acs.jpcb.5b01261

Chem. B , 2015, 119 (19), pp 5962–5969. DOI: 10.1021/acs.jpcb.5b01261. Publication Date (Web): April 23, 2015. Copyright © 2015 American Chemical ...
0 downloads 0 Views 567KB Size
Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)

Article

Quantum Effects in a Simple Ring with Hydrogen Bonds Alisher M Kariev, and Michael E. Green J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.5b01261 • Publication Date (Web): 23 Apr 2015 Downloaded from http://pubs.acs.org on May 4, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry B is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Quantum Effects in a Simple Ring with Hydrogen Bonds Alisher M. Kariev and Michael E Green* Department of Chemistry City College of New York New York, NY 10031

To whom correspondence should be addressed: [email protected]

1 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract:

Complexes containing multiple arginines are common in proteins. The

arginines are typically salt bridged or hydrogen bonded, so that their charges do not repel. Here we present a quantum calculation of a ring in which the components of a salt bridge composed of a guanidinium, the arginine side chain, and a carboxylic acid, are separated by water molecules. When one water molecule is displaced from the ring, atomic charges of the other water molecule, as well as other properties, are significantly affected. The exchange and correlation energy differences between optimized and displaced rings are larger than thermal energy at room temperature, and larger than the sum of other energy differences. This suggests that calculations on proteins and other systems where such a ring may occur must take quantum effects into account; charges on certain atoms shift as substituents are added to the system: another water molecule, or an –OH or –CN bonded to either moiety. Also, charge shifts accompany proton shifts from the acid to guanidinium to ionize the salt bridge. The consequences of moving one water out of the ring give evidence for electron delocalization. Bond order and atomic charges are determined using NBO (Natural Bond Orbital) calculations. The geometry of the complex changes with ionization as well as the –OH and –CN additions, but not in a simple manner. These results help in understanding the role of groups of arginines in salt bridged clusters in proteins. Keywords: exchange-correlation, hydrogen bonds, charge transfer, bond order, electron delocalization ****************************

2 ACS Paragon Plus Environment

Page 2 of 26

Page 3 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

INTRODUCTION: Arginine is known to form complexes in many proteins, whether rings, strings, or pairs1-2. Such complexes, and their role in protein stabilization and recognition, were earlier studied in detail by Scheraga and coworkers3 Typically, these arginines are salt-bridged with carboxylic acids, and involved in hydrogen bonds to water. Salt bridges may have single minima for the proton between the acid (in proteins, aspartate or glutamate) and base (in proteins, arginine; lysine salt bridges may also exist, but we do not consider them here, as the arginine groups are more common and of more interest), in which the proton is shared between the two moieties; this has been discussed in some detail for an isolated salt bridge with varying amounts of water4. Although we are concerned here primarily with the delocalized electron, it is worth noting that there is evidence that the proton that transfers is itself partially delocalized. This concerns the state of ionization of the salt bridge, which in turn affects the electron delocalization that is our prime concern. Examples of proton delocalization include a protein-water complex in which an IR continuum provides evidence of the existence of a delocalized proton, in either of two possible modes: it may be either on water molecules, or between two glutamates5. A cyclic system (tris(amino(R)methylidene)cyclohexane-1,3,5triones, R=H, methyl, or phenyl), in which the extent of delocalization is determined by steric and induction effects produced by the substituent R, was studied theoretically by Martyniak et al6. Theoretical and experimental demonstration of proton delocalization, or at least displacement, in bacteriorhodopsin was reported by Phatak et al7. Bakker and Nienhuys reported a relatively low energy delocalization of H in water; the delocalization came in the 2nd excited state, but not at high energy8. Similar phenomena in other non-protein systems include hydrogen phthalate anion9, formamidineformamide complex10, a Mannich base N-oxide11, and an aliphatic tertiary diamine12. Several 3 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

other computational studies, and studies that included mass spectroscopic and infrared data as well, have considered salt bridges including arginine, arginine and lysine, and sometimes water13-17. These studies refer principally to the position of the proton, not delocalization in the sense of the proton wavefunction itself spreading. From these studies it is clear that the state of ionization in a salt bridge is very dependent on the environment, including dielectric constant, hydration, and other factors. The energy involved in ionization is thus small. Water molecules are often involved in the protein cases. The position of the proton (i.e., ionization of the salt bridge) matters in determining the energy, bonding, and other properties of the ring. In a normal salt bridge, the guanidinium and carboxyl would be directly in contact, but there may also be solvent separated ion pairs and non-ionized pairs. Here we present evidence for electron delocalization that would suggest resonance for the electrons in some cases, especially if the ring is planar; when the ring is destroyed by moving a water molecule 3 Å out of the ring, the increase in energy is 60 to 120 kJ, larger than the loss of three hydrogen bonds that would be the maximum classical increase in energy; one of these hydrogen bonds appears especially weak, so that the displacement effect is relatively stronger. This is a property that does not appear to have been previously considered. Electron delocalization has profound effects on the properties of the system, affecting ionization as well as energy and charges on the atoms. It is possible that in some proteins as well, distances between arginine and an acid (either aspartate or glutamate) would be near the approximately 6 Å needed to fit in a water molecule with hydrogen bonds. It would be expected to have important consequences in naturally occurring systems, although we do not consider this question directly here. There have also been quantum studies of some salt bridges, albeit not with water in the ring formed by the base and the acid, nor considering electron delocalization13, 18-23. 4 ACS Paragon Plus Environment

