Systematic Study on Hydrated Arginine: Clear Theoretical Evidence

Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)

Article

Systematic Study on Hydrated Arginine: Clear Theoretical Evidence for the Canonical-to-Zwitterionic Structure Transition Hongbao Li, Andong Hu, Jun Jiang, and Yi Luo J. Phys. Chem. A, Just Accepted Manuscript • Publication Date (Web): 25 Apr 2017 Downloaded from http://pubs.acs.org on April 30, 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 24 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

Systematic Study on Hydrated Arginine: Clear Theoretical Evidence for the Canonical-to-Zwitterionic Structure Transition Hongbao Li,*, † Andong Hu, † Jun Jiang,*, † and Yi Luo †, ‡ †

Hefei National Laboratory for Physical Sciences at the Microscale, School of Chemistry and Materials Science,

University of Science and Technology of China, Hefei, Anhui, 230026, China. ‡

Guizhou Provincial Key Laboratory of Computational Nano-Material Science, and Guizhou Synergetic Innovation

Center of Scientific Big Data for Advanced Manufacturing Technology, Guizhou Education University, Guiyang, Guizhou, 550018, China

ABSTRACT Extensive ab initio investigations have been performed to characterize the stable conformers of hydrated arginine (Arg-H2O). Many new low-energy canonical Arg-H2O conformers were identified and they are more stable than previous results. The large energy differences (more than 5.00 kcal mol-1) between the canonical and zwitterionic Arg-H2O isomers calculated by the composite CBS-QB3 method confirmed the dominance of the zwitterions. The micro effects of corrections of the zero-point energy and the basis set superposition error on the stability of hydrated isomers were carefully examined for the first time. The infrared (IR) spectra were simulated at a recommended temperature and the notable spectral differences enable the unambiguous identification of the different hydrated forms. Further transition state calculations revealed that the canonical Arg-H2O can be transformed to the zwitterions using the amino group as a bridge. Our study thus shows valuable insights into the hydration of large flexible molecules and provides solid theoretical evidence for the canonical-to-zwitterionic structure transition of hydrated arginine.

1. INTRODUCTION Water plays a key role in determining the structures, properties and functions of crucial biological molecules in aqueous solution.1 It can influence the protonation2-5 or deprotonation6-8 site of molecules. It can also affect the stability of the zwitterionic form of amino acids and peptides, which is one of the most intriguing examples of dramatic effects of water on biomolecules. It is well-known that the charge separated zwitterionic conformers mostly exist in solid state or in solution within a narrow pH range, and can hardly be found dominantly in the gas phase.9-13 Therefore, the canonical-to-zwitterionic 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

Page 2 of 24

structure transition of amino acids becomes an important issue of biological phenomena triggered by the presence of water molecules.14-17 This also raises the question of how many water molecules are required to induce such structure transition for a given amino acid molecule.14 One advantage of studying the hydration of gaseous molecules is that the minor effects of each water molecule on the structure of hydrated clusters can be probed in detail. In fact, the hydration of several amino acids has been studied up to now.18-22 Through a systematic theoretical study on Gly(H2O)n (Gly=glycine, n=1-8),18 Gordon et al. had found that at least seven water molecules are required to set the zwitterionic form to the global minimum, lying ca. 1.3 kcal mol-1 below the canonical conformer. For phenylalanine, tryptophan and cysteine, at least three,19 four20 and six21 water molecules should be included to make the zwitterions energetically competitive with the canonical conformers. Besides, proton transformation between different isomers through hydrogen bond (HB) has been recognized as a fundamental process through which many chemical reactions are carried out.23,

24

The hydration

information thus plays an important role in revealing the intrinsic proton migration chemistry in the canonical-to-zwitterionic process of biomolecules. Recently, a theoretical study on the hydration of gaseous arginine (Arg)25, 26 has predicted that only one single water molecule is required to make the zwitterionic Arg-H2O more stable than its canonical form. However, the conclusion is not so compelling because only several arginine monomers were taken into account in the generation of the trial structures of Arg-H2O, which leads to an uncompleted search on the potential energy surface (PES) of this cluster. Another problem is the important energy barriers between the canonical and zwitterionic hydrated isomers were only qualitatively predicted mainly due to the lack of adequate structure information. This may lead to wrong spectral predictions since the lowenergy hydrated isomers are not always from the hydration of the most stable arginine monomers. Furthermore, previous studies12, 27 have verified that, compared to the advanced coupled-cluster single and doubles (CCSD) results, the DFT-B3LYP method is always misleading with respect to the energy ordering while the MP2 method usually overestimates the canonical isomers and underestimates the zwitterions. Therefore, high-level computational methods are required to achieve reliable conclusions. To solve the above problems, based on our previous study on the structures and spectra of gaseous arginine,10, 11 we performed a more comprehensive theoretical study to explore the coordinates of one water molecule connecting to three different arginine isomers. A large number of trial structures of hydrated arginine were generated; from these, a geometric search using first-principles calculations produced many new low-energy conformers. The effects of the dispersion interaction and the basis set superposition error (BSSE) were calculated and evaluated for the first time, to the best of our knowledge, 2 ACS Paragon Plus Environment

Page 3 of 24 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


as almost all previous studies had omitted them.18-22, 25-26 The unique infrared (IR) spectral features of the hydrated conformers have been assigned for different isomers, which can be used to determine the dominant form in future experiments. Accurate energy barriers between different isomers of Arg-H2O were predicted to give more reliable IR spectral results. This article is organized as follows: In Section 2, we describe the generation of the trial structures, the theoretical methods for high-level calculations, the transition state calculation, as well as the IR spectra simulations. Section 3 presents the optimized structures and the simulated IR spectra, based on which we discuss the canonical-to-zwitterionic structure transformation and the spectral differences. Concluding remarks are finally given in Section 4.

