Dialkylaluminum N, O -dimethylhydroxylamine complex as a reagent to mask reactive carbonyl groups in situ from nucleophiles

Article abstract of DOI:10.1021/ol102495v

Aluminum complexes of N,O-dimethylhydroxylamine are effective reagents to mask carbonyl groups in situ from nucleophilic addition by organolithiums, Grignard reagents, and borohydrides. The utility of this process by selectively adding nucleophiles into carbonyl groups on a variety of structures as well as distinguishing between carbonyl groups on a sensitive natural product is demonstrated. ¹H NMR analysis supports the in situ masking of the more reactive carbonyl group.

Full text of DOI:10.1021/ol102495v

ORGANIC
LETTERS
2010
Vol. 12, No. 23
5588-5591
Dialkylaluminum N,O-Dimethylhydroxylamine
Complex as a Reagent to Mask Reactive
Carbonyl Groups in Situ from Nucleophiles
Francis J. Barrios,† Xuechao Zhang,† and David A. Colby*,†,‡
Department of Chemistry and Department of Medicinal Chemistry and Molecular
Pharmacology, Purdue UniVersity, West Lafayette, Indiana 47907, United States
dcolby@purdue.edu
Received October 15, 2010
ABSTRACT
Aluminum complexes of N,O-dimethylhydroxylamine are effective reagents to mask carbonyl groups in situ from nucleophilic addition by
organolithiums, Grignard reagents, and borohydrides. The utility of this process by selectively adding nucleophiles into carbonyl groups on
1
a variety of structures as well as distinguishing between carbonyl groups on a sensitive natural product is demonstrated. H NMR analysis
supports the in situ masking of the more reactive carbonyl group.
The addition of organometallic nucleophiles to Weinreb
amides is a powerful strategy for the synthesis of aldehydes
or ketones.1-3 The key tetrahedral intermediate that is formed
during the process is exceptionally stable and guards the
trapped carbonyl precursor from nucleophilic attack. Indeed,
the masked carbonyl group can be carried through additional
synthetic manipulations prior to its unveiling.4,5 Cossy and
co-workers have exploited the stability of the tetrahedral
intermediate using a Birch reduction following nucleophilic
addition to a Weinreb amide4 and extended this strategy to
the synthesis of zoapatanol.5 During the course of our work
to produce semisynthetic derivatives of biologically active
natural products, the direct addition of a nucleophile to a
less reactive carbonyl group (i.e., lactone) in the presence
of a more reactive carbonyl group (i.e., ketone) was
necessary. Common strategies to achieve this goal require
the use of protecting groups and/or reduction/oxidation
sequences. However, neither of these multistep synthetic
strategies were attractive to us due to poor step economy.6
Herein we report a general strategy to mask a reactive
carbonyl group in the presence of a nucleophile using
complexes of dialkylaluminum and N,O-dimethylhydroxyl-
amine.
† Department of Chemistry.
‡ Department of Medicinal Chemistry and Molecular Pharmacology.
(1) Nahm, S.; Weinreb, S. M. Tetrahedron Lett. 1981, 22, 3815–3818
.
(2) Recent examples in synthesis: (a) Malathong, V.; Rychnovsky, S. D.
Org. Lett. 2009, 11, 4220–4223. (b) Clive, D. L. J.; Pham, M. P. J. Org.
Chem. 2009, 74, 1685–1690. (c) Kokotos, C. G.; Baskakis, C.; Kokotos,
G. J. Org. Chem. 2008, 73, 8623–8626. (d) Yang, S.-B.; Gan, F.-F.; Chen,
G.-J.; Xu, P.-F. Synlett 2008, 2532–2534. (e) Pippel, D. J.; Mapes, C. M.;
Mani, N. S. J. Org. Chem. 2007, 72, 5828–5831. (f) Spletstoser, J. T.; White,
J. M.; Tunoori, A. R.; Georg, G. I. J. Am. Chem. Soc. 2007, 129, 3408–
3419. (g) Murphy, J. A.; Commeureuc, A. G. J.; Snaddon, T. N.; McGuire,
T. M.; Khan, T. A.; Hisler, K.; Dewis, M. L.; Carling, R. Org. Lett. 2005,
7, 1427–1429
.
(3) Recent preparations of Weinreb amides: (a) Davis, F. A.; Theddu,
N. J. Org. Chem. 2010, 75, 3814–3820. (b) Krishnamoorthy, R.; Lam, S. Q.;
Manley, C. M.; Herr, R. J. J. Org. Chem. 2010, 75, 1251–1258. (c) Ghosh,
A. K.; Banerjee, S.; Sinha, S.; Kang, S. B.; Zajc, B. J. Org. Chem. 2009,
(4) Taillier, C.; Bellosta, V.; Meyer, C.; Cossy, J. Org. Lett. 2004, 6,
2145–2147
.
(5) Taillier, C.; Gille, B.; Bellosta, V.; Cossy, J. J. Org. Chem. 2005,
70, 2097–2108
74, 3689–3697
.
.
10.1021/ol102495v  2010 American Chemical Society
Published on Web 11/01/2010

