Urea cooler

1. Abstract

We made urea cycle in E.coli by introducing rocF gene encoding arginase in order to obtain urea to make urea cooler. To make urea cooler, we need large amount of urea. But just by introducing rocF, only a little amount of urea can be produced because arginine biosynthesis is repressed. Therefore, we tried to derepress the effect of repression. Furthermore, we researched flux to provide more urea. As a result, we found that the artificial urea production system is robust in a stoichometrically point of view. The analysis also found that supplementation of arginine, glutamic acid and aspartic acid would increase urea production rate.

Fig.4 Urea concentration in growth media 1 hour after IPTG induction	Fig.5 The reactions related with the urea cycle

2. Genetic Engineering for Urea Production

2.1 Introduction

Coolers can be made by adding urea to water, since dissolving urea in water is an endothermic reaction (-57.8 cal/g). However, E. coli does not synthetize urea naturally, so we attempted to complete the urea cycle inside E. coli and get urea.
Originally, E.coli has all enzymes of the urea cycle except for the arginase. In this work, introduction of the Bacillus subtilis rocF gene on a standardized plasmid completed urea cycle and enabled E.coli to produce urea as reported by TUCHMAN et al., (1997)
(Fig.1).

Fig.1 Addition of a gene which encoding arginase completes urea cycle in E.coli.

However, just by introducing arginase , E.coli, produces only a little amount of urea. TUCHMAN et al. proposed that catabolite repression in arginine biosynthesis pathway is the main reason for the low efficiency of production(TUCHMAN et al., 1997) The bacterial arginine biosynthetic genes are all regulated via a common repressor protein encoded by the argR gene and activated in the presence of arginine . (Fig.3)They circumvented the arginine repression by introduction of arginine operator sequences (Arg boxes), which bind the arginine repressor. Upon arginine repressor binding to Arg boxes, the amount of the arginine repressor which can repress arginine biosynthesis is reduced. In this work, we tried two ways of solving this problem. One way is introducing the Arg boxes as previous work. The other way is using an E. coli that has an argR deletion genotype so that the repressor is not synthetized.

Fig.2 Arginine biosynthesis is repressed by arginine repressor in the presence of arginine.

2.2 Results

Bacterial strains and plasmids The bacterial strains and plasmids used in this study are listed in Table 1 and Table 2, and the constructions are shown in Fig.3.

TABLE 1. E.coli strains used in this study

Strain	argR
MG1655	+
JD24293	-

JD24293 was obtained from National Institute of Genetics.

TABLE2. Expression plasmids used in this study

Designation	vector	rocF	Arg box
Ptrc-rocF	pSB3K3	+	-
Ptrc-rocF-Arg Box	pSB3K3	+	+

Fig.3 Plasmids used in this study
The details of the constructions are here.

MG1655 and JD24293 were transformed separately with pSB3K3, Ptrc-rocF or Ptrc-rocF-Arg box.

Urea concentrations detected in growth media of bacterial samples 1 hour after IPTG induction are shown in Fig.5.
A detailed method is described here

Fig.4 Urea concentration in growth media 1 hour after IPTG induction

In MG1655(ArgR+), addition of Trc promoter-rocF led to more production of urea compared to the bare backbone pSB3K3 as expected. These results show that insertion of rocF resulted in arginase production as expected, therefore completing the urea cycle in E.coli. In the same strain, however, addition of Arg box sequence led to little change in urea production. The reason why the effect of Arg boxes was not apparent is probably that pSB3K3 is a low-copy-number plasmid, in contrast to high-copy number used in the previous report. A low-copy-number plasmid is not capable of introducing enough number of Arg boxes to effectively deactivate the arginine repressor. Both of the plasmids containing rocF gene in the stain JD24293(Arg-) produce urea more efficiently than those in MG1655.

These results are in line with the fact that JD24293 carries argR (a gene which codes arginine repressor) loss-of-function mutant, which means deactivation of arginine repressor by Arg boxes is not needed and addition of the Arg box does not result in a significant increase of urea production.

3. Flux analysis for providing more urea

*The orange letters are the abbreviated names of the enzymes involved. The red letters are the enzyme expressed by introducing rocF gene. For complete names of the enzymes see Table 3.

3.1 Introduction

In metabolic engineering, mathematical modeling is the effective way to increase the products. Flux analysis, based on the hypothesis that the system is in steady state, is the effective way to find expect how to increase the products. In this work, we firstly focused on the concept of 'elementary flux modes' (Schuster et al, 2000), which provides metabolic routes both stoichiometrically and thermodynamically feasible.
One elementary mode shows that the carbon atom of urea derives from HCO₃^- which abounds in bacterial cytoplasm. Furthermore, in spite of L-glutamine consumption to transfer the side-chain ammonium group for production of carbamoyl phosphate which transfers the ammonium group to the urea cycle, ammonium ion can restore L-glutamine from L-glutamate which is a byproduct of the carbamoyl phosphate production. These findings suggest that there was little difference in the condition of containing NH₃ or L-glutamine in the culture to obtain more urea. This finding is proven in previous report's experiment. (Mendel et al, 1996) We also confirmed that the urea cycle in E.coli is well designed in a stoichiometrically point of view.
To provide more urea, there are two strategies. First one is to increase the amount of carbamoyl phosphate which is the reactant of the rate-limiting step of the urea cycle. The second one is to increase the amount of components of the urea cycle. We focused on the second one. We identified the elementary flux modes which produce these compounds from L-glutamine or compounds in TCA cycle. In all modes, ornithine was intermediate or final product to produce the components of the urea cycle. We also confirmed that E. coli have no feasible route for production of the four compounds other than those indicated in Fig.5. Ornithine production, which requires ATP, NADPH, Acetyl-CoA, and L-glutamine, is thus the necessary step in this strategy.
Considering that L-arginine, L-glutamate, and L-aspartate are consumed in protein biosynthesis, these compounds should be supplied from medium or produced by E. coli itself not only for increase but for maintenance of the cycle. Positions of L-arginine and L-aspartate in the reaction network show supplement of these compounds has similer effect on urea production.

