Plant arabinogalactan protein (AGP) genes

FIELD OF THE INVENTION

The subject matter of the invention relates to the isolation of arabinogalactan proteins (AGPs) from plants, e.g., Nicotiana alata, Nicotiana plumbaginafolia and Pyrus communis, and the utilization of amino acid sequences of various AGP fragments for the isolation of corresponding plant genes encoding the protein backbone of AGPs.

BACKGROUND OF THE INVENTION

Arabinogalactan proteins (AGPs) are found in flowering plants from every taxonomic group tested. These proteoglycans are widely distributed in most higher plants, occurring in almost all tissues including leaves, stems, roots, floral parts, seeds, and in many of their secretions. These macromolecules are found predominantly in soluble form in the intercellular wall space [Clarke et al. (1975) J. Cell Sci. 19:157-167; Clarke et al. (1978) Aust. J. Plant Physiol, 5:707-722], and are also localized in cytoplasmic organelles [Anderson et al. (1977) Aust. J. Plant Physiol. 4:143-158], at the protoplast surface [Clarke et al. (1975), supra and (1978), supra; Komalavilas et al. (1991) J. Biol. Chem. 266:15956-15965; Pennell et al. (1991) Plant Cell 3:1317-1326; Kieliszewski et al. (1992) Plant Physiol. 99:538-547] and in the cell wall [Bacic et al. (1988), The Biochemistry of Plants, Preiss, J. ed., Vol. 14, pp. 297-371 Academic Press, San Diego; Knox (1990) J. Cell Science 96:557-561; Roberts (1990) Current Opinion in Cell Biology 2:920-928; Knox (1992) Protoplasma 167: 1-9; Pennell (1992) Soc. Expt. Biol. Seminar Series 48: Perspectives in Plant Cell Recognition (ed. J. A. Callow and J. R. Green) Cambridge University Press pp. 105-121; Showalter (1993) Plant Cell 5:9-23; Wycoff et al. (Ref.)].

In cell cultures, AGPs are secreted into the medium [Fincher et al. (1983), Ann. Rev. Plant Physiol. 34:47-70]. Several AGPs from culture media have been investigated, including those from ryegrass cells [Anderson et al. (1977), supra; Glesson et al. (1989) Biochem. J. 264:857-262]; tobacco cells [Akiyama et al. (1981) Phytochemistry 20:2507-2510]; blackberry cells [Cartier et al. (1987) Carbohydrate Res. 168:275-283]; sycamore cells [Aspinall et al. (1969) Can. J. Biochem. 47:1063-1070]; carrot cells [Jermyn et al. (1985) AGP News 5:4-25; Kreuger et al. (1993) Planta 189:243-248]; Rosa cell suspension culture [Komalavilas et al. (1991) J. Biol. Chem. 266:15956-15965]; gladiolus cells [Glesson et al. (1979) Biochem. J. 181:607-621]; and maize cells [Kieliszewski et al. (1992) Plant Physiol. 99:538-547].

The multi-site localization of AGPs appears to be analogous to the multi-site localization of some animal proteoglycans. As regards chemical structure, however, little similarity seems to exist between plant AGPs and animal proteoglycans.

The AGPs are a family of structurally related glycosylated molecules containing high proportions of carbohydrate and usually less than 10 percent by weight of protein [Clarke et al. (1979), supra; Fincher et al. (1983), supra], although AGPs having a protein content of about 59% are known [Fincher et al. (1983), supra; Anderson et al. (1979) Phytochem. 18:609-610]. The carbohydrate consists of polysaccharide chains having a 1,3-.beta.-D-galactopyranosyl backbone and side chains of (1,3-.beta.- or 1,6-.beta.-)D-galactopyranosyl (Gal) residues and often terminating in .beta.-D-Galp and .alpha.-L-arabinofuranosyl Araf residues [Kreuger et al. (1993) Planta 189:243-248]. Other neutral sugars and uronic acids have also been detected, although at low levels. Monosaccharides which can be present are L-rhamnopyranose, D-mannopyranose, D-xylopyranose, D-glucopyranose, D-glucuronic acid and its 4-0-methyl derivative and D-galacturonic acid and its 4-0-methyl derivative [Fincher et al. (1983), supra]. In most cases, however, Gal and Ara predominate.

The protein content is usually between two and ten percent [Fincher et al. (1983), supra]. In contrast with the polysaccharide component, relatively little is known about the structure and organization of the protein core of AGPs, except that the protein appears to have domains rich in alanine, hydroxyproline, serine, and threonine [Fincher et al. (1983), supra]. This is reflected in the amino acid sequences that were obtained for AGP peptide fragments from carrot [Jermyn et al. (1985) supra]; Italian ryegrass [Glesson et al. (1989), supra]; and Rose (Komalavilas et al. (1991), supra]. A common feature of many of these isolated peptide fragments is the dipeptide Ala-Hyp, which is directly repeated in various AGP peptide fragments. To date, the entire amino acid sequence of an intact isolated AGP is not available publicly. The high carbohydrate content of AGPs appears to cause difficulties in sequencing; attempts to chemically remove the carbohydrate moiety usually results in incomplete deglycosylation and products with variable levels of carbohydrate content. The carbohydrate-protein linkage has been identified as a .beta.-galactosyl-hydroxyproline linkage in AGPs isolated from wheat and ryegrass [Glesson et al. (1985) AGP News 5:30-36 and McNamara and Stone (1981) Lebensm.-Wiss. u-Technol. 14:182-187].

AGPs are components of Gum arabic, a gummy exudation originating from the Acacia tree and known to be produced by stress conditions such as heat, drought, and wounding [Clarke et al. (1979) Biochemistry 18:520-540]. The gum finds wide use as a flavor encapsulator in dry mix products such as puddings, desserts, cake mixes and soup mixes, and is also used to emulsify essential oils in soft drinks and to prevent sugar crystallization in confectionery products [Randall et al. (1989) Food Hydrocolloids 3:65-75]. More recently, the significance of the protein component to the overall structural and functional characteristics of gums has been realized [Vandevelde et al. (1985) Carbohydr. Polymers 5:251-273; Connolly et al. (1987) Food Hydrocolloids 1:477-480 and Connolly et al. (1988) Carbohydr. Polymers 8:23-32]. The importance of the protein-rich fraction to the emulsification properties of the gum has been demonstrated [Randall et al. (1988) Food Hydrocolloids, 2:131-140].

AGPs function in several biological processes including plant development, cell-cell adhesion, pollen-stigma recognition, water retention, and disease resistance. AGPs may serve as glues or provide nutrients for growing pollen tubes. It has been suggested [Fincher et al. (1983) supra] that AGP proteins may interact with lectins or other proteins in the extracellular spaces and may be involved in the cellular response to extracellular oligosaccharide signal molecules [Norman et al. (1990) Planta 181:365-373]. Since AGPs interact with Yariv antigens and flavonol glycosides [Jermyn (1978) J. Plant Physiol. 5:563-571], they have been thought to have lectin-like properties. The molecular structure of AGPs has been proposed [Randall et al. (1989) Food Hydrocolloids 3:65-75] to resemble a type of block copolymer wherein carbohydrate blocks are covalently linked to a central polypeptide chain, thus explaining its ability to sterically stabilize emulsions and dispersions.

Plant AGP genes are not known in the prior art and the nucleotide sequence of a plant AGP gene has not been published to date. Very recently, it was reported [Sheng et al. (1993) Abstract no. 639 in Supplement to Plant Physiol. 102, Number 1, May 1993] that a PCR strategy is being used to clone potato tuber lectin, extensins and AGP sequences from a potato tuber cDNA library. It was reported that PCR products which hybridized to a carrot extensin probe gave several putative clones which are currently under investigation. No clones corresponding to AGP genes were disclosed.

The process of obtaining an AGP clone has been found to be complex and problematic. Two of the problems associated with AGPs and their genes are (1) the very high redundancy associated with the characteristic amino acid sequence of an AGP peptide, i.e., (a) a high hydroxyproline content and (b) regions containing a high content of hydroxyproline, alanine, serine, and threonine (OAST); and (2) the GC-richness of corresponding oligonucleotides leading to problems with the specificity of hybridization. Indistinct and imprecise alignment during nucleic acid hybridization, for example, in the PCR technique, has resulted in lack of success in the ability to obtain an AGP clone. This results in the amplification of incorrect sequences when compared to the original template. Plants are also known to contain a variety of glycine-rich proteins which are also encoded by GC-rich DNA. Applicants' disclosure circumvents this problem and enables the isolation of AGP genes.

Two approaches to the isolation of the AGPs from plant extracts have been used in previous studies. One approach consists of classical fractionation of plant extracts [Fincher et al. (1974) Aust. J. Biol. Sci. 27:117-132; Aspinall (1969) Adv. Carbohydrate Chem. 24:333-379]. A convenient initial fractionation of extracts is treatment to saturation with (NH.sub.4).sub.2 SO.sub.4, which does not usually precipitate AGPs. Subsequent ion-exchange and affinity chromatography can be used to isolate the AGPs.

Another approach to the isolation of AGPs from plant extracts is precipitation with a class of dyes prepared by coupling diazotized 4-aminophenyl glycosides to phloroglucinol [Jermyn et al. (1975), supra]. These dyes were first prepared by Yariv et al. (1962) Biochem. J. 85:383-388) as precipitating antigens for antibodies to glycoside determinants, and the .beta.-glycosyl artificial carbohydrate antigen was shown to precipitate an arabinose-and-galactose-containing polymer from soya bean, jack bean and maize [Yariv et al. (1967) Biochem. J. 105:1c-2c]. Since then, this precipitation reaction has been widely used to isolate AGPs from extracts of seeds of every taxonomic group of flowering plants, as well as leaf extracts and callus-culture filtrates [Jermyn & Yeow (1975) Aust. J. Plant Physiol. 2:501-531; Anderson et al. (1977), supra; and review by Clarke et al. (1979), Phytochemistry 18:521-540].

These dyes have also been used as cytochemical reagents for the localization of AGPs in plant tissues [Clarke et al. (1975), J. Cell Sci. 19:157-167; Clarke et al. (1978), Q. Rev. Biol. 53:3-28]. The nature of the binding of AGP to the Yariv reagent is not understood, but it is likely to involve both carbohydrate and protein residues. The binding of Yariv's reagent to AGP is not affected by removal of the arabinose residues [Glesson et al. (1979), supra; Akiyama et al. (1981), supra], but is abolished by progressive acid hydrolysis of the AGP [Fincher et al. (1983), supra].

In higher plants AGPs are also classified as belonging to a group of proteins characterized by hydroxyproline-rich domains. These hydroxyproline-rich glycoproteins (HRGPs) are also characterized by carbohydrate side chains that contain arabinose and galactose. The group has been traditionally divided into three main classes: the cell wall associated extensins; the soluble arabinogalactan-proteins (AGPs), and the solanaceous lectins. The differences between these groups are summarized in Table 1.0. The most important factors in the classification of the HRGPs are: the amount, composition, and sequence of their carbohydrate component, the sequence and composition of the polypeptide backbone, the linkage between carbohydrate and protein and its localization.

A new group of proteins, the proline-rich proteins, has been described recently. The proline-rich proteins (PRPs) have also been referred to as the hydroxyproline/proline-rich proteins or the repetitive proline-rich proteins. Amino acid compositions of some PRPs [Averyhart-Fullhard et al. (1988) Proc. Natl. Acad. 85:1082-1085; Datta et al. (1989) Plant Cell 1:945-952; Kleis-San Francisco et al. (1990) Plant Physiol. 94:1897-1902] indicated equimolar amounts of proline and hydroxyproline. However, the PRPs do not appear to be glycosylated and, in this way, are distinguished from the HRGPs (hydroxyproline-rich glycoproteins).

As indicated in Table 1.0, AGPs are readily distinguished from extension and lectin HRGPs. Extensins are highly positively charged HRGPs, are rich in hydroxyproline, lysine, tyrosine, serine, and proline, possess carbohydrate side chains that are rich in arabinose, and are tightly associated with cell walls. The high lysine content of the extensins contributes to their positive charge, and the tyrosine in extensin may form intermolecular [Stafstrom and Staehelin (1986), Plant Physiol. 81:234-241] and intermolecular isodityrosine linkages that have been implicated in cross-linking extensin in vitro [Everdeen et al. (1988) Plant Physiol. 87:616-621] and in vivo [Cooper and Varner (1983) Biochem. Biophys. Res. Comm. 112:161-167; Biggs and Fry (1990) Plant Physiol. 92:197-204].

Hydroxyproline accounts for 30-50% of the amino acids in extensin and is found in short peptides that are repeated a number of times in the molecule. The core peptide that is most commonly encoded by extensin genes is Ser(Pro).sub.4 (Table 1.0), which may be post-translationally modified to Ser(Hyp).sub.4. Recently, amino acid sequences have been obtained from extensin-like molecules that do not contain the Ser(Hyp).sub.4 peptides [Kieliszewski et al. (1990) Plant Physiol. 92:316-326; Li et al. (1990) Plant Physiol. 92: 327-333].

The carbohydrate side chains of extensins consist of short arabinosides linked to hydroxyproline, and single galactose residues linked to serine. The function of the carbohydrate side chains of extensins is not clear, but there is some evidence that they stabilize the polyproline II helix, which gives extensin its characteristic rod-like shape; Stafstrom and Staehelin (1986) Plant Physiol. 81:242-246.

The solanaceous lectins are positively charged glycoproteins that are identical to the extensins in the composition and structure of their carbohydrate side chains. Two important features discriminate the solanaceous lectins from extensins; their localization in the vacuole and cytoplasm [Millar et al. (1992) Biochem. J. 283:813-821], and their relatively high cysteine content (10-12 Mol %; Showalter (1993) Plant Cell 5:9-23). The cysteine in the potato lectin is concentrated in a single domain of the molecule that contains the carbohydrate binding site, and is distinct from the domain that is rich in hydroxyproline and glycosylated [Ashford et al. (1982) Biochem. J. 201:641-645]. The different lectins are immunologically cross-reactive [Kilpatrick et al. (1980), Biochem. J. 185:269-272], and contain both carbohydrate and protein epitopes [Ashford et al. (1982) supra].

The features that distinguish the AGPs from the extensins and solanaceous lectins are listed in Table 1.0. The AGPs usually have a negative to neutral overall charge, and are soluble in aqueous buffers. A characteristic feature of the AGPs is their ability to bind .beta.-glucosyl Yariv reagent, whereas extensins and lectins do not bind the Yariv reagent.

Carbohydrate forms a major portion of the mass of AGPs [Clarke et al. (1979), supra; Fincher et al. (1983), supra]. The majority of the AGPs that have been chemically characterized contain less than 10% (w/w) protein [Clarke et al. (1979), supra; Fincher et al. (1983), supra], but the AGPs from Cannabis sativa leaves (25% [w/w] protein), rice bran (27% [w/w] protein), and sycamore suspension cultures (19-38% [w/w] protein), are notable exceptions [Clarke et al. (1979), Phytochem. 18:521-540]. The protein backbones of AGPs often contain domains that are rich in hydroxyproline, alanine, serine, and threonine.

SUMMARY OF THE INVENTION

The present invention provides for the first time DNA fragments encoding protein backbones of plant arabinogalactan proteins (nonglycosylated AGPs). Specific embodiments of the invention present cDNA clones encoding nonglycosylated AGPs from cell suspension cultures of Nicotiana alata (NaAGP1), Nicotiana plumbaginafolia (NpAGP1), and Pyrus communis (PcAGP23 and PcAGP9) and from Nicotiana alata styles (Na35.sub.-- 1). Full length and partial nucleotide sequences of the cDNAs encoding said nonglycosylated AGPs are disclosed. DNA recombinant vectors containing these cDNAs are also provided. In further embodiments of the invention, genomic DNAs encoding plant nonglycosylated AGPs and recombinant vectors containing said genomic DNAs are provided. This invention further contemplates the use of oligonucleotide probes based on the amino acid sequence of plant AGPs for the detection of hybridizing sequences and the isolation of monocot and dicot AGP genes.

TABLE 1.0 __________________________________________________________________________ Biochemical and structural features of hydroxyproline-rich glycoproteins (HRGPs) HRGPs Arabinogalactan- Solanaceous Property Extensins proteins (AGPs) Lectins __________________________________________________________________________ % Protein (w/w) 40-50 2-10 50-60 Galactose/ <1 >1 <1 Arabinose Galactose terminal 1,3-linked terminal Linkage Types 1,3,6-linked 1,6-linked terminal Arabinose 1,2-linked terminal 1,2-linked Linkage Types 1,3-linked 1,3-linked terminal terminal Glycopeptide O-linked: O-linked: O-linked: linkages Ara--Hyp & Gal--Ser Gal--Hyp Ara--Hyp & Gal--Ser Abundant Amino Hyp, Lys, Tyr, Hyp, Ala & Ser Hyp, Cys, Gly & Acids Ser & Pro Ser mol % Hyp (of >30 >15 >13 protein domains) Amino Acid Ser(Hyp).sub.4 ? ? Repeats Isolectric Point 9.5-11 2-5 9.5 Localization Cell wall Extracellular Cytoplasm & matrix; plasma vacuole membrane .beta.-glucosyl Yariv No Yes No reagent binding __________________________________________________________________________

The invention also provides isolated plant AGP peptides and amino acid sequences of AGP peptide fragments. AGP peptides were isolated from Nicotiana alata, Nicotiana plumbaginafolia, and Pyrus communis. The amino acid sequences obtained from isolated AGP peptide fragments were either enriched in hydroxyproline or not enriched in hydroxyproline. In particular, hydroxyproline-enriched sequences were characterized by having (i) a high content of hydroxyproline and/or (ii) a high content of hydroxyproline, alanine, serine, and threonine (OAST-enriched). The sequences that were immediately useful in obtaining an AGP gene were those sequences that were not enriched in hydroxyproline, and not enriched in hydroxyproline, alanine, serine, and threonine content (not OAST-enriched). To date the amino acid sequence of an intact plant AGP is not publicly available. cDNAs thought to encode AGPs have been described, but evidence of a match between these sequences and amino acid sequence data from isolated AGPs is missing in these cases.

The invention further provides a substantially pure AGP having an amino acid sequence which is essentially that derived from a nucleotide sequence of an AGP gene. Specific embodiments of the invention provide an AGP comprising an amino acid sequence consisting essentially of that derived from the nucleotide sequence of an AGP gene from Nicotiana alata, Nicotiana plumbaginafolia, or Pyrus communis.

It is also an object of the invention to provide a method for obtaining a plant AGP gene. This method comprises the step of obtaining from an AGP peptide a fragment having an amino acid sequence that is hydroxyproline-poor, e.g., not enriched in OAST content. This hydroxyproline-poor sequence is then used to design a nucleotide primer which can be used to obtain, for example, a PCR fragment useful in screening a plant gene library for a hybridizing clone. Applicants' approach is novel and contrary to that generally used. Usually, a sequence which particularly characterizes an AGP (i.e., a sequence that is hydroxyproline-rich or enriched in OAST content) is utilized to design an oligonucleotide primer for use in obtaining a hybridizing clone. In Applicants' approach, a hydroxyproline-rich peptide sequence which particularly characterizes an AGP protein is not utilized, and is avoided; instead, a sequence which does not comprise a characterizing sequence of an AGP (i.e., a hydroxyproline-poor sequence) is utilized for the isolation of an AGP gene. In specific embodiments of the invention, peptide sequences which were not enriched in hydroxyproline or OAST content were isolated from N. alata, N. plumbaginafolia, and P. communis. These peptide sequences enabled the isolation of corresponding cDNA clones.

The present invention also provides a method for obtaining an AGP gene by utilizing a hydroxyproline-rich AGP sequence. Prior to the instant disclosure, public knowledge of hydroxyproline-rich AGP fragments has not enabled the isolation of corresponding AGP genes, due to difficulties imposed by resultant GC-rich domains. A method is provided herein that enables the use of a specific hydroxyproline-rich AGP peptide sequence for the isolation of a corresponding gene. The approach for using a hydroxyproline-rich sequence comprises the use of long guessmers combined with single-stranded antisense RNA probes for the screening of a library. The use of a long guessmer together with an RNA probe overcomes the problems presented upon using short oligonucleotide probes. A long guessmer can more easily accommodate mismatches and the use of an antisense RNA probe allows "U" to be used at the third position of the anticodon for AST amino acids, thus increasing the likelihood of the guessmer hybridizing to the target sequence. The resultant RNA molecule can be heavily labeled, permitting greater levels of detection, and also can bind more strongly to its target sequences than a DNA probe.

The invention also provides specific AGP cDNA sequences and specific oligonucleotide probe sequences for screening cDNA libraries to isolate specific plant AGP genes. For example, in specific embodiments, the following cDNA clones are provided:

______________________________________ Source cDNA clone ______________________________________ N. alata cell suspension culture NaAGP1 (SEQ ID NO:24) N. plumbaginafolia cell NpAGP1 (SEQ ID NO:25) suspension culture P. communis cell suspension PcAGP23 (SEQ ID NO:49) culture P. communis cell suspension PcAGP9 (SEQ ID NO:66) culture P. communis cell suspension PcAGP2 (SEQ ID NO:91) N. alata style Na35.sub.-- 1 (SEQ ID NO:63) N. alata style AGPNal 1 (SEQ ID NO:72) ______________________________________

The invention further provides antisense RNA probes designed such that they comprise one or more nucleotide sequences encoding amino acid sequences that are OAST-rich, representing the same or different AGPs. Also provided are RNA probes comprising a nucleotide sequence encoding an OAST-rich consensus sequence for plant AGPs. A guessmer-antisense RNA probe approach may also be used with an OAST-poor AGP sequence to isolate a corresponding AGP gene.

It is also an object of the present invention to provide an antibody to a substantially pure plant AGP, or fragment thereof, comprising an amino acid sequence consisting essentially of a whole or partial amino acid sequence derived from a plant AGP gene. Also provided is an antibody to an isolated AGP peptide fragment that is not enriched in hydroxyproline. Also provided by the invention is an antibody to a synthetic AGP peptide, or fragment thereof.

This invention further contemplates the use of antibodies to substantially pure AGP peptides, AGP peptide fragments not enriched in hydroxyproline or OAST content, or synthetic AGP peptides for (a) the detection, isolation, or diagnosis of AGPs in AGP-containing mixtures or tissues, and (b) in reducing or inhibiting natural biological and chemical AGP activities. It is recognized that polyclonal and monoclonal antibodies to AGPs or AGP peptides are most effective in AGP-containing specimens that are deglycosylated or otherwise preconditioned to expose the protein backbone of the AGP.

This invention also provides a genetically-engineered DNA molecule comprising a plant AGP gene under control of a heterologous promoter such that a nonglycosylated AGP is expressed. In a specific embodiment of the invention, an AGP gene obtained from N. alata., N. plumbaginafolia, or P. communis is inserted behind a heterologous promoter (e.g. a bacterial, viral, plant, etc., promoter) in a host cell such that a nonglycosylated AGP is expressed.

It is also an object of the invention to provide a recombinant DNA molecule comprising a plant AGP gene under control of a heterologous promoter such that a glycosylated AGP is expressed. For example, this invention contemplates the utilization of the expressed nonglycosylated AGP as a substrate for glycosylating and carbohydrate-protein linking enzymes (e.g., prolyl hydroxylase, glycosyl transferase, etc., to produce a glycosylated AGP). It is also an object of the invention to provide a host cell (for example, monocots, dicots, etc.) transformed with genetically-engineered DNA comprising a plant AGP gene under control of a heterologous promoter such that a glycosylated AGP is expressed. It is a further object of the invention to provide a plant AGP gene-transformed host cell capable of over-producing or under-producing nonglycosylated AGP. It is an additional object of the invention to provide an AGP gene-transformed host cell capable of further metabolic processing of an expressed nonglycosylated AGP.

This invention further provides a DNA fragment comprising a plant AGP promoter. In specific embodiments of the invention, AGP promoters are isolated from N. alata, N. plumbaginafolia, and P. communis. Subsequently, a recombinant DNA molecule is genetically engineered to comprise a plant AGP promoter situated adjacent to a heterologous structural gene such that the structural gene is expressed under the control of the plant AGP promoter. Also, the coding region of the gene could be used behind tissue-specific promoters to express the AGP at particular sites in a whole plant. This could change the phenotypes with respect to such functions as pest resistance, for example.

The instant invention provides a source of AGP that is not dependent upon its isolation from plant exudates, e.g., gum arabic, guar gum, etc. The availability of natural sources of AGP-containing gums, e.g., from trees, roots, seeds, seaweed, microbes, etc., present problems associated with harvesting, climate, man-power, fermentation, isolation, purity, and high costs. The production of AGPs using recombinant gene technology ensures (a) a method of supplying AGP that is independent of harvesting or fermentation requirements and problems, (b) that enables high levels of quality control, (c) that provides a supply of substantially pure AGP product, (d) that permits an overproduction of AGP in a host cell, and (e) that can be adapted to produce a specifically engineered AGP having desired properties. Thus, this invention provides a means for supplying the functions and utilities of plant gums, e.g., gum arabic, etc., without the need for finding renewable but shrinking natural sources of plant gums. These functions find wide applications as thickening, gelling, emulsifying, dispersing, suspending, stabilizing, encapsulating, flocculating, film-forming, sizing, adhesive, binding and/or coating agents, and/or as lubricants, water-retention agents, and coagulants.

BRIEF DESCRIPTION OF DRAWINGS

FIG. 1 presents different strategies for the preparation of single-stranded antisense RNA probes from oligonucleotides. A. Single oligonucleotide probes: FIG. 1A-1 presents two complementary guessmers annealed to each other to form a double-stranded construct containing the T7 promoter. FIG. 1A-2 presents a short primer annealed to form double-stranded T7 promoter sequence. B. Double oligonucleotide probes: FIG. 1B-1 presents two guessmers annealed to each other through the complementary adaptor sequences at their 3'-ends. FIG. 1B-2 presents two guessmers annealed to a mediator DNA through their adaptor sequences .rect-hollow.: adaptor sequence. Other promoters, for example, T3 or Sp6 RNA polymerases, may also be used.

FIG. 1C presents a Coomassie blue stained SDS-PAGE gel blot of deglycosylated and non-deglycosylated AGPs from various sources. AGPs were isolated from suspension culture filtrates of N. alata; N. plumbaginafolia and pear (Pyrus communis) by Yariv precipitation and deglycosylated with trifluoromethanesulfonic acid (TFMS). The deglycosylated and non-deglycosylated AGPs were separated on a 17.5% SDS-PAGE gel and blotted onto a PVDF membrane. After staining with Coomassie blue, the major band (MW 20-30 kD, indicated by an arrow) from deglycosylated N. alata AGPs was excised and sequenced.

FIGS. 1D-1 and 1D-2 present a PCR strategy for cloning of the NaAGP1 gene corresponding to an amino acid sequence of a deglycosylated AGP backbone from N. alata cell suspension culture. The sequences of the NaR1, NaF1, and NaF2 primers used to isolate the clone for NaAGP1 are given in Table 1.1 and FIG. 1E.

FIG. 1E presents a nucleotide and the derived amino acid sequences (SEQ ID NO:21) of the 160-bp primer extension fragment. The derived amino acid sequence corresponding to the peptide sequence by protein microsequencing is underlined. The asterisks (*) indicate the amino acids of the peptide obtained by direct microsequencing which are identical with the derived sequence. The sequences of the two oligonucleotides (NaF1, NaF2) designed for the amplification of the 3'-fragment of the AGP gene are double-underlined. The nucleotide sequence corresponding to the primers (NaR1) is underlined.

FIGS. 1F-1 and 1F-2 present the nucleotide and predicted amino acid sequences (SEQ ID NO:24) the NaAGP1 cDNA from N. alata cell suspension culture (NaAGP1). The nucleotide sequence obtained by PCR, which does not overlap with the cDNA clone, is in italics. The derived amino acid sequence corresponding to the peptide sequence by protein microsequencing is underlined. The asterisks (*) indicate the amino acids of the peptide obtained by direct microsequencing which are identical with the derived sequence. A predicted signal sequence is dot-underlined. X=undetermined residue.

FIG. 1G presents a summary of key structural features of the derived amino acid sequence of the NaAGP1 cDNA. The hydropathy values of each amino acid have been determined using an interval of nine amino acids according to the weight system of Kyte and Doolittle (1982). Values above the dotted line indicate hydrophobic regions, and the values below the dotted line represent hydrophilic regions.

FIGS. 1H-1 and 1H-2 present the nucleotide and predicted amino acid (SEQ ID NO:25) sequences of an N. plumbaginafolia AGP derived from cell suspension culture (NpAGP1). The derived amino acid sequences corresponding to the peptide sequence by protein microsequencing is underlined. The asterisks (*) indicate the amino acids of the peptide obtained by direct microsequencing which are identical with the derived sequence. O: hydroxyproline.

FIGS. 1I-1 and 1I-2 present the alignment of the derived amino acid sequences of the NaAGP1 (SEQ ID NO:24) and NpAGP1 (SEQ ID NO:25) cDNAs. The derived amino acid sequence of NaAGP1 cDNA is shown in the upper line and that of the NpAGP1 shown in the lower line. Identical aligned residues are indicated with `.vertline.`. Gaps were introduced when required to maximize the alignment.

FIGS. 1J-1, 1J-2, 1J-3 and 1J-4 present the alignment of the NaAGP1 (SEQ ID NO:24) and the NpAGP1 (SEQ ID NO:25) cDNA sequences. The nucleotide sequence of the NaAGP1 cDNA is shown in the upper line and that of the NpAGP1 shown in the lower line. Identical aligned residues are indicated with `.vertline.`. Gaps were introduced when required to maximize the alignment.

FIGS. 1K-1 and 1K-2 present northern blot analyses of the NaAGP1 and NpAGP1 genes.

FIG. 1K-1: Total RNA was isolated from N. alata (1) leaves, (2) pollen, (3) styles, (4) stems, (5) petals, (6) roots and (7) suspension-cultured cells. Equal amounts (10 .mu.g/lane) of RNA were fractionated on formaldehyde agarose gels, transferred to Hybond-N membranes, and hybridized with .sup.32 P-labeled 5'-probe (1-540 bp) and 3'-probe (541-1700 bp) of the NaAGP1 cDNA respectively.

FIG. 1K-2: Total RNA (10 .mu.g/lane) isolated from suspension-cultured cells of N. alata and N. plumbaginafolia was blotted and hybridized with the NaAGP1 cDNA.

