Synthesis and conformational analysis of the first 3-oxa-7-thia-1-r- azabicyclo[3.3.0]-c-5-octane single functionalised at the c-5 position

Article abstract of DOI:10.2174/157017810791130658

Starting from TRIS [2-amino-2-(hydroxymethyl)propane-1,3-diol] via its thia analogue, 2-amino-2- (mercaptomethyl)propane-1,3-diol' double cyclocondensation with formaldehyde, we report a four steps synthesis of the first term in the title series of compounds, bearing the exploitable hydroxymethyl functionality at the C-5 position. Conformational analysis is discussed based on DFT calculations and ¹H-DNMR.

Full text of DOI:10.2174/157017810791130658

Letters in Organic Chemistry, 2010, 7, 283-290
283
Synthesis and Conformational Analysis of the First 3-Oxa-7-thia-1-
r-azabicyclo[3.3.0]-c-5-octane Single Functionalised at the C-5 Position
Andrada Buta, Pedro Lameirasb, Ioan Silaghi-Dumitrescua, Carmen Batiua, Sophie Guillardb,
Yvan Ramondencb and Mircea Darabantu*,a
a”Babes-Bolyai” University, Department of Organic Chemistry, 11 Arany Jànos str., 400028 Cluj-Napoca, Romania
b
and Université de Rouen, I. R. C. O. F. (Institut de Recherche en Chimie Organique Fine), BP 08, 76131 Mont Saint-
Aignan Cedex, France
Received February 22, 2009: Revised February 26, 2010: Accepted March 02, 2010
Abstract: Starting from TRIS [2-amino-2-(hydroxymethyl)propane-1,3-diol] via its thia analogue, 2-amino-2-
(mercaptomethyl)propane-1,3-diol’ double cyclocondensation with formaldehyde, we report a four steps synthesis of the
first term in the title series of compounds, bearing the exploitable hydroxymethyl functionality at the C-5 position.
Conformational analysis is discussed based on DFT calculations and 1H-DNMR.
Keywords: Cysteinols, 1,3-oxazolines, NMR, serinols, thionation.
INTRODUCTION
chemistry based on 2-hydroxymethyl-2-aminopropane-1,3-
diol (IIa) [TRIS, “2-(hydroxymethyl)serinol” (IIa)] [4], for
The 3-Oxa-7-Thia-1-AzaBicyclo[3.3.0]Octane (OTABO)
heterocyclic system (Ia) (Fig. 1) is little known since the
period of 1970s being currently available from R or S
cysteine (optionally ethyl cysteinate) via 4-oxo precursors of
type (Ib) [1-3].
example
c-5-hydroxymethyl-3,7-DiOxa-r-1-AzaBicyclo
[3.3.0]Octane [DOABO-CH2OH, (IIIa)] [5], prompted us to
prepare its thia analogue (IIIb) (OTABO-CH2OH) using a
similar as classic chemistry.
OH
1
OH
1
O
O
H
H
3
3
6
6
4
3
7
7
4
2
NH2
IIb
3
2
5
N
5
N
HO
OH HO
SH
HS
OEt
S
O
S
O
1
1
NH2
IIa
NH2
8
2
8
2
Carbonyl
Carbonyl
Ib
Ia
electrophiles
electrophiles
Fig. (1).
CH2OH
5
CH2OH
H
Before 1976, but one C-2, -8-disubstituted derivative of
type (Ia) was published, as a result of the double
cyclocondensation of the in situ reduced form of cysteine,
2-amino-3-mercaptopropanol (“cysteinol”) by treatment
with o-phthaldialdehyde [1a]. Pharmaceutical interest was
claimed regarding this “cysteine synthetic approach”, also
focussed on Ia itself, one year later [1b]. In the period of
1980’, Seebach’s attractive strategy was based as well on R-
cysteine’ diastereoselective cyclisation with pivalaldehyde,
then formaldehyde, followed by the asymmetric
functionalisation, via metallation at the deprotonable C-5
position of the resulting Ib skeleton, with (hetero)aryl-
aldehydes [2]. It highlighted derivatives of type Ib as
enantiopure intermediates in the total syntheses of Biotin,
Micacocidine, Farnesyl Transferase and C-2-functionalised
cysteines [2, 3].
3
7
7
3
5
N
O
S
O
O
3
7
1
1
1
N
N
Pyrrolizidine
DOABO-CH2OH
c-5-Hydroxymethyl-3-Oxa-7-
Thia-r-1-AzaBicyclo[3.3.0]
*Compound IIIb is intrinsic chiral; hereafter Octane, OTABO-CH2OH*
IIIa
only its 1S*,5R* enantiomer will be depicted
in order to show the cis relationship between
ligands at positions 1 and 5.
IIIb
Fig. (2).
In this purpose, we needed 2-amino-2-(mercapto-
methyl)propane-1,3-diol [“2-(hydroxymethyl)cysteinol”] (IIb)
(Fig. 2) as starting material. To the best of our knowledge,
there are only two old patents [6a, 6b] from which just one
[6b] entirely dedicated to the conversion of C-2(CH2OH,
Me, Et)-substituted-serinols into the corresponding C-2-
substituted-cysteinols (e.g. IIa ꢀ IIb). Moreover, cysteinol
(IIb) was much less documented with respect to the C-2(Me,
Et)-substituted cysteinols [6b]. They all were claimed to be
useful in protection of mammalians against radiation, as free
bases or hydrochlorides [6a, 6b].
Our previous expertise in pyrrolizidine’ 3,7-dioxa-
analogs (Fig. 2) functionalised at the C-5 position
*Address correspondence to this author at the “Babes-Bolyai” University,
Department of Chemistry, 11 Arany Jànos str., 400028 Cluj-Napoca,
Romania; Tel: 00 40 264 59 38 33; Fax: 00 40 264 59 08 18;
E-mail: darab@chem.ubbcluj.ro; darabantu@cluj.astral.ro
1570-1786/10 $55.00+.00
© 2010 Bentham Science Publishers Ltd.

284 Letters in Organic Chemistry, 2010, Vol. 7, No. 4
But et al.
CH2OH
CH2OH
1.00 eq. Ph-CO-SH / Py
95 oC / 5 min.
0.75 eq. Ac2O / Toluene
0 oC ꢀ r.t. / 12 hrs.
Toluene
Dean-Stark trap / 24 hrs.
H2N-C(CH2OH)3
[ Ac-NH-C(CH2OH)3 ]
O
N
A - 89%
B - 60%
TRIS (IIa)
1
2
CH3
0.45 eq. aq. K2CO3 / r.t.
10.5 M aq. HCl / Reflux
16 hrs.
HS
OH
Ph-CO-S
OH
OH
HS
OH
OH
C - 90%
Cl- +H3N
OH
Ac-HN
H2N
IIb
3
4
CH2OH
CH2S]2
2.10 eq. H2C=O /
Benzene / Reflux / 4 hrs.
O
S
+
O
O
N
D - 60%
N
OTABO-CH2OH
[DOABO-CH2S]2
IIIb (94%)
5 (6%)
Fig. (3).
