Chiral Cyclic Aliphatic Linkers as Building Blocks for Selective Dopamine D₂ or D₃ Receptor Agonists

Abstract

Linkers are emerging as a key component in regulating the pharmacology of bitopic ligands directed toward G-protein coupled receptors (GPCRs). In this study, the role of regio- and stereochemistry in cyclic aliphatic linkers tethering well-characterized primary and secondary pharmacophores targeting dopamine D₂ and D₃ receptor subtypes (D₂R and D₃R, respectively) is described. We introduce several potent and selective D₂R (rel-trans-16b; D₂R K_i = 4.58 nM) and D₃R (rel-cis-14a; D₃R K_i = 5.72 nM) agonists, while modulating subtype selectivity in a stereospecific fashion, transferring D₂R selectivity toward D₃R via inversion of the stereochemistry around these cyclic aliphatic linkers (e.g., (−)-(1S,2R)-43 and (+)-(1R,2S)-42). Pharmacological observations were supported with extensive molecular docking studies. Thus, not only is an innovative approach to modulate the pharmacology of dopaminergic ligands described, but a new class of optically active cyclic linkers are introduced that can be used to expand the bitopic drug design approach toward other GPCRs.

Graphical Abstract

INTRODUCTION

G-protein coupled receptors (GPCRs) are essential to the transduction of cellular signaling and regulation of a wide range of physiological processes. Indeed, over a third of all FDA approved drugs target one or more GPCRs¹. Achieving subtype selective GPCR ligands remains challenging due to the high degree of homology among orthosteric binding sites (OBS) of receptors binding the same endogenous ligand². The expanding use of X-ray crystallography and cryoEM³ has revealed the structures of multiple GPCRs in their active, inactive, or transitional states, also uncovering potential secondary binding sites as an exploitable feature to improve ligand selectivity. As a result, the use of bitopic ligands capable of identifying a receptor OBS and secondary binding pocket (SBP) simultaneously has been recognized as an increasingly powerful strategy (Figure 1).⁴ Greater structural information of the receptor inner topology highlights the importance of architecturally complex ligands which exploit key structural differences among receptor binding sites.

Figure 1.

A) General conceptual diagram of a bitopic ligand: dark blue receptor represents generalized GPCR, orange shading represents receptor OBS, light blue shading represents receptor SBP. Orange circle represents PP of bitopic ligand, green line represents linker of bitopic ligand, blue circle represents SP of bitopic ligand. B) D₂R and D₃R primary pharmacophore scaffolds. C) and D) Representative examples of known dopaminergic bitopic antagonists or full/partial agonists, respectively; orange is PP, green is linker, blue is SP. (R)-4^{5, 6} (eutomer of (±)-4) and (R)-5⁷ are known D₃R selective antagonists. (R)-6⁶ is a potent and selective D₃R antagonist, meanwhile it’s corresponding (S)-6 enantiomer is an efficacious partial agonist. (+)-3⁸ is a D₃R selective non-competitive antagonist. 7a^{9, 10} and rel-trans-8¹¹ are known D₃R and D₂R selective agonists, respectively.

Bitopic ligands can be conceptually divided into three regions, namely the primary pharmacophore (PP) corresponding to the portion of the molecule which recognizes the OBS, the secondary pharmacophore (SP) corresponding to the portion of the molecule which recognizes the SBP, and a linker which unites PP and SP fragments into a single entity (Figure 1A). Indeed, the prerequisite of occupying both OBS and SBP simultaneously imposes a geometrical constraint on these ligands.^12–16

Structural chemoinformatic studies have shown how the chemical diversity of GPCR ligands can be efficiently exploited to achieve receptor subtype selectivity, but at the same time structural ligand-receptor complex analyses highlight how GPCR binding site similarities allow for common molecular substructures/synthons to be designed for multitarget pharmacology.¹⁷ Identifying common chemical scaffolds as components in a toolbox of molecular backbones¹⁷ upon which to construct medicinal chemistry can lead to receptor subtype selectivity and/or functional selectivity.

While unfunctionalized alkyl chains as linkers have dominated the literature on bitopic ligands,⁴ recent work from our laboratory has shown the linker portion of the molecule plays a pivotal role in the behavior of bitopic ligands: modulating affinity, efficacy, and selectivity.^{5, 6, 9, 11, 18–20} There is limited precedence for the inclusion of chiral and optically resolved cyclic moieties embedded within the linkers of bitopic ligands.^{16, 19, 21–27} Thus, our primary aim was to design synthetic routes that provide straightforward syntheses of a variety of chiral aliphatic/aromatic cyclic/heterocyclic linkers, which can act as building blocks for insertion into a wide variety of PPs and SPs for studying GPCRs of interest via bitopic drug design.

The dopamine receptor subtypes (D₁R, D₂R, D₃R, D₄R, and D₅R) are GPCRs that mediate various neurological functions such as motivation, learning, and motor control, among others; the dysregulation of which is associated with multiple neuropsychiatric and neurological disorders such as schizophrenia, Parkinson’s Disease (PD) and substance use disorders (SUD).^28–33 The individual receptor subtypes are expressed distinctly in specific brain regions and as a result modulate discrete physiological processes. Consequently, there is an ever-increasing demand for novel and subtype selective dopamine receptor ligands with improved affinities, selectivities and functional efficacies, as molecular tools and leads for medication development.

The dopamine D₃ receptor (D₃R) has become a target of great interest for the treatment of SUD.^34–42 Selective D₃R antagonists, such as (±)-4 and (R)-6 (Figure 1), have demonstrated significant attenuation of drug seeking behaviors in various animal models and may serve as promising non-opioidergic medications for the prevention and/or treatment of opioid use disorder.^{34, 43–47} Despite the success achieved in the discovery of highly selective D₃R antagonists/partial agonists, highly selective agonists remain elusive (Figure 1). D₃R-selective agonists are essential to study D₃R activation in vivo and thus extensive efforts have been aimed at developing D₃R selective agonists, yet only modest progress has been made. Moreover, the low expression level of D₃R, relative to D₂R, in brain regions controlling reward and locomotor processes^48–53 further increases the challenge, underscoring the need for highly D₃R selective agonists in order to discern the roles of D₃R from D₂R activation pathways. In this pursuit, we have recently started developing structure-activity relationships (SAR) resulting in agonists with significant D₃R (7a in Figure 1D) or D₂R (rel-trans-8 in Figure 1D) subtype selectivity.^9–11

In this extensive chemical/stereochemical SAR study we investigated and highlighted the importance of composition and chirality in the linkers of bitopic ligands. Our efforts were centered around incorporating a (1,3)-substituted cyclopentyl ring, (1,4)-substituted cyclohexyl ring, or (1,2)-substituted cyclopropyl ring in the linker (Figure 2). We uncovered that the cyclopentyl moiety introduces greater flexibility to the molecule as well as a new unique angular range between the relative positions of the PP and SP within the receptor. The stereochemistry about the linker region of the molecule, particularly around di-substituted cyclopropyl and cyclopentyl rings, tuned to the spatial environment of the D₂R and D₃R binding regions, was discovered to be a significant contributor to binding affinity and subtype selectivity as it enables accommodation of the primary and secondary pharmacophores in ideal binding poses.

Figure 2.

General templates for the design and synthesis of D₂R and D₃R selective agonists. SAR studies and chemical manipulations have been performed in the linker region of the bitopic molecules, to evaluate the effects of different aliphatic cyclic linkers, and their relative regio- and stereochemistry (bottom panel), to the pharmacological selectivity towards D₂-like receptors. (2S,5S)-1 and 2 fragment-based PP have been chosen to preferentially target D₃R and D₂R OBS, respectively. The privileged indole-2-amide (blue) SP was chosen to bind the SBP.

RESULTS AND DISCUSSION

Chemistry

Previous studies reported how a ring system in the linker adds structural constraint to the overall pose of the ligand within the GPCR. The increased rigidity can be advantageous when in harmony with the structural requirements of the binding pocket. For example, compound 7a (53a in reference⁹) ((1R, 2S) stereochemistry in the cyclopropyl ring; Figure 1D) shows a 150-fold improvement in affinity at D₃R and a 16-fold increase in selectivity for D₃R over D₂R with respect to its (1S,2R) enantiomer. This observation inspired us to further explore the chemical space around the linker, keeping constant the (2S,5S)-1 as PP and indole-2-amide as SP.

As depicted in Figure 2, our primary focus was centered around a cyclopentyl motif embedded within the linker, bridging the PP and SP of the bitopic ligands. Scheme 1 describes the general synthetic approach for compounds containing an alicyclic linker, whereas Scheme 2 describes the synthesis of compounds containing heterocyclic 5-membered rings.

Scheme 1.

a) Et₃N^.HCl, NaCN, MeOH:H₂O (1:2), followed by AcOH (pH = 7); b) methyl 2-(dimethoxyphosphoryl)acetate, NaH, MeOH; c) LAH, THF, 45 °C; d) indole-2-carboxylic acid, EDC^.HCl, HOBt, DIPEA, DCM; e) DMP, DCM, followed by nor-(2S,5S)-1⁹, cat. AcOH, STAB, DCE; f) Pd°/C, EtOH, H₂ (30 psi), followed by preparative chiral HPLC; g) DMP, DCM, followed by 2²⁰, cat. AcOH, STAB, DCE, and preparative chiral HPLC.

Scheme 2.

Panel A): a) 2-indole carboxylic acid, EDC^.HCl, HOBt, DIPEA, DCM; b) TFA, DCM; c) 2-chloroacetyl chloride, DIPEA, DCM; d) nor-(2S,5S)-1, K₂CO₃, MeCN, Δ. Panel B): a) TFA, DCM, followed by DIPEA; b) Indole-2-carboxylic acid, EDC^.HCl, HOBt, DIPEA, DCM; c) NaN₃, acetone, Δ, sealed under pressure; d) sodium ascorbate (5% mol), CuSO₄^.5H₂O (5% mol), THF:H₂O (1:1); e) DMP, DCM; f) nor-(2S,5S)-1, cat. AcOH, STAB, DCE.

The starting material for all compounds with an aliphatic cyclopentyl ring within the linker is 2-cyclopenten-1-one. Conjugate addition of the enone with sodium cyanide yields the β-cyano ketone, 9. From 9, standard Horner-Wadsworth-Emmons conditions provides 10. Subsequently, global reduction with LAH affords amino alcohol, 11. EDC mediated amide coupling of 11 with indole-2-carboxylic acid introduces the desired SP extending from the C1 position of the cyclopentyl ring. Finally, the PP is introduced in a one pot step oxidation and subsequent reductive amination with the nor-(2S,5S)-1 scaffold, synthesized as previously reported.⁹ Reduction of the exocyclic olefin of 12 via standard hydrogenation yields the saturated analog 13. Analytical chiral HPLC studies, and 1D/2D homonuclear NMR analyses, performed on this intermediate showed the cis and trans diastereomers can be readily separated using preparative chiral HPLC (NMR spectra and HPLC chromatograms in Supporting Information). Indeed, cis- and trans-13 can be readily oxidized and/or coupled with a large variety of different PPs, suitable for different GPCR targets, in all the possible stereochemical conformations. We prepared the desired 14, via DMP oxidation and reductive amination of 13 diastereomers mixture, as described above, then we proceeded to efficiently separate the cis and trans isomers at this stage via preparative chiral HPLC and assigned relative stereochemistry through NOE studies. Moreover, we also observed olefin 15 can be readily hydrogenated to yield 14, making both synthetic routes depicted in Scheme 1 converge toward the same product, in a single process if necessary. Analogously, the versatility of 13 as a key intermediate, and its suitability to high yielding oxidation, followed by reductive amination, is demonstrated by the preparation of 16. In this case, the diastereomeric mixture of cis and trans is efficiently resolved into the corresponding cis-16a and trans-16b via preparative HPLC. Determination of the relative stereochemistry was obtained via 1D-NOESY of each isomer (Supporting Information).

