|Home | About | Journals | Submit | Contact Us | Français|
Ehrlichia chaffeensis is an obligate, intracellular bacterium, transmitted by the tick Amblyomma americanum, and is the causative agent of human monocytic ehrlichiosis infections. We previously demonstrated that E. chaffeensis is capable of growing in Drosophila S2 cells. Therefore, we tested the hypothesis that E. chaffeensis can infect adult Drosophila melanogaster. Adult Drosophila organisms were experimentally challenged with intra-abdominal injections of bacteria. Ehrlichia-infected flies showed decreased survival compared to wild-type flies, and bacteria isolated from flies could reinfect mammalian macrophages. Ehrlichia infections activated both the cellular and humoral immune responses in the fly. Hemocytes phagocytosed bacteria after injection, and antimicrobial peptide pathways were induced following infection. Increased pathogenicity in flies carrying mutations in genes in both the Toll and Imd pathways suggests that both immune defense pathways participate in host defense. Induction of Drosophila cellular and humoral responses and the in vivo replication of E. chaffeensis suggests that D. melanogaster is a suitable host for E. chaffeensis. In the future, it will be a useful tool to unlock some of the in vivo mysteries of this arthropod-borne bacterium.
Drosophila melanogaster is a valuable tool for studies focused on innate immune responses. It is especially attractive because of the ability to study innate host defense without the complicating variables of acquired immunity (28, 58). Drosophila innate immunity involves both cellular and humoral components. The cellular immune response involves phagocytosis, encapsulation, and/or melanization of pathogens via hemocytes, plasmatocytes, or crystal cells, respectively (18, 31, 63). Humoral immunity involves the production of antimicrobial peptides through either the Toll or immune deficiency (Imd) pathway (18, 31, 63). The Toll pathway is activated by gram-positive bacteria or fungi and elicits production of the antimicrobial peptide Drosomycin (18, 31, 63). The Imd pathway is activated by gram-negative bacteria and is characterized by the production of antimicrobial peptides such as Attacin and Diptericin (18, 31, 63). Additionally, D. melanogaster's completed genome, ease of manipulation, availability of mutants, and homology to vertebrate systems make it an attractive tool as a model system for detailing the innate immune responses to various pathogens. In particular, it has been used to characterize immune reactions elicited in response to Erwinia carotovora, Mycobacterium marinum, Plasmodium gallinaceum, Francisella tularensis, Serratia marcescens, Listeria monocytogenes, and Salmonella enterica serovar Typhimurium (3, 7, 13, 40, 42, 52, 64).
Ehrlichia chaffeensis is an obligate, intracellular bacterium and is vectored by Amblyomma americanum (lone star tick). It is the causative agent of human monocytic ehrlichiosis, which can be particularly life-threatening in young, elderly, and/or immunocompromised patients. In 2006, the CDC reported an infection rate of 0.2/100,000 persons in the United States (41). E. chaffeensis is classified as a gram-negative bacterium, but it lacks the genes necessary for the synthesis of peptidoglycan or lipopolysaccharide (LPS) (37). Therefore, many questions exist about early host resistance to Ehrlichia as well as about the host genetic requirements for bacterial growth. Drosophila melanogaster could be a useful tool to address these questions. We have established that E. chaffeensis can infect and replicate in the hemocytic, macrophage-like Drosophila S2 cell line (38). We hypothesized that E. chaffeensis would infect adult flies and activate host defenses. We present evidence indicating that Ehrlichia can infect and replicate in adult Drosophila, that hemocytes respond to the infection, and that Drosophila humoral immune pathways are activated.
The canine macrophage cell line DH82 was maintained at 37°C in Dulbecco's modified Eagle's medium with 3.5% fetal bovine serum and 3.5% Nu serum (DMEM-7). The E. chaffeensis Arkansas isolate was continuously cultivated in the DH82 cell line at 37°C with 8% CO2 in DMEM-7. Bacteria were passaged when infectivity reached 80 to 90% as visualized by using cytospin-prepared slides (stained with Hema3 fixative and Dif-Quik stain) to monitor the formation of morulae in the cells. Infected cells were removed by scraping each plate with a cell scraper, transferring the culture to a conical tube, and shaking the suspension with glass beads. The liberated bacteria were purified by centrifuging the suspension at 600 × g for 20 min to remove cell debris. The supernatant containing bacteria was removed, transferred to a sterile conical tube, and centrifuged at 15,000 × g for 20 min to pellet the free bacteria. The final supernatant was removed and discarded, and the pellet was resuspended in sterile phosphate-buffered saline (PBS).
For preparation of dead bacterial cultures, the final pellet was resuspended in sterile PBS and the tube was subsequently placed in a boiling water bath for 15 min. The boiled bacteria was then centrifuged at 15,000 × g for 20 min to repellet the bacteria. The supernatant was removed, and the dead bacterial pellet was resuspended in 5 ml sterile PBS.
