Ether modifications to 1-[2-(3,4-dimethoxyphenyl)ethyl]-4-(3-phenylpropyl)piperazine (SA4503): effects on binding affinity and selectivity for sigma receptors and monoamine transporters

Abstract

Two series of novel ether analogs of the sigma (σ) receptor ligand 1-[2-(3,4-dimethoxyphenyl)ethyl]-4-(3-phenylpropyl)piperazine (SA4503) have been prepared. In one series, the alkyl portion of the 4-methoxy group was replaced with allyl, propyl, bromoethyl, benzyl, phenethyl, and phenylpropyl moieties. In the second series, the 3,4-dimethoxy was replaced with cyclic methylenedioxy, ethylenedioxy and propylenedioxy groups. These ligands, along with 4-O-des-methyl SA4503, were evaluated for σ₁ and σ₂ receptor affinity, and compared to SA4503 and several known ether analogs. SA4503 and a subset of ether analogs were also evaluated for dopamine transporter (DAT) and serotonin transporter (SERT) affinity. The highest σ₁ receptor affinities, K_i values of 1.75 nM – 4.63 nM, were observed for 4-O-des-methyl SA4503, SA4503 and the methylenedioxy analog. As steric bulk increased, σ₁ receptor affinity decreased, but only to a point. Allyl, propyl and bromoethyl substitutions gave σ₁ receptor K_i values in the 20 nM – 30 nM range, while bulkier analogs having phenylalkyl, and Z- and E-iodoallyl, ether substitutions showed higher σ₁ affinities, with K_i values in the 13 nM – 21 nM range. Most ligands studied exhibited comparable σ₁ and σ₂ affinities, resulting in little to no subtype selectivity. SA4503, the fluoroethyl analog and the methylenedioxy congener showed modest six- to fourteen-fold selectivity for σ₁ sites. DAT and SERT interactions proved much more sensitive than σ receptor interactions to these structural modifications. For example, the benzyl congener (σ₁ K_i = 20.8 nM; σ₂ K_i = 16.4 nM) showed over 100-fold higher DAT affinity (K_i = 121 nM) and 6-fold higher SERT affinity (K_i = 128 nM) than the parent SA4503 (DAT K_i = 12650 nM; SERT K_i = 760 nM). Thus, ether modifications to the SA4503 scaffold can provide polyfunctional ligands having a broader spectrum of possible pharmacological actions.

1. Introduction

Sigma (σ) receptors can be classified into distinct σ₁ and σ₂ subtypes that play prominent roles in certain cancers and in central nervous system (CNS) disorders. Malignancies often express high levels of σ₁ and σ₂ receptors with respect to the normal tissue, the σ₂ receptors serve as markers of cell proliferation, and σ₁ antagonists / σ₂ agonists have potential for cancer therapy.^1–3 In the CNS realm, σ₁ receptor agonists improve cognition, show antidepressant and anxiolytic properties, and promote neuroregeneration.^4–6 Moreover, selective σ₁, and perhaps σ₂, receptor antagonists show potential as medications for treatment of psychostimulant abuse.^7–10 Compounds from many different chemical classes show varying degrees of affinity and selectivity for σ receptors, and structurally diverse ligands including progesterone, N,N′-dimethyltryptamine and sphingolipids have been suggested as endogenous ligands.^11,12

The σ₁ receptor has been cloned,^13,14 while the primary structure of the σ₂ receptor remains elusive.³ Protein biochemistry studies and homology modeling efforts are now revealing specific features of the σ₁ receptor binding site at the amino acid level.¹⁵ Less is known about σ₂ receptors, but recent studies show the orthosteric binding site to be located within the progesterone receptor membrane component 1 (PGRMC1) protein complex.^3,16 Nonetheless, medicinal chemistry studies show a degree of commonality, with both σ₁ and σ₂ receptor pharmacophores simply described by an amine binding site flanked by two hydrophobic binding pockets.^17,18 These basic models are fit well by N,N′-disubstituted piperazines having phenethyl and phenylpropyl groups to occupy the hydrophobic regions.

Two active series from this structural class of σ receptor ligands have attracted attention as potential pharmaceuticals. These are denoted as either “SA” compounds^19,20 or “YZ” compounds (Fig. 1).^21,22 They differ in the patterns and types of substituents on the phenethyl and phenylpropyl moieties, and typically display relatively high σ₁ receptor binding affinity accompanied by modest selectivity against σ₂ sites. Subtle structural differences influence their σ₁ and σ₂ receptor binding, as well as the behavioral pharmacology observed in animal models.^22–27 SA4503 (Fig. 1), the most widely studied ligand from this class, is also known as cutamesine and is regarded as a σ₁ receptor agonist.^4–6,19 SA4503 causes dissociation of immunoglobulin binding protein from the σ₁ receptor, a definitive biochemical test of stimulation.²⁸ SA4503 has been evaluated in Phase II clinical trials for treatment of major depressive disorder²⁹ and to aid in functional recovery after stroke.^30,31 In the stroke trial,³⁰ SA4503 proved well tolerated, and the data indicated improvements in individuals who received a 3 mg/d dosage for 28 days. Interestingly, SA4503 shows mixed properties in recent behavioral studies, and can either potentiate or attenuate certain effects of cocaine and methamphetamine in mice as a function of dose.^23,24 By contrast, the YZ analogs profile as σ₁ receptor antagonists based upon their ability to block cocaine-induced convulsions in mice.²²

Figure 1.

Representative “SA” and “YZ” sigma receptor ligands.

Although a number of different SA / YZ compounds have been studied, relatively little information is available regarding the effects of ether modifications on σ receptor binding. Findings come primarily from the development of radiolabeled SA4503 analogs for molecular imaging applications.^32,33 The 4-methoxy position of SA4503 has been isotopically radiolabeled with carbon-11 for positron emission tomography (PET) studies of central and peripheral σ₁ receptors in human beings and animal models of disease.^34,35 The fluorine-18 labeled 4-fluoroethoxy³⁶ (FE-SA4503, Fig. 1) and fluoromethoxy³⁷ (FM-SA4503, Fig. 1) analogs have also been prepared, and used for PET studies of cerebral σ₁ receptors in conscious monkeys.^36,37 In our hands, SA4503 displays a 4.6 nM apparent affinity (K_i) for σ₁ receptors with 14-fold selectivity for σ₁ over σ₂ sites.³⁸ In side-by-side binding comparisons, the fluoroethyl modification caused a two-fold reduction in affinity for both σ₁ and σ₂ receptors while preserving the σ₁ subtype selectivity.³⁸ As previously discussed in detail,³⁸ these σ receptor subtype selectivities are quite different from those initially reported for SA4503¹⁹ and FE-SA4503,³⁶ primarily as a consequence of the σ₂ receptor binding results. In work aimed toward radioiodinated SA4503 analogs, the 4-methoxy group was replaced by (Z)- and (E)-iodoallyloxy substituents (IA-SA4503, Fig. 1).³⁹ These modifications reduced σ₁ affinity and increased σ₂ affinity, leading to a pair of non-selective isomeric ligands having K_i values of 11 – 20 nM for both subtypes.