Previous studies had shown the inclusion of a hydroxy moiety in the linker resulted in an improved antagonist selectivity for D₃R (i.e. (R)-4 and (R)-5 (R-22); Figure 1).^{5–7, 54} Computational studies have shown when this hydroxy is β to the basic nitrogen of (R)-5 it participates in favorable hydrogen bonding with Tyr373^7.43, forming a six-membered hydrogen bond network between Tyr373^7.43, Asp110^3.32 and the basic nitrogen of (R)-5.⁵⁵ For these reasons and given the precedence that introduction of various polar functionalities in that position proved successful in generating D₃R ligands, we also sought to include an amide function and its planar heterocyclic bioisosteric triazole,^{22, 56, 57} in the linker (Figure 2, compounds 20 and 23). Scheme 2 details the synthetic strategies applied for the synthesis of the compounds featuring a heterocyclic cyclopentyl moiety in the linker. Scheme 2A provides the synthetic route to compound 20. Intermediate 18 is prepared from the commercially available tert-butyl 3-(aminomethyl)pyrrolidine-1-carboxylate upon introduction of the SP via EDC-mediated amide coupling with indole-2-carboxylic acid, and subsequent deprotection. From 18, acetylation with 2-chloroacetyl chloride (19), followed by alkylation with nor-(2S,5S)-1 provides 20, featuring the exocyclic amide embedded within the cyclic linker.

For the final compound in the heterocyclic series (Scheme 2B), we wanted to evaluate the effect of a planar heteroaromatic triazole moiety in the linker. Starting from the commercially available Boc protected propargylamine: deprotection is followed by EDC-mediated amide coupling with indole-2-carboxylic acid to afford intermediate 21. A 1,3-dipolar cycloaddition with 2-azidoethanol provides 22 which could then be converted to 23 following the same one pot DMP oxidation and subsequent reductive amination with nor-(2S,5S)-1, as described above.

It has been demonstrated in the literature how symmetric 1,3- or 1,4-substituted cyclobutyl and cyclohexyl linkers, respectively, are suitable for the preparation of bitopic D₂R/D₃R competitive and non-competitive antagonists and partial/full agonists (i.e. SB269652⁵⁸, CJ-1882⁵⁹, CJ-1639²¹). We have also recently published the use of the trans-cyclohexyl moiety in the design of D₃R selective agonists.⁶⁰ In order to complete the investigation of the cis- vs trans- cyclic stereochemistry, we prepared the cis-1,4-cyclohexyl linker building block. Scheme 3A shows the synthetic route designed to obtain cis-27a. The corresponding trans-isomer trans-27b was synthesized again using this approach and compared with the same compound previously prepared in a diastereospecific synthesis,⁶⁰ used here as the standard to assign the relative stereochemistry. The commercially available tert-butyl (4-oxocyclohexyl)carbamate was first converted into 24, subsequent deprotection, olefin hydrogenation and LAH reduction yielded the amino alcohol 25. Then, amide formation followed by oxidation and reductive amination, and separation by preparative chiral HPLC, afforded the desired cis-27a, and trans-27b whose spectroscopic and chromatographic data matched with the literature⁶⁰ readily allowing the assignment of the relative stereochemistry for both diastereomers.

Scheme 3.

Panel A: a) methyl 2-(dimethoxyphosphoryl)acetate, NaH, MeOH; b) TFA, DCM; c) Pd⁰/C, EtOH, H₂ (50 psi); d) LAH, THF; e) Indole-2-carboxylic acid, EDC^.HCl, HOBt, DIPEA, DCM; f) DMP, DCM; g) nor-(2S,5S)-1, cat. AcOH, STAB, DCE, followed by preparative chiral HPLC. Panel B: a) benzyl chloride, NaOH, tetrabutylammonium hydrogen sulfate, DCM:H₂O (3:1); b) HCl (37% in H₂O):THF (1:5), cat. TFA; c) PPh₃, DIAD, phthalimide, THF; d) hydrazine, EtOH, Δ; e) Indole-2-carboxylic acid, EDC^.HCl, HOBt, DIPEA, DCM; f) Pd⁰/C, EtOH, H₂ (50 psi); g) DMP, DCM; h) nor-(2S,5S)-1, cat. AcOH, STAB, DCE, followed by preparative chiral HPLC.

Lastly, we turned our attention to the cyclopropyl ring embedded in the linker. First, we separated both single trans-enantiomers of our functionally selective D₂R agonists 8¹¹, obtaining (−)-(1S,2R)-43 and (+)-(1R,2S)-42, with their absolute configurations assigned following Scheme S1, analogous to what was previously reported.^{8, 9} Furthermore, we pursued syntheses of the cis-relative configuration to fully investigate the importance of the linker’s chirality, particularly in very rigid cyclopropyl scaffolds and determine whether or not different stereochemistry in the linker can by itself modulate/shift D₂R and D₃R subtype selectivity.

Hence, in Scheme 3B, we began from cis-28, prepared in Kumar et al.¹⁹ The free hydroxyl was benzylated and the THP alcohol deprotected. Mitsunobu reaction in the presence of phthalimide yielded the key intermediate 31. Treatment with hydrazine followed by amide coupling with indole-2-carboxylic acid, and final O-debenzylation afforded the desired cis-alcohol 33. Desired product cis-34, was again prepared via one-pot oxidation of 33, followed by reductive amination.

Radioligand Binding Studies

Previous studies by our group have identified (Figure 1B) scaffold 1 to be a versatile PP for the development of D₃R agonists. Compound 1 on its own shows only modest affinity for the dopamine receptors, with a slight preference for the D₃R^{9, 60} (Table 1). However, the utility of 1 is as a PP in a bitopic molecule, especially when in its (2S,5S)-conformation, which instead seems to be the least favorable stereochemistry when in a simple OBS monotopic (i.e. no linker and SP) interaction ((2S,5S)-1 in Table 1). It is possible the pose undertaken by the 1 unit within the OBS allows the linker and SP to accommodate within the receptor more preferably than we have observed with other canonical D₃R agonist scaffolds.⁶⁰ The indole moiety as SP has been used extensively in dopaminergic ligands,^{4, 23, 25} and in numerous GPCR drug discovery campaigns,^{61, 62} both in the context of agonists, antagonists and bitopic allosteric modulators (Figure 1). The SBP in dopamine receptors is known to accommodate aromatic moieties, consequently a wide array of aromatic groups can be used in this position, however in this study we focused exclusively on the indole-2-amide as the SP.

Table 1. Binding affinity data.

Radioligand binding assays performed on HEK293 cells stably expressing hD₂R and hD₃R

Compound		[3H]-(R)-(+)-7-OH-DPAT Ki (nM) ± SEM			[3H]-N-methylspiperone Ki (nM) ± SEM
		hD2LR	hD3R	D2R/ D3R or (D3R/D2R)	hD2LR	hD3R	D2R/ D3R or (D3R/D2R)
1 (mixture) a		1740 ± 348 (n=5)	185 ± 20.5 (n=8)	9.4	ND	ND	ND
(2S,5S)-1		1320 ± 197 (n=3)	625 ± 118 (n=3)	2.1	ND	ND	ND
2 a		80.6 ± 15.4 (n=9)	784 ± 131 (n=7)	(9.7)	16300 ± 2930 (n=4)	6330 ± 653 (n=5)	2.6
rel-trans-7 a		106 ± 7.94 (n=5)	2.84 ± 0.462 (n=5)	37	8280 ± 462 (n=3)	34 ± 2.23 (n=3)	243
7a a		87.8 ± 9.81 (n=3)	1.85 ± 0.137 (n=3)	48	1610 ± 94 (n=3)	5.36 ± 0.522 (n=3)	300
7b a		831 ± 99.5 (n=3)	282 ± 24 (n=3)	2.9	70600 ± 15900 (n=3)	2900 ± 463 (n=3)	24
rel-trans-8		14.5 ± 3.10 (n=3)	39.9 ± 3.26 (n=3)	(2.8)	ND	ND	ND
14		86.7 ± 6.14 (n=6)	8.13 ± 0.630 (n=6)	11	15000 ± 4760 (n=3)	114 ± 13.3 (n=4)	132
rel-cis-14a		80.8 ± 11 (n=5)	5.72 ± 0.744 (n=5)	14	22600 ± 3940 (n=4)	87.6 ± 6.83 (n=4)	258
rel-trans-14b		115 ± 5.30 (n=5)	22.6 ± 2.01 (n=5)	5.1	25400 ± 2670 (n=4)	362 ± 34.6 (n=4)	70
15		55.9 ± 4.78 (n=4)	7.41 ± 0.323 (n=4)	7.5	7920 ± 772 (n=3)	104 ± 7.65 (n=4)	76
rel-cis-16a		2.03 ± 0.190 (n=3)	19.5 ± 0.450 (n=3)	(9.6)	ND	ND	ND
rel-trans-16b		4.58 ± 0.671 (n=3)	77.1 ± 10.6 (n=3)	(17)	ND	ND	ND
20		1870 ± 299 (n=3)	1380 ± 326 (n=3)	1.4	7930 ± 1190 (n=3)	2410 ± 287 (n=3)	3.3
23		13500 ± 7070 (n=3)	6860 ± 248 (n=3)	2	32300 ± 12700 (n=3)	2780 ± 1020 (n=3)	12
cis-27a		110 ± 9.51 (n=3)	31 ± 2.70 (n=3)	3.5	ND	ND	ND
trans-27b		86.2 ± 18.7 (n=3)	21 ± 2.26 (n=3)	4.1	ND	ND	ND
rel-cis-34		371 ± 47.3 (n=3)	154 ± 16.7 (n=3)	2.4	ND	ND	ND
trans-(−)-43		9.56 ± 1.63 (n=3)	106 ± 7.19 (n=3)	(11)	ND	ND	ND
trans-(+)-42		49.1 ± 11.4 (n=3)	22.1 ± 0.856 (n=3)	2.2	ND	ND	ND
44 a		1010 ± 162 (n=5)	423 ± 164 (n=5)	2.4	ND	ND	ND

Each K_i value represents the arithmetic mean ± S.E.M; n = number of independent experiments, each performed in triplicate. ND = Not Determined.

^adata previously reported^{9, 11, 20, 60}

To generate SAR, the new library of compounds were tested in radioligand binding assays at both human cloned D₂R (hD₂R) and D₃R (hD₃R) receptor subtypes, using agonist [³H]-(R)-(+)-7-OH-DPAT and/or antagonist [³H]-N-methylspiperone (Table 1). We have found that D₂R active state agonist binding is significantly more sensitive^{9, 11, 63, 64} to the radioligand used with respect to D₃R. As a consequence, the use of [³H]-(R)-(+)-7-OH-DPAT allows an accurate determination of apparent affinities for both targets, and selectivity, valuable to better correlate measured binding values with functional profiles,^{10, 11, 64, 65} which would be otherwise dramatically overestimated in favor of the D₃R when the antagonist [³H]-N-methylspiperone is used (Table 1, comparison between D₃R and D₂R selectivity ratios between K_is obtained using [³H]-N-methylspiperone or [³H]-(R)-(+)-7-OH-DPAT).

To better compare their pharmacological profiles and focus the discussion on the effects of different linkers’ stereo- and regio-chemistry on receptor subtype affinity and selectivity, the binding curves for the most potent and interesting derivatives are reported in Figure 3, and grouped based on the alicyclic linker used for tethering PP and SP.

Figure 3.

Radioligand inhibition binding experiments performed in presence of [³H]-(R)-(+)-7-OH-DPAT in competition with agonist analogues at hD₂R and hD₃R. The curves represent the mean ± S.E.M of 3 to 9 independent experiments, each performed in triplicate. Equilibrium dissociation constants (K_i) were derived from IC₅₀ values using the Cheng−Prusoff equation,⁷⁰ and are reported in Table 1. In each panel are compared binding curves at both receptor systems for each specific aliphatic cyclic linker, in all its stereochemical combinations, tethered with D₂ 2-like (Panel A and C) or D₃ (2S,5S)-1-like (Panel B and D) PP. In each panel solid lines are used for representing binding curves at the main receptor subtype target of interest, based on the PP (2-like D₂R in panel A and C; (2S,5S)-1-like D₃R in panel B and D). In each panel, shift comparison between same color curves is representative of receptor subtype selectivity for a single compound (D₂R/D₃R and/or D₃R/D₂R selectivity values for all the tested compounds are reported in Table 1), meanwhile shift between solid and dashed curves represents changes in affinity of two diastereomers at the same receptor subtype.

Compound 1 (diastereomers mixture) and its (2S,5S)-enantiomer, 2, 7a, 7b and 8, alongside other reference compounds, have been prepared, characterized, and their affinity data freshly collected or previously reported^{9, 11, 20} (Table 1), as useful comparison.

We proceeded to separate both trans-diastereomers of 8, which has been extensively characterized as a trans-diastereomeric mixture and reported as a first-in-class potent G-protein functionally selective D₂R full agonist.¹¹ We observed how (−)-(1S,2R)-43 shows not only high D₂R affinity (Figure 3A), but also D₂R selectivity (Table 1). Surprisingly, the opposite trans-diastereomer (+)-(1R,2S)-42, highlighted an interesting inversion in enantio-specificity, acting as a preferential D₃R agonist, in comparison to trans-(−)-43 (Figure 3A). To our knowledge, this is the first time that linker chirality in a bitopic molecule, not only modulates binding poses, affinity, and selectivity, but actively redirects the ligand toward a new preferential target, independent from the nature of the PP.

Expanding the size of the ring within the linker to a cyclopentyl, and consequent separation of cis and trans isomers, allowed us to observe interesting nuances in pharmacological effects, often underestimated in the literature of dopaminergic ligands, and never observed before. In regard to the D₃R agonists presenting 1-based PP, when cis-14a and trans-14b isomers were separated, we noticed that the cis configuration (14a) favors high D₃R binding affinity and selectivity, with respect to its trans analog (Figure 3B). Additionally, the exocyclic olefin analog 15 shows similar binding affinity and D₃R selectivity profile as its saturated counterparts 14a and 14b.

Analogously, when the 1 PP was switched for the 2 PP, as expected, the new ligands directed predominantly toward D₂R (Figure 3C), but we again observed that the cis-cyclopentyl ring (cis-16a) despite yielding the highest D₂R agonist affinity in this series (Table 1), also favored D₃R affinity, with consequent moderate D₂R selectivity. However, the trans counterpart (trans-16b), with a significant reduction of D₃R affinity, yielded the most selective D₂R agonist reported to date, with low nanomolar D₂R affinity (Figure 3C and Table 1). Further ring expansion from cyclopentyl to cyclohexyl, flattened the SAR, with both cis-27a and trans-27b showing almost identical affinities at both targets and moderate selectivity (Figure 3D).

Shifting the trans-cyclopropyl stereochemistry towards rel-cis-34, resulted in a significant loss of affinity at D₃R, while maintaining a similar D₂R binding, thus leading to a loss of subtype selectivity (Table 1). This suggests the design of D₂R agonists, and the space between D₂R active OBS and SBP, might be able to accommodate the presence of either cis and trans-cyclopropyl linkers, meanwhile the trans-configuration is the privileged architecture to achieve high D₃R affinity and selectivity, for agonist⁹ or antagonist^{8, 19} alike. This unique cyclopropyl SAR, slightly diverging from what was observed with the more flexible 1,3-disubstituted cyclopentyl, 1,4-disubstituted cyclohexyl, or even from the previous studies focusing on 1,3-disubstituted cyclobutyl,²¹ are not surprising, given the peculiar bonding properties^66–68, rigidity, constrained poses, and spatial orientation⁶⁹ of PP and SP that only 1,2-cis or trans disubstituted cyclopropyl may induce. We were further able to provide confirmation of the observed effects through molecular docking analysis provided below (Figures 4,5 and 1).

Figure 4.

Docking poses for cyclopropyl-linked (2S,5S)-1 and 2 compounds inside active state D₂R (pink/light green)) and D₃R (white/cyan). A) and E) Overlapped docking pose for 7a (orange sticks) and 7b (green sticks) inside D₂R and D₃R, respectively, B) and F) Overlapped docking pose for the two chiral isomers of rel-cis-34 (orange/green sticks) inside D₂R and D₃R, respectively, C) and G) Docking poses of trans-(−)-43 (orange sticks) inside active state D₂R and D₃R, respectively, D) and H) Docking poses of trans-(+)-42 (green sticks) inside active state D₂R and D₃R, respectively.

Figure 5.

Docking poses for cyclopentyl-linked (2S,5S)-1 and 2 compounds inside active state D₂R (pink)) and D₃R (white). A) and E) Docking poses for 14a (orange sticks) inside D₂R and D₃R, respectively, B) and F) Docking poses for 14b (green sticks) inside D₂R and D₃R, respectively, C) and G) Docking poses of 16a inside active state D₂R and D₃R, respectively, D) and H) Docking poses of 16b inside active state D₂R and D₃R, respectively.

Molecular Docking

Extensive all-atom molecular docking simulations were performed to develop a molecular understanding of the effect of linker characteristics on the D₂ and D₃ selectivity profiles of these new bitopic ligands. Energy-based flexible docking was used to find binding poses of appropriately charged and chirally restrained bitopic ligands inside active state cryo-EM structures for dopamine D₂ (PDBID: 7JVR)⁷¹ and dopamine D₃ (PDBID: 7CMU, 7CMV)⁷² (Figure S2). To take into account potential flexibility of the binding pocket, we used docking with full flexibility in functionally relevant side chains, including Asp^3.32, Ser^5.42, and Ser^5.46, in addition to standard rigid docking. The top-scored docking poses for the bitopic ligands inside both D₂R and D₃R were then further optimized via extensive minimization and Monte Carlo optimization protocols that take into account the full flexibility of the binding pocket.

Analysis of the top-scored optimized docking poses (Table S1) reveals that the PP of (2S,5S)-1-type ligands form a highly consistent interaction network with residues of TM3-TM5-TM6 in both D₂R and D₃R. The basic morpholine nitrogen forms salt bridges with acidic anchor Asp^3.32, and the amino group attached to the pyridine ring forms hydrogen bonds to one or both Ser^5.42 and Ser^5.46 residues. Further, the methyl-morpholine ring and aromatic pyridine ring are nested inside a cavity lined by conserved hydrophobic side-chains of Val^3.33, Cys^3.36, Phe^6.51, Phe^6.52, Thr^7.39, and Tyr^7.43 residues of both D₂R and D₃R. Similarly, the bulky dihydro-imidazo quinoline-2-one of 2-type ligands occupies the same conserved hydrophobic cavity of D₂R and D₃R with the benzo-imidazol-2-one N-1 atom forming hydrogen bonds with one of the two TM5 serine (Ser^5.42 or Ser^5.46) residues, while the attached amine anchors these ligands to the acidic Asp^3.32 of D₂R and D₃R. Interestingly, the consistent positioning of the PP in both D₂R and D₃R is in contrast with the linker-dependent positioning of the aromatic indole SP of the bitopic ligands (Figure 4).

In docking modes inside the D₃R (Figure 4), compounds with a trans cyclopropyl linker attached to (2S,5S)-1, such as in 7a and 7b, the aromatic indole SP is placed in a primarily hydrophobic pocket between ECL1-TM2 lined by Leu89^2.64, Gly93^2.68, and Gly94^ECL1. Further, the indole nitrogen atom of both 7a and 7b forms hydrogen bonds with Glu90^2.65 of the D₃R. In contrast, the cis form of these cyclopropyl-linked 1 compounds, rel-cis-34, places the indole SP close to ECL2-TM6-TM7. Although the hydrophobic indole ring SP can stack against Tyr365^7.35 for rel-cis-34, the ECL2-TM6-TM7 subpocket is lined by Ser182^ECL2(45.51), Ser184^ECL2, His349^6.55, Asn352^6.58, Thr353^6.59, and Ser366^7.36, therefore, is considerably less hydrophobic than the ECL1-TM2 subpocket engaged by the trans-cyclopropyl linked 1 compounds. However, the amide and indole heteroatoms can form hydrogen bonds with backbone atoms of ECL2 or Ser366^7.36 for the rel-cis-34.

Among the cyclopropyl-linked 2 compounds, the trans-(−)-43 adopts a similar conformation for indole SP; however, it fails to engage Glu90^2.63 of D₃R via a hydrogen bond, and interestingly the 2 PP adopts an alternate flipped conformation in the conserved PP pocket. The PP of trans-(+)-42, on the other hand, preserves usual PP conformation, and the indole SP adopts a conformation similar to the cis isomer of cyclopropyl-linked (2S,5S)-1, i.e., stacking against Tyr365^7.35 and forming hydrogen bonds with the backbone of ECL2.

The pattern of linker-dependent placement of the indole SP is also maintained for docking modes of bitopic ligands inside the D₂R (Figure 4). In the analogous docking solutions for cyclopropyl-linked bitopic ligands inside the D₃R, the indole moiety of 7a and 7b is located in the ECL1-TM2 of the D₂R. However, indole nitrogen for both the ligands fails to engage conserved Glu95^2.63; furthermore, this ECL1-TM2 subpocket of the D₂R is slightly less hydrophobic due to the presence of a non-conserved Glu181^ECL2 residue nearby. Like the D₃R, the docking mode of rel-cis-34 inside the D₂R also places the indole SP in the ECL2-TM6-TM7 sub-pocket. This ECL2-TM6-TM7 subpocket is, conversely, more hydrophobic in the D₂R, due to a replacement of non-conserved Ser182 ^ECL2(45.51) residue of the D₃R with hydrophobic Ile183^ECL2(45.51) in ECL2 of the D₂R. Interestingly, both the 2 linked cyclopropyl compounds, trans-(−)-43 and trans-(+)-42, adopt a very similar conformation inside the D₂R with the indole SP in hydrophobic ECL2-TM6-TM7 pocket lined with I183^ECL2(45.51), F389^6.51, and Y408^7.51 neighboring residues. Trans-(−)-43, however, forms additional hydrogen bonds to the backbone ECL2 residues of the D₂R. Interestingly, the indole SP of trans-(+)-42 engages π-stacking with Tyr365^7.35 and hydrogen bonds with the ECL2 backbone while maintaining relative distance from other polar residues in the ECL2-TM6-TM7 subpocket of the D₃R; however, the ligand fails to engage additional hydrogen bonding with the SP in the D₂R. This result provides a qualitative model for PP independent selectivity inversion for trans-(+)-42 for the D₂R and D₃R.

The chirality-based discrete conformation preference for the indole SP observed for cyclopropyl linkers is also observed for bitopic ligands with the cyclopentyl linked ligands on the D₃R (Figure 5). For example, 14a and 16a locate the indole SP in the hydrophobic ECL1-TM2 region with hydrogen bonding to Glu90^2.65, whereas the same indole SP is located in the polar ECL2-TM6-TM7 subpocket for 14b and 16b. In the case of the D₂R, however, extension of the non-conserved acidic residue Glu181^ECL2 in the SBP, hinders the placement of the indole SP in the ECL1-TM2 sub-pocket, while the non-conserved I183^ECL2(45.51) favors placement of the indole SP in the ECL2-TM6-TM7 region. Therefore, both cis and trans isomers of the cyclopentyl-linked indole SP occupy hydrophobic ECL2-TM6-TM7 of the D₂R with potential hydrogen bonding to the backbone atoms of ECL2. In general, the linkers that favor the ECL1-TM2 region while allowing additional hydrogen bonding engagement of the indole SP promote D₃R selectivity (e.g., 14a), and linkers that favor the ECL2-TM6-TM7 region while allowing additional hydrogen bonding engagement of the indole SP promote D₂R selectivity (e.g., 16b).

Like cyclopropyl-linked and cyclopentyl-linked bitopic ligands inside the D₃R, the cyclohexyl-linked bitopic ligands also plant the indole SP in distinct conformations, albeit with slight variation (Figure S1). The trans-27b places the indole SP in a hydrophobic region where the indole nitrogen atom forms hydrogen bonds with the ECL1 backbone and the indole ring stacks against non-conserved V180^ECL2, thus eliminating hydrogen bonding potential with Glu90^2.65. For cis-27a inside the D₃R the indole SP, instead of stacking against the aromatic ring of Y365^7.35, is orthogonal to the aromatic ring with the indole nitrogen and amide forming a hydrogen-bonding network to the Y365^7.35 hydroxyl. This hydrogen bonding network perhaps compensates for positioning of the indole SP in a relatively polar pocket of the D₃R and somewhat flattening SAR for cyclohexyl-linked bitopic ligands. Inside the D₂R, as discussed earlier for cyclopentyl-linked compounds, both the cis and trans cyclohexyl linked compounds dock with the indole SP lodged in the hydrophobic pocket between ECL2-TM6-TM7.

