- Email: [email protected]
The effect of arcelin-1 on the structure of the midgut of bruchid larvae and immunolocalization of the arcelin protein
Journal of Insect Physiology 46 (2000) 393–402 www.elsevier.com/locate/jinsphys
The effect of arcelin-1 on the structure of the midgut of bruchid larvae and immunolocalization of the arcelin protein Norma S. Paes a, Isabel R. Gerhardt a, Marise V. Coutinho a, Massaru Yokoyama a,b, Eliana Santana a, Nicholas Harris a,c, Maarten J. Chrispeels a,d, M. Fatima Grossi de Sa a,* a
d
Embrapa/Cenargen, PO Box 02372, Brasilia, DF, Brazil b EMBRAPA/CNPAF, Goiania, GO, Brazil c Department of Biological Sciences, University of Durham, South Road, Durham, DH1 3LE, UK Department of Biology, University of California San Diego, 9500 Gilman Drive, La Jolla, CA, 92093-0116, USA Received 6 January 1999; accepted 9 March 1999
Abstract Some wild accessions of the common bean (Phaseolus vulgaris) contain a family of proteins called arcelins, that are toxic to the larvae of certain bruchid species. Among the six allelic variants of arcelin tested so far, arcelin-5 and arcelin-1 confer the highest level of resistance against the Mexican bean weevil, Zabrotes subfasciatus. The same proteins are not toxic to the bean weevil, Acanthoscelides obtectus, which is also a serious pest of cultivated beans. Arcelins belong to the bean lectin family that includes phytohemaggutinins and α-amylase inhibitors. Although homologous to lectins, arcelins are themselves only very weak lectins, and their binding properties have not been clearly established. The toxic properties of arcelins may be related to their recognition of and interaction with the glycoproteins and other constituents of the membranes along the digestive tract of insects. Since arcelin-1 was shown to have growth inhibitory effects for the larvae of Z. subfasciatus but not of A. obtectus, we examined the effect of an arcelin-1 containing diet on the structure of the cells that line the intestinal tract of the larvae of these two bruchid species, and used antibodies against arcelin to examine the distribution of arcelin within the cells and tissues. Here we show that dietary arcelin-1 caused an alteration of the gut structure and the penetration of arcelin into the haemolymph in Z. subfasciatus but not in A. obtectus. These results lead us to suggest that arcelins exert their toxic effect by severely damaging the epithelial cells. 2000 Elsevier Science Ltd. All rights reserved. Keywords: Bruchid beetles; Zabrotes subfasciatus; Arcelin; Immunolocalization; Midgut
1. Introduction Bruchids are the most serious pests attacking food legume seeds during storage. Substantial amounts of stored common bean, Phaseolus vulgaris (L.), are lost due to the damage caused by the two main storage pests, the Mexican bean weevil, Zabrotes subfasciatus (Boheman) and the bean weevil Acanthoscelides obtectus (Say) (Cardona et al., 1989). In some wild accessions of P. vulgaris that show variable levels of resistance to Z. subfasciatus, arcelin seed proteins replace the normal seed storage protein, phaseolin (Osborn et al., 1986).
* Corresponding author. Fax: +55-61-340-3624. E-mail address: [email protected] de Sa)
(M.F. Grossi
Subsequently, arcelins were shown to have inhibitory effects on the larval growth of Z. subfasciatus and to account for the resistance of the seeds of these wild accessions (Osborn et al., 1988a; Cardona et al., 1990). Although some wild bean accessions containing arcelin are also highly resistant to A. obtectus (Schoonhoven et al., 1983; Cardona et al., 1989; Kornegay and Cardona, 1991), arcelins appear not to be the factor involved in this resistance (Osborn et al., 1988a). Six electrophoretic variants of arcelin, designated arcelin-1 to 6, have been reported (Osborn et al., 1986; Lioi and Bollini, 1989; Santino et al., 1991) and among them, arcelin-5 and arcelin-1 exhibit the highest level of bruchid resistance (Cardona et al., 1990; Fory et al., 1996). A comparison of the amino acid sequences of arcelins with other seed proteins shows that they belong to the bean lectin family that includes phytohemagglutinin and
0022-1910/00/$ - see front matter 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 2 - 1 9 1 0 ( 9 9 ) 0 0 1 2 2 - 5
394
N.S. Paes et al. / Journal of Insect Physiology 46 (2000) 393–402
α-amylase inhibitor. Arcelins are not strong lectins, but may have weak lectin activity (Osborn et al., 1988b). Xray diffraction analysis shows that arcelin-1 molecules form a lectin-like dimer (Mourey et al., 1997). The dimeric structure is particularly suited to a function in molecular recognition, since it might allow the bridging of cells through interactions with membrane-associated glycoproteins or glycolipids (Mourey et al., 1998). The epithelial cells along the digestive tract of phytophagous insects are directly exposed to the contents of the diet, and therefore, are possible target sites for plant lectins and lectin-like proteins (Peumans and Damme, 1995). The mechanism of action of arcelins is not known, but in analogy with phytohemagglutinin, a protein that is toxic to mammals by virtue of its binding to the intestinal epithelium (Bardocz et al., 1995), we postulate that the toxicity of arcelins may be based on specific binding to glycoconjugates or proteins somewhere in the gut of the insect. If arcelins behave as lectins, then two types of interactions could be possible: binding of arcelins to glycoconjugates exposed on the epithelial cells along the digestive tract, and binding of arcelins to glycosylated digestive enzymes (Peumans and Damme, 1995). Fabre et al. (1998) proposed that the Asn-linked glycans of arcelin-1 that are responsible for its binding to mannosespecific lectins, known to occur in various insects, may account for its toxicity. To understand why arcelin-1 is toxic for Z. subfasciatus larvae but not for A. obtectus larvae, we conducted a study of the effects of dietary arcelin on the structure of the larval gut and determined the distribution of arcelin in the larval tissues with specific antibodies. Our results show that arcelin-1 had severe deleterious effects on the gut of Z. subfasciatus, but not on the gut of A. obtectus, and that arcelin was present in the haemolymph of Z. subfasciatus, indicating that it had traversed the cells that line the gut.
2. Materials and methods 2.1. Feeding tests Bruchid cultures were reared on the susceptible P. vulgaris cultivar Goiano Precoce and on the resistant wild accession of P. vulgaris G12882 (containing arc-1) and the wild accession G02771 (containing arc-5) The cultures were mantained in petri dishes in a controlled environment at 27°C and 60% relative humidity. 2.2. Protein extraction and immunoblots Proteins were extracted from bean seeds according to Grossi de Sa´ et al. (1997), separated by SDS–PAGE 13%, and transferred to nitrocellulose. The immunoblots were made with a polyclonal mouse serum against chemically deglycosylated purified arc-1. The antibodies
bound to the immunoblot were visualized with horseradish–peroxidase-conjugated secondary antibody (BioRad) using the manufacturer’s specifications. 2.3. Histology and immunoblots Larvae of 19 days of Z. subfasciatus and A. obtectus, fed on susceptible and resistant seeds, were washed in N-heptan saturated with fixation buffer (3% paraformaldehyde, 0.5% glutaraldehyde in sodium cacodylate buffer 0.05 M, pH 7.2) for 5–10 min, followed by fixation in the same fixation buffer for 24 h, alternating between shaking and vacuum. The tissue was then washed in sodium cacodylate buffer (3×15 min each) and dehydrated in ethanol/water at 10, 20, 30, 50, 70, 90 and 100% ethanol, and finally embedded in the hydrophilic resin LR White. Semi-thin sections (approx. 1 µm) were cut for light microscopy and stained with toluidine blue for structural examination of the tissues. Unstained sections were used for immunolabelling to investigate the effects of arcelin on the growth and development of the larvae of Z. subfasciatus and A. obtectus. The distribution of specifically-bound primary antibody (anti-arc-1) was determined using secondary antibody conjugated with colloidal gold (5 nm) with the signal amplified by silver enhancement.
