A Decapeptide Hydrated by Two Waters: Conformers Determined by

Subscriber access provided by READING UNIV

Article

A Decapeptide Hydrated by Two Waters: Conformers Determined by Theory and Validated by Cold Ion Spectroscopy Tapta Kanchan Roy, Natalia S. Nagornova, Oleg V. Boyarkin, and Robert Benny Gerber J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/acs.jpca.7b10357 • Publication Date (Web): 01 Nov 2017 Downloaded from http://pubs.acs.org on November 9, 2017

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 A 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 25

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

A Decapeptide Hydrated by Two Waters: Conformers Determined by Theory and Validated by Cold Ion Spectroscopy Tapta Kanchan Roy,1 Natalia S. Nagornova,2 Oleg V. Boyarkin2*, and R. Benny Gerber3* 1

Department of Chemistry & Chemical Sciences, Central University of Jammu, Jammu, 180011 India;

2

Laboratoire de Chimie Physique Molèculaire, École Polytechnique Fèdèrale de Lausanne, 1015 Lausanne, Switzerland; 3

Institute of Chemistry, The Hebrew University Jerusalem 91904, Israel; Department of

Chemistry, University of California Irvine, CA 92697, USA; Department of Chemistry, University of Helsinki P.O. Box, 55, 00014 Helsinki, Finland.

*

Corresponding authors: [email protected]; [email protected]

ACS Paragon Plus Environment

1


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

Page 2 of 25

ABSTRACT

The intrinsic structures of biomolecules in gas phase may not reflect their native solution geometries. Microsolvation of the molecules bridges the two environments, enabling a tracking of molecular structural changes upon hydration at the atomistic level. We employ density functional calculations to compute a large pool of structures and vibrational spectra for a gasphase complex, in which a doubly protonated decapeptide, gramicidin S, is solvated by two water molecules. While most vibrations of this large complex are treated in a harmonic approximation, the water molecules and the vibrations of the host ion coupled to them are locally described by a quantum mechanical vibrational self-consistent field theory with second order perturbation correction (VSCF-PT2). Guided and validated by the available cold ion spectroscopy data, the computational analysis identifies structures of the three experimentally observed conformers of the complex. They, mainly, differ by the hydration sites, of which the one at the Orn side chain is the most important for reshaping the peptide toward its native structure. The study demonstrates the ability of a quantum chemistry approach that intelligently combines the semi-empirical and ab initio computations to disentangle a complex interplay of intra and intermolecular hydrogen bonds in large molecular systems.


2

Page 3 of 25

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


INTRODUCTION Determining native three-dimensional (3D) structures of large biomolecules remains one of the most demanding and challenging objectives in life sciences. Although NMR and X-ray diffraction are the main methods for structural determination of biomolecules, they each have known limitations. New complementary approaches to structural determination are therefore highly desired. Over the last decade, high level computations, validated by spectroscopy of cryogenically cooled ions, have demonstrated increasing power in determining intrinsic 3D geometries of small to midsize biological ions, isolated in the gas phase.1-9 A way to link native and intrinsic structures is to study biomolecules in the gas phase, solvating them with a few water molecules.1, 10-18 This approach relies on the assumption that even a complex with a few water molecules may already exhibit the most important features of the structure in solution. On the experimental side this approach was successfully proven for a doubly protonated cyclic decapeptide, gramicidin S (GS; cyc-VOLFPVOLFP), using UV and IR cold ion spectroscopy.18 The study provided vibrationally resolved UV and IR spectroscopic fingerprints for the peptide gradually solvated by a known number of water molecules. In particular, three highly abundant conformers of the [GS + 2H]2+--(H2O)2 complex (GSW2) were identified from the spectra. It was suggested that these first two water molecules make the main structural rearrangement of the GS intrinsic structure toward its native-like geometry in condensed phase. However, the exact geometries of these doubly hydrated complexes and, in particular, the sites of hydrations, were not established due to the required computational complexity. Here we report on computations of the geometries of GSW2 conformers and their IR spectra. We use the available spectroscopic data and the computed spectra for identification of the experimentally observed conformers and,


3


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

Page 4 of 25

in particular, for revealing the most hydrophilic binding sites of [GS + 2H]2+ that accommodate the first two water molecules. A common approach for calculating IR spectra of large molecules is the harmonic oscillator (HO) approximation enhanced with empirical (single or multiple) scaling of the frequencies to match with experimental data.19 However, empirical scaling reduces confidence in validating geometries, specially for small molecules with fewer frequencies to analyze. Herein we employ a hybrid approach, where the vibrational frequencies of the (large) peptide in the complex are calculated using the quantum harmonic approximation, while the two attached water molecules and a few vibrational modes of the peptide coupled to them are treated from a firstprinciples based quantum anharmonic approximation, vibrational self-consistent field theory (VSCF)20,

21

with second order perturbation correction (VSCF-PT2).22 Such calculated

anharmonic vibrational frequencies of water can be directly compared with the experiment, which greatly raises confidence in assignment of hydration sites of the complex, as no scaling is used. THEORETICAL METHODS VSCF approximation. The VSCF approximation is based on the separable ansatz where the Nmode trial wave function is approximated as product of single mode wave function ( 𝜓! ! ) !