Finally, linkers containing heteroatoms (20, 23 and 44) showed limited binding to D₂R and D₃R. The docking and structural analyses of these ligands do not indicate adverse interactions commensurate with the loss of binding; however, they show reduced clogP values (>2) (Table S1) compared to the rest of the series (2.6–4.6),

CONCLUSIONS

In this study, by chemical and stereochemical manipulation exclusively around the linkers, i) we prepared and identified high affinity subtype selective D₂R and D₃R agonists (16b and 14a, respectively; Figure 3); ii) in line with previous studies based around 1,3-disubstituted cyclobutyls,²¹ we observed that cyclopentyl relative cis spatial conformation between PP and SP increases D₃R agonist affinity/selectivity, meanwhile a relative trans stereochemistry between PP and SP maintains high D₂R binding and simultaneously allows for higher D₂R agonist selectivity; iii) for the first time we described how inversion in cyclopropyl linker stereochemistry can direct a compound toward D₂R or D₃R preferential binding in a diastereospecific way, independently from the PP e.g., (trans-(−)-(1S,2R)-43: D₂R preferential agonist; trans-(+)-(1R,2S)-42: D₃R preferential agonist); iv) we further observed the uniqueness of 1,2-disubstituted cyclopropyl SAR, diverging from the other more flexible aliphatic rings, with exclusively trans-stereochemistry yielding high D₃R affinity and selectivity (rel-trans-7 versus rel-cis-34; Table 1); v) classic substitutions, or chemical modifications, that usually favor D₃R antagonist selectivity (i.e. hydroxybutyl, trans-cyclohexyl linker, and in general trans PP-SP relative orientations around alicyclic rings;⁵⁹ Table 1) are not directly translatable to an agonist profile.

Importantly, we have developed efficient synthetic routes for new series of cyclic linkers, readily resolved in their relative stereochemistry, and suitable to diastereospecific and enantio-specific processes, which can act as building blocks for future SAR campaigns targeting D₂-like receptors, and potentially any other GPCRs for which a bitopic drug design is suitable. By introducing varieties of PPs (effecting desired efficacy) and appropriate SPs for SBP allosteric interactions, novel GPCR bitopic ligands can readily be attained using this versatile synthetic route. Specifically, the versatility of the cyclopentyl linker will allow the design and study of unexplored chemical space, by altering the ring regiochemistry from 1,3- to 1,2-disubstitutions, and potentially introducing new functional groups in the other ring positions, exploiting unique protein-ligand interactions, guided by computer-aided drug design, and with the potential to lead innovative structural biology investigations. Ultimately, in the series described herein, we show that previous demonstrations of SAR developed for D₂-like antagonists don’t necessarily translate in the same way to agonist drug design, for the same receptor family. Indeed, we show that agonist affinities prefer opposite stereochemical requirements from what are commonly observed for antagonists, and that opposite diastereomers can also preferentially target different receptor subtypes. Further, in the computational structural studies, we identify a region of non-conservation in the ECL2 region of these two highly homologous receptors, specifically a switch in polarity from E181^ECL2 and I183^ECL2(45.51) of the D₂R, to V180^ECL2 and S182^ECL2(45.51) of the D₃R, respectively at the analogous positions, which can be exploited to develop future ligands with greater selectivity for the desired subtype. Our docking studies provide models for linker-specific positioning of the hydrophobic indole SP in this small non-conserved subregion of the pocket, directing the selectivity profile for these bitopic agonists.

CryoEM studies to observe the unique protein conformations induced by our ligand-receptor complexes, for both D₂R and D₃R active states, and how they ultimately modulate downstream cellular second messenger’s mediated responses is a future goal.

EXPERIMENTALS METHODS

Chemistry

All chemicals and solvents were purchased from chemical suppliers unless otherwise stated and used without further purification. The ¹H and ¹³C NMR spectra were recorded on a Varian Mercury Plus 400 instrument. Proton chemical shifts are reported as parts per million (δ ppm) relative to tetramethylsilane (0.00 ppm) as an internal standard, or to deuterated solvents. Coupling constants are measured in Hz. Chemical shifts for ¹³C NMR spectra are reported as parts per million (δ ppm) relative to deuterated CHCl₃ or deuterated MeOH (CDCl₃ 77.5 ppm, CD₃OD 49.3 ppm). Chemical shifts, multiplicities and coupling constants (J) have been reported and calculated using Vnmrj Agilent-NMR 400MR or MNova 9.0 software. Gas chromatography-mass spectrometry (GC/MS) data were acquired (where obtainable) using an Agilent Technologies (Santa Clara, CA) 7890B GC equipped with an HP-5MS column (cross-linked 5% PH ME siloxane, 30 m × 0.25 mm i.d. × 0.25 μm film thickness) and a 5977B mass-selective ion detector in electron-impact mode. Ultrapure grade helium was used as the carrier gas at a flow rate of 1.2 mL/min. The injection port and transfer line temperatures were 250 and 280 °C, respectively, and the oven temperature gradient used was as follows: the initial temperature (70 °C) was held for 1 min and then increased to 300 °C at 20 °C/min and maintained at 300 °C for 4 min, total run time 16.5 min. Column chromatography was performed using a Teledyne Isco CombiFlash RF flash chromatography system, or a Teledyne Isco EZ-Prep chromatography system. Preparative thin layer chromatography was performed on Analtech silica gel plates (1000 or 2000 μm). When DMA is reported as solebtm eluting system, it refers to MeOH (+1% ammonium hydroxide) in DCM. Preparative chiral HPLC was performed using a Teledyne Isco EZ-Prep chromatography system with DAD (Diode Array Detector) and ELS detectors. HPLC analysis was performed using an Agilent Technologies 1260 Infinity system coupled with DAD (Diode Array Detector). For each analytical HPLC run multiple DAD λ absorbance signals were measured in the range of 230–280 nm. Separation of the analyte, determination of purity and relative excess of enantiomers/diastereomers were achieved at 40 °C using the methods reported in each detailed reaction description. Preparative and analytical HPLC columns were purchased from Daicel corporation or Phenomenex. HPLC methods and conditions are reported in the descriptions of the chemical reactions where they were applied. For each compound, at least two independent HPLC methods were used to determine purity, enantiomeric and/or diastereomeric excess. The most successful methods used to obtain chiral resolutions, in both preparative and analytical HPLC systems, are the ones described in the experimental section. HRMS (mass error within 5 ppm) and MS/MS fragmentation analysis were performed on a LTQ-Orbitrap Velos (Thermo-Scientific, San Jose, CA) coupled with an ESI source in positive ion mode to confirm the assigned structures and regiochemistry. Optical rotations were determined using a Rudolph Research Analytical – Autopol I automatic polarimeter. Unless otherwise stated, all the test compounds were evaluated to be >95% pure on the basis of NMR, GC-MS, and HPLC-DAD. The detailed analytical results are reported in the characterization of each final compound. Figures of all the NMR spectra, NOESY used to determine relative stereochemistry, HRMS MS/MS analyses, and HPLC chromatograms are reported for each intermediate and final product and included the SI material.

5-((2S,5S)-5-Methyl-4-propylmorpholin-2-yl)pyridin-2-amine ((2S,5S)-1)

5-((2S,5S)-5-methylmorpholin-2-yl)pyridin-2-amine (nor-(2S,5S)-1; 20 mg, 0.10 mmol) was added to a solution of propionaldehyde (30 mg, 0.52 mmol) in DCE (5 mL) and cat. AcOH (2 drops). The mixture was stirred at RT for 30 min, followed by addition of STAB (22 mg, 0.10 mmol). The reaction was stirred overnight at RT. The solvents were evaporated and the residue purified by flash chromatography eluting with 10% DMA to yield 5 mg (21%) of the desired product to be used as reference standard in in vitro binding studies. Spectroscopic data are consistent with the literature⁷³.

3-Oxocyclopentane-1-carbonitrile (9)

Triethylamine hydrochloride (940 mg, 6.83 mmol) and sodium cyanide (335 mg, 6.83 mmol) were dissolved in a 30 mL H₂O:Methanol (MeOH) (2:1) solution. A prepared solution of cyclopent-2-en-1-one (520 μL, 6.21 mmol) in 1 mL of MeOH was subsequently added to the reaction mixture and the reaction was allowed to stir overnight at RT. Following overnight stirring, MeOH was removed under reduced pressure (in an ice bath) and the aq layer was brought to a pH of ~7 with dropwise addition of AcOH. The aq solution was then extracted (x3) with DCM, the combined organics were dried with Na₂SO₄ and concentrated under reduced pressure (in an ice bath). The crude material was subsequently purified via quick filtration over silica, eluting with solvent system ramping from 0% to 80% EtOAc in Hexanes, to yield 306.0 mg (45% yield) of partially purified product as a colorless oil, which was used in the following step without further purification.

Methyl-2-(3-cyanocyclopentylidene)acetate (10)

In a round bottom flask, methyl 2-(dimethoxyphosphoryl)acetate (2.98 g, 16.4 mmol) was dissolved in 100 mL of MeOH and the solution cooled to 0 °C in an ice bath. To the solution was added sodium hydride (95%, 433 mg, 17.1 mmol) portion-wise and the reaction was allowed to stir for 1 h and allowed to warm to RT. A prepared solution of 9 (1.70 g, 15.6 mmol) in 5 mL of MeOH was then added dropwise to the reaction mixture and was allowed to stir overnight at RT. Following overnight stirring, the reaction was quenched with sat. NH₄Cl solution (50 mL), MeOH was removed under reduced pressure and the aq layer was extracted with DCM. The combined organic layers were then dried with Na₂SO₄, concentrated, and the crude material purified via flash chromatography with solvent system ramping from 0% to 60% EtOAc in Hexanes, to yield 1.80 g (70% yield) of the desired product, a colorless oil, as a mixture of E and Z isomers. ¹H NMR (400 MHz, CDCl₃) δ 5.83 (p, J = 2.3 Hz, 1H, E isomer), 5.53 – 5.46 (m, 1H, Z isomer), 3.66 (d, J = 1.5 Hz, 3H), 3.29 – 3.06 (m, 2H), 3.05 – 2.83 (m, 1H), 2.83 – 2.63 (m, 2H), 2.57 – 2.30 (m, 1H), 2.24 – 1.93 (m, 1H).

2-(3-(Aminomethyl)cyclopentylidene)ethan-1-ol (11)

Compound 10 (1.96 g, 11.9 mmol) was dissolved in 100 mL of tetrahydrofuran (THF), the solution was cooled to 0 °C and subsequently was added lithium aluminum hydride (LAH; 3.15 g, 82.8 mmol) portion-wise. Upon full addition of the LAH, the reaction was heated to 45 °C and allowed to stir overnight. Following overnight stirring, the reaction was returned to RT and quenched with 4 mL of H₂O, followed by 4 mL of 2N NaOH, followed by an additional 12 mL of H₂O, and then allowed the biphasic solution to stir vigorously for 1 h. The suspension was filtered through celite, rinsing with EtOH, and the organic solvents removed under reduced pressure. The remaining crude was resuspended with 10 mL of H₂O and extracted (x3) with CHCl₃:iPrOH (3:1). The combined organic layers were dried with Na₂SO₄ and concentrated to yield the crude product in quantitative yield requiring no further purification.

N-((3-(2-Hydroxyethylidene)cyclopentyl)methyl)-1H-indole-2-carboxamide (12)

In a round bottom flask, EDC (1.36 g, 7.12 mmol), HOBt (962 mg, 7.12 mmol), and indole-2-carboxylic acid (994 mg, 6.17 mmol) were dissolved in 100 mL of DCM and stirred for 25 min. A prepared solution of 11 (670 mg, 4.74 mmol) and DIPEA (1.66 mL, 9.49 mmol) in 15 mL of DCM was then added to the reaction mixture and the reaction subsequently stirred for 5 h. The reaction solvent was then removed under reduced pressure and the resulting crude mixture purified via flash chromatography with solvent system ramping from 0% to 100% EtOAc in Hexanes, to yield 604 mg (49% yield) of the desired product as a white solid. ¹H NMR (400 MHz, CDCl₃) δ 9.97 (d, J = 8.8 Hz, 1H), 7.62 (d, J = 8.0 Hz, 1H), 7.42 (dd, J = 8.3, 3.3 Hz, 1H), 7.25 (dd, J = 7.1, 1.7 Hz, 1H), 7.15 – 7.09 (m, 1H), 6.87 (td, J = 7.9, 7.3, 2.2 Hz, 1H), 6.62 – 6.42 (m, 1H), 3.73 (d, J = 5.2 Hz, 1H), 3.63 (t, J = 6.8 Hz, 1H), 3.58 – 3.31 (m, 2H), 2.65 – 2.19 (m, 4H), 2.17 – 1.97 (m, 2H), 1.93 – 1.33 (m, 3H).