Thus, the aim of the present preliminary report consists
of the synthesis and conformational analysis of the
previously unknown OTABO-CH2OH derivative (IIIb) (Fig.
2) starting from the much cheaper TRIS (IIa) via IIb in a
“serinolic synthetic approach”. Except Crabb’s pioneering
studies in conformational analysis of some pyrrolizidine’
3,7-heteroanalogous [7], no similar investigation was
reported so far if the heteroatoms were O-3, S-7.
one week of storageꢁ. Fortunately, even in the last situation,
the mixture (1 + 2) could be resubmitted to the dehydrating
cyclisation conditions (1 ꢀ 2 + H2O, Fig. (3)) affording
again, quantitatively, the above crude product (2) (96%
1H-NMR purity, 4% 1). So, we had to use oxazoline (2) as
such, directly, in the next step of the synthesis (B, Fig. (3))
with no other purification than a simple trituration with
excess of ether, intended to discard the residual amounts of
acetic acid.
In step B, ring cleavage of 2 yielded the S-, N-protected
form (3) of the cysteinol (IIb) (60% yield, lit. 57%, [6a]). In
spite of the excellent NMR appearance of crude 3 as unique
product (71% yield), it reached analytical purity only after
two recrystallisations from hot ethanol.
RESULTS AND DISCUSSION
1. Synthesis
Since we first targeted cysteinol (IIb) (Fig. 2), we had to
re-examine its three steps obtention (A - C) (Fig. 3) by
which to access the starting material for the OTABO-
CH2OH (IIIb) derivative preparation (step D, Fig. (3)). All
syntheses were carried out in habitual conditions and rapidly
mandatory to essential improvements in comparison with
earlier literature descriptions [6a, 6b].
In step C, we N-, S-deprotected the thioamidoester (3) in
boiling 10.5 M aq. HCl with 90% yield (lit. 68%, [6b]) and
isolated the cysteinol IIb as hydrochloride (4) by applying
one of our previous patented methodologies [9] (see
Experimental Part). Compound (4) was stable on storage in
normal conditions indefinitely.
In step A, one of the three methylenes groups of TRIS
was activated by incorporation into a 1,3-oxazoline ring (2).
Rather than manipulating, as previously reported [6a, 6b],
with excess of boiling acetic acid and azeotropic removal of
water followed by fractional distillation of crude 2 under
reduced pressure, then recrystallisation from THF (52%
yield), we opted for a more reactive electrophile, acetic
anhydride in toluene.
Although 0.5 molar equivalents of acetic anhydrideꢀ had
theoretically ensured that all acetyl groups of TRIS were
included in the 1,3-oxazoline (2) thus preventing any
statistically favoured side O-acylation, we used 0.75 eq. of
Ac2O and separated the resulting water with a Dean-Stark
trap. Despite of the yield as 93 % of crude 2 (96 % 1H-NMR
purity and 4% TRIS’ amide (1)), in our hands, oxazoline (2)
was unstable [8]. For example, in the crude freshly isolated
(2), the content of 1 increased to 15% within drying 24 hr. at
room temperature to reach, in the same conditions, 40% after
With 4 in our hands, cysteinol (IIb) was generated in situ
(step D), then trapped with paraformaldehyde in a standard
protocol, earlier used for the double cycloaminalisation of
TRIS [4, 5] giving the desired OTABO-CH2OH derivative
(IIIb) as an non-crystallisable oil in 60% yield after
purification by flash column chromatography. The NMR
spectra of the crude reaction mixture showed IIIb as largely
major (94%) with respect to the side product, the disulphide
5 (6%) of type [DOABO-CH2S]2. The latter originated most
likely from a preliminary partial oxidation of IIIb followed
by the double cyclisation involving its remaining
nucleophilic sites, the hydroxymethyl groups.
2. Conformational Analysis
Having two stereogenic centres, N-1 and C-5, the
OTABO bicyclic skeleton is both chiral and heterofacial,
ꢀIn both Patents [6] the use of 1.5 eq. of anhydrous AcOH is recommended for the
ꢁIt appeared to us that 2 was very sensitive to moisture, especially if residual AcOH
synthesis of oxazoline (2).
was present in the crude material.

Synthesis and Conformational Analysis
Letters in Organic Chemistry, 2010, Vol. 7, No. 4
CH2OH
CH2OH
285
CH2OH
CH2OH
4
6
Cis face
7
3
7
7
3
7
5
N
7
X
3
3
3
O
O
X
X
O
X
O
X
O
1
N
N
N
N
Trans face
2
8
O-3-anti-X-7-syn
IIIa-a-s
IIIb-O-a-S-s
O-3-anti-X-7-anti
IIIa-a-a
IIIb-a-a
O-3-syn-X-7-anti
IIIa-s-a
IIIb-O-s-S-a
O-3-syn-X-7-syn
IIIa-s-s
IIIb-s-s
IIIa: X = O (DOABO-CH2OH)
IIIb: X = S (OTABO-CH2OH)
Fig. (4).
Fig. (5).
shaped as an expected cis fused oxazolidine / thiazolidine
system (lone pair at N-1 and C-5-hydroxymethyl group as
references, Fig. (4)) [7]. As in the case of the earlier
elucidated by us 3,7-dioxa analogue (IIIa) (Fig. 2) [4], we
primarily applied DFT computational methods on IIIb, then
attempted to corroborate the results with those provided by
high resolution 1H-DNMR Experiments.
enantiomeric conformers (IIIa-a-s) / (IIIa-s-a) both in gas
phase and in solution. Therefore, the dynamic behaviour of
IIIa could be limited to two diastereomeric inversions, a-s
ꢀ a-a ꢀ s-a, in an overall enantiomeric interconversion.
However, the magnitude of the calculated ꢀꢀE = ꢀE(IIIa-a-
a) – ꢀE(IIIa-s-a) positive values (0.84 – 5.48 kJ/mol) did
not allow an a priori assignment of the IIIa frozen
conformation both in gas phase and in solution [4b]ꢀ.
2.1. Conformational Analysis Based on DFT Calculation at
B3LYP/6-31G(d) Level
When one of the oxygen atoms (atomic radius 60 pm) in
IIIa was replaced by a sulphur (atomic radius 100pm, IIIb),
some of the above considerations also applied to conformers
of IIIb. Except IIIb-s-s and IIIb-O-a-S-s, only those having
the relative energy less than 4.00 kJ/mol were taken into
account and depicted in Fig. (5).
First of all, the lowest energy conformers of IIIb,
generated by Spartan’04 with the MMFF force field, have
been subjected to the full geometry optimisation at the
B3LYP/6-31G(d) level of theory. Further, the effect of
1
solvent (THF, also used in H-DNMR investigations) has
been taken into account by performing SCRF calculation
with COSMO (CPCM) option in Gaussian 98 [10] (Fig. (4),
Fig. (5)).
As for IIIa, a typical O-3-syn-S-7-syn geometry (Fig. (5),
IIIb-(s-s)), requiring cis protons pseudo-equatorial and trans
protons pseudo-axial, was very disfavored. The relative
energy of other (IIIb-s-s) conformers, exhibiting both thia-
and oxazolidine rings very distorted from the model
envelope shape with the heteroatoms O-3, S-7 out of plane,
ranged between 12 – 15 kJ/mol. That is, the occurrence of a
(s-s) spatial arrangement of the OTABO skeleton was once
again ruled out. The same, in series (IIIb-a-a), the lowest
energy conformers, (IIIb-a-a-1) and (IIIb-a-a-2), had no
In an identical approach, we previously established for
IIIa an oriented conformational flexibility comprising three
diastereomeric equilibria (Fig. 4) discriminated by the sense
of the puckering of the two oxazolidine rings. They were
disclosed to be syn or anti O-3/O-7 envelope diastereomers
with respect to the fiducial substituent at N-1 [4b, 11]. In our
analysis, we considered a single oxazolidine ring inversion /
equilibrium and discriminated the stereoisomers of IIIa in
terms of conformational chirality [4b, 4c]. Thus, the meso-
form (IIIa-s-s) was ruled out since it was higher in energy
by ꢀꢀE = 12.3 – 18.9 kJ/mol with respect to the
ꢀIIIa-a-a was determined to be the frozen conformation both in solution (1H-DNMR,
ꢀGꢁ = 53.252 kJ/mol) and in solid state (single crystal X-ray diffractometry) [4b]: the
global shape of IIIa as two fused envelope oxazolidine rings having the oxygen atoms
O-3, -7 out of plane was found.

286 Letters in Organic Chemistry, 2010, Vol. 7, No. 4
But et al.
authentic O-3-anti-S-7-anti fused envelope form with cis
protons pseudo-axial and trans protons pseudo-equatorial.
This was exhibited only by the conformers in series (IIIb-a-
a) having the relative energy in the domain 4.39 – 11.92
kJ/mol.
Before checking the validity of the above calculation, we
overall concluded that: i) The ꢂꢂE values between the four
lowest energy conformers of IIIb (Fig. 5) were, certainly,
too small (less than 0.8 kcal/mol) precluding a definite a
priori assignment of the frozen structure; ii) The
conformational behavior of the oxazolidine ring can be
described simply as an (envelope-O-3-syn) ꢀ (envelope-O-
3-anti) equilibrium meanwhile for the thiazolidine ring, an
When only the two pentaheterocycles adopted an
opposite orientation anti/syn, as in IIIb-O-a-S-s, IIIb-O-s-S-
a-1 and IIIb-O-s-S-a-2, one can assign two genuine fused
envelope conformations with a different as clear sense of
their puckering.
(envelope-S-7-anti)
ꢀ
(distorted
envelope-S-7-anti)
equilibrium was to be anticipated; iii) A preferred “in”
orientation of the hydroxymethyl group should not be ruled
out.
2.2. Conformational Analysis Based on 1H-DNMR Results
In the most stable conformer, (IIIb-O-s-S-a-1), the
hydroxymethyl group was positioned towards the cis face of
the OTABO system suggesting a stabilizing intramolecular
hydrogen bond with the lone pair of the bridged nitrogen in a
five membered internal chelate. We will hereafter entitle this
orientation as “in” rotamer. Indeed, following this idea, in
IIIb-O-s-S-a-1, the exocyclic methylene group at C-5-
position encountered no steric hindrance with the pseudo-
equatorial protons H-2-c and H-4-c. An alternative “in”
orientation, found in IIIb-O-s-S-a-2, promoted a + 2.11
kJ/mol energetic gain, presumably because of a possible
steric repulsion with the pseudo-axial protons H-6-c and H-
8-c.
1
1D-1H-, and 2D-1H, H-NMR data to be discussed are
collected in Table 1 and 2.
1
2D-1H, H-NOESY Experiments confirmed the cis-fused
construction of the OTABO skeleton by the dipolar
interactions of the hydroxymethyl group with four
heterocyclic protons, normally located on the same
molecular side, hence homofacial and designed as cis (H-c,
Fig. (5), Tables 1, 2). Typically, their ꢀ values were
established throughout downfield with respect to those of the
corresponding geminal H-t. Geminal couplings exhibited
2
Nevertheless, not the “in” rotamerism, exhibited by
conformers (IIIb-O-a-S-s), (IIIb-a-a-1) (or -2), (IIIb-O-s-S-
a-1) (or -2) against the “out” one, as in IIIb-s-s, seemed to
be the major factor in discriminating stability but the O-3-
syn-S-7-anti spatial arrangement of the OTABO skeleton.
normal values except the unexpected as very small J at C-2
proving a highly increased H-C-H bond angle at this position
[7, 14].
Next, not only that the two pentaheterocycles were very
1
different from H-NMR point of view (Table 1) but also the
1
Thus, the lowest relative energy of IIIb-O-s-S-a-1 could
be more plausibly explained according to expected
stereoelectronic effects in the O-3-C-2-N-1-C-8-S-7
(thio)aminalic part. The calculated bond lengths (Fig. 5) in
this region revealed noticeable contractions of the bonds O-
3-C-2 vs. C-4-O-3 (- 0.016 Å) and N-1-C-8 vs. N-1-C-5 (-
0.049 Å). The same type and magnitudes of contractions
were previously observed in X-ray crystallographically
determined structures of a variety of DOABO derivatives,
including IIIa [4b, 4c, 12]. They were rationalized recently
by Pavia [12] then by us [4b, 4c] in terms of a cross endo-
anomeric effect seen, in the case of IIIb-O-s-S-a-1, as two
identically oriented hyperconjugative interactions lp-O-3-eq.
ꢀ ꢁ*(C-2-N-1) and lp-N-1-ax. ꢀ ꢁ*(C-8-S-7) (Fig. 6).
oxazolidine zone of IIIb exhibited dissimilar H-NMR data
with respect to IIIa [4b] and even against 5.
Differently than IIIa, by progressive decreasing of the
temperature, coalescence occurred at none of the
heterocyclic methylenes of IIIb, preventing us to establish
its frozen conformation. The single dynamic modification
observed consisted of a slow exchange status reached by the
hydroxyl proton whose initial broad singlet (293 K), was
converted into a broad triplet (173 K, decoalescence at 193
K) in tandem with the expected shifting downfield of the
signal as + 0.87 ppm. The supplementary splitting of the
original AX system into AMX involving the adjacent
methylene group occurred accordingly. From the evolution
of the hydroxyl proton, whose lifetime in a spin state at
coalescence was ꢃ = 0.082 sec., we deduced ꢃ-1 = 13
interchanges / sec. at 193 K meanwhile, at 173 K, the same
value should be greater than 0.18 sec [13] in a chelatizing
environment ꢁ. That is, at 173 K, the OTABO skeleton was
flipping faster than the hydroxyl proton interchange.
Furthermore, regardless the temperature, methylene protons
of the C-5-CH2OH group developed NOESY interactions
with almost all H-c nuclei in agreement with the free rotation
about C-5-CH2 bond, obscuring any assignment of a
dominant “in” vs. “out”rotamerism.
