Asymmetrized tris(hydroxymethyl)methane as a precursor of N- And O-containing 6-membered heterocycles through ring-closing metathesis

Article abstract of DOI:10.1039/b502952j

A novel synthetic application of asymmetrized tris(hydroxymethyl)methane (THYM*) 1, obtained in both enantiomeric forms in high e.e. via a chemoenzymatic procedure, is described. Starting from the common precursor 3, N- and O-containing 6-membered heteroc

Full text of DOI:10.1039/b502952j

View Article Online / Journal Homepage / Table of Contents for this issue
A R T I C L E
Asymmetrized tris(hydroxymethyl)methane as a precursor of N- and
O-containing 6-membered heterocycles through ring-closing
metathesis†
Luca Banfi, Giuseppe Guanti, Monica Paravidino and Renata Riva*
Dipartimento di Chimica e Chimica Industriale, Via Dodecaneso 31, I-16146, Genova, Italy.
E-mail: riva@chimica.unige.it
Received 28th February 2005, Accepted 14th March 2005
First published as an Advance Article on the web 1st April 2005
A novel synthetic application of asymmetrized tris(hydroxymethyl)methane (THYM*) 1, obtained in both
enantiomeric forms in high e.e. via a chemoenzymatic procedure, is described. Starting from the common precursor 3,
N- and O-containing 6-membered heterocycles have been prepared exploiting ring-closing metathesis as the key step.
Possible elaborations of the double bond in 6 and 28 have been explored and, in the case of 28, conversion into the
glycosidase inhibitor isofagomine 53 has been achieved.
of this class. Branched derivatives are clearly more difficult to
Introduction
prepare in enantiomerically pure form, since they can not be
The synthesis of O- and N- containing heterocycles in enantiop-
easily derived from natural sugars. The synthesis of branched
ure form is still an important goal for organic chemists, since
N- or O-heterocycles in enantiomerically pure form through
compounds bearing this basic structure can be exploited as
RCM requires an optically active acyclic precursor. We have
precursors of important biologically active molecules. Among
them, polyhydroxylated non-aromatic heterocycles have been
previously described a new polyfunctionalized branched chiral
building block, asymmetrized tris(hydroxymethyl)methane 1
often used as enzyme inhibitors (in particular against glycosi-
(THYM*),23 that is particularly well suited for the preparation
dases and glycosyltransferases).1,2 This inhibition is of great
of several acyclic polyoxygenated compounds. In this paper we
importance for the development of compounds displaying
will describe its use, in conjunction with RCM, in the efficient
therapeutic activity against inflammatory diseases,2 diabetes,3
assembly of branched N- or O-heterocycles that can be in turn
viral infections such as HIV4 and influenza,5 and cancer (both
employed in the synthesis of polyhydroxylated glycomimetics.
in the stimulation of immune system against tumoural cell
proliferation and in contrasting metastasis formation).6 While
O-heterocycles can be exploited for the preparation of modified
Results and discussion
sugars, functionalized N-containing cyclic compounds can be
converted into iminosugars, a family of small molecules that
mimic the cyclic alkoxycarbenium-like transition state occurring
during the glycosidic bond cleavage.
An impressive number of syntheses of these derivatives has
been reported in recent years. In these synthetic approaches, vari-
ous methods have been employed for the formation of the hetero-
cyclic ring, including lactamization,7 Mitsunobu’s cyclization,8
isoxazoline elaboration,9 intramolecular amidomercuration10
(for N-heterocycles) and hetero Diels–Alder reactions11 for O-
heterocycles.
In this project we planned to use monoacetate 3 as chiral building
block (Scheme 1). This compound is a synthetic equivalent
of asymmetrized tris(hydroxymethyl)methane 1 (THYM*) or
of the corresponding aldehyde, bis(hydroxymethyl)acetaldehyde
2 (BHYMA*). Both enantiomers of monoacetate 3 can
be produced on a multigram scale, by two complementary
chemoenzymatic procedures: the R enantiomer in 98% e.e. by
monoacylation of the corresponding diol catalyzed by lipase
from porcine pancreas (PPL) supported on Celite,24 and the S
enantiomer in 97% e.e. by monohydrolysis of the corresponding
diacetate catalyzed by commercially available PPL.25
However, one of the most useful methods for generating
partially saturated heterocycles with different ring sizes is ring-
closing metathesis (RCM), since the endo double bond resulting
from the cyclization step can be functionalized in many different
ways, with high diastereoselection in some cases. Through
this procedure12 5-,13,14 6-,14,15 7-16,17 and even 8-membered16,18
partially saturated N-heterocycles are readily availabe. Tandem
stereoselective ring-opening metathesis – ring-closing metathesis
(ROM–RCM) has been employed for 6-membered ring forma-
tion during the synthesis of swainsonine analogues.19 Finally,
RCM has been also successfully applied to the synthesis of 5-,20,21
6-,21,22 and 7-membered partially saturated O-heterocycles.20
Many of the glycomimetics of this type prepared so far have
not been branched: the carbon atoms have all been included in
the ring and the hydroxy groups have all been directly bonded
to the ring. However, branched analogues, where at least one
of the hydroxy groups is on a side arm, are very interesting, as
demonstrated by the biological activity of some natural members
Scheme 1
In actual fact, the double bond of 3 behaves as a masked
aldehyde, since this group can be restored through a stereo-
conservative ozonolysis/reduction, and can be further reduced
to the corresponding alcohol. THYM* and BHYMA* have
been previously submitted to several transformations involving:
a) the substitution of one oxygenated moiety with a suitable
nucleophile without the formation of new stereocentres;23 b) the
stereoselective functionalization of one or two branches with
the creation of new stereogenic centres;23 c) the stereoselective
elaboration, which has also been performed intramolecularly,26
of suitable alkenes obtained after olefination of BHYMA*.23 In
all these applications the double bond of 3 was always cleaved
by ozonolysis/reduction. Now we report a new application of 3,
† Electronic supplementary information (ESI) available: Experimental
details. See http://www.rsc.org/suppdata/ob/b5/b502952j/
T h i s j o u r n a l i s
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 7 2 9 – 1 7 3 7
1 7 2 9
©

View Article Online
in which the double bond is not oxidatively cleaved, but, on the
contrary, used as a functional group for RCM, to give both 2,3-
dihydropyrans and 1,2,3,6-tetrahydropyridines.27,28 The second
double bond needed for RCM is introduced on one of the two
hydroxymethyl side arms by nucleophilic substitution.28
As a first goal we studied the independent transformation
of both enantiomers of 3 into 6 (Scheme 2). The introduction
of an allyl group onto an oxygen requires the formation of an
alkoxide.29 These conditions are, however, not compatible with
the acetyl group. Therefore, the better-suited THP group was
employed. High-yielding protection–deprotection furnished the
two enantiomers of 4, which were uneventfully converted into
the allyl ethers 5. RCM on 5, performed in the presence of 5%
Grubbs’ first-generation catalyst, required careful monitoring of
the reaction conditions, as shown in Table 1, since the formation
of 7 (as an 8 : 2 E : Z mixture), arising from an intermolecular
process, is in some cases significant. The amount of 7 cannot be
suppressed by high dilution conditions (entries 2 and 3), carried
out by the slow addition of the catalyst through a syringe pump.
While RCM usually involves two terminal double bonds, in this
case a terminal alkene and a substituted alkene are implicated.
The higher steric requirements of the intramolecular reaction,
makes the intermolecular reaction between two terminal olefins
a competitive process even under high dilution. However, by
simply changing the solvent from benzene to CH2Cl2, this side
reaction was nearly completely suppressed.
Scheme 3 Reagents and conditions: (a) MsCl, Et3N, CH2Cl2, −30 ◦C;
(b) allylamine, 80 ◦C; (c) KOH, MeOH, 0 ◦C; (d) DHP, p-TSA, CH2Cl2,
0
◦C; (e) TIPS-Cl, imidazole, DMF, rt; (f) Boc2O, 1,2-dichloroethane,
reflux.
to convert this compound directly into the O-acetyl derivative
of 21 failed: during the Staudinger reduction of the azide
an intramolecular acyl transfer took place to give acetamide
17. Most likely, this transfer occurs during the hydrolysis of
the intermediate iminophosphorane, as previously experienced
by us with similar compounds.30 However, after hydrolyzing
the acetate by means of Pseudomonas cepacia lipase,31 the
conversion of the azide 18 into 21 occurred cleanly, using
a modified literature procedure32 that involved trapping of
the intermediate amino alcohol with Boc-ON.33 From this
intermediate, three differently protected dienes (15, 24 and 25)
have been prepared by protection followed by allylation. The
allylation of the carbamate was quite troublesome, with the
yields strictly depending upon the nature of the O-protecting
group. Compound 15 was obtained in an unexpectedly low
yield,34 while the moderate yield for 24 is in part due to
the competitive intramolecular silyl migration from oxygen to
nitrogen with concomitant O-allylation, following a behaviour
previously observed on other THYM* derivatives. This compet-
itive process was indeed completely suppressed when the bulkier
TIPS protecting group was employed.