N-((3-(2-Hydroxyethyl)cyclopentyl)methyl)-1H-indole-2-carboxamide (13)

In a hydrogenation flask, compound 12 (500.0 mg, 1.76 mmol) was dissolved in 10 mL of EtOH and to the mixture was added Pd⁰/C (187 mg, 0.176 mmol, 10% wt. on carbon). The reaction was shaken in a Parr apparatus, under a 30 psi atmosphere of H_2, overnight. The suspension was filtered through celite and the solvent removed under reduced pressure. The crude residue was first purified via flash chromatography with solvent system ramping from 0% to 5% DMA, to yield 501 mg (99% yield) of the desired product as a white solid mixture of diastereomers. ¹H NMR (400 MHz, CDCl₃) δ 9.22 (s, 1H), 7.64 (d, J = 8.3 Hz, 1H), 7.43 (d, J = 8.3 Hz, 1H), 7.33 – 7.24 (m, 1H), 7.20 – 7.11 (m, 1H), 6.81 (d, J = 1.7 Hz, 1H), 6.17 (s, 1H), 3.67 (t, J = 6.8 Hz, 2H), 3.52 – 3.38 (m, 2H), 2.22 (dt, J = 16.9, 7.5 Hz, 1H), 2.11 – 1.93 (m, 2H), 1.91 – 1.77 (m, 2H), 1.64 (q, J = 6.8 Hz, 2H), 1.47 – 1.36 (m, 1H), 1.31 – 1.23 (m, 5H), 0.96 – 0.84 (m, 2H). The cis and trans diastereomers are separable via preparative HPLC using a combination of the reported methods A (Chiralpak AD-H 21 mm x 250 mm x 5 μm; flow rated 18 mL/min) and B (Chiralcel OD-H 20 mm x 250 mm x 5 μm; flow rated 18 mL/min). Analytical chiral HPLC analysis method A: AD-H analytical column (4.5mm x 250mm – 5 μm particle size); mobile phase: isocratic 10% 2-PrOH in hexanes; flow rate: 1 mL/min; injection volume: 20 μL; sample concentration: ~1 mg/mL. Multiple DAD λ absorbance signals measured in the range of 210–280 nm. HPLC Rt 87.572 min (peak 1), 91.605 min (peak 2) and 98.432 min (peak 3). Diastereomeric peak 3 (dr >99:1) can be efficiently isolated using the method A reported above. Analytical chiral HPLC analysis method B: OD-H analytical column (4.5mm x 250mm – 5 μm particle size); mobile phase: isocratic 10% 2-PrOH in hexanes; flow rate: 1 mL/min; injection volume: 20 μL; sample concentration: ~1 mg/mL. Multiple DAD λ absorbance signals measured in the range of 210–280 nm. HPLC Rt 44.602 min (peak 1) and Rt 54.310 min (peak 2). Diastereomeric peaks 1 (dr ~80:20) and 2 (dr >99:1) can be more efficiently separated using the method B reported above.

N-((3-(2-((2S,5S)-2-(6-Aminopyridin-3-yl)-5-methylmorpholino)ethyl)cyclopentyl)methyl)-1H-indole-2-carboxamide (rel-cis-14a and rel-trans-14b)

The compound was prepared following the same procedure described for 15, starting from 13 (100 mg, 0.349 mol). The crude material was purified by flash chromatography with solvent system ramping from 0% to 10% DMA, to yield 90 mg (79%) of the desired product, as a distereomeric mixture of cis and trans-cyclopentyl scaffolds. The cis and trans diastereomers were separated by preparative chiral HPLC: Chiralpak AD-H 21 mm x 250 mm x 5 μm): mobile phase: isocratic 40% 2-PrOH in hexanes; temperature: 25 °C; flow rate: 15–18 mL/min; injection volume: 3 mL (~15–20 mg/mL sample concentration); detection at λ 254 nm and 280 nm with the support of ELS detector. The desired product cis-14a eluted first, followed by trans-14b. Analytical chiral HPLC analysis performed using Chiralpak AD-H analytical column (4.5mm x 250mm – 5 μm particle size); mobile phase: isocratic 40% 2-PrOH in hexanes; flow rate: 1 mL/min; injection volume: 20 μL; sample concentration: ~1 mg/mL. Multiple DAD λ absorbance signals measured in the range of 210–280 nm. HPLC cis-14a Rt 40.395 min, purity of cis diastereomeric mixture >95%; HPLC trans-14b Rt 42.955 min and 47.023 min, purity of trans diastereomeric mixture >99%; dr of the two trans isomers 28:72. cis-14a: ¹H NMR (400 MHz, CDCl₃) δ 9.35 (s, 1H), 8.07 (d, J = 2.3 Hz, 1H), 7.64 (dd, J = 8.1, 1.0 Hz, 1H), 7.49 (dd, J = 8.4, 2.3 Hz, 1H), 7.43 (dd, J = 8.2, 0.9 Hz, 1H), 7.35 – 7.26 (m, 1H), 7.14 (ddd, J = 8.0, 7.0, 1.0 Hz, 1H), 6.82 (dd, J = 2.2, 0.9 Hz, 1H), 6.56 – 6.42 (m, 1H), 6.23 (d, J = 6.1 Hz, 1H), 4.52 – 4.34 (m, 2H), 3.87 (dd, J = 11.1, 2.8 Hz, 1H), 3.75 (dd, J = 11.1, 2.0 Hz, 1H), 3.43 (q, J = 6.7 Hz, 2H), 2.87 (dt, J = 6.6, 2.5 Hz, 1H), 2.61 – 2.49 (m, 2H), 2.48 – 2.34 (m, 2H), 2.20 (dd, J = 16.2, 7.8 Hz, 2H), 2.04 (dd, J = 8.9, 4.0 Hz, 2H), 1.82 (tt, J = 12.9, 6.6 Hz, 4H), 1.52 (dd, J = 13.9, 7.0 Hz, 3H), 1.45 – 1.35 (m, 1H), 1.31 – 1.22 (m, 3H), 1.21 (d, J = 6.2 Hz, 3H), 1.09 (d, J = 6.6 Hz, 3H), 0.96 – 0.79 (m, 2H). trans-14b: ¹H NMR (400 MHz, CDCl₃) δ 9.24 (s, 1H), 8.11 – 8.00 (m, 1H), 7.64 (d, J = 8.0 Hz, 1H), 7.54 – 7.35 (m, 2H), 7.30 (d, J = 7.4 Hz, 1H), 7.14 (t, J = 7.4 Hz, 1H), 6.82 (d, J = 2.3 Hz, 1H), 6.52 – 6.41 (m, 1H), 6.21 (s, 1H), 4.44 (s, 3H), 3.88 (d, J = 11.0 Hz, 1H), 3.75 (dd, J = 11.0, 2.0 Hz, 1H), 3.54 – 3.33 (m, 3H), 3.20 (dd, J = 13.7, 7.1 Hz, 1H), 2.88 (d, J = 7.1 Hz, 1H), 2.63 – 2.49 (m, 2H), 2.49 – 2.35 (m, 2H), 2.29 – 2.07 (m, 3H), 1.97 – 1.75 (m, 2H), 1.52 (q, J = 5.6, 3.9 Hz, 3H), 1.31 – 1.22 (m, 4H), 1.21 (dd, J = 6.0, 3.2 Hz, 1H), 1.09 (d, J = 6.6 Hz, 3H), 0.94 – 0.76 (m, 2H). HRMS MS/MS (C₂₇H₃₅N₅O₂ + H⁺): calculated 462.28635, found 462.28546. The relative stereochemistry at the cyclopentyl ring was assigned via COSY and NOESY experiments performed on each isomer.

N-((3-(2-((2S,5S)-2-(6-Aminopyridin-3-yl)-5-methylmorpholino)ethylidene)cyclopentyl)methyl)-1H-indole-2-carboxamide (15)

Compound 12 (150 mg, 0.523 mmol) was dissolved in 7 mL of DCM and cooled to 0 °C, subsequently Dess-Martin periodinane (DMP; 250 mg, 0.59 mmol) was added in one portion to the reaction mixture. The reaction was removed from the ice bath and allowed to stir for 25 min at RT after which saturated NaHCO₃ (3 mL) aq solution was added to the reaction mixture. The aq and organic layers were separated, the aq layer extracted (x3) with DCM and the combined organic layers were dried with Na₂SO₄ and concentrated. The crude residue was used without further purification. The crude residue was redissolved in 10 mL of DCE followed by the addition of nor-(2S,5S)-1 (50.0 mg, 0.258 mmol) and 3–4 drops of acetic acid. The reaction mixture was allowed to stir for 25 min followed by the addition of STAB (220 mg, 1.03 mmol). The reaction was subsequently stirred overnight. The solvent was removed under reduced pressure and the crude material purified via flash chromatography with solvent system ramping from 0% to 15% DMA, to yield 91.8 mg (77%) of the desired product, as a diastereomeric mixture of E and Z. ¹H NMR (400 MHz, CDCl₃) δ 9.49 (s, 1H), 8.15 – 8.05 (m, 1H), 7.67 – 7.59 (m, 1H), 7.51 – 7.40 (m, 2H), 7.31 – 7.24 (m, 2H), 7.17 – 7.08 (m, 1H), 6.91 – 6.77 (m, 1H), 6.51 – 6.44 (m, 1H), 6.44 – 6.19 (m, 1H), 5.36 (s, 1H), 4.47 (br s, 2H), 3.93 – 3.62 (m, 2H), 3.56 – 3.39 (m, 2H), 3.04 (s, 1H), 2.86 (dt, J = 10.1, 7.0 Hz, 1H), 2.67 – 2.46 (m, 4H), 2.45 – 2.21 (m, 2H), 1.86 (ddd, J = 21.2, 17.1, 9.2 Hz, 2H), 1.71 – 1.50 (m, 1H), 1.49 – 1.34 (m, 1H), 1.14 – 1.05 (m, 3H). Analytical chiral HPLC analysis performed using chiral OD-H analytical column (4.5mm x 250mm – 5 μm particle size); mobile phase: isocratic 30% 2-PrOH in hexanes +0.1% DIPEA; flow rate: 1 mL/min; injection volume: 20 μL; sample concentration: ~1 mg/mL. Multiple DAD λ absorbance signals measured in the range of 210–280 nm. HPLC Rt 24.218 min and 28.175 min; purity >99%, dr 70:30. HRMS MS/MS (C₂₇H₃₃N₅O₂ + H⁺): calculated 460.27070, found 460.26947.

N-((3-(2-(Methyl((R)-2-oxo-1,2,5,6-tetrahydro-4H-imidazo[4,5,1-ij]quinolin-5-yl)amino)ethyl)cyclopentyl)methyl)-1H-indole-2-carboxamide (rel-cis-16a and rel-trans-16b)

