Field of the invention The invention relates to the achievement of an increased content of Vitamin C in plants or products derivable therefrom, by methods of genetic modification.

Background of the invention Vitamin C (L-ascorbic acid) is essential to man, however an evolutionary modification to the L-gulono-1,4-lactone oxidase gene prevents humans and other primates from synthesising this vitamin, thereby necessitating its acquisition from dietary sources. Plant-derived vitamin C is the major source of vitamin C in humans.

D-arabinono-1,4-lactone oxidase is an enzyme which has been isolated from the yeast. (Huh W. K. et al., Eur. Journal Of Biochemistry 1994 225 1073-1079) In its native cellular environment this enzyme is known to catalyse the oxidation of D-arabinono-1,4-lactone to D-erythroascorbic acid, an analogue of vitamin C. In doing so this completes the final step in this biosynthetic pathway (Huh W. K. et al., Biochimica et Biophysica Acta 1996 1297 1-8).

The identification of analogous enzymes has given rise to some confusion as to the specific identity of the yeast D-

arabinono-1,4-lactone oxidase and this has been documented in the prior art.

The present consensus and experimental evidence supports the understanding that different analogues can be attributed to yeast, plant and animal sources and that the yeast derived enzyme cited in the prior art as L-galactono-1,4-lactone oxidase is synonymous with the yeast enzyme D-arabinono-1,4- lactone oxidase (Huh W. K. et al., European Journal of Biochemistry 1994 225 1073-1079).

The role of the plant enzyme L-galactono-1,4-lactone dehydrogenase in the vitamin C pathway of plants has been investigated in two separate studies using systems based on tissue derived from maize and potato plants. (de Gara L. et al., Journal of Plant Physiology 1994 144 649-653 and Viola R et al., Journal of Plant Physiology 1998 152 58-63).

Herein plant material was supplied with an excess of the plant substrate, L-galactono-1,4-lactone and the production of vitamin C monitored. A resulting increase in the amount of vitamin C synthesis was demonstrated by these experiments thereby providing a clear indication that this final step in vitamin C synthesis would not be expected to be rate limiting.

The implications of these two experimental studies are significant in that such data would strongly discourage the skilled person in the art from exploring the effects of overexpressing this plant gene or analogous yeast and animal genes that encode polypeptides that catalyse this step in the reaction pathway. It would appear from this research in

plant systems and by analogy in yeast and animal systems that the expression, under normal cellular conditions, of these enzymes is at a level that does not pose any significant limitation to the production of vitamin C or its yeast and animal cell analogues.

The cloning, sequencing and expression of the plant enzyme L-galactono-1,4-lactone dehydrogenase has been described WO 98/50558. This document also describes the overexpression of L-galactono-1,4-lactone dehydrogenase in plant systems wherein a small enhancement in vitamin C synthesis of 34% greater than non-modified plants has been shown. Such a small increase supports the understanding that this enzyme does not catalyse a significantly rate determining step as similar changes have been seen by simple variation in propagation conditions, such as light, iron deficiency and salt stress (Badiani M. et al., Agrochimica 1994 38 25-36, Iturbe-Ormaetxe I. et al., Plant Cell and Environment 1995 18 421-429, Lechno S. et al., Journal of Plant Physiology 1997 150 206-211.).

D-arabinono-1,4-lactone oxidase has been shown to have a low substrate specificity by way of its ability to take the plant substrate L-galactono-1,4-lactone and catalyse its conversion to vitamin C. (Huh W. K. et al., Eur. Journal Of Biochemistry 1994 225 1073-1079). However the enzyme activity of D-arabinono-1,4-lactone oxidase is that of oxidation simply requiring oxygen and thereby differing significantly from the dehydrogenase activity of galactono- 1,4-lactone dehydrogenase which is co-factor reliant,

requiring probably cytochrome C reduction in order to catalyse this final step in plant vitamin C synthesis.

