AU4317299A - Cotton cells, plants, and seeds genetically engineered to express insecticidal and fungicidal chitin binding proteins (lectins) - Google Patents

Description

WO 99/61635 PCT/US99/11766 DESCRIPTION COTTON CELLS. PLANTS, AND SEEDS GENETICALLY ENGINEERED 5 TO EXPRESS INSECTICIDAL AND FUNGICIDAL CHITIN BINDING PROTEINS (LECTINS) Field of the Invention The present invention is directed to chimeric genes that express in cotton cells, plants 10 and seeds, and encode insecticides and fungicides having substantially the insect toxicity and fungal toxicity of barley, nettle, and hevein lectins. Background of the Invention Chitin-binding proteins (lectins) are present in a wide range of plant species, including 15 both monocots and dicots, even though these plants contain no chitin. They are believed to be defense-related, and many exhibit insecticidal and/or anti-fungal activities (Murdock et. al., 1990; Lerner, D.R. and Raikhel, N.V., 1992). Lectins exhibit specific carbohydrate-binding properties. Lectins are presumably defense-related proteins in plants that exert their effect by binding to N-acetylglucosamine in susceptible pest species (Schroeder, M.R. and Raikhel, N. 20 V. 1992). In purified form, barley, nettle, and hevein lectins have shown insecticidal and fungicidal activity against certain species of pests which are known to attack cotton (for example, Heliothis and Fusarium). Various methods are available to utilize lectins to control such pests, but all require providing these proteins sufficiently pure and in sufficient quantity 25 to effect control of the target insect or pathogen. Even when available in sufficient purity or quantity, they must be applied to the crop in such a way so as to effectively reach the target species. Furthermore, because they are proteins, if topically applied to crops they are subject to light and protease inactivation before they can exert their controlling effect. Root associated pathogens are not readily treated with such preparations. Hence, lectins have not been available 30 for use in controlling many serious pests of cotton, even though they might be effective were they available in pure enough and high enough concentrations. By taking advantage of genetic engineering, a gene responsible for the production of a useful polypeptide can be transferred from a donor cell, in which the gene naturally occurs, to a host cell, in which the gene does not naturally occur; Cohen and Boyer, U.S. Patent Nos.

WO 99/61635 PCTIUS99/11766 2 4,237,224 and 4,468,464. There are, in fact, few inherent limits to such transfers. Genes can be transferred between viruses, bacteria, plants, and animals. In some cases, the transferred gene is functional, or can be made to be functional, in the host cell. When the host cell is a plant cell, whole plants can sometimes be regenerated from the cell. 5 Genes typically contain regions of DNA sequences including a promoter and a transcribed region. The transcribed region normally contains a 5' untranslated region, a coding sequence, and a 3' untranslated region. The promoter contains the DNA sequence necessary for the initiation of transcription, during which the transcribed region is converted into mRNA. In eukaryotic cells, the promoter 10 is believed to include a region recognized by RNA polymerase and a region which positions the RNA polymerase on the DNA for the initiation of transcription. This latter region, which is referred to as the TATA box, usually occurs about 30 nucleotides upstream from the site of transcription initiation. Following the promoter region is a sequence that is transcribed into mRNA but is not 15 translated into polypeptide. This sequence constitutes the so-called 5' untranslated region and is believed to contain sequences that are responsible for the initiation of translation, such as a ribosome binding site. The coding region is the sequence that is just downstream from the 5' untranslated region in the DNA or the corresponding RNA. It is the coding region that is translated into 20 polypeptides in accordance with the genetic code. Bacillus thuringiensis, for example, has a gene with a coding sequence that translates into the amino acid sequence of an insecticidal crystal protein. The coding region is followed by a sequence that is transcribed into mRNA, but is not translated into polypeptide. This sequence is called the 3' untranslated region and is believed 25 to contain a signal that leads to the termination of transcription and, in eukaryotic mRNA, a signal that causes polyadenylation of the transcribed mRNA strand. Polyadenylation of the mRNA is believed to have processing and transportation functions. Natural genes can be transferred in their entirety from a donor cell to a host cell. It is often preferable, however, to construct a gene containing the desired coding region with a 30 promoter and, optionally, 5' and 3' untranslated regions that do not, in nature, exist in the same gene as the coding region. Such constructs are known as chimeric genes. Barley lectin is a vacuolar protein synthesized with an amino-terminal signal sequence for entering the secretory pathway and a carboxyl-terminal propeptide necessary for proper targeting to the vacuole (Bednarek, S.Y., and Raikhel, N.V., 1991). The glycosylated carboxyl- WO 99/61635 PCT/US99/11766 3 terminal propeptide (CTPP) is removed before or concomitant with the deposition of the mature, active protein in vacuoles (Bednarek, et al., 1990). Mature barley lectin is a dimeric protein composed of two identical 18-kilodalton polypeptides (Wilkins, T.A., Bednarek, S.Y. and Raikhel, S.V., 1990). The nucleotide sequence and deduced amino acid sequence of the barley 5 lectin coding region (barley lectin cDNA clone BLc3) has been reported (see Lerner and Raikhel, 1989; and U.S. Patent No. 5,276,269, incorporated herein by reference). A chimeric gene construct was created by fusing the BLc3 coding region to the CaMV 35S promoter, and transferring the chimeric gene construct into tobacco plants via Agrobacterium tumefaciens mediated transformation (U.S. Patent No. 5,276,269). Plants were reported to exhibit 10 insecticidal and fungicidal properties. A full length cDNA clone (HEV 1) encoding Hevea brasiliensis lectin was isolated from a H brasiliensis latex cDNA library, sequenced, and characterized (see Broekaert et al., 1990; Lee et al., 1991; and U.S. Patent No. 5,187,262, incorporated herein by reference). Briefly, HEVI is 1018 nucleotides long and includes an open reading frame of 204 amino acids. The 15 deduced amino acid sequence contains a putative signal sequence of 17 amino acid residues followed by a 187 amino acid polypeptide. The amino-terminal region of 43 amino acids is identical to hevein and shows homology to several chitin-binding proteins and to the amino termini of wound-induced genes in potato and poplar. Northern blots, using HEVI cDNA as a probe, showed that the gene is induced by wounding and the plant hormones abscisic acid and 20 ethylene. Accumulation of these transcripts was seen in leaves, stems, and latex, but not in roots. Chimeric gene constructs fusing the hevein coding region with heterologous promoters were not reported. However, tests with hevein protein showed antifungal activity against Trichoderma, Phycomyces, Botrytis, Septoria, Pyricularia, and Fusarium. The observed activities differed from those of wheat germ aglutinin (another lectin). Furthermore, hevein anti 25 fungal activity was found to be stable even after heating to 900 C, a condition under which certain chitinase activities are completely destroyed. A full length cDNA encoding the nettle lectin (Urtica dioica agglutinin) has been cloned, sequenced, and characterized (Lerner and Raikhel, 1992). The protein is made up of 374 amino acids. 21 are a putative signal sequence and 86 amino acids encode the two chitin-binding 30 domains of nettle lectin. These are fused to a 19 amino acid "spacer" domain and a 244 amino acid carboxyl extension with partial identity to a chitinase catalytic domain. This gene represents another lectin heretofore unavailable as a source for resistance to important cotton insect and fungal pathogens. The studies noted above underscore the complexity of the biochemistry of plant lectins.

