Any actual understanding of trypanosomatids in general requires a comprehensive analysis of the less‐specialized species as thorough as our knowledge of the more specialized Leishmania and Trypanosoma. In this context, we have shown by antibody cross‐reactivity that purified extracellular metallopeptidases from Phytomonas françai, Crithidia deanei (cured strain) and Crithidia guilhermei share common epitopes with the leishmanial gp63. Flow cytometry and fluorescence microscopy analyses indicated the presence of gp63‐like molecules on the cell surface of these lower trypanosomatids. Binding assays with explanted guts of Aedes aegypti incubated with purified gp63 and the pretreatment of trypanosomatids with anti‐gp63 antibodies indicated that the gp63‐like molecules are involved in the adhesive process of these trypanosomatids to the A. aegypti gut wall. In addition, our results indicate for the first time that the gp63‐like molecule binds to a polypeptide of 50 kDa on the A. aegypti gut epithelium extract.
Protozoa of the Trypanosomatidae family comprise a large number of species, some of which are agents of important illnesses, such as leishmaniasis, Chagas' disease and African trypanosomiasis, affecting men and animals of economic interest. Phytomonas spp. comprise heteroxenous (two hosts) trypanosomatids found within plant fluids and tissues. These microorganisms are transmitted between plant hosts by phytophagous insects (Dollet, 1984; Camargo, 1999). Phytomonads can be differentiated from one another and from monoxenous (single host) trypanosomatids (as Crithidia spp.), which can also multiply in fruits and be transmitted by the same vectors, in their biological, biochemical and genetic aspects (Dollet, 1984; Camargo, 1999). These trypanosomatids have been isolated from the latex of laticiferous plants, the phloem of trees and from the mature fruits and seeds of many plant families. Phytomonas françai has been found to be associated with a disease known as ‘chochamento das raízes’, which means ‘empty roots’, in the latex of cassava (Manihot esculenta Crantz); the disease is characterized by poor root system development and general chlorosis of the aerial part of the plant (Vainstein & Roitman, 1986). However, further studies on the transmission are required to demonstrate the organism's pathogenicity conclusively.
The genus Crithidia comprises single‐host parasites of insects that present an amastigote and a barleycorn form, called choanomastigote, in their life cycle (McGhee & Cosgrove, 1980). Some of the plant and insect trypanosomatids are used routinely as laboratory models for biochemical and molecular studies because they are easily cultured under axenic conditions, and they contain homologues of virulence factors from the classic human pathogens (Inverso et al., 1993; d'Avila‐Levy et al., 2003, 2005). In addition, trypanosomatids that are not normally infectious to humans (lower trypanosomatids) are also emerging as pathogens in immunosupressed patients, mainly in HIV‐infected individuals (reviewed in Chicarro & Alvar, 2003). Therefore, the importance of studying lower trypanosomatids arises. Particularly relevant is their interaction with hematophagous insects (Chicarro & Alvar, 2003).
Proteolytic enzymes have been implicated in various facets of host–parasite relationships, including parasite attachment, survival and pathogenesis (reviewed in Sajid & McKerrow, 2002). The leishmanial cell surface metallopeptidase, gp63 (EC 184.108.40.206), is a major glycoprotein and is mainly anchored to the plasma membrane of promastigotes by a glycosylphosphatidylinositol (GPI) anchor. This peptidase is the founding member of the M8 class of the zinc peptidase family (McHugh et al., 2004). Several early studies showed that purified gp63, or a monoclonal antibody to gp63, inhibits parasite attachment to or phagocytosis by macrophages and facilitates complement inactivation in serum (Chang & Chang, 1986; Russell & Wilhelm, 1986; Wilson & Hardin, 1988; Brittingham et al., 1995).
In this work, we have assessed by antibody cross‐reactivity the relationship of the purified extracellular metallopeptidases from P. françai (Almeida et al., 2003), Crithidia guilhermei (Melo et al., 2001) and Crithidia deanei (cured strain) (d'Avila‐Levy et al., 2003) with the well‐described gp63 from Leishmania spp. We also searched for a similar protein in these cells by western blotting, flow cytometry and fluorescence microscopy analyses. In addition, we have analyzed the effect of anti‐gp63 antibodies and purified gp63 on the adhesion index of these trypanosomatids to explanted guts from adult female mosquitoes of Aedes aegypti, which is a dipteran vector of prevalent human diseases such as dengue and yellow fever (Gubler, 2004). Finally, we searched for the presence of a receptor capable of binding the gp63‐like molecule in an A. aegypti gut extract.
