Effect of Bacillus thuringiensis Toxins on the Membrane Potential of ...

3 downloads 152 Views 400KB Size Report
OLIVIER PEYRONNET,1 VINCENT VACHON,1 ROLAND BROUSSEAU,2 DANICA BAINES,3. JEAN-LOUIS SCHWARTZ,1,2. AND RAYNALD LAPRADE1*.
APPLIED AND ENVIRONMENTAL MICROBIOLOGY, May 1997, p. 1679–1684 0099-2240/97/$04.0010 Copyright © 1997, American Society for Microbiology

Vol. 63, No. 5

Effect of Bacillus thuringiensis Toxins on the Membrane Potential of Lepidopteran Insect Midgut Cells OLIVIER PEYRONNET,1 VINCENT VACHON,1 ROLAND BROUSSEAU,2 DANICA BAINES,3 JEAN-LOUIS SCHWARTZ,1,2 AND RAYNALD LAPRADE1* Groupe de Recherche en Transport Membranaire, Universite´ de Montre´al, Montreal, Quebec H3C 3J7,1 Biotechnology Research Institute, National Research Council, Montreal, Quebec H4P 2R2,2 and Great Lakes Forestry Center, Natural Resources Canada, Sault Ste. Marie, Ontario P6A 5M7,3 Canada Received 24 October 1996/Accepted 18 February 1997

To test whether the ability of Bacillus thuringiensis toxins to form pores in the midgut epithelial cell membrane of susceptible insects correlates with their in vivo toxicity, we measured the effects of different toxins on the electrical potential of the apical membrane of freshly isolated midguts from gypsy moth (Lymantria dispar) and silkworm (Bombyx mori) larvae. In the absence of toxin, the membrane potential, measured with a conventional glass microelectrode, was stable for up to 30 min. It was sensitive to the K1 concentration and the oxygenation of the external medium. Addition of toxins to which L. dispar is highly [CryIA(a) and CryIA(b)] or only slightly [CryIA(c) and CryIC] sensitive caused a rapid, irreversible, and dose-dependent depolarization of the membrane. CryIF, whose toxicity towards L. dispar is unknown, and CryIE, which is at best poorly active in vivo, were also active in vitro. In contrast, CryIB and CryIIIA, a coleopteran-specific toxin, had no significant effect. The basolateral-membrane potential was unaffected by CryIA(a) or CryIC when the toxin was applied to the basal side of the epithelium. In B. mori midguts, the apical-membrane potential was abolished by CryIA(a), to which silkworm larvae are susceptible, but CryIA(b) and CryIA(c), to which they are resistant, had no detectable effect. Although the technique discriminated between active and inactive toxins, the concentration required to produce a given effect varied much less extensively than the sensitivity of gypsy moth larvae, suggesting that additional factors influence the toxins’ level of toxicity in vivo. ertheless been shown to bind with high affinity to its midgut cell brush border membrane (10). Similarly, resistance to B. thuringiensis toxins can in some cases (3, 9, 40), but not in all (6, 12, 25), be attributed to altered receptor binding properties. The interpretation of results concerning the role of altered toxin binding characteristics in resistant insects is further complicated by the fact that the ability of the same strain of CryIA(c)resistant Plutella xylostella to bind this toxin was found to be either almost completely lost (34) or only slightly altered (26) when measured with two different techniques. Toxin binding thus appears to be necessary but not sufficient for toxicity and, therefore, does not provide a reliable index of toxin potency. In the present study, a new and simple in vitro technique was developed to measure the electrical membrane potential of the epithelial cells in freshly isolated lepidopteran larval midguts. The ability of different B. thuringiensis toxins to depolarize the apical membranes of gypsy moth and silkworm larval midguts was compared with their in vivo toxicities. Although only active toxins were able to depolarize the membrane, their in vitro activities varied much less extensively than their relative in vitro toxicities towards the gypsy moth.

