Supplying extra metabolites

Question

• Are the interactions deletrious for E. coli?
• If so, is there any way to prevent this deletrious effect?

Observations

The three reactions with a non-null flux on the previous simulation were the PheA, TrpE and PabA/B reactions. All these reactions use the CHORISMATE as a precursor.

Setup

model: E. coli iJR904 wild type

In order to understand the type of perturbation happening we simulated the decrease of the fluxes through the TrpE, PheA and PabA/B reactions by progressively reducing the upper bound of the fluxes going through those reactions. The response of the system (the biomass function) is plotted against these changes in upper bound flux value.

Upper bound modification for the PheA reaction

In this part we modify the upper bound flux allowed to go through the PheA reaction.

Simulation

The results of the simulation are presented in Figure 1.

Figure 1:

CHORISTMATE MUTASE reaction flux and growth rate as a function of the upper bound of the CHORISMATE MUTASE flux.

• Setup: upperbound from 1 mmol/gDW/h to 0 with a step of 0.01
• The code for this model can be found at the bottom of the page

Interpretation

As can be seen on Figure 1, the relationship between the PheA flux and the growth rate is linear, hence this reaction is central to the metabolism of E. coli.

It appears that the PheA reaction takes CHORISMATE and produces PREPHENATE, a compound participating the production of tryptophan, an amino acid. Hence perturbating this flux may result in scarcity in the tryptophan pool empeding the cell from growing normally. Note that when the flux is not perturbated, the growth rate value is the maximal predicted one.

Upper bound modification for the TrpE and PabA/B

In this section, we modify the maximal value of the fluxes allowed to go through the TrpE and PabA/B reactions.

Simulation

The results of the simulation are presented in Figure 2.

Figure 2:

PabA/B reaction flux and growth rate as a function of the upper bound of the PabA/B flux.

• Setup: upperbound from 1 mmol/gDW/h to 0 with a step of 0.01
• The code for this model can be found at the bottom of the page

Interpretation

The dynamic of the curve in Figure 2 is the same as the one in Figure 1, except that the dependency relation between the growth rate and the PabA/B is higher (as seen by the slope of the growth rate curve in its ascending part).

Again, this may be explained by the fact that the PabA/B reaction produces L-GLUTAMATE whichi is a central metabolite of the bacteria.

Conclusion on the perturbations

We observed linear dependencies of the growth rate for the two perturbated reactions PheA and PabA/B which means that reducing the flux of any of these reactions may have very important consequences on the growth rates of the bacterias.

Hence, the CHORISMATE metabolite can be considered as a limiting factor of the enterobacting bionsythesis pathway.

Does adding chorismate prevent the negative effects on the growth rate ?

model: E. coli iJR904 with addition of ENTSYNTH, ENTOUT and CHORIN (intake of CHORISMATE) reactions.

This section is about finding the good quantity of CHORISMATE to give to the bacteria for having a maximal production of enterobactin.

To answer this question, we set up a simulation using the same objective function as the first simulation :

At the same time we simulate an addition of CHORISMATE into the medium by adding an input flux of CHORISMATE named CHORIN. The ENTOUT flux (production of ENTEROBACTIN) is then monitored as a function of both parameters in order to get the best quantity of CHORISMATE to add in all conditions.

Simulation

The results of the simulation are presented in Figure 3.

Figure 3:

ENTOUT flux (ENTEROBACTIN output flux) as a function of CHORIN (CHORISMATE input flux) and i coefficient of the modified objective function.

• Setup: upperbound from 1 mmol/gDW/h to 0 with a step of 0.01
• The code for this model can be found at the bottom of the page

Interpretation

As we can see on Figure 3, the relationship between the ENTOUT flux and the CHORIN flux is linear whatever the i choosen. A stable region is reached for the ENTOUT flux at a flux for CHORIN of about 17 mmol/gDW/h for all the i. Hence this result gives an idea of the quantity of CHORISMATE to add to the medium in order to maximize the production of ENTEROBACTIN.

Plus, the dynamic of the graph becomes more regular as more CHORISMATE is added, this means that adding a CHORISMATE input flux successfuly counter the effect of the consumption of CHORISMATE by the enterobacting production pathway.

Another interesting point to analyze is the presence of a plateau when saturating the medium with CHORISMATE precusor. It means that the presence of this reagent is no longer the limiting step. As we can see on Figure 1 on the interraction model, the pathway uses NAD+ for the production of the non ribosomal peptide. We can assume therefore that the origin of the limiting step is the necessity for the cell to regenerate its Red/Ox potential, which is coupled to other essential metabolic transformations, and therefore to growth. Unless we add a regenerative reaction, our production of enterobactin, may be theoretically limited to the flux we calculated when the chorismate is supplied.

Conclusion

This model gives two interesting results :

1. The dynamic of the ENTOUT flux as a function of CHORIN and i is very regular;
2. A flat region for the ENTOUT flux is reached for all values of i for about 17 mmol/gDW/h of CHORISMATE.
3. There is probably no way we can improve this theoric ration unless we change the red-ox potential of the cell or change host.