3. Results 3.1. Arc-1 slows the development of Z. subfasciatus larvae To investigate the effects of arcelins on growth and development of larvae of Z. subfasciatus, we examined larvae that were allowed to develop normally in a control variety of beans (Goiano Precoce) and larvae that were allowed to develop in the seeds of wild accessions containing arcelin-1 or arcelin-5. We compared larval sizes 19 days after infestation (egg laying) with those of a second bruchid, A. obtectus. The results (Fig. 1) show that the larvae of both bruchid species were largest on the domesticated bean variety that does not contain any arcelins [Fig. 1(B)a and (C)a]. The growth of Z. subfasciatus larvae was substantially inhibited on the arc-1 containing seeds and even much more so on the arc-5 containing seeds. This bean accession clearly has the highest level of antibiosis factors against Z. subfasciatus. Larvae fed with seeds of acessions G12882 and G02771 had an average weight of 1.63 and 0.35 mg, respectively. In contrast, larvae fed with control seeds had an average weight of 4.96 mg (Gerhardt et al., unpublished results). The growth of the A. obtectus larvae was slightly inhibited on the two wild accessions, but there was no difference between the two accessions. These results
N.S. Paes et al. / Journal of Insect Physiology 46 (2000) 393–402
395
(control) and on arc-1 containing seeds is shown in Fig. 2(A), (C) and (E) and 2(B), (D) and (F), respectively. Fig. 2(A), a section through the midgut of a control larva, shows two cross-sections of the gut wall surrounding the intestinal lumen (L). Two cross-sections appear in a single micrograph because the gut forms a loop when the larvae are fixed and embedded. In Fig. 2(C), the epithelial cells that form the gut wall have a normal polar appearance, with prominent nuclei (N) in the center and numerous small translucent vacuoles/droplets toward the apical side. The cells appear to be secreting large dense-staining droplets (Dr). The epithelial cells have microvilli (MV) that stick out into the crypts (Cr) and these areas are more lightly stained than the cells themselves. Microvilli are clearly visible in Fig. 2(E). Adjacent epithelial cells appear to be closely appressed and only a narrow white line of extracellular matrix is visible between the cells [arrows in Fig. 2(E)]. The lumen contains some stained material that may be food in the process of being digested and/or digestive enzymes secreted by the cells that line the intestine. Some regions of the midgut [Fig. 2(B)] of larvae allowed to develop on arc-1 containing seeds were similar in appearance to the midgut of the larvae on control seeds, except that the lumen contained more stained material and that crystalline inclusions (CI) were present. Secretory droplets (Dr) were present near the surface of the epithelial cells (Ec). In other regions shown in Fig. 2(D) and (F), the cells had a much less regular appearance and were not closely appressed. They were separated by broad channels, and the appearance of the translucent droplets/vacuoles was much less regular. The densely staining secreted droplets in the lumen were similar in size and distribution as in the controls. Thus, it appeared as if secretion was intense, but digestion was slow as evidenced by the accumulation of undigested food in the lumen. In addition, the structure of the epithelium appeared disrupted. 3.3. Arc-1 penetrates into the haemolymph in Z. subfasciatus larvae fed with arc-1 seeds Fig. 1. (A) Goiano Precoce seed damaged by the larva of a bruchid beetle Z. subfasciatus. Inhibitory effect of arc-1 and arc-5 protein on the growth of larvae of Z. subfasciatus (B) and A. obtectus (C) after 19 days of infestation. a: Larvae allowed to develop on P. vulgaris cultivar Goiano Precoce seeds; b: larvae allowed to develop on wild bean P. vulgaris seeds containing arc-1; c: larvae allowed to develop on wild bean P. vulgaris containing arc-5. Bars=1 mm.
indicate that these two accessions contain no or very low levels of antibiosis factors against A. obtectus. 3.2. Arc-1 disrupts the integrity of the midgut epithelium of Z. subfasciatus The structure of the midgut of Z. subfasciatus larvae allowed to develop for 19 days on Goiano Precoce seeds
Because the distribution of arcelin within the tissues of the larvae may tell us something about its mode of action, we determined the location of arcelin by immunocytochemistry with a serum prepared against deglycosylated arcelin. The specificity of this serum is shown in Fig. 3. Fig. 3(A) shows the proteins present in extracts of G12882 (a) and Goiano Precoce (b) seeds separated by SDS–PAGE. In the adjacent immunoblot, the antiserum recognizes arcelin in extracts of the G12882 seeds, but does not recognize any protein in the Goiano Precoce seeds [Fig. 3(B)], indicating that the serum is specific for arcelin and does not react with other members of this protein family such as phytohemagglutinin and α-amylase inhibitor. Both of these proteins are
396
N.S. Paes et al. / Journal of Insect Physiology 46 (2000) 393–402
Fig. 2. Structure of the midgut epithelium of 19 day-old larvae of Z. subfasciatus allowed to develop on cultivar Goiano Precoce seeds (A, C, E) or on arc-1 containing seeds (B, D, F) visualized with toluidine blue. N=nuclei; L=lumen; MV=microvilli; Ec=gut epithelial cells; CI=cristalline inclusions; Cr=crypts; MU=muscle; Dr=dense droplets. Arrowheads in (E) indicate close apposition of adjacent cells. Arrowheads in (F) indicate broad channels between adjacent cells. Bars in (A)–(D)=150 µm; in (E) and (F)=50 µm.
present in cultivars of the common bean such as Goiano Precoce. We used this serum to localize arcelin-1 in the tissues of Z. subfasciatus larvae allowed to develop on the arc-1 containing seeds. As a control, we could not use larvae that develop on Goiano Precoce seeds, because those seeds have no arcelin. We therefore determined the distribution of arc-1 in the tissues of A. obtectus larvae that were allowed to develop on arc-1 seeds. We considered this to be a suitable control because these larvae are not or only minimally affected by arc-1 and develop normally [Fig. 1(B)b]. The structure of the gut of A. obtectus larvae as seen at low magnification in toluidine blue stained sections [Fig. 4(A) and (C)] is similar to that of Z. subfasciatus. After immunostaining with antibodies to arc-1, we found, as expected, strong immunolabeling of the gut content in the lumen of the gut [Fig. 4(B) and (D)] but without apparent staining of the cells. Fig. 4(A) and (B),
and (C) and (D) are matched pairs in which one is stained with toluidine blue and the other one is immunostained. This immunostaining is thought to represent undigested arc-1. The justaposition of unstained cells and stained lumen content is most clearly visible in Fig. 4(B) (see asterisk). Labeling of arcelin is also visible in crypts (Cr) of the epithelium [Fig. 4(D)]. Immunostaining is also found in structures that we interpret as vesicles (Vs), containing arc-1 that has been internalized by the cells. The haemolymph that surrounds the gut is completely free of immunostaining. Fig. 4(E) is a section through the midgut of an A. obtectus larvae that has been allowed to develop on Goiano Precoce seeds. There is as expected no immunostaining because this bean variety has no arcelin or cross reacting proteins. When examining the Z. subfasciatus larvae, we also found strong labeling of the gut content in the lumen.
N.S. Paes et al. / Journal of Insect Physiology 46 (2000) 393–402
Fig. 3. (A) Coomassie blue stained gel of total protein extract from a: arc-1 seeds and b: cultivar Goiano Precoce seeds. (B) Same gel as in (A), transferred to nitrocellulose membrane and immunoreacted with anti-arc-1 serum. Arrows correspond to molecular weight markers of 66, 45, and 29 kDa.
