Acid-promoted reaction of the stilbene antioxidant resveratrol with nitrite ions: Mild phenolic oxidation at the 4′-hydroxystiryl sector triggering nitration, dimerization, and aldehyde-forming routes

Article abstract of DOI:10.1021/jo060482i

In 0.1 M phosphate buffer, pH 3.0, and at 37 °C, resveratrol ((E)-3,4′,5-trihydroxystilbene, 1a), an antioxidant and cancer chemopreventive phytoalexin, reacted smoothly at 25 μM or 1 mM concentration with excess nitrite ions (NO₂^-) to give a complex pattern of products, including two novel regioisomeric α-nitro (3a) and 3′-nitro (4) derivatives along with some (E)-3,4′,5-trihydroxy-2, 3′-dinitrostilbene (5), four oxidative breakdown products, 4-hydroxybenzaldehyde, 4-hydroxy-3-nitrobenzaldehyde, 3,5- dihydroxyphenylnitromethane, and 3,5-dihydroxybenzaldehyde, two dimers, the resveratrol (E)-dehydrodimer 6 and restrytisol B (7), and the partially cleaved dimer 2. The same products were formed in the absence of oxygen. ¹H, ¹⁵N HMBC and LC/MS analysis of the crude mixture obtained by reaction of 1a with Na¹⁵NO₂ suggested the presence of 3,4′,5,β-tetrahydroxy-α-nitro-α,β-dihydrostilbene (8) as unstable intermediate which escaped isolation. Under similar conditions, the structurally related catecholic stilbene piceatannol ((E)-3,3′,4, 5′-tetrahydroxystilbene, 1b) gave, besides (E)-3,3′,4,5′- tetrahydroxy-β-nitrostilbene (3b), 3,4-dihydroxybenzaldehyde and small amounts of 3,5-dihydroxybenzaldehyde. Mechanistic experiments were consistent with the initial generation of the phenoxyl radical of 1a at 4′-OH, which may undergo free radical coupling with NO₂ at the α- or 3′-position, to give eventually nitrated derivatives and/or oxidative double bond fission products, or self-coupling, to give dimers. The oxygen-independent, NO₂^--mediated oxidative fission of the double bond under mild, physiologically relevant conditions is unprecedented in stilbene chemistry and is proposed to involve breakdown of hydroxynitro(so) intermediates of the type 8.

Full text of DOI:10.1021/jo060482i

Acid-Promoted Reaction of the Stilbene Antioxidant Resveratrol
with Nitrite Ions: Mild Phenolic Oxidation at the 4′-Hydroxystiryl
Sector Triggering Nitration, Dimerization, and Aldehyde-Forming
Routes
Lucia Panzella,† Maria De Lucia,† Carmine Amalfitano,‡ Alessandro Pezzella,†
Antonio Evidente,‡ Alessandra Napolitano,† and Marco d’Ischia*,†
Department of Organic Chemistry and Biochemistry and Dipartimento di Scienze del Suolo, delle Piante e
dell’Ambiente, UniVersity of Naples “Federico II”, Naples, Italy
dischia@cds.unina.it
ReceiVed March 6, 2006
In 0.1 M phosphate buffer, pH 3.0, and at 37 °C, resveratrol ((E)-3,4′,5-trihydroxystilbene, 1a), an
antioxidant and cancer chemopreventive phytoalexin, reacted smoothly at 25 µM or 1 mM concentration
-
with excess nitrite ions (NO2 ) to give a complex pattern of products, including two novel regioisomeric
R-nitro (3a) and 3′-nitro (4) derivatives along with some (E)-3,4′,5-trihydroxy-2,3′-dinitrostilbene (5),
four oxidative breakdown products, 4-hydroxybenzaldehyde, 4-hydroxy-3-nitrobenzaldehyde, 3,5-
dihydroxyphenylnitromethane, and 3,5-dihydroxybenzaldehyde, two dimers, the resveratrol (E)-dehy-
drodimer 6 and restrytisol B (7), and the partially cleaved dimer 2. The same products were formed in
the absence of oxygen. 1H,15N HMBC and LC/MS analysis of the crude mixture obtained by reaction of
1a with Na15NO2 suggested the presence of 3,4′,5,â-tetrahydroxy-R-nitro-R,â-dihydrostilbene (8) as
unstable intermediate which escaped isolation. Under similar conditions, the structurally related catecholic
stilbene piceatannol ((E)-3,3′,4,5′-tetrahydroxystilbene, 1b) gave, besides (E)-3,3′,4,5′-tetrahydroxy-â-
nitrostilbene (3b), 3,4-dihydroxybenzaldehyde and small amounts of 3,5-dihydroxybenzaldehyde.
Mechanistic experiments were consistent with the initial generation of the phenoxyl radical of 1a at
4′-OH, which may undergo free radical coupling with NO2 at the R- or 3′-position, to give eventually
nitrated derivatives and/or oxidative double bond fission products, or self-coupling, to give dimers. The
-
oxygen-independent, NO2 -mediated oxidative fission of the double bond under mild, physiologically
relevant conditions is unprecedented in stilbene chemistry and is proposed to involve breakdown of
hydroxynitro(so) intermediates of the type 8.
Introduction
duced by various plants, including grapes, berries, and peanuts,
in response to microbial attack.1 Currently, they are under
intensive study for their many potent biological activities,
suggesting a potential role in the prevention of coronary heart
disease and as cancer chemopreventive agents.2 Besides acting
Resveratrol (5-[2-(4-hydroxyphenyl)vinyl]benzene-1,3-diol,
1a) and its metabolite piceatannol (4-[2-(3,5-dihydroxyphenyl)-
vinyl]benzene-1,2-diol, 1b) are polyphenolic phytoalexins pro-
* Corresponding author. Phone: +39-081674132. Fax: +39-081674393.
† Department of Organic Chemistry and Biochemistry.
‡ Dipartimento di Scienze del Suolo.
(1) Ku, K. L.; Chang, P. S.; Cheng, Y. C.; Lien, C. Y. J. Agric. Food
Chem. 2005, 53, 3877-3881 and references therein.
