Discovery of MGS0274, an ester prodrug of a metabotropic glutamate receptor 2/3 agonist with improved oral bioavailability

Highlights

• MGS0008, an mGlu2/3 receptor agonist, was expected to show low oral bioavailability (BA) in humans.

• The prodrug 4h exhibited a 15-fold improvement in oral BA compared to MGS0008 in a monkey pharmacokinetic study.

• The prodrug was scarcely detected in the plasma.

• The besylate salt of compound 4j (MGS0274), a diastereomer of 4h, is under development for the treatment of schizophrenia.

Abstract

We previously reported that MGS0008 is a selective group II metabotropic glutamate receptor (mGlu2/3 receptor) agonist that is effective in animal models of schizophrenia. MGS0008 is a highly hydrophilic glutamate analog and is therefore expected to show low oral bioavailability in humans. To improve the oral bioavailability of MGS0008, ester prodrugs of MGS0008 were synthesized and their usefulness was evaluated. Among the prodrugs, the l-menthol-ester prodrug 4h demonstrated preferable lipophilicity, good chemical stability, and a high conversion rate to MGS0008 in human and monkey liver microsomes. A pharmacokinetic study in monkeys revealed that the oral bioavailability of MGS0008 after oral dosing of compound 4h was approximately 15-fold higher than that after oral dosing of MGS0008. Based on these findings, a diastereomer of compound 4h (compound 4j, or MGS0274), was selected as a candidate for clinical drug development, and its besylate is currently under development for the treatment of schizophrenia (Development code: TS-134).

Introduction

Schizophrenia is a devastating mental illness that affects approximately 1% of the global population. Its symptoms fall into three categories: positive, negative, and cognitive. While positive symptoms respond, at least to some extent, to antipsychotics, negative symptoms and cognitive dysfunction show poor responses to currently available medications, most of which have been generated based on the “dopamine hypothesis” [1]. Therefore, novel antipsychotics developed using an alternate hypothesis are needed to fulfill the existing medical needs of schizophrenia patients.

Increasingly, evidence suggests that hypofunction of the N-methyl-d-aspartate (NMDA) receptor expressed on GABAergic interneurons in the medial prefrontal cortex may play an important role in the pathophysiology of schizophrenia. Hypofunction of the NMDA receptor results in the disinhibition of pyramidal neurons, leading to increased glutamate release in the medial prefrontal cortex [2]. Since metabotropic glutamate (mGlu) 2 and mGlu3 receptors negatively regulate glutamate release as autoreceptors [3], activation of mGlu2 and mGlu3 receptors may be an attractive strategy for the treatment of disorders associated with excessive glutamatergic neurotransmission, such as schizophrenia [4,5]. Indeed, in a phase II trial, LY2140023 (a methionine amide prodrug of the mGlu2/3 receptor agonist, LY404039), developed by Eli Lilly, demonstrated efficacy in patients with schizophrenia (see Fig. 1 for chemical structures) [6]. Although its efficacy could not be confirmed in subsequent phase II and III clinical trials, a post hoc analysis of the data suggests that LY2140023 was effective in patients treated within the first three years of onset (early-in-disease) and those previously treated with drugs having predominantly dopamine D2 receptor antagonist activity [7]. These post hoc analyses are consistent with higher levels of glutamate measured in the brains of ultra-high-risk schizophrenia patients and early-in-disease schizophrenia patients than in control subjects [8]. Therefore, activation of mGlu2 and mGlu3 receptors provides a novel therapeutic target, at least for subgroups of schizophrenia patients, particularly those with excessive glutamate neurotransmission.

We previously reported MGS0008 as a selective mGlu2/3 receptor agonist [9]. The chemical structure of MGS0008 is shown in Fig. 2. In vitro profiles of the affinities of MGS0008 for the mGlu2/3 receptors are shown in Table 1, and their affinities for other mGlu receptor subtypes are shown in Table 2. MGS0008 is expected to show poor oral bioavailability in humans on the basis of its similarity to LY404039, a glutamate analog with low human oral bioavailability (6%), owing to its low membrane permeability arising from its high hydrophilicity [10]. Therefore, we endeavored to improve MGS0008 oral bioavailability in humans by transforming the compound into a prodrug.

To date, numerous prodrugs have been developed to improve the oral bioavailability of parent compounds [11]. Prodrugs are formed by attaching a small modifying group, such as an alkyl or acyl group, to a reactive substituent of the parent compound, such as a carboxyl group or amino group.

LY2140023, a methionine amide prodrug of LY404039, is designed to be absorbed via intestinal peptide transporter 1 and then hydrolyzed by peptidases, thereby enhancing the oral bioavailability of LY404039 [10,12,13]. Although LY2140023 improved the oral bioavailability of LY404039 in humans, the prodrug was still detected in plasma and the area under the concentration-time curve was approximately 30% of that of LY404039. To achieve presystemic and nearly complete conversion to the parent compound after gastrointestinal absorption in humans, we prepared an ester prodrug of MGS0008.

We prepared the ester prodrug by esterification of the hydrophilic carboxylic group of MGS0008 in order to increase lipophilicity; log D (pH 6.8) values were then measured. To estimate the stability of the prodrug in the stomach and small intestine, stability tests were performed in buffer solutions with pH values of 1.2 and 6.5 to mimic gastric juice and intestinal fluid, respectively. The rates of conversion to the parent compound were evaluated in both human and monkey liver microsomes containing carboxylesterases (CESs) [14]. The liver is considered to be a major site of presystemic hydrolysis of many ester-prodrugs, mainly by CESs, and their activities are reported to be relatively close between humans and monkeys [15]. Finally, the bioavailability of MGS0008 after oral administration of the selected prodrug candidates was evaluated in monkeys for an appropriate prediction of human pharmacokinetics, since tissue distribution patterns of CESs in monkeys are considered to be relatively close to those of humans [16]. As a result, we identified a diastereomeric compound, 4h, that exhibited a 15-fold improvement in oral bioavailability over the parent compound, MGS0008, and achieved complete conversion of the prodrug into the parent compound. Based on these findings, a diastereomer of compound 4h (compound 4j) was selected as a promising candidate for clinical drug development.