Article | Open

Isotope effect on the circular dichroism spectrum of methyl α-D-glucopyranoside in aqueous solution

  • Scientific Reports 5, Article number: 17900 (2015)
  • doi:10.1038/srep17900
  • Download Citation
Received:
Accepted:
Published online:

Abstract

H/D isotope effect on the circular dichroism spectrum of methyl α-D-glucopyranoside in aqueous solution has been analyzed by multicomponent density functional theory calculations using the polarizable continuum model. By comparing the computational spectra with the corresponding experimental spectrum obtained with a vacuum-ultraviolet circular dichroism spectrophotometer, it was demonstrated that the isotope effect provides insights not only into the isotopic difference of the intramolecular interactions of the solutes, but also into that of the solute–solvent intermolecular interaction.

Introduction

Circular dichroism (CD) is one of the most utilized spectroscopic properties for the assignment of the biomolecule structure. Following the recent development of spectroscopic and computational techniques, the combination of measurement and the computation of CD spectra has become an increasingly important approach for the detailed elucidation of molecular structure. Beside proteins and nucleic acids, CD has recently been applied for structural analysis of carbohydrates using vacuum-ultraviolet circular dichroism (VUVCD) spectrophotometer1,2,3,4,5. For example, Matsuo et al. compared the experimental CD spectrum of methyl α-D-glucopyranoside (methyl α-D-Glc, Fig. 1) in aqueous solution with the spectrum obtained by quantum mechanical (QM) calculations based on time-dependent density-functional theory (TD-DFT) with the polarizable continuum model (PCM)6,7 for the solvent effect and molecular dynamics (MD) simulation for conformational sampling3. Their pioneering analysis demonstrated that the static computation of CD spectrum for three representative equilibrium geometries resulted in a substantial difference between the experimental and computational spectra that was then successfully reduced by the use of broad conformational sampling for the accumulation of the spectrum. Matsuo et al. revealed that the fluctuation of the intramolecular hydrogen bond orientation strongly influenced the methyl α-D-Glc CD spectrum.

Figure 1: Methyl-α-D-glucopyranoside.
Figure 1

As seen in the above case, the fluctuation of hydrogen atom in hydrogen bonding systems can have a dominant effect on the molecular properties of hydrogen-bonded systems. Such systems are often associated with a significant change in properties on deuteration, i.e., the H/D isotope effect that arises from the isotopic difference of a quantum mechanical nature8. The isotope effect is known to slightly modulate the molecular geometry of hydrogen bonded systems, and it sometimes results in a drastic change of the phase-transition temperature9, chemical reaction rate10, and the nuclear magnetic resonance and infrared absorption spectra11. These isotopic difference effects can be used to obtain further detailed information on the behavior of molecules11. Although scarcely explored so far, a significant isotope effect can be expected for the CD spectrum of methyl α-D-Glc because of its dependence on the conformational fluctuations shown by Matsuo et al.

The H/D isotope effect cannot be predicted by the simple application of the conventional electronic structure calculation and classical MD simulation that ignore the quantum nature of the hydrogen nuclei. As an alternative choice for computing the molecular spectroscopic properties that efficiently considers the significant fluctuation of the hydrogen nuclei in hydrogen bonds and the concomitant H/D isotope effect, Tachikawa et al. have developed a multicomponent (MC) scheme for the feasible extension of the QM calculation12. Quantum mechanical calculations using this scheme (MC_QM) incorporate the quantum deviation of nuclei from equilibrium geometry, i.e., the nuclear quantum effect, into the molecular property calculation with only slight additional cost13. This approach has been recently extended to the combination with the PCM [MC_QM/PCM] for the analysis of the condensed phase14.

Results and Discussion

We have applied the MC_QM/PCM approach for the calculation of the CD spectrum of methyl α-D-Glc in aqueous solution as a tentative probe into the isotope effect on CD, with the calculated spectra shown in Fig. 2(a). We focused on methyl α-D-Glc (C7H14O6) and its isotopologue (C7H10D4O6) with deuterated hydroxyl groups. The shape of the obtained spectra agreed with that of the previous study3, demonstrating that the present approach is as reasonable as the combination of MD sampling and conventional QM calculation for modeling the fluctuating methyl α-D-Glc in aqueous solution. The isotopic difference of the spectra in Fig. 2(a) shows that the deuteration of the solute will lead to only a tiny shift. Thus, we predict that H/D isotope effect on the CD spectrum can be marginally detectable.

