WO2005054442A2

WO2005054442A2 - Yeast strains resistant to purine base analogues, process for the production thereof, yeast extracts obtainable from said yeast strains and use of said yeast strains or yeast extracts in food, drink and beverabes

Info

Publication number: WO2005054442A2
Application number: PCT/EP2004/013221
Authority: WO
Inventors: Petrus Jacobus Theodorus Dekker
Original assignee: Dsm Ip Assets B.V.
Priority date: 2003-11-21
Filing date: 2004-11-18
Publication date: 2005-06-16
Also published as: WO2005054442A3

Abstract

The present invention describes a process for the preparation of a yeast strain resistant to a purine base analogue comprising: a) subjecting a parent yeast strain to mutagenesis; b) selecting among the yeast strains obtained in a) a yeast strain which is resistant to a purine base analogue; c) selecting among the yeast strains obtained in b) a yeast strain which has a higher intracellular amount of purine-containing ribonucleotide when compared with the parent strain. The yeast strains resistant to a purine base analogue obtainable by the process of the invention accumulate purine-containing ribonucleotides and preferably also pyrimidine-containing ribonucleotides intracellularly and can be advantageously used in the preparation of yeast extracts, in particular of yeast extracts comprising 5'-ribonucleotides.

Description

Yeast strains resistant to purine base analogue and their use in the preparation of yeast extracts

The present invention relates to yeast strains resistant to a purine-base analogue, and to a process for the production of said yeast strains. The invention further relates to a yeast extract comprising 5'-ribonucleotides derived from said yeast strains, to a process for the production of said yeast extract and to the use of said yeast strain or yeast extracts in food or feed. Background of the invention

Generally, two types of yeast extracts are recognized. Autolytic yeast extracts are concentrates of the soluble materials obtained from autolysis of yeast i.e. a process by which degradation of the cells and of the polymeric yeast material is at least partially effected by the native yeast enzymes released in the medium after (partial) damaging and/or disruption of the microbial cell wall. This type of yeast extract, which is rich in amino acids, is used in the food industry as basic taste provider. The amino acids present in the yeast extract add a bouillon-type brothy taste to the food without adding specific notes. Hydrolytic yeast extracts are concentrates of the soluble materials obtained from hydrolysis of yeast i.e. a process by which degradation of the cells and of the polymeric yeast material is effected by addition of exogenous proteases and/or peptidases and especially nucleases capable of degrading RNA into 5'-ribonucleotides. Generally the native yeast enzymes are inactivated prior to the lysis. During this process 5'- ribonucleotides of guanine (5'-GMP), uracil (5'-UMP), cytosine (5'-CMP) and adenine (5'-AMP) are formed. When adenylic deaminase is added to the mixture, 5'-AMP is transformed into 5'-inosine mono phosphate (5'-IMP). The hydrolytic yeast extracts obtained by this method are therefore rich in 5'-ribonucleotides, especially rich in 5'- GMP and 5'-IMP. Often yeast extracts are also rich in mono sodium glutamate (MSG). 5'-IMP, 5'-GMP and MSG are known for their flavor enhancing properties. They are capable of enhancing the savory and delicious taste in certain types of food. This phenomenon is described as 'mouthfeel' or umami. The natural 5'-ribonucleotides of these yeast extracts demonstrate a synergistic effect with the glutamate present in the extract as well as in the food substrates to provide the enhanced savoury attributes to processed food. These yeast extracts rich in δ'-ribonucleotides and, optionally, rich in MSG, are usually applied in e.g. soups, sauces, marinades, and flavor seasonings. Yeast extracts rich in 5'-ribonucleotides are up to date produced using yeast strains with high RNA/ribonucleotide content and/or by partial extraction of the cell content. For this purpose, strains of the genera Saccharomyces, Kluyveromyces and Candida can for example be used. Partial extraction of the cell content is used to obtain a higher 5'-ribonucleotide content in the yeast extract than that which would be obtained based on the RNA/ribonucleotide content of the strain. However a problem of partial extraction of the cell content is that this process is economically not very advantageous. Therefore it would be preferred to use yeast strains with an increased RNA/ribonucleotide content. EP299078 describes the production of yeast extract with high 5'-ribonucleotide content using e.g. a boric-acid resistant mutant of a Candida utilis yeast strain. It is thus an object of the present invention to find alternative yeast strains with increased ribonucleotide and/or RNA content which can be advantageously used in the production of 5'-ribonucleotide containing yeast extracts. It is known that yeast strains resistant to purine base analogues excrete purines in the culture medium when grown in the presence of a purine base analogue like 8-aza guanine (from now on abbreviated as 8-AG). With excretion it is herewith intended either an active or a passive process by which a substance is released outside a cell. The de novo biosynthesis of purines in yeast is a highly regulated process and it has been extensively studied (Guetsova, M. L, K. Lecoq, and B. Daignan-Fornier Genetics 147:383-397 1997). In the biosynthetic pathway phosphoribosyl-1- pyrophosphate is converted via a series of steps, into 5'-IMP which in turn may be converted via a series of steps into 5'-AMP or into 5'-GMP. 5'-AMP and 5'-GMP are substrates for the final synthesis of ATP and GTP, which are substrates for the synthesis of RNA. The intracellular production of purine-containing 5'-ribonucleotides is regulated by a feedback inhibition mechanism. If an excess purine bases is present in the cell, intracellular production thereof is blocked. Therefore yeast cells grown in the presence of 8-AG and in the absence of other purines are not able to grow. Mutants which are resistant to 8-AG are able to grow in the presence of 8-AG and in the absence of other purines. Generally in said mutants the feedback inhibition mechanism in the de novo synthesis of purine is disturbed and the yeast cells continue to produce purine- containing 5'-ribonucleotides which are hydrolysed and the excess purine bases are excreted outside the cell. For example, Sato et al (Sato, N., Shimosaka, M., Fukuda, Y., Murata, K.,

Kimura, A., Agric. Biol. Chem. 50(5): 1339-1340, 1986) describe that some Saccharomyces cerevisiae mutants, which are resistant to 8-azaguanine and 8- azaadenine, are excreting hypoxanthine and inosine in the culture fluid while no hypoxanthine / inosine is detected in the culture fluid of the parent strain. Also Woods et al (Woods, R.A., Roberts, D.G., Friedman, T., Jolly, D., Filpula, D., Mol. Gen. Genet. 191 : 407-412, 1983) describe that Saccharomyces cerevisiae mutants resistant to 8-azaguanine and 8-azaadenine, excrete purines in the medium. The purine secreting yeast strains described in the art have the drawback that they hydrolyse and excrete excess of purine-containing ribonucleotides produced in the cell. As a consequence their use in the production of 5'-ribonucleotide-rich yeast extracts does not present any particular advantage when compared with the use of the parent strain. We have surprisingly found yeast strains resistant to purine base analogues accumulating purine-containing ribonucleotides intracellularly. These strains are highly suitable for the preparation of yeast extracts, in particular yeast extracts comprising 5'- ribonucleotides.

Description of the invention

In a first aspect, the present invention discloses a process for the preparation of a yeast strain resistant to a purine base analogue comprising: a) subjecting a parent yeast strain to mutagenesis; b) selecting among the yeast strains obtained in a) a yeast strain which is resistant to a purine base analogue; c) selecting among the yeast strains obtained in b) a yeast strain which has a higher intracellular amount of purine-containing ribonucleotide when compared with the parent strain. A purine base analogue is herewith defined as a molecule, which has a purine backbone and which is not naturally occurring in the (yeast) cell. Generally the purine base analogue is a derivative of adenine, guanine, urate, xanthine, hypoxanthine. In a preferred embodiment of the invention the purine-base analogue is 8-aza guanine (abbreviated as 8- AG) or 8-aza adenine (abbreviated as 8-AA). More preferably the purine-base analogue is 8-AG. Strains of the genera Saccharomyces, Kluyveromyces or Candida can for example be used as parent strains. However, yeast strains derived from the genus Saccharomyces are preferred. Yeast strains derived from the species Saccharomyces cerevisiae are most preferred. Generally, the parent yeast strain is a yeast strain with a high RNA content. A parent yeast strain as mentioned above is subjected in step a) of the process of the invention to mutagenesis. Said operation may be performed by subjecting the parent yeast strain to a mutagen according to methods known to those skilled in the art. As a mutagen UV radiation may be used, or a chemical agent such as N-methyl-N'-nitro- N-nitrosoguanidine or ethyl methanesulfonate. The use of other mutagens is also possible. In step b) of the invention yeast strains are selected among the yeast strains obtained in step a), which are resistant to a purine base analogue. The purine base analogues mentioned above can be used in the method of the invention. Also the isolation of spontaneous variants of the parent yeast strain is covered by the invention, i.e. yeast strains may be subjected to the selection step b) without prior being exposed to a mutagen. Step b) may be performed by firstly incubating strains obtained in step a) for a sufficient time on a nutrient medium comprising a purine base analogue in a concentration sufficient to give growth inhibition of the parent strain and not comprising any other sources of purine bases and secondly selecting the yeast strains growing under these conditions. If required, single colonies of purine base analogue resistant yeast strains can be isolated according to methods known to those skilled in the art and step b) of the method of the invention may be optionally repeated. Typically step b) of the invention may be performed at a temperature which may vary between 20 and 40°C. Typically the incubation time which is sufficient to obtain growth of purine base analogue resistant yeast strains may vary between 2-20 days. Generally the nutrient medium comprises a source of assimilable carbon, a source of assimilable nitrogen, a source of assimilable phosphorus, and may further comprise vitamins and trace elements. Generally a synthetic medium is used without complex carbon or nitrogen sources. Generally this step may be performed on a solid medium. Alternatively, said step may be performed in a liquid medium. The nutrient medium used in this step does not comprise any source of purine base apart for the purine base analogue. Such sources of purine base may be a complex nitrogen source, like yeast extract comprising 5 '-ribonucleotides. When 8-AG is used as purine base analogue in a process of the invention, depending on the yeast strain used, a concentration of 8-AG is used of at least 0.2 g/l, preferably at least 0.5 g/l, more preferably at least 1 g/l and generally not higher than 2 g/l. This concentration is significantly higher then that described in literature, for the selection of 8-AG resistant yeast strains of Saccharomyces cerevisiae excreting purines (Woods, R.A., Roberts, D.G., Friedman, T., Jolly, D., Fi lpula, D., Mol. Gen. Genet. 