10.1021/jo060482i CCC: $33.50 © 2006 American Chemical Society
Published on Web 04/26/2006
4246
J. Org. Chem. 2006, 71, 4246-4254

Acid-Promoted Reaction of ResVeratrol with Nitrite
as inhibitors of specific enzymes such as cyclooxygenase,3,4 1a
has been shown to be an efficient scavenger of cytotoxic oxygen
and nitrogen species.5-9 Studies of structure-activity relation-
ships indicate that the antioxidant activity of 1a stems from the
peculiar oxygenation pattern on the planar stilbenic skeleton,
featuring as a crucial determinant of the radical scavenger
activity the 4′-OH group, synergistically supported by the 3-
and 5-OH groups on the resorcin moiety. The efficiency of the
4′-OH group as a hydrogen donor is enhanced by the trans
double bond, which increases both its acidity10 and the resonance
stabilization energy of the phenoxyl radical derived from H-atom
abstraction, as confirmed by semiempirical (PM3)11 and DFT12-14
analysis.
considering that NO-derived species are important mediators/
contributory factors in the inflammatory response and in
carcinogenesis.22 The major physiologic metabolite of NO is
nitrite (NO2-), which is present at high levels (30-210 µM) in
saliva and is also found in polluted drinking waters, vegetables
(e.g. spinaches), fertilizers, and preserved/pickled meats.23,24
Within the stomach and other acidic compartments supporting
nitrous acid (HNO2) formation,25 NO2- may cause nucleobase
deamination and interstrand cross-link formation and production
of mutagenic N-nitrosamines.26 Determining the susceptibility
of 1a to react with acidic NO2- and the identity of the reaction
products is therefore of particular interest to predict the possible
transformations and fate of 1a in the stomach in the presence
-
of high NO2 levels. In this connection it is worth noting that
1b has recently been shown to react efficiently with acidic nitrite
via a regioselective nitration at the double bond sector.27 Studies
of the nitr(os)ation chemistry of 1a are also expected to provide
a convenient entry to novel stilbene derivatives of potential
synthetic and pharmacological interest, e.g. in the field of
cyclooxygenase inhibitors and antiestrogenic compounds.3,4,28
This study describes the isolation and structural characteriza-
tion of the main products formed by acid-promoted reaction of
1a with NO2- under mild conditions, with the view to filling a
major gap in the chemistry of this bioactive stilbene and to
gaining an improved background for further studies of the
biological activity of this phytoalexin. Further interest of this
study stems from the potential bioactivity of the oxidation/
nitration products of 1a against phytopathogenic fungi which
is currently under scrutiny.
Because of the central relevance to the antioxidant activity
of 1a as well as to the process of biotransformation by the plant
pathogens into a range of oligomer species,15-17 the oxidation
chemistry of 1a has been the subject of considerable interest,
and several aspects have been clarified.18-20 Little is known,
by contrast, on the reaction of 1a with reactive nitrogen species
derived from nitric oxide (NO),21 the only relevant paper dealing
with the peroxynitrite-induced conversion to oxidation products.5
This represents a considerable gap in stilbene chemistry
Results and Discussion
(2) Jang, M.; Cai, L.; Udeani, G. O.; Slowing, K. V.; Thomas, C. F.;
Beecher, C. W. W.; Fong, H. H. S.; Farnsworth, N. R.; Kinghorn, A. D.;
Mehta, R. G.; Moon, R. C.; Pezzuto, J. M. Science 1997, 275, 218-220.
(3) Murias, M.; Handler, N.; Erker, T.; Pleban, K.; Ecker, G.; Saiko, P.;
Szekeres, T.; Ja¨ger, W. Bioorg. Med. Chem. 2004, 12, 5571-5578.
(4) Szewczuk, L. M.; Fori, L.; Stivala, L. A.; Penning, T. M. J. Biol.
Chem. 2004, 279, 22727-22737.
Acid-Promoted Reaction of 1a with NO2-. Product Char-
acterization. In a first series of experiments 1a (1 × 10-3 M)
-
was reacted with NO2 (5 molar equiv) in 0.1 M phosphate
buffer, pH 3.0. Reverse phase HPLC analysis of the reaction
mixture after 3 h indicated complete substrate consumption and
the presence of a complex pattern of products (Figure 1, plot
A), two of which (tR 29.2, product V, and tR 37.9 min, product
VII) displayed intense UV absorption at 320 nm.
(5) Brito, P.; Almeida, L. M.; Dinis, T. C. P. Free Radical Res. 2002,
36, 621-631.
(6) Lorentz, P.; Roychowdhury, S.; Engelmann, M.; Wolf, G.; Horn, T.
F. W. Nitric Oxide 2003, 9, 64-76.
(7) Burkitt, M. J.; Duncan, J. Arch. Biochem. Biophys. 2000, 381, 253-
263.
The complexity of the reaction mixture was confirmed by
TLC analysis of the ethyl acetate extractable fraction which
showed seven bands at Rf 0.36, 0.40, 0.48, 0.55, 0.69, 0.78,
and 0.84, two of which (Rf 0.36 and 0.55) exhibited a marked
bathochromic shift on exposure to alkali. At lower concentra-
tions of both 1a (2.5 × 10-5 M) and NO2- (8 molar equiv added
with stirring over 2 h at 15 min intervals of time), that is, under
conditions aimed to model interactions that may occur in the
gastric compartment during digestion following continuous
elevated nitrite intake, the product pattern was slightly different
(Figure 1, plot B). In particular, formation of VII and the product
(8) Olas, B.; Nowak, P.; Kolodziejczyk, J.; Ponczek, M.; Wachowicz,
B. J. Nutr. Biochem. 2006, 17, 96-102.
(9) Amorati, R.; Ferroni, F.; Pedulli, G. F.; Valgimigli, L. J. Org. Chem.
2003, 68, 9654-9658.
(10) Caruso, F.; Tanski, J.; Villegas-Estrada, A.; Rossi, M. J. Agric. Food
Chem. 2004, 52, 7279-7285.
(11) Stivala, L. A.; Savio, M.; Carafoli, F.; Perucca, P.; Bianchi, L.; Maga,
G.; Forti, L.; Pagnoni, U. M.; Albini, A.; Prosperi, E.; Vannini, V. J. Biol.
Chem. 2001, 276, 22586-22594.
(12) Leopoldini, M.; Marino, T.; Russo, N.; Toscano, M. J. Phys. Chem.
A 2004, 108, 4916-4922.
(13) Cao, H.; Pan, X.; Li, C.; Zhou, C.; Deng, F.; Li, T. Bioorg. Med.
Chem. Lett. 2003, 13, 1869-1871.
(14) Amorati, R.; Lucarini, M.; Mugnaini, V.; Pedulli, G. F.; Roberti,
M.; Pizzirani, D. J. Org. Chem. 2004, 69, 7101-7107.
(15) Cichewicz, R. H.; Kouzi, S. A.; Hamann, M. T. J. Nat. Prod. 2000,
63, 29-33.
(22) Ohshima, H.; Tatemichi, M.; Sawa, T. Arch. Biochem. Biophys.
2003, 417, 3-11.
(23) Duncan, C.; Dougall, H.; Johnston, P.; Green, S.; Brogan, R.; Leifert,
C.; Smith, L.; Golden, M.; Benjamin, N. Nat. Med. 1995, 1, 546-551.
(24) Coss, A.; Cantor, K. P.; Reif, J. S.; Lynch, C. F.; Ward, M. H. Am.
