INTRODUCTION
Chagas disease, caused by the protozoan parasite Trypanosoma cruzi, is the main neglected vector borne illness in terms of public health burden in the Latin American region (Hotez et al. Reference Hotez, Bottazzi, Franco-Paredes, Ault and Periago2008). Moreover, it has become an emerging threat in many countries worldwide, where infection by blood transfusion might happen in regions where screening for the disease is not routinely performed in blood banks. In recent times, the massive increase in human migration patterns makes it extremely dangerous to overlook this mode of Chagas disease transmission (Bonney, Reference Bonney2014). The kissing bug Rhodnius prolixus (Hemiptera: Reduviidae) is an important vector of Chagas disease in Northern South America (Schofield and Galvão, Reference Schofield and Galvão2009; Hashimoto and Schofield, Reference Hashimoto and Schofield2012). All developmental stages of this triatomine bug are blood-feeders and several vertebrate species can serve as their food sources. If they feed on mammals infected with T. cruzi, they may become vectors for this parasite (Lent and Wygodzinsky, Reference Lent and Wygodzinsky1979).
Bug infection begins when T. cruzi trypomastigotes are ingested by a triatomine during a blood meal on an infected mammal. After entering the triatomine gut, these forms are transformed into epimastigote forms, which colonize the entire intestinal tract and later differentiate into metacyclic trypomastigotes in the rectum (Garcia et al. Reference Garcia, Genta, de Azambuja and Schaub2010). Trypanosoma cruzi trypomastigotes must experience substantial physiological changes upon arrival to the anterior midgut (AM), the place inside the insect in which parasite–vector interactions develop. Trypomastigotes will be subjected to environmental stresses including osmolarity changes, lower temperature, as well as the presence of digestive enzymes (Kollien and Schaub, Reference Kollien and Schaub2000). Thus, parasites need to react to these new conditions, adapt swiftly or probably be eliminated. The transformations that parasites need to go through inside triatomine gut might define infection success. Therefore, studying the initial phase of gut colonization becomes relevant to understand parasite population dynamics which will influence future successful transmission events. Regardless of the importance of this initial phase for the infection process, few studies about it have been developed (Dias, Reference Dias1934; Eichler and Schaub, Reference Eichler and Schaub2002).
The main aspects of the T. cruzi life cycle were described in 1909 (Chagas, Reference Chagas1909). Since then, several aspects of the interaction of T. cruzi and its invertebrate hosts have been studied (see review in Kollien and Schaub, Reference Kollien and Schaub2000; Garcia et al. Reference Garcia, Genta, de Azambuja and Schaub2010). Dias (Reference Dias1934) described the development of T. cruzi in the invertebrate host, suggesting a reduction in the initial population of parasites before its establishment in the gut. In the same work, the author also mentioned that trypomastigotes differentiated to epimastigotes in the posterior midgut (PM). Nevertheless, posterior studies have assumed that trypomastigotes differentiate to the replicative forms, i.e. the epimastigotes, and to amastigotes/spheromastigotes in the AM of the insect (Azambuja et al. Reference Azambuja, Ratcliffe and Garcia2005; Garcia et al. Reference Garcia, Genta, de Azambuja and Schaub2010). Aiming to clarify some controversial points concerning T. cruzi development in the insect, we evaluated T. cruzi mortality and differentiation in the AM and PM of infected R. prolixus nymphs between 0 and 96 h post-infection (pi)-feeding.
MATERIALS AND METHODS
Organisms
Rhodnius prolixus used in this study were obtained from a laboratory colony derived from insects collected in Honduras around 1990. The colony was maintained by the Vector Behavior and Pathogen Interaction Group from CPqRR-FIOCRUZ. Triatomines were reared at 25 ± 1 °C, 60 ± 10% relative humidity and natural illumination. Insects were fed on chicken and mice anesthetized with intraperitoneal injections of ketamine (150 mg kg−1; Cristália, Brazil) and xylazine (10 mg kg−1; Bayer, Brazil) mixture. Fifth instar nymphs, starved for 30 days after ecdysis, were used in the assays.