191 : 407-412, 1983). In step c) of the invention yeast strains are selected among the yeast strains obtained in b) which have a higher intracellular amount of purine-containing ribonucleotide when compared with the parent strain, Step c) of the process of the invention may be performed by culturing the yeast strains obtained in b) in a nutrient medium for a determined period of time and subsequently selecting the yeast strains which grow in the nutrient medium after said period of time and which have a higher intracellular amount of purine-containing ribonucleotide when compared with the parent strain. Generally this step is performed by culturing the yeast strain obtained in b) and the parent strain under the same conditions in the nutrient medium. In the context of the present invention culturing under the same conditions means that the yeast strains obtained in b) and the parent strain are inoculated at the same biomass concentration, cultured in the same type of nutrient medium, at the same temperature and for the same time period. The nutrient medium used in this step comprises as above described a source of assimilable carbon, a source of assimilable nitrogen, a source of assimilable phosphorus, and may further comprise vitamins and trace elements. Examples of sources of assimilable carbon which may be used include: glucose, fructose, sucrose or molasses. Typical sources of assimilable nitrogen which may be used include ammonia, ammonium salts, such as ammonium sulphate, ammonium carbonate and ammonium acetate, or other nitrogen compounds such as yeast extract and corn steep liquor. Examples of sources of assimilable phosphorus, which may be used, include phosphoric acid, phosphate salts like sodium phosphate, potassium phosphate etcetera. It is understood that the nutrient medium used in this step may comprise sources of purine base. Culturing of the yeast strains may be advantageously performed in a shake flask but other methods known to those skilled in the art are also possible. Typically culturing of the yeast strains is performed at a temperature varying between 20° and 40°C. Preferably culturing of the yeast strains is performed for a period of time sufficient to reach early stationary phase growth. The latter is dependent on the inoculation size and can be easily determined by those skilled in the art. In the context of the present invention purine-containing ribonucleotide is defined as a ribonucleotide selected from the group consisting of GMP, AMP and IMP. The intracellular amount of purine-containing ribonucleotide is defined as the total amount of purine-containing ribonucleotide inside the cell, i.e. the sum of purine-containing ribonucleotide present in the free form in the cytoplasm (as free 5'-ribonucleotide), the purine-containing ribonucleotide incorporated in the cellular RNA and the purine- containing ribonucleotide bound to phosphate as di- and triphosphate. This step may be done by measuring the intracellular amount of purine- containing ribonucleotide in a precise amount of biomass of the yeast strain obtained in b) and comparing this value with the intracellular amount of purine-containing ribonucleotide measured in the same amount of biomass in the parent strain. Yeast strains are selected which have a higher intracellular amount of purine-containing ribonucleotide when compared with the parent strain. Measuring of the intracellular amount of purine-containing ribonucleotide can be done according to a method which is described later. Preferably, yeast strains are selected wherein the intracellular amount of purine- containing ribonucleotide is at least 5 % higher, preferably at least 10 % higher when compared with the parent strain. There is no real upper limit for how much higher the intracellular amount of purine-containing ribonucleotide in the yeast strain can be when compared with the parent strain. Generally the upper limit of the intracellular amount of purine-containing ribonucleotide may be around a level that is 2 to 5 times the level of the parent yeast strain. In a second aspect the present invention discloses a yeast strain resistant to a purine base analogue which accumulates purine-containing ribonucleotides in the cell. Said yeast strain is derived from a parent yeast strain, and has a higher intracellular amount of purine-containing ribonucleotide when compared with the parent strain. The yeast strain according to the invention is obtainable by a process of the first aspect. Preferably the yeast strain according to the invention has a higher intracellular amount of GMP when compared with the parent yeast strain. Surprisingly yeast strains according to the invention may also have a higher amount of pyrimidine-containing ribonucleotide when compared with the parent strain. Therefore in an embodiment of the invention, the yeast strain obtainable by the process of the invention has a higher intracellular amount of purine-containing ribonucleotides when compared with the parent strain and of pyrimidine-containing ribonucleotides when compared with the parent strain. Intracellular amount of purine-containing ribonucleotides has been defined above. In the context of the present invention pyrimidine-containing ribonucleotide is defined as a ribonucleotide selected from the group consisting of UMP and CMP. The intracellular amount of pyrimidine-containing ribonucleotide is defined as the total amount of pyrimidine-containing ribonucleotide inside the cell, i.e. the sum of pyrimidine-containing ribonucleotide present in the free form in the cytoplasm (as free 5'- ribonucleotide), the pyrimidine-containing ribonucleotide incorporated in the cellular RNA and the pyrimidine-containing ribonucleotide bound to phosphate as di- and triphosphate. In another preferred embodiment the yeast strain has a higher