Highly diastereoselective synthesis of β-hydroxy amides from β-keto amides

Article abstract of DOI:10.1055/s-2004-829192

β-Hydroxy amides with stereodefined geometry represent an important unit present in various natural products. The diastereoselective preparation of amides carrying a secondary or tertiary alcohol in β-position is described here.

Full text of DOI:10.1055/s-2004-829192

3092
PRACTICAL SYNTHETIC PROCEDURES
Highly Diastereoselective Synthesis of b-Hydroxy Amides from b-Keto Amides
H
ighly Diastereose
i
lective
u
Synthesis
o
s
f
b
-Hydro
e
xy
A
mide
s
ppe Bartoli,*a Marcella Bosco,a Enrico Marcantoni,b Massimo Massaccesi,a Samuele Rinaldi,c Letizia Sambria
a
b
c
Dipartimento di Chimica Organica ‘A. Mangini’, v. le Risorgimento 4, 40136 Bologna, Italy
Fax +39(51)2093654; E-mail: giuseppe.bartoli@unibo.it
Dipartimento di Scienze Chimiche, Università di Camerino, v. S. Agostino 1,
62032 Camerino (MC), Italy
Dipartimento di Scienze e Materiali della Terra, Università di Ancona, v. Brecce Bianche,
60131 Ancona, Italy
PSP
Received 1 June 2004
No35
Abstract: b-Hydroxy amides with stereodefined geometry represent an important unit present in various natural products. The diastereo-
selective preparation of amides carrying a secondary or tertiary alcohol in b-position is described here.
Key words: diastereoselectivity, hydroxy amides, TiCl4, organocerium reagents, boranes
1) TiCl4, CH2Cl2, –30 °C
2) R2MgX-CeCl3,
THF or Et2O, –78 °C
3) HCl aq
O
OH
R2
R1
N
( )-2
O
O
Procedure 1
N
R1
Procedure 2
( )-1
O
OH
R1
1) TiCl4, CH2Cl2, –30 °C
2) BH3 Me2S, –78 °C
N
.
3) HCl aq
( )-3
Scheme 1
Introduction
In the design of a stereoselective process the formation of
a cyclic metal-chelate intermediate is frequently planned.6
The construction of b-hydroxy amides with stereodefined In fact, the interaction of a bidentate system like 1
geometry (Scheme 1) is an important target in organic (Scheme 2) with an appropriate chelating agent leads to
synthesis since these units are present in various natural the formation of a cyclic complex, whose structure can be
products. For instance, an amidic group carrying a tertiary accounted for by an equilibrium between 4A and 4B con-
alcohol in the b-position is a structural unit in a series of formations. Owing to the A1,2 strain present in 4B, the
inhibitors of MMPs enzymes;1 moreover, b-hydroxy equilibrium should be shifted towards the most stable con-
amides can be used as useful building blocks for the syn- formation 4A.7
thesis of biologically active compounds.2 Therefore, vari-
ous strategies have been planned and developed for their
synthesis.
H
N
O
O
Me
R1
Among these, the most direct approaches are the aldolic
MLn
condensation3 between an amide enolate and a ketone, or
( )-4B
O
O
an aldehyde, and the stereoselective addition of an orga-
nometallic species4 or of a reducing agent5 to the carbonyl
of a b-keto amide, carrying a stereocenter in a-position.
Since the stereochemical outcome of the aldol condensa-
tion is hardly controllable, the latter approach seems to be
a more promising and practicable tool.
MLn
N
R1
Nu-
( )-1
N
MLn
O
OH
O
O
MLn= TiCl4
Nu-= "H-", "R-"
Nu
H
N
R1
R1
Me
SYNTHESIS 2004, No. 18, pp 3092–3096
x
x.
x
x
.
