Diastereoselective synthesis of lincosamine precursors

Article abstract of DOI:10.1002/ejoc.201001740

The stereoselective syntheses of the four aminodiol precursors of the diastereomers of lincosamine are reported. The procedure is based on the initial two-carbon elongationof 1,2;3,4-di-O-isopropylidene-α-D- galactohexodialdo-1,5-pyranose, followed by the stereocontrolled introduction of the amino group by nucleophilic amination. Two complementary approaches have been investigated and compared: The first one is the direct transformation of α-chloroglycidic ester into β-amino-α-keto ester. The second strategy is a three-step synthesis that is based on the treatment of the β-iodo-α-keto ester with dibenzylamine. Subsequent reduction of the β-amino-α-keto ester provides the pure D-erythro, L-threo, L-erythro, and D-threo aminodiols after chromatographic purification. Further classical transformations afford the N-acetyl derivatives, which are key precursors of the lincosamine diastereomers. Copyright

Full text of DOI:10.1002/ejoc.201001740

SHORT COMMUNICATION
DOI: 10.1002/ejoc.201001740
Diastereoselective Synthesis of Lincosamine Precursors
Fiona Serra,[a] Philippe Coutrot,[b] Mélanie Estève-Quelquejeu,[c] Patrick Herson,[c]
Tomasz K. Olszewski,[a] and Claude Grison*[a]
Keywords: Natural products / Aminodiols / Boranes / Amination / Carbohydrates
The stereoselective syntheses of the four aminodiol precur-
sors of the diastereomers of lincosamine are reported. The
procedure is based on the initial two-carbon elongation
ester into β-amino-α-keto ester. The second strategy is a
three-step synthesis that is based on the treatment of the β-
iodo-α-keto ester with dibenzylamine. Subsequent reduction
of
1,2;3,4-di-O-isopropylidene-α-D-galactohexodialdo-1,5-
of the β-amino-α-keto ester provides the pure
D-erythro, L-
pyranose, followed by the stereocontrolled introduction of
the amino group by nucleophilic amination. Two comple-
mentary approaches have been investigated and compared:
The first one is the direct transformation of α-chloroglycidic
threo, -erythro, and -threo aminodiols after chromato-
L
D
graphic purification. Further classical transformations afford
the N-acetyl derivatives, which are key precursors of the lin-
cosamine diastereomers.
suitably protected d-dialdogalactopyranose is the starting
material with subsequent sequential introduction of the C-
6 to C-8 lateral chain, or the starting substrate is a pro-
tected or masked β-hydroxyaminoaldehyde representing the
future lateral carbon chain followed by several steps leading
to the building of the pyranic ring with complete stereocon-
trol of its five chiral centers.
Introduction
Lincosamine (6-amino-6,8-dideoxy-d-erythro-d-galacto-
octopyranose, 1) represents a synthetic precursor of methyl
thiolincosaminide (2), the carbohydrate moiety of anti-
biotics lincomycin (3) and its 7-chloro analogue clindamy-
cin (4; Figure 1). Extensive research on the preparation of
3 and 4 has mainly been focused on the synthesis of 1,
which suffers from low steric control in the introduction of
two chiral centers at C-6 and C-7. The introduction of the
glycoside thiomethyl group to obtain 2 seems resolved,[1] as
is the syntheses of the proline moiety[2] and coupling with
2.[3]
Illustrating the first strategy, numerous interesting syn-
theses of lincosamine precursors from 1,2;3,4-di-O-isopro-
pylidene-α-d-galactohexodialdo-1,5-pyranose (5) are re-
ported. These methods proceed through initial cyanoamin-
ation,[4] cyanohydrin formation,[5] or nitroaldol condensa-
tion,[6] or with the use of 2-methyl-1,3-dithiane-2-yllithium
as a reagent.[7] Initial Wittig carbonylolefination of 5 has
also been used,[8] as well as its Horner variant.[1,9] Reactions
between the N-benzylaldimine derivative of 5 and Grignard
reagents[10] or the cycloaddition of methoxyacetyl chlo-
ride[11] have given rise to several other approaches. Reaction
between the nitrone derivative of 5 and 2-lithiothiazole is
another initial step.[12] Interesting sophisticated works re-
lated to the second strategy concern the application of suit-
ably protected threoninal as the starting material.[13] A total
synthesis of racemic lincosamine by [4+2] cycloaddition for
building the pyranose moiety has been proposed[14] and ap-
plied to the asymmetric synthesis of lincosamine.[13d,15]
Nevertheless, all of these methods rely upon scarce syn-
thetic precursors and multistep, tedious reaction sequences,
and therefore, there is still room for improvement, and the
search for new methods leading to 1 and its derivatives is
highly desirable.
Figure 1. Structures of lincosamine (1), methyl thiolincosaminide
(4), and antibiotics lincomycin (3) and clindamycin (4).
