Determination of human F_aF_g of polyphenols using allometric scaling

Abstract

Certain polyphenols exhibit low permeability; precise prediction of their intestinal absorption is important for understanding internal exposure in humans. Intestinal availability, which represents the fraction of administered compounds that reach the portal blood (F_aF_g), is calculated by dividing bioavailability (F) by hepatic availability (F_h), and F is obtained from pharmacokinetic data from both intravenous (i.v.) and oral (p.o.) administration. However, human F_aF_g of polyphenols is hardly reported, as human i.v. data are extremely scarce. In this study, we developed an estimation method for F_aF_g of polyphenols in humans based on the extrapolation of rat clearance using allometric scaling (allometric scaling-based F_aF_g calculation method, AS- F_aF_gCM). First, for quercetin, for which human i.v. data have been reported, we compared the F_aF_g obtained by AS-F_aF_gCM with the traditional approach using human i.v. and p.o. data. Less than two-fold difference in FaFg values was observed between the two approaches. Next, we obtained F_aF_g of structurally diverse polyphenols (genistein, baicalein, resveratrol, and epicatechin) using AS-F_aF_gCM, demonstrating that all of them were poorly absorbable. Furthermore, to utilize the pharmacokinetic data of the total concentration, including aglycones and metabolites, we modified the AS-F_aF_gCM to focus on their excretion. The F_aF_g value of naringenin was obtained using modified AS-F_aF_gCM and was nearly equal to that of baicalein, a structural isomer of naringenin. This study provides quantitative information on the intestinal absorption of polyphenols using comprehensive estimation methods.

INTRODUCTION

Polyphenols are naturally occurring organic compounds present abundantly in plants, including fruits, vegetables, and herbs. They comprise a wide family of molecules containing one or more phenolic rings. Accumulating evidence has highlighted the beneficial effects of polyphenols on human health, including their antioxidant, anti-obesity, anti-inflammatory, and anti-carcinogenic properties (Joseph et al., 2016; Singh et al., 2020; Bonfili et al., 2008). Recently, growing awareness regarding the importance of health and wellness among individuals has compelled the interest in applications of polyphenols in functional foods and dietary supplements (de Araújo et al., 2021). Despite their beneficial effects, certain polyphenols at excessive dosage have been found to cause side effects, including hepatotoxicity (Scalbert et al., 2005; Mennen et al., 2005; Seeff et al., 2015). Therefore, extensive research is required to examine their internal exposure to determine the optimal dosage for effective and safe use.

The concentration and duration of bioactive compounds in the circulation and tissues depend on their absorption, distribution, metabolism, and excretion (ADME) (Selick et al., 2002). Generally, phase II metabolic enzymes and transporters are considered crucial for the ADME of polyphenols (Bohn, 2014; Yang et al., 2008). Numerous pharmacokinetic studies have revealed low plasma concentrations of polyphenol aglycones following oral ingestion in humans and animals, although their metabolites have been well observed in the plasma (Manach et al., 2005; Scalbert et al., 2002). Recent studies have made great progress in improving our understanding of polyphenol behaviors related to ADME. One of the major factors contributing to their poor oral bioavailability is their low intestinal absorption due to low solubility, low permeability, efflux transport, and extensive gut metabolism (Truzzi et al., 2021). Therefore, the intestinal absorption of certain polyphenols could be a key rate-limiting process in polyphenol kinetics. However, the quantitative evaluation of polyphenol absorption in humans remains to be carried out because of the analytical difficulties in the human gastrointestinal lumen and portal vein blood, as well as ethical issues.

To quantitatively determine the contribution of intestinal absorption to absolute bioavailability (F), fractions of the administered compounds that reached the portal blood (F_aF_g) were evaluated. In general, F_aF_g is calculated by dividing F by hepatic availability (F_h), and F is obtained from pharmacokinetic data of both intravenous (i.v.) and oral (p.o.) administration (Kato, 2008; Shibata et al., 2003). Human i.v. drug data are well obtained in the process of drug application and are based on the balance of risks and benefits. However, human i.v. studies of polyphenols are rarely implemented if they are not for medical use (Hu et al., 2017), creating a situation in which the F_aF_g values of polyphenols are currently scarcely available.

To predict intestinal absorption, pharmacokinetic models like Advanced Compartmental Absorption and Transit model (Agoram et al., 2001), Q_Gut model (Gertz et al., 2010), and translocation model (Ando et al., 2015) were used. These sophisticated models have the potential to precisely predict the intestinal absorption of drugs; however, multiple scaling factors such as kinetic parameters of metabolic enzymes or transporters (Ando et al., 2015) are required to address the gaps between in vitro and in vivo data. Considering that the absorption mechanism of polyphenols is not well elucidated as that of drugs, it is practically difficult to apply these pharmacokinetic models for estimating the F_aF_g of polyphenols.

Over the years, interspecies allometric scaling methods have been developed to predict pharmacokinetic parameters (Dedrick, 1973). These equations relate body weight to specific pharmacokinetic parameters, which were originally derived from empirical observations relating to anatomical, physiological, and biochemical similarities across species (Choi et al., 2019). A simple allometric equation can predict the human total clearance (CL_tot) of drugs from data obtained in rats (Chiou et al., 1998). This study exhibited good correlation coefficient between humans and rats using 54 extensively metabolized drugs (Chiou et al., 1998), whose pharmacokinetic features were similar to those of polyphenols (Di Lorenzo et al., 2021). One of the pharmacokinetic parameters used to calculate F is CL_tot, which is generally calculated from i.v. data. Together, it was expected that human F_aF_g of polyphenols could be predicted without human i.v. data using the F value calculated from human CL_tot estimated by allometric scaling.

In this study, we propose a practical method to estimate the F_aF_g of polyphenols by extrapolating rat i.v. data to human pharmacokinetic parameters (allometric scaling-based F_aF_g calculation methods, AS-F_aF_gCM). Using quercetin, for which human i.v. data are reported, we confirmed the validity of AS-F_aF_gCM by comparing the F_aF_g values obtained by AS-F_aF_gCM with the traditional approach using human i.v. data. Moreover, to utilize a broader range of existing data on polyphenols, we next focused on the pharmacokinetic data of the total concentration, including both aglycons and their metabolites. We further modified AS-F_aF_gCM by focusing on its excretion (allometric scaling-based F_aF_g calculation method focusing on excretion, AS-F_aF_gCME). Collectively, this study provides promising information on the intestinal absorption of polyphenols using comprehensive calculation methods that can be implemented without human i.v. data.

