The undersigned, an independent panel of recognized experts (hereinafter referred
to as the Expert Panel)3, qualified by their scientific training and relevant national and
international experience to evaluate the safety of food and food ingredients, was
convened by EAS Consulting Group LLC at the request of ATLA Holdings, LLC, USA,
to determine the Generally Recognized As Safe (GRAS) status of trans-Resveratrol as a
nutrient supplement [21 CFR 170.3(0)(20)]~ at levels up to 10 ppm in bottled water (near
waters). A comprehensive search of the scientific literature for safety and toxicity
information on resveratrol was conducted through October 2006 and made available to
the Expert Panel. The Expert Panel independently and critically evaluated materials
submitted by ATLA Holdings, LLC and other materials deemed appropriate or necessary.
Following an independent, critical evaluation, the Expert Panel conferred and
unanimously agreed to the decision described herein.
Historical Perspective
trans-Resveratrol, a polyphenol, occurs naturally in grapes, peanuts, and a number
of other plants. It is commonly found in foods/drinks made from grapes and peanuts, and
in a number of herbal remedies. As a constituent of red wine, resveratrol has been
identified as one possible explanation for the 'French paradox,' i.e the finding that the
incidence of coronary heart disease is relatively low in southern France despite the high
intake of dietary saturated fats or similar risk factor profile (Hendler and Rorvik, 2001).
Recent reports on the potential for resveratrol to inhibit the development of cancer and
extend life expectancy in animal and cell culture models have continued to generate
scientific interest. Available scientific evidence show that resveratrol has a wide range of
desirable biological effects such as cardioprotection (Hung et al., 2000),
chemoprevention (Jang and Pezzuto, 1999), anticancer (Gusman et al, 200) and
prolongation of life-span in several species (Howitz et al., 2003; Valenzano et al., 2006;
Horn, et al, 2007).
Description, occurrence, manufacturing process and specifications
Resveratrol (CAS No. 501-36-0) is chemically known as 3,4',5-stilbenetriol or
3,4',5-trihydroxystilbene (Figure 1). It is produced by various plants to help defend
against invading fungi, stress, injury, infection, and too much sunlight. Resveratrol may
exist in both cis- and trans-stereoisomeric forms. Both cis- and trans-resveratrol also
occur in their respective glucoside forms (bound to a glucose molecule). Resveratrol-3-O-beta-glucoside is also called piceid. Several methods including high-performance liquid
L, chromatography (HPLC), gas chromatography (GC), gas chromatography-mass
spectrometry (GC-MS), and capillary electrophoresis (CE) have been employed to extract
resveratrol from wine and to isolate the trans- and cis- isomers of resveratrol (NIEHS,
2002). Resveratrol (>98% purity) can be isolated from Polygonum cuspidatum by
employing high-speed counter-current chromatography (Yang et al., 2001).
Figure 1. Chemical structure of resveratrol
The ATLA Holdings, LLC�s trans-resveratrol is a purified, off-white to creamcolored
solid product in powder form. It is extracted from the roots of P. cuspidatum Sieb
Et Zucc (Knotweed herb; bushy knotweed, Japanese knotweed). trans-Resveratrol is
isolated by solvent extraction and the purified product is reported to contain >99% transresveratrol.
General descriptive parameters and properties of trans-resveratrol are
summarized in Table 1.
Ocurrence
Resveratrol is found in over 70 common plant species (NIEHS, 2002). The
presence of trans-resveratrol in varying concentrations has been reported in grapes,
peanuts, eucalyptus, spruce, lily, mulberries, groundnut, members of the knotweed and
hellebore genera (Polygonum and Helleborus), and fescue grass. The highest
concentration of resveratrol has been reported in P cuspidatum. Resveratrol is primarily
found in lignified plant tissues, in leaves, and in berries of Vaccinum species, including
blueberries, bilberries, and cranberries. trans-Resveratrol was also detected in vines, and
leaf tissues of Vitis vinifera infected with fungi or exposed to UV light. Infection of
grapes with the fungus Botrytis cinerea (gray mold) has been reported to result in
increased concentrations of resveratrol in nearby (unaffected) grapes. Stimulating a grape
plant's production of resveratrol and other defense chemicals has been shown to increase
its resistance to fungal infection.
The concentration of resveratrol in food substances varies widely. Grape skin
(fresh) contains approximately 50 to 100 pg/g (0.22 to 0.44 pmol/g) of trans-resveratrol
(NIEHS, 2002). Stewart et a1 (2003) reported that resveratrol accounts for 5-10% of the
grape skin biomass. Generally, non-muscadine red wine is reported to contain between
0.2 and 5.8 mg resveratrol/L (Gu et al., 1999). The amount of reseveratrol in wine
depends on the grape variety. White wine has much less resveratrol compared to red, the
reason being that red wine is fermented with the skins, allowing the release of resveratrol
from skin, whereas white wine is fermented after the skin has been removed. Wines
produced from muscadine grapes, however, both red and white, contain higher levels of
resveratrol, up to 13.4 mg/L (NIEHS, 2002).
Manufacturing process
Resveratrol is extracted from roots of the plant P cuspidaturn. Before the
extraction process, the plant root is identified as P. cuspidaturn Sieb. Et Zucc. In plants,
the majority of the resveratrol present occurs as glucoside called piceid. The
concentration of piceid and resveratrol in P. cuspidaturn root is generally between 1-2%
and 0.1-0.3%, respectively. The root is dried and cut to small pieces. The extraction of
resveratrol from the P. cuspidaturn root also results in extraction of piceid. Piceid is
extracted from the pieces of roots with ethanol. The piceid extract is evaporated to
dryness under reduced pressure. The isolated piceid is hydrolyzed with sulfuric acid.
Hydrolyzation of 1 g piceid yields approximately 0.58 g resveratrol. Normally hydrolysis
produces a product containing approximately 50% resveratrol. Subsequent purification
steps results in highly purified resveratrol (>990/) concentrate. The solvents and acids
used in the isolation and purification process are food grade.
Identity and specifications
Physical characteristics and specifications of resveratrol from ATLA Holdings,
LLC, are presented in Table 2. Analytical results of five lots from non-consecutive
batches indicate that the resveratrol product meets these specifications.
Technical effects
Resveratrol is intended for uses in food as a nutrient supplement, for individuals
who wish to increase their daily intake of resveratrol. The nutritive contribution of
resveratrol is well recognized. It is naturally present in a variety of commonly consumed
foods such as grapes, peanut, blueberries and mulberry. As indicated earlier, resveratrol
has wide-ranging biological activities including inhibition of lipid peroxidation, freeradical
scavenging activity, anti-inflammatory activity, modulation of lipid metabolism,
and anticancer activity. Its antioxidant activities are believed to be due to its protective
effects on the cellular membranes. Although resveratrol is present at certain amount in
foods, its supplementation to food is aimed at gaining certain health benefits.
Resveratrol�s effects are reported to be due to its amphiphatic character (as the structure
has both hydrophilic and hydrophobic sites), which allows the protection of cellular and
subcellular components.
Historical and current uses
Humans have been exposed to resveratrol since ancient times, at least as long as
grapes, wine-making and the consumption of other resveratrol-containing plants have
been popular. In traditional Asian remedies, the root of P. cuspidaturn, a source of
resveratrol, has long been used as a circulatory tonic, among other uses (NIEHS, 2002).
Approximately 4500 years ago, Ayurveda, the ancient medicinal book of Hindus
described �Darakchasava� (fermented juice of red grapes) as a heart tonic (Hendler and
Rorvik, 2001). At present, several dietary supplements marketed in the United States
contain trans-resveratrol. The source of these supplements is primarily an extract of P.
cuspidaturn or ground dried red grape skins. The P. cuspidaturn extracts are usually
standardized to deliver approximately 8% resveratrol. Some supplements deliver 16 mg
resveratrol per serving or higher.
Intended uses and estimated daily intake
ATLA Holdings, LLC, intends to use trans-resveratrol as a nutrient supplement
[2lCFR170.3(0)(20)]~ at levels up to 10 ppm in bottled water (�near water�) [21CFR
170.3(n)(3)I6. The daily estimated intake of trans-resveratrol was calculated using the
intended maximum use level value and the consumption of water (bottled) containing
resveratrol. To obtain the estimated daily intake of resveratrol from the intended uses,
CanTox Health Sciences International was commissioned to undertake these
determinations. The CanTox report is attached as Appendix 11.
As resveratrol is intended to be marketed in bottled water, intake analysis based
on consumption of bottled water was considered as the most appropriate and applicable
method for exposure analysis. The representative data for bottled water consumption
from individual dietary records in the CSFII 1994-1996, 1998 is lacking. Hence, the
methodology presented by the Environmental Protection Agency for the calculation of
direct bottled water consumption from the CSFII 1994-1996 was replicated in the intake
assessment performed by CanTox using data from the CSFII 1994-1996, 1998.
Household and sample person data from additional record types (Record types 15 and 25,
respectively) available within the CSFII dataset were merged with the individual dietary
record file (Record type 30) to determine the amount of plain bottled water consumed by
survey respondents. The method used differed only in the manner in which files were
merged. A new food code was generated to represent bottled water intake (95000000,
Water, bottled) in the current intake assessment. The intakes of bottled water represented
by this new food code were then added to the individual dietary record for each bottled
water consumer, in order to allow estimation of the intake of resveratrol from bottled
water, and to derive the total intake of resveratrol from all currently identified food-uses.
Based on this analysis, the estimated consumption of added resveratrol from the intended
use levels of 10 ppm at mean and 90th percentile for all-users total population (�users
only�) is determined as 2.09 and 3.99 mg resveratrol/person/day or 0.02 and 0.07
mg/kg/day. The CanTox derived estimated daily intake of resveratrol by the US
population from the intended food uses of resveratrol is summarized in Table 3.
Categories in the US by Population Group (1994-1996,1998 USDA CSFII Data)">
In a report prepared in 2000 for the Rockefeller University and conducted by
Yankelovich Partners, it was determined that of the total daily estimated intake of 6.1
servings (each serving consisting of eight ounce fluid) of water consumed in the US, 2.3
servings were consumed daily as bottled water (Saint-Jean, 2001). Based on this report,
the daily consumption of bottled water in the US is approximately 544 ml(2.3 servings X
8 ounce = 18.4 ounce; 18.4 ounces X 29.6 ml = 544 ml). Based on the results of this
report, the intended use of resveratrol at use levels of 10 ppm in bottled drinking water
will result in an estimated intake of 5.44 mg resveratrol/person/day or 0.09 mg/kg/day
(for an individual weighing 60 kg). In this survey, adults were the only population group
from which consumption data were obtained. Additional details of the report and
methods of determination used were not available. However, the resulting estimated daily
intake is lower than the tolerable daily intake.
Daily intake via natural occurrence in food
Humans have been exposed to resveratrol since ancient times primarily through
the widespread consumption of grapes and red wine. The current and primary sources of
human exposure to trans-resveratrol are primarily via ingestion of peanuts, grapes, and
the products containing either these products or derived from them (particularly wine)
and via consumer use of dietary supplements. The concentration of resveratrol in grapes,
grape juice, and wine is highly variable. Per capita wine consumption in the US during
2001 was estimated as 8.55 liters (Wine Institute, 2006). The highest per capita
consumption of wine is reported in Luxembourg (64.02 literdyear). Based on CSFII
�eaters only� data (2002), per cupifa total wine consumption from all sources (including
wine used in food preparation) in the US is reported as 16 ml/person/day. Details about
the levels of trans-resveratrol in different wines are presented in Table 4. Compared to
grape products, the concentration of resveratrol in peanuts and peanut products is
relatively low. In roasted peanuts, boiled peanuts and peanut butter, resveratrol
concentrations were reported as 0.05, 5.14, and 0.32 pg/g, respectively (NIEHS, 2002).
In other reports, the levels of trams-resveratrol in peanuts ranged from 0.02 -1.79 pg/g
(0.09-7.84 mol/g).
Analysis of exposure from the reports described above indicate that the daily
intake of resveratrol from existing food and beverage sources will be a small amount and
difficult to determine accurately. As noted earlier, red wine is much higher than white
wine in resveratrol as red wine is prepared by fermentation with the skins. If one
considers for exposure estimate purposes that all the wine consumed is red wine with
maximum levels of resveratrol (13.4 mg/L) and per capita consumption of wine (16 ml),
the resulting intake of resveratrol in US will be 0.21 mg/person/day. As the reported
intake of wine is high in Luxembourg, the likely intake of resveratrol is expected to be
eight-fold greater than in the US.
