Novel Apoptosis Suppressor Apsup from the Baculovirus Lymantria dispar Multiple Nucleopolyhedrovirus Precludes Apoptosis by Preventing Proteolytic Processing of Initiator Caspase Dronc

Abstract

We previously identified a novel baculovirus-encoded apoptosis suppressor, Apsup, from the baculovirus Lymantria dispar multiple nucleopolyhedrovirus (LdMNPV). Apsup inhibits the apoptosis of L. dispar Ld652Y cells triggered by infection with p35-defective Autographa californica MNPV (vAcΔp35) and exposure to actinomycin D or UV light. Here, we examined the functional role of Apsup in apoptosis regulation in insect cells. Apsup prevented apoptosis and the proteolytic processing of L. dispar initiator caspase Dronc (Ld-Dronc) in Ld652Y cells triggered by overexpression of Ld-Dronc, LdMNPV inhibitor-of-apoptosis 3 (IAP3), or Hyphantria cunea MNPV IAP1. In vAcΔp35-infected apoptotic Ld652Y cells, Apsup restricted apoptosis induction and prevented processing of endogenous Ld-Dronc. Conversely, upon RNA interference (RNAi)-mediated silencing of apsup, LdMNPV-infected Ld652Y cells, which typically support high-titer virus replication, underwent apoptosis, accompanied by the processing of endogenous Ld-Dronc. Furthermore, endogenous Ld-Dronc coimmunoprecipitated with transiently expressed Apsup, indicating that Apsup physically interacts with Ld-Dronc. Apsup prevented the apoptosis of Sf9 cells triggered by vAcΔp35 infection but did not inhibit apoptosis or activation of caspase-3-like protease in vAcΔp35-infected Drosophila melanogaster S2 cells. Apsup also inhibited the proteolytic processing of L. dispar effector caspase Ld-caspase-1 in the transient expression assay but did not physically interact with Ld-caspase-1. These results demonstrate that Apsup inhibits apoptosis in Ld652Y cells by preventing the proteolytic processing of Ld-Dronc. Together with our previous findings showing that Apsup prevents the processing of both overexpressed Ld-Dronc and Bombyx mori Dronc, these results also demonstrate that Apsup functions as an effective apoptotic suppressor in various lepidopteran, but not dipteran, insect cells.

INTRODUCTION

Apoptosis is a genetically and biochemically controlled process of cell death that plays a vital role in the development, tissue homeostasis, and immunity of multicellular organisms by eliminating unwanted and damaged cells (1, 2). In virus-infected animals, early apoptosis contributes to the inhibition of virus replication, whereas late apoptosis functions to promote dissemination of replicated progeny viruses within the infected animal (2).

In baculovirus-infected lepidopteran insect cells, apoptosis constitutes a major intracellular immune mechanism, together with global protein synthesis shutdown (3–6). A number of insect cell lines undergo apoptosis upon infection with specific baculoviruses (5). Among the insect cell lines examined to date, Ld652Y cells from the gypsy moth Lymantria dispar are particularly sensitive to apoptotic stimuli and undergo apoptosis upon infection with various nucleopolyhedroviruses (NPVs), including those from Bombyx mori, Hyphantria cunea, Orgyia pseudotsugata, Spodoptera exigua, and Spodoptera litura (7). Studies on the mechanisms of baculovirus-induced apoptosis collectively indicate that apoptosis is typically induced in insect cells upon baculovirus infection and that in cell lines permissive for productive baculovirus infection, virus-induced apoptosis is precluded by baculovirus-encoded apoptosis suppressors (8–10). Thus, apoptosis suppressors are essential for baculoviruses to survive intracellular antiviral immunity of host cells.

Baculoviruses encode apoptosis suppressor proteins belonging to two different families and include P35, P49, and inhibitor-of-apoptosis (IAP) (6, 11, 12). P35 is an inhibitor substrate for effector caspases that was first identified in Autographa californica multiple NPV (AcMNPV) (13) and has since been shown to have limited distribution among baculovirus members (11). P35 suppresses apoptosis through direct inhibition of the proteolytic activity of effector caspases, which cleave P35 into polypeptide products that irreversibly associate with the effector caspases' active sites (14–18). More recently, a P35 homologue, P49, was identified in Spodoptera littoralis NPV (19, 20) and in several other baculoviruses, including Spodoptera litura MNPV and Leucania separata MNPV (11, 21–24). Unlike P35, however, P49 inhibits the proteolytic activity of both initiator and effector caspases (20, 25, 26), although the overall mechanism of caspase activity inhibition by P49 is similar to that of P35.

