ArticlePDF AvailableLiterature Review

Role of resveratrol in prevention and therapy of cancer: Preclinical and clinical studies

Authors:

Abstract and Figures

Resveratrol, trans-3,5,4'-trihydroxystilbene, was first isolated in 1940 as a constituent of the roots of white hellebore (Veratrum grandiflorum O. Loes), but has since been found in various plants, including grapes, berries and peanuts. Besides cardioprotective effects, resveratrol exhibits anticancer properties, as suggested by its ability to suppress proliferation of a wide variety of tumor cells, including lymphoid and myeloid cancers; multiple myeloma; cancers of the breast, prostate, stomach, colon, pancreas, and thyroid; melanoma; head and neck squamous cell carcinoma; ovarian carcinoma; and cervical carcinoma. The growth-inhibitory effects of resveratrol are mediated through cell-cycle arrest; upregulation of p21Cip1/WAF1, p53 and Bax; down-regulation of survivin, cyclin D1, cyclin E, Bcl-2, Bcl-xL and clAPs; and activation of caspases. Resveratrol has been shown to suppress the activation of several transcription factors, including NF-kappaB, AP-1 and Egr-1; to inhibit protein kinases including IkappaBalpha kinase, JNK, MAPK, Akt, PKC, PKD and casein kinase II; and to down-regulate products of genes such as COX-2, 5-LOX, VEGF, IL-1, IL-6, IL-8, AR and PSA. These activities account for the suppression of angiogenesis by this stilbene. Resveratrol also has been shown to potentiate the apoptotic effects of cytokines (e.g., TRAIL), chemotherapeutic agents and gamma-radiation. Phamacokinetic studies revealed that the target organs of resveratrol are liver and kidney, where it is concentrated after absorption and is mainly converted to a sulfated form and a glucuronide conjugate. In vivo, resveratrol blocks the multistep process of carcinogenesis at various stages: it blocks carcinogen activation by inhibiting aryl hydrocarbon-induced CYP1A1 expression and activity, and suppresses tumor initiation, promotion and progression. Besides chemopreventive effects, resveratrol appears to exhibit therapeutic effects against cancer. Limited data in humans have revealed that resveratrol is pharmacologically quite safe. Currently, structural analogues of resveratrol with improved bioavailability are being pursued as potential therapeutic agents for cancer.
Content may be subject to copyright.
Abstract. Resveratrol, trans-3,5,4'-trihydroxystilbene, was first
isolated in 1940 as a constituent of the roots of white hellebore
(Veratrum grandiflorum O. Loes), but has since been found
in various plants, including grapes, berries and peanuts.
Besides cardioprotective effects, resveratrol exhibits anticancer
properties, as suggested by its ability to suppress proliferation
of a wide variety of tumor cells, including lymphoid and
myeloid cancers; multiple myeloma; cancers of the breast,
prostate, stomach, colon, pancreas, and thyroid; melanoma;
head and neck squamous cell carcinoma; ovarian carcinoma;
and cervical carcinoma. The growth-inhibitory effects of
resveratrol are mediated through cell-cycle arrest; up-
regulation of p21
Cip1/WAF1
, p53 and Bax; down-regulation of
survivin, cyclin D1, cyclin E, Bcl-2, Bcl-x
L
and cIAPs; and
activation of caspases. Resveratrol has been shown to suppress
the activation of several transcription factors, including NF-
Î B, AP-1 and Egr-1; to inhibit protein kinases including IÎ
kinase, JNK, MAPK, Akt, PKC, PKD and casein kinase II;
and to down-regulate products of genes such as COX-2,
5-LOX, VEGF, IL-1, IL-6, IL-8, AR and PSA. These
2783
Correspondence to: Bharat B. Aggarwal, Cytokine Research
Laboratory, Department of Bioimmunotherapy, The University of
Texas M. D. Anderson Cancer Center, Box 143, 1515 Holcombe
Boulevard, Houston, Texas 77030, U.S.A. Tel: 713-792-3503/6459,
Fax: 713-794-1613, e-mail: aggarwal@mdanderson.org
Key Words: Resveratrol, cell signaling, chemoprevention, metastasis,
transformation, invasion, tumorigenesis, apoptosis, review.
Abbreviations: TNF, tumor necrosis factor; NF-ÎB, nuclear factor kappa
B; PKC, protein kinase C; UV, ultraviolet; NOS, nitric oxide synthase;
COX, cyclooxygenase; PMA, phorbol myristate acetate; LDL, low-
density lipoprotein; PBMC, peripheral blood mononuclear cells; PMN,
human polymorphonuclear leukocytes; GSH, reduced glutathione; AP-1,
activator protein-1; MAPK, mitogen-activated protein kinase; ERK,
extracellular signal-regulated kinase; TGF, transforming growth factor;
PKA, protein kinase A; DMBA, 7,12-dimethylbenzoic acid; B[a]P,
benzo[·]pyrene; BPDE, B[a]P diol epoxides; AhR, aryl hydrocarbon
receptor; PhiP, 2-amino1-methyl-6-phenylimidazo[4,5-b]pyridine; AOM,
azoxymethane; NNK, 4-(methyl-nitrososamine)-1-(3-pyridyl)-1-butanone;
ODC, ornithine decarboxylase; B-CLL, B-cell chronic lymphocytic
leukemia; CTL, cytotoxic T lymphocyte; NQO, NAD(P)H quinone
oxidoreductase; HMG-CoA, 3-hydroxy-3-methylglutaryl coenzyme A;
SBP, systolic blood pressure; EWP, extract of wine phenolics ; SMC,
smooth muscle cells; ROS, reactive oxygen
species; EGFR, epidermal
growth factor receptor; HUVEC, human umbilical vein endothelial
cells; 8-OHdG, 8-hydroxydeoxyguanosine; TBARS, thiobarbituric acid-
reactive substances; AAPH, 2,2'-azobis-(2-amidinopropane)
dihydrochloride; IC
50
, concentration causing 50% inhibition; ICV,
intracerebroventricular; STZ, streptozotocin; HMG, half-mustard gas;
LLC, Lewis lung carcinoma; VEGF, vascular endothelial growth factor;
BHA, butylated hydroxyanisole; ICAM, intracellular adhesion
molecule; VCAM, vascular cell adhesion molecule; MMP, matrix
metalloproteinase; IL, interleukin; PARP, poly(ADP-ribose)
polymerase; Egr, early growth response gene; ER, estrogen receptor;
CYP, cytochrome P450; IFN, interferon; NSAID, nonsteroidal anti-
inflammatory drug; H
2
O
2
, hydrogen peroxide; Cdk; cyclin-dependent
kinases; PDGF, platelet-derived growth factor; PSA, prostate-specific
antigen; ACF, aberrant crypt foci; Ach, acetylcholine; MDA,
malondialdehyde; SHRSP, stroke-prone hypertensive rats; Ïmax,
wavelength maxima; HPLC, high-pressure (performance) liquid
chromatography; MS, mass spectrometric; CoA, coenzyme A; NO,
nitric oxide; AIF, apoptosis-inducing factor; AML, acute myeloid
leukemia; DISC, death-inducing signal complex; AR, androgen
receptor; ALL, acute lymphocytic leukemia; Rb, retinoblastoma; SPT,
serine palmitoyltransferase; PDE, phosphodiesterase; AZT,
zidovudine; ddC, zalcitabine; ddI, didanosine; PKD, protein kinase D;
LPS, lipopolysaccharide; PI3K, phosphoinositide 3-kinase; TRAIL,
tumor necrosis factor-related apoptosis-inducing ligand; FADD, Fas-
associated death domain.
ANTICANCER RESEARCH 24: 2783-2840 (2004)
Review
Role of Resveratrol in Prevention and Therapy of Cancer:
Preclinical and Clinical Studies
BHARAT B. AGGARWAL
1
, ANJANA BHARDWAJ
1
, RISHI S. AGGARWAL
1
,
NAVINDRA P. SEERAM
2
, SHISHIR SHISHODIA
1
and YASUNARI TAKADA
1
1
Cytokine Research Laboratory, Department of Bioimmunotherapy,
The University of Texas M. D. Anderson Cancer Center, Box 143, 1515 Holcombe Boulevard, Houston, Texas 77030;
2
UCLA Center for Human Nutrition, David Geffen School of Medicine,
900 Veteran Avenue, Los Angeles, CA 90095-1742, U.S.A.
0250-7005/2004 $2.00+.40
activities account for the suppression of angiogenesis by this
stilbene. Resveratrol also has been shown to potentiate the
apoptotic effects of cytokines (e.g., TRAIL), chemotherapeutic
agents and Á-radiation. Phamacokinetic studies revealed that
the target organs of resveratrol are liver and kidney, where it is
concentrated after absorption and is mainly converted to a
sulfated form and a glucuronide conjugate. In vivo, resveratrol
blocks the multistep process of carcinogenesis at various
stages: it blocks carcinogen activation by inhibiting aryl
hydrocarbon-induced CYP1A1 expression and activity, and
suppresses tumor initiation, promotion and progression.
Besides chemopreventive effects, resveratrol appears to exhibit
therapeutic effects against cancer. Limited data in humans
have revealed that resveratrol is pharmacologically quite safe.
Currently, structural analogues of resveratrol with improved
bioavailability are being pursued as potential therapeutic
agents for cancer.
Contents
Introduction
A. Source of resveratrol
B. Chemistry of resveratrol
C. Preclinical Studies
C1. In vitro effects
C1a. Antiproliferative effects of resveratrol
B-cell lymphoma
T-cell lymphoma
Myeloid leukemia
Breast cancer
Colon cancer
Pancreatic cancer
Gastric cancer
Prostate cancer
Melanoma
Lung cancer
Liver cancer
Thyroid and head and neck cancers
Ovarian and endometrial tumors
C1b. Resveratrol induces apoptosis
Fas/CD95 pathway
Mitochondrial pathway
Rb-E2F/DP pathway
p53 activation pathway
Ceramide activation pathway
Tubulin polymerization pathway
Adenylyl-cyclase pathway
C1c. Suppression of NF-Î B activation by resveratrol
C1d. Suppression of AP-1 by resveratrol
C1e. Suppression of Egr-1 by resveratrol
C1f. Suppression of mitogen-activated protein kinases by
resveratrol
C1g. Suppression of protein kinases by resveratrol
C1h. Suppression of NO/NOS by resveratrol
C1i. Suppression of growth factor protein tyrosine kinases
by resveratrol
C1j. Suppression of COX-2 and lipooxygenase by
resveratrol
C1k. Suppression of cell-cycle proteins by resveratrol
C1l. Suppression of adhesion molecules by resveratrol
C1m. Suppression of androgen receptors by resveratrol
C1n. Suppression of PSA by resveratrol
C1o. Suppression of inflammatory cytokine expression by
resveratrol
C1p. Suppression of angiogenesis, invasion and metastasis
by resveratrol
C1q. Effect of resveratrol on bone cells
C1r. Effects of resveratrol on expression of cytochrome
p450 and metabolism of carcinogens
C1s. Suppression of inflammation by resveratrol
C1t. Antioxidant effects of resveratrol
C1u. Suppression of transformation by resveratrol
C1v. Induction of cellular differentiation by resveratrol
C1w. Estrogenic/antiestrogenic effects of resveratrol
C1x. Effect of resveratrol on normal cells
C1y. Suppression of mutagenesis by resveratrol
C1z. Radioprotective and radiosensitive effect of resveratrol
C1aa. Chemosensitization by resveratrol
C1ab. Direct targets of resveratrol
C1ac. Immunomodulatory effects of resveratrol
C1ad. Modulation of gene expression by resveratrol
C2. In vivo animal studies of resveratrol
C2a. Metabolism, pharmacokinetics, tissue distribution and
clearance of resveratrol
C2b. Chemopreventive effects of resveratrol in animals
C2c. Antitumor effects of resveratrol in animals
D. Clinical studies with resveratrol
Conclusion
References
Introduction
The history of resveratrol can be traced back thousands of
years. Perhaps the first known use of grape extracts for
human health can be dated over 2000 years ago, to
"darakchasava", a well-known Indian herbal preparation of
which the main ingredient is Vitis vinifera L. This
"Ayurvedic" medicine is prescribed as a cardiotonic and also
given for other disorders (1). The use of dried grapes (also
called manakka) as a cardiotonic is well documented. High-
performance liquid choromatography (HPLC) analysis of
darakchasava revealed polyphenols such as resveratrol and
pterostilbene. This age-old formulation became interesting
in the light of recently acquired knowledge on resveratrol.
ANTICANCER RESEARCH 24: 2783-2840 (2004)
2784
Resveratrol (3,5,4’-trihydroxystilbene) is a naturally
occurring phytoalexin produced by a wide variety of plants,
such as grapes (Vitis vinifera), peanuts (Arachis hypogea),
and mulberries in response to stress, injury, ultraviolet (UV)
irradiation, and fungal (e.g., Botrytis cinerea) infection.
Although phytoalexins have long been inferred to be
important in the defense of plants against fungal infection,
few reports show that they provide resistance to infection.
Several plants, including grapevine, synthesize the stilbene-
type phytoalexin resveratrol when attacked by pathogens.
Stilbenes with fungicidal potential are formed in several
unrelated plant species, such as peanut, grapevine, and pine
(Pinus sylvestris) (Figure 1). Stilbene biosynthesis specifically
requires the presence of stilbene synthase. Furthermore, the
precursor molecules for the formation of hydroxy-stilbenes
are malonyl-coenzyme A (CoA) and p-coumaroyl-CoA,
both present in plants. Hain et al. isolated the stilbene
synthase gene from grapevine, transferred it into tobacco,
and found that regenerated tobacco plants containing this
gene are more resistant to infection by Botrytis cinerea (2).
Resveratrol was first identified in 1940 as a constituent of
the roots of white hellebore (Veratrum grandiflorum O. Loes),
and later in the dried roots of Polygonum cuspidatum, called
Ko-jo-kon in Japanese, which is used in traditional Chinese
and Japanese medicine to treat suppurative dermatitis,
gonorrhea favus, athlete’s foot (tinea pedis), and hyperlipemia
(3-6). In 1976, resveratrol was detected in the leaf epidermis
and the skin of grape berries but not in the flesh (7-9). Fresh
grape skins contain 50-100 mg resveratrol per gram, and the
concentration in wine ranges from 0.2 mg/l to 7.7 mg/l. The
epidemiological finding of an inverse relationship between
consumption of red wine and incidence of cardiovascular
disease has been called the "French paradox" (10, 11). For a
variety of reasons, the cardioprotective effects of red wine
Aggarwal et al: Resveratrol Inhibits Tumorigenesis
2785
Figure 1. Sources of resveratrol from different plants.
ANTICANCER RESEARCH 24: 2783-2840 (2004)
2786
Table I. Sources of Resveratrol and its analogues.
