ORIGINAL PAPER

Enhanced uridine 50-monophosphate production by whole cellof Saccharomyces cerevisiae through rational redistributionof metabolic flux

Dong Liu • Yong Chen • An Li • Jingjing Xie •

Jian Xiong • Jianxin Bai • Xiaochun Chen •

Huanqing Niu • Tao Zhou • Hanjie Ying

Received: 15 September 2011 / Accepted: 3 November 2011 / Published online: 15 November 2011

� Springer-Verlag 2011

Abstract A whole-cell biocatalytic process for uridine

50-monophosphate (UMP) production from orotic acid by

Saccharomyces cerevisiae was developed. To rationally

redistribute the metabolic flux between glycolysis and

pentose phosphate pathway, statistical methods were

employed first to find out the critical factors in the process.

NaH2PO4, MgCl2 and pH were found to be the important

factors affecting UMP production significantly. The levels

of these three factors required for the maximum production

of UMP were determined: NaH2PO4 22.1 g/L; MgCl22.55 g/L; pH 8.15. An enhancement of UMP production

from 6.12 to 8.13 g/L was achieved. A significant redis-

tribution of metabolic fluxes was observed and the under-

lying mechanism was discussed.

Keywords Uridine 50-monophosphate � Saccharomyces

cerevisiae � Metabolic flux analysis � Whole-cell

biocatalysis � Response surface method

Introduction

Uridine 50-monophosphate (UMP) is widely used as an

important pharmaceutical intermediate and food additive.

It is a membrane phosphatide precursor and can increase

brain cytidine diphosphate choline (CDP-choline) and

acetylcholine levels [1, 2]. UMP can also activate the

activity of mitochondrial ATP-dependent potassium chan-

nel (mitoKATP) of the heart, playing an essential role in

cardioprotection [3]. As a product of both the biosynthetic

and salvage pathways, UMP serves as the precursor for all

other pyrimidine nucleotides and contributes to pyrimidine

nucleotides-related oligosaccharides synthesis [4].

Generally, 50-nucleotides can be obtained by the reac-

tion of corresponding nucleosides with phosphorylating

agents [5]. However, due to the poor selectivity of the

reaction, in which 20-nucleotides and 30-nucleotides are

often produced, chemical synthesis is not suitable for UMP

production in large scales. UMP can also be produced by

fermentation and biocatalytic process. Corynebacterium

ammoniagenes have been exploited for fermentation, in

which UMP was accumulated in the culture broth con-

taining orotic acid (OA). The yield of UMP produced with

C. ammoniagenes by the fermentative process was 4.3 g/L

from 2 g/L orotic acid [6]. While in the biocatalytic pro-

cess, the strain was first cultivated in a mineral salt med-

ium, and then cells were harvested and used as the

biocatalyst in the reaction. The concentration of UMP

reached 10.4 g/L after optimization of cultivation and

reaction conditions by C. ammoniagenes ATCC 6872 [7].

Much attention has been paid to whole-cell biocatalyst

used for nucleotides and oligosaccharides production

[7–9]. Compared to chemical catalysts, enzymes have two

key advantages. One is their high selectivity; the other is

their great diversity in nature [10]. Whole-cell biocatalysis

combines these benefits with simple, low-cost catalyst

preparation and the possibility to develop efficient pro-

cesses for cofactor regeneration and multistep conversions.

Although the process using C. ammoniagenes as whole-cell

biocatalyst for UMP production was efficient, it was not

suitable for industrial application due to the high cell

concentrations required to provide enough enzymes [11].

D. Liu � Y. Chen � A. Li � J. Xie � J. Xiong � J. Bai � X. Chen �H. Niu � T. Zhou � H. Ying (&)

State Key Laboratory of Materials-Oriented Chemical

Engineering, College of Life Science and Pharmaceutical

Engineering, Nanjing University of Technology,

Nanjing 210009, People’s Republic of China

e-mail: yinghanjie@njut.edu.cn

123

Bioprocess Biosyst Eng (2012) 35:729–737

DOI 10.1007/s00449-011-0653-5

Since Saccharomyces cerevisiae is a by-product of estab-

lished fermentation process of making beer and can be

easily obtained in considerably substantial quantities at low

costs, it was used as the enzyme source to produce UMP

from orotic acid for the first time. The biological conver-

sion of orotic acid to UMP is catalyzed by two enzymes.

