EP1097229A1

EP1097229A1 - Process to collect metabolites from modified nectar by insects

Info

Publication number: EP1097229A1
Application number: EP99935173A
Authority: EP
Inventors: Jantina St. Ctr. voor Plantenveredel. CREEMERS; Gerrit C. St. Ctr. voor Plantenveredel. ANGENENT; Martin M. St. Ctr. voor Plantenveredel. KATER
Original assignee: Stichting Centrum voor Plantenveredelings- en Reprodukteonderzoek (CPRO-DLO)
Current assignee: Plant Research International BV
Priority date: 1998-07-16
Filing date: 1999-07-15
Publication date: 2001-05-09
Also published as: AU5070699A; AU772119B2; NZ509813A; IL140911A0; WO2000004176A1; JP2002520064A; CA2333897A1

Abstract

The invention relates to a recombinant double-stranded DNA molecule comprising an expression cassette comprising the following constituents: i) a promoter functional in nectaries of plants, ii) a DNA sequence encoding a protein which is fused to the promoter, iii) a DNA sequence encoding a signal peptide that targets the recombinant protein to nectar, which is translationally fused to the DNA sequence encoding the recombinant protein, and optionally iv) a signal sequence functional in plants for the transcription termination and polyadenylation of an RNA molecule. The invention further relates to a process for producing a recombinant gene product from honey, comprising: i) producing a transgenic plant by introducing in a plant cell a recombinant double-stranded DNA molecule, regenerating plants from the transgenic cell, and selecting modified plants exhibiting excretion of the recombinant gene product in nectar, ii) allowing insects, preferably bees, to collect nectar from the transgenic plants and to process the nectar into honey, and iii) isolating and purifying the gene product from the honey.

Description

Process to collect metabolites from modified nectar by- insects .

Field of the invention

The present invention relates to isolated, purified DNA sequences which can act as promoters in eukaryotic cells. More specifically, the present invention is related to such DNA sequences which act as promoters to express genes in nectaries of plants. The present invention also relates to chimerical gene constructs comprising a structural or a synthetic gene under the control of a promoter that effects expression of said genes in nectaries. This invention also relates to a process for producing metabolites in honey by allowing insects, preferably bees, to collect and process nectar from plants that excrete said metabolites in nectar or other exudates. Further, this invention relates to plant cells, plants or derivatives therefrom, that express the said chimerical gene.

Background of the invention

Nectaries are nectar secreting organs or tissues that can be located inside (floral) or outside (extrafloral) the flower. The main component of nectar is sugar, the variation between nectars of flowers from different species mainly being the concentration and ratio of glucose, fructose and sucrose (Baker and Baker, 1982) . In addition, depending on the plant species, varying amounts of polysaccharides, lipids, organic acids, volatiles, minerals, phosphates, alkaloids, amino acids and proteins have been detected (Baker and Baker, 1982). Being a specialised sink organ, the nectaries are supplied with sucrose by phloem unloading (Davis et al . , 1985, Hagitzer and Fahn, 1992).

The mechanisms of sugar accumulation and nectar secretion have been described for several plant species (Fahn et al . , 1979) . Sugar transport to the nectaries is achieved by active transport mechanisms and/or osmotic and chemical gradients. In the nectaries of many plants sucrose is converted to glucose and fructose, resulting in a hexose dominant nectar. Part of the hexoses are converted to starch, which is hydrolysed prior to anthesis and nectar secretion. Cell to cell transport of nectar in the nectary parenchyma tissue is mainly symplastic, as demonstrated by the presence of many plasmodesmata between these cells (Fahn et al . , 1979) . Nectar is secreted from secretory cells via the cell membrane (eccrine secretion) or via the Golgi and endoplas atic reticulum vesicles (granulocrine secretion) . Research on the molecular regulation of nectary development and nectary biochemistry has not been reported.

The main function of floral nectar is to reward pollinating insects. Insects collect nectar to meet their short-term energy requirements. Colony-living honeybees process large quantities of nectar into honey, which is stored in honeycombs of the beehive and is used as food supply during the winter period. Within the bee colony different classes of worker bees cooperate in the honey production process. Foraging bees collect pollen and nectar from the flowers and bring it to the hive. On returning to the hive they give most of it up to household bees. Pollen is used as a protein source, especially to feed the brood. Adult nurse and worker bees use little protein, their capacity to digest proteins being very low (Crailsheim et al . , 1993). Honey processing takes place by repeated swallowing and bringing up of the nectar from the honey stomach. In the first process 15% of the water content is lost. This semi- processed nectar is temporarily stored in a honeycomb cell and taken out later for further processing. The final process includes filtering the honey to discard small particles like pollen grains. Sugar metabolising enzymes (invertase, amylase) are added and the honey is concentrated to an average water content of 20%. Most nectars and honeys only contain traces of protein (<0.2%). However, Calluna vulgaris (heather) honey can contain up to 1.8% protein, giving it thixotropic properties (Butler, 1962) . It is known that bees add enzymes like invertase to nectar during the honey processing. Therefore, the probability that proteases are also added is very low. Protein digestion does not take place in the honey stomach but in the intestine of the honeybee. However, the ability of adult worker bees to digest proteins is very low, their main requirement being energy which they obtain from nectar. Until now, it was not established which proteins are present in heather honey and whether these originate from floral heather nectar or are added to honey by honeybees.

In the present invention it was established that heather honey contains two unique proteins that originate from floral nectar of heather. Based on these results a production system for proteins in nectar and honey was estab- lished.

It is an object of the present invention to show that recombinant proteins can be secreted in nectar of transgenic plants, that this nectar is collected by honeybees and that the bees process this nectar into honey that contains the unaltered protein in a concentrated form.

Definitions

Honey: A substance that contains approximately 80% sugar and varying amounts of other components and that is produced by insects, preferably bees, that collect and process nectar from floral or extrafloral nectaries, from honeydew, other plant exudates or artificial sugar solutions.

MADS box gene : a gene coding for a transcription factor having a region of 56 amino acids which is homologous to a similar region in the Arabidopsis AGAMOUS protein and Antirrhinum DEFICIENS protein. This region is called the 'MADS box' . At least 50% of the amino acids in this region should be identical to the amino acid composition in the MADS boxes of AGAMOUS and/or DEFICIENS.

Nectary: secretory organ or secretory tissue of plants, located in the flowers (floral nectaries) or outside the flower (extrafloral nectaries) that excrete nectar.

Nectar: sugar containing fluid that is secreted by nectaries. Nectar can also contain substances like minerals, amino acids, proteins, organic acids, volatiles, alkaloids etc .

Recombinant protein: the gene product of a recombinant DNA molecule .

Recombinant DNA molecule: A DNA molecule in which sequences which are not naturally contiguous have been placed next to each other by in vitro manipulations.

Promoter : The DNA region, usually upstream to the coding sequence of a gene, which binds RNA polymerase and directs the enzyme to the correct transcriptional start site.

Summary of the invention

The production of recombinant proteins for pharmaceutical purposes is a growing market. Until now, mainly bacterial and yeast systems have been used for bulk production of proteins. Recently animal production systems have also been developed. With the availability of efficient transformation techniques for plants, procedures to use plants for the production of proteins are now in progress. In plants, the recombinant proteins are targeted to sink organs like tubers and seeds. A serious draw-back of these production methods is that the recombinant protein can only be obtained after extended, and therefore expensive, purification steps .

The present invention provides a method to produce metabolites, preferably recombinant proteins in honey, which is manufactured by insects, preferably honeybees, that collect floral nectar of transgenic plants. Harvesting of honey is very simple and purification of the protein is very straight forward and requires no advanced purification steps. To give an estimation of the protein yield in a crop like rapeseed, we suggest an average protein production of 2% in honey, as has been found in honey of heather. If one hectare of rapeseed yields 100-500 kilo honey in one season, a yield of 2 to 10 kilo protein can be obtained. In addition, the present invention provides a method to collect metabolites from honey that is derived from non- transgenic plants that secrete these metabolites in nectar. An example are secondary metabolites like acetylandromedol , a diterpine compound, that is excreted in nectar of Rhododendron arboreum and Rhododendron barbatum and of Piptanthus nepalensis (Martini et al . , 1990).

This invention provides a gene from petunia, NECl , that is highly expressed in the nectaries of petunia and weakly expressed in the stamens. It also provides another gene from petunia, FBP15 , that encodes a MADS box protein and which is specifically expressed in the nectaries of petunia. Further, it provides the isolated DNA sequences of the promoters of the NECl and the FBP15 genes. Furthermore, this invention provides an isolated DNA sequence expressed in nectaries encoding a signal peptide that is translation- ally fused to a germin-like protein (Lane et al . , 1993, Dumas et al . , 1995), having the function to target the mature germin-like protein to nectar of heather ( Calluna vulgaris) . This invention gives proof that protein-contai- ning sugar solution is collected by honeybees to produce honey that has a higher protein content than the sugar solution itself, the protein having undergone no qualitative alterations. This invention also proofs that a recombinant protein can be produced in nectar of transgenic plants and that this protein is present in honey produced by honeybees that collected this nectar.

Accordingly, this invention provides an isolated DNA sequence which encodes a protein indicated NECl and having the amino acid sequence given in SEQ ID NO : 1 of the sequence listing hereafter or homologs of NECl. A homolog of NECl is predominantly expressed in nectaries and/or has at least 60% homology with the amino acid sequence given in SEQ ID NO:l. Further this invention provides an isolated DNA sequence which encodes a protein indicated FBP15 and having the amino acid sequence given in SEQ ID NO : 2 of the sequence listing hereafter or a homolog of FBP15. A homolog of FBP15 is specifically expressed in nectaries and belonging to the MADS box family. Furthermore, a homolog is also a gene sequence that has at least 80% homology within the

MADS box region and a 60% overall homology with the amino acid sequence given in SEQ ID NO : 2. Further this invention provides the characterisation and the isolation of a DNA sequence which encodes a signal peptide indicated "CVSP" (Calluna vulgaris signal peptide) , wherein the information contained in the DNA sequence permits, upon translational fusion with a DNA sequence encoding a protein that is expressed in nectaries, targeting of the protein to nectar.

