Restoring Agonist Function at a Chemogenetically Modified M₁ Muscarinic Acetylcholine Receptor

Abstract

Designer receptors exclusively activated by designer drugs (DREADDs) have been successfully employed to activate signaling pathways associated with specific muscarinic acetylcholine receptor (mAChR) subtypes. The M₁ DREADD mAChR displays minimal responsiveness to the endogenous agonist acetylcholine (ACh) but responds to clozapine-N-oxide (CNO), an otherwise pharmacologically inert ligand. We have previously shown that benzyl quinolone carboxylic acid (BQCA), an M₁ mAChR positive allosteric modulator (PAM), can rescue ACh responsiveness at these receptors. However, whether this effect is chemotype specific or applies to next-generation M₁ PAMs with distinct scaffolds is unknown. Here, we reveal that new M₁ PAMs restore ACh function at the M₁ DREADD while modulating ACh binding at the M₁ wild-type mAChR. Importantly, we demonstrate that the modulation of ACh function by M₁ PAMs is translated in vivo using transgenic M₁ DREADD mice. Our data provide important insights into mechanisms that define allosteric ligand modulation of agonist affinity vs efficacy and how these effects play out in the regulation of in vivo responses.

Introduction

Muscarinic acetylcholine receptors (mAChRs) mediate the majority of physiological actions of the neurotransmitter acetylcholine (ACh). Of the five subtypes of mAChRs, M₁−M₅, M₁ has the highest expression levels in the brain, particularly in regions involved in cognition and memory.¹ The subtype-selective activation of M₁ mAChRs is important for achieving therapeutic outcomes in cognitive-deficit disorders while avoiding off-target effects mediated by other mAChR subtypes.^2,3 However, traditional approaches to develop subtype-selective M₁ agonists that target the endogenous ligand (orthosteric) binding site have been largely unsuccessful due to the very high sequence conservation within this site across the five subtypes.⁴

Chemogenetically modified mAChRs, commonly referred to as DREADDs (designer receptors exclusively activated by designer drugs), are powerful tools to selectively target and define functions of a given receptor subtype in vitro and in vivo.^5,6 DREADDs contain mutations of two conserved orthosteric site residues, Y^3.33C and A^5.46G, which cause a dramatic loss of responsiveness to the endogenous agonist, ACh, but can be potently activated by an otherwise pharmacologically inert metabolite of clozapine, clozapine-N-oxide (CNO).⁵ DREADDs can also be used to provide insights into the mode of action of ligands targeting allosteric sites, which are less conserved and spatially distinct regions from the orthosteric binding sites.⁷ We have previously shown that the function of ACh at the M₄ DREADD can be rescued by an M₄-positive allosteric modulator (PAM), LY2033298.⁸ Similarly at the M₁ DREADD, BQCA, an M₁-selective PAM, enhanced the potency of ACh.⁹ The combination of chemogenetic modification and allosteric modulation may therefore provide subtype selectivity in response to endogenous ACh, enabling an understanding of the in vivo consequence of physiological selective receptor activation.

The poor physiochemical and pharmacokinetic properties of BQCA^10,11 are barriers to the in vivo use of this compound to study selective ACh-mediated signaling. This highlights the need for the discovery of new M₁ PAMs with improved pharmacokinetic properties. Recently, several novel M₁ PAM scaffolds were disclosed, including PF-0676783, VU6004256, VU0486846, and MIPS1780. We previously revealed that these next-generation M₁ PAMs exhibit a common mode of binding and action to that seen with BQCA at wild-type (WT) receptors, whereby they mainly modulate ACh affinity rather than efficacy.¹² Additionally, our previous studies demonstrated that at the M₁ DREADD, BQCA acts as an efficacy modulator, having no effects on the ACh affinity.⁹ It remains unknown whether the effects of BQCA at the DREADD are chemotype-specific or whether distinct PAM scaffolds can also modulate the ACh response at the M₁ DREADD in a similar manner to that observed with BQCA.

Herein, we sought to elaborate the pharmacology of the M₁ DREADD using a diverse range of orthosteric and allosteric ligands and, in particular, investigate the potential of next-generation PAMs to modulate the interaction of ACh with this receptor. We compared the activity of three M₁ PAMs with distinct scaffolds, PF-0676783, VU0486846, and MIPS1780, as well as the first generation M₁ PAM, BQCA, (Figure 1) at the human M₁ WT and DREADD mAChRs.

Figure 1. Chemical structures of the M₁ mAChR positive allosteric modulators used in the current study.

Our results reveal that, at the M₁ DREADD, in contrast to the WT receptor where M₁ PAMs primarily enhanced the affinity of ACh, each of the chemically distinct M₁ PAMs augmented the efficacy of ACh in functional assays with minimal effects on its affinity. Moreover, we demonstrate that the M₁ PAM VU0486846 that has the best pharmacokinetic profile can enable the physiological activation of the M₁ DREADD in vivo to modulate locomotor activity in transgenic mice.

Results and Discussion

Prototypical Orthosteric and Next-Generation Allosteric Ligands Have Reduced Affinity and Agonist Activity at the hM₁ DREADD

To provide a detailed pharmacological characterization of the M₁ DREADD, we first investigated the binding and signaling properties of structurally diverse ligands at these mutant receptors. Saturation binding experiments indicated a significant reduction in the cell-surface receptor expression in cells stably expressing the M₁ DREADD (Bmax = 386 549 ± 145 652 sites per cell) compared with that of cells expressing the WT (2 580 165 ± 185 544 sites per cell) receptors. In addition, there was a significant reduction in the affinity of [³H]NMS at the M₁ DREADD (pK_A 7.9 ± 0.16) compared with that at the WT (pK_A 9.0 ± 0.10) receptors (Table 1).