𝜓! ! (𝑄! ) , (1)

𝛹 𝑄!,…., 𝑄! = !!!

and the single mode VSCF equation thus can be obtained using the variational principle 1 𝜕! − ! + 𝑉! 2 𝜕𝑄!

!

𝑄!

𝜓! ! = 𝜀! ! 𝜓! ! 𝑄! . 3


4

Page 5 of 25

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


Here 𝑉!

!

𝑄! is the effective potential for mode Qi which is given by !

𝑉!

!

𝑄! =

!

𝜓!

!

𝑄!

𝜓! ! 𝑄!

𝑉 𝑄!,…., 𝑄!

!!!

. (4)

!!!

The single mode wave function and energies can be obtained by solving eq. 3 and 4 by selfconsistent method. Details of the VSCF theory can be found in a recent review.23 Pair-wise coupling approximations. The numerical evaluation of the multidimensional integral in the above equations is the main computational bottleneck. The difficulty of the integrals depends on the form of the potential. As a good approximation, VSCF potential can be written as a sum of single-mode terms (diagonal approximation) and pair-wise coupling terms. Following this approximation, the equation can be written in normal coordinate (Q) as, N

N

i =1

j

V (Q1 , Q2 ...QN ) = ∑Vi diag (Qi ) + ∑∑Vij2coup (Qi Q j ) i> j

(5)

This type of potential needs multi-dimensional grid point calculations. The computations of these integrals use grid representation for the 1-D (Vidiag) and 2-D (Vij2coup) potentials. It is already found that, points in the PES with 16 grid for 1-D and 16×16 for 2-D grid are sufficient in accuracy.16, 24 But for larger number of grid point per normal modes along with large number of vibrational modes, the total number of grid points on the PES for vibrational spectroscopy calculations is huge.. If NG is the number of grid points along a normal mode and NV is the number of vibrational modes, then the number of total grid points (NP) is

NP = [ NV × NG] +

[ NV ( NV − 1) × NG 2 ] 2

(6)


5


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

Page 6 of 25

That shows that, with the increase of NV and NG the number of required grid points increase rapidly. In this paper 10 grid points per normal modes are used. To further improve the energy levels given by VSCF, vibrational levels are computed using second order perturbation theory (PT2) where the PT2 correction on VSCF is described by the following equation:

EnPT 2 = EnVSCF + ∑

m≠ n

Π Nj=1ϕ (j n ) (Q j ) ΔV Π Nj=1ϕ (j m ) (Q j )

2

En( 0) − Em( 0)

(7)

where EnPT2 is the correlation corrected energy of the state n. first, the desired level of electronic structure calculation is used to find the equilibrium structure followed by second derivatives matrix of the potential from which normal modes were obtained which then further used for the point chosen for the grid. Each conformer of GSW2 consists of as large as 540 vibrational modes. It is not feasible to calculate a full pair-wise PES for this system. In a prototype 16 processors CPU system it will take approximately 500 years at B3LYP/6-31G(d,p)25-27 level of theory using 10 grid points per normal modes which is clearly not a realistic time scale. To avoid such circumstances for the computations of the anharmonic potential, a further approximation is adopted where the modes near to the water molecules are considered for pair-wise coupling potential which are interacting/coupling strongly with water vibrations. This protocol of truncated pair-wise potential implemented in this study was successfully applied earlier for bare GS.28 Accordingly, the computational time is reduced considerably to finish the computations of the construction of PES in a reasonable time scale. The coupling of water stretches and bends along with the adjacent modes of bare GS coupled near to water, such as N-H and C=O stretches and bends, were considered to calculate the anharmonic vibrational spectra of GSW2 structures.


6

Page 7 of 25

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


Next we turn to choose the quantum mechanical electronic structure method to compute the potential. For such flexible systems with several different types of interactions (very week or strong), use of highly accurate method is desirable. However, as the size of the system is very large for any level of quantum mechanical methods, only computationally fast algorithms such as HF and some suitable functionals from DFT methods are feasible to use with moderate basis set. Even a suitable DFT method, as a better choice over HF, requires very long computational time with higher basis sets. Hence, to make a balance between the level of theory (accuracy) and computational time one must be rational to choose the same so that first-principles based quantum anharmonic calculations can be performed in a reasonable time scale with present stateof-the-art computers. We used B3LYP/6-31G(d,p)25-27 level of theory for all the anharmonic calculations.

Conformational analysis. A routine practice to perform conformational analysis for structure prediction is generally based on the empirical parameterization or force-field model of PES. Several authors successfully carried out the conformational analysis using different protocols for biological molecules.29-36 In this work quantum mechanical static approach, guided by the experimental data is applied, to find several low-energy candidate structures. In case of micro-solvation of biomolecules, first, the knowledge of the gas phase global minima structure of biomolecules is crucial. If one knows the global minima precisely, then, with the help of experimental data, the quantitative prediction of the energy landscape of a given micro-solvated biomolecules is feasible, at least for a few solvent molecules. Such calculations are further less complicated if the number of the solvent, and the probable interacting sites of the solute-solvent system are countable. We began with a previously established C2 symmetric global minima of bare GS28, 37-39 and exploited the experimental data to reduce significantly the number of possible


7


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

Page 8 of 25

conformers. While the linear peptides posit very rich conformational flexibility, the cyclic peptides like GS are conformationally more rigid and hence the PESs underlying the cyclic peptides are rather less stringent. However, we must expect, if not huge, but a large number of different conformers for GSW2. To this end, with the help of experimental data, prediction of the number of possible conformers is reduced to a great extent. First, experimental IR spectra showed that all the conformers of the solvated structures of GSW2 retain higher symmetry than C1 (C2 in this case similar as bare GS) by giving single peak for a similar pair of stretches. This shows, two water molecules in each conformer must interact exactly opposite in position with the cyclic bare GS to retain the symmetry (fig. 1). Second, the first two waters, apparently, can form non-covalent bonds only with H of NH (total 4 pairs), NH3+(1 pair), CO groups (total 5 pairs) and π−electrons of Phe rings (1 pair) of the protonated GS or with each other keeping the symmetry constraint unchanged. Out of them, 4 antiparallel beta-sheet interactions due to 2 pairs of N-H of Val and Leu and C=O of Leu and Val bonds, are always present in all the conformers, and the interactions between two water molecules can be opted out due to symmetry constraints which further reduce the possible combinations of the sites for the interaction with water (fig. 1). Third, the hydration must not disrupt the four core hydrogen bonds (Fig. 1) that are known to be present in both the native and the intrinsic structures of GS18, 38, 39 and wrapped by hydrophobic groups. Water molecules must interact on the surface of the peptide and consequently only 2 pairs of N-H, 3 pairs of C=O and 1 pair each of π-electron of Phe and NH3+ were available to interact with water which already formed H-bonded interactions in bare GS. Thus, water molecules first need to break some of those H-bonds to interact, followed by formation of waterpeptide H-bonding. Additionally, there is a high possibility that the existing strong interaction between C=O of Phe and one N-H of the NH3+ group will also survive in all the solvated


8

Page 9 of 25

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


conformers (these are lying much inside of the cyclic peptide) leaving effectively only 2 pairs of N-H (N-H of Phe and N-H of Orn), 2 pairs of C=O (C=O of Pro and C=O of Orn) and 1 pair each of π-electron of Phe and NH3+ (NH3+ group of Orn side chain) are available for the interactions with water. All these constraints largely reduced the possible combinations of the interaction of bare GS with water to a great extent and greatly reduced the initial pool of structures. In Fig. 1 some of the possible sites of such GS-water interactions are given by black arrows. This opens up the possibility to use more accurate explicit quantum mechanical description for static conformer search. We studied each possible conformer systematically with a first-principles based DFT method in conjunction with different basis sets. In the first step, one water molecule was added in different positions of the C2-symmetric doubly protonated bare GS structure and freely optimized at B3LYP/6-31G(d) level of theory. Next, another water is placed in the symmetrically opposite position of this previously optimized structure (GSW1) and optimized it again at the same level for a potential conformer of GSW2. Then, the optimized structure of GSW2 at lower level further optimized at higher B3LYP/631G(d,p) level of theory where the C2-symmetry was retained. Following this protocol approximately 125 different structures of GSW2 with different positions of water molecules were investigated. In many cases slightly different initial pool of structures converged to some particular conformers which further reduced the conformational flexibility for such large systems.

For each case, tight optimization with ultrafine numerical integration grid was

performed and subsequent frequency calculations were done to check the structure is a true minima. As this system is very large for any level of quantum mechanical methods and extremely flexible with too many soft vibrations, the harmonic frequency calculations are essential to pre-select suitable candidate structures, which comply with the experimental


9


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

Page 10 of 25

constraints. From the initial pool of ~125 conformers, 14 lowest energy conformers were selected given at B3LYP/6-31G(d,p) level. Next, comparing the computed equilibrium structures, harmonic vibrational spectra and experimental data we have chosen 7 potential conformers which are also lowest in energy hierarchy. Finally, three structures, named A, B and C (Figure 1) were assigned to the experimentally observed conformers, earlier named as 2a, 2b and 2c, respectively. The assignment was based on (i) correlation of the experimentally observed red shifts of UV band origins and the theoretically determined hydration sites; (ii) comparison of the calculated and measured vibrational spectra; (iii) comparison of the relative potential energies of calculated geometries A, B, C with the measured relative intensities of UV band origins for conformers 2a, 2b, 2c. For further validation of the structures, we have tested the three conformers using dispersion corrected M06-2X/6-31G(d,p) and B3LYP-D3/6-31+G(d,p) level of theory and found the similar structures with same energy hierarchy. The detail analysis for the exclusion of other 4 potential structures out of 7 chosen one is described in the supporting information (S9). An artifact is observed when dispersion corrected B3LYP functional is used (B3LYP-D3) with 6-31G(d,p) basis which yields a small imaginary frequency and that cannot be removed using tighter gradient convergence and ultrafine numerical integration grid. It is also found that, these imaginary frequencies can be avoided by using higher basis like 6-31+G(d,p), however, which takes much longer computational time. Consequently, we had to restrict the calculations without dispersion correction and at B3LYP/6-31G(d,p). All the conformational analysis were carried out using Gaussian09 program package.40


10

Page 11 of 25

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


Fig. 1. The computed (B3LYP/6-31G(d,p) level) lowest-energy geometry of doubly protonated gramicidin S. Color-coded boxes indicate different intramolecular hydrogen bonds. Black arrows point to the potential sites of hydrations.

