ABSTRACT
Human defensins play a fundamental role in the initiation of innate immune responses to some microbial pathogens. Here we show that colonic epithelial model HCT116 cells respond to Trypanosoma cruzi infection by secreting defensin α-1, which reduces infection. We also report the early effects of defensin α-1 on invasive trypomastigotes that involve damage of the flagellar structure to inhibit parasite motility and reduce cellular infection. Short exposure of defensin α-1 to trypomastigotes shows that defensin α-1 binds to the flagellum, resulting in flagellar membrane and axoneme alterations, followed by breaking of the flagellar membrane connected to the trypanosome body, leading to detachment and release of the parasite flagellum. In addition, defensin α-1 induces a significant reduction in parasite motility in a peptide concentration-dependent manner, which is abrogated by anti-defensin α-1 IgG. Preincubation of trypomastigotes with a concentration of defensin α-1 that inhibits 50% trypanosome motility significantly reduced cellular infection by 80%. Thus, human defensin α-1 is an innate immune molecule that is secreted by HCT116 cells in response to T. cruzi infection, inhibits T. cruzi motility, and plays an important role in reducing cellular infection. This is the first report showing a novel cellular innate immune response to a human parasite by secretion of defensin α-1, which neutralizes the motility of a human parasite to reduce cellular infection. The mode of activity of human defensin α-1 against T. cruzi and its function may provide insights for the development of new antiparasitic strategies.
INTRODUCTION
Trypanosoma cruzi is a blood and tissue protozoan parasite that causes the debilitating Chagas disease, which affects millions of people, producing significant morbidity and mortality. Chagas disease is a neglected disease that causes abnormal heart rhythm, heart failure, neurological disorders, sudden cardiac death, and significant pathological alterations in the intestinal tract, such as megacolon and megaesophagus, resulting in serious digestive problems. The disease is transmitted in nature via triatomine or “kissing bug” insects, but transmission can occur through infected blood or organ donation, transplacentally from mother to child, and orally through infected food and drinks. The disease has spread from South and Central America to other continents, where hundreds of thousands of people are now infected, and represents a new serious global health threat due to migration of infected Latin Americans, most of whom are unaware of their infection (1, 2). Unfortunately, the few drugs available to treat Chagas disease are highly toxic for humans and do not cure the chronic phase of the disease, and therapeutic development for Chagas disease remains highly neglected (3, 4). The development of promising new effective and inexpensive anti-T. cruzi pharmacophores specific to T. cruzi targets is in the early stages of investigation (5, 6–9). One highly expensive drug with serious side effects is being tested in clinical trials in Spain (10), and the development of an efficacious prophylactic vaccine faces many challenges and progress is slow, despite several years of effort (11).
The infective trypomastigote form of T. cruzi must rapidly evade the barrage of innate immune mechanisms mounted against it in the host circulatory system by entering and establishing infection in several host mammalian cells, including epithelial cells. During this journey, invasive trypomastigotes manage to evade some innate immune molecules such as complement (12); however, it is unknown whether cells respond to T. cruzi infection by manipulating the expression and secretion of innate immune molecules, such as defensins, to affect trypanosome flagellar structure and motility and reduce infection. The flagellum of invasive trypomastigotes plays an important role in the infection process, propelling the trypanosome to initiate and disseminate the infection in the body. An immune response that damages this critical trypanosome structure would be beneficial for the host.
Defensins are small peptides that are produced particularly by leukocytes and epithelial cells as well as by other cells and have important effector roles in innate immunity against some microbes (13–18). Six human α-defensins have been identified (19), including α-1, α-2, α-3, α-4, HD5, and HD6. Human α-defensins are expressed in neutrophils, Paneth cells, certain macrophage populations, and epithelial cells of multiple tissues in the body (13, 14, 16, 20). α-Defensins are also engaged in cross talk with cells of the immune system. Defensin α-1 is also present in human natural killer cells, B cells, and γδ T cells (21), suggesting that defensin α-1 may regulate these cells. Defensins α-1 to α-3 are chemotactic for human monocytes, human immature dendritic cells, and naive human CD4+ CD45RA+ and CD8+ T cells; enhance systemic IgG responses fostering B and T cell interactions linking innate and adaptive immunity; and bind to the chemokine receptors CCR6 and CCR2 to induce chemotactic responses (reviewed in reference 15). Although significant knowledge on the antiviral and antibacterial properties of defensins is accumulating (22–24), much less is known of their activities against parasites, especially T. cruzi (25).
Human defensin α-1 is a 3.5-kDa, 30-amino-acid peptide that has shown effector functions in host innate immunity against some microorganisms (26, 27), and its secretion is induced in epithelial cells after stimulation with muramyl-dipeptide (MDP) (28), yet the role of human defensin α-1 in neutralizing T. cruzi motility has never been investigated. Here we report novel findings showing that defensin α-1 secretion is induced in colonic epithelial model HCT116 cells after T. cruzi infection. We also report novel findings showing that defensin α-1 damages the flagellar structure of invasive trypomastigote forms of T. cruzi to inhibit parasite motility and reduce cellular infection. This is the first report showing that human defensin α-1 plays an important innate immune role by neutralizing the motility of a human pathogen to reduce cellular infection.
