ENZYME CASCADES BASED ON SUCROSE SYNTHASE AND PYROPHOSPHORYLASE FOR CONVERSION OF ADP TO ATP

Abstract

The present invention relates to a process for the multi-step enzymatic conversion of adenosine diphosphate (ADP) to adenosine triphosphate (ATP), the process comprising the steps of: a) enzyme-catalyzed conversion of adenosine diphosphate in the presence of sucrose and a sucrose synthase to adenosine diphosphate-glucose; and b) enzyme-catalyzed conversion of the adenosine diphosphate-glucose formed in process step a) in the presence of inorganic pyrophosphate and a pyrophosphorylase to adenosine triphosphate and glucose-1-phosphate. Furthermore, the invention relates to the use of the process for the preparation of sugar phosphates, nucleotide sugars, glycans, glycoproteins, glycolipids or glycosaminoglycans.

Claims

1. A process for the multistage enzymatic conversion of adenosine diphosphate to adenosine triphosphate, characterized in that the process comprises at least the steps: (a) enzyme-catalyzed conversion of adenosine diphosphate in the presence of sucrose and a sucrose synthase to adenosine diphosphate-glucose; and b) Enzyme-catalyzed conversion of the adenosine diphosphate-glucose formed in process step a) in the presence of inorganic pyrophosphate and a pyrophosphorylase to adenosine triphosphate and glucose-1-phosphate; wherein process steps a) and b) are carried out in aqueous solution and simultaneously or successively, wherein process steps a) and b) are carried out in a common aqueous reaction solution.

2. (canceled)

3. Process according to claim 1, wherein method steps a) and b) are carried out simultaneously.

4. Process according to claim 1, wherein the pH in process step a) and/or b) is greater than or equal to 5.0 and less than or equal to 8.5.

5. Process according to claim 1, wherein process step a) and/or b) is carried out in the presence of fructose-1,6-bisphosphate at a concentration greater than or equal to 5 μM and less than or equal to 200 μM.

6. Process according to claim 1, wherein the inorganic pyrophosphate is formed in process step b) by an enzyme-catalyzed reaction in the aqueous solution.

7. Process according to claim 1, wherein the glucose-1-phosphate formed in process step b) is removed from the aqueous solution by a further enzymatic reaction.

8. Use of the process of claim 1 for in situ provision of adenosine triphosphate in multistage adenosine triphosphate-consuming enzyme cascades in the preparation of compounds selected from the group consisting of sugar phosphates, nucleotide sugars, glycans, glycoproteins, glycolipids, glycosaminoglycans, phospho-adenosine phosphosulfate, nucleotide-activated compounds, or mixtures of at least two compounds from this group.

9. The use according to claim 8, wherein the adenosine triphosphate-consuming enzyme reaction comprises converting N-acetylglucosamine and adenosine triphosphate to N-acetylglucosamine 1-phosphate and adenosine diphosphate by means of an N-acetylhexosamine 1-kinase.

10. The use according to claim 9, wherein the N-acetylglucosamine-1-phosphate is converted in a further enzymatic reaction using uridine triphosphate by means of a uridine diphosphate-N-acetylglucosamine diphosphorylase to uridine diphosphate-N-acetylglucosamine and inorganic pyrophosphate.

11. The use according to claim 10, wherein the glucose 1-phosphate formed in process step b) is converted to uridine 5′-diphosphoGlucose and inorganic pyrophosphate by further enzymatic reaction using uridine triphosphate with a uridine triphosphate monosaccharide 1-phosphate uridylyltransferase.

12. The use according to claim 10, wherein the glucose 1-phosphate formed in process step b) is converted to glucose 6-phosphate by further enzymatic reaction with a phosphoglucomutase.

Description

[0038] The figures show:

[0039] FIG. 1 a two-step enzyme cascade according to the invention;

[0040] FIG. 2-4 the analytical results of the two-step enzyme cascade;

[0041] FIG. 5 a two-step enzyme cascade according to the invention with in-situ generation of PP.sub.i;

[0042] FIG. 6-9 the analytical results of the two-step enzyme cascade;

[0043] FIG. 10 a 3-step cascade based on the individual steps according to the invention;

[0044] FIG. 11-13 the analytical results of the three-step enzyme cascade;

[0045] FIG. 14 a 4-step cascade based on the individual steps according to the invention;

[0046] FIGS. 15-16 the analytical results of the 4-step enzyme cascade;

[0047] FIG. 17 a 5-step cascade based on the individual steps according to the invention;

[0048] FIG. 18-21 the analytical results of the 5-step enzyme cascade;

[0049] FIG. 22 a 5-step cascade based on the individual steps according to the invention.