To gain a better understanding of the qualitative structure - affinity relationships associated with ether modifications of SA4503, we prepared nine novel analogs (Fig. 2) for a more systematic investigation. Two types of structural modifications were investigated. The first involved replacements of the 4-methoxy group with ethers of increasing steric bulk and hydrophobicity, as well as differing potential for π and other interactions (1 – 6). The second involved replacement of the 3,4-dimethoxy moiety with cyclic ethers of different ring sizes (7 – 9). Each compound was tested in vitro for affinity and selectivity of binding to σ₁ and σ₂ receptors using standard techniques. The σ receptor binding affinities of the known 4-O-des-methyl-SA4503^36b (Fig. 1) have not been reported, and this material was tested as well.

Figure 2.

Novel SA4503 ether analogs 1–9.

In previous work, SA4503 showed low affinity for about 40 different receptors, ion channels and second messenger systems,¹⁹ although a significant secondary interaction with the vesicular acetylcholine transporter, K_i = 50 nM, has been observed.⁴⁰ Interactions of SA4503 with dopamine transporters (DAT) and serotonin transporters (SERT) have not been reported to our knowledge. Several of the novel SA4503 ether analogs were designed to bear a structural resemblance to ligands such as GBR 12909 (vanoxerine) and certain rimcazole analogs that show high DAT and SERT affinities accompanied by weaker interactions with the σ receptor subtypes.^41,42 Considering the importance of σ receptors and the DAT in mediating the effects of cocaine, compounds having actions at both sites are of interest from the viewpoint of psychostimulant medications development.^41–43 As exemplified by fluvoxamine,⁴ ligands with actions at both σ receptors and the SERT are of interest as medications for treatment of depression and anxiety.^4–6 Accordingly, we determined the affinities of SA4503 and a subset of the ether analogs for DAT and SERT with the aim of identifying structural modifications that would provide polyfunctional ligands having a broader spectrum of pharmacological actions.

2. Results and discussion

2.1. Chemistry

Target compounds 1 – 9 were prepared by the routes shown in Scheme 1. Direct reaction of the phenolate anion generated in situ from 4-O-des-methyl SA4503^36b with primary alkyl, allyl or benzyl bromides provided piperazines 1 – 6 (Scheme 1). Isolated yields from the Williamson ether synthesis ranged from 44 – 90% for 1, 2, 5 and 6 using phase transfer conditions, 40% KOH / tetra-n-butylammonium hydroxide (TBAH), at 50 °C. This set of reaction conditions emerged from studies on the synthesis of compound 1 by varying the type of base (NaOH, KOH) and the number of molar equivalents of base and alkylating agent. For the more reactive electrophile, allyl bromide, the reaction was performed from 0 °C to ambient temperature to give 3 in 57% yield. The phase transfer method was not amenable to use with benzyl bromide, and a number of products were formed that were not characterized. However, treatment of 4-O-des-methyl SA4503 with benzyl bromide and K₂CO₃ in absolute EtOH at ambient temperature overnight provided 4 in 35% yield. These reactions were not optimized further.

Scheme 1.

Synthesis of SA4503 ether analogs 1 – 6. Reagents and reaction conditions: (a) 1, 2, 5, 6: corresponding alkyl bromide, KOH / TBAH, 50 °C, 50 min; 3: allyl bromide, KOH / TBAH, 0 °C to ambient temperature, 50 min; 4: benzyl bromide, EtOH / K₂CO₃, 25 °C, N₂ overnight.

Cyclic ethers 7 – 9 were synthesized by coupling substituted phenylacetic acids to 1-(3-phenylpropyl)piperazine to yield amide intermediates that were subsequently reduced with LiAlH₄ (Scheme 2). The 3,4-(methylenedioxy)phenylacetic acid 10 required for preparation of 7 was commercially available. The known 3,4-(ethylenedioxy)phenylacetic acid 11⁴⁴ and 3,4-(propylenedioxy)phenylacetic acid 12⁴⁵ required for preparation of 8 and 9, respectively, were synthesized in 52% and 42% yields by treatment of 3,4-dihydroxyphenylacetic acid with K₂CO₃ and the appropriate alkyl dibromide in ethylene glycol at 120 °C for 4.5 h. Kashima and colleagues⁴⁶ reported these conditions for the synthesis of cyclic ethers from catechols. This onestep route is a convenient alternative to the multi-step method of Sasamoto,⁴⁴ and uses milder conditions than the method of Alker et al.⁴⁵ that calls for KOH at reflux for 17 h. Coupling reactions were then performed using 1-hydroxybenzotriazole monohydrate (HOBt·H₂O), 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDC·HCl) and 4-methylmorpholine in CH₂Cl₂ at 0 °C under N₂ to give 13 – 15. These amides were isolated, characterized and then reduced with LiAlH₄ in THF to give 7 – 9. Coupling reactions were accomplished in > 90% yields, while the reduction steps provided 33 – 85% isolated yields of the final targets.

Scheme 2.

Synthesis of SA4503 ether analogs 7 – 9. Reagents and reaction conditions: (a) BrCH₂CH₂Br, K₂CO₃ / HOCH₂CH₂OH, 120 °C, 4.5 h; (b) Br(CH₂)₃Br, K₂CO₃ / HOCH₂CH₂OH, 120 °C, 4.5 h; (c) 1-(3-phenylpropyl)piperazine, HOBt•H₂O / EDC•HCl / 4-methylmorpholine, 0 °C, N₂ overnight; (d) LiAlH₄ / THF, 0 °C - ambient temperature, 7.5 h.

2.2. Sigma receptor binding

Table 1 shows binding affinities of 4-O-des-methyl SA4503 and novel ether analogs 1 – 9 for σ₁ and σ₂ receptors in guinea pig brain membranes. Published data, obtained using the same assay conditions, for SA4503 and FE-SA4503,³⁸ as well as (Z)- and (E)-IA-SA4503,³⁹ are included for comparison. Literature binding data for several YZ analogs and other reference ligands are also included. In addition, Table 1 shows a spot test of interactions with the DAT using a single high concentration (10000 nM) of SA4503 and analogs. This test was performed to prioritize the selection of compounds for more detailed studies (see Section 2.3).

Table 1.

σ Receptor affinity of SA4503 and novel ether analogs^a,^b, along with literature data for selected YZ and reference compounds.

	R	Sigma receptor binding		DATc
		σ1 Ki (nM)	σ2 Ki (nM)	σ2 / σ1	% Inh.
	—H (des-methyl SA4503)	1.75 ± 0.17	6.72 ± 0.21	3.8	20
	—CH3 (SA4503)d	4.63 ± 0.21	63.09 ± 4.33	13.6	23
	—CH2CH2F (FE-SA4503)d	8.03 ± 0.41	113.2 ± 11.7	14.1	19
	—CH2F (FM-SA4503) e	6.4 ± 1.3	250 ± 16	39.0	--
	(E)—CH2CH=CHI f	16.91 ± 1.28	18.20 ± 1.99	1.1	69
	(Z)—CH2CH= CHI f	10.67 ± 1.32	16.71 ± 1.20	1.6	94
1	—CH2CH2CH3	25.60 ± 2.56	33.40 ± 1.11	1.3	43
2	—CH2CH2Br	25.37 ± 2.02	27.12 ± 0.96	1.1	38
3	—CH2CH=CH2	30.13 ± 1.24	42.10 ± 0.66	1.4	44
4	—CH2Ph	20.79 ± 1.10	16.43 ± 1.13	0.8	93
5	—CH2CH2Ph	13.26 ± 0.82	16.57 ± 0.64	1.2	89
6	—CH2CH2CH2Ph	18.45 ± 0.48	23.31 ± 3.34	1.3	97
7	—Methylenedioxy—	2.55 ± 0.19	14.83 ± 0.43	5.8	42
8	—Ethylenedioxy—	10.20 ± 0.52	35.20 ± 1.69	3.4	33
9	—Propylenedioxy—	19.10 ± 3.11	52.10 ± 4.91	2.7	43
	YZ-184 (2-OCH3) g	3.9 ± 0.6	7.5 ± 0.2	1.9	--
	YZ-185 (3-OCH3) g	1.4 ± 0.2	10.2 ± 0.5	7.3	--
	YZ-067 (4-OCH3) g	1.3 ± 0.3	28.6 ± 1.9	22	--
	YZ-069 (3,4-di-Cl) g	2.2 ± 0.2	38.8 ± 1.6	24	--
	(+)-Pentazocine d	1.62 ± 0.15	728.4 ± 48.8	450	--
	Ditolylguanidine (DTG) d	35.45 ± 1.20	39.87 ± 2.61	1.1	--

^aK_i values are means ± SEM for 3 – 6 trials, each performed in duplicate.

^b1.0 nM [³H](+)-pentazocine (σ₁), 3.0 nM [³H]ditolylguanidine / 200 nM (+)-pentazocine (σ₂); guinea pig brain membranes.

^c% Inhibition for a 10000 nM concentration against [¹²⁵I]RTI-121 specific binding in mouse striatal membranes. Binding data reported in

^dRef. 38,

^eRef. 37,

^fRef. 39 and

^gRef. 22.

Of the fourteen compounds studied, 4-O-des-methyl SA4503 showed the highest binding affinity for both σ₁ and σ₂ receptors (Table 1). The σ₂ receptor binding is sensitive to this modification, with a ten-fold increase in affinity for 4-O-des-methyl SA4503 compared to SA4503. These findings are consistent with our recent report of a structurally similar phenol that showed much higher σ₂ receptor affinity than the corresponding ligand bearing a methoxy group.⁴⁷ 4-O-des-methyl SA4503 is the only ligand tested that can be a hydrogen bond donor. By contrast, the hydrogen bond acceptors FE-SA4503 and FM-SA4503 showed the poorest σ₂ receptor affinities of the series (Table 1). YZ-184 and YZ-185 have only hydrogen at the 4-position, but still display high σ₂ receptor affinity, while YZ-067 and YZ-069 that have a chlorine or methoxy at the 4-position show a somewhat lower σ₂ receptor affinity (Table 1). The data indicate that either having a hydroxyl group or no substituent at the 4-position promotes σ₂ binding. Rhoades et al.⁴⁸ have developed a three-dimensional model for the σ₂ receptor pharmacophore, based upon ligands from seven different structural classes, that includes at least four hydrogen bond acceptor groups located at different positions of the binding pocket. Thus, the hydrogen bond donor properties of the phenol may contribute to high affinity σ₂ binding.

For the SA4503 analogs described thus far, a smaller substituent volume leads to higher affinity binding to both σ₁ and σ₂ receptors that is accompanied by modest selectivities for the σ₁ subtype. In this regard, oxygen substituents at both the 3- and 4-positions of the phenethyl ring are not essential for high affinity σ receptor binding. For example, monomethoxy analog YZ-185 binds to σ₁ and σ₂ receptors with K_i values of 1.4 nM and 10.2 nM, respectively (Table 1).^21,22 The bulkier and more hydrophobic propyl, bromoethyl and allyl ether substitutions of analogs 1 – 3 proved detrimental to σ₁ receptor binding, but slightly advantageous for σ₂ receptor binding as compared to SA4503 (Table 1). These three compounds showed the poorest σ₁ receptor affinities of the compounds tested, and displayed K_i values of 25 nM – 30 nM (Table 1). Their σ₂ receptor affinities of 27 nM – 42 nM were also nearly equivalent, and engendered no appreciable subtype selectivity. The similarity of binding data across analogs 1 – 3 suggests that neither halogen bonding⁴⁹ of bromine, nor π effects of the ethylene system, contribute significantly to the interactions. Further, these data indicate that the high electronegativity of fluorine, rather than a steric constraint, is responsible for the low σ₂ receptor affinity of FE-SA4503 and FM-SA4503. Fluorine participation in hydrogen bonding acceptor or other dipole-dipole interactions⁵⁰ that inhibit σ₂ receptor binding fits with the notion that hydrogen bonding donor properties help promote the high affinity σ₂ receptor binding of 4-O-des-methyl SA4503.

The homologous series of phenylalkyl ethers 4 – 6 had no appreciable subtype selectivity, and displayed a tight range of K_i values, 13 nM – 23 nM, for binding to both σ₁ and σ₂ sites (Table 1). As observed for analogs 1 – 3, incorporation of these bulkier and more hydrophobic substituents reduced σ₁ but increased σ₂ receptor affinity with respect to SA4503. The presence of an aromatic ring modestly increased the affinities of 4 – 6 for both σ₁ and σ₂ receptors as compared to alkyl and allyl analogs 1 – 3 (Table 1). Phenethyl ether analog 5 gave the highest overall σ₁ and σ₂ affinity of this series, but the chain length variations had small effects.

The previously reported (Z)- and (E)-iodoallyl ether analogs show σ₁ and σ₂ receptor affinities comparable to those of novel phenylalkyl ether analogs 4 – 6 (Table 1). Notably, the presence of iodine confers 2- to 3-fold higher σ₁ and σ₂ binding affinities for the iodoallyl analogs over the simple allyl analog 3. The similarity of σ receptor affinities for the (Z)- and (E)- configurations indicates little sensitivity to these particular geometric constraints. Thus, the extended ether analogs having bulky, hydrophobic and polarizable aromatic ring or iodine substituents are readily accommodated by the σ receptor binding pockets. While discrete π–π stacking interactions for the phenylalkyl ether moieties of 4 – 6 may contribute to binding, the overall data suggest that substituent bulk and hydrophobicity are main factors. These properties confer higher σ₁ and σ₂ receptor binding affinities for 4 – 6, as well as (Z)- and (E)- IA-SA4503, than observed for analogs 1 – 3 that have smaller molecular volumes.

