Please visit the new BioModels platform to access the latest content. This website is no longer updated and will be retired on 31 May 2019.
BioModels Database logo

BioModels Database


Nguyen LK and Kholodenko BN (2016). Akt/mTOR and RTK/MAPK signalling model may serve as a valuable tool for therapeutic research

March 2018, model of the month by Thawfeek M Varusai
Original model: BIOMD0000000651


Cellular signal transduction networks are capable of a range of dynamical behaviours that make them robust and adaptable. Feedback mechanisms and pathway crosstalk are fundamental features of such networks controlling the nature of the signalling response. Recent studies show the existence of hidden feedback and feedforward regulations in signalling networks that contribute to system dynamics [1,2,3,4,5].Taken together these explicit and hidden feedback mechanisms and pathway crosstalk can generate complex behaviours in the system.


Signalling networks regulate several molecular processes that make up the physiological state of the cell. Consequently, perturbations in the networks result in disease conditions. It is essential to understand the mechanisms of signal transduction and the resulting response in cells. This aids in gaining knowledge about the basic biology of the system and developing ways to interfere with the process in case of diseases. The presence of feedback regulations and crosstalk generate intricate signalling responses. If not accounted, this may cause adverse effects during therapeutic interventions such as drug resistance. Thus, there is a dire need to understand the effects of feedback systems and crosstalk in signalling networks. To address this, Nguyen L. K. and Kholodenko B. N. computationally study the Akt/mTOR and RTK/MAPK pathway crosstalk [6].

Figure 1

Figure 1. Computational model of the IR(IGFR)/mTORC1 and RTKs/ERK pathway crosstalk and model predictions. (a) Schematic diagram of the pathway crosstalk including the major negative feedback. (b) Reaction scheme of the model, based on which differential ordinary equations (ODEs) are formulated. (c) Time-course simulations of active S6K, PI3K and ERK levels for increasing Akt inhibition (time unit is in minutes). (d) Steady-state simulations of active S6K, PI3K and ERK levels in response to increasing dose of the Akt inhibitor (AktI total) for high/low FOXO expression. Figures adapted from [6].

Mathematical Modelling

The kinetic reaction scheme of the Akt/mTOR and RTK/MAPK signalling network used for the model is shown in Figure 1b. The authors deterministically model the Ring1B/Bmi1/H2A system using ordinary differential equations. Mass-action (MA) kinetics were used to represent protein-protein interactions and activation/deactivation/inhibition reactions. Hypothetical yet biologically feasible parameter values were used for simulations. Variable initial concentrations of inhibitors (iAkt and iMEK) were used to simulate different drug dosage. This is a computational study with no experiments involved. However, the model is able to explain experimental data from other studies.


The authors use the Akt/mTOR and RTK/MAPK signalling model to simulate real-life therapeutic scenarios – Akt inhibition and MEK inhibition. Firstly, when an Akt inhibitor is used, the model predicts that mTORC1 activity (active S6K) is reduced but PI3K and ERK activity increases. This behaviour can be explained by the two negative feedback regulations in the system: S6K-to-IRS and Akt-to-FOXO-to-RTKs (Figure 1c). Furthermore, the model is also able to predict the effect of variable FOXO levels on Akt inhibitor simulations. Higher FOXO levels result in stronger enhancement of PI3K and ERK activation by the Akt inhibitor (Figure 1d). Secondly, when an MEK inhibitor is used, the model predicts that ERK signalling is suppressed but PI3K/Akt and mTORC1 activity is increased (Figure 1e). The reason for this is the negative feedback regulation from ERK-to-RTK: MEK inhibitors weakens this regulation resulting in RTK upregulation and thus compensatory PI3K/Akt and S6K signalling.

Predictive Power of the Model

The Akt/mTOR and RTK/MAPK signalling model takes into account the key topological features in the system by including feedback regulations and pathway crosstalk. Simple mass action kinetics is used to describe the various events in the signalling process. These factors make the model very useful to predict the qualitative behaviour of the system. However, the values used for the parameters in the model are hypothetical and not cell/tissue-specific. Thus, the model may not be able to precisely predict the quantitative signalling responses in different cells.

Significance of the Model

Drug discovery and development is a long, expensive and laborious process with a low chance of a fruitful result. Any attempt to alleviate this process will be very useful to the community. Here, the authors develop a model to predict the response of Akt/mTOR and RTK/MAPK signalling under the influence of different inhibitors. The model is able to explain the robustness upon PI3K/Akt inhibititon in the system. Moreover, the model also has predictions about signalling behaviour when MEK is inhibited. This is valuable knowledge when formulating intervention strategies to cure relevant diseases. Computational models make this possible whereas obtaining such information experimentally is an endeavour.

Scientific Value Added

Multifactorial diseases such as cancer have proven to be resistant and adaptable to therapy. Complex cell signalling dynamics arising from intricate network topology is a main reason for this. In this paper [6], the authors develop a model to study the Akt/mTOR and RTK/MAPK signalling network. This model gives insights on the signalling response under Akt inhibition – FOXO expression levels may dictate the degree of resistance to Akt inhibitors. Furthermore, the model makes novel predictions on the impact of MEK inhibitors – higher MEK inhibitor doses lower active ERK levels but increase active PI3K/Akt and mTORC1 levels.


  1. Sauro HM, Kholodenko BN. (2004) Quantitative analysis of signaling networks.. Prog Biophys Mol Biol. 86(1):5-43.
  2. Del Vecchio D, Ninfa AJ, Sontag ED. (2008) Modular cell biology: retroactivity and insulation.. Mol Syst Biol. 4:161.
  3. Varusai TM, Kolch W, Kholodenko BN, Nguyen LK. (2015). Protein-protein interactions generate hidden feedback and feed-forward loops to trigger bistable switches, oscillations and biphasic dose-responses. . Mol Biosyst. 11(10):2750-62.
  4. Markevich NI, Hoek JB, Kholodenko BN. (2004). Signaling switches and bistability arising from multisite phosphorylation in protein kinase cascades.. J Cell Biol. 164(3):353-9.
  5. Bhalla US, Iyengar R (1999). Emergent properties of networks of biological signaling pathways. Science, 283(5400):381–387.
  6. 6. Nguyen LK, Kholodenko BN. (2016). Feedback regulation in cell signalling: Lessons for cancer therapeutics.. Semin Cell Dev Biol. 50:85-94.