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 This website is no longer updated and will be retired on 31 May 2019.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


Lai et al., (2015). Modulation of calmodulin lobes by different targets: an allosteric model with hemiconcerted conformational transitions.

October 2017, model of the month by Lu Li
Original models: BIOMD0000000574

Calmodulin is an ubiquitous signalling molecule exerting diverse biological functions. It contains four calcium-binding domains (EF-hand structures), organized in two lobes exhibiting significant autonomy [1]. Calcium binding to these motifs favour the conformational transition from close to open state for each calmodulin lobe [2]. Most target proteins preferably associate with one conformation of calmodulin over the other. In turn, they modulate the affinities between calmodulin and calcium [3].

To understand the general principles underlying calmodulin function and its regulation by binding targets, Lai et al. set up a mechanistic model based on the MWC (Monod-Wyman-Changeux) framework to simulate allosteric transitions of the two calmodulin lobes independently (Fig.1, BIOMD0000000574) [4].

Although conceptually simple, to implement the hemiconcerted allosteric transition of calmodulin requires explicitly setting up all combinational states of calmodulin lobes, their binding to calcium and downstream targets. Additional complexity arises from estimating the intrinsic affinities between calcium and calmodulin at different states. In the past, Stefan et al. have modelled the regulation of calmodulin while only considering its concerted conformation transitions [5] (BIOMD0000000183).

Lai et al. firstly developed a model representing the C-lobe of calmodulin as a concerted allosteric protein with two identical calcium binding sites. They estimated key allosteric parameters concerning the ratio of calmodulin in T over R state when there is no ligands, and ligand affinities of the R and T state, based on three independent calcium saturation curves of isolated C-lobe. The simulation results were then compared with additional experimental data (Fig.2).

Using this model, Lai et al. showed that the truncated C-lobe has similar calcium binding affinities as the intact calmodulin. Importantly, they highlighted that competing calmodulin targets, favouring binding to either R or T state, generate opposing effects on binding affinities between calmodulin and calcium (Fig.3).

Figure 1

Figure 1Allosteric model of calmodulin with hemiconcerted conformational transitions. A,B,C and D represent calcium binding sites organized on N-lobe (pink) and C-lobe(blue) of calmodulin; K depicts dissociation constant of calcium and each binding site; L represents ratio of calmodulin in T over R state when calcium is not present.

Figure 2

Figure 2 Calcium saturation curves on isolated C-lobe. Comparison between fitted (lines) and experimental (dots) calcium saturation of calmodulin C-lobe alone (black), in presence of the R-state binding peptide (WFF, cyan) and T-state binding peptide (NaV1.2IQp, red).

Figure 3

Figure 3 Regulatory effect on calcium saturation of isolated C-lobe by its competing targets. Simulation results show that the binding affinities between calcium and isolated C-lobe of calmodulin have been modified by R-state calmodulin-binding protein (RBP, blue), T-state binding protein (TBP, red) or both (purple).

Figure 4

Figure 4 Effect of skMLCK peptides on the saturation curve of calmodulin. Calcium saturation predicted by the model (lines), in comparison with experimental measurements (dots) of the whole calmodulin alone (green), in presence of WFF (full-length calmodulin binding domain of skMLCK, blue) and WF10 (truncated WFF, interacting with the C-lobe of calmodulin, cyan).

The authors set up a similar model for the N-lobe of calmodulin, assuming that the affinity between calcium and N-lobe in R state is the same as of C-lobe. They therefore, only estimated the calcium affinity to N-lobe in T state. Lai et al. combined this model with its C-lobe counterpart and predicted the effect of skMLCK peptides on calcium saturation of the whole calmodulin, which is in agreement with experimental observations (Fig.4).

In conclusion, the hemiconcerted calmodulin model, set up by Lai et al., successfully explained the cooperative calcium binding as emergent property resulted from state transitions. They also showed the effect on calcium binding by calmodulin binding targets and their preference to specific conformational state of calmodulin. In the future, this model can be extended to understand how calmodulin responses to different calcium spike patterns and selectively binds to different targets in order to exert opposing functional roles (BIOMD0000000628) [6].


  1. Faas GC, Raghavachari S, Lisman JE, & Mody I (2011) Calmodulin as a direct detector of Ca2+ signals.. . Nat Neurosci, 14, , 301-304.

  2. Zhang M, Tanaka T,& IkuraM (1995) Calcium-induced conformational transition revealed by the solution structure of apo calmodulin.. Nat Struct Mol Biol, 2, , 758-767.

  3. Nelson MR, & Chazin WJ (1998). Conformational changes in Ca2+ sensor proteins. Protein Sci, 7, , 270-282.

  4. Lai M, Brun D, Edelstein SJ,& Le Novère N (2015). Modulation of Calmodulin Lobes by Different Targets: An Allosteric Model with Hemiconcerted Conformational Transitions.. PLoS Comput Biol, 11(1), , e1004063.

  5. Stefan MI, Edelstein SJ,& Le Novère N (2008). An allosteric model of calmodulin explains differential acti- vation of PP2B and CaMKII.. PNAS, 105 (31), , 10768–10773.

  6. Li L, Stefan MI,& Le Novère N (2012). Calcium Input Frequency, Duration and Amplitude Differentially Modulate the Relative Activation of Calcineurin and CaMKII. . PLoS ONE, 7(3), , e43810.