Desymmetrization of 4-hydroxy-2,5-cyclohexadienones by radical cyclization: Synthesis of optically pure γ-lactones

Article abstract of DOI:10.1039/b920884d

Optically pure γ-lactones fused onto a cyclohexenone are available by a radical-based method for desymmetrizing 4-hydroxy-2,5-cyclohexadienones.

Full text of DOI:10.1039/b920884d

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Desymmetrization of 4-hydroxy-2,5-cyclohexadienones by radical
cyclization: synthesis of optically pure c-lactonesw
Rajesh Sunasee and Derrick L. J. Clive*
Received (in Cambridge, UK) 8th October 2009, Accepted 27th November 2009
First published as an Advance Article on the web 16th December 2009
DOI: 10.1039/b920884d
Optically pure c-lactones fused onto a cyclohexenone are
available by radical-based method for desymmetrizing
stereochemistry of the acetal center influences the diastereo-
selectivity of cyclization in the sense shown by eqn (1).6
a
4-hydroxy-2,5-cyclohexadienones.
ð1Þ
As part of exploratory studies to evaluate potential synthetic
routes to the antileukemic agent sorbicillactone A (1),1 we
considered free radical cyclization approaches that would
desymmetrize2 the cross-conjugated ketone 2 along the lines
summarized in Scheme 1.3 In this scheme, the group X is a
homolyzable substituent and the substructure C–C–X is part
of an enantiomerically pure unit that can subsequently be
modified. Implementation of this plan requires a method
for functionalizing the tertiary—and hindered—hydroxyl
of 24 in a way that sets the stage for a diastereoselective
The result with systems containing a ring is illustrated in
Scheme 2 for the conversion of the racemic iodoacetal 7 into
the lactol ether (ꢀ)-9, a process which occurs with very high
diastereoselectivity (498%) at ꢁ78 1C, via the reacting
conformation 8.7 Another relevant observation is that the
optically pure vinyl ether 10 gave a 1 : 1 diastereoisomeric
mixture (12) when treated with NBS in the presence of the
symmetrical alcohol 1,4-pentadien-3-ol (11), but each diastereo-
isomer of 12 was found to undergo highly diastereoselective
(ds 4 98%) ring closure.7a,b Chromatography of the cycliza-
tion products derived from 12 allowed isolation of optically
pure 13, which was then used for the synthesis of a natural
product.7a,b,8
radical cyclization (3-4-5) which must afford
a
product that can be elaborated into an optically pure g-lactone
(5-6).
The information summarized in eqn (1) and Schemes 2 and
3 suggested that the use of a glycal (14) would serve our needs
(Scheme 4), especially as cyclization of carbohydrate-derived
radicals onto simple anomeric olefinic pendants is known to
proceed efficiently.9 We expected that the approach of
Scheme 4 would avoid the formation of diastereoisomeric
mixtures by controlling the stereochemical outcome in both
the haloetherification (2-15) and the radical cyclization
(15-17) steps, and would also provide opportunities for
degradation to release the required lactone (17-18). Accordingly,
a brief experimental survey of several glycals (19–22) was
made, beginning with triacetate 19, which is commercially
The formation of g-lactones via radical cyclization is a
standard outcome of processes that begin with cyclization of
a Stork haloacetal,5 and prior literature provided information
on the stereochemical outcome to be expected from the
ring closure of relevant haloacetals. In acyclic systems, the
Scheme 2 Stereochemical outcome for cyclic compounds.
Scheme 1 General plan.
Chemistry Department, University of Alberta, Edmonton, Alberta,
Canada T6G 2G2. E-mail: derrick.clive@ualberta.ca;
Fax: +1 780-492-8231; Tel: +1 780-492-3251
w Electronic supplementary information (ESI) available: Experimental
procedures as well as X-ray data for 28. CCDC 754501. For ESI and
crystallographic data in CIF or other electronic format see DOI:
10.1039/b920884d
Scheme
3 Diastereoselectivity of radical closure. Reagents and
conditions: (a) NBS, 1,4-pentadien-3-ol (11), 75%. (b) Bu3SnH, Et3B,
O2, 85%; separation.
ꢂc
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 701–703 | 701

