The Role of Thionins in Plant Protection.pdf

Critical Reviews in Plant Sciences ISSN: 0735-2689 (Print) 1549-7836 (Online) Journal homepage: https://www.tandfonline.com/loi/bpts20 The Role of Thionins in Plant Protection Holger Bohlmann & Dr.WilliamBroekaertF. A. To cite this article: Holger Bohlmann & Dr.WilliamBroekaertF. A. (1994) The Rol...

Critical Reviews in Plant Sciences ISSN: 0735-2689 (Print) 1549-7836 (Online) Journal homepage: https://www.tandfonline.com/loi/bpts20 The Role of Thionins in Plant Protection Holger Bohlmann & Dr.WilliamBroekaertF. A. To cite this article: Holger Bohlmann & Dr.WilliamBroekaertF. A. (1994) The Role of Thionins in Plant Protection, Critical Reviews in Plant Sciences, 13:1, 1-16, DOI: 10.1080/07352689409701905 To link to this article: https://doi.org/10.1080/07352689409701905 Published online: 30 Mar 2011. Submit your article to this journal Article views: 111 View related articles Citing articles: 15 View citing articles Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=bpts20 Critical Reviews in Plunr Sciences, 13(1):1-16 (1994) The Role of Thionins in Plant Protection Holger Bohlmann lnstitut fur Pflanzenwissenschaften, ETH Zurich, LFW D.58, Universitatsstrasse 2, CH-8092 Zurich, Switzerland Referee: Dr. William Broekaert, F. A. Janssens Laboratory of Genetics, Catholic University of Leuven, 8-3001 Heverlee, Belgium ABSTRACT: Thionins are a group of small (SO00 Da), sulfur-rich plant proteins found mainly in cereals and mistletoes. Their three-dimensional structures are very compact and amphipathic, stabilized by three or four disulfide bridges. Thionins are usually basic and exert toxicity in various biological systems by destroying membranes. Thionins are synthesized as preproproteins and secreted into vacuoles, protein bodies, and the cell wall. Their antibacterial and antifungal activities point to a role as plant defense proteins. Support for this possible function comes particularly from work on the leaf thionins of barley, showing that these proteins can be induced by several stress factors. Infection of barley with mildew, one of its most devastating pathogens, leads to an incorporation of leaf thionins into papillae in incompatible interactions. The possible role of thionins to enhance the resistance of crop plants by genetic engineering is discussed. KEY WORDS: toxin, plant defense proteins, sulfur-rich proteins, secretion, preproprotein. 1. INTRODUCTION Plants are continuously exposed to a multitude of pests and pathogens that attack them mechanically and chemically. In particular, bacteria and fungi use a variety of chemical weapons, including toxins and hydrolytic enzymes, to attack plants. The plants, on the other hand, try to establish physical barriers, e.g., papillae (Aist, 1983), against the introduction of pathogens and also counterattack with chemical weapons of their own. The plants’ resistance compounds can be divided into a constitutive or preinfection class, and an inducible or postinfection class if the accumulation is initiated after recognition of the pathogen. Both groups include a variety of low molecular weight chemicals and proteins; Harborne (1988) lists about 20,000 different “secondary” plant compounds involved in plant-animal interactions. The low molecular weight compounds that are induced after attack by microorganisms are termed phytoalexins. These compounds are chemically diverse and differ according to the systematic classification of the plant. The proteins involved in the resistance of plants include enzyme inhibitors and lectins, which are widely distributed in the seeds as preinfection defense proteins. Examples of inducible defense proteins include the protease inhibitors of tomato (Ryan, 1990). Another well-known, diverse group of inducible proteins are the so-called PR (“pathogenesis-related”) proteins (for a recent review, see Linthorst [19911). These include chitinases and glucanases, whose enzymatic activities are thought to be mainly directed against the cell walls of attacking fungi. The exact function of the other PR-proteins is not yet known. Another group of proteins that seems to be involved in the resistance of plants against fungal and bacterial pathogens is the thionins. These proteins are small (5000 Da), basic, cysteine-rich, and have toxic and antimicrobial activities. The thionins are the subject of this review, with particular emphasis on their presumed function as defense proteins. In addition, because there is considerable interest in increasing the resistance of crop plants to pathogens by genetic engineering, the prospects of using thionins in this regard also are discussed. Two other reviews concerning thionins have recently appeared (Garcia-Olmedo et al., 1989; Bohlmann and Apel, 1991).The review by GarciaOlmedo et al. (1989) focuses mainly on endosperm thionins and should be consulted for more information about the genetics of endosperm thionins in cereals. Another review by Garcia-Olmedo et al. (1992) is restricted to the barley thionins. 0735-2689/94/$.50 0 1994 by CRC Press, Inc. 1 II. DISTRIBUTION teins, which they named viscotoxin. Later, Samuelsson purified three different viscotoxins from this mixture and determined their amino acid sequences (see Figure 1). The sequences clearly show that these viscotoxins are homologous to the cereal thionins and thus can be grouped together in the thionin protein family. Konopa et al. (1980) claimed to have separated the crude viscotoxin preparation into four distinct viscotoxins, but their sequences were not determined; it seems possible that even more viscotoxins might be found in V . album upon further investigation. Samuelsson (1966, 1969) also found basic proteins with similar toxic effects in other members of the family Viscaceae (also called Loranthaceae). Some of these proteins have been isolated and sequenced by his research group (see Figure l), demonstrating that they are very similar to the thionins from V.album. This makes the viscotoxins from mistletoes the second large group of thionins besides the cereal thionins. Another thionin was found in the seeds of the parasitic plant Pyrularia pubera (Vernon et al., 1985), which belongs to the family Santalaceae and thus to the same order (Santales) as the mistletoes of the family Viscaceae. The primary structure of this Pyrularia thionin is somewhat intermediate between that of the viscotoxins and that of the cereal thionins (see Figure 1). Preliminary results indicate that the Pyrularia thionin or related thionins also may occur in the leaves and roots of this plant (Vernon et al., 1985). A special position among the thionins is occupied by crambin (actually at least two different proteins; see Figure 1) because it is neutral and Subsequent to the first report of the isolation of a thionin from wheat flour by Balls et al. (1942), several other thionins have been isolated from the endosperm of related cereals (see Figure 1). In recent years, thionins also have been identified in other organs of different Hordeum species, especially in leaves (Gausing, 1987; Bohlmann and Apel, 1987; Bunge, 1991; Bunge et al., 1992), although other organs of cereals also seem to contain thionins. Steinmiiller et al. (1986) showed that roots of Hordeum vulgare cv. Carina have very low levels of leaf-thionin-related transcripts. Bohl and Apel (submitted) recently isolated a cDNA from H. murinum for what seems to be a root-specific thionin (see Figure 1). Castagnaro et