Compound 15 was also prepared by introducing the THP
protecting group prior to azide reduction (Scheme 4). In
conclusion, 15 has been obtained following three different
routes (taking into account also the one depicted in Scheme 3),
equivalent in terms of steps. Among them, the one described in
Scheme 3 turned out to be the best, because of the low yield for
the allylation of carbamate 20.
With an analysis of the influence of both the O- and N-
protecting groups on the RCM reaction in mind, we also
synthesized an analogue of 25, compound 31 (Scheme 5). This
was obtained from 18 by the same sequence, but with somewhat
lower overall yield.
Scheme 2 Reagents and conditions: (a) (i) DHP, p-TSA, CH2Cl2, 0 ◦C;
(ii) KOH, MeOH, 0 ◦C; (b) allyl bromide, NaH, DMF, rt.
The analogous N-heterocycles were synthesized following
two strategies, differing in the order of introduction of the
allyl moiety and the carbamate protecting group. Initially the
more expeditious route depicted in Scheme 3 was explored.
Nucleophilic displacement on mesylates 8, 11 and 12, by means
of allylamine both as the solvent and reagent, was successful
◦
only if performed at 80 C in a sealed tube, giving the desired
secondary amine in good yield only when THP was used as
the O-protecting group. However, in view of further synthetic
elaborations, we wanted to develop an efficient protocol able
to produce a small library of differently O-protected starting
materials for RCM. For this reason we turned our attention to
the second strategy, shown in Scheme 4.
The four O-protected carbamates 15, 24, 25 and 31 were
then submitted to RCM in the presence of first-generation
Grubbs’ catalyst, under the conditions found optimal for 5. Boc-
protected olefins reacted smoothly to give the corresponding
tetrahydropyridines in excellent yield, thus suggesting in this
case a negligible influence of the branched unsaturated moiety
The nucleophilic displacement with sodium azide on mesylate
8 gave the desired azide 16 in excellent yield. The attempts
Table 1 Ring-closing metathesis on 5, using 5% Grubbs’ first-generation catalyst
Entry
Solvent
Concentration of 5/M
Temperature/◦C
Yield of 6 (%)
Yield of 7 (%)
1
2
3
4
5
Benzene
Benzene
Benzene
1,2-Dichloroethane
CH2Cl2
0.020
0.034a
0.034a
0.019
0.020
50
44
77
57
74
80
6.5
10
5.3
2.8
trace
40 → 50
60
50
reflux
a Catalyst added through a syringe pump over a period of 6–7 h.
1 7 3 0
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 7 2 9 – 1 7 3 7

View Article Online
Scheme 4 Reagents and conditions: (a) (i) MsCl, Et3N, CH2Cl2, −30 ◦C; (ii) NaN3, DMF, 50 ◦C; (b) PPh3, THF–H2O, 55 ◦C; (c) PCL, THF–H2O
1 : 3, pH 7, rt; (d) DHP, p-TSA, CH2Cl2, 0 ◦C; (e) (i) PPh3, THF–H2O, rt; (ii) Boc-ON, Et3N, rt; (f) allyl bromide, NaH, DMF, rt; (g) R1 R2SiCl,
2
imidazole, rt [R1 = Me, R2 = tBu (22), R1 = R2 = iPr (23)]; (h) Grubbs’ catalyst, 0.028 M (26), 0.014 M (27) or 0.036 M (28) in CH2Cl2, reflux.
during the cyclization process. In contrast to the O-containing
dienes, in no cases could we detect acyclic derivatives analogous
to 7, and this fact allowed us to work, for preparative purposes,
using more concentrated solutions. Finally, we were surprised
by the different behaviour of the Cbz-protected diene, which
was transformed into 32 in only moderate yield (Scheme 5),
thus indicating an influence of the N-protecting group on the
outcome of the reaction. An important feature of this reaction
is the procedure necessary for elimination of the Ru derivatives
responsible for the deeply coloured crude mixture. Several
reported methods have been tested to remove these coloured
impurities [including treatment with Pb(OAc)4,35 Me2SO, and
Ph3PO36]; the most efficient additive in our case was shown to
be triphenylphosphine oxide.
as starting materials for more funtionalized heterocycles, due to
the presence of the double bond and a pre-existing stereocentre.
As a first attempt at functionalization, we studied the
epoxidation of both N- and O-heterocycles, choosing 28 as
the model substrate (Scheme 6). The reaction, performed under
usual peracid conditions, was rather slow, requiring at least rt
in order to start, with several additions of oxidant necessary
in order to obtain a complete conversion of the alkene into
the epoxide. However, we found that the best, though modest,
results in terms of yield, reproducibility and d.r. could be
achieved when operating in refluxing CH2Cl2 (entry 1, Table 2),
while higher boiling solvents such as 1,2-dichloroethane were
Scheme 5 Reagents and conditions: (a) (i) PPh3, THF–H2O, rt;
(ii) BnOCOCl, pH 10, rt; (b) TIPS-Cl, imidazole, DMF, rt; (c) allyl
bromide, NaH, DMF, rt; (d) Grubbs’ catalyst, 0.010 M, CH2Cl2, reflux.
Although several unsaturated six-membered oxygen and
nitrogen heterocycles have been built up by a RCM reaction,
our strategy is, to the best of our knowledge, the only one using
a chemoenzymatic strategy, involving an asymmetrization reac-
tion for the preparation of the acyclic precursor.37 This allows
also the synthesis of both enantiomers by two possible pathways:
a) starting from R- or S-3, readily accessible as described above,
or b) exploiting the “enantiodivergency” of either R- or S-
3, a property arising from the presence of two differentiated
hydroxymethyl groups thatare, however, syntheticallyequivalent
and can be manipulated independently to establish at will the
absolute configuration of the stereogenic centre.
Scheme 6 Reagents and conditions: (a) m-CPBA, CH2Cl2, reflux, (b)
(i) CF3CO2H, CH2Cl2, 0 ◦C; (ii) Troc-Cl, H2O, pH 10, rt; (c) n-Bu4NF,
THF, rt; (d) MeOH, p-TSA, rt.
Both synthons 6 and 28, obtained in an overall yield of 66%
(4 steps) and 64% (8 steps) from 3, respectively, can be envisaged
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 7 2 9 – 1 7 3 7
1 7 3 1

View Article Online
Table 2 Epoxidation of N- and O-containing cyclic alkenes
Yield
Solvent Temperature/ C Time/h Products (% anti + syn) D.r.
◦
Entry Alkene Reagent
1
2
3
4
5
6
7
8
9
28
28
28
41
42
43
43
6
m-CPBA
CH2Cl2 reflux
5
33,37
33,37
33,37
34,38
35,39
36,40
36,40
44,46
45,47
40
trace
18
66
—
—
72
41
77
58 : 42a
—
NaClO, Jacobsen catalyst [(R,R)- or (S,S)-]
CH2Cl2 rt
24
22
5.5
24
24
5.5
6
◦
m-CPBA, NMO, Jacobsen catalyst [(R,R)- or (S,S)-] CH2Cl2 −78 C → rt
—
m-CPBA
CH2Cl2 reflux
CH2Cl2 rt
CH2Cl2 rt
CH2Cl2 reflux
CH2Cl2 reflux
CH2Cl2 reflux
53 : 47b
—
t-BuOOH/VO(acac)2
t-BuOOH/VO(acac)2
m-CPBA
m-CPBA
m-CPBA
—
75 : 25c
50 : 50a
72 : 28c
48
5
a By 13C NMR. b By 1H NMR. c By GC-MS.
less satisfactory. The low stereoselectivity confirms previous
results on similar compounds reported by Bols,38 thus proving a
negligible influence of the bulky O-protecting group on directing
the attack of m-CPBA.
In order to improve this reaction we turned our attention to
the employment of both enantiomers of a Mn(III)-based Jacob-
sen catalyst, using hypochlorite as a stoichiometric oxidant,39
hoping to observe a different stereoselectivity depending upon
the absolute configuration of the catalyst. In this case the
reaction was also very slow, whatever the amount of oxidant
employed, and only traces of product were obtained. When a
more efficient oxidant system (m-CPBA/N-methylmorpholine-
N-oxide) was used,40 an improved but always unsatisfactory
yield was obtained.
The results for the epoxidation of 28, in terms of yield,
were in contrast with literature data, which could be attributed
to the (possibly) low thermal stability of N-Boc derivatives
in the presence of peracids.41 To support this hypothesis we
converted 28 into 2,2,2-trichloroethylcarbamate 41, a derivative
very similar to the one described by Bols (with a TIPS instead of
a TBDPS group).38 In this case an improved yield was observed
(entry 4), but the results were hardly reproducible, and moreover
the reaction was almost completely non-stereoselective.
Working on the unprotected alcohol 43, we reasoned
that the syn-stereoisomer could be favoured using the t-
BuOOH/VO(acac)2 system, as a consequence of a possible cyclic
transition state involving vanadium, t-BuOOH, the primary
alcoholic function and the olefin. Indeed, this strategy has
been successfully applied to other THYM*-derived alkenes.42
However, in this case this approach failed: either starting
from 42 or 43, a low reactivity was observed; using forcing
reaction conditions, only decomposition of starting material was
obtained.
only conclusion allowed by the collected data is that the
stereoselectivity can not be attributed to steric effects.
