Phagocytosis is an ancient mechanism central to both tissue homeostasis and immune defense. Both the identity of the receptors that mediate bacterial phagocytosis and the nature of the interactions between phagocytosis and other defense mechanisms remain elusive. Here, we report that Croquemort (Crq), a Drosophila member of the CD36 family of scavenger receptors, is required for microbial phagocytosis and efficient bacterial clearance. Flies mutant for crq are susceptible to environmental microbes during development and succumb to a variety of microbial infections as adults. Crq acts parallel to the Toll and Imd pathways to eliminate bacteria via phagocytosis. crq mutant flies exhibit enhanced and prolonged immune and cytokine induction accompanied by premature gut dysplasia and decreased lifespan. The chronic state of immune activation in crq mutant flies is further regulated by negative regulators of the Imd pathway. Altogether, our data demonstrate that Crq plays a key role in maintaining immune and organismal homeostasis.
Phagocytosis is a first-line host defense mechanism against microbes. Interactions between phagocytosis and other immune mechanisms, such as the humoral response, however, remain elusive. Defective phagocytosis can lead to immune deficiencies and chronic auto-inflammation. Here, we show that Croquemort (Crq), a Drosophila member of the CD36 family of scavenger receptors, plays a role in microbial phagocytosis. Crq is required in phagocytes for efficient uptake of bacteria and fungi, and mutants for crq succumb to both environmental microbes and infections. Crq is also required for phagosome maturation, and crq mutants lack the ability to fully clear bacterial infection. As a result, crq mutant flies enter a state of chronic immune activation. Notably, they show increased production of the cytokine Upd3 that induces intestinal stem cell proliferation. Consequently, crq mutant flies show early signs of gut dysplasia, as well as a shortened lifespan. Altogether, our study demonstrates a new link between phagocytosis and tissue homeostasis, and illustrates how the chronic induction of cytokine production secondary to defective phagocytosis can alter gut homeostasis and shorten lifespan.
Citation: Guillou A, Troha K, Wang H, Franc NC, Buchon N (2016) The Drosophila CD36 Homologue croquemort Is Required to Maintain Immune and Gut Homeostasis during Development and Aging. PLoS Pathog 12(10): e1005961. https://doi.org/10.1371/journal.ppat.1005961
Editor: Mary O'Riordan, University of Michigan Medical School, UNITED STATES
Received: May 19, 2016; Accepted: September 29, 2016; Published: October 25, 2016
Copyright: © 2016 Guillou et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: This work was funded by the Empire State Stem Cell Fund through New York State Department of Health NYSTEM contract C029542 and from NSF1354421a grants to NB. It was also funded by a TSRI Advanced Discovery Institute grant provided by Novartis and a National Institute of Health grant of the National Institute of Allergy and Infectious Diseases R01 AI093687-01A1 to NCF. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Mounting appropriate immune responses against pathogens is critical for the survival of all animals. Mechanisms to both eliminate microbes and resolve infection by returning the immune system to basal activity are necessary to maintain an adequate and balanced immune response [1,2]. Alterations in these responses can lead to immune deficiency or auto-inflammation [3–5]. Yet, to date, how these mechanisms are coordinated upon infection remains unclear.
Drosophila is a prime model to genetically dissect humoral and cellular innate immune responses to a variety of pathogens [6–8]. Humoral responses include the pro-phenoloxidase (PO) cascade, which leads to the generation of reactive oxygen species and melanization, and the rapid production of antimicrobial peptides (AMPs) regulated by the Toll and Imd pathways . Upon recognition of microbial lysine (Lys)-type peptidoglycan (PGN), damage-associated molecular patterns (DAMPs), or exogenous protease activity, the Toll pathway promotes the nuclear translocation of the NF-κB-like transcription factor Dorsal-related Immune Factor (Dif) to induce AMP genes, such as Drosomycin [6,9]. In contrast, detection of bacterial meso-diaminopimelic acid (DAP)-type peptidoglycan activates the Imd pathway and leads to the nuclear translocation of the NF-κB-like transcription factor Relish (Rel) to induce transcription of AMP genes, such as Diptericin [10,11]. It has also been shown that proteases, such as Elastase and Mmp2, can activate the Imd pathway through cleavage of the receptor PGRP-LC . As in mammals, chronic activation of immune responses is deleterious to the fly, and negative regulators are required to maintain immune homeostasis [13–15]. For instance, amidase PGN recognition proteins (PGRPs), such as PGRP-LB and PGRP-SC, negatively regulate the Imd pathway by enzymatically degrading PGN [14–16].
Phagocytosis and encapsulation are key cellular innate immune responses . Phagocytosis allows for the uptake and digestion of microbes and apoptotic cells by phagocytes, including specialized immune cells called plasmatocytes [7,17]. Encapsulation results in the isolation and melanization of large materials, such as wasp eggs or damaged tissues, by dedicated immune cells named lamellocytes . Both phagocytosis and humoral responses are required to fight infection. Indeed, decreasing the phagocytic ability of plasmatocytes by pre-injecting latex beads, which they take up, impairs fly survival upon infection with Gram-positive bacteria . Similarly, inhibiting phagocytosis increases the susceptibility of Imd pathway-deficient flies to Escherichia coli (E. coli) infection, arguing that phagocytosis and the humoral response act in parallel . Plasmatocytes were proposed to activate the production of AMPs by releasing immunostimulatory pathogen-associated molecular patterns (PAMPs) following phagocytosis . They also express cytokines such as Unpaired 3 (Upd3), a ligand of the JAK-STAT pathway, which regulates immune-related genes . Yet, ablation of the majority of plasmatocytes by targeted apoptosis has only a moderate effect on the fly’s ability to fight infection [23,24]. Therefore, the role of phagocytosis in the regulation of the humoral response and the resolution of infection remains unclear.
Several plasmatocyte receptors promote the recognition and engulfment of bacteria . The scavenger receptor dSR-CI and a transmembrane protein, Eater, bind to both Gram-negative and -positive bacteria [26,27]. The membrane receptor PGRP-LC binds to and engulfs Gram-negative but not Gram-positive bacteria, and its membrane localization is dependent on the nonaspanin TM9SF4 [28,29]. Draper (Drpr) promotes clearance and degradation of neuronal debris and apoptotic cells via phagosome maturation, as well as phagocytosis of Staphylococcus aureus (S. aureus) together with the integrin βv and PGRP-SC1 [30–32]. Nimrod C1, which is related to Eater and Drpr, promotes phagocytosis of both S. aureus and E. coli by Drosophila S2 cells, and suppression of its expression in plasmatocytes inhibits phagocytosis of S. aureus . Peste, a member of the CD36 family of scavenger receptors plays a role in the recognition and uptake of Mycobacterium by S2 cells . Finally, croquemort (crq), another CD36 family member, promotes apoptotic cell clearance by embryonic plasmatocytes  and phagosome maturation of neuronal debris by epithelial cells .
In mammalian immunity, CD36 promotes the uptake of oxidized low density lipoproteins (oxLDLs) [37,38] and also regulates the host inflammatory response [39,40]. In addition, it is required to fight Mycobacteria and S. aureus infections in mice  and to induce pro-inflammatory cytokines in response to Plasmodium falciparum infection . Using two lethal deficiencies that delete crq (as well as other genes), we previously proposed that Crq was specific to apoptotic cell clearance, as crq-deficient embryonic plasmatocytes retained some ability to engulf both E. coli and S. aureus in vivo . However, Crq was subsequently implicated in phagocytosis of S. aureus by S2 cells, a heterogeneous cell line with phagocytic abilities derived from late embryonic stages . Thus, we generated a knock-out of crq and further investigated its role in microbial phagocytosis and its relationship with the humoral response at larval and adult stages in vivo.
Drosophila plasmatocytes derive from pro-hemocytes originating either in the procephalic mesoderm of the embryo, with some further expanding by self-renewal in larval hematopoietic pockets, or from a second hematopoietic organ, the larval lymph glands, that persist to adulthood, or finally from adult hematopoietic hubs [43–51]. Here, we show that Crq is a major marker of plasmatocytes that is not required for hematopoiesis. The survival to adulthood of crq knock-out (crqko) mutants allowed us to quantitatively demonstrate that crq is required for pupae to survive environmental microbe infections and for adults to resist infection against Gram-negative and Gram-positive bacteria and fungi. crqko flies tolerate infections as well as control flies, but are unable to efficiently eliminate microbes. Indeed, crqko plasmatocytes are poorly phagocytic and defective in phagosome maturation. Crq acts parallel to the Imd and Toll pathways in eliminating pathogens, and crqko flies display elevated and persistent Dpt and upd3 expression, demonstrating that mutating crq promotes a state of chronic immune activation. As a consequence, crqko flies die prematurely with early signs of gut dysplasia and premature intestinal stem cell hyperproliferation. Therefore, we propose a model wherein crq is central to immune and organismal homeostasis. Overall, our results shed new light on the links between phagocytes, commensal microbes, gut homeostasis, and host lifespan.
Croquemort is a major plasmatocyte marker and not required for hematopoiesis
In Drosophila adults, plasmatocytes (the phagocytic hemocyte lineage) originate from both embryonic and larval hematopoiesis . crq is expressed in embryonic and larval plasmatocytes, as well as in S2 cells . To test whether crq is expressed in adult plasmatocytes, we performed dual staining with combinations of GFP or dsRed and Crq antibodies of hemocytes bled from previously characterized transgenic plasmatocyte-reporter lines: eater-nls::GFP, eater-dsRed, and Hml-Gal4>UAS-GFP (Hemolectin-positive hemocytes) [54,55] (Fig 1A and 1B). We found that 83.3±4.4% of hemocytes of Hml-Gal4>UAS-GFP and eater-dsRed carrying flies were positive for both markers, while 16.7±4.4% were positive for eater-dsRed alone (Fig 1C). Crq immunostaining of hemocytes bled from eater-nls::GFP flies revealed that 85.2±2.6% of them were Crq and eater-dsRed positive, while 14.8±2.6% were Crq-positive but did not express eater-dsRed (Fig 1C). From this, we extrapolated that about 72.4% of circulating hemocytes are positive for all three markers, 12.8% are double positive for Crq and Eater, and 14.8% solely express Crq (Fig 1C). Therefore, crq is expressed in all Eater and Hml-positive hemocytes and marks the majority, if not all, adult plasmatocytes.
(A-B) Crq and GFP immunostainings of eater-nlsGFP (A) and GFP immunostaining of eater-dsRed; hml-gal4>UAS-GFP plasmatocytes (B). (C) Quantification of experiments in A and B reveals subpopulations of Crq-positive plasmatocytes, of Crq- and Eater-positive plasmatocytes, and of a majority of plasmatocytes expressing all three markers Crq, Eater and HML. (D) Relative hemocyte numbers (in %) of crqko larvae and 3-to-5- day-old adults compared to wild-type controls. Mean values of at least 5 repeats are represented ±SE. ***p<0.001 (Student’s T-test). (E) Percentages of homozygous crqko versus CyO-GFP-positive L3 larvae or adult flies emerging from crqko/CyO,GFP heterozygous stock maintained on conventional or antibiotic-supplemented medium. (F) Schematic of health status of crqko homozygous individuals emerging from cross of crqko homozygous males and females on conventional, antibiotic-supplemented or axenic medium.