The trapping of carbonyl groups with lithium amides has
been advocated by Comins,7 but the major drawback is that
lithium amides are strong bases. So deprotonation is a major
side reaction, if acidic protons are present. Hoffman and co-
workers have used the lithiated amide derivative of N,O-
dimethylhydroxylamine to mask an aldehyde from reduc-
tion.8 This process was successful on a single substrate but
failed on more complex systems due to unavoidable depro-
tonation at other sites.9 Lithium N,O-dimethylhydroxylamide
was subsequently used to protect aromatic aldehydes where
R-deprotonation is not an issue.10 To our knowledge, no other
general strategies to mask reactive carbonyl groups to
nucleophiles as tetrahedral intermediates with N,O-dimeth-
ylhydroxylamines are known.11 We have found that alumi-
num complexes of N,O-dimethylhydroxylamine are powerful
and versatile reagents to mask reactive carbonyl groups in
the presence of nucleophiles without the drawback of high
basicity associated with lithium amides (Scheme 1).
Table 1. Optimization of in Situ Carbonyl Group Masking
yield (%)a
entry
additives
EtMgBr
1
2
3
4
1
2
3
none
none
DIBALH,
1 equiv
6 equiv
5 equiv
10
0
0
49
0
0
0
82
0
0
0
47
HN(OMe)Me·HCl
DIBALH,
4
4 equiv
0
0
0
67
HN(OMe)Me·HCl
then i-PrMgCl
(1 equiv)
a Isolated yields.
Scheme 1. Strategies to Trap Carbonyl Groups in Situ
Thus, we turned our attention to i-PrMgCl as a base,12 and
this strategy not only spared an additional equivalent of
nucleophile but also increased the isolated yield of 4 to 67%
(entry 4).
To determine the scope of this strategy, we explored
compounds 1 and 5-8 as substrates because each has two
different types of carbonyl groups (Table 2). Indeed, using
our optimized protocol to mask the more reactive carbonyl
group followed by addition of an organolithium, a Grignard
reagent, or borohydride as a nucleophile, selective addition
to the less reactive carbonyl group was observed. Specifically,
for substrate 1, the double addition of n-BuLi, MeMgBr, or
EtMgBr occurrs preferentially at the ester group after
pretreatment with the dialkylaluminum complex. In the cases
of the Grignard additions, the use of THF lead to lower yields
due to unwanted carbonyl reduction; however, this competing
process was eliminated using Et2O as the primary solvent.13
The synthetic utility of Grignard reagents continues to grow
due to recent advances in the preparation of magnesium
compounds.12,14 Super hydride proved to be the optimum
reagent to promote reduction of ester 1, and the aldehyde
was returned after aqueous workup. This strategy was
subsequently applied to substrates 5-6 bearing a ketone and
ester and substrates 7-8 with an aldehyde and ketone.
Synthetically useful yields were rountinely isolated, and in
each case, the more reactive carbonyl groups were unscathed.
This strategy using aluminum complexes represents a
significant advance because it is fully compatible with
ketones and other carbonyl groups with acidic R-protons,
unlike previous protocols.8-10 Another application is the one-
step preparation of 5-hydroxy-5-methylhexan-2-one (12)
To determine the feasibility of this strategy, we selected
methyl 4-formylbenzoate 1 as a substrate, because it has two
different types of carbonyl groups (Table 1). The addition
of 1 equiv of EtMgBr occurs at the aldehyde to give adduct
2, and the addition of excess Grignard reagent consumes both
carbonyl groups to provide the diol 3. We hypothesized that
using a combination of DIBALH and HN(OMe)Me·HCl, the
more reactive carbonyl group would be masked in situ as
an aminal and protected from the subsequent nucleophilic
attack. Accordingly, upon pretreatment of 1 with a combina-
tion of DIBALH and HN(OMe)Me·HCl, the major product
isolated after adding EtMgBr was the aldehyde 4 resulting
from selective double addition to the less reactive carbonyl
group (i.e., the ester). Further optimizations lead to the
conclusion that the first equivalent of Grignard reagent
removes the last proton attached to the amine (the first acidic
proton of the amine hydrochloride is consumed by DIBALH)
and fully stabilizes the aminal from nucleophilic addition.
(6) (a) Wender, P. A.; Miller, B. L. Nature 2009, 460, 197–201. (b)
Young, I. S.; Baran, P. S. Nature Chem. 2009, 1, 193–205.
(7) Comins, D. L. Synlett 1992, 615–625.
(8) Hoffmann, R. W.; Munster, I. Tetrahedron Lett. 1995, 36, 1431–
1434.
(9) Kruger, J.; Hoffmann, R. W. J. Am. Chem. Soc. 1997, 119, 7499–
7504.
(12) Krasovskiy, A.; Krasovskaya, V.; Knochel, P. Angew. Chem., Int.
Ed. 2006, 45, 2958–2961.
(10) Roschangar, F.; Brown, J. C.; Cooley, B. E.; Sharp, M. J.; Matsuoka,
R. T. Tetrahedron 2002, 58, 1657–1666.
(13) For an example of carbonyl reduction with a Grignard reagent, see:
Hoye, T. R.; Aspaas, A. W.; Eklov, B. M.; Ryba, T. D. Org. Lett. 2005, 7,
2205–2208.
(11) Titanium tetrakis(dialkylamides) are known to protect carbonyl
groups in situ. Reetz, M. T.; Wenderoth, B.; Peter, R. J. Chem. Soc., Chem.
Commun. 1983, 8, 406–408.
(14) Fleury, L. M.; Ashfeld, B. L. Tetrahedron Lett. 2010, 51, 2427–
2430
.
Org. Lett., Vol. 12, No. 23, 2010
5589