[0050] FIG. 23-25 the analytical results of the 5-step enzyme cascade;

[0051] FIGS. 26-28 the analytical results for characterization of EcAGPase.

EXAMPLES

I. The Enzymes Used

[0052] In the following examples, the following enzymes were used:

TABLE-US-00001 EC- MWCO Vector Production Enzyme Number Origin [kDa] pET master Tag EcAGPase 2.7.7.27. E. coli 49 22b E. coli His.sub.6 CgAGPase 2.7.7.27. Corynebacterium 44 22b BL21 (DE3) glutamicum NeSuSy 2.4.1.13. Nitrosomonas 91 22b europea AtUSP 2.7.7.64. Arabidopsis 68 16b thaliana SzGlmU 2.7.7.23. Streptococcus equi 49 22b zooepidemicus BlNahK 2.7.1.162. Bifido- 40 22b bacterium longum PmPpA 3.6.1.1. Pasteurella 19 22b multocida

[0053] Cells were transformed with the appropriate vector via heat shock and proteins were expressed in “terrific broth” (TB) medium overnight using isopropyl-β-D-thiogalactopyranoside (IPTG) induction. Cell disruption was performed by sonication and the expressed enzymes were purified by Ni.sup.2+-immobilized metal ion affinity chromatography (IMAC) on HisTrap™ HP columns (GE Healthcare, Chicago, USA) on an AKTApurifier™ (GE Healthcare, Chicago, USA) system. Subsequently, the eluate was dialyzed using dialysis tubing (C. Roth, Karlsruhe, Germany) overnight in the respective storage buffer of the enzyme. EcAGPase and CgAGPase were stored in 100 mM HEPES pH 8; NeSuSy in 100 mM Tris-HCl pH 7 and the remaining enzymes in 100 mM HEPES pH 7.5. Protein concentrations of the eluates were performed after dialysis by Bradford assay using RotiQuant solution (C. Roth, Karlsruhe, Germany).

II. The Two-Step Enzyme Cascade According to the Invention

[0054] The reaction scheme of ATP synthesis reaction from sucrose (Suc) and inorganic pyrophosphate (PP.sub.i) by coupling EcAGPase and CgAGPase results in substeps a) and b) and in the overall summary to:

ADP+Suc .Math.ADP-Glc+Fru  a)

ADP-Glc+PP.sub.i.Math.Glc-1-P+ATP  b)

ADP+Suc+PP.sub.i.Math.ATP+Fru+Glc-1-P  Σ)

[0055] The interaction of the individual enzymatic cascade elements is shown in FIG. 1.

[0056] EcAGPase (2.9 mg/mL) and NeSuSy (0.1 μg/mL) are combined in a one-pot synthesis to synthesize ATP from sucrose and PP.sub.i. For this purpose, ADP and PP.sub.i are present in the experimental series in a concentration ratio of 1:1 at different concentrations (2 mM to 15 mM), respectively. In addition, the synthesis batch contains a MgCl.sub.2 concentration corresponding to the sum of the concentration of ADP and PP.sub.i in the respective experiment (4 mM to 30 mM). In addition, 1 mM fructose bisphosphate (Fru-1,6-P.sub.2) is added to the reaction. The batch is buffered with 100 mM MOPS-NaOH buffer at pH 8 and ATP synthesis is performed at 37° C. The synthesis is stopped at the respective measurement points with 28 mM SDS (final concentration 7 mM) and the analysis of the nucleotides (ADP-Glc, AMP, ADP, ATP) is performed by multiplex capillary electrophoresis (internal standards (final concentrations): 1 mM para-aminobenzoic acid and 1 mM 4-aminophthalic acid). The analytes are detected at 254 nm.

[0057] The analytical results of this conversion are shown in FIGS. 2-4. FIG. 2 shows the course of the ADP concentration, FIG. 3 the course of the ATP concentration and FIG. 4 the course of the ADP-Glc concentration over time.

[0058] The reaction shown in FIGS. 2-4 contained 2930 μg/mL EcAGPase, 0.1 μg/mL NeSuSy, a gradient of ADP and PP.sub.i (1 mM to 15 mM), while the concentrations of sucrose (200 mM) and the activator F-1,6-P.sub.2 (1 mM) were kept constant. The concentrations of MgCl.sub.2 were equal to the sum of the concentrations of ADP and PP.sub.i. The reaction was carried out at 37° C. in 100 mM MOPS buffer, pH 7 for 24 h.

