WO1999040206A1

WO1999040206A1 - Plant alkaline and neutral invertases

Info

Publication number: WO1999040206A1
Application number: PCT/EP1999/000623
Authority: WO
Inventors: Arnd Sturm
Original assignee: Novartis Ag; Novartis-Erfindungen Verwaltungsgesellschaft M.B.H.
Priority date: 1998-02-03
Filing date: 1999-02-01
Publication date: 1999-08-12
Also published as: AR014528A1; JP2002505084A; EP1053334A1; GB9802249D0; CN1289367A; AU742048B2; CA2319316A1; AU2721799A

Abstract

The present invention provides DNA comprising a sequence of nucleotides which can be translated into a protein with invertase activity, wherein highest activity is observed in the range of pH 6.0 to 7.5, as well as the encoded protein produced recombinantly.

Description

Plant Alkaline and Neutral Invertases

The present invention relates to DNA encoding proteins which hydrolyze sucrose. In particular, the present invention describes DNA which can be translated into a protein with neutral invertase activity.

In most plant species sucrose (α-D-glucopyranosyl β-D-fructofuranoside) is the first non- phosphorylated product of photoassimiiation and serves as a mobile source of energy and carbon for heterotrophic plant tissues. Metabolism of sucrose is an absolute requirement for the survival of heterotrophic plant organs, where sucrose can only be utilized after cleavage by invertase or sucrose synthase.

Invertases hydrolyze sucrose into glucose and fructose thus feeding the sucrose into various biochemical pathways. There exist several isoforms of invertases with different biochemical properties and subcellular locations. Acid invertases are characterized by acidic pH optima in the range of 4.0 to 5.5. They are found ionically bound to the cell wall (cell wall invertases) or as soluble proteins in the vacuole (vacuolar invertases). The amino acid sequences of said acid invertases share conserved motifs. Analysis of sequence similarities suggests that they are evolutionary related to invertases from yeast and bacteria whereas no counterparts have been found in animal cells. Technically acid invertase is used in the confectionary industry to convert easily crystallized sucrose into the less easily crystallized glucose-fructose mixture. Thereby a hard sucrose core coated for example with chocolate can be turned into the soft center ate. Depending on the specific use invertases with neutral or alkaline pH optima would be preferred.

Neutral and alkaline invertases are characterized by pH-optima in the range of 6.0 to 7.5 and 7.5 to 8.5, respectively. They are thought to be confined to mature tissues and it is generally assumed that they accumulate in the cytoplasm which is supported by the fact that no N-linked glycans have been detected. Up to now the molecular structure of neutral invertases and their genes had not been elucidated. This, however, is a prerequisite for the biotechnological exploitation of neutral or alkaline invertases. The corresponding enzymes from carrot (Daucus carota cv Queen Anne's Lace) have been purified (Lee and Sturm, Plant Physiol 112: 1513-1522, 1996) and biochemically characterized recently. Cells of a suspension culture of carrot contain soluble sucrose-cleaving activities with distinct pH optima above and below pH 6 (alkaline and acid invertase, respectively). The two activities were efficiently separated by an ammonium sulphate precipitation at 20-45% saturation. Activity of neutral and alkaline invertase was detected in the protein pellet, whereas that of acid invertase remained in the supernatant. The 20-45% ammonium sulphate fraction was chromatographed on Q-Sepharose and two peaks of invertase activity with only poor separation were obtained. Fractions containing activity were combined and further purified by chromatography on HA-Ultrogel followed by affinity chromatography on Green 19, leading to the efficient separation of the two activities. A sucrose-cleaving activity with a neutral pH optimum (neutral invertase, Hi) was identified in the non-bound protein fraction. An activity with a more basic pH optimum (alkaline invertase, H₂) bound to the HA-Ultrogel and Green 19 dye columns and could be eluted with salt-containing buffers. At this stage of the purification, neutral invertase accounted for about one-third of the invertase activity, and alkaline invertase for two-thirds. The pooled fractions containing neutral and alkaline invertase, respectively, were individually purified further by gel filtration chromatography, a second ion-exchange chromatography, a second gel filtration chromatography and hydrophobic interaction chromatography. Macro-Prep anion-exchange chromatography for neutral invertase was the most effective procedure for the removal of contaminating proteins from the preparations. Although propyl agarose chromatography did not increase the specific activity of alkaline invertase, it was required to obtain electrophoretically pure enzyme. At the end of the purifications less than 10% of the two enzymes were recovered. Said losses are considered to be a consequence of the various purification steps employed and low enzyme stabilities.