WO 99/61635 PCT/US99/11766 4 These are proteins which must be processed properly and transported into the proper subcellular compartment, usually a vacuole, where they are stored. In order to make use of these proteins in combating cotton pests, one viable approach is to generate chimeric gene constructs using various lectin genes and then transfer these into cotton using available transformation systems 5 (see for example, Rangan et al., U.S. Patent No. 5,244,802). Achieving an effective level of expression is not a given in heterologous systems. There would be no guarantee that the proteins would not have some unexpected toxic effect on the cotton plant itself, or that the proteins would exhibit the predicted pattern of activity. Furthermore, as noted above, some target pests attack plant tissues (for example, roots) in which some of these lectins are not normally expressed in 10 the plants from which they come. Hence, a lectin which might have activity against a given pest in a feeding assay following topical application to plant tissue (see, for example, Cavalieri et. al., U.S. Patent No. 5,407,454), may not exhibit that same activity when expressed in vivo. Cavalieri et al. provides somewhat suggestive evidence that a broad range of plant lectins may provide a level of control against certain corn pests. Unfortunately, those studies 15 were carried out using isolated lectin preparations for which essentially no biochemical characterization was provided. Some may even have been from commercial providers, where composition can vary from preparation to preparation. Hence, commercial providers include lot numbers with their products so that problems can be traced back on a lot by lot basis. Purity of the preparations was not discussed by Cavalieri, nor did they provide information on how they 20 obtained their lectins or discuss the actual number of different lectins which may have been present in a given preparation. Any plant species may produce several different lectins, and protein preparations are readily contaminated with multiple protein species which may be present in trace amounts, but have a significant effect, positive or negative, on observed activity. Hence, the preparations tried may have actually been mixtures of lectins and even other proteins 25 derived from the plants in question. No data were provided on the source of the lectin preparations used, on their purity, or hence on which of the lectin genes in a given plant the actual activity observed was based. Such preparations could have distinctly different insecticidal and fungicidal activities than a lectin provided in purified form from the expression in planta of a single lectin gene. 30 The best way to provide a protein in purified form, and therefore be certain of its activity against a given pest, is to isolate the gene and express the protein in an in vitro system. Since genes for most of the lectins cited in their study have still not been cloned as of this date, in vitro expression of single, purified lectins for analysis was not possible at the time Cavalieri et al. reported their data. Suggestive as their data is with respect to certain corn pests, Cavalieri et al.

WO 99/61635 PCTIUS99/11766 5 do not provide a single example of activity against a serious pest of cotton. Hence, their study is suggestive, but does not disclose a single lectin , in purified form, which one might use to control a significant pest of cotton. Conversely, proteins which do not have activity in a feeding assay following topical 5 application to plant tissues, may have activity when expressed in vivo. This could particularly be true in cotton, where plants normally express a compound called gossypol which is known to suppress feeding of certain insect pests. Thus, there could be synergistic effects between gossypol and lectins in such a way so as to enhance the insecticidal activity of a given lectin against important cotton pests. Alternatively, gossypol expression could suppress feeding just 10 enough so that the target insect might never consume a potentially lethal amount of lectin. Hence, one could not know the insecticidal or fungicidal effect of a lectin gene transferred into cotton until such cotton cells, plants, and seeds were created. Raikhel (U.S. Patent No. 5,276,269) showed that a chimeric barley lectin gene under control of the CaMV 35S promoter could be transferred into tobacco plants to produce a single 15 species lectin protein which was transported properly and thereby create a plant with new insecticidal and fungicidal properties. With the further availability of the hevein (Raikhel, U.S. Patent No. 5,187,262) and nettle genes due to cloning (Lerner and Raikhel, 1992), it has now become possible to create cotton plants expressing in highly purified form each of these lectins and to test those cells, plants, and seeds for the presence of new insecticidal and fungicidal 20 activities. Objects of the Invention It is an object of the present invention to provide cotton cells, plants, and seeds expressing chimeric barley, nettle, and hevein lectin genes in amounts and under conditions 25 which are sufficient to impart substantially the pesticidal properties such as insecticidal and fungicidal properties of barley, nettle, and hevein lectins to said cotton cells, plants, and seeds. It is a further object of the present invention to provide a method for killing cotton insect pests and pathogens by feeding them cotton cells, plants, and seeds containing chimeric genes 30 that express pesticidal (for example, insecticidal and fungicidal) amounts of a toxin having substantially the insect toxicities and fungal toxicities of barley, nettle, and hevein lectins. It is an additional object of the present invention to provide the genes and other DNA segments within the cotton cells, plants, and seeds associated with the above methods.