As a possible application of our building blocks, we checked
the possibility to transform the nitrogen derivatives into two
diastereomeric iminosugars, namely isofagomine 5344 and its
‘gulo’ analogue 54.45 For this purpose we had to hydrolyze
the oxirane ring and to remove the carbamate. For the first
goal, both acidic and basic conditions were tested. The acidic
hydrolysis was tested on both 33,37, and 36,40, and turned out
to be not very satisfactory in terms of yield. In both cases, the
diastereomeric ratio of the resulting diols was identical to that
of the starting epoxides. Thus we presume that the reaction is
regioselective and stereospecific (anti opening). A reasonable
assumption is that water attacks the less crowded C3 atom.
In this hypothesis the major starting epoxides should be anti.
However, since we were not able to perform the reaction on
the diastereomerically pure oxiranes, we cannot definitely prove
that each isomer affords a single diol. Diols 49 (major) and 50
(minor) were readily separated. On the contrary, separation of
triols 51 (major) and 52 (minor) was not possible.46 The chemical
correlations depicted in Scheme 7 demonstrated that the major
isomers 49 and 51 have the same relative configuration.
Finally, the best results, also in terms of d.r., were obtained
on the unprotected homoallylic alcohol 43 (entry 7).43 In all
cases we were unable to separate the diastereomeric epoxides
by chromatography, and for this reason we could not assign the
relative configuration to the prevailing isomer (which was always
the same), although the anti-epoxide seems to be the most likely
(vide infra).
A behaviour similar to the pair 28,43 was observed when 6 and
48 were submitted to epoxidation by means of m-CPBA. While
the THP-protected compound 6 was converted into 44,46 in
moderate yield and without stereoselection (entry 8), the free
alcohol 48 gave 45,47 in quite good yield and with a 72 : 28
diastereomeric ratio (entry 9); however, we were again unable to
separate the diastereoisomers of the O-derivatives.
Scheme 7 Reagents and conditions: (a) 2.3% aq. HClO4, reflux; (b) 3%
HClO4 in Me2CO, rt; (c) 1% aq. KOH, reflux; (d) n-Bu4NF, THF, rt; (e)
TIPS-OTf, 2,6-lutidine, CH2Cl2, 0 ◦C; (f) AcOEt–HCl (3 M), 2 : 1, rt.
As a matter of fact, both types of heterocycles showed a
low reactivity when treated with peracids, which is in contrast
with the usual behaviour of olefins, and this reactivity seems
not to depend upon the nature of the heteroatom. A more
important influence on the yield and the d.r. can be attributed
to the presence or absence of an O-protecting group. The
Thanks to the better overall yield (from 28) and the higher
diastereomeric ratio, acidic opening of epoxyalcohols 36,40 is to
be preferred for the synthesis of isofagomine 53.
Regarding the mixture of triols 51,52 deriving from hydrolysis
of 36,40, the Boc group was removed under different conditions,
affording the final compounds 53 and 54, again with the
1 7 3 2
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 7 2 9 – 1 7 3 7

View Article Online
same diastereomeric ratio of the starting expoxides. At this
step the diastereomers were separated by chromatography
and independently fully characterized and compared with the
literature data, which allowed the unambigous assignment of
relative and absolute configuration. In particular, the optical
rotatory power of 53 is consistent with literature data,38,47 and
this fact demonstrated that the whole sequence proceeded from
THYM* without racemization.
Other methods for epoxide opening under mild conditions,
highly employed in the field of sugars (such as water in the
presence of sodium acetate48 or benzoate49), were shown to be
ineffective, while prolonged reaction time induced only extensive
decomposition of starting material. On the contrary, when
the the mixture of 36,40 was treated in aqueous alkali, a
reproducible one-pot transformation into 53 and 54 in good
yields was realized. However, in this case an unexpected reversal
of diastereoselectivity was observed, with compound 54, having
the same relative stereochemistry of 50 or 52, prevailing.50 These
experimental data demonstrated that under basic conditions the
oxirane opening is most likely to be a non-regioselective process.
This base-catalysed process is therefore to be preferred for the
synthesis of the ‘gulo’ isomer.
always re-extracted thrice with the appropriate organic solvent.
Organic extracts were washed with brine, dried over Na2SO4
and filtered, before evaporation of the solvent under reduced
pressure. All reactions employing dry solvents were carried
out under a nitrogen atmosphere, while RCM reactions were
performed under ultra-pure argon. Lipase from Pseudomonas
cepacia was a kind gift from Amano P, while PPL was purchased
from Sigma and supported on celite following our procedure.
Note: in this section, only selected experimental data are
reported. An exhaustive report is available as electronic sup-
plementary information.†
General procedure for RCM under optimized conditions
Argon was bubbled into a solution of the desired diene [5
(0.020 M), 15 (0.028 M), 24 (0.014 M), 25 (0.036 M), 31 (0.010
M)] in dry CH2Cl2 for 15 min. Then Grubbs’ first-generation
catalyst (5% mmol with respect to substrate) was added and the
reaction was refluxed until complete (1–2.5 h). After cooling to
rt, triphenylphosphine oxide (50 molar equiv. with respect to
catalyst) was added and the mixture was stirred overnight. After
solvent removal under reduced pressure, the crude was directly
chromatographed with the appropriate eluent.
(3S)- and (3R)-3-[(Tetrahydro-2H-pyran-2-yloxy)methyl]-3,6-
dihydro-2H-pyran 6. Prepared starting from both enantiomers
of 5. Chromatography with PE–Et2O 8 : 2 → 7 : 3 gave 6
[(S)- from (S)-5 and (R)- from (R)-5] as a yellow oil in 80%
yield. Rf 0.41 (PE–Et2O 7 : 3, B). Anal. found C, 66.45; H,
9.20. C11H18O3 requires C, 66.64; H, 9.15. [a]D (2S-6) = + 79.9
(CHCl3, c 1.22); [a]D (2R-6) = −78.8 (CHCl3, c 1.30). IR: mmax
2946, 2865, 1351, 1118, 1073, 1022. GC-MS: Rt 5.10; m/z 169
(M+ − 29, 0.084), 86 (5.6), 85 (100), 69 (5.8), 67 (20), 57 (8.9),
Conclusions
In conclusion, we have demonstrated the possibility to transform
THYM*, a building block easy to obtain on a multigram scale,
into a series of heterocycles, differing by the nature of the
heteroatom and by the ring size, which can be decided when
preparing the acyclic precursor. After RCM, the endocyclic
double bond can be exploited for different functionalizations,
including epoxidations followed by oxirane opening by means
of different nucleophiles, dihydroxylations followed by transfor-
mation into cyclic sulfates, and so on. By this strategy, a large
variety of iminosugars and artificial sugars should be accessible.
Moreover, one or both oxygen arms of THYM* can be
sequentially elaborated through diastereoselective procedures to
give a series of dienes suited for RCM, allowing a new entry to
carbasugars. Our results in this field will be reported in due
course.
1
55 (8.2), 43 (12), 41 (20). H NMR: 1.43–1.88 [6H, m, 3 CH2
of THP]; 2.47 [1H, centre of m, CHCH2OTHP]; 3.30–3.91 [6H,
m, CH2OTHP, OCHOCH2, CH2OCHCH2OTHP]; 4.10–4.12
=
[2H, m, CHCH2O]; 4.57–4.63 [1H, m, OCHO]; 5.72–5.85 [2H,
m, CH CH]. 13C NMR: 19.34 and 19.64 [CH2CH2(CH2)2O];
=
25.44 [(CH2)2CH2CH2O]; 30.57 [CH2(CH2)3O]; 35.36 and 36.60
[CHCH2OTHP]; 62.02 and 62.44, 65.59 and 65.57, 66.26 and
66.61, 68.04 and 68.32 [4C, CH2O]; 98.47 and 99.40 [OCHO];
=
125.32 and 125.41, 127.86 [2C, C C].
Experimental
Compound 7 from intermolecular metathesis. Obtained, as
an 8 : 2 E : Z mixture, during RCM of diene 5. The yield
depended upon the reaction conditions (trace–10%) Rf 0.36
(PE–Et2O 7 : 3, B). IR: mmax 2927, 2854, 1455, 1363, 1190, 1119,
NMR spectra were taken in CDCl3 at 200 MHz (1H) and 50 MHz
(13C) (unless otherwise stated), using TMS as internal standard.
Chemical shifts are reported in ppm (d scale), coupling constants
are reported in hertz. Peak assignment in 1H NMR spectra was
also made with the aid of double resonance experiments. Peak
assignment in 13C NMR spectra was made with the aid of DEPT
experiments. GC-MS were carried out on a HP-5971A instru-
ment, using an HP-1 column (12 m long, 0.2 mm wide), electron
impact at 70 eV, and an ionization chamber temperature of about
170 ◦C. Unless otherwise indicated, analyses were performed
with a constant He flow of 0.9 ml min−1, init. temp. 100 ◦C,
init. time 2 min, rate 20 ◦C min−1, final temp. 260 ◦C, final time
4 min, inj. temp. 250 ◦C, det. temp. 280 ◦C. Rt values are in min.
