Synthesis of Aminoethyl‐Substituted Piperidine Derivatives as σ₁ Receptor Ligands with Antiproliferative Properties

Abstract

A series of novel σ₁ receptor ligands with a 4‐(2‐aminoethyl)piperidine scaffold was prepared and biologically evaluated. The underlying concept of our project was the improvement of the lipophilic ligand efficiency of previously synthesized potent σ₁ ligands. The key steps of the synthesis comprise the conjugate addition of phenylboronic acid at dihydropyridin‐4(1H)‐ones 7, homologation of the ketones 8 and introduction of diverse amino moieties and piperidine N‐substituents. 1‐Methylpiperidines showed particular high σ₁ receptor affinity and selectivity over the σ₂ subtype, whilst piperidines with a proton, a tosyl moiety or an ethyl moiety exhibited considerably lower σ₁ affinity. Molecular dynamics simulations with per‐residue binding free energy deconvolution demonstrated that different interactions of the basic piperidine‐N‐atom and its substituents (or the cyclohexane ring) with the lipophilic binding pocket consisting of Leu105, Thr181, Leu182, Ala185, Leu186, Thr202 and Tyr206 are responsible for the different σ₁ receptor affinities. Recorded logD_7.4 and calculated clogP values of 4a and 18a indicate low lipophilicity and thus high lipophilic ligand efficiency. Piperidine 4a inhibited the growth of human non‐small cell lung cancer cells A427 to a similar extent as the σ₁ antagonist haloperidol. 1‐Methylpiperidines 20a, 21a and 22a showed stronger antiproliferative effects on androgen negative human prostate cancer cells DU145 than the σ₁ ligands NE100 and S1RA.

Seven to eight reaction steps led to novel σ₁ receptor ligands with a 4‐(2‐aminoethyl)‐2‐phenylpiperidine scaffold. 1‐Methylpiperidine derivatives show high σ₁ affinity, considerable selectivity over the σ₂ subtype, and an improved Lipophilicity‐corrected Ligand Efficiency (LELP) index. Molecular dynamics simulations with per‐residue binding free energy deconvolution demonstrated that different interactions of the basic piperidine N atom and its substituents with the lipophilic binding pocket are responsible for the different σ₁ receptor affinities. 1‐Methylpiperidine derivatives show antiproliferative activity against the human prostate cancer cell line DU145.

Introduction

The σ₁ and σ₂ receptor subtypes differ in their molecular weight, tissue distribution and ligand binding profile.[ ¹ , ² , ³ , ⁴ ] Both σ₁ and σ₂ receptors are expressed in fast proliferating cells such as prostate cancer, breast carcinoma or leukemia cells. ^[5] Since this project was focused on σ₁ receptor ligands, only the σ₁ receptor subtype will be discussed further.

Twenty years after the first postulation of σ receptors by Martin et al., ^[1] the σ₁ receptor was cloned from various tissues including liver (guinea pig), brain (rat, mouse), kidney (rat) and chorioncarcinoma cells (human).[ ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ ] The membrane bound protein consists of 223 amino acids resulting in a molecular weight of 25.3 kDa. The σ₁ receptor proteins of different species have a high level of sequence identity (>93 % identity) yet they do not show any similarity to any other mammalian protein. Interestingly, a similarity of 65 % with the yeast sterol‐Δ⁸/Δ⁷‐isomerase has been detected although the σ₁ receptor is devoid of sterol isomerase activity. On the other hand, some sterol‐Δ⁸/Δ⁷‐isomerase inhibitors bind with high affinity at the σ₁ receptor.

In 2016, Kruse and coworkers produced the crystal structure of the σ₁ receptor, revealing a trimeric form of the receptor protein. ^[11] The N‐terminus consists of the unique transmembrane domain and a short extracellular peptide sequence. The C‐terminus is located on the cytosolic side of the membrane and forms a β‐barrel, which contains the ligand binding site. Intriguingly, the X‐ray‐determined protein structure differs considerably from that originally derived on the base on homology modeling techniques, nuclear magnetic resonance experiments and molecular biological investigations, all of which supported the existence of two transmembrane domains with both the C‐ and N‐terminal ends located on the cytosolic side.[ ¹² , ¹³ ] Two years later, the same group reported the structure of the σ₁ receptor in complex with the prototypical agonist (+)‐pentazocine and the prototypical antagonist haloperidol. ^[14]

The σ₁ receptor is not only expressed in the central nervous system (CNS), but also in some peripheral tissues including liver, heart and kidney. ^[4] On the intracellular level, the σ₁ receptor is predominantly found at the mitochondria‐associated membranes and at the endoplasmic reticulum (ER).[ ¹⁵ , ¹⁶ ] It plays a key role in the regulation of ion channels (K⁺, Na⁺, Cl⁻ channels), the release and reuptake of neurotransmitters and the intracellular signaling through modulation of Ca²⁺ levels. As a chaperone in the ER, the σ₁ receptor influences the activity of IP₃ receptors and the transfer of Ca²⁺‐ions from the ER to mitochondria.[ ¹⁷ , ¹⁸ , ¹⁹ ] Pharmacologically, the σ₁ receptor is involved in several neurological disorders including depression, alcohol and drug dependence, Parkinson's, Alzheimer‘s, Huntington's disease and neuropathic pain.[ ²⁰ , ²¹ , ²² , ²³ , ²⁴ ]

I In addition to its high concentration in the CNS, the expression level of σ₁ receptor in various human tumor cells is significantly increased compared to non‐tumor cells. This overexpression makes the σ₁ receptor an attractive target for the development of novel antitumor strategies. Specifically, the σ₁ receptor appears to be involved in programmed cell death (apoptosis). An increased σ₁ receptor expression was associated with a poor clinical outcome and high risk of metastasis. Antiproliferative effects were observed after treatment of human tumor cells with various σ₁ receptor antagonists. Moreover, the high density of σ₁ receptors in tumor tissue can be exploited for the development of novel diagnostic tools to image tumor cells, to evaluate the treatment with anticancer drugs and to increase the understanding of tumor physiology and pathophysiology.[ ⁵ , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ ] Several human tumor cells express an even higher amount of σ₂ receptors, which represents the rationale to develop σ₂ receptor‐based anticancer drugs and imaging tools.[ ³¹ , ³² , ³³ ]

In literature, a large number of structurally diverse ligands interacting with the σ₁ receptor is reported.^[24.28] Recently, the aminoethyl substituted 1,3‐dioxane 1 revealing low nanomolar σ₁ affinity (K _i=19 nM) and high antiallodynic activity in vivo (mouse capsaicin assay), which indicates σ₁ antagonistic activity, was reported.[ ³⁴ , ³⁵ ] (Figure 1) With respect to σ₁ affinity, the enantiomer (2S,4R)‐1 represents the eutomer (K _i=6.0 nM). ^[36] However, the acid‐labile acetalic substructure of 1 limits its further development. Therefore, ligands with a tetrahydropyran ring containing only one O‐atom ((2R,6S)‐2, K _i=5.4 nM, (2S,6R)‐2, K _i=1.6 nM) ^[37] and ligands with a cyclohexane ring without O‐atom ((1R,3S)‐3, K_i=0.61 nM, (1S,3R)‐3, K_i=1.3 nM, see Table 1), ^[38] which could not undergo further hydrolysis, were designed, synthesized and pharmacologically evaluated. Both tetrahydropyrans 2 and cyclohexanes 3 exhibit high σ₁ affinity, selectivity over the σ₂ subtype and, importantly, growth inhibition of the androgen negative human prostate cancer cell line DU145.[ ³⁷ , ³⁸ ]

Figure 1.

Reported σ₁ receptor ligands 1–3 with 1,3‐dioxane (1), tetrahydropyran (2) and cyclohexane (3) scaffold in comparison with the designed piperidine‐based ligands 4 reported in this manuscript. clogP values were calculated using ChemAxon.

However, the penalty for increased hydrolytic stability and σ₁ receptor affinity of tetrahydropyrans 2 and cyclohexanes 3 is an increased molecular lipophilicity. In Figure 1 the calculated clogP values of the σ₁ ligands 1–3 and one designed piperidine 4 a are summarized. In order to confirm the predicted clogP values, the logD_7.4 values of the σ₁ ligands 1–4 a were also determined experimentally using the micro shake flask method.[ ³⁹ , ⁴⁰ ]

In an effort to maintain high σ₁ affinity of the lead compounds 1–3 and, simultaneously, increase hydrolytic stability and reduce lipophilicity, the central 1,3‐dioxane, tetrahydropyran or cyclohexane ring of the lead compounds 1–3 was replaced by a piperidine ring (4, Figure 1). The calculated clogP value of −0.51 for the designed piperidine 4a is rather low, which is due to the additional secondary amine in the piperidine ring. Furthermore, the additional N‐atom in the piperidine ring entails the possibility to introduce further and diverse substituents at this position and this, in turn, allows for the modulation of σ₁ affinity, selectivity over the σ₂ subtype, lipophilicity, polarity and finally pharmacokinetic properties of the ligands 4.

Herein, we present the synthesis, receptor affinity and structure activity relationships of novel piperidine derivatives of type 4. Moreover, the effects on tumor cell growth will be reported.

Results and Discussion

Synthesis

Piperidines of type 4 were obtained by conjugate addition of a phenyl nucleophile at the α,β‐unsaturated ketones 7 and subsequent introduction of a C₂ chain by a Wittig reaction (Scheme 1). Transformation of the ester group into an amino moiety and removal of the N‐protective group represent the final steps of the synthesis (Schemes 2 and 3).

Scheme 1.

Synthesis of α,β‐unsaturated esters 9a,b: (a) pTsCl, K₂CO₃, CH₃CN, 18 h, rt, 6a, 95 %. (b) IBX, NMO, DMSO, 3 d, 30 °C, 7a, 77 %, 7b, 83 %, 7c, 77 %. (c) Phenylboronic acid, [Rh(cod)₂]BF₄, dioxane/H₂O, 8a, 34 %. (d) Phenylboronic acid, [Rh(cod)₂]BF₄, dioxane/KOH, 8b, 71 %. (e) Ph₃P=CHCO₂Et, toluene, 18 h, reflux; 9a, 103 % (contains small amounts of Ph₃PO); 9b, 98 %.

Scheme 2.

Synthesis of σ receptor ligands 4 from tosyl derivative 9a: (a) H₂ (balloon), Pd/C (10 %), CH₃OH, 20 h, rt, 81 %. (b) LiAlH₄, THF, 2.5 h, rt, 89 %. (c) CH₃SO₂Cl, Et₃N, CH₂Cl₂, 18 h, rt, 94 %. (d) Benzylamine or 3‐phenylpropan‐1‐amine, CH₃CN, 18 h, reflux, 60 % (13a), 87 % (13b). (e) Mg⁰ turnings, CH₃OH, ultrasonic irradiation, 5 h, rt, 80 % (4a), 37 % (4b).

Scheme 3.

