DK153571B - Process for the preparation of D-glucosone - Google Patents

Description

DK 153571 B

The present invention relates to a process for the preparation of D-glucosone.

D-glucosone is a suitable starting material for preparing crystalline fructose. The unique physical, chemical and metabolic properties of crystalline fructose provide special applications not available for g 1ucose isomerase-produced high fructose corn syrup (HFCS), typically containing 42% fructose, 50% dextrose and 8% polysaccharides. While HFCS is about as sweet as sucrose, crystalline fructose is approx. 50% sweeter and can be used at lower concentrations to produce the same sweetness. HFCS primarily competes on the direct cost of liquid sugars in a wide variety of traditional food uses in either dry or liquid form, primarily for pharmaceutical applications and in foods where a reduction in calorie content per day is desired. sødhedsenhed. Crystalline fructose is used in diet sweets and ice creams and can conveniently supplement or replace glucose, sucrose or HFCS in beverages.

20

The production of crystalline fructose on a commercial scale has encountered a number of serious problems.

One of the biggest problems is the manufacturing cost. The only commercially acceptable production method so far was based on first preparing a glucose fructose syrup mixture from glucose (using either glucose isomerase or alkaline isomerization) or sucrose (via invert sugar), then physically separating the two sugars (ion exchange or selective calcium salt precipitation) and finally to recover the fructose from the aqueous solution in crystalline form (grafting or methanol precipitation). The physical separation treatment of the two sugars is necessary since crystalline fructose can be difficult and frequently impossible to recover from an aqueous solution at best, unless substantially all ionic substances, residual glucose and other contaminants are removed.15 Such treatment is expensive and therefore results in a high market price for fructose - a price that cannot compete with cane sugar.

2

DK 153571B

The literature indicates that low concentrations of fructose can be prepared from glucose. As early as 1889, fructose was formed by reduction of D-glucosone with zinc powder in aqueous acetic acid.6 This reduction has been used as an analytical test for the determination of D-glucosone in several biological substances.8 »'12 Reduction of D-glucosone Glucosone for D-fructose has also been carried out with sodium borohydride. Conversion of D-glucosone to fructose is a possible enzymatic reaction, as reductases which effect the reduction of -CHO - CH2OH are known.

The literature also indicates that low concentrations of glucosone can be prepared from glucose. Several methods have been described for oxidizing glucose - with 1 ^ 024 or with 15 copper acetate.5 In each of these reactions, the conversion yields are low (<30%) and many by-products occur. Glucose has been converted to D-glucosone by first preparing a derivative of the glucose (for example, by reaction with phenyl hydrazine to produce glucosazone 6 or by reaction with p-to-luidin7) and then chemically treating this derivative to form D-glucosone. In these reactions, the yields were not more than 50% and the non-recovered reagents are too expensive for commercial use. The use of flavoring reagents could additionally cause problems in obtaining sugars of reasonable quality.

Conversion of glucose to D-glucosone is also a known enzymatic reaction. As early as 1932, it was stated that glucose was oxidized to D-glucosone by the crystalline form of a mollusk, Saxidomus giganteous.8 In 1937, 30 was stated that D-glucosone is formed by the oxidation of glucose, starch, maltose or sucrose with plasmolyzed preparations of two molds, Aspergillus parasiticus and Aspergillus flavus oryzae.

In 1956, the enzymatic oxidation of glucose to glucosone in a red alga, Iridophycus flaccidum, 10 was described. A carbohydrate oxidase was isolated from the mycelium of the Basidiomycete, Polypyporus obtusus, which oxidized glucose to D-glucosone. ** There 3

DK 153571B

nothing was stated about yields. Finally, in 1978, glucose-2-oxidase activity was detected in the basidiomycetine, Oudeman-siella mucida, as well as in other wood-decomposing fungi, and the best yields reported were not more than 50%. In all 5 of these cases, it is believed that D-glucosone production is accompanied by hydrogen peroxide formation.

It has now been found that in high yield, D-glucosone can be prepared from glucose by a process which is property-like by providing an aqueous solution of glucose and converting ca. 95% of the glucose in solution to D-gluconone by enzymatic oxidation with an oxidoreductase enzyme in the presence of oxygen while removing or utilizing co-produced hydrogen peroxide.

15

The invention is explained in more detail with reference to the accompanying drawings, in which fig. 1 is a schematic diagram of a preferred example of the method of the invention; FIG. Figure 2 shows the molecular structure of D-glucose; 3 shows the molecular structure of D-glucosone; and FIG. 4 shows the molecular structure of D-fructose.

25

As to the molecular structure of D-glucosone, those skilled in the art will know that it has been claimed that several other forms of the molecule exist in aqueous solution. A number of these are cyclic.23 In the present specification, the terms "glucose", "D-glucose" and "dextrose" are used interchangeably to include this monosaccharide in some form - solution or dry. Figure 2 represents glucose.

In the present specification the terms "D-glucosone" and "D-arabino-2-hexosulose" are used interchangeably. Figure 3 represents D-glucosone.

4

DK 153571B

In the present description, the terms "fructose", "D-fructose" and "levulose" mean the isomer of glucose which is sweeter than glucose. The term "crystalline fructose" is used in the present text to denote this monosaccharide in anhydrous form. Figure 4 shows fructose.

In general, according to the present invention, glucose in aqueous solution is enzymatically converted to D-glucosone with a suitable enzyme such as carbohydrate oxidase or giucose-2 oxidase. j0 This transformation is allowed to proceed spontaneously, quickly and substantially completely. Then, without further isolation, D-glucosone can be converted to fructose by appropriate chemical hydrogenation.

As mentioned and referred to above, the concentration of glucose to D-glucosone is known. According to the present invention, glucose is readily and completely converted to D-glucosone using a purified oxidoreductase enzyme such as carbohydrate oxidase or giucose-2 oxidase. The yield obtained by the present invention exceeds that described in literature 20 for this enzyme. This high conversion to D-glucosone eliminates the need to physically separate any residual unchanged glucose.

Certain oxidoreductases have a catalytic effect in oxidizing the hydroxyl group on the second carbon of glucose, promoting the oxidation of this hydroxyl group to a keto group, e.g. by converting the structure of FIG. 2 to the structure of FIG. 3. These specific oxidoreductase enzymes described in the Examples herein are variously referred to as "glucose-2-oxidase", "pyranose-2-oxidase" and "carbohydrate oxy-dase", but the invention is not necessarily limited to such enzymes. Giucose-2 oxidase has a high specificity to glucose as a substrate, whereas carbohydrate oxidase, although it has glucose as the preferred substrate, has a broader substrate specificity. Any enzyme capable of converting the hydroxyl group on the second carbon of glucose into a keto group and not otherwise significantly affecting the remainder of the giucose molecule falls within the present invention.

DK 153571B

scope of invention. Such an enzyme can be specified as an enzyme having glucose-2-oxidase activity.

A preferred carbohydrate oxidase enzyme is derived from the microorganism Polyporus obtusus. Sources of glucose-2 oxidase include several other microorganisms, mollusks and red algae, as referenced above. These enzymes and their sources are indicative only and are not to be considered as encompassing all suitable enzymes and their sources within the scope of the present invention.

10

In order to facilitate the discussion, various features of the present invention will be described specifically but not exclusively in connection with the use of the preferred carbohydrate oxidase from Polyporus obtusus and the glucose-2 oxidase from Aspergil-15 lus oryzae. The microorganisms can be grown in stirred, immersed culture at room temperature by conventional methods. The enzyme is prepared from mycelia of the microorganism grown under conditions of moving, immersed culture.

The enzyme is preferably used in immobilized form, although free enzyme can also be used. The methods of enzyme immobilization are well known to those skilled in the art and consist of reacting a solution of the enzyme with one of a large number of surface-treated or untreated organic and inorganic support materials. Included herein are polyacrylamide, ethylene-maleic acid copolymers, methacrylic-based polymers, polypeptides, styrene-based polymers, agarose, cellulose, dextran, silica, porous beads, charcoal or carbon black, wood and sawdust, hydroxyapatite and aluminum or titanium hydroxide. Enzymes in this form have increased stability, extended life and usefulness and extensibility. Reactions using immobilized enzymes can be carried out in columns or reaction vessels or other suitable reactors.