Cyclic ether 7, with its compact methylenedioxy group, showed 2- to 4-fold increases in both σ₁ and σ₂ receptor affinity with respect to SA4503 (Table 1). Compared to each other, cyclic ethers 7 – 9 showed progressive 7.5- to 3.5-fold decreases in both σ₁ and σ₂ receptor affinities when moving from methylenedioxy (7) to the bulkier ethylenedioxy (8) and propylenedioxy (9) congeners. Thus, these three constrained ether analogs of SA4503 follow the pattern discussed earlier where a smaller substituent volume leads to relatively higher σ₁ and σ₂ receptor affinities accompanied by modest σ₁ selectivity.

2.3. DAT and SERT binding

SA4503 and analogs varied widely in their potential to interact with the DAT (Table 1). Using a 10,000 nM concentration, their ability to inhibit [¹²⁵I]-RTI-121 binding ranged from 19% to 97%. SA4503, FE-SA4503 and 4-O-des-methyl SA4503 exhibited the lowest inhibition (19% – 23%), while the alkyl / allyl ethers 1 – 3 and the cyclic ethers 7 – 9 showed a moderate level of inhibition (38% – 44%). Phenylalkyl ethers 4 – 6 and the (Z)- and (E)-iodoallyl ethers analogs gave the highest degree of inhibition (69% – 97%). Thus, the presence of large, hydrophobic ether substituents at the 4-O-position of the SA4503 phenethyl leads to stronger DAT binding.

Based upon these results, SA4503, as well as selected ether analogs displaying moderate (1, 3, 7) or high (4, 6, (Z)- and (E)-IA-SA4503) potential for DAT binding, were studied in more detail. Over a 100-fold range of DAT affinities was observed, while SERT affinities varied by only 7-fold (Table 2). SA4503 displayed the weakest interactions with both the DAT and SERT, with a decided 16-fold higher affinity for SERT over DAT. The DAT K_i of 12650 nM proved about 10-fold lower than the DAT K_i values previously reported for monomethoxy analogs YZ-185 and YZ-067 (Table 2).²² By contrast, SERT affinities for SA4503 (K_i = 760 nM) and YZ-185 (K_i = 985 nM)²² were quite similar. Somewhat surprisingly, the DAT affinity for SA4503 proved close to that previously reported by Matsumoto et al.²² for dichloro analog YZ-069 (Table 2). The DAT and SERT binding data for SA4503 were replicated, with no significant differences (n = 4; t-test, P > 0.05), using fresh serial dilutions prepared from powder.

Table 2.

DAT and SERT affinity of SA4503 and novel ether analogs^a,^b,^c, along with literature data for selected YZ and reference compounds.

	R	DAT Ki (nM) b	DAT / σ1f	SERT Ki (nM) c	SERT / σ1f
	—CH3 (SA4503)	12650 ± 832	2732	760 ± 31	164
	(E) —CH2CH=CHI	560 ± 57	33	174 ± 11	10.3
	(Z) —CH2CH=CHI	237 ± 13	22	ND	ND
1	—CH2CH2CH3	1821 ± 174	71	124 ± 13	4.8
3	—CH2CH=CH2	2083 ± 156	69	295 ± 19	9.8
4	—CH2Ph	121 ± 16	6	128 ± 6	6.2
6	—CH2CH2CH2Ph	87 ± 5	5	ND	ND
7	—Methylenedioxy—	1840 ± 311	721	468 ± 41	184
	YZ-185 (3-OCH3) d	1408 ± 164	1006	985 ± 126	703
	YZ-067 (4-OCH3) d	1466 ± 19	1128	1726 ± 57	1328
	YZ-069 (3,4-di-Cl) d	8196 ± 1005	3725	2865 ± 484	1302
	GBR 12909 e	12.0 ± 1.9	0.04 f	24.1 ± 2.3	0.78 f

^aK_i values are means ± SEM for n = 3 – 6 trials, each performed in duplicate, except for DAT assays of compounds 6 and (Z) —CH₂CH=CHI, where n = 2. ND = not determined.

^b15 pM [¹²⁵I]RTI, mouse striatal membranes.

^c0.3 nM [³H]paroxetine, mouse whole brain membranes. Binding data reported in

^dRef. 22 and

^eRef. 41.

^fBased upon the σ₁ K_i values reported in Table 1, and the σ₁ K_i value of 318 nM reported for GBR 12909 in Ref. 41.

All novel ether analogs tested proved more potent than SA4503 for binding to both DAT and SERT (Table 2). Ligands 1, 3 and 7 displayed nearly equal affinities, approximately 2000 nM, for DAT sites, while SERT binding showed more sensitivity to structure. Their SERT K_i values of 124 nM to 468 nM varied by 4-fold. Benzyl analog 4 and the (E)-iodoallyl congener displayed SERT K_i values of 128 nM and 174 nM, respectively (Table 2). Thus, further increases in ether substituent size and hydrophobicity preserved, but did not enhance, SERT affinity. By contrast, DAT affinities of the SA4503 ether analogs increased substantially in conjunction with increasing substituent size and hydrophobicity. Phenylalkyl ethers 4 and 6 gave the highest DAT affinities of the ligands tested, with K_i values of 121 nM and 87 nM (Table 2), and exhibited over 100-fold higher DAT affinity than SA4503. Considering the lower DAT affinity of the (Z)- and (E)-iodoallyl ether analogs (Table 2), the terminal aromatic rings on the ether substituents of 4 and 6 play an important role in this interaction.

We anticipated the possibility of finding moderately high DAT and SERT affinities for the phenylalkyl ether analogs of SA4503 based upon their structural resemblance to GBR 12909 (Fig. 3), a ligand that shows high DAT (K_i = 12 nM) and SERT (K_i = 24 nM) affinities accompanied by weaker interactions with both σ₁ (K_i = 318 nM) and σ₂ (K_i = 116 nM) sites.⁴¹ A more recent determination puts the apparent affinity of GBR 12909 for σ₁ receptors at 50.8 nM.⁴² SA4503 and GBR 12909 are flexible molecules that can adopt hundreds of closely related low energy conformations.^20,51 In both cases, the piperazine rings are thought to take a chair conformation, and the phenyl alkyl chains are fully extended. These features have been observed in crystal structures of SA4503 and analogs.²⁰ One possible structural superimposition of GBR 12909 with 4, the benzyl ether analog of SA4503, is given in Fig. 3. Poses shown have the phenylpropyl piperazine moieties aligned, and were minimized using the MMFF94 force field routine of program Avogadro.⁵² Qualitatively, the overall molecular dimensions of the two compounds are comparable, and there is substantial spatial overlap between the terminal aromatic ring of the ether substituent of 4 and the diphenylmethoxy ring system of GBR 12909. GBR 12909,⁴¹ as well as the YZ-series of analogs,²² typically display low affinities, > 1000 nM, for the norepinephrine transporter. Consequently, the affinities of SA4503 and analogs were not determined for this site.

Figure 3.

Comparison of GBR 12909 with 4, the benzyl ether analog of SA4503, and superimposition of one of many possible low energy conformers from each using program Avogadro.

