Synthesis and elimination of C-3-labeled thiosialosides

Article abstract of DOI:10.1021/ol500917k

The synthesis of C-3-labeled phenylthio sialic acid derivatives and an investigation of stereoselectivity in elimination reactions for the synthesis of 2,3-dehydro derivatives (glycals) is described. The experimental results are consistent with the existence of a conformational change and may be indicative of the intermediacy of an all-axial oxacarbenium ion.

Full text of DOI:10.1021/ol500917k

Letter
pubs.acs.org/OrgLett
Synthesis and Elimination of C‑3-Labeled Thiosialosides
Cristina De Meo,* Clare E. Wallace, and Scott A. Geringer
Department of Chemistry, Southern Illinois University Edwardsville, Edwardsville, Illinois 62026, United States
S
* Supporting Information
ABSTRACT: The synthesis of C-3-labeled phenylthio sialic
acid derivatives and an investigation of stereoselectivity in
elimination reactions for the synthesis of 2,3-dehydro
derivatives (glycals) is described. The experimental results
are consistent with the existence of a conformational change
and may be indicative of the intermediacy of an all-axial
oxacarbenium ion.
N-Acetylneuraminic acid (Neu5Ac) is the most common
member of the sialic acid family, which is found at the
terminal position of glycoconjugate chains.1,2 Its involvement
in numerous biological phenomena, ranging from cell−cell
communication to pathogen recognition and oncogenesis, has
been a driving force in the search of efficient synthetic
methodologies to obtain oligosaccharides and glycoconjugates
containing α-sialic acid residues. In the field of sialylation
reactions, a dramatic improvement in the stereoselectivity
toward the natural α-glycosidic linkage has been recorded in
the past decades.3−5 Another major research effort is the
synthesis of 2,3-dehydro derivatives of sialic acids and
evaluation thereof as neuraminidase inhibitors and substrates.
In particular, a fast growing field has been the search for
neuraminidase influenza inhibitors.6−8
development of improved methodologies. To begin our
investigation, we assumed that the major elimination pathway
would be via an E1 mechanism, where the oxacarbenium ion,
formed upon the departure of the leaving group, would
eliminate mainly (or only) the pseudo-axial H-3 due to its
coplanarity in respect to the empty orbital of the anomeric
carbon (A, Scheme 1). It should be noted that, through the
text, 3-axial and 3-equatorial are referring to the substituent
2
orientation in the C5 chair conformation, standard for most
sialic acids, whereas pseudo-axial/equatorial refer to the
orientation in the oxacarbenium intermediates.
Scheme 1. Elimination Reactions of Labeled and Non-
labeled Sialosides 1a−d
Regardless of the synthetic target, it is clear that the
development of efficient methodologies in glycochemistry,
specifically in sialic acid chemistry, depends on extensive
knowledge of the mechanism of substitution reactions, to
form glycosidic bonds, or elimination reactions for the
synthesis of glycals. Complete understanding of a reaction
mechanism is in some cases a utopian task, as several
mechanisms might compete and the prevalence of one over
another might be decided by small alterations in reaction
features.9,10 On the other hand, for some systems, it is
possible to narrow down the mechanism to one specific
pathway and, in that case, gain a much better control of it.
As part of a research program dedicated to investigating
reaction mechanisms in sialic acid chemistry, herein we report
a study on the stereoselectivity of elimination of C-3-labeled
thiophenyl sialosides. For this purpose, labeled compounds 1a
and 1b, bearing deuterium at the C-3 axial and C-3 equatorial
positions, respectively, were synthesized and their reactivity/
selectivity was studied in the presence of N-iodosuccinimide
(NIS) and trifluoromethanesulfonic acid (TfOH) in dichloro-
methane as solvent (Scheme 1).
To evaluate the viability of this assumption, sialoside 1a
bearing an axial deuterium at the C-3 position was obtained
following the protocol described by Friebolin and Schmidt,
i.e., treatment of sialic acid 3 in alkaline solution with
deuterium oxide (Scheme 2).11 Under these reaction
conditions, the monosubstitution occurs exclusively in the
axial position.
The labeling reaction can be monitored by 1H NMR
following the disappearance of the signal corresponding to the
We anticipated that knowing whether the axial or equatorial
hydrogen at C-3 was eliminated would give some useful
insights into the course of the elimination reaction, its
outcome, and mechanism. This, in turn, might allow for the
Received: March 27, 2014
Published: April 30, 2014
© 2014 American Chemical Society
2676
dx.doi.org/10.1021/ol500917k | Org. Lett. 2014, 16, 2676−2679