The size of RNA transcripts is indicated at the right.

FIGS. 2A-2D present a flow chart describing the isolation and sequencing of AGP peptides from cell suspension culture filtrates of Nicotiana plumbaginafolia. FIG. 2A indicates that four sequences (SEQ ID NOS:26, 27, 28 and 29, respectively) were obtained from Peak RT21 and a sequence (SEQ ID NO:30) was obtained from peak RT32.

FIGS. 3A-3F present a flow chart describing the isolation and sequencing of AGP peptides from cell suspension culture filtrates of Pyrus communis. FIG. 3B indicates the resolution of Peak RT16.4 into three sequences (SEQ ID NOS:31, 32 and 33, respectively). FIG. 3C present the resolution of Peak RT18.2 into six sequences, namely, SEQ ID NOS:34, 35, 36, 37, 38 and 31, respectively. FIGS. 3E and 3F presents the resolution of the deglycosylated Peak RT7.8 into six peptides having sequences SEQ ID NOS:39-44, respectively.

FIG. 3G presents the nucleotide and derived amino acid sequences (SEQ ID NO:48) of the 350-bp PCR fragment. The derived amino acid sequence matching the peptide sequence by protein sequencing is underlined. The nucleotide sequence corresponding to the PcA23F2a primers is double-underlined.

FIGS. 3H-1 and 3H-2 present the nucleotide and predicted amino acid sequences (SEQ ID NO:49) of PcAGP23 cDNA clone encoding an AGP backbone from pear cell suspension culture. The translational initiation and stop sites are in bold-face. The predicted secretion signal is underlined with dots. The two potential N-glycosylation sites are double-underlined. The sequence matching the peptide sequences obtained from the AGP protein backbone are underlined. The proline residues which are hydroxylated, as identified by protein sequencing, are indicated by an "O" underneath.

FIGS. 4A-4C present a flow chart describing the isolation and sequencing of AGP peptides from style extract of Nicotiana alata. Peak RT25 comprised four sequences (SEQ ID NOS:50, 51, 52 and 53) and Peak RT35 also comprised four sequences (SEQ ID NOS:54, 55, 56 and 57).

FIG. 4D presents the cloning strategy of the Na35.sub.-- 1 gene.

FIG. 4E presents the nucleotide sequence (SEQ ID NO:62) of the PCR fragment by using RT35-specific primer and the predicted amino acid sequences. The derived amino acid sequence corresponding to the peptide sequence by protein microsequencing is underlined. The RT35 specific primer sequence is double underlined.

FIGS. 4F-1 and 4F-2 present the nucleotide sequence of NA35.sub.-- 1 cDNA clone and the predicted amino acid sequences (SEQ ID NO:63). The derived amino acid sequence corresponding to the peptide sequence by protein microsequencing is underlined.

FIG. 4G presents northern blot analyses of the NA35.sub.-- 1 gene expression in various parts of N. alata. Total RNAs from N. alata styles (S.sub.2 S.sub.2, S.sub.3 S.sub.3, S.sub.6 S.sub.6 ; 10 .mu.g each), leaves (S.sub.6 S.sub.6, 10 .mu.g), stems (S.sub.6 S.sub.6, 10 .mu.g) and roots (S.sub.6 S.sub.6, 6.3 .mu.g) were fractionated on a formaldehyde agarose gel, transferred to a nylon membrane, and hybridized with .sup.32 p labeled NA35.sub.-- 1 probe. The size of the RNA transcripts is indicated in kilo nucleotides.

FIG. 4H presents northern blot analyses of the NA35.sub.-- 1 gene expression in various suspension-cultured cells and plants. Total RNAs (10 .mu.g/lane) isolated from suspension cultured cells of N. alata and N. plumbaginafolia, Pyrus, and styles of N. alata (S.sub.6 S.sub.6) and L. peruvianum were blotted and hybridized with the NA35.sub.-- 1 probe. The size of the RNA transcripts is indicated in kilo nucleotides.

FIG. 4I presents reversed phase HPLC (RP-HPLC) separation of thermolysin cleavage products of the RT25 protein backbone. RT25 protein backbone (5-10 .mu.g) was digested with thermolysin and loaded onto an RP-300 column (2.1.times.100 mm, C8, ABT) equilibrated in 0.1% TFA at 1 ml/min. Unbound material was collected and bound material eluted with a linear gradient (0-60% acetonitrile in 0.1% TFA; 60 min; 100 .mu.l/min). Peptides (P1-6) eluted from the column were monitored at A.sub.215nm. Thermolysin was eluted after retention time 40 min. Individual peptides were subjected to amino acid sequencing.

FIG. 4J presents reversed phase HPLC separation of endoproteinase Asp-N cleavage products of the RT25 protein backbone. RT25 protein backbone was digested with endoproteinase Asp-N. The resulting peptides were loaded onto an RP-300 column (2.1.times.100 mm, C8, ABI) equilibrated in 0.1% TFA at 1 ml/min. Unbound material was collected and bound material eluted with a linear gradient (0-60% acetonitrile in 0.1% TFA; 60 min; 100 .mu.l/min). Peptides eluted from the column were monitored at A.sub.215nm. Peptides, A1 and A2 were subject to amino acid sequencing. Undigested starting material (RT25) was also detected.

FIGS. 4K-1 and 4K-2 present nucleotide and deduced amino acid sequences (SEQ ID NO:72) of the AGPNal 1 cDNA clone. The putative secretion signal (dot underlined) was predicted by using the PSIGNAL program (PC/Gene software, IntelliGenetics) based on the method described by Von Heijne (1986) Nucl. Acids Res. 14:4683-4690. Internal peptide sequences from amino acid sequencing are indicated by solid underlines and Hyp is shown encircled. Dash (-) indicates the stop codon.

FIG. 4L presents a hydropathy plot of the deduced amino acid sequence from the AGPNal 1 cDNA clone. The hydrophobicity of the deduced amino acid sequence was calculated by the SOAP program (PC/Gene software, IntelliGenetics) based on the method developed by Kyte and Doolittle (1982) J. Mol. Biol. 157:105-132. The putative secretion signal (shadowed) was predicted by using the PSIGNAL program (PC/Gene software, IntelliGenetics) based on the method described by Von Heijne (1986) supra.

FIGS. 4M-1 and 4M-2 present an RNA blot analysis of expression of the AGPNal 1 gene in N. alata and other plants. Total RNA (10 .mu.g/lane) isolated from (FIG. 4M-1) tissues of N. alata (genotype S.sub.6 S.sub.6): style, ovary, petal, anther, stem, leaf and root; and (FIG. 4M-2) styles of N. alata, N. sylvestris, N. tabacum, N. glauca, L. peruvianum and leaves of Arabidopsis and rye grass were run in a 2% agarose gel (15% formaldehyde; 40 mM MOPS buffer, pH 7.0) and blotted onto a Hybond-N nylon membrane (Amersham). AGPNal 1 cDNA fragment was labeled to 10.sup.8 cpm/.mu.g with .sup.32 P-dCTP. Hybridization was performed at 60.degree. C. overnight in 0.22M NaCl, 15 mM NaH.sub.2 PO.sub.4, 1.5 mM EDTA, 1% SDS, 1% BLOTTO and 4 mg/ml herring sperm DNA. The membrane was washed for 2.times.10 min., at room temperature, in 2.times.SSC, 1% SDS; 2.times.10 min., 60.degree. C., in 0.2.times.SSC, 1% SDS.

FIGS. 4N-1 and 4N-2 present an SDS-PAGE analysis of N. alata style AGPs at various stages of purification. SDS-PAGE (10% gel) followed by (N-1) silver staining and (N-2) staining with .beta.-glucosyl Yariv reagent. Lane 1, total style extract (1 .mu.g AGP). Lane 2, 95% (NH.sub.4).sub.2 SO.sub.4 -supernatant (4 .mu.g AGP). Lane 3, Mono Q-bound AGP-containing fraction (4 .mu.g AGP). Lane 4, Superose 6 AGP-containing fraction (4 .mu.g AGP). Lane 5, as Lane 3, but containing 20 .mu.g AGP. Lane 6, as Lane 4, but containing 20 .mu.g AGP. Protein molecular weight markers (M) are shown on the left.

FIGS. 4O-1 through 4O-4 present crossed-electrophoresis of AGPs from styles of N. alata during fractionation. AGPs from (FIG. 4O-1) crude style extract, (FIG. 4O-2) 95% (NH.sub.4).sub.2 SO.sub.4 -supernatant, (FIG. 4O-3) Mono Q-unbound AGP-containing fraction, and (FIG. 4O-4) Mono Q-bound fraction were first electrophoresed in a 1% agarose gel horizonatally then vertically into a gel containing the .beta.-glucosyl Yariv reagent.

FIGS. 5A-1 and 5A-2 present the nucleotide and predicted amino acid sequences of PcAGP9 (SEQ ID NO:66) encoding the protein backbone of an AGP from Pyrus communis cell suspension culture. The putative secretion signal peptide is underlined with dots. The sequences which match the peptide sequences obtained by protein sequencing are underlined. The proline residues which are modified post-translationally to hydroxyprolines are indicated by "O" underneath. X: undetermined residue.

FIGS. 5B-1 and 5B-2 present northern blot analyses of the PcAGP9 gene. FIG. 5B-1: Total RNA was isolated from pedicels (1) and cultured cells (2) of Pyrus communis; cultured cells of Nicotiana plumbaginafolia (3), shoots of Brassica napus (4), Arabidopsis thaliana (5) and Lycopersicon esculentum (6) and leaves of Lolium temulentum (7). Equal amounts (10 .mu.g/lane) of RNA were fractionated on formaldehyde agarose gels, transferred to Hybond-N membranes, and hybridized with .sup.32 P-labeled PcAGP9 cDNA at 55.degree. C. The final wash was carried out at 55.degree. C. for 30 min with 1.times.SSC+0.1% SDS. FIG. 5B-2: The same RNA blot was hybridized and washed at higher stringency (65.degree. C.). The size of the PcAGP9 RNA transcript in Pyrus communis cultured cells is indicated at the left.

FIG. 5C presents a hydropathy plot of the deduced amino acid sequence of PcAGP9 (SEQ ID NO:66). The hydropathy values of each amino acid have been determined by using an interval of five amino acids according to Kyte and Doolittle (1982) supra. Values above the dotted line indicate hydrophobic regions and values below the dotted line represent hydrophilic regions.

FIGS. 5D-1 and 5D-2 present a flow chart of the separation of AGPs from Pyrus communis (pear) cell suspension culture and the isolation of their protein backbones.

A. RP-HPLC (RP-300 column, 4.6.times.100 mm) profile of AGPs prepared by precipitation with the .beta.-glucosyl Yariv reagent. AGPs were loaded and the column washed with solvent A (0.1% TFA in H.sub.2 O). The unbound fraction was collected (not shown). The bound material was eluted with a linear gradient (0-100% solvent B; flow rate 1 ml/min; 60 min) (solvent B: 60% acetonitrile in solvent A). Individual fractions from five separate runs were pooled for subsequent purification.

B. RP-HPLC (RP-300 column, 4.6.times.100 mm) profile of AGPs from the major bound peak shown in A (retention time 5.0-10.57 min). Bound material was eluted with a shallow gradient (0-15% solvent B; flow rate 1 ml/min; 60 min). Two fractions (1 and 2) were separately collected and subjected to size-exclusion FPLC.

C. Superose-6 FPLC profiles of AGPs in the unbound fraction from A and two eluted fractions from B. Samples were eluted in 25% acetonitrile, 0.2M KCl, 5 mM KH.sub.2 PO.sub.4 (flow rate 0.4 ml/min). The unbound fraction and Fraction 1 gave single peaks; Fraction 2 resolved into two peaks (Peak 2A and 2B).

D. Superdex-75 FPLC profiles of protein backbones derived from AGPs in C by HF deglycosylation. Samples were eluted in the same buffer used in C (flow rate of 0.8 ml/min). The size of the protein was estimated from standard protein markers (Pharmacia).

The x axis is retention time (min). The pathway for purification of the AGP fractions, from which peptide sequences were obtained, is stippled.

FIGS. 5E-1 and 5E-2 present the nucleotide and predicted amino acid sequence of PcAGP2 cDNA (SEQ ID NO:91) encoding a putative AGP backbone from suspension cultured cells of P. communis. The translational initiation and stop sites are in bold-face. The predicted secretion signal is underlined with dots. The two long direct repeats are double-underlined. The sequence matching the peptide sequences obtained from the AGP protein backbone are underlined. The proline residues modified to HYP are indicated by an "O."

DETAILED DESCRIPTION OF THE INVENTION

The following definitions are provided in order to provide clarity as to the intent or scope of their usage in the specification and claims.

The term arabinogalactan protein or AGP as used herein refers to a Yariv reagent-precipitable, glycosylated molecule in which the protein constituent typically accounts for approximately 2 to 10% of the molecular weight of the molecule [although AGPs having protein values outside this range are known (Anderson et al. (1979) supra)] and in which carbohydrate usually accounts for most of the weight of the molecule. Galactose and arabinose form the major carbohydrate constituents with other monosaccharides and uronic acids as minor components; the galactosyl residues are organized to form a backbone of 3-linked galactose with branches through c(0)6; the arabinosyl residues are predominantly in terminal positions. AGPs specifically bind to and are precipitated by .beta.-glycosyl-Yariv reagents as a red colored complex. AGPs usually comprise a domain(s) enriched in hydroxyproline, alanine, serine, and threonine.

The term Yariv reagent-precipitable as used herein refers to an AGP that is capable of being precipitated by .beta.-glucosyl-Yariv reagents.

The term native AGP as used herein refers to an AGP in its native state, i.e., glycosylated.

The term glycosylated AGP as used herein refers to an AGP molecule comprising a carbohydrate component.

The term deglycosylated AGP as used herein refers to a native AGP or a glycosylated AGP which has been subjected to treatment for removal of carbohydrate and as a result of which contains a decreased but variable carbohydrate content.

The term nonglycosylated AGP or AGP backbone as used herein refers to a protein skeleton or backbone of an AGP molecule which is not glycosylated.

The term synthetic AGP as used herein refers to an AGP molecule which is chemically synthesized.

The term synthetic nonglycosylated AGP as used herein refers to a peptide backbone of an AGP which is chemically synthesized.

The term enriched in hydroxyproline or hydroxyproline-enriched or hydroxyproline-rich as used herein refers to a region or domain or segment of an amino acid sequence that has a hydroxyproline content of greater than 15%, and usually about 50% or greater.

The term OAST-enriched or high content of hydroxyproline, alanine, serine, and threonine or enriched in OAST content as used herein refers to a region of an amino acid sequence wherein the sum of the hydroxyprolyl, alanyl, seryl, and threonyl residues constitutes at least about 35%, and preferably at least about 60%, of the total amino acyl residues.

The term hydroxyproline-poor or not enriched in hydroxyproline as used herein refers to a region or domain of a peptide sequence that has a hydroxyproline content that is preferably less than 15%, more preferably less than 10% and most preferably less than 5%. A hydroxyproline-poor region may also have an OAST content that is preferably less than 50%, more preferably less than 35% and most preferably less than 20%.

The term a characterizing sequence or a sequence characterizing an AGP as used herein refers to a sequence that is hydroxyproline-rich and/or sequences that are enriched in OAST content.

The term a guessmer as used herein refers to an oligonucleotide that contains only a subset of the possible codons at each position. Guessmer is a term used routinely in the art and is thoroughly elucidated in Molecular Cloning, A Laboratory Manual, J. Sambrook, E. F. Fritsh and T. Maniatis, 2nd edition. Cold Spring Harbor Laboratory Press, 1989, pp. 11.11-11.16. In many cases, a guessmer is a chemically-synthesized, single oligonucleotide, 30-70 nucleotides in length, that contains the combination of codons most likely to match the authentic gene.

The term antisense RNA probe as used herein refers to a RNA strand produced from a DNA template encoding a desired amino acid sequence. The nucleotide sequence of the RNA is complementary to the coding strand of the DNA template sequence.

The term substantially pure as used herein refers to a protein that is substantially free of other proteins with which it is associated in nature.

The isolation of AGP genes from N. alata, N. plumbaginafolia, and P. communis suspension cultures and N. alata styles, as illustrated herein, exemplifies Applicants' invention which embraces the utilization of an amino acid sequence of a region of an AGP peptide from a plant cell to isolate a corresponding plant AGP gene. Not all regions or domains of AGP peptide sequences can be used equivalently to produce viable oligonucleotide primers for the isolation of AGP genes. Applicants succeeded in isolating AGP genes by using two different strategies:

(A) the use of a non-hydroxyproline-rich sequence as a primer template to obtain a corresponding AGP gene, and

(B) the use of a guessmer-antisense RNA probe approach wherein the guessmer can comprise a nucleotide sequence encoding a hydroxyproline-rich segment to obtain an AGP gene encoding the sequence of the hydroxyproline-rich segment.

In strategy A, the preferred sequences are those that have a low content, or are deficient in, hydroxyproline. Hydroxyproline-poor sequences are found in terminal regions as well as in internal domains of AGP peptides. It is also preferable that sequences of AGP peptides or fragments thereof, selected for synthesis of synthetic oligonucleotide primers, have a low hydroxyproline, alanine, serine, and threonine (OAST) content. It is particularly preferable that the content of the sum of these four amino acid residues be less than 50%, and more preferably less than 35%, and most preferably less than 20%, of the total amino acid residues. AGP sequences that are useful in isolating an AGP gene using PCR technology, i.e., sequences that are hydroxyproline-poor, or OAST-poor, are not available in the prior art.

The amino acid sequence selected as a template for the synthesis of an oligonucleotide primer should not be one that gives PCR degenerate primers having concentrated "GC-rich" regions. Primers having concentrated "GC-rich" sequences frustrate and make futile the attempts to obtain cDNA by the PCR technique. For example, AGP peptide fragments published in the art are the following:

from carrot (Jermyn, 1985, supra)* ##STR1## from ryegrass, Lolium multiflorum (Gleeson et al., 1989, supra)* ##STR2## and from rose (Komalavilas et al., 1990, supra)* ##STR3##

Although these amino acid sequences of AGP peptide fragments from carrot, ryegrass, and rose are known in the art, AGP genes corresponding to these peptide fragments are still not known in the art. All of these art-known plant AGP peptide fragments have amino acid sequences that are characterized by a high content of hydroxyproline, alanine, serine, and threonine. These amino acid partial sequences are such that they give GC-rich oligonucleotide primers. For this reason, no one to date has been successful in obtaining AGP cDNAs directly from these sequences.

Initially, Applicants attempted to obtain plant AGP genes using hydroxyproline-rich sequences obtained from isolated AGP fragments. The following sequences were utilized unsuccessfully: ##STR4## Each of the sequences (i), (ii) and (iii), which are found in both N. alata and N. plumbaginafolia AGPs, were used in both N. alata and N. plumbaginafolia to isolate an AGP gene. Sequence (iv) was used to obtain an AGP gene from P. communis. None of these sequences led to the isolation of a corresponding gene. All of these sequences produced oligonucleotide primers that were highly redundant and very GC-rich (in some cases greater than 80%). Consequently, a problem appeared that at high stringency, hybridization bands were obtained which, on sequencing, had no relationship to the amino acid sequence. On examination of the above sequences, it may be seen that all four of these sequences are OAST-enriched, i.e., (i) 50%, (ii) 85.7%, (iii) 64.3%, and (iv) 84.2%, respectively.

The instant disclosure overcomes this problem. Whereas isolated plant AGPs in the art have been characterized exclusively by peptide fragments having high hydroxyproline or OAST contents (AGP sequences having a low content of hydroxyproline, or a low OAST content are not available in the prior art), the AGPs isolated and described in Applicants' disclosure are characterized not only by peptide fragments that are hydroxyproline-rich but also by peptide fragments that are hydroxyproline-poor, if not hydroxyproline-deficient. The fact that an AGP peptide fragment that was not enriched in hydroxyproline had been isolated and sequenced and the fact that this sequence, which is also low in hydroxyproline, alanine, serine, and threonine content, had been utilized to synthesize degenerate primers, enabled circumvention of the problems associated with GC-rich primers and led to the isolation of a corresponding AGP cDNA.

The N-terminal region of an isolated plant AGP can be used to obtain a corresponding plant AGP gene. In a particular embodiment of the invention, the N-terminal region of an AGP peptide obtained from N. alata suspension culture comprised a hydroxyproline-poor region. The N-terminal peptide sequence, A-K-S-K-F-M-I-I-P-A-S-X-T-X-A (SEQ ID NO:11) was used as a template for the synthesis of an oligonucleotide primer which was further utilized for the isolation of a hybridizing AGP gene from both N. alata and N. plumbaginafolia.

In other specific embodiments of the invention, hydroxyproline-poor sequences from internal regions of AGPs from P. communis suspension culture and from N. alata style were used to obtain corresponding AGP genes. For example, in the case of P. communis, the AGP backbone encoded by the PcAGP23 gene (SEQ ID NO:49) is hydroxyproline-poor not only at the terminal regions but also internally, and an internal sequence (SEQ ID NO:41) was used to obtain a pear AGP gene. Similarly, for the N. alata style AGP backbone encoded by the Na35.sub.-- 1 cDNA clone (FIGS. 4F-1 and 4F-2) the N-terminal region and internal regions have low hydroxyproline contents, and, internal sequence (SEQ ID NO:58) was used to obtain an N. alata style AGP gene.

This basic approach (strategy A) for obtaining a plant AGP gene enabled the successful isolation of AGP genes from N. alata, N. plumbaginafolia, and P. communis cell suspension cultures, as well as from N. alata styles. In each case, the cDNA clone comprised a derived amino acid sequence which contained a hydroxyproline-poor domain and a hydroxyproline-enriched domain (a region enriched in OAST content).

In strategy B, a method is provided that enables the use of a specific OAST-rich AGP peptide sequence for the isolation of a corresponding gene. This method involves the screening of libraries with RNA probes prepared from a single long guessmer (oligonucleotides containing only a subset of the possible codons at each position) encoding a desired specific OAST-rich AGP peptide sequence. In order to produce an RNA probe from a DNA oligonucleotide, a bacteriophage promoter (e.g., T7 or T3 RNA polymerase promoter) is linked at the 5'-end of the oligonucleotide. In addition, the oligonucleotide, which is single-stranded, must be converted into a partial or complete, double-stranded DNA fragment, because the T7 (or T3) RNA polymerase will not recognize single-stranded promoter sequences. Relevant procedures for obtaining either DNA or RNA probes from a DNA template are known in the art [Berger and Kimmel (1987) Methods in Enzymology 152].

FIG. 1 presents schematically several ways of producing an RNA probe involving the use of a single (FIG. 1A) or double (FIG. 1B) oligonucleotide probe. For example, in FIG. 1A-1 a second oligonucleotide, which is complementary to the guessmer encoding a desired AGP peptide, is synthesized and the two oligonucleotides are annealed to form double-stranded DNA. Alternatively, as shown in FIG. 1A-2, a short complementary primer is annealed to the promoter sequence of the guessmer to form a double-stranded RNA polymerase promoter sequence. Using a double oligonucleotide probe approach (FIG. 1B-1), an adaptor sequence (15-18 bp long) is added to the 3'-end of the guessmer (oligonucleotide 1) and a second guessmer (oligonucleotide 2), which encodes a different OAST-rich AGP peptide sequence, with an adaptor sequence complementary to the adaptor of the first oligonucleotide, is synthesized. The two guessmers are thus annealed through their complementary adaptor sequences and the protruding single-stranded regions filled in by primer extension to produce a double-stranded DNA fragment. FIG. 1B-2 further demonstrates a method whereby the adaptor sequences are designed in such a way that they bind to opposite strands of a mediator DNA, enabling the two guessmers to be joined together by a PCR reaction to form a double-stranded DNA fragment. The double oligonucleotide probe can be used to screen a library for two AGP genes simultaneously.

Single-stranded RNA probes are superior to DNA probes for the screening of libraries. RNA probes can be labeled to much higher specific activity and bind more tightly to a target DNA, thus yielding stronger signals in hybridization reactions. The greater stability of hybrids involving RNA enables the use of higher hybridization stringency, thus increasing hybridization specificity. Unhybridized RNA probes can be removed by RNase digestion further reducing the background.

A single long guessmer (40-70 bp) rather than short degenerate oligonucleotides is used to avoid the extremely high degeneracy associated with OAST-rich AGP peptide sequences. It is preferred that the guessmer be longer than 40 bp in order that the increased stability of hybrids formed by the long oligonucleotide out-weigh the detrimental effects of mismatches. Anti-codons GGU, CGU, UGU, and AGU should be used for Pro (Hyp), Ala, Thr, and Ser, respectively. This is based on the consideration that the nucleotide base "A" is the preferred base in the third position of codons for Pro, Ala, and Thr. The other consideration is that the nucleotide base "U" can pair not only with "A" but also with "G" to some extent, hence GGU can pair with CCA or CCG for proline residues, for example. Therefore, it is further contemplated that antisense RNA rather than the sense RNA probes be used for the screening of libraries.

AGP peptides were isolated from plant cell suspensions by precipitation with Yariv reagent (a red dye, .beta.-glucosyl reagent described by Yariv in 1967). This dye was prepared by coupling diazotized 4-aminophenyl glucopyranoside to phloroglucinol and the reagent was used to precipitate AGPs. The AGPs from suspension-cultured cells were prepared by precipitation of AGPs from either the culture medium or from the Biopolymer products (the high molecular weight materials precipitated with four volumes of ethanol from a cell suspension culture filtrate). An isolation procedure independent of the Yariv reagent was also used to obtain AGPs from plant cells. (The Yariv reagent was used later in the isolation procedure to identify fractions containing AGPs.) The AGPs from N. alata style extracts were prepared by (NH.sub.4).sub.2 SO.sub.4 precipitation and further fractionation of the AGP-containing supernatant by Mono Q (Pharmacia) anion-exchange chromatography. In a different procedure, AGPs were initially fractionated by immunoaffinity chromatography using the J539 myeloma antibody (specific for Gal 1-6 .beta. Gal sequences).

As is known in the art, AGPs can be isolated by several methods, including affinity chromatography using, for example, galactose binding proteins, classical chromatography, for example, gel filtration, ion-exchange, etc., and also precipitation by selective reagents, for example, Yariv reagents, lectins, for example, lectins that binds galactosyl residues, including but not limited to, tridacnin, peanut agglutinin, the Ricinus communis (RCA.sub.120) lectins and myeloma protein J539 [Clarke et al. (1979) Phytochemistry 18:521-540; Fincher et al. (1983) Ann. Rev. Plant Physiol. 34:58], or antibodies to specific carbohydrate epitopes [Pennell et al. (1989) J. Cell Biol. 108:1967-1977 and Norman et al. (1990) Planta 181:365-373].

AGP fractions were deglycosylated by treatment with trifluoromethane sulfonic acid (TFMS) or by treatment with anhydrous hydrogen fluoride (HF). Additionally, other methods for separating the protein and the carbohydrate components from each other that are known in the art are contemplated by the invention [see Jermyn et al. (1975) Aust. J. Plant Physiol. 2:501].

AGPs and AGP fragments, glycosylated or deglycosylated, were separated by known separation techniques, for example, SDS-PAGE, HPLC reverse phase chromatography, etc. In some cases, the peptides were further fragmented by thermolysin digestion before separation. Separated peptides obtained off HPLC reverse phase and ion-exchange columns were sequenced directly, although in some cases the separated peptides were transferred to PVDF membranes for amino acid sequencing [Ward et al. (1990) in Electrophoresis 11:883-891]. The use of other known proteases, instead of or in addition to thermolysin, is contemplated by this invention. Similarly, this invention contemplates the use of other techniques known in the art for the preparation of pure peptide samples for amino acid sequencing.

From every source examined, multiple AGP peptides were observed. Multiple peptides were reproducibly obtained whether the AGPs were separated first and then individually deglycosylated or whether the whole AGP preparation was deglycosylated first and then the individual peptides separated.

In a specific embodiment of the invention, total native AGPs were isolated by Yariv reagent precipitation from the suspension culture filtrate of N. alata and deglycosylated using TFMS. The resulting peptides were separated on a 17.5% SDS-PAGE gel and blotted to a PVDF membrane. The major band (MW: 20-30 kD; FIG. 1C) was excised and sequenced. An N-terminal peptide sequence, A-K-S-K-F-M-I-I-P-A-S-X-T-X-A (SEQ ID NO:11), was obtained.

In a particular embodiment of the invention, the N. alata AGP N-terminal peptide sequence (SEQ ID NO:11) was used to isolate AGP genes from N. alata and N. plumbaginafolia libraries (FIGS. 1D-1 and 1D-2). Degenerate reverse primers corresponding to part of the AGP N-terminal amino acid sequence, i.e., K-F-M-I-I-P were synthesized (Table 1.1) and used to obtain a 160-bp primer extension product (FIG. 1E) which was then amplified by PCR. The 160-bp extension

TABLE 1.1 __________________________________________________________________________ A:Oligo primers used in the primer extension experiments AlaLysSerLysPheMetIleIleProAlaSerXThrXAla(SEQ ID NO:11) ##STR5## B:Oligonucleotide primers designed ##STR6## ##STR7## C:Subgroups of the group 1 primers NaR15'GGT/G/AATGATCATAAACTT3'(SEQ ID NO:15) NaR25'GGT/G/AATAATCATAAACTT3'(SEQ ID NO:16) NaR35'GGT/G/AATTATCATAAACTT3'(SEQ ID NO:17) NaR45'GGT/G/AATGATCATGAACTT3'(SEQ ID NO:18) NaR55'GGT/G/AATAATCATGAACTT3'(SEQ ID NO:19) NaR65'GGT/G/AATTATCATGAAGTT3'(SEQ ID NO:20) __________________________________________________________________________ A: Amino acid sequence obtained from deglycosylated AGPs isolated from N. alata cell suspension culture and the corresponding codons. B: The two groups of degenerate reverse primers designed for the primer extension experiment. C: Subgroups of the group 1 primers.

fragment was subcloned and sequenced. The nucleotide sequence (SEQ ID NO:21) included a derived peptide which matched with the peptide sequence SEQ ID NO:11 isolated from N. alata suspension culture.

Additional primers, corresponding in sequence to parts of the 160-bp fragment (e.g., NaF1 and NaF2; FIG. 1E), were synthesized and used to amplify the 3'-part of the AGP gene by nested PCR. A 1.6 kb fragment was amplified and sequenced. The alignment of the sequences obtained from the two PCR reactions gave rise to a DNA sequence of 1679 bp (FIGS. 1F-1 and 1F-2). The PCR fragment encoded a protein which contained the isolated peptide sequence (SEQ ID NO:11) with two mismatches: Arg for Ala at position 1 and Pro for His at position 12 (FIGS. 1F-1 and 1F-2).

The 1.6 kb PCR fragment was used to screen a cDNA library made from RNA isolated from N. alata cells in suspension culture and three positive clones were isolated and sequenced. The alignment of the PCR sequences with the cDNA sequence gave rise to a 1700-bp sequence (SEQ ID NO:24) including a poly(A) tail of 7 bp (FIGS. 1F-1 and 1F-2). This sequence was designated NaAGP1. Further primer extension experiments suggested that the 1.7 kb NaAGP1 cDNA (SEQ ID NO:24) represented the full-length sequence of the AGP transcript.

The NaAGP1 cDNA comprised an open reading frame spanning 1383 nucleotides. The open reading frame encoded a polypeptide containing 461 amino acid residues with a calculated molecular weight of 51.8 kD and a predicted pI of 3.84. The protein was highly rich in asparagine (25%), and relatively rich in serine (8.9%), tyrosine (7.5%), proline (7.2%) and glutamine (7.0%) (Table 1.2), and could be divided into four domains (FIG. 1G). At the N-terminus (residues 1-25), there was a putative transmembrane helix which was very hydrophobic.

TABLE 1.2 __________________________________________________________________________ Comparison of derived amino acid composition of NaAGP1 and NpAGP1. Full Pro-rich Asn-rich Amino sequence (Mol %).sup.1 domain (Mol %).sup.2 domain (Mol %).sup.3 acid NaAGP1 NaAGP1 NaAGP1 NaAGP1 NaAGP1 NaAGP1 __________________________________________________________________________ Asn 25.0 26.2 4.7 3.3 44.1 43.4 Ser 8.9 9.8 8.7 9.4 9.8 10.3 Tyr 7.5 7.7 1.3 1.3 12.1 11.9 Pro 7.2 7.9 20.2 20.8 0.0 0.3 Glu 7.0 7.7 6.7 6.7 5.7 6.3 Gly 6.0 5.4 6.7 6.0 6.0 5.5 Phe 5.8 4.7 6.0 6.7 3.8 3.9 Thr 5.4 4.5 10.8 10.7 1.5 1.1 Asp 3.9 3.1 4.7 5.4 3.8 1.9 Ala 3.5 4.1 8.7 8.7 1.5 1.5 Leu 3.3 2.9 5.4 4.0 1.5 1.5 Val 3.3 3.1 4.7 4.0 2.2 2.3 Gln 3.1 2.9 2.7 3.3 1.9 1.5 Ile 2.7 2.9 4.0 4.7 0.7 1.1 Lys 2.5 2.5 2.0 1.3 2.2 3.1 Arg 1.6 1.5 1.3 1.3 1.5 1.5 Met 1.2 1.1 0.6 0.6 0.7 0.7 His 0.8 0.6 0.6 0.6 0.3 0.7 Cys 0.4 0.0 0.0 0.0 0.0 0.0 Trp 0.0 0.0 0.0 0.0 0.0 0.0 __________________________________________________________________________ .sup.1 The NaAGP1 derived amino acid sequence is incomplete as the clone is approximately 100 bp short. .sup.2 The prolinerich domain is defined by amino acid residues 26173 in NaAGP1 and 14161 in NpAGP1. .sup.3 The Asnrich domain is defined by amino acid residues 174436 in NaAGP1 and 162412 in NpAGP1.

The next one-third of the protein (residues 26-173) was also hydrophobic and contained most of the proline (93.8%), alanine (76.5%) and threonine (76.2%) residues. These three amino acids accounted for 39.7% of all the amino acids in this domain (Pro, 20.2%; Thr, 10.8% and Ala, 8.7%) (FIG. 1G). This domain is predicted to be the site of glycosylation by Gal/Ara containing chains, linked through hydroxyproline residues. The proline residues (residues 37, 39, 41, and 43 in FIGS. 1F-1 and 1F-2) are known to be hydroxylated, as they appear as hydroxyproline (residues 25, 27, 29, and 31 in FIGS. 1H-1 and 1H-2) in the peptide sequence obtained from deglycosylated AGPs of N. plumbaginafolia. Such hydroxylation and glycosylation would make the molecule considerably more hydrophilic.

The portion of the protein corresponding to amino acid positions 174-436 was hydrophilic and contained most of the asparagine (95.1%) and tyrosine (94.1%) residues which accounted for 44.1% and 12.1%, respectively, of all amino acids in this domain (FIGS. 1F-1 and 1F-2 and FIG. 1G). The asparagine residues were distributed in clusters (residues 2-10) along the polypeptide chain. This domain contained no proline residues. The final 25 residues at C-terminus were hydrophilic (FIG. 1G).

An N. plumbaginafolia cell suspension cDNA library was also screened with the PCR fragment, and four cDNA clones were isolated and sequenced. The four clones were identical and contained an insert of 1430 bp (SEQ ID NO:25; FIGS. 1H-1 and 1H-2). This AGP gene was designated NpAGP1. These cDNAs were incomplete and predicted to be about 100 bp shorter at the 5'-end than the full-length sequence of the transcript. The NpAGP1 was not identical, but very similar to the NaAGP1 at both the nucleotide and derived amino acid sequence level (86% and 84.7% identify, respectively) (FIGS. 1I-1 and 1I-2, FIGS. 1J-1 through 1J-4, and Table 1.2). The transmembrane helix was missing in the NpAGP1 cDNA due to the incomplete sequence. The difference between the two AGP genes was mainly in the middle one-third of the sequence while the N-terminal and C-terminal parts were highly conserved (FIGS. 1I-1 and 1I-2 and FIGS. 1J-1 through 1J-4).

The NaAGP1 cDNA was cut into a 5'-half (residues 1-540) corresponding to the 5'-nontranslated part, the transmembrane helix and the proline-rich domain and a 3'-half (residues 541-1700) including the asparagine-rich domain, C-terminus, and the 3'-nontranslated part. These two parts of the cDNA were used separately to probe northern blots of RNA [Sambrook et al. (1989) supra] isolated from suspension cultured cells of N. alata and N. plumbaginafolia and various tissues of N. alata plants. The two probes gave an identical hybridization pattern, confirming that these two distinct domains are parts of the same transcript (FIGS. 1K-1 and 1K-2). The NaAGP1 cDNA probes hybridized to the RNA samples from all the tissues of N. alata tested, although the degree of hybridization and size of transcripts are different in different tissues. The highest signal was detected in RNA from N. alata suspension cultured cells whereas the signal in petals is barely detected. Pollen and style tissues have a smaller transcript of approximately 1.0 kb compared with 1.6 kb in N. plumbaginafolia cultured cells and 1.7 kb in all other tissues (FIGS. 1K-1 and 1K-2). Genomic southern blot analysis indicated that the AGP gene is a single copy or low copy gene in the genome of N. alata.

In a preferred embodiment of the invention, the cDNA library was screened with the labeled synthetic oligonucleotide probe derived from the hydroxyproline-poor or the N-terminal AGP protein sequence. In an alternative embodiment of the invention, individual recombinants within the cDNA library can be screened for expression of an antigen (antibody recognition). Procedures for selecting cloned sequences from a recombinant cDNA library are described in Kimmel (1987) Meth. Enzymol. 152:393-399.

This invention also contemplates the use of oligonucleotide probes, e.g., AGP cDNA, etc., for the detection of hybridizing sequences and the isolation of monocot and dicot AGP genes. Pear AGP (PcAGP9) transcripts were detected in RNA prepared from dicots as well as from a monocot.

cDNA clones which show a strong hybridization signal are sequenced to confirm complimentarity to the AGP amino acid sequence. In addition, the protein encoded by the cDNA is shown to possess AGP characteristics. This is done, for example, by transcribing the clone sequence with an appropriate RNA polymerase, then translating the mRNA in, for example, a commercially available wheat germ extract in vitro translation system. Thus, the identity of a clone is confirmed by transformation into a suspension-cultured cell and identifying the product using a suitable tag.

In another embodiment of the invention, the presence of AGP protein is detected immunologically. For example, antibodies raised to an AGP peptide, or fragment thereof, purified and isolated from an SDS-PAGE gel are shown to cross-react with the purified AGP peptide. AGP-specific antibodies are also utilized to bind and precipitate AGP from plant extracts as well as the product of the cloned AGP gene. Polyclonal and monoclonal antibodies specific to AGP peptide are prepared according to standard methods in the art. This type of immunological testing is further utilized, for example, for optimization of expression of the cloned AGP gene in a recipient organism.

This invention further contemplates the isolation of a genomic clone of AGP. Genomic DNA is isolated according to the methods described by Herrmann and Frischauf (1987) Methods Enzymol. 152:180-189. A PCR-based method is used to clone a gene from genomic DNA using partial protein sequence (e.g., Aarts et al. (1991) Plant Mol. Biol. 16:647) or cDNA fragment probes (e.g., King et al. (1988) Plant Mol. Biol. 10:401-412). The genomic AGP gene may be utilized instead of the cDNA to express AGP, in particular, in host systems where it appears that the native promoter or post-translational system is required for full expression, e.g., plant monocot or dicot cells, mammalian cells, e.g., COS cells, etc.

As is well known in the art [see, for example, Glover (1984) Gene Cloning, Brammar and Edidin (eds.), Chapman and Hall, NY], there are various strategies for generation of cDNA libraries and for the cloning of the cDNA into an appropriate DNA recombinant vector, e.g., the pUC family of plasmids or .lambda.gt10 or .lambda.gt11 phage vectors. In an embodiment of the invention, a DNA recombinant vector carries a constitutive or inducible promoter adjacent to the cloning site such that a transcript is made specifically to either strand of the cDNA simply by using different RNA polymerases. RNAs produced in this way can be used as hybridization probes or can be translated in cell-free protein synthesis systems.

It is understood in the art that modifications may be made to the structural arrangement and specific elements of a genetically-engineered recombinant DNA molecule described herein without destroying the activity of gene expression. For example, it is contemplated that a substitution may be made in the choices of enhancer regulatory elements and/or promoters [e.g., preferably, an inducible promoter (e.g., AdH1)] without significantly affecting the function of the recombinant DNA molecule of this invention. It will also be understood that optimization of gene expression also results from the use of preferred codons, the arrangement, orientation, and spacing of the different regulatory elements as well as the multiple copies of a particular element with respect to one another, and with respect to the position of the TATA box, as will be apparent to those skilled in the art using the teachings of this disclosure.

In another embodiment of the invention, AGPs were isolated from N. plumbaginafolia suspension cultures. The medium from the cell suspension culture of N. plumbaginafolia was separated from the cells by filtration and the high molecular weight materials precipitated with four volumes of ethanol. The total native AGPs were purified from the Biopolymer product by precipitation with the Yariv reagent after depleting the starting material of pectins by CTAB (hexadecyl trimethyl ammonium bromide) precipitation prior to Yariv precipitation. The total native AGPs were treated by two paths:

Path 1: Deglycosylation followed by reverse phase HPLC fractionation before direct sequencing, or sequencing after enzymatic (proteolytic) digestion [detailed in Example 2(c)2-5].

Path 2: Reverse phase HPLC fractionation followed by deglycosylation and further reverse phase HPLC fractionation [detailed in Example 1(c) 6-8].

Path 1 (deglycosylation followed by separation of AGPs) produced an unbound peak and two major bound peaks, RT21 and RT32, with retention times of 21 min and 32 min, respectively, in reverse phase HPLC (see FIG. 2A). Peak RT21 was digested with thermolysin and refractionated by RP-HPLC prior to amino acid sequencing. The sequences (SEQ ID NOS:26-29) obtained from peak RT21 exhibited a high content of hydroxyproline, alanine, serine, and threonine (OAST-rich sequences).

Peak RT32 was sequenced directly and gave the sequence R-K-S-K-F-M-I-I-P-A-S-O-T-O-A-O-T-O-I-N-E-I-S-F (SEQ ID NO:30) which, at the 5'-end, very closely matched the N-terminal sequence (SEQ ID NO:11) obtained from N. alata cell cultures, and which did not show a high content of hydroxyproline nor of OAST, i.e., hydroxyproline, alanine, serine, and threonine. The 3'-end of the peak RT32 sequence (SEQ ID NO:30) comprised a domain characterized by a high OAST content.

The results of amino acid analyses of chromatographic fractions from N. plumbaginafolia AGPs are presented in Table 2.1.

TABLE 2.1 __________________________________________________________________________ Amino acid analyses of fractions from N. plumbaginafolia AGPs Fractions of N. plumbaginafolia AGPs after deglycosylation and RT-300 separation (FIG. 2A) Native N. plumbaginafolia AGPs fractionated on RP-300 (FIG. 2B) Unbound RT21 RT32 Unbound RT5 RT6 RT10 RT21-23 RT34 __________________________________________________________________________ Hyp 9.6 20.8 18.2 16.2 19.6 16.0 14.6 2.3 1.0 Asx 9.6 3.3 4.9 8.7 5.7 1.6 6.4 8.1 9.2 Thr 7.6 8.1 8.9 4.3 7.3 11.1 8.5 8.3 7.0 Ser 9.2 16.0 13.0 12.0 15.0 18.2 11.6 10.6 10.3 Glx 7.8 5.6 5.5 9.0 6.9 5.0 5.9 6.7 6.4 Pro 6.2 2.5 2.2 4.5 3.4 4.4 8.3 5.0 13.3 Gly 10.1 6.5 7.9 6.9 6.0 5.1 5.5 7.3 9.6 Ala 12.7 18.5 24.6 20.8 22.7 22.3 12.3 18.3 16.1 Cys 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.4 0.1 Val 4.5 4.5 6.7 2.4 4.2 6.0 4.8 7.7 4.8 Met 0.5 0.0 0.2 2.2 1.5 1.5 1.0 1.5 0.7 Ile 4.8 1.2 1.8 1.2 1.1 1.5 4.6 1.8 2.4 Leu 6.41 3.6 3.2 3.2 2.6 2.4 5.0 8.8 7.9 Tyr 0.0 0.0 0.0 0.0 0.0 0.3 0.4 0.0 1.6 Phe 7.0 1.9 1.2 3.3 2.3 1.5 5.7 4.4 3.3 His 1.5 1.5 0.9 0.3 0.2 0.7 1.4 1.2 1.4 Lys 1.3 5.2 0.7 0.9 1.2 2.1 2.5 4.8 3.5 Arg 1.3 0.8 0.0 0.9 0.4 0.3 1.6 0.9 2.1 Trp ND ND ND ND ND ND ND ND ND __________________________________________________________________________ Hyp: hydroxyproline, ND: not determined

All AGPs that initially bound to the chromatography columns showed an enrichment in hydroxyproline, alanine, serine, and threonine residues.

In another embodiment of the invention, the total native AGPs were isolated from Pyrus communis (pear) Biopolymer by Yariv precipitation.

The AGPs were either deglycosylated first and then separated by reverse phase HPLC (RP-300) (Path 1), or alternatively, the total native AGPs were fractionated first by reverse phase HPLC (RP-300), and then deglycosylated, digested with thermolysin, and purified for sequencing (Path 2).

Path 1 (HPLC separation of deglycosylated AGPs) gave the profile shown in FIG. 3A. The results of amino acid analyses of major peaks (i.e., unbound, peak RT16.4 and peak RT18.2), as summarized in Table 3.1, indicated enrichment of hydroxyproline, alanine, serine, and threonine residues in the bound fractions. The RT16.4 and the RT18.2 peaks from FIG. 3A were subjected to thermolysin digestion and the digestion products were separated on a RP-300 column. The RP-300 profile for digested RT16.4 is shown in FIG. 3B and for RT18.2 is shown in FIG. 3C.

In all, only one peak (peak 1 of thermolysin-digested RT16.4, FIG. 3B) was a pure peptide and gave a clear sequence, L-S-O-K-K-S-O-T-A-O-S-O-S-(S)-T-O-O-T-(T) (SEQ ID NO:31), which showed a high content of alanine, hydroxyproline, serine, and threonine. Peaks 3 and 5 of RT16.4 (FIG. 3B) comprised sequences (SEQ ID NO:11 and SEQ ID NO:12, respectively) that also exhibited high contents of hydroxyproline, alanine, serine, and threonine.

Peaks from thermolysin-digested RT18.2 (FIG. 3C) were resolved into several peaks (SEQ ID NOS:31, 34-38). These sequences also were characterized by a high OAST content.

Path 2 (fractionation of the total native pear AGP fraction by reverse phase HPLC) gave the profile presented in FIG. 3D. Peak RT7.8 and the unbound fraction were analyzed for amino acid composition and both were found to be enriched in hydroxyproline, alanine, serine, and threonine as shown in Table 3.1. Peak RT7.8 and the unbound fraction were deglycosylated and fractionated on HPLC. The profile for the deglycosylated Peak RT7.8 (FIG. 3E) showed a major peak (Peak RT23) which, after thermolysin digestion and further purification on reverse phase HPLC (RP-300), gave six peptide sequences. Five sequences (SEQ ID NOS:39-44) were OAST-enriched, whereas one of the sequences, L-V-V-V-V-M-T-P-R-K-H (SEQ ID NO:41) was also present in sequence obtained by direct sequencing of the native AGP in RT7.8.

The unbound fraction of FIG. 3D after deglycosylation and further fractionation on HPLC (Path 2), gave the profile presented in FIG. 3F. The major peaks RT16-19 in FIG. 3F [obtained by Path 2 (separation followed by deglycosylation)] had retention times similar to those of peaks RT16-19.9 in FIG. 3A [obtained by Path 1 (deglycosylation followed by separation)].

It would appear from FIG. 3D that Peak RT7.8 represents about 27% of the total AGPs from pear. At least four N-terminal were observed in one fraction which may represent multiple chains. The unbound fraction represents about 67% of the total AGPs from pear and gives peaks which correspond to the RT16.4-19.9 of FIG. 3A which gave several OAST-enriched sequences. Thus, the invention provides amino acid sequence data from each of the two major AGPs from Pyrus communis.

In a particular embodiment of the invention, an AGP gene was obtained from P. communis.

TABLE 3.1 __________________________________________________________________________ Fractions of pear AGPs Fractions of native AGPs RT7.8 after deglycosyla- after deglycosylation and produced by HPLC (RT-300) tion and RP-300 separa- RP-300 separation (FIG. 3A) fractionation (FIG. 3D) tion (FIG. 3E) Unbound RT16.4 RT18.2 Unbound RT7.8 RT23 __________________________________________________________________________ Hyp 17.68 24.5 24.4 20.8 17.2 23.0 Asx 3.2 3.1 2.9 6.0 4.0 9.7 Thr 7.9 9.3 10.8 8.4 7.0 8.7 Ser 10.3 22.1 17.6 16.4 10.5 20.1 Glx 7.1 4.9 3.6 8.1 6.5 14.0 Pro 4.8 1.3 1.8 3.0 5.2 1.0 Gly 7.7 2.3 4.0 4.1 6.7 2.0 Ala 15.8 19.7 21.6 19.4 17.3 15.1 Cys 0.0 0.0 0.0 0.0 0.0 0.0 Val 8.3 3.2 4.1 2.8 7.4 0.5 Met 0.5 0.0 0.0 0.1 0.6 0.0 Ile 2.8 0.2 0.8 1.7 3.0 0.1 Leu 4.4 1.7 1.9 2.3 4.7 0.2 Tyr 0.9 1.0 0.3 2.3 1.7 0.0 Phe 0.9 0.0 0.1 0.0 0.7 0.9 His 2.0 0.5 0.8 0.6 1.9 3.4 Lys 4.4 5.1 4.7 3.5 4.0 1.6 Arg 1.6 1.2 0.6 0.6 1.6 0.0 Trp ND ND ND ND ND ND __________________________________________________________________________ Hyp: hydroxyproline, ND: not determined

The sequence L-V-V-V-V-M-T-P-R-K-H (SEQ ID NO:41), which was hydroxyproline-poor and OAST-poor, was selected as template for obtaining an AGP gene from pear cell suspension culture.

A number of primers corresponding to the L-V-V-V-V-M-T-P-R-K-H sequence (SEQ ID NO:41) was designed and synthesized for PCR experiments (Table 3.2).

TABLE 3.2 __________________________________________________________________________ Sequences of the oligonucleotide primers used in PCR __________________________________________________________________________ Peptide sequenceL--V--V--V--V--M--T--P--R--K--H(SEQ ID NO:41) Primer designation: PcA23F15'GTNGTNGTNGTNATGAC3'(SEQ ID NO:45) ##STR8## PcA23F2b5'GTAGTNATGACNCCNCGNAA3'(SEQ ID NO:47) __________________________________________________________________________ N = A, T, G, or C

The same nested PCR procedure used for the cloning of the NaAGP1 gene (FIG. 1D-2) was used to clone the gene encoding the above peptide, except that the annealing temperature was 52.degree. C. in this case. A 350-bp fragment was amplified after two successive PCR reactions using the PcA23F1 as the first primer and the PcA23F2a as the second primer. The fragment was sequenced and found to encode the correct peptide sequence (SEQ ID NO:48; FIG. 3G).

The PCR fragment was used to screen a cDNA library made from mRNA from pear cell suspension culture, as described above for N. alata cell suspension. One positive clone (PcAGP23) was isolated and sequenced. This clone contained an insert of 760 bp and matched the PCR sequence.

The PcAGP23 cDNA (SEQ ID NO:49) encodes an open reading frame, which starts with an initiation codon (ATG) at position 20 and ends with a termination codon (TAG) at position 560 (FIGS. 3H-1 and 3H-2). The open reading frame encodes a polypeptide containing 180 amino acid residues with a calculated molecular weight of 19.2 kD and a predicted pI of 8.46. The predicted amino acid sequence contains the peptide sequence, L-V-V-V-V-M-T-P-R-K-H (SEQ ID NO:41), which was used for the cloning of the PCR fragment. In addition, another peptide sequence, L-G-I-S-O-A-O-S-O-A-G-E-V-D-(G) predicted from nucleotides 428-472, matches SEQ ID NO:34 obtained from RT18.2 (FIG. 3C). However, other sequences from peak RT7.8 (SEQ ID NOS:39-44) are absent from the PcAGP23 sequence, indicating they are from different AGP backbones.

The most abundant amino acid residues in the predicted protein sequence are Ser (12.2%), Gly (10.5%), Leu (9.4%), Val (8.8%), Ala (7.2%) and Lys (7.2%) Table 3.3. The PcAGP23 contains 5.5% Pro residues, some of which are post-translationally modified to hydroxyproline, as identified by peptide sequencing. The Pro and Ala residues are, relatively speaking, concentrated in the last one-third of the sequence (at C-terminus).

In the sequence of the PcAGP23 cDNA (SEQ ID NO:49), there is a putative secretion signal at the N-terminus (1-27) with a potential cleavage site between Ala.sup.27 and Arg.sup.28. There are also two potential N-glycosylation sites at amino acid positions 36 and 87 (FIGS. 3H-1 and 3H-2).

In another embodiment of the invention, the AGPs in a pear cell culture filtrate were further purified as illustrated in the flow chart of FIGS. 5D-1 and 5D-2. The unbound fraction and the two minor bound fractions (FIG. 5D-1) which accounted for 72%, 0.9% and 0.1%, respectively, of total AGPs loaded on the column, were purified as described above and in Example 3(a).

TABLE 3.3 ______________________________________ Amino acid composition of the predicted PcAGP23 protein Mol % Amino Acid +SP -SP ______________________________________ Ser 12.2 9.8 Gly 10.5 11.1 Leu 9.4 7.8 Val 8.8 9.1 Ala 7.2 6.5 Lys 7.2 7.8 Thr 5.5 6.5 Pro 5.5 6.5 Glu 5.0 5.2 Phe 4.4 2.6 Asp 3.8 4.5 Asn 2.7 3.2 Tyr 2.7 2.6 Arg 2.7 3.2 Ile 2.7 3.2 Gln 2.2 2.6 Trp 2.2 2.6 Cys 1.6 1.9 His 1.6 1.9 Met 1.1 0.6 ______________________________________ +SP: The putative secretion signal peptide is included. -SP: The putative secretion signal peptide is excluded.

The major peak of FIG. 5D-1, which accounted for approximately 27% of the AGPs, was collected and reapplied to the same column. Upon elution with a shallow gradient, two peaks (Fractions 1 and 2) were resolved (FIG. 5D-2). The AGPs in Fraction 1 were described above and in Example 3(a).

Size-exclusion FPLC fractionation of Fraction 2 resolved two components (peaks 2A and 2B, FIGS. 5D-2,C3). Arabinose and galactose were the major monosaccharides of each fraction (Table 3.4).

TABLE 3.4 ______________________________________ Linkage analysis of AGP fractions Monosaccharide Unbound Fraction 1 Fraction 2 and deduced fraction (FIG. (FIG. 5D- Peak 2A Peak 2B linkage (mol %) 5D-2, C1) 2, C2) (FIG. 5D-2, C3) ______________________________________ Araf: terminal 34 36 24 18 3- 3 3 4 4 5- 2 3 1 1 Galp: terminal 7 8 12 14 3- 5 4 8 5 6- 10 10 8 23 3,6- 38 36 44 35 ______________________________________ Araf: Arabinofuranose; Galp: Galactopyranose

Arabinose was present mainly in the terminal position with small amounts of 3-linked and 5-linked residues. Galactose was present mainly as 3,6-linked and terminal residues in both peaks. However, the proportion of 6-linked galactosyl residues was greater in Peak 2B than 2A, and both had small proportions of 3-linked residues. Amino acid composition analysis of the AGPs in Peaks 2A and 2B are shown in Table 3.5. N-terminal amino acid sequencing of material in Peak 2B gave the sequence A-E-A-E-A-X-T-X-A-L-Q-V-V-A-E-A-X-E-L (SEQ ID NO:74).

TABLE 3.5 __________________________________________________________________________ Amino acid composition (mol %) of AGPs in Peaks 2A and 2B, their deglycosylated backbone and the protein deduced from PcAGP2 cDNA. Deglycosylated Deglycosylated Deduced Amino acid Peak 2A Peak 2A Peak 2B Peak 2B peptide Deduced (Mol %) (FIG. 5D-2, C3) (FIG. 5D-2, D3) (FIG. 5D-2, C3) (FIG. 5D-2, D4) fragment* protein** __________________________________________________________________________ Hyp 28.2 27.5 19.2 19.5 Pro 3.3 4.3 2.8 2.8 19.4 5.4 Asx 1.2 2.0 2.2 2.0 Asn 0.0 Asn 14.2 Asp 2.7 Asp 5.4 Glx 6.9 6.6 14.9 14.3 Glu 13.8 Glu 8.0 Gln 2.7 Gln 4.0 Ser 18.6 18.4 6.6 6.0 5.5 9.1 Gly 4.1 4.5 3.4 5.5 2.7 10.5 His 0.4 0.6 1.3 1.9 2.7 2.5 Arg 0.6 0.0 2.5 2.7 2.7 3.6 Thr 11.6 10.7 10.4 9.6 11.1 6.5 Ala 13.1 12.5 16.7 16.6 16.6 3.2 Tyr 0.2 0.1 2.5 0.1 2.7 7.6 Val 4.9 4.2 9.3 10.0 11.1 4.7 Met 0.4 1.8 0.3 0.2 0.0 1.0 Ile 2.8 2.1 0.8 0.8 0.0 4.0 Leu 1.3 1.5 5.9 6.4 5.5 2.9 Phe 0.0 0.0 0.4 0.5 0.0 2.5 Lys 2.9 2.1 1.0 1.3 0.0 4.0 Cys nd nd nd nd 0.0 0.0 Trp nd nd nd nd 0.0 0.0 __________________________________________________________________________ *Amino acid residues 53-88 (FIGS. 2A-2D). **Complete deduced protein sequence excluding the 20amino acid signal sequence (FIGS. 2A-2D). nd: Not determined

AGPs in Peaks 2A and 2B were separately deglycosylated and the resulting protein backbones isolated by size-exclusion FPLC (FIG. 5D-2, D1-D4). The apparent M.sub.r of the proteins was different for each fraction. Peak 2B gave one protein backbone (M.sub.r 10k), Peak 2A resulted in two protein peaks (M.sub.r 10k and 54k). The 10k protein backbone in Peak 2A is a contamination from Peak 2B. N-terminal amino acid sequencing of the 54k protein backbone gave the sequence T-O-A-O-A (SEQ ID NO:75) while the 10k protein backbone in Peak 2B gave the sequence A-E-A-E-A-O-T-O-A-L-Q-V-V-A-E-A-O-E-L (SEQ ID NO:76). The latter sequence is identical to the N-terminal sequence obtained from the AGP in Peak 2B before deglycosylation, assuming the unassigned residues "X" are Hyp. The amino acid compositions of the 54k and 10k protein backbones are very similar to that of their parent AGPs in Peaks 2A and 2B, respectively. The 54k protein backbone contained a higher proportion of Hyp (27.5%), Ser (18.4%) than the 10k protein backbone in Peak 2B (Hyp, 19.5%; Ser, 6.0%). On the other hand, the 10k protein backbone had a higher content of Glx (14.3%) and Val (10.1%) than the 54k protein backbone in Peak 2A (Glx, 6.6%; Val, 4.2%) (Table 3.5). The 10k and 54k protein backbones were digested separately with thermolysin and the resulting peptides purified by RP-HPLC for sequencing. Sequences of eight peptides were obtained from the 54k protein Peak 2A and three from the 10k protein in Peak 2B (Table 3.6). Two of the three sequences and the N-terminal sequence overlap to give a sequence A-E-A-E-A-O-T-O-A-L-Q-V-V-A-E-A-O-E-L-V-O-T-O-V-O-T-O-S-Y (SEQ ID NO:88) for the 10k protein in Peak 2B.