Synthesis of σ receptor ligands from Cbz derivative 9b: (a) H₂ (3 bar), Pd/C (10 %), CH₃OH, 20 h, rt, 70 %. (b) formalin or CH₃CH=O, NaBH(OAc)₃, CH₂Cl₂, 18 h, rt, 66 % (15a), 68 % (15b) (c) LiAlH₄, THF, 2 h, rt, 85 % (16a), 85 % (16b). (d) DMP, CH₂Cl₂, 2 h, rt, 62 % (17a), 93 % (17b). (e) R₂NH, NaBH(OAc)₃, CH₂Cl₂, 3 h, rt, 11–69 %.

Whereas Cbz‐ and Boc‐protected piperidin‐4‐ones 6b and 6c were commercially available, the tosyl‐protected piperidin‐4‐one 6a was prepared by tosylation of piperidin‐4‐one (5). Oxidation of piperidin‐4‐ones 6a‐c with iodoxybenzoic acid (IBX) ^[41] provided the α,β‐unsaturated ketones (vinylogous amides) 7a–c in 77–83 % yield. Addition of N‐methylmorpholin‐N‐oxide (NMO) allowed conducting the oxidation under very mild reaction conditions (30 °C), which gave high yields (Scheme 1).

The conjugate addition of a phenyl nucleophile at the α,β‐unsaturated ketones 7 a–c served as the key step in the synthesis of the designed ligands. The Rh‐catalyzed ([Rh(cod)₂]BF₄) conjugate addition of phenylboronic acid ^[42] at the tosyl‐protected dihydropyridin‐4‐one 7a in a dioxane/KOH mixture led to decomposition of 7a. To prevent decomposition of 7a the reaction was performed in a dioxane/water mixture without addition of a base. After optimization of the reaction conditions, the addition product 8a was isolated in 34 % yield. The Cbz‐protected dihydropyridine 7b turned out to be more stable and tolerated well the conjugate addition in dioxane/KOH resulting in 71 % yield of the addition product 8b. Despite thorough modification of the reaction conditions, the Rh‐catalyzed conjugated addition of phenylboronic acid at the Boc‐protected dihydropyridine 7c did not lead to the addition product 8c (Scheme 1).

Next, the ketones 8a and 8b were expanded by a two‐carbon chain. To this purpose, the ketones 8a and 8b were reacted with the stabilized P‐ylide Ph₃P=CHCO₂Et to form α,β‐unsaturated esters 9a and 9b. The α,β‐unsaturated esters 9a and 9b were obtained as mixtures of (E)‐ and (Z)‐diastereomers. The ratio was 60 : 40 for the tosyl derivatives (Z)‐9a : (E)‐9a and 55 : 45 for the Cbz derivatives (Z)‐9b : ( E)‐9b (Scheme 1).

The α,β‐unsaturated ester 9a with a tosyl protective group was hydrogenated using the catalyst Pd/C. The saturated ester was isolated as a mixture of cis‐10 : trans‐10 (35 : 65). LiAlH₄ reduction of the ester 10 led to the primary alcohol 11 (cis‐11 : trans‐11=83 : 17), which was reacted with methanesulfonyl chloride to afford the mesylate 12 (cis‐12 : trans‐12=83 : 17). Nucleophilic substitution of the mesylate 12 with benzylamine or 3‐phenylpropan‐1‐amine led to the secondary amines 13a and 13b in 60 % and 87 % yield, respectively. Both amines were isolated as 65 : 35‐mixture of cis‐ and trans‐configured diastereomers. Finally, the tosyl moiety of 1 a and 1 b was removed with Mg⁰ in methanol to provide the diamines 4a and 4b. The benzylamine 4a was isolated in 80 % yield (cis‐4a : trans‐4a=75 : 25) and the phenylpropylamine 4b in 37 % yield (cis‐4b : trans‐4b=65 : 35) (Scheme 2).

The structure of the signal for the axially oriented proton in 3‐position of the main diastereomer unequivocally proves its cis‐configuration. As an example, a dt (J=13.5/10.3 Hz) at 1.54 ppm and a broad q (J=12.1 Hz) at 1.20 ppm are observed for 3‐H_ax of 13a and 4b, respectively. The large coupling constants originate from germinal coupling with 3‐H_eq and vicinal couplings with two axially oriented protons in 2‐ and 4‐pposition indicating the equatorial orientation of both substituents at 2‐ and 4‐position at the piperidine ring. Since the signal structures for 2‐H_ax (dd, J=9.7–11.3 Hz and 2.4‐2.9 Hz) of both diastereomers of 4a and 4b are identical, the phenyl ring of both diastereomers adopts the equatorial orientation. Thus, cis‐ and trans‐configured diastereomers differ in the orientation of the aminoethyl moiety at the 4‐position.

During hydrogenation of the α,β‐unsaturated ester 9b, hydrogenolytic cleavage of the Cbz moiety at the piperidine ring occurred as well. The saturated ester 14 was isolated in 70 % yield as mixture of cis‐ and trans‐diastereomers (ratio 75 : 25). The secondary amine 14 was reductively alkylated with formalin or acetaldehyde using NaBH(OAc) ^[43] as reducing agent to afford the methyl and ethyl derivatives 15a and 15b, respectively. LiAlH₄ reduction of the esters 15a and 15b provided the primary alcohols 16a and 16b. Activation of the primary alcohol 16a with methanesulfonyl chloride as shown for the alcohol 11 led to a mesylate, which reacted directly with the tertiary amino moiety of the piperidine ring to form a 1‐azoniabicyclo[2.2.2]octane derivative. Therefore, the alcohols 16a and 16b were oxidized with Dess‐Martin‐Periodinane (DMP) ^[44] to give the aldehydes 17a and 17b, which were reductively aminated with various primary and secondary amines and NaBH(OAc)₃ ^[43] to provide the secondary and tertiary amines 18–22 (Scheme 3). The final amines 18–22 were isolated as mixtures of diastereomers (cis : trans=60 : 40 to 85 : 15). The quartet‐like structure or the dt structure (J >11 Hz, respectively) of the signal for the axially oriented proton in 3‐postion confirms the cis‐configuration of the main diastereomer.

Since for tetrahydropyrans and cyclohexanes the σ₁ affinities of cis‐ and trans‐configured diastereomers were very similar[ ³⁷ , ³⁸ ] and, moreover, the separation of cis‐ and trans‐configured piperidines turned out to be very difficult, mixtures of diastereomers 4, 13 and 18–22 were tested, respectively.

σ₁ and σ₂ receptor affinity

The affinity of the synthesized piperidines towards σ₁ and σ₂ receptors was determined in radioligand receptor binding assays. In the σ₁ assay, homogenates of guinea pig brains were used as receptor material and [³H]‐(+)‐pentazocine as σ₁ selective radioligand. The receptor material in the σ₂ assay was a membrane preparation from rat liver. As a σ₂ selective radioligand is not available, the assay was performed with the non‐selective radioligand [³H]‐1,3‐di(o‐tolyl)guanidine ([³H]DTG). In order to occupy σ₁ receptors and render the assay selective for the σ₂ subtype, an excess of non‐tritiated (+)‐pentazocine was added.[ ⁴⁵ , ⁴⁶ , ⁴⁷ ] Affinity data obtained with receptor preparations containing guinea pig and human σ₁ receptors are well comparable, since the amino acid sequences of guinea pig and human σ₁ receptors are 93 % identical. ^[48] Furthermore, binding studies with rat and human σ₂ receptors result in comparable affinity data for potent and selective σ₂ ligands.[ ⁴⁹ , ⁵⁰ ] In Table 1, the σ affinity of the synthesized compounds is compared with the σ affinity of some lead and reference compounds.

Table 1.

σ₁ and σ₂ receptor affinity of synthesized piperidines and some lead and reference compounds.

Compd	R1	NR2	K i±SEM [nM] (n=3)		σ1 : σ2
			σ1	σ2	selectivity
136	1,3‐dioxane[a]	NHCH2Ph	6.0±1.0	4200	14
237	tetrahydropyran[a]	NHCH2Ph	1.6±0.2	378	236
338	cyclohexane[a]	NHCH2Ph	0.61±0.1	49±31	80
4a	H	NHCH2Ph	165	372	2.3
4b	H	NH(CH2)3Ph	849	0 %[b]	–
13a	Ts	NHCH2Ph	57±21	763	13
13b	Ts	NH(CH2)3Ph	291±139	567	2
18a	CH3	NHCH2Ph	7.9±0.2	483	61
18b	CH2CH3	NHCH2Ph	129±38	131	1
19a	CH3	NH(CH2)3Ph	50±17	0 %[b]	>20
19b	CH2CH3	NH(CH2)3Ph	2400	334	0,14
20a	CH3	NHCH2C6H11	16±5	285	18
21a	CH3	N(CH3)CH2Ph	19±9	77±5	4
22a	CH3		27±11	1600	59
(+)‐pentazocine		–	5.4±0.5	–	–
haloperidol		–	6.6±0.9	125±33	19
di‐o‐tolylguanidine		–	71±7.9	54±8	0.76

[a] Structures of compounds 1–3 are shown in Figure 1. [b] For compounds with low affinity the inhibition (in %) of radioligand binding at a test compound concentration of 1 μM is given. The piperidines were tested as mixtures of cis‐ and trans‐configured diastereomers. cis : trans=60 : 40 to 85 : 15.

Replacement of the central cyclohexane ring of the lead compound 3 (K _i(σ₁)=0.61 nM) by a piperidine ring without N‐substituent led to remarkably reduced σ₁ affinity of the secondary amine 4 a (K _i(σ₁)=165 nM). Introduction of an ethyl (18b) or tosyl moiety (13a) increased the σ₁ affinity slightly, but a small methyl moiety at the piperidine N‐atom resulted in rather high σ₁ affinity. The σ₁ affinity of the piperidine 18a (K _i(σ₁)=7.9 nM) is only 10‐fold lower than the σ₁ affinity of the cyclohexane derivative 3 and equipotent with the 1,3‐dioxane derivative 1. It has to be noted that 18a was tested as mixture of diastereomers cis‐18a : trans‐18a=85 : 15.

Extension of the distance between the basic N‐atom and the terminal phenyl moiety from one methylene moiety (benzylamines) to three methylene moieties (3‐phenylpropylamines) led to reduced σ₁ affinity of 4b, 13b, 19a and 19b. As observed for the benzylamine 18a, the piperidine derivative 19a with the small N‐methyl moiety showed the highest σ₁ affinity (K _i(σ₁)=50 nM) of the series of 3‐phenylpropylamines. Therefore, further variations at the terminal N‐atom were performed starting with the piperidine ring bearing the small methyl moiety. Although the cyclohexylmethylamine 20a and the tertiary amines 21a and 22a revealed slightly reduced σ₁ affinity compared to the benzylamine 18a, their K _i values are still in the low nanomolar range (K _i(σ₁) <27 nM).