Vitamin C primarily functions as an antioxidant and itself provides an important cofactor for a number of enzymatic reactions e. g. in the production of collagen in vertebrates.

Abnormal collagen formed in the absence of vitamin C cannot properly form fibres and this causes the skin lesions and blood-vessel fragility that are prominent symptoms of scurvy. It would be clearly beneficial to be able to increase the vitamin C content of plants which are intended for human consumption.

Vitamin C also carries out an important role in plants systems, where its antioxidant capacity reduces the impact of conditions of stress such as excessive heat, cold or drought. Stress resistance in plants is of great agricultural significance and it would be desirable if this could be improved by increasing cellular expression of vitamin C.

There is a clear need to establish a means by which the vitamin C content of plants can be significantly increased.

In particular there is a need to establish such a means for application to those plants which are intended for human consumption to improve their nutritional value with respect to vitamin C.

SUMMARY OF THE INVENTION The present invention provides a way to genetically modify plants to significantly increase the content of vitamin C relative to non-modified plants.

A process for transforming a plant by inserting a gene that encodes an enzyme capable of catalysing the conversion of the plant substrate L-galactono-1,4-lactone to vitamin C, wherein said gene is derivable from a non-plant source, can be used to achieve significantly higher increases in vitamin C than are possible by conventional means.

Therefore the present invention comprises a process for producing plants or plant tissues having an elevated content of vitamin C comprising the steps, (i) transformation of a plant cell with a gene construct, followed by (ii) the regeneration of a genetically modified plant or plant tissue from the transformed plant cell of the transformation step (i), wherein the gene construct comprises a polynucleotide sequence capable of expressing a polypeptide product with an ability to catalyse conversion of L-galactono-1,4-lactone to vitamin C, characterised in that said polynucleotide sequence is derived from a non-plant source.

This and other aspects of the invention will be described in further detail in the description that will follow hereinafter.

DETAILED DESCRIPTION OF THE INVENTION.

Figure 1 shows the gene sequence of D-arabinono-1,4-lactone oxidase of Saccharomyces cerevisiae (sequence ID No. 1).

Figure 2 shows the amino acid sequence for D-arabinono-1,4- lactone oxidase of Saccharomyces cerevisie (sequence ID No. 2).

For the purpose of this invention an unmodified control plant or such plant tissues are taken to be that/those of an equivalent plant of identical type which has not undergone any transformation and is therefore in a native state.

It should be appreciated that for the present invention any polynucleotide sequence encoding a polypeptide that has an ability to convert the plant substrate L-galactono-1,4- lactone to vitamin C may be used in the above process however it is preferable that said polynucleotide sequence is derivable from yeast or animal cells.

Accordingly an aspect of the invention comprises a process for increasing content of vitamin C in plants or plant tissues as described above, wherein the non-plant source for the polynucleotide sequence is yeast or animal cells.

While not wishing to be bound to any theory the applicants believe that by introducing a polynucleotide sequence from a non-plant source, as described, the expression and/or activity of the encoded enzyme is less susceptible to the normal cellular regulatory control mechanisms for enzyme expression and enzyme activity rates, that function within plant cells.

The applicants believe that the heterologous overexpression of the plant polynucleotide encoding galactono-1,4-lactone dehydrogenase may be limited in its effect of increasing vitamin C synthesis to such low levels by the susceptibility of the enzyme to product inhibition. Hence as the concentration of vitamin C increases it is believed that there is a tendency for product retention within the active site of the galactono-1,4-lactone dehydrogenase thereby preventing the entry of new substrates.

Further to this understanding and in accordance with the present invention the applicants now believe that an analogous enzyme encoded by a polynucleotide from a non- plant source has a lower tendency for such inhibition due to evolutionary differences in the protein conformation of the enzyme active site which prevent the product from being an exact fit and facilitate the release thereof.

The applicants also believe that polynucleotide sequences from a non-plant source are less susceptible to feedback inhibition on enzyme expression and of enzyme activity rates than galactono-1,4-lactone dehydrogenase.