Previously reported approaches to lincosamine or lincos-
amine precursors depend on two main strategies. Either
[a] UMR 5175, CNRS-CEFE
19 Route de Mende, 34293 Montpellier Cedex 5, France
Fax: +33-467 61 32 44
E-mail: claude.grison@cefe.cnrs.fr
[b] UMR 7565, CNRS-Université Henri Poincaré
Nancy-I, BP 239, 54506 Vandœuvre-lès-Nancy, France
[c] Université Pierre et Marie Curie Paris 6
4 Place Jussieu, 75252 Paris, Cedex 5, France
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201001740.
We have previously reported on several applications of
dihalogenoacetate carbanions[16] and developed a general
synthesis of α-keto esters from carbonyl compounds by
using potassium alkyl dichloroacetates.[17] Applied to dial-
Eur. J. Org. Chem. 2011, 1841–1847
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1841

C. Grison et al.
SHORT COMMUNICATION
dosugar derivatives, the method is convenient for extension C-6. It clearly appears from these data that oxirane opening
of the chain with a α-keto ester unit[18] and to introduce and concomitant substitution at C-6 of chloroglycidic ester
functionality at the C-6 position.[19] In this paper we de- 6 is not a simple SN2 process. Easy fractional crystallization
scribe a novel variant of this method illustrated by the ster- in pentane allows pure diastereomers d-glycero 7 and l-gly-
eoselective synthesis of lincosamine precursors from 5.
cero 7 to be obtained, which is critical to the success of our
strategy concerning the synthesis of lincosamine and epi-
lincosamine. Single-crystal X-ray diffraction study of the
purified major isomer obtained after complete epimeriz-
Results and Discussion
Darzens condensation between isopropyl potassium ation over 50 d allowed the assignment of the l-glycero con-
dichloroacetate and readily available protected dialdose 5[19] figuration. This could be conducted further to obtain the
cleanly afforded a diastereomeric mixture of α-chlorogly- d-glycero configuration at C-6, that of C-6 of lincosamine,
cidic esters 6 (84:16, Scheme 1) in high yield (92%) and with through SN2 substitution with a suitable amino reagent. As
strong diastereoselection. The trans relationship of the ester a consequence, our first investigations concerned the direct
group and the pyranic moiety linked to the oxirane ring was substitution of the iodine atom of 7 by sodium azide or
assigned by analogy with previous results in the aliphatic dibenzylamine. With sodium azide, the sole formation of
series.[16,20] The configuration at C-6 of the major dia- acyl azide was observed without any substitution at C-6.
stereomer is believed to be d-glycero on the basis of both On the other hand, the reaction with dibenzylamine was
1H NMR spectroscopic data and mechanistic considera- difficult and needed to be heated at reflux for a longer time
tions.[18] Treatment of 6 with anhydrous magnesium iodide (3 equiv., pentane, reflux, 28 h). Under these conditions, the
in ether yielded an epimeric mixture (57:43) of 7 (95%, total conversion of 7 into 8 was obtained in moderate yield
1
Scheme 1). Both epimers were easily characterized by H after purification (42–54%). As previously reported, the
NMR spectroscopy. In the major epimer, H-1 appears as a very hindered nature of C-6 and the difficulty of introduc-
doublet at δ = 5.41 ppm (3JH1,H2 = 4.7 Hz), whereas in the ing the amino group in this position on lincosamine precur-
minor epimer the H-1 resonance shifts toward a higher field sors also came to be true in the presented case.[1–15] The
at δ = 5.62 ppm (3JH1,H2 = 4.7 Hz). It can be noted that the fragility of product 8 also needed to be taken into account.
epimeric ratio of purified product 7 varies in time and after Simple washing with dilute aqueous NaOH and subsequent
50 d at 4 °C the minor stereoisomer becomes the major fast chromatographic purification or rapid direct purifica-
product (11:89), indicating a configurational instability at tion of the reaction mixture by silica gel chromatography
Scheme 1. General presentation of the synthetic approach.
1842
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 1841–1847

Diastereoselective Synthesis of Lincosamine Precursors
were the best conditions to limit product degradation. Sur- stereoselectivity observed could therefore also result from
prisingly, the reaction was 100% stereoselective in favor of kinetic resolution, where 7-d-glycero would react faster than
sole epimer 8-l-glycero [which presented the (R) configura- 7-l-glycero with Bn2NH. The least-stable epimer would be
tion at C-6] whatever the ratios of epimers present in start- more reactive. Then, the Walden inversion would explain
ing 7, as exemplified by two significant attempts (Scheme 2, why a majority of 8-l-glycero is obtained. This phenome-
Table 1).
non, and partial in situ epimerization of amino ester 8-d-
glycero into 8-l-glycero, which is thermodynamically more
stable, would compete for the almost exclusive formation of
8-l-glycero.
