WO2000068391A1 - Expression of a mannan binding domain to alter plant morphology - Google Patents

Abstract

Novel compositions and methods for their use are provided comprising recombinant nucleic acids coding for a mannan binding domain or functional portions thereof. The compositions can be synthesized or prepared by recombinant DNA technology. The compositions can be used to create transgenic plants expressing said mannan binding domain. The mannan binding domain can be expressed in a tissue specific manner and thereby modify plant properties including mannan-containing plant components.

Description

EXPRESSION OF A MANNAN BINDING DOMAIN TO ALTER PLANT MORPHOLOGY

INTRODUCTION

Field of the Invention

This invention relates to plants having altered properties as a result of expression of a intracellular mannan binding domain (MBD).

Background

Mannan is a β-l,4-linked heteropolysaccharide consisting of a backbone of β- linked mannose residues which carries substitutions in the form of other sugars or of acids. Mannan is found abundantly in seeds of leguminous plants such as lucerne (Medicago sativa) and in the beans of carob trees (Ceratonia siliqua), in the form of galactomannan. In the softwood of gymnosperms, O-acetylgalactoglucomannan is the major component of their hemicellulose fraction (Dekker, (1985) In T. Higuchi (ed.) Biosynthesis and biodegradation of wood components p.505-533). β-mannans are found in moderate amounts in certain plant secondary cell walls.

Biodegradation of β-mannans is carried out by β-mannanases (1,4-β-D-mannan mannanohydrolase; EC 3.2.1.78), which attack the backbone of the polysaccharide, yielding short-chain oligomannosides. These sugars can then be hydrolyzed further to mannose by the action of β-mannosidases (1,4-β-D-mannosidase; EC 3.2.1.25). Endo- β-mannanases have been purified and characterized from bacteria such as Bacillus pumilus (Akino et al, (1989) Appl. Env. Microbiol. 55:3178-3183), Caldocellum saccharolyticum (Lϋthie et al, (1991) Appl. Environ. Microbiol. 57:694-700) and Streptomyces spp. (Kusabe and Takahasi, (1988) Methods Enzymol. 160:611-614). The enzyme also has been isolated from fungi such as Polyporus versicolor (Johnson and Ross, (1990)), Thiclavia terrestris (Araujo and Ward, (1990) J. Appl. Bacteriol. 68:253- 261) and Trichoderma reesei (Ratto and Poutanen, (1988)). The heterologous cloning of mannanase-encoding genes has been reported for a Bacillus sp.(Akino et al, (1989) Appl. Env. Microbiol. 55:3178-3183) and for Caldocellum saccharolyticum (Ltithi et al, (1991) Appl. Environ. Microbiol. 57:694-700)). Another polysaccharide that is abundant in the plant cell walls is cellulose, and cellulose binding domains (CBDs) which have been reported to affect plant growth and morphology. The gram-negative bacterium Acetobacter xylinum has long been regarded as a model of cellulose synthesis, at least in part because cellulose microfibril synthesis and cell wall formation are separate (Ross et al (1991) Microbiological Reviews, 55:35-58). Polymerization and crystallization of cellulose are coupled processes in A. xylinum cellulose synthesis, therefore interference with crystallization results in acceleration of polymerization (Benziman et al (1980) Proc. Nat'l. Acada. Sci (USA) 77:6678-6682). Some cellulose-binding organic substances also can alter cell growth and cellulose-microfibril assembly in vivo. Direct dyes, carboxymethyl cellulose (CMC) and fluorescent brightening agents (FBAs, e.g. calcofluor white ST) prevent Acetobacter xylinum microfibril crystallization, thereby enhancing polymerization. These molecules bind to the polysaccharide chains immediately after their extrusion from the cell surface, preventing normal assembly of microfibrils and cell walls (Haigler, (1991) In: Biosynthesis and Biodegradation of Cellulose (Haigler and Weimer eds Marcel Dekker, Inc., N.Y.) pp. 99-124). CBD enhances cellulose synthase activity in A. xylinum by interference with the crystallization process (Shpigel et al ( 1998) Plant Physiol. 777:1185-1194). Modifications in cell shape were observed when red alga (Waaland and Waaland (1975) and root tips (Hughes and McCully (1975) Stain Technology 50:319-329) were grown in the presence of dyes. It is now evident that these molecules can bind to the cellulose chains immediately upon their extrusion from the cell surface of prokaryotes and eukaryotes (Haigler and Brown (1979) J Cell Biol. 83:70a); Benziman et al, (1980) supra; Brown et al (1982) Science 218:1141-1142) and prevent crystal-structure formation (Haigler and Chanzy (1988) J. Ultrastruct. Mol. Struct. Res. 98:299-311). The effect of CBD on cellulose structure as observed by electron microscopy is comparable to the effect of CMC (carboxy methyl cellulose) rather than to the effect of calcofluor (Shpigel et al (1998) supra; Haigler (1991) supra) in both cases the cellulose ribbon only splayed. Expression oϊcbd modulated the growth of transgenic tobacco and poplar plants. Biomass production was significantly higher in selected clones as compared with the control (Shoseyov et al (1998) (abstract) 8^th International Cell Meeting, John Innes Centre, Norwich l*-5* September 1998). It therefore is of interest to determine whether expression of mannan binding domains (MBDs) may affect secondary cell wall deposition and synthesis and consequently plant growth, morphology, mechanical properties, and pulping properties of wood and other mannan-containing plants and plant parts, as well as the digestibility of certain plants.

RELEVANT LITERATURE The primary sequences of 10 mannanases have been determined for enzymes from eubacteria (Braithwaite, et al (1995) Biochem. J. 305:1005-1010; Gibbs et al (1992) Appl. Environ. Microbiol. 55:3864-3867; Arcand et al (1993) Biochem. J. 290:857-863; Akino et al (1989) Appl. Environ. Microbiol. 55:3178-3183; Mendoza et al (1995) Biochim. Biophys. Acta 1243:552-55A) and anaerobic (Fanutti et al (1995) J. Biol. Chem. 270:29 14-29322; Milward-Sadler et al (1996) FEMS Microbiol. Lett. 141(2-3):IS3-ISS) and aerobic fungi (Stalbrand et al (1995) Appl. Environ. Microbiol. 57:1090-1097; Chήsigau et al (199A) Mol. Biol. Int. 55:917-925). Hydrophobic cluster analysis (HCA) and comparison of the primary structures of these enzymes have placed them into two distinct families, namely families 5 and 26 of glycosyl hydrolases (Henrissat & Bairoch (1993) Biochem. J. 293:781-788).

Membrane-bound mannan synthase were reported in Pinus sylvestris L. (Dalessandro et al (1986) Planta 169:56A-7A) and sycamore (Smith et al (1976) Bichimie 58:1195-221). Levitov et al (1995) Annals ofBotony 76:1-6 report on the involvement of endomannanase in the control of tomato seed germination under low temperature conditions.

SUMMARY OF THE INVENTION

Plants and plant components having altered mannan-containing structures as a result of exposure to a peptide composition comprising an endo-β-l,4-mannananse (mannanase) and/or the mannan binding domain of a mannanase, together with methods for their preparation and use, are provided. In the method, plants are provided with mannanase or a non-catalytic mannan binding domain (MBD), so that mannan- containing plant components are altered as compared to control plant components. Where the mannan is a part of a living plant, such as a transgenic organism, the altered structure or morphology can be associated, for example, with altered biomass, growth, yield, greater or less resistance to biodegradation, more or less digestible to ruminants, altered mannan content as compared to an organism of the same species. The alteration can be by contacting the structure with a composition comprising an MBD, or by expressing a composition comprising the MBD in the plant so that the physiology of the plant can be modified so as to produce a plant with altered properties including morphological characteristics and physical characteristics. The surface properties of organisms having mannan-containing cell walls also can be altered and/or the mannan in the cells walls can be used as a target for fusion protein or conjugates which contain a mannan binding domain and an antifungal agent or a mannan binding domain and a detectable label. Also included are nucleic acids which encode the polypeptide compositions and transformed plant cells containing the nucleic acids. The invention finds use for example in modulating the physiology of cellular components which contain mannan to increase digestibility, to facilitate extraction of plant materials from plant tissues and/or parts, for attaching a compound of interest to plant parts and/or to cellular components and for immobilizing and/or purifying the cellular components and/or the plant cells and/or tissues containing them.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 shows a restriction enzyme map of pCMan2 and pCMan4. Figure 1A: Restriction digests of pCMan4 (Lane 1) and pCMan2 (Lane 2) separated on a 0.9 % agarose gel. Restriction endonucleases used are indicated, λ Hind III DNA was used as size standard. Figure IB: Plasmid map ofpCMan2. The restriction map of the 4.3 kbp C. fimi genomic DNA insert of pCMan2 is shown. Not all the fragments could be mapped.

Figure 2 shows a schematic representation of the 6.8 kbp region of C. fimi genomic DNA containing the man26A open reading frame. The open reading frame is represented as a box, starting with the ATG codon at +1 and ending with the TGA codon at +3033. The genomic DNA inserts of the three plasmids, ρCMan4, pCMan2 and pCMan30 and their relative positions are represented by arrows. The arrowheads show the start and the end of the inserts, relative to the man26A open reading frame. Figure 3 shows the nucleotide sequence (SEQ ID NO:l) and deduced amino acid sequence (SEQ ID NO:2) oiman26A. Putative promoter sequences, identified based on similarities to other C. fimi promoters are underlined. The putative ribosome binding site is shown in bold. A possible transcription termination hairpin is overlined. The PCR primers, MBD11 (SEQ ID NO:4) and MBD12 (SEQ ID NO:3), used for subcloning of mbd_III2 are indicated by arrows. ♦: Predicted signal peptide processing site, ♦: experimentally identified N-terminus of Man26A, Υ: predicted C-terminus of Man26A catalytic domain, *: N-terminus of Mannan Binding Domain, ♦: N-terminus of 29 kDa SLH domain fragment, •: N-terminus of 21 kDa SLH domain fragment and C-terminus of Mannan Binding Domain. Figure 4 shows the results from a protein sequence similarity search with

Man26A. Each bar represents an amino acid sequence which is similar to the sequence ofMan26A. The position of each bar is aligned for the homology with Man26 A. The alignment score is color-coded and the identities and alignment lengths are indicated for each protein sequence. The numbers in or beside the bars are the sequence Ids. (Altschul et al. , (1997) Nucleic Acids Res. 25:3389-3402).

Figure 5 shows the alignment of the first 460 amino acids from C. fimi mannanase (SEQ ID NO: 5), Cf Man26A with the entire sequence of the Pseudomonas fluorescens mannanase (SEQ ID NO:6), PF ManA. Identical residues are highlighted. The two family 26 proteins share 42 % identity. For Pf ManA, Glu 212 and Glu 320 were experimentally determined to be the catalytic residues acting as acid/base catalyst and nucleophile, respectively. In Man26A these residues are conserved and correspond to Glu 225 and Glu 332. The catalytic residues are underlined. Alignment was performed with Clustal W (Tompson et al, 1994). Figure 6 shows affinity gel electrophoresis (AFGE) of protease treated

Man26A. (Lanes 1 and 2), BSA (Lane 3) and intact Man26A (Lanes 4 and 5) without (Figure 12A) or with 1 % (Figure 12B) macromolecular affinity ligand, i.e., azo-carob galactomannan. To detect mannanase activity the gel was incubated in excess potassium-phosphate buffer (pH 7.0) at 37°C until clearings were visible. Protein bands were then visualized by Coomassie blue staining.

Figure 7 shows affinity gel electrophoresis of CBD_CenD (Lane 1), Man26A (Lane 2), MBD_{1] 12} (Lane 3), and Man2A (Lane 4). The concentrations of macromolecular affinity ligand, i.e., locust bean gum, are 1 % (Figure 13 A), 5.0 x 10^{" 4} % (Figure 13B) and 2.5 x 10^"3 % (Figure 13C). Figure 8 shows the nucleotide sequence (SEQ ID NO: 7) and deduced amino acid sequence (SEQ ID NO:8) of the Mannan Binding Domain of Man26A. The numbering corresponds to the numbering of the sequences in Figure 3. BRIEF DESCRIPTION OF THE SPECIFIC EMBODIMENTS

Compositions are provided which include a novel modular mannanase, and a non-catalytic binding domain obtainable from the mannanase. Also provided are nucleic acids which encode the mannanase and its non-catalytic binding domain, as well as fusion proteins and conjugates comprising the binding domain and nucleic acids encoding them, in which the non-catalytic binding domain is joined to a second molecule which can be a heterologous protein or a non-proteinaceous molecule. The second molecule can be joined to the non-catalytic binding domain chemically or made by recombinant means. In some instances, more than one molecule can be joined to the non-catalytic binding domain; the molecules that are joined can be the same or different. The binding domain, while lacking in the hydrolytic enzymatic activity ofa glycosyl hydrolase, retains the substrate binding activity of the enzyme. The terms "protein", "polypeptide" and "peptide" are used interchangeably throughout the specification and claims. These terms also encompass glycosylated proteins, i.e., glycoproteins. The term "mannan binding protein" refers to any protein which specifically binds to mannan and can be a region or portion ofa larger protein which binds specifically to mannan. The mannan binding domain (MBD) may be a part or portion of a mannanase or a non-catalytic mannan binding protein.

Methods for the preparation of mannanase binding domains are also provided, where the mannanase binding domain can be any mannanase binding domain ofa glycosyl hydrolase, a binding domain of a mannanase binding protein or a protein designed and engineered to be capable of binding to a mannan, particularly polymeric mannan or mannan containing structures such as cellular components, for example, cell walls and seed coats. By a mannanase is intended a glycosyl hydrolase which contains an amino acid sequence which binds specifically to mannan, in particular to the backbone ofa β-l,4-mannanase polymer rather than to terminal sugars of the polymer. By a mannan is intended a polymer which is a hemicellulose constituent with a β-1,4- linked backbone of mannose or of mannose and glucose residues. Other molecules, such as acetyl and galactosyl, may be attached to the sugars of the backbone. Included within the definition ofa mannan are glucomannans and galactomannans. The mannan can be in any form, soluble, insoluble, amorphous, crystalline, isolated or part ofa structure such as a component of an organism. The mannanase and/or the mannan binding domain can be used to modify mannan containing structures, including cellular components of higher plants. By "altered structure or morphology" is intended any microscopic or macroscopic change in the structure or morphology of mannan- containing tissues when compared to untreated mannan.

The subject invention offers several advantages over existing techniques used to alter plant morphology and physiology. For example, mannan deposition occurs primarily in the secondary cell wall, and therefore use of MBD has less drastic morphological effects than does use of cellulose binding domain (CBD). Modification of cellulose does not affect the germination rate of seeds, such as tomato seeds, whereas modification of the mannan in the endosperm cell wall does. The mannan binding domain provides a means for attaching a polypeptide or a non-protein molecule of interest to any plant structures which contain mannan, including seeds and roots. Due to the prevalence of mannan in a variety of plant structures, use of the mannan binding domain therefore has wide application, and avoids the need for preparing constructs that are tailor-made for particular plant structures and/or components. Furthermore, a variety of mannanases bind specifically to mannans and can be used as the source of MBD for the subject invention. Thus, compositions comprising the binding domain can be used to impart a desirable new physical property to a mannan-containing plant surface or part by transgenic expression of a mannanase or its non-catalytic binding domain in an organism of interest, or by contacting the mannan-containing composition with a mannanase or the non-catalytic binding domain.

Novel polypeptide compositions can include those having the following formula:

MBD - MR - X wherein: MBD can be either the N-terminal or the C-terminal region, or both, ofa mannanase and its characterized as having a sufficient amount of consecutive sequence of amino acids from the substrate binding region of a polysaccharidase to provide for high affinity binding to a substrate where the substrate can be a natural substrate of the mannanase or a different substrate. MR is the middle region, and can be a bound; short linking group of from 2 to

30 carbon atoms, or have from about 2 to about 20 amino acids. The region can include an amino acid sequence providing for specific cleavage of the fusion protein, usually a sequence corresponding to that recognized by a proteolytic enzyme of high specificities such as an IgAl protease or Factor Xa; and

X can be either the N-terminal or the C-terminal region and can be any molecule, and includes any peptide of interest or any non-protein of interest that can be linked to the MBD and/or the MR by any means, including recombinantly or by chemical means. When X is a peptide of interest, X is characterized as having up to the entire sequence of a peptide of interest, and can be an enzyme, a hormone, an immunoglobulin, a protein dye, and the like.

Novel conjugate compositions include those having the formula: MBD-Z or

MBD-MR-Z wherein: the mannan binding domain (MBD) is characterized as (1) obtainable from a mannanase; and (2) capable of binding to mannan. The MBD is at least as large as the minimum number of amino acids in a sequence required to bind a mannan; MBD-MR is defined as above; and Z is a moiety that is attached to the polysaccharide binding domain and can be either a peptide or a non-peptide moiety. Z indicates only the moiety, not the stoichiometry of the moiety. The stoichiometry can be variable.

A variety of mannan substrates are of interest, both soluble, exemplified by locust bean gum, and insoluble, exemplified by ivory nut mannan. The MBD from C. fini or related enzymes can be used in the subject invention. By related enzymes is intended enzymes which share some amino acid sequence identity with the C. fini mannanase binding domain. Relatedness between these groups of proteins can be calculated using the NBRF ALIGN program BLAST search. As applied to these polypeptides and peptides thereof, the terms "substantial sequence identity" or "homology" or "homologous" mean that two amino acid sequences, when optimally aligned, such as by the programs GAP or BESTFIT using default gap penalties, share at least 80 percent sequence identity, preferably at least 90 percent sequence identity, more preferably at least 95 percent sequence identity or more. "Percentage amino acid identity" or "percentage amino acid sequence identity" refers to a comparison of the amino acids of two polypeptides which, when optimally aligned, have approximately the designated percentage of the same amino acids. Preferably, residue positions which are not identical differ by conservative amino acid substitutions. For example, the substitution of amino acids having similar chemical properties such as charge or polarity are not likely to substantially effect the properties of a protein. Examples include glutamine for asparagine or glutamic acid for aspartic acid. The term mannanase refers not only to the amino acid sequences disclosed herein, but also to other proteins that are allelic or species variants of these amino acid sequences. It is also understood that these terms include nonnatural mutations introduced by deliberate mutation using recombinant technology such as single site mutation or by excising short sections of DNA encoding mannanase or by substituting new amino acids or adding new amino acids. Such minor alterations substantially maintain the immunoidentity of the original molecule and/or its biological activity. The biological properties of the altered proteins can be determined by expressing the protein in an appropriate cell line and by determining the ability of the protein to bind to mannan. Particular protein modifications considered minor include substitution of amino acids of similar chemical properties, e.g., glutamic acid for aspartic acid or glutamine for asparagine.

In other embodiments the present invention provides isolated nucleic acid molecules which encode a mannanase and/or a MBD. The term "isolated" as applied to nucleic acid molecules means those which are separated from their native environment, and preferably free of non-mannanase-related DNA or coding sequences with which they are naturally associated. The term "nucleic acids", as used herein, refers to either DNA or RNA. "Nucleic acid molecule" or "polynucleotide sequence" refers to a single- or double-stranded polymer of deoxyribonucleotide or ribonucleotide bases read from the 5' to the 3' end. The nucleic acid molecules of the invention, whether RNA, cDNA, or genomic DNA, may be isolated from natural sources or may be prepared in vitro. The nucleic acids may be present in transformed or transfected whole cells, in a transformed or transfected cell lysate, or in a partially purified or substantially pure form.