IAPs were discovered in O. pseudotsugata MNPV (OpMNPV) and Cydia pomonella granulovirus as apoptosis suppressors in Sf21 and Sf9 cells infected with a mutant AcMNPV lacking a p35 gene (27, 28). Homologues of IAP have been found in nearly all baculoviruses and are classified based on sequence homology into five groups, IAP1 to -5 (29). However, only IAPs from a few members of baculoviruses have been confirmed to function as apoptotic suppressors. Although cellular homologues of baculovirus IAPs have also been identified in various organisms, including yeast, nematodes, insects, and mammals (30, 31), only a few cellular IAPs have the capacity to suppress apoptosis (5, 11). Nonapoptotic cellular IAPs play important regulatory roles in immune responses, signaling, proliferation, and mobility of animal cells (32–35).

Recently, we identified a novel baculovirus apoptosis suppressor protein, Apsup, encoded by L. dispar MNPV (LdMNPV) (10). Apsup is expressed early in infection in LdMNPV-infected L. dispar Ld652Y cells and suppresses apoptosis triggered by infection with p35-defective AcMNPV or exposure to actinomycin D or UV light (10). In the present study, we demonstrated that Apsup precludes apoptosis by preventing the proteolytic processing and activation of the lepidopteran initiator caspase Dronc. Our results also reveal that Apsup inhibits apoptosis of various lepidopteran cells but does not function as an apoptosis suppressor in dipteran Drosophila melanogaster S2 cells.

MATERIALS AND METHODS

Cells, viruses, and infection procedure.

Ld652Y cells from the gypsy moth, L. dispar (36), and Sf9 cells from the fall armyworm, Spodoptera frugiperda (37), were cultured at 28°C in TC100 medium (AppliChem) supplemented with 10% fetal bovine serum (FBS) and 0.26% tryptose broth (Sigma). S2 cells from the fruit fly, D. melanogaster (38), were cultured at 28°C in Schneider's Drosophila Medium (Invitrogen) supplemented with 10% FBS. The viruses used were AcMNPV E2 (AcMNPV; 39) and L. dispar multiple-NPV A21-MPV (LdMNPV; 40). Recombinant viruses vAcΔp35, a p35-defective AcMNPV (41), and vAchrf-1, an AcMNPV harboring the LdMNPV hrf-1 gene (42, 43), were also used.

Virus infection was performed as described previously (44). Time zero of the infection was defined as the time when the viral inoculum was replaced with fresh TC100 medium.

Plasmids.

The expression plasmids pIE1-2/Apsup and pIE1-2/CmycEgfp were constructed for this study. pIE1-2/Apsup, an apsup gene cloned into the expression vector pIE1-2 (Novagen), was generated by inverse PCR using pIE1-2/CmycApsup as the template and primer pair 5′-ATGGCCCGACGACCTGACGCT-3′ and 5′-AGTCACTTGGTTGTTCACGAT-3′ with a KOD-Plus-Mutagenesis kit (Toyobo), followed by self-ligation. pIE1-2/CmycEgfp was constructed by inserting the egfp gene into the NcoI-NotI sites of pIE1-2/Cmyc-mcs (45). The egfp gene was prepared by digesting pIE1-2/Egfp with NcoI and NotI. Construction of plasmids pIE1-2/CmycLdDronc, pIE1-2/HALdCasp1His, pIE1-2/Egfp, pIE1-2/CmycApsup, pIE1-2/HA-ld-iap3, and pIE1-2/HA-hycu-iap1 was described previously (8, 10, 46–48).

Coimmunoprecipitation.

Coimmunoprecipitation was performed using a ProFound c-Myc Tag IP/CoIP kit (Pierce) as described by the manufacturer. Briefly, Ld652Y cells (5 × 10⁵) in 35-mm-diameter culture dishes (Falcon 3001) were transfected with 100 ng pIE1-2/CmycEgfp or 1 μg pIE1-2/CmycApsup and harvested after 24 h. The cells were washed with TBS (25 mM Tris, 0.15 M NaCl, pH 7.2) and then lysed in 350 μl M-PER Mammalian Protein Extraction Reagent (Pierce) containing 1/100 (vol/vol) of Protease Inhibitor Cocktail (Sigma). Aliquots (250 μl) of the cell lysates were reacted with 10 μl agarose-coupled anti-c-Myc antibody (Ab) for 2 h at 4°C in Handee Spin Columns (Pierce). After the columns were washed three times with TBS-T (TBS containing 0.05% Tween 20), immunoprecipitates with the c-Myc tag were eluted with 50 μl of 2× nonreducing buffer. The immunoprecipitates and cell lysates were boiled for 5 min, mixed with a 1/10 (vol/vol) dilution of 1 M dithiothreitol (DTT), and subjected to SDS-polyacrylamide gel electrophoresis and immunoblot analysis.

Immunoblot analysis.