MATERIALS AND METHODS
Parasites.The highly infective trypomastigote clone MMC 20A, derived from the Tulahuen strain of T. cruzi (29), was used. Pure-culture trypomastigotes were obtained from the supernatant of heart myoblast monolayers as described previously (29). Trypomastigotes expressing green fluorescent protein (GFP), generated as described previously (7, 30), were also used.
Cell culture.Human colon epithelial carcinoma (HCT116) cells were obtained from N. Nagathihalli at the Vanderbilt University School of Medicine. These cells were selected for this study since the expression of defensin α-1 in these cells is induced after bacterial MDP stimulation (28), and colonic epithelial model HCT116 cells permitted us to test the hypothesis of whether T. cruzi infection induces the secretion of defensin α-1 in these cells.
Peptides.Mature human defensin α-1 (ACYCRIPACIAGERRYGTCIYQGRLWAFCC) (31, 32) was synthesized and highly purified by reverse-phase high-performance liquid chromatography (HPLC), resulting in a single sharp chromatographic peak with approximately 98% purity (see Fig. S2 in the supplemental material). Defensin α-1 was synthesized with three disulfide bridges: Cys2-Cys30, Cys4-Cys19, and Cys9-Cys29. The demonstration that the three cysteine bridges of defensin α-1 were fully formed with a mass of 3,442.1 was provided by mass spectrometry (see Fig. S3 in the supplemental material). Scrambled defensin α-1 control peptide (CACRPGCRIQYECRARLTAICIGYFAWYCG) was designed with no discernible similarity to existing human proteins by using the control peptide software (http://bioware.ucd.ie/~cyclops/Fergal/tags/PepControls_release_versions/1.3/control_pep_website/control_peptides.html). This control peptide was synthesized, purified to a single chromatographic peak by reverse-phase HPLC with a purity of 98%, and analyzed by mass spectrometry. The peptides were synthesized, HPLC purified, and analyzed by mass spectroscopy by Genemed Synthesis (San Francisco, CA).
Defensin α-1 secretion.A human defensin α-1 enzyme-linked immunosorbent assay (ELISA) kit (MyBiosource, San Diego, CA) was used to determine the level of defensin α-1 secreted into the supernatant of HCT116 cell cultures infected with T. cruzi for 24 h, as described in detail by the manufacturer. Briefly, HCT116 cell monolayers were washed in phenol-red free minimal essential medium (MEM) (Life Technologies, Carlsbad, CA) to remove 10% fetal bovine serum (FBS) and were serum starved overnight in phenol-red free MEM at 37°C with 5% CO2. Starved HCT116 cells were exposed in triplicate to live or 4% paraformaldehyde-fixed trypomastigotes resuspended in phenol-red free MEM at a ratio of 10 parasites per cell for 24 h. Starved cell monolayers were also incubated in phenol-red free MEM alone for 24 h. Test and control cell supernatants were collected in triplicate for evaluation of the levels of defensin α-1 secretion, as described in detail by the manufacturer.
Immunoblots.The expression of defensin α-1 in HCTL116 cells infected with T. cruzi for 24 h or in mock-treated cells was evaluated by immunoblotting as described previously (33, 34). Cell monolayers were solubilized in 0.1% 3-[(3-cholamidopropyl)-dimethylammonio]-1-propanesulfonate (CHAPS) containing protease inhibitors (GE Healthcare, Piscataway, NJ). The same protein concentrations of solubilized samples (5 μg) were separated by SDS-polyacrylamide gel electrophoresis, transferred onto nitrocellulose membranes, probed with goat immunoglobulin G (IgG) specific to defensin α-1, and developed with peroxidase-conjugated mouse anti-goat IgG by enhanced chemiluminescence (GE Healthcare, Piscataway, NJ), as previously described (33, 34). A positive control for defensin α-1 expression in immunoblots included samples of solubilized HCTL116 cells that were previously stimulated with 1 μg of MDP for 24 h, as described in detail previously (28). Five micrograms of solubilized MDP-stimulated HCTL116 cells was used for immunoblot analyses. Blots were stripped and reprobed with monoclonal IgG to β-actin, developed as described above, and scanned by using an Odyssey CLX infrared imaging system (Li-Cor Biosciences, Lincoln, NE).
Motility assays.To investigate the ability of human defensin α-1 to inhibit T. cruzi motility, trypomastigotes (2 × 106 parasites/ml) were incubated in Dulbecco's modified Eagle medium (DMEM) with several concentrations of human defensin α-1 alone or in the presence of goat anti-defensin α-1 IgG or goat preimmune IgG (MBL International Corp., Woburn, MA) for 5 min at 37°C. Control experiments were performed by exposing the same number of parasites to the same concentrations of a scrambled defensin α-1 control peptide or medium alone for the same incubation period. The percentage of mobile trypomastigotes was microscopically determined (35, 36) by using the following formula: percent mobile trypomastigotes = (number of mobile trypomastigotes after treatment/number of motile trypomastigotes in DMEM) × 100. Three independent experiments were performed, each in triplicate. To evaluate whether the short treatment of trypomastigotes with defensin α-1, scrambled defensin α-1, or medium alone affected parasite viability, trypanosomes were exposed to propidium iodide and analyzed by flow cytometry. The positive control included nonviable heated trypanosomes.
