A convenient preparation of taxoid right-half building blocks

Article abstract of DOI:10.1016/S0957-4166(98)00289-4

Wieland-Miescher ketone derived unsaturated diols 4 reacted with Pb(OAc)₄ to furnish tricyclic enolether intermediate 6 which upon ozonolysis gave access to useful synthetic intermediates such as bicyclic lactone 7, methyl furanoside 8 and triol 9 (taxoid right-half precursors), depending on the solvent used during the ozonolysis and the nature of the following synthetic operation.

Full text of DOI:10.1016/S0957-4166(98)00289-4

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 9 (1998) 2871–2878
A convenient preparation of taxoid right-half building blocks
Siméon Arseniyadis,a,∗ José Quilez del Moral,a Rosemeire Brondi Alves,a Pierre Potier a and
Loïc Toupet b
aInstitut de Chimie des Substances Naturelles, CNRS, F-91198 Gif-sur-Yvette, France
bURA 804 au CNRS, Université de Rennes I, F-35042 Rennes, France
Received 26 June 1998; accepted 20 July 1998
Abstract
Wieland–Miescher ketone derived unsaturated diols 4 reacted with Pb(OAc)4 to furnish tricyclic enolether
intermediate 6 which upon ozonolysis gave access to useful synthetic intermediates such as bicyclic lactone 7,
methyl furanoside 8 and triol 9 (taxoid right-half precursors), depending on the solvent used during the ozonolysis
and the nature of the following synthetic operation. © 1998 Elsevier Science Ltd. All rights reserved.
1. Introduction
Elaborated six-membered ring systems containing multiple stereogenic centers constitute the backbone
of several biologically important natural compounds. The need for constructing such derivatives for the
synthesis of A- and C-ring components of taxoids, provided a strong incentive for the development of
improved synthetic methods1 during the era of taxol synthesis and considerable efforts have been devoted
to meet this challenge.2
Through our efforts towards the total synthesis of taxoids, using the (S)-(+)-Hajos–Parrish ketone 1
as precursor for the entire taxoid framework,3 we investigated the lead tetraacetate mediated oxidative
cleavage of the unsaturated diols 3. This provided a convenient route for elaboration of the right
half (C2–C10) of the taxoid diterpene skeleton I and prompted us to use similar methodology for
the preparation of the taxoid C-ring subunit starting from its higher analog 4, itself obtained from
the (S)-(+)-Wieland–Miescher ketone 2. As we pointed out,4 two main differences were observed in
the oxidative fragmentations of the unsaturated diols 3 and 4. The one of interest is related to the
stability of the intermediate tricyclic enol ether (or enol acetal) 6 which unlike its lower homolog is
perfectly stable. The latter, obtained in one synthetic operation upon treatment of 4 with 1.2 equiv. of
Pb(OAc)4 in acetonitrile, can be easily isolated and stored without any special care, thus allowing for a
∗
Corresponding author. E-mail: Simeon.Arseniyadis@icsn.cnrs-gif.fr
0957-4166/98/$ - see front matter © 1998 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(98)00289-4

2872
S. Arseniyadis et al. / Tetrahedron: Asymmetry 9 (1998) 2871–2878
differential functional group interconversion. This clearly means that the cascade can be ‘interrupted’ on
the Wieland–Miescher series while this is not the case in the Hajos–Parrish series, where the full cascade
product 5 is obtained straightforwardly (Scheme 1). In a preliminary communication, dealing with the
Pb(OAc)4 mediated one-pot multi-stage transformations, we reported that if the sequence is interrupted
after the intramolecular bis-hetero Diels–Alder reaction, the method also serves, following ozonolytic
cleavage, for the preparation of methyl furanoside 8 or formylacetal-aldehyde 10 depending on the
solvent used (Scheme 2).5 Reported here is the synthesis and characterization of the bicyclic lactone
7, methyl furanoside 8 and triol 9, homochiral taxoid C-ring precursors, offering linking possibilities
at C-2 and C-10. The key to these transformations is the Pb(OAc)4 mediated oxidative cleavage of
the Wieland–Miescher ketone derived bicyclic diol 4, allowing the synthesis of tricyclic enol ether 6.