Neutral invertase was found to elute from a gel-filtration column as a polypeptide with approximately 456 kD, whereas purified enzmye migrated as a single band of about 57 kD on SDS polyacrylamide gel electrophoresis. Alkaline invertase was found to elute as a polypeptide with approximately 504 kD, whereas purified enzmye migrated as a single band of about 126 kD on SDS polyacrylamide gel electrophoresis. The results suggested that neutral invertase constitutes an octamer and alkaline invertase a tetramer of the corresponding enzymes. Their pH optima were determined to be at pH 6.8 and 8.0, respectively. In addition neutral invertase was shown to cleave raffinose and stacchyose suggesting a β-fructofuronidase activity of the enzyme.

It is the main object of the present invention to provide DNA comprising a nucleotide sequence which can be translated into a protein with invertase activity, wherein highest activity is observed in the range of pH 6.0 to 8.5, preferably 6.0 to 7.5. Although neutral and - 3 -

alkaline invertase are believed to be products of different genes, they appear to be immunologically related.

Dynamic programming algorithms yield different kinds of alignments. In general there exist two approaches towards sequence alignment. Algorithms as proposed by Needleman and Wunsch and by Sellers align the entire length of two sequences providing a global alingment of the sequences. The Smith-Waterman algorithm on the other hand yields local alignments. A local alignment aligns the pair of regions within the sequences that are most similiar given the choice of scoring matrix and gap penalties. This allows a database search to focus on the most highly conserved regions of the sequences. It also allows similiar domains within sequences to be identified. To speed up alignments using the Smith- Waterman algorithm both BLAST (Basic Local Alignment Search Tool) and FASTA place additional restrictions on the alignments.

Within the context of the present invention alignments can be conveniently performed using BLAST, a set of similarity search programs designed to explore all of the available sequence databases regardless of whether the query is protein or DNA. Version BLAST 2.0 (Gapped BLAST) of this search tool has been made publicly available on the internet (currently http://www.ncbi.nlm.nih.gov/BLAST/). It uses a heuristic algorithm which seeks local as opposed to global alignments and is therefore able to detect relationships among sequences which share only isolated regions. The scores assigned in a BLAST search have a well-defined statistical interpretation. Particularly useful within the scope of the present invention are the blastp program allowing for the introduction of gaps in the local sequence alignments and the PSI-BLAST program, both programs comparing an amino acid query sequence against a protein sequence database, as well as a blastp variant program allowing local alignment of two sequences only. Said programs are preferably run with optional parameters set to the default values.

Global or local alignment of the amino acid sequences according to the present invention with known sequences shows less than 40% sequence identity to known sequences of acid invertases or other sucrose-metabolizing enzymes. Examples of DNA comprising a nucleotide sequence which can be translated into a protein with neutral invertase activity are described in SEQ ID NO: 1 and SEQ ID NO: 3. The amino acid sequence of the encoded invertase is given in SEQ ID NO: 2. Related proteins showing more than 40% - 4 -

sequence identity to SEQ ID NO: 2 and their corresponding genes can be isolated from at least any plant from which seeds, fruits or storage organs are harvested. Examples are protein crops, oil crops, and starch storing crops, sugar beet, corn, sweet corn, soybean, sunflower, grasses, oilseed rape, wheat, barley, sorghum, rice, melon, watermelon, squash, chicory, tomato, pepper, broccoli, cauliflower, cabbage, cucumber, daikon, benas, and lettuce.

The protein described in SEQ ID NO: 2 lacks a signal peptide and is very hydrophilic. Furthermore it contains 18 cystein and 15 methionine residues. It shows highest global sequence identity (47%) after alingment to the LIM17 protein which is encoded by a partial cDNA clone obtained from Lilium longiflorum. Global alingment to other protein sequences results in less than 40% sequence identity. The DNA sequences encoding LIM proteins were originally identified when screening a library obtained from cDNA derived from microsporocytes of Lilium longiflorum in meiotic prophase using a substraction probe specific to meiotic prophase (Kobayashi et al, DNA Research 1 : 15-26, 1994). Using the computer program GAP the amino acid sequence deduced from the partial sequence of the Lilium longiflorum LIM17 protein is 47% identical (58% similiar) to the carrot protein. The related LIM17 protein encoded by the genome of the unicellular cyanobacterium Synechocystis (ORF sll0626) is 37% identical (47% similiar) to the sequence of the carrot enzyme after several large gaps had to be introduced for optimal alignment. Thus, it is possible that related proteins, that is proteins showing a sequence identity to the invertases of the present invention of more than 40%, might be found in photosynthetic bacteria. Like the carrot protein, the LIM17 proteins from Lilium and Synechocystis are rich in Cys and Met but their positions within the polypeptide chains do not seem to be conserved. The functions or enzymatic activites of the LIM17 proteins, which are smaller than the carrot sequence homologues, are not known.

