Iron deficiency is one of the most common nutrition disorders worldwide. In addition to affecting a large proportion of infants, children and women in the developing world, iron deficiency is the only nutrient deficiency of significant prevalence in all developed nations as well. Deficiency of iron and probably zinc are highly prevalent in developing countries, where the diet is based on cereals and legumes but also in vulnerable population groups, in industrialized countries, with high requirements such as women of fertile age, infants and adolescents. In developing countries iron deficiency, due to poor bioavailability, retards normal brain development in infants and effects the success of a pregnancy by increasing premature deliveries, as well as morbidity of mother and child at or around child-birth. Zinc deficiency prevents normal child growth and greatly weakens the immune system leading to more infections. The body requires a high amount of calcium during growth. The amount of bone in later years is determined by the starting amount (the peak bone mass) and it's subsequent loss. Both are directly relevant for the subsequent development of osteoporosis.

Mild to moderate iron deficiency is often not recognized, but nevertheless may affect poorly defined parameters such as normal vigour, physical and mental endurance, and quality of life.

Iron deficiency is caused primarily by chronic blood loss. Using approximate values, inescapable loss of iron in sweat and cellular desquamation amounts to 1-2 mg per day. This is readily balanced in men by the absorption of 10% of the average dietary intake of 10-15 mg daily. In women of fertile ages, however, menstrual bleeding adds another 1 mg to the daily loss, making it more difficult to achieve an adequate balance.

Several haematologic and biochemical tests are well established for screening or diagnosis of iron deficiency individuals as well as for population based assessment.

The amount of iron absorbed from the diet at any one time is dependent on three factors: the quantity of iron, the composition of the diet and the behavior of the mucosa of the upper small bowel. Variation in the bioavailability (the portion of total iron in the diet absorbed and utilized by the organism) of food iron are of greater importance for iron nutrition than is the amount of iron in the diet. The haem iron in meat, poultry and fish is easily absorbed whatever the dietary composition whereas non-heme iron is markedly influenced by other ingredients in the diet. A number of promoters and inhibitors of iron absorption have been identified. The bioavailability of the iron in any particular diet ultimately depends on the relative quantities of promoters and inhibitors of iron absorption present in that diet.

Although the element is the second most abundant metal in the earth's crust, its low solubility makes its acquisition for metabolic use a major challenge. Most environmental iron exists as insoluble salts. Gastric acidity assists the conversion to absorbable forms, but the efficiency of this process is limited. Many plants produce

powerful chelators, such as the phytate (organic polyphosphate), which are potent inhibitors of iron absorption found in e. g. , cereals and legumes.

Zinc is a component of more than 300 enzymes needed to repair wounds, maintain fertility in adults and growth in children, synthesize protein, help cells reproduce, preserve vision, boost immunity, and protect against free radicals, among other functions.

Zinc deficiency is common in individuals or populations whose diets are low in sources of readily bioavailable zinc, such as meat, and high in unrefined cereals that are rich in phytate, a diet common in the poorer areas of the world. Vegetarian or lacto- vegetarian diets overall result in high intake of phytate, which is accompanied by a greater risk of zinc deficiency. Zinc deficiency in human caused by nutrition was first described in adolescent boys and girls in Egypt and Iran in the late 1960: s and early 1970's. The symptoms were severe growth failure, hypogonadism with delayed sexual maturation.

The cause of the deficiency was a diet based on unleavened wholemeal bread with a high content of phytate and low intake of animal protein. Treatment with zinc and an adequate intake of other nutrients improved growth and sexual maturation. Dietary zinc deficiency has later been described among the children of many countries, such as Turkey, China and Yugoslavia, . Another group at risk are pregnant women.

Other clinical manifestations of zinc deficiency than mentioned above are suppressed immunity, poor healing, dermatitis and impairments in neuropsychological functions.

The diagnosis of zinc deficiency is a problem while sensitive indices of zinc status are lacking. The most widely used indices of zinc status are levels of zinc in plasma or serum. These parameters may be decreased in cases of severe and moderate deficiency.

Dietary data and indirect measures of bone health indicate that the bioavailability of calcium is important when habitual intakes are low, especially during periods of bone

growth or loss. Bioavailability of calcium was found to be low in diets high in unrefined cereals, that are rich in phytate.

Cereals, together with oil seeds and legumes, supply a majority of the dietary protein, calories, vitamins, and minerals to the bulk population. Phosphorous, potassium, magnesium, calcium and traces of iron and other minerals are found in cereals. Barley and wheat provide 50 and 36 mg Ca/100 g respectively. Barley provides 6 mg of iron per 100 g; millet provides 6.8 ; oats, 4.6 and wheat, 3. 1.

Inositol hexaphosphate, IP6, is ubiquitous in nature and comprises the bulk of eukaryotic cell inositol phosphate content. In plants, IP6 constitutes the principal storage form of phosphorous, in particular whole grains of cereals and legumes are rich in IP6. In cereals phytate is located in bran and germ, whereas in legume seeds phytate occurs in the protein bodies in the endosperm. The highly negatively charged IP6 forms various complexes with minerals and proteins, commonly known as phytate. Phytate is considered an anti-nutrient due to the formation of precipitated complexes that strongly reduces the absorption of essential dietary minerals such as iron, zinc, calcium and magnesium. A dose dependent inhibition of iron, zinc and calcium absorption by phytate has been demonstrated in humans. Moreover, inositol penta phosphate has been identified as an inhibitor of iron and zinc absorption. In addition, it has been suggested that IP6 may influence negatively on solubility, digestibility and activity of proteins such as digesting enzymes Reddy et al. Reduction in antinutritional and toxic components in plant foods by fermentation. Food Res. Int. 27,281-290. Degradation of phytate to low levels in cereal and legume meals was demonstrated to markedly improve iron and zinc absorption in humans. To improve iron absorption the degradation has to be virtually complete. Thus, once phytate is degraded these foods become a good source of dietary minerals.

Degradation of phytate is possible with phytases. Native phytase in some cereals may be active during traditional processing such as soaking, germination, malting and fermentation and phytate degradation can be improved by process optimization, selection of specific starter cultures, or phytase can be added. It has also been demonstrated that phytate degradation in the stomach and small intestine of humans occurs as a result of activity of dietary phytase of plant or microbial origin thereby improving iron and zinc absorption. Consumption of foods containing active phytase enzymes is therefor an alternative to phytate removal during food processing. The plant phytase is less stable than the microbial phytase at the physiological conditions of the gastro-intestine and is therefor less effective in this respect.

Different feedstuffs having an improved phytase activity are available containing phytase derived from Aspergillus species, in order to reduce the need of inorganic phosphorous addition, thereby reducing overload of phosphorous present in faeces from animals, (e. g., EP-B 1-0 649 600).

Since IP6 is such a common compound in nature some microorganisms would be expected to have the ability to degrade IP6 and utilize the hydrolyzed phosphorous. This is true for certain bacteria, Riedereret al. (1991) Removal ofN-glycosylation sites of the yeast acidphosphatase severely affects protein foldiyzg. J. Bacteriol. 173,3539-3546 and many fungi, Shieh et al. (1968) Survey of microorganism for the production of extracellular phytase. Appl. Microbiol. 16, 1348-1351 ; Dvorakova et al. (1997) Characterization of phytase produced by Aspergillus niger. Folia Microbiol. 42, 349-352 ; Wyss et al. (1998) Comparison of the thermostability properties of three acid phosphatases fro7n molds : 24spergillus fumigatus phytase, A. niger phytase, and A. niger pH 2. 5 acid phosphatase. Appl. Environ. Microbiol. 64, 4446-4451, that are well known to synthesize secretory so called phytases, i. e. phosphatases hydrolyzing IP6 to inositol pentaphosphate (IP5) and inorganic ortho-phosphate (Pi). Phytases, derived from

Aspergillus sp are frequently used in animal feeds to improve phosphorous and mineral availability, and much research is devoted to further improve these by e. g. molecular enzyme engineering, Wyss et al. (1999) Biophysical characterization offurzgalphytases (nzyo-i7iositol hexaldsphosphate phospholiydrolases) : molecular size, glycosylatioll pattern, and engineering ofproteolytic resistance. Appl. Environ. Microbiol. 65,359-366.

For humans, however, no corresponding approved (food grade) enzyme is available. One way to improve the mineral state in vulnerable human populations may be to explore yeasts, that with their GRAS- (generally regarded as safe) status are preferable microorganisms in human foods. The natural yeast phytase activity during for instance bread leavening, however, seems too low to significantly influence the iron absorption when eating foods fermented by yeast, such as bread, Turk et al. (1996) Reduction in the levels of phytate during wholemeal bread making ; E/0fect of yeast and wheat phytases. J.

Cereal Science. 23,257-264.