Feedback inhibition typically involves the binding of a molecule of a process intermediate or product to surface sites on the quaternary structure of an enzyme protein away from the active site. These interactions alter the shape of the active site and reduce the catalytic ability of the enzyme, thereby reducing protein expression and/or enzyme activity rates.

The applicants believe that the expression of a heterologous plant gene and enzyme activity rate for galactono-1,4- lactone dehydrogenase will be susceptible to the feedback inhibition that naturally occurs in plant systems, thereby restricting the potential for increased vitamin C synthesis.

The applicants therefore see the introduction into the plant genome of analogous genes encoding analogous enzymes from non-plant sources as avoiding these plant control mechanisms.

The enzymes that have been isolated from yeast and animal cells, namely D-arabinono-1,4-lactone oxidase and L-gulono- 1,4-lactone oxidase carry out oxidation of their substrates which are analogous to the plant substrate L-galactono-1,4- lactone. This is in contrast to the galactono-1,4-lactone dehydrogenase which has activity for catalysing a dehydrogenation step requiring the co-factor cytochrome C (Mapson L. W. et al., Biochemical Journal 1958 68 395-406).

This co-factor is simultaneously reduced during vitamin C synthesis and it is likely that an excess of cytochrome C in its reduced form has a negative impact, by feedback inhibition, on the enzyme rate of galactono-1,4-lactone dehydrogenase. Whereas D-arabinono-1,4-lactone oxidase and

L-gulono-1,4-lactone oxidase will not demonstrate such sensitivity and are therefore able to produce more vitamin C.

The analogous yeast or animal polynucleotide sequences and enzymes are thought less likely to have the correct binding sites that would allow control of expression and enzyme activity rates by the binding of plant metabolic intermediates. Hence more vitamin C can be synthesised.

For the present invention it is preferred that the polynucleotide sequence in the gene construct used for transformation comprises a polynucleotide sequence encoding an enzyme with either D-arabinono-1,4-lactone oxidase or L- gulono-1,4-lactone oxidase activity. Therefore another aspect the invention comprises a process as previously described, wherein said polynucleotide sequence encodes an enzyme with either D-arabinono-1,4-lactone oxidase or L- gulono-1,4-lactone oxidase activity.

Most preferably said polynucleotide sequence encodes D- arabinono-1,4-lactone oxidase and comprises the coding part or parts of the polynucleotide sequence shown in figure 1 (SEQ ID No 1).

The process as outlined above can be used to increase the content of vitamin C in the genetically modified plants or plant tissues to at least 150 wt%, preferably at least 200 wt%, more preferably to at least 300 wt%, still more preferred to at least 500 wt%, most preferred vitamin C is

increased to at least 700 wt % of that in unmodified control plants or plant tissues.

The process of the present invention can be used to increase the content of vitamin C in all plants or plant tissues, however it is preferred that the plant or plant tissues for transformation are selected from the group comprising pea, tea, spinach, potato, beans, carrot, tomato, pepper, asparagus, borlotti beans, citrus fruit, Brassicaceae, maize and berries. Most preferably said plants or plant tissues are selected from the group comprising pea, spinach, tomatoes, potatoes, beans, carrot, borlotti beans and Brassicaceae.

Consumers commonly demand the storage stability and general convenience of frozen fruits and vegetables and by way of the present invention such frozen products can be provided with increased levels of vitamin C. Therefore a further embodiment of the invention resides in a process as described above comprising the further steps of blanching and/or freezing and packaging of the genetically modified plant or plant tissue.

Although it is recognised that such further processing steps can be applied to most plants or plant tissues the applicants see this aspect of the invention as being particularly applicable to plants or plant tissues that are selected from the group comprising pea, spinach, potatoes, beans, carrot, borlotti beans and Brassicaceae.

A further aspect of the invention comprises a genetically modified plant or plant tissue obtainable by the process that has been described above.