The compound was prepared following the same procedure described for 15, starting from 13 (150 mg, 0.524 mol) and 2 (106 mg, 0.524 mol). The crude material was purified by flash chromatography with solvent system ramping from 0% to 15% DMA, to yield 28 mg (11%) of the desired product, as a distereomeric mixture of cis and trans-cyclopentyl scaffolds. The cis and trans diastereomers were separated by preparative chiral HPLC: Chiralpak AD-H 21 mm x 250 mm x 5 μm): mobile phase: isocratic 60% 2-PrOH in hexanes; temperature: 25 °C; flow rate: 15–18 mL/min; injection volume: 3 mL (~10 mg/mL sample concentration); detection at λ 254 nm and 280 nm with the support of ELS detector. The desired product cis-16a eluted first, followed by trans-16b. Analytical chiral HPLC analysis performed using Chiralpak AD-H analytical column (4.5mm x 250mm – 5 μm particle size); mobile phase: isocratic 60% 2-PrOH in hexanes; flow rate: 1 mL/min; injection volume: 20 μL; sample concentration: ~1 mg/mL. Multiple DAD λ absorbance signals measured in the range of 210–280 nm. HPLC cis-16a Rt 32.811 min, purity of cis diastereomeric mixture >99%; HPLC trans-16b Rt 38.686 min, purity of trans diastereomeric mixture >95%; de 93%. cis-16a: ¹H NMR (400 MHz, CDCl₃ + CD₃OD) δ 7.60 (dt, J = 8.1, 1.0 Hz, 1H), 7.41 (dt, J = 8.3, 0.9 Hz, 1H), 7.24 – 7.19 (m, 1H), 7.08 (ddd, J = 8.0, 7.0, 1.0 Hz, 1H), 6.97 – 6.86 (m, 1H), 6.86 – 6.79 (m, 2H), 6.77 – 6.69 (m, 1H), 4.13 (dd, J = 11.7, 4.3 Hz, 1H), 3.52 (dd, J = 12.1, 10.2 Hz, 1H), 3.47 – 3.23 (m, 2H), 3.17 (dt, J = 10.1, 5.2 Hz, 1H), 3.04 (d, J = 9.9 Hz, 1H), 3.00 – 2.81 (m, 1H), 2.65 – 2.47 (m, 1H), 2.39 (br s, J = 27.2 Hz, 4H), 2.27 – 2.07 (m, 1H), 2.07 – 1.93 (m, 1H), 1.91 – 1.67 (m, 2H), 1.68 – 1.42 (m, 2H), 1.42 – 1.19 (m, 4H), 0.92 – 0.72 (m, 1H); ¹³C NMR (101 MHz, CDCl₃ + CD₃OD) δ 162.04, 154.60, 136.39, 130.78, 127.53, 127.28, 125.91, 124.35, 121.83, 121.59, 120.50, 119.86, 118.46, 112.04, 107.25, 102.44, 65.72, 57.38, 53.37, 44.79, 39.90, 39.76, 38.15, 38.10, 37.97, 34.07, 31.79, 31.62, 29.72, 29.21, 27.47, 26.89, 24.10. trans-16b: ¹H NMR (400 MHz, CDCl₃ + CD₃OD) δ 7.58 (d, J = 8.0 Hz, 1H), 7.46 – 7.31 (m, 1H), 7.21 (ddd, J = 8.3, 7.0, 1.2 Hz, 1H), 7.06 (ddd, J = 8.0, 7.0, 1.0 Hz, 2H), 7.02 – 6.68 (m, 3H), 4.19 – 4.01 (m, 1H), 3.50 (dt, J = 12.0, 10.3 Hz, 1H), 3.42 – 3.20 (m, 2H), 3.14 (dq, J = 10.0, 4.6 Hz, 1H), 2.96 – 2.80 (m, 2H), 2.64 – 2.41 (m, 1H), 2.34 (d, J = 5.2 Hz, 3H), 2.30 – 2.15 (m, 1H), 1.85 (dp, J = 19.4, 7.4 Hz, 3H), 1.63 – 1.16 (m, 6H), 0.93 – 0.67 (m, 1H); ¹³C NMR (101 MHz, CDCl₃ + CD₃OD) δ 157.32, 154.68, 145.15, 143.64, 136.53, 134.70, 127.58, 126.00, 124.35, 121.83, 121.57, 120.50, 119.87, 118.49, 118.47, 114.49, 114.42, 114.34, 112.08, 110.67, 110.27, 109.64, 107.26, 102.60, 102.31, 70.98, 65.99, 65.84, 60.97, 60.33, 57.46, 57.32, 53.41, 49.76, 49.46, 44.80, 44.51, 42.00, 40.03, 39.75, 38.75, 36.69, 36.16, 34.09, 32.88, 30.27, 29.74, 27.06, 25.22, 24.16. HRMS MS/MS (C₂₈H₃₃N₅O₂ + H⁺): calculated 472.27070, found 472.27029. The relative stereochemistry at the cyclopentyl ring was assigned via COSY and NOESY experiments performed on each isomer.

tert-Butyl 3-((1H-indole-2-carboxamido)methyl)pyrrolidine-1-carboxylate (17)

To a round bottom flask were added indole-2-carboxylic acid (2.00 g, 12.0 mmol), N-(3-Dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride (EDC^.HCl; 2.40 g, 12.0 mmol), and hydroxybenzotriazole hydrate (HOBt; 1.90 g, 12.0 mmol) and suspended in 150 mL of dichloromethane (DCM), N,N-Diisopropylethylamine (DIPEA; 2.20 mL, 12.0 mmol) was added to the reaction mixture and the mixture was allowed to stir at RT for 10 min, allowing the reactants to solubilize. To the reaction mixture was added tert-butyl 3-(aminomethyl)pyrrolidine-1-carboxylate (2.50 g, 12.0 mmol) and the reaction mixture was allowed to stir for 3 h. The reaction solvent was subsequently removed under reduced pressure and the crude mixture was purified via flash column chromatography with the desired product eluting with 40–50% Ethyl Acetate (EtOAc) in Hexanes, to yield 2.80 g (65% yield) of the desired product. ¹H NMR (400 MHz, CDCl₃) δ 9.74 (s, 1H), 7.64 (d, J = 8.1 Hz, 1H), 7.44 (d, J = 8.3 Hz, 1H), 7.28 (t, J = 9.5 Hz, 1H), 7.13 (t, J = 7.5 Hz, 1H), 6.88 (s, 1H), 6.61 (d, J = 63.3 Hz, 1H), 3.67 – 3.27 (m, 5H), 3.16 (dt, J = 50.8, 7.1 Hz, 1H), 2.62 – 2.50 (m, 1H), 2.09 – 1.96 (m, 1H), 1.69 (s, 1H), 1.46 (s, 9H).

N-(Pyrrolidin-3-ylmethyl)-1H-indole-2-carboxamide (18)

Compound 17 (2.80 g, 8.20 mmol) was dissolved in 20 mL of DCM and to the solution was added trifluoroacetic acid (TFA; 6.30 mL, 82.0 mmol) dropwise. The reaction mixture was allowed to stir at RT overnight. Solvent and TFA were subsequently removed under reduced pressure, the crude material was basified with 2N NaOH and the aq solution was then extracted (x3) with DCM:2-propanol (iPrOH) (3:1). The organic fractions were then dried with Na₂SO₄, concentrated, and the crude was purified via flash chromatography with the desired product eluting with 25% DMA, to yield 1.40 g (71% yield) of the desired product. ¹H NMR (400 MHz, CDCl₃) δ 7.64 (dd, J = 8.0, 1.0 Hz, 1H), 7.43 (dd, J = 8.3, 1.0 Hz, 1H), 7.30 – 7.25 (m, 2H), 7.20 (s, 1H), 7.13 (ddd, J = 8.0, 7.0, 1.0 Hz, 1H), 6.88 (s, 1H), 3.56 – 3.40 (m, 2H), 3.14 – 3.01 (m, 2H), 2.97 – 2.84 (m, 2H), 2.47 (dq, J = 13.2, 6.2 Hz, 1H), 2.10 – 1.92 (m, 1H), 1.63 – 1.49 (m, 1H).

N-((1-(2-Chloroacetyl)pyrrolidin-3-yl)methyl)-1H-indole-2-carboxamide (19)

Compound 18 (700 mg, 2.87 mmol) was dissolved in 20 mL of DCM. To the solution of 18 was added 2-chloroacetyl chloride (300 mg, 2.65 mmol) dropwise, followed by the addition of DIPEA (2.70 mL, 15.5 mmol). The reaction mixture was stirred for 3 h after which the reaction solvent was removed under reduced pressure and the crude material was purified via flash chromatography with the desired product eluting with 5% DMA, to yield 0.350 g (39% yield) of the desired product. ¹H NMR (400 MHz, CDCl₃) δ 9.39 (d, J = 26.3 Hz, 1H), 7.65 (dd, J = 7.7, 2.7 Hz, 1H), 7.33 – 7.27 (m, 2H), 7.15 (dddd, J = 8.0, 7.0, 5.0, 1.0 Hz, 1H), 6.91 (ddd, J = 6.8, 2.2, 0.9 Hz, 1H), 6.69 (dt, J = 61.3, 5.6 Hz, 1H), 4.03 (d, J = 2.6 Hz, 2H), 3.77 – 3.62 (m, 2H), 3.58 – 3.46 (m, 2H), 3.44 – 3.31 (m, 2H), 2.64 (dp, J = 42.1, 7.1 Hz, 1H), 2.24 – 2.06 (m, 1H), 1.80 (ddq, J = 47.6, 12.8, 7.9 Hz, 1H).

N-((1-(2-((2S,5S)-2-(6-Aminopyridin-3-yl)-5-methylmorpholino)acetyl)pyrrolidin-3-yl)methyl)-1H-indole-2-carboxamide (20)

To a round bottom flask were added compound 19 (50.0 mg, 0.156 mmol), nor-(2S,5S)-1 (30.0 mg, 0.155 mmol), and K₂CO₃ (210 mg, 1.52 mmol), the reactants were then dissolved in 10 mL of MeCN. The reaction mixture was stirred at reflux for 5 h and subsequently stirred at RT overnight. The suspension was then filtered, the solvent was evaporated under reduced pressure, and the crude residue was purified via flash chromatography with the desired product eluting with 10–15% DMA, to yield 26.0 mg (35% yield) of the desired product as a mixture of diastereomers. Mixture of isomer A and B: ¹H NMR (400 MHz, CDCl₃) δ 9.96 (s, 1H, isomer A), 9.86 (d, J = 27.9 Hz, 1H, isomer B), 8.11 (dd, J = 38.1, 2.3 Hz, 1H, isomer A), 7.99 (dd, J = 5.2, 2.3 Hz, 1H, isomer B), 7.61 (dt, J = 8.1, 2.6 Hz, 1H), 7.50 – 7.38 (m, 3H), 7.31 – 7.22 (m, 1H), 7.12 (ddddd, J = 7.8, 6.8, 4.4, 2.3, 1.0 Hz, 1H), 7.00 (dd, J = 4.8, 1.1 Hz, 1H, isomer A), 6.94 (dd, J = 21.7, 1.5 Hz, 1H, isomer B), 6.52 – 6.40 (m, 1H), 4.58 – 4.49 (m, 2H), 4.42 (ddd, J = 12.6, 8.8, 3.3 Hz, 1H), 3.84 – 3.57 (m, 4H), 3.56 – 3.46 (m, 2H), 3.45 – 3.29 (m, 3H), 3.27 – 3.03 (m, 2H), 2.90 – 2.70 (m, 2H), 2.69 – 2.62 (m, 1H), 2.11 (td, J = 11.0, 9.5, 5.1 Hz, 1H), 2.05 – 1.98 (m, 1H), 1.77 (ddt, J = 27.1, 13.0, 7.1 Hz, 1H), 1.66 – 1.52 (m, 1H), 1.11 (t, J = 6.8 Hz, 3H, isomer A), 1.04 (dd, J = 20.6, 6.5 Hz, 3H, isomer B). ¹³C NMR (101 MHz, CDCl₃) δ 171.32, 169.12, 168.89, 168.83, 162.44, 162.27, 162.25, 162.19, 158.26, 158.19, 158.15, 158.06, 146.73, 146.53, 146.51, 146.37, 140.33, 137.46, 137.36, 136.85, 136.60, 136.59, 136.57, 136.53, 130.89, 130.85, 130.79, 127.74, 127.72, 127.70, 125.88, 125.81, 125.78, 125.68, 124.71, 124.68, 124.62, 124.58, 122.11, 122.06, 120.82, 120.79, 120.73, 120.70, 112.15, 112.13, 112.11, 112.10, 108.79, 108.67, 108.57, 108.44, 103.41, 103.31, 103.06, 102.86, 75.72, 75.23, 74.86, 74.43, 71.62, 71.13, 70.43, 70.17, 60.54, 59.77, 58.41, 57.95, 57.86, 54.65, 53.52, 53.39, 53.17, 53.07, 52.96, 52.90, 52.59, 50.90, 49.75, 49.61, 45.81, 45.66, 45.27, 41.96, 41.78, 41.74, 40.24, 39.64, 37.83, 37.54, 29.82, 29.71, 27.88, 27.84, 21.20, 14.33, 10.24, 9.64, 8.97. Analytical chiral HPLC analysis performed using Chiralpak AD-H analytical column (4.5mm x 250mm – 5 μm particle size); mobile phase: isocratic 50% 2-PrOH in hexanes; flow rate: 1 mL/min; injection volume: 20 μL; sample concentration: ~1 mg/mL. Multiple DAD λ absorbance signals measured in the range of 210–280 nm. Rt 31.281 min and 36.464 min, purity >99%; dr 50:50. HRMS MS/MS (C₂₆H₃₂N₆O₃ + H⁺): calculated 477.26087, found 477.26064.

N-(Prop-2-yn-1-yl)-1H-indole-2-carboxamide (21)

To a solution of tert-butyl prop-2-yn-1-ylcarbamate (5.00 g, 32.2 mmol) in 100 mL of DCM was added TFA (3.70 g, 32.5 mmol) and the reaction was stirred at RT for 1 h. The reaction was subsequently basified with the addition of DIPEA (4.20 g, 32.5 mmol). This reaction mixture was then added portion-wise to a separate flask containing a solution of indole-2-carboxylic acid (5.20 g, 32.2 mmol), EDC^.HCl (6.20 g, 32.4 mmol), and HOBt (4.90 g, 32.0 mmol) in 100 mL of DCM. The reaction was stirred overnight at RT, the solvent was later removed under reduced pressure and the crude residue was purified via flash chromatography with the desired product eluting with 30% EtOAc in Hexanes, to yield 200 mg (3.1% yield) of the desired product (significant yield loss observed during flash chromatography due to precipitation of product). ¹H NMR (400 MHz, CDCl₃) δ 9.35 (s, 1H), 7.66 (dq, J = 8.1, 1.0 Hz, 1H), 7.45 (dd, J = 8.3, 1.0 Hz, 1H), 7.31 (ddd, J = 8.3, 7.0, 1.2 Hz, 1H), 7.15 (ddd, J = 8.0, 7.0, 1.0 Hz, 1H), 6.91 – 6.86 (m, 1H), 6.34 (s, 1H), 4.30 (dd, J = 5.4, 2.6 Hz, 2H), 2.32 (t, J = 2.6 Hz, 1H). GC/MS (EI), Rt 10.248 min; 198.1 (M⁺).

