Salt or ion bridges in biological systems: a study employing quantum and molecular mechanics

Effect of Divalent Cations on the Interaction of Carboxylate Self-Assembled Monolayers

Interactions between organic molecules in aqueous environments, whether in the fluid phase or adsorbed on solids, are often affected by the cations present in the solution. We investigated, at nanometer scale, how surface carboxylate interactions are influenced by dissolved divalent cations: Mg²⁺, Ca²⁺, Sr²⁺, and Ba²⁺. Self-assembled monolayer (SAM) surfaces with exposed terminations of alkyl, -CH₃, carboxylate, -COO^- , or dicarboxylate, -DiCOO^-, were deposited on gold-coated tips and substrates. We used atomic force microscopy (AFM), in chemical force mapping (CFM) mode, to measure adhesion forces between various combinations of SAMs on the tip and substrate, in solutions of 0.5 M NaCl, that contained 0.012 M of one of the divalent cations. The type of cation, the number of carboxyl groups that interact, and their structure on the SAM influenced adhesion between the surfaces. The effect of the reference solution, which only contains Na⁺ cations, on adhesion force was mainly attributed to van der Waals and hydrophobic forces, explaining the lower force in systems that are more hydrophilic, i.e., -COO^--COO^-, and higher force for more hydrophobic systems. For charged surfaces, i.e., -COO^- and -DiCOO^-, in divalent cation solutions results were consistent with ion bridging. The inclusion of a hydrophobic surface, i.e., the -CH₃-COO^- or -CH₃-DiCOO^- system, decreased the possibility for strong cation bridging with the charged surface, resulting in lower adhesion. For systems including -COO^-, the adhesion force series followed the inverse cation hydrated radius trend (Na⁺ ≈ Mg²⁺ < Sr²⁺ < Ca²⁺ < Ba²⁺) whereas -DiCOO^- was responsible for lower adhesion force and modified trends, depending on the corresponding surface in the system. Differences in force magnitude between the monolayers were correlated with lower charge availability on the -DiCOO^- surface as a result of fewer active sites, probably because of the tendency of exposed malonate surface groups to interact between them, as well as high rigidity, resulting from the molecule structure. The characteristic response of the -DiCOO^- surface in solutions of Sr²⁺ and Ca²⁺ was correlated with possible malonate complexation modes. Comparison with previous studies suggested that the strong response of a -DiCOO^- surface to Sr²⁺ resulted from bidentate chelation, whereas Ca²⁺ response was attributed to alpha-mode association to malonate.

Strengths of hydrogen bonds involving phosphorylated amino acid side chains

Post-translational phosphorylation plays a key role in regulating protein function. Here, we provide a quantitative assessment of the relative strengths of hydrogen bonds involving phosphorylated amino acid side chains (pSer, pAsp) with several common donors (Arg, Lys, and backbone amide groups). We utilize multiple levels of theory, consisting of explicit solvent molecular dynamics, implicit solvent molecular mechanics, and quantum mechanics with a self-consistent reaction field treatment of solvent. Because the approximately 6 pKa of phosphate suggests that -1 and -2 charged species may coexist at physiological pH, hydrogen bonds involving both protonated and deprotonated phosphates for all donor-acceptor pairs are considered. Multiple bonding geometries for the charged-charged interactions are also considered. Arg is shown to be capable of substantially stronger salt bridges with phosphorylated side chains than Lys. A pSer hydrogen-bond acceptor tends to form more stable interactions than a pAsp acceptor. The effect of phosphate protonation state on the strengths of the hydrogen bonds is remarkably subtle, with a more pronounced effect on pAsp than on pSer.

Effect of citrulline for arginine replacement on the structure and turnover of phosphopeptide substrates of protein phosphatase-1

Phosphorylated and nonphosphorylated forms of a decapeptide corresponding to residues 9 to 18 of glycogen phosphorylase were compared using two-dimensional nuclear magnetic resonance with assignment of both peptides done by the sequential method. Both forms had little secondary structure, but there was evidence for an interaction between arginine-16 and phosphorylated serine at position 14. A change in the chemical shift for the epsilon-nitrogen hydrogen of arginine in position 16 was observed in the spectrum of the phosphorylated peptide and was not evident in a phosphopeptide having citrulline in place of arginine-16. Hydrolysis catalyzed by protein phosphatase-1 was decreased with the citrulline-containing phosphopeptide compared to the arginine-containing phosphopeptide with effects observed on both kcat and Km of the phosphatase reaction. Alkaline phosphatase hydrolyzed these peptides and a di-citrulline peptide equally well. These results are consistent with arginine being favorable in the recognition of substrates by phosphatase-1, possibly recognition as an arginine-phosphoserine complex. As a model study, arginine and two analogs, citrulline and canavanine, were examined for association with inorganic phosphate by nuclear magnetic resonance spectrometry. 31P-NMR measurements showed that arginine and canavanine caused a shift in the phosphate resonance at 20 degreesC. Citrulline caused no change. Changes in chemical shift were measured over the pH range 5-9 with arginine and canavanine both causing a slight decrease in the apparent pKa of inorganic phosphate (DeltapKa approximately 0.15). NaCl, NH4Cl, and guanidine hydrochloride showed little effect on the resonance signal position of inorganic phosphate at pH 6.5, consistent with selectivity for the guanidino group. Temperature (6 degrees, 20 degrees, and 37 degreesC) caused little change in the effect of arginine, but there was some dependency with canavanine, decreasing with temperature. Citrulline caused no change in the chemical shift of phosphate at any temperature. It was concluded that hydrogen bonded complexes were formed between the dianion of phosphate and the protonated form of arginine or canavanine with a bifurcated structure having preference for the omega-hydrogens.