Ozonolytic cleavage of the latter leading to 10, sets the stage for an intramolecular Cannizzaro type
oxidoreduction6 which proceeds under very mild conditions allowing for the construction of the bicyclic
lactone 7.
Scheme 1. Reactions of unsaturated 1,2-diols with lead tetraacetate; taxoid C-ring precursors
Scheme 2. (a) 1.2 equiv. Pb(OAc)4, CH3CN, −10°C for 5 min, then rt, 3 h; (b) O3, CH2Cl2, −78°C then Me2S; (c)
K2CO3–MeOH–H2O, rt, 6 h; (d) LiAlH4, THF, reflux; (e) O3, MeOH, −78°C then Me2S, rt; (f) LiOH–MeOH–H2O, rt; (g)
SiO2, heptane:ethyl acetate (2:1), 12 h; (h) acetone, pTsOH, rt
2. Results and discussion
The preparation of the requisite tricyclic enol ether 6 was accomplished by lead tetraacetate mediated
oxidative cleavage of the unsaturated bicyclic diol 4. Treatment of the latter with Pb(OAc)4 (1.2 equiv.),
in dry acetonitrile (5 ml per mmol) at −10°C to rt, for 3 h under an inert atmosphere, afforded 6 in 91%
yield after silica gel flash chromatography (heptane:ether, 1:1).
This one-pot sequence leading to 6 (the ‘interrupted cascade’), also works well at room temperature
or at below zero temperatures, though longer periods of stirring are needed for low temperature runs

S. Arseniyadis et al. / Tetrahedron: Asymmetry 9 (1998) 2871–2878
2873
Fig. 1. Perspective drawing of the X-ray structure of 7 and spatial proximities (diagnostic NOEs) of 12 shown on the lowest
energy conformer as determined by molecular mechanics calculations
(−40°C for example), and tolerates various solvents. By appropriate selection of the next operation,
taxoid C-ring building blocks such as 7, 8, 9 and 11 could then be prepared. The procedure used
to prepare these C-ring precursors is outlined in Scheme 2. When ozonolysis of 6 was carried out
in methanol (10 ml per mmol), at −78°C, a diastereomeric mixture of methyl furanosides 8a and
8b (ca. 1:4 ratio) was obtained following reductive treatment with excess Me2S, in 78% combined
yield. On the other hand, when ozonolysis was performed in dichloromethane, formylacetal-aldehyde
10 was obtained. Reduction of the latter with LiAlH4 in refluxing THF afforded triol 9 which upon
room temperature treatment with dry acetone in the presence of a catalytic amount of p-toluenesulfonic
acid led to the corresponding isopropylidene derivative 11 in 68% overall yield (from 6). Upon mild
base treatment (K2CO3, MeOH–H2O, 6 h at room temperature), formylacetal-aldehyde 10 afforded
bicyclic lactone 7 as a crystalline solid in 67% yield, along with 12 (11%), via an intramolecular
Cannizzaro type oxidoreduction. The two bicyclic lactones, 7 and 12, could be separated by silica gel
column chromatography (eluent: heptane:ethyl acetate, 3:1). However, we observed that 12, originally
obtained as a minor product, was gradually transformed into 7 upon a prolonged stay in an SiO2 column.
Isopropylidene derivative 11 could also be obtained in comparable yields from bicyclic lactone 7 using
the same protocol as from 10. Complete acid promoted transesterification could also be achieved by
stirring pure bicyclic lactone 12 in heptane:ethyl acetate (2:1) in the presence of silica gel for a few hours
at room temperature, resulting in the exclusive formation of 7. On the other hand, upon subjection of
pure bicyclic lactone 7 to a basic treatment (10 equiv. of LiOH or KOH or NaOH, in MeOH:H2O (10:1),
10 ml/mmol), at room temperature, an equilibrium mixture of 7:12 (ca. 1:1.1) resulted. Interestingly,
treatment of 10 with LiCO3–MeOH–H2O at room temperature for up to 48 h produced solely lactol
aldehyde 13 as an unseparable epimeric mixture at the C-10 position, which was characterized as such.
Briefly, bicyclic lactone 7 was the final product of acid promoted transesterification, while the base
promoted transesterification led to an equilibrium mixture of 7 and 12 (nearly 1:1), regardless of the
starting point, which could be pure 7, pure 12, or any random mixture of 7 and 12. Single-crystal X-
ray diffraction analysis of bicyclic lactone 7 enabled unequivocal stereochemical assignment and further
confirmed the correctness of previous assignments by NMR techniques. The ORTEP representation of 7
along with the lowest energy conformer of 12 is shown in Fig. 1.
Repeated in the presence of a tenfold excess of formalin (37% HCH_O in H2O) or paraformaldehyde,
basic treatment of 10 did not furnish any crossed-Cannizzaro product; a mixture of the two bicyclic lac-
tones 7 and 12 was obtained instead, in an approximately 1:1 ratio. The site-selective disproportionation
of the Cannizzaro oxidoreduction is consistent with direct hydride shift on the lactol-aldehyde 13, leading
to the bicyclic lactone 7 which in turn in a basic medium suffers ring opening followed by partial translac-
tonization affording the bicyclic lactone 12. A mechanistic framework that accounts for these sequential

2874
S. Arseniyadis et al. / Tetrahedron: Asymmetry 9 (1998) 2871–2878
transformations is portrayed in Scheme 3. Saponification of the starting formyl acetal 10, obtained from
ozonolysis of 6 in methylene chloride, occurs in the presence of base to give the lactol-aldehyde 13.
Removal of the acidic proton then generates 14 which undergoes a subsequent intramolecular hydride
transfer leading to 15. The proposed sequence can be monitored by TLC (heptane:ethyl acetate, 2:1),
with intermediates 10, 13, 17, 7 and 12 possessing distinct Rf values. Soon after addition of the base in
MeOH–H2O, two new lower Rf spots appear on the TLC plate; the higher one is lactol-aldehyde 13, the
lower being bicyclic lactone 7. The spot corresponding to lactol-aldehyde 13 is gradually transformed to
7 upon 6–7 h stirring at room temperature. Further stirring then gives rise to a third spot, lower than the
bicyclic lactone 7, corresponding to the transesterified bicyclic lactone 12.
Scheme 3. Proposed mechanism for the base induced disproportionation of formyl-aldehyde 10
On the other hand, dihydroxy-carboxylic acid intermediate 17 can be observed by TLC as a spot
characteristic to acids, near the baseline, upon treatment with KOH–H2O or LiOH–H2O. The formation
of bicyclic lactones 7 and 12 could also be envisioned to occur via ring-opened intermediate 16 shown
below.
This second pathway, which involves disproportionation on the open form of the lactol-aldehyde
13, could be reasonably discarded as selectivity in disproportionation is complete. Neither the ring-
opened dihydroxy-acid corresponding to the Cannizzaro product resulting from HO− addition to the
C-2 carbonyl of 16, nor its hydroxy-lactone form were detected. Furthermore, it should be pointed
out that the nature of the counter ion (Li+, Na+, K+ and Cs+ were used) does not effect selectivity in
disproportionation.
3. Conclusion
In summary, we have described a facile preparation of bicyclic lactone 7, methylfuranoside 8 and triol
9 from the unsaturated bicyclic 1,2-diol 4, thus proving that a large number of skeletal and functional
changes can be obtained in a few synthetic operations. Tricyclic enol ether 6 proved to be a useful
intermediate for preparing the above cited taxoid C-ring building blocks. The observed site-selective
oxidoreduction is noteworthy, since it appears to have little, if any, precedence in the literature, insofar as
the extremely mild conditions are concerned.
Obtained in three (for 7), two (for 8) and three (for 9) steps, respectively, from bicyclic diol 4, they can
serve as either C-10 nucleophiles or C-2 electrophiles (and vice versa) in an ‘A+C’ approach towards the
total synthesis of taxoid analogs.7

S. Arseniyadis et al. / Tetrahedron: Asymmetry 9 (1998) 2871–2878
2875
4. Experimental
Experimental protocols such as the drying and purification of reaction solvents, instrumentation and
other such details are identical with those previously described.8 Key intermediate 6 was prepared
according to our earlier work.8
4.1. Ozonolysis of 6 in CH2Cl2
Ozone was passed into a stirred solution of 6 (805 mg, 3.19 mmol) in 40 ml of methylene chloride
and 1 ml of pyridine at −78°C until a blue color persisted, the reaction stirred for 5 additional minutes
at this temperature after which 6.4 mmol of Me2S were added dropwise. It is worth mentioning that
although it caused no harm, reductive work up of the ozonides (either Ph3P or Me2S) is unnecessary;
TLC control before Me2S addition indicated total destruction of the ozonides before reductive treatment.