Fig. 5(A) and (C), and (B) and (D) are once again matched pairs with one set stained with toluidine blue and the other one immunostained. The black labeling of the lumen (L) is very obvious in Fig. 5(B) and (D). A lower level of staining is readily apparent in the haemolymph (HE). Higher magnification of cells with crypts [Fig. 5(E) and (F)] shows labeling of the crypts and of apparent vesicles (Vs) in the cells. Internalization of arcelin and presumably other dietary components into vesicles is a feature of both the sensitive and the resistant bruchid species. 3.4. Arc-1 is present in cytoplasmic vesicles We used immuno-electron microscopy to examine the ultrastructure of the epithelial cells of Z. subfasciatus allowed to develop on resistant seeds and to confirm the existence of vesicles that internalize arcelin. The results are shown in Fig. 6(A) and (B). Fig. 6(A) shows a low magnification image of the interface between the lumen (L) and the epithelial cells (Ec). The peritrophic membrane (PM) separates the food in the lumen from the zone with the microvilli (MV). The microvilli are embedded in a clear zone underneath the peritrophic membrane. Underneath the microvillar zone is an area of cytoplasm that is extremely rich in different vesicles that range in electron density from completely clear to quite dense. The nucleus (N) is below this area. Adjacent
397
cells appear to be closely apposed with a narrow line of extracellular matrix separating two cells (arrows). This particular section is immunostained, but at this magnification the gold particles are too small to be seen. A portion of the same cell is shown at a higher magnification in Fig. 6(B) with the lumen area in the upper right hand corner. The material in the lumen is abundantly labeled with individual gold particles. Among the numerous cytoplasmic vesicles (Vs) is one that is clearly labeled with gold particles. The same vesicle is shown at a higher magnification in the inset. Fig. 7 shows a cell without microvilli at the interface of the epithelial cell and the lumen. The translucent zone is much narrower and the peritrophic membrane, if it is intact and present, is less well defined. The cytoplasm underneath this zone is completely filled with vesicles of varying electron density. Three vesicles with a uniform gray content (arrowheads) are in the cytoplasm. These images clearly confirm our light microscopy results indicating that arcelin is internalized by the Z. subfasciatus larvae. Although internalization could be the consequence of a specific interaction of arc-1 with plasma membrane components, it could also be the result of fluid phase endocytosis of digested food. Since arcelins and other plant lectins are quite refractory to proteolysis, the protein would end up intact in the pinocytotic vesicles. King et al. (1986) observed internalization of phytohemagglutinin by epithelial cells of the rat small intestine and suggested that this process may serve to limit the absorption of such dietary toxins.
4. Discussion In the field, resistance involves a reduction and/or a delay in adult emergence, usually the result of a prolongation of the life cycle. Both an increase in the duration of the first and second instars of larval development and actual mortality of the larvae in the seeds have been observed. Examination of the Z. subfasciatus larvae developing on arc-1 seeds showed that the larvae were still in the first or second instar stage after 19 days (1.0 mm in length instead of the expected size of 2.5 mm for fourth instar larvae). These results were confirmed by feeding experiments in which purified arc-1 was admixed with the flour of Goiano Precoce to produce artificial seeds (data not shown). A dosage of 6.5% of arc-1 in artificial seeds was shown earlier to cause 50% mortality of the larvae (Posso et al., 1989). Furthermore, backcrossing arc-1 containing lines such as G12882 with domesticated lines, followed by selection for the presence of arc-1, results in the transfer of the arc-1 locus to the domesticated variety, and in the transfer of the resistance (Osborn et al., 1988a). Together these results support the interpretation that arc-1 is the major antibiosis factor for Z. subfasciatus in the G12882 wild
398
N.S. Paes et al. / Journal of Insect Physiology 46 (2000) 393–402
Fig. 4. Immunocytochemical detection of arc-1 in the midgut of larvae of A. obtectus allowed to develop on arc-1 containing seeds. (A) and (C): sections of the midgut stained with toluidine blue; (B), (D) and (F): sections of midgut immunolabelled with anti arc-1; (E): section of midgut of larvae allowed to develop on control (Goiano Precoce) seeds; L=lumen; HE=haemolymph; MU=muscle region; Ec=gut epithelial cells; Cr=crypts; Vs=vesicles. Asterisks in (B) indicate two adjacent epithelial layers [see (A)]. Bar=200 µm.
accession. To understand the effect of arcelin on the growth of the Z. subfasciatus larvae, we focused on arc1 because arc-5 is so toxic [Fig. 1(B)c] that it almost kills the larvae. We reasoned that it would be more instructive to examine larvae severely retarded in their growth by arc-1. One of the major findings of this work is that arc-1 disrupts the epithelial structure. Disruption of epithelium structure was also observed by Sauvion (1995, as referenced in Powell et al., 1998) who studied the effect of concanavalin A on the digestive tract of the aphid Acyrthosiphon pisum. They found epithelium cell distension, enlargement and shedding. Powell et al. (1998) also observed alteration of the structure of the midgut epithelium in brown planthoppers fed on a mannose specific
lectin (snowdrop lectin), with disruption of the microvilli and abnormalities in epithelial cells. The major difference between the arc-1 sensitive bruchid and the resistant one appears to be the presence of arc-1 in the haemolymph of the sensitive species. How can we account for the transport of arc-1 through the epithelium? It seems unlikely that internalization by small vesicles serves as the basis for transcellular transport because vesicles were observed in both species of bruchids. However, the disruption of epithelial structure in the sensitive bruchid may well permit the passage of arc-1 into the haemolymph. It is not clear whether the presence of arc-1 in the haemolymph contributes to poor larval development or is simply the result of epithelial disruption.
N.S. Paes et al. / Journal of Insect Physiology 46 (2000) 393–402
399
Fig. 5. Immunocytochemical detection of arc-1 in the midgut of larvae of Z. subfasciatus allowed to develop on arc-1 containing seeds. (A) and (C): sections of the midgut of larvae stained with toluidine blue; (B), (D), (E) and (F): sections of midgut of larvae immunolabelled with antibody anti-arc-1. L=lumen; HE=haemolymph; MU=muscle region; Ec=gut epithelial cells; Cr=crypts; Vs=vesicles. Bar=100 µm.
It would be of interest to search for arc-1 binding sites or receptors. Gatehouse et al. (1989) examined phytohemagglutinin binding sites on the epithelial cells of A. obtectus and C. maculatus larvae and found that this lectin binds only to the epithelial membranes of A. obtectus. Arc-1, although homologous to phytohemaglutinin, is at best a very weak lectin because it lacks most of the amino acids essential for metal binding and for sugar binding (Fabre et al., 1998). Hartweck et al. (1997) observed that arcelins do not agglutinate untreated erythrocytes and weakly agglutinate pronase-treated erythrocytes. Whether arcelins bind to the fucose and xylose containing complex glycans of insect glycoproteins remains to be determined. We are presently undertaking a search for arcelin receptors in extracts of Z. subfasciatus larvae on the assumption that proteins
that have a biological action must have receptors. Such receptors could be located on the plasma membranes of the epithelial cells (microvilli), in the peritrophic membrane or in the haemolymph. Mourey et al. (1998) recently suggested, on the basis of binding experiments of arc-1 and various glycoproteins (fetuin, thyroglobulin), that the protein may have an extended carbohydrate-binding site in the vicinity of the unreactive sugar-binding pocket. It is important to establish whether or not arcelins exert their biological activity as lectins, because it is well established in mammals that dietary lectins are powerful exogenous growth factors for the small intestine (for review, see Pusztai, 1993). Minney et al. (1990) suggested that the arcelin-containing beans may be resistant to Z. subfasciatus because the larvae lack the necessary gut proteases to digest this
400
N.S. Paes et al. / Journal of Insect Physiology 46 (2000) 393–402
Fig. 6. Ultrastructure of epithelial cells of Z. subfasciatus allowed to develop on arc-1 containing seeds and immunocytochemical localization of arc-1. (A) Low magnification view. The polar epithelial cell is vertically oriented with the food containing lumen (L) at the top. The peritrophic membrane (PM) separates the food in the lumen from the clear zone that contains the microvilli (MV). Arrowheads indicate the close apposition of adjacent cells. N=nucleus. Bar=1 µm. (B) A magnification of a portion of (A). Immunogold particles clearly label the lumen and a vesicle in the cytoplasm (arrowhead). Bar=1 µm.
abundant protein and that this may result in starvation. Studies with arc-1 revealed that this protein is indeed highly resistant to proteolysis (Fabre et al., 1998). However, in view of our finding that arc-1 damages the epithelial cells, we do not believe that protein starvation alone can account for the effect of arcelin of Z. subfasciatus larvae.