2. COMPUTATIONAL DETAILS

Figure 1. Geometric structures of the hydrated arginine conformer Arg-H2O in three different possible forms: two are canonical forms (R1 and R2) and one is the zwitterionic form (Rz).

Arg has three different neutral isomers in the gas phase: two are canonical forms and one is a zwitterion.10, 11, 27 To achieve reliable structural information, the most stable monomers of arginine in each neutral isomer should be included. Arg contains three unique functional groups: the carboxylate group, the amino group and the guanidine group in the side chain. As depicted in Figure 1, all the three functional groups can be used as HB donors or acceptors to interact with the water molecule while the water molecule has one HB acceptor (O atom) and two HB donors (H atom). In principle, a donor can interact with more than three acceptors simultaneously,28, 29 but in practice, the hydrogen atom can only be bonded to no more than one acceptor due to the high demand of spatial density. Similarly, one acceptor atom (N or O) can be bonded to two donors at most. For small amino acids such as glycine, only eight different monomers exist, all of which can be used to generate trial hydrated structures.18 But for Arg with large and flexible structure, a huge number of 3 ACS Paragon Plus Environment


Page 4 of 24

local minima on its PES with varying geometries exist,27 making it impossible to take all of them into account in the initial generation process. In this work, we selected 80, 60 and 80 low-energy arginine monomers (more than 5 kcal mol-1) for the two canonical and one zwitterionic isomers.27 For each form, one representative conformer (c5, c12 and z22 from ref. 27) was chosen, and a water molecule was added to interact with its three functional groups by exhaustively scanning over configuration space of possible sites. The generated Arg-H2O structures were then optimized at the HF/3-21G (d) level of theory. 20, 18 and 14 unique hydrated structures for the representative conformer of the three isomers were obtained after geometry optimizations. Then a water molecule was put at the same positions as in the unique hydrated conformers (comparing to the three functional groups) to interact with all the other selected monomers. More than 3800 trial structures of Arg-H2O were finally generated by this strategy. All the trial hydrated structures were first optimized at the HF/3-21G (d) level of theory and the unique conformers were then optimized at the B3LYP level with the basis set of 6-31G (d). Conformers with energy difference within 8 kcal mol-1 were further optimized at the B3LYP/6-311++G (d, p) level. All the local minima were confirmed by ascertaining that all the harmonic vibrational frequencies are real. The electronic energies were finally determined at the advanced CCSD/6-31++G (d, p) level. The seven most important conformers were further calculated at an even more extensive CCSD/aug-ccpVDZ level30-32(with 443 basis functions considered) and by the composite CBS-QB3 method. A calculation with higher level of theory and bigger basis sets (such as CCSD (T)/aug-cc-pVTZ) may provide a more conclusive support, but it is unfortunately prohibitively expensive to do with such a large system (up to 966 basis functions included). The low-energy conformers and their frequencies were also calculated at the M062X/6-311++G (d, p) level of theory to better describe the HBs and the dispersion interactions.33, 34 For the M062X calculation, an “ultrafine” numerical integration grid has been used to ensure reliable results for systems with noncovalent interactions. The counterpoise procedure was employed for the hydrated clusters to estimate the basis set superposition error (BSSE).35 The calculated vibrational frequencies of the hydrated conformers are all scaled by uniform factors of 0.981330,36 and 0.95411,12,37 for the B3LYP and M062X methods, respectively. The transition states for the transformation from the most populated canonical conformers to the zwitterion were identified at the B3LYP/6-311++G (d, p) level of theory. The nature of the stationary and saddle points was verified by the Hessian calculations. The electronic energies of the transition states were also determined at the CCSD/6-31++G (d, p) level and the Gibbs free energy correction at 133 K has been used to calculate the reaction barriers. The reaction rate constant k has been calculated by38 4 ACS Paragon Plus Environment

Page 5 of 24 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


=

∆

ℎ

where kb denotes the Boltzmann’s constant, h is Plank’s constant, ∆G is the Gibbs free energy differences between the reactant and the transition state, R is the gas constant and T is the temperature in Kelvin. The Gibbs free energy is defined as G = H – TS, with H = E + RT. The relative Gibbs free energies are calculated by ∆G = ∆E – T∆S, while the Gibbs free energy corrections were obtained from vibrational frequency results. All the above calculations were carried out using the Gaussian 09 software package.39

3. RESULTS AND DISCUSSION 3.1. Conformations: Structure, Energy and the Correction Effects As listed in Table 1, the relative electronic energies, the relative zero-point vibrational energy correction (∆ZPVE) and the basis set superposition error correction (∆BSSE), the corrected relative electronic and Gibbs free energies of Arg-H2O are given. The five most stable conformers in each isomer, denoted Rzn, R1n and R2n (n=1-5), were selected for comparison. C5-1 and C4-1 are the two most stable canonical conformers reported in ref. 25. The thermodynamic parameters are calculated at a temperature of 133 K. This is in accordance with the recent infrared photodissociation (IRPD) experiment on the hydration of gaseous m-aminobenzonic acid,40 which also contains the amino group and the carboxylate group. Our previous study on deprotonated tyrosine and cysteine also suggested that the simulated IR spectra in the gas phase at around 150 K may coincide better with the experimental measurements.36 Many new low-energy hydrated conformers were identified, which indicates the effectiveness of our searching strategy. The most stable zwitterions Z22-1 and Z21-2 in ref. 25 corresponded to our conformer Rz1 and Rz3, and a new conformer Rz2 was found by this work. It is striking to find that the previously reported most stable canonical conformers C5-1 and C4-1, belonging to the R1 form, no longer hold the lowest electronic energies at the accurate CCSD level, although they came from the hydration of the two most populated canonical Arg monomers.27 In fact, the lowest-energy conformer (Rz1, R11 and R21) in each isomer originated from the combination of a water molecule with the Arg monomer z22, c6 and c14, respectively, as defined in ref. 27. This indicates that the previous strategy25 considering the hydration of only a few most stable monomers at the first step is actually inadequate, especially for the canonical isomers. 5 ACS Paragon Plus Environment