J. Epidemiol. 2004, 159, 693-701.
(25) Samouliov, A.; Kuppusamy, P.; Zweier, J. L. Arch. Biochem.
Biophys. 1998, 357, 1-7.
(26) Bartsch, H.; Oshima, H.; Pignatelli, B.; Calmels, S. Pharmacoge-
netics 1992, 2, 272-277.
(27) De Lucia, M.; Panzella, L.; Crescenzi, O.; Napolitano, A.; Barone,
V.; d′Ischia, M. Bioorg. Med. Chem. Lett. 2006, 16, 2238-2242.
(28) Gehm, B. D.; McAndrews, J. M.; Chien, P. Y.; Jameson, J. L. Proc.
Natl. Acad. U.S.A. 1997, 94, 14138-14143.
(16) Huang, K. S.; Lin, M.; Yu, L. N.; Kong, M. Tetrahedron 2000, 56,
1321-1329.
(17) Breuil, A.-C.; Adrian, M.; Pirio, N.; Meunier, P.; Bessis, R.; Jeandet,
P. Tetrahedron Lett. 1998, 39, 537-540.
(18) Nicotra, S.; Cramarossa, M. R.; Mucci, A.; Pagnoni, U. M.; Riva,
S.; Forti, L. Tetrahedron 2004, 60, 595-600.
(19) Sako, M.; Hosokawa, H.; Ito, T.; Iinuma, M. J. Org. Chem. 2004,
69, 2598-2600.
(20) del C. Pinto, M.; Garc´ıa-Barrado, J. A.; Mac´ıas, P. J. Agric. Food
Chem. 2003, 51, 1653-1657.
(21) Eiserich, J. P.; Patel, R. P.; O′Donnell, V. B. Mol. Aspects Med.
1998, 19, 221-357.
J. Org. Chem, Vol. 71, No. 11, 2006 4247

Panzella et al.
-
FIGURE 1. HPLC elution profile of the reaction mixture of 1a with NO2 in 0.1 M phosphate buffer, pH 3.0, at 37 °C, at 3 h reaction time at
different concentrations: plot A, 1.0 × 10-3 M 1a, 5.0 × 10-3 M NO2-; plot B, 2.5 × 10-5 M 1a, 2.0 × 10-4 M NO2-. Elution conditions: eluant
A, detection at 280 nm.
at tR 24.2 min (IV) was enhanced, while an abatement of those
at tR 12.2 (I), 15.5 (II), 20.9 (III), and 32.3 (VI) min was
observed; moreover, novel species, for example that eluted at
tR 38.7 min (VIII), were present. Under such conditions (2.5 ×
10-5 M 1a, 1 mM NO2-), a pseudo-first-order rate constant of
(9.9 ( 0.5) × 10-3 s-1 for 1a decay was determined.
phenol moieties but lacked the pair of doublets for the trans
protons on the stilbene double bond. These were replaced by a
1H singlet appearing downfield at δ 8.14, suggesting a strong
deshielding effect caused by a spatially close nitro group. A
distinct cross-peak in the 1H, 13C HMBC spectrum between the
resorcin proton resonances at δ 6.36 and a deshielded carbon
signal at δ 147.7 supported nitration on the R-position of the
stilbene system. On this basis, the product was formulated as
(E)-3,4′,5-trihydroxy-R-nitrostilbene (3a). The isolated yield of
3a was 4%.
For products isolation, the reaction of 1a (1 × 10-3 M) with
-
NO2 (5 molar equiv) was run on a preparative scale and the
ethyl acetate extractable fraction was subjected to careful TLC
fractionation.
The compound at Rf 0.40 (VI) was identified as the dimer 2,
previously isolated from Smilax bracteata rhizomes,29 while
products at Rf 0.69 and 0.84, corresponding to II and III in the
elutogram in that order, were identified as 4-hydroxybenzalde-
hyde (4%) and 4-hydroxy-3-nitrobenzaldehyde (1%), respec-
tively, by comparison with authentic samples.
The band eluting at Rf 0.48 proved positive to the Griess
reagent for nitroso compounds or nitrite-releasing species.30 On
NMR analysis it was shown to consist of an intimate mixture
of two products at a 2:1 ratio. The major product was identified
as 3,5-dihydroxybenzaldehyde (compound eluting under peak
I in the elutogram of Figure 1) by comparison of the spectral
features with those of an authentic sample. The signals of the
1H NMR spectrum pertaining to the minor component included
those typical of a resorcin moiety and a 2H singlet at δ 5.48
for a methylene group apparently linked to a nitro group (13C
NMR: δ 81.8). Accordingly, the compound was assigned the
structure of (3,5-dihydroxyphenyl)nitromethane. LC/ESI+/MS
analysis of the Rf 0.48 band showed, in addition to a peak due
to 3,5-dihydroxybenzaldehyde (tR 15.1 min), a peak at tR 13.9
min displaying a pseudomolecular ion peak [M + Na]+ at m/z
192.
The product at Rf 0.55 (VII) was evidently an isomer of 3a,
as inferred from the pseudomolecular ion peak [M - H]- in
1
the ESI-/MS spectrum at m/z 272. Inspection of the H NMR
spectrum revealed the characteristic signals for the trans double
bond (2H singlet at δ 7.14 in (CD3)2CO, appearing as a couple
of doublets (J ) 16.4 Hz) at δ 6.96 and 7.01 in CD3OD) and
the resorcin ring but indicated a substituted phenol ring, as
denoted by an ABX spin system (doublet at δ 7.20, double
The product at Rf 0.36 (V) gave a pseudomolecular ion peak
[M - H]- in the ESI-/MS spectrum at m/z 272, suggesting a
1
nitrated derivative of 1a. The H NMR spectrum featured the
expected resonances for unchanged resorcin and 4-substituted
(29) Li, S. Y.; Fuchino, H.; Kawahara, N.; Sekita, S.; Satake, M. J. Nat.
Prod. 2002, 65, 262-266.
(30) Klebanoff, S. J. Free Radical Biol. Med. 1993, 14, 351-360.
4248 J. Org. Chem., Vol. 71, No. 11, 2006

Acid-Promoted Reaction of ResVeratrol with Nitrite
TABLE 1. NMR Spectral Data (ppm) for 3a, 4, and 5 ((CD3)2CO)
the splitting patterns of the aliphatic protons at δ 3.40 (t, J )
9.2 Hz), 3.96 (t, J ) 9.2 Hz), 5.00 (d, J ) 9.6 Hz), and 5.50 (d,
J ) 8.6 Hz) were in agreement with those reported for a cis-
trans-trans configuration15 as illustrated in structure 7.
