Infection in Insects Coagulation, and Prolong Survival Upon Enhance

Home , Galleria mellonella

Host-Derived Extracellular Nucleic Acids Enhance Innate Immune Responses, Induce Coagulation, and Prolong Survival upon Infection in Insects This information is current as of September 28, 2021. Boran Altincicek, Sabine Stötzel, Malgorzata Wygrecka, Klaus T. Preissner and Andreas Vilcinskas J Immunol 2008; 181:2705-2712; ; doi: 10.4049/jimmunol.181.4.2705 http://www.jimmunol.org/content/181/4/2705 Downloaded from

References This article cites 58 articles, 17 of which you can access for free at:

http://www.jimmunol.org/content/181/4/2705.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication by guest on September 28, 2021

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2008 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Host-Derived Extracellular Nucleic Acids Enhance Innate Immune Responses, Induce Coagulation, and Prolong Survival upon Infection in Insects

Boran Altincicek,* Sabine Sto¨tzel,* Malgorzata Wygrecka,† Klaus T. Preissner,† and Andreas Vilcinskas1*

Extracellular nucleic acids play important roles in human immunity and hemostasis by inducing IFN production, entrapping pathogens in neutrophil extracellular traps, and providing procoagulant cofactor templates for induced contact activation during mammalian blood clotting. In this study, we investigated the functions of extracellular RNA and DNA in innate immunity and hemolymph coagulation in insects using the greater wax moth Galleria mellonella a reliable model host for many insect and human pathogens. We determined that coinjection of purified Galleria-derived nucleic acids with heat-killed bacteria synergistically Downloaded from increases systemic expression of antimicrobial peptides and leads to the depletion of immune-competent hemocytes indicating cellular immune stimulation. These activities were abolished when nucleic acids had been degraded by nucleic acid hydrolyzing enzymes prior to injection. Furthermore, we found that nucleic acids induce insect hemolymph coagulation in a similar way as LPS. Proteomic analyses revealed specific RNA-binding proteins in the hemolymph, including apolipoproteins, as potential mediators of the immune response and hemolymph clotting. Microscopic ex vivo analyses of Galleria hemolymph clotting reactions revealed that oenocytoids (5–10% of total hemocytes) represent a source of endogenously derived extracellular nucleic acids. http://www.jimmunol.org/ Finally, using the entomopathogenic bacterium Photorhabdus luminescens as an infective agent and Galleria caterpillars as hosts, we demonstrated that injection of purified nucleic acids along with P. luminescens significantly prolongs survival of infected larvae. Our results lend some credit to our hypothesis that host-derived nucleic acids have independently been co-opted in innate immunity of both mammals and insects, but exert comparable roles in entrapping pathogens and enhancing innate immune responses. The Journal of Immunology, 2008, 181: 2705–2712.

ucleic acids, DNA and RNA, are universal in all living (NETs)2 (9, 10). The importance of this defense mechanism has, in organisms and are polyanionic macromolecules that parallel, been highlighted by the observation that bacterial patho- by guest on September 28, 2021 N carry the whole genetic information of every cell. They gens expressing nucleic acid hydrolyzing enzymes are capable of are usually intracellular, but upon wounding or injury, nucleic escaping entrapment by these NETs during infection (11–14). acids are released from damaged tissue to the extracellular envi- We have recently demonstrated that extracellular nucleic acids ronment where they exert unexpected functions. In vertebrates, derived from damaged or necrotic cells particularly under patho- host- or pathogen-derived nucleic acids have been recognized as logical conditions or severe tissue damage induce blood clotting in immunostimulatory factors because their extracellular presence in- mammals (15). They may provide physiologically relevant tem- duces IFN production by fibroblasts or immune cells (1–3). TLR- plates for the factors XII/XI-induced contact activation/amplifica- dependent (e.g., TLR3, TLR7, and TLR9) and TLR-independent tion during human hemostasis (16). Invertebrates lack acquired (e.g., RIG-I/MDA5 and DNA-dependent activator of IFN-regula- immunity, which evolved during vertebrate evolution, and rely on tory factors) signaling pathways have been described that are ac- innate immunity to control pathogens. Therefore, we investigated tivated by nucleic acids from both pathogens and hosts (4–7). In potential functions of extracellular nucleic acids in innate immu- addition, human granular immune cells (neutrophils) were discov- nity and hemolymph coagulation in a particularly suited insects ered to generate and to weave tangled webs of extracellular fibers model, the greater wax moth Galleria mellonella. G. mellonella composed of nucleic acids and proteins with antimicrobial capac- caterpillars have widely been used as convenient and reliable ities when stimulated by cytokines or bacterial immune elicitors model hosts for many insect and human pathogens (17–25). like LPS (8). Because these webs are capable of entrapping mi- Among the advantages provided by the Galleria model, it is of crobes, they have been named neutrophil extracellular traps particular importance to note that the caterpillars can be reared at mammalian physiological temperatures (around 37°C) to which human pathogens are adapted and which are essential for synthesis of many microbial virulence/pathogenicity factors. Furthermore, *Institute of Phytopathology and Applied Zoology and †Institute of Biochemistry, Galleria represents a classical model for the investigation on insect Justus-Liebig-University of Giessen, Heinrich-Buff-Ring 26-32, Giessen, Germany hemolymph clotting (26–29), and it has recently been used to elu- Received for publication April 25, 2008. Accepted for publication June 10, 2008. cidate the mechanisms mediating sensing of infection by danger The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 Address correspondence and reprint requests to Dr. Andreas Vilcinskas, Institute 2 Abbreviations used in this paper: NET, neutrophil extracellular trap; w/v, weigh to of Phytopathology and Applied Zoology, Justus-Liebig-University of Giessen, volume ratio; PVDF, polyvinylidene difluoride membrane. Heinrich-Buff-Ring 26-32, D-35392 Giessen, Germany. E-mail address: [email protected] Copyright © 2008 by The American Association of Immunologists, Inc. 0022-1767/08/$2.00 www.jimmunol.org 2706 EXTRACELLULAR NUCLEIC ACIDS IN INSECT INNATE IMMUNITY signals (30–33), similar to signals proposed in the danger model of Determination of humoral and cellular immune responses mammalian immunity (34). Humoral antimicrobial activity was measured by an inhibition zone assay The insect innate immune system recognizes microbe and dam- using a LPS-defective, streptomycin- and ampicillin-resistant mutant of E. age associated pattern molecules by germline encoded receptors coli K12 strain D31 (38, 39). Using the inhibition zone assay with Esch- (e.g., Toll receptors and peptidoglycan recognition proteins) which erichia coli bacteria and gentamicin as an external standard, induction lev- engage potent defense reactions such as hemolymph coagulation, els of antimicrobial peptides within the hemolymph 24 h post treatment were quantified which is a degree of the systemic immune responses in cellular phagocytosis, nodulation, encapsulation, and phenoloxi- insects (32). Cellular immune stimulation in vivo was determined by count- dase activation leading to melanization (35). These reactions are ing circulating hemocytes in the hemolymph 4 h upon treatment of larvae often divided into cellular and humoral immune responses, al- because immune stimulation correlates with the switch from nonadherent, though it is somewhat arbitrary, as many humoral factors affect resting hemocytes to activated, adherent cells (40). hemocyte function and hemocytes are an important source of many humoral molecules. In Lepidoptera, granulocytes and plasmato- Two-dimensional gel electrophoresis of hemolymph proteins cytes are the hemocyte types responsible for phagocytosis of mi- Cell-free hemolymph samples were isolated by bleeding injected larvae 2 h crobes and become adherent upon stimulation (35, 36). The other post injection, or untreated larvae, into plastic tubes with traces of phenyl- hemocytes are nonadhesive spherule cells, oenocytoids, and pro- thiourea to prevent phenoloxidase activation followed by a centrifugation step of the hemolymph at 2,000 ϫ g for 5 min. The supernatant (cell-free hemocytes (35, 36). Spherule cells have been suggested to trans- hemolymph) was precipitated by the addition of 3 volumes of 100% ace- port cuticular components, while oenocytoids are fragile cells con- tone and 0.4 volumes of 100% trichloroacetic acid followed by incubation taing cytoplasmic phenoloxidase precursors that likely play a role at Ϫ20°C for 1 h. After centrifugation at 20,000 ϫ g for 10 min, the pellet in melanization of hemolymph. Prohemocytes are hypothesized to was washed three times with 100% acetone and resolved under agitation in Downloaded from be stem cells that can differentiate into one or more of the afore- 8 M urea at 22°C for 16 h. Protein concentrations were determined using the Micro BC assay kit (Uptima). Two-dimensional gel electrophoresis, mentioned hemocyte types (35, 36). in-gel digestion, and protein identification by matrix-assisted laser desorp- In insects, killing of invading pathogens is achieved similar to tion ionization–time of flight mass spectrometry were performed as de- mammals by enzymes (e.g., lysozymes), by reactive oxygen spe- scribed (32). cies, and by antimicrobial peptides (e.g., defensins) (35). These

defense reactions rely on both the cellular and humoral immune Northwestern blot analysis http://www.jimmunol.org/ responses. In this study, we report that survival of G. mellonella Cell-free hemolymph samples from LPS-challenged larvae were isolated larvae infected with the entomopathogen Photorhabdus lumine- by bleeding larvae 24 h post injection with 100 ␮g LPS (Sigma-Aldrich) scens can be prolonged when host-derived extracellular nucleic similarly as described (32). Obtained hemolymph proteins were separated acids are simultaneously injected in their hemocoels. This protec- by Tris-Tricine-SDS-PAGE and blotted on to a polyvinylidene diﬂuoride (PVDF) membrane according to the manufacturer’s instructions (Amer- tive role was putatively mediated by the ability of host-derived sham Biosciences). Apolipoprotein III (ApoLp-III) was partially puriﬁed as nucleic acids to synergistically enhance both induced expression of described (41) prior to SDS-PAGE and blotting on the PVDF membrane. antimicrobial peptides and activation of immune cells. Further- After precipitation of most other hemolymph proteins by heat treatment at more, we discovered that addition of extracellular RNA or DNA to 90°C for 30 min, the supernatant was obtained by centrifugation at

ϫ by guest on September 28, 2021 hemolymph samples resulted in the formation of net-like coagu- 20,000 g for 30 min. The supernatant is mainly composed of heat-stable arylphorin protein (ϳ80 kDa) and apolipoprotein III (41). Membranes were lation fibers that efficiently entrap bacteria. Consequently, we ad- incubated in buffer A containing 10 mM Tris-HCl (pH 7.5), 50 mM NaCl, dressed the question whether RNA is actively released by hemo- 1 mM EDTA, and 1ϫ Denhardt’s solution (0.02% weight-to-volume ratio cytes upon immune stimulation, and from which could hemocyte (w/v) Ficoll, 0.02% w/v,polyvinylpyrrolidone, and 0.02% w/v BSA) at 4°C type. Proteomic analyses revealed that corresponding RNA-bind- overnight. Subsequently, the membranes were incubated1hat25°C in 30 ml buffer A containing 3 ␮g biotin-labeled total RNA. Labeling of total ing proteins, particularly apolipoproteins, are potentially involved RNA was conducted using EZ-Link Psoralen-PEO3-Biotin (Pierce) ac- in nucleic acid-mediated defense reactions. cording to the manufacturer’s instructions. After washing overnight with buffer A, the membranes were subjected to chemiluminescent detection using x-ray films (Amersham Biosciences) and the SuperSignal West Pico Complete Biotinylated Protein Detection Kit (Pierce) according to the in- Materials and Methods structions of the manufacturers. Insect rearing and manipulation Galleria mellonella larvae were reared on an artificial diet (22% maize Coagulation assay meal, 22% wheat germ, 11% dry yeast, 17.5% bee wax, 11% honey, and 11% glycerin) at 32°C in darkness. Photorhabdus luminescens strain DSM Larvae were punctured with a sterile needle, and obtained hemolymph ϳ ␮ ␮ 12205 was purchased from DSMZ. Last instar larvae, each weighing be- samples ( 10 l) were dropped directly on test solutions (5 l) on the tween 250 and 350 mg, were used for injection experiments. Ten micro- microscope slides. Coagulation was monitored with an Axioplan 2 micro- Ј liters of sample volume per larva were injected dorsolaterally into the scope (Zeiss). DAPI (4 ,6-diamidino-2-phenylindole dihydrochloride) hemocoel using 1-ml disposable syringes and 0.4 ϫ 20 mm needles (Sigma-Aldrich) and SYTOX Green were used for staining of nucleic acids mounted on a microapplicator. Viable or heat-inactivated bacteria were according to the instructions of the manufacturers. FITC-labeled bacteria were prepared by coupling FITC to amine groups of bacteria using fluo- washed three times in sterile PBS (20 mM Na3PO4 buffer, 100 mM NaCl (pH 7.0)) and subsequently mixed with DNA or RNA solutions immedi- resceinisothiocyanate similar as proposed by the manufacturers’ instruc- ately prior application. tions (KMF Laborchemie Handels-GmbH).

Results Preparation of RNA and DNA from G. mellonella larvae Extracellular nucleic acids enhance systemic and cellular innate RNA was extracted from whole larvae using the TriReagent isolation re- immune responses and hemolymph coagulation in insects agent (Molecular Research Centre) and Qiagen RNeasy kit (Qiagen) according to the instructions of the manufacturers. DNA was extracted from To analyze potential functions of host-derived nucleic acids in whole larvae using Qiagen DNeasy kit. Integrity of RNA and DNA was Galleria, we tested their immunostimulatory activities. Using the confirmed on ethidium bromide agarose gels, and quantities were deter- inhibition zone assay with Escherichia coli bacteria and gentami- mined spectrophotometrically (37). Hydrolysis of DNA (1 mg/ml) was performed with 1 ␮g/ml DNase I (Qiagen) 16 h at 37°C and of RNA (1 cin as an external standard, induction levels of antimicrobial pep- mg/ml) with 1 ␮g/ml RNase A (Qiagen) for4hat60°C. Hydrolysis ef- tides within the hemolymph 24 h post treatment were quantified, ficiencies were confirmed by ethidium bromide agarose gels (37). which correlate to the activation level of the systemic immune The Journal of Immunology 2707

FIGURE 1. Extracellular presence of RNA synergistically induces humoral and cellular immune responses against bacteria. A, Induced systemic expression levels of antimicrobial peptides in the hemolymph is shown as gentamicin equivalents (U/ml were calculated using a calibration curve with Downloaded from gentamicin). RNA injection did not result in significant systemic immune responses leading to increased antimicrobial peptide levels in larval hemolymph (gentamicin aquivalents 2.5 Ϯ 0.5 U/ml) when compared with PBS injection (gentamicin aquivalents 2.4 Ϯ 0.4 U/ml). Larvae that had been injected with 10 ␮g RNA along with heat-killed P. luminescens bacteria (gentamicin aquivalents 13 Ϯ 0.8 U/ml) have a synergistically increased immune response when compared with PBS plus heat-killed P. luminescens bacteria injected animals (gentamicin equivalents 7.9 Ϯ 0.9 U/ml). This effect was abolished when RNase was coinjected along with RNA and heat-killed bacteria (gentamicin aquivalents 8.5 Ϯ 0.8 U/ml). B, In vivo hemocyte stimulation was quantified by the depletion rate of circulating nonactivated hemocytes. A synergistically increased cellular immune stimulation was observed when 10 ␮g RNA were coinjected with heat-killed P. luminescens bacteria (0.78 Ϯ 0.18 ϫ 103 hemocytes/␮l) when compared with injected bacteria alone (1.26 Ϯ 0.34 ϫ 103 http://www.jimmunol.org/ hemocytes/␮l). This activation of the cellular immunity was abolished when the RNA had been hydrolyzed by RNase A prior use (1.1 Ϯ 0.28 ϫ 103 hemocytes/␮l). RNA injection alone resulted in no significant cellular immune activation (3.9 Ϯ 0.7 ϫ 103 hemocytes/␮l) when compared with PBS injection (4 Ϯ 0.57 ϫ 103 hemocytes/␮l). Results represent mean values of at least three independent determinations Ϯ SD. Statistically significant .(p Ͻ 0.005 ,ءءء ;p Ͻ 0.05 ,ء) differences were determined using Student‘s t test and are indicated response in insects (32). Injection of purified RNA (mainly com- tion (hemocyte count of 3.9 Ϯ 0.7 ϫ 103 cells/␮l) when compared posed of rRNA and mRNA with an average chain length of ap- with PBS injection (hemocyte count of 4 Ϯ 0.57 ϫ 103 cells/␮l) proximate 1000–4000 nt) alone, up to 20 ␮g per animal, did not (Fig. 1B). by guest on September 28, 2021 result in significant systemic immune responses (gentamicin Moreover, high amounts of RNA (more than 50 ␮g per larvae) aquivalents 2.5 Ϯ 0.5 U/ml) when compared with PBS injected resulted in activation of the prophenoloxidase cascade in vivo, a animals (gentamicin aquivalents 2.4 Ϯ 0.4 U/ml) (Fig. 1A). How- serine-proteinase cascade, leading to melanisation of hemolymph ever, a synergistically increased immune response was observed in in the larvae (data not shown). However, the biological signifi- larvae that had been injected with 10 ␮g RNA along with heat- cance of this prophenoloxidase cascade activation at high concen- killed P. luminescens bacteria (gentamicin aquivalents 13 Ϯ 0.8 trations of nucleic acids is not clear and will be investigated in U/ml) when compared with larvae that had been injected with PBS future studies. In general, similar immunostimulatory activities along with heat-killed P. luminescens bacteria injected animals were determined when DNA was used instead of RNA (data not (gentamicin aquivalents 7.9 Ϯ 0.9 U/ml) (Fig. 1A). The difference was statistically significant ( p Ͻ 0.001) as calculated with a Stu- dent’s t test. Furthermore, this synergistic effect was abolished when RNase was coinjected along with RNA and heat-killed bacteria (gentamicin aquivalents 8.5 Ϯ 0.8 U/ml). We obtained similar synergistic effects when we examined hemocyte stimulation. To quantify cellular immune stimulation, we counted circulating hemocytes in the hemolymph of animals 4 h after injection with different solutions. Depletion of circulating nonactivated hemocytes indicate cellular stimulation in vivo because immune activated hemocytes become highly adhesive and attach to each other forming multicellular aggregates and to inter- nal organs of the larvae (40). When 10 ␮g RNA were coinjected with heat-killed P. luminescens bacteria, an increased cellular innate immune response was determined (hemocyte count of 0.78 Ϯ 0.18 ϫ 103 cells/␮l) when compared with animals injected with FIGURE 2. The presence of nucleic acids induces the formation of net- like fibrillar coagulation strands in the insect hemolymph. A, Larvae were bacteria along with PBS (hemocyte count of 1.26 Ϯ 0.34 ϫ 103 ␮ ϭ pierced with a sterile needle, and hemolymph samples collected from the cells/ l) (Fig. 1B). The difference was statistically significant ( p wound were directly applied on microscope slides containing 5 ␮l water 0.02) as determined with a Student’s t test. This effect was abol- (A) or solutions of 1 mg/ml LPS (B), RNA (C), or DNA (D). Within ished when the RNA had been hydrolyzed by RNase treatment minutes the formation of net-like structures were detected (indicated by 3 prior to use (hemocyte count of 1.1 Ϯ 0.28 ϫ 10 cells/␮l). RNA arrows) that bind to surrounding hemocytes. Differential interference con- injection alone resulted in no significant cellular immune activa- trast (Nomarski) image, ϫ 100. Scale bars, 50 ␮m. 2708 EXTRACELLULAR NUCLEIC ACIDS IN INSECT INNATE IMMUNITY

In a next step, we investigated the procoagulant potential of nucleic acids. Larvae were pierced with a sterile needle, and hemolymph samples (ϳ10–15 ␮l) collected from the wound were directly applied on microscope slides containing 5 ␮l of test solutions with LPS, water, or nucleic acids. Within minutes, the formation of net-like structures were induced by RNA and DNA, respectively, in the hemolymph (Fig. 2, C and D). This nucleic acid-dependent process appeared to be similar to the LPS-induced clotting reactions that we determined in our analysis (Fig. 2B). In contrast, coagulation induction was reduced when nucleic acids had been hydrolyzed prior to analysis because less coagulation strands could be observed under the microscope within 5–10 min (data not shown).