Electron microscopy.To investigate the ability of human defensin α-1 to bind to the flagellum of trypomastigotes and cause early and intermediate ultrastructural alterations in the flagellum and flagellar membrane bound to the body of trypomastigotes, samples containing 107 parasites/ml under different conditions, as indicated below, were subjected to scanning electron microscopy (SEM), negative-stain electron microscopy (nsEM), negative-stain immunogold electron microscopy (EM), and transmission electron microscopy (TEM), as described previously (37).
For SEM, trypomastigotes (107 parasites/ml) were exposed to various concentrations of human defensin α-1 (3.7, 5, and 10 nM), 10 nM scrambled defensin α-1 control peptide, or 10 nM defensin α-1 preincubated with anti-defensin α-1 IgG or preimmune IgG for 2 to 7 min, as described above. Cells were processed as described previously (35) and imaged by using a Hitachi 2700 scanning electron microscope.
For nsEM, trypomastigotes (107 parasites/ml) were exposed to 5 nM defensin or 5 nM scrambled defensin α-1 control peptide for 2 to 7 min. Samples were stained with phosphotungstic acid (38) and observed with a Phillips CM-12 electron microscope.
For negative-stain immunogold EM, trypomastigotes (107 parasites/ml) were treated with 3.7 nM defensin α-1 for 2 min and processed as described above for nsEM. Samples were incubated with anti-defensin α-1 IgG or preimmune IgG and probed with donkey anti-goat IgG conjugated with 12-nm gold particles (Amersham Bioscience, Piscataway, NJ). Samples were stained with phosphotungstic acid and observed with a Phillips CM-12 electron microscope.
For TEM, trypomastigotes (107 parasites/ml) were exposed to 10 nM defensin α-1, 10 nM scrambled defensin α-1 control peptide, 10 nM defensin α-1 preincubated with anti-defensin α-1 IgG, 10 nM defensin α-1 preincubated with preimmune IgG, or medium alone at 37°C for different periods of time starting at 2 min. Samples were processed as described previously (39), and ultrathin sections were observed with a Phillips CM-12 electron microscope.
Infection assays.To investigate whether prior exposure of T. cruzi trypomastigotes to defensin α-1 under conditions that inhibit 50% trypomastigote motility would inhibit parasite infection of HCT116 cells, the ability of human defensin α-1 to block T. cruzi infection of these cells was evaluated as described previously (7, 30). GFP-expressing trypomastigotes, obtained as described previously (7), resuspended in phenol red-free DMEM were preincubated in triplicate for 5 min with 10 nM human defensin α-1, 10 nM scrambled defensin α-1 control peptide, medium alone, or 10 nM defensin α-1 preincubated with either anti-defensin α-1 IgG or preimmune IgG. Parasites were washed with phenol red-free DMEM and exposed in triplicate to HCT116 cell monolayers in 24-well plates at a ratio of 10 parasites/cell. After the incubation period, unbound parasites were removed, complete fresh medium was added to the cocultures, and parasite multiplication within cell monolayers at 72 h was determined fluorometrically as relative fluorescence units (RFU), as previously described (7, 30). To microscopically visualize the effect of pretreatment of trypomastigotes with defensin α-1 on cellular infection, we also exposed GFP-expressing trypomastigotes to defensin α-1, scrambled defensin α-1, or medium alone, as described above, and the parasites were then exposed to HCT116 cells at a parasite-to-cell ratio of 10:1 for 72 h, as described previously (7, 30). Cells were fixed and stained with 4′,6-diamidino-2-phenylindole (DAPI) and Alexa Fluor 546 phalloidin for fluorescence confocal microscopy evaluation of infection (7, 30).
To test whether secretion of defensin α-1 induced by T. cruzi infection in HCT116 cells affects T. cruzi infection, we collected supernatants of cells infected with T. cruzi for 24 h or supernatants of noninfected cells, exposed to GFP-expressing trypomastigotes for 5 min at 37°C and then exposed to HCT116 cell cultures at a ratio of 10 parasites per cell in Lab Tech chambers in triplicate. Cocultures were then incubated for 72 h, and cellular infection was quantified as RFU (7, 30), as described above. Additional controls included supernatants of infected cells preincubated with either specific anti-defensin α-1 IgG or the IgG isotype control before exposure to trypanosomes, followed by incubation with HCT116 cells to evaluate the infection under the same conditions as those described above. The culture supernatants from infected HCT116 cells presented a defensin α-1 concentration of 12 ± 1.1 nM (mean ± standard error of the mean of three independently measured supernatants with each experiment being performed in triplicate), as determined by a defensin α-1 secretion ELISA kit, whereas culture supernatants from noninfected cells presented a minimal concentration of 0.06 ± 0.004 nM.
Presentation of results and statistical analysis.All experiments represent three biological replicates, with each experiment being performed in triplicate. Values were expressed as means ± standard error of the means. More than two groups were compared for significance by using one-way analysis of variance (ANOVA), and two groups were compared for significance by using the t test. Differences were considered to be statistically significant if the P value was <0.05.