The most potent ligands bearing a methyl moiety at the piperidine N‐atom reveal high selectivity for the σ₁ over the σ₂ receptor. In particular, the benzylamine 4a, the cyclohexylmethylamine 20a and the phenylpiperazine 22a exhibit 60‐, 18‐ and 60‐fold σ₁:σ₂ selectivity, respectively. The lowest σ₁:σ₂ selectivity (4‐fold) was found for the N‐benzyl‐N‐methylamine 21a.

In contrast, piperidine derivatives 18b and 19b bearing an ethyl moiety at the piperidine N‐atom display higher σ₂ affinity. Whereas the benzylamine 18b has the same affinity towards both σ₁ and σ₂ receptors, the homologous phenylpropylamine 19b reveals a 7‐fold preference for the σ₂ receptor over the σ₁ receptor.

Molecular dynamics simulation

Piperidine 4a and the methylated derivatives 20a, 21a, and 22a are provided with high σ₁ affinity. Accordingly, we carried out Molecular Dynamics (MD) simulations to investigate the interactions of these compounds with the σ₁ receptor. Initially, the putative binding modes were identified using a well‐validated docking protocol.[ ³⁷ , ³⁸ ] Next, MD simulations of the resulting σ₁ receptor/piperidine derivative complexes were carried out, and the corresponding ligand/protein free energy of binding (ΔG_bind) values were obtained via the Molecular Mechanics/Poisson‐Boltzmann Surface Area (MM/PBSA) approach. ^[51] According to the simulations, and in agreement with the corresponding experimental profiles, binding of the piperidine derivatives at σ₁ receptor provided a lower Gibbs free energy of binding ΔG_bind than binding of the previously reported cyclohexane derivatives. ^[37] In terms of enthalpic and entropic contributions, the piperidines exhibit a similar thermodynamics trend as cyclohexane 3 (Figure 2A), but their corresponding ΔG_bind values are more than 1 kcal/mol higher (Figure 2A, Table S1, ΔG_bind(3)=−11.31 kcal/mol; ΔG_bind(4a)=−9.48 kcal/mol; ΔG_bind(20a)=−10.12 kcal/mol; ΔG_bind(21a)=−10.06 kcal/mol; ΔG_bind(22a)=−9.97 kcal/mol).

Figure 2.

(A) Calculated free energy of binding (ΔG_bind), and enthalpic (ΔH_bind) and entropic (‐TΔS_bind) components for the σ₁ receptor complexed with 3, 4a, 20a, 21a and 22a (B) Details of compound 4a in the binding pocket of the σ₁ receptor. 4a is shown as atom‐colored sticks‐and‐balls (C, grey, N, blue, O, red) while the side chains of the protein residues mainly interacting with 4a are depicted as colored sticks and labelled. Hydrogen atoms, water molecules, ions, and counterions are omitted for clarity. (C) Per‐residue binding free energy decomposition of the main involved amino acids of the complex between σ₁ receptor and 3, 4a, 20a, 21a and 22a.

To explain the lower σ₁ binding capability of the new piperidine derivatives, the individual intermolecular interactions were analyzed by performing a per‐residue binding free energy deconvolution (PRBFED) of the enthalpic terms ΔH_res (Figures 2B, 2C, Table S2). As expected, the 4a/σ₁ receptor complex revealed the prototypical pattern of intermolecular interactions underlying σ₁ receptor ligand binding (Figure 2B). Specifically, the N‐atom of the basic benzylamino moiety of 4a is engaged in two interactions in the σ₁ binding site: i) a persistent salt bridge with the carboxylate moiety of Glu172, stabilized by an internal hydrogen bond with Tyr103 (ΣΔH_res=−4.93 kcal/mol, Figure 2C and Table S2); and ii) a π‐cation interaction with the phenyl ring of Phe107 (ΔH_res=−1.23 kcal/mol). Moreover, the side chain of Ile124 can support the appropriate orientation of the benzylamino moiety of 4a in the receptor binding cavity with favorable hydrophobic interactions (ΔH_res=−1.27 kcal/mol). On the other hand, the highly hydrophobic σ₁ receptor binding site should assist nestling of the phenylpiperidine moiety of 4a, but the presence of a further protonated amino moiety in this apolar region interferes with the lipophilic interactions with σ₁ receptor residues (Figure 2B). Accordingly, a considerable decrease of the corresponding enthalpic contribution is detected by our PRBFED analysis compared to the cyclohexane derivative 3 (4a: ΣΔH_{L105,T181,A185}=−2.23 kcal/mol, ΣΔH_{L182,L186,T202,Y206}=−2.68 kcal/mol; 3: ΣΔH_{L105,T181,A185}=−3.03 kcal/mol, ΣΔH_{L182,L186,T202,Y206}=−3.57 kcal/mol; Figure 2C and Table S2).

The N‐methylpiperidine derivatives 20a, 21a, and 22a show very similar binding modes as 4a and their interactions with σ₁ receptor residues Tyr103, Phe107, Ile124 and Glu172 are practically unchanged (Figures 2C and S1). The presence of the small CH₃ group on the N‐atom of the piperidine ring increases the lipophilic interactions with the σ₁ receptor binding pocket compared with the secondary amine 4a, but does not achieve the same value as the cyclohexane derivative 3. Accordingly, the favorable enthalpic contribution provided by the interactions with the hydrophobic cavity of the σ₁ receptor is significantly lower than the contribution of the analogous cyclohexane derivative 3 (20a: (ΣΔH_{L105,T181,A185}=−2.34 kcal/mol, ΣΔH_{L182,L186,T202,Y206}=−3.09 kcal/mol; 21a: ΣΔH_{L105,T181,A185}=−2.38 kcal/mol, ΣΔH_{L182,L186,T202,Y206}=−3.16 kcal/mol; 22a: ΣΔH_{L105,T181,A185}=−2.42 kcal/mol, ΣΔH_{L182,L186,T202,Y206}=−2.95 kcal/mol; Figure 2C and Table S2).

Lipophilicity and lipophilic ligand efficiency

In order to argue with reliable lipophilicity values, the logD_7.4 values of key compounds were determined experimentally following our micro‐shake‐flask protocol.[ ³⁹ , ⁴⁰ ] According to this method, each compound of interest was distributed between an n‐octanol layer and an aqueous MOPS buffer pH 7.4. Subsequently, the amount of compound in the buffer layer was determined by mass spectrometry.

In Table 2, the experimentally determined logD_7.4 values for lead compounds 1–3 and piperidines 4a and 18a are summarized. As expected, the most lipophilic compound is the cyclohexane derivative 3 with a logD_7.4 value of 3.25. Introduction of one O‐atom into the cyclohexane ring (tetrahydropyran 2) reduces the logD_7.4 value by one order of magnitude. A second O‐atom as in 1,3‐dioxane 1 further reduces the lipophilicity by one order of magnitude. However, introduction of an NCH₃ (18a) or NH (4a) moiety into the cyclohexane ring instead of one O‐atom resulted in very low logD_7.4 values of 0.52 and −0.79.

Table 2.

Experimentally determined logD_7.4 values, calculated clogP values and lipophilic ligand efficiency indices, LLE and LELP.

no.	Compd	σ1 affinity K i [nM]	logD7.4 (exp., n=3)	clogP[a] (calcd.)	LLE[b]	LELP[c]
1		6.0	1.36±0.02	1.30	6.92	3.48
2		1.6	2.52±0.05	2.05	6.75	5.13
3		0.61	3.25±0.02	3.10	6.11	7.40
4a		165	−0.79±0.07	−0.51	7.29	1.65[d]
18a		7.9	0.52±0.01	0.01	8.09	0.03

[a] clogP values were calculated with ChemAxon. [b] Lipophilic Ligand Efficiency (LLE) index is defined as: LLE=pK _i−clogP. [c] Lipophilicity‐corrected Ligand Efficiency (LELP) index is defined as: LELP=clogP : LE; LE=pK _i : HAC (HAC: number of non‐H‐atoms of a drug). [d] For the calculation, the negative sign of the clogP value was ignored.

In addition, the corresponding clogP values for the same set of compounds were calculated by ChemAxon. As shown in Table 2, the calculated clogP values correlate well with the experimentally recorded logD_7.4 values indicating that ChemAxon is a method leading to reliable predicted clogP values for this type of compounds.

Improving the potency of compounds is commonly achieved by increasing the molecular complexity in order to find the adequate interactions of the molecule with its target protein. However, addition of unnecessary molecular complexity often leads to “molecular obesity”. ^[52] Obese molecules, i. e., rather complex molecules with high lipophilicity, often suffer from unfavorable pharmacokinetics (poor bioavailability) and non‐acceptable toxicological profile. Lipinski's “rule of five” is one of the earliest attempts to overcome the risk of obese drugs.[ ⁵² , ⁵³ ] In order to quickly analyze the impact of molecular complexity and lipophilicity for the quality of drugs at an early stage during the drug discovery process, several ligand efficiency indices have been defined and validated.[ ⁵⁴ , ⁵⁵ , ⁵⁶ ] The Lipophilic Ligand Efficiency (LLE) index describes the contribution of the lipophilicity of a drug in form of the clogP value to its biological activity in form of K _i, K _d, or IC ₅₀ value (LLE=pK _i or pK _d or pIC ₅₀–clogP). ^[57] Since the LLE index is not useful for very small and polar drugs, the Lipophilicity‐corrected Ligand Efficiency (LELP) index was defined taking the number of non‐H atoms (HAC) of a drug into account in addition to its clogP value (LELP=(clogP ⋅ HAC) : pK _i). ^[58] The LELP index describes the reduction of the drug efficiency of even very potent drugs by increasing their lipophilicity and size. ^[59] Promising physicochemical properties are usually expected for drugs with a LLE index >5 and a LELP index <10.

With respect to efficiency, the benzylamines of all four compound classes fulfill the criteria of LLE>5 and LELP<10 (Table 2). However, the novel piperidines 4a and 18a show considerably higher LLE values than the corresponding 1,3‐dioxane 1, tetrahydropyran 2 and cyclohexane 3. Analogously, the LELP values of the piperidines 4a and 18a are very low, thereby rendering 4a and 18a particularly efficient drugs. The low σ₁ affinity of 4a (K _i=165 nM) is compensated by its high polarity (low lipophilicity, logD_7.4=−0.79).

Growth inhibition of human tumor cell lines

In a preliminary experiment, the human non‐small cell lung cancer cell line A427 ^[60] was incubated with the low affinity σ₁ ligand 4a and the proliferation of the tumor cells was observed using the Live Cell Imager IncuCyte® allowing the continuous observation of the morphology, behavior and growth of the tumor cells. In this assay 4a (IC ₅₀=17 μM) showed comparable growth inhibition as the prototypical σ₁ antagonist haloperidol (IC ₅₀=16 μM, see Table S3 in Supporting Information). The effects of both compounds on A427 cells were considerably reduced in the presence of the prototypical σ₁ agonist (+)‐pentazocine (10 μM) indicating a contribution of σ₁ receptors to this effect (Table S3 in Supporting Information). Moreover, 4a behaved as σ₁ receptor antagonist in this A427 tumor cell proliferation assay.

