WO1999054480A1 - Transgenic plants presenting a modified inulin producing profile - Google Patents

Abstract

A method is disclosed for producing a transgenic plant with a modified inulin producing profile comprising in its genome a combination of one or more expressible 1-SST enzyme encoding genes and one or more expressible 1-FFT enzyme encoding genes, wherein either of these genes or both of them comprise one or more recombinant genes containing one or more 1-SST, respectively 1-FFT, enzyme encoding DNA sequences of plant origin or an expressible homologous sequence thereof. The invention also relates to a method for modifying and controlling the inulin profile of plants and to a method for producing inulin from said transgenic plants. Furthermore, a novel cDNA sequence of a 1-SST enzyme encoding gene of Helianthus tuberosus and a novel cDNA sequence of a 1-FFT enzyme encoding gene of Cichorium intybus are disclosed, novel recombinant DNA constructs and genes derived thereof, as well as novel combinations of expressible 1-SST and 1-FFT enzyme encoding genes. Moreover, the invention also relates to novel polypeptides, homologues thereof and fragments thereof, which have 1-SST activity of 1-FFT activity, and to antibodies capable of specifically binding one or more of them.

Description

1

TRANSGENIC PLANTS PRESENTING A MODIFIED INULIN PRODUCING

PROFILE

Field of the invention

The present invention relates to transgenic plants presenting a modified inulin producing profile, to a method for producing said plants, to a method for modifying and controlling the inulin producing profile of plants and to a method for producing inulin from said transgenic plants. Furthermore, the present invention relates to novel 1-SST and 1-FFT enzyme encoding DNA sequences, to novel recombinant DNA constructs and recombinant genes derived thereof, to novel combinations of expressible 1-SST and 1-FFT enzyme encoding genes, as well as to novel polypeptides or fragments thereof presenting 1-SST and /or 1-FFT activity, and to antibodies capable of binding to them.

Background and prior art

Inulin is a fructan type carbohydrate polymer which occurs as a polydisperse composition in many plants and can also be produced by certain bacteria and fungi. Inulin from plant origin consists of a polydisperse composition of mainly linear chains composed of fructose units, mostly terminating in one glucose unit, which are linked to each other through β(2-l) fructosyl-fructose linkages.

Inulin can be generally represented, depending from the terminal carbohydrate unit, by the formulae GF_n and F_m , wherein G and F respectively represent a glucose unit and a fructose unit, n is an integer representing the number of fructose units linked to the terminal glucose unit, and m is an integer representing the number of fructose units linked to each other in the polyfructose chain. The number of saccharide units (fructose and glucose units) in one molecule, i.e. the values n+1 and m in the above formulae, are commonly referred to as the degree of polymerisation, represented by DP. Often also the parameter (number) average degree of polymerisation, represented by ( DP) is used, which is the value DPn calculated, after complete hydrolysis and considering that in native inulins the F_m fraction is negligible, as follows:

-^ totaPAF P = + 1 total%G

In the equation % refers to weight per cent (wt%). Furthermore, in this 2 calculation the saccharides glucose (G), fructose (F) and saccharose (GF) which are present in the polydisperse polysaccharide, should not be taken into account. The average degree of polymerisation is thus the ( DPn) of inulin, herein interchangeably referred to in short as {DP) inulin or {DP) (De Leenheer, 1996) .

The polysaccharide chains of native inulin from plant origin generally have a degree of polymerisation (DP) ranging from 3 to about 100, whereas the ( DP) of the native inulin largely depends from the plant source, the growth phase of the plant, the harvesting time and the storage conditions. The ( DP) of isolated inulin largely depends on the ( DP) of the native inulin and on the process conditions used for the extraction, purification and isolation of the inulin from the plant or plant parts.

By native inulin or crude inulin is meant herein inulin that has been extracted from plants or parts of plants, without applying any process to increase or decrease the ( DP), while taking precautions to inhibit the plant's own hydrolase activity and to avoid hydrolysis. The ( DP) of native inulin thus essentially corresponds to the {DP) of the inulin as present in the plant or plant parts.

The isolated inulin obtained from plants or plant parts through conventional manufacturing techniques, commonly including extraction, purification and isolation, without any process to modify the {DP) of the native inulin, is termed herein, interchangeably, standard ( DP) grade inulin or standard grade inulin. As a consequence of the manufacturing process, the ( DP) of standard grade inulin is usually about 1 to 1.5 lower than the ( DP) of the native inulin.

Inulin molecules with a DP < 10 are commonly termed oligofructose, inulo-oligosaccharides or fructo-oligosaccharides (in short FOS). Both, inulin chains with a DP < 10 and inulin chains with a DP > 10, are embraced herein by the term inulin . By inulin profile is meant the relative composition of the polydisperse inulin as formed by the individual components including glucose, fructose, sucrose and individual inulin chains, including the distribution pattern of the polyfructose (inulin) chains.

Linear inulin is common in a specific plant family, the Asteraceae, including the plant species Jerusalem artichoke {Helianthus tuberosus) and chicory {Cichorium intybus). Inulin is commonly stored in tap roots (chicory) or in tubers (Jerusalem artichoke) and acts as a storage reserve for regrowth of the sprout after the winter period. 3

Accordingly, typical sources for the production of inulin at industrial scale are roots of chicory and, on a much smaller scale, tubers of Jerusalem artichoke, in which inulin can be present in concentrations of about 14 % to 18 % by weight on fresh weight. Inulin can be readily extracted from these plant parts, purified and optionally fractionated in order to remove impurities, monosaccharides, disaccharides and undesired oligosaccharides, as for example described in PCT patent application WO 96/01849.

Conventional processing of roots of chicory yields a standard grade inulin, containing about 8 wt % of mono- and di-saccharides (including glucose, fructose and sucrose) and a polydisperse mixture of inulin molecules with a DP ranging from 3 to about 60 and a ( DP) of about 10. The DP of the inulin molecules of standard grade inulin from Jerusalem Artichoke tubers ranges from 3 to about 40 whereas the ( DP) is about 7.

It is known that in Asteraceaous plants, including Jerusalem artichoke and chicory, inulin molecules are synthesised by the concerted action of two enzymes: sucrose : sucrose 1-fructosyltransf erase (in short 1-SST enzyme or 1-SST, used interchangeably) and fructan : fructan 1-fructosyltransf erase (in short 1-FFT enzyme or 1-FFT, used interchangeably) (Koops and Jonker, 1994 and 1996). Both 1-SST and 1-FFT are active during the period of inulin synthesis and accumulation:

1-SST catalyses the initial reaction of inulin biosynthesis, the conversion of sucrose into the smallest inulin molecule, the trisaccharide kestose (GFF), according to:

GF + GF — > GFF + G (1) 1-FFT catalyses the redistribution of terminal fructosyl units (-F) between inulin molecules, which results in a stepwise increase in chain length, according to:

GFF_n + GFF_m — > GFF_n-i + GFF_m+ι, (2)

(wherein n and m are integers >0) Some examples of this type of reaction are

GFF + GFF — > GFFF + GF (2a)

GFFF + GFFF — > GFFFF + GFF (2b)

GFFFF + GFFFF — > GFFFFF + GFFF (2c)

An essential difference between the 1-SST enzyme and the 1-FFT enzyme is that the 1-FFT enzyme cannot catalyse reaction (1). In contrast, the 1-SST enzyme, next to reaction (1), can catalyse reactions of type (2) yielding inulin molecules with a low DP (catalysis by known 1-SST enzymes being able to yield inulin molecules with a DP up to about 5). 4

Accordingly, in plants both the 1-SST and the 1-FFT enzymes are contributing to inulin synthesis and the profile of native inulin is determined, inter alia, by sucrose supply, expression of the 1-SST enzyme encoding genes and 1-FFT enzyme encoding genes and the kinetic properties and relative activity of the 1-SST and 1-FFT enzymes which may be controlled by the relative expression of the 1-SST and the 1-FFT enzyme encoding genes.

Inulin is an edible, water soluble polydisperse polysaccharide composition which is used in the manufacture of many food and feed products, drinks and non-food products. In food, feed and drinks, inulin can be used, inter alia, as a bulking agent as well as a total or partial substitute for sugar and /or fat.

Furthermore, inulin can be added to food, feed and drinks to enrich them with soluble fibres having prebiotic properties. Moreover, inulin can also be used as a component of prophylactic and therapeutic compositions. Besides, inulin with a ( DP) of about 10 and more is commonly used at industrial scale as starting material for the manufacture of oligofructose and of fructose, which both are increasingly used in industry as sweeteners, particularly in drinks and fruit compositions.

Usually different applications require inulin with a different profile. For example for use as (oligofructose) sweetener, the inulin molecules should have a low DP, preferably about 3 to 8, whereas for use as fat replacer inulin should preferably have a {DP) higher than 15. For non-food applications inulin may have to be derivatised, for example to obtain carboxymethylated inulin which can be used as sequestering agent for divalent cations. Inulin suitable as starting material in derivatisation reactions should preferably have a ( DP) of at least 20, whereas its level of low molecular weight sugars should be very low.

Standard grade inulin from chicory or Jerusalem artichoke has a too low ( DP) and a too high level of low molecular weight sugars, which makes derivatisation of said inulins difficult.

To prepare inulin which is low in mono- and disaccharides and has a high average degree of polymerisation, preferably a ( DP) of at least 20, various techniques have already been disclosed, for example, a method of manufacture involving a directed crystallisation starting from native or standard grade chicory inulin as described in PCT patent application WO 96/01849. Inulin with such a profile is also very suitable as ingredient in various food, feed, drinks and non-food applications, and as starting material for the manufacture of hydrolysates and derivatives of inulin.

To prepare oligofructose, usually standard grade chicory inulin or preferably chicory inulin with a higher ( DP) , e.g. a {DP) of at least 20, is 5 subjected to partial, enzymatic hydrolysis, whereas to prepare fructose, typically in the form of a fructose syrup, said inulins are subjected to complete enzymatic or acidic hydrolysis, as for example described in patent applications PCT/BE97/00087 and EP 97870111.8 However, every treatment of native inulin or standard grade inulin to reduce the content of low molecular weight sugars, to increase the ( DP) of the inulin, to modify the inulin profile, particularly to modify the distribution pattern of the inulin chains of the source inulin, or to transform inulin into oligofructose, requires one or more additional processing steps, such as, for example, size fractioning or hydrolysis. These additional process steps inevitably result in technical and economical disadvantages.

Accordingly, in the search for methods for producing inulin with a predetermined profile also an other approach is being prospected which envisages the direct production of inulin with a desired profile from genetically modified plants showing a modified inulin producing profile.

Herein the terms genetically modified, transformed and transgenic are used interchangeably; the terms 1-SST or 1-SST enzyme and 1-FFT or 1-FFT enzyme refer to the respective enzymes, whereas the terms sstl03 or sstl03 sequence and fftlll or fftll 1 sequence indicate an example of a 1-SST, respectively 1-FFT encoding DNA sequence, and the terms sstl03 gene and fftlll gene indicate an example of a 1-SST, respectively a 1-FFT enzyme encoding gene.

PCT patent application WO 96/01904 claims a method for producing fructo-oligosaccharides from a transgenic plant containing a gene construct comprising a fructosyl transferase encoding/*¹/ gene from Streptococcus mutans _ a fructosyl transferase encoding Sac B gene from Bacillus subtilus or a mutated version of said genes. The invention aims particularly low molecular fructo-oligo-saccharides having a DP 3 to 4.