Infected flies were used to reinfect DH82 cells. Ten flies that had been infected for 168 h were anesthetized using CO2 and transferred to a sterile, 1.5-ml conical tube (Kimble Kontes no. 749510-1500). Flies were homogenized with a disposable pestle (Kimble Kontes no. 749521-1590) in 1 ml of sterile PBS. The homogenate was spun at 10,000 × g for 10 s in a microcentrifuge, and the resulting supernatant was added to a 150-mm plate of 40% confluent DH82 cells (treated with 0.13 μg per ml of amphotericin B [Fungizone]). Cells were observed for the formation of morulae, using cytospin slide preparations. At 2 weeks after infection, morula formation was observed in 90 to 100% of the cells. RNA extraction and reverse transcription-PCR (RT-PCR) analysis for the Ehrlichia 16S ribosomal gene was performed on the infected and on uninfected DH82 cells as previously described (38).
The number of bacteria used for infection experiments with DH82 cells and flies was estimated using TaqMan-based real-time RT-PCR as previously described (55). This TaqMan-based assay targets the E. chaffeensis 16S rRNA gene transcripts. Detection of the 16S rRNA is 100 times more sensitive than detection of rRNA genes (20, 55). We have confirmed that there is no difference in the relative levels of 16S rRNA and rRNA genes during the course of an infection using real-time quantitative RT-PCR (qRT-PCR) and quantitative PCR. qRT-PCR was performed on 10-fold serial dilutions of RNA extracted from 80 to 100% infected DH82 cells (three different samples) using a Smart Cycler system (Cepheid, Sunnyvale, CA). Standard curves were generated by plotting the log number of bacteria versus the corresponding threshold cycle value (mean of the results of three experiments). The lowest detection limit, or the presence of 1 bacterium (100 rRNA copies = 1 bacterium), was considered to be the dilution at which the threshold cycle value approaches 40 (zero).
Cell-free E. chaffeensis was purified as described above and was fluorescein treated as previously described (10) with modifications as follows. Fluorescein isothiocyanate (FITC) was dissolved into 0.2 M Na2CO3 (pH 9.5) buffer to a final concentration of 5 mg per ml. One milliliter of the FITC solution was added to the cell-free E. chaffeensis pellet, mixed gently, and incubated with rocking for 15 min at room temperature (protected from light). The FITC-bacterium mixture was then added to 4 ml of 135 mM NaCl-10 mM phosphate buffer (pH 7.4) and incubated for 5 min at room temperature. Following incubation, the mixture was centrifuged at 10,000 × g for 5 min, the supernatant was removed, the pellet was resuspended in 2.83% Na2HPO4 (pH 8.5), and the mixture was centrifuged again. The resulting pellet was resuspended in sterile PBS and subsequently washed three times with sterile PBS. The bacterial pellet was resuspended in sterile PBS for fly injections.
Bacteria were also labeled following the protocol of the pHrodo phagocytosis particle labeling kit for flow cytometry (Invitrogen, Carlsbad, CA; no. A10026). Briefly, cell-free E. chaffeensis was obtained as described above. The bacterial pellet was resuspended in 3 ml of 100 mM sodium bicarbonate buffer (component F) and centrifuged at 16,000 × g in a microcentrifuge for 60 s. The pellet was then suspended in 750 μl of component F, 18.25 μl of the pHrodo dye (component D) was added, and the bacteria were incubated with rocking for 45 min at room temperature (protected from light). Next, 750 μl of wash buffer (component C) was added to the bacterial mixture and centrifuged at 16,000 × g for 60 s, and the supernatant was removed. The bacterial pellet was then resuspended in 1.5 ml of 100% methanol, vortexed for 30 s, and centrifuged at 16,000 × g for 60 s, and the supernatant was removed. The bacteria were resuspended in 1.5 ml of component C, vortexed for 30 s, and centrifuged at 16,000 × g for 60 s. The wash step was repeated, and the final bacterial pellet was resuspended in sterile PBS for injection.
For fly injection, equal volumes of the FITC-labeled and pHrodo-labeled E. chaffeensis were mixed and simultaneously injected into yellow white (yw) flies. Flies were viewed on an Olympus SZX12 fluorescence dissecting microscope with emission filters of 535/50 for FITC detection and 620/60 for pHrodo detection.
S. enterica serovar Typhimurium (strain KSU-7) (15) was grown and streaked for isolation on MacConkey agar plates. Nutrient broth was inoculated with a single colony and grown overnight at 37°C. The cultures were centrifuged at 10,000 rpm for 10 min and rinsed with PBS. Bacterial pellets were resuspended in PBS, absorbance was measured on a NanoDrop spectrophotometer (Thermo Scientific, Wilmington, DE), and the bacterial concentration was estimated as previously described (9).