N-((1-(2-Hydroxyethyl)-1H-1,2,3-triazol-4-yl)methyl)-1H-indole-2-carboxamide (22)

In a pressure vessel, sodium azide (1.05 g, 16.1 mmol) was added portion-wise to a solution of 2-bromoethan-1-ol (2.00 g, 16.0 mmol) in 50 mL of acetone, the vessel was sealed and the reaction stirred for 3 h with gentle heating. The reaction was subsequently cooled down to RT, the suspension was filtered and the solvent evaporated under reduced pressure. The resulting residue was then dissolved in 30 mL of (1:1) THF:H₂O, to the solution was added 21 (200 mg, 1.01 mmol), sodium ascorbate (9.99 mg, 0.050 mmol) and copper sulfate pentahydrate (8.05 mg, 0.050 mmol), and the reaction was further stirred for an additional 3h at RT. The organic and aq layers were then separated and the aq layer extracted (x3) with DCM:iPrOH (3:1). The combined organic layers were dried with Na₂SO₄ and concentrated under reduced pressure. The crude residue was purified via flash chromatography with the desired product eluting with 5% DMA, to yield 140 mg (49% yield) of the desired product. ¹H NMR (400 MHz, CDCl₃) δ 9.31 (s, 1H), 7.64 (d, J = 5.8 Hz, 2H), 7.43 (dd, J = 8.3, 1.0 Hz, 1H), 7.29 (ddd, J = 8.3, 6.9, 1.1 Hz, 1H), 7.17 – 7.12 (m, 1H), 7.05 (t, J = 5.8 Hz, 1H), 6.89 (d, J = 2.2 Hz, 1H), 4.71 (d, J = 5.9 Hz, 2H), 4.49 – 4.42 (m, 2H), 2.84 (br s, 1H).

N-((1-(2-((2S,5S)-2-(6-Aminopyridin-3-yl)-5-methylmorpholino)ethyl)-1H-1,2,3-triazol-4-yl)methyl)-1H-indole-2-carboxamide (23)

Dess-Martin periodinane (220 mg, 0.52 mmol) was added to a solution of 22 (140 mg, 0.49 mmol) in 20 mL of DCM. The reaction mixture was stirred at RT for 20 min and subsequently quenched with 5 mL of sat. NaHCO₃. The aq and organic layers were separated, and the organic phase concentrated under reduced pressure. The residue was redissolved in 10 mL of dichloroethane (DCE), to the solution was then added nor-(2S,5S)-1 (20.0 mg, 0.103 mmol) and three drops of acetic acid (AcOH). The reaction mixture was stirred for 20 min and then sodium triacetoxyborohydride (STAB; 26.0 mg, 0.120 mmol) was added in one portion. The reaction mixture was stirred overnight at RT; the solvent was subsequently removed under reduced pressure and the crude residue purified via flash chromatography with the desired product eluting with 15% DMA, to yield 10.0 mg (21% yield) of the desired product. ¹H NMR (400 MHz, CDCl₃) δ 9.36 (s, 1H), 7.98 (d, J = 2.3 Hz, 1H), 7.74 (s, 1H), 7.64 (d, J = 8.0 Hz, 1H), 7.42 (ddd, J = 8.5, 5.9, 1.7 Hz, 2H), 7.32 – 7.24 (m, 2H), 7.13 (ddd, J = 8.0, 6.9, 1.0 Hz, 1H), 6.92 – 6.89 (m, 1H), 6.44 (d, J = 8.4 Hz, 1H), 4.78 (dd, J = 5.8, 1.9 Hz, 2H), 4.50 – 4.33 (m, 5H), 3.70 (ddd, J = 43.2, 11.2, 2.7 Hz, 2H), 2.91 (ddt, J = 20.3, 13.7, 7.4 Hz, 2H), 2.78 – 2.66 (m, 2H), 2.50 (dd, J = 11.8, 3.2 Hz, 1H), 1.10 (d, J = 6.6 Hz, 3H). ¹³C NMR (101 MHz, CDCl₃) δ 161.52, 146.26, 136.63, 130.32, 125.52, 124.67, 122.99, 122.09, 120.73, 111.85, 108.90, 108.42, 80.61, 54.21, 52.78, 48.51, 35.06, 34.13, 31.75, 29.70, 29.51, 29.21, 25.32, 22.59, 20.91, 20.71. Analytical chiral HPLC analysis performed using Chiralpak AD-H analytical column (4.5mm x 250mm – 5 μm particle size); mobile phase: isocratic 50% 2-PrOH in hexanes; flow rate: 1 mL/min; injection volume: 20 μL; sample concentration: ~1 mg/mL. Multiple DAD λ absorbance signals measured in the range of 210–280 nm. Rt 57.575 min, purity >99%; de >99%. HRMS MS/MS (C₂₄H₂₈N₈O₂ + H⁺): calculated 461.24080, found 461.24025.

Methyl 2-(4-((tert-butoxycarbonyl)amino)cyclohexylidene)acetate (24)

The compound was prepared following the same procedure described for 10, starting from tert-butyl (4-oxocyclohexyl)carbamate (3.00 g, 14.1 mmol). The reaction was quenched by addition of sat. NH₄Cl aq. solution. MeOH and H₂O were evaporated under reduced pressure, the residue absorbed over silica and purified by flash chromatography eluting with 40% EtOAc in hexanes, to yield 3.60 g (95% yield) of the desired product, as a white solid. ¹H NMR (400 MHz, CDCl₃) δ 5.64 (s, 1H), 4.43 (s, 1H), 3.68 (s + m, J = 0.6 Hz, 3H + 1H), 2.40 – 1.89 (m, 6H), 1.44 (s, 9H), 1.31 (dddd, J = 24.5, 12.3, 10.8, 4.5 Hz, 2H).; GC/MS (EI), Rt 9.848 min; 269.1 (M⁺).

2-(4-Aminocyclohexyl)ethan-1-ol (25)

TFA (5.00 mL, 60.0 mmol) was added to a solution of 24 (2.50 g, 9.30 mmol) in DCM (50 mL), at 0 °C. The reaction mixture was stirred at RT for 2h, solvent and excess of TFA were evaporated under reduced pressure, and the residue used in the following step, as TFA salt, without further purification. Pd⁰/C (40.0 mg, 0.40 mmol) was added to a solution of intermediate TFA salt (2.00 g, 7.00 mmol) in EtOH (15 mL). The suspension was shaken in a Parr apparatus, under 50 psi of H₂ pressure, for 30 min. The mixture was filtered over celite and the solvent evaporated. The residue was then dissolved in THF (20 mL), followed by portion-wise addition of LAH (1.00 g, 30.0 mmol) at 0 °C. The suspension was stirred overnight at RT, quenched by addition of 2N NaOH/MeOH (10 mL, 1:1 ratio), and filtered. The solvent was evaporated and the crude material was used in the following step without further purification.

N-(4-(2-Hydroxyethyl)cyclohexyl)-1H-indole-2-carboxamide (26)

The compound was prepared following the same procedure described for 17, starting from 25 (0.800 g, 5.59 mmol). The crude material was purified by flash chromatography eluting with 90% EtOAc in hexanes, to yield 0.330 g (20.6%) of the desired product, as a colorless oil. GC/MS (EI), Rt 14.783 and 14.814 min; 286.1 (M⁺). The presence of both cis and trans diastereomers was confirmed by analytical HPLC. Analytical chiral HPLC analysis performed using Chiralcel OD-H analytical column (4.5mm x 250mm – 5 μm particle size); mobile phase: isocratic 20% iPrOH in hexanes; flow rate: 1 mL/min; injection volume: 20 μL; sample concentration: ~1 mg/mL. Multiple DAD λ absorbance signals measured in the range of 210–280 nm. cis isomer: Rt 12.898 min, trans isomer: Rt 15.479 min; dr cis:trans 35:65. The relative stereochemistry was assigned using the previously synthesized trans-N-(4-(2-hydroxyethyl)cyclohexyl)-1H-indole-2-carboxamide⁶⁰ as HPLC reference standard (Rt 15.906).

cis-N-((1R,4s)-4-(2-((2S,5S)-2-(6-Aminopyridin-3-yl)-5-methylmorpholino)ethyl)cyclohexyl)-1H-indole-2-carboxamide (27a) and trans-N-((1S,4r)-4-(2-((2S,5S)-2-(6-Aminopyridin-3-yl)-5-methylmorpholino)ethyl)cyclohexyl)-1H-indole-2-carboxamide (27b)

The compounds were prepared following the same procedure described for 14, starting from 26 (330 mg, 1.15 mmol). The crude material and the two cis and trans diastereomers were purified by preparative chiral HPLC: Chiralpak AD-H 21 mm x 250 mm x 5 μm): mobile phase: isocratic 60% iPrOH in hexanes; temperature: 25 °C; flow rate: 15–18 mL/min; injection volume: 3 mL (~10 mg/mL sample concentration); detection at λ 254 nm and 280 nm with the support of ELS detector. The desired product cis-27a eluted first (96 mg), followed by trans-27b (96 mg). Analytical chiral HPLC analysis performed using Chiralpak AD-H analytical column (4.5mm x 250mm – 5 μm particle size); mobile phase: isocratic 60% iPrOH in hexanes; flow rate: 1 mL/min; injection volume: 20 μL; sample concentration: ~1 mg/mL. Multiple DAD λ absorbance signals measured in the range of 210–280 nm. HPLC cis-27a Rt 16.462 min, purity >99%, de >99%; HPLC trans-27b Rt 27.838 min, purity >99%; de >99%. The relative stereochemistry was assigned using the previously synthesized

trans-N-((1S,4r)-4-(2-((2S,5S)-2-(6-Aminopyridin-3-yl)-5-methylmorpholino)ethyl)cyclohexyl)-1H-indole-2-carboxamide⁶⁰ as NMR and HPLC reference standard (Rt 29.504).

cis-27a: ¹H NMR (400 MHz, CDCl₃) δ 9.10 (s, 1H), 8.08 (d, J = 2.3 Hz, 1H), 7.65 (d, J = 8.0 Hz, 1H), 7.50 (dd, J = 8.5, 2.4 Hz, 1H), 7.43 (d, J = 8.3 Hz, 1H), 7.29 (t, J = 7.7 Hz, 1H), 7.15 (t, J = 7.5 Hz, 1H), 6.82 (s, 1H), 6.49 (d, J = 8.4 Hz, 1H), 6.19 (d, J = 7.7 Hz, 1H), 4.52 – 4.44 (m, 1H), 4.42 (br s, 2H), 4.28 – 4.17 (m, 1H), 3.89 (d, J = 11.1 Hz, 1H), 3.77 (d, J = 11.0 Hz, 1H), 2.96 – 2.83 (m, 1H), 2.58 (d, J = 7.6 Hz, 2H), 2.52 – 2.39 (m, 2H), 1.78 – 1.68 (m, 4H), 1.64 – 1.50 (m, 6H), 1.36 – 1.27 (m, 2H), 1.12 (d, J = 6.6 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃ + CD₃OD) δ 161.86, 161.78, 159.05, 158.98, 147.33, 147.30, 147.25, 137.28, 137.24, 137.21, 132.01, 131.94, 128.50, 128.47, 128.42, 127.11, 127.04, 125.20, 125.13, 122.67, 122.61, 121.45, 121.38, 112.90, 112.87, 112.82, 109.20, 109.13, 102.49, 102.41, 73.13, 73.05, 53.97, 53.90, 53.26, 53.19, 52.28, 52.21, 47.03, 34.51, 33.19, 30.22, 30.15, 30.08, 29.24, 29.17, 8.74, 8.67. HRMS MS/MS (C₂₇H₃₅N₅O₂ + H⁺): calculated 462.28635, found 462.28608. trans-27b: ¹H and ¹³C NMR identical to the literature⁶⁰; HRMS MS/MS (C₂₇H₃₅N₅O₂ + H⁺): calculated 462.28635, found 462.28620.

cis-2-(2-(2-((Benzyloxy)methyl)cyclopropyl)ethoxy)tetrahydro-2H-pyran (29)

Predominantly cis-28 (2.00 g, 10 mmol) was isolated as previously described¹⁹, and dissolved in a mixture of DCM (30 mL), H₂O (10 mL), NaOH (1.20 g, 30 mmol) and tetrabutylammonium hydrogen sulfate (1.70 g, 5 mmol), at 0 °C. Benzyl chloride (1.26 g, 10 mmol) was then added dropwise, and the mixture was stirred at RT for 3 h. The organic phase was separated from the aq one, dried with Na₂SO₄, filtered and evaporated. The crude residue was purified via flash chromatography eluting with 40% EtOAc in hexanes, to yield 650 mg (22.4%) of the desired product, as a colorless oil. ¹H NMR (400 MHz, CDCl₃) δ 7.42 – 7.27 (m, 5H), 4.74 – 4.59 (m, 1H), 4.55 (d, J = 14.3 Hz, 2H), 3.97 – 3.81 (m, 3H), 3.81 – 3.66 (m, 2H), 3.67 – 3.57 (m, 1H), 3.53 (ddd, J = 12.1, 7.0, 3.1 Hz, 2H), 3.29 – 3.09 (m, 2H), 2.01 – 1.70 (m, 4H), 1.62 – 1.36 (m, 4H), 1.35 – 1.15 (m, 2H), 0.97 – 0.54 (m, 4H).

cis-2-(2-((Benzyloxy)methyl)cyclopropyl)ethan-1-ol (30)

Compound 29 (650 mg, 2.24 mmol) was dissolved in THF (20 mL), followed by addition of 37% aq. HCl (4 mL) and TFA (1 mL). The solvents excess was evaporated under reduced pressure, the residue basified with sat. aq. NaHCO₃ and extracted with DCM:iPrOH (3:1). The combined organic layers were dried with Na₂SO₄, filtered and evaporated to yield the crude material which was sued in the next step without further purification.

cis-2-(2-(2-((Benzyloxy)methyl)cyclopropyl)ethyl)isoindoline-1,3-dione (31)

A solution of 30 (650 mg, 3.15 mmol), triphenylphosphine (PPh₃; 826 mg, 3.15 mmol) and DIAD (637 mg, 613 μL, 3.15 mmol) in THF (10 mL) was stirred at 0 °C for 10 min, followed by dropwise-wise addition of a phthalimide (464 mg, 3.15 mmol) solution in THF (10 mL). The mixture was stirred at RT for 2 h, the solvents were evaporated and the residue purified via flash chromatography eluting with 30% EtOAc in hexanes to yield 600 mg (56.8%) of desired product, which, based on ¹H NMR analysis, co-eluted together with a phthalimide byproduct impurity. The partially pure isolated product was used in the following step without further purification.

cis-N-(2-(2-((Benzyloxy)methyl)cyclopropyl)ethyl)-1H-indole-2-carboxamide (32)

Hydrazine (2.00 mL, 63.7 mmol) was added to a solution of 31 (650 mg, 1.94 mmol) in EtOH (20 mL). The reaction mixture was stirred at reflux for 2h, until complete consumption of the starting material, as monitored by TLC. EtOH was evaporated, the residue suspended in 20% aq. K₂CO₃, and extracted with DCM:iPrOH (3:1). The combined organic layers were dried with Na₂SO₄, filtered and evaporated to yield 120 mg (30.2%) of crude product, which was used in the following step without further purification.

The compound was prepared following the same procedure described for 26, starting from 120 mg of primary amine crude intermediate (0.633 mmol). The crude material was purified via flash chromatography eluting with 40% EtOAc in hexanes, to yield 90 mg (41%) of desired product. The pure diastereomeric mixture was used in the following steps. ¹H NMR (400 MHz, CDCl₃) δ 9.79 (s, 1H), 7.58 (dt, J = 17.7, 9.1 Hz, 2H), 7.49 – 7.40 (m, 1H), 7.39 – 7.21 (m, 3H), 7.17 – 7.04 (m, 3H), 7.02 – 6.94 (m, 1H), 6.78 – 6.67 (m, 1H), 4.73 – 4.45 (m, 2H), 4.05 (dt, J = 13.6, 6.5 Hz, 1H), 3.87 (dd, J = 10.3, 5.2 Hz, 1H), 3.75 – 3.45 (m, 1H), 3.21 – 2.94 (m, 1H), 1.98 – 1.61 (m, 1H), 1.51 – 1.36 (m, 1H), 1.32 (dd, J = 9.5, 4.7 Hz, 1H), 1.22 – 1.11 (m, 1H), 1.03 – 0.66 (m, 2H).

cis-N-(2-(2-(Hydroxymethyl)cyclopropyl)ethyl)-1H-indole-2-carboxamide (33)

Pd⁰/C (2.70 mg, 0.026 mmol) was added to a solution of 32 (90.0 mg, 0.26 mmol) in EtOH (10 mL). The mixture was shaken in a Parr apparatus, under 50 psi of H₂ pressure, for 1 h. The suspension was filtered over celite, the solvent evaporated, and the residue used in the following step without further purification.

rel-cis-N-(2-((1r,2r)-2-(((2S,5S)-2-(6-Aminopyridin-3-yl)-5-methylmorpholino)methyl)cyclopropyl)ethyl)-1H-indole-2-carboxamide (34)

The compound was prepared following the same procedure described for 27, starting from 33 (60.0 mg, 0.23 mmol). The crude material was initially purified by flash chromatography eluting with 5% DMA, followed by preparative chiral HPLC to isolate 10 mg (10% yield) of the cis diastereomers mixture. Chiralpak AD-H 21 mm x 250 mm x 5 μm: mobile phase: isocratic 50% 2-PrOH in hexanes; temperature: 25 °C; flow rate: 15–18 mL/min; injection volume: 3 mL (~10 mg/mL sample concentration); detection at λ 254 nm and 280 nm with the support of ELS detector. Analytical chiral HPLC analysis performed using Chiralpak AD-H analytical column (4.5mm x 250mm – 5 μm particle size); mobile phase: isocratic 50% 2-PrOH in hexanes, multiple DAD λ absorbance signals measured in the range of 210–280 nm; Rt 24.256 min and 25.316 min, purity % >95%; dr 40:60, de cis>trans >99:1; ¹H NMR (400 MHz, CDCl₃) δ 9.11–9.23 (2x s, 1H), 8.20 – 7.99 (m, 1H), 7.65 (d, J = 8.1 Hz, 1H), 7.56 – 7.46 (m, 1H), 7.47 – 7.36 (m, 1H), 7.30 (dd, J = 8.4, 1.4 Hz, 1H), 7.14 (dd, J = 8.0, 7.0 Hz, 1H), 7.04 – 6.72 (m, 1H), 6.49 (dd, J = 8.3, 4.2 Hz, 2H), 4.42 (s, 2H), 3.98 – 3.41 (m, 5H), 3.05 – 2.70 (m, 1H), 2.71 – 2.21 (m, 2H), 1.80 (m, 1H), 1.37 – 1.19 (m, 2H), 1.08 (dd, J = 20.5, 6.6 Hz, 3H), 1.04 – 0.66 (m, 3H). HRMS MS/MS (C₂₅H₃₁N₅O₂ + H⁺): calculated 434.25505, found 434.25472.

Molecular modeling methods

The active state cryoEM structures of dopamine D₂ (PDBID: 7JVR) and dopamine D₃ (PDBID: 7CMU, 7CMV) were extracted from the RCSB server. All the objects except the receptor protein subunit and the crystallized ligand were deleted, and this was followed by the addition of hydrogens and optimization of the side-chain residues. Ligands were sketched with chiral definitions, assigned formal charges, and energy-optimized prior to molecular docking. The ligand docking box for potential grid docking was defined to contain the extracellular half of the protein, and all-atom docking was performed using the energy minimized structures for all ligands with an effort value of 10. Several combinations of rigid-protein and side chain flexible docking were performed (including Asp^3.32, Ser^5.42, and Ser^5.46 residues). The best-scored docking solutions were obtained from these docking runs were further optimized by iterative rounds of minimization and Monte Carlo sampling of the ligand conformation, including the surrounding side-chain residues (within 5 Å of the ligand) in the dopamine receptor orthosteric site. All the above-mentioned molecular modeling operations were performed in ICM-Pro v3.9 molecular modeling and drug discovery suite (Molsoft LLC)⁷⁴.

Radioligand Binding Assay Protocols

HEK293 cells stably expressing human D_2LR or D₃R were grown in a 50:50 mix of DMEM and Ham’s F12 culture media, supplemented with 20 mM HEPES, 2 mM L-glutamine, 0.1 mM non-essential amino acids, 1X antibiotic/antimycotic, 10% heat-inactivated fetal bovine serum, and 200 μg/mL hygromycin (Life Technologies, Grand Island, NY) and kept in an incubator at 37 °C and 5% CO₂. Upon reaching 80–90% confluence, cells were harvested using pre-mixed Earle’s Balanced Salt Solution (EBSS) with 5 mM EDTA (Life Technologies) and centrifuged at 3,000 rpm for 10 min at 21 °C. The supernatant was removed, and the pellet was resuspended in 10 mL hypotonic lysis buffer (5 mM MgCl₂, 5 mM Tris, pH 7.4 at 4 °C) and centrifuged at 14,500 rpm (~25,000 g) for 30 min at 4 °C. The pellet was then resuspended in fresh binding buffer. A Bradford protein assay (Bio-Rad, Hercules, CA) was used to determine the protein concentration. For [³H]-(R)-(+)-7-OH-DPAT binding studies, membranes were harvested fresh; the binding buffer was made from 50 mM Tris, 10 mM MgCl₂, 1 mM EDTA, pH 7.4. On test day, each test compound was diluted into half-log serial dilutions using 30% DMSO vehicle. When it was necessary to assist solubilization of the drugs at the highest tested concentration, 0.01%−0.1% acetic acid (final concentration v/v) was added alongside the vehicle. Membranes were diluted in fresh binding buffer. Radioligand competition experiments were conducted in 96-well plates containing 300 μl fresh binding buffer, 50 μl of diluted test compound, 100 μl of membranes (40–80 μg/well and 20–40 μg/well for hD_2LR and hD₃R, respectively), and 50 μl of radioligand diluted in binding buffer ([³H]-(R)-(+)-7-OH-DPAT: 1.5 nM final concentration for hD_2L, and 0.5 nM final concentration for hD₃, American Radiolabeled Chemicals or Perkin Elmer). Aliquots of [³H]-(R)-(+)-7-OH-DPAT solution were also quantified accurately to determine how much radioactivity was added, taking in account the experimentally determined counter efficiency. Nonspecific binding was determined using 10 μM (+)-butaclamol (Sigma-Aldrich, St. Louis, MO) and total binding was determined with 30% DMSO vehicle. All compound dilutions were tested in triplicate and the reaction incubated for 90 min at RT. The reaction was terminated by filtration through Perkin Elmer Uni-Filter-96 GF/B, presoaked for 90 min in 0.5% polyethylenimine, using a Brandel 96-Well Plates Harvester Manifold (Brandel Instruments, Gaithersburg, MD). The filters were washed 3 times with 3 mL (3 × 1 mL/well) of ice-cold binding buffer. 65 μl Perkin Elmer MicroScint 20 Scintillation Cocktail was added to each well and filters were counted using a Perkin Elmer MicroBeta Microplate Counter. IC₅₀ values for each compound were determined from dose-response curves and K_i values were calculated using the Cheng-Prusoff equation. Alternatively, K_i values were also directly obtained from the one site competition fitting of the binding dose-response curves. K_d values for [³H]-(R)-(+)-7-OH-DPAT were determined via separate homologous competitive binding experiments. When a complete inhibition couldn’t be achieved at the highest tested concentrations, K_i values have been extrapolated by constraining the bottom of the dose-response curves (= 0% residual specific binding) in the non-linear regression analysis. These analyses were performed using GraphPad Prism version 9.00 for Macintosh (GraphPad Software, San Diego, CA). All the results were rounded to the third significant figure. K_i values were determined from at least 3 independent experiments and are reported as mean ± SEM. Binding experiments using [³H]-N-methylspiperone as antagonist radioligand were performed following the previously reported protocol⁶, using the same stable cell lines, and membrane preparations described above for both hD_2LR and hD₃R.

ABBREVIATIONS

GPCRs

G protein-coupled receptors

OBS

Orthosteric binding site

PP

Primary pharmacophore

SBP

Secondary binding pocket

SP

Secondary pharmacophore