The DNA sequences of the invention can also be characterised in that they comprise the NECl gene and the FBP15 gene having the nucleotide sequences given in SEQ ID NO : 4 and SEQ ID NO : 5 respectively, or a functionally homologous gene or an essentially identical nucleotide sequence or part thereof or derivatives thereof which are derived from said sequences by insertion, deletion or substitution of one or more nucleotides, said derived nucleotide sequences being obtainable by hybridisation with the nucleotide sequences given in SEQ ID NO : 4 and 5 respectively.

Furthermore, the DNA sequences of the invention can also be characterised in that they comprise signal sequence CVSP having the nucleotide sequence given in SEQ ID NO : 6 , or an essentially identical nucleotide sequence or part thereof or derivatives thereof which are derived from said sequences by insertion, deletion or substitution of one or more nucleotides.

Further, this invention provides an isolated DNA sequence from the promoter region upstream of a nectary-specific expressed sequence, which nectary-specific expressed sequence encodes a protein comprising the amino acid sequence given in SEQ ID NO : 1 , or a homologous protein.

Furthermore, this invention provides an isolated DNA sequence from the promoter region upstream of a nectary- specific expressed sequence, which nectary-specific expres- sed sequence encodes a protein comprising the amino acid sequence given in SEQ ID NO : 2 , or a homologous protein.

In a further aspect, the invention provides a protein encoded by any of the above defined DNA sequences. Further, the invention provides processes of producing transgenic plants exhibiting excretion of recombinant proteins in nectar, the expression of the chimerical genes and the targeting of the recombinant proteins being under the control of promoter sequences and a signal sequence as described in this invention. Still further, the invention provides processes of producing transgenic plants that produce recombinant proteins in nectar, the expression of the chimerical genes being under the control of promoter regions upstream of other genes that are expressed in nectaries. Still further, the invention provides processes of producing transgenic plants that produce recombinant proteins in nectar, the expression of these proteins being under the control of any signal peptide that affects targeting of a protein in nectar.

Also, the invention provides recombinant double stranded DNA molecules comprising expression cassettes to be used in the above process. Further, the invention provides transgenic bacteria, transgenic plants producing recombinant proteins in nectar, and also plant cells, tissue culture, plant parts or progeny plants derived from said transgenic plants. Finally, the invention provides a process to produce recombinant gene products in honey, produced by bees that collect nectar from transgenic plants and process this nectar into honey.

Brief description of the figures

Figure 1 shows a polyacrylamide gel with PCR products after Differential Display mRNA amplification. PCR reactions were performed with the oligo-dT primer T12MG in combination with 5 different random primers Apll-AP15 on cDNA samples of pistils without nectaries (two independent samples) , nectaries (two independent samples) , leaves and a mixture of sepals (s) , petals (p) and stamens (a) . Bold arrow depicts the cloned fragment DD18.

Figure 2 is the DNA sequence of the Differential Display RT-PCR clone DD18a. The primers prat 122 and prat 119 that were used for 5' RACE PCR reactions are underlined. Figure 3 is the DNA sequence of clone RC8, obtained by RACE PCR with gene specific primers prat 122 and prat 119 (Fig. 2) in combination with adapter primers. Primer prat 129 (underlined) is used in the next step together with primer prat 122 to amplify the coding region of the NECl cDNA.

Figure 4 is the full length sequence of NECl cDNA. The translation start (ATG) and translation stop (TAA) are depicted bold.

Figure 5 shows the expression of NECl (A) and FBP15 (B) in wild type petunia plants (line W115) as determined by Northern blot analysis. Blot A contains total RNA, while blot B is enriched for mRNA. The tissues are indicated as: 1= leaf, 2= sepal, 3= petal, 4= stamen, 5= pistil, 6= nectary. For blot A the Hindlll/EcoRI fragment of pDD18a was used as a probe. For blot B the full length cDNA of FBP15 was used as a probe .

Figure 6 Expression of NECl by in situ localisation of NECl transcripts (A) and activity of the NECl promoter in the nectaries (B) and the stamen (C) as shown by GUS expression driven by the NECl promoter. The GUS assay used for the stamens was incubated overnight without modificati- ons to prevent diffusion (example 8) . The GUS assay for the nectaries was incubated for 5 hrs, using an assay mixture to prevent diffusion (example 8). For in situ localisation longitudinal sections of flowers of Petunia hybrida were hybridised with digoxigenin- labeled antisense NECl RNA

Figure 7 is the DNA sequence from the promoter region upstream of a sequence encoding the NECl protein. Underlined is the translation start of NECl cDNA.

Figure 8 depicts a schematic presentation of the T-DNA region between the borders of the binary vector pBNEPl, containing the NECl promoter (Figure 7), the GUS reporter gene and the nos terminator in pBINPLUS . This vector was used to generate transgenic plants to study the expression of the NECl promoter.

Figure 9 shows the SDS-PAGE separation of proteins that are present in commercial honey samples from different flowers. M= marker, lane 1: wattle bark, lane 2: flower mixture, lane 3: heather, lane 4: clover, lane 5: rapeseed.

Figure 10 shows the SDS-PAGE separation of proteins that are present in commercial honey samples of rapeseed (RH2x, RHIOx) and heather (HH2x, HHlOx) and of nectar samples of rapeseed (RN2x, RNlOx) and heather (HN2x, HNlOx) . M= molecular weight marker. Two (2x) or ten (lOx) fold diluti- ons were used.

Figure 11 shows the SDS-PAGE separation of proteins present in dilutions of the sugar/BSA feeding solution (A) and of honey from bees that had collected the sugar/BSA solution (B) . The dilutions of the sugar/BSA and honey/BSA solution was the same for both gels: 1= 15x, 2= 30x, 3= 60x, 4= 75x, 5= 75x, 6= 90x, 7= 105x, 8= 120x, 9= 135x. M= marker

Figure 12 shows the sequence homology of the N-terminal protein sequence of CVH29, a unique protein present in heather honey and nectar, with a germin-like protein GER1 from a gene bank homology search (BLAST) .

Figure 13 shows the deduced DNA sequence of the N-terminal protein sequence of CVH29. The degenerated primers prat 176 and prat 177 are underlined (A) . The DNA sequence of the PCR product obtained with prat 176 and prat 177 performed on genomic DNA of heather is shown in B. The gene-specific primers prat 207 and prat 206 used to perform 5 ' RACE PCR reactions on cDNA from heather flowers are underlined. Figure 14 shows the DNA sequence of four independent clones obtained by 5 ' RACE PCR with prat 207 and prat 206 on cDNA of heather flowers. The ATG translation start of the putative signal sequence is boxed. The end of the putative signal sequence and the start of the mature protein are indicated by arrows.

Figure 15 is the sequence of the synthetically produced DNA molecule encoding the signal sequence CVSP (boxed) with linkers.

Figure 16 is the schematic representation of the plasmid pCVl . Not all restriction sites are indicated.

Figure 17 is the schematic representation of the plasmid pCV2. Not all restriction sites are indicated.

Figure 18 is the schematic representation of the plasmid pCV3. Not all restriction sites are indicated.

Figure 19 is the DNA sequence of the full length cDNA of FBP15. The translation start (ATG) and translation stop (TAA) are boxed. The MAD-box and K-box region are underlined.

Detailed description of the invention

This invention provides processes of producing transgenic plants that produce recombinant proteins in nectaries and nectar that is collected by foraging honeybees. This invention gives evidence that honeybees process protein containing nectar into honey that contains the unaltered protein in a concentrated form. Subsequently, the desired protein can be purified from the honey. To express recombinant proteins in nectaries of transgenic plants, a translational fusion of an isolated DNA sequence from a promoter region upstream of a sequence encoding a protein that is expressed in nectaries with a sequence encoding the recombinant protein has to be carried out. Preferably, the isolated DNA sequence from a promoter region is upstream of a sequence that is specifically or highly expressed in nectaries.

The invention relates to a DNA sequence isolated from Petunia hybrida that encodes a protein indicated NECl or a homologous protein or part thereof. A homologous protein has at least 65% homology with the amino acid sequence given in SEQ ID NO : 1. The cDNA sequence of the NECl gene is given in Fig. 4 and in SEQ ID NO : 4. The deduced amino acid sequence of the NECl gene is given in SEQ ID NO : 1. The NECl gene shows strong expression in the nectaries and in a very localised region of the anther filaments of Petunia hybri da . The deduced amino acid sequence of NECl predicts a membrane bound protein. The precise function of the gene has not been elucidated yet, but considering the phenotype of transgenic plants that ectopically express NECl in the leaves, a role in sugar metabolism of NECl is apparent.

The present invention also relates to homologous DNA sequences that can be isolated from other organisms, preferably plants, using standard methods and the already known DNA sequence of the NECl gene. More precisely, it is also possible to use DNA sequences which have a high degree of homology to the DNA sequence of the NECl gene, but which are not completely identical, in the process according to the invention. The use of sequences having homologies between 85 and 100 % is to be preferred. DNA sequences can also be used which result from the sequence shown in SEQ ID NO: 4 by insertion, deletion or substitution of one or more nucleotides. This includes naturally occurring variations or variations introduced through targeted mutagenesis or recombination. The DNA sequence shown in SEQ ID NO : 4 can also be produced by using DNA synthesis techniques.

The invention also relates to a DNA sequence isolated from Petunia hybrida that encodes a MADS box protein indicated FBP15 or a homologous protein or part thereof. The cDNA sequence of FBP15 is given in SEQ ID NO : 5. FBP15 shows exclusively expression in the nectaries of Petunia nyjbrida. The function of FBP15 is unknown. The present invention also relates to homologous DNA sequences that can be isolated from other organisms, preferably plants, using standard methods and the already known DNA sequence of FBP15. More precisely, it is also possible to use DNA sequences which have a high degree of homology to the DNA sequence of FBP15, but which are not completely identical, in the process according to the invention. The use of sequences having homologies between 85 and 100 % is to be preferred. DNA sequences can also be used which result from the sequence shown in SEQ ID NO : 5 by insertion, deletion or substitution of one or more nucleotides. This includes naturally occurring variations or variations introduced through targeted mutagenesis or recombination. The DNA sequence shown in SEQ ID NO : 5 can also be produced by using current DNA synthesis techniques.