Immunoblot analysis was performed as described previously (8, 44, 48). Briefly, polypeptides were resolved on SDS-polyacrylamide gels and transferred onto Immobilon-P transfer membranes (Millipore) using a Trans-Blot Turbo Blotting System (Bio-Rad). The blots were probed with monoclonal antibodies (MAbs) against hemagglutinin (HA) (Babco), His (Clontech), and c-Myc (Clontech) tags and with polyclonal antibodies against Ld-Dronc, Ld-caspase-1, Apsup, and B. mori NPV (BmNPV) polyhedrin. The anti-Ld-Dronc antibody was raised in a rabbit against Ld-Dronc protein expressed in Escherichia coli (46), according to methods described previously for anti-Bm-Dronc antibody (45). To raise anti-Ld-caspase-1 polyclonal antibody in a rabbit, a mixture of partial amino acid sequences of Ld-caspase-1, 1-MEDESRNNGFEETEQRPNGGGGD-23, 67-EGFNIHSLKSRTGTNVD-83, and 218-YYSWRNTTRGSWFMQALC-235, was used. Antibody against purified BmNPV polyhedrin was prepared in a rabbit as described previously (49, 50). Preparation of polyclonal antibody against Apsup (Ld109) protein has been described previously (10). Horseradish peroxidase-conjugated goat anti-rabbit IgG and goat anti-mouse IgG (Zymed Laboratories) were used as the secondary antibodies. Positive signals were visualized using ECL Western Blotting Detection reagents (GE Healthcare). A BenchMark prestained protein ladder (Invitrogen) was used as the protein size marker.

RNAi-mediated silencing of apsup.

Double-stranded RNAs (dsRNAs) against apsup and egfp were synthesized using a MEGAscript T7 kit (Ambion) as described previously (10, 45). The synthesized dsRNA was introduced into Ld652Y cells by transfection.

Caspase-3-like protease activity assay.

Monolayer cell cultures were scraped into the culture medium with a rubber policeman, and the cells were harvested by centrifugation at 2,000 × g for 2 min at 4°C. The harvested cells were suspended in cell lysis buffer (Medical & Biological Laboratories) and incubated on ice for 10 min. The resultant cell lysates were centrifuged at 12,000 × g for 3 min at 4°C, and the obtained supernatants were used for the assay of caspase-3-like protease activity. Briefly, the supernatants (12.5 μl) were mixed with the same volume of substrate solution containing 10 mM DTT and 0.1 mM Ac-DEVD-MCA (Peptide Institute Inc.) in 2× reaction buffer (Medical & Biological Laboratories) and incubated at 37°C for 30 min. Accumulation of fluorescent product was monitored with a Fluoroskan Ascent Microplate Fluorometer (Thermo Labsystems) using an excitation wavelength of 360 nm and an emission wavelength of 460 nm.

Transfection.

Transfection of expression plasmids and dsRNAs was performed using Lipofectin (Invitrogen) for Ld652Y and Sf9 cells and Cellfectin II (Invitrogen) for S2 cells as described previously (48, 51).

RESULTS

Apsup prevents processing of caspases in the transient expression assay.

We previously demonstrated that the transient expression of an L. dispar effector caspase, Ld-caspase-1, triggers the apoptosis of lepidopteran and dipteran insect cells, including Ld652Y cells (47). To determine whether Apsup suppresses Ld-caspase-1-induced apoptosis, Ld652Y cells were cotransfected with pIE1-2/HALdcasp1His, which expresses Ld-caspase-1 with an N-terminal HA tag and a C-terminal His tag, and either the enhanced green fluorescent protein (EGFP) expression vector pIE1-2/Egfp or pIE1-2/CmycApsup, which expresses Apsup with an N-terminal c-Myc tag. Ld652Y cells cotransfected with both pIE1-2/HALdcasp1His and pIE1-2/Egfp underwent apoptosis (Fig. 1A) accompanied by the activation of caspase-3-like protease (Fig. 1B). In contrast, transiently expressed Apsup inhibited apoptosis and the activation of caspase-3-like protease triggered by Ld-caspase-1 (Fig. 1A and B). Immunoblot analysis of Ld652Y cells transiently expressing EGFP using anti-HA-tag MAb showed that full-length Ld-caspase-1 was detected at a high level only at 24 h posttransfection (Fig. 1C, left panel), indicating that nearly all transiently expressed Ld-caspase-1 underwent proteolytic processing releasing the HA-tagged N-terminal prodomain by 48 h posttransfection. In Ld652Y cells expressing Apsup, full-length Ld-caspase-1 was observed at both 24 and 48 h posttransfection (Fig. 1C, left panel), suggesting that Apsup prevented the proteolytic processing of Ld-caspase-1 and activation of caspase-3-like protease (Fig. 1B).

Fig 1.