Orotate phosphoribosyltransferase (OPRTase, EC 2.4.2.10)

catalyzes the formation of orotidine 50-monophosphate

(OMP, EC 4.1.1.23) from orotic acid and 5-phosphor-

ibosyl-1-pyrophosphate (PRPP). OMP is subsequently

decarboxylated by OMP decarboxylase (ODCase) to form

UMP (Fig. 1). The availability of intracellular PRPP is

critical for UMP production. PRPP is synthesized from

ribose 5-phosphate (R5P) and ATP, which are derived from

the pentose phosphate pathway (PPP) and glycolytic

pathway, respectively. So, a rational flux distribution

between PPP and glycolytic pathway is of essential

importance for UMP production.

However, in our preliminary experiments, we had found

that the flux through the glycolytic pathway was larger than

expected. The ATP produced in glycolysis was much more

than the R5P produced in PPP. Due to the imbalance

between ATP and R5P, PRPP supply was not sufficient and

UMP production was low. Therefore, glycolysis should be

suppressed and more carbon source should be redirected

toward PPP for R5P synthesis. To rationally redistribute

the metabolic flux between glycolysis and PPP, statistical

methods were employed first to find out the critical factors

in the process. NaH2PO4, MgCl2 and pH were found to be

the important factors affecting UMP production signifi-

cantly. Then, we further investigated the levels of these

three factors required for the optimal distribution of met-

abolic flux with a response surface analysis. A significant

redistribution of metabolic fluxes was observed and the

mechanism was studied.

Materials and methods

Strain, media and cultivation

The strain used for UMP production was S. cerevisiae 1002

(China Center of Industrial Culture Collection).

The seed culture medium contained (g/L): yeast extract

10; peptone 20; glucose 20. The pH of the medium was

adjusted to 5.8 with 1.0 M NaOH and 0.1 M HCl. The

prepared seed culture was inoculated (5%, v/v) into a 5-L

fermenter (NBS Bioflo-110) containing 3 L of fermenta-

tion medium. The fermentation medium contained (g/L):

glucose 50; peptone 5; yeast extract 2; (NH4)2HPO4 2;

MgSO4�7H2O 1; KH2PO4 2. The pH of the medium was

adjusted to 7.4 with 1.0 M NaOH and 0.1 M HCl. Culti-

vation was carried out at 30 �C for 48 h.

The cultured yeast cells were collected by centrifugation

(8,0009g, 10 min at 4 �C) and washed twice with distilled

water. Then the harvested cells were lyophilized and stored

at -20 �C.

Biocatalytic reaction

The biocatalytic reaction mixture contained the permeabi-

lized cells, necessary substrates (orotic acid and glucose),

phosphate and other required ingredients. The basic bio-

catalytic reaction mixture (300 mL) contained 24 g glu-

cose, 6 g NaH2PO4, 0.75 g MgCl2, 3 g orotic acid, 6 mL

glycerol, 6 mL dodecyl dimethyl benzyl ammonium bro-

mide (I5) and 80 g cells, at pH 8.0 and temperature 30 �C.

However, precise compositions were described in the

related sections. The reactions were performed at 30 �C in

500-mL flasks containing 300 mL of the reaction mixture

on a reciprocal shaker at 100 rpm for 20 h.

Analytical method

The glucose concentration was measured using an SBA-

40C biosensor analyzer (Institute of Biology, Shandong

Province Academy of Sciences, P.R. China). Concentra-

tions of UMP, UDP, UTP and UDPG in the reaction wereFig. 1 Schematic illustration of the biosynthesis of UMP from orotic

acid

730 Bioprocess Biosyst Eng (2012) 35:729–737

123

measured by high-performance liquid chromatography

(HPLC). The HPLC system was equipped with a SepaxHP-

C18 column (250 mm 9 4.6 mm 9 5 lm) and a UV

detector operating at a wavelength of 260 nm. The column

was eluted with 6% (v/v) phosphoric acid (adjusted to pH

6.6 with triethylamine) at a flow rate of 1 mL min-1 and

room temperature.