content of RNA when compared with the parent strain. The latter means that the higher intracellular amount of purine-containing or pyrimidine-containing ribonucleotide in the resistant yeast strain when compared with the parent strain is also due to a higher amount of purine- containing or pyrimidine containing ribonucleotide bound into the RNA in the resistant yeast strain when compared with the parent strain. To verify whether the yeast strains selected in step c) also have a higher intracellular amount of pyrimidine-containing ribonucleotide when compared with the parent strain, the intracellular amount of pyrimidine-containing ribonucleotide in a precise amount of biomass of the yeast strain obtained in b) can be measured and compared with the intracellular amount of pyrimidine-containing ribonucleotide measured in the same amount of biomass in the parent strain. Analogously it can be verified whether the yeast strains selected in c) has a higher RNA content when compared with the parent strain. According to a preferred embodiment of the invention, the yeast strain according to the invention has an intracellular amount of pyrimidine-containing ribonucleotide, which is at least 5% higher, preferably at least 10 % higher when compared with the parent strain. Also in this case, there is no real upper limit for how much higher the intracellular amount of pyrimidine-containing ribonucleotide in the yeast strain according to the invention can be when compared with the parent strain. Generally the upper limit will be as indicated for purine containing ribonucleotide. Measuring of the intracellular amount of GMP, AMP and IMP in the yeast strains resistant to the purine analogue and in the parent strain can be performed with methods known to those skilled in the art. A method which can advantageously be used when a sufficient amount of yeast biomass is available (e.g. when culturing is performed in a shake flask or at a larger scale) comprises hydrolysing a specific amount of yeast biomass under alkaline conditions and determining the corresponding 3'-GMP and 2'-GMP formed by hydrolysis of RNA and the 5'-GMP already present via a HPLC method. A similar method can be used for the determination of AMP or IMP or for the determination of the pyrimidine-containing ribonucleotides, i.e. CMP and UMP. It will be clear to those skilled in the art that different HPLC conditions may be necessary to determine the different ribonucleotides. Optionally the amount of intracellular purine-containing and pyrimidine- containing ribonucleotides in the yeast strain or in the parent strain may for example be measured by determining the amount of the corresponding 5'-ribonucleotides in a yeast extract comprising 5'-ribonucleotides and obtained from the yeast strain or parent strain as described below. The amount of 5'-ribonucleotides in the yeast extract may be measured as well via a HPLC method known to those skilled in the art. In order to determine the specific amount of yeast biomass to be used in the measurement methods known to those skilled in the art may be advantageously used, e.g. measurement of the optical density at 600 nm and determination of the yeast dry weight. To verify whether the yeast strains selected in step c) have a higher RNA content when compared with the parent strain it can be verified whether the increased purine- containing or pyrimidine-containing ribonucleotide content in the yeast strain selected in c) is also due to an increase in the RNA-bound purine-containing or RNA-bound pyrimidine- containing ribonucleotide or whether it is only due to an increase in the non bound purine- containing or pyrimidine-containing ribonucleotide. The latter can for instance be measured by determining e.g. the purine-containing ribonucleotide content of a preparation obtained with and without an alkaline hydrolysis of RNA. The method is further exemplified in the experimental section. Preferably the yeast strain according to the invention does not excrete purine base. With purine base it is herewith intended a molecule with a purine backbone which is usually present in the yeast cell and which is preferably selected from the group consisting of: guanine, adenine, xanthine, hypoxanthine, urate and a riboside thereof. To determine whether a yeast stain excretes purine bases a method as described by Lecoq et al. (2000) Genetics, 156; 953-961 may be used. In a third aspect the invention provides a process for the preparation of a yeast extract comprising 5'-ribonuceotides comprising: a) fermenting a yeast strain of the second aspect obtainable by the method of the first aspect in a culture medium until a desired cell density is achieved; b) separating the yeast cells from the fermentation broth; c) treating the yeast biomass to inactivate at least a substantial part of the native RNA degrading yeast enzymes and of the native 5'- ribonucleotide degrading yeast enzymes; d) subsequently treating the yeast biomass to open up the yeast cells; e) treating the opened yeast cells with an enzyme suitable to convert RNA into 5'-ribonucleotides. In step a) of the process of the invention the yeast strain is fermented in a culture medium until a desired cell density is achieved. This step may be performed by conventional methods. Typically a culture medium as already described above may advantageously be used in this step. It is understood that the culture medium may comprise sources of purine bases. Typically the yeast strain is fermented aerobically at a temperature generally from 29 to 36°C, and a pH of