Comparative Toxicokinetics of Trans-resveratrol and Its Major Metabolites in Harlan Sprague Dawley Rats and B6C3F1/N Mice Following Oral and Intravenous Administration

Abstract

Trans-resveratrol (RES) is a naturally occurring stilbene found in numerous plants and foods. Due to its widespread human exposure and lack of toxicity and carcinogenicity data, RES was nominated to the National Toxicology Program for testing. To aid the toxicology studies, the dose, sex, and species differences in RES toxicokinetics was investigated in Harlan Sprague Dawley rats and B6C3F1/N mice following single intravenous (IV) (10 mg/kg) or oral gavage administration (312.5, 625, and 1250 mg/kg and 625, 1250, and 2500 mg/kg in rats and mice, respectively).

Following IV and gavage administration, systemic exposure of RES based on AUC was trans-resveratrol-3-O-β-D-glucuronide (R3G)>> trans-resveratrol-3-sulfate (R3S)>RES in both species. Following gavage administration T_{max_predicted} values were ≤ 263 min for both species and sexes. RES elimination half-life was longer in rats than mice, and shortest in male mice. Clearance was slower in mice with no apparent sex difference in both species. In both rats and mice, following gavage administration AUC increased proportionally to the dose. After gavage administration, enterohepatic recirculation of RES was observed in both rats and mice with secondary peaks occurring around 640 min in the concentration-time profiles. RES was rapidly metabolized to R3S and R3G in both species. Extensive first pass conjugation and metabolism resulted in low levels of the parent compound RES which was confirmed by the low estimates for bioavailability. The bioavailability of RES was low, ~12–31% and ~2–6% for rats and mice, respectively, with no apparent difference between sexes.

Introduction

Trans-resveratrol (RES) is a polyphenol found in many plant species, particularly grapes, peanuts, pistachios and berries (Burns et al., 2002; Zamora-Ros et al., 2008). It has long been used in herbal medicines (Arichi et al., 1982; Bret C. Vastano et al., 2000; Renaud and de Lorgeril, 1992). RES is purported to have multiple health benefits due to its anti-cancer (Delmas et al., 2006; Signorelli and Ghidoni, 2005; Su et al., 2007), antioxidant (Olas et al., 2010), and anti-inflammatory properties (Cui et al., 2010).

To date, no adverse health effects of RES in humans have been reported. In rabbits, repeated oral exposure to RES (1 mg/kg) promoted atherosclerosis in hypercholesterolemic animals (Wilson et al., 1996). Subchronic oral exposure to RES in CD rats (200 – 1000 mg/kg/day) and dogs (200 – 1200 mg/kg/day) did not show any treatment-related mortality or toxicity but decreases in mean body weight at 1200 mg/kg/day were observed in dogs (Johnson et al., 2011). Following subchronic oral exposure to RES (0, 300, 1000, and 3000 mg/kg/day) in CD rats, there were no observed effects at 300 mg/kg/day; but several adverse effects, including nephrotoxicity, labored breathing, decreased food consumption and body weight, rough coat, and diarrhea were observed at 3000 mg/kg/day (Crowell et al., 2004). Reduced overall body weight, increased ovarian weight, and disrupted estrous cycle were observed in gonadally intact female rats following oral exposure to RES (Henry and Witt, 2002), although the increased ovarian weight was considered to be due to decreased body weights in a subsequent study (Johnson et al., 2011). The no observed adverse effect level (NOAEL) of RES in dogs was 600 mg/kg/day while in rats reported values range between 200 and 700 mg/kg/day (Crowell et al., 2004; Johnson et al., 2011; Williams et al., 2009).

Available toxicokinetic (TK)/pharmacokinetic data for RES in animals and humans vary somewhat between studies. Based on the reported data, RES absorption in rodents varies, with peak plasma concentration between 15 min and 6 h (Bertelli et al., 1998, 1996a, 1996b; Kapetanovic et al., 2011; Soleas et al., 2001; Williams et al., 2009); a second peak observed between 4 and 24 h in rats (Marier et al., 2002) and in humans (Walle, 2011) is likely due to enterohepatic recirculation. The study-to-study variability may be related to differences in the experimental parameters, which may be confounded by the presence of enterohepatic recirculation, as a review of the study data does not suggest that the differences are dose related.

RES is rapidly and extensively metabolized (Abd El-Mohsen et al., 2006; Baur and Sinclair, 2006; Calamini et al., 2010; Gambini et al., 2015; Goldberg et al., 2003; Hoshino et al., 2010; Marier et al., 2002; Polycarpou et al., 2013; Sharan et al., 2012; Vitrac et al., 2003; Walle et al., 2004; Wenzel et al., 2005; Wenzel and Somoza, 2005) and excreted (Asensi et al., 2002; Soleas et al., 2001; Vitrac et al., 2003; Yu et al., 2002) in both rodents and humans; the four metabolites with the highest circulating concentrations reported were trans-resveratrol-4’-sulfate (R4’S), trans-resveratrol-3-sulfate (R3S), trans-resveratrol-4-O-β-D-glucuronide (R4’G) and trans-resveratrol-3-O-β-D-glucuronide (R3G) (Figure 1) (Boocock et al., 2007; Johnson et al., 2011; Kapetanovic et al., 2011; Sharan et al., 2012; Xiaofeng Meng et al., 2004). Unconjugated RES levels were reported to be ≤ 5% of total RES (free and conjugated forms) in rodent plasma due to the phase II conjugation of RES in the small intestine (Andlauer et al., 2000; Aumont et al., 2001; Kuhnle et al., 2000). Low levels of unconjugated RES (<1.7%) was also reported in urine and feces of rats following a single intragastric gavage administration of RES (100 mg/kg) (Liang et al., 2013). The glucuronidation and sulfation of RES and polyphenols by liver and intestinal microsomes and isoforms were previously investigated (Aumont et al., 2001; Bode et al., 2013; Brill et al., 2006; Furimsky et al., 2008; Iwuchukwu and Nagar, 2008; Marvalin and Azerad, 2011; Miksits et al., 2005; Wenzel et al., 2005; Wenzel and Somoza, 2005). While there are variations in hepatic glucuronidation of RES between species (Maier-Salamon et al., 2011), in general the UDP-glucuronosyltransferase enzymes preferentially form R3G (Aumont et al., 2001). Several microbial-derived metabolites of RES such dihyroresveratrol, lunularin, and 3,4’-dihydroxy-stilbene also were also identified in vivo and in vitro (Azorín-Ortuño et al., 2011; Bode et al., 2013; Walle et al., 2004; Wang et al., 2005).