Apsup prevents apoptosis, caspase-3-like protease activation, and proteolytic processing of transiently expressed Ld-caspase-1 in Ld652Y cells. Ld652Y cells were cotransfected with 1 μg of pIE1-2/HALdCasp1His (Ld-caspase-1), which expresses Ld-caspase-1 protein harboring HA and His tags at the N and C termini, respectively, and 1 μg of either pIE1-2/Egfp (EGFP) or pIE1-2/CmycApsup (Apsup), which expresses EGFP or Apsup with a c-Myc tag at the N terminus. Ld652Y cells transfected with pIE1-2/Egfp alone were used as a control. (A) Apoptosis. Apoptosis was examined at 48 h postcotransfection. The scale bar represents 30 μm. (B) Caspase-3-like protease activity at 0, 24, and 48 h postcotransfection. The vertical bars indicate standard deviations of averages of results from three determinations. (C) Ld-caspase-1 expression. At 0, 24, and 48 h postcotransfection (h p.t.), proteins from cotransfected Ld652Y cells were subjected to immunoblot (IB) analysis using anti-HA-tag (left panel) or anti-His-tag (right panel) MAbs. The numbers on the left of the left panel indicate the molecular masses (kDa) of the marker proteins. The positions of full-length Ld-caspase-1 and its cleaved forms are indicated on the right of the panels. (D) Expression of Apsup. Proteins from cotransfected Ld652Y cells were examined at 0, 24, and 48 h postcotransfection by immunoblot analysis using anti-c-Myc MAb.

Consistent with the results described above for anti-HA-tag MAb, immunoblot analysis with anti-His-tag MAb detected two polypeptides, corresponding to full-length Ld-caspase-1, which contains a N-terminal prodomain followed by a large subunit (P20) and a small subunit (P10), and prodomain-defective Ld-caspase-1, which consists of P20 and P10 subunits, at 24 h posttransfection in the Ld652Y cells expressing EGFP. By 48 h posttransfection, prodomain-defective Ld-capsase-1 and a small polypeptide consistent with the p10 domain were observed, and full-length Ld-caspase-1 was no longer detected (Fig. 1C, right panel). In contrast, Ld652Y cells coexpressing Ld-caspase-1 and Apsup exhibited full-length Ld-caspase-1 only at 24 h posttransfection (Fig, 1C, right panel). At 48 h posttransfection, a relatively high level of prodomain-defective Ld-caspase-1 was observed in addition to full-length Ld-caspase-1, but the C-terminal p10 domain polypeptide was undetectable (Fig. 1C, right panel), indicating that the prodomain of Ld-caspase-1 was cleaved without further proteolytic processing in the presence of overexpressed Apsup. In the cotransfected Ld652Y cells, we confirmed that Apsup was expressed at high levels at both 24 and 48 h posttransfection (Fig. 1D).

We previously demonstrated that transiently expressed Ld-Dronc undergoes proteolytic processing and triggers apoptosis in several lepidopteran insect cell lines (46, 52). Here, to determine whether Apsup prevented processing of Ld-Dronc in the transient expression assay, Ld652Y cells were cotransfected with pIE1-2/CmycLdDronc, which expresses Ld-Dronc with an N-terminal c-Myc tag, and either pIE1-2/Egfp or pIE1-2/Apsup. Cotransfection of Ld652Y cells with pIE1-2/CmycLdDronc and pIE1-2/Egfp resulted in Ld-Dronc-triggered apoptosis (Fig. 2A) and caspase-3-like protease activation (Fig. 2B). In contrast, Ld652Y cells cotransfected with pIE1-2/CmycLdDronc and pIE1-2/Apsup did not undergo apoptosis (Fig. 2A) and had no detectable caspase-3-like protease activity (Fig. 2B). Immunoblot analysis demonstrated that in the Ld652Y cells expressing Apsup, full-length Ld-Dronc accumulated at markedly elevated levels and did not undergo significant proteolytic processing. These results contrasted with those of EGFP-expressing Ld652Y cells, in which mainly cleaved Ld-Dronc consisting of prodomain and p20 domain was observed and only small amounts of full-length Ld-Dronc were detected (Fig. 2C). In the Ld652Y cells cotransfected with pIE1-2/CmycLdDronc and pIE1-2/Apsup, high levels of Apsup were expressed at 24 and 36 h posttransfection (Fig. 2D). Together, these results indicated that Apsup precluded the apoptosis of Ld652Y cells by preventing the proteolytic processing and activation of Ld-Dronc.

Fig 2.

Apsup prevents apoptosis, caspase-3-like protease activation, and proteolytic processing of transiently expressed Ld-Dronc in Ld652Y cells. Ld652Y cells were transfected with 1 μg of pIE1-2/CmycLdDronc (Ld-Dronc) and 1 μg of either pIE1-2/Egfp (EGFP) or pIE1-2/CmycApsup (Apsup). (A) Apoptosis. Apoptosis was examined at 48 h postcotransfection. (B) Caspase-3-like protease activity. Caspase-3-like protease activity was assayed at the indicated postcotransfection times. (C) Immunoblot analysis of Ld-Dronc expression using anti-c-Myc MAb. (D) Immunoblot analysis of Apsup expression using anti-Apsup polyclonal antibody. For details, see the legend to Fig. 1.

Apsup prevents processing of endogenous Ld-Dronc in vAcΔp35-infected Ld652Y cells.

