Search

Hydrogen/deuterium exchange (HDX) using matrix-assisted laser desorption ionization time-of-flight (MALDI-ToF) mass spectrometry is a sensitive, salt-tolerant and high-throughput method useful to probe protein conformation and molecular interactions. However, a drawback of the MALDI HDX technique is that sample preparation methods can typically result in higher levels of artificial deuterium in-exchange and/or hydrogen back-exchange just prior to or during mass analysis; this may impair data reproducibility and impede structural and kinetic data interpretation. While methods to minimize effects of back-exchange during protein analyte deposition on MALDI plates have been reported, this study presents a readily available, highly sensitive protein control set to facilitate rapid MALDI HDX protocol workup. The Ca2+-induced solvent accessible surface area (ASA) changes of calmodulin (CaM) and S100 proteins were employed to monitor and optimize HDX protocol efficiency. Under non-stringent room temperature conditions, the Ca2+-induced deuterium exchange of CaM, ΔDca2+, MH+ shifts −17 Da to −24 Da, while S100 ΔDca2+ MH+ shifts +8 Da to +12 Da. By comparing the divergent CaM and S100 Ca2+-induced deuterium mass shift differences, HDX sample workup and MALDI plate spotting conditions can easily be monitored.

S100a is a heterodimeric, acidic calcium-binding protein that interacts with calmodulin antagonists in a Ca2(+)-dependent manner. In order to study the behavior of the hydrophobic domain on S100a when bound to Ca2+, its interaction with trifluoperazine (TFP) was investigated using 16F nuclear magnetic resonance (NMR) and circular dichroism (CD) spectroscopy. The dissociation constant (Kd) values of TFP, as estimated from the chemical shifts of 19F NMR, were 191 and 29 microns in the absence and presence of Ca2+, respectively, and were similar to those previously reported for S100b. However, the TFP linewidth in the presence of Ca2(+)-bound S100a was 65 Hz greater than in the presence of Ca2(+)-bound S100b. This suggests a slower TFP exchange rate for S100a than for S100b. Thus, the TFP linewidths observed for each isoform may reflect differences in structural and modulatory properties of the Ca2(+)-dependent hydrophobic domains on S100a and S100b. Additionally, the presence of magnesium had no effect on the observed Ca2(+)-induced TFP spectral changes in S100a solutions. Circular dichroism studies indicate that Ca2+ induces a small transition from alpha-helix to random coil in S100a; in contrast, the opposite transition is reported for calmodulin (Hennessey et al., 1987). However, TFP did not significantly alter the secondary structure of Ca2(+)-bound S100a; this observation is similar to the effect of TFP on Ca2(+)-bound calmodulin and troponin C (Shimizu and Hatano, 1984; Gariépy and Hodges, 1983). It is, therefore, proposed that TFP binds to a hydrophobic domain on S100a in a fashion similar to other calcium-modulated proteins.

S100b is a calcium-binding protein that will bind to many calmodulin target molecules in a Ca2+-dependent manner. In order to study the Ca2+-dependent binding properties of S100b, its interaction with a calmodulin antagonist, trifluoperazine (TFP), was investigated using [19F]- and [1H]-NMR and UV-difference spectroscopy. It was estimated from [19F]-NMR that in the absence of Ca2+, the k1/2 value of TFP was 130 microM, while its k1/2 value decreased to 28 microM in the presence of Ca2+. The addition of KCl was not antagonistic to the Ca2+-dependent interaction of TFP to S100b. The chemical exchange rate of TFP with Ca2+-bound S100b was estimated to be 9 x 10(2) sec-1. By comparison with TFP-calmodulin exchange rates, it is suggested that the TFP-binding site on S100b is structurally different from its binding sites on calmodulin. Proton NMR resonance broadening in the range 6.8-7.2 ppm, corresponding to phenylalanine nuclei of S100b, indicates that these residues may be involved in TFP binding. Addition of Ca2+ to a 1:1 mixture of S100b and TFP resulted in a red-shifted UV-difference spectrum, while no significant difference spectrum was detected when Mg2+ was added to a S100b-TFP solution. Thus, we suggest that Ca2+ induces the exposure of a hydrophobic domain on S100b containing one or more phenylalanine residues that will bind TFP but that this domain is different from the hydrophobic domain on calmodulin.