic examination and PCR using the 121/122 primer set. Results: Among the 515 specimens tested, 58 (11.3%) were positive by microscopy and 101 (19.61%) were positive by PCR and there was an association between the results of the techniques (χ2 = 53.354, p = 0.001). The main species of triatomine identifi ed was T. sordida (95.5%) Conclusions: The use of PCR in entomological surveillance may contribute to a better assessment of the occurrence of T. cruzi in triatomine populations.

Trypanosoma cruzi is a protozoan belonging to the order Kinetoplastida and the Trypanosomatidae family.T. cruzi is the causative agent of Chagas disease and is transmitted by triatomines, which are Hemiptera insects of the Reduviidae subfamily that are characterized by hematophagy (both males and females) from juvenile to adult stages 1 .
T. cruzi in triatomines is identifi ed using optical microscopy and material extracted from the digestive tract of the insect.Although this method is inexpensive and widely used, various drawbacks have been reported, particularly in relation to sensitivity and specifi city 4,5 .
The aim of the present study was to evaluate the frequency of infection with T. cruzi in triatomines in the State of Mato Grosso do Sul, Brazil using polymerase chain reaction (PCR) and microscopic examination (ME).

Triatomine identifi cation and microscopic examination
Triatomines were identifi ed using the dichotomous keys proposed by Carcavallo et al.Flagellated protozoa were detected using the method described by Souza.

DNA extraction and PCR
Genomic deoxyribonucleic acid (DNA) was extracted from the insects as described by Westenberger et al.The integrity of the DNA sample was determined using 0.8% agarose gel electrophoresis, staining with ethidium bromide (0.5µg/ml) and visualization under ultraviolet light.Additionally, one glass slide containing material from the digestive tract of the insects was analyzed.After the fi xed material was scraped from the slide, DNA extraction was performed as described above.
The amplifi cation products were visualized under ultraviolet light after electrophoresis on an agarose gel (2%) and staining with ethidium bromide.

Statistical analysis
The results were compared using the chi-squared test (χ 2 ) with a signifi cance level of 5%.Statistical analysis was performed using MedCalc 12.4.0.0.
In total, 515 samples were analyzed.Among the samples, 58 (11.3%) were positive for fl agellated protozoa as determined by optical microscopy, and 101 (19.6%) were positive for T. cruzi as determined by PCR.
The main species of triatomine identifi ed was T. sordida, which represented approximately 95.5% of the specimens collected, thus confi rming the fi ndings of Almeida et al.The frequency of T. cruzi infection in this species was 10.7% when assessed by microscopy and 18.1% when assessed by PCR.
Association was found between the results obtained by the techniques (χ 2 = 53.354,p = 0.001) (Table 1) and the number of positive samples by PCR was higher than by ME.
PCR was the only technique that has found T. cruzi in triatomines in the municipalities of Rochedo, Corumbá, Aquidauana and Terenos (Table 2) , since ME only confi rms the cases in which the parasite is visible in the test, and the PCR also identifi es the parasite in samples where this was not seen.
Reactions using primers for T. rangeli (data not shown) produced no overlapping data, thereby confi rming that the amplicons were specifi c to T. cruzi.
In Brazil, measures to control triatomine populations are enacted after the detection of insect outbreaks 2 and involve the use of insecticides in the infested areas.This practice has been effective for many years 13,14 .Microscopic examination of material from the digestive tract of the insects is routinely performed to monitor the distribution of fl agellate protozoa (possibly T. cruzi).However, microscopy does not permit The authors declare that there is no confl ict of interest.Advantages of PCR over ME with respect to sensitivity have been reported for T. cruzi 4 , Acanthamoeba spp. 15, Plasmodium spp. 16, Babesia spp. 17, Trypanosoma evansi 18 and Leishmania spp. 19.Advantages related to specificity have also been reported 20 .Furthermore, PCR is more effi cient than ME for the detection of T. cruzi in Triatoma infestans feeding on patients with Chagas diseases, with a difference of 46% between PCR and ME 20 .Previous studies have reported differences of 34.9% (Rhodnius prolixus), 14.5% (Triatoma dimidiata) 21 and 24.6% (T.infestans) 5 between PCR and ME.In the present study, the difference between the techniques (with respect to the frequency of positive results) was 8.4%.Although this value is lower than that of previous studies, PCR still exhibited highest number of triatomines infected by T. cruzi than ME.

CONFLICT OF INTEREST
A combination of PCR and optical microscopy to monitor T. cruzi in triatomines may ensure better results in terms of the presence of the parasite, as shown by the results in the municipalities of Rochedo, Dourados, Corumbá, Aquidauana and Terenos.However, considering the true sensitivity and specifi city of microscopic examination, questions may arise regarding the current distribution of fl agellated protozoa among triatomines, e.g., what is the true distribution of the parasite among triatomines?
In the present study, the proportion of false-negative results in microscopic examination was approximately 8.2%.Considering this issue and taking the data published by Almeida et al. regarding the frequency of triatomines positive for T. cruzi by ME in the State of Mato Grosso do Sul as an example, the number of insects positive for the parasite should be 724 (8.2%) rather than 15 (0.2%) found by Almeida et al.Similar or higher values have been reported in other studies comparing these techniques 4,5,20,21 , which demonstrates that the PCR results may be different from the results reported by the current offi cial surveys.Additionally, results obtained from fi rst-and secondstage nymphs using ME are underestimated because of the diffi culty in obtaining feces during these stages.This diffi culty does not exist when PCR is performed 5 .
In the municipality of Dourados, ME was performed using fi xed material on a microscope slide that was previously classifi ed as inconclusive for T. cruzi.Inconclusive results in ME may be caused by parasite deformation during preparation on the microscope slide, which hinders the identifi cation of the protozoan.In the present study, the identity of the fl agellated protozoa found by microscopy was confi rmed by PCR, which demonstrates the effectiveness of this tool, particularly with respect to controversial points that may arise during the identifi cation process.Other studies have shown that PCR may detect the DNA of parasites on microscope slides even years after preparation 22,23 .
False-positive results are a potential problem with PCR when used for identifi cation of T. cruzi 24 .To reduce the possibility of false-positive results, a negative control was used in the present study.In addition, the different steps of the technique were performed in different rooms.
Although the epidemiological importance of ME is clear in terms of disease control, the use of sensitive techniques, such as PCR, can increase the accuracy of the epidemiological investigation and enable the optimal allocation of fi nancial resources.Furthermore, PCR may be used in epidemiological surveys of other agents of public health importance without signifi cant additional implementation costs.
The frequency of capture of T. sordida was higher than that of the other triatomines, which confi rms the results of previous entomological surveys in which this species was frequently found 3,14,25,26 .Although previous studies have reported a low rate of T. cruzi infection 2,3,25 , the present study suggests that these values may be underestimated and that PCR is essential as a tool for detection in offi cial surveys.
Clearly, one detection technique does not exclude the other.The utilization of ME and PCR in combination contributes greater effi cacy in identifying outbreaks of the parasite and also a more accurate mapping of its distribution.

TABLE 2 -
Triatomines positive for T. cruzi by PCR and microscopy in municipalities of the State of Mato Grosso do Sul, Brazil.
*number of triatomines captured; ** inconclusive; ME: microscopic examination; PCR: polymerase chain reaction; P: Panstrongylus; T: Triatoma.theaccurate identifi cation of T. cruzi because other species of trypanosomes are morphologically indistinguishable from T. cruzi, thereby generating false-positive results.Furthermore, microscopy does not detect the presence of the fl agellates in triatomine as PCR, evidenced in the present study.