The patent application WO 96/01904 describes the isolation of an SST enzyme from onion and shows the activity of the purified enzyme in vitro by incubation with sucrose with the formation of 1-kestose only. Neither the DNA sequence coding for this SST enzyme nor the amino acid sequence of the purified SST enzyme were disclosed. The patent application also discloses the sequences of two other fructosyl-transferases and the use of the 6-sft gene isolated from barley, where it is involved in the biosynthesis of non-inulin type branched fructans, in an heterologous screening leading to the isolation of the cDNA sequences of two virtually identical genes from the flowers of onion, respectively pAC22 and pAC92, which have been tested separately in 6 protoplasts of tobacco. In the patent application (p. 19, line 27 to p. 20, line 2) it is indicated that in this way a fructosyltransferase activity could be shown but no experimental data were presented. Later experiments (Vijn et al., 1997) have revealed that pAC22 (designated pAC2 in Vijn et al., 1997) in fact encodes a 6- G-FFT enzyme (EMBL accession No Y07838) (fructan:fructan 6G-fructosyl transferase), which catalyses the transfer of a fructosyl residue to the carbon 6 of the glucose moiety of sucrose, resulting in the formation of the trisaccharide neokestose (F2-6G1-2F) according to: GFF + GF — > FGF + GF , so that, although presenting fructosyltransferase activity, the disclosed sequences of pAC22 and pAC92 thus do not represent 1-SST coding sequences.

On the one hand there is an essential difference between a 6-G-FFT enzyme and a 1-SST enzyme since (i) the 6-G-FFT enzyme can not use saccharose as donor of a fructosyl residue (as a 1-SST enzyme does) but needs kestose as a fructosyl unit donor, and (ii) the 6-G-FFT enzyme catalyses the synthesis of neokestose (FGF) constituting the starting moiety for the building up of a fructan of the inulin neoseries class, whereas the 1-SST enzyme catalyses the synthesis of kestose (GFF) constituting the starting moiety for the building up of a fructan of the inulin class.

On the other hand there is an essential difference too between a 6-G-FFT enzyme and a 1-FFT enzyme, since the 6-G-FFT enzyme can catalyse only the production of polysaccharide chains with a low DP, i. e. a DP < about 10, whereas a 1-FFT enzyme can catalyse the synthesis of polysaccharide chains with a higher DP, i.e. a DP up to about 70 and even up to about 100.

Said difference between the 6-G-FFT enzyme and the 1-SST enzyme, respectively the 1-FFT enzyme, is also reflected in the polysaccharide molecules built up through catalysis by said enzymes. The polysaccharide chains built up via a 6-G-FFT enzyme catalysis are of the inulin neoseries class, reach a DP of only up to about 10, have not a terminal glucose moiety, and are not strictly linear as a result of the 1-6 disubstituted glucose moiety in the molecule, whereas the polysaccharide molecules built up via 1-SST enzyme and 1-FFT enzyme catalysis can reach a DP of up to about 70, even up to about 100, have a terminal glucose moiety, and present an essentially linear structure. Furthermore, the use of the individual sequences in different DNA constructs to produce transgenic plants, in particular different crops, is mentioned but no experimental data on the oligosaccharides obtained were given.

Furthermore, Vijn et al., 1997 disclosed that introduction of the onion 6-G-FFT enzyme encoding sequence in chicory resulted in a transgenic plant which made linear inulins (i.e. genuine chicory inulin) and in addition fructans 7 of the inulin neoseries. Apparently the native 1-SST enzyme encoding sequences together with the native 1-FFT enzyme encoding sequences of the chicory ensured via the corresponding enzymes the synthesis of the native inulin molecules, whereas the native 1-SST enzyme encoding sequences together with the onion 6-G-FFT enzyme encoding sequence lead via the corresponding enzymes to the synthesis of inulin neoseries of low molecular weight. However, a combination in the genome of one and the same transgenic plant of a 1-SST encoding DNA sequence and a 1-FFT encoding DNA sequence, wherein either or both of said sequences are part of a recombinant construct, which combination is ensuring via the respective enzymes the production of inulin with a modified inulin profile (the host plant being an inulin or a non- inulin producing plant), has not been disclosed yet.

PCT patent application WO 96/21023 discloses a 1-SST encoding DNA sequence and a 1-FFT encoding DNA sequence both from J. artichoke, designated sstl03 (or pSST103 when referring to the original clone being the sstl03 inserted in the pBluescript SK vector) and fftl 11 (or pFFTlll when referring to the original clone being the fftlll inserted in the pBluescript SK vector), respectively, the construction of recombinant genes comprising said 1-SST enzyme or 1-FFT enzyme encoding sequence, and the transformation of plants (petunia and potato plants) by insertion of said sstl03 gene or fftlll gene in the plant genome. Expression of the sst 103 gene in the transgenic plants was shown by analysis of the carbohydrate composition of the plants, revealing the presence of fructo-oligosaccharides with a DP up to 5. Expression of the fftlll gene in the transgenic plants was demonstrated in vitro on the basis of the ability of an extract of the transgenic plant to catalyse the synthesis of a polysaccharide G-(F)_n (π>4) at the expense of submitted G -(F)4 (G = glucosyl; F = fructosyl). No fructans were formed in the latter plants because the FFT enzyme needs fructo-oligosaccharides (such as FOS with DP of 3 or 4) for the synthesis of oligofructans and fructans with a higher DP. Based on the above, the patent application claims a method for producing a transgenic plant showing a modified inulin profile. A combination of DNA constructs comprising the 1-SST encoding DNA sequence, respectively the 1-FFT encoding DNA sequence, in one and the same transgenic plant has not been disclosed. In food, feed, drinks and non-food applications, as well as for the manufacture of fructo-oligosaccharides, fructose and various derivatives of inulin, industry is increasingly making use of inulin. As a result thereof industry is continuously confronted with problems regarding the supply of inulin, particularly the supply of inulin compositions with desirable inulin 8 profiles of highly linear polysaccharide chains. Said inulin compositions should preferably be readily and directly producible at industrial scale from plant sources at economically interesting costs. Preferably said production should be possible by conventional manufacturing processes and without additional process steps to modify the inulin profile of the native inulin, since each additional process step would inevitably increase the manufacturing costs and reduce the overall yield of suitable inulin. Object of the invention.

The object of the present invention is, inter alia, to provide a method by which said and other problems can be solved, as well as to provide means for use in said method. Description of the invention.

As indicated above, in WO 96/21023 a 1-SST encoding DNA sequence, termed sstl03 , from the genome of J. artichoke is disclosed which codes for a 1-SST that catalyses in a plant the conversion of sucrose into kestose, thus enabling the plant to synthesise kestose (GFF), the smallest inulin molecule, from sucrose.

The inventors have now discovered the existence of a further 1-SST encoding DNA sequence in the genome of J. artichoke which apparently is part of a sleeping gene of said genome. The sequence has been identified in the form of its cDNA by DNA sequencing and this cDNA sequence, termed a33 , is given in SEQ ID NO: 1 and is shown in Fig 1 as A33. In Fig. 1 also the corresponding amino acid sequence is indicated, given in SEQ ID NO: 2 .

Furthermore, the inventors have been able to identify a 1-FFT encoding DNA sequence in the genome of chicory and to identify its DNA sequence in the form of its cDNA by DNA sequencing. This cDNA sequence termed c86b , is given in SEQ ID NO: 3 and is shown in Fig 2 A as C86B, together with the corresponding amino acid sequence, given in SEQ ID NO: 4.

The inventors have found that the a33 cDNA sequence can be expressed in a host organism, particularly a plant or a plant part, to produce active 1-SST, when it is inserted in reading frame in a proper DNA construct in the genome of the organism. Said construct typically comprises the a33 cDNA sequence operably linked in the normal orientation to a promoter sequence and to a terminator sequence which are active in said host organism. When the host organism is a plant, the construct is preferably further comprising an operably linked DNA sequence encoding a targeting signal or a transit peptide which ensures targeting of the a33 encoded 1-SST enzyme to a specific subcellular compartment. The constructs, thus constituting a 1-SST enzyme encoding 9 recombinant gene (in short herein 1-SST a33 gene), can be introduced into the genome of the host organism by conventional techniques.

Besides, the inventors have surprisingly found that said a33 cDNA sequence codes for an 1-SST enzyme which catalyses not only the transformation of sucrose to kestose and to lower oligofructoses (DP < about 5), as do the known 1-SST enzyme encoding DNA sequences such as e.g. sst!03 from J. artichoke , but encodes an 1-SST enzyme which is also able to catalyse the synthesis of higher fructo-oligosaccharides (DP up to about 10) in a plant. The a33 cDNA thus encodes in fact a 1-SST enzyme which has an activity which extends beyond the one of the known 1-SST enzymes, e.g. the 1-SST enzyme encoded by the sstl03 cDNA, and has also an activity which is with respect to the building up of inulin chains functionally comparable to an aspect of a 1-FFT enzyme.

The inventors have also found that the c86b cDNA sequence can be expressed in a host organism, particularly a plant or a plant part, to produce active 1-FFT, when it is inserted in reading frame in a proper DNA construct in the genome of the host organism. Said construct typically comprises the c86b cDNA sequence operably linked in the normal orientation to a promoter sequence and to a terminator sequence which are active in said host organism. When the host organism is a plant, the construct is preferably further comprising an operably linked DNA sequence encoding a targeting signal or a transit peptide which ensures targeting of the c86b encoded 1-FFT enzyme to a specific subcellular compartment. The constructs, thus constituting a 1-FFT enzyme encoding recombinant gene (in short herein l-bblc86b gene), can be introduced into the genome of the host organism by conventional techniques.

Expression of said recombinant gene in a host plant results in the production of 1-FFT, which catalyses, as mentioned above, the stepwise synthesis of inulin by transformation of an oligofructose chain or inulin chain into an inulin molecule with a higher DP. Furthermore, the inventors have found that a host organism, in particular a plant, can be transformed to comprise in its genome a combination of one or more expressible 1-SST enzyme encoding genes and one or more expressible 1-FFT enzyme encoding genes, wherein either the 1-SST enzyme encoding genes or the 1-FFT enzyme encoding genes or both comprise one or more recombinant genes containing one or more 1-SST enzyme encoding DNA sequences, respectively one or more 1-FFT enzyme encoding DNA sequences, of plant origin, resulting in a transgenic organism, particularly a transgenic plant, with a modified inulin producing profile. 10

Moreover, the inventors have found that the inulin producing profile of a plant and the profile of the inulin produced by a plant can be modified and even controlled, i.e. modified to yield a desired inulin profile, by producing a transgenic plant which comprises in its genome a combination of one or more expressible 1-SST enzyme encoding genes and one or more expressible 1-FFT enzyme encoding genes which code for 1-SST and 1-FFT enzymes with different kinetic properties. Either the 1-SST enzyme encoding genes or the 1-FFT enzyme encoding genes or both comprise one or more recombinant genes containing one or more 1-SST enzyme encoding DNA sequences, respectively one or more 1-FFT enzyme encoding DNA sequences, from plant sources, the latter 1-SST encoding sequences and 1-FFT encoding sequences being from the same or from different plant sources.

On the basis of said findings, the inventors have been able to provide a solution to the above mentioned and other problems by the present invention. Accordingly, in a first embodiment, the invention relates to a method for producing a transgenic plant by transforming a host plant to comprise in its genome a combination of one or more expressible 1-SST enzyme encoding genes and one or more expressible 1-FFTenzyme encoding genes, wherein either the 1-SST enzyme encoding genes or the 1-FFT enzyme encoding genes or both comprise one or more expressible recombinant genes containing one or more expressible 1-SST enzyme encoding DNA sequences, respectively one or more expressible 1-FFT enzyme encoding DNA sequences which are of plant origin, comprising transforming a host plant by inserting one or more of said 1- SST enzyme encoding genes and /or one or more of said 1-FFT enzyme encoding genes into the genome of the host plant, yielding a transgenic plant with a modified inulin producing profile.

In accordance with the present invention, said 1-SST encoding DNA sequence(s) and said 1-FFT encoding DNA sequence(s) of said recombinant genes are not restricted to the sequences as obtained from the plant sources, but also include expressible homologous sequences thereof with a degree of homology of at least 70 %, preferably at least 75%, more preferably at least 80% , even more preferably at least 85%, and most preferably at least 90% , respectively, irrespective whether or not the homologous sequences are derived from plant sources or are obtained by mutagenesis of DNA sequences from plant sources or from micro-organisms.

By inulin is meant herein fructans of the inulin class, i.e. molecules of the general formulae GF_n and F_m , as defined above, exclusive of fructans of the inulin neoseries class corresponding to the general formula 11

F2 - (lF2)m' - 6G1 - (2Fl)n' - 2F wherein F and G respectively represent fructose and glucose, and m' and n' represent integers which can be the same or different.

By modified inulin producing profile is meant the production of inulin by a transgenic plant which is quantitatively and /or qualitatively different from the production of inulin by the non-transformed host plant.

By modified inulin profile is meant an inulin profile which is qualitatively different from the one of the inulin produced by the host plant, i.e. an inulin composition wherein the ratio of monosaccharides, disaccharides, oligo- saccharides and /or the distribution pattern of the chain length of the individual inulin molecules, i.e. the DP and the {DP), are different from the ones of the inulin composition produced by the non-transformed host plant. For the sake of convenience, the term modified inulin producing profile is embracing herein a modified inulin producing profile, a controlled inulin producing profile as well as a modified inulin profile and a controlled inulin profile.

The non-transformed host plant suitable for the invention can be an inulin producing plant containing in its genome one or more 1-SST encoding genes and 1-FFT encoding genes or only 1-SST encoding genes, or a non-inulin producing plant. If in the genome of the host plant a 1-FFT encoding gene and /or a 1-SST encoding gene is present, when producing the desired transgenic plant according to the present invention said gene or genes can be maintained or their expression can be totally or partially suppressed by known techniques. For example the expression of said genes can be suppressed through anti-sense expression or co-suppression strategies.

The said 1-SST enzyme encoding genes, respectively the said 1-FFT enzyme encoding genes, of the genome of the transgenic plant in accordance with the present invention, can consist of a native gene or a mixture of different native genes, of a mixture of native and recombinant genes, or of one or more different recombinant genes.

If said combination of 1-SST encoding gene(s) and 1-FFT encoding gene(s) comprises known 1-SST encoding gene(s), respectively known 1-FFT encoding gene(s), these known genes may be the ones which are present in the genome of the non-transformed host plant. If said combination comprises recombinant genes, their 1-SST enzyme encoding sequence, respectively the 1-FFT enzyme encoding sequence, can be a known one or a novel one, from a plant source, or a homologous sequence 12 thereof, as defined above, which encodes a 1-SST enzyme, respectively a 1-FFT enzyme.

If both the 1-SST enzyme encoding genes and the 1-FFT enzyme encoding genes comprise a recombinant gene, the 1-SST enzyme encoding sequence(s) and the 1-FFT enzyme encoding sequence(s) of said recombinant genes can be from the same or from different plant species.

If more than one 1-SST enzyme encoding DNA sequence, respectively more than one 1-FFT enzyme encoding DNA sequence, is present in the genome of the transgenic plant, the 1-SST encoding sequences, respectively the 1-FFT encoding sequences, may be identical or not, may be present in one or in different genes, and these 1-SST encoding genes, respectively 1-FFT encoding genes, may be present on one or on different chromosomes.

In a preferred embodiment of the invention, the recombinant 1-SST encoding gene comprises said a33 cDNA sequence or an expressible homologous sequence thereof as defined above.

In an other preferred embodiment, the recombinant 1-FFT encoding gene comprises said c86b cDNA sequence or an expressible homologous sequence thereof as defined above.

In a further preferred embodiment, the recombinant 1-SST encoding gene comprises said a33 cDNA sequence or said homologous sequence and the 1-

FFT encoding gene comprises said c86b cDNA sequence or said homologous sequence.

Said homologous sequences are at least 70 % identical to said a.33 cDNA, respectively to said c86b cDNA, irrespective of whether or not the homologous sequences are derived from an other plant species or are obtained by mutagenesis of fructosyltransferase-encoding sequences from plant sources or from micro-organisms. Preferably the degree of homology is at least 75%, more preferably at least 80% , and even more preferably at least 85%. Most preferably the degree of homology is at least 90%. In a further preferred embodiment, the transgenic plant has been produced by inserting into the genome of a host plant a combination of one or more genes with a known 1-FFT enzyme encoding DNA sequence and one or more genes with the 1-SST enzyme encoding a33 cDNA sequence or said respective homologous sequences thereof which encode a 1-FFT enzyme, respectively, a 1-SST enzyme. In another preferred embodiment, the transgenic plant has been produced by inserting into the genome of a host plant a combination of one or more genes with a known 1-SST encoding DNA sequence and one or more genes with the 1-FFT enzyme encoding c86b cDNA, 13 or said respective homologous sequences thereof. In a further preferred embodiment, the transgenic plant has been produced by inserting into the genome of a host plant a combination of one or more genes with the 1-SST enzyme encoding a33 cDNA sequence or a said homologous sequence thereof and one or more genes with the 1-FFT enzyme encoding c86b cDNA or a said homologous sequence thereof. In a still further preferred embodiment of said execution forms of the invention, the host plant is a non-inulin producing plant.

Typical combinations of 1-SST encoding genes and 1-FFT encoding genes according to the invention comprise respectively a 1-SST enzyme encoding sequence or a said homologous sequence thereof and a 1-FFT enzyme encoding sequence or a said homologous sequence thereof selected from plant species of the Asteraceae family, with the 1-SST encoding sequence and the 1-FFT encoding sequence being selected from the same or from different plant species. Further typical combinations of 1-SST encoding genes and 1-FFT encoding genes according to the invention comprise respectively a 1-SST enzyme encoding sequence or a said homologous sequence thereof and a 1-FFT enzyme encoding sequence or a said homologous sequence thereof selected from plant species from the same or different plant families of the group consisting of the Asteraceae (Compositae) and Campanulaceae, comprising plant species such as, for example, Echinops, species, Helianthus tuberosus, Cichorium intybus, Dahlia species, Cynara species, Viguiera species, such as e.g.Viguiera discolor, Viguiera deltoida, Viguiera annua, Viguiera lanata, Viguiera multiflora, Veronia herbacea, Scorzonera hispanica, Tragopogon porriflorus, Taraxacum, species, Arctium lappa, Campanula rapuncoloϊdes and Bellis perennis. Typically suitable DNA sequences include, for example, the 1-SST encoding sequences sstl03 (J. artichoke), a33 (J. artichoke), c33 (chicory), Genbank accession No U81520 (chicory), Genbank accession No Y09662 {Cynara scolymus), and the 1-FFT encoding sequences c86b (chicory) and fftlll (J. artichoke). By the selection of a proper combination of one or more of said expressible

1-SST encoding genes and one or more of said expressible 1-FFT encoding genes, in combination with the selection of a proper ratio between the expression of the 1-SST encoding and 1-FFT encoding genes and the selection of a suitable host plant, a transgenic plant can be produced according to the invention with a desired modified inulin producing profile. In fact, the selection of proper combinations of said parameters by routine experiments, makes it possible to produce inulin with an almost tailored profile and to modify in a desired manner the inulin producing profile of a given host plant. 14

In a preferred embodiment, the genome of the transgenic plant comprises a combination of said expressible 1-SST encoding genes and said expressible 1-FFT encoding genes which induces the synthesis of native inulin with a degree of polymerisation which is higher, respectively lower, than the one of the native inulin produced, if any, by the non-transformed host plant.

When an inulin essentially composed of fructo-oligosaccharides with inulin chains with a DP≤ about 10 is desired, a method is provided according to a particular embodiment of the present invention, for producing a transgenic plant which produces inulin with such profile. This is very advantageous because no conventional, partial hydrolysis step of inulin with longer carbohydrate chains, such as e.g. standard grade chicory inulin with a ( DP) of about 10 , needs to be included in the production process of said inulo- oligosaccharides. Accordingly, in a first variant, a method is provided for producing a transgenic plant by transforming a host plant to comprise in its genome a combination of one or more expressible 1-SST encoding genes, originating from the host plant or from a different plant source and one or more l-SSTα33 genes. If the host plant contains a native 1-SST encoding gene and a native 1-FFT encoding gene in its genome, the expression of the native 1-FFT encoding gene or of both the native 1-SST encoding gene and the native 1-FFT encoding gene can optionally be suppressed by known techniques, for example through anti-sense expression. In a second variant, a transgenic plant is produced by inserting into the genome of a host plant which does not contain a 1-SST encoding gene nor a 1-FFT encoding gene, one or more 1-SST a33 genes or genes which contain a cDNA sequence which is an homologous sequence thereof, as defined above, which encodes an a33 1-SST enzyme. This particular embodiment of the present invention has become possible as a result of the fact that the 1-SST a33 gene codes for an enzyme which presents an activity of a 1- SST enzyme, i.e. catalysing the synthesis of kestose from sucrose, but also presents a moderate activity with respect to the building up of inulin chains which is comparable to an aspect of a 1-FFT enzyme activity, i.e. catalysing the synthesis of fructo-oligosaccharides from kestose or fructo-oligosaccharides with a low DP.

The selection of the optimal combination and ratio of the number of concerned 1-SST and 1-FFT enzyme encoding DNA sequences in combination with the selection of the most suitable plant species for the production of a desired inulin profile can be made by the skilled person according to conventional techniques, for example through routine experiments.

In the method according to the invention, the transgenic plant can be 15 produced by inserting into the genome of the host plant by conventional techniques one or more expressible 1-SST«33 genes or expressible homologues thereof, or one or more of said expressible 1-SST encoding genes from plant sources or said homologues thereof and /or one or more of said expressible 1-FFT encoding genes from plant sources or said expressible homologues thereof, resulting in a transgenic plant which comprises in its genome one or more l-SSTα33 enzyme encoding genes or combination of said 1-SST encoding genes and said 1-FFT encoding genes as defined herein above.

In a typical execution form, a cell of a host plant is transformed to comprise in its genome a said combination of 1-SST encoding genes and 1-FFT encoding genes as defined above, by inserting by conventional techniques one or more of said genes into the genome of the cell, followed by regenerating a transgenic plant from said transformed cell. If the host plant is transformed with both said 1-SST encoding genes and said 1-FFT encoding genes, the genes can be inserted into the host genome simultaneously or in subsequent steps.

In a typical execution form, said method comprises the following subsequent steps which can be carried out by conventional techniques: i) the preparation of a recombinant gene construct comprising one or more 1-SST enzyme encoding DNA sequences, respectively 1-FFT enzyme encoding DNA sequences as defined above, operably linked to a promoter sequence active in said host plant and a terminator sequence active in said host plant, ii) introduction of the recombinant gene construct obtained in step i) into the genome of a cell of the host plant, and iii) regeneration of the transformed plant cell obtained in step ii) to the corresponding transgenic plant. More specifically the method of the invention comprises preferably the following subsequent steps: a) the construction of a recombinant gene, i.e. a recombinant DNA construct, comprising essentially the following sequences:

1. a promoter which ensures the formation of a functional RNA or a functional protein in the intended target plant, target organs, tissues or cells thereof,

2. one or more copies of a DNA sequence encoding respectively 1-SST or 1-FFT enzyme, functionally connected to said promoter,

3. a transcription terminator operationally connected to said 1-SST or 1-FFT enzyme encoding DNA sequence, 16

4. a DNA sequence encoding a targeting signal or a transit peptide which ensures targeting of the 1-SST enzyme, respectively the 1- FFT enzyme to a specific subcellular compartment, b) the introduction of the recombinant gene obtained in a) into the genome of the host plant yielding genetically modified material, typically a cell, comprising said combination of 1-SST enzyme and 1- FFT enzyme encoding sequences, and regeneration of the genetically transformed material in the corresponding transformed host plant. In the recombinant DNA construct according to the present invention, one or more copies of the 1-SST and 1-FFT enzyme encoding DNA sequences are preferably linked to one or more regulatory sequences which are operational in the host plant ensuring proper expression of said DNA sequence at a sufficiently high expression level in the host plant, in the different plant organs, tissues or cells. Regulatory sequences are, for example, a promoter, a termination signal and a transcription or translational enhancer. A promoter can, for example, be the 35S promoter of the cauliflower mosaic virus (CaMV) or an organ-specific promoter like the tuber-specific potato proteinase inhibitor II promoter, or any other inducible or tissue-specific promoter.

The production of inulin being particularly advantageous in organs storing large amounts of sucrose, such as the tap roots of sugar beet or the stems of sugar cane, a highly preferred promoter is a promoter which is active in organs and cell types which normally accumulate sucrose (the primary substrate for inulin synthesis).

The production of inulin is particularly advantageous in the vacuole which can accumulate very high concentrations of sucrose (e.g. up to about 500 mol m^{" 3}). Accordingly, in the recombinant DNA according to the present invention, the DNA sequence encoding 1-SST or 1-FFT enzyme is preferably linked to a sequence encoding a transit peptide which directs the mature 1-SST or 1-FFT enzyme protein to a subcellular compartment containing sucrose, such as for example said vacuole.

In a preferred embodiment, the host plant is a non-inulin producing plant; in another preferred embodiment, the host plant is an inulin producing plant. Typical host plants for use in the method according to the invention are plants which can easily be grown, which are rather resistant to attack by injurious organisms such as insects and fungi, which give a high yield of plant material per hectare and which can be easily harvested and processed. Besides, said plants should after transformation be able to produce large amounts of inulin, give a high content of produced inulin based on fresh plant material 17 and preferably be able to deposit said inulin in a concentrated manner in parts of the plant, preferably in tap roots or tubers, which can be easily harvested, stored and processed.

Other typical host plants for use in the method of the invention are non- inulin producing plants which are quite sensitive to abiotic stresses such as, for example, drought, cold, and other ones. Transformation of said host plants according to the present invention resulting in a transgenic plant which is producing inulin, even in a quite low level, may significantly increase the resistance of the plant against said abiotic stresses, particularly against drought and/or cold.

Typical host plants suitable for use in the method according to the invention include corn, wheat, rice, barley, sorghum, millets, sunflower, cassava, canola, soybean, oil palm, groundnut, cotton, sugar cane, chicory, bean, pea, cawpea, banana, tomato, beet, sugar beet, Jerusalem artichoke, tobacco, potato, sweet potato, coffee, cocoa and tea.

In a further embodiment, the present invention provides a method for modifying the inulin producing profile of a plant and for controlling the profile of the inulin produced by a plant, which comprises genetically transforming a plant by inserting into its genome one or more expressible l-SSTα33 genes or expressible homologues thereof, or one or more expressible 1-SST encoding genes, and/or one or more expressible 1-FFT encoding genes, as defined above, yielding a transgenic plant comprising in its genome a combination of said 1- SST encoding genes and said 1-FFT encoding genes, as defined herein above, which plant, when cultured under conventional conditions, shows a modified inulin producing profile.

In still another embodiment, the present invention relates to a method for producing inulin from plant material, particularly inulin with a modified profile, by conventional techniques, wherein the source plant material for the method is material, typically tap roots or tubers, from a transgenic plant which comprises in its genome one or more expressible

1-SST a.33 enzyme encoding genes, or a combination of one or more expressible 1-SST encoding genes and one or more expressible 1-FFT encoding genes, or expressible homologues of said genes, as defined above. Such transgenic plant is obtainable by a method of the present invention as described herein before. The production of inulin from plant parts is well known in the art and commonly comprises (i) an isolation step, wherein the crude, native inulin is isolated from the plant material (typically involving extraction of shredded plant parts e.g. tap roots or tubers, with warm water ), followed by (ii) a 18 purification step, typically comprising a depuration treatment (involving liming and carbonatation or another flocculation technique and filtration) followed by a refining treatment (involving treatment over ion-exchangers, treatment with active carbon and filtration) and optionally a concentration of the purified inulin solution, and (iii) an isolation step wherein the inulin is isolated in particulate form from the purified inulin solution obtained in (ii), for example by spray-drying or by directed crystallisation, filtration and drying.

The invention further embraces the novel DNA sequences a.33 cDNA and c86b cDNA , and homologous sequences thereof, as defined above, encoding respectively 1-SST enzyme and 1-FFT enzyme, as well as novel recombinant

DNA constructs, novel recombinant genes and vectors comprising one or more of said novel cDNA sequences. Said novel cDNA sequences are suitable intermediates for the construction of said novel recombinant DNA constructs and recombinant genes, which are on their turn, suitable intermediates and tools for the production of transgenic plants presenting a modified inulin producing profile according to the present invention.

The invention furthermore embraces a novel combination of one or more expressible 1-SST encoding genes and one or more expressible 1-FFT encoding genes, wherein either the 1-SST encoding genes or the 1-FFT encoding genes or both comprise one or more recombinant genes of which one or more of the

1-SST enzyme encoding DNA sequences and one or more of the 1-FFT enzyme encoding DNA sequences are of plant origin or homologues thereof, as defined above. Said novel combination of genes constitutes a suitable intermediate and tool for the production of transgenic plants according to the present invention. The invention also embraces a transgenic plant, producible by a method according to the present invention, the genome of which comprises a combination of one or more expressible 1-SST encoding genes and one or more expressible 1-FFT encoding genes, as defined above.

The present invention also includes cells, plant tissue, plant parts, roots and shoots of transgenic plants according to the invention as well as seeds thereof, which comprise in their genome a combination of one or more expressible 1-SST encoding genes and one or more expressible 1-FFT encoding genes, as defined above.

The present invention also includes novel inulin compositions, i.e. inulin having a novel inulin profile, obtainable by a method according to the present invention, and the use thereof in the manufacture of food, feed, drinks, and non-food compositions, of derivatives of inulin , and of partial and complete hydrolysates of inulin . 19

In still a further embodiment, the present invention relates to a purified and isolated polypeptide having an amino acid sequence as shown in SEQ ID NO:2, respectively in SEQ ID NO: 4, and to purified and isolated respective homologues thereof having at least 70% homology, preferably at least 75%, more preferably at least 80% , even more preferably at least 85%, and most preferably at least 90% homology, as well as to a fragment of said polypeptides, which polypeptides, said homologues thereof and said fragments thereof have 1-SST, respectively 1-FFT, activity. Said polypeptides, homologues and fragments can be obtained according to conventional techniques. For example the polypeptides of SEQ ID NO: 2 and of

SEQ ID NO: 4 can be obtained from chicory roots through a purification procedure based on ammonium sulphate precipitation, followed by lectin affinity chromatography, anion- and cation exchange chromatography.

In an ultimate embodiment, the present invention relates to antibodies capable of specifically binding one or more polypeptides and/or homologues and /or fragments thereof as defined above.

The invention is further disclosed in Figures 1 to 18, wherein: Fig. 1: shows the nucleotide sequence (SEQ ID NO: 1) (A33) and deduced amino acid sequence (SEQ ID NO: 2) of the isolated a.33 cDNA .

Fig. 2A: shows the nucleotide sequence (SEQ ID No. 3) (C86B) and deduced amino acid sequence (SEQ ID No. 4) of the isolated c86b cDNA. Fig. 2B: shows the nucleotide sequence (SEQ ID No. 5) (C33) and deduced amino acid sequence (SEQ ID No. 6) of the isolated c33 cDNA. Fig. 3: depicts a Southern blot analysis of Helianthus tuberosus genomic DNA probed with an a33 PCR fragment. H. tuberosus DNA was digested with EcoRI, £coR5, Xbal, AflHI, Oral, Ndel, Hindi and, Xhol. Plasmid pA33 (the a33 cDNA in the pBluescript vector) was used as a positive control, potato DNA digested with EcoRI was used as negative control.

Fig. 4: depicts a Northern blot analysis of RNA isolated from various tissues from Helianthus tuberosus. : RNA was isolated from dormant tubers (lane 1), sprouting tubers ( lane 2), stolons (lane 3), tubers with a 2-5 mm diameter (lane 4), tubers with a 2-2.5 cm diameter (lane 5), tubers with a 5-7 cm diameter (lane 6), leaves (lane 7), stems (lane 8), fibrous roots (lane 9), receptacle (lane 10) and flower (lane 11). RNA was probed with an sstl03 (Fig. 4A), fftlll (Fig. 4B) or a33 PCR fragment (Fig. 4C). 20

Fig. 5: presents the chimeric gene construct pA33.236 consisting of the enhanced CaMV35S promoter (Penh35S) with a ALMV translational enhancer (amv) , the coding sequence of a33 (a33) and the nos- termination signal (Tnos). Fig. 6.: shows a TLC analysis of leaves and tubers of transgenic potato harbouring the pA33.236 construct. TLC plates were developed twice in 90% aqueous acetone., G=glucose standard, F=fructose standard GF=sucrose standard, GF₂ is the GF₂ standard, H.t.=standard fructan mixture from mature tubers of H. tuberosus. Plant no 27B, 50, 83, 93 and 23 represent individual potato plants harbouring the pA33.236 construct. Control is a control plant harbouring the AGL0 construct. Form each plant (control plant as well as transgenic plants) two tubers were analysed (lanes 1 and 2) and one leaf (lane 3). Fig. 7: shows HPAEC separations of carbohydrates extracted from leaves of control potato transformed with AGL0 lacking the binary vector (A), from leaves of transgenic potato harbouring the pA33.236 construct (B) and a standard inulin extracted from chicory tap roots (C), Fig. 8: represents the chimeric gene construct pUCPA33.342 consisting of the coding sequence of a33 (a33) under control of the enhanced CaMV35S promoter (Penh35S) and the nos -termination signal

(Tnos), the ALMV translational enhancer (amv), and the herbicide resistance gene (pat) under control of the promoter (P35S) and terminator sequences (T35S) of the CaMV35S gene. Fig. 9: represents the chimeric gene construct pUCPCA342.25 harbouring the coding sequences of a33 (a33) and c86b (c86b), each under control of the enhanced CaMV35S promoter (Penh35S) and the nos terminator (Tnos)_v the ALMV translational enhancer (amv), and the herbicide resistance gene (pat) under control of the promoter (P35S) and terminator sequences (T35S) of the CaMV35S gene. Fig. 10: represents the chimeric gene construct pUCPSF344.18 harbouring the coding sequences of fftlll (1-fft) and sstl03 (1-sst), each under control of the enhanced CaMV35S promoter (Penh35S) and the nos - termination signal (Tnos), the ALMV translational enhancer (amv), and the herbicide resistance gene (pat) under control of the promoter (P35S) and terminator sequences (T35S) of the CaMV35S gene.

Fig. 11: represents the chimeric gene construct pUCPSC344.14 harbouring the coding sequences of c86b (c86b) and sstl03 (1-sst), each under control of the enhanced CaMV35S promoter (Penh35S) and the nos - 21 termination signal (Tnos), the ALMV translational enhancer (amv), and the herbicide resistance gene (pat) under control of the promoter (P35S) and terminator sequences (T35S) of the CaMV35S gene. Fig. 12: represents the chimeric gene construct pUCPAF21.17 harbouring the coding sequences of a33 (a33) and fftlll (1-fft), each under control of the enhanced CaMV35S promoter (Penh35S) and the nos -termination signal (Tnos), the ALMV translational enhancer (amv), and the herbicide resistance gene (pat) under control of the promoter (P35S) and terminator sequences (T35S) of the CaMV35S gene. Fig. 13: depicts a Northern blot analysis of RNA isolated from leaves of sugar beets: RNA was isolated from transgenic sugar beet line 7PSF22 (lane 1), transgenic line 9PSC2 (lane 2), transgenic line 8PAF1 (lane 3), transgenics line 8PAF5 (lane 4) and a non-transgenic control sugar beet (lane 5). The RNA was probed with DNA fragments specific for sstl03, a33, c86b or fftlll.

Fig. 14: shows HPAEC separations of carbohydrates extracted from tap root of transgenic sugar beet line 7PSF22 expressing the sstl03 and fftlll genes. Fig. 15: shows HPAEC separations of carbohydrates extracted from tap root of chicory.

Fig. 16: shows HPAEC separations of carbohydrates extracted from tap root of transgenic sugar beet line 9PSC2 expressing the sstl03 and c86b genes. Fig. 17: shows HPAEC separations of carbohydrates extracted from tap root of transgenic sugar beet line 8PAF5 expressing the a33 and fftlll genes. Fig. 18: shows HPAEC separations of carbohydrates extracted from tap root of transgenic sugar beet line 8PAF1 expressing the α33 and fftlll genes.

The present invention results in significant technical and economical advantages. At first, the invention provides novel plant sources of inulin thus helping to solve the problem of sufficiency of inulin supply. Said plant sources may comprise known inulin producing plants which as a result of a genetically modification according to the invention produce more inulin and /or inulin of a preferred modified profile entailing improved functional properties, as well as non-inulin producing plants which have been transformed to produce inulin, particularly inulin having a desired profile. Secondly, the method according to 22 the present invention enables to obtain directly, i.e. without any additional process steps such as e.g. size fractioning or partial hydrolysis, inulin with a certain desirable inulin profile, such as inulin essentially composed of inulin chains with a DP < 10. This obviously has considerable advantages with respect to manufacturing and manufacturing costs.

Furthermore, the invention provides the possibility of producing many transgenic plant species and variants thereof within the scope of the present invention, resulting from the flexibility of the method of the present invention for producing a transgenic plant with a modified inulin producing profile. Indeed, the possibility to produce many different transgenic plants, starting from different host plant species and by using various different combinations of the 1-SST and/or 1-FFT enzyme encoding sequences according to the invention, i.e. including different 1-SST, respectively 1-FFT enzyme encoding DNA sequences originating from various different plant species or homologous sequences thereof, as defined above, and different ratio's of said 1-SST, respectively 1-FFT enzyme encoding DNA sequences, enables to produce various inulin compositions with a different, desired profile. Said desired inulin profile is often translated in improved functional properties of the inulin and commonly in improved physico-chemical and /or or anoleptic properties of end products containing said inulin, and of hydrolysates and of derivatives of said inulin.

The invention enables furthermore to produce transgenic plant species like for example cultural crops and ornamental plants with an increased resistance against abiotic stresses like e.g. drought and /or cold, which also results in economically important advantages.

The polypeptides, homologues and fragments thereof, as defined above, may be used to catalyse specific in vitro chemical reactions, such as, for example, the use of polypeptide of SEQ ID NO: 2 in the synthesis of 1-kestose from sucrose by a fructosyl transfer reaction; the use of polypeptide of SEQ ID NO:4 in the synthesis of fructo-oligosaccharides or inulin by a fructosyl transfer reaction from fructo-oligosaccharides; and the use of a combination of the polypeptides of SEQ ID NO:2 and SEQ ID NO:4, or respective homologues or fragments thereof as defined above, for the synthesis of fructo-oligosaccharides and inulin from one or more carbohydrates selected from the group consisting of sucrose, 1-kestose and oligofructoses.

The invention is further described in detail in the experimental part below, comprising the methodology and examples. It is emphasised that the examples are given as merely illustrative, non-limitative examples of the invention. 23

DNA METHODOLOGY

DNA and RNA isolation, subcloning, restriction analysis and sequencing were performed using standard methods described in molecular biology manuals (e.g. Sambrook et al. 1989; Ausubel et al. 1994). DNA and protein alignments were performed using the DNA analysis software Genworks 2.5.1. (Intelligenetics, Inc., CA). This program uses a progressive alignment method, building the alignment in approximate phylogenetic order using an algorithm similar to FASTA. Protein alignments use a PAM-250 scoring matrix. Construction and screening of a cDNA library from H. tuberosus

Ten μg of poly(A)⁺RNA isolated from tubers of Helianthus tuberosus 'Colombia', harvested in July, was used as starting material for the construction of an Uni-ZAP XR cDNA library (Stratagene, La Jolla, California, USA). Construction, plating and screening of the library were performed according to the protocols developed by Stratagene (USA, cat. no. 237211). About 100.000 plaques of the unamplified cDNA library of H. tuberosus were blotted onto Hybond N⁺ (Amersham) membranes and screened with a probe comprising two DNA fragments. A 840 bp DNA fragment was synthesised by PCR using primers 5'-TACGCTGTCAACTCGTCGG-3' and 5'-TTGAATAGAACATCGGGCTCTAGCG-3' and plasmid pSST103 {sstl 03 cDNA in pBluescript phagemid, Van der Meer et al. 1998) as a template. A 860 bp DNA fragment was obtained by PCR using primers 5'- GTTCAACGCTGCTTGATCCACC-3' and 5'-ACCACGGTCCTTCCAAACGG-3' and plasmid pFFTlll (fftlll cDNA in pBluescript phagemid Van de Meer et al., 1998) as a template. The two PCR fragments were radiolabelled by the RadPrime DNA labelling system (Life Technologies). Hybridisation was at 50°C in 500 mM NaP-buffer, pH 7.2, 7% SDS, 1% BSA, 1 mM EDTA. Filters were washed to a final stringency of 2xSSC, 0.1% SDS at 50°C. (2 x 30 min) and finally exposed to X-omat AR (Kodak). Positive plaques were purified in 3-4 rounds of plaque hybridisation. The pBluescript phagemids were excised from the Uni-ZAP vector using the Exassist/Solr system (Stratagene). The inserts were analysed by restriction enzyme analysis and sequencing. Construction and screening of a cDNA library from Cichorium int bus Ten μg of poly(A) RNA isolated from tap roots of Cichorium intybus 'Cassel', which was harvested in July, was used as starting material for the construction of a lambda TriplEx cDNA library (Clontech laboratories). Construction, plating and screening of the library were performed according to the protocols 24 developed by Clontech (Palo Alto, California, cat. no. CSlOlOt). About 60.000 plaques of the unamplified chicory library were screened with a mixture of the two P-labelled DNA probes obtained by PCR and RadPrime labelling as described above. Hybridisation and washing of Hybond-N⁺ membranes were performed under low stringency conditions (hybridisation at 50°C, final wash step with 2xSSC, 0.1% SDS, 50°C). Positive plaques were purified in 3-4 rounds of plaque hybridisation. The lambda TriplEx clones were converted into pTriplEx clones using the cre-recombinase mediated site-specific recombination at the loxP sites flanking the embedded pTriplEx (Clontech). The pTriplEx clones were analysed by restriction enzyme analysis and DNA sequencing.

PCR

PCR was performed in 50 μl PCR buffer (Life Technologies), containing 100 pmol plasmid as a template, and 100 pmol of gene specific primers (specific for sstl03, fftlll, a33, c33, or c86b, depending on the experiment). Amplification involved 30 cycles of denaturing (0.5 min, 92°C), annealing (1 min, 55°C) and amplification (1 min, 72°C). The resulting fragments were checked by DNA sequencing and restriction digestion to confirm the identity. Transformation of potato Transformation of potato was performed according to Visser (1991), with the following modifications. Stem internodes, cut from in vitro grown Solanum tuberosum, were placed in R3B agar plates, on top of filter paper, which was soaked with 2 ml PACM medium. The internodes were incubated 24 h at 21°C. The binary plasmid pA33.236 was introduced into the Agrobacterium tumefaciens strain 'AGLO' (Lazo et al. 1991) by adding 0.5 microgram of plasmid DNA to 200 microliter of competent Agrobacterium cells. Competent cells were prepared according to Sambrook (1989). The plasmid ON A- Agrobacterium mixture was incubated on ice, for 30 min, then frozen in liquid nitrogen and thawed in a water bath at 37°C for 5 min. After addition of 1 ml YEP medium, the bacteria were incubated at 28°C for 2 hours with gentle shaking. Cells were pelleted and resuspended in 100 μl YEP-medium. Finally, transformed bacteria were selected on YEP-agar plates containing 25 mg/1 kanamycin. The presence of pA33.236 was tested by restriction enzyme analysis.

A. tumefaciens cells, transformed with pA33.236 were grown at 30°C in 5 ml LB medium, containing 50 μg/ml kanamycin and 100 μg/ml rifampicin. After 48 h of growth, the cells were washed twice in 5 ml 2 mM MgSθ4 , then suspended in 5 ml MS medium. Stem internodes were incubated in 20 ml of a diluted (lOOx in MS) A. tumefaciens suspension, and gently shaken for 30 min. After incubation, the stem internodes were blotted dry on filter paper and placed on 25 top of PCM-agar plates (48 h at 21°C). Callus formation was induced by transferring the internodes to PCM agar plates, containing 200 mg/1 cefotaxime and 100 mg/1 kanamycine (96 h at 21°C). Shoot formation was induced at 21°C by transferring the stem internodes to PSM-agar, containing 200 mg/1 cefotaxime and 100 mg/1 kanamycin. Regenerating explants were transferred to fresh medium every three weeks. Root formation was induced by transferring the transgenic shoots to MS 30 medium, containing 200 mg/1 cefotaxime and 50 mg/1 kanamycine. Rooted plantlets of 5-10 cm were transferred to the greenhouse. DNA and RNA analysis of Terusalem artichoke

Total DNA was isolated from mature young mature leaves which were harvested 3 months after transfer of the plants to the greenhouse. Aliquots of the DNA was digested with a number of restriction enzymes. Total RNA was isolated from stolons, tubers of various ages, stems, leaves and flower tissues. After electrophoresis on 1% agarose, DNA or RNA was blotted onto Hybond-

N⁺ (Amersham) and UV cross-linked. Filters were hybridised with either an fftlll, sstl03 (see Construction and screening of a cDNA library from H. tuberosus) or an a33 probe. A 1140 bp DNA fragment specific for a33 was synthesised by PCR using primers 5'-CAACCCAATTCTCTTCCCTCCTCCG-3' and 5'-ACAAACACTTTGGGCGGC-3' and plasmid pA33 {a33 cDNA in pBluescript) as a template. The gene specific DNA fragments were radio labelled with alpha P- ATP using the RadPrime kit (Life Technologies). Hybridisation was at 65°C in 500 mM NaP-buffer, pH 7.2, 7% SDS, 1% BSA, 1 mM EDTA. Filters were washed to a final stringency of 0.1 x SSC, 0.1% SDS at 65°C (2 x 15 min) and finally exposed to X-omat AR (Kodak).

Transformation of sugar beet

Transgenic sugar beets {Beta vulgaris) were generated by a stomatal guard cell based transformation system (Hall et. al. 1996). Guard cell protoplasts were obtained from shoot cultures of the diploid breeding line Bv-NF (Hall et. al. 1996). One million guard cell protoplasts were transformed with 50 μg of the plasmid in the presence of PEG. Regeneration and selection of transformants was as described (Hall et. al. 1996). Plants were grown in a greenhouse at 18/15°C (day/night) under a 16 h photoperiod. Two months old leaves were cut to prevent spread with powdery mildew. DNA and RNA analysis of transgenic plants

Total DNA was isolated from young mature leaves which were harvested 3 months after transfer of the plants to the greenhouse. DNA was digested with specific restriction enzymes. Total RNA was isolated from 2 - 5 month old tap 26 roots or tubers. After electrophoresis on 1% agarose, DNA or RNA was blotted onto Hybond-N+ (Amersham) and UV cross-linked. Filters were hybridised with a gene specific probe. The DNA fragment was radio labelled with alpha⁰ P- ATP using the RadPrime kit (Life Technologies). Hybridisation was at 65°C in 500 mM NaP-buffer, pH 7.2, 7% SDS, 1% BSA, 1 mM EDTA. Filters were washed to a final stringency of 0.1 x SSC, 0.1% SDS at 65°C (2 x 15 min) and finally exposed to X-omat AR (Kodak). Enzyme activity: protein extraction and assay

Leaf and root tissues were ground in liquid nitrogen to a fine powder. Five hundred mg of the powdered tissue were extracted in 1 ml 50 mM BisTris pH

5.5, 1 mM EDTA, 1 mM MgSθ4, 1 mM DTT, 1 mM PMSF, 20 mM Na- metabisulfite and 15g 1 PVPP. The extract was centrifuged (20 min at 10 000 x g) and the supernatant desalted and concentrated by centrifugation through Centriprep 30 ultrafiltration devices (Amicon, Breda, The Netherlands). A 240 μl aliquot was mixed with 240 μl assay mixture, containing either 100 mM sucrose, GF2 or GF3 in 20 mM BisTris pH 5.5, 2 mM DTT and 0.01% Na-azide

(w/v). After 8, 16 and 24h of incubation, an aliquot of the reaction mixture was boiled for 5 min and store at -20°C. The assay mixtures were then analysed by HPAEC as described below. Analysis of sugars and inulin by TLC and HPAEC

The inulin composition of transgenic plants beet was measured by TLC (Thin Layer Chromatography) and HPAEC (High Performance Anion Exchange Chromatography). Two to five months after transfer of the plants to the greenhouse, fresh plant material (500 mg) was collected and cut into 2 mm thick slices with a sterile razor blade. A 20 mM phosphate buffer, pH 7.0, of 100°C was added to the slices to a final volume of 3 ml, and kept at 85°C for 30 min, with occasional vortexing. The extract was centrifuged at 14000 rpm and the supernatant collected. TLC analysis was performed on silica gel G (Schleicher and Schuell Nederland BV, The Netherlands) developed two times in a mixture of l-butanol:2-propanol:water 3:12:4. Carbohydrates were stained with a fructose-specific urea phosphoric spray (Wise et al., 1955). For HPAEC analysis, the extracts were deionized with a mixture of equal amounts of Q- sepharose and S-sepharose Fast How (Pharmacia, Upssala, Sweden), which were pre-equilibrated in 20 mM phosphate buffer, pH 7.0. The ion exchanger was added to 50% of the total extract volume, mixed for 5 min at 600 rpm, then centrifuged at 14000 rpm for 2 min. The supernatant was analysed by High Pressure Anion Exchange Chromatography /Pulsed Amperometric Detection (HPAEC-PAD, Dionex, The Netherlands) equipped with 250 x 4 mm CarboPac 27

PA1 anion exchange column and a 25 x 3 mm CarboPac PA guard column.

Inulins were separated with a 80 min linear gradient of 0 to 0.4 mol m NaAc in 0.1 mol m 3 NaOH at a flow rate of 1 ml min 1 , or over 85 minutes with an aqueous gradient (A: NaOH 0.1 mol m^"3 ; B: (0.1 mol m^"3 NaOH + 0.4 mol m^"3 NaAc); C: NaOH 1 mol m^"3 ) as follows: min 0-5: A:B 96:4; min 5-15 linear gradient A:B from 96:4 to 60:40; min 15-35: linear gradient A:B from 60:40 to

30:70; min 35-50 linear gradient A:B from 30:70 to 10:90; min 50-60 linear gradient A:B from 10:90 to 0: 100; min 60-85 A:B 0:100; followed by regeneration min 0-5 with A:B:C: 0:0:100 and min 5-30 with A:B 96:4, at a flow rate of 1 ml min^" . Standard grade inulin was used as a standard.

CGC analysis was made according to the method described by L. De Leenheer et al, Starch - Starke, (1994), Vol. 46, p. 193-196. EXAMPLES Example 1. Isolation of a new cDNA from the cDNA library of Helianthus tuberosus homologous to sst!03 and fftlll from Helianthus tuberosus An Uni-ZAP cDNA library, constructed from mRNA isolated from H. tuberosus tubers, was screened with a mixture of a 840 bp sstl03 and a 860 bp fftlll fragment. The sstl03 specific fragment was obtained by PCR using primers 5'-TGTCAGCCCATCCCTTGGAAAGG-3' and

5'- TACGCTGTCAACTCGTCGG-3' and pSST103 (Van der Meer et al, 1998) as template. The fftlll specific fragment was obtained by PCR using 5'-GTTCAACGCTGCTTGATCCACC-3' and 5'-ACCACGGTCCTTCCAAACGG-3' and pFFTlll (Van der Meer et al, 1998) as template. Screening of about 100.000 cDNA clones yielded about 1200 positive clones, most of them most probably representing either a sstl03 or a fftlll cDNA. Only the lambda ZAP clones giving a weak positive signal (84) were picked from the primary screening. After 3-4 rounds of purification, 11 clones were left. From these clones, the pBluescript phagemids were excised from the uni-ZAP vector and the cloned insert characterised by restriction enzyme analysis and sequencing. One clone with an about 2 kb insert, designated a.33, was fully sequenced. The DNA sequence of a.33 (Sequence ID. No. 1) and the corresponding amino acid sequence (Sequence ID. No. 2) is presented in Fig. 1. Sequence ID. No. 1, designated a33 has an open reading frame of 1845 base pairs and encodes a protein of 615 amino acid residues

(Sequence ID. No. 2). On DNA level, a33 shows a 70% identity with the 1-SST encoding sstl03 cDNA sequence from H. tuberosus and 54% identity with the 1- FFT encoding fftlll cDNA sequence from H. tuberosus. At the amino acid level, 28 33 shows a 75% similarity with sstl03 fromH. tuberosus and a 58% similarity with fftlll from H. tuberosus.

Example 2. Isolation of a new cDNA from the cDNA library of Cichorium intybus homologous to sstl 03 and fftlll horn Helianthus tuberosus A Lambda TriplEx cDNA library, constructed from mRNA isolated from the tap roots of C. intybus, was screened with a mixture of a 840 bp sstl03 and a 860 bp fftlll fragment (see example 1). The primary screening of about 60.000 cDNA clones yielded about 700 positive clones, of which 96 clones were picked.

After 3-4 rounds of purification 81 clones were left. From 55 clones, the pTriplEx phagemids were excised from lambda TriplEx vector and the cloned insert characterised by restriction enzyme analysis and sequencing. The DNA sequence of one of the clones, designated c86b (Sequence ID. No. 3) and the corresponding amino acid sequence (Sequence ID. No. 4) is presented in

Fig. 2A. Sequence ID. No. 3, designated c86b, has an open reading frame of 1851 base pairs and encodes a protein of 617 amino acid residues (Sequence ID. No. 4). On DNA level, c86b shows a 61% identity with sstl03 from H. tuberosus and 77% identity with fftlll from H. tuberosus. At the amino acid level, c86b shows a 53% similarity with sstl03 from H. tuberosus and a 78% similarity with fftlll from H. tuberosus . The DNA sequence of another clone, designated c33 (Sequence ID. No. 5) and the corresponding amino acid sequence (Sequence ID. No. 6) is presented in Fig 2B. Sequence ID. No. 5 designated c33, has an open reading frame of 1920 base pairs and encodes a protein of 640 amino acid residues (Sequence ID. No. 6). On DNA level, c33 shows a 97% identity with

1-SST from chicory (Genbank accession no U81520). On amino acid level c33 shows a 88% similarity with 1-SST from chicory (Genbank no U81520).

Example 3. Analysis of genomic organisation and expression of a33 in Helianthus tuberosus

To estimate the number of a33 genes present in the Jerusalem artichoke genome,

Southern blot analysis of digested H. tuberosus DNA was performed, using a radio labelled a33 fragment of 1140 bp as a probe. (Plant Journal 15, 489-500).

Under stringent conditions (hybridisation at 65°C and washing at 65°C, O.lxSSC, 0.1% SDS), only one, maybe two fragments of the AflU digest hybridises to the a33 probe (Fig. 3), suggesting that only one, maybe two a33 genes are present in the Jerusalem artichoke genome.

Transcript levels of a.33, sst!03 andfftlll were studied in different organs and 29 different developmental stages of the tubers of Jerusalem artichoke (Fig. 4). In correspondence to earlier experiments (Van der Meer 1998), data in Fig. 4 Ashows that sstl03 is highly expressed in tubers, and to a less extent in fibrous roots, flowers and receptacle. The expression pattern oi fftlll (Fig. 4 B) resembles that of sstl03, although in general the expression oi fftlll is 3-10 times lower than that of sst!03. In contrast to sstl03 and fftlll, a.33 is not expressed to a measurable level in any of the tested tissues (Fig. 4 C). To quantify the a.33, sstl03 and fftlll expressing in tubers, 100.000 plaques of the cDNA library of Jerusalem artichoke tubers were blotted onto Hybond filters, hybridised to either an a33, sstl03 or fftlll probe at 65°C, then washed at high stringency (65°C, O.lxSSC, 0.1% SDS). Expression levels of a33, sstl03 or fftlll in tubers were 0.001, 1 and 0.2%, respectively. If a33 encodes a functional fructosyltransferase protein, it does not effectively contribute to inulin synthesis in tubers of Jerusalem arti-choke. From a functional point of view, a33 can be considered as a silent gene.

Example 4.

Construction of a chimeric α33 gene construct for transformation into potato

By PCR, using primers S'-ATCAACATGTCTTCCACCCC-S' and 5'-TGGATCCTCAAGGCCGCCCTC-3', an Λ ZIII site was introduced at the first

ATG (start of the open reading frame) and a BαmHI site was introduced downstream of the stop codon of the full length α33 cDNA clone (ρA33), isolated from Jerusalem artichoke. The newly obtained α33 PCR fragment was cloned into pMOSblue (Amersham Life Science) yielding pMOSA33.10. From the plasmid pFFT405 (see further, for a complete description), which contains the enhanced Cauliflower Mosaic Virus 35S promoter {enh35S), Alfalfa Mosaic Virus RNA4 leader sequence {αmv), cDN A fftlll from Jerusalem Artichoke and the nos terminator sequence {Tnos), the complete fftlll sequence was replaced by the α.33 PCR fragment. The complete α33 PCR fragment, resulting from a digestion of pMOSA33.10 with AfUJI and Bαrriϋl (partial), was ligated into pFFT405, from which the fftlll was removed by a digestion with Ncol and BαmHΪ (partial). Replacing the fftlll cDNA in pFFT405 with the α33 PCR fragment yielded pA33.102. The eH/z35S-αrau-α33-Tttθs-fragment was cut from pA33.102 with Sfld and Sail, and ligated into the plant transformation vector pBLNPLUS (Van Engelen et al., 1995) digested with Sad/ Sail, which resulted in pA33.236 (Fig. 5). pFFT 405 was obtained by cloning the Penh35S-amv-fftlll-Tnos-fragment cut 30 from pFFT209 (Van der Meer et al, 1998), with EcoRI (partial digest) and H dIII into pBluescript SK+ (Stratagene, USA) cut with EcoRI/H dffl.

Example 5. Analysis of transgenic plants expressing the a33 gene

About 25 transgenic potato plants were generated harbouring the pA33.236 construct. Ten potato plants were transformed with the Agrobacterium strain AGLO lacking a binary vector. These plants were used as a control. Southern blot analysis of genomic DNA isolated from 22 transformed plants showed that 17 plants has integrated into their genome less than 3 copies of the introduced chimeric gene (data not shown). Plant numbers 23, 27B, 74, 84 and 93 had integrated one copy. Northern analysis showed that a33 was highly expressed in plants numbers 27B, 74 and 93, and less, but still clearly expressed in plant numbers 23 and 84. The carbohydrate composition of the a33 harbouring plants was analysed by two essentially different techniques: thin layer chromatography (TLC) and high pressure anion exchange chromatography (HPAEC). Analysis of leaf extracts from the potato harbouring the pA33.26 construct showed that plant numbers 23, 27B, 50, 83 and 93 accumulate a range of fructose containing compounds in leaves and tubers (Fig. 6). A comparison with the inulin standard extracted from Jerusalem Artichoke, containing inulins up to a DP of 30, shows that the tuber extracts accumulate inulins up to a degree of polymerisation of at least 10 (Fig. 6), whereas the leaves accumulate inulins with a DP of 3, 4 and 5. The presence of inulins in leaf extract and tuber extracts of the α33-harbouring plants is confirmed through HPAEC analysis. HPAEC indicates that plant numbers 23, 27B, 93 (only 93 is shown in Fig. 7) accumulate inulins up to a degree of polymerisation of 11. This clearly indicates that a33 encodes a fructosyltransferase enzyme, able to synthesise inulins up tot a DP of at least 11. Since the a.33 encodes an enzyme that can catalyse all steps required for inulin synthesis, including reaction (1), the conversion of sucrose into GFF, the a33 encoded enzyme (A33) belongs to the group of SST encoding genes. Example 6.

Construction of a chimeric a33 gene construct for transformation into sugar beet The complete chimeric a33 gene {Penh35S-amv-a33-Tnos) was cut from pA33.102 with Noil and Sail and ligated into pUCM2, digested with Notl/Sall, resulting in pUCA33.1. Plasmid pUCM2 is derived from the cloning vector pUCAP (Van Engelen et al. 1995). 31

To construct plasmid pUCM2 the multiple-cloning-site of plasmid pUCAP was modified by the insertion of two adapters. First, adapter 5'-TCGACCATATCGATGCATG-375'-CATCGCTATTGG-3', containing the restriction sites Sall/Clal/SphI, was cloned in the pUCAP plasmid digested with Sall/Sphl. This resulted in plasmid pUCMl. Next, adapter

5'-TAAGCGGCCGC AGATCTGG-3' /

5'-AATTCCAGATCTGCGGCCGCTTAAT-3', consisting of Pacl/Notl/Bgia/ EcoRI restriction sites, was cloned in plasmid pUCMl digested with Pad/ EcoRI. This resulted in plasmid pUCM2. The complete chimeric a33 gene {Penh35S- amv-a33-Tnos) was cut from pUCA33.1 with NotI and ΛscI and ligated into pUCPAT 34 digested with Notl/Ascl, yielding pUCPA33.342 (Fig. 8). pUCPA33.342 was used to transform guard cell protoplasts of sugar beet.

Example 7. Construction of a chimeric a33-c86b gene for transformation into sugar beet

The complete chimeric C86B gene {Penh35S -amv-c86b-Tnos), which was cut from pC86B.2 with EcoRI and Clal, and cloned into pUCPAT.34 (see example 8), was digested with EcoRI (partial digest) and Clal. This resulted in pUCPC86B.342. The Notl/Ascl fragment of pUCA33.1, containing the full length A33-cDNA {a33) was ligated into pUCPC86B.342 cut with Notl/Ascl, yielding pUCPCA342.25. (Fig. 9). pUCPCA342.25 was used to transform guard cell protoplasts of sugar beet.

Example 8. Construction of a chimeric sstl03-fftlll gene for transformation into sugar beet

The pat gene, encoding phosphinotricin acetyl transferase (AgrEvo, Berlin, Germany), which confers bialaphos resistance, was cut from pIGPD7 (Hall et al., 1996) with EcoRI and ligated into pUCM3, digested with EcoRI, which yielded pUCPAT34. To construct plasmid pUCM3, plasmid pUCAP was digested with Pacl/Ascl and the whole multiple-cloning-site was replaced by adapter 5'- TAAGGGGTACCACCATCGATACCGAATTCTACATGCATGCATGGAGATC TCCCAAGCTTCTAAGATGCGGCCGCTAAACATGG-375'- CGCGCCATGTTTAGCGGCCGCATCTTAGAAGCTTGGGAGATCTCCATGC ATGCATGTAGAATTCGGTATCGATGGTGGTACCCCTTAAT-3'

The complete chimeric sstl03 gene {Penh35S-amv-sstl03-Tnos) was cut from pSST403 ( Van der Meer et al., 1998) with EcoRI and Clal and ligated into pUCMl, digested with EcoRI/C/αl, which resulted in pUCST21. 32

The complete chimerical 22 gene {Penh35S-amv-fftlll-Tnos) cut from pFFT405 with Noil and Clal was ligated into pUCM2, cut with Notl/Clal, resulting in pUCFT21. The complete chimeric sstl03 gene was cut from pUCST21 with Pad and Clal and introduced into pUCPAT34, digested with Pad/ Clal, yielding pUCPS34.4. The complete chimeric fft 111 gene was cut from pUCFT21 with

Noil and Ascl and ligated into pUCPS34.4, digested with Notl/Ascl, yielding pUCPSF344.18 (Fig. 10). pUCPSF 344.18 was used to transform guard cell protoplasts of sugar beet.

Example 9.

Construction of a chimeric sst!03-c86b gene for transformation into sugar beet Using PCR and primer 5'-CCTCGAACCATGGAAACAGC-3' an Ncol was introduced at the first ATG (start of the open reading frame) of the full length c86b cDNA clone. A BamHl site was introduced downstream of the stop codon of c86b, using primer 5'-TAATAAAAGAGGATCCTCATGAAACG-3'. This c86b

PCR fragment (1895 bp) was cloned into pCR-Script Amp SK(+) (Stratagene, USA) yielding p86BpCRscp. From the plasmid pFFT405, which contains the enhanced Cauliflower Mosaic Virus 35S promoter {Penh35S), Alfalfa Mosaic Virus RNA4 leader sequence {amv) , fftlll cDNA clone from Jerusalem Artichoke and the nos terminator sequence {Tnos) , the completej^iϊll sequence was replaced by the c86B PCR fragment. The c86B fragment, cut from p86BpCRscp with Ncol and BamHl, was ligated into pFFT405 cut with Ncol and BamHl, yielding ρC86B2. The Notl/Clal fragment of ρC86B2 {enh35S-amv- sstl03-Tnos) was cloned into pUCM2 cut with Notl/Clal, yielding pUC86B2. The Notl/Ascl fragment of pUC86B2 was cloned into the pUCPS344 digested with Notl/Ascl, resulting in pUCPSC344.25 (Fig. 11). pUCPSC344.25 was used to transform guard cell protoplasts of sugar beet.

Example 10. Construction of a chimeric a33-fftlll gene for transformation into sugar beet

From the plasmid pUCPA33.342, the complete recombinant a33 gene {Penh35S- amv-sstl03-Tnos) and the complete pat cassette were excised with Clal, then introduced into pUCFT21 cut with Clal, yielding pUCPAF21.17 (Fig. 12)

Example 11.

Analysis of sugar beet plants comprising the pUCPSF344.18 plasmid

Two fructosyltransferase encoding cDNAs from H. tuberosus (sst 103 and fftlll), and the pat gene conferring bialaphos resistance were cloned into the pUCM3 33 vector. The resulting plasmid pUCPSF344.18 (Fig. 10) was used to transform stomatal guard cell protoplasts of sugar beet.

Bialophos-resistent calli were obtained after bialophos selection. Regeneration and selection of transformants was as described (Hall et al. 1996). Transgenic sugar beets were analyzed by Northern analysis (Fig. 13): as the sstl03 probe a

DNA fragment was used which was prepared by PCR using primers 5'-TTGAATAGAACATCGGGCTCTAGCG-3' and

5'-TACGCTGTCAACTCGTCGG-3' and plasmid pUCST21 (example 8) as a template; as the fftlll probe a DNA fragment was used which was prepared by PCR using primers 5'-GTTCAACGCTGCTTGATCCACC-3' and

5ΑCCACGGTCCTTCCAAACGG-3' and plasmid pUCFT21 (example 8) as a template. Line 7PSF22 expressed both the sstl03 and the fftlll gene (Fig. 13, lane 1). This line was further analyzed by HPAEC. The inulin profile can be described as a mixture of inulin molecules up to a DP of 9 (Fig. 14). The inulin profile of sugar beet line 7PSF22 is essentially different from the inulin profile of transgenic sugar beet lines comprising only the sst!03 gene (Severύer et al. 1998). The sugar beet comprising only the sst!03 gene accumulates inulins up to a DP of 5 (GF₂, GF₃ and GF₄; Sevenier et al. 1998). The inulin profile of line 7PSF22 is also different from the standard grade inulin from chicory ( Fig 15), which comprises of a mixture of inulins up to a DP of about 55.

Example 12

Analysis of sugar beet plants comprising the pUCPSC 344.14 plasmid

The plasmid prepared according to the description in example 9, was used to transform stomatal guard cell protoplasts of sugar beet. Bialophos-resistent calli were obtained after bialophos selection. Regeneration and selection of transformants was as described (Hall et al. 1996). Transgenic sugar beets were analysed by Northern analysis (Fig. 13): as a c86b probe a DNA fragment was used, which was prepared by PCR using primers 5'-GTGACCTTGAGGATGCATCC-3'and 5'-TCGGTTGCACCCGCGCTCG-3' and plasmid pUC86B2 (example 9) as a template.

Line 9PSC2, expressing both the sst!03 and the c86b gene (Fig 13, lane 2), was selected for further analysis by HPAEC. The inulin profile of sugar beet line 9PSC2 can be described as a polydisperse mixture of inulin molecules with a DP ranging from 3 to 15 (Fig. 16). The inulin profile of sugar beet line 9PSC2 is essentially different from transgenic sugar beet lines comprising only the sst!03 gene (Sevenier et al. 1998), but also from line 7PSF22 comprising a combination of sst!03 anάfftll, which accumulate inulin molecules up to a DP of 9 (Fig 14). 34

The inulin profile of line 9PSC2 is also different from the standard grade inulin from chicory ( Fig. 15), which comprises of inulins up to a DP of about 55.

Example 13 Analysis of sugar beet plants comprising the pUCPAF21.17 plasmid

The plasmid prepared according to the description in example 10, was used to transform stomatal guard cell protoplasts of sugar beet. Bialophos-resistent calli were obtained after bialophos selection. Regeneration and selection of transformants was as described (Hall et al. 1996). Transgenic sugar beets were analysed by Northern analysis: as the a.33 probe a DNA fragment was used, which was prepared by PCR using primers 5'-

CAACCCAATTCTCTTCCCTCCTCCG -3' and 5'-ACAAACACTTTGGGCGGC -3' and plasmid pUCA33.1 (example 6) as a template. Lines 8PAF1 and 8PAF5 which were expressing both the a33 and the fftlll gene (Fig. 13, lanes 3 and 4, respectively), were analysed by HPAEC. Sugar beet line 8PAF5 accumulates inuline molecules ranging from DP 3-22 (Fig. 17). Sugar beet line 8PAF1 accumulates a mixture of inulin molecules with a DP ranging from 3 to 26 (Fig. 18). In contrast to lines 7PSF22 (example 11), 9PSC2 (example 12) and 8PAF5 (example 13), in which GF₂ is invariably the most abundant inulin molecule, in line 8PAF1, the concentrations of GF₂, GF₃, GF₄ and GF₅ are in the same range, being 2.5. 2.5. 2.4 and 2.2% of total dry matter, respectively. Surprisingly, line 8PAF1 also accumulates significant amounts of F₂, F₃, F₄, F₅ F₆, F₇, F₈ and F₉, being 0.4, 0.4, 0.5, 0.4, 0.4, 0.3, 0.2, 0.1% of total dry weight, respectively (CGC- analysis results). The inulin profiles of sugar beet lines 8PAF1 and 8PAF5 are essentially different from from sugar line 7PSF22 comprising a combination of sstl03 and fftll, which accumulate inuline molecules up to a DP of only 9. The inulin profile of lines 8PAF1 and 8PAF5 are also different from the standard grade inulin from chicory (Fig. 15), which comprises of inulins up to a DP of about 55. 35

REFERENCES

- Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, and Struhl K. (1994).Current protocols in molecular biology. John Wiley & Sons.