Materials and methods
Parasites and cultivation
The trypanosomatid Phytomonas françai was provided by Dr Maria Auxiliadora de Sousa (Fundação Oswaldo Cruz, Brazil). Crithidia deanei (cured strain) and C. guilhermei were provided by Dr Maria Cristina M. Motta and Dr Wanderley de Souza, respectively (Instituto de Biofísica Carlos Chagas Filho, Universidade Federal do Rio de Janeiro, Brazil). The trypanosomatids were cultivated as described previously (Melo et al., 2001; Almeida et al., 2003; d'Avila‐Levy et al., 2003). Leishmania (L.) major (MHOM/SU/1973/5‐ASKH) was grown for 4 days at 26°C in Schneider's medium, in order to obtain promastigote forms.
Immunoblot analysis was performed with total cellular extracts from the parasites (200 μg of protein), obtained as described previously (d'Avila‐Levy et al., 2005), and with 1 μg of the purified extracellular metallopeptidases obtained previously from P. françai (Almeida et al., 2003), C. guilhermei (Melo et al., 2001) and C. deanei (cured strain) (d'Avila‐Levy et al., 2003). The primary antibody used was a rabbit antiserum raised against recombinant gp63 (H50) from L. mexicana at 1 : 500 dilution (kindly provided by Dr Peter Overath, Max‐Planck‐Institut für Biologie, Abteilung Membranbiochemie, Germany) (Ilg et al., 1993). The secondary antibody used was horseradish peroxidase‐conjugated goat antirabbit IgG at 1 : 25 000 (Pierce). The membranes were developed by chemiluminescence followed by exposure to an X‐ray film (d'Avila‐Levy et al., 2005).
Flow cytometry and fluorescence microscopy for cell surface gp63
For fluorescence‐activated cell sorter (FACS) and fluorescence microscopy analyses, the parasites (1.0 × 107 cells) were fixed for 30 min at room temperature in 0.4% (volume in volume (v/v)) paraformaldehyde diluted in phosphate‐buffered saline (PBS), followed by extensive washing in the same buffer. The fixed cells maintained their morphological integrity, as verified by microscopic observation. The cells were then incubated for 1 h at room temperature with a 1 : 1000 dilution of rabbit anti‐gp63 antiserum or rabbit preimmune serum, and then incubated for an additional 1 h with a 1 : 250 dilution of fluorescein isothiocyanate‐labeled goat antirabbit immunoglobulin G (IgG). The cells were then washed three times in PBS and observed under a Zeiss epifluorescence microscope (Axioplan 2). The images were digitally recorded using a cooled CCD camera (Color View XS, Analysis GmBH, DE), and analyzed with AnalySIS system software (AnalySIS, DE). Alternatively, the parasite‐associated fluorescence was excited at 488 nm and quantified on a FACSCalibur (BD Bioscience). Nontreated cells, those treated with the preimmune serum (kindly provided by Dr Peter Overath) and those treated with the secondary antibody alone were run in parallel as controls. Each experimental population was then mapped using a two‐parameter histogram of forward‐angle light scatter vs. side scatter. The mapped population (n=50 000) was then analyzed for log green fluorescence using a single‐parameter histogram.
Protozoa–insect gut interaction
Adult female mosquitoes (A. aegypti) were reared and maintained at Laboratório de Fisiologia e Controle de Vetores, Departamento de Entomologia (Instituto Oswaldo Cruz, Rio de Janeiro, Brazil). Live parasites were previously incubated in the presence of anti‐gp63 antibodies or in the presence of rabbit preimmune serum, both at 1 : 2500 dilution, for 1 h at room temperature. Alternatively, the dissected guts were incubated for 15 min in PBS (100 μL) containing 10, 100 or 1000 ng of the proteolytically active purified gp63, heat‐inactivated gp63 or an inactive apoenzyme (Melo et al., 2001; Almeida et al., 2003; d'Avila‐Levy et al., 2003). After that, binding of protozoa to insect guts was performed according to Fampa et al. (2003). Briefly, trypanosomatids (2.0 × 106 in 100 μL) were added to dissected guts that were sliced open longitudinally (10 per group) and incubated for 1 h at room temperature in PBS. The guts were then extensively washed with PBS as found by examination of the explanted guts under an Olympus inverted microscope, and two guts were transferred to microcentrifuge tubes containing 40 μL of PBS and homogenized. The released trypanosomatids were counted in a Neubauer chamber. Trypanosomatid viability was not affected when preincubated with preimmune serum. The results are shown as the mean±standard error (SE) of three independent experiments.
