Asymmetric homoenolate additions to acyl phosphonates through rational design of a tailored N-heterocyclic carbene catalyst

Article abstract of DOI:10.1021/ja410932t

A highly selective NHC-catalyzed synthesis of γ-butyrolactones from the fusion of enals and α-ketophosphonates has been developed. Computational modeling of competing transition states guided a rational design strategy to achieve enhanced levels of enanti

Full text of DOI:10.1021/ja410932t

Communication
pubs.acs.org/JACS
Asymmetric Homoenolate Additions to Acyl Phosphonates through
Rational Design of a Tailored N‑Heterocyclic Carbene Catalyst
Ki Po Jang,† Gerri E. Hutson,† Ryne C. Johnston,‡ Elizabeth O. McCusker,† Paul H.-Y. Cheong,*,‡
and Karl A. Scheidt*,†
†Department of Chemistry, Center for Molecular Innovation and Drug Discovery, Chemistry of Life Processes Institute,
Northwestern University, Silverman Hall, Evanston, Illinois 60208, United States
‡Department of Chemistry, Oregon State University, 153 Gilbert Hall, Corvallis, Oregon 97331, United States
S
* Supporting Information
ABSTRACT: A highly selective NHC-catalyzed synthesis
of γ-butyrolactones from the fusion of enals and α-
ketophosphonates has been developed. Computational
modeling of competing transition states guided a rational
design strategy to achieve enhanced levels of enantio-
selectivity with a new tailored C1-symmetric biaryl-
saturated imidazolium-derived NHC catalyst.
he ability to forge complex hetero- and carbocyclic systems
Tconvergently with high levels of selectivity remains a
challenge in organic synthesis. Over the past two decades, N-
heterocyclic carbenes (NHCs) have been explored intensely for
the synthesis of various structural manifolds through Umpolung
reactivity.1 Since the seminal reports of NHC-homoenolate
reactions, there have been significant advancements in this area,
with a primary focus on additions to various activated π
systems.2,3 While aldehydes and aldimines are productive
substrates in a variety of reactions to produce γ-lactones and
Figure 1. NHC annulations of acyl phosphonates.
lactams, respectively,3a−c,4 more sterically hindered, less reactive
ketones and imines impose severe limitations on single NHC
nates bearing α-hydroxy10 and α-amino groups11 has received
significant attention.
Here we report the addition of α,β-unsaturated aldehydes to α-
ketophosphonates under carbene catalysis to afford enantio-
catalyst systems.5,6 To address this challenge, new strategies and/
or new catalyst design approaches are needed. One strategy is
cooperative catalysis with NHCs integrating Lewis and Brønsted
acids to enhance reactivity, modulate stereocontrol, and access
enriched γ-butyrolactones with a previously unreported core
new reactivity with previously inactive electrophiles.4c,7 How-
structure (Figure 1, X = P(O)(OR)2). The discovery that
ever, there are few examples involving the synthesis of γ-
butyrolactones with high levels of enantioselectivity.6,7b These
motifs are highly attractive given the large number of bioactive
molecules and natural products with this key architectural feature
(Figure 1, X = H, C, or heteroatom). As a complementary strategy,
new advances in rational N-heterocyclic carbene structure/reactivity
development could provide new solutions for NHC annulations of π
electrophiles.
Acyl phosphonates are potential electrophiles for an NHC-
catalyzed [3+2] annulation process, but are particularly
challenging given the steric environment around the target
carbonyl group. These phosphonates have been employed in a
wide range of enantioselective processes.8 Many methods have
been developed to incorporate phosphonate groups due to the
prevalence of the phosphinic acid group in naturally occurring
and pharmaceutical compounds with antiviral, antibacterial, and
anticancer properties.9 In particular, the synthesis of phospho-
standard chiral NHC catalysts are ineffective for this trans-
formation resulted in the design of several new C1-symmetric
biaryl-saturated imidazolium catalysts. This scaffold was
introduced by Hoveyda in an enantioselective, metal-free C−B
bond-forming reaction.12 Computationally guided rational
catalyst design was employed to improve selectivity for this
new formal [3+2] annulation process. This strategy is under-
utilized in the field of NHC catalysis, but several groups, most
prominently Sigman et al., have demonstrated the use of
correlative methods to dictate improved catalyst scaffolds for
epoxidation, hetero-Diels−Alder, and other catalytic asymmetric
reactions.13 These efforts have resulted in enhanced catalyst
performance and relate catalyst structure to the observed
enantioselectivity.