35

In addition to the carbohydrate oxidase or giucose-2 oxidase enzymes, a source of oxygen is needed. A process for hydrogen peroxide removal or utilization is also required in the reaction to 6

DK 153571B

conversion of glucose to D-glucosone in the most efficient way. This is because H2O2 oxidizes certain critical sites on the enzyme molecule, which damages its function. Methods for hydrogen peroxide removal include (1) decomposition 5 by the enzyme catalase, (2) decomposition by known chemical means and (3) decomposition using decomposition matrices such as manganese oxide or carbon black16'17 as the immobilizing support material for the oxidoreductase enzyme. In a preferred alternative method, the hydrogen peroxide formed, rather than being decomposed, can be consumed to produce a valuable co-product. Coupling D-glucosone production in 11 with propylene halohydrin or propylene oxide preparation is a preferred example as shown in FIG. First

15

The enzymatic conversion of glucose to D-glucosone is preferably carried out in water at approximately neutral pH, but can be carried out within the pH value range easily from ca. three to approx. eight using appropriate buffers. This conversion is preferably carried out at ambient temperature, but can be carried out within the temperature range of from approx. 15 ° C to approx. 65 ° C. The pressure is preferably atmospheric pressure, but may vary from below pressure to atmospheric pressure. Any carbohydrate material that provides glucose or chemically through glucose is a suitable source of glucose for conversion to D-glucose. These substances include, but are not limited to, the following: cellulose, starch, sucrose, corn syrup, HFCS and other syrups containing various amounts of glucose and fructose.

30

The samples used to analyze the sugars are given below: 1. Thin Layer Chromatography (TLC). Avicelbel like glass plates are developed with an 8: 8: 2: 4 (by volume) isopropanol: pyridine: acetic acid: water-solvent system. Glucose has an Rf value of 0.5-0.6. D-glucosone has an Rf of 0.4-0.5 bar. (Rf = the distance that the substance travels from the starting point / solvent front distance from the starting point). When pla- 7

DK 153571B

there is sprayed with tri phenyltetrazolium chloride (2% TTC in 0.5N NaOH), immediately giving D-glucosone red spots; glucose gives only a red spot after heating for 1Q min at 100 ° C, When the plates are sprayed with diphenylamine / ani1 in / phosphoric acid / ethyl 5 acetate reagent (0.15 g / 0.8 ml / 11 ml / 100 ml) , glucose gives a brown stain, and D-glucosone gives a purple streak.

2. High performance liquid chromatography (HPLC). A μ-Bondapak carbohydrate column purchased from Waters Associates is run with 15% aqueous acetonitrile containing 0.001 M potassium phosphate buffer, pH 7, at a flow rate of 2 ml / min. Glucose has a retention time R * of 11.5 and D-glucosone an R * of 14.0. Experiments are performed on a spectra physics SP8000 instrument using both a Waters Associates refractive index detector and a Schoeffel UV variable wavelength detector set to 192 nm.

3. Mass Spectrometry (MS). The following derivatization protocol is used to make the chemical components volatile: 20 (a) 100 mg of the lyophilized 1 sample is added 110 mg of N, N-di phenylhydraziη (H2NNO2) and 1 ml of 75% aqueous ethanol. The reaction mixture is vortexed and then allowed to stand overnight at room temperature.

(b) 3 ml of water is added to the reaction mixture and the resulting precipitate is separated from the supernatant by centrifugation and decanting. To this precipitate is added 1 ml of a 1: 1 mixture of pyridine / acetic anhydride. The reaction mixture is placed in a 35-40 ° C water bath for 15 minutes with occasional shaking.

(C) 2 ml of water is added to quench the reaction and then the mixture is extracted twice with 3 ml portions of ethyl ether. The ether is dried over a small amount of anhydrous sodium sulfate and then the ether is stripped off by gentle heating (-40 ° C) and flowing nitrogen.

(D) The resulting solid or resultant syrup is ready for mass spectrometric analysis.

8

DK 153571B

Expected reactions: -CHO -CH = NNo2> C = O (1) h2NNo2- ^ C = O (unaffected) 5 -OH (2) Ac2O, pyridine -OAc

Glucose gives a molecular ion at mass 556 (C28H32N2 ° lo) ° S a base ion at mass 168 (No2 ion fragment); D-glucosone provides a molecular ion at mass 512 (C2gH28N2Og) and an intense diagnostic fragment ion at mass 223 (OC-CH = N-No2 ion fragment). Experiment 10 is performed on a Finnigan GC / MS / DS Model 4023 instrument set for 70 eV electron impact ionization and to a transmit temperature of 220 ° C.