Bacillus thuringiensis is the most widely used environmentfriendly alternative to chemical insecticides for the biological control of forest and agricultural pests and vectors of human and animal diseases (22). During sporulation, different strains of this gram-positive bacterium produce crystalline parasporal inclusion bodies, composed of one or more toxic proteins known as d-endotoxins, with a high level of specificity against different species of lepidopteran, dipteran, and coleopteran insects (17) and certain parasitic nematodes and protozoan pathogens (8). Following their ingestion by susceptible insect larvae, these crystalline inclusions are solubilized in the highly alkaline midgut lumen and converted to active toxins by trypsin-like proteases. The activated toxins cross the peritrophic membrane, bind to specific receptors on the brush border apical membrane of midgut columnar cells, and insert into the membrane. Pore formation disrupts the ionic gradients and osmotic balance across the apical membrane and eventually causes the epithelial midgut cells to lyse. This leads to a massive disruption of the epithelium and ultimately to the death of the larvae by starvation or septicemia (20). Binding of B. thuringiensis toxins to specific receptors undoubtedly plays an important role in their mode of action (15, 31). In several cases, the binding characteristics of different toxins correlate well with their toxicities (4, 6, 16, 38, 39). Such correlations are not always found, however. For instance, less potent toxins have been shown to bind with affinities similar to (5, 18) or higher than (9, 42) those of more potent toxins, and CryIA(c), which does not kill Spodoptera frugiperda, has nev-

MATERIALS AND METHODS Insects. Second-instar larvae of the European gypsy moth, Lymantria dispar, and the Chinese silkworm, Bombyx mori, were obtained from the insect-rearing facility of the Great Lakes Forestry Center. L. dispar larvae were reared at 18°C on a standard synthetic medium, and B. mori larvae were reared at room temperature on mulberry (Morus alba) leaves. Third- and fourth-instar larvae which were actively feeding were used for the electrophysiological experiments. Solutions. Experiments were conducted in the standard 32K solution composed of 32 mM KCl, 5 mM CaCl2, 5 mM MgCl2, 166 mM sucrose, and 5 mM Tris-HCl (pH 8.0) (30). The solution was filtered through a 0.2-mm-pore-diameter membrane (Gelman Science). For some experiments, KCl was replaced by N-methyl-D-glucamine–HCl or was added to achieve a concentration of 72 or 128

* Corresponding author. Mailing address: Groupe de Recherche en Transport Membranaire, Universite´ de Montre´al, P.O. Box 6128, Centre Ville Station, Montreal, Quebec H3C 3J7, Canada. Phone: (514) 343-7960. Fax: (514) 343-7146. E-mail: [email protected]. 1679

1680

PEYRONNET ET AL.

APPL. ENVIRON. MICROBIOL.

RESULTS

FIG. 1. Experimental setup for the measurement of membrane potentials. For the measurement of Vam, the midgut was aspirated from the posterior end into a glass pipette until its anterior end curled over the pipette tip and exposed the epithelial cells (A). For the measurement of Vbm, each end of the midgut was aspirated into a separate holding pipette (B). Cells were impaled with conventional microelectrodes filled with 1 M KCl. AMP, recording amplifier; HP, holding pipette; ME, microelectrode; MG, midgut; MO, microscope objective; PERF, perfusion apparatus; REF, reference electrode. The temperature was 24°C in all experiments.

mM. For most experiments, O2 was bubbled vigorously through the solutions for 30 min before use. Toxins. CryIA(a), CryIA(b), CryIA(c), CryIB, CryIC, CryIE, CryIF, and CryIIIA toxins were trypsin activated and high-pressure liquid chromatography purified as described previously (27, 33). Stock solutions were prepared in 25 mM Tris-HCl (pH 9.4) for the CryI toxins or in 25 mM 2-(N-cyclohexylamino)ethanesulfonic acid (CHES)–KOH (pH 10.5) for CryIIIA at a concentration of 2 mg/ml and kept at 4°C. They were diluted daily to appropriate concentrations in the 32K solution. Midgut isolation. The larvae were pinned ventral side down, and the body wall was cut with fine microdissection scissors on the dorsal side along the longitudinal axis from the last abdominal segment to the first thoracic segment. Midguts were transected at each end and rinsed with the 32K solution. The peritrophic membrane, with its food content, and the Malpighian tubules were removed with forceps. B. mori larvae were chilled on crushed ice for 10 min before dissection. Membrane potential measurements. When the midgut is cut transversely, both ends curl back onto themselves. For the measurement of the apical-membrane potential (Vam), the isolated midgut was aspirated from its posterior end into a glass pipette until its anterior end curled around the pipette tip, thus exposing the apical surface (adjacent to the lumen) of the epithelial cells. The pipette was lowered near the bottom of the perfusion chamber, and midgut cells were impaled with a glass microelectrode filled with 1 M KCl (Fig. 1A). For the measurement of the basolateral-membrane potential (Vbm), each end of the midgut was aspirated into a holding pipette and the epithelial cells were impaled from the basal side (facing the hemolymph) (Fig. 1B). Electrode resistance was between 120 and 200 MV. On the basis of earlier studies (29, 30, 35), the impalements were considered successful when the initial membrane potential was 220 mV or less. The signal was amplified with a KS-700 microprobe system (WP Instruments) apparatus, monitored on an oscilloscope (Kikusui Electronics), and recorded on a strip chart recorder (Houston Instruments). The bath was perfused continuously with the 32K solution at approximately 1 ml/min until the membrane potential was stable over 5 min. Then the perfusion was stopped, and 1 ml of the 32K solution containing the appropriate toxin concentration was added directly to the bath. After 5 min, the preparation was rinsed by resuming the perfusion with toxin-free 32K solution. All experiments were carried out at room temperature (24°C).