3. Conclusions

SA4503, 4-O-des-methyl-SA4503 and two series of ether analogs have been prepared and evaluated for σ₁ and σ₂ receptor binding. The 14 compounds tested could be divided into two groups having different structure affinity relationships. The first group included 6 compounds, SA4503, 4-O-des-methyl SA4503, FE-SA4503 and cyclic ether analogs 7 – 9. These displayed relatively high σ₁ receptor affinities, K_i values 1.75 nM – 19 nM, and variable σ₂ receptor affinities, K_i values 6.7 nM – 113 nM, leading to modest σ₁ receptor selectivities. The other group of eight SA4503 analogs had bulkier, hydrophobic ether substituents, and included alkyl / allyl ethers 1 – 3, phenylalkyl ethers 4 – 6 and the (Z)- and (E)-iodoallyl ether analogs. These compounds showed moderately high, nearly equal affinities for both σ₁ and σ₂ receptors, leading to ligands with little to no subtype selectivity. Considering there are two piperazine nitrogens in each ligand and two main hydrophobic binding pockets for each σ receptor subtype, at least eight different binding motifs are possible. Thus, one set of motifs may be favored for the binding of SA4503 and the compact analogs described above, while different motifs may be preferred as the ether substituents become larger and more hydrophobic. In this regard, some effects observed are likely to be unique to the exact position and type of substituent on the SA4503 scaffold. Diverse substituent pattern effects on σ receptor binding affinity and subtype selectivity are often found between studies of piperazines that have differing arylalkyl systems.^{17,18,20–22,53–58}

SA4503 and a subset of ether analogs were also evaluated for DAT and SERT interactions. SA4503 displayed very weak binding to the DAT, K_i = 12650 nM. Affinity increased with progressively larger ether substitutions, and the phenylpropyl ether analog 6 showed a 145-fold higher DAT affinity of 87 nM. SERT binding varied 7-fold between SA4503 and analogs. Overall, monoamine transporter binding proved more sensitive than σ receptor binding to these structural modifications of SA4503.

Together, the data provide some insight into how manipulations of the substituents at the 4-position of the phenethyl side chain of SA4503 modulate σ receptor subtype affinity and selectivity. None of the novel ether analogs displayed higher σ receptor subtype affinity or selectivity than SA4503 or known congeners. On the other hand, ether substituents can be chosen that greatly increase DAT and SERT binding affinities, and the work identifies SA4503 as a scaffold for the preparation of ligands that can target multiple receptors. For instance, benzyl ether 4 shows moderately high affinities for σ₁ and σ₂ receptors, as well as for the DAT and SERT. Thus, certain ether analogs of SA4503 may well be polyfunctional in vivo, and would be of interest for behavioral testing in animal models of psychostimulant abuse, depression and anxiety.

4. Experimental section

4.1. Chemistry

4.1.1. General methods

Chemicals and solvents were reagent grade, and used as received from commercial sources. 4-O-des-methyl-SA4503 was prepared according to the literature procedure.^{36b 1}H NMR spectra were obtained on Bruker ARX-250 (250 MHz), DRX-300 (300 MHz) or DRX-500 (500 MHz) spectrometers. Chemical shifts (¹H) are reported for the free base in ppm (δ) relative to internal Me₄Si in CDCl₃. Splitting patterns are identified as singlet (s), broad singlet (br s), doublet (d), triplet (t), apparent pentet (app p), multiplet, (m), doublet of doublets of doublets, (ddd), doublet of doublets of triplets, (ddt). Elemental analyses were determined by Atlantic Microlab, Inc., Norcross GA. Analytical TLC was conducted on Polygram SIL G/UV254 plates, with additional visualization by staining with cerium molybdate. Preparative TLC was conducted on silica gel plates (20 × 20 cm, 1000 μm) from Analtech Inc., Newark DE. Column chromatography was performed using Silicycle^® ultra pure silica gel (230–400 mesh) under N₂ pressure. For storage and ease of handling, final targets were converted into the di-HCl salts. The free bases were dissolved in EtOH, concentrated HCl was added, the mixtures were evaporated to dryness, and the solid white products were recrystallized. Compounds 1 – 9 showed > 95% purity by reversed-phase HPLC analysis.

General procedure for alkylation of 4-O-des-methyl-SA4503 (1, 2, 3, 5 and 6)

Mixtures of 4-O-des-methyl-SA4503 (200 – 400 mg, 0.56 – 1.12 mmol), 40% KOH (1.1 – 2.2 mL), tetra-n-butylammonium hydroxide (1 M, CH₃OH; 0.1 – 0.2 mL), and excess alkylating agent were heated at 50 °C for 50 min, and then diluted with H₂O (30 mL) and extracted with CH₂Cl₂ (3 × 20 mL). The organic extracts were washed with saturated NaCl solution, dried over MgSO₄, and concentrated under reduced pressure. The crude oily residues, pale yellow to dark brown in color, were purified by silica gel column chromatography (CHCl₃ : CH₃OH = 20 : 1).

1-(3-methoxy-4-propoxyphenethyl)-4-(3-phenylpropyl)piperazine (1)

Treatment of 4-O-des-methyl-SA4503 (200 mg, 0.56 mmol) with 40% KOH (1.1 mL), tetra-n-butylammonium hydroxide (0.1 mL) and 1-bromopropane (1200 mg, 9.74 mmol) according to the general procedure gave 1 (201 mg, 90%) ¹H NMR: (250 Mz, CDCl₃, δ) 1.03 (t, 3H, J = 7.5 Hz, CH₃), 1.77–1.89 (m, 4H, CH₂), 2.35–2.78 (m, 16H, CH₂), 3.84 (s, 3H, OCH₃), 3.94 (t, 2H, J = 7.5 Hz, OCH₂), 6.70–6.80 (m, 3H, CH), 7.13–7.29 (m, 5H, CH). ¹³C NMR: (62.5 MHz, CDCl_3, δ) 10.4, 22.5, 28.6, 33.2, 33.7, 53.2, 55.9, 57.97, 60.6, 70.6, 112.6, 113.2, 120.5, 125.7, 128.2, 128.3, 133.0, 142.1, 146.8, 149.3. Anal. Calcd for the di-HCl salt (C₂₅H₃₈Cl₂N₂O₂): C, 63.96; H, 8.16; N, 5.97. Found: C, 63.91; H, 8.13; N, 5.91; m.p. 212 °C.

