PH-Dependent mutarotation of 1-thioaldoses in water. Unexpected behavior of (2 S)-d-aldopyranoses

Article abstract of DOI:10.1021/jo100826e

Figure presented. The pH-dependent mutarotation of 1-thioaldopyranoses in aqueous media has been investigated. Anomerization readily occurred at lower and neutral pH for all aldopyranoses studied, whereas mainly for (2S)-d-aldopyranoses at higher pH. 1-Thio-d-mannopyranose and 1-thio-d-altropyranose showed very strong pH dependence where the anomeric equilibrium ratios changed dramatically from a preference for the β-anomer at lower pH to the α-anomer at higher pH.

Full text of DOI:10.1021/jo100826e

pubs.acs.org/joc
pH-Dependent Mutarotation of 1-Thioaldoses in Water. Unexpected
Behavior of (2S)-D-Aldopyranoses
‡
Remi Caraballo, Lingquan Deng, Luis Amorim, Tore Brinck, and Olof Ramstrom*
‡
ꢀ
€
KTH-Royal Institute of Technology, Department of Chemistry, Teknikringen 30,
S-10044 Stockholm, Sweden
ramstrom@kth.se
Received April 30, 2010
The pH-dependent mutarotation of 1-thioaldopyranoses in aqueous media has been investigated.
Anomerization readily occurred at lower and neutral pH for all aldopyranoses studied, whereas
mainly for (2S)-D-aldopyranoses at higher pH. 1-Thio-D-mannopyranose and 1-thio-D-altropyra-
nose showed very strong pH dependence where the anomeric equilibrium ratios changed dramati-
cally from a preference for the β-anomer at lower pH to the R-anomer at higher pH.
Introduction
In aqueous media, the mutarotation process of aldoses has
been extensively explored, resulting in mixtures of the dif-
ferent R- and β-pyranoses and furanoses in equilibrium with
the corresponding open-chain aldehyde and hydrate.1,3,6,11,12
The aldose stereochemistry can also dramatically affect the
anomeric composition at equilibrium, and therefore, a single
configurational change can yield the reverse anomeric pre-
ference for a given structure, as, for example, in the case of
the epimerization of D-glucose (β-preference) to D-mannose
(R-preference).1,6 In addition, replacement of the endocyclic
oxygen and/or the anomeric hydroxyl oxygen by another
heteroatom also leads to mutarotation.1,6,13 However, for
sulfur analogues, existing studies on mutarotation are
mainly confined to structures with an endocyclic sulfur at
the C-5 position.1,3,11,14 Mutarotation of 1-thioaldoses, on
the other hand, has not been thoroughly investigated,11,15-17
and it is still relatively unclear under what conditions glycosyl
Mutarotation of carbohydrate anomers is a fundamental
phenomenon in chemistry1-6 and of particular importance
in synthesis when stereochemically pure products are tar-
geted. The different anomers frequently possess distinct
biological activities,7-9 but similar physicochemical proper-
ties that can hamper the isolation of each stereoisomer.
Although mutarotation has been studied for many decades,
it still remains difficult to predict. Multiple factors, such as
steric hindrance, electronic properties, solvent effects, and
experimental conditions, have to be considered. Thus, the
anomerization of aldoses most likely arises from a combina-
tion of these different factors.10
‡ These authors contributed equally to this work.
(1) Sinnott, M. L. Carbohydrate Chemistry and Biochemistry: Structure
and Mechanism; RSC Publishing: Cambridge, 2007.
(2) Lindhorst, T. K. Essentials of Carbohydrate Chemistry and Biochem-
istry, 3rd ed.; Wiley-VCH: Chichester, 2007.
(3) Davis, B. G.; Fairbanks, A. J. Carbohydrate Chemistry; Oxford
University Press: Cary, NC, 2002.
(4) Ferrier, R. J.; Collins, P. M. Monosaccharide Chemistry; Penguin:
Harmondsworth, 1972.
(5) Capon, B. Chem. Rev. 1969, 69, 407–498.
(6) Pigman, W.; Isbell, H. S. Adv. Carbohydr. Chem. Biochem. 1968, 23, 11–57.
(7) Ishihara, H.; Miwa, I.; Nagano, H.; Okuda, J. Experientia 1979, 35,
470–472.
(8) Wu, J. H.; Wu, A. M.; Yang, Z.; Chen, Y.-Y.; Singha, B.; Chow, L.-P-;
Lin, J.-Y. Biochimie 2010, 147–156.
(11) Isbell, H. S.; Pigman, W. Adv. Carbohydr. Chem. Biochem. 1969, 24,
13–65.
(12) Silva, A. M.; Da Silva, E. C.; Da Silva, C. O. Carbohydr. Res. 2006,
341, 1029–1040.
(13) Raymond, A. L. J. Biol. Chem. 1934, 107, 85–96.
(14) Grimshaw, C. E.; Whistler, R. L.; Cleland, W. W. J. Am. Chem. Soc.
1979, 101, 1521–1532.
(15) Nakamura, H.; Tejima, S.; Akagi, M. Chem. Pharm. Bull. 1966, 14,
648–657.