WO 99/61635 PCTIUS99/11766 6 Summary of the Invention These and other objects of the present invention have been achieved by providing chimeric genes capable of expressing in cotton cells, plants, and seeds a polypeptide having substantially the pesticidal toxicity (for example, the insect toxicity) and fungal toxicity of 5 barley, nettle, and hevein lectins, in plant cells in culture and plant cells in living plants and seeds; as well as methods for producing a toxin having substantially the pesticidal properties (for example, the insect toxicity and fungal toxicity) of barley, hevein, and nettle lectins in cotton cells, plants, and seeds; and methods for killing cotton pests such as insects by feeding them cotton cells, plants, and seeds containing genes that express these toxins. 10 Brief Description of the Drawings Figure 1 illustrates the gene map and 35S promoter region of the binary plant expression vector pGA643 (described in An et al., 1988), which is useful for expression in plants of lectin genes (Wilkins et al., 1990; Raikhel U.S. Patent No. 5,276,269, incorporated herein by 15 reference). Figure 2 shows the nucleotide sequence of barley lectin cDNA clone BLc3 (Lerner and Raikhel, 1989; Raikhel U.S. Patent No. 5,276,269). Figure 3 shows the nucleotide sequence of the hevein cDNA clone "HEVI" (Broekaert et al., 1990; Raikhel U.S. Patent No. 5,187,262, incorporated herein by reference). 20 Figure 4 shows the nucleotide sequence of the nettle lectin cDNA clone MK209 (Urtica dioica agglutinin; Lerner, D. R. and Raikhel, N.V., 1992, incorporated herein by reference). Detailed Description The present invention is directed to a chimeric gene that expresses in cotton cells, plants, 25 and seeds, and encodes pesticides such as insecticides and fungicides having substantially the insect toxicity and fungal toxicity of barley, nettle, and hevein lectins. The cotton plant cells contemplated include cells from any and all cotton plants into which foreign DNA can be introduced, replicated, and expressed. Some suitable examples of cotton plant species include Gossypium hirsutum, Gossvpium arboreum, and Gossypium 30 barbadense. The term "plant cell" refers to any cell derived from a cotton plant. Some examples of cells encompassed by the present invention include differentiated cells that are part of a living plant; undifferentiated cells in culture; the cells of undifferentiated tissue such as callus or tumors; seeds; embryos; propagules; and pollen.

WO 99/61635 PCT/US99/11766 7 The chimeric gene of this invention contains a promoter region that functions efficiently in cotton plants and a coding region that codes for the barley lectin encoded in pBLc3, the hevein lectin encoded in the cDNA clone HEVI, and/or the nettle lectin encoded in the cDNA clone MK209. The coding sequence of the chimeric gene is not known to be associated with the 5 promoter in natural genes. The 5' and/or 3' untranslated regions may, independently, be associated in nature with either the promoter or the coding region, or with neither the promoter or the coding region. Preferably, either the 5' or the 3' untranslated region is associated with the promoter in natural genes, and most preferably both the 5' and 3' regions are associated with the promoter in natural 10 genes. One could not predict, based on the state of the art at the time this invention was made, that a chimeric barley, hevein, or lectin gene could be functionally introduced into cotton cells. It was even less predictable that such cells would express such lectins at sufficient levels to impart pesticidal (for example, insecticidal or fungicidal) properties to the cells. 15 In order to be considered pesticidal (for example, insecticidal or fungicidal), the plant cells must contain an insecticidal or fungicidal amount of lectin having substantially the insecticidal and fungicidal activity of purified lectin from barley, rubber, or nettle. Having "substantially the insecticidal and fungicidal activity of purified lectin" means exhibiting activity against substantially the same range of insects or fungi as does the corresponding lectin purified 20 from its native host. An insecticidal or fungicidal amount is an amount which, when present in plant cells, kills insects or fungi or at least significantly inhibits a function necessary for growth, such as feeding. Such inhibition is that which can be measured as statistically significant when compared with a control. Accordingly, the plant cells, plants, or seeds of the present invention are able to withstand attacks by cotton pests such as insects, nematodes, or 25 fungi without, or with less, application of purified barley lectin, hevein, nettle lectin, or other insecticides or fungicides when compared with plant cells, plants, or seeds that do not contain a gene producing barley lectin, hevein, or nettle lectin. Following are examples which exemplify certain embodiments of the subject invention. These examples are illustrative and should not be construed as limiting the subject invention in 30 any manner. Example 1: The Genes Three different chimeric plant lectin genes (barley, hevein, and nettle), were evaluated in this study. Each comprised a cDNA for a given specific lectin driven by a promoter active WO 99/61635 PCT/US99/11766 8 in cotton. For convenience, the CaMV 35S promoter was used, but any promoter proven to be active in cotton, such as the A. tumefaciens T-DNA promoters, A. rhizogenes T-DNA promoters, or the cotton chlorophyll A/B binding protein gene promoter (Anderson, et al., 1993) would be useful. This list is exemplary, but not intended to be all inclusive. One skilled in the 5 art will recognize other useful promoters which can be used to express barley, hevein, and nettle lectins in appropriate cotton cells, plants, and seeds to control problematic cotton pests such as insects and fungi. An expression cassette comprising the coding region for barley lectin operably linked to the CaMV 35S promoter was created by ligating the pBLc3 cDNA sequence (Figure 2) into 10 the plant cloning vector pGA643 (Figure 1; An et al., 1988) as described in Raikhel, U.S. Patent No. 5,276,269 and incorporated herein by reference, taking advantage of the Xbal restriction endonuclease sites in pBLc3 and pGA643. Transformation was into the E. coli strain DH5a. Proper orientation of the coding region of the insert relative to the promoter region was confirmed by restriction endonuclease mapping and DNA sequence analysis. The clone 15 comprising the coding region barley lectin cDNA pBLc3 in pGA643 can be obtained from Dr. N. Raikhel, MSU-DOE Plant Research Laboratory, Michigan State University, East Lansing, MI, 48824. An expression cassette comprising the coding region for hevein (Hevea brasiliensis aglutinin) operably linked to the CaMV 35S promoter was created by ligating the hevein cDNA 20 sequence HEVI (Figure 3; Broekaert et al., 1990; Raikhel, U.S. Patent No. 5,187,262) into the plant cloning vector pGA643 (Figure 1; An et al., 1988) taking advantage of the Xbal and BglII restriction endonucleases which release the insert from HEV 1 and cleave within the polylinker region of pGA643. Transformation was into the K coli strain DH5a. Proper orientation of the coding region of the insert relative to the promoter region was confirmed by restriction 25 endonuclease mapping and DNA sequence analysis. The clone comprising the HEVI cDNA inserted into pGA643 can be obtained from Dr. N. Raikhel, MSU-DOE Plant Research Laboratory, Michigan State University, East Lansing, MI, 48824. An expression cassette comprising the coding region for nettle lectin operably linked to the CaMV 35S promoter was created by ligating the nettle cDNA sequence (Figure 4) into the 30 plant cloning vector pGA643 (Figure 1; An et al., 1988). This was accomplished by releasing the insert from the nettle cDNA clone MK209 with Xbal and ligating this fragment into the Xbal restriction endonuclease site within the polylinker region of pGA643. Transformation was into the K coli strain DH5a. Proper orientation of the coding region of the insert relative to the promoter region was confirmed by restriction endonuclease mapping and DNA sequence WO 99/61635 PCT/US99/11766 9 analysis. The nettle cDNA clone MK209 and the clone comprising the nettle coding region inserted into pGA643 can be obtained from Dr. N. Raikhel, MSU-DOE Plant Research Laboratory, Michigan State University, East Lansing, MI, 48824. All three binary vector constructs were mobilized from the E. coli strain DH5a into A. 5 tumefaciens LBA4404 by triparental mating (Hooykaas, P.J.J., 1988) using the E. coli strain HB101 harboring the wide-host range mobilizing plasmid pRK2013 (Clonetech, Palo Alto, California). Transconjugates were selected on minimal nutrient plates (An et al., 1988) containing kanamycin (5 pg/ml) and tetracycline (12.5 ptg/ml). 10 Cotton Transformation with Chimeric Lectin Genes Example 2: Cotton Regeneration The establishment and maintenance of cotton embryogenic suspension cultures was as described in Rangan et al. (U.S. Patent No. 5,244,802, incorporated herein by reference), as further modified in Rajasekaran et al., 1996 (incorporated by reference). For convenience, the 15 cotton line B1654 was used. Many other upland or Pima cotton varieties will work equally well, and those skilled in the art would make their variety selection on the basis of the needs of their program. Seeds were surface sterilized by first treating with 70% ethanol for 3 min, followed by a 20 min treatment with a 20% CLOROX solution (1% available chlorine) containing 0.01% of 20 the surfactant TWEEN-20. Seedlings were grown under 16 h light (40-60 1 IE m- 2 s- 1 ) and 8 h dark at 26 ± 20 C on agar-solidified (TC Agar, Hazleton Biologics, Lenexa, KS) White's medium (Singh and Krikorian, 1981) containing 1 mg/l kinetin. Embryogenic callus cultures were first established from seedling explants according to the procedures of Rangan (U.S. Patent No. 5,244,802). Briefly, cotyledon and hypocotyl explants from 7- to 10-day old seedlings were 25 placed on a callus induction medium (MS, Murashige and Skoog, 1962) supplemented with 0.4 mg/l thiamine HCl, 30 g/l glucose, 2.0 mg/l a-naphthaleneacetic acid (NAA), 1.0 mg/l kinetin, 100 mg/l myo-inositol and 0.8% (w/v) agar. The cultures were incubated at 27 ± 2 "C under conditions of 16h light and 8h dark, light intensity at 60 pE m- 2 s-1, in an environmentally controlled incubator (Percival, Boone, IA). Callus formed on these explants within three to four 30 weeks. Callus pieces were selectively subcultured to enrich for friable, yellowish-green callus every three to four weeks on the same medium, except the carbon source was sucrose (20 g/l) instead of glucose. Depending on the variety, embryogenic callus capable of forming small globular somatic embryos appeared one to four subcultures after initiation. Embryogenic callus was maintained and multiplied by routine subculture every three to four weeks on MS medium WO 99/61635 PCT/US99/11766 10 containing 100 mg/i myo-inositol, 20 g/l sucrose, 2.0 mg/l NAA and 0.8% (w/v) agar (maintenance medium). Cell suspension cultures were initiated from finely dispersed embryogenic callus cultures in liquid maintenance medium agitated (120 rpm, 27 2' C) on a gyratory shaker (New 5 Brunswick G-10, Edison, NJ). The suspension cultures were enriched for small, isodiametric, densely cytoplasmic and highly embryogenic cells by periodically discarding free floating cells and large aggregates (> 840 ptm) every week. Two days before use, these cultures were subcultured in 250 ml Erlenmeyer flasks containing 40 ml of maintenance medium. The cell suspension cultures used 10 in our experiments were rapid growing embryogenic cells that exhibited a doubling of fresh weight in four to six days (the logarithmic phase of growth begins two days after subculture). All cell suspension cultures used for biolistic transformation experiments had a cumulative age of three to four months. 