Acknowledgements We are thankful to Dr C. Cardona (CIAT, Colombia) for providing us with the wild bean accession G12882 seeds and Dr M. Yokoyama (CNPAF, Brazil) for Z. subfasciatus and A. obtectus larvae. This work has been supported by grants from the Brazilian government
N.S. Paes et al. / Journal of Insect Physiology 46 (2000) 393–402
401
Fig. 7. Immunocytochemical localization of arc-1 in epithelial cell of Z. subfasciatus larva. The lumen is in the left-hand upper corner and the micrograph shows primarily the area with vesicles. Some vesicles (arrowheads) are clearly labeled by gold particles. Bar=1 µm.
(EMBRAPA, CNPq and FAP-DF) and from the International Foundation for Science (IFS).
References Bardocz, S., Grant, G., Ewer, S.W., Duguid, T.J., Brown, D.S., Englyst, K., Pusztai, A., 1995. Reversible effect of phytohemagglutinin on the growth and metabolism of rat gastrointestinal tract. Gut 37, 353–360. Cardona, C., Posso, C.E., Kornegay, J., Valor, J., Serrano, M., 1989. Antibiosis effects of wild dry bean accessions on the Mexican bean weevil and the bean weevil (Coleoptera: Bruchidae). Journal of Economic Entomology 82, 310–315. Cardona, C., Kornegay, J., Posso, C.E., Morales, F., Ramirez, H., 1990. Comparative value of four arcelin variants in the development of dry bean lines resistant to the Mexican bean weevil. Entomologia experimentalis et applicata 56, 197–206. Fabre, C., Causse, H., Mourey, L., Koninkw, J., Rivie`re, M., Hendriks, H., Puzo, G., Samama, J.P., Rouge´, P., 1998. Characterization and sugar-binding properties of arcelin-1, an insecticidal lectin-like protein isolated from kidney bean (Phaseolus vulgaris L. cv. RAZ-2) seeds. Biochemical Journal 329, 551–560. Fory, L.F., Finardi-Filho, F., Quintero, C.M., Osborn, T.C., Cardona, C., Chrispeels, M.J., Mayer, J.E., 1996. α-Amylase inhibitors in resistance of common beans to the Mexican bean weevil and the bean weevil (Coleoptera: Bruchidae). Journal of Economic Entomology 89, 204–210. Gatehouse, A.M.R., Shackley, S.J., Fenton, K.A., Bryden, J., Pusztai, A., 1989. Mechanism of seed lectin tolerance by a major insect storage pest of Phaseolus vulgaris, Acanthoscelides obtectus. Journal of the Science of Food and Agriculture 35, 373–380. Grossi De Sa´, M.F., Mirkov, T.E., Ishimoto, M., Colluci, G., Bateman, K.S., Chrispeels, M.J., 1997. Molecular characterization of a bean α-amylase inhibitor that inhibits the α-amylase of the Mexican bean weevil Zabrotes subfasciatus. Planta 203, 295–303. Hartweck, L.M., Cardona, C., Osborn, T.C., 1997. Bruchid resistance
of common bean lines having an altered seed protein composition. Theoretical and Applied Genetics 95, 1018–1023. King, T.P., Pusztai, A., Grant, G., Slater, D., 1986. Immunogold localization of ingested kidney bean (Phaseolus vulgaris) lectins in epithelial cells of the rat small intestine. Histochemical Journal 18, 413–420. Kornegay, J., Cardona, C., 1991. Inheritance of resistance to Acanthoscelides obtectus in a wild common bean accession crossed to commercial bean cultivars. Euphytica 52, 103–111. Lioi, L., Bollini, R., 1989. Identification of a new arcelin variant in wild bean seeds. Bean Improvement Cooperative 32, 28. Minney, B.H.P., Gatehouse, A.M.R., Dobie, P., Dendy, J., Cardona, C., Gatehouse, J.A., 1990. Biochemical bases of seed resistance to Zabrotes subfasciatus (Bean weevil) in Phaseolus vulgaris (common bean); a mechanism for arcelin toxicity. Journal of Insect Physiology 36, 757–767. Mourey, L., Pe´delacq, J.-D., Fabre, C., Causse, H., Rouge´, P., 1997. Small-angle X-ray scattering and crystallographic studies of arcelin-1: an insecticidal lectin-like glycoprotein from Phaseolus vulgaris L. Proteins: Structure, Function and Genetics 29, 433–442. Mourey, L., Pe´delacq, J.-D., Birck, C., Fabre, C., Rouge´, P., Samama, J.-P., 1998. Crystal structure of the arcelin-1 dimer from Phaseolus ˚ resolution. Journal of Biological Chemistry 273, vulgaris at 1.9A 12914–12922. Osborn, T.C., Blake, T., Gepts, P., Bliss, F.A., 1986. Bean arcelin 2. Genetic variation, inheritance and linkage relationships of a novel seed protein of Phaseolus vulgaris L. Theoretical and Applied Genetics 71, 847–855. Osborn, T.C., Alexander, D.C., Sun, S.S.M., Cardona, C., Bliss, F.A., 1988a. Insecticidal activity and lectin homology of arcelin seed protein. Science 240, 207–210. Osborn, T.C., Burow, M., Bliss, F.A., 1988b. Purification and characterization of arcelin seed protein from common bean. Plant Physiology 86, 399–405. Peumans, W.J., Damme, E.J.M.V., 1995. Lectin as plant defense proteins. Plant Physiology 109, 347–352. Posso, C.E., Cardona, C., Valor, J.F., Morales, H., 1989. Caracterizacion de una nueva proteina como factor responsable de la resisten-
402
N.S. Paes et al. / Journal of Insect Physiology 46 (2000) 393–402
cia de Phaseolus vulgaris a Zabrotes subfasciatus (Boheman) (Coleoptera: Bruchidae). Revista Colombiana de Entomologia 15, 3–9. Powell, K.S., Spence, J., Bharathi, M., Gatehouse, J.A., Gatehouse, A.M.R., 1998. Immunohistochemical and developmental studies to elucidate the mecanism of action of the snowdrop lectin on the rice brown planthopper Nilaparvata lugens (Stal). Journal of Insect Physiology 44, 529–539. Pusztai, A., 1993. Dietary lectins are metabolic signals for the gut
and modulate immune and hormone functions. European Journal of Nutrition 47, 691–699. Santino, A., Valsasina, B., Lioi, L., Vitale, A., Bollini, R., 1991. Bean (Phaseolus vulgaris L.) seed lectins: a novel electrophoretic variant of arcelin. Plant Physiology (Life Sc. Adv.) 10, 7–11. Schoonhoven, A.V., Cardona, C., Valor, J., 1983. Resistance to the bean weevil and the Mexican bean weevil (Coleoptera: Bruchidae) in noncultivated common bean accessions. Journal of Economic Entomology 76, 1255–1259.
