A mechanism of cell death involving an adenylyl cyclase/PKA signaling pathway is induced by the Cry1Ab toxin of Bacillus thuringiensis

Abstract

Many pathogenic organisms and their toxins target host cell receptors, the consequence of which is altered signaling events that lead to aberrant activity or cell death. A significant body of literature describes various molecular and cellular aspects of toxins associated with bacterial invasion, colonization, and host cell disruption. However, there is little information on the molecular and cellular mechanisms associated with the insecticidal action of Bacillus thuringiensis (Bt) Cry toxins. Recently, we reported that the Cry1Ab toxin produced by Bt kills insect cells by activating a Mg²⁺-dependent cytotoxic event upon binding of the toxin to its receptor BT-R₁. Here we show that binding of Cry toxin to BT-R₁ provokes cell death by activating a previously undescribed signaling pathway involving stimulation of G protein (G_αs) and adenylyl cyclase, increased cAMP levels, and activation of protein kinase A. Induction of the adenylyl cyclase/protein kinase A pathway is manifested by sequential cytological changes that include membrane blebbing, appearance of ghost nuclei, cell swelling, and lysis. The discovery of a toxin-induced cell death pathway specifically linked to BT-R₁ in insect cells should provide insights into how insects evolve resistance to Bt and into the development of new, safer insecticides.

Cell death appears to involve a relatively limited number of evolutionarily conserved mechanisms associated with cell signaling pathways, which, until provoked, lie inactive (1, 2). These tightly regulated pathways govern a variety of cellular activities critical to the survival and fate of the cell. Interference with the regulation of cell signaling can alter gene expression and bring about such structural and functional disarray within the cell that it may divide abnormally or die. Particular ligand–receptor complexes are known to generate specific cellular signals essential to differentiation, proliferation, pathogen recognition, immune response, and cell death (3, 4). Aberration in ligand–receptor interactions are implicated in a number of disease states such as cancer, chronic inflammation, and various allergies. Interestingly, many pathogenic organisms and their toxins target host cell receptors, the consequence of which is altered signaling events that lead to aberrant activity or cell death (5–7).

The Cry toxin proteins produced by the soil bacterium Bacillus thuringiensis (Bt) represent >100 phylogenetically related toxins, which specifically target insects and nematodes but not mammals (8). The specificity of Cry toxins depends on individual cell surface receptors, which represent a family of cadherins expressed in the midgut epithelium of various insects (9–13). Cytotoxicity and cell death are the direct result of univalent binding of a Cry toxin monomer to its respective cadherin receptor (14). Impeding the toxin–receptor interaction by receptor modification has been linked to the development of resistance to Cry toxins (12). However, recent studies reveal that neither resistance nor cytotoxicity can be explained solely by toxin binding. For example, both the number of Cry toxin binding receptors and the affinity of toxin to receptor were similar in the brush border membrane vesicles isolated from resistant and susceptible European corn borer larvae (15). Removing Mg²⁺ by EDTA completely blocks Cry1Ab toxin-induced cell death of cabbage looper cells but does not interfere with toxin–receptor binding (14). Evidently, the interaction of toxin with the receptor is prerequisite, but not sufficient, to induce cell death.

Until very recently, Cry proteins were believed to be pore-forming toxins that kill cells by osmotic lysis. Changes in membrane permeability were correlated with the incorporation of Cry toxin oligomers into lipid bilayer rafts and brush border membrane vesicles (8, 16). However, studies of mutated Cry toxin proteins have shown that neither the toxin oligomer complex nor commensurate changes in membrane vesicle permeability correlate directly with toxicity (17–19). Interestingly, Cry toxin oligomers also are incorporated into the cell membrane of nonsusceptible cabbage looper cells and are carried by the cells for several generations with no adverse effect (14). Apparently, toxin action is much more complicated than simply osmotic lysis. Recently, we discovered that the Cry1Ab toxin produced by Bt kills insect cells by activating a Mg²⁺-dependent cytotoxic event upon binding of the toxin to its receptor BT-R₁ (14). Here, we report that such binding provokes cell death in insect cells by activating a previously undescribed signaling pathway involving stimulation of the stimulatory G protein α-subunit (G_αs) and adenylyl cyclase (AC), increased cyclic adenosine monophosphate (cAMP) levels, and activation of protein kinase A (PKA). Activation of the AC/PKA signaling pathway initiates a series of cytological events that include membrane blebbing, appearance of nuclear ghosts, and cell swelling followed by cell lysis. That Cry toxins of Bt aggravate critical intracellular signaling pathways through receptor-coupled interactions has implications in insecticide and drug discovery (20).