15 Example 3: Biolistic transformation of embryogenic cotton cultures The three plasmids (barley lectin coding region in pGA643; hevein coding region in pGA643; and nettle lectin coding region in pGA643) were used to coat 1.0 uM gold particles, and then projected into embryogenic cotton suspension cell cultures using an improved helium driven biolistic device (PDS 1000/He; BioRad). Briefly, 50 pl of a gold micro-carrier 20 suspension (1p gold particles) in water was used. In an 1.5ml micro-centrifuge tube, under continuous vortexing, the following were added in order: 5pl DNA (Ig/ptl), 50p1l of 2.5M CaCl, and 20pl of 0.1M spermidine (free base, tissue culture grade). Vortexing was continued for 3 minutes, the micro-carriers were spun down at 10,000 rpm for 10 seconds, and as much of the supernatant was removed as possible. The micro-carriers were washed with 250 pl of 100% 25 ethanol (HPLC or spectrophotometric grade) by vortexing briefly, followed by centrifugation and removal of the supernatant. The micro-carriers were resuspended in 60 d of 100% ethanol. 7.5ptl of this DNA coated micro-carrier mix were used per macro-carrier disk. The bombardments were performed using a membrane rupture pressure of 1550 psi and other device settings as described by Hamilton et al. (8). The cell suspensions established as 30 described above, (<840 ptm fraction), subcultured two days earlier, were vacuum-deposited as a thin layer onto moist filter paper (Whatman No. 1; 3.5 cm diameter) in sterile Petri dishes (5.5 cm diameter). One ml of suspension cells (1 X 106 cells) was transferred to each dish. A 400 mesh nylon screen was placed over the surface of the suspensions to serve as a baffle. The optimal bombardment conditions included the use of 10 MPa rupture disks, a distance between WO 99/61635 PCTIUS99/11766 11 the stopping screen and the cell suspensions of 7.5 cm and a macro-carrier travel distance of 10 mm. During the bombardment, the vacuum in the sample chamber was 95 kPa. Bombardment of the cells was repeated three to five times at two-day intervals to maximize the transformation frequency. 5 Following particle bombardment, the cell suspension cultures were grown for a week without any selection in maintenance medium. The pGA643 binary vector carries a neomycin phosphotransferase II gene for selection of transformed cells (Figure 1). Accordingly, cell suspensions were selected with the antibiotic G418 (10ptg/ml). Selection with the antibiotic G418 was applied by gradually increasing the concentration each week. Selection with G418 10 was initiated at 10 pg/ml and increased by 10 pg/ml increments at five to seven day intervals to achieve a final concentration of 50pg/ml after three to four weeks. Alternatively, in some experiments, cells were directly exposed to only one high level of antibiotic (G418 at 50 pg/ml) at the beginning of the selection process. Independent transformation events arose as separated growing colonies in the presence of the selective agent. Each colony so arising was maintained 15 separately and verified as a true transformant via NPTII ELISA (Firoozabady et al., 1987). Cotton plants are regenerated from embryogenic suspension cultures as described in Rangan et al., U. S. Patent No. 5,593,036 (incorporated herein by reference). Example 4: Agrobacterium transformation with lectin genes 20 The three binary vector plasmids (barley lectin coding region in pGA643; hevein coding region in pGA643; and nettle lectin coding region in pGA643) were mobilized into the binary A. tumefaciens host strain LBA4404 by triparental mating as previously described. Transformation of cotton primary explants can be accomplished by a number of approaches (Firoozabady et al., 1987; Umbeck et al., 1987; Rangan et al., U. S. Patent No. 5,244,802). For 25 convenience, the method of Rangan et al., U. S. Patent No. 5,244,802, as modified by Rajasekaran et al., 1996, is briefly described. Agrobacterium cultures for transformation experiments were initiated in 50ml of YEB liquid medium using frozen glycerol stocks (500il) as inoculum. These cultures were grown overnight for about 18 h at 26 ± 2' C on a gyratory shaker. The optical density (A600) values 30 were adjusted to 0.6-0.8 in liquid MS medium prior to use. Cotyledon (1 cm 2 ) explants for Agrobacterium transformations are prepared from 5- to 7-day old seedlings. The explants are treated with an Agrobacterium suspension as prepared above for 15 to 30 min, blotted dry, and then plated on 12 cm diameter filter paper (Whatman No. 1) placed on freshly made, agar-solidified callus induction medium (Rangan U. S. Patent WO 99/61635 PCT/US99/11766 12 No. 5,244,802) in 15 cm diameter Petri dishes containing 60 ml of medium. Cocultivation is carried out for 48 h in a Percival incubator maintained at 26 ± 20 C, 16 h light, 60-90 piE m-s 1 . Following cocultivation, the explants are thoroughly washed in MS liquid medium containing 200 mg/l cefotaxime (Cal-Biochem) and 200 mg/I carbenicillin (Sigma), blotted dry, and placed 5 on freshly prepared callus induction medium containing the antibiotic G418 (10 mg/l; Gibco BRL, Life Technologies, Gaithersburg, MD) as the selection agent and the same concentrations of cefotaxime and carbenicillin as above to control bacterial growth. Cotyledon segments are plated at seven per Petri dish (9 cm diameter) containing 25 ml callus induction medium. After the first subculture the explants are transferred to freshly made callus induction medium to 10 encourage more callus production in the presence of selection pressure. Transformed (antibiotic resistant) callus develops 3-8 weeks after transformation. Individual callus colonies are subcultured separately to maintain identity of separate integration events. NPT II ELISAs are carried out according to the procedures of Firoozabady et al., 1987 to confirm that antibiotic resistant callus colonies are transformed. Transformed colonies are regenerated to plants as 15 described (Rangan, et al., U. S. Patent No. 5,244,802) Results Example 5: Confirmation of Cotton Transformation with Lectin Genes Cotton cell lines (embryogenic colonies) transformed with barley, nettle, or hevein lectin 20 genes in pGA643 were maintained as independent colonies in culture and confirmed to be transformed by NPTII ELISA as described above. To verify the co-transformation of the appropriate lectin gene along with the selectable marker in the transformation system employed, several NPTII ELISA positive colonies transformed with BLc3 were assayed using double-bind ELISA in methods similar in principle to those of Raikhel et al., 1984, but modified to be more 25 suitable for transformed cotton cells. Wheat germ agglutinin antibody, which is available commercially, will cross react with barley lectin (Wilkins et al., 1990) and hence can be used in detecting expression of BLc3 protein in transformed cotton cells using WGA ELISA. It was observed in initial studies with transformed cotton cells that cotton extracts give a high background reading when in these WGA 30 ELISA tests for transformation. The following protocol was developed which overcomes this background problem and enabled the confirmation of co-transfer of lectin genes along with the antibiotic marker gene using the methods in the present invention. Rabbit anti-wheat germ aglutinin (6mg/ml) and biotinylated rabbit anti WGA (3.5mg/ml) were purchased from E.Y. Laboratories. Primary antibody solution (1p g/ml) was WO 99/61635 PCT/US99/11766 13 prepared by diluting 1.8 tl of rabbit anti-WGA stock with 11 ml carbonate binding buffer (Na 2