What is elementary flux mode

"The concept of elementary flux mode provides a mathematical tool to define and comprehensively describe all metabolic routes that are both stoichiometrically and thermodynamically feasible for a group of enzymes". (Schuster et al, 2000) We can determine the modes which are able to work at steady state by this analysis.
As application of elementary flux modes, we can expect what substrates are needed to produce to the substances of interest. Furthermore, we can find expect which enzymes to overexpress or knockout so as to maximize the products we want.

3.2 Results

Overall reactions related with the urea cycle We considered the enzymatic reactions shown in Table 3 to determine the elementary flux modes. The scheme of the reaction is shown in Fig.5.

*The orange letters are the abbreviated names of the enzymes involved. The red letters are the enzyme expressed by introducing rocF gene. For complete names of the enzymes see Table 3.

One of the urea producing cycles without supplying the intermediates

At first, we attempted to get the elementary flux modes in the condition whose input is L-glutamine like previous reports. (Mendel et al, 1996) We determined the elementary flux modes by calculating matrix like. The initial tableau is shown below.

We calculated and got the final tableau.
The way to calculate is here..

We got eight modes shown in Fig.6. Each reaction formula is shown in Table 4. We focused on one of the urea producing modes in these eight modes as shown in Fig.7.

If we compared Fig.5 and Fig.7, we can see that in the mode displayed in Fig.7 the reaction which converts L-glutamate to L-ornithine is not needed for urea production.
As shown in Fig.7, the carbon atom of urea derives from HCO₃^- which abounds in bacterial cytoplasm.

Considering about nitrogen sources of urea, one of the nitrogen is derived from carbamoyl phosphate. Carbamoyl phosphate transfers the ammonium group to the urea cycle from L-glutamine. Therefore, it seems that L-glutamine is the effective nitrogen source. However, free ammonium ion can restore L-glutamine from L-glutamate which is a byproduct of the carbamoyl phosphate production. It means that L-glutamine and NH₃ have the same role to be the nitrogen source for urea. These finding suggest that there was little difference in the condition of containing NH₃ or L-glutamine in the culture to obtain more urea. This finding is proven in previous report's experiment. (Mendel et al, 1996) We also confirmed that the urea cycle in E. coli is well designed in a stoichiometrically point of view.

Increasing the four components of the urea cycle to provide more urea
As mentioned before, there are two ways to increase the urea production: increasing the amount of the carbamoyl phosphate and increasing the amount of the four components of the urea cycle. We focused on the second one. All elementary flux modes which produce these compounds from L-glutamine or compounds in TCA cycle produce L-ornithine as intermediate or final product. These modes are shown in Fig.8. Each reaction formula is shown in Table 5. One of the L-ornithine producing modes is shown in Fig.9.

As shown in Fig.9, we can provide L-ornithine by using a reaction which converts L-glutamate to L-ornithine. This is a key reaction to increase the reaction rates in the urea cycle. Therefore supplying ATP, NADPH, acetyl-CoA and L-glutamine or compounds in TCA cycle, is thus necessary step in this strategy. We also confirmed that E. coli have no feasible route for production of the four compounds other than those indicated in Fig 5.

The effect of protein biosynthesis to urea production
Considering that L-arginine, L-glutamate, and L-aspartate are consumed in protein biosynthesis, these compounds should be supplied from medium or produced by E. coli itself not only for increase but for maintenance of the cycle. However, L-glutamate is easily synthesized from L-glutamine. About L-arginine, it is synthesized through L-ornithine mentioned before. Therefore we focused on synthesizing L-aspartate we determined three modes which produce L-aspartate. (Fig.10) One of the modes is shown in Fig11. Each reaction formula is shown in Table 6.

Since L-aspartate and L-arginine are made from TCA cycle, we need pyruvate which is the substrates of the TCA cycle. Pyruvate is provided by glycolysis whose substrate is glucose. Therefore, adding glucose in the media and culturing in aerobic condition is effective way to maintain the urea cycle.

In conclusion, we determined that culturing the bacteria under aerobic conditions (to activate TCA cycle) are effective ways to increase the production of urea.

3.3 Future work

According to our results, we can say that three cycle or pathways are rerated to increase urea production: TCA cycle, glycolysis, the reaction which converts L-glutamate to L-ornithine. Therefore, the next step is the overexpression of the enzymes related with these reactions to increase urea production.

Reference
[1] Stefan Schuster, et al. A general definition of metabolic pathways useful for systematic organization and analysis of complex metabolic network, Nat Biotechnol(2000) 18:326-32
[2] Mendel Tuchman, et al. Enhanced production of arginine and urea by genetically engineered Escherichia coli K-12 strains, Apple Environ Microbiol(1997) 63: 38-8