Results

Cytological Changes Associated with the Progression of Cry1Ab Toxin-Induced Cell Death.

Cultured High Five (H5) cells, which originated from ovarian cells of the cabbage looper, Trichoplusia ni, were not affected by the Cry1Ab toxin because they do not express a receptor for the toxin. However, heterologous expression of the cadherin receptor BT-R₁ in transfected H5 cells (S5) rendered them susceptible to the toxin (14). In the present work, we examine the morphological changes associated with toxin-treated S5 cells. As seen in Fig. 1A Left Upper, toxin-treated S5 cells exhibited dramatic cytological changes including altered size, shape, and overall appearance compared with untreated viable cells that remained compact and uniformly round (Fig. 1A Left Lower). Time-lapse microscopy showed that S5 cells underwent sequential cytological changes upon toxin exposure (Fig. 1B). Cytotoxicity and cell death involved two rather distinct sequential events or stages: (i) membrane blebbing and ruffling that occurred within 20 min after toxin exposure (Fig. 1B Center) and (ii) cell swelling and lysis within 40 min (Fig. 1B Right), apparently the result of increased membrane permeability. Movie 1, which is published as supporting information on the PNAS web site, shows the sequence of morphological changes.

Fig. 1.

Cytological changes associated with the progression of Cry1Ab toxin-induced cell death. (A) S5 cells exposed to toxin in the presence of EDTA (Center Lower), EDTA + Ca²⁺ (Right Lower), EGTA (Center Upper), and EDTA + Mg²⁺ (Right Upper). S5 cells treated with Cry1Ab toxin (Left Upper) and untreated S5 cells (Left Lower). (B) Sequence of cytological changes during the course of toxin-induced cell death as viewed by phase-contrast microscopy. The long arrow beneath the photographs indicates the relative time for each stage of cell death, i.e., toxin binding, membrane blebbing, and cellular swelling. The symbols below the time line show the points at which cell death can be blocked by EDTA and cell swelling, but not cell death, can be blocked by raffinose (Raf). A time-lapse movie is provided as Movie 1. (C) Effect of osmotic protectants raffinose (Left), sucrose (Suc, Center), and glucose (Glu, Right) on membrane blebbing and cellular swelling. (Scale bars: 15 μm.)

Because the two cytological stages were distinct, we wanted to understand how membrane blebbing and cellular swelling relate to the progression of cytotoxicity and cell death. To do so, we first examined the effect of osmotic protectants on the swelling of S5 cells induced by the Cry1Ab toxin (Fig. 1C). Osmotic protectants such as glucose, sucrose, and raffinose can counter osmotic pressure produced by a drastic ion flux across cell membrane (21, 22). The protectants prevent cell swelling only when their molecular size is larger than the active ion channels in the cell membrane. The molecular diameters of raffinose, sucrose, and glucose are ≈1.2–1.4, 0.9, and 0.7 nm, respectively. In the presence of raffinose (30 mM), toxin-exposed cells were arrested in the blebbing stage and did not swell (Fig. 1C Left). Sucrose (30 mM) partially prevented cell swelling (Fig. 1C Center), whereas glucose (30 mM) did not interfere with either cell blebbing or swelling (Fig. 1C Right). Although raffinose-treated S5 cells did not undergo swelling after exposure to the Cry1Ab toxin, these cells eventually died, as did those cells treated with sucrose and glucose. Apparently, cytotoxicity is related to certain cellular events upstream of the swelling stage.