typically from 3 to 7 and the fermentation is stopped when a desired cell density is reached. The pH and/or the temperature may or may not vary during the fermentation. The fermentation can be performed on any suitable scale, i.e. from a laboratory scale up to a production scale, in a fermentor of suitable size. In step b) of the process of the invention the yeast strain is separated from the fermentation broth. This step can be performed by conventional methods, for example by centrifugation or filtration. In step c) of the invention the yeast biomass is treated to inactivate at least a substantial part of the native RNA degrading yeast enzymes and at least a substantial part of the native 5'-ribonucleotide degrading yeast enzyme. In the context of the present invention native RNA degrading yeast enzymes are defined as native yeast enzymes that break down the RNA and/or modify the RNA in such a way that the RNA cannot be degraded by δ'-phosphodiesterase (5'-Fdase) into 5'-ribonucleotides. Examples thereof are RNAse and some phosphatases. In the context of the present invention native 5'-ribonucleotide degrading yeast enzymes are defined as native yeast enzymes that degrade 5-'ribonucleotides i nto riboside and phosphate. Examples thereof are 3'-Fdase and phosphatases. Even more preferably in step c) at least a substantial part of all native yeast enzymes is inactivated. With "at least a substantial part" of the native RNA degrading yeast enzymes, of the native 5'-ribonucleotide degrading yeast enzymes or of all the native yeast enzymes it is herewith intended at least 50% of these native yeast enzymes, preferably at least 60% of these native yeast enzymes, more preferably at least 80%, even more preferably at least 90%, most preferably approximately all these native yeast enzymes. Inactivation of native yeast enzymes in the fermentation broth is possible with for example a heat shock, for example 5 to 10 minutes at a temperature of 80° - 97°C. After this treatment the RNA of the yeast will not be degraded by unwanted enzyme activity during open ing up of the yeast cells. In step d) the yeast biomass is subsequently treated to open up the yeast cells. In order to open up the yeast cells, the cells can be treated chemically, mechanically or enzymatically. Preferably enzymes are used for this solubilisation or cell wall lysing step. In general the microbial cells are enzymatically treated for 1-24 hours, at a pH between 4 and 10 and a temperature of 40-70°C. For example a protease such as endoprotease can be used. The reaction conditions for the protease depend on the enzyme used. After the chemical or enzymatic treatment, the chemicals or enzymes should preferably be neutralised. For example the enzymes should be inactivated. Inactivation of the enzymes can be done by pH treatment or preferably by a heat treatment whereby the enzyme is inactivated. After liberation of the cell contents a composition is obtained which comprises yeast cell walls, carbohydrates, proteins and amino acids, RNA, minerals, lipids, and vitamins. In step e) of the process of the invention the opened yeast cells are treated with an enzyme suitable to convert RNA into 5'-ribonucleotides. RNA can be converted into 5'-ribonucleotides enzymatically. 5'- Phosphodiesterase (5'-Fdase) can be used. 5'-Phosphodiesterase can be obtained from a microbial or a vegetable source (for example a malt root extract). An example of commercially available microbial product is Enzyme RP-1 produced by Amano. In a preferred embodiment of the invention, the process of the invention further comprises a step f) wherein the mixture obtained in e) is treated with an enzyme suitable to convert 5'-AMP into 5'-IMP. The invention also provide the possibility of performing step f) and e) at the same time. Deaminase, for example adenylic deaminase, can be used to convert 5'-AMP into 5'-IMP. An example of a commercial available product is Deaminase 500 produced by Amano. In a preferred embodiment of the invention the process of the invention further comprises a step g) wherein the solid fraction is separated from the liquid fraction and the liquid fraction is concentrated or dried. The liquid fraction can be concentrated to yield a product in liquid form (generally with a dry matter content of approximately 40- 65% w/w) or further concentrated to yield a product in a form of a paste (generally with a dry matter content of approximately 70-80% w/w). The product can be dried to for example a dried powder with a dry matter content of approximately 95% w/w or higher. Optionally the suspension of solid and liquid fraction may be concentrated or dried without prior separation of the solid phase. The separation of the solid fraction from the liquid fraction may be achieved with known methods, for example via centrifugation or filtration. Drying of the liquid fraction or of the solid/liquid suspension can also be achieved with conventional methods like for example spray drying. In a fourth aspect the invention provides a yeast extract comprising 5'- ribonucleotides obtainable by the process of the invention. A yeast extract obtainable by the process of the invention has the advantage of containing a higher amount of purine- containing 5'-ribonucleotide and preferably of pyrimidine-containing 5'-ribonucleotide than a yeast extract comprising 5'-ribonucleotides obtained from the parent strain. The latter is advantageous for the taste improving effect of the yeast extract. Generally the yeast extracts comprise an amount of 5'-ribonucleotides between