The nucleic acid molecules of the invention are typically identical to or show substantial sequence identity or homology (determined as described herein) a nucleic acid which encodes a mannanase or MBD or the complement thereof. Of particular interest are nucleic acid molecules that show substantial sequence to identify a homology to a nucleic acid having a sequence shown in SEQ ID NO:7. The nucleic acid molecules include those which are equivalent to native or allelic sequences due to the degeneracy of the genetic code as well as sequences which are introduced to provide codon preference in a specific host cell. Nucleic acids encoding a mannanase will typically hybridize to the nucleic acid sequences that encode a mannanase under stringent hybridization conditions. Less stringent hybridization conditions may also be selected. Generally, stringent conditions are selected to be about 5°C lower than the thermal melting point (T_m ) for the specific sequence at a defined ionic strength and pH. The T_m is the temperature (under defined ionic strength and pH) at which 50% of the target sequence hybridizes to a perfectly matched probe. Typically, stringent conditions will be those in which the salt concentration is at least about 0.02 M at pH 7 and the temperature is at least about 60°C. As other factors may significantly affect the stringency of hybridization, including, among others, base composition and size of the complementary strands, the presence of organic solvents and the extent of base mismatching, the combination of parameters is more important than the absolute measure of any one. Thus, the phrase "selectively hybridizing to" refers to a nucleic acid probe that hybridizes, duplexes or binds preferentially to a particular target DNA or RNA sequence when the target sequence is present in a preparation of total cellular DNA or RNA. "Complementary" or "target" nucleic acid sequence refers to a nucleic acid sequence which selectively hybridizes to a nucleic acid probe. For discussions of nucleic acid probe design and annealing conditions, see, for example, Sambrook et al, Molecular Cloning: A Laboratory Manual (2^nd ed.), Vols. 1-3, Cold Spring Harbor Laboratory, (1989) or Current Protocols in Molecular Biology, F. Ausubel et al, ed., Greene Publishing and Wiley-Interscience, New York (1987), each of which is incorporated herein by reference. Techniques for manipulation of nucleic acids encoding mannanase such as subcloning nucleic acid sequences encoding polypeptides into expression vectors, labelling probes, DNA hybridization, and the like are described generally in Sambrook, supra.

The compositions can be prepared by transforming into a host cell a DNA construct comprising DNA encoding at least a functional portion of the mannanase or the MBD. Alternatively, a DNA sequence encoding a heterologous protein can be ligated to the MBD DNA sequence to produce a fused gene. The fused gene, the MBD DNA sequence alone, or the mannanase gene, can be expressed in a host cell, either a eukaryotic or a prokaryotic cell. Expressed and isolated fusion proteins, mannanases, and MBD's optionally can be conjugated to chemical moieties. Many of the enzymes which bind to mannans comprise discreet catalytic and binding domains. However, the invention is not limited to the use of binding domains from such enzymes; those in which the binding and catalytic domains are one and the same also can be used. The techniques used in isolating mannanase genes and MBDs are known in the art, including synthesis, isolation from genomic DNA, preparation from cDNA, or combinations thereof. For example, DNA can be isolated from a genomic or cDNA library using labeled oligonucleotide probes having sequences complementary to the sequence disclosed herein as SEQ ID NO:7. Full-length probes may be used, or oligonucleotide probes also may be generated by comparison of the sequences of SEQ ID NO:7. Such probes can be used directly in hybridization assays to isolate DNA encoding mannanase. Alternatively, probes can be designed for use in amplification techniques such as PCR (Mullis, et al, US Patent Nos. 4,683,195 and 4,683,202, incorporated herein by reference), and DNA encoding mannanase be isolated by using methods such as PCR. Nucleic acid probes may be DNA or RNA fragments. DNA fragments can be prepared, for example, by digesting plasmid DNA, or by use of PCR, or synthesized by either the phosphoramidite method described by Beaucage and Carruthers, (1981) Tetrahedron Lett. 22:1859-1862, or by the triester method according to Matteucci et al, (1981) J. Am. Chem. Soc.l03:3l 5, both incorporated herein by reference. A double stranded fragment may then be obtained, if desired, by annealing the chemically synthesized single strands together under appropriate conditions or by synthesizing the complementary strand using DNA polymerase with an appropriate primer sequence. Where a specific sequence for a nucleic acid probe is given, it is understood that the complementary strand is also identified and included. The complementary strand works equally well in situations where the target is a double- stranded nucleic acid.

To prepare a cDNA library, mRNA is isolated from an organism which expresses a mannanase such as C. fimi. Other organisms include Streptomyces lividans, Bacillus spp, Pseudomonas fluorescens, Caldocellum saccharolyticum, Trichoderma reesei, Piromyces, Aspergillus aculeatus, Aspergillus niger, and Penicillium purporgenum. cDNA is prepared from the mRNA and ligated into a recombinant vector. The vector is transfected into a recombinant host for propagation, screening and cloning. Methods for making and screening cDNA libraries are well known. See Gubler and Hoffman (1983) Gene 25:263-269 and Sambrook, et al, supra.

For a genomic library, the DNA is extracted and either mechanically sheared or enzymatically digested to yield fragments of about 12-20kb. The fragments are then separated by gradient centrifugation from undesired sizes and are constructed in bacteriophage lambda vectors. These vectors and phage are packaged in vitro, as described in Sambrook, et al. Recombinant phage are analyzed by plaque hybridization as described in, for example, Benton and Davis (1977) Science, 196: 180-182. Colony hybridization is carried out as generally described in, for example, Grunstein, et al, (1975) Proc. Natl. Acad. Sci. USA., 72:3961-3965. DNA encoding a mannanase and/or a MBD is identified in either cDNA or genomic libraries by its ability to hybridize with nucleic acid probes, for example on Southern blots, and these DNA regions are isolated by standard methods familiar to those of skill in the art. See Sambrook, et al., supra. Various methods of amplifying target sequences, such as the polymerase chain reaction, can also be used to prepare nucleic acids encoding mannanse. PCR technology is used to amplify such nucleic acid sequences directly from mRNA, from cDNA, and from genomic libraries or cDNA libraries. The isolated sequences encoding mannanase may also can be used as templates for PCR amplification. In PCR techniques, oligonucleotide primers complementary to the two 3' borders of the DNA region to be amplified are synthesized. The polymerase chain reaction is then carried out using the two primers. See PCR Protocols: A Guide to Methods and Applications. Innis, M., Gelfand, D., Sninsky, J. and White, T., eds., Academic Press, San Diego (1990). Primers can be selected to amplify the entire regions encoding a full-length mannanase protein or to amplify smaller DNA segments as desired. PCR can be used in a variety of protocols to isolate cDNAs encoding a mannanase. In these protocols, appropriate primers and probes for amplifying DNA encoding mannanase are generated from analysis of the DNA sequences listed herein. Once such regions are PCR-amplified, they can be sequenced and oligonucleotide probes can be prepared from the sequences obtained. These probes can then be used to isolate DNAs encoding mannanase. Mannanase can be isolated from a variety of different organisms using this procedure. Oligonucleotides for use as probes are chemically synthesized according to the solid phase phosphoramidite triester method first described by Beaucage and Carruthers, Tetrahedron Lett. (1981) 22:1859-1862 , using an automated synthesizer, e.g., as described in Needham-VanDevanter et al, (1984) Nucleic Acids Res.72:6159-6168. Purification of oligonucleotides is by either native acrylamide gel electrophoresis or by anion-exchange HPLC as described in Pearson and Regnier (1983) J. Chrom., 255:137-149. The sequence of the synthetic oligonucleotide can be verified using the chemical degradation method of Maxam and Gilbert (1984) Meth. Enzymol., 55:499-560. Other methods known to those of skill in the art also can be used to isolate DNA molecules encoding mannanase and/or MBD. See Sambrook, et al, supra.

The present invention includes nucleotide sequences that have substantial sequence identity or homology to the mannanase nucleotide sequences described in SEQ ID NO:l. For substantial sequence identity or homology the polynucleotide comprises a sequence that has at least 80 percent sequence identity, preferably at least 90 percent sequence identity, and more preferably at least 95 percent sequence identity. The comparison is made to a reference sequence over a comparison window of at least 20 nucleotide positions, frequently over a window of at least 25-50 nucleotides, wherein the percentage of sequence identity is calculated by comparing the reference sequence to the polynucleotide sequence, which may include deletions or additions which total 20 percent or less of the reference sequence over the window of comparison. The reference sequence may be a subset of a larger sequence, for example, as a segment of the C. fimi mannanase sequences described herein. Optimal alignment of sequences for aligning a comparison window may be conducted according to the local homology algorithm of Smith and Waterman (1981) Adv. Appl. Math. 2:482, by the homology alignment algorithm of Needleman and Wunsch (1970) J. Mol. Biol. 48:AA3, by the search for similarity method of Pearson and Lipman (1988) Proc. Natl. Acad. Sci. (USA) 55:2444, or by computerized implementations of these algorithms (GAP, BESTFIT, FASTA, and TFASTA in the Wisconsin Genetics Software Package Release 7.0, Genetics Computer Group, 575 Science Dr., Madison, WI).

Once DNA encoding mannanase, MBD or a homologous sequence is isolated and cloned, a mannanase or a homologous protein can be expressed in a recombinantly engineered plant cells for the purpose of altering the structure or morphology or other characteristics of the transgenic cell itself (or progeny thereof). It is a theory of the invention that MBD may interfere with the synthesis and/or assembly of cell wall components and thereby modify growth and/or fiber characteristics in higher plants. These changes can result in changes in biomass, yield, growth rate, size, etc. Particular tissues also can be affected, in particular roots, leaves, fruit, and seed either directly or by the use of appropriate promoters.

Numerous expression systems are available for expression of DNA encoding a mannanase or MBD. The expression of natural or synthetic nucleic acids encoding a mannanase or MBD is typically achieved by operably linking the DNA to a promoter (which is either constitutive or inducible) within an expression vector. By expression vector is meant a DNA molecule, linear or circular, that comprises a segment encoding a mannanase or polypeptide of interest, operably linked to additional segments that provide for its transcription. Such additional segments include promoter and terminator sequences. An expression vector also may include one or more origins of replication, one or more selectable markers, an enhancer, a polyadenylation signal, etc. Expression vectors generally are derived from plasmid or viral DNA, and can contain elements of both. The term "operably linked" indicates that the segments are arranged so that they function in concert for their intended purposes, for example, transcription initiates in the promoter and proceeds through the coding segment to the terminator. See Sambrook et al, supra.