4. Colometric test tube sample. The analysis of D-glucosone in the presence of an excess of glucose is readily determined by two colometric methods. Using triphenyl tetrazolium chloride (TTC), the method of determination is based on a differential rate of reduction of the two sugars. To a 20 ml test tube is added 0.5 ml of sample plus 0.1 ml of 1% aqueous TTC plus 0.4 20 ml of 6N NaOH. After exactly 5 minutes, 15 ml of acetic acid / ethanol (1: 9) is added and the contents of the test tube are swirled. With water as the blank, absorbance is measured at 480 nm using a Varian 635 UV / VIS spectrometer. Glucose reduces TTC to a red pigment, a triphenylformazan, approx. 100 times 25 slower than an equivalent amount of D-glucosone. Using diphenylamine / aniline / phosphoric acid reagent, the assay method is based on the different colors formed with sugars of different structures. To a 20 ml test tube is added 0.2 ml of sample plus 5 ml of the following reagent mixture: 30 diphenylamine 0.15 g aniline 0.80 ml isopropanol 100 ml phosphoric acid 11 ml

The test tubes are placed in a 37 ° C water bath for 60 minutes. Glucose gives a yellow-colored solution; D-glucosone provides a purple solution.

9

DK 153571B

Sources of the pure sugars used in the various analytical aspects are listed below: 1. D-glucose was purchased from Applied Science Laboratories, c 99% pure.

ISLAND

2. D-glucosone was chemically synthesized by the following method: Mix 20 g of glucose with 1 liter of distilled water containing 27 ml of glacial acetic acid. 44 g of phenylhydrazine are added. The reaction is carried out for 3 hours at 80 ° C with vigorous stirring 10 by means of a mechanical stirrer and then cooled to room temperature overnight. The solid is filtered off and washed with 10% acetic acid, water and then ethyl ether. The solid is carefully dried in a vacuum oven at 50 ° C. The experimental yield is 16.1 g of glucosazone. The glucose zone is placed in a 3-neck, 3-liter flask and 450 ml of ethanol, 750 ml of distilled water and 9 ml of glacial acetic acid are added. 27.8 g of fresh benzaldehyde are added and refluxed with vigorous stirring by means of a mechanical stirrer. The reaction is refluxed for 5 hours. The condenser is inverted and 450 ml is distilled over, adding 2 750 ml of distilled water (via a drip funnel) to the flask. The reaction mixture is cooled overnight to allow precipitation of benzaldehyde phenylhydrazone. The solution is filtered and the residue is washed with distilled water (~ 1 liter) until the water becomes clear. The filtrate plus the washings are concentrated to 25,500 ml and then extracted with 10 x 300 ml portions of ethyl ether. To get rid of residual ethyl ether in the aqueous solution, place it on a rotary evaporator for 30 minutes. The aqueous solution is passed through a 4 x 100 cm column containing carefully acetone-washed Amberlite XAD-2.

The column is washed with an additional 200 ml of water to remove residual glucosone. The combined aqueous portions are lyophilized. The experimental yield of glucosone is 3.4 g (16% overall yield).

The following examples illustrate the method of the invention. Unless otherwise stated, all temperatures are ambient temperature (about 25 ° C) and all pressure ambient pressure (about 1 atm).

10

DK 153571B

Example 1 Substantially complete conversion of glucose to D-glucosone using immobilized carbohydrate oxidase is shown in this example.

Glucose (2 g) is added to 20 ml of distilled water in a 100 ml Pyrex flask and the sugar solution is stirred. Oxygen gas is bubbled into the flask and 3 mg of catalase (Sigma Chemical Co., C-40, from bovine liver) is added. Agarose-immobilized carbohydrate oxidase 10 prepared as indicated below from 200 ml of culture is also added to the flask.

For the preparation of the enzyme, mycelium pads of Polyporus obtusus ATCC No. 26733 are grown on oblique yeast / malt extract agar substrates as follows: yeast extract (3 g), malt extract (3 g), agar (20 g), peptone (5 g) and glucose (10 g) is added to distilled water (1 liter) and the pH is adjusted to 6.7. The medium is sterilized at 121 ° C for 15 min. The pH is then adjusted to 6.4. The organism is seeded on the oblique agar substrates and 20. grown for 7 days at 25 ° C. The organism grown on the inclined substrate is then used to inoculate yeast / malt extract medium (20 ml of medium into 125 ml of Erlenmeyer flask) prepared as above (but without the addition of agar). The organism is grown for 9 days on a rotary shaker at 25 ° C. The culture is vacuum filtered through Whatman paper # 541 in a Buchner funnel. The mycelia retained on the filter paper contain the enzyme.

