Reversal of the regioselectivity in a cycloaddition of o-quinones by varying the position of alkoxy substituents

Article abstract of DOI:10.1016/j.tetlet.2008.02.109

We have investigated the regioselective cycloaddition of o-quinones 1b-e with the protected sinapyl alcohol 2. It was found that the position of the alkoxy substituent on the o-quinone ring controlled the regioselectivity of the cycloaddition. In addition

Full text of DOI:10.1016/j.tetlet.2008.02.109

Available online at www.sciencedirect.com
Tetrahedron Letters 49 (2008) 2558–2561
Reversal of the regioselectivity in a cycloaddition of o-quinones
by varying the position of alkoxy substituents
Atsuhito Kuboki *, Toru Yamamoto, Mamie Taira, Tetsuya Arishige, Rina Konishi,
Mami Hamabata, Mutsumi Shirahama, Takeya Hiramatsu, Kohei Kuyama, Susumu Ohira *
Department of Biochemistry, Okayama University of Science, 1-1 Ridai-cho, Okayama 700-0005, Japan
Received 15 January 2008; revised 12 February 2008; accepted 18 February 2008
Available online 23 February 2008
Abstract
We have investigated the regioselective cycloaddition of o-quinones 1b–e with the protected sinapyl alcohol 2. It was found that the
position of the alkoxy substituent on the o-quinone ring controlled the regioselectivity of the cycloaddition. In addition, our reported
procedure for determining the location of the side chains on 1,4-benzodioxanes has been improved.
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: o-Quinone; 1,4-Benzodioxane; Cycloaddition; Regioselectivity
R
Considerable interest has been shown in neolignans hav-
ing 1,4-benzodioxane moiety in the structure due to their
potent biological activities. We first synthesized 1,4-benzo-
dioxane 3a and its regioisomer 30a as a 2:1 mixture by a
cycloaddition of o-quinone 1a with the protected sinapyl
alcohol 2.1 Subsequently, it was found that the reaction
of 1b, bearing a methoxy substituent at C-3, with 2 afforded
3b as the sole product in a much better yield than that of
3a1 (Scheme 1).
We have now investigated the effect of alkoxy substitu-
ents attached to the o-quinone component upon the regio-
selectivity of the cycloaddition, using o-quinones 1b–e as
shown in Figure 1. In addition, a plausible mechanism
for the cycloaddition is proposed to explain the results.
Preparation of o-quinone 1c is shown in Scheme 2. Con-
version of one acetyl group in the para-position of triace-
tate 4 into a PMB group afforded 5 in 77% yield. The
remaining two acetyl groups of 5 were exchanged for
benzyl groups to give 6 in 86% yield, which was trans-
formed into 8 almost quantitatively via the Weinreb amide
7. Simultaneous protection of the carbonyl moiety of 8 as a
OTBS
O
O
a
+
OMe
OH
O
O
2
1a (R = H)
(R = OMe)
OMe
1b
OMe
OH
R
O
O
OTBS
O
OMe
+
OMe
OH
O
OTBS
O
O
OMe
3
3'
Scheme 1. Reagents and conditions: (a) THF, rt [R = H: 3 h, 69% (3a/3a0
2:1), R = OMe: 83% (3b only)].
cyclic acetal and deprotection of the PMB group provided
phenol 9 in 81% yield. Subsequent oxidation of 9 with
IBX2,3 gave 1c in 56% yield.
Synthesis of o-quinone 1d commenced with TBS protec-
tion of commercially available 4-hydroxy-2-methoxybenz-
aldehyde (10) (Scheme 3). The protected 10 was treated
with methyllithium and then subjected to TPAP-catalyzed
*
Corresponding authors. Tel./fax: +81 86 2569489 (A.K.).
E-mail address: kuboki@dbc.ous.ac.jp (A. Kuboki).
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.02.109