RESULTS
HCT116 cells respond to T. cruzi infection by secreting defensin α-1, and exposure of trypanosomes to their cellular supernatants containing secreted defensin α-1 reduces cellular infection.We used colonic epithelial model HCT116 cells to determine whether T. cruzi infection of these cells induced the secretion of defensin α-1 into cellular supernatants. These cells were selected for this study because it is known that the expression of defensin α-1 in these cells is induced by bacterial muramyl-dipeptide (MDP) stimulation (28). This cellular model is useful to study the defensin α-1 innate immune response, with implications for megacolon pathology caused by T. cruzi. We found that HCT116 cells responded to T. cruzi infection by inducing the secretion of defensin α-1, as determined by ELISA. Figure 1A shows the defensin α-1 response of HCT116 cells to T. cruzi infection at 24 h. T. cruzi infection induced the secretion of up to 12 nM defensin α-1 by HCT116 cells into cellular supernatants, which represents a 16-fold increase with respect to the minimal basal levels of defensin α-1 present in the cellular supernatants of uninfected cells. The T. cruzi-induced secretion of human defensin α-1 was seen only when HCT116 cells were exposed to live invasive trypomastigotes but not to fixed trypomastigotes (Fig. 1A). We also assessed the expression of defensin α-1 protein in HCT116 cells infected and not infected with T. cruzi by immunoblot analyses. Figure 1B (top) shows that defensin α-1 protein expression was significantly increased in HCT116 cells during T. cruzi infection at 24 h, as evidenced by immunoblotting. The known induction of defensin α-1 protein expression by stimulation of HCT116 cells with MDP (28), detected by immunoblotting (Fig. 1B), represents a positive control of defensin α-1 expression. Scanning of immunoblots normalized with β-actin showed that the expression level of defensin α-1 in solubilized infected HCT116 cells was increased 20 times with respect to the expression level in solubilized noninfected cells (Fig. 1B, bottom). MDP induced expression of defensin α-1 in HCT116 cells by 11-fold (Fig. 1B, bottom). No induction of defensin α-1 was seen in these cells at 2 or 12 h of infection (results not shown). These results indicate that HCT116 cells respond to T. cruzi infection by inducing the expression and secretion of defensin α-1 (Fig. 1A and B). To test the ability of secreted defensin α-1 from culture supernatants of HTC116 cells infected with T. cruzi at 24 h to inhibit cellular infection, we exposed trypomastigotes to cellular supernatants from infected and noninfected cells for 5 min before evaluating the infection at 72 h. Our results, presented in Fig. 1C, indicate that a short exposure (5 min) of trypomastigotes to supernatants containing secreted defensin α-1 at a concentration of 12 nM reduced the infection by 50%, whereas supernatants from uninfected cells did not affect the infection. Preincubation of supernatants containing secreted defensin α-1 with anti-defensin α-1 IgG abolished the inhibitory effect on infection, whereas supernatants from uninfected cells did not have any consequences for cellular infection. Figure 1C indicates the specificity of this effect.
HCT116 cells respond to T. cruzi infection by inducing the expression and secretion of defensin α-1, which reduces cellular infection. (A) HCT116 cells were infected with T. cruzi for 24 h, and secretion of defensin α-1 was measured by ELISA. ∗, P < 0.05. (B) Immunoblot showing defensin α-1 protein expression in HCT116 cells not infected, infected with T. cruzi for 24 h, or stimulated with 1 μg of MDP for 24 h (positive control for defensin α-1 protein expression). Five micrograms of solubilized samples was separated by SDS-PAGE, blotted onto nitrocellulose membranes, probed with IgG specific to defensin α-1, and developed as described in Materials and Methods. The bottom panel shows defensin α-1 expression levels normalized to β-actin expression levels from three independent scanned immunoblots. ∗, P < 0.05. (C) Short exposure of trypomastigotes to supernatants from HCT116 cells containing secreted defensin α-1 reduced cellular infection, and anti-defensin α-1 IgG abolished the inhibitory effect. ∗, P < 0.05.
Human defensin α-1 blocks T. cruzi trypomastigote motility in a concentration-dependent manner.We exposed trypomastigotes to increasing concentrations of human defensin α-1 for 5 min. Our results, shown in Fig. 2, indicate that human defensin α-1 inhibited T. cruzi trypomastigote motility in a peptide concentration-dependent manner. Preincubation of defensin α-1 with IgG against anti-human defensin α-1 abolished the ability of defensin α-1 to block trypomastigote motility, whereas preincubation of defensin α-1 with preimmune IgG alone did not (Fig. 2). Furthermore, preincubation of trypomastigotes with the same concentration of a scrambled defensin α-1 control peptide did not have any consequences for trypomastigote motility (Fig. 2). Defensin α-1 at a concentration of 10 nM blocked trypomastigote motility by 50% with respect to parasites treated with scrambled defensin α-1 or medium alone. Parasite washing did not reverse this effect. We found that the short treatment of trypomastigotes with defensin α-1 that affected trypanosome motility did not alter its viability, as evaluated by staining with propidium iodide and analysis by flow cytometry (see Fig. S1 in the supplemental material).
Defensin α-1 inhibits T. cruzi trypomastigote motility. Trypomastigotes were incubated with several concentrations of defensin α-1 alone or in the presence of defensin α-1 preincubated with anti-defensin α-1 IgG or preimmune IgG for 5 min. Controls included scrambled defensin α-1 control peptide or medium alone. The number of immobilized trypomastigotes was microscopically determined as described in Materials and Methods. Differences in trypomastigote motility between parasites exposed to defensin α-1 or to defensin α-1 in the presence of preimmune IgG and parasites exposed to the scrambled defensin α-1 peptide control, defensin α-1 in the presence of anti-defensin α-1 IgG, or medium alone were statistically significant (P < 0.05). Data are from three independent experiments performed in triplicate.