Flies were maintained on standard dextrose-molasses-yeast medium at 18 to 29°C. For initial E. chaffeensis infections and dose-response infections, w;Hemese-Gal4, UASGFP flies (from Michael J. Williams, Umea Centre for Molecular Pathogenesis, Umea University, Umea, Sweden) (66) were used as the wild type (WT) in these experiments. yw flies are continuously maintained at Kansas State University for use as a WT stock. Transgenic flies expressing green fluorescent protein (GPF) under the control of the Attacin [Attacin-GFP(II)] (attacin-GFP), Diptericin [Pw+/Pw+ III (Dipt-GFP-Drom)] (diptericin-GFP), or Drosomycin [ywP(w+,Drom-GFP)D4] (drosomycin-GFP) promoter and dredd (ywD44) (dreddD44) and relish (relE20, e+) mutants (from Bruno Lemaitre, Ecole Polytechnique Federale de Lausanne, Lausanne, France) (35, 62) were used to study the Imd and Toll pathways. Toll-constitutive flies (Tl3,ri+e−/TM3,Ser) (Tl3) were received from Claudia Mizutani and Ethan Bier (University of California at San Diego) (1). pelle mutant (pll2) stocks were obtained from the Bloomington Drosophila Stock Center at Indiana University, Bloomington, IN (e1 pll2 ca1/TM3, Sb1) (no. 3111) (2). cactus-deficient flies (cactD11/CyO; cactD13/CyO) (cactD11 and cactD13) have been previously described (51).
Flies were transferred to fresh food at least 24 h prior to injection/infection. For injection/infection, adult male and female flies were anesthetized with CO2 (for no longer than 15 min at a time). Flies were injected with approximately 50 nl of sterile PBS with or without bacteria, using pulled glass capillary needles. Injections were made in the abdomen of the fly, close to the junction of the thorax, and ventral to the junction between the dorsal and ventral cuticles. Following injection, flies were maintained in clean bottles with molasses caps that were changed every other day throughout the course of the experiments. Survival was monitored daily.
In order to inhibit phagocytosis by hemocytes (13, 16, 47, 52), flies were injected with a 2% solution of PBS mixed with FluoSpheres carboxylate-modified microspheres (0.2 μm, red fluorescent [580/605]) (Invitrogen no. F8810) or PBS alone. Injections were performed 4 hours or 24 h prior to injection with E. chaffeensis. Survival was monitored for 96 h postinfection (hpi).
Live, adult flies were assayed for enhanced GFP expression by placing each fly into a 0.5-ml microcentrifuge tube containing 120 μl of sterile PBS. The flies were pulverized using a Dremel 100 series rotary tool with a conical grinding bit. The bit was sterilized by flaming the bit after dipping it in 70% ethanol. The bit was cooled between each sample. After pulverization, the tubes were briefly microcentrifuged at 12,000 × g, and 50 μl of each supernatant was transferred to one well of a 96-well plate. The fluorescence was quantitated using a Perkin-Elmer Wallac Victor3 Multilabel Counter.
To quantitate transcript levels, flies were anesthetized, placed in 1.5-ml tubes (Kimble Kontes no. 749510-1500), and homogenized with disposable pestles in 1 ml of TriReagent (Molecular Research Center) (Kimble Kontes no. 749521-1590). Homogenates were transferred to 2.0-ml Heavy Phase Lock Gel tubes (5 Prime/Eppendorf, Westbury, NY; no. 2302830). Three hundred microliters of chloroform was added, and the mixture was shaken (not vortexed) for 15 s. The samples were then centrifuged at 12,000 × g for 10 min at 4°C, and the aqueous phase was transferred to clean 1.5-ml tubes. Five hundred microliters of isopropanol was added, and RNA was precipitated at −20°C for 24 h. Samples were subsequently centrifuged at 12,000 × g for 10 min. The RNA pellet was washed with 1 ml of 70% ethanol, and samples were centrifuged at 7.4 × g for 5 min. The 70% ethanol was decanted from the pellet, and the pellet was allowed to slightly air dry and was resuspended in 50 μl of nuclease-free water.
qRT-PCR was performed using the Invitrogen's one-step Platinum qRT-PCR kit (no. 11732) or Invitrogen's Superscript III Platinum SYBR green one-step qRT-PCR kit (no. 11732) in a Cepheid Smart Cycler. E. chaffeensis was detected as described above. Drosophila ribosomal protein 15a (accession no. NM_136772) was detected using left primer TGGACCACGAGGAGGCTAGG, right primer GTTGGTGCATGGTCGGTGA, and TaqMan probe TGGGAGGCAAAATTCTCGGCTTC (13). Antimicrobial peptides were detected using the following primer sets: diptericin (accession no. NM_057460), 5′-ACCGCAGTACCCACTCAATC and 5′-CCCAAGTGCTGTCCATATCC; and drosomycin (accession no. NM_079177), 5′-GTACTTGTTCGCCCTCTTCG and 5′-CTTGCACACACGACGACAG (52).