(16) Schneider, W.; Leonhardt, H. Ber. Dtsch. Chem. Ges. B 1929, 62B,
1384–1389.
(9) Kalyuzhin, O. V.; Zemlyakov, A. E.; Kalina, N. G.; Mulik, E. L.;
Kuzovlev, F. N.; Makarova, O. V. Bull. Exp. Biol. Med. 2008, 145, 623–625.
(10) Juaristi, E.; Cuevas, G. Tetrahedron 1992, 48, 5019–5087.
(17) Yu, H. N.; Ling, C.-C.; Bundle, D. R. J. Chem. Soc., Perkin Trans. 1
2001, 832–837.
DOI: 10.1021/jo100826e
r
Published on Web 08/25/2010
J. Org. Chem. 2010, 75, 6115–6121 6115
2010 American Chemical Society

Caraballo et al.
JOCArticle
FIGURE 1. 1H NMR kinetic studies: (a) fraction of β-anomer formed from 1-thio-R-D-mannopyranose, 1-thio-β-D-galactopyranose, 1-thio-
β-D-glucopyranose and 1-thio-R-L-fucopyranose under acidic conditions (pD 4); (b) fraction of β-anomer formed from 1-thio-R-D-
mannopyranose at different pD; (c) fraction of β-anomer formed from 1-thio-R-L-fucopyranose and 1-thio-β-L-fucopyranose under basic
conditions (pD 9); (d) pD-dependence for the mutarotation of 1-thio-D-mannopyranose. Only 1-thioglycopyranose species taken into
consideration.
thiols effectively undergo mutarotation. This is, however, of
high importance, since sulfur-containing aldoses are increas-
ingly used in synthesis owing to their chemical stability, their
compatibility with numerous protection/deprotection steps,
and the enhanced thiol nucleophilicity compared to the corre-
sponding oxygen analogues.18-23 Furthermore, this class of
compounds is highly compatible with biological systems, yield-
ing stable mimics of natural glycosides due to the improved
resistance toward hydrolytic enzymes.24-27 Therefore, glycosyl
thioethers are widely used as glycosyl donors and provide easy
and rapid synthetic access to, for example, synthetic vaccines
and drugs28-33 and to neoglycoproteins and neoglycopep-
tides.19,34,35 This increasing popularity of glycosyl thiols com-
bined with the need for new synthetic methodologies and the
perpetual quest for fundamental understanding of chemical
processes led us to investigate and clarify whether 1-thioaldoses
undergo mutarotation. In the present study, we thus report on
the pH-dependent mutarotation of 1-thioaldoses and their
equilibrium anomeric ratios in aqueous media.
Results and Discussion
Four stereochemically pure 1-thioaldoses were initially
1
monitored by H NMR spectroscopy under acidic condi-
tions (Figure 1a). Thus, the mutarotation behavior of 1-thio-
D-mannopyranose (1),36 1-thio-D-galactopyranose (2), 1-thio-
D-glucopyranose (3), and 1-thio-L-fucopyranose (4)37 was
successively evaluated. The expected favored configurational
isomers in aqueous media were used, i.e., the R-anomer for the
manno derivative, and the β-anomer for the galacto and gluco
derivatives, except for the fuco derivative, likely to possess a
β-anomer preference similarly to L-fucopyranose, which was
initially tested starting with its R-form. At equilibrium, the
resulting mixtures were primarily composed of the cyclic
1-thioaldopyranoses. Disulfide products were generally
formed from the concomitant oxidation process but were
not taken into consideration in the analyses since they did
not influence the R/β anomeric ratio. Control experiments in
(18) Crich, D.; Li, M. Org. Lett. 2007, 9, 4115–4118.
(19) Pachamuthu, K.; Schmidt, R. R. Chem. Rev. (Washington, DC) 2006,
106, 160–187.
(20) Crich, D.; Smith, M. J. Am. Chem. Soc. 2001, 123, 9015–9020.
(21) Horton, D.; Wander, J. D. Thio sugars and derivatives. In The
Carbohydrates; Pigman, W.; Horton, D., Eds., Academic Press: New York,
1980; Vol. 1B, pp 799-842.
(22) Driguez, H. Top. Curr. Chem. 1997, 187, 85–116.
(23) Kiefel, M. J.; Thomson, R. J.; Radovanovic, M.; Von Itzstein, M.
J. Carbohydr. Chem. 1999, 18, 937–959.
(24) Roy, R.; Hernandez-Mateo, F.; Santoyo-Gonzalez, F. J. Org. Chem.
2000, 65, 8743–8746.
(25) Cohen, S. B.; Halcomb, R. L. J. Org. Chem. 2000, 65, 6145–
6152.
(26) MacDougall, J. M.; Zhang, X.-D.; Polgar, W. E.; Khroyan, T. V.;
Toll, L.; Cashman, J. R. J. Med. Chem. 2004, 47, 5809–5815.
(27) Munoz, J. L.; Garcia-Herrero, A.; Asensio, J. L.; Auzennaeau, F. I.;
Canada, F. J.; Jimenez-Barbero, J. J. Chem. Soc., Perkin Trans. 1 2001, 867–
872.
