Delineating the physical organic profile of the 6-fluoro glycosyl donor

Full text of DOI:10.1016/j.jfluchem.2015.06.004

G Model
FLUOR-8577; No. of Pages 6
Journal of Fluorine Chemistry xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Fluorine Chemistry
journal homepage: www.elsevier.com/locate/fluor
Delineating the physical organic profile of the 6-fluoro glycosyl donor
Nico Santschi a,b, Nuria Aiguabella a,b, Vanessa Lewe a, Ryan Gilmour a,b,
*
a Institute for Organic Chemistry, Westfa¨lische Wilhelms-Universita¨t Mu¨nster, Correnstrasse 40, 48149 Mu¨nster, Germany
b Excellence Cluster EXC 1003 ‘‘Cells in Motion’’, Westfa¨lische Wilhelms-Universita¨t Mu¨nster, Correnstrasse 40, 48149 Mu¨nster, Germany
A R T I C L E I N F O
Article history:
Received 16 April 2015
Received in revised form 26 May 2015
Accepted 3 June 2015
Available online xxx
Dedicated to Professor Ve´ronique
Gouverneur on her receipt of the 2015 ACS
Award for Creative Work in Fluorine
Chemistry.
Keywords:
Carbohydrates
Fluorine
Glycosylation
Physical organic chemistry
Stereoelectronic effects
1. Introduction
derivative that is routinely used in preparative glycochemistry.
6-Deoxy-6-fluoroglucose building blocks have found widespread
application in a variety of disciplines ranging from mechanistic
enzymology [6], and glycopeptide antigen design [7] through to
radiotracers for nuclear medicine [8]. Moreover, from the
perspectives of conformation and reactivity, they are intriguing
candidates for investigation on account of the vicinal relationship
that links the fluorine atom to an electropositive centre (F–C –C –
Fluorinated glycostructures are emerging as valuable mimics of
natural carbohydrates on account of their enhanced lipophilicity
and the minimal steric disruption that results from hydroxyl to
fluorine substitution [1–3]. The ability to subtly modulate the
hydrogen bond network of the glycostructure without altering the
overall topology of the scaffold renders fluorine incorporation
attractive as a molecular editing strategy [4]. Effective synthetic
methods to construct complex fluorinated carbohydrates are
required to meet the demands of this expanding research field.
Unlike their natural counterparts, very little is known about the
behavior of fluorinated glycosyl donors, and the effect this
seemingly innocuous [OH ! F] switch has on a glycosylation
process. To reconcile this dearth of information with the growing
interest in fluorinated carbohydrates [5], the 6-deoxy-6-fluor-
oglycosyl donor 1 was selected as a model compound for a physical
organic analysis, and compared to the common 6-benzyloxy
b
a
O–) via a freely rotatable (exocyclic) bond. Fluorinated scaffolds
bearing an electron withdrawing substituent in the -position are
known to adopt conformations that benefit from stabilising
C–F*) and electrostatic interactions
b
hyperconjugative (sC–H
!
s
d
d
(F ꢀꢁ ꢁ ꢁO +): this is commonly referred to as the gauche effect
[9]. These effects may conceivably be exploited to control
glycosylation selectivity: this notion has been validated in the
corresponding 2-deoxy-2-fluoro systems [10]. To explore the role
of the 6-fluoro substituent in
a conventional glycosylation
reaction, the trichloroacetimidate donor 1 was selected as a
platform for investigation (Fig. 1). Stereoelectronic considerations
suggest that the conformational equilibrium in 1 would likely favor
conformers I and II over III. Indeed, of the three staggered
conformers partitioned by 1208, the syn-clinal endo arrangement I
*
Corresponding author at: Westfa¨lische Wilhelms-Universita¨t Mu¨nster,
Institute for Organic Chemistry Correnstrasse 40, 48149 Mu¨nster, Germany.
Tel.: +49 251 83 33271.
would likely predominate on account of the stabilizing sC–H
F* hyperconjugative interaction. Assuming a mechanism with
!
sC–
E-mail address: ryan.gilmour@uni-muenster.de (R. Gilmour).
http://dx.doi.org/10.1016/j.jfluchem.2015.06.004
0022-1139/ß 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: N. Santschi, et al., J. Fluorine Chem. (2015), http://dx.doi.org/10.1016/j.jfluchem.2015.06.004

G Model
FLUOR-8577; No. of Pages 6
2
N. Santschi et al. / Journal of Fluorine Chemistry xxx (2015) xxx–xxx
Fig. 1. The fluorine gauche effect in the 6-deoxy-6-fluoro glycosyl donor 1. LG = leaving group (trichloroacetimidate).