Plasmids carrying GFP reporter-AMP promoter constructs for Attacin, Diptercin, and Drosomycin were obtained from the lab of Bruno Lemaitre (62). Competent Escherichia coli cells (Invitrogen no. C404003) were transformed and plated on nutrient agar containing 50 μg per ml of ampicillin. The presence of the attacin, diptericin, or drosomycin plasmid was confirmed using colony PCR. For detection of diptericin and drosomycin, the previously mentioned primer sets were used. For the detection of attacin (accession no. NM_079021), left primer CAATGGCAGACACAATCTGG and right primer ATTCCTGGGAAGTTGCTGTG (13) were used. PCRs were performed using the Platinum Taq polymerase kit (Invitrogen no. 10966-034). Upon confirmation of the plasmid of interest, nutrient broth cultures containing 50 μg per ml of carbenicillin were inoculated and grown overnight at 37°C. Plasmid DNA was extracted using the Qiagen Midi-Prep kit (no. 12145). For transfection, 5 × 105 S2 cells were plated in 60-mm2 tissue culture wells of six-well plates and incubated overnight. One microgram of plasmid DNA was mixed with 15 μl of Cellfectin reagent (Invitrogen no. 10362) and subsequently mixed in incomplete S2 medium (no serum) in a total volume of 200 μl. The mixture was incubated at room temperature for 30 min. The S2 cells were rinsed with incomplete S2 medium, and a final volume of 800 μl of incomplete medium was added to the cells. Plasmid/Cellfectin complexes were added to the cells, and the cells were briefly shaken and incubated with the complexes for 5.5 to 6 h. Cells were then washed with complete S2 medium and resuspended in a final volume of 2 ml. To test for activation of the transgene, viable or heat-killed cell-free E. chaffeensis was added to the cells at 48 h posttransfection, and fluorescence was measured at 24 h later using a Perkin-Elmer Wallac Victor3 Multilabel Counter.
Data are presented as the means ± standard errors of the means (SEM) of independent experiments. Differences in means were determined by using Student's t test (two tailed, general) (StatMost statistical package; Data XIOM, Los Angeles, CA). Survival data were analyzed for significance using the log rank test of Kaplan-Meier plots using Prism Graphpad software (La Jolla, CA). P values of <0.05 were considered highly significant.
We previously demonstrated that E. chaffeensis replicated in the Drosophila hemocyte-like S2 cells (38). These data raised the issue of whether E. chaffeensis could grow in vivo. To test the hypothesis that adult D. melanogaster can be infected with E. chaffeensis, we injected cell-free E. chaffeensis into the abdomens of WT adult male and female D. melanogaster organisms. Negative control male and female flies were injected with sterile PBS. Survival of a total of 20 to 25 flies per experiment was monitored for 120 h. There was a significant amount of death in the infected flies compared to those injected with PBS (Fig. (Fig.1).1). For example, at 24 hpi, a mean of 86% of flies injected with Ehrlichia survived, compared to 97% of the PBS-injected controls. By 120 hpi, only 25% of the flies injected with Ehrlichia were still alive, compared to 85% of the PBS-injected controls (Fig. (Fig.1)1) (P < 0.05). To confirm that our results were not particular to the w;Hemese-Gal4, UASGFP flies we chose to use as WT, we compared their survival with that of yw flies. We detected no significant differences in survival between the yw and WT flies that were challenged with Ehrlichia (Fig. (Fig.1).1). Therefore, Ehrlichia pathogenesis was not unique to the w;Hemese-Gal4, UASGFP flies.
The number of flies that survived was dependent on the bacterial dose (Fig. (Fig.2).2). At 96 hpi, a mean of 47% of flies survived at the highest challenge dose, with 68%, 72%, and 77% survival when flies were challenged with decreasing doses (Fig. (Fig.2).2). Bacteria replicated in the flies over time. RNA was collected from live flies at several time points for subsequent qRT-PCR to measure E. chaffeensis 16S rRNA and Drosophila ribosomal protein 15a. By 24 h, there were usually more bacteria detected than were originally injected, and the number of bacteria increased and decreased in a cyclical fashion over a period of 120 h (Fig. (Fig.3).3). By 96 hpi there were higher bacterial loads in the flies than in the original injections for 9 of the 11 measurements made (Fig. (Fig.4).4). Therefore, these data indicate that E. chaffeensis is capable of infecting D. melanogaster in a dose-dependent fashion and that the bacteria were actively replicating in adult flies. Furthermore, bacteria isolated from adult flies were used to reinfect DH82 cells in vitro, as determined by RT-PCR and by the identification of morulae in the infected cell after infection (data not shown). Therefore, the bacteria that replicate in adult flies are capable of reinfecting mammalian cells.