Concentrations of organic acids (succinate, citrate,

acetate, malate) were determined using an Aminex HPX-

87H ion exclusion column (300 9 7.8 mm; Bio-Rad Lab-

oratories, Hercules, CA, USA), with 5.0 mM H2SO4

solution used as the mobile phase (0.6 mL/min) at 50 �C,

and with a UV detector at 210 nm. A refractive index

detector was used instead to measure the concentrations of

ethanol and glycerol. Quantitative data were obtained by

comparing the peak areas of the query compounds with

those of standards of known concentrations.

The concentrations of glucose 6-phosphate (G6P),

fructose 6-phosphate (F6P) and fructose 1, 6-bisphosphate

(FBP) were determined according to our previously

described method based on the specificity of enzymatic

conversion [12]. Inorganic phosphate concentrations were

measured by the spectrophotometric molybdenum blue

method [13].

Experimental design and data analysis

As a first step, the components were tested individually

(one-factor-at-a-time optimization). Fractional factorial

design (FFD) was then used to identify the factors affecting

UMP production significantly. The steepest ascent was

generated by the first-order equation obtained by FFD and

led to optimization by a central composition design (CCD),

which was used to evaluate the quadratic effects and two-

way interactions among these variables.

Design expert, version 7.1 (STATEASE Inc., Minne-

apolis, USA) was used for the experimental designs and

regression analysis of the experimental data. Statistical

analysis of the model was performed to evaluate the

analysis of variance (ANOVA). The statistical significance

of the polynomial model equation was judged statistically

by R2 and was determined by an F-test.

Metabolic flux analysis

The stoichiometric model assumed to describe the meta-

bolic network of S. cerevisiae was adapted from Kyoto

Encyclopedia of Genes and Genomes (KEGG) pathway

database. The production rates of glucose, ethanol,

glycerol, succinate, citrate, acetate, malate, glucose

6-phosphate (G6P), fructose 6-phosphate (F6P), fructose

1,6-bisphosphate (FBP), uridine diphosphoglucose

(UDPG), UMP, uridine diphosphate (UDP), uridine

triphosphate (UTP) and inorganic phosphate were mea-

sured to calculate or test the fluxes. The production rates of

the other intermetabolites were assumed to be zero. The

calculation of fluxes was performed in MATLAB (Math-

works Inc.). Flux for glucose uptake was set to 100 and the

other fluxes in the network were given as relative molar

flux normalized to the flux for glucose uptake.

Results and discussion

Optimization of reaction components

using one-factor-at-a-time method

Saccharomyces cerevisiae cells, orotic acid, glucose, phos-

phate and surfactants are essential ingredients in the bio-

catalytic reaction mixture. Orotic acid is the substrate of the

reaction. Glucose is the source of energy for ATP regener-

ation. Phosphate is the phosphate donor for PRPP biosyn-

thesis and is essential for nucleotides synthesis [14, 15]. The

cells used as the source of enzymes were permeabilized with

the surfactant I5 (dodecyl dimethyl benzyl ammonium bro-

mide). The surfactant was used to facilitate the substrates and

products to permeate across the cell membrane [16].

Apart from the basic reaction components mentioned

above, several effector molecules were also investigated,

including divalent magnesium Mg2?, NH4? and citrate. An

effector molecule is an allosteric inhibitor or activator

that can bind to an enzyme and exert its effect by changing

the conformation of the enzyme to decrease or enhance the

activity of the enzyme [17]. Mg2? is necessary for the

activities of most enzymes in UMP production, such as

hexokinase, phosphofructokinase and orotate phosphor-

ibosyltransferase [18, 19]. NH4? is a typical activator of

phosphofructokinase in S. cerevisiae. Citrate decreases the

activities of both phosphofructokinase and pyruvate kinase,

being used as a powerful tool to regulate the metabolic

fluxes distribution among glycolysis and PPP [20].

