Synthesis and biological evaluation of novel N-substituted nipecotic acid derivatives with a cis-alkene spacer as GABA uptake inhibitors

Highlights

• New, additional cis-alkene derivatives of the lead compound DDPM-1457.

• Simple and selective synthesis of cis-alkene products via catalytic hydrogenation.

• The subtype selectivity of the new compounds generally favors mGAT4.

• Variation in the spacer length led to a surprisingly high potency at mGAT1.

Abstract

To discover new, potent, and selective inhibitors for the murine gamma-aminobutyric acid transporter 4 (mGAT4), the structure-activity relationship (SAR) study of a new cis-alkene analog family based on DDPM-1457 [(S)-2], which previously showed promising inhibitory potency at and subtype selectivity for mGAT4, was conducted. To uncover the importance of the differences between the trans- and the cis-alkene moiety in the spacer, the present publication describes the synthesis of the new compounds via catalytic hydrogenation with Lindlar’s catalyst. The biological results collected by the SAR study revealed that analog rac-7j characterized by a four-instead of a three-carbon atom spacer with a cis double bond applying to the majority of the studied compounds displays a surprisingly high potency at mGAT1 (pIC₅₀ = 6.00 ± 0.04) and at the same time a reasonable potency at mGAT4 (pIC₅₀ = 4.82).

Introduction

Disorders in the GABAergic neurotransmission can cause severe neurological illnesses like Alzheimer’s disease,¹ depression,² epilepsy,3, 4 and neuropathic pain.⁵ These conditions are closely related to the levels of gamma-aminobutyric acid (GABA), the major inhibitory neurotransmitter in the central nervous system (CNS).⁶ GABA concentration in the synaptic cleft, amongst other factors, is regulated by the GABA transporter proteins (GATs).⁷ Four different subtypes of these membrane-bound proteins exist,8, 9 which belong to the solute carrier 6 (SLC6) family.10, 11 Depending on the species they are cloned from, different nomenclatures can be applied for these transporters. The Human Genome Organization (HUGO) denotes them as GAT1, BGT-1, GAT2, and GAT3. Alternatively, if the transporters are originating from mice, they are termed as mGAT1 (GAT1), mGAT2 (BGT-1), mGAT3 (GAT2), and mGAT4 (GAT3).¹² mGAT1 and mGAT4 have been found to be clearly predominating in the CNS,¹³ from which mGAT1 is mainly accountable for the neuronal uptake of GABA in presynaptic cells and mGAT4, in particular, mediates GABA transport from the synaptic cleft into the adjacent glial cells.14, 15 The next two subtypes (mGAT2–mGAT3) are mainly observed in the kidneys and liver¹⁶ and are playing no significant role in the termination of the GABAergic neurotransmission in the brain.¹⁷

The inhibition of mGAT1 and mGAT4 leads to elevated GABA concentrations in the synaptic cleft, which can be used as a treatment option in the above mentioned diverse neurological diseases. The highly potent and selective mGAT1 inhibitor tiagabine (5) is already in medical use, but its main drawbacks are the commonly observed side effects (asthenia, depression, diarrhea, dizziness, nervousness, and tremor).18, 19, 20 Additionally to mGAT1 selective reuptake inhibitors, a large selection of ligands for mGAT2–mGAT4 were identified over the last years. However, these compounds in general display only mediocre affinities and selectivities for their target.21, 22, 23, 24, 25 Hence, there is a need for more potent and subtype-selective GAT inhibitors on the one hand for mGAT2–mGAT4, but also for mGAT1. This would allow a more in-depth study of the physiological role of these proteins, that could serve as alternative treatment options for tiagabine (5), which might give rise to fewer side effects. (S)-SNAP-5114 [(S)-1] was the first prototypic mGAT4 inhibitor with reasonable potency at and selectivity for this target.²⁶ Based on the structure of (S)-1, carba-analogs such as DDPM-1457 [(S)-2]²⁷ were developed, the latter of which displays a similar potency and subtype selectivity for mGAT4 as (S)-SNAP-5114 [(S)-1], and, in addition, a significantly enhanced chemical stability. Later on, a series of substances including compound 3 (Fig. 1) which is similar to (S)-2, but with an alkyne spacer instead of a trans-alkene moiety were also synthesized. These compounds, however, showed significantly lower potencies at the mGATs as compared to (S)-2.²⁸ More recently, the compound family represented by a trans-alkene spacer (S)-2 was expanded with analogs (4, Fig. 1) by a variation of the triaryl moiety, i.e. by substituting one of the three aryl rings by a variety of different substituents (Table 1).²⁹ Finally, as a representative of a new class of hGAT3 (mGAT4) inhibitors isatin derivative 6 is to be mentioned which represents the most potent compound of this set of inhibitors.³⁰

In this study, we aimed at the development of an additional carba-analog family of (S)-SNAP-5114 [(S)-1] with a cis-configured alkene spacer to supplement the already published results regarding structure–activity relationship (SAR) of the alkyne (3, Fig. 1) and the trans-alkene analogs ((S)-2 and 4, Fig. 1) and to possibly identify more potent and selective inhibitors for mGAT4. The structure of this new cis-alkene analog family is related to DDPM-1007 (rac-2), the racemic form of DDPM-1457 [(S)-2], and the applied modifications are shown in Fig. 2. On the one hand, the spacer between the nipecotic acid and the aromatic lipophilic residue should be modified by replacing the trans-alkene moiety by a cis-alkene unit. In addition, as a major modification, one of the aromatic moieties of the lipophilic triarylmethane unit should be replaced with a series of different residues, such as aromatic and heteroaromatic rings, benzylic residues or sterically less demanding groups (rac-7a–h, Table 1). Finally, we intended to increase the spacer length by one methylene group either by insertion of this unit between the cis-alkene group and the lipophilic residue (rac-7i–j, Table 1) or between the nipecotic acid and the cis-alkene moiety (rac-7k–l, Table 1). This should uncover which linker length would be most beneficial regarding biological activity.