Glc1.8 from Microplitis demolitor Bracovirus Induc
http://www.100md.com
病菌学杂志 2005年第3期
Department of Entomology and Center for Emerging and Tropical Diseases, University of Georgia, Athens, Georgia
ABSTRACT
Polydnaviridae is a unique family of DNA viruses that are symbiotically associated with parasitoid wasps. Upon oviposition, wasps inject these viruses into their hosts, where they cause several physiological alterations, including suppression of the cellular immune response. Here we report that expression of the glc1.8 gene from Microplitis demolitor bracovirus (MdBV) causes a loss of adhesion by two hemocyte-like cell lines, namely, High Five cells from the lepidopteran Trichoplusia ni and S2 cells from the dipteran Drosophila melanogaster. The expression of recombinant Glc1.8 also greatly reduced the ability of these cells to phagocytize foreign targets. Glc1.8 is characterized by a signal peptide at its N terminus, an extracellular domain comprised of five nearly perfect tandem repeats of 78 amino acids, and a C-terminal hydrophobic domain that encodes a putative membrane anchor sequence. The expression of a Glc1.8 mutant lacking the anchor sequence resulted in a secreted protein that had no effect on adhesion or phagocytosis. In contrast, sequential deletion of the repeats in the extracellular domain resulted in a progressive reduction in immunosuppressive activity. Since each repeat and its associated glycosylation sites are nearly identical, these results suggested that adhesion-blocking activity depends more on the overall number of repeats in the extracellular domain than on the specific determinants within each repeat. While it severely compromised adhesion and phagocytic functions, Glc1.8 did not cause cell death. Collectively, these results indicate that Glc1.8 is a major pathogenic determinant of MdBV that is involved in suppression of the insect cellular immune response.
INTRODUCTION
The family Polydnaviridae consists of segmented, double-stranded DNA viruses that are associated specifically with certain types of parasitoid wasps (46). Polydnaviruses (PDVs) are divided into two types, bracoviruses (BVs) and ichnoviruses, that coexist with wasps in the families Braconidae and Ichneumonidae, respectively (45). All PDVs persist as stably integrated proviruses in the genomes of wasps and replicate asymptomatically in ovarial cells that form a region of the female reproductive tract called the calyx. Virions accumulate in the lumens of the oviducts, and the resulting suspension of virus and protein is called calyx fluid. PDV-carrying wasps parasitize the larval stage of different moth species (Lepidoptera). When a wasp oviposits into a host larva, she deposits one or more eggs and a quantity of virus. PDVs do not replicate in the wasp's host, but the expression of PDV-carried genes causes physiological alterations in the host that are essential for the survival of the parasitoid's progeny. Thus, a true mutualism exists between PDVs and wasps, as viral transmission depends on parasitoid survival and parasitoid survival depends on infection of the wasp's host by the virus.
A key function of PDVs in this regard is the suppression of the host's immune system (24, 30, 43, 45). Insects typically eliminate parasitoids via encapsulation, a defense response in which circulating immune cells (hemocytes) adhere to and form a multicellular sheath around the parasitoid egg or larva (17). Studies with several parasitoid-host systems have indicated that PDV-infected hosts do not mount an encapsulation response, which in turn allows the parasitoid's offspring to successfully develop (1, 7, 11, 19, 28, 31, 44). Hosts parasitized by the braconid Microplitis demolitor are unable to encapsulate parasitoid eggs or other foreign targets because hemocytes infected by M. demolitor bracovirus (MdBV) lose the capacity to adhere to foreign surfaces by 4 to 6 h postinfection (31, 33). Hemocytes in primary culture and the hemocyte-like cell line BTI-TN-5B1-4 (High Five) from Trichoplusia ni also lose the ability to adhere to foreign surfaces after infection by MdBV (3, 31).
Several MdBV genes are expressed in hemocytes and High Five cells, including glc1.8, which encodes a 514-amino-acid mucin-like protein that localizes to the surfaces of infected cells (30, 40). Glc1.8 is characterized by a signal peptide at its N terminus, an extracellular domain comprised of five 78-amino-acid repeats arranged in a tandem array, and a C-terminal hydrophobic domain that encodes a putative anchor sequence (40). In addition, multiple N- and O-glycosylation sites are predicted in the extracellular domain of the protein. The pathogenic determinants of MdBV have yet to be fully described, although recent studies suggested that the Glc1.8 protein is involved in a loss of adhesion by infected immune cells and the inability of these cells to form capsules (3, 40). A knockdown of glc1.8 expression in MdBV-infected High Five cells by RNA interference (RNAi) restores the ability of cells to adhere to and spread on foreign surfaces, but it is unknown whether this RNAi effect is due solely to the activity of glc1.8 or if glc1.8 must interact with other MdBV gene products to cause the loss-of-adhesion phenotype (3). The Glc1.8-associated suppression of adhesion may also play a role in modulating other aspects of the host immune response since cell-cell contact is essential for several cell-mediated immune responses in insects.
In the present study, we report that the expression of recombinant Glc1.8 alone induced rounding and detachment of High Five cells as well as of S2 cells from Drosophila melanogaster. Glc1.8 also greatly reduced the ability of High Five and S2 cells to phagocytize foreign targets, which suggests that this protein is a major pathogenic determinant of MdBV that is involved in suppression of the insect cellular immune response.
MATERIALS AND METHODS
Insects and cell cultures. M. demolitor was reared in the lepidopteran host Pseudoplusia includens at 27 ± 1°C with a 16-h light (L)-8-h dark (D) photoperiod as previously described (28). High Five cells (Invitrogen) were cultured in TC-100 modified medium (Sigma) supplemented with 5% fetal calf serum (HyClone), while Drosophila S2 cells (American Type Culture Collection) were maintained in HyQ medium (HyClone). Cells were maintained and passaged as adherent cells in Corning 75-cm2 tissue culture flasks. Most experiments were conducted in 12-well culture wells (Corning). Cells were seeded into wells at specific densities, allowed to settle and spread for 6 to 12 h, and then used for particular assays.
Calyx fluid collection, virus purification, and cell infection. Calyx fluid was collected from wasps and virus particles were purified on sucrose gradients by established methods (29). Following the convention of PDV literature, the amount of MdBV collected from a single wasp was defined as one wasp equivalent, and viral doses used to infect cells during this study were expressed in wasp equivalents. For infections of cells, MdBV isolated from a cohort of 100 wasps was resuspended in 1 ml of TC-100 medium, yielding a concentration of one wasp equivalent per 10 μl. High Five cells in 12-well culture plates were infected with 0.1 wasp equivalents of MdBV. At a cell density of 2 x 105 cells per well, a prior study demonstrated that 0.1 wasp equivalents of MdBV resulted in >95% of High Five cells being infected (3).
Plasmids and expression. glc1.8 coding sequences were cloned into the insect expression plasmid pIZT/V5-His (Invitrogen). This vector uses the OpIE2 promoter from the Orgyia pseudotsugata baculovirus for constitutive expression of the gene of interest and encodes a Zeocin-green fluorescent protein (GFP) gene fusion under control of the OpIE1 promoter. Wild-type glc1.8 was PCR amplified by the use of MdBV genomic DNA as a template and the primers 5'-AATCTAGAATGGCGCAAATTACTT-3' (forward) and 5'-ATACCGCGGTAACTCGTGAGAAC-3' (reverse). XbaI and SacII sites (underlined) were incorporated into the forward and reverse primers for directional cloning. As recommended by the manufacturer, we modified the sequence around the start codon in the forward primer (bold) to create a Kozak sequence for proper initiation of translation. We also mutated the glc1.8 stop codon from TGA to TAA in the reverse primer (bold) to express the gene product in frame with the vector-encoded V5 epitope and six-His tag. Standard 50-μl reactions contained 2 μl of Elongase enzyme mix (Invitrogen), 1.5 mM MgCl2, a 250 μM deoxynucleoside triphosphate mix, a 200 nM concentration of each primer, and 100 ng of genomic MdBV DNA. After a 3-min initial denaturation step at 94°C, the target sequence was amplified by 30 cycles of 30 s of denaturation at 94°C, 30 s of annealing at 50°C, and 4 min of extension at 68°C. The PCR product was first cloned into pCR2.1-TOPO (Invitrogen). The resulting plasmid DNA was then digested with XbaI and SacII to release the insert, which was cloned into pIZT/V5-His to yield the wild-type expression construct pIZT/Glc1.8 (Fig. 1).
A mutant lacking the C-terminal hydrophobic domain (mutC) was produced by digestion of the wild-type expression construct with NsiI (nucleotide position 1346 of the previously published sequence for glc1.8 [GenBank accession no. AF267175]) and SacII (multiple cloning site of the vector) (Fig. 1). Mutants lacking one or more of the glycosylated tandem repeats in the extracellular domain of Glc1.8 were produced by partial digests of pIZT/Glc1.8 with AflII, which cuts once in every 234-bp repeat unit (nucleotide positions 147, 381, 615, 849, and 1083) (Fig. 1). After religation of the partially digested DNAs, plasmids were transformed into Top10 cells (Invitrogen) and replated on Luria-Bertani agar plates containing Zeocin. Forty colonies were randomly selected and miniprepped, and the sizes of the plasmid inserts were examined in 1% agarose gels. This resulted in the identification of clones lacking one (mut1), two (mut2), three (mut3), or four (mut4) repeats (Fig. 1). The sequences of these and all other constructs used for this study were confirmed by partial or full sequencing by the chain termination method with an ABI Prism cycle sequencing kit (Perkin-Elmer). Sequence reactions were run at the University of Wisconsin—Madison sequencing facility.
Constructs were transiently expressed in High Five or S2 cells by cationic lipid-mediated transfection. Cells were seeded at 70 to 80% confluence in 12-well culture plates (Corning) 24 h prior to transfection. High Five cells were transfected by the use of 6 μl of Lipofectin (Invitrogen) plus 0.02 to 8.0 μg of DNA per ml of TC-100 medium without serum, while S2 cells were transfected by the use of 16 μl of Cellfectin (Invitrogen) plus 2 μg of DNA per ml of HyQ medium. After a 6-h incubation period, the transfection medium was removed. High Five cells were then cultured in TC-100 medium plus serum, and S2 cells were maintained in fresh HyQ medium. Transfection efficiencies for experiments in both cell lines ranged from 50 to 75%, as measured by GFP expression.
Cell adhesion and viability assays. Adhesion assays were conducted in 12-well culture plates seeded with High Five or S2 cells as previously described (3). Briefly, cells were infected with MdBV or transfected as described above. Forty-eight hours later, cells were collected from wells, washed two times in medium, and then seeded at a density of 104 cells per well in new culture plates containing 500 μl of TC-100 medium plus serum (High Five) or HyQ (S2 cells) medium/well. After 2 h, the assays were stopped by placing the culture plates on ice. Although GFP served as a marker for transfected cells, we also labeled living cells with an anti-Glc1.8 antibody (see below) to unambiguously distinguish cells expressing Glc1.8 from cells that did not. The percentage of rounded, nonadherent cells relative to adherent attached cells expressing Glc1.8 was then scored by counting 200 cells per well from four randomly selected fields of view by use of a Leica TCS inverted epifluorescence microscope. In separate replicates, the viability of cells expressing Glc1.8 was determined by a dye exclusion assay using propidium iodide (2 μg/ml).
Western blotting. Cells and the culture medium were first separated by centrifugation. Cell pellets were washed three times in phosphate-buffered saline (PBS) (pH 7.2) and then placed directly in sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) loading buffer. The culture medium was likewise added at a ratio of 1:1 to SDS-PAGE loading buffer. Discontinuous polyacrylamide gel electrophoresis and Western blotting were performed with precast 8 to 16% gradient gels (Gradipore) and the subsequent transfer of proteins to nitrocellulose membranes. Recombinant Glc1.8 was detected with a murine anti-V5 antibody (Invitrogen) (1:10,000) and a goat anti-mouse horseradish peroxidase-conjugated secondary antibody (Jackson Laboratory) (1:50,000). Glc1.8 expressed in MdBV-infected cells was detected with a previously developed anti-Glc1.8 monoclonal antibody (55F2G3) (40). Bands were visualized by chemiluminescence with an ECL Advance Western blotting detection kit (Amersham Biosciences) and a GeneGenome bio-imaging system (Syngene).
Immunofluorescence staining. Cells were processed for immunofluorescence microscopy as previously described (3). Briefly, living cells were stained with anti-Glc1.8 and a Texas Red- or fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse secondary antibody. Other cells were fixed by first being washed with PBS and then fixed by the addition of 2% paraformaldehyde in PBS for 20 min. After being washed in PBS, fixed cells were permeabilized with PBS-0.1% Triton X-100 for 10 min, followed by staining with the anti-Glc1.8 antibody and a Texas Red- or FITC-conjugated secondary antibody. Actin microfilaments were labeled by the incubation of fixed and permeabilized cells with Alexa 488-conjugated phalloidin (Molecular Probes) in PBS (1:200). Samples were examined under a Leica TCS scanning confocal microscope.
FITC labeling of bacteria. FITC labeling of bacteria was performed by the method of Hed and Stendahl (13). Heat-killed Escherichia coli (109/ml) cells were incubated in 0.5% carbonate buffer (pH 9.5) containing FITC (0.1 mg/ml) for 30 min at 37°C. Thereafter, the FITC-conjugated bacteria were washed four times in PBS (pH 7.2) and then kept at –20°C until use.
Phagocytosis assays. We assessed the ability of High Five and S2 cells to phagocytize bacteria and inert polystyrene microspheres (0.5-μm-diameter) (Polysciences) by using a fluorescence quenching method (8). Cells were first infected with MdBV or transfected with different Glc1.8 expression constructs. Forty-eight hours later, the cells were collected and added to new 12-well culture plates in medium without serum at a density of 103 cells per well. After a 2-h preincubation period, bacteria or microspheres were added to each culture well at a ratio of 15:1. Cells were allowed to phagocytize for 2 h at 25°C, followed by transfer of the culture plate to ice. At this time, cells in some replicates were labeled with the anti-Glc1.8 antibody as described above to identify cells expressing Glc1.8 on their surfaces. We first scored the percentage of cells with one or more particles adhered to their surface by counting 200 cells per well from four randomly selected fields of view by use of a Leica TCS inverted epifluorescence microscope. Particles were green, while cells expressing Glc1.8 were red. We then scored the percentage of cells that had phagocytized particles by adding a drop of trypan blue (1 mg/ml in citrate buffer; pH 4.4). Intracellular particles continued to fluoresce green after the addition of trypan blue, while the fluorescence of extracellular particles was quenched. We also noted that the intensity of staining of the FITC-labeled particles made it easy to distinguish them from the much weaker GFP signals in the cytoplasm and/or nuclei of transfected cells.
Statistical analysis. The number of samples used for each experimental condition is indicated in the figure legends. Statistical analysis was performed by one-way analysis of variance followed by the Student-Neumann-Keuls (SNK) multiple comparison procedure.
RESULTS
Glc1.8 induces detachment and rounding of High Five cells. High Five cells, transfected with empty vector, readily attached and spread on the surfaces of culture plates, whereas cells transfected with 2 μg of pIZT/Glc1.8 began detaching from culture plates at 18 to 24 h posttransfection (Fig. 2A and B). By 36 h, a large proportion of cells were nonadhesive and rounded (Fig. 2B). Antibody labeling revealed that the Glc1.8 protein was strongly expressed on the surfaces of transfected cells (Fig. 2C) in a manner similar to the pattern of Glc1.8 expression on cells that had been infected with 0.1 wasp equivalents of MdBV (Fig. 2D). Increasing the amount of pIZT/Glc1.8 DNA used to transfect cells above 2 μg/ml resulted in a small decrease in the percentage of adhesive cells, while decreasing the amount of DNA increased the percentage of adhesive cells to levels similar to those for cells transfected with an empty vector (Fig. 2E). Staining with anti-Glc1.8 antibody indicated that reducing the amount of plasmid used for transfections resulted in both weaker staining and a more patchy distribution of Glc1.8 protein on the surfaces of cells (data not presented). Time course studies also indicated that the proportion of adhesive cells progressively declined from 0 to 48 h posttransfection and then remained near a steady state for a 5-day period (Fig. 2F). During this time, viability assays using propidium iodide indicated that >95% of cells expressing Glc1.8 were viable.
Both the transmembrane and extracellular domains of Glc1.8 are required for adhesion-blocking activity. A previous analysis using TMHMM, version 2.0 (http://www.cbs.dtu.dk/services), predicted the presence of a transmembrane domain in the C terminus of Glc1.8 (Fig. 1) (40). NetNGlyc (version 1) (http://www.cbs.dtu.dk/services/NetNGlyc/) and NetOGlyc (version 2) (http://www.cbs.dtu.dk/services/NetOGlyc/) also predicted multiple N- and O-glycosylation sites in association with the five tandem repeats that make up the majority of the extracellular domain of the protein (Fig. 1). The transfection of High Five cells with a mutant lacking the hydrophobic domain (mutC) resulted in a complete loss of the nonadhesive, cell-rounding phenotype produced by wild-type Glc1.8 (Fig. 3A). A Western blot analysis detected the mutC protein primarily in the culture medium, whereas wild-type Glc1.8 was detected almost exclusively in cell extracts (Fig. 3B). An immunocytochemical analysis confirmed this observation, as cells transfected with the mutC construct exhibited no staining of the cell surface with an anti-Glc1.8 antibody and only weak staining of the cytoplasm (data not presented). Likewise, we did not detect any Glc1.8 on the surfaces of nontransfected cells cultured in conditioned medium containing the secreted mutC form of Glc1.8 or when nontransfected cells were cocultured with High Five cells expressing wild-type Glc1.8. Collectively, these results indicated that the hydrophobic domain is required for the retention of Glc1.8 at the cell surface and that anchoring of the protein to the cell membrane is essential for the adhesion-blocking activity.
To map additional determinants required for the adhesion-blocking activity of Glc1.8, we also generated mutants (mut1 to mut4) lacking one or more of the tandem repeats of the extracellular domain. The sequential removal of these repeats led to a progressive increase in the percentage of High Five cells that remained adhesive following transfection (Fig. 3A). Mutants lacking one or two repeats (mut1 and mut2) caused a similar proportion of cells to become nonadhesive as did wild-type Glc1.8, whereas mut4, which had only one repeat, caused few cells to become nonadhesive (Fig. 3). Western blotting confirmed that mutants lacking one or more of these repeats remained associated with transfected cells but resulted in a proportional reduction in the molecular mass of the protein (Fig. 3B). Immunocytochemical staining of living cells also indicated that although the mutant proteins were smaller than the wild-type protein, each of them localized to the cell surface (see below). Since the repeats and their associated glycosylation sites are nearly identical to each other, these data suggested that the adhesion-blocking activity requires a minimum of three repeats in the extracellular domain.
Glc1.8 expression disrupts the actin cytoskeleton. Along with cell rounding and a loss of adhesion, wild-type Glc1.8 also disrupted the formation of actin stress fibers in transfected cells. The cytoskeletons of uninfected High Five cells and of cells transfected with an empty vector exhibited well-developed actin filaments (Fig. 4A and B). In contrast, cells infected with MdBV or transfected with wild-type Glc1.8 led to the disruption of filamentous actin (F-actin) in the cytoplasm and induced the accumulation of F-actin at the periphery of cells (Fig. 4C to F). Cells transfected with mut4 exhibited intermediate effects (Fig. 4G and H). Most cells were still adherent but showed reductions in the lengths of actin stress fibers and the appearance of cellular protrusions such as microspikes or a ruffled appearance. Some cells transfected with mut4 also showed aggregates of actin in their cytoplasm.
