Carbohydrate Research 396 (2014) 9–13

Contents lists available at ScienceDirect

Carbohydrate Research

journal homepage: www.elsevier .com/locate /carres

Does intramolecular hydrogen bond play a key rolein the stereochemistry of a- and b-D-glucose?

http://dx.doi.org/10.1016/j.carres.2014.06.0130008-6215/� 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding author. Tel.: +55 35 3829 1891.E-mail address: matheus@dqi.ufla.br (M.P. Freitas).

Josué M. Silla a, Rodrigo A. Cormanich b, Roberto Rittner b, Matheus P. Freitas a,⇑a Department of Chemistry, Federal University of Lavras, PO Box 3037, 37200-000 Lavras, MG, Brazilb Chemistry Institute, State University of Campinas, PO Box 6154, 13083-970 Campinas, SP, Brazil

a r t i c l e i n f o a b s t r a c t

Article history:Received 27 March 2014Received in revised form 19 May 2014Accepted 11 June 2014Available online 19 June 2014

Keywords:D-GlucoseHydrogen bondNBONCIQTAIM

Four a- and three b-isomers of the D-glucose were optimized in gas phase using ab initio (MP2) and DFT(xB97X-D) methods, both using the aug-cc-pVDZ basis set. While earlier works suggest that the orienta-tion of the hydroxyl groups is due to intramolecular hydrogen bonds (H-bonds), the present study revealsthat most H-bonds forming five-membered rings are either weak or even do not exist. The quantum the-ory of atoms in molecules (QTAIM) analysis showed only a few cases of H-bond in D-glucose, particularlyfor those H-bonds forming six-membered rings, while the non-covalent interactions (NCI) analysis indi-cated that most intramolecular H-bonds are not strong enough to justify the counter-clockwise arrange-ment of the OH� � �O chains. Natural bond orbital analysis supported the findings obtained from QTAIMand NCI analyses and indicated that the anomeric effect for D-glucose in the gas phase is governed bya balance of steric, electrostatic, and hyperconjugative interactions.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Carbohydrates are biomolecules widely studied due to numer-ous applications, that range from pharmacological action, forexample, the anticoagulant activity of heparin,1 to technologicalinterest in developing cellulose nanofibers2–4 that can exhibitmechanical properties comparable to carbon nanotubes.5 Suchproperties are related to the molecular architecture, which is gov-erned by intra- and intermolecular interactions, such as hydrogenbond. In addition to carbohydrates, many other compounds andbiological molecules exhibit hydrogen bonding as generally oneof the main stabilizing interactions, generating great interest ofthe scientific community for new studies with emphasis on thistype of interaction, for example, in supramolecular chemistry.

Phenyl-substituted monosaccharides (mannose, galactose, andglucose) and their single hydrated complexes have been studiedin gas-phase by UV and IR double resonance hole burning spectros-copy techniques, revealing that the most stable conformations arethose having a sequence of H-bonds (OH� � �O) surrounding the car-bohydrate.6 Microwave studies in isolated phase (LA-MB-FTMWexperimental technique) and ab initio calculations suggest thatthe most stable conformers of a-and b-2-deoxy-D-ribose are con-trolled by the anomeric effect and H-bond cooperativity.7 Alonsoet al.8 identified four conformers for a-D-glucopyranose and three

for b-D-glucopyranose, indicating the importance of hyperconjuga-tive factors, like those associated with anomeric or gauche effects,as well as the cooperative OH� � �O chains extended along the entiremolecule, as the main factors driving the conformational behavior.However, H-bond forming five-membered rings has not shown tobe stable in a variety of 2-substituted alcohols.10

The presence or not of these OH� � �O interactions, which areassumed to give rise to the counter-clockwise arrangement ofthe network of intramolecular H-bonds in D-glucose, were theoret-ically investigated using natural bond orbital analysis (NBO),11

noncovalent interactions (NCI) method,12 and the quantumtheory of atom in molecules (QTAIM),13 in order to check theimportance of these and other interactions on the conformationalequilibrium of the seven energy minima experimentally foundfor a- and b-D-glucose (Fig. 1).