Page 4 of 26

Page 5 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

COMPUTATIONAL DETAILS: Each system was optimized using B3LYP/6-311G** in Gaussian0924. Atomic charges and bonding were then determined, on the optimized structure, and the configurations with one water displaced 1 Å, and 3 Å, from the ring position, with the other atoms not moved from the original optimized position; these were also calculated using Natural Bond Orbital (NBO) calculations, again with Gaussian09, still using B3LYP, but with a larger basis set, 6-311++G**. Natural Atomic Orbitals analysis was used. The use of NBO analysis was reviewed by Reed et al25-26. With 2 water molecules, and no addition to the rings, there are 26 atoms; three waters, 29 atoms; with –OH and –CN, 2 waters, 27 atoms (these groups replace one H atom). The single point calculations were repeated using NWChem, also using B3LYP/6-311++G**27. This allowed a check on whether the results would be the same as in Gaussian09. The difference in total energy for all two water cases (optimized, 1 Å down, and 3 Å down) was 0.0067 to 0.0070 Ha (NWChem lower), for all three water cases, 0.0075 Ha (again, NWChem lower) for the two water case (no substituent). Separate total energy and xc energy values are from the NWChem calculation. All differences are the same in NWChem and Gaussian09 to the accuracy shown in the Tables, so that all conclusions that depend on energy differences are independent of whether NWChem or Gaussian09 was used. Each optimization was an independent calculation, as was each single point calculation, so the consistency of the results suggests that whatever errors may remain in the calculation are too small to affect conclusions. Counterpoise corrections were all calculated using Gaussian09, using the counterpoise keyword in single point computations on the converged structures; as the corrections were small, this procedure is valid. Fragments were chosen as each water molecule and each of the larger moieties (with guanidinium, and with the acid, as the other two fragments for counterpoise corrections). It is not clear why the three water ionized case failed to converge.

5 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

However, given the consistency of all the other results, it is impossible to see how this could affect the conclusions. The Tables therefore have NWChem energies, and Gaussian BSSE corrections, but, as the differences between NWChem and Gaussian were entirely consistent, this affects no conclusions. This made it possible to do the calculations using the software most appropriate to the particular calculation; the counterpoise computations worked well in Gaussian, and the xc calculations were much more direct in NWChem. The consistency check allowed us to use the results together. CALCULATION RESULTS: In a normal salt bridge, the transfer of the proton depends on the presence of at least one neighboring water, and additional water forms rings that in turn produce cyclic variation (with respect to water molecules, complete rings forming the cycle) in the energy and other properties4. In the calculation presented here, an energy minimum is found that places water directly between the two components of the salt bridge, an arrangement that might occur if the acid and base were on separate α-helices in a well-structured protein; it would be difficult to see that such a ring exists in an X-ray structure, as the water oxygen is sometimes not well resolved, and the hydrogens not seen at all. Only the distance between the heavy atoms, and the orientation of side chains, is resolved, so the acid and base may be considered simply separated, without the bridge water being seen. A) Systems with water, guanidinium, and an acid group, not modified. i)

Two water molecules: The first calculation optimized the ring with two water molecules, with the movable proton on the carboxyl group in one calculation, and on the guanidinium in another. The water optimized in both cases by inserting between the acid and base groups, forming hydrogen bonds as shown in Fig.1. Table 1 gives the properties of the ring. To help find whether there is electron delocalization in the ring formed by the 6 ACS Paragon Plus Environment

Page 6 of 26

Page 7 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

acid, base, and water, we do a single point calculation with one water that forms part of the optimized ring structure pulled down out of the ring by 1 Å, and in a separate calculation, 3 Å, thereby destroying the ring, and with it presumably electron delocalization that might have existed as a circuit around the ring; there may be remaining delocalization along the one surviving branch of the ring. Only the one water molecule is moved, and the subsequent single point calculations leave all other atoms in the same position they hold in the optimized structure. Since even non-planar rings show a similar effect of withdrawing a water molecule, it appears that there must be electron delocalization even in non-planar rings. The non-planarity depends mainly on the ionization of the salt bridge. In the ionized cases, the dihedral angle, φ, defined by the last oxygen on the acid, the two water atoms in hydrogen bonds (H to the acid oxygen, O to the base proton that has transferred) and the proton that has transferred, is in the range 25o < φ < 32o. When not ionized, the definition is last N atom on the base, through the two atoms on water, to the proton on the acid, and the angle becomes 74o < φ