Organic Letters
Letter
Scheme 2. Synthesis of D-3ax Sialoside 1a
Table 1. Elimination of D-3ax Sialoside 1a
entry solvent/temp (°C) ratio 2a:2b yield (%) major elimination
1
2
3
CH2Cl2/rt
CH2Cl2/0
1:5
1:6
1:6
98
87
84
H-3eq
H-3eq
H-3eq
CH2Cl2/−40
to be the major product of the reaction performed at room
temperature (2a:2b = 1:5, entry 1). Somewhat surprisingly,
this ratio was practically unaffected when the elimination
reaction was performed at a lower temperature of 0 or −40
°C (entries 2 and 3).
The preference of 1a to eliminate the pseudo-equatorial H-
3 can be rationalized by the isotopic effect created by the
stronger C−D bond at C-3 rather than an intrinsic tendency
of this system to eliminate in any case from this position.12
To confirm this possibility, we proceeded to the synthesis and
elimination of thiosialoside 1b having inverted configuration
at C-3 (D-3eq, H-3ax) with respect to that of 1a. The direct
introduction of deuterium in the equatorial position at C-3 of
sialic acid 3 did not seem feasible. Hence, we chose to
originate 1b from the dideuterated intermediate 1c. As
depicted in Scheme 3, elimination of 1c in the presence of
H-3ax in 3 (Figure 1a−c). Concomitant conversion of the H-
3eq signal from a double doublet to a doublet is indicative of
monodeuteration at the C-3 axial position. In the case of
overdeuteration, which leads to the formation of 1c, complete
disappearance of H-3 signals is observed. In order to stop the
reaction at the monodeuteration level, the pD value was
found to be of critical importance. Thus, after thorough fine-
tuning of the reaction conditions, we concluded that pD = 8
allows for a nearly quantitative monodeuteration. It should be
noted that the reaction at pD = 12 predominantly yields the
dideuterated product, as previously reported.11
Scheme 3. Synthesis of D-3eq Sialoside 1b
Figure 1. Monitoring labeling process by 1H NMR in D2O (400
MHz, a = 0 min; b = 60 min; c = 120 min).
Upon labeling, esterification of the reaction mixture in an
acidic solution of MeOH, followed by acetylation, gave either
the mono- or dideuterated products 4a or 4c, respectively.
Conversion of 4a in the corresponding chloride by treatment
with acetyl chloride in methanol, followed by the reaction
with thiophenol in the presence of diisopropylethylamine
(DIPEA), gave thioglycoside 1a in 81% yield over two steps
(Scheme 2). Having obtained the labeled thiosialoside 1a, we
turned our attention to studying its properties in the
elimination reaction in the presence of NIS/TfOH as the
promoter system in dichloromethane. The ratio between H-3
glycal 2a and D-3 glycal 2b, the two possible elimination
NIS/TfOH yielded glycal 2b in 96% yield. Treatment of 2b
with N-bromosuccinimide (NBS), in the presence of sodium
acetate and acetic acid, gave two 3-brominated diastereomers
5a and 5b, which could be separated by silica gel
chromatography. Reduction of either 5a or 5b with tributyltin
hydride occurred with high stereoselectivity, and in both cases,
the hydride was delivered in the axial position almost
exclusively, probably due to steric hindrance. Therefore,
both intermediates 5a and 5b could be used as precursors
for the synthesis of D-3eq thiosialoside 6. Conversion of 6
into the corresponding chloride, followed by the reaction with
thiophenol in the presence of DIPEA, afforded the desired D-
3eq thiosialoside 1b in 71% yield (Scheme 3).
1
products, was determined by H NMR analysis of the crude
reaction mixture. As summarized in Table 1, D-3 glycal 2b,
resulting from the elimination of the H-3eq in 1a, was found
With the D-3eq thiosialoside 1b in hand, we turned our
attention to studying its elimination reaction in the presence
2677
dx.doi.org/10.1021/ol500917k | Org. Lett. 2014, 16, 2676−2679