Previously, we observed that Cry1Ab cytotoxicity is Mg²⁺-dependent (14). To determine the effect of Mg²⁺ on the cytological changes of cells exposed to Cry1Ab toxin, we microscopically monitored cellular responses to Cry1Ab toxin in the presence of the divalent cation chelators EDTA and EGTA. EDTA, but not EGTA, prevented toxin-induced cellular responses, including both membrane blebbing and cell swelling, in S5 cells (Fig. 1A Center Lower and Center Upper, respectively). EDTA chelates Mg²⁺ and Ca²⁺, whereas EGTA preferentially chelates Ca²⁺. Addition of Mg²⁺ (5 mM) to EDTA-treated cells restored toxin-induced cellular blebbing and swelling (Fig. 1A Right Upper), whereas Ca²⁺ (5 mM) had no such effect (Fig. 1A Right Lower). Removal of Mg²⁺ by EDTA prevented the S5 cells from blebbing (Fig. 1A Center Lower) and inhibited cell death completely. Evidently, a Mg²⁺-dependent intracellular pathway is established upstream of the blebbing stage (within 5 min after toxin binding to BT-R₁ on S5 cells; see Fig. 1B Left) and is critical to toxin-induced cell death.

Effect of Endocytosis and Apoptosis Inhibitors on Cry1Ab Toxin-Induced Cell Death.

A number of bacterial toxins kill target cells by receptor-mediated endocytosis and disruption of essential cytosolic functions (23, 24). Some pathogenic bacteria also produce pore-forming toxins or protein synthesis inhibitors that are associated with apoptosis in the target cells (6, 25). To determine whether Cry1Ab toxicity involves endocytosis, we studied the effect of several endocytosis inhibitors on toxin action. The inhibitors included nocodazole, cytochalasin D, phenylarsine oxide, and bafilomycin A (Table 1). S5 cells, treated with each inhibitor before toxin addition, underwent the same morphological changes (blebbing and swelling) as those cells not exposed to the inhibitors (data not shown). Furthermore, none of the inhibitors precluded cell death as indicated by trypan blue exclusion analysis (Table 1). It is noteworthy that direct delivery of Cry1Ab toxin into receptor-free H5 cells by the protein delivery reagent Chariot did not induce death (data not shown). As was shown previously by fluorescence microscopy (14), the toxin is not internalized at either stage of cytotoxicity (blebbing or swelling). We also investigated the effect of specific protease inhibitors on S5 cells treated with the toxin. The broad-spectrum caspase inhibitor z-VAD-fmk, which prevents apoptosis in many cell types (26), did not suppress Cry1Ab-induced toxicity and cell death (Table 1). Likewise, the serine protease inhibitors Pefabloc and l-1-chloro-3-[4-tosylamido]-4-phenyl-2-butanone (TPCK) and the cathepsin inhibitor E-64 (27) had no effect. The S5 cells treated with any of the inhibitors underwent cell death upon exposure to Cry1Ab in a manner similar to cells exposed to toxin alone (Table 1). In addition, we did not observe externalization of phosphatidylserine or DNA fragmentation in toxin-treated cells (data not shown). Evidently, the Cry1Ab toxin of Bt does not cause death in insect cells with features of either endocytosis or apoptosis. Rather, the morphological changes observed in Cry1Ab toxin-treated cells (Fig. 1B) are similar to those associated with oncosis, which involves cellular blebbing, swelling, and increased membrane permeability (28).

Table 1.

Characterization of toxin-induced cell death

Inhibitor	Cellular targets	Effect(s)	Cell death*
Control	n.a.†	n.a.†	+
Nocodazole	Microtubule	Endocytosis	+
Cytochalasin D	Actin	Endocytosis	+
Phenylarsine oxide	n.d.‡	Endocytosis	+
Bafilomycin A	H+-ATPases	Endocytosis	+
Z-VAD-fmk	Caspases	Apoptosis	+
Pefabloc	Serine proteases	Apoptosis	+
TPCK	Serine proteases	Apoptosis	+
E-64	Cathepsins	Apoptosis	+
EGTA	Metalloproteins	Ca2+	+
EDTA	Metalloproteins	Mg2+, Ca2+	−

TPCK, l-1-chloro-3-[4-tosylamide]-4-phenyl-2-butanone.