CO

3 1.59g, NaHCO 2.93g, H20 to 1L, pH 9.6) and kept on ice. 100pl were applied to each well of a 96 well ELISA plate (Corning #25805-96), sealed and kept overnight at 40 C. Pre adsorbed antibody was then prepared as follows. Four grams of control (non-transformed) callus 5 was homogenized in 6ml of PBS Tween prepared from 50X concentrate (Agdia, Elckhart, Indiana) containing 1% PVP 40,000 and centrifuged at 8,000 rpm for 10' to pellet cell debris. 5.5ml of the supernatant was mixed with 5.5m] of PBS TWEEN containing 0.1% BSA and 4% PEG 8,000. To this was added 9.4pl biotinylated rabbit anti- WGA (E.Y. Laboratories, 3.5mg/ml) for a final antibody concentration of 3ig/ml. This was then incubated on ice for 3 10 hours to preadsorb the antibody. ELISA plates were removed from the overnight incubation and washed thoroughly (4X) with PBS TWEEN. A blocking step was performed by filling each well of the plate with 1% BSA in PBS without TWEEN. PBS without TWEEN is prepared by combining 5ml of a 10% stock w/v of Bovine Serum Albumin (Fraction V, ICN Pharmaceuticals #81-066 in water) with 15 5 ml of PBS(NaCl 8.0g, Na 2

HPO

4 .2H 2 0 1.44g, KH 2

PO

4 0.2g, KCI 0.2g, H20 to IL, adjusted to pH 7.4). The plates were incubated at room temperature (22 0 C-24'C) for 1 hour and then washed 4X with PBS TWEEN. Extracts from embryogenic cell lines transformed with BLc3 were prepared as follows. About 0.5g of callus was homogenized in 130p PBS Tween containing 1% PVP 40,000 in a 20 1.5ml micro-centrifuge tube, centrifuged at 10,000 rpm to pellet cell debris, and held on ice. 1 00[d of the supernatant was added to each well of the ELISA plates following the 1 hr blocking, washing step noted above. Plates were incubated for 3 hours at room temperature and washed 4X with PBS TWEEN. 100ptl of pre-adsorbed, biotinylated antibody were then added to each well of the plate, the plates were incubated overnight at 4'C, and washed 4X with PBS TWEEN. 25 Eleven ml of a 1:3000 dilution of streptavidin/alkaline phosphatase conjugate (from 5 'to 3') was prepared in PBS (no TWEEN) containing 1% BSA. 100p were applied to each well of the ELISA plates and the plates were incubated for 1 hour at room temperature. The plates were washed 4X with PBS TWEEN. 200pl of PNP (paranitrophenyl phosphate; Sigma 104 phosphate substrate #104-0) in 10% diethanolamine + 0.5mM MgCl 2 , pH 9.8 (prepared 30 immediately prior to use) was added per well and the color reaction was allowed to develop for 20 minutes at room temp. The reaction was stopped by the addition of 50pl of 3N NaOH and the plates were read in a microplate reader at a wavelength of 4102. Results of assays with several transformed embryogenic lines are presented in the following Table.

WO 99/61635 PCT/US99/11766 14 Table 1. Results of Immunoassays in WGA ELISA with cotton cells transformed with Blc3 in pGA643 Sample # Colony # Lectin DNA ELISA Result 1 control none 5 2 control none 3 control none 4 control none 5 control none 6 75 BLC 10 7 95 BLC 8 105 BLC 9 138 BLC 10 158 BLC 11 171 BLC 15 12 173 BLC 13 175 BLC 14 176 BLC ++ 15 177 BLC + 16 178 BLC ++ 20 17 180 BLC 18 181 BLC 19 183 BLC 20 184 BLC + 21 185 BLC 25 22 186 BLC + 23 187 BLC 24 188 BLC ++ 25 189 BLC 26 190 BLC ++ 30 27 191 BLC 28 192 BLC 29 194 BLC + 30 195 BLC ++ 31 197 BLC ++ 35 32 198 BLC + 33 200 BLC ++ 35 203 BLC 36 205 BLC ++ 37 207 BLC 40 38 209 BLC 39 211 BLC ++ 40 216 BLC + 41 25pe/ml WGA ++ -= no signal detected. 45 +, ++, +++ indicates a signal detected and gives the relative intensity, with +++ being most intense.

WO 99/61635 PCTIUS99/11766 15 The data in Table 1 confirm co-transfer of the lectin gene along with the NPTII selectable marker. Nearly 50% of the transformed embryogenic cell lines expressed sufficient lectin protein to be detectable in this assay. However, it is also evident that there was variability 5 in the extent of the detectability of the BLc3 protein in these assays. This could be due to differences in the level of lectin protein expression in the separate transformation events represented by the different lines assayed. Example 6: Cotton cells transformed with BLc3 are insecticidal 10 In order to confirm the insecticidal nature of cotton cells transformed with BLc3, feeding assays were preformed with larvae of the genus Heliothis. Heliothis species are economically important pests of cotton. Transformed embryogenic callus cultures which scored positive in both NPTII ELISA and WGA ELISA were selected for assay. Colonies were divided in two, with one half maintained in culture for regeneration to plants and one half used for the feeding 15 assay. For the assay, >lg of tissue was frozen in liquid nitrogen, lyophilized, and stored at -75' C until used in the feeding assay. Because corn ear worm is a significant pest of cotton, corn ear worm diet was used for this example. One skilled in the art would know to use an appropriate insect diet should one desire to assay transformed tissues for insecticidal activity against other insects. Insect diet was 20 prepared as follows. 2.6g agar was dissolved in 157ml H 2 0, boiled for 1 minute, and 40.6g of corn ear worm diet (Bioserv Product #9394) was added and the mixture was stirred well. 1.5ml of this mixture was added to each well of a 16 well insect feeding tray. To each well was then added 25mg of lyophilized callus sample or control (non-transformed) sample as appropriate. Callus transformed with a truncated Bacillus thuringiensis (B.L) crystalline endotoxin (CrylAb) 25 gene ("pPHY3") was used as a positive control. This callus expresses Bt protein at a low level and is toxic to Heliothis larvae. Callus transformed with an NPTII gene ("pUC/NEO"), but without a lectin gene, served as the negative control. Heliothis virescens larvae, a serious pest of cotton were used in this example. Eggs were allowed to hatch, and newly emerged larvae were applied, one per well of the insect feeding tray. 30 After 6 days, larvae were weighed in order to determine extent of growth. The data are summarized in Table 2, which follows.