Stimulated by the promising antiproliferative effect of 4a on human non‐small cell lung cancer cells A427, the growth inhibition of the androgen negative human prostate cancer cells DU145 ^[61] was investigated. For this purpose, the methylated piperidines 20a, 21a and 22a were selected, due to their promising σ₁ affinity. In the assay, DU145 tumor cells were incubated in 96‐well plates for 24 h. Different concentrations of the test compounds were added and after incubation for additional 72 h, the amount of living cells was recorded by staining with Sulforhodamine B. ^[62] In Table 3 the activity of the prototypical σ₁ antagonists NE‐100 and S1RA is included. Figure S1 in the Supporting Information displays the corresponding graphics.

Table 3.

Growth inhibition of androgen negative human prostate cancer cells DU145 by potent σ₁ ligands.

Compd	NR2	σ1 affinity K i±SEM [nM]	cytotoxicity (DU145) IC 50 [μM]
20a	NHCH2C6H11	16±5	4.9
21a	N(CH3)CH2Ph	19±9	5.5
22a		27±11	4.0
NE‐100		1.3 [63]	>10
S1RA		17±7.0 [64]	>10

σ₁ affinity and cytotoxicity represent the data of three experiments (n=3).

The methylated piperidines 20a, 21a, and 22a inhibit the growth of DU145 tumor cells with IC₅₀ values in the low micromolar range. Both the σ₁ affinity and the antitumor activity of the three compounds are very similar. In this assay, the piperidines 20a, 21a, and 22a are more potent than the reference σ₁ antagonists NE‐100 and S1RA.

Conclusion

Saturated six‐membered rings bearing an aminoethyl side chain show high σ₁ receptor affinity and high selectivity over the σ₂ subtype. However, 1,3‐dioxane 1 containing an acetal is not stable under acidic conditions (e. g., stomach) and the cyclohexane derivative 3 is rather lipophilic. Therefore, piperidines of type 4 were designed, which are hydrolytically stable and rather polar.

Piperidines 4 and 18–22 were prepared in a nine‐step synthesis. Piperidines with a methyl moiety at the piperidine N‐atom show high σ₁ receptor affinity and σ₁:σ₂ selectivity indicating that it is possible to replace bioisosterically the 1,3‐dioxane ring of 1 or the cyclohexane ring of 3 by the piperidine ring with only slightly reduced σ₁ affinity.

In addition to the high σ₁ affinity, the piperidines 4a and 18a are polar compounds with very low experimentally determined logD_7.4 values. As a result, the lipophilic ligand efficiency (LLE) index of the piperidines is considerably higher than the LLE of the lead compounds 1–3 even for 4a exhibiting only low σ₁ affinity (K _i=165 nM). In case of 4a, the low σ₁ affinity is compensated by the low lipophilicity.

Molecular dynamics simulations and analysis of the per‐residue binding free energy revealed that the very polar protonated piperidine ring of 4a reduces crucial lipophilic interactions within the lipophilic binding pocket of the σ₁ receptor. Introduction of a NCH₃ moiety (compounds 20a, 21a, 22a) compensates partially these unfavorable interactions. However, the σ₁ receptor affinity of the very lipophilic cyclohexane derivative 3 could not be achieved.

Due to their promising physicochemical properties, the inhibition of tumor cell growth by selected piperidines was investigated. The piperidine 4a reduced the proliferation of non‐small cell lung cancer A427 cells similar to the σ₁ antagonist haloperidol and the σ₁ agonist (+)‐pentazocine abolished its effect. The methylated piperidines 20a, 21a and 22a inhibited the growth of the androgen negative human prostate cancer cell line DU145. The piperidines are more active than the prototypical σ₁ antagonists NE100 and S1RA, which underlines the favorable physicochemical properties of the piperidine‐based σ₁ ligands.

Experimental Section

Chemistry, general

Unless otherwise noted, moisture sensitive reactions were conducted under dry nitrogen. CH₂Cl₂ was distilled over CaH₂. THF was distilled over sodium/benzophenone. Et₂O and toluene were dried over molecular sieves 4. Thin layer chromatography (tlc): Silica gel 60 F254 plates (Merck). Flash chromatography (fc): Silica gel 60, 40–64 μm (Merck); parentheses include: diameter of the column (d), length of the stationary phase (l), fraction size (V), eluent. Melting point: Melting point apparatus Mettler Toledo MP50 Melting Point System, uncorrected. MS: microTOF−Q II (Bruker Daltonics); APCI, atmospheric pressure chemical ionization. IR: FT‐IR spectrophotometer MIRacle 10 (Shimadzu) equipped with ATR technique. Nuclear magnetic resonance (NMR) spectra were recorded on Agilent 600‐MR (600 MHz for ¹H, 151 MHz for ¹³C) or Agilent 400‐MR spectrometer (400 MHz for ¹H, 101 MHz for ¹³C); δ in ppm related to tetramethylsilane and measured referring to CHCl₃ (δ=7.26 ppm (¹H NMR) and δ=77.2 ppm (¹³C NMR)), CHD₂OD (δ=3.31 ppm (¹H NMR) and δ=49.0 ppm (¹³C NMR)) and DMSO‐d₆ (δ=2.54 ppm (¹H NMR) and δ=39.5 ppm (¹³C NMR)); coupling constants are given with 0.5 Hz resolution; the assignments of ¹³C and ¹H NMR signals were supported by 2‐D NMR techniques where necessary.

HPLC equipment and methods

HPLC method to determine the purity of compounds: Pump: L‐7100, degasser: L‐7614, autosampler: L‐7200, UV detector: L‐7400, interface: D‐7000, data transfer: D‐line, data acquisition: HSM‐Software (all from LaChrom, Merck Hitachi); Equipment 2: Pump: LPG‐3400SD, degasser: DG‐1210, autosampler: ACC‐3000T, UV‐detector: VWD‐3400RS, interface: DIONEX UltiMate 3000, data acquisition: Chromeleon 7 (Thermo Fisher Scientific); column: LiChropher® 60 RP‐select B (5 μm), LiChroCART® 250–4 mm cartridge; flow rate: 1.0 mL/min; injection volume: 5.0 μL; detection at λ=210 nm; solvents: A: demineralized water with 0.05 % (V/V) trifluoroacetic acid, B: acetonitrile with 0.05 % (V/V) trifluoroacetic acid; gradient elution (% A): 0–4 min: 90 %; 4–29 min: gradient from 90 % to 0 %; 29–31 min: 0 %; 31–31.5 min: gradient from 0 % to 90 %; 31.5–40 min: 90 %. Unless otherwise noted, the purity of all test compounds is greater than 95 %.

Synthetic procedures

The compounds 6a and 7a have been reported in ref.. ^[65] The procedures have been modified and are described in the Supporting Information.

Benzyl 4‐oxo‐3,4‐dihydropyridine‐1(2H)‐carboxylate (7b)

Iodoxybenzoic acid (IBX with 20 % benzoic acid as stabilizer, 2.77 g, 11.8 mmol, 1.3 eq) and 4‐methylmorpholin‐4‐oxide (NMO), (4.32 g, 36 mmol, 3.4 eq) were dissolved in DMSO (15 mL) and the piperidone 6b (2.73 g, 10.8 mmol) dissolved in DMSO (20 mL) was added to the solution. The mixture was stirred for 72 h at 30 °C in the dark. The reaction mixture was poured into a saturated solution of NaHCO₃ (50 mL), the mixture was extracted with Et₂O (3×50 mL) and the combined Et₂O layers were washed with NaHCO₃, brine and water. The organic layer was dried (Na₂SO₄) and concentrated in vacuo. The crude product was purified by automated fc (Snap, 340 g, V=1600 mL, CH₂Cl₂ : ethyl acetate=9 : 1, Rf=0.44). Colorless solid, mp 65 °C, yield 2.26 g (83 %) C₁₃H₁₃NO₃ (231.3 g/mol). HR‐MS (APCI): m/z=232.0975 (calcd. 232.0968 for C₁₃H₁₄NO₃ [M+H]⁺). ¹H NMR (600 MHz, CDCl₃): δ (ppm)=2.56 (t, J=7.3 Hz, 2H, 3‐H), 4.04 (t, J=7.4 Hz, 2H, 2‐H), 5.26 (bs, 2H, CH₂‐bnz), 5.34 (bs, 1H, 5‐H), 7.33–7.44 (m, 5H, H_arom.), 7.85 (bs, 1H, 6‐H). ¹³C NMR (151 MHz, CDCl₃): δ (ppm)=35.8 (C‐3), 42.7 (C‐2), 69.2 (CH₂‐bnz), 107.9 (C‐5), 127.1, 128.6, 128.7, 128.9, 129.0 (5 C, C_arom.), 135.0 (C‐1_arom.), 141.1 (NCOO‐benz), 143.3 (C‐6), 193.5 (C‐4). Purity (HPLC): 87.2 %, t _R=17.6 min.

tert‐Butyl 4‐oxo‐3,4‐dihydropyridine‐1(2H)‐carboxylate (7c)

Iodoxybenzoic acid (IBX with 20 % benzoic acid as stabilizer, 4.3 g, 13 mmol, 1.3 eq) and 4‐methylmorpholin‐4‐oxide (NMO), (3.5 g, 30 mmol, 3.0 eq) were dissolved in DMSO (15 mL) and the piperidone 6c (1.99 g, 10.1 mmol) dissolved in DMSO (20 mL) was added to the solution. The mixture was stirred for 70 h at 30 °C in the dark. The reaction mixture was poured into a saturated solution of NaHCO₃ (50 mL), the mixture was extracted with Et₂O (3×50 mL) and the combined Et₂O layers were washed with NaHCO₃, brine and water. The organic layer was dried (Na₂SO₄) and concentrated in vacuo. The crude product was purified by automated fc (Snap, 340 g, V=540 mL, CH₂Cl₂ : ethyl acetate=9 : 1, R_f=0.44). Colorless solid, mp 53 °C, yield 1.51 g (77 %) C₁₀H₁₅NO₃ (197.2 g/mol). HR‐MS (APCI): m/z=198.1125 (calcd. 198.1161 for C₁₀H₁₅O₃ [M+H]⁺). ¹H NMR (400 MHz, DMSO‐d₆): δ (ppm)=7.83 (d, J=8.2 Hz, 1H, 6‐H), 5.18 (d, J=8.2 Hz, 1H, 5‐H), 3.92–3.84 (m, 2H, 2‐CH₂), 2.48–2.41 (m, 2H, 3‐CH₂), 1.48 (s, 9H 3x CH₃). ¹³C NMR (151 MHz, DMSO‐d₆): δ (ppm)=28.2 (3 C, CH₃), 35.8 (C‐3), 42.4 (C‐2), 83.7 (C‐(CH₃)₃) 106.4 (C‐5), 144.2 (C‐6), 154.6 (COOR), 193.8 (C‐4). Purity (HPLC): 86.7 %, t _R=16.5 min.