Defensin α-1 binds to the flagellar membrane of T. cruzi trypomastigotes and damages the trypanosome flagellar structure.A short exposure of trypomastigotes to defensin α-1 for 2 min showed that defensin α-1 binds to the flagellum as evidenced by negative-stain immunogold EM (Fig. 3A) compared to the control (Fig. 3B) and causes early trypanosome cell alterations in the flagellum, as evidenced by nsEM and TEM approaches (Fig. 4). Figure 4 shows the intact flagellum of trypomastigotes exposed to control scrambled defensin α-1 (Fig. 4A) or medium alone (Fig. 4B) observed by nsEM, the normal flagellum of trypomastigotes exposed to scrambled defensin α-1 observed by TEM (Fig. 4E), or normal flagellar axonemes when trypomastigotes were exposed to scrambled defensin α-1 observed by nsEM (Fig. 4G). The very early alterations of the trypanosome flagellum induced by defensin α-1 can be observed by nsEM in Fig. 4C (black arrows) and by TEM in Fig. 4F (white arrows). A longer exposure of defensin α-1 (5 nM) to trypanosomes (5 min) induced more pronounced alterations in the flagellar membrane, as evidenced by the appearance of many pores along the parasite flagellar membrane observed by nsEM (Fig. 4D), compared to trypomastigotes exposed to either scrambled defensin α-1 control peptide or medium alone showing the intact trypanosome flagellum and intact flagellar membrane (Fig. 4A and B). Exposure of trypomastigotes to defensin α-1 (5 nM) for 5 min also damaged the flagellar axonemes at the origin of the trypomastigote flagellum, as indicated by nsEM (Fig. 4H, white arrows), compared to trypomastigotes exposed to scrambled defensin α-1 control peptide, which showed intact flagellar axonemes (Fig. 4G, white arrows). We observed that the early effect of defensin α-1 is at the level of the trypanosome flagellum (Fig. 4C, D, F, and H), while the parasite body remains unaltered (Fig. 4C, D, and H). Analysis of a longer exposure of defensin α-1 (5 nM) to trypanosomes (7 min) by nsEM shows that defensin α-1 breaks the flagellar membrane connected to the body of the trypanosome, causing progressive detachment of the flagellum from the parasite body (Fig. 5B and C), compared to trypomastigotes exposed to scrambled defensin α-1 control peptide, which shows intact trypomastigotes with the flagellum attached to the parasite body by the flagellar membrane (Fig. 5A). The separation of the flagellum from the parasite body induced by defensin α-1 is due to the disruption of the membrane that attaches the flagellum to the body of the trypanosome (Fig. 5D, white arrowheads). We observed that this longer exposure breaks the detached flagellum (Fig. 5E), resulting in release of the flagellum from the trypanosome body (Fig. 5F). Analysis of this longer exposure of trypomastigotes to defensin α-1 by an alternative SEM approach showed that defensin α-1 also breaks the flagellum of the trypanosome while attached to the trypanosome, as indicated in Fig. 5G (arrow), compared to trypomastigotes exposed to scrambled defensin α-1 control peptide, which showed intact trypanosomes (Fig. 5H). Defensin α-1 also causes holes in the flagellum released from the parasite, as observed by SEM (Fig. 5I, arrows). These results show that human defensin α-1 binds directly to the flagellar membrane of trypomastigotes, causes pore formation in the flagellar membrane, damages the flagellar axonemes, breaks, detaches, and releases the flagellum.
Defensin α-1 binds to the flagellum of invasive trypomastigotes. (A) Trypomastigotes were incubated with defensin α-1 for 2 min followed by anti-defensin α-1 IgG, developed with protein A linked to gold particles, and observed by scanning electron microscopy. Arrows point to the defensin α-1 bound to the trypanosome flagellum. (B) Trypomastigotes were incubated with defensin α-1 followed by preimmune IgG incubation, developed with protein A linked to gold particles, and observed by scanning electron microscopy. Bars, 500 nm. These results are representative of three independent experiments with similar results.
The early alterations induced by defensin α-1 in the trypanosome are in the trypomastigote flagellar membrane and its flagellar structure. (A) nsEM of trypomastigotes exposed to scrambled defensin α-1 control peptide for 2 min showing the normal flagellum (f) and normal parasite body (pb). (B) nsEM of trypomastigotes exposed to medium alone for 2 min. (C) Trypomastigotes exposed to defensin α-1 for 2 min, observed by nsEM, showing early flagellar alterations. Arrows point to early flagellar membrane damage. Bars, 500 nm (A to C). (D) Trypomastigotes exposed to defensin α-1 for 5 min and observed by nsEM. A longer exposure to defensin α-1 causes many pores visible along the parasite flagellar membrane. (E) TEM of trypomastigotes exposed to scrambled defensin α-1 control peptide for 2 min showing the normal flagellum. (F) TEM of trypomastigotes exposed to defensin α-1 control peptide for 2 min showing the early flagellar membrane damage, indicated by arrows. Bars, 100 nm (D to F). (G) nsEM of trypomastigotes exposed to scrambled defensin α-1 control peptide for 5 min showing normal flagellar axonemes (indicated by arrows). (H) nsEM of trypomastigotes exposed to defensin α-1 for 5 min showing alterations in the axoneme flagellar structures, indicated by arrows. Bars, 500 nm (G and H). These results are representative of three independent experiments with similar results.