Effect of B. thuringiensis toxins on Vam of L. dispar midguts. In the absence of toxin, Vam measured in the anterior region of freshly isolated L. dispar larval midguts bathed in the standard 32K solution was 273.4 6 1.4 mV (mean 6 standard error of the mean [SEM]; n 5 50) and remained stable for up to 30 min. It increased when KCl was replaced by N-methyl-D-glucamine–HCl and decreased progressively as the K1 concentration was raised from 32 to 72 and 128 mM (Fig. 2). Saturating the bathing solution with O2 increased Vam to 280.6 6 1.7 mV (n 5 84). The bathing medium was therefore oxygenated by vigorous bubbling of O2 for 30 min prior to its use in all subsequent experiments. Under these conditions, the measured potential ranged from 250 to 2115 mV. The membrane potential was also sensitive to the rate at which the bath was perfused. It decreased by 5 to 10 mV when the perfusion was stopped but returned to its original level soon after the perfusion was resumed (Fig. 3). The addition of 1 or 10 mg of trypsin-activated CryIA(a), CryIA(b), CryIA(c), CryIC, CryIE, or CryIF per ml caused a rapid and irreversible depolarization of the apical membrane (Fig. 3A to C and E to G). In contrast, CryIB (Fig. 3D) and CryIIIA (Fig. 3H) had little effect even at 50 mg/ml. Despite their inactivity in L. dispar midguts, the CryIB and CryIIIA toxin preparations used for these experiments were shown to be functional by demonstrating their ability to form ionic channels in planar lipid bilayer membranes (results not shown). The rate of depolarization depended strongly on the concentration of active toxins but varied relatively little among the different toxins tested. At the highest toxin concentration (10 mg/ml), depolarization started soon after the toxin was added and became evident within 1 to 2 min. The rate of depolarization was considerably reduced when the toxin concentration was reduced to 1.0 mg/ml and became negligible at 0.1 mg/ml. In the presence of 10 mg of CryIA(a), CryIA(b), or CryIC per ml, the membrane potential was completely abolished after about 10 min (Fig. 3A, B, and E). On the other hand, the same concentration of CryIA(c), CryIE, or CryIF caused only a partial reduction of the membrane potential, which remained stable after the bathing solution was replaced with toxin-free medium (Fig. 3C, F, and G). Increasing the time of exposure to the toxin from 5 to 15 min, however, caused a complete depolarization of the cells (Fig. 3G). The effects of these toxins on Vam are summarized in Table

FIG. 2. Effect of external K1 concentration on Vam of isolated L. dispar midguts. Values are means 6 SEM for at least 10 epithelial cells from different midguts.

FIG. 3. Effect of various B. thuringiensis toxins on Vam of midgut epithelial cells from L. dispar. Midguts were mounted as illustrated in Fig. 1A and perfused for 5 min with the 32K solution. The perfusion was stopped, and the preparation was incubated for 5 min (as indicated by the width of the box above each panel) with the indicated trypsin-activated toxins in the 32K solution. The concentrations used were 0 (F), 0.1 (ç), 1.0 (å), 10.0 (■), or 50.0 (}) mg/ml. Midguts were finally rinsed with the 32K solution by resuming the perfusion. In panel G, the dashed line represents an experiment in which the midguts were incubated with 10 mg of CryIF/ml for 15 min. V is Vam measured at the indicated times, and Vo is the mean membrane potential measured during the 5 min preceding the addition of toxin. Values are means 6 SEM for 3 to 10 independent experiments.

1681

1682

PEYRONNET ET AL.

APPL. ENVIRON. MICROBIOL.