2
0
0
4
( )-5
Advanced online publication: 13.08.2004
( )-4A
DOI: 10.1055/s-2004-829192; Art ID: Z10804SS
© Georg Thieme Verlag Stuttgart · New York
Scheme 2

PRACTICAL SYNTHETIC PROCEDURES
Highly Diastereoselective Synthesis of b-Hydroxy Amides
3093
At variance with 4B, 4A presents a significant stereofacial Organocerium derivatives, however, can be prepared only
discrimination. In fact the attack from the lower side of in ethereal solvents. Hence, to form a cyclic intermediate,
the molecule is prevented by the axial methyl group posi- we needed a Lewis acid able to chelate even in coordinat-
tioned in the same direction of the ideal trajectory of an in- ing solvents (THF, Et2O). Between various Lewis acids,
coming nucleophile. In conclusion, the winning strategy TiCl4 proved to be the best choice. In fact we found that it
for a high stereocontrol of the process is, in principle, to is strong enough to ensure the formation of a stable and
pursue the most appropriate reaction conditions so as to rigid chelate with b-keto amides 1 even in the presence of
favor the shift of the equilibrium towards the 4A confor- ethereal cosolvents. The reaction was tested on two sub-
mation.
strates 1a (R1 = Ph) and 1b (R2 = Et), carrying a phenyl
group and an ethyl group bound to the carbonyl, respec-
tively. Our results are reported in Table 1.12
The choices of the carbanionic moiety and the hydride ion
sources are also a crucial point. These reagents in fact
should have a high nucleophilic character associated with
a low basicity, so as to minimize proton abstraction phe-
nomena at the a-position.8 Although this undesirable side
reaction should easily occur only at the axial hydrogen of
the minor conformer 4B, however, it could become com-
petitive when the addition process is slow on account of
steric factors and/or the equilibrium is not appreciably
shifted towards 4A (small A1,2 strain).9
Table 1 Diastereoselective Addition of Organocerium Reagents to
b-Keto Amides 1 in THF at –78 °C, Unless Otherwise Mentioned
O
O
O
OH
1) TiCl4, CH2Cl2, –30 °C
R2
R1
N
R1
N
2) R2MgX-CeCl3, THF, –78 °C
3) HCl aq
( )-1
( )-2
Entry R1
R2 in R2MgX- Product
CeCl3
Yield (%)a de (%)b
In this ‘Practical Synthetic Procedure’ we present the best
reaction conditions for the stereocontrolled synthesis of b-
hydroxy amides.
1
2
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Et
Et
Et
Et
Et
Et
Et
Me
2aa
2ab
2ac
2ad
2ae
>99
65
85
90
85
>99
60c
60c
85
60
85
97
99
0e
>98
>98
>98
>98
>98
>98
>98
>98
>98
94
Et
3
n-Bu
PhCH2
PhC≡C
Scope and Limitations
4
Our first goal was to set up a general and efficient protocol
(Procedure 1) to obtain b-hydroxy amides having a tertia-
ry alcoholic fragment by addition of the appropriate orga-
nometallic species to a metal complex of the starting b-
keto amide 1 (Scheme 1).
5
6
CH2=CHCH2 2af
7
i-Pr
2ag
2ah
2ba
2bc
2bi
8
t-Bu
Med
Although the addition of organometallic reagents to oxo-
amides had been previously studied, the few methods
present in the literature did not supply general conditions.4
In these works, in order to obtain high stereoselectivity it
was indeed necessary to tune the choice of the reagent to
the structure of the carbon chain to be introduced at the
prostereogenic carbonylic group. We were looking for an 12
organometallic species able to transfer to the carbonyl
various carbon frameworks, independently from the na-
9
10
11
n-Bud
Phd
>98
>98
>98
–
PhCH2
2bd
13
CH2=CHCH2 2be
14
15
i-Prd
2bf
ture of the substrate, in order to furnish general applicabil-
ity to the proposed protocol.10
t-Bud
2bg
0f
–
Organocerium reagents completely fulfill these require-
ments.11 In fact, a large variety of carbon frameworks is
available since they can be easily prepared from the cor-
responding RMgX derivatives. Moreover, they generally
exhibit a high nucleophilicity, in some cases superior to
a Yields refer to pure isolated products.
b Determined on the crude product.
c Together with small amounts of starting material.
d Reaction carried out in Et2O.
e 85% of starting material recovered.
that of the parent compounds. At the same time, their low f 90% of starting material recovered.
basic character avoids the occurrence of extensive eno-
lization processes.
When R1 = Ph, the reaction shows in all cases a high ste-
reochemical control. Yields vary from good to excellent.
