AU603778B2 - Self-incompatibility gene - Google Patents

Description

I

9m.

i ii OF AUSTRALIA CO M M O N W E A L T H PATENTS ACT 1952 COMPLETE SPECIFICATION (Original) FOR OFFICE USE Class Int. Class Application Number: 2s Lodged: Complete Specification Lodged: Accepted: Published: Priority: Related Art: This document contains the amendments made under Section 49 anrd is correct for printing.

,,Name of Applicant: THE UNIVERSITY OF MELBOURNE, of Pti! I~c Molrno- t Address of Applicant:Grattan Street, Parkville, Victoria, 3052, Australia.

Actual Inventor(s): Adrienne E. CLARKE, Shaio-Lim MAU, Marilyn ANDERSON, Edwina CORNISH, Geoffrey W. TREGEAR, Robert J. CRAWFORD and Hugh D. NIALL.

Address for Service: DAVIES COLLISON, Patent Attorneys, 1 Little Collins Street, Melbourne, 3000.

C t C t t Complete specification for the invention entitled: "SELF-INCOMPATIBILITY GENE" The following statement is a full description of this invention, including the best method of performing it known to US SELF-INCOMPATIBILITY GENE Field of the Invention This invention relates to the identification and isolation of cDNA and genomic DNA coding sequences of an S-gene which controls self-i'ncompatibil ity in a wide variety of self-incompatible plants, particularly exemplified by members of the Solonaceae and Cruciferae. Studies of S-gene products, Sproteins, indicate that they are associated with the expression of the selfincompatibility genotype of such self-incompatible plants.

S-proteins are useful in control of pollen tube growth, for example as natural gametocides to control, induce or promote self-incompatibility and interspecific incompatibility. S-genes and their products can also be used in genetic manipulation of plants to create self-incompatible cultivars. Plants engineered in this way will be valuable for the economic production of hybrid seed.

Background of the Invention Many plant species, including Nicotiana alata and Lycopersicon peruvianum, are self-incompatible, that is they cannot be fertilized by pollen from themselves or by that of a plant of the same S- (or self-incompatibility) genotype. The molecular basis of self-incompatibility is believed to arise from the presence of S-protein in the mature styles of plants; in particular, as exemplified by N. alata and L. peruvianum, S-protein has now been shown to be present in extracts of plant styles at the developmental stages of buds at first show of petal color, and at the subsequent stages of maturation of open i i I :i I V r t I I IP I: t I 1 1 It I II I I 4L

IC

IL i2 I I I IC but immature flowers, and flowers having mature glistening styles. On the other hand, S-protein is not present in the earlier developmental stages of green bud and elongated bud.

For general reviews of self-incompatibility, see de Nettancourt (1977) Incompatibility in Angiosperms, Springer-Verlag, Berlin; Heslop-Harrison (1978) Proc. Roy. Soc. London B, 202:73; Lewis (1979) N.Z. J. Bot. 17:637; Pandey (1979) N.Z. J. Bot. 17:645 and Mulcahy (1983) Science 220:1247. Selfincompatibility is defined as the inability of female hermaphrodite seed plants to produce zygotes after self-pollination. Two types of selfincompatibility, gametophytic and sporophytic, are recognized. Gametophytic incompatibility is most common and in many cases is controlled by a single nuclear gene locus (S-locus) with multiple alleles. Pollen expresses its haploid S-genotype and matings are incompatible if the S-allele exoressed is the same as either of the S-alleles expressed in the diploid tissue of the 15 pistil. During both incompatible and compatible matings, pollen tubes germinate and grow through the stigma into the transmitting tissue of the style. Tube growth from incompatible pollen grains is arrested in the upper third of the style.

In sporophytic incompatibility, pollen behavior is determined by the genotype of the pollen-producing plant. If either of the two S-alleles in the pollen parent is also present in the style, ;.ollen tube growth is inhibited.

Unlike the gametophytic systems, inhibition usually occurs at the stigma surface and not in the style. In sporophytic incompatibility, S-protein may be concentrated at or near the stigma surface. The gametophytic polyalle'ic 25 system is considered to be the ancestral form of self incompdtibility i', flowering plants with the sporophytic system being derived from it (d Nettancourt 1977). The products of the S-gene in the two systems are considered to be structurally related.

There are five species of gametophytically self-incompatible plants and two species of sporophytically incompatible plants in which style or stigma proteins apparently related to S-genotype have been detected by either Selectrophoretic or.immuhological methods-. In N. alata, an association between specific protein bands and three S-allele groups was demonstrated by isoelectric focussing of stylar extracts (Bredemeijer and Blaas (1981) Theor.

I r Appl. Genet. 59:185). Two major antigenic components have been identified in mature styles of a Prunus avium cultivar of S 3 S4 genotype, one of which (Santigen) was specific to the particular S-allele group (Raff, et al. (1981) Planta 153:125; and Mau, et al. (1982) Planta 156:505). The S-antigen, a glycoprotein, was a potent inhibitor of the in vitro growth of pollen tubes from a S3S cultivar (Williams, et al. (1982) Planta 156:577). The glycoprotein was resolved into two components, purportedly representing the S and S4 products of the S 3 S4 genotype. Stylar protein components which have been associated with the S-allele group or the self-incompatibility genotype are reported in Petunia hybrida (Linskens (1960) Z. Bot. 48:126), Lilium longiflorum and Trifolium pratens (Heslop-Harrison (1982) Ann. Bot. 49:729).

A glycoprotein corresponding to genotype S 7 of Brassica campestris has been isolated from extracts of stigmas by gel-filtration followed by affinity chromatography and isoelectric focussing (Nishio and Hinata (1979) Jap. J.

Genet. 54:307). Similar techniques were used to isolateS-specific glycoproteins from stigma extracts of Brassica oleracea plants homozygous for S-alleles S39, S2 and S 7 (Nishio and Hinata (1982) Genetics 100:641).

Antisera raised to each isolated S-specific Brassica oleracea glycoprotein not only precipitated its homologous glycoprotein but also reacted with the other two S-specific glycoproteins of B. oleracea and the S 7 -specific glycoprotein of B. campestris (Hinata et al. (1982) Genetics 100:649). An S-specific glycoprotein was isolated by Ferrari et al. (1981) Plant Physiol. 67:270 from a stigma extract of B. oleracea using sucrose gradient sedimentation and double diffusion tests in gels in which the proteins were identified by Coomassie Blue staining. This preparation was shown to be biologically active since pretreatment of S2 pollen with the glycoprotein prevented the pollen from germinating on normally compatible stigmas. Recently a cDNA clone encoding part of an S-locus specific glycoprotein from B. oleracea stigmas has been described (Nashrallah et al. (1985) Nature 318:263-267.

In recent work that is detailed in Australian Patent Application No.

28736/84, filed May 25 1984 which is expressly incorporated by reference herein, stylar extracts of several self-incompatibility genotypes from both Nicotiana alata and Lycopersicon peruvianum were examined for the presence of S-gene associated protein. Glycoprotein materials were identified in the 30,000 MW region of stylar extracts of genotypes $Si11, S -S S and S3 of N. alata and of genotypes SI S 2

S

3 11-SI S2S2' S 3

S

3 and S 3 S4 of L.

peruvianum. By comparing two-dimensional gel electrophoresis of stylar extracts of the different genotypes, closely related, but distinct glycoproteins were found to segregate with the individual S-alleles. For example, the N. alata S-protein was found only in stylar extracts of the genotypes containing the 52-alleles (52S3 and S For each genotype, the genotype specific glycoprotein only appeared as the flower matured, and was detected only in stylar extracts of buds at first show of petal color and in later stages of maturation, but not in earlier bud stages. Therefore, the appearance of these glycoproteins is temporally coincident with the appearance of the self-incompatibility phenotype. The S2-glycoprotein of N. alata and the S and S3-proteins of L. peruvianum were shown to be more highly concentrated in the upper style sections, which is the zone in which pollen tube inhibition occurs. Therefore, the appearance of these glycoproteins is spacially coincident with the self-incompatibility reaction. Further, corroboration of the biological activity of.S2-protein of N. alata was demonstrated by its inhibition of pollen tube growth in an in vitro assay i (Williams, et al. (1982)).

SA significant aspect of the work disclosed in Application No.

28736/84 was the discovery that rabbit antisera and monoclonal antibodies raised to individual S-proteins or stylar extracts showed immunoloyical crossreaction between S-proteins of different genotype within the same species, between S-proteins of different species and also between species having gametophytic incompatibility and sporophytic incompatibility. It was concluded therein that there is structural homology among S-proteins, and that despite apparent differences in molecular weight and pl, these proteins are a recognizable structural class in addition to their functional similarities.

Patent Application No. 28736/84 also disclosed a method of purification for S-proteins which included fractionation of stylar extracts by 30 ion exchange chromatography followed by a second fractionation by affinity chromatography. The method of purification was exemplified with the isolation i of the 32

K-

2 -glycoprotein from Nicotiana alata styles.

The S-proteins that have been identified are glycoproteins, which are proteins that have been modified by covalent bonding of one or more -4-

L-

r r carbohydrate groups. Little is known of the composition and structure of the carbohydrate portion of S-proteins. It is, as yet, unclear what contribution, if any, the carbohydrate portion of the S-protein makes to biological activity in the incompatibility reaction. Petunia hybrida stylar mRNA is translated in Xenopus laevis (frog) egg cells to produce active proteins which induce the incompatibility reaction. The relative glycosylation of S-proteins produced in frog egg cells to that of the S-proteins produced in the plant is unknown; however, the post-translational processing in the foreign system is adequate to produce biologically active proteins (Donk, van der J. A. W. (1975) Nature 256:674-675).