In addition to the gene(s) to be expressed in the proposed transgenic plants, several nucleotide sequences are required for the insertion and/or expression of the structural gene(s). The required sequences for insertion of the gene(s) into the plant to be transformed vary with the method of transformation chosen. Many plasmids and vectors effective for use in the transformation of plants are commercially available.

The choice of a promoter will depend in part upon whether constitutive or inducible expression is desired and whether it is desirable to produce the mannanase or MBD at a particular stage of plant development and/or in a particular tissue.

Expression can be targeted to a particular location within a host plant such as seed, leaves, fruits, flowers, and roots. Still other promoters have been described that are inducible by chemical, mechanical or other types of stimulation. Many of these promoters are available commercially, usually as part of a plasmid/vector. Constitutive promoters have been isolated from both plants and viruses. The most widely-used constitutive promoters are derived from the cauliflower mosaic virus (CaMV). The CaMV 35S and CaMV 19S promoters are described in USPN 5,352,605 and related patent USPN 5,530,196. USPN 5,196,525 describes increased transcription efficiency of CaMV 35S and other promoters by incorporating duplicates of a transcription activating sequence of the promoter ("double 35S"). USPN 5,491,288 describes a double CaMV 35S promoter and an H4 histone promoter from the Arabidopsis thaliana plant sequentially linked, or multiple H4 histon promoters so linked, which provides even higher levels of expression. Other plant-derived, constitutive promoters include AHAS (ALS) promoters isolated from corn (USPN 5,750,866) and an ALS promoter (ALS3) derived from Brassica napus (USPN 5,659,026). Other constitutive promoters include the rice actin 1 (Act 1) gene promoter (e.g., Wanggen Zhang et al (1991) The Plant Cell 3:1155-1165) and the corn ubiquitin 1 gene (Ubi 1) promoter (e.g., Christensen et al (1992) Plant Mol. Biol. 18:675-689). Many promoters that direct expression in specific plant tissue or parts have been described, including those for specific expression in fruit, sink organ, vasculature, phloem, epidermis roots, seeds and flowers. Fruit-Specific Promoters are described in USPN 4,943,674 and Deekman et al (1988) EMBO. J. 7:3315-5330. Sink-Organ-Specific Promoters are described in USPN 5,723,757 (promoter for the patatin gene from potatoes). Promoters directing expression to subparts found throughout a plant include USPN 5,495,007 (a phloem protein gene promoter derived from squash that directs expression of a marker gene specifically in the phloem of transgenic tobacco plants); USPN 5,391,725 (a promoter from the cytosolic glutamine synthase gene of the pea, which directs phloem/vasculature-specific marker gene expression in transgenic tobacco plants and seedlings and transgenic alfalfa plants): vasculature-specific expression is found in the leaves, roots, stems and cotyledons of the transgenic plants); USPN 5,646,333 (a promoter ofa lectin gene that directs expression ofa linked heterologous marker gene in the epidermis, as well as somatic embryos and growing shoot tips, of transgenic alfalfa plants). Root-Specific Promoters: USPN 5,635,618 (a promoter from a corn α- tubulin gene which directs expression of a heterologous marker gene preferentially in roots, as well as developing root tips and pollen, in transformed tobacco plants); USPN 5,633,363 (a promoter derived from a corn gene which is preferentially expressed in roots). Seed-Specific Promoters: USPN 5,608,152 (a napin gene promoter derived from a B. napus which directs the expression of a heterologous gene in seeds, but not leaves); USPN 5,677,474 (a barley α-amylase gene promoter). Inducible Promoters: several promoters that direct expression of linked genes in response to external stimulation, e.g., chemically inducible or inducible by mechanical stimulation, such as wounding or pest infestation, have been described and include Chemical Induced Promoters: USPN 5,589,614 (a promoter from a corn glutathione-S- transferase (GST) gene that promotes gene expresses in response to exposure to specific chemicals); USPN 5,608,143 (promoters of genes induced by a chemical in corn roots (4), petunia roots (1) and tobacco roots). Pest or Wound-Inducible Promoters:

USPN 5,689,056 (a 3-hydroxy-3-methylglutaryl CoA reductase (HMGR), isolated from tomatoes, which is expressed in response to various pest infestations and wounding of plant tissue in hypocotyledons, trichomes (leaf surface hairs) and pollen.; USPN 5,684,239 (promoter from a potato proteinase inhibitor gene (PIN II) and the 5' intron of the rice actin 1 (Act 1) gene which is expressed in response to mechanical wounding of shoots); USPN 5,428,146 (a promoter (wun 1) isolated from a gene in potatoes; USPN 5,677,175; and USPN 5,750,399 describes promoters from genes expressed in response to fungal infestation. Growth-Stage-Specific Promoters: disclose promoters that regulate expression in seeds and in immature or mature fruit (USPN 5,753,475 and USPN 5,633,440, respectively). USPN 5,177,011 (an elongation factor (EF-lά) gene promoter, derived from tomato plants, which directs expression of a marker gene in the shoot tips and root tips of transgenic tomato and tobacco plants); USPN 5,589,583 (a promoter from a gene expressed primarily in the meristematic tissue of the Arabidopsis thaliana plant. This promoter directed the expression of heterologous genes in the flowers and root and stem apical meristems of transgenic tobacco and Brassica napus plants). Other promoters from genes which have a differential pattern of expression in a specific tissue can be identified by differential screening of a tissue of interest, using methods described for example in USPN 4,943,674 and EP-A 0255378. The regulatory regions may be homologous (derived from the plant host species) or heterologous (derived from source foreign to the plant host species), to the plant host or a synthetic DNA sequence. The term "homologous" includes both indigenous and endogenous sequences. In order to joint the promoter(s) to the structural gene, the non-coding 5' region upstream from the structural gene may be removed by endonuclease restriction. Alternatively, where a convenient restriction site is present near the 5' terminus of the structural gene, the gene can be restricted and an adapter employed for linking the structural gene to a promoter region, where the adapter provides for any lost nucleotides of the structural gene. The termination region can be derived from the 3 '-region of the gene from which the initiation region was obtained or from a different gene. The termination region can be derived from a plant gene, particularly from the same plant gene used as a source of sequences to initiate transcription and translation. Other 3'-regions include the tobacco ribulose bisphosphate carboxylase small subunit (SSU) termination region; a gene associated with the Ti-plasmid such as the octopine synthase termination region; the tml termination region; and other 3 '-regions known to those skilled in the art.

Structures of interest for modification include the secondary cell wall. Therefore, it is desirable that the expression product of the MBD transgene be translocated to the secondary cell wall. Accordingly, DNA coding for a translocation or transit peptide optionally including a processing signal, recognized by the plant host can be included in the construct. The transit peptide, also known as a leader sequence, and processing signal may be derived from gene encoding any plant protein which is expressed in the cytoplasm and translocated to the secondary cell wall. For some applications, the leader sequence can be combined in a DNA construct with a seed specific promoter, for example to increase expression in seed, and for others, the leader sequence can be provided under the regulatory control of a more constitutive transcription initiation region. By "transit peptide" is meant a sequence capable of translocating a peptide joined to the transit peptide to a particular organelle. For the most part, the transit peptide is from one plant, but is generally recognized by other plants. Thus, the DNA encoding the transit peptide may be native to or heterologous to the ultimate host in which the chimeric gene is introduced. DNA encoding transit peptides may come from soybean, corn, petunias, tobacco, Brassica, tomato, wheat, pea and the like. The DNA encoding the transit peptide may be the complete transit-peptide- encoding sequence lacking from about 1 to 10 codons, or a portion ofa codon, from the 3' terminus. In addition, one or more changes may be made in the nature of mutations, deletions or insertion in the transit peptide and processing signal, where such change may provide for convenience in construction by providing for a convenient restriction site, or removing an inconvenient restriction site. The mutations may be conservative or non-conservative, so that the transit peptide may be the same or different from the wild-type transit peptide. An additional peptide segment between the transit peptide and the MBD may be useful. Such peptide may be the mature (post-processing) amino- terminal portion of the structural peptide or any other peptide providing the appropriate structure features recognized and needed by the plastid translocation system. When post processing amino acids from the SSU leader sequence are included, it is preferred that no more than 50, preferably no more than 40, and most preferably no more than 30 post processing amino acids are present at the N-terminal of the desired MBD peptide. In developing the expression cassette, the various fragments comprising the regulatory regions and open reading frame may be subjected to different processing conditions, such as ligation, restriction, resection, in vitro mutagenesis, primer repair, use of linkers and adapters, and the like. Thus, nucleotide transitions, transversions, insertions, deletions, or the like, may be performed on the DNA which is employed in the regulatory regions and/or open reading frames.

During the construction of the expression cassette, the various fragments of the DNA will usually be cloned in an appropriate cloning vector, which allows for amplification of the DNA, modification of the DNA or manipulation by joining or removing of the sequences, linkers, or the like. Normally, the vectors will be capable of replication in at least a relatively high copy number in E. coli. A number of vectors are readily available for cloning, including such vectors as pBR322, pUC series, Ml 3 series, etc. The clonging vector will have one or more markers which provide for selection or transformants. The markers will normally provide for resistance to cytotoxic agents such as antibiotics, heavy metals, toxins, or the like. By appropriate restriction of the vector and cassette, and as appropriate, modification of the ends, by chewing back or filling in overhangs, to provide for blunt ends, by addition of linkers, by tailing, complementary ends can be provided for ligation and joining of the vector to the expression cassette or component thereof. After each manipulation of the DNA in the development of the cassette, the plasmid will be cloned and isolated and, as required, the particular cassette component analyzed as to its sequence to ensure that the proper sequence has been obtained. Depending upon the nature of the manipulation, the desired sequence may be excised from the plasmid and introduced into a different vector or the plasmid may be restricted and the expression cassette component manipulated, as appropriate.