RESULTS AND DISCUSSION The experiment suggests near equal population of the conformers 2a and 2b, but lower population of the conformer 2c.18 The relative energies, calculated at B3LYP/6-31G(d,p) level show that, structure B has the lowest energy among all tested geometries (Table S1). Structures A and C are higher in energy than structure B in energy by ~4.5 and ~10 kcal/mol, respectively. Similar observations were found for use of the dispersion-corrected functionals with higher basis set (table S1). The calculated energetics of the structures (EA, EB, EC), as well as the known energies of [GS+2H]2+(EGS) and H2O (EH2O) allow for evaluating the binding energies (BE) of two water molecules in the complex. The formula BE = [EGSW2 - (EGS - 2EH2O)], yields -32.0, 36.9 and -25.6 kcal/mol for BE of H2O in structures A, B and C, respectively. Interestingly, these three, as well as all other considered GSW2 structures retain the C2 symmetry of the bare


11


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

Page 12 of 25

protonated GS. This leads to a (near) degeneracy of similar IR transitions in identical paired residues.38,

39

Consistent with this, the experiment, indeed, observes smaller number of N-H

stretch transitions than are expected for an asymmetric structure.

A"

B"

C"

Fig. 2 Optimized structures (omitting non-polar hydrogens) of three predicted conformers of [GS + 2H]2+--(H2O)2 complex with the expanded views of protonation sites (only one out of two sites for each conformer). In structure B (Fig. 2) each water molecule is flanked by the C=O(Pro) and N-H(Orn) groups, forming the hydrogen bonds of 1.82 Å and 1.81 Å, respectively and keeping one O-H completely free (Table S2). Structure C also exhibits a free O-H of each H2O, flanked between N-H and O=C of Orn. The corresponding hydrogen bonds (1.90 Å and 1.98 Å, respectively) are both weaker than in B. In conformer A, the water molecules are more tightly placed between NH3+ and O=C of Orn (1.79 Å and 1.68 Å, respectively), but with a longer (2.22 Å as acceptor) H2O—NH(Phe) bond. Unlike for the structures B and C, all O-H stretches of waters are bound in


12

Page 13 of 25

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


A either by very strong or very week H-bond interactions. This result is consistent with the experimental observation of nearly free O-H stretch in conformer 2a only. Another observation that we use for tentative assignment of the conformers relates to their UV spectroscopy. The strong interaction of waters with Orn in A lengthens the NH3+....π(Phe) bond to 2.55 Å (vs. 2.45 Å in bare GS and structures B and C), making the aromatic ring less coupled to the charge. Such decoupling should red shift the UV band origin of GSW2 towards its position in isolated neutral Phe.18, 38, 39 The only conformer of GSW2, where a significant red shift was observed is 2a (~100 cm-1). We therefore assign the conformer 2a to the calculated structure A and structures B and C to conformers 2b/2c. Firm validation of this assignment has been finally carried out based on comparison of the calculated and measured vibrational spectra of the GSW2 complex. Table 1 (see also Table S4 for more details) compares the scaled harmonic frequencies computed at B3LYP/6-31G(d,p) level for structures A, B and C with the measured IR frequencies of conformers 2a, 2b and 2c. The weakly bound O-H stretch of water is red-shifted by ~17 cm-1 in conformer 2a relative to the free O-H stretches in conformers 2b and 2c. The calculated harmonic frequencies reproduce this pattern with the red shifts of ~10 cm-1 and ~23 cm-1 for A relative to B and C, respectively. The bound O-H stretch in 2a exhibits a large red shift of ~45 cm-1 relative to the positions of the same band in two other conformers. That implies stronger hydrogen bonding of the O-H in conformer 2a, compared with 2b and 2c. Consistently, the computed bound O-H bond length in A(0.993 Å) is much longer than in B (0.981 Å) and C (0.973 Å), and OH....O=C hydrogen bond length is significantly shorter in A(1.676 Å), than in B(1.816 Å) and C(1.978 Å). Other scaled frequencies match the experiment only satisfactorily, reflecting the fundamental deficiency of harmonic approximation in accounting for very different anharmonicity of vibrations for hydrogen-bonded and for “free” O-H groups of the water


13


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

Page 14 of 25

molecules. An anharmonic treatment is therefore particularly highly desirable to confirm the structures and the locations of hydration sites for GSW2 complex, which has many intra- and inter-molecular H-bonds. To this end, we employ the B3LYP/6-31G(d, p) electronic structure method for the anharmonic potential energy surface, and the VSCF-PT2 algorithm for ab initio calculations of anharmonic frequencies and intensities of the waters in GSW2. The VSCF method has been proven to perform well in several applications determining the vibrational spectroscopy of biological molecules,41-44 including molecules solvated by H2O. 45, 46 As discussed earlier, the main computational cost for the VSCF method is in calculating grid points for the multi-dimensional PES. The cost scales up rapidly (Eq. 5 and 6) upon increasing the number of vibrational modes, number of the grid points per vibrational mode and the dimensionality of PES. GSW2 (182 atoms with 636 electrons) is very large for the calculations at a good level of ab initio quantum chemical method. We have computed water fundamentals of GSW2 in VSCF-PT2 approximation,23, 24, 47 by including in the construction of VSCF potential (considering 10 grid points along each normal modes) only those few stretching and bending modes of the peptide near the water molecules, which are, most likely, strongly coupled to the vibrations of H2O. For example, the pair-wise coupling potential of conformer A is constructed by considering total 15 normal modes including the stretching and bending modes of N-H: Orn side chain, NH3+ and all of the O-H. Frequencies of these vibrations are sensitive to the location of waters on GS and therefore can serve for an additional unbiased verification of the binding sites, proposed above.