IR spectra were measured in CHCl3 solution with a Perkin–
Elmer 881 instrument. [a]D values were determined on a Jasco
DIP 181 polarimeter, in CHCl3 (containing 0.75–1% EtOH)
solution. TLC analyses were carried out on silica gel plates,
which were developed by the following detection methods: A)
UV; B) dipping into a solution of (NH4)4MoO4·4H2O (21 g)
and Ce(SO4)2·4H2O (1 g) in H2SO4 (31 ml) and H2O (469 ml)
and warming; C) dipping into a solution of 4% aq. KMnO4
and warming; D) dipping into a solution of p-anisaldehyde
(5.5 ml) in H2SO4 (7.5 ml), AcOH (2.2 ml) and EtOH (500 ml)
and warming. Rf were measured after an elution of 7–9 cm.
Chromatographies were carried out on 220–400 mesh silica gel
using “flash” methodology. Petroleum ether (40–60 ◦C) is ab-
breviated as PE. In extractive work-up, aqueous solutions were
1
1019, 974. GC-MS: compound unsuitable for this analysis. H
NMR: 0.97 [12H, d, (CH3)2CH, J 6.6]; 1.28–1.87 [12H, m, 3
CH2 of THP]; 2.58 [2H, octet, (CH3)2CH, J 6.7]; 2.55 [2H,
centre of m, CHCH2OTHP]; 3.35–3.91 [8H, m, CH2OTHP
=
=
and CH2OCH2CH ]; 3.97 [4H, d, OCH2CH CHCH2O (E),
J 3.8]; 4.04 [4H, d, OCH2CH CHCH2O (Z), J 4.4]; 4.59
=
[2H, broad t, OCHO, J 3.0]; 5.32 and 5.52 [4H, ddd
and dd, CH CH–iPr, J 2.6, 7.8, 15.8 and 6.2, 15.6]; 5.68
=
=
[2H, broad t, OCH2CH CHCH2O (Z), J 3.9]; 5.78 [2H,
broad s, OCH2CH CHCH2O (E)]. 13C NMR: 19.34 [2C,
=
CH2CH2(CH2)2O]; 22.52 and 22.58 [4C, CH(CH3)2]; 25.55
[2C, (CH2)2CH2CH2O]; 30.61 [2C, CH2(CH2)3O]; 31.19 [2C,
CH(CH3)2]; 42.95 [2C, CHCH2OTHP]; 61.92, 68.19 and 68.33
[4C, CH2OTHP and (CH2)3CH2O]; 71.04 and 71.35 [4C,
=
CH2OCH2CH ]; 98.70 [2C, OCHO]; 125.46 and 139.74 [4C,
CH CH–iPr]; 129.29 [2C, OCH2CH CHCH2O].
=
=
(R)-3-[(Tetrahydro-2H-pyran-2-yloxy)methyl]-3,6-dihydro-2H-
pyridine-1-carboxylic acid tert-butyl ester 26. Prepared
starting from 15. Chromatography with PE–Et2O 7 : 3 gave 26
as a yellow oil in 90% yield. Rf 0.40 (PE–Et2O 7 : 3, B). Anal.
found C, 64.80; H, 9.25; N, 4.60. C16H27NO4 requires C, 64.62;
H, 9.15; N, 4.71. [a]D = −55.0 (CHCl3, c 0.84). IR: mmax 2943,
1675, 1365, 1192, 1119. GC-MS: Rt 7.92; m/z 242 (M+ − 55,
0.079), 157 (16), 156 (19), 141 (6.4), 140 (18), 139 (19), 138 (7.5),
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 7 2 9 – 1 7 3 7
1 7 3 3

View Article Online
=
127 (18), 112 (28), 97 (8.2), 96 (16), 95 (11), 94 (19), 86 (5.7),
85 (96), 82 (9.0), 80 (9.5), 68 (15), 67 (30), 57 (100), 56 (7.2), 55
5.67–5.84 [2H, m, CH CH]; 7.31–7.38 [5H, m, aromatics]. 13C
NMR (DMSO-d6; temp. = 100 ◦C): 11.00 [3C, Si(CH(CH3)2)3];
17.10 [6C, Si(CH(CH3)2)3]; 37.56 [CHCH2O]; 42.10 and 42.78
1
(9.5), 43 (16), 42 (5.3), 41 (47), 39 (9.4). H NMR (DMSO-d6;
temp. = 100 ◦C): 0.83–1.90 [6H, m, 3 CH2 of THP]; 1.43 [9H, s,
OC(CH3)3]; 2.45 [1H, centre of m, CHCH2OTHP]; 2.97–3.80
[6H, m, CH2OTHP, OCHOCH2, CH2NCHCH2O]; 3.82–3.83
[2C, CH2NCH2CH CH]; 64.00 and 65.60 [2C, CH2OSi and
CH2OPh]; 124.76 and 125.62 [2C, C C]; 126.69, 126.98 and
127.60 [5C, CH of Ph]; 136.46 [ipso-C of Ph]; 154.28 [CO].
=
=
=
[2H, m, CHCH2N]; 4.57–4.60 [1H, m, OCHO]; 5.69–5.82
[2H, m, CH CH]. 13C NMR (DMSO-d6; temp. = 100 ◦C):
=
General procedure for the epoxidation of RCM-derived products
18.37 and 18.48 [CH2CH2(CH2)2O]; 24.43 [(CH2)2CH2CH2O];
27.47 [3C, OC(CH3)3]; 29.66 [CH2(CH2)3O]; 34.95 and 35.28
[CHCH2OTHP]; 41.89 and 42.09, 42.59 [2C, CH2NCH2]; 60.55
and 60.88, 67.29 and 67.64 [2C, CH2O]; 78.04 [OC(CH3)3];
97.37 and 98.14 [OCHO]; 125.05 and 125.14, 125.65 [2C, C C];
153.66 [CO].
A solution of the alkene (1.00 mmol) in dry CH2Cl2 (8 ml)
was cooled to 0 ◦C and treated with m-CPBA (1.50 mmol).
After 5 min the solution was refluxed for the required time (5–
6 h). Usually, after about 2 h, a further addition of peracid was
required in order for the reaction to go to completion. After
cooling the reaction mixture to 0 ◦C, Me2S (1 molar equiv. with
respect to the acid employed) was added. This was followed by
the addition of sat. aq. NaHCO3 solution (15 ml). The biphasic
system was vigorously stirred at rt for 15 min and then extracted
with ether.
=
(R)-3-[(tert-Butyldimethylsilyloxy)methyl]-3,6-dihydro-2H -
pyridine-1-carboxylic acid tert-butyl ester 27. Prepared
starting from 24. Chromatography with PE–CH2Cl2 1 :
1 → CH2Cl2 and then Et2O gave 27 as a yellow oil in 94%
yield. Rf 0.18 (PE–Et2O 97 : 3, B). Anal. found C, 62.60;
H, 10.05; N, 4.35. C17H33NO3Si requires C, 62.34; H, 10.16;
N, 4.28. [a]D = −34.5 (CHCl3, c 1.42). IR: mmax 2928, 2427,
(1S,5R,6R)- and (1R,5R,6S)-5-[(Triisopropylsilyloxy)methyl]-
7-oxa-3-azabicyclo[4.1.0]heptane-3-carboxylic acid tert-butyl es-
ter 33 and 37. This diastereomeric mixture was prepared
starting from 28. Chromatography with PE–Et2O 9 : 1 → 7 :
3 gave 33,37 as an inseparable 58 : 42 (13C NMR) diastereomeric
mixture (as a pale yellow oil) in 40% yield. Rf 0.55 (PE–Et2O 7 :
3, B, C). IR: mmax 2939, 2864, 1673, 1367, 1252, 1115. GC-MS: Rt
9.50; m/z 328 (M+ − 57, 0.25), 312 (6.0), 288 (5.8), 287 (20), 286
(100), 256 (9.2), 242 (26), 131 (6.3), 119 (6.8), 103 (6.6), 94 (5.6),
75 (9.2), 61 (6.9), 59 (6.8), 57 (43), 56 (4◦2), 44 (9.4), 42 (7.4), 41
(11). 1H NMR (DMSO-d6; temp. = 100 C): 0.95–1.25 [21H, m,
TIPS]; 1.40 [9H, s, OC(CH3)3]; 2.10–2.23 [1H, m, CHCH2O];
2.83–2.95 and 3.12–3.30 [2H, 2 m, 2 CH–O]; 3.40–3.90 [6H,
m, CH2OSi and CH2NCH2]. 13C NMR (DMSO-d6; temp. =
100 ◦C): [N.B.: where possible the signals in the 13C NMR
spectrum have been attributed to the major diastereoisomer (M)
or to the minor diastereoisomer (m)] 11.00 [3C, Si(CH(CH3)2)3];
17.15 [6C, Si(CH(CH3)2)3]; 27.47 [3C, OC(CH3)3]; 35.81 (m)
and 36.83 (M) [CHCH2O]; 39.23, 41.36 (M) and 41.70 (m) [2C,
CH2NCH2]; 48.69 (m) and 49.15 (M), 50.44 (M) and 51.15 (m)
[2C, 2 CH–O]; 62.34 (m) and 62.70 (M) [CH2OSi]; 78.22 (m)
and 78.23 (M) [C(CH3)3]; 153.53 (m) and 153.92 (M) [CO].
