Advertisement
Research Article

Tsetse Immune System Maturation Requires the Presence of Obligate Symbionts in Larvae

  • Brian L. Weiss mail,

    brian.weiss@yale.edu

    Affiliation: Department of Epidemiology and Public Health, Division of Epidemiology of Microbial Diseases, Yale University School of Medicine, New Haven, Connecticut, United States of America

    X
  • Jingwen Wang,

    Affiliation: Department of Epidemiology and Public Health, Division of Epidemiology of Microbial Diseases, Yale University School of Medicine, New Haven, Connecticut, United States of America

    X
  • Serap Aksoy

    Affiliation: Department of Epidemiology and Public Health, Division of Epidemiology of Microbial Diseases, Yale University School of Medicine, New Haven, Connecticut, United States of America

    X
  • Published: May 31, 2011
  • DOI: 10.1371/journal.pbio.1000619

Abstract

Beneficial microbial symbionts serve important functions within their hosts, including dietary supplementation and maintenance of immune system homeostasis. Little is known about the mechanisms that enable these bacteria to induce specific host phenotypes during development and into adulthood. Here we used the tsetse fly, Glossina morsitans, and its obligate mutualist, Wigglesworthia glossinidia, to investigate the co-evolutionary adaptations that influence the development of host physiological processes. Wigglesworthia is maternally transmitted to tsetse's intrauterine larvae through milk gland secretions. We can produce flies that lack Wigglesworthia (GmmWgm) yet retain their other symbiotic microbes. Such offspring give rise to adults that exhibit a largely normal phenotype, with the exception being that they are reproductively sterile. Our results indicate that when reared under normal environmental conditions GmmWgm adults are also immuno-compromised and highly susceptible to hemocoelic E. coli infections while age-matched wild-type individuals are refractory. Adults that lack Wigglesworthia during larval development exhibit exceptionally compromised cellular and humoral immune responses following microbial challenge, including reduced expression of genes that encode antimicrobial peptides (cecropin and attacin), hemocyte-mediated processes (thioester-containing proteins 2 and 4 and prophenoloxidase), and signal-mediating molecules (inducible nitric oxide synthase). Furthermore, GmmWgm adults harbor a reduced population of sessile and circulating hemocytes, a phenomenon that likely results from a significant decrease in larval expression of serpent and lozenge, both of which are associated with the process of early hemocyte differentiation. Our results demonstrate that Wigglesworthia must be present during the development of immature progeny in order for the immune system to function properly in adult tsetse. This phenomenon provides evidence of yet another important physiological adaptation that further anchors the obligate symbiosis between tsetse and Wigglesworthia.

Author Summary

Beneficial bacterial symbionts, which are ubiquitous in nature, are often characterized by the extent to which they interact with the host. In the case of mutualistic symbioses, both partners benefit so that each one can inhabit diverse ecological niches where neither could survive on its own. Unfortunately, little is known about the functional mechanisms that underlie mutualistic relationships. Insects represent a group of advanced multi-cellular organisms that harbor well-documented symbiotic associations. One such insect, the tsetse fly, harbors a maternally transmitted bacterial mutualist called Wigglesworthia that provides its host with essential metabolites missing from its vertebrate blood-specific diet. In this study, we further examine the relationship between tsetse and Wigglesworthia by investigating the interaction between this bacterium and its host's immune system. We have found that when Wigglesworthia is absent from tsetse during the maturation of immature larval stages, subsequent adults are characterized by an underdeveloped cellular immune system and thus highly susceptible to infection with a normally non-pathogenic foreign microbe. These findings represent an additional adaptation that further anchors the steadfast relationship shared between tsetse and its obligate symbiont.

Introduction

Bacteria comprise the most abundant and diverse life form on earth. The ubiquity of bacteria means they have colonized virtually every ecological niche, including habitation within more evolutionarily sophisticated multi-cellular animals. Co-evolution over millions of years has provided an opportunity for beneficial symbiotic associations to develop between phylogenetically distant taxa. Such affiliations are often mutualistic, meaning both partners benefit so that each can successfully inhabit diverse environments that neither could survive in on its own [1],[2]. Deciphering the mutualistic relationships between prokaryotic bacteria and multi-cellular eukaryotic animals is a rapidly advancing field of research.

Performing detailed investigations into the relationships between symbiotic bacteria and higher eukaryotes often involves costly and complex procedures. However, insects have well-documented symbioses that are attractive to study because they have relatively short generation times and are easy and less costly to rear. One insect that harbors multiple symbionts is the tsetse fly, Glossina morsitans. These microbes include the commensal Sodalis, the parasite Wolbachia, and the obligate mutualist Wigglesworthia glossinidia [3]. Molecular phylogenetic analysis indicates that tsetse's symbiosis with Wigglesworthia is ancient, dating back 50–80 million years [4]. The concordant nature of tsetse's obligate association with Wigglesworthia has driven the co-evolution of biological adaptations that are beneficial to both partners. For example, the localization of Wigglesworthia cells within host bacteriocytes provides a protective and metabolically favorable niche for this bacterium [5]. In return tsetse derives benefit from Wigglesworthia in at least two distinct ways. First, tsetse feeds exclusively on vertebrate blood, which is deficient in vitamins essential for survival. In accordance, a large proportion of Wigglesworthia's streamlined (700 kB) genome encodes vitamin biosynthesis pathways that presumably supplement tsetse's restricted diet [5],[6]. Second, more recent studies indicate that Wigglesworthia may serve an immunologic role in tsetse. These flies are the sole vector of pathogenic African trypanosomes, the causative agent of sleeping sickness in humans [7]. In laboratory experiments infection with immunogenic trypanosomes results in a decrease in tsetse fecundity [8]. Furthermore, parasite infection prevalence is higher in flies that lack Wigglesworthia when compared to age-matched wild-type (WT) individuals [9]. Wigglesworthia is thought to influence tsetse's vectorial competence by modulating its host's humoral immune system [10]. Thus, by preventing energetically costly parasite infections, this obligate symbiont may indirectly benefit the reproductive fitness of tsetse.

Symbiotic bacteria are rapidly gaining recognition for their important contributions to host development and immunity. The most well-known example of this type of interaction involves the mammalian microbiome, which modulates gut development during early postnatal life and subsequently shapes our mucosal and systemic immune systems [11],[12]. Symbionts also serve similar functions in invertebrate hosts. For example, light organ morphogenesis in juvenile bobtail squid (Euprymna scolopes) initiates only after symbiotic Vibrio fischeri cells have stably colonized this tissue [13]. The pea aphid, Acyrthosiphon pisum, harbors a secondary symbiont, Hamiltonella defensa, which can be infected with a lysogenic bacteriophage (A. pisum secondary endosymbiont; APSE). A. pisum that harbor both H. defensa and APSE are protected from being consumed by a parasitic wasp larvae through the action of phage-encoded toxins [14].

To the best of our knowledge no evidence exists that demonstrates insect symbionts can confer an impending protective phenotype in their adult host by directing immune system development during immature stages. In the present study we investigate the mechanism by which Wigglesworthia contributes to the development and function of cellular and humoral immune responses in adult tsetse. Our results show that the maturation and normal function of adult cellular immune responses in tsetse are severely compromised when Wigglesworthia is absent during larval development. This study reveals an important new facet that further anchors the obligate relationship between tsetse and Wigglesworthia and may serve as a useful model to understand the highly integrated and dynamic relationship between hosts and their beneficial bacterial fauna.

