• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Aug 9, 2005; 102(32): 11426–11431.
Published online Aug 1, 2005. doi:  10.1073/pnas.0505240102
PMCID: PMC1183600

Inhibitor κB-like proteins from a polydnavirus inhibit NF-κB activation and suppress the insect immune response


Complex signaling pathways regulate the innate immune system of insects, with NF-κB transcription factors playing a central role in the activation of antimicrobial peptides and other immune genes. Although numerous studies have characterized the immune responses of insects to pathogens, comparatively little is known about the counterstrategies pathogens have evolved to circumvent host defenses. Among the most potent immunosuppressive pathogens of insects are polydnaviruses that are symbiotically associated with parasitoid wasps. Here, we report that the Microplitis demolitor bracovirus encodes a family of genes with homology to inhibitor κB (IκB) proteins from insects and mammals. Functional analysis of two of these genes, H4 and N5, were conducted in Drosophila S2 cells. Recombinant H4 and N5 greatly reduced the expression of drosomycin and attacin reporter constructs, which are under NF-κB regulation through the Toll and Imd pathways. Coimmunoprecipitation experiments indicated that H4 and N5 bound to the Rel proteins Dif and Relish, and N5 also weakly bound to Dorsal. H4 and N5 also inhibited binding of Dif and Relish to κB sites in the promoters of the drosomycin and cecropin A1 genes. Collectively, these results indicate that H4 and N5 function as IκBs and, circumstantially, suggest that other IκB-like gene family members are involved in the suppression of the insect immune system.

Keywords: immunity, virus, virulence, parasite

Key regulators of the insect and mammalian immune systems are NF-κB transcription factors that are homo- or heterodimers of proteins belonging to the Rel family (15). In Drosophila, three Rel proteins (Dif, Dorsal, and Relish) form NF-κBs that regulate the expression of antimicrobial peptide (AMP) genes and numerous other molecules with immune functions (6, 7). The Toll pathway activates NF-κBs comprised of Dif or Dorsal, whereas the Imd pathway activates NF-κBs comprised of Relish (3, 79). Outside of Drosophila and, to a more limited degree, mosquitoes (10, 11), little is known about NF-κB signaling in other insects. However, it is widely assumed that NF-κBs play a conserved role in immune regulation because of similarities in the inducible expression of effector molecules, like AMPs, across many invertebrates (3, 4). Parallels have also been noted between the Toll and Imd pathways of Drosophila and immune signaling pathways that activate NF-κBs in mammals (3, 7).

In resting cells, NF-κBs exist in an inactive state through noncovalent association with inhibitor κB (IκB) proteins. These complexes reside primarily in the cytoplasm, although some also enter the nucleus (12). Mammals encode seven IκB family members (IκBα, IκBβ, IκBε, IκBγ, Bcl-3, and the precursor Rel proteins p100 and p105), whereas Drosophila encodes two, called Cactus and Relish (Rel-110) (7, 12). The structural motif shared among all IκBs is an ankyrin-repeat domain (ARD) that is required for NF-κB binding (1315). Cactus complexes with Dif and/or Dorsal, but, upon immune challenge, is phosphorylated and degraded (16), allowing the NF-κB dimer to translocate to the nucleus, bind to κB promoter elements, and interact with target genes (3, 7). Rel-110 is a compound protein with an N-terminal Rel homology domain and C-terminal ARD, similar to mammalian p100 and p105 (17, 18). Immune challenge results in cleavage of Rel-110, such that dimers containing the N-terminal fragment (Rel-68) translocate to the nucleus, whereas the C-terminal fragment remains in the cytoplasm (18).

Equally important to understanding how the insect immune system is regulated is understanding how pathogens evade host immune defenses (1921). Among the most potent immunosuppressive pathogens of insects are viruses in the family Polydnaviridae. Polydnaviruses (PDVs) are divided into ichnoviruses and bracoviruses (BV), on the basis of their association with parasitoid wasps in the families Ichneumonidae and Braconidae (22). PDVs are the only viruses with segmented DNA genomes, and all appear to share a similar life cycle. PDVs persist as stably integrated proviruses in the genomes of wasps and replicate in the ovaries of females. When a wasp lays an egg into its insect host, she also injects a quantity of virus that infects host immune cells and other tissues. PDVs do not replicate in the wasp's host, but viral gene expression suppresses the host's immune response, allowing the parasitoid to successfully develop (22). In some PDV systems, viral infection protects only the wasp, whereas other PDVs have global immunosuppressive effects on hosts. These effects include disrupting the ability of host immune cells to encapsulate or phagocytize foreign targets and altering humoral immune responses, such as inducible expression of antimicrobial peptides (summarized in refs. 22 and 23).