1-(4-(2-bromoethoxy)-3-methoxyphenethyl)-4-(3-phenylpropyl)piperazine (2)

Treatment of 4-O-des-methyl-SA4503 (400 mg, 1.13 mmol) with 40% KOH (2.2 mL), tetra-n-butylammonium hydroxide (0.2 mL) and 1,2-dibromoethane (690 mg, 3.66 mmol) according to the general procedure gave 2 (430 mg, 83%). ¹H NMR: (250 Mz, CDCl₃, δ) 1.77–1.89 (m, 2H, CH₂), 2.35–2.78 (m, 16H, CH₂), 3.62 (t, 2H, J = 7.5 Hz, CH₂), 3.84 (s, 3H, CH₃), 4.28 (t, 2H, J = 7.5 Hz, CH₂), 6.70–6.84 (m, 3H, CH), 7.14–7.30 (m, 5H, CH). ¹³C NMR: (62.5 MHz, CDCl_3, δ) 28.6, 29.0, 33.2, 33.7, 53.2, 56.0, 58.0, 60.5, 69.5, 113.0, 115.2, 120.7, 125.7, 128.2, 128.3, 134.7, 142.1, 145.7, 149.8; Anal. Calcd for the di-HCl salt (C₂₄H₃₅BrCl₂N₂O₂): C, 53.94; H, 6.60; N, 5.24. Found: C, 53.94; H, 6.70; N, 5.24; mp 248–251 °C.

1-(4-allyloxy)-3-methoxyphenethyl)-4-(3-phenylpropyl)piperazine (3)

4-O-des-methyl-SA4503 (250 mg, 0.070 mmol) was treated with 40% KOH (1.4 mL) and tetra-n-butylammonium hydroxide (0.12 mL), followed by dropwise addition of allyl bromide (1500 mg, 12.40 mmol) at 0 °C. The mixture was then stirred at room temperature for 50 min, and worked up according to the general procedure to give 3 (160 mg, 57%). ¹H NMR: (300 Mz, CDCl₃, δ) 1.81–1.86 (app p, 2H, J = 7.5 Hz, CH₂), 2.36–2.76 (m, 16H, CH₂), 3.85 (s, 3H, OCH₃), 4.55–4.58 (ddd, 2H, CH₂, J = 5.4, 1.5, 1.5 Hz), 5.23–5.28 (ddt, 1H, J = 10.5, 1.5, 1.5 Hz), 5.35–5.42 (ddt, 1H, J = 17.1, 1.5, 1.5 Hz), 6.00–6.13 (ddt, 1H, J = 17.4, 10.5, 5.4 Hz), 6.70–6.81 (m, 3H, CH), 7.17–7.29 (m, 5H, CH). ¹³C NMR: (75 MHz, CDCl_3, δ) 28.6, 33.2, 33.7, 53.2, 55.8, 58.0, 60.6, 69.9, 112.4, 113.6, 117.7, 120.4, 125.7, 128.3, 128.4, 133.4, 133.5, 142.1, 146.3, 149.3; Anal. Calcd for the di-HCl salt (C₂₅H₃₆Cl₂N₂O₂): C, 64.23; H, 7.76; N, 5.99. Found: C, 64.33; H, 7.71; N, 5.96. m.p. 222 °C.

1-(4-(benzyloxy)-3-methoxyphenethyl)-4-(3-phenylpropyl)piperazine (4)

To a stirred solution of 4-O-des-methyl-SA4503 (250 mg, 0.70 mmol) in absolute EtOH (2.5 mL) was added K₂CO₃ (193 mg, 1.40 mmol) and benzyl bromide (192 mg, 1.12 mmol). The mixture was left under N₂ overnight, and then filtered through Celite. Concentration under reduced pressure followed by preparative TLC (CHCl₃ : CH₃OH = 10 : 1) gave 4 as a pale yellow solid (110 mg, 35%). ¹H NMR: (250 Mz, CDCl₃, δ) 1.82–1.88 (m, 2H, CH₂), 2.38–2.77 (m, 16H, CH₂), 3.89 (s, 3H, CH₃), 5.13 (s, 2H, CH₂), 6.68–6.83 (m, 3H, CH), 7.19–7.46 (m, 10H, CH). ¹³C NMR: (62.5 MHz, CDCl_3, δ) 28.6, 33.2, 33.7, 53.2, 55.9, 58.0, 60.6, 71.2, 112.6, 114.2,120.5, 125.70 127.2, 127.7, 128.2, 128.3, 128.4, 133.6, 137.4, 142.1, 146.5, 149.5. Anal. Calcd for the di-HCl salt (C₂₉H₃₈Cl₂N₂O₂·0.5 H₂O): C, 66.15; H, 7.47; N, 5.32. Found: C, 66.00; H, 7.34; N, 5.38. m.p. 213 °C.

1-(3-methoxy-4-phenylethoxyphenethyl)-4-(3-phenylpropyl)piperazine (5)

Treatment of 4-O-des-methyl-SA4503 (200 mg, 0.56 mmol) with 40% KOH (1.1 mL), tetra-n-butylammonium hydroxide (0.1 mL) and (2-bromoethyl)benzene (1800 mg, 9.733 mmol) according to the general procedure gave 5 (113 mg, 44%). ¹H NMR: (250 Mz, CDCl₃, δ) 1.82–1.89 (m, 2H, CH₂), 2.38–2.78 (m, 16H, CH₂), 3.17 (t, 2H, J = 7.5 Hz, CH₂), 3.86 (s, 3H, CH₃), 4.22 (t, 2H, J = 7.5 Hz, CH₂), 6.74–6.83 (m, 3H, CH), 7.19–7.31 (m, 10H, CH). ¹³C NMR: (62.5 MHz, CDCl_3, δ) 28.6, 33.2, 33.7, 35.8, 53.2, 56.0, 58.0, 60.6, 70.0, 112.8, 113.6, 120.7, 125.7, 125.8, 126.4, 127.3, 128.3, 128.4, 128.4, 129.0, 133.4, 138.1, 142.1, 146.6, 149.4. Anal. Calcd for the di-HCl salt (C₃₀H₄₀Cl₂N₂O₂): C, 67.79; H, 7.58; N, 5.27. Found: C, 67.53; H, 7.64; N, 5.24. m.p. 252 °C.

1-(3-methoxy-4-(3-phenylpropoxy)phenethyl)-4-(3-phenylpropyl)piperazine (6)

Treatment of 4-O-des-methyl-SA4503 (200 mg, 0.56 mmol) with 40% KOH (1.1 mL), tetra-n-butylammonium hydroxide (0.1 mL) and 1-bromo-3-phenylpropane (1460 mg, 7.31 mmol) according to the general procedure gave 6 (214 mg, 81%). ¹H NMR: (300 MHz, CDCl₃, δ) 1.87 (app p, 2H, J = 6.0 Hz, CH₂), 2.15 (app p, 2H, J =6.0 Hz, CH₂), 2.43–2.85 (m, 18H, CH₂), 3.86 (s, 3H, CH₃), 3.99 (t, 2H, J = 6.0 Hz, CH₂), 6.71–6.80 (m, 3H, CH), 7.16–7.31 (m, 10H, CH). ¹³CNMR: (75 MHz, CDCl_3, δ) 28.2, 30.8, 31.3, 32.1, 32.2, 32.3, 33.0, 33.6, 34.4, 52.8, 56.0, 58.5, 61.5, 68.2, 112.7, 113.6, 120.6, 125.8, 125.9, 125.9, 128.4, 128.4, 128.4, 128.5, 132.8, 141.5, 141.8, 146.9, 149.5; Anal. Calcd for the di-HCl salt (C₃₁H₄₂Cl₂N₂O₂): C, 68.24; H, 7.76; N, 5.13. Found: C, 67.94; H, 7.77; N, 5.12. m.p. 227 °C.