Results

Tsetse's Susceptibility to E. coli Infection Is Dependent Upon Age and Symbiotic Status

Insects are normally capable of mounting an immune response that combats infection with various groups of bacteria. Interestingly, in comparison to Drosophila, tsetse flies are uniquely susceptible to septic infection with 103 colony-forming units (CFU) of normally non-pathogenic Escherichia coli (E. coli) K12 [15]. In the present study we further investigated tsetse's unique susceptibility to E. coli infection by subjecting wild-type (GmmWT) and adults from two age groups to hemocoelic infections with varying quantities of E. coli K12. Three-day-old GmmWT individuals (flies from this age group are hereafter referred to as “young”) were highly susceptible to this treatment, as 103 CFU resulted in the death of all flies by 8 d post-infection (dpi; Figure 1A, top graph). In contrast, 77% and 55% of 8-d-old WT individuals (flies 8 d old and older are hereafter referred to as “mature”) survived for 14 dpi with 103 and 106 CFU of E. coli K12, respectively (Figure 1A, middle graph).

thumbnail

Figure 1. Host survival correlates with symbiont status following septic infection with E. coli K12.

(A) The effects of age and symbiont status on the survival of tsetse following systemic infection with E. coli K12. Mature adult GmmWgm- flies were significantly more susceptible to infection with 106 CFU of E. coli than were their wild-type counterparts (bottom and middle panels; p<0.001). (B) GmmWT/Wgm flies harbored Wigglesworthia during immature development but lacked the bacteria as mature adults. (C) GmmWT/Wgm adults were infected with tetracycline resistant E. coli 1 d after their last antibiotic-supplemented blood meal. Unlike their counterparts that lacked Wigglesworthia throughout immature development, GmmWT/Wgm was able to survive infection with E. coli. No significant difference in survival was observed between mature adult GmmWT versus GmmWT/Wgm adults infected with 106 CFU of E. coli (Figure 1A middle panel and Figure 1C; p = 0.07). (D) Relative Sodalis and Wolbachia densities in 40-d-old GmmWT and GmmWgm adults (n = 5 of each) were normalized against host β-tubulin copy number. (E) Analysis of bacterial 16s rRNA clone libraries indicates that GmmWT larvae harbored Wigglesworthia, Sodalis, and Wolbachia, while their counterparts from ampicillin treated females harbored only Sodalis and Wolbachia. No other bacteria were identified from either fly line. (F) Average number (±SEM) of recE. colipIL per tsetse strain over time (n = 3 individuals per strain per time point) following septic infection with 103 CFU of bacteria. Values shown in red represent lethal infections.

doi:10.1371/journal.pbio.1000619.g001

We demonstrated previously that feeding pregnant female tsetse the antibiotic ampicillin results in the generation offspring that lack Wigglesworthia (GmmWgm) but still harbor Sodalis [9] and presumably Wolbachia. In the present study we determined the survival outcome of mature GmmWgm inoculated with 106 CFU of E. coli K12 (the dose required to kill ~50% of mature WT flies; Figure 1A, middle graph). In comparison to age-matched WT tsetse, GmmWgm flies were highly susceptible to this infection. In fact, at 14 dpi, only 1% of these individuals remained (Figure 1A, bottom graph). We reasoned that the dramatic susceptibility of mature GmmWgm flies to infection with E. coli could result from one of two scenarios. In the first scenario Wigglesworthia may directly contribute to the adult immune response to a foreign micro-organism. In the second scenario, the absence of Wigglesworthia during larvagensis may give rise to GmmWgm adults with compromised immune functions. To determine if the first scenario is correct we treated mature WT tsetse with the antibiotic tetracycline to eliminate all of their microbiota, including bacteriome-associated Wigglesworthia (Figure 1B), and subsequently challenged these adult flies (GmmWT/Wgm) with 106 CFU of tetracycline resistant E. coli K12. We found that, similar to their WT counterparts (GmmWT; Figure 1A, middle panel), about 70% of GmmWT/Wgm individuals survived this infection (Figure 1C). Conversely, GmmWgm adults were highly susceptible (Figure 1A, bottom graph). This result suggests that when Wigglesworthia is absent from tsetse during larval development subsequent adults are severely immuno-compromised.

We demonstrated that when larval tsetse lacks exposure to Wigglesworthia in utero their immune system is highly compromised during adulthood. This finding does not exclude the possibility that treatment of female flies with ampicillin to produce GmmWgm offspring induces perturbations in other constituents of tsetse's microbiota. Because tsetse's other two known endosymbionts, Sodalis and Wolbachia, are present at similar densities in GmmWT and GmmWgm adults (Figure 1D), we looked for the presence of uncharacterized endosymbionts or digestive tract-associated microbes that are potentially passed on to developing intra-uterine larvae where they subsequently impact immunity. We generated a clone library containing 16s rRNA gene sequences PCR amplified from 3rd instar GmmWgm and GmmWT larvae and then sequenced multiple clones from each tsetse line. Our results indicate that the proportion of Wigglesworthia, Sodalis, and Wolbachia in 3rd instar GmmWT larvae is 19%, 75%, and 6%, respectively. As expected no Wigglesworthia 16s rRNA sequences were present in GmmWgm larvae, and the proportion of Sodalis and Wolbachia sequences present was 71% and 29%, respectively (Figure 1E). No uncharacterized endosymbionts or gut-associated environmental microbes were present in any sample. Taken together, these results suggest that the presence of Wigglesworthia specifically is responsible for enabling immune system maturation in WT tsetse.

Our host survival curves indicate that mature GmmWT and GmmWT/Wgm can survive infection with E. coli while young GmmWT and mature GmmWgm perish. To determine a cause for the variation in survival we observed between the four groups following infection with E. coli, we monitored bacterial growth dynamics over the course of the experiment in each group. Bacterial number within mature GmmWT and GmmWT/Wgm peaked at 3.8×104 and 1.9×104 E. coli, respectively, over the 2-wk period, indicating that these groups appeared able to control their infections. In contrast, E. coli increased exponentially in young GmmWT and mature GmmWgm and reached a maximum density at 6 dpi of 4.2×106 and 8.8×106, respectively (Figure 1F). These results implicate bacterial sepsis as the cause of high mortality observed in young GmmWT and mature GmmWgm. All of the above-mentioned results taken together indicate that mature GmmWT are considerably more resistant to infection with a foreign microbe than are their younger counterparts. Furthermore, tsetse's obligate mutualist, Wigglesworthia, must be present during the development of immature stages so that mature adults are able to overcome infection with E. coli.

Expression of Immunity-Related Genes in Adult Tsetse Varies Depending on Symbiont Status

To understand the basis of the compromised immunity we observed in mature GmmWgm flies, we evaluated the expression profile of a set of immunity-related genes from age-matched mature adult GmmWT and GmmWgm flies that were either uninfected or 3 dpi with E. coli K12. We included the antimicrobial peptides (AMPs) attacin, cecropin, and defensin, which distinctly target gram-negative (attacin and cecropin) and gram-positive bacteria (defensin) [16]. Our analysis also included thioester-containing proteins (tep2 and tep4), prophenoloxidase (PPO), and inducible nitric oxide synthase (iNOS). In insects TEPs likely function as pathogen-specific opsonins that bind to bacteria or parasites and promote their phagocytosis/encapsulation [17], while PPO initiates a proteolytic cascade that results in melanin deposition [18]. iNOS catalyzes synthesis of the signaling molecule nitric oxide (NO), which plays a role in humoral and cellular immunity in Anopheline mosquitoes [19], reduvid bugs [20], and Drosophila [21],[22] by inducing AMP expression and recruiting hemocytes to the site of infection.