At 3 × 10-6 M concentration, 1a reacted with NO2- (0.2 ×
10-3 M, added in eight portions at 15 min intervals) to give
mainly the two nitration products 3a and 4 and 3,5-dihydroxy-
benzaldehyde and 4-hydroxybenxaldehyde, with little or no
detectable dimers formation (HPLC evidence).
3a
4
5
1H (J, Hz)
13C
1H (J, Hz)
13C
1H (J, Hz)
13C
1
2
3
4
5
133.9
139.8
106.0
159.6
103.4 6.53 (d, 2.0)
159.6
106.0 6.79 (d, 2.0)
130.5 7.42 (d, 16.0)
126.4 7.15 (d, 16.0)
131.3
123.3 8.27 (d, 2.0)
135.3
136.8
130.5
156.0
103.0
162.0
107.5
125.5
130.0
129.8
123.5
135.0
154.0
120.5
6.36 (d, 2.0) 109.3 6.61 (d, 2.0)
160.1
6.53 (t, 2.0) 104.6 6.33 (t, 2.0)
160.1
6.36 (d, 2.0) 109.3 6.61 (d, 2.0)
147.7 7.14 (s)
6
R
â
1′
8.14 (s)
134.8 7.14 (s)
123.3
Close inspection of the aqueous phase after extraction and
workup revealed the presence in all cases of chromatographically
ill-defined materials, presumably oligomers and polymers, which
could not be identified. Whether other dimers, e.g. restrytisols
A and C,15 are produced in the mixture remains uncertain,
although, if present, they would be only minor constituents. In
no case, however, could trans-ꢀ-viniferin32,33 be detected (HPLC
evidence in mixtures spiked with an authentic sample).
Acid-Promoted Reaction of 1b with NO2-. Product
Characterization. At 2 × 10-5 M concentration, 1b reacted
2′ 7.21 (d, 8.8) 134.3 8.23 (d, 2.4)
3′ 6.79 (d, 8.8) 116.4
4′
160.9
154.5
121.0 7.25 (d, 8.0)
5′ 6.79 (d, 8.8) 116.4 7.20 (d, 8.8)
6′ 7.21 (d, 8.8) 134.3 7.97 (dd, 8.8, 2.4) 135.5 8.00 (dd, 8.0, 2.0) 134.8
doublet at δ 7.97, and deshielded doublet at δ 8.23). These data,
along with 2D NMR analysis, allowed formulation of the
product as (E)-3,4′,5-trihydroxy-3′-nitrostilbene (4).31 The prod-
uct was isolated in 1% yield. Notably, 4 was also obtained in
45% formation yield by reaction of 1a (3.5 × 10-2 M) with
-
with NO2 (4 molar equiv added in four portions at 30 min
intervals of time) in 0.1 M phosphate buffer, pH 3.0, at 37 °C
to give the nitro derivative 3b as the major product.27 Close
inspection of the mixture revealed small amounts of two
additional species, of which one was isolated and identified as
3,4-dihydroxybenzaldehyde (19%) while the other was identified
as 3,5-dihydroxybenzaldehyde by comparison of chromato-
graphic properties with an authentic sample.
Effects of Oxygen, Oxidant, and pH on Product Stability
and Distribution from 1a. No significant change in product
distribution was observed when the reaction of 1 × 10-3 M 1a
with NO2- was run under an oxygen-depleted atmosphere, care
being taken to avoid contact with air prior to workup.
Furthermore, the reaction of 1a with NO2- was also run under
an 18O2 atmosphere and products were analyzed for incorpora-
tion of the label. LC/ESI+/MS analysis of the reaction mixture
confirmed the expected lack of incorporation of 18O within the
main reaction products, including notably the aldehyde deriva-
tives. Taken together, these observations ruled out any signifi-
cant involvement of O2 in the nitration, dimerization, and
aldehyde-forming pathways.
To gain some insights into the mechanism of dimerization
and oxidative cleavage, the behavior of 1a with one-electron
oxidants at acidic pH was investigated. Oxidation of 0.25 ×
10-3 M 1a with 0.25 × 10-3 M CAN in 0.1 M phosphate buffer,
pH 3.0, resulted in the formation of 6 and 7 as the main products
as well as of 4-hydroxybenzaldehyde and 3,5-dihydroxyben-
zaldehyde in comparable amounts but in much lower yields than
in the reaction with NO2-. Interestingly, oxidation of 1a with
K3Fe(CN)6 at pH 7.0 led to the formation of both aldehyde
products and the dimer 6 but no detectable 7, suggesting that
the latter reflects a specific acid-mediated oxidation pathway
of 1a.
To establish possible relationships between reaction products,
additional experiments were directed to investigate the fate of
isolated aldehydes, nitrated derivatives, and dimers on exposure
to NO2- under the usual reaction conditions. Careful monitoring
of the reaction course by HPLC or TLC at various intervals of
time showed that 4-hydroxybenzaldehyde, 3,5-dihydroxyben-
-
NO2 (0.35 M) in acetonitrile containing 2.5% acetic acid. In
this conditions no detectable formation of 3a was observed.
The compound at Rf 0.78 was characterized as a dinitro
compound (ESI+/MS: pseudomolecular ion peaks [M + H]+
and [M + Na]+ at m/z 319 and 341, respectively). The 1H NMR
spectrum showed the signals for a trans double bond, a
substituted phenol ring (doublet at δ 7.25, double doublet at δ
8.00 and deshielded doublet at δ 8.27), and two doublets (J )
2.0 Hz) at δ 6.53 and 6.79. On this basis the product was
formulated as (E)-3,4′,5-trihydroxy-2,3′-dinitrostilbene (5) (1%
yield). NMR data assignments for 3a, 4, and 5 are reported in
Table 1.
With the attempt to isolate the products IV and VIII, the
reaction of 1a (2.5 × 10-5 M) and NO2 (8 molar equiv) was
-
run on preparative scale. TLC fractionation of the ethyl acetate
extracts allowed isolation of four main bands at Rf 0.09, 0.36,
0.55, and 0.69. The product at Rf 0.09, corresponding to VIII,
was identified as the resveratrol (E)-dehydrodimer 6 (4% yield)
by comparison with literature data,17,18 whereas the species at
Rf 0.36, 0.55, and 0.69 were identified as 3a (3%), 4 (5%), and
4-hydroxybenzaldehyde (4%), in that order.
Because of the difficulties to isolate product IV by the above
procedure, an alternative approach was pursued, involving
column chromatography of the ethyl acetate extract of the
reaction mixture on Sephadex LH-20 followed by preparative
HPLC. This methodology allowed isolation of the product that
could be identified as restrytisol B15 (7) (1%) by NMR analysis
and comparison with literature data. The stereochemical features
1
of 7 were deduced from the H NMR spectrum: in particular,
(32) Amalfitano, C.; Evidente, A.; Surico, G.; Tegli, S.; Bertelli, E.;
Mugnai, L.Phytopathol. Mediterr. 2000, 39, 178-183.