1682, 1414, 1194, 1104. GC-MS: Rt 7.29; m/z 270 (M+
−
57, 0.24), 215 (15), 214 (100), 170 (16), 105 (7.5), 96 (6.5), 95
(5.2), 94 (6.3), 89 (20), 88 (6.3), 75 (27), 73 (25)◦, 67 (8.0), 57
1
(42), 41 (10). H NMR (DMSO-d6; temp. = 100 C): 0.06 and
0.08 [6H, 2 s, Si(CH3)2tBu]; 0.91 [9H, s, SiMe2C(CH3)3]; 1.43
[9H, s, OC(CH3)3]; 2.32 [1H, centre of m, (CH3)2CH]; 3.28
and 3.55 [2H, AB part of ABX system, CH2O, JAB 13.0, JAX
6.2, JBX 4.3]; 3.38–3.55 [2H, m, CH2NCHCH2O]; 3.81 [2H,
broad s, NCH2CH ]; 5.67–5.80 [2H, m, CH CH]. 13C NMR
(DMSO-d6; temp. = 100 ◦C): −6.12 [2C, Si(CH3)2C(CH3)3];
17.24 [Si(CH3)2C(CH3)3]; 25.31 [3C, Si(CH3)2C(CH3)3]; 27.54
[3C, OC(CH3)3]; 37.50 [CHCH2O]; 41.87 and 42.67 [2C,
=
=
=
CH2NCH2CH CH]; 62.52 [CH2OSi]; 78.06 [C(CH3)3]; 125.18
and 125.2 [2C, C C]; 153.73 [CO].
=
(R)-3-[(Triisopropylsilyloxy)methyl]-3,6-dihydro-2H-pyridine-
1-carboxylic acid tert-butyl ester 28. Prepared starting from
25. Chromatography with PE–Et2O 98 : 2 → 9 : 1 gave 28
as a yellow oil in 95% yield. Rf 0.36 (PE–Et2O 95 : 5, B,
C). Anal. found C, 64.75; H, 10.80; N, 3.65. C20H39NO3Si
requires C, 64.99; H, 10.64; N, 3.79. [a]D = −27.7 (CHCl3,
c 1.15). IR: mmax 2938, 2865, 1679, 1159, 1110. GC-MS: Rt
8.81; m/z 312 (M+ − 57, 0.73), 272 (5.8), 271 (20), 270 (100),
226 (13), 182 (5.4), 145 (5.4), 131 (12), 119 (7.2), 103 (12),
75 (21), 73 (7.0), 67 (9.5), 61 (11), 59 (15), 57 (44), 45 (7.5),
41 (22). 1H NMR (300 MHz, DMSO-d6; temp. = 100 ◦C):
1.04–1.11 [21H, m, TIPS]; 1.42 [9H, s, OC(CH3)3]; 2.37 [1H,
centre of m, CHCH2O]; 3.30 and 3.58 [2H, AB part of ABX
system, CH2O, JAB 13.0, JAX 6.3, JBX 8.2]; 3.51–3.70 [2H, m,
(1S,5R,6R)- and (1R,5R,6S)-5-[(Triisopropylsilyloxy)methyl]-
7-oxa-3-azabicyclo[4.1.0]heptane-3-carboxylic acid 2,2,2-
trichloroethyl ester 34 and 38. This diastereomeric mixture
was prepared starting from 41. Chromatography with PE–Et2O
95 : 5 → 7 : 3 gave 34,38 as an inseparable 53 : 47 (1H NMR)
diastereomeric mixture (as a pale yellow oil) in 66% yield. Rf
0.29 (PE–Et2O 9 : 1, B, C). IR: mmax 2920, 2860, 1710, 1601,
1193, 1123. GC-MS: Rt 11.17; m/z 418 [M+ − 43 (2 36Cl and
1
35Cl, 21)], 416 [M+ − 43 (3 35Cl, 20)], 388 (12), 386 (12), 312
=
CH2NCHCH2O]; 3.73–3.89 [2H, m, NCH2CH ]; 5.70–5.80
(8.6), 272 (9.3), 270 (28), 268 (29), 234 (7.7), 232 (23), 230 (22),
222 (5.2), 220 (9.8), 218 (6.0), 211 (14), 199 (6.5), 157 (16), 145
(8.2), 139 (5.8), 138 (11), 137 (11), 135 (12), 133 (38), 131 (56),
129 (14), 127 (8.4), 123 (5.0), 121 (14), 120 (10), 119 (83), 115
(26), 114 (5.0), 113 (9.4), 111 (8.6), 108 (6.6), 103 (26), 101 (11),
99 (15), 98 (6.3), 97 (20), 96 (8.3), 95 (33), 94 (28), 93 (14), 89
(5.8), 88 (5.0), 87 (18), 85 (6.9), 83 (7.3), 82 (12), 81 (6.0), 80 (16),
79 (5.3), 77 (11), 75 (46), 73 (21), 71 (6.7), 69 (7.7), 67 (13), 63
(5.2), 61 (43), 60 (6.3), 59 (45), 57 (6.8), 56 (59), 55 (16), 54 (6.6),
[2H, m, CH ;CH]. 13C NMR (300 MHz, DMSO-d6; temp. =
=
100 ◦C): 11.02 [3C, Si(CH(CH3)2)3]; 17.14 [6C, Si(CH(CH3)2)3];
27.50 [3C, OC(CH3)3]; 37.73 [CHCH2O]; 41.97 and 42.65 [2C,
CH2NCH2CH CH]; 64.06 [CH2OSi]; 78.03 [C(CH3)3]; 125.13
=
=
and 125.54 [2C, C C]; 153.75 [CO].
(R)-3-[(Triisopropylsilyloxy)methyl]-3,6-dihydro-2H-pyridine-
1-carboxylic acid benzyl ester 32. Prepared starting from 31.
Chromatography with PE–Et2O 9 : 1 → 8 : 2 gave 32 as a yellow
oil in 73% yield. Rf 0.43 (PE–Et2O 8 : 2, A, B). Anal. found
C, 68.60; H, 9.15; N, 3.60. C23H37NO3Si requires C, 68.44; H,
9.24; N, 3.47. [a]D = −47.4 (CHCl3, c 1.94). IR: mmax 2941, 2862,
1685, 1431, 1192, 1107. GC-MS: Rt 11.14; m/z 360 (M+ − 43,
10), 317 (6), 316 (22), 182 (7), 100 (6), 92 (9), 91◦100), 75 (6), 65
1
53 (5.0), 45 (25), 44 (12), 43 (23), 42 (30), 41 (26), 39 (6.9). H
NMR (DMSO-d6; temp. = 100 ◦C): 0.95–1.25 [21H, m, TIPS];
2.20–2.35 [1H, m, CHCH2O]; 3.06–3.39 [2H, m, 2 CH–O];
3.57–4.07 [6H, m, CH2OSi and CH2NCH2]; 4.78 and 4.83,
4.80 and 4.82 [2H, 2 AB system, CO2CH2, J 12.3 and 12.3].
◦
1
(7), 59 (5). H NMR (DMSO-d6; temp. = 100 C): 1.06 [21H,
13C NMR (DMSO-d6; temp. = 100 C): [N.B.: where possible
apparent s, TIPS]; 2.40 [1H, centre of m, CHCH2O]; 3.37 and
3.70 [2H, AB part of ABX system, CH2O, JAB 13.0, JAX 6.6,
the signals in the 13C NMR spectrum have been attributed to
the major diastereoisomer (M) or to the minor diastereoisomer
(m)] 11.00 [3C, Si(CH(CH3)2)3]; 17.17 [6C, Si(CH(CH3)2)3];
35.66 (m) and 36.68 (M) [CHCH2O]; 39.59, 41.75 (m) and
J
BX 4.9]; 3.32–3.75 [2H, m, CH2NCHCH2O]; 3.78–4.01 [2H, m,
=
NCH2CH ]; 5.09 and 5.13 [2H, AB system, CH2OPh, J 12.6];
1 7 3 4
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 7 2 9 – 1 7 3 7

View Article Online
42.07 (M) [2C, CH2NCH2]; 48.47 (m) and 48.93 (M), 50.42
(M) and 51.17 (m) [2C, 2 CH–O]; 62.38 (m) and 62.67 (M)
[CH2OSi]; 73.97 [CO2CH2]; 95.68 [CCl3]; 152.48 [CO].
was suspended in 2.3% aq. HClO4 (3 ml) and refluxed for 1 h. The
solution was neutralized with solid K2CO3 and, after saturation
with solid NaCl, an extraction was performed with AcOEt.
Chromatography with PE–Et2O 25 : 75 gave 49 (8.9 mg, 12%
yield) and 50 (6.5 mg, 9% yield) as colourless oils. Compound
49: Rf 0.36 (PE–Et2O 75 : 25, C, D). [a]D = –9.6 (CHCl3, c 0.73).
