Double stereodifferentiating aldol reactions based on chiral ketones derived from lactic acid: Synthesis of C1-C6 fragment of erythronolides

Article abstract of DOI:10.1055/s-2004-831332

Highly stereoselective titanium-mediated aldol additions of ethyl ketones derived from lactic acid to α-methyl-β-OTBDPS chiral aldehydes are documented. One of these double stereodifferentiating processes represents the key step of a straightforward and e

Full text of DOI:10.1055/s-2004-831332

LETTER
2127
Double Stereodifferentiating Aldol Reactions Based on Chiral Ketones
Derived from Lactic Acid: Synthesis of C1–C6 Fragment of Erythronolides
D
ouble St
o
ereodifferent
a
iatingAldo
n
l
Reactions G. Solsona, Joaquim Nebot, Pedro Romea,* Fèlix Urpí*
Departament de Química Orgànica, Universitat de Barcelona, Martí i Franqués 1-11, 08028 Barcelona, Catalonia, Spain
Fax +34(93)3397878; E-mail: felix.urpi@ub.edu; E-mail: pedro.romea@ub.edu
Received 4 June 2004
Dedicated to the memory of recently deceased Professor Satoru Masamune.
Given that the above-mentioned titanium-mediated aldol
Abstract: Highly stereoselective titanium-mediated aldol additions
of ethyl ketones derived from lactic acid to a-methyl-b-OTBDPS
chiral aldehydes are documented. One of these double stereodiffer-
entiating processes represents the key step of a straightforward and
efficient synthetic approach to the C1–C6 fragment of erythrono-
lides.
methodologies3 only afford syn relationships, four stereo-
chemistries represented in Scheme 1 were expected to
predominate.
Chiral ketones, 1–3, and aldehydes, 4 and ent-4, were pre-
pared in enantiomerically pure form by well-known pro-
cedures.7,8 Aldol reactions were then carried out
according to the protocols previously reported.3 The re-
sults summarized in Scheme 2 show that most reactions
are highly stereoselective irrespective of the configuration
of the aldehyde (compare eq 1 and 2, 5 and 6, and 7 and 8
in Scheme 2), which confirms the high stereocontrol ex-
erted by ketones 1–3. It is worth mentioning that virtually
a single isomer is obtained in many cases (see eq 2, 5, 6,
and 7 in Scheme 2).9,10 Remarkably, an outstanding
stereocontrol is achieved both in matched and mis-
matched pairs based on ketone 2 when the process is car-
ried out in the presence of an additional equivalent of
TiCl4 (see eq 5 and 6 in Scheme 2). As expected, the less
selective process involves TiCl4-mediated aldol reaction
of benzyloxy ketone 2 and aldehyde 4, which gives the
mismatched Felkin adduct 7 in a poor diastereomeric ratio
(dr 70:30).4,11 With the exception of this case, appropriate
choice of the protecting group (ketones 1–3) and the eno-
lization procedure permits all the stereochemical relation-
ships described in Scheme 1 to be obtained. Therefore,
these double stereodifferentiating aldol reactions give ac-
cess to a wide array of enantiopure intermediates useful
for stereoselective syntheses.
Key words: aldol reactions, chiral ketones, erythronolides, stereo-
selective synthesis, titanium
Pioneering studies by Masamune and Heathcock revealed
the synthetic potentiality of a-hydroxy ketones for the ste-
reoselective construction of carbon–carbon bonds through
aldol type reactions.1 They paved the way for the develop-
ment of a plethora of asymmetric methodologies, and,
particularly, inspired our endeavors to devise new stereo-
selective titanium-mediated aldol transformations based
on chiral ketones derived from lactic acid.2,3 Accumulated
evidence so far points out that there are three identifiable
stereochemical determinants that influence the reaction
diastereoselectivity: (i) the hydroxyl protecting group, (ii)
the titanium Lewis acid used in the enolization step, and
(iii) the addition of a supplementary Lewis acid to the re-
action mixture.3 Furthermore, conventional wisdom states
that chirality on the aldehyde must also play an important
role.4–6 Therefore, we decided to evaluate double stereo-
differentiating titanium-mediated aldol reaction between
ketones 1–3 and chiral aldehydes 4 and ent-4 (see
Scheme 1), aiming to discover simple and highly stereo-
selective processes useful for the synthesis of polypro-
pionate-like natural products.