14

Page 15 of 25

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


Table 1. Harmonic frequencies in GSW2 conformers.

Vib. modes Free OH

2a ~3694

Experiment3 2b 2c ~3711 ~3711

Bound OH

~3379

~3421

~3424

N-H Phe

~3441

~3409

~3404

N-H Orn

~3371

~3309

--

N-H Val

~3355

~3326

~3332

N-H Leu

~3294

~3289

~3307

NH3+

~3181

~3259

~3255

1 3

1, 2

A 3694 3694 3170 3173 3377 3377 3468 3468 3389 3389 3251 3257 3275 3275

Calculations3, 4 B C 3704 3717 3703 3717 3407 3538 3407 3538 3447 3462 3447 3462 3154 3255 3156 3255 3364 3399 3365 3399 3329 3272 3329 3272 3277 3023 3277 3024

Only vibrations of water and NH-stretch vibrations of GSW2 are shown. 2 The assignment is tentative.

Frequencies are in cm-1.

4

Scaled by the factor of 0.956. The two lines for the calculated bands are due

to same two vibrations (e.g two free O-H stretches) which are in resonance.

Table 2 compares the calculated anharmonic and measured frequencies and relative intensities of the O-H stretching vibrations of the water molecules in the three conformers of GSW2. The computed anharmonic frequencies of the bound O-H stretches match the experiment within ~30 cm-1is several times better than the scaled harmonic frequencies which matches to ~210 cm-1(see Table S5 for more details). The most striking improvement provided by the anharmonic treatment is the near spectroscopic accuracy achieved in reproducing the difference between frequencies of the two O-H vibrations in each conformer (5, 2 and 2 cm-1 discrepancy for the conformers 2a, 2b and 2c, respectively). In comparison to the scaled harmonic frequencies, these differences, on average, are greater than100 cm-1. Figure 3 summarizes our computational results and compares them with the experimental spectra. For each conformer we plot the scaled harmonic frequencies of NH-stretch vibrations of the complex (Table 1), and the calculated


15


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

Page 16 of 25

anharmonic OH-stretch frequencies of water molecules. The match of the calculated spectra to the experiment is quite good for the conformers 2a and 2b, except, perhaps, the couple of strongly blue-shifted transitions (NH-stretches of NH3+ and Orn) in 2a and a strongly red-shifted (NH-stretches of Orn) in 2b which are known to be intrinsically less accurately calculated in the HO approximation. Most of these deviations are satisfactorily corrected by anharmonic calculations (see S10 for details). The match between the experimental and the calculated spectra (Fig. 3c) is however less convincing for the minor conformer 2c; the only nicely reproduced experimental observation is the VSCF-PT2 calculated O-H stretch transitions and difference between the free and the bound O-H stretch vibrations of the waters (287 cm-1 in the experiment and 285 cm-1 in calculations). This fact suggests that the location of water molecules in the complex 2c is likely correct, although the overall geometry may not be accurately reproduced by the calculations.

Table 2. Frequencies of OH-stretch modes of water molecules in the GSW2 conformers. Experiment1, 2, 3

Calculations1, 4

Stretch mode

2a

2b

2c

A

B

C

Free OH

~3694

~3711

~3711

3719

3718

3740

(1.0)

(1.0)

(1.0)

(1.0)

(1.0)

(1.0)

~3379

~3421

~3424

3409

3430

3455

(0.97)

(1.46)

(1.0)

(9.2)

(5.5)

(1.56)

Bound OH

1

In cm-1. 2 Reference(18). 3Intensities of hydrogen-bonded OH stretch transitions relative to the intensities of the respective free OH transitions are shown in brackets. 4VSCF-PT2 anharmonic level.


16

Page 17 of 25

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


Being held by four strong intramolecular hydrogen bonds, the calculated structures of GS in the GSW2 complex do not differ significantly from the intrinsic structure of the peptide.18, 38, 39 The calculated RMSD of the optimized coordinates of GS in the three structures from the optimized coordinates of the bare doubly protonated peptide at B3LYP/6-31G(d, p) level suggest that GS in complex B resembles the best lowest energy structure of the fully dehydrated peptide (see tables in S3 for more details). The largest structural changes, induced by the microsolvation of GS, are calculated for structure A, where each of the two water molecules inserts into the tight space between the Phe ring and the Orn residue. Although this structure is not the most stable, it exhibits the largest among all the three observed conformers red shift of its UV band origin.18 Interestingly, exactly the same position of the UV band origin was observed for GSW4 complex, for which no other strong UV absorptions were detected, in particular at the positions of the band origins in the structures B and C of GSW2.18 This observation suggests that local structure of the GSW4 complexes around Phe chromophore is similar to that in structure A of the GSW2 complex. We therefore speculate that, similar to the structure A, the two hydration sites, each between the Phe ring and the Orn side chain, are always occupied in GSW4 by two of the four water molecules. Regarding the relative remoteness of the water binding sites in structures A from the sites in B/C, these should not significantly influence the binding energy of the two other waters in GSW4 at the sites, similar to the ones in B or C. We thus propose that may be the hydration of GS by four water molecules may ensures it's the most essential structural change toward the solution-like geometry though its validation is required as future study. If that is the case, this will complementary to the earlier qualitative suggestion that two waters determined, essentially, the structure:18 two more molecules are required to ensure occupation of the structurally critical binding sites (A) in all the peptides.