It is appreciated that the form of gene construct used in the transformation step may vary with the optional presence of transciption and translation regulation factors linked to the polynucleotide sequence capable of expressing a polypeptide product with the catalytic abilities described above.

For the purpose of the present invention it is preferable that the gene construct for the transformation of a plant cell comprises a pGPTV vector with Kanamycin resistance gene, nos terminator sequence, 2x35S promoter sequence and a polynucleotide sequence encoding a polypeptide with D- arabinono-1,4-lactone oxidase activity. More preferably the construct takes the form of the plasmid pArLo see figure 3.

In an important embodiment, the present invention comprises a genetically modified plant or plant tissue having at least 150 wt%, preferably at least 200 wt%, more preferably at least 300 wt%, still more preferred at least 500 wt%, most preferred at least 700 wt % of the vitamin C content of unmodified control plants or plant tissues.

A genetically modified plant or plant tissue of the present invention may also comprise a polynucleotide sequence capable of expressing a polypeptide product with an ability to catalyse conversion of L-galactonono-1,4-lactone to

vitamin C. It is preferred that said polynucleotide sequence is derivable from yeast or animal cells.

In a further aspect of the invention it is preferred that the genetically modified plant or plant tissue comprises a polynucleotide sequence encoding an enzyme with D-arabinono- 1,4-lactone oxidase or L-gulono-1,4-lactone oxidase activity. Most preferred said polynucleotide sequence encodes an enzyme with D-arabinono-1,4-lactone oxidase activity and comprises the coding part or parts of the nucleotide sequence shown in figure 1 (SEQ ID No 1). This nucleotide sequence corresponds to the amino acid sequence for D-arabinono-1,4-lactone oxidase shown in figure 2 (SEQ ID No. 2).

A genetically modified plant or plant tissue according to the present invention may be any so modified plant or plant tissue, however preferably said plant or plant tissue is selected from the group comprising pea, tea, spinach, potato, beans, carrot, tomato, pepper, asparagus, borlotti beans, citrus fruit, Brassicaceae, maize and berries. More preferably said modified plant or plant tissue is selected from the group comprising pea, spinach, tomatoes, potatoes, beans, carrot, borlotti beans and Brassicaceae.

Another embodiment of the present invention resides in a frozen packaged portion of a genetically modified plant or plant tissue according to the above description.

The present invention will be further enabled with reference to the following non-limitative example.

EXAMPLE 1. Isolation of polynucleotide sequence from yeast: The yeast D-arabinono-1,4-lactone oxidase (ArLO) gene sequence can be amplified by the polymerase chain reaction (PCR) from Saccharomyces cerevisiae (NCYC 957).

The yeast is grown overnight at 30 °C with shaking in a medium consisting of 2 % (w/v) glucose, 2 % (w/v) Bactopeptone and 1 % (w/v) yeast extract, sterilised by autoclaving after adjusting the pH to 4.0). Cells are pelleted by centrifugation 4 ml buffer (50 mM Tris. HCl, pH 8,200 mM sodium chloride, 100 mM ethylenediaminetetracetic acid (EDTA), 1 % sodium dodecyl sulphate) added.

The suspension is then heated at 60 °C for 15 minutes, following which, 10 g/ml 1 RNase and 10 mg/ml Proteinase K are added and the suspension heated at 50 °C for a further 15 minutes. The suspension is extracted twice with phenol/chloroform (1: 1) and once with chloroform. The aqueous layer is then added to 0.7 volumes of isopropanol and 0.1 volumes of 3 M sodium acetate, pH 5.2.

After incubation at room temperature for 1 minute, the suspension is centrifuged, the supernatant removed and 2 ml 70 % ethanol is added. The suspension is again centrifuged and the ethanol removed. After air-drying the precipitate for 60 minutes, the pellet is resuspended in 100 jj. l TE buffer (10 mM Tris. HCl (pH 7.4), 1 mM EDTA) and diluted to 50,100,150 and 200 ng/1.