Further, this invention provides an isolated DNA sequence from the promoter region upstream of a nectary-specific expressed sequence, which nectary-specific expressed sequence encodes a protein comprising the amino acid sequence given in SEQ ID NO : 1 , or a homologous protein that is expressed in nectaries. Furthermore, this invention provides an isolated DNA sequence from the promoter region upstream of an isolated DNA sequence from the promoter region upstream of a nectary-specific expressed sequence, which nectary specific sequence encodes a protein compri-

SUBST1TUTE SHEET (RULE 26) sing the amino acid sequence given in SEQ ID NO : 2 , or a homologous protein that is expressed in nectaries.

More specifically this invention provides an isolated DNA sequence from the promoter region upstream of a nectary- specific expressed sequence, which nectary-specific expressed sequence has : a) a nucleotide sequence given in SEQ ID NO : 4 , or b) a nucleotide sequence obtainable by hybridisation with the nucleotide sequence of (a) or with a fragment of (a) .

In a more specific embodiment this invention provides an isolated DNA sequence from the promoter region upstream of a nectary-specific expressed sequence, obtained from a plant of Petunia hybrida , the sequence consisting essentially of the sequence given in SEQ ID NO : 7 , or a functional fragment thereof having promoter activity.

In a further aspect, the invention provides an isolated DNA sequence from the promoter region upstream of a nectary- specific expressed sequence, which nectary-specific expressed sequence has : a) a nucleotide sequence given in SEQ ID NO : 5 , or b) a nucleotide sequence obtainable by hybridisation with the nucleotide sequence of (a) or with a fragment of (a) .

In a more specific embodiment this invention provides an isolated DNA sequence from the promoter region upstream of a nectary-specific expressed sequence, obtained from a plant of Petunia hybrida , the sequence consisting essentially of the sequence given in SEQ ID NO : 8 , or a functional fragment thereof having promoter activity.

Further, this invention provides an isolated DNA sequence comprising the coding region for a signal peptide, wherein the information contained in the DNA sequence permits, upon translational fusion with a DNA sequence encoding a protein that is expressed in nectaries, targeting of the protein to nectar. More specifically, the DNA sequence comprises the nucleotide sequence given in SEQ ID NO : 6 obtained from a plant of Calluna vulgaris, or a nucleotide sequence obtai- nable by hybridisation with the nucleotide sequence given in SEQ ID NO : 6. The use of sequences having homologies between 95 and 100 % is to be preferred. DNA sequences can also be used which result from the sequence shown in SEQ ID NO: 6 by insertion, deletion or substitution of one or more nucleotides. This includes naturally occurring variations or variations introduced through targeted mutagenesis or recombination. The DNA sequence shown in SEQ ID NO : 6 can also be produced by using DNA synthesis techniques. The signal peptide CVSP was isolated from nectar of Calluna vulgaris flowers and from honey processed by honeybees that collected the nectar. The function of CVSP in heather nectaries is to target the germin-like protein to nectar. The DNA sequence CVSP can also be used to target other proteins to nectar in plant species.

A subject of the present invention is the use of DNA sequences for producing recombinant proteins in nectar of plants, wherein the protein is produced in nectaries and targeted to nectar, and wherein expression in nectaries is achieved by using a DNA sequence consisting of the promoter region upstream of a DNA sequence that is expressed in nectaries, and wherein secretion in nectar is achieved by using a DNA sequence that encodes a signal sequence that targets the recombinant protein to nectar. In a further aspect the present invention relates to processes wherein a recombinant protein is expressed in other plant tissues than the nectaries and wherein the biochemical composition of nectar is changed as a consequence of the recombinant gene expression. The present invention also relates to processes wherein a recombinant protein is expressed in nectaries of a transgenic plant, wherein the biochemical composition of nectar or the nectar secretion is changed as a consequence of this protein expression. In particular, it relates to processes where the recombinant protein is an enzyme that interferes with the sugar metabolism in nectaries .

The production of a recombinant protein in nectaries and nectar is achieved by integrating into the genome of the plants a recombinant double-stranded DNA molecule comprising an expression cassette having the following constitu- ents and expressing it: i) a promoter functional in nectaries of plants, ii) a DNA sequence encoding a protein which is fused to the promoter, iii) a DNA sequence encoding a signal peptide that targets the recombinant protein to nectar, which is translationally fused to the DNA sequence encoding the recombinant protein, and optionally iv) a signal sequence functional in plants for the transcription termination and polyadenylation of an RNA molecule.

Such DNA molecules are also subject of the invention. The present invention provides an example of such a DNA molecule that contains the described expression cassettes in the form of plasmid pCV3 (Fig. 18), which comprises the promoter region of the NECl gene from petunia, the signal sequence CVSP from heather, the coding region of the reporter gene GUS and the NOS terminator. In principle, any promoter that is active in the nectaries of plants can be used as promoter. The promoter is to ensure that the chosen gene is expressed in nectaries. Also, in principle, any signal sequence that targets the expressed protein to nectar can be used as a signal sequence. The signal sequence is to ensure that the protein is excreted in nectar. Furthermore, any sequence that encodes a recombinant protein in nectaries can be used in the present invention. Preferably, the subject of this invention relates to DNA sequences that encode proteins to be used for pharmaceutical purposes. It is also possible to use the invention to produce proteins for other purposes, e.g. enzymes for biotests or antioxidants for food additives. Furthermore, it is possible to use the invention to produce metabolites in nectar that attract predators of pest insects or that kill or repel pest insects. In another aspect it is possible to use the invention to produce metabolites in nectar that modify the attractiveness of the plant for pollinating insects or improve the health of pollinating insects.

It is also possible to use DNA sequences that encode proteins that modify the nectar composition or the sink strength of nectaries . This means that the recombinant protein interferes with metabolic pathways in the necta- ries, resulting in changed levels of compounds that are already present in nectar, or the formation of new compounds in nectar.

In addition, the present invention also relates to expression cassettes that contain the above mentioned DNA sequen- ces, except for a signal sequence. The recombinant protein is then only expressed in the nectaries, but not targeted to the nectar. Consequently, the expression of the recombinant protein in the nectaries can still affect nectar composition. In a further aspect, the present invention also relates to expression cassettes that contain DNA sequences coding for a protein that is expressed in other tissues than the nectaries. The expression of the recombinant protein affects changes in the biochemical composition of nectar or in nectar secretion.

Finally, the present invention also relates to non-transge- nic plants that produce metabolites in nectar that can be harvested and purified from honey that is produced by honeybees that collect this nectar. Examples for these metabolites are alkaloids, terpines, amino acids, proteins, pigments and volatiles. A preferred embodiment of the process discussed above provides that the expression cassette is transformed to a plant species that produces nectar. Preferably, the recombinant protein is produced in nectar of plants that are visited by honeybees that collect the nectar. Honeybees collect floral as well as extrafloral nectar. The present invention relates to plants that produce recombinant proteins in floral or extrafloral nectar. In addition, the present invention also relates to plants that produce recombinant proteins in other plant organs, said plant organs producing an exudate that is collected by insects, preferably bees, and processed into honey. A particularly preferred embodiment of the present invention are plants that can be grown under controlled conditions. Controlled conditions are greenhouses or field facilities where transgenic plants can be grown according to the safety rules that are required. Preferably, the controlled conditions are such that bee colonies that perform normal foraging behaviour can be maintained in the same compart- ment during the flowering period. Preferred plants originate from the Brassicaceae family, in particular Brassica napus .

Examples

Example 1:

Cloning of NECl

The NECl cDNA was isolated using the mRNA Differential Display system (Gen unter Corporation, Brookline USA) . The isolation of total RNA from nectaries, sepals, petals, stamens and pistils from open flowers and from young leaves of Petunia hybrida was done according to Verwoerd et al .

(1989) . Two independent RNA isolations were performed on nectaries as well as on pistils. A DNase treatment was carried out on each RNA sample, using the RNA MessageClean™

Kit (Genhunter Corporation Brookline USA, cat. No. M601) . A reverse transcription reaction was carried out on 0.1 μg RNA of each sample, using the oligo-dT primer T12MG from the Genhunter Kit. Following the protocol, PCR reactions were carried out using the arbritary primers AP11-AP15 in combination with primer T12MG from the Kit. The PCR products were loaded on a sequencing gel and after electrophoresis the gel was blotted on 3M paper, dried and exposed to X-ray film (Figure 1) . Two adjacent nectary-specific bands were cut out from the blot and the DNA was purified accor- ding to the manual. Reamplification of the fragment was carried out using the oligo-dT primer T12MG and the arbritary primer AP15. After electrophoresis, the PCR product was extracted from the agarose gel by freezing the isolated fragment in liquid nitrogen, followed by centrifugation. DNA was precipitated by adding 1/10 volume 1% HAc, 0. IM MgCl₂ and 2.5 volume of 96% ethanol to the supernatant. The pellet was dissolved in 10 μl TE buffer. The fragment, now called DD18a, was cloned into a PMOSBlue T-vector (RPN 1719, Amersham Little Chalfont UK) giving the vector pDD18a.

The nucleotide sequence of this 3' cDNA clone was determined by the dideoxynucleotide chain termination method (ABI PRISM™ Ready Reaction DyeDeoxy™ Terminator Cycle Sequencing Kit, P/N 402078, Perkin Elmer) and is shown in Figure 2. The DNA fragment has a length of 460 nucleotides. The missing 5' part of the cDNA was isolated using the Marathon TM cDNA Amplification Kit of Clontech (catalog K1802-1) and following the procedure as described in the manual. Briefly, Poly A+ RNA was isolated from nectaries of Petunia hybrida flowers. After double stranded cDNA synthesis, adapters were ligated and a 5 ' RACE reaction was carried out using the adapter primer API supplied in the kit and a gene-specific primer prat 122. The nucleotide sequence of prat 122 is: 5 ' -gtgggaaggctatgctacaagc-3 ' (Figure 2). The PCR product was diluted lOx and 1 μl was used in a second 5' RACE reaction with the nested adapter primer supplied by the kit (AP2) and the nested gene-specific primer prat 119 (Figure 2) . The nucleotide sequence of prat 119 is: 5'- ccttctccatggactgcaatgcg- ' 3 . After gel electrophoreses a fragment of +850 bp was obtained that hybridised with clone DD18a. The fragment, now called RC8 , was extracted from the gel, purified and cloned into a PMOSBlue T-vector as described above. The sequence is shown in Figure 3. The combined (overlapping) sequences of clones DD18a and RC8 are shown in Figure 4, comprising the full length cDNA of a gene called NECl hereafter. The NECl clone has a length of 1205 nucleotides and encodes for a polypeptide of 265 amino acid residues. Based on the deduced amino acid sequence, high homology was found with a cDNA that is associated with Rhizobium- induced nodule development in the legume Medicago trunculata (MtN3, gene bank number: gnl/PID/e274341) . The percentages of identity and similarity are 47% and 72% respectively. Analysis of the predicted protein, using the CAOS/CAMM programme (Protein analysis 1991, Genetics Computer Group inc., Wisconsin USA), shows that the putative protein structure resembles membrane proteins, having six evenly spaced hydrophobic loops that traverse the cell membrane. In addition, a signal sequence is predicted at the N-terminus, while the C-terminus is highly hydrophilic. Highest homology with MtN3 is found in the N-terminal signal sequence, the first two membrane-spanning loops and the last two membrane-spanning loops. The C-terminal hydrophilic part shows the lowest homology (28% identity, 30% similarity) . The function of NECl has not yet been determined.

