Discovery of oxazolidinone-based heterocycles as subtype selective sigma-2 ligands

The sigma-2 (σ2) receptor, also known as the Transmembrane Protein 97 (TMEM97, and MAC30 (Meningioma-associated protein), has been linked to a number of conditions are disease states such as schizophrenia, cancer, Alzheimer’s disease, traumatic brain injury, and neuropathic pain. As part or our on-going effort identify novel σ2 ligands, we have identified a series of novel, functionalized oxazolidin-2-one sigma-2 ligands (4). Our lead compound (4h) demonstrated high affinity (Ki = 36 nM) and excellent σ1/σ2 selectivity (79-fold). Evaluation of its affinity at key CNS targets via the Psychoactive Drug Screening Program (PDSP) also indicated a high degree of selectivity for σ2 over other receptors. Graphical Abstract Graphical Abstract


Introduction
The identification and characterization of the sigma receptors began in 1976 when Martin et. al. described their attempts to classify opioids based on pharmacological responses produced in chronic spinal dogs. Morphine (1), ketocyclazocine, (2), and SKF-100047 (3) (Fig. 1) each produced different results in this animal model. They proposed that each compound engaged a different receptor that they labeled the µopioid receptor (morphine type, MOR), the κ-opioid receptor (ketocyclazocine type, KOR), and the σ-opioid receptor (SKF-100047 like) [1]. These studies were conducted with racemic SKF-100047, and in the early 1980s, follow-up studies with (−)-SKF-100047 and (+)-SKF-100047 demonstrated that each enantiomer engaged different targets. While the (−)-SKF-100047 elicited opioid type responses through MOR and KOR, (+)-SKF-100047's pharmacological activity was mediated by a previously unknown, nonopioid receptor that was designated the sigma receptor (σR) [2,3]. It was later demonstrated by that there were two subtypes of this receptor, sigma-1 (σ 1 ) and sigma-2 (σ 2 ) [4]. The mammalian σ 1 receptor was cloned and expressed in yeast cells in 1976 [5], and a crystal structure of the human σ 1 was reported in 2016 [6], To date, however, there is no known ligand for this receptor.
Characterization of σ 2 was significantly more challenging. The true nature of this protein was not determined until 2017, over 40 years after the original discovery of the sigma receptors, when it was demonstrated that σ 2 is Transmembrane Protein 97 (TMEM97, also known as MAC30 (Meningioma-associated protein) [7]. There are no known natural functional ligands for σ 2 , but this protein is present in the endoplasmic reticulum and lysosomes where it binds to cholesterol [8]. Although the role of this protein in physiological processes has as yet to be determined, σ 2 has been linked to numerous disease states and conditions such as Niemann-Pick disease [9], schizophrenia [10], Alzheimer's disease [11][12][13], neuropathic pain [14], traumatic brain injury [15], and cancer [16,17]. Efforts to develop novel therapeutic agents based on these links has led to the identification of numerous, highly potent σ 2 ligands (Fig. 2). Many have in vivo efficacy in animal models, and some have advanced to human clinical trials. Siramesine (4, σ 2 K i = 0.12 nM), for example, was demonstrated to have antidepressant and anxiolytic properties [18], while UKH-1114 (5, σ 2 K i = 46 nM) and CB-184 (6, σ 2 K i = 13.4 nM) are both efficacious in animal models of pain [19,20]. Agents that are the subject of clinical investigation include the radioligand [ 18 F]ISO-1 (7, σ 2 K i = 6.9 nM) which is being examined as a PET ligand in breast cancer [21] and CT1812 (8, σ 2 K i = 8.5 nM), a potential Alzheimer's disease therapy [22]. Homology models [23,24] have been developed to facilitate the discovery of novel σ 2 ligands and the recent disclosure of ligand bound crystal structures (Fig. 3) [25] should further accelerate these efforts.
As part of an effort to identify novel, biologically active small molecules, we have been exploring the chemical space associated with a series of 5-(piperazin-1-ylmethyl) oxazolidin-2-one (9) (Fig. 2). Preliminary assessment of initial compounds prepared in this series by the Psychoactive Drug Screening Program (PDSP) demonstrated that members of this family are potent σ 2 binders with varying levels of σ 2 /σ 1 selectivity. Given the therapeutic potential of this receptor, a drug discovery program focused on the identification of novel, drug-like functionalized 5-(piperazin-1-ylmethyl)oxazolidin-2-ones with potent σ 2 binding was established. The synthesis, characterization and preliminary evaluation of these compounds as potential selective σ 2 ligands will be presented.