Search for an insect gut receptor for gp63
To obtain the soluble gut extract, 30 insect guts were resuspended in 50 μL of PBS, homogenized with a Teflon‐coated microtissue grinder and vigorously vortexed (three times for 30 s) in the presence of 1% sodium dodecyl sulfate (SDS), followed by centrifugation at 10 000 ×g for 10 min at 4°C. The supernatant obtained after centrifugation corresponded to an insect gut extract and contained approximately 1 μg protein per μL (Lowry et al., 1951). The insect gut extract (50 μg) was separated in 10% SDS‐polyacrylamide gel electrophoresis (PAGE) and transferred onto nitrocellulose membranes. Blotted proteins were blocked in PBS‐containing 0.05% Tween 20% and 10% low‐fat dried milk overnight at 4°C. After blocking, the membrane strips were incubated with soluble gp63 from C. deanei, C. guilhermei or P. françai (5 μg) (Melo et al., 2001; Almeida et al., 2003; d'Avila‐Levy et al., 2003), and sequentially with an anti‐gp63 antibody (1 : 2500) and peroxidase‐conjugated secondary antibody (1 : 25 000). The membrane was revealed by chemiluminescence followed by exposure to X‐ray films.
The experiments were performed in triplicate, in three independent experimental sets. The analysis of variance between groups was performed by means of anova test using EPI‐INFO 6.04 (Database and Statistics Program for Public Health) computer software. P values of 0.05 or less were considered statistically significant.
Results and discussion
Previous work by our group showed that Phytomonas françai (Almeida et al., 2003), C. guilhermei (Melo et al., 2001) and C. deanei cured strain (d'Avila‐Levy et al., 2003) release peptidases during in vitro growth. Metallopeptidases in the 62–67 kDa range have been purified and biochemically characterized (Melo et al., 2001; Almeida et al., 2003; d'Avila‐Levy et al., 2003). The biochemical properties of these enzymes, such as acidic pH, inhibition profile and molecular mass, suggest that they are related to leishmanial gp63, which was confirmed only in C. deanei (d'Avila‐Levy et al., 2003).
To investigate whether the purified peptidases released by P. françai and C. guilhermei cross‐react with the anti‐gp63 polyclonal antibody, we performed an immunoblot assay (Fig. 1). This antibody recognized a single 62 kDa band in the purified fractions from C. deanei and C. guilhermei, and a band of 67 kDa in the purified fraction from P. françai (Fig. 1, lanes P). In an attempt to detect the gp63 homologues in cells, we performed western blotting using total cellular extracts from the three species, which revealed the existence of two protein bands migrating at 67 and 62 that were recognized by the anti‐gp63 antibody (Fig. 1, lanes W). It is possible that the former may correspond to the major secreted isoform detected in the same molecular range in P. françai, while the latter may correspond to the major secreted isoforms detected in C. guilhermei and C. deanei. The promastigote forms of L. major were included as a positive control (Fig. 1).
Released gp63 homologues have been identified in other trypanosomatids, including T. cruzi (Cuevas et al., 2003), Herpetomonas (Elias et al., 2005), Leptomonas (Jaffe & Dwyer, 2003) and Blastocrithidia (d'Avila‐Levy et al., 2005). The present work describes the presence of cell‐associated and released gp63 homologues in P. françai, C. guilhermei and C. deanei. This is the first report on the presence of gp63 homologues in a member of the Phytomonas genus and also in a member of the Angomonas group. The Crithidia genus is highly heterogeneous and comprises at least three distinct groups (Angomonas, Strigomonas and Crithidia) (Du & Chang, 1994; Brandão et al., 2000; d'Avila‐Levy et al., 2001, 2004). Collectively, these data suggest that the secretion of gp63 homologues is a feature common to the Trypanosomatidae family. Moreover, we may speculate that gp63 is subjected to the selective pressures of similar environments (invertebrate host) and it is expected to be structurally and functionally conserved in distinct trypanosomatids.