Example 2 :

Cloning of FBP15

Petunia MADS box cDNA clones were isolated from a cDNA library made from nectaries of Petunia hybrida flowers. The cDNA library was constructed using the lambda ZAP cloning vector (Stratagene, La Jolla USA, catalog nr. 200400- 200402) . The library was screened under low stringency hybridisation conditions with a mixed probe comprising the MADS box regions of Floral binding protein gene FBP2, FBP6 and pMADS3 (Angenent et al . , 1993, 1994, Tsuchimoto 1993). The hybridizing phage plaques were purified using standard techniques. Using the in vivo excision method, E.coli clones which contain a double- stranded Bluescript SK- plasmid with the cDNA insertion between the EcoRI and Xhol cleavage site of he polylinker were generated. Cross- hybridisation of the purified clones revealed 3 independent clones that did not cross hybridise with previously isolated FBP cDNA' s and which were designated FBP15, FBP16 and FBP17. The nucleotide sequence of FBP15 was determined by the dideoxynucleotide-mediated chain termination method and is depicted in SEQ ID NO : 5. The FBP15 cDNA clone has a length of 1157 nucleotides and encodes a peptide of 222 amino acid residues. All characteristics of a MADS box protein are present in FBP15 : a N-terminal located MADS box region which shows a high degree of similarity with other MADS box proteins, and a K-box in the middle of the protein with an alpha helical structure. FBP15 is most similar to the tobacco MADS box protein NAG1, which is an Agamous homolog and expressed in whorl 3 and 4 (Huang et al . , 1996, Mizukami et al . , 1996).

Example 3 : Expression of FBP15

Expression of FBP15 was determined by standard Northern blot hybridisation experiments. A DNA fragment comprising the complete cDNA of FBP15 was used as a probe. High stringency hybridisation and washing conditions were used. Using 10 μg of total RNA from various petunia tissues, expression of FBP15 was only detectable in nectaries. Using 10 μg of mRNA from various tissues, prepared by using the kit and protocol of the Quickprep Micro mRNA Purification Kit (Pharmacia Biotech) , expression of FBP15 was only detectable in nectaries as shown in Figure 5B. The expression in the ovary and nectaries was determined by in situ hybridisation using a DIG labelled antisense RNA probe corresponding to the full length cDNA of FBP15. In vitro antisense RNA transcripts were made using T7 RNA polymerase. A standard protocol for in situ hybridisation was used as described by Canas et al . , 1994. A hybridizing signal was observed evenly strong in all cells of the nectary tissue.

Example 4 : Expression of NEC!

The RΝA expression of NECl was determined by standard Northern blot hybridisation experiments. A DNA fragment comprising the complete sequence of the Differential Display clone DD18 (Figure 2) was used as a probe. Using 10 μg of total RNA from various petunia tissues, strong expression of NECl was detectable in nectaries and weak expression in anthers. No expression was detectable in other floral organs, in leaves or in roots (Figure 5A) . The expression in the ovary and nectaries was determined by in situ hybridisation using a DIG labelled antisense RNA probe corresponding to the nucleotides 79 to 1036 of NECl cDNA, comprising the coding region and part of the 3 ' untranslated region. A clone containing this sequence was obtained by PCR on adapter-ligated cDNA, using two gene- specific primers prat 122 and prat 129 (Figure 4) . The nucleotide sequence of prat 122 is: 5 ' -gtgggaaggctatgctaca- agc-3'", comprising the nucleotides 1015 to 1036 of the NECl cDNA. The nucleotide sequence of prat 129 is: 5'- gqqatccatqqcqcaattacgtqctqatq-3 ' , comprising the nucleotides 79 to 100 of the NECl cDNA. The gene-specific region of the primers is underlined. The primer contains an extra BamHI and Ncol site at the 5 'end. A PCR fragment of 958 nucleotides was obtained and cloned into a PMOSBlue vector. The fragment was subcloned in a vector containing the T7 promoter and in vitro antisense RNA transcripts were made using T7 RNA polymerase. A standard protocol for in situ hybridisation was used as described by Canas et al . , 1994. Strong hybridizing signals were observed in the outer cell layers of the nectaries (Figure 6A)

Example 5 :

Isolation of NECl promoter fragment

The promoter fragment of NECl was cloned using the genome walker protocol (PT3042-1) and kit as provided by Clontech Laboraties. Briefly, genomic DNA from Petunia hybrida was digested with 5 different blunt cutting restriction enzymes. GenomeWalker adapters were ligated and PCR reactions were carried out on each GenomeWalker "library" with a gene specific, reversed primer prat 148 and the adapter primer from the kit (API) . The nucleotide sequence of prat 148 is: 5 ' -ccaagaaggccaaatatgaaagac-3 ' comprising the nucleotides 105 to 128 of the NECl cDNA (Figure 4) . PCR products were subjected to a second round of PCR, using the nested adapter primer AP2 and the nested gene specific, reversed primer prat 149. The nucleotide sequence of prat 149 is: 5' -aagtcatcagcacgtaattgcgcc-3 ' , comprising the nucleotides 81 to 104 of the NECl cDNA. From the second PCR a 2 kb fragment was isolated from the Stul library, which was cloned in the PMOSBlue T-vector, yielding the construct pMA5-10. Figure 7 (SEQ ID NO: 7) shows the DNA sequence of the NECl promoter in the construct pMA5-10, including the translation start of NECl cDNA. Example 6 :

Construction of NECl promoter-GUS

A PCR reaction was performed on pMA5-10 (example 5) , using the forward vector primer U19 of pMOSBlue and the gene- specific primer prat 169. The nucleotide sequence of prat 169 is:

5' -cgctqcaqcqccatqqttttttttaqtgaaqcccc-3 ' . The gene-specific region is underlined. The primer contains an Ncol and Bglll restriction site at the 3' end. The PCR product was digested with Kpnl and Ncol and ligated into a pBluescript- derived vector (pM04) that contains the NTM19 promoter

(Custers et al . , 1997), the reporter gene GUS and the nos terminator. The Kpnl/Ncol NTM19 promoter fragment was replaced, resulting in a NECl-promoter/ GUS translational fusion. The resulting plasmid pΝEPl was digested with Smal to release the NECl promoter/ GUS /nos fragment and this fragment was ligated into a derivative of the binary plasmid pBIΝ (Bevan, 1984) yielding the binary plasmid pBNEPl (Figure 8) . pBNEPl was introduced into Agrobacterium tumefaciens strain LBA4404 or C58pMP90 by electroporation. Plasmid DΝA from the Agrobacterium transformants was isolated and the structure of the binary vector was verified by restriction analysis and PCR.

Example 7 :

Generation of transgenic Petunia plants

Agrobacterium strain LBA4404 transformants were used to transform Petunia hybrida using leaf discs as described by

Horsch et al . (1985). After shoot and root induction on kanamycin selection media, plants were transferred to soil in the greenhouse . Example 8 :

Histochemical GUS assay

Different plant parts of Kanamycin-resistant plants trans- formed with the pBNEPl construct were analysed for the distribution of /3-glucuronidase activity (GUS) using the method described by (Jefferson et al . , 1987) . In transgenic plants with high expression levels diffusion of reaction products to other tissues was observed. To avoid this spreading a modified GUS assay was used. Briefly, tissues were pre-treated with 90% cold acetone at -20°C for 1 h, then rinsed three times 20' with 100 mM phosphate buffer containing 1 mM potassium ferricyanide . After this treatment the standard GUS assay was performed with the modifi- cation that ferricyanide was excluded from the reaction mixture .

Example 9 :

Results histochemical GUS assay

In very young flowers (<1,4 cm) no blue staining was observed, in flowers of 2-4 cm weak blue staining of the nectaries was observed. In flowers of (4-6 cm) strong blue staining was observed in the nectaries (figure 6B) and in a very restricted region of the upper part of the anther filaments (Figure 6C) . GUS expression was highest in the outer cell layers of the nectary parenchyma. In cross sections of the anther filaments GUS expression was observed in all cells except in the xylem of the inner vascular bundle.

Example 10 :

Protein analysis of heather honey and nectar

Samples of pure heather honey, together with samples of rapeseed, clover, wattle bark and lavender honey were diluted, dialyεed and loaded on a 12% SDS page gel (Laemmli, 1970) . All honey samples showed several identical high molecular weight protein bands. Heather honey contained 2 unique protein bands of 29 and 50 kDa (Figure 9) . The proteins were named CVH29 and CVH50 (CVH stands for Calluna vulgaris honey) . To determine the origin of the proteins, nectar and honey samples of rapeseed and heather were prepared and loaded on a 12% SDS page gel. The high molecular weight protein bands of around 70 kDa that are present in all honey samples were not observed in rapeseed or heather nectar (Figure 10) . These proteins are added by honeybees during honey processing. Proteins CVH29 and CVH50 are present in heather honey and heather nectar, but not in nectar of rapeseed. Therefore, it was concluded that CVH29 and CVH50 are secreted in nectar of heather and can be recovered from honey derived from this nectar. The protein concentration in the heather honey we tested was around 0.5%.