E. chaffeensis infects monocytes and macrophages of vertebrate hosts (45). E. chaffeensis can also infect and replicate in phagocytic Drosophila S2 cells (38). Consequently, we hypothesized that interfering with hemocyte/phagocyte function in the adult flies would have an impact on E. chaffeensis infections. Injection of polystyrene beads has been shown to disrupt Drosophila hemocyte/phagocyte function in other studies (13, 47, 52). Therefore, we used that technique to explore the role of hemocytes in host resistance to an E. chaffeensis challenge. We injected WT adult male and female flies with a 2% solution of 0.2-μm polystyrene beads in PBS or with PBS alone. Flies were challenged with Ehrlichia either 4 or 24 h after injection of the beads-PBS or PBS alone. Although bead injection did not affect fly survival more than the PBS injection (Fig. (Fig.5),5), the bead injection had an immediate impact on flies challenged with Ehrlichia (P < 0.05, log rank test). When flies were challenged 4 hours after bead injection, 60% of the flies died, compared to 30% of the flies challenged with Ehrlichia alone, at 24 hpi. The impact of the beads diminished over time. By 96 hpi, there were no differences in survival (P > 0.05, t test) between bead-injected and non-bead-injected flies that were challenged with Ehrlichia. Flies challenged with Ehrlichia 24 h after bead injection had a pattern similar to that of the flies that were challenged 4 h after bead injection (Fig. 5A and B). In addition, we performed qRT-PCR to determine if bead injection had an impact on the number of bacteria present in surviving flies. We observed an increase in the number of bacteria present per fly in those flies that were injected with the beads prior to Ehrlichia challenge (Fig. 5C and D). For example, flies injected with beads 4 hours prior to challenge had an average of 40 bacteria per surviving fly, compared to 23 bacteria per control fly at 24 h after Ehrlichia challenge. The increased bacterial load was observed when the beads were injected 4 or 24 hours before Ehrlichia challenge (Fig. (Fig.55).
To determine if the Drosophila hemocytes were phagocytosing the injected E. chaffeensis, we utilized pHrodo dye, an amine-reactive succinimidyl ester, which becomes fluorescent as the local environment becomes more acidic. Drosophila hemocyte phagosomes have a lower pH than the extracellular environment (26). Therefore, we hypothesized that pHrodo-labeled E. chaffeensis would fluoresce only if it was phagocytosed by hemocytes. Additionally, we FITC labeled E. chaffeensis to visualize bacterial trafficking and localization. FITC-labeled bacteria were identified in 98% of the flies (three experiments, 20 flies/experiment) examined 4 h after injection and in 93% of the flies examined at 24 h after injection. We observed pHrodo-labeled bacteria in 82% of the flies at 4 h and 88% of the flies at 24 h after injection. Therefore, it appeared that the lower-pH environment of the hemocytes caused the bacteria to fluoresce red. Hemocytes and extracellular bacteria were distinctly observed in the central, dorsal abdomen at 4 and 24 h postinjection (Fig. (Fig.6),6), and this was the location where hemocytes and bacteria were most frequently observed. We also observed dissemination of the hemocytes and bacteria throughout the fly (Fig. (Fig.66).
Our data suggested that hemocytes were mobilized after injection of E. chaffeensis. Drosophila innate immunity also includes humoral components, principally the production of antimicrobial peptides through either the Toll or Imd pathway. The Toll pathway is most often activated by gram-positive bacteria and fungi, which elicit production of the antimicrobial peptide Drosomycin (18, 31, 63). The Imd pathway is most often activated by gram-negative bacteria and is characterized by production of antimicrobial peptides such as Attacin and Diptericin (18, 31, 63). Although E. chaffeensis is gram negative, it lacks the genes for synthesis of both LPS and peptidoglycan (37). Because E. chaffeensis has this atypical outer membrane, we were particularly interested in determining if the host defense pathways (Imd and/or Toll) were activated in response to E. chaffeensis challenge.
