Chemicals.

We purchased 1′-hydroxylmidazolam and STZ from Sigma-Aldrich Co. (Shanghai, China); atorvastatin, midazolam, and repaglinide from the National Institute for the Control of Pharmaceutical and Biologic Products (Beijing, China); parahydroxy atorvastatin (P-OH-Ator) calcium and ortho-hydroxy atorvastatin (O-OH-Ator) calcium from TLC Pharmaceutical Standards Ltd. (Aurora, ON, Canada); β-actin antibody from Bioworld (Louis Park, MN); mouse monoclonal CYP3a1, OATP1b2 (M-50), OATP1a5, (H-82), and BCRP (M-70) antibodies from Santa Cruz Technology (Santa Cruz, CA); and mouse monoclonal antibody specific for P-gp (clone C219) from Calbiochem-Novabiochem (Seattle, WA). All other reagents were commercially available.

Animals.

Male Sprague-Dawley rats were purchased from Sino-British Sipper & BK Laboratory Animal Ltd. (Shanghai, China) and housed in an environment of controlled temperature (22 ± 2°C) and relative humidity (50% ± 5%) with 12-hour light/dark cycle. Water and food were allowed ad libitum. Animal experiments were carried out according to institutional guidelines for the care and the use of laboratory animals and approved by the Animal Ethics Committee of China Pharmaceutical University (no. CPU-PCPK-1631010070). Experimental procedures were conducted in accordance with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health.

Development of DM Rats.

DM rats were induced by HFD feeding plus low-dose STZ intraperitoneal injection according to the method described previously (Shu et al., 2016) (Supplemental Methods). We once investigated effect of diabetic progression on the pharmacokinetics of atorvastatin after an oral dose of atorvastatin using diabetic rats induced by combination of HFD feeding and low-dose STZ. Compared with CON rats, the DM rats on day 10 (termed 10-day DM rats) after STZ injection showed significantly higher oral plasma exposure of Ator. By contrast, the DM rats on day 22 (termed 22-day DM rats) after STZ intraperitoneal injection showed significantly lower oral plasma exposure of Ator. Therefore, both 10-day DM rats and 22-day DM rats were used for the following studies. The age-matched HFD rats and the age-matched CON rats served as the controls.

Ator Metabolism in Rat Intestinal and Hepatic Microsomes.

Ator is metabolized mainly to O-OH-Ator and P-OH-Ator by intestinal and hepatic CYP3a. Here, we detected Ator metabolism in hepatic and intestinal microsomes. We prepared hepatic and intestinal microsomes from experimental rats according to the methods described previously (Hu et al., 2011). The incubation mixture consisted of 40 μl of rat hepatic microsomes (final level, 0.2 mg/ml) or intestinal microsomes (final level, 1 mg/ml), 40 μl of the NADPH-regenerating system (0.5 mM NADP, 10 mM glucose 6-phosphate, 1 U/ml glucose-6-phosphate dehydrogenase, and 5 mM MgCl₂), and phosphate-buffered saline (PBS, pH = 7.4; final volume, 200 μl). After a 10-minute preincubation at 37°C, the reaction was initiated by the addition of 10 μl of Ator solution and incubated for 10 minutes. The reaction was terminated by adding 1 ml of ethyl acetate. The formations of O-OH-Ator and P-OH-Ator were measured liquid chromatography-mass spectrometry (LC-MS) (Shu et al., 2016). All the preceding microsomal incubation conditions were in the linear range of the reaction rate. The final concentrations of Ator were set at 2.5, 5, 10, 20, and 40 μM, respectively. Formation of 1′-hydroxylmidazolam (1′-OH-MDZ) from midazolam (MDZ, 10 μM) in intestinal and hepatic microsomes was also measured to assess CYP3A activity after 15-minute incubation using the previously described incubation system. Free fractions (f_umic) of Ator in intestinal and hepatic microsome systems were measured using ultrafiltration on centrifugal filters (YM-30; Merck Millipore Ltd., Burlington, MA), and K_m values were corrected by f_umic.

