|Home | About | Journals | Submit | Contact Us | Français|
To characterize the subchronic oral toxicity of resveratrol, CD rats received daily gavage doses of 0, 200, 400, or 1000 mg resveratrol/kg/day, and beagle dogs received daily capsule doses of 0, 200, 600, or 1200 mg resveratrol/kg/day for 90 days. Resveratrol induced only minimal toxicity, consisting of dose-related reductions in body weight gain in female rats and both sexes of dogs, and a statistically significant increase in bilirubin levels in rats at the 1000 mg/kg/day dose. Clinical observations, hematology, ophthalmology, neurotoxicity evaluations (functional observational batteries), organ weights, and gross pathology provided no biologically significant evidence of resveratrol toxicity in either species. In rats, the high dose of resveratrol reduced the incidence of cardiomyopathy; no other microscopic changes were seen. Histopathologic changes in dogs were limited to minimal inflammatory infiltrates in the kidney and urinary bladder, which were not considered toxicologically significant. A cardiovascular safety pharmacology (telemetry) study in dogs revealed no evidence of resveratrol toxicity. Based on body weight effects, the No Observed Adverse Effect Level (NOAEL) for resveratrol was 200 mg/kg/day in rats and 600 mg/kg/day in dogs. The apparent cardioprotective activity of resveratrol in rats demonstrates that its potentially beneficial activities may extend beyond efficacy in cancer prevention.
Resveratrol (3,5,4’-trihydroxy-trans-stilbene; Fig. 1) is a naturally occurring polyphenol phytoalexin that is present in grapes and several other common foodstuffs (Sanders et al., 2000). Resveratrol demonstrates a broad range of potentially desirable biological activities, including cancer prevention (reviewed by Kraft et al., 2009 and Brisdelli et al., 2009); cardioprotection (Hung et al., 2000; Das and Maulik, 2006); and life span prolongation (Howitz et al., 2003; Valenzano et al., 2006). Resveratrol has been studied extensively for its activity in cancer prevention: in laboratory animal models, resveratrol confers significant protection against cancer induction in the skin (Jang et al., 1997; Roy et al., 2009); mammary gland (Banerjee et al., 2002; Provinciali et al., 2005); prostate (Harper et al., 2009); liver (Biyashee and Dhir, 2009); oral cavity (Berta et al., 2010); esophagus (Woodall et al., 2010); and colon (Tessitore et al., 2000; Sale et al., 2005), among other tissues. In addition to its cancer chemopreventive activity, resveratrol can inhibit the growth and/or induce the regression of established tumors in xenograft models for cancer of the breast (Garvin et al., 2006); prostate (Wang et al., 2008); pancreas (Oi et al., 2010); ovary (Lee et al., 2009); urinary bladder (Bai et al., 2010); and stomach (Zhou et al. 2005).
Resveratrol demonstrates a broad range of biological activities that could potentially underlie its anticarcinogenic efficacy, and the biochemical and molecular mechanism(s) of resveratrol action remain under intensive investigation. Activities of resveratrol that may be relevant to cancer chemoprevention include modulation of the activity of carcinogen metabolizing enzymes (Jang et al., 1997); scavenging of free radicals (Manna et al., 2000); anti-inflammatory activity and inhibition of oxidative stress (Bishayee et al., 2010); induction of growth arrest and apoptosis in preneoplastic and neoplastic cells (Joe et al., 2002; Whyte et al., 2007; Kai et al., 2010); inhibition of cyclooxygenase (Subbaramaiah et al., 1998; Szewczuk et al., 2004); inhibition of aromatase (Sun et al., 2010); and inhibition of angiogenesis (Tseng et al., 2004; Garvin et al., 2006).
