Drosophila as a model for antiviral immunity

Abstract

The fruit fly Drosophila melanogaster has been successfully used to study numerous biological processes including immune response. Flies are naturally infected with more than twenty RNA viruses making it a valid model organism to study host-pathogen interactions during viral infections. The Drosophila antiviral immunity includes RNA interference, activation of the JAK/STAT and other signaling cascades and other mechanisms such as autophagy and interactions with other microorganisms. Here we review Drosophila as an immunological research model as well as recent advances in the field of Drosophila antiviral immunity.

Key Words: Antiviral, Drosophila, Genetics, Immune signaling, Infection, JAK/STAT, RNA interference, Virus

DROSOPHILA AS A RESEARCH MODEL

The fruitfly Drosophila melanogaster has been an important animal model in laboratory research since Thomas Hunt Morgan started using it to study chromosomes at the beginning of the 20th century. Drosophila is easy and inexpensive to rear in the laboratory, produces numerous progeny and has a short (about 10 d) generation time. As an invertebrate, Drosophila is considered an ethically acceptable animal model. Drosophila is an extremely useful tool for studying various biological processes, since many Drosophila external features such as compound eyes, wing veins and bristles can be affected by mutations and are easily visualized with a microscope[1].

The Drosophila genome is compact with low redundancy; single mutants are likely to reveal phenotypes of interest in contrast to mammals, whose genomes are more redundant and single mutation often does not produce a clear phenotype. Moreover, an analysis of the Drosophila euchromatin sequence revealed a high degree of similarity between flies and mammals[2-4] indicating a high degree of conservation of a very large number of biological processes. Moreover, 75% of known human genetic disease genes have homologues in the fly[5]. An online database named Homophila[5,6], which allows searching for human disease gene homologues in flies and vice versa, is available at http://superfly.ucsd.edu/homophila/.

The major advantage of flies is the simplicity and scale for genetic analysis, which has been undertaken for a century with a large number of sophisticated genetic tools[7]. One of the most important tools in Drosophila, successfully used for decades for new gene discovery, is the possibility of carrying out genetic screens for mutations that affect chosen biological processes. The traditional forward genetic screen (Figure 1A) involves mutagenesis, which is most commonly carried out with ethyl methane sulphonate or X-rays[1]. The phenotype of interest is screened in the mutated population followed by mapping and identifying the gene(s) causing the phenotype[1] (Figure 1A). In addition, mutations caused by P-element insertions[8] may also be screened; the commonest approach is to screen existing P-element insertion collections from the Berkeley Genome project[9].

Figure 1 Methods for the systematic study of Drosophila gene function. A: Forward genetic screen. Male flies subjected to ethyl methane sulphonate (EMS)-induced mutagenesis are crossed to virgin females that carry a balancer for the chromosome (indicated as B) to be screened. In the F1 progeny, each male inherits a mutagenized chromosome with a different spectrum of mutations. Individual F1 males are backcrossed to balancer stock. F2 male and F2 female carrying the same mutagenized chromosome are crossed with each other. 25% of the F3 are homozygous for the mutagenized chromosome. These are screened for interesting phenotype(s) followed by mapping and identification of corresponding genes; B: In vitro RNA interference (RNAi) screen. Drosophila cells are grown in multiwell-plates and incubated with known double-stranded RNA (dsRNA) to knock-down a specific gene. Thereafter, RNAi-treated cells are subjected to selected experimental procedures (e.g. viral infection). Interesting phenotype(s) can be immediately linked to the corresponding gene; C: Reverse genetic screen. Flies carrying dsRNA insertion for a particular gene driven by UAS promoter is crossed to a tissue-specific driver fly line. In the F1 progeny, the expressed GAL4 bind to UAS to drive the expression of the dsRNA to silence the expression of the selected gene. Flies are then tested in selected experimental conditions (for example viral infection).

In addition to traditional screens, many molecular genetic techniques have been developed in the last few decades that allow e.g. germ-line transformation[10], homologous recombination[11], and RNA interference (RNAi)[12,13]. The discovery of RNAi in Drosophila[14], and the completion of the Drosophila genome sequence[2] boosted a large number of large-scale in vitro RNAi screenings[15-20] (Figure 1B) including microarray-based screens[21,22] in Drosophila macrophage-like S2 cells[23,24]. Recently, libraries of transgenic Drosophila based on the binary GAL4/UAS system[25], have been constructed, making large-scale in vivo RNAi screening studies feasible e.g.[26-28] (Figure 1C). The first generation of RNAi fly lines included the genome-wide VDRC lines[29] and lines from the NIG-FLY (National Institutes of Genetics, Japan). The second generation of RNAi lines from the TRiP (Transgenic RNAi project) collection at Harvard Medical School[30] and from the VDRC phiC31 RNAi library[31] contain improvements: for example, the insertion of the RNAi construct is site-specific and the efficacy of the RNAi phenotypes is improved[30,31]. It is likely that the availability of these GAL4/UAS RNAi-based libraries will spur publications in the field of genome-wide in vivo screens in the immediate future.

USING DROSOPHILA AS A MODEL TO STUDY IMMUNE RESPONSE

In the defense against pathogenic microorganisms, Drosophila relies on innate immunity including epithelial and systemic responses generating antimicrobial peptides (AMPs), a phenoloxidase reaction resulting in melanin deposition, and a cellular response leading to the encapsulation and phagocytosis of intruding microbes[32-34]. Several signaling pathways implicated in Drosophila immunity have been identified, such as Imd, Toll, JNK, JAK/STAT and RNAi pathways. Upon microbial invasion, corresponding pathway(s) are activated specifically or in combination, to mount anti-microbial responses. The evolutionarily conserved innate immunity between Drosophila and humans makes Drosophila a valuable model for deciphering the mechanisms underlying human immunity[35,36]. This is exemplified by the discovery of the Drosophila Toll receptor[37], which stimulated the identification of the human Toll-like receptors[38]. In addition, the discovery of Drosophila peptidoglycan recognition proteins PGRPs[39] and the immune functions of PGRP-SA[40] and PGRP-LC[15,41,42] prompted the identification of a human family of PGRPs[43].

In Drosophila innate immune signaling, it is well established that the Imd and Toll pathways respond to bacterial and fungal infections via production of AMPs[33]. AMPs are mostly cationic, small molecules with an activity range directed against a variety of microorganisms[34]. Drosophila JAK/STAT signaling is required to control immune and stress responses[34,44,45]. After septic injury, activation of the JAK/STAT pathway leads to the expression of a number of genes including Turandot stress genes in fat body[46-48]. The Drosophila cellular response involves Drosophila professional phagocytes i.e. plasmatocytes, which phagocytose pathogens and locally secrete extracellular matrix components, AMPs, clotting factors and signaling molecules[49]. Another cellular response mechanism at the larval stages is the encapsulation of microbes that are too large for phagocytosis; a specialized group of hemocytes called lamellocytes is responsible for this[33,50].

In comparison to bacteria and fungal pathogens, viral evoked immune response in Drosophila is less known. It is apparent that both local and systemic immune responses are involved in virus clearance. A detailed description of Drosophila viruses has recently been reviewed by Huszar et al[51]. In this article we will review the interactions of viruses with Drosophila at the molecular level.

DROSOPHILA ANTIVIRAL RESPONSE

Being the most abundant infectious agents, viruses cause diseases which represent a constant threat and cause significant mortality worldwide. They evolve rapidly to adapt to the changing environment of host cells and are a great challenge to their host as well as to the development of efficient therapies and vaccines. By 2005, more than 25 distinct Drosophila viruses had been identified and were all RNA viruses[33,51]. A large proportion of all flies are infected with viruses via horizontal transmission between any two individuals, and infection through vertical transmission from parent to offspring is also common in Drosophila[33]. The innate antiviral immunity in Drosophila has recently been reviewed by Sabin et al[52]. Drosophila can also be used as a model to study human pathogenic viruses, such as the human immunodeficiency virus 1 (HIV-1). It has been demonstrated that the HIV-1 gene Vpu in Drosophila fat body cells inhibits the activity of the Toll pathway[53]. In addition, the HIV-1 gene Nef in the wing disc inhibits the activity of the Imd pathway[54]. Bearing in mind the crucial contribution of the innate immune response both to fighting HIV infection and activating the adaptive immune response, it appears advantageous for HIV to produce proteins that interfere with innate immunity pathways[53]. Therefore, using a genetically tractable model, such as Drosophila, is very useful in investigating these important evolutionarily conserved processes.