2‐Phenyl‐1‐tosylpiperidin‐4‐one (8a)

Phenylboronicacid (1.46 g, 11.9 mmol, 3.0 eq) and [Rh(cod)₂]BF₄ (64.0 mg, 0.16 mmol, 0.04eq) were dissolved in degassed H₂O/dioxane (1 : 11, 12 mL) and the mixture was stirred for 30min. Enone 7a (1.02g, 4.06mmol, 1.0 eq) dissolved in H₂O/dioxane (1 : 11, 8 mL) was added dropwise and the mixture was heated to 85 °C for 5 h. The mixture was filtered through a short pad of silica gel with Et₂O washing, the filtrate was dried (Na₂SO₄), concentrated in vacuo and the crude product was purified by automated fc (Snap, 100 g, V=200 mL, diethyl ether/cyclohexane=4 : 1, R_f=0.55). Yellow resin, yield 460mg (34 %). C₁₈H₁₉NO₃S (329.4 g/mol). HR‐MS (APCI): m/z=330.1184 (calcd. 330.1158 for C₁₈H₂₀NO₃S [M+H]+). ¹H NMR (400 MHz, CDCl3): δ [ppm]=2.25 (ddt, J=15.5/3.8/2.1 Hz, 1H, 5‐H_eq), 2.39–2.45(m, 1H, 5‐H_ax), 2.45 (s, 3H, CH₃), 2.72 (dd, J=15.3/7.0 Hz, 1H, 3‐H_ax), 2.93 (dt, J=15.3/1.9 Hz, 1H, 3‐H_eq), 3.14 (ddd, J=14.5/12.1/3.6 Hz, 1H, 6‐H_ax), 4.01 (ddt, J=14.5/6.9/2.1 Hz, 1H, 6‐H_eq), 5.63 (d, J=7.1 Hz, 1H, 2‐H), 7.21–7.37 (m, 7H, 3‐H_Tos, 5‐H_Tos, 5× H_phenyl), 7.79–7.86 (m, 2H, 2‐H_Tos, 6‐H_Tos).¹³C NMR (151 MHz, CdCl₃): δ [ppm]=δ 21.7 (CH₃), 40.4 (C‐5), 40.4 (C‐6), 43.5 (C‐3), 56.6 (C‐2), 127.3 (2 C, C‐2_Tos, C‐6_Tos), 127.5 (2 C, C‐2_phenyl, C‐6_phenyl), 128.2 (C‐4_phenyl), 128.9 (2 C, C‐3_phenyl, C‐5_phenyl), 130.2 (2 C, C‐3_Tos, C‐5_Tos), 137.6 (C‐4_Tos), 138.5 (C‐1_Tos), 144.2 (C‐1_phenyl), 206.4 (C‐4). Purity (HPLC): 96.6 %, 22.1 min. FT‐IR (neat): ν [cm⁻¹]=2971 (C‐Harom.), 1715 (C=O), 1152 (SO2 N).

Benzyl 4‐oxo‐2‐phenylpiperidine‐1‐carboxylate (8b)

Phenylboronicacid (688 mg, 5.6mmol, 1.3eq) and [Rh(cod)₂]BF₄ (36.8 mg, 0.09 mmol, 0.02eq) were dissolved in a mixture of degassed KOH (1.5 M, 2 mL) and dioxane (6 mL) and the mixture was stirred for 30min at rt. Enone 7b (1.0g, 4.3mmol, 1.0eq) dissolved as well in a mixture of KOH (2 mL) and dioxane (6 mL) was added dropwise to the first mixture and heated to 90 °C for 7 h. After cooling down to rt, brine (45 mL) was added and the mixture was extracted with CH₂Cl₂ (4×, 40 mL). The combined organic layers were dried (Na₂SO₄), concentrated in vacuo and the crude product was purified by automated fc (Snap, 100 g, V=400 mL, cyclohexane : ethyl acetate=75 : 25, R_f=0.26). Yellow resin, yield 954mg (71 %). C₁₉H₁₉NO₃ (309.4). HR‐MS (APCI): m/z=310.1446 (calcd. 310.1438 for C₁₉H₂₀NO₃ [M+H]⁺). ¹H NMR (400 MHz, CDCl₃): δ [ppm]=2.30–2.43 (m, 1H, 5‐H), 2.54 (ddd, J=15.9/11.3/6.7 Hz, 1H, 5‐H), 2.86 (ddd, J=15.5/6.9/0.8 Hz, 1H; 3‐H_ax), 2.99 (ddd, J=15.5/3.3/1.4 Hz, 1H, 3‐H_eq), 3.20 (ddd, J=14.5/11.3/3.9 Hz, 1H, 6‐H), 4.29 (d, J=10.7 Hz, 1H, 6‐H), 5.16–5.29 (m, 2H, CH₂‐bnzl), 5.84 (bs, 1H, 2‐H), 7.19–7.39 (m, 10H, H_arom.).¹³C NMR (101 MHz, CdCl₃): δ [ppm]=39.1 (C‐6), 40.7 (C‐5), 44.3 (C‐3), 54.8 (C‐2), 68.0 (CH₂‐ph), 126.8, 127.1, 127.8, 127.9, 128.1, 128.4, 128.68, 128.71, 129.0 (10 C, C_arom.), 136.4 (C‐1_benzyl), 139.8 (C‐1_phenyl), 155.6 (NCOO‐bnz), 207.4 (C‐4). Purity (HPLC): 99.9 %, t_R=20.5 min.

Ethyl (E)‐ and (Z)‐2‐(2‐phenyl‐1‐tosylpiperidin‐4‐ylidene)acetate (9a)

Piperidone 8a (624 mg, 1.89 mmol) was dissolved in dry toluene (8 mL) Then Ph₃P=CHCO₂Et (1.05g, 3.01mmol, 1.6eq) was added and the mixture was heated to reflux for 18h. The solvent was removed in vacuo and the crude product was purified by automated fc (Snap 100 g, V=200 mL, cyxlohexane : ethyl acetate=75 : 25, R_f=0.77 and 0.67). The diastereomers (Z)‐9a and (E)‐9a were not separated. Colorless resin, yield 782 mg (103 %, Ph₃P=O impurity). C₂₂H₂₅NO₄S (399.5). HR‐MS (APCI): m/z=400.1601 (calcd. 400.1577 for C₂₂H₂₆NO₄S [M+H]+). ¹H NMR (600 MHz, CDCl₃): δ [ppm]=1.22 (t, J=7.2 Hz, 1.2H, *OCH₂ CH₃), 1.27 (t, J=7.2 Hz, 1.8H, ^#OCH₂ CH₃), 2.03 (d, J=13.9 Hz, 0.6H, 5‐H_eq ^#), 2.19–2.28 (m, 1H, 5‐H_ax ^#, 5‐H_ax*), 2.30 (dd, J=14.5/6.2 Hz, 0.6H, 3‐H_ax ^#), 2.42 (s, 1.2 H, *CH_3Tos), 2.44 (s, 1.8H, ^#CH_3Tos), 2.61 (dd, J=14.5/6.2 Hz, 0.4H, 3‐H_ax*), 2.74 (dd, J=14.5/2.9 Hz, 0.4H, 3‐H_eq*), 3.05 (ddd, J=14.1/12.1/3.3 Hz, 0.6H, 6‐H_ax ^#), 3.12 (ddd, J=14.1/12.1/3.3 Hz, 0.4H, 6‐H_ax*), 3.49 (dt, J=15.5/3.0 Hz, 0.6H, 5‐H_eq*), 3.86 (ddd, J=14.1/5.0/3.1 Hz, 0.4H, 6‐H_eq*), 3.92 (dtd, J=14.0/3.5/2.0 Hz, 0.6H, 6‐H_eq ^#), 4.04–4.12 (m, 0.8H, *OCH₂ CH₃), 4.16 (qd, J=7.1/3.1 Hz, 1.2H, *OCH₂ CH₃), 4.42 (d, broad, J=15.0 Hz, 0.6H, 3‐H_eq ^#), 5.33 (dd, J=6.3/2.8 Hz, 0.4H, 2‐H_eq*), 5.42 (d, J=6.0 Hz, 0.6H, 2‐H_eq ^#), 5.64 (s, 0.6H, ^#=CHCO₂R), 5.70 (s, 0.4H, *=CHCO₂R), 7.19–7.25 (m, 1H, H‐4_phenyl ^#*), 7.25–7.34 (m, 4.8H, 2‐H_phenyl*, 6‐H_phenyl*, 3‐H_phenyl ^#*, 5‐H_phenyl ^#*, 3‐H_Tos ^#*, H‐5_Tos ^#*), 7.40 (d, J=8.2 Hz, 1.2H, 2‐H_phenyl ^#, 6‐H_pheny ^# _l), 7.69–7.74 (m, 0.8H, 2‐H_Tos*, 6‐H_Tos*), 7.77–7.84 (m, 1.2H, 2‐H_Tos ^#, 6‐H_Tos ^#). The ratio of diastereomers (Z)‐9a:(E)‐9a is 60 : 40. Signals of (Z)‐9a are marked with ^#, signals of (E)‐9a with *. ¹³C NMR (151 MHz, CDCl₃) δ [ppm]=14.3*, 14.4 (OCH₂ CH₃ ), 21.67*, 21.69 (CH_3Tos ), 28.4*, 34.9 (C‐5), 30.4, 38.5* (C‐3), 41.1*, 41.6 (C‐6), 56.37, 56.42* (C‐2), 60.01*, 60.04 (OCH₂CH₃), 117.6, 117.8* (=CHCO₂R), 127.19, 127.21, 127.23, 127.31, 127.44, 127.57 (5 C, C‐4_‐phenyl, C‐2_phenyl, C‐6_phenyl, C‐2_Tos, C‐6_Tos), 128.6, 128.7* (2 C, C‐3_phenyl, C‐5_phenyl), 130.0*, 130.1 (2 C, C‐3_Tos; C‐5_Tos), 137.8*, 138.2 (C‐1_Tos), 138.7, 139.4* (C‐1_phenyl), 143.6*, 143.7 (C‐4_Tos), 154.3 (C‐4), 166.0*, 166.2 (CO₂R). Signals of the minor diastereomer (E)‐9a are marked with *. Purity (HPLC): 96.2 %, t_R=23.4 min and 23.8 min.