A. Kuboki et al. / Tetrahedron Letters 49 (2008) 2558–2561
2559
OR
oxidation to provide acetophenone 11 in 75% yield. Protec-
tion of the ketone moiety of 11 as a cyclic acetal and sub-
sequent removal of the TBS ether group gave phenol 12 in
71% yield. Oxidation of 12 with IBX2 afforded o-quinone
1d in 82% yield.
O
O
RO
O
O
2
3 from 1b,c
3' from 1d,e
O
O
O
O
1b (R = Me)
1c (R = Bn)
1d (R = Me)
1e (R = Bn)
o-Quinone 1e was prepared from the commercially
available 20,40-dihydroxyacetophenone (13) by a similar
route as used for the synthesis of 1d (Scheme 4). Regio-
selective protection of 13 with MOM and benzyl groups,4
followed by the replacement of the MOM group by a
TBS group, afforded 14 in 81% yield over four steps. A
sequence involving the protection of the carbonyl moiety
of 14 as a cyclic acetal, removal of the TBS group, and
oxidation with IBX3,4 provided 1e in 80% yield over three
steps.
Fig. 1. Investigation of regioselectivity into the cycloaddition of 1b–e
with 2.
OAc
OBn
b,c
e
OR
OPMB
OBn
MeO2C
R = Ac: 4
OAc
R
The reactions of 1b–e with 2 selectively afforded 3b,1 3c,5
30d,6 and 30e,7 respectively, the structures of which were
determined through a modification of our previous pro-
cedure1 (Scheme 5). Iodides derived from the cycloadducts
were briefly treated with excess n-butyllithium. In situ
acetylation of the two phenoxy groups that were liberated
by ring cleavage and elimination of the methanesulfonyl
group gave the corresponding diacetates 15c, 150d,8 and
150e9 with good reproducibility. The substitution patterns
of the benzene rings of 150d,e were determined by NOE
experiments. However, the structure of 15c could not be
assigned on the basis of NOE experiments at this point.
Therefore, it was transformed into triacetate 1610 through
R = CO2Me: 6
R = CON(OMe)Me: 7
a
d
R = PMB: 5
OBn
OBn
g
OH
f
OPMB
OBn
1c
OBn
O
O
O
9
8
Scheme 2. Reagents and conditions: (a) PMBCl, KI, K2CO3, acetone,
50 °C, 19 h (77%); (b) K2CO3, H2O, MeOH, 50 °C, 0.5 h (quant.); (c)
BnBr, K2CO3, TBAI, DMF, 50 °C, 0.5 h (86%); (d) MeHN(OMe)HCl,
MeMgBr, THF, rt, 0.5 h (7: 50%, 8: 49%); (e) MeMgBr, THF, 0 °C to rt,
0.5 h (99% from 7); (f) ethylene glycol, PPTS, benzene, reflux, 1 h (81%);
(g) IBX, DMSO, rt, 0.5 h (56%).
OMe
OAc
R
O
H
H
a-c
MeO
OHC
OH
MeO
O
OTBS
O
a-d
OMe
3'd-e
O
O
O
Ac
10
11
NOE
15'd (R = Me)
15'e (R = Bn)
d,e
f
MeO
OH
1d
OBn
OAc
O
O
O
a-d
3c
OMe
12
O
O
Scheme 3. Reagents and conditions: (a) TBSCl, imidazole, DMF, rt, 0.5 h
(94%); (b) MeLi, THF, ꢀ78 °C to rt, 1 h (85%); (c) TPAP, NMO, MS4A,
CH2Cl2, rt, 0.5 h (88%); (d) ethylene glycol, p-TsOH, benzene, reflux, 3 h
(90%); (e) TBAF, THF, rt, 0.5 h (79%); (f) IBX, DMSO, rt, 0.5 h (82%).
OAc
OMe
15c
δ 2.28 ppm
2.29 ppm
2.32 ppm
OAc
OAc
e-f
HO
O
OH
BnO
O
OTBS
a-d
e-g
OMe
O
1e
O
O
OAc
16
OMe
13
14
Scheme 5. Reagents and conditions: (a) TBAF, THF, rt, 2 h; (b) MsCl,
Et3N, CH2Cl2, 0 °C, 30 min (c: 96%, d: 83%, e: 91% in 2 steps); (c) NaI,
i-Pr2NEt, DMF, 80 °C, 1 h (c: 66%, d: 75%, e: 64%); (d) n-BuLi, THF,
ꢀ78 °C, 1 min, then Ac2O, ꢀ78 °C, 15 min (c: 65%, d: 19%, e: 35%); (e)
H2, Pd/C, EtOAc, rt, overnight; (f) Ac2O, Et3N, CH2Cl2, 0 °C to rt, 1.5 h
(61% in 2 steps).
Scheme 4. Reagents and conditions: (a) MOMCl, K2CO3, acetone, 0 °C
to rt, 0.5 h (94%); (b) BnBr, NaH, TBAI, DMF, 0 °C to rt, 1 h (95%); (c)
AcCl, MeOH, 0 °C to rt, 0.5 h (95%); (d) TBSCl, imidazole, DMF, 0 °C to
rt, 0.5 h (95%); (e) ethylene glycol, p-TsOH, benzene, reflux, 0.5 h (89%);
(f) TBAF, THF, 0 °C to rt, 0.5 h (95%); (g) IBX, DMSO, rt, 0.5 h (95%).