al. (1992) recently published a cDNA sequence for a neutral thionin from wheat endosperm whose sequence differs in many respects from that of the other endosperm thionins known thus far (see Figure 1). In addition to the cereal thionins, whose sequences have been compiled in Figure 1, endosperm thionins seem to be present in all species of the genera Hordeum, Triticum, and Aegilops (Carbonero and Garcia-Olmedo, 1969; Okadaet al., 1970; Jones et al., 1982; Bunge, 1991). Thionins also may be present in the endosperm of rye (Hernandez-Lucas et al., 1978) and maize (Jones and Cooper, 1980; Wada and Buchanan, 1981b), but their sequences have not been reported yet. In 1948, Winterfeld and Bijl reported that they had isolated from the leaves and stems of Viscum album a toxic mixture of small basic pro- P x G F P x G F PI( 0 F P I D F P I( D F P T G F P S 5 F L P A pm20 G P x P K PI( P X P I( P K P I - - - - - - - - - ~ o n c aand M k 115771, O h t a m e l a1 ,15751. Ohfan1 er al. 119771 Jones and M k 115711 m k and Jones 1 1 9 7 6 ! . Ohtan1 ct al. 119751. Ohtan> ef al. 119771 Bekes and La51tlZY 115811 Bckes and Laszrlty 119811 Ponr CL -1. 119861. Rodriguez-Palenruela eL a 1 i i 5 8 8 1 Hernandez-Lucas eL el. 119861 Cascagnaro e t a l . 1 1 5 5 2 1 m4 Dc4 PKG1348 Dc3 ph16 mm X K X X K K X X S S S S S S S S C C C C C C C C C C C C C C C C P P P P P P P P W Y W N T T S T T T T T T T T T T T T T T T T A C O C G A I A A R R R R R I R R N N N N N N N N I I I I I I I C Y Y Y Y Y I Y Y N T C R F N N N N N N N A T T T T T I C C C C C C C R l R R R R R L L F F F L L FIGURE 1. The amino acid sequences of thionins. Cysteine residues are boxed. The sequences of viscotoxin A2 and viscotoxin 1-PS have been corrected for the last 4 amino acid residues. (From Bohlmann and Apel, 1991.) 2 not known to be toxic (Van Etten et al., 1969). Crambin was first described by Van Etten et al. (1965) and is found in the seeds of Crambe abyssinica, family Brassicaceae. Taken together, it seems that the distribution of thionins in the plant kingdom is very sporadic in distantly related plant families. However, the majority of thionins described thus far has been found in plants that have agricultural or pharmaceutical importance: extracts from mistletoes are used for cancer treatment (Selawry et al., 1961), cereals are important crop plants, and Crambe species have some prospects as oil-producing plants. It therefore seems possible that, upon a thorough search, thionins might be found in other plant families as well. Indeed, Daley and Theriot (1987) have demonstrated the existence of proteins with characteristics similar to thionins in tomato (Lycopersicum esculentum, Fam. Solanaceae), mango (Mansifera indica, Fam. Anacardiaceae), papaya (Carica papaya, Fam. Caricaceae), and walnut (Juglans regia, Fam. Juglandaceae). These data and the recent discoveries of new thionin variants (Castagnaro et al., 1992; Bohl and Apel, submitted) indicate that thionins might be more ubiquitously distributed than previously thought. 111. STRUCTURE The published sequences of the different thionins are compiled in Figure 1. Many amino acid sequences have been determined by directly sequencing the isolated polypeptides, while in recent years the amino acid sequences for several thionins, especially those from Hordeum species, have been deduced from the correspondingcDNA or genomic clones. It is obvious that the six or eight cysteine residues are always conserved. Several other amino acids, especially those surrounding the cysteine residues, also are conserved (e.g., tyrosine or phenylalanine at position 13), whereas at other positions, the amino acids vary. Most of the thionins known thus far are highly basic, but there are two exceptions of neutral thionins: first, the crambins from the seeds of Crambe abyssinica (Van Etten et al., 1965), consisting of at least two different isoforms (Vermeulen et al., 1987); second, the wheat endo- sperm thionin deduced from the cDNA clone pTTH20, which was discovered very recently (Castagnaro et al., 1992). This putative thionin has only 37 amino acids due to a C-terminal deletion relative to other thionins, only 6 cysteines compared to the 8 cysteine residues of the other endosperm thionins, and little sequence similarity with the other thionins. Nevertheless, it is clearly a thionin, as shown by the alignment of the cysteine residues and several other amino acids and particularly by the high homologies between the signal peptide and the acidic domain of the precursor (see Section VII) and those of the other endosperm thionins (Figure 2). The amino acid composition of a putative thionin from corn endosperm has been determined (Wada and Buchanan, 1981a), revealing that this thionin might actually be acidic and would contain only four cysteine residues. In all cases in which the three-dimensional structure of thionins has been determined, the studies indicate a compact, L-shaped molecule (for references to these studies see Bohlmann and Apel [ 19911). The long arm of the L is formed by two a-helixes and the short arm by two short antiparallel P-sheets, while approximately the last 10 amino acids form a loop-like structure. In those cases investigated, the cysteines form three or four disulfide bridges giving the thionins a pronounced stability; viscotoxins, for example, can be heated at 100°C for 30 min without loss of their toxic properties (Samuelsson, 1974).A similar stability against heat denaturation also has been demonstrated for purothionin (Okada and Yoshizumi, 1970) and Pyrularia thionin (Vernon et al., 1985). Thionins have an amphipathic structure; hydrophobic residues are found primarily at the outer surface of the long arm of the L, whereas hydrophilic residues are found mainly at the inner surface of the L and at the outer surface of the comer of the L. IV. EFFECTS ON BIOLOGICAL SYSTEMS A main characteristic of most thionins is their toxic effect on different biological systems. This toxicity is the reason why these polypeptides were studied quite thoroughly in the past. The first 3 Posit ion : Precursor Precursor Precursor Precursor Precursor Precursor DB4 (Hordeum vulgare) mu16 (Hordeum murinum) HTHl (Hordeum vulgare) alpha1 (Triticum aestivum) pTTH20 (Triticum aestivum) Viscotoxin A3 (Viscum album) -10 -20 M M M M M M Leaf Leaf Endosperm Endosperm Endosperm Leaf -1 A A G G G E P - S K T - N K L K S K G G Q K V V - R --- S T G G G G I K S V I K S J V L K G V L E S A - S S L --- 11 V L L V L L L G A - L L V S Q V E S K ' S C C P N T T G R N I Y N A 21 31 "' B =I 61 T Q Y C I E Y C S L G C Y C T M G C -- 101 41 71 51 91 81 D D D D D N N Y Y N -- M M M M M A D D V V I T N N N N N N T S A A A - V V A A D - F F A A N N R R D D S G G G D D T D Q Q E E E A E Q E E E E M M M M M A K K K K K - F I L L L - D D Y Y Y - M M L V V V 0 G L C G L C E N C E N C K R C - R C 111 (Bohlmann and Apel 1987) (Bunge 1991) (Rodriguez-Palenzuela et al. 1988) (Castagnaro et al. 1992) (Castagnaro et al. 1992) (Schrader and Apel 1991) FIGURE 2. The amino acid sequences of thionin preproproteins. Cysteine residues are boxed and the putative processing sites are indicated. report that wheat flour contained a substance that was toxic to yeast seems to be that of Jag0 and Jag0 in 1885 (cited in Ohtani et al. [1977]), a report that has since been confirmed repeatedly (cited in Okada et al. [1970]). Soon after the discovery of purothionins as the active agent in wheat flour that is toxic to brewer's yeast (Balls et al., 1942), it was found that purothionins were bactericidal and fungicidal (Stuart and Harris, 1942) and toxic to mammals if injected intraperitoneally or intravenously (Coulson et al., 1942), whereas oral administration of up to 229 mgjkg body weight had no effect on guinea pigs. Similar toxic effects have since been reported for other 4 endosperm thionins and for viscotoxins. Viscotoxins also are toxic to mammals when administered intraperitoneally (LD50 in mice: 0.5 m a g body weight) or intravenously (LD,, in cats: 0.1 mgjkg body weight) but are not toxic when eaten (Samuelsson, 1974). The LD50 of Pyruluriu thionin for mice (administered intraperitoneally) is 1.5 mg/kg body weight (Evett et al., 1986). Several authors have investigated the effects of viscotoxins on the blood circulation of mammals. For example, Rose11 and Samuelsson (1966) showed that low doses of viscotoxin A3 or phoratoxin, the latter being less potent, produced reflex bradycardia and had a negative inotropic effect on the heart, whereas high doses led to vasoconstriction of vessels in skin and skeletal muscles. Kramer et al. (1979) reported the toxic effects of several endosperm thionins on insects. Toxicity against yeast, in addition to the reports cited earlier, has been described by several authors (e.g., Okada et al. [1970], Okada and Yoshizumi [1973], Hernandez-Lucaset al. [1974], Wada et al. [1982]) and toxicity against bacteria has been reported by Stuart and Harris (1942), Fernandez de Caleya et al. (1972), and Cammue et al. (1992). The latter authors also found p-purothionin to display antifungal activity to about a dozen phytopathogenic fungi with IC,, values of approximately 1 to 5 pdrnl. In addition to the toxicity of thionins on whole organisms, several authors have reported cytotoxic effects for several thionins, as well as several inhibitory effects in in vitro systems. Nakanishi et al. (1979) found that purothionin had the most pronounced cytotoxic effect on mammalian cells during the S-phase. Kashimoto et al. (1979) demonstrated leakage of cytoplasmic enzymes and ions after perfusion of bovine adrenal glands with 4 pg/ml purothionin. The same symptoms also have been demonstrated for brewer’s yeast (Okada and Yoshizumi, 1973). Carrasco et al. (1981) also found an increased permeability of several mammalian cell cultures after application of several types of endosperm thionins and viscotoxins. Cytotoxicity on mammalian cells was further reported for p-purothionin (Cammue et al., 1992) and for Pyruluriu thionin (Vernon et al., 1985; Evett et al., 1986; Shaw et al., 1987). Hemolytic activity of thionins has been demonstrated by Lankisch and Vogt (1971) for viscotoxin B and by Osorio e Castro et al. (1989) for Pyrulariu thionin. Inhibitory effects on in vitro systems have been reported by Garcia-Olmedo et al. (1983), who showed that purothionin inhibits cell-free protein synthesis, and by Diaz et al. (1992), who found an irreversible inactivation of Escherichiu coli P-glucuronidase by purothionins. The only thionin known thus far to have no toxic effects is the neutral seed protein crambin (Van Etten et al., 1969; Teeter et al., 1981; Schrader, 1988). Nothing is known yet about the toxic properties of the second neutral thionin, the newly discovered wheat endosperm thionin en- coded by the cDNA clone pTTH20 (Castagnaro et al., 1992). The common denominator for these toxic effects seems to be a destruction of membranes, and two mechanisms have been proposed as possible explanations. The amphipathic structure of the thionins indicates that the toxicity might be exerted by a direct, detergent-like interaction with the lipid bilayers of biological membranes. The hydrophobic face of the thionins could interact with the hydrophobic aliphatic chains of the membrane lipids, whereas the positively charged basic amino acids could interact with the negatively charged phosphate groups of the phospholipids. These intermolecular salt bridges cannot be formed by crambin, which has only two basic residues (Arg 10 and Arg 17), both of which are blocked through intramolecular bonds (Whitlow and Teeter, 1985). Work on Pyruluria thionin led Vernon and co-workers to propose a different model, which includes the specific binding to a membrane receptor, after showing that Pyruluriu thionin has specific binding sites on the surface of erythrocytes (Osorio e Castro et al., 1989, 1990; Osorio e Castro and Vernon, 1989; Vernon and Rogers, 1992b). First proposed to be a membrane protein (Osorio e Castro et al., 1989), the binding site is now thought to be a specific phospholipid (Vernon and Rogers, 1992a). Interestingly, the binding site for Pyruluria thionin seems to be the same as that of the cardiotoxins, small (about 60 amino acids), basic proteins with 8 cysteine residues, from the venoms of cobras and related snakes (Harvey, 1985,1991). Iodinated Pyruluriu thionin, which is no longer toxic (Evans et al., 1989), competitively inhibits hemolysis by both native Pyruluriu thionin and cardiotoxin, whereas the hemolytic activity of melittin from bee venom is not inhibited (Osorio e Castro and Vernon, 1989). In addition, both toxins are inhibited by extracellular calcium ions and are stimulated by external phosphate ions (Vernon and Rogers, 1992a). That higher concentrations (5 mM) of calcium ions and other divalent metal ions abolish the toxic activity of purothionins has been known for quite some time (e.g., see Okada et al. [1970]). Binding of Pyruluriu thionin to the membrane leads to the stimulation of an internal phospholipase, probably PLA2 (Evans et al., 1989; Angerhofer et al., 5 1990). Pyrularia thionin itself has no phospholipase activity (Osorio e Castro et al., 1989; Angerhofer et al., 1990). This also holds for cardiotoxins (e.g., see Jiang et al. [1989]), but snake venom also contains phospholipases, which might act synergistically with the cardiotoxins (Louw and Visser, 1978). Such a synergistic effect also has been demonstrated for viscotoxin B and phospholipase A (Lankisch and Vogt, 1971). Another effect mediated by Pyrularia thionin was the increase in cytosolic calcium ions in mouse P388 cells (Evans et al., 1989). Taken together, there is good evidence that the mechanism by which Pyrularia thionin exerts its toxicity starts with the binding to a membrane receptor, probably a specific phospholipid, and that cardiotoxins use a similar mechanism involving the same receptor. The tyrosine residue in position 13 seems to be crucial not only for the toxicity of the Pyrularia thionin based on the iodination studies, but also for the other toxic thionins inasmuch as this residue and the region around it (position 10-14) are conserved in all toxic thionins (see Figure 1). The importance of the tyrosine residue in purothionin has been demonstrated - also using iodination - by Wada et al. (1982), supporting the model of Vernon and co-workers of specific binding sites on the membrane for Pyrularia thionin and perhaps for other thionins as well. In a simple detergent-like interaction, the crucial role for this residue would be hard to imagine. On the other hand, all toxic thionins have a conserved basic nature and amphipathic structure, which suggests that the second step after the specific binding might still be a detergent-like interaction leading to destruction of the membranes and all of the well-known effects of the toxic thionins. Oka et al. (1992) recently proposed that thionins have some homology with the epidermal growth factor (EGF) domain, found, for example, it^ the mammalian perforins. This homology is vely weak because the alignment of the cysteine residues requires the introduction of many gaps in the sequences and does not include the first two cysteine residues of the thionins, which have been shown to form disulfide bridges with the last two cysteine residues. Also, the EGF-like domains do not have the conserved tyrosine residue of the toxic thionins. 