Most proteins, such as the S-proteins, that are excreted from or transported within cells have signal or transit sequences that function in the translocation of the protein, for example see: Perlman, D. and Halverson, (1983) J. Mol. Biol. 167:391-409; Edens, L. et al- (1984) Cell 37:629- 633.; and Messing, J. et al. in Genetic Engineering of Plants, ed. Kosuge, T.

et al. (1983) Plenum Press, New York, pp. 211-227. Signal or transit DNA sequences are generally adjacent to the 5' end of the DNA encoding the mature protein, are co-transcribed with the mature protein DNA sequence into mRNA and are co-translated to give immature proteins with the signal or transit peptide attached. During the translocation process the signal or transit peptide is cleaved to produce the mature protein.

The expression of S-genes in self-incompatible plants shows very complex regulation, with S-gene products appearing in only certain tissues at certain times. The mechanism of this regulation is not yet known in detail, but Sinvolves the presence of specific regulatory DNA sequences in close proximity to the genomic DNA that encodes the S-protein. Adjacent to the structural gene and signal or transit sequences, are promoter sequences that control the initiation of transcription and exert control over protein expression levels.

3 Su ,mary of the Invention.

According to one aspect of the present invention, there h 'is provided a recombinant vector comprising a DNA sequence encoding an S-protein of a self-incompatible plant. The plant may display either gametophytic or sporophytic self-incompatibility, and may be a member of the family Solanacae (for example, the genus Nicotiana.or Lycopersicon), or of the family Cruciferae (for example, the genus Brassica).

tl_ S 4 In one particular embodiment of this aspect of the invention, the S-protein is the S2-protein of Nicotiana alata, or an S-protein substantially homologous therewith or immunologically cross-reactive with the antibody therewith. In another embodiment, the S-protein is the S_-protein or the S3-protein of Lycopersicon peruvianum.

Preferably, the recombinant vector is AgtlO or M13 mp 8.

j 10 According to another aspect of this invention, there is provided a recombinant DNA molecule comprising a DNA sequence which encodes an S-protein, or a promoter or signal sequence of an S-protein, of a self-incompatible plant.

The promoter sequence may, for example, be that of the S2-allele of Nicotiana alata. Similarly, the signal sequence may be that of the S2-protein of N.alata or an S-protein substantially homologous thereto. In one preferred embodiment the signal sequence is encoded by DNA comprising the nucleotide sequence: CCG GCA TCA AAA CCT CCC GCC AAG CTG GAC AGA CTC CAG CTA ACG TCA GTT TTC TTC ATT TTG CTT TGT GCT CTT TCA CCG ATT TAT GGG -3' Finally, the present invention provides a method for isolating and identifying an S-gene comprising the step of screening clones which comprise DNA from a self-incompatible plant with a hybridization proble, the DNA sequence of which is based on the DNA sequence encoding the S2-protein of N.alata.

The clones may comprise DNA from a genomic DNA library of a self-incompatible plant, for example, DNA from a cDNA library which is prepared by reverse transcription on a template of mRNA of a self-incompatible plant.

r The present work describes the isolation and identification of cDNA encoding an S-gene protein and is exemplified with the 2 -allele associated glycoprotein of Nicotiana alata. In summary, the initial isolation involved the following general steps: 6 1. isolation and purification of mature S-protein from styles; 2. N-terminal amino acid sequencing of the mature S-protein; 3. chemical deglycosylation of mature S-protein and identification of I deglycosylated S-protein translation products in plant tissue; 4. preparation of a cDNA library from poly(A RNA of mature styles followed by differential screening of the library with radioactively labelled cONA from ovary and green bud style to remove non-mature d style specific cDNA and obtain mature style specific clones; preparation of an S-protein specific oligonucleotide probe, based on either the amino acid sequence of the S-protein or the nucleotide sequence of a cDNA fragment produced from stylar mRNA by specific priming with mixed oligonucleotide primers which are based on the Sprotein aniino acid sequence. Alternatively, the specifically primed cDNA fragment can be used directly as the S-protein probe; 6. screening of the mature style specific clones with an S-protein specific DNA probe to obtain full or nearly full length clones which encode the S-protein and its attendant signal or transit sequence.

It is an important aspect of this invention that a nearly full length N.

aLata 2-protein cDNA clone, obtained as summarized above, hybridizes in Northern blot experiments with N. alata mature style poly(A RNA not only from the SS and 12S3 gehotypes, but also from the SS3 and S_5S genocypes.

Further, the N. alata S-protein specific clone also hybridizes with style mRNA from L. peruvianum (genotype SIS) and B. oleracea (mixed genotype).

These observations demonstrate that substantial DNA homology exists among Sgenes in addition to the protein homology evidenced by the existence of immunological cross-reaction among S-proteins. A further confirmation of substantial amino acid sequence homology among S-proteins is shown herein by d comparison of the N-terminal amino acid sequence of the N. alata S,-protein with the N. alata S6-protein and the L. peruvianum SI and S3-proteins. In the 1 I region sequenced (amino acids 1-15, Figure the N. alata S2-protein is homologous to the N. alata S6-protein, 67% homologous to the L. peruvianum Sprotein, and 53% homologous to the L. peruvianum S 3 -protein.

A further aspect of this invention is the use of the nearly full length N. alata S-protein specific cDNA clone as a hybridization probe for the isolation and identification of cDNA clones specific for other S-proteins.

Although the procedure summnarized above (steps 1-6) is generally applicable to any S-protein, a major improvement to the procedure is the use of the N. alata I2-protein clone in step 6, as a probe in a hybridization screen to detect clones which contain S-protein specific cDNA. The availability of the S2protein specific probe obviates the need to isolate each individual S-protein, determine the N-tenninal protein sequence and prepare a unique probe for each _S-protein. 'he use of the S-specific cDNA clone will greatly facilitate the S' isolation of cDNA clones for other S-proteins, particularly when the limited I 15 availability of pure S-protein makes protein sequencing difficult if not tci impossible. There are several recent examples of the application of cDNA Shybridization probes for the identification of other cDNA clones which encode S, homologous proteins. The basic procedures employed are described in Staswich, P. and Chrispeels, M. (1984) J. Molecular Appl. Genet. 2:525; Dennis, E. S. et al. (1984) Nucleic Acids Research 12:3983; and Dennis, E. S. et al. (1985) SNucleic Acids Research 13:727, for example.

A full-length cDNA clone of the N. alata S--gene was obtained by hybridization screening using the nearly-full length N. alata S-cDNA as a probe. This full-length clone contains a sequence that encodes the mature Sprotein as well as the complete signal or transit polypeptide sequence. This signal sequence functions in the extracellular translocation of the mature protein from the transmitting tract cells. The transmitting tract is the tissue through which the pollen tubes grow on their way to the ovary, and it is in this tissue that the S-gene is expected to be expressed.

Signal or transit DNA sequences are useful in combination with the adjacent DNA sequences of the mature protein in affecting the excretion or translocation of the mature protein in heterologous expression systems.

Further, the presence of signal or transit DNA sequences have been shown to enhance protein expression levels (Edens, L. et al., 1984). Signal or transit

I

IDIC-I1~ji i 1__ l I sequences are useful in the construction of chimaeric genes in which they are fused to a heterologous protein, for example in a recombinant vector to direct translocation of the protein. Examples of such heterologous fusions are found in prokaryotic systems, Palva, I. et al., (1982) Proc. Natl. Acad. Sci. USA 79:5582-5586 and Palva I. et al., (1983) Gene 22:229-235; and eukaryotic systems, Edens, L. et al. (1984) and Rose, M. et al. (1981) Proc. Natl. Acad.

Sci USA 78:2460-2464.

Plant signal or transit sequences are particularly important for use in combination with their adjacent DNA sequences or in chimaeric gene fusions with heterologous proteins to target mature protein to specific organelles in plant cells or for excretion from the cell. (Van den Bioeck, G. et al., (1985) Nature 313:358-363). For example, such plant chimaeras can be used for the targeting of proteins toxic to plant pathogens or of gene products associated with pesticide resistance in plant tissue.

15 A full or nearly full-length cDNA clone coding for the N. alata S2- Xl protein, is also useful as a probe of a style genomic library in order to S isolate and identify a chromosomal gene encoding an S-protein. In addition to the mature protein sequence and signal or transit sequence, a genomic clone Scan be obtained which includes the promoter sequences involved in regulating the timing and rate of transcription and expression of the S2-protein. The screening of genomic libraries with cDNA probes is well known in the art. The basic procedures as applied to plant genes are described in Hoffman, L. M.

(1984) J. Mol. Genet. 2:447-453 and Mazure, B. J. and Chui, (1985) Nucleic Acids Research 13:2373, for example.

The S-protein DNA coding sequences that are isolated by the techniques described herein can be used in heterologous in vivo expression systems to direct synthesis of S-protein which can thereby be produced in significant C (amounts in biologically active form to be used, for example, as natural gametocides. The DNA sequence encoding the mature S-protein can be so.

employed separately or in combination with its attendant regulatory sequences.

Promoter sequences are likewise useful in combination with DNA sequences encoding proteins in effecting the transcription of DNA sequences and exerting control over protein expression levels in heterologous expression systems.

Plant promoter sequences are particularly useful for the expression of heterologous proteins in plant cells, since promoters derived from prokaryotes or non-plant eukaryotes may not function efficiently in plant cells.

The present invention provides novel genecic constructs containing DNA encoding an S.-protein associated with the se! -incompatibility phenotype.

Knowledge of the amino acid sequence of the mature S2-protein and the successful development of a differential screening procedure were important factors in the initial isolation of the cONA clone. Hybridization experiments and amino acid sequence data have confirmed the existence of amino acid and nucleotide sequence homology among S-proteins and demonstrate that the Sprotein specific clone can be employed as a hybridization probe in the isolation of other S-genes.