To determine whether Apsup also prevents the processing and activation of endogenous Ld-Dronc, Ld652Y cells were first transfected with pIE1-2/Egfp or pIE1-2/CymycApsup and at 24 h posttransfection, the cells were infected with vAcΔp35 to induce apoptosis. The vAcΔp35-infected Ld652Y cells expressing EGFP underwent extensive apoptosis, whereas only a few vAcΔp35-infected, Apsup-expressing Ld652Y cells exhibited apoptotic morphology (Fig. 3A). Immunoblot analysis using anti-Ld-Dronc polyclonal antibody detected two polypeptides, corresponding to full-length Ld-Dronc and cleaved Ld-Dronc consisting of the prodomain and p20 domain, at high levels in Ld652Y cells expressing EGFP from 24 h posttransfection onward (Fig. 3B). This result indicated that the activation of Ld-Dronc is responsible for apoptosis induction in vAcΔp35-infected Ld652Y cells. In contrast, although full-length Ld-Dronc accumulated at high levels in Ld652Y cells expressing Apsup, cleaved Ld-Dronc was detected only at very low levels from 24 h postinfection. The levels of full-length Ld-Dronc in EGFP-expressing Ld652Y cells decreased gradually during vAcΔp35 infection, whereas they remained at high levels in Apsup-expressing Ld652Y cells throughout the posttransfection period, similar to the results seen with Ld652Y cells transfected with pIE1-2/Egfp alone (Fig. 3B).

Fig 3.

Apsup prevents apoptosis and proteolytic processing of endogenous Ld-Dronc in Ld652Y cells infected with vAcΔp35. Ld652Y cells were first transfected with pIE1-2/Egfp (EGFP) or pIE1-2/CmycApsup (Apsup), and at 24 h posttransfection, the cells were infected with vAcΔp35 at an MOI of 0.1. The transfected and infected cells were examined for apoptosis (indicated by arrowheads) at 24 h post-vAcΔp35 infection (A) and for Ld-Dronc processing by immunoblot analysis using anti-Ld-Dronc polyclonal antibody at the indicated times postinfection (B). Ld652Y cells transfected with pIE1-2/Egfp alone were used as a control. The scale bar in panel A represents 30 μm. In panel B, molecular masses (kDa) of the marker proteins are indicated on the left, and the positions of full-length Ld-Dronc and cleaved Ld-Dronc consisting of the prodomain and p20 subunit are indicated on the right.

Apsup prevents processing of Ld-Dronc in LdMNPV-infected Ld652Y cells.

Infection of Ld652Y cells with AcMNPV results in global translation arrest, whereas vAchrf-1, a recombinant AcMNPV harboring the LdMNPV hrf-1 and original p35 genes, and LdMNPV replicate to high titers and produce large numbers of polyhedra in Ld652Y cells (5, 6). Apoptotic cells were not observed after infection with each of these viruses. Immunoblot analysis with anti-Ld-Dronc polyclonal antibody showed that endogenous Ld-Dronc underwent proteolytic processing upon infection of Ld652Y cells with AcMNPV or vAchrf-1 (Fig. 4A and B), indicating that the baculovirus-induced apoptosis was precluded by AcMNPV-encoded P35, the substrate inhibitor of effector caspases (15, 18, 26). In vAchrf-1-infected Ld652Y cells, the levels of both full-length and cleaved Ld-Dronc decreased gradually from 24 to 120 h postinfection (Fig. 4B), likely due either to apoptosis induction late in the infection cycle (51, 53) or to the shutdown of host protein synthesis, which is often observed during progeny virus replication (54–56). In contrast, cleaved Ld-Dronc corresponding to prodomain and p20 domain was not detected in LdMNPV-infected Ld652Y cells (Fig. 4C), suggesting that LdMNPV-encoded Apsup prevented the processing of Ld-Dronc and inhibited apoptosis. However, a polypeptide of approximately 24 kDa was specifically detected in the LdMNPV-infected Ld652Y cells from 24 to 96 h postinfection and its numbers subsequently decreased gradually to a negligible level at 120 h postinfection (Fig. 4C). No further characterization of the 24-kDa polypeptide was performed in the present study.

Fig 4.

Proteolytic processing of Ld-Dronc in Ld652Y cells infected with various NPVs. Ld652Y cells were mock infected or infected with AcMNPV (A), vAchrf-1 (B), or LdMNPV (C). At 0, 24, 48, 72, 96, and 120 h postinfection, Ld-Dronc was examined by immunoblotting using anti-Ld-Dronc polyclonal antibody. The molecular masses (kDa) of the marker proteins are indicated to the left and the positions of full-length and cleaved Ld-Dronc are indicated to the right of the panels.