The gp63‐like molecules were detected on the cell surface of P. françai, C. guilhermei and C. deanei (cured strain), as demonstrated by fluorescence microscopy and flow cytometry analyses (Figs 2 and 3). Curiously, C. deanei and C. guilhermei presented two distinct populations with different affinities for anti‐gp63 antibody (Fig. 3). One explanation for this observation would be that the lack of equal expression is correlated with the parasite growth phase, since flagellate cultures were not synchronized. However, the occurrence of distinct subpopulations with differential expression of surface molecules is well documented in trypanosomatids (Matta et al., 1999; Santos et al., 2002), including surface gp63 (Elias et al., 2005), which could alternatively denote a different expression of surface gp63 molecules or even a diminished accessibility to external ligands in cell subsets or evolutive forms. For instance, it was shown that C. fasciculata presents in the log phase of culture a nonmotile adhesive form and a more elongated form that swims freely (Scolaro et al., 2005). Finally, this could also reflect the presence of a mixed culture with distinct genotypes, which could be assessed by cell cloning.
The gp63 of Leishmania spp. is involved in many basic processes including cellular adhesion (reviewed in Yao et al., 2003). Although direct evidence has never been obtained, it has been suggested that the gp63‐like molecules from lower trypanosomatids might fulfill a nutritional role in these trypanosomatids in the vector midgut, since insect colonization is believed to be the only lifecycle stage common to the monoxenous and heteroxenous flagellate trypanosomatids (Inverso et al., 1993). Accordingly, it has been demonstrated that the down‐regulation of gp63 in L. amazonensis reduces its early development in Lutzomyia longipalpis (Hajmová., 2004). Structural and biochemical similarities exist between gp63 and members of the matrix metallopeptidases (Button et al., 1993; Stocker et al., 1995; McGwire et al., 2003). Parasites must interact with and overcome a variety of obstacles to establish the infection, including extracellular matrix (ECM) proteins. Recently, it was reported that gp63 enhanced the capacity of Leishmania migration through the ECM in vitro. Parasites lightly fixed with glutaraldehyde or the purified gp63 enzyme digested collagen type IV and fibronectin (McGwire et al., 2003). Attachment of Leishmania to native or degradation products of ECM proteins may help facilitate parasite migration and host macrophage activation or migration (Yu & Stamenkovic, 2000).
In order to assess a potential function for the gp63 homologues in lower trypanosomatids, we tested the influence of the anti‐gp63 antibodies on the interaction of P. françai, C. deanei (cured strain) and C. guilhermei with A. aegypti guts. At first, we determined the suitability of A. aegypti as an experimental model to study the interaction between these lower trypanosomatids and hematophagous insects. This particular insect was chosen because it has been shown that Herpetomonas sp. is found repeatedly in this insect (Weinman & Cheong, 1978). It has also been established that Blastocrithidia culicis and C. deanei are able to bind to A. aegypti‐dissected guts in vitro, and that B. culicis can colonize A. aegypti guts (Fampa et al., 2003). Our data showed that the studied protozoa readily adhered to midguts of A. aegypti, possibly mimicking the natural recognition of the midgut epithelium by the parasites. The plant trypanosomatid P. françai interacted more readily with A. aegypti guts. The mean numbers of adhered protozoa were 4.4 × 103, 4.6 × 103 and 1.5 × 104 in C. deanei (cured strain), C. guilhermei and P. françai, respectively (Fig. 4). Curiously, P. françai presented the strongest surface labeling with the anti‐gp63 antibodies, as judged by the mean of fluorescence intensity in the FACS analysis (Fig. 3), which is indicative of large amounts of surface gp63‐like molecules. The binding of anti‐gp63 antibody to P. françai cells was 3.4‐ and 2.2‐fold higher when compared with C. deanei (cured strain) and C. guilhermei, respectively (Fig. 3). The correlation between expression of surface gp63‐like proteins and trypanosomatid adhesion is suggestive of a potential role in cell adhesion.