significant SN1 character [10c], the incipient oxocarbenium ion
explored relying on the addition of DeoxofluorTM at 0 8C, followed
by stirring at 70 8C [17]. This proved successful in delivering the
fluorinated product 6 in 76% yield as a white solid. Attempts to
hydrolyse the methyl acetal of 6 to the corresponding lactol (C1–
OH) 8 proved difficult [18]. A two-step protocol was therefore
implemented, beginning with the treatment of 6 with BCl3ꢁSMe2
in CH2Cl2 for 30 min at ambient temperature [19]. This success-
would likely benefit from dipole minimization thus placing
the highly electronegative fluorine substituent proximal to the
electrophilic (sp2) anomeric centre. Extensive studies by the
Woods and Woerpel groups have demonstrated that the 3H4
conformation of the per-benzylated glucose-derived oxocarbe-
nium ion benefits from electrostatic stabilization arising from the
axial benzyloxy groups [12]; this would enforce an axial
arrangement of the mono-fluoromethyl substituent thus augment-
ing the [C–F ! C55O+] dipole minimization and increasing prox-
imity to the anomeric carbon (Fig. 1, right). In the case of the 6-OBn
analog, conformer I should also predominate on account of the
fully generated the chloroacetal 7 exclusively as the a-anomer in
46% isolated yield, which was then exposed to a mixture of DMF
and saturated aqueous NaHCO3 to give the desired lactol 8 in 92%
yield. Finally, preparation of the trichloroacetimidate donor was
straightforward and proceeded in almost quantitative yield (99%,
sC–H
!
s
C–OBn* hyperconjugative interaction. However this con-
formational preference may be less pronounced on account of the
the poorer acceptor capabilities of the C–OBn* orbital, and the
exclusively the a-anomer), thereby completing the synthesis
sequence in 24% over 6 steps. For a comparative selectivity
analysis, the C6-OBn substituted derivative 13 was also required:
this was accessed from commercial 2,3,4,6-tetra-O-benzyl-
s
larger steric demands of the benzyloxy group (cf. fluorine). These
considerations may lead to the 3H4 transition state being favored
over the 4H3 when substituting benzyloxy by fluorine: this ought to
D
-glucopyranose, as previously described [10a], in an a:b ratio
of 11:1 as determined by 1H NMR spectroscopy (see Table 1)
(Scheme 1).
be reflected in enhanced b-selectivity. Since the fluorine substitu-
ent is generally perceived as having a small steric footprint, a top-
face attack on the 3H4 oxocarbenium ion intermediate according to
the Fu¨rst–Plattner rule should be less hindered than in the
corresponding OBn-substituted system (see Scheme 3).
To glean insights into the conformational behavior of the 6-
fluoro oxocarbenium ion that is central to this study (Fig. 1, Scheme
2) the corresponding gluconolactone was selected as a mimic for
NMR conformer population analysis. Lactones of this nature have
frequently been employed as convenient models for the study of
ephemeral oxocarbenium ions on account of the planar geometry
around the O–C–O fragment and the partial positive charge on the
pyranose-oxygen (Scheme 2, left) [20]. It was envisaged that
analyses of these systems might also provide useful information
regarding the conformational equilibrium that results from
rotation around the C5–C6 bond: the populations of the three
limiting staggered conformations partitioned by 1208 were of
particular interest (Scheme 2, right. I, II and II, Fig. 1). Based on the
Herein, we report
a variable temperature glycosylation
selectivity study for the 6-deoxy-6-fluoroglucose donor 1 in a
model reaction and validate the fluorine substituent as a useful
steering group in chemical glycosylation. By extrapolation of the
associated entropic (DD
S
ba
z) and enthalpic (DD
H
z) parameters,
ba
a comparison can be made with the corresponding 6-benzyloxy
system that is commonly employed in preparative glycochemistry.
2. Results and discussion
findings that 2,3,4,6-tetra-O-benzyl-D-gluconolactone (11) exists
The synthesis of fluorinated donor 9 began with the per-
benzylation of commercially available methyl 4,6-O-benzyli-
in a 4H3 conformation, rather than a B2.5 boat [21], the half-chair
transition states were preferentially considered.
dene-
a
-
D
-glucopyranoside (3). Under standard reaction condi-
To perform the solution phase conformer population analysis,
the experimental JHH coupling constants were required. Com-
3
tions with NaH, BnBr and TBAI in DMF, the desired product 4 was
isolated in 85% yield after low-temperature recrystallization from
ether/pentane [13]. Subsequently, regioselective reduction of the
benzylidene acetal afforded the free 6-hydroxyl derivative in 90%
yield relying on a combination of catalytic amounts of Cu(OTf)2
and 5 equivalents of BH3ꢁTHF in CH2Cl2 at 0 8C [14]. The
fluorination of compound 5 had previously been described with
DAST at 70 8C [15]. However, due to the thermal instability of this
reagent at elevated temperatures [16], an alternative protocol was
pound 10 was synthesized starting from structure 8 in 92% by
3
Albright–Goldman oxidation, and the JHH constants were deter-
mined to be 2.9 and 1.7 Hz. For compound 11 the literature values
of 3.2 Hz and 2.4 Hz were used [22]. Finally, a simplified structural
model was employed (12) (Scheme 2) and the theoretical 3JgHH and
3JaHH were calculated based on estimated group electronegativities
of F = 1.71, OBn = 1.20, OAc = 1.57 and CH2OMe = 0.13 [23]. Under
these assumptions, molar fractions (xI, xII, xIII) of (0.74, 0.23, 0.03)
Please cite this article in press as: N. Santschi, et al., J. Fluorine Chem. (2015), http://dx.doi.org/10.1016/j.jfluchem.2015.06.004

G Model
FLUOR-8577; No. of Pages 6
N. Santschi et al. / Journal of Fluorine Chemistry xxx (2015) xxx–xxx
3
Table 1
Summary of selectivity data for donors 9 and 13 at different temperatures.