The ArLO gene is amplified by PCR as follows: 10 u. l of lOx PCR buffer, 3 pu ouf 50 mM magnesium chloride, 1 pu ouf nucleotide triphosphates (10 LM each), 1 pl (50 pmoles) of primer 1 (5'-TTTTCACCCCATGGCTACTATCCC-3'Seq. ID No. 3), 1 µl of primer 2 (50 pmoles) (5'-TATTGAGAGAATTCGTTACTAGTCGCAC- 3'Seq. ID No. 4), 1 pl of yeast DNA (50,100,150 or 200 ng), 0.5 u. l of Taq DNA polymerase (5 U/pl), 0.05 pl of pfu DNA polymerase (5 U/tl), 83.5 µl of sterile water are put into a PCR tube and overlaid with 2 drops mineral oil.

The PCR is performed using the following conditions: 2.00 minutes at 94 °C (denature), 0.50 minutes at 55 °C (anneal), 3.00 minutes at 72 °C (extend) and 0.75 minutes at 94 °C (denature, remaining 30 cycles).

Agarose gel electrophoresis shows that a 1598 base pair fragment is produced, and this can be purified using a QIA quick PCR purification kit, eluting the purified DNA in 50 µl EB buffer. [Primers 1 and 2 are based on the ArLO and L- galactono-1,4-lactone oxidase sequences published by and Huh et al, (Genebank Accession N°: U40390) and Nishikimi et al (Biochem. Molec. Biol. Int. (1998) 5 907-913)].

2. Ligation into vector pT7: The fragment can then be ligated into the pT7 vector by taking 15 pl of the purified fragment, 1 µl of pT7 vector (lu. l/ml), 2 pl of 10x ligation buffer and 2 pl T4 DNA ligase (1 pl U/pl) and incubating at 4 °C overnight.

3. Transformation into Escherichia coli cells and purification of transformed plasmid: The resulting ligation is then transformed into competent XLl-blue Escherichia coli cells, prepared from 100 tl of an E. coli cell culture added to 25 ml LB medium containing Tetracycline (12.5 g/ml) and incubated at 37 °C overnight with shaking. 1 ml of this culture is added to 100 ml Lennox broth (LB) medium (0.5 % (w/v) sodium chloride, 1 % (w/v) yeast extract and 1 % (w/v) Bactotryptone) without tetracycline and incubated for a further 2-3 hours.

Bacteria are centrifuged and washed in 100 mM CaCl2, then resuspended in 5 ml CaCl2 and left on ice for 60 minutes. 4 pl of the ligation is added to 200 1 competent E. coli cells and left on ice for 30 minutes.

The cells are then heat shocked at 42 °C for exactly 40 seconds and 300; j. l LB medium added. The cells are then incubated at 37 °C for 30 minutes with shaking and are subsequently plated onto LB agar plates containing tetracycline and Ampicillin (50 g/ml) and incubated overnight at 37 °C.

The presence of the recombinant plasmid in the bacterial colonies is checked by PCR using T7 and U19 as primers, and positive clones grown in 2xTY medium containing Ampicillin.

Colonies demonstrating the correct fragment are grown in 50 ml 2xTY medium with Ampicillin overnight at 37 °C with shaking.

The plasmid is purified from the cell culture as follows: bacteria are pelleted by centrifugation, resuspended in 4 ml solution 1 (50 mM glucose, 25 mM Tris. HCl (pH 8.0), 10 mM EDTA (pH 8.0)). 8 ml of solution 2 (0.2 M sodium hydroxide, 1 % sodium dodecyl sulphate) is added and the mixture incubated at room temperature for 5 minutes. 6 ml buffer solution 3 (5 M potassium acetate, 11.5 ml glacial acetic acid, 28.5 ml water) is added and the mixture incubated at 4 °C for 15 minutes. The mixture is then strained through 2 layers of Miracloth and 10 ml isopropanol added. The solution is centrifuged at and the pellet resuspended in 1 ml TE buffer containing 10 pl/ml RNase.