TABLE 1. Comparison of the activity of B. thuringiensis toxins determined by in vivo bioassays on L. dispar larvae and by in vitro depolarization assays on isolated L. dispar larval midguts Toxin

Bioassay (FFD50 [ng/larva])a

Depolarization assay (TD50 [min])b

CryIA(a) CryIA(b) CryIA(c) CryIB CryIC CryIE CryIF CryIIIA

56.2 22.0 2,484.0 .208.0 948.0 .696.0 ND ND

2.9 2.5 2.7 NS 2.7 1.7 4.8 NS

a The 50% frass failure dose (FFD50), determined with force-feeding bioassays by van Frankenhuyzen et al. (36, 37). ND, not determined. b Time required for 10 mg of toxin/ml to cause 50% depolarization of the midgut apical membrane. NS, not sensitive.

1 and compared with published values on their in vivo toxicity towards L. dispar (36, 37). Among the active toxins, the time required for 10 mg of toxin per ml to depolarize the membrane by 50% varied only about 2.8-fold. In contrast, the 50% frass failure dose, measured with force-feeding bioassays, varied about 113-fold. The difference in activity is especially noteworthy for CryIA(c) and CryIC, which were much less active than CryIA(a) and CryIA(b) in the bioassays, and for CryIE, which was even found to be inactive in vivo, although it was not tested at very high concentrations. Effect of B. thuringiensis toxins on Vbm of L. dispar midguts. Because some of the differences between the in vivo and in vitro toxicities of the tested toxins could possibly be due to the fact that, in the experimental setup used to measure Vam (Fig. 1A), toxins could have access to the basolateral side of the epithelium by infiltrating the space between the midgut tissue and the pipette, Vbm was also measured (Fig. 1B). In the absence of toxin, the potential measured in oxygenated 32K solution was 232.0 6 2.0 mV (n 5 31). It ranged from 220 to 250 mV and was stable over 15 min. Addition of a 10-mg/ml solution of CryIA(a) or CryIC, which were highly active on the apical membrane, to the basolateral side of the epithelium had no significant effect on Vbm (Fig. 4). The toxins thus appear to act specifically on the apical membrane of the epithelial cells. Effect of B. thuringiensis toxins on Vam of B. mori midguts. In the absence of toxin, Vam measured in the anterior region of freshly isolated B. mori midguts bathed in O2-saturated 32K solution was 284.5 6 3.2 mV (n 5 63). It ranged from 250 to 2135 mV. The membrane potential of B. mori midgut cells was more sensitive to the rate of bath perfusion than that of L. dispar. When the perfusion was stopped, the membrane potential dropped about 20 to 30 mV, but as was observed with midguts isolated from L. dispar, it returned rapidly to its original level when the perfusion was resumed (Fig. 5). Addition of CryIA(a) caused a rapid and irreversible depolarization of the membrane, but CryIA(b) and CryIA(c) were inactive (Fig. 5). These results correlate well with published lethal doses established with bioassays (37), which demonstrate a high level of toxicity for CryIA(a) but no detectable activity for the other two toxins. DISCUSSION The larval midgut epithelium of lepidopteran insects actively transports K1 ions from the hemolymph to the lumen (14). This activity, which is mediated by a vacuolar-type proton AT-

FIG. 4. Effect of B. thuringiensis toxins on Vbm of midgut epithelial cells from L. dispar. Midguts were mounted on two holding pipettes as illustrated in Fig. 1B and perfused for 5 min with the 32K solution. After the perfusion was stopped, the midguts were incubated for 5 min with trypsin-activated CryIA(a) (ç) or CryIC (å) at 10 mg/ml for another 5 min (as indicated by the width of the box). The midguts were then washed by resuming the perfusion. Control experiments were done without toxin (F). V is Vbm measured at the indicated times, and Vo is the mean membrane potential measured during the 5 min preceding the addition of toxin. Values are means 6 SEM for three independent experiments.

Pase coupled with an electrogenic K1/H1 exchanger, both located in the apical membrane of goblet cells (41), maintains a strong potential difference across the epithelium. In turn, the electrical component of the K1 electrochemical gradient generated across the apical membrane serves as the main driving force for the absorption from the lumen of solutes such as amino acids by columnar cells (11). By forming channels in the apical membrane, B. thuringiensis toxins are thought to cause the electrochemical gradient to collapse, thus abolishing the capacity of the cells to transport solutes and allowing equilibration of the pH between the cytoplasm and the highly alkaline content of the lumen (21, 43). In the present study, we used a standard microelectrode technique to measure the membrane potential in freshly isolated midguts from lepidopteran larvae. In the absence of toxin, Vam was large (about 280 mV), stable for up to 30 min,

FIG. 5. Effect of B. thuringiensis toxins on Vam of midgut epithelial cells from B. mori. Membrane potential was measured as described in the legend to Fig. 3. During the period indicated by the width of the box, bath perfusion was stopped and CryIA(a) (■), CryIA(b) (å), or CryIA(c) (ç) was added to the bath at 10 mg/ml or was omitted (F). Values are means 6 SEM for four to seven independent experiments.