Thus, according to the present invention a family of neutral invertases can be defined the members of which after global alingment show a 40% or higher amino acid sequence identity to SEQ ID NO: 2. Preferably the amino acid sequence identity is higher than 50% or even higher than 55%. Sequences more than 55% identical might be considered a subfamily. Sequences according to the present invention can also comprise component sequences of at least 330, 450 or 510 basepairs length which are at least 60%, 70% or even more than 75% identical to locally aligned component sequences of SEQ ID NO: 2. - 5 -

When making multiple sequence alignments certain algorithms can take into account sequence similarities such as same net charge or comparable hydrophobicity/hydrophilicity of the individual amino acids in addition to sequence identities. Thus, said algorithms evaluate whether the substitution of one amino acid for another is likely to conserve the physical and chemical properties necessary to maintain the structure and function of the protein or is more likely to disrupt essential structural and functional features of a protein. Such sequence similarity is quantified in terms of of a percentage of positive amino acids as compared to the percentage of identical amino acids and can help to assign a protein to the correct protein family in border-line cases. Proteins of particular interest within the scope of the present invention are invertases the amino acid sequence of which comprises at least one of the following characteristic amino acid subsequences:

(a) VGTVAA (SEQ ID NO: 4)

(b) AIGRV (SEQ ID NO: 5)

(c) DFGESAIGRVAPVDSGLWWIIL (SEQ ID NO: 6)

(d) CMIDRRMGI (SEQ ID NO: 7)

(e) PTLLVTDGSCMIDRRMGIHGHPLEIQAL (SEQ ID NO: 8)

(f) GGYLIGN (SEQ ID NO: 9)

(g) DFRFFTLGN (SEQ ID NO: 10)

DNA encoding invertases belonging to said new family of proteins can be produced by the following general method. A single stranded fragment of SEQ ID NO: 1 consisting of at least 15, prefeably 20 to 30 or even more than 100 consecutive nucleotides is used as a probe to screen a DNA library for clones hybridizing to said fragment. The factors determining hybridization are described in Sambrook et al, Molecular cloning: A laboratory manual, Cold Spring Harbor Laboratory Press, chapters 9.47-9.57 and 11.45-1 1.49, 1989. Hybridizing clones are sequenced and DNA of clones comprising an open reading frame encoding a protein with more than 40% sequence identity to SEQ ID NO: 2 are purified. Said DNA can then be further processed by a number of routine methods of recombinant DNA such as restriction enzyme digestion, ligation, or polymerase chain reaction.

DNA comprising a sequence of nucleotides defined by SEQ ID NO: 1 can be cloned in the following way:

Comparison of a partial internal tryptic peptide sequence (XNIYPDQIPPWLV, SEQ ID NO: 11) of purified carrot neutral invertase with an est database revealed its presence in the amino acid sequence encoded by est t88552 from Arabidopsis (1026 bp 3'-end of an Arabidopsis cDNA). A DNA fragment encoding this peptide sequence (nucleotides 100 to 410) can be isolated by PCR using the primers 5'-TCTAAGGATCTAGAAAGAGCCATTA-3' (SEQ ID NO: 12) and 5'-TTCAATTGAATTCAATATAGCTTC-3' (SEQ ID NO: 13). The PCR product is after cleavage with Xba\ and EcoRI ligated into the respective sites of the E. coli plasmid pBluescript II KS (Stratagene). After amplification and purification of the plasmid, the fragment is excised, purified by agarose gel electrophoresis and electroelution, and randomly labeled with [ -³²P]ATP. The labeled DNA is used as a probe to screen a library made from rapidly growing suspension cultures of wild carrot (Daucus carota cv Queen Anne's Lace, W001C). Clones obtained are sequenced and for example might reveal a clone comprising a 2447 nucleotide insert containing 29 bp of 5' and 393 bp of 3' non- coding sequences whereas the ORF codes for a protein with 675 amino acids sharing 80% identity (86% similarity) with the deduced amino acid sequence of the Arabidopsis est t88552.

A person skilled in the art is able to modifiy said process to make it applicable to any gene encoding a protein belonging to the new family of invertases. Additionally, disclosing SEQ ID NO: 3 enables a person skilled in the art to design oligonucelotides for polymerase chain reactions which attempt to amplify DNA fragments from templates comprising a sequence of nucleotides characterized by any continuous sequence of 15 and preferably 20 to 30 or more basepairs in SEQ ID NO: 1. Said nucleotides comprise a sequence of nucleotides which represents 20 and preferably 20 to 30 or more basepairs of SEQ ID NO: 1 . Polymerase chain reactions performed using at least one such oligonucleotide and their amplification products constitute another embodiment of the present invention. Further, the disclosed nucleotide sequences enable a person skilled in the art to design transformation vectors which can be used to generate transgenic plants applying art-recognized transformation techniques as described for example in WO 96/27673 (pages 17-20).