CH2OH
CH2OH
+
-
O
S
O
S
-
N
+
N
IIIb
non-bonding-oxa
IIIb-O-s-S-a-1
IIIb
bonding
non-bonding-thia
Upon cooling, 2D-1H, 1H-NOESY Experiments also
evidenced the flipping of IIIb because simultaneous dipolar
interactions (symbolised hereafter as “ꢄ”) between pseudo-
Fig. (6).
These delocalisations, illustrated by bonding / non-
bonding formulae [13] (oxa- and thia) increased the O-3-C-2
and N-1-C-8 bond orders. They should due to the near
antiperiplanar arrangement between the lp-O-3-eq. vs. bond
C-2-N-1 and lp-N-1-ax. vs. bond C-8-S-7 in an O-3-syn-S-7-
anti benefic geometry.
ꢁCalculated based on the 3J = 5.5 Hz value (Table 1, 173 K) of the OH proton as ꢃ (life
time of the spin state at coalescene) = 20..5 / J ꢅ = 20.5 / 5.5 ꢅ = 0.082 sec. and ꢃ (life
time of the spin state) at 173 K as ꢃ > J-1 = 0.18 sec [14].

Synthesis and Conformational Analysis
Letters in Organic Chemistry, 2010, Vol. 7, No. 4
287
Table 1. 1H-NMR Data of Compound (IIIb) [ꢀ (ppm), J (Hz)] (500 MHz Time Scale, [D8]THF, Appendix 1, 2)
OH
H-2-ca
H-2-ta
H-8-c
H-4-c
H-8-t
CH2OH
H-4-t
CH2OH
H-6-c
H-6-t
293 K
4.03
(bs)
-
4.58
(d)
4.13
(d)
4.04
(d)
3.97
(d)
3.87
(d)
3.49
(d)
3.46
(d)
3.38
(d)
3.03
(d)
2.78
(d)
ꢀ
J
173 K
ꢀ
2.8
1.5
10.8
3.93
(d)
8.5
10.8
3.91
(d)
10.5
8.5
10.5
11.5
12.0
4.90
(dd)
5.5
4.62
(d)
3.96
(d)
4.01
(d)
3.46
3.24
(d)
3.28
2.95
(d)
2.82
(d)
(dd)
(dd)
J
11.0
- 0.11
8.5
11.0
+ 0.04
6.3, 10.3
- 0.03
8.5
4.3, 10.3
- 0.10
ꢀꢀb
+0.87
+ 0.04
- 0.17
+ 0.04
- 0.22
- 0.08
+ 0.04
aSpatial location of the heterofacial protons (Fig. 5) deduced from 2D-1H, 1H-NOESY Charts (Table 2). bFluctuations of the chemical shift upon cooling from 293 K to 173 K:
ꢀꢀ(ppm) = ꢀ(173 K) - ꢀ(293 K).
Table 2. Dipolar Interactions in 2D-1H, 1H-NOESY Experiments Performed on Compound IIIba (Appendix 3, 4)
OH
H-2-c
H-2-t
H-8-c
H-4-c
H-8-t
CH2O
H-4-t
CH2O
H-6-c
H-6-t
ꢀ
ꢁ
OH
ꢀ
ub
u
u
u
u
s
-
ꢀ
s
-
s
-
-
-
sb
-
-
s
-
s
-
-
-
mb
-
-
-
-
-
H-2-c
H-2-t
H-8-c
H-4-c
H-8-t
CH2O
H-4-t
CH2O
H-6-c
H-6-t
ꢀ
-
-
s
-
s
-
-
w
w
-
ꢀ
-
-
s
s
-
m
m
-
s
-
-
ꢀ
-
-
s
s
-
-
s
s
ꢀ
-
-
-
-
-
wb
-
w
-
w
s
ꢀ
-
u
ꢀ
-
s
s
-
-
-
s
-
u
-
s
s
w
-
-
s
s
-
s
ꢀ
s
s
-
s
-
s
-
-
w
w
-
ꢀ
s
m
-
-
-
-
s
s
-
s
ꢀ
aRight up zone vs. diagonal: dipolar interactions detected at 293 K; left down zone vs. diagonal (italicised and boldface): dipolar interactions detected at 173 K; bu (uncertain), s
(strong), m (medium), w (weak) interactions.
axial homofacial oxazolidine protons and some of
thiazolidine, in spite of their cis or trans location (e.g. H-2-c
ꢁ H-4-c, H-2-t ꢁ H-4-t and H-8-c ꢁ H-6-c) were
observed. However, relevant for an oriented flipping of IIIb
were the spatial interactions implying oxazolidine vs.
thiazolidine homofacial H-t protons (H-2-t ꢁ H-8-t and H-4-
t ꢁ H-6-t).
geminal anisochrony as 0.43 ppm at C-2 and 0.51 ppm at C-
4 was observed. It appreciably increased at 173 K as 0.66
ppm at C-2 and 0.77 ppm at C-4. In contrast, in the
thiazolidine counterpart, not only that geminal anisochrony
was smaller, 0.17 ppm at C-8 and 0.25 ppm at C-6 (293 K),
0.02 ppm at C-8 and 0.13 ppm at C-6 (173 K) but also the
influence of decreasing the temperature was highly reversed.
ꢁꢀ values of oxazolidine protons H-2-t and H-4-t were
shifted upfield the most, - 0.17 and - 0.22 ppm respectively,
suggesting, at 173 K, their dominant location in pseudo-axial
positions, remote from the equatorial lone pair at O-3, for
example as in a conformer of type IIIb-O-s-S-a (Fig. (5),
Table 1). Mutatis-mutandis, the ꢁꢀ values of thiazolidine
In this last context, since all chemical shifts of IIIb in
Table 1 were, in fact, mediated ꢀ values between more or
less equally populated conformational equilibria, we focused
on their fluctuations following the diminishing of the
temperature (only fluctuations greater than ± 0.04 ppm were
considered). At 293 K, in the oxazolidine ring, an important

288 Letters in Organic Chemistry, 2010, Vol. 7, No. 4
But et al.
protons H-8-c (- 0.11 ppm) and H-6-c (- 0.08 ppm), denoting
a weaker shielding, were symptomatic as well for a pseudo-
axial orientation. That is, in Newman projection (Table 1),
the bonds C-8-H-8-c and C-6-H-6-c do not more bisect,
progressively, the angle between the lone pairs at S-7. If so,
the prevailing geometry of the thiazolidine motif was, once
again, S-7-anti, hence, a global IIIb-O-s-S-a stereochemistry
of IIIb.
was observed on TLC. At this step, starting from refluxing
toluene, the reaction mixture, as a fine colourless emulsion,
was let very gently to reach room temperature under
vigorous stirring in order to ensure successfully the obtention
of a fine white suspension. After filtering, well washing with
dry ether (ꢂ 25 mL) and drying at room temperature within
one hour, 9.45 g of the crude title compound 2 were obtained
(yield 89% with respect to TRISꢀ; 1H-NMR purity 96%; 4%
compound 1).