2. Theoretical calculations

Optimization calculations in the gas phase at the MP214 andDFT/xB97X-D15 methods using the aug-cc-pVDZ16 basis set wereperformed for the most stable rotamers of a- and b-D-glucose(Cartesian coordinates are given in Supplementary material),experimentally determined elsewhere,8 using the Gaussian 09 pro-gram.17 Natural bond orbital (NBO)10 analysis at the xB97X-D/aug-cc-pVDZ level was carried out over the optimized geometries,using the NOSTAR and STERIC keywords as inputs to determinethe energy of the hypothetical Lewis structure and the steric

Figure 1. Hydrogen bonds (dashed lines) and their respective nO(LP(1) + LP(2)) ? r⁄OH interactions (kcal mol�1), according to NBO analysis.

10 J. M. Silla et al. / Carbohydrate Research 396 (2014) 9–13

exchange energy, respectively (minimum energy cut-off of0.5 kcal mol�1). Non-covalent interactions (NCI, using the NCIPLOTprogram18) and quantum theory of atoms in molecules (QTAIM,using the AIMAll program13) calculations were used to analyzethe formation of five- and six-membered rings due to intramolec-ular OH� � �O H-bonds.

3. Results and discussion

Four out 7 stable rotamers of isolated d-glucose found in the gasphase are a anomers (G�g+/cc/t, G+g�/cc/t, Tg+/cc/t and G�g+/cl/g�), while 3 are b anomers (G�g+/cc/t, G+g�/cc/t and Tg+/cc/t).The rotamers were optimized using both MP2 and DFT (xB97X-D) methods and the aug-cc-pVDZ basis set. The relative energies(Erel) are similar for both theoretical levels (Table 1) and, therefore,the remaining calculations were carried out using the xB97X-D/aug-cc-pVDZ level only. In agreement with the anomeric effect,the 4a-D-glucose rotamers are more stable than the b. The hyper-conjugative and electrostatic contributions for this effect havebeen widely discussed elsewhere for 2-substituted tetrahydropy-rans.19,20 The origin of such effect for D-glucose was investigatedusing the G�g+/cc/t conformation (the most stable one) for botha and b anomers, which differ from one another only by the orien-tation of the hydroxyl group bonded to the anomeric carbon (axialin a and equatorial in b). According to the hyperconjugation model,the nO ? r⁄CO electron delocalization stabilizes the a-anomer by18.4 kcal mol�1. This hyperconjugative nature is determined byexamining all possible interactions between ‘filled’ (donor)Lewis-type NBOs and ‘empty’ (acceptor) non-Lewis NBOs, and esti-mating their energetic importance by second-order perturbationtheory. In addition, the contributions from Lewis-type (steric andelectrostatic interactions) and non-Lewis-type (such as hypercon-jugation) interactions to the overall energy of a molecular system

Table 1Full electronica (MP2 and DFT xB97X-D/aug-cc-pVDZ level), steric,b electrostatic,b and hywater (w), for the rotamers of D-glucose

Parameter Rotamers of a-D-glucose

G�g+/cc/t G+g�/cc/t Tg+/cc/t

E(MP2) 0.0 0.2 0.5E(DFT) 0.0/0.0(w) 0.1/0.1(w) 0.1/0.4(w)Esteric 8.2/9.7(w) — —Eelectrostatic 0.0/0.0(w) — —Ehyperconjugation 0.0/0.0(w) — —

a For all rotamers.b For most stable rotamers of a- and b-D-glucose (positive values are destabilizing an

can be estimated, in such a way that the full energy can be decom-posed according to Efull = ELewis + Enon-Lewis, where ELewis = Esteric +Eelectrostatic. According to this decomposition scheme, the a-G�g+/cc/t rotamer is sterically destabilized by 8.2 kcal mol�1 relative tob-G�g+/cc/t, but the latter is electrostatically destabilized by11.7 kcal mol�1 relative to a-G�g+/cc/t. Overall, Lewis-typeinteractions favor the a-anomer by 3.5 kcal mol�1, while non-Lewis-type interactions favor the b-anomer by 2.6 kcal mol�1

(despite the individual nO ? r⁄CO interaction in favoring thea-anomer by 18.4 kcal mol�1). Thus, in the gas phase, the anomericeffect for D-glucose is mainly described by electrostatic (dipolar)interactions, but they are competitive with hyperconjugation, inagreement with the results for 2-OH-tetrahydropyran obtainedelsewhere.19

The anomeric effect operates in D-glucose in the isolated state,while the preferred counter-clockwise arrangement for thehydroxyl groups has been claimed to be due to a network ofintramolecular H-bonds.8 Actually, the a-G�g+/cc/t and a-G�g+/cl/g� conformations have the same orientation for the CH2OHhydroxyl group, but the O–H� � �O arrangements in the otherhydroxyl groups are counter-clockwise (‘cc’) in the former andclockwise (‘cl’) in the latter (less stable). Thus, the preferencefor the a-G�g+/cc/t and all other ‘cc’ conformations over thea-G�g+/cl/g� arrangement is expected to be due to the stericrelief involving the lone pairs of electrons in the endocyclicoxygen (O5) and the anomeric oxygen (O1) in ‘cc’ rather thanintramolecular H-bonds. This is supported by the premise thatH-bond forming four-membered rings in O1–H� � �O5 is not stable(is not formed) and, consequently, this should not dictate the‘cc’ or ‘cl’ arrangements along with the other O–H� � �O arrange-ments in D-glucose. In order to search for the rule of intramolec-ular H-bonds in the stereochemistry of D-glucose, particularly theorientation of the O–H� � �O arrangements, 3 widely used

perconjugationb energies (xB97X-D/aug-cc-pVDZ level) in the gas phase and implicit

Rotamers of b-D-glucose

G�g+/cl/g� G�g+/cc/t G+g�/cc/t Tg+/cc/t

0.9 1.3 1.5 1.90.6/1.5(w) 0.9/0.7(w) 1.1/0.8(w) 1.2/1.1(w)— 0.0/0.0(w) — —— 11.7/17.9(w) — —— �2.6/�7.5(w) — —

d negative ones are stabilizing interactions).

J. M. Silla et al. / Carbohydrate Research 396 (2014) 9–13 11

computational tools to identify H-bonds were applied to studythe seven main conformations of D-glucose in the gas phase.

The non-covalent interactions (NCI) method introduced byJohnson et al.10 can serve as a complement to the Bader’s theory(QTAIM)21,22 to characterize, for example, weak H-bonds in 3Dspace of five-membered rings, since there has been controversialcases in the literature in which a bond critical point (BCP) cannotbe found for compounds where a H-bond is expected to be formed,such as in the QTAIM analysis of 2-fluorophenol23 and 3-fluoro-1,2-propanediol.9 The NCI method is based on the analysis of theelectron densities q(r) and their reduced gradients, s(r). In thissense, the non-covalent interactions will be predicted in regionswhere the q(r) and s(r) are low. Attractive interactions (blue), suchas H-bonds, Van der Waals contact (green) and repulsive interac-tions/electron density depletion regions (red) can be visualizedon isosurfaces of s(r), as shown for the energy minima of D-glucosein Figure 2. Here, the attractive and repulsive interactions aremixed in the region between the (H)O� � �H contacts. However, theblue volume corresponding to H-bonding appears when formingsix-membered rings, but the corresponding region in five-mem-bered rings is modest. The red volume appears as a consequenceof ring formation on the five- and six-membered rings, beingrelated to the QTAIM ring critical points (RCPs). The NCI isosurfacesfor five-membered rings in Figure 2 suggest very weak intramolec-ular H-bonds, since such isosurfaces are single volumes encom-passing both the H-bond isosurface (blue color) and itscorresponding RCP isosurface (red color), that cannot bedetected by other theoretical methods (vide infra). Indeed, such

Figure 2. NCI isosurfaces indicating some intramolecular non-covalent interactions in a-colors scaling from �0.02 au <sign(k2)q <+0.02 au. Peaks at negative sign(k2)q values indH-bonds and peaks at positive sign(k2)q values indicate ring critical point regions. (For inthe web version of this article.)

five-membered rings are similar fragments of the 1,2-butanediolmolecule, which was previously studied by Lane et al.,24 by theNCI method and despite not detected by the QTAIM, they werecharacterized as weak H-bonds having the same nature as H-bondsin 1,3-butanediol and 1,4-butanediol. Accordingly, the six-membered H-bonds are similar to a 1,3-propanediol fragment,which forms a stronger H-bond in the NCI point of view, sincethe H-bond blue isosurfaces and RCP red isosurfaces are notcompletely mixed up as for the five-membered rings and suggeststronger and more stable H-bonds.