(28) Driguez, H. ChemBioChem 2001, 2, 311–318.
(29) Nicotra, F. Top. Curr. Chem. 1997, 187, 55–83.
(30) Witczak, Z. J. Curr. Med. Chem. 1999, 6, 165–178.
(31) Knapp, S.; Myers, D. S. J. Org. Chem. 2001, 66, 3636–3638.
(32) Knapp, S.; Myers, D. S. J. Org. Chem. 2002, 67, 2995–2999.
(33) Bousquet, E.; Spadaro, A.; Pappalardo, M. S.; Bernardini, R.;
Romeo, R.; Panza, L.; Ronsisvalle, G. J. Carbohydr. Chem. 2000, 19, 527–
541.
(34) Bernardes, G. J. L.; Grayson, E. J.; Thompson, S.; Chalker, J. M.;
Errey, J. C.; ElOualid, F.; Claridge, T. D. W.; Davis, B. G. Angew. Chem.,
Int. Ed. 2008, 47, 2244–2247.
(35) Bernardes, G. J. L.; Gamblin, D. P.; Davis, B. G. Angew. Chem., Int.
Ed. 2006, 45, 4007–4011.
(36) Pei, Z.; Larsson, R.; Aastrup, T.; Anderson, H.; Lehn, J.-M.;
€
Ramstrom, O. Biosens. Bioelectron. 2006, 22, 42–48.
(37) Hashimoto, H.; Shimada, K.; Horito, S. Tetrahedron: Asymmetry
1994, 5, 2351–2366.
6116 J. Org. Chem. Vol. 75, No. 18, 2010

Caraballo et al.
JOCArticle
presence of a reducing agent (dithiothreitol or phosphino-
triacetic acid) led to the same distributions.
TABLE 1. Anomeric Composition of 1-Thioglycopyranoses and Their
Corresponding Glycopyranoses (Percentage of β-Anomer Shown)
1-β-thioaldosea
1-β-hydroxyaldose
pD 4
pD 7
pD 9
ref 2 and 40
pD 4
Man
Fuc
Gal
Glc
66.2
74.3
78.1
74.1
60.9
79.1
77.7
75.2
24
<5
>95
93.6
34.5
69.3
68.1
62
33.3
71.4
68.0
62.5
aInitial anomers: 1-thio-R-D-mannopyranose (1), 1-thio-R-L-fucopyr-
anose (4), 1-thio-β-D-galactopyranose (2), and 1-thio-β-D-glucopyra-
nose (3).
The four glycosyl thiols reached their respective anomeric
equilibrium within comparable time lengths but showed
slightly different anomeric ratios. Thus, at room tempera-
ture, the equilibria were reached within 15 h, resulting in
approximately 78% β-anomer for the galactopyranose, 74%
β-anomer for the fuco- and the glucopyranose structures,
and surprisingly, 66% β-anomer for the manno derivative
(Table 1). This indicates the noninfluence of the carbohy-
drate structure on the equilibration time but an effect on the
anomeric configuration ratio, as expected.
conformer. For the mannose species, this result clearly
contrasts with the anomeric equilibrium ratio of the corre-
sponding 1-hydroxyaldose (Table 1).1,2,40 For the 1-thio-D-
mannopyranose, the anomeric distribution totally shifted in
favor of the β-configuration, resulting in an R/β ratio of 33.8/
66.2 at pD 4. For the normal mannopyranose, the distribu-
tion is in favor of the R-anomer (66.7/33.3) under the same
conditions. For the other 1-thioaldoses tested, the ratios also
shifted in favor of the β-anomer compared to the 1-hydro-
xyaldose analogues, albeit to a lower extent. For fucose, the
increase in β-species amounted to a few percent, whereas for
glucose and galactose, the preference for the β-anomers was
significantly higher. These results indicate that replacement
of a hydroxyl group at the anomeric center by a larger and
more polarizable thiol group can dramatically affect the
anomeric composition.
This R- and β-character of 1-thio-D-mannopyranose could
also be turned on/off by manipulation of the pH (Figure 1d).
Thus, the reversibility and adaptability of the mutarotation
process toward its environment were demonstrated. This
experiment, where acidic or basic conditions were initially
chosen, and subsequently switched to basic or acidic media,
respectively, clearly confirmed the faster equilibration at
low pH (∼15 h) compared to the slower rates at higher pH
(∼60 h). The mutarotation rates and the anomeric equilib-
rium ratios also proved identical after several rounds of pH
switching. In contrast, the observation of restricted mutar-
otation at basic conditions for the other thioaldoses tested
excludes any reversibility of the anomerization process for
these species.
