Regioselective asymmetric α,α-bisalkylation of ketones via complex-induced Syn -deprotonation of chiral N -amino cyclic carbamate hydrazones

Article abstract of DOI:10.1021/ja202267k

The first general method for the asymmetric α,α-bisalkylation of ketones having both α- and α′-protons is described. Both excellent regio- and stereoselectivity result. The transformation is enabled by complex-induced syn-deprotonation (CIS-D), which completely reverses the inherent preference of lithium diisopropylamide (LDA) to remove the less sterically hindered of two similarly acidic protons. CIS-D also overrides the normal tendency of LDA to remove the more strongly acidic proton in a substrate having protons differing significantly in their acidity. The regiochemical outcome is, thus, the opposite of that normally obtained for kinetic LDA-mediated deprotonation of ketones and (S)-1-amino-2- methoxymethylpyrrolidine/(R)-1-amino-2-methoxymethylpyrrolidine (SAMP/RAMP)hydrazones. Conveniently, this strategy allows access to either ketone enantiomer using a single enantiomer of the auxiliary. The utility of this method is demonstrated by a concise and highly efficient formal synthesis of both (R)- and (S)-stigmolone.

Full text of DOI:10.1021/ja202267k

ARTICLE
pubs.acs.org/JACS
Regioselective Asymmetric r,r-Bisalkylation of Ketones via
Complex-Induced Syn-Deprotonation of Chiral N-Amino Cyclic
Carbamate Hydrazones
Sarah E. Wengryniuk, Daniel Lim, and Don M. Coltart*
Department of Chemistry, Duke University, Durham, North Carolina 27708, United States
S Supporting Information
b
ABSTRACT: The first general method for the asymmetric
R,R-bisalkylation of ketones having both R- and R0-protons is
described. Both excellent regio- and stereoselectivity result.
The transformation is enabled by complex-induced syn-depro-
tonation (CIS-D), which completely reverses the inherent
preference of lithium diisopropylamide (LDA) to remove the
less sterically hindered of two similarly acidic protons. CIS-D
also overrides the normal tendency of LDA to remove the more
strongly acidic proton in a substrate having protons differing
significantly in their acidity. The regiochemical outcome is, thus,
the opposite of that normally obtained for kinetic LDA-mediated deprotonation of ketones and (S)-1-amino-2-methoxymethyl-
pyrrolidine/(R)-1-amino-2-methoxymethylpyrrolidine (SAMP/RAMP)hydrazones. Conveniently, this strategy allows access to
either ketone enantiomer using a single enantiomer of the auxiliary. The utility of this method is demonstrated by a concise and
highly efficient formal synthesis of both (R)- and (S)-stigmolone.
’ INTRODUCTION
of ketones having both R- and R0-protons (cf. 7 f 8 f 9 f 5).
Regiocontrol is achieved using chiral N-amino cyclic carbamate
(ACC) hydrazones through a process we term complex-induced
syn-deprotonation (CIS-D) (Scheme 2b). This bisalkylation
method is remarkable in producing nearly perfect levels of
both stereo- and regioselectivity. Moreover, the regiochemical
outcome is the opposite of that normally obtained for kinetic
LDA-mediated deprotonation of ketones and SAMP/RAMP
hydrazones. Conveniently, this strategy allows access to either
ketone enantiomer using a single enantiomer of the auxiliary by
simply altering the alkylation sequence. To highlight the effec-
tiveness and efficiency of this R,R-bisalkylation method, a formal
asymmetric synthesis of both (R)- and (S)-stigmolone was
carried out.
We recently described the development of ACC auxiliaries for
asymmetric ketone alkylation.5ꢀ8 In contrast to other methods,
the auxiliaries are both easily introduced into and removed from
ketones, with nearly quantitative recovery. Furthermore, depro-
tonation is rapid, and alkylation does not require extremely low
temperature, yet it proceeds with excellent stereoselectivity and
substantially greater yields. A key design feature of these aux-
iliaries was the placement of a carbonyl group adjacent to the
hydrazone moiety for enhanced R-proton acidity and tight
chelation at the level of the azaenolate.5 The carbonyl was also
intended to influence the regiochemistry of deprotonation to
The asymmetric R-alkylation of ketones is an important yet
highly challenging transformation.1 Indeed, with regard to the
latter assertion, in contrast to the R-alkylation of carboxylic acid
derivatives, for which numerous chiral auxiliary-based asym-
metric methods are available,2 only a single method is available
for ketone R-alkylation that has found application in the synth-
esis of natural products, albeit in only a handful of cases.1c This
method employs the well-known Enders (S)-1-amino-2-meth-
oxymethylpyrrolidine/(R)-1-amino-2-methoxymethylpyrroli-
dine (SAMP/RAMP) chiral auxiliaries.1aꢀc As in situations
involving chiral carboxylic acid derivatives,2 both azaenolate
geometry and facial selectivity must be controlled during the
alkylation of hydrazones. However, adding to the complexity of
hydrazone alkylation is the need for tight regiochemical control
during deprotonation of nonsymmetrical ketone derivatives
having both R- and R0-protons (cf. 2, Scheme 1a). For SAMP/
RAMP hydrazones this is achieved kinetically using lithium
diisopropylamide (LDA) and generally results in removal of
the more sterically accessible proton.1aꢀc,3 As a result, the
controlled asymmetric R,R-bisalkylation of ketone derivatives
having indistinguishable R- and R0-protons is not possible (cf. 2
f 3 þ 4). However, if such a transformation could be achieved,
then, given the expansive body of literature on ketone manipula-
tion, the bisalkylated products4 would provide access to an
unusually wide array of chiral, nonracemic intermediates for
use in asymmetric synthesis. In what follows, we describe our
studies on the first method for the asymmetric R,R-bisalkylation
Received: March 11, 2011
Published: April 21, 2011
r
2011 American Chemical Society
8714
dx.doi.org/10.1021/ja202267k J. Am. Chem. Soc. 2011, 133, 8714–8720
|

Journal of the American Chemical Society
ARTICLE
Scheme 1. Regioselectivity of Bisalkylation of (a) Chiral
Dialkyl Hydrazones Compared to (b) ACC Hydrazones
Scheme 3. Studies on Complex-Induced Syn-Deprotonation
Scheme 2. (a) Syn-Dianion Effect in Sulfonyl Hydrazones
and (b) R,R-Bisalkylation via Complex-Induced
Syn-Deprotonation (CIS-D) of ACC Hydrazonesa
sulfonyl hydrazone directs an incoming base to remove a proton
on the same side of the CdN bond that it is on. This results in the
formation of a configurationally stable dianion that can be
alkylated (12 f 13). The process may then be repeated (13 f
14), again in a configurationally controlled manner with regard to
the hydrazone CdN bond geometry. We reasoned that a similar
directed deprotonation might be possible by simply utilizing the
carbonyl lone pair10 of an ACC hydrazone via CIS-D. Here, the
carbonyl oxygen electrons would coordinate with the base, direct-
ing deprotonation to the same side of the carbonꢀnitrogen double
bond (cf. 15 f 16, 17f 18). If such a directed deprotonation
could be achieved, then providing that the azaenolates involved (cf.
16, 18) were configurationally stable and that the monoalkylation
product (cf. 17) did not isomerize once formed, asymmetric R,R,-
bisalkylation would be possible.11
a S = small substituent; L = large substituent.
enable the R,R-bisalkylation of ketones having both R- and R0-
protons. Our inspiration for this was the syn-dianion effect of N-
sulfonyl hydrazones.9 In the syn-dianion effect (Scheme 2a), an
N-centered monoanion (11) obtained from deprotonation of a
8715
dx.doi.org/10.1021/ja202267k |J. Am. Chem. Soc. 2011, 133, 8714–8720

Journal of the American Chemical Society
ARTICLE
’ RESULTS AND DISCUSSION
Table 1. Survey of Conditions for the Regioselective
Allylation of 10
Given the novelty of the proposed transformation, at the
outset of our studies we did not know if the intended CIS-D
would occur, or, if it did, whether the resulting azaenolate
intermediates (16, 18) would be configurationally stable about
the original carbonꢀnitrogen double bond. This would be
essential for the proposed asymmetric R,R-bisalkylation process
for two reasons. First, since the monoalkylated hydrazone (17,
Scheme 2) would have to undergo a second directed deprotona-
tion for regiocontrolled access to the second azaenolate (17 f
18), it would be critical that isomerization of the hydrazone
double bond did not occur during the first alkylation (15 f 17),
so that the auxiliary carbonyl could again direct deprotonation to
the R-poistion. Second, since addition of the second electrophile
would have to occur in a diastereoselective fashion, the auxiliary
would need to be positioned on the same side of the carbonꢀ
nitrogen double bond from which the electrophile was approach-
ing to ensure high facial selectivity.11
entry
temp (°C)
time (min)
22:23
yield (%)
1
2
3
4
5
ꢀ78
ꢀ20
ꢀ20
4
30
30
60
60
90
n.d.a
<5
70
84
93
99
>99:1
>99:1
>99:1
>99:1
4
a Not determined.
As an initial test of the mechanistic course of this reaction with
regard to CIS-D, 3-pentanone-derived ACC hydrazone 20 was
prepared and then alkylated with p-bromobenzyl bromide
Table 2. Regioselective Methylation of ACC Hydrazones 10,
30ꢀ32
1
(Scheme 3). A single alkylated product was detected by H
NMR from this reaction and was subsequently analyzed by X-ray
crystallography (Scheme 3) and confirmed to be 21. We were
pleased to find that the p-bromobenzyl substituent and the
auxiliary were positioned on the same side of the hydrazone
CdN bond in this compound, supporting the notion of CIS-D.
Theoretical studies were also carried out on a simplified model
system corresponding to 20 that provided support for intramo-
lecular deprotonation occurring preferentially at the R-position
(cf. 15 f 16) over the R0-position.6
With these encouraging results as a basis, we undertook an R,
R-bisalkylation sequence starting from acetone-derived hydra-
zone 10. To do so, allyl bromide was added to a ꢀ78 °C THF
solution of the lithium azaenolate derived from 10. The cold bath
was removed immediately following addition, and the mixture
was allowed to stir for 20 min and then quenched with H2O. We
were pleased to find that this gave exclusively the R-regioisomer
(22) in excellent yield. Since the allylation product was not
crystalline, its structure was deduced to be 22 by independently
preparing a 15:85 mixture of 22 and 23, respectively, via acid-
catalyzed condensation of 5-hexen-2-one and auxiliary 26.12 The
NMR data of the minor, thermodynamically less favored isomer
of this reaction matched the NMR data of the allylation product
obtained from 10, strongly suggesting this compound to be 22. In
the key part of this bisalkylation study, 22 was subjected to a
second alkylation reaction using p-bromobenzyl bromide as the
alkylating agent. As hoped, the major product was indeed the R,
R-bisalkylated compound (24), which was confirmed by X-ray
crystallography (Scheme 3). Once again, alkylation had occurred
on the same side of the CdN bond as the auxiliary carbonyl. The
diastereoselectivity of the transformation leading to 24 was also
excellent (dr = 97:3). To our knowledge, this was the first
instance of asymmetric R,R-bisalkylation of a ketone having both
R- and R0-protons.
a
entry ACC acetone hydrazone methylated hydrazone R:R0 yield (%)
1
2
3
4
26
27
28
29
10
30
31
32
33
34
35
36
>99:1
91:9
98:2
91:9
96
90
82
93
a Determined by HPLC.
an effort to develop a more reliable protocol for the monoalk-
ylation, we conducted a survey of the reaction conditions, as
outlined in Table 1. We eventually found that, by holding the
alkylation temperature constant at 4 °C and allowing the reaction
to proceed for 1.5 h, an excellent yield of 22 could reliably be
produced in a fully regiocontrolled manner (entry 5). Using
these conditions, we were able to prepare compound 22 starting
with up to 5 g of 10, without compromising the outcome of the
reaction.12 This does not represent the upper limit of the
reaction, but simply the largest scale on which we have conducted
it to date.