To demonstrate a major application of this method, we
turned our attention to natural products, as our research
interests are the selective modification of these compounds
for structure-activity investigations.16 Typically, natural
products have a diverse array of functional groups, and the
significance of selectively modifying them has been recently
demonstrated by others.17,18 The sesquiterpene lactone,
R-santonin (17), is a key scaffold that has been used on
numerous occasions to build synthetic derivatives16,19 and
access other natural products.20,21 The molecule has two
carbonyl groups: (1) a highly reactive cyclic ketone and (2)
a less reactive lactone. Our initial synthetic strategy was to
protect the ketone as an ether and then modify the lactone
(Scheme 2). Thus, DIBALH readily reduced the ketone to
Table 2. Selective Additions to Carbonyl Groups
Scheme 2. Nucleophilic Additions to R-Santonin (17)
the alcohol 18 without affecting the lactone.19 Unfortunaetly,
the next steps of this strategy proved to be quite difficult
due to the facile carbon-oxygen bond cleavage that con-
comitantly promotes the formation of the aromatic compound
19 under a variety of conditions.22 Because of the sensitive
nature of R-santonin (17), a stepwise approach ultimately
proved intractable, and we turned our attention to the in situ
protection strategy. We treated 17 with DIBALH/
HN(OMe)Me·HCl and MeLi, and the lactol 20 was produced
in 69% yield (see Scheme 2). To contrast this result, the
addition of only MeLi to 17 provided the unstable tertiary
a Isolated yields. b i-PrMgCl was not used as a base prior to the addition
of the nucleophile.
(16) Han, C.; Barrios, F. J.; Riofski, M. V.; Colby, D. A. J. Org. Chem.
2009, 74, 7176–7179.
from ethyl levulinate (5) in 65% yield because 12 is
routinely used in synthesis, but existing syntheses require
multiple steps.15 The nucleophiles MeLi, n-BuLi, MeMg-
Br, EtMgBr, and LiBEt3H readily participate in the
reaction to provide a high level of preferential selectivity
for the less reactive carbonyl group when added after
DIBALH and HN(OMe)Me·HCl.
(17) Lewis, C. A.; Longcore, K. E.; Miller, S. J.; Wender, P. A. J. Nat.
Prod. 2009, 72, 1864–1869
(18) Erb, J.; Alden-Danforth, E.; Kopf, N.; Scerba, M. T.; Lectka, T. J.
Org. Chem. 2010, 75, 969–971
.
.
(19) Arantes, F. F. P.; Barbosa, L. C. A.; Alvarenga, E. S.; Demuner,
A. J.; Bezerra, D. P.; Ferreira, J. R. O.; Costa-Lotufo, L. V.; Pessoa, C.;
Moraes, M. O. Eur. J. Med. Chem. 2009, 44, 3739–3745
(20) Zhang, W.; Luo, S.; Fiang, F.; Chen, Q.; Hu, H.; Jia, X.; Zhai, H.
J. Am. Chem. Soc. 2005, 127, 18–19
(21) Suzuki, T.; Miyashita, M. Heterocycles 2001, 54, 805–870
.
.
.
(15) (a) Masuno, H.; Yamamoto, K.; Wang, X.; Choi, M.; Ooizumi,
H.; Shinki, T.; Yamada, S. J. Med. Chem. 2002, 45, 1825–1834. (b) Shono,
T.; Kashimura, S.; Mori, Y.; Hayashi, T.; Soejima, T.; Yamaguchi, Y. J.
Org. Chem. 1989, 54, 6001–6003.
(22) (a) T. Kawamata, T.; Nagashima, K.; Nakai, R.; Tsuji, T. Synth.
Commun. 1996, 26, 139–148. (b) Banerjee, A. K.; Vera, W. Synth. Commun.
1996, 26, 4641–4646. (c) Huffman, J. W. J. Org. Chem. 1987, 52, 2901–
2904.
5590
Org. Lett., Vol. 12, No. 23, 2010