The broad chemopreventive activity of resveratrol in experimental models, and its relatively modest toxicity profile following short-term administration in rodents (Juan et al., 2002; Crowell et al., 2004), suggests that this natural product merits consideration for study in clinical trials for cancer prevention. The present studies were conducted to provide a systematic evaluation of the subchronic toxicity of resveratrol in rats and dogs; to characterize its plasma drug levels in these species, and to generate data to support the identification of appropriate starting dose levels of resveratrol for use in clinical trials with this agent.
Prior to the initiation of in vivo experimentation, study protocols were reviewed and approved by the IIT Research Institute Animal Care and Use Committee. All aspects of the program involving animal care, use, and welfare were performed in compliance with United States Department of Agriculture regulations and the Guide for the Care and Use of Laboratory Animals (National Research Council, 1996). Both studies were conducted in full compliance with the Good Laboratory Practice Regulations of the United States Food and Drug Administration (21 CFR Part 58).
Male and female CD rats (Crl:CD®[SD]IGS) were received at approximately four weeks of age from virus-free colonies maintained at Charles River Laboratories (Raleigh, NC). Rats were housed individually in suspended stainless steel cages in a temperature-controlled room maintained on a 12 h light/dark cycle, and were held in quarantine for two weeks prior to the initiation of dosing. With the exception of scheduled fasting periods, rats were allowed free access to Certified Rodent Diet 5002 (PMI Nutrition International, Inc., Brentwood, MO) throughout the study. City of Chicago drinking water was supplied to rats ad libitum using an automatic watering system.
After release from quarantine, rats were assigned to groups of 20 per sex using a computer-based randomization procedure that blocks for body weights. Rats received daily oral (gavage) administration of resveratrol (Pharmascience, Inc. [formerly Royalmount Pharma], Montreal, Quebec) at doses of 200, 400, or 1000 mg/kg/day (1200, 2400, or 6000 mg/m2/day) for a minimum of 90 days; resveratrol doses were selected on the basis of the results of a previous 28-day toxicity study in rats (Crowell et al., 2004). Resveratrol was administered in a vehicle of 0.5% (w/v) aqueous methycellulose containing 0.2% (w/v) Tween 80 (Sigma Chemical Co., St. Louis, MO); a dosing volume of 10 ml/kg/day was used. Rats in the vehicle control group received daily oral (gavage) administration of vehicle only (10 ml/kg/day) for the same period.
Throughout the study, rats were observed a minimum of twice daily to monitor their general health status; detailed clinical examinations and measurements of body weight and food consumption were performed weekly. Indirect funduscopic ophthalmic examinations were performed on all animals during the quarantine period (pre-test) and during the final week of the treatment period. A functional observational battery (FOB) was performed on 5 rats/sex/group at pretest and during week 13 of dosing; the week 13 FOB was performed at 2 to 4 hours post-dosing. Evaluations in the FOB included: home cage observation, hand-held observation, open field observation (mobility/gait), body weight, body temperature, eye blink, pupil response, tail pinch, hindlimb extension, hearing (click response), vision, catalepsy, righting reflex, grip strength (forelimb and hind limb), and foot splay. Blood samples for clinical chemistry and hematology evaluations were collected from fasted rats (10/sex/group) during weeks 4 and 12 of dosing; coagulation parameters were evaluated in blood samples obtained from the same animals at study termination. Clinical pathology assays were performed using automated instruments (Synchron CX5 Clinical Chemistry Analyzer [Beckman Instruments, Brea, CA]; Advia System 120 Hematology Analyzer [Bayer Corp., Tarrytown, NY]; MLA Electra 900 Automatic Coagulation Timer [Hemoliance, Raritan, NJ]).
Levels of free (unconjugated) resveratrol and total (unconjugated + conjugated) resveratrol were quantitated in plasma samples collected at 1 h post-dosing on day 1 and during week 13. Plasma levels of free resveratrol were measured by HPLC, using an analytical method developed in our laboratories (Muzzio et al., submitted for publication). Total resveratrol values were quantitated by HPLC after enzymatic hydrolysis of plasma samples with β-glucuronidase/sulfatase (from Helix pomatia; Type H-5, lot number 084K3795, Sigma-Aldrich, St. Louis, MO).