Detailed Description of the Invention Details of the isolation and identification of cDNA clones specific to Sproteins and the use of full or nearly full length cDNA clones specific to the N. alata S 2 -protein as hybridization probes are given in the following detailed description and in the accompanying drawings: Figure 1 illustrates the separation of stylar extracts of N. alata genotypes S212S, _S 3 and 3S 3 by selected 2-dimensional gel electrophoresis.

i The protein bands associated with the two alleles are identified.

Figure 2 provides a comparison of the chemically deglycosylated.

mature S2 glycoprotein of N. alata of molecular weight 26 kd, with the in vitro translation products of style poly(A RNA, by SDS-gel electrophoresis. Note the presence of the 27K kd molecular weight protein I band only in the translation products from mature style poly(A RNA. The 27 kd molecular weight translation product is slightly larger than the chemically i deglycosylated mature S2 protein, consistent with the presence of a signal sequence in the 27 kd protein.

Figure 3 presents a comparison of the SDS-polyacrylamide gel electrophoresis of protein extracts from ovary, style and other N. alata (S23) tissue. There is more similarity between the extracts of ovary and style than between extracts of other organs and style, as shown by the protein bands visualized by Coomassie Blue staining.

-J

Figure 4 provides the N-terminal amino acid sequences of the mature N.

alata S2- and S6-proteins and the mature L. peruvianum S and 3 proteins.

Amino acid homology is indicated by solid lines.

Figure 5 shows the production of a 100 bp cDNA fragment from mature style poly(A RNA using synthetic oligonucleotide 14-mers as primers. One oatch primed synthesis of a single 100 bp fragment (tracks 1, 2 and Trdcks 4, and 6 show that only the 100 bp fragment is produced with mature style poly (A RNA when pooled synthetic primers are used. Only traces of the 100 bp fragment are detected from ovary and green bud style poly RNA.

Figure 6 provides the partial nucleotide sequence of the 5 -protein specific 100 bp cDNA fragment on which the preparation of the 30-mer synthetic i probe was based. fucleotides on the 5' side of residue 1 of the mature S- Sprotein are indicated by negative numbers. The deduced amino acid sequence is shown above the nucleotide sequence. The probe sequence is also provided, and is complementary to the underlined nucleotide sequence. A reading error at I nucleotide 24 (amino acid was later discovered. The correct nucleotide at position 24 is A.

Figure 7 provides the nucleotide sequence of the near full-length cDNA coding for the 32K molecular weight-S 2 -protein. cDNA contains the full mature protein encoding sequence and part of the signal encoding sequence.

I Nucleotide residues are numbered in the 5' to 3' direction. Nucleotides on the 5' side of residue 1 of the mature S-protein are indicated by negative i numbers. The deduced amino acid sequence is shown above the nucleotide j sequence. The partial signal sequence is underlined.

I 25 Figure 8 is a Northern blot analysis of mature style poly(A RNA from N.

alata genotypes S 3

S

3 1SS 3 2S2 and 253 L. peruvianum genotypes S~3 and mixed genotypes from B. oleraced. Poly(A RNA from N. alata S2S 3 green bud style and ovary are also included. All tracks dre probed with 32 p-labelled probe from the NA-2-1 clone cDNA insert encoding the N. alata S-protein described infrd.

Figure 9 provides the nucleotide sequence of-the full-length cDNA coding for the 32K molecular weight S2-protein. This cDNA contains the full mature protein encoding sequence and the full signal encoding sequence. Nucleotide -11residues are numbered in the 5' to 3' direction. Nucleotides on the 5' side of residue 1 of the mature S-protein are indicated by negative numbers. The deduced k.ino acid sequence is shown above the nucleotide sequence. The signal sequence is underlined.

The following definitions apply in the specification and claims: The Sgene protein is the product of the S-gene or S-allele. The term protein as used herein also includes glycoprotein. Although the biochemical mechanism of the self-incompatibility reaction is not fully understood, the S-protein is associated with the presence of self-incompatibility. Accordingly, the Sprotein must show segregation with the S-allele; be localized in the tissue where the incompatibility reaction is localized and occur in the appropriate plant tissue in coincidence with the expression of selfincompatibility. In addition, it will be understood that the biological activity of the S-protein in an in vitro assay will provide corroboration that the S-protein is itself functionally active for pollen inhibition. However, it is possible that the active component is a modified protein or a secondary S' product. In such cases, biological activity of the S-protein may require the activity of other components in order to be manifested in a bio-assay system. A mature S-protein is the processed form of the S-protein from which the signal or transit peptide has been cleaved. This is the form of the protein isolated from stylar tissue.

The S-gene or S-allele contains the DNA coding sequences for the mature i S-proteins defined above. Further, the S-gene contains the coding region for a signal or transit peptide and other information necessary to the translation and processing of the S-protein. Further, the S-gene is likely to contain promoter sequences involved in the transcription and expression and processing of the S-protein. A full length cDNA clone comprises the DNA sequence i encoding a mature protein and the entire regulatory signal -or transit sequence.

The term recombinant vector is used herein to designate a DNA molecule capable of autonomous replication in a host eukaryotic or prokaryotic cell, into which heterologous DNA sequences can be inserted, so that the heterologous sequences are replicated in the host cell. Conventional techniques known to those of ordinary skill in the art are used to introduce -12the vector into its host cell (Maniatis eL al., 1982). Recombinant vectors often contain a marker displaying a selectable phenotype such as antibiotic resistance to allow selection of transformed cells.

A DNA molecule that is substantially pure will migrate as a single band in agarose or polyacrylamide gel electrophoresis, using conventional procedures described in Maniatis et al. (1982) and exemplified in Figure The term homology is used in the art to describe a degree of amino acid or nucleotide sequence identity between polypeptides or polynucleotides. The presence of sequence homology is often used to support a genetic or functional relationship between polypeptides or nucleotide sequences. The presence of amino acid sequence homology between polypeptides implies homology between the DNA sequences that encode the individual polypeptides. Since the genetic code is degenerate the degree of homology between polypeptides or proteins is not Snecessarily the same as that between the DNA sequences-that encode them. The degree of homology between polypeptides or polynucleotides can be Squantitatively determined as a percent homology if the sequences are known.

In the absence of sequence information for comparison, the presence of homology is usually determined operationally by experiment. In the case of uNA or RNA sequences, hybridization experiments are used to determine the presence or absence of homology. Since the strength of a particular hybridization signal depends on the experimental conditions used as well as the degree of homology, it is convenient to define homology in relation to the experimental conditions used. We use the term substantially homologous as the degree of homology that must exist between the hybridization probe and a target RNA or DNA sequence in order to select the target sequence from a background of undesired sequences using hybridization experiments as described herein. Based on our knowledge of the degree of amino acid homology between the protein of N. alata andtheher S-proteins, substantial homology is herein quantitatively defined as equal to or greater than about 53% homology.

The term cDNA is understood in the art to denote the single stranded complementary DNA copy made by action of reverse transcriptase on an mRNA template. Hcrein, the term cDNA is also used to denote any single or double stranded DNA that is replicated from this first complementary copy. cDNA coding sequences are distinguished from genomic DNA sequences by the potential -13- -2 presence of intron non-coding sequences in the genomic DNA. In vivo, introns are removed from messenger RNA by splicing events that produce mature mRNA.

It is mature mRNA that is uzsd in the initial preparation of cDNA by reverse transcription.

The term recombinant DNA molecule is used herein to distinguish DNA molecules in which heterologous DNA sequences have been artificially ligated together by the techniques of genetic engineering, for example by in vitro ligation using DNA ligase (Maniatis, T. et al. (1982) Molecular Cloning, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York). Heterologous DNA sequences are derived from different genetic entities.

The process of cloning a DNA fragment involves excision and isolation of the DNA fragment from its natural source, insertion of the DNA fragment into a recombinant vector and incorporation of the vector into a microorganism or cell where the vector and inserted DNA fragment are replicated during proliferation of the microorganism or cell. The term clone is used to designate an exact copy of a particular DNA fragment. The term is also used to designate both the microorganism or cell into which heterologous DNA at C t fragments are initially inserted and the line of genetically identical organism or cells that are derived therefrom.

St 20 The plant materials used in the present work were based on Nicotiana L alata self incompatible heterozygous genotypes S S and SS_ 3 (Dr. K. R. T.

t c Pandey, Genetics Unit, Grasslands Division, DSIR Palmerston North, New Zealand) and on L. peruvianum heterozygous genotypes S-,S and .S_ 3 (Victorian t State Department of Agriculture, Burnley, Victoria, Australia) and Brassica oleracea mixed genotype. Self-incompatibility genotype was confirmed by hand ,t pollination. Homozygous self-incompatibility genotype5 were generated, as described in Patent Application No. 28736/84, by self-pollination of flower buds at the early bud stage before expression of selfincompatibility.

Mature non-pollinated styles were obtained from flowers that had been s, emasculated at the onset of petal coloration or from yellow buds. Thesemature styles were removed and used immediately or stored at -70 0 C. Styles refer to stigmas and style which were excised together. Ovary was separated -14-

I

from styles. Green bud styles refer to immature styles before the onset of self-incompatibility.

Except as noted hereafter, standard techniques for cloning, DNA isolation, amplification and purification, for enzymatic reactions involving DNA ligase, DNA polymerase, restriction endonucleases and the like, and various separation techniques are those known and commonly employed by those skilled in the art. A number of standard techniques are described in: Maniatis et al. (1982) Molecular Cloning, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York; Wu (1979) Meth. Enzymol. 68; Wu et al. (eds.) (1983) Meth. Enzymol. 100 and 101; Grossman and Moldave (eds.) Meth. Enzymol.

Miller (1972) Experiments in Molecular Genetics, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York; Old and Primrose (1981) Principles of Gene Manipulation, University of California Press, Berkeley; Schleif and S; Wensink (1982) Practical Methods in Molecular Biology; Glover (1985) DNA Cloning Vol. I and II, IRL Press, Oxford, UK; Hames and Higgins (eds.) (1985) r Nucleic Acid Hybridization, IRL Press, Oxford, UK; Sellow and Hollaender (1979) Genetic Engineering: Principles and Methods, Vols. 1-4, Plenum Press, New York, which are expressly incorporated by reference herein. Abbreviations C cand nomenclature, where employed, are deemed standard in the field and commonly used in professional journals such as those cited herein.