*Cells were switched to inhibitor-free buffer before Cry1Ab toxin treatment. Cell death is the percent of dead cells in the total population as determined by trypan blue staining of nuclei. +, 80 ± 5% cell death; −, no cell death. Data are the mean ± SD of six experiments.

^†n.a., not applicable.

^‡n.d., not determined.

Correlation of Cytotoxicity and Elevated cAMP Levels Stimulated by Cry1Ab.

Second messenger cAMP has been implicated in modulation of signaling related to cell death in a wide variety of cells (29, 30). cAMP-dependent signal transduction involves a multiprotein cascade of which many members are Mg²⁺-dependent. Because Mg²⁺ is a key component in Cry1Ab toxin cytotoxicity and because it is required for cell death (Table 1), we hypothesized that a cAMP pathway is induced by Cry toxin action, involving stimulation of G protein, AC, and PKA. The production of cAMP is controlled by activation of AC, which is stimulated by G_αs (31). cAMP then binds to cytoplasmic PKA, activating the catalytic subunit(s) of the protein that, in turn, phosphorylates effector proteins (29, 32). Mg²⁺ is required for G protein stimulation through GTP exchange and subunit dissociation and ATP binding as well as for catalytic synthesis of cAMP by AC (31, 33, 34).

To test our hypothesis, we used a specific set of compounds that inhibits the activity of these particular signaling molecules (Fig. 2). Because elevation of intracellular cAMP levels is a hallmark of activation of AC/PKA pathway (29), we measured intracellular cAMP production in Cry1Ab toxin-treated S5 cells over time. In toxin-exposed S5 cells, cAMP production increased in a time-dependent manner (Fig. 2A). The magnitude of the cAMP response was similar to that elicited by forskolin (FSK) (Fig. 2B), a direct activator of membrane ACs. Because Mg²⁺ is essential to Cry1Ab-induced cell death (Table 1), we also tested the effect of EDTA on Cry1Ab-induced cAMP levels in S5 cells. Pretreatment of S5 cells with EDTA (5 mM) virtually precluded any increase in cAMP levels in response to either Cry1Ab toxin or FSK treatment (Fig. 2B). EDTA treatment of S5 cells also prevented cell death when cells were exposed to Cry1Ab (results not shown), indicating that increase in cAMP correlates with toxin-induced cell death. Significantly, cAMP levels did not change in receptor-free H5 cells (Fig. 2A), substantiating that Cry1Ab toxin binding to BT-R₁ is critical to the stimulation of cAMP production.

Fig. 2.

Involvement of AC/PKA pathway in Cry1Ab toxin-induced cytotoxicity. (A) Stimulation of cAMP production by Cry1Ab toxin. Cells were incubated with isobutyl-1-methylxanthine (0.5 mM) before the addition of Cry1Ab, and cAMP was measured by using the Bridge-It cAMP Assay (Mediomics) in S5 and H5 cells (3 × 10⁴). (B) Effect of Cry1Ab, FSK, and EDTA on cAMP production. cAMP levels were measured in S5 cells after 10 min of treatment with either Cry1Ab (Cry, 180 nM) or FSK in the absence or presence of EDTA. (C) Percent relative cytotoxicity of Cry1Ab in the presence of NF023 (1 μM) or NF449 (1 μM). (D) Effect of ddADP on Cry1Ab cytotoxicity. S5 cells were incubated for 30 min with ddADP (0–40 μM), a cell-permeable inhibitor of AC, before the addition of Cry1Ab toxin (60 nM). Cell death was determined by trypan blue exclusion analysis (14). (E) Percent relative cytotoxicity of Cry1Ab in the presence of cell-permeable inhibitors of PKA (H-89, ≈0–50 μM, and PKAI 14–22-amide, ≈0–8 μM) for 30 min at the specified concentration before addition of Cry1Ab toxin. (F) Synergetic effect of pCPT-cAMP (200 μM) and FSK (0.2 μM) on Cry1Ab toxin-induced cell death. The concentration of ddADP was 40 μM. Cry1Ab toxin was administered at 60 nM, the concentration required to kill 50% of S5 cells (14). Data are presented as the mean ± SD of six experiments.