The mycelia, obtained from 400 ml of culture, are washed twice with 0.05 M 30 potassium phosphate buffer at pH 7.0. The mycelia are then placed in a Waring mixer containing 70 ml of 0.05M potassium phosphate buffer at pH 7.0 and then homogenized for 3 minutes. The mixture is then centrifuged at 6000 rpm. min for 20 min, and the supernatant is decanted from the solid. To the supernatant, placed in a 500 ml Erlenmeyer flask, add 19 g of polyethylene glycol (weight 4000) and the solution is stirred for 30 minutes. The suspension is then centrifuged at 7000 rpm. min for 20 min. The supernatant is decanted off and discarded. 15 ml 0.2M Na

DK 153571B

π of tri-chloride plus 15 ml of 0.05 M potassium phosphate buffer at pH 7.0 is then added to the precipitate and swirled. The solution is allowed to stand for 30 minutes, during which time a precipitate is formed. The mixture is centrifuged at 14000 rpm.

5 pr. min for 20 min. Oer is decanted into an opaque white supernatant containing cell-free, purified enzyme.

Immolization of the enzymes on agarose can be carried out as follows: The cell-free purified enzyme is dialyzed against 500 ml of distilled water overnight. Then 5 ml of 0.1 M sodium bicarbonate at pH 8.0 is added. To this solution is added 5 g of activated CH-sepharose 4B (washed and repurposed on a sintered glass filter using 500 ml of ImM HCl). Using an end-over-end mixing apparatus, the gel-gel pension is mixed for one hour at 25 ° C. The gel suspension is then washed first with 40 ml of 0.1M sodium bicarbonate at pH 8.0, then with 40 ml of 0.05M Tris buffer at pH 8.0 containing Q, 5M sodium chloride and then with 0.5M sodium formate buffer at pH 4.0 also containing 0.5M sodium chloride.

20

Samples of the reaction mixture are taken at different times and analyzed for glucose and D-glucosone. Using HPLC, quantitative determination of the peak areas of the peaks is done at Rt of 11.5 min (glucose) and Rt of 14.0 min (D-glucosone) and concentrations of sugar present are calculated. The following results are obtained:

Reaction time, Glucose, Q-glucosone,% glucose converted hours_g_g_to D-glucosone 0 2.0 0.0 0 30 1 1.2 0.8 40 2 0.5 1.5 75 3 0.2 1.8 90 4 0 The significant conversion of glucose to D-glucosone is also shown by the disappearance of the spot at Rf of 0.58 (glucose) and the appearance of the dash at Rf of 0.43-0.49 (D-glucosone ) during the reaction and by increasing the absorbance at 480 nm i

DK 15357TB

12 during the reaction using the TTC colorimetrit tube sample.

That the product is definitely D-glucosone is confirmed by derivatization followed by mass spectrometry as described previously, The diagnostic D-glucosone mass ions at mass 512 (molecular ion) and mass 223 (0C-CH = N-No 2 ion fragment) are obtained for product.

Example 2 10

The following represents an example of substantially complete conversion of glucose to D-glucosone by the action of immobilized carbohydrate oxidase.

Glucose (1 g) is added to 50 ml of distilled water in a 250 ml 1 o

Pyrex flask and the sugar solution is stirred. Agarose immobilis in serous carbohydrate oxidase, prepared as in Example 1 of 50 ml of cell-free, purified enzyme, is then added to the flask with 1 mg of catalase (as in Example 1).

20

Eighteen hours later, the aqueous solution is decanted from the solid. Analysis of this solution shows that 99% of the gi ucose has been converted to D-glucosone.

Example 3 25

This example shows essentially complete enzymatic conversion of glucose to D-glucosone using giucose-2 oxidase.

The reaction and conditions of Example 1 (using agarose as the immobilizing support) are repeated, using giucose-2 oxidase instead of carbohydrate oxidase. After 5 hours of reaction, more than 99% of the glucose was converted to D-glucosone.