The one-factor-at-a-time method was adopted at first to

optimize the reaction components. The effects of these com-

ponents on UMP production are shown in Figs. 2 and 3. The

optimum concentrations of the components in the reaction

mixture were determined: orotic acid 10 g/L; NaH2PO4

20 g/L; glucose 80 g/L; surfactant I5, 20 mL/L; cells 267 g/L;

magnesium chloride 2.5 g/L; ammonium chloride 0.4 g/L;

citrate 1.2 g/L; glycerol 20 mL/L; pH 8.0; temperature 30 �C.

Screening key factors with the fractional factorial

design

After the one-factor-at-a-time optimization, fractional fac-

torial design (FFD) was used to pick out the key factors

that significantly influenced UMP production. The factors

Bioprocess Biosyst Eng (2012) 35:729–737 731

123

and the two levels for each factor were chosen based on the

one-factor-at-a-time optimization. Results of the design are

given in Table 1. Regression analysis resulted in the fol-

lowing equation, which describes UMP production (Y g/L)

as a function of the coded values of all factors:

Y ¼ 4:13� 0:11 X1 þ 0:13 X2 � 0:12 X3 � 0:092 X4

þ 0:031 X5 þ 0:17 X6 þ 0:61 X7 þ 0:74 X8 þ 0:49 X9

� 0:026 X10 þ 0:32 X11 ð1Þ

where, X1, X2,…, X11 are coded values of orotate, glu-

cose, citrate, glycerol, temperature, NH4Cl, NaH2PO4,

MgCl2, pH, surfactant I5 and cell concentration,

respectively.

Statistical analysis of the data showed that NaH2PO4,

MgCl2 and pH were the most significant factors for UMP

production. These three key effectors were selected for

further study.

a b

c d

e f

Fig. 2 Effects of a temperature, b pH, c orotic acid, d NaH2PO4, e glucose and f surfactant on UMP production

732 Bioprocess Biosyst Eng (2012) 35:729–737

123

a b

c d

Fig. 3 Effects of effector molecules: a MgCl2, b NH4Cl, c citrate and d glycerol on UMP production

Table 1 The results of the fractional factorial design

Run X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 UMP (g/L)

Observed Predicted

1 5 40 0.6 20 26 0.4 20 2.5 6 20 260 5.496 5.426

2 10 80 1.8 10 30 0.2 10 1.5 6 20 260 2.506 2.436

3 5 80 0.6 20 30 0.2 20 1.5 8 10 260 4.532 4.967

4 10 80 1.8 20 30 0.4 20 2.5 8 20 260 6.618 6.269

5 5 80 1.8 10 26 0.2 20 2.5 8 10 180 6.012 5.685

6 5 80 0.6 10 30 0.4 10 2.5 6 10 260 4.782 4.766

7 10 40 0.6 10 30 0.2 20 2.5 6 10 180 4.627 4.535

8 5 40 0.6 10 26 0.2 10 1.5 8 20 260 3.915 3.566

9 10 40 0.6 20 30 0.4 10 1.5 8 10 180 3.299 2.972

10 10 40 1.8 20 26 0.2 20 1.5 6 10 260 3.216 3.200

11 10 80 0.6 10 26 0.4 20 1.5 8 20 180 4.275 4.516

12 5 80 1.8 20 26 0.4 10 1.5 6 10 180 2.241 2.149

13 10 80 0.6 20 26 0.2 10 2.5 6 20 180 3.101 3.278

14 5 40 1.8 20 30 0.2 10 2.5 8 20 180 3.796 4.037

15 10 40 1.8 10 26 0.4 10 2.5 8 10 260 4.537 4.972

16 5 40 1.8 10 30 0.4 20 1.5 6 20 180 3.125 3.302

Bioprocess Biosyst Eng (2012) 35:729–737 733

123

Prediction of the optimal conditions using central

composite design

Based on the FFD experiments, NaH2PO4 (X7), MgCl2 (X8)

and pH (X9) were recognized as the most significant fac-

tors. Experiments of central composite design (CCD) were

carried out to predict the optimal values of these factors.

The design and experimental results are shown in Table 2.