The mutarotation of normal aldoses is usually both acid-
and base-catalyzed.1,38 The pH-dependence for the mutar-
otation of the thio analogues was thus addressed, and the pD
profile of anomerization was investigated at three different
1
pDs (Figure 1b). The results from the H NMR analyses
were in this case highly conspicuous, and the mutarotation
rate of 1-thioaldoses proved pH dependent.12 Thus, at a
lower pD than 7.7, a value corresponding to the thiol pKa,39
relatively rapid mutarotation occurred, whereas at higher
pD, the mutarotation proceeded at a (considerably) slower
rate. For the 1-thio-D-mannopyranose species, half-times of
3.1 and 3.3 h, respectively, were recorded under acidic (pD 4)
and neutral (pD 7) conditions, whereas the mutarotation rate
diminished under basic conditions (pD 9; t1/2 = 14.8 h). For
the other aldoses tested, the mutarotation process was al-
most completely blocked at higher pD. This was most
dramatic for the fucose and galactose species, in which case
very low anomerization from the initial configurations were
observed over the time range. Therefore, the anomerization
of the β-anomer of 1-thio-L-fucopyranose was subsequently
monitored under acidic (pD 4) and basic (pD 9) conditions to
confirm that the mutarotation process of these 1-thioglyco-
pyranoses is restricted under basic conditions. While the
anomeric equilibrium composition at pD 4 provided an R/β
ratio (∼20/80) similar to the one obtained starting with the
opposite anomeric form, 1-thio-R-L-fucopyranose (4) (∼26/
74), the results at basic pD proved to be different (Figure 1c).
Thus, the R-anomer was only observed in trace amounts,
supporting the hypothesis that the mutarotation of these
1-thioglycopyranoses is generally restricted under basic con-
ditions.
The structural analysis of the different 1-thioaldoses re-
sults in a clear pattern that may lead to prediction of the main
conformer, at equilibrium, for any given 1-thioaldopyra-
nose. First, carbohydrate derivatives from both the D- and
the L-series have been monitored with very similar results.
Thus, the anomerization of 1-thio-β-L-fucopyranose (6-
deoxy-1-thio-β-L-galactopyranose) and 1-thio-β-D-galacto-
pyranose (2) reached similar equilibria under both acidic and
basic conditions, suggesting the noninfluence from the D- or
L-character of the carbohydrate derivative on the mutarota-
tion process. Second, the comparison of the different epimers
Surprisingly, the β-anomer was observed as the major
mutarotation product in all cases at acidic and neutral
pH, independently of the R- or β-nature of the starting
(38) Imamura, A.; Okajima, T.; Kanda, K. Bull. Chem. Soc. Jpn. 1986, 59,
943–945.
(40) Ryu, K.-S.; Kim, C.; Park, C.; Choi, B.-S. J. Am. Chem. Soc. 2004,
126, 9180–9181.
€
(39) Larsson, R.; Ramstrom, O. Eur. J. Org. Chem. 2005, 285–291.
J. Org. Chem. Vol. 75, No. 18, 2010 6117

Caraballo et al.
JOCArticle
SCHEME 1. Synthesis of 1-Thio-R-D-altropyranose (11)a
aReagents and conditions: (a) I2, Ac2O, 0 °C to rt, 1 h (quant); (b) AcSH, BF3 Et2O, DCM, 0 °C to rt, 22 h (95%); (c) NaOMe, MeOH, 5 h (quant).
3
SCHEME 2. Proposed 1-Thio-D-altropyranose Mutarotation
Pathway
tested (the glucose/galactose and glucose/mannose deriva-
tives) offered valuable information. The structural changes
from an equatorial hydroxyl group at the C-4 carbon of the
carbohydrate ring (1-thio-β-D-glucopyranose (3)) to the
corresponding axial derivative (1-thio-β-D-galactopyranose
(2)) did not significantly affect the mutarotation behavior,
whereas the epimerization of the C-2 hydroxyl group of the
glucose derivative leading to an axial C-2 hydroxyl group
(D-mannose derivative, 2S-configuration) completely affec-
ted the anomeric mixture as well as the reversibility proper-
ties of the carbohydrate. Finally, the R- or β-preference of the
natural carbohydrates could be important to consider for the
prediction of the anomeric composition of their correspond-
ing 1-thio analogues. Of the four different 1-thioaldoses,
only the mannose derivative possesses, as a natural carbohy-
drate, an R-preference.
In order to determine which of these structural factors
mainly governs both the anomeric composition at equilibri-
um and the reversibility properties, 1-thio-R-D-altropyra-
nose (11) was synthesized and monitored under acidic (pD
4), neutral (pD 7), and basic (pD 9) conditions (Scheme 1).
1-Thio-R-D-altropyranose possesses an axial hydroxyl group
at the C-2 carbon of the carbohydrate ring (2S-configur-
ation), but contrary to 1-thio-R-D-mannopyranose, its cor-
responding natural derivative has a β-preference at equilib-
rium (R/β ratio of 44/56).41,42 The results were in this case
highly intriguing, and at acidic or neutral pDs, 1-thio-R-D-
altropyranose was rapidly (<10 min), and entirely, converted
to the β-anomer (12), whereas under basic conditions, the
altrose derivative remained (∼100%) in the R-form. Unfor-
tunately, 1-thio-R-D-altropyranose, similarly to D-altropyr-
anose (5), isomerized from the 4C1 chair conformation to the
1C4 chair conformation with time (Scheme 2).41,42 This was
of the chair conformations, which enabled the pD switch
experiment to be performed and, therefore, allowed the
observation of the reversibility properties. Thus, compound
12 instantaneously anomerized back to its original R-con-
formation after changing the pD of the solution from acidic
to basic conditions. These results supported the importance
of an axial hydroxyl group at the C-2 carbon of the carbohy-
drate ring over the anomeric preference of the corresponding
natural aldoses.