In addition the protocol has a general applicability, in fact
a large variety of carbon frameworks can be introduced,
including saturated alkyl chains, as well as alkynylic, al-
lylic and benzylic moieties (Table 1, entries 1–6). More-
over, we wish to outline that this is the first methodology
that allows bulky alkyl chains, such as i-Pr and t-Bu, to be
The high nucleophilicity of the organometallic species is
crucial for the success of the reaction since it allows the
addition to be carried out at low temperature. In fact, ac-
cording to the Boltzmann law, the population of the most
stable conformer 4A increases with lowering of the tem-
perature.
Synthesis 2004, No. 18, 3092–3096 © Thieme Stuttgart · New York

3094
G. Bartoli et al.
PRACTICAL SYNTHETIC PROCEDURES
introduced with very satisfactory yields (Table 1, entries We tested various hydride donors. The best choice was the
7 and 8).
BH3·Me2S complex16 since it can act at low temperatures
in poorly coordinating solvents, such as CH2Cl2. The rea-
sons of this choice are as follows: the small hydride ion is
less sensitive, with respect to bulky carbanionic nucleo-
philes, to the stereofacial discrimination exerted by the
axial methyl group in 4A conformation. It was therefore
necessary to increase the stability and rigidity of the che-
late complex by carrying out the reaction in poorly coor-
dinating solvents which cannot compete with the substrate
in the coordination of Ti(IV).17
When R1 is a linear alkyl group, as in 1b, some problems
arise in the organometallic addition to the carbonyl. Ow-
ing to the presence of a less hindered group bound to the
carbonyl, the A1,2 strain decreases with a consequent in-
creased amount of 4B at the equilibrium. Yields are excel-
lent with benzylic and allylic derivatives (Table 1, entries
12 and 13), but they remarkably decrease with nonstabi-
lized carbanionic moieties. In these cases, yields can be
improved using a less coordinating and less polar solvent,
such as Et2O instead of THF, (Table 1, entries 9–11). The results reported in Table 2 show that the reaction pro-
However, when highly sterically hindered and highly ba- ceeds with high diastereoselectivity and in almost quanti-
sic carbanionic moieties, such as i-Pr and t-Bu, are present tative yields. It is interesting to note that, contrary to other
in the organocerium reagents the enolization process borane complexes,16 the reduction with BH3·Me2S does
largely prevails even carrying out the reaction in Et2O, not proceed on to g-aminols, but it stops at b-hydroxy
(Table 1, entries 14 and 15).
amides.
The reaction proceeds in all cases with very high diaste-
reoselectivity; indeed the exclusive formation of the dia-
stereoisomer derived from the attack of the carbanionic
moiety opposite to the a-methyl group was observed ac-
cording to the mechanism depicted in Scheme 2. An im-
portant feature of this methodology is the opportunity to
obtain both diastereoisomers of a certain b-hydroxy
amide simply by exchanging the residue R1 on the b-keto
amide with R2 in the organometallic reagent (Table 1, en-
try 1 and entry 9).
Table 2 Diastereoselective Reduction of b-Keto Amides 1 with
BH3·Me2S, in CH2Cl2 at –78 °C
O
O
O
OH
R1
1) TiCl4, CH2Cl2, –30 °C
N
R1
N
2) BH3-Me2S, –78 °C
3) aq HCl 4%
( )-1
( )-3
Entry
R1
Product
3a
Yield (%)a de (%)b
1
2
3
4
5
6
7
Ph
Et
>99
>99
>99
>99
>99
>99
>99
96
90
Organocerium reagents are frequently employed in large
amount, and also in our procedure an excess of
R2MgX·CeCl3 is required. This can be partially due to in-
teractions of the organocerium reagents with TiCl4
present in the reaction mixture to give organotitanium
species. However, the addition to the carbonyl proceeds
through a direct attack of R2MgX·CeCl3 on the intermedi-
ate 4. A convincing evidence for this assumption arises
from results obtained in the reaction with i-
PrMgCl·CeCl3.13 It is well known that i-propylmagnesium
bromide smoothly rearranges, even at low temperatures,
to n-propylmagnesium chloride in the presence of small
amounts of TiCl4.14 In the reaction of i-PrMgCl·CeCl3
with 1a (Table 1, entry 7) the exclusive incorporation of
the a-branched chain was found in the final product 2ag.