The mixture was allowed to reach room temperature within approximately 30 min, concentrated under
reduced pressure, and filtered through silica gel with heptane:ethyl acetate (4:1) to give crude formyl-
aldehyde 10 which, although unstable, can be isolated through rapid SiO2 column chromatography
(heptane:ether, 9:1) in 66% yield, and characterized. 10: IR (film): 2975, 2948, 2875, 1731, 1467, 1390,
1366, 1255, 1191, 1174, 1161, 1130, 1104, 1073, 1057, 1024, 1010, 966, 938, 881 cm−1. 1H-NMR (250
MHz): 1.07 (3H, s, Me-19), 1.21 (9H, s, tBu), 1.10–1.90 (7H, m), 2.75 (1H, dd, J=6.2, 14.1, H-9), 3.28
(1H, dd, J=3.6, 10.9, H-7), 6.55 (1H, t, J=5.9, H-10), 8.14 (1H, s, H-20), 9.82 (1H, s, H-2). 13C-NMR
(62.5 MHz): 13.0 (Me-19), 19.0 (C-4), 24.9 (C-5), 29.2 (tBu), 29.0 (C-6), 43.0 (C-9), 49.0 (Cq-8), 72.1
(C-7), 73.2 (Cq-tBu), 93.7 (Cq-3), 97.4 (C-10), 160.4 (C-20), 202.2 (C-2). Formyl-aldehyde 10 needs no
purification and can be carried out as such thus increasing the yields.
4.2. Cannizzaro type base-catalyzed oxidoreduction of 10; preparation of the bicyclic lactones 7 and 12
Formylacetal-aldehyde 10 (224 mg, 0.78 mmol) was dissolved in a mixture of methanol (5 ml) and
water (0.5 ml), and 200 mg (1.45 mmol, 1.17 equiv.) of potassium carbonate was added. The mixture
was stirred at room temperature for 6 h, diluted with water, and extracted with ether. Note that in large
scale preparations methanol should be removed first under reduced pressure. Following usual work up
the residue was purified by flash chromatography on silica gel to afford 134 mg (0.522 mmol, 67%) of
7 and 22 mg (0.086 mmol, 11%) of 12. Elution with heptane:ethyl acetate (3:1) easily separated the two
bicyclic lactones. (3S)-(7S)-(8R)-7: Mp 104–105°C (ether–heptane): [α]D +34 (c 1.4). IR (CHCl3): 3409,
3020, 2978, 2944, 2938, 1765, 1216 cm−1. 1H-NMR (300 MHz): 1.15 (3H, s, Me-19), 1.23 (9H, s, tBu),
1.45–1.95 (6H, m), 2.42 (1H, A of AB, JAB=17.2), 2.75 (1H, B of AB, JAB=17.2, H-9), 3.50 (1H, dd,
J=2.6, 6.8, H-7), 3.71 (2H, s, H-2). 13C-NMR (62.5 MHz): 15.6 (Me-19), 18.5 (CH2), 28.2 (CH2), 28.8
(tBu), 29.1 (CH2), 42.0 (C-9), 45.9 (Cq-8), 66.2 (C-2), 72.3 (C-7), 74.1 (Cq-tBu), 89.3 (Cq-3), 176.7 (C-
10). EIMS: 256 (M+, 1), 200 (60), 182 (96), 169 (100), 138 (66), 57 (72). HREIMS: calcd for C14H24O4
m/z 256.1674, found 256.1660. Anal.: calcd for C14H24O4 C 65.60, H 9.44, found: C 65.44, H 9.46.
(3S)-(7S)-(8R)-12: Mp 146–148°C (heptane). [α]D +18 (c 0.51). IR (CHCl3): 3402, 2979, 1705, 1466,
1
1389, 1359, 1246, 1192, 1067, 1031, 900, 864 cm−1. H-NMR (400 MHz): 1.05 (3H, s, Me-19), 1.17
(9H, s, tBu), 1.33 (1H, m, H-4αeq), 1.43 (1H, m, H-6βax), 1.55 (1H, m, H-4βax), 1.63 (1H, m, H-5),
1.74 (1H, m, H-6αeq), 1.80 (1H, m, H-5), AB-system at 2.49 (1H, d, J=17.6, H-9βeq), and 2.72 (1H, d,
J=17.6, H9αax), 3.69 (1H, dd, J=4.3, 11.4, H-7), AB-system at 4.00 (1H, A of AB, JAB=12.2, H-2αeq),
4.30 (1H, B of AB JAB=12.2, H-2βax). Diagnostic NOEs: {Me-19}: H-9βeq, H-6βax, H-2βax, H-4βax;
{H-9αax}: H-7; {H-2βax}: H-2αeq (NOE gem), Me-19, H-4βax. 13C-NMR (75 MHz): 16.1 (Me-19),

2876
S. Arseniyadis et al. / Tetrahedron: Asymmetry 9 (1998) 2871–2878
18.8 (C-5), 28.4 (C-4), 28.9 (tBu), 29.6 (C-6), 39.1 (C-9), 40.4 (Cq-8), 70.9 (C-7), 71.6 (Cq-3), 73.3
(Cq-tBu), 76.2 (C-2), 171.7 (C-10). EIMS: 256 (M+, 2), 200 (36), 182 (100), 57 (46). CIMS: 257 (M+1,
100). HRCIMS: calcd for C14H25O4 m/z 257.1753, found 257.1755.
4.3. X-Ray structure determination of 7
C14H24O4: Mr=256.35, triclinic, P-1, a=7.876(4), b=9.020(4), c=10.705(3) Å, α=89.95(2),
β=102.29(3), γ=106.77(3)°, V=710.0(5) Å−3, Z=2, Dx=1.199 Mg m−3, λ(MoKα)=0.70926 Å,
µ=0.81 cm−1, F(000)=280, T=294 K. The sample (0.45×0.45×0.55 mm) was studied on an automatic
diffractometer CAD4 Enraf–Nonius with graphite monochromatized MoKα radiation. The cell
parameters were obtained by fitting a set of 25 high-theta reflections. The data collection (2θmax=50°,
scan ω/2θ=1, tmax=60 s, range HKL: H 0,9 K −10,10 L −12,12, intensity controls without appreciable
decay (0.2%) gave 2504 reflections of which 1693 were independent with I>3.0σ(I). After Lorenz and
polarization corrections the structure was solved with Direct Methods which revealed the non-hydrogen
atoms of the molecule. After isotropic (R=0.12), then anisotropic refinement (R=0.10), the hydrogen
atoms were found with a Fourier difference (between 0.50 and 0.27 eÅ−3). The whole structure was
refined by the full-matrix least-square techniques (use of F magnitude; x, y, z, βij for C and O atoms
and x, y, z for H atoms; 236 variables and 1693 observations; w=1/σ(Fo)2=[σ2(I)+(0.04Fo2)2]−1/2) with
the resulting R=0.035, Rw=0.033 and Sw=0.85 (residual ∆ρ=0.14 eÅ−3). Atomic scattering factors were
taken from International Tables for X-Ray Crystallography (1974). All the calculations were performed
on a Digital MicroVAX 3100 computer with the MOLEN package (Enraf–Nonius, 1990).
4.4. Attempted crossed-Cannizzaro reactions in the presence of formalin and paraformaldehyde
To a stirred solution of formylacetal-aldehyde 10 (1.4 mmol) in 20 ml of MeOH, and 2 ml of H2O,
were added 10 equiv. of K2CO3 and 10 equiv. of paraformaldehyde, and stirring at room temperature
was continued for 46 h (TLC monitoring). Work up as above gave an 80% isolated yield of 7 and 12 in
nearly a 1:1 ratio. Repeated in the presence of formalin (37% in H2O) the same experiment furnished
the intramolecular Cannizzaro products 7 and 10 in the same yield and ratio while, again, no crossed-
Cannizzaro products were detected.