Uptake and Metabolism of Ator in Primary Rat Hepatocytes.

The isolation of primary rat hepatocytes was operated according to the method reported previously (Jia et al., 2014). The isolated hepatocytes (viability >85%), assessed by trypan blue assay, were suspended in plating medium [Dulbecco’s modified Eagle’s medium (DMEM) containing 5% fetal bovine serum, 1 mM dexamethasone, and 4 mg/l insulin] at a density of 2.5 × 10⁵ cells/ml. A 500-μl aliquot of the hepatocytes’ suspension was seeded in collagen precoated plates for each well. After a 4-hour incubation in a humidified atmosphere containing 5% CO₂, the hepatocytes were used for investigating the Ator metabolism and uptake.

The primarily cultured hepatocytes of experimental rats were rinsed twice with 500 μl of warm Hanks’ balanced salt solution (HBSS) and incubated with 500 μl of the same buffer for 5 minutes at 37°C. For uptake, hepatocytes were incubated in 500 μl of HBSS containing Ator (0.1, 0.5, 2, 10, and 40 μM, respectively) for 2 minutes, and the reaction was terminated by washing three times with ice-cold HBSS. Intracellular concentrations of Ator in hepatocytes were measured by LC-MS. Kinetics of Ator uptake is characterized by eq. 1 (Vildhede et al., 2014):(1)where V and CL_{up,int, Pas} are initial uptake velocity and the clearance via passive diffusion, respectively. K_m and V_max are the Michaelis constant and the maximal uptake rate for the saturable uptake, respectively. Kinetic parameters were estimated using Phoenix WinNonlin 8.1 (Pharsight, St. Louis, MO). The CL_{up, int, active} was defined as V_max/K_m, and the total uptake clearance (CL_int,uptake) was calculated by CL_{up,int,active} plus CL_{up, int,Pas}.

It is generally accepted that Ator uptake in hepatocytes is mediated mainly by OATP1b2. Rifampicin is a typical inhibitor of OATP1b2. Several reports have demonstrated that 100 μM rifampicin could completely inhibit OATP1B1-mediated uptake of estradiol-17β-glucuronide (Gui et al., 2008) and OATP1b2-mediated uptake of rosuvastatin and olmesartan acid (Ishida et al., 2018). Ator uptake in the primary hepatocytes of normal rats was also measured in the presence of the OATP1b2 inhibitor rifampicin (200 μM) to investigate the contribution of OATP1b2 to the hepatic uptake of Ator. The OATP1b2-mediated intrinsic uptake clearance (CL_{int,up,Oatp1b2}) of Ator was calculated using eq. 1.

For metabolism, hepatocytes were incubated in 500 μl of HBSS containing Ator (0.1, 0.5, 2, 10, and 40 μM, respectively) for 30 minutes. The reaction was terminated by rinsing three times with ice-cold HBSS. The formations of O-OH-Ator and P-OH-Ator were measured by LC-MS. Formation of 1′-hydroxymidazolam from midazolam (10 μM) after 120 minutes of incubation and 2 minutes of uptake of repaglinide (20 μM) were used to index the functions of hepatic CYP3a and OATP1b2, respectively.

Intestinal Absorption of Ator in Rats.

In situ single-pass intestinal perfusion experiments were performed to evaluate the intestinal absorption of Ator, as well as functions of intestinal P-gp and BCRP in DM rats according to the method described previously (Doluisio et al., 1969; Cummins et al., 2003; Zhong et al., 2016) (Supplemental Methods). The perfusion Krebs-Henseleit buffer containing tested agents (0.5 μM Ator, 0.5 μM prazosin, or 0.5 μM rhodamine 123) prewarmed at 37°C was perfused at 0.2 ml/min for about 30 minutes until steady state. The outlet perfusate samples were collected every 15 minutes for 120 minutes. The areas of the perfused segments (A) were measured at the end of the experiment. The apparent effective permeability (P_eff, cm/min) was calculated according to the equation P_eff = −Q·ln(C_out/C_in)/A, where C_out and C_in were concentrations of the tested agents in the output and input buffers, respectively. Q (ml/min) was the flow rate corrected using a gravimetric method (Zhong et al., 2016).

Contributions of OATP1a5, BCRP, and P-gp to intestinal absorption of Ator in normal rats (weighing 230–250 g) were estimated using in situ single-pass intestinal perfusion in the presence of corresponding transporter inhibitors, respectively. The isolated jejunum of rats were perfused with perfusion buffer containing Ator (0.5 μM), Ator (0.5 μM) + verapamil (P-gp inhibitor, 200 μM), Ator (0.5 μM) + Ko143 (BCRP inhibitor, 1 μM), and Ator (0.5 μM) + naringin (OATP1a5 inhibitor, 200 μM), respectively. Concentrations of verapamil, Ko143, and naringin were taken from previous reports (Jiang et al., 2015; Duan et al., 2017). Assuming their effects were additive, the contribution of each transporter to intestinal absorption of Ator was estimated according to eqs. 2–5:(2)(3)(4)(5)where P_eff,ator, P_eff,+ver, P_eff,+Ko143, and P_eff,+Nar represent apparent effective permeability of Ator in rats without transporter inhibitor, with verapamil, Ko143, and naringin, respectively. P_eff,Pas, P_eff,Oatp1a5, P_eff,Bcrp, and P_eff,P-gp represent the P_eff values of Ator-transported process, OATP1a5-mediated Ator uptake, BCRP-mediated Ator efflux, and P-gp–mediated Ator efflux across the apical membrane of enterocytes, respectively.

Development of a Semi-PBPK Model Involving Both Metabolic Enzyme and Drug Transporters.

A semi-PBPK model involving both intestinal or hepatic metabolic enzyme and transporters (Fig. 1) was constructed to describe the pharmacokinetic profiles of Ator after oral or intravenous administration to CON, HFD, and DM rats. The essential structure of the PBPK model included stomach, gut lumen, gut wall, portal vein, hepatic blood, hepatocytes, and system compartments. The gut lumen and gut wall were anatomically divided into duodenum, jejunum, and ileum, with corresponding compartment volume, luminal radius, and transit time.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 1.

Schematic diagram of semi-PBPK model involving both enzyme and transporter. A_i and Q_i indicate the drug amount and blood flow in the corresponding compartment, respectively. CL_int,gw,i shows the intrinsic clearance in the ith enterocyte compartment. K_i, k_a,i, and k_b,i represent the transit rate constant, drug absorption rate constant, and efflux rate constant from enterocyte to gut lumen, respectively. CL_up,ce, CL_ef,ce, and CL_h,met were uptake clearance, efflux clearance, and metabolic clearance in hepatocytes.

Stomach.

Assuming neither absorption nor metabolism occurred in the stomach, the amount of drug (A₀) in the stomach was governed by the gastric emptying rate (eq. 6):(6)where k₀ represents the gastric emptying rate constant.

Gut Lumen.

Absorption of Ator was often involved in intestinal BCRP, P-gp, and OATP1a5. Transport of Ator from the gut lumen to enterocytes was assumed to be governed by the diffusion process and OATP1a5-mediated uptake (Generaux et al., 2011), whose transfer rate constant was termed k_a,i. By contrast, efflux of Ator from enterocytes into the lumen was mediated by intestinal P-gp and BCRP (Generaux et al., 2011; Kellick et al., 2014), whose transfer rate constant was termed k_b.i.