It is further noted that it is known from the patent literature that yeasts, such as Pichia rhodanensis (JP2000050864), Arxula adeninivorans (JP2000050863), Candida boidinii (EP 0 931 837), Saccharomyces cerevisiae (W02001036607) may comprise genes for expression of phytase.

Greiner et al, J. Agric. Food Chem. (2001) 49 (5), 2228-2233 relates to production of stereospecific myo-inositol hexaphosphate by using baker's yeast phytase.

Nakamura, et al, Biosci. Biotechnol. Biochem (2000), 64 (4), 841-844 describe dephosphorylation of inositol using baker's yeast.

Turk et al, J. Agric. Food Chem. (2000), 48 (1), 100-104 describes that phosphorous is stored in plants as phytates, and discuss the negative effect on mineral bioavailability and shows the ability of two commercial strains of Saccharomyces cerevisiae to reduce phytate by its phytase production. Turk et al express a need for more

efficient phytase to be screened for or developed, although there is only evidence that phytase produced utilizes phosphorous from inositol when there is a lack of other phosphorous present.

Moore, E. , et al, J. Ind. Microbiol. 14 (5): 396-402 (1995, May) discuss molecular cloning, expression and evaluation of phosphohydrolases for phytate degrading activity whereby four acid phosphatase genes, pho3, pho5, and phol 1 from Saccharomyces cerevisiae, and a phosphate repressible acid phosphatase from Aspergillus niger were cloned by PCR and inserted into a A. 07Syzae, A07. At culturing the A. oryzae thus transformed a two to six fold increase of phytase activity was monitored. The enzymes so produced were included into poultry diets. An available phosphorous increase was obtained. Thus Moore et al propose transformation of phosphohydrolase expressing genes from Saccharomyces cerevisiae to an Aspergillus specie in combination with an A. niger gene. There is no proposal to overexpress such phosphohydrolase producing genes in Saccharomyces cerevisiae, although the genes were picked from S. cerevisiae.

Ogawa, N. , et al, Mol. Biol. Cell, 11 (12): 4309-21, (2000, Dec) discuss new components of a system for phosphate accumulation and polyphosphate metabolism in Saccharomyces cerevisiae revealed by genomic expression analysis, whereby it is stated that the PHO regulatory pathway is involved in the acquisition of phosphate in the yeast. When extracellular phosphate is low several genes are transcriptionally induced by the pathway, including pho4, pho80, pho85. There is no indication or proposal of overexpressing any of the pho family genes or deleting any of these to facilitate phytase activity on inositol under all conditions.

The PHO gene family has been extensively studied in Saccharomyces cerevisiae, Yoshida et al. (1989) Function of the PHO regulatory genes for repressible acid

phosphatase synthesis in Saccharomyces cerevisiae. Mol. Gen. Genet. 217,40-46 ; Ogawa et al. (2000) New co7nponents of a system for phosphate accumulation and polyphosphate metabolism i7l Saccharomyces cerevisiae revealed by gelzomic expression analysis. Mol.

Biol. Cell. 11,4309-4321, however, not in the context of IP6 as the substrate. The secretory acid phosphatases encoded by the structural genes PH03, PH05, PH010 and PH011 all seem to be aimed for hydrolyzing extracellular organic phosphorous compounds allowing the yeast to grow in the absence of inorganic phosphate. All except PH03 are repressed by inorganic phosphate, whereas PH03 is repressed by thiamine present in the external environment, Praekelt et al. (1994) Regulation of THI4 (MOL1), a thiamine-biosynthetic gene of Saccharomyces cerevisiae. Yeast. 10, 481-490. PH03 is therefore suggested to primarily hydrolyze phosphate from thiamine phosphate or thiamine pyro-phosphate ; Nosakaet al. (1989) A possible role for acid phosphatase with thiamin-binding activity encoded by PH03 in yeast. FEMS Microbiol. Lett. 51,55-59 ; Nosaka (1990) High affinity of acid phosphatase encoded by PHO3 gene in Saccharomyces cerevisiae for thiamin phosphates. Biochim. Biophys. Acta. 1037,147- 154 PH05 is often described as responsible for the major fraction of secretory acid phosphatase activity whereas PH010 and PHOI I encode a minor fraction of secretory phosphatase, Rogers, D. T. , Lemire, J. M. and Bostian, K. A. (1982) Acid phosphatase polypeptides in Saccharomyces cerevisiae are encoded by a differentially regulated multigene family. Proceedings of the National Academy of Sciences of the United States of America. 79,2157-2161 ; Lemire et al. (1985) Regulation of repressible acid phosphatase gene transcription in Saccharomyces cerevisiae. Mol. Cell Biol. 5, 2131- 2141, that may be lowly and constitutively expressed. In addition to the enzymes, several components within the PHO system are regulatory proteins, such as Pho4p and Pho2p

which are transcriptional activators for PH05, PH010 and PH011, Vogel et al. (1989) The two positively acting regulatory proteins Pho2p and Pho4p physically interact with PH05 upstream activation regions. Mol. Cell. Biol. 9,2050-2057 A single study approaches the question of level of phytase activity derived from specific S. cerevisiae phosphatase encoding genes, Moore et al. (1995) Molecular cloning, expression and evaluation of phosphohydrolases for phytate-degrading activity. J. Ind.

Microbiol. 14,396-402. In that work, however, the yeast genes were all expressed in Aspergillus leading to uncertainty in factors such as copy number, levels of glycosylation and secretion as well as to what extent the genes contribute to phytase activity in S. cerevisiae. The opposite approach was taken by Han et al, Han et al. (1999) Expression of an Aspergillus niger phytase gene (phyA) in Saccharomyces cerevisiae. Appl. Environ.

Microbiol. 65,1915-1918, who instead expressed the Aspergillus 7liger phytase gene (phyA) in S. cerevisiae, to explore an alternative expression system for a well-known phytase.

SUMMARY OF INVENTION Broadly, the invention includes a process for preparing a Saccharomyces cerevisiae yeast strain having improved phytase activity comprising modifying the strain by inserting one or more of the PHO5, PHOIO, PHOll, DIA3, PH02 and/or PH04 genes, and/or by deleting one or more of the negative regulatory genes PH080, PHO85, a7ld/orPHO23, whereby the insertion is either chromosomal or by transforming the strain using a plasmid containing the gene/s. The modified strain overxpresses one or more the PHO5, PHOlO, PHOll, DIA3, PHO2 and/or PH04 genes.

A preferred embodiment encompasses Process for preparing a Saccharomyces cerevisiae yeast strain having improved phytase activity, characterized in that the strain is

modified by inserting one or more of the genes PHOS, PHO10, PHO11, PHO2, PHO4, PHO23, and/or DIA3 to overexpress said genes, and by deleting one or more of the regulatory, repressor genes PH080 and/or PHO85, whereby the insertion is either chromosomal or by transforming the strain using a plasmid containing said gene/s.

In a preferred embodiment the modified Saccharomyces cerevisiae overexpresses PHOS.

In another preferred embodiment the modified Saccharomyces cerevisiae overexpresses PHO10.

In a further preferred embodiment the modified Saccharomyces cerevisiae overexpresses PHOll.

In another preferred embodiment the modified Saccharomyces cerevisiae overexpresses PHO2.

In another preferred embodiment the modified Saccharomyces cerevisiae overexpresses PHO4.

In another preferred embodiment the modified Saccharomyces cerevisiae overexpresses DIA3.

In a further preferred embodiment of the invention the modified Saccharomyces cerevisiae is a strain, wherein the negatively regulating gene PHO80 has been deleted to form a mutant pho80#.

In another preferred embodiment the modified Saccharomyces cerevisiae overexpresses PH02 and PHO80 is deleted.

In another preferred embodiment the modified Saccharomyces cerevisiae overexpresses PH04 and PHO80 is deleted.

In another preferred embodiment the modified Saccharomyces cerevisiae overexpresses PHO5 and PHO80 is deleted.

In another preferred embodiment the modified Saccharomyces cerevisiae overexpresses PHO10 and PH080 is deleted.

In another preferred embodiment the modified Saccharomyces cerevisiae overexpresses PHO11 and PH080 is deleted.

In another preferred embodiment the modified Saccharomyces cerevisiae overexpresses PH023 and PH080 is deleted.

In another preferred embodiment the modified Saccharomyces cerevisiae overexpresses DIA3 and PH080 is deleted.

In another preferred embodiment the modified Saccharomyces cerevisiae overexpresses PH02 and PH05 and PH080 is deleted.

In another preferred embodiment the modified Saccharomyces cerevisiae overexpresses PH05 and PHO10 and PH080 is deleted.

In another preferred embodiment the modified Saccharomyces cerevisiae overexpresses PH04 and PH05 and PHO80 is deleted.

In another preferred embodiment the modified Saccharomyces cerevisiae overexpresses PH04 and PHO10 and PH080 is deleted.