IR: mmax 3424, 2934, 2866, 1680, 1420, 1191. GC-MS: Rt 10.15
(1S,5R,6R)- and (1R,5R,6S)-5-Hydroxymethyl-7-oxa-3-aza-
bicyclo[4.1.0]heptane-3-carboxylic acid tert-butyl ester 36 and
40. This diastereomeric mixture was prepared starting from
43. Chromatography with Et2O gave 36,40 as an inseparable
75 : 25 (GC-MS) diastereomeric mixture (as a pale yellow oil)
in 72% yield. Rf 0.46 (Et2O, C, D). IR: mmax 3454, 2920, 1679,
1413, 1366, 1162, 1019. GC-MS (◦usual method, but init. temp.
80 ◦C, init. time 2 min, rate 10 C min−1): Rt 10.63 (minor);
m/z 229 (M+, 1.3), 172 (5.8), 156 (5.2), 142 (5.9), 98 (5.1), 83
(8.6), 82 (6.3), 70 (6.1), 68 (5.6), 58 (5.1), 57 (100), 56 (9.0),
55 (6.6), 43 (26), 42 (7.8), 41 (23), 39 (5.0); Rt 10.82 (major);
m/z 229 (M+, 0.32), 173 (5.0), 172 (5.5), 98 (12.8), 83 (6.3),
82 (5.6), 80 (5.5), 70 (6.0), 68 (5.9), 58 (5.3), 57 (100), 56
◦
(or 18.82 with ◦usual method, but with init. temp. 130 C, init.
time 0, rate 5 C min−1 to establish d.r.); m/z 342 (M+ − 61,
0.093), 304 (14), 260 (17), 242 (13), 112 (10), 103 (6.7), 83 (5.3),
77 (8.8), 75 (15), 73 (5.0), 61 (11), 59 (10), 58 (5.2), 57 (100),
1
56 (6.9), 45 (6.2), 44 (66), 43 (8.2), 42 (18), 41 (18). H NMR
(DMSO-d6; temp. = 100 ◦C); 0.95–1.20 [21H, m, TIPS]; 1.41
[9H, s, OC(CH3)3]; 1.47–1.62 [1H, m, CHCH2O]; 2.45 [1H, dd,
H
6ax, J 12.8, 10.2]; 2.56 [1H, dd, H2ax, J 13.2, 11.4]; 3.55–3.65
[2H, m, H3, H4]; 3.96 [1H, ddd, H6eq, J 12.4, 4.8, 2.2]; 3.97–
4.04 [2H, m, CH2OSi]; 4.12 [1H, ddd, H2eq, J 13.6, 4.4, 2.6];
4.36 [1H, d, OH, J 4.6]; 4.61 [1H, d, OH, J 4.4]. 13C NMR
(DMSO-d6; temp. = 100 ◦C): 11.04 [3C, Si(CH(CH3)2)3]; 17.21
[6C, Si(CH(CH3)2)3]; 27.51 [3C, OC(CH3)3]; 44.25 [CHCH2O];
44.82 and 47.73 [2C, CH2NCH2]; 62.14 [CH2OSi]; 70.66 and
73.31 [2C, 2 CHOH]; 78.08 [C(CH3)3]; 153.41 [CO]. Compound
50: Rf 0.24 (PE–Et2O 75 : 25, C, D). [a]D = −13 (CHCl3, c
0.77). IR: mmax 3416, 2923, 2865, 1680, 1416, 1157, 1088. GC-
MS:◦Rt 10.09 (or 18.49 with usual method, but with init. temp.
1
(8.4), 55 (5.5), 44 (5.0), 43 (21), 42 (10), 41 (24), 39 (5.2). H
NMR (DMSO-d6; temp. = 100 ◦C): 1.41 [9H, s, OC(CH3)3];
2.08 [1H, centre of m, CHCH2O]; 2.77–3.91 [8H, m, 2 CH–
O, CH2OSi and CH2NCH2]. 13C NMR (DMSO-d6; temp. =
100 ◦C): [N.B.: where possible the signals in the 13C NMR
spectrum have been attributed to the major diastereoisomer (M)
or to the minor one (m)] 27.53 [3C, OC(CH3)3]; 35.53 (m) and
36.61 (M) [CHCH2O]; 39.21 (M) and 40.40 (m), 41.39 (M) and
41.76 (m) [2C, CH2NCH2]; 48.73 (m) and 49.11 (M), 50.75 (M)
and 51.51 (m) [2C, 2 CH–O]; 60.23 (m) and 60.54 (M) [CH2OSi];
78.22 (m) and 78.26 (M) [C(CH3)3]; 153.62 [CO].
◦
130 C, init. time 0, rate 5 C min−1 to establish d.r.); m/z 342
(M+ − 73, 1.14), 304 (23), 261 (5.5), 260 (29), 112 (9.3), 103
(7.0), 83 (5.3), 77 (8.8), 75 (16), 73 (5.6), 72 (5.6), 61 (13), 59
(11), 58 (5.4), 57 (100), 56 (10), 45 (6.4), 44 (50), 43 (6.5), 42
(1S,5R,6R)- and (1R,5R,6S)-5-[(tetrahydro-2H-pyran-2-yloxy)-
methyl]-3,7-dioxabicyclo[4.1.0]heptane 44 and 46. This di-
astereomeric mixture was prepared starting from 6. Chromatog-
raphy with PE–Et2O 6 : 4 gave 44,46 as an inseparable ≈1 : 1
(13C NMR) diastereomeric mixture (as a pale yellow oil) in 41%
yield. Rf 0.57 (PE–Et2O 4 : 6, B). GC-MS: Rt 6.06; m/z 129
(M+ − 85, 0.22), 101 (36), 86 (6.1), 85 (100), 84 (16), 83 (18),
69 (10), 67 (16), 57 (20), 56 (9.8), 55 (31), 43 (19), 41 (29),
(19), 41 (17). H NMR (DMSO-d6; temp. = 100 ◦C); 0.94–
1
1.20 [21H, m, TIPS]; 1.40 [9H, s, OC(CH3)3]; 2.00–2.11 [1H,
m, CHCH2O]; 2.80–3.81 [8H, m, CH2NCH2CHCH2OSi, CH–
O]; 4.41 [2H, broad s, OH]. 13C NMR (DMSO-d6; temp. =
100 ◦C): 11.04 [3C, Si(CH(CH3)2)3]; 17.21 [6C, Si(CH(CH3)2)3];
27.58 [3C, OC(CH3)3]; 40.79 [CHCH2O]; 44.96 and 51.52 [2C,
CH2NCH2]; 62.05 [CH2OSi]; 67.22 and 68.56 [2C, 2 CHOH];
77.55 [C(CH3)3]; 154.48 [CO].
1
39 (9.8). H NMR: 1.43–1.90 [6H, m, 3 CH2 of THP]; 2.28–
2.49 [1H, centre of m, CHCH2OTHP]; 3.15–4.08 [10H, m, 4
CH2O, 2 CH–O]; 4.57–4.67 [1H, m, OCHO]. 13C NMR: 19.40
and 19.45 [CH2CH2(CH2)2O]; 25.38 [(CH2)2CH2CH2O]; 30.51
[CH2(CH2)3O]; 34.71, 34.78, 35.04 and 35.11 [CHCH2OTHP];
50.28, 50.69, 50.98 and 51.80 [2C, 2 CH–O]; 62.18, 62.35 and
62.39, 63.49, 63.53 and 63.69, 64.90 and 65.11, 65.87, 65.97,
66.04 and 66.19 [4C, 4 CH2O]; 99.04 [OCHO].
(3R,4R,5R)- and (3S,4S,5R)-3,4-Dihydroxy-5-(hydroxymethyl)-
piperidine-1-carboxylic acid tert-butyl ester 51 and 52 from
36,40. Diastereomeric mixture 36,40 (94 mg, 410 lmol) was
◦
dissolved in acetone (8 ml) and cooled to 0 C. 3% aq. HClO4
(90 ll) was added and the rection was stirred at rt for 31 h.