3,4-Ethylenedioxyphenylacetic acid (11) and 3,4-propylenedioxyphenylacetic acid (12)

To a solution of 3,4-dihydroxyphenylacetic acid (500 mg, 3.0 mmol) and either 1,2–dibromoethane (1.13 g, 6.0 mmol) or 1,3–dibromopropane (1210 mg, 6.0 mmol) in ethylene glycol (5 mL) was added anhydrous K₂CO₃ (1240 mg, 9.0 mmol). After heating at 120 °C for 4.5 h, the mixtures were cooled, diluted with H₂O (50 mL), acidified to pH < 1 (2 N HCl), and extracted with EtOAc (3 x 20 mL). The organic layers were concentrated under reduced pressure, and residues were purified by column chromatography (EtOAc : Hexane = 50 : 50 for 11; 40 : 60 for 12) to give 11⁴⁴ (301 mg, 52%) or 12⁴⁵ (260 mg, 42%). 11 ¹H NMR: (300 MHz, CDCl₃, δ) 3.56 (s, 2H, CH₂), 4.23 (s, 4H, CH₂), 6.75–6.87 (m, 3H, CH), 11.64 (br s, 1H, -COOH). 12 ¹H NMR: (250 MHz, CDCl₃, δ) 2.19 (app p, 2H, J = 7.5 Hz, CH₂), 3.55 (s, 2H, CH₂), 4.21 (m, 4H, CH₂), 6.81–6.95 (m, 3H, CH), 10.50 (br s, 1H, -COOH).

General procedure for phenylacetic acid coupling to 3-phenylpropylpiperazine (13 – 15)

Equimolar amounts of 3-phenylpropylpiperazine (0.83 – 1.55 mmol), HOBt·H₂O, EDC·HCl and the substituted phenylacetic acid (10 – 12) were treated with 3 equivalents of 4-methylmorpholine in CH₂Cl₂ (7 – 15 mL) under N₂ at 0 °C. Mixtures were stirred at ambient temperature overnight, and then evaporated to dryness under reduced pressure. Residues were diluted with saturated NaHCO₃ and extracted with EtOAc. Organic layers were washed with saturated NaCl solution, dried over MgSO₄, filtered and evaporated to dryness under reduced pressure. Residues were purified by column chromatography (CHCl₃ : CH₃OH = 20 : 1).

2-(3,4-methylenedioxy)phenyl-1-(4-(3-phenylpropyl)piperazin-1-yl)ethanone (13)

The general procedure was followed with 3-phenylpropylpiperazine (170 mg, 0.83 mmol) and 3,4-(methylenedioxy)phenylacetic acid (10, 150 mg, 0.83 mmol) to provide 13 (288 mg, 94%) ¹H NMR: (300 MHz, CDCl₃, δ) 1.76–1.81 (m, 2H, CH₂), 2.25–2.39 (m, 6H, CH₂), 2.62 (t, 2H, J = 6.0 Hz, CH₂), 3.45 (t, 2H, J = 6.0 Hz, CH₂), 3.63 (t, 4H, J = 6.0 Hz, CH₂), 5.92 (s, 2H, OCH₂O), 6.66–6.75 (m, 3H, CH), 7.15–7.27 (m, 5H, CH). ¹³C NMR: (75 MHz, CDCl_3, δ) 28.3, 33.4, 40.4, 41.7, 46.0, 52.6, 53.0, 57.5, 100.9, 108.3, 109.0, 121.5, 125.7, 128.2, 128.3, 128.6, 141.8, 146.31, 147.8, 169.3. Anal. Calcd for C₂₂H₂₆N₂O₃·0.25 H₂O: C, 71.23; H, 7.20; N, 7.55. Found: C, 71.41; H, 7.18; N, 7.54.

2-(3,4-ethylenedioxy)phenyl-1-(4-(3-phenylpropyl)piperazin-1-yl)ethanone (14)

The general procedure was followed using 3-phenylpropylpiperazine (316 mg, 1.55 mmol) and 3,4-(ethylenedioxy)phenylacetic acid (11, 301 mg, 1.55 mmol) to yield 14 (577 mg, 98%). ¹H NMR: (250 MHz, CDCl₃, δ) 1.78 (app p, 2H, J = 7.5 Hz, CH₂), 2.24–2.39 (m, 6H, CH₂), 2.62 (t, 2H, J = 7.5 Hz, CH₂), 3.43 (t, 2H, J = 5.0 Hz, CH₂), 3.63 (t, 2H, J = 5.0 Hz, CH₂), 3.59 (s, 2H, CH₂), 4.20 (s, 4H, OCH₂CH₂O), 6.66–6.81 (m, 3H, CH), 7.15–7.30 (m, 5H, CH). ¹³C NMR: (62.5 MHz, CDCl_3, δ) 28.4, 33.4, 40.0, 41.7, 46.0, 52.7, 53.1, 57.5, 64.2, 64.3, 117.3, 121.5, 125.8, 128.2, 128.3, 128.3, 141.9, 142.3, 143.5, 169.4. Anal. Calcd for C₂₃H₂₈N₂O₃·0.5H₂O: C, 70.93; H, 7.50; N, 7.19. Found: C, 70.73; H, 7.31; N, 7.20.

2-(3,4-propylenedioxy)phenyl-1-(4-(3-phenylpropyl)piperazin-1-yl)ethanone (15)

The general procedure was followed using 3-phenylpropylpiperazine (255 mg, 1.25 mmol) and 3,4-(ethylenedioxy)phenylacetic acid (12, 260 mg, 1.25 mmol) to yield 15 (443 mg, 90%) ¹H NMR: (500MHz, CDCl₃, δ) 1.63 (app p, 2H, J = 7.5 Hz, CH₂), 1.98 (app p, 2H, J = 6.0 Hz, CH₂), 2.09 (t, 2H, J = 5.0 Hz, CH₂), 2.16 (t, 2H, J = 7.5 Hz, CH₂), 2.21 (t, 2H, J = 5.0 Hz, CH₂), 2.47 (t, 2H, J = 7.5 Hz, CH₂), 3.27 (t, 2H, J = 5.0 Hz, CH₂), 3.45 (s, 2H, CH₂), 3.48 (t, 2H, J = 5.0 Hz, CH₂), 3.98–4.01 (m, 4H, CH₂), 6.62–6.77 (m, 3H, CH), 7.00–7.15 (m, 5H, CH). ¹³C NMR: (125 MHz, CDCl_3, δ) 28.3, 31.8, 33.4, 39.8, 41.7, 46.0, 52.6, 53.0, 57.5, 70.4, 121.6, 121.7, 123.4, 125.7, 128.2, 128.3, 130.2, 141.9, 149.9, 151.1, 169.2. Anal. Calcd for C₂₃H₂₈N₂O₃·0.5H₂O: C, 68.64; H, 7.19; N, 6.27. Found: C, 69.09; H, 7.31; N, 6.66.