This mixture is incubated at 50 °C for 20 minutes and chloroform/phenol extracted once and chloroform extracted once. 0.7 volumes of isopropanol and 0.1 volumes of sodium acetate (3M, pH 5.2) is added and mixed. The suspension is then centrifuged at 13,000 rpm, room temperature for exactly 5 minutes and the precipitate air-dried.

The resultant DNA is diluted to 3 g/ml and restriction digests are set up. This requires 2 jul DNA, 3 1 lOx buffer, 3 tl 0.1 % bovine serum albumin, 1 1 of each restriction enzyme (10 U/pl) and sterile water added to 30 J. The mixtures are incubated at 37 °C for 45 minutes.

Treatments are as follows (1) Hin DIII, buffer B, (2) Eco R1, buffer H, (3) Sac 1, buffer A, (4) Sal 1/Mun 1, buffer H, (5) Nco 1/Sna Bl, buffer M and (6) Kpn 1, buffer L.

From the size of the resultant fragments on agarose gels, the insertion and orientation of the DNA are determined.

Restriction digests are set up as follows: (1) 2 pl DNA (5 pg), 3 pl lOx H buffer, 3 pl 0.1 % bovine serum albumin 1 pl Eco Rl (10 U/pl), 1 pl Nco 1 (10 U/pl) and 20; j. l sterile water (2) 0.5 pl pP5LN vector (1 µg/ml), 3 pl lOx M buffer, 3 jul 0.1 % bovine serum albumin 1 jj. l Mun 1 (10 U/pl), 1 pl Nco 1 (10 U/pl) and 21.5 pl sterile water. The mixtures are digested for 45 minutes at 37 °C and run on an agarose gel for 2 hours. Bands at approximately 1600 base pairs are cut out from digestion 1 preparations and the band for preparation 2 is cut out.

The DNA is purified using a QIAquick gel extraction kit, eluting the purified DNA in 50 µl EB buffer. Fragments (1) and (2) are ligated using a mixture of 2 µl lOx ligation buffer, 2 pl T4 DNA ligase (U/pl), 5 pl of (2) and 11 pl of (1). The mixture is incubated overnight at 4 °C. The ligations are inserted into competent E. coli cells as before. After checking for insertion of the fragment by PCR as before, bacteria containing the recombinant plasmid are grown up overnight as before.

Plasmid DNA is again purified, diluted to 0.5 pg/pl and restriction digests set up using a mixtures of 7 µl DNA (3.5 µg), 3 pl lOx buffer, 3 pl 0.1 % bovine serum albumin, 1 pl of each restriction enzyme (10 U/pl) and sterile water to 30 µl. These mixtures are incubated at 37 °C for 45 minutes.

Treatments are as follows (1) Hin DIII/Eco Rl, buffer B, (2) Hin DIII/Nco 1, buffer H, (3) Hin DIII/Sac 1, buffer A, (4) Nco 1/Sac 1, buffer A, (5) Kpn 1/Eco Rl, buffer A and (6) Spe 1/Nco 1, buffer H. The resulting solutions are run on an agarose gel to determine correct insertion of the ArLO DNA.

The fragments are sequenced using the following primers: T7 5'-TAATACGACTCACTATAGGG-3'Seq. ID No. 5 U19 5'-TTTCCCAGTCACGACGTTGT-3'Seq. ID No. 6 (3) 5'-GTTGAATATCCTGAGTTAC-3'Seq. ID No. 7 (4) 5'-GGCTTCAATATTAAATCCAC-3'Seq. ID No. 8 (5) 5'-TACCGATTCTATCAGCGG-3'Seq. ID No. 9 and (6) 5'-CCATTTATAGGCCGTTTGG-3'Seq. ID 10.