Page 6 of 24

Table 1. The computed relative electronic energies, relative zero-point vibrational energy correction (∆ZPVE) and the basis set superposition error correction (∆BSSE), the corrected relative electronic and Gibbs free energies (in kcal mol-1) of the most stable conformers of the three neutral Arg-H2O isomers at the CCSD/6-31++G(d, p) and B3LYP (or M062X)/6-311++G(d, p) levels. C5-1 and C4-1 are the two most stable canonical conformers from ref. 25. Conf.

Relative electronic energy

∆ZPVE

∆BSSE

Corrected relative electronic energy

Corrected relative Gibbs free energy

ECCSD

EB3LYP

EM062X

∆B3LYP

∆M062X

∆B3LYP

∆M062X

ECCSD

EB3LYP

EM062X

GCCSD

GB3LYP

GM062X

Rz1

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

Rz2

0.70

1.00

0.79

0.29

0.09

0.06

-0.01

1.05

1.35

0.88

1.16

1.46

0.93

Rz3

1.12

0.49

1.30

0.08

-0.16

-0.09

-0.14

1.11

0.48

1.00

1.07

0.44

0.91

Rz4

2.96

2.35

2.31

-0.21

-0.11

0.02

-0.08

2.77

2.16

2.12

2.69

2.08

2.10

Rz5

3.15

1.19

1.30

-0.32

-0.16

-0.02

-0.14

2.81

0.85

1.00

2.65

0.69

0.91

R11

2.51

6.88

6.56

-0.92

-0.68

-0.31

-0.36

1.28

5.65

5.51

0.83

5.20

5.27

R12

3.09

7.43

7.41

-0.57

-0.50

-0.24

-0.27

2.28

6.62

6.65

2.02

6.36

6.42

R13

3.39

6.94

6.48

-0.89

-0.75

-0.18

-0.21

2.32

5.87

5.53

1.97

5.52

5.35

R14

3.50

7.18

6.94

-0.72

-0.37

-0.16

-0.05

2.62

6.30

6.52

2.09

5.76

6.37

R15

3.23

8.03

7.49

-0.68

-0.54

0.17

0.92

2.71

7.51

7.87

2.38

7.18

7.70

R21

2.37

5.31

5.50

-0.23

-0.32

-0.06

-0.07

2.08

5.02

5.11

1.86

4.80

4.86

R22

3.01

6.45

6.16

-0.36

-0.55

0.00

-0.02

2.65

6.09

5.59

2.35

5.79

5.31

R23

3.94

6.54

6.75

-0.37

-0.41

-0.01

-0.01

3.56

6.17

6.33

3.32

5.93

6.12

R24

5.50

7.12

7.82

-0.57

-0.84

0.03

-0.05

4.95

6.57

6.93

4.56

6.19

6.45

R25

5.88

8.94

11.66

-0.67

-0.76

-0.11

-0.53

5.11

8.16

10.37

4.68

7.73

9.82

C5-1

5.19

3.93

8.34

-2.09

-1.96

-0.07

-0.21

3.02

1.77

6.17

2.45

1.20

5.15

C4-1

4.39

3.69

8.51

-1.74

-2.00

0.00

-0.14

2.66

1.96

6.37

2.19

1.49

5.84

Using the global minimum Rz1 as the benchmark, the relative energies of conformers in each isomer exhibit small changes for different DFT methods. These energy differences become even much smaller at the advanced CCSD level. For example, the electronic energies of conformers in R1 are usually 6-8 kcal mol-1 higher than the zwitterions at the B3LYP (or M062X) level, but only 2-4 kcal mol-1 energy differences were found at the CCSD level. Generally, the energy difference changes about 4 (or 3) kcal mol-1 for R1 (or R2) at different methods. It is interesting to find that the previously reported conformers 6 ACS Paragon Plus Environment

Page 7 of 24 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 2. Corrections of the zero-point vibrational energy (∆ZPVE) and the basis set superposition error (∆BSSE) of the 20 most stable conformers in isomer R1. (a) The relative energies of the corrections at the B3LYP and M062X levels; (b-d) Comparison of the relative energies of the 20 conformers before and after the corrections at the CCSD, B3LYP and M062X levels; (e-f) Comparison of the relative energies for the three different methods on the 20 conformers before and after the corrections. Note: the corrections used in the CCSD method are from the B3LYP level and all the relative energies are in kcal mol-1.

C5-1 and C4-1 have lower energies than our canonical results at the B3LYP level. However, their relative values become higher at the M062X and CCSD level, which again indicates the unreliability of the DFT/B3LYP method. The corrected values for the zero-point vibrational energy and the basis set superposition error of conformers in Rz are larger than those in R1 and R2, which low down the energy differences between the canonical and zwitterionic isomers (decrease by 0.5-1 kcal mol-1). As shown in Figure 2, the contributions from corrections of the zero-point vibrational energy (∆ZPVE) and the basis set superposition error (∆BSSE) of the 20 most stable conformers in isomer R1 by the three different methods are plotted and compared. These conformers are ordered by the corrected relative electronic energies at the accurate CCSD/6-31++G (d, p) level. The average value of the absolute ZPVE is 155.38 (or 157.70) kcal mol-1 at the B3LYP (or M062X) level, while the corresponding value of the absolute BSSE is only 1.08 (or 1.40) kcal mol-1. Although the BSSE correction is very small, it is still 7 ACS Paragon Plus Environment


Page 8 of 24

very important since the electronic energy difference between the canonical and zwitterionic isomers is only 2-4 kcal mol-1 before the correction (see Table 1). From Figure 2(a), the relative BSSE values exhibit small changes comparing with the relative ZPVE values. It seems that the B3LYP method performs better than the M062X for both of the two corrections. Therefore, the B3LYP corrections were employed in the following calculations. Comparing the relative energies of these conformers before and after the corrections by the three different methods (Figure 2b, 2c and 2d), we found that due to the positive values of the corrections, most of the corrected relative energies are a little larger than before. Although the global minimum at different theoretical levels remains unchanged, the energies of some conformers become much larger, especially at the DFT levels. Through comparison of the relative energies by different methods before and after the corrections (Figure 2e and 2f), we found that the B3LYP method is the most unreliable one and inclusion of the dispersion energy can improve the theoretical results. All the above analysis indicates that such corrections are indispensable and can greatly influence the relative energies of different conformers.

Figure 3. Structures of the most stable conformers of Arg-H2O in the three neutral isomers and their relative electronic energies (in kcal mol-1) based on temperature-independent electronic energies calculated at the CCSD/6-31++G (d, p) level with the zero-point energy correction and the BSSE correction from B3LYP/6-311++G (d, p). The relative energies in parentheses are determined by further CCSD/aug-cc-pVDZ (the former) and the CBS-QB3 (the latter) calculations. The intermolecular HB interactions are depicted by dotted lines in angstrom (Å).

8 ACS Paragon Plus Environment

Page 9 of 24 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 geometric structure of the most stable conformers of Arg-H2O in the three neutral isomers, together with the intermolecular HB interactions and the relative electronic energies are displayed in Figure 3. All the low-energy conformers fold spherically using the water molecule as a bridge connecting the carboxylate and the guanidine group. It is interesting to find that these two functional groups in our results are in two nearly parallel planes and three intermolecular HBs have been formed in most cases. However, for the previously reported canonical conformers C5-1 and C4-125, these two groups and the water molecule are almost in the same plane and only two HBs exist. These structure differences will definitely lead to the dominance of our structural results. At the high CCSD/6-31++G (d, p) level of theory, the relative electronic energies of our newly found canonical conformers are all lower than the previously reported ones. Considering the zero-point and the BSSE correction, the smallest electronic energy difference between the canonical and zwitterionic isomers is only 1.28 kcal mol-1 while the Gibbs free energy difference is even 0.83 kcal mol-1(see Table 1). To collect more convincing evidence, further CCSD/aug-cc-pVDZ and CBS-QB3 calculations for the seven most important conformers (Rz1-Rz3, R11, R21, C5-1 and C4-1) were performed. It is found that the global minimum is still the zwitterionic conformer Rz1. Furthermore, it is interesting to find that the energy differences between the zwitterionic and canonical forms become larger with the improved theoretical methods (more than 3.18 and 5.00 kcal mol-1), which is opposite to the situation on dipeptide arginylglycine.12 It seems that with the improvement in the computational accuracy, the identification of the dominant form from the energetic point of view becomes possible for the hydrated molecules. Therefore, we believed that, at least for this hydrated molecule, calculations at the CCSD/6-311++G (d, p) are good enough to give reliable theoretical values. A higher level of theory and bigger basis sets were not attempted for this large molecule due to the expensive computational cost (up to 966 basis functions included at the CCSD (T)/aug-cc-pVTZ level of theory). However, previous CCSD (T) calculations with very large basis sets (up to 1380 functions included) for three zwitterionic and five canonical arginine conformers provided almost the same relative stabilities as that at a cheaper

CCSD/6-31++G(d, p) approach (only 312 basis functions

included).27, 41 At present, both the measurement of energy in several kcal mol-1 and probing the molecular geometry of the isomers are still difficult in experiment. So various spectroscopic techniques1, 10-13, 42 should be applied to identify the dominant form in future experiment. The simulated spectral differences between the canonical and zwitterionic isomers can help one to deeply understand the structure-property relationship. 9 ACS Paragon Plus Environment


Page 10 of 24

3.2. Equilibrium Distributions and Infrared Spectra Table 2. Conformers with low Gibbs free energy of the three hydrated isomers, together with their respective percent shares at 133 K.40

a, b

R1

Percenta(%)

R2

Percenta(%)

Rz

Percenta(%)

R11

95.90

R21

86.23

Rz1

97.10

R12

1.06

R22

13.42

Rz2

1.20

R13

1.29

R23

0.34

Rz3

1.70

R14

0.81

Sum

100

Sum

99.99

R15

0.27

Sum

99.33

R1

Percentb(%)

R2

Percentb(%)

Rz

Percentb(%)

R11

88.20

R21

84.81

Rz1

90.50

R12

3.19

R22

14.93

Rz2

3.82

R13

1.78

R23

0.26

Rz3

5.67

R14

2.55

Sum

100

Sum

99.99

R15

3.67

Sum

99.40

The percent shares based on the CCSD/6-31++G(d, p) electronic energies and frequency calculations at (a)

B3LYP/6-311++G(d, p) and (b) M062X/6-311++G(d, p) levels of theory.