To confirm the speculated mode of action of Apsup, Ld652Y cells were transfected with dsRNA against egfp or apsup and at 24 h posttransfection, the transfected cells were then infected with LdMNPV. The apsup-silenced Ld652Y cells underwent apoptosis accompanied by the activation of caspase-3-like protease upon infection with LdMNPV, whereas the egfp-silenced Ld652Y cells exhibited no detectable apoptotic morphology or caspase-3-like protease activity (data not shown) following LdMNPV infection and produced a number of polyhedra, in accordance with the previous results (10). Immunoblot analysis showed that cleaved Ld-Dronc was first detected in apsup-silenced Ld652Y cells at 24 h postinfection as a weak but clear band that increased in intensity at 48 and 72 h postinfection, whereas cleaved Ld-Dronc was not detected in egfp-silenced Ld652Y cells, even at 72 h postinfection (Fig. 5A). Apsup was detected only in egfp-silenced LdMNPV-infected Ld652Y cells (Fig. 5B), similar to the results of a previous study (10). These results provide evidence that Apsup precludes apoptosis by preventing the proteolytic processing and activation of Ld-Dronc in LdMNPV-infected Ld652Y cells.

Fig 5.

RNAi-mediated silencing of apsup expression promotes processing of endogenous Ld-Dronc in LdMNPV-infected Ld652Y cells. Ld652Y cells were transfected with 1 μg of dsRNA against egfp or apsup. At 24 h posttransfection, the cells were mock infected (mock) or infected with LdMNPV at an MOI of 1 (LdMNPV). At 0, 24, 48, and 72 h postinfection, the cells were examined for proteolytic processing of Ld-Dronc expression (A) and Apsup expression (B) by immunoblot analysis using polyclonal antibodies against Ld-Dronc and Apsup, respectively. In panel A, full-length and cleaved Ld-Droncs are indicated. In panel B, Apsup is indicated by white dots.

Apsup prevents processing of endogenous Ld-Dronc triggered by baculovirus IAPs.

We previously found that IAP1s from diverse NPVs and IAP2 and IAP3 from LdMNPV trigger apoptosis when transiently expressed in several lepidopteran insect cells, accompanied by the activation of caspase-3-like protease and proteolytic processing of endogenous Dronc (48, 51). These baculovirus IAPs are unable to suppress apoptosis triggered by baculovirus infections and may trigger apoptosis by replacing the cellular IAPs, which presumably bind to Dronc and negatively regulate the activation of Dronc in nonapoptotic cells (48, 51, 57).

To determine whether Apsup precluded apoptosis triggered by baculovirus IAPs and the proteolytic processing of Ld-Dronc, cotransfection experiments were performed using pIE1-2/HA-ld-iap3 and pIE1-2/HA-hycu-iap1, which express LdMNPV IAP3 (Ld-IAP3) and Hyphantria cunea MNPV IAP1 (Hycu-IAP1), respectively. Ld652Y cells cotransfected with pIE1-2/Egfp and either pIE1-2/HA-ld-iap3 or pIE1-2/HA-hycu-iap1 showed marked activation of caspase-3-like protease (Fig. 6, left panels) and proteolytic processing of endogenous Ld-Dronc, as indicated by the faint but clear polypeptide bands of cleaved Ld-Dronc at 24 and 48 h posttransfection (Fig. 6, right panels). In contrast, in Ld652Y cells cotransfected with pIE1-2/CmycApsup and pIE1-2/HA-ld-iap3 or pIE1-2/HA-hycu-iap1, no accumulation of cleaved Ld-Dronc or significant activation of caspase-3-like protease was observed (Fig. 6). These results indicated that Apsup affected the processing and activation of Ld-Dronc to prevent apoptosis.

Fig 6.

Apsup prevents apoptosis and proteolytic processing of endogenous Ld-Dronc in Ld652Y cells triggered by overexpression of Ld-IAP3 and Hycu-IAP1. Ld652Y cells were cotransfected with 1 μg of pIE1-2/HA-ld-iap3 (Ld-IAP3) (A) or pIE1-2/HA-hycu-iap1 (Hycu-IAP1) (B) and either pIE1-2/Egfp (EGFP) or pIE1-2/CmycApsup (Apsup). At 0, 24, and 48 h postcotransfection, the Ld652Y cells were examined for caspase-3-like protease activity (left panels) and proteolytic processing of endogenous Ld-Dronc by immunoblot analysis using anti-Ld-Dronc antibody (right panels). Ld652Y cells transfected with pIE1-2/Egfp alone were used as a control. The vertical bars in the left panels indicate standard deviations of averages of results from three determinations. In the right panels, molecular masses (kDa) of the marker proteins are indicated on the left, and the positions of bands of full-length and cleaved Ld-Droncs are indicated on the right.

Apsup physically interacts with endogenous Ld-Dronc but not with Ld-caspase-1.