Our expression results indicate that symbiont status plays little or no role in the expression of immunity-related genes in uninfected adults. In fact, with the exception of the AMP defensin, no significant differences in immunity-related gene expression between mature uninfected GmmWT and GmmWgm adults (Figure 2A). However, we observed a considerably different profile of immunity-related gene expression when these different fly strains were infected with E. coli K12. Under these circumstances, all of the genes evaluated (with the exception of defensin) were expressed at significantly higher levels in GmmWT compared to GmmWgm individuals (Figure 2B). Particular striking was the fact that the induction of pathways associated with cellular immunity, such as pathogen recognition (tep2 and tep4) and melanization (PPO), were significantly compromised in mature GmmWgm adults. The absence of a robust cellular immune response is likely the cause of high mortality among these individuals following E. coli infection. This analysis indicates that Wigglesworthia must be present during the development of immature tsetse in order for immune-related genes to subsequently be expressed in mature E. coli–infected adults.

thumbnail

Figure 2. The effect of symbiont status on the expression of immunity-related genes in adult tsetse.

(A) Target gene expression in uninfected GmmWT and GmmWgm adults is indicated relative to the constitutively expressed tsetse β-tubulin gene. (B) Fold-change in the expression of immunity-related genes in GmmWT and GmmWgm tsetse 3 dpi with E. coli K12. All values for both tsetse strains are represented as a fraction of average normalized gene expression levels in bacteria-infected flies relative to expression levels in PBS-injected controls. In (A) and (B), quantitative measurements were performed on three biological samples in duplicate. Values are represented as means (±SEM). * p<0.05, ** p<0.005 (Student's t-test).

doi:10.1371/journal.pbio.1000619.g002

Hemocytes Play an Integral Role in Controlling E. coli Infection Outcomes in Tsetse

Our analysis of immunity-related gene expression in tsetse suggests that cellular immune pathway functions in adult tsetse are particularly compromised when Wigglesworthia is absent during immature development. The most prominent cellular immune mechanisms include melanization and phagocytosis. These processes, which ultimately result in the removal of foreign invaders in Drosophila [23] and A. aegypti [24], both arise from distinct crystal cell and plasmatocyte hemocyte lineages, respectively [25].

We investigated the role hemocytes might play in determining the susceptible phenotype we observed in tsetse following infection with E. coli. We infected mature GmmWT individuals with GFP-expressing E. coli and were able to observe that hemocytes had engulfed a large number of the introduced cells by 12 hpi (Figure 3A). We next inhibited phagocytosis by introducing blue fluorescent microspheres directly into tsetse's hemocoel and 12 h later infected the bead-treated individuals with GFP-expressing E. coli. Microscopic inspection of hemocytes harvested 12 hpi with E. coli revealed the presence of internalized microspheres and the absence of engulfed E. coli. This observation indicated that we were successful in blocking hemocyte phagocytosis (Figure 3B). We subsequently maintained our microsphere-injected tsetse for 2 wk with the intention of determining the impact of impaired phagocytosis on host survival outcome. Mature GmmWT flies exhibiting impaired phagocytosis were highly susceptible to infection with both 1×103 and 1×106 E. coli K12. In fact, by day 12 post-infection, all of these flies had perished regardless of the initial dose used for infection (Figure 3C). This observation contrasts starkly with infection outcome in mature GmmWT that exhibit normally functioning hemocytes (Figure 1A, middle graph). Our results suggest that defects in phagocytosis severely compromise the ability of these tsetse flies to overcome bacterial infection.

thumbnail

Figure 3. Hemocyte-mediated phagocytosis is a critical component of tsetse's immune response.

(A) 8-d-old GmmWT were subjected to septic infection with GFP-expressing E. coli K12. Twelve hours post-infection hemolymph was collected, fixed on glass slides using 2% paraformaldehyde, and microscopically examined for the presence of hemocyte-engulfed bacterial cells. Scale bar = 10 µm. (B) The process of hemocyte-mediated phagocytosis in tsetse was blocked by micro-injecting polystyrene beads into the hemocoel of 8-d-old WT individuals. In consecutive 12 h intervals following bead injection, flies were infected with GFP-expressing E. coli K12 and then hemolymph was collected and fixed as described above. Hemocytes appear to have engulfed the beads, thus prohibiting the subsequent uptake of bacterial cells. The inset in each panel shows a higher magnification image of one hemocyte, which is identified by a white triangle in the left-most panel. Scale bar = 20 μm. (C) Tsetse flies that harbor hemocytes incapable of engulfing E. coli are susceptible to septic infection with this bacterium while their wild-type counterparts are not. The susceptible phenotype is exhibited regardless of whether phagocytosis-inhibited tsetse were inoculated with 103 or 106 CFU of E. coli. Beads alone had no effect on tsetse mortality. No significant difference existed in survival outcome between mature GmmWT phagocytosis inhibited flies infected with 103 versus 106 CFU of E. coli (p = 0.47, log-rank analysis). Furthermore, no significant difference was present between mature GmmWgm flies with uninhibited hemocytes (Figure 1A, bottom panel) and mature GmmWT phagocytosis inhibited flies (p = 0.11) infected with 106 CFU of E. coli.

doi:10.1371/journal.pbio.1000619.g003

A notable result of our immunity-related gene expression analysis was a 37-fold decrease in PPO levels in GmmWgm flies. This enzyme is an essential component of the melanization pathway, and its expression ultimately results in host wound healing and the melanization, encapsulation, and subsequent removal of foreign microorganisms [26][28]. In conjunction with the remarkable variation in PPO expression observed between GmmWT and GmmWgm, we were also able to visually observe the absence of a melanization response to E. coli infection in flies lacking Wigglesworthia. In fact, 30 min post-injection with E. coli, hemolymph was still actively exuding from the inoculation wound of GmmWgm flies. Conversely, in WT individuals no hemolymph was detectable and melanin was deposited at the wound site (Figure 4). These results further suggest that hemocyte-mediated cellular immunity provides an imperative defense against the establishment of bacterial infections in tsetse's hemocoel. Furthermore, the absence of this response in GmmWgm individuals was likely responsible for the compromised host survival phenotype we observed following infection with E. coli.

thumbnail

Figure 4. The effect of symbiont status on melanization in tsetse.