iPrOH (1.2 eq)
TMSOTf (0.1 eq)
X
X
O
BnO
F
O
BnO
BnO
BnO
BnO
OiPr
4Å MS
DCM (0.05 M)
BnO
O
CCl3
NH
9
14 F
, X =
15, X = OBn
,
X =
13, X = OBn
b
Donor
T (K)a
r
ba
sba
r0
r
/r0
ba ba
ba
9 (X5F)
299.15 (25)
274.15 (0)
1.14
0.07 (6.5%)
0.09 (8.6%)
0.08 (5.5%)
0.13 (5.2%)
0.10 (2.7%)
0.83
1.07
1.56
2.53
3.80
0.73
0.99
1.03
0.98
1.00
9
9
9
9
1.08c
1.52c
2.58
3.79
244.15 (ꢀ30)
214.15 (ꢀ60)
194.15 (ꢀ80)
13 (X5OBn)
299.15 (25)
274.15 (0)
1.22
1.44
3.27
5.52
6.82
0.06 (4.7%)
0.10 (6.7%)
0.19 (5.7%)
0.22 (4.1%)
0.33 (4.9%)
1.24
1.69
2.68
4.84
7.94
1.01
1.18
0.82
0.88
1.16
13
13
13
13
244.15 (ꢀ30)
214.15 (ꢀ60)
194.15 (ꢀ80)
a
In parentheses, values in 8C.
Relative standard deviations sba/r
ba
Two reactions run at 0.05 M and two reactions run at 0.03 M.
b
.
c
Scheme 1. Synthetic route to fluorinated glycosyl donor 9.
and (0.79, 0.20, 0.01) were determined for lactones 10 and 11,
respectively. For compound 10, xII may also be estimated based on
attention was then turned to the performance of the glycosyl donor
in a model glycosylation reaction. To that end, the trichloroace-
timidate donors 9 and 13 were independently subjected to
standard, commonly employed glycosylation conditions using
iPrOH as a glycosyl acceptor and TMSOTf as Lewis acid catalyst.
Furthermore, in order to preclude solvent participation, all
3
3 g
the JCF coupling constant according to
x
II = (3JexpCF ꢀ J CF)/
(3JaCF ꢀ J CF) with standardized values of 3JgCF = 1.2 ꢂ 1.0 Hz and
3JaCF = 11.2 ꢂ 2.0 Hz [24]. Thus, for 3JexpCF = 5.8 Hz a higher popula-
tion xII = 0.46 results. These data indicate that the major solution
phase conformer has the highly electronegative fluorine substituent
in close proximity to the electron deficient pyranose oxygen (I): this is
consistent with an earlier conformational analysis of 6-deoxy-6-
fluoro-D-glucose [25]. It is interesting to note that an elegant
reactivity study by Bols and co-workers implicated rotamer I as
being the most reactive in a study of glycoside hydrolysis [26]. This
study indicates that both donors (X55F or OBn) display comparable
conformational behavior with respect to rotation about C5–C6. This is
consistent with the working stereoelectronic hypothesis and can be
3 g
rationalized based on hyperconjugative interactions (sC–H ! s*C–X
;
X55F or OBn)—albeit a stronger bias was initially expected for the
fluorinated system. For conformers I and II, a stabilizing Coulombic
interaction can also be invoked on account of the proximity of the
partially negatively charged substituent X (X55F or OBn) to the
pyranose oxygen: this should be more pronounced in the 3H4 half-
chair than in the 4H3 half-chair (closer spatial alignment; Scheme 3).
Having analyzed the solution phase conformational behavior of
the oxocarbenium ion mimics 10 and 11 by NMR spectroscopy,
Scheme 2. Lactones 10 and 11 used as models for 1H NMR conformer population
analysis.
Please cite this article in press as: N. Santschi, et al., J. Fluorine Chem. (2015), http://dx.doi.org/10.1016/j.jfluchem.2015.06.004

G Model
FLUOR-8577; No. of Pages 6
4
N. Santschi et al. / Journal of Fluorine Chemistry xxx (2015) xxx–xxx
Scheme 3. Tentative SN1 type glycosylation mechanism with b a
-TS (3H4) and -TS (4H3) and possible nucleophile trajectories (solid and dashed lines). The solid gray arrows
corresponds to trajectories that are consistent with the Fu¨rst–Plattner rule (solid gray).
z
experiments were conducted in CH2Cl2 (c 0.05 M). After quenching
the reaction mixtures by addition of NEt3 and subsequent
note that when the data point was included, slightly different DDH
ba
and DDS z values of ꢀ6.00 ꢂ 0.72 kJ molꢀ1 and ꢀ20.0 ꢂ 3.1 J molꢀ1
ba
Kꢀ1 were obtained with a goodness of fit of 0.95. Next, predicted
concentration under reduced pressure, the selectivities [
b
:
a
ratios
(r )] were determined directly either by 19F or 1H NMR
selectivity ratios r0
were calculated based on all values derived
ba
ba
spectroscopy for 14 and 15, respectively (Table 1). Each experi-
ment was repeated at least 3 times to ensure reproducibility and
estimate the accuracy of the method in the form of standard
deviations (sba); these ranged from 2.7 to 8.6%. Moreover, the
glycosylation reactions were run at five discrete temperatures,
ranging from ambient temperature to ꢀ80 8C in order to derive the
using rearranged Eq. (1) [28]. These selectivity ratios closely matched
the experimental values, as evidenced by the ratio (r /r0 ) ranging
ba ba
from 0.88 to 1.16.