(33) Li, W.-W.; Ding, L.-S.; Li, B.-G.; Chen, Y.-Z. Phytochemistry 1996,
42, 1163-1165.
(31) For the sake of simplicity, throughout this paper, the same numbering
system as for 1a was adopted for 4, which assigns numbers 1-6 to the
resorcin moiety and numbers 1′-6′ to the phenol moiety.
J. Org. Chem, Vol. 71, No. 11, 2006 4249

Panzella et al.
SCHEME 1
zaldehyde, and (3,5-dihydroxyphenyl)nitromethane in the Rf 0.48
band remained unchanged in 0.1 M phosphate buffer, pH 3.0,
at 37 °C, with or without added NO2-, after 24 h. Conversely,
under the typical reaction conditions, 4 smoothly decayed to
give 4-hydroxy-3-nitrobenzaldehyde and (3,5-dihydroxyphenyl)-
nitromethane, while 3a gave rise to 4-hydroxybenzaldehyde as
the main product. Under the same conditions 6 gave the cleaved
dimer 2.
Mechanistic Issues. The products obtained by exposure of
1a to acidic NO2- suggest competing reaction channels that lead
to nitration, dimerization, and cleavage of the stilbene double
bond. When NO2- is exposed to acids, nitrous acid (HNO2, pKa
) 3.25)34 is formed which decomposes according to the
following equilibria:
via an initial oxidative step (probably via H-atom transfer,10,35
though electron transfer has also been proposed12) leading to
the delocalized 4′-phenoxyl radical as a common intermediate
from which the nitration, dimerization, and aldehyde-forming
paths depart. Oxidation of 1a (Eox ) +1.14 V vs saturated
p
calomel electrode in CH3CN)19 may be brought about by HNO2
(the reduction potential for the equation HNO2 + H+ + e-
)
NO + H2O is +0.996 V at pH ) 0)36 or by the NO2 produced
by decomposition of HNO2 (the reduction potential for NO2 +
e- ) NO2 is 0.99 V).37
-
Formation of nitration products would involve coupling of
the phenoxyl radical with NO2 at the 3′- and R-positions38,39
(Scheme 1). According to this scheme, double bond nitration
to give 3a follows from a nitro quinone methide intermediate
and is in line with the reactivity of the 4′-phenoxyl radical at
the R-position predicted by computational studies at the
semiempirical11 and the DFT12,13 levels. This mechanism is akin
to that proposed for nitration of 1b leading to 3b27 and reflects
again the dominant role of the 4-OH group in directing the
reactivity of phenolic stilbenes toward the double bond.
However, at variance with 1a, 1b does not undergo significant
ring nitration.
NO2- + H+ a HNO2
HNO2 + H+ a NO+ + H2O
NO+ + NO2- a N2O3
N2O3 a NO + NO2
Oxygen-independent formation of aldehydes by oxidative
fission of the double bond under mild conditions of physiologi-
To distinguish between several possible mechanistic path-
ways, the O,O,O-trimethyl derivative of 1a, prepared by a
reported procedure,11 was allowed to react with acidic NO2
-
(35) Wright, J. S.; Johnson, E. R.; DiLabio, G. A. J. Am. Chem. Soc.
2001, 123, 1173-1183.
under the same conditions used for 1a, and HPLC analysis did
not reveal appreciable conversion to products. This observation
would suggest that reaction of 1a with acidic NO2 proceeds
(36) Takahama, U.; Oniki, T.; Hirota, S. J. Agric. Food Chem. 2002,
50, 4317-4322.
-
(37) van der Vliet, A.; Eiserich, J. P.; Halliwell, B.; Cross, C. E. J. Biol.
Chem. 1997, 272, 7617-7625.
(38) Al-Obaidi, U.; Moodie, R. B. J. Chem. Soc., Perkin Trans. 2: Phys.
Org. Chem. (1972-1999) 1985, 3, 467-472.
(34) In CRC Handbook of Chemistry and Physics, 75th ed.; Lide, D. R.,
Ed.; CRC Press: Boca Raton, FL, 1995.
4250 J. Org. Chem., Vol. 71, No. 11, 2006

Acid-Promoted Reaction of ResVeratrol with Nitrite
of (3,5-dihydroxyphenyl)nitromethane and 8 (mixture of di-
astereoisomers). Consistent with this interpretation, LC/ESI+/
MS analysis of two separate mixtures obtained from reaction
-
of 1a with 15N-labeled and unlabeled NO2 indicated in both
cases a species eluted at tR 14.1 min giving a pseudomolecular
ion peak [M + Na]+ at m/z 315 and 314, respectively,
confirming the presence of 8 (Figure 3). Label incorporation
was observed also in the case of 3a (tR 31.8 min) and 4 (tR
46.5 min), showing pseudomolecular ion peaks [M + H]+ and
[M + Na]+ at m/z 275 and 297, respectively.
Cleavage of a nitrohydroxy derivative akin to 8 from 4 could
account for the formation of 4-hydroxy-3-nitrobenzaldehyde.
Formation of 3,5-dihydroxybenzaldehyde is however incom-
patible with reaction pathways involving 8, since it does not
seem to arise from (3,5-dihydroxyphenyl)nitromethane. A
plausible route would be through hydrolysis of an oxime
intermediate43 produced by cleavage of a nitrosohydroxy species
akin to 8, arising by coupling of NO with the 4′-phenoxyl radical
of 1a. To test the proposed route, 3,5-dihydroxybenzaldoxime
was prepared by reaction of 3,5-dihydroxybenzaldehyde with
NH2OH in 1.2 M sodium acetate at 80 °C44 and exposed to
FIGURE 2. 1H,15N HMBC spectrum of the ethyl acetate extractable
fraction of the reaction mixture of 1a with [15N]NaNO2 at pH 3.0.
Arrows indicate cross-peaks of identified products (see text).
cal relevance has apparently escaped the attention of previous
workers in the chemistry on 1a and related natural stilbene
antioxidants. To the best of our knowledge, formation of
4-hydroxybenzaldehyde and 3,5-dihydroxybenzaldehyde (as
O-methyl derivatives) was reported only by harsh ozonolytic
splitting of the double bond of the O,O,O-trimethyl derivative
of 1a.40 It is also noteworthy that aldehyde formation is
enhanced in the NO2--induced reaction compared to the CAN
oxidation at pH 3.0, suggesting a specific NO2--dependent
mechanism. The identification of (3,5-dihydroxyphenyl)ni-
tromethane among the products argues strongly for an oxidative
fission pathway involving nucleophilic attack of water to the
nitro quinone methide intermediate in Scheme 1 to give 3,4′,5,â-
tetrahydroxy-R-nitro-R,â-dihydrostilbene (8), which would un-
dergo fragmentation to give 4-hydroxybenzaldehyde and (3,5-
dihydroxyphenyl)nitromethane. Unfortunately, all attempts to
isolate the postulated hydroxynitro derivative 8 proved unsuc-
cessful, due to the apparent instability of this species during
chromatographic separation and workup. In an attempt to
demonstrate its formation, the reaction was carried out with 2.5
-
NO2 under the usual reaction conditions: HPLC analysis of
the reaction mixture showed a ca. 50% consumption of the
oxime after 2 h, with concomitant formation of 3,5-dihydroxy-
benzaldehyde. 3,5-Dihydroxybenzaldoxime was also detected
in trace amounts by careful HPLC analysis (tR 11.4 min, eluant
A) in the reaction mixture of 1a (1 × 10-3 M) with NO2- (5 ×
10-3 M).