VOL. 63, 1997

EFFECT OF B. THURINGIENSIS TOXINS ON INSECT MIDGUTS

and sensitive to the K1 and O2 concentrations in the bathing solution. The viability of the midgut cells and the activity of their ionic pumps were thus well preserved following isolation of the tissue. Because B. thuringiensis toxins directly affect the ionic permeability of the apical membrane, such membrane potential measurements provide a rapid and sensitive in vitro assay for toxin potency in their normal target cells. The membrane potential decreased rapidly in the presence of active toxins and was unaffected by CryIB and CryIIIA, two inactive toxins. Furthermore, all tested toxins with known activity against B. mori or L. dispar larvae were active in their respective isolated midguts. In the case of B. mori midguts, the in vitro activity of the toxins was consistent with their in vivo toxicities (37). Surprisingly, however, the ability of active toxins to depolarize the apical membrane of L. dispar midgut cells correlated only partially with their respective in vivo toxicities, established by force-feeding bioassays (Table 1) (36, 37). Despite their relatively poor activity in the bioassays, CryIA(c), CryIC, and CryIE were about as effective in vitro as the more potent CryIA(a) and CryIA(b) toxins. In vitro and in vivo data should, however, be compared with caution. While in the in vitro experiments the effects of the toxins are measured directly on their primary target cells during a 15-min interval, in the in vivo assays their effects are examined on whole larvae after several days. Regeneration of the midgut epithelium could thus contribute to the survival of the larvae without affecting the results of the in vitro assay. Such comparisons could also be misleading in view of the rather large differences in the relative in vivo toxicities towards L. dispar larvae reported by different laboratories for different toxins. For instance, CryIA(b) was found to be about 394 (42), 113 (37), 4 (23), and 1.2 (24) times more potent than CryIA(c) against this insect. It should be pointed out, however, that such variability may in large part be due to different bioassay conditions and, in particular, to the use of different toxicity indices: e.g., mortality (42), frass failure (37), and growth inhibition (23, 24). Our results correlate well with those of Liang et al. (24) for the three CryIA toxins, but comparison with the bioassay data of van Frankenhuyzen et al. (36, 37), in addition to being more complete, appears to be more appropriate, since potential sources of variability have been minimized by use of larvae from the same source, identical diets, and toxins purified and trypsin activated by the same method. As was mentioned above, binding of B. thuringiensis toxins to brush border membrane vesicles often fails to correlate with their toxicities. This difference is usually taken as an indication of the importance of postbinding events such as pore formation in the mode of action of these toxins. Although the results of the present study cannot exclude the possibility that a toxin could bind to the membrane without forming a channel, they do indicate that, at least in the case of L. dispar, the ability of a toxin to form a channel in the apical membrane does not necessarily reflect its potency towards whole larvae. An even more striking case of discrepancy between the in vivo and in vitro toxicities of B. thuringiensis toxins was demonstrated when a number of toxins were tested against live silkworm larvae and against dissociated midgut epithelial cells from the same insect (27). For example, CryIA(c), which was inactive in B. mori larvae, was among the most effective toxins tested, including CryIA(a), when assayed against the epithelial cells. The differences in apparent potency summarized in Table 1 are likely related to the fact that in the in vitro assay, as in most binding assays, the toxins have unhindered access to their receptors on the surface of the apical membrane. In live larvae, on the other hand, numerous additional factors could affect the potency of different toxins. Although they are potentially im-