The effect of arcelin-1 on the structure of the midgut of bruchid larvae and immunolocalization of the arcelin protein Norma S. Paes a, Isabel R. Gerhardt a, Marise V. Coutinho a, Massaru Yokoyama a,b, Eliana Santana a, Nicholas Harris a,c, Maarten J. Chrispeels a,d, M. Fatima Grossi de Sa a,* a
d
Embrapa/Cenargen, PO Box 02372, Brasilia, DF, Brazil b EMBRAPA/CNPAF, Goiania, GO, Brazil c Department of Biological Sciences, University of Durham, South Road, Durham, DH1 3LE, UK Department of Biology, University of California San Diego, 9500 Gilman Drive, La Jolla, CA, 92093-0116, USA Received 6 January 1999; accepted 9 March 1999
Abstract Some wild accessions of the common bean (Phaseolus vulgaris) contain a family of proteins called arcelins, that are toxic to the larvae of certain bruchid species. Among the six allelic variants of arcelin tested so far, arcelin-5 and arcelin-1 confer the highest level of resistance against the Mexican bean weevil, Zabrotes subfasciatus. The same proteins are not toxic to the bean weevil, Acanthoscelides obtectus, which is also a serious pest of cultivated beans. Arcelins belong to the bean lectin family that includes phytohemaggutinins and α-amylase inhibitors. Although homologous to lectins, arcelins are themselves only very weak lectins, and their binding properties have not been clearly established. The toxic properties of arcelins may be related to their recognition of and interaction with the glycoproteins and other constituents of the membranes along the digestive tract of insects. Since arcelin-1 was shown to have growth inhibitory effects for the larvae of Z. subfasciatus but not of A. obtectus, we examined the effect of an arcelin-1 containing diet on the structure of the cells that line the intestinal tract of the larvae of these two bruchid species, and used antibodies against arcelin to examine the distribution of arcelin within the cells and tissues. Here we show that dietary arcelin-1 caused an alteration of the gut structure and the penetration of arcelin into the haemolymph in Z. subfasciatus but not in A. obtectus. These results lead us to suggest that arcelins exert their toxic effect by severely damaging the epithelial cells. 2000 Elsevier Science Ltd. All rights reserved. Keywords: Bruchid beetles; Zabrotes subfasciatus; Arcelin; Immunolocalization; Midgut
1. Introduction Bruchids are the most serious pests attacking food legume seeds during storage. Substantial amounts of stored common bean, Phaseolus vulgaris (L.), are lost due to the damage caused by the two main storage pests, the Mexican bean weevil, Zabrotes subfasciatus (Boheman) and the bean weevil Acanthoscelides obtectus (Say) (Cardona et al., 1989). In some wild accessions of P. vulgaris that show variable levels of resistance to Z. subfasciatus, arcelin seed proteins replace the normal seed storage protein, phaseolin (Osborn et al., 1986).
* Corresponding author. Fax: +55-61-340-3624. E-mail address: [email protected] de Sa)
(M.F. Grossi
Subsequently, arcelins were shown to have inhibitory effects on the larval growth of Z. subfasciatus and to account for the resistance of the seeds of these wild accessions (Osborn et al., 1988a; Cardona et al., 1990). Although some wild bean accessions containing arcelin are also highly resistant to A. obtectus (Schoonhoven et al., 1983; Cardona et al., 1989; Kornegay and Cardona, 1991), arcelins appear not to be the factor involved in this resistance (Osborn et al., 1988a). Six electrophoretic variants of arcelin, designated arcelin-1 to 6, have been reported (Osborn et al., 1986; Lioi and Bollini, 1989; Santino et al., 1991) and among them, arcelin-5 and arcelin-1 exhibit the highest level of bruchid resistance (Cardona et al., 1990; Fory et al., 1996). A comparison of the amino acid sequences of arcelins with other seed proteins shows that they belong to the bean lectin family that includes phytohemagglutinin and
0022-1910/00/$ - see front matter 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 2 - 1 9 1 0 ( 9 9 ) 0 0 1 2 2 - 5
394
N.S. Paes et al. / Journal of Insect Physiology 46 (2000) 393–402
α-amylase inhibitor. Arcelins are not strong lectins, but may have weak lectin activity (Osborn et al., 1988b). Xray diffraction analysis shows that arcelin-1 molecules form a lectin-like dimer (Mourey et al., 1997). The dimeric structure is particularly suited to a function in molecular recognition, since it might allow the bridging of cells through interactions with membrane-associated glycoproteins or glycolipids (Mourey et al., 1998). The epithelial cells along the digestive tract of phytophagous insects are directly exposed to the contents of the diet, and therefore, are possible target sites for plant lectins and lectin-like proteins (Peumans and Damme, 1995). The mechanism of action of arcelins is not known, but in analogy with phytohemagglutinin, a protein that is toxic to mammals by virtue of its binding to the intestinal epithelium (Bardocz et al., 1995), we postulate that the toxicity of arcelins may be based on specific binding to glycoconjugates or proteins somewhere in the gut of the insect. If arcelins behave as lectins, then two types of interactions could be possible: binding of arcelins to glycoconjugates exposed on the epithelial cells along the digestive tract, and binding of arcelins to glycosylated digestive enzymes (Peumans and Damme, 1995). Fabre et al. (1998) proposed that the Asn-linked glycans of arcelin-1 that are responsible for its binding to mannosespecific lectins, known to occur in various insects, may account for its toxicity. To understand why arcelin-1 is toxic for Z. subfasciatus larvae but not for A. obtectus larvae, we conducted a study of the effects of dietary arcelin on the structure of the larval gut and determined the distribution of arcelin in the larval tissues with specific antibodies. Our results show that arcelin-1 had severe deleterious effects on the gut of Z. subfasciatus, but not on the gut of A. obtectus, and that arcelin was present in the haemolymph of Z. subfasciatus, indicating that it had traversed the cells that line the gut.
2. Materials and methods 2.1. Feeding tests Bruchid cultures were reared on the susceptible P. vulgaris cultivar Goiano Precoce and on the resistant wild accession of P. vulgaris G12882 (containing arc-1) and the wild accession G02771 (containing arc-5) The cultures were mantained in petri dishes in a controlled environment at 27°C and 60% relative humidity. 2.2. Protein extraction and immunoblots Proteins were extracted from bean seeds according to Grossi de Sa´ et al. (1997), separated by SDS–PAGE 13%, and transferred to nitrocellulose. The immunoblots were made with a polyclonal mouse serum against chemically deglycosylated purified arc-1. The antibodies
bound to the immunoblot were visualized with horseradish–peroxidase-conjugated secondary antibody (BioRad) using the manufacturer’s specifications. 2.3. Histology and immunoblots Larvae of 19 days of Z. subfasciatus and A. obtectus, fed on susceptible and resistant seeds, were washed in N-heptan saturated with fixation buffer (3% paraformaldehyde, 0.5% glutaraldehyde in sodium cacodylate buffer 0.05 M, pH 7.2) for 5–10 min, followed by fixation in the same fixation buffer for 24 h, alternating between shaking and vacuum. The tissue was then washed in sodium cacodylate buffer (3×15 min each) and dehydrated in ethanol/water at 10, 20, 30, 50, 70, 90 and 100% ethanol, and finally embedded in the hydrophilic resin LR White. Semi-thin sections (approx. 1 µm) were cut for light microscopy and stained with toluidine blue for structural examination of the tissues. Unstained sections were used for immunolabelling to investigate the effects of arcelin on the growth and development of the larvae of Z. subfasciatus and A. obtectus. The distribution of specifically-bound primary antibody (anti-arc-1) was determined using secondary antibody conjugated with colloidal gold (5 nm) with the signal amplified by silver enhancement.