An example of a PDV with broad immunosuppressive activity is Microplitis demolitor bracovirus (MdBV) which is carried by the wasp Microplitis demolitor (2227). The MdBV genome consists of 16 double-stranded, circular DNAs that range in size from 6.1 kb (segment A) to 34.3 kb (segment O) (22). A characteristic of MdBV and other PDV genomes is that several genes have undergone duplication and divergence to produce families of related gene variants (22, 28). Functional studies indicate that one MdBV gene family (glc genes) inhibits host encapsulation and phagocytosis (29, 30). Glc proteins also have similar effects in other insects, including Drosophila, suggesting that these viral proteins interact with conserved molecules required for capsule formation and phagocytosis (30). Complete sequencing of the MdBV genome has identified other potentially immunosuppressive factors, including a gene family with homology to IκBs. Here, we report the outcome of functional studies on two MdBV IκB-like genes, named H4 and N5. Our results indicate that these viral proteins function as IκBs and are potent inhibitors of the insect immune system.

Materials and Methods

Sequence Analysis. MdBV IκB-like gene sequences were identified from viral genomic clones by using the blast program and lasergene software (DNASTAR, Madison, WI). The ARDs of the IκB-like gene family members were aligned to selected insect and mammalian IκBs by using the clustalw method with gap-creation penalties of 10.00 and gap-extension penalties of 0.20. Signal peptides were sought by using the SignalP server (www.cbs.dtu.dk/services/SignalP). N- and C-flanking motifs were identified by using the PFAM database (www.sanger.ac.uk/Software/Pfam).

Expression and Reporter Constructs. H4 and N5 expression plasmids were constructed in the vector pIZT/V5-His (Invitrogen), which uses the OpIE2 promoter from the Orgyia pseudotsugata baculovirus for constitutive expression of the gene of interest and encodes a Zeocin GFP fusion gene under control of the OPIE1 promoter. The H4 and N5 genes were PCR amplified by using MdBV genomic DNA as a template. H4 primers were 5′-CATCTAGAGGTCACGATGGTGCGATACTA-3′ (forward) and 5′-GACCGCGGATACA TTTTTTTCGATCTTTCTTC-3′ (reverse), whereas N5 primers were 5′-CATTCTAGAACGATGGAGCGTGCAGATAATTC-3′ (forward) and 5′-GTCCGCGGATACTTGAACAATTGCATCATATAAG-3′ (reverse). XbaI and SacII sites (underlined) were incorporated into the forward and reverse primers for directional cloning (start and previous stop codons in bold). As recommended by the manufacturer, we modified the sequence around the start codons in the forward primers (bold) in each gene to create a Kozak sequence for proper initiation of translation. We also mutated the stop codons to TAT in the reverse primers (bold) to express the gene products in frame with the vector-encoded V5 epitope. PCR products were first cloned into pCR2.1-TOPO (Invitrogen). Plasmid DNA was then digested with XbaI and SacII to release the insert, which was cloned into pIZT/V5-His to yield the constructs pIZT-H4 and pIZT-N5. Epitope-tagged expression constructs of Dorsal (pSHhis-Dorsal), Dif (pSHhis-Dif) and Relish (pSHflag-Relish) under the control of the metallothionein promoter were provided by T. Ip (31). A TollΔLRR expression construct using the actin 5C promoter (pPAC-TollΔLRR), and Drosomycin-luciferase (pGL3-Drosomycin) and Attacin-luciferase (pGL3-Attacin) reporter constructs were provided by J.-L. Imler (32). A phRL Renilla luciferase reporter using the cytomegalovirus promoter (Promega) served as an internal standard for reporter assays.

S2 Cell Culture and Transfections. Drosophila S2 cells were maintained in HyQ medium (HyClone) (30). The cells were seeded at 70–80% confluence in 12-well (25 mm) plates (Corning) and transfected, as described in ref. 30, by using Cellfectin (Invitrogen). For the pSH-based vectors, expression was induced by 500 μM CuSO4 24 h before the preparation of cell lysates or immune challenge.