After 13 weeks of resveratrol exposure, all rats were euthanized with sodium pentobarbital followed by exsanguination, and received a complete gross necropsy with tissue collection. At necropsy, weights of the adrenals, brain, heart, kidneys, liver, ovaries/testes, spleen, thyroids and uterus were collected. All gross lesions plus approximately 45 tissues per rat were collected and fixed in 10% neutral buffered formalin. All tissues from all rats in the high dose and vehicle control groups underwent histologic processing and were evaluated microscopically. Histologic processing and microscopic evaluation of tissues from rats in the middle and low dose groups were limited to gross lesions and identified target tissues.
Male and female purebred beagle dogs were received at approximately five months of age from Ridglan Farms, Inc. (Mount Horeb, WI), and were held in quarantine for approximately four weeks prior to randomization into experimental groups. Dogs were housed individually in floor-level pens in a temperature-controlled room maintained on a 12 h light/dark cycle. Dogs were provided with 400 g of Certified Canine Diet 5007 (PMI Nutrition International, Inc.) for a minimum of 2 h each day, and were permitted free access to City of Chicago drinking water supplied via an automatic watering system.
After release from quarantine, dogs were assigned to experimental groups of four dogs per sex using a computerized randomization procedure that blocks for body weight. Dogs received daily oral (capsule) exposure to resveratrol at doses of 200, 600, or 1200 mg/kg/day (4000, 12,000 or 24,000 mg/m2/day) for 91 consecutive days. Dogs in the control group received empty capsules only. Dose levels used in the study were selected on the basis of the results of a preliminary 28-day oral toxicity study in dogs (unpublished).
Throughout the study, dogs were observed a minimum of twice daily to monitor their general health status. Detailed clinical examinations and body weight measurements were performed once weekly, and food consumption was quantitated daily. Indirect funduscopic ophthalmic examinations and electrocardiographic (ECG) evaluations were performed on all dogs during quarantine (pre-test) and during the final week of the treatment period. ECGs were reviewed for heart rate and rhythm, amplitude of the P wave and QRS complex, and duration of the P wave, PR, QRS, and QT intervals. To identify possible neurotoxic effects of resveratrol, all dogs underwent FOB evaluations at pretest and during the final week of resveratrol administration. FOB parameters included body weight, body temperature, cage-side observation, gait and posture, wheelbarrow test, hopping tests (forelimb and hindlimb), placing response, patellar reflex, perineal reflex, hindlimb and forelimb flexor reflexes, menace reflex, papillary reflex, and righting reflex.
Blood samples for determination of levels of free resveratrol, resveratrol glucuronide, and resveratrol sulfate were collected from all resveratrol-treated animals at one hour post-dosing on day 1 and during week 13. Plasma levels of free resveratrol and resveratrol conjugates were quantitated using a tandem mass spectrometer (API 3000; Applied Biosystems/MDS Sciex, Foster City, CA) equipped with a high performance liquid chromatograph (Agilent 1100; Agilent Technologies, Wilmington, DE) and methods established in our laboratories. Pharmacokinetic analyses were performed using WinNonlin Professional Edition software, version 5.0.1 (Pharsight Corporation, Mountain View, CA).
Blood samples for clinical chemistry, hematology and coagulation evaluations were collected from fasted dogs during pre-test and weeks 4 and 13 of dosing. Clinical pathology assays were performed as described for the 90-day toxicity study in rats, except that clinical pathology parameters were evaluated using a Synchron LX20 and coagulation parameters were measured using the STA Compact CT automatic coagulation instrument (Diagnostica Stago, Parsippany, NJ). Urine samples were collected from fasted dogs pre-test and during weeks 4 and 13 of dosing, and were analyzed by dipstick and microscopy.