The manner of transformation of E. coli with the various DNA constructs (plasmids and viruses) for cloning is not critical to this invention. Conjugation, transduction, transfection or transformation, for example, calcium phosphate mediated transformation, may be employed.

In addition to the expression cassette, depending upon the manner of introduction of the expression cassette into the plant cell, other DNA sequences may be required. For example, when using the Ti- or Ri-plasmid for transformation of plant cells, as described below, at least the right border and frequently both the right and left borders of the T-DNA of the Ti- or R-plasmids will be joined as flanking regions to the expression cassette. The use of T-DNA for transformation of plant cells has received extensive study and is amply described in Genetic Engineering, Principles and Methods (1984) Vol. 6, (Eds. Setlow and Hollaender) pp. 253-278 (Plenum, NY); A Hoekema, in: The Binary Plant Vector System (1985) Offsetdrukkerij Ranters, 8.V. Alblasserdam.

Alternatively, to enhance integration into the plant genome, terminal repeats of transposons may be used as borders in conjunction with a transposase. In this situation, expression of the transposase should be inducible, so that once the expression cassette is integrated into the genome, it should be relatively stably integrated and avoid hopping.

The expression cassette will normally be joined to a marker for selection in plant cells. Conveniently, the marker may be resistant to a biocide, particularly an antibiotic, such as Kanamycin, G418, Bleomycin, Hygromycin, Chloramphenicol, or the like. The particular marker employed will be one which will allow for selection of transformed plant cells as compared to plant cells lacking the DNA which has been introduced.

A variety of techniques are available for the introduction of DNA into a plant cell host. These techniques include transformation with Ti-DNA employing A. tumefaciens or A. rhizogenes as the transforming agent, protoplast fusion, injection, electroporation, DNA particle bombardment, and the like. For transformation with Agrobacterium, plasmids can be prepared in E. coli which plasmids contain DNA homologous with the Ti-plasmid, particularly T-DNA. The plasmid may be capable of replication in Agrobacterium, by inclusing of a broad spectrum prokaryotic replication system, for example RK290, if it is desired to retain the expression cassette on an independent plasmid rather than having it integrated into the Ti-plasmid. By means of a helper plasmid, the expression cassette may be transferred to the A. tumefaciens and the resulting transformed organism used for transforming plant cells. Conveniently, explants may be cultivated with the A. tumefaciens ox A. rhizogenes to allow for transfer of the expression cassette to the plant cells, and the plant cells dispersed in an appropriate selection medium. The Agrobacterium host will contain a plasmid having the vir genes necessary for transfer.

In other embodiments, various alternative methods for introducing recombinant nucleic acid constructs into plants and plant cells may also be utilized. These other methods are particularly useful where the target is a monocotyledonous plant or plant cell. Alternative gene transfer and transformation methods include, but are not limited to, protoplast transformation through calcium-polyethylene glycol (PEG) - or electroporation-mediated uptake of naked DNA (see Paszkowski et al (1984) EMBOJ 5:2717-2722, Potyrkus et al (1985) Molec. Gen. Genet 799:169-177; Fromm et al (1985) Proc. Nat. Acad. Sci USA 52:5824-5828; and Shimamoto (1989) Nature 555:274-276) and electroporation of plant tissue (D'Halluin et al (1992) Plant Cell 4:1495-1505). Additional methods for plant cell transformation include microinjection, silicon carbide mediated DNA uptake (Kaeppler et al (1990) Plant Cell Reporter 9:415- 418), and microprojectile bombardment (see Klein et al (1988) Proc. Nat. Acad. Sci. USA 55:4305-4309; and Gordon-Kamm et al (1990) Plant Cell 2:603-618.

After transformation, the cell tissue (for example protoplasts, explants or cotyledons) is transferred to a regeneration mediu, such as Murashige-Skoog (MS) medium for plant tissue and cell culture, for formation ofa callus. Cells which have been transformed may be grown into plants in accordance with conventional ways. See, for example, McCormick et al (1986) Plant Cell Reports 5:81-84. The transformed plants may then be analyzed to determine whether the desired gene product is still being produced in all or a portion of the plant cells. After expression of the desired product has been demonstrated in the plant, the plant can be grown, and either pollinated with the same transformed strain or different strains and the resulting hybrid having the desired phenotypic characteristic identified. Two or more generations may be grown to ensure that the subject phenotypic characteristic is stably maintained and inherited.

To identify the desired phenotypic characteristic, transgenic plants which contain and express a given MBD transgene are compared to control plants. By MBD transgene is intended a transgene encoding a peptide comprising an MBD. Preferably, transgenic plants are selected by measurement of MBD activity in leaf, fruit and/or root. The MBD activity may be periodically measured from various stages of growth through senescence and compared to that of control plants. Plants or plant parts having increased or decreased MBD activity compared to controls at one or more periods are selected. The activity can be compared to one or more other traits including SPS type, transcription initiation type, translation mitiation type, termination region type, transgene copy number, transgene insertion and placement.

When evaluating a phenotypic characteristic associated with expression of the MBD transgene, the transgenic plants and control plants are preferably grown under growth chamber, greenhouse, open top chamber, and/or field conditions. Identification of a particular phenotypic trait and comparison to controls is based on routine statistical analysis and scoring. Statistical differences between plants lines can be assessed by comparing MBD activity between plant lines within each tissue type expressing MBD. Expression and activity are compared to growth, development and yield parameters which include plant part morphology, color, number, size, dimensions, dry and wet weight, ripening, above and below-ground biomass ratios, and timing, rates and duration of various stages of growth through senescence, including vegetative growth, fruiting, flowering, and soluble solid content including sucrose, glucose, fructose and startch levels. To identify transgenic plants having other traits, the plants can be tested for photosynthetic and metabolic activity, as well as end-product synthesis. For example, material isolated from transgenic plant cells and plant parts such as leaf, fruit and root are measured for end-products such as starch, sucrose, glucose, fructose, sugar alcohols, and glycine and serine from photorespiratory metabolism following standard protocols. Sweetness based on sugar content, particularly fructose, can be tested as well. For some plants, it may be necessary to modify growth conditions to observe the phenotypic effect. As an example, oxygen, carbon dioxide and light can be controlled and measured in an open gas chamber system, and carbon partitioning measured by C¹⁴ labeling of carbon dioxide or other metabolic substrates. Carbon partitioning also can be determined in extracts from fruit, leaf and/or root by chromatographic techniques or by Brix using a sugar refractometer. These characteristic also can be compared against or induced by growth conditions which vary gas exchange parameters, light quality and quantity, temperature, substrate and moisture content between lines within each type of growing condition. Morphological changes can be monitored as follows. Light microscopy of plant material cross sections, or alternatively sections stained with white fluorescent brightner (0.1% (w/v) calcofluor, in 0.1 M K3PO3), can be used to reveal cell-wall components under a fluorescent light microscope. Methods to determine changes in fiber characteristics, changes in mannan content and other plant polysaccharides are included and referenced in: "The Growing Plant Cell Wall:

Chemical and Metabolic analysis. (1988) Longman Scientific & Technical, UK. ISBN 0-582-01897-8. Insertion of MBP into cellular structures can be detected by gold- immunolabling using anti MBD antibodies as described in Shpigel et al. (1988), Plant Physiol. 117: 1185-1194. Changes in seed cost hardness can be tested by Instrone. Determination of seed germination rate is described in Levitov et al. (1995) Annals of Botany. 76: 1-6.

According to the present invention, a wide variety of plants and plant cell systems may be engineered for the desired physiological and agronomic characteristics described herein using the nucleic acid constructs of the present invention and the various transformation methods mentioned above. In preferred embodiments, target plants and plant cells for engineering include, but are not limited to, those monocotyledonous and dicotyledonous plants, such as crops including grain crops (e.g., wheat, maize, rice, millet, barley), fruit crops (e.g., tomato, apple, pear, strawberry, orange), forage crops (e.g., alfalfa), root vegetable crops (e.g., carrot, potato, sugar beets, yam), leafy vegetable crops (e.g., lettuce, spinach); flowering plants (e.g., petunia, rose, chrysanthamum), conifers and pine trees (e.g., pine fir, spruce); plants used in phytoremediation (e.g., heavy metal accumulating plants); oil crops (e.g., sunflower, rape seed) and plants used for experimental purposes (e.g., Arabidopsis).

The transgenic cells also can be used to produce MBD-containing peptides. The recombinantly produced mannanase and MBD produced as described above, generally are at least substantially purified following expression. The phrase "substantially purified" when referring to mannanase means a composition which is essentially free of other cellular components with which the mannanase is associated in its native environment or the environment in which it is produced, for example, a transgenic cell. Purified protein is preferably in a homogeneous state although it can be in either a dry state or in an aqueous solution. Purity and homogeneity are typically determined using analytical chemistry techniques such as polyacrylamide gelelectrophoresis or high performance liquid chromatography. Generally, a substantially purified protein comprises more than 80% of all macromolecular species present in the preparation. Preferably, the protein is purified to represent greater than 90% of all proteins present. More preferably the protein is purified to greater than 95%, and most preferably the protein is purified to essential homogeneity, wherein other macromolecular species are not detectable by conventional techniques.

The mannanase, MBD, or fusion proteins can be purified to substantial purity by standard techniques well known in the art, by a combination of cell lysis (e.g., sonication) and affinity chromatography, including selective precipitation with such substances as ammonium sulfate; column chromatography; affinity methods, including immunopurification methods; and others. See, for instance, R. Scopes, Protein

Purification: Principles and Practice, Springer- Verlag: New York (1982), incorporated herein by reference. For example, the mannanase can be purified by binding to immobilized mannan from which it can be removed. As necessary, mannanase or MBD can then be further purified by standard protein chemistry techniques. For fusion products, subsequent digestion of the fusion protein with an appropriate proteolytic enzyme releases the desired polypeptide.