Compound Sources References
Resveratrol Japanese knotweed (Polygonum cuspidatum); Vitis spp. (incl. (26-28, 33-38)
(trans-3,5,4’–trihydroxystilbene) grape-vines, leaves and berryskin); Vaccinum spp. (incl. blueberry,
bilberry, cranberry); Morus spp. (incl. mulberry); Lily (Veratrum spp.);
Legumes (Cassia spp., Pterolobium hexapetallum); Peanuts (Arachis
hypogaea); Rheum spp.(incl. Rhubarb); Eucalyptus; Spruce (Picea spp.);
Pine (Pinus spp.); Poaceae (grasses incl. Festuca, Hordeum, Poa, Stipa
and Lolium spp.); Trifolium spp.; Nothofagus spp.; Artocarpus spp.;
Gnetum spp.; Pleuropterus ciliinervis; Bauhinia racemosa; Paeonia
lactiflora; Scilla nervosa; Tetrastigma hypoglaucum; Synthetic
Dihydroresveratrol Dioscorea spp.; Bulbophyllum triste; Synthetic (39, 40)
(trans-3,5,4’–
trihydroxybibenzylstilbene)
Piceatannol or astringinin White tea tree (Melaleuca leucadendron); Asian legume (28, 40-45)
(trans-3,4,3’,5’- (Cassia garrettiana), C. marginata; Rhubarb (Rheum spp.);
tetrahydroxystilbene) Euphorbia lagascae; Polygonum cuspidatum; Vitis vinifera
Dihydropiceatannol Cassia garrettiana; Synthetic (42)
(trans-3,4,3’,5’-
tetrahydroxybibenzylstilbene)
Gnetol (trans-2,6,3’, Gnetum spp. (incl. G. monatum, G. africanum, G. gnemon, G. ula) (36, 46, 47)
5’,-tetrahydroxystilbene)
Oxyresveratrol (trans-2,3’,4, Morus spp.; Maclura pomifera; Artocarpus gomezianus; (38, 48-50)
5’-tetrahydroxystilbene) Schoenocaulon officinale
Hydroxyresveratrol (trans-2,3,5, Polygonum cuspidatum (28)
4’–tetrahydroxystilbene)
Trans-3,4,5, Synthetic (51)
4’–tetrahydroxystilbene
Trans-3,3’,4’,5, Eucalyptus wandoo; Vouacapoua americana, V. macropetala; Synthetic (52, 53)
5’-pentahydroxystilbene
Pinosylvin (trans-3,) Gnetum cleistostachyum; Alpinia katsumadai; Polyalthia longifolia; (51, 54-59, 361)
5-dihydroxystilbene Polygonum nodosum; Pinus spp.(incl. Scottish pine, P. sylvestris); Synthetic
Dihydropinosylvin (trans-3, Dioscorea batatas; Synthetic (60-62)
5-dihydroxybibenzylstilbene)
Trans-2,4,4'-trihydroxystilbene Synthetic (61, 62)
Trans-3,5,3'-trihydroxystilbene Synthetic (63, 64)
Trans-3,4,5-trihydroxystilbene Synthetic (65)
Trans-3,4,4'-trihydroxystilbene Synthetic (65, 66)
Trans-3,4-dihydroxystilbene Synthetic (61, 62, 66)
Trans-3,4'-dihydroxystilbene Synthetic (63, 64)
Trans-3,3'-dihydroxystilbene Synthetic (63, 64)
Trans-2,4-dihydroxystilbene Synthetic (61, 62)
Trans- 4,4'-dihydroxystilbene Synthetic (61, 62, 65, 66)
continued
Aggarwal et al: Resveratrol Inhibits Tumorigenesis
2787
Table I. continued.
Compound Sources References
Trans-3-hydroxystilbene Synthetic (63, 64)
Trans-4-hydroxystilbene Synthetic (61, 62, 65)
(p-hydroxystilbene)
Trans-halogenated-3,5, Synthetic (67, 68)
4’–trihydroxystilbenes
(fluoro-, chloro- and
iodo-resveratrols)
Dimethoxypinosylvin Synthetic (51)
(trans-3,5-dimethoxystilbene)
Rhapontigenin or Rheum spp. (incl. R. rhaponticum, R. undulatum); (35, 69, 70)
3-methoxyresveratrol Scilla nervosa; Synthetic
(trans-3,5,3',-trihydroxy-
4'-methoxystilbene)
Isorhapontigenin (trans-3,5,4',- Gnetum spp.; Belamcanda chinensis; Synthetic (36, 71, 72)
trihydroxy-3'-methoxystilbene)
Desoxyrhapontigenin or Gnetum cleistostachyum; Rheum undulatum; (54, 73-75)
4-methoxyresveratrol Knema austrosiamensis; Rumex bucephalophorus
(trans-3,5-dihydroxy-
4'-methoxystilbene)
Pinostilbene or Rumex bucephalophorus (75)
3-methoxyresveratrol
(trans-5,4'-dihydroxy-
3-methoxystilbene)
Trans-3,4'-dimethoxy- Knema austrosiamensis; Synthetic (73, 74)
5-hydroxystilbene
Cis-3,5,3',-trihydroxy- Synthetic (76)
4'-methoxystilbene
Trimethylresveratrol Pterolobium hexapetallum; Synthetic (37 , 51, 77)
(trans-3,5,4’–trimethoxystilbene)
Gnetucleistol D or Gnetum cleistostachyum (54)
2-methoxyoxyresveratrol
(trans-2-methoxy-3’,4,
5-trihydroxystilbene)
Gnetucleistol E or Gnetum cleistostachyum (54)
3-methoxy-isorhapontigenin
(trans-3,3'-dimethoxy-5,
4'-dihydroxystilbene)
Trans- and cis-3,5, Synthetic (76)
4’-trimethoxy-3’-hydroxystilbene
Trans- and cis-3,5, Synthetic (76)
3’-trimethoxy-4’-hydroxystilbene
Trans- and cis-3,5-dimethoxy-3’, Synthetic (76)
4’-dihydroxystilbene
continued
ANTICANCER RESEARCH 24: 2783-2840 (2004)
2788
Table I. continued.
Compound Sources References
Trans- and cis-3,5-dihydroxy-3’- Synthetic (76)
amino-4’-methoxystilbene
Trans- and cis-3,5-dimethoxy- Synthetic (76)
4’-aminostilbene
Trans-and cis-3,4’,5-trimethoxy- Synthetic (76)
3’-aminostilbene
Trans-and cis-3, Synthetic (76)
5-dimethoxy-4’-nitrostilbene
Trans-and cis-3,4’, Synthetic (76)
5-trimethoxy-3’-nitrostilbene
Trans-5,4'-dihydroxy- Rumex bucephalophorus (75)
3-methoxystilbene
Pterostilbene (trans-3, Dracena cochinchinensis; Pterocarpus spp. (37, 76, 78)
5-dimethoxy-4'-hydroxystilbene) (incl. P. santalinus, P marsupium); Vitis vinifera;
Pterolobium hexapetallum; Synthetic
Cis-3,5-dimethoxy-4'-hydroxystilbene Synthetic (76)
3,4,5,4'-tetramethoxystilbene Synthetic (51)
3,4,5,3'-tetramethoxystilbene Synthetic (51)
3,4,5,3',4'-pentamethoxystilbene Synthetic (51)
Trans-3,4,3',5'-tetra methoxystilbene Crotalaria madurensis (80)
Trans-and cis-3,3',5, Yucca periculosa, Y. schidigera; (81-83)
5'-tetrahydroxy-4-methoxystilbene Cassia pudibunda
Trans-4,4'-dihydroxystilbene Yucca periculosa (81)
Trans-3-hydroxy-5-methoxystilbene Cryptocarya idenburgensis (84)
Trans-4,3'-dihydroxy- Dracaena loureiri (85)
5'-methoxystilbene
Trans-4-hydroxy-3', Dracaena loureiri, D. cochinchinensis (85, 86)
5'-dimethoxystilbene
Piceid or polydatin or resveratrol-3- Polygonum cuspidatum; Rheum rhaponticum; (27, 35, 87, 88)
glucoside (trans-3,5, Picea spp.; Lentils (Lens culinaris)
4'-trihydroxystilbene-3-
O-‚-D-glucopyranoside)
Rhapontin or rhaponticin Rheum spp.; Eucalyptus (27, 35)
(trans-3,3',5-trihydroxy-4'-
methoxystilbene -3-O-‚-D-
glucopyranoside)
Deoxyrhapontin Rheum rhaponticum (35)
(trans-3,5-dihydroxy-4'-
methoxystilbene-3-O-‚-D-
glucopyranoside)
Isorhapontin Pinus sibirica; Picea spp. (35, 87)
(trans-3,4',5-trihydroxy-3'
methoxystilbene-3-
O-‚-D-glucopyranoside)
continued
Aggarwal et al: Resveratrol Inhibits Tumorigenesis
2789
Table I. continued.
Compound Sources References
Piceatannol glucoside Rheum rhaponticum; (27, 35)
(3,5,3',4'-tetrahydroxystilbene-4'- Polygonum cuspidatum; Spruce
O-‚-D-glucopyranoside)
Pinostilbenoside Pinus koraiensis (89)
(trans-3-methoxy-5-hydroxystilbene-
4'-O-‚-D-glucopyranoside)
Resveratroloside or resveratrol-4'- Polygonum cuspidatum; Pinus spp.; (27, 28, 35, 90)
glucopyranoside (trans-3,5,4'- Vitis vinifera
trihydroxystilbene-4'-O-‚-
D-glucopyranoside)
Astringin (trans-3,4,3',5'- Picea spp., Vitis vinifera (28, 87, 90)
tetrahydroxystilbene-3'-O-‚-
D-glucopyranoside)
Piceid-2''-O-gallate and -2''- Pleuropterus ciliinervis (91)
O-coumarate
Rhaponticin-2''-O- Rhubarb (Rheum undulatum) (92)
gallate and -6''-O-gallate
Piceatannol-6''-O-gallate Chinese rhubarb (Rhei rhizoma) (93)
Cis-resveratrol-3,4'-O-‚-diglucoside Vitis vinifera (cell suspension culture) (94)
Combretastatins and their glycosides Synthetic (95)
(e.g. combretastain A= trans-2',3'-
dihydroxy-3,4,4',
5-tetramethoxystilbene)
5-methoxy-trans-resveratrol-3- Elephantorrhiza goetzei (96)
O-rutinoside
Oxyresveratrol-2-O-‚- Schoenocaulon officinale (50)
glucopyranoside
Resveratrol-3,4'-O,O'-di-‚- Schoenocaulon officinale (50)
D-glucopyranoside
Mulberrosides (e.g. cis- Morus alba (cell cultures), Morus lhou (97, 98)
oxyresveratrol diglucoside)
Gnetupendins (isorhapontigenin Gnetum pendulum, G. gnemon (98, 99)
dimer glucosides); Gnemonosides
(resveratrol oligomer glucosides)
Gaylussacin Gaylussacia baccata, (100)
[5-(b-D-glucosyloxy)- G. frondosa
3-hydroxy-trans-stilbene-2-
carboxylic acid]
Resveratrol oligomers and Dipterocarpaceae, Gnetaceae, Vitaceae, Cyperaceae (6, 101-103)
oligostilbenes (incl. viniferins) and Leguminosae plants (incl. Vatica pauciflora, V. rassak,
V. oblongifolia; Vateria indica; Shorea laeviforia,
S. hemsleyana; Paeonia lactiflora; Sophora moorcroftiana,
S. leachiana; Gnetum venosum; Cyperus longus;
Upuna borneensis; Iris clarkei
1,5,7-trimethoxy-9,10 Nidema boothii (104)
dihydrophenanthrene-2,6-diol
have been attributed to resveratrol (12). These effects include
suppression of lipid peroxidation and eicosanoid synthesis,
inhibition of platelet aggregation, and antioxidant, anti-
inflammatory and vasorelaxant activities (13). Numerous
reports indicate that resveratrol has antiviral effects against
HIV-1 (14) and the herpes simplex virus (15, 16). Heredia et
al. reported that resveratrol synergistically enhances the anti-
HIV-1 activity of the nucleoside analogues zidovudine (AZT),
zalcitabine (ddC) and didanosine (ddI) (14).
Resveratrol also exhibits antibacterial effects (17),
including inhibition of growth of different strains of
Helicobacter pylori (18-20).
Extensive research during the last two decades has
suggested that, besides cardioprotective effects, resveratrol
also exhibits anticancer activities. How resveratrol manifests
its anticancer properties, the cell signaling pathways
affected, the transcription factors modulated, the genes
induced, the enzyme activities regulated, the protein
interactions, and the types of in vitro and in vivo model
systems in which resveratrol has been examined are the
focus of this review. Although several reviews have been
written on resveratrol (21-28), none covers the aspects of
this polyphenol discussed here.
A. Sources of Resveratrol
That red grapes or red wine are sources of resveratrol is well
known (29). However, resveratrol has been identified in a
wide variety of plants, including Japanese knotweed
(Polygonum cuspidatum) (4); the peanut (30, 31); Vaccinum
spp. (including blueberry, bilberry, and cranberry) (32, 33);
Reynoutria japonica; and Scots pine (Figure 1). Other plant
sources of resveratrol include Vitis spp. (including grapevines,
leaves, and berryskins); Morus spp. (including mulberry); lilies
(Veratrum spp.); legumes (Cassia spp., Pterolobium
hexapetallum); Rheum spp. (including rhubarb); eucalyptus;
spruce (Picea spp); pine (Pinus spp.); grasses (Poaceae
including Festuca, Hordeum, Poa, Stipa and Lolium spp.);
Trifolium spp.; Nothofagus spp.; Artocarpus spp; Gnetum spp.;
Pleuropterus ciliinervis; Bauhinia racemosa; Paeonia lactiflora;
Scilla nervosa; and Tetrastigma hypoglaucum. Isorhapontigenin,
isolated from Belamcanda chinensis, is a derivative of stilbene.
Its chemical structure is very similar to that of resveratrol and
it has a potent anti-oxidative effect. Compounds that are
closely related to resveratrol structurally, and thus may have
similar biological effects, have been identified in a wide variety
of plants (Table I).
B. Chemistry of Resveratrol
Resveratrol (Figure 2) is found widely in nature, and a
number of its natural and synthetic analogues and their
isomers, adducts, derivatives and conjugates are known (6,
26-28, 33-104) (Table I). It is an off-white powder (extracted
by methanol) with a melting point of 253-255ÆC and
molecular weight of 228.25. Reveratrol is insoluble in water
but dissolves in ethanol and dimethylsulphoxide. The
stilbene-based structure of resveratrol consists of two
phenolic rings linked by a styrene double bond to generate
3,4’,5, -trihydroxystilbene. Although the presence of the
double bond facilitates trans- and cis-isomeric forms of
resveratrol [(E)- and (Z)-diasteromers, respectively], the
trans-isomer is sterically the more stable form (105). On
spectrophotometric analysis in ethanol, trans-resveratrol
absorbs maximally at 308 nm and cis-resveratrol at 288 nm,
which allows for their separation by HPLC with UV
detection (105, 106). Absorptivity is greater in an ethanol:
water solution (1:9 v/v), but with a small shift in Ïmax (trans-
resveratrol Ïmax, 306 nm; cis-resveratrol Ïmax, 286 nm).
Besides their differences in spectrophotometric UV
absorptions, trans- and cis-resveratrol are also clearly
distinguished by their chemical shifts in nuclear magnetic
resonance spectroscopy (106).