Mechanisms similar to those described above can be envis-
aged for cleavage of 1b.
Dimerization of 1a is well documented18,19,45 and occurs via
coupling of the resulting phenoxyl radical. Formation of 7 in
small amounts by NO2--induced oxidation of 1a is however
noteworthy, since this dimer was previously described only by
enzymatic oxidation produced by a fungal grapevine pathogen15
but was never obtained by chemical oxidation under mild
conditions. Present data indicate that one-electron oxidants, like
CAN, can induce formation of 7 and that the heterocyclic
oxygen derives from H2O rather than O2, because of the apparent
formation of this dimer under O2-depleted atmosphere. A
possible mechanism is depicted in Scheme 2.
× 10-5 M 1a and 15N-labeled NO2 under the usual reaction
-
This mechanism is akin to that proposed for the biogenesis
of tricuspitadol A, a diastereoisomer of 7 isolated from
Parthenocissus tricuspidata,46 and for the formation of related
tetrahydrofuran derivatives by chemical oxidation of caffeic acid
under acidic conditions.47 Formation of 7 in acidic but not in
neutral medium is in line with previous observations and
indicates that H2O can act as a nucleophile toward the quinone
methide only when acidic catalysis is provided.
Cleaved dimer 2 may derive at least in part by NO2--induced
oxidation of 6. The likely mechanism would involve free radical
addition of NO2 to the double bond followed by recombination
of the â-nitroalkyl radical with another molecule of NO2 to give
a nitronitrito adduct,41,48,49 which would suffer subsequent
conditions. Direct analysis of the crude ethyl acetate extractable
fraction by 1H,15N HMBC revealed, as expected, two significant
series of cross-peaks, correlating the proton resonances of 4 at
δ 8.23 (appearing as a triplet (J ) 2.4 Hz) because of the further
splitting by coupling with 15N) and δ 7.20 with a nitrogen signal
at δ 373 and the proton signals of 3a at δ 8.14 (doublet, J )
4.0 Hz) and δ 6.36 with a nitrogen resonance at δ 378 (Figure
2). Another cross-peak correlating a proton signal at δ 6.95 with
a nitrogen signal at δ 383 (Figure 2, region A) was considered
to be indicative of the presence of the Z isomer of 3a.41
In addition, two intense 15N resonances were detectable at δ
388 and 392 (Figure 2, region B), denoting nitro groups linked
to sp3 carbons.41,42 These resonances (cross-peaks with proton
signals in the range δ 5.4-5.6) may be attributed to the nitrogens
(43) Lee, J. G.; Kwak, K. H.; Hwang, J. P. Tetrahedron Lett. 1990, 31,
6677-6680.
(39) Oldreive, C.; Zhao, K.; Paganga, G.; Halliwell, B.; Rice-Evans, C.
Chem. Res. Toxicol. 1998, 11, 1574-1579.
(44) Ley, J. P.; Bertram, H.-J. Bioorg. Med. Chem. 2001, 9, 1879-1885.
(45) Takaya, Y.; Terashima, K.; Ito, J.; He, Y.-H.; Tateoka, M.;
Yamaguchi, N.; Niwa, M. Tetrahedron 2005, 61, 10285-10290.
(46) Lins, A. P.; Felicio, J. D.; Braggio, M. M.; Roque, L. C.
Phytochemistry 1991, 30, 3144-3146.
(40) Fronza, G.; Fuganti, C.; Serra, S.; Cisero, M.; Koziet, J. J. Agric.
Food Chem. 2002, 50, 2748-2754.
(41) Napolitano, A.; Panzella, L.; Savarese, M.; Sacchi, R.; Giudicianni,
I.; Paolillo, L.; d’Ischia, M. Chem. Res. Toxicol. 2004, 17, 1329-1337.
(42) Napolitano, A.; Crescenzi, O.; Camera, E.; Giudicianni, I.; Picardo,
M.; d’Ischia, M. Tetrahedron 2002, 58, 5061-5067.
(47) Fulcrand, H.; Cheminat, A.; Brouillard, R.; Cheynier, V. Phytochem-
istry 1994, 35, 499-505.
J. Org. Chem, Vol. 71, No. 11, 2006 4251

Panzella et al.
FIGURE 3. (A) LC/ESI+/MS elution profile of the ethyl acetate extractable fraction of the reaction mixture of 1a with 15N-labeled NO2 at pH
-
3.0. (B) ESI+/MS spectrum of the species eluting at tR 14.1 min in the ethyl acetate extractable fraction of the reaction mixture of 1a with 15N-
labeled NO2-at pH 3.0. (C) ESI+/MS spectrum of the species eluting at tR 14.1 min in the ethyl acetate extractable fraction of the reaction mixture
-
of 1a with unlabeled NO2 at pH 3.0.
cleavage.50 This mechanism, which does not require a phenolic
oxidation step, would become operative under forcing conditions
such as those leading to the formation of 2.
Experimental Section
Preparation of 3,5-Dihydroxybenzaldoxime. 3,5-Dihydroxy-
benzaldoxime was prepared by a general procedure reported in the
literature.44 Briefly, to 3,5-dihydroxybenzaldehyde (50 mg, 0.36
mmol) dissolved in water (4.2 mL) was added a solution of NH2-
OH‚HCl (14 mg, 0.20 mmol) and CH3COONa × 3 H2O (27 mg,
0.20 mmol) in water (4.8 mL), and the mixture was taken under
stirring at 80 °C. After 2 h, the mixture was cooled and extracted
with ethyl acetate (3 × 3 mL). The combined organic extracts were
dried over Na2SO4 and evaporated to dryness. The residue was
dissolved in ethyl acetate and fractionated by preparative TLC to
give 3,5-dihydroxybenzaldoxime (Rf 0.43, 30 mg, 54% yield).
3,5-Dihydroxybenzaldoxime: 1H NMR δ 6.38 (1H, t, J ) 2.0
Hz), 6.63 (2H, d, J ) 2.0 Hz), 7.96 (1H, s); ESI+/MS:m/z 154
([M + H]+). 3,5-Dihydroxybenzaldoxime was reacted with NaNO2
under the same conditions as for 1a, and the reaction mixture was
periodically analyzed by HPLC (gradient elution: water, solvent
A; acetonitrile, solvent B; from 2 to 30% B, 0-25 min; from 30 to
60% B, 25-70 min; 60% B, 70-75 min, eluant A).
Reaction of 1a with NaNO2. General Procedure. To a solution
of 1a (10 mg, 44 µmol) in methanol (0.5 mL) was added 0.1 M
phosphate buffer (pH 3.0) (44 mL) followed by NaNO2 (15 mg,
0.22 mmol), and the mixture was taken under vigorous stirring at
room temperature. After 3 h, at complete consumption of the
substrate (HPLC analysis, eluant A), the mixture was extracted with
ethyl acetate (3 × 30 mL) and the combined organic layers were
dried over Na2SO4 and taken to dryness. The residue was dissolved
in methanol and analyzed by HPLC (eluant A), TLC, and LC/MS.
In other experiments, the reaction of 1a was run (i) as above with
The reaction pathways illustrated in Schemes 1 and 2 entail
that product distribution mirrors the relative concentrations of
reacting free radical species in the medium. The incomplete mass
balance, due to the presence of other ill-defined species that
escaped isolation and characterization, prevents a more detailed
mechanistic analysis, so it is possible that other reaction
pathways of 1a are operative. However, the above schemes
establish the central role of the 4′-OH group in directing the
-
main reaction pathways of 1a with acidic NO2 toward the
double bond, which is a most significant outcome of this study.
In conclusion, the chemistry described in this paper discloses
-
unusual reactions of phenolic stilbene antioxidants with NO2
in acidic medium, highlighting the hitherto overlooked suscep-
tibility of 1a,b to oxidative cleavage under mild conditions of
physiological relevance, and the formation of 7 under nonen-
zymatic conditions. Nitration of 1a is also of chemical interest
as it may provide an entry to novel functionalized stilbene
derivatives via proper manipulation of 3a and 4.
(48) Napolitano, A.; Camera, E.; Picardo, M.; d’Ischia, M. J. Org. Chem.
2000, 65, 4853-4860.
(49) Golding, P.; Powell, J. L.; Ridd, J. H. J. Chem. Soc., Perkin Trans.
2 1996, 813-819.
(50) Bryant, D. K.; Challis, B. C.; Iley, J. J. Chem. Soc., Chem. Commun.
1989, 1027-1028.
4252 J. Org. Chem., Vol. 71, No. 11, 2006

Acid-Promoted Reaction of ResVeratrol with Nitrite
SCHEME 2
1a at 3 × 10-6 or 25 × 10-6 M concentration, with 0.2 × 10-3
M
(3 × 50 mL), and the combined organic layers were dried over
Na2SO4 and taken to dryness. The residue was analyzed by HPLC
(eluant A).
NaNO2 added in eight portions at 15 min intervals, and at 37 °C,
(ii) under an argon atmosphere, and (iii) under an 18O2 atmosphere.
When required, Na15NO2 was used in the reaction of 25 × 10-6
M
Reaction of 1b with NaNO2. Isolation of (E)-3,3′,4,5′-Tet-
rahydroxy-â-nitrostilbene (3b) and 3,4-Dihydroxybenzaldehyde.
The reaction of 1b with NaNO2 was run as previously described,27
and the mixture was analyzed by HPLC (gradient elution: 3% TFA,
solvent A; acetonitrile, solvent B; from 2 to 30% B, 0-15 min;
from 30 to 60% B, 15-45 min; 60% B, 45-55 min, eluant B).
For preparative purposes the reaction was carried out using 45 mg
of the starting material. After workup of the reaction mixture, the
residue (65 mg) was fractionated by preparative HPLC (3% TFA-
acetonitrile, 70:30 v/v) to give 3b27(tR 18.0 min eluant B, 36 mg,
68% yield) and 3,4-dihydroxybenzaldehyde (tR 11.0 min eluant B,
5 mg, 19% yield).
Isolation of rac-(2R,3R)-3-(3,5-Dihydroxyphenyl)-2-(4-hy-
droxyphenyl)-2,3-dihydrobenzofuran-5-carbaldehyde (2), (E)-
3,4′,5-Trihydroxy-R-nitrostilbene (3a), (E)-3,4′,5-Trihydroxy-3′-
nitrostilbene (4),31 (E)-3,4′,5-Trihydroxy-2,3′-dinitrostilbene (5),
3,5-Dihydroxybenzaldehyde, (3,5-Dihydroxyphenyl)nitromethane,
4-Hydroxybenzaldehyde, and 4-Hydroxy-3-nitrobenzaldehyde.
For preparative purposes, the reaction of 1a with NaNO2 was carried
out as in the general procedure using 400 mg of starting material.
After workup of the reaction mixture, the residue (380 mg) was
fractionated by preparative TLC to give 3a (Rf 0.36, 18 mg, 4%
yield, >95% purity), 229 (Rf 0.40, 10 mg, 2% yield, >90%purity),
4 (Rf 0.55, 5 mg, 1% yield, >98% purity), 4-hydroxybenzaldehyde
(Rf 0.69, 8 mg, 4% yield), 5 (Rf 0.78, 4 mg, 1% yield, >90% purity),
and 4-hydroxy-3-nitrobenzaldehyde (Rf 0.84, 4 mg, 1% yield). The
1a and the mixture was worked up as above and directly analyzed
by NMR and LC/MS. For kinetic experiments 1a (2.5 × 10-5 M)
was reacted with 1 × 10-3 M NaNO2 added in one portion. In
control experiments, the reaction was carried out under the
conditions of the general procedure without added NaNO2. Reaction
of 1a (3.5 × 10-2 M) with NaNO2 (0.35 M) was also run in
acetonitrile containing 2.5% acetic acid; the reaction course was
followed by HPLC (eluant A). Reaction of 3,4′,5-trimethoxystilbene
(2.5 × 10-4 M) with NaNO2 (1 × 10-3 M) was carried out at pH
3.0, and the reaction course was followed by HPLC (gradient
elution: water, solvent A; acetonitrile, solvent B; from 20 to 80%
B, 0-45 min; 80% B, 45-55 min).
Reaction of 1a with CAN. To a solution of 1a (10 mg, 44 µmol)
in methanol (0.5 mL) was added 0.1 M phosphate buffer (pH 3.0)
(175 mL) followed by CAN (24 mg, 44 µmol), and the mixture
was taken under vigorous stirring. After 1 h, the mixture was
extracted with ethyl acetate (3 × 50 mL) and the combined organic
layers were dried over Na2SO4 and taken to dryness. The residue
was analyzed by HPLC (eluant A).