WO 99/61635 PCT/US99/11766 16 Table 2. Results of Growth of Heliothis larvae on diet supplemented with cotton tissue transformed with BLc3 Callus #. WGA ELISA Avg. Larval Wt. Comment Result Increase (% of Control) 5 pUC/NEO 100% Transformed with NPTII gene only BLC 194 + 97% 11% of the suppression achieved with pPHY3. 10 BLC 178 +++ 90% 37% of the suppression achieved with pPHY3. BLC 195 +++ 89% 40% of the suppression achieved with pPHY3. 15 pPHY3 73% 27% suppression of growth of negative control BLC numbers correspond to sample numbers in Table 1. Eight Heliothis larvae were 20 tested for each callus test sample prepared as described in the text. The % growth weight increases shown are the average for the eight larvae after 6 days of feeding on the indicated test sample mix. The negative control sample was prepared using callus transformed with WO 99/61635 PCT/US99/11766 17 the NPTII marker alone (no lectin gene). The positive control was from tissue transformed with pPHY3. The data in Table 2 show that cotton embryogenic callus transformed with BLc3 suppresses the growth of Heliothis larvae, and indeed killed some larvae, even with the relatively 5 small amounts (25mg) of lyophilized transformed callus mixed into the artificial diet in these studies. Example 7: Cotton cells transformed with HEVI (hevein) and MK209 (nettle lectin) are insecticidal 10 Corn ear worm diet supplemented with lyophilized callus was prepared as described in Example 6, except that the callus samples were derived from transformations carried out with HEVI and with MK209. Newly hatched larvae (1 per feeding test plate well) were placed on the test medium, incubated at room temperature, and then scored after 7 days. The data are summarized in the following Table 3. 15 WO 99/61635 PCTIUS99/11766 18 Table 3. Results of Growth of Heliothis larvae on diet supplemented with cotton tissue transformed with either HEVI (hevein) or MK209 (nettle lectin) 5 Callus #. Avg. Larval Wt. Comment Increase (% of Control) pUC/NEO 100% Transformed with NPTII gene only 10 HEV 30 95% 35% of suppression achieved with Bt. MK209 34 86% 108% of suppression achieved with Bt. 15 pPHY3 87% 13% suppression of growth versus negative control The insect diet formulations employed in the present study included a very small percentage by weight of the test callus. Accordingly, the extent of insecticidal activity observed is to be deemed significant when one considers the relative activity versus the positive control. 20 Although all lectin genes tested showed significant activity against Heliothis, the nettle lectin MK209 demonstrated the highest level of activity relative the B.t. endotoxin in this study. One skilled in the art would know to use these methods to test cotton tissues transformed with the exemplified or other lectin genes for insecticidal activity against other insects and pests of economic importance in cotton production. Examples of such insects and pests include 25 cutworms (Agrotis spp., Paridroma spp., Euxoa spp., Feltia spp.), thrips (Franklinialla spp.), aphids (Aphis gossvpii), bollworms (Heliothis spp., Pectinophora spp., Helicoverpa spp.), budworms (Heliothis spp.), plant bugs (Lygus spp., Euschistus spp.), boll weevil (Anthonomus grandis), arnyworms (Spodoptera spp.), loopers (Alabama spp.), caterpillars (Estigmene spp.), WO 99/61635 PCT/US99/11766 19 cotton leaf perforator (Bacculatrix spp.), spider mites (Tetranychus spp.), whiteflies (Bemisia spp., Trialeurodes spp.), nematodes (Meloidogyne spp., Rotylenchulus spp. Hoploaimus spp.), and fungal pathogens (Verticillium spp., Fusarium spp., Pvthium spp., Rhizoctonia spp., Thielaviopsis spp. Phytophthora spp.). In view of the foregoing examples, the skilled artisan 5 will expect such tests with other lectins to be successful. Accordingly, it is clear that the invention is one encompassing embodiments other than those presented in the illustrative examples, and is to be construed by reference to the appended claims.

WO 99/61635 PCT/US99/11766 20 References An G., Ebert, P.R., Mitra, A., and Ha, S.B (1988) Binary vectors, Plant Mol. Biol. Manual A3:1 19. Anderson, D.M., Hudspeth, R.L., Hobbs, S.L., and Grula, J.W. (1993) Plant Physiol. 102:1047 1048. Blake, M.S., Johnston, K.H., Russel-Jones, G.J., and Gotschlich, E.C. (1984) Anal. Biochem. 136:175-179. Broekaert, W., Lee, H.-i., Kush, A., Chua, N.-H., and Raikhel, N. (1990) Proc. Natl. Acad. Sci. US.A. 87:7633-7637. Firoozabady, E., DeBoer, D., Merlo, D., Halk, E. Amerson, L., Rashka, K., and E. Murray (1987) Plant Mol. Biol. 10:105-116. Hooykaas, P.J.J. (1988) Plant Molec. Biol. Manual A4:1-13. Lee, Hyung-il, Broekaert, W F., and Raikhel, N.V. (1991) J. Biol. Chem. 266:15944-15948. Lerner, D.R., and Raikhel, N.V. (1989) Plant Physiol. 91:124-129. Lerner, D.R., and Raikhel, N.V. (1992) J. Biol. Chem. 267:22694. Murashige, T. and Skoog, F. (1962) Physiol. Plant. 15:4730497 Murdock et. al. (1990) Phytochemistry 29:85-89. Raikhel, N.V., Mishkind, M.L., and Palevitz, B.A. (1984) Planta 162:55-61. Rajasekaran, K. Grula, J.W., Hudspeth, R.L., Pofelis, S., and Anderson, D.M. (1996) Molecular Breeding 2:307-319. Schroeder, M.R. and Raikhel, N. V. (1992) Protein Expr. Purif 3:508-511. Singh, M. and Krikorian, A.D. (198 1) Ann. Bot. 47:133-139. Umbeck, P., Johnson, G., Barton, K., and W. Swain (1987) Bioltechnology 5:263-266. Wilkins, T.A., Bednarek, S.Y., and Raikhel, N.V. (1990) The Plant Cell 2:301-313.