Received: November 1, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja410932t | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Table 1. Optimization of Reaction Conditions
a
b
NMR yield with internal standard. Diastereomeric ratio determined
c
1
by H NMR spectroscopy. Enantiomeric ratio determined by HPLC
analysis.
Figure 2. Transition structures with catalyst D. The first Re/Si notation
refers to the homoenol face, and the second Re/Si refers to the
electrophile face. Et groups abbreviated to Me for computational
efficiency. Cyan region marks stabilizing nonclassical hydrogen-bonding
interactions in the major TS, while pink marks those found in the minor.
Green lines indicate electrostatic interactions. Distances are in Å and
energies in kcal/mol.
We began our studies by combining α-ketophosphonate 1
with cinnamaldehyde in the presence of imidazolium precatalyst
A and 20 mol% 1,8-diazabicyclo[5.4.9]undec-7-ene (DBU, Table
1, entry 1). Under these conditions, the desired formal [3+2]
annulation product was formed in 35% yield with good selectivity
for the trans diastereomer. To improve the efficiency, various
other NHC precatalysts were examined for this transformation:
the use of saturated imidazolium B resulted in a decrease in
diastereoselectivity but a significant increase in yield (entry 2).
Notably, while saturated imidazoliums are frequently employed
as ligands for transition metals,14 there are limited reports on the use
of these NHCs as organocatalysts.7d
Table 2. Catalyst Optimization
With this backdrop, we turned to rendering this reaction
enantioselective. The use of 1,2-aminoindanol-derived triazo-
lium catalyst C4b resulted in poor conversion to the lactone
product with low levels of enantioselectivity (67:33 er, entry 3).
Based on the success of achiral saturated imidazolium B, we
explored the C1-symmetric biaryl-saturated imidazolium pre-
catalyst D.12 Under the previously explored conditions,
precatalyst D afforded the product in excellent yield and good
diastereo- and enantioselectivity (94%, 85:15 er, entry 4). A
further increase in enantioselectivity was observed by switching
the base from DBU to 7-methyl-1,5,7-triazabicyclo[4.4.0]dec-5-
ene (MTBD), affording the lactone in 79% yield and 90:10 er
(entry 5).
At this stage, a new direction was required to achieve higher
levels of selectivity. Faced with the typical arduous empirical
approach of evaluating many new analogues of D, we initiated a
computational investigation of the stereodetermining homoaldol
step with density functional theory, B3LYP/6-31G*.15 The
transition structures (TSs) that correspond to the formation of
the major (S,S) and minor (R,R) enantiomer products are shown
in Figure 2. In both cases, nucleophilic attack of the homoenol to
the acyl phosphonate carbonyl is concomitant with deprotona-
tion of the enol proton by the forming alkoxide.16 Stereo-
selectivity arises from differential stabilization of the phosphonyl
oxygen by the catalyst aryl protons through nonclassical hydrogen
bonds (NCHBs).17 The preorganization and anion stabilization
present in these TSs are reminiscent of the oxyanion hole
stabilization found in enzymes, where an array of backbone
amide protons provide stabilization of the developing alkoxide.18
a
b
NMR yield with an internal standard. Diastereomeric ratio was
c
determined by 1H NMR spectroscopy. Enantiomeric ratio was
determined by HPLC analysis.
It was anticipated that selective destabilization of the minor TS
could be achieved because the stabilization sites differ between
the major and minor enantiomers (Table 2). In the minor TS, the
catalyst NCHB sites are only on the terminal biphenyl meta
positions (shown in pink). This is in contrast to the major TS,
where there are multiple stabilizing NCHB interactions, located
at the ortho positions of the catalyst backbone phenyl and the
internal N-biphenyl (shown in cyan). The increased number of
strong NCHB interactions is responsible for the computed 1.7
kcal/mol selectivity (95:5 er), which compares favorably with the
experimental selectivity of ∼1.0 kcal/mol (90:10 er). While both
TSs are stabilized by the mesityl methyl C−H, these protons are
inferior NCHB donors compared to aryl C−H’s.17 Since the
NCHB stabilization motif was clearly different between the
B
dx.doi.org/10.1021/ja410932t | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
a
major and minor TSs, we envisioned that the installation of
methyl groups at the terminal biphenyl meta positions would
disrupt the phosphonyl stabilization of the minor TS but leave
the major TS unchanged, thus increasing enantioselectivity.