TABLE 3.6 __________________________________________________________________________ Peptide sequences obtained from Peaks 2A and 2B Peak Peptide sequence __________________________________________________________________________ Peak 2A T--O--A--O--A (N-terminal) (SEQ ID NO:75) V--S--X--O--V--Q--S--O--A--X--O (SEQ ID NO:77) V--X--X--O--V--Q--S--O--A--S--O--O--O--T--T (SEQ ID NO:78) I--S--O--A--S--T--O--O--T-- (SEQ ID NO:79) I--S--O--A--S--T--O--O--T--O--A--S--O--O--T (SEQ ID NO:80) F--S--O--T--I--S--O--A (SEQ ID NO:81) X--A--(A)--T--O--S--L--D--V--G--I--O--S--S--N--A--T (SEQ ID NO:82) T/P--S--O--A--T--O--O--A--T (SEQ ID NO:83) X--A--A--O--A--O--S--(O)--X--P--T--(N)--T (SEQ ID NO:84) Peak 2B A--E--A--E--A--X--T--X--A--L--Q--V--V--A--E--A--X--E--L (N-terminal)* # (SEQ ID NO:74) A--E--A--E--A--O--T--O--A--L--Q--V--V--A--E--A--O--E--L (N-terminal)** # (SEQ ID NO:76) V--V--A--E--A--O--E--L--V--O--T--O--V--O--T--O--S-# (SEQ ID NO:85) L--V--O--T--O--V--O--T--O--S--Y # (SEQ ID NO:86) Y--T--E--R-- # (SEQ ID NO:87) __________________________________________________________________________ Note: All the residues of ambiguous assignments are shown, uncertain residues are in brackets. "X" indicates no signal or an unknown residue. "O" represents hydroxyproline. Sequences included in the cDNA are marked #. *Obtained from the AGP in Peak 2B before deglycosylation. **Obtained from the deglycosylated protein backbone of the AGP in Peak 2B

In another embodiment of the invention, AGPs were isolated from N. alata styles. In this example, the total native N. alata style AGPs were not purified by the Yariv reagent precipitation technique, but by ion exchange chromatography (IEC) followed by gel filtration chromatography (GFC). The presence of AGP in column fractions was verified by precipitation of AGP with a Yariv reagent. The AGPs were then deglycosylated by HF and fractionated by reversed phase HPLC.

Two major peaks: RT25 and RT35 (FIG. 4C) were obtained after deglycosylation and HPLC fractionation. Amino acid analysis of each fraction and the native materials are shown in Table 4.1.

TABLE 4.1 ______________________________________ Amino acid analysis of AGPs isolated from the styles of N. alata (genotype S.sub.6 S.sub.6) MonoQ bound AGPs separated by GPC chromatography Total native AGPs separated deglycosylated and further Amino by MonoQ column fractionated by RP300 HPLC Acid MonoQ Unb. MonoQ Bound RP-Unbound RT25 RT35 ______________________________________ Hyp 19.2 7.5 ND 18.1 0.4 Asx 6.3 13.0 8.8 3.6 14.8 Thr 3.9 7.7 4.0 6.9 5.3 Ser 10.3 6.0 11.5 14.8 6.2 Glx 8.8 11.2 13.3 6.2 14.0 Pro 9.2 3.0 3.6 2.7 2.7 Gly 6.1 5.5 20.7 9.3 7.9 Ala 16.6 15.9 6.2 20.1 10.6 Val 3.7 3.4 4.4 6.3 3.8 1/2Cys ND ND ND ND ND Met 2.1 2.6 0.9 1.5 2.7 Ile 2.4 4.3 3.6 1.7 5.2 Leu 4.2 5.9 7.0 2.7 6.9 Tyr 1.2 3.4 3.4 0.5 4.2 Phe 1.3 2.7 2.2 1.4 3.7 Lys 0.4 3.5 1.8 2.3 4.6 His 1.1 1.0 3.2 1.0 1.7 Arg 2.7 2.2 4.7 0.8 3.4 Trp ND ND ND ND ND ______________________________________ Hyp: Hydroxyproline, ND: not determined

Distinct differences are apparent in the amino acid composition between the three fractions. The unbound fraction contains little Hyp but is rich in Gly, Glx, Ser and Asx. The RT35 fraction is also Hyp-poor but rich in Asx, Glx and Ala. Together, these two fractions account for the bulk of the Asx and Glx detected in the native and deglycosylated AGPs. The amino acid composition of the material in fraction RT25 is dominated by Hyp (18%), Ala (20%) and Ser (15%) with very little Tyr. This RT25 protein backbone was thus selected for further analyses.

Peak RT25 gave four sequences (SEQ ID NOS:50-53) which are OAST-enriched. Three of these sequences (SEQ ID NOS:50, 51, and 52) closely matched SEQ ID NOS:27-29, respectively, for RT21 from N. plumbaginafolia.

An N-terminal sequence was not obtained for the RT25 peak. Pyroglutamate aminopeptidase was then used to remove the N-terminal blocked pyroglutamate residue and the sequence Ala-Hyp-Gly was obtained. The RT25 backbone was also fragmented by treatment with the endoproteinase thermolysin. The resulting peptides were separated and further purified by RP-HPLC. Six major peptides (FIG. 4I) were subjected to amino acid sequencing and four sequences were obtained (SEQ ID NOS:50, 51, 53, 67). All the sequences were rich in Hyp, Ser and Ala (33 of 52 amino acid residues).

Endoproteinase Asp-N was also used to cleave the RT25 protein backbone at the Asp residues. Two major peptides were produced (A1 and A2; FIG. 4J) indicating that there is only one Asp residue in the RT25 protein. The cleavage was incomplete as indicated by the presence of the starting material (RT25 protein; FIGS. 4K-1 and 4K-2). Peptide sequence was obtained for A2 (SEQ ID NO:68). The other peptide (A1) gave no sequence data, indicating a blocked N-terminal residue. Overlaps were identified between A2 (SEQ ID NO:68) (FIG. 4J) and Peak 3 (SEQ ID NO:51) (FIG. 4I) and gave a continuous amino acid sequence of 26 reisdues: LASOOAOOTADTOAFAOSGGVALPOS (SEQ ID NO:69).

Peak RT35 gave four sequences (SEQ ID NOS:54-57) which had a low OAST content. Three of these sequences (SEQ ID NOS:55-57) were characterized by the sequence T-A-I-N-T-E-F-G-P (SEQ ID NO:58).

In an alternative method of preparation, N. alata style AGPs were isolated according to Bacic et al. (1988) Phytochem. 27:679-684. The sample was deglycosylated with TFMS, separated and blotted onto a PVDF membrane as described previously for N. alata suspension-cultured cells. An approximately 30 kD band, running at the same position as the major band prepared by Yariv precipitation from N. alata suspension cultured cells, was sequenced. The sequence A-V-F-K-N-K-X-X-L-T-X-X-P-X-I-I (SEQ ID NO:59) was obtained.

In other embodiments of the invention, AGP genes from N. alata style were isolated. The cloning strategy of FIG. 4D was used to obtain the genes. Several of the peptide sequences of peak RT35 isolated from N. alata style contained the sequence T-A-I-N-T-E-F-G-P (SEQ ID NO:58). In a specific embodiment, gene-specific degenerate oligonucleotide primers were designed based on the sequence A-I-N-T-E-F-G (SEQ ID NO:60) and a PCR fragment was amplified in vitro from style RNA of N. alata. A 380-bp PCR fragment (SEQ ID NO:62; FIG. 4E) was used to screen a style cDNA library and a cDNA clone was isolated and fully sequenced. The N. alata style cDNA clone was designated Na35.sub.-- 1. The insert of the cDNA clone was 800 bp in length with a poly(A) tail at the 3'-end. The cDNA sequence (SEQ ID NO:63) matched the PCR sequence except that it was 3 bp shorter at the 3'-end (FIG. 4E and FIGS. 4F-1 and 4F-2).

The Na35.sub.-- 1 sequence (SEQ ID NO:63) had an open reading frame starting with an initiation codon (ATG) at position 21 and ending with a termination codon (TAA) at position 530 (FIGS. 4F-1 and 4F-2). The open reading frame encoded a polypeptide containing 169 amino acid residues with a calculated molecular weight of 19.5 kD and a predicted pI of 8.1. The most abundant residues in the sequence were: proline (11.2%), phenylalanine (9.5%), alanine (7.7%), leucine (7.7%), and lysine (7.7%) (Table 4.2).

TABLE 4.2 ______________________________________ Amino acid composition of the predicted Na35.sub.-- 1 polypeptide and purified RT35 peptide peak Amino Acid Na35.sub.-- 1 cDNA RT35 peptide ______________________________________ Asn/Asp 11.8 14.5 Thr 3.0 6.7 Ser 5.9 5.5 Gln/Glu 7.1 12.4 Pro/Hyp 11.2 7.8 Gly 1.8 4.3 Ala 7.7 10.8 Val 2.4 4.2 1/2 Cys 4.7 1.1 Met 2.4 3.0 Ile 7.7 4.8 Leu 7.7 6.6 Tyr 3.0 3.1 Phe 9.5 4.2 Trp 1.8 ND Lys 6.5 5.3 His 1.8 2.1 Arg 4.1 3.4 ______________________________________ N.D.: not determined

The amino acid sequence derived from the N. alata style cDNA (SEQ ID NO:63) comprised regions that matched peptide fragments of peak 35 isolated from N. alata styles, i.e., SEQ ID NO:55, SEQ ID NO:56, and SEQ ID NO:57. Northern blot analyses of the Na35.sub.-- 1 gene (FIG. 4G) indicated a specificity of the gene to N. alata and to style tissue. Signals were not detected in transcripts from tomato style, N. alata cell suspension, N. plumbaginafolia cell suspension, and pear cell suspension (FIG. 4H) indicating that the Na35.sub.-- 1 PCR fragment was specific for an N. alata style AGP gene.

Further, the isolation of a different gene for an N. alata style AGP is described in another specific embodiment of the invention. The five peptides isolated from fragments of the AGP protein backbone (SEQ ID NOS:50,51,53,67 and 68) together gave 52 amino acid residues. Much of the sequence contained adjacent residues of Hyp, Ser and Ala for which the codons are highly redundant and GC-rich. These sequences are not useful for cloning. However, the sequence TADTOAF from the continuous 26 amino acid sequence resulting from the overlaps of the isolated peptides contains two amino acids which are not GC-rich and only have two degenerate codons. This TADTOAF sequence allowed design of an oligonucleotide suitable for PCR and the eventual cloning of the AGPNal 1 cDNA.

A gene-specific oligonucleotide (20 nucleotides) was designed from one region of the continuous 26 amino acid sequence: TADTOAF (SEQ ID NO:70). Inosine was used at the third position of the first two codons to reduce the degeneracy of the oligonucleotide to 128. The resulting oligonucleotide contained 60% GC. cDNA was synthesized from total style RNA using poly T linked with an adaptor sequence. Rapid amplification of the cDNA 3' end (3' RACE) was performed using the gene-specific primer together with a 3' primer in the adaptor sequence. A PCR fragment of 400 base pairs (bp) was produced. This PCR fragment was cloned and sequenced. The deduced amino acid sequence from this PCR clone matched isolated AGP sequences, i.e., SEQ ID NOS: 50, 51, 53, 67, 68.

The PCR clone was then used as a probe to screen a style cDNA library (300,000 plaques). Two cDNA clones were obtained which differ only in the length of the 3' and 5' ends. One of the clones, designated AGPNal 1 (SEQ. ID NO:72; FIGS. 4K-1 and 4K-2) was used for all subsequent analyses. The 3' end of the AGPNal 1 cDNA clone was identical to the PCR clone except that the PCR clone was 20 bp shorter and contained a poly A tail. The 712-bp AGPNal 1 clone encodes a putative protein of 12.5 kD (FIGS. 4K-1 and 4K-2). The derived amino acid sequence includes sequences identical to isolated AGP peptides (SEQ ID NOS:50, 51, 53, 67 and 68). Most of the proline residues in the peptide sequences obtained by amino acid sequencing are hydroxylated. A secretion signal peptide is predicted (FIGS. 4K-1, 4K-2 and 4L). The deduced N-terminus of the mature protein (10 kD; pI 6.8) is Gln-Ala-Pro-Gly which matches the N-terminal sequence data obtained. The Pro residue in the N-terminal sequence is also hydroxylated. The amino acid composition of the deduced mature protein and the isolated RT25 protein backbone are in general agreement (Table 4.1). The C-terminus of the deduced protein is very hydrophobic and predicted to be a transmembrane helix.

The cDNA clone obtained (FIGS. 4K-1 and 4K-2) predicts a 132 amino acid protein characterized by hydrophobic stretches at both the N-and C-termini (FIG. 4L). The N-terminal hydrophobic sequence corresponds to a signal peptide which would lead to secretion of the encoded protein. This is consistent with the known secretion and extracellular localization of the style AGPs [Sedgley et al. (1985) Micron Microscop. Acta 16:247-254]. Modification of the N-terminal residue, Gln, by intra-molecular cyclization to form pyroglutamate is not unusual. The cyclization could occur during purification, or it could occur in situ and might be involved in the stabilization of the AGP backbones. The same N-terminal sequence: Gln-Ala-Pro-Gly-Ala is also present in the AGP backbone isolated from pear (FIGS. 5A-1 and 5A-2). The C-terminal hydrophobic sequence is predicted to be a transmembrane helix (FIG. 4L) which might anchor the AGP in plasma membrane. The hydrophobic C-terminal region could also potentially enable the interaction of the AGP with other proteins, such as S-RNase which also contains a very hydrophobic sequence (in this case at the N-terminus of the mature protein; Mau et al. (1986) Planta 169:184-191. The central part of the protein contains most of the Hyp, Ala, Ser residues. The fact that most of the Pro residues within the peptide sequences are hydroxylated suggests extensive O-glycosylation in the central part of the protein. No potential N-glycosylation sites are present. The abundance of potential O-glycosylation sites is consistent with the high content of carbohydrate (85% w/w). Individual AGPs may differ in the types of saccharide chains and in the number and location of glycosylation sites along the protein backbone.

mRNA hybridizing to AGPNal 1 cDNA is present in most tissues of N. alata and in the styles of related solanaccous species (FIGS. 4M-1 and 4M-2) suggesting a general role of this transcript (or closely related transcripts) in plant development. Various tissues from N. alata were examined for the expression of the AGPNal 1 gene. As shown in FIG. 4M-1, mRNA transcripts of similar length of about 700-750 nucleotide were detected in all tissues examined. This suggests that the AGPNal 1 gene or its homologs are expressed in many parts of the plant. Style, ovary, petal, leaf and stem have similar levels of transcript, but the highest level of mRNA expression is found in roots.

Some expression of hybridizing transcript was detected in the styles of N. sylvestris and N. tabacum and a lower level in N. glauca and Lycopersicon peruvianum (FIG. 4M-2). Arabidopsis and rye grass (Lolium perenne, a monocot) leaves had no detectable hybridizing transcript.

In another embodiment of the invention, an AGP gene was isolated from P. communis using a guessmer oligonucleotide sequence encoding a hydroxyproline-rich pear AGP segment and linked to a double-stranded promoter sequence for RNA polymerase, allowing the synthesis of an antisense RNA probe (see FIG. 1) (strategy B). Strategy B thus enabled the isolation of an AGP gene (SEQ ID NO:66) that specifically encodes a particular hydroxyproline-rich peptide segment (see FIGS. 5A-1 and 5A-2). Hydroxyproline-rich and OAST-rich domains appear to represent characterizing features of AGPs.

AGP peptide fragments were isolated and sequenced essentially as described in Example 3(a). The sequence A-K-S-O-T-A-T-O-O-T-A-T-O-O-S-A-V (SEQ ID NO:37) of an isolated pear AGP fragment exhibited hydroxyproline-enrichment and OAST-enrichment. This sequence was selected for the isolation of a corresponding pear AGP gene. The codon usage for proline is strongly biased towards CCA which accounts for 73.3% of all proline codons; the codon for alanine is biased, to a lesser extent, to CCT (44.8%); there is no significant bias in codon usage for other amino acids.

Two hybrid oligonucleotides (AF1T3 and AR2T7), each comprising a GC-enriched sequence encoding a hydroxyproline-rich AGP segment, were constructed as primers. The sequences of primers AF1T3 and AR2T7, each comprising a GC-rich domain, are presented in Table 5.1. AF1T3 (SEQ ID NO:64) includes a T3 promoter sequence, a 42-bp GC-enriched nucleotide sequence corresponding to an isolated N. plumbaginafolia AGP peptide fragment (SEQ ID NO:27), that is OAST-enriched, and an 18-bp sequence corresponding to position 150-167 from the NaAGP1 (SEQ ID NO:24). The AR2T7 primer (SEQ ID NO:65) consists of a T7 promoter, a 47-bp GC-enriched nucleotide sequence corresponding to a hydroxyproline-rich (OAST-enriched) AGP sequence from pear (SEQ ID NO:37) and another 18-bp sequence corresponding to position 444-461 from the NaAGP1 cDNA (SEQ ID NO:24).

An antisense RNA probe was synthesized from the guessmer oligonucleotide template by using T7 polymerase, and was used to screen a cDNA library prepared from pear cell suspension culture essentially as described in Example 3(b). Three cDNA clones were isolated and sequenced. The sequence of the longest clone PcAGP9 (SEQ ID NO:66) is shown in FIGS. 5A-1 and 5A-2. The cDNA clone contains an insert of 893 bp and encodes an open reading frame of 145 amino acid residues. There is a putative secretion signal peptide at the N-terminus. The predicted polypeptide is highly rich in Pro, Ala, Ser, and Thr (Table 5.2) and contains two sequences which match exactly two peptide sequences obtained previously from pear AGPs by protein sequencing: AKSOTATOOTATOOSAV (SEQ ID NO:37) and VTAOTOSASOOSSTOA(S)TXA (SEQ ID NO:38). The PcAGP9 sequence (with the secretion signal included) gave an estimated pI of 10.79 and an apparent molecular weight of 13.622 kD. The PcAGP9 sequence (excluding the secretion signal) gave an estimated pI of 11.07 and an apparent molecular weight of 11.238 kD.

TABLE 5.1 __________________________________________________________________________ Nucleotide sequences of the primers AF1T3 and AR2T7 __________________________________________________________________________ AF1T3:(Forward primer) (SEQ ID NO: 64) ##STR9## AR2T7 (Reverse primer) (SEQ ID NO:65) ##STR10## __________________________________________________________________________ Promomter sequences are underlined once. Sequences from NaAGP1 cDNA are double underlined.

TABLE 5.2 ______________________________________ Amino acid composition of the PcAGP9 sequence +Secretion Signal -Secretion Signal Amino acid No. Mol % No. Mol % ______________________________________ Pro 30 20.6 30 24.5 Ala 29 20.0 26 21.3 Ser 25 17.2 24 19.6 Thr 17 11.7 16 13.1 Val 8 5.5 6 4.9 Gly 8 5.5 5 4.1 Leu 6 4.0 2 1.6 Ile 5 3.4 4 3.2 Lys 4 2.7 3 2.4 Phe 4 2.7 2 1.6 Met 3 2.0 0 0.0 Gln 2 1.3 1 0.8 Asn 1 0.6 1 0.8 Asp 1 0.6 1 0.8 Arg 1 0.6 1 0.8 Cys 1 0.6 0 0.0 Glu 0 0.0 0 0.0 His 0 0.0 0 0.0 Tyr 0 0.0 0 0.0 Trp 0 0.0 0 0.0 ______________________________________

As illustrated in the hydropathy profile of FIG. 5C, the cDNA has three domains, an N-terminal hydrophobic sequence encoding a secretion signal, a central hydrophilic domain containing most of the proline residues and a hydrophobic C-terminal domain which is predicted to be a transmembrane helix. The N-terminus of the mature protein corresponds to the sequence predicted from processing of the secretion signal. The proline residues within the central region are mainly hydroxylated and would bear the glycosyl chains. A cDNA encoding the protein backbone of an AGP from the styles of Nicotiana alata, has three domains with similar characteristics. Although the amino acid composition of the proteins encoded by these cDNAs is similar, the only common sequence is at the N-terminal sequence of the mature proteins, Q-A-P-G-A-A (SEQ ID NO:73). The cDNAs encode protein backbones of single AGPs from several present in the plant extracts which are quantitatively a minor part of these proteoglycans.

The central part (amino acids 24-123) of the sequence is dominated by four amino acids (Pro, 29%; Ala, 19%; Ser, 23% and Thr 15%). The dominant feature of this part of the sequence is that the four residues are interspersed with each other; there are no obvious motifs and few runs of any single amino acid. There are no predicted N-glycosylation sites.

The C-terminal region of 22 amino acid residues is very hydrophobic and is predicted to be a transmembrane helix [Eisenberg et al. (1984) J. Mol. Biol. 179:125-142; Klein et al (1985) Biochem Biophys. Acta 815:468-476; Rao et al. (1986) Biochem Biophys. Acta 869:197-214]. There are several potential sites for proteolytic cleavage (Endoprotease Asp-N, Ala.sup.114 /Asp.sup.115 ; V8 protease, Asp.sup.115 /Ala.sup.116 ; Clostripain and Trypsin, Arg.sup.127 /Val.sup.128) around the border between the C-terminal transmembrane helix and the extracellular domain [Allen et al. (1989) Sequencing of Proteins and Peptides (2nd ed.); Drapeau (1978) Can. J. Chem. 56:534-544; (1980) J. Biol. Chem. 255:839-840]. These represent single cleavage sites, with the exception of trypsin for which there are several cleavage sites within the sequence.

The PcAGP9 cDNA was used to probe northern blots containing RNA from six plants representing both dicotyledonous (Pyrus, Nicotiana, Brassica, Arabidopsis, and Lycopersicon) and monocotyledonous (Lolium) plants (FIGS. 5B-1 and 5B-2). At high stringency (65.degree. C.), a 0.9 kb transcript was detected in an RNA sample from suspension culture cells of Pyrus communis. A smaller transcript was also detected in pedicels of the same plant together with a larger transcript in N. plumbaginafolia suspension culture cells (FIG. 5B-2). Under reduced stringency conditions (55.degree. C.), RNA transcripts were also detected in all other RNA samples tested indicating the expression of AGP genes homologous to PcAGP9 in both dicotyledonous and monocotyledonous plants tested (FIG. 5B-1).

The PcAGP9 cDNA has similarity to the N. alata sytle cDNA (AGPNal 1 clone). In both cases the cDNA clones predict protein sequences composed mainly of Pro, Ala, Ser and Thr. Despite the similarity in amino acid composition, these cDNA clones have little sequence identity. In fact, the AGPNal 1 cDNA and PcAGP9 cDNA did not cross hybridize at medium to high stringency on RNA blot analysis; the AGPNal 1 detected a single 700-750 nt transcript in most tissues examined while the PcAGP9 detected a 800-900 nt mRNA. Other AGP-like peptide sequences have also been reported from N. plumbaginifolia, pear, L. multiflorum and a histidine-rich HRGP from maize suspension cell culture filtrate [Kieliszewski et al (1992) Plant Physiol. 99:538-547]. Again, these peptides are composed mainly of Hyp, Ala and Ser residues yet the exact sequences is different. For example, the Ala-Pro-Ala-Pro repeats present in L. multiflorum are not present in the deduced amino acid sequence from the AGPNal 1 and PcAGP9 cDNA.

In another embodiment of the invention, another P. communis cDNA (PcAGP2; SEQ ID NO:91) was isolated and shown to be distinct from both the PcAGP9 (SEQ ID NO:66) and the PcAGP23 (SEQ ID NO:49) clones. The approach to cloning the PcAGP2 cDNA was essentially the same as for the PcAGP9 cDNA (Example 5).

The 10k protein purified in FPLC as Peak 2B (FIG. 5D-2,D4) and having the amino acid sequence of AEAEAOTOALQVVAEAOELVOTOVOTOSY (SEQ ID NO:88) was selected for the isolation of a corresponding pear AGP gene. Two reverse and partially complementary long "guessmers" [AcF1 (SEQ ID NO:89) and AcR2 (SEQ ID NO:90), Table 5.3] were synthesized.

TABLE 5.3 __________________________________________________________________________ Nucleotide and corresponding peptide sequences of the "guessmers" AcF1 and AcR2 __________________________________________________________________________ AcF1 (SEQ ID NO:89) ##STR11## AcR2 (SEQ ID NO:90) 5'CTGGAGCTCATATGATGGTGTTGGTACTGGTGTTGGTACTAG TTCTGGTGCTTCTGCTAC3' __________________________________________________________________________ Note: Restriction enzyme cut sites incorporated into the guessmer for subcloning are underlined. Reversecomplementary regions are doubleunderlined.

In the "guessmers," nucleotide A was used at the third codon position for all amino acids, and CTA and TCA were assigned for Leu and Ser residues, respectively. The last 18 bp sequence at the 3' of the two "guessmers" were reverse-complementary, and they were annealed to each other in PCR to produce a double-stranded DNA fragment of 101 bp encoding the amino acid sequence A-E-A-E-A-O-T-O-A-L-Q-V-V-A-E-A-O-E-L-V-O-T-O-V-O-T-O-S-Y (SEQ ID NO:88). The PCR fragment was subcloned into the pBluescriptII (Ks) vector. A .sup.32 P-labeled anti-sense RNA probe was synthesized using T3 RNA polymerase from the 101-bp DNA fragment and used to screen a pear cDNA library. Five cDNA clones were isolated and sequenced. The consensus sequence of 1040 bp is shown in FIGS. 5E-1 and 5E-2. This cDNA is referred to as PcAGP2 (SEQ ID NO:91).

The PcAGP2 cDNA sequence encodes a polypeptide of 294 residues and can be divided into four domains (FIGS. 5E-1 and 5E-2). The first 20 amino acid sequence is hydrophobic and predicted to be a secretion signal with a potential cleavage site between Ser.sup.20 and Phe.sup.21. The second domain (residues 21-51) is rich in Asn and contains a stretch of five Asn residues. The third domain (residues 52-135) is rich in Pro, Ala, Thr, and Glu. Most of these four residues are located in this domain. This domain also includes all the peptide sequences obtained by protein sequencing. The fourth domain (residues 136-294) is rich in Asn and Gly and contains two direct repeated sequences of 34 residues. The amino acid composition of the deduced protein, excluding the signal sequence, differs from that obtained from the glycosylated and deglycosylated AGP in Peak 2B in that it is rich in Asn (14.2%), Glu (8.0%), Gly (10.5%) and Ser (9.1%) (Table 3.5). However, the sequence from residues 53 to 88 has an amino acid composition closely matching that obtained from the AGP in Peak 2B.

Except as noted hereafter, standard techniques for isolation and purification of proteins and protein fragments, sequencing, chromatography, cloning, DNA isolation, amplification and purification, for enzymatic reactions involving DNA ligase, DNA polymerase, restriction endonuclease and the like, the PCR technique and various protein separation and purification techniques are those known and commonly employed by those skilled in the art. A number of standard techniques are described in Deutscher (1990) Methods in Enzymology 182:309-539; Maniatis et al. (1982) Molecular Cloning, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.; Wu (ed.) (1979) Meth. Enzymol. 68; Sambrook, et al. (1989) supra; Wu et al. (1983) Meth. Enzymol. 100 and 101; Grossman and Moldave (eds.) (1980) Meth. Enzymol. 65; Miller (ed.) (1972) Experiments in Molecular Genetics, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.; Old and Primrose (1981) Principles of Gene Manipulation, University of California Press, Berkeley; Schleif and Wensink (1982) Practical Method of Molecular Biology; Glover (ed.) (1985) DNA Cloning Vols. I and II, IRL Press, Oxford, UK; Hames and Higgins (eds.) (1985) Nucleic Acid Hybridization, IRL Press, Oxford UK; Setlow, Hollaender (1979) Genetic Engineering: Principles and Methods, Vols. 1-4, Plenum Press, New York and Deutscher (ed.) (1990) Guide to Protein Purification, Academic Press, New York. Abbreviations and nomenclature, where employed, are deemed standard in the field and commonly used in professional journals such as those cited herein.

It will be appreciated by those of ordinary skill in the art that the objects of this invention can be achieved without the expense of undue experimentation using well known variants, modifications, or equivalents of the methods and techniques described herein. The skilled artisan will also appreciate that alternative means, other than those specifically described, are available in the art to attain protein purification and to achieve the functional features of the molecules described herein and how to employ those alternatives to achieve functional equivalents of the molecules of the present invention. It is intended that the present invention include those variants, modifications, alternatives, and equivalents which are appreciated by the skilled artisan and encompassed by the spirit and scope of the present disclosure.

The following examples are provided to better elucidate the practice of the present invention and should not be interpreted in any way to limit the scope of the present invention. Those skilled in the art will recognize that various modifications can be made to the methods and genes described herein while not departing from the spirit and scope of the present invention.

EXAMPLE 1

General Method for the Isolation and Purification of AGP Peptides from Plant Cells Comprising AGP

1. Preparation of Cell Suspension Cultures

Suspension cultures of plant cells comprising AGP were initiated from cotyledons of seedlings germinated in the medium of Murashige and Skoog (1977) Physiol. Plant 15:473-497 supplemented with plant hormones, factors, buffers, salts, etc., as are routinely used in the art to enhance and improve the quality of cell growth.

2. Preparation of Plant Tissue Extracts

Plants were grown from commercial seed stock and were maintained under standard glass house conditions.

3. Isolation of Total AGPs

Total AGPs were prepared from suspension-cultured cells by precipitation of AGPs from the culture medium with Yariv reagent [Yariv et al. (1967), Biochem. J. 105:1C], followed by dissociation of the AGP-Yariv reagent complex and recovery of the AGP. Alternatively, total AGPs were prepared from plant tissue extracts by (NH.sub.4).sub.2 SO.sub.4 precipitation, anion exchange chromatography and/or immunoaffinity chromatography with, for example, an antibody specific for Gal 1-6-B-Gal sequences, followed by gel filtration chromatography using, for example, a superose matrix.

The AGPs of the total AGP fraction were separated using either ion exchange or reverse phase HPLC. The individual AGPs were then subjected to amino acid sequencing. Alternatively, the total AGP fraction was subjected to deglycosylation using, for example, TFMS or HF, and the deglycosylated AGPs were separated either on SDS-PAGE or reverse phase HPLC and prepared for amino acid sequencing. In some cases, the peptides were digested by treatment with proteolytic enzymes before separation of the different deglycosylated peptides.

Hydroxyproline-rich AGP fragments are separated from hydroxyproline-poor fragments by chromatographic methods based on differentiating characteristics, e.g., polarity, immunogenicity, etc. For example, affinity chromatography supports to which are attached ligands specific for amino acid R-group hydroxyls or antibodies to a hydroxyproline-rich peptide fragment that is OAST-enriched are used to retain preferentially hydroxyproline-rich peptides. Other protein purification techniques useful in the separation of hydroxyproline-rich and hydroxyproline-poor fragments are found in Deutscher, Guide to Protein Purification (1990) Methods in Enzymology 182.