One day after the final dose of resveratrol or vehicle, dogs were euthanized by barbiturate overdose and subjected to a complete necropsy with tissue collection. Weights of the adrenals, brain, heart, kidneys, liver, spleen, testes, thymus and thyroids were collected, and all gross lesions and approximately 45 tissues were collected from each animal and fixed in 10% neutral buffered formalin. All tissues collected from dogs in the high dose and vehicle control groups were processed by routine histologic methods and evaluated microscopically. Histologic processing and microscopic evaluation of tissues from dogs in the middle and low dose groups were limited to gross lesions and identified target tissues.
A separate cardiovascular safety pharmacology (telemetry) study was performed using a balanced Latin square design. In this study, eight naïve beagle dogs (4 per sex; Ridglan Farms) were implanted with telemetry transmitters (Data Sciences International, St. Paul, MN [DSI]) for remote monitoring of cardiovascular function. Telemetry data acquisition was performed using a signal receiver, ambient pressure monitor, DSI Data Exchange Network, and the DSI data acquisition system based on DataquestART™ software.
After recovery from implantation surgery, groups of two dogs (one per sex) received a single oral (capsule) dose of resveratrol at one of four dose levels (0 mg/kg [empty capsules; control], 200 mg/kg, 400 mg/kg, and 1000 mg/kg). After a one-week washout period, animals were rotated to a different dose level; weekly dosing was continued in this fashion until each animal had received a single capsule dose of resveratrol at each dose level.
General toxicology evaluations performed during the safety pharmacology study included twice daily mortality/moribundity observations, daily clinical observation, and weekly body weights. Body temperature and cardiovascular function were monitored continuously via telemetry for at least 2 hours prior to and 24 hours after each dose of resveratrol; cardiovascular parameters monitored included heart rate; diastolic, systolic and mean blood pressure; and electrocardiography. ECG waveforms were evaluated using DSI ECG Analysis 4.0 software, and were reviewed by a veterinary cardiologist. The ECG Analysis software recognizes ECG waveforms and measures the distance between waveform peaks to determine PR, RR, QRS and QT interval times within each 30-second interval. The QT interval was normalized for changes in heart rate by conversion to the corrected QT (QTc) interval using Fridericia’s formula [QTcF=QT/(RR)1/3].
Statistical evaluations of continuous data from toxicology studies were performed by analysis of variance (ANOVA), with post-hoc analyses performed using Dunnett’s test. Incidence data were compared by χ2 analysis or Fischer’s Exact test. A minimum significance level of p < 0.05 was used in all comparisons.
Statistical analyses of plasma drug level data (Tmax, Cmax, AUC) were performed using Systat software, version 10.2 (Systat Software Inc., Chicago, IL); data were analyzed either by t-test or by analysis of variance (ANOVA) followed, as necessary, by the post hoc Tukey’s test. Data for Cmax and AUC were normalized to the dose prior to log-transformation for statistical evaluation of dose-proportionality responses. A minimum significance level of p < 0.05 was used in all comparisons.
Daily gavage administration of resveratrol to rats for at least 90 days at doses of up to 1000 mg/kg/day (6000 mg/m2/day) induced no mortality in either sex. No evidence of treatment-related gross toxicity was identified during clinical observations of male or female rats exposed to any dose of resveratrol used in the study.
No effects of resveratrol on body weight were seen in male rats at any time in the study (Fig. 2). By contrast, a modest (8 to 9%) but statistically significant suppression of group mean body weight was observed during weeks 10 to 13 in female rats exposed to the high dose of resveratrol (1000 mg/kg/day; Fig. 3). Although absolute reductions in mean body weight were also seen at these times in female rats exposed to lower doses of resveratrol, mean body weights in the middle and low dose groups were not significantly different from the control group. Total body weight gain in female rats was decreased in dose-related fashion by resveratrol; decreases in total body weight gain were statistically significant at resveratrol doses of 400 and 1000 mg/kg/day (total body weight gains of 140 and 132 g, respectively, versus 158 grams in vehicle controls). Total body weight gain in female rats receiving resveratrol at 200 mg/kg/day (143 g) was not different from total body weight gain in the vehicle control group at the 5% level of confidence.