A full second-order polynomial model was obtained from

regression analysis of the experimental data of CCD:

Y¼7:89þ0:27A�0:051Bþ0:20C�0:16AB

þ0:23AC�0:026BC�0:49A2�0:46B2�0:49C2 ð2Þ

where, Y is the predicted response, and A, B and C are

coded values of NaH2PO4, MgCl2 and pH, respectively.

The statistical significance of Eq. 2 was checked by

F test, and the ANOVA for response surface quadratic

model is summarized in Table 3. The 3D response surface

curves were then plotted to illustrate the interactions of the

factors and the optimal value of each factor required for

UMP production (Figs. 4, 5, 6). Each figure presents the

effect of two factors when the other one is held at zero

level. These 3D plots and their contour plots provide visual

interpretation of the interaction between the two factors

and facilitate location of the optimum experimental con-

ditions. The optimum values of NaH2PO4, MgCl2 and pH

predicted by the model were 22.10 g/L, 2.55 g/L and 8.15,

respectively. The predicted UMP production under this

condition was 7.97 g/L.

Redistribution of the metabolic flux

under the optimized condition

To confirm the model adequacy for predicting the maxi-

mum UMP production, three additional experiments under

the optimized condition were performed. The mean value

of UMP concentration was 8.03 g/L, which was in excel-

lent agreement with the predicted value 7.97 g/L. The

results of FFD showed that UMP production was mainly

affected by NaH2PO4, MgCl2 and pH. Indeed, after opti-

mization of these three factors, UMP production was

increased by 31%, from 6.12 to 8.03 g/L.

To evaluate the flux redistribution induced by optimiza-

tion of the key factors, the metabolic fluxes at 5, 10 and 15 h

before and after optimization were calculated. At the early

production stage (5 h), the flux through glycolytic pathway

was lowered by 24% after optimization, while the flux

through PPP was 2.5-fold higher (Fig. 7a). Although UMP

synthesis flux was increased, UDP and UTP synthesis fluxes

were decreased significantly, suggesting that ATP supply

was insufficient. Thus, a large amount of ribulose 5-phos-

phate (R5P) was converted back to the glycolytic pathway by

the non-oxidative branch of the PPP, probably to generate

ATP. At both middle and late stages, the fluxes in glycolysis

Table 2 Experimental results

of the central composite designRun Type A (NaH2PO4) B (MgCl2) C (pH) UMP (g/L)

1 Fact 18 (-1) 2.2 (-1) 8.5 (?1) 6.105

2 Axial 21 (0) 2.6 (0) 8.8409 (?1.68) 6.894

3 Center 21 (0) 2.6 (0) 8 (0) 7.980

4 Center 21 (0) 2.6 (0) 8 (0) 7.887

5 Fact 24 (?1) 2.2 (-1) 7.5 (-1) 6.511

6 Fact 24 (?1) 3 (?1) 8.5 (?1) 6.922

7 Fact 18 (-1) 2.2 (-1) 7.5 (-1) 5.982

8 Fact 18 (-1) 3 (?1) 8.5 (?1) 6.115

9 Center 21 (0) 2.6 (0) 8 (0) 7.892

10 Center 21 (0) 2.6 (0) 8 (0) 7.802

11 Axial 21 (0) 1.9273 (-1.68) 8 (0) 6.682

12 Fact 24 (?1) 3 (?1) 7.5 (-1) 5.970

13 Axial 21 (0) 2.6 (0) 7.1591 (-1.68) 6.252

14 Center 21 (0) 2.6 (0) 8 (0) 7.961

15 Axial 15.955 (-1.68) 2.6 (0) 8 (0) 6.107

16 Center 21 (0) 2.6 (0) 8 (0) 7.791

17 Axial 21 (0) 3.2727 (?1.68) 8 (0) 6.589

18 Fact 24 (?1) 2.2 (-1) 8.5 (?1) 7.306

19 Fact 18 (-1) 3 (?1) 7.5 (-1) 6.361

20 Axial 26.045 (?1.68) 2.6 (0) 8 (0) 7.017

734 Bioprocess Biosyst Eng (2012) 35:729–737

123

under the optimized condition were decreased by nearly

10%, while the fluxes through PPP were increased by 40%

(Fig. 7b, c). Consequently, more glucose was metabolized

via PPP and a higher UMP production was achieved.