CL strain T. cruzi, originally isolated from naturally infected Triatoma infestans (Brener and Chiari, Reference Brener and Chiari1963) was used. Epimastigote forms were cultured by two weekly passages in liver-infusion tryptose supplemented with 15% fetal bovine serum, 100 mg mL−1 streptomycin and 100 units mL−1 penicillin (Fellet et al. Reference Fellet, Lorenzo, Elliot, Carrasco and Guarneri2014). To ensure strain infectivity, parasites were submitted to cycles of triatomine-mice infection every 6 months (Elliot et al. Reference Elliot, Rodrigues, Lorenzo, Martins-Filho and Guarneri2015).
Knockout interferon gamma (INF-γ) mice (B6·129S7) infected with T. cruzi were used in the assays with live hosts, in order to obtain high levels of parasites in circulating blood. All experiments using live animals were performed in accordance to FIOCRUZ guidelines and were approved by its Ethics Committee on Animal Experimentation (CEUA/FIOCRUZ) under protocol number L-058/08.
Insect infection and dissection
Insects were infected by either of two methods: (1) nymphs were fed using an artificial feeder containing citrated heat-inactivated (56 °C 30 min−1) rabbit blood added with a suspension of culture epimastigotes; (2) nymphs were fed on infected mice (therefore ingesting trypomastigote forms). The blood used for insect infection had 1 × 104 epimastigotes µL−1 (artificial feeder) or between 1 × 103 and 4 × 104 trypomastigotes µL−1 (live hosts). Each insect was allowed ingesting 20–30 µL of blood. The AM and PM of fed nymphs were individually dissected and homogenized in 20 µL of phosphate buffered saline (PBS; 0·15 m NaCl at 0·01 m sodium phosphate, pH 7·4) at different times pi to determine the abundance of different parasite forms.
Estimation of the number of parasites recovered in the midgut after bug ingestion
The volume of blood ingested was estimated by weighing nymphs before and immediately after feeding. The number of parasites µL−1 found in the AM, PM, live hosts and artificial feeder blood was assessed by counting them in 5 mm3 of blood as previously described by Brener (Reference Brener1962). In experiments using anesthetized mice as infective sources, a sample of blood was collected from the host tail. The amounts of parasites found in each portion of the insect midgut after blood ingestion (calculated as described immediately above) were determined after adjusting sample volumes according to their dilution in PBS. In the assays evaluating the amount of parasites 3 h after bug ingestion we also estimated the percentage of parasites found in the AM in relation to that present in the initial blood solution.
Approximately 10 µL of each sample were transferred to a 1·5 mL microtube containing 20 µL of sterile PBS and kept at −20 °C for subsequent DNA extraction and quantification by quantitative polymerase chain reaction (qPCR). AM and PM portions were evaluated at 3 h, 1, 2, 4, 7 and 15 d pi.
DNA extraction and qPCR
DNA extraction was performed using the Wizard Genomic DNA Purification Kit (Promega) following manufacturer's instructions for DNA extraction of blood samples. DNA samples extracted from parasites maintained in culture and from the intestinal tract of non-infected triatomines were used for positive and negative controls, respectively. The products obtained from DNA extractions were amplified by qPCR using specific primers for the T. cruzi gene TCZ (5′ GCTCTTGCCCACAMGGGTGC 3′; 5′ CCAAGCAGCGGATAGTTCAGG 3′; Cummings and Tarleton, Reference Cummings and Tarleton2003; Caldas et al. Reference Caldas, Caldas, de Figueiredo Diniz, de Lima, de Paula Oliveira, Cecílio, Ribeiro, Talvani and Bahia2012) in an ABI Prism 7500 Sequence Detection System (Applied Biosystems). The qPCR was performed in a final volume of 25 µL containing 50 ng of DNA, 300 nm of each primer and 12·5 µL Power SYBR® Green PCR Master Mix (Applied Biosystems) under the following conditions: initial denaturation at 95 °C for 10 min followed by 40 cycles of denaturation at 95 °C for 15 s, primer binding and extension at 60 °C for 1 min. Standard curves were generated from five serial dilutions in water (1:10) of a mixture of DNA extracted from blood and cultured parasites following previously described methodology (Caldas et al. Reference Caldas, Caldas, de Figueiredo Diniz, de Lima, de Paula Oliveira, Cecílio, Ribeiro, Talvani and Bahia2012). Blood from non-infected INF-γ knockout mouse was used, and the first point of the curve contained DNA amounts equivalent to 1 × 105 parasites 0·1 mL−1 of blood.