In addition to producing fusion proteins from fused genes, the fusion protein can be made chemically. The substrate binding region or multiples thereof is produced on its own, purified and then chemically linked to a polypeptide of interest or a non protein molecule (chemical moiety) using techniques known to those skilled in the art. Methods of protein conjugation include use of glutaraldehyde to couple MBD to a second protein. (Reichlin Methods ofEnzymology ( 180) 70: 159-165).

Conjugation of MBD to a protein of interest can be used as a general method for purifying a protein of interest. The purification can be carried out using for example standard affinity chromatography techniques in which an insoluble or immobilized mannan is contacted with the MBD conjugate or material "tagged" with MBD. The conjugate or the protein of interest can then be dissociated from the MBD and purified as necessary. Alternatively, an aqueous phase separation system can be used for purification and/or separation of MBD-tagged compositions or MBD-conjugates. The MBD-tagged composition or MBD conjugate is combined with a phase-forming polymer to which the MBD binds specifically and an incompatible polymer such as PEG. A conjugate between a MBD and chemical moiety (formulas (2) and (3) above) can be prepared as follows. The MBD can be obtained as described above, or by other means known to those skilled in the art. The chemical moiety then can be attached to the amino acid sequence obtainable from the polysaccharide binding domain by a variety of chemical methods including covalent modification, ionic bonding, hydrophobic bonding, hydrogen bonding, protein translation, protein expression or combination thereof. Where the catalytic domain remains bound to the binding domain, if desired, specific inhibitors of catalytic activity may be used to inactivate the catalytic unit without affecting binding of the MBD. Covalent modification reactions can involve terminal amines, sulfhydryl groups, azido groups and other commonly used biochemical covalent bonding reagents. Non- covalent modification reactions can involve anionic bonding, hydrophobic bonding, hydrogen bonding and other commonly used non-covalent bonding reagents. If the MBD is sensitive to a particular covalent or ionic reagent, the essential residues that make up the MBD can be protected by incubating the domain with a ligand capable of binding the domain during the modification reaction. This technique protects the MBD from reacting with the chemical agents used to modify other parts of the domain.

The conjugation ofa chemical moiety to the fusion protein can occur both in vivo and in vitro. Typically, reactions can be carried out in vitro but on occasion in vivo conjugation can occur in the form of glycosylation and the like. In vitro conjugation chemical reactions to modify the polysaccharide binding domain can be carried out while the MBD is either bound to the mannan matrix or free from the mannan matrix. Examples include the use of gluteraldehyde conjugation as described by Reichlin (1980), supra to couple a protein of interest such as protein A to the MBD. When the MBD is bound to the matrix, it offers the advantage of protecting the site that actually binds to the matrix while leaving other residues to react with the chemical moiety.

If bonding of the chemical moiety to the MBD results in a diminished capacity to bind the mannan substrate, a reaction procedure requiring the presence of the mannan matrix is preferred to retain the binding characteristics of the domain. Conjugation reactions can be carried out with fusion proteins when it is desired to obtain heterobifunctional properties of the fusion protein or chemical conjugate. For example, the fusion protein can comprise both an enzyme such as alkaline phosphatase, β- glucosidase or trypsin and a dye such as Coomassie blue or amido black. Depending on the use of a MBD-chemical conjugate, the chemical moiety can be selected from a variety of compounds, including dyes, chromophores, isotopic chemicals, proteins, fats, liquids, carbohydrates, pigments and the like. It also is desirable to use chemical moieties that are stable in both non-aqueous and aqueous environment. Thus, preferred are dyes, markers and tags that dry on mannan matrices. When it is important to reduce non-specific, background binding, reagents should be used that are easily removed from the reaction mixture. For example, reagents that do not bind to a mannan matrix, by themselves, can be used to reduce nonspecific, background binding. For example, a MBD fusion protein or chemical conjugate can be passed over a mannan matrix and the unreacted reagent washed through while the MBD-containing composition remains bound to the matrix. Alternatively, the conjugated MBD can be removed from the unreacted by centrifugal techniques using mannan matrix beads where the supernatant can contain the reagent and the MBD-conjugate can be sedimented along with the mannan matrix beads. Depending upon the particular protocol and the purpose of the reagent, the polypeptide of interest, or the chemical moiety, or the MBD alone can be tagged with a conjugate or untagged. A wide variety of tags have been used which provide for, directly or indirectly, a detectable signal. These tags include radionuclides, enzymes, fluorescors, particles, chemiluminesors, enzyme substrates or co-factors, enzyme inhibitors, magnetic particles, dyes, etc. Various methods exist for attaching or linking the tags to the polypeptides or MBDs and are well known to those skilled in the art. For example, the N-terminal amino groups of the polypeptide of interest can be derivatized to form a pyrolezone, while other free amino groups are protected, where the pyrolezone can then be contacted with various reagents to link a detectable signal generating moiety. Alternatively, labels can also be attached to the MBD by using protein modifying reagents such as sulfhydryl or azido groups.

In general, the MBD conjugates can be bound to mannan-containing plant tissues and/or plant parts at neutral pH in a medium ionic strength buffer of from about 10-³M to about 1M. Binding can be performed at temperatures from 4°C to at least 45°C depending on the conjugate. Binding is virtually instantaneous and the temperature is not critical. Once the MBD-conjugate is bound to the mannan- containing plant component, the component can be dried or remain in an aqueous environment, depending upon the intended use. The buffer used should be one which does not damage the tissue and /or plant part so as render it unsuitable for its intended purpose.

A conjugate between a MBD and a detectable ligand can be used as selective tags to indicate where mannans are located on various mannan containing plant surfaces and as removable dyes and stains of mannan-containing plant surfaces. Chemical agents such as antibiotics, fungicides, insecticides, texturizing agents, and peptides conjugated or fused to the MBD can be bound to such mannan-containing plant surfaces.

To debind the MBD conjugate from the matrix, a low ionic strength buffer or water is required or a buffer of alkaline pH or a chaotrophic salt. The temperature for desorption is not critical and generally in the range of 10°C-40°C, although ambient temperatures are generally preferred, i.e., about 20°. The bound MBD conjugate is washed repeatedly in water or diluted by a continuous stream of water. Generally, pH 9.5 a carbonate buffer or 6M guanidine HC1 can be used for this desorption step. Dilute sodium hydroxide (about 0.1M) may be the preferred treatment in some cases. The nature of the MBD can be modified to alter its adherence properties so that it can be, or, if desired, cannot be, desorbed by water. Application of the desorption medium to the matrix causes the MBD of the fusion protein to release the MBD-conjugate from the matrix. The present invention can be used to provide transgenic plants with modified cell walls possessing different properties, such as, plants having longer or shorter fibers; plants which are either more or less digestible in the rumen of animals; and plants which are either more or less resistance to pests, such as insects, fungi, viruses and bacteria. The invention also can provide a means for modification of growth and/or fibre characteristics in higher plants. These include agricultural crops and trees of commercial importance. Mannans/glucomannans/galactomannans are prevalent in both the primary and secondary cell walls of woody plants and transgenic species may have useful properties, e.g. in pulp and paper manufacture. Seeds such as coffee beans which have a high mannan content can be modified so as to improve extractability of e.g. beans for instant coffee manufacture.

The invention also finds use for applying an antifungal agent, alone or fused/cross-linked to an antibiotic to plant surfaces, such as roots and seeds. Properties of other plant materials containing mannan also can be altered to improve the extraction or process of plant components such as polysaccharides, phenolic compounds and proteins as well as quality of the food products produced by these plants.

MBD-fusion proteins produced in plants can be readily purified using the mannan produced by the plant. Composite materials containing mannan can be produced by crosslinking mannan to other materials using MBD fused to other domains such as CBD to improve/alter the mechanical properties of these materials.

The following examples are offered by way of illustration of the present invention, not limitation.

EXAMPLES

Example 1 Screening of a C. fimi genomic DNA library for β-mannanase genes.

A C. fimi genomic library was prepared by inserting genomic DNA fragments (2 to 5 kbp) into the EcoR I site of the multiple cloning site of λ-ZAPII, prepared by Stratagene. This created translational fusions of the genomic inserts with the first 36 amino acids of and E. coli (lacZ) β-galactosidase coding sequence transcribed from the lacZ promoter (Stratagene; Meinke et al., (1993) J Bacteriol. 775(7):1910-1918). Therefore, the C. fimi λ-ZapII library could be screened for IPTG-inducible β- mannanase activity on azo-carob galactomannan plates. Eight plaques with mannanase activity were isolated, their phagemids (pBluescript SK+C. fimi DNA insert) excised, and transferred into E. coli XLOLR cells. All eight mannanase clones, CManl-8, secreted mannanase activity into the culture supernatant. In the supernatant of each clone at least two bands with mannanase activity could be detected by zymograms. The only clone that produced higher molecular weight bands with significantly more mannanase activity than the other clones was Cman2. From the other seven clones, slightly more mannanase activity could be detected in the supernatant of Cman4. Cman2 and Cman4 therefore were chosen for further analysis. Restriction maps of their plasmids were established (Figure 1). pCMan2 had a 4.3 kbp pCMan4 a 6.3 kbp insert. The similarity of the restriction pattern from pCMan2 and pCMan4 suggested that their inserts had a DNA fragment in common (Figure 1). By DNA sequencing it was shown that these two plasmids carried inserts with identical 5' ends except for 65 extra bp in pCMan4 (Figure 2). This suggested that the 6.3 kbp long insert of pCMan4 contained DNA encoding the same mannanase as pCMan2. The insert in pCMan4, however, was 65 bp longer at the 5' end and 2 kbp longer at the 3' end than pCMan2 (Figure 2). Hence, pCMan2 was chosen for DNA sequencing. The subclones and oligodeoxyribonucleotide primers used for sequencing are MBD 11:5' AGC GCG CAG CTC GAC AAC AGC ACC TAC ACC GTC ACC GCG ACG 3' (SEQ ID NO:3) and MBD12:5' GCG TCG GGC TCG CTC GTC GTC GAC GAC ATC GCC GCC CAC CCC 3' (SEQ ID NO:4). No putative start codon was found in either pCMan2 or in pCMan4.