2560
A. Kuboki et al. / Tetrahedron Letters 49 (2008) 2558–2561
R1
debenzylation and acetylation. The unsymmetrical
structure of 16 proved the structure of 15c.
OTBS
OTBS
R2
O
O
SET
The yields and ratios of 3 and 30 obtained from the
cycloaddition of o-quinones (1b–e) with 2 are summarized
in Table 1. o-Quinones 1b and 1c predominantly gave 3b
and 3c, respectively. In contrast, 1d and 1e provided 30d
and 30e. The regioselectivity was thus reversed by changing
the position of the alkoxy group on the o-quinone ring. The
reactivity, yield, and regioselectivity were somewhat lower
with the benzyl ethers (1c,e) than with the methyl ethers
(1b,d). With 1c,e, the addition of base (K2CO3) proved to
be crucial for the progress of the reaction.
+
OMe
OH
OMe
O
H
e
O
O
1
2
C
OMe
OMe
SET
e
R1
R1
R1
R2
R2
R2
O
O
O
O
O
O
O
O
O
O
O
O
To explain these results, we propose a stepwise reaction
mechanism for the cycloaddition, which includes a single-
electron transfer (SET), as outlined below (Fig. 2). Depro-
tonation11 of 2 and subsequent SET to 1 gives radical C as
well as anion radicals A or B, each of which may be char-
acterized by three major resonance forms (A1–3, B1–3).
When R1 is an alkoxy group and R2 is hydrogen (1b,c),
anion radical A is more stable than B due to the contribu-
tion of form A3, in which the alkoxy group stabilizes the
adjacent carbon radical. Similarly, anion radical B is more
stable than A when R2 is an alkoxy group and R1 is hydro-
gen (1d,e) due to the contribution of form B2. Radical
coupling of C with either A or B gives an anionic inter-
mediate D or D0, which undergoes conjugate addition to
afford product 3 or 30, respectively. The intermediates,
which include para-quinomethide structure, may lose an
information of the geometry to give thermodynamically
stable trans adducts preferentially. In the case of 1c and
1e, elimination of a benzyl radical from A3 or B2 may lead
to the decomposition of the anion radical intermediates,
which may lower the yield.
A1
A2
A3
R1
R1
R1
R2
R2
R2
O
O
O
O
O
O
O
O
O
O
O
O
B1
C
B2
B3
R1
R2
O
H
OTBS
OMe
A
+
3
O
O
O
O
D
OMe
OMe
O
R1
R2
O
O
H
OMe
+
B
C
3'
OTBS
O
O
D'
In summary, we have demonstrated regioselective
cycloadditions of 1b–e with the protected sinapyl alcohol
2. The selectivity is controlled by the position of the alkoxy
group on the o-quinone ring. Moreover, a plausible
mechanism has been proposed, and the procedure for
determining the structures of the resulting cycloadducts
has been improved in the course of this work. Our current
efforts are directed toward the regiocontrolled synthesis
of natural products that have the 1,4-benzodioxane
structures.
Fig. 2. A plausible mechanism for the cycloaddition of 1 with 2.
Acknowledgment
The authors thank Mr. Y. Fukuda of Okayama Univer-
sity of Science for assistance with the NMR measurements.
References and notes
1. Kuboki, A.; Yamamoto, T.; Taira, M.; Arishige, T.; Ohira, S.
Tetrahedron Lett. 2007, 48, 771.
2. Magdziak, D.; Rodriguez, A. A.; Water, R. W. V. D.; Pettus, T. R. R.
Org. Lett. 2002, 4, 285.
Table 1
Cycloadditions of o-quinones (1b–e) with 2
3. Ozanne, A.; Pouysegu, L.; Depernet, D.; Franßcois, B.; Quideau, S.
Org. Lett. 2003, 5, 2903.
additive, THF
3a
3'a
+
+
o-quinone (1)
2
4. Methylation instead of benzylation at this stage led to the C-alkylated
product as the major product. Thus, we prepared 1c from 10, not 13.
5. Spectral data for 3c: 1H NMR (400 MHz, CDCl3): d 0.06 (s, 3H), 0.10
(s, 3H), 0.90 (s, 9H), 1.63 (s, 3H), 3.58 (dd, J = 2.4, 11.7 Hz, 1H),
3.74–3.78 (m, 2H), 3.90 (s, 6H), 3.90–3.95 (m, 2H), 3.97–4.00 (m, 2H),
5.00 (d, J = 7.8 Hz, 1H), 5.11 (d, J = 11.7 Hz, 1H), 5.17 (d,
J = 11.7 Hz, 1H), 5.61 (s, 1H), 6.71 (s, 2H), 6.72 (d, J = 2.0 Hz,
1H), 6.80 (d, J = 2.0 Hz, 1H), 7.30–7.40 (m, 3H), 7.46–7.50 (m, 2H);
13C NMR (100 MHz, CDCl3): d ꢀ5.4, ꢀ5.0, 18.3, 25.7, 25.9, 27.5,
56.3, 62.2, 64.41, 64.43, 71.1, 76.2, 78.5, 104.1, 104.2, 107.3, 108.5,
127.6, 127.7, 128.3, 133.6, 134.8, 135.1, 136.9, 143.9, 146.9, 147.5.
r.t., time
Entry
Quinone
Additive
Time (h)
Yield (%)
Ratio (3:30)
1
2
3
4
a
1bb
1cc
1d
None
K2CO3
None
3
10
14
14
83
57
82
59
1:0
5:1
1:15
1:8
1e
K2CO3
A trace amount of cis isomers was observed.
See Ref. 1.
The reaction was carried out at 50 °C.
b
c