z
For analysis of the entropic parameter DD
S
ba
a fully
dissociative, SN1-type mechanism was assumed (Scheme 3),
proceeding via an intermittent oxocarbenium ion followed by
reaction with the acceptor. This is consistent with the observed
thermodynamic parameters DD
H
and DD
S
. These data are
ba
ba
summarized in Table 1 and include the predicted selectivity ratios
erosion of the
b:a ratio associated with reaction of the
(r0 ) as determined based on the aforementioned values of
(a
-enriched) trichloroacetimidate donor to form the isopropyl-
ba
DD
H
and DD
S
.
ba
glycoside (i.e. consistent with a mechanism which has significant
SN1 character) [see Ref. [10c]]. In contrast, a reaction with
significant SN2-character would be expected to furnish a product
ba
Whereas both donors performed comparably at ambient
temperature with r
values of 1.14 ꢂ 0.07 and 1.22 ꢂ 0.06 for
ba
14 and 15, respectively, a clear selectivity differences were noted at
lower temperatures (6-OBn > 6-F). To extrapolate the thermody-
namic parameters DDH and DDS , the data was then subjected to
further investigations on the basis of an Eyring plot according to the
following equation [27].
b:a ratio that reflects the associated inversion of configuration at
the anomeric centre (i.e. an inverted
donors employed). In the case of donor (
-selectivity would be expected.
In this SN1 paradigm, a negative entropy of activation for the
b
:a
ratio of that of the
a
)-9, almost exclusive
ba
ba
b
associative second step can rezasonably be assumed (
D
Sz < 0).
DS
z) < 0 implicates
a
lnðr Þ ¼ ꢀDD
ba
H
zðRTÞꢀ1
þ
DDS
ba
z ꢃ Rꢀ1
(1)
The z finding that
(
DD
S
ba
=
D
S
b
z ꢀ
ba
D
S
<
D
S
z < 0, i.e. that for both substitutions (X55F or OBn)
b
a
By means of linear regression (ln(r ) ꢄ Tꢀ1), the enthalpic
the
(4H3).
b a-TS
-TS (3H4) is more constrained than the corresponding
ba
z) contributions to the difference in
(
DD
H
z) and entropic (DD
S
ba
ba
free energy between the
b
-
and
a-transition states
Moreover, since the donors vary only at the C6 locus, differences
z
z ꢀ
DG
a
z) were calculated to be ꢀ7.02 ꢂ 0.17 kJ
z
(
DD
G
=
D
G
in DD
S
are expected to primarily stem from rotational
ba
b
ba
molꢀ1 and ꢀ25.0 ꢂ 0.8 J molꢀ1 Kꢀ1 for 14, and ꢀ8.55 ꢂ 1.06 kJ molꢀ1
flexibility about the C5–C6ꢀ1 bond. The finding that
and ꢀ26.8ꢂ4.5 J molꢀ1 Kꢀ1 for 15 with correlation coefficients R2 of
ba
ꢀ1
DD
S
z(14) = ꢀ25.0ꢂ0.8 J mol
ba
K
is almost equal to the value
ba
of DDS z(15) = ꢀ26.8ꢂ4.5 J molꢀ1 Kꢀ1 is also in line with compara-
0.99 and 0.96, respectively (Fig. 2). As the selectivity ratio (r ) for
product 14 was higher at 25 8C than at 0 8C, this data point was not
considered in the initial analysis. This reversal in selectivity could
point toward a general change in mechanism, but in the absence of
additional data points this is purely speculative. It is important to
ble solution phase populations of the model systems 10 and 11. It is
therefore postulated that the difference in selectivity observed
between the C6-F (9) and C6-OBn (13) glycosyl donors result from
enthalpic contributions.
3. Conclusion
The selectivity of two structurally similar glucose-derived
donors differing only at the 6-position (X55F versus OBn) was
i
studied in a model glycosylation reaction with PrOH at varying
temperatures. At ꢀ80 8C the highest selectivities were observed
with
b:a ratios of 3.79:1 and 6.82:1 for donors 9 (X55F) and 13
(X55OBn), respectively. This selectivity data was then subjected to
an Eyring analysis to extrapolate the enthalpic (DD
z) and
H
ba
z) parameters associated with the transition states
entropic (DD
S
ba
that lead to the
destabilization of the
both donors, the enthalpic stabilization of the
was larger for the 6-benzyloxy system. This comparable values for
z support the findings that rotation around the C5–C6 bond
b
- or
a
-anomer. Whereas the relative entropic
-TS versus the -TS was comparable for
-TS versus the -TS
b
a
b
a
DD
S
ba
is equally hindered in both oxocarbernium ion mimics 10 and
11. This can be rationalized by invoking stabilizing hyperconju-
Fig. 2. Eyring plot of the selectivity data summarized in Table 1. Error bars are
presented as 2 ꢃ s0ba with s0ba
=
sba/r
.
gative interactions (sC–H
!