Example 11: N-terminal sequence analysis of CVH29 and CVH50

Honey samples were loaded on an SDS PAGE gel and after electrophoreses the gel was blotted on a PVDF membrane. After staining the CVH29 and CVH50 bands were cut out from the blot and N-terminal sequencing was performed on both proteins. The N-terminal sequence of CVH50 is: SVLDFCVADPS- LPDGPAGYSCTEPSTVTSQDF . The N-terminal sequence of CVH29 is: SVLDFCVADPSLPDGPAGYSCKEPAKVTVDDFVFHGLGTA. A gene bank homology search (BLAST) showed high amino acid sequence homology (63%) with germin-like proteins isolated from Arabidopsis (Figure 12). Example 12 :

Identification signal sequence of CVH29

Because the germin-like protein CVH29 is excreted in heather nectar it was expected that part of the cDNA encodes a signal sequence. Based on the N-terminal amino acid sequence, degenerated primers were designed. The sequence of the forward primer prat 176 is: 5'-gayttyt- gygtngcngaycc-3 ' (y= c or t, n= c, t, a or g) . The sequence of the reversed primer prat 177 is: ccrtgraanacraartcrtc (r= g or a) . A PCR reaction performed on genomic DNA of heather yielded a 99 bp DNA fragment. The fragment was sequenced and two reversed, gene-specific 5' primers were designed to clone the 5' cDNA by "Marathon cDNA racing" using the kit and protocol of Clontech laboratories (protocol PT1115-1, Clontech Palo Alto USA). The sequence of gene-specific primer prat 207 that was used is: 5'- ggtgactttagagggctccttgc-3 ' , the sequence of gene-specific nested primer prat 206 is: 5 ' -gctccttgcaggagtagcctgc-3 ' (Figure 13). RNA was isolated from open flowers of heather and mRNA was prepared using the Pharmacia quickprep micro mRNA kit. After cDNA synthesis and adapter ligation a PCR reaction was performed, using the adapter primer API and the gene- specific primer prat 207. The PCR product was used for a second PCR, using adapter primer AP2 and the nested gene-specific primer prat 206. A single fragment of around 300 nucleotides was obtained and cloned in a PMOSBlue T-vector. Four clones were sequenced. Figure 14 shows that three clones were identical and one clone had two different nucleotides in the untranslated 5' region. A putative signal sequence of 17 amino acids was identified between the ATG start codon and the first codon of the mature protein CVH29 that was identical in all four clones. The nucleotide sequence of the putative signal sequence (SEQ ID NO: 6) is:

5 ' -atgtttcttccaattctcttcaccatttccctcctcttctcctcctcccatgct- 3' . Example 13 :

Construction of an expression cassette for excretion of proteins in nectar

To clone the NECl promoter into a PMOSBlue vector a PCR reaction was carried out on pMA5-10 (example 5) using the forward primer prat 247 and the reversed primer prat 248 (Fig. 7) . Prat 247 contains an extra Pstl restriction site. The Ndel restriction site of prat 248 coincides with the ATG translation start of NECl . The nucleotide sequence of prat 247 is: 5 ' -ggctgcaggagtgttctttgatagaatg-3 ' , the nucleotide sequence of prat 248 is: 5'-cgcca- tatgtttttttatggaagcccc-3 ' . Gene-specific regions are underlined. A 1,8 kb promoter fragment was obtained and cloned into a pMOSBlue vector, yielding the plasmid pΝECP .

A DΝA molecule encoding the signal sequence CVSP as depicted in SEQ ID NO: 6 was produced by synthesis and subsequent annealing of two oligo molecules prat 245 and prat 246. The sequence of prat 245 is: 5 ' tatgttccttccaattcttttcactatttct- cttcttttctcttcttctcatgcttctqttcttqatttc 3 , the sequence of prat 246 is: 5 ' gatccgaaatcaagaacagaagcatgagaagaagagaaaagaa- gagaaatagtgaaaagaattggaaggaaca ' 3. The region encoding the signal sequence CVSP is underlined. To ensure correct cleavage of the signal peptide, the linkers were extended with the coding region for the first five amino acids of the mature germin-like protein (Fig. 13). The codon usage of the signal peptide sequence was optimised for Arabidop- sis. By addition of a BamHI restriction site at the 3' end, 2 extra amino acids were linked in frame to the mature protein. The resulting DNA molecule is shown in Figure 15. The fragment was ligated into a Ndel/BamHl cut PMOSBlue vector, yielding the plasmid pCVSP.

pNECP was digested with Ndel and Pstl to release the NECl promoter fragment which was cloned into the Pstl/Ndel cut pCVSP, yielding the plasmid pCVl . A schematic representation of pCVl is given in Figure 16.

A 250 bp long fragment containing the NOS terminator sequence (NOST) was obtained by PCR, using the forward primer prat 251 and the reversed primer prat 252 on DNA of pRAP 33, which is a pUC 19 derived plasmid. Prat 251 adds a Sad and Xhol site, prat 252 adds a Smal and EcoRI site. The sequence of prat 251 is: 5 ' -gggagctcgagtcgttcaaa- catttggcaataaag-3 ' . The sequence of prat 252 is: 5'-cgaatt- cccgqqatctaqtaacataqatqacac-3 ' . The NOST-specific regions are underlined. The PCR product was cloned into pCR-Script™ Amp SK(+) Cloning Kit (Catalog 21188-21190, Stratagene La Jolla USA) , yielding the plasmid pCR-NOST. pCR-NOST was digested with Sacl and EcoRI and the resulting fragment was cloned into the pUC 19 (ClonTech) , derived plasmid pUCAP yielding the plasmid pCVNOS .

The plasmid pGUSN358 was purchased from Clontech (catalog 6030-1) containing the reporter gene GUS in pUC 119, modified to destroy the N-linked glycosylation site within the 1.814 Kb GUS coding sequence. A PCR reaction was carried out with gene-specific primers prat 249 and prat 250, yielding a fragment that contains the GUS gene coding region and a BamHI restriction site at the 5 ' end and a Sacl restriction site at the 3 'end. The sequence of prat 249 is: 5' -ccqqatccatqttacgtcctqtaqaaacc-3 ' . The sequence of prat 250 is: 5' -qqqagctcccaccqaqqctqtaq-3 ' . The GUS specific regions are underlined. Subsequently, the PCR fragment was digested with BamHI and Sacl and ligated into the BamHI/Sa- cl cut plasmid pCVNOS , yielding the plasmid pCV2. A schematic representation of pCV2 is given in Figure 17.

pCVl is digested with Pstl and BamHI and the resulting fragment is cloned into the Pstl/BamHI cut plasmid pCV2 , yielding the plasmid pCV3. A schematic representation of pCV3 is given in Figure 18. pCV3 is digested with Ascl and Smal and the resulting fragment is cloned into a derivative of the binary plasmid pBIN, yielding the binary plasmid pBCV3. pBCV3 was transferred from Escherichia coli to the AgroJacteriu-T? tumefaciens strain LBA4404 and C58pMP90 by electroporation. The transformed Agrobacterium strain was used to transform Arabidopsis and petunia.

Example 14 :

Protein production in nectar

Using the GUS reporter gene, GUS activity in nectar of transgenic plants was measured according to the method as described by Jefferson et al . , (1987). Briefly, the assay was carried out by measuring the amount of methyl umbelli- ferone (MU) produced by GUS fluorometrically by emission of light of 455 nm. The absolute emission was corrected for artificial quenching using an internal standard of InM MU (Angenent et al . , 1993).

Example 15:

Feeding experiments with honeybees

In September 1996 a beehive located outside was supplied with a 25% sucrose solution supplemented with 2% BSA (bovine serum albumin) . After 3 weeks the bees had consumed 15 litters of the feeding solution and honey was harvested from the hive. Although the flowering season had mostly past, bees still foraged on flowers to collect nectar outside. Therefore, the honey produced during this period is derived from a mixture of the feeding solution and nectar from flowers. An SDS page protein gel was loaded with dialysed honey samples and sugar/BSA solutions. Figure 11 shows that the protein band of BSA was present in all the samples tested and no qualitative changes were observed in the honey samples compared to the sugar/BSA solutions. The BSA concentration in honey was 1.5 times higher than in the feeding samples, demonstrating that protein is concen- trated in honey. Honeybees that foraged on the sugar/BSA solution did not show any aberrant behaviour and the colony developed normally.

Example 16 :

Process of honey production from transgenic plants

Twohundred and fifty transgenic plants that each produce recombinant protein in nectar were grown in a greenhouse of 25 square meters. The facilities were adjusted according to the safety rules according to European law, including safety measures to prevent in- or outflow of insects. A beehive adjusted for small populations, containing around 200 worker honeybees and a queen, was placed in the greenhouse when the plants were flowering. When a queen is present, she will start laying eggs and larvae will come out. The presence of brood stimulates the bees to collect nectar and process it into honey. After 2-3 weeks bees processed the nectar into honey and stored in sealed cells of the honeycomb. Under the described conditions the amount of honey that can be harvested is 250-1000 grams.

Example 17 : Ablation of nectaries

By introducing the highly sensitive Rnase BARNASE in plant cells, under the control of a tissue-specific promoter, cell ablation can be achieved in very specific tissues or organs. Ablation of nectaries can be applied to decrease the attractiveness of plants for pest insects that forage on the nectar that is secreted by nectaries. In addition, plants without nectaries can be obtained that are more resistant to bacterial and fungal infections. An example is given for the ablation of nectary tissue by expressing bacterial BARNASE in nectaries, using the NECl promoter. Plasmid DNA of pNEPl (example 6) was digested with Kpnl and Ncol to release the 1800 bp NECl promoter fragment. The purified promoter fragment was ligated into a pWP90 derived vector, upstream of the BARNASE-BARSTAR bacterial operon construct (Hartley, 1988) . The construct contains a 35SCaMV terminator of polyA signal cauliflower mosaic virus terminator sequence downstream of the BARNASE-BARSTAR operon. The resulting plasmid pWP126 was digested with Kpnl/ Xhol to release the NECl-promoter/BARNASE-BARSTAR/CamVpolyA fragment and this fragment was ligated into a pBIN-derived vector pBIN Plus. The recombinant vector was transferred via Agrobacterium tumefaciens (LBA4404) to petunia variety W115. Transgenic petunia plants were selected with flowers without nectaries or underdeveloped nectaries.

Many promoters are less specific than can be concluded based on promoter/GUS expression is concluded. Because the bacterial BARNASE is highly cytotoxic at very low concentrations it can be preferred to protect other plant tissues by expression of a ribonuclease inhibitor gene under the control of a weak, constitutive promoter (e.g. NOS promoter) or a tissue-specific promoter that is not active in the tissues where cell ablation is to be achieved (Mariani et al., 1992, Beals et al . , 1997).