EXAMPLE 2

Cloning of Genes Encoding a Protein Backbone of an AGP from Nicotiana alata, and N. plumbaginafolia

(a) Isolation and Purification of AGP Peptides from Suspension Cultures of Nicotiana alata

1. Preparation of Suspension Cultures

Suspension cultures of N. alata cells were initiated from cotyledons of seedlings germinated in the medium of Murashige and Skoog (1977), supra, supplemented with 1 g/l myo-inositol, 2g/l Mes/KOH pH 5.7, 4% (w/v) sucrose, 0.1 mg/l gibberellic acid and 5 mg/1 .alpha.-napthalene-acetic acid. The cells were subcultured weekly in this medium without gibberellic acid.

2. Purification and Deglycosylation and Sequencing of AGPs

Cells of N. alata were removed from the culture medium by filtration through two layers of Miracloth. The supernatant was centrifuged (10,000.times.g; 50 min) to remove any cell debris. To the supernatant, NaCl and .beta.-glucosyl Yariv reagent Yariv et al. 1967) were added to a final concentration of 1% and 0.2%, respectively. The AGP-Yariv complex was pelleted by centrifugation (10,000.times.g; 50 min), washed twice with 1% NaCl, followed by centrifugation as above. The pellet was dissolved in H.sub.2 O and undissolved material removed by centrifugation (10,000.times.g; 20 min). The AGP-Yariv complex was re-precipitated by adding NaCl to 1%, and the precipitate washed and redissolved in H.sub.2 O. The Yariv precipitation and NaCl wash steps were repeated twice. The AGP-Yariv precipitate was finally dissolved in H.sub.2 O and sodium dithionite (30%) was added to disrupt the AGP-Yariv complex. The volume of the sample was reduced by Diaflo (YM30 membrane; Mr 30,000 Dalton cut off) filtration and the solution desalted by passage through a PD10 column (Pharmacia) equilibrated with 10 mM NH.sub.4 HCO.sub.3.

AGPs from N. alata were deglycosylated by trifluoromethane sulphonic acid (TFMS) using a modification of the procedure of Edge et al. (1981). The deglycosylated AGPs were separated on 17.5% SDS-PAGE according to Laemmli (1970). The 17.5% SDS-PAGE gels were run at 200 V with thioglycollic acid (1 mM) in the upper reservoir until the tracking dye reached the bottom of the gel. The peptides were transblotted onto a PVDF membrane with blotting buffer [10 mM 3-(cyclohexylamino)-1-propanesulfonic acid (CAPS) buffer pH 11, 15% methanol, thioglycollic acid (70 .mu.1/l)]. Blotting was for 1.5 h at 90 V with cooling. The blot was stained with 0.1% Coomassie Blue in 50% methanol, 10% acetic acid for 5 min and de-stained in 50% methanol, 10% acetic acid for 5 min. The blot was washed with distilled water overnight and bands excised and sequenced. A major band having a molecular weight of approximately 20-30 kD was obtained from the deglycosylated N. alata AGPs.

3. Sequencing

Purified protein was chromatographed on a reverse phase HPLC microbore column prior to automated Edman degradation on a gas phase sequencer [Mau et al. (1986), Planta 169:184-191]. Phenylthiohydantoin amino acids were analyzed by HPLC, as described by Grego et al. (1985), Eur. J. Biochem. 264:857-862. An N-terminal amino acid sequence, A-K-S-K-F-M-I-I-P-A-S-X-T-X-A (SEQ ID NO:11) was obtained.

(b) Cloning of Genes from N. alata and N. plumbaginafolia Cell Cultures

1. In Vitro Amplification of 5' End of the cDNA

Total RNA (10 .mu.g) from N. alata suspension cultured cells was mixed with 1.0 pmoles gene specific radioactive primers in 10 .mu.l of 40 mM PIPES (pH 6.0), 1 mM EDTA and 0.4M NaCl. The mixture was heated at 80.degree. for 5 min and incubated at 37.degree. overnight. The RNA/primer mixture was precipitated by ethanol and resuspended in 20 .mu.l of reverse transcription buffer containing: 50 mM Tris-HCl (pH 8.3), 60 mM KCl, 10 mM MgCl.sub.2, 1 mM DTT, 20 U RNase inhibitor and 50 U AMV reverse transcriptase. After 1 h incubation, the reaction was stopped by addition of EDTA. The RNA was removed by treatment with RNase and the primer extension product was purified by polyacrylamide gel electrophoresis.

The primer extension product was tailed with dGTP by terminal transferase and amplified by PCR using a (dC).sub.15 -adaptor primer and the gene specific primers. The PCR was carried out in 100 .mu.l solution containing: 1.times. PCR buffer (100 mM Tris-HCl pH 8.3, 500 mM KCl), 2 mM MgCl.sub.2, 200 .mu.M dNTPs, 100 ng poly dC primer, 100-200 ng of gene-specific primer and 2.5 U of Taq DNA polymerase. Samples were denatured by boiling for 5 min and then cooled to 80.degree. before Taq DNA polymerase was added. The PCR cycles are: 25.times.: 93.degree., 30 sec.; 42.degree., 30 sec.; 72.degree. C., 2 min; 4.times.: 93.degree., 30 sec.; 42.degree., 30 sec.; 72.degree., 5 min and 1.times.: 93.degree., 30 sec; 42.degree., 30 sec; 72.degree., 10 min. The PCR product was subcloned and sequenced.

2. In Vitro Amplification of 3'-End of the cDNA

cDNA was synthesized in a volume of 20 .mu.l solution containing 10 .mu.g total RNA, 1.times. PCR buffer, 50 mM MgCl.sub.2, 10 mM dNTPs, 5 .mu.M of dT(.sub.17)+adaptors, 30 U of RNasin and 50 U AMV reverse transcriptase at 42.degree. for 1 h. cDNA (2 .mu.l) was subjected to PCR reaction described as above, except that the annealing temperature was 60.degree. in this case.

3. Screening of cDNA Libraries with the PCR Fragment

About 5.times.10.sup.4 pfu phage/plate of cDNA libraries (in .lambda.zap) were plated out. After overnight growth at 37.degree., phage were blotted onto nitrocellulose membranes and hybridized with .sup.32 p-labeled DNA fragment at 68.degree. overnight in a hybridization buffer containing 2.times. SSPE, 1% SDS, 0.5% BLOTTO, 1% PEG and 0.5 mg/ml carrier DNA [Sambrook et al. (1989) supra]. The membranes were washed at 68.degree. for 30 min in 1.times. SSC+0.1% SDS and exposed to X-ray film. Positive .lambda.zap clones were converted into plasmid DNA by in vivo excision as described in the Stratagene's instruction manual for the sequence analysis.

4. Purification and N-terminal Sequencing of AGPs from the Cell Suspension Culture

The purified AGPs were deglycosylated with TFMS and the resulting peptides separated on a 17.5% SDS-PAGE gel and blotted onto a PVDF membrane. The major band (MW: 20-30 kD) (FIG. 1C) was excised and sequenced. An N-terminal peptide sequence: A-K-S-K-F-M-I-I-P-A-S-X-T-X-A (SEQ ID NO:11), was obtained.

5. In Vitro Amplification of an AGP Gene from N. alata cDNA by PCR

The strategy to clone the gene corresponding to the peptide sequence is illustrated in FIGS. 1D-1 and 1D-2. Two groups of degenerate reverse primers of 17 bp corresponding to part of the AGP amino acid sequence were synthesized (Table 1.1). When the group 1 primers were used in a primer extension experiment (FIG. 1D-1), a single 160-bp cDNA fragment was obtained (FIG. 1E). The primers of group 1 were further divided into six subgroups each containing three 17-mers (Table 1.1). Primer extension experiments showed that group NaR1 gave the highest yield of the 160-bp fragment and these oligonucleotides were therefore used as the gene-specific primer in subsequent scale-up preparation of primer extension product and PCR experiments. The 160-bp primer extension product was purified and tailed with dGTP. The tailed, single-stranded cDNA was then amplified by PCR with the oligo NaR1 and a (dC).sub.15 -adaptor as primers (FIG. 1D-1). The PCR fragment was subcloned and sequenced (SEQ ID NO:21; FIG. 1E). The sequence included a derived peptide which matched with the sequence obtained from the isolated AGP peptide (SEQ ID NO:11). There was one mismatch, the Ala obtained from the peptide sequencing was replaced with an Arg in the cDNA derived sequence. On the basis of this close match (8/9 amino acids), the 160-bp fragment was concluded to represent a correct sequence for part of the gene.

Two specific primers with sequences: 5'CATTATGGGTCATTTCACTAAGC3' (SEQ ID NO:22) (NaF1); 5 5'GGTGATCTCAACTCCATTGGTGC3' (SEQ ID NO:23) (NaF2), corresponding to positions 56-78 and 101-123 (FIG. 1E) were then designed and used in conjunction with the two 3'-end nonspecific primers (Ad1 and Ad2) to amplify the 3'-part of the AGP gene by nested PCR (FIG. 1D-2). A 1.6-kb fragment was amplified and sequenced. The alignment of the sequences obtained from the two PCR reactions gave rise to a DNA sequence of 1679 bp (FIGS. 1F-1 and 1F-2). The PCR fragment encodes a protein containing the peptide obtained by protein sequencing with two mismatches: Arg for Ala at position 1 and Pro for His at position 12 (FIGS. 1F-1 and 1F-2).

6. Isolation and Sequence Analysis of cDNA Clones from N. alata and N. plumbaginafolia cDNA Libraries

The 1.6-kb PCR fragment was used to screen a cDNA library made from RNA isolated from N. alata cells in suspension culture and three positive clones were isolated and sequenced. The alignment of the PCR sequence with the cDNA sequences gave rise to a 1700-bp sequence including a poly(A) tail of 7 bp (FIGS. 1F-1 and 1F-2). This sequence is designated NaAGP1 (SEQ ID NO:24). Further primer extension experiments suggested that the 1.7 kb NaAGP1 cDNA represents the full-length sequence of the AGP transcript.

The NaAGP1 cDNA encodes an open reading frame, which starts with an initiation codon (ATG) at position 60 and ends with a termination codon (TAA) at position 1443 (FIGS. 1F-1 and 1F-2). The open reading frame encodes a polypeptide containing 461 amino acid residues with a calculated molecular weight of 51.8 kD and a predicted pI of 3.84. The protein is highly rich in asparagine (25%), and relatively rich in serine (8.9%), tyrosine (7.5%), proline (7.2%) and glutamine (7.0%) (Table 1.2), and can be divided into four domains (FIG. 1G). There is a putative transmembrane helix at the N-terminus (1-25), which is very hydrophobic. The next one-third of the protein (26-173 aa) is also hydrophobic and contains most of the proline (93.8%), alanine (76.5%) and threonine (76.2%) residues. These three amino acids account for 39.7% of all the amino acids in this domain (Pro, 20.2%; Thr, 10.8% and Ala, 8.7%) (FIG. 1G). This domain is predicted to be the site of glycosylation by Gal/Ara containing chains, linked through hydroxyproline residues. The proline residues (Nos. 25, 27, 29 and 31 in FIGS. 1H-1 and 1H-2) are known to be hydroxylated, as they appear as hydroxyproline in the peptide sequence obtained from deglycosylated AGPs of N. plumbaginafolia. Such hydroxylation and glycosylation would make the molecule considerably more hydrophilic.

The portion of the protein corresponding to amino acid positions 174-436 is hydrophilic and contains most of the asparagine (95.1%) and tyrosine (94.1%) residues which account for 44.1% and 12.1%, respectively, of all amino acids in this domain (FIGS. 1F-1, 1F-2 and 1G). The asparagine residues are distributed in clusters (2-11 amino acids) along the polypeptide chain. It contains no proline residues. The final 25 residues at C-terminus are hydrophilic (FIG. 1G).

An N. plumbaginafolia cell suspension cDNA library was also screened with the PCR-fragment, and four cDNA clones were isolated and sequenced. The four clones were identical and contained an insert of 1430 bp (FIGS. 1H-1 and 1H-2). This AGP gene is designated NpAGP1 (SEQ ID NO:25). These cDNAs are incomplete and predicted to be about 100 bp shorter at the 5'-end than the full-length sequence of the transcript. The NpAGP1 is not identical, but very similar to, the NaAGP1 at both the nucleotide and derived amino acid sequence levels (86% and 84.7% identity, respectively) (FIGS. 1I-1, 1I-2, and 1J-1 through 1J-4, and Table 1.2). The transmembrane helix is missing in the NpAGP1 cDNA due to the incomplete sequence. The difference between the two AGP genes is mainly in the middle one-third of the sequence while the N-terminal and C-terminal parts are highly conserved (FIGS. 1I-1, 1I-2 and 1J-1 through 1J-4).

7. Northern and Southern Blot Analyses of the Putative AGP Gene

The NaAGP1 was cut into a 5' half (1-540 bp) corresponding to the nontranslated part, the transmembrane helix and the proline-rich domain and a 3'-half (541-1700 bp) including the asparagine-rich domain, C-terminus and the 3'-nontranslated part. These two parts of the cDNA were used separately to probe northern blots of RNA [Sambrook et al. (1989) supra] isolated from suspension-cultured cells of N. alata and N. plumbaginafolia and various tissues of N. alata plants. The two probes gave an identical hybridization pattern, confirming that these two distinct domains are parts of the same transcript (FIGS. 1K-1 and 1K-2). The NaAGP1 cDNA probes hybridized to the RNA samples from all the tissues tested, although the degree of hybridization and size of transcripts are different in different tissues. The highest signal was detected in RNA from N. alata suspension-cultured cells whereas the signal in petals was barely detected. Pollen and style tissues had a smaller transcript of approximately 1.0 kb compared with 1.6 kb in N. plumbaginafolia cultured cells and 1.7 kb in all other tissues (FIGS. 1K-1 and 1K-2). Genomic southern blot analysis indicated that the AGP gene is a single copy or low copy gene in the genome of N. alata.

(c) Isolation and Purification of AGP Peptides from Suspension Cultures of Nicotiana plumbaginafolia

1. Isolation of Total Native AGPs from N. plumbaginafolia Biopolymer

The total native AGPs were purified from the Biopolymer product by precipitation with the Yariv reagent after depleting the starting material of pectins by CTAB (hexadecyl trimethyl ammonium bromide) precipitation prior to Yariv precipitation. The medium from the cell suspension culture was separated from the cells by filtration and the high molecular materials precipitated with four volumes of ethanol. This is referred to as the Biopolymer product.

Biopolymer product (1 g) was dissolved in 1% NaCl solution (100 ml) and filtered through two layers of Miracloth. The filtrate was centrifuged (10,000.times.g, 10 min) and the supernatant collected. An equal volume of CTAB solution (2% CTAB in 20 mM Na.sub.2 SO.sub.4) was added. After 1 h incubation at 37.degree., the solution was filtered through two layers of Miracloth and then centrifuged (10,000.times.g, 20 min) to remove any remaining precipitate. Four volumes of ethanol were then added to the supernatant and centrifuged at 10,000.times.g for 20 min. The pellet was dissolved in 100 ml of 1% NaCl solution and AGPs precipitated with Yariv reagent as described in Example 2(a) 2. The desalted AGP sample was re-dissolved in 6M guanidinium-HCl and incubated at 50.degree. for 15 min. The sample was then chromatographed on a FPLC Superdex.TM.75 column equilibrated with 6M urea and 20 mM Tris-HCl, pH 8.8. The void (Vo) fraction was collected, dialysed against distilled water and freeze dried. This sample is the total native AGPs. The total native AGPs were treated by one of two paths:

Path 1: Deglycosylation followed by reverse phase HPLC fractionation before direct sequencing, or sequencing after enzymatic (proteolytic) digestion.

Path 2: Reverse phase HPLC fractionation followed by deglycosylation and further separation by reverse phase HPLC fractionation.

Path 1 [Comprising Steps (2)-(5)]

(2) Deglycosylation of Total Native AGPs Using Anhydrous HF

The AGP sample was dried in a vacuum oven at 40.degree. in the presence of P.sub.2 O.sub.5 overnight; 0.2 ml anhydrous MeOH and 1 ml of anhydrous HF [Mort and Lamport (1977) Anal. Biochem. 82:289-309] was added and mixed well to dissolve all the sample. This mixture was incubated at room temperature, under argon, for 3 h and the HF removed by vacuum aspiration. Ice cold TFA (0.5 ml) was added and the sample desalted on a PD10 column equilibrated with 0.1% TFA, and freeze dried. This sample is referred to as the total deglycosylated AGPs.

(3) Reduction and Carboxymethylation of the Total Deglycosylated AGP Sample

The total deglycosylated AGP sample was dissolved in 6M guanidinium-HCl (in 0.2M Tris-HCl, pH 8.5 and 20 mM DTT; 600 .mu.l); and incubated at 25.degree. under argon for 2 h. Freshly prepared iodoacetic acid (100 .mu.l) was added. The mixture was incubated for 3 h at 25.degree. and stopped by addition of DTT to 100 mM and following dilution was chromatographed as above.

(4) HPLC Separation of the Total Deglycosylated AGPs

After reduction and carboxymethylation, the total deglycosylated AGPs were separated on a RP-300 HPLC column with a linear gradient (60 ml) (0-100% solvent B; flow rate 1 ml/min) (solvent A: 0.1% TFA in water, solvent B: 60% acetonitrile in solvent A). The profile is shown in FIG. 2A. Two major peaks RT21 and RT32 (retention times 21 min and 32 min, respectively) were collected for further analysis. Amino acid analysis was performed on both peaks (see Table 2.1). The RT32 peak was sequenced without further treatment. The RT21 peak was subjected to thermolysin digestion before sequencing.

(5) Thermolysin digestion of RT21

RT21 sample (12 .mu.g) was concentrated and Tween 20 added to give a final volume of 100 .mu.l with a final concentration of 0.01% Tween 20. NH.sub.4 HCO.sub.3 (1% in 0.01% Tween 20; 500 .mu.l), CaCl.sub.2 (0.1M; 7 .mu.l) and thermolysin (1 mg/ml; 7 .mu.l) were added and the mixture incubated at 55.degree. for 3 h. The products were purified on reverse phase HPLC and sequenced. The peptide sequences obtained are shown in FIG. 2A and were used to construct primers for cloning. The sequences L--A--S--O--O--A--O--O--T--A (SEQ ID NO:26), L--A--S--O--O--A--O--O--T--A--D--T--O--A (SEQ ID NO:27), F--A--O--S/N--G--G--V--A--L--P--O--S (SEQ ID NO:28), and I--G--A--A--O--A--G--S--O--T--S--S--P--N (SEQ ID NO:29) from RT21 are either similar to or identical with that obtained from fraction RT25 of N. alata styles (FIGS. 4A-4C) and represent conserved, tissue-nonspecific N. alata AGP fragments.

Peak 32 gave the sequence R--K--S--K--F--M--I--I--P--A--S--O--T--O--A--O--T--O--I--N--E--I--S--F (SEQ ID NO:30) which at the 5'-end, matched very closely the N-terminal sequence (SEQ ID NO:1) obtained from N. alata suspension culture.

Path 2 [Comprising Steps (6)-(8)]

(6) HPLC fractionation of total native AGPs

The total native AGPs sample was dissolved in 6M guanidinium-HCl and left at 50.degree. for 15 min. The sample was then fractionated on reverse phase HPLC (RP-300; 4.6 mm.times.10 cm column) with a linear gradient (60 ml) (0-100% solvent B; flow rate 1 ml/min) (solvent A: 0.1% TFA in water, solvent B: 60% acetonitrile in solvent A). A number of major peaks were obtained from this separation all of which reacted with Yariv reagent in a gel diffusion test (van Holst and Clarke, 1985) (unbound, RT5, RT6, RT10, RT21-23 and RT34) (FIG. 2B). Each fraction was quantified for AGP content (Table 2.1) as described by van Holst and Clarke (1985). Amino acid analyses of each fraction of native AGPs are shown in the Table 2.1.

(7) Deglycosylation of native AGP fractions from HPLC

Individual native AGP fractions from reverse phase HPLC (FIG. 2B) were deglycosylated using anhydrous HF as described above.

(8) HPLC separation of the deglycosylated AGPs

After deglycosylation, each sample was reduced and carboxymethylated before reverse phase HPLC separation (FIGS. 2C and 2D). The fractions obtained were reserved for further sequencing.

EXAMPLE 3

Cloning of a Gene Encoding a Protein Backbone of an AGP from P. communis Suspension Cultured Cells

(a) Isolation and Purification of AGP Peptides From Cell Cultures of Pyrus communis (Pear)

1. Isolation of total native AGPs from Pyrus communis (pear) Biopolymer

The total native AGPs were purified by Yariv precipitation from pear Biopolymer as described for AGPs of Nicotiana plumbaginafolia in Example 2(c)1. The AGPs were deglycosylated and resulting peptides separated by reverse phase HPLC (RP-300) (Path 1). Alternatively, the total native AGPs were fractionated by reverse phase HPLC (RP-300), deglycosylated, digested with thermolysin and peptides purified for sequencing.

Path 1 [Comprising Steps (2) and (3)]

(2) HPLC separation of total deglycosylated AGPs for sequencing

The total native AGPs were deglycosylated using HF. The sample was reduced and carboxymethylated before separation on reverse phase HPLC (RP-300) as described in Example 2(c)(2). The profile is shown in FIG. 3A. The results of amino acid analysis of major peaks are summarized in Table 3.1.

(3) Separation of thermolysin digested peaks on a C18 microbore column

Deglycosylated AGP fractions (unbound, RT16.4 and RT18.2 from FIG. 3A) were subjected to thermolysin digestion. The products were separated on an RP-300 column (2.1 mm.times.10 cm); linear gradient (6 ml) (0-100% B; flow rate at 0.1 ml/min) (solvent A: 0.1% TFA in water, solvent B: 60% acetonitrile in solvent A). The unbound fraction after digestion remained unbound, i.e., gave no peptide which bound to the RP-300 column. The RP-300 profile for digested RT16.4 is shown in FIG. 3B and for RT18.2 is shown in FIG. 3C.

Individual peaks (peaks 1-5, FIG. 3B) from thermolysin digested RT16.4 (FIG. 3A) were separated on a C18 microbore column (2.1 mm.times.10 cm) and resolved on a linear gradient (50 ml, 0-50% B; flow rate 0.1 ml/min) (solvent A: 1% NaCl, solvent B: 100% acetonitrile). Peaks were further separated on the same column with TFA-acetonitrile system (solvent A: 0.1% TFA, solvent B: 60% methanol in solvent A; 0-100% B in 60 min at 0.1 ml/min). Neither solvent system gave further separation of peaks. Three of the peaks (peaks 1, 3 and 5) were subjected to amino acid sequencing. Peak 1 was a pure peptide and gave clear sequence L--S--O--K--K--S--O--T--A--O--S--O--S--(S)--T--O--O--T--(T) (SEQ ID NO:31). Peaks 3 and 5 were not single peptides and at least two stretches of sequence were obtained from each of these two peaks with less certainty. Peak 3 gave the sequence: ##STR12## whereas Peak 5 gave the sequence: ##STR13##

Individual peaks (peaks 1-5, FIG. 3C) from thermolysin digested RT18.2 were separated as described above for RT16.4. A number of peptides were obtained and sequenced: ##STR14## The last sequence is identical to the sequence obtained from Peak 1 of RT16.4.

Path 2 [Comprising Steps (4)-(7)]

(4) Fractionation of total native AGPs by reverse phase HPLC

The total native AGPs samples were separated by reverse phase HPLC essentially as described in Example 2(c)2-4. A number of major peaks were obtained from this separation all of which reacted with Yariv reagent in a gel diffusion test (van Holst and Clarke, 1985) (unbound, RT7,8, RT17.2 and RT19.1) (FIG. 3D). Amino acid analyses of unbound and RT7.8 fractions are shown in the Table 3.1.

(5) Deglycosylation of native AGP fractions from HPLC

Individual native AGP fractions from reverse phase HPLC were deglycosylated using anhydrous HF as in Example 2(c)(7).

(6) HPLC separation of the deglycosylated AGPs

After deglycosylation, each sample was reduced and carboxymethylated before separation on reverse phase HPLC (RP-300) as described previously. The profiles of each sample are shown in FIG. 3E and FIG. 3F. The major peaks RT16-19 in FIG. 3F have similar retention times with the group of peaks RT16-19.9 in FIG. 3A. These peaks may arise from the one component or a group of closely related components.

(7) Thermolysin digest of deglycosylated pear AGPs

Peak RT23 from FIG. 3E was digested with thermolysin and the resulting peptides were further purified on reverse phase HPLC (RP-300). Six peptides were selected for sequencing and gave the following amino acid sequences (also shown in FIGS. 3A-3F): ##STR15##

(b) Cloning of Genes From P. communis Cell Suspension Culture

To obtain an AGP gene from P. communis the methods and procedures essentially as described for the cloning of genes from N. alata and N. plumbaginafolia were followed.

A number of primers corresponding to the L--V--V--V--V--M--T--P--R--K--H (SEQ ID NO:41) sequence (FIGS. 3D and 3E) were designed and synthesized for PCR experiments (Table 3.2). The same nested PCR procedure used for the cloning of the NaAGP1 gene (FIG. 1D-2) was used to clone the gene encoding the above peptide, except that the annealing temperature was 52.degree. C. in this case. A 350-bp fragment was amplified after two successive PCR reactions using the PcA23F1 as the first primer and the PcA23F2a as the second primer. The fragment was sequenced and found to encode the correct peptide sequence (SEQ ID NO:48; FIG. 3G).

The PCR fragment was used to screen a cDNA library made from mRNA from pear cell suspension culture, as described above for N. alata cell suspension. One positive clone (PcAGP23) was isolated and sequenced (SEQ ID NO:49; FIGS. 3H-1 and 3H-2). This clone contains an insert of 760 bp and matches the PCR sequence.

The PcAGP23 cDNA (SEQ ID NO:49) encodes an open reading frame, which starts with an initiation codon (ATG) at position 20 and ends with a termination codon (TAG) at position 560 (FIGS. 3H-1 and 3H-2). The open reading frame encodes a polypeptide containing 180 amino acid residues with a calculated molecular weight of 19.2 kD and a predicted pI of 8.46. The predicted amino acid sequence contains the peptide sequence, L--V--V--V--V--M--T--P--R--K--H (SEQ ID NO:41), which was used for the cloning of the PCR fragment. In addition, another peptide sequence, L--G--I--S--O--A--O--S--O--A--G--E--V--D--(G) obtained from RT18.2 (SEQ ID NO:34) is also present. However, other sequences, e.g., I--S--O--A--S--T/Q--O--O--T--T--S--O--A--S--O--O--T (SEQ ID NO:39); and V--S--P/S--O--V--Q--S--O--A--S--O--O--O--T--(T) (SEQ ID NO:40), obtained from peak RT7.8, are absent from the PcAGP23 sequence, indicating they are from different AGP backbones.

EXAMPLE 4

Cloning of Genes Encoding a Protein Backbone of an AGP From Nicotiana alata Style

(a) Isolation and Purification of AGP Peptides From the Styles of Nicotiana alata

Total native AGPs of N. alata styles were purified by ion exchange chromatography (IEC) and gel filtration chromatography (GFC). The AGPs were then deglycosylated by HF and fractionated by reverse phase HPLC. Peptide sequence data were obtained after thermolysin digestion of these deglycosylated fractions.

1. Purification of total native AGPs

Styles (500-1000 styles including the stigma) were collected fresh or were stored at -70.degree. C. The styles were ground with polyvinyl pyrrolidone (1% w/v) in the presence of liquid nitrogen, and extraction buffer (50-100 ml; 100 mM Tris pH8, 1 mM EDTA, 14 mM .beta.-mercaptoethanol) was added. The mixture was centrifuged (10,000 xg, 20 min) and cell debris discarded. The extract was brought to 95% ammonium sulfate at 4.degree., centrifuged (10,000 xg, 20 min) and the supernatant collected and concentrated by ultrafiltration using a Diaflo system (YM-30 membrane, Mr 30 kD cut off) to about 10-20 ml. The solution was desalted on a PD-10 column (Pharmacia) equilibrated with 10 mM Tris pHS. The sample was applied to a FPLC Mono Q column (Pharmacia; buffer A: 10 mMTris pH8; buffer B: 10 mM Tris pH8, 1M NaCl; gradient: 0-30% B 15 min, 30-100% B 0.1 min). The bound AGP fractions were detected by the Yariv reagent gel diffusion test on samples of each fraction; AGP containing fractions eluted at about 5-15% buffer B (FIG. 4A). The AGP fractions were pooled, equilibrated into 10 mM NH.sub.4 HCO.sub.3 with a PD-10 desalting column, freeze dried, and further purified on a Superose 6 .beta. column (Pharmacia) in 6M urea, 10 mM Tris pH8 (FIG. 4B). The AGP containing fractions were exchanged as above into 10 mM NH.sub.4 HCO.sub.3 and freeze dried.

Recovery of style AGP during the purification procedure is as follows: crude style extract (1000 styles), 100% ; 95% (NH.sub.4).sub.2 SO.sub.4 -supernatant, 68.2% ; Mono Q anion-exchange column, Unbound AGPs 5.4% , Bound AGPs 44.5% ; Superose 6 gel filtration column, 25.4% . The presence of AGPs at different stages of purification is demonstrated on SDS-PAGE gels in FIGS. 4N-1 and 4N-2. Crossed-electrophoresis of AGPs from styles of N. alata during fractionation is presented in FIGS. 40-1 through 40-4.

2. Deglycosylation of total native AGPs and sequencing of peptides

Deglycosylation, peptide cleavage and sequencing were performed as described in Example 2(c)2. Two major peaks, RT25 and RT35 (FIG. 4C), were obtained after deglycosylation as well as an unbound fraction. Amino acid analysis of each fraction and the native materials are shown in Table 4.1. Each fraction was digested with thermolysin. No peptide which bound to the RP-300 column (2.1.times.100 mm) was obtained from the unbound fraction. Three of the sequences from RT25, F--A--O--S--G--G--V--A--L--P--O--S (SEQ ID NO:50), L--A--S--O--O--A--O--O--T--A--D--T--O--A (SEQ ID NO:51), and I--G--S--A--O--A--G--S--O--T--S--S--P--N (SEQ ID NO:53) match closely that obtained for RT21 from N. plumbaginafolia (SEQ ID NOS:27-29, respectively; FIG. 2A). A fourth fragment gave the sequence I/V--G/S--A/S--A/O--O/S--A/Q--G/S--S/O--O/S--T/A--S/A--S/A--P--O (SEQ ID NO:52).