Expression of the intestinal P-gp, BCRP, and OATP1a5 were often region-dependent. Relative expression of intestinal P-gp, BCRP, and OATP1a5 progressively increased from proximal to distal regions (Kawase et al., 2009; MacLean et al., 2010; Yu et al., 2010). Therefore, relative scaling factors (SC_P-gp,i, SC_Bcrp,i, and SC_Oatp1a5,i) for P-gp, BCRP, and OATP1a5 were incorporated into the current PBPK model to correct P_eff,P-gp, P_eff,Bcrp, and P_eff,Oatp1a5 in corresponding gut segments. In general, values of P_eff were often estimated using jejunum; thus, P_eff,P-gp,i, P_eff,Bcrp,i, and P_{eff,Oatp1a5,i} in the other gut segment were estimated using P_{eff,P-gp, jejunum} × SC_P-gp,i, P_{eff,Bcrp, jejunum} × SC_Bcrp,i, and P_{eff,Oatp1a5, jejunum} × SC_Oatp1a5,i, respectively. Assuming the expression of jejunal P-gp, BCRP, and OATP1a5 proteins were set to be 1, SC_{P-gp,doudenum,}:SC_{P-gp, jejunum}:SC_{P-gp, ileum} were estimated to be 0.578:1:1.883 using our previous report (Yu et al., 2010); SC_{Bcrp,doudenum}:SC_Bcrp,jejunum:SC_Bcrp,ileum were calculated to be 0.399:1:0.878 according to a previous report (Kawase et al., 2009); and SC _{Oatp1a5,doudenu}:SC _{Oatp1a5, jejunum}:SC _{Oatp1a5, ileum} were estimated to be 0.84:1:1.73 based on data reported by MacLean et al. (2010).

Thus, the k_a,i and k_b,i. in the ith gut lumen were estimated using eqs. 7 and 8, respectively:(7)(8)where i = 1, 2, and 3 represent duodenum, jejunum and ileum, respectively. r_i represents the radius of the ith gut lumen.

The drug amount (A_i) in the gut lumen is described by eq. 9:(9)where k_i and A_i,gw represent the transit rate constant and drug amount in the ith enterocyte compartment, respectively; f_ugut represents the unbound drug fraction in the enterocytes and is assumed to be 1 (Yang et al., 2007; Guo et al., 2013).

Enterocyte Compartment.

We assumed that each luminal compartment was associated with a unique enterocyte compartment, with no transit of drug between adjacent enterocyte compartments. The drug amount in the ith enterocyte compartment is illustrated by eqs. 10 and 11:(10)(11)where V_i,gw, Q_i,gw, A_p, V_p, and K_pgut represent the volume of the ith enterocyte compartment, its blood flow, the drug amount in portal vein compartment, the volume of portal vein, respectively, and intestine-to-plasma drug concentration ratio. K_m,gw,j and V_max,gw,j are kinetic parameters for formation of O-OH-Ator (j = 1) and P-OH-Ator (j = 2) from Ator in intestinal microsomes. PBSF_i is physiologically based scaling factor (microsome content) in the ith enterocyte compartment. R_B is Ator blood/plasma ratio, which was set to be 1.5 (Paine et al., 2008). fu_at,gut refers to the free fraction of atorvastatin in rat intestinal microsomes, measured to be 0.89 ± 0.01.

Expression and activity of intestinal Cyp3a were also region-dependent (Takemoto et al., 2003; Mitschke et al., 2008). Here, Ator metabolism was documented in jejunal microsomes. Therefore, Ator metabolism in other gut segments was corrected by relative scaling factor (SC_CYP3a). Relative activity of CYP3a in the jejunum was set to be 1, and SC_{CYP3a,duodenum}, and SC_{CYP3a, ileum} were estimated to be 1.293 and 0.742, respectively, based on the reported data (Takemoto et al., 2003).

Portal Vein.

The drug amount in the portal vein was described by eq. 12:(12)where Q_p, A_c1, and V_c1 are the blood flow of portal vein, the drug amount, and its apparent distribution volume in the system compartment, respectively.

Hepatic Compartment.