Example 18 :

Ectopic nectary development

MADS box genes regulate floral meristem and floral organ identity. Ectopic expression of MADS box genes can change the developmental fate of floral organs or cells. Transgenic petunia plants ectopically expressing FBP11, an ovule- specific MADS box gene, develop ovule-like structures on sepals and petals ( Colombo et al . , 1995) . FBP15 is a nectary-specific MADS box gene, involved in the molecular regulation of nectary development. In petunia nectaries develop at the base of the carpel. Ectopic expression of FBP15 in petunia may result in the development of nectaries on other organs of the flower or on vegetative parts of the plant. An example is given of a gene construct that, when transformed to a plant, results in ectopic expression of FBP15 .

FBP15 was amplified using a 5' primer that hybridises with FBP15 sequences just upstream of the ATG translation start site and a 3' primer that hybridises with FBP15 sequences just downstream the translation stop site. The 5 'primer contains a Ncol recognition site, the 3 'primer contains a BamHI recognition site. After the sequence was confirmed, the amplified FBP15 fragment was inserted as a BamHI/NcoI fragment into the binary vector pCP031. This binary vector was derived from pPCV708, as described by Florack et al .

(1994), and contains three expression cassettes with a multiple cloning site between the left and right T-DNA borders. The cDNA was cloned in sense orientation between a modified CaMV 35S promoter and the nopaline synthase terminator sequence. The chimerical gene construct was transferred via Agrobacterium GV3101 to petunia variety W115, using the transformation method as described in example 7. Transgenic petunia plants were selected that show ectopic nectary development .

Example 19 :

Modification of sugar composition and nectar secretion

Although sugar content of nectar from different petunia W115 flowers shows some variation, the ratio between hexoses and sucrose is very stable. Down-regulation or up- regulation of genes involved in the establishment of the ratio between hexoses and sucrose in nectar will therefore modify nectar composition. An example is given for anti- sense expression of a petunia-derived invertase gene.

PCR primers were designed that hybridise with the cDNA of an invertase gene cloned from Solanum tuberosum . The 5' primer 5 ' -AAGGACTTTAGAGAGACCCGACCACTGCTGG-3 ' and the 3' primer 5 ' -AAATGTCTTTGATGCATAATATTTCCCATAATC-3 ' were used for a PCR reaction on genomic DNA of petunia to yield a fragment of around 420 bp . The fragment was sequenced and cloned into a pMOSBlue vector to used as a probe to screen a petunia nectary-specific cDNA library. Hybridizing phage plaques were purified and cDNAs were retrieved by in vivo excision as described in example 2. The expression of the cDNA' s was determined by Northern blotting as described in example 3 and the sequence of a nectary-specific invertase was determined as described in example 2. The invertase gene was amplified using a 5' primer that hybridises with sequences just upstream of the ATG translation start site and a 3' primer that hybridises with sequences just downstream of the translation stop site. Extra restriction enzyme recognition sites were generated to allow cloning of the cDNA in sense (overexpression) or antisense direction into the binary vector pCP031 as described in example 18. The chimerical gene constructs are transferred via Agrobacterium GV3101 to petunia variety W115, using the transformation method as described in example 7. Transgenic petunia plants were selected that exhibit modified sugar composition in nectar.

Example 20

Modification of plant development

A DNA which is the NECl gene or a homologous gene is introduced into a plant cell, the said DNA being induced by promoter elements controlling the expression of the introduced DNA in such a way that transcription produces sense RNA. Plants were regenerated from the transgenic cells as described in example 7. Plants that ectopically express the NECl gene exhibited modified leaf morphology and modified sugar composition. Furthermore, plants that ectopically express the NECl gene showed a delay in flowering time.

REFERENCES :

Angenent G.C., Franken J., Busscher M., Colombo L. and van Tunen A.J., 1993. Petal and stamen formation in petunia is regulated by the homeotic gene fbpl . Plant J. 4, 101- 112.

Angenent G.C., Franken J., Busscher M., Weiss D. and van Tunen A.J., 1994. Co-suppression of the petunia homeotic gene fbp2 affects the identity of the generative meri- stem. Plant J. 5, 33-44.

Baker H.G. and Baker I., 1982. Floral nectar constituents in relation to pollinator type. In: Handbook of experimental pollination biology (ED. by CE Jones and RJ Little), pp 117-141. Scientific and Academic Edition, Van Nostrand Reinhold, New York.

Beals T.P. and Goldberg R.B. (1997). A novel cell ablation strategy blocks tobacco anther dehiscence. The Plant Cell 9, 1527-1545.

Bevan M. (1984) . Binary Agrobacterium vectors for plant transformation. Nucl. Acids Res. 12, 8711-8721

Butler C.G. (1962) . The world of the honeybee. London: Collins . Canas L.A., Busscher M., Angenent G.C., Beltran J-P and van Tunen A.J. (1994). Nuclear localisation of the petunia MADS box gene protein FBP1. Plant J. 6, 597-604.

Colombo L., Franken J., Koetje E., van Went J., Dons H.J.M, Angenent G.C. and van Tunen A.J. (1995) . The petunia MADS Box gene FBP11 determines ovule identity. The Plant Cell 7, 1859-1868.

Crailsheim k., Hrassnigg N. , Lorenz W and Lass A. Protein consumption and distribution in a honeybee colony (Apis mellifera carnica Pollm) . Apidologie 24 (5), 510-511.

Custers J.B.M., Oldenhof, Schrauwen J.A.M., Cordewener J.H.G., Wullems G.J. and Lookeren Campagne M.M. (1997). Analysis of microspore-specific promoters in transgenic tobacco. Plant molecular biology 35, 689-699.

Davis A.R., Peterson R.L. and Shuel R.W. (1986). Anatomy and vasculature of the floral nectaries of Brassica napus (Brassicaceae) . Can. J. Bot . vol. 64, 2508-2516.

Dumas B., Freyssinet G. and Pallett E. (1995). Tissue- specific expression of germin-like oxalate oxidase during development and fungal infection of barley seedlings. Plant Physiol. 107: 1091-1096.

Fahn A. (1979) . Ultrastructure of nectaries in relation to nectar secretion. Amer. J. Bot. 66(8), 977-985.

Florack D.E.A, Dirkse W.G., Visser B., F. Heidekamp and Stiekema W.J. (1994). Expression of biologically active hordothionins in tobacco: Effects of pre- and pro- sequences at the amino and carboxyl termini of the hordothionin precursor on mature protein expression and sorting. Plant Mol. Biol. 24. 83-96. Hartley R.W. (1988) . Barnase and Barstar: Expression of its cloned inhibitor permits expression of a cloed ribonu- clease. J. Mol. Biol. 202, 913-915

Horsch R.B., Fry J.E., Hofman N.L., Eichholz D., Rogers S.G. and Fraley R.T. (1985) . A simple and general method for transferring genes into plants. Science 227, 1229- 1231.

Huang H., Tudor M. , Su T., Hu Y. and Ma H. (1996). DNA binding properties of two Arabidopsis MADS domain proteins: Binding consensus and dimer formation. The Plant Cell 8, 81-94.

Jefferson R.A., Kavanagh T.A. and Bevan M. (1987). GUS fusion: J-glucuronidase as a sensitive and versatile gene fusion marker in higher plants. EMBO J. 6, 3901- 3907.

Laemmli U.K. (1970). Cleavage of structural proteins during the assembly of the head bacteriophage T4. nature 227, 680-685.

Lane B.G., Dunwell J.M. , Ray J.A. , Schmitt M.R. and Cuming A.C. (1993). Germin, a protein marker of early plant development, is an oxalate oxidase. The Journal of Biological Chemistry 268 (17), 12239-12242.

Mariani C, Gossele V., De Beuckeleer M. , De Block M. , Goldberg R.B., De Greef W. and Leemans J. (1992). a chimaeric ribonuclease-inhibitor gene restores fertility to male sterile plants. Nature 357, 384-387.

Martini M. , Schmid A. and Hess D. (1990). Antibiotics, and amino acids in nectar of Rhododendron and Piptanthus species from Nepal. Bot. Acta 103, 343-348. Mizukami Y., Huang H., Tudor M. and Ma H. (1996). Functional domains of the floral regulator AGAMOUS: Characterisation of the DNA binding domain and analysis of dominant negative mutations. The Plant Cell 8, 831-845.

Tsuchimoto S., van der Krol A.R. and Chua N.-H, 1993. Ectopic expression of pMADS3 in transgenic petunia phenocopies the petunia blind mutant. Plant Cell 5, 843- 853.

Verwoerd T.C., Dekker, B.M.M. and Hoekema A. (1989). A small-scale procedure for the rapid isolation of plant RNAs. Nucl. Acids Res. 17, 2362.

Zer H. and Fahn A. (1992). Floral nectaries of Rosmarinus officinalis L. Structure, Ultrastructure and Nectar Secretion. Annals of Botany 70, 391-397.