The mutarotation of 1-thioaldoses is most likely to follow
an exo-ring-opening mechanism in analogy to 1-hydroxy- or
1-aminoaldoses (Scheme 3).43 Protonation of the ring oxygen
and formation of a thiocarbenium ion intermediate would
rationalize the acid-catalyzed character of the reaction.44
Under basic conditions, the ring opening is strongly hampered
and the mutarotation rate is decreased. However, slow forma-
tion of 1-hydroxyaldoses (19) as well as the appearance of the
characteristic smell of H2S points to the existence of a com-
peting process other than mutarotation at lower pH. The
exo-ring-opening mechanism also enables the formation of
1
most dramatic at low pD where 12 isomerized to the C4
chair conformation (13), followed by its mutarotation to the
1
more favored C4 1-thio-R-D-altropyranose (14). However,
the mutarotation kinetics was faster than the isomerization
(43) Smiataczowa, K.; Kosmalski, J.; Nowacki, A.; Czaja, M.; Warnke,
Z. Carbohydr. Res. 2004, 339, 1439–1445.
(44) Indrugalla, D.; Bennet, A. J. J. Am. Chem. Soc. 2001, 44, 10889–
10898.
(41) Angyal, S. J. Angew. Chem., Int. Ed. 1969, 8, 157–166.
(42) Vijayalakshmi, K. S.; Rao, V. S. R. Carbohydr. Res. 1972, 22, 413–
424.
6118 J. Org. Chem. Vol. 75, No. 18, 2010

Caraballo et al.
JOCArticle
SCHEME 3. Proposed Mutarotation Mechanism
hemithioacetal derivative 17, a potential intermediate in the
formation of aldose species 19. The very fast ring closure of
the thiocarbenium ion intermediate compared to the addi-
tion of water would in this case justify the slow formation of
the corresponding normal aldoses. A second mechanism
involving solvolysis of the 1-thioaldose via oxocarbenium
ion intermediate 18 may, however, also be involved. The
slower rate of the oxocarbenium ion formation and its con-
version to 19 would induce an increasing but still low concen-
tration of 19 compared to 15 and 16. In addition, during the
evaluation of 1-thio-R-D-altropyranose (11) anomerization,
furanose derivatives (e.g., 10) could unambiguously be iden-
tified by 1H NMR spectroscopy pointing to the existence of
an acyclic intermediate. Therefore, the existence of a ring-
opening mechanism was supported.
The pH-dependent stereochemistry of the process, espe-
cially for the mannose and altrose species, is highly conspic-
uous. A tentative explanation for the results would concern
the effects directing the preferential R-anomer formation:
steric hindrance, electronic effects (especially the anomeric
effect), and solvation effects. The presence of a thiol group at
the anomeric center would thus partially reduce or promote
these effects, leading to a configurationally more stable
β-anomer at low or neutral pH or R-anomer at higher pH
exclusively for carbohydrates bearing an axial hydroxyl
group at the C-2 carbon of the ring. Considering the first
effect, the steric hindrance resulting from a sulfhydryl group
is likely to be comparable to a hydroxyl group, as reflected
by the conformational energy values for these groups.
A β-preference would thus generally be expected, in part
counteracted by an axial hydroxyl group at the C-2 carbon of
the ring.10 Steric effects are however strongly context depen-
dent, and the effect of sulfur may be larger than predicted.
Comparing the influence of the anomeric effect on the axial
versus the equatorial preferences of the 1-thioaldoses with
the normal oxygen-containing aldoses, the effects are likely
subtle and generally difficult to predict. For 5-thiopyranosyl
compounds, it has been proposed that the anomeric effect is
larger than for the oxygen analogues, resulting in increased
R-preference for these compounds.45,46 The combined endo-
(nS f σ*C-O) and exo- (nO f σ*C-S) anomeric effects in the
axial conformer was in this case reported to be more stabiliz-
ing than the exo- (nO f σ*C-S) anomeric interaction in the
equatorial conformer. In the present study, the groups are
reversed, but with the same reasoning the combined anome-
ric effects in the axial conformer (nO f σ*C-S and nS f
σ*C-O) would be more stabilizing than the exo- (nS f σ*C-O
)
anomeric interaction in the equatorial conformer. This is,
however, contradicted by the experimental results at low pH,
where the β-anomers are preferred. The combined anomeric
effect may on the other hand be more pronounced in the
thiolate species. Both the exo-anomeric and the endo-anome-
ric effects are likely to result in slightly better matching upon
deprotonation, potentially resulting in a higher degree of
R-anomer. Besides steric and electronic effects, solvation
effects are likely to be involved, and the effect of water as a
solvent is important. Further, in order to explain the mutaro-
tation of aldoses in aqueous media, it is well accepted that
one or several water molecules are involved in the ring-
opening and ring-closing mechanisms.12,51 Walkinshaw first
described the preferential conformation of aldopyranoses in
terms of hydrophilicity, a property directly related to the
number of solvent molecules surrounding and stabilizing the
carbohydrate structure.47 It was also generally observed for
aldoses that there is an increase in the proportion of the
R-anomer with decreasing dielectric constant of the solvent.47
Such a solvent-induced shift is, however, likely to be reduced
with the less polar sulfhydryl group compared to the hydro-
xyl moiety at lower pD. At higher pD, on the other hand, the
considerably lower solvation energy of the thiolate anion
may invoke additional steric effects, especially for the struc-
tures carrying an axial OH-group at C-2.48-50 Therefore, the
(45) Pinto, B. M.; Leung, R. Y. N. In The Anomeric Effect and Associated
Stereoelectronic Effects; Thatcher, G. R. J., Ed.; American Chemical Society:
Washington, DC, 1993; pp 126-155.