In the whole-cell biocatalytic process, the cells used as

the source of enzymes were freeze-thawed and permeabi-

lized with the surfactant I5 to ensure an efficient perme-

ation. The intracellular pH homeostasis might be disrupted

and the biocatalytic performance was thus significantly

affected by pH. pH has a marked effect on the stability,

structure and function of many proteins due to their ability

to influence the state of ionization of the enzymes. In

S. cerevisiae, the optimal pH of yeast hexokinase shared by

both glycolysis and PPP is 7.5–8.5. The optimum pH of

other key enzymes in the glycolytic pathway and PPP are

different. The optimal pH of glucose-6-phosphate dehy-

drogenase is 7.4–9.2 [21, 22] and orotate phosphoribosyl-

transferase 7.8 [23], whereas the optimal pH of phos-

phofructokinase is about 7.0 [24] and pyruvate kinase

6.5–7.5. Hence, one can infer that PPP will be most active

under slightly alkaline conditions (pH 7.4–9.2), while

glycolysis is most active under neutral conditions

(pH 6.5–7.5). The optimum pH predicted by the model was

8.15. During the biocatalytic process, the pH decreased, but

not significantly, from 8.15 to about 7.4. At these pH

values, glycolysis might be slightly repressed, while PPP

exhibited greatest activity, which redirected the carbon flux

from glycolysis toward PPP and resulted in an enhanced

UMP production.

Phosphate is the phosphate donor for ATP regeneration

and PRPP biosynthesis. Its significant effect was also

confirmed by other studies [7, 8]. Also, phosphate can

affect the metabolism by altering enzyme activity. Shi-

mano et al. [15] found that during phosphate starvation,

there was an increase in the activities of various

Table 3 Regression results of the central composite design

Factor Coefficient F value P value

Model – 111.88 \0.0001a

A 0.27 94.74 \0.0001a

B -0.051 3.36 0.0967

C 0.20 51.24 \0.0001a

AB -0.16 20.66 0.0011a

AC 0.23 41.84 \0.0001a

BC -0.026 0.54 0.4802

A2 -0.49 330.5 \0.0001a

B2 -0.46 296.34 \0.0001a

C2 -0.49 325.27 \0.0001a

Lack of fit – 2.41 0.1780

Adj R2: 0.9813a Statistically significant at a confidence level of 95%

Fig. 4 Response surface curve for UMP production as a function of

NaH2PO4 and MgCl2 concentrations, when pH was maintained at 8.0

Fig. 5 Response surface curve for UMP production as a function of

NaH2PO4 concentration and pH, when MgCl2 concentration was

maintained at 2.6 g/L

Fig. 6 Response surface curve for UMP production as a function of

MgCl2 concentration and pH, when NaH2PO4 concentration was

maintained at 21.0 g/L

Bioprocess Biosyst Eng (2012) 35:729–737 735

123

nucleotidases and other hydrolyzing enzymes participating

in the hydrolysis of nucleotides. Phosphate behaves not

only as a classical allosteric activator of yeast phospho-

fructokinase, but also modifies significantly the qualitative

features of the regulatory properties. At pH 6.8, the activity

of yeast phosphofructokinase was stimulated by phosphate

up to 25 mM. However, in the presence of phosphate the

optimum pH of the enzyme would be displaced toward a

more acidic region. At pH 7.5, the phosphofructokinase

was inhibited by 10 mM phosphate instead [25–27]. In

contrast to phosphofructokinase, glucose 6-phosphate

dehydrogenase was not be inhibited by phosphate less than

50 mM. The Ki for commercial glucose 6-phosphate

dehydrogenase from yeast was about 100 mM in vitro, and

200 mM phosphate only partially reduced the activity (less

than 14%) of yeast glucose-6-phosphate dehydrogenase at

pH 7.5 [22, 28]. In the experiments, the initial phosphate

concentration was 142 mM, which then decreased gradu-

ally as phosphate was consumed. The final phosphate

concentration was about 24 mM. Considering the pH val-

ues and phosphate concentrations in the experiments, the

inhibition of glycolysis by phosphate might be more severe

than the inhibition of PPP, especially in early stage. This

was in excellent agreement with the enhanced flux through

PPP calculated at 5 h (Fig. 7a). Consequently, more carbon

source was used to produce R5P, which led to a higher

UMP yield from glucose.