Isolation of cDNA encoding the 32K S-gene protein of N. alata A method for isolating and purifying the S-gene associated glycoproteins from mature styles had been established using a combination of ion exchange and affinity chromatography Patent Application Serial No. 615,079).

it This method had been applied to the isolation and purification of N. alata S2protein. More recently, purified protein yield improvements have been obtained by using a less basic buffer (pH 7 rather than pH 7.8) in affinity

B

chromatography. The S-protein appears to be more stable at lower pH. As illustrated in Fig. 1, it was possible to isolate a single component of MW 32K associated with the -allele of Nicotiana alata. Chemical deglycosylation of this component yielded a single product of approximately 26 kd in molecular weight, shown in Figure 2a. The results of in vitro translation of mRNA from mature styles, green bud style' and ovary are shown in:Fig. 2b. Total RNA was isolated by conventional methods. Since most mRNA is polyadenylated, poly(aT) cellulose chromatography was used to isolate mRNA, as poly(A RNA. The various poly(A RNA fractions were translated using an amino acid depleted rabbit reticulocyte lysate kit (Amersham No. N.150, Arlington Heights, Ill.) in the presence of tritiated amino acids. An in vitro translation product of approximately 27 kd molecular weight is detected only from mature style mRNA. This product is slightly larger than the chemically deglycosylated protein. It is therefore identified as the full length immatureS2-protein, which is composed of mature S-protein and its signal peptide.

Based on this finding, a protocol of differential screening was adopted as the initial part of the strategy to isolate cDNA coding for S2-protein. A cDNA library was prepared in Xgtl0 phage using mature style poly(A RNA of N.

Salata genotype 213. Mature style poly(A RNA was transcribed into double stranded cDNA by conventional methods (Maniatis et al., 1982). End-repair, EcoRI methylation and EcoRI linker ligation reactions were carried out and the cDNA was cloned into the EcoRI site of the Xgtl0 vector (Huynh, T. et al.

(1985) in Practical Approaches in Biochemistry, DNA Cloning Vol. 1 ed. Glover, D. IRL Oxford, pp. 49-78). This library was subjected to differential screening using 3 2 P-labelled cDNA from mature and green bud styles. The lambda-phage was used to infect Escherichia coli C600 cells. Plaques that hybridized strongly only to the mature style cDNA were selected and differentially screened a second time using 32P-labelled cDNA prepared from either mature style or ovary mRNA. Again plaques that hybridized strongly only to the mature style cDNA were selected. Ovary cDNA was used in this second screen because SDS-gel electrophoresis indicated that extracts of mature style and ovary had some common proteins which were not expressed in green bud styles (Figure Surprisingly, tissues other than ovary and green bud were found to be unsuitable sources of cDNA for differential screening Ssince the protein profiles of other organs were found to be too diverse from that of mature style to be useful. Therefore, differential screening with ovary and green bud cDNA, although considerably less convenient, was necessary to discriminate mature style-specific cDNA. The resultant cDNA clones were specific for mature style.

Once the cDNA mature style library had'been differentially screened, a S2-protein specific DNA probe was required for final screening of the clone library. The first step in the preparation of the probe was the determination -16of the N-terminal amino acid sequence of the N. alata S2-protein (Figure 4).

Conventional microsequencing techniques were used (Hewick, R.M. et al. (1981) J. Biol. Chem 256:7990-7997). The several alternative methods of screening the mature style specific library (cited supra) rely on a knowledge of the amino acid sequence. S- n is made in minuscule amounts at limited times and in limited tissue. Several hundred styles must be dissected from the flowers in order to obtain sufficient pure S-protein for micro-anino acid sequencing. Consequently only short segments of N-terminal sequence could be determined with the amounts of material available for analysis, using conventional microsequencing techniques. Unfortunately, the N-terminal amino acid sequence proved to have highly redundant coding oligonucleotide possibilities. Nevertheless, a partial-length cDNA was isolated by the following procedure. A set of synthetic mixed oligonucleotide primers were prepared based on the partial amino acid sequence. A set of 24 14-mers, covering all the codon ambiguities at amino acids 4-8,-was synthesized. These synthetic mixed oligonucleotides were then used in three batches of eight 14mers each, to prime synthesis of cDNA from N. alata (SS3) mature style poly(A RNA.

As shown in Figure 5, only one batch (No. 165) was found to be specific for the priming reaction. Surprisingly, a single cDNA band 100 nucleotides in length was identified in this reaction. A 100 bp-nucleotide band was also *observed when the pooled 14-mers were used to prime poly RNA from mature styles; only traces of this fragment were detected in priming from ovary or green bud style mRNA.

The 100 nucleotide long band was eluted from an acrylamide gel and sequenced yielding the S2-protein coding sequence fran amino acid -12 in the signal sequence, up to amino acid 2 of the mature protein, as shown in Figure 6. From this sequence a single 30-mer was synthesized which covered the part S',of the signal sequence to -9 and included the first amino acid codon of the coding sequence (Figure This amino acid region was chosen in order to insure that the synthetic probe would identify cDNA clones that extended into the signal sequence codons.- This .strategy was,. adopted .for convehienc,- since adequately l-arge amounts of the synthetic probe could be prepared in a single synthesis. Alternatively, the 100. bp fragment could have been cloned, amplified, purified and radioactively labelled for use as a probe.

-17- The 30-mer was used as an 52-protein specific probe to screen the mature style-specific clones previously identified by differential screening. One of the clones obtained was chosen for further study. The clone, designated NA-2- 1, contained a cDNA insert of 877 bp which could be excised as a single fragment from the lambda vector by EcoRI digestion. The 877bp insert has been cloned into M13 phage (M13mp8) and has been deposited with the American Type Culture Collection, 12301 Parklawn Drive, Rockville, Maryland 20852 (Accession No. 40201).

The nucleotide sequence of the 877bp cDNA insert from NA-2-1 for the 32K style glycoprotein associated with the S-allele of N. alata is given in Figure 7. Nucleotide residues are numbered in the 5' to 3' direction. The deduced amino acid sequence of the protein is shown above the nucleotide sequence and the partial signal sequence is underlined.

K In sequencing of the NA-2-1 insert, a stop codon was identified in the middle of what was believed to be the protein-coding sequence. Protein sequencing of the polypeptide fragment corresponding to the coding region in question revealed that an extra adenine nucleotide had been inserted in the region 171-182 (Fig. 7) of the clone, most likely the result of a cloning artifact. The extra adenine nucleotide has not been included in the sequence of Figure 7. The amino acid sequence of nucleotides 1-30 of the NA-2-1 insert sequence is identical to the N-terminal sequence of the first 10 amino acids of the mature S protein. With exclusion of the extra adenine nucleotide, the mature protein coding sequence extends to a TGA stop codon (577-579) and encodes a 192 amino acid mature protein. The cDNA insert from NA-2-1 extends 171 nucleotides from this TGA stop codon and includes a poly(A tail 18 residues long. The sequence following the poly(A tail is also believed to be the result of a cloning artifact. The partial signal sequence that is included by the NA-2-1 insert has the typical features described for eukaryotic signal sequences (von Heijne, G. (1983) Eur. J. Biochem. 133:17-21; S 30 von Heijne, G. (1985) J. Mol. Biol. 184:99-105) with a relatively hydrophylic sequence of 5 amino acids at the C-terminal end, an extremely hydrophobic sequence to -14 amino acids) and a more hydrophylic sequence an the Nterminal end.

-18r J The particular clone insert chosen for sequencing (NA-2-1) does not extend in the 5' direction to an ATG initiation codon, so it is not believed to contain the full signal sequence. It would be a matter of ordinary skill in the art, however, to use the 30-mer probe or the NA-2-1 insert as a probe to isolate the full length clone with the complete signal sequence.

A full-length clone was obtained from a second cDNA library which had been prepared using a method (Okayama et al. (1982) Mol. Cell Biol. 2:161-170) which optimizes the recovery of full length clones. This library was screened with the 30-mer probe as well as with the cDNA insert from clone NA-2-1 (described above). A clone designated NA-2-2 was obtained which hybridized to both probes. Figure 9 shows the nucleotide sequence of the cDNA insert from NA-2-2. The NA-2-2 clone insert has been subcloned into M13 phage (M13mp8), designated pAEC9, and has been deposited with the American Type Culture Collection, 12301 Parklawn Drive, Rockville, Maryland 20852 on April 16, 1986, and has been given Accession No. 40233.

The sequence of the cDNA insert of clone NA-2-2 (Figure 9) includes an ATG at its 5' end that is a potential initiation codon and exactly matches the sequence of NA-2-1 (Figure 7) from nucleotide -54 to 682 of the NA-2-2 cDNA.

The sequence contains an open reading frame of 642 bp which encodes a protein with a predicted molecular weight of 24,847 that includes a putative signal sequence of 22 amino acids. The sequence of Figure 9 encodes the mature S protein with a signal sequence that would direct the extracellular transfer of the S2 glycoprotein from the transmitting tract cells. The full-length signal sequence has the typical features described for eukaryotic signal sequences (von Heijne, 1983, and von Heijne, 1985).

Apart from the differences at the 5' end, clones NA-2-1 and NA-2-2 also differ in the length of their 3' untranslated sequence. They are identical to nucleotide 682, which is the polyadenylation site in clone NA-2-2. The clone i insert from NA-2-1 has an additional 50 nucleotides of untranslated mRNA and a polyA tail of 18 residues. This difference at the 3' end suggests that there are alternative polyadenylation sites in S2 RNA transcripts.