Intermediate alterations caused by defensin α-1 in trypomastigote flagellar structures showing that defensin α-1 breaks, detaches, and releases the trypomastigote flagellum. (A) Trypomastigotes exposed to scrambled defensin α-1 control peptide for 5 min showing the normal flagellum attached to the trypanosome body, as observed by nsEM. (B) Trypomastigotes exposed to defensin α-1 showing early detachment of part of the flagellum from the trypanosome body (indicated by the arrow), as observed by nsEM. (C) Trypomastigotes exposed to defensin α-1 showing a pronounced detachment of part of the flagellum from the trypanosome body (indicated by an arrow). Bars, 2 μm (A to C). (D) Defensin α-1 induces the rupture of the flagellar membrane (arrowheads) that attaches the flagellum (f) to the parasite body (pb). (E) Defensin α-1 induces the break of the detached flagellum from the trypanosome body. The arrow points to the broken flagellum. Bars, 100 nm (D and E). (F) Exposure of trypomastigotes to defensin α-1 releases the base of the flagellum from the trypanosome. Bar, 0.50 μm. (G and H) Defensin α-1 also induces the break of the flagellum while attached to the trypanosome body (G), whereas control scrambled defensin does not (H). The arrow in panel G points to the broken flagellum. (I) Defensin α-1 induces the release of the flagellum, as observed by SEM. Arrows point to the holes in the flagellum. Bars, 2 μm (G to I). These results are representative of three independent experiments with similar results.
Preincubation of trypomastigotes with defensin α-1 under conditions that inhibit trypanosome motility followed by exposure to HCT116 cells significantly reduces T. cruzi infection.We preincubated GFP-expressing trypomastigotes with a concentration of defensin α-1 that inhibits 50% of trypomastigote motility followed by exposure to HCT116 cells and examined cellular infection at 72 h fluorometrically. We observed that exposure of trypomastigotes to 10 nM human defensin α-1 caused a significant reduction of T. cruzi cellular infection at 72 h compared to trypomastigotes exposed to medium alone or scrambled defensin α-1 control peptide (Fig. 6A). Anti-defensin α-1 IgG but not preimmune IgG abolished the inhibitory effect of defensin α-1 in cellular infection under the same conditions, indicating the specificity of defensin α-1 effects on trypomastigote infection (Fig. 6A). Figure 6B shows the inhibition of T. cruzi cellular infection induced by pretreatment of trypomastigotes with defensin α-1 compared to trypomastigotes pretreated with the scrambled defensin α-1 control or medium alone by fluorescence microscopic observation.
Preexposure of trypomastigotes to a concentration of defensin α-1 that inhibits 50% trypanosome motility significantly reduced infection of HCT116 cells at 72 h. (A) GFP-expressing trypomastigotes were pretreated with defensin α-1 alone or in the presence of defensin α-1 preincubated with anti-defensin α-1 IgG or preimmune IgG for 5 min. Controls included scrambled defensin α-1 control peptide or medium alone. T. cruzi infection was determined as RFU. ∗, P < 0.05. (B) Fluorescence microscopic observation of the effect of defensin α-1 pretreatment of GFP-expressing trypomastigotes on infection of HCT116 cells at 72 h. GFP-expressing amastigotes are seen inside host cells: host cell nuclei are stained blue, and their actin filaments are stained red.
DISCUSSION
This is the first report showing a novel innate immune process whereby the host responds to T. cruzi infection by inducing the secretion of defensin α-1, which inhibits cellular infection. Defensin α-1 binds to the trypanosome flagellar membrane, leading to damage of the trypanosome flagellar structure and neutralizing parasite motility. This report expands the biological role of human defensin α-1 against pathogens. Thus, human defensin α-1 plays a beneficial immune role by targeting a critical structure of the invasive trypanosome that propels this organism to initiate and disseminate infection.
Our results elucidate the initial and intermediate effects of defensin α-1 on the trypanosome leading to neutralization of parasite motility, which results in inhibition of cellular infection. The very early events in the toxicity of defensin α-1 against the trypanosome first involve binding of defensin α-1 molecules to the trypanosome flagellar membrane (Fig. 3A). Upon binding, we observed pore formation in the flagellar membrane (Fig. 4C and D) and damage of the flagellar axonemes (Fig. 4H), which are essential core structures of the trypanosome flagellum. The intermediate effects of defensin α-1 in T. cruzi involve detachment of the flagellum from the body of the trypanosome (Fig. 5B and C), mediated by the rupture of the flagellar membrane that attaches the flagellum to the body of the parasite (Fig. 5D), facilitating the initial separation of the flagellum from the trypanosome body (Fig. 5B and C) and resulting in a broken trypanosome flagellum (Fig. 5E) and flagellum release (Fig. 5F). Importantly, defensin α-1 also induces breakage of the flagellum while attached to the body of the trypanosome (Fig. 5G) as well as the presence of holes, which we observed on the released flagellar structure (Fig. 4I). Thus, we conclude that defensin α-1 mounts a potent attack against the trypanosome flagellum. The flagellum is an important organelle that provides motility to the trypanosome, with an average speed of 20 μm/s (40), to allow the trypanosome access to all parts of the body via the blood. Before this report, the role of the T. cruzi flagellum in infectivity was unclear. Thus, we suggest that the generation of motility-deficient strains of T. cruzi by targeting genes for critical trypanosome flagellar proteins generated through genetic manipulation is important to an understanding of the role of flagellar motility in infectivity. Our findings provide direct evidence that neutralization of trypanosome motility compromises its cellular infectivity. Our results show that trypanosomes are still viable after a short treatment with defensin α-1 that affects trypanosome motility (see Fig. S1 in the supplemental material). This indicates that under our experimental conditions, the defensin α-1-induced decrease in cellular infection is not due to trypanosome death.
Since our findings show that human defensin α-1 limits the expansion of invasive T. cruzi trypomastigotes, we suggest that this innate effector process appears to be a frontline defense, which may provide time for more effective adaptive immunity to be generated in the host. Our findings also indicate that the induction of defensin α-1 in cells by T. cruzi requires live organisms. The concentrations of defensin induced by T. cruzi in cells (Fig. 1A) are comparable to plasma levels of α-defensins, which normally vary between 11.6 and 58 nM. Higher levels of defensin α-1 exist at sites of infection, and plasma levels of up to 29,000 nM may occur during infections (41–43). The concentrations of human defensin α-1 that inhibit the motility of T. cruzi trypomastigotes in vitro (Fig. 2) are comparable to the concentrations of human defensin α-1 present in the serum of healthy individuals and to the serum concentrations that are present during inflammation (44, 45). We demonstrate that human defensin α-1 inhibits trypomastigote motility at low-nanomolar concentrations, and this requires a direct association of defensin α-1 with the flagellar membrane of the trypanosome, resulting in flagellar damage. The fact that the innate immune response mediated by defensin α-1 damages the flagellar structure and prevents T. cruzi infection suggests that the flagellar structure may be a target for intervention against T. cruzi infection. The molecules that defensin α-1 targets in the trypanosome flagellar membrane remain unknown and require additional detailed investigation. Small-molecule inhibitors specific to antimicrobial peptide-sensitive sites on the trypanosome flagellar membrane would also likely be a target for interventional applications. As there are currently no effective drugs to cure the chronic phase of the neglected Chagas disease, and due to the fact that T. cruzi has developed resistance to two toxic and commonly used drugs and to experimental drugs (8), we suggest that the core defensin α-1 structure is an attractive candidate for the development of new antitrypanosomal molecules, particularly due to its small size and its potent early effects in neutralizing trypanosome motility and infection.
Since Chagas disease is an inflammatory disease, we suggest that the parasite may be exposed to these concentrations of defensin α-1 in the bloodstream or produced by intestinal mucosal immunity. The colonic epithelial model HCT116 cells used in this work are useful to study the defensin α-1 innate immune response, with implications for megacolon pathology caused by T. cruzi. Furthermore, our results indicate that defensin α-1 is an innate immune molecule that plays a role in neutralizing the motility of invasive T. cruzi trypomastigotes to reduce infection. Therefore, we suggest that human defensin α-1 may play a beneficial role in the host by reducing early infection or during the chronic phase of the disease. The fact that, during the early acute phase of the disease, the peak of parasitemia is followed by a reduction of blood trypomastigotes and that there is very little blood and tissue parasites in the chronic phase may suggest that defensin α-1 may act at these levels. In the other hand, defensin α-1 could interact with serum proteins and other host proteins that may partially reduce it effects in vivo. Our findings indicate that exposure of trypomastigotes to 10 nM defensin α-1 reduces 50% of infection and that a further reduction of cellular infection requires a noticeable increase in the level of defensin α-1 (Fig. 2), which would suggest that exposure of the parasite to a >10 nM concentration of defensin α-1 may stimulate a trypanosome endocytosis response through its cytostome to incorporate and degrade some defensin α-1. A potential way to bolster defensin α-1 to ameliorate infection might be the activation of neutrophils in the host with colony-stimulating factor (CSF), interleukin-8 (IL-8), or tumor necrosis factor alpha (TNF-α) to increase the production of defensin α-1 in the body (46–48). It might also be possible to engineer defensin α-1 molecules to avoid potential undesirable interactions with the host and retain parasite-inhibitory motility, which might also be useful for preventing Chagas disease transfusion.
The fact that defensin α-1 induces a large number of pores of different sizes in the flagellar membrane (Fig. 4D), increasing its fragility while the body of the trypanosome appears normal (Fig. 4C, D, and H), may facilitate the initial separation of the flagellum from the body of the trypanosome (Fig. 5B and C). This separation is due to the rupture of the membrane that attaches the flagellum to the body of the trypanosome, as documented in Fig. 5D. In an previous study, we reported the late toxicity of defensin α-1 on T. cruzi (37). However, the early action of defensin α-1 on T. cruzi, its flagellar membrane, and its motility were unknown and are documented in detail in the present study. The fact that human defensin α-1 causes pores of different sizes in the flagellar membrane (Fig. 4D) is in agreement with the observation that human defensin α-1 forms pores of high conductance in the trypanosome membranes, with diameters ranging from 3 to 200 nm and with a pronounced 10- to 20-nm peak. Single pores formed by peptide monomers tend to fuse, which explains their difference in size (37). Our findings showing that defensin α-1 forms pores in the trypanosome flagellar membrane are also consistent with previous studies showing that defensin α-1 causes pore formation in phospholipid bilayer membrane models in vitro (49). The fact that defensin α-1 induces large pores (50), damages the flagellar membrane of invasive trypomastigotes, and neutralizes their motility and infection also makes this molecule suitable for nanotechnology-based devices for potentially controlling infection. Pore-forming proteins are important components of the immune system. Our results therefore add defensin α-1 to the list of cell-induced pore-forming cytotoxic molecules, such as perforins, which are efficient in primary host protection against invasive microorganisms. In addition to defensin α-1 neutralization of trypanosome motility and infection, this innate immune molecule plays other immunological roles in cellular cross talk (15).