Because the mannanase activities detected in the supernatants of all the other clones, CManl, 3, and 5 to 8, were very similar to CMan4 (vide supra) the library was rescreened for more mannanase encoding clones. The extracellular mannanase activities of five clones, CManl 0, 20, 30, 40 and 50 were analyzed on zymograms. CMan30 secreted two active polypeptides that were very similar in size to those secreted by C.fimi. The plasmid insert of CMan30 was 3 kbp in size. Comparisons of restriction digests of the plasmids pCMan30, pCMan2 and pCMan4 showed that all three plasmids had a DNA region in common. An ATG start codon was found in pCMan30, 540 bp downstream from the 5' end of the insert. This start codon was found to be only 6 bp and 72 bp upstream from the 5' ends of the inserts in pCMan4 and pCMan2, respectively (Figure 2).

Example 2 Nucleotide and deduced amino acid sequence of the C. fimi mannanase.

The nucleotide sequence (SEQ ID NO:l) of its gene and the deduced amino acid sequence (SEQ ID NO:2) of the C. fimi mannanase are shown in Figure 3. The open reading frame was 3033 bp long, which translated into a 1011 amino acid long protein with a calculated MW of 107,033. The N-terminus of the C. fimi mannanase had an amino acid composition rich in positively charged residues, a characteristic of secretion signal peptides (Nielsen et al., (1997) Prot. Eng. 10(l):l-6). To determine the N-terminus of the secreted and processed C. fimi mannanase, concentrated C. fimi LBG culture supernatant was separated and analyzed by a non-reducing PAGE-zymogram and blotted onto a PVDF membrane for N-terminal sequence analysis. The N-terminal sequence of the processed mannanase, corresponding to a 75 kDa active polypeptide, was determined by Edman degradation as ₅₀APADET₅₅ (SEQ ID NO:9), with the starting methionine being position 1. The cleavage between Thr 49 and Ala 50 was not in agreement with the signal peptide cleavage site as predicted by computer analysis using the program SignalP (Nielsen et al. , 1997). Cleavage was predicted to occur between Ala 40 and Ala 41 in the sequence ₃₇PAPA APV₄₃ (SEQ ID NO: 10) (O-: indicating the cleavage site). This prediction was in agreement with the (-3, -1) rule (Nielsen et al, 1997) and with the consensus cleavage sequence A/VχA O- A (χ can be any amino acid) from secreted C. fimi glycanases. Secreted C. fimi protease preferentially cleaves the C-terminal to threonines (Gilkes et al, (1989) J Biol. Chem. 25 :17802-17808), suggesting that the experimentally determined N-terminus APADET (SEQ ID NO:9) (cleaved after Thr 49) of the secreted mannanase resulted from signal peptidase processing followed by proteolysis by the secreted C. fimi protease.

The amino acid sequence of the C. fimi mannanase was compared to the amino acid sequences of other proteins. The sequence of the N-terminal half of the C. fimi mannanase was similar to those of the catalytic domains of mannanases in glycosyl hydrolase family 26 (EC 3.2.1.78). The highest identity was with mannanase ManA from Pseudomonas fluorescens ssp. cellulosa (Pf ManA) (Braithwaite et al, (1995) Biochem. J. 505:1005-1010). The two proteins were 46 % identical over a sequence of 328 amino acids. The C. fimi mannanase was also similar to mannanases ManB from Bacillus subtilis (Mendoza et al, (1995) Biochim. Biophys. Acta 72 5:552-554), ManB from Caldocellulosiruptor saccharolyticus (Ltithi et al, (1991) Appl. Environ. Microbiol. 57:694-700) and ManA, ManB and ManC from Piromyces sp. (Millward- Sadler et al, (1996) FEMS Microbiol. Letters 747:183-188), all members of family 26 (Millward-Sadler et al (1996) Supra) (Figure 4).

The best studied enzyme in family 26 is Pf ManA. This enzyme, representative for all family 26 members, cleaves its substrate via a double displacement mechanism, with a net retention of the configuration at the anomeric center. The catalytic residues in Pf ManA were determined by site-directed mutation of conserved family 26 carboxylic residues and kinetic studies of these mutants. Glu 212 was identified as the acid-base catalyst, and Glu 320 as the catalytic nucleophile (Bolam et al, (1996) Biochemistry 55:16195-16204). Both catalytic residues from Pf ManA are conserved in the C. fimi mannanase and correspond to Glu 225 as acid/base catalyst and to Glu 320 as the catalytic nucleophile (Figure 5). In accordance with the nomenclature proposed recently the C. fimi mannanase, the first family 26 enzyme from C. fimi to be described was named Man26A (or Cf Man26A) (Henrissat et al, (1998) FEBS Letters 425:352- 354).

In Man26A, between residues 680 and 860, another region with homology to other proteins was found. All the proteins sharing identities with Man26A in this region are either S-layer proteins, or proteins with a S-layer homology (SLH) domain, e.g., Bacillus anthracis S-layer protein (24 % identical residues over a sequence of 174 amino acids), Bacillus sp. SprB (31 % identity over 116 amino acids), Clostridium thermocellum ORF3p, also reported as ANCA (30 % identity over 130 amino acids), and the endoglucanase from Clostridium josui (23 % identity over 119 amino acids), to name just a few (Figure 4). SLH domains are generally involved in anchoring S-layer proteins to the bacterial cell wall. They also occur in other secreted proteins, such as xylanases, pullulanases and cellulosome anchoring proteins (Ries et al, (1997) J. Bacteriol. 779(12):3892-3898; Lemaire et al, (1993) J. Bacteriol. 177(9):2451-2459; Fujino et al, (1993a) J. Bacteriol. 775:1891-1899). Cf Man26A is the first mannanase reported to have a SLH domain.

The sequences in Man26A between amino acid 470 to 680 and 940 to 1011 did not show any significant homologies to other protein sequences.

Example 3 Identification of Mannan Binding Domain. Sequence of binding domain

Binding of Man26A to soluble mannan was studied by affinity gel electrophoresis (AFGE). In this technique native protein samples are electrophoretically separated in polyacrylamide gels containing soluble substrate. Reversible binding of the protein to the soluble substrate would reduce its migration distance (mobility). The change in mobility is proportional to the substrate concentration. The mobility of proteins without binding specificity should not change, even in the presence of high concentrations of the polymeric substrate (Nakamura et al, (1992) J. Chrom. 597:351-356).

The AFGE-zymogram method was used to test whether the catalytic domain of Man26A, or the non-catalytic portion of the enzyme was involved in substrate binding. The catalytic domain was obtained by C. fimi protease treatment of Man26A. Changes in relative mobility of undigested Man26A were compared to changes in relative mobility of the Man26A catalytic domain on AFGE gels with 0 % and 1 % mannan. To confirm that the proteolytic band corresponded to the catalytic domain, the 1 % azo- carob galactomannan gel was incubated in phosphate buffer (1 h at 37°C) prior to Coomassie blue staining. The presence of mannan did not affect the relative mobility of the catalytic domain, seen as a zone of clearing on the 1 % substrate gel, whereas the intact enzyme did not even migrate into the gel (Figure 6). It appeared that mannan binding was due to a domain other than the catalytic domain. The region between the catalytic domain and the SLH domain in Man26A has no significant sequence similarity to any other known protein sequences (see Example 2). It was hypothesized, therefore, that a mannan binding domain might be present between the catalytic domain and the SLH domain.

Example 4 Sub-Cloning of mbd_UΩ The DNA fragment encoding the protein portion between the catalytic and SLH domain, the putative mannan binding domain (MBD), was cloned into the pET28a (Novagen) as described: The primers MBDl 1 (Nco I) and MBD 12 (Not I) (Example 1) were used to amplify the MBD encoding DNA fragment by the polymerase chain reaction (PCR) using pCMan2 as template (see Example 1). The PCR product was cloned into the pZErO™ 1.1 vector (Invitrogen) at the EcoR V restriction site. From this plasmid, pOMBD_]112, the 570 pb mbdlll2 DNA fragment was excised by Nco I and Not I restriction endonuclease digestion, and ligated into the expression vector pΕT28a to obtain pET28MBD₁₁₁₂, encoding the putative MBD translationally fused to a C-terminal sequence of six histidines.

The mbdliπ coding region was expressed in E. coli BL21(DE3) cells, producing a protein, MBD,,₁₂, with a calculated molecular weight of 20,990. Initial expression levels were low and purification by MCAC metal chelate affinity chromatography was unsuccessful because of poor binding of MBD_U12 to the affinity column.

Example 5 Characterization of recombinant mannan binding domain (MBD_m2).

Binding of MBD_1U2 to mannan was analyzed by AFGE. The MBD_{] 112} protein bands from the partially purified sample were detected after AFGE separation on Western blots, using oligohistidine specific antibodies (Figure 7). Man2A was used as the non-mannan binding control. To mark the top of the AFGE gels, CBD_CenD, a protein that does not migrate into the separating gels under native conditions was applied. The relative mobilities of Man26A and MBD_{I 112} were compared in gels including 0 %, 5 x 10-* %, 7.5 x 10^"4 %, 1.25 x 10^"3 %, 2.5 x 10^"3 % and 1.25 x 10^"2 % locust bean gum and 1.0 x 10^'3 %, 1.5 x 10^"3 %, 2.5 x 10^"3 %, 5.0 x 10^"3 % and 2.5 x 10^'2 % azo-carob galactomannan. The relative mobilities of Man26A and MBD_{1 U2} decreased with increasing substrate concentrations. The decrease was similar for both proteins. However, slightly stronger protein-substrate interactions were detected for Man26A (Figure 7). From the double reciprocal plots of l/(R-r) vs. 1/c (See Equation 1) the negative reciprocal of the dissociation constant could be determined as the intercept on the abscissa.

d: migrating distance of protein (MBD_{1 I 12}) D migrating distance of reference (Man2A) r relative mobility (d/D) of MBD_U12 in the presence of mannan R relative mobility of unbound MBD_{1] 12}

Re relative mobility of MDB₁₁₁₂-mannan complex, or that value obtained in the presence of an excess amount of mannan with all MBD molecules bound K_d: dissociation constant of protein for affinity ligand

The dissociation constants for MBD_{U 12} were determined to be K_d = 4.6 x 10^"4 % (± 4 x 10-⁵) for locust bean gum and K_d = 5.5 x 10^-3 % (± 1 x 10^-3) for azo-carob galactomannan. The 12 fold weaker binding of MBD_{U 12} to the azo-substrate could be caused by the Remazol brilliant-Blue R molecules linked to the backbone to an extent of about one dye molecule per 20 sugar residues reducing the accessibility of binding sites. Furthermore, a lower viscosity, as found for the azo-carob galactomannan, could be indicative of molecules with a lower degree of polymerization (DP). Shorter molecules would have fewer potential binding sites, assuming MBD_{U 12} binds to the mannan backbone (see Example 4).