Reporter Assays. For the Drosomycin–luciferase reporter assays, cells were cotransfected with 1 μg of the reporter plasmid, 0.01 μg of phRL Renilla, and 1 μg each of different expression vectors (pPAC-TollΔLRR, pIZT-H4, and pIZT-N5). After 36 h, cell lysates were prepared in Reporter Lysis Buffer (Promega). For the Attacin–luciferase reporter assays, cells were cotransfected with 1 μg of the reporter plasmid, 0.01 μg of phRL Renilla, and 1 μg of pIZT-H4 or -N5. At 24 h posttransfection, the cells were incubated with 10 μg/ml LPS (Escherichia coli serotype 026:B6 Sigma) for 12 h before cell-lysate preparation in Reporter Lysis Buffer (Promega). Luciferase activity was measured with the Dual Luciferase assay system (Promega) by using a luminometer (Molecular Dynamics). Renilla luciferase activity was used to normalize variability among replicates. Reporter activity was then normalized to one in control cells without TollΔLRR or LPS treatment, whereas other treatments were recorded as higher or lower values relative to the controls. Each data point shown in Results is the mean ±SD of six independent assays. Data were analyzed by one-way ANOVA and Tukey's multiple-comparison procedure by using the program jmp 3.0 (SAS Institute, Cary, NC).

Immunoprecipitations. Cells cotransfected with 1 μg of each of the different Rel and IκB-like expression constructs were lysed 48 h posttransfection on ice in 50 mM Tris (pH 7.5), 150 mM NaCl, 1% Nonidet P-40 vol/vol (Pierce), and protease-inhibitor mixture (Roche). Cell lysate (500 μl) was prepared from two 25-mm culture wells. Protein concentrations were adjusted to 1 μg/μl and precleared by using protein A/G-agarose beads (Santa Cruz Biotechnology). Lysates were incubated for 2 h with 1 μl of an anti-V5 antibody (Invitrogen) and then with 10 μl (bed volume) of protein A/G-agarose beads for 3 h. After washing in lysis buffer, the beads were resuspended in 60 μl of SDS/PAGE buffer and boiled, and the antigen–antibody–protein A/G complexes were separated by SDS/PAGE. After transfer to poly(vinylidene difluoride) membrane, the blots were probed with anti-V5, anti-FLAG (Sigma), or anti-RGS-His (Qiagen) primary antibodies and a horseradish-peroxidase-conjugated goat anti-mouse secondary antibody. Western blots were visualized by using the ECL system (Amersham Pharmacia) and a PhosphorImager (Syngene).

EMSAs. Cells cotransfected with 1 μg of each of the different Rel and IκB-like expression constructs were LPS-challenged 24 h postinduction with CuSO4. Eight hours later, nuclear extracts were prepared in NE-PER extraction buffer (Pierce). Oligonucleotides corresponding to κB target sequences for drosomycin (ggccttcggtagGGGAACTACTtgtacgc) or cecropinA1 (tcgagacacGGGGATTTTTgcac) (κB sites in uppercase) were 3′-end-labeled with biotin by using a commercially available kit (Pierce). Control reactions included: (i) the use of a mutant drosomycin (ggccttcggtagATTAACTACTtgtacgc) or cecropinA1 (tcgagacacGTTGATTTTTg cac) probe (mutations underlined), (ii) the addition of 50-fold excess unlabeled wild-type probe to the reaction, and (iii) supershift assays. Binding reactions were carried out by adding 2 ng of labeled probe to 4 μg of nuclear protein extract. For supershift assays, extracts were preincubated with 1 μg of anti-RGS-His or anti-FLAG antibody for 10 min at room temperature before the addition of the probe. After a 30-min incubation at room temperature, binding complexes were separated on 5% native acrylamide gels and visualized by using the LightShift EMSA kit (Pierce) and autoradiography.


MdBV Encodes Multiple IκB-Like Genes. Analysis of the MdBV genome identified seven ORFs on segments F (GenBank accession no. AY875682), G (accession no. AY875684), H (accession no. AY875685), J (accession no. AY875686), and N (accession no. AY875689) that we designated as the IκB-like gene family. Individual family members were named on the basis of the genomic segment upon which they were located (i.e., MdBV IκB-like F5, -G3, -G4, -H4, -J4, -N1, and -N5). The seven ORFs ranged from 420–579 bp and encoded predicted proteins that were 15.8–22.2 kDa. Each gene lacked introns, possessed consensus 5′ TATA boxes that were 15–41 bp upstream of the initiation methionine, and had 3′ polyadenylation signal sequences that were 23–115 bp downstream of the stop codon. None of the IκB-like genes had predicted signal peptides.

