The signaling circuit

The signaling circuit 3 described in the Circuit Section has been modeled using Matlab Simbiology desktop. The following diagram shows the different parts of the model we have simulated:

For the level of detail considered, the main parts simulated are the following (the number correspond to the equations listed in the table in the next section):

1. Generation of L_aspartate induced by AAL
2. Diffusion of L_aspartate through the cell wall
3. Transcription of the aspA, promoted by FecI_a (active)
4. Translation of aspA
5. Activation of FecI, induced by the activation of FecR
6. Activation of FecR induced by FecA-PrhA
7. Plant cell wall lingand, FecA-PrhA binding

Reactions

The reaction equations for the previous parts, and the reactions rates associated, are summarized in the following table:

Simulations

The following figure shows the typical evolution of the output of the system (the generated chemoattractant medium.L_aspartate) againts the inputs (the wall cells ligand and the FecA-PrhA components on the outer membrane)

Analysis

Sensibility

Simbiology allows to compute the sensibility of the system against the different parameters.

The following figure shows the sensibility of all state variables (molecules of the different species considered) with respect to all the parameters.

What the analysis reveal is that the system is quite insensitive to changes in the parameters. This is due mainly to the nature of the transduction signals. The promoters act as a kind of "switch". This means that, provided these promoters reach certain levels, the other parts of the circuits are activated, even if the levels are not equal.

This analysis is referred to the steady-state of the system. Some parameters do not affect the final steady-state number of molecules, but on the other hand they affect the velocity of the system in the transitory. This can be seen in the following paragraphs.

Scanning of Parameters

If we perform a scan over several values of some parameters, it can be seen the influence of these parameters on the output. For instance, in the following figure it can be seen how the parameter kCellBinding (the "force" of the binding with the cell wall) affects the final output of the system (medium.L_aspartate, the amount of chemoattractant). The parameter has been scanned through several orders of magnitude.

It can be seen that this parameter affects the velocity of the system, but not in great deal. In this case, the parameter mainly influences the speed of the signaling circuit on the cell wall. But once this signal is activated, the rest of the system (in this case, the final production of aspartate) is not affected very much.

In the following, the same scan of parameters is done for the four main steps in the generation of the chemoattractant:

1. Transcription of the aspA, promoted by FecI_a (active)
2. Translation of aspA
3. Activation of FecI, induced by the activation of FecR
4. Activation of FecR induced by FecA-PrhA

Again, it can be seen that the evolution of the output is quite unsensitive to the changes of parameters. In order to illustrate this, a simple mathematical analysis is performed in the next paragraphs.

The system can be seen as a set of smaller subsystems (listed above) that are coupled by the enzymatic promoters. This subsystems are qualitatively quite similar. They are usually modeled as mass action driven kynetic equations.

If we analyze the FecR activation reaction, induced by the binding between the FecA-PrhA and the cell wall ligand:

The differential equation governing the reaction is:

were we have changed the names of some quantities for the sake of clarity:

Knowing that FecR+FecRa=FecR₀ (FecR₀ being the initial amount of FecR), and some simple calculations, then:

The steady-state solution of this equation is k₉FecR₀/(k₉+k₁₀), and it does not depend on bind.

The time constant of the system (its speed) is governed by (k₉+k₁₀) and bind.

In the actual system, bind evolves with time. If its evolution is much quicker than this system, reaching its steady-state, it can be considered constant and the same analysis can be applied. It acts as an activator of the system; moreover, the higher the quantity, the quicker the system.

If bind evolves with time the solution of the equation is bit more involved. A simplified analysis can be done assuming that bind is timewise constant for small amounts of time, and applying the same idea. The main effect of bind is to accelerate or decelerate the evolution.

The simulation of the FecR system, with no approximation, is very similar to the first order system described: