Specific AIM

To identify small molecule positive allosteric modulators and/or allosteric agonists of the M4 muscarinic acetylcholine receptor that are cell permeable, possess submicromolar potency and show greater than 10-fold selectivity over the other mAChRs (M1, M2, M3 and M5) employing a functional HTS approach.

Significance

To date, five muscarinic acetylcholine receptor (mAChR) subtypes have been identified (M1–M5) and play important roles in mediating the actions of ACh in the peripheral and central nervous systems. (1) Of these, M1 and M4 are the most heavily expressed in the CNS and represent attractive therapeutic targets for cognition, Alzheimer’s disease, and schizophrenia. (2–4) In contrast, the adverse effects of cholinergic agents are thought to be primarily due to activation of peripheral M2 and M3 mAChRs. (5,6) Due to the high sequence homology and conservation of the orthosteric ACh binding site among the mAChR subtypes, development of chemical agents that are selective for a single subtype has been largely unsuccessful, and in the absence of highly selective activators of M4, it has been impossible to test the role of selective M4 activation. Clinical trials with xanomeline, a M1/M4-preferring orthosteric agonist, demonstrated efficacy as both a cognition-enhancing agent and an antipsychotic agent. (7–9) In follow-up studies in rats, xanomeline displayed an antipsychotic-like profile comparable to clozapine. (10) However, a long standing question concerned whether or not the antipsychotic efficacy or antipsychotic-like activity in animal models is mediated by activation of M1, M4, or a combination of both receptors. Data from mAChR knockout mice led to the suggestion that a selective M1 agonist would be beneficial for cognition, whereas an M4 agonist would provide antipsychotic activity for the treatment of schizophrenia. (5,6,11) This proposal is further supported by recent studies demonstrating that M4 receptors modulate the dynamics of cholinergic and dopaminergic neurotransmission and that loss of M4 function results in a state of dopamine hyperfunction. (12) These data, coupled with findings that schizophrenic patients have altered hippocampal M4 but not M1 receptor expression, (13) suggest that selective activators of M4 may provide a novel treatment strategy for schizophrenia patients. However, multiple studies suggest that M1 may also play an important role in the antipsychotic effects of mAChR agonists and that the relative contributions of M1 and M4 to the antipsychotic efficacy of xanomeline or antipsychotic-like effects of this compound in animal models are not known. However, highly selective centrally penetrant activators of either M1 or M4 have not been available, making it impossible to determine the in vivo effects of selective activation of these receptors. Only recently did we develop a highly selective M1 allosteric agonist probe (CID 25010775) to study the role of selective M1 activation in vitro and in vivo; however, no such probe existed for M4.

Rationale

In recent years, major advances have been made in the discovery of highly selective agonists of other GPCRs that act at an allosteric site rather than the orthosteric site, (14) as well as positive allosteric modulators (PAMs). (15,16) By screening for compounds that act at an allosteric site on the receptor, it is anticipated that compounds that can selectively activate the M4 receptor versus the other mAChR subtypes may be identified. In conjunction with our MLPCN M1 allosteric agonist probe (CID 25010775), development of a potent and selective M4 allosteric agonist or PAM will enable the biomedical community to dissect the pharmacology of xanomeline and determine the pharmacology and therapeutic potential of selective M1 and M4 activation.

PubChem Bioassay Name

Discovery of Novel Allosteric Agonists of the M4 Muscarinic Receptor: Primary Screen

List of PubChem bioassay identifiers generated for this screening project (AIDs)

AID-1921, AID-1923, AID-1928, AID-1929, AID-1930, AID-1932, AID-1938, AID-1939, AID-2616, AID-626, AID-643

PubChem Primary Assay Description

Chinese hamster ovary (CHO K1) cells stably expressing rat (r)M1 were purchased from the American Type Culture Collection (ATCC, Manassas, VA) and cultured according to their recommendations. CHO cells stably expressing human (h) M2, hM3, and hM5 were generously provided by A. Levey (Emory University, Atlanta, GA); rM4 cDNA provided by T. I. Bonner (National Institutes of Health, Bethesda, MD) was used to stably transfect CHO-K1 cells purchased from the ATCC using Lipofectamine 2000. To make stable hM2 and rM4 cell lines for use in calcium mobilization assays, cell lines were cotransfected with a chimeric G protein (G_qi5) using Lipofectamine 2000. rM2, hM3, and hM5 cells were grown in Ham’s F-12 medium containing 10% heat-inactivated fetal bovine serum, 2 mM GlutaMax I, 20 mM HEPES, and 50 μg/ml G418 sulfate. hM2-Gqi5 cells were grown in the same medium supplemented with 500 μg/ml hygromycin B. Stable rM4 cells were grown in Dulbecco’s modified Eagle’s medium containing 10% heat-inactivated fetal bovine serum, 2 mM GlutaMax I, 1 mM sodium pyruvate, 0.1 mM nonessential amino acids, 20 mM HEPES, and 400 μg/ml G418 sulfate; rM4-G_qi5 cells were grown in the same medium supplemented with 500 μg/ml hygromycin B. CHO cells stably expressing rM1, hM3, or hM5 were plated at a seeding density of 50,000 cells/100 μl/well. CHO cells stably coexpressing hM2/G_qi5 and ratM4/G_qi5 were plated at a seeding density of 60,000 cells/100 μl/well. For calcium mobilization, cells were incubated in antibiotic-free medium overnight at 37°C/5% CO₂ and assayed the next day.

Calcium Mobilization Assay

Cells were loaded with calcium indicator dye [2 μM Fluo-4 acetoxymethyl ester (50 μl/well) prepared as a stock in DMSO and mixed in a 1:1 ratio with 10% Pluronic acid F-127 in assay buffer (1×Hanks’ balanced salt solution supplemented with 20 mM HEPES and 2.5 mM probenecid, pH 7.4)] for 45 min at 37°C. Dye was removed and replaced with the appropriate volume of assay buffer. All compounds were serially diluted in assay buffer for a final 2× stock in 0.6% DMSO. This stock was then added to the assay plate for a final DMSO concentration of 0.3%. Acetylcholine (EC₂₀ concentration or full dose-response curve) was prepared at a 10× stock solution in assay buffer before addition to assay plates. Calcium mobilization was measured at 25°C using a FLEXstation II (Molecular Devices, Sunnyvale, CA). Cells were preincubated with test compound (or vehicle) for 1.5 min before the addition of the agonist, acetylcholine. Cells were then stimulated for 50 s with a submaximal concentration (EC₂₀) or a full dose-response curve of acetylcholine. The signal amplitude was first normalized to baseline and then as a percentage of the maximal response to acetylcholine.

Summary of Screen

This screen was performed in the pilot phase, the MLSCN, when the MLSMR compound collection at Vanderbilt only contained 12,364 compounds. From the primary M4 screen of 12,364 compounds, 69 putative M4 activators were identified with an average Z′ score of 0.67±0.093. The confirmation screen (singles at 10 μM) produced 25 active compounds. After a selectivity screen versus M1, only 4 compounds (Figure 1) appeared to be M4 selective. Chemistry was pursued around these four screening hits 1–4, and 48 analogs were synthesized (25 analogs of 1–3 and 23 analogs of 4); unfortunately, none of these confirmed as selective M4 agonist or PAMs.

Figure 1

M4 HTS hits.

The project was shelved until a preliminary report from Eli Lilly (C.C. Felder, Eli Lilly, personal communication) showed that a new compound, LY2033298, was a robust, but weak, M4 PAM that is highly selective for human M4 (Figure 2). However, to study selective M4 activation in a preclinical, academic setting, we, and the biomedical community at large, require a potent and selective M4 for rat M4. Therefore, we initiated a cheminformatics and database mining effort based on the LY2033298 scaffold, to attempt to deliver a useful M4 PAM probe for the MLPCN Network that would possess appropriate potency (rM4 EC₅₀ < 500 nM), and be free from patent issues. (16) A substructure search for commercially available compounds in the ChemBridge Corporation’s chemical database containing a core similar to LY2033298 delivered CID 714286 and related analogs, which were free of intellectual property issues. CID 714286 was a potent PAM of rat M4 (EC₅₀ = 400 nM) which induced a 47-fold leftward shift in the ACh CRC. Moreover, CID 714286 was highly selective for rat M4 (EC₅₀ > 50 μM for rM1, hM2, hM3 and hM5), binds at an allosteric site on the M4 receptor, increases affinity for ACh and increases coupling to G proteins. (17) However, CID 714286 possessed a high log P (4.6) and was virtually insoluble in anything except DMSO. (17)

Figure 2

LY2033298 and cheminformatics and database mining approach leading to CID 714286.

Probe Chemical Lead Optimization Strategy

For the chemical lead optimization of CID 714286, we employed a parallel synthesis approach to rapidly explore alternative amide moieties (Figure 3). Readily available acid 5 was treated with polymer-supported DCC, HOBt and one of 24 different benzyl or heteroarylmethyl amines, followed by mass-directed preparative HPLC purification to deliver 24 analogs 6 of CID 714286. (18) Clear SAR was observed for this set of analogs, with EC₅₀s for rM4 potentiation ranging from 370 nM to >100 μM, and with fold-shifts ranging from 3.4- to 70-fold (Table 1). Within this library, CID 713286, upon re-synthesis, displayed an EC₅₀ of 339 nM and 29-fold leftward shift of the ACh CRC. Attempts to install a basic nitrogen atom in the form of pyridyl analogs CID 44176129 and CID 20417337 to improve solubility led a diminution in rM4 potency (rM4 EC₅₀s of 2.82 μM and 5.59 μM, respectively). Ultimately, ether-containing analogs such as CID 864492 (rM4EC₅₀ = 390 nM, 70-fold shift) and CID 1541501 (rM4EC₅₀ = 369 nM, 29-fold shift) emerged as attractive probe candidates, as they met the potency requirements for an MLPCN probe. Moreover, the log P values for both CID 864492 and CID 1541501 were measured as 3.4 – a full order of magnitude lower than the initial lead CID 714286. (18) This reduced lipophilicity directly translated into improved solubility across a panel of 22 pharmaceutically acceptable vehicles – including aqueous vehicles such as 20% β-cyclodextrin, wherein both provided homogeneous solutions at 10 mg/mL. (18)

Figure 3

Optimization of CID 7142z86.

Table 1

Structures and SAR for Analogs of CID714286.

Upon further study, both CID 864492 and CID 1541501 were found to be potent rM4 positive allosteric modulators, but due to the larger fold-shift observed with CID 864492 and the lack of significant ancillary pharmacology (Table 2), we pursued CID 864492 as a potential MLPCN probe. As shown in Figure 4A, CID 864462 failed to elicit rM4 receptor activation in the absence of ACh (Flat line in A); however, when a submaximal (EC₂₀) concentration of ACh is applied, concentration dependent rM4 activation is observed with an EC₅₀ of 390 nM. CID 864492 potentiates the response of rM4 to ACh, as manifested by a dose-dependent leftward shift of the ACh CRC. At the highest concentration tested (10 μM), CID 864492 induced a 70-fold shift of the ACh CRC. CID 864492 was also found to be highly selective for rM4 versus rM1, hM2, hM3 and hM5, as no shift in the ACh CRC was observed with 30 μM CID 864492 (Figure 5A).(18)

Table 2

MDS Lead Profiling Screen for CID 1541501 and CID 864492.

Figure 4

A) rM4 CRC in presence and absence of ACh EC₂₀; B) ACh Fold-shift assay with CID 864492.

Figure 5

A) mAChR selectivity of CID 864492 versus M1, M2, M3 and M5 at 30 μM; B) [³H]-NMS competition binding experiments with the orthosteric ligand atropine and CID 864492 and CID1541501.

Figure 5B shows that neither CID 864492 nor CID 1541501 displace the mAChR orthosteric ligand [³H]-NMS in radioligand competition binding experiments (0.1 nM, [³H-NMS]), further supporting the allosteric mode of receptor activation, since the orthosteric antagonist readily competes for [³H-NMS] with a K_i of 0.54 nM. (17)

Due to the unprecedented mAChR and ancillary target selectivity, we felt that CID 864492, while a valuable in vitro probe to study selective rM4 activation, would be more valuable to the translational biomedical community if it were also an in vivo probe for rM4 activation like our earlier M1 probes. Thus, we performed a brain/plasma study to determine if our rM4 PAM probe (CID 864492) was centrally penetrant after systemic dosing. When dosed at 56.6 mg/kg i.p. to male rats, CID 864492 displayed a 0.87 brain_AUC/plasma_AUC ratio (Figure 6) with comparable plasma t_1/2 (1.6 hr) to brain t_1/2 (1.25 hr). CID 864492 was cleared from plasma and brain after 4 hours with levels BLQ. The brain AUC far exceeded the 390 nM EC₅₀ for rM4 activation. This is an excellent profile for a CNS agent. Importantly, the animals were closely monitored, and were healthy, with no signs of classical pan-mAChR activation (SLUD – salivation, lacramation, urination and defecation), indicating that the in vitro selectivity profile was mirrored in vivo. (18)

Figure 6

Brain/plasma study with CID 864492 when dosed i.p. at 56.6 mg/kg.

As an ongoing debate in the field has concerned the antipsychotic potential of selective M4 activation, especially in light of the recent positive PhII data in schizophrenics with xanomeline, an M1/M4 preferring agonist,, we studied CID 864492 in a preclinical antipsychotic model (amphetamine-induced hyperlocomotion) were all known atypical and typical antipsychotic agents show efficacy. In the event, we dosed both CID 864492 and CID 1541501 at 56.6 mg/kg (the same dose that provided excellent brain levels in the brain/plasma study), and found that both compounds inhibit amphetamine-induced hyperlocomtion (Figure 7). In confirm efficacy, we ran, in parallel, a rotorod experiment with CID 864492 and CID 1541501 at doses of 30, 56.6 and 100 mg/kg, and observed no sedation. Thus, CID 864492, a selective rM4 positive allosteric modulator, demonstrated that selective M4 activation has therapeutic potential for a novel mechanism for the treatment of schizophrenia. Thus, CID 864492 has utility as both an in vitro and in vivo probe for selective rat M4 activation by an alloosteric mechanism, positive allosteric modulation (PAM). (18) This probe nicely compliments other highly selective M1 MPLCN probes: CID 25010775, an M1 allosteric agonist (19) and CID 24768606, a highly selective M1 antagonist. (20)

Figure 7

Reversal of amphetamine-induced hyperlocomotion with CID 864492 and CID 1541501.

Synthetic procedure (large scale) and spectral data for CID 864492

MLS#s

002474502 (CID 864492, 500 mg), 002474499, 002474500, 002474501

Solubility

Solubility in PBS (at pH = 7.4) for ML108 was 0.78 μM.

Stability

Stability (at room temperature = 23 °C) for ML108 in PBS (no antioxidants or other protectorants and DMSO concentration below 0.1%) is shown in the table below. After 48 hours, the percent of parent compound remaining was 102%, but the assay variability over the course of the experiment ranged from a low of 93% (at 1 hour) to a high of 102% (at 48 hours).

Reactivity

As assessed through a glutathione (GSH) trapping experiment in phosphate buffered saline (with a substrate concentration of typically 5–50 μM and a GSH concentration of 5 mM, at t = 60 minutes) ML108 was found to not form any detectable GSH adducts.^*