All known IκBs have a central ARD that usually consists of six ankyrin repeats (5, 15). Family members such as Cactus have signal-receiving domains (SRDs), N-terminal to the ARD, that accept phosphorylation and ubiquitination signals, whereas the C-terminal domains contain proline, glutamic acid, serine, and threonine (PEST) residues implicated in protein turnover (5, 12). Inactive Relish also has a C-terminal ARD and PEST domain, but its N terminus contains a Rel homology domain. Each predicted protein encoded by the MdBV IκB-like genes was much smaller than Cactus or Relish because of a reduced ARD and the absence of regulatory regions in their N and C termini (Fig. 1A). Predicted viral proteins had centrally located ARDs comprised of four ankyrin repeats that aligned with ankyrin repeats 3–6 of Cactus and IκBα (Fig. 1 A and B). Comparisons with Cactus indicated that percent identities were lowest for the third ankyrin repeat (6–21%), intermediate for the sixth repeat (15–42%), and highest for the fourth and fifth repeats (21–41%) (Fig. 1B). All of the predicted viral proteins lacked SRDs and consensus protein kinase sequences (e.g., SDYK, TPPD, SDIE, KHS, or SRGE) in their N termini. The C termini of each viral protein contained glutamic acid and serine residues but lacked the proline and threonine residues typical of PEST domains.

Fig. 1.
MdBV encodes multiple IκB-like genes. (A) Domain structure of the deduced IκB-like proteins F5, G3, G4, H4, J4, N1, and N5 in comparison with Cactus (Dm-Cac) and Relish (Dm-Rel) from Drosophila. Each viral protein contains a reduced ARD ...

H4 and N5 Suppress Inducible Expression of Attacin and Drosomycin. Northern blotting and real-time PCR indicated that the MdBV IκB-like genes are expressed in virus-infected hosts such as the moth Pseudoplusia includens (H.T., M.H.B., and M.R.S., unpublished work). These studies also indicated that the IκB-like genes H4 and N5 are preferentially expressed in immune tissues such as fat body and hemocytes. Combined with the above sequence analysis, these observations suggested that H4, N5, and, possibly, other IκB-like genes, are functional. As previously discussed, however, NF-κB signaling pathways have not been characterized in insects outside of Drosophila and, to a lesser degree, the mosquitoes Anopheles gambiae and Aedes aegypti (3, 10, 11). We, therefore, conducted functional experiments with the MdBV IκB-like genes H4 and N5 in Drosophila S2 cells, which have been used previously in studies of NF-κB signaling (3133). Immune effector genes under NF-κB control include the antimicrobial peptide genes drosomycin, which is regulated by the Toll pathway, and attacin, which is stimulated by LPS challenge via the Imd pathway (6, 3133). If H4 and N5 encode products with IκB activity, we reasoned that the expression of these factors should disrupt the inducible expression of luciferase reporter constructs under the control of the drosomycin and attacin promoters. As reported in ref. 32, S2 cells expressing a constitutively active form of the Toll receptor TollΔLRR exhibited high levels of expression from the drosomycin–luciferase reporter, when compared with control cells transfected with the reporter alone (Fig. 2A). In contrast, the expression of the drosomycin reporter remained at low levels in cells expressing TollΔLRR and H4 or N5 (Fig. 2A). LPS similarly induced the expression of the attacin–luciferase reporter, compared with nontreated cells transfected with the reporter alone. LPS treatment, however, did not induce the expression of the attacin reporter in cells transfected with pIZT-H4 or pIZT-N5, (Fig. 2B).

Fig. 2.
Induction of antimicrobial peptide promoters is inhibited by H4 and N5. (A) S2 cells were cotransfected with plasmids expressing TollΔLRR, H4, or N5 and a luciferase reporter under the control of the drosomycin promoter. Negative control cells ...

H4 and N5 Complex with Dif and Relish. We next asked whether these viral proteins complex with one or more Rel proteins by transfecting resting-state S2 cells with different combinations of pIZT-H4, pIZT-N5, and epitope-tagged constructs of Dorsal, Dif, and Relish (Fig. 3A). After induction with CuSO4, cell extracts were prepared, and recombinant Dif and Dorsal were detected by using an anti-RGS-His antibody, Relish was detected with an anti-FLAG antibody, and H4 and N5 were detected with an anti-V5 antibody. We readily detected recombinant Dif, Dorsal, Relish, H4, and N5 in extracts prepared from cells transfected individually with each construct (Fig. 3 B and C, lanes 1–4 and lane 1, respectively). The anti-V5 antibody immunoprecipitated H4 and N5 from samples transfected with only the H4 and N5 expression constructs (Fig. 3B, lanes 7 and 8) but did not immunoprecipitate Dif, Dorsal, or Relish from samples transfected with only the Dif, Dorsal, or Relish constructs (Fig. 3 B and C, lanes 5 and 6 and lane 2, respectively). In cotransfected samples, however, the anti-V5 antibody coimmunoprecipitated Dif and Relish when H4 or N5 was present (Fig. 3 B and C, lanes 9 and 11 and lanes 3 and 4, respectively). N5, but not H4, also weakly pulled down Dorsal (Fig. 3B, lanes 10 and 12).