Table 1. Inhibition of [³H]NMS Binding by Orthosteric or Allosteric Ligands at the hM₁ WT and DREADD mAChRs Stably Expressed in FlpIn CHO Cells^a.

ligands			pKIb (nH)c
			M1WT	M1DREADD
orthosteric ligands	CNO		4.9 ± 0.05	6.9 ± 0.04d
			(0.9 ± 0.09)e	(0.8 ± 0.05)
	ACh		5.0 ± 0.03	3.0 ± 0.03d
			(0.8 ± 0.04)	(0.9 ± 0.06)e
	iperoxo		7.2 ± 0.06	5.0 ± 0.06d
			(0.6 ± 0.04)	(0.8 ± 0.09)e
	xanomeline		6.8 ± 0.04	5.5 ± 0.03d
			(1.1 ± 0.08)e	(1.5 ± 0.14)
	oxotremorine		6.0 ± 0.04	4.3 ± 0.05d
			(0.8 ± 0.05)	(1.0 ± 0.15)e
	atropine		8.6 ± 0.03	7.3 ± 0.05d
			(0.9 ± 0.04)	(0.7 ± 0.05)
allosteric ligands	BQCA		5.7 ± 0.18	5.4 ± 0.21
	MIPS1780		6.0 ± 0.10	5.3 ± 0.09d
	PF-06767832		6.3 ± 0.16	5.3 ± 0.12d
	VU0486846		6.1 ± 0.19	4.6 ± 0.25d

^aValues for orthosteric ligands were estimated by fitting the data to eq 1, and those for allosteric modulators were estimated using an allosteric ternary complex model (eq 4). Data are the mean ± SEM of four experiments performed in duplicate.

^bNegative logarithm of the equilibrium dissociation constant for each ligand.

^cSlope factor (Hill coefficient) values for orthosteric ligands. For allosteric ligands, data were analyzed using an allosteric ternary complex model (eq 4), which has a slope of 1.

^dSignificantly different compared with the M₁ WT; p < 0.05, unpaired Student’s t-test with the Holm—Sidak post hoc test.

^eNot significantly different from 1 as determined by the F-test; hence n_H was constrained to 1 to estimate the K_I values.

Equilibrium binding assays, with [³H]NMS as the probe, were performed to determine the relative affinity (pK_I) of structurally diverse ligands. As shown in Table 1, the binding curves for the competition of most orthosteric agonists for the radio-labeled antagonist binding site at the M₁ WT had shallow slopes (n_H). The most pronounced deviation from a unit slope was noted for iperoxo, but the mechanistic significance remains unclear because these deviations vary between studies even for the same ligands^9,13,14 and are not necessarily indicative of multiple binding sites or states.¹³ At the M₁ DEADD, while CNO binding curve was flattened, the calculated slope for most orthosteric agonists was not significantly different from unity. For all the prototypical orthosteric ligands, ACh, iperoxo, xanomeline, oxotremorine, and atropine, the pK_I values at the M₁ DREADD were significantly reduced compared with the values at the M₁ WT. In contrast, as previously reported,⁹ the affinity of CNO was enhanced at the M₁ DREADD. The allosteric ligands each partially inhibited [³H]NMS binding at the M₁ WT and to a lesser extent at the M₁ DREADD, indicating a negative binding cooperativity with the orthosteric antagonist (Figure 2 and Table 1). Interestingly, while the affinity of BQCA was equivalent at the M₁ WT and DREADD, which was consistent with our previous studies,⁹ PF-06767832, VU0486846, and MIPS1780 all exhibited reduced binding affinities at the M₁ DREADD relative to those at the WT receptor (Figure 2 and Table 1).

Figure 2. Equilibrium binding inhibition studies reveal the altered affinity of orthosteric and allosteric ligands at the M₁DREADD.

Inhibition of [³H]NMS binding by unlabeled ligands at the (A, C) hM₁ WT and (B, D) DREADD mAChRs stably expressed in FlpIn CHO cells. Data points represent the mean + SEM of four experiments performed in duplicate. The estimated affinity values from these experiments are listed in Table 1.

We next investigated the response of the different ligands in the IP₁ accumulation and pERK1/2 assays at the M₁ WT and DREADD mAChRs. In the IP₁ accumulation assay as a canonical measure of the G_q-mediated M₁ mAChR activation, all orthosteric and allosteric ligands were full agonists, while CNO displayed a weak inverse agonism at the M₁ WT (Figure 3 and Table 2). At the M₁ DREADD, CNO was a potent agonist; however, all other ligands displayed reduced potency and/or efficacy (Figure 3 and Table 2). In pERK1/2 assays, CNO activated the M₁ DREADD at nanomolar concentrations, whereas ACh, xanomeline, oxotremorine, and the allosteric modulators were inactive at the highest concentration used; meanwhile, iperoxo only weakly activated these receptors (Figure 4 and Table 2). The low potency or efficacy values for the orthosteric and allosteric ligands at the M₁ DREADD in the IP₁ or pERK1/2 assays are likely due to the significantly lower expression of the M₁ DREADD compared with that of the M₁ WT as well as the reduced binding affinity of these ligands at the M₁ DREADD. However, in our previous study⁹ where the level of the M₁ DREADD expression was higher and comparable to that of the M₁ WT, BQCA still displayed minimal efficacy at the M₁ DREADD in its own right. This suggests that the DREADD mutations may also affect the efficacy of ligands by promoting conformational changes less favorable for the binding of these ligands and more favorable for CNO binding.

Figure 3. M₁ DREADD receptors display reduced ligand-mediated IP₁ accumulation in response to prototypical orthosteric and allosteric ligands.

Concentration−response curves for individual ligands in IP₁ accumulation assays at the (A, C) hM₁ WT and (B, D) DREADD mAChRs stably expressed in FlpIn CHO cells. Data points are the mean + SEM of at least four experiments performed in duplicate. The estimated potency values from these experiments are listed in Table 2.

Table 2. Potency Values (pEC₅₀) for Orthosteric or Allosteric Ligands in IP₁ Accumulation and pERK1/2 Assays at the hM₁ WT and DREADD mAChRs Stably Expressed in FlpIn CHO Cells^a.

ligands		pEC50b
		IP1			pERK1/2
		M1WT	M1DREADD		M1WT	M1DREADD
orthosteric ligands	ACh	7.90 ± 0.08	ND		5.98 ± 0.14	ND
	CNO	ND	8.27 ± 0.04		ND	7.46 ± 0.24
	iperoxo	9.81 ± 0.10	5.59 ± 0.08c		8.09 ± 0.12	6.04 ± 0.36c
	xanomeline	8.97 ± 0.15	4.62 ± 0.28c		5.96 ± 0.15	ND
	oxotremorine	7.68 ± 0.12	ND		6.56 ± 0.28	ND
allosteric ligands	BQCA	6.33 ± 0.22	ND		5.85 ± 0.32	ND
	MIPS1780	6.86 ± 0.19	ND		6.25 ± 0.39	ND
	PF-06767832	7.89 ± 0.15	ND		6.00 ± 0.27	ND
	VU0486846	6.43 ± 0.17	ND		7.04 ± 0.56	ND

^aValues were estimated using a three-parameter logistic equation and represent the mean ± SEM of at least four experiments performed in duplicate. ND means not detectable.

^bNegative logarithm of the concentration of each ligand required to produce half the maximal response.

^cSignificantly different compared with the M₁ WT; p < 0.05, unpaired Student’s t-test with the Holm—Sidak post hoc test.

Figure 4. M₁ DREADD receptors display reduced ligand-mediated pERK1/2 in response to prototypical orthosteric and allosteric ligands.

Concentration−response curves to individual ligands in pERK1/2 assays at the (A, C) hM₁ WT and (B, D) DREADD mAChRs stably expressed in FlpIn CHO cells. Data points are the mean + SEM of at least four experiments performed in duplicate. Quantitative data from these experiments are listed in Table 2.

Consistent with the results of previous studies at the M₁ and M₄ DREADD,^8,9 we identified different profiles of receptor binding and activation that suggest multiple modes of receptor engagement. Altering the binding and signaling properties of both orthosteric and allosteric ligands at the M₁ mAChRs by the DREADD mutations, which are located in the orthosteric binding site, suggests that the change in the allosteric ligand affinity is likely due to an indirect conformational effect of the DREADD mutations on the allosteric site.

M₁ PAMs Do Not Modulate ACh Affinity but Do Modulate ACh Efficacy to Restore Signaling at the hM₁ DREADD

Given that the combination of the chemogenetic DREADD approach and allosteric modulation offers a great opportunity to study selective M₁ mAChR functions in response to the cognate agonist, we determined the modulation of ACh binding and signaling by structurally distinct M₁ PAMs. Our previous studies revealed that BQCA did not modulate the affinity of ACh at the M₁ DREADD.⁹ To investigate whether this loss of affinity modulation also applies to the M₁ PAMs with distinct chemical structures, the ACh-mediated inhibition of [³H]NMS binding was studied in either the absence or presence of increasing concentrations of each modulator. At the M₁ WT, BQCA (Figure S1) and the next generation M₁ PAMs (Figure 5) displayed strong positive binding cooperativity with the agonist, ACh, and negative cooperativity with the antagonist, [³H]NMS (Table 3). These effects were severely abrogated or abolished at the M₁ DREADD, which is consistent with the previous observations for BQCA.⁹

Figure 5. Next-generation M₁ PAMs, PF-06767832, VU0486846, or MIPS1780, do not enhance ACh affinity at the M₁ DREADD.

Inhibition of [³H]NMS binding by ACh in the absence or presence of increasing concentrations of each modulator at the (A-C) hM₁ WT and (D-F) DREADD mAChRs stably expressed in FlpIn CHO cells. Data points are the mean + SEM of four experiments performed in duplicate. The curves were generated by fitting the data to an allosteric ternary complex model (eq 5). Quantitative data from these experiments are listed in Table 3.

Table 3. Quantitation of Binding Parameters for the Interactions between ACh and M₁ PAMs, BQCA, PF-06767832, VU0486846, or MIPS1780, at the hM₁ WT and DREADD mAChRs Stably Expressed in FlpIn CHO Cells^a.

PAM	receptor	pKBb	logαNMS (αNMS)c	logαACh (αACh)d
BQCA	WT	5.77 ± 0.06	−0.60 ± 0.03 (0.25)	1.45 ± 0.10 (28)
	DREADD	5.41	−0.13 ± 0.02e (0.74)	0.17 ± 0.11e (1.5)
PF-06767832	WT	6.59 ± 0.05	−0.85 ± 0.03 (0.14)	1.52 ± 0.09 (33)
	DREADD	5.30e	−0.30 ± 0.05e (0.50)	0.29 ± 0.16e (1.9)
VU0486846	WT	5.96 ± 0.07	−0.41 ± 0.02 (0.39)	1.33 ± 0.08 (21)
	DREADD	4.57e	−0.50 ± 0.15 (0.32)	0.42 ± 0.17e (2.6)
MIPS1780	WT	6.24 ± 0.04	−1.23 ± 0.09 (0.06)	1.73 ± 0.08 (54)
	DREADD	5.27e	−0.53 ± 0.06e (0.29)	0.81 ± 0.12e (6.5)

^aValues were estimated from binding interaction studies using an allosteric ternary complex model (eq 5). The pK_A of [³H]NMS was constrained to the values obtained from saturation binding assays, and the pK_B for each modulator at the M₁ DREADD was constrained to the corresponding value listed in Table 1. Values represent the mean ± SEM of four experiments performed in duplicate.

^bNegative logarithm of the allosteric modulator equilibrium dissociation constant.

^cLogarithm of the binding cooperativity between [³H]NMS and each modulator.

^dLogarithm of the binding cooperativity between ACh and each modulator

^eSignificantly different compared with the M₁ WT; p < 0.05, unpaired Student’s t-test with the Holm—Sidak post hoc test.

To determine whether the M₁ PAMs could modulate ACh-mediated signaling, interaction studies were performed in the IP₁ accumulation (Suppl. Figure 2 and Figure 6) and pERK1/2 (Suppl. Figure 2 and Figure 7) assays at the M₁ WT and DREADD. Data were analyzed using an operational model of allosterism¹⁵ to estimate the efficacy (τ_B) of PAMs and their functional cooperativity (αβ) with ACh. Applying the operational model of allosterism allowed us to normalize the derived efficacy parameter based on the ratio of the receptor expression levels (B_max) between the two cell lines, thus allowing direct comparisons that would not otherwise be possible using empirical estimates of potency and E_max. As shown in Tables 4 and 5, each of the PAMs strongly potentiated the ACh response at the M₁ WT in both pathways in addition to the robust agonism. The degrees of functional cooperativity between ACh and each modulator were comparable to their binding cooperativity estimates, indicating that the PAMs were primarily modulating the ACh affinity at the M₁ WT (Tables 3−5).

Figure 6. Next-generation M₁ PAMs, PF-06767832, VU0486846, or MIPS1780, enhance the ACh-mediated IP₁ accumulation at the M₁ DREADD.

Interaction between ACh and each modulator at the (A−C) hM₁ WT and (D−F) DREADD mAChRs stably expressed in FlpIn CHO cells. Data points are the mean + SEM of four experiments performed in duplicate. The curves were generated by fitting the data to an operational model of allosterism and agonism (eq 6). The E_max value is shared for the interactions between each modulator and ACh. Quantitative data from these experiments are listed in Table 4.

Figure 7. Next generation M₁PAMs, PF-06767832, VU0486846 or MIPS1780 enhance ACh-dependent pERK1/2 at the M₁ DREADD.

Interaction between ACh and each modulator at the (A−C) hM₁ WT and (D−F) DREADD mAChRs stably expressed in FlpIn CHO cells. Data points are the mean ± SEM of four experiments performed in duplicate. The curves were generated by fitting the data to an operational model of allosterism and agonism (eq 6). The E_max value is shared for the interactions between each modulator and ACh. Quantitative data from these experiments are listed in Table 5.

Table 4. Efficacy and ACh Cooperativity Estimates for the Interaction between ACh and M₁ PAMs, BQCA, PF-06767832, VU0486846, or MIPS1780, in IP₁ Accumulation Assays in FlpIn CHO Cells Stably Expressing hM₁ WT or DREADD Receptors^a.

PAM	receptor	logαβACh (αβACh)b	logτB (τB)c
BQCA	WT	1.41 ± 0.26 (26)	1.13 ± 0.10 (13)
	DREADD	1.18 ± 0.07 (15)	−0.09 ± 0.18d (0.81)
PF-06767832	WT	1.62 ± 0.23 (42)	1.28 ± 0.10 (19)
	DREADD	1.96 ± 0.08 (91)	−3
VU486846	WT	1.48 ± 0.32 (30)	1.10 ± 0.13 (13)
	DREADD	1.92 ± 0.09 (83)	−3
MIPS1780	WT	1.87 ± 0.27 (74)	1.34 ± 0.11 (22)
	DREADD	2.41 ± 0.07 (257)	0.09 ± 0.15d (1.2)

^aValues were calculated using an operational model of allosterism and agonism (eq 6). The pK_B for each modulator was constrained to the corresponding binding affinity value listed in Table 1. Values represent the mean ± SEM of at least four experiments performed in duplicate.

^bLogarithm of the functional cooperativity between Ach and each modulator.

^cLogarithm of the functional efficacy of the modulator corrected for receptor expression levels relative to the WT. Where determined as the preferred model by the F-test, the value of logτ was constrained to −3 (τ = 0.001), which is consistent with no detectable agonist activity.

^dSignificant differences in the efficacy values of each ligand at the M1 DREADD compared with the M₁ WT; p < 0.05, unpaired Student’s t-test with the Holm—Sidak post hoc test.

Table 5. Efficacy and ACh Cooperativity Estimates for the Interaction between ACh and M₁ PAMs, BQCA, PF-06767832, VU0486846, or MIPS1780, in pERK Assays in FlpIn CHO Cells Stably Expressing hM₁ WT or DREADD Receptors^a.

PAM	receptor	logαβACh (αβACh)b		logτB (τB)c
BQCA	WT	1.41 ± 0.10 (25)		0.24 ± 0.08 (1.7)
	DREADD	1.35 ± 0.06 (23)		−3
PF-06767832	WT	1.64 ± 0.11 (35)		0.29 ± 0.08 (1.9)
	DREADD	1.79 ± 0.06 (62)		−3
VU486846	WT	1.31 ± 0.09 (20)		0.06 ± 0.06 (1.1)
	DREADD	1.56 ± 0.06 (36)		−3
MIPS1780	WT	1.85 ± 0.09 (71)		0.32 ± 0.10 (2.1)
	DREADD	2.01 ± 0.11 (102)		0.53 ± 0.15 (3.4)

^aValues were estimated using an operational model of allosterism and agonism (eq 6). The pK_B of each modulator was constrained to the corresponding binding affinity listed in Table 1. Values represent the mean ± SEM of at least four experiments performed in duplicate.

^bLogarithm of the functional cooperativity between ACh and each modulator cLogarithm of the functional efficacy of the modulator corrected for receptor expression levels relative to the WT. Where determined as the preferred model by the F-test, the value of logτ was constrained to −3 (τ = 0.001), which is consistent with no detectable agonist activity.

Strikingly, despite the significant loss of either allosteric agonist activity or ACh-affinity modulation at the M₁ DREADD, all the PAMs markedly enhanced the ACh function in the IP₁ and pERK1/2 assays (Figures S2, 6, and 7), with functional cooperativity estimates that were similar to the values at the M₁ WT (unpaired Student’s t-test with the Holm-Sidak post hoc test, Tables 4 and 5). This indicates a switch from the allosteric modulation of ACh affinity at the M₁ WT to the modulation of ACh efficacy at the M₁ DREADD. The modulatory effects of the PAMs on the ACh response were comparable between the IP₁ and pERK1/2 pathways, which is consistent with a mechanism operating at the level of receptor transitioning between inactive and active states as opposed to promoting different active states, which we had previously observed for the interaction between BQCA and CNO at the M₁ DREADD.⁹ Thus, while the PAMs almost exclusively modulate ACh binding at the M₁ WT, they act predominantly as efficacy modulators at the M₁ DREADD. The chemotype-independent switch from affinity to efficacy modulation highlights a fundamental role of the DREADD mutations in signaling, over and above any direct effects on the orthosteric binding site. Taken together with the effects observed on the allosteric ligand affinity, these findings suggest that the DREADD mutations form part of a universal network that links the orthosteric pocket to both the allosteric site and the G protein binding site.

VU0486846 Selectively Enables hM₁ mAChR-Mediated Locomotor Effects in hM₁ DREADD Transgenic Mice

Chemogenetic DREADDs are powerful tools for elucidating the in vivo consequences of the selective pharmacological activation of closely related receptor sub-types.^5,6,16,17 However, dissecting the contribution of individual receptor subtypes to the physiological or pathological signaling of the natural agonist remains challenging. To provide a proof-of-concept that M₁ PAMs can activate M₁ DREADD receptors in vivo, we performed an open-field test on humanized M₁ WT and M₁ DREADD mice 30 min after the injection of either vehicle or VU0486846 (10 mg/kg). We selected this M₁ PAM as it has been proven to be a valuable in vivo M₁ PAM tool compound with suitable pharmacokinetic properties and devoid of adverse effect liability in rodents.^18,19 As shown in Figure 8, the hM₁ DREADD mice showed a hyperactivity phenotype compared with the hM₁ WT mice. This is consistent with our previous findings where the M₁ DREADD mice were hyperactive, similar to the M₁ knockout (KO) mice.²⁰ The hyperactive phenotype of M₁ KO mice was attributed to elevated dopaminergic transmission in the striatum.²¹ Of note, our Western blot analysis showed that the total receptor expression levels in hippocampal membranes from hM₁ WT vs hM₁ DREADD mice were equivalent (Figure S3). VU0486846 significantly reduced the locomotor activity in the hM₁ DREADD mice (hM₁ DREADD vehicle vs hM₁ DREADD VU0486846, n > 50; p < 0.001, 2-way ANOVA); however, interestingly, despite restoring DREADD functionality, the PAM did not have any significant effects on the locomotor activity in mice expressing the WT M₁ mAChR. This may suggest that the effect of the PAM on M₁ mAChR-mediated locomotion is “context-specific”, i.e., normalizing an ACh response deficit manifests as a reduction in hyperactivity toward normal, but the enhancement of an ambient ACh tone in WT mice does not perturb the system any further, presumably due to a homeostatic ceiling to this effect mediated by the M₁ mAChR. Taken together, our in vitro finding that M₁ PAMs can rescue ACh function at the M₁ DREADD was also validated in vivo since we found that M₁ PAMs can partially balance locomotor disturbances in the M₁ DREADD mice.

Figure 8. VU0486846 reduces the hyperactivity phenotype displayed by hM₁ DREADD, but not WT, mice.

The average distance traveled by hM₁ WT or hM₁ DREADD mice treated with either vehicle or VU0486846 (10 mg/kg) 30 min prior to the test. Locomotor activity was recorded over a 10 min period. Data shown are the mean ± SEM for 48−57 individual mice. Data were analyzed using a two-way ANOVA with Tukey’s multiple comparison test (***p < 0.001 and ****p < 0.0001).

Conclusions

The potential to combine DREADDs with allosteric modulators (in the absence of CNO) may offer a great opportunity to selectively “reactivate” native ACh signaling. The current study, using a range of M₁ MAChR PAM chemotypes, has revealed that this is possible. Our findings indicate that a modulator that mediates effects on ACh efficacy at the DREADD in vitro may restore the receptor function in vivo. DREADDs may thus open up opportunities to differentiate the the in vivo action of M₁ PAMs that operate via efficacy vs affinity modulation.

In summary, restoring the endogenous ligand signaling at the chemogentically modified receptors in vitro and in vivo will allow the study of the physiological relevance of the activation of specific receptor subtypes, which may be implicated in the treatment of diseases in which the endogenous ligand activity has decreased.

Methods

Materials

FlpIn Chinese hamster ovary (CHO) cells, Dulbecco’s modified Eagle’s medium (DMEM), fetal bovine serum (FBS), and hygromycin B were purchased from Invitrogen (Carlsbad, CA), ThermoTrace (Melbourne, Australia), and Roche (Mannheim, Germany), respectively. [³H]N-Methylscopolamine ([³H]NMS) (specific activity of 80 Ci/mmol) and Ultima gold were purchased from PerkinElmer (Waltham, MA). The IP-One assay kit and reagents were purchased from Cisbio (Codolet, France). The AlphaScreen Surefire phospho-ERK1/2 (pERK1/2) reagents were kindly provided by TGR Biosciences (Adelaide, Australia), whereas the AlphaScreen anti-IgG (protein A) acceptor beads and streptavidin donor beads for the detection of pERK1/2 were purchased from PerkinElmer. BQCA and MIPS1780 were synthesized at the Monash Institute of Pharmaceutical Sciences,^22,23 and VU0486846 was synthesized at Vanderbilt University¹⁸ as described previously. All other chemicals were purchased from Sigma-Aldrich (St. Louis, MO), or as otherwise stated below.

Cell Culture

The hM₁ WT or DREADD mAChRs constructs²⁴ were transfected into FlpIn CHO cells and selected using 400 μg/mL hygromycin for stable expression. Cells were maintained in DMEM supplemented with 5% FBS, 16 mM HEPES, and 400 μg/mL hygromycin B at 37 °C in a humidified atmosphere containing 5% CO₂.

Whole-Cell Radioligand Binding Assays

Saturation binding assays were performed to estimate the expression levels and equilibrium dissociation constant of the radioligand (K_A). FlpIn CHO cells stably expressing the hM₁ WT mAChR and the hM₁ DREADD mAChRs were seeded at 10 000 and 50 000 cells per well, respectively, of 96-well white clear bottom plates (Greiner Bio-one, Kremsmünster, Austria) and grown overnight at 37 °C, 5% CO₂. The following day, cells were washed twice with phosphate buffer saline (PBS) and incubated with 0.03−10 nM [³H]NMS in Hanks’ balanced salt solution (HBSS) containing 10 mM HEPES (pH 7.4) in a final volume of 100 μL for 5 h at room temperature. Equilibrium inhibition binding assays were performed to determine the affinity of the ligands.

Cells were incubated with increasing concentrations of each ligand in the presence of [³H]NMS (at concentrations approximately equal to its K_A values at the M₁ WT or DREADD). For binding interaction assays, cells were incubated with increasing concentrations of ACh in the absence or presence of an increasing concentration of each modulator. In all experiments, atropine at the final concentration of 10 μM (for the M₁ WT) or 100 μM (for the M₁ DREADD) was used to determine nonspecific binding. The assays were terminated by the rapid removal of the unbound radioligand and two washes with 100 μL per well of ice-cold 0.9% NaCl buffer. Radioactivity was measured following addition of 100 μL per well of Ultima gold (PerkinElmer, Waltham, USA) and counting in a MicroBeta plate reader (PerkinElmer, Waltham, USA).

IP-One Accumulation Assays

Inositol 1 phosphate (IP₁) levels were measured using the IP-One assay kit (Cisbio, Codolet, France). Cells were seeded at 10 000 cells per well into 96-well transparent cell culture plates and incubated overnight at 37 °C, 5% CO₂. The following day, cells were stimulated with increasing concentrations of ligands or ACh in the absence or presence of increasing concentrations of each modulator in the IP₁ stimulation buffer (1 mM CaCl₂, 0.5 mM MgCl₂, 4.2 mM KCl, 146 mM NaCl, 5.5 mM D-Glucose, 10 mM HEPES, and 50 mM LiCl pH 7.4) for 1 h at 37 °C. Cells were then lysed with the IP₁ lysis buffer (50 mM HEPES pH 7.0, 15 mM KF, 1.5% v/v Triton-X-100, 3% v/v FBS, and 0.2% w/v BSA). Cell lysates were incubated with homogeneous time-resolved FRET (HTRF) reagents (the cryptate-labeled anti-IP₁ antibody and the d2-labeled IP₁ analogue) for 1 h at 37 °C. The emission signals were measured at 590 and 665 nm after excitation at 340 nm using the Envision plate reader (PerkinElmer, Waltham, USA). The signals were expressed as the HTRF ratio values and normalized to the maximum response to ACh or CNO.