6 V. BIOLOGICAL FUNCTION Several biological functions have been proposed for thionins, based mostly on in vitro observations. The finding that purothionin can be reduced in vitro by the thioredoxin system of wheat seeds (Wada and Buchanan, 1981a; Johnson et al., 1987) led the authors to propose that purothionin could function as a secondary thiol messenger, with the sulfhydryl form of purothionin reducing and thereby activating fructose- 1,6biphosphatase. In a similar reaction, purothionin reduced by thioredoxin was shown to block DNA synthesis in vitro by inhibiting the enzyme ribonucleotide reductase. It has been proposed that the inhibition of DNA synthesis may be the reason for the toxic effect that thionins have on mammalian cells undergoing chromosome duplication (Nakanishi et al., 1979).On the other hand, it was shown that the interaction of viscotoxins with DNA in vitro leads to a protection of the DNA double helix against thermal denaturation (Woynarowski and Konopa, 1980). Whether these in vitro observations have any significance for the in vivo situation, in which the subcellular compartmentalization of the cell hinders the interaction of compounds that can be brought together easily in vitro, is not known. The seed-specificmembers of this gene family could have a function as storage proteins, particularly for sulfur. Evidence for a function as sulfurstorage proteins has been obtained for viscotoxins (Schrader and Apel, 1993. Plant Physiol. 101:745749). These authors showed that viscotoxins are produced in high amounts in young leaves, which appear in May. They are detectable in these leaves until the next summer. Viscotoxins are no longer detectable in senescent leaves, which are lost in AugusVSeptember, indicating that the plant is reutilizing the sulfur of the viscotoxins. Crambin, which is neutral and has no known toxic effects, also may function as a storage protein. On the other hand, its nitrogen content is significantly lower than that of the toxic seed-specificthionins and this argues against a role as a general storage protein. However, one should keep in mind that seed proteins can have dual functions as storage proteins and as, among other functions, defense proteins (Bohlmann, 1992).The demonstrationthat the wellknown 2s seed-storage albumins of radish (Raphanus sativus L.) seeds have antifungal activity supports this view (Terras et al., 1992). There is one short report about purothionin having a-amylase-inhibiting activity (Jones and Meredith, 1982), but the inhibition was only partial. Nevertheless, because many enzyme inhibitors have been found in seeds and are - like thionins - usually cysteine-rich (the cysteine residues forming disulfide bridges), a thorough investigation of the potential of thionins as enzyme inhibitors might be interesting. Another possible defense function was first described by Fernandez de Caleya et al. (1972). Their finding that purothionins are toxic to phytopathogenic bacteria in vitro led them to suggest that thionins may function as defense proteins against phytopathogens. Recently, this has been supported by work on the leaf-specific thionins of barley (Bohlmannand Apel, 1987),described next, which has produced experimental evidence that these proteins might indeed function as defense proteins against microorganisms. VI. LEAF THlONlNS OF BARLEY Leaf thionins of barley can be isolated from cell walls (Bohlmann et al., 1988) and vacuoles (Reimann-Philipp et al., 1989b) of barley leaves. This distribution is reminiscent of that of two PRproteins, glucanases and chitinases (Linthorst, 1991). Although in the case of the PR-proteins, different isofoms are present in the vacuole and the extracellular space, this does not seem to be the case with the leaf-specific thionins of barley (Bohlmann et al., 1988; Reimann-Philipp et al., 1989b); sequencing of thionins isolated from cell walls and vacuoles indicates similar mixtures of leaf thionins in both compartments. As shown by immunocytochemistry, all cell walls of barley leaves have thionins incorporated, but the outer cell walls of epidermal cells have the highest concentration (Reimann-Philipp et al., 1989a). Leaf thionins isolated from cell walls and vacuoles of barley leaves were tested in a plate diffusion assay against two phytopathogenic fungi, Thielaviopsis paradoxa, a pathogen of sugarcane, and Pyrenophora (Drechslera) teres, a pathogen of barley. The growth of both fungi was inhibited by the thionin preparations (Bohlmann et al., 1988). In vitro toxicity against phytopathogenic microorganisms also has been demonstrated for other thionins. Fernandez de Caleya et al. (1972), as previously mentioned, found the growth of several phytopathogenic bacteria to be inhibited by purothionins, and Cammue et al. (1992) showed that P-purothionin is toxic against several phytopathogenic fungi and some phytopathogenic bacteria. Because the cellular distribution of leaf thionins and their toxicity against fungi suggested that they were defense proteins, it was expected that leaf thionins would be induced by pathogen attack, as has been shown for the PR-proteins, whereas the seed-specific thionins are most probably preinfection defense proteins. Indeed, an induction of leaf thionins was demonstrated at the mRNA and the protein levels after infection with barley mildew (Erysiphe graminis f.sp. hordei), perhaps the most important pathogen of barley. Following infection, a transient rise in the mRNA level with a peak after about 2 days was detected in both compatible and incompatible interactions (Bohlmann et al., 1988), with the mRNA accumulation being somewhat higher in the incompatible interaction. A difference between compatible and incompatible interactions was clearly demonstrated at the protein level with immunogold labeling using an antibody obtained against a fusion protein of P-galactosidase and the leaf-specific thionin precursor DB4 (Bohlmann and Apel, 1987). Mildew-infectedbarley epidermal cells try to counteract the invading infection hyphae of the fungus by producing cell wall appositions called papillae at the point of invasion (Aist, 1983). After inoculation with the mildew race C17, no thionins could be detected in the papillae of the susceptible cultivar “Peruvian” and only reduced levels were present in the cell wall surrounding the infection site compared to cell walls of noninfected cells. On the other hand, in the resistant cultivar “Stamm 41 ,” the cell wall surrounding the infection site showed normal levels of thionins, and thionins also were detectable in the papillae. The leaves of the adult-plant resistant cultivar “Osiris” showed a differential response, depending on the age of the leaf. Although the primary leaves are susceptible, showing the same very low level of thionins in the cell wall as in the susceptible cultivar “Peruvian,” thionins are de- 7 tectable in the papillae of the fifth leaves, which are resistant to this race of mildew (EbrahimNesbat et al., 1989). These