2-Acetamido-2-(hydroxymethyl)propane-1,3-diol (1)
CONCLUSION
1H-NMR (300 MHz, [D6]DMSO): ꢀ = 1.79 (s, 3 H, CH3),
3.51 (s, 6 H, CH2OH), 4.54 (bs, 3 H, CH2OH), 7.23 (bs, 1 H,
NH) ppm. 13C-NMR (75 MHz, [D6]DMSO) : ꢀ = 24.0 (1 C,
CH3), 60.0 (1 C, Cq., C-2), 60.8 (3 C, CH2OH), 175.2 (1 C,
>C=O) ppm.
In a four step synthesis, TRIS provided the first C-5
hydroxymethylated 3-oxa-7-thia-1-azabicyclo[3.3.0]octane
(OTABO-CH2OH) in 29% overall yield. The key
intermediate, 2-amino-2-(mercaptomethyl)propane-1,3-diol
hydrochloride, was obtained in a 48 % overall yield (lit. 20%
[6a, 6b]). DFT optimisation of the new derivative OTABO-
CH2OH predicted its oriented conformational flexibility, as a
cis-fused structure in which oxazolidine and thiazolidine
rings have an opposite envelope orientation, O-3-syn vs. S-7-
anti. 1H-DNMR and NOESY Experiments confirmed the
flipping of the molecule. Although no frozen stereoisomer
was assigned, one can presume the shifting of the
conformational equilibria of the OTABO skeleton towards
an O-3-syn-S-7-anti dominant geometry.
4,4-Bis(hydroxymethyl)-2-methyl-1,3-oxazoline (2)
Yield 89%, (9.45 g, 62.3 mmol 2 starting from 8.48 g,
o
TRIS) white crystalline powder, m.p. 82–84 C (Et2O) [lit.
o
o
[6a] 95 – 97 C (CHCl3/Et2O/AcOEt); lit. [6b] 88 – 89 C
(THF)]. Rf (75% toluene/EtOH) = 0.45. IR (KBr): ꢃ = 3360
(s), 3271 (s), 3108 (s), 2982 (s), 2941 (s), 2872 (s), 1670 (s),
1629 (m), 1572 (m), 1459 (m), 1384 (s), 1270 (s), 1229 (w),
1187 (w), 1143 (w), 1029 (s), 994 (s), 972 (w), 942 (w), 882
(w), 843 (w), 691 (w), 649 (w), 625 (w), 588 (w), 525 (w)
1
cm–1. H-NMR (300 MHz, [D6]DMSO): ꢀ = 1.84 (s, 3 H,
2
CH3), 3.30 (d, 2 H, JH,H = 11.0 Hz, CH2OH), 3.36 (d, 2 H,
EXPERIMENTAL PART
General
2JH,H = 11.0 Hz, CH2OH), 4.02 (s, 2 H, H-5), 4.54 (bs, 2 H,
CH2OH ) ppm. 13C-NMR (75 MHz, [D6]DMSO): ꢀ = 13.6 (1
C, CH3), 63.9 (2 C, CH2OH), 70.3 (1 C, C-4), 76.4 (1 C, C-
5), 163.4 (1 C, C-2) ppm. MS (positive CI, isobutane, 200
eV): m/z (%) = 202 (17) [M+i-BuH-2]+, 188 (7) [M+42]+,
164 (15) [M+18]+, 146 (100) [M+1]+, 114 (6), 73 (< 5);
C6H11NO3 (145.07).
Melting points are uncorrected; they were carried out on
ELECTROTHERMALꢀ instrument. Conventional NMR
spectra were recorded on a Brukerꢀ AM 300 instrument
operating at 300 and 75 MHz for 1H and 13C nuclei
respectively. 1H-DNMR spectra were recorded on a Brukerꢀ
1
DMX500 instrument operating at 500 and 125 MHz for H
Preparation of 2-acetamido-2-(benzoylthiomethyl)propane-
1,3-diol (3)
and 13C nuclei respectively. All NMR spectra were measured
in anhydrous commercially available deuteriated solvents.
No SiMe4 was added; chemical shifts were measured against
the solvent peak. All chemical shifts (ꢁ values) are given
throughout in ppm; all coupling patterns (nJH,H values) are
given throughout in Hz. TLC was performed by using
aluminium sheets with silica gel 60 F254 (Merckꢀ); flash
column chromatography was conducted on Silica gel Si 60
(40 – 63 μm, Merckꢀ). IR spectra were performed on a
Perkin-Elmerꢀ Paragom FT-IR spectrometer. Only relevant
absorptions are listed [throughout in cm-1: weak (w), medium
(m) or (s) strong]. Microanalyses were performed on a Carlo
Erbaꢀ CHNOS 1160 apparatus. Mass spectra (MS) were
recorded on a Brukerꢀ Esquire Instrument.
In a vigorously stirred solution prepared from 4,4-
bis(hydroxymethyl)-2-methyl-1,3-oxazoline (2) (3.68 g,
25.35 mmol) dissolved in dry pyridine (10.00 mL, 10.50 g,
133 mmol), fresch thiobenzoic acid (3.30 mL, 3.50 g 100%,
25.35 mmol) was rapidly injected at room temperature. The
resulted yellow solution was heated at 95 oC and kept at this
temperature for 5 min. A clear orange-reddish solution was
o
obtained which was cooled at 0 C for 30 min. then poured
on aq. HCl (45.00 mL, 47.70 g, 4 M aq. HCl, 179 mmol).
Crude 3 crystallised as a yellow mass (pH = 0.5 - 1) which
o
was cooled for additional 12 hrs. at 0 C. After filtering,
washing with cooled water (3 ꢁ 15 mL) and drying at r.t.,
6.90 g crude 3 were obtained as a yellow amorphous powder.
This was twice recrystallised from boiling ethanol (15 mL)
to yield 4.30 g pure 3 as a white crystalline powder (60%
yield with respect to 2).
Preparation of 4,4-bis(hydroxymethyl)-2-methyl-1,3-oxazo-
line (2)
o
To TRISꢀ (8.48 g, 70 mmol) suspended in cooled (0 C)
dry toluene (125 mL), acetic anhydride (5.00 mL, 5.36 g,
52.5 mmol) was rapidly injected under vigorous stirring. The
resulted suspension was stirred and let to reach room
temperature overnight (12 hrs.) then heated at reflux for
additional 12 hrs. with continuous removal of water in a
Dean-Stark trap (TLC monitoring, eluent toluene : ethanol
3/1 v/v, visualisation in a I2 bath). The reaction was stopped
when no more water was separated and no more evolution
2-Acetamido-2-(benzoylthiomethyl)propane-1,3-diol (3)
Yield 60% (4.30 g, 15.21 mmol 3 starting from 3.68 g,
o
25.35 mmol 2) white crystalline powder, m.p. 143–145 C
o
o
(EtOH) [lit. [6a] 147 – 148 C (-); lit. [6b] 146 – 147 C
(EtOH)]. C13H17NO4S (283.09): calcd. C 55.11, H 6.05, N
4.94; found C 54.88, H 5.88, N 5.29. Rf (75% toluene/EtOH)
= 0.75. IR (KBr): ꢃ = 3380 (s), 3344 (m), 3153 (m), 2942

Synthesis and Conformational Analysis
Letters in Organic Chemistry, 2010, Vol. 7, No. 4
289
(m), 2896 (m), 2837 (m), 1662 (s), 1648 (s), 1557 (s), 1450
(m), 1372 (m), 1323 (w), 1205 (s), 1176 (m), 1079 (m), 1064
(s), 969 (w), 913 (s), 774 (m), 690 (m), 645 (m), 596 (w),
mL) and the resulted solution was added to anh. potasssium
carbonate (0.380 g, 2.75 mmol) and paraformaldehyde (0.38
g, 12.72 mmol) suspended in benzene (40 mL). The reaction
mixture was heated with stirring at reflux with continuous
removal of water (Dean-Stark trap). When no more water
separated (about. 4 hrs.), the reactoion mixture was cooled at
room temperature then the benzenic solution was decanted
and evaporated under reduced presure to give crude IIIb as
an oily mass. This was purified by flash column
chromatography (eluent ligroin : acetone 2:1 v/v) to yield
pure IIIb (0.567 g, 60% yield withy respect to 4) as the first
fraction; complete elution of the column provided additional
0.100g as a equimolar mixture mixture IIIb : 5 (1H-NMR
monitoring).