Identification of apolipoproteins as potential mediators of extracellular RNA-mediated immune responses FIGURE 3. Interaction of Galleria hemolymph proteins with labeled To identify host proteins that directly interact with and thereby RNA. Galleria hemolymph proteins obtained from LPS-challenged ani- mediate the enhancing effects of extracellular nucleic acids on hu- mals were separated by Tris-tricine-SDS-PAGE and blotted on to a PVDF Downloaded from moral and cellular immune responses a combination of proteomic membrane. Proteins were stained with Coomassie Blue, photographed (lane 1), completely destained with 70% (v/v) ethanol, and equilibrated in approaches was used. First, Galleria hemolymph samples obtained binding buffer. The membrane was incubated with biotin-labeled total from LPS-challenged larvae were analyzed by Northwestern blot RNA. Labeled RNA bound to the interacting proteins on the membrane analysis. Five protein bands within the hemolymph were detected was detected by chemoluminescent reaction and x-ray films, resulting in that interact with labeled RNA (Fig. 3, lane 2). The RNA binding dark bands (lane 2). The protein interacting with RNA at ϳ17 kDa cor- protein with ϳ17 kDa may correspond to apolipoprotein-III responds to apolipoprotein-III (ApoLp-III). The potential SDS-stable com- (ApoLp-III), whose role in ␤ 1,3-glucan pattern recognition and http://www.jimmunol.org/ plex of ApoLp-II and ApoLp-III (ϳ85 kDa) is indicated by an asterisk. The cellular encapsulation in G. mellonella larvae has recently been bands corresponding to 50 and 70 kDa proteins are not known. Molecular established by the group of Norman Ratcliffe (42). To confirm its mass standards are shown in kDa. identity ApoLp-III was partially purified by precipitation of most other hemolymph proteins by heat treatment. The supernatant still contained binding activity of ApoLp-III for RNA as demonstrated shown). The DNA mediated effects were also abolished when by Northwestern blot analysis (data not shown). In addition, it has DNA had been degraded by DNA hydrolyzing enzyme prior to recently been demonstrated that ApoLp-II forms partially SDS- injection as similarly shown above for RNA. stable complexes with ApoLp-III in the Galleria hemolymph with by guest on September 28, 2021

FIGURE 4. Proteomic analysis of G. mellonella hemolymph proteins in the presence of nucleic acids. A, Incubation of 1 to 100 diluted Galleria cell-free hemolymph in 120 mM NaCl and 20 mM Na-phosphate buffer (pH 7.0) for 16 h at 25°C results in the formation of a small amount of SDS-stable aggregates (lane 1) which are strongly enhanced in the presence of 20 ␮g/ml LPS (lane 2) similar as also reported by others (31). The induced aggregate formation correlates well with decrease of ApoLp-I (ϳ 250 kDa) and ApoLp-II (ϳ 85 kDa) bands (indicated by arrows). Comparable aggregate formation is detected in the presence of 20 ␮g/ml RNA (lane 3) and DNA (lane 5), respectively. The DNA/RNA-induced aggregate formation is abolished in the presence of enzymes that hydrolyze nucleic acids (lanes 4 and 6). The presence of reducing agent (2 mM 2-ME) resulted in complete inhibition of aggregate formation by nucleic acids and LPS (data not shown). B, Hemolymph protein samples (1 mg) from 30 untreated and 30 RNA injected larvae were loaded on 24-cm pH 3–11 NL-IEF strips followed by Tris-tricine-SDS-PAGE on a 15% gel, respectively. Four out of ϳ500 abundant hemolymph proteins with increased and one with reduced abundance after 2 h RNA-injection were identiﬁed. Spot 1 may correspond to ApoLp-II because ApoLp-II is an abundant hemolymph protein with an estimated molecular mass of 85 kDa. Molecular mass standards are indicated in kDa. The Journal of Immunology 2709 a molecular mass of 80–90 kDa (43). These stable complexes may correspond to the detected protein band at ϳ85 kDa in the North- western blot analysis (Fig. 3, lane 2). The identities of the proteins with an estimated molecular mass of 70 and 50 kDa, respectively, remained unclear. Because LPS has been shown to mediate the formation of ApoLp-I/II/III-containing lipoprotein aggregates in Galleria (44), nucleic acids were tested for similar activities in the hemolymph. In accordance with published information (44), 20 ␮g/ml LPS induced the formation of detergent-stable aggregates in cell-free diluted hemolymph which correlated well with the decrease of ApoLp-I and ApoLp-II protein bands (Fig. 4A, lanes 1 and 2). Comparable aggregate formation was found in the presence of 20 ␮g/ml RNA or DNA (Fig. 4A, lanes 3 and 5). As expected, presence of RNase or DNase, respectively, prevented enhanced aggregate formation (Fig. 4A, lane 4 and 6). Furthermore, the reducing agent 2-ME inhibited both LPS- and nucleic acid-mediated lipoprotein aggregate formation indicating that processes that are essential for defense reactions are activated in a similar way by an Downloaded from endogenous danger signal (nucleic acids) and an exogenous immune elicitor (LPS). In a subsequent step, RNA was injected into larvae to analyze changes of hemolymph proteins in vivo. As compared with control animals, the spectrum of ϳ500 most abundant hemolymph pro-