In order to calculate the molar dissociation constant, the molarity of the locust bean gum solution was determined by total and reducing sugar analysis, assuming a galactose to mannose ratio of 1:5 (McCleary et al, 1985) in the galactomannan. The soluble galactomannan molecules had an average DP of 130 mannose residues, which meant that a 1 % solution was 0.39 mM. This gave a molar dissociation constant for MBD_{I 112} on locust bean gum of K_d = 1.8 x 10^"7 M. Although only an approximation, it suggests a high affinity of MBD_1U2 for soluble galactomannan.

Example 6 Transgenic plants expressing MBDs in different tissues and different cell compartments.

Transgenic plants expressing MBDs can be constructed as follows. The coding sequences for the mannan binding domain (see Examples 1 and 5) can be excised from the pET28aMBD_IU2 vector previously described (see Example 4) and recombined into the pBIlOl .1 vector. Tissue specific expression can be derived by many promoters known in the literature. For example, an Arabidopsis thaliana Cell promoter can be used to express MBDs in elongating fast growing tissues (Shani et al, 1997). The 1.6 kb ce/1 promoter region (base 5-1618 accession # X98543) is cloned into the binary vector pBI 101.1 at the 5'- end of the mbd coding sequence. The construct is mobilized into disarmed LB4404 Agrobacterium tumefaciens by triparental mating An (1987) Meth. Enzymol. 153: 292-305. Transformation of tomato plants for example can be done according to Beaudoin and Rothstein (1997) Plant Mol Biol 33: 835-46. The MBD and MBD-fusion proteins can be targeted to the cell wall using an appropriate signal peptide such as the Ce/1 signal peptide.

Construction of CaMV35S promoter-ce/1

vector can be accomplished as follows: A DNA fragment encoding the ce/1 signal peptide is cloned into pUCl 8 (Promega, Madison, WI, USA). To this end, a PCR fragment using the following primers: (Sphl) 5'-AAAAGCATGCCGCGAAAATCCCTAATTT-3' (SEQ ID NO: 11) and (S I):5'-AAAAGTGACTTTACGG AGAGCGTCGC-3' (SEQ ID NO: 12) is generated, and following restriction with Sphl and Sail is cloned into the Hindlϊl and Sail cloning sites of pUC18. Inclusion of the Sphl restriction site replace the first amino acid after the initiation site from Alanine to Proline. A mbd 1 coding DNA fragment is generated by PCR amplification using primers which include Sail (at the 5' end of the sense primer) and EcoRI (at the 5' end of the antisense primer) restriction sites. Following Sail and EcoRI restriction, the mbd coding DNA fragment is cloned into the Sail and EcoRI sites of the above modified pUC18 vector, fused to, and in frame with, the signal peptide of ce/1. The DNA containing the ce/1 signal- mbd fusion is cloned into pCd cloning cassette (Broido et al, 1993) using the Sphl and EcoRI cloning sites. The pCd vector contains a polylinker downstream of a CaMV35S promoter (Gulley et al, (1982) Cell, 30: 763-773) and the omega-DNA sequence from the coat protein gene of tobacco mosaic virus (Gallie et al, (1987) Nucl. Acids Res. 15: 3257-3273). A DNA fragment containing a CaMV35S-omega-ce/l signal peptide-mbd and the octopine polyadenylation site is excised using BamHl and Sacl and thereafter subcloned into the binary vector pBHOl (CLONTECH) using BamHl and S cl cloning sites. Alternatively, MBD or MBD fusion proteins can be expressed specifically during secondary cell wall deposition using the parsley 4CL promoter (Hauffe et al, (1991) The Plant Cell 3 : 435-443). Constitutive expression can be achieved with a CaMV35 S promoter, for example. Accumulation of MBD or MBD-fusion protein in the ER can be achieved by addition of nucleotides at the 3' end of the gene that encodes for either HDEL or KDEL sequence.

Example 7 Analysis of transgenic plants.

Transgenic plants transformed with the pBHOl construct previously described (see Example 6) or a similar vector capable of expression of MBD in plants can be tested for expression of MBD. Northern blot analysis of MBD can be carried out using the cDNA fragment shown in Figure 3 (SEQ ID NO:l) as a probe. Furthermore, RTPCR can be performed on RNA obtained from such transgenic plants using PCR primers MBDl 1 (SEQ ID NO:3) and MBD12 (SEQ ID NO:4) to amplify the MBD transgene. Additional methods for analysis of transgenic plants include Southern analysis RNAse protection, protein gel electrophoresis and Western blot techniques, immunoprecipitation, enzyme-linked immunoassays or mannan binding activity. In addition, techniques such as in situ hybridization and immunostaining also may be used to detect the presence or expression of the recombinant construct in specific plant organs and tissues. The methods for performing all of these assays are well known to those skilled in the art. In addition to analyzing expression of the MBD transgene, transgenic plants can also be evaluated phenotypically. For example, transgenic plants can be compared to their wild-type counterparts with respect to plant growth, morphology, mechanical properties and cell-wall composition.

Claims

What is claimed is:

1. A method for modifying one or more mannan-containing plant components, said method comprising: expressing in a transgenic plant a recombinant nucleic acid that encodes a polypeptide comprising a non-catalytic binding domain that binds specifically to mannan in said plant components, whereby modified mannan-containing plant components are obtained as compared to an identically cultivated progenitor plant which does not contain said recombinant nucleic acid.

2. The method according to Claim 1 , wherein said mannan is selected from the group consisting of mannan, glucomannan and galactomannan.

3. The method according to Claim 1, wherein said plant component is a cell wall.

4. The method according to Claim 1 , wherein said plant component is a seed coat.

5. The method according to Claim 1 , wherein said plant is a higher plant.

6. The method according to Claim 1, wherein as a result of modifying said plant component said transgenic plant has an altered physical characteristic.

7. The method according to Claim 6, wherein said altered physical characteristic is selected from the group consisting of plant growth rate, height of mature plant, size of plant canopy, or size of plant roots

8. A transgenic plant comprising : one or more altered mannan-containing components as a result of expression of a recombinant nucleic acid encoding a polypeptide comprising a non- catalytic binding domain that binds specifically to other than terminal sugars of mannan in said components.

9. A seed produced by a transgenic plant according to Claim 8, wherein said seed comprises said recombinant nucleic acid construct.

10. A progeny, clone, cell line or cell ofa transgenic plant according to

Claim 8, wherein said progeny, clone, cell line or cell comprises said recombinant nucleic acid construct.

11. An isolated nucleic acid comprising: as operably linked components, a promoter functional in a higher plant cell, a first nucleotide sequence encoding a glycosyl hydrolase comprising a catalytic domain, a non-catalytic mannan binding domain, wherein said binding domain binds to the backbone of a polymeric β-l,4-mannanose, and an SLH domain.

12. The isolated nucleic acid molecule according to Claim 11, wherein said glycosyl hydrolase has an amino acid sequence as shown in SEQ ID NO:l.

13. The isolated nucleic acid according to Claim 11, wherein said promoter is a constitutive plant promoter.

14. The isolated nucleic acid according to Claim 13, wherein said constitutive plant promoter is a CaMV 35S promoter.

15. The isolated nucleic acid according to Claim 13, wherein the promoter is a tissue specific plant promoter.

16. The isolated nucleic acid according to Claim 13, wherein said promoter is from a gene which is expressed during secondary cell wall deposition in a higher plant cell.

17. The isolated nucleic acid according to Claim 16, wherein said promoter is a parsley 4CL promoter.

18. The isolated nucleic acid according to Claim 11 , further comprising: an operably linked component, a second nucleotide sequence encoding a secretion signal peptide.

19. The isolated nucleic acid according to Claim 18, wherein said secretion signal peptide is a ce/1 secretion signal peptide.

20. The isolated nucleic acid according to Claim 19, wherein said ce/1 secretion signal peptide is obtainable from Arabidopsis thaliana.

21. A vector comprising: an isolated nucleic acid according to Claim 11.

22. A higher plant cell comprising: a vector according to Claim 21.

23. A method for obtaining seeds having an altered germination rate, said method comprising: growing a transgenic plant containing a DNA construct comprising as operably linked components, a promoter functional in said plant, a nucleotide sequence encoding a seed coat modifying polypeptide and a transcriptional and translational termination region functional in said plant under conditions whereby seed are produced which contain said DNA construct.

24. The method according to Claim 23, wherein said coat modifying polypeptide is a glycosyl hydrolase comprising a non-catalytic domain that binds to the backbone for polymeric β-l,4-mannanose.

25. The method according to Claim 23, wherein said coat modifying polypeptide is a non-catalytic binding domain from a glycosyl hydrolase, wherein said non-catalytic binding domain binds to the backbone of a polymeric β-l,4-mannanose.

6. The method according to Claim 23, wherein said promoter is expressed preferentially in seed as compared to other tissues of said plant.