The solution was neutralized with solid NaHCO3 and, after
saturation with solid NaCl, an extraction was performed with
AcOEt. Chromatography with PE–AcOEt 2 : 8 → AcOEt–
MeOH 9 : 1 gave inseparable mixture 51,52 (45 mg, 44% yield)
as a white solid with a 75 : 25 d.r. (by 13C NMR). Rf 0.43
(AcOEt–MeOH 9 : 1, D). GC-MS (for 51,52): Rt 7.57; m/z 247
(M+, 0.34), 243 (6.2), 190 (6.7), 154 (7.0), 112 (5.4), 98 (13), 72
(5.3), 70 (11), 60 (6.7), 59 (7.6), 58 (6.7), 57 (100), 56 (10), 55
(6.3), 45 (7.6), 44 (16), 43 (18), 42 (14), 41 (33), 39 (8.5). The
following spectroscopic data have been collected on the separate
diastereoisomers obtained after removal of TIPS from samples
of 49 and 50 (see supplementary material†). Compound 51: IR:
(1S,5S,6R)- and (1R,5S,6S)-3,7-dioxabicyclo[4.1.0]heptan-5-
ylmethanol 45 and 47. This diastereomeric mixture was pre-
pared starting from 48. Chromatography with Et2O gave 45,47
as a 72 : 28 (GC-MS) inseparable diastereomeric mixture (as a
pale yellow oil) in 77% yield. Rf 0.25 (PE–AcOEt 15 : 85, C). IR:
mmax 3450, 2954, 2889, 1243, 1135, 1109, 1022. GC-MS (usual
method, but init. temp. 60 ◦C, init. time 2 min, rate 2 ◦C min−1
until 120 ◦C, then 20 ◦C min−1 until 260 ◦C): Rt 10.08 (minor);
m/z 99 (M+ −31, 60), 87 (7.0), 83 (12), 82 (16), 81 (14), 74 (21), 73
(20), 71 (30), 70 (15), 69 (56), 68 (5.5), 58 (29), 57 (100), 56 (25),
55 (33), 54 (9.5), 53 (17), 45 (16), 44 (29), 43 (35), 42 (11), 41 (56),
40 (6.6), 39 (28); Rt 10.38 (major); m/z 129 (M+ − 18, 2.3), 100
(5.3), 99 (84), 87 (13), 83 (22), 82 (60), 81 (31), 74 (9.0), 73 (12),
71 (29), 70 (81), 69 (100), 68 (6.1), 66 (8.6), 58 (8.2), 57 (99), 56
(23), 55 (51), 54 (16), 53 (20), 45 (24), 44 (38), 43 (63), 42 (25), 41
(80), 40 (15), 39 (43). 1H NMR: 2.05–2.34 [2H, m, CHCH2OH];
3.24–4.00 [8H, m, 3 CH2O, 2 CH–O]. [N.B.: where possible the
signals in the 13C NMR spectrum have been attributed to the
major diastereoisomer (M) or to the minor diastereoisomer (m)]
13C NMR: 36.31 (M) and 36.46 (m) [CHCH2OH]; 50.44, 51.09
and 51.62 [2C, 2 CH–O]; 61.65 (M) and 62.04 (m), 63.36 (M)
and 63.92 (m), 64.86 (M) and 64.91 (m) [3C, 3 CH2O].
1
mmax 3404, 2958, 2851, 1673, 1235, 1091. H NMR (DMSO-d6;
temp. = 100 ◦C); 1.43 [9H, s, OC(CH3)3]; 1.80–2.15 [1H, m,
CHCH2O]; 2.42–2.58 [2H, m, H6ax, H2ax]; 2.76–4.13 [6H, m,
CH2OH, H2eq, H3, H4, H6eq]; 4.33 [1H, d, OH, J 3.3]; 4.38 [1H,
broad s, OH]; 4◦.57 [1H, d, OH, J 3.4]. 13C NMR (DMSO-
d6; temp. = 100 C): 27.62 [3C, OC(CH3)3]; 43.85 [CHCH2O];
44.69 and 47.82 [2C, CH2NCH2]; 59.94 [CH2OH]; 70.69 and
73.72 [2C, 2 CHOH]; 78.06 [C(CH3)3]; 153.48 [CO]. Compound
52: IR: mmax 3367, 2962, 2845, 1674, 1249, 1146, 1097. 1H NMR
(DMSO-d6; temp. = 100 ◦C); 1.41 [9H, s, OC(CH3)3]; 1.92–2.06
[1H, m, CHCH2O]; 2.89–3.05 [2H, m, H6ax, H2ax]; 3.20–3.62
[9H, m, OH, CH2OH, H2eq, H3, H4, H6eq]. 13C NMR (DMSO-
◦
(3R,4R,5R)- and (3S,4S,5R)-3,4-Dihydroxy-5-[(triisopropyl-
silyloxy)methyl]piperidine-1-carboxylic acid tert-butyl ester 49
and 50. The diastereomeric mixture 33,37 (70 mg, 181 lmol)
d6; temp. = 100 C): 27.54 [3C, OC(CH3)3]; 43.88 [CHCH2O];
45.09 and 47.81 [2C, CH2NCH2]; 59.95 [CH2OH]; 67.25 and
68.87 [2C, 2 CHOH]; 77.56 [C(CH3)3]; 153.49 [CO].
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 7 2 9 – 1 7 3 7
1 7 3 5

View Article Online
(3R,4R,5R)- and (3S,4S,5R)-5-(Hydroxymethyl)piperidine-
3,4-diol 53 and 54. Method 1: By removal of Boc from 51,53: a)
Under acidic conditions: a solution of 51,52 (33 mg, 132 lmol)
in AcOEt (1 ml) was treated with 3 M aq. HCl (1.5 ml) and
the biphasic system was stirred at rt for 1 h. The solvent was
removed by distillation under vacuum. The residue was taken
up with 100 ll of Et3N and purified by preparative thin layer
chromatography (thickness of silica on the plate: 0.5 mm) using
MeOH–NH3 (1%), 7 : 3, to give 53 (6.5 mg, 33% yield) and 54
(2.4 mg, 12% yield), confirming by weight the d.r. of the starting
mixture of triols. b) Under basic conditions: the diastereomeric
mixture 51,52 (18 mg, 73 lmol) was suspended in 1% aq.
KOH (1 ml) and refluxed for 12 h. Water was removed under
reduced pressure and the crude was directly chromatographed
with MeOH–NH3 (1%), 7 : 3, to give 53,54 (8.7 mg, 81% yield)
as a 75 : 25 diastereomeric mixture (by 13C NMR). Method 2:
By oxirane opening under basic conditions: the diastereomeric
mixture 36,40 (69 mg, 300 lmol) was suspended in 1% aq.
KOH (2 ml) and refluxed for 3 h. Water was removed under
reduced pressure and the crude was directly chromatographed
with MeOH–NH3 (1%), 7 : 3, to give 53 (13 mg, 30% yield) and
54 (18 mg, 41% yield). Compound 53: Rf 0.41 (MeOH–NH3
(1%), 7 : 3, B). [a]D = +18 (CHCl3, c 0.16).51 Compound 54: Rf
0.22 (MeOH–NH3 (1%), 7 : 3, B). [a]D = +22 (CHCl3, c 0.11).52
5582; S. Schleich and G. Helmchen, Eur. J. Org. Chem., 1999, 2515–
2521; A. J. Souers and J. A. Ellman, J. Org. Chem., 2000, 65, 1222–
1224; T. Taniguchi and K. Ogasawara, Org. Lett., 2000, 2, 3193–3195;
C. Agami, F. Couty and G. Evano, Tetrahedron: Asymmetry, 2000,
11, 4639–4643; H. Mues and U. Kazmaier, Synthesis, 2001, 487–498;
Y. Banba, C. Abe, H. Nemoto, A. Kato, I. Adachi and H. Takahata,
Tetrahedron: Asymmetry, 2001, 12, 817–819; S. Varray, R. Lazaro, J.
Martinez and F. Lamaty, Eur. J. Org. Chem., 2002, 2308–2316; H.
Han, Tetrahedron Lett., 2003, 44, 1567–1569.
16 S. J. Miller, H. E. Blackwell and R. H. Grubbs, J. Am. Chem. Soc.,
1996, 118, 9606–9614.
17 H. Li, Y. Ble´riot, C. Chantereau, J.-M. Mallet, M. Sollogoub, Y.
Zhang, E. Rodr´ıguez-Garc´ıa, P. Vogel, J. Jime´nez-Barbero and P.
Sinay¨, Org. Biomol. Chem., 2004, 2, 1492–1499.
18 G. Godin, E. Garnier, P. Compain, O. R. Martin, K. Ikeda and N.
Asano, Tetrahedron Lett., 2004, 45, 579–581.
19 U. Voigtmann and S. Blechert, Org. Lett., 2000, 2, 3971–3974.
20 B. Schmidt and H. Wildemann, J. Chem. Soc., Perkin Trans 1, 2000,
2916–2925.
21 B. Schmidt and H. Wildemann, J. Org. Chem., 2000, 65, 5817–5822;
S. Aragone`s, F. Bravo, Y. D´ıaz, M. I. Matheu and S. Castillo´n,
Tetrahedron: Asymmetry, 2003, 14, 1847–1856.
22 B. Schmidt and H. Wildemann, Eur. J. Org. Chem., 2000, 3145–3163;
I. Izzo, P. Del Gaudio, C. Della Monica, M. Squitieri and F. De
Riccardis, Heterocycles, 2003, 60, 2057–2062.
23 L. Banfi and G. Guanti, Eur. J. Org. Chem., 1998, 745–757, and
references therein.
24 G. Guanti, L. Banfi and R. Riva, Tetrahedron: Asymmetry, 1994, 5,
9–12.
25 G. Guanti, L. Banfi and E. Narisano, J. Org. Chem., 1992, 57, 1540–
1554.
Acknowledgements
26 G. Guanti, A. Moro and E. Narisano, Tetrahedron Lett., 2000, 41,
The authors gratefully thank the University of Genova
and M.I.U.R. for financial support and Amano for a generous
gift of lipase from Pseudomonas cepacia.
3203–3207.
27 Part of this work has been presented at the IUPAC 14th International
Conference on Organic Synthesis, Christchurch (NZ), 14–18 July
2002, and in a preliminary communication (ref. 28).