View Article Online
Scheme 5 Use of acetate-protected galactal.
Scheme 4 Use of glycals.
compatible with the observed coupling constants between the
hydrogens on the iodine-bearing carbon and the adjacent
acetal carbon. Acid hydrolysis (aqueous CF3CO2H) of 28
served to release the diol 29. This hydrolysis was also tried
with TsOH–MeOH and PPTS–MeOH, but neither of these
systems gave the diol. With aqueous CF3CO2H, however, the
hydrolysis was efficient (81%) and the diol was then easily
cleaved by treatment with Pb(OAc)4 to dialdehyde 30 (85%).
The indicated stereochemistry at C* in 30 is an arbitrary
assignment, but is of no consequence. We initially experienced
difficulty in further degrading the dialdehyde. Experiments
using CF3CO2H in aqueous THF, aqueous HCl–THF or
BiCl3ꢃH2O–MeOH were all unsuccessful, but we eventually
found that acid hydrolysis with 0.1 M H2SO4, followed by
Jones oxidation of the resulting crude product, gave the
lactone 31 in 65% yield over the two steps, which must be
done separately.15 Evidently, facile decarboxylation occurs
during treatment with the Jones reagent.
available. The cross-conjugated ketone 2 reacted with 19 in the
presence of NIS (Scheme 5) to give iodide 23 as a single isomer
whose full stereochemistry was not established. The
compound underwent radical cyclization, giving again a single
isomer (75%) whose stereochemistry was partially established
(TROESY). However, attempts to degrade10 the derived triol
25 to a lactone were unsuccessful.
Suspecting that degradation of a diol, rather than a triol,
might be simpler, we decided to repeat the sequence of
Scheme 5 with the D-arabinal diacetate11 20. However, the
first step (the iodoetherification) produced a mixture of two
isomers, suggesting that the presence of a C(6) substituent
on the glycal is important for stereochemical control. The
di-O-acetyl-D-glucal derivative 21 was examined next, this
compound being chosen because it is easily prepared and the
methanesulfonyloxy group might facilitate12 the degradation
process, after radical cyclization. Surprisingly, the iodoetheri-
fication (cf. 2-23) did not work—at least, in the two attempts
we made—and so we returned to the galactal series and used
compound 22.13 Our hope was that the bulky ketal unit would
foster a high degree of stereoselectivity in both the iodoetheri-
fication and the radical cyclization steps, and that the ketal
might also be easily removable to afford a diol with the
original C(6) acetoxy group intact. In the presence of NIS
(Scheme 6) the tertiary alcohol 2 reacted with 22 to provide a
single iodoether (69%) that underwent stannane-mediated
radical cyclization under standard conditions (slow addition
of Bu3SnH and AIBN in PhH to a refluxing solution of the
substrate in PhH). The stereochemistry of the crystalline
product 28 was determined by TROESY measurements and
the assignment was later confirmed by X-ray analysis. On this
basis, the stereochemistry of its precursor 27 is assumed to be
as shown. The stereochemical course proposed for the
iodoetherification (2-27) is in line with prior observations
made with a related galactal,14 and the assignment to 27 is
Scheme 6 Use of ketal-protected galactal.
ꢂc
This journal is The Royal Society of Chemistry 2010
702 | Chem. Commun., 2010, 46, 701–703