3. Results 3.1. Arc-1 slows the development of Z. subfasciatus larvae To investigate the effects of arcelins on growth and development of larvae of Z. subfasciatus, we examined larvae that were allowed to develop normally in a control variety of beans (Goiano Precoce) and larvae that were allowed to develop in the seeds of wild accessions containing arcelin-1 or arcelin-5. We compared larval sizes 19 days after infestation (egg laying) with those of a second bruchid, A. obtectus. The results (Fig. 1) show that the larvae of both bruchid species were largest on the domesticated bean variety that does not contain any arcelins [Fig. 1(B)a and (C)a]. The growth of Z. subfasciatus larvae was substantially inhibited on the arc-1 containing seeds and even much more so on the arc-5 containing seeds. This bean accession clearly has the highest level of antibiosis factors against Z. subfasciatus. Larvae fed with seeds of acessions G12882 and G02771 had an average weight of 1.63 and 0.35 mg, respectively. In contrast, larvae fed with control seeds had an average weight of 4.96 mg (Gerhardt et al., unpublished results). The growth of the A. obtectus larvae was slightly inhibited on the two wild accessions, but there was no difference between the two accessions. These results
N.S. Paes et al. / Journal of Insect Physiology 46 (2000) 393–402
395
(control) and on arc-1 containing seeds is shown in Fig. 2(A), (C) and (E) and 2(B), (D) and (F), respectively. Fig. 2(A), a section through the midgut of a control larva, shows two cross-sections of the gut wall surrounding the intestinal lumen (L). Two cross-sections appear in a single micrograph because the gut forms a loop when the larvae are fixed and embedded. In Fig. 2(C), the epithelial cells that form the gut wall have a normal polar appearance, with prominent nuclei (N) in the center and numerous small translucent vacuoles/droplets toward the apical side. The cells appear to be secreting large dense-staining droplets (Dr). The epithelial cells have microvilli (MV) that stick out into the crypts (Cr) and these areas are more lightly stained than the cells themselves. Microvilli are clearly visible in Fig. 2(E). Adjacent epithelial cells appear to be closely appressed and only a narrow white line of extracellular matrix is visible between the cells [arrows in Fig. 2(E)]. The lumen contains some stained material that may be food in the process of being digested and/or digestive enzymes secreted by the cells that line the intestine. Some regions of the midgut [Fig. 2(B)] of larvae allowed to develop on arc-1 containing seeds were similar in appearance to the midgut of the larvae on control seeds, except that the lumen contained more stained material and that crystalline inclusions (CI) were present. Secretory droplets (Dr) were present near the surface of the epithelial cells (Ec). In other regions shown in Fig. 2(D) and (F), the cells had a much less regular appearance and were not closely appressed. They were separated by broad channels, and the appearance of the translucent droplets/vacuoles was much less regular. The densely staining secreted droplets in the lumen were similar in size and distribution as in the controls. Thus, it appeared as if secretion was intense, but digestion was slow as evidenced by the accumulation of undigested food in the lumen. In addition, the structure of the epithelium appeared disrupted. 3.3. Arc-1 penetrates into the haemolymph in Z. subfasciatus larvae fed with arc-1 seeds Fig. 1. (A) Goiano Precoce seed damaged by the larva of a bruchid beetle Z. subfasciatus. Inhibitory effect of arc-1 and arc-5 protein on the growth of larvae of Z. subfasciatus (B) and A. obtectus (C) after 19 days of infestation. a: Larvae allowed to develop on P. vulgaris cultivar Goiano Precoce seeds; b: larvae allowed to develop on wild bean P. vulgaris seeds containing arc-1; c: larvae allowed to develop on wild bean P. vulgaris containing arc-5. Bars=1 mm.
indicate that these two accessions contain no or very low levels of antibiosis factors against A. obtectus. 3.2. Arc-1 disrupts the integrity of the midgut epithelium of Z. subfasciatus The structure of the midgut of Z. subfasciatus larvae allowed to develop for 19 days on Goiano Precoce seeds
Because the distribution of arcelin within the tissues of the larvae may tell us something about its mode of action, we determined the location of arcelin by immunocytochemistry with a serum prepared against deglycosylated arcelin. The specificity of this serum is shown in Fig. 3. Fig. 3(A) shows the proteins present in extracts of G12882 (a) and Goiano Precoce (b) seeds separated by SDS–PAGE. In the adjacent immunoblot, the antiserum recognizes arcelin in extracts of the G12882 seeds, but does not recognize any protein in the Goiano Precoce seeds [Fig. 3(B)], indicating that the serum is specific for arcelin and does not react with other members of this protein family such as phytohemagglutinin and α-amylase inhibitor. Both of these proteins are
396
N.S. Paes et al. / Journal of Insect Physiology 46 (2000) 393–402
Fig. 2. Structure of the midgut epithelium of 19 day-old larvae of Z. subfasciatus allowed to develop on cultivar Goiano Precoce seeds (A, C, E) or on arc-1 containing seeds (B, D, F) visualized with toluidine blue. N=nuclei; L=lumen; MV=microvilli; Ec=gut epithelial cells; CI=cristalline inclusions; Cr=crypts; MU=muscle; Dr=dense droplets. Arrowheads in (E) indicate close apposition of adjacent cells. Arrowheads in (F) indicate broad channels between adjacent cells. Bars in (A)–(D)=150 µm; in (E) and (F)=50 µm.
present in cultivars of the common bean such as Goiano Precoce. We used this serum to localize arcelin-1 in the tissues of Z. subfasciatus larvae allowed to develop on the arc-1 containing seeds. As a control, we could not use larvae that develop on Goiano Precoce seeds, because those seeds have no arcelin. We therefore determined the distribution of arc-1 in the tissues of A. obtectus larvae that were allowed to develop on arc-1 seeds. We considered this to be a suitable control because these larvae are not or only minimally affected by arc-1 and develop normally [Fig. 1(B)b]. The structure of the gut of A. obtectus larvae as seen at low magnification in toluidine blue stained sections [Fig. 4(A) and (C)] is similar to that of Z. subfasciatus. After immunostaining with antibodies to arc-1, we found, as expected, strong immunolabeling of the gut content in the lumen of the gut [Fig. 4(B) and (D)] but without apparent staining of the cells. Fig. 4(A) and (B),
and (C) and (D) are matched pairs in which one is stained with toluidine blue and the other one is immunostained. This immunostaining is thought to represent undigested arc-1. The justaposition of unstained cells and stained lumen content is most clearly visible in Fig. 4(B) (see asterisk). Labeling of arcelin is also visible in crypts (Cr) of the epithelium [Fig. 4(D)]. Immunostaining is also found in structures that we interpret as vesicles (Vs), containing arc-1 that has been internalized by the cells. The haemolymph that surrounds the gut is completely free of immunostaining. Fig. 4(E) is a section through the midgut of an A. obtectus larvae that has been allowed to develop on Goiano Precoce seeds. There is as expected no immunostaining because this bean variety has no arcelin or cross reacting proteins. When examining the Z. subfasciatus larvae, we also found strong labeling of the gut content in the lumen.
N.S. Paes et al. / Journal of Insect Physiology 46 (2000) 393–402
Fig. 3. (A) Coomassie blue stained gel of total protein extract from a: arc-1 seeds and b: cultivar Goiano Precoce seeds. (B) Same gel as in (A), transferred to nitrocellulose membrane and immunoreacted with anti-arc-1 serum. Arrows correspond to molecular weight markers of 66, 45, and 29 kDa.