The equilibrium distributions of the most populated conformers in the three neutral isomers of Arg-H2O corrected by the B3LYP and M062X methods at 133 K are listed in Table 2. Due to the low temperature, the calculated percentage shares at the advanced CCSD level are mainly distributed over the lowestenergy conformer (R11, R21 and Rz1) of each isomer. This is rather different from the results of gaseous arginine, for which the most abundant conformers are always c5 and c4.27 As the three neutral forms of Arg-H2O may coexist in future experiment, the IR spectra for all of the populated conformers listed in Table 2 were simulated and are displayed in Figure 4a (by B3LYP method) and 4b (by M062X method). More than 99% of all conformers in each isomer were included in the averaged theoretical spectra (the “SUM”). Each of the three isomers produces several unique IR spectral features, providing a convenient way to distinguish between them. In the 500-1000 cm-1 region, all the three isomers have at least two peaks, which are mainly attributable to the NH and NH2 out-of-plane bending mode of the guanidine group. However, due to their more anharmonic behavior,43, 44 these modes cannot be observed by experiment at present. The first peak of R1 at 612 cm-1 in Figure 4(a) is contributed by the OH out-of-plane bending mode of the water 10 ACS Paragon Plus Environment

Page 11 of 24 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 4. Simulated IR spectra of dominant conformers in the three neutral isomers of Arg-H2O system R1, R2 and Rz at 133 K40 calculated with (a) B3LYP and (b) M062X functional, as well as their summations (SUM) calculated using the percentage shares listed in Table 2. A Lorentzian profile with a full width at half maximum (FWHM) of 20 cm-1 is used to convolute the calculated spectra.

molecule, which is blue-shifted to 877 cm-1 for R2 due to the different intensity of the intermolecular HB interactions involving H2O (see Figure 3). In the 1400-1800 cm-1 region, more unique spectral features of the canonical conformers have been observed, which is thus useful for distinguishing them from the zwitterions. For example, the strong peak at 1410 cm-1 of R1 (or 1416 cm-1 of R2) originates from the OH in-the-plane bending mode of the carboxylate group. Another strong peak at 1761 cm-1 of R1 (or 1758 cm-1 of R2) is due to the C=O stretching mode of the same group. But these two peaks are totally absent in the zwitterions. Two strong peaks at 1666 and 1693 cm-1 appeared in the IR experiment of gaseous arginine at 443 K45 and have been explained by a combination of the three neutral arginine isomers.11 For R1, the experimental peak (at 1693 cm-1) red-shifted to 1678 cm-1, with the contributions from the C=N stretching mode and NH2 scissoring mode of the guanidine group. For R2 and Rz, the experimental peak (at 1666 cm-1) red-shifted to 1642 and 1634 cm-1, respectively, but with different contributions. For R2, it is resulted from the NH2 scissoring mode of the amino group and the C=N stretching mode of the guanidine group. But for Rz, it is attributable to the NH2 scissoring mode of the amino group and the guanidine group and the O=C=O asymmetrical stretching mode in the deprotonated carboxylate group. After the adsorption of a water molecule, all these peaks have red-shifted ca. 15-30 cm-1 mainly due to the newly formed intermolecular interactions (Figure 3). 11 ACS Paragon Plus Environment


Page 12 of 24

In the 3000-4000 cm-1 region, two unique peaks at 3077 and 3162 cm-1 of Rz were observed and are attributable to the NH stretching mode of the NH2 group in the guanidine side-chain and the OH stretching mode of the water molecule. Such OH stretching mode of H2O blue-shifted to 3532 and 3523 cm-1 for R1 and R2 due to the different intensity of the O-H···O interaction. This can be used to distinguish them from the zwitterions. For R1 and R2, a peak from the HOH asymmetrical stretching mode of the H2O has been found at 3818 cm-1, which is blue-shifted ca.100 cm-1 comparing with the IRPD experimental measurement of the hydrated protonated m-aminobenzoic acid.40 This is caused by their different HB types between the two constructed monomers, for which only the oxygen atom of the H2O has involved in the intermolecular interaction of the experimental example. Both of the two canonical isomers display a unique peak at 3315 cm-1, but the origins of them are different. For R1, it is resulted from the OH stretching mode of the carboxylate group, while for R2, it is attributable to the NH stretching mode of the amino group. Similar to Arg and ArgGly,11,

12

although the dispersion interactions are included in the M062X

method, which induced small changes in the conformational distributions of stable conformers in each isomer as listed in Table 2, no effective spectral changes were observed in the simulated IR spectra, as depicted in Figure 4b. The previously reported low-energy conformers C5-1 and C4-1 depicted a strong peak at 2409 cm-1, which is attributable to the OH stretching mode of the carboxylate group. However, due to their high relative Gibbs free energies, little contribution from them has been found to the averaged spectrum, which may lead to wrong conclusions.

3.3. Transition State Calculation Due to the limited search on the PES of Arg-H2O, the structure interconversions among the different canonical and zwitterionic isomers were only qualitatively predicted in the previous work.25 In fact, the arginine monomers in the hydrated clusters C5-1 and C4-1 exhibit only subtle differences comparing with their gaseous naked states.27 The predicted unique IR peak at 2409 cm-1 discussed above is actually corresponding to contributions from the intramolecular HB interaction OH···N between the carboxylate group and the guanidine group, which may make the proton transformation easily. From our study on arginine,11 only small energy barriers were found between such canonical and zwitterionic conformers. Furthermore, the large energy barrier of 6 kcal mol-1 predicted in previous work is too high for transformation at such a low temperature 5 K according to the transition state theory (the reaction rate constant is only 4.02 × 10-252 s-1 in this case).