(46) Wolfe, S.; Shi, Z. Isr. J. Chem. 2000, 40, 343–355.
(47) Walkinshaw, M. D. J. Chem. Soc., Perkin Trans. 2 1987, 1903–1906.
(48) Cramer, C. J.; Truhlar, D. G. J. Comput. Chem. 1992, 13, 1089–1097.
J. Org. Chem. Vol. 75, No. 18, 2010 6119

Caraballo et al.
JOCArticle
methods. In the case of the neutral form, the calculations
predict a preference for the R-isomer by 0.5 kcal/mol. This is
in disagreement with the experiments, since the R/β ratio of
34:66 obtained from the NMR experiments shows that the
β-isomer is slightly lower in free energy. The computed R/β
preference is sensitive to the solvation description, and chan-
ging the cavity representation from the Bondi cavity to the
UA0 cavity-model reduces the energy difference to 0.1 kcal/
mol. For the deprotonated 1-thio-D-mannose, the calcula-
tions predict the β-isomer to be 2.0 kcal/mol lower in free
energy than the R-isomer, which is in contrast to the dom-
inance of the R-isomer in the NMR experiments. However,
the computational results are strongly dependent on the
description of solvent effects. The gas-phase energies of the
most favored conformers in solution show the β-isomer to be
favored by 2.6 kcal/mol. Also here the predictions are im-
proved by changing the cavity representation to the UA0
model; the energy difference is reduced to 0.6 kcal/mol. The
computations provide no simple explanation for the obser-
ved shift in R/β preference with pH. It seems that the R/β
ratio is determined by specific solvents interactions, which
cannot be reproduced by an implicit solvent model, such as
PCM.
FIGURE 2. Solution structures for the most stable conformers of
the R- and β-anomers of the deprotonated (top) and neutral (bottom)
forms of 1-thio-D-altrose. Geometries were optimized at the PCM-
MP2/6-31þG(d,p) level.
R- and β-preferences observed at high and low pD, respec-
tively, are likely to be affected by solvation effects.
To understand the molecular basis for the shift in R/β
preference with protonation state for 1-thio-D-altrose, ab
initio calculations with an implicit solvent model (PCM) at
the MP2/6-31G(d,p) level were performed. The optimized
structures for the most stable conformers in solution are
depicted in Figure 2. On the basis of the computed free ener-
gies of the anomers in solution, 9:91 and 89:11 distributions
between the R- and β-anomers for the neutral and ionic
forms of 1-thio-D-altrose, respectively, were predicted. This
is in good agreement with NMR measurements, which show
the β-anomer to dominate for the neutral form (>95%) and
the R-anomer for the ionic form (>95%). Focusing first on
the ionic structures, a strong and short internal hydrogen
bond between the axial hydroxyl group at the C-3 position
and the negatively charged thiolate group of the R-anomer
was found. A similar hydrogen bond cannot be formed in the
β-anomer. If the gas phase energies of the two conformers are
considered, the R-anomer is favored by 8.9 kcal/mol. This is
reduced to a free energy difference of 1.3 kcal/mol after inclu-
sion of solvent effects. The stabilizing effect of the internal
hydrogen bond of the R-anomer is partly compensated for by
a more effective solvation of the β-anomer due to its less
shielded thiolate group. However, the remaining effect is suf-
ficient to account for the strong dominance of the R-anomer
also in solution. It should be noted that it is not possible to
form an internal hydrogen bond of comparable strength in
the other 1-thioaldopyranoses, which explains why 1-thio-D-
altrose has the most strongly favored R-anomer. Considering
the solution structures of the R- and β-anomers of neutral
1-thio-D-altropyranose, no strong internal hydrogen bonds
involving the sulfhydryl group could be found; it is more
favorable to interact with the solvent. The β-anomer is favored
over the R-anomer by a free energy difference of 1.40 kcal/mol
in solution. In the gas phase, the energy difference between the
corresponding structures is slightly larger, 1.64 kcal/mol.
We have also investigated the shift in R/β preference with
protonation state for 1-thio-D-mannose by computational
Conclusions
In conclusion, it has been demonstrated that 1-thioaldo-
pyranoses undergo mutarotation in aqueous media, a pro-
cess that proceeds with comparable rates at lower and
neutral pH and is slower or almost blocked at higher pH.