Fig. 3.
H4 and N5 bind Rel proteins. (A) Maps of recombinant proteins used in immunoprecipitation experiments: D. melanogaster Relish (Rel), Dif, and Dorsal (Dor) and H4 and N5. Epitope tags (FLAG, RGS-His, and V5) on each protein are indicated. (B) Outcome of ...

H4 and N5 Inhibit κB-Binding Activity. Immune challenge induces the dissociation of Cactus from Dif and Dorsal and cleavage of Rel-110 to active Rel-68, allowing NF-κB dimers to translocate to the nucleus and interact with target genes. The promoters of several AMP genes from Drosophila contain κB sites that bind to NF-κBs (31, 34), including a κB site in the drosomycin promoter that binds to NF-κBs containing Dif and a κB site in the cecropinA1 promoter that binds Relish (7, 18, 3134). Because H4 and N5 formed stable complexes with Dif and Relish, we tested whether these viral proteins also inhibited the binding of corresponding NF-κBs to drosomycin and cecropinA1 κB sequences after immune challenge. EMSAs detected strong binding to an oligonucleotide containing the κB site of the drosomycin gene by using nuclear extracts from cells transfected with Dif (Fig. 4A, lane 1). By using the same extracts, no binding activity was detected to a mutant drosomycin probe (Fig. 4A, lane 2), and binding activity was competed off by a 50-fold excess of unlabeled wild-type probe (Fig. 4A, lane 5). The NF-κB–DNA complex we detected by EMSA also contained recombinant Dif, because it was supershifted by the addition of anti-RGS-His antibody to the reaction (Fig. 4A, lane 6). These results were consistent with prior studies and indicated that Dif specifically bound to the drosomycin κB sequence. As observed in resting cells (see Fig. 3), Western blotting confirmed that cells cotransfected with Dif, H4, or N5 and immune challenged expressed each recombinant protein (data not shown). More significant, however, was the finding that coexpression of H4 or N5 blocked Dif binding to the drosomycin κB probe (Fig. 4A, lanes 3 and 4). EMSAs conducted with extracts from cells expressing recombinant Relish and H4 or N5 resulted in a similar outcome. Rel specifically bound to the cecropinA1 κB probe in the absence of H4 and N5, as evidenced by control experiments using a mutant cecropinA1 κB probe, excess unlabeled competitor, and supershift assays (Fig. 4B). However, Rel did not bind to the cecropinA1 κB probe when H4 or N5 were present (Fig. 4B, lanes 3 and 4).

Fig. 4.
H4 and N5 inhibit binding of Dif and Relish to κB sites. (A) Outcome of EMSAs using nuclear extracts from cells expressing Dif with or without H4 and N5 and a drosomycin probe. Lane 1, extracts from cells expressing Dif alone and drosomycin wild-type ...


Considerable progress has been made in recent years in understanding how the insect immune system is regulated, yet little is known about how insect pathogens evade host defenses. PDVs are important subjects for such studies because of their diversity and potent immunosuppressive activity (22, 23, 30). Here, we focused on the MdBV IκB-like gene family because of the important role NF-κB signaling plays in insect immunity. Our results strongly indicate that H4 and N5 function as IκBs. Although the expression of epitope-tagged proteins in cell lines may not fully reflect the interactions that occur in whole insects, our results also suggest that these and other IκB-like family members are likely capable of disrupting NF-κB signaling in a diversity of insects and other organisms, including mammals.

The most detailed structure–function data on IκB–NF-κB complexes derive from mammalian IκBα, which binds homo- and heterodimers of the Rel proteins p50 and p65 (5, 1215, 35). Structural studies indicate that the ARD is essential for IκBα–NF-κB binding, whereas the N-terminal SRD and C-terminal PEST domains are not. Within the ARD, ankyrin repeats 3–6 form the largest number of interactions with NF-κBs, leading to the suggestion that these regions alone may be sufficient to form stable complexes with NF-κBs (13). Similar studies have not been conducted with insect IκBs, but our results clearly suggest an important role for ankyrin repeats 3–6 in binding of insect NF-κBs. To our knowledge, no IκBs have been identified from other insect pathogens outside of PDVs. However, an unrelated mammalian virus (African swine fever virus) encodes an IκB-like gene with a reduced ARD that also disrupts the function of the mammalian NF-κB p50/p65 (36).