In vitro evaluation of mortality for T. cruzi blood trypomastigotes
To evaluate if compounds originated from R. prolixus tissues could kill parasites, 100 µL of blood from a T. cruzi infected mouse were transferred to 1·5 mL microtubes containing 100 µL Roswell Park Memorial Institute (RPMI) medium without serum (Chuenkova and Pereira, Reference Chuenkova and Pereira2000; Roffê et al. Reference Roffê, Rothfuchs, Santiago, Marino, Ribeiro-Gomes, Eckhaus, Antonelli and Murphy2011) plus:
-
(a) one AM from a unfed nymph;
-
(b) one AM from a nymph recently fed on an uninfected mouse;
-
(c) one salivary gland;
-
(d) only RPMI medium.
The AMs and salivary glands used, respectively, in (a), (b) and (c) were dissected and had their lumens exposed. The number of parasites was quantified by fresh counting (Brener, Reference Brener1962) at 0 and 24 h of incubation. Cell viability was assessed by existence of flagellar movements and using the vital dye Erythrosin B (Sigma-Aldrich, St. Louis, MO, USA). Five replicates were tested for each treatment.
Differentiation of trypomastigotes into epimastigotes in the intestinal tract of R. prolixus
Midgut samples were isolated at different times pi to evaluate the differentiation of trypomastigotes into epimastigote forms. Thin smears of AMs and PMs obtained at different times pi (3, 24, 48, 96 h and 7 days for AM and 24, 48, 96 h and 15 days pi, for PM) were stained by the Giemsa method. The numbers of parasites in 50 microscopic fields were counted under a light microscope (1000× magnification; n = 10–15 AM and PM samples for each experimental point). The distinction between epimastigote and trypomastigote forms was performed according to the classification for Trypanosomatidae (Hoare and Wallace, Reference Hoare and Wallace1966), considering the relative position of both the kinetoplast and the flagellum in the parasite cell body. The forms in which it was not possible to determine the position of the kinetoplast were called intermediate forms. The round-shaped parasites found on triatomine guts were regarded as amastigote-like forms (Tyler and Engman, Reference Tyler and Engman2001; reviewed by Contreras et al. Reference Contreras, Lima and Navarro2006). The proportion of each form was calculated on the total of cells evaluated for each slide.
Statistical analysis
Data normality was tested using the Kolmogorov–Smirnov test. Data showing a normal distribution were analysed by means of t tests or analysis of variance (ANOVA). In the case of ANOVA, pairwise comparisons were performed by means of Tukey post hoc tests. Non-parametric data were analysed by Mann–Whitney or Kruskal–Wallis tests. In the case of Kruskal–Wallis tests, pairwise comparisons were performed by means of Dunn's tests. The correlation between mice parasitaemia and number of parasites recovered in AMs was performed by means of Spearman correlation. The results from the in vitro experiment were analysed by means of an ANOVA, followed by a subsequent Dunnett's test for post hoc comparisons between treatments and the control group. In all cases, differences were considered significant when P < 0·05.
RESULTS
Parasite population profile following infection
Parasite numbers were counted in the insects’ AM 3 h after feeding. Interestingly, only 23% of the estimated ingested epimastigotes and 25% of the trypomastigotes were found. The percentage of T. cruzi recovered from the AMs of insects fed on the artificial feeder was not significantly different from that of bugs fed on live hosts (t test, P = 0·37; n = 20 AMs for each group).