teins was analyzed by two-dimensional gel electrophorsis. In http://www.jimmunol.org/ agreement with data obtained in vitro, a dominant protein spot at ϳ80 kDa disappeared 2 h post RNA injection that may most probably correspond to ApoLp-II and, additionally, four other protein spots were found to be induced in the hemolymph (Fig. 4B). Un- fortunately, matrix-assisted laser desorption ionization–time of flight mass spectrometry analysis resulted in no positive identifi- cations because the corresponding sequences from these Galleria proteins were probably not yet in public databases. In addition, the two-dimensional protein map from G. mellonella larvae 24 h post by guest on September 28, 2021 RNA injection showed no significant differences to hemolymph proteins from untreated animals (data not shown) suggesting no induced or repressed expression of abundant hemolymph proteins FIGURE 5. Detection of intrinsic extracellular nucleic acids derived 24 h upon RNA injection in Galleria. from oenocytoids during insect hemolymph clotting. A, Microscopic examination of Galleria hemolymph samples containing live hemocytes re- Discovery of intrinsic extracellular nucleic acids in insect sulted in the observation that oenocytoids (Oc) rupture within 10–60 s hemolymph clotting reactions upon bleeding. B, The rupture of oenocytoids results in visible fibrillar strands that contain nucleic acids because they are positively stained by Using SYTOX Green, a selective dye that visualizes the extracel- SYTOX Green nucleic acid stain (indicated by an arrow). The “naked” lular presence of nucleic acids, we searched for intrinsic nucleic nucleus is intensively stained. The granulocyte (Gc) is unstained because acids during insect clotting. We performed an ex vivo analysis the cell is intact. C and D, Within 1–3 min the granulocytes degranulate without any permeabilization or fixation steps by directly applying substances that entrap FITC-labeled bacteria and results in fibrillar net-like hemolymph samples (10–15 ␮l) collected from the wound site of coagulation structures. E and F, By the hanging drop method (26) we found larvae on microscope slides containing 5 ␮l of test solutions with 3–5 min after bleeding that coagulation strands developed including he- appropriate stains or FITC-labeled bacteria. We found that Galle- mocytes and entrapped FITC-labeled bacteria. G, Within 5–15 min we observed multicell aggregates with enhanced activation of the phe- ria oenocytoids which constitute ϳ5–10% of total circulating he- noloxidase cascade resulting in melanisation (brownish color). H,In mocytes represent a source of endogenously derived extracellular these aggregates we detected intense DAPI staining of free cell nuclei nucleic acids in the hemolymph. Oenocytoids rupture after 10 to of ruptured oenocytoids and faint nuclei staining of intact cells. I,To 60 s upon immune stimulation by bleeding on microscope slides identify NETs from insect cells we incubated hemolymph samples for and are the primary source for extracellular nucleic acids as dem- 2–5 h in Schneider’s insect medium in the presence of FITC-labeled onstrated by SYTOX staining (Fig. 5, A and B). Within 1–3 min, bacteria and DAPI stain. Hemocytes coagulated with hemolymph pro- the granulocytes became degranulated most probably triggered by teins and bacteria but no cells were found producing NETs as known oenocytoid-derived factors including extracellular nucleic acids. from human neutrophils. J, Overlay of coagulon including DAPI Degranulation of granulocytes results in efficient microbial entrap- stained nuclei and FITC-labeled bacteria is shown. Aggregated and ment at their surface and at fibrillar structures appearing between fragmented nuclei of granulocytes (Gc) and plasmatocytes (Pc) (shown magnified in insets) indicate apoptotic processes after 3–4 h stimulation these cells (Fig. 5, C and D). Examination of the hemolymph by whereas “naked” nuclei of oenocytoids (Oc) are highly condensed dur- the hanging drop approach (26) led to the observation that within ing incubation period. Note that many bacteria are entrapped by gran- 3–5 min coagulation strands including hemocytes and bacteria ulocytes and only few are bound to plasmatocytes. Differential inter- were formed (Fig. 5, E and F). Additionally, within 5–15 min ference contrast (Nomarski) and fluorescence imaging; A–D and insets multicell aggregations with increased activation of the serine pro- in J, ϫ1.575; E–J, ϫ630. Scale bars, 10 ␮m. teinase cascade leading to melanisation were detected (Fig. 5G). 2710 EXTRACELLULAR NUCLEIC ACIDS IN INSECT INNATE IMMUNITY

FIGURE 6. Survival curve of G. mellonella larvae infected with P. luminescens. A, All Galleria larvae (n ϭ 10) that have been injected with P. luminescens (103 CFU/larvae) died after 24 h incubation at 32°C (E). Injection of 10 ␮g RNA along with P. luminescens signiﬁcantly prolonged survival of infected Gal- leria larvae up to 35 to 42 h (F). B, Prolonged survival of larvae following their treatment was only marginal when a 100-fold higher inoculum of P. luminescens (105 CFU/larvae) was injected along with RNA. There was no killing of caterpillars that received heat-killed bacterial cells of the same strain. The experiment was repeated at least three times, with similar results.

DAPI staining under used conditions resulted in strong fluores- antimicrobial defense is mediated by their capacity to synergisti- cence of “naked” nuclei of oenocytoids and of faint staining of cally induce humoral and cellular immune responses during infec- nuclei of other hemocytes which are covered by a surrounding cell tion and to induce net-like coagulation structures that entrap in- membrane (Fig. 5H). To identify potential NETs formed by insect vading microbes. Results obtained by two-dimensional gel Downloaded from cells, we incubated hemolymph samples (10 ␮l) mixed with electrophoresis and Northwestern-blot analyses identified apoli- Schneider’s insect medium (200 ␮l) (BioWhittaker) containing poproteins as potential components involved in these processes. FITC-labeled bacteria and DAPI stain on microscope slides in a The first defense response to wounding in insects is hemolymph humidity chamber for 2–5 h. Hemocytes coagulated with hemo- coagulation. This clotting reaction shares functional similarities lymph proteins and bacteria but no cells were found producing with vertebrate hemostasis: to seal wounds and to prevent life- NETs as known from human neutrophils (Fig. 5, I and J). threatening loss of blood (35, 45). In vertebrates, blood and lymph are confined to vessels, whereas in insects an open circulatory http://www.jimmunol.org/ Beneficial role of extracellular nucleic acids in the in vivo system provides access of hemolymph to other tissue cells. Yet, in immune defense of Galleria against Photorhabdus infection both systems, the contribution of circulating cells is of great im- To elucidate in vivo roles of extracellular RNA/DNA in insect portance for efficient clot formation to occur (26, 46). Upon immune defense, the influence of host-derived nucleic acids during wounding, circulating hemocytes in insects immediately switch from infection of Galleria larvae with Photorhabdus bacteria was ana- a resting, nonadherent state to activated cells that are highly adhesive lyzed. Survival times of larvae receiving 10 ␮g RNA in combina- like activated platelets in mammals. However, unlike platelets, insect tion with 103 CFU of Photorhabdus cells were significantly longer hemocytes contribute to microbial clearance by engulfment within