A second interesting outcome of our study is the finding that H4 and N5 bind inactive Relish (Rel-110), which consists of an N-terminal Rel homology domain and C-terminal ARD (17, 18). Whether Rel-110 functions as both an IκB and a specific homo- or heterodimeric partner for another Rel protein is unclear, although studies with a related Rel-precursor protein from mammals (p100) suggest that dual activity is possible (37). Regardless, the ability of H4 and N5 to complex with Rel-110 suggests that these viral proteins compete with Rel-110's own ARD for binding. A final feature of the MdBV IκB-like genes, compared with known IκBs, is the absence of SRD and PEST domains. This finding suggests that H4, N5, and other IκB-like family members are insensitive to host signaling factors that regulate phosphorylation, degradation, or cleavage of endogenous IκBs after immune challenge. If so, viral IκBs may function as irreversible inhibitors of host NF-κB signaling. Circumstantial evidence for this interpretation derives from mutagenesis studies of Cactus, where deletion of the SRD and PEST domains inhibits phosphorylation, which, in turn, results in the accumulation of Cactus protein and phenotypic alterations consistent with disrupted NF-κB function (38, 39).

An important question is why the MdBV IκB-like genes have diversified to form multiple gene variants. Because mammalian NF-κB-family members have cell-lineage-specific functions (21), one possibility is that MdBV IκB-like variants have functional activities that differ between insect tissues. Our own studies with MdBV indicate that some IκB-like family members are preferentially expressed in host hemocytes and fat body, whereas others are expressed elsewhere (H.T., M.H.B., and M.R.S., unpublished data). Tissue-specific patterns of expression have also been noted for IκB-like genes encoded by PDVs from the ichneumonid wasp Campoletis sonorensis (40) and the braconids Toxoneuron nigriceps and Cotesia congregata (F. Pennacchio, personal communication).

A second possibility is that sequence variation among viral IκB-like proteins affects binding preferences, given that: (i) mammalian Rel proteins (p65, RelB, cRel, p50, and p52) form different homo- and heterodimers that differentially interact with κBsitesin target gene promoters (5, 12, 21), and (ii) mammalian IκBs exhibit NF-κB-binding preferences that correlate with amino acid substitutions in different ankyrin repeats (1315). For example, IκBα and IκBβ preferentially bind p50-p65 and p50-cRel heterodimers, IκBε complexes bind with only homo- and heterodimers containing p65, and Bcl-3 interacts with p50 and p52 homodimers (5, 41). Drosophila encodes three Rel protein genes, whereas the mosquito Anopheles gambiae encodes only two (10). Similar to mammals, however, Dif, Dorsal, and Relish from Drosophila appear to be capable of forming different homo- and heterodimeric combinations that have different DNA-binding affinities (31, 34). Insect Rel-protein diversity may also be further increased by alternative splicing (11). The impact that these potentially different NF-κBs have on the regulation of the insect immune system is unclear. Nonetheless, the picture that collectively emerges is that NF-κB diversity may be more complex in insects than is generally recognized and that pathogens like MdBV have evolved multiple IκB-like genes to selectively interact with this complexity. Results of the current study already suggest that MdBV IκB-like variants have different NF-κB-binding preferences, given that N5 weakly bound Dorsal, whereas H4 did not. It is also possible that some MdBV IκB-like family members bind other host proteins (42).

Whereas AMPs are clearly important in insect immunity, they are often used in studies of NF-κB signaling because of their characteristic κB-binding sites, which allows them, in effect, to serve as marker genes. By no means, though, are they the only effector molecules under NF-κB regulation (6, 8, 43). For example, De Gregorio et al. (6) identified 230 genes in Drosophila whose expression was significantly up-regulated and another 170 genes that were down-regulated in association with NF-κB signaling. These factors ranged from recognition molecules and factors involved in cellular immune responses to many humoral factors and antimicrobial peptides. Since PDVs immunosuppress insect hosts for the benefit of their associated parasitoid (22), disrupting NF-κB signaling is likely significant to more than suppression of AMPs. Of particular importance would be the disruption of NF-κB-regulated molecules involved in encapsulation (43) or antiviral responses toward the PDV itself, given the recent implication of the Toll pathway in viral defense (44). Tissue-specific differences in the expression of host Rel proteins and viral IκBs could, in turn, further refine the types of effector genes affected by disrupted NF-κB function.