MATERIALS AND METHODS

Overview of calculation methods (AS-F_aF_gCM and AS-F_aF_gCME)

AS-F_aF_gCM and AS-F_aF_gCME were used to estimate human F_aF_g using rat i.v. and human p.o. data, respectively (Fig. 1). AS-F_aF_gCM focused on the metabolism and excretion kinetics of polyphenol aglycones (Fig. 2A). In contrast, AS-F_aF_gCME focused on the excretion kinetics of polyphenol aglycone and its metabolites, which are mainly applied to the total blood concentration data, including both aglycone and its metabolites (Fig. 2B). AS-F_aF_gCME assumed that all metabolites produced in the enterocytes were excreted into the lumen and were not absorbed into the portal vein. This assumption could be applied to baicalein and its structural analogues, based on the following findings:

Fig. 1

Procedure for human F_aF_g estimation using AS-F_aF_gCM and AS-F_aF_gCME. A diagram describing the procedure of AS-F_aF_gCM for estimating human F_aF_g. (1) Rat i.v. data were collected, and total rat clearance from the i.v. study (rCL_tot,iv) was estimated using Eqs. 1 and 7 for AS-F_aF_gCM and AS-F_aF_gCME, respectively. (2) The human total clearance (hCL_tot,iv) was extrapolated using the allometric approach in Eqs. 2 and 8 for AS-F_aF_gCM and AS-F_aF_gCME, respectively. (3) Human p.o. data were collected, and human F was estimated by Eqs. 3 for AS-F_aF_gCM and human F' was Eq. 9 for AS-F_aF_gCME. (4) Human F_aF_g was estimated from F and F_h using Eqs. 4 and 5 for AS-F_aF_gCM and from F' and F_h' using Eqs. 10 and 11 for AS-F_aF_gCME.

Fig. 2

Schematic diagram of the compartment models for AS-F_aF_gCM and AS-F_aF_gCME. The compartment models used for AS-F_aF_gCM (A) and AS-F_aF_gCME (B) are shown. (A) AS-F_aF_gCM focuses on the absorption, distribution, metabolism, and excretion (ADME) of aglycone. CL_h, CL_r, and clearance of aglycone (CL_g) reflect hepatic metabolism of aglycone, renal excretion of aglycone, and intestinal efflux transport and metabolism of aglycone, respectively. CL_h and CL_r were extrapolated from the rat data using an allometric approach. F_aF_g and F_h are the fractions of the administered compounds that reach the portal blood and the hepatic bioavailability of aglycone, respectively. (B) AS-F_aF_gCME focuses on the ADME of both aglycone and its metabolites; a single quotation (') of CL_h', CL_r', and CL_g' indicates the clearances of both aglycone and its metabolites. CL_h' and CL_r' were extrapolated from rat data using an allometric approach. CL_g' is based on the assumption that all intestinal metabolites are excreted into the gut lumen by efflux transporters. F_h' refers to hepatic bioavailability, focusing on the bile excretion of aglycone and its metabolites.

i) Almost all the glucuronated baicalein was excreted into the intestinal lumen by the intestinal mucosal cells through multidrug resistance associated protein 2 (MRP2) (Akao et al., 2004).

ii) Baicalein metabolites appeared only in the apical side, when baicalein was applied to the Caco-2 cell monolayer model (Zhang et al., 2007).

In this study, AS-F_aF_gCME was applied to naringenin, one of the structural analogues of baicalein.

Collection of the rat intravenous data

The pharmacokinetic data in which rats received i.v. polyphenol were collected by searching PubMed and Google Scholar for relevant publications.

Using data from previous studies, in which aglycone was administered and the blood concentration of the unchanged form was analyzed, were used for AS-F_aF_gCM. The data were excluded if the reliability of their plasma concentration-time profile was insufficient (e.g., blood sampling was performed for less than 30 min, stabilization of the sample was not performed).

The data from this study, in which aglycone was administered and the total blood concentration of aglycone and its metabolites were analyzed, were collected for AS-F_aF_gCME as the following examples:

i. The blood samples were deconjugated and the aglycone concentration was quantified.

ii. Blood concentrations of aglycone and its metabolites were quantified using reference standards.

iii. The total blood concentration was examined using radio isotope-labelled compound.

Collection of the human oral data

Pharmacokinetic data in which polyphenols were orally administered to humans were collected by searching PubMed and Google Scholar for relevant publications.

The human p.o. reference data used in this study were obtained from studies in which aglycone was administered and the blood concentration of the unchanged form were analyzed. For F_aF_g calculation, the data were excluded from these studies if any of the following criteria were met: (i) repeated dose study, (ii) controlled absorption using technologies such as drug delivery systems, (iii) the chemical structure was changed before absorption (e.g., gut microbial metabolism and natural degradation), and (iv) the reliability of their plasma concentration-time profile was not sufficient (e.g., blood sampling was performed for less than 30 min, stabilization of the sample was not performed).

The procedure of AS-F_aF_gCM

AS-F_aF_gCM is the method to obtain human F_aF_g using the following procedure.

1. Calculation of rat total clearance (rCL_tot,iv)

2. Estimation of human total clearance (hCL_tot,iv) by allometric scaling

3. Estimation of human F

4. Estimation of human F_aF_g

The detailed procedure is described below.

1. Calculation of rCL_tot,iv

The RCL_tot,iv of polyphenols was estimated using the following equation:

where CL_tot,iv, Dose_iv, AUC_iv, are total clearance obtained using i.v. data, dose of an i.v. study, AUC of blood concentration versus time after dosage in an i.v. study, respectively. CL_tot,iv of polyphenols was calculated based on the “blood concentration”. The blood-to-plasma ratios (Rb) of quercetin and resveratrol were 0.68 and 0.78, respectively (Furukawa et al., 2012); the plasma concentration was converted to the blood concentration by multiplying Rb value. For the other compounds with unknown Rb, Rb was assumed to be 1. In cases where i.v. data (e.g., dose, area under the curve (AUC), and body weight) of a specific polyphenol were available in several reports, CL_tot,iv was estimated from each report, and the average value was used for the estimation of human clearance. In cases where several doses were examined in a report, the data of the minimum dose were adopted, considering the solubility, saturation of enzyme, transporter kinetics, and low intestinal absorption of polyphenols.

2. Estimation of hCL_tot,iv, by allometric scaling

Extrapolation of rat clearance to humans was performed by the extrapolation of clearance based on the relationship between pharmacokinetics and body weight of the animal.

The extrapolation of rCL_tot,iv to hCL_tot,iv was performed using the following allometric equation (Chiou et al., 1998):

where 0.66 is the allometric exponent and BW is body weight.

3. Estimation of human F

The human F value was obtained from estimated hCL_tot,iv, and human p.o. data. F was estimated from the following equation:

where F, hCL_tot,iv, AUC_po and Dose_po are absolute bioavailability, human total clearance obtained from i.v. data, AUC of blood concentration versus time after dosage in a p.o. study, and dose of a p.o. study, respectively. In cases where p.o. data (e.g., dose, AUC, and body weight) in several reports were available, human F was estimated from each report, and the average value was used. If several doses were examined in a previous study (Barnett et al., 2015; Boocock et al., 2007; Li et al., 2014), the data of the minimum dose were adopted, considering the solubility and saturation of enzyme and transporter kinetics. In studies where only the dose per body weight was described, the dose was calculated using the average body weight of the participants. In the instances where body weight data were not available in the study, the average body weight in the area where the study was performed (Walpole et al., 2012) was used for calculation.

4. Estimation of human F_aF_g

The F_aF_g value was estimated using the following equations (Kato, 2008):

where F_h, F_aF_g, hCL_h, and hQ_h are hepatic bioavailability, fraction of the administered compounds that reach the portal blood, human hepatic clearance, and human hepatic blood flow rate, respectively.