To address this question, we challenged transgenic flies expressing GPF under the control of the attacin, diptericin, or drosomycin promoter (62) with E. chaffeensis. After injection with bacteria or sterile PBS, GFP-expressing flies were examined under a fluorescence dissecting microscope at 24 hpi as described in Materials and Methods. There were distinct differences in attacin-GFP and diptericin-GFP expression between the flies that received bacteria and those that received PBS or no injection at all (Fig. (Fig.7).7). Fluorescence was also quantitated in the transgenic flies that were injected with Salmonella enterica serovar Typhimurium, as described in Materials and Methods. Salmonella is a gram-negative bacterium that is lethal to flies (47). No significant differences in antimicrobial peptide induction were detected between flies that were challenged with Ehrlichia and Salmonella (Fig. (Fig.8).8). For the Ehrlichia-injected flies, the attacin-GFP flies had a significant increase (P < 0.05, n = 3 independent experiments) compared to control flies (injected with PBS only) at both 6 and 24 hpi (Fig. (Fig.8).8). For example, at 6 hpi, attacin-GFP expression in the Ehrlichia-injected flies had a mean of 95,009 arbitrary fluorescent units (AFU), compared to 54,894 AFU in the PBS-injected flies. Significant increases in GFP expression were also observed in the diptericin-GFP flies that were injected with Ehrlichia compared to controls (Fig. (Fig.7).7). At 6 hpi, the mean diptericin expression was 66,635 AFU in the Ehrlichia-injected flies and 48,309 AFU in the PBS-injected flies. The activation of drosomycin was distinctly different from the activation of diptericin and attacin. There were minimal visual and quantitative (Fig. (Fig.77 and and8)8) differences in GFP expression in the drosomycin-GFP flies that received Ehrlichia and those that received PBS or no injection. GFP expression was also significantly lower in the drosomycin-GFP flies than in the attacin-GFP and diptericin-GFP flies.
The apparent upregulation of attacin and diptericin in adult flies challenged with E. chaffeensis suggested that the Imd pathway was activated in response to the E. chaffeensis and might be important to host resistance. To test this hypothesis, we experimentally challenged flies that carried mutations in the genes that encode specific proteins of the Imd pathway. These mutations included those that affected the Dredd or Relish protein. WT flies were concurrently infected as controls in these experiments. Both the relish and dredd mutants had significantly increased mortality compared to the control flies (Fig. (Fig.9).9). Statistically significant differences in survival between relish mutants and WT flies were observed at 24, 48, 72, 96, and 120 hpi (P < 0.05, t test and log rank test). For the dredd mutants, significant differences in survival compared to the WT flies were confirmed at 24 and 96 hpi (t test, P < 0.05), and a significant difference in the survival curves was confirmed by the log rank test. We confirmed that the dredd and relish mutants maintained their phenotypes. When we used qRT-PCR to measure the AMP transcript levels in the mutants, there were decreased transcript levels for diptericin in the dredd and relish mutants (compared to WT flies); drosomycin transcript levels were not altered in either of these mutants.
We also used qRT-PCR to measure the bacteremia in the mutant flies after infection to give us a second measure of pathogenesis. Infection was detected in all mutant flies and was higher in the relish mutants than in the WT flies. These results suggested that Relish was a key component of the Imd pathway necessary for Drosophila to combat E. chaffeensis infections.
The data from the drosomycin-GFP transgenic flies suggested that the Toll pathway was activated to a lesser extent after Ehrlichia infection than attacin and diptercin. To directly assess the role of the Toll defense pathway in host resistance to E. chaffeensis, we also challenged adult pll2, Tl3, cactD11, and cactD13 flies. Tl3 mutant flies have constitutively active Toll and enhanced Drosomycin activity. pll2 flies are mutants lacking the Pelle kinase and have an impaired ability to make Drosomycin. cactD11 and cactD13 flies are mutants lacking the Cactus protein, which is a negative inhibitor of the Toll pathway. Therefore, these flies have enhanced Drosomycin activity. We observed decreased mortality (P < 0.05, log rank test) in the Tl3 flies compared to WT flies (Fig. (Fig.10).10). We observed no significant difference in death at any time point in the cactus mutants (Fig. (Fig.11).11). In contrast to the Toll and cactus mutants, the pelle mutants displayed significantly decreased survival (P < 0.05) compared to WT flies at all time points (Fig. (Fig.12).12). At 24 hpi, a mean of 75% of the pelle mutants were alive, compared to 92% of WT flies. By 120 hpi, a mean of 25% of pelle mutants survived, compared to 53% of WT flies.
In mammalian systems, E. chaffeensis and other rickettsiae activate macrophages through the Toll-like receptor 4 pathway (11, 22, 27), despite the fact that the bacteria do not synthesize LPS. Therefore, the atypical outer membrane of Ehrlichia likely contains molecular patterns that can serve as alternative ligands for Toll-like receptor 4. Since disruption of components of both the Toll and the Imd pathways led to decreased fly survival after an Ehrlichia challenge, we wanted to determine if activation was dependent on bacterial replication/infection or if exposure to the bacteria and their respective molecular patterns was sufficient for activation of these pathways. To test this hypothesis, we used the AMP-GFP reporters used in the fly experiments described above as an in vitro readout system, using Drosophila S2 cells as outlined in Materials and Methods. Results from a microarray analysis (data not shown) revealed that several antimicrobial peptide genes were upregulated in S2 cells after infection with E. chaffeensis. attacin, diptericin, and drosomycin were among those genes. Therefore, we anticipated that the upregulation of GFP in S2 cells would occur in response to the appropriate signals delivered either by exposure to bacteria alone or by infection by live bacteria. attacin-GFP and diptericin-GFP expression were both upregulated in the transfected S2 cells in response to E. chaffeensis infection by viable bacteria (Fig. (Fig.13).13). drosomycin was also upregulated in response to infection (Fig. (Fig.13).13). Interestingly, we also observed induction of the antimicrobial peptides when heat-killed E. chaffeensis was added to the cells (Fig. (Fig.13).13). However, the intensity of the activation was lower than that with viable bacteria. These results suggest that some component of the Ehrlichia membrane activates the AMP pathways and that replication is able to augment that response.