It was noted that 100% diastereoselectivity of the reac-
tion of 7 with Bn2NH leading to sole epimer 8-l-glycero
is only interesting for further synthesis of protected 6-epi-
lincosamine14-d-threo or 14-l-erythro. Thus, this substitu-
tion does not allow access to protected lincosamine 14-d-
erythro, which is the active epimer. Thus, another approach
was studied based on direct nucleophilic attack of 6 by
Bn2NH with the aim to obtain major epimer 8-d-glycero
(Scheme 3, Table 2).[22–24] An epimeric mixture of 8-d-gly-
cero/8-l-glycero (51:49) was so obtained from 6 (79:21), af-
ter 32 h in refluxing toluene (77% yield, Scheme 3). Conse-
quently, this reaction appeared more interesting than the
previous one in terms of profitability, being clearly less
stereoselective in favor of epimer 8-l-glycero. However, in
this case also, it was noted over a continuous period of time
that there was a tendency for the epimeric mixture 8 to iso-
merize to the more stable epimer 8-l-glycero up to an equi-
Scheme 2. Reaction β-iodo-α-keto ester 7 with dibenzylamine.
Table 1. Reaction of β-iodo-α-keto ester 7 with dibenzylamine.
Entry
Epimeric ratio[a]
7-l-glycero/7-d-glycero 8-l-glycero/8-d-glycero
Epimeric ratio[a]
Yield
[%][b]
1
2
70:30
25:75
99:1
100:0
54
42
[a] Determined by 1H NMR spectroscopic analysis of the crude
reaction mixture. [b] Yield of isolated product.
In fact, the absolute configuration at C-6 of epimers 8 librium ratio of 8-d-glycero/8-l-glycero = 28:72 after 40 h,
remained equivocal at this stage but was proved by de- as illustrated in Table 2. So, if the target is protected lincos-
termining the structure of aminodiol 9-d-threo derived from amine 14-d-erythro, the conditions of the Entry 5 in Table 2
1
8, as will be described later on. In the H NMR spectrum, should be chosen because they represent the best compro-
the H-1 signal in the 8-l-glycero epimer appears as a doub- mise between the rate of conversion of starting ester 6 and
let at δ = 5.69 ppm (3JH1,H2 = 5.3 Hz), whereas in the 8-d- the amount of epimer 8-d-glycero produced. After chroma-
glycero epimer the H-1 doublet shifts toward a lower field tographic purification, pure epimer 8-d-glycero could be ob-
at δ = 5.37 ppm (3JH1,H2 = 4.8 Hz).[21] A first explanation
of the stereoselectivity observed in favor of the sole epimer
8-l-glycero was a quasitotal epimerization of 8-d-glycero oc-
curring in the reaction medium. In order to test this hy-
pothesis, a study of the stability of several mixtures of epi-
mers 8 in different ratios was undertaken in refluxing tolu-
ene over time. Whatever the composition of the mixture, a
continuous variation was observed in favor of 8-l-glycero
up to a ratio of 75:25 after a 40-h reaction time; this ratio
Scheme 3. Reaction of 6 with dibenzylamine.
never changed afterwards. This result also confirms the
known natural trend of lincosamine precursors to undergo
epimerization at C-6 in favor of the l-glycero configuration
(see above epimerization of β-iodo-α-keto ester 7).[5,10,21]
However, such a process alone does not account for the
total formation of the 8-l-glycero epimer in the substitution
Table 2. Reaction of α-chloroglycidic ester 6 with dibenzylamine in
refluxing toluene over time.
of 7 by Bn2NH, and another phenomenon is probably in- Entry Time
volved. As the reaction most likely follows an SN2 mecha-
Yield [%][b]
Epimeric ratio[c]
[h]
6[a] 8-d-glycero 8-l-glycero 8-d-glycero/8-l-glycero
nism, the amounts of epimers 8 produced should reflect the
reverse composition of the starting diastereomeric mixture
of 7 or lead to relatively more 8-d-glycero than expected,
given the natural trend of 7-d-glycero to partially undergo
epimerization into 7-l-glycero in the reaction medium. This
is never the case, as the reaction of 7 with Bn2NH always
yields predominantly or exclusively 8-l-glycero, regardless
1
2
3
4
5
6
3
8
22
26
32
40
74
47
17
15
9
20
40
54
50
47
28
6
77:23
75:25
65:35
59:41
51:49
28:72
13
29
35
44
72
0
[a] Starting substrate epimeric ratio 6-d-glycero/6-l-glycero was
79:21. [b] Yield of purified product. [c] Determined by 1H NMR
of the composition of starting epimeric mixture 7. The total spectroscopic analysis of the crude reaction mixture.