Page 13 of 24 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 5. (a) Transition states (TS) for the interconversion between the most populated conformer R21 (in isomer R2) and Rz1 (in isomer Rz) at 133 K.40 The relative Gibbs free energies were determined at the CCSD/6-31++G(d, p) level, with the Gibbs free energy corrections at the B3LYP/6-311++G(d, p) level. Here I is an intermediate structure. TS1: R21 to I; TS2: I to Rz1. (b) Simulated IR spectra of the most populated conformers of Arg-H2O at 133 K.40 The conformational distributions are also given in parentheses. A Lorentzian profile with a FWHM of 20 cm-1 is used to convolute the calculated spectra.

In this work, the structure transformation between the canonical and zwitterionic isomers has been accurately investigated. The most populated conformer (R11, R21 and Rz1 of isomers R1, R2 and Rz, respectively), as corrected by the B3LYP method, was selected as the representative structures of each isomer. Here we use the transformation between R21 and Rz1 as an example to explain this problem. The relative Gibbs free energies of the transition states and the corresponding geometric structures are depicted in Figure 5(a). Unlike the gas-phase condition that only one transition state exist,11 due to the adsorption of a water molecule, the proton donor in the carboxylate group will be transformed to the guanidine group through two steps using the protonated amino group as a bridge. Two transition states and one intermediate state, labelled as TS1, TS2 and I, were identified in this process. The calculated energy barriers from the canonical R21 to the zwitterionic Rz1 are 7.37 and 3.81 kcal mol-1 at the advanced CCSD/6-31++G (d, p) level. From the transition state theory, the transition rate of 1 s-1 (indicating one unit of reaction per second) corresponds to an energy barrier of 7.57 kcal mol-1 at 133 K. It means the proton in the carboxylate group has large possibility to be transformed to the guanidine group and the transformation between the canonical and zwitterionic forms is feasible. This also indicates that for Arg-H2O, the proton transformation bridge of the canonical-to-zwitterionic 13 ACS Paragon Plus Environment


Page 14 of 24

process is actually the amino group, rather than the traditional water molecule as in other hydrated molecules.46, 47 For the transformation from R11 to Rz1, only the first energy barrier about 7.31 kcal mol-1 was identified. The constructed intermediate structure and the second transition state structure will directly be transformed to the zwitterion, which overcome almost no energy barrier at the second step. Therefore, conformers in the three neutral isomers are believed to reach their equilibrium eventually at 133 K. The theoretical IR spectra of the most populated neutral Arg-H2O conformers at 133 K are predicted, as shown in Figure 5(b), together with the corresponding distribution of each conformer and the averaged spectrum (SUM). Due to the large structural population (92.94%) of the zwitterionic Rz, the total spectrum is almost the same as that in Figure 4 (the SUM of Rz). The conformational distributions of the two canonical isomers are so small (only occupied 4.13%) that the above analyzed unique features can hardly be observed in the averaged spectrum. We believe that with the increase of the temperature, such peaks will appear gradually and enable people to distinguish the different isomers both in theory and experiment.

4. CONCLUSIONS To summarize, we have conducted a systematic search on the potential energy surface of Arg-H2O, computed their IR spectra from the chemical structure of its conformers, and the structural transformation in the isomerization process, to provide valid theoretical evidence for the canonical-tozwitterionic structure transition of hydrated arginine. Many new low-energy hydrated canonical conformers of Arg-H2O were found by our strategy. The energy difference between the canonical and zwitterionic isomers by the composite CBS-QB3 method is as large as 5.00 kcal mol-1, which provides a way to determine the dominant form of the zwitterions from the energetic point of view. The micro effects of the zero-point energy corrections and the basis set superposition error corrections on the hydrated molecules have been carefully examined. The unique features appeared in the simulated IR spectra, especially the strong resonant ones in the 1400-1800 cm-1 region, can be used to distinguish the three different neutral isomers of Arg-H2O. Two energy barriers have been identified in the canonicalto-zwitterionic process and it is found that the amino group actually acts as a bridge for the proton transformation of this hydrated system. Our study thus shed light on the exploration of the hydration of large flexible molecules and the structural information, the correction effects, the simulated spectra, as well as the predicted energy barriers among different isomers will help researchers in the future to determine the dominant isomer of Arg-H2O both in theory and experiment. 14 ACS Paragon Plus Environment

Page 15 of 24 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


AUTHOR INFORMATION Corresponding Author *E-mail: [email protected], [email protected] Notes The authors declare no competing financial interest.

ACKNOWLEDGMENTS The work is supported by National Natural Science Foundation of China (Grant No. 21603206 and 21473166, Key program 21633006 and 21633007) and Program for Innovative Research Team of Guizhou Province of China (QKTD-[2012]4009). The density functional theoretical calculations were performed at the Supercomputing Center of the University of Science and Technology of China.