Involvement of G_αs and ACs in Cry1Ab Cytotoxicity.

Membrane-bound ACs are activated by G_αs (31). To ascertain whether G protein activity is involved in the Cry toxin-induced pathway, we used a cell-permeable inhibitor, NF449 (35), which selectively antagonizes G_αs. S5 cells preincubated (30 min) with NF449 (1 μM) were less sensitive to the Cry1Ab toxin than untreated cells (Fig. 2C). In fact, toxicity of Cry1Ab decreased by 50% when S5 cells were incubated with NF449. Another G protein antagonist for inhibitory G protein α-subunit (G_αi), NF023 (36), had no effect on toxicity when used at the same concentration (1 μM) as NF449 (Fig. 2C). We also investigated whether 2′,5′-dideoxy-3′-ADP (ddADP), which interferes with AC activity by blocking substrate utilization by the enzyme, can prevent Cry1Ab-induced cell death. As can be seen in Fig. 2D, ddADP interfered with Cry1Ab cytotoxicity in a dose-dependent manner, dramatically reducing the number of cells undergoing cell death. We conclude from these results that activation of G_αs protein and stimulation of AC, which lead to an increase in production of the second messenger cAMP, are involved directly in toxicity and cell death.

Requirement of PKA Activity in Cry1Ab Toxin-Induced Cell Death.

To determine whether the toxicity of Cry1Ab is mediated by an AC/PKA signaling event, we tested the effects of two potent cell-permeable PKA inhibitors, H-89 and myristoylated amide 14–22 (PKAI 14–22-amide). H-89 is a competitive inhibitor that interferes with the utilization of ATP by PKA (37), whereas PKAI 14–22-amide is a peptide substrate inhibitor (38). The PKA inhibitors were introduced to S5 cells in a preincubation step followed by addition of Cry1Ab toxin. H-89 (50 μM) and PKAI 14–22-amide (8 μM) fully blocked cell death, indicating that inhibition of PKA abolishes Cry toxin action (Fig. 2E). Moreover, the characteristic morphological changes, including membrane blebbing and cellular swelling along with cell death, were completely prevented by PKA inhibition (results not shown), demonstrating that cAMP-dependent PKA is cardinal in Cry toxin action. Indeed, when S5 cells were pretreated with the cAMP analog pCPT-cAMP (200 μM), even in the presence of the AC inhibitor ddADP (40 μM), the number of dead cells increased (75% cell death) relative to toxin-treated cells without ddADP (50% cell death; Fig. 2F). A more dramatic effect (almost 100% cell death) was observed when S5 cells were pretreated with FSK (0.2 μM). Neither FSK nor pCPT-cAMP alone was sufficient to mediate toxicity (Fig. 2F). Possibly, a specific PKA-dependent effector(s) stimulated by Cry1Ab, but not by FSK, mediates downstream cell death activity.

Discussion

In the present work, we have demonstrated that the binding of Cry1Ab toxin to BT-R₁ provokes cytotoxicity via a signaling pathway in insect cells, which involves stimulation of G_αs and AC along with an increased level of cAMP (8- to 10-fold) and activation of PKA (Fig. 2). An increase in the level of intracellular cAMP is a hallmark of activation of cAMP-related signal transduction pathways that can either promote cell death or protect cells from death, depending on the cell type and the triggering stimulus (39). Preventing cAMP production by inhibitors of G_αs (NF449) and AC (ddADP) substantially reduced the Cry1Ab cytotoxicity, whereas the activator (FSK) and potentiator (pCPT-cAMP) of cAMP sensitized the cells and enhanced cytotoxicity. The most common downstream effector of cAMP is PKA (29, 32). Pretreatment of S5 cells with inhibitors of PKA (H89 and PKAI 14–22-amide) protected the cells from Cry toxin action.