1683

portant sources of variation in the activities of different toxins in nature, protoxin solubilization (1) and activation (13, 19, 32) cannot explain the differences presented here, since in the bioassays the larvae were force-fed with trypsin-activated toxins. It remains possible, however, that the toxins may have different effects on the feeding behavior of the larvae (7), thus affecting the results of the bioassays based on growth inhibition or frass failure. The activated toxins could also differ in their susceptibility to further proteolysis in the midgut environment. The possibility that the solubility of different toxins, even when activated, may vary in the midgut is suggested by the recent isolation from spruce budworm gut juice of a protein complex that causes selective toxin precipitation (28). This particular complex, however, cannot explain the relatively poor activity of CryIA(c) against L. dispar larvae, since it does not precipitate this toxin and appears to be specific for CryIA(a). Gut juice from L. dispar was also shown to have a toxinprecipitating activity (28), but since it was tested against the toxins from the HD-1 strain of B. thuringiensis, which produces a mixture of CryIA(a), CryIA(b), and CryIA(c), it remains possible that it could differ in its specificity and that additional toxin-precipitating proteins could be present. Diffusion across the peritrophic membrane could also affect differentially the in vivo activity of the toxins, as suggested by the rather strong accumulation of toxins at the level of this membrane that was demonstrated by immunocytochemical studies (2, 3) and by the fact that such accumulation varies greatly for different toxins. Passage through the peritrophic membrane could also be restricted by the binding of the toxins to specific proteins in the midgut, as was suggested previously (28). In conclusion, the results of the present study and their comparison with published bioassay data illustrate the complexity of the interaction of B. thuringiensis toxins with their target tissue. Although the results clearly emphasize the importance of specific binding and pore formation in the midgut brush border membrane, as well as the usefulness of the in vitro techniques used to characterize them, they stress the need for further research on the potential role of prebinding events in modulating the potency of these toxins against the live insect. ACKNOWLEDGMENTS We are grateful to Manon Beauchemin of the Biotechnology Research Institute for preparing the toxins and to Ray Wilson of the Great Lakes Forestry Center for his assistance in rearing the insects. This work was supported by strategic grant STR0167557 from the Natural Sciences and Engineering Research Council of Canada to R. Laprade and J.-L. Schwartz. REFERENCES 1. Aronson, A. I., E.-S. Han, W. McGaughey, and D. Johnson. 1991. The solubility of inclusion proteins from Bacillus thuringiensis is dependent upon protoxin composition and is a factor in toxicity to insects. Appl. Environ. Microbiol. 57:981–986. 2. Bravo, A., K. Hendrickx, S. Jansens, and M. Peferoen. 1992. Immunocytochemical analysis of specific binding of Bacillus thuringiensis insecticidal crystal proteins to lepidopteran and coleopteran midgut membranes. J. Invertebr. Pathol. 60:247–253. 3. Bravo, A., S. Jansens, and M. Peferoen. 1992. Immunocytochemical localization of Bacillus thuringiensis insecticidal crystal proteins in intoxicated insects. J. Invertebr. Pathol. 60:237–246. 4. Denolf, P., S. Jansens, M. Peferoen, D. Degheele, and J. Van Rie. 1993. Two different Bacillus thuringiensis delta-endotoxin receptors in the midgut brush border membrane of the European corn borer, Ostrinia nubilalis (Hu ¨ber) (Lepidoptera: Pyralidae). Appl. Environ. Microbiol. 59:1828–1837. 5. Escriche, B., A. C. Martinez-Ramirez, M. D. Real, F. J. Silva, and J. Ferre´. 1994. Occurrence of three different binding sites for Bacillus thuringiensis d-endotoxins in the midgut brush border membrane of the potato tuber moth, Phthorimaea operculella (Zeller). Arch. Insect Biochem. Physiol. 26: 315–327.

1684

PEYRONNET ET AL.