Recently; a cDNA clone encoding part (about 30%) of an S-locus specific glycoprotein from B. oleracea has been described (Nashrallah et al., 1985).

19- .Il This clone is reported to represent the 3' end of the coding sequence of the B. oleracea S-protein. There is no apparent homology between this partial sequence and that of the N. alata S -cDNA (Figure 7 and The sequence homology between N. alata and Brassica S-genes indicated by Southern blots may then reside in the as yet undescribed 60% of the Brassica oleracea S-protein clone.

It will be obvious to one of ordinary skill in the art that the DNA sequence information provided herein can be used for the chemical synthesis of oligonucleotide probes that can be used in the hybridization screens described herein. See, for example, Caruthers, M.H. (1984) Contemp. Top. Polym. Sci.

5:55-71; Eisenbeis, S.J. et al. (1985) Proc. Natl. Acad. Sci. USA 82:1084- 1088.

Hybridization of the N. alata S,-protein cDNA clone to poly(A RNA from I mature styles of N. alata, L. peruvianum and Brassica oleracea I A 32 P-labelled copy of the cDNA insert from the NA-2-1 clone, which contains the S-protein coding region, was used in Northern blot hybridization experiments with poly(A RNA prepared from mature styles of N. alata genotypes SS3 S3 SZ S2 and S 3 as well as mature styles of L. peruvianum genotype S 1 3 and green bud styles and ovaries of N. alata genotype SS 3 Figure 8. The size of the major transcript in mature styles bearing the S2allele was 940 bases, based on comparison to 5' end labelled-HindIII-EcoRI markers, with two minor transcripts at 1500 and 3500 bp. The 940 base transcript was also present in RNA from S3S3 and S 1

S

3 styles but at a much reduced frequency, that is 1% or less than the level in S12 orS S S styles.

The major transcript was not present in green bud RNA but was detected in RNA from ovaries of mature flowers, again at a much lower concentration than that

LI

of mature styles (less than Lycopersicon peruvianum genotype SiS contains readily detectable levels of a 2.5 kb mRNA that hybridizes with the NA-2-1 cDNA insert. TheS, and S3 proteins from L. peruvianum both have estimated molecular weights of 28 kd; the RNA blot analysis indicates that the mRNA transcripts encoding these proteins are identical in size. Hybridization with Brassica oleracea mature style mRNA was faint under the conditions used. However, Southern blots of i-LiL-L-~i-LliliiiYW1LC_1-- I i I1 restricted genomic DNA from leaves of Brassica campestris show bands winch strongly hybridize to the NA-2-1 clone insert.

These results indicate homology between the DNA coding sequences of the N. alata S 1 and S 3 proteins and the S protein of N. alata. Further, they indicate that there is homology between the coding sequences of the N. alata S2 protein and those of Lycopersicon peruvianum SI and S3 protein. The weak hybridization of the S2-protein cDNA probe to poly(A RNA from B. oleracea may be due to lower levels of S-protein specific message in whole styles or to more limited homoloyy between the S 2 -gene and S-genes in Brassica.

Isolation of cDNA clones of S-alleles using the S.-allele specific cDNA clone as a hybridization probe The homology disclosed herein can be exploited to isolate DNA encoding ,e 1 other S-proteins, without the need for protein microsequencing and oligonucleotide primers, making the isolation of other S-protein coding DNA ST available to those of ordinary skill in the art.

Total RNA and poly(A RNA can be isolated by conventional methods using appropriate tissue frma a self-incompatible plant of known genotype. Poly(A RNA from mature styles can then be used to create a cDNA library as exemplified herein in the phage vector XgtlO. The protocol of differential screening described above can then be used to obtain mature style specific cDNA clones.

The cDNA NA-2-i clone insert described above can then be used as a hybridization probe, for example, in a colony hybridization screen. The plaques selected as mature style specific cDNA clones can be picked to agarose plates for regrowth and replicated onto nitrocellulose filters. The NA-2-1 877bp insert can be labelled with 3 2 P- by nick translation, end labelling or S ,e random priming. Plaques which hybridized to the labelled NA-2-1 insert can S 30 then be selected for further study. These clones contain cDNA which encodes the desired S-protein although control screening to eliminate false positives is recommended. Standard hybridization'conditions for screens of this type have been described (Maniatis et al., 1982). We have recently used this screening procedure to isolate several cDNA clones from cDNA libraries prepared from L. peruvianum (SIS) stylar mRNA.

-21- ,i Isolation of chromosomal S-genes using the S2-allele specific cDNA clone as a hybridization probe DNA can be isolated from a self-incompatible plant of known genotype by conventional methods as for example those described by Rivin, C. J. et al.

(1982) in Maize for Biological Research F. Sheridan, ed.) pp. 161-164, Plant Mol, Biol. Assn. Charlottesville, Virginia; and Mazure, B. J. and Chui, (1985). A genomic DNA library can then be constructed in an appropriate vector. This involves cleaving the genomic DNA with a restriction endonuclease, size selecting DNA fragments and inserting these fragments into a cloning site of the chosen vector. A description of the construction of a Nicotiana tabacum genomic library in the phage x has been given by Mazure, B.

J. and Chui, (1985).

S'i The 3 2 P-labelled cDNA NA-2-1 clone insert described above can be used as S. a hybridization probe, for example in a filter hybridization screen. An appropriate microorganism is infected with the phage X containing the genomic I 1 library. The infected organisms can be plated on agarose at a concentration St of several thousand plaque forming units/plate and replicated onto nitrocellulose filters. The labelled probe can then be applied to the filter and allowed to hybridize. Plaques that show hybridization to the probe are selected, replated and rehybridized until a pure phage is isolated. DNA from

S

c selected phage can then be purified, restricted, separated on agarose gels and transferred by blotting to nitrocellulose filters. These filters can then be reprobed with the labelled cDNA _2-allele probe to identify those restriction S a fragments that contain S-protein coding sequences. Standard hybridization conditions for such screens have been described (Maniatis et al., 1982).

Synthesis of S-protei in heterologous in vivo expression systems The S-protein DNA coding sequences whose isolation is described herein can be used to direct synthesis of significant amounts of active S-protein.

The DNA encoding the S-protein can be inserted into a recombinant.vect6r so that it is under the control of its own regulatory sequences, an endogenous regulatory region of the vector or an inserted regulatory region by -22conventional recombinant DNA techniques. The choice of recombinant vector is not crucial. A partial list of vectors includes lambda or M13 bacteriophage, Ti or Ri-plasmids of Agrobacterium, pBR322 derived plasmids, and plant viral vectors such as brome mosaic virus (BMV) or tobacco mosaic virus (TMV). An appropriate host microorganism or plant cell is then transformed with the vector containing S-protein coding sequences. Transformed organisms or cells are selected by conventional means and assayed for the expression of active Sprotein, for example as in an in vitro pollen tube inhibition assay or by imunoassay. Transformants which produce active protein can then be grown in liquid medium for an appropriate time to allow synthesis of S-protein which is Ithen isolated and subject to further purification, if .necessary. S-protein sequences can be maintained on the vector or integrated into the chromosomal DNA of the host, where the S-protein sequences will be flanked by DNA sequences of the host.

Yeast expression systems are particularly useful for the expression of plant proteins since correct post-translational processing of plant proteins Shas been observed in such systems. Detailed descriptions of the expression of plant proteins in yeast are given in Rothstein, S.J. et al. (1984) Nature S308:662-665; Langridge, P. et al. (1984) EMBO J. 3:2467-2471; Edens, L. et al.

(1984) Cell 37:629-633; and Cramer, J.A. et al. (1985) Proc. Natl. Acad. Sci.

82:334-338.

Alternatively, plant proteins can be expressed using similar techniques in bacteria as exemplified in Edens, L. et al. (1982) Gene 18:1-12, which described the expression of the plant protein thaumatin.in Escherichia coli.

When a bacterial system is employed, the DNA encoding the S-protein shoula be free of introns, as will be the case with cDNA.

While the presence of a complete signal sequence is not essential to obtain expression of active prntein in either yeast or bacteria, more Sefficient protein synthesis has been observed in yeast when signal sequences are present (Edens, L. et al., 1984).

Example 1: Purification of 32K Sg-protein from Nicotiana alata styles Flowers from N. alata (genotype S2s) were emasculated at the onset of petal coloration. Two days later, the fully mature styles were removed and stored at -70 0 C. (Styles refer to the style and stigma which were removed together; ovary is not included.) Frozen styles (3g) were ground to a fine powder in liquid nitrogen using a mortar and pestle; this was followed by further grinding in 50ml of extracting buffer (50mm Tris-HCl NH 8.5, ImM CaCl 2 20mM NaCl, 1mM DTT, 10mM EDTA and 1% insoluble polyvinylpyrollidone. The homogenate was centrifuged (12,000g; 15 minutes) and the supernatant (11ml) was collected.

Prior to ion exchange chromatography the style extract (11ml) was equilibrated with NH 4

HCO

3 (5mM, pH NaC1 (1mM), CaC1 2 (ImM), EDTA (ImM) by passage through a Sephadex G-25 (Trademark, Pharmacia Inc., Uppsala, 15 Sweden) column (1.6cm diameter; 22cm long, void volume llml). The first 16ml St eluted after the void volume was collected and applied to DEAC-Sepharose ci; (-Trademark, Pharmacia Inc., Uppsala, Sweden) (bed volume 26ml, 1.6cm diameter x 13cm long) which was equilibrated with the same ammonium bicarbonate buffer. The column was then washed with this buffer (50ml) before the application of a NaCl gradient The S 2 -protein was present in the unbound fractions which were combined and concentrated to a final volume of 16ml by rotary evaporation at room temperature. The S2-protein was further purified by affinity chromatography using ConA-Sepharose (Tradneark, Pharmacia Inc. Uppsala, Sweden) followed by gel filtration.