To study its role in vivo, we generated a knock-out allele of crq (crqko) by homologous recombination . This mutant deletes the entire crq open reading frame (S1A Fig), and thus abolishes its expression . As previously reported , crq was not required for embryonic hematopoiesis. As for crq deletion mutants, crqko embryonic plasmatocytes were less efficient at clearing apoptotic cells, having a phagocytic index of 1.6±0.2 versus 2.45±0.3 apoptotic cells/plasmatocyte for wild-type embryos (p<0.05, S1B Fig). Homozygous crqko flies were viable and appeared morphologically normal. To ask whether crq is required for hematopoiesis at later developmental stages, we recombined an eater-nls::GFP transgene (i.e., the broadest plasmatocyte reporter after Crq) (Fig 1C) into the crqko mutants, bled larvae and adults, and semi-automatically scored their eater-nls::GFP positive plasmatocytes by microscopy (S1C Fig and Fig 1D). As previously reported for wild-type [56,57], adult crqko flies had about 5-fold less plasmatocytes than larvae, and their number of eater-nls::GFP-positive plasmatocytes at both larval and adult stages were similar to that of wild-type flies (Fig 1D). Pro-hemocytes that differentiate into plasmatocytes can also differentiate into crystal cells, which are involved in melanization . Furthermore, self-renewing plasmatocytes of the embryonic lineage can also differentiate into crystal cells by trans-differentiation [59,60]. Thus, we tested whether crqko flies have differentiated crystal cells by scoring the melanotic dots formed following heat-induced crystal cell lysis. We found no significant difference between crqko and wild-type larvae (S1D Fig). Therefore, Crq is a major plasmatocyte marker that is not required for hematopoiesis or hemocyte differentiation.
croquemort mutant flies are susceptible to environmental microbes
While crqko homozygous flies were viable to adulthood, we could not maintain a homozygous stock on conventional fly food. We found that 36±3.2% homozygous crqko larvae arose from crosses between crqko heterozygous flies over GFP-marked CyO balancer chromosome, indicating full viability of the homozygous larvae (Fig 1E). However, only 18±1.7% of emerging adults were homozygous crqko flies, indicating that half of the crqko homozygous progeny died during pupariation. Because flies with decreased plasmatocyte counts undergo pupal death associated with the presence of otherwise innocuous environmental microbes , we asked whether supplementing the food with antibiotics could rescue crqko lethality. With this treatment, we recovered 29±3.6% of crqko homozygous adults (Fig 1E), indicating a partial rescue of pupal lethality (homozygous vs balanced adults, p = 0.021). These results suggest that crqko pupae are susceptible to environmental microbes.
No adult progeny could be recovered from crqko homozygous crosses on conventional fly food, but crqko adults emerged in the presence of antibiotics that gave rise to a second adult progeny (Fig 1E and 1F). Maintaining a homozygous viable stock with antibiotics, however, remained difficult. We next bleached homozygous crqko embryos and raised them on sterile food. Under these axenic conditions, we successfully cultured a homozygous crqko line (Fig 1F). Therefore, environmental microbes represent a health constraint for crqko homozygous flies.
croquemort mutant flies are broadly susceptible to infection
The susceptibility of crqko pupae to environmental microbes suggested that crq is required to mount an appropriate immune response. We next asked whether crq was up-regulated in flies injected with the Gram-negative bacterium Pectinobacterium (previously known as Erwinia) carotovora 15 (Ecc15) or the Gram-positive Enterococcus faecalis (E. faecalis). As anticipated, there was no crq expression in unchallenged (UC) or infected crqko flies as detected by RT-qPCR (Fig 2A). While crq was expressed in both UC pXH87-crq transgenic (the parental transgenic strain used for the generation of crqko flies, hereafter referred to as PXH87) and Canton S (Cs) control flies, it was not up-regulated within the first 24hrs of infection with Ecc15 or E. faecalis (Fig 2A). However, we cannot exclude the possibility that crq may be up-regulated in plasmatocytes specifically at these early time points after infection. Its expression was also not altered in mutant flies for the NF-κB-like transcription factor Relish (RelE20) downstream of the Imd pathway, or in flies mutant for the Toll ligand spz (spzrm7), upstream of the Toll pathway during that time-frame . Surprisingly, we did observe an increase in crq mRNA levels at 36 (p = 0.0076) and 132 hrs (p = 0.0213) post Ecc15 infection (S2A Fig), but did not detect any upregulation of crq mRNA levels at 36 and 132 hours post E. faecalis infection (S2B Fig) (p>0.05). Altogether, our data show that crq does not appear to be induced by infection in whole adult extracts during the first 24 hours post infection with Ecc15 and E. faecalis, and its expression appears independent of the Toll and Imd pathways. However, at later time-points after infection crq can be upregulated in a pathogen-specific manner, as seen with Ecc15 here.
(A) Relative percentages of crq mRNA levels of UC Cs and PXH87 controls, crqko, RelE20, and spzrm7 mutant flies at 4, 10 or 24 hrs after Ecc15 or E. faecalis infections when compared to that of UC Cs flies. Mean values of at least 3 repeats are represented ± SE. (B-G) Survival curves (in %) over time of Cs and PXH87 control flies, crqko, and RelE20 or spzrm7 homozygous male flies after septic injury with Ecc15 (B), E. coli (C), E. faecalis (D), C. albicans (E), S. aureus (G), or after spore coating with B. bassiana (F). The curves represent the average percent survival ±SE. **p<0.01 ***p<0.001 in a log rank test.
To assess the susceptibility of crqko male (Fig 2B–2G) and female (S2B–S2G Fig) flies to a variety of pathogens, we monitored their survival to these infections over time. When challenged by septic injury with the Gram-negative bacterium Ecc15, male (Fig 2B) and female (S2C Fig) crqko flies were more susceptible than Cs and PXH87 control flies to this infection (p<0.0001). crqko flies all died within 336 hrs post-infection (hpi), while only 64±6.8% and 67±6.5% of PXH87 and Cs flies had died by that time-point. crqko flies were, however, less susceptible than RelE20 mutants (p<0.0001), which are defective in the production of AMPs downstream of Imd . All RelE20 flies died within 72 hpi, while only 56±7.7% of crqko flies had succumbed by that same time-point (Fig 2B). To verify that the susceptibility to Ecc15 infection was due to the crqko mutation and not to a background mutation, we infected trans-heterozygous flies for crqko and Df(2L)BSC16, which deletes crq, with Ecc15 (S2D Fig). These flies were as susceptible to Ecc15 infection as the crqko homozygous flies; they all died within 288 hpi, indicating that the crq mutation is responsible for this phenotype (S2D Fig). crqko flies also succumbed to infection with E. coli (39±8.1% survival at 336 hpi), a Gram-negative bacterium that does not kill Cs (97±2.5% survival) or PXH87 (86±5.6% survival) flies. However, crqko flies were less susceptible to E. coli infection than RelE20 flies, which all died within 312 hpi (p<0.0001) (Fig 2C and S2E Fig). Therefore, crqko flies are susceptible to various Gram-negative bacterial infections.
Similarly, crqko flies were more susceptible to infection with the Gram-positive bacterium E. faecalis than controls (p = 0.0006) (Fig 2D and S2F Fig) and died in 312 hpi. However, they were less susceptible than spzrm7 flies (p<0.0001), which are defective in the production of AMPs downstream of Toll and died within 72 hpi (Fig 2D and S2F Fig). crqko flies also died with intermediate susceptibility between that of control and spzrm7 flies (p<0.0001 for both) after septic injury with the pathogenic yeast Candida albicans (Fig 2E and S2G Fig). Similarly, crqko flies were significantly more susceptible to exposure to spores of the entomopathogenic fungus Beauveria bassiana than Cs and PXH87 flies (p<0.0001), but less susceptible than spzrm7 flies (p<0.0001) (Fig 2F and S2H Fig). Finally, crqko flies were more susceptible to S. aureus infection than spzrm7 flies (p = 0.0073) (Fig 2G and S2I Fig), and spzrm7 flies were only slightly more susceptible than Cs and PXH87 flies (p = 0.0006 and p<0.0001 respectively). Therefore, crqko flies are susceptible to Gram-positive bacteria and fungal infections and strongly susceptible to infection with S. aureus, a bacterium specifically cleared by phagocytosis [19,62,63].
These results argue that crq is required to fight infection. To further confirm this, we drove the expression of a UAS-crq transgene under the control of a crq promoter-Gal4 driver in the crqko flies (crqko; crq-Gal4>UAS-crq). These rescue flies were no longer susceptible to Ecc15 (S3A Fig), E. faecalis (S3B Fig), and B. bassiana (S3C Fig) infections (non-significant (ns) compared to PXH87, and p<0.0001 when compared to crqko flies) (S3A–S3C Fig). To assess the possible requirement of crq in hemocytes, we drove the expression of a UAS-crq transgene under the control of a hemocyte-specific serpent promoter-Gal4 driver in the crqko flies (crqko; srp-Gal4>UAS-crq). These flies were significantly less susceptible to Ecc15, E. coli, E. faecalis and C. albicans infections than crqko flies (p<0.0001, p<0.0001, p = 0.0004 and p<0.0001, respectively) (S3D–S3G Fig). We did not observe any significant differences between rescue experiments with the crq-Gal4 or srp-Gal4 drivers after infection with Ecc15, E. coli, or E. faecalis (p>0.05). The hemocyte-specific rescue of crqko flies infected with C. albicans, however, was slightly less efficient than the rescue with the crq-Gal4 driver (p = 0.0269). Thus, crq appears to be required mostly in phagocytes to fight infection by both Gram negative and Gram positive bacteria, although it appears to also be required in other tissues to fight C. albicans infection.
croquemort mutant flies are tolerant but poorly resistant to infection
Multi-cellular organisms use two complementary strategies to fight infection: resistance, to eliminate microbes, and tolerance, to allow them to endure the infection and/or its deleterious effects [64,65]. Compared to controls, crqko flies die prematurely at around 552 hours even in the absence of infection (S4A Fig), suggesting these flies could be generally unfit or susceptible to damage. To test their response to abiotic damage, we pricked crqko flies with sterile needles at two separate thoracic sites. These flies did not die any earlier than non-pricked crqko flies (S4A Fig). Thus, despite their decreased lifespan, crqko flies are not susceptible to aseptic wounds.
To date, few studies have quantified the tolerance of immune-deficient flies [66,67]. Tolerance can be measured as the dose response curve relating health to microbe load. This curve takes the shape of a sigmoid; life expectancy in unchallenged conditions is considered as vigor, and the slope of the response curve (the portion of the health/load curve which is linear) estimates the ability to tolerate infection (S4B Fig) . crqko flies have shortened lifespan and therefore an altered vigor (S4A Fig). We further aimed to estimate whether crqko flies show a decrease in tolerance by measuring the relationship (statistical interaction) between microbial load and the corresponding health of the host [64,67]. We used three approximations to relate health to microbe load of crqko flies and focused on the linear part for each regression. First, we estimated the regression between the LT50 (time at which 50% of the flies are dead) of Ecc15 or E. faecalis-infected flies and the number of bacteria injected (measured as colony forming units or CFUs) (S4C and S4D Fig). We did not detect any significant LT50~Time interaction between PXH87 and crqko flies (p = 0.21782 for E. faecalis, p = 0.55800 for Ecc15) (S4C and S4D Fig). However, this measure of bacterial load does not take into account the growth of the pathogen within the host. We therefore also quantified the regression between LT50 and the number of bacteria in the flies at 24 hpi (Fig 3A and 3B). We detected significant LT50~Time interaction between PXH87 and crqko flies (p = 0.008486 for E. faecalis, p = 0.018965 for Ecc15), with PXH87 flies having lower tolerance than crqko flies (Fig 3A and 3B). Finally, to get another estimate of the health of the flies, we plotted the health/bacterial load curve using survival at 3 time-points post Ecc15 infection and their corresponding bacterial load (S4E Fig). We did not detect any significant survival-time interaction between PXH87 and crqko flies (p = 0.335111). Thus, while crqko flies die prematurely in the absence of infection, they do not show any decreased tolerance to infection when compared to control flies.
(A-B) Tolerance graphs given as the plot of regression between LT50 and the number of CFUs of Ecc15 (A) or E. faecalis (B) found in flies at 24hrs after septic injury for crqko homozygous and PXH87 flies. (C-D) Resistance graphs given as the log number of CFUs of Ecc15 (C) or E. faecalis (D) per crqko homozygous and PXH87 flies over time after septic injury. **p<0.01 ***p<0.001.
These data suggest that the increased susceptibility of crqko flies to infection is due to their inability to control bacterial growth. In order to test this hypothesis, we monitored bacterial load during the course of Ecc15 and E. faecalis infections. In PXH87 flies, Ecc15 is eliminated within the first 48hrs of infection to reach an apparent plateau of low number of CFUs that persist at 72hrs post-infection (Fig 3C). crqko flies were less able to clear Ecc15 than controls with higher bacterial loads throughout the infection (p<0.001 for 24, 48 and 72hrs) (Fig 3C). In contrast, despite an initial decline of CFUs at 48hrs, E. faecalis grew within control flies at 96 and 168hrs (Fig 3D). During the whole course of infection with E. faecalis, the bacterial loads were significantly lower in wild-type control flies than in the crqko flies (p<0.001 at 48, 96 and 168hrs) (Fig 3D). These data indicate that crq is required for efficient elimination of both Ecc15 and E. faecalis.
croquemort is required for engulfment of bacteria and phagosome maturation
crq is required for efficient phagocytosis of apoptotic cells (also known as efferocytosis) in vivo, and phagocytosis of S. aureus by S2 cells (S1B Fig and [35,41]). In addition, rescue of crq expression in hemocytes improved survival to various infections (S3D–S3G Fig), suggesting that crq could alter microbial phagocytosis. To test this hypothesis, we first compared the susceptibility of crqko flies to infection with that of mutants for two phagocytic receptors, Eater and Drpr [26,31]. crqko flies succumbed to Ecc15 infection significantly faster than eater-deficient (p = 0.0002) and drprrec8Δ5 loss-of-function flies (p<0.0001). 90±3.58% of crqko flies died within 192 hpi, while only 60±6.77% of drpr and eater mutants died in that same time (Fig 4A). However, the crqko flies were significantly less susceptible to Ecc15 than RelE20 flies (p<0.0001), which all died within 48 hpi (Fig 4A). In contrast, crqko flies succumbed to E. faecalis infection at a similar pace to that of both eater-deficient and drpr rec8Δ5 flies with 80–90% of all strains dying within 240 hpi (Fig 4B). However, all mutants were significantly less susceptible than spzrm7 flies, which died within 48 hrs of E. faecalis infection (p<0.0001) (Fig 4B).
(A-B) Survival curves (in %) over time of Canton S, crqko, drprΔ5rec8, eater deficient flies, and spzrm7 or RelE20 mutant flies after septic injury with Ecc15 (A) and E. faecalis (B). (C-D) Representative fluorescent images of abdomen sections of Canton S, PXH87, crqko mutant and crqko; crq-gal4>UAS-eGFP rescue flies at 45min after injection of Alexa 488-labeled E. coli (C) and S. aureus (D). (E-F) Quantifications of Alexa 488 fluorescent integrated density (IntDen) of experiments highlighted in C and D. respectively. (G) Confocal micrographs of plasmatocytes from wild-type and crqko homozygous flies carrying the eater-GFP transgene (in green) bled either before (UC) or at 3 and 5hrs after injection with Rhodamine-labeled E. coli (in red). Scale bar, 10μm. (H) Quantification of the experiment in (G) given as the percentage of Rhodamine fluorescence of bacteria contained within plasmatocytes relative to control plasmatocytes bled at 3hrs after challenge of wild-type and crqko homozygous flies. Results are presented for UC flies and flies at 45 min, 3 and 5hrs post-injection with Rhodamine E. coli. (I) Same quantification as in (H) for experiment with pHrodo red-labeled E. coli. (J) Confocal micrographs of plasmatocytes (in red) bled from PXH87 control and crqko flies carrying the eater::dsred transgene 4 days post-injury with GFP-expressing Ecc15.
To examine the precise role of crq in phagocytosis, we compared the amount of bacteria engulfed within 45min of thoracic injections of dead, Alexa 480-labeled E. coli and S. aureus in Cs, PXH87, and crqko flies as previously described [20,26] (Fig 4C and 4D). The crqko flies engulfed both E. coli and S. aureus bacteria with on average 66% less efficiency than control flies (Fig 4C–4F, respectively). This phenotype was completely rescued in crqko flies expressing a UAS-crq transgene under a crq-Gal4 driver (crqko, crq-Gal4>UAS-crq), which appeared to engulf more efficiently than control PXH87 flies (S5A Fig and S5B Fig). We speculate that this difference was due to the overexpression of crq in those flies.
To further assess the phagocytosis phenotype, wild-type and crqko flies carrying the eater-nls::GFP plasmatocyte-reporter were injected with dead rhodamine-labeled E. coli, bled, and their plasmatocytes were analyzed by confocal microscopy for internalized bacteria (Fig 4G). The rhodamine-fluorescence per eater-nls::GFP plasmatocyte was quantified at 45min, 3hrs, and 5hrs post-injection and normalized to that of WT plasmatocytes at 3hrs post-injection (Fig 4H). In WT plasmatocytes, the relative rhodamine-fluorescence increased as early as 45min, peaked at 3hrs, and decreased after 5hrs, as bacteria were presumably digested in mature phagosomes (Fig 4H). In contrast, crqko plasmatocytes accumulated about 2-fold fewer bacteria than controls at 45min and 3hrs post-injection, but accumulated 1.7-fold more bacteria by 5hrs post-injection. In addition, at 45min post-injection, most bacteria were internalized within wild-type plasmatocytes (S5C Fig), whereas bacteria were often bound to the cell surface of crqko plasmatocytes without being internalized (S5D Fig). Thus, crqko plasmatocytes can engulf bacteria but are less efficient at it than controls at early time-points; they also appear to accumulate internalized bacteria over time. These results are consistent with a role for crq in promoting efficient uptake of bacteria. Moreover, the observed accumulation of bacteria in crqko plasmatocytes at 5hrs post-injection suggested that crq could also be required for phagosome maturation and digestion of bacteria. To test this, we injected control, crqko, and rescue flies with pH-sensitive pHrodo E. coli and S. aureus. pHrodo bacteria fluoresce when engulfed into a fully mature, acidified phagosome  (S5E and S5F Fig). After quantification, we observed about 50% less fluorescence in crqko when compared to controls at 1, 3, and 5hrs post-injection (S5G and S5H Fig, p<0.5 when comparing PXH87 and crqko flies). This phenotype was again completely rescued in crqko, crq-Gal4>UAS-crq flies (S5E and S5F Fig and S5G and S5H Fig, p>0.5 when comparing PXH87 and rescue flies). At the single cell level, crqko plasmatocytes had up to 63±5.66% and 55±7.46% less pHrodo E. coli than controls at 3 and 5hrs, respectively (Fig 4I).
Finally, to ask whether mutating crq resulted in persistence of pathogenic bacteria, we injected live GFP-labeled Ecc15 in control and crqko flies carrying the eater-dsred plasmatocyte reporter (Fig 4J and S5I Fig). Control PXH87 plasmatocytes had little to no GFP signal at 4 days post-infection, indicating that most bacteria had been engulfed and digested (Fig 4J and S5I Fig). In contrast, crqko plasmatocytes had a 6-fold higher GFP signal, demonstrating that live Ecc15 accumulate in crqko plasmatocytes (Fig 4J and S5I Fig). Taken together, these results show that crq is required for efficient microbial phagocytosis by playing a role in bacterial uptake and phagosome maturation.
Croquemort acts in parallel to the Imd and Toll pathways
Phagocytosis has been proposed as a key step to initiate AMP production . To assess the effect of mutating crq on AMP production downstream of both the Imd and Toll pathways, we next quantified the expression of Diptericin (Dpt) and Drosomycin (Drs)-encoding genes by RT-qPCR after Ecc15 or E. faecalis infections (Fig 5A and 5B). As previously reported, septic injury of control flies with Ecc15 induced Dpt expression, which peaked at 10 hpi and returned to near-basal levels within 48 hpi (Fig 5A). In crqko flies, Dpt expression was 2-fold higher than in control flies at 10hrs post-infection and failed to return to basal levels within 48hrs (Fig 5A). In contrast, there was no significant difference in Drs induction between control and crqko flies after E. faecalis inoculation (Fig 5B). Survival curves indicated that crqko flies were less susceptible to a non-pathogenic E. coli infection than RelE20 flies, while double mutants for crqko and RelE20 were statistically more susceptible than RelE20 or crqko mutants alone (Fig 5C). The extreme sensitivity of RelE20 flies to infection with pathogenic bacteria prevented us from carrying out these experiments with Ecc15. Instead, we inoculated the flies with 20 times fewer E. coli than previously used in Fig 2. Similarly, crqko and spzrm7 double mutants were also statistically more susceptible to C. albicans infection than the spzrm7 or crqko mutants alone (Fig 5D). Therefore, crq is not required for the induction of AMPs and acts in parallel to the Toll and Imd pathways.