Based on the computational prediction, catalyst E was
employed in the annulation reaction. Under the previously
optimized conditions with the more sterically demanding 3,5-
dimethylphenyl substituent we observed an increase in enantio-
selectivity to 92:8 er (1.5 kcal/mol, Table 2, entry 2). This
correlates well with the computationally predicted er of 99:1 (3.0
kcal/mol). Extension of the catalyst to include ethyl groups at the
3- and 5-positions of the aryl ring was explored to determine if
the enantioselectivity could be further enhanced. We were
pleased to find that this simple catalyst modification afforded the
product in 80% yield and 94:6 er (entry 3).
Table 3. Lactone Substrate Scope
We surveyed several α,β-unsaturated aldehydes in the reaction
with α-ketophosphonate 1 with the optimized reaction
conditions (Table 3). Both electron-withdrawing and -donating
substituents were tolerated, providing the γ-butyrolactones in
good to excellent yield and enantioselectivity (3−7, >91:9 er). In
addition, enals bearing a pyridine and N-Me indole at the β-
position were also accommodated. 3-Pyridyl-substituted lactone
11 was formed in slightly diminished yield but with good
enantioselectivity. Likewise, the indole-substituted lactones were
both produced in >65% yield, with 2-indoyl substrate 12
generated with slightly higher enantioselectivity. Variation of the
α-ketophosphonate was also explored. We found that electron-
donating groups performed well, with lactones 14 and 15 formed
in good yield and 93:7 er. Electron-withdrawing substituents
afforded the products in excellent yield and enantioselectivity but
with moderate diastereoselectivity (16−18). Other aromatic
groups, such as naphthyl, also performed well in the reaction.
Currently, alkyl-substituted substrates (either aldehyde or acyl
phosphonate) do not provide practical levels of efficiency.
Based on our calculations, the proposed mechanism for
formation of the lactone products involves initial addition of the
catalyst to the aldehyde and subsequent formal 1,2-H migration
to generate the extended Breslow intermediate (I, Scheme 1).
Based on computational modeling, H-bonding between the
ketone of the acyl phosphonate and the enol results in an
organized transition state. Following C−C bond formation to
generate enol II, tautomerization occurs, giving rise to acyl
a
Scheme 1. Proposed Reaction Mechanism
Yields are of isolated product after column chromatography.
Diastereomeric ratio determined by 1H NMR spectroscopy.
Enantiomeric excess determined by chiral HPLC.
azolium III, which then undergoes O-acylation to afford lactone
2 and release the NHC catalyst.
The lactone products can be converted to enantioenriched 1,4-
ketoesters in a single, mild operation with almost complete
transfer of chirality. Treatment of lactone 2 (93:7 er) with
potassium phosphate dibasic afforded ketoester 21 in 68% yield
and 92:8 er through initial opening of the lactone and a
subsequent retro-phospho aldol reaction (Scheme 2). Related
Scheme 2. Transformation to Stetter Products
C
dx.doi.org/10.1021/ja410932t | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
K. A. J. Am. Chem. Soc. 2007, 129, 5334. (g) Phillips, E. M.; Reynolds, T.
E.; Scheidt, K. A. J. Am. Chem. Soc. 2008, 130, 2416. (h) Nair, V.; Babu,
B. P.; Vellalath, S.; Varghese, V.; Raveendran, A. E.; Suresh, E. Org. Lett.
2009, 11, 2507. (i) Izquierdo, J.; Orue, A.; Scheidt, K. A. J. Am. Chem.
Soc. 2013, 135, 10634.
(4) (a) Nair, V.; Varghese, V.; Babu, B. P.; Sinu, C. R.; Suresh, E. Org.
Biomol. Chem. 2010, 8, 761. (b) Raup, D. E. A.; Cardinal-David, B.;
Holte, D.; Scheidt, K. A. Nat. Chem. 2010, 2, 766. (c) Zhao, X. D.;
DiRocco, D. A.; Rovis, T. J. Am. Chem. Soc. 2011, 133, 12466.
(5) For NHC-homoenolate annulations of ketones and ketimines, see:
(a) Burstein, C.; Tschan, S.; Xie, X. L.; Glorius, F. Synthesis 2006, 2418.
(b) Nair, V.; Vellalath, S.; Poonoth, M.; Mohan, R.; Suresh, E. Org. Lett.