Protein structure of some enzymes can be modified by

magnesium ion, and these structural alterations can cause

modifications in protein function. The Mg2?–ATP complex,

formed by Mg2? and ATP, was a biologically active form for

most enzymes involved in the formation and utilization of

ATP, such as hexokinase [18], phosphofructokinase [25]

and pyruvate kinase [29]. The divalent magnesium is

Fig. 7 Metabolic flux distributions at a 5 h, b 10 h, c 15 h. Numbers indicate flux ratios between before and after optimization

736 Bioprocess Biosyst Eng (2012) 35:729–737

123

indispensable for the formation of the OA/Mg2?–PRPP

ternary complex in S. cerevisiae. In the absence of Mg2?,

PRPP cannot bind to OPRTase [19, 23]. This may explain

why Mg2? effected UMP production significantly.

Conclusions

In the present study, S. cerevisiae was used as whole-cell

biocatalyst for UMP production from orotate. Among the

factors in the biocatalytic process, NaH2PO4, MgCl2 and

pH were found to be the most important ones affecting

UMP production significantly. Their optimum values were

finally determined: NaH2PO4 22.1 g/L; MgCl2 2.55 g/L;

pH 8.15. Under the optimized condition, the metabolic flux

was successfully redistributed and the UMP production

was enhanced from 6.12 to 8.13 g/L.

Acknowledgments This work was supported by a grant from the

National Outstanding Youth Foundation of China (Grant No.:

21025625), the Major Basic R&D Program of China

(2007CB707803), National Key Technology R&D Program

(2008BAI63B07) and Natural Science Foundation of Jiangsu Prov-

ince (BK2007527).

References

1. Wang L, Albrecht MA, Wurtman RJ (2007) Dietary supple-

mentary with uridine 50-monophosphate (UMP), a membrane

phosphatide precursor, increases acetylcholine level and release

in striatum of aged rat. Brain Res 1133:42–48

2. Cansev M, Watkins CJ, van der Beek EM, Wurtman RJ (2005)

Oral uridine-50-monophosphate (UMP) increases brain CDP-

choline levels in gerbils. Brain Res 1058:101–108

3. Krylova IB, Kachaeva EV, Rodionova OM, Negoda AE, Evd-

okimova NR, Balina MI, Sapronov NS, Mironova GD (2006) The

cardioprotective effect of uridine and uridine-50-monophosphate:

the role of the mitochondrial ATP-dependent potassium channel.

Exp Gerontol 41:697–703

4. Varki A (1993) Biological roles of oligosaccharides: all of the

theories are correct. Glycobiology 2:97–130

5. Haze A, Hatano H (1997) Method of producing 50-nucleotide.

United State Patent: 5623069

6. Nakayama K, Tanaka H (1971) Production of nucleic acid-related

substances by fermentative processes XXXVIII. Production of

uridine 50-monophosphate and orotidine 50-monophosphate by

Brevibacterium ammoniagenes. Agr Biol Chem 35:518–525

7. Wang X, Wang XW, Yin MX (2007) Production of uridine 50-monophosphate by Corynebacterium ammoniagenes ATCC 6872

using a statistically improved biocatalytic process. Appl Micro-

biol Biotechnol 76:321–328

8. Ying HJ, Chen XC, Cao HP, Xiong J, Hong Y, Bai JX, Li ZJ (2009)

Enhanced uridine diphosphate N-acetylglucosamine production

using whole-cell catalysis. Appl Microbiol Biotechnol 84:677–683

9. Tang JP, Yao YL, Ying HJ, Xiong J, Zhang L, Li ZJ, Bai JX,

Zhang YY, Ouyang PK (2009) Effect of NH4? and glycerol on

cytidine 50-diphosphocholine synthesis in Saccharomyces cere-visiae. Bioresour Technol 100:4848–4853