[0059] The combination of NeSuSy with EcAGPase in an enzyme cascade for ATP synthesis from sucrose and PP.sub.i shows that after only a few minutes, the concentration of ADP in the reaction decreases and ATP is formed. However, the synthesis of ATP reaches an ATP synthesis limit between 1.96 mM and 2.14 mM after 30 min of reaction time at ADP and PP.sub.i starting concentrations of 5 mM and 10 mM, respectively, corresponding to ATP yields of 39% (5 mM ADP/PP.sub.i) and 21% (10 mM ADP/PP.sub.i), respectively. The experiments with EcAGPase show that the synthesis of ATP from PP.sub.i and ADP-glucose is subject to the reaction equilibrium of EcAGPase (FIG. 1). From the course of ATP synthesis it can be seen (FIG. 3) that ADP-Glc can be converted very rapidly into ATP by EcAGPase in the presence of PP.sub.i. However, the example of the experiment with 2 mM ADP/PP.sub.i shows that ATP is converted again over a longer period (4 h) when the ADP and PP.sub.i concentrations decrease. During the same period, the ADP-Glc concentration increases (FIG. 4). In ATP synthesis, Glc-1-P is also produced along with ATP and leads to the increase of ADP-Glc concentration in the reaction with increasing Glc-1-P concentration. This means that the enzyme EcAGPase sets the reaction equilibrium with increasing Glc-1-P and synthesizes less ATP.

II. In-Situ Generation of PP.SUB.i

[0060] This reaction sequence involves the in-situ generation of PP.sub.i through the use of a complex cascade involving the use of an AtUSP.

[0061] This ATP synthesis reaction can be represented as follows:

Glc-1-P+UTP .Math.UDP-Glc+PP.sub.i

ADP-Glc+PP.sub.i.Math.Glc-1-P+ATP

ADP-Glc+UTP .Math.UDP-Glc+ATP

[0062] The interaction of the individual enzymatic cascade elements is shown in FIG. 5.

[0063] FIGS. 6-9 show the analytical results of the above implementation.

[0064] AtUSP and EcAGPase are combined in a one-pot synthesis to use PP.sub.i formed in the UDP-Glc synthesis for the synthesis of ATP in a subsequent reaction. The resulting Glc-1-P is again used by AtUSP for the synthesis of UDP-Glc. This attenuates the back reaction of EcAGPase towards ADP-Glc synthesis. This allows a more accurate description of the effect of Glc-1-P on ATP synthesis.

[0065] The experimental setup is as follows: 2.9 mg/mL EcAGPase and 0.5 mg/mL AtUSP are combined in a reaction with 3 mM ADP-Glc, 3 mM UTP, and 1 mM F-1,6-P.sub.2. UDP-Glc is synthesized with starting concentrations ranging from 0.5 mM to 10 mM Glc-1-P over a 10 min period at 37° C. in 100 mM HEPES buffer (pH 8). In addition, the reaction contains MgCl.sub.2 whose concentration was adjusted according to the sum of the concentrations of Glc-1-P and UTP. The synthesis is stopped at the respective measurement points with 28 mM SDS (final concentration 7 mM) and the analysis of the nucleotides (ADP-Glc, AMP, ADP, ATP, UTP, UDP, UMP, UDP-Glc) is performed by multiplex capillary electrophoresis (internal standards (final concentrations): 1 mM para-aminobenzoic acid and 1 mM 4-aminophthalic acid) and the analytes are detected at 254 nm.

[0066] The combination of AtUSP and EcAGPase shows that ATP is formed from the PP.sub.i of the UDP-Glc synthesis reaction and ADP-Glc (FIG. 6). In addition, it is shown that all reactions reach a synthesis limit of 1.2 mM ATP (approx. 40% yield relative to UTP) after only one minute and are hardly influenced by the initial Glc-1-P concentration. It is remarkable that with 0.5 mM Glc-1-P an ATP concentration of 1.1 mM is already reached after one minute. This is also reflected by the very rapid conversion of UTP (FIG. 7) and the very rapid synthesis of UDP-Glc (FIG. 8). On the one hand, this means that PP.sub.i is used by the AGPase for the synthesis of ATP from ADP-Glc (FIG. 9) after the initial Glc-1-P concentration has been converted. On the other hand, this means that the Glc-1-P formed is in turn used by the enzyme AtUSP for UDP-Glc synthesis. This creates a closed loop in which ATP is synthesized from ADP-Glc, UTP, and in situ generated PP.sub.i. However, even here, ATP synthesis is subject to the reaction equilibrium of EcAGPase when the experimental time is prolonged. The concentration of ATP decreases again in the course of the experiment (FIG. 6). ATP should therefore be removed from the synthesis reaction as quickly as possible by combining it with ATP-consuming enzymes in order to maintain the regeneration cycle.

[0067] Through this experiment, it is demonstrated that EcAGPase is able to synthesize ATP from in situ nascent PP.sub.i, in a coupled enzyme reaction. ATP can in turn be converted as a substrate by ATP-utilizing enzymes.