2006, 8, 507. (c) Li, Y.; Zhao, Z. A.; He, H.; You, S. L. Adv. Synth. Catal.
2008, 350, 1885. (d) Rommel, M.; Fukuzumi, T.; Bode, J. W. J. Am.
Chem. Soc. 2008, 130, 17266.
(6) (a) Lv, H.; Tiwari, B.; Mo, J. M.; Xing, C.; Chi, Y. R. Org. Lett. 2012,
14, 5412. (b) Sun, L. H.; Shen, L. T.; Ye, S. Chem. Commun. 2011, 47,
10136.
(7) (a) Cohen, D. T.; Scheidt, K. A. Chem. Sci. 2012, 3, 53. (b) Dugal-
Tessier, J.; O’Bryan, E. A.; Schroeder, T. B. H.; Cohen, D. T.; Scheidt, K.
A. Angew. Chem., Int. Ed. 2012, 51, 4963. (c) Mo, J. M.; Chen, X. K.; Chi,
Y. R. J. Am. Chem. Soc. 2012, 134, 8810. (d) A recent search (Reaxys) for
the general γ-lactone structure identified >6000 bioactive molecules and
>10 000 isolated natural products.
(8) For selected examples, see: (a) Evans, D. A.; Johnson, J. S. J. Am.
Chem. Soc. 1998, 120, 4895. (b) Evans, D. A.; Scheidt, K. A.; Fandrick, K.
R.; Lam, H. W.; Wu, J. J. Am. Chem. Soc. 2003, 125, 10780. (c) Takenaka,
N.; Abell, J. P.; Yamamoto, H. J. Am. Chem. Soc. 2007, 129, 742.
(9) (a) Collinsova, M.; Jiracek, J. Curr. Med. Chem. 2000, 7, 629.
(b) Kafarski, P.; Lejczak, B. Curr. Med. Chem.: Anti-Cancer Agents 2001,
1, 301. (c) Mucha, A.; Kafarski, P.; Berlicki, L. J. Med. Chem. 2011, 54,
5955.
(10) (a) Kolodiazhny, O. I. Tetrahedron: Asymmetry 2005, 16, 3295.
(b) Samanta, S.; Zhao, C. G. J. Am. Chem. Soc. 2006, 128, 7442.
(c) Gondi, V. B.; Hagihara, K.; Rawal, V. H. Angew. Chem., Int. Ed. 2009,
48, 776. (d) Hou, G.; Yu, J.; Yu, C.; Wu, G.; Miao, Z. Adv. Synth. Catal.
2013, 355, 589. (e) Corbett, M. T.; Johnson, J. S. J. Am. Chem. Soc. 2013,
135, 594.
(11) (a) Joly, G. D.; Jacobsen, E. N. J. Am. Chem. Soc. 2004, 126, 4102.
(b) Abell, J. P.; Yamamoto, H. J. Am. Chem. Soc. 2008, 130, 10521.
(c) Vicario, J.; Ezpeleta, J. M.; Palacios, F. Adv. Synth. Catal. 2012, 354,
2641.
(12) Lee, K.-s.; Zhugralin, A. R.; Hoveyda, A. H. J. Am. Chem. Soc. 2009,
131, 7253.
1,4-carbonyl systems are often prepared through a Stetter
reaction,19,20 but this particular class is difficult to access through
traditional methods due to the diminished reactivity of the ester
and the resulting acidity of the α-aryl ketone product.21 This
NHC-catalyzed annulation provides a potential new strategy for
accessing this challenging 1,4-carbonyl manifold which in theory
can be accessed through a Stetter reaction. However, to our
knowledge, no general, enantioselective intermolecular Stetter
reaction of aromatic aldehydes to α,β-unsaturated esters has been
reported that provides high levels of enantiocontrol at the β-
stereocenter.21
In conclusion, a highly selective NHC-catalyzed formal [3+2]
annulation of α,β-unsaturated aldehydes with acyl phosphonates
has been developed. The ineffectiveness of known NHC catalysts
for this transformation resulted in the creation of new C1-
symmetric biaryl-saturated imidazolium catalysts. Computation-
ally guided rational catalyst design resulted in enhanced
enantioselectivity, allowing for the synthesis of various γ-
butyrolactones with excellent selectivity. These saturated
imidazoliums provide a new catalyst scaffold for investigating
other NHC-catalyzed transformations. In addition, this new
enantioselective platform provides a distinct approach to 1,4-
carbonyl compounds that are difficult to access through
traditional methods, where various substituents can be
incorporated through the appropriate choice of the acyl
phosphonate and aldehyde starting materials. Continuing
investigations involving the use of these chiral saturated
imidazolium catalysts in NHC-catalyzed transformations and
the integration of computational methods to enhance selectivity
via specific catalyst tailoring are ongoing.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures and spectral data for new compounds,
geometries, and energies of all structures discussed. This material
is available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
scheidt@northwestern.edu
■
(13) Harper, K. C.; Sigman, M. S. J. Org. Chem. 2013, 78, 2813.