Next, we tested other ACC auxiliaries (27ꢀ29) for their
ability to effect regioselective R-alkylation, in the hopes of
providing increased flexibility in our later investigations on the
second, asymmetric alkylation step. To do so, acetone-derived
hydrazones 30ꢀ32 were prepared,12 and each of these
With proof of concept established, we began a further investi-
gation into this new asymmetric alkylation method. Unfortu-
nately, our attempts to prepare more of compound 22 using the
procedure describe above resulted in variable yields of product
and at times led to the formation of 23 as well, the latter
presumably obtained via in situ isomerization of 22. Thus, in
8716
dx.doi.org/10.1021/ja202267k |J. Am. Chem. Soc. 2011, 133, 8714–8720

Journal of the American Chemical Society
ARTICLE
Table 3. Scope and Regioselectivity of the r-Alkylation of 10
via CIS-D
Scheme 4. Studies on Regioselective Asymmetric R,R-
Bisalkylation
Table 4. Methylation of 37
entry
temp (°C)
time (h)
dra
yield (%)
1
2
3
4
ꢀ78
ꢀ78
ꢀ78
ꢀ78
1
6
n.d.b
n.d.b
<5
22
68
98
compounds, in addition to 10, was treated successively with LDA
and then MeI using the conditions established above (Table 1,
entry 5). As shown in Table 2, under these conditions all four of
the hydrazones tested underwent regioselective alkylation
strongly favoring formation of the R-product. However, of these,
hydrazone 10, formed from ACC auxiliary 26, gave the best
result, providing a single regioisomer (entry 1). Consequently,
auxiliary 26 was used for the remainder of our studies.
At this stage, the scope of the monoalkylation was examined
using hydrazone 10 and a variety of alkyl halides (Table 3). In
each case the alkylation proceeded with excellent yield and gave
only the R-regioisomer.13 The transformations proved highly
reliable and were very easy to carry out. These results provided
further evidence of CIS-D occurring during azaenolate forma-
tion. Moreover, they supported the notion that the azaenolate
intermediate was configurationally stable under the reaction
conditions and that isomerization of the hydrazone did not occur
in situ following alkylation.14
12
>99:1
24
>99:1
a Determined by HPLC analysis. b Not determined.
sterically accessible R0-methyl protons of the monoalkylated
compounds (cf. Table 3) and instead direct the removal of the
less accessible R-methylene protons. Moreover, since a stereo-
genic center is formed, the auxiliary must also provide high levels
of asymmetric induction. While the result of our preliminary test
of this transformation described above (22 f 24, Scheme 3) was
very promising, there was clearly room for improvement with
regard to both the regio- and stereochemical outcome.
To investigate the possibility of such improvement, we chose
to study the methylation of hydrazone 37 (Scheme 4). We began
by using the conditions employed above for the p-bromobenzy-
lation of 22 (Scheme 3), which had produced a 92:8 mixture of
24 and 25, respectively. We were pleased to find that, under these
conditions, the methylation of 37 produced only the R,R-
bisalkylation product 46, reproducibly and in excellent yield.
Inspired by this result, we next tried the allylation of 38 under the
same conditions. Unfortunately, in stark contrast to the methyla-
tion of 37, this reaction was extremely problematic, providing
variable yields and mixtures of products that appeared to include
the desired product 47, as well as the undesired (E)-isomer 48.