4.5. Translactonization experiments
Pure bicyclic lactone 12 (0.39 mmol, 100 mg) was stirred at 25°C, for 12 h (until lactone 12 had
been consumed) in the presence of SiO2 (ca. 100 mg) in heptane:ethyl acetate (2:1, 10 ml). The reaction
mixture was filtered and concentrated under reduced pressure to give quantitatively compound 7. In a
parallel experiment, 1.4 mmol of pure bicyclic lactone 7, in 20 ml of MeOH and 2 ml of H2O, was
treated with 10 equiv. of KOH (14 mmol) until the equilibrium ratio, 1.1:1, 12:7 was established upon
stirring for 4.5 h at rt. The ratio remains unchanged upon prolonged room temperature stirring (2 days).
The same equilibrium mixture was enriched upon subjection of either pure 7 or any random mixture of 7
and 12.
4.6. Ozonolysis of 6 in MeOH
Ozone was passed into a stirred solution of 102 mg (0.40 mmol) of 6 in methanol (8 ml) at −78°C
until a purple–blue color persisted. Me2S (ca. 3 ml) was added. The reaction mixture was allowed to

S. Arseniyadis et al. / Tetrahedron: Asymmetry 9 (1998) 2871–2878
2877
reach room temperature within 1 h, concentrated under reduced pressure, and the residue was flash
chromatographed on silica gel. Elution with heptane:ether (10:1) afforded 17 mg (16%) of 8a (H-10α)
and 67 mg (62%) of 8b (H-10β) in 78% combined yield and ca. 1:4 ratio. (3S)-(7S-)(8R)-(10R)-8a:
Mp 59–61°C (heptane–ether). [α]D −30 (c 0.8). IR (CHCl3): 2973, 2945, 1728, 1466, 1388, 1365,
1190, 1131, 1099, 1051, 1016, 973 cm−1. 1H-NMR (250 MHz): 0.99 (3H, s, Me-19), 1.20 (9H, s, tBu),
1.30–1.85 (7H, m), 2.63 (1H, dd, J=6.0, 14.0, H-90), 3.26 (1H, dd, J=3.5, 11.0, H-7), 3.50 (3H, s, OCH3),
5.23 (1H, t, J=5.8, H-10), 9.88 (1H, s, H-2). 13C-NMR (75 MHz): 13.0 (Me-19), 19.2 (C-4), 24.8 (C-5),
29.1 (tBu), 29.4 (C-6), 44.1 (C-9), 49.1 (Cq-8), 56.1 (OCH3), 72.3 (C-7), 73.3 (Cq-tBu), 92.4 (C-3),
105.2 (C-10), 204.8 (C-2). CIMS: m/z 271 [(M+H), 7], 239 [(M+H)−MeOH, 100], 183 (82), 165 (32),
57 (64). Anal.: calcd for C15H26O4 C 66.64, H 9.69, found: C 66.34, H 9.76. (3S)-(7S)-(8R)-(10S)-8b:
Mp 58–60°C (heptane–ether). [α]D +115 (c 1.0). IR (CHCl3): 3019, 2977, 2943, 2915, 2872, 1729, 1466,
1391, 1365, 1216, 1192, 1127, 1099, 1069, 1024, 970 cm−1. 1H-NMR (400 MHz): 1.03 (3H, s, Me-19),
1.20 (9H, s, tBu), 1.50 (1H, m, H-6β), 1.61 (1H, m, H-4), 1.70–1.82 (3H, m, H-5, 6α, 4), 1.77 (1H, dd,
J=13.7, 6.6, H-9β), 2.26 (1H, dd, J=2.2, 13.7, H-9α), 3.43 (3H, s, OCH3), 3.87 (1H, dd, J=2.5, 7.9, H-7),
5.28 (1H, dd, J=2.2, 6.6, H-10), 9.61 (1H, s, H-2). Diagnostic NOEs: {Me-19}: H-6β, H-9β, H-10; {H-
10}, H-9β, Me-19. 13C-NMR (75 MHz): 15.8 (Me-19), 19.0 (C-4), 26.8 (C-5), 29.1 (C-6), 29.0 (tBu),
43.2 (C-9), 46.6 (Cq-8), 55.7 (OCH3), 71.1 (C-7), 73.6 (Cq-tBu), 92.0 (C-3), 104.8 (C-10), 202.5 (C-2).