17


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

Page 18 of 25

CONCLUSIONS In this contribution, we have used a hybrid theoretical approach (scaled harmonic and local ab initio anharmonic approximations) and cold ion spectroscopy data to determine the three lowest energy structures of the non-covalent complex of a protonated decapeptide, GS, with two water molecules. The interplay of strong intramolecular and weaker intermolecular hydrogen bonds leads to accommodation of the first two water molecules at three distinct hydration sites on the surface of the peptide. Hydration of the protonated Orn side chains causes the most essential structural changes of the peptide, which nevertheless conserves its C2 symmetry. The presence of at least four water molecules most likely provides the major shift in geometry towards the solution structure for a part of the system and a detailed study in this direction is required in future. Overall, this study demonstrates how the theory can take up the challenge in solving geometry of large non-covalent complexes, for which cold ion spectroscopy can provide stringent structural constraints.


18

Page 19 of 25

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


Figure 3. IR spectra of conformers 2a, 2b and 2c (traces in panels A, B and C, respectively) of

doubly protonated GSW2 complex,18 the calculated scaled (factor 0.95) harmonic wavenumbers of NH-stretch vibrations (O, L, V, F and NH3+) in structures A, B, and C (vertical sticks in the panels A, B and C, respectively) and the calculated anharmonic frequencies of OH-stretch free (OHf) and bound (OHb) vibrations of water molecules in these structures (Lorentzian shapes). For each conformer the height of the sticks reflects the calculated relative intensities of NH transitions; the two OH bands reflect their relative intensities and are broaden to reproduce the widths of the respective experimentally observed transitions. The calculated scaled harmonic OH-stretch frequencies (red sticks) are shown for comparison where scaling factor is chosen such that to match experimental free OH-stretch for A exactly.

ASSOCIATED CONTENT Supporting Information: Experimental method, optimized structures, coordinates and parameters; relative energies, theory experiment comparison, comparisons of vibrational frequencies for conformers, comments on strong red-blue shifts are provided in the Supporting Information document. ACKNOWLEDGMENT TKR thanks the CSC-IT Center for Science, Finland for the computational resources and Central University of Jammu for infrastructural facility. RBG thanks the Academy of Finland for support in the framework of the FiDiPro program. OVB thanks Swiss National Science Foundation for supporting this work through the grant #200021_146389. Dr. B. Hirshberg, Dr. L. McCaslin and Dr. V. Sarkar are gratefully acknowledged for valuable comments.

REFERENCES


19


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

(1)

Page 20 of 25

Mercier, S. R.; Boyarkin, O. V.; Kamariotis, A.; Guglielmi, M.; Tavernelli, I.; Cascella,

M.; Rothlisberger, U.; Rizzo, T. R. Microsolvation Effects on the Excited-State Dynamics of Protonated Tryptophan. J. Am. Chem. Soc. 2006, 128, 16938-16943. (2)

Kopysov, V.; Nagornova, N. S.; Boyarkin, O. V. Identification of Tyrosine-

Phosphorylated Peptides Using Cold Ion Spectroscopy. J. Am. Chem. Soc. 2014, 136, 92889291. (3)

Kopysov, V.; Makarov, A.; Boyarkin, O. V. Colors for Molecular Masses: Fusion of

Spectroscopy and Mass Spectrometry for Identification of Biomolecules. Anal. Chem. 2015, 87, 4607-4611. (4)

Rizzo, T. R.; Boyarkin, O. V. Cryogenic Methods for the Spectroscopy of Large,

Biomolecular Ions. Top. Curr. Chem. 2015, 364, 43-97. (5)

Kopysov, V.; Boyarkin, O. V. Resonance Energy Transfer Relates the Gas-Phase

Structure and Pharmacological Activity of Opioid Peptides. Angew. Chem. Int. Ed. 2016, 55, 689-692. (6)

Burke, N. L.; Redwine, J. G.; Dean, J. C.; McLuckey, S. A.; Zwier, T. S. UV and IR

Spectroscopy of Cold Protonated Leucineenkephalin. Int. J. Mass Spectrom. 2015, 378, 196-205. (7)

Féraud, G.; Dedonder, C.; Jouvet, C.; Inokuchi, Y.; Haino, T.; Sekiya, R.; Ebata, T.

Development of Ultraviolet-Ultraviolet Hole-Burning Spectroscopy for Cold Gas-Phase Ions. J. Phys. Chem. Lett. 2014, 5, 1236-1240. (8)

Garand, E.; Kamrath, M. Z.; Jordan, P. A.; Wolk, A. B.; Leavitt, C. M.; McCoy, A. B.;

Miller, S. J.; Johnson, M. A. Determination of Noncovalent Docking by Infrared Spectroscopy of Cold Gas-Phase Complexes. Science 2012, 335, 694-698. (9)

deVries, M. S.; Hobza, P. Gas-Phase Spectroscopy of Biomolecular Building Blocks.

