Calcium dependence of CaM target interaction was
Calcium dependence of CaM/target interaction was analyzed using a titration matrix in which both Ca2+ and CaM concentrations were changed. This gave us access to different parameters characterizing the interaction and namely to the number of bound Ca2+ required for the interaction to occur.
Our approach illustrates the possible discrimination between the CaM–Can complexes involved in the interaction of CaM with its target proteins or its small molecule interactors and provides a new tool to elaborate a classification of CaM-dependent target peptides as a function of CaM–Can complexes (n=0–4) recognition.
Results and discussion In order to gain deeper insight into the mechanism of action of calmodulin (CaM), we analyzed the interaction of SynCaM, the synthetic CaM which activates all mammalian CaM dependent enzymes and plant NAD kinase, and of three SynCaM mutants (Table 1) with two different entities. One entity is constituted by a chemical compound, probe CHPO 199-5-B05 ortho isomer (16B05 ortho) selected from a previous fluorescence screening assay  and the second by peptide analogs of the DAPK–CaM binding domain both in its unphosphorylated (DAPK-Reg) and phosphorylated (DAPK-P-Reg) forms and of the CaM binding domain of EGFR. Probe CHPO 199-5-B05 ortho isomer (16B05 ortho) will be called probe S1 throughout this study. The structures of the different entities are given in Fig. 1.
Conclusion Several Thiamet G of models have successively emerged since the 1970s. Three main models describing Ca2+ binding to CaM are now surviving: All three models explain most of the experimental results obtained up to now. However, the two first ones, due to the symmetrical behavior of the two sites of the C- or the N-terminal lobes, cannot take into account the results obtained with tryptophan isofunctional mutants where unique tryptophan residues were used to follow the occupancy of specific sites in CaM . The asymmetric SOB model allows analyzing the interaction between CaM and specific targets as a function of the number of Ca2+ bound in a more general way. Our experimental data are well fitted with the SOB model. Moreover, our model allows a decoupling between the four calcium binding steps and therefore to fine tune the interaction and the activation steps in the molecular mechanism of CaM/target recognition. The set of CaM targets studied, interestingly shows a specific recognition of each of the different SynCaM–Can complex (n=1 to 4), namely DAPK-Reg with SynCaM–Ca1 complex, the fluorescent probe S1 with SynCaM–Ca2 complex, DAPK-Reg phosphorylated with SynCaM–Ca3 and the EGFR CaM binding domain with SynCaM–Ca4 complex. We have performed the same set of experiments with human CaM and obtained the same results (data to be published). The CaM binding domain of DAPK shows that the peptide binding behavior to CaM–Can is significantly altered by the presence of the phospho-Ser. This surprising result illustrates the exquisite modulation in the interaction between calmodulin and its targets. The regulatory domain of DAPK in the unphosphorylated form shows significant binding to CaM in the absence of Ca2+, but efficient binding requires that at least one Ca2+ be bound to CaM. Efficient binding of the phosphorylated peptide requires that at least three Ca2+ be bound to CaM. Hence, our hypothesis of CaM flexible interaction site is strengthened by the phosphorylation-driven alteration of the interplay between CaM, Ca2+ ions and the peptide. Depending of the phosphorylation state of the CaM regulatory domain of DAPK, the enzymatic response will differ depending upon the shape of the calcium signal. Finally, we propose the characterization of the known CaM target peptides by adopting our approach that allows the discrimination between CaM–Can complexes involved in the interaction and subsequently the activation of CaM-dependent target proteins. Hence, this analytical strategy provides a new tool to elaborate a kind of typology allowing the classification of CaM-dependent target peptides as a function of the CaM–Can complexes (n=0–4) recognized by the peptide target sequence. We believe that this typology could contribute to improve our understanding of the mechanism of CaM/target recognition and of how CaM deciphers the calcium signal.