CIMS: m/z 271 [(M+H), 1], 241 (15), 239 [(M+H)−MeOH, 27], 183 (100), 165 (17), 57 (99). Anal.:
calcd for C15H26O4 C 66.64, H 9.69, found: C 66.51, H 9.64.
4.7. Reduction of 10 and selective acetonide formation
To a stirred solution of 10 (284 mg, 1.0 mmol) in THF (20 ml), LiAlH4 (114 mg, 3 mmol) was added
slowly at room temperature. The mixture was stirred under reflux for 1 h, cooled to ambient temperature,
quenched with 0.2 ml of water, 0.2 ml of 15% NaOH, 0.6 ml of water, and stirring was continued at room
temperature for 1 h. Filtration and concentration under reduced pressure afforded crude triol 9, which
was purified through silica gel column chromatography, using ethyl acetate:methanol (9:1) as eluent, to
yield 68% (two steps; ozonolysis+reduction) of (3S)-(7S)-(8R)-9: mp 96–98°C (CH2Cl2). [α]D +35 (c
1.0). IR: 3252, 2969, 2934, 2872, 1451, 1423, 1363, 1226, 1191, 1086, 1018, 974, 908, 878, 754 cm−1
.
1H-NMR (300 MHz): 0.90 (3H, s, CH3-19), 1.17 (9H, s, tBu), 1.40–2.12 (8H, m), 3.50–3.75 (5H, m).
13C-NMR (75 MHz): 17.2 (Me-19), 18.7 (CH2), 29.1 (CH2), 29.2 (tBu), 30.2 (CH2), 37.4 (CH2), 44.3
(C-9), 58.6 (C-10), 65.4 (C-2), 72.0 (C-7), 73.3 (Cq-tBu), 76.4 (C-3). CIMS: m/z 261 [(M+H), 100], 243
[(M+H)−H2O, 14], 205 [(M+H)−Me2C_C, 16]. HRCIMS: calcd for C14H29O4 m/z 261.2065, found
261.2069. Anal.: calcd for C14H28O4 C 64.58, H 10.84, found: C 64.67, H 10.76. Triol 9 thus obtained,
was dissolved in dry acetone (5 ml) and a catalytic amount of p-toluenesulfonic acid (ca. 20 mg) was
added. The mixture was stirred under argon at room temperature for 5–10 min, diluted with ethyl acetate
and filtered through aluminum oxide to give acetonide 11 in quantitative yield. (3S)-(7S)-(8R)-11: [α]D
+1 (c 0.9). IR: 3423, 2973, 2941, 2872, 1465, 1368, 1252, 1212, 1197, 1111, 1065, 1049, 1017, 1007,
977, 918, 874 cm−1. 1H-NMR (300 MHz): 1.05 (3H, s), 1.17 (9H, s), 1.35 (3H, s), 1.41 (3H, s), 1.20–1.82
(8H, m), 2.22 (1H, m), 3.33 (1H, d, J=3.2, OH), 3.60 (1H, m), 3.79 (1H, m), 3.87 (1H, d, J=9.7), 4.10
(1H, dd, J=1.2, 9.7). 13C-NMR (75 MHz): 17.2 (Me), 19.0 (CH2), 26.0 (Me), 27.0 (CH2), 27.9 (Me),
28.7 (tBu), 32.4 (CH2), 39.1 (CH2), 42.8 (Cq-8), 58.9 (C-10), 69.8 (C-2), 73.6 (Cq-tBu), 76.2 (C-7),
85.7 (C-3), 107.1 (Cq-acetonide). CIMS: m/z 301 [(M+H), 100], 245 [(M+H)−Me2C_C, 40], 169 (53).
HRCIMS: calcd for C17H33O4 m/z 301.2378, found 301.2340. Triol 9 and hence acetonide 11 were also
obtained from 7 using the same procedure as above.