The value of hQ_h (96.6 L/h) was reported earlier (Kato et al., 2008). hCL_h was calculated as follows:

i. The contribution of human renal clearance (hCL_r) is less than 5% of hCL_tot,iv: hCL_h = ℎCL_tot,iv

ii. hCL_r is more than 5% of hCL_tot,iv: hCL_h = hCL_{tot, iv} − hCL_r

In cases where the value of hCL_r was available (resveratrol (Boocock et al., 2007) and quercetin (Moon et al., 2008), the values were used for the calculation. In contrast, in cases where only the urinary excretion data in a human p.o. study were available without hCL_r values (baicalein (Li et al., 2014)), hCL_r was calculated using the below equation:

where X_u,po is the total amount of excretion in urine. In cases where only renal clearance data in rats (rCL_r) were available (epicatechin (Yu, 2000)), the hCL_r value was estimated using the allometric equation (Eq. 2). In cases where both renal clearance data in rats and humans (genistein) were not available, hCL_{tot, iv} was regarded as ℎCL_h.

Traditional approach to estimate F_aF_g using human intravenous data for quercetin

The F value was estimated using Eq. 3, using the reported value of hCL_tot,iv in human i.v. studies. Human F_h and F_aF_g were estimated using Eqs. 4 and 5, considering the contribution of the renal clearance.

The procedure of AS-F_aF_gCME

AS-F_aF_gCME is the method to obtain human F_aF_g using the following procedure.

1. Estimation of rat CL_tot,iv'

2. Estimation of human CL_tot,iv' by allometric scaling

3. Estimation of human F^'

4. Estimation of human F_aF_g

The detailed procedure is described below.

1. Estimation of rat CL_tot,iv'

The rCL_tot,iv' of polyphenols was estimated using the following equation:

where the single quotation (') represents the value focusing on the excretion of both aglycone and its metabolites, and AUC_{iv, agri+metab} is AUC of total blood concentration of aglycone and its metabolites versus time after dosage in an i.v. study.

2. Estimation of hCL_tot,iv' by allometric scaling

The extrapolation of rCL_tot,iv' to hCL_tot,iv' was performed using the following allometric equation:

3. Estimation of human F'

The human absolute bioavailability, focusing on the excretion of aglycone and its metabolites (F'), was obtained by the equation below:

4. Estimation of human F_aF_g

The F_aF_g value was estimated using the following equations:

where a single quotation represents the values focusing on the excretion of both aglycone and its metabolites. The value of Qh (96.6 L/hr) was used as previously reported (Kato et al., 2008). hCL_h' was calculated using the equation below:

hCL_r' was estimated by the equation below:

where the single quotation represents the value focusing on the excretion of both aglycone and its metabolites, and X_{u,po,agri+metab} is the total amount of urinary excretion of aglycone and its metabolites.

Calculation of F, F_h, and F_aF_g in rats

If the data for oral administration in rats was available, F, F_h, and F_aF_g in rats were calculated, combined with the rat i.v. data. Collection of the rat oral data was performed in the same way described in Collection of the human oral data.

Rat F was calculated using the below equation:

where F, rCL_tot,iv, AUC_po, and Dose_po are absolute bioavailability, rat total clearance obtained from i.v. data, AUC of blood concentration versus time after dosage in a p.o. study, and dose of a p.o. study, respectively.

If several doses were examined in a previous study (Zhou et al., 2008), the data of the minimum dose were adopted, considering the solubility and saturation of enzyme and transporter kinetics.

Rat F_h and F_aF_g were calculated using the below equations:

where F_ℎ, F_aF_g, rCL_h, and rQ_h are hepatic bioavailability, fraction of the administered compounds that reach the portal blood, rat hepatic clearance and rat hepatic blood flow rate, respectively.

The value of rQ_h (0.853 L/hr) was described previously (Kato et al., 2008). rCL_h was estimated as follows:

i. The contribution of rCL_r is less than 5% of rCL_tot,iv: rCL_h = rCL_{tot, iv}

ii. The rCL_r is more than 5% of rCL_tot,iv: rCL_ℎ = rCL_{tot, iv} − rCL_r

In cases where the value of rCL_r was available (epicatechin (Yu, 2000)), the reported values were used for the calculation. In cases where the renal clearance data in rats (genistein) was not available, rCL_tot,iv was regarded as rCL_h.

RESULTS

Comparison of the F_aF_g values estimated by AS-F_aF_gCM and traditional approach

To check the validity of AS-F_aF_gCM, the estimated values of F_aF_g for quercetin were compared between AS- F_aF_gCM and the traditional approaches using human i.v. data. The values of hCL_tot,iv, human F, F_h, and F_aF_g estimated by AS-F_aF_gCM or the traditional approach are shown in Table 1. The values of hCL_tot,iv, human F, and F_h in AS-F_aF_gCM were within 2-fold of those of the traditional approach. Also, The F_aF_g value estimated by AS-F_aF_gCM was 1.2- and 2.0-fold lower than that estimated by the traditional approach with two reference datasets, respectively. These results suggest that AS-F_aF_gCM could estimate pharmacokinetic parameters, such as F_aF_g, comparable to the traditional approach.

Table 1. The values of quercetin for the traditional approach and AS-F_aF_gCM.

	AS-FaFgCM	Traditional approach
rCLtot,iv (L/hr)	1.57	-	-
Rat BW (kg)	0.263	-	-
hCLtot,iv (L/hr)	59.12*1	34.3*2	46.7*3
Human dosepo (mg)	500*4
Human AUCpo (mg*hr/L)†	0.0541*4
Human BW (kg)	64.1*4	-	-
Human F	0.0062	0.00545	0.00744
hCLr (L/hr)	1.68*5
hCLh (L/hr)††	59.12	34.3	46.7
hQh (L/hr)	96.6*6
Human Fh	0.388	0.645	0.516
Human FaFg	0.0165	0.00845	0.0144

The rCL_tot,iv was estimated using the rat dose_iv, AUC_iv, and BW from three rat i.v. studies (Chang et al., 2015; Khaled et al., 2003; Yang et al., 2005). Rat BW was the mean value of three rat i.v. studies. *¹: Estimated from “average rCL_tot,iv” with the average human BW in Europe (Walpole et al., 2012); *²*³: values from human i.v. studies (*²: (Ferry et al., 1996); *³: (Gugler et al., 1975)); *⁴: (Riva et al., 2019); *⁵: (Moon et al., 2008); *⁶: (Kato et al., 2008); ^†: The plasma concentration was converted to the blood concentration by multiplying Rb value of 0.68 (Furukawa et al., 2012); ^††: The value of hCL_tot,iv was directly applied to the hCL_h because hCL_r was less than 5% of hCL_{tot, iv}.

F_aF_g estimation of structurally different polyphenols by AS-F_aF_gCM

Next, we estimated the F_aF_g values of genistein, epicatechin, baicalein, and resveratrol using AS-F_aF_gCM. The values used for AS-F_aF_gCM and the estimated values for the four polyphenols are shown in Table 2. The estimated F and F_aF_g values of those polyphenols were less than 0.05 and 0.15, respectively. The F_aF_g value of epicatechin was lower than those of the other three polyphenols. These results suggest that these four polyphenols had low bioavailability and intestinal absorption, and epicatechin had the most incomplete intestinal absorption.