Since no N-terminal sequence was obtained for the RT25 protein backbone, pyroglutamate aminopeptidase was used to remove the N-terminal blocked pyroglutamate residue [20 .mu.g pyroglutamate aminopeptidase (Boehringer Mannheim) in 100 mM potassium phosphate buffer pH 8.0, 10 mM EDTA, 5 mM DTT, 5% glycerol at 37.degree. C. overnight; deblocked protein was separated by RP--HPLC and N-terminal amino acid sequencing was performed] and the sequence Ala--Hyp--Gly was obtained. The RT25 backbone was also fragmented by treatment with thermolysin [thermolysin (Boehringer Mannheim) at 0.2 .mu.g/.mu.g protein was added to RT25 protein backbone (2-10 .mu.g) and incubated at 55.degree. C. for 2 hours in 500 .mu.l of 1% ammonium bicarbonate, pH 7.8, 1 mM CaCl.sub.2 and 0.01% Tween 20]. The resulting peptides were separated by RP--HPLC. Six major peptides were obtained (FIG. 4I). Peak 2 gave the amino acid sequence VSAOSQSOSTAA (SEQ ID NO:67), as well as IGSAOAGSOTSSPN (SEQ ID NO:53) and IGSAOAGSO (contained in SEQ ID NO:53). Peak 3 gave the sequence LASOOAOOTADTOA (SEQ ID NO:51) and peak 5 gave the sequence FAOSGGVALPOS (SEQ ID NO:50). Both sequences were rich in Hyp, Ser and Ala (33 of 52 amino acid residues).

Endoproteinase Asp--N (Sigma; 0.1 .mu.g/.mu.g protein) was also used to cleave the RT25 protein backbone at the Asp residue [30.degree. C. overnight in 500 .mu.l of 1% ammonium bicarbonate, pH 7.8, and 0.01% Tween 20], followed by separation with RP--HPLC. Two major peptides were produced (peaks A1, A2; FIG. 4J), indicating that there is only one Asp residue in the RT25 protein. The cleavage was incomplete as indicated by the presence of the starting material. The first peptide eluted (peak A1) gave no sequence data indicating a blocked N-terminal residue. The A2 peak gave the sequence DTOAFAOSGGVAL (SEQ ID NO:68). The peptide sequence of A2 (FIG. 4J) overlaps with that of peak 3 (SEQ ID NO:51) (FIG. 4I) and yields a continuous amino acid sequence of 26 residues LASOOAOOTADTOAFAOSGGVALPOS (SEQ ID NO:69).

Four sequences were obtained from the RT35 peak of N. alata style: ##STR16## Three of these sequences were characterized by the sequence T--A--I--N--T--E--F--G--P (SEQ ID NO:58).

3. Purification of style AGPs by J539 affinity chromatography

AGPs were prepared from styles according to Bacic et al. (1988), Phytochem. 27:679-684. The sample was deglycosylated with TFMS, separated and blotted onto a PVDF membrane as described in Example 1(b). A 30 kD band, running at the same position as the major band prepared by Yariv precipitation from N. alata suspension cultured cells Example 1(b) was sequenced. The sequence A--V--F--K--N--K--X--X--L--T--X--X--P--X--I--I (SEQ ID NO:59) was obtained.

(b) Cloning of Genes From N. alata Styles

1. In vitro amplification of 3'-end of the cDNA

cDNA was synthesized in a volume of 20 .mu.l solution containing 5 .mu.g total style RNA from N. alata, 1.times.PCR buffer (10 mM Tris-HCl pH 8.3, 50 mM KCl), 5 mM MgCl.sub.2, 1 mM dNTPs; 5 .mu.M of dT.sub.(17) +adaptors, 30 U of RNasin and 50 U AMV reverse transcriptase at 42.degree. for 1 h cDNA (2 .mu.l) was subjected to polymerase chain reaction. The PCR was carried out in 100 .mu.l solution containing: 1.times.PCR buffer, 1.5 mM MgCl.sub.2, 200 .mu.M dNTPs, 30 pmole of the gene-specific primer (FIGS. 1D-1 and 1D-2) 30 pmole of adaptor primer and 2.5 U of Taq DNA polymerase. Samples were denatured by heating at 94.degree. for 2 min and then cooled to 80.degree. before Taq DNA polymerase was added. The PCR cycles are: 35 X: 94.degree., 30 sec; 52.degree., 30 sec; 72.degree., 1 min 30 sec. The PCR product was subcloned and sequenced.

2. Screening of cDNA library with the PCR fragment

About 5.times.10.sup.4 pfu phage/plate of cDNA libraries (in .lambda.zap) were plated out. After overnight growth at 37.degree., phage were blotted onto nitrocellulose membranes and hybridized with .sup.32 P-labeled PCR fragment at 65.degree. overnight in a hybridization buffer containing 0.22M NaCl, 15 mM NaH.sub.2 PO.sub.4, 1.5 mM EDTA, 1% SDS, 1% BLOTTO and 4 mg/ml carrier DNA [Sambrook et al. (1989) supra]. The membranes were washed at 65.degree. for 2.times.15 min in 0.2.times.SSC and 1% SDS and exposed to X-ray films. Positive .lambda.zap clones were converted into plasmid DNA by in vivo excision as described in the Stratagene's instruction manual for the sequence analysis.

3. Design of a gene-specific primer based on the AINTEFG sequence

As described in Example 2(c)2, pp. 37-38, the purified AGPs were deglycosylated with HF and the resulting AGP backbones were separated on reverse phase HPLC. Two major peaks: RT25 and RT35 were obtained after deglycosylation as well as an unbound fraction. Amino acid sequences were obtained from both peaks after protease digestion. Three of the four peptide sequences from peak RT35 contain the sequence: TAINTEFGP (SEQ ID NO:58). A degenerate oligonucleotide was synthesized based on the sequence: AINTEFG (SEQ ID NO:60). ##STR17## where I is an inosine residue

(a) In Vitro Amplification of an AGP Gene From N. alata cDNA by PCR

The strategy to clone the gene encoding the RT35 peptide sequence is illustrated in FIG. 4D. The RT35-specific primer was used in conjunction with the adaptor primer in a polymerase chain reaction and a single 380-bp DNA fragment was obtained. The PCR fragment was subcloned and sequenced (SEQ ID NO:62; FIG. 4E). The sequence included a derived peptide ID NO:62; FIG. 4E). The sequence included a derived peptide that matched with the sequence obtained from the isolated AGP peptide.

(b) Isolation and Sequence Analysis of a cDNA Clone From N. alata

The 380-bp PCR fragment (SEQ ID NO:62; FIG. 4E) was used to screen a cDNA library made from RNA isolated from N. alata styles and one positive clone was isolated and sequenced (SEQ ID NO:63; FIGS. 4F-1 and 4F-2). The insert of the cDNA clone is 800 bp in length with a poly(A) tail at the 3'-end. The cDNA sequence matches the PCR sequence except that it is 3 bp shorter at the 3'-end. The cDNA sequence is designated Na35.sub.-- 1.

The Na35.sub.-- 1 sequence (SEQ ID NO:63) has an open reading frame which starts with an initiation codon (ATG) at position 21 and ends with a termination codon (TAA) at position 530 (FIG. 4E). The open reading frame encodes a polypeptide containing 169 amino acid residues with a calculated molecular weight of 19.5 kD and a predicted pI of 8.1. The most abundant amino acid residues in this sequence are: proline (11.2% ), phenylalanine (9.5% ), alanine (7.7% ), leucine (7.7% ) and lysine (7.7% ) (Table 4.2).

(c) Northern Blot Analyses of the Na35.sub.-- 1 Gene

The Na35.sub.-- 1 PCR fragment was used to probe northern blots of RNA [Sambrook et al. (1989) supra] isolated from various parts of N. alata plants (FIG. 4G) L. peruvianum (tomato) style and suspension-cultured cells of N. alata, N. plumbaginafolia and pear (FIG. 4H). The Na35.sub.-- 1 probe hybridized to a style transcript of 800 nucleotide which corresponds to the length of the Na35.sub.-- 1 cDNA. Longer exposure of the northern blot did not reveal any signal in other parts of the plant (i.e., leaf, stem, root). The signal strength varies in different genotypes of N. alata. The strongest signal was detected in RNA from S.sub.6 S.sub.6 style. The same probe did not detect any transcript from tomato style or suspension-cultured cells (FIG. 4H).

4. Design of a gene-specific primer based on the TADTOAF sequence

(a) Oligonucleotide Design and Synthesis

A gene-specific primer of 20 nucleotides long was designed according to the overlapping peptide sequences of SEQ ID NOS:50, 51, 53, 67 and 68. Inosine was used to reduce the degeneracy as shown: ##STR18##

The oligonucleotide was synthesized on an Applied Biosystems DNA synthesizer (model 391, ABI).

(b) Rapid Amplification of 3' End of the cDNA (3' RACE)

Total RNA was isolated from N. alata styles as described by McClure et al. (1990) Nature 342:955-957. Complementary DNA (cDNA) was synthesized from total style RNA (5 .mu.g) in a 20 .mu.l solution containing 10 mM Tris-HCl, pH 8.3, 50 mM KCl, 5 mM MgCl.sub.2, 1 mM dNTPs, 5 .mu.M dT.sub.(17) +adaptors, 30 U RNasin, and 50 U AMV reverse transcriptase (Promega) at 42.degree. C. for 1 hour. cDNA (2 .mu.l) was subjected to polymerase chain reaction (PCR) in 100 .mu.l solution containing: 10 mM Tris-HCl, pH 8.3, 50 mM KCl, 1.5 mM MgCl.sub.2, 200 .mu.M dNTPs, 30 pmole of the gene-specific primer, 30 pmole of the adaptor primer and 2.5 U of Taq DNA polymerase (Perkin Elmer-Cetus). Samples were denatured by heating at 96.degree. C. for 2 min and then cooled to 80.degree. C. before Taq DNA polymerase was added. The PCR cycles were: 35.times.: 96.degree. C., 45 sec; 55.degree. C., 45 sec; 72.degree. C., 1 min. The PCR product (400 bp) was cloned and sequenced on an Applied Biosystems DNA sequencer (model 373A, ABI). The deduced amino acid sequence from this PCR clone matched isolated AGP sequences, i.e., SEQ ID NOS:50, 51, 53, 67, 68.

(c) cDNA Library Screening

A style cDNA library (.lambda.ZAP II; Stratagene) was constructed using mRNA from styles (6 hours after touching) of N. alata (S.sub.6 S.sub.6) by Dr. Joaquin Royo, Plant Cell Biology Research Center, School of Botany, The University of Melbourne, Parkville, Australia (PCBRC). cDNA library (300,000 pfu) was plated out and blotted onto Hybond-N nylon membranes (Amersham) according to the manufacturer's instruction. The PCR fragment was labeled to 10.sup.8 cpm/.mu.g with .sup.32 P-dCTP. Hybridization was carried out at 55.degree. C. overnight in 0.22M NaCl, 15mM NaH.sub.2 PO.sub.4, 1.5 mM EDTA, 1% SDS, 1% BLOTTO and 4 mg/ml herring sperm DNA. The membranes were washed for 2.times.10 min at room temperature in 2.times.SSC, 1% SDS followed by 2>10 min at 55.degree. C. in 0.2.times.SSC, 1% SDS. Positive .lambda.ZAP clones were in vivo excised (Stratagene) and DNA sequences were analyzed. The clone encoding the RT25 protein backbone was designated AGPNal 1 cDNA. The nucleotide and deduced protein sequences were analyzed using the PC/Gene software (IntelliGenetics).

The PCR clone, used as a probe to screen a style cDNA library (300,000 plaques), produced two cDNA clones which differed only in the length of the 3' and 5' ends. One of the clones, designated AGPNal 1 (FIGS. 4K-1 and 4K-2) was used for all subsequent analyses. The 3' end of the AGPNal 1 cDNA clone was identical to the PCR clone except that the PCR clone was 20 bp shorter and contains a poly A tail. The 712-bp AGPNal 1 clone encodes a putative protein of 12.5 kD (FIGS. 4K-1 and 4K-2). The derived amino acid sequence included sequences identical to isolated peptides (SEQ ID NOS:50, 51, 53, 67, 68). Most of the proline residues in the peptide sequences obtained by amino acid sequencing are hydroxylated. A secretion signal peptide is predicted (FIGS. 4K-1, 4K-2 and 4L). The deduced N-terminus of the mature protein (10 kD; pI 6.8) is Gln--Ala--Pro--Gly which matches the N-terminal sequence data obtained. The Pro residue in the N-terminal sequence is also hydroxylated. The amino acid composition of the deduced mature protein and the RT25 protein backbone are in general agreement. The C-terminus of the deduced protein is very hydrophobic and predicted to be a transmembrane helix.

(d) RNA Blot Analysis

RNA blot analysis was performed as described by Sambrook et al. (1989) supra. Hybridization and washing conditions were the same as described above except that the AGPNal 1 cDNA was used as probe and hybridization was carried out at 60.degree. C.

EXAMPLE 5

Cloning of an AGP Gene from P. communis Using an Antisense RNA Probe

1. The PcAGP9 cDNA Clone (SEQ ID NO:66)

(a) Isolation and purification of AGP peptides from cell cultures of Pyrus communis (pear)

The procedure essentially as described in Example 3(a) was followed to obtain amino acid sequences of AGP peptide fragments. The sequence A--K--S--O--T--A--T--O--O--T--A--T--O--O--S--A--V (SEQ ID NO:37) was selected as a template for the isolation of a corresponding AGP gene.

(b) Cloning of a Pear AGP Gene Encoding SEQ ID NO:37

In the previous examples of the invention (Examples 2, 3, and 4) AGP genes were isolated by utilizing a hydroxyproline-poor sequence of an isolated AGP peptide fragment to synthesize an oligonucleotide primer which was not enriched in GC. In contrast, in this example (Example 5), a hydroxyproline-rich peptide sequence is utilized for the construction of an antisense RNA probe.

The sequences of two oligonucleotide (AF1T3) and (AR2T7) used for the construction of a GC-rich probe are presented in Table 5.1 AF1T3 includes a T3 promoter sequence, 42-bp GC-enriched nucleotide sequence corresponding to an isolated AGP peptide fragment from N. plumbaginafolia, L--A--S--O--O--A--O--O--T--A--D--T--O--A (SEQ ID NO:27), and an 18-bp sequence corresponding to position 150-167 of the NaAGP1 (SEQ ID NO:24). The other oligonucleotide (AR2T7) consists of a T7 promoter, a 47-bp nucleotide sequence corresponding to a hydroxyproline-rich AGP sequence from pear, A--K--S--O--T--A--T--O--O--T--A--T--O--O--S--A--V (SEQ ID NO:37) and an 18-bp sequence corresponding to position 444-461 from the NaAGP1 cDNA (SEQ ID NO:24). These oligonucleotide primers were used to amplify the proline-rich domain (nucleotide position 150-461) of the NaAGP1 cDNA (SEQ ID NO:24).

An antisense RNA probe was synthesized from the PCR fragment by using T7 RNA polymerase (Promega) and used to screen a cDNA library prepared from pear cell suspension culture. The hybridization was carried out at 40.degree. C. in hybridization buffer containing 2.times.SSPE, 1% SDS, 0.5% BLOTTO, 50% formamide and 0.5 mg/ml denatured herring sperm DNA. After overnight hybridization, lifts were first rinsed at room temperature with 2.times.SSC, 0.1% SDS and then washed at 50.degree. C. with the same buffer for 30 min. The lifts were finally washed at 50.degree. C. with 1.times.SSC, 0.1% at 50.degree. C. for another 30 min. Three cDNA clones were isolated and sequenced. The sequence of the longest cDNA clone PcAGP9 (SEQ ID NO:66) is shown in FIGS. 5A-1 and 5A-2.

2. The PcAGP2 cDNA Clone (SEQ ID NO:91)

(a) Further purification of AGP peptides from cell cultures of Pyrus communis (pear)

AGPs in pear cell culture filtrate were purified by precipitation with the .beta.-glucosyl Yariv reagent and fractionated by HPLC as described in Example 3(a). A flow chart of the purification procedure is presented in FIGS. 5D-1 and 5D-2. The major peak of FIG. 5D-1, which accounted for approximately 27% of the AGPs loaded onto the column, was collected and reapplied to the same column. Upon elution with a shallow gradient, two peaks (Fractions 1 and 2) were resolved (FIG. 5D-2). The AGPs in Fraction 1 were described in Example 3 and Example 5[1].

Fraction 2 (FIG. 5D-2) was subjected to size-exclusion fractionation on superose-6 FPLC and was resolved into two components, peaks 2A and 2B (FIG. 5D-2,C3). N-terminal amino acid sequencing of material in Peak 2B gave the sequence AEAEAXTXALQVVAEAXEL (SEQ ID NO:74).

AGPs in Peaks 2A and 2B were separately deglycosylated and the resulting protein backbones were isolated by size-exclusion FPLC (FIG. 5D-2,D1-D4). Peak 2B gave one protein backbone with a molecular weight of 10 k. Peak 2A resulted in two protein peaks having molecular weights of 54 k and 10 k. N-terminal amino acid sequencing of the 54 k protein backbone gave the sequence TOAOA (SEQ ID NO:75), while the 10 k protein backbone in Peak 2B gave the sequence AEAEAOTOALQVVAEAOEL (SEQ ID NO:76).

The 10 k and 54 k protein backbones were digested separately with thermolysin and the resulting peptides were purified by RP-HPLC for sequencing. Sequences of eight peptides were obtained from the 54 k protein of Peak 2A and three from the 10 k protein in Peak 2B (Table 3.6). Two of the three sequences and the N-terminal sequence overlap to give a sequence AEAEAOTOALQVVAEAOELVOTOVOTOSY (SEQ ID NO:88) for the 10 k protein in peak 2B.

(b) Isolation of a cDNA encoding the 10 k AGP protein backbone

The approach to cloning of cDNA encoding the 10 k protein backbone was essentially the same as that used to clone the PcAGP9 cDNA. Two reverse and partially complementary long "guessmers" [AcF1 (SEQ ID NO:89) and AcR2 (SEQ ID NO:90), Table 5.3] were synthesized. In the "guessmers," nucleotide A was used at the third codon position for all amino acids, and CTA and TCA were assigned for Leu and Ser residues, respectively. The last 18 bp sequence at the 3' of the two "guessmers" were reverse-complementary, and they were annealed to each other in PCR to produce a double-stranded DNA fragment of 101 bp encoding the amino acid sequence A--E--A--E--A--O--T--O--A--L--Q--V--V--A--E--A--O--E--L--V--O--T--O--V--O- -T--O--S--Y (SEQ ID NO:88). The PCR fragment was subcloned into the pBluescriptII (Ks) vector. A .sup.32 P-labeled anti-sense RNA probe was synthesized using T3 RNA polymerase from the 101-bp DNA fragment and used to screen a pear cDNA library. Five cDNA clones were isolated and sequenced. The consensus sequence of 1040 bp is shown in FIGS. 5E-1 and 5E-2. This cDNA is referred to as PcAGP2.

EXAMPLE 6

Cloning and Expression of Genomic AGP Genes

(a) Cloning of Genomic AGP Genes and Identification of an AGP Promoter Region

The procedure essentially as used for the isolation of cDNA clones is used to obtain a genomic clone of a plant AGP. Whenever possible, AGP cDNA clones will be used to screen genomic libraries. The following procedure describing the isolation of a genomic AGP clone from suspension-cultured cells of N. alata and N. plumbaginafolia represents a general procedure which can be adapted for the isolation of a genomic AGP gene from a desired plant cell.

To isolate an AGP genomic clone, genomic DNA is isolated from suspension-cultured cells of N. alata and N. plumbaginafolia and partly digested with Sau3AI. After size selection by ultracentrifugation under a glycerol gradient, DNA fragments of 10-23 kb in size are ligated into vectors such as .lambda.Dash (Stratagene) to form a genomic library. The libraries are then screened with the NaAGP1 and NpAGP1 cDNAs, respectively, to isolate their corresponding genomic clones. The resulting genomic clones are studied by Southern analysis and some clones are sequenced. The promoter region of the AGP gene is then identified from the DNA sequence.

(b) Recombinant Gene Construction

The expression of a plant gene which exists in double-stranded DNA form involves transcription of messenger RNA (mRNA) from one strand of the DNA by RNA polymerase enzyme, and the subsequent processing of the mRNA primary transcript inside the nucleus. This processing involves a 3'-nontranslated region which adds polyadenylate nucleotides to the 3'-end of the RNA. Transcription of DNA into mRNA is regulated by a promoter. The promoter region contains a sequence of bases that signals RNA polymerase to associate with the DNA and to initiate the transcription of mRNA using one of the DNA strands as a template to make a corresponding strand of RNA.

A number of promoters which are active in plant cells have been described in the literature. These include the nopaline synthase (NOS) and octopine synthase (OCS) promoters (which are carried on tumor-inducing plasmids of Agrobacterium tumefaciens, the Cauliflower Mosaic Virum (CaMV) 19S and 35S promoters, the light-inducible promoter from the small subunit of ribulose bis-phosphate carboxylase (ssRUBISCO) and the mannopine synthase (MAS) promoter [Velten & Schell (1985) Nucl. Acids Res. 13:6981-6998]. All of these promoters have been used to create various types of DNA constructs which have been expressed in plants (see, e.g., PCT publication WO84/02913).

Promoters which are known or are found to cause transcription of RNA in plant cells can be used in the present invention. Such promoters may be obtained from plants or plant viruses and include, but are not limited to, the CAMV35S promoter and promoters isolated from plant genes such as ssRUBISCO genes. It is preferred that the particular promoter selected should be capable of causing sufficient expression to result in the production of an effective amount of protein.

The promoters used in the DNA constructs (i.e., chimeric plant genes) of the present invention may be modified, if desired, to affect their control characteristics. For example, the CaMV35S promoter may be ligated to the portion of the ssRUBISCO gene that represses the expression of ssRUBISCO in the absence of light, to create a promoter which is active in leaves but not in roots. The resulting chimeric promoter may be used as described herein. For purpose of this description, the phrase "CaMV35S" promoter thus includes variations of CaMV35S promoter, e.g., promoters derived by means of ligation with operator regions, random or controlled mutagenesis, etc. Furthermore, the promoters may be altered to contain multiple "enhancer sequences" to assist in elevating gene expression.

The RNA produced by a DNA construct of the present invention also contains a 5'-nontranslated leader sequence. This sequence can be derived from the promoter selected to express the gene, and can be specifically modified so as to increase translation of the mRNA. The 5'-nontranslated regions can also be obtained from viral RNAs, from suitable eukaryotic genes, or from a synthetic gene sequence. The present invention is not limited to constructs as presented in the following examples. Rather, the nontranslated leader sequence can be part of the 5'-end of the nontranslated region of the coding sequence for the virus coat protein, or part of the promoter sequence, or can be derived from an unrelated promoter or coding sequence in any case. It is preferred that the sequence flanking the initiation site conform to the translational consensus sequence rules for enhanced translation initiation reported by Kozak (1984) Nature 308:241-246.

The DNA construct of the present invention also contains a modified or fully-synthetic structural coding sequence which has been changed to enhance the performance of the gene in plants. For example, the enhancement method can be applied to design modified and fully synthetic genes encoding a plant AGP protein. The structural genes of the present invention may optionally encode a fusion protein comprising an amino-terminal chloroplast transit peptide or secretory signal sequence, etc.

The DNA construct also contains a 3'-nontranslated region. The 3'-nontranslated region contains a polyadenylation signal which functions in plants to cause the addition of polyadenylate nucleotides to the 3'-end of the viral RNA. Examples of suitable 3'- regions are (1) the 3'-transcribed, nontranslated regions containing the polyadenylation signal of Agrobacterium tumor-inducing (Ti) plasmid genes, such as the nopaline synthase (NOS) gene, and (2) plant genes like the soybean storage protein (7S) genes and the small subunit of the RuBP carboxylase (E9) gene. An example of a preferred 3'-region is that from the 7S gene.

(c) Plant Transformation

A chimeric plant gene containing a structural coding sequence of the present invention can be inserted into the genome of a plant by any suitable method. Suitable plants for use in the practice of the present invention include, but are not limited to, soybean, cotton, alfalfa, oilseed rape, flax, tomato, sugarbeet, sunflower, potato, tobacco, maize, rice and wheat. Suitable plant transformation vectors include those derived from a Ti plasmid of Agrobacterium tumefaciens, as well as those disclosed., e.g., by Herrera-Estrella et al. (1983) Nature 303:209, Bevan et al. (1983) Nature 304.:184, Klee et al. (1985) Bio/Technology 3:637-642, and EPO publication 120,516. In addition to plant transformation vectors derived from the Ti or root-inducing (Ri) plasmids of Agrobacterium, alternative methods can be used to insert the DNA constructs of this invention into plant cells. Such methods may involve, for example, the use of liposomes, electroporation, chemicals that increase free DNA uptake, free DNA delivery via microprojectile bombardment, and transformation using viruses or pollen.

A useful Ti plasmid cassette vector for transformation of dicotyledonous plants, for example, may consist of the enhanced CaMV35S promoter and the 3'-end including polyadenylation signals from a soybean gene encoding the alpha-prime subunit of beta-conglycinin. A multilinker containing multiple restriction sites for the insertion of genes may be positioned between these two elements.

(d) Over- and Under-production of AGPs by Transformed Cell Lines

It is generally acknowledged that all plant natural cell lines produce some AGPs, probably at the level of approximately 2-10% (w/w) of total structural complex carbohydrate [Showalter (1993) Plant Cell 5:9-23] These natural plant cells comprise all the regulatory factors (promoters, enhancers, enzymes, etc.) for transcription, translation and post-translational processing to produce a glycosylated AGP as the natural product. Glycosylation comprises the steps of (a) proline hydroxylation with a prolyl hydroxylase, (b) galactosylation using a unique .beta.-Hyp-galactosyl transferase, (c) the addition of galactose chains by a separate galactosyl transferase for each linkage type, and (d) the addition of arabinose by arabinosyl transferase. Thus, cultured natural plant cells (e.g., monocots or dicots) can be transformed with heterologous recombinant gene fragments and used for overproduction or underproduction of nonglycosylated AGPs. In some cases, a dicot host may be transformed with a monocot gene or, alternatively a monocot host may be transformed with a dicot gene. Alternatively, a host cell which normally does not produce glycosylated AGP (e.g., E. coli) may be transformed and used for the over- or under-production of a nonglycosylated AGP peptide backbone in which the proline residues have not been hydroxylated.

To transform a host cell for overproduction of AGP, an AGP cDNA (e.g., NaAGP1 or NpAGP1) is linked at the 5'-end with a heterologous promoter (e.g., CaMV 35S promoter) and at the 3'-end with a terminator (e.g., NOS-terminator). Thus, the AGP gene will be under the control of the CaMV 35S promoter, which is known to be a strong promoter. This expression cassette is then subcloned into a binary vector derived from the A. tumefaciens Ti plasmid to transform the cultured cells of either N. alata or N. plumbaginafolia to create cell lines that overproduce AGPs. The AGP is also tagged by histidines at the C-terminus by introducing a six-histidine coding DNA fragment into the AGP cDNAs. The six-histidine tagged AGP can then be readily isolated by using nickel-nitrolotriacetic acid Sepharose column (Hochuli et al., 1988, Bio/Technology 6:1321-1325). An alternative approach is to use the tag, Flag.TM., [Hopp, T. P. et al. (1988) Biotechnology 6:1204-1210], which can be incorporated into the AGP sequence to allow purification with an anti-Flag.TM. monoclonal antibody.

To transform a host cell for underproduction of AGP, an antisense construct is utilized. In this construct, the AGP cDNA is situated in the opposite direction of the CaMV 35S promoter so that an antisense transcript is produced. This transcript hybridizes to its corresponding sense mRNA eventually leading to the inhibition of gene expression.