(50% mortality occurred at ϳ37 h) than those of controls that were multicellular aggregate formation, known as nodules with subsequent by guest on September 28, 2021 injected with bacteria alone (50% mortality at ϳ21 h) or with melanization. Although mechanisms that initiate these processes are bacteria plus RNA that had been hydrolyzed prior injection (50% hardly defined, we propose that extracellular nucleic acids serve as mortality at ϳ21 h) (Fig. 6A). A similar protective effect was ob- important, but as yet unrecognized, cofactors. served when DNA was coinjected along with bacteria resulting in In this study, we discovered that oenocytoids release nucleic a 50% mortality of larvae at ϳ36 h (data not shown). However, acids upon immune activation. Oenocytoids in Lepidoptera show when a 100-fold higher inoculum of bacteria was used (105 CFU similarities to the crystal cells in the fruit fly Drosophila melano- of Photorhabdus cells per larvae), the prolonged survival rate in gaster because both are large, regular in shape, contain phenoloxi- the presence of 10 ␮g RNA was only marginally reduced with a dases, and rupture upon immune activation (35, 36, 47). However, 50% mortality at ϳ18 h vs 50% mortality at ϳ16 h without RNA Drosophila has obviously no granulocytes and no apolipoprotein (Fig. 6B). A further control group injected with heat-killed bacteria III gene which is present in many other insects (48), suggesting resulted in 100% survival, indicating that injection injury itself did striking differences between different insect species. Because in- not cause mortality. sects occupy a wide range of ecological niches, comparative studies using Galleria and other insects as models should enhance our Discussion understanding of blood cell-dependent innate immune responses in Our study identifies extracellular nucleic acids (naturally released terms of conserved and derived molecular mechanisms. by damaged tissues and by activated oenocytoids) as a novel dan- In a previous study, we provided evidence that extracellular nucleic ger signal in defense reactions of insects to protect them against acids, in particular RNA, represent the long-sought natural “foreign infecting bacteria. The role of nucleic acids as alarm signals in surface” in the vertebrate system to induce blood clotting, particularly

FIGURE 7. Schematic overview of comparable roles of extracellular nucleic acids in insects and mammals. The Journal of Immunology 2711 under pathological conditions (15). In analogy, we show in this study Disclosures that the exposure of isolated hemolymph samples from G. mellonella The authors have no financial conflict of interest. larvae to extracellular RNA or DNA resulted in hemocyte activation and formation of net-like coagulation structures. Our results provide References evidence that extracellular nucleic acids derived from ruptured oeno- 1. Rotem, Z., R. A. Cox, and A. Isaacs. 1963. Inhibition of virus multiplication by cytoids or tissue damage trigger coagulation and other immune de- foreign nucleic acid. Nature 197: 564–566. 2. Jensen, K. E., A. L. Neal, R. E. Owens, and J. Warren. 1963. Interferon responses fense mechanisms in Galleria. of chick embryo fibroblasts to nucleic acids and related compounds. Nature 200: Insect coagulation nets share similarities in appearance and 433–434. function (entrapment of microbes) with the indicated vertebrate 3. Ishii, K. J., K. Suzuki, C. Coban, F. Takeshita, Y. Itoh, H. Matoba, L. D. Kohn, and D. M. Klinman. 2001. Genomic DNA released by dying cells induces the NETs (27, 49–52) but have more similarities to mammalian fibrin- maturation of APCs. J. Immunol. 167: 2602–2607. platelet matrix. In contrast to human NETs, the extracellular nu- 4. Akira, S., S. Uematsu, and O. Takeuchi. 2006. Pathogen recognition and innate cleic acids in Galleria exhibit only weak bacterial entrapment ca- immunity. Cell 124: 783–801. 5. Karin, M., T. Lawrence, and V. Nizet. 2006. Innate immunity gone awry: linking pacity when compared with the binding capacity of the matrix that microbial infections to chronic inflammation and cancer. Cell 124: 823–835. derives from granulocytes (Fig. 5, C and D). Furthermore, oeno- 6. Ishii, K. J., and S. Akira. 2007. Innate immune recognition of, and regulation by, DNA. Trends Immunol. 27: 525–532. cytoids release nucleic acids within seconds upon stimulation 7. Chi, H., and R. A. Flavell. 2007. Immunology: sensing the enemy within. Nature whereas vertebrate NET formation is a slow process. In general, 448: 423–424. 180 min upon activation, NET components (including nucleic ac- 8. Brinkmann, V., U. Reichard, C. Goosmann, B. Fauler, Y. Uhlemann, D. S. Weiss, Y. Weinrauch, and A. Zychlinsky. 2004. Neutrophil extracellular traps kill bac- ids and cellular proteins) mix freely in the neutrophils and 15–60 teria. Science 303: 1532–1535. min after that, NETs are released (9). Our results indicate that 9. Brinkmann, V., and A. Zychlinsky. 2007. Beneficial suicide: why neutrophils die Downloaded from extracellular nucleic acids are at least as important as microbial to make NETs. Nat. Rev. Microbiol. 5: 577–582. 10. Urban, C. F., U. Reichard, V. Brinkmann, and A. Zychlinsky. 2006. Neutrophil derived molecular patterns for inducing hemolymph coagulation extracellular traps capture and kill Candida albicans yeast and hyphal forms. and melanization in vivo and provide a novel link between the Cell. Microbiol. 8: 668–676. 11. Beiter, K., F. Wartha, B. Albiger, S. Normark, A. Zychlinsky, and vertebrate and insect system in terms of molecular principles that B. Henriques-Normark. 2006. An endonuclease allows Streptococcus pneu- lead to engagement of the innate immunity. moniae to escape from neutrophil extracellular traps. Curr. Biol. 16: 401–407. 12. Buchanan, J. T., A. J. Simpson, R. K. Aziz, G. Y. Liu, S. A. Kristian, M. Kotb, Consistent with our results, a recent study demonstrates that http://www.jimmunol.org/ J. Feramisco, and V. Nizet. 2006. DNase expression allows the pathogen group mammalian LL37, an antimicrobial peptide released during skin A Streptococcus to escape killing in neutrophil extracellular traps. Curr. Biol. 16: injury, converts self-DNA into a “danger signal” that potently ac- 396–400. tivates innate immune responses (53). On the other side, circulat- 13. Walker, M. J., A. Hollands, M. L. Sanderson-Smith, J. N. Cole, J. K. Kirk, A. Henningham, J. D. McArthur, K. Dinkla, R. K. Aziz, R. G. Kansal, et al. 2007. ing nucleic acids in plasma and serum are prognostic markers for DNase Sda1 provides selection pressure for a switch to invasive group A strep- stroke, infarct, or cancer patients (54, 55), and it was recently tococcal infection. Nat. Med. 13: 981–985. 14. Sumby, P., K. D. Barbian, D. J. Gardner, A. R. Whitney, D. M. Welty, shown that in DNase II-deficient mice, extracellular DNA escapes R. D. Long, J. R. Bailey, M. J. Parnell, N. P. Hoe, G. G. Adams, et al. 2005. degradation and may cause chronic polyarthritis resembling hu- Extracellular deoxyribonuclease made by group A Streptococcus assists patho- man rheumatoid arthritis (56). These observations along with our genesis by enhancing evasion of the innate immune response. Proc. Natl. Acad.