To determine whether Apsup physically interacts with Ld-Dronc and/or Ld-caspase-1 proteins, Ld652Y cells were transfected with the EGFP expression vector pIE1-2/CmycEgfp or pIE1-2/CmycApsup expressing Apsup with an N-terminal c-Myc tag. At 24 h posttransfection, c-Myc-tagged proteins in the cell lysates of transfected cells were immunoprecipitated using agarose-coupled anti-c-Myc antibody. Analysis of cell lysates and immunoprecipitates prepared from Ld652Y cells transfected with pIE1-2/CmycEgfp or pIE1-2/CmycApsup using anti-Ld-Dronc and anti-Ld-caspase-1 antibodies showed that Ld-Dronc, but not Ld-caspase-1, coimmunoprecipitated with c-Myc-tagged Apsup (Fig. 7). In contrast, neither Ld-Dronc nor Ld-caspase-1 coimmunoprecipitated with c-Myc-tagged EGFP (Fig. 7). These results indicated that Apsup physically interacts with Ld-Dronc, but not with Ld-caspase-1, in the transfected Ld652Y cells.

Fig 7.

Apsup coimmunoprecipitates with endogenous Ld-Dronc but not with Ld-caspase-1. Ld652Y cells were transfected with 100 ng of pIE1-2/CmycEgfp (EGFP) or 1 μg of pIE1-2/CmycApsup (Apsup). At 24 h posttransfection, the transfected cells were harvested and lysed using M-PER Mammalian Protein Extraction Reagent (Lysate), and c-Myc-tagged proteins were then immunoprecipitated using agarose-coupled anti-c-Myc antibody (IP: anti-cmyc). Polypeptides were resolved on 9% SDS-polyacrylamide gels, blotted onto Immobilon-P membranes, and probed with polyclonal antibodies against Ld-Dronc and Ld-caspase-1. Apsup and EGFP with a N-terminal c-Myc tag were probed using MAb against a c-Myc tag. The positions of polypeptide bands corresponding to c-Myc-Apsup, c-Myc-EGFP, Ld-Dronc, and Ld-caspase-1 are indicated to the right of the panels. Molecular masses (kDa) of the marker proteins are indicated to the left of the panels.

Apsup precludes apoptosis of Sf9 cells, but not S2 cells, triggered by vAcΔp35 infection.

We previously showed that transiently expressed Apsup inhibits apoptosis and the proteolytic processing of Droncs triggered by both Ld-Dronc and Bm-Dronc in BM-N cells (52), suggesting that the antiapoptotic function of Apsup is not restricted to L. dispar cells. Here, we examined whether Apsup also inhibits apoptosis triggered by vAcΔp35 infection in Sf9 and S2 cells transfected with pIE1-2/Egfp or pIE1-2/CmycApsup. At 24 h posttransfection, Sf9 and S2 cells were infected with vAcΔp35 at multiplicities of infections (MOI) of 0.1 and 1 PFU, respectively, and were then examined for apoptosis at 72 and 36 h postinfection, respectively. Apsup inhibited apoptosis of vAcΔp35-infected Sf9 cells, with some of the infected cells producing polyhedrin and polyhedra (Fig. 8A). In contrast, vAcΔp35-infected S2 cells expressing Apsup underwent apoptosis, similar to that seen with EGFP-expressing S2 cells (Fig. 8B, left panel), and activation of caspase-3-like protease occurred similarly in Apsup- and EGFP-expressing S2 cells (Fig. 8B, right panel).

Fig 8.

Apsup prevents the apoptosis of Sf9 cells, but not S2 cells, triggered by infection with vAcΔp35. Sf9 cells (A) and S2 cells (B) were transfected with 1 μg of pIE1-2/Egfp (EGFP) or pIE1-2/CmycApsup (Apsup) and at 24 h posttransfection, transfected cells were then infected with vAcΔp35 at an MOI of 0.1 (Sf9) or 1 (S2). At 72 and 36 h postinfection, Sf9 and S2 cells, respectively, were examined for apoptosis (left panels). Sf9 cells with polyhedra are indicated by arrows (A, left panel). Polyhedrin accumulation in Sf9 cells at 72 h postinfection was examined by immunoblot analysis using anti-BmNPV polyhedrin antibody (A, right panel), and caspase-3-like protease activity at 0, 12, 24, and 36 h postinfection was examined for S2 cells (B, right panel). The vertical bars in panel B (right) indicate standard deviations of averages of results from three determinations.

DISCUSSION

In this study, we demonstrated that Apsup inhibits apoptosis by preventing the proteolytic processing and activation of the lepidopteran initiator caspase Dronc. The results from a transient expression assay in Ld652Y cells showed that Apsup inhibits apoptosis triggered by Ld-Dronc overexpression by preventing the proteolytic processing of Ld-Dronc and thereby abrogating the activation of caspase-3-like protease. In the vAcΔp35-infected apoptotic Ld652Y cells, Apsup markedly restricted apoptosis induction and prevented the proteolytic processing of endogenous Ld-Dronc and activation of caspase-3-like protease. Conversely, upon RNAi-mediated silencing of apsup expression in LdMNPV-infected Ld652Y cells, which results in the production of a high titer of progeny viruses, the cells undergo apoptosis accompanied by the proteolytic processing of endogenous Ld-Dronc and activation of caspase-3-like protease. Furthermore, endogenous Ld-Dronc coimmunoprecipitated with transiently expressed Apsup, indicating that Apsup and Ld-Dronc interact in LdMNPV-infected Ld652Y cells. In addition, our results demonstrate that Apsup prevents the processing of endogenous Ld-Dronc and activation of caspase-3-like protease during apoptosis triggered by overexpression of certain baculovirus IAPs, which likely replace and release Ld-Dronc from cellular IAP1, further indicating that Apsup acts at the level of Dronc processing and activation in the apoptotic pathway. Together, these results provide strong evidence that Apsup has the capacity to prevent proteolytic processing and activation of Ld-Dronc, thereby precluding the apoptosis triggered by various apoptotic stimuli, including baculovirus infection.