35 To prepare the enzyme giucose-2-oxidase, mycelial cultures are grown by Aspergillus oryzae ATCC no. 7252 in meat, yeast extract / tryptone medium as follows: meat extract (5 g), yeast extract (5 g), tryptone (3 g), dextrose (1 g) and Difco starch (24 g) are added to distilled water ( 1 liter) and the pH is adjusted to 7.3. The medium is sterilized at 121 ° C for 35 minutes. Using spores obtained in a generally known manner, the medium is germinated to reach βf ca. 3 x 10 4 spores / ml and grown in a rotary row 5 steamer (180 rpm) at 30 ° C for 2 days. Cultures are vacuum filtered through Whatman Paper No. 541 in a Buchner funnel and washed several times with water. The mycelia, which are baked on the filter paper, contain the enzyme.

Purification and immobilization of the enzyme can then proceed using the procedure of Example 1.

Example 4 In Example 1, the concentration of free hydrogen peroxide 15 was formed by the reaction of glucose with carbohydrate oxidase compound with the D-glucose formation, kept as low as possible using catalase to decompose the hydrogen peroxide. An alternative method of keeping the concentration of free hydrogen peroxide as low as possible is to link its preparation to a hydrogen peroxide utilizing reaction which yields a desirable (valuable) co-product.

In this example, hydrogen peroxide production is coupled to the production of propylene bromohydrin, an intermediate in the propylene oxide synthesis according to concepts set forth in detail U.S. Patent No. 4,284,723 and U.S. Patent No. 4,247,641. The reaction of glucose and the immobilized carbohydrate oxidase from Polyporus obtusus ATCC no. 26733 to form D-glucosone and hydrogen peroxide are coupled to turnover 30 of immobilized seaweed or algal peroxidase from Coralina sp.

presence of bromide and propylene to form propylene bromine hydrine. The end result of this coupled reaction is then the co-production of glucosone and of propylene bromohydrin, which is readily converted to propylene oxide, as described in U.S. Patent No. 4,284,723 and U.S. Patent No. 4,247,641. Any enzyme capable of oxidizing the hydroxy1 group on the 2-carbon atom in glucose with concomitant production of hydrogen peroxide can be coupled to any halogenating peroxidase and the compound

DK 15357 IB

14 substance is used for alkene halohydrin production in accordance with the teachings of the aforementioned US patents.

Cell-free, purified seaweed or algal peroxidase enzyme is prepared as follows:

Coralina sp., Obtained from the coast of La Jolla, California, is ground in a Virtis 45 homogeneous isator for 5 min in distilled water. The homogenate is rotated at 20,000 rpm. min for 20 min. The supernatant is decanted off and stored. The pellet is resuspended in distilled water and centrifuged again. The supernatant and the aforementioned supernatant are combined. The solution is first brought to 33¾ and then to 55¾ saturation in ammonium sulfate. Centrifugation and separation of the pill is performed at each step.

The 33 ^ 55¾ pill fraction is passed through a DEAE column using a 0.3M-1M phosphate buffer (pH 6.0) gradient. The fraction, eluting at 1M, is dialyzed against 20 mM pho-sphat buffer (pH 6) overnight.

The immobilis in serous seaweed or algal peroxidase is prepared as follows:

Glass beads (obtained from Sigma Chemical Company, PG-700-200) are activated by suspending 1 g of plastic beads in 18 ml of deionized water. 2 ml of 10¾ (v / v) α-amino-propyl triethoxysilane is added and the pH of the mixture is adjusted to 3-5 with 6N HCl. The mixture is shaken at 75 ° C for 2 hours. The plasma beads are then vacuum dried overnight at 80 ° C. 3.2 ml of purified Coralina sp. enzyme prepared as above and 50 mg of water-soluble carbodiimide are added to the beads. The pH is adjusted to 4.5 and the mixture is then shaken at 4 ° C overnight. The product (enzyme-coated beads) is washed with water. Activity is measured as 2 monochlorodimon units / g beads.

Immobilized carbohydrate oxidase on agarose is prepared as in Example 1 of 10 ml of cell-free, purified enzyme.