In general, hepatic clearance of some drugs, including statins, was often governed by hepatic uptake. Thus, a two-compartment model consisting of hepatic blood and hepatocytes was used to illustrate the disposition of drugs in the liver (i.e., in hepatic blood) (eqs. 13 and 14):(13)And in hepatocytes:(14)(15)where A_h,b, A_h,ce, V_h,b, and V_h,ce represent the amount of drug in the liver blood and liver cell compartment, volume of the liver blood and liver cell compartment, respectively. Q_h,at represents the blood flow of the hepatic artery, and Q_h = Q_pv + Q_h,at. f_u, and K_ph represents the unbound drug fraction in the blood and the liver-to-blood concentration ratio, separately. CL_up,ce, CL_ef,ce, and CL_h,m represent intrinsic clearances of Ator uptake, efflux, and metabolism in hepatocytes, respectively. K_m,h,j and V_max,h,j are kinetic parameters for the formation of O-OH-Ator (j = 1) and P-OH-Ator (j = 2) from Ator in hepatic microsomes, respectively. PBSF_h is physiologically based scaling factor of metabolism in liver; fu_at,h refers to the free fraction of atorvastatin in rat liver microsomes, which was measured to be 0.95 ± 0.01.

System Compartment and Peripheral Compartment.

Assuming urinary excretion of Ator was negligible, the drug kinetics in the compartment is described by eqs. 16 and 17:(16)(17)where A_c2 and V_c2 are the drug amount and its apparent distribution volume in the peripheral compartment, and k₁₂ and k₂₁ are the drug amount rate constants between the peripheral compartment and system compartment; respectively.

Coding and solving of the PBPK model were conducted by Phoenix WinNonlin 8.1 (Pharsight). Pharmacokinetic profiles of Ator in plasma of CON rats, HFD rats, and DM rats after oral or intravenous administration were predicted. Pharmacokinetic parameters (such as C_max, T_max, mean residence time, t_1/2, AUC_0–∞, volume clearance, CL, and steady-state apparent volume of distribution) of the predicted profiles were estimated using noncompartmental analysis.

Contribution of Altered Expression of Transporters and CYP3a in Intestine and Liver to Ator Pharmacokinetics.

Expression of transporters and CYP3a in intestine and liver of rats were measured using Western blotting (Supplemental Methods). Diabetes significantly altered the function and expression of transporters and CYP3a in the intestine and liver of rats in a different manner. We further investigated individual contribution of the altered expression of transporters and CYP3a to Ator pharmacokinetics and their integrated effects according to their alterations under diabetic status.

Pharmacokinetics of Ator after Oral Administrations to Rats.

Pharmacokinetic profiles of Ator in experimental rats were documented after oral administration of Ator (10 mg/kg) to 10-day DM rats, 22-day DM rats, the CON rats and HFD rats. Blood samples (250 μl) were collected in microcentrifuge tubes containing heparin under light ether anesthesia via the oculi chorioideae vein at 5, 10, 20, 30, 45, 60, 120, 240, and 360 minutes postdose. After three or four samplings, the appropriate amount of normal saline was administered to the rats via the tail vein to compensate for blood loss. Plasma samples were obtained by centrifugation for 10 minutes and stored at −80°C until analysis.

Drug Analysis.

Concentrations of Ator, O-OH-Ator and P-OH-Ator in biologic fluids were determined using an LC-MS method previously described (Shu et al., 2016). The linear ranges of Ator, O-OH-Ator, and P-OH-Ator were 3.9–500 ng/ml in plasma and hepatocytes and were 3.9–1000 ng/ml in microsomes. If the concentrations of analytes in biologic matrix exceeded the upper limit of quantification, analysis was operated following dilution using the blank matrix.

Statistical Analysis.

All results were expressed as mean ± S.D. Statistical differences among groups were evaluated using one-way ANOVA. If analysis results were significant, the differences between groups were estimated by using a Student-Newman-Keuls multiple-comparison post hoc test. P < 0.05 was considered statistically significant.