In vitro metabolism, reaction phenotyping, enzyme kinetics, CYP inhibition and induction potential of ataluren

Abstract Ataluren promotes ribosomal readthrough of premature termination codons in mRNA which result from nonsense mutations. In vitro studies were performed to characterize the metabolism and enzyme kinetics of ataluren and its interaction potential with CYP enzymes. Incubation of [14C]‐ataluren with human liver microsomes indicated that the major metabolic pathway for ataluren is via direct glucuronidation and that the drug is not metabolized via cytochrome P450 (CYP). Glucuronidation was also observed in the incubation in human intestinal and kidney microsomes, but not in human pulmonary microsomes. UGT1A9 was found to be the major uridine diphosphate glucuronosyltransferase (UGT) responsible for ataluren glucuronidation in the liver and kidney microsomes. Enzyme kinetic analysis of the formation of ataluren acyl glucuronide, performed in human liver, kidney, and intestinal microsomes and recombinant human UGT1A9, found that increasing bovine serum albumin (BSA) levels enhanced the glucuronidation Michaelis‐Menten constant (Km) and ataluren protein binding but had a minimal effect on maximum velocity (Vmax) of glucuronidation. Due to the decreased unbound Michaelis‐Menten constant (Km,u), the ataluren unbound intrinsic clearance (CLint,u) increased for all experimental systems and BSA concentrations. Human kidney microsomes were about 3.7‐fold more active than human liver microsomes, in terms of CLint,u/mg protein, indicating that the kidney is also a key organ for the metabolism and disposition of ataluren in humans. Ataluren showed no or little potential to inhibit or induce most of the CYP enzymes.


| INTRODUC TI ON
Ataluren (PTC124, Translarna™, Figure 1) is a small molecule drug that is being developed for the treatment of genetic disorders resulting from nonsense mutations. Nonsense mutations, which encode premature stop codons, result in the premature termination of protein translation in the coding region of an mRNA. 1 Ataluren promotes ribosomal read-through of a premature termination codon in the dystrophin mRNA, a cause of Duchenne muscular dystrophy, resulting in production of full-length protein. [1][2][3][4][5] Duchenne muscular dystrophy (DMD) is an X-linked genetic muscle disorder that results from the presence of mutations in the gene that encodes the dystrophin protein. Dystrophin provides stability to the muscle and acts as a shock absorber, bearing the mechanical stresses that occur during contraction, stabilizing the cell membranes, and protecting the muscle from injury. [6][7][8][9] Approximately, 10% to 15% of boys with DMD have the disease due to nonsense mutations. [10][11][12] Ataluren has shown to produce full length, functional dystrophin ( 2,4,13 ) in the nonsense mutation mdx mouse and sapje zebrafish models of DMD. Ataluren activity has also been demonstrated in multiple cell-based and animal disease models of other nonsense mutation genetic disorders, corroborating its ability to promote readthrough of premature stop codons and its potential for treating genetic disease caused by nonsense mutations. [14][15][16][17][18][19] Comprehensive nonclinical studies have been conducted in safety pharmacology and secondary pharmacodynamics, pharmacokinetics and metabolism, and toxicology programs. Following a single oral dose of [ 14 C]-ataluren in mice, rats, dogs, and humans, ataluren was well absorbed and cleared primarily by metabolism. Biliary secretion was the major route for elimination of drug-related radioactivity in bile-duct cannulated rats. Ataluren acyl glucuronide was the only detectable metabolite in human plasma, and the major metabolic and clearance pathways in humans are similar to animal species. 20 The acceptable pharmacokinetic, toxicokinetic, and safety profiles of ataluren support the use of ataluren for chronic administration to patients with genetic disorders resulting from nonsense mutations.
Ataluren clinical development program consists of Phase 1 studies characterizing the absorption, metabolism, and excretion profile of ataluren and evaluating the pharmacokinetics and safety profiles in healthy subjects, [20][21][22] Phase 2 studies in patients with nonsense mutation genetic disorders, 5 and long-term Phase 2b/3 studies in patients with nonsense mutation Duchenne muscular dystrophy (nmDMD) and nonsense mutation cystic fibrosis (nmCF). [23][24][25] Ataluren has shown good tolerability in healthy subjects and patients, and has demonstrated clinical benefit in two multicenter, randomized, clinical trials to slow disease progression in patients with nmDMD. 24,25 On this basis, ataluren received conditional marketing authorization for the treatment of nmDMD in ambulatory patients transporter-mediated drug-drug interaction potentials in healthy subjects, pharmacokinetics evaluation of ataluren in special population, and safety, pharmacokinetic, and pharmacodynamic assessments in nmDMD patients aged ≥ 2 to < 5 years, etc. The purpose of the current manuscript is to describe the in vitro metabolism and enzyme kinetics of ataluren and its interaction with CYP enzymes.
The results presented here have provided the understanding of ataluren disposition and metabolism in vitro and informed the need and design of additional clinical studies to investigate the potential drugdrug interactions in human.    After centrifugation, the entire ultrafiltrate was weighed and analyzed for radioactivity using liquid scintillation counting. The percent bound = (1-C u /C m )×100, where C u is the concentration of radioactivity in the ultrafiltrate and C m is the concentration of radioactivity in the plasma before centrifugation.