Due to point mutations in the sequences, restriction digests and re-ligation is carried out to get an error-free sequence. This is digested and ligated into the pGPTV vector as follows: (ArLO gene) 2 µl of DNA, 3 µl lOx B buffer, 3 pu 0.1 % bovine serum albumin 1 jul Hin DIII 10 U/p l Eco Rl 10 U/pl and 14 jul sterile water, and (vector) 2 pl pP5LN vector 1 µg/ml, 3 il lOx B buffer, 3 µl 0.1 % bovine serum albumin 1 µl Hin DIII 10 U/l, 1 µl Eco Rl 10 U/pl and 20 pl sterile water are incubated at 37 °C for 60 minutes.

After running on an agarose gel, the appropriate bands are cut out and purified using a QIAquick gel purification kit.

The DNA is ligated into the vector as follows: 3 µl 10 x

ligation buffer, 3 pl T4 DNA ligase, 4 pl vector and 20 pl of the ArLO DNA. This is incubated overnight at 4 °C and transformed into E. coli cells as before.

The cells are plated on LB agar plates containing Tetracycline and Kanamycin (50 g/ml) and incubated overnight at 37 °C. (The pGPTV vector contains the Kanamycin resistance gene necessary for clonal selection). Colonies are checked for the presence of the plasmid containing the ArLO gene by PCR as before, except 30035S 5'- CGCAAGACCCTTCCTCTATATAAG-3'Seq. ID No. 11 and nos-as'5- CCGGCAACAGGATTCAATCTT-3'Seq. ID No. 12 are used as primers, and the time for extension is 1 minute. Colonies containing the ArLO gene are grown in 50 ml 2xTY medium with Kanamycin, overnight at 37 °C with shaking and the plasmid purified as before. The plasmid is checked for correct insertion of the gene by restriction digests as follows: Plasmid DNA is purified as before, diluted to and restriction digests set up as follows: 7 u. l DNA (3 g/1), 3 jul lOx buffer, 3 il 0.1 % bovine serum albumin, 1 pu ouf each restriction enzyme 10 U/µl and sterile water to 30 J. The mixtures are incubated at 37 °C for 45 minutes. Treatments are as follows (1) Mun 1, buffer M, (2) Xmn 1, buffer B, (3) Hin DIII/Eco Rl, buffer B, (4) Hin DIII/Sac 1, buffer A, (5) Hin DIII/Sna Bl, and buffer M.

4. Agrobacterium transformation:

The recombinant plasmid pArLO (fig 3.) is then transformed into competent Agrobacterium tumefaciens (strain LBA 4404) and the bacteria grown on LB agar plates containing Kanamycin. Colonies are checked for the presence of the plasmid containing the ArLO gene by PCR as before, using 30035S and nos-as as primers. Bacteria containing the ArLO gene are grown overnight in LB medium at 28 ° C with shaking.

The overnight culture is centrifuged and the cell pellet is resuspended in liquid MS basal medium supplemented with 3 % sucrose.

5. Genetic modification of tobacco plant tissue and the propagation thereof: By way of example the applicants herein provide a detailed account of a process that can be followed for genetic modification, by agrobacterium transformation, of a model system in tobacco. It is recognised that it would be well within the capacity of the skilled person in the art, to make any minor modifications to this process that may be specific to the transformation of a particular plant or plant tissue.

Seeds of Nicotinia tabacum (var. Petit Havana SR1) are surface sterilised for 10 minutes in a solution of 10 % sodium hypochlorite, rinsed three times in sterile distilled water and planted in 50 ml of a MS basal media supplemented with 3 % sucrose and 0.9 % agar. The seedlings are grown aseptically at 26 °C, in a culture room at 3000 lux (16 hour day/8 hour night). After 2 weeks, the seedlings are

thinned and maintained by monthly apical meristem cuttings onto fresh MS basal media.

Discs (7-10 mm dia.) are punched from topmost leaves of young SR1 plants using a sterile cork borer and incubated with a suspension of the Agrobacterium strain containing the ArLO gene for 10 minutes. The discs are then removed, blotted dry and plated face down onto a nurse culture plate of tobacco (Nicotinia bethiana) cells.