1
554 (m), 533 (w) cm–1. H-NMR (300 MHz, [D6]DMSO): ꢀ
2
3
= 1.86 (s, 3 H, CH3), 3.62 (dd, 2 H, JH,H = 10.3 Hz, JH,H
=
2
3
5.7 Hz, CH2OH), 3.61 (dd, 2 H, JH,H = 10.3 Hz, JH,H = 4.6
Hz, CH2OH), 4.99 (dd as t, 3JH,H = 5.7 Hz, CH2OH ), 7.39 (s,
1 H, NH), 7.59 (t, 2 H, 3JH,H = 7.5 Hz, H-3, Ph), 7.72 (t, 1 H,
3JH,H = 7.4 Hz, H-4, Ph), 7.95 (d, 2 H, JH,H = 7.5 Hz, H-2,
3
Ph) ppm. 13C-NMR (75 MHz, [D6]DMSO): ꢀ = 23.4 (1 C,
CH3), 30.3 (1 C, CH2S), 60.3 (2 C, CH2OH), 61.2 (1 C, C-2),
126.8 (2 C, C-2, -6, Ph), 129.0 (2 C, C-3, -5, Ph), 133.7 (1 C,
C-4, Ph), 136.5 (1 C, C-1., Ph), 170.3 [1 C, >C(=O)-NH-),
191.1 [1 C, >C(=O)-S-] ppm. MS (positive CI, isobutane,
200 eV): m/z (%) = 340 (< 5) [M+i-BuH-1]+, 284 [M+1]+
(48), 266 (8), 252(< 5), 236 (5), 218 (7), 180 (100), 162 (10),
148 (8), 130 (< 5), 123 (10), 116 (< 5), 73 (< 5).
Rac-c-5-hydroxymethyl-3-oxa-7-thia-r-1-azabicyclo [3.3.0]
octane (IIIb)
Yield 60% (0.567 g, 3.52 mmol IIIb starting from 1.00 g,
5.78 mmol 4) yellowish oil (flash column chromatography,
eluent ligroin : acetone 2:1 v/v). Rf (66% ligroin/acetone) =
0.75. C6H11NO2S (161.05): calcd. C 44.70, H 6.88, N 8.69;
found C 45.03, H 6.68, N 8.82. IR (KBr): ꢀ = 3417 (s), 2931
(s), 2868 (s), 1715 (w), 1644 (w), 1455 (m), 1434 (m), 1398
(m), 1239 (m), 1208 (m), 1156 (m), 1122 (s), 1069 (s), 959
(s), 919 (m), 779 (w), 738 (m), 708 (m), 654 (m), 615 (w),
572 (w) cm–1. 1H-NMR (300 MHz, [D6]DMSO): ꢀ = 2.84 (d,
1 H, 2JH,H = 11.8 Hz, H-6-t), 3.00 (d, 1 H, 3JH,H = 11.8 Hz, H-
6-c), 3.42 (dd, 1 H, 2JH,H = 11.7 Hz, 3JH,H = 5.5 Hz, CH2OH),
3.33 (dd, 1 H, 2JH,H = 11.7 Hz, 3JH,H = 5.5 Hz, CH2OH), 3.48
Preparation of 2-amino-2-(mercaptomethyl)propane-1,3-
diol hydrochloride (4)
2-Acetamido-2-(benzoylthiomethyl)propane-1,3-diol (3)
(2.41 g, 8.50 mmol) was suspended in aq. HCl (7.40 mL,
8.57 g, 10.5 M aq. HCl, 77.8 mmol) and the reaction mixture
was heated at reflux with stirring for 16 hrs. After cooling at
room temperature, benzoic acid crystallised abundantly and
was filtered off, washed with aq. HCl (3 ꢁ 5 mL, 10.5 M aq.
HCl) to yield 0.93 g pure by product (90% with respect to
the theoretical amount). The combined aqueous filtrate was
added to benzene (100 mL) and the resulted mixture was
heated with stirring at reflux with azeotropic removal of
water (Dean-Stark trap). During anhydrification, compound
4 precipitated as an oily mass. When no more water was
separated, the mixture was kept at reflux for additional 2 hrs.
in order to eliminate thes excess of hydrochloric acid. After
cooling at room temperature, benzene was decanted and
2
2
(d, 1 H, JH,H = 8.5 Hz, H-4-t), 3.96 (d, 1 H, JH,H = 8.5 Hz,
2
H-4-c), 3.99 (d, 1 H, JH,H = 10.8 Hz, H-8-t), 4.04 (1 H, d,
2
2JH,H = 10.8 Hz, H-8-c), 4.13 (1 H, d, JH,H = 3.3 Hz, H-2-t),
3
4.57 (d, 1 H, 2JH,H = 3.3 Hz, H-2-c), 5.01 (1 H, dd as t, JH,H
= 5.5 Hz, CH2OH) ppm. 13C-NMR (75 MHz, [D6]DMSO): ꢀ
= 39.4 (1 C, C-6), 57.4 (1 C, C-8), 65.3 (1 C, CH2OH), 73.9
(1 C, C-4), 77.8 (1 C, C-5), 86.9 (1 C, C-2) ppm; 13C-NMR
(125 MHz, [D8]THF, 293 K): ꢀ = 40.2 (1 C, C-6), 58.4 (1 C,
C-8), 66.9 (1 C, CH2OH), 75.11 (1 C, C-4), 79.16 (1 C, C-5),
88.5 (1 C, C-2) ppm. MS (positive CI, isobutane, 200 eV):
m/z (%) = 218 [M+i-BuH-1]+ (13), 204 [M+43]+ (7), 178
[M+17]+ (<5), 162 [M+1]+ (100), 144 [M-17]+ (<5), 130
(10).
o
crude oily 4 was crystallised at 0 C from 10 mL 1:1 v/v
isopropanol : ether to give 1.32 g pure 4 (90% yield with
respect to 3).