ConA-Sepharose was washed with 5 volumes of methyl-a-D-mannoside (0.1M) in buffer:sodium acetate (10mM, pH 0.1M NaC1, ImM MgC12, 1mM CaC12, ImM MnCI 2 The washed ConA-Sepharose was then transferred to bicarbonate buffer, NaHCO 3 (0.25M, pH 8.8) for 1 hour at room temperature; the bicarbonate buffer was changed 4 times during the 1 hour period. Four volumes of NaHCO 3 (0.25M, pH 8.8) containing 0.03% glutaraldehyde were added and the ConA- Sepharose was then washed with NaHCO 3 (0.1M, pH containing 0.5M NaC1, resuspended in buffer: sodium acetate (10mM, pH 0.1M NaCl, ImM MgCI 2 1mM CaC1 2 ImM MnC12 and packed into a column (0.8cm diameter, 14cm long). The unbound fraction from DEAE-Sepharose was equilibrated in 10mM acetdte buffer, by passing through a G25-Sephadex column equilibrated with 1OmM acetate -24- I I _s

I

buffer, then applied to the column. Unbound material was collected, the column washed with 10 volumes of acetate buffer, and the bound material eluted with 0.1M or 0.2M methyl-a-D-mannoside in acetate buffer. Fractions containing S_-protein were identified by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), collected and concentrated to Iml by rotary evaporation. The use of a lower pH buffer represents an improvement over the method described in Patent Application 28736/84, and results in improved yields of purified _2-protein. The protein appears to be more stable at lower pH, The pooled fraction eluted by 0.1M methyl-a-D-mannoside was applied to a column of Biogel P150 (Trademark, Biorad Laboratories, Richmond, California) to separate the methyl-a-D-mannoside from the S2-protein. (Void volume 13ml, 1.6cm diameter, 36.5cm long equilibrated and run in NH 4

HCO

3 (10mM, EDTA, 0.1M NaCI ImM CaC1 2 A further passage through Biogel P2 (Trademark, Bio'-ad Laboratories, Richmond, California) in water was used to remove any trace of methyl-a-D-mannoside. The purified S2-protein was essentially homogenous by the criteria of SDS-PAGE (Figure 2a).

*z C r Ci L t I t t 't t I II SDS-PAGE was performed according to Laemli, U.K.

Mol. Biol. 80:575-583, using 12.5% acrylamide.

1.43M 2 mercaptoethanol in sample buffer with heating boiling water bath. After electrophoresis, gels were Blue.

and Favre, M. (1973) J Samples were reduced ii for 2 minutes in a stained with Coomassie I It LC C C

I.

IC

LI

C~ t Example 2: N-terminal amino acid sequence of the N. alata S 2 -protein N-terminal sequencing was performed using an Applied Biosystems (Pfungstadt, West Germany) Model 470A gas phase sequencer. Approximately 200 ug of purified S2-glycoprotein was applied in aqueous solution to a glass fibre disc and evaporated to dryness. The disc was placed in the reaction cell of the sequencer, the protein was eluted and then subjected to 20 cycles of automated Edman degradation by phenylisothiocyanate procedure. The resultant amino acid phenylthiohydantoin derivatives were identified by HPLC techniques on an IBM-CN column (IBM, Danbury, Connecticut) at 32 0 C using a sodium acetate-acetonitrile gradient, 20mM sodium acetate (pH 5-5.6) varying from 100%-65% over 30 minutes. The identity of derivates was confirmed by comparison to known standard refernnce compounds.

Example 3: Comparison of the deglycosylated S2_ genotype associated style glycoprotein with the in vitro translation products of style and ovary poly(A RNA Frozen mature styles of Nicotiana alata (S3 y enotype) were ground to a fine powder in liquid nitrogen using a mortar and pestle. Protein was extracted from this tissue and the S-allele associated glycoprotein was isolated by a combination of ion-exchange and affinity chromatography (U.S.

Patent Serial No. 615,079). This material was deglycosylated using a trifluoromethane sulphonic acid (TFMS) procedure modified for use with small quantities of protein (Edge et al. (1981) Annal. Biochem. 118:131-137).

Purified -associated glycoprotein (200pg) was lyophilized in a glass tube with Teflon-lined screw cap and dried over P 2 0 5 in a dessicator for 18 hours. Anisole (60pl) and TFMS (120p1) were added and the tube was flushed with N 2 for 30 seconds and sealed. After 90 minutes at 25 0 C, 10ml of a 1:9 mixture of n-hexane:diethyl ether, precooled on dry ice, was added. The solution was placed on dry ice for 60 minutes, centrifuged (500g, 5 minutes, 4°C) and the supernatant discarded. The pellet was air-dried, resuspended in buffer (300pl) and the pH was adjusted to 6.8 by addition of pyridine:H 2 0 The sample was boiled for 2 minutes before electrophoresis.

Total RNA was isolated from ovary, green bud style or mature style by conventional methods using guanidinium thiocyanate as a protein denaturant.

Oligo(dT)-cellulose chromatography was used to isolate mRNA which is 25 polyadenylated, poly(A RNA. This poly(A RNA (2.0 or 0.5ug) was translated using an amino acid depleted rabbit reticulocyte lysate kit (Amnersham, 1' Arlington Heights, Illinois) in the presence of 150mM K 1.2 mM Mg 2 and tritiated amino acids. Leucine, lysine, phenylalanine, proline and tyrosine were used at specific activities of 5.4, 3.1, 4.8, 3.8 and 4.0 TBq/mmol, respectively. The reaction volume was 25ul. After incubation for 90 minutes at 30°C, RNA was removed by treatment with bovine pancreatic ribonuclease 2mg/ml) for 20 minutes at 370° The glycosylated and deglycusylated samples of pure a-allele protein were analyzed by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) using acrylamide. The gels were stained with Coomassie Blue.

-26- Similarly, the translation products of mature style poly(A RNA were separated by SDS-PAGE using 10-15% acrylamide gradient gels. The products were visualized after treatment of the gel with Amplify (Trademark, Amersham, Arlington Heights, Illinois) and exposure to X-ray film. In both cases, molecular weight markers were included in adjacent lanes and visualized with Coomassie Blue.

Example 4: Preparation of a cDNA library in bacteriophage xgtlO Poly(A RNA was isolated from mature styles of N. alata (genotype S2S) as described above and transcribed into double stranded cDNA (Maniatis et al., 1982). Blunt-ended cDNA was prepared by end repair with DNA polymerase.

EcoRI sites contained in the cDNA were blocked by treatment with EcoRI methylase. Synthetic EcoRI linkers were then ligated to-the double stranded cDNA. The cDNA was then cloned into the EcoRI site of-XgtlO as described by i (Huynh, et al. 1985). This phage Was used to infect Escherichia coli C600 and plated.

i Example 5: Differential screening of mature style cDNA library Poly(A+) RNA was isolated from mature style, green bud style or ovary of N. alata genotype S 3 Single stranded 3 2 p-labelled cDNA hybridization probes were prepared by random priming from the individual RNA. Lambda gtlO containing the mature style library was used to infect E. coli C600 and plated at a density of about 1000 plaque forming units/150mm Petrie plate. Duplicate nitrocellulose lifts were prepared for hybridization (Mahiitis et al., S982). The plaques were first screened with labelled cDNA probe from mature style and green bud style. Plaques that hybridized strongly only to the mature style probe were selected, picked, purified and subjected to a second differential screening using the probes to mature style and ovary. The j resultant plaques represent mature style specific clones.

In these plaque hybridizations, the filters were treated prior to hybridization (prehybridized) for 2 hours and during hybridization for 16 hours at 42 0 C with 5X Denhardt's solution, 5X SSC (3M NaCI, 0.3M Trisocium citrate), 50g/ml sonicated salmon sperm DNA, 50mM sodium phosphate (pH 6.8), ImM sodium pyrophosphate, 100 M ATP and 50% deionized formamide. Probes were -27i p 2 used at a specific activity of 4 x 107 cpm/ml. Filters were washed in a 0.1X SSC solution containing 0.1% SDS (sodium dodecyl sulfate) at 42 0

C.

Example 6: Isolation of the cDNA clones specific for the S 2 -allele associated protein A set of 24 14-mer oligonucleotides was synthesized corresponding to all possible codon ambiguities at anino acids 4-8 in the N-terminal sequence of the S 2 -protein (Figure Oligonucleotides were synthesized by the solidphase phosphoramidite methodology (Beaucage and Caruthers, (1981) Tetrahedron Letters 22:1859) using an Applied Biosystems (Pfungstadt, West Ge'"-wy) ABI Model 380A DNA synthesizer. The 14-mers were end labelled using '4 inase in the presence of 32 P-ATP (5000 Ci/mmol). These labelled 14-mers (5ug/ml) were used in three batches of 8 14-mers to prime selective cDNA synthesis using mature style poly(A RNA. Reverse transcription react-ion volume was 40P1.

The reaction contained 0.75mM of dCTP, dGTP, dTTP and dATP, 75ug/ml poly(A RNA, 50mM Tris-HC1 (pH 10mM KC1, 8mM, MgC12, 0.4mM dithiothreitol, 500 U/ml placental RNAase inhibitor and 500U/ml AMV reverse transcriptase. After incubation at 42 0 C for 90 minutes, the reactions were stoppea by addition of EDTA to 50mM, extracted with phenol:chloroform 1:1 and the product, labelled cDNA, was precipitated with ethanol. The pellets were resuspended in of a solution of 100mM NaOH, 7M urea, and 10mM EDTA. Samples were heated at 90 0 C for 5 minutes before electrophoresis on an 8% acrylamide/7M urea gel. The gel was exposed to X-ray film for 5 minutes, to locate specifically primed cDNA products.