In another preferred embodiment the modified Saccharomyces cerevisiae overexpresses PH05, PHO10 and PHO11 and PH080 is deleted.

In order to further increase phytase activity, other genes belonging to the PHO family like the genes encoding for phytase or some transcription regulators, such as PH02 and PH04, may eventually have to be over-expressed alone or in combination with other genetic modifications, such as pho80 or pho85 mutations.

In another preferred embodiment of the invention the modified Saccharomyces cerevisiae is a strain, wherein the negatively regulating gene PHO85 has been deleted to form a mutantpho85D.

In a further preferred embodiment of the invention the modified Saccharomyces cerevisiae over expresses the positively regulating gene PH02.

In a further preferred embodiment of the invention the modified Saccharomyces cerevisiae over expresses the positively regulating gene PH04.

In accordance with a further aspect of the invention, the invention includes a fermented product having been fermented using a Saccharomyces cerevisiae strain identified above, and having a reduced content of inositol hexaphosphate.

In one embodiment the fermented product is a bakery product.

In accordance with another embodiment the fermented product is a gruel product, preferably for infants.

In a further embodiment the fermented product is a brewed product.

In yet another aspect of the invention, a method for the preparation of phytase is provided wherein one or more of the strains identified above are grown in a growth medium that facilitates phytase expression, whereupon produced phytase is isolated.

In accordance with a further aspect of the invention such a phytase preparation can be used as an additive in foodstuffs or feedstuffs to improve the utilization of minerals.

In accordance with a further aspect of the invention such a phytase preparation can be used to produce specific isomers of lower inositol phosphates In yet another aspect of the invention, the invention provides a method for the preparation of phytase wherein one or more of the strains identified above are grown in a medium under conditions optimized for phytase expression whereupon lower inositol

phosphates are produced as a result of yeast phytase activity on added inositol hexaphosphate.

Besides fermented products for human use the present Saccharomyces cerevisiae strains can be used for the production of animal feedstuffs to increase mineral uptake, and to reduce the phosphate amounts in faeces. Phosphorous in the form of IP6 can not be resorbed in the intestines and thus animal feedstuffs are enriched using inorganic phosphates. IP6 together with excess of phosphates follow the faeces, which leads to high amounts/concentrations of phosphorous in the manure.

Besides the foodstuff and feedstuff applications the enzyme phytase can be used for the manufacture of specific inositol phosphates for pharmacological use. One example is 1,2, 6IP3, which may stimulate the release of insuline, to reduce thrombocyte aggregation and to have anti-inflammatory properties. Such inositol phosphates may be developed into pharmaceuticals.

The enzyme phytase can also be added to foodstuffs and feedstuffs to obtain the effects specified above. Thus a further aspect of the present invention is to grow the modified Saccharomyces cerevisiae and to isolate the phytase enzyme produced. Such isolation is readily carried out as the phytase is present extracellularly. Primarily the addition of phytase is made to improve availability of iron and zinc to the body.

The invention also includes a modified strain of Saccharomyces cereviaisae exhibiting increased phytase activity, the strains being designated as YM-p5 and YD80.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a graph depicting the growth of S. cerevisiae SKQ2n (squares) and extracellular concentration of inositol hexaphosphate (IP6; triangles), measured by HPIC, as a function of time;

FIG. 2 is a graph depicting the chromatographic profiles (HPIC) of chemical acid hydrolysate of sodium phytate (first line), and supernatant from S. cerevisiae cultures (second and third lines) after 15 h of growth; FIGS. 3A and 3B are graphs depicting the growth (squares) of S. cerevisiae parent strain YS18 (open symbols) and S. cerevisiae YMR4 pho5Spho3X double deletion mutant (filled symbols) and extracellular concentration of IP6 (circles) as determined by HPIC; FIG. 4 is a graph depicting the concentration of inositol hexaphosphate (IP6) in yeast culture supernatant as a function of growth in CBS medium with IP6 as the sole phosphorous source; FIG. 5 is a graph depicting the concentration of inositol hexaphosphate (IP6) in yeast culture supernatant as a function of growth in complex YPD medium with IP6 as the added phosphorous source ; FIG. 6 is a vector construct consisting of plasmid pYX212 containing the insert PH05 used for transformation of YS 18 and YMR4; FIGS. 7A and 7B are graphs depicting the extracellular concentration of IP6 as a function of growth time of S. cerevisiae YMR4 (A) and YS 18 (B), with and without the PH05 containing plasmid pYX212; FIG. 8 is a graph depicting a direct comparison of phytate degrading capacity between PH05 overexpressing strains, and pho80d and pho85d deletions mutants; FIG. 9 is a graph depicting phytate degrading capacity of pho2Z1 deletion mutant; FIG. 10 is a graph depicting the extracellular concentration of para-nitrophenyl phosphate (pNPP, squares) and IP6 (triangles) as a function of growth of S. cerevisiae SKQ2n in CBS medium with 2% glucose as carbon and energy source;

FIG. 11. is a graph depicting the content of inositol hexaphosphate (IP6, triangles) and inorganic phosphate (Pi ; squares), expressed as Amol per gram wet dough, as a function of leavening time in two types of wheat based dough; and FIG. 12 is a graph depicting phytate degradation capacity of the phoNOzI deletion mutant SD80 (YS 18 without PH080) compared with S. cerevisiae YS 18 in YPD supplemented with 0.25 mM IP6 and 30 mM Pi.

DETAILED DESCRIPTION OF INVENTION The invention will now be described in reference to the following non-limiting examples.

Materials and methods Organisms Saccharomyces cerevisiae SKQ2n, Bataille et al. (1987) Identification of polypeptides of the carbon metabolism machinery on the two-dimensional map of Saccharomyces cerevisiae. Location of 23 additional polypeptides. Yeast. 3, 11-21,. S. cerevisiae YS 18 (MATa ; his3-11, 3-15 ; leu2-3, 2-112; ura3A5 ; canR) and S. cerevisiae YMR4 were used in most experiments as indicated. YS18 and YMR4 were previously constructed by Riederer and Hinnen ; Riederer et al. (1991) Removal of N-glycosylation sites of the yeast acid phosphatase severely affects protein folding. J. Bacteriol. 173, 3539-3546. YS18 and YMR4 were in the present work both transformed with a PH05 containing plasmid (see below) and the resulting transformant strains were designated YS-p5 and YM-p5, respectively. Strain YS-p5 was deposited with DSMZ-Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH, Mascheroder Weg lb, 38124 Braunschweig, Germany on April 18,2002 and was assigned accession number DSM-No. 14929.

YS18 and YMR4 were further deleted for PHOI 0 yielding strains named YS10 and YM10, respectively. YS18 was also used for construction of the pho80A strain labeled YD80. In addition, pho80A, pho851N andpho2Z1 mutant strains were retrieved from the German EUROSCARF collection. pho80 (YOL001w), pho85 (YPL031c) and pho2 (YDL106c) strains originates from the laboratory BY4741 strain (MATa ; his3Al, leu2#0, met15#0, ura3AO). The PHO80, PH085 and PHO2 genes were deleted with kan::MX4 disruption cassette.

Yeast cultures were maintained on YPD-plates (yeast extract 10g, peptone 20 g, glucose 20 g and agar 20 g in one liter of water) and stored long-term in glycerol at-70°C.

Escherichia coli DH5α was used for propagating the plasmid containing the inserted PHO5 (see below). The yeast strains and their expression profiles of relevant genes are listed in Table 1. Table 1. Strains of Saccharomyces cerevisiae used in this study, their relevant genotype and expression profiles of secretory phosphatases in repressing and non-repressing media. Secretory phosphates aSecretory phosphates Strain Relevant expressed in IPs expressed in IP6 + Pi Reference for genotype strain medium medium all PHO genes Bataille et al. SKQ2n Pho5p, Pho10p, Pho11p none intact (1987) all PHO genes YS18 intact. Parent Pho5p, PholOp, Phollp none (1991) (1991) strain for YMR4 Riedereret et al. YMR4 pho5,3::ura#1 Pho10p, Pho11p none (1991) pho5,3::ura#1 Pho5pb, Pho10p, Pho5pb This work pYX-PH05 Phollp YS-p5 pYX-PH05 PhoSp, Pho5p, PholOp, Pho5pb This work Poll YM10 pho5, 3 ::ura3#a Pho11p none This work pholO : : 11is3A YS10 pholO : : his3S Pho5p, Phol lp none This work BY4741 AllPHO genes pho5p, PholOp, Phollp none EUROSCARF intact YOL001w pho80# Pho5p, Pho10p, Pho11p Pho5p, Pho10p, Pho11p EUROSCARF YPL031c pho85# Pho5p, Pho10p, Pho11p Pho5p, Pho10p, Pho11p EUROSCARF YDL106c p/ ! p2AnonenoneEUROSCARF YD80 ho80# : : hphPho5p, PholOp, Phol lp Pho5p, PholOp, Phol lp This work

a) A small constitutive expression may occur b) On the plasmid pYX212 under the control of TP 11 promoter Growth media and culture conditions The growth medium used in most experiments was a modified version of the defined yeast minimal medium developed by CBS (Centralbueau voor Schimmelcultures, Delft, the Netherlands) which has been described previously, Albers et al. (1996) Influence of the nitrogen source ofa Sacchaomyces cerevisiae anaerobic growth and product formation. Appl. Environ. Microbiol. 62, 3187-3195 using ammonium sulfate (7.5 mg/ml) as nitrogen source. As phosphorus source, either IP6 in the form of sodium phytate (C6H6 (OPO3Na2) 6: 0.5 or 0.25 mg/ml), inorganic phosphate (KH2PO4 : 3.5 mg/ml), para-nitrophenyl phosphate (pNPP ; to a final concentration of 2.25 or 0.25 mM), or combinations of these were used. Depending on the phosphorous source used the media were designated CBSIP6, CBSPi, CBSIP6+Ps and CBSpNPP. CBSIP6 and