The defensin α-1 peptide used in these studies had approximately 98% purity (see Fig. S2 in the supplemental material) and was synthesized with three disulfide bridges (Cys2-Cys30, Cys4-Cys19, and Cys9-Cys29) fully formed with a mass of 3,442.1, as demonstrated by mass spectrometry (see Fig. S3 in the supplemental material). In addition, IgG specific to defensin α-1 but not the isotype control blocked the observed defensin α-1 activity. All of these observations together support the contention that it is a functional peptide. The defensin α-1 peptide was always handled under physiological conditions and was active at nM concentrations compared to scrambled defensin α-1 in terms of affecting trypanosome flagellum, motility, and cellular infection. Furthermore, the activity of defensin α-1 secreted from supernatants of T. cruzi-infected cells was similar to that the defensin α-1 peptide.
The fact that defensin α-1 at a concentration that inhibits 50% of trypomastigote motility can significantly reduce T. cruzi infection of cells by 80% through direct binding to the flagellar membrane of the trypanosome reinforces the concept that this immune molecule plays an important direct role in interference with T. cruzi infection. This is also supported by the fact that exposure of cellular supernatants from HCT116 cells infected with T. cruzi containing induced defensin α-1 reduces cellular infection (Fig. 1). Our findings showing that defensin α-1 IgG abolished the neutralizing action of human defensin α-1 in trypomastigotes and subsequent infection and that scrambled defensin α-1 does not affect trypanosome motility support the specificity of the effects observed (Fig. 6). The experimental conditions that we used did not evaluate the action of defensin α-1 on the release of trypomastigotes from infected cells in the same coculture, because when trypomastigotes are released from infected cells, there is cellular destruction, making it difficult if not impossible to evaluate cellular infection in the same coculture. Therefore, the infection was evaluated as the multiplication of intracellular amastigotes at 72 h. Under these conditions, there were no trypomastigotes in culture supernatants, which usually are released at 78 h of infection. Therefore, the defensin α-1 secreted from infected cells at 24 h under our conditions could not act on trypomastigotes because they were not released or present. To evaluate the effect of defensin secreted from infected cells on T. cruzi cellular infection, we exposed supernatants of infected cells containing defensin α-1 to trypomastigotes to assess the infection at 72 h in a separate coculture. Our results showed that defensin α-1 secreted from infected cells at 24 h acted on trypomastigotes to significantly reduce cellular infection at 72 h (Fig. 1C). Furthermore, the fact that pretreatment of trypomastigotes with defensin α-1 reduced cellular infection at 72 h (Fig. 6B) confirmed these observations.
Based on our findings, we propose a model of the role of defensin α-1 in the host response to T. cruzi infection (Fig. 7). Accordingly, T. cruzi infection of cells at 24 h induces the secretion of defensin α-1, which reduces cellular infection (Fig. 1). The early action of human defensin α-1 involves its insertion into the trypanosome flagellar membrane to form pores (Fig. 4 D), resulting in structural damage of the flagellum (Fig. 4 and 5), trypanosome paralysis (Fig. 2), and reduction of cellular infection (Fig. 6).
Proposed model of the defensin α-1 host response to T. cruzi infection. The cellular response to trypanosome infection induces the secretion of defensin α-1. Defensin α-1 binds to the flagellar membrane, causing pores and alterations in the flagellar structures, neutralizing parasite motility, and reducing cellular infection.
In summary, these results indicate that human defensin α-1 plays an important beneficial role in neutralizing the motility of trypomastigotes, reducing T. cruzi infection of cells. An understanding of the complex function of human defensins in innate immunity against human protozoan parasites has implications for the prevention and treatment of parasitic diseases, suggesting that defensin α-1 derivatives might be exploited for use as trypanocidal agents against T. cruzi.
ACKNOWLEDGMENTS
This work was supported by NIH grants AI080580, U54 MD007593 (F.V.), and AI007281 (C.A.J., T.C.C., and M.N.M.) and in part by NIH grants GM059994 (M.F.L.), AI083925 (P.N.N.), and MD007586 (S.P.).
We thank Shawn Goodwin for his help with confocal microscopy, Qiujia Shao for her assistance with flow cytometry, and Victor Paromov from our Proteomics Core for his assistance with mass spectrometry.
FOOTNOTES
- Received 30 December 2012.
- Returned for modification 11 March 2013.
- Accepted 14 August 2013.
- Accepted manuscript posted online 26 August 2013.
Supplemental material for this article may be found at http://dx.doi.org/10.1128/IAI.01459-12.
- Copyright © 2013, American Society for Microbiology. All Rights Reserved.
The authors have paid a fee to allow immediate free access to this article.