5-50 % w/w based on sodium chloride free dry matter of the yeast extract. The amount of 5'-ribonucleotides in the yeast extract comprising 5'-ribonucleotides may be measured via HPLC according to methods known to those skilled in the art. The weight percentage of 5'-ribonucleotide is calculated based on the disodium salt heptahydrate thereof. Sodium chloride free dry matter means that for the sole scope of calculations the amount of sodium chloride in the yeast extract dry matter is excluded from the calculation. In a fifth aspect the invention provides the use of the yeast strains according to the invention or of the yeast extracts comprising 5'-ribonucleotides according to the invention in food and feed products, including food and feed intermediate products. In the context of the present invention the word "food" means either a nutriment in solid form or a beverage. The yeast strains according to the invention can also be used for the production of autolytic as well as hydrolytic yeast extracts. The invention will now be illustrated by examples, which however are not intended to be limiting.

Example 1

Mutagenesis and selection The yeast Saccharomyces cerevisiae was grown in a shake flask in 100 ml YePD medium (10 g/l Yeast extract (Difco), 20 g/l Bacto-peptone (Difco), 20 g/l glucose (Merck)), for 16 hours at 30 °C under vigorous shaking. The cells of this culture were collected by centrifugation in 50 ml Greiner tubes and washed twice with 50 ml 0.9% NaCI. The cell pellet was resuspended in 0.9% NaCI to 10⁷ cells/ml. 10 ml of this cell suspension (10⁸ cells) was transferred to a plastic petri dish and exposed to UV using the DARK TOP UV (Osram HQV 125W) for 4 minutes while shaking at 6 rpm. After UV mutagenesis the suspension was kept in the dark for 16 hours at 4°C. This procedure yielded a killing of 90-98% of the cells, which was tested by plating dilutions of the mutagenized and non-mutagenized suspension on YePD-agar (YePD containing 15 g/l agar), and growth at 30 °C. UV dosage was adjusted when the killing was too low or too high. The next day the cells were resuspended and spread on YNB-D agar-plates containing 8-azaguanine (8-AG plates). The 8-AG plates were freshly prepared and contained 6.8 g/l Yeast Nitrogen Base without amino acids (YNB, Difco), 20 g/l glucose (Merck) and 15 g/l agar (Difco). After autoclaving at 1 10 °C for 15 minutes, the agar solution was cooled to 50 °C before 20 ml of a solution of 50 g/l 8-azaguanine (Sigma- Aldrich) in 1 N NaOH was added per litre of agar medium. Final concentration of 8- azaguanine in the agar plates was 1 g/l. This concentration is significantly higher then described in literature, for selection of 8-AG resistant yeast strains (Woods et al. (1983) Mol. Gen. Genet. 191 , 407-412). Before use the plates were stored at 30 °C since storage in the cold leads to crystallization in the plates. After spreading, the 8-AG plates were incubated at 30 °C for several days until colonies appeared (generally 5-17 days incubation were required). Single colonies were picked and re-streaked on YePD-agar plates and incubated at 30 °C until single colonies appeared (generally 1-2 days incubation). A single colony of each mutant was resuspended in 1 ml 0.9% NaCI. 10 μl of this suspension was spotted onto 8-AG plates. The parent Saccharomyces cerevisiae was used as control strain, and showed no growth on 8-AG plates. Plates were incubated at 30°C for 10 days. Strains that grew on 8-AG plates were marked as 8-AG resistant and kept for further analysis. Using this procedure 114 8-AG resistant yeast strains were isolated.

Example 2

Shake flask selection 100 ml YePD medium was inoculated with the parent S. cerevisiae or a 8-AG resistant strain from YePD plate, and grown for 24 hours at 30°C and 280 rpm. The optical density at 600 nm (OD₆₀₀) was determined in a UltroSpec 2000 (Pharmacia Biotech) of 50-fold diluted precultures. 100 ml YeP-molasses (10 g/l Yeast Extract, 20 g/l Bacto-peptone, 2% (v/v) beat-molasses) was inoculated with precultures untill OD₆oo=1.0. Cultures were incubated for 16 hours at 30°C and 280 rpm. After growth the OD₆oo of the cultures was determined and samples of 60 OD units (in x ml) were taken and centrifuged for 15 minutes at 5000 rpm at 4°C in Greiner tubes, x ml minus 0.6 ml of the supernatant was removed, and the cell pellet was resuspended in the remaining fluid. Using this procedure the final volume and cell concentration in the sample is always the same, irrespective of the OD₆₀o of the culture. The suspensions were incubated for 5 minutes at 95°C. 0.60 ml 4M NaOH was added and the suspension was mixed and incubated at 70°C for 30 minutes. After cooling down to room temperature, 1.2 ml 2M H₂SO₄ was added and mixed. 1.6 ml Pj buffer (0.05M KH₂PO₄, pH3.35) was added. The samples were filtered through a 0.45μm filter after being mixed very well and diluted 4 times in Pi buffer. GMP levels were determined by HPLC analysis of the alkaline hydrolysate of 78 strains. The rest of the strains had very poor growth in YeP-molasses. The RNA in yeast samples was hydrolysed during the above-mentioned alkaline treatment. The break down products 2'-GMP and 3'-GMP of RNA and 5'-GMP were quantified by means of HPLC, using a Whatman Partisil 10-SAX column, a posphate buffer pH 3.35 as eluent and UV detection. The surface area of 2'-GMP and 3'-GMP and 5'-GMP were calculated on basis of a 5'-GMP standard. GMP levels of the mutant strains varied between 72 and 160 % of the parent S. cerevisiae, when corrected for the OD₆₀₀. 18 Strains gave an increase in GMP level compared to the parent strain. These strains and 3 strains with GMP level just below the parent strain were re-tested in triplet in shake flask. Surprisingly, five of the mutant strains had an intracellular GMP level, which is significantly and reproducible higher than the parent strain; these strains were termed STRAIN-1 , STRAIN-2, STRAIN-3, STRAIN-4 and STRAIN-5 (Table I). It is striking that the three mutants with the highest GMP levels do not grow as good as the parent strain on YeP-molasses in shake flask. However, mutants with low OD₆₀o do not automatically have a higher GMP level.

Table I. Relative OD600 and GMP content of 8-AG resistant yeast strains grown in shake flask. Numbers are averages of 4 independent experiments for the mutants, and 18 independent experiments for the parent strain. Standard deviations are included in the table.

The accumulation of GMP in the biomass of the mutant strains was not expected since it is described in literature that 8-azaguanine resistant yeast strains excrete purines in the growth medium, with a marked reduction in resistance in the absence of purine excretion (Lomax and Woods Mol. Gen. Genet. 120:139-149 1973). To test if the selected mutants also excrete purines a plate assay as described by Lecoq et al. (Lecoq, K., Konrad, M., Daignan-Fornier, B., Genetics, 156:953-961 2000) was performed. Surprisingly, after spotting of the strains on a lawn of ade- cells, no growth of the tester strain was found (not shown). This suggests that the 8- azaguanine resistant strains according to the invention preferably do not excrete purines, in contrast to the purine-analogue resistant yeast described in the art.

Example 3 Yeast fermentation Several strains with increased GMP levels (STRAIN-1 , STRAIN-2, and STRAIN- 4), were tested in fed-batch fermentation in 10 litres laboratory fermentors with a net culture volume of 6 litres. For the complete description of a fed-batch process reference is made to [Fermentative Capacity in High-Cell-Density Fed Batch Cultures of Baker's Yeast, Pirn van Hoek, Erik de Hulster, Johannes P. van Dijken, Jack Pronk, Biotechnology and Bioengineering, Vol. 68, NO. 5, June 5, 2000]. Temperature and pH during the fermentation where kept in the range as described in the above-mentioned article. The cells where grown on industrial medium containing molasses. The fermentations were carried out under aerobic conditions and with incremental feeding of molasses and nitrogen. Yeast cells were fermented in fed-batch fermentations which were characterized by carbon limitation. Strains STRAIN-3 and STRAIN-5 did not grow as well as the other strains in shake-flask cultures and therefore these strains were not tested in fed-batch fermentations.

Table II: Fermentation yield (gram yeast dry-weight / gram consumed sugar) and GMP content (% of dry-weight) of different yeast strains was determined and shown as percentage of the parent strain. After fed-batch fermentation of the three mutant strains, the fermentation yield, dry-weight and GMP content of the yeast biomass was determined. Results of this analysis are depicted in Table II. From these results it is clear that all three analysed strains showed an increased GMP content in the fed batch fermentation, and had a good fermentation yield. From these results strain STRAIN-4 was selected for further analysis by extract preparation. Example 4

Yeast extract production and analysis Biomass of yeast strain STRAIN-4 was used in the production of a yeast extract. As a comparison, biomass of the corresponding parent strain was also used in the production of a yeast extract using the same conditions. The yeast biomass was heat treated in a continuous flow-through heat exchanger for 10 minutes at 95 °C in order to inactivate all yeast enzyme activity. Subsequently this inactivated yeast was treated batch-wise for 6 hours with Pescalase (endo-protease from Bacillus licheniformis, DSM N.V., The Netherlands) at pH 8.0 and 62 °C. Thereafter, the protease was inactivated by heat treatment for 1 hour at 70 °C (batch-wise) and the pH was lowered to 5.3 with hydrochloric acid. The mixture was then incubated batch-wise for 15 hours at pH 5.3 and 65 °C with the enzyme 5'-phosphodiesterase in order to hydrolyse the RNA into 5'- ribonucleotides. The solids were removed from the reaction mixture by continuous centrifugation. In this yeast extract the content of all 5'-ribonucleotides was quantified using

HPLC, using a Whatman Partisil 10-SAX column, a posphate buffer pH 3.35 as eluent and UV detection. Concentrations were calculated on basis of GMP, AMP, CMP and UMP standards. In Table III the results of this analysis is shown:

Table III: % ribonucleotide in different strains. Percentages are calculated as weight percentage of free 5'-ribonucleotides based on sodium chloride free yeast extract dry-matter. Surprisingly, in strain STRAIN-4 not only the percentage of purine-containing ribonucleotides (GMP + AMP) was increased when compared with the parent strain, but also the percentage of the pyrimidine-containing ribonucleotides (CMP + UMP). This was unexpected since these strains are isolated for increased 8-azaguanine resistance, interfering only with the purine nucleotide biosynthesis pathway. Since also the pyrimidine-containing 5'-ribonucleotides were increased in amount in the yeast extract, it is suspect that total RNA content of strain STRAIN-4 is increased. Apparently it is possible to isolate yeast strains with increased RNA content by selection for 8- azaguanine resistance.

