Engineering an Anti-arrhythmic Calmodulin

Calmodulin (CaM) is an integral part of Ca²⁺ signaling in all cell types, performing diverse functions through its interaction with target proteins such as ion channels, enzymes, and transcription factors. Hundreds of CaM targets have been identified to date, but many aspects of regulation by CaM remain a mystery. The goal of this work is to rationally design a CaM as a therapy for cardiac arrhythmia by elucidating the mechanism(s) by which CaM could be altered to control a specific branch of the cellular Ca²⁺/CaM signaling network. A major hurdle to engineering CaM is the ability to use that CaM to modulate a specific target rather than every target. We propose this problem can be overcome based on the strategies plants have evolved to differentially regulate enzymes and our knowledge of how to smartly formulate Ca²⁺ binding proteins.

Evolution has developed a family of CaM isoforms in plants whereas vertebrates have only a single isoform. We first investigated the differential regulation that has been reported for two CaMs from soybean, sCaM1 and sCaM4. The highly divergent sCaM4 isoform binds but does not activate specific targets in the plant that sCaM1 does activate; from this naturally-occurring CaM we learned that despite differences in Ca²⁺ binding affinity, both sCaMs respond to similar Ca²⁺ signals in the presence of Mg²⁺ and target peptide. Thus, when engineering CaMs, their behavior at multiple levels of complexity (isolated protein, presence of Ca²⁺/Mg²⁺ competition, presence of target, etc.) must be considered.

We next sought to determine the biochemical defect in recently-identified mutations in CaM (N54I, N98S) that have been associated with a type of inherited cardiac arrhythmia, CPVT. In CPVT, Ca²⁺ spontaneously "leaks" through the ryanodine receptor (RyR2), and under physical or emotional stress this additional cytosolic Ca²⁺ can trigger ventricular tachycardia in patients. CaM and other regulatory proteins serve to keep RyR2 quiescent during diastole (refractory). It has been proposed that the CPVT CaMs cause the disease by a disrupted interaction with RyR2. Compared to WT CaM, the N-domain of both CPVT CaMs displayed accelerated Ca²⁺ dissociation rates when bound to a peptide corresponding to the human RyR2 CaM binding domain. We therefore proposed that this accelerated Ca²⁺ dissociation is the mechanism for the dysregulation of RyR2 seen for the CPVT CaM mutants.

Earlier work in our lab dealt with modification of parvalbumin and troponin C (evolutionary relatives of CaM) to tune their Mg²⁺ or Ca²⁺ binding properties. As a result of these studies, we are now able to make point mutations in CaM that sensitize or desensitize the protein to Ca²⁺ over a broad range. We then tested our model for the mechanism of CaM-associated CPVT by designing CaMs with slower N-domain Ca²⁺ dissociation rates in order to restore RyR2 refractoriness. When we introduced one such CaM to a mouse model of CPVT using recombinant Adeno-Associated Virus, all treated mice were protected from arrhythmia, while 80% of the untreated mice experienced ventricular tachycardia. This study serves as a proof-of-concept that small modifications in CaM can potentially be therapeutic in vivo, opening up new possibilities for personalized medicine.