Reaction of 1a with K3Fe(CN)6. To a solution of 1a (10 mg,
44 µmol) in methanol (0.5 mL) was added 0.1 M phosphate buffer
(pH 7.0) (175 mL) followed by K3Fe(CN)6 (14 mg, 44 µmol), and
the mixture was taken under stirring. After 1 h, the mixture was
acidified with 0.5 M HCl to pH 3 and extracted with ethyl acetate
J. Org. Chem, Vol. 71, No. 11, 2006 4253

Panzella et al.
fraction (5 mg) eluting at Rf 0.48 was found to consist of
3,5-dihydroxybenzaldehyde and (3,5-dihydroxyphenyl)nitromethane.
3a, 4, 3,5-dihydroxybenzaldehyde, 4-hydroxybenzaldehyde, or
the Rf 0.48 band was exposed to NaNO2 under the standard reaction
conditions, and the products formed were analyzed by HPLC (eluant
A) and TLC.
3a. UV λmax: CH3OH, 276, 356 nm; CH3OH/0.1 M NaHCO3,
pH 8, 302, 451 nm. 1H and 13C NMR: see Table 1. HR ESI-/MS:
found m/z 272.0563 ([M - H]-), calcd for C14H10NO5 m/z
272.0559.
4. UV λmax: CH3OH 303, 323, 396 nm; CH3OH/0.1 M NaHCO3,
pH 8, 331, 464 nm. 1H and 13C NMR: see Table 1. 1H NMR (CD3-
OD): δ 6.20 (1H, t, J ) 2.0 Hz), 6.49 (2H, d, J ) 2.0 Hz), 6.96
(1H, d, J ) 16.4 Hz), 7.01 (1H, d, J ) 16.4 Hz), 7.13 (1H, d, J )
8.8 Hz), 7.83 (1H, dd, J ) 8.8, 2.0 Hz), 8.14 (d, 1H, J ) 2.0 Hz).
HR ESI-/MS: found m/z 272.0555 ([M - H]-), calcd for C14H10-
NO5 m/z 272.0559.
using 50 mg of the starting material. After workup of the reaction
mixture, the residue (45 mg) was fractionated on a Sephadex LH-
20 column (50 cm × 2 cm) using 95% ethanol as the eluant.
Fractions were collected on the basis of HPLC analysis (eluant A)
and further purified by preparative HPLC (water-acetonitrile, 70:
30 v/v) to give 715 (tR 24.2 min eluant A, 1 mg, 1% yield, >95%
purity).
Isolation of trans-ꢀ-Viniferin. Wood from grapewine plants
infected with fungi associated with esca (Phaeoacremonium aleo-
philum, Phaeomoniella chlamydospora, Fomitiporia mediterranea)
was lyophilized and milled under nitrogen to obtain a fine powder.
After treatment with petroleum ether 40-60 to remove lipids, the
powder was extracted twice (3 h and overnight) with methanol (1:
15 w/v) in the dark with stirring. The extracts were filtered and
taken to dryness. The residue was dissolved in methanol and
purified by silica gel column chromatography using chloroform-
methanol, 80:20 v/v, as the eluant. The fraction containing trans-
ꢀ-viniferin (HPLC analysis: water-acetonitrile gradient from 80:
20 to 33:67 in 90 min, tR 56.9 min) was further fractionated by
preparative TLC (chloroform-methanol, 80:20 v/v) to give the pure
compound33 (Rf 0.51).
5. UV λmax: CH3OH 300, 394 nm; CH3OH/0.1M NaHCO3, pH
1
8, 317, 399 nm. H and 13C NMR: see Table 1. HR ESI+/MS:
found m/z 319.0561 ([M + H]+), calcd for C14H11N2O7 m/z
319.0566; found m/z 341.0380 ([M + Na]+), calcd for C14H10N2O7-
Na m/z 341.0386.
Rf 0.48 Band. 1H NMR resonances for (3,5-dihydroxyphenyl)ni-
tromethane: δ 5.48 (2H, s), 6.42 (1H, t, J ) 2.0 Hz), 6.50 (2H, d,
J ) 2.0 Hz). 13C NMR resonances for (3,5-dihydroxyphenyl)ni-
tromethane: δ 81.8 (CH2), 103.9 (CH), 110.6 (2 × CH), 135.0
(C), 160.8 (2 × C). LC/ESI+/MS: tR 13.9 min, m/z 192 ([M +
Na]+). HR ESI+MS: found 192.0279 ([M + Na]+), calcd for C7H7-
NO4Na m/z 192.0273.
Isolation of rac-5-{(2R, 3R)-2-(4-Hydroxyphenyl)-5-[(1E)-2-
(3,5-dihydroxyphenyl)vinyl]-2,3-dihydrobenzofuran-3-yl}benzene-
1,3-diol (6). For preparative purposes the reaction of 1a (2.5 ×
10-5 M) with NaNO2 (8 molar equiv) was carried out using 50 mg
of starting material. After workup of the reaction mixture, the
residue (45 mg) was fractionated by preparative TLC to give 617,18
(Rf 0.09, 2 mg, 4% yield, purity >95%), 3a (2 mg, 3% yield), 4 (3
mg, 5% yield), and 4-hydroxybenzaldehyde (1 mg, 4% yield). 6
was exposed to NaNO2 under the standard reaction conditions, and
the products formed were analyzed by HPLC (eluant A) and TLC.
Isolation of rac-5,5′-[(2R,3R,4R,5S)-2,5-Bis(4-hydroxyphenyl)-
tetrahydrofuran-3,4-diyl]bis(benzene-1,3-diol) (7). The reaction
of 1a (2.5 × 10-5 M) with NaNO2 (8 molar equiv) was carried out
Acknowledgment. This study was carried out in the frame
of the MIUR projects “Sostanze naturali ed analoghi sintetici
con attivita` antitumorale” (Grant PRIN 2003) and was supported
in part by a grant from the Italian Ministry of Agrarian and
Forestal Politics-Tuscany Region (ARSIA). We thank the
“Centro Interdipartimentale di Metodologie Chimico-Fisiche”
(CIMCF, University of Naples Federico II) for NMR, mass,
and computational facilities. We thank Miss Silvana Corsani
for technical assistance.
Supporting Information Available: General experimental
1
1
methods, H NMR spectra of compound 2, 5, 6, and 7, H NMR,
13C NMR (and/or 13C DEPT), H,13C HMQC, and H,13C HMBC
1
1
spectra of compound 3a and 4, and H NMR, 13C NMR, H,1H
1
1
COSY, and H,13C HMBC spectra of a reaction mixture of 1a
1
obtained with 15NO2-. This material is available free of charge via
the Internet at http://pubs.acs.org.
JO060482I
4254 J. Org. Chem., Vol. 71, No. 11, 2006