Figure 2: CD spectra of the methyl α-D-Glc (C7H14O6, red) in H2O and its isotopologue (C7H10D4O6, dashed black) in D2O of (a) initial computational prediction, (b) experimental measurement, and (c) revised computation with the scaling of the solvation radius by 1.04.
Figure 2

To verify the computational prediction of the isotope effect, we measured the CD spectra of the isotopologues of methyl α-D-Glc by using a vacuum-ultraviolet circular dichroism (VUVCD) spectrophotometer1,2,3. Deuteration was introduced by the substitution of light water solvent with heavy one. Fig. 2(b) shows the observed CD spectra of methyl α-D-Glc in H2O and D2O. Contrary to computational prediction, a significant blue shift and an increase in the peak intensity were observed on deuteration.

What is the origin of the discrepancy between the predicted and the observed isotope shifts? To answer this question we focused on the isotopic difference of the solute-solvent interaction that was not considered in the above calculation. It is well known that the deuteration of hydrogen-bonding clusters involves the elongation of hydrogen-bonding distances; this is the so-called Ubbelohde effect11. Similar isotope effects have also been reported for liquid water, in which the length of the hydrogen-bond in light water was evaluated to be ~4% shorter than that in heavy water15. It can be expected that the interaction distances between the hydrophilic solutes and solvent water molecules undergo similar changes with isotopic substitution.

Therefore, we recalculated the CD of methyl α-D-Glc in D2O considering the isotope effect on the intermolecular interaction with the deuteration of the solute hydroxyl groups. To describe the elongation of the intermolecular distance on deuteration and the corresponding slight weakening of the solute-solvent interaction, the solvation radius used in PCM was scaled by factors of around 1.04 for D2O and alienated from the solute. The results of these calculations are shown in Fig. 2(c) and Supplementary Figure S1. The relevant isotopic difference of the OH-bond length, cavity volume, and the solvation energy are shown in Supplementary Table. We can see that the computational isotope effect in Fig. 2(c) was significantly enhanced to achieve better agreement with the experimental isotope shift shown in Fig. 2(b), which was brought by only a slight alienation of the solvation surface and the relevant change of the geometry and the energy. This result indicates that the obtained isotopic shift strongly depends on the isotopic difference of the solute-solvent interaction. Consequently, the isotope effect on the CD spectrum can offer insight into not only the solute conformations but also the carbohydrate hydration. Such insight is not available by other spectroscopic techniques such as IR and NMR.

Summary

To summarize, we have analyzed the isotope effect on the CD spectrum of methyl α-D-Glc in aqueous solution experimentally and theoretically. We have observed significant isotopic differences for the peak position and the intensity in spectra. The modification of the solvation surface was essential for reproducing the observed isotope effect on the CD spectrum by MC_QM/PCM calculation, indicating that the isotopic differences are strongly dependent on the solute-solvent interaction. The present results suggest that the isotope effect on CD spectra carries the information about the conformation of the hydroxyl groups and water molecules along the solvation surface; this will provide new insights into the activity of biomolecules including saccharides.

Method

VUVCD Measurement

Methyl α-D-Glc of high purity (>98%) was purchased from Sigma-Aldrich (St. Louis, MO) and used without further purification. The sample solutions were freshly prepared by dissolving in H2O or D2O at a concentration of 10.0 (w/v%). The obtained sample solutions were incubated at room temperature for 1 day prior to performing VUVCD measurements.

The VUVCD spectra of methyl α-D-Glc in H2O or D2O were measured in the 168–210 nm wavelength range at 25 °C using a VUVCD spectrophotometer at the Hiroshima Synchrotron Radiation Center. A detailed description of the spectrophotometer optical devices is available elsewhere16,17. The VUVCD measurement was performed using an assembled-type optical cell with CaF2 windows17. The path length of the cell was adjusted with a Teflon spacer to 50 μm for the measurements from 200 to 180 nm, and to reduce the effects of light absorption by the solvent, no spacer was used for the measurements below 180 nm. The spectra obtained without the spacer were calibrated by normalizing the ellipticities to the spectra measured using a 50 μm spacer in the overlapping wavelength region from 180 to 200 nm. The spectrum was recorded with a 1.0 mm slit, an 8 s time constant, and an 8 nm/min scan speed and by using four accumulations. All spectra were smoothed with the Savitsky–Golay filter. The molar ellipticity [θ] was calculated using the molecular weight of the solute. The ellipticity was reproducible within an error of ±5%, which can be mainly attributed to noise and the inaccuracy in the light path length.

