
Nyman et al., (2011). A Hierarchical Wholebody Modeling Approach Elucidates the Link between in Vitro Insulin Signaling and in Vivo Glucose Homeostasis.
November 2011, model of the month by Ishan Ajmera
Original model: BIOMD0000000356
Modeling approaches towards illustrating complex dynamics of glucose insulin system have drawn the attention of many theoreticians and experimentalists since last six decades. As a result, numerous characteristic mathematical models for diabetes simultaneously combined with highthroughput data have been developed with an aim to address this systemic disease. In spite of such tremendous efforts, the comprehensive overview of the glucose insulin regulation and its malfunction in diabetes has not yet completely emerged.
From the broader perspective, these models were mainly developed in two directions either considering whole body or specific subsystem involved in insulin glucose regulation. As both these model types are increasingly becoming more complete and realistic, detailed modules for crucial cellular processes exemplified by subsystem models can be potentially merged into wholebody organ based models. The result of such a merger is a multilevel model. Such a complex multilevel models where the same submodel, or module, may be described on different degrees of complexity, is often denoted as a hierarchical model. This model provides a link between different model systems and in vivo human conditions, thereby evaluating the relevance of in vitro data at physiological level.

Figure 1: Multi level modeling strategy. In Minimal modeling cycle, mechanistic hypotheses are tested against experimental dataset and conclusions are drawn in the form of core predictions and rejected hypotheses. The non rejected minimal models can be included as organ modules in multilevel models provided that the modules constraints are fulfilled. The minimal model can further be extended with more details as long as the submodules fit their relevant module constraints. The result is hierarchical multi level model with optional submodules of varying complexity[1]. Figure taken from [1].

Figure 2: Nyman Hierarchical model [1]. The top level glucose insulin whole body model (first panel) is from [5], with an adipose tissue module extracted from the original single insulin dependent tissue. In the next level of model (second panel), the adipose tissue module is expanded to depict insulin signalling enhancing glucose uptake via Glucose Transporter4 (GLUT4) translocation by including [4]. The insulin–insulin receptor binding is then further expanded (third panel) with insulin–insulin receptor binding model from [2] and insulininsulin receptor internalization/feedback model from [3]. Altogether, all the three panels above constitutes final hierarchical model. Figure taken from [1].

Nyman et al. [1, BIOMD0000000356] have developed a whole body hierarchical model which links intracellular insulininsulin receptor binding mechanism with the glucose transport mechanism in adipocytes and altogether with whole body glucose homeostasis. To be more precise in terms of models, Nyman et al have linked insulininsulin receptor dynamics models by Kiselyov et al [2] and Brannmark et al (Mifa model) [3, BIOMD0000000343] to insulin signalling based model by Sedaghat et al [4, BIOMD0000000137] and subsequently inserted this dynamic module in wholebody glucose homeostasis model by DallaMan et al [5, BIOMD0000000379]. Thus, this model elucidates the relation between in vitro insulin signalling in primary human adipocytes and the in vivo wholebody glucose homeostasis.
Plots in figure 3 represents simulation results from different levels of model described above. Plot A and B in figure 3 corresponds to the first panel of figure 2 representing the plasma glucose and insulin concentrations. Similarly, plot C, D and E, F of figure 3 corresponds to the second and the third panel of the figure 2 illustrating insulin signalling and insulininsulin receptor dynamics, respectively.

There are many subsystems involved in whole body glucose homeostasis, but currently there are no consensus regarding which of these subsystems are actually most important for overall regulation and play the most decisive role for its malfunctions in Type 2 Diabetes. In this condition, multi level modeling approach proposed by Nyman et al. (2011), potentially create bridges between different experimental model systems and the in vivo human situation, thereby offering a framework for systematic evaluation of the physiological relevance of in vitro obtained molecular/cellular experimental data [1].
Nyman et al. (2011), have represented an example of how pieces of knowledge can be merged together using a hierarchical modeling approach and also how such an approach efficiently can pinpoint important missing components in our understanding of a system. Thus, hierarchical multilevel modeling approach can be envisioned as an important step towards achieving more comprehensive view of glucose and energy homeostasis at cellular and wholebody level, which may eventual lead to better understanding and sound treatment of Type 2 Diabetes.

Figure 3: Simulations results of Nyman (M3) hierarchical model [1, BIOMD0000000356]. Simulation results of the M3 hierarchical model (which corresponds to Figure 9 of the [1]), using a parameterset of the adipose tissue module which can be used as core predictions to draw conclusions.

Bibliographic References

Nyman E, Brännmark C, Palmér R, Brugård J, Nyström FH, Strålfors P, Cedersund G. A hierarchical wholebody modeling approach elucidates the link between in Vitro insulin signaling and in Vivo glucose homeostasis. J Biol Chem. Jul 22;286(29):2602841, 2011. [CiteXplore]

Kiselyov VV, Versteyhe S, Gauguin L, De Meyts P. Harmonic oscillator model of the insulin and IGF1 receptors' allosteric binding and activation. Mol Syst Biol. 5:243. Epub 2009 Feb 17. 2009. [CiteXplore]

Brännmark C, Palmér R, Glad ST, Cedersund G, Strålfors P. Mass and information feedbacks through receptor endocytosis govern insulin signaling as revealed using a parameterfree modeling framework. J Biol Chem. Jun 25;285(26):201719. 2010. [CiteXplore]

Sedaghat AR, Sherman A, Quon MJ. A mathematical model of metabolic insulin signaling pathways. Am J Physiol Endocrinol Metab. Nov;283(5):E1084101. 2002. [CiteXplore]

Dalla Man C, Rizza RA, Cobelli C. Meal simulation model of the glucoseinsulin system. IEEE Trans Biomed Eng. Oct;54(10):17409. 2007.[CiteXplore]