O
OH
5
O
OH
5
2
4
6
2
4
6
OTBDPS
OTBDPS
PGO
PGO
O
O
O
2,4-syn-4,5-syn-5,6-syn
2,4-syn-4,5-syn-5,6-anti
H
OTBDPS
H
OTBDPS
PGO
O
OH
5
O
OH
5
4
ent-4
1 PG: TBDMS
2 PG: Bn
2
4
6
2
4
6
OTBDPS
OTBDPS
PG: PMB
3
PGO
PGO
2,4-anti-4,5-syn-5,6-syn
2,4-anti-4,5-syn-5,6-anti
Scheme 1
SYNLETT 2004, No. 12, pp 2127–2130
0
1
.1
0
.2
0
0
4
Advanced online publication: 31.08.2004
DOI: 10.1055/s-2004-831332; Art ID: D14404ST
© Georg Thieme Verlag Stuttgart · New York

2128
J. G. Solsona et al.
LETTER
O
O
O
O
O
O
O
O
OH
OH
OH
OH
OH
OH
OH
OH
c
d
c
d
e
71% dr 90:10
93% dr > 97:3
81% dr 70:30
95% dr 90:10
77% dr 99:1
79% dr 98:2
78% dr 99:1
75% dr 94:6
(eq 1)
OTBDPS
TiCln
O
O
TBDMSO
TBDMSO
BnO
5
a
TBDMSO
TBDMSO
(eq 2)
(eq 3)
(eq 4)
OTBDPS
1
6
Cln
Ti
OTBDPS
O
O
7
a
a
b
BnO
BnO
OTBDPS
2
BnO
8
(eq 5)
(eq 6)
(eq 7)
(eq 8)
Cln
Ti
OTBDPS
O
O
BnO
9
BnO
BnO
f
OTBDPS
2
BnO
10
c
(i-PrO)Cln
OTBDPS
Ti
O
O
PMBO
11
PMBO
PMBO
d
OTBDPS
3
PMBO
12
Scheme 212 Reagents and conditions: (a) TiCl4 (1.1 equiv), i-Pr2NEt (1.1 equiv), CH2Cl2, –78 °C, 1.5 h; (b) (i-PrO)TiCl3 (1.1 equiv), i-Pr2NEt
(1.1 equiv), CH2Cl2, –78 °C, 1.5 h; (c) 4 (1.5 equiv), –78 °C, 45 min; (d) ent-4 (1.5 equiv), –78 °C, 45 min; (e) TiCl4 (1 equiv), 4 (1.5 equiv),
–78 °C, 45 min; (f) TiCl4 (1 equiv), ent-4 (1.5 equiv), –78 °C, 45 min.
The utility of these double asymmetric processes has been fragment a crucial role in the overall strategy. Thus, it is
already demonstrated in the construction of the C18–C27 not surprising that methyl ketones such as that represented
fragment of superstolide A through the stereoselective in Scheme 3 have been considered as surrogates of ad-
reaction of ketone 3 and aldehyde 4 (see eq 7 in vanced C1–C6 intermediates and have concentrated the
Scheme 2).13 Looking for a more challenging case, we no- synthetic efforts of many groups.16
ticed that the 4,5-syn-5,6-syn relationship (see Scheme 1)
present in aldol 10 (see eq 6 in Scheme 2) would be a key
function in accessing an advanced intermediate in the
synthesis of erythromicins A and B.
The well known antibiotic macrolides erythromicins A ketone 13 (see Scheme 3) reported by Stork16a and
and B have attracted much attention because of their im- Yonemitsu16f attracted our attention because its synthesis
portant biological activity and their complex structure, might rely on the highly stereoselective titanium-mediat-
which contains most of the stereochemical motifs present ed aldol reaction represented in eq 6 of Scheme 2.