2-Amino-2-(mercaptomethyl)propane-1,3-diol hydrochlori-
de (4)
Yield 90% (1.32 g, 7.65 mmol 4 starting from 2.41 g,
8.50 mmol 3) white crystalline powder, m.p. 102 - 104 oC (i-
PrOH/Et2O 1:1 v/v) [lit. [6b] 104 – 105 oC (i-PrOH)].
C4H12ClNO2S (173.03): calcd. C 27.66, H 6.96, N 8.07;
found C 28.02, H 6.88, N 7.99. IR (KBr): ꢀ = 3344 (s), 3265
(s), 3008 (s), 2554 (m), 1602 (m), 1542 (m), 1514 (s), 1458
(m), 1399 (m), 1277 (m), 1247 (m), 1112 (m), 1062 (s), 918
(m), 671 (m), 546 9w) cm–1. 1H-NMR (300 MHz,
Bis(3,7-dioxa-r-1-azabicyclo[3.3.0]octan-c-5-yl)-disulphide
(5)
0.100 g equimolar mixture with IIIb, as a second fraction
in the elution of pure IIIb. Viscous oil. Rf (66%
1
ligroin/acetone) = 0.65. H-NMR (300 MHz, [D6]DMSO): ꢀ
= 3.21 (s, 4 H, CH2S), 3.73 (d, 4 H, 2JH,H = 9.0 Hz, H-4, -4’, -
6, -6’-t), 3.85 (d, 4 H, 2JH,H = 9.0 Hz, H-4, -4’, -6, -6’-c), 4.35
(d, 4 H, 2JH,H = 5.5 Hz, H-2, -2’, -8, -8’-t), 4.48 (d, 4 H, 2JH,H
= 5.5 Hz, H-2, -2’, -8, -8’-c) ppm. 13C-NMR (75 MHz,
[D6]DMSO): ꢀ = 45.9 (2 C, CH2S), 72.0 (4 C, C-4, -4’, -6, -
6’), 74.6 (2 C, C-5, -5’), 87.7 (4 C, C-2, -2’ -8, -8’) ppm. MS
(positive CI, isobutane, 200 eV): m/z (%) = 377 [M+i-BuH-
1]+ (15), 321 [M+1]+ (80).
3
[D6]DMSO): ꢀ = 2.74 (d, 2 H, JH,H = 9.2 Hz, CH2SH), 3.03
3
2
(d, 1 H, JH,H = 9.1 Hz, CH22SH), 3.52 (d, 2 H, JH,H = 11.6
Hz, CH2OH), 3.58 (d, 2 H, JH,H = 11.6 Hz, CH2OH), 5.40
(bs, 2 H, CH2OH), 8.03 (s, 3 H, NH3 ) ppm. 13C-NMR (75
+
MHz, [D6]DMSO): ꢀ = 24.5 (1 C, CH2S), 59.7 (1 C, C-2),
61.0 (2 C, CH2OH) ppm. MS (positive CI, isobutane, 200
eV): m/z (%) = 176 [M+3]+, 152 (7), 138 (100), 103 (9), 90
(6).
Preparation of c-5-hydroxymethyl-3-oxa-7-thia-r-1-azabi-
cyclo[3.3.0]octane (IIIb)
REFERENCES AND NOTES
[1]
a) Broadbent, H.S.; Burnham, W.S.; Sheely, R.M.; Olsen, R.K.
Novel heterotricyclic systems: 2,6-dioxa- and 2-oxa-6-thia-10-aza-
2-Amino-2-(mercaptomethyl)propane-1,3-diol hydroch-
loride (4) (1.00 g, 5.78 mmol) was dissolved in water (3.00
tricyclo-[5.2.1.04,10]decanes; 2,6-dioxa-11-azatricyclo[5.3.1.0.4,11
]

290 Letters in Organic Chemistry, 2010, Vol. 7, No. 4
But et al.
undecane; and 9,13-dioxa-14-azatetracyclo[6.5.1.02,7.011,14]tetra-
deca-2,4,6-triene (1a, b). J. Heterocycl. Chem., 1976, 13, 337-348;
b) Habernel, G.G.; Ecsy, W. Synthesis of 4-aza-2-oxa-6-
thiabicyclo[3.3.0]octane. Heterocycles, 1977, 7(2), 1027-1032; c)
Enz, W; Cecchinato, M. L-Cysteinol. Helv. Chim. Acta, 1961,
44(3), 706-709.
a) Seebach, D.; Weber, T. ꢀ-Alkylation of a cysteine derivative
without racemization and without the use of a chiral auxiliary.
Tetrahedron Lett., 1983, 24(32), 3315-3318; b) Seebach, D.;
Weber, T. Hydroxyalkylierungen von cystein über das enolat von
(2R,5R)-2(tert-Butyl)-1-aza-3-oxa-7-thiabicyclo[3.3.0]octan-4-on
und unter Selbstreproduktion des Ciralitätszentrums. Helv. Chim.
Acta, 1984, 67(6), 1650-1661.
Commun., 1955, 20, 1426-1438; i) Winters, G.; Nathansohn, G.;
Testa, E. Synthesis of ꢂ-(p-nitrophenyl)cysteine and of ꢀ-amino-ꢂ-
mercapto-ꢂ-(p-nitrophenyl)propanol. Gazz. Chim. Ital., 1964,
94(12), 1419-1429.
a) Cookson, R.C.; Crabb, T.A. Geminal coupling constants in
methylene groups ꢀ to nitrogen and oxygen atoms in 5- and 6-
membered ring compounds. Tetrahedron, 1968, 24(5), 2385-2397;
b) Crabb, T.A.; Hall, M.J.; Williams, R.O. Proton magnetic
resonance studies of compounds with bridgehead nitrogen atoms—
XXVI: configurational studies with derivatives of perhydro-
oxazolo[3,4-c]oxazole and the conformational analysis of
perhydro-imidazo[1,5-c]thiazole. Tetrahedron, 1973, 29(21), 3389-
3398.
[7]
[2]
[3]
a) Gonzales, A.; Lavilla, R.; Piniella, J.F.; Alvarez-Larena, A.
Protected derivatives of (R)-cysteine and (R)-cysteinol.
Tetrahedron, 1995, 51(10), 3015-3024; b) Deroose, F.D.; De
Clercq, P.J. Novel enantioselective syntheses of (+)-Biotin J. Org.
Chem., 1995, 60(2), 321-330; c) Ino, A.; Hasegawa, Y.;
Murabayashi, A. Total synthesis of the antimycoplasma antibiotic
Micacocidin. Tetrahedron Lett., 1998, 39(21), 3509-3516; d) Yang,
H.; Sheng, X.C.; Harrington, E.M.; Ackerman, K.; Garcia, A.M.;
Lewis, M.D. Synthesis of sulfur-containing olefinic peptide
[8]
a) This fact is, however, not surprising but quite similar to that of
4,4-bis(hydroxymethyl)-1,3-oxazolidines, see Buur, A.; Bundgaard,
H. Prodrugs as drug delivery systems 66. Hydrolysis of various
oxazolidines and N-acylated oxazolidines–potential prodrug types
for ꢂ-aminoalcohols or carbonyl-containig drugs. Arch. Pharm.,
Sci. Ed., 1987, 15, 76-86; b) for a more detailed discussion on this
topic see Darabantu.
