Driving Calmodulin Protein towards Conformational Shift by Changing Ionization States of Select Residues

Abstract

Proteins are complex systems made up of many conformational sub-states which are mainly determined by the folded structure. External factors such as solvent type, temperature, pH and ionic strength play a very important role in the conformations sampled by proteins. Here we study the conformational multiplicity of calmodulin (CaM) which is a protein that plays an important role in calcium signaling pathways in the eukaryotic cells. CaM can bind to a variety of other proteins or small organic compounds, and mediates different physiological processes by activating various enzymes. Binding of calcium ions and proteins or small organic molecules to CaM induces large conformational changes that are distinct to each interacting partner. In particular, we discuss the effect of pH variation on the conformations of CaM. By using the pKa values of the charged residues as a basis to assign protonation states, the conformational changes induced in CaM by reducing the pH are studied by molecular dynamics simulations. Our current view suggests that at high pH, barrier crossing to the compact form is prevented by repulsive electrostatic interactions between the two lobes. At reduced pH, not only is barrier crossing facilitated by protonation of residues, but also conformations which are on average more compact are attained. The latter are in accordance with the fluorescence resonance energy transfer experiment results of other workers. The key events leading to the conformational change from the open to the compact conformation are (i) formation of a salt bridge between the N-lobe and the linker, stabilizing their relative motions, (ii) bending of the C-lobe towards the N-lobe, leading to a lowering of the interaction energy between the two-lobes, (iii) formation of a hydrophobic patch between the two lobes, further stabilizing the bent conformation by reducing the entropic cost of the compact form, (iv) sharing of a Ca +2 ion between the two lobes.

Effect of Calcium Ion Removal, Ionic Strength, and Temperature on the Conformation Change in Calmodulin Protein at Physiological pH

“…This implies that the low ionic strength paves the pathways for the two lobes to come close to each other which lead to a more compact form as compared to the extended native configuration. This is in accordance with the case observed in [ 15 , 16 ] where a change in the pH to a lower value also showed similar results. This change is observed because of the loss of electrostatic interactions at a lower pH or lower ionic strength of the solution which gives good opportunity to the N- and C-lobes of the protein to come close to each other, thus leading to a conformation change.…”

“…For this we monitor the distance distribution between the residues numbers 34 and 110 throughout the simulation. A normalized cumulative distance distribution is then calculated which shows a trend similar to what is observed in the lower pH case [ 15 , 16 ]. These normalized distance distributions are calculated at the end of the 100 ns simulations of the native structure.…”

“…The distance distributions are calculated between the residues 34 and 110, observed at the end of 100 ns as in [ 12 , 15 , 16 ] which gives a measure of the compactness of the protein structure. The respective normalized distance distributions are as seen in Figure 3 with different stages of the calcium ion removal from the protein shown in different colors.…”

“…To further understand the dynamics of the protein at a higher temperature in detail, we perform an energy calculation using the “NAMD energy” plugin in the VMD program as in [ 15 , 16 ]. This plugin calculates the various interaction energy terms between the different parts of the protein.…”

“…In our earlier work [ 15 , 16 ] we studied the effect of varying pH and observed a conformation change in CaM triggered by protonating nine acidic residues of the protein. Four different servers H++ [ 17 , 18 ], pKd [ 19 , 20 ], propKa 3.1 [ 21 ], and PHEMTO [ 22 , 23 ] were used to calculate the pKa values of all the titratable groups.…”