Sci. USA 102: 1679–1684. by guest on September 28, 2021 results from insects indicate that extracellular nucleic acids exhibit 15. Kannemeier, C., A. Shibamiya, F. Nakazawa, H. Trusheim, C. Ruppert, evolutionarily conserved protective roles in the first line of defense P. Markart, Y. Song, E. Tzima, E. Kennerknecht, M. Niepmann, et al. 2007. which may turn to harmful effects when their degradation is Extracellular RNA constitutes a natural procoagulant cofactor in blood coagulation. Proc. Natl. Acad. Sci. USA 104: 6388–6393. dysregulated. 16. Pedicord, D. L., D. Seiffert, and Y. Blat. 2007. Feedback activation of factor XI We found that apolipoproteins are potentially involved in nu- by thrombin does not occur in plasma. Proc. Natl. Acad. Sci. USA 104: 12855–12860. cleic acid-mediated hemocyte activation and coagulation in in- 17. Aperis, G., B. B. Fuchs, C. A. Anderson, J. E. Warner, S. B. Calderwood, and sects. Insect ApoLp-III is a multifunctional protein that is involved E. Mylonakis. 2007. Galleria mellonella as a model host to study infection by the in lipid transport but also functions as pattern recognition receptor Francisella tularensis live vaccine strain. Microbes Infect. 9: 729–734. ␤ 18. Mylonakis, E., R. Moreno, J. B. El Khoury, A. Idnurm, J. Heitman, by binding to LPS, lipoteichoic acid, or to fungal 1,3-glucan to S. B. Calderwood, F. M. Ausubel, and A. Diener. 2005. Galleria mellonella as a induce cellular and humoral immune responses (42). Interestingly, model system to study Cryptococcus neoformans pathogenesis. Infect. Immun. mammalian apolipoproteins are analogously involved in LPS bind- 73: 3842–3850. 19. Jander, G., L. G. Rahme, and F. M. Ausubel. 2000. Positive correlation between ing and other roles in innate immunity (57). Because insect virulence of Pseudomonas aeruginosa mutants in mice and insects. J. Bacteriol. ApoLp-III and ApoLp-I/II is homologous to human apolipoprotein 182: 3843–3845. 20. Miyata, S., M. Casey, D. W. Frank, F. M. Ausubel, and E. Drenkard. 2003. Use E (58) and human apolipoprotein B (59), respectively, our results of the Galleria mellonella caterpillar as a model host to study the role of the type favor these molecules to be involved in the early defense mecha- III secretion system in Pseudomonas aeruginosa pathogenesis. Infect. Immun. 71: nisms in vertebrates as well and provides a starting point for the 2404–2413. 21. Scully, L. R., and M. J. Bidochka. 2005. Serial passage of the opportunistic investigation of interrelationships of innate immunity and physio- pathogen Aspergillus flavus through an insect host yields decreased saprobic ca- logical processes such as lipid storage and use. pacity. Can. J. Microbiol. 51: 185–189. The present findings are consistent with the hypothesis that ex- 22. Bergin, D., L. Murphy, J. Keenan, M. Clynes, and K. Kavanagh. 2006. Pre- exposure to yeast protects larvae of Galleria mellonella from a subsequent lethal tracellular nucleic acids promote host survival by improving de- infection by Candida albicans and is mediated by the increased expression of fense mechanisms against pathogens at sites of tissue damage/in- antimicrobial peptides. Microbes Infect. 8: 2105–2112. 23. Park, S. Y., K. M. Kim, J. H. Lee, S. J. Seo, and I. H. Lee. 2007. Extracellular fection in both insects and mammals (Fig. 7). Our data confirm that gelatinase of Enterococcus faecalis destroys a defense system in insect hemo- Galleria is a powerful model organism for analyzing host defense lymph and human serum. Infect. Immun. 75: 1861–1869. reactions in general and the role of extracellular nucleic acids in 24. Fedhila, S., N. Daou, D. Lereclus, and C. Nielsen-LeRoux. 2006. Identification of Bacillus cereus internalin and other candidate virulence genes specifically in- innate immunity in particular. duced during oral infection in insects. Mol. Microbiol. 62: 339–355. 25. Morton, D. B., G. B. Dunphy, J. S. Chadwick. 1987. Reactions of hemocytes of immune and non-immune Galleria mellonella larvae to Proteus mirabilis. Dev. Comp. Immunol. 11: 47–55. Acknowledgments 26. Rowley, A. F., and N. A. Ratcliffe. 1976. The granular cells of Galleria mel- We thank Meike Fischer for excellent technical assistance, Katja Altinci- lonella during clotting and phagocytic reactions in vitro. Tissue Cell 8: 437–446. 27. Bidla, G., M. Lindgren, U. Theopold, and M. S. Dushay. 2005. Hemolymph cek for critical reading of the manuscript, and Monica Linder for MALDI- coagulation and phenoloxidase in Drosophila larvae. Dev. Comp. Immunol. 29: TOF-MS analysis. 669–679. 2712 EXTRACELLULAR NUCLEIC ACIDS IN INSECT INNATE IMMUNITY