Benzyl (E)‐ and (Z)‐4‐(ethoxycarbonylmethylene)‐2‐phenylpiperidine‐1‐carboxylate (9b)

Piperidone 8b (904.8 mg, 2.92 mmol, 1.0 eq) was dissolved in dry toluene (6 mL) Then Ph₃P=CHCO₂Et (2.04g, 5.85mmol, 2.0eq) was added and the mixture was heated to 115°C for 18h. The solvent was removed in vacuo and the crude product was purified by automated fc (Snap, 100 g, V=1700 mL, CH₂Cl₂ : ethyl acetate=4 : 1, R_f=0.52 and 0.44). Colorless resin, yield 1.09 g (98 %). C₂₃H₂₅NO₄ (379.5). HR‐MS (APCI): m/z=380.1884 (calcd. 380.1856 for C₂₃H₂₆NO₄ [M+H]⁺). ¹H NMR (400 MHz, CDCl₃) δ [ppm]=1.27 (t, J=7.1 Hz, 1.35H, CH₃*), 1.29 (t, J=7.1 Hz, 1.65H, CH₃ ^#), 2.22 (d, J=14.2 Hz, 0.55H, 5‐H_ax ^#), 2.45 (td, J=13.3/12.8, 5.6 Hz, 0.55H, 5‐H_eq ^#), 2.54–2.68 (m, 1H, 3‐H^#, 5‐H*), 2.74–2.88 (m, 0.9H, 3‐H*, 6‐H*), 2.97 (ddd, J=13.3/11.9/3.5 Hz, 0.55H, 6‐H_eq ^#), 3.19 (td, J=12.5/2.5 Hz, 0.45H, 3‐H_eq*), 3.44 (d, J=16.8 Hz, 0.45H, 5‐H_eq*), 4.13–4.20 (m, 2.55H, 6‐H^#, CH₂ CH₃*^#), 4.44 (d, J=15.2 Hz, 0.55H, 3‐H_eq), 5.07–5.27 (m, 2H, CH₂ ‐bnz), 5.42 (s, broad, 0.45H, 2‐H*), 5.68 (s, broad, 0.55H, 2‐H^#), 5.74 (s, 0.55H, =CHCOOR^#), 5.79 (s, 0.45H, =CHCOOR*), 7.11–7.44 (m, 10H, H_arom.). The ratio of diastereomers (Z)‐9b:(E)‐9b is 55 : 45. Signals of (Z)‐9b are marked with ^#, signals of (E)‐9b with *.¹³C NMR (101 MHz, CdCl3): δ [ppm]=14.3, 14.4* (CH₃), 27.1*, 28.2 (C‐5), 31.02, 31.04* (C‐3), 37.3, 40.5* (C‐6), 51.4*, 54.9 (C‐2), 59.9*, 60.9 (CH₂CH₃), 66.8*, 67.5 (CH₂‐bnz), 117.4, 125.2* (=CHCOOR), 127.3, 127.7, 128.06, 128.111, 128.16, 128.40, 128.5, 128.60, 128.63, 128.76 (10 C, C_arom.), 131.2, 131.1* (C‐1_phenyl), 136.78, 136.80* (C‐1_bnz), 140.5, 140.7* (NCOOR), 155.22, 155.24* (C‐4), 171.1*, 171.2 (C=O). Signals of the minor diastereomer (E)‐9b are marked with *. Purity (HPLC): 98.4 %, t_R=23.5 min and 23.8 min.

Ethyl cis‐ and trans‐2‐(2‐phenyl‐1‐tosyl‐piperidin‐4‐yl)acetate (10)

A solution of α,β‐unsaturated ester 9a (738mg, 1.80mmol) in CH₃OH (20mL) was added to a suspension of Pd/C (10 %, 197mg, 0.18mmol, 0.1eq.) in CH₃OH (5mL) and the mixture was stirred for 20h under H₂ (1 bar). Then, the mixture was filtered through Celite® 45 and the filtrate was concentrated in vacuo. The crude product was purified by automated fc (Snap, 100 g, cyclohexane : ethyl acetate=75 : 25, V=360 mL, R_f=0.53 and 0.47). Colorless oil, yield 591mg (81 %). C₂₂H₂₇NO₄S (401.5). HR‐MS (APCI): m/z=402.1759 (calcd. 402.1734 for C₂₂H₂₈NO₄S [M+H]⁺). ¹H NMR (600 MHz, CDCl₃): δ [ppm]=0.85 (m, 0.35H, H‐5_ax*), 1.01 (qd, J=12.7/4.6 Hz, 0.65H, H‐5_ax ^#), 1.20 (t, J=7.1 Hz, 1H, OCH₂ CH3*), 1.23 (t, J=7.2 Hz, 2H, OCH₂ CH3^#), 1.33 (ddd, J=13.7/12.2,/5.3 Hz, 0.65H, 3‐H^# _ax), 1.37–146 (m, 1H, H‐5_eq ^#*), 1.67 (ddd, J=13.8/9.7/3.8 Hz, 0.35H, 3‐H*_ax), 1.85–2.04 (m, 1.35H 3‐H*_eq, 4‐H^#*), 2.03–2.20 (m, 2H, CH₂CO^#*), 2.31–2.38 (m, 0.65H, 3‐H^# _eq), 2.41 (s, 1.05H, CH_3Tos*), 2.44 (s, 1.95H, CH_3Tos ^#), 2.99 (ddd, J=14.7/13.1/2.9 Hz, 0.65H, 6‐H_ax ^#), 3.10 (ddt, J=12.9/8.4/4.2 Hz, 0.35H, 6‐H_ax*), 3.88 (m, 1H, 6‐H_eq ^#*), 4.02–4.15 (m, 2H, OCH₂CH₃ ^#*), 4.18 (dd, J=9.7/4.3 Hz, 0.35H, 2‐H_ax*), 5.34 (d, J=5.2 Hz, 0.65H, 2‐H_eq ^#), 7.17–7.39 (m, 7H, H_phenyl ^#*, H‐3_Tos ^#*, H‐5_Tos ^#*), 7.44–7.49 (m, 0.7H, H‐2_Tos*, H‐6_Tos*), 7.72–7.79 (m, 1.3H, H‐2_Tos ^#, H‐6_Tos ^#). The ratio of diastereomers cis‐10:trans‐10 is 35 : 65. Signals of trans‐10 are marked with ^#, signals of cis‐10 with *. ¹³C NMR (151 MHz, CDCl₃): δ [ppm]=14.3*, 14.4 (OCH₂ CH₃), 21.6*, 21.7 (CH_3Tos), 27.6, 30.9* (C‐4), 30.7*, 30.8 (C‐5), 33.1, 39.6* (C‐3), 39.7*, 41.0 (CH₂CO), 41.5, 45.0* (C‐6), 55.2, 60.8* (C‐2), 60.5, 60.6 (OCH₂CH₃), 127.0, 127.11, 127.14, 127.4, 127.8, 128.2, 128.8,128.9, 129.4, 129.9 (9 C, C_phenyl, C‐2_Tos, C‐3_Tos, C‐5_Tos, C‐6_To) 138.7, 138.6 (C‐1_phenyl), 141.6*, 141.7 (C‐1_Tos), 143.3, 143.4 (C‐4_Tos) 172.1, 172.6* (CO₂R). Signals of cis‐10 are marked with *. Purity (HPLC): 67.1 %, t_R=23.1 and 23.4 min.

cis‐ and trans‐2‐(2‐Phenyl‐1‐tosylpiperidin‐4‐yl)ethan‐1‐ol (11)

A mixture of LiAlH₄ (76.3 mg, 2.0 mmol, 2.0 eq) and THF (10 mL) was stirred for 10 min at 0 °C. Then a solution of 10 (405 mg, 1.05 mmol) in THF (25 mL) was added dropwise to the LiAlH₄ suspension under ice cooling. The mixture was stirred for 20 min at 0 °C and then at rt. for 2.5 h. Under ice cooling H₂O was added dropwise and the mixture was heated to reflux for 30 min. After cooling to rt, the mixture was filtered over Celite® 45 and the celite layer was washed with ethyl acetate. The solvent was removed in vacuo and the crude product was purified by automated fc (Snap 50 g, cylxohexane : ethyl acetate=Gradient 80 : 20 to 60 : 40, V=400 mL, R_f=0.18 (cylxohexane : ethyl acetate=60 : 40)). Colorless resin, yield 333.8 mg (89 %). C₂₀H₂₅NO₃S (359.5 g/mol). HR‐MS (APCI): m/z=360.1640 (calculated 360.1628 for C₂₀H₂₆NO₃S [M+H]+). ¹H NMR (600 MHz, CDCl₃): δ [ppm]=0.88 (m, 0.17H, 5‐H_ax*), 0.98 (tdd, J=13.1/11.9/4.6 Hz, 0.83H, 5‐H_ax ^#), 1.35 (ddd, J=13.8/12.2/5.4 Hz, 0.85 Hz, 3‐H_ax ^#), 1.30–1.40 (m, 1.83H, 3‐H_ax ^#*, CH₂CH₂OH^#), 1.41–1.48 (m, 1H, 5‐H_eq ^#*), 1.50–1.70 (m, 1H, 4‐H), 1.84–1.89 (m, 0.34H, CH₂CH₂OH*) 2.29 (ddt, J=13.8/3.5/2.0 Hz, 1H, 3‐H_eq), 2.40 (s, 0.5H, *CH₃), 2.44 (s, 2.5H, CH₃), 2.99 (ddd, J=14.5/13.1/2.9 Hz, 0.83H, 6‐H_ax ^#), 3.07 (ddd, J=12.9/8.7/4.3 Hz, 0.17H, 6‐H_ax*), 3.56–3.67 (m, 2H, CH₂OH^#*), 3.84–3.93 (m, 1H, 6‐H_eq ^#*), 4.10–4.14 (m, 0.17H, 2‐H*), 5.34 (d, J=4.8 Hz, 0.83H, 2‐H_eq ^#), 7.18–7.25 (m, 2H, 2x H_phenyl), 7.28–7.34 (m, 5H, 3‐H_Tos, 5‐H_Tos, 3x H_phenyl), 7.46 (d, J=8.3 Hz, 0.34H, 2‐H_Tos*, 6‐H_Tos*), 7.74–7.78 (m, 1.66H, 2‐H_Tos, 6‐H_Tos). The ratio of diastereomers trans‐11:cis‐11 is 83 : 17. Signals of trans‐11 are marked with ^#, signals of cis‐11 with *. ¹³C NMR (151 MHz, CDCl₃): δ [ppm]=24.1, 24.2 (CH₃), 29.7, 33.1* (C‐4), 33.2*, 33.6 (C‐5), 36.4, 43.0 (C‐3), 41.0*, 41.9 (CH₂CH₂OH), 44.2, 47.8* (C‐6), 57.8, 63.8* (C‐2), 62.7, 62.8* (CH₂OH), 129.4, 129.5, 129.7, 129.82, 129.85, 130.4, 130.6, 131.3, 131.8, 132.3 (9 C, 5x C_phenyl, 4× C_Tos), 138.3*, 141.3 (C‐1_Tos), 141.5, 144.6* (C‐1_phenyl), 145.6*, 145.7 (C‐4_Tos). Signals of cis‐11 are marked with *. Purity (HPLC): 82.3 %, t_R=20.2 min, 20.4 min.

cis‐ and trans‐2‐(2‐Phenyl‐1‐tosylpiperidin‐4‐yl)ethyl methanesulfonate (12)