REFERENCES (1) Zhong, L.; Hu, Y.; Xing, D.; Zou, H. Mocrosolvation Process of Biomolecules. Prog. Chem. 2010, 22, 1-8. (2) Schmidt, J.; Meyer, M. M.; Spector, I.; Kass, S. R. Infrared Multiphoton Dissociation Spectroscopy Study of Protonated p-Aminobenzoic Acid: Does Electrospray Ionization Afford the Amino- or Carboxy-Protonated Ion? J. Phys. Chem. A 2011, 115, 7625−7632. (3) Tian, Z.; Kass, S. R. Gas-Phase versus Liquid-Phase Structures by Electrospray Ionization Mass Spectrometry. Angew. Chem. Int. Ed. 2009, 48, 1321−1323. (4) Kumler, W. D.; Strait, L. A. The Ultraviolet Absorption Spectra and Resonance in Benzene DerivativesSulfanilamide, Metanilamide, p-Aminobenzoic Acid, Benzenesulfonamide, Benzoic Acid and Aniline. J. Am. Chem. Soc. 1943, 65, 2349−2354. (5) Chang, T. M.; Prell, J. S.; Warrick, E. R.; Williams, E. R. Where’s the Charge? Protonation Sites in Gaseous Ions Change with Hydration. J. Am. Chem. Soc. 2012, 134, 15805−15813. (6) Tian, Z. X.; Wang, X. B.; Wang, L. S.; Kass, S. R. Are Carboxyl Groups the Most Acidic Sites in Amino Acids? Gas-Phase Acidities, Photoelectron Spectra, and Computations on Tyrosine, p-Hydroxybenzoic Acid, and Their Conjugate Bases. J. Am. Chem. Soc. 2009, 131, 1174−1181. (7) Steill, J. D.; Oomens, J. Gas-Phase Deprotonation of p-Hydroxybenzoic Acid Investigated by IR Spectroscopy: Solution-Phase Structure Is Retained upon ESI. J. Am. Chem. Soc. 2009, 131, 13570−13571. (8) Yamdagni, R.; McMahon, T. B.; Kebarle, P. Substituent Effects on the Intrinsic Acidities of Benzoic Acids Determined by Gas Phase Proton Transfer Equilibria Measurements. J. Am. Chem. Soc. 1974, 96, 4035−4037. (9) Wang, C.; Lin, Z.; Zhang, R. Zwitterions Are the Most Stable Form for Neutral Arginylglycine in Gas Phase: Clear Theoretical Evidence. Comput.Theor.Chem. 2013, 1008, 96-102. (10) Li, H.; Hua, W.; Lin, Z.; Luo, Y. First-Principles Study on Core-Level Spectroscopy of Arginine in Gas and Solid Phases. J. Phys. Chem. B 2012, 116, 12641-12650.



Page 16 of 24

(11) Li, H.; Lin, Z.; Luo, Y. The Role of Dimerization on the Structure Transformation of Arginine in Gas Phase. Chem. Phys. Lett. 2014, 608, 398-403. (12) Li, H.; Li, L.; Jiang, J.; Lin, Z.; Luo, Y. Theoretical Spectroscopic Studies on Chemical and Electronic Structures of Arginylglycine. Phys. Chem. Chem. Phys. 2015, 17, 24754-24760. (13) Li, H.; Jiang, J.; Luo, Y. Identification of the Smallest Peptide with A Zwitterion as the Global Minimum: A FirstPrinciples Study on Arginine-Containing Peptides. Phys. Chem. Chem. Phys. 2017, DOI: 10.1039/C7CP10380A. (14) Xu, S.; Nilles, J. M.; Bowen, Jr, K. H. Zwitterion Formation in Hydrated Amino Acids, Dipole Bound Anions: How Many Water Molecules Are Required? J. Chem. Phys. 2003, 119, 10696-10701. (15) Chen, M.; Lin, Z. Ab Initio Studies of Aspartic Acid Conformers in Gas Phase and in Solution. J. Chem. Phys. 2007, 127, 154314. (16) Bachrach, S. M. Microsolvation of Glycine: A DFT Study. J. Phys. Chem. A 2008, 112, 3722-3730. (17) Jockusch, R. A., Lemoff, A. S. and Williams, E. R. Hydration of Valine−Cation Complexes in the Gas Phase: On the Number of Water Molecules Necessary to Form a Zwitterion. J. Phys. Chem. A 2001, 105, 10929-10942. (18) Aikens, C. M.; Gordon, M. S. Incremental Solvation of Nonionized and Zwitterionic Glycine. J. Am. Chem. Soc. 2006, 128, 12835-12850. (19) Rodziewicz, P.; Doltsinis, N. L. Ab Initio Molecular Dynamics Free-Energy Study of Microhydration Effects on the Neutral–Zwitterion Equilibrium of Phenylalanine. ChemPhysChem. 2007, 8, 1959-1968. (20) Blom, M. N.; Compagnon, I.; Polfer, N. C.; Helden, G. v.; Meijer, G.; Suhai, S.; Paizs, B.; Oomens, J. Stepwise Solvation of an Amino Acid: The Appearance of Zwitterionic Structures. J. Phys. Chem. A 2007, 111, 7309-7316. (21) Bachrach, S. M.; Nguyen, T. T.; Demooin, D. W. Microsolvation of Cysteine: A Density Functional Theory Study. J. Phys. Chem. A 2009, 113, 6172-6181. (22) Mullin, J. M. and Gordon, M. S., Mullin, J. M. Alanine: Then There Was Water. J. Phys. Chem. B 2009, 113, 8657-8669. (23) Bountis, T., Ed. Proton Transfer in Hydrogen-Bonded Systems, NANO ASI series, Series B. Physics Vol. 291, Plenum Press: New York, 1992. (24) Christopher, F. R.; Alwin, C.; Tamer, S.; Ivan, K. C.; Alan, C. H. and K. W. Michael, S. Proton Migration and Tautomerism in Protonated Triglycine. J. Am. Chem. Soc. 2001, 123, 3006-3012. (25) Im, S.; Jang, S-W.; Lee, S.; Lee, Y.; Kim, B. Arginine Zwitterion Is More Stable Than the Canonical Form when Solvated by a Water Molecule. J. Phys. Chem. A 2008, 112, 9767-9770. (26) Kim, J. Y.; Ahn, D. S.; Park, S. W.; Lee, S. Gas Phase Hydration of Amino Acids and Dipeptides: Effects on The Relative Stability of Zwitterion vs. Canonical Conformers. RSC Advances 2014, 4, 16352-16361. (27) Ling, S.; Yu, W.; Huang, Z.; Lin, Z.; Haranczyk, M.; Gutowski, M. Gaseous Arginine Conformers and Their Unique Intramolecular Interactions. J. Phys. Chem. A 2006, 110, 12282-12291. (28) Jeffrey, G. A.; Saenger, W. Hydrogen Bonding in Biological Structures, Springer, Berlin, 1991. (29) Jeffrey, G. A. An Introduction to Hydrogen Bonding, Oxford University Press, Oxford, 1997. (30) Rak, J.; Skurski, P.; Simons, J.; Gutowski, M. Low-Energy Tautomers and Conformers of Neutral and Protonated Arginine. J. Am. Chem. Soc. 2001, 123, 11695-11707.