Eur. J. Org. Chem. 2011, 1841–1847
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1843

C. Grison et al.
SHORT COMMUNICATION
tained, but purification was rather difficult.[25]Subsequently, of the relative difficulty in obtaining pure 8-d-glycero, the
the stereoselective reduction of pure 8-d-glycero and pure reaction with LiAlH4 was also studied for a mixture of ste-
8-l-glycero into aminodiols 9 was studied (Scheme 4).
reoisomers 8, enriched in 8-d-glycero (Scheme 5). The re-
duction was much faster than that with 8-l-glycero. The re-
action was complete after 4 h and yielded a mixture of ami-
1
nodiols 9 with good stereoselectivity (70%de). The H and
13C NMR spectroscopic data confirmed that that the major
stereoisomer present in the product was not 9-d-threo. As
it was not possible to obtain crystals, this major stereoiso-
mer was further transformed into final compound 14-d-
erythro, which was assigned the 9-d-erythro configuration
(see further sections of the text). Surprisingly, the (R)-C-6
of 8-l-glycero induced an (S) configuration at C-7 in 9-d-
threo, whereas reduction of epimer 8-d-glycero afforded the
same (S) configuration at C-7 in 9-d-erythro with a closely
related degree of stereoselectivity. These results demonstrate
that the observed selectivity is not the result of only 1,2-
stereoinduction with a dominant effect induced by the C-
6 stereocenter. We suppose that these results rather reflect
concomitant 1,2- and 1,3-asymmetric induction, where the
dominant effect is 1,3-induction. The data are consistent
with the works of Evans et al., where the stereogenic center
proximal to bond formation is not always the dominating
stereochemical determinant governing carbonyl facial in-
duction.[26] Caution must be exercised when extending ste-
reochemical models based on transformations of simple
substrates to polyfunctional complex systems such as gluc-
idic derivatives. So, it is difficult to consider only the Ev-
ans’s model to rationalize the stereochemical impact of the
β-stereocenter in dictating the sense of ketone facial selec-
tivity.[27] The influence of electronic and steric effects of the
nucleophiles can clearly be important stereochemical pa-
rameters and should be considered in reduction reactions.
Consequently, other reducing agents have been studied
(Table 3).
Scheme 4. Reduction of β-dibenzylamino-α-keto ester
aminodiols 9.
8 into
In previous work, we described the chemoselective re-
duction of different glycosyl α-keto esters into the corre-
sponding glycosyl α-hydroxy esters using various chiral or
achiral reagents and Baker’s yeast. The diastereomeric ex-
cess could exceed 98% in the galacto series.[23a] Conse-
quently, two possibilities were considered for obtaining 9:
either direct reduction of the aminopyruvic moiety of 8 into
aminodiol 9 (route A) or sequential reduction of 8 into ami-
nodiol 9 via α-hydroxy ester intermediate 10 (route B). For
route A (Scheme 4), the reduction of pure stereoisomer 8-
l-glycero with an excess amount of LiAlH4 (2 equiv. in
ether) was complete after 24 h at 25 °C and afforded a mix-
ture of stereoisomers 9 in excellent yield (95%) and with
good diastereomeric excess (76%de, Scheme 5). The reac-
tion was easily monitored by thin-layer chromatography
and revealed clearly the formation of intermediate 10,
For route B (Scheme 4), the reactions of sodium borohy-
which disappeared over time. Pure major stereoisomer was dride, in the presence or in the absence of cerium chloride,
obtained as crystals after chromatographic purification. X- sodium cyanoborohydride, n- or s-tributylborohydride, and
ray analysis revealed the 9-d-threo configuration. Because catecholborane, with or without (R)- or (S)-oxazaborolid-
Scheme 5. Reduction of β-dibenzylamino-α-keto ester 8 into aminodiols 9 by means of LiAlH4.
1844
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 1841–1847

Diastereoselective Synthesis of Lincosamine Precursors
ine, were studied. All these reagents selectively reduced the Entries 7–10). This surprising decrease in reactivity in the
α-keto moiety of 8 into corresponding α-hydroxy ester 10 presence of this activating agent certainly results from diffi-
with variable efficiency (33–91% yield) according to the na- cult formation of the Lewis acid–base complex between the
ture of the reagent (Table 3).
chiral ate complex catecholborane–oxazaborolidine and
All borohydrides reduced 8 into 10 (67–91% yield) under 8.[28,29] This hypothesis is confirmed by the stereochemical
mild conditions (–78 to 25 °C). The reaction with results where no influence of (R)- or (S)-oxazaborolidine
NaBH3CN was noted (Table 3, Entry 3). Although selective can be observed compared with the reaction using cate-
reduction of the α-keto moiety of glucidic α-keto esters was cholborane as a reducing agent. The absolute configura-
known to be efficient without any acidic activating addi- tions of C-6 and C-7 in stereoisomers 10 were determined
tive,[23a] it is noteworthy that in the presented case the re- by chemical correlation as a result of their stereospecific
duction required a slightly acidic pH as a result of the less reduction into the corresponding aminodiols 9 of known
reactive α-keto moiety due to the presence of the dibenz- configuration (see above) with LiAlH4 (2 equiv. in ether)
ylamino group at the β-site. Catecholborane presented a after 10 h at 25 °C (86–88% yield, Ͼ98:2de).
rather good efficiency (50–80% yield), whereas the re-
Having synthesized, in a selective fashion, all four pos-
duction with oxazaborolidine was more difficult (Table 3, sible diastereomers of aminodiol 9, that is, d-erythro, l-
Table 3. Diastereoselective reduction of β-dibenzylamino-α-keto ester 8 into β-dibenzylamino-α-hydroxy ester 10 using different reducing
agents.