Ann. Rev. Phys. Chem. 2007, 58, 585-612. (10) Chakraborty, T.; Holm, A. I. S.; Hvelplund, P.; Nielsen, S. B.; Poully, J. C.; Worm, E. S.; Williams, E. R. On the Survival of Peptide Cations After Electron Capture: Role of Internal Hydrogen Bonding and Microsolvation. J. Am. Soc. Mass Spectr. 2006, 17, 1675-1680. (11) Fricke, H.; Schwing, K.; Gerlach, A.; Unterberg, C.; Gerhards, M. Investigations of the Water Clusters of the Protected Amino Acid Ac-Phe-OMe by Applying IR/UV Double Resonance Spectroscopy: Microsolvation of the Backbone. Phys. Chem. Chem. Phys. 2010, 12, 3511-3521.


20

Page 21 of 25

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


(12) Scuderi, D.; Bakker, J. M.; Durand, S.; Maitre, P.; Sharma, A.; Martens, J. K.; Nicol, E.; Clavaguera, C.; Ohanessian, G. Structure of Singly Hydrated, Protonated Phospho-Tyrosine. Int. J. Mass Spectrom. 2011, 308, 338-347. (13) Milosavljevic, A. R.; Cerovski, V. Z.; Canon, F.; Nahon, L.; Giuliani, A. NanosolvationInduced Stabilization of a Protonated Peptide Dimer Isolated in the Gas Phase. Angew. Chem. Int. Ed. 2013, 52, 7286-7290. (14) Warnke, S.; von Helden, G.; Pagel, K. Protein Structure in the Gas Phase: The Influence of Side-Chain Microsolvation. J. Am. Chem. Soc. 2013, 135, 1177-1180. (15) Silveira, J. A.; Servage, K. A.; Gamage, C. M.; Russell, D. H. Cryogenic Ion MobilityMass Spectrometry Captures Hydrated Ions Produced During Electrospray Ionization. J. Phys. Chem. A 2013, 117, 953-961. (16) Chaban, G. M.; Gerber, R. B. Anharmonic Vibrational Spectroscopy of the Glycine– Water

Complex:

Calculations

for

ab

initio,

Empirical,

and

Hybrid

Quantum

Mechanics/Molecular Mechanics Potentials. J. Chem. Phys. 2001, 115, 1340-1348. (17) Jockusch, R. A.; Kroemer, R. T.; O. Talbot, F. O.; Simons, J. P. Hydrated Sugars in the Gas Phase: Spectroscopy and Conformation of Singly Hydrated Phenyl beta-D-Glucopyranoside. J. Phys. Chem. A 2003, 107, 10725-10732. (18). Nagornova, N. S.; Rizzo, T. R.; Boyarkin, O. V. Interplay of Intra- and Intermolecular HBonding in a Progressively Solvated Macrocyclic Peptide. Science 2012, 336, 320-323. (19). Scott, A. P.; Radom, L. Harmonic Vibrational Frequencies: An Evaluation of HartreeFock, Moller-Plesset, Quadratic Configuration Interaction, Density Functional Theory and Semiempirical Scale Factors. J. Phys. Chem. 1996, 100, 16502-16513. (20) Gerber, R. B.; Ratner, M. A. Semi-Classical Self-Consistent Field Approximation for Eigenvalues of Coupled-Vibration Systems. Chem. Phys. Lett. 1979, 68, 195-198. (21) Bowman, J. M. Self-Consistent Field Energies and Wavefunctions for Coupled Oscillators. J. Chem. Phys. 1978, 68, 608-610. (22) Norris, L. S.; Ratner, M. A.; Roitberg, A. E.; Gerber, R. B. Møller-Plesset Perturbation Theory Applied to Vibrational Problems. J. Chem. Phys. 1996, 105, 11261-11267. (23) Roy, T. K.; Gerber, R. B. Vibrational Self-Consistent Field Calculations for Spectroscopy of Biological Molecules: New Algorithmic Developments and Applications. Phys. Chem. Chem. Phys. 2013, 15, 9468-9492.