Regardless of which parasite form started infection, parasite number massively decreased over the first 3 days after infection (Fig. 1; Kruskal–Wallis; P = 0·0003 for epimastigotes, P < 0·0001 for trypomastigotes). However, the dynamics of parasite establishment was different depending on whether trypomastigote or epimastigote forms were used for infection. Insects fed on blood containing epimastigote forms suffered a significant reduction until 72 h pi when no parasites were found anymore (Fig. 1A; Dunn, P < 0·05). On the other hand, insects infected with trypomastigotes from infected mice showed a significant reduction in the amount of AM parasites between 3 and 48 h pi (Fig. 1B; Dunn, P < 0·05). No parasite was found in the AMs in late time points (96 and 168 h pi).
Fig. 1. Parasites are absent from the AM 72 h after infection. Temporal profile of the number of parasites µL−1 found in AMs after insects were fed on (A) blood containing Trypanosoma cruzi epimastigotes and (B) T. cruzi infected mice (trypomastigotes forms). The number of parasites µL−1 was estimated by counting parasites in 5 mm3 of blood (Brener, Reference Brener1962). Points represent the number of parasites µL−1 determined for each AM, while horizontal lines represent the median of each group (numbers of samples varied between 2 and 18). Abbreviation: AM, anterior midgut.
Mice used to feed and infect insects showed a large variation in their parasitaemia. To check whether this variation would influence the percentage of recovered parasites in the AM of infected insects, we performed the Spearman correlation test. We found that insect population in AM samples 3 h after ingestion was directly correlated with the number of circulating parasites in corresponding mice (Fig. 2; Spearman's test, r = 0·9148, P < 0·0001). Parasite load in AMs varied overtime regardless of the initial number of ingested parasites (Fig. 3; Kruskal–Wallis, P < 0·0001 for low and high parasitaemia). Insects fed on mice having low parasitaemia (up to 9 × 103 par mL−1) showed statistically significant reductions in parasite amounts from 3 to 48 h and from 24 to 96 h pi (Fig. 3A; Dunn, P < 0·05). Insects fed on mice with high parasitaemia (more than 9 × 103 par mL−1) showed a reduction in the numbers of parasites from 3 to 24 h pi (Fig. 3B; Dunn, P < 0·05). Nevertheless, no parasites were detected in the insect AMs after 96 h pi (Fig. 3A and B).
Fig. 2. Mouse parasitaemia influences the percentage of recovered parasites in insect AM. Correlation between the percentages of parasites found in the insect AM in relation to the amount expected according to the parasitaemia of the blood provided to the insect and the volume ingested and mice parasitaemia. Each point represents the percentage of parasites found in one AM (n = 42). Abbreviation: AM, anterior midgut.
Fig. 3. Number of parasites ingested does not affect infection profile over time. Temporal variation of the number of parasites µL−1 in the Rhodnius prolixus AM after insects had fed on Trypanosoma cruzi infected mice with low (A) and high parasitaemia (B). The number of parasites µL−1 was estimated by counting parasites in 5 mm3 of blood (Brener, Reference Brener1962). Points represent the log of the amount of parasites µL−1 determined for each AM, while each horizontal line represents the median of each group (numbers of samples varied between 5 and 27). Abbreviation: AM, anterior midgut.
Since the dynamics of colonization of T. cruzi in R. prolixus AMs was similar independently of mice parasitaemia, data were thereafter evaluated in combination. In order to confirm that the decrease in parasite number observed over time was not caused by an inability to visualize parasites hindered in the guts (for instance, parasites adhered to the epithelium) we also estimated parasite loads by qPCR. AM parasite numbers varied over time regardless of which quantification method was used (ANOVA, P = 0·0007 and P = 0·0026 for counting parasites in fresh samples and qPCR, respectively). Parasite density in fresh samples showed a reduction of approximately 80% in the first 24 h pi (Fig. 4A; Tukey test P < 0·05). On their turn, qPCR values remained similar during the first 48 h (Tukey, not significant); but a significant reduction in the amount of trypanosome DNA was observed from 48 to 96 h pi (Fig. 4B; Tukey test P < 0·05). Differently from fresh samples evaluation, qPCR estimation showed a remaining amount of parasite DNA in the AMs after 96 h of infection (medians of 136·2, 109·5 and 2·38 parasite DNA at 4, 7 and 15 days, respectively).