III. Three-Step Enzyme Cascade for the Synthesis of GlcNAc-1-P

[0068] This reaction sequence involves the BlNahK/NeSuSy/EcAGPase 3-enzyme cascade for the synthesis of GlcNAc-1-P using the new ATP regeneration system.

[0069] This ATP synthesis reaction can be represented as follows:

GlcNAc+ATP .Math.GlcNAc-1-P+ADP

ADP+Suc .Math.ADP-Glc+Fru

ADP-Glc+PP.sub.i.Math.Glc-1-P+ATP

GlcNAc+PP.sub.i+Suc .Math.GlcNAc-1-P+Fru+Glc-1-P

[0070] The interaction of the individual enzymatic cascade elements is shown in FIG. 10.

[0071] FIGS. 11-13 show the analytical results of the above implementation.

[0072] Reactions were based on 1.6 mg/mL BlNahK, 1.3 mg/mL EcAGPase, 25 μg/mL NeSuSy, 5 mM GlcNAc, 200 mM sucrose, 0.5 mM to 5 mM ATP, 5 mM PP.sub.i, 10 mM MgCl.sub.2, and 0.5 mM Fru-1,6-P.sub.2. The experiment was performed in 100 mM MOPS buffer, pH 7 at 37° C. for 24 h. GlcNAc-1-P and Glc-1-P were reacted separately to give UDP-GlcNAc and UDP-Glc, respectively, and measured by MP-CE. Here, 250 μg/mL AtUSP; 2.5 mg/mL SzGlmU and 2.6 mg/mL PmPpA with 10 mM UTP and 10 mM MgCl.sub.2 in 100 mM MOPS buffer, pH 7 at RT for 2 h were added to the reaction after removing the synthesis enzymes. ATP regeneration was calculated using the following formula:

[00001]$\mathrm{Reg}{.}_{\left[\mathrm{ATP}\right]}=\frac{c_{\left[\mathrm{UDP}-G1\mathrm{cNAc}\right]}}{c_{\left[\mathrm{ATP}\right]}}.$

[0073] The new ATP regeneration system NeSuSy/EcAGPase is combined with an ATP-consuming enzyme, such as a sugar-1-phosphate kinase (using BlNaHK as an example), for the synthesis of GlcNAc-1-P. ATP is consumed by the kinase BlNahK to form GlcNAc-1-P. The resulting ADP is converted by NeSuSy with sucrose to ADP-Glc and fructose. ADP-Glc is subsequently converted to Glc-1-P and ATP by EcAGPase with the addition of PP.sub.i. In this way, ATP is made available again (regenerated) from ADP for the sugar kinase reaction.

[0074] The experimental setup is as follows: The reaction contains 1.6 mg/mL BlNahK, 1.3 mg/mL EcAGPase and 25 μg/mL NeSuSy, 5 mM GlcNAc, 200 mM sucrose, 0.5 mM to 5 mM ATP, 5 mM PP.sub.i and 10 mM MgCl.sub.2 and 0.5 mM Fru-1,6-P.sub.2. Enzyme reactions are performed in 100 mM MOPS buffer, pH 7 at 37° C. for 24 h in a 96-microtiter plate in a volume of 200 μL. After each measurement time point, 150 μL is removed from each of the reaction mixtures and the enzymes are separated from the reaction by ultrafiltration (30 K filter, cut-off 30 kDa, AcroPrep™ Advance filters; Pall) for 15 min. Then, to 100 μL of sample, 50 μL of a solution of AtUSP, SzGlmU, PmPpA, 10 mM UTP and 10 mM MgCl.sub.2 are added to synthesize the nucleotide sugars UDP-GlcNAc and UDP-Glc. These are then analyzed by capillary electrophoresis. The reaction of the enzymes from the follow-up reaction for nucleotide sugar synthesis are stopped with 28 mM SDS (final concentration 7 mM) and the analysis of the nucleotides (ADP-Glc, ADP, ATP, UTP, UDP, UDP-Glc and UDP-GlcNAc) is performed by multiplex capillary electrophoresis (internal standards (final concentrations): 1 mM para-aminobenzoic acid and 1 mM 4-aminophthalic acid) and the analytes detected at 254 nm. The UDP-GlcNAc concentration is proportional to the GlcNAc-1-P formed. The UDP-Glc concentration is proportional to the ATP formed.