Fig. 5(A) and (C), and (B) and (D) are once again matched pairs with one set stained with toluidine blue and the other one immunostained. The black labeling of the lumen (L) is very obvious in Fig. 5(B) and (D). A lower level of staining is readily apparent in the haemolymph (HE). Higher magnification of cells with crypts [Fig. 5(E) and (F)] shows labeling of the crypts and of apparent vesicles (Vs) in the cells. Internalization of arcelin and presumably other dietary components into vesicles is a feature of both the sensitive and the resistant bruchid species. 3.4. Arc-1 is present in cytoplasmic vesicles We used immuno-electron microscopy to examine the ultrastructure of the epithelial cells of Z. subfasciatus allowed to develop on resistant seeds and to confirm the existence of vesicles that internalize arcelin. The results are shown in Fig. 6(A) and (B). Fig. 6(A) shows a low magnification image of the interface between the lumen (L) and the epithelial cells (Ec). The peritrophic membrane (PM) separates the food in the lumen from the zone with the microvilli (MV). The microvilli are embedded in a clear zone underneath the peritrophic membrane. Underneath the microvillar zone is an area of cytoplasm that is extremely rich in different vesicles that range in electron density from completely clear to quite dense. The nucleus (N) is below this area. Adjacent
397
cells appear to be closely apposed with a narrow line of extracellular matrix separating two cells (arrows). This particular section is immunostained, but at this magnification the gold particles are too small to be seen. A portion of the same cell is shown at a higher magnification in Fig. 6(B) with the lumen area in the upper right hand corner. The material in the lumen is abundantly labeled with individual gold particles. Among the numerous cytoplasmic vesicles (Vs) is one that is clearly labeled with gold particles. The same vesicle is shown at a higher magnification in the inset. Fig. 7 shows a cell without microvilli at the interface of the epithelial cell and the lumen. The translucent zone is much narrower and the peritrophic membrane, if it is intact and present, is less well defined. The cytoplasm underneath this zone is completely filled with vesicles of varying electron density. Three vesicles with a uniform gray content (arrowheads) are in the cytoplasm. These images clearly confirm our light microscopy results indicating that arcelin is internalized by the Z. subfasciatus larvae. Although internalization could be the consequence of a specific interaction of arc-1 with plasma membrane components, it could also be the result of fluid phase endocytosis of digested food. Since arcelins and other plant lectins are quite refractory to proteolysis, the protein would end up intact in the pinocytotic vesicles. King et al. (1986) observed internalization of phytohemagglutinin by epithelial cells of the rat small intestine and suggested that this process may serve to limit the absorption of such dietary toxins.
4. Discussion In the field, resistance involves a reduction and/or a delay in adult emergence, usually the result of a prolongation of the life cycle. Both an increase in the duration of the first and second instars of larval development and actual mortality of the larvae in the seeds have been observed. Examination of the Z. subfasciatus larvae developing on arc-1 seeds showed that the larvae were still in the first or second instar stage after 19 days (1.0 mm in length instead of the expected size of 2.5 mm for fourth instar larvae). These results were confirmed by feeding experiments in which purified arc-1 was admixed with the flour of Goiano Precoce to produce artificial seeds (data not shown). A dosage of 6.5% of arc-1 in artificial seeds was shown earlier to cause 50% mortality of the larvae (Posso et al., 1989). Furthermore, backcrossing arc-1 containing lines such as G12882 with domesticated lines, followed by selection for the presence of arc-1, results in the transfer of the arc-1 locus to the domesticated variety, and in the transfer of the resistance (Osborn et al., 1988a). Together these results support the interpretation that arc-1 is the major antibiosis factor for Z. subfasciatus in the G12882 wild
398
N.S. Paes et al. / Journal of Insect Physiology 46 (2000) 393–402
Fig. 4. Immunocytochemical detection of arc-1 in the midgut of larvae of A. obtectus allowed to develop on arc-1 containing seeds. (A) and (C): sections of the midgut stained with toluidine blue; (B), (D) and (F): sections of midgut immunolabelled with anti arc-1; (E): section of midgut of larvae allowed to develop on control (Goiano Precoce) seeds; L=lumen; HE=haemolymph; MU=muscle region; Ec=gut epithelial cells; Cr=crypts; Vs=vesicles. Asterisks in (B) indicate two adjacent epithelial layers [see (A)]. Bar=200 µm.
accession. To understand the effect of arcelin on the growth of the Z. subfasciatus larvae, we focused on arc1 because arc-5 is so toxic [Fig. 1(B)c] that it almost kills the larvae. We reasoned that it would be more instructive to examine larvae severely retarded in their growth by arc-1. One of the major findings of this work is that arc-1 disrupts the epithelial structure. Disruption of epithelium structure was also observed by Sauvion (1995, as referenced in Powell et al., 1998) who studied the effect of concanavalin A on the digestive tract of the aphid Acyrthosiphon pisum. They found epithelium cell distension, enlargement and shedding. Powell et al. (1998) also observed alteration of the structure of the midgut epithelium in brown planthoppers fed on a mannose specific
lectin (snowdrop lectin), with disruption of the microvilli and abnormalities in epithelial cells. The major difference between the arc-1 sensitive bruchid and the resistant one appears to be the presence of arc-1 in the haemolymph of the sensitive species. How can we account for the transport of arc-1 through the epithelium? It seems unlikely that internalization by small vesicles serves as the basis for transcellular transport because vesicles were observed in both species of bruchids. However, the disruption of epithelial structure in the sensitive bruchid may well permit the passage of arc-1 into the haemolymph. It is not clear whether the presence of arc-1 in the haemolymph contributes to poor larval development or is simply the result of epithelial disruption.
N.S. Paes et al. / Journal of Insect Physiology 46 (2000) 393–402
399
Fig. 5. Immunocytochemical detection of arc-1 in the midgut of larvae of Z. subfasciatus allowed to develop on arc-1 containing seeds. (A) and (C): sections of the midgut of larvae stained with toluidine blue; (B), (D), (E) and (F): sections of midgut of larvae immunolabelled with antibody anti-arc-1. L=lumen; HE=haemolymph; MU=muscle region; Ec=gut epithelial cells; Cr=crypts; Vs=vesicles. Bar=100 µm.
It would be of interest to search for arc-1 binding sites or receptors. Gatehouse et al. (1989) examined phytohemagglutinin binding sites on the epithelial cells of A. obtectus and C. maculatus larvae and found that this lectin binds only to the epithelial membranes of A. obtectus. Arc-1, although homologous to phytohemaglutinin, is at best a very weak lectin because it lacks most of the amino acids essential for metal binding and for sugar binding (Fabre et al., 1998). Hartweck et al. (1997) observed that arcelins do not agglutinate untreated erythrocytes and weakly agglutinate pronase-treated erythrocytes. Whether arcelins bind to the fucose and xylose containing complex glycans of insect glycoproteins remains to be determined. We are presently undertaking a search for arcelin receptors in extracts of Z. subfasciatus larvae on the assumption that proteins
that have a biological action must have receptors. Such receptors could be located on the plasma membranes of the epithelial cells (microvilli), in the peritrophic membrane or in the haemolymph. Mourey et al. (1998) recently suggested, on the basis of binding experiments of arc-1 and various glycoproteins (fetuin, thyroglobulin), that the protein may have an extended carbohydrate-binding site in the vicinity of the unreactive sugar-binding pocket. It is important to establish whether or not arcelins exert their biological activity as lectins, because it is well established in mammals that dietary lectins are powerful exogenous growth factors for the small intestine (for review, see Pusztai, 1993). Minney et al. (1990) suggested that the arcelin-containing beans may be resistant to Z. subfasciatus because the larvae lack the necessary gut proteases to digest this
400
N.S. Paes et al. / Journal of Insect Physiology 46 (2000) 393–402
Fig. 6. Ultrastructure of epithelial cells of Z. subfasciatus allowed to develop on arc-1 containing seeds and immunocytochemical localization of arc-1. (A) Low magnification view. The polar epithelial cell is vertically oriented with the food containing lumen (L) at the top. The peritrophic membrane (PM) separates the food in the lumen from the clear zone that contains the microvilli (MV). Arrowheads indicate the close apposition of adjacent cells. N=nucleus. Bar=1 µm. (B) A magnification of a portion of (A). Immunogold particles clearly label the lumen and a vesicle in the cytoplasm (arrowhead). Bar=1 µm.
abundant protein and that this may result in starvation. Studies with arc-1 revealed that this protein is indeed highly resistant to proteolysis (Fabre et al., 1998). However, in view of our finding that arc-1 damages the epithelial cells, we do not believe that protein starvation alone can account for the effect of arcelin of Z. subfasciatus larvae.