Sequences :

SEQ ID NO:l amino acid sequence NECl

1 MAQLRADDLS FIFGLLGNIV SFMVFLAPVP TFYKIYKRKS SEGYQAIPYM

51 VALFSAGLLL YYAYLRKNAY LIVSINGFGC AIELTYISLF LFYAPRKSKI

101 FTGWLMLLEL GALGMVMPIT YLLAEGSHRV MIVGWICAAI NVAVFAAPLS

151 IMRQVIKTKS VEFMPFTLSL FLTLCATMWF FYGFFKKDFY lAFPNILGFL

201 FGIVQMLLYF VYKDSKRIDD EKSDPVREAT KSKEGVEIII NIEDDNSDNA

2 51 LQSMEKDFSR LRTSK

SEQ ID No: 2 ammo acid sequence ofFBP15

Met Gly Arg Gly Lys lie Glu lie Lys Arg lie Glu Asn Thr Thr Asn 1 5 10 15

Arg Gin Val Thr Phe Cys Lys Arg Arg Asn Gly Leu Leu Lys Lys Ala 20 25 30

Tyr Glu Leu Ser Val Leu Cys Asp Ala Glu Val Ala Leu lie Val Phe 35 40 45

Ser Ser Arg Gly Arg Leu Tyr Glu Tyr Ala Asn Asn Ser Val Lys Ala 50 55 60

Thr lie Asp Arg Tyr Lys Lys Ala Ser Ser Asp Ser Ser Asn Thr Gly 65 70 75 80

Ser Thr Ser Glu Ala Asn Thr Gin Phe Tyr Gin Gin Glu Ala Ala Lys 85 90 95

Leu Arg Val Gin lie Gly Asn Leu Gin Asn Ser Asn Arg Asn Met Leu 100 105 110

Gly Glu Ser Leu Ser Ser Leu Thr Ala Lys Asp Leu Lys Gly Leu Glu 115 120 125

Thr Lys Leu Glu Lys Gly lie Ser Arg lie Arg Ser Lys Lys Asn Glu 130 135 140

Leu Leu Phe Ala Glu lie Glu Tyr Met Arg Lys Arg Glu lie Asp Leu 145 150 155 160

His Asn Asn Asn Gin Met Leu Arg Ala Lys lie Ala Glu Ser Glu Arg 165 170 175

Asn Val Asn Met Met Gly Gly Glu Phe Glu Leu Met Gin Ser His Pro 180 185 190

Tyr Asp Pro Arg Asp Phe Phe Gin Val Asn Gly Leu Gin His Asn His 195 200 205

Gin Tyr Pro Arg Gin Asp Asn Met Ala Leu Gin Leu Val 210 215 220

SEQ ID NO: 3 amino acid sequence CVSP

MFLPILFTISLLFSSSHA

SEQ ID NO: 4 Nucleotide sequence NECl

1 TCGAGCGGCC GCCCGGGCAG GTATTCAACA AGAGTATTCA CCACTTGAAC 51 TCAAAAGGGG CTTCACTAAA AAAAAATCAT GGCGCAATTA CGTGCTGATG 101 ACTTGTCTTT CATATTTGGC CTTCTTGGTA ATATTGTATC ATTCATGGTC 151 TTCCTAGCAC CCGTGCCAAC ATTTTACAAA ATATATAAAA GGAAATCATC

201 AGAAGGATAT CAAGCAATAC CATATATGGT AGCACTGTTC AGCGCCGGAC 251 TATTGCTATA TTATGCTTAT CTCAGGAAGA ATGCCTATCT TATCGTCAGC

301 ATTAATGGCT TTGGATGTGC CATTGAATTA ACATATATCT CTCTGTTTCT

351 CTTTTACGCG CCCAGAAAGT CTAAGATTTT CACAGGGTGG CTGATGCTCT

401 TAGAATTGGG AGCCCTAGGA ATGGTGATGC CAATTACTTA TTTATTAGCA

451 GAAGGCTCAC ATAGAGTGAT GATAGTGGGA TGGATTTGTG CAGCTATCAA 501 TGTTGCTGTC TTTGCTGCTC CTTTAAGCAT CATGAGGCAA GTAATAAAAA

551 CAAAGAGTGT AGAGTTCATG CCCTTCACTT TATCTTTGTT CCTCACTCTC

601 TGTGCCACTA TGTGGTTTTT CTATGGGTTT TTCAAGAAGG ACTTTTACAT

651 TGCGTTTCCA AATATACTGG GCTTTCTATT CGGAATCGTT CAAATGCTAT

701 TATATTTTGT TTACAAGGAT TCAAAGAGAA TAGATGATGA AAAATCTGAT 751 CCTGTTCGAG AAGCTACAAA ATCAAAAGAA GGTGTAGAAA TCATTATCAA

801 CATTGAAGAT GATAATTCTG ATAACGCATT GCAGTCCATG GAGAAGGATT

851 TTTCCAGACT GCGGACATCA AAATAAGCAA GAAGATGATC AAAAAATGAC

901 AAAGCTAAGG AGTTTGAAGT AAGGCAAGGA ACTTGACACT GAATATCTAA

951 GCTAATTAGC AAGACTTTAG CAGCTTGTAA TATTTAGTGT TTGTGAGGTG 1001 TTACCTTATA ATTAGCTTGT AGCATAGCCT TCCCACTAAT AATTCTGCTT

1051 AGCGAATCTT ATATATGGGA AATACTTACA CTAGTATGCA TCTTCTATAT

1101 ACATGTTTGG CACTTGACTA TACATAGAAA AATTAACAAG CATTTCTCAC

1151 CTCAATTTGT CACTTACTTA TAAGTAGCTG AATAATATAA TGCAATTTTC

1201 ACCCC SEQ ID NO: 5 Nucleotide sequence FBP15

1 TCTGAATACAAGCTGTGTGTGTAGAGAGATTTCATAAAGACAGCAAACAT 51 CCCTTCTTTTTGTTCTGTTTTAAAAGTTCCCTTCTTCAACCAGCTCTTTT 101 CCTCATCAGGGTAAGTTGCAAATAAAGGGGATGTTCCAGAATCAAGAAGA 151 GAAGATGTCAGACTCGCCTCAGAGGAAGATGGGAAGAGGAAAGATTGAGA

201 TTAAGAGGATTGAAAATACAACAAATCGTCAAGTCACTTTCTGTAAGAGA

251 AGAAATGGGTTGCTTAAAAAAGCTTATGAACTTTCTGTTCTTTGTGATGC 301 TGAAGTTGCTCTCATCGTTTTCTCAAGCCGTGGCCGCCTCTATGAATATG

351 CTAACAACAGTGTGAAGGCAACAATTGATAGATATAAGAAAGCATCCTCA

401 GATTCCTCCAACACTGGATCTACTTCTGAAGCTAACACTCAGTTTTATCA

451 ACAAGAAGCTGCCAAACTCCGAGTTCAGATTGGTAACTTACAGAACTCAA

501 ACAGGAACATGCTAGGCGAGTCTCTAAGTTCTCTGACTGCAAAAGATCTG 551 AAAGGCCTGGAGACCAAACTTGAGAAAGGAATTAGTAGAATTAGGTCCAA

601 AAAGAATGAACTCCTGTTTGCTGAGATTGAGTATATGCGAAAAAGGGAAA

651 TTGATTTGCACAACAACAATCAGATGCTTCGGGCAAAGATAGCTGAGAGT

701 GAAAGAAATGTGAACATGATGGGAGGAGAATTTGAGCTGATGCAATCTCA

751 TCCGTACGATCCAAGAGACTTCTTCCAAGTGAACGGCTTACAGCATAATC 801 ATCAATATCCACGCCAAGACAACATGGCTCTTCAATTAGTATAAGTTTAT

851 AATAAAATGCATGGTTTGAAGCACTCTGATTGTGGTGGATTTGGATTATG

901 TATAAGGGAGTGCAGGCCATTTGCCAATTATTGAAAGGTACTCAAACAGG

951 AAGTTGAAGAAGTTCATCATCTCTCTCATCTATATGTCTTAACAAAAGTC

1001 TTAGCTTATGGACTCTAAAACAAAGACTTAATTTAACATATAAATATAAT 1051 TGTGTAATGCTGTTGTATTGTATGGTATGTATCCAAAAACATTAATAACC

1101 TATCTTTTTCTTCAAATTATGTCTCCTTTGATACAAACTACTAACATATT

1151 TTCTTAT

SEQ ID NO : 6 Nucleotide sequence CVSP

ATGTTTCTTCCAATTCTCTTCACCATTTCCCTCCTCTTCTCCTCCTCCCATGCT

SEQ ID NO: 7 Nucleotide sequence NECl promoter

1 CCTAGGAGAAATCAAGCCTACTCTTAAGATGGATGACTCACTTGCCCCGA 51 TGGTAAGGTGAAGGATCTGTTGATTAGAGTTGGGAAGTTCATGTTCTCTG 101 CTGATTTTATTATTCTAGACTATGAAGAGGACCAAGAAGCTCCAATAATT 151 TTGGGAAGAGCATTCTTAATCACATCGATGGCAATTATTGACATGGAACT