[0075] Reduction of the ATP concentration (2 mM and 0.5 mM) below the substrate concentration (5 mM GlcNAc) shows that GlcNAc is phosphorylated with 41%-68% yield (FIG. 11). The control reaction with 5 mM ATP gives a yield of 77%. ATP is regenerated 1.7 and 4.1 times in the reactions with an initial concentration of 2 mM and 0.5 mM, respectively, in the NeSuSy/AGPase enzyme cascade (FIG. 13). Reduction of ATP concentration by more than 50% (from 5 mM to 2 mM) leads to similar product yields. The detected yield of Glc-1-P after 24 h indicates ATP regeneration over 24 h, which also occurred in reactions with ATP excess. Furthermore, these results support the assumption that during the production process of GlcNAc-1-P, ATP is recycled. However, it also shows that Glc-1-P undergoes saturation during ATP recycling and is degraded during the synthesis process (FIG. 12). Four out of a maximum of ten ATP regeneration cycles (at 0.5 mM ATP) are achieved (coupling efficiency 40%). A further increase is possible by optimizing the enzyme ratios.

[0076] In combination with a sugar kinase (ATP-consuming enzyme), the new ATP regeneration system NeSuSy/AGPase is capable of regenerating ATP from ADP with sucrose and PP.sub.i.

IV. 4-Enzyme Cascade for UDP-GlcNAc Synthesis with ATP Regeneration System According to the Invention

[0077] This ATP synthesis reaction using a 4-enzyme cascade BINahK/SzGlmU/NeSuSy/EcAGPase to synthesize UDP-GlcNAc with new ATP regeneration system can be shown as follows:

GlcNAc+ATP .Math.GlcNAc-1-P+ADP

GlcNAc-1-P+UTP .Math.UDP-GlcNAc+PP.sub.i

ADP+Suc .Math.ADP-Glc+Fru

ADP-Glc+PP.sub.i.Math.Glc-1-P+ATP

GlcNAc+Suc+UTP .Math.UDP-GlcNAc-1-P+Fru+Glc-1-P

[0078] The interaction of the individual enzymatic cascade elements is shown in FIG. 14.

[0079] FIGS. 15 and 16 show the analytical results of the above implementation.

[0080] The ATP regeneration system NeSuSy/EcAGPase is combined with the enzyme cascade BlNahK/SzGlmU for the synthesis of UDP-GlcNAc. GlcNAc is converted to GlcNAc-1-P and ADP with BlNahK consuming ATP. GlcNAc-1-P is then converted with SzGlmU to UDP-GlcNAc with release of PP.sub.i. NeSuSy converts ADP and sucrose to ADP-Glc and fructose. EcAGPase uses the released PP.sub.i and ADP-Glc to form Glc-1-P and ATP, which is thus regenerated.

[0081] The experimental setup is as follows: The synthesis reaction contains 57.5 μg/mL EcAGPase, 58 μg/mL NeSuSy, 84 μg/mL BlNahK, 94 μg/mL SzGlmU, 5 mM UTP, 200 mM sucrose, 0.5 mM Fru-1,6-P.sub.2, 10 mM MgCl.sub.2 and the ATP concentration is 0.5 mM to 5 mM. Reactions are performed on a 200 μL scale in a 96-microtiter plate at 37° C. for 24 h in 100 mM HEPES buffer pH 7. With each measurement time point, 150 μL of a reaction is removed and the enzymes are separated from the reaction by ultrafiltration (30 K filter, cut-off 30 kDa, AcroPrep™ Advance filters; Pall) for 15 min. To determine the resulting Glc-1-P concentration, 50 μL of a solution of AtUSP, PmPpA, 10 mM UTP, and 10 mM MgCl.sub.2 are then added to 100 μL of sample. The resulting nucleotide sugar UDP-Glc is analyzed by capillary electrophoresis. The UDP-Glc concentration is proportional to the ATP formed. The reaction of the enzymes from the follow-up reaction to the nucleotide sugar synthesis are stopped with 28 mM SDS (final concentration 7 mM) and the analysis of the nucleotides (ADP-Glc, ADP, ATP, UTP, UDP, UDP-Glc and UDP-GlcNAc) is performed by multiplex capillary electrophoresis (internal standards (final concentrations): 1 mM para-aminobenzoic acid and 1 mM 4-aminophthalic acid) and analytes are detected at 254 nm.

[0082] Reduction of ATP concentration (2 mM and 0.5 mM) below the substrate concentration (5 mM GlcNAc) shows that UDP-GlcNAc is synthesized with 25%-63% yield (FIG. 15). The control reaction with 5 mM ATP gives a yield of 74%. ATP is regenerated 1.6 and 2.5 times in the reactions with an initial concentration of 2 mM and 0.5 mM, respectively, in the NeSuSy/AGPase enzyme cascade (FIG. 16). Reduction of ATP concentration by more than 50% (from 5 mM to 2 mM) leads to similar product yields. Almost three out of a maximum of ten ATP regeneration cycles (at 0.5 mM ATP) are achieved (coupling efficiency 30%). Further increase should be possible by optimizing the enzyme ratios and by removing Glc-1-P from the reaction equilibrium of EcAGPase.