RNAi as a defense mechanism against viral infection

Initially described in plants[55,56], RNAi was later found in Caenorhabditis elegans with double-stranded RNA (dsRNA) as its initiating factor in silencing effects, and was named RNAi[57]. RNAi is an ancient, cell-intrinsic immune mechanism for the control of RNA viruses in plants and insects[58] as well as DNA viruses in mammals[59]. Three RNAi pathways have been identified in Drosophila[60], namely the small interfering RNA (siRNA) pathway, the micro-RNA (miRNA) pathway and the Piwi-interacting RNA (piRNA) pathway. The siRNA pathway utilizes AGO2 and Dicer2 and is activated by dsRNA. The miRNA pathway involves AGO1 and Dicer1 and regulates gene expression, particularly during development. The piRNA pathway involves three AGO proteins, is particularly active in the germline and seems to function in transposon silencing and epigenetic regulation[61,62]. In Drosophila, the exogenous part of the siRNA pathway is mainly responsible for the antiviral defense (Figure 2), whereas the endogenous siRNA pathway is involved in the regulation of transposons and transcripts[63].

Figure 2 Local (cellular) and systemic anti-viral response in Drosophila. In local anti-viral response, the viral particle infects host cell via endocytosis or is taken up by host receptor yet to be identified. Upon infection, viral RNA is processed to dsRNA[63], which evokes RNAi pathway response. The activation of the RNAi pathway, specifically small interfering RNA (siRNA) pathway, leads to the RNA-induced silencing complex (RISC)/siRNA-mediated degradation of viral RNA. It appears that the viral dsRNA can be taken up by un-infected neighboring cells to spread protective RNAi. The other systemic anti-viral mechanism is mediated by the JAK/STAT pathway. The infected cells produce signals that activate the JAK/STAT pathway by currently unknown mechanisms and lead to production of antiviral factors.

The RNAi mediated antiviral response in Drosophila has been studied with evolutionarily diverse viruses including Drosophila X virus (DXV)[64], Drosophila C virus (DCV)[65], Cricket paralysis virus (CrPV)[65,66] and Flock house virus (FHV)[66,67]. Both single-stranded and dsRNA viruses can infect Drosophila, but to date, no Drosophila DNA viruses have been identified. Both RNA virus types produce dsRNA as a replication intermediate, and this dsRNA activates the host RNAi pathway. RNAi is currently considered the major antiviral immune defense mechanism in Drosophila and the above-mentioned studies point to the significant role of Dicer2, AGO2 and Ars2 in this process[64-68]. In addition, Zambon et al[64] indicated that piwi, vasa intronic gene, aubergine, armitage, Rm62, and r2d2 also have vital roles in anti-DXV response. It is worth noting that a low level of infection, which can be cleared, or persistent infection of Drosophila by Nora virus[69,70] are not affected by RNAi machinery, JAK/STAT pathway or Toll pathway[71]. Detailed RNAi in the antiviral innate immune defense has recently been reviewed by Ulvila et al[72].

One virulence mechanism for viruses is suppression of the host RNAi pathway. DCV encodes a suppressor of RNAi, DCV-1A, that binds to long dsRNA and inhibits Dicer2-mediated processing of dsRNA into siRNA, but does not bind to siRNAs nor disrupt the miRNA pathway[65]. On the other hand, successful infection and killing of Drosophila by FHV strictly depends on the expression of the viral protein B2[67], which binds to dsRNA regardless of the length and inhibits cleavage of dsRNA by Dicer2 as well as incorporation of FHV siRNAs into the RNA-induced silencing complex[73]. The N-terminal fragment of 140 amino acids of the CrPV-A was identified as a viral suppressor of RNAi (VSR) for CrPV[66].

Previously, it was believed that RNA silencing-mediated antiviral response is systemic in plants, Caenorhabditis elegans and fungi but not in Drosophila[58,74]. However, this view has been challenged. Saleh et al[75] showed that antiviral immunity against DCV and Sindbis virus (SINV) in Drosophila requires systemic RNAi spread through the endocytosis pathway, which had been identified previously[76,77]. The endocytic pathway is also used in viral entry. Flies heterozygous for mutations in components of the endocytic pathway were protected from DCV-induced lethality in vivo[78]. The virus is suggested to bind to its cell surface receptor, which is captured in clathrin-coated vesicles and trafficked through the endocytic pathway. When virus escapes from the endocytic pathway upon cell death or lysis, dsRNA is released and may be taken up by uninfected cells thus inducing the systemic RNAi-mediated antiviral mechanisms[75] (Figure 2).