Salt bridge interactions: stability of the ionic and neutral complexes in the gas phase, in solution, and in proteins

A theoretical study on the stability of the salt bridges in the gas phase, in solution, and in the interior of proteins is presented. The study is mainly focused on the interaction between acetate and methylguanidinium ions, which were used as model compounds for the salt bridge between Asp (Glu) and Arg. Two different solvents (water and chloroform) were used to analyze the effect of varying the dielectric constant of the surrounding media on the salt bridge interaction. Calculations in protein environments were performed by using a set of selected protein crystal structures. In all cases attention was paid to the difference in stability between the ion pair and neutral hydrogen-bonded forms. Comparison of the results determined in the gas phase and in solution allows us to stress the large influence of the environment on the binding process, as well as on the relative stability between the ionic and neutral complexes. The high anisotropy of proteins and the local microenvironment in the interior of proteins make a decisive contribution in modulating the energetics of the salt bridge. In general, the formation of salt bridges in proteins is not particularly favored, with the ion pair structure being preferred over the interaction between neutral species.

Ion pair formation of phosphorylated amino acids and lysine and arginine side chains: a theoretical study

Protein phosphorylation is one of the major signal transduction mechanisms for controlling and regulating intracellular processes. Phosphorylation of specific hydroxylated amino acid side chains (Ser, Thr, Tyr) by protein kinases can activate numerous enzymes; this effect can be reversed by the action of protein phosphatases. Here we report ab initio (HF/6-31G and Becke3LYP/6-31G) and semiempirical (PM3) molecular orbital calculations pertinent to the ion pair formation of the phosphorylated amino acids with the basic side chains of Lys and Arg. Methyl-, ethyl-, and phenylphosphate, as well as methylamine and methylguanidinium were used as model compounds for the phosphorylated and basic amino acids, respectively. Phosphorylated amino acids were calculated as mono- and divalent anions. Our results indicate that the PSer/PThr ion pair interaction energies are stronger than those with PTyr. Moreover, the interaction energies with the amino group of Lys are generally more favorable than with the guanidinium group of Arg. The Lys amino groups form stable bifurcated hydrogen bonded structures; while the Arg guanidinium group can form a bidentate hydrogen bonded structure. Reasonable values for the interaction free energies in aqueous solution were obtained for some complexes by the inclusion of a solvent reaction field in the computation (PM3-SM3).

Molecular modeling studies of novel retro-binding tripeptide active-site inhibitors of thrombin

A novel series of retro-binding tripeptide thrombin active-site inhibitors was recently developed (Iwanowicz, E. I. et al. J. Med. Chem. 1994, 37, 2111(1)). It was hypothesized that the binding mode for these inhibitors is similar to that of the first three N-terminal residues of hirudin. This binding hypothesis was subsequently verified when the crystal structure of a member of this series, BMS-183,507 (N-[N-[N-[4-(Aminoiminomethyl)amino[-1-oxobutyl]-L- phenylalanyl]-L-allo-threonyl]-L-phenylalanine, methyl ester), was determined (Taberno, L.J. Mol. Biol. 1995, 246, 14). The methodology for developing the binding models of these inhibitors, the structure-activity relationships (SAR) and modeling studies that led to the elucidation of the proposed binding mode is described. The crystal structure of BMS-183,507/human alpha-thrombin is compared with the crystal structure of hirudin/human alpha-thrombin (Rydel, T.J. et al. Science 1990, 249,227; Rydel, T.J. et al. J. Mol Biol. 1991, 221, 583; Grutter, M.G. et al. EMBO J. 1990, 9, 2361) and with the computational binding model of BMS-183,507.

The first solvation shell of magnesium and calcium ions in a model nucleic acid environment: an ab initio study

The interaction of organophosphate anions with divalent metal ions is central to many biological catalytic events. While experimental structural studies can give insight into the likely geometries that can be adopted, quantum mechanics allows for a more complete exploration of the competing forms. Ab initio quantum mechanical calculations have been performed on a series of complexes comprised of dimethyl phosphate, a divalent metal ion (either Mg(II) or Ca(II)) and water of hydration. An additional series of complexes were studied that included a Cl(I) ion to provide for charge neutrality. The most stable orientation of the hydrated metal ion complexed with the phosphate anion occurs when the metal ion is in a unidentate, rather than bidentate, orientation. The question of whether the divalent metal ion is located in the phosphinyl (-PO2(-)-) plane depends on the identity of the divalent metal ion and on the charge state of the complex.