21


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

Page 22 of 25

(24) Roy, T. K.; Carrington Jr. T.; Gerber, R. B. Approximate First-Principles Anharmonic Calculations of Polyatomic Spectra Using MP2 and B3LYP Potentials: Comparisons with Experiment. J. Phys. Chem. A 2014, 118, 6730-6739. (25) Lee, C. T.; Yang, W. T. Parr, R. G. Development of the Colle-Salvetti Correlation-Energy Formula into a Functional of the Electron Density. Phys. Rev. B 1988, 37, 785-789. (26) Becke, A. D. Density-Fucntional Theormochemistry. III. The Role of Exact Exchange. J. Chem. Phys.1993, 98, 5648-5652. (27) Hariharan, P. C.; Pople, J. A. The Influence of Polarization Functions on Molecular Orbital Hydrogenation Energies, Theor. Chim. Acta 1973, 28, 213-222. (28) Roy, T. K.; Kopysov, V.; Nagornova, N. S.; Rizzo, T. R.; Boyarkin, O. V.; Gerber, R. B. Conformational Structures of a Decapeptide Validated by First Principles Calculations and Cold Ion Spectroscopy. ChemPhysChem. 2015, 16, 1374-1378. (29) Baldauf, C.; Pagel, K.; Warnke, S.; von Helden, G.; Koksc, B.; Blum, V.; Scheffler, M. How Cations Change Peptide Structure. Chem. Eur. J. 2013, 19, 11224-11234. (30) Chutia, S.; Rossi M.; Blum, V. Water Adsorption at Two Unsolvated Peptides with a Protonated Lysine Residue: From Self-Solvation to Solvation. J. Phys. Chem. B, 2012, 116, 14788−14804. (31) Ropo, M.; Blum, V.; Baldauf, C. Trends for Isolated Amino Acids and Dipeptides: Conformation, divalent ion binding, and remarkable similarity of binding to calcium and lead. Sci. Rep. 2016, 6, 3577. (32) Schubert, F.; Pagel, K.; Rossi, M.; Warnke, S.; Salwiczek, M.; Koksch, B.; von Helden, G.; Blum, V.; Baldauf, C.; Scheffler, M. Native Like Helices in a Specially Designed β Peptide in the Gas Phase. Phys. Chem. Chem. Phys. 2015, 17, 5376-5385. (33) Supady, A.; Blum, V.; Baldauf, C. First-Principles Molecular Structure Search with a Genetic Algorithm. J. Chem. Inf. Model. 2015, 55, 2338-2348. (34) Voronina, L.; Masson, A.; Kamrath, M.; Schubert, F.; Clemmer, D.; Baldauf, C.; T. Rizzo. Conformations of Prolyl-Peptide Bonds in the Bradykinin 1-5 Fragment in Solution and in the Gas Phase. J. Am. Chem. Soc. 2016, 138, 9224–9233. (35) Jaeqx, S.; Oomens, J.; Cimas, A.; Gaigeot, M.-P.; Rijs, A. M. Gas-Phase Peptide Structures Unraveled by Far-IR Spectroscopy: Combining IR-UV Ion-Dip Experiments with


22

Page 23 of 25

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


Born–Oppenheimer Molecular Dynamics Simulations. Angew. Chem. Int. Ed. 2014, 53, 36633666. (36) Cimas, A.; Vaden, T. D.; de Boer, T. S. J. A.; Snoek, L. C.; Gaigeot, M. P. Ab Initio Molecular Dynamics of Protonated Dialanine and Comparison to Infrared Multiphoton Dissociation Experiments. J. Chem. Theor. Comp. 2009, 5, 1068-1078. (37) Joshi, K.; Semrouni, D.; Ohanessian, G.; Clavaguera, C. Structures and IR Spectra of the Gramicidin S Peptide: Pushing the Quest for Low-Energy Conformations. J. Phys. Chem. B 2012, 116, 483-490. (38) Nagornova, N. S.; Rizzo, T. R.; Boyarkin, O. V. Highly Resolved Spectra of Gas-Phase Gramicidin S: A Benchmark for Peptide Structure Calculations. J. Am. Chem. Soc. 2010, 132, 4040-4042. (39) Nagornova, N. S.; Guglielmi, M.; Doemer, M.; Tavernelli, I.; Rothlisberger, U.; Rizzo, T. R.; Boyarkin, O. V. Cold-Ion Spectroscopy Reveals the Intrinsic Structure of a Decapeptide. Angew. Chem. Int. Ed. 2011, 50, 5383-5386. (40) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; et al. Gaussian 09, Revision D.01, Gaussian, Inc. Wallingford CT, 2010. (41) Adesokan, A. A.; Gerber, R. B. Anharmonic Vibrational Spectroscopy Calculations for Proton-Bound Amino Acid Dimers. J. Phys. Chem. A 2009, 113, 1905-1912. (42) Brauer, B.; Gerber, R. B.; Kabeláč, M.; Hobza, P.; Bakker, J. M.; Riziq, A. G. A.; de Vries, M. S. Vibrational Spectroscopy of the G•••C Base Pair: Experiment, Harmonic and Anharmonic Calculations, and the Nature of the Anharmonic Couplings. J. Phys. Chem. A 2005, 109, 6974-6984. (43) Roitberg, A.; Gerber, R. B.; Elber, R.; Ratner, M. A. Anharmonic Wave-Functions of Proteins - Quantum Self-Consistent-Field Calculations of BPTI. Science 1995, 268, 1319-1322. (44) Pele, L.; Brauer, B.; Gerber, R. B. Acceleration of Correlation-Corrected Vibrational Self-Consistent Field Calculation Times for Large Polyatomic Molecules. Theor. Chem. Acc. 2007, 117, 69-72. (45) Gregurick, S. K.; Fredj, E.; Elber, R.; Gerber, R. B. Vibrational Spectroscopy of Peptides and Peptide-Water Complexes: Anharmonic Coupled-Mode Calculations. J. Phys. Chem. B 1997, 101, 8595-8606.


23


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

Page 24 of 25

(46) Pincu, M.; Cocinero, E. J.; Mayorkas, N.; Brauer, B.; Davis, B. G.; Gerber, R. B.; Simons, J. P. Isotopic Hydration of Cellobiose: Vibrational Spectroscopy and Dynamical Simulations. J. Phys. Chem. A 2011, 115, 9498-9509. (47) Roy, T. K.; Sharma, R.; Gerber, R. B. First-Principles Anharmonic Quantum Calculations for Peptide Spectroscopy: VSCF Calculations and Comparison with Experiments. Phys. Chem. Chem. Phys. 2016, 16, 9468-9492.


24

Page 25 of 25

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


TOC GRAPHICS

Where to go ?


25

A Decapeptide Hydrated by Two Waters: Conformers Determined by

Recommend Documents