s*CꢀX; X55F or OBn). Consequently,
ba
Please cite this article in press as: N. Santschi, et al., J. Fluorine Chem. (2015), http://dx.doi.org/10.1016/j.jfluchem.2015.06.004

G Model
FLUOR-8577; No. of Pages 6
N. Santschi et al. / Journal of Fluorine Chemistry xxx (2015) xxx–xxx
5
selectivity differences likely arise from variances in enthalpic
contributions (DD
z) and may be tentatively rationalized by
electronic shielding of the planar oxocarbenium moiety. A possible
rationale for this may be attributed to the lower polarizability of
Rf (Cyhx/EtOAc 4:1) = 0.58; m.p. 66 8C; m/z (ESI) found:
H
517.2353 (M+Na)+, C30H35FO5Na requires 517.2361;
nmax (neat)/
ba
cmꢀ1 3030w, 2972w, 2867w, 1497w, 1454m, 1382w, 1357w,
1209w, 1069br, 1029br, 1013br, 912w, 736s, 698s; 1H NMR
fluorine (versus OBn) such that nucleophilic attack in the
b
-TS is
(400 MHz, CDCl3) d = 7.35–7.19 (15H, m, ar. H), 4.93 (1H, d,
enthalpically more costly for X55F than X55OBn. Intriguingly,
molecular editing at the 6-position of glucose [OBn ! F] is
J = 11 Hz, BnCH2), 4.90 (1H, d, J = 11 Hz, BnCH2), 4.83 (1H, d,
J = 11 Hz, BnCH2), 4.75 (1H, d, J = 11 Hz, BnCH2), 4.66 (1H, d,
accompanied by a reduction in
b-selectivity on account of the
J = 11 Hz, BnCH2), 4.64–4.39 (2H, m, H-C6
BnCH2), 4.44 (1H, d, J = 8 Hz, H-C1 ), 3.98 (1H, sept, J = 6 Hz, H-
C7 ), 3.83 (1H, sept, J = 6 Hz, H-C7 ), 3.79 (1H, dddd, J = 30, 10, 3,
2 Hz, H-C5 ), 3.62 (1H, t, J = 9 Hz, H-C3 ), 3.50 (1H, t, J = 9 Hz, H-
C4 ), 3.41 (1H, dddd, J = 25, 10, 4, 2 Hz, H-C5 ), 3.39 (1H, dd, J = 9,
8 Hz, H-C2 ), 1.26 (3H, d, J = 6 Hz, H-C8 ), 1.20 (3H, d, J = 6 Hz, H-
C8 ), 1.18 (3H, d, J = 6 Hz, H-C8 ), 1.14 (3H, d, J = 6 Hz, H-C8
ppm, only some characteristic
-signals are reported; 13C NMR
(101 MHz, CDCl3) = 138.9 (iPh ), 138.6 (iPh ), 138.5 (iPh ),
138.2 (iPh ), 138.1 (iPh ), 138.0 (iPh ), 128.6 (Ph), 128.6 (Ph),
b), 4.55 (1H, d, J = 11 Hz,
electronic shielding effect of this small substituent. Whilst rotamer
I is equally populated for X55F or OBn in a model system, and the
intermediate oxocarbenium ion likely resembles the 3H4 in both
scenarios, the study has demonstrated that fluorine disfavors top
b
b
a
a
b
b
b
face (
b
)-attack more strongly than OBn. Efforts to further explore
b
b
the utility of fluorine as a steering group in chemical glycosylation
are currently ongoing.
b
a
a)
a
d
a
b
b
a
a
b
4. Experimental
128.6 (Ph), 128.5 (Ph), 128.5 (Ph), 128.3 (Ph), 128.3 (Ph), 128.2 (Ph),
128.1 (Ph), 128.0 (Ph), 128.0 (Ph), 127.8 (Ph), 127.8 (Ph), 127.7 (Ph),
4.1. 2,3,4-Tri-O-benzyl-6-deoxy-6-fluoro-
trichloroacetimidate 9
a
-
D
-glucopyranosyl
102.2 (C1
b
), 95.0 (C1
a
), 84.6 (C3
b
), 82.3 (C2
), 76.8 (d, 3JCF = 6 Hz,
), 75.8 (BnCH2 and ), 75.4
), 74.9 (BnCH2 ), 73.0 (d, JCF = 18 Hz,
b), 82.2 (d,
1JCF = 174 Hz, C6
b
), 80.1 (d, 1JCF = 205 Hz, C6
a
3
C4
(BnCH2
C5 ), 73.3 (BnCH2
(C7 ), 23.8 (C8 ), 23.3 (C8
NMR (282 MHz, CDCl3)
= ꢀ232.34 (td, JFH = 48 Hz, JFH = 25 Hz
b
), 77.0 (d, JCF = 6 Hz, C4
), 75.1 (BnCH2
), 72.6 (C7
), 22.3 (C8
a
a
b
2
At 0 8C and under Ar, 0.1 mL (0.06 mmol, 0.1 eq.) of a 0.6 M
stock solution of DBU in CH2Cl2 was added to a solution of 2,3,4-tri-
a
b
b
2
b
a
a
b
), 69.9 (d, JCF = 18 Hz, C5
a), 69.4
O-benzyl-
a
/b
-D
-glucopyranoside (8) (30 mg, 0.6 mmol, 1.0 eq.)
b
a
b
), 21.3 (C8
a
) ppm; 19F
2
3
and trichloroacetonitrile (0.7 mL, 6 mmol, 10.0 eq..) in CH2Cl2
(1.2 mL). After 5 min the solution was allowed to warm to room
temperature and stirred for 2 h. The solvent was evaporated under
reduced pressure. Fast filtration over silica gel (SiO2, Cyhx/EtOAc
10:1) afforded trichloroacetimidate (9) as colorless oil (354 mg,
d
2
3
b
-anomer), ꢀ234.26 (td, JFH = 48 Hz, JFH = 30 Hz,
a-anom-
er) ppm.