- Hall, R.D., Riksen-Bruinsma, T., Weyens, G.J., Rosquin, I.J., Denys, P.N., Evans, I.J., Lathouwers, J.E., Lefebvre, M.P., Dunwell, J.M., van Tunen, A.J., and Krens, F.A. (1996). A high efficiency technique for the generation of transgenic sugar beets from stomatal guard cells. Nature Biotechnology 14, 1133-1138.

- Koops AJ and Jonker HH. (1994). Purification and characterisation of the enzymes of fructan biosynthesis in tubers of Helianthus tuberosus 'Colombia'.

I. Fructan:fructan fructosyltransferase. Journal of Experimental Botany 45, 1623-1631

- Koops AJ and Jonker HH. (1996). Purification and characterisation of the enzymes of fructan biosynthesis in tubers of Helianthus tuberosus 'Colombia'. II. Purification of sucrose:sucrose 1-fructosyltransferase and reconstitution of fructan synthesis in vitro with purified sucrose:sucrose 1 -fructosyltransferase and fructan:fructan 1 -fructosyltransferase. Plant Physiology, in press. . De Leenheer, L (1996). Production and use of inulin, in: Carbohydrates as

Organic Raw Materials, Vol. Ill, p. 67-92. - Sambrook J, Fritsch EF, Maniatis T. (1989) Molecular cloning. A laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor.

- Van Engelen FA, Molthof f JW, Conner, AJ, Nap J-P, Pereira A, and Stiekema WJ. (1995). pBLNPLUS: an improved plant transformation vector based on pBIN19. Transgenic Research 4, 288-290. - Van der Meer IM, Koops AJ, Hakkert JC and Van Tunen AJ (1998). Cloning of Fructan Biosynthesis genes of Jerusalem Artichoke. Plant Journal 1998.

- Vijn L, Van Dijken A, Sprenge N, Van Dun K, Weisbeek P, Wiemken A and Smeekens S (1997). Fructan of the inuline Neoseries is synthesized in transgenic chicory plants {Cichorium intybus L.) harbouring onion {Allium cepa L.) fructan:fructan 6G-fructosyltransf erase. Plant J. 11, 387-398.

- Visser RGF. (1991). Regeneration and transformation of potato by Agrobacterium tumefaciens. In: Plant Tissue Culture Manual B5, ed. Lindsey, K. Dordrecht, The Netherlands, Kluwer Academic Publishers, pp. 1-9.

- Wise CS, Dimler RJ, Davis HA, Rist CE. (1955). Determination of easily hydrolysable fructose units in dextran preparations. Anal. Chem. 27, 33-6.

- Sevenier R., Hall R.D., van der Meer I.M., Hakkert H.J.C., van Tunen A.J. and Koops A.J., 1998. High level fructan accumulation in a transgenic sugar beet. Nature Biotechnology 16: 843-846.

Claims

36CLAIMS

1. Method for producing a transgenic plant with a modified inulin producing profile comprising in its genome a combination of one or more expressible 1-SST (sucrose : sucrose 1 -fructosyltransferase) enzyme encoding genes and one or more expressible 1-FFT (fructan : fructan 1-fructosyl- transf erase) enzyme encoding genes, wherein either the 1-SST enzyme encoding genes or the 1-FFT enzyme encoding genes or both comprise one or more recombinant genes containing one or more 1-SST enzyme encoding DNA sequences, respectively one or more 1-FFT enzyme encoding DNA sequences, which are of plant origin or are an expressible homologous sequence thereof with a degree of homology of at least 70% which encodes an 1-SST enzyme, respectively an 1-FFT enzyme, said method comprising transforming a host plant by inserting one or more of said 1-SST enzyme encoding genes or said homologues thereof and/or one or more of said 1-FFT enzyme encoding genes or said homologues thereof into the genome of the host plant, said host plant being an inulin producing plant or a non-inulin producing plant.

2. Method according to claim 1 wherein said one or more 1-SST enzyme encoding genes comprise one or more recombinant 1-SST enzyme encoding genes.

3. Method according to claim 1 or claim 2 wherein said one or more

1-FFT enzyme encoding genes comprise one or more recombinant 1-FFT enzyme encoding genes.

4. Method according to claim 1 wherein both the said 1-SST enzyme encoding genes and the said 1-FFT enzyme encoding genes comprise one or more recombinant genes.

5. Method according to any one of claims 1 to 4 wherein one or more of the 1-SST enzyme encoding DNA sequences of the recombinant 1-SST enzyme encoding genes and one or more of the 1-FFT enzyme encoding DNA sequences of the recombinant 1-FFT enzyme encoding genes are originating from the same or different plant families of the group consisting of the Asteraceae

(Compositae) and Campanulaceae, or are homologous sequences thereof with a degree of homology of at least 70%.

6. Method according to any one of claims 1 to 5 wherein the 1-SST enzyme encoding DNA sequences and the 1-FFT enzyme encoding DNA sequences are originating from the same or from different plant species or are homologous sequences thereof.

7. Method according to any one of claims 1 to 6 wherein the recombinant 1-SST enzyme genes comprise one or more expressible a33 cDNA 37

sequences defined in Fig. 1 and corresponding to SEQ ID No. 1, or one or more expressible 1-SST enzyme encoding homologous sequences thereof with a degree of homology of at least 70 %.

8. Method according to any one of claims 1 to 7 wherein the recombinant 1-FFT enzyme encoding genes comprise one or more expressible c86b 1-FFT enzyme encoding cD A sequences defined in Fig. 2 A and corresponding to SEQ ID No. 3, or one or more expressible 1-FFT enzyme encoding homologous sequences thereof with a degree of homology of at least 70 %.

9. Method for producing a transgenic plant with a modified inulin producing profile comprising in its genome one or more expressible 1-SST enzyme encoding genes comprising one or more recombinant genes containing one or more expressible a33 cDNA sequences defined in Fig. 1 and corresponding to SEQ ID No 1, or expressible homologous sequences thereof with a degree of homology of at least 70 %, said sequences being coding for an enzyme which acts as a 1-SST enzyme as well as a 1-FFT enzyme, said method comprising transforming a host plant which is free of 1-FFT enzyme encoding genes by inserting one or more of said expressible l-SST╬▒33 enzyme encoding genes or said expressible homologues thereof into the genome of the host plant.

10. Method according to any one of claims 1 to 9 wherein said expressible homologues of the 1-SST enzyme encoding DNA sequences and/ or said expressible homologues of the 1-FFT enzyme encoding DNA sequences of said recombinant genes have a degree of homology of at least 80 %.

11. Method according to any one of claims 1 to 10 wherein the expression of the native 1-SST enzyme encoding DNA sequences and /or the native 1-FFT enzyme encoding DNA sequences, present in the genome of the host plant, have partially or completely been suppressed.

12. Method according to any one of claims 1 to 11 wherein the host plant is selected from the group consisting of corn, wheat, rice, barley, sorghum, millets, sunflower, cassava, canola, soybean, oil palm, groundnut, cotton, sugar cane, chicory, bean, pea, cawpea, banana, tomato, beet, sugar beet, Jerusalem artichoke, tobacco, potato, sweet potato, coffee, cocoa and tea.

13. Method according to any one of the proceeding claims wherein the native inulin produced by the transgenic plant are fructo-oligosaccharides.

14. Method according to any one of claims 1 to 8 and 10 to 12 wherein the native inulin produced by the transgenic plant has an average degree of polymerisation of at least 10. 38

15. Method according to any one of the proceeding claims wherein said method comprises the subsequent steps of

(i) preparing an expressible recombinant gene construct, comprising in an operably linked reading frame a promoter sequence, one or more 1-SST enzyme encoding and /or one or more 1-FFT enzyme encoding DNA sequences from plant origin, or one or more expressible homologous sequences thereof as defined in any one of claims 1 to 11, a terminator sequence, and optionally a DNA sequence encoding a targeting signal or a transit peptide, all active in said host plant, (ii) inserting said 1-SST and/or 1-FFT gene construct of step i) into the genome of a cell of the host plant, and

(iii) regenerating a corresponding transgenic plant from the transformed cell obtained in step ii).

16. Method for modifying the inulin producing profile of a plant comprising genetically transforming a host plant by a method defined in any of claims 1 to 15.

17. Method for producing inulin with a modified inulin profile from plant material by conventional techniques wherein the source material is plant material from a transgenic plant, comprising in its genome a combination of one or more expressible 1-SST enzyme encoding genes and one or more expressible 1-FFT enzyme encoding genes, wherein either the 1-SST enzyme encoding genes or the 1-FFT enzyme encoding genes or both comprise one or more recombinant genes as defined in any of claims 1 to 11, obtainable by a method defined in any of claims 1 to 15.

18. The a 33 1-SST enzyme encoding cDNA sequence defined in Fig. 1 and corresponding to SEQ ID No. 1 and 1-SST enzyme encoding homologous DNA sequences thereof with a degree of homology of at least 70 %.

19. The c86b 1-FFT enzyme encoding cDNA sequence defined in Fig. 2A and corresponding to SEQ ID No. 3 and 1-FFT enzyme encoding homologous DNA sequences thereof with a degree of homology of at least 70 %.

20. Recombinant DNA construct, recombinant gene or vector comprising one or more copies of a cDNA sequence or an homologous sequence thereof as defined in any of claims 18 and 19.

21. Combination of one or more expressible 1-SST enzyme encoding genes and one or more expressible 1-FFT enzyme encoding genes, as defined in any of claims 1 to 8 and 10 to 11.

22. Transgenic plant cell comprising in its genome an expressible combination of one or more expressible 1-SST enzyme encoding genes and one 39

or more expressible 1-FFT enzyme encoding genes as defined in any of claims 1 to 8 and 10 to 11.

23. Transgenic plant cell comprising in its genome one or more expressible 1-SST enzyme encoding genes as defined in claim 9.

24. Transgenic plant with a modified inulin producing profile , root, shoot, plant part, plant tissue, plant cell thereof or seed thereof with a genome corresponding to said transgenic plant, comprising in its genome a combination of one or more expressible 1-SST enzyme encoding genes and one or more expressible 1-FFT enzyme encoding genes, wherein either the 1-SST enzyme encoding genes or the 1-FFT enzyme encoding genes or both comprise one or more recombinant genes as defined in any of claims 1 to 11, obtainable by a method defined in any of claims 1 to 15.

25. Transgenic plant as defined in claim 24, which presents an improved tolerance against drought, cold and/or other abiotic stresses.

26. Use of inulin with a modified profile obtained by the method of claim 17 in the manufacture of food, feed, drinks, partial hydrolysates, complete hydrolysates and derivatives of inulin.

27. A purified and isolated polypeptide, having an amino acid sequence selected from the group consisting of the sequences SEQ ID NO: 2 and SEQ ID NO: 4; or having at least 70% homology with said respective sequences, or a fragment of said respective polypeptides, which polypeptide, said homologue and said fragment have 1-SST activity, respectively 1-FFT activity.

28. An antibody capable of specifically binding one or more polypeptides, homologues thereof or fragments thereof, as defined in claim 27.

29. Use of a polypeptide of sequence SEQ ID NO: 2 , or homologue thereof or fragment thereof as defined in claim 27 as a catalyst in the in vitro synthesis of 1 -kestose and oligofructoses from sucrose.

30. Use of a polypeptide of sequence SEQ ID NO: 4 , or homologue thereof or fragment thereof as defined in claim 27 as a catalyst in the in vitro synthesis of oligofructoses and inulin from 1-kestose and/or oligofructoses.

31. Use of a combination of a polypeptide of sequence SEQ ID NO: 2 and a polypeptide of sequence SEQ ID NO: 4 , or of the respective homologues thereof or fragments thereof, as defined in claim 27, as a catalyst in the in vitro synthesis of oligofructoses and/or inulin from one or more carbohydrates selected from the group consisting of sucrose, 1-kestose and oligofructoses.