This is the first demonstration that Ehrlichia chaffeensis is able to replicate in an arthropod other than the tick. We previously demonstrated that that Ehrlichia could grow in Drosophila S2 cells (38). Therefore, establishing infections at the organismic level represents an important step forward in our ability to work in an alternative arthropod system. We concede that the mode of Drosophila infection was not similar to the natural route of infection in a tick, which acquires the infection by a blood meal. However, the ability to grow the bacteria in the fly opens the biological and genetic tool box of Drosophila melanogaster to address questions about Ehrlichia that cannot be addressed in the tick.
E. chaffeensis infection was pathogenic to adult flies. It is not completely clear how this outcome differs when bacteria infect ticks. Little is known about tick survival after Rickettsia infection. Our results are consistent with the observation of a decrease in larval molting and overall survival among Ixodes scapularis ticks when allowed to feed on mice infected with various isolates of Anaplasma phagocytophilum (50). Increased death during molting to adult stage or before feeding of Dermacentor andersoni larvae and nymphs after experimental infection with Rickettsia rickettsii has also been reported (44). In the same study, fewer larvae developed from infected ticks. Decreases in molting success and increased death were also observed in a study in which Rhipicephalus sanguineus ticks were experimentally infected with Rickettsia conorii (36).
Although we observed a dose-dependent survival when flies were injected at any one particular time (Fig. (Fig.2),2), there was some experiment-to-experiment variability in the number of bacteria that caused death in the challenged flies. This variability probably resulted from a combination of factors, including differences in bacterial preparations (ratio of dense core to reticulate forms ), variability in the custom-made glass needles used to inject the bacteria, the ages and sexes of the flies (14, 19, 46), and the skill of the individual doing the injections. Therefore, the results of the experiments we presented were always reported as the outcome of multiple independent experiments. Control experiments and direct comparisons were done at the same time.
Hemocytes appear to participate in the Drosophila cellular immune response to E. chaffeensis. This is supported by three different experiments. There was significantly decreased survival when the flies received an injection of polystyrene beads prior to bacterial challenge. We also detected phagocytosis of Ehrlichia using the pHrodo dye and an increase in the number of bacteria in flies receiving the beads compared to the bead-free controls. It is possible that the beads inhibited hemocyte phagocytosis, rendering the fly less capable of controlling the initial infection of the bacteria. This hypothesis is supported by previous studies where polystyrene beads inhibited phagocytosis of Streptococcus pneumoniae, Plasmodium gallinaceum, Mycobacterium marinum, and Escherichia coli (13, 16, 47, 52). Alternatively, the beads may inhibit Ehrlichia-specific receptors on the hemocytes. This could lead to an overgrowth of bacteria in the hemocoel, resulting in tissue damage that may be fatal.
Our attempts to isolate hemocytes from infected or uninfected adult flies were unsuccessful, even after using several different techniques that were successful in other dipterans or Drosophila (8, 29). Therefore, we visualized hemocyte location within the fly with the pHrodo dye. Since not all the FITC-labeled bacteria were localized to hemocytes (identified with pHrodo-labeled bacteria), it appears that some of the bacteria (~25%) were not phagocytosed (extracellular bacteria). We observed an abundance of hemocytes as well as extracellular bacteria as early as 1 hour after and as late as 120 h after injection of bacteria. Hemocytes and extracellular bacteria were most often observed at the dorsal midline at the anterior end of the abdomen of the fly. There were also hemocytes and free bacteria distributed throughout the body of the fly. We suspect that ehrlichiae are capable of growing in hemocytes, since bacteria can grow in S2 cells (38). However, we have yet to determine whether the pathogenesis we see is due to bacteria growing inside the hemocyte, extracellularly, or by infecting other tissues. It has been reported that in Ixodid ticks, rickettsiae infect hemocytes and thus disseminate to all tissues and organs and can survive in the tissues and the body cavity for long periods of time (56). Additional experiments will be needed to determine if this dynamic is also the case in Drosophila, which was beyond the goals of this investigation. However, based on the survival kinetics after bead injections, it does appear that the early plasmatocyte/hemocyte response does work to control the infection.