A further object of the present invention is to provide recombinant plant invertase with a neutral pH optimum. This can be achieved by recombinant expression of DNA encoding said invertase, preferably cDNA, in a microbial host such as E.coli or yeast. For example recombinant invertase can be produced the following way: cDNA encoding the enzyme is engineered into an expression vector such as p7rc 99 A (Pharmacia Biotech). After - 7 -

transformation of bacteria such as E. coli and, if required, induction of protein synthesis with for example IPTG, bacteria are lysed. Neutral invertase activity is determined in the soluble lysate fraction. In particular about 100μl of soluble extract are mixed with 700μl of water, 100μl of 0.5 M potassium phosphate, pH 6.8, and 100μl of 0.5 M sucrose, and incubated for 30 min at 37°C. Aliquots of this solution are used for the determination of reducing sugars according to Somogyi (Somogyi, J Biol Chem 195: 19-23, 1952). For the determination of the pH dependence of the sucrose-cleaving activity, solutions of 0.5 M potassium phosphate with pH values between 4.5 and 8.5 are used.

Three key biochemical properties of the recombinantly produced invertase are very similiar to those of the enzyme purified from plants, namely a K_m value of about 20mM, a pH dependence with a sharp maximum between pH 6.5 and 7.0, and an inhibition by Cu²⁺ at micromolar concentrations. On the other hand, the recombinant enzyme unexpectedly hydrolyzes only sucrose without cleaving raffinose or stachyose. Thus, the recombinant protein is substantially devoid of β-fructofuranosidase activity.

EXAMPLES:

Example 1 : Purification of Carrot Neutral and Alkaline Invertase

Preparation of Extracts

Carrot cells (400g) collected from suspension cultures in the exponential growth phase are homogenized four times for 20 sec at full speed with a Polytron homogenizer in 2.5 volumes of ice-cold buffer A (50mM Hepes-KOH, pH 7.5, containing 0.5mM EDTA, 10mM lysine, 0.5mM MgCI₂, 0.5% 2-mercaptoethanol and 100mM phenylmethylsulfonyl fluoride). The homogenate is centrifuged for 20 min at 6000(7 in a Sorvall GSA-rotor. The supernatant is collected and kept cold. The 6000gr pellet is resuspended in 2.5 volumes of ice-cold buffer A, homogenized with a Polytron homogenizer three times for 20 sec at full speed and centrifuged for a further 20 min. The combined supernatants are centrifuged at 16,300g for 30 min and then poured through four layers of Miracloth (Calbiochem-Behring Corporation, La Jolla, USA). The filtrate is used for further protein purification. If not stated otherwise, all steps are carried out at 4°C. - 8 -

Ammonium Sulfate Precipitation

Solid ammonium sulfate is slowly added to the crude extract with gentle stirring, and the protein that precipitates between 20% and 45% saturation is collected by centrifugation for 30 min at 16,300α\ The precipitate is dissolved in 100ml of buffer B (25mM Hepes-KOH, pH 7.5, containing 190mM NaCI, 0.5% 2-mercaptoethanol and 100mM phenylmethylsulfonyl fluoride), and dialyzed against buffer B overnight.

Anion-Exchange Chromatography on Q-Sepharose

The dialysate is loaded onto a Q-Sepharose column (2.5cm x 25cm, Pharmacia LKB Biotechnology, Uppsala, Sweden) equilibrated with buffer B. The column is washed with buffer B until the absorbance at 280nm is less than 0.01. Bound protein is eluted with a linear gradient of 240ml of 190-550 mM NaCI in 25mM Hepes-KOH, pH 7.5, containing 0.5% 2-mercaptoethanol and lOOrnM phenylmethylsulfonyl fluoride. Active fractions (fraction size 5 ml) are pooled, precipitated with ammonium sulfate at 60% saturation, and centrifuged for 30 min at 16,300o\ The precipitate is dissolved in 5ml of buffer C (5mM K- phosphate buffer, pH 7.5, containing 0.1% 2-mercaptoethanol), and dialyzed against buffer C overnight.

Chromatography on HA-Ultrogel

The dialysate is applied to an HA-Ultrogel column (2.5cm x 25cm, Sigma, Buchs, Switzerland) equilibrated with buffer C. The column is washed with buffer C and eluted with 200ml of a linear gradient of 5-500 mM K-phosphate buffer, pH 7.5, containing 0.1 % 2-mercaptoethanol. The column is eluted at a flow rate of 40 ml/h and fractions of 5ml are collected. Fractions in the flow-through containing neutral invertase activity and fractions in the eluate containing alkaline invertase activity are combined separately, precipitated with ammonium sulfate at 60% saturation and centrifuged for 30 min at 16,300^. The two protein pellets are individually dissolved in 5ml of buffer D (25mM K-phosphate buffer, pH 7.5, containing 0.1% 2-mercaptoethanol) and dialyzed against buffer D overnight.