6. Estada, U., and J. Ferre´. 1994. Binding of insecticidal crystal proteins of Bacillus thuringiensis to the midgut brush border of the cabbage looper, Trichoplusia ni (Hu ¨bner) (Lepidoptera: Noctuidae), and selection for resistance to one of the crystal proteins. Appl. Environ. Microbiol. 60:3840–3846. 7. Farrar, R. R., Jr., and R. L. Ridgway. 1995. Feeding behavior of gypsy moth (Lepidoptera: Lymantriidae) larvae on artificial diet containing Bacillus thuringiensis. Environ. Entomol. 24:755–761. 8. Feitelson, J. S., J. Payne, and L. Kim. 1992. Bacillus thuringiensis: insects and beyond. Bio/Technology 10:271–275. 9. Ferre´, J., M. D. Real, J. Van Rie, S. Jansens, and M. Peferoen. 1991. Resistance to the Bacillus thuringiensis bioinsecticide in a field population of Plutella xylostella is due to a change in midgut membrane receptor. Proc. Natl. Acad. Sci. USA 88:5119–5123. 10. Garczynski, S. F., J. W. Crim, and M. J. Adang. 1991. Identification of putative insect brush border membrane-binding molecules specific to Bacillus thuringiensis d-endotoxin by protein blot analysis. Appl. Environ. Microbiol. 57:2816–2820. 11. Giordana, B., F. W. Sacchi, P. Parenti, and G. M. Hanozet. 1989. Amino acid transport systems in intestinal brush-border membranes from lepidopteran larvae. Am. J. Physiol. 257:R494–R500. 12. Gould, F., A. Martinez-Ramirez, A. Anderson, J. Ferre´, F. J. Silva, and W. J. Moar. 1992. Broad-spectrum resistance to Bacillus thuringiensis toxins in Heliothis virescens. Proc. Natl. Acad. Sci. USA 89:7986–7990. 13. Haider, M. Z., B. H. Knowles, and D. J. Ellar. 1986. Specificity of Bacillus thuringiensis var. colmeri insecticidal d-endotoxin is determined by differential proteolytic processing of the protoxin by larval gut proteases. Eur. J. Biochem. 156:531–540. 14. Harvey, W. R., and K. Zerahn. 1972. Active transport of potassium and other alkali metals by the isolated midgut of the silkworm. Curr. Top. Membr. Transp. 3:367–410. 15. Hofmann, C., P. Lu ¨thy, R. Hu ¨tter, and V. Pliska. 1988. Binding of the delta endotoxin from Bacillus thuringiensis to brush-border membrane vesicles of the cabbage butterfly (Pieris brassicae). Eur. J. Biochem. 173:85–91. 16. Hofmann, C., H. Vanderbruggen, H. Ho ¨fte, J. Van Rie, S. Jansens, and H. Van Mellaert. 1988. Specificity of Bacillus thuringiensis d-endotoxins is correlated with the presence of high-affinity binding sites in the brush border membrane of target insect midguts. Proc. Natl. Acad. Sci. USA 85:7844– 7848. 17. Ho¨fte, H., and H. R. Whiteley. 1989. Insecticidal crystal proteins of Bacillus thuringiensis. Microbiol. Rev. 53:242–255. 18. Ihara, H., E. Kuroda, A. Wadano, and M. Himeno. 1993. Specific toxicity of d-endotoxins from Bacillus thuringiensis to Bombyx mori. Biosci. Biotechnol. Biochem. 57:200–204. 19. Jaquet, F., R. Hu ¨tter, and P. Lu ¨thy. 1987. Specificity of Bacillus thuringiensis delta-endotoxin. Appl. Environ. Microbiol. 53:500–504. 20. Knowles, B. H. 1994. Mechanism of action of Bacillus thuringiensis insecticidal d-endotoxins. Adv. Insect Physiol. 24:275–308. 21. Knowles, B. H., and J. A. T. Dow. 1993. The crystal d-endotoxins of Bacillus thuringiensis: models for their mechanism of action on the insect gut. Bioessays 15:469–476. 22. Lambert, B., and M. Peferoen. 1992. Insecticidal promise of Bacillus thuringiensis. Facts and mysteries about a successful biopesticide. BioScience 42: 112–122. 23. Lee, M. K., A. Curtiss, E. Alcantara, and D. H. Dean. 1996. Synergistic effect of the Bacillus thuringiensis toxins CryIAa and CryIAc on the gypsy moth, Lymantria dispar. Appl. Environ. Microbiol. 62:583–586. 24. Liang, Y., S. S. Patel, and D. H. Dean. 1995. Irreversible binding kinetics of Bacillus thuringiensis CryIA d-endotoxins to gypsy moth brush border membrane vesicles is directly correlated to toxicity. J. Biol. Chem. 270:24719– 24724. 25. MacIntosh, S. C., T. B. Stone, R. S. Jokerst, and R. L. Fuchs. 1991. Binding of Bacillus thuringiensis proteins to a laboratory-selected line of Heliothis virescens. Proc. Natl. Acad. Sci. USA 88:8930–8933. 26. Masson, L., A. Mazza, R. Brousseau, and B. Tabashnik. 1995. Kinetics of