Table 2. The values of genistein, epicatechin, baicalein and resveratrol for AS-F_aF_gCM.

	Genistein	Epicatechin	Baicalein	Resveratrol
rCLtot,iv (L/hr)†	0.552	0.485	2.54	2.27
hCLtot,iv (L/hr)	21.9	25.9	79.2	78.5
Human dosepo (mg)	30.0*1	50.0*2	100*4	500*5
				400*6
				500*7
Human AUCpo (mg∙hr/L)†	0.0397*1	0.0199*2	0.0250*4	0.182*5
				0.092*6
				0.121*7
Human BW (kg)	62.7*1	80.7*2,3	62.2*4	75.8*3,5
				70.8*3,6
				69.1*7
Human F	0.029	0.0103	0.0198	0.0220
hCLr (L/hr)	N.A.	3.6*8	0.01*4	0.86*5
hCLh (L/hr)††	21.9	22.3	79.2	78.5
hQh (L/hr)	96.6*9
Human Fh	0.773	0.769	0.180	0.19
Human FaFg	0.0375	0.0134	0.110	0.116

The rCL_tot,iv was estimated using the rat dose_iv, AUC_iv, and BW from the following rat i.v. studies: genistein (Zhou et al., 2008; Coldham et al., 2002), epicatechin (Yu, 2000), baicalein (Lai et al., 2003), and resveratrol (Kapetanovic et al., 2011; Peñalva et al., 2018; Marier et al., 2002). *¹: (Metzner et al., 2009); *²: (Barnett et al., 2015); *³: (Walpole et al., 2012); *⁴: (Li et al., 2014); *⁵: (Boocock et al., 2007); *⁶: (Vaz-da-Silva et al., 2008); *⁷: (Sergides et al., 2016). The average BW in Europe was used in (Vaz-da-Silva et al., 2008); that in North America was used in (Barnett et al., 2015); the average value of the average European BW and the average North America BW was used in (Boocock et al., 2007); and the average BW in each region was obtained from (Walpole et al., 2012). *⁸: the values was estimated from rat CL_r reported in (Yu, 2000). ^*9: (Kato et al., 2008). ^†: In resveratrol, the plasma concentration was converted to the blood concentration by multiplying Rb value of 0.78 (Furukawa et al., 2012) ^††: The value of hCL_tot,iv was applied to the hCL_h because hCL_r was unavailable (baicalein) or less than 5% of hCL_{tot, iv} (baicalein and resveratrol). N.A.: Not available

F_aF_g estimation of naringenin by AS-F_aF_gCME

The F_aF_g value of naringenin was estimated by AS-F_aF_gCME using rat i.v. and human p.o. data of deconjugated samples by β-glucuronidase and sulfatase. The data used for the estimation and estimated values of naringenin are shown in Table 3. The estimated values of F', F_h', and F_aF_g were 0.134, 0.989, and 0.136, respectively. The F_aF_g value of naringenin was comparable to that of baicalein (0.11, Table 2), a structural isomer of naringenin. These results demonstrate the low intestinal bioavailability of naringenin.

Table 3. The values of naringenin for AS-F_aF_gCME

rCLtot,iv' (L/hr)	0.0559
Human BW (kg)*1	69.2
Human dose (mg)*1	135
Human AUCagri+metab (mg·hr/L)*1	9.42
Human REagri+metab (%)*1	5.81
hCLtot,iv' (L/hr)	1.92
Human F'	0.134
hCLr' (L/hr)	0.832
hCLh' (L/hr)	1.09
Human Fh'	0.989
Human FaFg	0.136

The estimated values (rCL_tot,iv', hCL_tot,iv', hCL_r', F', F_h', and F_aF_g) and the values from the human p.o. study (human dose_po, human AUC_po', human BW, and human renal excretion of aglycon and its metabolites (RE_agri+metab)) are shown. rCL_tot,iv' was estimated using the rat dose_iv, AUC_agri+metab, and BW from the rat i.v. study (Wang et al., 2006). *1: (Kanaze et al., 2007).

Comparison of F, F_h, and F_aF_g between rats and humans

The rat F, F_h, and F_aF_g values were compared with human values estimated for genistein and epicatechin, which have reliable rat p.o. data. The data used for the calculation and the calculated values are shown in Table 4. The rat F values were 0.20 and 0.102 for genistein, and 0.07 for epicatechin (Table 4), which were over 3-fold higher than that in humans (0.029 for genistein and 0.0103 for epicatechin, Table 2). In addition, rat F_aF_g values for genistein (0.57 and 0.29) and epicatechin (0.13) in Table 4 were approximately 10-fold higher than those in humans (0.0375 for genistein and 0.0134 for epicatechin in Table 2). However, rat F_h values for genistein (0.35) and epicatechin (0.51) in Table 4 were comparable to those in humans (0.773 for genistein and 0.769 for catechin in Table 2). These results suggested large species differences in intestinal availability.

Table 4. The values for genistein and epicatechin in rats

	Genistein		Epicatechin
rCLtot,iv (L/hr)	0.552		0.485
Rat dosepo (mg)	1.25*1	1*2	13.7*3
Rat AUCpo (mg∙hr/L)	0.452*1	0.205*2	1.88*3
Rat F	0.20	0.102	0.07
rCLr,iv (L/hr)	N.A.		0.0709*4
rCLh (L/hr)	0.552*5		0.414
rQh (L/hr)	0.853*4
Rat Fh	0.35		0.51
Rat FaFg	0.57	0.29	0.13

The rCL_tot,iv was calculated using the rat dose_iv, AUC_iv, and BW from the following rat i.v. studies: genistein (Zhou et al., 2008; Coldham et al., 2002), epicatechin (Yu, 2000) *¹: (Zhou et al., 2008); *²: (Coldham et al., 2002); *³: (Baba et al., 2001); *⁴: (Yu, 2000); *⁵: The value of rCL_tot,iv was applied to the rCL_h because rCL_r was unavailable. N.A.: Not available

DISCUSSION

In this study, the validity of AS-F_aF_gCM was examined of quercetin, for which human i.v. data were reported in two studies. The results showed that F_aF_g of AS-F_aF_gCM (0.0165) was approximately 2.0 or 1.2 times of F_aF_g obtained by the traditional approach (0.00845 and 0.0144, respectively) (Table 1). Previously, it has been mentioned that inter-individual variability of bioavailability would occur up to four times, especially when the mean bioavailability is low (e.g., 10%) (Toutain and Bousquet-Mélou, 2004). As shown in Table 1, quercetin had a low bioavailability (approximately 1%), suggesting that the difference in F_aF_g between AS-F_aF_gCM and the traditional approach could be within the range of variation. Indeed, there was a 1.7-fold difference in the F_aF_g values between the two i.v. studies obtained using the traditional approach (Table 1). Thus, AS-F_aF_gCM can be considered to estimate values that are practically comparable to those of the traditional approach.

Both methods showed extremely low intestinal absorption of quercetin (F_aF_g = 0.00845 to 0.0165 in Table 1). F_h values of quercetin (0.388 to 0.645) were more than 10-times higher than F_aF_g values, suggesting that the contribution of hepatic metabolism on bioavailability of quercetin was not dominant. When rat intestine was perfused in situ with quercetin, 90% of the formed quercetin conjugates existed in the intestinal eluent and only 10% of the quercetin conjugates were recorded in the bile (Crespy et al., 2003). In addition, in vitro studies using human intestinal Caco-2 cells showed rapid intestinal conjugation and MRP2 mediated-efflux transport of quercetin (Nait Chabane et al., 2009; Ikeno et al., 1999). These data suggested that quercetin was extensively metabolized and excreted in the luminal side of small intestine, which is consistent with the results obtained with AS-F_aF_gCM. Therefore, AS-F_aF_gCM is applicable to the F_aF_g estimation of polyphenols.

We also determined the F and F_aF_g values for genistein, epicatechin, baicalein, and resveratrol using AS-F_aF_gCM (Table 2). The results showed that the F values were less than 0.05. It has been reported that the plasma AUC ratios of aglycone to its metabolites are approximately 1.3, 4,3, 3.1, and 1.0% for epicatechin, baicalein, resveratrol, and genistein, respectively (Barnett et al., 2015; Li et al., 2014; Boocock et al., 2007; Metzner et al., 2009). These data imply that the F values of these polyphenols were very low, which is consistent with the results of AS-F_aF_gCM, suggesting that AS-F_aF_gCM can accurately obtain human F from rat i.v. data.

The F_aF_g values of these four polyphenols were less than 0.15 (Table 2), suggesting low intestinal absorption; however, the F_aF_g of epicatechin was extremely lower compared to the others. Extensive intestinal metabolism of these polyphenols has been suggested in human and rat intestinal perfusion studies (Andlauer et al., 2000; Actis-Goretta et al., 2013; Zhang et al., 2005). In contrast, Caco-2 studies have shown that the permeability of epicatechin is in the range of poor absorption (Vaidyanathan and Walle, 2001; Yee, 1997), while genistein, baicalein, and resveratrol are classified as well-absorbed compounds (Chen et al., 2014; Willenberg et al., 2015; Kitaguchi et al., 2022). Therefore, lower F_aF_g of epicatechin could be derived from a lower fraction of dose absorbed (F_a), but not the fraction of dose passing through the gut wall without metabolism (F_g).

AS-F_aF_gCM is applicable to the pharmacokinetic data of aglycone; however, only the total concentration data of aglycone and its metabolites have been reported for several polyphenols owing to their rapid metabolism. Thus, AS-F_aF_gCM was further modified for application to the total concentration data of both aglycones and metabolites, namely AS-F_aF_gCME (Fig. 2B). Naringenin is a structural isomer of baicalein; naringenin glucuronides were excreted through MRP2 (Xu et al., 2009), suggesting that naringenin meets the assumption for applying AS-F_aF_gCME described in Materials and Methods. Table 2 shows the F_aF_g value of naringenin obtained by AS-F_aF_gCME using the pharmacokinetic data of deconjugated samples, demonstrating its low intestinal bioavailability. Naringenin is rapidly metabolized into glucuronide and sulfate conjugates, and the main plasma metabolite is naringenin-o-β-D-glucuronide, which constitutes 98% of the total metabolites detected (Joshi et al., 2018; El Mohsen et al., 2004). A mouse perfusion study revealed that naringenin was absorbed well in the small intestine and colon, but was extensively metabolized by phase II enzymes. These observations show the poor intestinal bioavailability of naringenin, which is consistent with the F_aF_g value estimated by AS-F_aF_gCME.

The kinetic parameters of intestinal metabolism and permeability of naringenin are similar to that of baicalein. The human intrinsic intestinal clearance values of baicalein and naringenin in human intestinal microsomes are 125 µL/min/mg protein (Li et al., 2012) and 166.6 µL/min/mg protein (Isobe et al., 2018), respectively. The apparent permeability values (P_app) for baicalein and naringenin in Caco-2 cells are 1.7 × 10⁻⁵ cm/s (Chen et al., 2014) and 1.7 × 10⁻⁵ cm/s (Nait Chabane et al., 2009), respectively. These results suggested that the F_aF_g values of naringenin and baicalein were similar. Indeed, the F_aF_g of naringenin estimated by AS-F_aF_gCME (0.136) was close to that of baicalein estimated by AS-F_aF_gCM (0.110) (Table 2 and 3). Therefore, AS-F_aF_gCME would be applicable to the estimation of human F_aF_g of polyphenols.

F_aF_g values for genistein and epicatechin in rats were approximately10-fold higher than that in humans (Table 2 and 4), suggesting that there are large species differences in intestinal availability between rats and humans. These discrepancies could be due to the difference in the expression of key drug metabolizing enzymes. UDP-glucuronosyltransferase (UGT) 1A10 is involved in the metabolism of genistein in the human gut (Doerge et al., 2000). Rat intestine lacks UGT1A10 (Jeong et al., 2005); therefore, the large difference in F_aF_g of genistein between rats and humans could be attributed to the difference in the expression of UGT1A10. Similar species differences are observed for the UGT1A10-metabolized drug; raloxifene, whose bioavailability in humans (2%) is much lower than that in rats (39%) (Jeong et al., 2005). Epicatechin was metabolized by sulfotransferase (SULT) 1A1 and SULT1A3 in the human small intestinal microsome (Vaidyanathan and Walle, 2002). However, SULT1A3 has no known animal orthologue (Peters et al., 2016); therefore, the lack of SUT1A3 in rat intestine influenced the higher F_aF_g value in rats compared to that in humans. The method proposed in this study could more precisely estimate the human F_aF_g of polyphenols than the traditional i.v. approach using a rat model.

AS-F_aF_gCM and AS-F_aF_gCME were based on previously reported allometric scaling methods (Chiou et al., 1998). Chiou et al. discussed that the allometric approach may inaccurately predict clearance if the liver is not the main metabolic tissue in humans, or if a drug is mainly excreted unchanged in humans but is mainly metabolized in the rat liver. Various polyphenols absorbed in the intestine undergo rapid hepatic phase II metabolism in both humans and rats, because they contain one or more phenolic rings with hydroxy groups; thus, the allometric approach is considered applicable to polyphenols. However, the linearity or nonlinearity of kinetic parameters, such as gastrointestinal absorption, first-pass metabolism, and apparent volume of distribution may also need to be considered. Although we adopted the data of the lowest dose in the reference studies for F_aF_g estimation to avoid saturation in enzyme and transporter kinetics, the F_aF_g values varied depending on the dose used for estimation. Finally, the validity of the AS-F_aF_gCME could not be confirmed; however, the applicability of AS-F_aF_gCM was demonstrated in comparison to the traditional i.v. approach. This is a limitation of this study; further research is warranted to ensure the validity of AS-F_aF_gCME.

In conclusion, this study determined the intestinal bioavailability of structurally diverse polyphenols and quantitatively demonstrated their poor intestinal absorption. In this study, we developed AS-F_aF_gCM and AS-F_aF_gCME, which are comprehensive approaches that can be used to obtain F_aF_g of polyphenols without using human i.v. data. Scarce i.v. data of polyphenols in humans have prevented us from understanding their intestinal absorption; however, this study provides a solution to this insufficient knowledge and potentially contributes to the development of novel alternative methods that can quantitatively predict the internal exposure of polyphenols.