As shown in Figure 5, one of the batches of synthetic 14-mers primed synthesis of a 100 bp nucleotide specific for mature style. This 100 bp nucleotide cDNA band was excised from the gel and eluted overnight with shaking at 37 0 C in 0.5M ammonium acetate and 1mM EDTA. The elutant was concentrated by butanol extraction, phenol:chloroform extracted and ethanol precipitated. The i00 bp nucleotide was then sequenced using the technique of Maxam and Gilbert (1977), Proc. Natl. Acad. Sci. 74:560. The sequence of this nucleotide corresponding to the -12 to +8 amino acid of the protein is shown in Figure 6.

-28- ;1 i A 30 bp-long synthetic oligonucleotide probe based on the sequence of the 100 bp cDNA and covering the region -8 to +1 of the corresponding amino acid sequence was prepared as described above. The 30-mer probe was end-labelled with 3 2 P-ATP. This probe was then used to screen the mature style specific clones obtained by differential screening of the XgtiO library. The hybridization of the 3 2 P-labelled oligomer probe (4 x 107 cpm/ml) was done as described above except that the formamide concentration was decreased to and the temperature was decreased to 370C. Filters were washed using 2 x SSC at 37 0 C. Approximately 100,000 plaques from two separately prepared libraries were screened yielding 5 clones that strongly hybridized with the 30 mer probe. One Xgt1O clone, designated NA-2-1, was selected for further study.

This clone was found to contain a single 877 bp insert which could be excised from the lambda vector by EcoRI digestion. After sequencing of the NA-2-1 clone, it was found that an error had been made in reading the sequencing gel of the lOObp fragment. The sequence shown in Fig. 6 was used to prepare the probe. A comparison of Fig. 6 with Fig. 7 in the corresponding region shows a discrepancy between the two sequences at nucleotide 24 (Fig. The sequence of the 30-mer probe that was used in screening does not therefore exactly correspond to the NA-2-1 clone insert.

Example 7: Nucleotide sequence of NA-2-1 cDNA insert The excised 877 bp DNA insert was sequenced using the chain termination method Sanger et al. (1977) Proc. Natl. Acad. Sci. USA 74:5463-5467; Sanger et al. (1980) J. Mol. Biol. 143:161-178). Figure 7 shows the sequence of this insert containing the full structural coding region for the 2-protein as well as the partial signal sequence. The bases are numbered in the 5' to 3' direction starting with the first base of the structural coding region.

Nucleotides on the 5' side of residue 1 are indicated by negative numbers.

These negatively numbered bases are in the signal sequence. The protein amino acid sequence as deduced from the nucleotide sequence is also shown. The deduced sequence for amino acids 1-10 and 12-15 is identical to the sequence determined by protein microsequencing (Figure As noted above, an extra adenine nucleotide was found to be inserted in the protein coding sequence.

This was confirmed by protein microsequencing of the protein fragment which corresponded to the coding region in question. Several other regions of the nucleotide sequence were confirmed in a similar manner.

-29- 1 Example 8: Northern blot analysis A 3 2 P-labelled probe was prepared from the cDNA clone (NA-2-1) insert encoding the S2-allele associated protein by random priming. Aliquots of poly(A RNA were fractlonated on formaldehyde agarose gels as described by Maniatis, et al. (1982), except that the gel was run in morpholinopropane sulfonic acid (pH 5mM sodium acetate and O.1mM EDTA (pH 8.0) as a buffer. The gel was blotted directly onto nitrocellulose filters using 20X SSC. Klenow labelled-HindIII EcoRI lambda fragments were used as molecular weights markers. Prehybridization and hybridization were carried out at 420 as described for plaque hybridization.

SExample 9: Cloning of the nearly full length S9-protein clone from NA-2-1 I into M13mp8 The 877 bp NA-2-1 clone insert was excised from xgtlO with EcoRI restriction endonuclease. The DNA fragments generated were precipitated with i ethanol, dried in vaccuo and resuspended in water, to 0.25 pg DNA/ul. The DNA fragments (2.5 ug) were then subjected to end repair by incubation at 37 0 C for 1 hour in 25 il buffer containing: 2mM each of dATP, dCTP, dGTP and dTTP, units DNA polymerase I (Klenow fragment), 50mM Tris-HCl (pH 10mM MgC1 2 and 10mM dithiothreitol. The end-repaired fragments were reprecipitated, dried in vaccuo and again suspended in water to 0.25 pg DNA/ll.

The end repaired fragments were inserted into the commercially available vector M13mp8 which had been cut with SmaI restriction endonuclease and dephosphorylated (Amersham, Arlington Heights, Illinois). Blunt-end ligation was done using 1.25 ug of the end repaired fragments and 20 ng of M13mp8 in a buffer containing 1 U/pl T 4 ligase, ImM ATP 66mM Tris-HCl (pH 5mM MgC12 h 30 and 5mM dithiothreitol. The ligation mixture (total volume of 20 il) was incubated overnight at 4 0

C.

The ligation mixture (10 pl) was then used to transform 0.2 ml of competent E. coli JM101 cells (Messing, J. et al. (1981) Nucleic Acids Res.

9:309). Clones containing the 877 bp S2-protein DNA fragment were using the purified 877 bp_ 2-clone insert labeled with 3 2 P by random priming as a Y~ a~ul*Lli~?;~ r" ~Yjilli'i i-i-i- U~ IL' i iii-i 9 Q 0 S0 0 0 eo 00* a 0o a U 90 0 00 f hybridization probe. DNA was purified from one of the selected clones and a DNA molecule designated pAEC5 was isolated which consisted of the 877 bp fragment inserted in the Smal site of M13mp8.

Example 10: Isolation and sequencing of a full-length S2-protein cDNA clone Mature style poly(A RNA was used to prepare a second cDNA library in XgtlO. The library was constructed according to a method described by Okayama et al. (1982) Mol. Cell Biol. 2:161-170, which was designed to optimize isolation of full-length cDNA clones. A library containing 20,000 plaques was obtained from 5ig of poly(A RNA. This library was screened as described in .Example 6 using the 30-bp long synthetic oligonucleotide probe as well as the 877 bp cDNA insert from the NA-2-1 clone of Example 7. One clone, designated NA-2-2, which hybridized to both probes, was selected for further study.

The NA-2-2 cDNA insert was sequenced using the same methods employed to sequence the NA-2-1 insert. Figure 9 shows the sequence of the NA-2-2 cDNA insert which contains the full structural coding region for the mature Sprotein which is identical to that of the NA-2-1 except that there was no 20 extra adenine nucleotide in the NA-2-2 clone sequence. The NA-2-2 clone also encodes the full signal sequence, which extends 22 amino acids on the Nterminal end of the mature protein. The derived amino acid sequence of the signal peptide of both NA-2-1 and NA-2-2 is identical up to amino acid -18.

The reason for the discrepancy in sequence between the two clones is believed to be the result of a cloning artifact during the preparation of NA-2-1. The two clones are different in the length of their 3' untranslated sequence.

They are identical to the polyadenylation site in clone NA-2-2. The NA-2-1 clone contains an extra 50 nucleotides before the poly(A) tail.

Those skilled in the art will appreciate that the invention described 30 herein and the methods of isolation and identification specifically described are susceptible to variations and modifications other than as specifically described. It is to be understood that the invention includes all such variations and modifications which fall within its spirit and scope.

-31- Example 11 Isolation of N. alata S. and S6 cDNA clones cDNA libraries of genotypes S3S3 and S8 were prepared in A gtlO using mRNA from mature styles. Single Sstranded 32 P-labelled cDNA hybridization probes were prepared by random priming from the individual RNA.

Plaque hybridization screens were performed essentially as previously described.

The S3-clones were selected by differential screening of the S3S cDNA library with $3S3 cDNA and S,6 labelled cDNA. Plaques that hybridized strongly to S3S3 cDNA and I weakly to SS cDNA were selected and rescreened with the f' labelled S 2 cDNA clone (NA-2-1 or NA-2-2). Clones which hybridized to the S33 cDNA and the S 2 cDNA clone were then used as probes of northern blots containing RNA from several N. alata S-genotypes. Clones which hybridized most strongly to RNA from styles which carry the Sallele, and weakly to RNA from styles which do not carry the S3-allele were selected as S3 clones. The DNA sequence of one S3 clone selected by this procedure is provided in Table 1.

The S clone selected for sequencing in near fulllength but during subcloning into the pGEM vector for sequencing, a short EcoRI fragment at the 5'-end of the clone was inadvertently deleted. Sequence extending to the indicated EcoRI was determined by RNA sequencing and the N-terminal amino acid sequence was obtained by microsequencing analysis.

32 ~7 S6 cDNA clones were obtained using a similar differential screening procedure. Plaques were initially selected if they hybridized strongly to SS 6 cDNA and poorly to S3S3 cDNA. The DNA sequence of one S6 clone selected by this procedure is provided in Table 2. This clone contained the entire S gene coding sequence, but does not extend in the 5' direction to an ATG codon and so is not full length. Furthermore, the sequenced S clone does not contain a poly tail.

i r c Example 12 Isolation and characterisation of the ttr chromosomal q2 gene t.t Genomic DNA of the N. alata S282 genotype was isolated from leaves. The S2 cDNA clone was radioactively labelled and employed as a hybridization probe of Southern blots of EcoRI digested S 2

S

2 DNA. The S2 gene probe hybridized to a single approximately 3.1 kb EcoRI fragment. This fragment was isolated and cloned in 1 gtlO following ligation of EcoRI digested I gtlO with size fractionated (2.5 kb 4.0 kb) EcoRI. The 3.1 S2 gene fragment was sequenced and the sequence is given in Table 2. The fragment includes an open reading frame extending from nucleotide 1603 to 2338 which is interrupted by a single 94 bp intron (nucleotides 1833 1927). The sequence includes the two polyadenylation signals and T 2 which had been identified in the two S2 cDNA clones. Conventional primer extension techniques were employed to map the starting point of transcription y to a base 19 bp upstream of the ATG start codon.