Genomic studies indicate that PDVs other than MdBV encode IκB-like genes, suggesting that these factors are both widespread and functionally important (refs. 22, 28, 45, and 46; and F. Pennacchio, personal communication). Given that an estimated 40,000 species of PDV-carrying parasitoid wasps exist and that each species carries a unique viral isolate (22), the diversity of IκB-like and other immunosuppressive genes encoded by PDVs is potentially enormous. The possibility of using these genes as tissue-, cell-, or pathway-specific probes could also provide important new insights in understanding how the insect immune system is regulated.


We thank K. D. Clark, D. M. Donnell, and A. Pruijssers for suggestions during the study; B. A. Webb and F. Pennacchio for valuable discussions; and T. Ip (University of Massachusetts Medical School, Worcester, MA) and J.-L. Imler (Institut de Biologie Moléculaire et Cellulaire, Strasbourg, France) for generously providing the Rel, Toll, and luciferase constructs. This work was supported, in part, by grants from the U.S. Department of Agriculture, the National Institutes of Health, and the Georgia Experiment Station.


Author contributions: M.R.S. designed research; H.T., M.H.B., and M.R.S. performed research; and M.R.S. wrote the paper.

Abbreviations: AMP, antimicrobial peptide; ARD, ankyrin-repeat domain; BV, bracovirus; IκB, inhibitor κB; MdBV, Microplitis demolitor BV; PDVs, polydnaviruses; PEST, proline, glutamic acid, serine, and threonine; SRD, signal-receiving domain.

Data deposition: The sequences reported in this paper have been deposited in the GenBank database (accession nos. AY875682, AY875684, AY875685, AY875686, and AY875689 for segments F, G, H, J, and N of the MdBV genome).