Fig. 4. Temporal profile of the number of parasites µL−1 and parasite DNA µL−1 detected in the AM (A, B) and PM (C, D) of Rhodnius prolixus after feeding on Trypanosoma cruzi infected mice. (A, C) The numbers of parasites µL−1 were estimated by counting parasites in 5 mm3 of blood (Brener, Reference Brener1962). (B, D) Values represented were obtained by qPCR. Each point represents the numbers of parasites µL−1 or parasite DNA µL−1 in the specific portion of the gut and each horizontal line represents the median of the group (numbers of samples varied between 3 and 8). Abbreviations: AM, anterior midgut; PM, posterior midgut; qPCR, quantitative polymerase chain reaction.
The numbers of parasites from PMs were also measured by counting parasites on fresh samples and qPCR (Fig. 4C and D). In the fresh quantification, parasite load varied over time (Kruskal–Wallis test, P = 0·005), the values obtained at 15 days pi being significantly higher than those from 2 days pi (Fig. 4C; Dunn, P < 0·05). The amounts of parasite DNA measured by qPCR did not show statistically significant differences along time (Fig. 4D; Kruskal–Wallis test, P = 0·1).
In vitro evaluation of mortality in T. cruzi blood trypomastigotes
To evaluate whether the reduction of parasite numbers observed along the first days of infection was due to vector-produced factors, parasites were incubated for 24 h with extracts from different insect tissues. Results showed that T. cruzi mortality was altered by such treatments (Fig. 5; ANOVA, P < 0·0001). The number of parasites kept in pure RPMI medium for 24 h showed a reduction of about 6% (Fig. 5). Lower numbers of parasites were detected after incubation with salivary glands, AM from unfed insects and AM from fed insects, where a decrease of 15 ± 8, 19 ± 11 and 38 ± 6% was found, respectively. Those parasites incubated with AMs from fed insects showed a significant increase in mortality when compared with the control treatments (Fig. 5; Dunnett, P < 0·05).
Fig. 5. Tissue extracts from Rhodnius prolixus kill Trypanosoma cruzi trypomastigotes. Reduced T. cruzi abundance (percentage) relative to the initial number of parasites µL−1 after 24 h of incubation in RPMI media added with different tissues. The bars correspond to the mean ± s.e. of 5 replicates per treatment (Dunnett, P < 0·05).
Differentiation of T. cruzi trypomastigotes into epimastigotes
Evaluation of different parasite forms present in the AM (Fig. 6A) revealed that 3 h after being ingested, 100% of the T. cruzi observed therein were trypomastigotes (Figs 7A and 6B1). These forms were gradually being substituted by intermediate/amastigote-like forms (Fig. 6B2–4) which corresponded to more than 80% of the few remaining parasites by the 7th day pi (Fig. 7A). No epimastigote forms were observed in the AM during the evaluated period. It is worth mentioning that even if trypomastigote morphology is maintained, parasites could be already under differentiation, morphology per se not being the best signal (although the most widely used) of parasite functional phase.
Fig. 6. Trypanosoma cruzi development in triatomine gut. (A) Dissected intestinal tract of fifth instar nymph 3 days after a blood meal, under a stereomicroscope, showing the AM, PM and RC. (B) Giemsa stained smears showing T. cruzi forms found in AM (upper row, images 1–4) and PM (lower row, images 5–8); trypomastigotes (1), intermediate forms (2, 3), amastigote-like (4, 5) and epimastigotes (6–8). Abbreviations: AM, anterior midgut; PM, posterior midgut; RC, rectum.
Fig. 7. Differentiation into epimastigotes does not happen in the AM. Time course showing Trypanosoma cruzi differentiation (%) from trypomastigotes into intermediate forms and epimastigotes. The percentage of parasite forms was calculated for AMs (A) and PMs (B) by counting 50 microscopic fields in slides stained by the Giemsa method (n = 5 slides evaluated for each time). The mean number of parasites found in each intestinal portion was plotted on the right axis to represent the total parasite burden (all evolutive forms) in each portion and period. Abbreviations: AM, anterior midgut; PM, posterior midgut.