Then the productive attack of i-PrMgCl·CeCl3 on the car-
bonyl group has to occur before any interactions with
chlorotitanium species.
3b
i-Pr
3c
>98
96
C5H11
3d
3e
3f
t-Bu
80
c-C6H11
p-BrC6H4
>98
>98
3g
a Yields refer to pure isolated products.
b Determined on the crude product.
Due to the efficiency and the high diastereoselectivity ob-
served, the TiCl4-BH3·Me2S method represents a very
useful alternative to previously reported procedures.5
Experimental Procedures
Moreover, alkynyltitanium derivatives are not stable even
at low temperatures and cannot be successfully employed
in synthesis.15 The success of the reaction with the alkynyl
framework can be explained again in terms of direct at-
tack of the organocerium reagent at the carbonyl group.
Herein we describe the typical practical synthetic procedures for the
synthesis of b-hydroxy amides. Depending on the nature of the de-
sired hydroxy amide, the addition or the reduction procedure should
be chosen.
In fact, Procedure 1, carried out by addition of organocerium re-
agents to b-keto amides pretreated with TiCl4, is used for the stere-
ocontrolled preparation of amides having a tertiary alcohol in b-
position. The described typical procedure employs THF as the sol-
vent for the preparation of organocerium compounds. When a less
coordinating solvent is required, the anhydrous CeCl3 is suspended
in Et2O and a Et2O solution of Grignard reagent is added.
The same concepts can be applied to the synthesis of func-
tionalized amides carrying a secondary alcohol in b-posi-
tion. This target can be achieved by developing an
appropriate procedure for the reduction of the same inter-
mediate 4. In fact, the addition of an appropriate hydride
ion source to the less hindered side of the metal chelate 4
will lead to syn-b-hydroxy amides 3.
Synthesis 2004, No. 18, 3092–3096 © Thieme Stuttgart · New York

PRACTICAL SYNTHETIC PROCEDURES
Highly Diastereoselective Synthesis of b-Hydroxy Amides
3095
On the other hand, when a secondary alcohol is required, Proce-
dure 2 can be used to successfully reduce the b-keto amides 1 with
BH3·Me2S in CH2Cl2.
Acknowledgment
Work carried out in the framework of the National Project ‘Stereo-
selezione in Sintesi Organica. Metodologie e Applicazioni’ suppor-
ted by MIUR, Rome, by the University of Bologna, in the
framework of ‘Progetto di Finanziamento Pluriennale, Ateneo di
Bologna’.
(2R*,3R*)-3-Hydroxy-N,N,2,4-tetramethyl-3-phenylpentan-
amide (2ag); Typical Procedure 1
The organocerium reagent should be prepared8 separately.
CeCl3·7H2O (2.98 g, 8.0 mmol) was quickly and finely ground in a
mortar and placed in a three-necked flask equipped with a stirrer bar
and a three-way cock. The flask was immersed in an oil bath and
gradually heated to 135–140 °C under vacuum (<0.5 mmHg). After
1 h at this temperature, the cerium chloride was completely dried in
vacuo by stirring at the same temperature for an additional hour.
While the flask was still hot, argon gas was introduced and the flask
then cooled in an ice bath. THF (10 mL) freshly distilled from sodi-
um/benzophenone was added all at once with vigorous stirring. The
ice bath was removed and the suspension left to stir overnight under
argon at r.t. Then, a 2 M THF solution of i-PrMgCl (4 mL, 8 mmol)
was added dropwise to the stirring suspension of anhyd CeCl3 at
0 °C under argon and the stirring was continued for 1.5 h. The reac-
tion mixture resulted in a dark grey suspension. Meanwhile, an oven
dried 100 mL three necked flask, equipped with a stirring bar, was
charged with a solution of N,N,2-trimethyl-3-oxo-3-phenylpropan-
amide (1a, R1 = Ph; 205 mg, 1.0 mmol) in anhyd CH2Cl2 (10 mL)
under argon and cooled at –30 °C. Then a 1 M CH2Cl2 solution of
TiCl4 (1.05 mL, 1.05 mmol) was added. After 30 min at –30 °C, the
reaction mixture was cooled to –78 °C and the previously prepared
suspension of i-PrMgCl-CeCl3 (8 mmol) in THF was added with a
syringe. The mixture was left to stir at –78 °C for 30 min, and then
quenched with aq HCl (ca. 1 M, 20 mL). The aqueous layer was sep-
arated and washed with additional Et2O (20 mL). The combined or-
ganic layers were dried (MgSO4), filtered and concentrated by
rotary evaporation. The crude product was purified by silica gel col-
umn chromatography using a Et2O–petroleum ether (80:20) mix-
ture as eluent affording 2ag (150 mg, 60%) as a yellow solid; mp
102.3–102.5 °C.