Affinity Chromatography on Green 19 dye

The dialyzed protein solutions (5ml each) are divided into 0.5-ml aliquots, and then applied to 10 prepacked green 19 dye columns (4.5 x 0J cm, Sigma, Buchs, Switzerland) - 9 -

equilibrated with buffer D. The columns are washed with 15ml of buffer D, then step eluted with NaCI at 0.35 M and 1.5 M (15ml and 25ml, respectively), and the eluate is collected in 2ml fractions. Neutral invertase activity is detected in the flow-through, whereas alkaline invertase activity is eluted by 1.5 M NaCI. Fractions containing enzyme activity are pooled and precipitated with ammonium sulfate at 60% saturation. The precipitated proteins are collected by centrifugation for 30 min at 16,300g.

Gel-Filtration Chromatography I on Sephacryl S-300

Each protein pellet is dissolved in 7ml of buffer E (100 mM Hepes-KOH, pH 7.5, containing 0.1 % 2-mercaptoethanol). The protein solutions are individually applied to a Sephacryl S-300 column (2.6cm x 100cm, Pharmacia LKB Biotechnology, Uppsala, Sweden) equilibrated with buffer E and calibrated with blue dextran (V₀), thyroglobulin (669 kD), apoferritin (443 kD), β-amylase (200 kD), alcohol dehydrogenase (150 kD), BSA (66 kD), and carbonic anhydrase (29 kD). The column is eluted at a flow rate of 110 ml/h and fractions of 5ml are collected. Fractions containing enzyme activity are pooled, dialyzed overnight against buffer F for alkaline invertase (25mM Hepes-KOH, pH 8.0, containing 200mM NaCI and 0.1 % 2-mercaptoethanol), and buffer G for neutral invertase (25mM Hepes-KOH, pH 7.2, containing 275mM NaCI and 0.1 % 2-mercaptoethanol).

Anion-Exchange Chromatography II on Macro-Prep

For further purification of alkaline invertase, the dialysate is applied to a Macro-Prep column

(1.5cm x 20cm, Bio-Rad Laboratories, Richmond, CA, USA) equilibrated with buffer F. The column is washed with buffer F and eluted with 200ml of a linear gradient of 200-450 mM

NaCI in 25mM Hepes-KOH, pH 8, containing 0.1 % 2-mercaptoethanol.

For further purification of neutral invertase, the dialysate is applied to a Macro-Prep column

(1.2cm x 25cm, Bio-Rad Laboratories) equilibrated with buffer G. The column is washed with buffer G and eluted with 200ml of a linear gradient of 275-360 mM NaCI in 25mM

Hepes-KOH, pH 7.2, containing 0.1 % 2-mercaptoethanol.

Fractions containing the relevant enzyme activity are combined separately and precipitated with ammonium sulfate at 60% saturation and centrifuged for 30 min at 16,300g. - 10 -

Hydrophobic Interaction Chromatography on Propyl Agarose

The protein pellet with alkaline invertase activity is dissolved in 5ml of buffer H (25mM Hepes-KOH, pH 8.0, containing 1.5 M ammonium sulfate and 0.1 % 2-mercaptoethanol). The solution is applied to a Propyl Agarose column (10cm x 1.5cm, Sigma, Buchs, Switzerland) equilibrated with buffer H. The column is washed with buffer H, then eluted with 25mM Hepes-KOH, pH 8.0, containing 0.1 % 2-mercaptoethanol, and 3-ml fractions are collected. Fractions containing enzyme activity are pooled, dialyzed against 10mM Hepes-KOH, pH 8.0, containing 0.1% 2-mercaptoethanol and stored in 50% glycerol at -20°C.

Gel-Filtration Chromatography II on Sephacryl S-300

The protein pellet with neutral invertase activity is dissolved in 5ml of buffer I (100mM K-phosphate buffer, pH 7.0, containing 0.1% 2-mercaptoethanol) and applied to a Sephacryl S-300 column (2.6cm x 100cm, Pharmacia LKB) equilibrated with buffer I. Fractions of 5 ml are collected. Fractions containing enzyme activity are pooled, dialyzed against 10mM K-phosphate buffer, pH 7.0, containing 0.1 % 2-mercaptoethanol and stored in 50% glycerol at -20°C.

Example 2: Isolation of a cDNA Clone Encoding Carrot Neutral Invertase

Comparison of a partial internal tryptic peptide sequence (XNIYPDQIPPWLV, SEQ ID NO: 11 ) of the purified carrot neutral invertase with an est database identified its presence in the amino acid sequence encoded by est t.88552 from Arabidopsis (1026 bp 3'-end of an Arabidopsis cDNA). A DNA fragment encoding this peptide sequence (nucleotides 100 to 410) is isolated by PCR using the primers 5'-TCTAAGGATCTAGAAAGAGCCATTA-3' (SEQ ID NO: 12) and 5'-TTCAATTGAATTCAATATAGCTTC-3' (SEQ ID NO: 13). Amplification is achieved in a DNA Thermal Cycler (Perkin Elmer Cetus) with the following conditions: 10 cycles of denaturation at 95°C for 1 min, annealing at 40°C for 0.5 min, and elongation at 72°C for 1.5 min, followed by 20 cycles of denaturation at 95°C for 1 min, annealing at 60° for 0.5 min, and elongation at 72°C for 1.5 min. The PCR product is extracted with phenol/chloroform and after cleavage with Xba\ and EcoRI ligated into the respective sites of the E. coli plasmid pBluescript II KS (Stratagene). After amplification and purification of the plasmid, the fragment is excised, purified by agarose gel electrophoresis and - 1 1 -

electroelution, and randomly labeled with [ -³²P]ATP. The labeled DNA is used as a probe to screen a cDNA library in a lambda ZAP II vector (Stratagene) made with polyA⁺ mRNA from cells of a rapidly growing suspension culture of wild carrot (Daucus carota cv Queen Anne's Lace, W001C) and a single hybridizing clone is identified.

Example 3: Sequence Analysis of the Carrot Invertase Clone

The insert of the cDNA clone of Example 2 is ligated into the pBluescript II KS (+/-) vector (Stratagene) and both strands are automatically sequenced by the dideoxynucleotide chain- termination method. Computer-assisted analysis of DNA and protein sequences as described in Examples 2 and 3 is performed using the Wisconsin Package Version 9.0, Genetics Computer Group (GCG), Madison, Wisconsin.

Sequence comparisons are carried out with the computer program GAP, which uses the alignment algorithm of Needleman and Wunsch (J. Mol. Biol. 48; 443-453, 1970) to find the alignment of two complete sequences maximizing the number of matches and minimizing the number of gaps while allowing the introduction of gaps for optimal alignments. GAP considers all possible alignments and gap positions and creates the alignment with the largest number of matched bases and the fewest gaps. One provides a gap creation penalty and a gap extension penalty in units of matched bases. In other words, GAP must make a profit of gap creation penalty number of matches for each gap it inserts. If you choose a gap extension penalty greater than zero, GAP must, in addition, make a profit for each gap inserted of the length of the gap times the gap extension penalty. Typical values to use as a point of departure for the gap creation and gap extension penalties are 3.0 and 0.1 for protein sequence comparisons.

The cDNA clone of carrot neutral invertase is found to be 2447 nucleotides long (SEQ ID NO: 3) and contains 29 bp of 5' and 393 bp of 3' non-coding sequences. The ORF codes for 675 amino acids with a molecular mass of 75957 Dalton and a calculated isoelectric point of pi 8.01. The deduced amino acid sequence shares 80% identity (86% similarity) with the deduced amino acid sequence of the Arabidopsis est t88552.

A comparison of the deduced amino acid of the carrot cDNA clone for neutral invertase - 12 -

(car) with the sequences of the LIM17 proteins from L. longiflorum (IN) and Synechocystis (bac) (Table 1 ) identifies three conserved sequence domains (boxes 1 -3). When the sequence of box 2 is used for a database search, a protein of known function is identified, namely cellobiose phosphorylase from Clostridium stercorarium, which cleaves cellobiose [β-D-Glc(1→4)-D-Glc] in the presence of pyrophosphate into glucose 1 -phosphate and glucose. This suggests that box 2 may constitute the binding site for the glucose residues of the disaccharides.

Table 1 : Comparison of the cDNA-derived amino acid sequences of neutral invertase from carrot (car) with the amino acid sequences of the LIM17 proteins form Lilium longiflorum (HI) and Synechocystis (bac). The amino acid sequences are in one-letter-code and have been aligned by introducing gaps (..) to maximize identity. The amino acid residues in bold face indicate conserved domains (boxes 1 -3). The asterisks below the sequence mark amino acid residues identical in all three sequences.

lil RHGQRA. 60 car MNTTCIAVSN MRPCCPMLLS CKNSSIFGYS FRKCDHPM3T NLSKKQFKVY GLRGYVSGRG bac —, _. „._ , , lil PPQISARSAV DQVRNRFERP LSFWV PSW MDQGKN NTP DSGQDKPDEF DFSKL HIKP 120 car GKG GYRCGI DPNRKGFFGS GSTM3QPRVL TSGCRRVDSG QRSVLVNVAS bac ~ — ~ — — lil RVLNIDRQTS ..CDERSLLE HS GIGIIYP PLVFK PES SSRLLDHPEI VSTPGKRSAV 180 car DYENHSTSVE GHV DKSFER IYVRGGLNVK PLVIERVE.K GEKVKEEEGR VGV GSNVNI ac ~ ~~~ -~ lil NTPKAEN... YFEPHGQHEM MD.BGWDAL RSLVYFRGQP VG TAALDHS EEA..LNYNQV 240 car GDSKG NGGK VLSPKREVSE VEKEA ELLR GAWDΪCGNP VGTVAASDPA DSTPlXODQy bac MKSPQAQQI LDQARRLYE KΑ VKINSQY VGTVAAIPQS IJHHDlUIΪ EV