28. Li, D., C. Scherfer, A. M. Korayem, Z. Zhao, O. Schmidt, and U. Theopold. 43. Halwani, A. E., D. F. Niven, and G. B. Dunphy. 2001. Apolipophorin-III in the 2002. Insect hemolymph clotting: evidence for interaction between the coagula- greater wax moth. Galleria mellonella (Lepidoptera: Pyralidae). Arch. Insect Bio- tion system and the prophenoloxidase activating cascade. Insect Biochem. Mol. chem. Physiol. 48: 135–143. Biol. 32: 919–928. 44. Ma, G., D. Hay, D. Li, S. Asgari, and O. Schmidt. 2006. Recognition and inac- 29. Korayem, A. M., T. Hauling, C. Lesch, M. Fabbri, M. Lindgren, O. Loseva, tivation of LPS by lipophorin particles. Dev. Comp. Immunol. 30: 619–626. O. Schmidt, M. S. Dushay, and U. Theopold. 2007. Evidence for an immune 45. Theopold, U., O. Schmidt, K. So¨derha¨ll, and M. S. Dushay. 2004. Coagulation in function of lepidopteran silk proteins. Biochem. Biophys. Res. Commun. 352: arthropods: defense, wound closure and healing. Trends Immunol. 25: 289–294. 317–322. 46. Ruggeri, Z. M. 2007. Von Willebrand factor: looking back and looking forward. 30. Altincicek, B., and A. Vilcinskas. 2006. Metamorphosis and collagen-IV-frag- Thromb. Haemost. 98: 55–62. ments stimulate innate immune response in the greater wax moth, Galleria mel- 47. Wood, W., and A. Jacinto. 2007. Drosophila melanogaster embryonic haemo- lonella. Dev. Comp. Immunol. 30: 1108–1118. cytes: masters of multitasking. Nat. Rev. Mol. Cell Biol. 8: 542–551. 31. Griesch, J., M. Wedde, and A. Vilcinskas. 2000. Recognition and regulation of 48. Smith, A. F., L. M. Owen, L. M. Strobel, H. Chen, M. R. Kanost, E. Hanneman, metalloproteinase activity in the haemolymph of Galleria mellonella: a new path- and M. A. Wells. 1994. Exchangeable apolipoproteins of insects share a common way mediating induction of humoral immune responses. Insect. Biochem. Mol. structural motif. J. Lipid Res. 35: 1976–1984. Biol. 30: 461–472. 49. Karlsson, C., A. M. Korayem, C. Scherfer, O. Loseva, M. S. Dushay, and 32. Altincicek, B., M. Linder, D. Linder, K. T. Preissner, and A. Vilcinskas. 2007. U. Theopold. 2004. Proteomic analysis of the Drosophila larval hemolymph clot. Microbial metalloproteinases mediate sensing of invading pathogens and activate J. Biol. Chem. 279: 52033–52041. innate immune responses in the lepidopteran model host Galleria mellonella. 50. Agianian, B., C. Lesch, O. Loseva, and M. S. Dushay. 2007. Preliminary char- Infect. Immun. 75: 175–183. acterization of hemolymph coagulation in Anopheles gambiae larvae. Dev. Comp. 33. Altincicek, B., and A. Vilcinskas. 2008. Identiﬁcation of a lepidopteran matrix Immunol. 31: 879–888. metalloproteinase with dual roles in metamorphosis and innate immunity. Dev. 51. Scherfer, C., C. Karlsson, O. Loseva, G. Bidla, A. Goto, J. Havemann, Comp. Immunol. 32: 400–409. M. S. Dushay, and U. Theopold. 2004. Isolation and characterization of hemo- 34. Matzinger, P. 2002. The danger model: a renewed sense of self. Science 296: lymph clotting factors in Drosophila melanogaster by a pullout method. Curr. 301–305. Biol. 14: 625–629. 52. Haine, E. R., J. Rolff, and M. T. Siva-Jothy. 2007. Functional consequences of

35. Jiravanichpaisal, P., B. L. Lee, and K. So¨derha¨ll. 2006. Cell-mediated immunity Downloaded from blood clotting in insects. Dev. Comp. Immunol. 31: 456–464. in arthropods: hematopoiesis, coagulation, melanization and opsonization. Immu- 53. Lande, R., J. Gregorio, V. Facchinetti, B. Chatterjee, Y. H. Wang, B. Homey, nobiology 211: 213–236. W. Cao, Y. H. Wang, B. Su, F. O. Nestle, T. Zal, et al. 2007. Plasmacytoid Insect Science 36. Strand, M. R. 2008. The insect cellular immune response. 15: dendritic cells sense self-DNA coupled with antimicrobial peptide. Nature 449: 1–14. 564–569. 37. Ausubel, F. M., R. Brent, R. Kingston, D. Moore, J. Seidman, J. Smith, and 54. Vlassov, V. V., P. P. Laktionov, and E. Y. Rykova. 2007. Extracellular nucleic K. Struhl. 1987. Current Protocols in Molecular Biology, New York: John Wiley acids. Bioessays 29: 654–667. and Sons. 55. Fleischhacker, M., and B. Schmidt. 2007. Circulating nucleic acids (CNAs) and 38. Boman, H. G., I. Nilsson-Faye, K. Paul, and T. J. Rasmusen. 1974. Insect im- cancer: a survey. Biochim. Biophys. Acta 1775: 181–232. http://www.jimmunol.org/ munity I: characteristics of an inducible cell-free antibacterial reaction in hemo- 56. Kawane, K., M. Ohtani, K. Miwa, T. Kizawa, Y. Kanbara, Y. Yoshioka, lymph of Samia cynthia. Infect. Immun. 10: 136–145. H. Yoshikawa, and S. Nagata. 2006. Chronic polyarthritis caused by mammalian 39. Altincicek, B., and A. Vilcinskas. 2007. Analysis of the immune-inducible tran- DNA that escapes from degradation in macrophages. Nature 443: 998–1002. scriptome from microbial stress resistant, rat-tailed maggots of the drone ﬂy 57. Feingold, K. R., J. L. Funk, A. H. Moser, J. K. Shigenaga, J. H. Rapp, and Eristalis tenax. BMC Genomics 8: 326. C. Grunfeld. 1995. Role for circulating lipoproteins in protection from endotoxin 40. Ratcliffe, N. A., and S. J. Gagen. 1977. Studies on the in vivo cellular reactions toxicity. Infect. Immun. 63: 2041–2046. of insects: an ultrastructural analysis of nodule formation in Galleria mellonella. 58. Cole, K. D., G. P. Fernando-Warnakulasuriya, M. S. Boguski, M. Freeman, Tissue Cell 9: 73–85. J. I. Gordon, W. A. Clark, J. H. Law, and M. A. Wells. 1987. Primary structure 41. Wiesner, A., S. Losen, P. Kopa´cek, C. Weise, and P. Go¨tz. 1997. Isolated apo- and comparative sequence analysis of an insect apolipoprotein: apolipophorin-III lipophorin III from Galleria mellonella stimulates the immune reactions of this from Manduca sexta. J. Biol. Chem. 262: 11794–11800. insect. J. Insect Physiol. 43: 383–391. 59. Avarre, J. C., E. Lubzens, and P. J. Babin. 2007. Apolipocrustacein, formerly

42. Whitten, M. M., I. F. Tew, B. L. Lee, and N. A. Ratcliffe. 2004. A novel role for vitellogenin, is the major egg yolk precursor protein in decapod crustaceans and by guest on September 28, 2021 an insect apolipoprotein (apolipophorin III) in ␤-1,3-glucan pattern recognition is homologous to insect apolipophorin II/I and vertebrate apolipoprotein B. BMC and cellular encapsulation reactions. J. Immunol. 172: 2177–2185. Evol. Biol. 7: 3.