Trans-resveratrol is commercially available and converts
to the cis-form on exposure to UV irradiation (23, 24, 26-
28). Trela and Waterhouse conducted trials under various
conditions and showed that trans-resveratrol is stable for
months when protected from light, except in high pH
buffers (105). These workers also showed that the cis-isomer
is extremely light-sensitive but can remain stable in the dark
at ambient temperature in 50% ethanol for at least 35 days
over the range of 5.3-52.8 ÌM. Low pH also causes cis-
resveratrol to isomerize to trans-resveratrol. Recently, Deak
and Falk studied the reactions of commercially obtained
trans-resveratrol and photochemically prepared cis-
resveratrol (106). The free enthalpy difference between the
two isomers was estimated to be of the order of that of
common stilbenes, with the trans-isomer being more stable
by about 11-14 KJ/mol. These workers also reported that
the pK
a
values of trans-resveratrol, corresponding to the
mono, di- and tri-protonation of the system, were 9.3, 10.0,
and 10.6, respectively. Resveratrol occurs predominantly as
the trans-isomer, and reports of the presence of the cis-
isomer, for example in certain wines, are attributed to
photoisomeric conversion, enzyme action during
fermentation, or release from resveratrol oligomers
(viniferins) (23, 24, 26-28). Since reports about the cis-
isomer are limited, when the structure of resveratrol is not
specified, we refer here to trans-resveratrol.
Over the past decade, several HPLC and gas
chromatographic methods have been developed to detect
the presence and measure levels of resveratrol and its
analogues (23, 24, 26-28). Much attention has been focused
on method development, since studying the biological
properties of resveratrol requires analyses of complex
mixtures containing very low amounts of stilbenes, and
ANTICANCER RESEARCH 24: 2783-2840 (2004)
2790
complete and quick extractions are required to minimize
losses from isomerization or denaturation. Generally, HPLC
methods using reverse phase C18 columns coupled with UV
detection (photodiode array or diode array detectors) can
adequately distinguish resveratrol isomers and their
analogues on the basis of their different absorbance
maxima. However, the use of mass spectrometry (MS)
fluorimetric and electrochemical detectors, which are more
specific than UV detection, has considerably improved
sensitivity and decreased sample size (23, 24). Gas
chromatographic methods, with or without MS detection,
although not as popular as HPLC, have been frequently
employed but require trimethylsilyl derivatization of
resveratrol and its analogues.
Since the first reported detection of trans-resveratrol in
grapevines in 1976, and later in wine in 1992, and its
implications in relation to the "French paradox" (7, 10, 107),
there has been an explosion of interest in the various
biological activities of this natural phytoalexin. Given the
substantial number of reports on natural and synthetic
analogues of resveratrol (Table I), considerable attention
has been focused on structure-activity relationship studies
of these compounds. Natural and synthetic resveratrol
analogues include a myriad of compounds differing in the
type, number and position of substituents (hydroxyl,
methoxyl, halogenated, glycosylated, esterified, etc.),
presence or absence of stilbenic double bonds, modified
steroisomery, and oxidative dimerizations (to form
oligomers). Calculations based on density functional theory
studies have been used to study the structure-activity
relationships of resveratrol in the chain reaction of auto-
oxidation (108). The 4’-hydroxyl group of resveratrol was
reported to be more reactive than the 3’- and 5’-hydroxyl
groups becase of resonance effects and, in conjunction with
the trans-olefin structure of the parent stilbene skeleton,
were the most important determinants of bioactivity (61-63,
108-110). Ashikawa et al. reported that piceatannol (a
tetrahydroxyl resveratrol analogue) was considerably
different in biological activity to the stilbene and
rhaponticin (a methoxylated and glucosylated analogue of
resveratrol) (111). Similarly, structure-activity relationship
studies have shown distinct biological properties of
Aggarwal et al: Resveratrol Inhibits Tumorigenesis
2791
Figure 2. Resveratrol and its various analogues/derivatives.
resveratrol oligomers and resveratrol glycosides (called
polydatins and piceids) (6, 26-28). Much attention has been
focused on the chemistry of resveratrol and its natural and
synthetic analogues because of their biological properties
and their potential in the prevention and therapy of cancer.
C. Preclinical Studies
C1. In vitro effects
C1a. Antiproliferative effects of resveratrol
Resveratrol has been shown to suppress proliferation of a
wide variety of tumor cells, including lymphoid and myeloid
cancers; breast, colon, pancreas, stomach, prostate, head
and neck, ovary, liver, lung and cervical cancers; melanoma;
and muscles (112-188) (Table II). Besides inhibiting
proliferation, resveratrol also induces apoptosis either
through the caspase-8-dependent pathway (receptor-
mediated; type I) or the caspase-9-dependent pathway
(mitochondrial; type II), or both. The mechanisms of
suppression of cell growth and induction of apoptosis for
these cell types are described here.
B-cell lymphoma: Several studies have shown the
antiproliferative effects of resveratrol on B cells (112-115).
Billard et al. investigated the effects of resveratrol on
leukemic cells from patients with chronic B-cell
malignancies and found that resveratrol had
antiproliferative effects and induced apoptosis in leukemic
B-cells that correlated with activation of caspase-3, a drop
in the mitochondrial transmembrane potential, reduction in
the expression of the anti-apoptotic protein Bcl-2, and
reduction in expression of inducible nitric oxide synthase
(iNOS) (112). In contrast, resveratrol had little effect on the
survival of normal peripheral blood mononuclear cells
(PBMC). Roman et al. reported apoptotic and growth-
inhibitory effects of resveratrol in human B-cell lines
derived from chronic B-cell malignancies (113). Resveratrol
inhibited the expression of the antiapoptotic proteins Bcl-2
and iNOS in WSU-CLL and ESKOL cells and cells derived
from patient with B-cell choronic lymphocytic leukemia
(B-CLL). Dorrie et al. showed that resveratrol induced
extensive apoptotic cell death not only in Fas/CD95-
sensitive leukemia lines, but also in B-lineage leukemic cells
that are resistant to Fas signaling (114). They also found
that resveratrol had no cytotoxicity against normal PBMC.
In each acute lymphocytic leukemia (ALL) cell line,
resveratrol induced progressive loss of mitochondrial
membrane potential and increase in caspase-9 activity. No
evidence of caspase-8 activation or Fas signaling was
observed. In BJAB Burkitt-like lymphoma cells, Wieder et
al. demonstrated that resveratrol-induced cell death
accompanied an increase in mitochondrial permeability
transition and caspase-3 activation and was independent of
the Fas signaling pathway (115). Resveratrol was also found
to induce apoptosis in leukemic lymphoblasts isolated from
patients suffering from childhood ALL.
T-cell lymphoma: Several reports indicate that resveratrol
modulates the growth of T cells (116, 117). Hayashibara et al.
showed that resveratrol inhibited growth in five HTLV-1-
infected cell lines (adult T-cell leukemia) and induced
apoptosis, which correlated with a gradual decrease in the
expression of survivin, an anti-apoptotic protein (116).
Tinhofer et al. showed that resveratrol induced apoptosis in
the CEM-C7H2 T-ALL cell line. They also found that
resveratrol induced apoptosis via a novel mitochondrial
pathway controlled by Bcl-2 (117) and that resveratrol-
induced apoptosis was inhibited by Bcl-2. Resveratrol
stimulation of C7H2 cells produced reactive oxygen species
(ROS), and this production was significantly reduced
by Bcl-2. As expected, pretreatment of cells with
N-acetylcysteine protected cells from DNA fragmentation
induced by resveratrol. Interestingly, resveratrol-induced
apoptosis did not involve cytochrome c release, nor trigger
activation of death receptor type II pathways, as no early
processing of Bid could be detected. Resveratrol, however,
caused activation of caspase-9, -2, -3 and -6 in the control
cells, but not in the subclones overexpressing Bcl-2. These
authors also found that DNA cleavage by resveratrol
occurred downstream of mitochondrial signaling and was
significantly blocked in the Bcl-2-overexpressing subclones.
After various proapoptotic stimuli, the loss of mitochondrial
transmembrane potential led to the release of apoptosis-
inducing factor (AIF) from the mitochondrial intermembrane
space, thus representing the link between mitochondria and
nucleus in resveratrol-induced apoptosis. Resveratrol,
however, did not induce translocation of AIF, suggesting that
this pathway of caspase-independent activation of nucleases is
not involved in resveratrol-induced apoptosis.
Myeloid leukemia: Resveratrol can induce apoptosis in
myeloid cells (118-127). Clement et al. showed that
resveratrol triggered Fas signaling-dependent apoptosis in
HL-60 human leukemia cells (118). Resveratrol-treated cells
exhibited increases in externalization of inner membrane
phosphatidylserine and in cellular content of subdiploid
DNA, indicating loss of membrane phospholipid asymmetry
and DNA fragmentation. Resveratrol-induced cell death
was mediated by intracellular caspases, as indicated by the
increase in proteolytic cleavage of caspase substrate poly
(ADP-ribose) polymerase (PARP) and the ability of caspase
inhibitors to block resveratrol cytotoxicity. Furthermore,
resveratrol treatment enhanced Fas ligand (FasLCD95L)
expression on HL-60 cells, and resveratrol-mediated cell
death was specifically Fas signaling-dependent. The
expression of FasL was not unique to HL-60 cells but also
ANTICANCER RESEARCH 24: 2783-2840 (2004)
2792
was induced on T47D breast carcinoma cells. Resveratrol
treatment of normal human PBMC did not affect cell
survival for as long as 72 h, which correlated with the
absence of a significant change in either Fas or FasL
expression on treated PBMC. These data showed specific
involvement of the Fas-FasL system in the anticancer
activity of resveratrol (Table III).
Tsan found that, in human monocytic leukemia THP-1
cells, resveratrol induced apoptosis independently of Fas
signaling (119). The effect of resveratrol on THP-1 cells was
reversible after its removal from the culture medium. Surh et
al. found that resveratrol inhibited proliferation and DNA
synthesis in human promyelocytic leukemia HL-60 cells
(120). Resveratrol-induced cell death was characterized by
internucleosomal DNA fragmentation, an increased proportion
of the subdiploid cell population, and a gradual decrease in the
expression of anti-apoptotic Bcl-2. In histiocytic lymphoma
U-937 cells, Park et al. revealed that resveratrol treatment
caused apoptosis and DNA fragmentation, which are
associated with caspase-3 activation and phospholipase C-Á1
degradation. Bcl-2 was found to inhibit resveratrol-induced
apoptosis by a mechanism that interfered with cytochrome c
release and caspase-3 activity (121).
We examined the effect of resveratrol on fresh acute
myeloid leukemia (AML) cells (122). Interleukin (IL)-1‚
plays a key role in proliferation of AML cells, and we found
that resveratrol inhibited proliferation of AML by arresting
the cells at S-phase. Resveratrol significantly reduced
production of IL-1‚, suppressed IL-1‚-induced activation of
NF-Î B, and suppressed colony-forming cell proliferation of
fresh AML marrow cells.
Breast cancer: Several groups have investigated the effects of
resveratrol on breast cancer cells (128-138). Mgbonyebi et al.
showed that resveratrol had antiproliferative effects against
the breast cancer cell lines MCF-7, MCF-10F and MDA-
MB-231, and these effects were independent of the estrogen
receptor (ER) status of the cells (128). Serrero et al. found
that, in ER-positive MCF-7 breast cancer cells, resveratrol
inhibited estradiol-induced cell proliferation by antagonizing
the stimulation by estradiol of an ER element reporter gene
construct and of progesterone receptor (PR) gene expression
(129). Resveratrol also inhibited proliferation of the ER-
negative human breast carcinoma cell line MDA-MB-468 by
a mechanism other than ER antagonism, involving alteration
in autocrine growth modulators such as transforming growth
factor (TGF)-·, TGF-‚, PC cell-derived growth factor and
insulin-like growth factor I receptor mRNA. Nakagawa et al.
found that resveratrol at low concentrations caused cell
proliferation in ER-positive human breast cancer cell lines
(KPL-1, ≤ 22 ÌM; MCF-7, ≤ 4 ÌM), whereas it suppressed
cell growth at high concentrations (≥ 44 ÌM). Growth
suppression was due to apoptosis, as indicated by the
appearance of a sub-G1-phase fraction, up-regulation of Bax
and Bak proteins, down-regulation of Bcl-x
L
protein and
activation of caspase-3. Pozo-Guisado et al. examined the
effects of resveratrol in human breast cancer cell lines MCF-7
and MDA-MB-231 (131). They showed that, although
resveratrol inhibited cell proliferation and viability in both
cell lines, apoptosis was induced in a concentration- and cell-
specific manner. In MDA-MB-231, resveratrol (at
concentrations up to 200 ÌM) lowered the expression and
kinase activities of positive G1/S and G2/M cell-cycle
regulators and inhibited ribonucleotide reductase activity in
a concentration-dependent manner, without a significant
effect on the low expression of tumor suppressors
p21
Cip1/WAF1
, p27
Kip1
and p53. These cells died by a
nonapoptotic process in the absence of a significant change
in cell-cycle distribution. In MCF-7, resveratrol produced a
significant (< 50 ÌM) and transient increase in the
expression and kinase activities of positive G1/S and G2/M
regulators. Simultaneously, p21
Cip1/WAF1
expression was
markedly induced in the presence of high levels of p27
Kip1
and p53. These opposing effects resulted in cell-cycle
blockade at the S phase and induction of apoptosis in MCF-7
cells. Thus, the antiproliferative activity of resveratrol could
take place through the differential regulation of the cell-
cycle, leading to apoptosis or necrosis.
Colon cancer: Several reports suggest that resveratrol
suppresses proliferation of colon cancer cells (143-151). In
the human wild-type p53-expressing HCT116 colon
carcinoma cell line and HCT116 cells with both p53 alleles
inactivated by homologous recombination, Mahyar-Roemer
et al. showed that resveratrol induced apoptosis
independently of p53 and that the apoptosis was mediated
primarily by mitochondria and not by a receptor pathway
(143). Wolter and Stein determined that, in the colon
adenocarcinoma cell line Caco-2, resveratrol enhanced the
differentiation-inducing effect of butyrate, inhibited
butyrate-induced TGF-‚1 secretion, and did not elevate
alkaline phosphatase (ALP) activity or E-cadherin protein
expression (markers of epithelial differentiation) when
applied alone (144). Wolter et al. reported that resveratrol
inhibited growth and proliferation of Caco-2 cells through
apoptosis, which was accompanied by an increase in caspase-
3 activity and in the expression of cyclin E and cyclin A,
decrease in levels of cyclin D1 and cyclin-dependent kinase
(Cdk) 4, cell-cycle arrest in S- to G2-phases at lower
concentrations, and reversal of S-phase arrest at higher
concentrations (145). They observed similar results for the
colon carcinoma cell line HCT116 and found that cell-cycle
inhibition by resveratrol was independent of COX inhibition.
Delmas et al. analyzed the molecular mechanisms of
resveratrol-induced apoptosis in colon cancer cells, with
special attention to the role of the death receptor Fas in this
Aggarwal et al: Resveratrol Inhibits Tumorigenesis
2793
ANTICANCER RESEARCH 24: 2783-2840 (2004)
2794
Table II. Antiproliferative and pro-apoptotic effects of resveratrol against tumor cells and their mechanism.