Ehrlichia infection in flies appears to activate both the Toll and Imd host defense pathways. This conclusion is supported by several different experiments. For the Toll pathway, we saw increased survival in the Tl3 flies and decreased survival in the pelle mutant flies. The Tl3 flies were created with the mutagen ethyl methanesulfonate and have a mutation that renders Toll constitutively active (1, 32, 33, 49, 53). The pelle mutant flies were created with the mutagen ethyl methanesulfonate and have a mutation that ablates that function of the pelle kinase (54), which is believed to play a role in the degradation of Cactus (31, 61). The continuous expression of drosomycin in the Tl3 flies and the loss of drosomycin in the pelle mutant flies would support some role for this pathway after Ehrlichia challenge. Spaetzle is the ligand for Toll, and the interaction initiates Toll signaling (24, 33, 65), followed by interaction of MyD88/Pelle/Tube, which causes the degradation of Cactus (4, 59). Cactus is an inhibitor of the Toll pathway (43). Therefore, the loss-of-function Cactus mutants have a constitutively active Toll pathway (5, 51). Toll pathway involvement was also supported by activation of the drosomycin promoter when S2 cells were infected in vitro. It is not clear why activation of drosomycin was lower than that of attacin and diptericin in vivo. drosomycin expression has been shown to be upregulated later than diptericin expression (6, 34). Therefore, it is possible that our data reflect those kinetics.
The Imd pathway also appears to be activated after an Ehrlichia challenge. We observed activation of the promoters of the Imd-dependent antimicrobial peptides Attacin and Diptericin both in vivo and in vitro and increased transcript levels of diptericin in WT flies after infection (data not shown). We also observed significantly diminished survival of relish and dredd mutant flies after experimental challenge. Relish is a key component of the Imd pathway, which is activated by gram-negative bacteria (31, 48, 64). It is a homologous component to the p100 and p105 precursors in the NF-κB family of transcription factors and is essential for the transcription of antimicrobial peptides in D. melanogaster (57). The activation of the Imd pathway by the obligate intracellular pathogen E. chaffeensis is similar to observations made with another gram-negative, intracellular bacterium, Francisella tularensis (64). We also observed a significant decrease in the survival of dredd mutants. Dredd is the upstream caspase activator of Relish (17, 39), and it has been previously demonstrated that dredd mutants are more susceptible to gram-negative bacterial infections (35). Therefore, several different observations demonstrate that Imd plays a role in the host defense of E. chaffeensis.
Our data show that bacteria grow in adult flies, and the oscillatory nature of the bacterial load over time suggests that there is an active host response against the bacteria, just as there is in mice (21). Interestingly, the activation of AMPs did not appear to require bacterial replication. Exposure to boiled (dead) E. chaffeensis was sufficient for activation of AMPs in the S2 cell reporter system. The activation of AMP production is consistent with the control of these innate responses by pattern recognition receptors after they engage their ligands (31). In flies, the Toll and Imd pathways are activated by different forms of peptidoglycan but not by LPS (31). In rodents, E chaffeensis and other rickettsiae appear to engage Toll-like receptors even in the absence of LPS (22, 27, 60). Therefore, the Ehrlichia outer membrane components responsible for activation of antimicrobial peptides may be lipoproteins (25). The isolation of several immunogenic lipoproteins from the outer membrane of E. chaffeensis would support this hypothesis (25). Moreover, the activation of both the Imd and Toll pathways would be consistent with the atypical outer membrane of E. chaffeensis compared to the unique activation of Imd by classical, pyogenic gram-negative, LPS-containing bacteria (12, 23, 30).
In conclusion, we have demonstrated that E. chaffeensis is able to grow and replicate in adult D. melanogaster. Infection induces innate cellular and humoral responses in the fly that contribute to host resistance. These findings are significant because the Drosophila system will allow us to dissect the role of host genes in bacterial replication and elucidate which bacterial components contribute to the generation of innate resistance in arthropods.
We thank Tere Ortega for her help in the lab. We thank Roman Ganta, Department of Diagnostic Medicine and Pathobiology, and Carol Chitko-McKown, USDA Meat Animal Research Center, Hastings NE, for serving on the graduate supervisory committee of Alison Luce-Fedrow and for their constructive criticism of this work.
This project has been supported in part by Kansas Agriculture Experiment Station Animal Health Project grant 481848; the Kansas Space Grant Consortium; NIH grants AI55052, AI052206, RR16475, and RR17686; American Heart Association grant 0950036G; NASA grant NNX08BA91G; and the Terry C. Johnson Center for Basic Cancer Research.
This is Kansas Agricultural Experiment Station Publication no. 09-349-J.
Editor: R. P. Morrison
Published ahead of print on 17 August 2009.