Fig 1.

RES and its monoconjugated metabolites trans-resveratrol-3-O-β-D-glucuronide (R3G), trans-resveratrol-4-O-β-D-glucuronide (R4’G), trans-resveratrol-3-sulfate (R3S), and trans-resveratrol-4’-sulfate (R4’S).

Bioavailability of free RES was reported to be independent of dose and ranged between 29 and 38% in rodents (Kapetanovic et al., 2011; Marier et al., 2002) and < 1% in humans (Almeida et al., 2009; Cottart et al., 2010). Reported plasma elimination half-life of RES in rodents ranged between 180 and 1440 min depending on the study. RES was measured in multiple tissues following oral administration in red wine (~26–28 μg RES) to rats (Bertelli et al., 1996a). Highest concentrations of RES in the liver, kidney and heart were reached within ~1–2 h. Wide tissue distribution was reported following oral administration of radiolabeled RES to mice (Vitrac et al., 2003) and rats (Abd El-Mohsen et al., 2006) with highest levels observed at ~1.5–3 h after administration.

The National Toxicology Program (NTP) generated TK data following oral exposure in Harlan Sprague Dawley (HSD) rats and B6C3F1/N mice, the same two rodent species used in NTP toxicology studies (https://ntp.niehs.nih.gov/testing/status/agents/ts-m010090.html) to put toxicological findings into context. The objective of this study was to generate TK data for RES and 2 of the 4 major metabolites, R3S and R3G, following oral exposure in the same two rodent strains, and to facilitate extrapolation to human exposures. The doses selected for rats (312.5, 625, and 1250 mg/kg) and mice (625, 1250, and 2500 mg/kg) are the same doses used in the NTP 2-year bioassay. Limited studies were also conducted following a single intravenous (IV) administration in rats and mice to generate data to estimate absolute oral bioavailability. A liquid chromatography-tandem mass spectrometry (LC-MS/MS) method was developed and validated to simultaneously quantitate the 3 analytes, RES, R3S, and R3G in plasma.

Materials and methods

Chemicals and reagents

RES (CASRN 501–36-0, lot no.210AI) was obtained from Bayville Chemical Supply Company (New York, NY). Identity was confirmed by infrared spectroscopy and proton and carbon-13 nuclear magnetic resonance spectroscopy. The purity determined by high performance liquid chromatography (HPLC) with time-of-flight (TOF) mass spectrometry and ultraviolet spectroscopy was >99%. (¹³C₆)RES and sodium salts of R3S (CASRN 858127–11-4) and R3G (CASRN 387372–17-0) were obtained from Toronto Research Chemicals Inc. (Ontario, Canada). All other reagents were obtained from commercial sources.

Analytical method

An analytical method using protein precipitation followed by liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) was validated and utilized to quantitate resveratrol, R3S, and R3G in HSD rat plasma and B6C3F1/N mouse plasma.

Samples were prepared as follows. To each well in a 96-well Impact™ protein precipitation plate (Phenomenex, Torrance, CA), 100 μL plasma (standards, QCs or matrix blanks) and 400 μL of 250 ng/mL IS solution was added (except blanks without IS). The plate was covered, vortexed and placed on a positive pressure manifold (~8 psi) for 2 min and eluate was collected in a collection plate. A pierceable cover was placed over the collection plate and samples were analyzed by LC-MS/MS as described below.

Samples were analyzed using Shimadzu Prominence LC (Kyoto, Japan) coupled to a Sciex Triple Quad 5500 (Toronto, Ontario Canada) mass spectrometer using electrospray ionization in the negative ion mode. A Phenomenex Kinetex-XB C18 column (150 × 4.6 mm, 5 μ) (Torrance, CA) was used for analyte separation; column temperature was maintained at 40°C. The mobile phases used were (A) 5 mM ammonium acetate/2% 2-propanol in water and (B) 2% 2-propanol in methanol with the following gradient, isocratic at 10% B from 0 to 4 min, at 4 min changed to 20% B, linear gradient from 20% B to 45% B from 4 to 5 min, isocratic at 45%B from 5 to 12 min, linear gradient from 45% B to 60% B from 12 to 17 min, linear gradient from 60% B to 10% B from 17 to 19 min, equilibrated for 1 min at 10% B. The flow rate was 500 μL/min. The mass spectrometer operating conditions were; ion spray voltage, −4500 V; source temperature, 550 °C; sheath and auxiliary gas pressures, 50 arbitrary units, collision energy, 30 eV. The transitions monitored were: m/z 227→185, 307→227, 403→227, 233→191 for RES, R3S, R3G, and (¹³C₆)RES, respectively. The retention times for RES, R3S, and R3G were approximately 11.5, 9.0, and 8.5 min, respectively.

Animals and maintenance

All studies were conducted in Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC)-accredited facilities and approved by the Battelle (Columbus, OH) Animal Care and Use Committee and were conducted in compliance with the Food and Drug Administration Good Laboratory Practice Regulation (21 CFR Part 58). Animal care was performed in accordance with the “Guide for the Care and Use of Laboratory Animals” (Council, 2011). Male and female Hsd:Sprague Dawley^® SD^® (Sprague Dawley) (HSD) rats were obtained from Harlan Laboratories (Dublin, VA). B6C3F1/N mice were obtained from Taconic Biosciences (Germantown, NY). Animals were housed in humidity (55%±15%) and temperature-controlled (22±2°C) rooms which were maintained on a 12-h light/dark cycle and had free access to NTP-2000 diet (irradiated pellets; Zeigler Bros., Gardners, PA) and city water (West Jefferson, OH). Animals were quarantined for at least 1 week before they were used in a study, and were randomized into dosing groups. At the time of dosing, the average weights of male and female rats were 365.7 and 227.4 g, respectively, and male and female mice were 30.6 and 24.0 g, respectively. Rats were 14–15 weeks and mice were 13 weeks of age at the time of dosing.

Formulation preparation and analysis

Oral formulations of RES were prepared in 0.5% aqueous methylcellulose. IV formulations of RES were prepared in Cremophor®:ethanol:0.9% saline (1:1:8, v:v:v). All formulations were analyzed by HPLC with ultraviolet detection using a validated analytical method (RE%, ≤±10%; RSD ≤5%., r > greater than 0.99) and found to be within 10% of target.