results indicate that thionins are newly synthesized after infection by mildew and incorporated into the papillae. On the other hand, in compatible interactions, the fungus seems to have a mechanism to protect itself against the toxic thionins by somehow masking or destroying the proteins so that they are no longer recognized by the antibody. Such a mechanism also can be postulated for P . teres, which is inhibited by thionins in vitro, but is a pathogen of barley in vivo. In barley, leaf thionins are encoded by a large gene family containing as many as 50 genes that are differentially regulated (see later discussion). While in H . vulgare C.V. Carina the known DNA clones can be grouped into two subfamilies with microheterogeneity at the DNA and protein levels in each subfamily (Bohlmann and Apel, 1987), the cDNA clones from H . murinum are much more divergent and cannot be grouped into major subfamilies (Bunge et al., 1992). In the latter species, it also was demonstrated that the thionin domain shows a higher variability than the signal sequence or the acidic domain (Bunge et al., 1992). The reason for this could be the selective pressure of genetically variable pathogens, which try to overcome the toxic action of these thionins. A similar polymorphism of polypeptide toxins, this time as a consequence of the interplay between predator and prey, is for example seen in the case of the cone shell neurotoxins (Olivera et al., 1990). Castagnaro et al. (1992) recently described a cDNA for a novel wheat endosperm thionin. The precursor deduced from this cDNA shows more divergence from the other known purothionin and hordothionin precursors in the thionin region than in the signal peptide or the acidic domain (compare Figure 2), which agrees with the findings of Bunge et al. (1992) for H . murinum. Thus, it seems that leaf thionins are on one hand preinfectional compounds stored in the vacuole and the cell walls, but on the other hand they also can be induced by pathogen attack and several other external stimuli. Etiolated barley seedlings have a very high level of leaf thionin transcripts, which drops drastically after illumination, probably mediated by two photoreceptors, phytochrome and a blue-light-absorbing photoreceptor 8 (Reimann-Philipp et al., 1989a). Blue light leads to a much stronger decline of thionin mRNAs compared to the red-light effect, but only at high light intensities. Such high light intensities are not reached under normal light conditions in the meristematic zone at the leaf basis, which is covered by the sheath of the preceding leaves. Thionin synthesis is therefore possible in these young developing cells, and since these proteins are very stable, their concentration declines very slowly once these cells are no longer shaded by the leaf sheath but exposed to full light, which leads to a drastic reduction in thionin mRNA. Heavy metals increase the level of leaf thionin transcripts, whereas salicylic acid, which is discussed as a second messenger for plant defense responses (Malamy and Klessig, 1992), seems to have no effect (Fischer et al., 1989). On the other hand, methyl-jasmonate has a drastic influence and leads to an accumulation of leaf thionin transcripts and the mature proteins (Andresen et al., 1992). The accumulation in cut leaf segments occurs shortly after the start of the treatment with jasmonic acid, whereas in whole seedlings exposed to volatile methyl-jasmonate, the accumulation is much slower and requires higher concentrations. The reason may be that the uptake of volatile methyljasmonate is impeded in whole, uninjured barley leaves. In any case, the responsiveness to jasmonate or methyl-jasmonate is again indicative of a defense function of barley leaf thionins. Farmer and Ryan have shown that accumulation of proteinase inhibitors of tomato also can be induced by methyl-jasmonate (Farmer and Ryan, 1990) or its precursors (Farmer and Ryan, 1992). In addition, Gundlach et al. (1992) showed that jasmonic acid and its methyl ester accumulate transiently in plant cell cultures after treatment with elicitors. VII. BIOSYNTHESIS AND PROCESSING OF THlONlN PRECURSORS It has been shown that barley leaf thionins have a toxic effect on plant cells. The regeneration of tobacco protoplasts cultivated 2 weeks under dim light was inhibited if isolated barley leaf thionins were added to the medium (ReimannPhilipp et al., 1989b). Similarly, regeneration of barley protoplasts also is inhibited by barley leaf thionins (Dae-Won Lee et al., manuscript in preparation). These results emphasize that plants producing these toxic thionins must have a protective mechanism against their own toxins, a common principle among toxin producers. In the case of thionins, it has been demonstrated by in vitro translation and immunoprecipitation that hordothionins (Ponz et al., 1983), leaf thionins of barley (Bohlmann and Apel, 1987), and viscotoxins (Schrader and Apel, 1991)are synthesized as much larger precursors with a molecular weight of about 15,000 Da. The sequences of several precursors have been deduced from cDNA clones and are shown in Figure 2. As can be seen, all precursors known thus far have the same general structure of a preproprotein. N-terminal to the thionin domain is a typical signal sequence, which directs the proprotein into the endoplasmic reticulum. The proprotein consists of the actual thionin and a C-terminal extension, the so-called acidic domain, which contains a large number of acidic amino acids. Although the amino acids at most positions are variable (see Figure 2), some amino acids are highly conserved, including a tyrosine at position 61 and a glycine at position 65. The six cysteine residues are absolutely conserved, even in the two known viscotoxin precursors in which the acidic domains have several deletions (Schrader and Apel, 1991). Whether these cysteines form disulfide bridges is not known, and a protein corresponding to the acidic domain has never been isolated as such. Nevertheless, the homology of these acidic domains from diverse thionin precursors indicates an important role for these domains. One might postulate that they may neutralize the basic thionin domain and protect the cell against the toxic action of the thionins. In addition, the acidic domain also may contain information to guide the thionins through the secretory pathway to their final destinations. There is, however no experimental information available at the moment about possible functions of these acidic domains. Also, this does not explain how plants can tolerate the large amounts of thionins that can accumulate in the central vacuole as has been shown for barley (Reimann-Philipp et al., 1989b). It thus seems that beyond inactivation of the thionin during synthesis and transport by the action of the acidic domain, there must be an additional mechanism to protect the plant against the mature thionin in its final cellular compartment. Inasmuch as several authors have reported that thionins, especially endosperm thionins, can be isolated as lipid-protein complexes with petroleum ether (e.g., see Balls et al. [1942], Redman and Fisher [ 19681,Hernandez-Lucas et al. [ 19771, and Carbonero et al. [ 1980]), one might speculate that lipids could play such a role. However, these lipid-thionin complexes also could be an artefact due to the isolation procedure. Interestingly, the chitin-binding protein hevein (43 amino acids, 8 cysteine residues) from the laticifers