A solution of alcohol 11(130 mg, 0.36 mmol) in CH₂Cl₂ (10 mL) was cooled to 0 °C, Et₃N (170 μL, 1.23 mmol, 3.4 eq) was added and the solution was stirred for 10 min under ice cooling before methanesulfonyl chloride (40 μL, 0.52 mmol, 1.5 eq) was added. The reaction mixture was stirred at rt. for 18 h. Then, the mixture was washed with NaOH (2 x, 0.5 M, 5 mL) and NH₄Cl (5 mL), dried (Na₂SO₄) and the solvent was removed in vacuo. The crude product was purified with automated fc (Snap 50 g, cyclohexane : ethyl acetate=50 : 50, V=200 mL, R_f=0.48). Colorless resin, yield 148 mg (94 %). C₂₁H₂₇NO₅S₂ (437.6 g/mol). HR‐MS (APCI): m/z=438.1422 (calcd 438.1403 for C21H28NO5S2 [M+H]+). ¹H NMR (600 MHz, CDCl₃): δ [ppm]=0.88 (m, 0.17H, 5‐H_ax*), 1.03 (qd, J=13.1/4.6 Hz, 0.83H, 5‐H_ax ^#), 1.38 (ddd, J=13.7/12.3/5.4 Hz, 1H, 3‐H_ax ^#*), 1.46 (dt, J=13.3/2.9 Hz, 0.83H, 5‐H_eq ^#), 1.56 (q, J=6.8 Hz, 2H, CH₂ CH₂OSO₂ ^#*), 1.66 (dddt, J=14.4/11.8/5.3/2.7 Hz, 1H, 4‐H), 1.87 (m, 0.34H, 3‐H_eq*, 5‐H_eq*) 2.31 (dq, J=13.7, 2.2 Hz, 0.83H, 3‐H_eq ^#), 2.41 (s, 0.5H, *CH_3Tos), 2.44 (s, 2.5H, CH_3Tos), 2.93 (s, 2.5H, SO₂ CH₃), 2.95 (s, 0.5H, *SO₂ CH₃), 3.00 (ddd, J=14.5/13.1/2.9 Hz, 0.83H, 6‐H_ax ^#), 3.10 (ddd, J=12.9/8.5,/4.3 Hz, 0.17H, 6‐H_ax*), 3.87 (ddd, J=13.1/6.6/4.4 Hz, 0.17H, 6‐H_eq*), 3.89–3.95 (d, broad, J=14.2 Hz, 0.83H, 6‐H_eq ^#), 4.19 (ddt, J=10.0, 6.4 Hz, 2H, CH₂OSO₂), 4.11–4.17 (m, 0.17H, 2‐H*), 5.36 (d, J=4.9 Hz, 0.83H, 2‐H_eq ^#), 7.17–7.25 (m, 2H, 3‐H_Tos ^#*, 5‐H_Tos ^#*), 7.28–7.34 (m, 5H, H_pheny ^#*_l), 7.43–7.48 (m, 0.34H, 2‐H_Tos*, 6‐H_Tos*), 7.72–7.79 (m, 1.66H, 2‐H_Tos ^#, 6‐H_Tos ^#). The ratio of diastereomers trans‐12:cis‐12 is 83 : 17. Signals of trans‐12 are marked with ^#, signals of cis‐12 with *. ¹³C NMR (151 MHz, CDCl₃): δ [ppm]=21.68*, 21.70 (CH_3Tos), 27.1*, 27.2 (C‐4), 30.9 (C‐5), 33.5 (C‐3), 34.7*, 35.7 (CH₂CH₂OSO₂CH₃), 37.5, 37.6 (SO₂ CH₃), 41.5, 44.9* (C‐6), 55.1, 60.8* (C‐2), 66.9, 67.4* (CH₂OSO₂CH₃), 126.9, 127.0, 127.1, 127.3*, 127.4*, 127.8*, 128.2*, 128.9, 129.5*, 129.9 (9 C, 5x C_phenyl, 4× C_Tos), 135.8*, 138.5 (C‐1_Tos), 138.6, 141.7* (C‐1_phenyl), 143.33, 143.38* (C‐4_Tos). Signals of cis‐12 are marked with *. Purity (HPLC): 99.6 %, t_R=21.9 min, 22.1 min.

cis‐ and trans‐N‐Benzyl‐2‐(2‐phenyl‐1‐tosyl‐piperidin‐ 4‐yl)ethan‐1‐amine (13a)

Mesylate 12 (200 mg, 0.45 mmol) was dissolved in CH₃CN (15 mL), dest. benzylamine (147 μL, 1.35 mmol, 3.0 eq) was added and the reaction mixture was stirred under reflux for 18 h. The solvent was removed in vacuo, the crude product was dissolved in ethyl acetate and the solution was washed with 0.5 M NaOH (2 x, 10 mL), dried (Na2SO4) and concentrated in vacuo. The crude product was purified by fc (d=2 cm, l=18 cm, V=35 mL, CH₂Cl₂ : MeOH : NH₃=94 : 5 : 1, R_f=0.43 and 0.35). Colorless solid, mp 79 °C, yield 122 mg (60 %). C₂₇H₃₂N₂O₂S (448.6 g/mol). HR‐MS (APCI): m/z=449.2228 (calcd. 449.2257 for C₂₇H₃₃N₂O₂S [M+H]⁺). ¹H NMR (600 MHz, CDCl₃): δ [ppm]=0.86–0.98 (qd, J=12.1/4.9 Hz, 0.35H, 5‐H_ax*), 1.25–1.40 (m, 3.35H, H‐3_ax*, H‐5_ax ^#, 5‐H_eq*, CH₂ CH₂NH^#*), 1.43–1.50 (m, 1H, 4‐H^#*), 1.54 (dt, J=13.5/10.3 Hz, 0.65H, 3‐H_ax ^#), 1.78–1.85 (m, 1.3H, 5‐H_eq ^#, 3‐H_eq ^#), 2.24 (ddt, J=13.8/3.6/2.0 Hz, 0.35H, 3‐H_eq*), 2.39 (s, 1.95H, CH_3Tos), 2.43 (s, 1.05H, CH_3Tos*), 2.48–2.62 (m, 2H, CH₂ CH₂NH^#*), 2.96 (ddd, J=14.5/13.2/2.9 Hz, 0.35H, 6‐H_ax*), 3.08 (ddd, J=12.9/8.4/4.4 Hz, 0.65H, 6‐H_ax ^#), 3.71 (s, 1.3H, NCH₂‐ph), 3.72 (s, 0.7H, NCH₂‐ph), 3.84 (ddd, J=12.8/6.6/4.5 Hz, 0.65H, 6‐H_eq ^#), 3.86–3.91 (m, 0.35H, 6‐H_eq*), 4.13 (dd, J=10.1/4.5 Hz, 0.65H, 2‐H_ax ^#), 5.32 (d, J=4.5 Hz, 0.35H, 2‐H_eq*), 7.16–7.23–7.33 (m, 12H, 3‐H_Tos, 5‐H_Tos, 5x H_phenyl, 5x H_benzyl), 7.43–7.48 (m, 1.3H, 2‐H_Tos ^#, 6‐H_Tos ^#), 7.73–7.77 (m, 0.7H, 2‐H_Tos*, 6‐H_Tos*). The ratio of diastereomers cis‐13a:trans‐13a is 65 : 35. Signals of cis‐13a are marked with ^#, signals of trans‐13a with *. ¹³C NMR (151 MHz, CDCl₃): δ [ppm]=21.6, 21.7* (CH₃), 28.4*, 31.9 (C‐4*), 30.7, 33.7* (C‐3), 31.1, 40.4 (C‐5*), 36.0, 37.0* (CH₂CH₂NH), 41.8*, 45.2 (C‐2), 46.4*, 46.7 (CH₂ CH₂NH), 54.1, 54.2* (CH₂‐bnz), 55.3*, 61.2 (C‐2), 127.0, 127.1, 127.29, 127.31, 127.8, 128.1, 128.2, 128.3, 128.6, 128.8, 129.3, 129.8 (14 C, C_aromat), 135.9 (C‐1_Tos), 139.1*, 142.2 (C‐1_phenyl), 140.2 (C‐1_bnz), 143.1, 143.2* (C‐4_Tos). Signals of trans‐13a are marked with *. Purity (HPLC): 96.3 %, t_R=20.2 min, 20.5 min.

cis‐ and trans‐N‐(3‐Phenylpropyl)‐2‐(2‐phenyl‐1‐tosylpiperidin‐4‐yl)ethan‐1‐amine (13b)

Mesylate 12 (69.2 mg, 0.16 mmol,) was dissolved in CH₃CN (8 mL), phenylpropylamine (88 μL, 0.6 mmol, 3.8 eq) was added and the reaction mixture was stirred under reflux for 18 h. The solvent was removed in vacuo, the crude product was dissolved in ethyl acetate and the solution was washed with 0.5 M NaOH (2×, 10 mL), dried (Na₂SO₄) and concentrated in vacuo. The crude product was purified twice by fc. First column (d=1 cm, l=18 cm, V=25 mL, CH₂Cl₂ : MeOH : Et₃N=93 : 5 : 2). Second column (d=1 cm, l=18 cm, V=12 mL, CH₂Cl₂ : ethyl acetate:EtNMe₂=90 : 8 : 2, R_f=0.27). Light‐yellow resin, yield 65.6 mg (87 %). C₂₉H₃₆N₂O₂S (476.7 g/mol). HR‐MS (APCI): m/z=477.2568 (calcd. 477.2570 for C₂₉H₃₇N₂O₂S [M+H]⁺). ¹H NMR (600 MHz, CDCl₃) δ [ppm]=0.94 (qd, J=13.0/4.8 Hz, 0.35H, H‐5_ax*), 1.22–1.48 (m, 4H, 3‐H_ax*, 4‐H^#*, 5‐H_ax ^#, CH₂ CH₂NH−R), 1.54 (dt, J=13.5/10.3 Hz, 0.65H, H‐3_ax ^#), 1.68–1.91 (m, 3.65H, R‐NH‐CH₂ CH₂ CH₂‐phenyl, 5‐H_eq ^#*, 3‐H_ax ^#), 2.24 (d, J=13.3 Hz, 0.35H, 3‐H_eq*), 2.39 (s, 1.95H, CH_3Tos ^#), 2.43 (s, 1.05H, CH_3Tos*), 2.45–2.72 (m, 6H, CH₂ CH₂ NH−R, R‐NHCH₂ CH₂ CH₂ ‐phenyl), 2.96–2.98 (m, 0.35H, 6‐H_ax*), 3.08 (ddd, J=12.8/8.3/4.4 Hz, 0.65H, 6‐H_ax ^#), 3.84 (ddd, J=12.8/6.5/4.4 Hz, 0.65H, 6‐H_eq ^#), 3.88–3.94 (m, 0.35H, 6‐H_eq*), 4.08–4.19 (m, 0.65H, 2‐H_ax ^#), 5.32 (d, J=5.2 Hz, 0.35H, 2‐H_eq*), 7.11–7.33 (m, 12H, H_arom.), 7.38–7.48 (m, 1.3, 2‐H_Tos ^#, 6‐H_Tos ^#), 7.68–7.80 (m, 0.7H, 2‐H_Tos*, 6‐H_Tos*). The ratio of diastereomers cis‐13b:trans‐13b is 65 : 35. Signals of cis‐13b are marked with ^#, signals of trans‐13b with *. ¹³C NMR (151 MHz, CdCl₃) δ [ppm]=21.63, 21.68* (CH_3Tos), 30.7 (C‐5), 31.7 (R‐NHCH₂ CH₂CH₂‐phenyl), 32.0 (C‐4), 33.8 (R‐NHCH₂CH₂ CH₂‐phenyl), 40.5 (C‐3), 41.7*, 45.2 (C‐6), 47.0*, 47.3 (CH₂ CH₂NH−R), 49.6 (R‐NHCH₂CH₂CH₂‐phenyl), 55.3*, 61.2 (C‐2), 125.9, 126.96, 126.98, 127.2, 127.3, 127.4, 127.8, 128.1, 128.5, 128.8, 129.3, 129.8 (14 C, C_arom.), 136.0 (C‐1_Tos), 138.8 (C‐1_phenyl) 142.2, 142.3* (C‐1_arom.), 143.1, 143.2* (C‐4_Tos). Where signals of cis and trans could be distinguished, signals of trans‐13b are marked with *. Purity (HPLC): 87.7 %, t_R=20.9 min.