Resveratrol had no effect on food consumption in either sex during the 13-week dosing period. In addition, ophthalmologic examinations showed no evidence of ocular toxicity and FOB evaluations showed no evidence of any neurotoxicity following exposure to resveratrol.
At study termination, treatment-related alterations in clinical pathology parameters were limited to small, but statistically significant, increases in mean total bilirubin levels in the high dose group in both sexes (Table 1); significant increases in total bilirubin were also seen in high dose males at week 4 (Table 1). No other treatment-related changes were seen in any other hematology, clinical chemistry or coagulation parameters in rats exposed to resveratrol for 13 weeks; selected clinical chemistry data related to hepatic and renal function are provided in Table 1.
Organ weight measurements performed at the terminal necropsy identified a modest, but statistically significant and dose-related, hepatomegaly in both sexes (Table 2). Increased absolute liver weights were seen only in female rats in the high dose group. When normalized to body weight, increased relative liver weights were seen in female rats in all dose groups and in male rats in the middle and high dose groups. By contrast, however, these differences were not significant when normalized versus brain weights (Table 2). No other treatment-related changes in organ weights were seen in male or female rats. Ovarian and testes weights were comparable in all exposure groups, and do not support the hypothesis that resveratrol has phytoestrogenic or other hormonal activity in rats (Table 2).
Gross pathology at the terminal necropsy provided no evidence of resveratrol toxicity. All gross lesions that were observed at the terminal necropsy were consistent with findings that are commonly seen at the conclusion of 13-week toxicity studies in rats and, therefore, were interpreted as incidental findings.
Histopathologic evaluation of tissues in the high dose and control groups failed to identify any pattern of toxicity that was associated with subchronic exposure to resveratrol. No tissues were identified as targets of resveratrol toxicity; in particular, microscopic evaluation of the ovaries, testes, and other hormone-responsive tissues provided no evidence of any hormonal activity of resveratrol. Interestingly, however, a statistically significant decrease in the incidence of cardiomyopathy was observed in rats exposed to the high dose of resveratrol (Table 3).
Levels of both free and total (free + conjugated) resveratrol were below the limit of detection (< 0.035 μg/ml) in all animals in the vehicle control group. In groups receiving daily gavage exposure to resveratrol, plasma resveratrol levels increased with dose and duration of exposure. At both day 1 and week 13, ≥ 95% of total plasma resveratrol was in conjugated forms (sulfates or glucuronides) rather than as free drug (Figures 4 and and55).
Levels of free resveratrol were generally comparable in male and female rats at both Day 1 and Week 13 of exposure. On Day 1, mean plasma levels of free resveratrol in male rats were 0.166, 0.394, and 0.545 μg/ml in the low, middle, and high dose groups, respectively. Mean plasma levels of free resveratrol in male rats in these groups increased to 0.430, 0.912, and 1.87 μg/ml after 13 weeks of exposure (Fig. 4). In female rats, mean plasma levels of free resveratrol on Day 1 were 0.185, 0.308, and 0.547 μg/ml. After 13 weeks of resveratrol administration, exposure, mean plasma levels of free resveratrol in female rats increased to 0.486, 1.39, and 2.03 μg/ml (Fig. 5). Differences between mean plasma levels on Day 1 and Week 13 were statistically significant for all dose groups.
On day 1, mean total (free + conjugated) resveratrol in the plasma of male rats was 19.3, 23.4 and 28.8 μg/ml in the low, middle, and high dose groups, respectively (Fig. 4). Mean plasma levels of total resveratrol at 13 weeks were comparable to those measured on day 1 (18.4, 25.3, and 36.7 μg/ml). In female rats, mean plasma levels of total (free + conjugated) resveratrol in female rats on Day 1 were 17.0, 18.4, and 22.5 μg/ml in the low, middle, and high dose groups, respectively (Fig. 5). After 13 weeks of exposure, total resveratrol levels in female rats had increased to 19.0, 36.2, and 50.3 μg/ml.
Daily oral (capsule) administration of resveratrol to beagle dogs for 91 days at doses of 200, 600 or 1200 mg/kg/day (4000, 12,000 or 24,000 mg/m2/day) induced no treatment-related mortality or clinical signs of toxicity in either sex. Statistically significant, treatment-related decreases in mean body weight was seen in both sexes in the high dose group beginning at week 5 in male dogs and at week 6 in female dogs (Figures 6 and and7);7); at study termination, mean body weights in male and female dogs receiving the high dose of resveratrol were 15% and 17% below mean body weights in sex-matched controls. Decreases in body weight were associated with decreased food consumption in both sexes in the high dose group during most weeks of the study (data not shown).
Ophthalmic examinations, clinical pathology measurements, and electrocardiographic evaluations provided no evidence of resveratrol toxicity in any treated animal. At the terminal necropsy, gross pathology was unremarkable, and organ weights in all groups exposed to resveratrol were comparable to controls.
Histopathologic evaluation of tissues from the canine toxicity study provided no evidence of limiting toxicity of resveratrol; treatment-related changes were limited to inflammatory infiltrates in the urinary bladder and kidneys (Table 4). The severity of these changes ranged from minimal to mild, and most dogs demonstrated changes of minimal severity. The results of histopathologic evaluations demonstrated that all other tissues from dogs in all dose groups in both sexes were within normal limits.
Following oral (capsule) administration, resveratrol was absorbed systemically and most animals reached maximum plasma concentrations within 1 to 4 hours after dosing (data not shown). Maximum observed plasma concentration values (Cmax; mean ± SD) ranged from 1690 ± 622 to 9890 ± 1370 mg/ml (Table 5). Plasma Cmax increased proportionately with dose, and there were no significant differences between sexes or duration of exposure. Resveratrol AUC values (mean ± SD) ranged from 3620 ± 853 to 44,300 ± 8240 hr*ng/ml; AUC values increased proportionately with dose on day 1, but not at week 13.
As was seen in rats, dogs rapidly metabolized resveratrol to resveratrol glucuronide and resveratrol sulfate; both conjugates demonstrated similar plasma concentration time profiles (Table 5), but demonstrated no significant differences with either sex or duration of exposure. Overall, Cmax and AUC values for resveratrol glucuronide and resveratrol sulfate were greater than those for the parent compound. For resveratrol glucuronide, AUC values increased proportionately with dose on day 1, but not week 13. For resveratrol sulfate, increases in AUC on day 1 were not proportional to the increases in dose; however, AUC increases were proportional to the increase in dose at week 13. There were no significant changes in AUC values between day 1 and week 13 for any of the dose groups that indicated induction or accumulation for either resveratrol or its metabolites.
Administration of resveratrol orally via capsule using a balanced Latin-Square crossover design at doses of 200, 400 and 1000 mg/kg resulted in no treatment-related effects on any parameter investigated. No animals died during the study and no adverse effects were noted in any animal during clinical observations and physical examinations. There were no treatment-related differences in mean body weights or total body weight gains when compared to the control group (data not shown). Administration of resveratrol had no effect on body temperature, heart rate, blood pressure or ECG parameters (data not shown).
Daily oral administration of resveratrol to rats at doses of up to 1000 mg/kg/day and to dogs at doses of up to 1200 mg/kg/day resulted in dose-related increases in plasma levels of both free resveratrol and total (free + conjugated) resveratrol. In both species, more than 95% of plasma total resveratrol was present as resveratrol conjugates (resveratrol glucuronide and resveratrol sulfate), rather than as parent drug.
When administered to rats and dogs for 13 weeks, resveratrol induced only minimal toxicity. No mortality was seen in either species in any dose group, and clinical observations failed to identify any evidence of resveratrol toxicity. Body weight and body weight gain provided perhaps the most sensitive indicators of resveratrol toxicity. In rats, a modest (<10%) but significant reduction in mean body weight was seen between weeks 10 and 13 in females in the high dose (1000 mg/kg/day) group. Total body weight gain over the 13-week period of exposure was also significantly suppressed in female rats in the middle dose group (400 mg/kg/day; 11% suppression of body weight gain) and the high dose groups (1000 mg/kg/day; 15% suppression of body weight gain. When compared to vehicle controls, the high dose of resveratrol reduced mean body weights in male and female dogs by 15% and 17%, respectively.
Body weight effects in rats demonstrated no clear association with food consumption. By contrast, decreased body weights in dogs in the high dose group were associated with modest but statistically significant decreases in food consumption.
Clinical pathology provided little evidence of resveratrol toxicity in either species. In comparison to sex- and species-matched controls, statistically significant differences in clinical pathology parameters were seen only in total bilirubin levels in both sexes of rats in the high dose group. The observed increases in plasma bilirubin were not accompanied by gross or microscopic evidence of hepatic damage and, therefore, were interpreted to be of minimal toxicological significance. Increases in bilirubin were not seen in rats in either the mid or low dose groups, or in dogs exposed to resveratrol at any dose level.
Resveratrol did induce a dose-related, and apparently species-specific hepatomegaly in both sexes of rats. After 13 weeks of resveratrol exposure, statistically significant increases in absolute and/or relative (normalized to body weight) liver weights were present in both sexes of rats in the mid and high dose groups, and in female rats in the low dose group. Interestingly, differences in relative liver weights observed in rats when liver weight was normalized to body weight were not present when liver weight was normalized to brain weight. No treatment-related alterations in liver weight were observed in dogs in any dose group.
In spite of the observed hepatomegaly, no microscopic evidence of hepatoxicity was seen in either male or female rats exposed to resveratrol at any dose level. On this basis, the increases in liver weight that were observed in resveratrol-treated rats are interpreted as reflecting the activity of resveratrol as a modifier of hepatic enzymes involved in drug metabolism (Chun et al., 1999; Ciolino et al., 1999).
In a 4-week toxicity study of resveratrol in rats (Crowell et al., 2004), resveratrol toxicity was seen primarily in the high dose group receiving resveratrol at 3000 mg/kg/day. In that 4-week study, administration of the high dose of resveratrol induced decreases in body weight, decreases in food consumption, nephrotoxicity, increases in plasma total bilirubin levels, and increases in liver weight. In the present 13-week study, no evidence of renal toxicity was seen at the highest dose of resveratrol administered (1000 mg/kg/day). The lack of nephrotoxicity observed in the present study, when considered with the fact that the total doses of resveratrol administered in the two rat studies were comparable (1000 mg/kg/day for 13 weeks versus 3000 mg/kg/day for 4 weeks), suggests that resveratrol toxicity is unlikely to be cumulative and is more closely related to dose rate than total dose administered. This conclusion is consistent with short half-life of free resveratrol in the plasma, and its rapid metabolism of resveratrol to its glucuronide and sulfate conjugates.
Although experimental data are conflicting, the stilbenoid structure of resveratrol and its structural similarity to the potent synthetic estrogen, diethylstilbestrol, have generated concern that resveratrol may have estrogenic activity (Bowers et al., 2000). Freyberg and colleagues (2001) found no evidence of resveratrol estrogenicity in an in vivo uterotrophic assay in rats. Similarly, in a 13-week safety study of high purity trans-resveratrol in rats, no effects on reproductive markers were seen (Williams et al., 2009). However, Henry and Witt (2002) have reported that oral administration of resveratrol had effects on the estrous cycle, ovarian weight, and ovarian histopathology in adult female rats. In the present 13-week study, a statistically significant and apparently dose-related increase in relative ovarian weight was seen in female rats receiving resveratrol at 1000 mg/kg/day. However, the increase in relative ovarian weights appears to result primarily from dose-related decreases in body weight in resveratrol-treated females; absolute ovary weights in resveratrol-treated rats were not significantly different from those in vehicle controls. Furthermore, no effects of resveratrol on ovarian histopathology were seen in female rats at any dose level, nor were changes in testicular weights or histology seen in male rats. On this basis, it is concluded that the results of the present study do not provide compelling evidence to support the hypothesized activity of resveratrol as a phytoestrogen.
On the basis of statistically significant decreases in body weight gain in female rats exposed to resveratrol at 1000 and 400 mg/kg/day, the No Observed Adverse Effect Level (NOAEL) for subchronic oral administration of resveratrol in rats was defined as 200 mg/kg/day. The NOAEL for subchronic oral administration of resveratrol in dogs was 600 mg/kg/day; the NOAEL for dogs was defined on the basis of decreases in body weight seen in both sexes at 1200 mg/kg/day.
Williams et al. (2009) reported slightly lower food consumption and body weight in male and female rats receiving dietary supplementation with resveratrol at levels designed to provide 300 or 750 mg/kg body weight/day. These effects were interpreted as resulting from decreased palatability of the diet at these dose levels, rather than a specific adverse effect of resveratrol, and the NOAEL for dietary administration of resveratrol was determined to be 700 mg/kg/day. In our 13-week study, total body weight gain was statistically significantly decreased in female rats at the 400 and 1000 mg/kg/day levels; however, no decreased body weight gain was seen in male rats at any dose level. The effect on body weight gain in only one sex was not consistent with the Williams et al. (2009) study, but are similar to the data presented by Crowell et al. (2004), who reported a statistically significant reduction in body weight gain in female rats receiving daily gavage exposure to resveratrol for 28 days at 1000 mg/kg/day; no effects on body weight in male rats were observed by Crowell and colleagues at this resveratrol dose. The NOAEL in that study was determined to be 300 mg/kg/day. The differences in the NOAEL of resveratrol in the Williams et al. (2009) study (700 mg/kg/day), the Crowell et al. (2004) study (300 mg/kg/day) and our study (200 mg/kg/day) may have been due to the route of administration (dietary in the Williams et al. study) or the duration of exposure (4 weeks in the Crowell et al. study versus 13 weeks in our study).
An unexpected finding in the rat study was that administration of minimally toxic levels of resveratrol conferred significant protection against the appearance of cardiomyopathy in rats. When compared to vehicle controls, a statistically significant reduction in the total incidence of cardiomyopathy (males + females) was observed in rats exposed to the high dose of resveratrol. Although the incidences of cardiomyopathy in rats receiving the low and middle dose levels of resveratrol were not significantly reduced, the total incidences of cardiomyopathy in both groups were below that seen in vehicle controls.
The mechanism for the apparent cardioprotection conferred by resveratrol is unknown, but may be related to the antioxidant activity of this polyphenolic compound. Resveratrol has been proposed as the agent that may be responsible for the “French paradox”, the observation that the incidence of cardiac disease in France is lower than in other industrialized nations, in spite of the fact that the French consume a diet that is high in cholesterol and saturated fats (Renaud and Lorgeril, 1992). The observations from the rat study are consistent with this hypothesis.
This work was supported by contract N01-CN-43304 from the Chemoprevention Agent Development Research Group, Division of Cancer Prevention, National Cancer Institute. The authors thank Scott Garthwaite, Gail Dianis, Lynn Barron and Nicole Kozub for excellent technical assistance and Janet Mossman for assistance in preparation of the manuscript.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.