TABLE 2. Inhibition of TPA-Induced Ear Edema by
Hispanolone and Its Derivativesa
11), the TLC control also showed only one spot corre-
sponding to products. However, if the reaction was not
treated immediately, a new product was obtained after
3 days. After separation, product 917 was obtained in a
65% yield and only the 7c,7d pair of diastereoisomers
was obtained in a 35% yield. These diastereoisomers
corresponded to the major pair obtained during the
electrolysis and, in NMR, to those that have the S
configuration on C-16. Because of their stability, and the
disappearance of the other diastereoisomeric pair, the
proposal of a reversible cyclization reaction seems to be
supported. From the results obtained from the semiem-
pirical calculations (see section 3 of the Supporting
Information), the two diastereoisomers generated by the
attack over the re face (7c and 7d) corresponded to those
with the lowest total energy values. This observation and
the experimental results are in agreement with the
thermodynamic control of the reaction producing, as the
major compounds, the pair of diastereoisomers with
lowest energy (configuration 16S). The proposed stereo-
chemistry for the compounds observed in high yield at
the different temperatures is as follows: 7c 15S, 16S and
7d 15R, 16S (CH3O signals at 3.24 ppm in 1H NMR; 53.20
and 53.36 ppm in 13C NMR).
compd
µmol/ear % inhibition ( SEM IC50 (µmol/ear)
1a
7
0.316
1.000
3.160
0.316
1.000
3.160
0.316
1.000
3.160
0.316
1.000
3.160
0.13
21.44 ( 4.8*
40.76 ( 6.3*
83.26 ( 3.6*
29.89 ( 7.2*
35.65 ( 5.7*
56.35 ( 2.4*
12.87 ( 4.1*
24.88 ( 6.7*
56.35 ( 2.4*
5.10 ( 2.8*
21.09 ( 4.6*
31.96 ( 7.0*
35.15 ( 6.4*
48.18 ( 2.0*
56.29 ( 8.2*
69.42 ( 9.4*
90.35 ( 2.6*
1.05
2.26
8
>3.16
>3.16
9
0.24
indomethacin
0.42
0.27
0.75
1.30
a *p < 0.05.
12).13 The main spectroscopic signals for the spiro moiety
are equivalent to those described for 7a-7d.
Experiments at different temperatures (Table 1) showed
that at temperatures lower than 20 °C, the reaction was
very clean and the sole observed products in TLC were
compounds 7a-7d. At higher temperatures, other byprod-
ucts appeared on the TLC decreasing the chemical yield
of the reaction. This can be explained as a result of a
faster but less selective oxidation of the electrogenerated
bromine with the starting compound. The reactions at
low temperature neither favored a specific diastereoiso-
meric pair nor affected the ratio of the pairs of diaste-
reoisomers in a predictable way. Nevertheless, it was
clearly observed that one pair is almost always present
in a higher yield; it has the highest selectivity (93:7) at
room temperature. It is possible that this pair of diaste-
reoisomers is more stable than the other, and thus, its
formation is thermodynamically favored (vide infra).14
The electrochemical method was compared with the
typical furan oxidation methods. Compound 1a was
allowed to react with NBS15 (2 mol in 1,4-dioxane/water
at room temperature). At the end of the reaction, TLC
showed a complex mixture of products that was not
separated. GC-MS of the mixture demonstrated the
presence of bromated products. The TLC analysis of the
bromine oxidation of 1a in MeOH at room temperature
showed the spot for products 7a-7d, but other byprod-
ucts were also observed. These experiments showed the
high selectivity of the electrochemical method in the
furan oxidation. The use of distilled MeOH or anhydrous
MeOH at -22 °C (Table 1, entries 5 and 6) had practically
the same results.16
All the spectroscopic data of compound 9 point to the
presence of a butenolide ring, and the R- or â-substitution
was proposed by comparison with the typical 1H and 13
C
NMR signals for both substitutions in natural products.18
The observed signals corresponded with an R-substitu-
tion, and the proposal was confirmed by a NOESY NMR
experiment in which the interactions between the meth-
ylene of the butenolide ring at 4.79 ppm and the vinylic
proton at 7.1 ppm were clearly observed. In the literature,
this R-butenolide has not been described as the only
product reported is the product with a â-substitution,
Leopersin G;19 thus, product 9 was named iso-Leopersin
G. By taking into account that â-butenolide substitution
is most common in the natural products isolated from
plants of the Lamiaceae family,18 we can easily reach the
alternative substitution from the terminal furan by
means of this electrochemical reaction. Another possible
application of this methodology is to transform the furan
ring of furanic diterpenoids to 1,5-disubstituted pyrroli-
din-2-ones; this is a transformation in which the 2,5-
dialkoxy-2,5-dihydrofuran could be the key intermedi-
ate.20 Product 9 could be obtained by acidic hydrolysis
with subsequent rearrangements from 7a-7d catalyzed
by HBrO or HBr, generated from water and traces of the
halogen (see Scheme 1 in the Supporting Information for
a mechanism proposal). An analogous reaction between
phthalaldehyde and amines to obtain unsaturated γ-lac-
tams has been reported.21,22 To check this proposal, we
allowed the mixture of diastereoisomers 7a-7d to react
for 3 days with a catalytic quantity of p-TSA at room
When the quantity of hispanolone was increased to 1.6
mmol (500 mg) in the same electrolytic cell (Table 1, entry
(13) See Table 2 of the Supporting Information for the 1H and 13C
NMR data for this compound.
(14) Atkinson, R. S. Stereoselective Synthesis; John Wiley & Sons:
New York, 1995; pp 12-15.
(17) See Table 3 of the Supporting Information for the 1H and 13C
NMR data and the discussion for this compound.
(18) Rodr´ıguez-Hahn, L.; Esquivel, B.; Ca´rdenas, J. In Progress in
the Chemistry of Organic Natural Products; Herz, W., Kirby, G. W.,
Moore, R. E., Steglich, W., Tamm, Ch., Eds.; Springer-Verlag: New
York, 1994; Vol. 63, p 145.
(19) Tasdemir, D.; Sticher, O.; Calis, I.; Linden, A. J. Nat. Prod.
1997, 60, 874-879.
(20) Katritzky, A. R.; Mehta, S.; He, H.-Y.; Cui, X. J. Org. Chem.
2000, 65, 4364-4369.
(15) Ferland, J. M.; Lefevre, Y.; Deghenghi, R.; Wiesner, K. Tetra-
hedron Lett. 1966, 3617-3620.
(16) This observation was previously reported for furan electro-
oxidation (see ref 8). The water contained in the distilled MeOH is
about 0.01%, and for dry MeOH obtained using Na or Mg as the drying
agent, it is about 5 × 10-5% as reported by Perrin, D. D.; Armarego,
W. L. F. Purification of Laboratory Chemicals, 3rd ed.; Pergamon
Press: Oxford, U.K., 1988; p 217.
4540 J. Org. Chem., Vol. 70, No. 11, 2005