Conformational search

We have performed conformational search using CONFLEX18,19. After 457 precursory conformers were obtained by CONFLEX with MMFF94s force field and the GB/SA solvation model, 180 unique conformers optimized by CAM-B3LYP/6-31G(d) with the SMD variant of the PCM solvation model20 were obtained from the precursors. The obtained structures included 71 GG, 61 GT, and 48 TG hydroxymethyl rotamers of methyl α-D-Glc. TD-CAM-B3LYP/6-311++G(2d,2p)/SMD calculation was applied to obtain the rotatory strength of 40 GG, 40 GT, and 25 TG relatively stable conformers according to the relative energies evaluated by CAM-B3LYP/6-311++G(2d,2p)/SMD calculations for the conformations. All quantum mechanical calculations were performed with the Gaussian 09 package21 in which we have implemented the multicomponent scheme calculation.

Computational CD spectrum of representative GG and GT hydroxymethyl rotamers

We performed the geometry optimization and rotatory strength calculation of methyl α-D-Glc conformers using CAM-B3LYP with the multicomponent scheme (MC_CAM-B3LYP). In MC_CAM-B3LYP calculations, all hydrogen nuclei were treated quantum mechanically. Two isotopologues were investigated; the first isotopologue had only protons as the hydrogen nuclei and four OH groups of the molecule were replaced with the deuterons for the other isotopologue. We used 6-31G(d) electronic basis set for geometry optimization and 6-311++G(2d,2p) basis set for rotatory strength calculations. For both types of calculations, [1s] GTF calculations were used with 24.1825 a.u. exponent value for the proton and 35.6214 a.u. for deuteron optimized for the quantum treatment of hydrogen nuclei in the MC_HF scheme22. Electronic excitation calculation based on TD-DFT with CAM-B3LYP functional was performed to obtain the rotatory strength of the optimized geometry, where the occupied and virtual Kohn–Sham orbitals were evaluated with the multicomponent scheme and the excitation of the hydrogen nuclei was neglected. 30 rotatory strengths were calculated for each conformer to construct the CD spectrum according to the equation:

where is the excitation energy, is the corresponding electronic rotatory strength, and is the bandwidth of each signal that was set to 0.30 eV.

Average CD spectrum among the rotamers GG, GT, and TG hydroxymethyl rotamers have four OH groups that can rotate to give rise to different hydroxyl rotamers. To consider rotation, average CD spectra were constructed on the basis of the Boltzmann distribution of each OH rotamer according to the equation:

where Eis are the relative energies without the entropic contributions and the temperature in β is set to 298 K. The relative populations of GG, GT, and TG hydroxymethyl rotamers in the above averaging were [GG:GT:TG = 0.45:0.45:0.10], which is in reasonable agreement with the previously reported population of [GG:GT:TG = 0.48:0.48:0.04]23.

Additional Information

How to cite this article: Kanematsu, Y. et al. Isotope effect on the circular dichroism spectrum of methyl α-D-glucopyranoside in aqueous solution. Sci. Rep. 5, 17900; doi: 10.1038/srep17900 (2015).

References

  1. 1.

    & Vacuum-ultraviolet circular dichroism study of saccharides by synchrotron radiation spectrophotometry. Carbohydr. Res. 339, 591–597 (2004).

  2. 2.

    , , & Membrane-induced conformational change of alpha1-acid glycoprotein characterized by vacuum-ultraviolet circular dichroism spectroscopy. Biochemistry 48, 9103–9111 (2009).

  3. 3.

    , , & Vacuum-ultraviolet electronic circular dichroism study of methyl α-D-glucopyranoside in aqueous solution by time-dependent density functional theory. J. Phys. Chem. A 116, 9996–10003 (2012).

  4. 4.

    , , & Z-DNA: vacuum ultraviolet circular dichroism. Proc. Natl. Acad. Sci. USA 78, 4801–4804 (1981).

  5. 5.

    , , & Vacuum ultraviolet circular dichroism as an indicator of helical handedness in nucleic acids. Nucleic Acids Res. 14, 6649–6659 (1986).

  6. 6.

    , , & New developments in the polarizable continuum model for quantum mechanical and classical calculations on molecules in solution. J. Chem. Phys. 117, 43–54 (2002).

  7. 7.

    et al. Geometries and properties of excited states in the gas phase and in solution: theory and application of a time-dependent density functional theory polarizable continuum model. J. Chem. Phys. 124, 94107 (2006).

  8. 8.

    & Geometrical H/D isotope effect on hydrogen bonds in charged water clusters. J. Am. Chem. Soc. 127, 11908–11909 (2005).

  9. 9.

    et al. Geometric effects of deuteration on hydrogen-ordering phase transitions. Nature 348, 317–319 (1990).

  10. 10.

    & Carbon dioxide hydration activity of carbonic anhydrase in mixtures of water and deuterium oxide. Biochemistry 19, 4984–4989 (1980).

  11. 11.