10. Robertson DE, Steer BA (2004) Recent progress in biocatalyst

discovery and optimization. Curr Opin Chem Biol 8:141–149

11. Chen Y, Li SY, Xiong J, Li ZJ, Bai JX, Zhang L, Ye Q, Ouyang

PK, Ying HJ (2010) The mechanisms of citrate on regulating the

distribution of carbon flux in the biosynthesis of uridine 50-monophosphate by Saccharomyces cerevisiae. Appl Microbiol

Biotechnol 86:75–81

12. Ying HJ, OuYang PK (2000) Inquiry into mechanism of FDP

accumulating using model of metabolic pathway flux. J Chem Ind

Eng (China) 3:313–319

13. Murphy J, Riley JP (1962) A modified single solution method for

the determination of phosphate in natural waters. Anal Chim Acta

27:31–36

14. Maruyama A, Fujio T (2001) ATP production from adenine by a

self-coupling enzymatic process: high-level accumulation under

ammonium-limited conditions. Biosci Biotechnol Biochem

65:644–650

15. Shimano F, Ashihara H (2006) Effect of long-term phosphate

starvation on the levels and metabolism of purine nucleotides in

suspension-cultured Catharanthus roseus cells. Phytochemistry

67:132–141

16. Fujio T, Maruyama A (1997) Enzymatic production of pyrimi-

dine nucleotides using Corynebacterium ammoniagenes cells and

recombinant Escherichia coli cells: enzymatic production of

CDP-choline from orotic acid and choline chloride (Part I).

Biosci Biotechnol Biochem 61:956–959

17. Arkin MR, Wells JA (2004) Small-molecule inhibitors of pro-

tein–protein interactions: progressing towards the dream. Nat Rev

Drug Discov 3:301–317

18. Romero CS, Olmo R, Teijon C, Blanco MD, Teijon JM, Romero

A (2005) Structural and functional implications of the hexoki-

nase–nickel interaction. J Inorg Biochem 99:2395–2402

19. Segura LG, Witte JF, McClard RW, Hurley TD (2007) Ternary

complex formation and induced asymmetry in orotate phos-

phoribosyltransferase. Biochemistry 46:14075–14086

20. Brul S, Coote P (1999) Preservative agents in foods-mode of

action and microbial resistance mechanisms. Int J Food Microbiol

50:1–17

21. Anderson WB, Nordlie RC (1968) Glucose dehydrogenase

activity of yeast glucose 6-phosphate dehydrogenase. I. Selective

stimulation by bicarbonate, phosphate, and sulfate. Biochemistry

7:1479–1485

22. Ashihara H, Komamine A (1976) Characterization and regulatory

properties of glucose-6-phosphate dehydrogenase from black

gram (Phaseolus mungo). Physiol Plant 36:52–59

23. McClard RW, Holets EA, MacKinnon AL, Witte J (2006) Half-

of-sites binding of orotidine 50-phosphate and R-D-5-phospho-

rylribose 1-diphosphate to orotate phosphoribosyltransferase

from Saccharomyces cerevisiae supports a novel variant of the

Theorell-Chance mechanism with alternating site catalysis. Bio-

chemistry 45:5330–5342

24. Hofmann E, Kopperschler G (1982) Phosphofructokinase from

yeast. Methods Enzymol 90:49–60

25. Baiwjelos M, Gancedo C, Gancedo JM (1977) Activation by

phosphate of yeast phosphofructokinase. J Biol Chem 252:

6394–6398

26. Laurent M, Seydoux F (1977) Influence of phosphate on the

allosteric behavior of yeast phosphofructokinase. Biochem Bio-

phys Res Commun 78:1289–1295

27. Yoshino M, Murakami K (1981) AMP deaminase reaction as a

control system of glycolysis in yeast. J Biol Chem 257:2822–2828

28. Van Noorden C, Tas J (1980) Quantitative aspects of the cyto-

chemical demonstration of glucose-6-phosphate dehydrogenase

with tetra nitro BT studied in a model system of polyacrylamide

films. Histochem J 12:669–685

29. Kuczek M (1999) A hypothetical model of the influence of

inorganic phosphate on the kinetics of pyruvate kinase. BioSys-

tems 54:71–76

Bioprocess Biosyst Eng (2012) 35:729–737 737

123