| Correlation analysis
Ataluren at a concentration of 10 μmol/L, was incubated (n = 2) with 16 individual HLM at 0.5 mg protein concentration/mL for 0 and 60 minutes at 37°C. Enzyme activities for UGTs 1A1, 1A4, 1A6, 1A9, and 2B7 were precharacterized under optimized conditions by TA B L E 1 Analysis of the correlation between the rate of disappearance of ataluren, the rate of the glucuronide formation, and marker UGT enzyme activity in a bank (n = 16) of individual human liver microsomes    for hepatic, renal, and intestinal data, respectively. 26

| Correction for BSA binding
The high-throughput membrane ultrafiltration method 27 was used to determine the fraction unbound in the incubation (n = 3).
Incubation mixtures and procedures were essentially the same as the above except that UDPGA (5 mmol/L) was replaced with Tris-

| CYP inhibition
The metabolic reactions that were monitored and probe substrate concentrations used are shown in Table S1.  (Table S1).
Incubations were stopped with the addition of ice-cold acetonitrile.
Metabolite formation in incubations with test compound and control inhibitors was assessed with LC-MS/MS methods for each of the reaction products as described in Table S2. For inhibition constant     with validated LC-MS/MS methods for each of the reaction products as described in Table S2. Enzymatic activity for each CYP was calculated using the absolute amount of metabolite formed (pmol) divided by hepatocytes in the incubation (million cells) and the incubation time (minute). The relative fold induction in enzymatic activity was calculated by comparing the rate of metabolite formation for treatment groups to that of the solvent control group (reported as vehicle control%) or positive control group (reported as positive control%).

| Reaction phenotyping for the formation of ataluren acyl glucuronide
Quantification of ataluren loss and ataluren-O-1β-acyl glucuronide formation was carried out with an LC-MS/MS system, which con-

| Kinetics for the formation of ataluren acyl glucuronide
Quantification of ataluren-O-1β-acyl glucuronide formation was conducted on an Accela pump and a PAL autosampler equipped with a TSQ Quantum Ultra mass spectrometer (Thermo Fisher Scientific).
A Waters Acquity BEH column (C18, 1.7 µm, 50 × 2.1 mm; Waters Cooperation) was used and maintained at 50°C. The mobile phases were 0.1% formic acid in water (A) and acetonitrile (B). The flow rate was 0.6 mL/min and gradient program was as follows: linear gradient from 0% B to 95% B in 0.5 minutes and hold isocratic at 95% B

| Data analysis
The study design was not powered for statistical analysis. All assays were run in singlet, duplicate or triplicate and were used only to test

| Reaction phenotyping for ataluren glucuronidation
The greatest loss of ataluren and the greatest formation of ataluren-O-1β-acyl glucuronide was seen with recombinant human UGT1A9 (32.7% ataluren loss and 187 pmol/mg/min ataluren glucuronide formation after 60 minutes incubation, n = 2) followed by UGT1A7 significantly correlated with one another (r = 0.948) confirming that the same enzyme (UGT1A9) was responsible for both the loss of ataluren and ataluren acyl glucuronide formation (Table 1 and Figure 2).

| Ataluren glucuronidation kinetics
The results are shown in Table 2, Figure 3 and  (Table 2).

| CYP inhibition potential
The appropriate positive controls used in this study inhibited the enzyme activities at acceptable levels indicating that the test system was functional (data not shown). As summarized in Table S3,

| CYP induction potential
Hepatocytes from all three human donors used in the study responded well to exposure of all prototypical inducers. At the concentrations tested, ataluren was not cytotoxic as indicated by methylthiazolyldiphenyl-tetrazolium bromide (MTT) assay (data not shown). Ataluren did not induce the activities of CYPs 1A2, 2A6, 2C8, 2C19, 2E1, or 3A4 relative to the vehicle control, 1% DMSO, but showed approximately 1.7-and 1.5-fold induction (n = 3) over vehicle control at 400 µmol/L for CYP2B6 and CYP2C9 activity, respectively. The mean enzyme activities are summarized in Table S4.