Cell type Mechanism References
Leukemia
ñ
Inhibits proliferation of chronic B lymphocytic ñ caspase 3, Bcl-2; iNOS (112)
leukemia
ñ
Induces apoptosis in chronic B-cell leukemia ñ iNOS; Bcl-2 (112)
ñ
Inhibits growth and induces apoptosis in many ñ caspases; G2/M-phase (113)
lymphoid and myeloid leukemic cells
ñ
Induces apoptosis in promyelocytic leukemia ñ caspase-9 (114)
(HL-60) cells
ñ
Induce apoptosis in BJAB Burkitt-like lymphoma cells ñ caspases (115)
ñ
Induces apoptosis in adult T-cell leukemia ñ survivin (116)
ñ
Induces apoptosis in T-lymphoblastic leukemia ñ ROS; caspases (117)
CEM-C7H2 cells
ñ
Induces apoptosis in HL-60 cells ñ Fas signaling-dependent apoptosis (118)
ñ
Induces apoptosis in monocytic leukemia ñ caspases; PARP cleavage (119)
(THP-1) cells
ñ
Induces apoptosis in HL-60 cells ñ Bcl-2 (120)
ñ
Induces apoptosis in U-937 cells ñ cytochrome c; caspases (121)
ñ
Inhibits growth of acute myeloid leukemia (AML) ñ S phase; PARP cleavage; caspases (122)
OCIM2 and OCI/AML3
ñ
Induces apoptosis in HL-60 cells ñ Bax; cytochorome c; caspases (123)
ñ
Inhibits growth of HL-60 cells ñ CYP1B1; DNA damage (124)
ñ
Inhibits growth of THP-1 cells ñ tissue factor; NF-kB/Rel-dependent transcription (125)
ñ
Induces apoptosis in BJAB Burkitt-like lymphoma ñ Mitochondrial permeability transition; caspase-3 (125)
ñ
Inhibits cell adhesion U-937 cells ñ E-Selectin (125)
to endotherial cells
ñ
Inhibits proliferation of mitogen-, IL-2, or ñ NF-Î B, IFN-Á, IL-2, TNF and IL-12 (126)
alloantigen-induced splenic lymphocytes
Breast
ñ
Inhibits proliferation of breast epithelial ñ Mechanism is independent of ER status (128)
(MCF-7, MCF-10F and MDA-MB-231) cells
ñ
Inhibits growth of breast cancer (MCF-7, ñ Estradiol stimulation; TGF-·; TGF-‚2 (129)
MDA-MB-468) cells
ñ
Inhibits growth of KPL-1 and MCF-7 cells ñBax, Bak; Bcl-x
L
; caspase-3 (130)
ñ
Induces apoptosis in MCF-7 cells ñ G1/S, G2/M-phase; p21
Cip1/WAF1
; S-phase (131)
ñ
Inhibits growth of MCF-7 cells ñTGF-·; TGF-‚; IGF-1R (132)
ñ
Inhibits growth of 4T1 cells ñ Tumor take; Tumor growth; Metastasis (133)
ñ
Inhibits growth of MCF-7, T47D ñ ROS (134)
and MDA-MB-231 cells
ñ
Inhibits growth of MDA-MB-435 and MCF-7 cells ñ sub G1 phase; G2-phase; p53; cathepsin D (135)
ñ
Induces apoptosis in MCF-7 cells ñ cyclin D; Cdk4; p53, p21
Cip1/WAF1
; Bcl-2, Bax; caspase (136)
ñ
Induces apoptosis of MDA-MB-231 ñ nSMase; ceramide; serine palmitoyltransferase (137)
ñ
Inhibits growth of MCF-7 cells ñ Adenylyl-cyclase activity (138)
ñ
Inhibits growth of MCF-7 cells ñ TGF-·, IGR-R1 mRNA; TGF-‚2 mRNA (139)
ñ
Inhibits growth of MCF-7 and T47D cells ñ CYP1A1 (140)
Colon
ñ
Induces apoptosis of HCT116 cells ñ p53-independent apoptosis (143)
ñ
Enhances the differentiation of Caco-2 with butylate ñ TGF-‚; p27
Kip1
; p21
Cip1/WAF1
(144)
ñ
Induces apoptosis of Caco-2 and HCT116 cells ñ cyclin D1/Cdk4 complex; cyclin E and A (145)
ñ
Induce apoptosis SW480 ñ Redistribution of Fas receptor in membrane rafts (146)
ñ
Induces cell-cycle arrest ñ G2-phase; Cdk 7; Cdc2 (147)
ñ
Induces apoptosis in (col-2) cancer cells ñ sub G0-phase (148)
ñ
Inhibits colon carcinogenesis in F344 rats ñ p21
Cip1/WAF1
(149)
ñ
Induces apoptosis in colon cancer cells ñ DNA fragmentation (150)
ñ
Induces apoptosis of HCT116 cells ñ p53-independent apoptosis (151)
Pancreas
ñ
Induces apoptosis of PANC-1 and AsPC-1 cells ñ sub G0/G1-phase cells (152)
Gastric
ñ
Inhibits growth of KATO-III and RF-1 cells ñ G0/G1-phase (153)
ñ
Inhibits proliferation of human gastric ñ DNA synthesis, iNOS (154)
adenocarcinoma (SNU-1) cells
continued
Aggarwal et al: Resveratrol Inhibits Tumorigenesis
2795
Table II. continued.
Cell type Mechanism References
ñ
Induces apoptosis in esophageal carcinoma ñ Bcl-2; Bax (155)
(EC-9706) cells
Prostate
ñ
Inhibits growth of LnCaP ñ PSA (156)
ñ
Inhibits growth of LnCaP, DU145 and PC-3 cells ñ G1/S-phase; apoptosis; PSA (157)
ñ
Induces apoptosis in prostate cancer (DU145) cells ñ MAPK; cellular p53; p53 binding to DNA (158)
ñ
Inhibits androgen stimulated growth of LNCaP cells ñ PSA; kallikarin-2; ARA70 (159)
ñ
Inhibits growth of LnCaP, DU145 and PC-3 cells ñ NO secretion (160)
ñ
Inhibits growth of LnCaP ñ DNA synthesis; S-phase (161)
ñ
Inhibits growth of LnCaP ñ PSA; ARA; NF-kB (162)
ñ
Inhibits growth of PC-3 ñ PKCa; ERK1/2 (163)
Melanoma
ñ
Induces apoptosis in melanoma (A375 and ñPhosphorylates ERK1/2 (164)
SK-mel28) cells
ñ
Induces apoptosis in epidermoid carcinoma (A431) ñ p21
Cip1/WAF1
; G1-phase (165)
cells
ñ
Inhibits proliferation of epidermoid carcinoma ñ Hyperphosphorylated Rb; G0/G1-phase (166)
(A431) cells
ñ
Induces apoptosis in JB6 P+ mouse epidermal ñ p53-dependent apoptosis pathway (166)
cell line C1 41
ñ
Induces apoptosis of SK-Mel-28 ñ S-phase cyclins A, E, and B1 (167)
Lung
ñ
Induce apoptosis of A549 ñ p53; p21
Cip1/WAF1
; Bax/Bcl-2; caspase; NF-kB (168)
ñ
Induces apoptosis in Chinese hamster lung cell line ñ S-phase (169)
ñ
Inhibits growth of lung cancer (BEP2D) cells ñ CYP1A1 and CYP1B1 (170)
Liver
ñ
Inhibits proliferation in rat hepatoma Fao cells ñ S- and G2/M-phase (171)
ñ
Suppresses hepatoma cell invasion ñ ROS (172)
ñ
Decreases hepatocyte growth factor-induced HepG2 ñUses an unidentified post-receptor mechanism (173)
cell invasion
ñ
Inhibits hepatoma cell, AH 109A proliferation and ñAntioxidant involved in anti-invasive action (174)
invasion
Thyroid and Head &Neck
ñ
Induces apoptosis in thyroid cancer cell lines ñ p53 and MAPK (175)
ñ
Inhibits growth and proliferation of oral squamous ñ DNA synthesis (176)
carcinoma (SCC-25) cells (177)
ñ
Inhibits proliferation in human gingival epithelial ñ DNA synthesis (178)
S-G cells
ñ
Induces apoptosis in the neuroblastoma ñ ERK1/2 (179)
(SH-SY5Y) cell line ñ caspase-7, PARP cleavage (180)
ñ
Induces apoptosis in rat pheochromocytoma ñ DNA fragmentation; NF-kB; ROS (181)
(PC12) cells
Ovarian and Endometria
ñ
Inhibits proliferation of endometrial ñ cyclin A; cyclin E; Cdk2 (174)
adenocarcinoma cells
ñ
Inhibits cell growth and induces apoptosis in ñ NQO-1 (182)
ñ
varian cancer (PA-1) cells
ñ
Inhibits proliferation of endometrial ñ VEGF; EGF; p21
Cip1/WAF1
; Bax (183)
adenocarcinoma cells
ñ
Inhibited growth and induced death ñ cytochorome c; caspases; autophagocytosis (184)
of five human ovarian carcinoma cell
ñ
Inhibits proliferation of endometrial ñExerts estrogen -dependent and -independent effects, (185)
adenocarcinoma cells ñ S-phase, cyclins A and E
ñ
Inhibits proliferation in cervical tumor (HeLa and ñ prostaglandin biosynthesis; S-phase (186)
SiHa) cells
Muscle
ñ
Induces growth inhibition, apoptosis in various cell ñ S-phase; cyclin A1, B1, and D1; ‚-catenin (187)
lines (MCF-7, SW480, HCE7, Seg-1, Bic-1, and HL-60)
oSuppresses mitogenesis in smooth muscle cells ñ G1/S-phases (188)
pathway (146). They showed that, at concentrations of 10-
100 ÌM, resveratrol activated various caspases and triggered
apoptosis in SW480 human colon cancer cells. Caspase
activation was associated with accumulation of the pro-
apoptotic proteins Bax and Bak, which underwent
conformational changes and relocalization to the
mitochondria. Resveratrol did not modulate the expression
of Fas and Fas-ligand (FasL) at the surface of cancer cells,
and inhibition of the Fas/FasL interaction did not influence
the apoptotic response to the molecule. Resveratrol induced
the clustering of Fas and its redistribution in cholesterol and
sphingolipid-rich fractions of SW480 cells, together with
Fas-associated death domain protein (FADD) and
procaspase-8. This redistribution was associated with the
formation of a death-inducing signaling complex (DISC).
Transient transfection of a dominant-negative mutant of
FADD, E8, or viral protein MC159, that interfered with
DISC function, decreased the apoptotic response of SW480
cells to resveratrol and partially prevented resveratrol-
induced Bax and Bak conformational changes. Altogether,
these results indicated that the ability of resveratrol to
induce the redistribution of Fas in membrane rafts may
contribute to the molecule's ability to trigger apoptosis in
colon cancer cells.
Liang et al. found that resveratrol inhibited proliferation
of HT-29 colon cancer cells and resulted in their
accumulation in the G2-phase of the cell-cycle, and that this
was accompanied by inactivation of Cdc2/p34 protein kinase
and an increase in the tyrosine phosphorylated (inactive)
form of Cdc2 (147). Kinase assays revealed that the
reduction of Cdc2 activity by resveratrol was mediated
through inhibition of Cdk7 kinase activity, while Cdc25A
phosphatase activity was not affected. In addition,
resveratrol-treated cells were shown to have a low level of
ANTICANCER RESEARCH 24: 2783-2840 (2004)
2796
Table III. Effects of resveratrol on different cell signaling pathways.
Signaling pathway References
Up-regulate Fas pathway (118, 146, 191)
Inhibit mitochondrial pathway (114, 117, 192)
Inhibit Rb/E2FDP pathway (166, 168)
Activate p53 pathway (51, 162, 175, 193-198)
Activate ceramide pathway (137)
Inhibit tubulin polymerazation pathway (199)
Activate adenyl-cyclase pathway (138)
Inhibit NF-kB signaling pathway (120, 122, 125, 126, 168, 202-208)
Inhibit AP-1 signaling pathway (22, 120, 201, 209-214)
Regulate Egr-1 pathway (215, 216)
Inhibit MAPK pathway (163, 175, 179, 195, 196, 217, 218)
Suppression of protein kinases by resveratrol (127, 139, 153, 218-221)
Modulation of NO/NOS pathway (92, 154, 194, 222)
Suppression of growth factor (129, 173, 183, 223-226)
and associated protein tyrosine kinases
Suppression of COX-2 and lipooxygenase (141, 142, 212, 222, 227, 228)
Suppression of cell-cycle proteins (122, 135, 145, 147, 151, 161, 165, 167, 187, 191, 194, 229)
Suppression of adhesion molecules (230, 231)
Suppression of androgen receptors (159, 285)
Suppression of PSA (156)
Suppresion of inflammatory cytokine (211, 232-235)
Suppression of angiogenesis, invasion and metastasis (194, 218, 237-241, 243-246, 286)
Effect on bone cells (247, 278)
Inhibit the expression of cytochrome p450 (73, 140, 229, 248-258, 287)
and modulate metabolism of carcinogens:
Suppression of inflammation (198, 222, 259-261)
Antioxidant effects (71, 262-276)
Suppression of transformation (193, 226)
Induction of cellular differentiation (277-279)
Estrogenic/antiestrogenic effects (132, 174, 185, 280-284, 289)
Effect on normal cells (188, 194, 197, 237, 238, 245, 290-292)
Suppression of mutagenesis (169, 294-298)
Radioprotective and radiosensitive (186)
Chemosensitization (180, 181, 304-307)
Immunomodulatory effects (126, 236, 259, 314-316)
Cdk7 kinase-Thr(161)-phosphorylated Cdc2. These results
demonstrated that resveratrol induced cell-cycle arrest at
the G2 phase through inhibition of Cdk7 kinase activity,
suggesting that its antitumor activity might occur through
disruption of cell division at the G2/M-phase.
Pancreatic cancer: Ding and Adrian demonstrated that, in
human pancreatic cancer cell lines PANC-1 and AsPC-1,
resveratrol inhibited proliferation through apoptosis and
dramatically increased the fraction of sub-G0/G1-phase
cells (152).
Gastric cancer: Resveratrol has been shown to suppress
proliferation of gastric cancer cells (153-155). Atten et al.
reported that resveratrol inhibited proliferation of
nitrosamine-stimulated human gastric adenocarcinoma
KATO-III and RF-1 cells (153). It arrested KATO-III cells in
the G0/G1-phase of the cell-cycle and eventually induced
apoptotic cell death by utilizing a proteinase kinase C (PKC)-
mediated mechanism to deactivate these gastric
adenocarcinoma cells. Holian et al. demonstrated that, in
gastric adenocarcinoma cell line SNU-1, which was stimulated
by hydrogen peroxide (H
2
O
2
), resveratrol suppressed
synthesis of DNA and generation of endogenous O
2
-
but
stimulated NOS activity, which may have been responsible for
inhibition of SNU-1 proliferation (154). Resveratrol also
inhibited the growth of esophageal cancer cell line EC-9706
(155). Resveratrol-induced apoptosis of EC-9706 was
mediated by down-regulation of Bcl-2 and up-regulation of
the expression of the apoptosis-regulated gene Bax.
Prostate cancer: Proliferation of both androgen-dependent
and -independent prostate cancer cells is suppressed by
resveratrol (156-163). Using cultured prostate cancer cells
that mimic the initial (hormone-sensitive; LNCaP) and
advanced (hormone-refractory; DU-145, PC-3, and JCA-1)
stages of prostate carcinoma, Hsieh and Wu showed that
resveratrol caused substantial decreases in growth of LNCaP,
PC-3 and DU145 cells, but had only a modest inhibitory
effect on proliferation of JCA-1 cells, and that it partially
disrupted the G1/S transition in all three androgen-non-
responsive cell lines (157). It caused a significant percentage
of LNCaP cells to undergo apoptosis and significantly
lowered both intracellular and secreted prostate-specific
antigen (PSA) levels without affecting expression of the
androgen receptor (AR). Lin et al. also showed, in DU145
cells, that resveratrol induced apoptosis through activation
of mitogen-activated protein kinase (MAPK,) increases in
cellular p53 content, serine-15 phosphorylation of p53, p53
binding to DNA and p53-stimulated increase in p21
Cip1/WAF1
mRNA (158). Mitchell et al. found that, in a hormone-
sensitive prostate cancer cell line, resveratrol repressesed
different classes of androgen up-regulated genes at the
protein or mRNA level, including PSA, human glandular
kallikrein-2, AR-specific coactivator ARA70, and the Cdk
inhibitor p21
Cip1/WAF1
(159). This inhibition is probably
attributable to a reduction in AR at the transcription level,
inhibiting androgen-stimulated cell growth and gene
expression. Kampa et al. reported that the antiproliferative
effects of resveratrol on DU145 cells could have been
mediated through a decrease in NO, although resveratrol did
not affect growth of PC3 and LNCaP cells (160).
Kuwajerwala et al. showed that, in androgen-sensitive
LNCaP cells, the effect of resveratrol on DNA synthesis
varied dramatically depending on the concentration and the
duration of treatment (161). In cells treated for 1 h,
resveratrol had only an inhibitory effect on DNA synthesis,
which increased with increasing concentration (IC
50
, 20 ÌM).
However, when treatment duration was extended to 24 h,
resveratrol had a dual effect on DNA synthesis. At 5-10 ÌM
it caused a two- to three-fold increase in DNA synthesis,
while at ≥15 ÌM it inhibited DNA synthesis. The increase in
DNA synthesis was seen only in LNCaP cells, not in
androgen-independent DU145 prostate cancer cells or in
NIH/3T3 fibroblast cells. The resveratrol-induced increase in
DNA synthesis was associated with enrichment of LNCaP
cells in S-phase and concurrent decreases in nuclear
p21
Cip1/WAF1
and p27
Kip1
levels. Furthermore, consistent
with the entry of LNCaP cells into S-phase, there was a
dramatic increase in nuclear Cdk2 activity associated with
both cyclin A and cyclin E. Taken together, their
observations indicate that LNCaP cells treated with
resveratrol are induced to enter into S-phase, but subsequent
progression through S-phase is limited by the inhibitory
effect of resveratrol on DNA synthesis, particularly at
concentrations greater than 15 ÌM. This unique ability of
resveratrol to exert opposing effects on two important
processes in cell-cycle progression, induction of S-phase and
inhibition of DNA synthesis, may be responsible for its dual
apoptotic and antiproliferative effects.
Prostate cancer prevention by key elements present in
human nutrients derived from plants and fruits has been
confirmed in various cell cultures and tumor models.
Resveratrol has been shown to induce remarkable inhibitory
effects in prostate carcinogenesis via diverse cellular
mechanisms associated with tumor initiation, promotion and
progression. Narayanan et al. examined whether resveratrol
activates a cascade of p53-directed genes that are involved
in apoptosis mechanism(s) or modifies cell growth by
modifying AR and its co-activators directly or indirectly
(162). They demonstrated by DNA microarray, reverse
tanscriptase-polymerase chain reaction (RT-PCR), Western
blot and immunofluorescence analyses that treatment of
androgen-sensitive prostate cancer cells (LNCaP) with 10
ÌM resveratrol for 48 h down-regulated PSA, AR co-
activator ARA 24, and NF-Î B p65. Altered expression of
Aggarwal et al: Resveratrol Inhibits Tumorigenesis
2797
these genes is associated with activation of p53-responsive
genes such as p53, PIG 7, p21
Cip1/WAF1
, p300/CBP and
apoptosis protease activating factor-1 (Apaf-1). The effect
of resveratrol on p300/CBP plays a central role in its cancer-
preventive mechanisms in LNCaP cells. These results
implicate activation of more than one set of functionally
related molecular targets. At this point we have identified
some of the key molecular targets associated with the AR
and p53 target genes.
Melanoma: Several studies suggest that resveratrol is
effective against melanoma (164-167). Resveratrol inhibited
growth and induced apoptosis in human melanoma cell lines
A375 and SK-mel28 (164). It did not alter the
phosphorylation of p38 MAPK or c-Jun N-terminal kinase
(JNK) in either cell line. Resveratrol induced
phosphorylation of extracellular receptor kinase (ERK)1/2
in A375 but not in SK-mel28 cells. Ahmad et al.
demonstrated that resveratrol, via modulations in Cdk
inhibitor-cyclin-Cdk machinery, resulted in a G1-phase
arrest followed by apoptosis of human epidermoid
carcinoma (A431) cells (165). It caused an induction of
p21
Cip1/WAF1
that inhibited cyclin D1/D2-Cdk6, cyclin
D1/D2-Cdk4, and cyclin E-Cdk2 complexes, thereby
imposing an artificial checkpoint at the G1/S-phase
transition of the cell-cycle. These authors also showed, in
the same cell line, the involvement of the retinoblastoma
(Rb)-E2F/DP pathway in resveratrol-mediated cell-cycle
arrest and apoptosis (166). They suggested that resveratrol
caused a down-regulation of hyperphosphorylated Rb
protein with a relative increase in hypophosphorylated Rb
that, in turn, compromised the availability of free E2F,
which may have resulted in stoppage of cell-cycle
progression at the G1/S-phase transition, thereby leading to
a G0/G1 phase arrest and subsequent apoptotic cell death.
Larrosa et al. showed that resveratrol and the related
molecule 4-hydroxystilbene induced growth inhibition,
apoptosis, S-phase arrest and up-regulation of cyclins A, E
and B1 in human SK-Mel-28 melanoma cells (167).
Lung cancer: Several studies suggest that resveratrol is
effective against lung carcinoma (168-170). Kim et al. showed
that resveratrol inhibited the growth of human lung
carcinoma A549 cells and resulted in a concentration-
dependent induction of S-phase arrest in cell-cycle
progression, marked inhibition of phosphorylation of Rb and
concomitant induction of Cdk inhibitor p21
Cip1/WAF1
, which
is transcriptionally up-regulated and is p53-dependent (168).
In addition, fluorescence microscopy and flow cytometric
analysis showed that treatment with resveratrol resulted in
induction of apoptosis. These effects were found to correlate
with activation of caspase-3 and a shift in the Bax/Bcl-x
L
ratio toward apoptosis. Resveratrol treatment also inhibited
the transcriptional activity of NF-Î B. These findings suggest
that resveratrol has firm potential for development as an
agent for prevention of human lung cancer.
Liver cancer: Several studies suggest that resveratrol is
effective against liver cancer (171-174). Delmas et al.
examined the ability of resveratrol to inhibit cell proliferation
in the rat hepatoma Fao cell line and the human
hepatoblastoma HepG2 cell line (171). The results showed
that resveratrol strongly inhibited cell proliferation and that
Fao cells were more sensitive than HepG2 cells. Interestingly,
the presence of ethanol lowered the threshold of the
resveratrol effect. Resveratrol appeared to prevent or delay
the entry to mitosis, since no inhibition of
3
H-thymidine
incorporation was observed, while the number of the cells in
S- and G2/M-phases increased. Kozuki et al. revealed that 100
or 200 ÌM of resveratrol inhibited proliferation of AH109A
hepatoma cells and suppressed invasion of the hepatoma cells
even at a concentration of 25 ÌM (172). This anti-invasive
activity of resveratrol is independent of its antiproliferative
activity and may be related to its anti-oxidative action. De
Ledinghen et al. found that resveratrol decreased hepatocyte
growth factor-induced scattering of HepG2 hepatoma cells
and invasion by an unidentified postreceptor mechanism
(173). It decreased cell proliferation without evidence of
cytotoxicity or apoptosis, with no decrease in the level of the
hepatocyte growth factor receptor c-met, c-met precursor
synthesis, c-met autophosphorylation, or activation of Akt-1
or ERK1/2. Moreover, resveratrol did not decrease urokinase
expression and did not block the catalytic activity of
urokinase.
Thyroid and head and neck cancers: Several reports suggest
that resveratrol may suppress proliferation of thyroid and
other head and neck cancers (174-181). Shih et al. showed
that treatment of papillary thyroid carcinoma and follicular
thyroid carcinoma cell lines with resveratrol led to
apoptosis, which accompanied activation and nuclear
translocation of ERK1/2 (175). Resveratrol increased the
cellular abundance of p53, serine phosphorylation of p53,
and abundance of c-fos, c-Jun, and p21
Cip1/WAF1
mRNAs.
Elattar et al. reported that resveratrol led to inhibition of
human oral squamous carcinoma SCC-25 cell growth and
DNA synthesis (176, 177). Moreover, combining 50 ÌM
resveratrol with 10, 25, or 50 ÌM quercetin resulted in
gradual and significant increases in the inhibitory effects of
the two compounds. Babich et al. demonstrated that
resveratrol irreversibly caused arrest of human gingival
epithelial cell growth by inhibition of DNA synthesis (178).
Ovarian and endometrial tumors: Several studies suggest that
resveratrol is effective against ovarian and endometrial
tumors (174, 182-186). Yang et al. showed that resveratrol
ANTICANCER RESEARCH 24: 2783-2840 (2004)
2798
inhibited cell growth and induced apoptosis in PA-1 human
ovarian cancer cells and up-regulated the NAD(P)H
quinone oxidoreductase 1 (NQO-1) gene, which is involved
in p53 regulation (182). Bhat and Pezzuto reported that
treatment of human endometrial adenocarcinoma
(Ishikawa) cells with resveratrol did not significantly
increase the levels of the estrogen-inducible marker enzyme
ALP (174). On the contrary, it decreased 17‚-estradiol-
induced ALP and PR expression and thus its effects may be
mediated by both estrogen-dependent and -independent
mechanisms. It inhibited Ishikawa cell proliferation by
arresting cells at S-phase and increased expression of cyclins
A and E but decreased Cdk2. Kaneuchi et al. showed that
resveratrol suppressed the growth of Ishikawa cells through
down-regulation of epidermal growth factor (EGF) (183).
Opipari et al. showed that resveratrol inhibited growth and
induced death in a panel of five human ovarian carcinoma cell
lines and that this response was associated with mitochondrial
release of cytochrome c, formation of the apoptosome
complex, and caspase activation (184). Surprisingly, even with
these molecular features of apoptosis, analysis of the
resveratrol-treated cells by light and electron microscopy
revealed morphological and ultrastructural changes indicative
of autophagocytic, rather than apoptotic, death. This
suggested that resveratrol can induce cell death through two
distinct pathways. Consistent with resveratrol's ability to kill
cells via nonapoptotic processes, cells transfected to express
high levels of the antiapoptotic proteins Bcl-x
L
and Bcl-2 were
equally as sensitive as control cells to resveratrol. Together,
these findings show that resveratrol induces death in ovarian
cancer cells through a mechanism distinct from apoptosis,
suggesting that it may provide leverage to treat ovarian cancer
that is chemoresistant on the basis of ineffective apoptosis.
C1b. Resveratrol induces apoptosis
Apoptosis is a mode of cell death that differs from necrosis.
While the former is characterized by initiation of cell death
from the outside of the cell, the latter is a death mechanism
initiated from inside the cell, primarily from the
mitochondria (189). Apoptosis is usually mediated through
the activation of caspases. Mechanistically, two different
type of apoptosis have been described; one that is caspase-8-
dependent and receptor-mediated (type I), and the other
that is caspase-9-dependent and usually mediated through
the mitochondria (type II). Resveratrol has been shown to
mediate apoptosis through a variety of different pathways
(Figure 3) (51, 114, 117, 118, 131, 137, 138, 146, 148, 162,
166, 168, 175, 187, 190-199), as described below.
Fas pathway: Resveratrol has been shown to induce death
receptors, that in turn activate apoptosis, through the type I
pathway. Fas is one of the death receptors of the tumor
necrosis factor (TNF) superfamily (200). Clement et al.
showed that resveratrol triggered FasL signaling-dependent
apoptosis in human tumor cells (118). They showed that
resveratrol treatment enhanced FasL expression on HL-60
cells and T47D breast carcinoma cells, and that resveratrol-
mediated cell death was specifically dependent on Fas
signaling. Resveratrol treatment had no effect on normal
PBMC, which correlated with the absence of a significant
change in either Fas or FasL expression on treated PBMC.
These data showed specific involvement of the Fas-FasL
system in the anticancer activity of resveratrol. In contrast
to these results, those of Bernhard et al. found that
resveratrol caused arrest in the S-phase prior to Fas-
independent apoptosis in CEM-C7H2 ALL cells (191).
These findings indicate that the effect of resveratrol on Fas
signaling may depend on cell type. Delmas et al. showed
that resveratrol-induced apoptosis was associated with Fas
redistribution in the rafts and the formation of a DISC in
colon cancer cells (146). Resveratrol did not modulate the
expression of Fas and FasL at the surface of cancer cells,
and inhibition of the Fas-FasL interaction did not influence
the apoptotic response to the molecule. Resveratrol,
however, induced the clustering of Fas and its redistribution
in cholesterol- and sphingolipid-rich fractions of SW480
cells, together with FADD and procaspase-8. This
redistribution was associated with formation of a DISC.
Transient transfection of a dominant-negative mutant of
FADD, E8, or viral protein MC159 that interferes with
DISC function decreased the apoptotic response of SW480
cells to resveratrol and partially prevented resveratrol-
induced Bax and Bak conformational changes. Altogether,
these results indicate that the ability of resveratrol to induce
redistribution of the Fas receptor in membrane rafts may
contribute to the molecule's ability to trigger apoptosis in
colon cancer cells.
Mitochondrial pathway: Resveratrol has also been shown to
activate the type II pathway. This pathway for apoptosis is
mediated through the activation of the mitochondrial
pathway. Dorrie et al. showed that resveratrol induced
extensive apoptosis by depolarizing mitochondrial
membranes and activating caspase-9 in ALL cells and that
these effects were independent of Fas signaling (114).
Tinhofer et al. showed that resveratrol induced apoptosis via
a novel mitochondrial pathway controlled by Bcl-2 (117).
Mitochondrial proton F0F1-ATPase/ATP synthase
synthesizes ATP during oxidative phosphorylation. Zheng et
al. found that resveratrol inhibited the enzymatic activity of
both rat brain and liver F0F1-ATPase/ATP synthase (IC
50
,
12–28 ÌM) (192). The inhibition of F0F1-ATPase by
resveratrol was non-competitive in nature. Thus the
mitochondrial ATP synthase is a target for this dietary
phytochemical and may contribute to its potential
Aggarwal et al: Resveratrol Inhibits Tumorigenesis
2799
cytotoxicity. Zheng et al. also found that piceatannol, an
analogue of resveratrol, inhibited mitochondrial F0F1-
ATPase activity by targeting the F1 complex (192).
Piceatannol potently inhibited rat brain mitochondrial
F0F1-ATPase activity in both solubilized and
submitochondrial preparations (IC
50
, 8-9 ÌM) while having
a relatively small effect on Na
+
, K
+
-ATPase activity.
Piceatannol inhibited the ATPase activity of purified rat
liver F1 (IC
50
, 4 ÌM), while resveratrol was slightly less
active (IC
50
, 14 ÌM). These results indicated that
piceatannol and resveratrol inhibit the F-type ATPase by
targeting the F1 sector, which is located in the inner
membrane of mitochondria and the plasma membrane of
normal endothelial cells and several cancer cell lines.
Rb-E2F/DP pathway: Rb and the E2F family of transcription
factors are important proteins that regulate the progression
of the cell-cycle at and near the G1/S-phase transition
(Figure 4). Adhami et al. provided evidence for the
involvement of the Rb-E2F/DP pathway as an important
contributor to resveratrol-mediated cell-cycle arrest and
apoptosis (166). Immunoblot analysis demonstrated that
resveratrol treatment of A431 melanoma cells resulted in a
decrease in the hyperphosphorylated form of Rb and a
relative increase in hypophosphorylated Rb. This response
was accompanied by down-regulation of expression of all
five E2F family transcription factors studied and their
heterodimeric partners DP1 and DP2. This suggested that
resveratrol causes down-regulation of hyperphosphorylated
Rb protein with a relative increase in hypophosphorylated
Rb that, in turn, compromises the availability of free E2F.
These events may result in a stoppage of cell-cycle
progression at the G1/S-phase transition, thereby leading to
a G0/G1-phase arrest and subsequent apoptotic cell death.
Kim et al. showed that resveratrol treatment of A549 cells
resulted in a concentration-dependent induction of S-phase
ANTICANCER RESEARCH 24: 2783-2840 (2004)
2800
Figure 3. Various proposed mechanisms of apoptosis of tumor cells by resveratrol.
arrest in cell-cycle progression (168). This antiproliferative
effect of resveratrol was associated with a marked inhibition of
phosphorylation of Rb and concomitant induction of the Cdk
inhibitor p21
Cip1/WAF1
, which appears to be transcriptionally
up-regulated and p53-dependent. Fluorescence microscopy
and flow-cytometric analysis also revealed that treatment with
resveratrol resulted in induction of apoptosis. These effects
were found to correlate with activation of caspase-3 and a shift
in the Bax/Bcl-x
L
ratio toward apoptosis.
p53 activation pathway: p53 is a tumor suppressor gene.
There are numerous reports about the role of p53 in
resveratrol-induced apoptosis (51, 162, 175, 193-198).
Huang et al. found that resveratrol-induced apoptosis
occurred only in cells expressing wild-type p53 (p53
+/+
), but
not in p53-deficient (p53
-/-
) cells, while there was no
difference in apoptosis induction between normal
lymphoblasts and sphingomyelinase-deficient cell lines
(193). These results demonstrated for the first time that
resveratrol induces apoptosis through activation of p53
activity, suggesting that resveratrol’s antitumor activity may
occur through induction of apoptosis. Hsieh et al. showed
that resveratrol inhibited proliferation of pulmonary artery
endothelial cells, which correlated with suppression of cell
progression through the S- and G2-phases of the cell-cycle
and was accompanied by increased expression of p53 and
elevation of the level of Cdk inhibitor p21
Cip1/WAF1
(194).
Lu et al. showed that resveratrol analogues significantly
induced expression of p53, GADD45 and Bax genes and
concomitantly suppressed expression of the Bcl-2 gene in
human fibroblasts transformed with SV40 virus (WI38VA),
but not in nontransfected WI38 cells (51). A large increase
in p53 DNA-binding activity and the presence of p53 in the
Bax promoter binding complex suggested that p53 was
responsible for the Bax gene expression induced by
resveratrol in transformed cells.
She et al. elucidated the potential signaling components
underlying resveratrol-induced p53 activation and induction
of apoptosis (195, 196). They found that, in the JB6 mouse
epidermal cell line, resveratrol activated ERK1/2, JNK, and
p38 MAPK and induced serine-15 phosphorylation of p53.
Stable expression of a dominant-negative mutant of ERK2
or p38 MAPK or their respective inhibitors, PD98059 or
SB202190, repressed phosphorylation of p53 at serine-15. In
Aggarwal et al: Resveratrol Inhibits Tumorigenesis
2801
Figure 4. Effect of resveratrol on signaling proteins involved in apoptosis.
contrast, overexpression of a dominant-negative mutant of
JNK1 had no effect on the phosphorylation. Most
importantly, ERK1/2 and p38 MAPK formed a complex with
p53 after treatment with resveratrol. Strikingly, resveratrol-
activated ERK1/2 and p38 MAPK, but not JNKs,
phosphorylated p53 at serine-15 in vitro. Furthermore,
pretreatment of the cells with PD98059 or SB202190 or
stable expression of a dominant-negative mutant of ERK2
or p38 MAPK impaired resveratrol-induced p53-dependent
transcriptional activity and apoptosis, whereas constitutively
active MEK1 increased the transcriptional activity of p53.
These data strongly suggest that both ERK1/2 and p38
MAPK mediate resveratrol-induced activation of p53 and
apoptosis through phosphorylation of p53 at serine-15. Shih
et al. also showed that resveratrol acted via a Ras-MAPK
kinase-MAPK signal transduction pathway to increase p53
expression, serine phosphorylation of p53, and p53-
dependent apoptosis in thyroid carcinoma cell lines. Haider
et al. showed that resveratrol led to a reversible arrest in
early S phase of the vascular smooth muscle cell (VSMC),
accompanied by accumulation of hyperphosphorylated Rb
(197). Resveratrol decreased cellular levels of the
p21
Cip1/WAF1
and p27
Kip1
and increased the level of
phosphorylated p53 protein (serine-15). The authors found
that resveratrol only slightly inhibited phosphorylation of
ERK1/2, protein kinase B/Akt, and p70(S6) kinase upon
serum stimulation. Thus, inhibition of these kinases is not
likely to contribute to the effects of the polyphenol on the
cell-cycle. Importantly, the observed S-phase arrest was not
linked to an increase in apoptotic cell death: there were no
detectable increases in apoptotic nuclei or in levels of the
proapoptotic protein Bax. This was the first study to
elucidate the molecular pathways mediating the
antiproliferative properties of resveratrol in VSMCs.
The expression of the nonsteroidal anti-inflammatory
drug -activated gene-1 (NAG-1), a member of the TGF-‚
superfamily, has been associated with pro-apoptotic and
antitumorigenic activities. Baek et al. demonstrated that
resveratrol induced NAG-1 expression and apoptosis
through an increase in the expression of p53 (198). They
showed that p53-binding sites within the promoter region of
NAG-1 played a pivotal role in controlling NAG-1
expression by resveratrol. Derivatives of resveratrol were
examined for NAG-1 induction, and the data suggest that
induction of NAG-1 and p53 by resveratrol is not dependent
on its anti-oxidant activity. The data may provide a linkage
between p53, NAG-1 and resveratrol and, in part, a new clue
to the molecular mechanism of the antitumorigenic activity
of natural polyphenolic compounds.
Earlier studies showed that resveratrol alters the expression
of genes involved in cell-cycle regulation and apoptosis,
including cyclins, Cdks, p53, and Cdk inhibitors. However,
most of the p53-controlled effects related to the role of
resveratrol in transcription, either by activation or repression
of a sizable number of primary and secondary target genes,
have not been investigated. Narayanan et al. examined
whether resveratrol activates a cascade of p53-directed genes
that are involved in apoptosis mechanism(s) (162). They
demonstrated by DNA microarray, RT-PCR, Western blot
and immunofluorescence analyses that treatment of androgen-
sensitive prostate cancer cells (LNCaP) with resveratrol down-
regulated PSA, AR co-activator ARA 24, and NF-Î B p65.
Altered expression of these genes is associated with activation
of p53-responsive genes such as p53, PIG 7, p21
Cip1/WAF1
,
p300/CBP and Apaf-1.
Ceramide activation pathway: Apoptosis induction by various
cytokines has been shown to be mediated through
generation of ceramide. Whether resveratrol-induced
apoptosis also involves ceramide production has been
investigated. Scarlatti et al. showed that resveratrol can
inhibit growth and induce apoptosis in MDA-MB-231, a
highly invasive and metastatic breast cancer cell line, in
concomitance with a dramatic endogenous increase of
growth inhibitory/pro-apoptotic ceramide (137). They found
that accumulation of ceramide derives from both de novo
ceramide synthesis and sphingomyelin hydrolysis. More
specifically, they demonstrated that ceramide accumulation
induced by resveratrol can be traced to the activation of
serine palmitoyltransferase (SPT), the key enzyme of a de
novo ceramide biosynthetic pathway, and neutral
sphingomyelinase (nSMase), a main enzyme of the
sphingomyelin/ceramide pathway. By using specific
inhibitors of SPT (myriocin and L-cycloserine) and nSMase
(gluthatione and manumycin), however, they found that
only the SPT inhibitors could counteract the biological
effects induced by resveratrol. Thus, resveratrol seems to
exert its growth-inhibitory/apoptotic effect on the metastatic
breast cancer cell line MDA-MB-231 by activating the de
novo ceramide synthesis pathway.
Tubulin polymerization pathway: Certain chemotherapeutic
agents such as taxol induce apoptosis by interfering with
tubulin polymerization. Whether resveratrol could also
mediate apoptosis through this pathway has been
investigated. Schneider et al. found that a methylated
derivative of resveratrol (Z-3,5,4'- trimethoxystilbene; R3)
at a concentration of 0.3 ÌM, exerted an 80% growth-
inhibitory effect on human colon cancer Caco-2 cells and
arrested growth completely at a concentration of 0.4 ÌM
(R3 was 100-fold more active than resveratrol) (199). The
cis conformation of R3 was also 100-fold more potent than
the trans isomer. R3 (0.3 ÌM) caused cell-cycle arrest at the
G2/M-phase transition. The drug inhibited tubulin
polymerization in a dose-dependent manner (IC
50
, 4 ÌM),
and it reduced by half the activities of ornithine
ANTICANCER RESEARCH 24: 2783-2840 (2004)
2802
decarboxylase and s-adenosylmethionine decarboxylase.
This caused depletion of the polyamines putrescine and
spermidine, which are growth factors for cancer cells. R3
partially inhibited colchicine binding to its binding site on
tubulin, indicating that R3 either partially overlaps with
colchicine binding or binds to a specific site of tubulin that
is not identical with the colchicine binding site, modifying
colchicine binding by allosteric influences. R3 is an
interesting antimitotic drug that exerts cytotoxic effects by
depleting the intracellular pool of polyamines and by
altering microtubule polymerization. Such a drug may be
useful for the treatment of neoplastic diseases.
Adenylyl-cyclase pathway: Both cyclic GMP and cyclic AMP
(cAMP) are known to regulate proliferation of cells.
Whether resveratrol could modulate cell growth by
modulating the levels of these nucleotides has been
investigated (138). El-Mowafy et al. examined the effects
of resveratrol on the activity of the enzymes adenylate
cyclase and guanylate cyclase, two known cytostatic
cascades in MCF-7 breast cancer cells (138). Resveratrol
increased cAMP levels (t
1/2
, 6.2 min; EC50, 0.8 ÌM), but
had no effect on cGMP levels. The stimulatory effects of
resveratrol on adenylate cyclase were not altered either by
the protein synthesis inhibitor actinomycin-D (5 ÌM) or
the ER blockers tamoxifen and ICI182,780 (1 ÌM each).
Likewise, cAMP formation by resveratrol was insensitive
to both the broad-spectrum phosphodiesterase (PDE)
inhibitor IBMX (0.5 ÌM) and the cAMP-specific PDE
inhibitor rolipram (10 ÌM). Instead, these PDE inhibitors
significantly augmented maximal cAMP formation by
resveratrol. Parallel experiments showed that the
antiproliferative effects of resveratrol in these cells were
appreciably reversed by the protein kinase A inhibitors
Rp-cAMPS (100-300 ÌM) and KT-5720 (10 ÌM).
Pretreatment with the cPLA2 inhibitor arachidonyl
trifluoromethyl ketone (10 ÌM) markedly antagonized the
cytotoxic effects of resveratrol. With these findings, we
demonstrated that resveratrol is an agonist for the
cAMP/protein kinase A system.
C1c: Resveratrol suppresses NF-Î B activation
Because resveratrol exhibits anti-inflammatory, cell growth-
modulatory and anticarcinogenic effects, that it mediates
these effects by suppressing NF-Î B, a nuclear transcription
factor that regulates the expression of various genes involved
in inflammation, cytoprotection and carcinogenesis, has been
proposed (200, 201). We investigated the effect of
resveratrol on NF-Î B activation induced by various
inflammatory agents. Resveratrol blocked TNF-induced
activation of NF-Î B and suppressed TNF-induced
phosphorylation and nuclear translocation of the p65 subunit
of NF-Î B and NF-Î B-dependent reporter gene transcription
(22, 71, 73, 92, 120, 122, 125-127, 129, 132, 135, 139-142, 145,
147, 151, 153, 154, 156, 159, 161, 165, 167, 168, 173-175, 179,
182, 183, 185, 187, 191, 193-196, 198, 201-284). Suppression
of TNF-induced NF-Î B activation by resveratrol was not
restricted to myeloid cells (U-937); it was also observed in
lymphoid (Jurkat) and epithelial (HeLa and H4) cells.
Resveratrol also blocked NF-Î B activation induced by
phorbol myristate acetate (PMA), LPS, H
2
O
2
, okadaic acid
and ceramide. Holmes-McNary and Baldwin found
resveratrol to be a potent inhibitor of both NF-Î B activation
and NF-Î B-dependent gene expression through its ability to
inhibit IÎ B kinase activity, the key regulator in NF-Î B
activation, probably by inhibiting an upstream signaling
component (202). In addition, resveratrol blocked the
expression of mRNA-encoding monocyte chemoattractant
protein-1, a NF-Î B-regulated gene. Heredia et al. found that
resveratrol synergistically enhanced the anti-HIV-1 activity
of the nucleoside analogues AZT, ddC, and ddI (14).
Resveratrol at a concentration of 10 ÌM was not toxic to
cells, and by itself reduced viral replication by 20-30%. In
phytohemagglutinin (PHA)-activated PBMCs infected with
HTLV-IIIB, 10 ÌM resveratrol reduced the 90% inhibitory
concentrations (IC
90
) of AZT, ddC and ddI by 3.5-, 5.5- and
17.8-fold, respectively. Similar antiviral activity was
demonstrated when ddI was combined with 5 or 10 ÌM
resveratrol in PBMCs infected with clinical isolates of
HIV-1. The addition of resveratrol resulted in a >10-fold
augmentation of ddI antiviral activity in infected monocyte-
derived macrophages. In a resting cell model of
T lymphocytes infected with HTLV-IIIB, resveratrol plus
ddI in combination, but not individually, suppressed the
establishment of a productive viral infection. In addition,
resveratrol plus ddI markedly inhibited the replication of
four ddI-resistant viral isolates, three of which presented
mutations in the reverse transcriptase gene conferring
reverse transcriptase-multidrug resistance. Finally, 10 ÌM
resveratrol showed enhancement of ddI antiviral suppressive
activity similar to that of 100 ÌM of hydroxyurea. However,
resveratrol had less of a cellular antiproliferative effect than
hydroxyurea.
Pellegatta et al. reported different short- and long-term
effects of resveratrol on NF-Î B phosphorylation and nuclear
appearance in human endothelial cells (203). They found that
the nuclear appearance of p50 and p65 acutely induced by
TNF· was not modified by resveratrol, but was increased after
overnight incubation with resveratrol alone or in combination
with TNF·. Acute treatment with resveratrol did not modify
TNF·-induced cytoplasmic IÎ B· serine phosphorylation but
did increase IÎ B· tyrosine phosphorylation. Resveratrol
increased tyrosine phosphorylation (but not nitrosylation) of
immunoprecipitated NF-Î B, did not decrease cellular
p21
Cip1/WAF1
, and did not increase peroxisome proliferator-
Aggarwal et al: Resveratrol Inhibits Tumorigenesis
2803
activated receptor-· activity. They concluded that acute
resveratrol treatment does not inhibit the nuclear appearance
of NF-Î B in human umbilical vein endothelial cells
(HUVEC), but overnight treatment does.
We showed that resveratrol blocks IL-1‚-induced activation
of NF-Î B that leads to inhibition of proliferation, causes
S-phase arrest, and induces apoptosis of AML cells (122).
Adhami et al. showed the suppression of UV B exposure-
mediated activation of NF-Î B in normal human keratinocytes
by resveratrol (204). Kim et al. showed the involvement of NF-
Î B suppression in induction of growth arrest and apoptosis by
resveratrol in human lung carcinoma A549 cells (168). These
results indicate that NF-Î B suppression by resveratrol may be
essential for its antitumor activities.
C1d. Resveratrol suppresses AP-1 activation
Activator protein-1 (AP-1) is a transcription factor
transactivated by many tumor-promoting agents, such as
phorbol ester, UV radiation, asbestos and crystalline silica
(209, 210). AP-1 complexes are formed by dimers of Jun proto-
oncogene family members (c-Jun, JunB, and JunD) or
heterodimers of Jun family members with the Fos proto-
oncogene family members (c-Fos, FosB, Fra-1, and Fra-2).
AP-1 binds to a specific target DNA site (also known as TRE)
in the promoters of several cellular genes and mediates
immediate early gene expression involved in a diverse set of
transcriptional regulation processes (209, 210). Agents that
activate NF-Î B also activate AP-1. Both of these factors are
regulated by the redox status of the cell. AP-1 activation has
been implicated in cell proliferation and chemical
carcinogenesis. It has been shown to play a critical role in
proliferation of cells. Whether resveratrol affects activation of
AP-1 has been investigated by several groups. We showed that
suppression of NF-Î B by resveratrol coincided with
suppression of AP-1 (201). Resveratrol has been shown to
suppress activation of AP-1 by PMA, TNF and UV. It
inhibited PMA-induced IL-8 production in human monocytic
U-937 cells at protein and mRNA levels which was, at least
partly, due to inhibition of AP-1 activation (211). It also
suppressed PMA-mediated signaling events such as induction
of COX-2 and prostaglandin synthesis in human mammary and
oral epithelial cells (212). Moreover, it inhibited PMA-
mediated activation of PKC and induction of COX-2 promoter
activity by c-Jun. PMA-mediated induction of AP-1 activity was
blocked by resveratrol. Resveratrol also inhibited PMA- or
UV-induced AP-1-mediated activity through inhibition of c-Src
non-receptor tyrosine kinase and MAPK pathways and may
also regulate gene expression of cellular defensive enzymes
such as phase II detoxifying enzymes (213). It also suppressed
TNF-induced AP-1 activity in various cancer cell lines (201).
Resveratrol inhibited the TNF-induced activation of
MAPK and JNK, which are needed for AP-1 activation.
Yu et al. found that resveratrol inhibited phorbol ester and
UV-induced AP-1 activation by interfering with MAPK
pathways (213). They showed that pretreatment with
resveratrol also inhibited the activation of ERK2, JNK1 and
p38 MAPK. Selectively blocking MAPK pathways by
overexpression of dominant-negative mutants of kinases
attenuated the activation of AP-1 by PMA and UVC.
Interestingly, resveratrol had little effect on induction of the
AP-1 reporter gene by active Raf-1, MAPK/ERK kinase
kinase (MEKK)1, or MAPK kinase (MKK)6, suggesting that
it inhibited MAPK pathways by targeting the signaling
molecules upstream of Raf-1 or MEKK1. Indeed, incubation
of resveratrol with the isolated c-Src protein tyrosine kinase
and PKC diminished their kinase activities. Moreover,
modulation of ER activity by 17-‚-estradiol had no effect on
the inhibition of AP-1 by resveratrol. In contrast to these
studies, those of Wolter et al. showed that the AP-1
constituents c-Fos and c-Jun increased on resveratrol
treatment of cells (214). While the DNA-binding activity of
c-Jun remained unchanged, the DNA-binding activity of c-
Fos was significantly enhanced by resveratrol and
piceatannol.
C1e: Resveratrol suppresses Egr-1 activation
Early growth response–1 gene product (Egr-1) is another
transcription factor that plays an important role in
proliferation of cells. It is a member of a family of immediate
early response genes and regulates a number of
pathophysiologically relevant genes that are involved in
growth, differentiation, immune response, wound healing
and blood clotting. Resveratrol selectively up-regulates Egr-1
by an ERK1/2-dependent mechanism in human erythro-
leukemic K562 cells, induces Á-globin synthesis, and causes
erythroid differentiation due to impairment of cell
proliferation, increase in p21
Cip1/WAF1
expression and
inhibition of Cdk2 activity (215). Ragione et al. found that
resveratrol increases Egr-1 and causes differentiation of HL-60
cells (216) and examined its effects on this transcription
factor (215). Up-regulation of p21
Cip1/WAF1
transcription is
prevented by cycloheximide, indicating that an intermediate
protein(s) is required that, in turn, regulates gene expression.
Quantitative analysis of some transcription factors involved
in the erythroid lineage, namely GATA-1, GATA-2 and Egr-1,
indicated that resveratrol selectively up-regulates Egr-1 by an
ERK1/2-dependent mechanism. The presence of an Egr-1
consensus sequence in the p21
Cip1/WAF1
promoter suggests
that this transcription factor directly regulates the expression
of the Cdk inhibitor. Transfection studies with deleted gene
promoter constructs, as well as electrophoretic mobility shift
assay, pull-down and chromatin immunoprecipitation
experiments, substantiated this view, demonstrating that Egr-1
binds in vitro and in vivo to the identified consensus sequence
ANTICANCER RESEARCH 24: 2783-2840 (2004)
2804
of the p21Cip1/WAF1 promoter. Moreover, an Egr-1
phosphorothioate antisense construct hinders p21
Cip1/WAF1
accumulation and the antiproliferative effects of resveratrol.
C1f. Suppression of MAPK by resveratrol
Three different MAPK have been identified: ERK1/2, JNK
and p38 MAPK. While ERK1/2 have been implicated in the
proliferation of cells, JNK and p38 MAPK are activated in
response to different types of stress stimuli. JNK activation
is needed for activation of AP-1; it also mediates apoptosis
in some situations. Numerous studies suggest that
resveratrol modulates all three of these protein kinases
(163, 175, 179, 195, 196, 217, 218). Miloso et al. showed that
resveratrol induced activation of ERK1/2 in human
neuroblastoma SH-SY5Y cells (179). In undifferentiated
cells, resveratrol 1 ÌM induced phosphorylation of ERK1/2,
which was already evident at 2 min, peaked at 10 min and
still persisted at 30 min. A wide range of resveratrol
concentrations (from 1 pM to 10 ÌM) were able to induce
phosphorylation of ERK1/2, while higher concentrations
(50-100 ÌM) inhibited phosphorylation of MAPK. In
retinoic acid-differentiated cells, resveratrol (1 ÌM) induced
an evident increase in ERK1/2 phosphorylation. El-Mowafy
et al. found short-term treatment of porcine coronary
arteries with resveratrol substantially inhibited MAPK
activity (IC
50
, 37 ÌM) and reduced phosphorylation of
ERK1/2, JNK1 and p38 MAPK at active sites. Endothelin-1
enhanced, MAPK activity, phosphorylation and nuclear
translocation in a concentration-dependent manner, but
resveratrol reversed it (217). She et al. showed that
resveratrol activated ERK1/2, JNKs and p38 MAPK in the
JB6 mouse epidermal cell line and induced serine-15
phosphorylation of p53 (196). Stable expression of a
dominant-negative mutant of ERK2 or p38 MAPK
repressed phosphorylation of p53 at serine-15. In contrast,
overexpression of a dominant-negative mutant of JNK1 had
no effect on this phosphorylation. Most importantly,
Aggarwal et al: Resveratrol Inhibits Tumorigenesis
2805
Figure 5. Identification of molecular targets of resveratrol.
ERK1/2 and p38 MAPK formed a complex with p53 after
treatment with resveratrol. Strikingly, resveratrol-activated
ERK1/2 and p38 MAPK, but not JNKs, phosphorylated p53
at serine-15 in vitro. Shih et al. examined the effect of
resveratrol on papillary and follicular thyroid carcinoma cell
lines (175). They found that treatment with resveratrol
(1-10 ÌM) induced activation and nuclear translocation of
ERK1/2. Cellular abundance of the oncogene suppressor
protein p53, serine phosphorylation of p53, and abundance
of c-fos, c-Jun, and p21
Cip1/WAF1
mRNAs were also
increased by resveratrol. Inhibition of the MAPK pathway
by either H-Ras antisense transfection or PD 98059, MAPK
kinase inhibitor, blocked these effects. Thus, resveratrol
appears to act via a Ras-MAPK kinase-MAPK signal-
transduction pathway to increase p53 expression, serine
phosphorylation of p53 and p53-dependent apoptosis in
thyroid carcinoma cell lines.
She et al. showed the interesting involvement of JNK in
resveratrol-induced activation of p53 (195). They found that
resveratrol activated JNKs at the same dosage that inhibited
tumor promoter-induced cell transformation. Stable
expression of a dominant-negative mutant of JNK1 or
disruption of the Jnk1 or Jnk2 gene markedly inhibited
resveratrol-induced p53-dependent transcription activity and
induction of apoptosis. Furthermore, resveratrol-activated
JNKs were shown to phosphorylate p53 in vitro, but this
activity was repressed in the cells expressing a dominant-
negative mutant of JNK1 or in Jnk1 or Jnk2 knockout (Jnk1
-/-
or Jnk2
-/-
) cells. These data suggest that JNKs act as mediators
of resveratrol-induced activation of p53 and apoptosis, which
may occur partially through p53 phosphorylation. Woo et al.
showed that resveratrol inhibited PMA-induced matrix
metalloproteinase (MMP)-9 expression by inhibiting JNK
(218). From these results, it is clear that resveratrol can
modulate all three MAPKs, which leads to modulation of
gene expression. Resveratrol appears to cause activation of
MAPK in some cells and inhibition in others. This variability
may depend on the cell type and the dose of resveratrol used.
Stewart and O'Brian showed that resveratrol antagonized
EGFR-dependent ERK1/2 activation in human androgen-
independent prostate cancer cells with associated isozyme-
selective PKC-· inhibition (163). They found that
resveratrol suppressed EGFR-dependent ERK1/2 activation
pathways stimulated by EGF and PMA in human AI PrCa
PC-3 cells in vitro. Resveratrol abrogation of a PKC-
mediated ERK1/2 activation response in PC-3 cells
correlated with isozyme-selective PKC-· inhibition.
C1g. Suppression of protein kinases by resveratrol
PKC has been shown to play a major role in tumorigenesis.
The PKC isozyme subfamily consists of cPKC-·, -‚ and -Á,
nPKC-D and -Â, and ·PKC-˙. Numerous reports indicate
that resveratrol can inhibit PKC (127, 139, 153, 218-221).
Garcia-Garcia et al. showed that resveratrol was
incorporated into model membranes and inhibited PKC-·
activity (219). Resveratrol activated by phosphatidylcholine/
phosphatidylserine vesicles inhibited PKC-· with an IC
50
of
30 ÌM, whereas that activated by Triton X-100 micelles
inhibited PKC-· with an IC
50
of 300 ÌM. These results
indicate that the inhibition of PKC-· by resveratrol can be
mediated, at least partially, by membrane effects exerted
near the lipid-water interface. Stewart et al. showed that
resveratrol preferentially inhibited PKC-catalyzed
phosphorylation of a cofactor-independent, arginine-rich
protein substrate by a novel mechanism (139). While
resveratrol has been shown to antagonize both isolated and
cellular forms of PKC, the weak inhibitory potency observed
against isolated PKC cannot account for the reported
efficacy of the polyphenol against PKC in cells. Stewart et
al. analyzed the mechanism of PKC inhibition by resveratrol
and found that resveratrol has a broad range of inhibitory
potencies against purified PKC that depend on the nature
of the substrate and the cofactor dependence of the
phosphotransferase reaction. Resveratrol weakly inhibited
the Ca
2+
/phosphatidylserine-stimulated activity of a purified
rat brain PKC isozyme mixture (IC
50
, 90 ÌM) by
competition with ATP (K
i
, 55 ÌM). Consistent with the
kinetic evidence for a catalytic domain-directed mechanism
was resveratrol’s inhibition of the lipid-dependent activity
of PKC isozymes with divergent the regulatory domains, and
it was even more effective in inhibiting a cofactor-
independent catalytic domain fragment of PKC generated
by limited proteolysis. This suggested that regulatory
features of PKC might impede resveratrol inhibition of the
enzyme. To explore this, the authors examined the effects
of resveratrol on PKC-catalyzed phosphorylation of the
cofactor-independent substrate protamine sulfate, which is a
polybasic protein that activates PKC by a novel mechanism.
Resveratrol potently inhibited protamine sulfate
phosphorylation (IC
50
, 10 ÌM) by a mechanism that entailed
antagonism of the activation of PKC by protamine sulfate
and did not involve competition with either substrate.
Protein kinase D (PKD) is a member of the PKC
superfamily with distinctive structural, enzymic and
regulatory properties. Identification of the cellular
function(s) of PKD has been hampered by the absence of a
selective inhibitor. Stewart et al. compared the effects of
resveratrol against the autophosphorylation reactions of
PKC isozymes to those against the autophosphorylation
reactions of the novel phorbol ester-responsive kinase PKD
(127). They found that resveratrol inhibited PKD
autophosphorylation, but had only negligible effects against
the autophosphorylation reactions of representative
members of each PKC isozyme subfamily (cPKC-·, -‚1 and
-Á, nPKC-D and -Â, and ·PKC-˙). Resveratrol was
ANTICANCER RESEARCH 24: 2783-2840 (2004)
2806
comparably effective against PKD autophosphorylation
(IC
50
, 52 ÌM) and PKD phosphorylation of the exogenous
substrate syntide-2 (IC
50
, 36 ÌM). The inhibitory potency of
resveratrol against PKD is in line with those observed in
cellular systems and against other purified enzymes and
binding proteins that are implicated in the cancer
chemopreventive activity of the polyphenol. Thus, PKD
inhibition may contribute to the cancer chemopreventive
action of resveratrol. Haworth et al. showed inhibition of
PKD by resveratrol, not only in vitro but also in intact cells
(220). Atten et al. demonstrated that resveratrol treatment
significantly inhibited PKC activity of KATO-III human
gastric adenocarcinoma cells and of human recombinant
PKC-· (153). Woo et al. showed that resveratrol inhibited
PMA-mediated PKC-¢ activation, which led to suppression
of MMP-9 (218).
The COP9 signalosome (CSN), purified from human
erythrocytes, possesses kinase activity that phosphorylates
proteins such as c-Jun and p53, with consequences for their
ubiquitin-dependent degradation. Uhle et al. showed that
resveratrol could block the CSN-associated kinases protein
kinase CK2 and PKD and induce degradation of c-Jun in
HeLa cells (221).
C1h. Modulation of NO/NOS expression by resveratrol
Synthesis of NO is dependent on expression of an inducible
enzyme, iNOS. The expression of this enzyme is regulated
by the transcription factor NF-Î B. Production of NO has
been shown to mediate antiproliferative effects in various
cell types. NO also been linked with pro-inflammatory
effects. Resveratrol has been reported to both enhance and
suppress production of NO (92, 154, 194, 222). Kageura et
al. reported that resveratrol analogues had inhibitory
activity against NO production in LPS-activated
macrophages (IC
50
, 11-69 ÌM) (92). Furthermore, the active
stilbenes (rhapontigenin, piceatannol and resveratrol) did
not inhibit iNOS activity, but they inhibited NF-Î B
activation following expression of iNOS. Chung et al.
examined the effect of ·-viniferin, a trimer of resveratrol,
in a mouse model of carrageenin-induced paw edema (222).
They found that ·-viniferin at doses >30 mg/kg (p.o.) or >3
mg/kg (i.v.) showed significant anti-inflammatory activity on
this edema. ·-Viniferin at doses of 3-10 ÌM inhibited NO
production in LPS-activated Raw 264.7 cells when ·-
viniferin and LPS were applied simultaneously, but not
when ·-viniferin was applied 12 h after LPS stimulation. ·-
Viniferin inhibited synthesis of the iNOS transcript with an
IC
50
value of 4.7 ÌM.
Hsieh et al. found that resveratrol induced NOS in
cultured pulmonary artery endothelial cells, which led to
inhibition of their proliferation (194). Holian et al. found
that resveratrol stimulated NOS activity in human gastric
adenocarcinoma SNU-1 cells (154). They suggested that the
antioxidant action of resveratrol toward gastric
adenocarcinoma cells may reside in its ability to stimulate
NOS to produce low levels of NO, which, in turn, exerts
antioxidant action. Thus, whether resveratrol induces or
inhibits NO production depends on the cell system, inducer
and other conditions.
C1i. Suppression of growth factor and associated
protein tyrosine kinases by resveratrol
Because resveratrol exhibits antiproliferative effects against
a wide variety of tumor cells and the effects of various growth
factors are mediated through protein tyrosine kinases, it is
possible that resveratrol either down-regulates the expression
of growth factors and growth factor receptors or suppresses
the activity of protein tyrosine kinases required for their
activity. Kaneuchi et al. found that resveratrol treatment
significantly decreased EGF expression in Ishikawa
endometrial cancer cells (183). Palmieri et al. found that
tyrosine kinase activities from particulate and cytosolic
fractions of placenta were inhibited by resveratrol and
piceatannol (223). Oliver et al. showed that piceatannol
(3,4,3',5'-tetrahydroxy-trans-stilbene) preferentially inhibited
the activity of Syk protein tyrosine kinase as compared with
Lyn when added to in vitro assays with isolated enzymes
(224). Selective inhibition of Syk in this manner blocked
receptor-mediated downstream cellular responses (inositol
1,4,5-trisphosphate production, secretion, ruffling and
spreading). We showed that piceatannol inhibited H
2
O
2
-
induced NF-Î B activation through inhibition of Syk kinase
(225). These reports suggest that resveratrol and its
analogues can potentially suppress growth factors, growth
factor receptors and their associated protein tyrosine kinases.
Resveratrol exerts an inhibitory effect in EGF-induced
cell transformation (226). It also inhibits proliferation of the
breast cancer cell line MDA-MB-468 through alteration in
autocrine growth modulators such as TGF-·, TGF-‚, PC
cell-derived growth factor, and insulin-like growth factor I
receptor mRNA (129). Moreover, it decreases hepatocyte
growth factor-induced cell scattering and invasion by an
unidentified postreceptor mechanism in HepG2 cells (173).
C1j. Suppression of COX-2 and LOX by resveratrol
The enzymes COX-2 and lipooxygenase (LOX) play
important roles in inflammation. Both of these enzymes are
regulated by the transcription factors NF-Î B and AP-1. The
products of these enzymes also regulate proliferation of cells.
Whether resveratrol modulates expression of these enzymes
has been investigated by numerous groups (141, 142, 212,
222, 227, 228). Subbaramaiah et al. showed that resveratrol
inhibits COX-2 transcription and activity in phorbol ester-
Aggarwal et al: Resveratrol Inhibits Tumorigenesis
2807
treated human mammary epithelial cells (141). Transient
transfections utilizing COX-2 promoter deletion constructs
and COX-2 promoter constructs, in which specific enhancer
elements were mutagenized, indicated that the effects of
PMA and resveratrol were mediated via a cAMP response
element. Resveratrol inhibited the PMA-mediated activation
of PKC. Overexpressing PKC-·, ERK1 and c-Jun led to 4.7-,
5.1- and 4-fold increases in COX-2 promoter activity,
respectively. These effects were inhibited by resveratrol.
Resveratrol blocked PMA-dependent activation of AP-1-
mediated gene expression. In addition to these effects on
gene expression, we found that resveratrol also directly
inhibited the activity of COX-2. These data are likely to be
important for understanding the anticancer and anti-
inflammatory properties of resveratrol. Chung et al. showed
that ·-viniferin inhibited COX-2 activity with an IC
50
value
of 4.9 ÌM, and at doses of 3-10 ÌM, inhibited synthesis of
COX-2 transcript in LPS-activated murine macrophages Raw
264.7 (222). MacCarrone et al. demonstrated that resveratrol
acted as a competitive inhibitor of purified 5-LOX and 15-
LOX and prostaglandin H synthase, with inhibition constants
of 4.5 ÌM (5-LOX), 40 ÌM (15-LOX), 35 ÌM (COX activity
of prostaglandin H synthase), and 30 ÌM (peroxidase activity
of prostaglandin H synthase) (227).
C1k. Suppression of cell-cycle proteins by resveratrol
Numerous reports indicate that resveratrol inhibits
proliferation of cells by inhibiting cell-cycle progression (122,
135, 145, 147, 151, 161, 165, 167, 187, 191, 194, 229). Various
reports indicate that resveratrol inhibits different cells at
different stages of the cell-cycle. The arrest of cells in G1-
phase (165), S-phase (122, 151, 161, 187, 191), S/G2-phase
(194) and G2-phase (147) of the cell-cycle has been reported.
Why the effects of resveratrol on different cell types vary so
widely is not clear. Which cell-cycle proteins are modulated
by resveratrol has been investigated in detail. Wolter et al.
showed the down-regulation of the cyclin D1/Cdk4 complex
by resveratrol in colon cancer cell lines (145). Yu et al.
showed that, following treatment of H22 tumor-bearing mice
with resveratrol at 10 or 15 mg/kg bodyweight for 10 days, the
growth of transplantable liver cancers was inhibited by 36.3%
or 49.3%, respectively (229). The levels of expression of cyclin
B1 and Cdc2 protein were decreased in treated tumors,
whereas the expression of cyclin D1 protein did not change.
Liang et al. showed that resveratrol induced G2 arrest
through the inhibition of Cdk7 and Cdc2 kinases in colon
carcinoma HT-29 cells (147). Larrosa et al. showed that
resveratrol and the related molecule 4-hydroxystilbene
induced S- phase arrest and up-regulation of cyclins A, E and
B1 in human SK-Mel-28 melanoma cells (167). Thus, it is
clear that the effects of resveratrol on the cell-cycle are highly
variable. Kuwajerwala et al. showed that resveratrol had a
dual effect on DNA synthesis (161). At concentrations of 5-
10 ÌM, it caused a 2- to 3-fold increase in DNA synthesis, and
at doses ≥15 ÌM, it inhibited DNA synthesis. The increase
in DNA synthesis was seen only in LNCaP cells, not in the
androgen-independent DU145 prostate cancer cells or in
NIH/3T3 fibroblast cells. The resveratrol-induced increase in
DNA synthesis was associated with enrichment of LNCaP
cells in S-phase and concurrent decreases in nuclear
p21
Cip1/WAF1
and p27
Kip1
levels. Furthermore, consistent
with the entry of LNCaP cells into the S-phase, there was a
dramatic increase in nuclear Cdk2 activity associated with
both cyclin A and cyclin E.
C1l. Suppression of adhesion molecules by resveratrol
Various cell-surface adhesion molecules, including intracellular
adhesion molecule (ICAM)-1, vascular cell adhesion molecule
(VCAM)-1 and endothelial-leukocyte adhesion melecule
(ELAM)-1, are regulated by NF-ÎB. These molecules play an
essential role in adhesion of tumor cells to endothelial cells and
thus mediate tumor cell metastasis. Several groups have
examined the effect of resveratrol on the adhesion of cells to
the endothelial cells. Ferrero et al. examined the activity of
resveratrol on granulocyte and monocyte adhesion to
endothelium in vitro (230, 231). They showed that resveratrol, at
concentrations as low as 1 ÌM and 100 nM, significantly
inhibited ICAM-1 and VCAM-1 expression by TNF·-
stimulated HUVEC and LPS-stimulated human saphenous
vein endothelial cells (HSVEC), respectively. They also showed
that resveratrol induced significant inhibition of the adhesion
of U-937 monocytoid cells to LPS-stimulated HSVEC. Such
inhibition was comparable with that obtained when anti-
VCAM-1 monoclonal antibody was used instead of resveratrol.
Resveratrol also significantly inhibited the adhesion of
neutrophils to TNF·-stimulated NIH/3T3 ICAM-1-transfected
cells, whereas neutrophils activated by formyl-methionyl-leucyl-
phenylalanine did not significantly modify adhesion to NIH/3T3
ICAM-1-transfected cells. Pendurthi et al. also showed that
resveratrol suppressed agonist-induced monocyte adhesion to
cultured human endothelial cells (125). Thus, it is clear that
resveratrol affects the expression of adhesion molecules, most
likely through down-regulation of NF-ÎB.
C1m. Suppression of androgen receptors by resveratrol
Via their receptor AR, androgens play a role in prostate
cancer etiology (159, 285). Mitchell et al. demonstrated that
resveratrol had inhibitory effects on androgen action in the
LNCaP prostate cancer cell line (159). They found that
resveratrol repressed different classes of androgen up-
regulated genes at the protein or mRNA level, including
PSA, human glandular kallikrein-2, AR-specific coactivator
ARA70, and the Cdk inhibitor p21
Cip1/WAF1
. This inhibition
ANTICANCER RESEARCH 24: 2783-2840 (2004)
2808
is probably attributable to a reduction in AR level at the
transcription level, inhibiting androgen-stimulated cell growth
and gene expression. These results suggest that resveratrol
may be a useful chemopreventive / chemotherapeutic agent
for prostate cancer.
C1n. Suppression of PSA by resveratrol
Hsieh et al. demonstrated that resveratrol inhibited the
proliferation of LNCaP cells and expression of the prostate-
specific gene PSA. A 4-day treatment with resveratrol
reduced the levels of intracellular and secreted PSA by
approximately 80%, as compared to controls (156). They
found that this change in PSA was not due to a change in
AR expression. Thus, it would appear that the prostate
tumor marker PSA is down-regulated by resveratrol, by a
mechanism independent of changes in AR.
C1o. Suppression of inflammatory cytokine
expression by resveratrol
Because resveratrol down-regulates NF-Î B, which is known
to mediate inflammation, it is possible that resveratrol also
down-regulates the expression of inflammatory cytokines.
Wang et al. showed that resveratrol inhibited IL-6 production
in cortical mixed glial cells under hypoxic/hypoglycemic
conditions followed by reoxygenation (232). Zhong et al.
demonstrated the inhibitory effect of resveratrol on IL-6
release by stimulated peritoneal macrophages of mice (233).
Shen et al. found that resveratrol suppressed IL-8 gene
transcription in phorbol ester-treated human monocytic cells
(211). Wadsworthe et al. showed that resveratrol had no
effect on LPS-induced TNF· mRNA in the macrophage cell
line RAW 264.7, but decreased LPS-stimulated TNF·
release, as measured by ELISA (234). Culpitt et al.
determined whether resveratrol would inhibit cytokine
release in vitro by alveolar macrophages from patients with
chronic obstructive pulmonary disease (COPD) (235). They
showed that resveratrol inhibited basal release of IL-8 in
smokers and patients with COPD by 94% and 88%,
respectively, and inhibited granulocyte-macrophage colony-
stimulating factor (GM-CSF) release by 79% and 76%,
respectively. Resveratrol also inhibited stimulated cytokine
release. Resveratrol reduced IL-1‚-stimulated IL-8 and GM-
CSF release in both smokers and COPD patients to below
basal levels. Moreover, resveratrol inhibited cigarette smoke
media (CSM)-stimulated IL-8 release by 61% and 51%,
respectively, in smokers and COPD patients, and inhibited
GM-CSF release by 49% in both subject groups.
Boscolo et al. elucidated the "in vitro" effects of resveratrol
on human PBMC proliferation and cytokine release (236).
Spontaneous PBMC proliferation was unaffected by
resveratrol, while resveratrol at a concentation of 100 ÌM
inhibited PHA-stimulated PBMC proliferation by 69%. The
proliferation stimulation index (i.e., the ratio of PHA-
stimulated PBMC proliferation/spontaneous PBMC
proliferation) of cultures containing 100 ÌM resveratrol was
very low in relation to the control, while the proliferation
stimulation index values at resveratrol concentrations of
10 ÌM and 100 nM were similar and slightly higher (without
statistical significance), respectively. Resveratrol strongly
inhibited PHA-stimulated interferon (IFN)-Á and TNF·
release from PBMC at a concentration of 100 ÌM, but not
concentrations of 10 ÌM or 100 nM. The concomitant
immune effects of resveratrol on PBMC proliferation and
release of IFN-Á and TNF· may be explained by an inhibitory
effect on transcription factor NF-Î B.
C1p. Suppression of angiogenesis, invasion and