As already mentioned, a close inspection of the stereo-
chemical array embodied in such systems suggested that
three of their stereocenters might be easily installed
through a double asymmetric aldol process. Particularly,
in the polypropionate-like natural products. Therefore, to-
tal syntheses of both macrolides and their corresponding
aglycones, namely erythronolides A and B, have become
one of the cornerstones in organic synthesis for the last
In fact, our retrosynthetic analysis anticipated that methyl
ketone 13 would be available from diol 14 after a protec-
tion-deprotection sequence and final oxidation. Eventual-
ly, stereoselective reduction of aldol 10 would render the
decades and have stimulated the development of new con-
desired diol (see Scheme 3).
cepts and reactions for acyclic stereocontrol.14,15
Thus, such a synthetic sequence should permit us to obtain
13 in a highly economic and straightforward manner
Many of the retrosynthetic analyses applied to the seco-
acids of erythronolides A and B choose the C6–C7 bond
taking advantage of the substrate-controlled aldol process
as strategic disconnection, which confers to the C1–C6
that was previously developed.
Synlett 2004, No. 12, 2127–2130 © Thieme Stuttgart · New York

LETTER
Double Stereodifferentiating Aldol Reactions
2129
O
OH OH
O
OH OH
O
7
R
OH
OH
6
OH
13
R
1
OH
OH
O
O
OH
O
O
OH OH
O
1
OH OH
O
X
R = OH Erythronolide A
6
R = H
Erythronolide B
7
OTBDPS
OR
13
R
O
O
13
C1–C6 Fragment
O
OH
OH OH
OTBDPS
OTBDPS
BnO
BnO
10
14
Scheme 3
According to this approach, aldol 10 was routinely pre- Subsequently, removal of the benzyl group21 and Dess–
pared in 80% yield and 98:2 diastereomeric ratio on 5 Martin periodinane oxidation of the resulting alcohol af-
mmol scale using 1.2 equivalents of aldehyde ent-4. Sur- forded the desired ketone 1322 in five steps and 51% over-
prisingly, subsequent stereoselective syn reduction all yield from ketone 2.
proved to be more elusive. Initial attempts based on
In summary, highly stereoselective titanium-mediated al-
Narasaka–Prasad and zinc borohydride procedures17,18 af-
dol reactions based on chiral ketones derived from lactic
forded diol 14 in 95% yield but in a poor diastereomeric
ratio (dr 86:14 and 50:50, respectively). Fortunately, a
highly stereoselective reduction (82%, dr 96:4) was final-
acid, 1–3, and a-methyl-b-OTBDPS chiral aldehydes, 4
and ent-4, have been documented. Making the most of
these double stereodifferentiating processes, a new ap-
proach to the construction of a fully protected C1–C6
fragment of erythronolides has been disclosed. The
stereoselective sequence proceeds over five steps in 51%
overall yield, which proves the synthetic potentiality of
ly achieved with diisobutylaluminium hydride at –78 °C
(see Scheme 4).19 Protection of the diastereomeric mix-
ture furnished pure isopropylidene acetal 15 in 90%
yield.20
the aforementioned methodology even in the case of a re-
luctant 4,5-syn-5,6-syn relationship.
OH OH
a
10
OTBDPS
b
BnO
Acknowledgment
14
Financial support from the Ministerio de Ciencia y Tecnología and
Fondos FEDER (Grant BQU2002-01514), and the Generalitat de
Catalunya (2001SGR00051), as well as doctorate studentships
(Universitat de Barcelona) to J.G.S. and J.N. are acknowledged.
O
O
O
O
c
OTBDPS
OTBDPS
HO
BnO
16
15
References
d
O
O
(1) (a) Heathcock, C. H.; Pirrung, M. C.; Buse, C. T.; Hagen, J.
P.; Young, S. D.; Sohn, J. E. J. Am. Chem. Soc. 1979, 101,
7077. (b) Masamune, S.; Ali, S. A.; Snitman, D. L.; Garvey,
D. S. Angew. Chem., Int. Ed. Engl. 1980, 19, 557.
(c) Masamune, S.; Choy, W.; Kerdersky, F. A. J.; Imperiali,
B. J. Am. Chem. Soc. 1981, 103, 1566. (d) Van Draanen, N.
A.; Arseniyadis, S.; Crimmins, M. T.; Heathcock, C. H.
J. Org. Chem. 1991, 56, 2499.
OTBDPS
O
13
[α]D = +23.1 (c 1.8, CHCl3)
[α]D = +20.6 (c 1.2, CHCl3)16a
[α]D = +19.3 (c 1.45, CHCl3)16f
Scheme 4 Reagents and conditions: (a) DIBALH, THF, –78 °C,
82%; (b) cat. PPTS, (MeO)2CMe2–CH2Cl2 1:1, r.t., 90%; (c) H2, 10%
Pd/C, EtOAc, r.t., 91%; (d) DMP, CH2Cl2, 0 °C, 97%.