This nurse plate is prepared by adding 2 ml of Nicotinia bethiana cell suspension to a plate containing 25 ml of MS salts, B5 vitamins (1 pg/ml nicotinic acid, 1 g/ml pyridoxine, 10 g/ml thiamine, 100 g/ml inositol) supplemented with 1 pg/ml 2-4 D, 2 pg/ml 6-benzylaminopurine and 0.8 % agar. Swirl the cells to cover the agar then cover the cells with a sterile Whatman filter paper (N°. 7 cut to fit the plate).

Discs are incubated in low light (2000 lux) at 26 °C for 2 days to co-cultivate with the Agrobacterium. After 2 days, the discs are removed from the nurse plates and plated face upwards onto selection media (MS basal media supplemented with 3 % sucrose, 0.2 g/ml indole-3-acetic acid, 1 pg/ml 6- benzylaminopurine, 0.8 % agar, 500 g/ml Cefotaxime and 100 g/ml Kanamycin), and incubated at 26 °C, (16 hour day/8 hour night in a light intensity of 3000 lux). Discs are transferred to fresh selection media every 2 weeks. Shoots appearing on the cut edges of the leaf discs are removed into rooting selection medium (MS basal media supplemented

with 3 % sucrose, 0.9 % agar, 500 pg/ml Cefotaxime and 100 g/ml Kanamycin).

Rooted shoots are tested by PCR for the ArLO gene when the plants are large enough to remove a small sample. The sample is transferred to a homogenisation tube on ice and the material is ground using a pestle. 400 jul of Edward's buffer (0.2 M Tris. HCl (p H 7.5), 0.25 M sodium chloride, 25 mM EDTA, 0.5 % sodium dodecyl sulphate) is added to the tube and the sample is centrifuged at room temperature to pellet the plant cell debris. 300 pl of the supernatant is added to 300 pl of isopropanol to precipitate the nucleic acids.

The suspension is left for 5 minutes at room temperature and the sample is centrifuged again. The supernatant is discarded and the tube is inverted to drain. The pellet is rinsed in 500 pu ouf 70 % ethanol and the suspension centrifuged as before. The supernatant is discarded and the tube inverted drain. The DNA pellet is air dried for 60 minutes and the DNA re-dissolved in 100 pu ouf TE buffer.

PCR is performed as before, using the following primer pairs: 30035S/nos-as, primer 1/primer 2, primer 3/primer 2, primer 4/primer 2, primer 5/primer 2, primer 6, primer 2, 30035S/primer 2 and primer 1/nos-as.

Positive plantlets are removed to a mixture of John Innes N° 2 : Perlite (1: 1), gradually weaning plants to growth room conditions (22-26 °C, 16 hour day/8 hour night), while maintaining humidity. Plants are potted on in John Innes N° 2 to flower in 5"pots. Emerging flower spikes are bagged to prevent cross-pollination.

ArLO activity in leaf tissue is determined using the method of Huh et al (Eur. J. Biochem (1994) 225 1073-1079), based on the spectrophotometric measurement of the production of ascorbic acid from L-galactono-1,4-lactone or by the enzymatic production of hydrogen peroxide (Analytical Biochemistry (1980) 105 389-397). Ascorbic acid content is determined using the method of Foyer et a1 (Plant Physiol.

(1995) 109 1047-1057) based on the spectrophotometric measurement of ascorbic acid in the presence and absence of ascorbate oxidase.

ArLO enzyme activity is detected in plant leaf tissue containing the ArLO gene construct by the conversion of L- galactono-1,4-lactone into ascorbic acid and hydrogen peroxide in the absence of the cytochrome c cofactor necessary to demonstrate plant L-galactono-1,4-lactone dehydrogenase activity. These tissues also demonstrate a higher level of ascorbic acid (at least two-fold) than that observed in plant tissue transformed with an Agrobacterium construct lacking the ArLO gene.