Study design and dose administration

Male and female HSD rats (21 animals/species/sex/dose group) and B6C3F1/N mice (39 animals/species/sex/dose group) were given a single gavage dose of RES at 312.5, 625 or 1250 mg/kg and 625, 1250 or 2500 mg/kg, respectively. The dosing volume was 5 mL/kg for rats and 10 mL/kg for mice. Male and female HSD rats (24 animals/species/sex/dose group) and B6C3F1/N mice (36 animals/species/sex/dose group) were given a single IV dose of 10 mg/kg RES. The dose was delivered via the lateral tail vein using a dosing volume of 2 mL/kg for rats and 4 mL/kg for mice. The actual volume administered was based on the body weights on the day of dosing.

After dosing, blood was collected under CO₂:O₂ (~70:30) anesthesia via retro-orbital bleeding in rats and closed-chest cardiac puncture in mice into tubes containing K₃-ethylenediaminetetraacetic acid (K₃-EDTA) from up to three animals per time point. Rats were typically sampled twice. Immediately following collection, tubes were mixed gently, and placed on wet ice. Target times for blood collection following gavage administration were as follows: 0 (pre dose), 5, 10, 20, 30, 45, 60, 120, 180, 240, 480, 720 and 1440 (rats only) min. Target times for blood collection following IV administration were as follows: 0 (pre dose), 5, 10, 20, 30, 45, 60, 90, 120, 180, 240, and 480 and 720 (rats only) min; Plasma was separated from the blood within 60 min of collection by centrifugation for 10 min and then frozen at −70 °C. After blood collection, euthanasia was performed via CO₂ asphyxiation.

TK analysis

All samples, except two, were collected within 5% of target times for timepoints ≤ 4 h and within 15 min of target times for timepoints between 4 and 18 h. Therefore, target timepoints were used for the analysis except for two instances where the actual times were used. All analyte concentrations above LOD were used in the TK analysis.

Phoenix WinNonlin (Version 6.4, Certara, Princeton, NJ) was used for the TK analysis. A high degree of animal to animal variability was observed, likely due to extensive enterohepatic recirculation. Therefore, the data were analyzed using a naïve average method (Gabrielsson and Weiner, 2016). This approach allowed for better fits of the compartmental models to the plasma concentration-time data. TK parameters from the compartmental models are presented along with the standard errors of the parameter estimations.

The following two- and one-compartment models were selected to fit mean plasma concentration versus time data in IV (1) and gavage (2) groups:

$$\mathrm{C}(\mathrm{t})=\mathrm{D}\mathrm{o}\mathrm{s}\mathrm{e}*(\mathrm{\alpha }-{\mathrm{k}}_{21})/{\mathrm{V}}_1*(\mathrm{\alpha }-\mathrm{\beta })*{\mathrm{e}}^{-\mathrm{\alpha }*\mathrm{t}}+\mathrm{D}\mathrm{o}\mathrm{s}\mathrm{e}*({\mathrm{k}}_{21}-\mathrm{\beta })/{\mathrm{V}}_1*(\mathrm{\alpha }-\mathrm{\beta })*{\mathrm{e}}^{-\mathrm{\beta }*\mathrm{t}}$$ (1)

$$\mathrm{C}(\mathrm{t})=\mathrm{D}\mathrm{o}\mathrm{s}\mathrm{e}*{\mathrm{K}}_{01}/\mathrm{V}*({\mathrm{K}}_{01}-{\mathrm{K}}_{10})*({\mathrm{e}}^{-{\mathrm{K}}_{10}\mathrm{t}}–{\mathrm{e}}^{-{\mathrm{K}}_{01}\mathrm{t}})$$ (2)

Absolute bioavailability following gavage administration was estimated as:

$$\mathrm{F}\mathrm{\%}=[\mathrm{A}\mathrm{U}{\mathrm{C}}_{\mathrm{\infty }(\mathrm{g}\mathrm{a}\mathrm{v}\mathrm{a}\mathrm{g}\mathrm{e})})/(\mathrm{D}\mathrm{o}\mathrm{s}{\mathrm{e}}_{(\mathrm{g}\mathrm{a}\mathrm{v}\mathrm{a}\mathrm{g}\mathrm{e})}]/[\mathrm{A}\mathrm{U}{\mathrm{C}}_{\mathrm{\infty }(\mathrm{I}\mathrm{V}))}/(\mathrm{D}\mathrm{o}\mathrm{s}{\mathrm{e}}_{(\mathrm{I}\mathrm{V})}]*100$$

A glossary of TK parameters used are given in Supplementary Table 1.

Results

Analytical method

A linear regression with 1/x weighing was used to relate peak area response ratio of analyte to IS. The concentration of RES, R3S, and R3G was calculated as ng/mL using their individual response ratios, the regression equation, and any dilution factors. The plasma standard curves were linear with r values greater than or equal to 0.99 for RES, R3S, and R3G in all accepted runs. All concentrations above the limit of detection (LOD) were reported. The lower limit of quantitation (LLOQ) for this assay was 5 ng/mL for all analytes and LOD was estimated as the 3 times the standard deviation of the LLOQ. Quality control (QC) samples at 3 different concentrations and plasma blanks were included in each analytical run. Analytical method validation and stability data in HSD rats and B6C3F1/N mice plasma are reported in Table 1.

Table 1.

Analytical method validation and stability data in HSD rat and B6C3F1/N mice plasma^a

	RES		R3S		R3G
	Rat	Mouse	Rat	Mouse	Rat	Mouse
Concentration Range (ng/mL)	5 to 1000	5 to1000	5 to 1000	5 to1000	5 to1000	5 to1000
Linearity (r)	≥ 0.99	≥ 0.99	≥ 0.99	≥ 0.99	≥ 0.99	≥ 0.99
LLOQ (ng/mL)b	5	5	5	5	5	5
LOD (ng/mL)c	0.315	0.483	0.639	0.402	2.33	0.939
Accuracy (%RE)d
Intra-day Inter-day	−0.4 to 13.1 3.3 to 5.3	5.9 to 6.9	0.4 to 8.6 −2.4 to 4.3	−4.3 to 10.2	−9.1 to 0.7 −5.7 to −2.4	−9.5 to 3.2
Precision (%RSD)d
Intra-day, Inter-day	1.4 to 8.7, 3.4 to 7.8	0.7 to 3.0	2.1 to 9.4, 3.1 to 2.6	1.0 to 2.0	1.2 to 10.5, 3.7 to 7.8	1.1 to 5.2
Absolute Recovery (%)	89.5 to 94.2	94.8 to 123.1	167.0 to 198.0	146.4 to 205.5	98.1 to 140.0	88.2 to 115.2
Dilution Verification (%RE, %RSD)	7.0, 6.2	1.0, 1.5	0.0, 7.1	−6.6, 1.1	6.9, 8.8	−3.0, 1.6
Extracted Sample Stability,(%RE)e
Ambient temperature (7d)	−0.8 to 6.9	−6.1 to 2.8	−12.2 to −5.2	−6.7 to 1.1	−23.2 to −18.1	0.6 to 3.2
Matrix Stability (%RE)
Freeze-Thawf Storage (113d) Storage (l38d)	−4.9 to −11.4, −5.3 to 2.0	−6.3 to 2.5, −1.9 to 2.8	3.1 to 4.3, −9.4 to −1.5	−0.3 to −2.9, −10.8 to −8.7	−4.7 to −10.9, 11.5 to 22.3	1.1 to 5.3, 10.2 to 13.8

^aFull validation was performed in HSD rat plasma and the method was partially validated to B6C3F1/N mouse plasma

^bLLOQ, Lower limit of quantitation; experimental LLOQ, is the lowest concentration used in standard curve for both HSD rat and B6C3F1/N mouse plasma

^cLOD, limit of detection was estimated as the 3 times the standard deviation of the LLOQ (n=6 replicates)

^dRE, relative error and RSD, relative standard deviation

^eValues are given the range for 3 QC concentrations at ∼ 6, 450, and 750 ng/mL for rat plasma and ∼7, 600, and 1000 ng/mL for mouse plasma, n=4

^fValues given are for 2 target QC concentrations at ∼6 and 1000 ng/mL in HSD rat plasma (113 d) and B6C3F1/N mouse plasma (138 d), 4 freeze-thaw cycles

Gavage administration of RES in HSD Rats

RES, R3S and R3G were measurable at all time points except pre-dose. RES concentration versus time profiles exhibited at least one secondary peak after the initial absorption phase. Similar to the parent compound, R3S and R3G exhibited more than one peak in their concentration versus time profiles. The plasma concentration versus time profiles were best described by a one-compartmental model with 1/Ŷ² weighting (Eq 2). Representative model fits to the concentration-time data for RES, R3S, and R3G in male rats following administration of 625 mg/kg RES are shown in Figure 2. Toxicokinetic parameters estimated for all groups are presented in Table 2 for RES and Table 3 for R3S and R3G.

Fig 2.

a) RES, b) R3S, and c) R3G plasma concentrations versus time profiles following a single gavage administration of 312.5, 625, and 1250 mg/kg RES in male HSD rats. Data were fitted by a one-compartment model with 1/Ŷ² weighing. N≤3 per time point.

Table 2.

Plasma toxicokinetic parameters of trans-resveratrol (RES) in male and female HSD rats following gavage administration of trans-resveratrol (RES) ^a,b

Parameter	312.5 mg/kg		625 mg/kg		1250 mg/kg
	Male	Female	Male	Female	Male	Female
Cmax_predicted (ng/mL)	902±187	493±122	1170±290	723±149	1450±360	1230±250
Cmax_predicted /Dose (ng/mL)/(mg/kg)	2.89±0.60	1.58±0.39	1.87±0.46	1.16±0.24	2.12±0.29	0.984±0.200
Tmax_observed (min)	90.0	240	180	480	480	360
Tmax_predicted (min)	58.9±23.3	49.9±27.4	97.5±36.2	46.0±22.2	182±56	263±55
k01 half-life (min)	11.8±6.8	8.30±6.33	22.8±14.0	7.50±4.96	33.4±18.4	60.0±28.6
k10 half-life (min)	334±93	502±238	366±143	495±192	1330±1740	1060±1050
Cl_F (mL/min/kg)	636±126	817±277	839±209	1140±320	406±426	561±402
V_F (mL/kg)	307000±79000	592000±171000	443000±152000	810000±193000	782000±272000	855000±299000
AUClast (min*ng/mL)	386000	288000	518000	473000	1330000	1010000
AUC∞_predicted (min*ng/mL)	491000±97000	382000±130000	745000±185000	551000±154000	3080000±3220000	2230000±1600000
AUC∞/Dose (min*ng/mL)/(mg/kg)	1570±310	1220±420	1190±300	882±246	2460±2580	1780±1280

^aBased on one-compartmental model with $1/{\hat{\mathrm{Y}}}^2$ weighting.

^bValues given are mean (standard error, SE).

Table 3.

Plasma toxicokinetic parameters of trans-resveratrol-3-sulfate (R3S) and resveratrol-3-O-β-D-glucuronide (R3G) in male and female HSD rats following gavage administration of trans-resveratrol (RES)

Parameter	312.5 mg/kg		625 mg/kg		1250 mg/kg
	Male	Female	Male	Female	Male	Female
R3S
Cmax_predicted (ng/mL)	4860±1270	6960±2390	6040±1310	12200±2800	15600±3700	17300±4800
Cmax_predicted /Dose (ng/mL)/(mg/kg)	15.6±4.1	22.3±7.6	9.66±2.10	19.50±4.4	12.5±3.0	13.8±3.8
Tmax_predicted (min)	194±55	183±72	245±55	260±59	431±80	392±99
k01 half-life (min)	0.00454	0.014	0.00712	0.00955	0.0027	0.00573
k10 half-life (min)	119±265	500±423	337±217	659±542	351±1730	820±1380
Cl_F (mL/min/kg)	121±33	48.3±22.6	128±29	41.1±19.4	67.6±37.9	44.0±43.2
V_F (mL/kg)	20800±43500	34900±20200	62500±36500	39100±17500	34200±152000	52000±40600
AUClast (min*ng/mL)	2820000	4320000	4860000	8370000	13800000	16700000
AUC∞_predicted (min*ng/mL)	2580000±700000	6470000±3030000	4870000±1100000	15200000±7200000	18500000±10300000	28400000±27900000
AUC∞/Dose (min*ng/mL)/(mg/kg)	8260±2240	20700±9700	7790±1760	24300±11500	14800±8200	22700±22300
R3G
Cmax_predicted (ng/mL)	30100±5800	28600±5000	31700±4700	60100±10200	64200±9600	67000±11400
Cmax_predicted /Dose (ng/mL)/(mg/kg)	96.3±18.6	91.5±16.0	50.7±7.5	96.2±16.3	51.4±7.7	53.8±9.1
Tmax_predicted (min)	145±35	124±30	146±27	173±34	338±50	248±43
k01 half-life (min)	0.0167	0.0283	0.0172	0.0108	0.00801	0.0108
k10 half-life (min)	348±116	721±364	379±104	265±82	1040±890	742±474
Cl_F (mL/min/kg)	15.5±2.9	9.33±3.29	27.6±4.2	17.3±2.8	10.4±6.2	13.8±5.5
V_F (mL/kg)	7780±2430	9710±2290	15100±3600	6610±2300	15500±4900	14800±4600
AUClast (min*ng/mL)	19900000	23000000	23000000	35600000	52900000	53000000
AUC∞_predicted (min*ng/mL)	20200000±3800000	33500000±11800000	22700000±3500000	36100000±5700000	120000000±71000000	90400000±35900000
AUC∞/Dose (min*ng/mL)/(mg/kg)	64600±12200	107000±38000	36300±5600	57800±9100	96000±56800	72300±28700

In rats, RES exhibited an absorption phase with T_{max_observed} ranging between 90 and 480 min. Based on the model fits, T_{max_predicted} was reached at ≤ 263 min (Table 2). T_{max_predicted} and absorption half-life increased with dose, but the differences did not appear to be biologically relevant. The apparent volume of distribution values for male and female rats, ranging from 307000 to 855000 mL/kg far exceeded the reported body water volume in the rat (688 mL/kg) (Davies and Morris, 1993), indicating the distribution of RES into a peripheral compartment. In addition, these volume of distribution values increased as the dose was increased. Plasma elimination half-life (k₁₀_Half-life) was markedly increased for the 1250 mg/kg dose, with 334, 366 and 1330 min observed for 312.5, 625, and 1250 mg/kg, respectively, in males, and 502, 495, and 1060 min, respectively, in females. In general, C_{max_predicted} and AUC_∞ increased proportionally to the dose as seen with the similar values for dose normalized C_{max_predicted} and AUC_∞ (Table 2). While there were some apparent differences in RES TK parameters between male and female rats, e.g., faster clearance and larger volume of distribution in females, overall, gender did not impact the TK parameters for RES (Table 2).

In rats, RES was metabolized to R3S and R3G; T_max was reached ≤ 431 min for R3S and ≤ 338 min for R3G with a dose-dependent increase at the highest dose. In general, plasma elimination half-life increased with the dose similar to RES with the longest elimination half-life values for both R3S and R3G occurring after a 1250 mg/kg dose. Clearance was slower for the metabolites than for RES while the volume of distribution for the metabolites was smaller than that of RES. C_{max-predicted} and AUC_∞ increased with the dose for both R3S and R3G. Dose normalized C_{max-predicted} and AUC_∞ showed that systemic exposure was approximately dose proportional but with greater variability for R3G. There did not appear to be any gender effects on the TK parameters for either R3S or R3G (Table 3).

IV administration of RES in HSD Rats

Plasma concentrations versus time profiles for RES following IV administration of 10 mg/kg in male and female rats were biphasic and were best described by a two-compartmental model with first order elimination and 1/Ŷ² weighting (Eq 1). A representative model fit to the concentration-time data is given for male rats in Supplementary Fig 1 and parameters estimated are presented in Supplementary Table 2. The elimination half-life was 6.21 and 7.26 min for males and females, respectively. The systemic exposure parameters, C_{max_observed}, C_{max_predicted} and AUC_∞ for RES were higher in males (5960 ng/mL, 11600 ng/mL, and 104000 min*ng/mL, respectively) than in females (3870 ng/mL, 5560 ng/mL, and 58300 min*ng/mL, respectively).

The metabolites, R3S and R3G, were eliminated rapidly with a half-life ≤ 24.6 min (Supplementary Table 3). For R3S, AUC_∞ in females (360000 min*ng/mL) was greater than males (216000 min*ng/mL). Likewise, for R3G, AUC_∞ in females (1650000 min*ng/mL) was also higher than in males (941000 min*ng/mL). When comparing systemic exposure for the three analytes, C_{max_predicted} and AUC_∞ were greater for R3S and R3G than for RES for both males and females.

Gavage administration of RES in B6C3F1/N mice

RES, R3S, and R3G were measurable at all time points except pre-dose. At least one secondary peak ~500 min after initial absorption phase was observed in both sexes. The plasma concentration versus time data profiles were best described by a one-compartmental model with 1/Ŷ² weighting (Eq 2). Figs 3A, B, and C show the model fits to the concentration-time data for RES, R3S, and R3G in male mice following administration of 625 mg/kg RES. Toxicokinetic parameters estimated from these data are presented in Table 4.

Fig 3.

a) RES, b) R3S, and c) R3G plasma concentrations versus time profiles following a single gavage administration of 625, 1250 and 2500 mg/kg RES in male B6C3F1/N mice. Data were fitted by a one-compartment model with 1/Ŷ² weighing. N≤3 per time point.

Table 4.

Plasma toxicokinetic parameters of trans-resveratrol (RES) in male and female B6C3F1/N mice following gavage administration of trans-resveratrol (RES)

Parameter	625 mg/kg		1250 mg/kg		2500 mg/kg
	Male	Female	Male	Female	Male	Female
Cmax_predicted (ng/mL)	1320±630	1040±280	1850±770	1020±420	1890±530	1820±440
Cmax_predicted Dose (ng/mL)/(mg/kg)	2.11±1.01	1.66±0.45	1.48±0.62	0.816±0.336	0.756±0.212	0.728±0.176
Tmax_observed (min)	5.00	120	480	360	120	360
Tmax_predicted (min)	60.8±45.8	36.1±22.7	101±51	69.2±48.1	121±40	174±45
k01 half-life (min)	153±76	7.08±6.08	40.4±47.6	11.8±12.7	29.3±19.6	117±1410
k10 half-life (min)	17.1±21.8	217±72	137±84	642±915	424±321	124±1480
Cl_F (mL/min/kg)	1630±690	1710±410	2060±820	1230±1330	1780±780	2910±630
V_F (mL/kg)	40000±45000	535000±176000	406000±348000	1140000±600000	1090000±480000	521000±6210000
AUClast (min*ng/mL)	376000	407000	513000	520000	854000	824000
AUC∞_predicted (min*ng/mL)	384000±163000	366000±88000	607000±242000	1010000±1090000	1410000±620000	860000±187000
AUC∞/Dose (min*ng/mL)/(mg/kg)	614±261	586±1441	486±194	808±872	564±248	344±75

In mice, RES exhibited an absorption phase with Tmax_observed ranging between 5 and 480 min. Based on the model fits, T_{max_predicted} was reached at ≤ 174 min. The T_{max_predicted} increased with increasing dose for both males and females while absorption half-life increased with dose for just the females. Conversely, k₀₁_half-life increased with dose for the males from 17.1 to 424 min, but varied for females going from 217 to 642 to 124 min as the dose was increased. The volume of distribution for males and females, ranging from 40000 to 1140000 mL/kg, far exceeded the reported aqueous body water volumes in mice (725 mL/kg) (Davies and Morris, 1993), indicating the distribution of RES into a peripheral compartment. Dose-normalized C_{max_predicted} values decreased with increasing dose while dose normalized AUC_∞ suggested a near dose proportional increase in value. Taken as a whole, several of the TK parameters for RES in mice presented non-linear kinetics over the dose range (Table 4).

R3S and R3G formed from RES with T_max reached at ≤ 156 and 180 min, respectively. The elimination half-life for both R3S (< 200 min) and R3G (< 499 min) was rapid with no apparent differences between dose groups. In general, elimination half-life for R3S was shorter than RES, but for R3G was similar to RES. Both C_{max_predicted} and AUC_∞ increased with increasing dose for both R3S and R3G. TK parameters for R3S and R3G were typically similar (within 2-fold) between males and females suggesting no apparent sex-dependent differences (Table 5).

Table 5.

Plasma toxicokinetic parameters of trans-resveratrol-3-sulfate (R3S) and resveratrol-3-O-β-D-glucuronide (R3G) in male and female B6C3F1/N mice following gavage administration of trans-resveratrol (RES)

Parameter	625 mg/kg		1250 mg/kg		2500 mg/kg
	Male	Female	Male	Female	Male	Female
R3S
Cmax_predicted (ng/mL)	9660±2560	9160±2230	5250±1570	9560±3190	6470±1020	9770±1860
Cmax_predicted/Dose (ng/mL)/(mg/kg)	15.5±4.1	14.7±3.6	4.20±1.26	7.65±2.55	2.59±0.41	3.91±0.74
Tmax_predicted (min)	101±30	55.6±22.8	108±40	123±48	156±28	150±31
k01 half-life (min)	0.114	0.0497	0.0184	0.0158	0.00652	0.00717
k10 half-life (min)	95.7±51.8	177±48	182±89	200±128	110±1510	112±335
Cl_F (mL/min/kg)	226±65	215±45	601±157	296±86	913±139	625±113
V_F (mL/kg)	31200±23200	54900±18600	158000±91000	85300±58700	144000±2000000	101000±310000
AUClast (min*ng/mL)	1410000	2930000	1810000	3160000	2630000	4020000
AUC∞_predicted (min*ng/mL)	2770000±790000	2910000±610000	2080000±540000	4220000±1220000	2740000±420000	4000000±720000
AUC∞/Dose (min*ng/mL)/(mg/kg)	4430±1260	4660±980	1660±430	3390±980	1100±170	1600±290
R3G
Cmax_predicted (ng/mL)	24000±4700	32100±7800	38100±10700	30000±6500	56300±14400	76600±18600
Cmax_predicted /Dose (ng/mL)/(mg/kg)	38.4±7.5	51.4±12.5	30.5±8.6	24.0±5.2	22.5±5.8	30.6±7.4
Tmax_predicted (min)	28.3±17.1	49.7±22.5	105±37	60.6±23.3	160±42	180±43
k01 half-life (min)	0.172	0.0667	0.0251	0.0637	0.013	0.00998
k10 half-life (min)	499±255	248±86	302±170	472±263	290±203	258±205
Cl_F (mL/min/kg)	34.7±13.6	47.3±11.0	59.2±17.5	56.0±22.0	72.5±20.7	54.2±14.3
V_F (mL/kg)	25000±5500	16900±5300	25800±11600	38100±10400	30300±17000	20100±13500
AUClast (min*ng/mL)	11100000	12900000	16500000	14500000	33200000	31200000
AUC∞_predicted (min*ng/mL)	18000000±7000000	13200000±3100000	21100000±6200000	22300000±8800000	34500000±9900000	46200000±12200000
AUC∞/Dose (min*ng/mL)/(mg/kg)	28800±11200	21100±5000	16900±5000	17800±7000	13800±4000	18500±4900

IV administration of RES in B6C3F1/N mice

RES was administered to groups of male and female mice at 10 mg/kg. Plasma concentrations versus time profiles for RES were biphasic for both sexes and were best described by a two-compartmental model with first order elimination and 1/Ŷ² weighting (Eq 1). A representative model fit to the concentration-time data is given for male mice in Supplementary Fig 2 and toxicokinetic parameters estimated are presented in Supplementary Table 2.

Following administration of RES, there were no sex differences in distribution and elimination of RES, and all TK parameters were within 2-fold between males and females. The elimination half-life was 29.1 and 21.7 min for males and females, respectively. The volume of distribution exceeded the reported aqueous body water volumes in mice (725 mL/kg) (Davies and Morris, 1993) for both sexes, indicating distribution of RES into the extravascular tissues. There were no apparent differences in the elimination and clearance of RES between sexes. C_{max_predicted} (4410 and 4420 ng/mL) and AUC_∞ (185000 and 138000 ng/mL*min) values were similar between males and females further suggesting no sex effects on the TK parameters for RES in mice.

While the metabolite R3S half-life values were similar to RES, R3G had greater elimination half-life than RES. For R3S, AUC_∞ in females (359000 min*ng/mL) was greater than males (105000 min*ng/mL). Conversely, for R3G, AUC_∞ in males (4530000 min*ng/mL) was higher than in females (320000 min*ng/mL) (Supplementary Table 4). When comparing systemic exposure for the three analytes, C_{max_predicted} and AUC_∞ were greater for R3G than for RES for both males and females but R3S C_{max_predicted} and AUC_∞ were only greater than RES in females. There were no apparent sex differences for systemic exposure of R3G, but systemic exposure was greater in females than males for R3S.

Bioavailability

The absolute bioavailability for RES was ∼ 15, 12, and 24 for male rats and ~ 21, 15 and 31% for female rats at 312.5, 625, and 1250 mg/kg, respectively. The absolute bioavailability for RES was ~ 3.3, 2.6, and 3 for male mice and ~ 4, 6, and 3% for female mice at 625, 1250, and 2500 mg/kg, respectively. Additionally, bioavailibity for R3S and R3G and total bioavality were estimated in both rats and mice and the data is presented in Table 6.

Table 6.

Estimated bioavalibity of trans-resveratrol (RES), trans-resveratrol-3-sulfate (R3S) and resveratrol-3-O-β-D-glucuronide (R3G) in male and female HSD rats and B6C3F1/N mice following gavage administration of RES

	RES (%)		R3S (%)		R3G (%)
	Male	Female	Male	Female	Male	Female
Rats
312.5 mg/kg	15.1	21.0	38.2	57.5	68.7	65.0
625 mg/kg	11.5	15.1	36.1	67.6	38.6	35.0
1250 mg/kg	23.7	30.6	68.5	63.1	102.0	43.8
Mice
625 mg/kg	3.3	4.2	42.2	13.0	63.6	66.0
1250 mg/kg	2.6	5.9	15.8	9.4	37.3	55.8
2500 mg/kg	3.0	2.5	10.4	4.5	30.5	57.8

Discussion

While RES has long been identified for its health benefits, there are few comprehensive studies investigating the safety or toxicity of RES. Therefore, NTP evaluated the toxicity of RES in rodent models. To aid in interpretation of toxicity data, we have conducted TK studies in the two rodent species using the same doses used in NTP toxicity and carcinogenicity studies (https://ntp.niehs.nih.gov/testing/status/agents/ts-m010090.html).

Following a single gavage administration of RES in HSD rats, the absorption was very rapid. Of particular note, both the T_{max_predicted} and absorption half-life typically had a general trend of increasing as the dose was increased in both sexes. This presented the first indication of a lack of dose linearity for several of the TK parameters for RES. The presence of multiple peaks in the plasma concentration-time curves indicated enterohepatic recirculation similar to previous studies in animals and humans evaluating RES absorption and metabolism (Boocock et al., 2007; Marier et al., 2002; Sharan et al., 2012; Walle et al., 2004). The lack of dose linearity as shown in rats with longer elimination half-lives associated with the highest administered doses. Following both IV and gavage administration high volumes of distribution of RES were observed in rats. This high volume of distribution is likely due to high tissue distribution and enterohepatic recirculation.

C_max and AUC_∞ increased with increasing dose and were near dose proportional, although dose normalized C_max values did trend downwards with an increase in dose. These, combined with the lack of dose linearity seen in other TK parameters, along with previously reported work that the absorption of RES does not increase with dose but percent excreted in urine increases with increased dose (Wenzel and Somoza, 2005), suggests a possible saturation of the metabolism or elimination. An overall comparison of the TK parameters for male and female rats showed no apparent sex effect on RES.

While other minor metabolites of RES have also been identified (e.g. R4’S and R4’G) (Baur and Sinclair, 2006; Calamini et al., 2010; Hoshino et al., 2010; Polycarpou et al., 2013), the major metabolites of RES in plasma have been consistently reported to be glucuronide (R3G) and sulfate (R3S) conjugates (Boocock et al., 2007; Sharan et al., 2012; Walle et al., 2004; Yu et al., 2002). In this study, RES was rapidly metabolized to R3S and R3G in rats with T_max of ~60–420 min. Following single gavage administration of RES, plasma concentrations of R3G and R3S were much higher than RES (R3G>>R3S>RES), which agrees with previous reports of lower plasma levels of RES in humans (Walle et al., 2004) and animals in comparison to the major metabolites (Boocock et al., 2007; Wenzel and Somoza, 2005). Consistently higher plasma RES metabolite concentrations compared to RES plasma concentrations supports a rapid and extensive metabolism (Crowell et al., 2007; Goldberg et al., 2003; Marier et al., 2002; Meng et al., 2004). A similar trend was observed following a single IV administration in rats. While R3S and R3G half-life values were similar to RES, C_{max_predicted} and AUC_∞ were greater than RES. Systemic exposure to R3G and R3S was higher than RES in rats based on the higher plasma concentrations and AUC values at all exposure concentrations. Additionally, both R3S and R3G increased less than proportionally to the dose with no apparent sex difference.

Since several other polyphenols exhibit relatively low bioavailability (Lewandowska et al., 2013; Scalbert and Williamson, 2000), it is not surprising that RES also demonstrates low oral bioavailability. RES undergoes extensive first-pass glucuronidation and sulfonation and is likely to be absorbed in the form of R3G and R3S in the small intestine, which plays an important role in pre-systemic glucuronidation of RES. Oral bioavailability of RES in animals and humans was reported to be very low (Marier et al., 2002; Walle et al., 2004) due to rapid and extensive metabolism resulting in low concentrations of RES in systemic circulation. In this study, the bioavailability of RES was also low and fairly consistent for rats and mice. The absolute bioavailability for rats was ∼ 15, 12, and 24 for males and ~19, 15 and 31 % for females at 312.5, 625, and 1250 mg/kg, respectively. While bioavailability for female rats seems to be slightly higher than male rats, an increase in bioavailability at 1250 mg/kg dose group was observed for both sexes. Increases in the bioavailability of RES with increasing dose might be related to the saturation of metabolism (Kaldas et al., 2003). Sex differences in the glucuronidation of resveratrol in human liver microsomes has been previously reported (Dellinger et al., 2013), and may explain the slight increase of bioavailability in female rats.

Similar to rats, absorption following single gavage administration of RES in mice was very rapid and at least one secondary peak was observed after the initial inclined phase. Both mice and rats had similar T_max values (≤240 min). In mice, both the T_{max_predicted} and absorption half-life typically had a general trend of increasing as the dose was increased, indicating a lack of dose linearity in mice similar to that in rats. C_max and AUC_∞ increased with increasing dose, but in less than dose proportional manner for C_max. Similar to IV administration, a relatively high volumes of distribution of RES was observed in mice, as was also observed in rats following gavage administration. For mice, RES was metabolized to R3S and R3G even more rapidly than rats, with T_max ~30–180 min. As was seen in rats, R3S and R3G half-life values for mice were similar to RES, C_{max_predicted} and AUC_∞ were generally greater than RES. Systemic exposure to R3G and R3S was higher than RES in both species based on the higher plasma concentrations and AUC values at all exposure concentrations. Additionally, both systemic exposure of R3S and R3G increased less than proportionally to the dose with no apparent sex difference. Bioavailability for mice was similar regardless of sex and dose. Overall, mice had lower bioavailability compared to rats. Lower bioavailability in mice is likely due to faster metabolism and clearance leading to lower AUC levels. Differences in absolute bioavailability calculated relative to IV and differences in systemic exposure and TK parameters following IV administration is likely due to the differences in metabolism between species (Furukawa et al., 2014; Maier-Salamon et al., 2011; Shiratani et al., 2008; Wang and Sang, 2018)

A comparison of dose-normalized systemic exposure parameters C_max and AUC_∞ between rodent studies, including the current study, is presented in Table 7. For RES, dose normalized AUC values were somewhat similar, ranging from 13.5 to 53.3 (min*ng/mL)/(mg/kg) for all rat species, but dose normalized C_max values differed greatly between studies, from as little as 0.27×10⁻³ to as high as 29.98 (ng/mL)/(mg/kg). For R3S and R3G, differences were more prevalent between and within the studies. The reasons for these differences are unclear but potentially reflects a combination of species, dose, age of animals used, or enterohepatic recirculation which may be occurring at different rates in individual animals (Marier et al., 2002).