With a highly effective regioselective, isomerization-free mono-
alkylation procedure established, we began a further investiga-
tion of the regio- and stereocontrolled incorporation of the
second alkyl group at the R-position. This second alkylation
requires an even more demanding application of CIS-D than the
first, in that the ACC auxiliary in this instance must completely
reverse the inherent preference of LDA for removal of the more
8717
dx.doi.org/10.1021/ja202267k |J. Am. Chem. Soc. 2011, 133, 8714–8720

Journal of the American Chemical Society
ARTICLE
Scheme 5. Regioselective Asymmetric R,R-Bisalkylation
With regard to the formation of 48, while it was possible that
isomerization had occurred subsequent to the allylation, in
which case it would have no bearing on the overall level of
asymmetric induction (assuming no epimerization occurred
during isomerization), we could neither be certain of that nor
assume that, if that were the case, the same scenario would apply
for other alkylation reactions. We therefore decided to focus on
developing conditions that would enable regiocontrolled R,R-
bisalkylation without isomerization occurring. We suspected that
this might be possible simply by holding the reaction tempera-
ture at ꢀ78 °C for the entire course of the alkylation, although we
were concerned that the alkylation might not proceed appreci-
ably at such a low temperature. To test this we reverted to the
methylation of 37, which had previously worked extremely well,
and studied the time course of this reaction. We found the
reaction to be complete within 24 h and also determined that no
regioisomeric products had formed (Table 4).13 Moreover, we
were extremely pleased to find that the diastereomer ratio for this
reaction was >99:1.15 An initial test of the scale-up of this reac-
tion was conducted beginning with 4 g of 37 and proceeded
without compromising the outcome of the transformation.12
Gratifyingly, we found that the use of these conditions with a
range of hydrazones and alkylating agents gave the desired R,R-
bisalkylated products in excellent yield, with both complete regio-
and stereochemical control (Scheme 5).16 The absolute config-
uration of the products was inferred by analogy to our previous
experimental work5 and recent theoretical studies.6 An additional
convenience of this bisalkylation procedure is that either
enantiomeric ketone can be formed from the same auxiliary,
following its cleavage, by simply changing the alkylation se-
quence. To highlight this, the alkylations in Scheme 5 were
conducted in a pairwise fashion. In all cases, the R,R-bisalkylated
product was formed with essentially complete regio-13 and
stereochemical control and in excellent yield, and the transfor-
mations were independent of the alkylation order.
To further probe the effectiveness of CIS-D, we next attempted
the R,R-bisalkylation of hydrazone 54 (Scheme 6). The success of
such a transformation would require that CIS-D be able to
completely reverse the normal preference of LDA for removal of the
significantly more acidic (∼6ꢀ7 pKa units) benzylic (R0) protons
of compound 54 during the first alkylation. This would also be true
of the second alkylation. However, an additional challenge arises in
this case, as there would no longer be a steric preference for R-
deprotonation, since both the R and R0 acidic sites would now be
methylene groups. To test the feasibility of this challenging sequ-
ence of transformations, compound 54 was prepared and then
subjected to the first alkylation reaction using MeI as the electro-
phile. We were very pleased to find that only the R-product was
formed13 (54 f 55), thus indicating that CIS-D was indeed able
to completely overcome the normal pKa bias favoring R0 depro-
tonation. Even moreimpressively, the secondallylation (55 f 56)
also gave only R-alkylation,13 and with excellent (>99:1)15 di-
astereoselectivity. The R,R-bisalkylation was also conducted in the
opposite sense with regard to the order of alkylation (54 f 57 f
58), and was equally effective. Significantly, in contrast to these
results, it has been established that the SAMP hydrazone of ketone
8718
dx.doi.org/10.1021/ja202267k |J. Am. Chem. Soc. 2011, 133, 8714–8720

Journal of the American Chemical Society
ARTICLE
Scheme 6. (a) R,R-Bisalkylation via ACC Hydrazones and
(b) R0-Alkylation via SAMP Hydrazones
Scheme 7. Formal Asymmetric Synthesis of (R)- and (S)-
Stigmolone from a Common Intermediate
Table 5. Hydrolysis of Bisalkylated Hydrazones
At this point we turned our attention to the removal and
recovery of the auxiliary from the bisalkylated products (Table 5).