CBSpNPP were supplemented with KC1 (3 mg/ml) to compensate for the potassium present in the inorganic P-source (KH2PO4), but absent in the organic P-sources. In all experiments glucose (either 2% (w/v) or 1% (w/v) as indicated) was used as carbon and energy source. When appropriate, histidine, leucine and uracil were supplied at 120 mg/1.

Most experiments were also performed in YPD media (per liter: yeast extract, 10 g; peptone 20 g and glucose 20 g) supplemented with the appropriate P-sources as indicated in results and figure legends. Succinic acid/NaOH, pH 5.3, was used as buffer in the CBS based media. Medium components were autoclaved, with the exception for IP6, pNPP, vitamins, growth factors and FeCl2, which were sterilized by filtration through a 0. 2 Ilm filter. Experimental cultures containing 100 ml in 250 ml E-flasks were inoculated with primary cultures grown for approximately 20h in medium of the same composition as the respective experiment culture. The inoculation level was set to OD610 = 0. 2. The experimental cultures were grown at 30°C in a rotary shaker set to 210 rpm. The growth was monitored as optical density at 610 nm (OD610) using a spectrophotometer (Hitachi, model U-100). Experiments aimed at PH03 expression were performed in CBS medium as described above with the exception of excluding thiamine (to avoid repression of the thiamine phosphatase PH03), with and without addition of thiamine monophosphate.

Construction of PH05 overexpressing strain.

Chemicals and enzymes for the procedures were purchased from New England Biolabs Inc., USA, and the PCR primers used for amplification of PH05 were purchased from Life Technologies AB, Taby, Sweden. Extraction of genomic DNA from S. cerevisiae YS18 was carried out according to standard procedures. The nucleotide sequences of the primers are shown in upper case letters and the added restriction enzyme sites are shown in lower case letters.

Primers for overexpression of PH05 - Forward primer: 5' : CGGAATTCATGTTTAAATCTGTTGTTTATTC, appended as SEQ ID NO: 1 - Backwards printer : 3' : CGCTCGAGCTATTGTCTCAATAGACTGGC, appended as SEQ ID NO: 2 Underlined sequences are EcoRI and Aval (yields end compatible to XhoI used to cut the pYX212 plasmid) restriction sites, respectively. The remaining sequences complement the beginning (from ATG) and end of PH05.

The PCR reaction was performed in 100 ul reaction mixtures, using VENT polymerase and was carried out according to standard procedures.

The PCR produced PH05 was purified with a PCR purification kit (Qiagen, Cat, No. 28104, Merck Eurolab AB, Sweden) according to the manufacturer's protocol and 50 ul was cut with Aval and EcoRI in EcoRI buffer for 12h at 37°C. The cut PCR product was loaded on a preparative agarose gel and run together with a DNA ladder in TAE buffer. The appropriate band was cut out and purified using DNA gel extraction kit (Qiagen, Cat. No. 28704). A 20 p1 ligation mixture was prepared by mixing 16 µl purified insert (PH05), 2 ul T4 DNA ligase buffer (lOX) 1 . l ligase and 2 Ill vector DNA (pYX212). The plasmid pYX212 containing the selection markers AmpR for E. coli and MM3 for S. cerevisiae had previously been cut with EcoRI and XhoI (yields ends compatible with AvaI). Two ul of the DNA construct was added to 40 fil of cold competent E. coli DH50e and transformed by electroporation using a Gene Pulser II (Biorad) set to the 25 llE capacitator, 2.5 kV and the pulse controller to 200 Q. Cells were plated on NB plates containing 100 . g/ml ampicillin. Clones were cultivated over night in liquid NB medium and plasmids were prepared according to standard procedures.

Sequence of the plasmid pYX212 with insert PHO5:

li appended as SEQ ID NU : 3.

Underlined sequences show sites for primers Bolded nucleotides represent the inserted PH05 The presence of PH05 was verified by PCR using the same primers as described above as well as by cutting with EcoRV followed by gel electrophoresis, then transformed into yeast strains YS 18 and YMR4 using a standard LiOAc S. cerevisiae transformation protocol. Yeasts containing the vector with insert were selected for on uracil negative

YNB plates (DIFCO, USA) containing the appropriate amino acids. The resulting transformant strains were named YS-p5 and YM-p5, respectively.

Construction of deletion strains The PH010 deletions were generated by PCR-mediated gene replacement, Baudin et al.

(1993) A simple and efficient for direct gene deletion in Saccharomyces cerevisiae.

Nucleic Acids Res. 21, 3329-3330using HIS5 from Saccharomyces pombe (corresponding to HIS3 in Saccharomyces cerevisiae) as the selectable marker. As template DNA the plasmid pFA6a-HIS3MX6; Wach et al. (1994) New heterologous modules for classical or PCR-based gene disruptions in Saccharomyces cerevisiae. Yeast. 10, 1793-1808was used.

Primers for deletion of PHOIO <BR> <BR> <BR> <BR> <BR> <BR> 5' :<BR> <BR> <BR> CGATAGATTCAAGCTCAGTTTCGCCTTGGTTGTAAAGTAGGCAGCTGAAGCTT CGTACGC-3', appended as SEQ ID NO: 4 5' : GGTCTATTTACTGTTTTAATAAAGTGTCGTTGTAGTGCTTGGGGCAGATGATG TCGAGGCG-3', appended as SEQ ID NO 5 These oligonucleotides were used as primers for construction of deletion cassettes.

Underlined sequences are complementary to HIS5 in the template, and non-underlined parts to flanking regions of PH010. The PHOIO deletions were generated by PCR- mediated gene replacement using HIS5 from Saccharomyces pombe (corresponding to HIS3 in S cerevisiae) as the selectable marker. As template DNA the plasmid pFA6a- HIS3MX6 was used.

Sequence of the Resulting PCR product used for deletion of PH010 by homologous recombination: <BR> <BR> <BR> <BR> <BR> <BR> CGATAGATTCAAGCTCAGTTTCGCCTTGGTTGTAAAGTAGCCAGCTGAAGCTTCGT ACGCTGCAGGTCGACGGATCCCCGGGTTAATTAAGGCGCGCCAGATCTGTTTA

appended as SEQ ID NO: 6 Underlined sequences represent primers (or complementary strand to primer) used Italic shows sequence corresponding to PHOIO (SGD) The PCR reactions were performed in 100 u. l using VENT polymerase, the crude PCR product was purified by cutting the appropriate band from a preparative agorase gel- electrophoresis and YMR4 and YS 18 were transformed by the lithium acetate method resulting in strains named YM10 and YS10, respectively. The deletions were verified by PCR on whole yeast colonies as well as on extracted DNA.

Deletion of PH080 was performed by PCR-based gene disruption, mainly according to Baudin et al., (1993) "A simple and efficient for direct gene deletion in Saccharomyces cerevisiaw."Nucleic Acids Res. 21,3329-3330, and Goldstein and McCusker, (1999) "New heterologous modules for classical or PCR-based gene disruptions in Saccharomyces cerevisiae."Yeast, 10,1793-1808. A deletion fragment

consisting of a selectable marker flanked by 5'and 3'flanking sequences of PH080 was constructed. At transformation, the PH080 ORF was exchanged for hygromycin B phosphotransferase (/2ph) by homologous recombination. The plasmid pAG32 Goldstein et al. , containing the selectable marker hph (hygromycin B phosphotransferase) was used as template for the PCR based construction of the PH080 deletion fragment. For the PH080 region, primers were designed according to SGD (http ://genome- www. stanford. edu/Saccharomyces/) and for the hph region according to Gritz and Davis (1983) "Plasmid encoded hygromycin B resistance: the sequence of hygromycin B phosphotransferase gene and its expression in Escherichia coli and Saccharomyces cerevisiae."Gene. 25, 178-188..