Organic Letters
Letter
of NIS/TfOH. As listed in Table 2, elimination of 1b at room
temperature gave a mixture of H-3 glycal 2a and D-3 glycal
2b in a high yield (entry 1). The observed ratio of 2a:2b =
1:2 was indicative of the preferential elimination of the H-3ax
in 1b. The effect of the temperature, although still minimal,
was in this case more pronounced at −40 °C, with a ratio of
2a:2b = 1:4 being obtained (entry 3). These results prove
that elimination from the pseudo-axial position can occur, and
the overall stereoselectivity observed in the elimination of 1a
and 1b is mainly controlled by the relative position of
hydrogen vs deuterium.
thiosialoside 1a, two conformations of the oxacarbenium
intermediate A and B can be envisaged with the major
elimination product arising from B (Scheme 4). On the other
hand, starting from thiosialoside 1b, two conformations of the
oxacarbenium intermediate C and D can be envisaged, with
the main elimination occurring in this case from the all-
equatorial conformation C (Scheme 4).
To support this general concept and elucidate a possible
influence of the all-axial conformation, we decided to obtain
and study two new series of sialosides modified at C-5,
trifluoroacetamido sialosides 7a,b, and oxazolidinones 9a,b
(Scheme 5).
Table 2. Elimination of D-3eq Sialoside 1b
Scheme 5. Synthesis of Sialosides 7a,b and 9a,b
entry solvent/temp (°C) ratio 2a:2b yield (%) major elimination
1
2
3
CH2Cl2/rt
CH2Cl2/0
1:2
1:2
1:4
93
63
71
H-3ax
H-3ax
H-3ax
CH2Cl2/−40
One explanation of the cumulative experimental data
presented in Tables 1 and 2 invokes the existence of an
equilibrium between an all-equatorial and all-axial conforma-
tions of the oxacarbenium ion, the key intermediate en route
via the E1 pathway.
In this context, a conformational equilibrium of the
oxacarbenium ion, and its contribution to controlling the
overall efficiency and stereoselectivity of glycosylation
reactions with model compounds and hexoses, has been
acknowledged by many groups.13 Oxacarbenium ions and
conformations thereof within the sialic acid series are much
less studied. However, the ease of the synthesis of 2,7-
dehydro14 and 1,7-lactone15 derivatives (Scheme 4) is
Structural modifications at C-5 have found a broad
application in sialic acid chemistry due to their profound
effect on the reactivity, selectivity, and yields.4 For our testing,
sialosides 7a and 7b bearing a strongly electron-withdrawing
5-trifluoroacetamido group should provide additional stabiliza-
tion of the all-axial conformation while disfavoring the all-
equatorial counterpart.16 On the other hand, 4,5-oxazolidi-
none-substituted sialosides 9a and 9b having trans-fused rings
should limit the conformational freedom of the pyranose ring.
As a result of these induced structural modifications and
conformational preferences, we expected different results from
those acquired with standard 5-acetamido sialosides 1a and
1b, which would imply a profound effect of conformation on
stereoselectivity.
Scheme 4. Conformational Equilibrium at the
Oxacarbenium Ion Level and Known Sialosides with C2
5
Conformation
The syntheses of D-3ax sialosides 7a and 9a and their D-
3eq counterparts 7b and 9b were accomplished from the
respective labeled precursors 1a and 1b following standard
manipulations as depicted in Scheme 5.17−19 As a control
standpoint, we also obtained and tested 3-D,D and 3-H,H
sialosides 7c, 9c and 7d, 9d, respectively. Sialosides 7c and 7d
gave the corresponding glycals 8a,b within minutes and in
moderate/high yield (89% for 7c and 97% for 7d).
Conversely, elimination of the sialosides 9c and 9d afforded
the corresponding glycals 3-H 10a and 3-D 10b within
minutes in high yield (73% for elimination of 9c and 74% for
elimination of 9d).
Elimination reactions of D-3ax 7a gave results similar to
those of the corresponding N-acetyl donor 1a (8a:8b = 1:6,
entry 1, Table 3). On the other hand, when D-3eq sialoside
7b was tested, we obtained a mixture of glycals closer to a 1:1
ratio (8a:8b = 1:1.4, entry 2, Table 3). The difference in the
ratio observed between 7b and 1b (2a:2b = 1:2), although
certainly indicative of the existence of all-axial conformations
for sialic acid intermediates and substrates. In view of this
equilibrium, our data might be the result of isotopic
influenced pseudoaxial-eliminations from the two conforma-
tions.
In fact, regardless of the orientation of H-3 in the starting
thioglycoside, the system would adjust itself to provide a
suitable conformation to facilitate the elimination of more
weakly bonded H-3 rather than D-3. Thus, starting from
2678
dx.doi.org/10.1021/ol500917k | Org. Lett. 2014, 16, 2676−2679