For 1-thio-R-D-mannopyranose and 1-thio-R-D-altropyra-
nose, the process is fully reversible upon pH-switching. The
carbohydrates tested showed an interesting tendency to
preferentially form the β-anomer in acidic or neutral solu-
tion, whereas 1-thio-D-mannopyranose and 1-thio-D-altro-
pyranose displayed a strong pH dependence and a preference
for the R-configuration in a basic environment. Calculation
studies clearly displayed a strong solvation effect involved in
the mutarotation process and, in the case of the 1-thio-D-
altropyranose, the presence of internal hydrogen bonding at
basic conditions providing a conformation totally shifted in
favor of the R-anomer at equilibrium.
Experimental Section
General Methods. 1-Thio-R-D-mannopyranose (1),36 1-thio-
R-L-fucopyranose37 (4), and 1-thio-β-L-fucopyranose52 were
synthesized following procedures previously described. 1-Thio-
β-D-galactopyranose sodium salt (2), 1-thio-β-D-glucopyranose
sodium salt (3), and all other reactants and reagents were pur-
chased from commercial sources and used as received. 1H NMR
and 13C NMR spectra were recorded on 400 (100) MHz and/or
500 (125) MHz instruments at 298 K. 1H peak assignments were
made by first-order analysis of the spectra, supported by standard
1H-1H correlation spectroscopy (COSY). Buffer solutions were
prepared from D3PO4 (85%w/winD2O)andNaOD(40%w/win
D2O).
2,3,5,6-Tetra-O-acetyl-1-S-acetyl-1-thio-r-D-altrofuranose (8)
and 2,3,4,6-Tetra-O-acetyl-1-S-acetyl-1-thio-R-D-altropyranose
(9). Molecular iodine (5.08 mg, 0.02 mmol) was dissolved in
acetic anhydride (561 mg, 5.5 mmol), and the mixture was
(49) Cabani, S.; Gianni, P.; Mollica, V.; Lepori, L. J. Solution Chem.
1981, 10, 563–595.
€
(52) Andre, S.; Pei, Z.; Siebert, H.-C.; Ramstrom, O.; Gabius, H.-J.
Bioorg. Med. Chem. 2006, 14, 6314–6326.
(50) Pearson, R. G. J. Am. Chem. Soc. 1986, 108, 6109–6114.
(51) Yamabe, S.; Ishikawa, T. J. Org. Chem. 1999, 64, 4519–4524.
6120 J. Org. Chem. Vol. 75, No. 18, 2010

Caraballo et al.
JOCArticle
stirred at 0 °C. D-Altrose (100 mg, 0.55 mmol) (compound 5) was
added, and the reaction mixture was allowed to stir at 0 °C for
0.5 h and then at room temperature for another 0.5 h. After
detection of the reaction completion by TLC, DCM was added,
and the mixture was washed with saturated NaHCO3 and
extracted with DCM for three times. The organic phase was
collected and washed by Na2S2O5 and brine and dried over
Na2SO4. The solvent was then evaporated, and the crude com-
pounds 6 and 7 were used in the next step without further puri-
fication. Compounds 6 and 7 (210 mg, 0.54 mmol) and thioa-
cetic acid (165 mg, 2.16 mmol) were stirred in dry DCM (3 mL)
H-5, H-50), 5.10 (d, 2 H, J = 3.4 Hz, H-1, H-10). 13C NMR (125
MHz, D2O): δ 60.32, 65.03, 69.86, 70.92, 72.76, 90.45. MS (ESI)
measured for C12H22O10S2 ([M þ Na]þ): m/z 413.07.
Quantum Chemical Calculations. Conformational searches
were employed to identify the most stable conformation in solution
for each investigated isomer. The solution structures were obtained
by means of geometry optimization at the MP2/6-31þG(d,p) level
using Gaussian03.53 The effect of the solvent was modeled using
the standard PCM method of Gaussian 03 and the default united-
atom cavity model (UA0).54 Subsequent single point calculation
were performed at the PCM-MP2/6-31þG(d,p) level with an all-
atom cavity model, where the size of each atom was defined by its
van der Waals radius according to Bondi55 and scaled by the factor
1.2. We have found this cavity model to provide accurate relative
solvation energies for organic compounds, including neutral and
ionic compounds with sulfur.56 The final free energies in solution
included the nonelectrostatic components from the PCM calcula-
tions, but were not corrected for nuclear motions. It should be
noted that MP2 calculations employing large polarized basis sets
have been found to produce accurate relative energies in molecular
systems influenced by the anomeric effect.57
at 0 °C, and BF3 Et2O (461 mg, 3.24 mmol) was added dropwise
3
for 5 min. After removal of the ice bath, the reaction mixture was
warmed to rt and stirred for 22 h. The mixture was diluted with
DCM, washed with saturated NaHCO3, and extracted three
times. The organic phase was collected, washed with brine, and
dried over Na2SO4. The solvent was then evaporated, and the
crude was further purified by flash chromatography, providing
compound 8 (19%). 1H NMR (500 MHz, CDCl3): δ 2.06 (s, 3 H,
OAc), 2.10 (s, 3 H, OAc), 2.14 (s, 3 H, OAc), 2.15 (s, 3 H, OAc),
2.39 (s, 3 H, SAc), 4.11 (dd, 1 H, J = 5.7 and 12.3 Hz, H-6b), 4.19
(dd, 1 H, J = 2.8 and 7.6 Hz, H-4), 4.46 (dd, 1 H, J = 3.1 and
12.3 Hz, H-6a), 5.20 (d, 1 H, J = 2.8 Hz, H-3), 5.21 (s, 1 H, H-2),
5.26 (m, 1H, H-5), 6.04 (s, 1 H, H-1). 13C NMR (125 MHz,
CDCl3): δ 20.89, 20.93, 21.01, 31.17, 62.52, 69.79, 76.64, 81.26,
82.98, 86.39, 169.30, 169.55, 170.26, 170.66, 192.64. MS (ESI)
measured for C16H22O10S ([M þ Na]þ): m/z 429.13. Anal. Calcd
for C16H22O10S: C, 47.29; H, 5.46; S, 7.89. Found: C, 47.34; H,
Acknowledgment. This study was supported by the Swed-
ish Research Council, the Royal Institute of Technology,
and the European Commission (Contract No. MRTN-CT-
2005-019561). L.D. thanks the China Scholarship Council
for a special scholarship award.