A. Kuboki et al. / Tetrahedron Letters 49 (2008) 2558–2561
2561
6. Spectral data for 30d: 1H NMR (400 MHz, CDCl3): d 0.06 (s, 3H),
0.07 (s, 3H), 0.90 (s, 9H), 1.77 (s, 3H), 3.55 (dd, J = 2.4, 11.2 Hz, 1H),
3.81 (s, 3H), 3.82–3.88 (m, 4H), 3.91 (s, 6H), 4.02–4.06 (m, 2H), 4.97
(d, J = 7.8 Hz, 1H), 5.58 (br s, 1H), 6.60 (s, 1H), 6.68 (s, 2H), 7.12 (s,
1H); 13C NMR (100 MHz, CDCl3): d ꢀ5.4, ꢀ5.0, 18.3, 25.5, 25.9,
56.27, 56.31, 62.4, 64.58, 64.63, 76.1, 78.2, 101.6, 104.1, 108.1, 115.2,
123.8, 127.6, 134.9, 136.3, 143.4, 147.0, 151.2.
7. Spectral data for 30e: 1H NMR (400 MHz, CDCl3): d 0.06 (s, 3H), 0.08
(s, 3H), 0.90 (s, 9H), 1.83 (s, 3H), 3.54 (dd, J = 2.4, 12.2 Hz, 1H),
3.70–3.91 (m, 4H), 3.88 (s, 6H), 4.04–4.09 (m, 2H), 4.95 (d,
J = 7.8 Hz, 1H), 5.08 (s, 2H), 5.60 (s, 1H), 6.62 (s, 1H), 6.66 (s,
2H), 7.14 (s, 1H), 7.26–7.33 (m, 1H), 7.35–7.39 (m, 2H), 7.47–7.50 (m,
2H); 13C NMR (100 MHz, CDCl3): d ꢀ5.3, ꢀ5.0, 18.4, 25.7, 25.9,
56.3, 62.4, 64.5, 64.6, 71.0, 76.8, 78.2, 103.4, 104.2, 108.2, 115.4, 124.6,
126.9, 127.4, 127.6, 128.4, 135.0, 136.7, 137.4, 143.3, 147.0, 150.1.
8. Spectral data for 150d: 1H NMR (400 MHz, CDCl3): d 1.72 (s, 3H),
2.23 (s, 3H), 2.33 (s, 3H), 3.72 (s, 3H), 3.82 (s, 6H), 3.87–3.92 (m, 2H),
4.00–4.04 (m, 2H), 5.30 (d, J = 10.4 Hz, 1H), 5.30 (d, J = 17.1 Hz,
1H), 5.39 (d, J = 6.0 Hz, 1H), 6.05 (ddd, J = 6.0, 10.4, 17.1 Hz, 1H),
6.50 (s, 1H), 6.66 (s, 2H), 7.17 (s, 1H).
9. Spectral data for 150e: 1H NMR (400 MHz, CDCl3): d 1.78 (s, 3H),
2.23 (s, 3H), 2.31 (s, 3H), 3.80 (s, 6H), 3.83–3.86 (m, 2H), 4.01–4.05
(m, 2H), 5.01 (s, 2H), 5.22 (d, J = 10.8 Hz, 1H), 5.28 (d, J = 16.8 Hz,
1H), 5.33 (d, J = 6.0 Hz, 1H), 5.97 (ddd, J = 6.0, 10.8, 16.8 Hz, 1H),
6.49 (s, 1H), 6.60 (s, 2H), 7.19 (s, 1H), 7.28–7.38 (m, 5H).
10. Spectral data for 16: 1H NMR (400 MHz, CDCl3): d 1.01 (t,
J = 7.3 Hz, 3H), 1.51 (s, 3H), 2.01–1.87 (m, 2H), 2.28 (s, 3H), 2.29
(s, 3H), 2.32 (s, 3H), 3.29–3.33 (m, 1H), 3.47–3.52 (m, 1H), 3.79 (s,
6H), 5.04 (dd, J = 5.4, 7.3 Hz, 1H), 6.57 (s, 2H), 6.74 (d, J = 2.0 Hz,
1H), 6.75 (d, J = 2.0 Hz, 1H).
11. We assume that the initial deprotonation plays an important role in
the cycloaddition, because it was accelerated by addition of base in
case of 1b and 1d and could not proceed under acidic conditions in all
cases.