Results and discussion
Synthesis of the 5-(piperazin-1-ylmethyl)oxazolidin-2-ones was conducted using the methods described in Schemes 1 and 2. The unsubstituted oxazolidin-2-one (9a) was prepared from the known primary chloride (10). Conversion of this starting material to the corresponding iodide using NaI in acetone was followed by displacement of the iodide with (11) to provide (9a). Synthesis of the remaining compounds (9b-9q) began with either epibromohydrin (12) or glycidol (13). Reaction of (12) with an amine (14) in methanolic K 2 CO 3 provided oxazolidin-2-one (15). Alternatively, reaction of (13) with an amine (14), followed by ring closure with diethyl carbonate in the presence of sodium methoxide provided (15). Conversion of (15) to the corresponding tosylate (pTosCl, NEt 3 , THF), followed by displacement of the leaving group with an amine (11) provided the final target compounds (9b-9q). Table 1 includes the in vitro binding (K i at σ 2 and σ 1 ) as well as the physicochemical properties (MW, TPSA, LogP) of target compounds. The compounds prepared and tested have TPSA and MW values that suggest they will cross the BBB, and all but three cLogP values are higher than the desired range of 2-4 for BBB penetration. The structure-activity relationship began with the unsubstituted oxazolidin-2-one (9a), which demonstrated minimal affinity for σ 1 and σ 2 (K i = 10,000 nM at both σ 1 and σ 2 ). Capping of the amide with a 3 to 6 membered cycloalkane (9b-9e) or a benzene ring (9f) produced compounds with moderate σ 2 binding affinity (K i = 116-530 nM) that had limited capacity to bind to σ 1 (K i = 10,000 nM). Interestingly, the impact of a benzyl substituent on σ 2 affinity was highly dependent on the electronic character of the benzene ring. In the absence of a substituent, the benzyl derivative (9g) is approximately twice as potent at σ 2 (K i = 91 nM) as the corresponding phenyl derivative (9f), and the high degree of selectivity over σ 1 is maintained (K i = 10,000 nM). σ 2 affinity increased three-fold when a 4-F substituent was appended to the benzene ring (9h, σ 2 K i = 36 nM), and high degree of selectivity over σ 1 (K i = 2847 nM) was maintained. In contrast, incorporation of the electron donating 4-OMe (9i) or 4-Me (9j) produced compounds that had limited capacity to bind both sigma receptors (σ 2 and σ 1 K i = 10,000 nM). Replacing the benzene ring of (9g) with either a cyclohexane ring (9k) or tetrahydropyran ring (9l) also lead to a substantial drop in σ 2  affinity (K i = 1379 and 2428 nM, respectively). σ 2 binding affinity and selectivity over σ 1 were maintained when the benzyl group (9g) was extended by an additional methylene (9m, σ 2 K i = 49 nM, σ 1 K i = 10,000 nM), but incorporation of both electron withdrawing (9n) and electron donating substituents (9o, 9p) led to a significant loss in σ 2 binding affinity, as did the addition of a third methylene unit (9q).

Conclusion
In summary, we have identified a series of 5-(piperazin-1ylmethyl)oxazolidin-2-ones (9) σ 2 ligand that are highly selective for this receptor over σ 1 . In addition, we have established preliminary SAR that could be used to extend our exploration of this series and identified a novel compound (9h) that is selective for σ 2 over a range of key CNS targets. Future efforts are focused on determining the in vitro ADME Scheme 1 Synthesis of (9a) Scheme 2 Synthesis of (9b)-(9q)  [26]. (2-oxo-3phenyloxazolidin-5-yl)methyl-4-methylbenzenesulfonate was prepared according to previously published methods [27].
General procedure for the synthesis of toslyates from epibromohydrine and an amine To a suspension of K 2 CO 3 (2.5 g, 18.08 mmol) in anhydrous methanol (30 ml) containing epibromohydrin (4.965 g, 36.24 mmol) was added an amine (36.12 mmol) and the reaction was stirred overnight. The reaction mixture was then filtered and the organic solvent was removed under reduced pressure to obtain a liquid residue that was utilized for the tosylation step without further purification.
p-Toluene sulfonyl chloride (13.5 g, 7.096 mmol) in methylene chloride (50 ml) was added dropwise to a 0°C solution of the crude alcohol (35.52 mmol) and triethylamine (7.1874 g, 71.028 mmol) in methylene chloride (60 ml). The reaction was stirred at 0°C for 1 h followed by overnight stirring at room temperature. It was then quenched with ice water and the organic layer was washed successfully with 10% HCl, saturated aqueous sodium bicarbonate and brine. The organic layer was dried over MgSO 4 and concentrated under reduced pressure to obtain a thick oil which was Table 2 Off-target assessment of (9h) at key CNS targets BZP-B benzodiazepine brain receptor, BZP-P benzodiazepine peripheral receptor, DAT dopamine transporter, DOP δ-opioid receptor, KOP κopioid receptor, MOP µ-opioid receptor, NET norepinephrine transporter, SERT serotonin transporter purified on a silica gel column using hexane: ethyl acetate (0-100%) to obtain the desired product.
Yield radioligand binding observed in the absence of competitor. The log IC 50 (i.e., the log of the ligand concentration that reduces radioligand binding by 50%) is thus estimated from the data and used to obtain the K i by applying the Cheng-Prusoff approximation: where [ligand] equals the assay radioligand concentration and KD equals the affinity constant of the radioligand for the target receptor [28].
Radiolabel binding studies for the sigma-1 receptor Binding assays to determine test compound affinity at σ 1 was determined using the methods described for σ 2 using membrane fractions of cells expressing recombinant σ 1 . The radioligand employed was 3 H-DTG and the control compound was Haloperidol [28].