A reaction mixture containing the following ingredients is placed in a 100 ml Pyrex flask: 35 i. *

DK 15357TB

15 a) 1 g glass beads coated with seaweed or algae peroxidase, b) the immobilized carbohydrate oxidase prepared above, 5 c) 800 mg of potassium bromide and d) 20 ml of 0.01M potassium phosphate buffer, pH 7.0.

Both propylene and oxygen are bubbled to the flasks continuously. The reaction is started with 1 g glucose. After 20 hours, a sample of the reaction is analyzed which is analyzed for residual glucose, D-glucosone and propylene bromohydrin. The resulting propylene bromine hydrine is analyzed as follows: 5 μΐ of the reaction mixture is injected into a Hewlett-Packard Model 402 gas chromatograph fitted with a 1.8 m by 0.32 cm 15 glass column packed with Porapak R (80/100 mesh). The flow rate is set to 30 ml / minute for helium, increase the column temperature to 200 ° C. Retention times for the propylene bromo hydrines are 9 min for 1-bromo-2-propanol and 10 min for 2-bromo-1-propanol.

20

Product identity was confirmed by comparison with authentic samples of propylene bromohydrin: 1-bromo-2-prapanol provided by Pfaltz and Bauer, Inc.; 2-Bromo-1-propanol is synthesized by means of lithium aluminum hydride reduction of 1-bromopropionyl chloride. The reaction products and authentic samples show the same retention times and identical mass spectra: bromine is identified by the presence of M and (M + 2) isotope bonds of similar intensity. The molecular ion of both isomers is confirmed by chemical ionization with isobutane reagent gas (M +; m / e 138 + 140); for 30 l-bromo-2-propanol, the main fragmentation is the expected loss of CH2 = Br, while the main fragmentation of 2-bromo-1-propanol is the expected loss of CH3CHBr.

The analysis of the sample showed a conversion of over 93% of glucose 35 to D-glucosone and propylene bromohydrin conversion of 20 g / l.

DK 153571B

16

Example 5

This example serves to further elucidate the concepts set forth and shown in Example 4. In this case, immobilized carbohydrate oxidase is replaced with immobilized glucose-2-oxidase.

The immobilized seaweed or algal peroxidase enzyme is prepared as described in Example 4. The immobilized glucose-2-oxidase is prepared as described in Example 3.

10

The reaction mixture is prepared as in Example 4, using immobilized carbohydrate oxidase instead of immobilized giucose-2 oxidase.

After 20 hours, a sample of the reaction mixture is taken which is analyzed for residual glucose, D-glucosone and propylene bromohydrin. The results showed a conversion of over 09% of glucose to D-glucosone and a propylene bromohydrin production of 19.5 g / liter.

Example 6

This example illustrates a high conversion of glucose to D-glucose over an extended period of time using immobilized carbohydrate oxidase in a column reactor.

25

Carbohydrate oxidase (cell-free, purified enzyme) (10 ml), prepared as in Example 1, is immobilized on hydroxyapatite (calcium phosphate hydroxide) as follows:

To 100 ml of cell-free, purified enzyme, 20 g of hydroxyapatite is added to 100 ml of 1 mS *] of calcium phosphate buffer at pH 7.0. The mixture is stirred for 3Θ minutes, after which the solid is separated from the liquid by decantation and the solid is first washed with 200 ml of 10 mM potassium phosphate buffer at pH 7.0 and then with 200 ml of distilled water.

35

This fabric is then packed in a glass column (0.5 cm x 4.5 cm). A 1% giucose solution is passed through the column at a flow rate of 1.5 ml per ml. hour. The eluent is periodically analyzed for residual glucose and formed D-glucosone.

Claims (5)

A process for the preparation of D-glucosone, characterized in that an aqueous solution of glucose is obtained and converted to ca. 95% of the glucose in solution to D-glucosone by enzymatic oxidation with an oxidoredicttase enzyme in the presence of oxygen while removing or utilizing co-produced hydrogen peroxide. Process according to claim 1, characterized in that the oxidoreductase enzyme is glucose-2-oxidase from Aspergillus oryzae or carbohydrate oxidase from Polyporus obtusus. 35 Method according to claim 1 or 2, characterized in that the enzyme is immobilized. DK 153571 B Process according to claim 1, characterized in that the hydrogen peroxide is decomposed with catalase. Process according to claim 1, characterized in that it is carried out in a column reactor. 3 10 15 20 25 3 0 35