Table 7.

Comparison of Dose-Normalized C_max and AUC of trans-resveratrol (RES), trans-resveratrol-3-sulfate (R3S) and resveratrol-3-O-β-D-glucuronide (R3G)Following Single Dose Gavage Administration in Rats and Mice

Rodents	Dose (mg/kg)	RES		R3S		R3G
		Cmax/Dose (ng/mL)/(mg/kg)	AUC∞/Dose (min*ng/mL)/(mg/kg)	Cmax/Dose (ng/mL)/(mg/kg)	AUC∞/Dose (ng/mL)/(mg/kg)	Cmax/Dose (ng/mL)/(mg/kg)	AUC∞/Dose (ng/mL)/(mg/kg)
HSD Ratsa	312.5	2.89	26.2	15.6	138	96.3	1080
	625	1.87	19.8	9.66	130	50.7	605
	1250	1.16	41.0	12.5	247	51.4	1600
B6C3F1/Na	625	2.11	10.2	15.5	73.8	38.4	480
	1250	1.48	8.10	4.20	27.7	30.5	282
	2500	0.756	9.40	2.59	18.3	22.5	230
SD Ratsb	50	29.98	32.4	NA	NA	850.9	2626.2
CD Ratsc	50	1.5	27	4.4	252	4.6	232.0
	150	5.6	17.2	58.3	306.7	24.9	151.3
SD Ratsd	100	NA	53.3	NA	NA	NA	NA

^aCurrent study, AUC values were converted from min*ng/mL to hr*mg/mL for comparison.

^bMarier et al. 2002 (Marier, Vachon et al. 2002), C_max values of 6.57 and 105.2 μmol/L were converted to ng/mL and AUC_∞ values of 7.1 and 324.7 hr*μmol/L were converted to hr*ng/mL using MW of 228.25 and 404.4 g/mol for RES and R3G respectively.

^cKapetanovic 2011(Kapetanovic, Muzzio et al. 2011), C_maxvalues at 50 mg/kg were 76.7, 2020, and 2290 ng/mL and at 150 mg/kg were 847, 8710, and 3740 ng/mL for RES, R3S and R3G, respectively. AUC_∞ values at 50 mg/kg were 1350, 12600, and 11600 ng/mL and at 150 mg/kg were 2580, 46000, and 22700 hr*ng/mL for RES, R3S and R3G, respectively.

^dLiang et al 2013(Liang, Liu et al. 2013), AUC_∞ value of 320 min*mg/L was converted to 5333.3 hr*ng/L.

NA Not applicable

We compared the external dose (mg/kg and mg/m²) and systemic exposure parameters (C_max and AUC) of RES in our study to those reported in humans (Boocock et al., 2007) following a single oral administration of 2.5 or 5.0 g of RES (41.6 and 83.3 mg/kg (based on an average weight of 60 kg) or 1539.2 and 3082.1 mg/m² (using a K_m value of 37 kg/m² for a 60 kg human) (Reagan-Shaw et al., 2008) (Table 8). Rat doses of 312.5 and 625 mg/kg and mouse doses of 625 and 1250 mg/kg translate to 1875 and 3750 mg/m², assuming K_m values of 6 and 3 kg/m² for rats and mice, respectively. Using either the mg/kg or mg/m² human-equivalent dose, rat doses of 312.5 and 625 mg/kg were 7.5-fold and mouse doses of 625 and 1250 mg/kg were 15.0-fold higher than human doses of 41.6 and 83.3 mg/kg, respectively. Based on C_max, rat to human exposure multiples were <4.9 for RES. Rat to human exposures for AUC ranged between ~0.9–9.9, for RES (Table 8). The external dose and systemic exposure parameter comparisons suggested that systemic exposure to RES following single oral doses of 312.5, 625, or 1250 mg/kg doses in male HSD rats and 625, 1250, or 2500 mg/kg doses in male B6C3F1/N mice is similar to that following a single oral 2.5 and 5 g dose in humans.

Table 8.

Comparison Between Current Study and Humans

Current Study Doses mg/kg (mg/m2)	Human Doses mg/kg (mg/m2) c	Exposure multiple (rat/human)
		mg/kg	mg/m2	Cmax	AUC
				RES	RES
HSD Rats
312.5 (1875)a	41.6 (1539.2)	7.5	1.2	3.4	10.4
625 (3750)a	83.3 (3082.1)	7.5	1.2	2.2	9.4
B6C3F1/N mice
625 (1875) b	41.6 (1539.2)	15.0	1.2	4.9	8.1
1250 (3750)b	83.3 (3082.1)	15.0	1.2	3.4	7.7

^aCurrent study male HSD rats, Cmax (ng/mL) and AUC (ng*h/mL) values are presented in Table 2.

^bCurrent study male B6C3F1/N mice, Cmax (ng/mL) and AUC (ng*h/mL) values are presented in Table 4.

^cBoocock et al 2007 (Boocock, Faust et al. 2007), following C_max (ng/mL) and AUC_∞ (ng*h/mL) values reported following 2.5 and 5.0 g of RES administration were used, Cmax (2.5 and 5.0 g, respectively) 268.0 and 538.8 for RES; AUC (2.5 and 5.0 g, respectively) 786.5 and 1319 for RES, mg/kg doses were calculated based on 60 kg human body weight.

Conclusions

Our data demonstrated that RES is absorbed and metabolized to R3S and R3G in both species. Similar to the reported literature, enterohepatic recirculation of RES was also observed in both species with secondary increases of plasma concentrations ~640 min following oral administration. Comparison of data from rodents and humans administered a single oral dose of RES demonstrates that systemic exposure as measured by C_max and AUC in both rats (312.5, 625, and 1250 mg/kg) and mice (625, 1250, and 2500 mg/kg) was similar to systemic exposure determined in humans (2.5 and 5 g). Systemic exposure of R3S and R3G were typically greater than that of RES in both species. The lower systemic exposure of the parent compound combined with the rapid metabolism to metabolites was further evidenced by the low bioavailability of RES in both rats and mice. The bioavailability of RES was low, more so for mice than rats, and fairly consistent within the species (~12–31% and 2–6% for rats and mice, respectively).