(A-B) Relative levels (in %) of Dpt mRNA expression (normalized against RpL32) compared to Canton S at 10hrs after Ecc15 challenge, as determined by RT-qPCR on extracts of UC Canton S, PXH87, crqko, RelE20 and spzrm7 mutant flies, or at 4, 10, 24 or 48hrs after pricking with Ecc15 (A) or E. faecalis bacteria (B). Mean values of at least 3 repeats are represented ±SE. *p<0.05 ***p<0.001 (Student’s T-test). (C-D) Survival curves (in %) over time of Canton S, PXH87, crqko, RelE20 and crqko;RelE20 double mutant flies after septic injury with E. coli (C), or of Canton S, PXH87, crqko, spzrm7 and crqko; spzrm7 double mutant flies after septic injury with C. albicans (D). Curves represent average survivals ±SE. **p<0.01 ***p<0.001 in a log rank test. (E) Relative levels (in %) of Dpt mRNA expression (normalized against RpL32) compared to Canton S flies at 24hrs after Ecc15 challenge, as determined by RT-qPCR on extracts of UC Canton S, PXH87, crqko, PGRP-LBΔ and crqko; PGRP-LBΔ double mutant flies, or at 24, 48 or 72hrs after challenge with Ecc15. Mean values of at least 3 repeats are represented ±SE. ***p<0.001 (Student’s T-test). (F) Relative levels (in %) of upd3 mRNA expression (normalized against RpL32) compared to Canton S at 48hrs after Ecc15 challenge, as determined by RT-qPCR on extracts of UC Canton S, PXH87 and crqko mutant flies, or at 48 and 72hrs after challenge with Ecc15. Mean values of at least 3 repeats are represented ±SE. ***p<0.001 (Student’s T-test).
These results suggested that aberrant phagocytosis in crqko flies can result in enhanced and persistent Imd pathway activation. Multiple negative regulators of the Imd pathway help maintain immune homeostasis. For example, Peptidoglycan Recognition Proteins (PGRPs) with amidase activity, such as PGRP-LB, degrade immunostimulatory molecules . Thus, we next assessed Dpt expression levels by RT-qPCR in single and double PGRP-LBΔ and crqko mutants upon Ecc15 infection. Single crqko and PGRP-LBΔ mutants expressed statistically higher levels of Dpt than Cs and PXH87 controls at 48hrs (Fig 5E). The Dpt expression resolved back to basal levels within 72hrs post infection in control flies, but remained high in single PGRP-LBΔ or crqko mutants despite a steady decline in its expression (Fig 5E). Moreover, double mutants for crqko and PGRP-LBΔ expressed Dpt at levels 5-fold higher than controls at 24hrs post-infection, and levels remained high at 48 and 72hrs (Fig 5E). These results demonstrate the critical interplay between phagocytosis and negative regulators of the immune system to achieve proper resolution of AMP expression upon systemic infection.
Plasmatocytes are also a major source of cytokine production upon systemic infection. Upd3, the Drosophila analogue of IL-6, can induce the JAK-STAT pathway, which regulates the systemic immune response and metabolic homeostasis in the fat body, as well as gut homeostasis [6,22,69,70]. Using RT-qPCR, we asked whether crq is required for upd3 expression upon Ecc15 infection. Control flies displayed a small and temporary induction of upd3 expression that resolved within 72hrs (Fig 5F). In contrast, UC and Ecc15-challenged crqko flies showed a 1.5-fold stronger induction of upd3 expression, which further increased over 72hrs (Fig 5F). Thus crq is not required to induce upd3 expression, but crq mutation results in enhanced and continuously increasing upd3 expression. Altogether, these results demonstrate that crq is required for bacterial clearance and mutation of crq alters the resolution of AMPs and Upd3 cytokine production.
croquemort mutant flies have a short lifespan with early gut dysplasia
PGRP-LBΔ and RelE20 mutants all die prematurely, within about 696 hrs (29 days) of age, when compared to wild-type (p<0.0001) and PXH87 (p<0.0001) control flies, which die after about 912 hrs (37 days) on conventional food at 29°C  (Fig 6A). crqko flies died on average within 552 hrs (23 days), considerably earlier than RelE20 and PGRP-LBΔ mutants (p<0.0001). Double crqko and PGRP-LBΔ or crqko and RelE20 mutants died within about 480 hrs (20 days) and 408 hrs (17 days) of age, respectively (Fig 6A). Antibiotic treatment partially rescued these phenotypes, as the lifespan of crqko flies and the double mutants increased significantly (p<0.0001) (Fig 6B). To ask whether the premature aging of crqko flies might correlate with a loss of immune cells or their function, we estimated the number of plasmatocytes present in control and crqko flies using the eater-nlsGFP reporter (Fig 6C). As previously reported [56,57], the number of plasmatocytes was decreased by about 40% in 16-day-old control flies (Fig 6C), while similarly aged crqko flies had lost 80% of their plasmatocytes (Fig 6C). Treatment with antibiotics rescued this crqko phenotype but had no effect on the plasmatocyte counts of control flies. crqko flies also lost about 40% of their plasmatocytes at 4 days post-E. faecalis infection when compared to similarly challenged wild-type controls (Fig 6D). This loss of crqko hemocytes may be a consequence of accumulation of undigested bacteria inside their phagosomes. Thus, crq is required for plasmatocytes to survive innocuous or pathogenic bacterial infection.
(A) Survival curves (in %) over time of PXH87, crqko, RelE20, PGRP-LBΔ, crqko;RelE20 and crqko;PGRP-LBΔ double mutant flies. (B) Survival curves (in %) over time of PXH87, crqko, and crqko;PGRP-LBΔ double mutant flies raised on conventional or antibiotics-supplemented medium. Curves represent average survivals ±SE. **p<0.01 ***p<0.001 in a log rank test. (C) Hemocyte counts of 3–5 and 16 days old PXH87 and crqko adult flies raised in UC conditions on conventional or antibiotics-supplemented medium. Mean values of at least 3 repeats are represented ±SE. *p<0.05 **p<0.01. (D) Relative number (in %) of plasmatocytes in crqko flies compared to WT flies (normalized at 100%) 4 days post E. faecalis infection. Mean values of at least 3 repeats are represented ± SE. *p<0.05 ***p<0.001 (Student’s T-test). (E) Relative levels (in %) of Dpt mRNA expression (normalized against RpL32) compared to 3 days old PXH87 flies, as determined by RT-qPCR on extracts of UC 3 or 8-days old PXH87, crqko, PGRP-LBΔ, and crqko; PGRP-LBΔ double mutant flies. (F) Relative levels (in %) of upd3 mRNA expression levels (normalized against RpL32) compared to 3 days old PXH87 flies, as determined by RT-qPCR on extracts of UC 3, 8 and 16 days old PXH87 and crqko flies. (G) Number of mitotic PH3 positive cells per midgut of PXH87, crqko and rescue flies, and of crqko;PGRP-LBΔ and crqko;RelE20 double mutant flies. (H) Survival curves (in %) over time of PXH87, crqko, upd3;crqko double mutant flies, as well as crq-Gal4 and srp-Gal4 rescue flies. (I) Number of mitotic PH3-positive cells per midgut of PXH87, crqko, upd3;crqko double mutants and hemocyte-specific rescue flies. Mean values of at least 3 repeats are represented ±SE. **p<0.01 ***p<0.001.
Dpt expression in wild-type and PXH87 flies is relatively low and stable over the first 8 days of their lives and increases as flies age  (Fig 6E). Strikingly, Dpt expression was 70-fold higher in 8-day-old crqko flies and nearly 1,100-fold higher in the double mutants for crqko and PGRP-LBΔ compared to controls (Fig 6E). Thus, the Imd pathway is strongly up-regulated early on in the life of these mutant flies, even in the absence of infection. This points to a role for Crq in phagocytosis and in maintaining immune homeostasis. Likewise, upd3 expression steadily increased as PXH87 flies aged, and it was further enhanced by nearly 10-fold in 8- and 16-day-old crqko flies (Fig 6F). Antibiotic treatment partially rescued the levels of Dpt expression in crqko flies (S6A Fig), arguing that the hyper-activation of the Imd pathway in these flies results from their inability to control environmental microbes. To address this, we plated fly extracts on both LB (on which most pathogens can grow) and MRS (on which most Drosophila microbiota can grow) agar plates and quantified the resulting CFUs (S6B Fig). In line with previous studies, the CFUs obtained from 2 week-old control flies were in the range of 2,000 per fly (S6B Fig) [72–74]. Significantly fewer CFUs were recovered from PGRP-LBΔ mutants, while both crqko and RelE20 extracts showed a 10-fold increase. Double mutants for crqko and RelE20 had 50-fold more CFUs than controls (S6B Fig). Altogether, these results demonstrate that Crq and the Imd pathway act in parallel and are required for the management of environmental microbes.
Elevated levels of Upd3 are associated with midgut hyperplasia in aging flies [72,75]. In addition, loss of gut barrier integrity leads to early death in a microbiota-dependent manner [76,77]. Because 8-day-old crqko flies expressed high levels of upd3, we asked whether they also displayed premature gut hyperplasia by looking at the number of mitotic PH3-positive intestinal stem cells of their midgut. While PXH87 and crqko flies did not show any signs of midgut hyperplasia at day 7, midguts of 16 day-old crqko flies had a 2-fold increase in PH3-positive cells compared to that of similarly aged controls (p = 0.0109) (Fig 6G). This phenotype was completely rescued in crqko; crq-Gal4>UAS-crq flies (Fig 6G). The double mutants for crqko and PGRP-LBΔ or for crqko and RelE20 showed even higher levels of intestinal stem cell proliferation than controls (p = 0.03) and did so more prematurely (already in 7-day-old flies) (Fig 6G). The premature increase in midgut stem cell proliferation was partially dependent on Upd3, as upd3;crqko double mutant flies had significantly less mitotic cells (p = 0.04) and lived longer than crqko flies (p<0.0001) (Fig 6H and 6I). However, the lifespan of upd3;crqko double mutants flies was still shorter than that of PXH87 flies (p<0.0001), suggesting that additional mechanisms play a role in the shortened lifespan of crqko flies. We further asked whether crq is required in hemocytes to maintain intestinal homeostasis. Hemocyte-specific re-expression of crq led to a strong rescue of lifespan compared to crqko flies (p<0.0001) but not to the levels of PXH87 flies (p = 0.0462) and to a partial rescue of midgut hyperplasia in 16-day-old flies (p = 0.003 for crqko vs rescue and p = 0.0123 for rescue vs PXH87) (Fig 6H and 6I). Altogether, these results indicate that flies lacking crq display chronically elevated expression of upd3 that triggers early midgut hyperplasia and promotes premature death.