Our transient expression assay results also showed that Apsup prevented the proteolytic processing of Ld-caspase-1, the lepidopteran effector caspase, and activation of caspase-3-like protease, thereby inhibiting the apoptosis triggered by Ld-caspase-1 overexpression. Spontaneous proteolytic processing has been shown to occur for Sf-caspase-1 and Ld-caspase-1 expressed in E. coli cells (47, 58) and insect cells (47, 59). When overexpressed in insect cell cultures, full-length and prodomain-defective Ld-caspase-1s similarly undergo proteolytic processing, triggering apoptosis of Ld652Y cells (47), whereas only prodomain-defective Sf-caspase-1 triggers apoptosis in S. frugiperda Sf21 cells (59), indicating that prodomain-defective Sf-caspase-1 is activated by proteolytic processing in Sf21 cells. However, in the coimmunoprecipitation assays conducted in the present study, evidence was not found for the physical interaction between overexpressed Apsup and endogenous Ld-caspase-1. Furthermore, overexpression of Ld-caspase-1 in Ld652Y cells promoted the proteolytic processing of Ld-Dronc by an as-yet-unknown mechanism (H. Yamada, M. Kobayashi, and M. Ikeda, unpublished data). Thus, it is possible that Apsup indirectly inhibits the proteolytic processing of Ld-caspase-1 by preventing the processing and activation of upstream Ld-Dronc. Further analyses are required to determine whether Apsup directly prevents the proteolytic processing and activation of Ld-caspase-1.

Our results showed that Apsup inhibits the apoptosis of Sf9 cells, but not dipteran S2 cells, triggered by vAcΔp35 infection. In conjunction with the previous finding that Apsup prevents the processing and apoptosis induced by Ld-Dronc and Bm-Dronc when overexpressed in Ld652Y cells (52), our present results suggest that Apsup functions as an apoptosis suppressor not only in Ld652Y cells but also in other lepidopteran insect cells. Notably, however, Apsup does not exhibit antiapoptotic activity in dipteran S2 cells. Analogous results have been reported for OpMNPV IAP3 (Op-IAP3), which effectively prevents apoptosis and caspase processing in lepidopteran insect cells (27, 59–61) but is unable to suppress apoptosis in Drosophila S2 or DL-1 cells triggered by exposure to actinomycin D or UV light or by infection with an AcMNPV lacking p35 (26, 62). These facts suggest that the apoptosis suppressor Apsup, as well as Op-IAP3, from lepidopteran baculoviruses coevolved with lepidopteran cellular components and mechanisms involved in apoptosis regulation. Recently, we found that Bm-Dronc and Ld-Dronc undergo proteolytic processing, but do not trigger apoptosis, when overexpressed in S2 cells (45, 46). This finding indicates that although lepidopteran and Drosophila cells have undergone functional divergence with respect to the specificity of proteins functioning in apoptotic pathways, the caspase cascade that regulates apoptosis is evolutionarily conserved (63).

The molecular mechanisms by which Apsup prevents the proteolytic processing and activation of Ld-Dronc are not known. Apsup is a 336-amino-acid (aa) polypeptide and has no characteristic domains suggestive of possible function (10). Homologues of Apsup have been found in a number of insect-specific viruses, including NPVs, granuloviruses, and ascoviruses, although the amino acid sequence identities are less than 38.2%, with the exception of a homologue in Lymantria xylina MNPV that shares 87.6% amino acid sequence identity with Apsup. None of these Apsup homologues has been functionally characterized, except for an Apsup homologue in AcMNPV (Ac-Apsup). Ac-Apsup is a 258-aa protein that contains a truncated C terminus and is unable to suppress apoptosis of Ld652Y cells triggered by actinomycin D or UV light (10). Recently, we found that Ac-Apsup containing the C-terminal 79-aa residues from LdMNPV Apsup (Ld-Apsup) functions as an apoptotic suppressor in Ld652Y cells overexpressing Ld-Dronc, while Ld-Apsup lacking the C-terminal 79-aa residues has no antiapoptotic activity, indicating that these C-terminal residues of Ld-Apsup are indispensable for its antiapoptotic activity (Yamada et al., unpublished). Further structure-function analyses of both truncated and point mutants of Apsup are under way in our laboratory.