Since A. aegypti has proven to be a good experimental model to study the interaction between lower trypanosomatids and vectors, we have performed binding assays with parasites previously treated with the anti‐gp63 antibody, at a concentration that did not promote cell agglutination. This treatment led to a significant reduction (P<0.01) in the capacity of adhesion to the guts of C. deanei (cured strain), C. guilhermei and P. françai by approximately 53%, 42% and 48%, respectively. On the other hand, parasites treated with the preimmune serum adhered to the guts at a rate similar to that of the control (P>0.05) (Fig. 4). These observations indicated that the anti‐gp63‐treated trypanosomatids induced a significant reduction in the interaction with the gut epithelial cells. Although these data suggest a probable involvement of the gp63‐like molecules in the binding process of these insect trypanosomatids to the gut wall, it is possible that this inhibition may also be a result of steric hindrance against a different surface receptor. Additional evidence of the role of the gp63‐like molecules in the insect gut wall binding was demonstrated by treating insect guts with 10, 100 or 1000 ng of the purified metallopeptidases from each trypanosomatid before the interaction experiment. Only the treatment of insect guts with 1000 ng of the purified enzymes led to a significant reduction (P<0.01) in C. deanei (cured strain), C. guilhermei and P. françai adhesion (83%, 78% and 86%, respectively). These results indicated that the gp63‐like molecules are relevant in the binding of the trypanosomatid to the gut wall. Alternatively, it could represent the degradation of the insect gut surface receptors by the gp63‐like activity. Therefore, we have tested the effect of an apoenzyme, which retains the tertiary structure but loses the peptidase activity, on the binding process. The apoenzyme also led to a significant reduction (P<0.01) in C. deanei (cured strain), C. guilhermei and P. françai adhesion (79%, 72% and 80%, respectively). In addition, the effect of a heat‐inactivated metallopeptidase on the interaction also showed similar results (77%, 74% and 80%, respectively). These data suggest the saturation of insect gut cell receptors by the gp63‐like polypeptide. Moreover, the similar effects observed with the apoenzyme and the heat‐inactivated metallopeptidases suggest that the peptide sequence recognized by the A. aegypti gut receptor does not depend on the tertiary conformation of the gp63‐like molecule.
The trypanosomatid C. deanei adheres to the A. aegypti gut epithelium. At least one cellular receptor might be involved in this primary interaction stage. The Western blotting analysis identified a polypeptide band of 50 kDa in the insect gut extract as a potential A. aegypti cellular receptor able to recognize the gp63‐like proteins from C. deanei, C. guilhermei and P. françai (Fig. 5, left panel). Conversely, when only anti‐gp63 or secondary antibodies were used as controls, no polypeptide was recognized (Fig. 5, lanes C1 and C2, respectively). The purity of isolated gp63 from C. deanei, C. guilhermei and P. françai is shown by SDS‐PAGE (Fig. 5, right panel), and the protein profile of A. aegypti guts extract is shown for comparison (Fig. 5, lane G). The isolation and sequencing of the A. aegypti cellular receptor remain an open field.
This study does not implicate the gp63‐like molecules as the only adhesive molecule of lower trypanosomatids to A. aegypti gut, nor does it indicate the only biological role of this parasite surface molecule, but it suggests one of the important biological roles of the gp63‐like molecules in the interaction of trypanosomatids with the insect gut epithelium. The concerted action of other surface molecules of the parasite in stabilizing the gp63‐like peptidase activity in the insect gut is very possible. The generation of metallopeptidase knockout mutants of these trypanosomatids is an important direction of future research in our laboratory.
The authors wish to thank Ms. Iêda Coleto Miguel de Castro e Silva for her technical assistance. This study was supported by grants from the Brazilian Agencies: Conselho Nacional de Desenvolvimento Científico e Tecnológico (MCT/CNPq), Conselho de Ensino para Graduados e Pesquisa (CEPG/UFRJ), Fundação de Amparo à Pesquisa do Rio de Janeiro (FAPERJ) and Fundação Universitária José Bonifácio (FUJB).