* * *** ** *** * lil FVKDFFP.SGL AFI.MKGEPEI VKNF RILR QJ3WEKKIDR FKLGEGAMPA SFKV HDPV. 300 car IKL»VPS.AL AFIJUJGEX333I VKNFLH Q LQS4WEKTVDC HSPGQGLMPA SFKVKNVAID bac FIRENVFV I F UζgHE^El VQNFLEIC T LQS K GFPTYGIFPT SF

* ** * * * ** * *** * *** lil RNQETT. NADEXSEBAIG KVΑFVTΛ3GFW WIHiLE-AX K STGERS Xm PDCQRGMKLI 360 car GKIGESEDIL DPDFGF.SAIG KVΛFVDSG W WIHJ^RAYTK GDYG Q^R VDVQTGIR I bac . .VETE HEL KADYGQRAIG KVCSVTtAS W WPIIAYYYVQ RαXΞNEM*B*RQ OHVQLGLQKF

* * * *** ** ** ** * ** ** * * lil TLCL.SBGFD FPT LCAEtA CX3MIDRRMSI YGYPIEIQ^ FFMAIXCR MUKQ 420 car LNCL DG D MFPT V DG SCMΣDRRM3I HGHPLEIQA FXSM-RCSRE ML. TV bac NLILHFVFR DAPTLFVPDG AFMIDRPMDV WSAP EIQTI, ΩYGΑISKSΑΑG IiLLIDLKAKG - 13 -

lxl D DEGRE AERI A...QRL<_AL .SFH R,SYFWL I3FRRTJJDIΪR K EQYSDTA 480 car N DSTKNLVAAV N...NRSAL SFHIREYYWV EMKKINEIYR YK3EEYS DA bac YCSKDHPFD SFT EQSHQF iNLSVEWLKKL RTYIiiKHYWI NCWIVQaiΛR RPTE3QYGEEA

* * * * ** * lxl HSKETJVMPDS LPDWVFDFMP TRGGYFIGNV SPAR DFRWF C GNCIAIIS NIATAEQ.SEA 540 car IMKFNIYPDQ IPSWLVDWMP E QGYLIGM QPAHMDFRFF TLGNEiWSIVS .SLGTPKQNES bac .SNEHNVHTE IPNWLQDWLG DRGGYLIGNI RTGRPDFRFF STJGNCLGAIF DVTSIAQQRS

* * * * * ** *** *** * *** * lil IMΠU^EERWP ELVGEMP*— ~~ — 600 car I1JΛ IEDKWD DLVAHMPLKI CYPALEYEEW RVTTGSDPKN TPWSYHNGGS PT LWQFT bac FFEVUKNQR ELCΛOjP RI CHPP KDDDW RSKTGFDR N PWCYHNAGH PCLFWFLW

lxl 660 car ACIK MKKP ELARKAVALA EKKLSEDHWP EYYDTRRGRF IGKQSRLYQT bac AVLRHSCHSN YGTVEYAEMG NLIRNYEV LRRPKHKWA EYFDGPTGF VGQQSRSYQT lil ₇₁₆ car WTIAGFLTSK T. T.KNPEMAS KLF EEDYEL LESCVCAIGK SGRKKCSRFA AKSQλ/V bac WΓΓVGL LVH HFTEVNPDDA M

Example 4: Steady State Levels of Neutral Invertase mRNA

Steady-state levels of neutral invertase mRNA in leaves and roots at three different developmental stages and in reproductive organs of carrot plants are determined Total RNA is prepared by the method described by Prescott and Martin (Plant Molecular Biology Reporter 4: 219-224,1987) modified by adding 20 mg of Polyclar AT (Serva) per gram of tissue before grinding in liquid nitrogen. For RNA gel blot analysis, total RNA (10 mg/lane) is separated on a 1.2 % agarose gel, containing 6% formaldehyde The northern blot is loaded with total RNA (10 mg/lane) from 4-, 10-, and 16-week-old leaves, 4-, 10-, and 16- week-old roots, and flower buds (B), flowers (F), small developing seeds (G_s), large developing seeds (Gι), and mature seeds (S). The blot is hybridized with the ³²P-labeled cDNA for neutral invertase.