The analyses of the PM (Fig. 6A) were affected by the reduced number of parasites found in this intestinal portion (only 15 ± 6 flagellates could be counted per slide). One day after infection, the dissections showed that almost all parasites were intermediate/amastigote-like forms (Figs 7B and 6B5). The proportion of epimastigote forms started to increase at the 3rd day of infection (Fig. 7B), becoming the forms most frequently observed at 15 days pi (Figs 7B and 6B6–8).
DISCUSSION
When ingested by a triatomine, T. cruzi and blood quickly travel through the bug foregut to reach the AM, which seems to be the first environment where a steady interaction with vector-produced factors takes place. Since several factors that can affect parasite survival are present in this intestinal portion (Pereira et al. Reference Pereira, Andrade and Ribeiro1981; Mello et al. Reference Mello, Azambuja, Garcia and Ratcliffe1996; Azambuja et al. Reference Azambuja, Ratcliffe and Garcia2005), the time spent by parasites inside the AM should have critical implications for their future establishment as a viable/infective metacyclic trypomastigote form in the vector rectum. Although some aspects of the development of triatomine infection by T. cruzi have already been studied, to date most reports focused on times starting at 10 days of infection (Garcia and Gilliam, Reference Garcia and Gilliam1980; Schaub and Böker, Reference Schaub and Böker1986; Kollien et al. Reference Kollien, Schmidt and Schaub1998; Carvalho-Moreira et al. Reference Carvalho-Moreira, Spata, Coura, Garcia, Azambuja, Gonzalez and Mello2003).
As stated earlier, blood trypomastigotes of T. cruzi are the infective forms, acquired by triatomines while feeding on infected mammals. Despite few studies have used blood trypomastigotes to infect their vectors (Carvalho-Moreira et al. Reference Carvalho-Moreira, Spata, Coura, Garcia, Azambuja, Gonzalez and Mello2003; Cordero et al. Reference Cordero, Gentil, Crisante, Ramírez, Yoshida, Anez and da Silveira2008; Botto-Mahan, Reference Botto-Mahan2009), most published works that evaluated parasite–vector interactions used cultured epimastigotes as an infective form for triatomines (Mello et al. Reference Mello, Azambuja, Garcia and Ratcliffe1996; Cortez et al. Reference Cortez, Gonzalez, Cabral, Garcia and Azambuja2002; Azambuja et al. Reference Azambuja, Feder and Garcia2004; Araújo et al. Reference Araújo, Cabello and Jansen2007; Uehara et al. Reference Uehara, Moreira, Oliveira, Azambuja, Lima, Britto, dos Santos, Branquinha and d'Avila-Levy2012). Our analyses using trypomastigotes and epimastigotes for infection, to test whether different developmental forms could modify population dynamics and colonization of R. prolixus AM showed that both forms were severely reduced from the AM soon after ingestion. However, trypomastigotes remained for a shorter time in that intestinal portion. In all assays, flagellate populations decreased with time until it was not possible to detect any forms in fresh counts after 96 h of infection. This was our first indication suggesting that T. cruzi does not multiply in the AM.
The number of parasites ingested by triatomines showed a positive correlation with mice parasitaemia. However, independently of the number of parasites ingested by nymphs, a drastic decrease in their populations was observed in the first 24 h pi. Work done on Glossina morsitans infected with Trypanosoma brucei showed that 99% of ingested parasites were eliminated during the initial developmental stage inside the vector (Van Den Abbeele et al. Reference Van Den Abbeele, Claes, Van Bockstaele, Le Ray and Coosemans1999). Two hypotheses could explain the reduction observed after the first 24 h of infection in our study. The first one proposes that parasites would quickly move from the AM to the PM, where they would differentiate to epimastigotes and subsequently multiply. In this case, a relatively high number of parasites should have been found in the PM over the first hours after infection. However, our results obtained in the PM analyses did not support this hypothesis. Specifically, a small number of parasites was found in PM 24 h pi. Significant higher number of parasites was observed in the PM only after 15 days of infection, although still being greatly reduced in comparison with the initial infection. The evaluation by qPCR of AM and PM samples corroborated that parasites were not migrating to the PM, or staying adhered to the AM epithelium, but having their populations reduced at that portion of the digestive tract. It is worth mentioning that differently from evaluation of fresh samples, qPCR analyses were able to detect the presence of remaining parasite DNA in the AM after 96 h of infection. These results are in agreement with data from Dias (Reference Dias1934) and suggest that a residual population remains in the AM. Future studies will evaluate the role of these parasites in T. cruzi development.