The NCI results are consistent with the NBO findings, whichdescribe H-bonds as nO ? r⁄OH interactions. Indeed, importantelectron delocalizations from oxygen lone pairs to antibondingOH orbitals are found mostly for H-bonds forming six-memberedrings (involving CH2OH as proton donor and O4 as proton acceptor,e.g., the Tg+/cc/t conformations in Figs. 1 and 3). H-bonds formingfive-membered rings are weaker or even not detected by the NBOanalysis. The QTAIM analysis shows that a BCP is found onlybetween CH2O-H� � �O4 in the Tg+/cc/t conformation of both a andb anomers (Fig. 4), while H-bonds forming five-membered ringsare not observed at all, in agreement with the study of Lane et al.24

In water solution, where the stereochemistry of carbohydratesproves to be more interesting, intramolecular H-bonds areexpected to play a still weaker role. The seven main conformersof D-glucose were further optimized using implicit water accordingto the Polarizable Continuum Model.25 Even not consideringspecific solute–solvent interactions (which would compete withintramolecular interactions), the calculations indicate that the

and b-D-glucose. The isosurfaces were constructed with RGD = 0.5 au and blue–redicate attractive non-covalent interactions corresponding to five- and six-memberedterpretation of the references to colour in this figure legend, the reader is referred to

Figure 4. Molecular graphs obtained by AIM for the rotamers of a- and b-D-glucose in the gas phase.

Figure 3. Molecular graphics and NBO interactions indicate the formation of six membered rings.

12 J. M. Silla et al. / Carbohydrate Research 396 (2014) 9–13

conformational preferences are not substantially different fromthose obtained for the gas phase (Table 1), while NBO, QTAIM,and NCI exploration (Supplementary material) indicates that noadditional intramolecular H-bond appears in implicit water.2-Hydroxytetrahydropyran forms intermolecular H-bond withexplicit water molecule;19 the same is expected for D-glucose,which would compete with the six-membered ring-formingintramolecular H-bond to make it weaker or even absent.

4. Conclusions

Intramolecular hydrogen bonds are not responsible for thecounter-clockwise arrangement of hydroxyl groups in D-glucose.The repulsion between the lone pairs of electrons in the endocy-clic oxygen with oxygen bonded to the anomeric carbon dictatesthe preferred conformations of this compound, since this interac-tion induces the orientation of the remaining hydroxyl groups.These findings are instructive because they suggest that derivati-zation of D-glucose through replacement of hydroxyl by ORgroups, which should not exhibit H-bond, can lead to counter-clockwise arrangement, such as the prevalent rotamers ofD-glucose.

Acknowledgments

Authors thank FAPEMIG (Grant # CEX-PPM-00033-13) andFAPESP (Grant #2012/03933-5) for the financial support of thisresearch. CAPES and FAPESP are also gratefully acknowledged forthe studentships (to J.M.S. and R.A.C. #2011/01170-1, FAPESP), asis CNPq for the fellowships (M.P.F. and R.R.).

Supplementary data

Supplementary data associated with this article can be found,in the online version, at http://dx.doi.org/10.1016/j.carres.2014.06.013.

References

1. Green, D.; Hirsh, J.; Heit, J.; Prins, M.; Davidson, B.; Lensing, A. W. A. Pharmacol.Rev. 1994, 46, 89–109.

2. Kemmel, M.; Pore, V.; Ritala, M.; Leskelä, M.; Lindén, M. J. Am. Chem. Soc. 2005,127, 14178–14179.

3. Leskelä, M.; Kemmel, M.; Kukli, K.; Pore, V.; Santala, E.; Ritala, M.; Lu, J. Mater.Sci. Eng. C 2007, 27, 1504–1508.

4. Fernandes, A. N.; Thomas, L. H.; Altaner, C. M.; Callow, P.; Forssyth, V. T.;Apperley, D. C.; Kennedy, C. J.; Jarvis, M. C. Proc. Natl. Acad. Sci. U.S.A. 2011, 108,E1195–E1203.