Acknowledgements
99%), mainly as the
a-anomer. Rf (Cyhx/EtOAc 4:1) = 0.69; m/z
(ESI) found: 618.0984 (M+Na)+, C29H29NO5Cl3FNa requires
We acknowledge generous financial support from the Swiss
National Science Foundation (P2EZP2-148757), the DFG (SFB
858 and Excellence Cluster EXC 1003 ‘‘Cells in Motion—Cluster of
Excellence’’) and the European Research Council (ERC-2013-StG
Starter Grant. Project number 336376-ChMiFluorS).
618.0988; ½a 2D6
ꢅ
= +48 (c 1.00 in CH2Cl2); nmax (neat)/cmꢀ1
3338w, 3031w, 2871w, 1736w, 1670m, 1497w, 1454m, 1360m,
1287m, 1210w, 1156m, 1072s, 1005s, 908m, 859m, 830m, 795s,
737s, 698s; 1H NMR (400 MHz, CDCl3)
d = 8.61 (1H, s, NH), 7.40–
7.26 (15H, m, ar. H), 6.42 (1H, d, J = 4 Hz, H-C1), 4.98 (1H, d,
J = 11 Hz, BnCH2), 4.93 (1H, d, J = 11 Hz, BnCH2), 4.85 (1H, d,
J = 11 Hz, BnCH2), 4.75 (1H, d, J = 12 Hz, BnCH2), 4.69 (1H, d,
J = 13 Hz, BnCH2), 4.64 (1H, d, J = 11 Hz, BnCH2), 4.64 (1H, ddd,
J = 48, 11, 3 Hz, H-C6), 4.56 (1H, ddd, J = 48, 10, 2 Hz, H-C6), 4.08
(1H, t, J = 9 Hz, H-C3), 4.03–3.87 (1H, m, H-C5), 3.75 (1H, dd, J = 9,
3 Hz, H-C4), 3.74–3.65 (1H, m, H-C2) ppm; 13C NMR (101 MHz,
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jfluchem.2015.
06.004.
CDCl3)
d = 161.2 (C7), 138.4 (iPh), 137.8 (iPh), 137.7 (iPh), 128.5
(Ph), 128.4 (Ph), 128.4 (Ph), 128.1 (Ph), 128.0 (Ph), 128.0 (Ph), 127.8
(Ph), 127.7 (Ph), 127.7 (Ph), 94.1 (C1), 91.1 (C8), 81.4 (d,
References
3
1JCF = 174 Hz, C6), 81.1 (C3) 79.2 (C2), 75.9 (d, JCF = 6 Hz, C4),
[1] (a) For selected reviews on organofluorine chemistry see D. O’Hagan, Chem.
Soc. Rev. 37 (2008) 308–319;
2
75.7 (BnCH2), 75.5 (BnCH2), 73.0 (BnCH2), 72.6 (d, JCF = 18 Hz,
(b) S. Purser, P.R. Moore, S. Swallow, V. Gouverneur, Chem. Soc. Rev. 37 (2008)
320–330;
(c) V. Gouverneur, K. Mu¨ller (Eds.), Fluorine in Pharmaceutical and Medicinal
Chemistry, Imperial College Press, London, 2010;
(d) L.E. Zimmer, C. Sparr, R. Gilmour, Angew. Chem. Int. Ed. 50 (2011) 11860–
11871.
C5) ppm; 19F NMR (282 MHz, CDCl3)
3JFH = 30 Hz) ppm.
d
= ꢀ234.93 (td, 2JFH = 48 Hz,
4.2. 1-O-Isopropyl-2,3,4-tri-O-benzyl-6-deoxy-6-fluoro-a/b-D-
glucopyranoside 14
[2] (a) For selected reviews on glycochemistry see, Nature Insight:
Glycochemistry and Glycobiology, 446 (2007) 999–1051.
(b) N.S. Sampson, M. Mrksich, C.R. Bertozzi, Proc. Natl. Acad. Sci. U.S.A. 98
(2001) 12870–12871;
(c) C.R. Bertozzi, L.L. Kiessling, Science 291 (2001) 2357–2364;
(d) T.J. Boltje, T. Buskas, G.-J. Boons, Nat. Chem. 1 (2009) 611–622;
(e) P.H. Seeberger, Nat. Chem. Biol. 5 (2009) 368–372.
[3] For an excellent review of carbohydrate inspired glycomimetic drug
development see B. Ernst, J.L. Magnani, Nat. Rev. Drug Discovery, 8 (2009)
661–677.
[4] G.T. Giuffredi, V. Gouverneur, B. Bernet, Angew. Chem. Int. Ed. 52 (2013)
10524–10528.
[5] (a) For selected examples of fluorinated carbohydrates see I.P. Street, C.R.
Armstrong, S.G. Withers, Biochemistry, 25 (1986) 6021–6027;
(b) S.G. Withers, D.J. MacLennan, I.P. Street, Carbohydr. Res. 154 (1986) 127–
144;
˚
In a 10 mL Schlenk-tube containing 4A molecular sieves 2,3,4-
tri-O-benzyl-6-deoxy-6-fluoro- -glucopyranosyl trichloroacetimi-
D
date (9) (25 mg, 0.04 mmol, 1.0 eq.) was dissolved in CH2Cl2
(0.7 mL) under Ar. After cooling to ꢀ90 8C, 0.1 mL (0.05 mmol,
i
1.2 eq.) of an PrOH stock solution (0.5 M in CH2Cl2) was added
followed by 0.1 mL (0.005 mmol, 0.1 eq.) of a TMSOTf stock
solution (0.05 M in CH2Cl2). Upon reaction completion NEt3 was
added for quenching and the solvent was removed under reduced
pressure. The product was purified by column chromatography
(SiO2, Cyhx/EtOAc 10:1) affording 1-O-isopropyl-2,3,4-tri-O-ben-
zyl-6-deoxy-6-fluoro-
(15 mg, 76%, a:b = 1:10.7).
a/b-D-glucopyranoside (14) as a white solid
(c) M.D. Burkart, Z. Zhang, S.-C. Hung, C.-H. Wong, J. Am. Chem. Soc. 119
(1997) 11743–11746;
Please cite this article in press as: N. Santschi, et al., J. Fluorine Chem. (2015), http://dx.doi.org/10.1016/j.jfluchem.2015.06.004

G Model
FLUOR-8577; No. of Pages 6
6
N. Santschi et al. / Journal of Fluorine Chemistry xxx (2015) xxx–xxx
(d) M. Albert, B.J. Paul, K. Dax, Synlett 9 (1999) 1483–1485;
(e) T. Miyazaki, T. Sakakibara, H. Sato, Y. Kajihara, J. Am. Chem. Soc. 121
(1999) 1411–1412;
(f) A.G. Watts, I. Damager, M.L. Amaya, A. Buschiazzo, P. Alzari, A.C. Frasch,
S.G. Withers, J. Am. Chem. Soc. 125 (2003) 7532–7533;
(g) C.A. Tarling, S.G. Withers, Carbohydr. Res. 339 (2004) 2487–2497;
(h) V. Subramaniam, S.S. Gurcha, G.S. Besra, T.L. Lowary, Bioorg. Med. Chem.
13 (2005) 1083–1094;
(b) T.R. Neal, W.C. Schumann, M.S. Berridge, B.R. Landau, J. Labelled Compd.
Radiopharm. 48 (2005) 845–854.
[9] (a) For theoretical treatments of the gauche effect see L. Goodman, H. Gu, V.
Pophristic, J. Phys. Chem. A, 109 (2005) 1223–1229;
(b) P.R. Rablen, R.W. Hoffmann, D.A. Hrovat, W.T. Borden, J. Chem. Soc., Perkin
Trans. 2 8 (1999) 1719–1726;
(c) D.Y. Buissonnneaud, T. van Mourik, D. O’Hagan, Tetrahedron 66 (2010)
2196–2202;
´
(i) D. Crich, L. Li, J. Org. Chem. 72 (2007) 1681–1690;
(j) H.A. Chokhawala, H. Cao, H. Yu, X. Chen, J. Am. Chem. Soc. 129 (2007)
10630–10631;
(d) M. Baranac-Stojanovic, RSC Adv. 4 (2014) 43834–43838;
(e) For a recent study of the gauche effect in 2,3-difluorobutane see S.J. Fox, S.
Gourdain, A. Coulthurst, C. Fox, I. Kuprov, J.W. Essex, C.K. Skylaris, B. Linclau,
Chem. Eur. J. 21 (2015) 1682–1691.
`
´
´
(k) D. Benito, M.I. Matheu, A. Morere, Y. Dıaz, S. Castillon, Tetrahedron 64
(2008) 10906–10911;
[10] (a) C. Bucher, R. Gilmour, Angew. Chem. Int. Ed. 49 (2010) 8724–8728;
(b) C. Bucher, R. Gilmour, Synlett (2011) 1043–1046;
(c) E. Durantie, C. Bucher, R. Gilmour, Chem. Eur. J. 18 (2012) 8208–8215.
[11] D. Crich, Acc. Chem. Res. 43 (2010) 1144–1153.
[12] (a) R.J. Woods, C.W. Andrews, J.P. Bowen, J. Am. Chem. Soc. 114 (1992)
859–864;
(l) B.R. Rempel, S.G. Withers, Aust. J. Chem. 62 (2009) 590–599;
(m) S.A. Allman, H.H. Jensen, B. Vijayakrishnan, J.A. Garnett, E. Leon, Y. Liu,
D.C. Anthony, N.R. Sibson, T. Feizi, S. Matthews, B.G. Davis, ChemBioChem 10
(2009) 2522–2529;
(n) G.T. Giuffredi, S. Purser, M. Sawicki, A.L. Thompson, V. Gouverneur,
Tetrahedron: Asymmetry 20 (2009) 910–920;
(o) O. Boutureira, F. D’Hooge, M. Fernandez-Gonzalez, G.J.L. Bernardes, M.