ACKNOWLEDGMENTS

We would like to thank Editage (www.editage.com) for English language editing.

Conflict of interest

The authors declare that there is no conflict of interest.

REFERENCES

•  Actis-Goretta,  L.,  Lévèques,  A.,  Rein,  M.,  Teml,  A.,  Schäfer,  C.,  Hofmann,  U.,  Li,  H.,  Schwab,  M.,  Eichelbaum,  M. and  Williamson,  G. (2013): Intestinal absorption, metabolism, and excretion of (-)-epicatechin in healthy humans assessed by using an intestinal perfusion technique. Am. J. Clin. Nutr., 98, 924-933.
•  Agoram,  B.,  Woltosz,  W.S. and  Bolger,  M.B. (2001): Predicting the impact of physiological and biochemical processes on oral drug bioavailability. Adv. Drug Deliv. Rev., 50 (Suppl 1), S41-S67.
•  Akao,  T.,  Sakashita,  Y.,  Hanada,  M.,  Goto,  H.,  Shimada,  Y. and  Terasawa,  K. (2004): Enteric excretion of baicalein, a flavone of Scutellariae Radix, via glucuronidation in rat: involvement of multidrug resistance-associated protein 2. Pharm. Res., 21, 2120-2126.
•  Andlauer,  W.,  Kolb,  J.,  Stehle,  P. and  Fürst,  P. (2000): Absorption and metabolism of genistein in isolated rat small intestine. J. Nutr., 130, 843-846.
•  Ando,  H.,  Hisaka,  A. and  Suzuki,  H. (2015): A new physiologically based pharmacokinetic model for the prediction of gastrointestinal drug absorption: translocation model. Drug Metab. Dispos., 43, 590-602.
•  Baba,  S.,  Osakabe,  N.,  Natsume,  M.,  Muto,  Y.,  Takizawa,  T. and  Terao,  J. (2001): In vivo comparison of the bioavailability of (+)-catechin, (-)-epicatechin and their mixture in orally administered rats. J. Nutr., 131, 2885-2891.
•  Barnett,  C.F.,  Moreno-Ulloa,  A.,  Shiva,  S.,  Ramirez-Sanchez,  I.,  Taub,  P.R.,  Su,  Y.,  Ceballos,  G.,  Dugar,  S.,  Schreiner,  G. and  Villarreal,  F. (2015): Pharmacokinetic, partial pharmacodynamic and initial safety analysis of (-)-epicatechin in healthy volunteers. Food Funct., 6, 824-833.
•  Bohn,  T. (2014): Dietary factors affecting polyphenol bioavailability. Nutr. Rev., 72, 429-452.
•  Bonfili,  L.,  Cecarini,  V.,  Amici,  M.,  Cuccioloni,  M.,  Angeletti,  M.,  Keller,  J.N. and  Eleuteri,  A.M. (2008): Natural polyphenols as proteasome modulators and their role as anti-cancer compounds. FEBS J., 275, 5512-5526.
•  Boocock,  D.J.,  Faust,  G.E.,  Patel,  K.R.,  Schinas,  A.M.,  Brown,  V.A.,  Ducharme,  M.P.,  Booth,  T.D.,  Crowell,  J.A.,  Perloff,  M.,  Gescher,  A.J.,  Steward,  W.P. and  Brenner,  D.E. (2007): Phase I dose escalation pharmacokinetic study in healthy volunteers of resveratrol, a potential cancer chemopreventive agent. Cancer Epidemiol. Biomarkers Prev., 16, 1246-1252.
•  Chang,  L.W.,  Hou,  M.L.,  Hung,  S.H.,  Lin,  L.C. and  Tsai,  T.H. (2015): Pharmacokinetics of quercetin-loaded nanodroplets with ultrasound activation and their use for bioimaging. Int. J. Nanomedicine, 10, 3031-3042.
•  Chen,  H.,  Gao,  Y.,  Wu,  J.,  Chen,  Y.,  Chen,  B.,  Hu,  J. and  Zhou,  J. (2014): Exploring therapeutic potentials of baicalin and its aglycone baicalein for hematological malignancies. Cancer Lett., 354, 5-11.
•  Chiou,  W.L.,  Robbie,  G.,  Chung,  S.M.,  Wu,  T.C. and  Ma,  C. (1998): Correlation of plasma clearance of 54 extensively metabolized drugs between humans and rats: mean allometric coefficient of 0.66. Pharm. Res., 15, 1474-1479.
•  Choi,  G.W.,  Lee,  Y.B. and  Cho,  H.Y. (2019): Interpretation of non-clinical data for prediction of human pharmacokinetic parameters: in vitro-in vivo extrapolation and allometric scaling. Pharmaceutics, 11, 168.
•  Coldham,  N.G.,  Zhang,  A.Q.,  Key,  P. and  Sauer,  M.J. (2002): Absolute bioavailability of [14C] genistein in the rat; plasma pharmacokinetics of parent compound, genistein glucuronide and total radioactivity. Eur. J. Drug Metab. Pharmacokinet., 27, 249-258.
•  Crespy,  V.,  Morand,  C.,  Besson,  C.,  Cotelle,  N.,  Vézin,  H.,  Demigné,  C. and  Rémésy,  C. (2003): The splanchnic metabolism of flavonoids highly differed according to the nature of the compound. Am. J. Physiol. Gastrointest. Liver Physiol., 284, G980-G988.
•  de Araújo,  F.F.,  de Paulo Farias,  D.,  Neri-Numa,  I.A. and  Pastore,  G.M. (2021): Polyphenols and their applications: an approach in food chemistry and innovation potential. Food Chem., 338, 127535.
•  Dedrick,  R.L. (1973): Animal scale-up. J. Pharmacokinet. Biopharm., 1, 435-461.
•  Doerge,  D.R.,  Chang,  H.C.,  Churchwell,  M.I. and  Holder,  C.L. (2000): Analysis of soy isoflavone conjugation in vitro and in human blood using liquid chromatography-mass spectrometry. Drug Metab. Dispos., 28, 298-307.
•  Di Lorenzo,  C.,  Colombo,  F.,  Biella,  S.,  Stockley,  C. and  Restani,  P. (2021): Polyphenols and human health: the role of bioavailability. Nutrients, 13, 273.
•  El Mohsen,  M.A.,  Marks,  J.,  Kuhnle,  G.,  Rice-Evans,  C.,  Moore,  K.,  Gibson,  G.,  Debnam,  E. and  Srai,  S.K. (2004): The differential tissue distribution of the citrus flavanone naringenin following gastric instillation. Free Radic. Res., 38, 1329-1340.
•  Ferry,  D.R.,  Smith,  A.,  Malkhandi,  J.,  Fyfe,  D.W.,  deTakats,  P.G.,  Anderson,  D.,  Baker,  J. and  Kerr,  D.J. (1996): Phase I clinical trial of the flavonoid quercetin: pharmacokinetics and evidence for in vivo tyrosine kinase inhibition. Clin. Cancer Res., 2, 659-668.