Claims (49)

Claims

1. A plurality of cotton cells comprising a heterologous coding sequence encoding a lectin, said coding sequence being expressed in said cells, whereby said cells are rendered pesticidal.

2. The cotton cells of claim 1, wherein said cells are insecticidal.

3. The cotton cells of claim 1, wherein said cells are fungicidal.

4. The cotton cells of claim 1, wherein said cells are nematocidal.

5. The cotton cells of claim 1, wherein said lectin-encoding sequence is derived from barley.

6. The cotton cells of claim 1, wherein said lectin-encoding sequence is derived from nettle.

7. The cotton cells of claim 1, wherein said lectin-encoding sequence is derived from hevein.

8. A cotton plant, comprising cotton cells according to claim 1.

9. The cotton plant of claim 8, wherein said cells are insecticidal.

10. The cotton plant of claim 8, wherein said cells are fungicidal.

11. The cotton plant of claim 8, wherein said cells are nematocidal.

12. The cotton plant of claim 8, wherein said cells comprise a lectin-encoding sequence derived from barley.

13. The cotton plant of claim 8, wherein said cells comprise a lectin-encoding sequence derived from nettle.

14. The cotton plant of claim 8, wherein said cells comprise a lectin-encoding sequence derived from hevein.

15. A seed of the cotton plant of claim 8, said seed comprising cells which comprise a heterologous lectin-encoding sequence.

16. A seed according to claim 15, wherein said lectin-encoding sequence encodes an insecticidal lectin.

17. A seed according to claim 15 wherein said lectin-encoding sequence encodes a fungicidal lectin.

18. A seed according to claim 15, wherein said lectin-encoding sequence encodes a nematocidal lectin.

19. A seed according to claim 15, wherein said lectin-encoding sequence is derived from barley.

20. A seed according to claim 15, wherein said lectin-encoding sequence is derived from nettle.

21. A seed according to claim 15, wherein said lectin-encoding sequence is derived from hevein.

22. A method of producing a pesticidal lectin comprising the steps of: obtaining a pesticidal lectin-encoding polynucleotide; transforming a cotton cell with said polynucleotide; culturing said cell under conditions whereby descendant cotton cells comprising said polynucleotide or a plant comprising said descendant cells are produced; and verifying that said polynucleotide is expressed in said descendant cells, whereby a pesticidal lectin is produced.

23. The method of claim 22, wherein said lectin is insecticidal.

24. The method of claim 22, wherein said lectin is fungicidal.

25. The method of claim 22, wherein said lectin is nematocidal.

26. The method of claim 22, wherein said polynucleotide is derived from barley.

27. The method of claim 22, wherein said polynucleotide is derived from nettle.

28 The method of claim 22, wherein said polynucleotide is derived from hevein.

29. A method of producing a pest-resistant cotton plant comprising the steps of: obtaining a pesticidal lectin-encoding polynucleotide; transforming a cotton cell with said polynucleotide; culturing said cell under conditions whereby a cotton plant comprising descendant cells of said cell, which descendant cells comprise said lectin-encoding polynucleotide, is produced; and verifying that said polynucleotide is expressed in said descendant cells, whereby a pesticidal lectin is produced.

30. The method of claim 29, further comprising the steps of growing said plant under conditions whereby cotton seed are produced; harvesting at least one cotton seed from said plant; and producing descendant generations of pest resistant cotton plants from said plant.

31. The method of claim 29, wherein said lectin is insecticidal.

32. The method of claim 29, wherein said lectin is fungicidal.

33. The method of claim 29, wherein said lectin is nematocidal.

34. The method of claim 29, wherein said polynucleotide is derived from barley.

35. The method of claim 29, wherein said polynucleotide is derived from nettle.

36. The method of claim 29, wherein said polynucleotide is derived from hevein.

37. The method of claim 30, wherein said lectin is insecticidal.

38. The method of claim 30, wherein said lectin is fungicidal.

39. The method of claim 30, wherein said lectin is nematocidal.

40. The method of claim 30, wherein said polynucleotide is derived from barley.

41. The method of claim 30, wherein said polynucleotide is derived from nettle.

42. The method of claim 30, wherein said polynucleotide is derived from hevein.

43. A method of killing a cotton pest comprising the steps of: obtaining a pesticidal lectin-encoding polynucleotide; transforming a cotton cell with said polynucleotide; culturing said cell under conditions whereby descendant cotton cells comprising said polynucleotide or a plant comprising said descendant cells are produced; verifying that said polynucleotide is expressed in said descendant cells, whereby a pesticidal lectin is produced; and allowing said descendant cells to be contacted by a cotton pest.

44. The method of claim 43, wherein said lectin is insecticidal.

45. The method of claim 43, wherein said lectin is fungicidal.

46. The method of claim 43, wherein said lectin is nematocidal.

47. The method of claim 43, wherein said polynucleotide is derived from barley.

48. The method of claim 43, wherein said polynucleotide is derived from nettle.

49. The method of claim 43, wherein said polynucleotide is derived from hevein.