5. Momeni, K.; Yassar, R. S. J. Comput. Theor. Nanosci. 2009, 6, 1511–1518.6. Çarbaçal, P.; Jockusch, R. A.; Hünig, I.; Snoek, L. C.; Kroemer, R. T.; Davis, B. G.;

Gamblin, D. P.; Compagnon, I.; Oomens, J.; Simons, J. P. J. Am. Chem. Soc. 2005,127, 11414–11425.

7. Peña, I.; Cocinero, E. J.; Cabezas, C.; Lezarri, A.; Mata, S.; Écija, P.; Daly, A. M.;Cimas, A.; Bermúdez, C.; Basterretxea, F. J.; Blanco, S.; Fernández, J. A.; López, J.C.; Castaño, F.; Alonso, J. L. Angew. Chem., Int. Ed. 2013, 52, 11840–11845.

8. Alonso, J. L.; Lozoya, M. A.; Peña, I.; López, J. C.; Cabezas, C.; Mata, S.; Blanco, S.Chem. Sci. 2014, 5, 515–522.

9. Andrade, L. A. F.; Silla, J. M.; Duarte, C. J.; Rittner, R.; Freitas, M. P. Org. Biomol.Chem. 2013, 11, 6766–6771.

10. Cormanich, R. A.; Freitas, M. P.; Tormena, C. F.; Rittner, R. RSC Adv. 2012, 2,4169–4174.

11. NBO 6.0. Glendening, E. D.; Badenhoop, J. K.; Reed, A. E.; Carpenter, J. E.;Bohmann, J. A.; Morales, C. M.; Landis, C. R.; Weinhold, F. (TheoreticalChemistry Institute, University of Wisconsin, Madison, WI, 2013); http://nbo6.chem.wisc.edu/.

12. Johnson, E. R.; Keinan, S.; Sánchez, P. M.; Contreras-García, J.; Cohen, A. J.; Yang,W. J. Am. Chem. Soc. 2010, 132, 6498–6506.

13. AIMAll (Version 11.10.16), Todd A. Keith, TK Gristmill Software, Overland ParkKS, USA, 2011, aim.tkgristmill.com.

J. M. Silla et al. / Carbohydrate Research 396 (2014) 9–13 13

14. Head-Gordon, M.; Pople, J. A.; Frisch, M. J. Chem. Phys. Lett. 1988, 153, 503–506.15. Chai, J.-D.; Head-Gordon, M. Phys. Chem. Chem. Phys. 2008, 10, 6615–6620.16. Woon, D. E.; Dunning, T. H. J. Chem. Phys. 1993, 100, 2975–2988.17. 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.;Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.;Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven,T., ; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.;Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.;Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.;Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.;Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.;Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.;Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas,

O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, RevisionA.02; Gaussian: Wallingford, CT, 2009.

18. Contreras-García, J.; Johnson, E. R.; Keinan, S.; Chaudret, R.; Piquemal, J.-P.;Beratan, D. N.; Yang, W. J. Chem. Theory Comput. 2011, 7, 625–632.

19. Freitas, M. Org. Biomol. Chem. 2013, 11, 2885–2890.20. Mo, Y. Nat. Chem. 2010, 2, 666–671.21. Bader, R. F. W. Atoms in Molecules: A Quantum Theory; Clarendon: Oxford, 1990.22. Bader, R. F. W. Chem. Rev. 1991, 91, 893–928.23. Cormanich, R. A.; Moreira, M. A.; Freitas, M. P.; Ramalho, T. C.; Anconi, C. P. A.;

Rittner, R.; Contreras, R. H.; Tormena, C. F. Magn. Reson. Chem. 2011, 49,763–767.

24. Lane, J. R.; Contreras-García, J.; Piquemal, J.-P.; Miller, B. J.; Kjaergaard, H. G. J.Chem. Theory Comput. 2013, 9, 3263–3266.

25. Tomasi, J.; Mennucci, B.; Cammi, R. Chem. Rev. 2005, 105, 2999–3093.