Organic Letters
Letter
not very high, might be related to the effect of the C-5
modification on the conformational equilibrium of the
oxacarbenium ion.
ACKNOWLEDGMENTS
■
We acknowledge Southern Illinois University Edwardsville
College of Art and Science and Graduate School. We also
thank Dr. Leonid Kononov (N.D. Zelinsky Institute of
Organic Chemistry, Russian Academy of Sciences) for his
helpful feedback.
Encouraged by these results, we proceeded with testing the
elimination reactions of 9a and 9b having, at C-5,
oxazolidinone groups and D-3ax and H-3ax, respectively.
Elimination of thiosialoside 9a gave a mixture of glycals 10a,b
with a ratio of 2:1, showing a preference of this system to
cleave the stronger pseudo-axial C−D bond rather than the
weaker pseudo-equatorial C−H bond. Conversely, elimination
of 9b, with deuterium in the equatorial position, gave mainly
elimination of the pseudo-axial H-3 (10a:10b = 1:7). The
opposite stereoselectivity observed for the elimination of 9a vs
1a suggests that a conformational change of the oxacarbenium
ion intermediate might be a strong factor governing the
stereoselectivity of elimination reactions.
DEDICATION
■
This paper is dedicated to the late Dr. Masangu Shabangi and
to the late Dr. Kevin Johnson (Southern Illinois University
Edwardsville).
REFERENCES
■
(1) Rosenberg, A. Biology of Sialic Acids; Plenum Press: New York,
London, 1995.
(2) Varki, A. Trends Mol. Med. 2008, 14, 351−360.
(3) Ress, D. K.; Linhardt, R. J. Curr. Org. Synth. 2004, 1, 31−46.
(4) De Meo, C.; Priyadarshani, U. Carbohydr. Res. 2008, 343,
1540−1552.
(5) Boons, G. J.; Demchenko, A. V. Chem. Rev. 2000, 100, 4539−
4565.
Table 3. Elimination of C-5-Modified Thiosialosides
(6) Kiefel, M.; von Itzstein, M. Chem. Rev. 2002, 102, 471−490.
(7) von Itzstein, M. Nat. Rev. Drug Discovery 2007, 6, 967−974.
(8) Kulikova, N. Y.; Shpirt, A. M.; Kononov, L. O. Synthesis 2006,
2006, 4113−4114.
(9) Mydock, L. K.; Demchenko, A. V. Org. Biomol. Chem. 2010, 8,
497−510.
́
(10) Bohe, L.; Crich, D. C. R. Chim. 2011, 14, 3−16.
(11) Schmidt, H.; Friebolin, H. J. Carbohydr. Chem. 1983, 2, 405−
413.
(12) Gervay reported syn elimination of deuterated sialoside under
different reaction conditions. Horn, E. J.; Gervay-Hague, J. J. Org.
Chem. 2009, 74, 4357−4359.
entry sialoside
products
ratio
yield (%) major elimination
(13) Walvoort, M. T. C.; Dinkelaar, J.; van den Bos, L. J.; Lodder,
G.; Overkleeft, H. S.; Codee, J. D. C.; van der Marel, G. A.
Carbohydr. Res. 2010, 345, 1252−1263.
1
2
3
4
7a
7b
9a
9b
8a:8b
1:6
69
87
58
62
H-3eq
H-3ax
D-3ax
H-3ax
8a:8b
1:1.4
2:1
10a:10b
10a:10b
(14) Furuhata, K.; Ogura, H. Chem. Pharm. Bull. 1992, 40, 3197−
3200.
1:7
(15) Sugiyama, N. S. K.; Yamada, N.; Goto, M.; Ban, C.; Furuhata,
K.; Takayanagi, H.; Ogura, H. Chem. Pharm. Bull. 1988, 36, 1147−
1152.
(16) Yang, M. T.; Woerpel, K. A. J. Org. Chem. 2008, 74, 545−553.
(17) Tanaka, H.; Nishiura, Y.; Takahashi, T. J. Am. Chem. Soc.
2006, 128, 7124−7125.
(18) Farris, M. D.; De Meo, C. Tetrahedron Lett. 2007, 48, 1225−
1227.
(19) Crich, D.; Li, W. J. Org. Chem. 2007, 72, 2387−2391.
(20) An alternative E2 mechanism, prior leaving group full
departure, would lead to a syn elimination of the axial position,
which is not consistent with the data aquired. An E2 pathway from
triflate intermidiates seems unlikely at room temperature.
(21) Seeman, J. I. Chem. Rev. 1983, 83, 83−134.
In conclusion, novel C-3 deuterium-labeled sialosides 1−9
were synthesized, including C-5 modified sialosides with
oxazolidinone and trifluoroacetyl groups. Elimination reactions
of each of these sialosides suggested the presence of a
conformational equilibrium which is strongly influenced by
the isotopic effect, conformational constraints, and electro-
negativity of the C-5 substituent.20 Although the ratio of the
glycals obtained did not provide a quantitative measurement
of the conformational equilibrium (Curtin−Hammett princi-
ple),21 it did confirm, in our opinion, the presence and the
relevance of an all-axial oxacarbenium ion intermediate, which
should be taken into account when analyzing results of
elimination and sialylation reactions.
ASSOCIATED CONTENT
■
S
* Supporting Information
1
Experimental procedures and H and 13C NMR spectra for all
new compounds. This material is available free of charge via
the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: cdemeo@siue.edu.
Notes
The authors declare no competing financial interest.
2679
dx.doi.org/10.1021/ol500917k | Org. Lett. 2014, 16, 2676−2679