[0083] In combination with a sugar kinase (ATP-consuming enzyme) and a pyrophosphorylase (PP.sub.i generating enzyme), the new ATP regeneration system NeSuSy/AGPase is capable of regenerating ATP from ADP with sucrose and PP.sub.i.

V. 5 Enzyme Cascade for UDP-GlcNAc Synthesis

[0084] This ATP synthesis reaction using a 5-enzyme cascade BINahK/SzGlmU/NeSuSy/EcAGPase/AtUSP to synthesize UDP-GlcNAc with the ATP regeneration system of the invention can be described as follows:

GlcNAc+ATP .Math.GlcNAc-1-P+ADP

GlcNAc-1-P+UTP .Math.UDP-GlcNAc+PP.sub.i

ADP+Suc .Math.ADP-Glc+Fru

ADP-Glc+PP.sub.i.Math.Glc-1-P+ATP

Glc-1-P+UTP .Math.UDP-Glc+PP.sub.i

GlcNAc+Suc+2 UTP .Math.UDP-GlcNAc+Fru+UDP-Glc+PP.sub.i

[0085] The interaction of the individual enzymatic cascade elements is shown in FIG. 17.

[0086] FIGS. 18-21 show the analytical results of the above implementation.

[0087] The enzyme cascade for UDP-GlcNAc synthesis is completed with the enzyme AtUSP. Glc-1-P is converted to UDP-Glc with AtUSP and thus removed from the reaction of EcAGPase to suppress the back reaction of EcAGPase and provide more ATP for the enzyme cascade BINahK/SzGlmU. This results in a higher product yield for UDP-GlcNAc synthesis.

[0088] The experimental setup is as follows: The synthesis is performed in a one-pot procedure with five enzymes. 41.5 μg/mL NeSuSy, 834 μg/mL BlNahK, 960 μg/mL EcAGPase, 1.2 mg/mL SzGlmU and 75.5 μg/mL AtUSP are used. Synthesis was performed using 5 mM GlcNAc, 10 mM UTP, 0.5 mM F-1,6-P.sub.2, 10 mM MgCl.sub.2, and 0.25 mM to 2.5 mM ATP. The reaction was carried out in 200 μL in 100 mM HEPES pH 7 at 37° C. for 24 h in a 96-microtiter plate. The synthesis was stopped with 28 mM SDS (final concentration 7 mM) and the analysis of nucleotides (ADP-Glc, AMP, ADP, ATP, UTP, UDP, UMP, UDP-Glc, and UDP-GlcNAc) was performed by multiplex capillary electrophoresis (internal standards (final concentrations): 1 mM para-aminobenzoic acid and 1 mM 4-aminophthalic acid) and the analytes were detected at 254 nm.

[0089] Removal of Glc-1-P from the reaction equilibrium of EcAGPase by reaction with AtUSP significantly increases the yield for UDP-GlcNAc and regeneration of ATP. After 24 h, UDP-GlcNAc yields ranging from 61% (0.25 mM ATP) to 85% (2.5 mM ATP) are achieved (FIG. 18). ATP is regenerated 1.7 and 12.5 times in the reactions with initial concentrations of 2.5 mM and 0.25 mM, respectively, in the NeSuSy/AGPase enzyme cascade (FIG. 19). The UDP-Glc concentration after 24 h shows that this does not correspond to the theoretical maximum concentration of UDP-Glc (12.1 regeneration cycles of 0.25 mM ATP would correspond to 3 mM UDP-Glc, for example). It is expected that with each mole of regenerated ATP, one mole of Glc-1-P is produced by AGPase and subsequently converted to UDP-Glc by AtUSP (FIG. 20). The low amount of UDP-Glc is probably based on the hydrolysis activity of NeSuSy leading to the degradation of UDP-Glc, which is also evidenced by an increasing UDP concentration (FIG. 21). Reducing the ATP concentration by 50% (from 5 mM to 2.5 mM) leads to similar product yields (about 85%). 12 out of a maximum of 20 ATP regeneration cycles (at 0.25 mM ATP) are achieved (coupling efficiency 60%). A further increase should be possible by optimizing the enzyme ratios.