Mature GmmWT and GmmWgm tsetse (n = 10 of each strain) were intra-thoracically inoculated with 1×103 E. coli K12. Thirty minutes post-inoculation the wound site on individuals from each strain was inspected microscopically for the presence of hemolymph clotting and melanin deposition. Thirty minutes post-inoculation, neither hemolymph clotting nor melanin were observed at the wound site of GmmWgm individuals (indicated by a red arrow). In contrast, within the same amount of time, hemolymph no longer exuded from the wound of WT flies and melanin was present surrounding the site (indicted by a white arrow).

doi:10.1371/journal.pbio.1000619.g004

Symbiotic Bacteria Impact the Maturation of Tsetse's Cellular Immune Response During Development

We observed that young GmmWT were markedly more susceptible to infection with E. coli K12 than were their mature counterparts. Furthermore, symbiont status also altered infection outcome, as mature GmmWgm perished following E. coli infection while age-matched WT individuals survived. These differential infection outcomes appeared to result from disparities in cellular immune system function between the different tsetse lines we examined. Based on these observations we hypothesized that the obligate mutualist Wigglesworthia plays a crucial role in regulating the development of cellular immunity in tsetse during immature stages. To test this hypothesis we quantified the number of circulating and sessile hemocytes present in young and mature adult GmmWT and mature adult GmmWgm. Our results indicate that a 1.4-fold increase in circulating hemocyte number occurs between day 3 and day 8 in WT tsetse, while no significant change in circulating hemocyte number was observed between young and mature GmmWgm (Figure 5A). Interestingly, mature GmmWT adults harbored 3.4× more circulating hemocytes than did mature GmmWgm adults (Figure 5A). We also looked at sessile hemocyte abundance as a further indicator of Wigglesworthia's impact on the development of cellular immunity in tsetse. In Drosophila, this hemocyte subtype concentrates in large quantities around the anterior end of the fly's dorsal vessel [29]. Thus, we indirectly quantified sessile hemocyte abundance immediately adjacent to the anterior-most chamber of tsetse's dorsal vessel by measuring the fluorescent emission of microspheres that were found engulfed in this region. Young GmmWT adults engulfed 1.2× more microspheres than their mature counterparts and 15.7× more than mature GmmWgm adults. Furthermore, mature WT adults engulfed 13.2× more microspheres than did age-matched GmmWgm adults (Figure 5B).

thumbnail

Figure 5. The effect of age and symbiont status on the development of tsetse's cellular immune response.

(A) Number of hemocytes per µl of hemolymph in young (3 d) and mature (8 d) GmmWT and GmmWgm- tsetse (n = 5 individuals from each tsetse strain at both time points). (B) Quantitative analysis of sessile hemocyte abundance in young and mature GmmWT and mature GmmWgm- tsetse (n = 4 individuals from each tsetse strain and age point). All tsetse strains tested were subjected to hemocoelic injection with blue fluorescent microspheres. Twelve hours post-injection, flies were dissected to reveal tsetse's heart. The left-most panel is a Brightfield image of the three chambers that make up the dorsal vessel (DV; scale bar = 350 µm). The anterior-most chamber is indicated within a white circle. The three remaining panels are close-ups of the anterior chamber (scale bar for all 3 panels = 80 µm), visualized by excitation with UV light (365/415 nm). Relative fluorescence per tsetse group was determined using ImageJ software. (C) The presence of Wigglesworthia affects the expression of genes involved in hemocyte differentiation in immature larval tsetse. Target gene expression in GmmWT and GmmWgm larval instars 1–3 is indicated relative to the constitutively expressed tsetse β-tubulin gene. Quantitative measurements were performed on three biological samples in duplicate. All values in this figure are represented as means (±SEM). * p<0.05, ** p<0.005 (Student' t-test).

doi:10.1371/journal.pbio.1000619.g005

Our results demonstrate that Wigglesworthia must be present during the development of immature tsetse in order for cellular immunity to develop and function properly in adults. Thus, we speculate that the absence of hemocytes in adult tsetse should reflect a lack of blood cell differentiation during the development of immature stages. In Drosophila the process of blood cell differentiation, or hematopoiesis, begins in the embryo and proceeds through all larval stages [30]. During Drosophila embryogenesis early hematopoiesis can be distinguished by the expression of the zinc finger transcription factor “Serpent.” Subsequently, another transcription factor, “Lozenge,” directs the differentiation of serpent-expressing precursor cells into a specific lineage of hemocytes called crystal cells. To address the relationship between the presence of Wigglesworthia and early hematopoiesis in tsetse, we used qPCR to evaluate the relative number of serpent and lozenge transcripts present in 1st, 2nd, and 3rd instar larvae dissected from pregnant GmmWT and GmmWgm females. Larval instars L1, L2, and L3 from WT females expressed 1.7, 2.1, and 1.9 times more serpent transcripts, and 4, 4.4, and 3.9 times more lozenge transcripts, respectively, than did their counterparts from females that lacked Wigglesworthia (Figure 5C). This observed attenuated expression of both serpent and lozenge may account for the depleted hemocyte population, and compromised cellular immune function, we observed in adult GmmWgm individuals.

Discussion

In the present study we demonstrate that Wigglesworthia is intimately involved in regulating the maturation and function of tsetse's cellular immune system during immature larval development. We present a model that links the presence of Wigglesworthia in larval progeny with host immune system maturation during development and the subsequent ability of adult tsetse to overcome infection with foreign microbes (Figure 6). Obligate symbioses between intracellular bacteria and multi-cellular eukaryotes represent millions of years of co-evolution during which time both partners have adapted to increase each other's overall fitness. The association between tsetse and Wigglesworthia is an example of this reciprocal relationship in that neither organism can survive in the absence of the other.

thumbnail

Figure 6. Wigglesworthia modulates the development and function of tsetse's immune system.

Through a currently unknown mechanism, the presence of Wigglesworthia in GmmWT larvae stimulates hemocyte differentiation in a specialized organ that is homologous to Drosophila's lymph gland. Upon metamorphosis, specialized hemocyte subtypes are released from the lymph gland and carried over in a functional state to the adult. In the absence of Wigglesworthia, GmmWgm larvae produce significantly less hemocytes than their WT counterparts. Several innate immunity pathways are activated upon inoculation of E. coli into the hemocoel of mature adult GmmWT. Tsetse's preliminary line of defense against E. coli infection likely involves melanization at the wound site. This process is initiated by localized crystal cells, which instigate the melanization cascade by secreting prophenoloxidase (PPO) into the hemolymph. Pathogens that circumvent the wound site then encounter phagocyte-mediated cellular and humoral immune responses. Soluble TEPs likely opsonize bacterial cells, thus tagging them for engulfment by phagocytes. Lysis of engulfed bacteria causes the release of bacterial peptidoglycan PGN that subsequently stimulates the production of AMPs by the fat body. AMPs, which are also generated by hemocytes, are then secreted into the hemolymph where they further abrogate microbial proliferation. Hemocytes also produce reactive oxygen intermediates, such as iNOS, that exhibit direct bacterial toxicity and further stimulate humoral immunity. In the case of GmmWgm adults, incapacitated hematopoiesis during larval stages results in severely compromised immunity that renders these flies highly susceptible as adults to bacterial infection.

doi:10.1371/journal.pbio.1000619.g006

We present several lines of evidence indicating that tsetse's resistance to E. coli positively correlates with fly age and symbiont infection status during juvenile stages. Specifically, our results signify that mature GmmWT adults are resistant to E. coli infection, while young GmmWT adults are susceptible. In contrast, both young and old adult flies that lack Wigglesworthia throughout all developmental stages are killed by E. coli infections. Like their WT counterparts, GmmWgm larvae acquire both Sodalis and Wolbachia while in utero [31]. Although the intrauterine larval environment is otherwise aseptic, adult GmmWgm can be exposed to a wide range of environmental microbes during adulthood. However, neither the presence of other symbiotic bacteria (Sodalis and Wolbachia) in larvae, nor environmental microbes acquired during adulthood, appear to be sufficient to induce immunity in GmmWgm adults. Additionally, we show that mature GmmWT adults treated with antibiotics to eliminate Wigglesworthia (GmmWT/Wgm) remain resistant to E. coli infections. The resistant phenotype of GmmWT/Wgm further signifies that obligate Wigglesworthia is not directly responsible for the ability of mature adult GmmWT to overcome septic bacterial infection. Instead, it appears that Wigglesworthia's presence during the maturation of larval progeny stimulates the development of host immunity in adults.