__________________________________________________________________________ SEQUENCE LISTING (1) GENERAL INFORMATION: (iii) NUMBER OF SEQUENCES: 91 (2) INFORMATION FOR SEQ ID NO:1: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 11 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 2 (D) OTHER INFORMATION: /note="X=D/N" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4..8 (D) OTHER INFORMATION: /note="At positions 4, 6, 8 X=O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 9 (D) OTHER INFORMATION: /note="X=A/T" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 10 (D) OTHER INFORMATION: /note="X=O/S, where O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 11 (D) OTHER INFORMATION: /note="X=(O)=0 without absolute certainty, where O =Hydroxyproline" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:1: AlaXaaAlaXaaAlaXaaSerXaaXaaXaaXaa 1510 (2) INFORMATION FOR SEQ ID NO:2: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 11 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4..10 (D) OTHER INFORMATION: /note="At positions 4, 6, 8, 10 X=O=Hydroxyproline" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:2: AspGluAlaXaaAlaXaaAlaXaaSerXaaMet 1510 (2) INFORMATION FOR SEQ ID NO:3: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 10 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 1 (D) OTHER INFORMATION: /note="X=G/E" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 2..8 (D) OTHER INFORMATION: /note="At positions 2, 4, 6, 8 X=O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 9 (D) OTHER INFORMATION: /note="X=(Q)=Q without absolute certainty" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 10 (D) OTHER INFORMATION: /note="X=(V)=V without absolute certainty" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:3: XaaXaaAlaXaaAlaXaaAlaXaaXaaXaa 1510 (2) INFORMATION FOR SEQ ID NO:4: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 10 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4..8 (D) OTHER INFORMATION: /note="At positions 4, 6, 8 X=O=Hydroxyproline" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:4: AlaGluAlaXaaAlaXaaAlaXaaAlaSer 1510 (2) INFORMATION FOR SEQ ID NO:5: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 12 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 6..11 (D) OTHER INFORMATION: /note="At positions 6, 7, 9, 11 X=O=Hydroxyproline" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:5: LysAlaAlaAlaSerXaaXaaAlaXaaAlaXaaLys 1510 (2) INFORMATION FOR SEQ ID NO:6: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 10 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 2..8 (D) OTHER INFORMATION: /note="At positions 2, 4, 6, 8 X=O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 7 (D) OTHER INFORMATION: /note="X=V/H" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:6: AlaXaaAlaXaaAlaXaaXaaXaaGluAla 1510 (2) INFORMATION FOR SEQ ID NO:7: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 12 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 1 (D) OTHER INFORMATION: /note="X=S/L" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4..12 (D) OTHER INFORMATION: /note="At positions 4, 8, 11 X=O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 11 (D) OTHER INFORMATION: /note="X=(X)=without absolute certainty" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:7: XaaThrAlaXaaValAlaAlaXaaThrThrXaaXaa 1510 (2) INFORMATION FOR SEQ ID NO:8: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 6 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 2..5 (D) OTHER INFORMATION: /note="At positions 2 and 5 X=O=Hydroxyproline" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:8: SerXaaProAlaXaaAla 15 (2) INFORMATION FOR SEQ ID NO:9: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 6 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4 (D) OTHER INFORMATION: /note="X=(S)=S without absolute certainty" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 6 (D) OTHER INFORMATION: /note="X=(K)=K without absolute certainty" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:9: AlaAlaAlaXaaLeuXaa 15 (2) INFORMATION FOR SEQ ID NO:10: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 9 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 1 (D) OTHER INFORMATION: /note="X=(A)=A without absolute certainty" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4..8 (D) OTHER INFORMATION: /note="At positions 4, 6, 8 X=O=Hydroxyproline" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:10: XaaAspAlaXaaAlaXaaSerXaaVal 15 (2) INFORMATION FOR SEQ ID NO:11: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 15 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:11: AlaLysSerLysPheMetIleIleProAlaSerXaaThrXaaAla 151015 (2) INFORMATION FOR SEQ ID NO:12: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 39 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 7..9 (D) OTHER INFORMATION: /note="Positions 7-9 can also be AGY" (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 31..33 (D) OTHER INFORMATION: /note="Positions 31-33 can also be AGY" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:12: GCNAARTCNAARTTYATGATHATHCCNGCNTCNACNGCN39 (2) INFORMATION FOR SEQ ID NO:13: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (xi) SEQUENCE DESCRIPTION: SEQ ID NO:13: GGDATDATCATRAACTT17 (2) INFORMATION FOR SEQ ID NO:14: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (xi) SEQUENCE DESCRIPTION: SEQ ID NO:14: GGDATDATCATRAATTT17 (2) INFORMATION FOR SEQ ID NO:15: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 18 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear

(ii) MOLECULE TYPE: cDNA (xi) SEQUENCE DESCRIPTION: SEQ ID NO:15: GGDAATGATCATAAACTT18 (2) INFORMATION FOR SEQ ID NO:16: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 18 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (xi) SEQUENCE DESCRIPTION: SEQ ID NO:16: GGDAATAATCATAAACTT18 (2) INFORMATION FOR SEQ ID NO:17: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 18 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (xi) SEQUENCE DESCRIPTION: SEQ ID NO:17: GGDAATTATCATAAACTT18 (2) INFORMATION FOR SEQ ID NO:18: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 18 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (xi) SEQUENCE DESCRIPTION: SEQ ID NO:18: GGDAATGATCATGAACTT18 (2) INFORMATION FOR SEQ ID NO:19: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 18 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (xi) SEQUENCE DESCRIPTION: SEQ ID NO:19: GGDAATAATCATGAACTT18 (2) INFORMATION FOR SEQ ID NO:20: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 18 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (xi) SEQUENCE DESCRIPTION: SEQ ID NO:20: GGDAATTATCATGAACTT18 (2) INFORMATION FOR SEQ ID NO:21: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 161 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (ix) FEATURE: (A) NAME/KEY: CDS (B) LOCATION: 3..161 (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 56..78 (D) OTHER INFORMATION: /note="NaF1 primer" (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 102..123 (D) OTHER INFORMATION: /note="NaF2 primer" (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 138..161 (D) OTHER INFORMATION: /note="Amino acids of the peptide obtained by direct microsequencing which are identical with the derived sequence" (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 148..160 (D) OTHER INFORMATION: /note="Nucleotide sequence corresponding to the NaR1 primers" (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 161 (D) OTHER INFORMATION: /note="Nucleotide 161 was not included in the 160 base pair primer extension fragment" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:21: AGAGAACCAAGAAACCAACACATCAAATATTCTTCTTTCCCTTTTGT47 ArgThrLysLysProThrHisGlnIlePhePhePheProPheCys 151015 TCTATTTTCATTATGGGTCATTTCACTAAGCAAATGACATTCTTCTTG95 SerIlePheIleMetGlyHisPheThrLysGlnMetThrPhePheLeu 202530 TTCTTGGTGATCTCAACTCCATTGGTGCAAATTGAAGGTAGAAAAAGC143 PheLeuValIleSerThrProLeuValGlnIleGluGlyArgLysSer 354045 AAGTTTATGATCATACCC161 LysPheMetIleIlePro 50 (2) INFORMATION FOR SEQ ID NO:22: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 23 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (xi) SEQUENCE DESCRIPTION: SEQ ID NO:22: CATTATGGGTCATTTCACTAAGC23 (2) INFORMATION FOR SEQ ID NO:23: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 23 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (xi) SEQUENCE DESCRIPTION: SEQ ID NO:23: GGTGATCTCAACTCCATTGGTGC23 (2) INFORMATION FOR SEQ ID NO:24: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 1690 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (ix) FEATURE: (A) NAME/KEY: CDS (B) LOCATION: 60..1442 (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 1..38 (D) OTHER INFORMATION: /note="Nucleotide sequence obtained by PCR which does not overlap with the cDNA clone" (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 60..128 (D) OTHER INFORMATION: /note="Predicted transmembrane segment" (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 135..179 (D) OTHER INFORMATION: /note="Derived amino acid sequence corresponding to the peptide sequence by protein microsequencing" (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 135..179 (D) OTHER INFORMATION: /note="Amino acids 27 to 36, 38, and 40 are identical to that in the peptide obtained by direct microsequencing" (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 135..179 (D) OTHER INFORMATION: /note="Amino acid 26 may also be Ala; 37 and 39 can also be undetermined residues" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:24: AGAGAACCAAGAAACCAACACATCAAATATTCTTCTTTCCCTTTTGTTCTATTTTCATT59 ATGGGTCATTTCACTAAGCAAATGACATTCTTCTTGTTCTTGGTGATC107 MetGlyHisPheThrLysGlnMetThrPhePheLeuPheLeuValIle 151015 TCAACTCCATTGGTGCAAATTGAAGGTAGAAAAAGCAAGTTTATGATC155 SerThrProLeuValGlnIleGluGlyArgLysSerLysPheMetIle 202530 ATACCTGCATCTCCTACACCAGCTCCAACACCAATCAATGAAATTAGT203 IleProAlaSerProThrProAlaProThrProIleAsnGluIleSer 354045 TTTCCTCCATTTTCATCCCTTACTCCAACTCCATCACCAACACCAGCA251 PheProProPheSerSerLeuThrProThrProSerProThrProAla 505560 CCAGCAACAGCACCAACACCGTTTTTTAATGATTTTGCGTTTCCTCCA299 ProAlaThrAlaProThrProPhePheAsnAspPheAlaPheProPro 65707580 TTGTCATCTTTAAGTCCAACACCAGCACCAGTACCAGTAGGTAATGTT347 LeuSerSerLeuSerProThrProAlaProValProValGlyAsnVal 859095 CAAGATCCTGATGTGAATGGCGTACCTACGCCTGCATTGGCACCAGGA395 GlnAspProAspValAsnGlyValProThrProAlaLeuAlaProGly 100105110 GGGAGTGGTGAAGATCCAGAGGAAGGTGGCATTGAAGCGCCAGCACCA443 GlySerGlyGluAspProGluGluGlyGlyIleGluAlaProAlaPro 115120125 CTTTTGACTGATACTCCCTATGGACTTTATGGTCCTCATTCTCAGGAA491 LeuLeuThrAspThrProTyrGlyLeuTyrGlyProHisSerGlnGlu 130135140 ATTTCTTCTACTGTCACAAATCTTGATGAGGTTGAAACTCAAACTCCT539 IleSerSerThrValThrAsnLeuAspGluValGluThrGlnThrPro 145150155160 GCCAAGGAATTTCAAGGTGCTAGATTTAATACAGATGAGTCCTACAAT587 AlaLysGluPheGlnGlyAlaArgPheAsnThrAspGluSerTyrAsn 165170175 AACAATGGTTATGATTCCAACAACAACGACAACAACAATGGTTATGAT635 AsnAsnGlyTyrAspSerAsnAsnAsnAspAsnAsnAsnGlyTyrAsp 180185190 TCCAACAATAACAACAACAATAACGATGATGGCTTCTCCGAGAATTAC683 SerAsnAsnAsnAsnAsnAsnAsnAspAspGlyPheSerGluAsnTyr 195200205 AACAACAATGGCTACTCGGAGAATGCTAATAACAAAAATAACAATGGC731 AsnAsnAsnGlyTyrSerGluAsnAlaAsnAsnLysAsnAsnAsnGly 210215220 TACTCAGAGAATTACAACAACAATAACAACAATGGCTACGCCAAGAAT779 TyrSerGluAsnTyrAsnAsnAsnAsnAsnAsnGlyTyrAlaLysAsn 225230235240 TACAACAATGGCTACTCTCAGAGTTACAACAACAACAATAATTTTTAC827 TyrAsnAsnGlyTyrSerGlnSerTyrAsnAsnAsnAsnAsnPheTyr 245250255 TCGGAGAATTACAACAACAACAACAACAATGTTTTCTCGGAGAATTCC875 SerGluAsnTyrAsnAsnAsnAsnAsnAsnValPheSerGluAsnSer 260265270 AACAACAATGGCTACTCCAAAAAGATCAACAATAATGGCTACTCCCAG923 AsnAsnAsnGlyTyrSerLysLysIleAsnAsnAsnGlyTyrSerGln 275280285 AATTACATGAACAACAACAATGGCTTCTCCGAGAGTTACAACAACAAC971 AsnTyrMetAsnAsnAsnAsnGlyPheSerGluSerTyrAsnAsnAsn 290295300 AACAACAACAACAACAACAACAACGTTTTCTCTGAGAATTACAACAAC1019 AsnAsnAsnAsnAsnAsnAsnAsnValPheSerGluAsnTyrAsnAsn 305310315320 AACAATAACAATAATGTTTTCTCCGAGAATTACAACAACAACAATAAC1067 AsnAsnAsnAsnAsnValPheSerGluAsnTyrAsnAsnAsnAsnAsn 325330335 AACAATGCTTTCTACGAGAATTACAACAACAACAACAATGGCTACTCA1115 AsnAsnAlaPheTyrGluAsnTyrAsnAsnAsnAsnAsnGlyTyrSer 340345350 GAGAACTACAATCAGGCTAGCAGCTACAATAACAATGACAATACGGTG1163 GluAsnTyrAsnGlnAlaSerSerTyrAsnAsnAsnAspAsnThrVal 355360365 GAAAGGCAAGGATTAAGTGATACAAGATTCTTGGAAAATGGCAAGTAT1211 GluArgGlnGlyLeuSerAspThrArgPheLeuGluAsnGlyLysTyr 370375380 TATTATGATATCAAGAATGAGAATACCAACAACAATGGCTACTCTGAG1259 TyrTyrAspIleLysAsnGluAsnThrAsnAsnAsnGlyTyrSerGlu 385390395400 AATTACAACCATGTTAGCAGCTACAATAACAATAACAATATGGTGGAA1307 AsnTyrAsnHisValSerSerTyrAsnAsnAsnAsnAsnMetValGlu 405410415 AGGCAAGGATTGAGTGACACAAGATTCTTAGATAATGGTAACTACTTT1355 ArgGlnGlyLeuSerAspThrArgPheLeuAspAsnGlyAsnTyrPhe 420425430 TATAGCAACAATGGTGAGAAAATGTCAGTGGAAGAGTCTGAAAGACAG1403 TyrSerAsnAsnGlyGluLysMetSerValGluGluSerGluArgGln 435440445 CAGGAATATCCAGACACTGAAGATCAGTACGAACTTCCTTGAAGATAAA1452 GlnGluTyrProAspThrGluAspGlnTyrGluLeuPro 450455460 TATTATTAGTTGGTCCAGAGAAGAGGGACAAACGCAGAGGACGTGAAAATAGATTTAATG1512 ATTGAATTTTAAGTTATTTTGAGTGTTTGTTTCATTAGTTCCACTTGAGTCTGCAAACAC1572 CTTTTTTTCTTTTTTTATAGTTCTGCAAATCAGACCGAGGGAACTTTGAGTTGTTTAACA1632 CTTTTGGATTATTTTAAAAACTTTTTATAATGATCTTGAAGCTTCACGCCTAAAAAAA1690 (2) INFORMATION FOR SEQ ID NO:25: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 1430 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear

(ii) MOLECULE TYPE: cDNA (ix) FEATURE: (A) NAME/KEY: CDS (B) LOCATION: 2..1312 (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 41..112 (D) OTHER INFORMATION: /note="Derived amino acid sequence corresponding to the peptide sequence by protein microsequencing" (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 41..112 (D) OTHER INFORMATION: /note="Derived amino acids 14-24, 28, 30, 32- 37 are identical to sequences obtained by protein microsequencing" (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 25..31 (D) OTHER INFORMATION: /note="Amino acids 25, 27, 29, and 31 are hydroxylated proline residues; amino acid 26 can be T instead of A" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:25: GTTCTTGGTGATCTCAACTCCACTAGTACAAATTGAAGCAAGAAAA46 PheLeuValIleSerThrProLeuValGlnIleGluAlaArgLys 151015 AGCAAGTTTATGATCATTCCTGCATCTCCTGCACCAGCTCCAACTCCA94 SerLysPheMetIleIleProAlaSerProAlaProAlaProThrPro 202530 ATCAATGAAATTAGTTTTCCTCCATTTTCATCCTTTACTCCAACTCCA142 IleAsnGluIleSerPheProProPheSerSerPheThrProThrPro 354045 TCACCAACACCAACACCAACACCAACATCAGCACCAACACCGTTTTTT190 SerProThrProThrProThrProThrSerAlaProThrProPhePhe 505560 AATGATTTCGCGTTTCCTCCATTGTCATCTTTAAGTCCAACACCAGCA238 AsnAspPheAlaPheProProLeuSerSerLeuSerProThrProAla 657075 CCAGTAGGTAGTGATCAAGATCCTGATGTGAACGGTGTACCGGCGCCT286 ProValGlySerAspGlnAspProAspValAsnGlyValProAlaPro 80859095 GCAGTGGCACCAATAGGGAGTGGTCAAGATCCAGAAGAAGGTGGCATT334 AlaValAlaProIleGlySerGlyGlnAspProGluGluGlyGlyIle 100105110 GAAGCACCAGCACCACTTTTAACTGATACTCCTTATGGACTTTATGGT382 GluAlaProAlaProLeuLeuThrAspThrProTyrGlyLeuTyrGly 115120125 CCTCATTCTCAGGAAATTCCTTCAACTGTCACAAATCTTGATGAGGTT430 ProHisSerGlnGluIleProSerThrValThrAsnLeuAspGluVal 130135140 GAAACTCAAACTCCTGCCGAGGAATTCCAAGGTGCTAGATTTAATACA478 GluThrGlnThrProAlaGluGluPheGlnGlyAlaArgPheAsnThr 145150155 GATGAGTCCTACAATAACAATGGTTATGATTCCAACAACAATGGCTAC526 AspGluSerTyrAsnAsnAsnGlyTyrAspSerAsnAsnAsnGlyTyr 160165170175 TCGGAGAATAACAACAACAAGAACAACAATGGCTACTCGGAGAATTAC574 SerGluAsnAsnAsnAsnLysAsnAsnAsnGlyTyrSerGluAsnTyr 180185190 AACAACAACAACAACAATGGCTACTCCGAGAATTACAACAACAACAAC622 AsnAsnAsnAsnAsnAsnGlyTyrSerGluAsnTyrAsnAsnAsnAsn 195200205 AATGGCTACTCCAAGAATTACAACAACAATGGCTACTCCAAAAAAATC670 AsnGlyTyrSerLysAsnTyrAsnAsnAsnGlyTyrSerLysLysIle 210215220 AACAATAATGGTTACTCCCAGAATTACATGAACAACAACAACGGCTTC718 AsnAsnAsnGlyTyrSerGlnAsnTyrMetAsnAsnAsnAsnGlyPhe 225230235 TCCGAGAGTTACAACAGCAACAACAACAACAACAATATTTTCTCCGAG766 SerGluSerTyrAsnSerAsnAsnAsnAsnAsnAsnIlePheSerGlu 240245250255 AATTACAACAACAATAATAACAATAATGTTTTCTCCGAGAATTACAAC814 AsnTyrAsnAsnAsnAsnAsnAsnAsnValPheSerGluAsnTyrAsn 260265270 AACAATAATAACAATAATGTTTTCTCCGAGAATTACAACAACAATAAC862 AsnAsnAsnAsnAsnAsnValPheSerGluAsnTyrAsnAsnAsnAsn 275280285 AACAATGCTTTCTCCGAGAACTACAACAACAATAATGTTTTCTCCGAG910 AsnAsnAlaPheSerGluAsnTyrAsnAsnAsnAsnValPheSerGlu 290295300 AATTACAACAAAAACAATAACAACAATGCTTTCTCTGAGAATTACAAC958 AsnTyrAsnLysAsnAsnAsnAsnAsnAlaPheSerGluAsnTyrAsn 305310315 AACAAAAACAACAATGCCTACTCTGAGAACTACAATCAAGCTAGCAGC1006 AsnLysAsnAsnAsnAlaTyrSerGluAsnTyrAsnGlnAlaSerSer 320325330335 TACAATAACAATGGCAATACGGTGGAGAGGCAAGGATTAAGTGATACA1054 TyrAsnAsnAsnGlyAsnThrValGluArgGlnGlyLeuSerAspThr 340345350 AGATTCTTGGAGAATGGCAAGTACTATTATGATATCAAGAATGAGAAT1102 ArgPheLeuGluAsnGlyLysTyrTyrTyrAspIleLysAsnGluAsn 355360365 CCCAACCACAACAATGGCTACTCCGAGAACTACAATCATGTTAGCAGC1150 ProAsnHisAsnAsnGlyTyrSerGluAsnTyrAsnHisValSerSer 370375380 TACAATAACAATAACAATATGGTGGAAAGGCAAGGATTGAGTGACACA1198 TyrAsnAsnAsnAsnAsnMetValGluArgGlnGlyLeuSerAspThr 385390395 AGATTCTTAGATAATGGCAACTACTTTTATAGTAACAATGGTGAGAAA1246 ArgPheLeuAspAsnGlyAsnTyrPheTyrSerAsnAsnGlyGluLys 400405410415 ATGTCAATGGAAGAATCTGAAAGACAGCAGGAATATCCAAATACTGAA1294 MetSerMetGluGluSerGluArgGlnGlnGluTyrProAsnThrGlu 420425430 GATCAGTATGAACTTCCTTGAAGACTAACATTATCAGTTGGCTTAAAG1342 AspGlnTyrGluLeuPro 435 AAGAGGGACAAATGCAGGGAACATGAGAATAGATTTAATTTACAGAGTTTGATTGAATTT1402 TTAAGTTAAAAAAAAAAAAAAAAAAAAA1430 (2) INFORMATION FOR SEQ ID NO:26: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 10 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4..8 (D) OTHER INFORMATION: /note="At positions 4, 5, 7, 8 X=O=Hydroxyproline" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:26: LeuAlaSerXaaXaaAlaXaaXaaThrAla 1510 (2) INFORMATION FOR SEQ ID NO:27: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 14 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4..13 (D) OTHER INFORMATION: /note="At positions 4, 5, 7, 8, 13 X=O=Hydroxyproline" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:27: LeuAlaSerXaaXaaAlaXaaXaaThrAlaAspThrXaaAla 1510 (2) INFORMATION FOR SEQ ID NO:28: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 12 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 3..11 (D) OTHER INFORMATION: /note="At positions 3 and 11 X=O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4 (D) OTHER INFORMATION: /note="X=S/N" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:28: PheAlaXaaXaaGlyGlyValAlaLeuProXaaSer 1510 (2) INFORMATION FOR SEQ ID NO:29: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 14 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 5..9 (D) OTHER INFORMATION: /note="At positions 5 and 9 X=O=Hydroxyproline" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:29: IleGlyAlaAlaXaaAlaGlySerXaaThrSerSerProAsn 1510 (2) INFORMATION FOR SEQ ID NO:30: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 24 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 12..18 (D) OTHER INFORMATION: /note="At positions 12, 14, 16, 18 X=O=Hydroxyproline" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:30: ArgLysSerLysPheMetIleIleProAlaSerXaaThrXaaAlaXaa 151015 ThrXaaIleAsnGluIleSerPhe 20 (2) INFORMATION FOR SEQ ID NO:31: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 19 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 3..17 (D) OTHER INFORMATION: /note="At positions 3, 7, 10, 12, 16, 17 X=O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 14 (D) OTHER INFORMATION: /note="X=(S)=S without absolute certainty" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 19 (D) OTHER INFORMATION: /note="X=(T)=T without absolute certainty" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:31: LeuSerXaaLysLysSerXaaThrAlaXaaSerXaaSerXaaThrXaa 151015 XaaThrXaa (2) INFORMATION FOR SEQ ID NO:32: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 20 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 1 (D) OTHER INFORMATION: /note="X=V/A" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 2 (D) OTHER INFORMATION: /note="X=A/T" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4 (D) OTHER INFORMATION: /note="X=O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 5 (D) OTHER INFORMATION: /note="X=S/O, where O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 6 (D) OTHER INFORMATION: /note="X=O/Y, where O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 9 (D) OTHER INFORMATION: /note="X=T/A" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 11 (D) OTHER INFORMATION: /note="X=O=Hydroxyproline" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:32: XaaXaaAlaXaaXaaXaaSerSerXaaXaaXaaSerAlaThrXaaThr 151015 XaaXaaValAla 20

(2) INFORMATION FOR SEQ ID NO:33: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 10 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 1 (D) OTHER INFORMATION: /note="X=V/A" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 3 (D) OTHER INFORMATION: /note="X=D/A/O, where O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4 (D) OTHER INFORMATION: /note="X=S/O, where O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 5 (D) OTHER INFORMATION: /note="X=T/O/K, where O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 6 (D) OTHER INFORMATION: /note="X=O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 7 (D) OTHER INFORMATION: /note="X=S/O" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:33: XaaAlaXaaXaaXaaXaaXaaProGlnSer 1510 (2) INFORMATION FOR SEQ ID NO:34: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 15 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 5..9 (D) OTHER INFORMATION: /note="At positions 5, 7, 9 X=O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 15 (D) OTHER INFORMATION: /note="X=(G)=G without absolute certainty" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:34: LeuGlyIleSerXaaAlaXaaSerXaaAlaGlyGluValAspXaa 151015 (2) INFORMATION FOR SEQ ID NO:35: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 11 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 3..7 (D) OTHER INFORMATION: /note="At positions 3, 4, 7 X=O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 11 (D) OTHER INFORMATION: /note="X=O/S, where O=Hydroxyproline" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:35: XaaXaaXaaXaaAlaAlaXaaValXaaAlaXaa 1510 (2) INFORMATION FOR SEQ ID NO:36: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 20 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4..11 (D) OTHER INFORMATION: /note="At positions 4, 6, 10, 11 X=O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 15..19 (D) OTHER INFORMATION: /note="At positions 15 and 19 X=(T)=T without absolute certainty" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:36: ValThrAlaXaaThrXaaSerAlaSerXaaXaaSerSerThrXaaAla 151015 AlaThrXaaAla 20 (2) INFORMATION FOR SEQ ID NO:37: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 17 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4..14 (D) OTHER INFORMATION: /note="At positions 4, 8, 9, 13 and 14 X=O=Hydroxyproline" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:37: AlaLysSerXaaThrAlaThrXaaXaaThrAlaThrXaaXaaSerAla 151015 Val (2) INFORMATION FOR SEQ ID NO:38: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 20 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4..15 (D) OTHER INFORMATION: /note="At positions 4, 6, 10, 11, 15 X=O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 17 (D) OTHER INFORMATION: /note="X=(S)=S without absolute certainty" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:38: ValThrAlaXaaThrXaaSerAlaSerXaaXaaSerSerThrXaaAla 151015 XaaThrXaaAla 20 (2) INFORMATION FOR SEQ ID NO:39: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 17 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 3..16 (D) OTHER INFORMATION: /note="At positions 3, 7, 8, 12, 15 X=O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 6 (D) OTHER INFORMATION: /note="X=T/Q" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:39: IleSerXaaAlaSerXaaXaaXaaThrThrSerXaaAlaSerXaaXaa 151015 Thr (2) INFORMATION FOR SEQ ID NO:40: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 15 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 3 (D) OTHER INFORMATION: /note="X=P/S" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4..13 (D) OTHER INFORMATION: /note="At positions 4, 8, 11, 12, 13 X=O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 15 (D) OTHER INFORMATION: /note="X=(T)=T without absolute certainty" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:40: ValSerXaaXaaValGlnSerXaaAlaSerXaaXaaXaaThrXaa 151015 (2) INFORMATION FOR SEQ ID NO:41: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 11 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:41: LeuValValValValMetThrProArgLysHis 1510 (2) INFORMATION FOR SEQ ID NO:42: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 10 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 3..7 (D) OTHER INFORMATION: /note="At positions 3, 6, 7 X=O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 9 (D) OTHER INFORMATION: /note="X=T/K" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:42: XaaAsnXaaAlaThrXaaXaaAlaXaaPro 1510 (2) INFORMATION FOR SEQ ID NO:43: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 7 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 5 (D) OTHER INFORMATION: /note="X=O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 7 (D) OTHER INFORMATION: /note="X=(L)=L without absolute certainty" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:43: IleAlaAlaThrXaaSerXaa 15 (2) INFORMATION FOR SEQ ID NO:44: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 10 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 1 (D) OTHER INFORMATION: /note="X=(G)/(S)=G/S without absolute certainty" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4..8 (D) OTHER INFORMATION: /note="At positions 4, 6, 8 X=O=Hydroxyproline" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:44: XaaAsnAlaXaaAlaXaaXaaXaaLysPro 1510 (2) INFORMATION FOR SEQ ID NO:45: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (xi) SEQUENCE DESCRIPTION: SEQ ID NO:45: GTNGTNGTNGTNATGAC17 (2) INFORMATION FOR SEQ ID NO:46: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 20 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (xi) SEQUENCE DESCRIPTION: SEQ ID NO:46: GTAGTNATGACNCCNRGAAA20 (2) INFORMATION FOR SEQ ID NO:47: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 20 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (xi) SEQUENCE DESCRIPTION: SEQ ID NO:47: GTAGTNATGACNCCNCGNAA20 (2) INFORMATION FOR SEQ ID NO:48: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 350 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (ix) FEATURE: (A) NAME/KEY: CDS (B) LOCATION: 1..156 (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 1..24 (D) OTHER INFORMATION: /note="Derived amino acid sequence matching the peptide sequence by protein sequencing" (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 1..20 (D) OTHER INFORMATION: /note="Nucleotide sequence corresponding to the PcA23F2a primers" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:48: GTAGTGATGACGCCGAGGAAGCACCTCGGTATTTCTCCCGCTCCTTCT48 ValValMetThrProArgLysHisLeuGlyIleSerProAlaProSer 151015 CCGGCAGGGGAAGTCGACGGTCCTGCTATTGCTCCGACAAGCGGCGCT96 ProAlaGlyGluValAspGlyProAlaIleAlaProThrSerGlyAla 202530 ACAAGCTTGAAGGGTGGTGTTCTGACTGTGGTGGCATTGGGAGGGTTT144 ThrSerLeuLysGlyGlyValLeuThrValValAlaLeuGlyGlyPhe 354045 TGTCTGTGGTTTTAGCGAGGGGGAGATTTTTTGAACCGTGGTTGTTATCTTT196 CysLeuTrpPhe 50 CTGGGTTTTTGTTTTGAGAGTGGGGGATAATTATTTGTTTAATTCTTTATTTTTTTTATA256 CATATGAGACGAGATATTATGTAATTCTATTTCGAATGTCATAATATCAATATATTCATT316 TCCTAAATATAAAAAAAAAAAAAAAAAAAAAAAA350 (2) INFORMATION FOR SEQ ID NO:49: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 760 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (ix) FEATURE: (A) NAME/KEY: CDS (B) LOCATION: 20..559 (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 20..100 (D) OTHER INFORMATION: /note="The predicted secretion signal" (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 126..132 (D) OTHER INFORMATION: /note="Potential N-glycosylation site" (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 279..285 (D) OTHER INFORMATION: /note="Potential N-glycosylation site" (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 395..471 (D) OTHER INFORMATION: /note="Sequence matches the peptide sequences obtained from the AGP protein backbone" (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 441..453 (D) OTHER INFORMATION: /note="Proline residues are hydroxylated" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:49: CGCTCTCTAAAATTTTCAAATGGCTGGCTTTTCATCCTCAAAAGCTCTGTCA52 MetAlaGlyPheSerSerSerLysAlaLeuSer 1510 TACTCCTCTCTTCTCGTCGTCTTCCTCCTCTTCGGCTTCTCGGAAGCC100 TyrSerSerLeuLeuValValPheLeuLeuPheGlyPheSerGluAla 152025 AGAGAGATCACCGTCGGTGGCAAGAATGGCTCATGGGCAGTCCCCTCC148 ArgGluIleThrValGlyGlyLysAsnGlySerTrpAlaValProSer 303540 TCCGAATCGCAATCCCTCAACAAATGGGCCGAAAGCACCCGCTTTCGC196 SerGluSerGlnSerLeuAsnLysTrpAlaGluSerThrArgPheArg 455055 GTCGGCGACACTCTTGTGTGGAAGTACGACAGCGCCAAAGACTCAGTC244

