Probing Ca²⁺-Induced Conformational Changes in Porcine Calmodulin by H/D Exchange and ESI-MS:  Effect of Cations and Ionic Strength

Abstract

We applied a new method, "protein-ligand interaction using mass spectrometry, titration, and H/D exchange" (PLIMSTEX) [Zhu, M. M. (2003) J. Am. Chem. Soc. 125, 5252-5253], to determine the conformational changes, binding stoichiometry, and binding constants for Ca(2+) interactions with calmodulin (CaM) under varying conditions of electrolyte identity and ionic strength. The outcome shows that CaM becomes less solvent-accessible and more compact upon Ca(2+)-binding, as revealed by the PLIMSTEX curve. The formation of CaM-4Ca species is the biggest contributor to the shape of the titration curve, indicating that the formation of this species accounts for the largest conformational change in the stepwise Ca(2+) binding. The Ca(2+)-binding constants, when comparisons permit, agree with those in the literature within a factor of 3. The binding is influenced by ionic strength and the presence of other cations, although many of these cations do not cause conformational change in apo-CaM. Furthermore, Ca(2+)-saturated CaM exhibits larger protection and higher Ca(2+) affinity in media of low rather than high ionic strength. Both Ca(2+) and Mg(2+) bind to CaM with different affinities, causing different conformational changes. K(+), if it does bind, causes no detectable conformational change, and interactions of Ca(2+) with CaM in the presence of Li(+), Na(+), and K(+) occur with similar affinities and associated changes in solvent accessibility. These metal ion effects point to nonspecific rather than competitive binding of alkali-metal ions. The rates of deuterium uptake by the various CaM-xCa species follow a three-group (fast, intermediate, slow), pseudo-first-order kinetics model. Calcium binding causes the number of amide hydrogens to shift from the fast to the slow group. The results taken together not only provide new insight into CaM but also indicate that both PLIMSTEX and kinetic modeling of H/D exchange data may become general methods for probing protein conformations and quantifying protein-ligand interactions.

Protein-metal interactions of calmodulin and α-synuclein monitored by selective noncovalent adduct protein probing mass spectrometry

“…Hydrogen-deuterium exchange mass spectrometry (HDX MS) is another approach to examine conformational changes in proteins [39 -42]. For example, Gross et al found that the number of exchangeable hydrogens decreases as the concentration of Ca 2ϩ is increased up to ϳ0.25 mM, which they interpret as a tightening of the Ca 2ϩ binding domains of calmodulin [43,44]. Reactive labeling groups can also be used to monitor protein surface structure [45].…”

Application of SIMSTEX to oligomerization of insulin analogs and mutants

“…We also wish to resolve issues with solubility of insulin in the earlier kinetic results reported in [14], a report in which the kinetics of amide exchange was used without modeling to distinguish various forms of insulin. We now fit the data for r-human, lispro, porcine, and bovine insulins, whose sequences are shown in Table 1, to a kinetic model similar to the one used in [15]. Exchange reactions for proteins and peptides were successfully modeled as pseudo first-order reactions [16], and three exchange rate grouping models were previously applied to the kinetics of exchange of other proteins [15].…”

“…We now fit the data for r-human, lispro, porcine, and bovine insulins, whose sequences are shown in Table 1, to a kinetic model similar to the one used in [15]. Exchange reactions for proteins and peptides were successfully modeled as pseudo first-order reactions [16], and three exchange rate grouping models were previously applied to the kinetics of exchange of other proteins [15].…”

Modeling data from titration, amide H/D exchange, and mass spectrometry to obtain protein-ligand binding constants

“…Thus, an important question is: What is the sensitivity of K 3 and K 4 to the values of the values of K 1 and K 2 that are input into the 4-parameter model? We also posed this question and provided an answer in another article [30]. When the values of K 1 and K 2 were artificially increased or decreased by a factor of 8, both ␤ 3 and ␤ 4 obtained with the 4-parameter fit changed, but K 4 remained nearly constant (within a factor of 1.2), and K 3 only varied within a factor of 2.…”

“…The percentage of D 2 O was decreased from 99 to 90% in the H/D exchange media to minimize dilution of the protein and to conserve it. The new data were fit by using both the 4-parameter and the 7-parameter models; the results are in Table 3 (the curves were published in our recent article [30]). The relative standard deviation for each binding constant was 40 -70% less than those estimated when using the smaller number of data in the titration.…”