Our study shows that Crq is required for the engulfment of microbes by plasmatocytes and their clearance, and that the mild immune deficiency due to crq mutation is associated with increased susceptibility to infection, defects in immune homeostasis, gut hyperplasia, and decreased lifespan (S7 Fig). We have also re-confirmed a role for crq in apoptotic cell clearance, although the phagocytosis defect of crqko plasmatocytes is less severe than what had been previously observed with two lethal crq deficiency mutants, Df(2L)al and Df(2L)XW88 . A possible explanation is that these deficiencies may have deleted at least one other gene required for apoptotic cell clearance. Additionally, morphological defects associated with secondary mutations could have exacerbated the crq phagocytosis defect by preventing efficient plasmatocyte migration to apoptotic cells. These same deficiency mutants had been assessed qualitatively for phagocytosis of bacteria by injecting embryos with E. coli or S. aureus; their plasmatocytes had no obvious defect in their ability to engulf these bacteria . However, a role for crq in phagocytosis of S. aureus, but not that of E. coli, was subsequently proposed based on S2 cell phagocytosis assays following knock-down of crq by RNAi . Here, we show that crq is required in vivo for uptake and phagosome maturation of both S. aureus and E. coli. A simple explanation of this discrepancy with E. coli could be that knocking down crq by RNAi is not sufficient to affect its role in E. coli phagocytosis (but sufficient to affect its role in S. aureus phagocytosis), and that completely abrogating crq expression by in vivo knock-out leads to a stronger phenotype with both bacteria. Our in vivo data in crqko flies further demonstrate that crq is required to resist multiple microbial infections, such as Ecc15, E. faecalis, B. bassiana, and C. albicans. These data therefore argue that crq plays a more general role in microbial phagocytosis than was previously anticipated. Our previous experiments to test whether crq is required for bacterial phagocytosis in embryos were qualitative rather than quantitative, and did not allow us to identify a role for crq at that stage . In contrast, the experiments we now report in adult crqko flies are quantitative and allowed us to identify a delay in phagocytosis, followed by a defect in bacterial clearance in crqko hemocytes. A possible explanation for this discrepancy would be that hemocytes may differ in their expression profile, behavior, and phagocytic ability at various developmental stages due to differences in their microenvironment and/or sensitivity to stimuli. Accordingly, it has recently been shown that the phagocytic activity of embryonic hemocytes acts as a priming mechanism, increasing the ability of primed cells to phagocytose bacteria at later stages . It is therefore possible that embryonic, larval and adult hemocytes display very different levels of priming and bacterial phagocytic activity, and that crq is required mostly in larval/adult bacterial phagocytosis. Alternatively, a potential defect in phagocytosis of bacteria by embryonic hemocytes of the crq deficiencies may have been suppressed by the deletion of (an)other gene(s) in that genomic region.
Because the immune competence of hemocytes varies during development [50,79,80], we were prompted to re-examine the potential role for crq in innate immunity by knocking it out. Here, we show that Crq is a major plasmatocyte marker at all developmental stages of the fly. We have found that crqko flies are homozygous viable, but short-lived, and can hardly be maintained as a homozygous stock in a non-sterile environment; crqko pupae become susceptible to environmental bacteria and their microbiota during pupariation. In a recent study, Arefin and colleagues induced the pro-apoptotic genes hid or Grim in plamatocytes and crystal cells using the hml-gal4 driver (Hml-apo) and observed a similar pupal lethality, but also associated with an induction of lamellocyte differentiation, and the apparition of melanotic tumors of hemocyte origin . The authors therefore concluded that the death of hemocytes triggered lamellocyte accumulation and melanotic tumor phenotypes . In contrast, we did not observe any obvious melanotic tumors in crqko flies, despite observing a loss of hemocytes in aging crqko flies (Fig 6C) and crqko flies subjected to Ecc15 infection (Fig 6D). One possible explanation is that hemocytes do not die of apoptosis in crqko flies, but of a distinct mechanism. Alternatively, crq mutation could affect more hemocytes than Hml-apo flies, as crq is expressed in all plasmatocytes, while Hml is only expressed in 72.4% of all plasmatocytes expressing crq (from Fig 1C). Thus the 27.6% of non-Hml plasmatocytes (thus non induced for apoptosis, which is hml-Gal4 dependent ) may respond to the death of the other plasmatocytes by inducing a signal that triggers the induction of lamellocytes and the subsequent formation of melanotic tumors. Considering the role of crq in apoptotic cell clearance, this signal may require a functional crq, which could explain why crqko flies do not develop melanotic tumors. Strikingly, in the Arefin study, as well as in previous studies, targeted ablation of plasmatocytes also made resulting ‘hemoless’ pupae more susceptible to environmental microbes [23,24,81]. Extensive tissue remodeling takes place at pupariation, and plasmatocytes are essential to remove dying cells, debris, and bacteria. Thus, it was argued that this increased susceptibility was likely due to environmental bacteria invading the body cavity after disruption of the gut . In addition, it was found that the gut microbiome of Hml-apo flies could influence pupal lethality, as the eclosure rate of Hml-apo flies varied depending on the quality of the food they were reared on . Accordingly, our rescue of the crqko pupal lethality with antibiotics demonstrates that their premature aging and death are indeed due to infection by normally innocuous environmental bacteria. Altogether, these data suggest that phagocytes and crq are important actors regulating the interaction between a host and its microbiome.
Hosts use both resistance and tolerance mechanisms to withstand infection and survive a specific dose of microbes [65,83]. crqko flies exhibit a shorter lifespan when compared to control flies, but they are equally tolerant to aseptic wounds and infections. The crqko flies are less resistant to infection, as crq is required to promote efficient microbial phagocytosis. crqko plasmatocytes can still engulf bacteria, albeit at a lower efficiency than their controls. Our data also demonstrate that crq plays a major role in phagosome maturation during bacterial clearance. This is in agreement with a recent study showing that crq promotes phagosome maturation during the clearance of neuronal debris by epithelial cells . Thus, crq is required at several stages of phagocytosis. Similar observations have been made for the C. elegans Ced-1 receptor and for Drpr, as both promote engulfment of apoptotic corpses and their degradation in mature phagosomes [84,85].
‘Hemoless’, Hml-apo and crqko flies are all more susceptible to environmental microbes and their microbiota. While it is not known whether mutants of eater, which encodes a phagocytic receptor for bacteria but does not play a role in phagosome maturation, are more susceptible to environmental microbes during pupariation, both eater mutants and ‘hemoless’ flies showed either decreased or unaffected systemic responses [23,24,26]. Hml-apo larvae however, showed an upregulation in Toll-dependent constitutive Drs mRNA levels whereas Dpt expression was suppressed . In contrast, crqko flies showed no significant difference in constitutive or infection induced expression of Drs, but showed an increased expression of Dpt with age, and infection induced an increased and chronic expression of Dpt. Altogether our results argue that phagosome maturation defects in crqko flies lead to persistence of bacteria and thus to an increased and persistent systemic immune response via the Imd pathway. Such defects in phagosome maturation are not present in hemocyte ablation experiments, which could explain different outcomes for the host immunity and survival.
We have found that Crq acts in parallel to the Toll and Imd pathways. In the mealworm Tenebrio molitor, hemocytes and cytotoxic enzymatic cascades eliminate most bacteria early during infection, and AMPs are required to eliminate persisting bacteria . These data suggest that AMPs act in parallel with hemocytes to fight infections. We have also found that crqko flies are more susceptible to infection with S. aureus than wild-type and Toll pathway-deficient flies. These results are consistent with S. aureus infection being mainly resolved via phagocytosis and Crq having a major role in this process. Surprisingly, we have observed the opposite for infection with other Gram-negative or positive bacteria and fungi. Drosophila mutants for AMP production were more susceptible to infection than crqko flies, arguing that AMPs are critical to eliminate the bulk of pathogens. Indeed, crq (thus phagocytosis) is not essential for Ecc15 elimination, but accelerates bacterial clearance. Our results also suggest that the defects in phagosome maturation may allow some bacteria to persist and grow within hemocytes, where they are hidden from systemic AMPs. Thus, depending on the microbe, humoral and cellular immune responses can act at distinct stages of infection. In this context, phagocytosis acts as a main defense mechanism against pathogens that may escape AMPs or modulate their production.
Chronic activation of immune pathways can be detrimental to organismal health [13–15]. In Drosophila, multiple negative regulators of the Imd pathway, including PGRP-LB, act in concert to maintain immune homeostasis [14–16]. We have observed that crqko flies sustain high production levels of the AMP Dpt and the cytokine Upd3, demonstrating that defects in phagocyte function can lead to chronic immune activation. Notably, the level of Dpt expression induced by activation of the Imd pathway in unchallenged conditions is stronger in crqko flies than was previously observed in mutants of three negative regulators of the Imd pathway, namely pirkEY, PGRP-SCΔ, and PGRP-LBΔ , and over 1,000-fold higher in PGRP-LBΔ, crqko double mutants. This is despite the persistence of only a few hundred bacteria in these mutants. This phenotype may be due solely to the accumulation of these persistent bacteria, or Crq may also function in plasmatocytes to remove immunostimulatory molecules from the hemolymph. Nonetheless, our study shows that plasmatocytes, Crq, and phagocytosis are all key factors in the immune response, and that losing crq induces a state of chronic immune induction.
The ability of a host to control microbes decreases with age, a phenomenon called immune senescence . The causes of immune senescence remain elusive, but the loss of immune cells with age and a decline in their ability to phagocytose have been suggested [56,57]. Recent studies have argued that microbial dysbiosis and disruption in gut homeostasis contribute to early aging [76,77,87]. In addition, persistent activation of the JAK-STAT pathway in the gut has been linked to age-related decline in gut structure and function . Aging crqko flies lose a greater number of hemocytes than wild-type flies after infection, which may be the result of accumulating bacteria in these hemocytes in which phagosomes fail to mature. The premature death of crqko flies could be partially rescued by the presence of antibiotics. This demonstrates that phagocytosis, and phagosome maturation in particular, plays a crucial role in managing the response to environmental microbes and potentially, the gut microbiota directly to promote normal aging. We have also found that chronic upd3 expression in crqko flies triggers premature midgut hyperplasia, which is known to alter host physiology and promote premature aging [72,76,89]. It has recently been proposed that plasmatocytes can influence gut homeostasis by secreting dpp ligands and modulating stem cell activity . Our results reinforce the possibility of an interaction between plasmatocyte function and gut homeostasis, and suggests that cytokines derived from hemocytes can trigger cell responses in the gut. These results are also in agreement with a recent publication showing that Upd3 from hemocytes can trigger intestinal stem cell proliferation . Altogether, these results demonstrate that the interaction between hemocytes and the gut tissue are central to host health, and our data demonstrate that phagocytic defects can be associated with chronic gut inflammation and aberrant intestinal stem cell turn-over. As gut aging and barrier integrity are in turn important to maintain bodily immune homeostasis , we propose the following model: in crqko flies, plasmatocyte-derived cytokines accelerate gut aging promoting loss of gut homeostasis and microbial dysbiosis, with immune and plasmatocyte activation acting in a positive feedback loop (S7 Fig).