2878
S. Arseniyadis et al. / Tetrahedron: Asymmetry 9 (1998) 2871–2878
4.8. Intermediate hemiacetal-aldehyde 13
To a solution of 10 (450 mg, 1.58 mmol) in MeOH (20 ml) and H2O (2 ml) was added LiCO3 (1.17
g, 15.8 mmol). The resulting mixture was stirred at room temperature for 24 h at which point no starting
material remained. The reaction mixture was then diluted with ether and worked up as usual. Silica gel
column chromatography (eluent: ethyl acetate:heptane, 1:4) afforded 378 mg (94%) of an unseparable
6:1 epimeric mixture of 13. Major isomer: IR: 3427, 2972, 1727, 1467, 1391, 1361, 1190, 1095, 1060,
983 cm−1. 1H-NMR (300 MHz): 1.06 (3H, s, Me-19), 1.20 (9H, tBu), 1.33 (1H, dd, J=6.0, 14.2, H-9),
1.35 (2.01, 6H, m), 2.59 (1H, dd, J=6.1, 14.2, H-9), 3.26 (1H, dd, J=3.8, 11.0, H-7), 4.07 (1H, br s,
OH), 5.59 (1H, t, J=6.0), 9.77 (1H, s, H-2). 13C-NMR (62.5 MHz): 12.8 (Me-19), 19.5 (C-4), 24.8 (C-5),
29.2 (tBu), 29.6 (C-6), 45.1 (C-9), 46.6 (Cq-8), 71.6 (C-7), 73.5 (Cq-tBu), 91.5 (C-3), 98.4 (C-10), 206.6
(C-2). CIMS: m/z 257 [(M+H), 7), 239 (79), 183 (100), 165 (27).
Acknowledgements
The authors wish to thank Ministerio de Educacion y Cultura (Spain) and CNPq (Brazil) for fellow-
ships to Jose Quilez del Moral, and Rosemeire Brondi-Alves, respectively.
References
1. For synthetic studies on taxoid C-rings see: Brown, P. A.; Jenkins, P. R.; Fawcett, J.; Russel, D. R. J. Chem. Soc., Chem.
Commun. 1984, 254–255; Cairns, P. M.; Howes, C.; Jenkins, P. R. J. Chem. Soc., Perkin Trans. 1 1990, 627–632; Boa, A.
N.; Clark, J.; Jenkins, P. R.; Lawrence, N. J. J. Chem. Soc., Chem. Commun. 1993, 151–152; Magee, T. V.; Bornmann, W.
G.; Isaacs, R. C. A.; Danishefsky, S. J. J. Org. Chem. 1992, 57, 3274–3276.
2. Review articles: Boa, A. N.; Jenkins, P. R.; Lawrence, N. J. Contemporary Organic Synthesis 1994, 1, 47–75; Nicolaou,
K. C.; Dai, W. M.; Guy, R. K. Angew. Chem. Int. Ed. Engl. 1994, 33, 15–44.
3. Arseniyadis, S.; Yashunsky, D. V.; Pereira de Freitas, R.; Munoz Dorado, M.; Potier, P.; Toupet, L. Tetrahedron 1996, 52,
12443–12458.
4. Arseniyadis, S.; Toupet, L.; Yashunsky, D. V.; Wang, Q.; Potier, P. Tetrahedron Lett. 1995, 36, 8783–8786.
5. Arseniyadis, S.; Yashunsky, D. V.; Brondi Alves, R.; Wang, Q.; Toromanoff, E.; Toupet, L.; Potier, P. Tetrahedron Lett.
1994, 35, 99–102.
6. Swain, C. G.; Powell, A. L.; Sheppard, W. A.; Morgan, C. H. J. Am. Chem. Soc. 1979, 101, 3576–3583.
7. Arseniyadis, S.; Martin Hernando, J. I.; Quilez del Moral, J.; Yashunsky, D. V.; Potier, P. Tetrahedron Lett. 1998, 39,
3489–3492.
8. Arseniyadis, S.; Brondi Alves, R.; Quilez del Moral, J.; Yashunsky, D. V.; Potier, P. Tetrahedron 1998, 54, 5949–5958.