28 G. Guanti and R. Riva, Tetrahedron Lett., 2003, 44, 357–360.
29 Reactions involving the formation of the alkoxide of 3 are usually
accompanied by extensive racemization and low yield of the corre-
sponding alkylated products.
References and notes
1 Carbohydrate Mimetics. Concepts and Methods, ed. Y. Chapleur,
Wiley-VCH, Weinheim, 1998; A. E. Stu¨tz, in Iminosugars as Gly-
cosidase Inhibitors: Nojirimycin and Beyond, Wiley-VCH, Weinheim,
1999; V. H. Lillelund, H. H. Jensen, X. Liang and M. Bols,
Chem. Rev., 2002, 102, 515–553 and references therein;; N. Asano,
Glycobiology, 2003, 13, 93R–104R.
30 L. Banfi, G. Guanti and R. Riva, Tetrahedron: Asymmetry, 1999, 10,
3571–3592.
31 In this case, PPL gave poor results and the hydrolysis was unexpect-
edly slow.
2 E. E. Simanek, G. J. McGarvey, J. A. Jablonowski and C.-H. Wong,
Chem. Rev., 1998, 98, 833–862.
32 X. Ariza, F. Urp´ı, C. Viladomat and J. Vilarrasa, Tetrahedron Lett.,
1998, 39, 9101–9102.
3 A. Mitrakou, N. Tountas, E. Raptis, R. J. Bauer, H. Schulz and S. A.
Raptis, Diabetic. Med., 1998, 5, 657–660.
33 This procedure, notwithstanding the removal of the reagent-derived
by-product being quite troublesome, gave superior results to the
classical protection using Boc2O.
34 Yields using other conditions (LiHMDS, KHMDS or tBuOK as
base, and allyl iodide instead of allyl bromide as alkylating agent)
were even less satisfactory.
35 L. A. Paquette, J. D. Schloss, I. Efremov, F. Fabris, F. Gallou, J.
Me´ndez-Andino and J. Yang, Org. Lett., 2000, 2, 1259–1261.
36 Y. M. Ahn, K. Yang and G. I. Georg, Org. Lett., 2001, 3, 1411–
1413.
4 D. A. Winkler and G. Holan, J. Med. Chem., 1989, 32, 2084–2089; L.
Ratner, N. V. Heyden and D. Dedera, Virology, 1991, 181, 180–192.
5 C. U. Kim, W. Lew, M. A. Williams, H. Liu, L. Zhang, S.
Swaminathan, M. Bischofberger, M. S. Chen, D. B. Mendel, C. Y.
Tai, W. G. Laver and R. C. Stevens, J. Am. Chem. Soc., 1997, 119,
681–690.
6 P. E. Gross, M. A. Baker, J. P. Carver and J. W. Dennis, Clin. Cancer
Res., 1995, 1, 935–944.
7 V. H. Lillelund, H. Liu, X. Liang, H. Søhoel and M. Bols, Org.
Biomol. Chem., 2003, 1, 282–287; R. J. B. H. N. van den Berg, W.
Donker-Koopman, J. H. van Boom, H. M. F. G. Aerts and D. Noort,
Bioorg. Med. Chem., 2004, 12, 891–902.
37 Only recently, a compound related to 26–28,32 has been prepared
by enzymatic kinetic resolution, but without involving an RCM
protocol: H.-J. Altenbach and G. Blanda, Tetrahedron: Asymmetry,
1998, 9, 1519–1524.
8 M. Kummeter and U. Kazmaier, Eur. J. Org. Chem., 2003, 3330–
38 S. U. Hansen and M. Bols, J. Chem. Soc., Perkin Trans 1, 2000,
3334.
911–915.
9 V. Ja¨ger, R. Mu¨ller, T. Leibold, M. Hein, M. Schwarz, M. Fengler, L.
Jaroskowa, M. Pa¨tzel and P.-Y. LeRoy, Bull. Soc. Chim. Belg., 1994,
103, 491–507; M. Lemaire, N. Veny, T. Gefflaut, E. Gallienne, R.
Cheˆnevert and J. Bolte, Synlett, 2002, 1359–1361.
39 E. J. Jacobsen, W. Zhang, A. R. Muci, J. R. Ecker and L. Deng,
J. Am. Chem. Soc., 1991, 113, 7063–7064.
40 E. N. Jacobsen, L. Deng, Y. Furukawa and L. E. Mart´ınez,
Tetrahedron Lett., 1994, 50, 4323–4334; A. Scheurer, P. Mosset, M.
Spiegel and R. W. Saalfrank, Tetrahedron, 1999, 55, 1063–1078.
41 T. W. Greene and P. G. M. Wuts, in Protective Groups in Organic
Synthesis, Wiley, New York, 1999.
10 S. Singh, D. Chikkanna, O. V. Singh and H. Han, Synlett, 2003,
1279–1282.
11 M. Johannsen and K. A. Jørgensen, J. Org. Chem., 1995, 60, 5757–
5762.
12 Only selected references for the formation of heterocycles through
RCM are cited.
13 H. S. Overkleeft and U. K. Pandit, Tetrahedron Lett., 1996, 37, 547–
550; R. Kumareswaran and A. Hassner, Tetrahedron: Asymmetry,
2001, 12, 3409–3415; R. Mart´ın, M. Alco´n, M. A. Perica`s and A.
Riera, J. Org. Chem., 2002, 67, 6896–6901.
14 R. Kumareswaran, T. Balasubramanian and A. Hassner, Tetrahedron
Lett., 2000, 41, 8157–8162; F.-X. Felpin and J. Lebreton, Eur. J. Org.
Chem., 2003, 3963–3712.
42 G. Guanti, L. Banfi, V. Merlo, E. Narisano and S. Thea, Tetrahedron,
1993, 49, 9501–9516.
43 Our results are in agreement with, and even superior to the ones
reported on racemic 28 while our preliminary studies were in
progress: P. Jakobsen, J. M. Lundbeck, M. Kristiansen, J. Breinholt,
H. Demuth, J. Pawlas, M. P. T. Candela, B. Andersen, N. Wester-
gaard, K. Lundgren and N. Asano, Bioorg. Med. Chem., 2001, 9,
733–744.
44 T. M. Jespersen, W. Dong, M. R. Sierks, T. Skrydstrup, I. Lundt
and M. Bols, Angew. Chem., Int. Ed. Engl., 1994, 33, 1778–1779; Y.
Ichikawa, Y. Igarashi, M. Ichikawa and Y. Suhara, J. Am. Chem.
Soc., 1998, 120, 3007–3018; Y. J. Kim, M. Ichikawa and Y. Ichikawa,
15 S. J. Miller and R. H. Grubbs, J. Am. Chem. Soc., 1995, 117, 5855–
5856; E. Jo, Y. Na and S. Chang, Tetrahedron Lett., 1999, 40, 5581–
1 7 3 6
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 7 2 9 – 1 7 3 7

View Article Online
J. Org. Chem., 2000, 65, 2599–2602; G. H. Zhao, U. C. Deo and
B. Ganem, Org. Lett., 2001, 3, 201–203; G. Pandey and M. Kapur,
Synthesis, 2001, 1263–1267; J. Andersch and M. Bols, Chem. Eur. J.,
2001, 7, 3744–3747.
49 T. Hudlicky, M. Mandel, J. Rouden, R. S. Lee, B. Bachmann, T.
Dudding, K. J. Yost and J. S. Merola, J. Chem. Soc., Perkin Trans 1,
1994, 1553–1567; T. Hudlicky, X. Tian, K. Ko¨nigsberger, R. Maurya,
J. Rouden and B. Fan, J. Am. Chem. Soc., 1996, 118, 10752–10765;
P. Magnus and I. K. Sebhat, Tetrahedron, 1998, 54, 15509–15524.
50 Our results, confirmed by several experiments, are in contrast with
the data reported by Jakobsen et al.,43 who separated all the 4
stereoisomers of a racemic mixture of 53,54 (obtained by basic
hydrolysis of racemic 36,40) by chiral HPLC and observed that
stereoisomers with isofagomine relative stereochemistry prevailed,
although the diastereomeric ratio was not given.
45 U. Kazmaier and C. Schneider, Tetrahedron Lett., 1998, 39, 817–818.
46 The low-to-moderate yield in the acidic hydrolysis is a consequence
of the formation of less polar products, most likely arising from
the dehydration of the diols, followed by the tautomerization of the
intermediate enols.
47 T. M. Jespersen, M. Bols, M. R. Sierks and T. Skrydstrup, Tetrahe-
dron, 1994, 50, 13449–13460.
48 K. Tadano, Y. Ueno, C. Fukabori, Y. Hotta and T. Suami, Bull.
Chem. Soc. Jpn., 1987, 60, 1727–1739; K. Tadano, C. Fukabori, M.
Miyazaki, H. Kimura and T. Suami, Bull. Chem. Soc. Jpn., 1987, 60,
2189–2196.
51 Other spectroscopic data agree with those reported in ref. 47.
52 Other spectroscopic data agree with those reported in ref. 38, with
the exception of the 13C NMR, in which, we observed a 0.7 ppm
downfield shift for all the signals with respect to the reported data.
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 7 2 9 – 1 7 3 7
1 7 3 7