References
(1) Jacobson, I. C.; Reddy, P. G.; Wasserman, Z. R.; Hardman,
K. D.; Covington, M. B.; Arner, E. C.; Copeland, R. A.;
Decicco, C. P.; Magolda, R. Bioorg. Med. Chem. Lett. 1998,
8, 837.
(2) Kakei, H.; Nemoto, T.; Ohshima, T.; Shibasaki, M. Angew.
Chem. Int. Ed. 2004, 43, 317.
(3) (a) Heathcock, C. H. In Asymmetric Synthesis, Vol. 3;
Morrison, J. D., Ed.; Academic Press: New York, 1984,
111. (b) Evans, D. A.; Nelson, J. V.; Taber, T. R. Top.
Stereochem. 1984, 13, 1. (c) Evans, D. A.; McGee, L. R.
Tetrahedron Lett. 1980, 21, 3975. (d) Goasdoue, C.;
Goasdoue, N.; Gaudemar, M. J. Organomet. Chem. 1981,
208, 279. (e) Canceill, J.; Jacques, J. Bull. Soc. Chim. Fr.
1970, 2180. (f) Kamimura, A.; Omata, Y.; Mitsudera, H.;
Kakehi, A. J. Chem. Soc., Perkin Trans. 1 2000, 4499.
(4) (a) Taniguchi, M.; Fujii, H.; Oshima, K.; Utimoto, K. Bull.
Chem. Soc. Jpn. 1994, 67, 2514. (b) Molander, G. A.;
Andrews, S. W. Tetrahedron Lett. 1986, 27, 3115.
(c) Molander, G. A.; Etter, J. B.; Zinke, P. W. J. Am. Chem.
Soc. 1987, 109, 453.
(5) (a) Taniguchi, M.; Fujii, H.; Oshima, K.; Utimoto, K.
Tetrahedron 1993, 49, 11169. (b) Fujita, M.; Hiyama, T. J.
Am. Chem. Soc. 1985, 107, 8294. (c) Fujita, M.; Hiyama, T.
J. Org. Chem. 1988, 53, 5405. (d) Fujita, M.; Hiyama, T. J.
Org. Chem. 1988, 53, 5415. (e) Fujita, M.; Hiyama, T. Org.
Synth. 1990, 69, 44. (f) Fujii, H.; Oshima, K.; Utimoto, K.
Tetrahedron Lett. 1991, 32, 6147.
1H NMR (CDCl3, 300 MHz): d = 0.69 (d, 3 H, CH3, JH,H = 6.7), 0.79
(d, 3 H, CH3, JH,H = 7.0), 0.92 (d, 3 H, CH3, JH,H = 7.0), 1.6 (br s, 1
H, OH), 2.00–2.15 (m, 1 H, CH), 3.02 (s, 3 H, CH3), 3.20 (s, 3 H,
CH3), 3.40 (q, 1 H, CH, JH,H = 7.0), 7.15–7.40 (m, 5 H, C6H5).
13C NMR (CDCl3, 75 MHz): d = 13.4 (CH3), 17.5 (CH3), 18.4
(CH3), 35.6 (CH3), 37.5 (CH3), 37.7 (CH), 38.9 (CH), 80.0 (C),
126.2 (CH), 127.4(CH), 141.7 (C), 178.0 (C).
(6) Mengel, A.; Reiser, O. Chem. Rev. 1999, 99, 1191.
(7) Bartoli, G.; Bosco, M.; Dalpozzo, R.; Marcantoni, E.;
Sambri, L. Chem.–Eur. J. 1997, 3, 1941.
(8) Bartoli, G.; Bosco, M.; Di Martino, E.; Marcantoni, E.;
Sambri, L. Eur. J. Org. Chem. 2001, 2901.
(9) The proton abstraction in 4A is expected to be difficult due
to stereoelectronic factors (H in pseudo equatorial position):
Deslongchamps, P. In Stereoelectronic Effects in Organic
Chemistry, Vol. 1; Pergamon: New York, 1983, 274.
(10) Bartoli, G.; Marcantoni, E.; Sambri, L. In Seminars in
Organic Synthesis: XXV Summer School ‘A. Corbella’ 12-
16/6/2000, Gargnano; Società Chimica Italiana: Camerino,
2000, 117–138.
(11) (a) Imamoto, T. Lanthanides in Organic Synthesis;
Academic Press: New York, 1994, 68. (b) Imamoto, T. In
Comprehensive Organic Synthesis, Vol. 1; Trost, B. M.;
Fleming, I.; Schreiber, S. L., Eds.; Pergamon: London, 1991,
Chap. 1.8.
(2R*,3R*)-3-Hydroxy-N,N,2,-Trimethyl-3-phenylpropion-
amide (3a); Typical Procedure 2
An oven dried 100 mL three necked flask, equipped with a stirring
bar, was charged with a solution of N,N,2-trimethyl-3-oxo-3-phe-
nylpropanamide (1a, R1 = Ph; 205 mg, 1.0 mmol) in anhyd CH2Cl2
(10 mL) under argon and cooled at –30 °C. Then a 1 M CH2Cl2 so-
lution of TiCl4 (1.1 mL, 1.1 mmol) was added. After 30 min at
–30 °C, the reaction was cooled to –78 °C and a 10 M solution of
BH3·Me2S complex in Me2S (0.5 mL, 5 mmol) was added dropwise.
The mixture was left to stir at –78 °C for 3 h, and then quenched
with aq HCl (ca. 1 M, 20 mL) and extracted with CH2Cl2 (3 × 10
mL). The combined organic layers were dried (MgSO4), filtered and
concentrated by rotary evaporation. The crude product was purified
by flash column chromatography on aluminum oxide using a Et2O–
petroleum ether (80:20) mixture as eluent affording 3a5 (205 mg,
>99%) as a white solid.
(12) Bartoli, G.; Bosco, M.; Marcantoni, E.; Massaccesi, M.;
Rinaldi, S.; Sambri, L. Tetrahedron Lett. 2001, 42, 6093.
(13) Bartoli, G.; Bellucci, M. C.; Bosco, M.; Marcantoni, E.;
Sambri, L. Chem.–Eur. J. 1998, 4, 2154.
(14) Cooper, G. D.; Finkbeiner, H. L. J. Org. Chem. 1962, 27,
1493.
(15) Reetz, M. T. Organometallics in Synthesis, A Manual;
Schlosser, M., Ed.; 1994, 195.
(16) Bartoli, G.; Bosco, M.; Marcantoni, E.; Massaccesi, M.;
Rinaldi, S.; Sambri, L. Tetrahedron Lett. 2001, 42, 8811.
Synthesis 2004, No. 18, 3092–3096 © Thieme Stuttgart · New York

3096
G. Bartoli et al.
PRACTICAL SYNTHETIC PROCEDURES
(17) (a) This procedure can be successfully applied, with slight
modifications (the use of BH3·py complex instead of
BH3·Me2S) to closely related systems. (b) For the reduction
of b-keto sulfones, see: Bartoli, G.; Bosco, M.; Cingolani, S.;
Marcantoni, E.; Sambri, L. J. Org. Chem. 1998, 63, 3624.
(c) For the reduction of b-keto esters, see: Bartoli, G.;
Bellucci, M. C.; Alessandrini, S.; Malavolta, M.;
Marcantoni, E.; Sambri, L.; Dalpozzo, R. J. Org. Chem.
1999, 64, 1986. (d) For the reduction of a-nitroketones, see:
Ballini, R.; Bosica, G.; Marcantoni, E.; Vita, P.; Bartoli, G.
J. Org. Chem. 2000, 65, 5854. (e) For the reduction of b-
keto carbonitriles, see: Bartoli, G.; Bosco, M.; Dalpozzo, R.;
De Nino, A.; Procopio, A.; Sambri, L.; Tagarelli, A. Eur. J.
Org. Chem. 2001, 2971.
Synthesis 2004, No. 18, 3092–3096 © Thieme Stuttgart · New York