1. Gillespie, J. P., Kanost, M. R. & Trenczek, T. (1997) Annu. Rev. Entomol. 42, 611-643. [PubMed]
2. Lavine, M. D. & Strand, M. R. (2002) Insect Biochem. Mol. Biol. 32, 1237-1242. [PubMed]
3. Hoffmann, J. A. (2003) Nature 426, 33-38. [PubMed]
4. Loker, E. S., Adema, C. M., Zhang, S.-M. & Kepler, T. B. (2004) Immunol. Rev. 198, 10-24. [PubMed]
5. Ghosh, S., May, M. J. & Kopp, E. B. (1998) Annu. Rev. Immunol. 16, 225-260. [PubMed]
6. De Gregorio, E., Spellman, P. T., Rubin, G. M. & Lemaitre, B. (2001) Proc. Natl. Acad. Sci. USA 98, 12590-12595. [PMC free article] [PubMed]
7. Imler J.-L. & Bulet, P. (2005) in Mechanisms of Epithelial Defense, eds. Kabelitz, D. & Schroder, J. M. (Karger, Basel), Vol. 86, pp. 1-21.
8. Bourtros, M., Agaisse, H. & Perrimon, N. (2002) Dev. Cell 3, 711-722. [PubMed]
9. Agaisse, H., Petersen, U. M., Boutros, M., Mathey-Prevot, B. & Perrimon, N. (2003) Dev. Cell 5, 441-450. [PubMed]
10. Christophides, G. K., Vlachou, D. & Kafatos, F. C. (2004) Immunol. Rev. 198, 127-148. [PubMed]
11. Shin, S. W., Kokoza, V., Lobkov, I. & Raikhel, A. S. (2003) Proc. Natl. Acad. Sci. USA 100, 2616-2621. [PMC free article] [PubMed]
12. Ghosh, S. & Karin, M. (2002) Cell 109, S81-S96. [PubMed]
13. Huxford, T., Huang, D.-B., Malek, S. & Ghosh, G. (1998) Cell 95, 759-770. [PubMed]
14. Jacobs, M. D. & Harrison, S. C. (1998) Cell 95, 749-758. [PubMed]
15. Michel, F., Soler-Lopez, M., Petosa, C., Cramer, P., Siebenlist, U. & Muller, C. W. (2001) EMBO J. 20, 6180-6190. [PMC free article] [PubMed]
16. Nicolas, E., Reichhart, J. M., Hoffmann, J. A. & Lemaitre, B. (1998) J. Biol. Chem. 273, 10463-10469. [PubMed]
17. Dushay, M. S., Asling, B. & Hultmark, D. (1996) Proc. Natl. Acad. Sci. USA 93, 10343-10347. [PMC free article] [PubMed]
18. Stoven, S., Ando, I., Kadalayil, L., Engstrom, Y. & Hultmark, D. (2000) EMBO Rep. 1, 347-352. [PMC free article] [PubMed]
19. Tortorella, D., Gewurz, B. E., Furman, M. H., Schust, D. J. & Ploegh, H. L. (2000) Annu. Rev. Immunol. 18, 861-926. [PubMed]
20. Celli, J. & Finlay, B. B. (2002) Trends Microbiol. 10, 232-237. [PubMed]
21. Mason, N. J., Artis, D. & Hunter, C. A. (2004) Immunol. Rev. 198, 48-56. [PubMed]
22. Webb, B. A. & Strand, M. R. (2005) in Comprehensive Molecular Insect Science, eds. Gilbert, L. I., Iatrou, I. & Gill, S. S. (Elsevier, San Diego), Vol. 6, pp. 323-360.
23. Schmidt, O., Theopold, U. & Strand, M. R. (2001) BioEssays 23, 344-351. [PubMed]
24. Strand, M. R. & Noda, T. (1991) J. Insect. Physiol. 37, 839-850.
25. Strand, M. R. (1994) J. Gen. Virol. 75, 3007-3020. [PubMed]
26. Strand, M. R. & Pech, L. L. (1995) J. Gen. Virol. 76, 283-291. [PubMed]
27. Trudeau, D., Witherell, A. R. & Strand, M. R. (2000) J. Gen. Virol. 81, 3049-3058. [PubMed]
28. Kroemer, J. A. & Webb, B. A. (2004) Annu. Rev. Entomol. 49, 431-456. [PubMed]
29. Beck, M. & Strand, M. R. (2003) Virology 314, 521-535. [PubMed]
30. Beck, M. & Strand, M. R. (2005) J. Virol. 79, 1861-1870. [PMC free article] [PubMed]
31. Han, Z. S. & Ip, T. (1999) J. Biol. Chem. 274, 21355-21361. [PubMed]
32. Tauszig, S., Jouanguy, E., Hoffmann, J. A. & Imler, J.-L. (2000) Proc. Natl. Acad. Sci. USA 99, 10520-10525. [PMC free article] [PubMed]
33. Sun, H., Bristow, B. N., Qu, G. & Wasserman, S. A. (2002) Proc. Natl. Acad. Sci. USA 99, 12871-12876. [PMC free article] [PubMed]
34. Uvell, H. & Engstrom, Y. (2003) Mol. Cell. Biol. 23, 8272-8281. [PMC free article] [PubMed]
35. Sun S.-C., Elwood, J. & Greene, W. C. (1996) Mol. Cell. Biol. 16, 1058-1065. [PMC free article] [PubMed]
36. Revilla, Y., Callejo, M., Rodriquez, J. M., Culebras, E., Nogal, M. L., Salas, M. L., Vinuela, E. & Fresno, M. (1998) J. Biol. Chem. 273, 5405-5411. [PubMed]
37. Solan, N. J., Miyoshi, H., Carmona, E. M., Bren, G. D. & Paya, C. V. (2002) J. Biol. Chem. 277, 1405-1418. [PubMed]
38. Bergmann, A., Stein, R., Geisler, R., Hagenmaier, S., Schmid, B., Fernandez, N., Schnell, B. & Nusslein-Volhard, C. (1996) Mech. Dev. 60, 109-123. [PubMed]
39. Liu, Z.-P., Galindo, R. L. & Wasserman, S. A. (1997) Genes Dev. 11, 3413-3422. [PMC free article] [PubMed]
40. Kroemer, J. A. & Webb, B. A. (2005) J. Virol. 79, 7617-7628. [PMC free article] [PubMed]
41. Simeonidis, S., Liang, S., Chen, G. & Thanos, D. (1997) Proc. Natl. Acad. Sci. USA 94, 14372-14377. [PMC free article] [PubMed]
42. Bork, P. (1993) Proteins 17, 363-374. [PubMed]
43. Irving, P., Ubeda, J.-M., Doucet, D., Troxler, L., Lagueux, M., Zachary, D., Hoffmann, J. A., Hetru, C. & Meister, M. (2005) Cell. Immun. 7, 335-350. [PubMed]
44. Zambon, R. A., Nandakumar, M., Vakharia, V. N. & Wu, L. P. (2005) Proc. Natl. Acad. Sci. USA 102, 7257-7262.. [PMC free article] [PubMed]
45. Espagne, E., Dupuy, E. C., Huguet, E., Cattolico, L., Provost, B., Martins, N., Poirie, M., Periquet, G. & Drezen, J. M. (2004) Science 306, 286-289. [PubMed]
46. Malva, C., Varicchio, P., Falabella, P., La Scaleia, R., Graziani, F. & Pennacchio, F. (2004) Insect Biochem. Mol. Biol. 34, 177-183. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...