Entry
8[a]
Reducing agent (equiv.)
Conditions
10[b]
Yield
[%][c]
de
[%][b]
1
2
3
4
5
6
7
8
9
10
8-l-glycero
8-l-glycero
8-l-glycero
8-l-glycero
8-l-glycero
8-l-glycero
8-l-glycero
8-l-glycero
8-l-glycero
8-d-glycero
NaBH4 (1.7)
NaBH4, CeCl3 (2:1)
NaBH3CN (2.0)
LisBu3BH (1.1)
Zn(BH4)2 (2.0)
iPrOH, 2 h, 25 °C
MeOH, 3 h, –10 °C; 13 h, 20 °C
iPrOH, 3 h, 25 °C, pH 5.5
THF, 1.5 h, –78 °C
d-threo/l-erythro, 93:7
d-threo[a]
86
91
80
88
67
80
41
33
50
80
86
Ͼ98
Ͼ98
Ͼ98
90
d-threo[a]
d-threo[a]
Et2O, 3.5 h, –78 °C
CH2Cl2, 22 h, 25 °C
CH2Cl2, 50 h, 25 °C
CH2Cl2, 50 h, 25 °C
CH2Cl2, 30 h, 25 °C
CH2Cl2, 14 h, 25 °C
d-threo/l-erythro, 95:5
d-threo/l-erythro, 95:5
l-erythro[a]
NBu4BH4 (1.7)
90
catecholborane-(R) oxazaborolidine (7)
catecholborane-(S) oxazaborolidine (7)
catecholborane (5.2)
Ͼ98
Ͼ98
Ͼ98
Ͼ98
l-erythro[a]
l-erythro[a]
catecholborane (2.0)
d-erythro[a]
[a] Only one diastereomer was detected by 1H NMR spectroscopy. [b] Determined by 1H NMR spectroscopic analysis of the crude
reaction mixture. [c] Yield of purified product.
Scheme 6. Final reaction sequence leading to target lincosamine 14-d-erythro, the active epimer, and its C-6 epimer 14-d-threo.
Eur. J. Org. Chem. 2011, 1841–1847
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1845

C. Grison et al.
SHORT COMMUNICATION
threo (740 mg, 1.53 mmol) using the general hydrogenation–acety-
lation protocol described above (40 bars of H2 for 18 h) to obtain
intermediate 13-d-threo (465 mg, 100%). Next, 13-d-threo (465 mg,
1.53 mmol) was acetylated to afford 14-d-threo (431 mg, 81%) as a
white solid. M.p. 165–166 °C (ref.[31] 163–165 °C). Rf = 0.42
(EtOAc/MeOH, 19:1). [α]2D3 = –53.3 (c = 1.05, CHCl3) {ref.[31]
threo, l-erythro, and d-threo, we moved to the synthesis of
target 14-d-erythro and its C-6 epimer. For that purpose,
aminodiols 9-d-threo and 9-d-erythro were used as sub-
strates in the protocol presented in Scheme 6. First, both 9-
d-threo and 9-d-erythro were tosylated using standard con-
ditions to afford desired tosylated intermediates 11-d-threo
and 11-d-erythro in acceptable yields. Those compounds
were found to be very unstable, and therefore, they were
immediately reduced with LiAlH4 to produce desired prod-
1
[α]2D3 = –59.0 (c = 5.3, CHCl3)}. H NMR (400 MHz, CDCl3): δ =
6.13 (d, JH9,H6 = 6.7 Hz, 1 H, H9), 5.29 (d, JH1,H2 = 5.0 Hz, 1 H,
H1), 4.62 (dd, JH3,H4 = 8.0 Hz, JH3,H2 = 2.2 Hz, 1 H, H3), 4.38
(dd, JH4,H3 = 8.0 Hz, JH4,H5 = 1.5 Hz, 1 H, H4), 4.29 (dd, JH2,H1
ucts 12-d-threo and 12-d-erythro in good overall yields. = 4.9 Hz, JH2,H3 = 2.2 Hz, 1 H, H2), 4.21 (dd, JH5,H6 = 6.4 Hz,
JH5,H4 = 1.5 Hz, 1 H, H5), 4.17 (dd, JH7,H8 = 6.6 Hz, JH7,H6
=
Subsequent hydrogenation of compounds 12-d-threo and
12-d-erythro resulted in quantitative formation of desired
debenzylated intermediates 13, which were directly pro-
tected by acetylation to produce the final target com-
pounds, that is, 14-d-erythro and its C-6 epimer 14-d-threo,
in good yields.