Deletion of PH080 -Forwa1^dpri77zer : 5':CAGCGTATATTGGCTTTCCTTTAATCTAATGCCCCAAGCCCACAT ACGATTTAGGTGACAC-3'APPENDED AS SEQ ID NO: 10 - Backwards primer: 5'-GGAGTTCTCAAGCTCATCTCGAAGTGTTTTCTGTCGCTTATGAATACG ACTCACTATAGGGTG-3'APPENDED AS SEQ ID NO: 11 Underlined sequences complement 17ph in pAG32 (Goldstein and McCusker, 1999, Yeast 15: 1541-1553) The rest are deleting sequences that complement regions flanking PH080 (SGD). The plasmid pAG32 (Goldstein and McCusker, 1999), containing the selectable marker hygromycin B phosphotransferase (holz), was used as template. FW primer complements to non-coding strain. Starting at-483 bp upstream hph start (ATG) in pAG32 and 315 bp upstream PHO80 start. BW primer complements to coding strand. Oligo. (63 bp. ) starts 311 bp downstream hph end in pAG32, and 267 bp downstream PH080 end in S. cerevisiae.

The PCR mix consisted of Thermopol buffer, 0. 1 µg/µl acetylated BSA (Bovine Serum Albumin), 0.2 mM dNTPs, 0.5 µM of each primer and 2 units of VENT polymerase. To each reaction approximately 0,7 llg/ml template DNA was added.

Following the PCR reaction, products were separated by agarose gel electrophoresis (0,7 % agarose in TBE buffer. 100V) and purified by QIAquick Gel Extraction Kit (QIAGEN).

Resulting PCR product used for deletion of PH080 by homologous recombination:

Nucleotides in bold show the ORF of the lagla gene. Underlined sequences shows the PCR primers, partly complementary to the PHONO gene The purified deletion fragment was transformed into YS 18 by electroporation. The protcol for transformation was mainly based on a protocol at Gottschling Lab website (http : //www. there. org/gottschling/yeast/ytrans. html). Transformants were incubated in 30°C in 1 ml 1M sorbitol + 1 ml 2x YPD for 2h to allow expression of the drug resistance marker, then spread onto selective growth medium (YPD, 0.9 mg/ml hygromycin B).

Colonies appeared after a few days and putative transfonnants were tested for accuracy by PCR. The resulting strain was designated YD80 (MATa ; his3-11, 3-15 ; leu2-3, 2-112 ; ura3#5 ; can7 ; pho80#::hph) 2.5. Construction of combined pho800 deletion and PH04 and PH05 overexpressing strains.

The plasmid pYX212 containing the insert PH05 (described above) was transformed into strain YD80 using a standard LiOAc S. cerevisiae transformation protocol. Furthermore, PH04 was PCR cloned as described above using the primers : Primers for overexpression of PH04 -Forward pri7ner (JP42) 5'-CGGAATTCATGGGCCGTACAACTTCTGAGG-3', APPENDED AS SEQ ID NO : 7 - Backwards primers (JT42) 5'-CGCTCGAGTCACGTGCTCACGTTCTGCAG-3', APPENDED AS SEQ ID NO: 8

Underlined sequences complement EcoRI and XhoI restriction sites in pYX212 and the rest are sequences that complement start and end regions in PH04 (SGD).

Sequence of pYX212 containing the insert PH04 :

ACACAiACAU Appended as Seq 11) : 9 Bolded sequence shows PH04 Underlined sequences represent sites for primers.

Also pYX212 with PH04 was transformed into YD80 using the methods already described.

Sample preparation, inositol phosphates and pNPP 1.5 ml of cell suspension was withdrawn at intervals during the growth and centrifuged for 3 minutes at 13000 rpm to pellet the cells. One ml of the supernatant was

transferred to a new micro-tube and acidified by adding 150 1ll of 4 M HC1. This yields approximately 0.5 M HC1, which is enough to inactivate all enzymatic activity potentially left in the supernatant. Chemical acid hydrolysis of inositol phosphates does not take place in such solutions without strong heating. These samples were stored in freezer until use for analysis of IP : s and pNPP. A reference sample for identification of peaks was prepared by chemically hydrolyzing 1 mM sodium phytate (Na-IP6) in 2 ml of 0.5 M HC1. The solution was heated to 110°C for 15 h yielding a mixture of isomers of IP : s that was stored in freezer until used as reference sample.

Bread dough making, phosphate extraction and analysis To analyze the Pi and IP6 content in a typical environment, baker's yeast are exposed to during bread leavening two simple doughs were mixed, in which two types of flour were used: whole meal wheat flour (100% extraction rate; i. e. the proportion of the whole wheat grain obtained as finished flour), and normal wheat flour (extraction rate 60%). The dough mixture contained: 300g flour, lOg sucrose, 3g NaCl, 20g commercial baker's yeast (Swedish Yeast Company) and 200g tap-water. At time zero and thereafter every 15 min, 20 g of dough was withdrawn and mixed with 40 ml 0.5 M HC1 (at room temperature this HC1 concentration will not hydrolyze IP6, but is sufficient to inactivate enzymatic activity). Inorganic phosphate (P ;) and IP6 were extracted by magnetically stirring the mixture for three hours at room temperature. The extracts were centrifuged, the supernatant collected and frozen until use. For analysis, the samples were thawed and 1 ml was centrifuged in a micro-tube for 5 min at 13000 x g. The resulting supernatants were appropriately diluted with mQ-water and inositol phosphates were analyzed by HPIC chromatography as described in previous paragraph. The ion chromatographic analysis of Pi were performed using a Dionex (Sunnyvale, CA, USA) model 4500i

equipped with a 50 p. 1 loop injector, PAX-100 guard and analytical column, Anion Micro Membrane Supressor and a conductivity detector. The samples were eluted with a linear gradient of water and increasing portion of 200 mM NaOH, starting with 6 % NaOH and ending after 35 min at 50 % NaOH.

Results Expression and repression of phytase activity in wild-type yeasts IP6 is efficiently degraded by yeast in a synthetic medium with IP6 as the sole phosphorous source. Turk et al. (2000) Inositol hexaphosphate hydrolysis by baker's yeast. Capacity, kinetics, and degradation products. J. Agric. Food Chem. 48, 100-104 In order to test whether extracellular IP6 hydrolysis by S. cerevisiae was repressed by inorganic phosphate (Pi), synthetic yeast minimal medium containing Pi (26 mM) was supplemented with IP6 (CBSIP6+Pi). For all S. cerevisiae tested, that is SKQ2n, CBS 7764, CBS 7765, YS 18, BY4741 and YMR4 virtually no breakdown of IP6 was detected in the IP6+Pi cultures, showing an efficient repression of phytase activity by high levels of extracellular Pi. Referring to FIG. 1, growth of S. cerevisiae SKQ2n (squares) and extracellular concentration of inositol hexaphosphate (IP6; triangles), measured by HPIC, as a function of time is shown. Growth was performed in synthetic CBS medium with 2 % (wt/vol) glucose as carbon and energy source. Filled symbols: Inositol hexaphosphate (IP6) as the sole phosphorous source; open symbols: inositol hexaphosphate plus inorganic phosphate (Pi) as combined phosphorous sources. Data are from a representative experiment performed many times. Error bars on IP6 data show maximum variation between repeated HPIC analysis. The typical pattern for the IP6 concentration in the medium during growth in the absence and presence of Pi is depicted for strain SKQ2n in FIG. 1.

In this experiment, all the IP6 was depleted before 20h of growth in the CBSIP6 culture, while in the CBSIP6+Pi culture no IP6 was degraded at that time. For all strains, the growth rates, as calculated by regression in the exponential respiro-fermentative phase (glucose growth), were as rapid in CBSIP6 medium as compared with CBSIP6+Pi or CBSP ; media, showing efficient utilization of IP6 as P-source for growth by S. cerevisiae.

The same type of experiments were performed also in complex and rich YPD medium, either with the addition of only IP6 (YPDIP6) or with the addition of IP6 and Pi (YPDIP6+Pi). The YPD experiments yielded data consistent with the CBS cultures, that is IP6 degradation without loss in growth rate in YPDIP6 medium and repression of phytase activity in YPDIP6+Pi medium (data not shown). However, the Pi induced repression was for some strains (e. g. YS18) less pronounced in YPD as compared with CBS medium, that is a certain constitutive, not negligible, phytase activity was detected.

Inositol phosphates formed during IP6 degradation IP6 and the breakdown products formed as a result of IP6 hydrolysis by the yeast were identified and quantified by HPIC. Referring to FIG. 2, chromatographic profiles (HPIC) of chemical acid hydrolysate of sodium phytate (first line), and supernatant from S. cerevisiae cultures (second and third lines) after 15 h of growth is shown. S. cerevisiae was cultured in CBS medium containing either IP6 as the sole P-source (second line, thick) demonstrating yeast phytase activity, or in CBS medium with IP6 and Pi as combined P-sources (third line) showing Pi-induced repression of phytase activity.

Numbered peaks : 1: so4-2 ; 2: Ins (1, 2,6) P3; 3: Ins (1, 2,5, 6) P4; 4: Ins (1, 2,4, 5,6) P5 ; 5: IP6.

The chromatographic profile showed that the enzymes exerting the yeast phytase activity are very specific with respect to position on the inositol ring. Only one isomer of each IPs, IP4 and IP3 was detected for samples withdrawn during growth of S. cerevisiae

in CBSIP6 medium (FIG. 2, second line). These isomers aligned with an acid chemical hydrolysate (FIG. 2, first line, in which all peaks previously have been identified) proved to be I (1,2, 4,5, 6) P5, I (1,2, 5,6) P4, I (1,2, 6) P3, respectively. It is known that peak 2 in the hydrolysate may contain a mixture of I (1, 2,6) P3 and I (1,2, 3) P3. However, the I (1, 2,3) P3 isomer can be excluded in the yeast samples since position 3 was already missing. The same pattern of IP isomers was obtained for all strains tested in our collection, including laboratory strains and several wild-types. The lower, third line in FIG. 2 shows the extracellular composition of IP : s from a S. cerevisiae CBSIP6+Pi culture at 15 hrs of growth. In such cultures no lower inositol phosphates were detectable and the concentration of IP6 (peak 5) remained high through out the experiment proving the Pi induced repression of extracellular phytase activity.

PHO deletion mutants The major secretory acid phosphatase, encoded by PH05, may be involved in phytate hydrolysis. A PH05 deletion mutant may be less efficient in degrading extracellular IP6 as compared with the corresponding parent strain. For this purpose the pho3dpho5d double mutant YMR4 was used. By using this mutant the effect of missing PH03, in addition to PH05, was assessed. However, this gene may be less important since it has been shown to be repressed by thiamine, (which is present in the CBS medium) and its product is mainly a thiamine phosphatase, Praekelt et al. (1994) Regulation of THI4 (MOLI), a thiamine-biosynthetic gene of Saccharomyces cerevisiae.

Yeast. 10,481-490. Contrary to expectation, the mutant was not affected.

Referring to FIGs. 3A and 3B, growth (squares) of S cerevisiae parent strain YS 18 (open symbols) and S. cerevisiae YMR4 phoSdpho3d double deletion mutant (filled symbols) and extracellular concentration of IP6 (circles) as determined by HPIC.

FIG. 3A shows synthetic CBS medium with glucose as carbon and energy source and IP6 plus Pi as phosphorous sources (repressing conditions). FIG. B shows synthetic CBS medium with glucose as carbon and energy source and IP6 as the sole phosphorous source (de-repressing conditions). Represented data are means of double samples from a representative experiment run twice.

In repeated experiments strain YMR4 hydrolyzed extracellular IP6 at a rate equal to its parent strain YS18 (FIG. 3B). In addition, Figure 3A shows that in CBSIP6 + Pi medium (repressing conditions), the pattern of growth and the Pi induced repression was unaffected in the mutant strain YMR4, which closely follows the pattern for YS 18. For both strains at time 24 h after inoculation, 94% of the initial amount of IP6 was still present in the CBSIP6+Pi cultures (FIG. 3A). At this time, neither IP6 nor any other lower IP: s were detectable in the medium of the parallel CBSIP6 culture (FIG. 3B). Additional evidence for independence of PH05 in this context was provided by the rate of growth for the mutant in CBSIP6 medium. The calculated growth rate in the exponential respiro- fermentative phase (growth on glucose, CBSIP6 medium) was 0. 33h-1 (generation time of 2.1 h) for both YMR4 and its parent strain YS 18. This is not evidence that Pho5p and Pho3p are unable to hydrolyze IP6, it is only evidence that shows that they are not needed by the yeast for intact phytase activity.

By PCR and homologous recombination PH010 was deleted in YMR4 and YS 18 thereby obtaining a pho5dplao3dpholOzl triple mutant and pho10# single mutant labeled YM10 and YS10, respectively. Referring to FIG. 4, concentration of inositol hexaphosphate (IP6) in yeast culture supernatant as a function of growth in CBS medium with IP6 as the sole phosphorous source is shown. Grey circles: YS 18 parent strain with all PHO genes intact; filled squares: YMR4pho3Apho5A double mutant; filled triangles: YM10 pho3dplo5dphol Od triple mutant; and open diamonds: YS10 pholOd single

deletion mutant. Data are means of double samples from separate cultures and error bars show the variation between those. These strains were cultured in CBSIP6 and YPDIP6 medium, and the medium concentration of IP6 was monitored by HPIC during growth. In CBS, the rate of IP6 degradation was unaffected in YS10 showing that also PH010 was superfluous (FIG. 4) for intact phytase activity.

Even in YM10, in which the only remaining secretory phosphatase is Phol lp, the net extracellular phytase activity was unaffected as compared with YS 18 and YMR4.

Only a slight, but significant, decrease in the rate of IP6 degradation was in CBS detected in this mutant (FIG. 4). However, in the complex medium YPD, the triple mutant showed a strong reduction in rate of extracellular IP6 (FIG. 5) with maintained growth rate, suggesting phytase activity for PholOp and at least one more enzyme.

Referring to FIG. 5, concentration of inositol hexaphosphate (IP6) in yeast culture supernatant as a function of growth in complex YPD medium with IP6 as the added phosphorous source is shown. Grey circles: YS 18 parent strain with all PHO genes intact; filled squares: YMR4 pho3#pho5# double mutant; filled triangles: YM10 pho34pho5dpholOd triple mutant ; and open diamonds: YS l O phol OzI single deletion mutant. Data are means of double samples from separate cultures and error bars show the variation between those.

Overexpression of PH05 Referring to FIG. 6, a vector construct consisting of plasmid pYX212 containing the insert PH05 used for transformation of YS 18 and YMR4. The glycolytic promoter triose phosphate isomerase (TPI1) controlling expression of PH05, the yeast selection marker URA3 and the AmpR gene are depicted.

In order to assess the impact of exclusively PH05 expression on the extracellular IP6 degradation YMR4 and YS 18 were both transformed with pYX212 containing the insert PH05 controlled by the constitutive glycolytic promoter TPI1 (FIG. 6).

The resulting strains were designated YM-p5 and YS-p5, respectively (Table 1).

The experiments with these mutants were performed in YPD medium with the addition of the appropriate phosphorous source. Growth and extracellular IP6 concentration were assessed as a function of time. Referring to FIGs. 7A and 7B, extracellular concentration of IP6 as a function of growth time of S. cerevisiae YMR4 (A) and YS18 (B), with and without the PH05 containing plasmid pYX212. Open symbols : strains without plasmid; solid symbols: strains containing the plasmid pYX212-PHO5. The experiment was performed in YPD medium with 2% glucose supplemented with either only IP6 (circles) or with both IP6 and Pi (triangles). Represented data are means from two to three independent experiments with a standard deviation never exceeding 5%. From both YM- p5 and YS-p5 data it is evident that PH05 encodes an enzyme with phytase activity ( FIGs. 7A and B).

Under normally repressing conditions (YPDIP6+Pi) IP6 was by YM-p5 and YS- p5 degraded at a fairly high rate (filled triangles in FIGs. 7A and 7B), demonstrating that phytase activity was in these mutants constitutively expressed. The corresponding parent strains lacking the plasmid are shown as controls (open triangles). These controls were in YPD somewhat lesser repressed as compared with CBS medium, however, much more repressed than the cells with plasmid. After 24 h of growth strain YM-p5 had degraded 63 % of the initial IP6 whereas in the YMR4 control strain the corresponding value at 24 h was only 22.5 % (Fig. 6). The rate of IP6 degradation in YPDIP6 non-repressing medium was also analyzed. During these conditions PH010 and PH011 plus the introduced PH05 would be expressed in YM-p5, and PH05, PH010 and PH011 plus the introduced

PH05 in the YS-p5 strain. The data show that the plasmid located PH05 indeed increased the net rate of IP6 degradation (FIGs. 7A and 7B) in both strains. The effect was more pronounced in YS-p5, in which the PH05 gene is present both in the genome and on the plasmid. The combined data from the PH05 overexpression experiments and the experiments with deletion strains compared with their parent strains (FIGs. 3 and 4) shows that Pho5p, PholOp and at least one more enzyme, believed to be Phol lp, are by definition all phytases.

Deletion of the regulatory genes PH080, PH085 and PH02 The PHO system is involved in the yeast phytase activity. Accordingly, deletions in the regulatory genes PH080 and PH085 yield constitutive phytase activity.

Referring to FIG. 8, a direct comparison of phytate degrading capacity between PHO5 overexpressing strains, andpho80zS andpho8521 deletions mutants. Extracellular concentration of IP6 is plotted as a function of growth time of S. cerevisiae YS 1 S-p5 (triangles) with the PHO5 containing plasmid pYX212, S. cerevisiae pho80# deletion mutant (squares) andpho85A deletion mutant (circles) is shown. The experiment was performed in YPD medium with 2% glucose supplemented with either only IP6 (solid symbols) or with both IP6 and Pi (open symbols). Data are means from two independent experiments with a standard deviation never exceeding 5%.

As shown in FIG. 8 these mutants were highly effective in degrading IP6, and completely constitutive demonstrating that the PHO system indeed is involved in S. cerevisiae phytase activity. Since the net rate of IP6 degradation was very rapid, these strains were compared with the YS18-pS (PH05 overexpressing) strain. Both strain pho80A and pho85 showed a higher net rate in IP6 degradation as compared with the YS 18-pS (FIG. 8). The growth rate was unchanged in strain pho80#, however significantly reduced in strain pho85#. In addition to thepho80Z1 andpho85Z1 strains

retrieved from BY4741 (EUROSCARF collection), anotherpho804 refe7ring to Fig 12, mutant was constructed by deletion in YS18, yielding strain YD80. The highly efficient extracellular phytase activity was reconfirmed in this strain.

Referring to FIG. 9, phytate degrading capacity ofpho2A deletion mutant is shown. For comparison, pho80A deletion strain and BY4741 with all PHO genes intact are also shown. Extracellular concentration of IP6 is plotted as a function of growth time in YPD. For explanation of symbols see Figure. The P-source used is shown within parenthesis. In addition to IP6 data, growth data as OD610, are shown for thepho2A deletion mutant and for its parent strain BY4741. Data are means from two independent experiments with a standard deviation never exceeding 5%.

PH02 is together with PH04 known to positively regulate the expression of the secretory phosphatases PH05, PHO 10 and PH011. In a direct comparison the pho2 deletion mutant was demonstrated to lack virtually all phytase activity, demonstrating that yeast phytase activity is largely a matter of its PHO system. The phytase activity may therefore be further increased by overexpressing PH02 and/or PH04.

Expression of PH03 The expression of PH03, in addition to expression of PH05, PH010 and PH011 may increase the rate of extracellular IP6 degradation. For this purpose the parent strain YS18 was used in a CBS medium excluding thiamine, which is known to repress PH03 expression, Nosaka et al. (1989) A possible rolefor acidphosphatase with thiamiya- binding activity encoded by PH03 in yeast. FEMS Microbiol. Lett. 51,55-59 ; Nosaka (1990) High affinity of acid phosphatase encoded by PHO3 gene in Saccharomyces cerevisiaefor thianzin phosphates. Biochim. Biophys. Acta. 1037,147-154. As controls normal CBSIP6 medium (with thiamine) was used, with and without thiamine phosphate to ensure that the yeast was not negatively affected by the missing thiamine. Growth rates

and IP6 degradation were by YS 18 not different between the three media used, indicating that PH03 is of no or minor significance for yeast phytase activity (data not shown).

Cornpetitive degradation of IP6 andpNPP The PHO system in S. cerevisiae has frequently been studied using phenyl phosphate and para-nitrophenyl phosphate (pNPP) as substrates. Monod et al. (1989) Functional analysis of the signal-sequence processing site of yeast acid phosphatase. Eur.

J. Biochem. 182, 213-221 ; Slmyreva et al. (1996) Biochemicalproperties and excretion <BR> <BR> <BR> <BR> <BR> <BR> behavior of repressible acid phosphatases with altered subunit coniposition. Microbiol.

Res. 151, 291-300; Martinez et al. (1998) Identification, cloning and characterization of a derepressible Na'^-coupled phosphate transporter in Saccharomyces cerevisiae. Mol.

Gen. Genet. 258, 628-638. These compounds are used as model organic phosphorous source. Yeasts are often exposed to several organic phosphorous sources.

Accordingly, the impact of adding a second organic phosphorous compound (pNPP) on the rate of IP6 degradation was studied. Both compounds are readily degraded by S. cerevisiae and since the PHO system is involved in both cases a competitive situation arose.

Referring to FIG. 10, extracellular concentration of para-nitrophenyl phosphate (pNPP, squares) and IP6 (triangles) as a function of growth of S. cerevisiae SKQ2n in CBS medium with 2% glucose as carbon and energy source is shown. The CBS medium was supplemented either with IP6 as the sole P-source (open triangles), with pNPP as the sole P-source (open squares) or with both IP6 and pNPP as combined P-sources (Solid symbols).

In the experiment, with both IP6 and pNPP present and Pi absent, the rate of IP6 degradation was completely unchanged as compared with IP6 alone (FIG. 10). However, the rate of pNPP was delayed by presence of IP6 compared to pNPP alone. Most of the

degradation of pNPP occurred after IP6 was depleted from the medium. At 12 h of growth, IP6 was depleted in both cultures, whereas only 17 % of the initial pNPP was used in the mixed culture, and 49% in the sole pNPP culture. The experiment was also performed with the pNPP concentration set to 6 times the concentration of IP6 (2.25 mM) to obtain equal stoichiometry in number of available phosphate groups. Six times more frequently pNPP randomly met the enzymes. The experiment yielded similar data, that is no change in rate of IP6 hydrolysis in the presence of pNPP and delayed pNPP degradation as compared with pNPP alone was observed (data not shown).

Content of inorganic phosphate and IP6 in bread dough To understand the yeast environment during bread making, relative to the PHO system, the Pi and IP6 content were analyzed in two types of wheat based bread doughs.

In both doughs the level of Pi increased and the level of IP6 decreased as a function of leavening time (FIG. 11).

Referring to FIG. 11, the content of inositol hexaphosphate (IP6, triangles) and inorganic phosphate (Pi ; squares), expressed as jumol per gram wet dough, as a function of leavening time in two types of wheat based doughs is shown. Solid symbols: whole meal wheat flour (100% extraction rate; i. e. the proportion of the whole wheat grain obtained as finished flour), and open symbols: white wheat flour (extraction rate 60%). Error bars indicate the maximum variation between double samples.

The concentration of Pi was higher in the whole wheat flour dough starting at 4 MmolXg wet dough (approximately equal to mM), and ending after 60 min of leavening at 11.8 pmol/g dough, while in the white wheat flour the start and end concentrations were 1. 7 pmol/g and 5. 8 llmol/g, respectively. The initial concentrations of IP6 were 9 among and 3 llmol/g for whole-wheat flour and white wheat flour, respectively, decreasing to 4.8

and 0.4 among after 60 minutes of leavening. This demonstrates that phytases are active in the dough.

Most previous investigations in the context of yeast degrading IP6 refer to a"yeast phytase", frequently distinguishing it from phosphatases, without identifying genes and enzymes. Nayini et al. (1984) The phytase of yeast. Lebensm. Wiss. u-Technol. 17,24-26 ; Harland et al. (1989) Effects of phytase from three yeasts on phytate reduction in Norwegian whole wheat flour. Cereal Chem. 4, 357-358 ; Nair et al. (1991) Phytic acid content reduction in canola meal by various microorganisms in a solid state fermentation process. Acta Biotechnologica. 11,211-218. Lambrechts et al. (1992) Utilization of phytate by some yeasts. Biotechnol. Lett. 14,61-66 Attempts to assess the yeast contribution to the observed phytase activity during bread leavening has sometimes been performed in conditions such as starvation in combination with 50°C, Harland et al.

(1989) Effects of phytase from three yeasts on phytate reduction in Norwegian whole wheat flour. Cereal Chem. 4,357-358.

The yeast phytase activity in conditions suitable for yeast growth were studied and it was observed that, provided absence of Pi, S. cerevisiae is well adapted to utilize extracellular IP6 as a phosphorous source. The rate of IP6 degradation was rapid and supplied the yeast metabolic machinery with phosphorous without a decrease in growth rate, that arose from the increased energy cost of synthesizing the degrading enzymes.

The phosphate-induced repression of phytase activity indicated that the PHO gene family in S. cerevisiae is involved.

In low concentrations of inorganic phosphate S. cerevisiae is known to induce the synthesis of at least three phosphatases that are exported out through the cellular membrane. It appears that the enzymes work best in oligomeric organization, as studied by Shnyreva et al. Shnyreva et al. (1996) Biochemical properties and excretion behavior

of repressible acidphosphatases with altered subunit coniposition. Microbiol. Res. 151, 291-300, composed of a specific ratio between Pho5p (86%), PholOp and Phol lp (14% together). The so-called"yeast phytase"may be the concerted action of the components in this oligomeric enzyme.

However, when using pNPP as substrate Shnyreva et al. , Shnyreva et al. (1996) Biochemical properties and excretion behavior of repressible acidphosphatases with altered subunit composition. Microbiol. Res. 151, 291-300 have shown that also homopolymers of the individual secretory acid phosphatases have significant and different activities. In addition to being present in the largest amount, Pho5p was found to exert the highest specific activity. Fifteen times higher activity was obtained for Pho5p expressed alone as compared with PholOp and Phol lp, respectively.

The PHO5 deletion did not yield a less efficient strain with respect to IP6 degradation. Several explanations could fit the data: (i) PH05 was not important for phytase activity or (ii) the remaining level of PholOp and Phol lp, organized without Pho5p, was sufficient for unchanged phytase activity. It may be that (iii) PHO01 and PH011 became unregulated in the absence of PHO5. In fact, the absence of PH010 or PH011 has been shown to increase the Pho5p level Shnyreva et al. (1996) Biochemical <BR> <BR> <BR> <BR> <BR> <BR> properties an. d excretioya behavior of repressible acid phosphatases with altered subunit composition. Microbiol. Res. 151,291-300 and hence, the opposite may also be true. A (iv) different as yet unidentified phytase enzyme may exist. The S. cerevisiae gene DIA3 for instance, encodes a protein largely related to Pho3, Pho5, PholO and Phol 1 (see e. g. the YPD database). Dia3p may be extracellular.

Since the significance of the different secretory phosphatases in IP6 specific degradation is not known a strain was created lacking only PHO10 (YS10). Using strain YS10 it was demonstrated that PHOI 0 was also dispensable for yeast phytase activity,

shown by intact growth-and IP6 degradation rate. Thus, neither of PH03, PH05 or PHO10 was essential for high phytase activity. However, in a triple mutant (YM10), a very small, but significant decrease in rate of IP6 degradation was observed in CBS and a pronounced decrease in YPD, demonstrating two things. First, PholOp is an enzyme with phytase activity since a change was observed when PHO10 was the only difference and second, a strain with only Phol lp alone (or an unknown phytase) still has a remarkable phytase activity in defined medium. For all these deletion strains the growth rate remained unchanged, so the net phytase activity is comparable also if expressed as biomass specific phytase activity. This further means that phosphorous never became limiting, even in a strain lacking three secretory phosphatases during growth on IP6. It is interesting that the main secretory phosphatase (Pho5p) is dispensable and that a strain encoding only one of two so-called minor secretory phosphatases, (Phol lp), is alone able to keep almost unchanged phytase activity.

To assess the phytase potential of solely Pho5p, two strains were constructed which expressed almost exclusively PH05 during growth in repressing medium. The data obtained using these strains show that PH05 by itself encodes a protein with phytase activity and is therefore likely to participate when a wildtype strain degrades IP6. When growing YM-p5 and YS-p5 in de-repressing media, additional phytase activity was observed as compared with the corresponding strains lacking the plasmids (FIG. 6). This was not necessarily expected since the increased rate of IP6 degradation leads to increase in Pi concentration which in turn have a repressing impact on the genomic phytase encoding genes. The data reveals that yeast phytase activity is not the action of one gene product. However, simply by overexpressing one single gene strains were obtained with improved net phytase activity under both repressing and de-repressing conditions.

Furthermore, the regulatory components pho80p and pho85p are known to phosphorylate pho4p in conditions of high medium Pi levels. The phosphorylated pho4p retrieves affinity for a transporter and leaves the nucleus instead of acting as a transcription activator for several PHO genes, including the secretory phosphatases. If one or both of pho80p and pho85p is missing its kinase activity on pho4p is lost and pho4p will together with pho2p remain active as transcriptional activators for the secretory phosphatases. The resulting mutant constitutively expresses PHO5, PHO10 and PH011.

Since deletions of PH080 or PH085 both yielded very strong constitutive phytase activity the involvement of the PHO system was again demonstrated.

It has previously been shown that the contribution of yeast to the phytase activity observed during bread leavening is very small, and not sufficient for a significant increase in iron and zinc absorption, Harland et al. (1989) Effects of phytase from three yeasts on phytate reduction in Norwegian whole wheatflour. Cereal Chem. 4, 357-358 ; Turk et al.

(1992) P/ytate degradation during bread making: effect of phytase addition. J. Cereal Science. 15, 281-294 ; Turk et al. (1996) Reduction in the levels ofphytate during wholemeal bread making ; Effect of yeast and wheat phytases. J. Cereal Science. 23,257- 264. The IP6 level and mineral uptake in human intestine was, however, improved when adding commercial Aspergillus phytase to dough or wheat rolls, respectively, Turk et al.

(1992) Phytate degradatiora during bread making : effect ofphyt. ase addition. J. Cereal Science. 15,281-294 ; Turk et al. (1996) Reduction in the levels ofphytate during wholemeal bread making ; Effect of yeast and wheat phytases. J. Cereal Science. 23,257- 264. It has been shown that the phytase activity is Pi repressed. The concentration of Pi was shown to be in the millimolar range which is above the level known to induce repression, Yoshida et al. (1989) Function of the PHO regulatory genes for repressible acidphosphatase synthesis in Saccharomyces cerevisiae. Mol. Gen. Genet. 217,40-46.

(10 mM Pi is often used as"high"and 0.2 mM as"low" ; Ogawa et al. (2000) New comporaerats of a system for phosphate accurnulation and polyphosphate metabolism in Saccharornyces cerevisiae revealed by genomic expression analysis. Mol. Biol. Cell. 11, 4309-4321). Hence, in the typical bread making situation the PHO genes encoding PhoSp, PholOp and Phol lp are not very active and, as shown, phytase activity is not substantially expressed. Furthermore, it has been shown that the IP6 degradation in cereals must be extensive to improve the iron absorption, Brune et al. (1992) Human iron absorption from bread : Inhibiting effects of cerealfibre, phytate and inositol phosphates with different numbers of phosphate groups. J. Nutr. 122, 442-449. As little as 0.5 llmol/g dry sample of IP6 and IP5 is inhibitory, Sandberg et al. (1999) Inositol phosphates with different number ofphosphate groups influence iron absorption in humans. Am. J. Clin. Nutr. 70,240-246.

The data on the content of IP6 in dough were expressed per wet weight (i. e. the dry content is higher), and even at the end of the fermentation it exceeds iron inhibitory levels. The reduction in IP6 taking place during dough fermentation with commercial baker's strains has been shown to be almost exclusively due to plant phytases present in the flour. A constitutively expressing yeast strain is desirable.

It has further been demonstrated herein an apparent preference for IP6 over pNPP when these were given together. In Nayaini et al. , Nayini et al. (1984) The phytase of yeast. Lebensm. Wiss. u-Technol. 17,24-26. yeast phytase and phosphatase were purified from a total biomass extraction of a Baker's yeast cake. The distinction between phytase (which is a phosphatase, for which one of the substrates is phytate) and phosphatase activity was that a phosphatase was able to hydrolyze alpha-glycerophosphate but unable to hydrolyze IP6. The purified phytase was assessed for substrate specific activity using 11 different phosphorous sources. Phenyl-phosphate was one of these and the specific activity with this substrate was 40 times higher than for IP6. The data set forth above,

shows virtually no degradation of pNPP until IP6 is depleted. Even at six times higher pNPP concentration the data indicate an inhibition of pNPP hydrolysis by IP6. Although the yeast phytase is not one single enzyme the activity was shown to be very precise in regard to the position on the inositol ring to be hydrolyzed, consistent with previous data on one commercial baker's strain, Turk et al. (2000) Inositol hexaphosphate hydrolysis by baker's yeast. Capacity, kinetics, and degradation products. J. Agric. Food Chem. 48, 100-104. This specificity, the so-called 3-phytase activity, was true for all yeasts tested, including such different wild isolates of S. cerevisiae including those from fish intestine and clinical isolates of human pathogens, as well as for all deletion strains. Accordingly, the modified strains can be used to efficiently produce highly specific isomers of lower inositol phosphates.

In addition to the Saccharomyces phytase work we are screening non- Saccharomyces yeasts for the possibility of finding novel enzymes with desirable properties. Tested species are primarily from tropical-and cactus origin. Desired properties are high temperature optimum and/or a specificity yielding different isomers as compared with S. cerevisiae. So far, all tested species and strains have shown phytase activity. However, biochemical properties of the different enzymes have not yet been examined.

In addition to S. cerevisiae, other yeast species that are capable of expressing phytase can be transformed to exhibit increased phytase activity.

Although the present invention has been shown and described with respect to several preferred embodiments thereof, various changes, omissions and additions to the form and detail thereof, may be made therein, without departing from the spirit and scope of the invention.