Glc1.8-mediated adhesion loss extends to other hemocyte-like cell lines. Like High Five cells, the Drosophila S2 cell line exhibits several hemocyte-like properties, including the ability to adhere to foreign surfaces and to phagocytize a variety of targets (22, 23). S2 cells also possess functional Toll and immunodeficiency pathways that regulate inflammatory responses such as inducible antimicrobial peptide expression (12, 36). Although S2 cells are not permissive for MdBV infection, their hemocyte-like properties prompted us to ask whether the transfection of these cells with recombinant Glc1.8 would also disrupt the ability of these cells to adhere to cell culture plates. The transfection of S2 cells with pIZT/Glc1.8 resulted in the expression of Glc1.8 on cell surfaces and in a loss of adhesion, similar to what occurred with High Five cells (Fig. 5). The transfection of S2 cells with mut1 also reduced the percentage of adhesive cells compared to that for cells transfected with the empty vector (pIZT/V5-His). However, the percentage of cells expressing mut1 that remained adhesive was higher than that for cells expressing wild-type Glc1.8 (Fig. 5). The adhesion of S2 cells was unaffected by mutC (Fig. 5).
Glc1.8 expression blocks phagocytosis. Phagocytosis is a highly conserved defense response that relies on similar intracellular machineries across diverse taxa (4). Prior observations indicated that MdBV-infected hemocytes are unable to phagocytize bacteria (34). Since phagocytosis depends upon the binding of targets to the cell surface, we also examined whether Glc1.8 expression disrupted this important innate immune response. To answer this question, we infected High Five cells with MdBV or transfected them with wild-type Glc1.8, mut1, mutC, or an empty vector. We then incubated these cells with FITC-labeled E. coli or inert polystyrene beads. Our results indicated that High Five cells infected with MdBV or transfected with wild-type Glc1.8 showed a marked reduction in both adherence to and phagocytosis of E. coli cells (Fig. 6A and B). Wild-type Glc1.8 also almost completely inhibited the phagocytosis of E. coli by S2 cells (Fig. 6C and D). High Five and S2 cells transfected with mut1 exhibited intermediate responses, while a similar percentage of cells transfected with mutC bound and ingested particles as that for cells transfected with the empty pIZT/V5-His vector (Fig. 6). Figure 7 visually presents the dramatic differences in the phagocytic responses of High Five cells transfected with vector only and of those transfected with wild-type Glc1.8. The inhibition of phagocytosis was also not restricted to bacteria since Glc1.8 expression caused reduced binding to and phagocytosis of inert beads (Fig. 6). A comparison of the responses of High Five and S2 cells to these targets, however, suggested that Glc1.8 expression reduced the binding of E. coli more completely than that of inert beads (Fig. 6). This was most apparent for S2 cells, as the beads bound to 73% of S2 cells transfected with the mut1 mutant but E. coli bound to only 33% of these cells (Fig. 6C). In contrast, Glc1.8 reduced the percentages of cells that phagocytized (i.e., internalized) E. coli and beads nearly equally for both cell types. These results demonstrated overall that MdBV infection blocks phagocytosis in High Five cells and that this effect is reproduced in both High Five and S2 cells expressing wild-type Glc1.8.
DISCUSSION
Previous studies unambiguously demonstrated that MdBV plays a major role in the immunosuppression of hosts that are parasitized by M. demolitor (15, 28, 31, 32). After infection by MdBV, hemocytes lose the ability to adhere to foreign surfaces, which blocks encapsulation and allows the progeny of M. demolitor to develop. However, MdBV-mediated protection of the parasitoid also results in a more global suppression of the cellular immune response, since hemocytes and hemocyte-like cell lines are also unable to phagocytize small intruders such as bacteria or to encapsulate other foreign targets. In this study, we focused on the role of Glc1.8 in these responses since prior RNAi knockdown experiments implicated this gene in the loss of adhesion by MdBV-infected cells (3). Our results demonstrated that the expression of recombinant Glc1.8 reduces the ability of High Five cells to adhere to foreign surfaces and to phagocytize targets in a manner very similar to that which occurs after infection by MdBV itself. This suggests that the adhesion and phagocytosis-blocking activities of Glc1.8 do not depend on other viral gene products that are expressed in MdBV-infected immune cells. Our results also suggest that the biological activity of Glc1.8 is not restricted to Lepidoptera such as T. ni, which is a natural host of M. demolitor, since Glc1.8 also blocks adhesion and phagocytosis in Drosophila S2 cells.
Mucins are a diverse group of glycoproteins that can be either membrane bound or secreted. Cell surface mucins usually contain a high percentage (20 to 55%) of Pro, Thr, and Ser (PTS) residues arrayed in repeat domains and are anchored to the cell membrane by a transmembrane domain at the C terminus (14, 25, 26, 39, 42). Glc1.8 has a PTS content of 28.5%, 29 N-glycosylation sites, and 4 O-glycosylation sites in its repetitive extracellular domain and a membrane-spanning region at its C terminus (40). The expression of a mutant (mutC) lacking the transmembrane domain resulted in a secreted form of Glc1.8 that contained all of the glycosylation sites of the wild-type protein. However, mutC had no effect on either cell adhesion or phagocytosis, indicating that the membrane anchor is essential for the functions of Glc1.8. The immunosuppressive activity of Glc1.8 is also likely restricted to cells expressing the protein since secreted Glc1.8 did not affect the activity of nontransfected cells and since cells expressing wild-type Glc1.8 did not appear to alter the adhesion or phagocytosis of normal cells when they were cocultured together. While MdBV carries several genes whose products are secreted, prior studies have suggested that none of these factors alter adhesion, phagocytosis, or other cellular immune functions (3, 27, 32, 40).
The second region of Glc1.8 that is required for its biological activity is the extracellular domain of the protein, which is comprised of glycosylated repeats arranged in a tandem array. Interestingly, the envelope glycoprotein of Ebola virus and vertebrate cell surface mucins like episialin induce strikingly similar decreases in cell attachment that also require the presence of heavily glycosylated extracellular domains for their function (5, 26, 41, 47). Since surface mucins form rigid structures that extend several nanometers above the cell surface, they may block adhesion by physically hindering ligand-receptor interactions (14, 21, 47). The progressive reduction in adhesion-blocking activity with sequential deletions of the repeats from Glc1.8 was consistent with the idea that the overall length of the extracellular domain is more important for its function than the glycosylation sites that are present in each individual repeat. The results of our phagocytosis assays further suggested the possibility that Glc1.8 may interfere with multiple adhesion or pattern recognition receptors. Studies of mammalian immune cells indicate that the binding of bacteria and inert targets such as polystyrene beads involves different surface receptors (4, 10). The surface receptors that mediate the phagocytosis of foreign targets by insect immune cells are not as well characterized as those on mammalian immunocytes (17), but recent studies with S2 cells and Drosophila hemocytes have suggested that different receptors mediate the binding of gram-negative bacteria such as E. coli and of inert targets such as beads (22, 23).
In addition to possibly interfering with adhesion receptor-ligand interactions, Ebola virus GP downregulates ?1 integrins and several other cell surface molecules by unknown mechanisms. This transcriptional effect may also contribute to the inhibition of adhesion (26, 35, 37). We currently do not know whether the presence of Glc1.8 on the surfaces of insect cells alters the expression of other endogenous cell surface molecules. Given that Glc1.8 induces a loss of adhesion but not cell death, the evidence circumstantially suggests that this factor modulates the expression or activation of cell adhesion molecules. In addition, recent studies indicated that MdBV infection inhibits the inducible expression of selected - and ?-integrin subunits by lepidopteran hemocytes (16, 18). Glc1.8-mediated alterations of the cell surface also likely contributed to the changes in cytoplasmic actin distribution that we observed, since all adhesion and phagocytic processes involve complex rearrangements of the actin cytoskeleton (10).
Whether other PDVs carry genes with functions like those of glc1.8 is unclear. Bracoviruses associated with other parasitoids in the genus Microplitis disrupt adhesion and capsule formation by host hemocytes, but whether these alterations are mediated by glc1.8 orthologs is unknown (15, 38). One gene, P30, was also recently described for Hyposoter didymater ichnovirus and encodes a glycosylated protein with mucin-like motifs (9). P30 is expressed at high levels in Sf9 cells and hemocytes from infected Spodoptera larvae, but unlike Glc1.8, it lacks a transmembrane domain. The resemblance of the P30 mucin domain to that of Glc1.8, however, suggests the possibility that it may play a role in suppression of the host's cellular immune response. A few other PDV genes have also been implicated in the suppression of host immune responses (1, 2, 6, 20), but none of these factors share any structural similarities with Glc1.8.
In summary, we suggest that the primary function of Glc1.8 is the disruption of the host's immune response toward the parasitoid since the adhesion of immune cells to the target and to one another is essential for capsule formation. In contrast, we are less certain about why MdBV encodes a virulence gene like glc1.8 that also suppresses phagocytosis, as this may render parasitized hosts more susceptible to infection by other pathogens such as bacteria. The inhibition of phagocytosis may simply be a side effect of the way that Glc1.8 disrupts cell adhesion, while other arms of the host immune system, such as antimicrobial peptide synthesis, continue to confer some protection on the host against infection by other organisms as the parasitoid develops. Alternatively, since MdBV is injected directly into the hemocoel of the host at oviposition, the rapid suppression of phagocytosis may enhance the infection of other tissues in the host by preventing hemocytes from clearing MdBV itself from circulation. Testing these ideas as well as understanding the molecular mechanisms underlying how Glc1.8 suppresses immune cell functions will require more detailed information about the host molecules with which this virulence factor interacts.
ACKNOWLEDGMENTS
We thank K. Clark for comments on the manuscript and J. Johnson for assistance with the maintenance of insects and cell lines used for this study.
This study was supported by grants from the U.S. Department of Agriculture NRI Program and from the National Institutes of Health to M.R.S.
REFERENCES
Asgari, S., M. Hellers, and O. Schmidt. 1996. Host haemocyte inactivation by an insect parasitoid: transient expression of a polydnavirus gene. J. Gen. Virol. 77:2653-2662.