2.3 Hz, 1 H, H7), 3.78–3.73 (m, 1 H, H6), 2.05 (s, 3 H, H11), 1.54
(s, 3 H, CH3), 1.46 (s, 3 H, CH3), 1.34 (s, 3 H, CH3), 1.32 (s, 3 H,
CH3), 1.18 (d, JH8,H7 = 6.5 Hz, 3 H, H8) ppm. 13C NMR
(100 MHz, CDCl3): δ = 171.5 (1 C, C10), 109.3, 109.1 (2 C, C12,
C13), 96.5 (1 C, C1), 71.0 (1 C, C3), 70.9 (1 C, C2), 70.7 (1 C, C4),
66.0 (1 C, C5), 64.9 (1 C, C7), 56.6 (1 C, C6), 26.0, 25.9, 25.1, 24.1
(4 C, CH3*4), 23.3 (1 C, C11), 20.4 (1 C, C8) ppm. IR (disc, KBr):
ν = 3600–3050 (OH), 1651 (amide), 1556 (NH amide), 1456, 1381,
˜
1373 [C(CH3)2] cm–1. HRMS: calcd. for C16H28NO7 [M + H]+
346.1866; found 346.1859.
Conclusions
In conclusion, we have reported here a convenient and
stereoselective preparation of 14-d-erythro, the active epi-
mer, and its C-6 epimer 14-d-threo by using aminodiol 9 as
a key chiral intermediate. The stereochemistry of com-
pounds 14-d-erythro, 14-d-threo, and 9-l-erythro was con-
firmed by X-ray analysis.[30] All four possible diastereomers
of aminodiol 9, that is, d-erythro, l-threo, l-erythro, and d-
threo, were successfully prepared by stereoselective re-
duction of corresponding α-keto esters 8 with either cate-
cholborane or NaBH3CN followed by reduction of the re-
sulting α-hydroxy esters 10 to desired aminodiols 9 with
LiAlH4.
The route presented here constitutes an important syn-
thetic potential for the rapid and completely stereoselective
preparation of lincosamine 14-d-erythro and its epimers,
which, without a doubt, can find application as useful
building blocks in the construction of lincosamine-based
antibiotics.
6-N-Acetylamino-6,8-didesoxy-1,2:3,4-di-O-isopropylidene-
D-
erythro-α- -galacto-1,5-octopyranose (14- -erythro): Prepared from
D
D
12-d-erythro (157 mg, 0.32 mmol) using the general hydrogenation–
acetylation protocol described above (1 bar of H2 for 15 h) to ob-
tain intermediate 13-d-erythro (98 mg, 100%). Next, 13-d-erythro
(98 mg, 0.32 mmol) was acetylated to afford 14-d-erythro (91 mg,
82%) as a white solid. M.p. 162–163 °C (ref.[31] 164–167 °C). Rf =
0.25 (EtOAc/MeOH, 19:1). [α]2D3 = –51.0 (c = 0.44, CHCl3) {ref.[31]
[α]2D3 = –52.9 (c = 2.25, CHCl3)}. H NMR (400 MHz, CDCl3): δ
1
= 6.48 (d, JH9,H6 = 7.1 Hz, 1 H, H9), 5.53 (d, JH1,H2 = 5.1 Hz, 1
H, H1), 4.61 (dd, JH3,H4 = 8.1 Hz, JH3,H2 = 2.3 Hz, 1 H, H3), 4.45
(dd, JH4,H3 = 8.1 Hz, JH4,H5 = 1.3 Hz, 1 H, H4), 4.30 (dd, JH2,H1
= 5.1 Hz, JH2,H3 = 2.3 Hz, 1 H, H2), 4.17–4.10 (m, 2 H, H5, H6),
4.07–4.00 (m, 1 H, H7), 1.99 (s, 3 H, H11), 1.52 (s, 3 H, CH3), 1.50
(s, 3 H, CH3), 1.35 (s, 3 H, CH3), 1.32 (s, 3 H, CH3), 1.25 (d, JH8,H7
= 6.6 Hz, 3 H, H8) ppm. 13C NMR (100 MHz, CDCl3): δ = 170.6
(1 C, C10), 109.4, 108.9 (2 C, C12, C13), 96.7 (1 C, C1), 72.1 (1 C,
C4), 71.0 (1 C, C3), 70.6 (1 C, C2), 68.4 (1 C, C7), 65.6 (1 C, C5),
57.0 (1 C, C6), 26.0, 25.9, 25.0, 24.2 (4 C, CH3*4), 23.5 (1 C, C11),
20.1 (1 C, C8) ppm. IR (disc, KBr): ν
= 3600–3000 (OH), 1660
˜
max
(amide), 1531 (NH amide), 1460, 1381 [C(CH3)2] cm–1. HRMS:
calcd. for C16H27NO7 [M]+ 345.1788; found 345.1788.