    In Isotope Effects In Chemistry and Biology. (eds. & ) Ch. 5, 153–174 (CRC Press 2005).

  12. 12.

    , , & An extension of ab initio molecular orbital theory to nuclear motion. Chem. Phys. Lett. 290, 437–442 (1998).

  13. 13.

    , & Theoretical study of H/D isotope effects on nuclear magnetic shieldings using an ab initio multi-component molecular orbital method. Molecules 18, 5209–5220 (2013).

  14. 14.

    & Development of multicomponent hybrid density functional theory with polarizable continuum model for the analysis of nuclear quantum effect and solvent effect on NMR chemical shift. J. Chem. Phys. 140, 164111 (2014).

  15. 15.

    & Quantum differences between heavy and light water. Phys. Rev. Lett. 101, 065502 (2008).

  16. 16.

    et al. Vacuum-ultraviolet circular dichroism spectrophotometer using synchrotron radiation: optical system and on-line performance. Chem. Lett. 522–523 (2001).

  17. 17.

    , , , & Optical cell with a temperature-control unit for a vacuum-ultraviolet circular dichroism spectrophotometer. Anal. Sci. 19, 129–132 (2003).

  18. 18.

    & Corner flapping: a simple and fast algorithm for exhaustive generation of ring conformations. J. Am. Chem. Soc. 111, 8950–8951 (1989).

  19. 19.

    , , , & Origin of regioselectivity in the O-methylation of erythromycin as elucidated with the aid of computational conformational space search. J. Chem. Soc. Perkin Trans. 2 1647 (1993).

  20. 20.

    , & Universal solvation model based on solute electron density and on a continuum model of the solvent defined by the bulk dielectric constant and atomic surface tensions. J. Phys. Chem. B 113, 6378–6396 (2009).

  21. 21.

    et al. Gaussian09 Rev. C01 (2009).

  22. 22.

    , & Analysis of exponent values in Gaussian-type functions for development of protonic and deuteronic basis functions. Int. J. Quantum Chem. 106, 1465–1476 (2006).

  23. 23.

    & Effect of solvation on the rotation of hydroxymethyl groups in carbohydrates. J. Am. Chem. Soc. 120, 10953–10963 (1998).

Download references

Acknowledgements

We would like to thank Dr. Takumi Yamaguchi (Institute for Molecular Science) and Dr. Norio Yoshida (Kyushu University) for useful discussion. We also thank Professors Hirofumi Namatame and Masaki Taniguchi for the use of the VUVCD spectrophotometer at the Hiroshima Synchrotron Radiation Center. This study was partly supported by a JSPS/MEXT KAKENHI Grant-in-Aid for Scientific Research on Innovation Areas (25102001, 25102008, 26102518, and 26102539) and Grant-in-Aid for Challenging Exploratory Research (26560451 and 26620013). Theoretical calculations were partly performed at the Research Center for Computational Science, Institute for Molecular Science, Japan and Center of Computational Materials Science, Institute for Solid State Physics, The University of Tokyo, Japan.

Author information

Affiliations

  1. Quantum Chemistry Division, Yokohama City University, Seto 22-2, Kanazawa-ku, Yokohama 236-0027, Japan

    • Yusuke Kanematsu
    •  & Masanori Tachikawa
  2. Institute of Materials and Systems for Sustainability, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8603, Japan

    • Yukiko Kamiya
  3. Hiroshima Synchrotron Radiation Center, Hiroshima University, Higashi-Hiroshima 739-0046, Japan

    • Koichi Matsuo
    •  & Kunihiko Gekko
  4. Institute for Molecular Science and Okazaki Institute for Integrative Bioscience, National Institutes of Natural Sciences, 5-1 Higashiyama Myodaiji, Okazaki, Aichi 444-8787, Japan

    • Koichi Kato

Authors

  1. Search for Yusuke Kanematsu in:

  2. Search for Yukiko Kamiya in:

  3. Search for Koichi Matsuo in:

  4. Search for Kunihiko Gekko in:

  5. Search for Koichi Kato in:

  6. Search for Masanori Tachikawa in:

Contributions

Y.K.a performed the computational analysis. Y.K.b and K.K. conceived and designed the experiments. Y.K.b, K.M. and K.G. performed the experimental measurement. Y.K.a wrote the manuscript and prepared Figure 2(a–c). Y.K.b prepared Figure 2(b). K.K., K.M., K.G. and M.T. edited the manuscript. All authors reviewed and approved the manuscript.

Competing interests

The authors declare no competing financial interests.

Corresponding authors

Correspondence to Koichi Kato or Masanori Tachikawa.

Supplementary information

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Creative Commons BYThis work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/