An individual ingesting red wine (containing the highest levels of resveratrol)
regularly and in moderation (2 drinkdday = -300 ml) is likely to ingest 4 mg
resveratrol/day or 0.067 mg/kg/day for an individual weighing 60 kg. These values are
similar to the estimated daily intake of resveratrol (0.07 mg/kg/day at 90th percentile)
resulting from its intended use in bottled water. Although, exact figures are not available,
it is well known that there are significant numbers of individuals who consume red wine
daily in moderation. As noted earlier, the amount of resveratrol in dietary supplements
varies from product to product and depends on dosage recommendations. The dietary
supplement recommended dosage of resveratrol covered a range from 3 to 1000 mg
(NIEHS, 2002).
Above described information suggest that red wine is the major natural source of
resveratrol intake. Very small amount of resveratrol is ingested by other sources such as
peanut. Moderate consumers of red wine containing high levels of resveratrol are likely
to ingest about 4 mg resveratroUday or 0.067 mg/kg/day. The estimated 90th percentile
daily intake of resveratrol(3.99 mg/person) is similar to that of consumers of red wine (in
moderation) with high levels of resveratrol(4 mg/person).
Consumption summary
The intended use of resveratrol at levels of 10 ppm in bottled water (�near
waters�) will result in estimated mean and 90th percentile exposures of 2.09 and 3.99 mg
resveratrol/person/day or 0.04 and 0.07 mgikglday, respectively. The mean intake of
resveratrol from consumption of wine in US is estimated as 0.21 mg/person/day (0.004
mg/kg/day). The intake of resveratrol from other natural dietary sources is very small.
The total estimated intake of resveratrol from all sources including its natural
consumption is approximately 4 mglpersodday. The estimated daily intake from
intended uses of resveratrol is below acceptable daily intake determined on the basis of
available safety information on resveratrol (see discussion below). In summary the 90�
estimated intake of resveratrol from the intended uses is determined as 3.99
mglpersodday .
Toxicology
frans-Resveratrol is naturally present in a number of commonly consumed foods
such as grapes (red-wine) and peanuts. Because of its potential health benefits, there has
been a considerable effort to elucidate the potential biological effects of trans-resveratrol.
As a result, literature is full of information on resveratrol. The relevant biological and
toxicological studies on trans-resveratrol are included in the following section. The
results of metabolism, subchronic, chronic toxicity, and carcinogenicity studies are also
discussed from a mechanism of toxicity perspective.
Absorption, metabolism and excretion
Available animal studies and human data show that following oral ingestion,
resveratrol is absorbed from the gastrointestinal tract. A summary of the metabolic and
toxicokinetic studies of resveratrol is presented in Table 5. The human (in vivo) studies of
resveratrol detailing absorption, metabolism and excretion are described in section 2.4.
In vitro studies
Andlauer et al. (2000) investigated the bioavailablity of bans-resveratrol in the
perfused small intestine of the rat (Table 5). Following perfusion of the small intestine
with a synthetic perfusate containing either 28, 34, or 57 pM resveratrol, the majority of
the absorbed resveratrol was found in the luminal effluent (54%). Approximately 20% of
I, the administered resveratrol appeared at the vascular site, with the major product being
the conjugated glucuronide form. In another study, Kuhnle et al. (2000) reported that
small amounts of unmetabolized resveratrol were absorbed across the enterocytes of the
jejunum and ileum and significant amounts of its glucuronide were found in the serosal
fluid (Table 5). In this study, 100 pM of resveratrol was administered and the
glucuronide noted in jejunum and ileum was reported as 1.19 and -0.45 nmol/cm,
respectively.
In human liver microsomes, the maximum rate of trans-resveratrol
glucuronidation occurred at a neutral pH, and the resveratrol-glucuronide amount
increased linearly with time up to 40 min (Table 5). The highest rate of glucuronidation
was noted at concentrations of up to 1 mM resveratrol and 1 mM uridine 5�-
diphosphoglucuronic acid in incubation mixture. The reaction of resveratrol sulphation at
a concentration up to 2 pM resveratrol and 0.4 pM 3�-phosphoadenosine-5�-
phosphosulphate was also linear for about 40 min. The rates of resveratrol sulphation
were inhibited (mixed and noncompetitive) by flavonoids such as quercetin, fisetin,
myricetin, kaempferol and apigenin. Flavonoids also inhibited resveratrol glucuronidation,
but the extent of inhibition was less than that for sulphation (De Santi et al., 2000a,b,c).
In vivo studies
Oral administration of resveratrol (86 pg/kg or 43 pgkg) daily for 15 days in red
wine to rats indicates that resveratrol was rapidly absorbed from the intestine in rats
(Bertelli et al., 1998a) (Table 5). Within one hour of administration, maximum levels of
resveratrol were noted in the blood. Following single (86 pg/kg) or repeated (43
pgikg/day for 15 days) administration of resveratrol to rats, the highest amount was noted
in liver (20.7 and 53.5 ng/g). The investigators reported that the "main excretion
pathways appear to be renal." Kinetic studies revealed equilibrium between the absorbed
and the eliminated resveratrol. In another report by these investigators, significant cardiac
bioavailability was noted, as well as a strong affinity for the liver and kidneys (Bertelli et
al., 1998b). Zhu et al. (2000) reported that following intraperitoneal administration of
trans-resveratrol (2 mg/kg), it was rapidly absorbed and the concentration in rat blood
declined in a "two-exponential'' manner.
Abd El-Mohsen et al. (2006) investigated the distribution of [3H]-transresveratrol
and its metabolites, following gavage administration. Male Sprague Dawley
rats were gavaged with 50 mgkg ['H]-radiolabeled resveratrol. At 2 hours after the
injection, plasma concentrations of resveratrol reached 1.7% of the administered dose,
while liver and kidney concentrations were 1 .O and 0.6%, respectively. At the end of 18
hours, plasma levels were 0.5% in plasma and a total of 0.35% in tissues. At 18 hours,
L" kidney and liver concentrations fell to 10 and 25%, respectively, of concentrations at 2
hours, the brain retained 43% of that measured at 2 hours. The major metabolite was
identified as resveratrol-glucuronide, reaching 7 pm in plasma at 2 hours. However, at 18
hours the main form identified in liver, heart, lung and brain was native resveratrol
aglycone. Unlike flavonoids, resveratrol does not appear to be metabolized by colonic
microflora as no phenolic degradation products were detected in urine or tissues. The
results of this study indicate that resveratrol and not its metabolite might be responsible
for its in vivo biological effects.
Hebbar et a1 (2005) investigated the effects of resveratrol on stress-related genes
and drug detoxifying enzymes by using cDNA array analysis and Quantitative Real-Time
PCR. Male and female CD-rats were treated (gavage) with resveratrol (300, 1000 and
3000 mg/kg/day) for 28 days. The gene expression profiles of Phase I drug metabolizing
enzymes changed only slightly from control rats among the low and intermediate dose
groups. The induction in these dose groups did not change more than 2-fold from control
rats. Investigators used a cut-off of 1.5 X fold of control in at least one treatment dose
level, to explain gene expression changes in the array data. The investigators concluded
that resveratrol, especially at higher doses moderately induced Phase I1 enzyme activities
and inhibited Phase I enzyme activities in the rat liver. As evidenced from the cDNA
array data, resveratrol modulated the expression of certain Phase I drug metabolizing as
well as antioxidant genes. The clinical significance of the changes noted in a variety of
genes is difficult to interpret. In addition to gene array, the investigators also studied
various safety related parameters. Evaluation of toxicity parameters revealed that
administration of resveratrol at 300 mgikg/day for 28 days did not cause adverse effects.
The safety related results from this study are described below in short-term and
L subchronic studies (see section 2.3).
As cited in the NIEHS (2002) report, based on an abstract publication, oral
administration of cis- and trans-resveratrol (1 mg/kg/day for five or ten days) to CD2Flmice
caused inhibition of 7-ethoxyresorufin-o-dealkylation (EROD) activity (CYPlA2).
However, resveratrol did not affect ethoxycoumarin-odeethylation (ECOD) activity
(CYPlA2/2El) or benzo[a]pyrene metabolism. The details of this publication were not
available. In an in vitro study, also cited in NIEHS (2002) report, incubation of
resveratrol with human microsomes inhibited CYPlA2 (methoxyresorufin 0-
demethylation) and CYP3A4 (erythromycin demethylation) without affecting CYP2El
activity.
By inhibiting the expression and activity of certain cytochrome P450 enzymes,
resveratrol may prevent activation of some carcinogens. In contrast, increasing the
activity of phase I1 biotransformation enzymes generally promotes the excretion of
potentially toxic or carcinogenic chemicals. Although such interactions have not been
reported in humans, high intakes of resveratrol could theoretically increase the
bioavailability and the risk of toxicity of drugs that undergo extensive first-pass
metabolism by cytochrome P450. Available in vivo and in vitro studies indicate that
resveratrol-mediated changes in drug metabolizing enzymes occur at doses greater than 1
1 mg/kg/day. The intended use of resveratrol at the low doses in water (bottled) is
unlikely to alter the drug metabolizing enzymes. The available studies show that
resveratrol is rapidly absorbed from the intestine and eliminated via kidney. As discussed
in human, because of its rapid metabolism and elimination, bioavailability of resveratrol
is low.
Acute studies
In a brief report, Juan et a1 (2002) investigated the acute toxic effects of transresveratrol
in rats (apparently Sprague Dawley). The study was conducted according to
the Organization for Economic Cooperation and Development (OECD) guidelines at a
dose level of 2000 mg/kg. The investigators concluded that the absence of symptoms, the
lack of any negative effects on growth and the normal appearance of vital organs in the
gross necropsy suggest that trans-resveratrol was practically non-toxic even under these
exaggerated exposure conditions.
In unpublished studies cited by Boocock et al. (2006), reported as part of Phase I
human trials of trans-resveratrol, the toxicity of pure trans-resveratrol was studied in rats
and mice. In rats, administration of a single intravenous dose of resveratrol (80 mg/kg;
highest dose) failed to elicit any manifestation of toxicity, as reflected by lack of any
effects on weight loss or histological alterations investigated at a range of time points
post-dosing. Similarly, repeated oral dosing of resveratrol(250 mg/kg daily for 5 days) in
mice was without adverse effect.
In summary, results from an acute toxicity study show that the LD5o (median
lethal dose) of resveratrol is likely to be higher than 2 g/day. The results of the acute
toxicity study indicate that trans-resveratrol is practically nontoxic. Single intravenous
administration of resveratrol (SO mgkg) to rats or oral repeat-dose administration (250
mgkg/day for 5 days) to mice did not reveal any adverse effects.
Short term and suhchronic studies
Juan et a1 (2002) investigated the effects of repeated oral administration of transresveratrol
in male Sprague-Dawley rats. trans-Resveratrol dissolved in
carboxymethylcellulose was administered orally to male Sprague-Dawley rats at a dose
of 20 mg/kg/day for 28 days. Animals in the control group received
carboxymethylcellulose. Compared to the control group, resveratrol administration did
not affect body weight, food or water consumption. The results of the hematologic tests
did not differ between control and treatment groups. trans-Resveratrol treatment did not
affect serum lipids, enzymes, electrolytes or other metabolites except alanine
aminotransferase (ALT), which was 30% higher compared to the control group. trans-
Resveratrol treatment did not affect the final relative weights of lungs, spleen, heart, liver,
kidney or adrenal gland. However, relative brain weight was greater in the treatment
group compared to control rats. Histopathology examination of the organs did not reveal
any alterations between the groups. The increase in ALT noted in this study was not
associated with histological changes.
Crowell et al. (2004) investigated the potential adverse effects of resveratrol in
rats. Male and female CD-rats (20/sex/group) were dosed once daily with resveratrol (0,
L.- 300, 1000, or 3000 mg/kg/day) for 28 days via gavage. The majority of the adverse
effects were noted in the rats receiving 3000 mg/kg/day of resveratrol. Adverse effects
included decreased final body weights and food intake; increased clinical signs of
toxicity; increases in plasma levels of BUN, creatinine, alkaline phosphatase, ALT, total
bilirubin, and albumin; decreases in hemoglobin, hematocrit, and red cell counts; and
increased white cell counts. Additionally, at this dose level (3000 mg/kg/day) increased
kidney weights, gross renal pathology changes, and an increased incidence and severity
of histopathological changes in the kidneys were noted. Although clinical chemistry
changes suggest liver toxicity, this was not supported by histopathology. In rats treated
with 1000 mg/kg/day of resveratrol, reduced body weight gain (females only) and
elevated white blood cell count (males only) were noted. The investigators determined
the no-observed-adverse-effect level (NOAEL) as 300 mg/kg/day in rats. In a review
article of toxicological studies of resveratrol prepared for the NIEHS (2002), this article
was considered to determine the NOAEL for resveratrol (300 mg/kg/day). The study
referenced above was conducted in compliance with Good Laboratory Practices and
meets FDA core standards for safety studies of food additives.