A second hypothesis proposes that the majority of trypomastigotes entering the AM would be eliminated within 24 h pi, as it has been shown for tsetse flies infected by T. brucei blood forms (Turner et al. Reference Turner, Barry and Vickerman1988). Dias (Reference Dias1934) suggested that the initial development of T. cruzi in triatomine would be essentially a regression phase, which would precede the multiplication phase in PM. To test the hypothesis of trypomastigote lysis in AM, we developed an in vitro assay in which blood trypomastigotes were incubated with tissues extracts from different organs of the insect. Salivary glands and the AM from fed insects were included in the assay, since triatomines ingest part of the released saliva during the feeding process (Ribeiro and Francischetti, Reference Ribeiro and Francischetti2003; Soares et al. Reference Soares, Carvalho-Tavares, Gontijo, Dos Santos, Teixeira and Pereira2006). In addition, T. infestans saliva has been shown to contain a pore-forming molecule (trialysin) that lyses T. cruzi trypomastigotes (Amino et al. Reference Amino, Martins, Procopio, Hirata, Juliano and Schenkman2002). A significant parasite killing was only observed when parasites were incubated with the AM from a recently fed insect, suggesting that factors present in the AM after feeding, such as a trialysin-like molecule, play a role in parasite elimination. Several factors have been related to the establishment of T. cruzi infection in the vector, such as lytic factors (Azambuja et al. Reference Azambuja, Guimarães and Garcia1983), lectins (Pereira et al. Reference Pereira, Andrade and Ribeiro1981; Mello et al. Reference Mello, Azambuja, Garcia and Ratcliffe1996) and haemoglobin fragments (Garcia et al. Reference Garcia, Gonzalez, Deazambuja, Baralle, Fraidenraich, Torres and Flawia1995). The induction of immune response factors (Whitten et al. Reference Whitten, Sun, Tew, Schaub, Soukou, Nappi and Ratcliffe2007; Ursic-Bedoya et al. Reference Ursic-Bedoya, Nazzari, Cooper, Triana, Wolff and Lowenberger2008), which also occurs in order to control symbiont population growth (Garcia et al. Reference Garcia, Genta, de Azambuja and Schaub2010), would also contribute to the AM becomes an inhospitable environment for T. cruzi. Recent studies have shown that T. cruzi infection can modulate microbiota growth and suggest that this modulation is necessary to guarantee parasite development (Castro et al. Reference Castro, Moraes, Gonzalez, Ratcliffe, Azambuja and Garcia2012; Soares et al. Reference Soares, Buarque, Queiroz, Gomes, Braz, Araújo, Pereira, Guarneri and Tanaka2015). The population size of Rhodococcus rhodnii, a R. prolixus intestinal symbiont, can show an increase of almost 80 times in the anterior portions of the intestinal tract after the ingestion of a blood meal (Eichler and Schaub, Reference Eichler and Schaub2002). Whether the decrease in parasite population size could also be related with a symbiont-insect immune response-trypanosome interaction is an interesting question for further studies.
Most of the recent studies working with triatomine–T. cruzi interactions assume that after a few days at the AM trypomastigotes differentiate into epimastigotes and spheromastigotes (Garcia and Azambuja, Reference Garcia and Azambuja1991; Kollien and Schaub, Reference Kollien and Schaub2000; Azambuja et al. Reference Azambuja, Ratcliffe and Garcia2005; Garcia et al. Reference Garcia, Genta, de Azambuja and Schaub2010). Our data, however, showed that no T. cruzi epimastigotes can be found in the AM after blood trypomastigote ingestion, indicating that the differentiation into those multiplicative forms, also known as epimastigogenesis, occurs in the PM of R. prolixus. Classical work from Dias (Reference Dias1934) corroborate our data, showing that the differentiation to epimastigotes only begin in the AM, ending after the arrival of intermediate forms at PM, where epimastigotes will replicate.