The first solvation shell of magnesium ion in a model protein environment with formate, water, and X-NH3, H2S, imidazole, formaldehyde, and chloride as ligands: an Ab initio study

The first coordination shell of an Mg(II) ion in a model protein environment is studied. Complexes containing a model carboxylate, an Mg(II) ion, various ligands (NH3, H2S, imidazole, and formaldehyde) and water of hydration about the divalent metal ion were geometry optimized. We find that for complexes with the same coordination number, the unidentate carboxylate-Mg(II) ion is greater than 10 kcal mol-1 more stable than the bidentate orientation. Imidazole was found to be the most stable ligand, followed in order by NH3, formaldehyde, H2O, and H2S.

Ion pair formation involving methylated lysine side chains: a theoretical study

Lysine residues with one, two, or three methyl groups substituted on the epsilon-nitrogen atom are found in many proteins. To evaluate the effect of the posttranslational methylation on ion-pair formation we have performed semiempirical and ab initio molecular orbital calculations, using the AM1 method and the 6-31G* basis set, respectively. Combinations of various methylated forms of methylamine and ethylamine with formate, acetate, and dimethyl phosphate were studied as model compounds. This approach allowed us to obtain information relevant to the interaction of the modified Lys residues with carboxylate groups of proteins, and the backbone of nucleic acids. We have found that the interaction energy decreases with an increasing number of methyl groups. Inclusion of a solvent reaction field in the semiempirical calculations gave reasonable values for the interaction energy in aqueous solution, when formate and acetate were the counterions. These studies suggest that, in addition to other factors, a weakening of ionic interactions contributes to the various physiological effects of lysine methylation.

Modeling study on the cleavage step of the self-splicing reaction in group I introns

A three-dimensional model of the Tetrahymena thermophila group I intron is used to further explore the catalytic mechanism of the transphosphorylation reaction of the cleavage step. Based on the coordinates of the catalytic core model proposed by Michel and Westhof (Michel, F., Westhof, E. J. Mol. Biol. 216, 585-610 (1990)), we first converted their ligation step model into a model of the cleavage step by the substitution of several bases and the removal of helix P9. Next, an attempt to place a trigonal bipyramidal transition state model in the active site revealed that this modified model for the cleavage step could not accommodate the transition state due to insufficient space. A lowering of P1 helix relative to surrounding helices provided the additional space required. Simultaneously, it provided a better starting geometry to model the molecular contacts proposed by Pyle et al. (Pyle, A. M., Murphy, F. L., Cech, T. R. Nature 358, 123-128. (1992)), based on mutational studies involving the J8/7 segment. Two hydrated Mg2+ complexes were placed in the active site of the ribozyme model, using the crystal structure of the functionally similar Klenow fragment (Beese, L.S., Steitz, T.A. EMBO J. 10, 25-33 (1991)) as a guide. The presence of two metal ions in the active site of the intron differs from previous models, which incorporate one metal ion in the catalytic site to fulfill the postulated roles of Mg2+ in catalysis. The reaction profile is simulated based on a trigonal bipyramidal transition state, and the role of the hydrated Mg2+ complexes in catalysis is further explored using molecular orbital calculations.

Towards an understanding of the arginine-aspartate interaction

We have made a comparison of the geometries of intra- and intermolecular arginine-aspartate interactions by extracting orientation information from protein co-ordinate data. The results show a pronounced difference, with both types of interaction preferring to form twin N-H . . . O = C hydrogen bonds, but involving different nitrogen atoms. In intramolecular interactions, the aspartate favours a "side on" geometry, forming hydrogen bonds with N epsilon and N eta 2; in the intermolecular case, however, "end on" contacts involving N eta 1 and N eta 2 of the arginine are preferred. We have used Distributed Multipole Analysis of the methylguanidinium-acetate system to model the electrostatic component of the arginine-aspartate ion pair interaction in vacuo. We find, in agreement with the experimental arginine-aspartate distribution, that side on and end on doubly N-H . . . O = C hydrogen-bonded configurations are clearly the most favourable, with the side on being marginally lower in energy. Thus, despite the many competing side-chain interactions in proteins, many arginine-aspartate pairs adopt one of the minimum electrostatic energy conformations, or one close to a minimum. Within each of the two regions (side on and end on) we find only a small energy gap between the "symmetric" doubly hydrogen-bonded and slightly displaced "staggered" structures, again in agreement with the crystal structure data. Further calculations of the total ab initio interaction energy show that this follows the electrostatic term in its orientational variation, this phenomenon of "electrostatic domination" being well known in hydrogen-bonded systems. The end on arginine nitrogen atoms are observed to be more surface-exposed than N epsilon, as demonstrated by their greater accessibilities over a large sample of proteins. This helps explain the side on and end on preferences of intra- and intermolecular interactions, respectively. We also note the effect of short sequence intervals, particularly i in equilibrium with i + 2 relationships, in forcing many intramolecular contacts to be side on.