Asgari, S., O. Schmidt, and U. Theopold. 1997. A polydnavirus-encoded protein of an endoparasitoid wasp is an immune suppressor. J. Gen. Virol. 78:3061-3070.
Beck, M., and M. R. Strand. 2003. RNA interference silences Microplitis demolitor bracovirus genes and implicates glc1.8 in blocking adhesion of infected host cells. Virology 314:521-535.
Blander, J. M., and R. Medzhitov. 2004. Regulation of phagosome maturation by signals from Toll-like receptors. Science 304:1014-1018.
Chervenak, J. L., and N. P. Illsley. 2000. Episialin acts as an antiadhesive factor in an in vitro model of human endometrial-blastocyst attachment. Biol. Reprod. 63:294-300.
Cui, L., A. S. Soldevila, and B. A. Webb. 1997. Expression and hemocyte targeting of a Campoletis sonorensis polydnavirus cysteine-rich gene in Heliothis virescens larvae. Arch. Insect Biochem. Physiol. 36:251-271.
Davies, D. H., M. R. Strand, and S. B. Vinson. 1987. Changes in differential haemocyte count and in vitro behaviour of plasmatocytes from host Heliothis virescens caused by Campoletis sonorensis polydnavirus. J. Insect Physiol. 33:143-153.
Fallman, M., K. Andersson, S. Hakansson, K.-E. Magnusson, L. Stendahl, and H. W. Watz. 1995. Yersinia pseudotuberculosis inhibits Fc receptor-mediated phagocytosis in J774 cells. Infect. Immun. 63:3117-3124.
Galibert, L., J. Rocher, M. Ravallec, M. Duonon-Cerutti, B. A. Webb, and A. N. Volkoff. 2003. Two Hyposoter didymator ichnovirus genes expressed in the lepidopteran host encode secreted or membrane-associated serine- and threonine-rich proteins in segments that may be nested. J. Insect Physiol. 49:441-451.
Gruenheid, S., and B. B. Finlay. 2003. Microbial pathogenesis and cytoskeletal function. Nature 422:775-781.
Guzo, D., and D. B. Stoltz. 1987. Observations on cellular immunity and parasitism. J. Insect Physiol. 33:19-26.
Han, Z. S., and T. Ip. 1999. Interaction and specificity of Rel-related proteins in regulating Drosophila immunity gene expression. J. Biol. Chem. 274:21355-21361.
Hed, L., and O. Stendahl. 1982. Differences in the ingestion mechanisms of IgG and C3b particles in phagocytosis by neutrophils. Immunology 45:727-736.
Hilkens, J., J. L. Ligtenberg, H. L. Vos, and S. Litvinov. 1992. Cell membrane-associated mucins and their adhesion-modulating property. Trends Biochem. Sci. 17:359-363.
Kadash, K., J. A. Harvey, and M. R. Strand. 2003. Cross-protection experiments with parasitoids in the genus Microplitis (Hymenoptera; Braconidae) suggest a high level of specificity in their associated bracoviruses. J. Insect Physiol. 49:473-482.
Lavine, M. D. 2002. Mediators of hemocyte adhesion during encapsulation in the soybean looper, Pseudoplusia includens. Ph.D. thesis. University of Wisconsin, Madison.
Lavine, M. D., and M. R. Strand. 2002. Insect hemocytes and their role in cellular immune responses. Insect Biochem. Mol. Biol. 32:1237-1242.
Lavine, M. D., and M. R. Strand. 2003. Hemocytes from Pseudoplusia includens express multiple alpha and beta integrin subunits. Insect Mol. Biol. 12:441-452.
Lavine, M. D., and N. E. Beckage. 1996. Temporal pattern of parasitism-induced immunosuppression in Manduca sexta larvae parasitized by Cotesia congregata. J. Insect Physiol. 42:39-49.
Li, X., and B. A. Webb. 1994. Apparent functional role for a cysteine-rich polydnavirus protein in suppression of the insect cellular immune response. J. Virol. 68:7482-7489.
Ligtenberg, J. J. L., F. Buijs, H. L. Vos, and J. Hilkens. 1992. Suppression of cellular aggregation by high levels of episialin. Cancer Res. 52:2318-2324.
Ramet, M., A. Pearson, P. Manfruelli, X. Li, H. Koziel, V. Gobel, E. Chung, M. Krieger, and R. A. B. Ezekowitz. 2001. Drosophila scavenger receptor CI is a pattern recognition receptor for bacteria. Immunology 15:1027-1038.
Ramet, M., P. Manfruelli, A. Pearson, B. Mathey-Prevot, and R. A. B. Ezekowitz. 2002. Functional genomic analysis of phagocytosis and identification of a Drosophila receptor for E. coli. Nature 416:644-648.
Schmidt, O., U. Theopold, and M. R. Strand. 2001. Innate immunity and evasion by insect parasitoids. Bioessays 23:344-351.
Shen, Z., G. Dimopoulos, F. C. Kafatos, and M. Jacobs-Lorena. 1999. A cell surface mucin specifically expressed in the midgut of the malaria mosquito, Anopheles gambiae. Proc. Natl. Acad. Sci. USA 96:5610-5615.
Simmons, G., R. J. Wool-Lewis, F. Baribaud, R. C. Netter, and P. Bates. 2002. Ebola virus glycoproteins induce global surface protein down-regulation and loss of cell adherence. J. Virol. 76:2518-2528.
Strand, M. R. 1994. Microplitis demolitor polydnavirus infects and expresses in specific morphotypes of Pseudoplusia includens haemocytes. J. Gen. Virol. 75:3007-3020.
Strand, M. R., and T. Noda. 1991. Alterations in the haemocytes of Pseudoplusia includens after parasitism by Microplitis demolitor. J. Insect Physiol. 37:839-850.
Strand, M. R., D. I. McKenzie, V. Grassl, B. A. Dover, and J. M. Aiken. 1992. Persistence and expression of Microplitis demolitor polydnavirus in Pseudoplusia includens. J. Gen. Virol. 73:1627-1635.
Strand, M. R., and L. L. Pech. 1995. Immunological compatibility in parasitoid-host relationships. Annu. Rev. Entomol. 40:31-56.
Strand, M. R., and L. L. Pech. 1995. Microplitis demolitor polydnavirus induces apoptosis of a specific haemocyte morphotype in Pseudoplusia includens. J. Gen. Virol. 76:283.
Strand, M. R., S. A. Witherell, and D. Trudeau. 1997. Two related Microplitis demolitor polydnavirus mRNAs expressed in hemocytes of Pseudoplusia includens contain a common cysteine-rich domain. J. Virol. 71:2146-2156.
Strand, M. R., K. C. Clark, and E. M. M. Gardiner. 1999. Plasmatocyte spreading peptide does not induce Microplitis demolitor polydnavirus-infected plasmatocytes to spread on foreign surfaces. Arch. Insect Biochem. Physiol. 42:213-223.
Strand, M. R., and M. D. Lavine. Unpublished data.
Sullivan, N., Z.-Y. Yang, and G. J. Nabel. 2003. Ebola virus pathogenesis: implications for vaccines and therapies. J. Virol. 77:9733-9737.
Sun, H., B. N. Bristow, G. Qu, and S. A. Wasserman. 2002. A heterotrimeric death domain complex in Toll signaling. Proc. Natl. Acad. Sci. USA 99:12871-12876.
Takada, A., S. Watanabe, H. Ito, K. Ozazak, H. Kida, and Y. Kawaoka. 2000. Downregulation of ?1 integrins by Ebola virus glycoprotein: implication for virus entry. Virology 278:20-26.
Tanaka, T. 1987. Morphological changes in haemocytes of the host Pseudaletia separata parasitized by Microplitis mediator or Apanteles kariyae. Dev. Comp. Immunol. 1:57-67.
Theopold, U., C. Samakovlis, H. Erdjumentbromage, N. Dillon, B. Axelsson, O. Schmidt, P. Tempst, and D. Hultmark. 1996. Helix pomatia lectin, an inducer of the Drosophila immune response, binds to hemomucin, a novel surface mucin. J. Biol. Chem. 271:12708-12715.
Trudeau, D., A. R. Witherell, and M. R. Strand. 2000. Characterization of two novel Microplitis demolitor polydnavirus mRNAs expressed in Pseudoplusia includens haemocytes. J. Gen. Virol. 81:3049-3058.
van de Wiel-Kemanade, E., M. J. L. Ligtenberg, A. J. de Boer, F. Buijs, H. L. Vos, C. J. M. Melief, J. Hilkens, and C. G. Fifdor. 1993. Episialin (MUC1) inhibits cytotoxic lymphocyte-target cell interaction. J. Immunol. 151:767-776.
van Klinken, J.-W., J. Dekker, H. A. Buller, and A. W. C. Einerhand. 1995. Mucin gene structure and expression: protection vs. adhesion. Am. J. Physiol. 269:G613-G627.
Webb, B. A. 1998. Polydnavirus biology, genome structure, and evolution, p. 105-139. In L. K. Miller and A. L. Ball (ed.), The insect viruses. Plenum Press, New York, N.Y.
Webb, B. A., and S. Luckhart. 1996. Factors mediating short-term and long-term immunosuppression in a parasitized insect. J. Insect Physiol. 42:33-40.
Webb, B. A., N. E. Beckage, Y. Hayakawa, P. J. Krell, B. Lanzrein, M. R. Strand, D. B. Stoltz, and M. D. Summers. 2000. Polydnaviridae, p 253-259. In M. H. V. Regenmortel et al. (ed.), Virus taxonomy. Academic Press, San Diego, Calif.
Webb, B. A., and M. R. Strand. 2004. The biology and genomics of polydnaviruses. In L. I. Gilbert, I. Iatrou, and S. S. Gill (ed.), Comprehensive molecular insect science, vol. 6, p. 323-360. Elsevier, San Diego, Calif.