General procedure for amide reduction (7 – 9)

Solutions of substituted amides (13 – 15, 0.79 – 1.27 mmol) in THF (7 – 12 mL) were added to a suspension of 3 equivalents of LiAlH₄ in THF (7 – 12 mL) under N₂ at 0 °C. Mixtures were then stirred at ambient temperature for 7.5 h, cooled in an ice bath, and quenched by sequential addition of EtOAc (3 mL) and HCl (2 N, 1 mL), followed by addition of solid NaHCO₃ until CO₂ evolution ceased. Mixtures were filtered, diluted with saturated NaHCO₃ and extracted with EtOAc. Organic extracts were dried over MgSO₄, filtered and evaporated to dryness under reduced pressure. Cyclic ether 7 was isolated by preparative TLC (CHCl₃ : CH₃OH = 10 : 1) while 8 and 9 were purified on silica gel columns (CHCl₃ : CH₃OH = 30 : 1).

1-(3,4-methylenedioxyphenethyl)-4-(3-phenylpropyl)piperazine (7)

The general procedure was followed using amide 13 (288 mg, 0.79 mmol) to give 7 as a yellow oil (90 mg, 33%). ¹H NMR: (250 MHz, CDCl₃, δ) 1.75–1.87 (m, 2H, CH₂), 2.33–2.73 (m, 16H, CH₂), 5.85 (s, 2H, OCH₂O), 6.60–6.71 (m, 3H, CH), 7.15–7.28 (m, 5H, CH). ¹³C NMR: (75 MHz, CDCl_3, δ) 28.6, 33.3, 33.6, 53.2, 57.9, 60.7, 100.7, 108.1, 109.1, 121.3, 125.7, 128.2, 128.3, 134.0, 142.1, 145.7, 147.5; Anal. Calcd for the di-HCl salt (C₂₂H₃₀Cl₂N₂O₂): C, 62.12; H, 7.11; N, 6.59. Found: C, 61.85; H, 7.04; N, 6.55. m.p. 230 °C.

1-(3,4-ethylenedioxyphenethyl)-4-(3-phenylpropyl)piperazine (8)

The general procedure was followed using amide 14 (484 mg, 1.27 mmol) to give 8 as a yellow oil (340 mg, 73%). ¹H NMR: (250 MHz, CDCl₃, δ) 1.84 (app p, 2H, J = 7.5 Hz, CH₂), 2.36–2.74 (m, 16H, CH₂), 4.19 (s, 4H, OCH₂CH₂O), 6.65–6.80 (m, 3H, CH), 7.15–7.31 (m, 5H, CH). ¹³C NMR: (62.5 MHz, CDCl_3, δ) 28.6, 33.3, 33.6, 53.2, 57.9, 60.7, 100.7, 108.1, 109.1, 121.3, 125.7, 128.2, 128.3, 134.0, 142.1, 145.7, 147.5; Anal. Calcd for the di-HCl salt (C₂₃H₃₂Cl₂N₂O₂): C, 62.87; H, 7.34; N, 6.38. Found: C, 63.14; H, 7.49; N, 6.39. m.p. 230 °C.

1-(3,4-propylenedioxyphenethyl)-4-(3-phenylpropyl)piperazine (9)

The general procedure was followed using amide 15 (390 mg, 0.99 mmol) to give 9 as a yellow oil (321 mg, 85%). ¹H NMR: (300 MHz, CDCl₃, 1.84 (app p, 2H, J = 6.0 Hz, CH₂), 2.15 (app p, 2H, J = 6.0 Hz, CH₂), 2.37–2.75 (m, 16H, CH₂), 4.16 (t, 4H, J = 6.0 Hz, -CH₂O), 6.74–6.91 (m, 3H, CH), 7.15–7.30 (m, 5H, CH). ¹³C NMR: (75 MHz, CDCl_3, δ) 28.6, 32.0, 32.7, 33.7, 53.2, 58.0, 60.5, 70.4, 121.3, 121.6, 123.4, 125.7, 128.3, 128.4, 135.6, 142.1, 149.4, 151.0; Anal. Calcd for the di-HCl salt (C₂₄H₃₄Cl₂N₂O₂): C, 63.57; H, 7.56; N, 6.18. Found: C, 63.44; H, 7.71; N, 6.16. m.p. 212 °C.

4.2 Receptor Binding

Tritiated radioligands were obtained from PerkinElmer Life Sciences, Inc. (Boston, MA). [¹²⁵I]RTI-121 was synthesized as described previously.⁵⁹ Serial dilutions of test ligands were prepared in water containing 1% EtOH and 0.1% HOAc. Tritiated and [¹²⁵I]-labeled radioligands were prepared in Tris-HCl buffers (50 mM). Tritium radioactivity was measured using a Wallac 1409 liquid scintillation counter and OptiPhase^® HiSafe 2 cocktail at 44% efficiency. Fresh-frozen brains from male English Hartley guinea pigs were obtained from Rockland Immunochemicals, Inc. (Gilbertsville, PA) for σ receptor assays. Fresh male CD-1® mouse brains (Charles River Laboratories International, Inc., Wilmington, MA) were harvested, after animal euthanasia by cervical dislocation, as needed for the DAT and SERT assays. Studies involving living animals were performed humanely in compliance with the Guide for the Care and Use of Laboratory Animals promulgated by the U.S. National Institutes of Health, and with prior approvals from the Institutional Animal Care and Use Committees of the University of Missouri and the Harry S. Truman Memorial Veterans’ Hospital.

Assays for σ receptor binding were performed using 1.0 nM [³H](+)-pentazocine (σ₁), 3.0 nM [³H]ditolylguanidine ([³H]DTG) / 200 nM (+)-pentazocine (σ₂), and membranes from whole male guinea pig brains as previously described in detail.³⁸ Non-specific binding was defined by haloperidol (1.0 μM; σ₁) or by DTG (100 μM; σ₂). Binding assays for DAT and SERT were performed using literature procedures with minor modifications. In brief, DAT assays used [¹²⁵I]RTI at 15 pM and mouse striatal membranes, with non-specific binding defined by 0.5 μM GBR 12909.⁶⁰ SERT assays used [³H]paroxetine at 0.3 nM and mouse whole brain membranes, with non-specific binding defined by 10 μM fluoxetine.⁶¹ Experiments were conducted in duplicate, and replicated three to six times. Data were analyzed by non-linear curve-fitting using a sigmoidal four-parameter logistic fit (Prism 6.0c; GraphPad Software, Inc., La Jolla, CA). K_i values were derived from inhibition IC₅₀ data by the Cheng-Prusoff relationship,⁶² using reported K_d values of 2.3 nM for [³H](+)-pentazocine,³⁸ 23.9 nM for [³H]DTG,³⁸ 0.12 nM for [¹²⁵I]RTI,⁶⁰ and 0.13 nM for [³H]paroxetine.⁶¹