Experimental Section
General Procedure for the Hydrogenation–Acetylation of 12 Leading
to Products 14: To a solution of compound 12 in MeOH (22.0 mL)
was added Pd/C (10%), and the resulting reaction mixture was
stirred under an atmosphere of H2 for the required period of time
(see below). After completion of the reaction, the resulting mixture
was filtered through a pad of Celite, and the filtrate was concen-
trated under reduced pressure to afford 13 with a free amino group.
Those intermediates were unstable and, therefore, were immediately
acetylated using the following procedure. To a solution of crude
intermediate 13 (1 equiv.) in a mixture of dioxane (9.0 mL) and
H2O (11.0 mL) was added NaHCO3 (2.5 equiv.) and acetic anhy-
dride (2 equiv.). The resulting reaction mixture was vigorously
stirred for 1 h at room temperature and then the excess amount
of acetic anhydride was removed by co-evaporation with toluene.
Resulting crude acetylated products 14 were purified by column
chromatography.
Supporting Information (see footnote on the first page of this arti-
cle): General methods, experimental details, and characterization
data.
[1] a) S. Knapp, P. J. Kukkola, J. Org. Chem. 1990, 55, 1632–1636;
b) G. B. Howarth, W. A. Szarek, J. K. N. Jones, J. Chem. Soc.
C 1970, 2218–2224 and references cited therein.
[2] a) G. Slomp, F. A. Mackeller, J. Am. Chem. Soc. 1967, 89,
2454–2459; b) B. J. Magerlein, R. D. Birkenmeyer, R. R. Herr,
F. Kagan, J. Am. Chem. Soc. 1967, 89, 2459–2464.
[3] B. J. Magerlein, Tetrahedron Lett. 1970, 11, 33–36.
[4] a) S. Czernecki, J. M. Valéry, Carbohydr. Res. 1988, 184, 121–
130; b) M. A. Fontes Prado, J. Alves, A. Braga de Oliveira, J.
Dias de Souza Filho, Synth. Commun. 1996, 26, 1015–1022.
[5] H. Saeki, E. Ohki, Chem. Pharm. Bull. 1970, 18, 789–802.
[6] a) G. B. Howarth, D. G. Lance, W. A. Szarek, Can. J. Chem.
1969, 47, 75–79; b) G. R. Woolard, E. B. Rathbone, W. A. Sza-
rek, J. K. N. Jones, J. Chem. Soc. Perkin Trans. 1 1976, 950–
954.
6-N-Acetylamino-6,8-didesoxy-1,2:3,4-di-O-isopropylidene-
D-threo-
α- -galacto-1,5-octopyranose (14- -threo): Prepared from 12-d-
D
D
1846
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 1841–1847

Diastereoselective Synthesis of Lincosamine Precursors
[7]
[8]
[9]
[17] P. Coutrot, C. Legris, Synthesis 1975, 118–120.
[18] P. Coutrot, C. Grison, M. Tabyaoui, S. Czernecki, J. M. Valéry,
J. Chem. Soc., Chem. Commun. 1988, 1515–1516.
[19] a) O. T. Schmidt in Methods in Carbohydrate Chemistry, Aca-
demic Press, New York, 1963, vol. 2, pp. 318–319; b) A. J. Man-
cuso, S. L. Huang, D. Swern, J. Org. Chem. 1978, 43, 2480–
2482.
A. Gateau-Olesker, G. Sepulchre, G. Vass, S. D. Géro, Tetrahe-
dron 1977, 33, 393–397.
G. B. Hawarth, W. A. Szarek, J. K. N. Jones, J. Chem. Soc. C
1969, 1339–1340.
a) S. Danishefsky, M. P. DeNinno, J. Org. Chem. 1987, 52,
3171–3173; b) J. Gonda, E. Zavackà, M. Budesinsky, I. Cisa-
rovà, J. Podlaha, Tetrahedron Lett. 2000, 41, 525–529.
a) T. Fukumaru, T. Atsumi, T. Ogawa, M. Matsui, Agric. Biol.
Chem. 1973, 37, 2617–2619; b) T. Atsumi, T. Fukumaru, T.
Ogawa, M. Matsui, Agric. Biol. Chem. 1973, 37, 2621–2626; c)
T. Atsumi, T. Fukumaru, M. Matsui, Agric. Biol. Chem. 1973,
37, 2627–2630; d) F. L. van Delft, M. de Kort, G. A.
van der Marel, J. H. van Boom, J. Org. Chem. 1996, 61, 1883–
1885; e) F. L. van Delft, M. de Kort, G. A. van der Marel, J. H.
van Boom, Tetrahedron: Asymmetry 1994, 5, 2261–2264.
[20] J. Villieras, B. Castro, N. Ferracutti, Bull. Soc. Chim. Fr. 1970,
[10]
1450.