Wessling, J., S. W. van der Valk, H. L. Vos, A. Sonnenberg, and J. Hilkens. 1995. Episialin (MUC1) overexpression inhibits integrin-mediated cell adhesion to extracellular matrix components. J. Cell Biol. 129:255-265.(Markus Beck and Michael R)
ABSTRACT
Polydnaviridae is a unique family of DNA viruses that are symbiotically associated with parasitoid wasps. Upon oviposition, wasps inject these viruses into their hosts, where they cause several physiological alterations, including suppression of the cellular immune response. Here we report that expression of the glc1.8 gene from Microplitis demolitor bracovirus (MdBV) causes a loss of adhesion by two hemocyte-like cell lines, namely, High Five cells from the lepidopteran Trichoplusia ni and S2 cells from the dipteran Drosophila melanogaster. The expression of recombinant Glc1.8 also greatly reduced the ability of these cells to phagocytize foreign targets. Glc1.8 is characterized by a signal peptide at its N terminus, an extracellular domain comprised of five nearly perfect tandem repeats of 78 amino acids, and a C-terminal hydrophobic domain that encodes a putative membrane anchor sequence. The expression of a Glc1.8 mutant lacking the anchor sequence resulted in a secreted protein that had no effect on adhesion or phagocytosis. In contrast, sequential deletion of the repeats in the extracellular domain resulted in a progressive reduction in immunosuppressive activity. Since each repeat and its associated glycosylation sites are nearly identical, these results suggested that adhesion-blocking activity depends more on the overall number of repeats in the extracellular domain than on the specific determinants within each repeat. While it severely compromised adhesion and phagocytic functions, Glc1.8 did not cause cell death. Collectively, these results indicate that Glc1.8 is a major pathogenic determinant of MdBV that is involved in suppression of the insect cellular immune response.
INTRODUCTION
The family Polydnaviridae consists of segmented, double-stranded DNA viruses that are associated specifically with certain types of parasitoid wasps (46). Polydnaviruses (PDVs) are divided into two types, bracoviruses (BVs) and ichnoviruses, that coexist with wasps in the families Braconidae and Ichneumonidae, respectively (45). All PDVs persist as stably integrated proviruses in the genomes of wasps and replicate asymptomatically in ovarial cells that form a region of the female reproductive tract called the calyx. Virions accumulate in the lumens of the oviducts, and the resulting suspension of virus and protein is called calyx fluid. PDV-carrying wasps parasitize the larval stage of different moth species (Lepidoptera). When a wasp oviposits into a host larva, she deposits one or more eggs and a quantity of virus. PDVs do not replicate in the wasp's host, but the expression of PDV-carried genes causes physiological alterations in the host that are essential for the survival of the parasitoid's progeny. Thus, a true mutualism exists between PDVs and wasps, as viral transmission depends on parasitoid survival and parasitoid survival depends on infection of the wasp's host by the virus.
A key function of PDVs in this regard is the suppression of the host's immune system (24, 30, 43, 45). Insects typically eliminate parasitoids via encapsulation, a defense response in which circulating immune cells (hemocytes) adhere to and form a multicellular sheath around the parasitoid egg or larva (17). Studies with several parasitoid-host systems have indicated that PDV-infected hosts do not mount an encapsulation response, which in turn allows the parasitoid's offspring to successfully develop (1, 7, 11, 19, 28, 31, 44). Hosts parasitized by the braconid Microplitis demolitor are unable to encapsulate parasitoid eggs or other foreign targets because hemocytes infected by M. demolitor bracovirus (MdBV) lose the capacity to adhere to foreign surfaces by 4 to 6 h postinfection (31, 33). Hemocytes in primary culture and the hemocyte-like cell line BTI-TN-5B1-4 (High Five) from Trichoplusia ni also lose the ability to adhere to foreign surfaces after infection by MdBV (3, 31).
Several MdBV genes are expressed in hemocytes and High Five cells, including glc1.8, which encodes a 514-amino-acid mucin-like protein that localizes to the surfaces of infected cells (30, 40). Glc1.8 is characterized by a signal peptide at its N terminus, an extracellular domain comprised of five 78-amino-acid repeats arranged in a tandem array, and a C-terminal hydrophobic domain that encodes a putative anchor sequence (40). In addition, multiple N- and O-glycosylation sites are predicted in the extracellular domain of the protein. The pathogenic determinants of MdBV have yet to be fully described, although recent studies suggested that the Glc1.8 protein is involved in a loss of adhesion by infected immune cells and the inability of these cells to form capsules (3, 40). A knockdown of glc1.8 expression in MdBV-infected High Five cells by RNA interference (RNAi) restores the ability of cells to adhere to and spread on foreign surfaces, but it is unknown whether this RNAi effect is due solely to the activity of glc1.8 or if glc1.8 must interact with other MdBV gene products to cause the loss-of-adhesion phenotype (3). The Glc1.8-associated suppression of adhesion may also play a role in modulating other aspects of the host immune response since cell-cell contact is essential for several cell-mediated immune responses in insects.
In the present study, we report that the expression of recombinant Glc1.8 alone induced rounding and detachment of High Five cells as well as of S2 cells from Drosophila melanogaster. Glc1.8 also greatly reduced the ability of High Five and S2 cells to phagocytize foreign targets, which suggests that this protein is a major pathogenic determinant of MdBV that is involved in suppression of the insect cellular immune response.
MATERIALS AND METHODS
Insects and cell cultures. M. demolitor was reared in the lepidopteran host Pseudoplusia includens at 27 ± 1°C with a 16-h light (L)-8-h dark (D) photoperiod as previously described (28). High Five cells (Invitrogen) were cultured in TC-100 modified medium (Sigma) supplemented with 5% fetal calf serum (HyClone), while Drosophila S2 cells (American Type Culture Collection) were maintained in HyQ medium (HyClone). Cells were maintained and passaged as adherent cells in Corning 75-cm2 tissue culture flasks. Most experiments were conducted in 12-well culture wells (Corning). Cells were seeded into wells at specific densities, allowed to settle and spread for 6 to 12 h, and then used for particular assays.
Calyx fluid collection, virus purification, and cell infection. Calyx fluid was collected from wasps and virus particles were purified on sucrose gradients by established methods (29). Following the convention of PDV literature, the amount of MdBV collected from a single wasp was defined as one wasp equivalent, and viral doses used to infect cells during this study were expressed in wasp equivalents. For infections of cells, MdBV isolated from a cohort of 100 wasps was resuspended in 1 ml of TC-100 medium, yielding a concentration of one wasp equivalent per 10 μl. High Five cells in 12-well culture plates were infected with 0.1 wasp equivalents of MdBV. At a cell density of 2 x 105 cells per well, a prior study demonstrated that 0.1 wasp equivalents of MdBV resulted in >95% of High Five cells being infected (3).
Plasmids and expression. glc1.8 coding sequences were cloned into the insect expression plasmid pIZT/V5-His (Invitrogen). This vector uses the OpIE2 promoter from the Orgyia pseudotsugata baculovirus for constitutive expression of the gene of interest and encodes a Zeocin-green fluorescent protein (GFP) gene fusion under control of the OpIE1 promoter. Wild-type glc1.8 was PCR amplified by the use of MdBV genomic DNA as a template and the primers 5'-AATCTAGAATGGCGCAAATTACTT-3' (forward) and 5'-ATACCGCGGTAACTCGTGAGAAC-3' (reverse). XbaI and SacII sites (underlined) were incorporated into the forward and reverse primers for directional cloning. As recommended by the manufacturer, we modified the sequence around the start codon in the forward primer (bold) to create a Kozak sequence for proper initiation of translation. We also mutated the glc1.8 stop codon from TGA to TAA in the reverse primer (bold) to express the gene product in frame with the vector-encoded V5 epitope and six-His tag. Standard 50-μl reactions contained 2 μl of Elongase enzyme mix (Invitrogen), 1.5 mM MgCl2, a 250 μM deoxynucleoside triphosphate mix, a 200 nM concentration of each primer, and 100 ng of genomic MdBV DNA. After a 3-min initial denaturation step at 94°C, the target sequence was amplified by 30 cycles of 30 s of denaturation at 94°C, 30 s of annealing at 50°C, and 4 min of extension at 68°C. The PCR product was first cloned into pCR2.1-TOPO (Invitrogen). The resulting plasmid DNA was then digested with XbaI and SacII to release the insert, which was cloned into pIZT/V5-His to yield the wild-type expression construct pIZT/Glc1.8 (Fig. 1).
A mutant lacking the C-terminal hydrophobic domain (mutC) was produced by digestion of the wild-type expression construct with NsiI (nucleotide position 1346 of the previously published sequence for glc1.8 [GenBank accession no. AF267175]) and SacII (multiple cloning site of the vector) (Fig. 1). Mutants lacking one or more of the glycosylated tandem repeats in the extracellular domain of Glc1.8 were produced by partial digests of pIZT/Glc1.8 with AflII, which cuts once in every 234-bp repeat unit (nucleotide positions 147, 381, 615, 849, and 1083) (Fig. 1). After religation of the partially digested DNAs, plasmids were transformed into Top10 cells (Invitrogen) and replated on Luria-Bertani agar plates containing Zeocin. Forty colonies were randomly selected and miniprepped, and the sizes of the plasmid inserts were examined in 1% agarose gels. This resulted in the identification of clones lacking one (mut1), two (mut2), three (mut3), or four (mut4) repeats (Fig. 1). The sequences of these and all other constructs used for this study were confirmed by partial or full sequencing by the chain termination method with an ABI Prism cycle sequencing kit (Perkin-Elmer). Sequence reactions were run at the University of Wisconsin—Madison sequencing facility.
Constructs were transiently expressed in High Five or S2 cells by cationic lipid-mediated transfection. Cells were seeded at 70 to 80% confluence in 12-well culture plates (Corning) 24 h prior to transfection. High Five cells were transfected by the use of 6 μl of Lipofectin (Invitrogen) plus 0.02 to 8.0 μg of DNA per ml of TC-100 medium without serum, while S2 cells were transfected by the use of 16 μl of Cellfectin (Invitrogen) plus 2 μg of DNA per ml of HyQ medium. After a 6-h incubation period, the transfection medium was removed. High Five cells were then cultured in TC-100 medium plus serum, and S2 cells were maintained in fresh HyQ medium. Transfection efficiencies for experiments in both cell lines ranged from 50 to 75%, as measured by GFP expression.
Cell adhesion and viability assays. Adhesion assays were conducted in 12-well culture plates seeded with High Five or S2 cells as previously described (3). Briefly, cells were infected with MdBV or transfected as described above. Forty-eight hours later, cells were collected from wells, washed two times in medium, and then seeded at a density of 104 cells per well in new culture plates containing 500 μl of TC-100 medium plus serum (High Five) or HyQ (S2 cells) medium/well. After 2 h, the assays were stopped by placing the culture plates on ice. Although GFP served as a marker for transfected cells, we also labeled living cells with an anti-Glc1.8 antibody (see below) to unambiguously distinguish cells expressing Glc1.8 from cells that did not. The percentage of rounded, nonadherent cells relative to adherent attached cells expressing Glc1.8 was then scored by counting 200 cells per well from four randomly selected fields of view by use of a Leica TCS inverted epifluorescence microscope. In separate replicates, the viability of cells expressing Glc1.8 was determined by a dye exclusion assay using propidium iodide (2 μg/ml).
Western blotting. Cells and the culture medium were first separated by centrifugation. Cell pellets were washed three times in phosphate-buffered saline (PBS) (pH 7.2) and then placed directly in sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) loading buffer. The culture medium was likewise added at a ratio of 1:1 to SDS-PAGE loading buffer. Discontinuous polyacrylamide gel electrophoresis and Western blotting were performed with precast 8 to 16% gradient gels (Gradipore) and the subsequent transfer of proteins to nitrocellulose membranes. Recombinant Glc1.8 was detected with a murine anti-V5 antibody (Invitrogen) (1:10,000) and a goat anti-mouse horseradish peroxidase-conjugated secondary antibody (Jackson Laboratory) (1:50,000). Glc1.8 expressed in MdBV-infected cells was detected with a previously developed anti-Glc1.8 monoclonal antibody (55F2G3) (40). Bands were visualized by chemiluminescence with an ECL Advance Western blotting detection kit (Amersham Biosciences) and a GeneGenome bio-imaging system (Syngene).
Immunofluorescence staining. Cells were processed for immunofluorescence microscopy as previously described (3). Briefly, living cells were stained with anti-Glc1.8 and a Texas Red- or fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse secondary antibody. Other cells were fixed by first being washed with PBS and then fixed by the addition of 2% paraformaldehyde in PBS for 20 min. After being washed in PBS, fixed cells were permeabilized with PBS-0.1% Triton X-100 for 10 min, followed by staining with the anti-Glc1.8 antibody and a Texas Red- or FITC-conjugated secondary antibody. Actin microfilaments were labeled by the incubation of fixed and permeabilized cells with Alexa 488-conjugated phalloidin (Molecular Probes) in PBS (1:200). Samples were examined under a Leica TCS scanning confocal microscope.
FITC labeling of bacteria. FITC labeling of bacteria was performed by the method of Hed and Stendahl (13). Heat-killed Escherichia coli (109/ml) cells were incubated in 0.5% carbonate buffer (pH 9.5) containing FITC (0.1 mg/ml) for 30 min at 37°C. Thereafter, the FITC-conjugated bacteria were washed four times in PBS (pH 7.2) and then kept at –20°C until use.
Phagocytosis assays. We assessed the ability of High Five and S2 cells to phagocytize bacteria and inert polystyrene microspheres (0.5-μm-diameter) (Polysciences) by using a fluorescence quenching method (8). Cells were first infected with MdBV or transfected with different Glc1.8 expression constructs. Forty-eight hours later, the cells were collected and added to new 12-well culture plates in medium without serum at a density of 103 cells per well. After a 2-h preincubation period, bacteria or microspheres were added to each culture well at a ratio of 15:1. Cells were allowed to phagocytize for 2 h at 25°C, followed by transfer of the culture plate to ice. At this time, cells in some replicates were labeled with the anti-Glc1.8 antibody as described above to identify cells expressing Glc1.8 on their surfaces. We first scored the percentage of cells with one or more particles adhered to their surface by counting 200 cells per well from four randomly selected fields of view by use of a Leica TCS inverted epifluorescence microscope. Particles were green, while cells expressing Glc1.8 were red. We then scored the percentage of cells that had phagocytized particles by adding a drop of trypan blue (1 mg/ml in citrate buffer; pH 4.4). Intracellular particles continued to fluoresce green after the addition of trypan blue, while the fluorescence of extracellular particles was quenched. We also noted that the intensity of staining of the FITC-labeled particles made it easy to distinguish them from the much weaker GFP signals in the cytoplasm and/or nuclei of transfected cells.
Statistical analysis. The number of samples used for each experimental condition is indicated in the figure legends. Statistical analysis was performed by one-way analysis of variance followed by the Student-Neumann-Keuls (SNK) multiple comparison procedure.
RESULTS
Glc1.8 induces detachment and rounding of High Five cells. High Five cells, transfected with empty vector, readily attached and spread on the surfaces of culture plates, whereas cells transfected with 2 μg of pIZT/Glc1.8 began detaching from culture plates at 18 to 24 h posttransfection (Fig. 2A and B). By 36 h, a large proportion of cells were nonadhesive and rounded (Fig. 2B). Antibody labeling revealed that the Glc1.8 protein was strongly expressed on the surfaces of transfected cells (Fig. 2C) in a manner similar to the pattern of Glc1.8 expression on cells that had been infected with 0.1 wasp equivalents of MdBV (Fig. 2D). Increasing the amount of pIZT/Glc1.8 DNA used to transfect cells above 2 μg/ml resulted in a small decrease in the percentage of adhesive cells, while decreasing the amount of DNA increased the percentage of adhesive cells to levels similar to those for cells transfected with an empty vector (Fig. 2E). Staining with anti-Glc1.8 antibody indicated that reducing the amount of plasmid used for transfections resulted in both weaker staining and a more patchy distribution of Glc1.8 protein on the surfaces of cells (data not presented). Time course studies also indicated that the proportion of adhesive cells progressively declined from 0 to 48 h posttransfection and then remained near a steady state for a 5-day period (Fig. 2F). During this time, viability assays using propidium iodide indicated that >95% of cells expressing Glc1.8 were viable.
Both the transmembrane and extracellular domains of Glc1.8 are required for adhesion-blocking activity. A previous analysis using TMHMM, version 2.0 (http://www.cbs.dtu.dk/services), predicted the presence of a transmembrane domain in the C terminus of Glc1.8 (Fig. 1) (40). NetNGlyc (version 1) (http://www.cbs.dtu.dk/services/NetNGlyc/) and NetOGlyc (version 2) (http://www.cbs.dtu.dk/services/NetOGlyc/) also predicted multiple N- and O-glycosylation sites in association with the five tandem repeats that make up the majority of the extracellular domain of the protein (Fig. 1). The transfection of High Five cells with a mutant lacking the hydrophobic domain (mutC) resulted in a complete loss of the nonadhesive, cell-rounding phenotype produced by wild-type Glc1.8 (Fig. 3A). A Western blot analysis detected the mutC protein primarily in the culture medium, whereas wild-type Glc1.8 was detected almost exclusively in cell extracts (Fig. 3B). An immunocytochemical analysis confirmed this observation, as cells transfected with the mutC construct exhibited no staining of the cell surface with an anti-Glc1.8 antibody and only weak staining of the cytoplasm (data not presented). Likewise, we did not detect any Glc1.8 on the surfaces of nontransfected cells cultured in conditioned medium containing the secreted mutC form of Glc1.8 or when nontransfected cells were cocultured with High Five cells expressing wild-type Glc1.8. Collectively, these results indicated that the hydrophobic domain is required for the retention of Glc1.8 at the cell surface and that anchoring of the protein to the cell membrane is essential for the adhesion-blocking activity.
To map additional determinants required for the adhesion-blocking activity of Glc1.8, we also generated mutants (mut1 to mut4) lacking one or more of the tandem repeats of the extracellular domain. The sequential removal of these repeats led to a progressive increase in the percentage of High Five cells that remained adhesive following transfection (Fig. 3A). Mutants lacking one or two repeats (mut1 and mut2) caused a similar proportion of cells to become nonadhesive as did wild-type Glc1.8, whereas mut4, which had only one repeat, caused few cells to become nonadhesive (Fig. 3). Western blotting confirmed that mutants lacking one or more of these repeats remained associated with transfected cells but resulted in a proportional reduction in the molecular mass of the protein (Fig. 3B). Immunocytochemical staining of living cells also indicated that although the mutant proteins were smaller than the wild-type protein, each of them localized to the cell surface (see below). Since the repeats and their associated glycosylation sites are nearly identical to each other, these data suggested that the adhesion-blocking activity requires a minimum of three repeats in the extracellular domain.
Glc1.8 expression disrupts the actin cytoskeleton. Along with cell rounding and a loss of adhesion, wild-type Glc1.8 also disrupted the formation of actin stress fibers in transfected cells. The cytoskeletons of uninfected High Five cells and of cells transfected with an empty vector exhibited well-developed actin filaments (Fig. 4A and B). In contrast, cells infected with MdBV or transfected with wild-type Glc1.8 led to the disruption of filamentous actin (F-actin) in the cytoplasm and induced the accumulation of F-actin at the periphery of cells (Fig. 4C to F). Cells transfected with mut4 exhibited intermediate effects (Fig. 4G and H). Most cells were still adherent but showed reductions in the lengths of actin stress fibers and the appearance of cellular protrusions such as microspikes or a ruffled appearance. Some cells transfected with mut4 also showed aggregates of actin in their cytoplasm.