Example 5

Higher RNA in mutant strains To test whether the measured increase in purine and pyrimidine ribonucleotides was due to an increase in the pool of free ribonucleotides, or due to an increase in RNA content, an experiment was performed where this distinction could be made. For this experiment, the strains that were used in the yeast fermentation of Example 3, STRAIN-1 , STRAIN-2, STRAIN-4 and the parent strain, were used. All four strains were fermented in triplicate in shake flask as described in Example 2. Two samples were taken from each culture as described in Example 2. One of these samples was treated as described in Example 2, including an alkaline hydrolysis with NaOH and a neutralization step with H₂SO . These samples contained both the ribonucleotides derived from the hydrolysis of RNA and the ribonucleotides belonging to the free pool. The second sample was treated as above except that 1.2 ml 1M Na₂SO₄ was added instead of 0.6 ml 4M NaOH, and 0.6 ml water was added instead of 1.2 ml 2M H₂SO₄. In these samples the RNA was not hydrolysed and they contained only the ribonucleotides belonging to the free pool. GMP levels of all samples were determined exactly as described in Example 2. Total GMP levels of the three mutant strains and the parent strain were comparable to the values found in Example 2. Percentage of the increase in total GMP that was due to an increase in the free pool GMP was calculated with the formula 100*(a-b)/(c-d); a. is the GMP concentration of the free nucleotide pool from the mutant strain; b. is the GMP concentration of the free nucleotide pool from the parent strain; c. is the GMP concentration of the total nucleotide pool from the mutant strain; d. is the GMP concentration of the total nucleotide pool from the parent strain. Percentage of the increase in total GMP that was due to an increase in the bound GMP (RNA) was calculated as 100 - % free GMP. Results are represented in Table IV. From this it is clear that almost all increase in GMP in the mutants is due to an increase in RNA. Using the method of the invention yeast strains with higher RNA content when compared with the parent strain can be isolated.

Table IV: Total increase in GMP content in the mutants is divided in increase due to the free GMP pool, and increase due to GMP in RNA (bound GMP). Standard deviations are calculated from three independent samples.

Claims

CLAIMS 1. Process for the preparation of a yeast strain resistant to a purine base analogue comprising: a) subjecting a parent yeast strain to mutagenesis; b) selecting among the yeast strains obtained in a) a yeast strain which is resistant to a purine base analogue; c) selecting among the yeast strains obtained in b) a yeast strain which has a higher intracellular amount of purine-containing ribonucleotide when compared with the parent strain. 2. Process according to claim 1 , wherein the yeast strain belongs to the genus Saccharomyces. 3. Process according to claims 1 or 2, wherein the yeast strain belongs to the species Saccharomyces cerevisiae. 4. Process according to any one of claims 1 to 3, wherein the purine base analogue is 8-Aza-Guanine or 8-Aza-Adenine. 5. Process according to claim 4, wherein the purine base analogue is 8-Aza- Guanine. 6. Process according to any one of claims 1 to 5, wherein in step c) a yeast strain is selected with an intracellular amount of purine-containing ribonucleotide, which is at least 5% higher, preferably at least 10 % higher when compared with the parent strain. 7. Yeast strain resistant to a purine base analogue obtainable by the process according to any one of claims 1 to 6. 8. Yeast strain according to claim 7 which also has a higher intracellular amount of pyrimidine-containing ribonucleotide when compared with the parent strain. 9. Yeast strain according to claim 8, which has a higher RNA content when compared with the parent strain. 10. Process for the preparation of a yeast extract comprising 5'- ribonucleotides comprising: a) fermenting a yeast strain resistant to a purine base analogue according to any one of claims 7 to 9 in a culture medium until a desired cell density is achieved; b) separating the yeast cells from the fermentation broth; c) treating the yeast biomass to inactivate at least a substantial part of the native RNA degrading yeast enzymes and at least a substantial part of the native 5'-ribonucleotide degrading yeast enzymes; d) subsequently treating the yeast biomass to open up the yeast cells; e) treating the opened yeast cells with an enzyme suitable to convert RNA into 5'- ribonucleotides. 11. Process according to claim 10 which further comprises: treating the mixture obtained in e) with an enzyme suitable to convert 5'-AMP into 5'- IMP. 12. Process according to claim 10 or 11 which further comprises: separating the solid fraction from the liquid fraction and concentrating or drying the liquid fraction. 13. Yeast extract comprising 5'-ribonucleotides obtainable by the process according to any one of claims 10 to 12. 14. Use of a yeast strain according to any one of claims 7 to 9 or of a yeast extract comprising 5'-ribonucleotides according to claim 13 in food or feed. 15. Use of a yeast strain according to any one of claims 7 to 9 in the preparation of an autolytic yeast extract. 16. Use of a yeast strain according to any one of claims 7 to 9 in the preparation of a hydrolytic yeast extract.