Extracellular Signal-Regulated Kinase 1/2 Phosphorylation (pERK1/2) Assays

The AlphaScreen SureFire kit was used for the quantitative measurement of phosphorylated ERK1/2 (pERK1/2). Cells were plated at the density of 10 000 cells per well into 96-well transparent cell culture plates and incubated overnight at 37 °C, 5% CO₂. The following day, cells were washed twice with PBS and serum-starved for 5 h at 37 °C in 5% CO₂. Initial time course experiments were performed to determine the time of the peak pERK1/2 response for each agonist (5 min for all ligands tested). For concentration−response curves or functional interaction experiments, cells were stimulated with increasing concentrations of ligands or ACh in the absence or presence of increasing concentrations of each allosteric modulator at 37 °C for 5 min, the time at which the peak response was induced. FBS (10% v/v) was used as the positive control. The reaction was terminated by the removal of the media and the addition of the SureFire lysis buffer., and 5 μL of cell lysates were transferred to a 384-well Proxiplate (PerkinElmer, Waltham, USA). In reduced lighting conditions, 8 μL of the detection buffer (reaction buffer, activation buffer, acceptor beads, and donor beads at a ratio of 60:10:0.3:0.3) was added, and plates were incubated for 1.5 h at 37 °C. The fluorescence signal was measured using the Envision plate reader (PerkinElmer, Waltham, USA). Data were normalized to the maximum response elicited by ACh or 10% FBS.

Behavioral Pharmacology Studies

Animals

All mice were bred as homozygous onto a C57BL/6/J background. Male mice aged 8−12 weeks old were used for the behavioral studies outlined below. Mice were fed ad libitum with a standard mouse chow and were maintained within the animal facility at least 1 week prior to experiments. Animals were cared for in accordance with national guidelines on animal experimentation. All experiments were performed under a project license from the British Home Office (United Kingdom) under the Animals (Scientific Procedures) Act of 1986.

Open Field Test

General locomotor activity was assessed using the open-field test following overnight habituation in the behavioral testing suite. Mice were injected (i.p.) with either vehicle (10% Tween-80) or VU0486846 (10 mg/kg) 30 min prior to the open-field test. Mice were placed into a clear Perspex square arena (50 × 50 cm), and their activity was tracked for a 10 min period using the ANY-maze software.

Western Blotting

For the preparation of membrane extracts, frozen hippocampal tissue was suspended in a T/E buffer (10 mM Tris and 1 mM EDTA, pH 8.0) containing protease and phosphatase inhibitors. The tissue was homogenised, and the resulting homogenate was centrifuged at 15 000 × g for 1 h at 4 °C. The pellets were then solubilised in RIPA buffer (50 mM Tris-HCl pH 8, 150 mM NaCl, 0.5% (w/v) Na-deoxycholate, 1% (v/v) IGEPAL CA-630, and 0.1% (v/v) SDS) that included protease and phosphatase inhibitors and incubated for at least 2 h at 4 °C with end-over-end rotation. After the centrifugation of samples at 14 000 × g for 10 min at 4 °C, the supernatants (membrane extracts) were transferred to fresh microcentrifuge tubes and stored at −80 °C until use. Protein concentrations were determined using a Bradford assay and normalised. Samples were incubated with a Laemmli loading buffer containing 5% β-mercaptoethanol and heated for 30 min at 37 °C. Samples (10 μg) were then loaded onto 8% SDS-Tris-glycine polyacrylamide gels and run at ±100 V, following transfer onto nitrocellulose membranes. Membranes were blocked for 1 h with 5% fat-free milk in TBS-T (0.1% Tween-20 in TBS, pH 7.4). Membranes were then incubated with the Anti-HA (Roche; 1:1000) antibody (in 5% milk) overnight at 4 °C, then washed three times with TBS-T and incubated with an anti-rat secondary antibody (3:10,000) conjugated to horseradish peroxidase. Proteins were visualized with the Immobilon ECL detection system (Merck). After detection of HA, the membrane was stripped for 15 minutes at room temperature in a stripping buffer (ThermoFisher) and washed with TBS-T. The membrane was re-blocked with 5% milk, and the antibody incubations and visualization steps were repeated for the loading control Na/K-ATPase (Abcam; 1:100 000).

Data Analysis

All data were analyzed using GraphPad Prism 7 (San Diego, CA). Inhibition binding data between the radioligand antagonist, [³H]NMS, and unlabeled orthosteric ligands were analyzed according to the following equation:^25,26

$$Y=\text{basal}+\frac{\text{top}-\text{basal}}{1+{\left(\frac{\left[\text{I}\right]}{{\text{IC}}_{50}}\right)}^{n_{\text{H}}}}$$ (1)

where Y is the percentage binding, top and basal are the maximal and minimal plateaus of the curves, respectively, n_H is the Hill coefficient, [I] is the concentration of the orthosteric ligand, and IC₅₀ is the concentration producing a 50% inhibition of radioligand binding. The equilibrium dissociation constant of each ligand (K_I) was calculated using the Cheng and Prusoff equation²⁷

$$K_{\text{I}}=\frac{{\text{IC}}_{50}}{1+\left(\frac{[\text{A}]}{K_{\text{A}}}\right)}$$ (2)

where [A] is the concentration of the radioligand antagonist and K_A is its equilibrium dissociation constant obtained from the saturation binding experiments.