Steady-state transcript levels for carrot neutral invertase are found in all organs at different stages of development with slightly higher levels in developing organs This finding suggests a more general and possibly growth-related function of the enzyme in carrot sucrose metabolism - 14 -

Example 5: E. coli Expression of Carrot Neutral Invertase

To express the cDNA clone of carrot neutral invertase in E.co/ strain JM105 (Pharmacia Biotech) the ORF is amplified by PCR using the primers 5'-CGATTTAGCAAGGTACC ATAGATATGAATAC-3' (SEQ ID NO: 14) and 5'-CTTATCCTTAAACTAGATCTCCATT AGACC-3' (SEQ ID NO: 15). Amplification is achieved in a DNA Thermal Cycler (Perkin Elmer Cetus) under the following conditions: 30 cycles of denaturation at 95°C for 1 min, annealing at 55°C for 0.5 min, and elongation at 72°C for 1.5 min. The PCR product is extracted with phenol/chloroform and after cleavage with Kpn\ and Xba\ ligated into the respective sites of the expression vector pTrc 99 A (Pharmacia Biotech). Protein biosynthesis in transformed bacteria carrying the expression vector is induced with 1 mM IPTG for approximately 16 hours, and bacteria are lysed in a small volume of 50mM potassium phosphate, pH 6.8, by 1 cycle of freezing and thawing (Johnson and Hecht, Biotechnology 12: 1357-1360, 1994). Neutral invertase activity is determined in a soluble lysate fraction as described by Lee and Sturm, 1996, supra. Briefly, 100μl of soluble extract are mixed with 700μl of water, 100μl of 0.5 M potassium phosphate, pH 6.8, and 100μl of 0.5 M sucrose, and incubated for 30 min at 37°C. Aliquots of this solution are used for the determination of reducing sugars according to Somogyi. For the determination of the pH dependence of the sucrose-cleaving activity, solutions of 0.5 M potassium phosphate with pH values between 4.5 and 8.5 are used.

Example 6: Measurement of Invertase Activity

Invertase activity is determined in reaction mixtures containing 50mM K-phosphate buffer (pH 6.8 or 8.0), 100mM sucrose and an appropriate volume of enzyme in a final volume of 1 ml. The mixture is incubated at 37°C for 30 min. The amount of reducing sugar liberated is determined according to Somogyi. Enzyme activity (units) is expressed as the amount (μmol) of reducing sugar (glucose and fructose) released per minute. Invertase activity is inhibited by high concentrations of ammonium ions, which necessitates that protein solutions prepared after ammonium sulfate precipitation are dialyzed prior to activity determination. - 15 -

From measurement taken during the purification procedure according to example 1 it can be deduced that 400g of cells express about 240 units neutral invertase activity which means 0.6 units per gramm cells. During the purification procedure of the enzyme about 95% of the activity are lost.

Activity measurement of the recombinant enzyme expressed in E. coli detects about 0.4 units of neutral invertase activity in 100μl of the E. coli extract described in example 5.

Claims

- 16 -What is claimed is:

1. DNA comprising a sequence of nucleotides which can be translated into a protein with invertase activity, wherein highest activity is observed in the range of pH 6.0 to 7.5.

2. The DNA according to claim 1 coding for a plant protein.

3. The DNA according to claim 1 comprising a sequence of nucleotides encoding an amino acid sequence selected form the group of amino acid sequences described in SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO: 6, SEQ ID NO: 7, SEQ ID NO: 8, SEQ ID NO: 9, and SEQ ID NO: 10.

4. The DNA according to claim 1 comprising a sequence of nucleotides coding for a protein as described in SEQ ID NO: 2.

5. The DNA according to claim 1 comprising a sequence of nucleotides as described in SEQ ID NO: 1.

6. A protein having invertase activity but lacking ╬▓-fructofuranosidase activity, wherein highest invertase activity is observed between pH 6.0 and 7.5.

7. A plant protein according to claim 6 comprising an amino acid sequence selected from the group of amino acid sequences described in SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO: 6, SEQ ID NO: 7, SEQ ID NO: 8, SEQ ID NO: 9, and SEQ ID NO: 10.

8. A plant protein according to claim 6 having the amino acid sequence described in SEQ ID NO: 2.

9. A method of producing DNA according to claim 1 , comprising

- screening a DNA library for clones which are capable of hybridizing to a fragment of the DNA defined by SEQ ID NO: 3, wherein said fragment has a length of at least 15 nucleotides;

- sequencing hybridizing clones;

- purifying vector DNA of clones comprising an open reading frame encoding a protein with more than 40% sequence identity to SEQ ID NO: 2

- optionally further processing the purified DNA.

10. A polymerase chain reaction wherein at least one oligonucleotide used comprises a sequence of nucleotides which represents 15 or more basepairs of SEQ ID NO: 1 or SEQ ID NO: 3.