1
5.28; S, 7.77. Further elution yielded compound 9 (76%). H
Supporting Information Available: NMR spectra of 1-thio-D-
mannopyranose and 1-thio-D-altropyranose species, and compu-
tational data. This material is available free of charge via the
Internet at http://pubs.acs.org.
NMR (500 MHz, CDCl3): δ 2.02 (s, 3 H, OAc), 2.08 (s, 3 H,
OAc), 2.18 (s, 3 H, OAc), 2.20 (s, 3 H, OAc), 2.41 (s, 3 H, SAc),
4.11 (dd, 1 H, J = 2.2 and 12.3 Hz, H-6b), 4.23 (m, 1 H, H-5),
4.32 (dd, 1 H, J = 4.4 and 12.3 Hz, H-6a), 5.01 (dd, 1 H, J = 0.9
and 3.5 Hz, H-2), 5.21 (dd, 1 H, J = 3.2 and 10.1 Hz, H-4), 5.32
(bt, 1H, H-3), 5.99 (bs, 1 H, H-1). 13C NMR (125 MHz, CDCl3):
δ 20.78, 20.91, 20.95, 21.05, 30.96, 62.49, 65.02, 66.71, 67.67,
71.32, 78.77, 169.17, 169.37, 169.53, 170.88, 192.04. MS (ESI)
measured for C16H22O10S ([M þNa]þ): m/z 429.13.
(53) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery Jr., J. A.; Vreven, T.; Kudin, K. N.;
Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci,
B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada,
M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.;
Na- kajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox,
J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.; Gomperts, R.;
Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.;
Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain,
M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski,
J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen,
W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian, Inc., Pittsburgh PA,
2003.
1-Thio-r-D-altropyranose-sodium Salt (11). NaOMe (16 mg,
0.29 mmol) was added to a solution of compound 9 (120 mg,
0.29 mmol) in MeOH (2 mL), and the reaction mixture was
stirred at room temperature for 5 h. After evaporation of the
solvent, compound 11 was obtained quantitatively as white
1
solid. H NMR (500 MHz, D2O): δ 3.82-3.78 (dd, 1 H, J =
3.5 and 9.5 Hz, H-4), 3.85-3.82 (m, 2 H, H-6a, H-6b), 3.89 (m,
1 H, H-3), 4.04 (dd, 1 H, J = 2.2 and 4.1 Hz, H-2), 4.35 (m, 1H,
H-5), 5.29 (bt, 1 H, H-1). 13C NMR (125 MHz, D2O): δ 61.02,
65.30, 68.40, 72.48, 74.13, 79.82. Flash chromatography led to
oxidation of compound 11 and yielded the disulfide product,
which was reduced to compound 11 by addition of dithiothrei-
(54) Cossi, M.; Scalmani, G.; Rega, N.; Barone, V. J. Chem. Phys. 2002,
117, 43–54.
(55) Bondi, A. J. Phys. Chem. 1964, 68, 441.
(56) Brinck, T.; Carlqvist, P.; Holm, A. H.; Daasbjerg, K. J. Phys. Chem.
A 2002, 106, 8827–8833.
1
tol. H NMR (500 MHz, D2O): δ 3.89-3.80 (m, 4 H, H-6a,
H-6a0, H-6b, H-6b0), 3.98-3.92 (m, 4 H, H-3, H-30, H-4, H-40),
4.13 (dd, 2 H, J = 3.4 and 5.1 Hz, H-2, H-20), 4.19-4.14 (m, 2 H,
(57) Weldon, A. J.; Vickrey, T. L.; Tschumper, G. S. J. Phys. Chem. A
2005, 109, 11073–11079.
J. Org. Chem. Vol. 75, No. 18, 2010 6121