Collectively, our data show that Crq is essential in development and aging to protect against environmental microbes. Interestingly, the impact of mutating crq on host physiology is strikingly different from previously reported phagocytic receptor mutations. We speculate that this could be due to its dual role in uptake and phagosome maturation during phagocytosis. Crq is required for microbial elimination in parallel to the Toll and Imd pathways and acts to maintain immune homeostasis. This situation is surprisingly reminiscent of inflammatory disorders, such as Crohn’s disease, that result from primary defects in bacterial elimination and trigger chronic immune activation and disruption of gut homeostasis. Further characterization of the crq mutation in Drosophila will provide an interesting conceptual framework to understand auto-inflammatory diseases and their repercussions on immune homeostasis and host health.
Materials and Methods
Fly rearing, stocks, and mutant generation
All stocks were raised at 22°C on standard medium, unless otherwise specified. RelE20, spzrm7, and PGRP-LBΔ stocks were described in [15,61,91]. The crqko stock was generated by homologous recombination, which removed the majority of the crq open reading frame  and (S1B Fig).
Bacterial strains, infection experiments, and antibiotic treatment
For bacterial infections, males or females were pricked in the thorax with a needle previously dipped in a concentrated pellet of the tested pathogen. The following bacterial or yeast strains were used at the indicated optical density (OD) taken at 600 nm: Ecc15 (OD = 200), E. coli (OD = 200 and OD = 10), E. faecalis (OD = 5), S. aureus (OD = 0.5), C. albicans (OD = 200). For B. bassiana infection, flies were shaken in a petri dish with mature germinating Beauveria for spore coating. All infections and aging experiments were performed at 29°C. In antibiotic treatments, a cocktail of kanamycin, ampicillin, rifampicin, streptomycin, and spectinomycin (5mg/mL each) was added to the fly medium. Axenic stocks were generated as described in [72,73]. Survival experiments represent at least 3 independent repeats with 20 flies (60–100 flies tested). Survival was analyzed by a Log-rank test using the statistical programs R and Prism.
Quantification of bacterial CFUs
Flies were individually homogenized in 500 μl of sterile PBS using bead beating with a tissue homogenizer (OPS Diagnostics). Dilutions of the homogenate were plated onto LB agar or MRS agar with a WASP II autoplate spiral plater (Microbiology International), incubated at 29°C, and the CFUs counted. Results were analyzed using a Krustal-Wallis test in R.
Phagocytosis assays and plasmatocyte immunostaining
Flies were injected in their thorax with 69nl of pHrodo red or Alexa 488 bacteria (Life Technologies Inc.) using a nanoject injector (Drummond). The fluorescence within the abdomen of the flies was then imaged at 45min, 3hrs, and 5hrs post-injection with a Leica MZFLIII fluorescent microscope and DFC300 FX camera and quantified using Image J 2.0.0-rc-30/1.49s (NIH).
For ex vivo imaging, flies were injected with 46nl of PBS at 45min, 3hrs and 5hrs after infection to release all hemocytes, and 10 flies were bled on a lysine-coated slide by mechanically scraping their hemocytes onto a drop of PBS. Once settled for 10min on the slide, hemocytes were quickly dried and mounted with AF1 mounting solution (Citifluor Ltd). Slides were automatically scanned using a Zeiss LSM 700 confocal microscope, and the number of plasmatocytes and average fluorescence signal per plasmatocyte quantified.
For immunostaining, flies were bled as described above and the hemocytes fixed in a solution of PBS, Tween 0.1%, PFA 4% for 30min. The samples were incubated in PBT with 1% normal goat serum and Crq  and GFP antibodies (Roche) at 1:500 overnight at 4°C. Samples were washed at RT three times for 5min in PBS, incubated with the appropriate secondary antibodies at 1:1000 in PBT for 2hrs at RT, and washed three additional times in PBT. Samples were imaged with a Zeiss LSM 700 confocal microscope.
Total RNA was extracted from pools of 20 flies per time point using TRIzol (Invitrogen). RNA was reverse-transcribed using Superscript II (Invitrogen), and the qPCR was performed using SYBR green (Quanta) in a Biorad instrument. Data represent the ratio or relative ratio (in %) of mRNA levels of the target gene (crq, Dpt, Drs or upd3) and that of a reference gene (RpL32 also known as rp49). The primer sequences used in this study are provided in the supplementary material. All experiments were performed at least 3 times.
S1 Fig. crq is required for apoptotic cell clearance but not hematopoiesis.
(A) Schematic of wild-type (WT) versus crq-targeted allele in which most of the crq ORF was replaced by the FP-mini-w+ cassette. (B) Apoptotic cell phagocytosis indices of control PXH87 and crqko homozygous plasmatocytes of stage 13 embryos. (C) Characterization of the bleeding technique showing the average number of plasmatocytes per field of view in relation to the number of larvae bled. (D) Relative number of melanized dots following heat shock-induced crystal cell lysis in wild-type control (WT) versus crqko mutant larvae.
S2 Fig. crq is required in females to survive infection.
(A-B) Relative levels (in %) of crq mRNA expression (normalized against RpL32) as determined by RT-qPCR on extracts of flies after 12, 36 and 132 hrs post infection with Ecc15 (A) or E. Faecalis (B). (C) Percent survival over time of Canton S and PXH87 flies, crqko, RelE20 homozygous female flies upon septic injury with Ecc15. (D) Percent survival over time of PXH87, Df(2L)BSC16/CyO heterozygous, crqko homozygous and crqko/Df(2L)BSC16 trans-heterozygous male flies upon Ecc15 septic injury. (E-I) Percent survival over time of Canton S and PXH87 flies, crqko, RelE20 or spzrm7 homozygous female flies upon septic injury with E. coli (E), E. faecalis (F) or C. albicans (G), after natural infection with B. bassiana (H) or after infection with S. aureus (I). Curves represent average survival ±SE. *p<0.05 **p<0.01 ***p<0.001 in a log rank test.
S3 Fig. Rescue of crq expression ameliorates survival to infection.
(A-C) Percent survival over time of PXH87 control, crqko homozygous mutant and crqko; crq-Gal4>UAS-crq rescue flies upon septic injury with Ecc15 (A) and E. faecalis (B), as well as upon natural infection with B. bassiana (C). (D-G) Percent survival over time of PXH87 control, crqko homozygous mutant, crqko; crq-Gal4>UAS-crq and crqko; srp-Gal4>UAS-crq rescue flies upon septic injury with Ecc15 (D), E coli (OD200) (E), E. faecalis (F) and C. albicans (G). Curves represent average survival ±SE. *p<0.05 and ***p<0.0001 in a log rank test.
S4 Fig. crq mutation does not alter host tolerance to infection.
(A) % survival over time of PXH87 control and crqko homozygous flies with or without aseptic wound. In the shaded area are the survival curves of crqko flies upon multiple infections from Fig 1. (B) The relationship between health and bacterial load (tolerance curve) is depicted here. A tolerance curve adopts a sigmoid shape, and we focus on the linear part of the relationship, where tolerance is represented by the slope of the regression health/load. (C, D) Tolerance graph of PXH87 and crqko flies given as the plot of regression between LT50 and the log number of injected bacteria for Ecc15 (C) or E. faecalis (D) septic injury. (E) Tolerance graph for PXH87 and crqko flies given as the plot of regression of their survival at 3 timepoints post infection against the log number of Ecc15 CFUs present at the same timepoint.
S5 Fig. crq is required for bacterial phagocytosis.
(A, B) Fluorescent images of abdomen of control pXH87, crqko and crqko; crq-Gal4 > UAS-crq rescue flies at 3hrs after injection of Alexa488 E. coli and Alexa488 S. aureus, respectively. (C, D) 3D reconstruction and sections of confocal zeta-stacks scans of eater-nls::GFP hemocytes. In WT most hemocytes internalize rhodamine E. coli at 45min post injection. In crqKO flies, a number of hemocytes instead show contact with bacteria not fully internalized. (E, F) Fluorescent images of abdomen of control PXH87, crqko and crqko; crq-Gal4 > UAS-crq rescue flies at 3hrs after injection of pHrodo red-E. coli and pHrodo red-S. aureus, respectively. (G, H) Quantification of average E. coli or S. aureus pHrodo red fluorescence present per fly abdomen in control PXH87, crqko mutant and crqko;crq-Gal4>UAS-crq rescue flies, respectively. * p<0.5; ** p<0.01. (I) Average GFP fluorescence per plasmatocyte of UC or PXH87 and crqko flies at 4 days post Ecc15-GFP injection.
S6 Fig. crq is required to manage environmental microbes.
(A) Relative percentage of Dpt mRNA expression (normalized against RpL32) in PXH87 and crqko flies raised on conventional or antibiotics-supplemented medium compared to UC 16 days-old PXH87 flies raised on conventional medium. (B) Number of CFUs per fly of 14 days-old PXH87, crqko, RelE20, PGRP-LB single mutants and crqko; RelE20 or crqko; PGRP-LB double mutant flies. a, b, c represents statistical grouping.
S7 Fig. Model for crq requirement in immunity.
In absence of crq, phagocytic function is decreased and absence of phagosome maturation is associated with a defect in bacterial clearance. This mild immune-deficiency in turns triggers a chronic activation of immune pathways and cytokine production, potentially secondary to the decreased bacterial clearance. This hyperactive immune response includes the activation of the Toll and Imd pathways, and the induction of the cytokine Upd3. This chronic immune activation results in the induction of early midgut hyperplasia and promotes a decrease in lifespan.
We thank David Duneau for insightful discussions and help with the statistical analyses. We also thank members of the Buchon and Franc labs for helpful comments on the manuscript. Finally, we would like to thank Katja Brückner and Daria Siekhaus for the hemocyte-specific serpent-gal4 driver flies.
- Conceptualization: NB NCF.
- Data curation: AG KT HW NCF NB.
- Formal analysis: AG KT HW NCF NB.
- Methodology: AG KT HW NCF NB.
- Supervision: NCF NB.
- Validation: AG KT HW NCF NB.
- Visualization: AG KT HW NCF NB.
- Writing – original draft: NCF NB.
- Writing – review & editing: AG KT HW NCF NB.
- 1. Serhan CN. The resolution of inflammation: the devil in the flask and in the details. the FASEB Journal. The Federation of American Societies for Experimental Biology; 2011;25: 1441–1448.