Glc1.8-mediated adhesion loss extends to other hemocyte-like cell lines. Like High Five cells, the Drosophila S2 cell line exhibits several hemocyte-like properties, including the ability to adhere to foreign surfaces and to phagocytize a variety of targets (22, 23). S2 cells also possess functional Toll and immunodeficiency pathways that regulate inflammatory responses such as inducible antimicrobial peptide expression (12, 36). Although S2 cells are not permissive for MdBV infection, their hemocyte-like properties prompted us to ask whether the transfection of these cells with recombinant Glc1.8 would also disrupt the ability of these cells to adhere to cell culture plates. The transfection of S2 cells with pIZT/Glc1.8 resulted in the expression of Glc1.8 on cell surfaces and in a loss of adhesion, similar to what occurred with High Five cells (Fig. 5). The transfection of S2 cells with mut1 also reduced the percentage of adhesive cells compared to that for cells transfected with the empty vector (pIZT/V5-His). However, the percentage of cells expressing mut1 that remained adhesive was higher than that for cells expressing wild-type Glc1.8 (Fig. 5). The adhesion of S2 cells was unaffected by mutC (Fig. 5).
Glc1.8 expression blocks phagocytosis. Phagocytosis is a highly conserved defense response that relies on similar intracellular machineries across diverse taxa (4). Prior observations indicated that MdBV-infected hemocytes are unable to phagocytize bacteria (34). Since phagocytosis depends upon the binding of targets to the cell surface, we also examined whether Glc1.8 expression disrupted this important innate immune response. To answer this question, we infected High Five cells with MdBV or transfected them with wild-type Glc1.8, mut1, mutC, or an empty vector. We then incubated these cells with FITC-labeled E. coli or inert polystyrene beads. Our results indicated that High Five cells infected with MdBV or transfected with wild-type Glc1.8 showed a marked reduction in both adherence to and phagocytosis of E. coli cells (Fig. 6A and B). Wild-type Glc1.8 also almost completely inhibited the phagocytosis of E. coli by S2 cells (Fig. 6C and D). High Five and S2 cells transfected with mut1 exhibited intermediate responses, while a similar percentage of cells transfected with mutC bound and ingested particles as that for cells transfected with the empty pIZT/V5-His vector (Fig. 6). Figure 7 visually presents the dramatic differences in the phagocytic responses of High Five cells transfected with vector only and of those transfected with wild-type Glc1.8. The inhibition of phagocytosis was also not restricted to bacteria since Glc1.8 expression caused reduced binding to and phagocytosis of inert beads (Fig. 6). A comparison of the responses of High Five and S2 cells to these targets, however, suggested that Glc1.8 expression reduced the binding of E. coli more completely than that of inert beads (Fig. 6). This was most apparent for S2 cells, as the beads bound to 73% of S2 cells transfected with the mut1 mutant but E. coli bound to only 33% of these cells (Fig. 6C). In contrast, Glc1.8 reduced the percentages of cells that phagocytized (i.e., internalized) E. coli and beads nearly equally for both cell types. These results demonstrated overall that MdBV infection blocks phagocytosis in High Five cells and that this effect is reproduced in both High Five and S2 cells expressing wild-type Glc1.8.
DISCUSSION
Previous studies unambiguously demonstrated that MdBV plays a major role in the immunosuppression of hosts that are parasitized by M. demolitor (15, 28, 31, 32). After infection by MdBV, hemocytes lose the ability to adhere to foreign surfaces, which blocks encapsulation and allows the progeny of M. demolitor to develop. However, MdBV-mediated protection of the parasitoid also results in a more global suppression of the cellular immune response, since hemocytes and hemocyte-like cell lines are also unable to phagocytize small intruders such as bacteria or to encapsulate other foreign targets. In this study, we focused on the role of Glc1.8 in these responses since prior RNAi knockdown experiments implicated this gene in the loss of adhesion by MdBV-infected cells (3). Our results demonstrated that the expression of recombinant Glc1.8 reduces the ability of High Five cells to adhere to foreign surfaces and to phagocytize targets in a manner very similar to that which occurs after infection by MdBV itself. This suggests that the adhesion and phagocytosis-blocking activities of Glc1.8 do not depend on other viral gene products that are expressed in MdBV-infected immune cells. Our results also suggest that the biological activity of Glc1.8 is not restricted to Lepidoptera such as T. ni, which is a natural host of M. demolitor, since Glc1.8 also blocks adhesion and phagocytosis in Drosophila S2 cells.
Mucins are a diverse group of glycoproteins that can be either membrane bound or secreted. Cell surface mucins usually contain a high percentage (20 to 55%) of Pro, Thr, and Ser (PTS) residues arrayed in repeat domains and are anchored to the cell membrane by a transmembrane domain at the C terminus (14, 25, 26, 39, 42). Glc1.8 has a PTS content of 28.5%, 29 N-glycosylation sites, and 4 O-glycosylation sites in its repetitive extracellular domain and a membrane-spanning region at its C terminus (40). The expression of a mutant (mutC) lacking the transmembrane domain resulted in a secreted form of Glc1.8 that contained all of the glycosylation sites of the wild-type protein. However, mutC had no effect on either cell adhesion or phagocytosis, indicating that the membrane anchor is essential for the functions of Glc1.8. The immunosuppressive activity of Glc1.8 is also likely restricted to cells expressing the protein since secreted Glc1.8 did not affect the activity of nontransfected cells and since cells expressing wild-type Glc1.8 did not appear to alter the adhesion or phagocytosis of normal cells when they were cocultured together. While MdBV carries several genes whose products are secreted, prior studies have suggested that none of these factors alter adhesion, phagocytosis, or other cellular immune functions (3, 27, 32, 40).
The second region of Glc1.8 that is required for its biological activity is the extracellular domain of the protein, which is comprised of glycosylated repeats arranged in a tandem array. Interestingly, the envelope glycoprotein of Ebola virus and vertebrate cell surface mucins like episialin induce strikingly similar decreases in cell attachment that also require the presence of heavily glycosylated extracellular domains for their function (5, 26, 41, 47). Since surface mucins form rigid structures that extend several nanometers above the cell surface, they may block adhesion by physically hindering ligand-receptor interactions (14, 21, 47). The progressive reduction in adhesion-blocking activity with sequential deletions of the repeats from Glc1.8 was consistent with the idea that the overall length of the extracellular domain is more important for its function than the glycosylation sites that are present in each individual repeat. The results of our phagocytosis assays further suggested the possibility that Glc1.8 may interfere with multiple adhesion or pattern recognition receptors. Studies of mammalian immune cells indicate that the binding of bacteria and inert targets such as polystyrene beads involves different surface receptors (4, 10). The surface receptors that mediate the phagocytosis of foreign targets by insect immune cells are not as well characterized as those on mammalian immunocytes (17), but recent studies with S2 cells and Drosophila hemocytes have suggested that different receptors mediate the binding of gram-negative bacteria such as E. coli and of inert targets such as beads (22, 23).
In addition to possibly interfering with adhesion receptor-ligand interactions, Ebola virus GP downregulates ?1 integrins and several other cell surface molecules by unknown mechanisms. This transcriptional effect may also contribute to the inhibition of adhesion (26, 35, 37). We currently do not know whether the presence of Glc1.8 on the surfaces of insect cells alters the expression of other endogenous cell surface molecules. Given that Glc1.8 induces a loss of adhesion but not cell death, the evidence circumstantially suggests that this factor modulates the expression or activation of cell adhesion molecules. In addition, recent studies indicated that MdBV infection inhibits the inducible expression of selected - and ?-integrin subunits by lepidopteran hemocytes (16, 18). Glc1.8-mediated alterations of the cell surface also likely contributed to the changes in cytoplasmic actin distribution that we observed, since all adhesion and phagocytic processes involve complex rearrangements of the actin cytoskeleton (10).
Whether other PDVs carry genes with functions like those of glc1.8 is unclear. Bracoviruses associated with other parasitoids in the genus Microplitis disrupt adhesion and capsule formation by host hemocytes, but whether these alterations are mediated by glc1.8 orthologs is unknown (15, 38). One gene, P30, was also recently described for Hyposoter didymater ichnovirus and encodes a glycosylated protein with mucin-like motifs (9). P30 is expressed at high levels in Sf9 cells and hemocytes from infected Spodoptera larvae, but unlike Glc1.8, it lacks a transmembrane domain. The resemblance of the P30 mucin domain to that of Glc1.8, however, suggests the possibility that it may play a role in suppression of the host's cellular immune response. A few other PDV genes have also been implicated in the suppression of host immune responses (1, 2, 6, 20), but none of these factors share any structural similarities with Glc1.8.
In summary, we suggest that the primary function of Glc1.8 is the disruption of the host's immune response toward the parasitoid since the adhesion of immune cells to the target and to one another is essential for capsule formation. In contrast, we are less certain about why MdBV encodes a virulence gene like glc1.8 that also suppresses phagocytosis, as this may render parasitized hosts more susceptible to infection by other pathogens such as bacteria. The inhibition of phagocytosis may simply be a side effect of the way that Glc1.8 disrupts cell adhesion, while other arms of the host immune system, such as antimicrobial peptide synthesis, continue to confer some protection on the host against infection by other organisms as the parasitoid develops. Alternatively, since MdBV is injected directly into the hemocoel of the host at oviposition, the rapid suppression of phagocytosis may enhance the infection of other tissues in the host by preventing hemocytes from clearing MdBV itself from circulation. Testing these ideas as well as understanding the molecular mechanisms underlying how Glc1.8 suppresses immune cell functions will require more detailed information about the host molecules with which this virulence factor interacts.
ACKNOWLEDGMENTS
We thank K. Clark for comments on the manuscript and J. Johnson for assistance with the maintenance of insects and cell lines used for this study.
This study was supported by grants from the U.S. Department of Agriculture NRI Program and from the National Institutes of Health to M.R.S.
REFERENCES
Asgari, S., M. Hellers, and O. Schmidt. 1996. Host haemocyte inactivation by an insect parasitoid: transient expression of a polydnavirus gene. J. Gen. Virol. 77:2653-2662.
Asgari, S., O. Schmidt, and U. Theopold. 1997. A polydnavirus-encoded protein of an endoparasitoid wasp is an immune suppressor. J. Gen. Virol. 78:3061-3070.