| D ISCUSS I ON AND CON CLUS I ON
The findings of the current studies show that the direct conjugation of ataluren with glucuronic acid to form ataluren acyl glucuronide is the primary metabolic pathway of ataluren, consistent with in vivo observations.20 CYP plays only a minimal role in the metabolism of ataluren. UGT1A9 was the major enzyme responsible for the formation of ataluren-O-1β-acyl glucuronide. UGT1A7 was found to be the minor enzyme for producing ataluren acyl glucuronide.
Furthermore, in UDPGA fortified HLM, the loss of ataluren and the formation of ataluren acyl glucuronide correlated significantly with glucuronidation of propofol, a marker substrate for UGT1A9 activity. Human kidney microsomes were more active than liver microsomes and intestinal microsomes, whereas intestinal microsomes was least active in terms of unbound intrinsic clearance of ataluren glucuronidation.
Ataluren UGT reaction phenotyping results are similar to propofol glucuronidation in microsomes from human liver, intestine, and kidney. 26 They are also consistent with the fact that UGT1A9 is mainly expressed in liver and kidney and is more abundant in kidney than in liver, is expressed at low levels in the small intestine and is not expressed in the lung. 28,29 UGT1A7 is minimally expressed in the liver and at low levels in the kidney, small intestine, and similar to UGT1A9, it is not expressed in the lung. 28,29 In vitro kinetic data revealed that human kidney microsomes is about 3.7-fold more active than human liver microsomes in ataluren glucuronidation, in terms of CL int,u /mg protein, indicating that in addition to the liver, the kidney is also a key organ for the metabolism and disposition of ataluren in human. Though HIM showed relatively low CL int,u for ataluren glucuronidation activity, contribution of intestinal UGTs cannot be excluded for ataluren first pass metabolism in small intestine following oral dose. However, the maximal effect on CL int,u were system/tissue dependent and were increased approximately 1.6-, 10-, 4.8-, and 3.7-fold for recombinant UGT1A9, and human liver, kidney, and intestine microsomes, respectively. Regarding the use of BSA, 2% BSA is most frequently used in different test systems. However, based on limited data, the BSA concentration may also play a role in enzyme activation in a substrate/system-dependent manner. 32,35,36 For ataluren, a benzoic acid derivative, 1% BSA was the optimal concentration, which is in agreement with Gill et al that 1% BSA was optimal for acids and 2% was the best for base/neutral. 26 These results indicate that BSA titrations may be necessary for compounds with low depletion and high protein binding.
Ataluren did not inhibit enzyme activity for CYPs 1A2, 2B6, were all close to unity in human liver microsomes at 0.1 mg/mL protein concentration (internal data). Therefore, the unbound inhibition constants (K i,u ) are approximately equal to the respective total inhibition constants for both CYP2C8 and CYP2C9 under current incubation conditions. Following oral therapeutic doses of 10, 10, 20 mg/kg/day (morning, mid-day and evening dose, respectively) in DMD patients, the mean C max of ataluren at the steady state is around 79.9 μmol/L. Since the protein binding of ataluren in human plasma is high (99.6%), the free C max of ataluren at the steady state is around 0.32 µmol/L (using free fraction value of 0.4% as measured) or 0.80 µmol/L (to be conservative assuming free fraction value of 1%), and thus is much lower than K i,u values, 169 µmol/L for CYP2C8 and 75.4 µmol/L for CYP2C9. These data indicate that ataluren will likely not exhibit any clinically relevant effect due to CYP inhibition.
At concentrations up to 400 µmol/L in cultured primary human hepatocytes, ataluren did not induce the activities of CYPs 1A2, 2A6, 2C8, 2C19, 2E1 and 3A4, except for a mild 1.7-and 1.5-fold induction over vehicle control at 400 µmol/L for CYP2B6 and CYP2C9 activity, respectively. This induction is considered not significant or clinically relevant, and thus, ataluren interaction with other concomitantly administered drugs due to enzyme induction is less likely.
In summary, results of in vitro metabolism studies showed that CYP-mediated metabolism of ataluren is minimal, whereas acyl glucuronidation is the major metabolic pathway. UGT1A9 is the major enzyme that catalyzes glucuronidation of ataluren in liver and kidney. Enzyme kinetic studies indicated that in addition to the liver, the kidney is also a key organ for the metabolism and disposition of ataluren in humans, whereas other UGTs may be also involved in this reaction in the intestine. Ataluren showed no or little potential to inhibit or induce most of the CYP enzymes.

ACK N OWLED G EM ENTS
We thank Danielle Duva from PTC Therapeutics, Inc, and Kinapse Ltd for their involvement with the compilation of this manuscript.

D I SCLOS U R E
The study was sponsored by PTC Therapeutics. Authors are current employees of PTC Therapeutics and may hold stock or other equity positions with the company.

E TH I C S S TATEM ENT
All studies were conducted in accordance with all applicable ethical requirements.

DATA R E P O S I TO RY
Supplementary tables are listed in Appendix.