(M1130-02-12) A Mechanistic Rat PBPK Model for Genistein and Its Glucuronide Metabolite: Integrating Liver and Enteric Metabolism with Transporter-Mediated Biliary and Enteroluminal Clearance

Purpose: Glucuronide recycling profoundly affects systemic exposure of drug and its glucuronide, thus impacting clinical efficacy and/or toxicity of drugs. In addition, some drugs like morphine and buprenorphine form active glucuronide metabolites, which also contribute to the overall pharmacological and toxicological responses. Therefore, it is important to build mechanistic PBPK models for glucuronide disposition to predict the parameters that could significantly impact efficacy and safety of drugs undergoing glucuronidation. For this purpose, we used genistein (4', 5, 7-trihydroxyisoflavone), a naturally occurring soy isoflavone as a model compound. It has been studied extensively for its antioxidant, anti-inflammatory, and anticancer properties. Based on several published literatures, after oral dose, genistein is highly metabolized by UGT and SULT enzymes in the enterocytes of intestine before reaching liver and formed metabolites are majorly excreted into the bile and urine. The biliary excreted metabolites undergo deglucuronidation by intestinal glucuronidases to convert back to genistein which can be reabsorbed back into the systemic circulation thereby increasing its half-life. In this study, a mechanistic rat PBPK model was built to describe the hepatic and enteroluminal metabolism, the transporter-mediated biliary excretion, the passive and transporter-mediated intestinal absorption/efflux, heterogenous tissue distribution, and renal/fecal excretion of genistein and genistein glucuronide.

Methods: A whole-body rat PBPK model was built using the MATLAB SimBiology software (version R2023b). In addition to defining the PK of genistein, the PBPK model also captures the disposition of genistein glucuronide, a major circulating glycone metabolite in rats. Both physiological and drug related parameters were obtained from the literature. To define the distribution and elimination parameters, previously published PK data from intravenously-dosed rats were utilized [1]. In vitro intrinsic metabolic clearance values for bacterial beta-glucuronidase, liver, and intestine were obtained from rat fecal and tissue microsomal incubation data, respectively [2, 3]. Kidney metabolic clearance of genistein was included and estimated in the model based on the reported renal mRNA expression of UGTs, which has resulted in a better fit. Genistein glucuronide requires transporters for its movement in and out of cells due to its charge and hydrophilic nature. As such, organ-specific apical efflux transporters Bcrp and Mrp2 were used to outline the biliary excretion and intestinal efflux of the glucuronide [4, 5], whereas the basolateral efflux transporter Mrp3 was used to delineate the glucuronide uptake into the blood [6]. In vitro to in vivo extrapolation (IVIVE) was performed for all the in vitro values by scaling them based on the tissue weight, tissue enzyme/transporter protein expression abundance, and/or relative activity factors (RAF) where applicable for transporters and metabolic clearance values. Where experimental data were unavailable, these scaling factors were fitted to the PK data. Renal and fecal excretions were included as the possible elimination pathway for genistein and/or genistein glucuronide.

Results: The mechanistic PBPK model was fitted to the rat intravenous PK data to estimate distribution and elimination related parameters. The estimated values along with their coefficient of variances and confidence intervals were given in Table 1. As mentioned in the methods, RAFs for apical and basolateral transporters were estimated as a part of IVIVE. Permeability surface area products for slowly perfused tissues and liver were also estimated along with tissue partition coefficient factor to get suitable distribution profile for genistein. A renal elimination factor was implemented for the glucuronide to account for any transporter mediated excretion in the kidney. The overlay of observed and predicted plasma PK profiles given in Figure 1 shows that the final model fits well with the observed data. Subsequently, sensitivity analysis was performed to determine the influence of model parameters on AUC and/or C_max of genistein and its glucuronide. Mrp2-mediated biliary efflux of genistein glucuronide was shown to have a larger influence on the glucuronide PK profile than Mrp3-mediated basolateral efflux (Figure 2). Due to the lack of evidence for enteric apical uptake of genistein glucuronide, the second peak in the genistein glucuronide PK profile (Figure 1) could only be observed after genistein regeneration by enteroluminal beta-glucuronidase in the distal ileum and colon. As a result, genistein solubility and pKa could also influence genistein glucuronide PK profile (Figure 2) as genistein aqueous solubility is low and pH dependent and the regenerated genistein amount greatly exceeded its solubility inside the enteroluminal space. Additional factors, e.g., renal elimination factor, also showed significant influence on the genistein glucuronide plasma AUC (data not shown).

Conclusion: The newly developed mechanistic PBPK model for genistein and its glucuronide was able to capture their IV profiles in rats after incorporating transporter mediated clearance for biliary excretion and enterohepatic circulation of the glucuronide, intestinal glucuronidase mediated regeneration of genistein, and solubility limitation in the distal ileum and colon. This model will serve as the foundation to construct an oral PBPK model of genistein in rats, which can be further extrapolated to humans by using human specific physiological and drug related properties.

References: 1. Zhou, S., et al., Dose-dependent absorption, metabolism, and excretion of genistein in rats. Journal of Agricultural and Food Chemistry, 2008. 56(18): p. 8354-8359.
2. Chen, J., et al., Disposition of flavonoids via recycling: comparison of intestinal versus hepatic disposition. Drug metabolism and Disposition, 2005. 33(12): p. 1777-1784.
3. Ebuzoeme, C., et al., Glucuronides hydrolysis by intestinal microbial β-glucuronidases (GUS) is affected by sampling, enzyme preparation, buffer pH, and species. Pharmaceutics, 2021. 13(7): p. 1043.
4. Ge, S., et al., Transport–Glucuronidation Classification System and PBPK Modeling: New Approach To Predict the Impact of Transporters on Disposition of Glucuronides. Molecular Pharmaceutics, 2017. 14(9): p. 2884-2898.
5. Wang, M., et al., Establishment and use of new MDCK II cells overexpressing both UGT1A1 and MRP2 to characterize flavonoid metabolism via the glucuronidation pathway. Molecular nutrition & food research, 2016. 60(9): p. 1967-1983.
6. Van De Wetering, K., et al., Targeted metabolomics identifies glucuronides of dietary phytoestrogens as a major class of MRP3 substrates in vivo. Gastroenterology, 2009. 137(5): p. 1725-1735.


Table 1. PBPK estimated parameters.


Figure 1. Simulated plasma concentration time profiles overlaid with observed data obtained from intravenous dosing of 12.5 mg/kg genistein.


Figure 2. Sensitivity of genistein glucuronide plasma concentration towards efflux transporter expression in liver and based on solubility of genistein regenerated by intestinal beta glucuronidase.