Beck, M., and M. R. Strand. 2003. RNA interference silences Microplitis demolitor bracovirus genes and implicates glc1.8 in blocking adhesion of infected host cells. Virology 314:521-535.
Blander, J. M., and R. Medzhitov. 2004. Regulation of phagosome maturation by signals from Toll-like receptors. Science 304:1014-1018.
Chervenak, J. L., and N. P. Illsley. 2000. Episialin acts as an antiadhesive factor in an in vitro model of human endometrial-blastocyst attachment. Biol. Reprod. 63:294-300.
Cui, L., A. S. Soldevila, and B. A. Webb. 1997. Expression and hemocyte targeting of a Campoletis sonorensis polydnavirus cysteine-rich gene in Heliothis virescens larvae. Arch. Insect Biochem. Physiol. 36:251-271.
Davies, D. H., M. R. Strand, and S. B. Vinson. 1987. Changes in differential haemocyte count and in vitro behaviour of plasmatocytes from host Heliothis virescens caused by Campoletis sonorensis polydnavirus. J. Insect Physiol. 33:143-153.
Fallman, M., K. Andersson, S. Hakansson, K.-E. Magnusson, L. Stendahl, and H. W. Watz. 1995. Yersinia pseudotuberculosis inhibits Fc receptor-mediated phagocytosis in J774 cells. Infect. Immun. 63:3117-3124.
Galibert, L., J. Rocher, M. Ravallec, M. Duonon-Cerutti, B. A. Webb, and A. N. Volkoff. 2003. Two Hyposoter didymator ichnovirus genes expressed in the lepidopteran host encode secreted or membrane-associated serine- and threonine-rich proteins in segments that may be nested. J. Insect Physiol. 49:441-451.
Gruenheid, S., and B. B. Finlay. 2003. Microbial pathogenesis and cytoskeletal function. Nature 422:775-781.
Guzo, D., and D. B. Stoltz. 1987. Observations on cellular immunity and parasitism. J. Insect Physiol. 33:19-26.
Han, Z. S., and T. Ip. 1999. Interaction and specificity of Rel-related proteins in regulating Drosophila immunity gene expression. J. Biol. Chem. 274:21355-21361.
Hed, L., and O. Stendahl. 1982. Differences in the ingestion mechanisms of IgG and C3b particles in phagocytosis by neutrophils. Immunology 45:727-736.
Hilkens, J., J. L. Ligtenberg, H. L. Vos, and S. Litvinov. 1992. Cell membrane-associated mucins and their adhesion-modulating property. Trends Biochem. Sci. 17:359-363.
Kadash, K., J. A. Harvey, and M. R. Strand. 2003. Cross-protection experiments with parasitoids in the genus Microplitis (Hymenoptera; Braconidae) suggest a high level of specificity in their associated bracoviruses. J. Insect Physiol. 49:473-482.
Lavine, M. D. 2002. Mediators of hemocyte adhesion during encapsulation in the soybean looper, Pseudoplusia includens. Ph.D. thesis. University of Wisconsin, Madison.
Lavine, M. D., and M. R. Strand. 2002. Insect hemocytes and their role in cellular immune responses. Insect Biochem. Mol. Biol. 32:1237-1242.
Lavine, M. D., and M. R. Strand. 2003. Hemocytes from Pseudoplusia includens express multiple alpha and beta integrin subunits. Insect Mol. Biol. 12:441-452.
Lavine, M. D., and N. E. Beckage. 1996. Temporal pattern of parasitism-induced immunosuppression in Manduca sexta larvae parasitized by Cotesia congregata. J. Insect Physiol. 42:39-49.
Li, X., and B. A. Webb. 1994. Apparent functional role for a cysteine-rich polydnavirus protein in suppression of the insect cellular immune response. J. Virol. 68:7482-7489.
Ligtenberg, J. J. L., F. Buijs, H. L. Vos, and J. Hilkens. 1992. Suppression of cellular aggregation by high levels of episialin. Cancer Res. 52:2318-2324.
Ramet, M., A. Pearson, P. Manfruelli, X. Li, H. Koziel, V. Gobel, E. Chung, M. Krieger, and R. A. B. Ezekowitz. 2001. Drosophila scavenger receptor CI is a pattern recognition receptor for bacteria. Immunology 15:1027-1038.
Ramet, M., P. Manfruelli, A. Pearson, B. Mathey-Prevot, and R. A. B. Ezekowitz. 2002. Functional genomic analysis of phagocytosis and identification of a Drosophila receptor for E. coli. Nature 416:644-648.
Schmidt, O., U. Theopold, and M. R. Strand. 2001. Innate immunity and evasion by insect parasitoids. Bioessays 23:344-351.
Shen, Z., G. Dimopoulos, F. C. Kafatos, and M. Jacobs-Lorena. 1999. A cell surface mucin specifically expressed in the midgut of the malaria mosquito, Anopheles gambiae. Proc. Natl. Acad. Sci. USA 96:5610-5615.
Simmons, G., R. J. Wool-Lewis, F. Baribaud, R. C. Netter, and P. Bates. 2002. Ebola virus glycoproteins induce global surface protein down-regulation and loss of cell adherence. J. Virol. 76:2518-2528.
Strand, M. R. 1994. Microplitis demolitor polydnavirus infects and expresses in specific morphotypes of Pseudoplusia includens haemocytes. J. Gen. Virol. 75:3007-3020.
Strand, M. R., and T. Noda. 1991. Alterations in the haemocytes of Pseudoplusia includens after parasitism by Microplitis demolitor. J. Insect Physiol. 37:839-850.
Strand, M. R., D. I. McKenzie, V. Grassl, B. A. Dover, and J. M. Aiken. 1992. Persistence and expression of Microplitis demolitor polydnavirus in Pseudoplusia includens. J. Gen. Virol. 73:1627-1635.
Strand, M. R., and L. L. Pech. 1995. Immunological compatibility in parasitoid-host relationships. Annu. Rev. Entomol. 40:31-56.
Strand, M. R., and L. L. Pech. 1995. Microplitis demolitor polydnavirus induces apoptosis of a specific haemocyte morphotype in Pseudoplusia includens. J. Gen. Virol. 76:283.
Strand, M. R., S. A. Witherell, and D. Trudeau. 1997. Two related Microplitis demolitor polydnavirus mRNAs expressed in hemocytes of Pseudoplusia includens contain a common cysteine-rich domain. J. Virol. 71:2146-2156.
Strand, M. R., K. C. Clark, and E. M. M. Gardiner. 1999. Plasmatocyte spreading peptide does not induce Microplitis demolitor polydnavirus-infected plasmatocytes to spread on foreign surfaces. Arch. Insect Biochem. Physiol. 42:213-223.
Strand, M. R., and M. D. Lavine. Unpublished data.
Sullivan, N., Z.-Y. Yang, and G. J. Nabel. 2003. Ebola virus pathogenesis: implications for vaccines and therapies. J. Virol. 77:9733-9737.
Sun, H., B. N. Bristow, G. Qu, and S. A. Wasserman. 2002. A heterotrimeric death domain complex in Toll signaling. Proc. Natl. Acad. Sci. USA 99:12871-12876.
Takada, A., S. Watanabe, H. Ito, K. Ozazak, H. Kida, and Y. Kawaoka. 2000. Downregulation of ?1 integrins by Ebola virus glycoprotein: implication for virus entry. Virology 278:20-26.
Tanaka, T. 1987. Morphological changes in haemocytes of the host Pseudaletia separata parasitized by Microplitis mediator or Apanteles kariyae. Dev. Comp. Immunol. 1:57-67.
Theopold, U., C. Samakovlis, H. Erdjumentbromage, N. Dillon, B. Axelsson, O. Schmidt, P. Tempst, and D. Hultmark. 1996. Helix pomatia lectin, an inducer of the Drosophila immune response, binds to hemomucin, a novel surface mucin. J. Biol. Chem. 271:12708-12715.
Trudeau, D., A. R. Witherell, and M. R. Strand. 2000. Characterization of two novel Microplitis demolitor polydnavirus mRNAs expressed in Pseudoplusia includens haemocytes. J. Gen. Virol. 81:3049-3058.
van de Wiel-Kemanade, E., M. J. L. Ligtenberg, A. J. de Boer, F. Buijs, H. L. Vos, C. J. M. Melief, J. Hilkens, and C. G. Fifdor. 1993. Episialin (MUC1) inhibits cytotoxic lymphocyte-target cell interaction. J. Immunol. 151:767-776.
van Klinken, J.-W., J. Dekker, H. A. Buller, and A. W. C. Einerhand. 1995. Mucin gene structure and expression: protection vs. adhesion. Am. J. Physiol. 269:G613-G627.
Webb, B. A. 1998. Polydnavirus biology, genome structure, and evolution, p. 105-139. In L. K. Miller and A. L. Ball (ed.), The insect viruses. Plenum Press, New York, N.Y.
Webb, B. A., and S. Luckhart. 1996. Factors mediating short-term and long-term immunosuppression in a parasitized insect. J. Insect Physiol. 42:33-40.
Webb, B. A., N. E. Beckage, Y. Hayakawa, P. J. Krell, B. Lanzrein, M. R. Strand, D. B. Stoltz, and M. D. Summers. 2000. Polydnaviridae, p 253-259. In M. H. V. Regenmortel et al. (ed.), Virus taxonomy. Academic Press, San Diego, Calif.
Webb, B. A., and M. R. Strand. 2004. The biology and genomics of polydnaviruses. In L. I. Gilbert, I. Iatrou, and S. S. Gill (ed.), Comprehensive molecular insect science, vol. 6, p. 323-360. Elsevier, San Diego, Calif.
Wessling, J., S. W. van der Valk, H. L. Vos, A. Sonnenberg, and J. Hilkens. 1995. Episialin (MUC1) overexpression inhibits integrin-mediated cell adhesion to extracellular matrix components. J. Cell Biol. 129:255-265.(Markus Beck and Michael R)