ValGlyAspThrLeuValTrpLysTyrAspSerAlaLysAspSerVal 60657075 TTGCGAGTGACGAAAGAAGACTACTCAAACTGCAATGCGTCAAACCCA292 LeuArgValThrLysGluAspTyrSerAsnCysAsnAlaSerAsnPro 808590 ATTGAGCAGCTCAAGGACGGCGAAACAAAGCTCCACCTTGACCAGCCA340 IleGluGlnLeuLysAspGlyGluThrLysLeuHisLeuAspGlnPro 95100105 GGGCCTTACTACTTCATCAGCGGAACCAAGGGGCACTGCGAGAAGGGG388 GlyProTyrTyrPheIleSerGlyThrLysGlyHisCysGluLysGly 110115120 CAGAAACTGGTGGTGGTGGTTATGACTCCAAGGAAGCACCTCGGTATT436 GlnLysLeuValValValValMetThrProArgLysHisLeuGlyIle 125130135 TCTCCCGCTCCTTCTCCGGCAGGGGAAGTCGACGGTCCTGCTATTGCT484 SerProAlaProSerProAlaGlyGluValAspGlyProAlaIleAla 140145150155 CCGACAAGCGGCGCTACAAGCTTGAAGGGTGGTGTTCTGACTGTGGTG532 ProThrSerGlyAlaThrSerLeuLysGlyGlyValLeuThrValVal 160165170 GCATTGGGAGGGTTTTGTCTGTGGTTTTAGCGAGGGGGAGATTTTTT579 AlaLeuGlyGlyPheCysLeuTrpPhe 175180 GAACCGTGGTTGTTATCTTTCTGGGTTTTTGTTTTGAGAGTGGGGGATAATTATTTGTTT639 AATTCTTTATTTTTTTTATACATATGAGACGAGATATTATGTAATTCTATTTCGAATGTC699 ATAATATCAATATATTCATTTCCTAAATATAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA759 A760 (2) INFORMATION FOR SEQ ID NO:50: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 12 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 3..11 (D) OTHER INFORMATION: /note="At positions 3 and 11 X=O=Hydroxyproline" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:50: PheAlaXaaSerGlyGlyValAlaLeuProXaaSer 1510 (2) INFORMATION FOR SEQ ID NO:51: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 14 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4..13 (D) OTHER INFORMATION: /note="Positions 4-5, 7-8, 13 X=O=Hydroxyproline" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:51: LeuAlaSerXaaXaaAlaXaaXaaThrAlaAspThrXaaAla 1510 (2) INFORMATION FOR SEQ ID NO:52: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 14 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 1 (D) OTHER INFORMATION: /note="X=I/V" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 2 (D) OTHER INFORMATION: /note="X=G/S" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 3 (D) OTHER INFORMATION: /note="X=A/S" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4 (D) OTHER INFORMATION: /note="X=A/O, where O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 5 (D) OTHER INFORMATION: /note="X=O/S, where O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 6 (D) OTHER INFORMATION: /note="X=A/Q" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 7 (D) OTHER INFORMATION: /note="X=G/S" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 8 (D) OTHER INFORMATION: /note="X=S/O, where O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 9 (D) OTHER INFORMATION: /note="X=O/S, where O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 10 (D) OTHER INFORMATION: /note="X=T/A" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 11 (D) OTHER INFORMATION: /note="X=S/A" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 12 (D) OTHER INFORMATION: /note="X=S/A" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 14 (D) OTHER INFORMATION: /note="X=O=Hydroxyproline" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:52: XaaXaaXaaXaaXaaXaaXaaXaaXaaXaaXaaXaaProXaa 1510 (2) INFORMATION FOR SEQ ID NO:53: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 14 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 5..9 (D) OTHER INFORMATION: /note="At positions 5 and 9 X=O=Hydroxyproline" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:53: IleGlySerAlaXaaAlaGlySerXaaThrSerSerProAsn 1510 (2) INFORMATION FOR SEQ ID NO:54: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 12 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 7 (D) OTHER INFORMATION: /note="X=O=Hydroxyproline" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 10 (D) OTHER INFORMATION: /note="X=(D)=D without absolute certainty" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:54: XaaXaaXaaGlnSerAlaXaaAlaAlaXaaXaaAsn 1510 (2) INFORMATION FOR SEQ ID NO:55: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 18 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4 (D) OTHER INFORMATION: /note="X=S/A" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 5 (D) OTHER INFORMATION: /note="X=Y/L" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 6 (D) OTHER INFORMATION: /note="X=D/I" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 8 (D) OTHER INFORMATION: /note="X=K/E" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 9 (D) OTHER INFORMATION: /note="X=T/A" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 18 (D) OTHER INFORMATION: /note="X=(E)=E without absolute certainty" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:55: XaaThrPheXaaXaaXaaIleXaaXaaAlaIleAsnThrGluPheGly 151015 ProXaa (2) INFORMATION FOR SEQ ID NO:56: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 22 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4 (D) OTHER INFORMATION: /note="X=S/A" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 5 (D) OTHER INFORMATION: /note="X=Y/L/V" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 6 (D) OTHER INFORMATION: /note="X=D/I/A" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:56: XaaThrPheXaaXaaXaaIleGluThrAlaIleAsnThrGluPheGly 151015 ProXaaGluXaaXaaGln 20 (2) INFORMATION FOR SEQ ID NO:57: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 18 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 6 (D) OTHER INFORMATION: /note="X=D/I" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 7 (D) OTHER INFORMATION: /note="X=K/E" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 15 (D) OTHER INFORMATION: /note="X=G/M" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:57: XaaThrPheSerTyrXaaXaaThrAlaIleAsnThrGluPheXaaPro 151015 AlaGlu (2) INFORMATION FOR SEQ ID NO:58: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 9 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:58: ThrAlaIleAsnThrGluPheGlyPro 15 (2) INFORMATION FOR SEQ ID NO:59: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 16 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:59: AlaValPheLysAsnLysXaaXaaLeuThrXaaXaaProXaaIleIle 151015

(2) INFORMATION FOR SEQ ID NO:60: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 7 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:60: AlaIleAsnThrGluPheGly 15 (2) INFORMATION FOR SEQ ID NO:61: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 20 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 3 (D) OTHER INFORMATION: /note="N=I" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:61: GCNATHAAYACNCARTTYGG20 (2) INFORMATION FOR SEQ ID NO:62: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 350 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (ix) FEATURE: (A) NAME/KEY: CDS (B) LOCATION: 1..102 (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 1..19 (D) OTHER INFORMATION: /note="RT35-specific primer sequence" (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 1..24 (D) OTHER INFORMATION: /note="Derived amino acid sequence corresponding to the peptide sequence by protein microsequencing" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:62: GCGATCAACACGGAGTTCGGTCCTGAGGAATGTAACCAGTATGAATTT48 AlaIleAsnThrGluPheGlyProGluGluCysAsnGlnTyrGluPhe 151015 GCCATGATCAAAAATCAGTGTGCAAATCAAGCTGCTCCTCCTCCTACA96 AlaMetIleLysAsnGlnCysAlaAsnGlnAlaAlaProProProThr 202530 GATTACTAAGTTATTAAGGGGATGTATGTGTACGTGCGTGTTTATATAGCCGACCC152 AspTyr CAACTTGTTTGGACTTAGGCGTAGTTTGCATTATTGTTATTTTATATATGTGTATGTATT212 CATATAGTCGACCCCAACTTGTTTGGGATTGAGACGTAATTGGTGTTTATTAGTATGTAT272 GTGTGTGTATTTTGATGAGAATAAATTAATGAAGTGATTTGCTTATTGGGTTATCACAAA332 AAAAAAAAAAAAAAAAAA350 (2) INFORMATION FOR SEQ ID NO:63: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 762 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (ix) FEATURE: (A) NAME/KEY: CDS (B) LOCATION: 21..527 (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 342..449 (D) OTHER INFORMATION: /note="Amino acids 108-113, 128-132, and 135-143 correspond to the peptide sequences by protein microsequencing" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:63: TCTTTTTCATGTTATAAGCCATGTCTAGAGTAAGAAACTTATTTTCTTTC50 MetSerArgValArgAsnLeuPheSerPhe 1510 CTTATTTTCTTTCTCATTATTGCCTTAAATTTCACTAATGGATTAGCC98 LeuIlePhePheLeuIleIleAlaLeuAsnPheThrAsnGlyLeuAla 152025 ATTGATCATAAATCTGATGCTAATATTGCATTAATCCCACAAAAGAAA146 IleAspHisLysSerAspAlaAsnIleAlaLeuIleProGlnLysLys 303540 GGATTAAAATGGTTGCATTGGCCATTTGCACATGCACCACCACCACCT194 GlyLeuLysTrpLeuHisTrpProPheAlaHisAlaProProProPro 455055 TCATCTTTTTTTCCTAAGTTTCCATTTCCAAAAATATTTCCCTGGCCG242 SerSerPhePheProLysPheProPheProLysIlePheProTrpPro 606570 CGATTTTTGCCACCTAAGCCTTTTTCGCCTAGTGAAAAACGCGTCAGT290 ArgPheLeuProProLysProPheSerProSerGluLysArgValSer 75808590 GACATAAACATAGACAACAGTCAGAACGTGTTGGACAAGAAATATTAT338 AspIleAsnIleAspAsnSerGlnAsnValLeuAspLysLysTyrTyr 95100105 TGTGCTTTAATTATTGAGGCGTGTATGCTTGAGAGGGATATGCTTTGC386 CysAlaLeuIleIleGluAlaCysMetLeuGluArgAspMetLeuCys 110115120 GTTCGCAATAGATGTACCTTCTCTTATGATTGTTGTACTGCCATTAAT434 ValArgAsnArgCysThrPheSerTyrAspCysCysThrAlaIleAsn 125130135 ACTGAATTTGGTCCTGAGGAATGTAACCAGTATGAATTTGCCATGATC482 ThrGluPheGlyProGluGluCysAsnGlnTyrGluPheAlaMetIle 140145150 AAAAATCAGTGTGCAAATCAAGCTGCTCCTCCTCCTACAGATTAC527 LysAsnGlnCysAlaAsnGlnAlaAlaProProProThrAspTyr 155160165 TAAGTTATTAAGGGGATGTATGTGTACGTGCGTGTTTATATAGCCGACCCCAACTTGTTT587 GGGACTTAGGCGTAGTTTGCATTATTGTTATTTTATATATGTGTATGTATTCATATAGTC647 GACCCCAACTTGTTTGGGATTGAGACGTAATTGGTGTTTATTAGTATGTATGTGTGTGTA707 TTTTGATGAGAATAAATTAATGAAGTGATTTGCTTATTGGGTTATCAAAAAAAAA762 (2) INFORMATION FOR SEQ ID NO:64: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 80 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 1..20 (D) OTHER INFORMATION: /note="T3 promoter sequence" (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 63..80 (D) OTHER INFORMATION: /note="Adaptor sequence is equal to nucleotides 150-167 of the NaAGP1 cDNA" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:64: TGTTATTAACCCTCACTAAAGCATCACCACCAGCACCACCAACAGCAGACACACCAGCAG60 CTATGATCATACCTGCATCT80 (2) INFORMATION FOR SEQ ID NO:65: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 85 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 1..19 (D) OTHER INFORMATION: /note="T7 promoter sequence" (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 68..85 (D) OTHER INFORMATION: /note="Adaptor sequence is equal to nucleotides 444-461 of the NaAGP1 cDNA" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:65: NCTAATACGACTCACTATAGGCTGATGGTGGTGTTGCTGTTGGTGGTGTTGCTGTTGGTG60 ATTTTGCGGGAGTATCAGTCAAAAG85 (2) INFORMATION FOR SEQ ID NO:66: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 893 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (ix) FEATURE: (A) NAME/KEY: CDS (B) LOCATION: 70..504 (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 70..138 (D) OTHER INFORMATION: /note="Putative secretion signal peptide" (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 70..339 (D) OTHER INFORMATION: /note="Amino acids 70-138, 38-53, and 71-90 are sequences which match the peptide sequences obtained by protein sequencing" (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 87..324 (D) OTHER INFORMATION: /note="Amino acids 40, 44, 45, 49, 50, 74, 76, 80, 81, and 85 are hydroxylated prolines" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:66: CTCTCTCGCTCACTCATCAAATTCTCTCTCTCTCTCTCTCTCTCTCTTTCTCTCTCTCTC60 TCTCTAAAAATGAAGATGGGTTTTGCAGGGTTCCAAGTTTTGATGGTT108 MetLysMetGlyPheAlaGlyPheGlnValLeuMetVal 1510 TTGGGTCTGTTGGCCACATCATGCATAGCCCAAGCCCCAGGAGCAGCA156 LeuGlyLeuLeuAlaThrSerCysIleAlaGlnAlaProGlyAlaAla 152025 CCCACAGCTTCACCCCCAACCGCAAAGTCGCCAACCGCCACCCCACCA204 ProThrAlaSerProProThrAlaLysSerProThrAlaThrProPro 30354045 ACCGCCACACCGCCATCAGCCGTACCAGTTCCATCACCCAGCAAAACA252 ThrAlaThrProProSerAlaValProValProSerProSerLysThr 505560 CCAACCGCGTCACCAACTCCATCACCAGTGACAGCACCAACCCCAAGT300 ProThrAlaSerProThrProSerProValThrAlaProThrProSer 657075 GCCTCCCCACCATCTTCCACACCAGCTTCCACCCCAGCTTCCACTCCA348 AlaSerProProSerSerThrProAlaSerThrProAlaSerThrPro 808590 GCAGCTAAGTCTCCATCGTCGTCAGCTGCTCCCTCAGGCTCAAGCCCG396 AlaAlaLysSerProSerSerSerAlaAlaProSerGlySerSerPro 95100105 AACTCCCCACCGGCTGACGCTATTCCTCCAAGTGGCACCTCCGCCATC444 AsnSerProProAlaAspAlaIleProProSerGlyThrSerAlaIle 110115120125 AGCCGCGTTGCTATTGCTGGAACTGCTCTTGCTGGAGTTTTCTTCGCG492 SerArgValAlaIleAlaGlyThrAlaLeuAlaGlyValPhePheAla 130135140 ATTGTGTTGGCTTAGATTCATGGGATTTGCTCTTTCGGGTTTTCCTATTGGT544 IleValLeuAla 145 CCACGTGGAGACTCACATCTGCTCTTAGATCTGGGTTTTGATGGACGGTCGAGATCTATT604 AATTTCTTTTTATTTTGTTGCTTATTTTCGTTAATGTTTTTTGTATTTTTGTTTAACTCT664 GTTTTCATGCCATATGGTGATTATTGGTTTGGCAGTCTATGGTGGATTTGGACGGTCGTG724 ATGTGATTAATTATGGTGATTCATTGTTTTAGAGTTGACAAGTGCACCCATTTGTAGATG784 AGTCGTTGGATGTACATCTGTCCGATCATAGTTTAATAAAACAGTTTGTCATTCTTTTTC844 TTATGGATCTTCAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA893 (2) INFORMATION FOR SEQ ID NO:67: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 12 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4..8 (D) OTHER INFORMATION: /note="At positions 4 and 8 X=O=Hydroxyproline." (xi) SEQUENCE DESCRIPTION: SEQ ID NO:67: ValSerAlaXaaSerGlnSerXaaSerThrAlaAla 1510 (2) INFORMATION FOR SEQ ID NO:68: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 13 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 3..7 (D) OTHER INFORMATION: /note="At positions 3 and 7 X=O=Hydroxyproline" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:68: AspThrXaaAlaPheAlaXaaSerGlyGlyValAlaLeu 1510 (2) INFORMATION FOR SEQ ID NO:69: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 26 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4..25 (D) OTHER INFORMATION: /note="At positions 4, 5, 7, 8,

13, 17, and 25 X=O=Hydroxyproline" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:69: LeuAlaSerXaaXaaAlaXaaXaaThrAlaAspThrXaaAlaPheAla 151015 XaaSerGlyGlyValAlaLeuProXaaSer 2025 (2) INFORMATION FOR SEQ ID NO:70: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 7 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 5 (D) OTHER INFORMATION: /note="X=O=Hydroxyproline" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:70: ThrAlaAspThrXaaAlaPhe 15 (2) INFORMATION FOR SEQ ID NO:71: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 20 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 3..6 (D) OTHER INFORMATION: /note="At positions 3 and 6 N=I." (xi) SEQUENCE DESCRIPTION: SEQ ID NO:71: ACNGCNGAYACNCCNGCNTT20 (2) INFORMATION FOR SEQ ID NO:72: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 712 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (ix) FEATURE: (A) NAME/KEY: CDS (B) LOCATION: 85..480 (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 85..147 (D) OTHER INFORMATION: /note="Putative secretion signal." (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 154..393 (D) OTHER INFORMATION: /note= "154-156,244-246,256-258,298-300,301-303,307-309 = hydroxyproline." (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 235..408 (D) OTHER INFORMATION: /note="235-268, 289-330, 331-365, and 366-408 are internal peptide sequences from amino acid sequencing." (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 205..408 (D) OTHER INFORMATION: /note="235-268 =SEQ ID NO:67; 289-330 =SEQ ID NO:51; 319-357 =SEQ ID NO:68; 331-365 =SEQ ID NO:50; 366-408 =SEQ ID NO:53" (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 154..393 (D) OTHER INFORMATION: /note= "310-312,325-327,337-339,361-363,378-380,391-393 = Hydroxyproline." (xi) SEQUENCE DESCRIPTION: SEQ ID NO:72: TGAAGAAACTTACACTTTCTCTCTCTGAAAACTGCTCAACACTTCAAATCAGAGTTTTCG60 AAAAGCTTCTAGAGAGAGAAAGAAATGGCTTACTCAAGGATGATGTTCGCT111 MetAlaTyrSerArgMetMetPheAla 15 TTCATTTTCGCTTTGGTCGCCGGATCTGCTTTTGCTCAGGCTCCGGGA159 PheIlePheAlaLeuValAlaGlySerAlaPheAlaGlnAlaProGly 10152025 GCTTCTCCCGCAGCTTCACCGAAGGCATCACCGGTTGCACCAGTAGCA207 AlaSerProAlaAlaSerProLysAlaSerProValAlaProValAla 303540 TCACCTCCAACTGCTGTTGTTACACCGGTATCCGCTCCATCACAATCT255 SerProProThrAlaValValThrProValSerAlaProSerGlnSer 455055 CCTTCTACTGCTGCATCTCCTTCTGAATCTCCATTGGCATCTCCACCA303 ProSerThrAlaAlaSerProSerGluSerProLeuAlaSerProPro 606570 GCTCCACCAACTGCTGACACTCCAGCATTTGCTCCCTCCGGCGGCGTT351 AlaProProThrAlaAspThrProAlaPheAlaProSerGlyGlyVal 758085 GCTCTTCCTCCATCCATCGGCTCTGCTCCCGCCGGTTCTCCAACCTCG399 AlaLeuProProSerIleGlySerAlaProAlaGlySerProThrSer 9095100105 TCTCCTAACGCTGCTTCCTTGAACAGAGTCGCCGTCGCTGGATCTGCA447 SerProAsnAlaAlaSerLeuAsnArgValAlaValAlaGlySerAla 110115120 GTTGTAGCGATCTTCGCTGCATCTTTGATGTTTTAGATCTGAGGAGAGTTTGC500 ValValAlaIlePheAlaAlaSerLeuMetPhe 125130 ATTTTGGATTTTCACGAGATGTTTATTATTTTAGGATTTATTTAGTTCATCTTACTCGTT560 GATGTTTATTCGTTTTGTTTTACTTTTACCCGTGGGCGGTGGTGACTGCGTACATGCTAT620

TGATTTGATTTTTACTCTGGTTATTGTTTATTGTTACTACCACTATTATTATTATGGATT680 CTTTGTTTATTTATGAAGCACTATGATTTACA712 (2) INFORMATION FOR SEQ ID NO:73: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 6 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:73: GlnAlaProGlyAlaAla 15 (2) INFORMATION FOR SEQ ID NO:74: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 19 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:74: AlaGluAlaGluAlaXaaThrXaaAlaLeuGlnValValAlaGluAla 151015 XaaGluLeu (2) INFORMATION FOR SEQ ID NO:75: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 5 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 2..4 (D) OTHER INFORMATION: /note="In positions 2 and 4 X=O=Hydroxyproline." (xi) SEQUENCE DESCRIPTION: SEQ ID NO:75: ThrXaaAlaXaaAla 15 (2) INFORMATION FOR SEQ ID NO:76: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 19 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 6..17 (D) OTHER INFORMATION: /note="At positions 6, 8, and 17 X=O=Hydroxyproline." (xi) SEQUENCE DESCRIPTION: SEQ ID NO:76: AlaGluAlaGluAlaXaaThrXaaAlaLeuGlnValValAlaGluAla 151015 XaaGluLeu (2) INFORMATION FOR SEQ ID NO:77: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 11 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4..11 (D) OTHER INFORMATION: /note="At positions 4, 8, 11 X=O=Hydroxyproline." (xi) SEQUENCE DESCRIPTION: SEQ ID NO:77: ValSerXaaXaaValGlnSerXaaAlaXaaXaa 1510 (2) INFORMATION FOR SEQ ID NO:78: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 15 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4..13 (D) OTHER INFORMATION: /note="At positions 4,8,11-13 X=O=Hydroxyproline." (xi) SEQUENCE DESCRIPTION: SEQ ID NO:78: ValXaaXaaXaaValGlnSerXaaAlaSerXaaXaaXaaThrThr 151015 (2) INFORMATION FOR SEQ ID NO:79: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 9 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 3..8 (D) OTHER INFORMATION: /note="At positions 3,7,8 X=O=Hydroxyproline." (xi) SEQUENCE DESCRIPTION: SEQ ID NO:79: IleSerXaaAlaSerThrXaaXaaThr 15 (2) INFORMATION FOR SEQ ID NO:80: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 15 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 3..14 (D) OTHER INFORMATION: /note="X=O=Hydroxyproline." (xi) SEQUENCE DESCRIPTION: SEQ ID NO:80: IleSerXaaAlaSerThrXaaXaaThrXaaAlaSerXaaXaaThr 151015 (2) INFORMATION FOR SEQ ID NO:81: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 8 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 3..7 (D) OTHER INFORMATION: /note="At positions 3 and 7 X=O=Hydroxyproline." (xi) SEQUENCE DESCRIPTION: SEQ ID NO:81: PheSerXaaThrIleSerXaaAla 15 (2) INFORMATION FOR SEQ ID NO:82: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 17 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 3 (D) OTHER INFORMATION: /note="At position 3 X=(A)=A without absolute certainty." (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 5..12 (D) OTHER INFORMATION: /note="At positions 5 and 12 X=O=Hydroxyproline." (xi) SEQUENCE DESCRIPTION: SEQ ID NO:82: XaaAlaXaaThrXaaSerLeuAspValGlyIleXaaSerSerAsnAla 151015 Thr (2) INFORMATION FOR SEQ ID NO:83: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 9 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 1 (D) OTHER INFORMATION: /note="X=T/P" (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 3..7 (D) OTHER INFORMATION: /note="At positions 3,6,7 X=O=Hydroxyproline." (xi) SEQUENCE DESCRIPTION: SEQ ID NO:83: XaaSerXaaAlaThrXaaXaaAlaThr 15 (2) INFORMATION FOR SEQ ID NO:84: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 13 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 4..6 (D) OTHER INFORMATION: /note="At positions 4 and 6 X=O=Hydroxyproline." (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 8 (D) OTHER INFORMATION: /note="X=(O)=O without absolute certainty, where O=Hydroxyproline." (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 12 (D) OTHER INFORMATION: /note="X=(N)=N without absolute certainty." (xi) SEQUENCE DESCRIPTION: SEQ ID NO:84: XaaAlaAlaXaaAlaXaaSerXaaXaaProThrXaaThr 1510 (2) INFORMATION FOR SEQ ID NO:85: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 17 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 6..16 (D) OTHER INFORMATION: /note="At positions 6,10,12,14,16 X=O=Hydroxyproline." (xi) SEQUENCE DESCRIPTION: SEQ ID NO:85: ValValAlaGluAlaXaaGluLeuValXaaThrXaaValXaaThrXaa 151015 Ser (2) INFORMATION FOR SEQ ID NO:86: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 11 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 3..9 (D) OTHER INFORMATION: /note="At positions 3,5,7,9 X=O=Hydroxyproline." (xi) SEQUENCE DESCRIPTION: SEQ ID NO:86: LeuValXaaThrXaaValXaaThrXaaSerTyr 1510 (2) INFORMATION FOR SEQ ID NO:87: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 4 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:87: TyrThrGluArg (2) INFORMATION FOR SEQ ID NO:88: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 29 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE: (A) NAME/KEY: Peptide (B) LOCATION: 6..27 (D) OTHER INFORMATION: /note="X=O=Hydroxyproline." (xi) SEQUENCE DESCRIPTION: SEQ ID NO:88: AlaGluAlaGluAlaXaaThrXaaAlaLeuGlnValValAlaGluAla 151015 XaaGluLeuValXaaThrXaaValXaaThrXaaSerTyr 2025 (2) INFORMATION FOR SEQ ID NO:89: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 59 base pairs

(B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (xi) SEQUENCE DESCRIPTION: SEQ ID NO:89: TTCCTGCAGAAGCAGAAGCACCAACACCAGCACTACAAGTAGTAGCAGAAGCACCACAA59 (2) INFORMATION FOR SEQ ID NO:90: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 60 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (xi) SEQUENCE DESCRIPTION: SEQ ID NO:90: CTGGAGCTCATATGATGGTGTTGGTACTGGTGTTGGTACTAGTTCTGGTGCTTCTCCTAC60 (2) INFORMATION FOR SEQ ID NO:91: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 1040 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (ix) FEATURE: (A) NAME/KEY: CDS (B) LOCATION: 5..886 (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 5..63 (D) OTHER INFORMATION: /note="Predicted secretion signal." (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 452..552 (D) OTHER INFORMATION: /note="A long direct repeat." (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 572..672 (D) OTHER INFORMATION: /note="A long direct repeat." (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 161..255 (D) OTHER INFORMATION: /note="A sequence matching the peptide sequences obtained from the AGP protein backbone." (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 176..178 (D) OTHER INFORMATION: /note="Codes for Hydroxyproline." (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 182..184 (D) OTHER INFORMATION: /note="Codes for Hydroxyproline." (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 209..211 (D) OTHER INFORMATION: /note="Codes for Hydroxyproline." (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 221..223 (D) OTHER INFORMATION: /note="Codes for Hydroxyproline." (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 227..229 (D) OTHER INFORMATION: /note="Codes for Hydroxyproline." (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 233..235 (D) OTHER INFORMATION: /note="Codes for Hydroxyproline." (ix) FEATURE: (A) NAME/KEY: misc.sub.-- feature (B) LOCATION: 239..241 (D) OTHER INFORMATION: /note="Codes for Hydroxyproline." (xi) SEQUENCE DESCRIPTION: SEQ ID NO:91: AGCAATGGCTTCCTTTGCAAAACCACTTCCATTTTTCTTCCTTCTGGTG49 MetAlaSerPheAlaLysProLeuProPhePhePheLeuLeuVal 151015 CTAGTTTGCTCTTCCTTTATACACATCATTCATGCTAGACAGAGCGTG97 LeuValCysSerSerPheIleHisIleIleHisAlaArgGlnSerVal 202530 TCCTTCAGCAAGGTCACCCACAACGACAACAACAACAACAACAACGAT145 SerPheSerLysValThrHisAsnAspAsnAsnAsnAsnAsnAsnAsp 354045 AATTATGTTATGGAGGCGGAGGCCGAAGCACCAACGCCAGCACTACAA193 AsnTyrValMetGluAlaGluAlaGluAlaProThrProAlaLeuGln 505560 GTAGTAGCAGAGGCACCGGAACTAGTACCAACACCGGTACCGACACCA241 ValValAlaGluAlaProGluLeuValProThrProValProThrPro 657075 AGTTACACCGAAAGAGACCATGGCAGCAACAGCGCCCTGTATGGTCTT289 SerTyrThrGluArgAspHisGlySerAsnSerAlaLeuTyrGlyLeu 80859095 GGCTCGACCAATTCCCCTTCCACGAAGGAGACTCCAACCACAATTACT337 GlySerThrAsnSerProSerThrLysGluThrProThrThrIleThr 100105110 GATGTTGAAGATCAAATTTTGAGTGAAGAACTTAGCGGTGAAAGTTTT385 AspValGluAspGlnIleLeuSerGluGluLeuSerGlyGluSerPhe 115120125 GATCATCCGAAAGGTAATTACGAAAGCACCAACTTGTTCAACAAGGAC433 AspHisProLysGlyAsnTyrGluSerThrAsnLeuPheAsnLysAsp 130135140 AACATTAATCAAAACACTGGCTACACCGGCAACAGCTACTATGTCAAA481 AsnIleAsnGlnAsnThrGlyTyrThrGlyAsnSerTyrTyrValLys 145150155 AACTACGATGGCAGAGGAGGCTACAACCGCAATCCCCCGGGCGGAGGC529 AsnTyrAspGlyArgGlyGlyTyrAsnArgAsnProProGlyGlyGly 160165170175 AATGGGATTAGTGAACAGCAAGGGATTAGTAATCAGGACATTGGCTAC577 AsnGlyIleSerGluGlnGlnGlyIleSerAsnGlnAspIleGlyTyr 180185190 ACCGGCAACAGTTACTACGTCAAAAACTACGATGGCAGAGGAGGCTAC625 ThrGlyAsnSerTyrTyrValLysAsnTyrAspGlyArgGlyGlyTyr 195200205 AACCGCAATCCCCCGGGTGGAGGCAATGAGATTAGTGAACAGCAAGGG673 AsnArgAsnProProGlyGlyGlyAsnGluIleSerGluGlnGlnGly 210215220 ATGAGTGATACAAGGTTTCTGGAAAATGGTAAATACTATCATGATGTG721 MetSerAspThrArgPheLeuGluAsnGlyLysTyrTyrHisAspVal 225230235 AAGAATGAGATTAAAAATAATAATTTCAATGGTAACTCCGAATCAGAT769 LysAsnGluIleLysAsnAsnAsnPheAsnGlyAsnSerGluSerAsp 240245250255 GGGAGAGGAAGTAACAGAAATGATGTTGAGCGCTACTATGCCAACAGT817 GlyArgGlySerAsnArgAsnAspValGluArgTyrTyrAlaAsnSer 260265270 CACAGCTCCAATGAGTTCAACACCATGGAGGAGTATGATAAGTACCAG865 HisSerSerAsnGluPheAsnThrMetGluGluTyrAspLysTyrGln 275280285 AAGACCCAAGGATATGTGCCCTAAATGATATTCCATGTTTTTAGGTGTGCG916 LysThrGlnGlyTyrValPro 290 TTGAAAACTTAATCAATATATAAGAGATTTTATGGTTTGTTTTGGAATTCCATTTGTCTT976 TTGAATATGTTTTCGCTATAAAAATTAAACCCTTTCCACTCCAAAAAAAAAAAAAAAAAA1036 AAAA1040 __________________________________________________________________________