Sequence analysis identified a putative "TATA" box (nucleotides 1549 1559) in the 5' upstream region of the gene.

33 -ti Table 1: The nucleotide sequence of the Sj cDNA clone.

1

A

signal F E Y M Q L V L Q W P A A TTA CAA TGG CCA GCA GCC 147 F C H T TTT TGT CAC ACC 174 T P S P C K ACT CCT AGT CCT TGC AAA R I P N N AGA ATT CCA AAC AAC Eco RI F T I H TTC ACA ATT CAT 219 N Y C S AAT TAC TGC TGT 264 K K K K AAA AAG AAG AAA 309 A R A D GCC CGA GCT GAT 354 Y N K H TAC AAT AAG CAT 399 Q Y F D CAA TAT TTT GAT 444 L T S L TTG ACA TCT CTC 489 T V Q K ACC GTT CAA AAA 539 Y P N L TAT CCT AAC CTC 579 I G I C ATA GGC ATA TGT 624 P H. P K CCT CAT CCT AAG 669 G L W P D N V S T M L GGG CTT TGG CCG GAT AAC GTG AGC ACA ATG CTT G E D E Y E K L D D D GGC GAA GAT GAG TAC GAA AAA TTA GAT GAT GAT D L D D R W P D L T I GAT CTG GAT GAC CGC TGG CCT GAC TTG ACA ATT C I E H Q V F W K H E TGT ATC GAA CAT CAA GTT TTC TGG AAA CAT GAA G T C C S K S Y N L T GGA ACG TGT TGT TCC AAG AGC TAC AAT CTA ACA L A M A L K D K F D L TTA GCC ATG GCC TTA AAG GAC AAA TTT GAT CTT R K H G I I P G N S Y AGG AAG CAT GGC ATT ATT CCT GGA AAC AGT TAT I N S T I K A I T Q G ATC AAT AGC ACC ATA AAG GCA ATC ACG CAA GGG S C T K R Q M G L L E TCG TGC ACT AAA AGA CAA ATG GAG CTA TTG GAG F D S K V K N V I D C TTC GAC TCG AAG GTA AAA AAT GTG ATA GAT TGT T C K P M G N R G I K ACA TGC AAA CCT ATG GGA AAT AGG GGG ATT AAG

C

I 34 1i.

r L i: i, -i- Table 1 (Continued) F P TTT CCA TGA TTA TAA ATT TCT GTT TCT GTT GCT TTG AGC TGC CTA 714 AAA AAT AAT ACA AAA CTA ATA AGG GAT AAT CAG GAC CAT GGG ACA 759 ATT CTA TTA TGA AAG CCA ACA TTG TGG AAC CAT ATA TAA TTT CCA tr 44 4C TAT AAA TTT ATG AAA T ATT ATT GAA CTG ACA CTT ATT TTG TGT 849 CAA AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA 899 AAA AAA AAA AA 939 1. The isolated S3 cDNA clone is near full length, but part of the 3' end of the clone was removed during subcloning for sequencing due to the presence of an EcoRI site (196 201). The sequence 5' to this site was obtained by RNA sequencing. The N-terminal amino acid sequence was obtained by microsequencing analysis of the isolated S 3 protein.

35 Table 2: Nucleotide sequence of the S 6 -cDNA clone 1 M F N L P L T S V F V I F F A L S P1 I

ATGTTTAACTTACCACTCACGTCAGTTTTCGTCATATTT-TTTTTGCTCTTTCGCCCATTTAT

1 10 20 30 40 50 GCAF E Y M Q L V L Q W PT A F C H T T I' GGGGCTTTCGAATACATGCAACTTCTTTTACAATGGCCAACCGCTTTTTGCCACACTACT signal 70 80 90 100 110 120 P C K N IP S N F T I H G L W P D N V S

CCTTGCAAAAATATTCCAAGCAACTTTACAATCCATGGACTTTGGCCGGATAACGTGAGT

130 140 150 160 170 180 T T L N F C G K E DD Y NIlI MD G P E

ACAACGCTGAATTTCTGTGGTAAAGAAGATGACTATAACATTATAATGGATGGACCCGAG

190 200 210 220 230 240 K N G L Y V IZ W P D L I RE K A D C M K

AAGAATGGTCTGTATGTCCGCTGGCCTG.ACTTGATCAGAGAGAAAGCTGATTCTATGAAA

250 260 27 0 280 290 300 T Q N F W R R E Y I K H G T C C S E I Y

ACGCAAAATTTCTGGAGACGTGAATACATTAAGCATGGAACGTGTTGTTCAGAGATCTAC

310 320 330 340 350 360 N Q V QY F R L AMA L K D K F D L L T

AATCAAGTACAATATTTTCGTTTAGCCATGGCCTTAAAAGACAACTTTGATCTTCTGACT

370 380 390 400 410 420 S L K NH G I I R C Y K Y T V Q K IN N

TCTTTGU.AAAATCATGGAATTATTCGTGGTTACAAATATACCGTTCAGAAAATCAATAAC

'1430 440 450 460 470 480 T I K TV I1KG V P N L S C T iV G Q E L

ACGATCAAGACAGTAACAAAGGGTATCCTAACCTCTCGTGCACTAAAGGGCAAGA.ACTA

490 500 510 520 530 540 W E V G- I C F D S T A K N V I D C P N P

TGGGACGTTGGCATATGTTTCGACTCGACACCGAAAAATGTAATTGATTGTCCTAATCCT

550 560 570 580 590 600 36 Table 2 (cont.) K T C K T A S N G CI M F P*

AAGACATGCAAAACAGCGTCGAATCAGGGAATTATGTTTCCATGAACAAAATTGGCATTT

610 620 630 640 650 660

TTCTTGGTTTAGGCTACGTAAACCAAAATCCAAACCACACGAATAATCAAGAAAATCAAA

670 680 690 700 710 720

CAAAATTTTATTATGAAGATCAAATTGTCAAACCATATGTAAATTTGATAACAAATTTAT

730 740 750 760 770 780

GAAAAGTATTATTGAACTGCG

790 800 1 The S6 cDNA clone does not extend to an ATG codon at the end and does not contain a poly tail. It is believed that the clone is only 2 bases short at the 5' end with the f irst nucleotide of the sequence predicted to be the last base of the ATG start codon. The predicted bases at the 5' end of the sequence are underlined.

1; T t 37

Claims (18)

1. A recombinant vector comprising a DNA sequence encoding an S-protein of a self-incompatible plant wherein the self-incompatible plant displays gametophytic self-incompatibility. i 2. The recombinant vector of claim 1 wherein the self- i incompatible plant is of the family Solonacae.

3. The recombinant vector of claim 2 wherein the self- incompatible plant is of the genus Lycopersicon.

4. The recombinant vector of claim 2 wherein the self- incompatible plant is of the genus Nicotiana. The recombinant vector of claim 1 wherein the self- incompatible plant is Nicotiana alata.

6. The recombinant vector of claim 1 wherein the self- incompatible plant is Lycopersicon peruvianum.

7. The recombinant vector of claim 1 wherein the S- protein is immunologically cross-reactive with antibody to the S 2 -protein of Nicotiana alata.

8. The recombinant vector of claim 1 wherein the S- protein is substantially homologous to the S 2 -protein of Nicotiana alata.

9. The recombinant vector of claim 1 wherein the S- protein is the S2-protein of Nicotiana alata. The recombinant vector of claim 9 wherein the vector is IgtlO. 38 di~a~ c yl

11. The recombinant vector of claim 9 wherein the vector is M13mp8.

12. The recombinant vector of claim 1 wherein the S- protein is the S,-protein of Lycopersicon peruvianum.

13. The recombinant vector of claim 1 wherein the S- protein is the S-protein of Lycopersicon peruvianum.

14. A recombinant DNA molecule comprising a DNA sequence which encodes an S-protein of a self-incompatible plant displaying gametophytic self-incompatibility. The recombinant DNA molecule of claim 14 wherein the S-protein is the S2-protein of Nicotiana alata.

16. The recombinant DNA molecule of claim 15 which further comprises the promoter sequence of the S 2 -allele of Nicotiana alata.

17. The recombinant DNA molecule of claim 14 wherein the S-protein is the S 1 -protein of Lycopersicon peruvianum.

18. The recombinant DNA molecule of claim 14 wherein the S-protein is the S-protein of Lycopersicon peruvianum.

19. A recombinant DNA molecule comprising a DNA sequence which encodes a signal sequence of an S-protein of a self-incompatible plant displaying gametophytic self- incompatibility. The recombinant DNA molecule of claim 19 wherein the signal sequence is that of the S2-protein of Nicotiana alata or an S-protein with an amino acid sequence substantially homologous thereto. 39 i- S21. The recombinant DNA molecule of claim 20 wherein the signal sequence is encoded by DNA comprising the nucleotide sequence 5'-ATG TCT AAA TCA CAG CTA ACG TCA GTT TTC TTC ATT TTG CTT TGT GCT CTT TCA CCG ATT TAT GGG- 3'

22. A recombinant DNA molecule which comprises a Spromoter sequence of an S-allele of a self-incompatible plant displaying gametophytic self-incompatibility.

23. The recombinant DNA molecule of claim 22 wherein the S-allele is the S2- allele of Nicotiana alata. j 24. A method for isolating and identifying an S-gene i comprising the step of screening clones which comprise SDNA from a self-incompatible plant with a hybridization probe, the DNA sequence of which is based on the DNA sequence encoding the S 2 -protein of Nicotiana alata. The method of claim 24 wherein the clones comprise DNA from a genomic DNA library of a self-incompatible Splant.

26. The method of claim 24 wherein the clones comprise j DNA from a cDNA library which is prepared by reverse transcription on a template of mRNA of a self- i incompatible plant. i Dated this 10th day of August, 1990. THE UNIVERSITY OF MELBOURNE By Its Patent Attorneys DAVIES COLLISON S- 40 0-40-