M
(Masked)serinols: molecules,
biomolecules, building-blocks, supramolecules. Part (I): syntheses
based on serinols’ reactivity with carbonyl electrophiles. Curr.
Org. Synth., 2010, 7(2), 120-152.
mimetic
Farnesyl
Transferase
inhibitors
using
the
NozakiꢁHiyamaꢁKishi reaction and cuprate SN2‘ displacements. J.
Org. Chem., 1999, 64(1), 242-251.
[9]
Darabantu, M; Breazu, D.; Vasilioni, V.; Brodszky, M.; Cotora, I.;
Pap, Z. Method of isolation of Nortriptyline hydrochloride. RO Pat.
91181, June 27, 1986; cf. Chem. Abstr., 1991, 114, 121788.
a) Frisch, M.J.; Trucks, G.W.; Schlegel, H.B.; Scuseria, G.E.;
Robb, M.A.; Cheeseman, J.R.; Zakrzewski, V.G.; Montgomery,
J.A., Jr.; Stratmann, R.E.; Burant, J.C.; Dapprich, S.; Millam, J.M.;
Daniels, A.D.; Kudin, K.N.; Strain, M.C.; Farkas, O.; Tomasi, J.;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.;
Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G.A.; Ayala,
P.Y.; Cui, Q.; Morokuma, K.; Rega, N.; Salvador, P.; Dannenberg,
J.J.; Malick, D.K.; Rabuck, A.D.; Raghavachari, K.; Foresman,
J.B.; Cioslowski, J.; Ortiz, J.V.; Baboul, A.G.; Stefanov, B.B.; Liu,
G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.;
Martin, R.L.; Fox, D.J.; Keith, T.; Al-Laham, M.A.; Peng, C.Y.;
Nanayakkara, A.; Challacombe, M.; Gill, P.M. W.; Johnson, B.;
Chen, W.; Wong, M.W.; Andres, J.L.; Gonzalez, C.; Head-Gordon,
M.; Replogle, E.S.; Pople, J.A. Gaussian 98, Revision A.11.3.;
Gaussian: Pittsburgh, PA, 2002; b) Klamt, A.; Schüürmann, G.
COSMO: a new approach to dielectric screening in solvents with
explicit expressions for the screening energy and its gradient. J.
Chem. Soc., Perkin Trans. 2, 1993, 5, 799-805.
a) Eliel, E.; Wilen, S. H. Stereochemistry of Organic Compounds;
John Wiley & Sons, Inc. New York, 1994; pp. 1017, 1199, 221-
239, 488-507; b) Moss, G. P. Basic terminology in stereochemistry
(IUPAC Recommendations). Pure Appl. Chem., 1996, 68(12),
2193-2221.
Monge, S.; Selambaron, J.; Carré, F.; Verducci, J.; Roque, J.P.;
Pavia, A. A. Anomeric effects in non-carbohydrate compounds:
conformational differences between the oxazolidine rings of a cis-
fused bicyclic system. Carbohydr. Res., 2000, 328(2), 127-133.
Star, A.; Fuchs, B. Mechanism of formation and stabilities of the
new dioxadiazadecalin systems. Ringꢁchain tautomerism. J. Org.
Chem., 1999, 64(4), 1166-1172.
[4]
a) Darabantu, M.; Plé, G.; Maiereanu, C.; Silaghi-Dumitrescu, I.;
Ramondenc, Y.; Mager, S. Synthesis and stereochemistry of some
1,3-oxazolidine systems based on TRIS (ꢀ,ꢀ,ꢀ-trimethylo-
laminomethane) and related aminopolyol-skeletons (II). 1-Aza-3,7-
dioxabicyclo[3.3.0]octanes. Tetrahedron, 2000, 56(23), 3799-3916;
b) Darabantu, M.; Maiereanu, C.; Silaghi-Dumitrescu, I.; Toupet,
L.; Condamine, E.; Ramondenc, Y.; Berghian, C.;Plé, G.; Plé, N.
3,7-Dioxa-1-azabicyclo[3.3.0]octanes substituted at C-5 position.
From local to global stereochemistry. Eur. J. Org. Chem., 2004, 12,
2644-2661; c) Berghian, C.; Lameiras, P.; Toupet, L.; Condamine,
E.; Plé, N.; Turck, A.; Maiereanu, C.; Darabantu, M. ꢀ-(3,7-Dioxa-
r-1-azabicyclo[3.3.0]oct-c-5-ylmethoxy)-diazines (I): Synthesis
and stereochemistry. Extension to s-triazine series. Tetrahedron,
2006, 62(31), 7319-7338.
[10]
[5]
[6]
Senkus, M. Some new derivatives of amino hydroxy compounds. J.
Am. Chem. Soc., 1945, 67, 1515-1519.
a) Nys, J.; Libeer, J. 4-Hydroxymethyl- and 4-acetoxymethyl-2-
methyl-2-thiazolines. U. S. Pat. 2, 954, 376, September 27, 1960;
cf. Chem. Abstr., 1961, 55, 36770; b) Bresson, C.R. 2-
Aminoalkanethiols. U. S. Pat. 3, 351, 664, November 7, 1967; cf.
Chem. Abstr., 1968, 68, 95331; for related strategies involving 2-
amino-1-(4-nitrophenyl)propane-1,3-diol’ stereiosomers in Chloro-
mycetin syntheses see: c) Ykuma, S; Nagawa, M. Alkamines. III.
Steric inversion of dl-erythro-1-p-nitrophenyl-2-aminopropane-1,3-
diol. 1. Yakugaku Zasshi, 1952, 72, 950-951; d) Alberti, C.G.;
Camerino, B; Vercellone, A. Chloroamphenicol. XI. Oxazolines
and acyl migrations in nitrophenylaminopropanediols. Gazz. Chim.
Ital., 1955, 85, 324-344; e) Ikuma, S. Alkamines. VI. Synthesis of
chloramphenicol. Yakugaku Zasshi, 1953, 73, 284-287; f) Lafuente,
L.S.; Pastor, O.S. p-Nitro-DL-trans-1-phenyl-1,3-dihydroxy-2-
aminopropane: nitration of oxazolinic structures. Anales real soc.
espan, fis. y quim. (Madrid), 1956, 52B, 647-650; g) Ikuma, S.
Alkamines. VI. Synthesis of chloramphenicol. Yakugaku Zasshi,
1953, 73, 284-7; h) Svoboda, M; Sicher, J.; Farkas, J.; Pankova, M.
Stereochemical studies. IV: reaction of 2-oxazolines with thioacetic
acid. New syntheses of vic-amino thiols. Coll. Czech. Chem.
[11]
[12]
[13]
[14]
a) Friebolin, H. Basic One- and Two-Dimensional NMR
Spectroscopy. VCH Verlagsgesellschaft, Weinheim, New York,
1991; pp. 75, 93, 263–291; b) Silverstein, R.M.; Webster, F.X.;
Kiemle, D. J. Identification Spectrométrique de Composés
Organiques, De Boeck & Larcier s.a., 2007; pp. 172, 275, 374.