(14) Diez-Gonzalez, S.; Marion, N.; Nolan, S. P. Chem. Rev. 2009, 109,
3612.
(15) Frisch, M. J.; et al. Gaussian 09; Gaussian, Inc.: Wallingford, CT,
2009.
(16) The concerted, asynchronous proton transfer was verified with an
intrinsic reaction coordinate calculation.
(17) Johnston, R. C.; Cheong, P. H. Org. Biomol. Chem. 2013, 11, 5057.
(18) Bryan, P.; Pantoliano, M. W.; Quill, S. G.; Hsiao, H. Y.; Poulos, T.
Proc. Natl. Acad. Sci. U.S.A. 1986, 83, 3743.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Financial support was generously provided by the NIH (RO1-
GM073072) and Oregon State University (Stone Scholar
Funds). We thank Dr. Paul Siu (Northwestern University) for
assistance with X-ray crystallography.
(19) Stetter, H. Angew. Chem., Int. Ed. 1976, 15, 639.
(20) (a) Enders, D.; Breuer, K.; Runsink, J.; Teles, J. H. Helv. Chim.
Acta 1996, 79, 1899. (b) de Alaniz, J. R.; Rovis, T. Synlett 2009, 1189.
(21) For recent examples of asymmetric Stetter reactions utilizing
NHC catalysis, see: (a) DiRocco, D. A.; Rovis, T. J. Am. Chem. Soc. 2011,
REFERENCES
■
(1) For leading reviews, see: (a) Enders, D.; Niemeier, O.; Henseler, A.
Chem. Rev. 2007, 107, 5606. (b) Phillips, E. M.; Chan, A.; Scheidt, K. A.
Aldrichimica Acta 2009, 43, 55. (c) Biju, A. T.; Kuhl, N.; Glorius, F. Acc.
Chem. Res. 2011, 44, 1182. (d) Grossmann, A.; Enders, D. Angew. Chem.,
Int. Ed. 2012, 51, 314. (e) Bugaut, X.; Glorius, F. Chem. Soc. Rev. 2012,
41, 3511.
́
133, 10402. (b) Sanchez-Larios, E.; Thai, K.; Bilodeau, F.; Gravel, M.
Org. Lett. 2011, 13, 4942. (c) Fang, X. Q.; Chen, X. K.; Lv, H.; Chi, Y. R.
Angew. Chem., Int. Ed. 2011, 50, 11782. (d) DiRocco, D. A.; Noey, E. L.;
Houk, K. N.; Rovis, T. Angew. Chem., Int. Ed. 2012, 51, 2391. (e) For a
recent example of an intermolecular enantioselective Stetter reaction
with a β-alkyl substitution at 59% yield and 90:10 er, see: Wurz, N. E.;
Daniliuc, C. G.; Glorius, F. Chem.Eur. J. 2012, 18, 16297.
(2) Nair, V.; Menon, R. S.; Biju, A. T.; Sinu, C. R.; Paul, R. R.; Jose, A.;
Sreekumar, V. Chem. Soc. Rev. 2011, 40, 5336.
(3) (a) Burstein, C.; Glorius, F. Angew. Chem., Int. Ed. 2004, 43, 6205.
(b) Sohn, S. S.; Rosen, E. L.; Bode, J. W. J. Am. Chem. Soc. 2004, 126,
14370. (c) He, M.; Bode, J. W. Org. Lett. 2005, 7, 3131. (d) Chan, A.;
Scheidt, K. A. Org. Lett. 2005, 7, 905. (e) Nair, V.; Vellalath, S.; Poonoth,
M.; Suresh, E. J. Am. Chem. Soc. 2006, 128, 8736. (f) Chan, A.; Scheidt,
D
dx.doi.org/10.1021/ja410932t | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX