Butyrolactone synthesis via polar radical crossover cycloaddition reactions: Diastereoselective syntheses of methylenolactocin and protolichesterinic acid

Article abstract of DOI:10.1021/ol5022993

A direct catalytic synthesis of γ-butyrolactones from simple alkene and unsaturated acid starting materials is reported. The catalytic system consists of the Fukuzumi acridinium photooxidant and substoichiometric quantities of a redox-active cocatalyst. Oxidizable alkenes such as styrenes and trisubstituted aliphatic alkenes are cyclized with unsaturated acids via polar radical crossover cycloaddition (PRCC) reactions. This method has been applied to the diastereoselective total synthesis of methylenolactocin and protolichesterinic acid.

Full text of DOI:10.1021/ol5022993

Letter
pubs.acs.org/OrgLett
Terms of Use
Butyrolactone Synthesis via Polar Radical Crossover Cycloaddition
Reactions: Diastereoselective Syntheses of Methylenolactocin and
Protolichesterinic Acid
Mary A. Zeller,† Michelle Riener,† and David A. Nicewicz*
Department of Chemistry, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599-3290, United States
S
* Supporting Information
ABSTRACT: A direct catalytic synthesis of γ-butyrolactones
from simple alkene and unsaturated acid starting materials is
reported. The catalytic system consists of the Fukuzumi
acridinium photooxidant and substoichiometric quantities of a
redox-active cocatalyst. Oxidizable alkenes such as styrenes and
trisubstituted aliphatic alkenes are cyclized with unsaturated acids via polar radical crossover cycloaddition (PRCC) reactions.
This method has been applied to the diastereoselective total synthesis of methylenolactocin and protolichesterinic acid.
γ-Butyrolactones are prevalent heterocycles in naturally
occurring isolates that often possess a broad range of bioactivity
including antibacterial, antifungal, antiviral, and anti-inflamma-
tory properties.1 Increased bioactivity is commonly observed for
γ-butyrolactones possessing an α-methylene moiety, which is
often implicated as the active pharmacophore.2 The paraconic
acids are a family of γ-butyrolactones with extensively
documented medicinal properties and have been the target of
synthetic studies for a number of years (Figure 1).3
Methylenolactocin, isolated by Park and co-workers, was found
to possess unique antibacterial activity4 and has been previously
synthesized by Greene,5 and others,3,6 in diastereoselective and
Figure 1. Biologically active γ-butyrolactones.
enantioselective multistep syntheses. Similarly, protolichesterinic
acid exhibits antifungal activity and inhibitory activity toward
Markovnikov alkene hydroetherification,19a hydrotrifluorome-
thylation,19b hydroamination,19c,d and hydrohalogenation19e
reactions. We have also taken advantage of a polar radical
crossover cycloaddition mechanism (PRCC) to synthesize
tetrahydrofurans arising from the reaction of electron-rich
alkenes and allyl alcohols.19f This method employed phenyl-
malononitrile as the redox-active hydrogen atom source to
deliver tetrahydrofurans in excellent yields.
Extension of this work to include α,β-unsaturated acids as
nucleophiles could provide a direct synthesis of γ-butyrolactones
from simple and commercially available starting materials. The
attenuated nucleophilicity of α,β-unsaturated acids relative to
allyl alcohols was an initial concern; however, we have previously
demonstrated success in the anti-Markovnikov hydroacetox-
ylation of alkenes, where acetic acid was employed as the
nucleophile.19g We reported that (1) catalytic quantities of base
could increase the nucleophilicity of the acid, improving both
yield and rate and (2) benzenesulfinic acid, thiophenols, or
disulfides could be used in catalytic quantities as viable redox-
active cocatalysts.
colon carcinoma,7,8 and enantioselective syntheses have been
reported.9
Because of their pronounced medicinal function, γ-butyr-
olactones are appealing targets for study and have spawned
numerous general methods for their synthesis.10 Highlights for
the synthesis of γ-butyrolactones include variations of the
catalytic enal−aldehyde annulation11 and halolactonization.
Several examples for the direct synthesis of α-methylene γ-
butyrolactones are known, including the Drieding−Schmidt
reaction,12 allylboration,13 enyne cyclizations,14 and Pd(II)-
catalyzed acylation−lactonization reaction.15 Unfortunately,
these reports often require highly prefunctionalized starting
materials, which can often restrict the structural diversity of the
lactones constructed employing these methods.
We envisioned that a direct method to form γ-butyrolactones
with α-methylene or α-alkylidine functionalities might be
achieved via organic photoredox catalysis.16 We also hypothe-
sized that such a method could be used to access members of the
paraconic acid family by further elaboration of the γ-
butyrolactone products. We have previously reported the
addition of nucleophiles to oxidizable alkenes using the
Fukuzumi acridinium photooxidant17,18 in conjunction with a
redox-active hydrogen atom donor in the development of anti-
Received: August 2, 2014
Published: September 5, 2014
© 2014 American Chemical Society
4810
dx.doi.org/10.1021/ol5022993 | Org. Lett. 2014, 16, 4810−4813

Organic Letters
Letter
a
Scheme 1. Proposed PRCC Mechanism
Table 1. Optimization of Reaction Conditions
b
conversion
yield 5a (5b)
entry
R
conditions
(%)
(%)
c
1
2
3
4
5
6
7
8
9
Me
Me
Me
H
1 equiv of 6
96
88
20 (34)
18 (20)
2 (44)
2 (43)
62
c
1 equiv of 7
20 mol % of 8
20 mol % of 8
63
We propose a PRCC mechanism similar to our previous
tetrahydrofuran report (Scheme 1). Single electron oxidation of
the alkene by the acridinium excited state (*E1/2red = +2.06 V vs
SCE) would furnish an electrophilic alkene cation radical.
Addition of the carboxylic acid to the alkene cation radical would
afford radical 2 poised to undergo a 5-exo-trig radical cyclization
to give 3. Despite the unusual electronic constraints of this radical
cyclization, work by Clive supported the feasibility of this step.20
Following this, hydrogen atom transfer to carbon-centered
radical 3 would furnish the desired γ-butyrolactone. The success
of this proposal was contingent upon the careful selection of an
appropriate redox-active hydrogen atom source or surrogate
which would meet three criteria: (1) a pKa high enough to act as a
mild conjugate base; (2) a low X−H bond dissociation energy
(BDE = 70−80 kcal/mol); and (3) redox activity allowing the
resulting radical (4) to act as an oxidant to reset the
photooxidant.
We began by investigating the reaction between crotonic acid
and β-methylstyrene. Previously developed conditions for the
tetrahydrofuran PRCC methodology furnished the desired
lactone product 5a in a 20% yield, with significant amounts of
a byproduct (5b) resulting from anti-Markovnikov hydro-
acyloxylation (Table 1, entry 1). Additionally, a Michael addition
byproduct between crotonic acid and 6 was observed in small
quantities. This prompted us to consider a more sterically
hindered H atom donor, such as 9-cyanofluorene (7).
Unfortunately, 7 also gave modest amounts of lactone product
in equal ratio to uncyclized material (entry 2).
We considered that uncyclized byproduct formation was due
to the hydrogen atom transfer event occurring prior to the 5-exo
radical cyclization. When thiophenol (8) was employed with
crotonic acid, the uncyclized product 5b dominated (entry 3).
Acrylic acid was also found to give 5b (entry 4). Speculating that
the rate of the 5-exo radical cyclization might be increased by
either stabilizing the forming radical and/or by modulating the
polarity of the unsaturated acid, we turned our attention to
unsaturated acids bearing radical stabilizing groups. It was found
that formation of 5b could be minimized if a thiol hydrogen atom
donor was matched with an unsaturated acid bearing a radical
stabilizing group (ester, aryl, silyl, or halogen) at the β-position.21
Subjecting mono-tert-butyl fumarate to the standard conditions
with the inclusion of catalytic amounts of thiophenol produced
the lactone adduct in a promising 62% yield (entry 5). The solid,
nonvolatile thiol 9 gave the lactone product in similar yields
(entry 6). As thiyl radicals are prone to dimerize, we considered
50
CO2t-Bu 20 mol % of 8
CO2t-Bu 20 mol % of 9
CO2t-Bu 10 mol % of 10
CO2t-Bu 10 mol % of 11
CO2t-Bu 10 mol % of 10, no base
80
100
99
64
79
100
100
64
64
a
Reactions carried out on a 0.3 mmol scale in N2-sparged CH2Cl2
[0.15 M] at 23 °C with a 1.1:1 acid/alkene for 24 h unless otherwise
b
1
noted. Yield of lactone determined by H NMR analysis of crude
reaction. 4 days, 5:1 acid/alkene.
c
that the corresponding disulfides could form and were curious if
they would be catalytically active in this context. To test this, we
employed disulfides 10 and 11 as potential hydrogen atom
surrogates (entries 7 and 8). When 10 was employed in catalytic
amounts, the desired adduct was observed in 79% yield (entry 7).
As with the previously reported anti-Markovnikov hydro-
acetoxylation, it was found that catalytic amounts of base could
improve reactivity depending on the nucleophilicity of the acid
(entry 9).
Having identified the optimal reaction conditions, we turned
our attention to the substrate scope, where β-methylstyrene was
investigated in combination with various α,β-unsaturated acids
(Figure 2). Commercially available monoethyl fumarate was a
viable substrate for this reaction, giving the expected lactone 12a
in 66% yield (2.2:1 dr); however, mono-tert-butyl fumarate was
favored due to the simplicity of characterization. The nature of
the ester functionality had limited effect on either yield or
diastereoselectivity. Cinnamic acid and p-chlorocinnamic acid
were found to be suitable nucleophiles, producing the desired
lactones in 81% (12c) and 80% (12d) yields, respectively. We
were pleased to find that our method was also tolerant to
substrates possessing a heteroaromatic group, forging lactone
12e in moderate yield (44%).
We attempted the use of propiolic acid as the PRCC partner,
with the hope of directly forging α-methylene-γ-butyrolactones.
Only trace quantities of the desired lactone product were
observed. We speculated that the radical cyclization was again the
problematic step in this process. However, we were pleased to
find that when 3-(trimethylsilyl)propiolic acid was employed as
the acid, γ-butyrolactone 12f was isolated in excellent yield
(85%) as a 1.3:1 mixture of E/Z isomers. Other alkynoic acids,
including 3-phenylpropiolic acid and 3-(thiophene-3-yl)-
propiolic acid, could also be employed and afforded the
4811
dx.doi.org/10.1021/ol5022993 | Org. Lett. 2014, 16, 4810−4813

Organic Letters
Letter
Figure 2. Reaction scope of α,β-unsaturated acids. Reactions carried out
in in N2-sparged CH2Cl2 [0.15 M], average of two isolated yields on 100
mg scale. Conditions A: 1:1.1 alkene/acid, 10 mol % of 2,6-lutidine, 10
mol % of 10. Conditions B: 3:1 alkene/acid, 15 mol % of 10. Conditions
C: 2−3:1 alkene/acid, 10 mol % of 2,6-lutidine, 15 mol % of 10 carried
out for 2−4 days. (a) Reaction carried out for 48 h.
Figure 3. Reaction scope of oxidizable alkenes. Reactions carried out in
N2-sparged CH2Cl2 [0.15 M], average of two isolated yields on 100 mg
scale. Conditions A: 1:1.1 alkene/acid, 10 mol % of 2,6-lutidine, 10 mol
% of 10. Conditions B: 3:1 alkene/acid, 10 mol % of 2,6-lutidine, 5 mol
% of 9. Conditions C: 3:1 alkene/acid, 15 mol % of 10
anticipated lactone adducts 12g and 12h in 41% and 44% yields,
respectively. Neither silyl groups nor aromatic groups at the β-
position are required to stabilize the subsequent radical, as seen
by the reaction of 3-cyclohexylpropiolic acid with β-methylstyr-
ene in an 83% yield (12i).
1). Not surprisingly, we were able to obtain the thermodynamic
isomers in high levels of diastereomeric purity. This epimeriza-
We next tested the scope of oxidizable alkenes in conjunction
with either mono-tert-butyl fumarate, cinnamic acid, or 3-
(trimethylsilyl)propiolic acid for the synthesis of γ-butyrolac-
tones (Figure 3). Because of the scrambling of alkene geometry
through the formation of the radical cation intermediate,
oxidizable olefins could be employed as a mixture of E/Z
isomers without affecting the high (>20:1) diastereoselectivity
with respect to the starting alkene substituents. Mono-tert-butyl
fumarate, in combination with β-methylstyrene derivatives of
varying electron density, forged the desired lactones (13a, 13b,
and 13c) in good yields. Trisubstituted alkenes such as 1-
phenylcyclohexene and 2-methyl-2-butene gave lactones 13d
and 13f in good yields. α-Methylstyrene was a viable substrate,
providing the lactone 13e in 81% yield, albeit with almost no
relative stereocontrol (1.3:1 dr). We found that 3-(trimethyl-
silyl)propiolic acid reacted with a variety of styrene derivatives.
Electron-withdrawing β-methylstyrenes reacted to furnish γ-
butyrolactones (13g, 13l) in 73% and 72% yield, respectively.
Moderately electron-donating styrenes were also tolerated, as 4-
tert-butyl-β-methylstyrene gave the corresponding lactone in a
78% yield (13k). When 3-(trimethylsilyl)propiolic acid was used
with 2-methyl-2-butene (13j), only the Z-isomer was observed. It
is possible that the fleeting vinyl radical intermediate equilibrates
to relieve nonbonding interactions between the neighboring
gem-dimethyl group and the bulky trimethylsilyl group.22
To upgrade the diasteromeric ratio of the lactone adducts, we
submitted lactones 12b and 12c to epimerization conditions (eq
tion study also confirmed the relative stereochemistry of our
major adduct, as the all-trans isomer has been noted previously as
the most stable conformer.23
In an effort to better understand the potential reversibility of
the putative radical cyclization step, we synthesized bromide 14
from β-bromoacrylic acid and β-methylstyrene using our typical
reaction protocol and subjected a single diastereomer of 14 to
classical radical hydrodehalogenation conditions (eq 2). If the
radical cyclization was reversible, variable quantities of 5b (R =
H) would be expected (Table 1). We observed only clean
hydrodebromination of 14 (73% yield) and isolated 15 as a single
diastereomer. This result suggests that the radical cyclization step
of the PRCC is irreversible and that the predominance of
uncyclized adduct, particularly in the reaction of acrylic acid with
β-methylstyrene (Table 1, entry 4), is likely the result of a slow
radical cyclization relative to hydrogen atom abstraction.
4812
dx.doi.org/10.1021/ol5022993 | Org. Lett. 2014, 16, 4810−4813

Organic Letters
Letter
(3) Bandichhor, R.; Nosse, B.; Reiser, O. Top. Curr. Chem. 2005, 243,
43.
(4) Park, B. K.; Nakagawa, M.; Hirota, A.; Nakayama, M. J. Antibiot.
1988, 41, 751.
(5) De Azevedo, M. B. M.; Murta, M. M.; Greene, A. E. J. Org. Chem.
1992, 57, 4567.
We then elected to apply our new method to the synthesis of
two bioactive paraconic acids. Initially we proposed desilylation
of a vinyl TMS adduct (e.g., 13g) and oxidation of an electron-
rich aromatic to a carboxylic acid with a RuO4/NaIO4 system.
However, this proved challenging as many conditions used to
desilylate the vinyl TMS resulted in unreacted E-isomer and
alkene isomerization of the desilylated Z-isomer. Furthermore,
the RuO4/NaIO4 system reacted preferentially with the α-
methylene moiety. These issues prompted us to consider β-
haloacrylic acids as reaction partners. Initial results with (E)-β-
chloroacrylic acid produced the desired adduct in low yields,
where substantial amounts of Michael addition byproduct with
thiophenol were obtained. However, the (Z)-isomer provided
the lactone in good yields and a 1.1:1 dr, while formation of the
uncyclized byproduct was suppressed (Scheme 2). Nevertheless,
treatment of the chloride isomers with mild basic conditions
afforded the desired paraconic acids as single diastereomers.
(6) (a) Braukmuller, S.; Bruckner, R. Eur. J. Org. Chem. 2006, 2110.
̈
̈
(b) Blanc, D.; Madec, J.; Popowyck, F.; Ayad, T.; Phansavath, P.;
Ratovelomanana-Vidal, V.; Genet, J.-P. Adv. Synth. Catal. 2007, 349,
̂
943. (c) Hajra, S.; Karmakar, A.; Giri, A.; Hazra, S. Tetrahedron Lett.
2008, 49, 3625. (d) Ghosh, M.; Bose, S.; Maity, S.; Ghosh, S.
Tetrahedron Lett. 2009, 50, 7102. (e) Jongkoi, R.; Choommongkol, R.;
Tarnchompoo, B.; Nimmanpipug, P.; Meepowpan, P. Tetrahedron
2009, 65, 6382. (f) Fernandes, R. A.; Chowdhury, A. K. Tetrahedron:
Asymmetry 2011, 22, 1114.
(7) (a) Brisdelli, F.; Perilli, M.; Sellitri, D.; Piovano, M.; Garbarino, J.
A.; Nicoletti, M.; Bozzi, A.; Amicosante, G.; Celenza, G. Phytother. Res.
2012, 27, 431. (b) Goel, M.; Dureja, P.; Rani, A.; Uniyal, P. L.; Laatsch,
H. J. Agric. Food Chem. 2011, 59, 2299.
(8) Kumar KC, S.; Muller, K. J. Nat. Prod. 1999, 62, 817.
̈
(9) For representative enantioselective syntheses, see: (a) Saha, S.;
Roy, S. C. Tetrahedron 2010, 66, 4278. (b) Chhor, R. B.; Nosse, B.;
Sorgel, S.; Bohm, C.; Seitz, M.; Reiser, O. Chem.Eur. J. 2003, 9, 260.
Scheme 2. Synthesis of Methylenolactocin and
Protolichesterinic Acid
̈
̈
(10) For selected recent examples of catalytic γ-butyrolactone
synthesis, see: (a) McInturff, E. L.; Mowat, J.; Waldeck, A. R.;
Krische, M. J. J. Am. Chem. Soc. 2013, 135, 17230. (b) Montgomery, T.
P.; Hassan, A.; Park, B. Y.; Krische, M. J. J. Am. Chem. Soc. 2012, 134,
11100. (c) Steward, K. M.; Corbett, M. T.; Goodman, G.; Johnson, J. S.
J. Am. Chem. Soc. 2012, 134, 20197. (d) Steward, K. M.; Gentry, E. C.;
Johnson, J. S. J. Am. Chem. Soc. 2012, 134, 7329.
(11) (a) Burstein, C.; Tschan, S.; Xie, X.; Glorius, F. Synthesis 2006,
2418. (b) Sohn, S. S.; Rosen, E. L.; Bode, J. W. J. Am. Chem. Soc. 2004,
126, 14370−14371.
In summary, our method provides a modular approach to the
synthesis of γ-butyrolactones from a variety of simple, often
commercially available, oxidizable olefins and unsaturated acids.
Perhaps most importantly, this method could be utilized to
rapidly generate libraries of unique γ-butyrolactone structures for
biological testing. A highly diastereoselective synthesis of two
important α-methylene paraconic acids, methylenolactocin and
protolichesterinic acid, was accomplished. Enantioselective
variants of this PRCC reaction are the focus of current research
efforts.
̈
(12) (a) Ohler, E.; Reininger, K.; Schmidt, U. Angew. Chem., Int. Ed.
1970, 9, 457. (b) Loffler, A.; Pratt, R. J.; Ruesch, H. P.; Dreiding, A. S.
̈
̈
Helv. Chim. Acta 1970, 53, 383.
(13) Kennedy, J. W. J.; Hall, D. G. J. Am. Chem. Soc. 2002, 124, 11586.
(14) (a) Lu, X.; Zhu, G.; Wang, Z. Synlett 1998, 115−121. (b) Zhu, G.;
Zhang, Z. J. Org. Chem. 2005, 70, 3339. (c) Lu, X.; Zhu, G.; Wang, Z.;
Ma, S.; Ji, J.; Zhang, Z. Pure Appl. Chem. 1997, 69, 553.
(15) (a) Masuyama, Y.; Nimura, Y.; Kurusu, Y. Tetrahedron Lett. 1991,
32, 225. (b) Tamaru, Y.; Hojo, M.; Yoshida, Z. J. Org. Chem. 1991, 56,
1099.
(16) Recently, Wu and co-workers have reported a direct synthesis of
γ-butyrolactones using an iridium photoredox catalyst, oxidizing α-
bromoesters for radical addition to alkenes: (a) Wei, X.-J.; Yang, D.-T.;
Wang, L.; Song, T.; Wu, L.-Z.; Liu, Q. Org. Lett. 2013, 15, 6054. For an
electrochemical method, see: (b) Shono, T.; Ohmizu, H.; Kawakami, S.;
Sugiyama, H. Tetrahedron Lett. 1980, 21, 5029.
ASSOCIATED CONTENT
* Supporting Information
Experimental procedures and spectral data. This material is
available free of charge via the Internet at http://pubs.acs.org.
■
S
(17) Fukuzumi, S.; Kotani, H.; Ohkubo, K.; Ogo, S.; Tkachenko, N. V.;
Lemmetyinen, H. J. Am. Chem. Soc. 2004, 126, 1600.
AUTHOR INFORMATION
Corresponding Author
■
(18) Nguyen, T. M.; Nicewicz, D. A. ACS Catal. 2014, 4, 355.
(19) (a) Hamilton, D. S.; Nicewicz, D. A. J. Am. Chem. Soc. 2012, 134,
18577. (b) Wilger, D. J.; Gesmundo, N. J.; Nicewicz, D. A. Chem. Sci.
2013, 4, 3160. (c) Nguyen, T. M.; Nicewicz, D. A. J. Am. Chem. Soc.
2013, 135, 9588. (d) Nguyen, T. M.; Manohar, N.; Nicewicz, D. A.
Angew. Chem., Int. Ed. 2014, 53, 6198. (e) Wilger, D. J.; Grandjean, J.-M.
M.; Lammert, T. R.; Nicewicz, D. A. Nat. Chem. 2014, 6, 720.
(f) Grandjean, J.; Nicewicz, D. A. Angew. Chem., Int. Ed. 2013, 52, 3967.
(g) Perkowski, A. J.; Nicewicz, D. A. J. Am. Chem. Soc. 2013, 135, 10334.
(20) Clive, D. L. J.; Beaulieu, P. L. J. Chem. Soc., Chem. Commun. 1983,
307.
*E-mail: nicewicz@unc.edu.
Author Contributions
†These authors contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
The project described was supported by the David and Lucile
Packard Foundation.
■
(21) Bordwell, F. G.; Zhang, X.-M.; Satish, A. V.; Cheng, J. P. J. Am.
Chem. Soc. 1994, 116, 6605.
(22) Singer, L. A.; Kong, N. P. Tetrahedron Lett. 1966, 19, 2089.
(23) Shimada, S.; Hashimoto, Y.; Saigo, K. J. Org. Chem. 1993, 58,
5226.
REFERENCES
■
(1) (a) Kitson, R. R. A.; Millemaggi, A.; Taylor, R. J. K. Angew. Chem.,
Int. Ed. 2009, 48, 9426. (b) Seitz, M.; Reiser, O. Curr. Opin. Chem. Bio.
2005, 9, 285.
(2) Hoffmann, H. M. R.; Rabe, J. Angew. Chem., Int. Ed. 1985, 24, 94.
4813
dx.doi.org/10.1021/ol5022993 | Org. Lett. 2014, 16, 4810−4813