of rubber trees also is produced as a preproprotein with a C-terminal extension of 144 amino acids that also contains 6 cysteine residues (Broekaert et al., 1990). The preproproteins of the defensins (small, basic, cysteine-rich polypeptides with antimicrobial activities found in mammals and insects) on the other hand have a different precursor structure. Here, the prodomain is located N-terminally to the mature defensin, between the signal sequence and the mature defensin (Daher et al., 1988; Dimarcq et al., 1990). As in the case of the thionins, however, the prodomain also is acidic and neutralizes the basic, toxic defensin. VIII. y -THIONINS Recently, Mendez and his group isolated several new proteins from barley and wheat endosperm (Colilla et al., 1990; Mendez et al., 1990). Like the hordothionins and purothionins, these proteins are small, basic, have eight cysteine residues, and inhibit protein synthesis in vitro. These features may have prompted the authors to classify these proteins as thionins and to name them y-thionins. However, an alignment of these proteins with the classical thionins is possible only for five of the eight cysteine residues, and to do so, several gaps have to be introduced in the amino acid sequences. The comparison of the amino acid sequences clearly shows that the y-thionins do not belong to the classical thionin protein family but to a different protein family. This also is supported by inspection of the sequences of other members of this gene family that have been described since the first reports by the 9 Mendez group. Figure 3 shows an alignment of the sequences for all y-thionins. Sequences for y-hordothionins, y-purothionins, and sorghuminhibitors were obtained directly from the isolated proteins, whereas the other sequences were derived from cDNA clones. All cDNAs encode a typical signal sequence in addition to the y-thionin sequences, indicating that all of these proteins are secreted. Interestingly, the precursor for the y-thionin from tobacco flowers (Gu et al., 1992) has an acidic extension, but this acidic domain has no cysteine residues and no apparent homology with the acidic domain of the thionin precursors, again indicating that these y-thionins belong to a different gene family. Not included in Figure 3 are the partial sequences of two R . sativus antifungal proteins (Terras et al., 1992), which probably belong to the y-thionin protein family. The in vivo function of the y-thionins is not yet clear. The y-hordothionins have been shown to inhibit protein synthesis in vitro (Mendez et al., 1990) and the sorghum-inhibitors inhibit a-amylases from insects (Bloch and Richardson, 1991), whereas the R . sativus antifungal proteins are fungistatic, suggesting that they are defense proteins. Elevation of the mRNA levels for the pea y-thionins after fungal attack also indicates that these proteins may play a role in plant defense. Considering the homology between the amino acid sequences of the y-thionins, it is quite possible that they all have a similar three-dimensional structure; it will be interesting to see if they function via a common mechanism in their suspected role as defense proteins and if this mechanism is different from that of the classical thionins. In addition to classical thionins and y-thionins, there are several other groups of small, basic, and cysteine-rich antimicrobial polypeptides or proteins. The cardiotoxins have already been mentioned. Insects and other animals contain cecropins (Boman and Hultmark, 1989) and mammals and insects produce defensins in certain cell types (Lehrer et al., 1991; Lambert et al., 1989). Recently, Cammue et al. (1992) isolated 2 basic polypeptides (36/37 amino acids) with 4 disulfide bridges from the seeds of Mirabilis jalapa L. (called Mj-AMP1 and Mj-AMP2). Both have strong antimicrobial activities in a concentration range similar to purothionin, but they were not 10 toxic to the two tested mammalian cell lines. Another example is the antifungal protein, isolated from the fungus Aspergillus giganteus (Nakaya et al., 1990), which shows some sequence similarity with phospholipases A2. IX. THE USE OF THlONlNS TO INCREASE THE RESISTANCE OF CROP PLANTS During evolution, parasites and pathogens developed diverse mechanisms to break the resistance of certain plants and became specialized to live on those host plants. This is especially evident for insects, many of whom feed on only one or a few host plants, but also holds true for bacterial and fungal phytopathogens. Many fungi, for example, have adapted to their host plants by developing mechanisms to inactivate the phytoalexins produced by their specific host (Van Etten et al., 1989), and it seems reasonable that pathogens also have evolved mechanisms to cope with the plant’s defense proteins. With the advent of gene technology, it now seems possible to increase the resistance of crop plants by introducing genes that would lead to the production of new defense compounds to which the specific pathogens are susceptible. This would be much more difficult for low molecular weight “secondary” metabolites because their production would often require the introduction of several genes coding for different enzymes of the biosynthetic pathway leading to that product. On the other hand, the engineering of plants to produce new defense proteins with antimicrobial activities or with enzyme inhibitory functions should be easier, requiring only the introduction and expression of the gene coding for that specific defense protein. Theoretically, this seems to be straightforward and several groups have indeed expressed proteins directed against insects and proteins with antimicrobial activities in transgenic plants. Johnson et al. (1989) expressed proteinase inhibitors from tomato in transgenic tobacco plants and found the plants to be at least partially protected against Manduca sexta larvae. Several groups have used the expression of the insecticidal B.t. (Bacillus thuringiensis) toxin to protect plants from insect attack (e.g., see Vaeck et al. [1987]). Pro- -1 -11 -21 Posit ion : SI1 SI2 S13 Gamma-1P Gam-2P G a m -H FST p322 pSASl0 pI39 pI230 1 M A R S L O F M A F A I L A M M L F V A Y E V Q A M R F F A T F F L L A M L V V A T K M G P M R I A E A M E K K S I A G L O F L - F L V L F V A Q E V V V Q S E A M E K K S L A A L S F L L L L V L F V A Q E I V V T - E A M E K K S L A B L S F L - L L V L F V A Q E I V V S - E A 11 n 21 31 n n n R V C M G K S Q H H S F P C I S D R L C S N E C V K E E G G W T A G R R V C M G K S A G F K G L C M R D Q N C A Q V C L Q E - - G W G G G N R V C R R R S A G F K G L C M S D H N C A Q V C L Q E - - G W G G G N 51 41 61 1 l:I C D G P F R R C K C I R Q C ICI K lC/ ICI D G P L R R IT K P k - V F D E K M I K T G A E T L V 1 ' 1 pi f 1:1 1C1 R D D V R W C H D - W K - C F C T Q N C 71 81 Bloch and Richardson (1991) Bloch and Richardson (1991) Bloch and Richardson (1991) Colilla et al. (1990) Colilla et al. (1990) Mendez et al. (1990) E E A K T L A A A L L E E E I M D N Gu et al. (1992) Stiekerna et al. (1988) Ishibashi et al. (1990) Chiang and Hadwiger (1991) Chiang and Hadwiger (1991) FIGURE 3. The amino acid sequences of y-thionins. The sequences of sorghum inhibitors (SI), y-purothionins (gamma-P), and gamma-hordothionins (gamma-H) were derived from the isolated proteins. Sequences for FST (tobacco, Nicotiana tabacurn),p322 (potato, Solanurn tuberosurn),pSAS10 (cowpea, Vigna unguiculata), and pl39/pl230 (pea, Pisurn sativurn) are deduced from cDNA clones. teins with antimicrobial activities expressed in transgenic plants include, for instance, chitinases and ribosome inactivating proteins. Broglie et al. (1991) found that transgenic tobacco and canola plants expressing a bean chitinase have an increased resistance against the fungal pathogen Rhizoctonia soluni. Logemann et al. (1992) demonstrated that transgenic tobacco plants expressing a barley ribosome inactivating protein also show an increased resistance against R. soluni. Other prime candidates for such purposes would be antimicrobial proteins such as thionins, 11 y-thionins, defensins, and cecropins. Jaynes et al. are attempting to express cecropins in crop plants (Jaynes et al., 1987), and several groups are using thionin genes in a similar fashion. Preliminary results regarding the expression of thionins in tobacco using the CaMV promoter have been reported. While Florack et al. (1991) claim expression levels of up to 3% total protein for a synthetic hordothionin sequence, Mollenhauer and Apel (personal communication) have obtained much lower protein levels with a barley leaf thionin cDNA. Results cited by Carmona et al. (1993. Plant J . 3:457-462) claim that tobacco plants expressing hordothionins are less sensitive to bacterial pathogens than control plants. Nevertheless, it seems that this approach is not as straightforward as previously thought. Until now, only y-thionins and not classical thionins have been found in tobacco and it is not clear how tobacco cells, which are sensitive to thionins (see earlier discussion), can cope with these proteins. It also may be possible that the thionins are correctly processed but proteolytically cleaved in the vacuole. Phytotoxicity also seems to be a problem for expression of cecropins, at least in some plant species (Nordeen et al., 1992). In insects, cecropins are produced as preproproteins, and it may be necessary to use a similar strategy for the expression of these proteins in plants. It is not known, however, if the plants can process cecropin precursors correctly, whereas the thionin precursors from barley were correctly processed in tobacco (Florack et al., 1991; Mollenhauer and Apel, personal communication). In the case of the thionins, it also would be useful to carefully look for endogenous thionins in those plants in which exogenous thionins are to be expressed. Overexpressionof such endogenous thionins also may lead to increased resistance to pathogens. Unfortunately, this cannot be done with those plants in which thionins have been found to date because these species cannot be transformed. Taken together, thionins show promise to increase the disease resistance in transgenic plants due to their antibacterial and antifungal activities, but more work is needed to achieve an optimal expression of these proteins. 12 X. CONCLUSIONS During the last few years, evidence has accumulated that thionins are one part of the multifaceted resistance response of plants, but many unanswered questions remain. Unfortunately, the barley system, which has provided much of that evidence, is not amenable to genetic transformation and other modem genetic methods. The large number of thionin genes present in barley also hinders a detailed analysis of their possible functions. The mode of action of thionins, the processing of the precursors, and the manner in which plants protect themselves against their own toxic thionins are unclear, but such information may be very important for an optimized expression of thionins in transgenic plants to increase their resistance. ACKNOWLEDGMENTS I thank Drs. Klaus Apel and Gregory Armstrong for a critical review of the manuscript and Dr. Dieter Rubli for help with the figures. REFERENCES Aist, J. R. 1983. Structural responses as resistance mechanisms. In: The Dynamics of Host Defence. pp. 33-70 (Bailey, J. A,, Ed.). New York: Academic Press. Andresen, I., Becker, W., Schliiter, K., Burges, J., Parthier, B., and Apel, K. 1992. The identification of leaf thionin as one of the main jasmonate-induced proteins of barley (Hordeum vulgare). Plant Mol. Biol. 19: 193204. Angerhofer, C. K., Shier, W. T., and Vernon, L. P. 1990. Phospholipase activation in the cytotoxic mechanism of thionin purified from nuts of Pyrularia pubera. Toxicon. 28547-557. Balls, A. K., Hale, W. S., and Harris, T. H. 1942. A crystalline protein obtained from a lipoproteinof wheat flour. Cereal Chem. 19:279-288. Bekes, F. and Lasztity, R. 1981. Isolation and determination of amino acid sequence of avenothionin, a new purothionin analogue from oat. Cereal Chem. 58:360361. Bloch, C., Jr. and Richardson, M. 1991. A new family of small (5 kDa) protein inhibitors of insect a-amylases from seeds of sorghum (Sorghum bicolor (L) Moench) have sequence homologies with wheat y-purothionins. FEBS Lett. 279:lOl-104. Bohlmann, H. 1992. Significance of sulfur-rich proteins in seeds and leaves. Manuscript submitted. Bohlmann, H. and Apel, K. 1987. Isolation and characterization of cDNAs coding for leaf-specificthionins closely related to the endosperm-specific hordothionin of barley (Hordeum vulgare L.). Mol. Gen. Genet. 207:44&454. Bohlmann, H. and Apel, K. 1991. Thionins. Annu. Rev. Plant Physiol. Plant Mol. Biol. 42:227-240. Bohlmann, H., Clausen, S., Behnke, S., Giese, H., Hiller, C., Reimann-Philipp, U., Schrader, G., Barkholt, V., and Apel, K. 1988. Leaf-specific thionins of barley - a novel class of cell wall proteins toxic to plant-pathogenic fungi and possibly involved in the defense mechanism of plants. EMBO J . 7:1559-1565. Boman, H. G. and Hultmark, D. 1987. Cell-free immunity in insects. Annu. Rev. Microbiol. 41: 103-126. Broekaert, W. F., Lee, H.-I., Kush, A., Chua, N.-H., and Raikhel, N. 1990. Wound-induced accumulation of mRNA containing a hevein sequence in laticifers of rubber tree (Hevea brasiliensis). Proc. Natl. Acad. Sci. U S A . 87:7633-7637. Broglie, K., Chet, I., Holliday, M., Cressman, R., Biddle, P., Knowlton, S., Mauvais, C. J., and Broglie, R. 1991. Transgenic plants with enhanced resistance to the fungal pathogen Rhizoctonia solani. Science. 254: 1194-1 197. Bunge, S. 1991. Vergleichende Untersuchungen an BlattThionin-Genen in der Gattung Hordeum. Ph.D. thesis, University of Kiel. Bunge, S., Wolters, J., and Apel, K. 1992. A comparison of leaf thionin sequences of barley cultivars and wild barley species. Mol. Gen. Genet. 231:460468. Cammue, B. P. A., De Bolle, M. F. C., Terras, F. R. G., Proost, P., Van Damme, J., Rees, S. B., Vanderleyden, J., and Broekaert, W. F. 1992. Isolation and characterization of a novel class of plant antimicrobial peptides from Mirabilis jalapa L. seeds. J. Biol. Chem. 267: 2228-223 3. Carbonero, P. and Garcia-Olmedo, F. 1969. Purothionins in Aegilops-Triticum spp. Experientia. 25: 1110-1 111. Carbonero, P., Garcia-Olmedo, F., and Hernandez-Lucas, C. 1980. External association of hordothionin with protein bodies in mature barley. J. Agric. Food Chem. 28: 399402. Carrasco, L., Vazques, D., Hernandez-Lucas, C., Carbonero, P., and Garcia-Olmedo, F. 1981. Thionins: plant peptides that modify membrane permeability in cultured mammalian cells. Eur. J. Biochem. 116:185-189. Castagnaro, A., Marana, C., Carbonero, P., and GarciaOlmedo, F. 1992. Extreme divergence of a novel wheat thionin generated by a mutational burst specifically affecting the mature protein domain of the precursor. J. Mol. Biol. 224:1003-1009. Chiang, C. C. and Hadwiger, L. A. 1991. The Fusarium solani-induced expression of a pea gene family encoding high cysteine content proteins. Mol. PlantMicrobe Interact. 