Our quantitative analysis of gene expression indicates that GmmWT and GmmWgm adults exhibit no significant difference in the expression of genes that serve as hallmarks of humoral and cellular immunity in the absence of microbial challenge. However, following infection with E. coli, all pathways were significantly compromised in GmmWgm versus GmmWT adults. The most notable discrepancy observed between the two tsetse lines involved the expression of prophenoloxidase. Following E. coli infection, the expression of this gene increased 37-fold in WT tsetse but remained virtually unchanged in individuals that lacked Wigglesworthia. In WT insects, PPO, which is an inactive zymogen, is proteolytically cleaved to produce phenoloxidase upon mechanical injury or the presence of foreign pathogens. Phenoloxidase then facilitates the process of melanin deposition [18]. A functional melanotic pathway would likely increase tsetse's resistance to E. coli infection in several ways. First, sequestration of E. coli in a melanotic capsule at the site of inoculation would likely help prevent their dissemination into adjacent host tissues. This phenomenon has been observed in both the hawk moth, Manduca sexta, and Drosophila following infection with pathogenic bacteria (Photorhabdus luminescens) and parasitic wasp eggs, respectively [32],[33]. In both cases the lack of melanization at the wound site severely compromised the host's ability to subsequently overcome the foreign invader. Second, the melanization cascade results in the production of reactive oxygen intermediates that are directly toxic to foreign pathogens. In the flesh fly Sarcophaga peregrina and M. sexta, melanization intermediates such as DOPA exhibit direct antimicrobial activity [34],[35]. Finally, melanin at the site of inoculation expedites wound healing that could prevent the spread of secondary infections [26].

The E. coli–susceptible phenotype of mature GmmWgm adults is likely reflective of their blood cell (hemocyte) deficit in comparison to E. coli–resistant WT flies. In Drosophila 90%–95% of the total hemocyte population is composed of sessile and circulating plasmatocytes, which are a distinct hemocyte lineage predominantly responsible for engulfing and digesting foreign pathogens [36]. The inability of mature GmmWgm tsetse to survive infection with E. coli may result from their significantly reduced population of phagocytic hemocytes available to engulf bacteria injected into the hemocoel. In fact, by injecting polystyrene beads as a means of blocking this physiological process, we demonstrate that phagocytosis is a critical component of tsetse's ability to manage septic infection with E. coli in WT flies. Similarly, mutant Drosophila that contain a depleted plasmatocyte population, or have been injected with beads to prohibit phagocytosis, exhibit a remarkable susceptibility to a variety of gram positive and negative bacteria [37],[38]. Interestingly, Drosophila that lack functional plasmatocytes also exhibit a reduced capacity to activate humoral immune responses, further inhibiting their ability to fight bacterial infection [29].

In Drosophila the crystal cell hemocyte lineage controls the humoral melanization cascade via the release of PPO stored in large cytoplasmic inclusion bodies [30],[39]. In addition to the dramatic reduction in PPO expression in GmmWgm tsetse, two further lines of evidence indicate that this fly strain harbors a significantly reduced population of hemocytes that function in a homologous manner to Drosophila crystal cells. First, lozenge expression in all three larval instars from GmmWgm females is significantly lower than in their GmmWT counterparts. In Drosophila larval crystal cells fail to form in the absence of lozenge expression, yet the differentiation of other hemocyte types proceeds normally [40]. Second, we observed that hemocyte-deficient GmmWgm flies were unable to produce a viable clot at the site of bacterial inoculation. In mutant Drosophila strains that lack crystal cells, PPO is absent from the hemolymph. Consequently, the melanization cascade fails to initiate and hard clots do not form at wound sites [41].

We discovered that young GmmWT adults harbor significantly less circulating hemocytes than do mature WT adults. This observation implies that circulating hemocyte number in adult WT tsetse increases as a function of age, although the specific mechanism underlying this process in tsetse is currently unknown. In adult Drosophila intact lymph glands are absent and no evidence exists for the de novo synthesis of hemocytes following metamorphosis [30],[40]. Interestingly, we did observe more subepidermal sessile hemocytes in young compared to mature GmmWT adults, although the number was not statistically significant between the two groups. We speculate that the increased abundance of circulating hemocytes we observed in mature compared to young GmmWT may reflect a shift in the proportion of sessile to circulating cells instead of de novo production of new hemocytes. A comparable process occurs in Drosophila larvae and adults, where the proportion of sessile to circulating hemocytes changes following immune stimulation [42],[43]. Furthermore, after larval Drosophila receive an epidermal wound, circulating hemocytes are rapidly recruited to the site of injury. These circulating cells are phagocytically active and likely function as a front-line surveillance system against tissue damage and microbial infection [44]. In the mosquito malaria vector Anopheles gambiae, exposure to Plasmodium parasites stimulated an increase in the number of granulocytes circulating in the hemolymph. These primed mosquitoes were subsequently more resistant to infection with pathogenic bacteria than their wild-type counterparts [45]. We propose that young GmmWT adults and mature GmmWig adults may be susceptible to E. coli infection in part because they harbor significantly less circulating hemocytes than do older WT individuals that are resistant. Our previous studies indicated that GmmWig adults are highly susceptible to trypanosome infection [9],[10] and that this phenotype may be modulated by tsetse's humoral immune system [46][48]. Studies are ongoing to determine if cellular immunity also modulates tsetse's trypanosome vectorial competence.

Neonatal humans (and presumably other mammals as well) acquire probiotic gut microbes through the act of breast feeding [49]. These bacteria subsequently stably colonize the naïve intestine where they promote immune system maturation and enhance defense against infection with pathogenic microbes [50]. Most lower eukaryotes, including insects, hatch from an egg deposited into the environment and thus rely mainly on environmental microbes to stimulate their innate immune systems during the development of immature stages. For example, Anopheline mosquitoes and Drosophila are exposed to environmental microbes throughout all stages of their development. Under these natural conditions subsequent adults exhibit potent innate immunity [51],[52]. However, when these insects are reared under germ-free conditions, they exhibit a severely compromised humoral immune response [51],[53]. Tsetse flies are unique among other insects because they lead a relatively aseptic existence. Not only do they feed exclusively on sterile vertebrate blood, but they also exhibit a unique viviparous reproductive strategy. During this type of reproduction, all embryonic and larval stages develop within the female's uterus. The protected in utero environment in which viviparous offspring develop limits their exposure to environmental microbes. However, throughout tsetse larvagenesis maternal milk gland secretions provide the developing offspring with nourishment as well as Wigglesworthia, Sodalis, and Wolbachia [31]. Interestingly, recently established Sodalis and Wolbachia appear unable to influence their host's physiology to the same extent as mutualist Wigglesworthia despite their presence throughout larval development. Future studies on this system will focus on determining the chemical and/or metabolic elements provided by Wigglesworthia that stimulate the maturation of its host's immune system during larval development.

The results of this study provide evidence of a novel functional role for obligate insect symbionts–host immune system activation during immature developmental stages to ensure robust function during adulthood. We demonstrate that the obligate Wigglesworthia provides tsetse offspring the stimuli necessary for immune system development, a process that exhibits functional parallels with the mammalian system following the transfer of beneficial microbes from mothers to their neonates during that act of breast feeding. This phenomenon represents an adaptation that further anchors the steadfast relationship shared between tsetse and its obligate mutualist. The essential nature of tsetse's dependence on Wigglesworthia provides a potentially exploitable niche to experimentally modulate host immunity, with the intention of diminishing this insect's capacity to vector deadly trypanosomes.

Materials and Methods

Tsetse and Septic Infections

Throughout the text, 3-d-old and 8-d-old tsetse are referred to as “young” and “mature,” respectively. Wild-type G. m. morsitans (GmmWT) were maintained in Yale's insectary at 24°C with 50%–55% relative humidity. These flies received defibrinated bovine blood every 48 h through an artificial membrane feeding system [54]. Two Wigglesworthia-free tsetse lines were generated for use in our experiments. The first, GmmWgm, was generated by supplementing the diet of pregnant females with 25 µg of ampicillin per ml of blood. GmmWgm (offspring of ampicillin-fed females) adult are devoid of Wigglesworthia throughout all developmental stages [9]. PCR using Wigglesworthia thiamine C-specific primers (forward, 5′-TGAAAACATTTGCAAAATTTG-3′; reverse, 5′-GGTGTTACATAGCATAACAT-3′) confirmed that this tsetse strain lacked Wigglesworthia. The second Wigglesworthia-free tsetse line, GmmWT/Wgm, was generated by feeding mature WT adults three blood meals supplemented with 40 µg/ml tetracycline. GmmWT/Wgm individuals thus harbored Wigglesworthia, Sodalis, and Wolbachia throughout the development of all immature stages and early adulthood, but lacked these symbionts thereafter. The absence of Wigglesworthia from mature GmmWT/Wgm adults was confirmed microscopically by comparing the bacteriome (Wigglesworthia-harboring organ) contents of mature WT and GmmWT/Wgm adults.

Septic infection of tsetse was achieved by anesthetizing flies with CO2 and subsequently injecting individuals with live bacterial cells using glass needles and a Narashige IM300 micro-injector. The methods used to produce luciferase-expressing E. coli K12 (recE. colipIL), and the assay used to quantify luciferase expression in vivo, were described previously [15]. GFP-expressing and tetracycline-resistant E. coli K12 were produced via electroporation with pGFP-UV (Clontech, Mountain View, CA) and pBR322 (Promega, Madison, WI) plasmid DNA, respectively. All flies treated with tetracycline were subsequently infected with tetracycline resistant E. coli K12. The number of bacterial cells injected and control group designations for all infection experiments are indicated in the corresponding figures and their legends. For all survival experiments, treatments were performed in triplicate, using 25 flies per E. coli treatment replicate. LB media controls were performed once using 25 flies.

Real-Time Quantitative PCR (qPCR)

For analysis of immunity-related gene expression, sample preparation and qPCR were performed as described previously [15]. Quantitative measurements were performed on three biological samples in duplicate and results were normalized relative to tsetse's constitutively expressed β-tubulin gene (determined from each corresponding sample). Fold-change data are represented as a fraction of average normalized gene expression levels in bacteria-infected flies relative to expression levels in corresponding uninfected controls.

For symbiont quantification, total RNA was prepared from 40-d-old adult GmmWT and GmmWgm flies. Symbiont genome numbers were quantified using single-copy Sodalis fliC and Wolbachia groEL. Relative symbiont densities were normalized to tsetse β-tubulin. All qPCR was performed with an icycler iQ real time PCR detection system (Bio-Rad). Values are represented as the mean (±SEM). qPCR primer sequences are shown in Table S1.

Bacterial 16s rRNA Clone Libraries

Universal bacterial 16S rRNA gene primers 27F (5′-AGAGTTTGATCCTGGCTCA G-3′) and 1492R (5′-GGTTACCTTGTTACGACTT-3′) [55] were used to produce a clone library of 16s rRNA gene sequences found in 3rd instar GmmWT and GmmWgm larvae. Five individual larvae from each tsetse line were dissecting under sterile conditions and washed in DNase (Ambion, Austin, TX) to remove any surface contamination prior to DNA extraction. Genomic DNA was isolated using Holmes-Bonner buffer (0.1 mol/L Tris-HCl, pH 7.5; 0.35 mol/L NaCl; 10 mmol/L EDTA, pH 8.0; 2% SDS; 7 mol/L Urea), purified via phenol-chloroform extraction and precipitated in 100% EtOH. PCR was performed under standard reaction conditions [15], and the resulting products were cloned in the pGEM-T vector (Promega, Madison, WI). Twenty clone inserts from each larvae (100 in total from each tsetse line) were sequenced using the T7 vector specific primer, and homology to previously described 16s rRNAs was determined using the blastn database.

Hemolymph Collection and Hemocyte Quantification

Depending on the subsequent experiment, tsetse hemolymph was collected using one of two methods. For hemocyte quantification, undiluted hemolymph was collected by removing one front fly leg at the joint nearest the thorax and then applying gentle pressure to the distal tip of the abdomen. Hemolymph exuding from the wound was collected using a glass micro-pipette and placed into a microfuge tube on ice. Hemocytes were quantified microscopically using a Bright-Line hemocytometer, and hemocyte numbers are represented as cells per µl of hemolymph.

When hemocytes were required for microscopic visualization, hemolymph was collected by employing a modified version of the high injection/recovery method previously developed for use in mosquitoes [56]. In brief, tsetse flies were sedated on ice and injected with 25 µl of chilled anticoagulant buffer [70% MM medium, 30% anticoagulant citrate buffer (98 mM NaOH, 186 mM NaCl, 1.7 mM EDTA, and 41 mM citric acid, buffer pH 4.5), vol/vol] between the last two abdominal schlerites using a glass needle and a Narashige IM300 micro-injector. Following a 30 min incubation on ice, a front leg was removed at the joint most proximal to the thorax. At this point internal pressure forced hemolymph diluted with anticoagulant buffer to be expelled from the wound site (more liquid could be recovered by applying gentle pressure to the distal end of the abdomen). Liquid was collected using a pipette and either placed into a chilled microfuge tube or directly into a 24-well cell culture plate. In the latter case, cells were allowed to adhere to the plate bottom, after which anticoagulant buffer was replaced with MM media.

Sessile hemocytes were observed by intra-thoracically injecting young and mature GmmWT, and mature GmmWgm (n = 5 of each strain), with 2 µl of blue fluorescent (365/415 nm) 0.2 µm carboxylate-modified beads (Invitrogen corp.). Prior to use, beads were washed once in PBS and resuspended in 100% of their original volume. Flies were dissected 12 h post-injection to reveal their dorsal vessel and surrounding tissue, which was gently washed 3 times with PBS to remove any potentially contaminating circulating (non-adherent) hemocytes. Engulfed microspheres were visualized using a Zeiss steriomicroscope (Discovery v8) equipped with a coaxial fluorescence module. Semi-quantitative comparison of sessile hemocyte number between young and mature GmmWT adults, and mature GmmWT and GmmWgm adults, was performed by quantifying fluorescent signal intensity (n = 4 individuals from each group) using ImageJ software (http://rsbweb.nih.gov/ij/).

Phagocytosis and Melanization Assays

Phagocytosis by circulating tsetse hemocytes was observed by intra-thoracically infecting mature GmmWT flies (n = 10) with 1×106 GFP-expressing E. coli K12. Twelve hours post-infection, hemolymph was collected and hemocytes monitored to determine if they had engulfed the GFP-expressing bacterial cells. Hemolymph samples were fixed on glass microscope slides via a 2 min incubation in 2% PFA. Prior to visualization using a Zeiss Axioscope microscope, slides were overlayed with VectaShield hard set mounting medium containing DAPI (Vector Laboratories, Burlingame, CA).

Phagocytosis by tsetse hemocytes was inhibited with blue fluorescent (365/415 nm) 0.2 µm carboxylate-modified beads (Invitrogen Corp.). Prior to use, beads were washed once in PBS and resuspended in 100% of their original volume. Inhibition assays were performed by inoculating 8-d-old GmmWT with 2 µl of beads via their thoracic compartment. Twelve hours later, these flies were similarly infected with 1×103 and 1×106 GFP-expressing E. coli K12 (experiment was performed in triplicate; n = 25 flies per replicate). Finally, 12 hours post-infection with E. coli, hemolymph was collected and processed as described above (these samples were overlayed with VectaShield hard set mounting medium that lacked DAPI).

Melanization assays were performed by intra-thoracically inoculating mature GmmWT and GmmWgm (n = 10 of each strain) with 1×103 E. coli K12. Subsequently, three individuals from each group were monitored microscopically every 10 min for the presence of melanin at the wound site. The remaining seven flies from each group were maintained for 2 wk in order to observe infection outcome.

Statistics

Statistical significance of survival curves was determined by log-rank analysis using JMP (v8.02) software (www.jmp.com). Statistical analysis of qPCR data was performed by Student's t test using JMP (v8.02) software. Statistical significance between various treatments, and treatments and controls, is indicated in corresponding figure legends.

Supporting Information

Table S1.

Primers used for quantitative real-time PCR.

doi:10.1371/journal.pbio.1000619.s001

(0.02 MB XLS)

Acknowledgments

We are grateful to Dr. Yineng Wu for assistance with qPCR, Dr. Corey Brelsford for assistance with statistical analysis, and Dan Summers for his help processing bacterial 16s rRNA clone libraries. We also thank members of the Aksoy lab for critical review of the manuscript.

Author Contributions

The author(s) have made the following declarations about their contributions: Conceived and designed the experiments: BLW SA. Performed the experiments: BLW JW. Analyzed the data: BLW SA. Contributed reagents/materials/analysis tools: BLW SA. Wrote the paper: BLW SA.

References

  1. 1. Moran N. A (2006) Symbiosis. Curr Biol 16: R866–R871.
  2. 2. Wernegreen J. J (2002) Genome evolution in bacterial endosymbionts of insects. Nat Rev Genet 3: 850–861.
  3. 3. Aksoy S (2000) Tsetse – a haven for microorganisms. Parasitol Today 16: 114–118.
  4. 4. Chen X, Li S, Aksoy S (1999) Concordant evolution of a symbiont with its host insect species: molecular phylogeny of genus Glossina and its bacteriome-associated endosymbiont, Wigglesworthia glossinidia. J Mol Evol 48: 49–58.
  5. 5. Akman L, Yamashita A, Watanabe H, Oshima K, Shiba T, et al. (2002) Genome sequence of the endocellular obligate symbiont of tsetse flies, Wigglseworthia glossinidia. Nat Genet 32: 402–407.
  6. 6. Nogge G (1976) Sterility in tsetse flies (Glossina morsitans Westwood) caused by loss of symbionts. Experientia 32: 995–996.
  7. 7. Aksoy S, Gibson W. C, Lehane M. J (2003) Interactions between tsetse and trypanosomes with implications for the control of trypanosomiasis. Adv Parasitol 53: 1–83.
  8. 8. Hu C, Rio R. V, Medlock J, Haines L. R, Nayduch D, et al. (2008) Infections with immunogenic trypanosomes reduce tsetse reproductive fitness: potential impact of different parasite strains on vector population structure. PLoS Neg Trop Dis 2: 1–10. doi:10.1371/journal.pntd.0000192.
  9. 9. Pais R, Lohs C, Wu Y, Wang J, Aksoy S (2008) The obligate mutualist Wigglesworthia glossinidia influences reproduction, digestion, and immunity processes of its host, the tsetse fly. Appl Environ Microbiol 74: 5965–5974.
  10. 10. Wang J, Wu Y, Yang G, Aksoy S (2009) Interactions between mutualist Wigglesworthia and tsetse peptidoglycan recognition protein (PGRP-LB) influence trypanosome transmission. Proc Natl Acad Sci U S A 106: 12133–12138.
  11. 11. Hooper L. V (2004) Bacterial contributions to mammalian gut development. Trends Microbiol 12: 129–134.
  12. 12. Macpherson A. J, Harris N. L (2004) Interactions between commensal intestinal bacteria and the immune system. Nat Rev Immunol 4: 478–485.
  13. 13. Koropatnick T. A, Engle J. T, Apicella M. A, Stabb E. V, Goldman W. E, et al. (2004) Microbial factor-mediated development in a host-bacterial mutualism. Science 306: 1186–1188.
  14. 14. Oliver K. M, Degnan P. H, Hunter M. S, Moran N. A (2009) Bacteriophages encode factors required for protection in a symbiotic mutualism. Science 325: 992–994.
  15. 15. Weiss B. L, Wu Y, Schwank J. J, Tolwinski N. S, Aksoy S (2008) An insect symbiosis is influenced by bacterium-specific polymorphisms in outer membrane protein A. Proc Natl Acad Sci U S A 105: 15088–15093.
  16. 16. Hoffmann J. A (2003) The immune response of Drosophila. Nature 426: 33–38.
  17. 17. Blandin S. A, Levashina E. A (2007) Phagocytosis in mosquito immune responses. Immunol Rev 219: 8–16.
  18. 18. Cerenius L, Söderhäll K (2004) The prophenoloxidase-activating system in invertebrates. Immunol Rev 198: 116–126.
  19. 19. Dimopoulos G, Richman A, Muller H. M, Kafatos F. C (1997) Molecular immune responses of the mosquito Anopheles gambiae to bacteria and malaria parasites. Proc Natl Acad Sci U S A 94: 11508–11513.
  20. 20. Whitten M, Sun F, Tew I, Schaub G, Soukou C, et al. (2007) Differential modulation of Rhodnius prolixus nitric oxide activities following challenge with Trypanosoma rangeli, T. cruzi and bacterial cell wall components. Insect Biochem Mol Biol 37: 440–452.
  21. 21. Foley E, O'Farrell P. H (2003) Nitric oxide contributes to induction of innate immune responses to Gram-negative bacteria in Drosophila. Genes Dev 17: 115–125.
  22. 22. Carton Y, Frey F, Nappi A. J (2009) Parasite-induced changes in nitric oxide levels in Drosophila paramelanica. J Parasitol 95: 1134–1141.
  23. 23. Matova N, Anderson K. V (2006) Rel/NF-κB double mutants reveal that cellular immunity is central to Drosophila host defense. Proc Natl Acad Sci U S A 103: 16424–16429.
  24. 24. Hillyer J. F, Schmidt S. L, Christensen B. M (2003) Rapid phagocytosis and melanization of bacteria and Plasmodium sporozoites by hemocytes of the mosquito Aedes aegypti. J. Parasitol 89: 62–69.
  25. 25. Meister M, Lagueux M (2003) Drosophila blood cells. Cell Microbiol 5: 573–580.
  26. 26. Galko M. J, Krasnow M. A (2004) Cellular and genetic analysis of wound healing in Drosophila larvae. PLoS Biol 2: 1114–1126. doi:10.1371/journal.pbio.0020239.
  27. 27. Christensen B. M, Li J, Chen C. C, Nappi A. J (2005) Melanization immune responses in mosquito vectors. Trends Parasitol 21: 192–199.
  28. 28. Ayres J. S, Schneider D. S (2008) A signaling protease required for melanization in Drosophila affects resistance and tolerance of infections. PLoS Biol 6: 2764–2773. doi:10.1371/journal.pbio.0060305.
  29. 29. Elrod-Erickson M, Mishra S, Schneider D. S (2000) Interactions between the cellular and humoral immune responses in Drosophila. Curr Biol 10: 781–784.
  30. 30. Evans C. J, Sinenko S. A, Mandal L, Martinez-Agosto J. A, Hartenstein V, et al. (2008) Genetic dissection of hematopoiesis using Drosophila as a model system. In: Bodmer , R , editors. Advances in developmental biology, Volume 18. Elsevier. pp. 259–298.
  31. 31. Attardo G. M, Lohs C, Heddi A, Alam U. H, Yildirim S, et al. (2008) Analysis of milk gland structure and function in Glossina morsitans: milk protein production, symbiont populations and fecundity. J Insect Physiol 54: 1236–1242.
  32. 32. Eleftherianos I, Boundy S, Joyce S. A, Aslam S, Marshall J. W, et al. (2007) An antibiotic produced by an insect-pathogenic bacterium suppresses host defenses through phenoloxidase inhibition. Proc Natl Acad Sci U S A 104: 2419–2424.
  33. 33. Rizki T. M, Rizki R. M (1990) Encapsulation of parasitoid eggs in phenoloxidase-deficient mutants of Drosophila melanogaster. J Insect Physiol 36: 523–529.
  34. 34. Leem J. Y, Nishimura C, Kurata S, Shimada I, Kobayashi A, et al. (1996) Purification and characterization of N-beta-alanyl-5-S-glutathionyl-3,4-dihyd​roxyphenylalanine,a novel antibacterial substance of Sarcophaga peregrina (flesh fly). J Biol Chem 271: 13573–13577.
  35. 35. Zhao P, Li J, Wang Y, Jiang H (2007) Broad-spectrum antimicrobial activity of the reactive compounds generated in vitro by Manduca sexta phenoloxidase. Insect Biochem Mol Biol 37: 952–959.
  36. 36. Lebestky T, Chang T, Hartenstein V, Banerjee U (2000) Specification of Drosophila hematopoietic lineage by conserved transcription factors. Science 288: 146–149.
  37. 37. Charroux B, Royet J (2009) Elimination of plasmatocytes by targeted apoptosis reveals their role in multiple aspects of the Drosophila immune response. Proc Natl Acad Sci U S A 106: 9797–9802.
  38. 38. Pham L. N, Dionne M. S, Shirasu-Hiza M, Schneider D. S (2007) A specific primed immune response in Drosophila is dependent on phagocytes. PLoS Pathog 3: 1–8. doi:10.1371/journal.ppat.0030026.
  39. 39. Bidla G, Dushay M. S, Theopold U (2007) Crystal cell rupture after injury in Drosophila requires the JNK pathway, small GTPases and the TNF homolog Eiger. J Cell Sci 120: 1209–1215.
  40. 40. Lanot R, Zachary D, Holder F, Meister M (2001) Postembryonic hematopoiesis in Drosophila. Dev Biol 230: 243–257.
  41. 41. Bidla G, Lindgren M, Theopold U, Dushay M. S (2005) Hemolymph coagulation and phenoloxidase in Drosophila larvae. Dev Comp Immunol 29: 669–679.
  42. 42. Zettervall C, Anderl I, Williams M. J, Palmer R, Kurucz E, et al. (2004) A direct screen for genes involved in Drosophila blood cell activation. Proc Natl Acad Sci U S A 101: 14192–14197.
  43. 43. Brandt S. M, Schneider D. S (2007) Bacterial infection of fly ovaries reduces egg production and induces local hemocyte activation. Dev Comp Immunol 31: 1121–1130.
  44. 44. Babcock D. T, Brock A. R, Fish G. S, Wang Y, Perrin L, et al. (2008) Circulating blood cells function as a surveillance system for damaged tissue in Drosophila larvae. Proc Natl Acad Sci U S A 105: 10017–10022.
  45. 45. Rodrigues J, Brayner F. A, Alves L. C, Dixit R, Barillas-Mury C (2010) Hemocyte differentiation mediates innate immune memory in Anopheles gambiae mosquitoes. Science 329: 1353–1355.
  46. 46. Hao Z, Kasumba I, Lehane M. J, Gibson W. C, Kwon J, et al. (2001) Tsetse immune responses and trypanosome transmission: implications for the development of tsetse-based strategies to reduce trypanosomiasis. Proc Natl Acad Sci U S A 98: 12648–12653.
  47. 47. Hu Y, Aksoy S (2005) An antimicrobial peptide with trypanocidal activity characterized from Glossina morsitans morsitans. Insect Biochem Mol Biol 35: 105–115.
  48. 48. Hu C, Aksoy S (2006) Innate immune responses regulate trypanosome parasite infection of the tsetse fly Glossina morsitans morsitans. Mol Microbiol 60: 1194–1204.
  49. 49. Martin R, Langa S, Reviriego C, Jimenez E, Marin M. L, et al. (2003) Human milk is a source of lactic acid bacteria for the infant gut. J Pediatr 143: 754–758.
  50. 50. Lara-Villoslada F, Olivares M, Sierra S, Rodriguez J. M, Boza J, et al. (2007) Beneficial effects of probiotic bacteria isolated from breast milk. Br J Nutr 98: S96–S100.
  51. 51. Dong Y, Manfredini F, Dimopolous G (2009) Implications of the mosquito midgut microbiota in the defense against malaria parasites. PLoS Pathog 5: 1–10. doi:10.1371/journal.ppat.1000423.
  52. 52. Charroux B, Royet J (2010) Drosophila immune response: from systemic antimicrobial peptide production in fat body cells to local defense in the intestinal tract. Fly 4: 40–47.
  53. 53. Ryu J, Kim S, Lee H, Bai J, Nam Y, et al. (2008) Innate immune homeostasis by the homeobox gene caudal and commensal-gut mutualism in Drosophila. Science 319: 777–782.
  54. 54. Moloo S. K (1971) An artificial feeding technique for Glossina. Parasitol 68: 507–512.
  55. 55. Dale C, Beeton M, Harbison C, Jones T, Pontes M (2006) Isolation, pure culture, and characterization of “Candidatus arsenophonus arthropodicus,” and intracellular secondary endosymbiont from the hippoboscid louse fly Pseudolynchia canariensis. App Environ Microbiol 72: 2997–3004.
  56. 56. Castillo J. C, Robertson A. E, Strand M. R (2006) Characterization of hemocytes from the mosquitoes Anopheles gambiae and Aedes aegypti. Insect Biochem Mol Biol 36: 891–903.