Thus, each of the R,R-bisalkyated hydrazones prepared in
Scheme 5 was individually treated with p-TsOH H2O in acet-
3
one/H2O, and in each case the desired ketone and 10
were obtained in excellent yield, the latter conveniently set for a
second round of asymmetric R,R-bisalkylation. The er of the
resulting ketones was determined via chiral HPLC, by reference
to a mixture of enantiomers for which baseline separation conditions
had been established. From these experiments it was determined
that no epimerization occurred during auxiliary removal. Unfortu-
nately, we were unable to effect baseline separation of com-
pounds 64 and 65 utilizing the analytical equipment at our
disposal. While we have no reason to expect that anything other
than complete stereochemical integrity was maintained from this
transformation, presently we cannot confirm this.
Finally, in order to demonstrate the utility of the above asym-
metric R,R-bisalkylation method in a synthetic context, a formal
asymmetric total synthesis of both (R)- and (S)-stigmolone was
carried out utilizing the same ACC chiral auxiliary, 26 (Scheme 7).18
Stigmolone is a highly potent pheromone secreted by the
myxobacterium Stigmatella aurantica, which induces formation
of fruiting bodies.19 The synthesis of (þ)-stigmolone began with
the acid-catalyzed condensation of 4-methyl-2-pentanone and
26, which provided ACC hydrazone 71 in 96% yield. This was
then subjected to asymmetric R,R-bisalkylation by treatment
with LDA and prenyl bromide, giving 72, which then underwent
LDA-mediated methylation to produce 73 in both excellent
overall yield and diastereoselectivity (dr > 99:1).15 Auxiliary
hydrazone
dra
ketone yield
entry
hydrazone (R1; R2)
45 (Me; Bn)
ketone
era
(%)
1
2
3
4
5
6
7
8
>99:1
>99:1
>99:1
>99:1
>99:1
>99:1
>99:1
>99:1
62
63
64
65
66
67
68
69
>99:1
>99:1
n.d.b
98
98
95
96
98
97
98
98
46 (Bn; Me)
47 (Et; allyl)
49 (allyl; Et)
n.d.b
50 (4-Br-C6H4CH2; allyl)
24 (allyl; 4-Br-C6H4CH2)
51 (4ꢀ-Br-C6H4CH2; prenyl)
52 (prenyl; 4-Br-C6H4CH2)
>99:1
>99:1
>99:1
>99:1
a Determined by HPLC analysis. b Not determined.17
53 (i.e., 59) undergoes alkylation preferentially at the R0-position
(59 f 60), not the R-position, albeit with low diastereoselectivity.
As such, the present method provides a convenient complemen-
tarystrategy tothe Endersmethod for the asymmetric alkylation of
ketones possessing activating substituents.
8719
dx.doi.org/10.1021/ja202267k |J. Am. Chem. Soc. 2011, 133, 8714–8720

Journal of the American Chemical Society
ARTICLE
cleavage proceeded efficiently to produce ketone 74 (er = 99:1).20
Compound 74 has been shown to undergo cobalt-medi-
ated oxidation to give (S)-stigmolone (75) in 75% yield, without
epimerization occurring.18c Access to the ketone precursor needed
for the preparation of (R)-stigmolone simply entailed switching
the order of alkylation of hydrazone 71, and also proceeded with
excellent yield and diastereoselectivity (dr > 99:1).15 Hydrolysis of
bisalkylated product 76 produced ketone 77 (er = 99:1),20 which
has previously been converted to (R)-stigmolone (78) without
epimerization occurring.18c Utilizing our R,R-bisalkylation meth-
od, ketones 74 and 77 were prepared from ketone 70 in overall
yields of 89% and 87%, respectively, and with excellent enantios-
electivity (er = 99:1). This compares favorably to a prior asymmetric
synthesis of these ketones using the SAMP and RAMP auxiliaries.18c
In that case, ketones 74 and 77 were generated from 70 in 27% and
14% overall yields, respectively, and with enantiomer ratios of
96.5:3.5 and 96:4, respectively.21
(2) See, for example: (a) Abiko, A.; Moriya, O.; Filla, S. A.; Masamune,
S. Angew. Chem., Int. Ed. Engl. 1995, 34, 793–795. (b) Evans, D. A.; Ennis,
M. D.; Mathre, D. J. J. Am. Chem. Soc. 1982, 104, 1737–1739. (c) Evans,
D. A.; Takacs, J. M. Tetrahedron Lett. 1980, 21, 4233–4236. (d) Evans,
D. A.; Urpi, F.; Somers, T. C.; Clark, J. S.; Bilodeau, M. T. J. Am. Chem. Soc.
1990, 112, 8215–8216. (e) Jeong, K. S.; Parris, K.; Ballester, P.; Rebek, J.
Angew. Chem., Int. Ed. Engl. 1990, 29, 555–556. (f) Kawanami, Y.; Ito, Y.;
Kitagawa, T.; Taniguchi, Y.; Katsuki, T.; Yamaguchi, M. Tetrahedron Lett.
1984, 25, 857–860. (g) Larcheveque, M.; Ignatova, E.; Cuvigny, T.
J. Organomet. Chem. 1979, 177, 5–15. (h) Lin, J.; Chan, W. H.; Lee,
A. W. M.; Wong, W. Y. Tetrahedron 1999, 55, 13983–13998. (i) Myers,
A. G.; Yang, B. H.; Chen, H.; McKinstry, L.; Kopecky, D. J.; Gleason, J. L.
J. Am. Chem. Soc. 1997, 119, 6496–6511. (j) Oppolzer, W.; Rodriguez, I.;
Starkemann, C.; Walther, E. Tetrahedron Lett. 1990, 31, 5019–5022.
(k) Sonnet, P. E.; Heath, R. R. J. Org. Chem. 1980, 45, 3137–3139.
(3) The opposite regioselectivity results for hydrazones having
electronically activating groups (e.g., phenyl, ester). See refs 1a, 1b.
(4) Such compounds may be prepared indirectly by the asymmetric
alkylation of a chiral carboxylate species (or synthetic equivalent) and
then stepwise conversion to the ketone. See, for example: (a) Myers,
A. G.; McKinstry, L. J. Org. Chem. 1996, 61, 2428–2440. (b) Palomo, C.;
Oiarbide, M.; Mielgo, A.; Gonzꢀalez, A.; García, J. M.; Landa, C.;
Lecumberri, A.; Linden, A. Org. Lett. 2001, 3, 3249–3252.
(5) Lim, D.; Coltart, D. M. Angew. Chem., Int. Ed. 2008, 47, 5207–
5210.
’ CONCLUSION
We have developed the first general method for the asym-
metric R,R-bisalkylation of ketones having both R- and R0-
protons, via CIS-D of ACC hydrazones. The transformation is
efficient and proceeds with both excellent regio- and stereose-
lectivity. Significantly, CIS-D completely reverses the inherent
preference of LDA to remove the least sterically hindered of two
similarly acidic protons. It also overrides the normal tendency of
LDA to remove the more strongly acidic proton in a substrate
having protons differing significantly in their acidity. Conse-
quently, the regiochemical outcome of this method is the
opposite of that normally obtained for kinetic LDA-mediated
deprotonation of ketones and SAMP/RAMP hydrazones. The
method was successfully utilized for the concise and efficient
asymmetric formal synthesis of both (R)- and (S)-stigmolone.
(6) Krenske, E. H.; Houk, K. N.; Lim, D.; Wengryniuk, S. E.; Coltart,
D. M. J. Org. Chem. 2010, 75, 8578–8584.
(7) Garnsey, M. R.; Lim, D.; Yost, J. M.; Coltart, D. M. Org. Lett.
2010, 12, 5234–5237.
(8) Garnsey, M. R.; Matous, J. A.; Kwiek, J. J.; Coltart, D. M. Bioorg.
Med. Chem. Lett. 2011, 21, 2406–2409.
(9) For lead references on the syn-dianion effect in N-sulfonyl
hydrazones, see: (a) Shapiro, R. H.; Lipton, M. F.; Kolonko, K. J.;
Buswell, R. L.; Capuano, L. A. Tetrahedron Lett. 1975, 1811–1814. (b)
Dauben, W. G.; Rivers, G. T.; Zimmerman, W. T.; Yang, N. C.; Kim, B.;
Yang, J. Tetrahedron Lett. 1976, 2951–2954. (c) Adlington, R. M.;
Barrett, A. G. M. Acc. Chem. Res. 1983, 16, 55–59.
(10) For a review of metal-mediated complex-induced proximity
effects in deprotonation, see: Whisler, M. C.; MacNeil, S.; Snieckus, V.;
Beak, P. Angew. Chem., Int. Ed. 2004, 43, 2206–2225.
(11) For an explanation of the stereochemical course of ACC
alkylations, see refs 5 and 6.
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental procedures and
b
analytical data for all new compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
(12) See the Supporting Information for details.
(13) Determined by 1H NMR.
’ AUTHOR INFORMATION
(14) Isomerization does occur if the hydrazone is exposed to acidic
conditions.
(15) Determined by HPLC. See the Supporting Information for
details.
Corresponding Author
don.coltart@duke.edu
(16) A third regioselective R-alkylation was attempted using 45 but
resulted in the recovery of only nonepimerized starting material,
suggesting that deprotonation had not occurred.
’ ACKNOWLEDGMENT
(17) Baseline resolution of the enantiomers could not be achieved
using the equipment at our disposal. However, based on the other data in
Table 5, we have no reason to suspect that the stereochemical integrity of
64 and 65 was compromised during hydrolysis.
We thank the NSF for a Graduate Fellowship for S.E.W. This
work was supported by NSF (1012287) and NCBC (2008-IDG-
1010).
(18) For prior total syntheses of (R)- and (S)-stigmolone, see: (a)
Mori, K. Eur. J. Org. Chem. 1998, 1479–1489. (b) Mori, K. Eur. J. Org.
Chem. 1998, 2181–2184. (c) Enders, D.; Ridder, A. Synthesis
2000, 1848–1851.
(19) Plaga, W.; Stamm, I.; Schairer, H. U. Proc. Natl. Acad. Sci. U.S.A.
1998, 95, 11263–11267.
(20) Determined by chiral HPLC. See the Supporting Information
for details.
(21) The preparation of ketone 73 according to this method
required HPLC purification of an advanced intermediate from dr =
92.5:7.5 to 99.5:0.5. See ref 18c.
’ REFERENCES
(1) (a) Enders, D.; Eichenauer, H.; Baus, U.; Schubert, H.; Kremer,
K. A. M. Tetrahedron 1984, 40, 1345–1359. (b) Enders, D. In Asymmetric
Synthesis, 1st ed.; Morrison, J. D., Ed.; Academic Press: New York, 1984;
Vol. 3, pp 275ꢀ339. (c) Job, A.; Janeck, C. F.; Bettray, W.; Peters, R.;
Enders, D. Tetrahedron 2002, 58, 2253–2329. (d) Meyers, A. I.;
Williams, D. R.; Druelinger, M. J. Am. Chem. Soc. 1976, 98, 3032–
3033. (e) Meyers, A. I.; Williams, D. R. J. Org. Chem. 1978, 43, 3245–
3247. (f) Meyers, A. I.; Williams, D. R.; Erickson, G. W.; White, S.;
Druelinger, M. J. Am. Chem. Soc. 1981, 103, 3081–3087. (g) Hashimoto,
S.; Koga, K. Tetrahedron Lett. 1978, 573–576. (h) Hashimoto, S.; Koga,
K. Chem. Pharm. Bull. 1979, 27, 2760–2766.
8720
dx.doi.org/10.1021/ja202267k |J. Am. Chem. Soc. 2011, 133, 8714–8720