The morphological changes brought about by the Cry1Ab toxin occurred in a sequential manner, i.e., membrane blebbing and ruffling along with the appearance of ghost nuclei followed by cell swelling and lysis (Fig. 1B and Movie 1). Cell swelling and lysis occurred within a short time frame (30–40 min after toxin exposure), a phenomenon also observed in other insect cells (9, 40) as well as in mammalian cells transfected with the cDNA of BT-R receptors (11). The morphological changes observed in Cry1Ab toxin-treated cells (Fig. 1B) are strikingly similar to the phenotypic changes associated with oncosis (28), whereas DNA (or nuclear) fragmentation and caspase activation, typical of apoptotic cell death, were not detected.

Interestingly, the sequential cellular events associated with the progression of cell death induced by the Cry1Ab toxin could be interrupted at different stages. EDTA prevented membrane blebbing and cell swelling and protected the cells from death (Fig. 1B). The osmotic protectant, raffinose, arrested the cells at the membrane blebbing stage and prevented cell swelling (Fig. 1C) even though the cells eventually died. Remarkably, the distinctive phenotypic changes associated with Cry1Ab cytotoxicity also can be blocked by inhibiting the AC/PKA signaling pathway. Together, these results support the notion that cell death, occasioned by Cry toxins, is a complex cellular response and is not simply osmotic lysis (14, 41).

In light of these observations, we propose a previously undescribed model for Cry toxin action (Fig. 3). The model portrays a series of events that are confined to or associated with the cell membrane and that are related spatially and temporally to G protein activation, AC stimulation, and cAMP production. Accordingly, the Cry1Ab toxin binds specifically to BT-R₁, stimulating G protein and AC, which brings about accumulation of cAMP and activation of PKA. PKA is the key component in this cell death pathway. Activated PKA alters downstream effectors that, in turn, actually dismantle the cell by destabilizing both the cytoskeleton and ion channels in the cell membrane. Such impairment of the structural and functional integrity of the cell leads to cell death as manifested by membrane blebbing and cellular swelling (Fig. 1B). The model agrees with the paradigm for many bacterial toxins that challenge host cells by targeting cell surface receptors and manipulating critical reactions associated with various cellular responses (5, 42). Certainly, the involvement of an AC/PKA signal pathway in the cytotoxic action of Cry1Ab argues against the previously postulated lytic-pore formation model (8, 16) and explains why the toxin–receptor interaction is prerequisite to cytotoxic action (14, 15). Events that may originate directly from the receptor, before G protein stimulation, or that may link G_αs to effectors other than AC, could be implicated. Nevertheless, the Cry1Ab toxin does activate a pathway involving G protein through BT-R₁, a single-pass membrane receptor that does not resemble classical G protein-coupled receptors.

Fig. 3.

Proposed model for the action of Cry toxin. Cry toxin binds to BT-R and stimulates G protein and AC, which promotes production of intracellular cAMP. In turn, PKA activation destabilizes the cytoskeleton and ion channels, leading to cell death.

Induction of cell death through stimulation of an AC/PKA pathway(s) is a powerful evolutionary strategy for a pathogen to overcome a host (5, 42). Continuous exposure of an insect population to Cry toxins indeed may prompt alterations in midgut cadherins and/or the coupled signaling components, which, in turn, disable the cell death machinery and render an insect resistant to Cry toxin action. Therefore, elucidation of cellular signal cascading involved in Cry toxin action should provide new insights into how insects evolve resistance to Bt and into the development of new, safer insecticides. No equivalent toxicity has been observed in mammals apparently because of the absence of appropriate Cry toxin receptors and because of the strikingly different conditions in the mammalian gastrointestinal tract. However, cell death seems to be evolutionarily conserved with mechanisms involving shared components common among diverse pathways and in different cell types (1, 3, 4). It will be interesting to find out whether the AC/PKA signaling described in this work operates in other cells.

Materials and Methods

Cell Cultures.