[0090] In combination with the UDP-sugar pyrophosphorylase AtUSP (PP.sub.i generating enzyme), a sugar kinase (ATP-consuming enzyme) and another pyrophosphorylase (PP.sub.i generating enzyme), the new ATP regeneration system NeSuSy/AGPase is capable of regenerating ATP from ADP with sucrose and PP.sub.i very efficiently and achieving high product yields. A key function is assigned to the nascent Glc-1-P in the reaction of EcAGPase. Glc-1-P should be removed from the reaction equilibrium. This can be achieved, for example, with the enzyme AtUSP, which converts Glc-1-P with UTP to UDP-Glc and PP.sub.i. AtUSP thus additionally forms PP.sub.i, which drives ATP synthesis and thus ATP regeneration.

[0091] Other enzymes such as sugar P mutases (phosphoglucomutase, formation of Glc-6-P) and sugar phosphate isomerases (Fru-6-P isomerase, formation of Fru-6-P) would also be suitable to remove Glc-1-P from the equilibrium of EcAGPAse.

VI. Further 5-Enzyme Cascade with ATP Regeneration System According to the Invention

[0092] The ATP synthesis reaction can proceed in the context of one of the phosphate-free UDP-GlcNAc synthesis with PGM to remove Glc-1-P according to the following equations:

GlcNAc+ATP .Math.GlcNAc-1-P+ADP

ADP+Suc .Math.ADP-Glc+Fru

GlcNAc-1-P+UTP .Math.UDP-GlcNAc+PP.sub.i

ADP-Glc+PP.sub.i.Math.Glc-1-P+ATP

Glc-1-P .Math.Glc-6-P

GlcNAc+Suc+UTP .Math.UDP-GlcNAc-1-P+Fru+Glc-6-P

[0093] The interaction of the individual enzymatic cascade elements is shown in FIG. 22.

[0094] FIGS. 23 to 25 show the analytical results of the above implementation.

[0095] Using AtUSP to reduce Glc-1-P in the reaction resulted in an additional unit of PP.sub.i, which could be used by EcAGPase to regenerate ATP. Therefore, this experiment tests how the system behaves when only one unit of PP.sub.i is provided during UDP-GlcNAc synthesis.

[0096] The experimental setup is as follows: The synthesis was performed in a one-pot procedure on a 96-well plate with a volume of 200 μL. Each reaction batch contained ATP at different concentrations (0.25 mM-2.5 mM), UTP (7 mM), GlcNAc (5 mM), sucrose (200 mM), F-1,6-P.sub.2 (0.5 mM), and MgCl.sub.2 (10 mM). The reactions were additionally carried out at 37° C. in 100 mM MOPS-buffer pH 7 for 24 h. BlNahK and SzGlmU enzymes were added at concentrations of 0.5 mg/mL and 5 μg/mL, respectively. The enzymes of the ATP regeneration cascade were used at the concentrations of 23.5 μg/mL (NeSuSy) and 175 μg/mL. The enzyme phosphoglucomutase (PGM) from hare muscle (Sigma Aldrich, USA) was added to the cascade at a concentration of 600 μg/mL. The reactions were stopped at the respective measurement points using a stop solution (28 mM SDS, 5 mM PAPA, 1 mM PABA) and analyzed by MP-CE. The nucleotides (ATP, ADP, AMP, UTP, UDP and UMP) and nucleotide sugars (UDP-GlcNAc and ADP-Glc) were detected on MP-CE using UV at 254 nm.

[0097] By replacing the AtUSP with PGM, UDP-GlcNAc yields of up to 73% could be achieved from 5 mM GlcNAc and 2.5 mM ATP after 24 h (FIG. 23). Further reduction of the ATP amount resulted in product yields between 60% (2 mM ATP) and at least 37% (0.5 mM ATP). The regeneration of ATP was most evident in the synthesis with 0.25 mM ATP. Here, 48% (2.4 mM) of the GlcNAc used was converted to UDP-GlcNAc, corresponding to a 9.5-fold ATP regeneration (FIG. 24). However, the reduction of the ATP concentration again resulted in an increase of the UDP concentration (FIG. 25).

[0098] The use of PGM for the reduction of Glc-1-P shows that coupling of the system according to the invention with further enzymes for Glc-1-P reduction is possible.

VII. Characterization of AGPase from E. coli

VII.1 Influence of PP.SUB.i .Concentration

[0099] The influence of PP.sub.i concentrations on the ATP synthesis activity of the EcAGPase used was investigated.

[0100] The experiment included 4.35 μg/mL EcAGPase, 2.5 mM ADP-Glc, 10 mM MgCl.sub.2, 1 mM Fru-1,6-P.sub.2 and PP.sub.i, at concentrations ranging from 0.75 mM to 10 mM, and a control reaction without PP.sub.i. ATP synthesis was performed in 600 μL of 100 mM HEPES buffer, pH 8 at 37° C. for 12 min. The synthesis was stopped at the respective measurement points with 28 mM SDS (final concentration 7 mM) and the analysis of the nucleotides (ADP-Glc, AMP, ADP, ATP) was performed by multiplex capillary electrophoresis (internal standards (final concentrations): 1 mM para-aminobenzoic acid and 1 mM 4-aminophthalic acid) and the analytes were detected at 254 nm.