•  Furukawa,  T.,  Nakamori,  F.,  Tetsuka,  K.,  Naritomi,  Y.,  Moriguchi,  H.,  Yamano,  K.,  Terashita,  S. and  Teramura,  T. (2012): Quantitative prediction of intestinal glucuronidation of drugs in rats using in vitro metabolic clearance data. Drug Metab. Pharmacokinet., 27, 171-180.
•  Gertz,  M.,  Harrison,  A.,  Houston,  J.B. and  Galetin,  A. (2010): Prediction of human intestinal first-pass metabolism of 25 CYP3A substrates from in vitro clearance and permeability data. Drug Metab. Dispos., 38, 1147-1158.
•  Gugler,  R.,  Leschik,  M. and  Dengler,  H.J. (1975): Disposition of quercetin in man after single oral and intravenous doses. Eur. J. Clin. Pharmacol., 9, 229-234.
•  Hu,  M.,  Wu,  B. and  Liu,  Z. (2017): Bioavailability of polyphenols and flavonoids in the era of precision medicine. Mol. Pharm., 14, 2861-2863.
•  Ikeno,  A.,  Hashimoto,  K.,  Shimizu,  M.,  Terao,  J.,  Kumazawa,  S. and  Nakayama,  T. (1999): Absorption and metabolism of quercetin in Caco-2 cells. Food Sci. Technol. Res., 5, 90-92.
•  Isobe,  T.,  Ohkawara,  S.,  Ochi,  S.,  Tanaka-Kagawa,  T.,  Jinno,  H. and  Hanioka,  N. (2018): Naringenin glucuronidation in liver and intestine microsomes of humans, monkeys, rats, and mice. Food Chem. Toxicol., 111, 417-422.
•  Jeong,  E.J.,  Liu,  Y.,  Lin,  H. and  Hu,  M. (2005): Species- and disposition model-dependent metabolism of raloxifene in gut and liver: role of UGT1A10. Drug Metab. Dispos., 33, 785-794.
•  Joseph,  S.V.,  Edirisinghe,  I. and  Burton-Freeman,  B.M. (2016): Fruit polyphenols: a review of anti-inflammatory effects in humans. Crit. Rev. Food Sci. Nutr., 56, 419-444.
•  Joshi,  R.,  Kulkarni,  Y.A. and  Wairkar,  S. (2018): Pharmacokinetic, pharmacodynamic and formulations aspects of Naringenin: an update. Life Sci., 215, 43-56.
•  Kanaze,  F.I.,  Bounartzi,  M.I.,  Georgarakis,  M. and  Niopas,  I. (2007): Pharmacokinetics of the citrus flavanone aglycones hesperetin and naringenin after single oral administration in human subjects. Eur. J. Clin. Nutr., 61, 472-477.
•  Kapetanovic,  I.M.,  Muzzio,  M.,  Huang,  Z.,  Thompson,  T.N. and  McCormick,  D.L. (2011): Pharmacokinetics, oral bioavailability, and metabolic profile of resveratrol and its dimethylether analog, pterostilbene, in rats. Cancer Chemother. Pharmacol., 68, 593-601.
•  Kato,  M. (2008): Intestinal first-pass metabolism of CYP3A4 substrates. Drug Metab. Pharmacokinet., 23, 87-94.
•  Kato,  M.,  Shitara,  Y.,  Sato,  H.,  Yoshisue,  K.,  Hirano,  M.,  Ikeda,  T. and  Sugiyama,  Y. (2008): The quantitative prediction of CYP-mediated drug interaction by physiologically based pharmacokinetic modeling. Pharm. Res., 25, 1891-1901.
•  Khaled,  K.A.,  El-Sayed,  Y.M. and  Al-Hadiya,  B.M. (2003): Disposition of the flavonoid quercetin in rats after single intravenous and oral doses. Drug Dev. Ind. Pharm., 29, 397-403.
•  Kitaguchi,  T.,  Mizota,  T.,  Ito,  M.,  Ohno,  K.,  Kobayashi,  K.,  Ogawa,  I.,  Qiu,  S.,  Iwao,  T.,  Hanioka,  N.,  Tanaka,  M. and  Matsunaga,  T. (2022): Simultaneous evaluation of membrane permeability and UDP-glucuronosyltransferase-mediated metabolism of food-derived compounds using human induced pluripotent stem cell-derived small intestinal epithelial cells. Drug Metab. Dispos., 50, 17-23.
•  Lai,  M.Y.,  Hsiu,  S.L.,  Tsai,  S.Y.,  Hou,  Y.C. and  Chao,  P.D. (2003): Comparison of metabolic pharmacokinetics of baicalin and baicalein in rats. J. Pharm. Pharmacol., 55, 205-209.
•  Li,  C.R.,  Zhang,  L.,  Wo,  S.K.,  Zhou,  L.M.,  Lin,  G. and  Zuo,  Z. (2012): Pharmacokinetic interactions among major bioactive components in Radix Scutellariae via metabolic competition. Biopharm. Drug Dispos., 33, 487-500.
•  Li,  M.,  Shi,  A.,  Pang,  H.,  Xue,  W.,  Li,  Y.,  Cao,  G.,  Yan,  B.,  Dong,  F.,  Li,  K.,  Xiao,  W.,  He,  G.,  Du,  G. and  Hu,  X. (2014): Safety, tolerability, and pharmacokinetics of a single ascending dose of baicalein chewable tablets in healthy subjects. J. Ethnopharmacol., 156, 210-215.
•  Manach,  C.,  Williamson,  G.,  Morand,  C.,  Scalbert,  A. and  Rémésy,  C. (2005): Bioavailability and bioefficacy of polyphenols in humans. I. Review of 97 bioavailability studies. Am. J. Clin. Nutr., 81 (Suppl), 230S-242S.
•  Marier,  J.F.,  Vachon,  P.,  Gritsas,  A.,  Zhang,  J.,  Moreau,  J.P. and  Ducharme,  M.P. (2002): Metabolism and disposition of resveratrol in rats: extent of absorption, glucuronidation, and enterohepatic recirculation evidenced by a linked-rat model. J. Pharmacol. Exp. Ther., 302, 369-373.
•  Mennen,  L.I.,  Walker,  R.,  Bennetau-Pelissero,  C. and  Scalbert,  A. (2005): Risks and safety of polyphenol consumption. Am. J. Clin. Nutr., 81 (Suppl), 326S-329S.
•  Metzner,  J.E.,  Frank,  T.,  Kunz,  I.,  Burger,  D. and  Riegger,  C. (2009): Study on the pharmacokinetics of synthetic genistein after multiple oral intake in post-menopausal women. Arzneimittelforschung, 59, 513-520.
•  Moon,  Y.J.,  Wang,  L.,  DiCenzo,  R. and  Morris,  M.E. (2008): Quercetin pharmacokinetics in humans. Biopharm. Drug Dispos., 29, 205-217.
•  Nait Chabane,  M.,  Al Ahmad,  A.,  Peluso,  J.,  Muller,  C.D. and  Ubeaud,  G. (2009): Quercetin and naringenin transport across human intestinal Caco-2 cells. J. Pharm. Pharmacol., 61, 1473-1483.
•  Peñalva,  R.,  Morales,  J.,  González-Navarro,  C.J.,  Larrañeta,  E.,  Quincoces,  G.,  Peñuelas,  I. and  Irache,  J.M. (2018): Increased oral bioavailability of resveratrol by its encapsulation in casein nanoparticles. Int. J. Mol. Sci., 19, 2816.