As mentioned earlier, Hebbar et al. (2005) studied the dose-response effects of
resveratrol in rats. Male and female CD-rats were gavaged with resveratrol (300, 1000
and 3000 mg/kg/day) for 28 days. At termination, clinical chemistry parameters such as
ALT, alkaline phosphatates, total bilirubin, blood urea nitrogen, creatinine were not
affected at low (300 mg/kg/day) and intermediate (1000 mg/kg/day) doses of resveratrol.
At the high dose level (3000 mg/kg/day), nephrotoxicity was observed in the male and
female rats and dehydration and labored breathing were seen among the female rats at the
intermediate dose level (1000 mg/kg/day). At the highest dose level, slight but significant
increases in ALT and alkaline phosphatatse in male and female rats were noted. Increases
in total bilirubin, blood urea nitrogen and creatinine were noted only in female rats at the
highest dose of resveratrol. Based on the results of this study, the NOAEL of resveratrol
is determined as 300 mg/kg/day in rats. The results from these investigations support the
observation from the Crowell et a1 (2004) study.
In a four week study, C57BL/6-mice (lO/sex/group) were treated daily via gavage
with resveratrol at doses of 0 (vehicle control), 1000, 2000, 3000, 4000, or SOOO
mg/kg/day for 28 consecutive days (Horn et al., 2007). Animals were observed twice
daily for morbidity and mortality, and weighed weekly. On study day 29, blood was
drawn for clinical pathology evaluations and a limited gross necropsy was performed
with collection of gross lesions data. In male mice receiving the highest dose of
resveratrol (SOOO mg/kg /day), a 40% mortality was noted. Necropsy in all early deaths
observed in the high-dose group revealed a large amount of unabsorbed resveratrol in the
stomach andor intestine. Except for the mortality associated with the impaction of
resveratrol in the gastrointestinal tract, no clear evidence of resveratrol toxicity was
identified in any dose group. The study was designed as part of an oncogenicity study
(Horn et al., 2007). In other NCI-sponsored preclinical toxicity studies of resveratrol
(cited in Boocock et al, 2006), doses of up to 1000 mg/kg/day in rats treated for 28 days,
or doses up to 2000 mg/kg/day in dogs treated for 14 days, did not reveal any toxic
effects.
In a subchronic study, male Sprague-Dawley rats were administered 20
mg/kg/day of resveratrol via gavage for 90-days (Juan et al., 2005). The investigators
reported that the selected dosage is not harmful and corresponds to 1000 times the
amount of resveratrol consumed by a 70-kg person who drinks 250 mL of red wine a day
containing 1.4 mg of trans-resveratrol. The objective of this study was to investigate the
effect of pans-resveratrol on the testis and on spermatogenesis (See Section 2.5.
Reproductive and Developmental Studies). Although the study was designed to
investigate reproductive toxicity of resveratrol, it does provide additional information on
the safety of the resveratrol. During the course of study, rats were examined daily for any
changes in skin, eyes, mucous membranes, respiratory system, autonomic and central
nervous system conditions, somatomotor pattern, and behavior. Body weight and food
and water consumption were recorded daily. The growth rate was calculated as the
difference between the final weight and the initial weight divided by 90 day. The food
and water intake and body weight gain of the resveratrol-treated group did not differ from
that of the controls. Resveratrol administration did not adversely affect any of the
parameters studied. As discussed later in Section 2.5, resveratrol did not affect testicular
gross anatomy, wet weight or relative weight.
In summary, available short-term studies of resveratrol indicate that daily
administration of resveratrol at doses up to 300 mg/kg/day for 28 days to rats did not
reveal any adverse effects. In one study (Juan et al. 2002), resveratrol administration at a
dose of 20 mgikg/day for 28 days revealed increased serum ALT levels without any
histopathological changes. Contrary to the increased serum ALT levels noted by Juan et
a1 (2002), results from two separate subsequent well-designed studies by Crowell et al.
(2004) and Hebbar et al. (2005) did not observe any increase in ALT levels at dose levels
of 300 mg/kg/day (1 5-fold higher). In both these subsequent studies, elevation of ALT
levels was noted at a very high dose of 3000 mg/kg/day, but without any
histopathological alteration. Additionally, in an NCI sponsored study, resveratrol did not
reveal any adverse effects at dose level of 1000 mg/kg/day. In a study in mice, resveratrol
at doses up to 4000 mgkg/day did not reveal any adverse effects. Based on the results of
two separate studies in rats, the NOAEL of resveratrol is determined as 300 mgkg/day.
Chronic toxicity and carcinogenicity studies
In an extensive and well designed study, Baur et al. (2006) investigated the longterm
effects of trans-resveratrol on the health and survival of mice. Cohorts of one year
old male C57BL/6NIA-mice were maintained on a standard diet or an otherwise
equivalent high-calorie diet alone (60% calories from fat) or high caloric diet with
resveratrol that provided an average 5.2 or 22.4 mg resveratrolkg body weightlday for
the remainder of their life. At the end of six months, a clear trend towards increased
survival and insulin sensitivity was noted and hence the investigators published an
interim report of the study at the end of one year. Several parameters, including safetyrelated
evaluations such as clinical pathology and histopathology were investigated. The
published report contained results from three groups of mice that received one of the
following regimens: standard diet (SD); high calorie diet (HCD) and; high calorie diet
plus resveratrol (HCDR). The HCDR diet provided an average of 22.4 mg resveratrolkg
L body weighuday. No significant effects of resveratrol on body weight, body temperature,
food consumption, total fecal output or lipid content, or post-mortem body fat distribution
were noted compared to animals receiving HCD. At 60 weeks of age, the survival in
HCDR mice was 3-4 months longer than SD or HCD mice. At the end of 114 weeks,
58% of the HCD-control mice died as compared to 42% HCDR and 42% of the SDcontrols.
Mice in HCDR-group steadily improved their motor skills compared to HCD
group.
Administration of resveratrol for one year at a dose of 22.4 mg/kg/day did not
affect serum clinical chemistry parameters including safety related indices such as ALT,
AST, creatinine phosphokinase, alkaline dehydrogenase, bilirubin, albumin, and
creatinine. The HCDR mice group had significantly lower levels of insulin, glucose, and
IGF-1 compared to mice in HCD-group. The levels of these markers in the HCDR-group
paralleled the SD-group. Mice receiving resveratrol had higher levels of insulin
sensitivity and low insulin resistance. Plasma amylase was elevated in the HCD-group
and was significantly reduced by resveratrol feeding. At 18 months of age, resveratrol
prevented an increase in size and weight of livers of mice in HCDR-group compared to
HCD-group. Histological examinations of the liver revealed a loss of cellular integrity
and the accumulation of large lipid droplets in the liver of the HCD but not the HCDRgroup.
The livers of mice receiving resveratrol had more mitochondria compared to mice
receiving HCD alone.
Recently, Horn et al. (2007) investigated the carcinogenic potential of resveratrol
in p53 knockout mice. This model is accepted by the FDA as an alternative model for
oncogenicity bioassay. TSG-p53(%) (heterozygous p53 knockout) mice (25/sex/group)
received daily via gavage either vehicle only (negative control), resveratrol doses of 1000,
2000, or 4000 mg/kg/day, or p-cresidine (400 mg/kg/day; positive-control) for six months
(180 days). Because of high mortality in the high-dose group, the 4000 mgkgiday dose
was reduced to 3000 mg/kg/day at the end of study week 4. Animals were observed twice
daily for mortality or evidence of toxicity. Body weights, food consumption, and detailed
clinical observations were performed once per week. Blood samples were collected
immediately prior to the terminal necropsy for clinical evaluations. At termination,
weights of the adrenals, brain, heart, kidneys, liver, spleen, testedovaries, and thymus
were collected. Histopathology evaluations on approximately 45 tissues per animal were
performed in control, low and mid-dose groups (Horn et al., 2007).
No mortality was noted at 1000 mgkglday. Administration of resveratrol at 2000
and 4000/3000 mg/kg/day induced mortality in both sexes of p53(i) mice. At the
termination of the study (26 weeks), survival in male mice was 64 and 28% in the middle
and high dose groups, respectively. In females the survival at mid- and high-dose was 60
and 24%, respectively. The early mortality in groups receiving the mid- and high-doses
of resveratrol appears to be related to the mass of resveratrol administered, rather than a
result of any agent-specific toxicity. Dose- and time-related mortality patterns were
noted in both male and female mice. Clinical observations, body weight measurements,
and quantitation of food intake failed to identify any evidence of toxicity in animals
exposed to resveratrol. Blood analysis for clinical pathology did not reveal any adverse
effects at the lowest dose (1000 mg/kg/day). Modest (<40%) increases in serum
cholesterol were seen in male mice at the highest dose and in female mice at the mid- and
high dose. At the low dose, resveratrol administration did not affect serum cholesterol
levels. The investigators did not consider these changes of any toxicological significance
Small (<15%) but significant reductions in red blood cell count, hematocrit, and
hemoglobin were seen only in male mice receiving the high dose (4000/3000 mg/kg/day)
of resveratrol (Horn et al., 2007). These changes were not noted in male mice receiving
lower doses or in female mice in any dose group. Occasional significant differences from
control were noted in several other clinical pathology parameters in mice exposed to
resveratrol. However, these changes were small (
With the exception of a dose-related increase in absolute and relative liver
weights (without any microscopic changes) in both sexes of mice, resveratrol
administration did not affect organ weight. At the lowest dose of resveratrol, the effects
on liver weight were not significant. Histopathology evaluation of 45 tissues did not
reveal any evidence of oncogenicity of resveratrol. In the positive-control group, pcresidine
induced urinary bladder neoplasms in both sexes of mice. Resveratrol
treatment resulted in a dose-related hydronephrosis and epithelial cell hyperplasia of the
urinary bladder. Neither hydronephrosis nor urothelial hyperplasia were linked to any
functional alteration. Hydronephrosis was noted at 1000 and 2000 mglkglday dose, while
hyperplasia was noted only at 2000 mg/kg/day. Hydronephrosis was noted in 20% lowdose
male, 12% low-dose female, 18% mid-dose male and 60% mid-dose female.
Epithelial hyperplasia of the urinary bladder was noted in 44% surviving males and 33%
surviving females in the group exposed to mid-dose of resveratrol. The results of this
study demonstrate that chronic oral administration of resveratrol at its maximum tolerated
dose (3000 mg/kg/day) does not increase the incidence of any malignant or benign
neoplasm in p53 knockout mice (Horn et al., 2007). The results of this study also
demonstrate that except for a hydronephrosis in few animals without any functional
alterations, resveratrol administration at a dose of 1000 mg/kg/day for 180 days did not
cause any toxicity.
In summary, the results from a well-designed chronic toxicity study in mice
demonstrate that administration of resveratrol (22.4 mg/kg/day) had no adverse effects on
overall health and lifespan as determined by several indicators including survival, motor
function, insulin sensitivity, organ pathology, and mitochondrial number. Resveratrol
administration for one year did not affect serum clinical chemistry parameters such as
ALT, AST, creatinine phosphokinase, alkaline dehydrogenase, bilirubin, albumin, and
creatinine. Based on the results of the chronic study, a NOAEL for resveratrol of 22.4
mg/kg/day can be determined. The carcinogenicity study in p53 knockout mice shows
that resveratrol does not possess carcinogenic potential. Hydronephrosis (at 1000 and
2000 mg/kg/day) and epithelial cell hyperplasia (at 2000 mg/kg/day) of the urinary
bladder noted in the carcinogenicity study appear to be related to the elimination of high
doses of resveratrol. Based on the results of this study, a NOAEL for resveratrol of 1000
mg/kg/day can be determined. FDA has accepted the p53 mouse model as an alternative
model for an oncogenicity bioassay. Based on the information provided in chronic and
carcinogenecity studies, the data quality and study design for both these studies is very
good and both studies appear to have followed Good Laboratory Practices.
Reproduction and developmental studies
Nikaido et a1 (2005) investigated the effect of prepubertal exposure to resveratrol
on development of the reproductive tract and mammary glands in female mice.
Beginning at 15 days of age, female CD-I mice were administered four daily
subcutaneous injections of 10 mg/kg/day of resveratrol. Vaginal opening was checked
and estrous cyclicity was monitored from 5, 9 or 21 weeks of age for 21 consecutive days
At 4, 8 and 24 weeks of age, animals were necropsied. Prepubertal exposure to
resveratrol did not affect body weight gain, puberty onset (vaginal opening) or the estrous
cycle length. Necropsy at 4, 8 and 24 weeks of age did not reveal any significant adverse
effects on corpora lutea. Resveratrol did not alter the uterine or vaginal morphology or
mammary gland growth. The results of this study indicate that resveratrol does not
exhibit estrogenic biological activity. In a similar previously reported study by this group,
resveratrol at same dose (10 mg/kg/day) caused transient effects (increased length of
estrus cycle by prolonging diestrus during week 9-1 1 on the reproductive tract (Nikaido
et al., 2004). However, as described above, these effects were not confirmed in the more
recent study by the same group.
In a multigenerational mouse study, Kyselova et a1 (2003) investigated the
effects of resveratrol on the body weight, organ weight, histology of testes and ovaries,
acrosomal integrity of the spermatozoa, and litter size. Adult CD-1 outbred mice twomonths
old (parental or P-generation) were exposed to resveratrol (3 mg/L) in drinking
water for four weeks, and then mated. Based on an average daily intake of water and an
average body weight of mice, resveratrol intake was determined as 0.75 mg/kg/day. The
FI generation was exposed to resveratrol during gestation, lactation, the prepubertal
period and the pubertal period, up to adulthood. Exposure to resveratrol did not affect
body weight in either males or females in the P or F1 generation. Resveratrol decreased
the absolute (20%) and relative seminal vesicle (12%) weight and spleen weight (18%) as
compared to controls in the P generation and the absolute vesicle (18%) and kidney
(10%) weight in the F1 generation compared to control. Although, resveratrol
administration resulted in decreased absolute and relative seminal vesicle weight and
spleen weight in the P generation, similar decreases, in absolute and relative seminal
vesicle weight, and spleen weight were not noted in FI generation. The absolute and
relative ovary weight was decreased in the P-generation females treated with resveratrol.
However, such a decrease was not noted in FI generation. The differential changes noted
in the P and FI generations indicate that the decreases noted in the P generation may not
be related to the treatment. It is anticipated that changes noted in the P generation should
also be present in FI generation. Resveratrol did not affect litter size or sex ratio
(female/male). Resveratrol exposure had no affect on sperm number or sperm quality.
Histological examinations did not reveal any alteration of testes or ovarian morphology in
either generation. The investigators concluded that resveratrol had no effects on in vivo
fertility.
In a comparative study, Kubo et a1 (2003) investigated the effects of resveratrol,
bisophenol-A, diethylstilbestrol or vehicle (control) on the sexual differentiation of openfield
behavior and the sexually dimorphic nuclei in the brain in the offspring of rats
exposed to these compounds during the fetal and suckling periods. After copulation,
female Wistar rats received water containing trans-resveratrol (5 mg/L) until their pups
were weaned on postnatal day 21. The dose of resveratrol for dams was determined as 1.5
mg/kg/day. Resveratrol exposure did not affect male sexual development, whereas
resveratrol delayed the day of vaginal opening in females (their body weights on the day
of vaginal opening increased in comparison to the control females). However, in a mouse
study described above, Nikaido et al. (2005) reported that resveratrol did not affect
vaginal opening. In the open-field test, compared to control group, resveratrol exposure
did not show any effects. In male rats, resveratrol did not affect the number of mounts or
intromission. The intromission rate, however, was decreased. The investigators
concluded that resveratrol did not change sexual behavior in male rats. In females,
resveratrol did not affect the number of ear wiggles, but decreased the lordosis7 quotient
and increased the rejection score suggesting reduced receptivity.
Resveratrol did not affect the weight of epidydymis, ventral prostate, seminal
vesicle or qualitylquantity of sperms in males as determined by the total spermatid
present in the left testis, the caudal sperm, or the percentage of motile sperms. In either
male or female offspring, resveratrol exposure had no significant effects on serum
hormone levels of luteinizing hormone, follicle-stimulating hormone, prolactin,
testosterone or 17P-estradiol. In female offspring, resveratrol did not affect weight of the
uterus or bilateral ovaries. Resveratrol exposure resulted in reduced number of estrus
cycles. Resveratrol treatment had no effect on the sex differences in the brain weight. The
volume of female locus coeruleus' was significantly larger than that of males in the
respective control groups, and resveratrol administration abolished this difference.
Comparatively the estrogenicity of resveratrol was less than bisophenol-A or
diethylstilbestrol. Interestingly, the delayed vaginal opening in female offspring noted in
this study appears to contradict the claimed estrogenic activity, as estrogens are known to
accelerate the vaginal opening (Kubo et al., 2003).
Juan et al. (2005) investigated the effect of trans-resveratrol on testis and
spermatogenesis. Male Sprague-Dawley rats were administered 20 mglkglday of
resveratrol via gavage for 90-days. A period of 90-days was chosen to cover the complete
spermatogenesis cycle of rats. The food intake and body weight gain of the resveratroltreated
group did not differ from that of the controls. Resveratrol did not affect testicular
gross anatomy, wet weight or relative weight. Histological examinations of testis did not
reveal microscopic lesions such as cytoarchitectural alterations or disorganization of the
tubular elements. Resveratrol treatment reduced the seminiferous tubules diameter and
increased the tubular density. Moreover, sperm counts were significantly greater in the
resveratrol-treated group compared to control, but sperm quality did not differ. Serum
levels of gonadotrophins and testosterone were higher in the resveratrol-treated group.
The investigators concluded that resveratrol enhanced sperm production by stimulating
the hypothalamic-pituitary-gonadal axis, without inducing adverse effects. These
investigators also suggested that resveratrol might be useful in the treatment of male
infertility.
Henry and Witt (2006) investigated the effects of maternal resveratrol exposure to
male and female offspring. In this study, reproductive physiology, behavior and brain
morphology were examined in adult offspring of dams exposed to resveratrol throughout
the lactation period. Time-mated female Sprague Dawley rats following both delivery
and during the entire lactation period received water (control), 10% ethanol (vehicle), or
resveratrol (5, 50 or 100 pM in 10% ethanol) as a daily drinking water solution. At
weaning, rats were housed in same sex groups and were provided water and food ad
libitum. The mean daily consumption of drinking solution was similar (-70 ml) in all
groups and ranged from 61-79 ml. The resulting daily resveratrol intake was
approximately 0.08, 0.8 and 1.6 mg/day, respectively. During adulthood, female
offspring exposed to resveratrol throughout nursing exhibited reduced body weight and
increased ovarian weight. However, these offspring exhibited normal estrous cyclicity
and sociosexual behavior, without changes in the volume of the sexually dimorphic
nucleus of the preoptic area or the anteroventral periventricular nucleus of the
hypothalamus. Adult males exposed to resveratrol during nursing exhibited decreased
body weight and plasma testosterone concentration, increased testicular weight, and
reduced sociosexual behavior. These males also had significantly smaller sexually
dimorphic nucleus of the preoptic area volumes and larger anteroventral periventricular
nucleus volumes compared to male controls. The investigators suggested that postnatal
exposure to resveratrol may affect estrogenic activity in specific peripheral tissues (e.g.,
the gonads), while inducing antiestrogenic effects in the brain.
In a previous study, Henry and Witt (2002) reported that resveratrol acts as a
possible agonisthtagonist depending on the availability of specific estrogen receptor
isoforms localized in the reproductive tract and brain of the female rats. In this previous
study, exposure of resveratrol (100 pM in 10% ethanol) in drinking water to female
Sprague-Dawley rats for seven days decreased body weight, disrupted estrous cyclicity,
and induced ovarian hypertrophy. In ovariectomized females, subcutaneous resveratrol
(0.01-1 mg) injection did not appear to mimic 17P-estradiol benzoate induced behavioral
responses and had no subsequent estrogen sensitivity or sociosexual behavior
The results of the Henry and Witt (2002; 2006) studies are difficult to interpret
because of confounding factors. Although the investigators used an ethanol vehicle group,
it is not clear how high levels of ethanol (10%; 10,000 mgkg/day) in combination with
resveratrol may interact. Studies have shown that heavy alcohol consumption results in
reduced testosterone levels in the blood. Alcohol is also known to impair the function of
the testicular Sertoli cells that play an important role in sperm maturation (Emanuele and
Emanuele, 1998). Ethanol used in Henry and Witt (2002; 2006) study is likely to cause
several unknown effects that were not investigated in the study. To put this into
perspective, the ethanol used in this study is equivalent to a human weighing 70 kg
drinking 70 ml of ethanol daily. A typical alcoholic beverage contains 12 g of alcohol,
corresponds to a dose of about -12 ml/day for a 70-kg adult, and produces a peak blood
ethanol concentration of approximately 25 mg/dl.
In summary, prepubertal exposure to resveratrol did not affect development of the
reproductive tract and mammary glands in female mice. In a multigenerational mouse
study, resveratrol treatment resulted in decreased absolute and relative seminal vesicle
weight and spleen weight in the P generation and the absolute vesicle and kidney weight
in the FI generation. Some of the changes noted in the P generation were not found in the
FI generation. For example, resveratrol administration resulted in decreased absolute and
relative seminal vesicle weight and spleen weight in the P generation, similar decreases,
in absolute and relative seminal vesicle weight, and spleen weight were not noted in FI
generation. The results of this multigeneration study show that resveratrol has no effect
on in vivo fertility. In a comparative study of different endocrine disruptors, resveratrol
exposure did not affect male sexual development, but delayed the day of vaginal opening
in females with an increase observed in body weight on the day of vaginal opening.
However, in another study, resveratrol did not affect vaginal opening. As estrogens are
known to accelerate vaginal opening, the delayed vaginal opening noted in one of the
study contradicts the claimed estrogenic activity. In a 90-day study, gavage
administration of resveratrol at a dose of 20 mgikdday enhanced sperm production by
stimulating the hypothalamic-pituita-gonadal axis, without inducing adverse effects.
Maternal resveratrol exposure in drinking water with 10% ethanol to male and female
offspring, indicate that postnatal exposure to resveratrol along with ethanol may affect
estrogenic activity in specific peripheral tissues (e.g., the gonads), while inducing
antiestrogenic effects in the brain. The results of this study are difficult to interpret as
very high levels of ethanol used in this study could compound any specific effects
observed.
Estrogenic activity
The structural similarity of resveratrol to that of the synthetic estrogen agonist,
diethylstilbestrol, suggest that resveratrol might also function as an estrogen agonist. In
vitro experiments in cell culture indicate that resveratrol acts as an estrogen agonist under
some conditions, and an estrogen antagonist under other conditions. Initial concerns
regarding resveratrol's effect on breast cancer arose from a paper by Gehm et al (1997)
describing resveratrol's estrogenic actions. These investigators found that resveratrol
bound specifically to estrogen receptors and stimulated proliferation of a breast cancer
cell line (T47D). In swift succession, several papers found the exact opposite for
resveratrol: inhibition of several breast cancer cell lines with and without estrogen
receptors, including the T47D breast cancer cell line (Mgbonyebi et al., 1998; Hsieh et al ,
1999; Lu and Serrero, 1999; Damianaki et al., 2000).
Additional in vitro studies with mammary cancer cell lines also indicated that
resveratrol acts as a mixed estrogen agonist/antagonist. However in the presence of 17pestradiol,
it was an antiestrogen (Bhat et al., 2001; Gehm et al., 1997). For example, in
MCF-7 and S30 cells, resveratrol alone showed weak estrogenic response, but when
combined with estradiol (1 nM), a dose-dependent antagonism occurred. In addition,
progesterone receptor (PR) protein expression was induced with resveratrol alone, but
when combined with estradiol, the expression was suppressed. In T47-D and LY2 cells,
resveratrol was a pure estrogen antagonist and it significantly down regulated steady state
and estradiol-induced PR protein levels. With LY2 and S30 cells, presnelin 2-protein
expression was down-regulated (Bhat et al., 2001).
Nakagawa et al. (2001) reported that exposure to resveratrol at pharmacological
doses (52-74 pM [12-17 pg/mL]) suppressed the growth of ER-positive breast cancer
cells (KPL-1 and MCF-7) and ER-negative breast cancer cells (MKL-F) stimulated by
linoleic acid, a potent stimulator of these cells. High fat diet, particularly linoleic acid has
been shown to play a key role in breast cancer stimulation. Resveratrol(1 pM-1 pM [2.28
x lO-'-O.2 pg/mL]) was also an agonist of steroid receptors. In the MCF-7 cells,
resveratrol at the nanomolar range interacted with estradiol simultaneously with PRs (at
the picomolar range). In T47-D hormone-sensitive breast cancer cell line, the same
interactions were seen but to a lesser extent. In MDA-MB-23 1 hormone-independent
breast cancer cell line, no steroid binding was noted observed (Damianaki et al., 2000).
Both trans- and cis-resveratrol (10 and 25 pM [2.3 and 5.7 pg/mL]) significantly
increased the growth of MCF-7 cells. Cell growth was decreased at a high dose of 50 pM
(1 1 pg/mL), and this concentration was found to be cytotoxic. In the presence of estradiol,
and at 25 and 50 pM trans-resveratrol, and 50 pM cis-resveratrol, significant reductions
in cell proliferation were observed. In MVLN cells, trans-resveratrol (10 and 25 pM) and
cis-resveratrol (25 pM) significantly increased luciferase activity compared to estradiol.
In the presence of estradiol, both isomers at the same doses acted as superagonists of
estradiol. In both cell lines, cis-resveratrol was less effective than trans-resveratrol (Basly
et al., 2000). Resveratrol was observed to exhibit estradiol antagonist activity for ER-a
with select estrogen response elements and no such activity was noted with ER-P
(Bowers el al., 2000). In human endometrial adenocarcinoma (Ishikawa) cells at
concentrations of 10 pM (2.3 pg/mL), resveratrol mediated antiestrogenic effects by
selective down-regulation of ER-a but not ER-P (Bhat and Pezzuto, 2001).
Contrary to some of the above described in vitro studies, an in vivo study with
oral administration of resveratrol (1, 4, 10, 40, and 100 pg/day for six days) to weanling
rats had no effect on estrogen target tissues (bone formation and mineralization rates)
versus the estrogen 17p-estradiol. Resveratrol administration had no significant effect on
body weight, serum cholesterol, radial bone growth, epithelial cell height, or messenger
RNA levels for insulin-like growth factor I. Resveratrol treatment resulted in slight
increases in uterine wet weight, but significance was only achieved at the 10 pg/day dose.
These observations suggest that resveratrol was not an agonist at the estrogen receptor.
Simultaneous administration of resveratrol (1 000 pg) and 17p-estradiol (1 00 pg) showed
a synergistic effect (i.e., a significant decrease in cholesterol levels was seen in the
animals). The inability of low doses (1 and 10 pg, respectively) of resveratrol to lower
serum cholesterol levels suggested antagonism by resveratrol at the estrogen receptor
(Turner et al., 1999). In other studies in rats, oral or subcutaneous administration of
trans-resveratrol (0.03-575 mgkg) had no estrogenic response in uterine tissue (Ashby et
al., 1999; Freyberger et al., 2000). Observations from in vivo studies raise questions
regarding the extent to which estrogenicity data from in vitro assays using MCF-7 or
other cell lines can be used to predict the hormonal effects likely to occur in animals or
humans.
In summary, available in vitro data indicate that resveratrol has complex
estrogenic and antiestrogenic effects depending on the biological environment.
Resveratrol appears to act consistently as an antiestrogen to breast tumors under
physiological conditions. These data suggest that resveratrol may have beneficial effects
if used as a chemopreventive agent for breast cancer. Results from an in vivo study
demonstrate that resveratrol does not stimulate indices of uterine growth and
differentiation in immature rats. Even very high doses of resveratrol generally had
insignificant effects on uterine wet weight, epithelial cell height, and IGF-I gene
expression. Additionally, resveratrol had no effect on other estrogen target tissues.
Resveratrol treatment did not alter cortical bone growth, serum cholesterol concentration ,
or body weight. The available evidence shows that resveratrol is not an estrogen agonist.
Cytotoxicity
In an in vitro study, Brakenhielm et al. (2001) investigated dose-related effects of
resveratrol in bovine capillary endothelial (BCE) and porcine aortic cell lines. Resveratrol
dose-dependently (1 -10,000 nm) inhibited the cell growth of capillary endothelial cells
stimulated with fibroblast growth factor-2 (FGF-2). Resveratrol also inhibited the
phosphorylation of mitogen-activated protein kinases (MAPKs) (10 and 20 pM), and
FGF-2 and vascular endothelial growth factor (VEGF)-induced proliferation of porcine
aortic cell lines expressing PAEEGFR-1 and PAENEGFR-2, respectively, in a dosedependent
manner (0.5-10 pM).
Babich et al (2000) investigated sensitivity of human gingival epithelial S-G cells
to resveratrol using the neutral red dye uptake assay. The following sequence of
sensitivity to resveratrol (doses up to 500 pM) was determined: tongue squamous
carcinoma SCC-25 cells > Smulow-Glickman (S-G) human gingival epithelial cells >
RHEK-1 keratinocytes >> fibroblasts. In a 3-day continuous exposure to resveratrol
assay (5-150 pM) with S-G cells, toxicity was found to level off between day 2 and 3. At
concentrations >75 pM, irreversible damage to cell proliferation was noted, and the
presence of hepatic S9 microsomal fraction did not affect the cytotoxicity.
Genotoxicity
Matsuoka et al. (2001) investigated the genotoxic potential of trans-resveratrol in
bacterial reverse mutation assay, in the in vitro chromosome aberration test, the in vifro
micronucleus test, and in the sister chromatid exchange test (Table 6). Resveratrol was
negative in the bacterial reverse mutation assay (Salmonella typhimurium strains TA98
and TAlOO and E coli WP2uvrA) in the absence and presence of a microsomal
metabolizing system. Resveratrol induced structural chromosomal aberrations (mainly
chromatid breaks and exchanges) and showed weak aneuploidy induction in a Chinese
hamster lung cell line. In the in vitro micronucleus test, resveratrol induced mononuclear,
polynuclear and karyorrhectic cells at the end of 48 h treatments. In the sister chromatid
exchange test, resveratrol induced sister chromatid exchanges dose-dependently at dose
levels up to 10 pg/ml. Cell cycle analysis indicated that resveratrol caused S-phase arrest,
and 48 h treatment induced apoptosis.
In another study, Fukuhara and Miyata (1998) reported that trans-resveratrol
cleaved plasmid DNA in the presence of Cu2+at neutral pH and under aerobic conditions.
Under anaerobic conditions, however, increasing the concentration of resveratrol failed to
enhance the efficiency of DNA cleavage, suggesting the cleavage to be dependent on the
presence of both Cu2' and oxygen. Ahmad et al. (2005) also found resveratrol induced
DNA changes in plasmid DNA in the presence of elevated levels of copper ions.
The positive genotoxicity findings are understood to be an artifact of the test
systems as physiological concentrations of ascorbic acid or glutathione, normally present,
are lacking in the test system. In these test s stems, resveratrol was found to promote
hydroxyl-radical formation by DNA-bound Cu ions, thereby acting as a reducing agent.
Burkitt and Duncan (2000) have shown that in the presence of either ascorbic acid or
glutathione (ix., under more physiological conditions), resveratrol lost this property and
behaved as an antioxidant. In the ascorbate system, resveratrol had no effect on the rate
of hydroxyl radical formation, but protected DNA from damage by acting as a radical
scavenging antioxidant. Further, in the glutathione system, resveratrol inhibited hydroxyl
radical formation via a novel mechanism involving the inhibition of glutathione disulfide
formation. These investigators concluded that the DNA-damaging properties of
resveratrol will be of no significance under physiological conditions. Burkitt and Duncan
(2000) demonstrated that resveratrol behaves as a powerful antioxidant, both via classical,
hydroxyl-radical scavenging and via a novel, glutathione-sparing mechanism.
The mechanistic data of Burkitt and Duncan (2000) are consistent with majority of
L" the published data which demonstrate an antimutagenic effect of resveratrol in
combination with known mutagens and carcinogens, usually thought to be associated
with its antioxidant properties. For example, two studies found that resveratrol inhibited
the mutagenic effects of the carcinogens Trp-P-1 and MNNG on Salmonella ryphimurium
bacteria (Uenobe et a1 , 1997; Kim et al., 2002).
Studies in animal cell lines have also largely corroborated these findings.
Sgambato et al. (2001) found that resveratrol reduced DNA fragmentation in a variety of
cell lines (including rat fibroblast, mouse mammary cells, and human cell lines) that were
exposed to mutagenic agents such as tobacco-smoke and hydrogen peroxide. Topical
administration of resveratrol to mouse epidermis inhibited binding of a carcinogen,
DMBA, to DNA (Szaefer et al., 2004). Resveratrol also inhibited binding of the
carcinogen benzo[a]pyrene to DNA of human oral epithelial cells (Walle et al., 2006).
In summary, the genotoxicity studies of resveratrol clearly indicate that it is
generally antimutagenic and anticarcinogenic. Although, in in vitro assays, DNA
damaging effects of resveratrol have been reported, these effects are not physiologically
comparable to in vivo biochemistry and allow for Cu2' -induced hydroxyl radical
formation by resveratrol. Under physiological conditions that have ascorbate and
glutathione present, resveratrol does not induce DNA damage by Cu2+ but acts as a potent
antioxidant. Therefore, resveratrol is not considered to pose any genotoxic risk in vivo.
Angiogenesis
Some studies indicate that resveratrol inhibits angiogenesis. Because angiogenesis
plays an important role in physiologic process of wound healing and pathophysiologic
cancer growth processes, it is important to determine under what conditions resveratrol
inhibits angiogenesis. It is important to note that resveratrol potently inhibits
angiogenesis related to tumor growth. Garvin et a1 (2006) demonstrated significantly less
angiogenesis in human breast cancer grafts in mice. Similarly, other investigators found
an inhibitory effect of resveratrol on angiogenesis in Kaposi's Sarcoma (Tseng et al.,
2004), brain glioma (Baliestrieri et aL, 2006), ovarian cancer (Cao et al. 2004), and lung
cancer (Kimura ef ~1,2001).
Brakenhielm et a1 (2001) reported that oral administration of resveratrol
significantly inhibits the growth of a murine fibrosarcoma in mice, and delays
angiogenesis-dependent wound healing in mice. Resveratrol(5.7 pg/ml) was added to the
drinking water of mice with a 5 mm surgical wound 2 days before the operation and
throughout the experiment. Resveratrol significantly delayed wound healing in mice, as
determined by measuring the sizes of wounds and the percentage of animals with healed
wounds. In another experiment with corneal micropockets of the mice, oral
administration of resveratrol (0.4 pg/mL) given three days before growth factor
implantation and for 15 days after surgery significantly inhibited VEGF- and FGF-2-
induced corneal neovascularization compared to the control (Brakenhielm et a1 ,2001).
Experimental studies generally do not indicate an adverse effect of resveratrol.
Although resveratrol was found to inhibit corneal neorevascularization in mice (induced
by the vascular endothelial growth factor [VEGF] (Oak et a1 , 2005), other studies clarify
this by showing that resveratrol appears to act as a �brake� to unregulated angiogenesis
(in tumors), but does not inhibit normal angiogenesis. Wound healing, in particular,
requires neorevascularization and resveratrol appears to accelerate dermal wound healing
(Khanna et al,, 2002) by actually upregulating VEGF in skin cells (Khanna et a1 , 2001;
Sen et al, 2002). These observations contradict the above described findings from
Brakenhielm et al. (2001). Similarly, resveratrol upregulates VEGF in rat models of
myocardial infarction, leading to improved collateral formation (Fukuda et al., 2006;
Kaga et al., 2005). Thus, based on the above, resveratrol is not considered to pose any
risk of altering normal angiogenesis.
Platelet aggregation
The effects of resveratrol on platelet aggregation and cholesterol metabolism are
controversial. Some in vitro studies indicate that resveratrol may block platelet
aggregation (Kirk et al, 2002; Pace-Asciak et a1 , 1995). Kirk et al. (2002) reported that
resveratrol, at 10-50 pM, blocked aggregation of washed platelets induced by collagen (5
pg/ml), thrombin (0.2 units/ml), and ADP (10 pM). Compared to washed platelets, in
whole blood, resveratrol has poor antiplatelet activity and platelet aggregation was not
affected by 50-100 pM resveratrol. Concentrations of 200 pM resveratrol were required
to cause a decrease in platelet aggregation in whole blood.
Freedman et a1 (2002) suggested that quercetin or resveratrol are not the main
substances of wine responsible for the platelet inhibition or nitric oxide-releasing effects
(Violi et a l , 2002). Although, m vitro studies have shown that flavonoids including
quercetin, resveratrol, and catechin inhibit platelet aggregation (Pace-Asciak et al., 1995),
the physiological relevance of these findings has been questioned in humans because oral
supplementation with quercetin causes markedly increased plasma levels but does not
alter total, LDL, or HDL cholesterol levels or change thrombogenic markers including
platelet aggregation and platelet thromboxane B2 production (Conquer et al., 1998). The
inhibitory effects on platelet aggregation were observed in cells cultured in the presence
of resveratrol.
Observations in humans
In human studies, orally administered trans-resveratrol appears to be well-absorbed.
Walle et al. (2004) studied absorption, metabolism, and excretion of ��C-resveratrol in six
human volunteers following both oral and intravenous administration. Following an oral
dose of 25 mg of trans-resveratrol, the absorption was at least 70%, with peak plasma
levels ofresveratrol and metabolites of 491 ng/ml (peaked at 30-60 min) and a plasma
half-life of 9.2 h. Only trace amounts of unchanged resveratrol (<5 ng/ml) were detected
in plasma. The majority of the oral dose was recovered inurine as sulfate and glucuronic
acid conjugates of the phenolic groups of resveratrol.
Recently, the National Cancer Institute (NCI) initiated clinical toxicity studies on
trans-resveratrol (Abrams ef al. 2004; Boocock et al. 2006). In a single dose pilot trial,
four human participants were treated with 1 g of resveratrol and the samples assayed
using a high performance liquid chromatography assay with metabolite identification by
liquid chromatography-mass spectrometry. Data from this study indicate that at a C,,
(between 60 -120 min), the mean concentration of resveratrol was 103.2 ng/mL with
minimum and maximum concentrations of 58.8 and 208.0 ng/mL, respectively (Abrams
et al. 2004). In a subsequent clinical study protocol, Boocock et al. (2006) reported that
administration of a single oral dose of resveratrol at levels of 0.5, 1.0, 2.5, 5.0 g
demonstrated bioavailability, albeit not greatly, with a C, between 30 and 90 minutes
and peak plasma levels for the 1.0 g dose of -1 pM. No serious adverse events have
been recorded during this study.
Zamora-Ros ef al. (2006) conducted two randomized, cross-over trials and a
cohort study to determine whether urinary resveratrol metabolites could serve as a
biomarker of moderate wine consumption. In this study investigators used liquid
chromatography-tandem mass spectrometry to analyze urinary total resveratrol
metabolites In the first study, 10 healthy men consumed 30 g of ethanol/day as sparkling
wine or gin for 28 days. In the second trial, 10 healthy women consumed 20 g of
ethanol/day as white or red wine for 28 days. Urinary total resveratrol metabolites were
analyzed using liquid chromatography-tandem mass spectrometry. A significant increase
in total resveratrol metabolite (72.4, 211.5, and 560.5 nmol/g of creatinine) was noted
after consumption of sparkling, white, or red wine, respectively, but no changes after the
washout or gin periods. In the cohort study, the reported daily dose of wine consumption
correlated directly with total resveratrol metabolites. These investigators concluded that
resveratrol metabolites in urine may be useful biomarkers of wine intake.
In summary, available pharmacokinetic and safety studies in human subjects
indicate that orally administered ham-resveratrol is well absorbed by humans. Due to its
rapid metabolism and elimination, bioavailability of resveratrol is low. Resveratrol
determination in urine samples may serve as a marker of wine consumption. In a pilot
study, a single oral dose of resveratrol (1 g) was without any serious adverse effects. As
the majority of the basic research on resveratrol has been conducted in cultured cells
exposed to unmetabolized resveratrol at concentrations that are often 10-100 times
greater than peak concentrations observed in human plasma after oral consumption,
Gescher and Steward (2003) reported that bioavailability is important for its action. In a
review article, Wenzel and Somoza (2005) reported that the oral bioavailability of
resveratrol is almost zero due to rapid and extensive metabolism and the consequent
formation of various metabolites as resveratrol glucuronides and resveratrol sulfates.
SUMMARY AND DISCUSSION
Resveratrol, a polyphenol, is commonly found in grapes, red wine, purple grape
juice, peanuts and some berries. Various plants produce resveratrol to help defend against
invading fungi, stress, injury, infection, and too much sunlight. Resveratrol may exist in
both cis- and trans-stereoisomeric forms. In nature, both the forms also occur as the
glucoside (bound to a glucose molecule). The richest source of resveratrol is the roots of
the plant Polygonum cuspidatum (KO-jo-kon), mainly cultivated in China and Japan.
frans-Resveratrol (>98% pure) can be isolated from this plant by employing high-speed
counter-current chromatography. In recent years, resveratrol has been extensively studied
for its wide range of desirable biological effects such as cardioprotection,
chemoprevention, anti-cancer, and prolongation of life span in several species.
ATLA Holdings, LLC, intends to use standardized Pans-resveratrol (>99% pure),
extracted from P cuspidaturn, as a dietary nutrient at levels up to 10 ppm in bottled water
(�near waters�). trans-Resveratrol is isolated by solvent extraction and the purified
product contains >99% trans-resveratrol. The intended use levels in bottled water will
result in an estimated daily mean and 90th percentile exposure of 2.09 and 3.99
mg/person/day or 0.04 and 0.07 rngkdday. Moderate consumers of red wine containing
high levels of resveratrol are likely to ingest about 4 mg resveratrol/day or 0.067
mgkg/day. The estimated 90th percentile daily intake of resveratrol (3.99 mg/person) is
similar to that of consumers of red wine (in moderation) with high levels of resveratrol (4
mg/person). The dietary supplement recommended dosages of resveratrol have been
reported to range from 3 to 1000 mg/person/day.
Following oral ingestion in animals and humans, resveratrol is rapidly absorbed
from the gastrointestinal tract. Although resveratrol appears to be well-absorbed by
humans, its bioavailability is low because it is rapidly metabolized and eliminated. The
plasma half-life of frans-resveratrol following oral administration (25 mg) has been
determined as 9.2 hours. The pharmacokinetic studies indicate that resveratrol is rapidly
eliminated from the body and is unlikely to accumulate.
In an acute toxicity study, the LDso of resveratrol was found to be greater than 2
g/day indicating that trans-resveratrol is practically nontoxic. Single intravenous
administration of resveratrol (80 mgkg) to rats or oral repeat dose administration (250
mg/kg/day for 5 days) to mice did not reveal any adverse effects. In two separate studies,
daily administration of resveratrol at doses up to 300 mg/kg/day for 28 days did not
reveal any adverse effects. In one study, oral administration of resveratrol (20 mgkglday
for 28 days) revealed some adverse effects such as increased serum ALT levels without
liver histological changes. Contrary to these results, other studies did not reveal increases
in ALT levels at a dose level of 300 mg/kg/day (a 15-fold higher dose). Additionally, in
NCI sponsored study, resveratrol did not reveal any adverse effects at dose level of 1000
mgkglday. Oral administration of resveratrol to mice at a dose of 4000 mg/kg/day for 28
days did not reveal any adverse effects. In a review article prepared for the NIEHS,
NOAEL for resveratrol was determined as 300 mgkg/day on the basis of 28 day toxicity
study in rats.
In a chronic toxicity study in mice, feeding of resveratrol (22.4 mg/kg/day) in the
diet for one year had no adverse effects as evaluated by survival, motor function, insulin
sensitivity, organ pathology, and mitochondrial number. Based on the results of this
study, a NOAEL for resveratrol of 22.4 mgkg/day can be determined. Results of a
carcinogenicity study in p53 knockout mice show that resveratrol is not a carcinogen. In
this extensive and well designed study, over 45 organs were studied for histopathology.
Hydronephrosis (at dose levels of 1000 and 2000 mgkg/day for 6 months) and epithelial
cell hyperplasia (at dose level of 2000 mg/kg/day) of the urinary bladder noted in the
carcinogenicity study appear to be related to the elimination of high doses of resveratrol.
Blood analysis for clinical pathology at resveratrol dose of 1000 mg/kg/day for six
months did not reveal any adverse effects. Based on the results of this study a NOAEL
for resveratrol of 1000 mg/kg/day can be determined.
Results of reproductive and developmental toxicity studies indicate that
resveratrol is unlikely to cause toxicity. Prepubertal exposure to resveratrol did not affect
development of the reproductive tract as well as the mammary glands in female mice. In
a multigeneration mouse study, although resveratrol decreased absolute and relative
seminal vesicle weight and spleen weight in the P generation and the absolute vesicle and
kidney weight in the FI generation, some of the changes noted in P generation were not
found in the Fl generation. Although, resveratrol administration resulted in decreased
absolute and relative seminal vesicle weight and spleen weight in the P generation,
similar decreases, in absolute and relative seminal vesicle weight, and spleen weight were
not noted in Fl generation. In another multigeneration study, resveratrol did not affect
fertility. Resveratrol exposure did not affect male sexual development. The available
studies contradict the claimed estrogenic activity of resveratrol. In a 90-day study, oral
administration of resveratrol at a dose of 20 mg/kg/day enhanced sperm production by
stimulating the hypothalamic-pituitary-gonadal axis, without inducing adverse effects.
Available in vitro data suggest that depending on the biological environment,
resveratrol can have both complex estrogenic and antiestrogenic effects. Under
physiological conditions, resveratrol appears to act consistently as an antiestrogen
relative to inducing breast tumors. Resveratrol does not stimulate indices of uterine
growth and differentiation in immature rats. Resveratrol administration to rats even at
high doses did not significantly affect uterine wet weight, epithelial cell height, IGF-I
gene expression, or other estrogen target tissues. The available evidence shows that
resveratrol is not an estrogen substance.
The structural similarity between resveratrol and some estrogenic substances such
as diethylstilbestrol have raised questions about their pharmacological and toxicological
similarity. However, available studies show that resveratrol does not possess estrogenic
properties. In contrast to trans-resveratrol, diethylstilbestrol is known to have deleterious
effects on the male reproductive tract. The distinct activity of trans-resveratrol and
diethylstilbestrol can be explained by subtle differences in their molecules. Compared
with trans-resveratrol, diethylstilbestrol lacks the 3-OH and 5-OH groups, but possesses a
4-OH group and two additional ethyl groups. These features provide differential binding
characteristics to estrogen receptors. Diethylstilbestrol has an affinity to estrogen
receptors similar to estradiol and acts as a potent agonist.
In an in vitro bacterial reverse mutation assay in the presence or absence
microsomal activation system, resveratrol was not mutagenic. However, it induced
structural chromosomal aberrations in a Chinese hamster lung cell line and mononuclear,
polynuclear and karyorrhectic cells in micronucleus assay. In the sister chromatid
exchange test, resveratrol induced sister chromatid exchanges. Contrary to these positive
results, resveratrol is known for its antimutagenic and anticarcinogenic properties. The in
vitro DNA damaging effects of resveratrol are not physiologically comparable to in vivo
conditions where resveratrol acts as a potent antioxidant. The available evidence shows
that under m vivo conditions, resveratrol is unlikely to be genotoxic.
Resveratrol has been reported to inhibit corneal neorevascularization in mice.
However, other studies have shown that resveratrol appears to act as a �brake� to
unregulated angiogenesis (in tumors), but does not inhibit normal angiogenesis. Based
on the available evidence, resveratrol is not considered to pose any risk of altering normal
angiogenesis. Some in vitro studies indicate that resveratrol may inhibit platelet
aggregation. However, available evidence from other similar antioxidants indicates that
under in vivo conditions, resveratrol is unlikely to inhibit platelet aggregation. In in vivo
conditions, resveratrol is rapidly metabolized and eliminated and it is unlikely to reach
the concentrations required to inhibit platelet aggregation.
There is sufficient qualitative and quantitative scientific evidence, including
animal and human data, to determine safety-in-use or acceptable daily intake (ADI) for
resveratrol. Using the experimental data and an uncertainty factor (to bridge the lack of
safety information in humans) an AD1 of resveratrol for human use can be determined
Although, no systematic or typical 90-day toxicity study of resveratrol is available to
determine an ADI, there are several studies, such as short-term toxicity, subchronic
toxicity, chronic toxicity and carcinogenicity, which support the safety-in-use and the
AD1 determinations. In two separate 28 day studies, a NOAEL for resveratrol of 300
mg/kg/day was established. Using this experimental data and an uncertainty factor of
1000 (10 for interspecies, 10 for intraspecies, and 10 for chronic extrapolation), an AD1
of 0.3 mgkg/day (18 mg/person/day) for resveratrol is determined. This AD1
determination is supported by an extensive and well designed chronic study, in which
mice were fed a high calorie diet and resveratrol for one year. In this study, 22.4
mgkg/day was the highest dose of resveratrol used and the results support a NOAEL of
22.4 mgkg/day. Application of an uncertainty factor of 100 to the NOAEL of 22.4
mgkg/day from the chronic mouse study yields an AD1 of 0.224 mg of resveratrolkgper
day. An additional investigation that supports the above AD1 is a well designed six month
carcinogenicity study in p53 knockout mouse. In the carcinogenicity study, the NOAEL
for resveratrol of 1000 mgkgiday was established. In this extensive study, clinical
pathology and histology of 45 organs were evaluated. Using an uncertainty factor of 1000
(10 for interspecies, 10 for intraspecies and 10 for database deficiency), an AD1 of 1
mgkg/day for resveratrol is determined.
From the above discussion and supporting evidence, an AD1 of 0.3 mgkg/day or
18 mg/person/day (for an individual weighing 60 kg) is established. This daily level of
intake, if ingested daily over a lifetime, is projected to be safe. The estimated 90th
percentile intake of resveratrol (3.99 mg/day or 0.07 mgkg/day) from the intended uses
is approximately 4-fold lower than the AD1 (0.3 mg/kg/day) determined on the basis of
available safety data.
Because of lack of typical toxicity studies, in addition to the above described
standard approach of AD1 determination, a �weight of the evidence� approach was also
used to determine safety in use of resveratrol at the estimated daily intake level. In this
context, any body of literature on biological effects and experience of historical ingestion
of resveratrol from food is taken in to account to provide adequate assurance that any
potential to produce adverse effects in humans at the estimated daily intake levels is
understood and predictable.
All available studies on resveratrol were critically reviewed for quality and
suitability for use in this safety assessment. Particular emphasis was placed on studies
that focused on safety related parameters. The review yielded ten metabolism/
pharmacokinetic (in vitro and in vivo) studies, four short-term toxicity studies, one
chronic study, one carcinogenicity, six reproductioddevelopmental studies, three
mutagenicity studies, and several in vitro studies on estrogen-like activity, antiangiogenesis,
and platelet aggregation effects. Regarding these end points the review
revealed that:
1. Following oral administration, resveratrol is rapidly absorbed, metabolized
and eliminated. The bioavailability of resveratrol is low. The plasma half life
of resveratrol in human was reported as 9.2 hours.
2. Short-term (28-day) toxicity studies did not reveal adverse effects at doses up
to 300 mg/kg/day.
3. In a chronic one-year study in mice, resveratrol at a dose level of 22.4
mg/kg/day did not cause any adverse effects.
4. Resveratrol was not carcinogenic in a mouse study. In this study, safety
related parameters such as clinical pathology and histology of 45 organs did
not reveal toxicity at doses up to 1000 mg/kg/day.
5. The available evidence supports the conclusion that at low dose levels,
resveratrol is unlikely to cause reproductive and developmental toxicity.
6. Depending on the biological environment, resveratrol has complex estrogenic
and antiestrogenic effects. The available evidence shows that resveratrol is not
estrogenic.
7. Based on the available evidence, resveratrol is not considered to pose any risk
of altering normal angiogenesis or platelet aggregation.
8. There is a long history of resveratrol consumption by humans with written
records dating back to ancient times.
Across the dataset, adverse effects of resveratrol on relevant endpoints were noted
generally at doses greater than 300 mg/kg/day. The intended use levels of resveratrol,
resulting in 90th percentile intake of 0.07 mg/kg/day is over 4000-fold lower than the
lowest dose at which no significant adverse effects were noted. The available dataset
show that adverse effects of resveratrol are not likely to occur at resveratrol exposure
levels of 0.07 mg/kg/day. It is concluded that, under the intended and expected conditions
of use, resveratrol does not pose a health risk to humans.
Moderate consumers of red wine (such as muscadine) with high levels of
resveratrol are likely to ingest 4 mg resveratrol/day or 0.067 mg/kg/day. Although, exact
figures are not available, it is well known that there are significant numbers of individuals
who consume red wine daily in moderation for health benefits. The resulting 90th
percentile intake of resveratrol (3.99 mg/person/day) from the intended uses is similar to
individuals consuming red wine (in moderation) with high levels of resveratrol.
In summary, on the basis of scientific procedures�, and history of exposure from
natural sources, the consumption of resveratrol as an added food ingredient is considered
safe at levels up to 10 ppm in bottled water (%ear waters�). The intended uses are
compatible with current regulations, i e , resveratrol is used in bottled drinking water
under the broad category non-alcoholic beverages [21 CFR �170.3(n)(3)], and is
produced according to current good manufacturing practices (cGMP).
CONCLUSION
Based on a critical evaluation of the publicly available data summarized above,
the Expert Panel members whose signatures appear below, have individually and
collectively concluded that resveratrol, meeting the specifications cited above, and when
used as a nutrient supplement [21 CFR 170.3(0)(20)] at maximum use levels of up to 10
ppm in bottled water (�near waters�) [21 CFRl70.3(n)(3)] is safe.
It is also our opinion that other qualified and competent scientists reviewing the
same publicly available toxicological and safety information would reach the same
conclusion. Therefore, we have also concluded that resveratrol, when used as descnhed,
is GRAS based on scientific procedures.
REFERENCES
Abd El-Mohsen, M., Bayele, H., Kuhnlem G., Gibson, G., Debnam, E., Kaila Srai, S.,
Rice-Evans, C., Spencer, J.P. (2006) Distribution of [3H]-trans-resveratrol in rat tissues
following oral administration. British Journal ofNutritron 96:62-70.
Abrams, T.A., Boocock, D.J., Brenner, D.E., Faust, G.E.S., Gescher, A.J., Lao, C.D.,
Normolle, D., Sharma, R.A. and Steward, W.P. (2004). Phase I single-dose safety and
pharmacokinetics clinical study of resveratrol. Contract Number: NCI-NO1 -CN-25025
Ahmad, A,, Syed, F.A., Singh, S. and Hadi, S.M. (2005) Prooxidant activity of
resveratrol in the presence of copper ions: mutagenicity in plasmid DNA. Toxicology
Letters 159: 1-1 2.
Andlauer, W., Kolb, J., Siebert, K. and Furst, P. (2000) Assessment of resveratrol
bioavailability in the perfused small intestine of the rat. Drugs Experimental and Clinical
Research 26:47-55.
Ashby, J., Tinwell, H., Pennie, W., Brooks, A.N., Lefevre, P.A., Beresford, N. and
Sumpter, J.P. (1999) Partial and weak oestrogenicity of the red wine constituent
resveratrol: Consideration of its super agonist activity in MCF-7 cells and its suggested
cardiovascular protective effects. Journal ofApplied Toxicologv 19:39-45.
Baur, J.A., Pearson, K.J., Price, N.L., Jamieson, H.A., Lerin, C., Kalra, A,, Prabhu, V.V.,
Allard, J.S., Lopez-Lluch, G., Lewis, K., Pistell, P.J., Poosala, S., Becker, K.G., Boss, O.,
Gwinn, D., Wang, M., Ramaswamy, S., Fishbein, K.W., Spencer, R.G., Lakatta, E.G., Le
Couteur, D., Shaw, R.J., Navas, P., Puigserver, P., Ingram, D.K., de Cabo, R. and Sinclair,
D.A. (2006) Resveratrol improves health and survival of mice on a high-calorie diet.
Nature 444:337-342.
Bertelli, A., Bertelli, A.A.E., Gozzini, A. and Giovannini, L. (1998a) Plasma and tissue
resveratrol concentrations and pharmacological activity. Drugs Experimental and
Clinical Research 24: 133-1 38.
Bertelli, A.A.E., Giovannini, L., Stradi, R., Urien, S., Tillement, J.-P. and Bertelli, A.
(1998b) Evaluation of kinetic parameters of natural phytoalexin in resveratrol orally
administered in wine to rats. Drugs Experimental and Clinical Research 24:5 1-55.
Bhat, K.P, Lantvit, D., Christov, K., Mehta, R.G., Moon, R.C. and Pezzuto, J.M. (2001)
Estrogenic and antiestrogenic properties of resveratrol in mammary tumor models.
Cancer Research 61:7456-7463.
Bhat, K. P. and Pezzuto, J. M. (2001) Resveratrol exhibits cytostatic and antiestrogenic
properties with human endometrial adenocarcinoma (Ishikawa) cells. Cancer Research
Boocock, D.J., Brenner, D.E., Gescher, A.J., Steward, W.P. Brown, V. and Normolle, D.
(2006) Phase I repeat-dose clinical study of safety, pharmacokinetics and
pharmacodynamics of resveratrol. Contract Number: NCI-NO 1 -CN-25025.
Brakenhielm, E., Cao, R. and Cao, Y. (2001) Suppression of angiogenesis, tumor growth,
and wound healing by resveratrol, a natural compound in red wine and grapes. FASEB
Journal 15(8):U568-U581.
Burkitt, M.J. and Duncan, J. (2000) Effects of frans-resveratrol on copper-dependent
hydroxyl radical formation and DNA damage: Evidence for hydroxyl-radical scavenging
and a novel, glutathione-sparing mechanism of action. Archives of Biochemisfry
Biophysics 38 1 :253-263.
Cao, Z., Fang, J., Xia, C., Shi, X. and Jiang, B.H. (2004) frans-3,4,5'-Trihydroxystibene
inhibits hypoxia-inducible factor 1 alpha and vascular endothelial growth factor
expression in human ovarian cancer cells. Clinical Cancer Research 10:5253-5263.
Conquer J, Maiani G, Azzini E, Raguzzini, A. and Holub, B.J. (1998) Supplementation
with quercetin markedly increases plasma quercetin concentrations without effect on
selected risk factors for heart disease in healthy subjects. Journal of Nutrition 128593-
597.
Crowell, J. A., Korytko, P. J., Morrissey, R. L., Booth, T. D. and Levine, B. S. (2004)
Resveratrol-associated renal toxicity. Toxicological Sciences 82:614-619.
CSFII (2002). Foods Commonly Eaten in the United States. Quantities Consumed Per
Eating Occasion and in a Day, 1994-96. Eds: Smiciklas-Wrightl, H., Mitchelll, D.C.,
Mickle, S.J., Cook, A.J. and Goldman, J.D.
Damianaki, A., Bakogeorgou, E., Kampa, M., Notas, G., Hatzoglou, A,, Panagotou, S.,
Gemetzi, C., Kouroumalis, E., Martin, P. and Castanas, E. (2000) Potent inhibitory action
of red wine polyphenols on human breast cancer cells. Journal of Cellular Biochemistry
De Santi, C., Pietrabissa, A,, Mosca, F. and Pacifici, G.M. (2000a) Glucuronidation of
resveratrol, a natural product, present in grape and wine, in the human liver. Xenobiotica
De Santi, C., Pietrabissa, A,, Spisni, R., Mosca, F. and Pacifici, G.M. (2000b) Sulphation
of resveratrol, a natural compound present in wine, and its inhibition by natural
flavonoids. Xenobiotica 302357466.
De Santi, C., Pietrabissa, A,, Spisni, R., Mosca, F. and Pacifici, G.M. (2000~)S ulphation
of resveratrol, a natural product present in grapes and wine, in the human liver and
duodenum. Xenobiotica 30:609-617.
Emanuele, M.A. and Emanuele, N.V. (1998) Alcohols effects on male reproductions.
Alcohol Health and Research World 22: 195-201.
Freyburger, A,, Hartmann, E., Hildebrand, H. and Krotlinger, F. (2001) Differential
response of immature rat uterine tissue to ethinylestradiol and the red wine constituent
resveratrol. Archives of Toxicology 74: 709-71 5
Fukuda, S., Kaga, S., Zhan, L., Bagchi, D., Das, D.K., Bertelli, A. and Maulik, N. (2006)
Resveratrol ameliorates myocardial damage by inducing vascular endothelial growth
factor-angiogenesis and tyrosine kinase receptor Flk-1 . Cell Biochemistry and Biophysics
44:43-49.
Fukuhara, K. and Miyata, N. (1998) Resveratrol as a new type of DNA-cleaving agent.
Bioorganic Medicinal Chemistry Letters 8:3 187-3 192.
Garvin S, Ollinger K, Dabrosin C. (2006) Resveratrol induces apoptosis and inhibits
angiogenesis in human breast cancer xenografts in vivo. Cancer Letters 231: 113-122.
Gehm, B.D., J.M. McAndrews, P.-Y. Chien, and J.L. Jameson. (1997) Resveratrol, a
polyphenolic compound found in grapes and wine, is an agonist for the estrogen receptor.
Proceeding ofNational Academy ofSciences 94(25):14138-14143.
Gescher, A.J. and Steward, W.P. (2003) Relationship between mechanisms,
bioavailibility, and preclinical chemopreventive efficacy of resveratrol: a conundrum.
Cancer Epidemiology Biomarkers Prevention 121953-957.
Gu, X., Creasy, L., Kester, A. and Zeece, M. (1999) Capillary electrophoretic
determination of resveratrol in wines. Journal of Agriculture and Food Chemistry 47:
Gusman, J., Malonne, H. and Atassi, G. (2001) A reappraisal of the potential
chemopreventive and chemotherapeutic properties of resveratrol. Carcinogenesis
22:1111-1117.
Hebbar V, Shen G, Hu R, Kim BR, Chen C, Korytko PJ, Crowell JA, Levine BS, Kong
AN. (2005) Toxicogenomics of resveratrol in rat liver. Life Sciences. 76:2299-23 14.
Hendler, S.S. and Rorvik, D. (2001) Resveratrol. In: PDR for Nutritional Supplements.
Medical EconomicsTM Thomson Healthcare, Montvale, NJ, pp. 397-401.
Henry, L.A. and Witt, D.M. (2006) Effects of neonatal resveratrol exposure on adult male
and female reproductive physiology and behavior. Developmental Neurosciences 28: 186-
195.
Henry, L.A. and Witt, D.M. (2002) Resveratrol: phytoestrogen effects on reproductive
physiology and behavior in female rats. Hormones and Behavior 41 (2):220-8.
Horn, T.L., Cwik, M.J., Monissey, R.L., Kapetanovic, I., Crowell, J.A., Booth, T.D. and
McCormick, D.L. (2007). Oncogenicity evaluation of resveratrol in p53(+/-) @53
knockout) mice. Food and Chemical Toxicology 45:55-63.
Howitz, K.T., Bitterman, K.J., Cohen, H.Y., Lamming, D.W., Lavu, S., Wood, J.G.,
Zipkin, R.E., Chung, P., Kisielewski, A., Zhang, L.L., Scherer, B. and Sinclair, D.A.
(2003) Small molecule activators of sirtuins extend Saccharomyces cerevisiae lifespan.
Nature 425: 191-196.
Hung, L.M., Chen, J.K., Huang, S.S., Lee, R.S. and Su, M.J. (2000) Cardioprotective
effect of resveratrol, a natural antioxidant derived from grapes. Cardiovascular Research
Jang, M. and Pezzuto, J.M. (1999) Cancer chemopreventive activity of resveratrol. Drugs
Experimental and Clinical Research 25:65-77.
Juan, M.E., Gonzalez-Pons, E., Munuera, T., Ballester, J., Rodriguez-Gil, J.E. and Planas,
J.M. (2005) trans-Resveratrol, a natural antioxidant from grapes, increases sperm output
in healthy rats. Journal ofNutrition 135:757-760.
Juan, M.E., Vinardell, M.P. and Planas, J.M. (2002) The daily oral administration of high
doses of trans-resveratrol to rats for 28 Days is not harmful Journal of Nutrition
132:257-260.
Kaga, S., Zhan, L., Matsumoto, M. and Maulik, N. (2005) Resveratrol enhances
neovascularization in the infarcted rat myocardium through the induction of thioredoxin-
1, heme oxygenase-1 and vascular endothelial growth factor Journal of Molecular and
Cellular Cardiology 39:813-822.
Khanna, S., Roy, S., Bagchi, D., Bagchi, M. and Sen, C.K. (2001) Upregulation of
oxidant-induced VEGF expression in cultured keratinocytes by a grape seed
proanthocyanidin extract. Free Radical Biology and Medicine 3 1 :38-42.
Khanna, S., Venojarvi, M., Roy, S., Sharma, N., Trikha, P., Bagchi, D., Bagchi, M. and
Roy, S. (2002) Dermal wound healing properties of redox-active grape seed
proanthocyanidins. Free Radical Biology and Medicine 33: 1089- 1096.
Kim, H.J., Chang, E.J., Bae, S.J., Shim, S.M., Park, H.D., Rhee, C.H., Park, J.H. and
Choi, S.W. (2002) Cytotoxic and antimutagenic stilbenes from seeds of Paeonia
lactijlora Archives of Pharmaceutical Research 25:293-299.
Kimura, Y. and Okuda, H. (2001) Resveratrol isolated from Polygonum cuspidatum root
prevents tumor growth and metastasis to lung and tumor-induced neovascularization in
Lewis lung carcinoma-bearing mice Journal ofNutrition 131: 1844-1 849.
Kirk, R.I., Deitch, J.A., Wu, J.M. and Lerea, K.M. (2000) Resveratrol decreases early
signaling events in washed platelets but has little effect on platelet in whole blood. Blood
Cells Molecules and Disease 26: 144- 150.
Kubo K, Arai 0, Omura M, Watanabe R, Ogata R, Aou S. (2003) Low dose effects of
bisophenol A on sexual differentiation of the brain and behavior in rats. Neuroscience
Research 45:345-56.
Kuhnle, G., Spencer, J.P.E., Chowrimootoo, G., Schroeter, H., Debnam, E.S., Srai,
S.K.S., Rice- Evans, C. and Hahn, U. (2000) Resveratrol is absorbed in the small
intestine as resveratrol glucuronide. Biochemical and Biophysical Research
Communication 272:212-217.
Kyselova, V., Peknicova, J., Buckiova, D. and Boubelik, M. (2003). Effects of pnonylphenol
and resveratrol on body and organ weight and in vivo fertility of outbred
CD-1 mice. Reproductive Biology and Endocrinology 1:30 (10 pages).
Lu, R. and Serrero, G. (1999) Resveratrol, a natural product derived from grape, exhibits
antiestrogenic activity and inhibits the growth of human breast cancer cells. Journal of
Cell Physiologv 179:297-304.
Matsuoka, A,, Furuta, A,, Masayasu, O., Fukuhara, K. and Miyata. N. (2001) Resveratrol,
a naturally occurring polyphenol, induces sister chromatid exchanges in a Chinese
hamster lung (CHL) cell line. Mutation ofResearch 494:107-113.
Mgbonyebi, O.P., Russo, J. and Russo, I.H. (1998) Antiproliferative effect of synthetic
resveratrol on human breast epithelial cells. International Journal of Oncology 12:865-
869.
Nakagawa, H., Kiyozuka, Y., Uemura, Y., Senzaki, H., Shikata, N., Hioki, K. and
Tsubura, A. (2001) Resveratrol inhibits human breast cancer cell growth and may
mitigate the effect of linoleic acid, a potent breast cancer cell stimulator. Journal of
Cancer Research and Clinical Oncology 127:258-264.
NIEHS (2002) National Institute of Environmental Health Sciences. trans-Resveratrol:
CAS No.501-36-0. Review of Toxicological Literature under Contract No. NOl-ES-
65402, Integrated Laboratory Systems, P.O. Box 13501, Research Triangle Park, North
Carolina. pp. 1-64. Article available on line at website:
http://ntp.niehs.nih.gov/ntp/htdocs/Chem Background/ExSumPdf/Resveratrol.pdf
(Website visited on December 5, 2006.)
Nikaido, Y., Danbara, N., Tsujita-Kyutoku, M., Yuri, T., Uehara, N. and Tsubura, A.
(2005) Effects of prepubertal exposure to xenoestrogen on development of estrogen
target organs in female CD-1 mice. In Vivo 19:487-494.
Nikaido Y; Yoshizawa K; Danbara N; Tsujita-Kyutoku M; Yuri T; Uehara N; Tsubura A
(2004) Effects of maternal xenoestrogen exposure on development of the reproductive
tract and mammary gland in female CD-1 mouse offspring. Reproductive Toxicology
Oak, M.H., El Bedoui, J. and Schini-Kerth, V.B. (2005) Antiangiogenic properties of
natural polyphenols from red wine and green tea. Journal of Nutritional Biochemistry
Pace-Asciak C, Hahn S, Diamandis E, et al. The red wine phenolics trans-resveratrol and
quercetin block human platelet aggregation and eicosanoid synthesis: implications for
protection against coronary heart disease. Clinica Chimrca Acta. 1995;235:207-219.
Saint-Jean, N. (2001) Volume of bottled water consumption. In Physics Factbook. (Ed.
Elert, G.). Article available on line at website:
http://hyuertextbook.co~facts/2001/NaomiSaintJean.sh(twmelb site visited on February
28,2007).
Sen, C.K., Khanna, S., Gordillo, G., Bagchi, D., Bagchi, M. and Roy, S. (2002) Oxygen,
oxidants, and antioxidants in wound healing: an emerging paradigm. Annals of New York
Academy of Sciences 957:239-249.
Sgambato, A., Ardito, R., Faraglia, B., Boninsegna, A,, Wolf, F.I. and Cittadini, A.
(2001) Resveratrol, a natural phenolic compound, inhibits cell proliferation and prevents
oxidative DNA damage. Mutation Research 496:171-180.
Stewart, J.R., Artime, M.C. and O�Brian, C.A. (2003) Resveratrol: a candidate nutritional
substance for prostate cancer prevention. Journal of Nutrition 133(7 Suppl):244OS-2443S.
44 000047
Szaefer, H., Cichocki, M., Brauze, D. and Baer-Dubowska, W. (2004) Alteration in phase
I and I1 enzyme activities and polycyclic aromatic hydrocarbons-DNA adduct formation
by plant phenolics in mouse epidermis. Nutrition and Cancer 48:70-77.
Turner, R.T., Evans, G.L., Zhang, M., Maran, M. and Sibonga, J.D. (1999) Is resveratrol
an estrogen agonist in growing rats? Endocrinology 140:50-54.
Uenobe, F., Nakamura, S. and Miyazawa, M. (1997) Antimutagenic effect of resveratrol
against Trp-P-1. Mutation Research 373:197-200.
Valenzano, D.R., Terzibasi, E., Genade, T., Cattaneo, A,, Domenici, L., Cellerino, A,,
(2006) Resveratrol prolongs lifespan and retards the onset of age-related markers in a
short-lived vertebrate. Current Biology 16:296-300.
Violi, F., Pignatelli, P. and Pulcinelli, F.M. (2002) Synergism Among Flavonoids in
Inhibiting Platelet Aggregation and H202 Production. Circulation. 105:e53
Vyas, S., Asmerom, Y. and De Leon, D.D. (2005) Resveratrol regulates insulin-like
growth factor-I1 in breast cancer cells. Endocrinology 146:4224-4233.
Walle, T., Hsieh, F., Delegge, M.H., Oatis, J.E. Jr. and Walle, U.K. (2004) High
absorption but very low bioavailability of oral resveratrol in humans. Drug Metabolism
and Disposition 32: 1377-1 3 82.
Walle, T., Walle, U.K., Sedmera, D. and Klausner, M. (2006) Benzo[A]pyrene-induced
oral carcinogenesis and chemoprevention: studies in bioengineered human tissue. Drug
Metabolism and Disposition 34:346-350.
Wenzel, E. and Somora, V. (2005) Metabolism and bioavailability of trans-resveratrol.
Molecular Nutrition and Food Research 49:472-48 1
Wine Institute (2006). Key facts: per capita wine consumption in selected countries
1 997-200 1. Internet address: http://www.wineinstitute.org/communications/siatistics/
keyfacts~worldpercapitaconsumption02.htmW. ebsite accessed on November 17,2006.
Yang, F., Zhang, T. and Ito, Y. (2001) Large-scale separation of resveratrol,
anthragylcoside A and anthraglycoside B from Polygonum cuspidatum Sieb. Et Zucc by
high-speed counter-current chromatography. Journal ofChromatography A 91 9: 443-448.
Zamora-Ros, R., Urpi-Sarda, M., Lamuela-Raventos, R.M., Estruch, R., Vazquez-Agell,
M., Serrano-Martinez, M., Jaeger, W. and Andres-Lacueva, C. (2006) Diagnostic
performance of urinary resveratrol metabolites as a biomarker of moderate wine
consumption. Clinical Chemistry 52: 1373-1380.
Appendix: I
FOOTNOTES:
3 Modeled after that described in section 201(s) ofthe Federal Food, Drug, and Cosmetic Act, As Amended.
See also attachments (curriculum vitae) documenting the expertise of the Panel members.
4 Nutrient supplements. Substances which are necessary for the body's nutritional and metabolic processes.
5 Nutrient supplements Substances which are necessary for the body�s nutritional and metabolic processes.
6 The categoy includes Beverages and beverage bases, nonalcoholic, including only special or spiced teas,
soft drinks, coffee substitutes, and fruit and vegetable flavored gelatin drinks
7 abnormal forward curvature of the spine in the lumbar region
8 a bluish area of the brain stem with many norepinephrine-containing neuron
9 21 CFR 5170.3 Definitions (h) Scientific procedures include those human, animal, analytical, and other
scientific studies, whether published or unpublished, appropriate to establish the safety of a substance.
- To bookmark this page (add it to your favorites), please click the image to the left.
- If you wish to link to this page, you can do so by referring to the URL address below this line.
https://theodora.com/drugs/abilify_tablets_otsuka_america.html