- 2. Ortega Gómez A, Perretti M, Soehnlein O. Resolution of inflammation: an integrated view. EMBO Molecular Medicine. WILEY VCH Verlag; 2013;5: 661–674. pmid:23592557
- 3. Hayee B, Rahman FZ, Sewell G, Smith AM, Segal AW. Crohn's disease as an immunodeficiency. Expert Review of Clinical Immunology. 2010;6: 585–596. pmid:20594132
- 4. Maggadottir SM, Sullivan KE. The intersection of immune deficiency and autoimmunity.—PubMed—NCBI. Current Opinion in Rheumatology. 2014;26: 570–578. pmid:25014038
- 5. Danzer C, Mattner J. Impact of microbes on autoimmune diseases.—PubMed—NCBI. Arch Immunol Ther Exp. SP Birkhäuser Verlag Basel; 2013;61: 175–186.
- 6. Buchon N, Silverman N, Cherry S. Immunity in Drosophila melanogaster—from microbial recognition to whole-organism physiology. Nat Rev Immunol. 2014;14: 796–810. pmid:25421701
- 7. Lemaitre B, Hoffmann JA. The host defense of Drosophila melanogaster. Annu Rev Immunol. 2007;25: 697–743. pmid:17201680
- 8. Royet J, Reichhart JM, Hoffmann JA. Sensing and signaling during infection in Drosophila. Current Opinion in Immunology. 2005;17: 11–17. pmid:15653304
- 9. Lemaitre B, Nicolas E, Michaut L, Reichhart JM, Hoffmann JA. The dorsoventral regulatory gene cassette spätzle/Toll/cactus controls the potent antifungal response in Drosophila adults. Cell. 1996;86: 973–983. pmid:8808632
- 10. Kaneko T, Golenbock D, Silverman N. Peptidoglycan recognition by the Drosophila Imd pathway. J Endotoxin Res. 2005;11: 383–389. pmid:16303095
- 11. Leulier F, Parquet C, Pili-Floury S, Ryu J-H, Caroff M, Lee W-J, et al. The Drosophila immune system detects bacteria through specific peptidoglycan recognition. Nat Immunol. 2003;4: 478–484. pmid:12692550
- 12. Schmidt R, Trejo T, Plummer T, Platt J, Tang A. Infection-induced proteolysis of PGRP-LC controls the IMD activation and melanization cascades in Drosophila. FASEB J. 2007.
- 13. Royet J, Gupta D, Dziarski R. Peptidoglycan recognition proteins: modulators of the microbiome and inflammation. Nat Rev Immunol. 2011;11: 837–851. pmid:22076558
- 14. Bischoff V, Vignal C, Duvic B, Boneca IG, Hoffmann JA, Royet J. Downregulation of the Drosophila immune response by peptidoglycan-recognition proteins SC1 and SC2. PLoS Pathog. Public Library of Science; 2006;2: e14. pmid:16518472
- 15. Paredes JC, Welchman DP, Poidevin M, Lemaitre B. Negative Regulation by Amidase PGRPs Shapes the Drosophila Antibacterial Response and Protects the Fly from Innocuous Infection. Immunity. 2011;35: 770–779. pmid:22118526
- 16. Zaidmanremy A, Hervé M, Poidevin M, Pili-Floury S, Kim M- S, Blanot D, et al. The Drosophila amidase PGRP-LB modulates the immune response to bacterial infection. Immunity. 2006;24: 463–473. pmid:16618604
- 17. Honti V, Csordás G, Kurucz E, Márkus R, Ando I. The cell-mediated immunity of Drosophila melanogaster: Hemocyte lineages, immune compartments, microanatomy and regulation. 2014;42: 47–56. Available: http://linkinghub.elsevier.com/retrieve/pii/S0145305X13001687 pmid:23800719
- 18. Krzemień J, Crozatier M, Vincent A. Ontogeny of the Drosophila larval hematopoietic organ, hemocyte homeostasis and the dedicated cellular immune response to parasitism.—PubMed—NCBI. Int J Dev Biol. 2010;54: 1117–1125. pmid:20711989
- 19. Nehme NT, Quintin J, Cho JH, Lee J, Lafarge M- C, Kocks C, et al. Relative roles of the cellular and humoral responses in the Drosophila host defense against three gram-positive bacterial infections. Gay N, editor. PLoS ONE. 2011;6: e14743. pmid:21390224
- 20. Elrod-Erickson M, Mishra S, Schneider DS. Interactions between the cellular and humoral immune responses in Drosophila. 2000;10: 781–784. pmid:10898983
- 21. Brennan CA, Delaney JR, Schneider DS, Anderson KV. Psidin is required in Drosophila blood cells for both phagocytic degradation and immune activation of the fat body. Curr Biol. 2007;17: 67–72. pmid:17208189
- 22. Agaisse H. Signaling Role of Hemocytes in Drosophila JAK/STAT-Dependent Response to Septic Injury. Dev Cell. 2003;5: 441–450. pmid:12967563
- 23. Crozatier M. Genetic ablation of Drosophila phagocytes reveals their contribution to both development and resistance to bacterial infection. J Innate Immun. Karger Publishers; 2009;1: 322–334. pmid:20375589
- 24. Charroux B, Royet J. Elimination of plasmatocytes by targeted apoptosis reveals their role in multiple aspects of the Drosophila immune response. Proceedings of the National Academy of Sciences. 2009;106: 9797–9802.
- 25. Chung Y-SA, Kocks C. Phagocytosis of bacterial pathogens. Fly. 2012;6: 21–25. pmid:22223092
- 26. Kocks C, Cho JH, Nehme N, Ulvila J, Pearson AM, Meister M, et al. Eater, a transmembrane protein mediating phagocytosis of bacterial pathogens in Drosophila. Cell. 2005;123: 335–346. pmid:16239149
- 27. Rämet M, Pearson A, Manfruelli P, Li X, Koziel H, Göbel V, et al. Drosophila scavenger receptor CI is a pattern recognition receptor for bacteria. Immunity. 2001;15: 1027–1038. Available: http://www.ncbi.nlm.nih.gov/pubmed/11754822 pmid:11754822
- 28. Perrin J, Mortier M, Jacomin A-C, Viargues P, Thevenon D, Fauvarque M-O. The nonaspanins TM9SF2 and TM9SF4 regulate the plasma membrane localization and signalling activity of the peptidoglycan recognition protein PGRP-L…—PubMed—NCBI. J Innate Immun. Karger Publishers; 2015;7: 37–46. pmid:25139117
- 29. Rämet M, Manfruelli P, Pearson A, Mathey-Prevot B, Ezekowitz RAB. Functional genomic analysis of phagocytosis and identification of a Drosophila receptor for E. coli. Nature. 2002;416: 644–648. pmid:11912489
- 30. Garver LS, Wu J, Wu LP. The peptidoglycan recognition protein PGRP-SC1a is essential for Toll signaling and phagocytosis of Staphylococcus aureus in Drosophila. Proceedings of the National Academy of Sciences. 2006;103: 660–665.
- 31. Hashimoto Y, Tabuchi Y, Sakurai K, Kutsuna M, Kurokawa K, Awasaki T, et al. Identification of Lipoteichoic Acid as a Ligand for Draper in the Phagocytosis of Staphylococcus aureus by Drosophila Hemocytes. J Immunol. 2009.
- 32. Shiratsuchi A, Mori T, Sakurai K, Nagaosa K, Sekimizu K, Lee BL, et al. Independent recognition of Staphylococcus aureus by two receptors for phagocytosis in Drosophila. Journal of Biological Chemistry. American Society for Biochemistry and Molecular Biology; 2012;287: 21663–21672. pmid:22547074
- 33. Kurucz E, Márkus R, Zsámboki J, Folkl-Medzihradszky K, Darula Z, Vilmos P, et al. Nimrod, a putative phagocytosis receptor with EGF repeats in Drosophila plasmatocytes. Curr Biol. 2007;17: 649–654. pmid:17363253
- 34. Philips J, Rubin E, Perrimon N. Drosophila RNAi screen reveals CD36 family member required for mycobacterial infection. Science. 2005;309: 1251–1253. pmid:16020694
- 35. Franc NC, Heitzler P, Ezekowitz R, White K. Requirement for croquemort in phagocytosis of apoptotic cells in Drosophila. Science. 1999;284: 1991–1994. pmid:10373118
- 36. Han C, Song Y, Xiao H, Wang D, Franc NC, Jan LY, et al. Epidermal cells are the primary phagocytes in the fragmentation and clearance of degenerating dendrites in Drosophila. Neuron. 2014;81: 544–560. pmid:24412417
- 37. Endemann G, Stanton LW, Madden KS, Bryant CM, White RT, Protter AA. CD36 is a receptor for oxidized low density lipoprotein. J Biol Chem. 1993;268: 11811–11816. pmid:7685021
- 38. SILVERSTEIN RL. Inflammation, atherosclerosis, and arterial thrombosis: role of the scavenger receptor CD36. Cleveland Clinic Journal of Medicine. Cleveland Clinic; 2009;76 Suppl 2: S27–30.
- 39. Sharif O, Matt U, Saluzzo S, Lakovits K, Haslinger I, Furtner T, et al. The scavenger receptor CD36 downmodulates the early inflammatory response while enhancing bacterial phagocytosis during pneumococcal pneumonia. J Immunol. American Association of Immunologists; 2013;190: 5640–5648. pmid:23610144
- 40. Stewart CR, Stuart LM, Wilkinson K, van Gils JM, Deng J, Halle A, et al. CD36 ligands promote sterile inflammation through assembly of a Toll-like receptor 4 and 6 heterodimer. Nat Immunol. 2010;11: 155–161. pmid:20037584
- 41. Stuart L, Deng J, Silver J, Takahashi K, Tseng A, Hennessy E, et al. Response to Staphylococcus aureus requires CD36-mediated phagocytosis triggered by the COOH-terminal cytoplasmic domain. J Cell Biol. 2005;170: 477–485. pmid:16061696
- 42. Gowda NM, Wu X, Kumar S, Febbraio M, Gowda DC. CD36 contributes to malaria parasite-induced pro-inflammatory cytokine production and NK and T cell activation by dendritic cells. Stager S, editor. PLoS ONE. Public Library of Science; 2013;8: e77604. pmid:24204889
- 43. Tepass U, Fessler L, Aziz A, Hartenstein V. Embryonic origin of hemocytes and their relationship to cell death in Drosophila. 1994;120: 1829–1837. pmid:7924990
- 44. Gold KS, Brückner K. Macrophages and cellular immunity in Drosophila melanogaster.—PubMed—NCBI. Seminars in Immunology. 2015;27: 357–368. pmid:27117654
- 45. Holz A, Bossinger B, Strasser T, Janning W, Klapper R. The two origins of hemocytes in Drosophila. Development. 2003;130: 4955–4962. pmid:12930778
- 46. Petraki S, Alexander B, Brückner K. Assaying Blood Cell Populations of the Drosophila melanogaster Larva. JoVE. 2015;: e52733–e52733.