APPL. ENVIRON. MICROBIOL.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

Bacillus thuringiensis toxin binding with brush border membrane vesicles from susceptible and resistant larvae of Plutella xylostella. J. Biol. Chem. 270:11887–11896. Masson, L., A. Mazza, L. Gringorten, D. Baines, V. Aneliunas, and R. Brousseau. 1994. Specificity domain localization of Bacillus thuringiensis insecticidal toxins is highly dependent on the bioassay system. Mol. Microbiol. 14:851–860. Milne, R. E., A. S. D. Pang, and H. Kaplan. 1995. A protein complex from Choristoneura fumiferana gut-juice involved in the precipitation of d-endotoxin from Bacillus thuringiensis subsp. sotto. Insect Biochem. Mol. Biol. 25:1101–1114. Moffett, D. F., R. L. Hudson, S. B. Moffett, and R. L. Ridgway. 1982. Intracellular K1 activities and cell membrane potentials in a K1-transporting epithelium, the midgut of tobacco hornworm (Manduca sexta). J. Membr. Biol. 70:59–68. Moffett, D. F., and A. R. Koch. 1988. Electrophysiology of K1 transport by midgut epithelium of lepidopteran insect larvae. I. The transbasal electrochemical gradient. J. Exp. Biol. 135:25–38. Oddou, P., H. Hartmann, F. Radecke, and M. Geiser. 1993. Immunologically unrelated Heliothis sp. and Spodoptera sp. midgut-proteins bind Bacillus thuringiensis CryIA(b) d-endotoxin. Eur. J. Biochem. 212:145–150. Oppert, B., K. J. Kramer, D. E. Johnson, S. C. MacIntosh, and W. H. McGaughey. 1994. Altered protoxin activation by midgut enzymes from a Bacillus thuringiensis resistant strain of Plodia interpunctella. Biochem. Biophys. Res. Commun. 198:940–947. Schwartz, J.-L., L. Garneau, L. Masson, and R. Brousseau. 1991. Early response of cultured lepidopteran cells to exposure to d-endotoxin from Bacillus thuringiensis: involvement of calcium and anionic channels. Biochim. Biophys. Acta 1065:250–260. Tabashnik, B. E., N. Finson, F. R. Groeters, W. J. Moar, M. W. Johnson, K. Luo, and M. Adang. 1994. Reversal of resistance to Bacillus thuringiensis in Plutella xylostella. Proc. Natl. Acad. Sci. USA 91:4120–4124. Thomas, M. V., and T. E. May. 1984. Active potassium ion transport across the caterpillar midgut. II. Intracellular microelectrode studies. J. Exp. Biol. 108:293–304. van Frankenhuyzen, K., J. L. Gringorten, D. Gauthier, R. E. Milne, L. Masson, and M. Peferoen. 1993. Toxicity of activated CryI proteins from Bacillus thuringiensis to six forest Lepidoptera and Bombyx mori. J. Invertebr. Pathol. 62:295–301. van Frankenhyzen, K., J. L. Gringorten, R. E. Milne, D. Gauthier, M. Pusztai, R. Brousseau, and L. Masson. 1991. Specificity of activated CryIA proteins from Bacillus thuringiensis subsp. kurstaki HD-1 for defoliating forest Lepidoptera. Appl. Environ. Microbiol. 57:1650–1655. Van Rie, J., S. Jansens, H. Ho¨fte, D. Degheele, and H. Van Mellaert. 1989. Specificity of Bacillus thuringiensis d-endotoxins. Importance of specific receptors on the brush border membrane of the mid-gut of target insects. Eur. J. Biochem. 186:239–247. Van Rie, J., S. Jansens, H. Ho¨fte, D. Degheele, and H. Van Mellaert. 1990. Receptors on the brush border membrane of the insect midgut as determinants of the specificity of Bacillus thuringiensis delta-endotoxins. Appl. Environ. Microbiol. 56:1378–1385. Van Rie, J., W. H. McGaughey, D. E. Johnson, B. D. Barnett, and H. Mellaert. 1990. Mechanism of insect resistance to the microbial insecticide Bacillus thuringiensis. Science 247:72–74. Wieczorek, H., M. Putzenlechner, W. Zeiske, and U. Klein. 1991. A vacuolartype proton pump energizes K1/H1 antiport in an animal plasma membrane. J. Biol. Chem. 266:15340–15347. Wolfersberger, M. G. 1990. The toxicity of two Bacillus thuringiensis d-endotoxins to gypsy moth larvae is inversely related to the affinity of binding sites on midgut brush border membranes for the toxins. Experientia 46:475– 477. Wolfersberger, M. G. 1992. V-ATPase-energized epithelia and biological insect control. J. Exp. Biol. 172:377–386.