[21] R. V. Hoffman, M. C. Johnson, J. F. Okonya, J. Org. Chem.
1997, 62, 2458–2465.
[22] a) P. Coutrot, S. Dumarçay, C. Finance, M. Tabyaoui, B. Ta-
byaoui, C. Grison, Bioorg. Med. Chem. Lett. 1999, 9, 949–952;
b) P. Coutrot, C. Grison, M. Tabyaoui, B. Tabyaoui, S. Duma-
rçay, Synlett 1999, 792–794.
[23] a) C. Grison, F. Coutrot, P. Coutrot, Tetrahedron 2002, 58,
2735–2741; b) C. Grison, F. Coutrot, P. Coutrot in Peptides
2000 (Eds.: J. Martinez, J. A. Fehrentz), EDK, Paris, 2001, pp.
135–136; c) C. Grison, F. Coutrot, C. Comoy, C. Le Milbeau,
P. Coutrot, Tetrahedron Lett. 2000, 41, 6571–6574.
[24] J. Villieras, N. Ferracutti, J. C. Combret, C. R. Acad. Sci. C
1970, 270, 2083–2085.
[25] We thank Dr Mélanie Etève-Quelquejeu, UMR CNRS 7743,
University Paris 6, for HPLC and preparative thin-layer
chromatography of pure 8-d-glycero (773 mg).
[26] D. A. Evans, M. J. Dart, J. L. Duffy, M. G. Yang, J. Am. Chem.
Soc. 1996, 118, 4322–4343.
[27] J. Seyden-Penne in Réductions par les alumino- et borohydrures
en synthèse organique, Technique et documentation, Lavoisier,
Paris, 1988.
[28] E. J. Corey, C. J. Helal, Angew. Chem. 1998, 110, 2092; Angew.
Chem. Int. Ed. 1998, 37, 1986–2012.
[29] W. Harb, M. F. Ruiz-Lopez, F. Coutrot, C. Grison, P. Coutrot,
J. Am. Chem. Soc. 2004, 126, 6996–7008.
[11]
A. K. Bose, C. Mathur, D. R. Wagle, R. Naqvi, M. Manhas,
Heterocycles 1994, 39, 491–496.
[12] a) A. Dondoni, S. Franco, F. L. Merchan, P. Merino, T. Tejero,
Synlett 1993, 78–80; b) P. Merino, T. Tejero, Molecules 1999,
4, 169–179.
[13] a) J. A. Marshall, S. J. Beaudouin, J. Org. Chem. 1996, 61, 581–
586; b) B. Szechner, O. Achmatowicz, J. Carbohydr. Chem.
1992, 11, 401–406; c) T. L. Ho, S. G. Sapp, Synth. Commun.
1983, 13, 207–211; d) A. Golebiowski, T. Jurczak, Tetrahedron
1991, 47, 1037–1044.
[14] a) S. J. Danishefsky, M. DeNinno, Angew. Chem. 1987, 99, 15;
Angew. Chem. Int. Ed. Engl. 1987, 26, 15–23; b) S. J. Danishef-
sky, E. Larson, J. P. Springer, J. Am. Chem. Soc. 1985, 107,
1274–1280; c) E. Larson, S. J. Danishefsky, J. Am. Chem. Soc.
1983, 105, 6715–6716.
[15] a) J. Jurczak, A. Golebiowski in Antibiotics and Antiviral Com-
pounds: Chemical Synthesis and Modification (Eds: K. Krohn,
H. A. Kirst, H. Maag), VCH, Weinheim, 1993, pp. 353–357; b)
A. Golebiowski, J. Jurczak, Tetrahedron 1991, 47, 1045–1052;
c) A. Golebiowski, J. Raczko, U. Jacobsson, J. Jurczak, Tetra-
hedron 1991, 47, 1053–1064.
[30] CCDC-772930 (for 9-l-erythro), -772931 (for 14-d-threo), and
-772933 (for 14-d-erythro) contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
[31] T. Moriwake, S. Hamano, D. Miki, S. Saito, S. Torii, Chem.
Lett. 1986, 815–818.
[16] a) P. Coutrot, A. El Gadi, J. Organomet. Chem. 1985, 280, C11;
b) P. Coutrot, A. El Gadi, Synthesis 1984, 115–117; c) P. Co-
utrot, Bull. Soc. Chim. Fr. 1974, 1965; d) P. Coutrot, C. Legris,
J. Villieras, Bull. Soc. Chim. Fr. 1974, 1971; e) P. Coutrot, Bull.
Soc. Chim. Fr. 1974, 1965; f) P. Coutrot, C. Legris, J. Villieras,
P. Coutrot, J. C. Combret, C. R. Acad. Sci. 1970, 270, 1250.
Received: December 29, 2010
Published Online: February 25, 2011
Eur. J. Org. Chem. 2011, 1841–1847
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1847