Studies analysing the differentiation of cultured cell derived trypomastigotes to epimastigotes have showed that in vitro epimastigogenesis occurs by amastigote-like (Albesa and Eraso, Reference Albesa and Eraso1981; De Lima et al. Reference De Lima, Aparicio, Berrocal, Navarro, Graterol and Contreras2007; Graterol et al. Reference Graterol, Arteaga, Navarro, Domínguez, De Lima and Contreras2013) and spheromastigote (Rondinelli et al. Reference Rondinelli, Silva, Carvalho, de Almeida Soares, de Carvalho and de Castro1988) forms that, in turn, give rise to epimastigotes. Furthermore, the occurrence of these intermediary forms seems to be related to nutritional composition of culture medium (Albesa and Eraso, Reference Albesa and Eraso1981). Our results from in vivo experiments suggest a similar differentiation route with the presence of non-replicative amastigote-like intermediary forms. The Giemsa-stained smear analysis suggests that trypomastigotes transform into amastigote-like and intermediate forms, but these amastigote-like forms did not seem to replicate, since cells in division were not found. Considering in vivo observations (Dias, Reference Dias1934) and in vitro studies (Albesa and Eraso, Reference Albesa and Eraso1981; Rondinelli et al. Reference Rondinelli, Silva, Carvalho, de Almeida Soares, de Carvalho and de Castro1988; De Lima et al. Reference De Lima, Aparicio, Berrocal, Navarro, Graterol and Contreras2007; Graterol et al. Reference Graterol, Arteaga, Navarro, Domínguez, De Lima and Contreras2013) together with our results, we propose that differentiation of T. cruzi blood trypomastigotes into epimastigotes occurs mostly through intermediate amastigote-like forms. Nevertheless, distinctly from in vitro studies, in vivo epimastigogenesis show the uniqueness of spatial separation, since complete differentiation and epimastigotes replication occurs only in PM.
One should acknowledge that T. cruzi represents a very diverse group of strains composed of six discrete typing units (DTUs) (reviewed by Zingales et al. Reference Zingales, Miles, Campbell, Tibayrenc, Macedo, Teixeira, Schijman, Llewellyn, Lages-Silva, Machado, Andrade and Sturm2012), indicating that results obtained with specific strains may not be generalized. In our study, we have used the CL strain (group VI), which has been shown to produce similar infection profiles in R. prolixus, when compared with DM28c, a T. cruzi I (TCI) strain (Mello et al. Reference Mello, Azambuja, Garcia and Ratcliffe1996; Uehara et al. Reference Uehara, Moreira, Oliveira, Azambuja, Lima, Britto, dos Santos, Branquinha and d'Avila-Levy2012). We suggest that our findings represent relevant information for understanding the development of T. cruzi in its invertebrate host. However, future work addressing such parasite diversity would be desirable.
In conclusion, we suggest including parasite transformations taking place during the first hours of triatomine infection in the current description of the T. cruzi cycle. Our current model shows that immediately after ingestion, factors present in the AM would be responsible for a significant reduction of the incoming trypomastigote population, leading to a bottleneck in parasite population. The surviving trypomastigotes transform into amastigote-like and intermediate forms that will differentiate into epimastigotes in the PM. Since the process leading to this extensive parasite elimination inside the triatomine AM is still largely unknown, it will be interesting to investigate whether the mechanisms promoting such events are based on immune response activation, antimicrobial factors or other biochemical/physiological processes.
FINANCIAL SUPPORT
This work was supported by Instituto Nacional de Ciência e Tecnologia em Entomologia Molecular (INCT-EM/CNPq), under grant 573959/2008-0 (MGL, AAG) and Programa Estratégico de Apoio a Pesquisa em Saúde (PAPES VI/FIOCRUZ, under grant 407614/2012-5 (AAG).