201 TGGGGAGATGACTGTGAGAGCGCATGGAGAAAAGGTTACTTTCAAGGTTT

251 ATAATAAAAAGGATCATATGGCTAAGTTTGAAGAGTGTTCTTTGATAGAA 301 TGTGTCAGACGAGAACATGAAAGTAAACCGAAAGAGGTGTTTGAGCGGAA

351 TGTAGAACAAAGTGACCACGGCACAATAATTGACAAGTTGAAGGAAAATT

401 CACCTAAAGGAAGGAAGAAGACAAAAGTTCGTCGTAACAAGAGGAGACGT

451 AAATGCTGGAAGTGAGCTTAAAGGTGTTGTCGTACTACGACGTTAACTAA

501 GGCGCTTGTCGGGAGGCAACCCTAGCTTTGTATGTAAATGTAAAAGTAAA 551 AAATATATATATAGAAAAAGGAAAATACAAAAAGAGTCGTGCCGCGACGT

601 TAAATCAAGCGCTTGTTGGAAGGCAACCCAATTTTTATTGTTTTAGTTGT

651 TTTACTTATTTAGTATTACGTAGTTTCTTGTTGTTTTTGTAGGGCTCGGG

701 ACTTTCGGAAGGTGAGGTAATTTCAAGGCATCGCGGTGTGTATTGCAGCG

751 AGGTAAGTGTAAGAGTTGAGTTGGAAGCGTTTGGCCAAGTGTTGCACCGT 801 GAGAGGCTTTCAACCTGTTGCGACACGTGAAAAATTAAGAGCCAGATCTG

851 CTACATTAGCACTGAAGCATCGCTTGGCCAATAGCTTGGAATGGAAGCAA

901 GAATTCAAACCAAAATCAGAAACGCCACAAGAGATGTGTCGCACACTGCA

951 AAGCTTTGTGCAAACTAGTGAACGCAGAAATAGAAATGCTACAGCCCATG

1001 CGTCGCTTGGCTTATGGCAGGCAGCAAAAATTCAGCAGCAAAACAGAAAC 1051 GCTGCGAGAAACGCGTCGCATACGCCATAGCTTTGTGTCAAACAGAACGT

1101 CCAGAAATTGAAAAGCTATAAGCCTGCGTCGCTTGGCTCATGGCGTGCAG

1151 ACTAGAAAAGCTCTAGCAGATGCGTCGCGTATTGTATAGCTTGGTGTGAA

1201 ACAGAAAGTTCGAAACTTGGAAAACGATAACCCAGCGTCGCCTCTTCAAC

1251 CGCGTCCAGGTAAGTTCAAGATTCTTACGGGTTGACCCATTAACCCATTG 1301 ATCGGCTGATTATAAACAATAAAACATCACCTTCAACTATCACATGATTT

1351 CATAAGTTTGACCTAGGATATTTTATATATATATATATATATATACACAC

1401 ACACACCATTTCC GCGATCTTACCTCATTTTTATTCAAACCATTTTTCT 1451 GCTTCAAAAGTTTAAATTATTAATATGATAAGTCATCCATAGTCAAACAA

1501 GATTTTCTATACTATTTTGTCCCTTGTAATTTTAAAAAAAAAATGAGCGA

1551 TGGTAAGATAAACATTGTTTGCAAGTGTACAATTTTAGTATATGCAAACC

1601 AACGCTTCTTCTTCCAACTATCACCTAAAACTACATCATTTATGGCGGGC

1651 GGACTAGACGTAGCCAAATATAAAAACGCAATGGCCATTCAGTTCATGTC

1701 ATTTTTATATCCTTCATCCAATAATATTACTCAAAATTGATGTACAGTTT

1751 GGTCTCTGATGTGCACTTTACTATACGTAATACGGAATTTACATTATAAT 1801 TAAAGAGAACTGTTCCACTAAATTTTAATGATTTAATTAATTTAACTCGG

1851 TTACTTGTATTATTATTATTGCTGTATTTGTTTGTCATTTGAATTTGGCA

1901 CCGCAGATTTTTGTATGCAATTAACCCTCATATATCTTTTGGCCAAATAA

1951 AGAAAAAGTCTGCATATTTCTTGCCAAACATTTATCATACTTTACCGAAT

2001 TCTTGTTTTTTGTTTCTCTGTTGTTGTTCTCCACTATAAATAACATTTGC 2051 AGTGAGTAAAGTTTCTTCAGGTCTCTTTTGTAGATTCAACAAGAGTATTC

2101 AGCACTTGAACTCAAAAGGGGCTTCACTAAAAAAAATCATG

SEQ ID NO: 8 Nucleotide sequence FBP15 promoter

Claims

1. An isolated DNA sequence from the promoter region upstream of a nectary-specific expressed sequence, which nectary-specific expressed sequence encodes a protein comprising the amino acid sequence given in SEQ ID NO : 1, or a homologous protein.

2. An isolated DNA sequence according to claim 1, wherein the nectary-specific expressed sequence has: a) a nucleotide sequence given in SEQ ID NO : 4 , or b) a nucleotide sequence obtainable by hybridisation with the nucleotide sequence of (a) or with a fragment of (a) .

3. An isolated DNA sequence according to claim 1 or 2 , obtained from a plant of Petunia hybrida , the sequence consisting essentially of the sequence given in SEQ ID NO: 7, or a functional fragment thereof having promoter activity.

4. An isolated DNA sequence from the promoter region upstream of a nectary-specific expressed sequence, which nectary-specific expressed sequence encodes a protein comprising the amino acid sequence given in SEQ ID NO : 2 , or a homologous protein.

5. An isolated DNA sequence according to claim 4, wherein the nectary-specific expressed sequence has: a) a nucleotide sequence given in SEQ ID NO: 5, or b) a nucleotide sequence obtainable by hybridisation with the nucleotide sequence of (a) or with a fragment of (a) .

6. An isolated DNA sequence according to claim 4 or 5 , obtained from a plant of Petunia hybrida , the sequence consisting essentially of the sequence given in SEQ ID NO : 8 or a functional fragment thereof having promoter activity.

7. An isolated DNA sequence encoding a protein comprising the amino acid sequence given in SEQ ID NO:l, or a homologous protein, the expression of the DNA sequence being predominantly confined to the nectaries of a plant.

8. An isolated DNA sequence according to claim 7 having: a) a nucleotide sequence given in SEQ ID NO: 4, or b) a nucleotide sequence obtainable by hybridisation with the nucleotide sequence of (a) or with a fragment of (a) .

9. An isolated DNA sequence that results from the sequence shown in SEQ ID NO : 4 by insertion, deletion or substitution of one or more nucleotides, including naturally occurring variations or variations introduced by targeted mutagenesis or recombination.

10. An isolated DNA sequence according to claim 7 having a nucleotide sequence given in SEQ ID NO : 4 , said sequence being produced by current DNA synthesis techniques.

11. An isolated DNA sequence encoding a protein comprising the amino acid sequence given in SEQ ID NO : 2 , or a homologous protein, the expression of the DNA sequence being predominantly confined to the nectaries of a plant.

12. An isolated DNA sequence according to claim 11, having: a) a nucleotide sequence given in SEQ ID NO: 5, or b) a nucleotide sequence obtainable by hybridisation with the nucleotide sequence of (a) or with a fragment of (a) .

13. An isolated DNA sequence that results from the sequence shown in SEQ ID NO : 5 by insertion, deletion or substitution of one or more nucleotides, including naturally occurring variations or variations introduced by targeted mutagenesis or recombination.

14. An isolated DNA sequence according to claim 11 having a nucleotide sequence given in SEQ ID NO : 5 , said sequence being produced by current DNA synthesis techniques.

15. Any sequence that encodes a nectary-specific MADS box gene .

16. An isolated DNA sequence comprising the coding region for a signal peptide, wherein the information contained in the DNA sequence permits, upon translational fusion with a DNA sequence encoding a protein that is expressed in nectaries, targeting of the protein to nectar.

17. An isolated DNA sequence according to claim 16, compri- sing the nucleotide sequence given in SEQ ID NO: 6 obtained from a plant of Calluna vulgaris , or a nucleotide sequence obtainable by hybridisation with the nucleotide sequence given in SEQ ID NO : 6.

18. A recombinant double-stranded DNA molecule comprising an expression cassette comprising the following constituents: i) a promoter functional in plants, ii) a DNA sequence coding for a protein as defined in any of claims 7 to 15 which is fused to the promoter sequence in sense or antisense orientation, and optionally iii) a signal sequence functional in plants for the transcription termination and polyadenylation of an RNA molecule.

19. A recombinant double-stranded DNA molecule comprising an expression cassette comprising the following constituents: i) a promoter functional in nectaries of plants, ii) a DNA sequence coding for a protein which is fused to the promoter sequence in sense or antisense orientation, and optionally iii) a signal sequence functional in plants for the transcription termination and polyadenylation of an RNA molecule .

20. A recombinant double-stranded DNA molecule comprising an expression cassette comprising the following constituents: i) a promoter functional in nectaries of plants, ii) a DNA sequence encoding a protein which is fused to the promoter, iii) a DNA sequence encoding a signal peptide that targets the recombinant protein to nectar, which is translati- onally fused to the DNA sequence encoding the recombinant protein, and optionally iv) a signal sequence functional in plants for the trans- cription termination and polyadenylation of an RNA molecule .

21. A recombinant double-stranded DNA molecule according to claim 19 or 20 wherein the promoter is as defined in any of claims 1-6.

22. A recombinant double-stranded DNA molecule according to claim 20 or 21 wherein the DNA sequence encoding a signal peptide is as defined in claim 16 or 17.

23. A process for producing a transgenic plant exhibiting excretion of a recombinant protein in its nectar, comprising: i) introducing in a plant cell a recombinant double-stran- ded DNA-molecule as defined in any of claims 20 to 22, wherein the recombinant protein is excreted in nectar ii) regenerating plants from the transgenic cell, and iii) selecting transgenic plants.

24. A process for producing a transgenic plant exhibiting a modified nectar composition, comprising: i) introducing in a plant cell a recombinant double-stranded DNA-molecule as defined in any of claims 19 to 22, wherein the recombinant protein interferes with metabolic pathways in the nectaries, ii) regenerating plants from the transgenic cell, and iii) selecting transgenic plants.

25. A process for producing a transgenic plant exhibiting a modified nectar secretion, comprising: i) introducing in a plant cell a recombinant double-stranded DNA-molecule as defined in any of claims 19 to 22, wherein the recombinant protein interferes with sink strength of nectaries ii) regenerating plants from the transgenic cell, and iii) selecting transgenic plants.

26. A process for producing a transgenic plant exhibiting a modified nectary development, comprising: i) introducing in a plant cell a recombinant double-stranded DNA-molecule as defined in claims 19 or 22, wherein the recombinant protein interferes with the development of nectaries ii) regenerating plants from the transgenic cell, and iii) selecting transgenic plants.

27. A process for producing honey from modified nectar of transgenic plants, comprising: i) producing a transgenic plant by introducing in a plant cell a recombinant double-stranded DNA molecule as defined in any of claims 19 to 22, regenerating plants from the transgenic cell, and selecting modified plants exhibiting the excretion of nectar with a modified composition, ii) allowing insects, preferably bees, to collect nectar from the transgenic plants and to process the nectar into honey.

28. A process for producing a recombinant gene product from honey, comprising: i) producing a transgenic plant by introducing in a plant cell a recombinant double-stranded DNA molecule as defined in any of claims 20 to 22, regenerating plants from the transgenic cell, and selecting modified plants. exhibiting excretion of the recombinant gene product in nectar, ii) allowing insects, preferably bees, to collect nectar from the transgenic plants and to process the nectar into honey, and iii) isolating and purifying the gene product from the honey.

29. A process for producing a metabolite from honey, comprising: i) producing a plant that excretes this metabolite in nectar and which plant has been produced by current breeding and selection methods, ii) allowing insects, preferably bees, to collect nectar from the selected plants and to process the nectar into honey, and iii) isolating and purifying the metabolite from the honey.

30. Micro organisms containing DNA sequences according to one or more of claims 1 to 17.

31. Micro organisms containing recombinant DNA molecules according to any of claims 18 to 22.

32. A plant cell or plant cell culture transformed with one or more DNA sequences according to claims 1 to 17.

33. A plant cell or plant cell culture transformed with recombinant DNA molecules according to any of 18 to 22.

34. A plant consisting essentially of the plant cells of claims 32 or 33.

35. A transgenic plant obtained by the process of any of claims 23 to 26.

36. Seeds, tissue culture, plant parts or progeny plants derived from a transgenic plant according to claim 35.

37. Honey obtained from nectar from transgenic plants, which nectar has a modified composition.

38. Honey obtained from nectar from transgenic plants, which nectar comprises a recombinant gene product.