Acknowledgements We are thankful to Dr C. Cardona (CIAT, Colombia) for providing us with the wild bean accession G12882 seeds and Dr M. Yokoyama (CNPAF, Brazil) for Z. subfasciatus and A. obtectus larvae. This work has been supported by grants from the Brazilian government
N.S. Paes et al. / Journal of Insect Physiology 46 (2000) 393–402
401
Fig. 7. Immunocytochemical localization of arc-1 in epithelial cell of Z. subfasciatus larva. The lumen is in the left-hand upper corner and the micrograph shows primarily the area with vesicles. Some vesicles (arrowheads) are clearly labeled by gold particles. Bar=1 µm.
(EMBRAPA, CNPq and FAP-DF) and from the International Foundation for Science (IFS).
References Bardocz, S., Grant, G., Ewer, S.W., Duguid, T.J., Brown, D.S., Englyst, K., Pusztai, A., 1995. Reversible effect of phytohemagglutinin on the growth and metabolism of rat gastrointestinal tract. Gut 37, 353–360. Cardona, C., Posso, C.E., Kornegay, J., Valor, J., Serrano, M., 1989. Antibiosis effects of wild dry bean accessions on the Mexican bean weevil and the bean weevil (Coleoptera: Bruchidae). Journal of Economic Entomology 82, 310–315. Cardona, C., Kornegay, J., Posso, C.E., Morales, F., Ramirez, H., 1990. Comparative value of four arcelin variants in the development of dry bean lines resistant to the Mexican bean weevil. Entomologia experimentalis et applicata 56, 197–206. Fabre, C., Causse, H., Mourey, L., Koninkw, J., Rivie`re, M., Hendriks, H., Puzo, G., Samama, J.P., Rouge´, P., 1998. Characterization and sugar-binding properties of arcelin-1, an insecticidal lectin-like protein isolated from kidney bean (Phaseolus vulgaris L. cv. RAZ-2) seeds. Biochemical Journal 329, 551–560. Fory, L.F., Finardi-Filho, F., Quintero, C.M., Osborn, T.C., Cardona, C., Chrispeels, M.J., Mayer, J.E., 1996. α-Amylase inhibitors in resistance of common beans to the Mexican bean weevil and the bean weevil (Coleoptera: Bruchidae). Journal of Economic Entomology 89, 204–210. Gatehouse, A.M.R., Shackley, S.J., Fenton, K.A., Bryden, J., Pusztai, A., 1989. Mechanism of seed lectin tolerance by a major insect storage pest of Phaseolus vulgaris, Acanthoscelides obtectus. Journal of the Science of Food and Agriculture 35, 373–380. Grossi De Sa´, M.F., Mirkov, T.E., Ishimoto, M., Colluci, G., Bateman, K.S., Chrispeels, M.J., 1997. Molecular characterization of a bean α-amylase inhibitor that inhibits the α-amylase of the Mexican bean weevil Zabrotes subfasciatus. Planta 203, 295–303. Hartweck, L.M., Cardona, C., Osborn, T.C., 1997. Bruchid resistance
of common bean lines having an altered seed protein composition. Theoretical and Applied Genetics 95, 1018–1023. King, T.P., Pusztai, A., Grant, G., Slater, D., 1986. Immunogold localization of ingested kidney bean (Phaseolus vulgaris) lectins in epithelial cells of the rat small intestine. Histochemical Journal 18, 413–420. Kornegay, J., Cardona, C., 1991. Inheritance of resistance to Acanthoscelides obtectus in a wild common bean accession crossed to commercial bean cultivars. Euphytica 52, 103–111. Lioi, L., Bollini, R., 1989. Identification of a new arcelin variant in wild bean seeds. Bean Improvement Cooperative 32, 28. Minney, B.H.P., Gatehouse, A.M.R., Dobie, P., Dendy, J., Cardona, C., Gatehouse, J.A., 1990. Biochemical bases of seed resistance to Zabrotes subfasciatus (Bean weevil) in Phaseolus vulgaris (common bean); a mechanism for arcelin toxicity. Journal of Insect Physiology 36, 757–767. Mourey, L., Pe´delacq, J.-D., Fabre, C., Causse, H., Rouge´, P., 1997. Small-angle X-ray scattering and crystallographic studies of arcelin-1: an insecticidal lectin-like glycoprotein from Phaseolus vulgaris L. Proteins: Structure, Function and Genetics 29, 433–442. Mourey, L., Pe´delacq, J.-D., Birck, C., Fabre, C., Rouge´, P., Samama, J.-P., 1998. Crystal structure of the arcelin-1 dimer from Phaseolus ˚ resolution. Journal of Biological Chemistry 273, vulgaris at 1.9A 12914–12922. Osborn, T.C., Blake, T., Gepts, P., Bliss, F.A., 1986. Bean arcelin 2. Genetic variation, inheritance and linkage relationships of a novel seed protein of Phaseolus vulgaris L. Theoretical and Applied Genetics 71, 847–855. Osborn, T.C., Alexander, D.C., Sun, S.S.M., Cardona, C., Bliss, F.A., 1988a. Insecticidal activity and lectin homology of arcelin seed protein. Science 240, 207–210. Osborn, T.C., Burow, M., Bliss, F.A., 1988b. Purification and characterization of arcelin seed protein from common bean. Plant Physiology 86, 399–405. Peumans, W.J., Damme, E.J.M.V., 1995. Lectin as plant defense proteins. Plant Physiology 109, 347–352. Posso, C.E., Cardona, C., Valor, J.F., Morales, H., 1989. Caracterizacion de una nueva proteina como factor responsable de la resisten-
402
N.S. Paes et al. / Journal of Insect Physiology 46 (2000) 393–402
cia de Phaseolus vulgaris a Zabrotes subfasciatus (Boheman) (Coleoptera: Bruchidae). Revista Colombiana de Entomologia 15, 3–9. Powell, K.S., Spence, J., Bharathi, M., Gatehouse, J.A., Gatehouse, A.M.R., 1998. Immunohistochemical and developmental studies to elucidate the mecanism of action of the snowdrop lectin on the rice brown planthopper Nilaparvata lugens (Stal). Journal of Insect Physiology 44, 529–539. Pusztai, A., 1993. Dietary lectins are metabolic signals for the gut
and modulate immune and hormone functions. European Journal of Nutrition 47, 691–699. Santino, A., Valsasina, B., Lioi, L., Vitale, A., Bollini, R., 1991. Bean (Phaseolus vulgaris L.) seed lectins: a novel electrophoretic variant of arcelin. Plant Physiology (Life Sc. Adv.) 10, 7–11. Schoonhoven, A.V., Cardona, C., Valor, J., 1983. Resistance to the bean weevil and the Mexican bean weevil (Coleoptera: Bruchidae) in noncultivated common bean accessions. Journal of Economic Entomology 76, 1255–1259.