Synlett 2004, No. 12, 2127–2130 © Thieme Stuttgart · New York

2130
J. G. Solsona et al.
LETTER
(2) Paterson nicely established the synthetic utility of lactate-
derived ketones in stereoselective syn and anti boron-
mediated aldol reactions. See (a) Paterson, I.; Wallace, D.
J.; Velázquez, S. M. Tetrahedron Lett. 1994, 35, 9083.
(b) Paterson, I.; Wallace, D. J. Tetrahedron Lett. 1994, 35,
9087. (c) Paterson, I.; Wallace, D. J.; Cowden, C. J.
Synthesis 1998, 639.
(3) (a) Figueras, S.; Martín, R.; Romea, P.; Urpí, F.; Vilarrasa,
J. Tetrahedron Lett. 1997, 38, 1637. (b) Solsona, J. G.;
Romea, P.; Urpí, F.; Vilarrasa, J. Org. Lett. 2003, 5, 519.
(c) Solsona, J. G.; Romea, P.; Urpí, F. Tetrahedron Lett.
2004, 45, 5379.
(16) (a) Stork, G.; Paterson, I.; Lee, F. K. C. J. Am. Chem. Soc.
1982, 104, 4686. (b) Burke, S. D.; Schoenen, F. J.;
Murtiashaw, C. W. Tetrahedron Lett. 1986, 27, 449.
(c) Stork, G.; Rychnovsky, S. D. J. Am. Chem. Soc. 1987,
109, 1565. (d) Nakata, M.; Arai, M.; Tomooka, K.; Ohsawa,
N.; Kinoshita, M. Bull. Chem. Soc. Jpn. 1989, 62, 2618.
(e) Kochetkov, N. K.; Sviridov, A. F.; Ermolenko, M. S.;
Yashunsky, D. V.; Borodkin, V. S. Tetrahedron 1989, 45,
5109. (f) Tone, H.; Nishi, T.; Oikawa, Y.; Hikota, M.;
Yonemitsu, O. Chem. Pharm. Bull. 1989, 37, 1167.
(g) Mulzer, J.; Kirstein, H. M.; Buschmann, J.; Lehmann, C.;
Luger, P. J. Am. Chem. Soc. 1991, 113, 910. (h) Hoffmann,
R. W.; Stürmer, R. Chem. Ber. 1994, 127, 2511.
(17) (a) Narasaka, K.; Pai, F.-C. Tetrahedron 1984, 40, 2233.
(b) Chen, K.-M.; Hardtmann, G. E.; Prasad, K.; Repic, O.;
Shapiro, M. J. Tetrahedron Lett. 1987, 28, 155.
(18) Nakata, T.; Tani, Y.; Hatozaki, M.; Oishi, T. Chem. Pharm.
Bull. 1984, 32, 1411.
(19) Kiyooka, S.-i.; Kuroda, H.; Shimasaki, Y. Tetrahedron Lett.
1986, 27, 3009.
(20) The relative 3,4-syn-4,5-syn configuration was secured by
NMR studies of the dioxane moiety.
(4) Masamune, S.; Choy, W.; Petersen, J. S.; Sita, L. R. Angew.
Chem., Int. Ed. Engl. 1985, 24, 1.
(5) (a) Heathcock, C. H.; White, C. T. J. Am. Chem. Soc. 1979,
101, 7076. (b) Evans, D. A.; Dart, M. J.; Duffy, J. L.; Rieger,
D. L. J. Am. Chem. Soc. 1995, 117, 9073. (c) Marco, J. A.;
Carda, M.; Díaz-Oltra, S.; Murga, J.; Falomir, E.; Roeper, H.
J. Org. Chem. 2003, 68, 8577.
(6) For an early example based on a-OTBDPS chiral aldehyde,
see: Esteve, C.; Ferreró, M.; Romea, P.; Urpí, F.; Vilarrasa,
J. Tetrahedron Lett. 1999, 40, 5083.
(7) Ferreró, M.; Galobardes, M.; Martín, R.; Montes, T.; Romea,
P.; Rovira, R.; Urpí, F.; Vilarrasa, J. Synthesis 2000, 1608.
(8) Roush, W. R.; Palkowitz, A. D.; Ando, K. J. Am. Chem. Soc.
1990, 112, 6348.
(9) All new compounds have analytical and spectroscopic data
consistent with the assigned structure. The absolute
configurations were initially established by analogy and
those of 5, 6, 9, 10, and 11 have been later confirmed by
spectroscopic analysis of cyclic derivatives. See, for
instance, ref. 3c.
(10) As expected, the minor diastereomer observed in all cases is
the alternative 4,5-syn isomer. No significant amounts of
anti adducts were indeed detected throughout the overall
study.
(11) Roush, W. R. J. Org. Chem. 1991, 56, 4151.
(12) Overall yields of purified materials are indicated.
Diastereomeric ratios have been established through HPLC
and NMR analysis.
(13) Solsona, J. G.; Romea, P.; Urpí, F. Org. Lett. 2003, 5, 4681.
(14) Tatsuta, K. In Recent Progress in the Chemical Synthesis of
Antibiotics; Lukacs, G.; Ohno, M., Eds.; Springer-Verlag:
Berlin, 1990.
(15) For recent examples, see the following reports and
references therein: (a) Evans, D. A.; Kim, A. S.; Metternich,
R.; Novack, V. J. J. Am. Chem. Soc. 1998, 120, 5921.
(b) Hergenrother, P. J.; Hodgson, A.; Judd, A. S.; Lee,
W.-C.; Martin, S. F. Angew. Chem. Int. Ed. 2003, 42, 3278.
(c) Peng, Z.-H.; Woerpel, K. A. J. Am. Chem. Soc. 2003,
125, 6018.
(21) Unexpectedly, this step turned out to be troublesome.
Debenzylation with Pd(OH)2/C was sluggish in EtOAc and
afforded a complex mixture in MeOH. Better results (60%
yield) were obtained with 10% Pd/C in EtOH, although it
was not possible to prevent partial removal of acetonide
protecting group and the corresponding triol was produced
in 25% yield. Finally, alcohol 16 was isolated in excellent
yield (91%) after 3 h in EtOAc.
(22) Physical and spectroscopic data of ketone 13 are in
agreement with those previously reported. See ref. 16a, 16f.
Compound 13: colorless oil. Rf (hexanes–EtOAc 85:15) =
0.45. [a]D +23.1 (c 1.8, CHCl3). IR (film): n = 2933, 1717,
1111, 1017 cm–1. 1H NMR (400 MHz, CDCl3): d = 7.66–
7.54 (4 H, m, ArH), 7.43–7.36 (6 H, m, ArH), 4.22 (1 H, d,
J = 2.5 Hz, CH3COCHO), 3.73 (1 H, dd, J = 9.6 Hz, J = 1.9
Hz, CHOCHCH2OSi), 3.57 (1 H, dd, J = 10.3 Hz, J = 4.3
Hz, CHxHyOSi), 3.49 (1 H, dd, J = 10.3 Hz, J = 5.7 Hz,
CHxHyOSi), 2.12 (3 H, s, CH3CO), 2.03–1.95 [1 H, m,
OHCCH(CH3)CHO], 1.83–1.78 (1 H, m, CHCH2OSi), 1.48
(3 H, s, CH3CCH3), 1.41 (3 H, s, CH3CCH3), 1.06 [9 H, s,
SiC(CH3)3], 1.05 (3 H, d, J = 6.8 Hz, CH3CHCH2OSi), 0.70
[3 H, d, J = 6.6 Hz, OHCCH(CH3)CHO]. 13C NMR (100.6
MHz, CDCl3): d = 209.3 (C), 135.6 (CH), 135.5 (CH), 133.5
(C), 133.4 (C), 129.7 (CH), 127.7 (CH), 127.6 (CH), 99.4
(C), 79.4 (CH), 75.2 (CH), 64.9 (CH2), 36.7 (CH), 32.4
(CH), 29.8 (CH3), 27.0 (CH3), 26.8 (CH3), 19.3 (C), 19.1
(CH3), 14.2 (CH3), 6.5 (CH3). HRMS (+FAB): m/z calcd for
C28H41O4Si [M + H]+: 469.2774. Found: 469.2762.
Synlett 2004, No. 12, 2127–2130 © Thieme Stuttgart · New York