(A-14) Physiologically-based pharmacokinetic computational modeling of sodium fluorescein blood clearance for the assessment of the degree of hepatic ischemia-reperfusion injury

Introduction:: Introduction: Hepatic ischemia-reperfusion injury (IRI) inherent to liver transplantation affects physiological processes including bile formation (PMID: 32246581). Quantitative analysis of the bile formation machinery has the potential to provide an assessment of the viability of a liver before its transplantation. Recently, we demonstrated the utility of the fluorescent dye sodium fluorescein (SF) by showing that the bile-to-plasma SF fluorescence ratio is inversely correlated with warm ischemia time in rats in vivo, and we associated this finding with IRI-induced intracellular internalization of the canalicular MRP2 transporter (PMID: 35700191). However, it is unclear how much SF is cleared by active MRP2 transporter in relation to other contributing factors associated with complex in vivo physiology. We hypothesize that physiologically-based pharmacokinetic (PBPK) computational modeling can account for dominant vascular, tissue, and biochemical processes contributing to SF blood clearance through hepatocytes into bile, including MRP2 activity, and simulate diminished liver function in IRI leading to decreased disposition of SF in the bile.

Materials and Methods:: Materials and Methods: We considered previously published in vivo data of SF fluorescence intensity in blood, plasma, and bile over time in rats under control and IRI conditions (PMID: 35700191). To model these data, we first performed SF calibration experiments and converted these data from SF fluorescence intensity to concentration using standard curves generated from nonlinear regression. We then fitted an empirical triple-exponential function to the calibrated control and IRI blood SF concentration data and used this function as an input for a mechanistic liver-centric PBPK model. The PBPK model accounts for three regions in the liver (blood, hepatocytes, bile) and major physiological processes, including SF metabolism into SF glucuronides (SFG) in hepatocytes and the kinetics of basolateral and canalicular transporters governing SF and SFG dynamics in different liver regions. We fitted this model to the control and IRI SF concentration data in the bile using a pseudo-Monte Carlo parameter estimation strategy to characterize alterations in the basolateral and canalicular transporter activities associated with IRI. Correspondingly, model parameters relating to maximal velocity for the transporters were allowed to vary, while all other parameters were fixed at their standard physiologic values.

Results, Conclusions, and Discussions:: Results and Discussion: The liver-centric PBPK model was able to provide good fits to the control and IRI bile SF concentration data, providing estimates of the alterations in the basolateral and canalicular transporter activities associated with IRI. The input model for blood SF concentration was able to predict plasma SF concentration related by a variable partition coefficient, which led to a large difference between plasma and blood SF concentrations at high SF concentrations but diminished at lower SF concentrations. These results were not significantly altered in IRI. We also simulated the IRI condition by decreasing the canalicular MRP2 transporter activity and potentially the basolateral OATP1B2 and MRP3 transporters activities in the control condition, which showed that SF blood clearance through hepatocytes into bile is highly sensitive to changes in the activity of MRP2, consistent with our hypothesis that SF clearance data and PBPK modeling could quantify the effect of IRI on MRP2 activity. These simulations showed a larger decrease in SF concentration in bile relative to the increase in blood resulting in a lower bile-to-blood SF concentration ratio. The simulation results were consistent with measured bile SF concentration data from livers with IRI and with decrease in value of the model parameter descriptive of MRP2 activity estimated by fitting the PBPK model solution to bile SF concentration data from livers with IRI. The model is also able to predict the corresponding dynamics of SF glucuronides (SFG) in blood, plasma, and bile, and how SFG importantly contributes towards the overall SF disposition in liver in IRI.

Conclusions: Based on our experimental data, we developed a liver-centric PBPK model for the blood clearance of SF through hepatocytes into bile and used this model to simulate the contribution of MRP2 transporter to diminished secretion of SF into the bile as hypothesized following liver IRI.


Acknowledgements (Optional): :

References (Optional): :