alcohol which rapidly converted to the triene 21 during
purification (77% yield). These data illustrate the effectiveness
of the in situ trapping of a reactive carbonyl group with a
dialkylaluminum complex because the carbonyl groups can be
accessed selectively. This strategy not only avoids multistep
protection/deprotection sequences but also is superior in the case
of sensitive and complex natural products, such as 17, where
multistep sequences can not be used.
parisons of the spectra of substrate 1 with treated 1 show
that the proton of the aldehyde at 9.34 ppm is no longer
present when the dialkylaluminum reagent is added. Also,
the protons on the aromatic ring have shifted, likely due to
the absence of the carbonyl group of the aldehyde after
addition of the reagent.
In conclusion, we have demonstrated a general strategy
to trap reactive carbonyl groups with aluminum complexes
of N,O-dimethylhydroxylamine. This protocol is a significant
advancement over previously described lithium amides
because competing deprotonation with the strongly basic
lithium amides is limited. We have demonstrated the scope
of this process by selectively adding nucleophiles into
carbonyl groups on a variety of structures. We have applied
our strategy to distinguish selectively between carbonyl
groups on a sensitive natural product in which multistep
strategies fail. Additionally, we have provided key NMR data
to support the in situ masking of the more reactive carbonyl
group. Overall, this process is a powerful alternative to
traditional multistep reduction/oxidation processes and pro-
tection/deprotection strategies to distinguish between reactive
carbonyl groups on natural products and other complex
molecules.
To gain further insight into the process, we treated
substrate 1 with DIBALH/HN(OMe)Me·HCl in THF and
1
acquired H NMR data at rt (Figure 1). Because we were
Acknowledgment. We are grateful to Purdue University
for funding. X.Z. is the recipient of a Dean’s Summer
Fellowship.
Figure 1.
Comparison of 1H NMR data of compound 1 at 500 MHz.
(A) Compound 1 in THF. (B) Compound 1 in THF after treatment
with DIBALH/HN(OMe)Me·HCl.
Supporting Information Available: Full experimental
details, characterization data, and H and 13C NMR spectra
for all new compounds. This material is available free of
charge via the Internet at http://pubs.acs.org.
1
only interested in the protons downfield, these data were
quickly gathered using the “no-D” NMR method.23 Com-
(23) For a leading reference, see Hoye, T. R.; Eklov, B. M.; Voloshin,
M. Org. Lett. 2004, 6, 2567–2570.
OL102495V
Org. Lett., Vol. 12, No. 23, 2010
5591