Page 17 of 24 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


(31) Nguyen, D. T.; Scheiner, A. C.; Andzelm, J. W.; Sirois, S.; Salahub, D. R.; Hagler, A. T. A Density Functional Study of the Glycine Molecule: Comparison with Post-Hartree–Fock Calculations and Experiment. J. Comp. Chem. 1997, 18, 1609-1631. (32) Fogarasi, G. High-level Electron Correlation Calculations on Some Tautomers of Cytosine. J. Mol. Struct. 1997, 413-414, 271-278. (33) Zhao, Y.; Thuhlar, D. G. Density Functionals with Broad Applicability in Chemistry. Acc. Chem. Rsc. 2008, 41, 157-167. (34) Hphenstein, E. G.; Chill, S. T.; Sherrill, C. D. Assessment of the Performance of the M05-2X and M06-2X Exchange-Correlation Functionals for Noncovalent Interactions in Biomolecules. J. Chem. Theor. Comput. 2008, 4, 1996-2000. (35) Boys, S. F.; Bernardi, F. The Calculation of Small Molecular Interactions by the Differences of Separate Total Energies. Some Procedures with Reduced Errors. Mol. Phys. 1970, 19, 553-566. (36) Li, H.; Lin, Z.; Luo, Y. Gas-phase IR Spectroscopy of Deprotonated Amino Acids: Global or Local Minima? Chem. Phys. Lett. 2014, 598, 86-90. (37) Li, H. Conformations of Amino Acids Characterized by Theoretical Spectroscopy. Ph. D. dissertation, Royal Institute of Technology, Stockholm, Sweden. 2014, pp41. (38) Truhlar D, Garrett B, Klippenstein S. Current Status of Transition-State Theory. J. Phys. Chem. 1996, 100, 1277112800. (39) 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 C.01; Gaussian Inc.: Wallingford, CT, 2009. (40) Chang, T. M.; Chakrabarty, S.; Willians, E. R. Hydration of Gaseous m‑Aminobenzoic Acid: Ionic vs.Neutral Hydrogen Bonding and Water Bridges. J. Am. Chem. Soc. 2014, 136, 10440−10449. (41) Robert, J. G.; Wim, C.; Theresa. L. W.; Jack. S. Very Large Scale Computations of the Free Energies of Eight Low-Lying Structures of Arginine in the Gas Phase. J. Phys. Chem. A 2003, 107, 7915-7922. (42) Hu, Y.; Bernstein, E. R. Vibrational and Photoionization Spectroscopy of Biomolecules: Aliphatic Amino Acid Structures. J. Chem. Phys. 2008, 128, 164311. (43) Sinclair, W. E.; Pratt, D. W. Structure and Vibrational Dynamics of Aniline and Aniline–Ar From High Resolution Electronic Spectroscopy in the Gas Phase. J. Chem. Phys. 1996, 105, 7942-7956. (44) Oomens, J.; Steill, J. S.; Redlich, B. Gas-Phase IR Spectroscopy of Deprotonated Amino Acids. J. Am. Chem. Soc. 2009, 131, 4310-4319. (45) Chapo, C. J.; Paul, J. B.; Provencal, R. A.; Roth, K.; Saykally, R. J. Is Arginine Zwitterionic or Neutral in the GasPhase? Results from IR Cavity Ringdown Spectroscopy. J. Am. Chem. Soc. 1998, 120, 12956-12957. (46) Kim, J. Y.; Im, S.; Kim, B.; Desfrancois, C.; Lee, S. Structures and Energetics of Gly–(H2O)5: Thermodynamic and Kinetic Stabilities. Chem. Phys. Lett. 2008, 451, 198-203. (47) Yamabe, S.; Ono, N.; Tsuchida, N. Molecular Interactions between Glycine and H2O Affording the Zwitterion. J. Phys. Chem. A 2003, 107, 7915-7922.



TOC Graphic


Page 18 of 24

Page 19 of 24 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 1 19x7mm (300 x 300 DPI)

ACS Paragon Plus Environment




Page 20 of 24

Page 21 of 24 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 4 177x124mm (300 x 300 DPI)


Page 22 of 24

Page 23 of 24 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 Graphic 43x26mm (300 x 300 DPI)


Page 24 of 24

Systematic Study on Hydrated Arginine: Clear Theoretical Evidence

Recommend Documents