View Article Online
Table 1 Etherification and radical cyclization
cyclohexadienone, and the final lactones have functionality
that is appropriate for further elaboration.
We thank the Natural Sciences and Engineering Research
Council of Canada for financial support.
Notes and references
1 (a) G. Bringmann, G. Lang, T. A. M. Gulder, H. Tsurata,
J. Muhlbacher, K. Maksimenka, S. Steffens, K. Schaumann,
¨
R. Stohr, J. Wiese, J. F. Imhoff, S. Perovic-Ottstadt, O. Boreiko
¨
and W. E. G. Muller, Tetrahedron, 2005, 61, 7252–7265;
¨
(b) G. Bringmann, T. A. M. Gulder, G. Lang, S. Schmitt,
R. Stohr, J. Wiese, K. Nagel and J. F. Imhoff, Mar. Drugs, 2007,
¨
K. Schaumann, S. Steffens, P. G. Rytik, U. Hentschel,
¨
5, 23–30; (c) G. Bringmann, G. Lang, J. Muhlbacher,
32
33
34
R = Pr, R1 = H, R2 = H
R = Me, R1 = H, R2 = Me
R = Me, R1 = Me, R2 = H
32a 64%
33a 67%
34a 63%
32b 80%
33b 80%
34b 74%
J. Morschhauser and W. E. G. Muller, Marine Molecular
¨
¨
Biotechnology, in Sponges (Porifera), ed. W. E. G. Muller,
¨
Springer, Berlin, 2003, vol. 1, pp. 231–253; (d) W. E. G. Muller,
G. Bringmann, G. Lang, J. Muhlbacher, K. Schaumannand
¨
S. Steffens, WO 026 854 A1, 2004; (e) G. Bringmann, G. Lang,
¨
Conditions: (a) galactal 22, NIS, MeCN; (b) Bu3SnH, AIBN, PhH,
85 1C.
J. Muhlbacher, K. Schaumann, S. Steffens, P. G. Rytik,
¨
U. Hentschel, J. Morschhauser and W. E. G. Muller, Prog. Mol.
¨
¨
Subcell. Biol., 2003, 37, 231–253.
2 Review on desymmetrization of 1,4-cyclohexadienes: A. Studer
and F. Schleth, Synlett, 2005, 3033–3041.
3 Ionic methods for desymmetrizing 2,5-cyclohexadienones to
products with high ee: (a) Q. Liu and T. Rovis, J. Am. Chem.
Soc., 2006, 128, 2552–2553; (b) E. Merino, R. P. A. Melo,
M. Ortega-Guerra, M. Ribagorda and M. C. Carreno, J. Org.
Chem., 2009, 74, 2824–2831; (c) Y. Hayashi, H. Gotoh, T. Tamura,
H. Yamaguchi, R. Masui and M. Shoji, J. Am. Chem. Soc., 2005,
127, 16028–16029; (d) N. T. Vo, R. D. M. Pace, F. O’Hara and
M. J. Gaunt, J. Am. Chem. Soc., 2008, 130, 404–405.
´ ´
4 M. C. Carreno, M. Gonzalez-Lopez and A. Urbano, Angew.
Chem., 2006, 45, 2737–2741.
5 (a) G. Stork, R. Mook, Jr., S. A. Biller and S. D. Rychnovsky,
J. Am. Chem. Soc., 1983, 105, 3741–3742; (b) Y. Ueno, K. Chino,
M. Watanabe, O. Moriya and M. Okawara, J. Am. Chem. Soc.,
1982, 104, 5564–5566.
6 (a) F. Villar and P. Renaud, Tetrahedron Lett., 1998, 39,
8655–8658; (b) A. L. J. Beckwith and D. M. Page, J. Org. Chem.,
1998, 63, 5144–5153; (c) O. Corminboeuf, P. Renaud and
C. H. Schiesser, Chem.–Eur. J., 2003, 9, 1578–1584.
7 (a) F. Villar, O. Equey and P. Renaud, Org. Lett., 2000, 2,
1061–1064; (b) F. Villar, T. Kolly-Kovac, O. Equey and
The above reaction sequence for desymmetrization of cross-
conjugated ketones of type 2 is general, as it was successfully
applied to the three additional substrates 32,16 33,4 and 344
listed in Table 1, which shows the iodoetherification and
radical closure steps. The stereochemical outcome follows
the pattern established for 2-28, as indicated by TROESY
measurements on 32b–34b. In the case of the propyl series
1
(Table 1, compound 32) the H and 13C NMR spectra of the
initial iodoether showed a small amount of a contaminant
(ca. 5%, assuming it is an isomer), but the radical cyclization
product was pure, as was the final lactone (32d, 64% from 32c,
Table 2).
Each of the radical cyclization products 32b, 33b, and 34b
was degraded to the corresponding lactone shown in Table 2,
using conditions established in making lactone 31. The inter-
mediate dialdehydes 33c and 34c were sensitive, and we were
unable to obtain them pure, but the crude materials gave the
desired pure lactones 33d and 34d in the overall yields
indicated in Table 2. We did not establish the stereochemistry
at C* in 32c–34c, nor at C(4) in 33d; but 33d was shown to be a
single isomer (as were all the other lactones we made).
Our results show that the present method for
desymmetrization tolerates alkyl substitution on the starting
P. Renaud, Chem.–Eur. J., 2003, 9, 1566–1577; (c) J. C. Lo
A. M. Gomez and B. Fraser-Reid, Aust. J. Chem., 1995, 48,
333–352.
´
pez,
´
8 Cf. R. Nouguier, S. Gastaldi, D. Stien, M. Bertrand, F. Villar,
O. Andrey and P. Renaud, Tetrahedron: Asymmetry, 2003, 14,
3005–3018.
9 (a) A. De Mesmaeker, P. Hoffmann, T. Winkler and A. Waldner,
Synlett, 1990, 201–204; (b) A. De Mesmaeker, P. Hoffmann and
B. Ernst, Tetrahedron Lett., 1989, 30, 57–60.
10 Attempted cleavage of the triol, using NaIO4 or Pb(OAc)4, gave
complex mixtures.
Table 2 Formation of g-lactones
11 Made by the method of: B. K. Shull, Z. Wu and M. Koreeda,
J. Carbohydr. Chem., 1996, 15, 955–964.
12 By acetate hydrolysis, elimination of OMs to form an enol ether,
diol cleavage and acid hydrolysis.
13 A. Fernandez-Mayoralas, A. Marra, M. Trumtel, A. Veyrieres and
P. Sinay, Carbohydr. Res., 1989, 188, 81–95. We used a different
¨
preparation (see ESIw).
14 (a) R. A. Alonso, G. D. Vite, R. E. McDevitt and B. Fraser-Reid,
J. Org. Chem., 1992, 57, 573–584; (b) J. Thieme and B. Meyer,
Chem. Ber., 1980, 113, 3067–3069.
15 In principle, the Jones reagent should serve for both the hydrolysis
of 30 and its subsequent oxidation, but this procedure gave a lower
yield (one experiment).
16 J. McKinley, A. Aponick, J. C. Raber, C. Fritz, D. Montgomery
and C. T. Wigal, J. Org. Chem., 1997, 62, 4874–4876. We used the
preparative method of ref. 4.
32b R = Pr, R1 = H, R2 = H
32c 86%a, 80%b 32d 64%c
33b R = Me, R1 = H, R2 = Me 33c -%
34b R = Me, R1 = Me, R2 = H 34c -%
33d 53%d
34d 62%e
a
b
For hydrolysis with CF3CO2H. For diol cleavage with Pb(OAc)4.
c
d
From 32c. Yield from 33b. From 34b.
e
ꢂc
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 701–703 | 703