•  Peters,  S.A.,  Jones,  C.R.,  Ungell,  A.L. and  Hatley,  O.J. (2016): Predicting drug extraction in the human gut wall: assessing contributions from drug metabolizing enzymes and transporter proteins using preclinical models. Clin. Pharmacokinet., 55, 673-696.
•  Riva,  A.,  Ronchi,  M.,  Petrangolini,  G.,  Bosisio,  S. and  Allegrini,  P. (2019): Improved oral absorption of quercetin from quercetin phytosome^®, a new delivery system based on food grade lecithin. Eur. J. Drug Metab. Pharmacokinet., 44, 169-177.
•  Scalbert,  A.,  Manach,  C.,  Morand,  C.,  Rémésy,  C. and  Jiménez,  L. (2005): Dietary polyphenols and the prevention of diseases. Crit. Rev. Food Sci. Nutr., 45, 287-306.
•  Scalbert,  A.,  Morand,  C.,  Manach,  C. and  Rémésy,  C. (2002): Absorption and metabolism of polyphenols in the gut and impact on health. Biomed. Pharmacother., 56, 276-282.
•  Seeff,  L.B.,  Bonkovsky,  H.L.,  Navarro,  V.J. and  Wang,  G. (2015): Herbal products and the liver: a review of adverse effects and mechanisms. Gastroenterology, 148, 517-532.e3.
•  Selick,  H.E.,  Beresford,  A.P. and  Tarbit,  M.H. (2002): The emerging importance of predictive ADME simulation in drug discovery. Drug Discov. Today, 7, 109-116.
•  Sergides,  C.,  Chirilă,  M.,  Silvestro,  L.,  Pitta,  D. and  Pittas,  A. (2016): Bioavailability and safety study of resveratrol 500 mg tablets in healthy male and female volunteers. Exp. Ther. Med., 11, 164-170.
•  Shibata,  N.,  Ishida,  M.,  Prasad,  Y.V.,  Gao,  W.,  Yoshikawa,  Y. and  Takada,  K. (2003): Highly sensitive quantification of vancomycin in plasma samples using liquid chromatography-tandem mass spectrometry and oral bioavailability in rats. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci., 789, 211-218.
•  Singh,  M.,  Thrimawithana,  T.,  Shukla,  R. and  Adhikari,  B. (2020): Managing obesity through natural polyphenols: a review. Future Foods, 1-2, 100002.
•  Toutain,  P.L. and  Bousquet-Mélou,  A. (2004): Bioavailability and its assessment. J. Vet. Pharmacol. Ther., 27, 455-466.
•  Truzzi,  F.,  Tibaldi,  C.,  Zhang,  Y.,  Dinelli,  G. and  D  Amen, E. (2021): An overview on dietary polyphenols and their biopharmaceutical classification system (BCS). Int. J. Mol. Sci., 22.
•  Vaidyanathan,  J.B. and  Walle,  T. (2001): Transport and metabolism of the tea flavonoid (-)-epicatechin by the human intestinal cell line Caco-2. Pharm. Res., 18, 1420-1425.
•  Vaidyanathan,  J.B. and  Walle,  T. (2002): Glucuronidation and sulfation of the tea flavonoid (-)-epicatechin by the human and rat enzymes. Drug Metab. Dispos., 30, 897-903.
•  Vaz-da-Silva,  M.,  Loureiro,  A.I.,  Falcao,  A.,  Nunes,  T.,  Rocha,  J.F.,  Fernandes-Lopes,  C.,  Soares,  E.,  Wright,  L.,  Almeida,  L. and  Soares-da-Silva,  P. (2008): Effect of food on the pharmacokinetic profile of trans-resveratrol. Int. J. Clin. Pharmacol. Ther., 46, 564-570.
•  Walpole,  S.C.,  Prieto-Merino,  D.,  Edwards,  P.,  Cleland,  J.,  Stevens,  G. and  Roberts,  I. (2012): The weight of nations: an estimation of adult human biomass. BMC Public Health, 12, 439.
•  Wang,  M.-J.,  Chao,  P.-D.,  Hou,  Y.-C.,  Hsiu,  S.-L.,  Wen,  K.-C. and  Tsai,  S.-Y. (2006): Pharmacokinetics and conjugation metabolism of naringin and naringenin in rats after single dose and multiple dose administrations. Yao Wu Shi Pin Fen Xi, 14, 247-253.
•  Willenberg,  I.,  Michael,  M.,  Wonik,  J.,  Bartel,  L.C.,  Empl,  M.T. and  Schebb,  N.H. (2015): Investigation of the absorption of resveratrol oligomers in the Caco-2 cellular model of intestinal absorption. Food Chem., 167, 245-250.
•  Woo,  S. and  Jusko,  W.J. (2007): Interspecies comparisons of pharmacokinetics and pharmacodynamics of recombinant human erythropoietin. Drug Metab. Dispos., 35, 1672-1678.
•  Xu,  H.,  Kulkarni,  K.H.,  Singh,  R.,  Yang,  Z.,  Wang,  S.W.,  Tam,  V.H. and  Hu,  M. (2009): Disposition of naringenin via glucuronidation pathway is affected by compensating efflux transporters of hydrophilic glucuronides. Mol. Pharm., 6, 1703-1715.
•  Yang,  C.S.,  Sang,  S.,  Lambert,  J.D.,  Lee,  M.J.,  Tsai,  S.-Y.,  Hou,  Y.-C. and  Chao,  P.-D. (2008): Bioavailability issues in studying the health effects of plant polyphenolic compounds. Mol. Nutr. Food Res., 52 (Suppl 1), S139-S151.
•  Yang,  C.-Y.,  Hsiu,  S.-L.,  Wen,  K.-C.,  Lin,  S.-P., Thai, S.-Y., Hou, Y.-C. and Chao, P.-D.L. (2005): Bioavailability and metabolic pharmacokinetics of rutin and quercetin in rats. Yao Wu Shi Pin Fen Xi, 13, 244-250.
•  Yee,  S. (1997): In vitro permeability across Caco-2 cells (colonic) can predict in vivo (small intestinal) absorption in man--fact or myth. Pharm. Res., 14, 763-766.
• Yu, C. (2000): Pharmacokinetics of tea catechins in the rat, Chinese University of Hong Kong. https://repository.lib.cuhk.edu.hk/en/item/cuhk-323641.
•  Zhang,  L.,  Lin,  G.,  Chang,  Q. and  Zuo,  Z. (2005): Role of intestinal first-pass metabolism of baicalein in its absorption process. Pharm. Res., 22, 1050-1058.
•  Zhang,  L.,  Lin,  G.,  Kovács,  B.,  Jani,  M.,  Krajcsi,  P. and  Zuo,  Z. (2007): Mechanistic study on the intestinal absorption and disposition of baicalein. Eur. J. Pharm. Sci., 31, 221-231.
•  Zhou,  S.,  Hu,  Y.,  Zhang,  B.,  Teng,  Z.,  Gan,  H.,  Yang,  Z.,  Wang,  Q.,  Huan,  M. and  Mei,  Q. (2008): Dose-dependent absorption, metabolism, and excretion of genistein in rats. J. Agric. Food Chem., 56, 8354-8359.