[0101] EcAGPase already shows a reduction in specific activity at PP.sub.i concentrations above 3 mM (FIG. 26). At a PP.sub.i concentration of 7 mM, the specific activity of the enzyme is only about 1% (0.6 U/mg) of the activity compared to the non-inhibited reaction (50 U/mg).

[0102] EcAGPase is subject to substrate inhibition by PP.sub.i in the ATP synthesis direction. Therefore, ATP regeneration by this enzyme may be less efficient in reactions that involve a lot of PP.sub.i or release a lot of PP.sub.i very rapidly. This disadvantage can be compensated by increasing the EcAGPase concentration.

VII.2 Influence of Fructose (Fru)-1,6-P.SUB.2 .Concentration

[0103] Fru-1,6-P.sub.2 was used as an activator for EcAGPase. It is Fru-1,6-P.sub.2 a relatively expensive compound and its use in the EcAGPase reaction should be reduced as much as possible. Therefore, the activity of EcAGPase was investigated for different Fru-1,6-P.sub.2 concentrations to determine a minimum concentration of the activator.

[0104] The experiment included 3.9 μg/mL EcAGPase, 3 mM ADP-Glc, 3 mM PP.sub.i, 10 mM MgCl.sub.2 and Fru-1,6-P.sub.2 at concentrations ranging from 0.1 mM to 1 mM, as well as a control reaction without Fru-1,6-P.sub.2. The synthesis was performed in 600 μL of 100 mM HEPES buffer, pH 8 at 37° C. for 12 min. The synthesis was stopped at the respective measurement points with 28 mM SDS (final concentration 7 mM) and the analysis of the nucleotides (ADP-Glc, AMP, ADP, ATP) was performed by multiplex capillary electrophoresis (internal standards (final concentrations): 1 mM para-aminobenzoic acid and 1 mM 4-aminophthalic acid) and the analytes were detected at 254 nm.

[0105] With 0.5 mM Fru-1,6-P.sub.2, a high EcAGPase activity (60 U/mg, 12-fold higher than without activator) is still achieved (FIG. 27). At 0.25 mM, the activity decreases sharply (20 U/mg). EcAGPase still shows an activity of 5 U/mg even in the absence of the activator.

[0106] The activator Fru-1,6-P.sub.2 significantly increases the activity of EcAGPase. The concentration of the activator can be significantly reduced. Therefore, it is also possible to perform ATP regeneration with EcAGPase efficiently with very small Fru-1,6-P.sub.2 concentrations and even without activator.

VII.3 Influence of the Shift of the Reaction Equilibrium

[0107] The EcAGPase prefers the ADP-Glc synthesis. Thus, the accumulation of Glc-1-P negatively affects the reaction equilibrium for ATP synthesis. Therefore, the effect of Glc-1-P on the ATP synthesis activity of EcAGPase was investigated.

[0108] The experiment included 8.5 μg/mL EcAGPase, 3 mM ADP-Glc, 3 mM PP.sub.i, 0.5 mM Fru-1,6-P.sub.2, 10 mM MgCl.sub.2 and Glc-1-P at concentrations ranging from 1 mM to 10 mM, and a control reaction without Glc-1-P. The synthesis was performed in 600 μL of 100 mM HEPES buffer, pH 8 at 37° C. for 12 min. The synthesis was stopped at the respective measurement points with 28 mM SDS (final concentration 7 mM) and the analysis of the nucleotides (ADP-Glc, AMP, ADP, ATP) was performed by multiplex capillary electrophoresis (internal standards (final concentrations): 1 mM para-aminobenzoic acid and 1 mM 4-aminophthalic acid) and the analytes were detected at 254 nm.

[0109] EcAGPase shows a strongly reduced ATP synthesis activity at increasing Glc-1-P concentrations (FIG. 28). The activity of the enzyme already drops at 1 mM Glc-1-P to 71% of the activity without the addition of Glc-1-P. With 5 mM Glc-1-P, the residual activity is only 27%. The IC50 value determined for EcAGPase ATP synthesis activity is 1.78 mM Glc-1-P. The Glc-1-P concentration has a strong influence on the efficiency of ATP regeneration by EcAGPase. Therefore, it is recommended for the ATP regeneration system that Glc-1-P is actively removed from the process with increasing cascade duration.

[0110] The invention underlying this patent application was developed in a project funded by the BMBF under the grant number 031B0104B.