Signaling cascades regulating antiviral immunity

JAK/STAT pathway in antiviral response: Known earlier as a regulator of multiple aspects of development, the JAK/STAT pathway was shown to partly contribute to the antiviral response in Drosophila, because DCV triggered STAT (a signal transducer and activator of transcription) DNA-binding activity in whole fly nuclear extract[79]. Moreover, loss-of-function mutation of the only Drosophila Jak kinase, Hopscotch resulted in increased viral replication upon DCV infection, and using a dominant negative form of the receptor Domeless, decreased the expression of a virus-specific target gene virus-induced RNA 1 (vir-1)[79]. It is suggested that DCV infection triggers the induction of an unidentified cytokine, possibly a member of the Upd family, which activates the Domeless receptor and Hopscotch, leading to activation of STAT and induction of a set of genes[80] (Figure 2). This set of induced genes appears to be dependent on the specific virus, at least in part, since the RNA level of a Drosophila JAK/STAT target gene Turandot M, although induced in FHV infection, was not induced upon DCV infection[79]. The role of JAK/STAT in the control of viral infection is further supported by a recent study[81] showing that flies heterozygous for a stat mutation displayed increased SINV replication. Together, these studies indicate an evolutionarily conserved involvement of the JAK/STAT pathway in antiviral response[79,82].

Toll and Imd pathways in antiviral response: Previously it was suggested that the fly nuclear factor (NF)-κB pathways, namely Toll and Imd pathways, have no role in antiviral defense, e.g.[80], but this view has been revised. The Toll pathway was shown to control the survival of and be required for the inhibition of the viral replication of DXV-infected flies[64,83]. Also, a role for the Toll pathway in the control of Dengue virus infection in mosquito has been shown[84]. Recently, two research groups demonstrated the involvement of the Imd pathway in antiviral defenses in Drosophila. Avadhanula et al[81] measured the SINV RNA change in transgenic flies expressing SINV replicon RNA and compared levels from flies with knock-down of different Imd pathway genes to that of wild-type flies. Their results showed that the Imd but not the Toll pathway plays a role in anti-SINV response. Later, Costa et al[85] demonstrated a role for hemocytes and the Imd pathway in combating against CrPV infection. Both studies showed increased viral loads in flies with knock-down of Imd pathway genes, which suggests a role for the Imd pathway at least in SINV and CrPV infections in Drosophila[81,85]. However, in all these studies it was agreed that AMPs are not involved, at least not directly, in the control of viral infection in Drosophila[79,81,83,85].

The importance of NF-κB signaling for antiviral defense was further highlighted by identification of viral proteins that suppress IMD and Toll pathway-mediated immune response. In Drosophila S2 cells, H4 and N5, two of the proteins encoded by Microplitis demolitor bracovirus (McBV), homologous to inhibitor κB (IκB) from insects and mammals, reduce the expression of Drosomycin and Attacin reporter constructs[86]. Moreover, H4 and N5 are able to bind to Dif and Relish and inhibit binding of Dif and Relish to κB sites in the promoters of the Drosomycin and Cecropin A1 genes[86]. Mimicking IκB factors thus appears to be one way for the virus to evade the insect immune system.

Other Drosophila antiviral defense: Induction of host antiviral responses via mechanisms independent of the RNAi, JAK/STAT, Toll and Imd pathways have also been found in Drosophila. One such inducible gene is Vago[80], identified in the screen for DCV-induced genes[79], which negatively controls DCV load in the fat body in a tissue-autonomous way. Vago induction by DCV requires Dicer2, but not AGO2 or r2d2[68,79]. This therefore, suggests a novel role for Dicer2 in sensing viral dsRNA in virus-infected cells, leading to induction of antiviral genes, in addition to its involvement in RNAi[80]. As well as DCV, Vago was also induced by SINV, but not FHV[80].

One other important factor in antiviral immunity appears to be dSUR, a homolog of the mammalian SUR2 protein, which was identified to be important for mammalian antiviral immunity in mapping the mice mayday phenotype[87]. SUR2 is a subunit of the ATP-sensitive potassium channel complex expressed in smooth muscle cells of coronary arteries. RNAi knock-down of Drosophila SUR (dSUR, CG5772) led to increased susceptibility to infection with FHV, but not for DCV, the bacterial species Enterococcus faecalis and Enterobacter cloacae, nor the fungus Beauveria bassiana[87]. Moreover, flies fed with tolbutamide, a member of the sulfonylurea class of channel blockers, showed increased susceptibility to FHV infection, indicating that the heart is one of the tissues maintaining homeostasis during the innate immune response[87].