H5 cells (Invitrogen) were cultured as a monolayer in 25-cm² tissue culture flasks containing 5 ml of insect-Xpress medium (Cambrex, East Rutherford, NJ) supplemented with gentamycin (10 μg/ml; Sigma). H5 cells stably expressing BT-R₁ cDNA (GenBank accession no. AF319973), designated as S5 cells (14), were maintained in the same medium plus G418 (800 μg/ml; ISC BioExpress, Kaysville, UT).

Preparation of Cry1Ab Toxin.

Cry1Ab toxin was prepared as described (43) by trypsin digestion of the protoxin obtained from parasporal crystals of Bt subsp. berliner. Trypsin-activated Cry proteins were purified by anion-exchange chromatography using a MonoQ HR 10/10 column with an AP-Biotech FPLC system. Quantification of purified Cry toxin protein was performed by the bicinchoninic acid method (Pierce) with BSA (fraction V) as a standard.

Cell Treatment.

Cells were seeded in 96-well plates (1 × 10⁴ cells per well) and allowed to grow attached to the bottom surface of the plate. Cell monolayers were preincubated for 30 min with raffinose (30 mM), glucose (30 mM), sucrose (30 mM), nocodazole (20 μM), cytochalasin D (10 μM), pherylarsine oxide (10 μM), bafilomycin A (200 nM), NH₄Cl (10 mM), z-VAD-fmk (100 μM), pefabloc (300 μM), E-64d (300 μM), l-1-chloro-3-[4-tosylamido]-4-phenyl-2-butanone (TPCK; 300 μM), EDTA (5 mM), EGTA (5 mM), NF023 (1 μM), NF449 (1 μM), 3-isobutyl-1-methylxanthine (IBMX; 0.5 mM), FSK (0.2 μM), ddADP (0–40 μM), H89 (0–50 μM), and PKAI 14–22-amide (0–8 μM), respectively, before the addition of Cry1Ab toxin. For the combined treatments, EDTA-treated cells were incubated with MgCl₂ (5 mM), CaCl₂ (5 mM), and FSK (0.2 μM), respectively, for 15 min; ddADP-treated cells were incubated with pCPT-cAMP (200 μM) for 30 min before cytotoxicity assays were performed.

Assay for Cytotoxicity.

Insect cells were incubated for 1 h in PBS buffer containing Cry1Ab toxin at various concentrations (60–180 nM). Cell death was determined by trypan blue exclusion. Ten microliters of trypan blue (0.4%, wt/vol) was added directly to each well and incubated for 5 min. Stained cells were viewed immediately under a microscope (Nikon TE600), and photomicrographs were taken with an RTE/CCD-1300 camera (Roper Scientific, Trenton, NJ) at 200×. Photomicrographs were analyzed by using imaging software (metamorph 4; Universal Imaging, Downington, PA) to count the number of blue-stained, dead cells (NB) and transparent, viable cells (NT), respectively. Cytotoxicity was calculated by the ratio NB/(NB + NT).

Microscopy.

Time-lapse phase-contrast microscopy was performed with a Nikon TE600 microscope and an RTE/CCD-1300 camera (Roper Scientific) at 200×. After addition of Cry1Ab toxin, real-time images of S5 cells were recorded every 20 s for a 40-min period.

Intracellular cAMP Assay.

Cells (3 × 10⁴) were planted in 96-well plates. After experimental treatment, the cells were harvested in microcentrifuge tubes and washed in buffered saline solution twice, followed by 5 min of centrifugation (8,000 × g). Then, 100 μl of assay solution (Bridge-It; Mediomics, St. Louis, MO) was added to the cell pellet and vortexed for 1 s. The contents of each tube were transferred to a single well of a black polypropylene 96-well plate and incubated for 30 min at room temperature. The intensity of fluorescence (excitation, 480 nm; emission, 520 nm) was read with a fluorescence plate reader (DTX 880; Beckman Coulter). The cAMP concentration was determined based on a standard curve.

Abbreviations

AC

adenylyl cyclase

Bt

Bacillus thuringiensis

ddADP

2′,5′-dideoxy-3′-ADP

FSK

forskolin

G_αs

stimulatory G protein α-subunit

H5

High Five

S5

BT-R₁-transfected H5 cells

PKA

protein kinase A.