4324-33 1. Colilla, F. J., Rocher, A., and Mendez, E. 1990. yPurothionins: amino acid sequence of two polypeptides of a new family of thionins from wheat endosperm. FEBS Lett. 270:191-194. Coulson, E. J., Harris, T. H., and Axelrod, B. 1942. Effect on small laboratory animals of the injection of the crystalline hydrochloride of a sulfur protein from wheat flour. Cereal Chem. 19:301-307. Daher, K. A., Lehrer, R. I., Ganz, T., and Kronenberg, M. 1988. Isolation and characterizationof human defensin cDNA clones. Proc. Natl. Acad. Sci. U.S.A.8573277331. Daley, L. S. and Theriot, L. J. 1987. Electrophoretic analysis, redox activity, and other characteristics of proteins similar to purothionins from tomato (Lycopersicum esculenta), mango (Mangifera indica), papaya (Carica papaya), and walnut (Juglans regia). J . Agric. Food Chem. 35:680687. Diaz, I., Carmona, M. J., and Garcia-Olmedo, F. 1992. Effects of thionins on P-glucuronidase in vitro and in plant protoplasts. FEBS Lett. 296:279-282. Dimarcq, J.-L., Zachary, D., Hoffmann, J. A., Hoffmann, D., and Reichhart, J.-M. 1990. Insect immunity: expression of the two major inducible antibacterial peptides, defensin and diptericin, in Phormia terranovae. EMBO J. 9~2507-2515. Ebrahim-Nesbat, F., Behnke, S., Kleinhofs, A., and Apel, K. 1989. Cultivar-related differences in the distribution of cell-wall-bound thionins in compatible and incompatible interactions between barley and powdery mildew. Planta. 179:203-210. Evans, J., Wang, Y., Shaw, K.-P., and Vernon, L. P. 1989. Cellular responses to Pyrularia thionin are mediated by Ca2+influx and phospholipase A, activation and are inhibited by thionin tyrosine iodination.Proc. Natl. Acad. Sci. U.S.A. 865849-5853. Evett, G. E., Donaldson, D. M., and Vernon, L. P. 1986. Biological properties of Pyrularia thionin prepared from nuts of Pyrularia pubera. Toxicon. 24622425. Farmer, E. E. and Ryan. C. A. 1990. Interplant communication: airborne methyl jasmonate induces synthesis of proteinase inhibitors in plant leaves. Proc. Natl. Acad. Sci. U S A . 87:7713-7716. Farmer, E. E. and Ryan, C. A. 1992. Octadecanoid precursors of jasmonic acid activate the synthesis of woundinducible proteinase inhibitors. Plant Cell. 4: 129-1 34. Fernandez de Caleya, R., Gonzales-Pascual, B., GarciaOlmedo, F., and Carbonero, P. 1972. Susceptibility of phytopathogenic bacteria to wheat purothionins in vitro. Appl. Microbiol. 23:998-1000. Fischer, R., Behnke, S., and Apel, K. 1989. The effect of chemical stress on the polypeptide composition of the intercellular fluid of barley leaves. Planta. 178:6168. Florack, D., Visser, B., and Stiekema, W. 1991. Synthetic hordothionin genes for bacterial disease resistance breeding in Solanaceae. Abstr. 3rd Int. Congr. Plant Molecular Biology. Hallick, R. B., Ed., Tucson. 13 Garcia-Olmedo, F., Carbonero, P., Hernandez-Lucas, C., Paz-Ares, J., Ponz, F., Vicente, O., and Sierra, J. M. 1983. Inhibition of eukaryotic cell-free protein synthesis by thionins from wheat endosperm. Biochim. Biophys. Acta. 740:52-56. Garcia-Olmedo, F., Salcedo, G., Sanchez-Monge, R., Hernandez-Lucas, C., Carmona, M. J., Lopez-Fando, J. J., Fernandez, J. A., Gomez, L., Royo, J., GarciaMaroto, F., Castagnaro, A., and Carbonero, P. 1992. Trypsinla-amylase inhibitors and thionins: possible defence proteins from barley. In: Barley: Genetics, Biochemistry, Molecular Biology and Biotechnology. pp. 335-350 (Shewry, P. R., Ed.). Wallingford: CAB International. Garcia-Olmedo, F., Rodriguez-Palenzuela, P., HernandezLucas, C., Ponz, F., Marana, C., Carmona, M. J., Lopez-Fando, J., Fernandez, J. A., and Carbonero, P. 1989. The thionins: a protein family that includes purothionins, viscotoxins and crambins. Oxford Sum. Plant Mol. Cell Biol. 6:3140. Gausing, K. 1987. Thionin genes specifically expressed in barley leaves. Planta. 171:241-246. Gu, Q., Kawata, E. E., Morse, M.-J., Wu, H.-M., andcheung, A. Y. 1992.A flower-specific cDNA encoding a novel thionin in tobacco. Mol. Gen. Genet. 234:89-96. Gundlach, H., Miiller, M. J., Kutchan, T. M., and Zenk, M. H. 1992. Jasmonic acid is a signal transducer in elicitor-induced plant cell cultures. Proc. Natl. Acad. Sci. U.S.A. 89:2389-2393. Harborne, J. B. 1988. Introduction to Ecological Biochernistry. London: Academic Press. Harvey, A. L. 1985. Cardiotoxins from cobra venoms: possible mechanisms of action. J . Toxicol. Toxin Rev. 4:41-69. Harvey, A. L. 1991. Cardiotoxins from cobra venoms. In: Handbook of Natural Toxins, Vol. 5 . Reptile Venoms and Toxins. pp. 85-106 (Tu, A. T., Ed.). New York: Marcel-Dekker. Hernandez-Lucas, C., Carbonero, P., and Garcia-Olmedo, F. 1978. Identification and purification of a purothionin homologue from rye (Secale cereale L.). J . Agric. Food Chem. 26:794-796. Hernandez-Lucas, C., Fernandez de Caleya, R., Carbonero, P., and Garcia-Olmedo, F. 1977. Reconstitution of petroleum ether soluble wheat lipopurothionin by binding of digalactosyl diglyceride to the chloroformsoluble form. J. Agric. Food Chem. 251287-1289. Hernandez-Lucas, C., Fernandez de Caleya, R., and Carbonero, P. 1974. Inhibition of brewer’s yeasts by wheat purothionins. Appl. Microhiol. 28:165-168. Hernandez-Lucas,C., Royo, J., Paz-Ares, J., Ponz, F., GarciaOlmedo, F., and Carbonero, P. 1986. Polyadenylation site heterogeneity in mRNA encoding the precursor of the barley toxin P-hordothionin.FEBS Lett. 200: 103106. Ishibashi, N., Yamauchi, D., and Minamikawa, T. 1990. Stored mRNA in cotyledons of Vigna unguiculata seeds: nucleotide sequence of cloned cDNA for a stored mRNA and induction of its synthesis by precocious germination. Plant Mol. Biol. 1 5 5 9 4 4 . 14 Jaynes, J. M., Xanthopoulos, K. G., Destefano-Beltran, L., and Dodds, J. H. 1987. Increasing bacterial disease resistance in plants utilizing antibacterial genes from insects. BioEssays. 6:263-270. Jiang, M.-S., Fletcher, J. E., and Smith, L. A. 1989. Factors influencing the hemolysis of human erythrocytes by cardiotoxins from Naja naja kaouthia and Naja naja atra venoms and a phospholipase A, with cardiotoxinlike activities from Bungarus fasciatus venom. Toxicon. 27:247-257. Johnson, R., Narvaez, J., An, G., and Ryan, C. 1989. Expression of proteinase inhibitors I and I1 in transgenic tobacco plants: effects on natural defense against Manduca sexta larvae. Proc. Natl. Acad. Sci. U.S.A. 86:9871-9875. Johnson, T. C., Wada, K., Buchanan, B. B., and Holmgren, A. 1987. Reduction of purothionin by the wheat seed thioredoxin system. Plant Physiol. 8544645 1. Jones, B. L. and Cooper, D. B. 1980. Purification and characterization of a corn (Zea mays) protein similar to purothionins. J . Agric. Food Chem. 28:904-908. Jones, B. L., Lookhart, G. L., Mak, A. S., and Cooper, D. B. 1982. Sequences of purothionins and their inheritance in diploid, tetraploid, and hexaploid wheats. J . Hered. 73:143-144. Jones, B. L. and Mak, A. S. 1977. Amino acid sequences of the two a-purothionins of hexaploid wheat. Cereal Chem. 54:511-523. Jones, B. L. and Meredith, P. 1982. Inactivation of a-amylase activity by purothionins. Cereal Chem. 59:321. Kashimoto, T., Sakakibara, R., Huynh, Q. K., Wada, H., and Yoshizumi, H. 1979. The effect of purothionin on bovine adrenal medullary cells. Res. Commun. Chem. Pathol. Pharmacol. 26:22 1-224. Konopa, J., Woynarowski, J. M., and LewandowskaGumieniak, M. 1980. Isolati