Autophagy, an intrinsic mechanism that can degrade cytoplasmic components, was found to play a direct antiviral role against the mammalian viral pathogen, vesicular stomatitis virus (VSV), in Drosophila[88,89]. Autophagy decreased viral replication in a cell-autonomous manner and its activation does not require viral replication. Repression of autophagy led to increased viral replication and pathogenesis in cells and animals. Furthermore, the antiviral response of autophagy was controlled by the phosphatidylinositol 3-kinase (PI3K)-Akt-signaling pathway, which normally mediates autophagy in response to nutrient availability. On the other hand, flies depleted of Atg18, a component of the PI3K-Akt-signaling pathway, had a normal life span and were not more sensitive to infection with DCV, suggesting that the autophagy-mediated antiviral response is protective against VSV but not DCV infection. It is likely that the surface glycoprotein, VSV-G, which serves as the pathogen-associated molecular pattern initiated this cell-autonomous response. However, the intracellular sensor for VSV-G has not been identified in Drosophila.

ASPECTS AFFECTING ANTIVIRAL RESPONSES OF DROSOPHILA

Microarray studies have examined changes in gene expression in Drosophila infected with DCV[79,90], DXV[83], SIGMAV[91,92] and DCV/FHV/SINV[80]. It appears that Drosophila antiviral immune response is virus specific. Carpenter et al[92] also reported that there was a difference in gene regulation between male and female SIGMAV infected flies. On the other hand, although both studies from Tsai et al[91] and Carpenter et al[92] used SIGMAV, little overlap of regulated genes was observed. Therefore, it is apparent that the genetic background of Drosophila affects host-parasite interactions[93-95]. Genetic background may also explain, in part, the variable Nora virus titers observed between fly stocks in a recent study[70].

The genetic composition of alleles encoding a polymorphic gene ref(2)P appear important in Drosophila SIGMAV infection and have been intensively studied. Ref(2)P is evolutionarily conserved and its mammalian orthologue, p62, serves as an adaptor for the activation of the NF-κB pathway by aPKC (atypical protein kinase)[96-98]. A similar role for Ref(2)P has been proposed in Drosophila, where Ref(2)P/DaPKC activates the NF-κB proteins Dorsal and DIF in the Toll pathway[99]. In nature, there are ref(2)p alleles that are either permissive or restrictive for SIGMAV multiplication[51,100,101]. The mechanism by which the restrictive ref(2)p allele interacts with and blocks SIGMAV replication is not known, but it was shown that all allele encoded proteins can interact with SIGMAV P protein and share conformation-dependent epitopes with the N protein[102].

The interactions of Drosophila with other microorganisms also play a role in antiviral response. Drosophila infected with Wolbachia, a common bacterium in natural Drosophila populations, increases resistance to RNA viruses, namely DCV, Nora, FHV and CrPV, but not to a DNA virus insect iridescent virus 6[103,104]. Also, preinfection of flies with another virus, namely FHV, led to an increased induction of Vir-1 and more modest upregulation of Vago by DCV infection[80]. Antiviral silencing against FHV in S2 cells induced by FR1gfp, a construct defective in suppressing RNA silencing-based antiviral response in Drosophila cells, as described previously[105] was suppressed by CrPV superinfection[66].

CONCLUSION

In conclusion, the antiviral response in Drosophila is mainly mediated by the siRNA pathway, which functions against a broad range of viruses. Viruses have evolved ways to circumvent RNAi to survive in their host cells, exemplified by viral genome encoded VSRs. The JAK/STAT, and possibly Toll and Imd pathways are involved in the control of viral growth for specific virus(es) by mechanisms yet to be clarified. The viral infection could induce the expression of genes which are involved in decreasing the viral infection, and Dicer2 is needed for gene induction. Other aspects, including e.g. autophagy, also appear to play a role in the control of viral infection in Drosophila. Of note, the genetic background of Drosophila as well as infection with other microorganisms has effects on the antiviral responses.

Despite the progress in recent years in understanding fly antiviral immunity, there are still a lot of unanswered questions. Currently, dsRNA-triggered RNAi is considered the major antiviral immune response in Drosophila, and it has been intensively studied. What is the viral trigger for other pathways in antiviral response, such as the JAK/STAT pathway? What is the ligand of the virally-induced JAK/STAT pathway, and what do the activated genes do in combat against viruses at the molecular level? Do the fly NF-κB pathways mount a response specifically against some viruses, and what are the factors induced, since in all the studies it was agreed that AMPs are not directly involved[79,81,83,85]? What is the route of viral spreading in Drosophila[95]? What is the role of the newly identified Nora virus, which is present in large numbers in the intestine of infected flies[69-71]? In addition, it will be interesting to investigate the interactions between bacteria and viruses in the light of the new findings about Wolbachia conferring resistance against viral infections in Drosophila[103,104]. Further work is likely to uncover additional mechanisms, factors and pathways in the Drosophila antiviral immune response as well as insights into host-pathogen interactions.