- 47. Makhijani K, Alexander B, Tanaka T, Rulifson E, Brückner K. The peripheral nervous system supports blood cell homing and survival in the Drosophila larva. 2011;138: 5379–5391. pmid:22071105
- 48. Ghosh S, Singh A, Mandal S, Mandal L. Active Hematopoietic Hubs in Drosophila Adults Generate Hemocytes and Contribute to Immune Response. Dev Cell. 2015;33: 478–488. pmid:25959225
- 49. Honti V, Kurucz E, Csordás G, Laurinyecz B, Márkus R, Ando I. In vivo detection of lamellocytes in Drosophila melanogaster. Immunology Letters. 2009;126: 83–84. pmid:19695290
- 50. Rus F, Flatt T, Tong M, Aggarwal K, Okuda K, Kleino A, et al. Ecdysone triggered PGRP-LC expression controls Drosophila innate immunity. EMBO J. 2013.
- 51. Tan KL, Vlisidou I, Wood W. Ecdysone Mediates the Development of Immunity in the Drosophila Embryo. Current Biology. 2014.
- 52. Hartenstein V. Thicker than blood: conserved mechanisms in Drosophila and vertebrate hematopoiesis. Dev Cell. 2003;5: 673–690. pmid:14602069
- 53. Franc NC, Dimarcq JL, Lagueux M, Hoffmann J, Ezekowitz RA. Croquemort, a novel Drosophila hemocyte/macrophage receptor that recognizes apoptotic cells. Immunity. 1996;4: 431–443. pmid:8630729
- 54. Kroeger PT Jr., Tokusumi T, Schulz RA. Transcriptional regulation of eater gene expression in Drosophila blood cells.—PubMed—NCBI. Genesis. 2011;50: 41–49. pmid:21809435
- 55. Sinenko SA, Mathey-Prevot B. Increased expression of Drosophila tetraspanin, Tsp68C, suppresses the abnormal proliferation of ytr-deficient and Ras/Raf-activated hemocytes.—PubMed—NCBI. Oncogene. 2004;23: 9120–9128. pmid:15480416
- 56. Horn L, Leips J, Starz-Gaiano M. Phagocytic ability declines with age in adult Drosophila hemocytes. Aging Cell. 2014.
- 57. Mackenzie DK, Bussière LF, Tinsley MC. Senescence of the cellular immune response in Drosophila melanogaster. Exp Gerontol. 2011;46: 853–859. pmid:21798332
- 58. Lebestky T, Chang T, Hartenstein V, Banerjee U. Specification of Drosophila hematopoietic lineage by conserved transcription factors. Science. 2000;288: 146–149. Available: http://eutils.ncbi.nlm.nih.gov/entrez/eutils/elink.fcgi?dbfrom=pubmed&id=10753120&retmode=ref&cmd=prlinks pmid:10753120
- 59. Leitão AB, Sucena E. Drosophila sessile hemocyte clusters are true hematopoietic tissues that regulate larval blood cell differentiation. elife. eLife Sciences Publications Limited; 2015;4: e06166.
- 60. Bretscher AJ, Honti V, Binggeli O, Burri O, Poidevin M, Kurucz E, et al. The Nimrod transmembrane receptor Eater is required for hemocyte attachment to the sessile compartment in Drosophila melanogaster. Biol Open. The Company of Biologists Ltd; 2015;4: 355–363. pmid:25681394
- 61. Hedengren M, Asling B, Dushay MS, Ando I, Ekengren S, Wihlborg M, et al. Relish, a central factor in the control of humoral but not cellular immunity in Drosophila. Molecular Cell. 1999;4: 827–837. Available: http://eutils.ncbi.nlm.nih.gov/entrez/eutils/elink.fcgi?dbfrom=pubmed&id=10619029&retmode=ref&cmd=prlinks pmid:10619029
- 62. Atilano ML, Yates J, Glittenberg M, Filipe SR, Ligoxygakis P. Wall Teichoic Acids of Staphylococcus aureus Limit Recognition by the Drosophila Peptidoglycan Recognition Protein-SA to Promote Pathogenicity. PLoS Pathog. Public Library of Science; 2011;7: e1002421. pmid:22144903
- 63. Chung Y- SA, Kocks C. Recognition of pathogenic microbes by the Drosophila phagocytic pattern recognition receptor eater. Journal of Biological Chemistry. 2011.
- 64. Ayres JS, Schneider DS. Tolerance of Infections. Annu Rev Immunol. 2011.
- 65. Medzhitov R, Schneider DS, Soares MP. Disease tolerance as a defense strategy. Science. 2012;335: 936–941. pmid:22363001
- 66. Ayres JS, Schneider DS. A signaling protease required for melanization in Drosophila affects resistance and tolerance of infections. Promislow D, editor. PLoS Biol. 2008;6: 2764–2773. pmid:19071960
- 67. Louie A, Song KH, Hotson A, Thomas Tate A, Schneider DS. How Many Parameters Does It Take to Describe Disease Tolerance? PLoS Biol. 2016;14: e1002435. pmid:27088212
- 68. Silva EA, Burden J, Franc NC. In vivo and in vitro methods for studying apoptotic cell engulfment in Drosophila. Meth Enzymol. Elsevier; 2008;446: 39–59. pmid:18603115
- 69. Chakrabarti S, Dudzic JP, Li X, Collas EJ, Boquete J-P, Lemaitre B. Remote Control of Intestinal Stem Cell Activity by Haemocytes in Drosophila. PLoS Genet. Public Library of Science; 2016;12: e1006089. pmid:27231872
- 70. Woodcock KJ, Kierdorf K, Pouchelon CA, Vivancos V, Dionne MS, Geissmann F. Macrophage-derived upd3 cytokine causes impaired glucose homeostasis and reduced lifespan in Drosophila fed a lipid-rich diet. Immunity. 2015;42: 133–144. pmid:25601202
- 71. Zerofsky M, Harel E, Silverman N, Tatar M. Aging of the innate immune response in Drosophila melanogaster. Aging Cell. Blackwell Science Ltd; 2005;4: 103–108. pmid:15771614
- 72. Buchon N, Broderick NA, Chakrabarti S, Lemaitre B. Invasive and indigenous microbiota impact intestinal stem cell activity through multiple pathways in Drosophila. Genes Dev. Cold Spring Harbor Lab; 2009;23: 2333–2344. pmid:19797770
- 73. Broderick NA, Buchon N, Lemaitre B. Microbiota-induced changes in drosophila melanogaster host gene expression and gut morphology. MBio. American Society for Microbiology; 2014;5: e01117–14.
- 74. Wong CNA, Ng P, Douglas AE. Low-diversity bacterial community in the gut of the fruitfly Drosophila melanogaster. Environ Microbiol. 2011;13: 1889–1900. pmid:21631690
- 75. Arshad Ayyaz HJ. Intestinal inflammation and stem cell homeostasis in aging Drosophila melanogaster. Front Cell Infect Microbiol. Frontiers Media SA; 2013;3: 98. pmid:24380076
- 76. Rera M, Clark RI, Walker DW. Intestinal barrier dysfunction links metabolic and inflammatory markers of aging to death in Drosophila. Proceedings of the National Academy of Sciences. 2012;109: 21528–21533.
- 77. Clark RI, Salazar A, Yamada R, Fitz-Gibbon S, Morselli M, Alcaraz J, et al. Distinct Shifts in Microbiota Composition during Drosophila Aging Impair Intestinal Function and Drive Mortality.—PubMed—NCBI. Cell Rep. 2015;12: 1656–1667. pmid:26321641
- 78. Weavers H, Evans IR, Martin P, Wood W. Corpse Engulfment Generates a Molecular Memory that Primes the Macrophage Inflammatory Response. Cell. 2016;165: 1658–1671. pmid:27212238
- 79. Sampson CJ, Williams MJ. Real-Time Analysis of Drosophila Post-Embryonic Haemocyte Behaviour. Skoulakis EMC, editor. PLoS ONE. 2012;7: e28783. pmid:22242151
- 80. Regan JC, Brandão AS, Leitão AB, Mantas Dias AR, Sucena E, Jacinto A, et al. Steroid hormone signaling is essential to regulate innate immune cells and fight bacterial infection in Drosophila. PLoS Pathog. Public Library of Science; 2013;9: e1003720. pmid:24204269
- 81. Arefin B, Kucerova L, Krautz R, Kranenburg H, Parvin F, Theopold U. Apoptosis in Hemocytes Induces a Shift in Effector Mechanisms in the Drosophila Immune System and Leads to a Pro-Inflammatory State. PLoS ONE. Public Library of Science; 2015;10: e0136593. pmid:26322507
- 82. Takashima S, Younossi-Hartenstein A, Ortiz PA, Hartenstein V. A novel tissue in an established model system: the Drosophila pupal midgut. Development Genes and Evolution. 2011;221: 69–81. pmid:21556856
- 83. Schneider DS, Ayres JS. Two ways to survive infection: what resistance and tolerance can teach us about treating infectious diseases. Nat Rev Immunol. 2008;8: 889–895. pmid:18927577
- 84. Kurant E, Axelrod S, Leaman D, Gaul U. Six-microns-under acts upstream of Draper in the glial phagocytosis of apoptotic neurons. Cell. 2008;133: 498–509. pmid:18455990
- 85. Yu X, Lu N, Zhou Z. Phagocytic receptor CED-1 initiates a signaling pathway for degrading engulfed apoptotic cells. PLoS Biol. Public Library of Science; 2008;6: e61. pmid:18351800
- 86. Haine ER, Moret Y, Siva-Jothy MT, Rolff J. Antimicrobial defense and persistent infection in insects. Science. 2008;322: 1257–1259. pmid:19023083
- 87. Guo L, Karpac J, Tran SL, Jasper H. PGRP-SC2 Promotes Gut Immune Homeostasis to Limit Commensal Dysbiosis and Extend Lifespan. Cell. 2014;156: 109–122. pmid:24439372
- 88. Li H, Qi Y, Jasper H. Preventing Age-Related Decline of Gut Compartmentalization Limits Microbiota Dysbiosis and Extends Lifespan. Cell Host Microbe. 2016;19: 240–253. pmid:26867182
- 89. Biteau B, Hochmuth CE, Jasper H. JNK activity in somatic stem cells causes loss of tissue homeostasis in the aging Drosophila gut. Cell Stem Cell. 2008;3: 442–455. pmid:18940735
- 90. Ayyaz A, Li H, Jasper H. Haemocytes control stem cell activity in the Drosophila intestine. Nat Cell Biol. 2015;17: 736–748. pmid:26005834
- 91. Michel T, Reichhart JM, Hoffmann JA, Royet J. Drosophila Toll is activated by Gram-positive bacteria through a circulating peptidoglycan recognition protein. Nature. 2001;414: 756–759. pmid:11742401