Sanchez-Navarro, J.R. Koeppe, B.G. Davis, Chem. Commun. 46 (2010)
8142–8144;
(p) A. Hoffmann-Ro¨der, A. Kaiser, S. Wagner, N. Gaidzik, D. Kowalczyk, U.
Westerlind, B. Gerlitzki, E. Schmitt, H. Kunz, Angew. Chem. Int. Ed. 49 (2010)
8498–8503;
(b) J.A.C. Romero, S.A. Tabacco, K.A. Woerpel, J. Am. Chem. Soc. 122 (2000)
168–169.
[13] M. Matwiejuk, J. Thiem, Chem. Commun. 47 (2011) 8379–8381.
[14] R. Johnson, D. Olsson, U. Ellervik, J. Org. Chem. 73 (2008) 5226–5232.
[15] P. Kova´, H.J.C. Yeh, G.L. Jung, J. Carbohydr. Chem. 6 (1987) 423–439.
[16] P.A. Messina, K.C. Mange, W.J. Middleton, J. Fluorine Chem. 42 (1989)
137–143.
(q) S. Wagner, C. Mersch, A. Hoffmann-Ro¨der, Chem. Eur. J. 16 (2010) 7319–
7330;
[17] A.M. Esmurziev, N. Simic, B.H. Hoff, E. Sundby, J. Carbohydr. Chem. 29 (2010)
348–367.
(r) S. Bresciani, T. Lebl, A.M.Z. Slawin, D. O’Hagan, Chem. Commun. 46 (2010)
5434–5436;
[18] (a) For select attempted protocols consider M. Matwiejuk, J. Thiem, Eur. J. Org.
Chem. (2011) 5860–5878;
(s) A. Ioannou, E. Cini, R.S. Timofte, S.L. Flitsch, N.J. Turner, B. Linclau, Chem.
Commun. 47 (2011) 11228–11230;
(t) M.J. Corr, D. O’Hagan, J. Fluorine Chem. 155 (2013) 72–77;
(u) L. Mtashobya, L. Quiquempoix, B. Linclau, J. Fluorine Chem. 171 (2014)
92–96;
(v) K.A. van Straaten, J.R. Kuttiyatveetil, C.M. Sevrain, S.A. Villaume, J. Jimenez-
Barbero, B. Linclau, S.P. Vincent, D.A.R. Sanders, J. Am. Chem. Soc. 137 (2015)
1230–1244;
(b) M.E. Jung, M.A. Lyster, J. Org. Chem. 42 (1977) 3761–3764;
(c) A. Kumar, V.R. Doddi, Y.D. Vankar, J. Org. Chem. 73 (2008) 5993–5995.
[19] S. Halazy, V. Berges, A. Ehrhard, C. Danzin, J. Fluorine Chem. 59 (1992)
185–192.
[20] S. Chamberland, J.W. Ziller, K.A. Woerpel, J. Am. Chem. Soc. 127 (2005)
5322–5323.
[21] J. Gratien, M.-P. Heck, C. Mioskowski, Carbohydr. Res. 343 (2008) 18–30.
[22] S.Y. Kang, K.-S. Song, J. Lee, S.-H. Lee, J. Lee, Bioorg. Med. Chem. 18 (2010)
6069–6079.
[23] S. Hamman, C. Be´guin, C. Charlon, C. Luu-Duc, Org. Magn. Reson. 21 (1983)
361–366.
(w) Also see the special edition Fluoro Sugars, Carbohydr. Res. 327 (2000) 1.
[6] (a) M.N. Namchuck, J.D. McCarter, A. Becalski, T. Andrews, S.G. Withers, J. Am.
Chem. Soc. 122 (2000) 1270–1277;
(b) M.D. Percival, S.G. Withers, Biochemistry 31 (1992) 505–512.
[7] (a) T. Oberbillig, C. Mersch, S. Wagner, A. Hoffmann-Ro¨der, Chem. Commun.
48 (2012) 1487–1489;
[24] A. Dahbi, S. Hamman, C.G. Be´guin, Magn. Reson. Chem. 24 (1986) 337–342.
[25] R.J. Abraham, E.J. Chambers, W.A. Thomas, Magn. Reson. Chem. 30 (1992)
S60–S65.
(b) T. Oberbillig, H. Lo¨we, A. Hoffmann-Ro¨der, J. Flow Chem. 2 (2012) 83–86.
[8] (a) B.R. Landau, C.L. Spring-Robinson, R.F. Muzic, N. Rachdaoui, D. Rubin, M.S.
Berridge, W.C. Schumann, V. Chandramouli, T.S. Kern, R. Ismail-Beigi, Am. J.
Physiol. Endocrinol. Metab. 293 (2007) 237–245;
[26] H.H. Jensen, L.U. Nordstrøm, M. Bols, J. Am. Chem. Soc. 126 (2004) 9205–9213.
[27] H. Buschmann, H.-D. Scharf, N. Hoffmann, P. Esser, Angew. Chem. Int. Ed. 30
(1991) 477–515.
[28] r0 = exp(ꢀDDH z ꢃ (RT)ꢀ1 + DDS z ꢃ Rꢀ1).
ba
ba
ba
Please cite this article in press as: N. Santschi, et al., J. Fluorine Chem. (2015), http://dx.doi.org/10.1016/j.jfluchem.2015.06.004