The inhibition of [³H]NMS binding by allosteric modulators was analyzed using an allosteric ternary complex model^28,29

$$Y=Y_0\frac{[\text{A}]+K_{\text{A}}}{[\text{A}]+K_{\text{APP}}}$$ (3)

Where

$$K_{\text{APP}}=\frac{K_{\text{A}}\left(1+[\text{B}]/K_{\text{B}}\right)}{\left(1+\alpha [\text{B}]/K_{\text{B}}\right)}$$ (4)

Y is the percentage of binding; Y₀ is the amount of radioligand bound in the absence of the allosteric modulator; [A] denotes the concentration of the radioligand; K_A is its dissociation constant; and K_App is the apparent K_A consisting of K_A modified by the modulator concentration [B], the binding cooperativity factor, α, and the dissociation constant of the the modulator, K_B.

Binding interaction studies between orthosteric agonists and allosteric modulators were fitted to the following allosteric ternary complex model:³⁰

$$Y=\frac{B_{\max }[\text{A}]}{[\text{A}]+\left[\frac{K_{\text{A}}{K}_{\text{B}}}{{\alpha }^{\prime }[\text{B}]+K_{\text{B}}}\right]\left[1+\frac{[\text{I}]}{K_{\text{I}}}+\frac{[\text{B}]}{K_{\text{B}}}+\frac{\alpha [\text{I}][\text{B}]}{K_{\text{I}}{K}_{\text{B}}}\right]}$$ (5)

where Y is percentage of binding; B_max is the total number of receptors; [A], [B], and [I] denote the concentrations of the radioligand, the allosteric ligand, and the orthosteric ligand, respectively; K_A, K_B, and K_I represent their respective equilibrium dissociation constants; and α′ and α are the affinity cooperativity factors between the allosteric ligand and the radioligand or between the allosteric ligand and orthosteric ligand, respectively. Values of α or α′ > 1 denote positive cooperativity, values between 0 and 1 denote negative cooperativity, and a value of 1 indicates neutral cooperativity.

Functional interaction studies between ACh and allosteric modulators in IP₁ and pERK1/2 assays were analyzed using the following operational model of allosterism and agonism¹⁵ :

$$\begin{array}{l}E=\text{basal}\\ +\frac{\left(E_{\text{m}}-\text{basal}\right)\left([\text{A}]\left(K_{\text{B}}+\text{αβ}[\text{B}]\right)+{\tau }_{\text{B}}[\text{B}]{\text{EC}}_{50}\right)}{{\text{EC}}_{50}\left(K_{\text{B}}+[\text{B}]\right)+\left([\text{A}]\left(K_{\text{B}}+\alpha \beta [\text{B}]\right)+{\tau }_{\text{B}}[\text{B}]{\text{EC}}_{50}\right)}\end{array}$$ (6)

where E_m is the maximal possible system response; basal is the response in the absence of an agonist; [A] and [B] denote the concentrations of orthosteric and allosteric ligands, respectively; K_B is the equilibrium dissociation constant of the allosteric ligand; and EC₅₀ is the concentration of the orthosteric agonist required to achieve a half-maximal response. α and β denote allosteric effects on the orthosteric ligand binding affinity and efficacy, respectively, and τ_B denotes the efficacy of the allosteric ligand and incorporates the total receptor density and the efficiency of the stimulus-response coupling. This model assumes that ACh is able to attain a maximal system response at saturating concentrations in both the absence and presence of the modulator, which was the case for this study.

All affinity, efficacy, and cooperativity values were estimated as logarithms³¹ and, where appropriate, were compared using the unpaired Student’s t-test with the Holm-Sidak multiple comparison test. A p-value <0.05 was considered statistically significant.

For the open field studies, data were analyzed using a two-way ANOVA with Tukey’s post hoc test to determine statistical differences between the genotype (M₁ WT vs M₁ DREADD) and the effects of VU0486846 on the locomotor activity in the M₁ WT and M₁ DREAD mice.

Abbreviations

ACh

acetylcholine

mAChR

muscarinic acetylcholine receptor

DREADD

designer receptors exclusively activated by designer drugs

PAM

positive allosteric modulator

CHO

Chinese hamster ovary

WT

wild type

DMEM

Dulbecco-modified eagle medium

FBS

fetal bovine serum

BSA

bovine serum albumin

PBS

phosphate-buffered saline

HBSS

Hanks’ balanced salt solution

IP₁

inositol monophosphate

NMS

N-methylscopolamine

BQCA

benzylquinolone carboxylic acid 1-(4-methoxybenzyl)-4-oxo-1,4-dihydroquinoline-3-carboxylic acid

PF-06767832

N-((3R,4S)-3-hydroxytetrahydro-2H-pyran-4-yl)-5-methyl-4-(4-(thiazol-4-yl)benzyl)pyridine-2-carboxamide

VU0486846

(2R)-N-[(1S,2S)-2-hydroxycyclohexyl]-4-[(4-pyrazol-1-ylphenyl)-methyl]-2,3-dihydro-1,4-benzoxazine-2-carboxamide

MIPS1780

3-(2-hydroxycyclohexyl)-6-(2-((4-(1-methyl-1H-pyrazol-4-yl)-benzyl)oxy)phenyl)-pyrimidin-4(3H)-one

CNO

clozapine-N-oxide

ANOVA

analysis of variance