cis‐ and trans‐N‐Benzyl‐2‐(2‐phenylpiperidin‐4‐yl)ethan‐1‐amine (4a)

Sulfonamide 13a (30.5 mg, 0.07 mmol) and Mg⁰ turnings (27.5 mg, 1.13 mmol, 16.0 eq) were suspended in in MeOH (5 mL) and the mixture was stirred under irradiation with ultrasound for 8 h. Then the mixture was acidified with HOAc to pH=5 and then the pH‐value was adjusted to pH 10 with NH₃. The organic layer was separated and the aqueous layer was extracted with CH₂Cl₂ (3×5 mL). The combined organic layers were dried (Na₂SO₄) and concentrated in vacuo. The crude product was purified by fc (d=1 cm, l=15 cm, V=30 mL, CH₂Cl₂ : MeOH : NH₃=94 : 5 : 1, R_f=0.15 (CH₂Cl₂:CH₃OH:NH₃=93 : 5 : 2)). Colorless resin, yield 16.1 mg (80 %). C₂₀H₂₆N₂ (294.4 g/mol). HR‐MS (APCI): m/z=295.2177 (calcd. 295.2169 for C₂₀H₂₇N₂ [M+H]⁺). ¹H NMR (600 MHz, CDCl₃): δ [ppm]=0.79–0.96 (m, 0.2H, 5‐H*), 1.17–2.03 (m, 7H, 3‐H_ax, 3‐H_eq, 4‐H, 5‐H_ax, 5‐H_eq, CH₂CH₂NH), 2.68 (q, J=7.7 Hz, 2H, CH₂ CH₂NH), 2.80 (td, J=12.1/2.6 Hz, 0.8, 6‐H), 2.90–2.31 (m, 0.2H, 6‐H*), 3.21 (ddd, J=11.8/4.2/2.4 Hz, 0.8H, 6‐H), 3.45–3.52 (m, 0.2H, 6‐H*), 3.63 (dd, J=11.3/2.5 Hz, 0.8H, 2‐H), 3.79 (s, 1.6H, CH₂‐ph), 3.81 (s, 0.4H, CH₂‐ph), 3.89 (dd, J=9.7/2.9 Hz, 0.2H, 2‐H*), 7.18–7.45 (m, 10H, H_arom). The ratio of diastereomers cis‐4a:trans‐4a is 80 : 20. Due to low intensity, some of the signals for trans‐4a could not be detected. Signals of trans‐4a are marked with *. ¹³C NMR (101 MHz, CDCl₃): δ [ppm]=32.1 (C‐5), 34.8 (C‐4), 37.9 (CH₂CH₂NH), 41.2 (C‐3), 46.4 (CH₂ CH₂NH), 47.1 (C‐6), 53.8 (CH₂‐ph), 61.9 (C‐2), 127.0, 127.4, 127.5, 128.5, 128.6, 128.7 (10 C, C_arom), 139.1 (C‐1_bnz), 141.3 (C_phenyl). Signals of trans‐4 a are not visible in the ¹³C NMR spectrum. Purity (HPLC): 99.0 %, t_R=11.9 min.

cis‐ and trans‐N‐(3‐Phenylpropyl)‐2‐(2‐phenyl‐piperidin‐ 4‐yl)ethan‐1‐amine (4b)

Sulfonamide 13b (27.6 mg, 0.06 mmol) and Mg⁰ turnings (28.3 mg, 1.16 mmol, 20.0 eq) were suspended in in MeOH (5 mL) and the mixture was irradiated with ultrasound for 5 h. Then, the mixture was acidified with HOAc to pH=5 and then, the pH‐value was adjusted to pH =10 with NH₃. The organic layer was separated and the aqueous layer was extracted with CH₂Cl₂ (3×5 mL). The combined organic layers were dried (Na₂SO₄) and concentrated in vacuo. The crude product was purified by fc (d=1 cm, l=18 cm, V=25 mL, CH2Cl2 : ethyl acetate:EtNMe₂=90 : 8 : 2, R_f=0.13). Brown resin, yield 6.9 mg (37 %). C₂₂H₃₀N₂ (322.5 g/mol). HR‐MS (APCI): m/z=323.2486 (calcd. 323.2482 for C C₂₂H₃₁N₂ [M+H]⁺). ¹H NMR (600 MHz, CDCl₃) δ [ppm]=1.16–1.26 (m, 2H, 3‐H_ax, 5‐H_ax), 1.48 (tq, J=13.8/6.7/6.2 Hz, 2H, CH₂CH₂NH−R), 1.59 (m, 0.75H, H‐4^#), 1.72 (d, broad, J=15.4 Hz, 1H, 5‐H_eq), 1.75‐1.85 (m, 3H, 3‐H_eq, R‐NH‐CH₂ CH₂ CH₂‐ph), 1.90–1.94 (m, 0.25H, 4‐H*), 2.64 (m, 6H, R‐NH‐CH₂ CH₂ CH₂ ‐ph, CH₂ CH₂ NH−R), 2.80 (td, J=12.1/2.6 Hz, 0.75H, 6‐H_ax ^#), 2.93 (dt, J=12.2/4.3 Hz, 0.25H, 6‐H_ax*), 2.99 (td, J=11.7/2.9 Hz, 0.25H, 6‐H_eq*), 3.22 (ddd, J=11.6/4.1/2.6 Hz, 0.75H, 6‐H_eq ^#), 3.60 (dd, J=11.2/2.4 Hz, 0.75H, 2‐H_ax ^#), 3.89 (dd, J=10.4/2.8 Hz, 0.25H, 2‐H_ax*), 7.13–7.38 (m, 10H, H_arom ^#*). The ratio of diastereomers cis‐4b:trans‐4b is 75 : 25. Signals of cis‐4b are marked with ^#, signals of trans‐4b with *. ¹³C NMR (151 MHz, CDCl₃) δ [ppm]=30.4*, 32.7 (C‐5), 31.6*, 31.7 (R‐NHCH₂ CH₂CH₂‐ph), 33.8, 33.8* (R‐NHCH₂CH₂ CH₂‐ph), 35.1 (C‐4), 37.4 (CH₂CH₂NH−R), 38.7*, 41.8 (C‐3), 42.3*, 47.4 (C‐6) 47.3, 48.4* (CH₂ CH₂NH−R), 49.7, 49.8 (R‐NHCH₂CH₂CH₂‐ph), 56.1*, 62.1 (C‐2), 125.9, 126.8, 127.3, 128.48, 128.50, 128.55 (10 C, C_arom, C_phenyl), 142.2 (C‐1_arom), 145.0*, 145.3 (C‐1_phenyl). Signals of trans‐4b are marked with *, where they could be distinguished from signals of cis‐4b. Purity (HPLC): 97.8 % t_R=14.2 min.

Ethyl cis‐ and trans‐2‐(2‐phenylpiperidin‐4‐yl)‐acetate (14)

A solution ion of α,β‐unsaturated ester 9b (2.99 g, 1.0mmol) in CH₃OH (27mL) was added to a suspension of Pd/C (10 %, 841.0 mg, 0.79mmol, 0.1eq.) in CH₃OH (3mL) and the mixture was stirred for 20h under H₂ (3 bar). Then, the mixture was filtered through Celite® 45 and concentrated in vacuo. The crude product was purified by fc (d=6 cm, h=16 cm, V=1500 mL, CH₂Cl₂ : MeOH : dimethylethylamin,=97 : 2 : 1, R_f=0.27). Colorless resin, yield 1.37 g (70 %). C₁₅H₂₁NO₂ (247.3). HR‐MS (APCI): m/z=248.1635 (calcd. 248.1645 for C₁₅H₂₂NO₂ [M+H]⁺). ¹H NMR (600 MHz, CH₃OD): δ [ppm]=1.22–1.29 (m, 3H, CH₃ ^#*), 1.26–1.35 (m, 1.5H, 5‐H_ax ^#, 3‐H_ax ^#), 1.51–1.58 (d, broad, J=13.4 Hz, 0.25H, 5‐H_ax*), 1.68–1.73 (m, 0.25H, 3‐H_ax*), 1.78 (dt, J=13.8/2.7 Hz, 0.75H, 5‐H_eq ^#), 1.81–1.90 (m, 1H, 5‐H_eq*, 3‐H_eq ^#), 1.91–2.03 (m, 0.25H, 3‐H_eq*), 2.09 (dddt, J=15.6/11.5/7.9/3.9 Hz, 0.75H, 4‐H^#), 2.24–2.35 (m, 1.5H, CH₂ COOR^#), 2.39–2.49 (m, 0.25H, 4‐H*), 2.59 (d, broad J=7.6 Hz, 0.5H, CH₂ COOR*), 2.82 (td, J=12.5/2.9 Hz, 0.75H, 6‐H_ax ^#), 2.90–3.00 (m, 0.5H, 6‐H_ax*,6‐H_eq*), 3.14–3.28 (d, broad J=12.5 Hz, 0.75H, 6‐H_eq ^#), 3.66 (dd, J=11.6/2.5 Hz, 0.75H, 2‐H_ax ^#), 3.92 (dd, J=10.6/2.9 Hz, 0.25H, 2‐H_ax*), 4.14 (dqd, J=14.2/7.1/1.2 Hz, 2H, CH₂ CH₃ ^#*), 7.14–7.41 (m, 5H, H_phenyl ^#*). The ratio of diastereomers cis‐14:trans‐14 is 75 : 25. Signals of cis‐14 are marked with ^#, signals of trans‐14 with *. ¹³C NMR (151 MHz, CD₃OD): δ [ppm]=14.57*, 14.60 (CH₃), 30.4*, 32.6 (C‐5), 30.5*, 35.1 (C‐4), 38.2*, 42.4 (CH₂COOR), 38.3*, 41.3 (C‐3) 42.5*, 47.5 (C‐6), 56.6*, 62.6 (C‐2), 61.4, 61.5* (CH₂CH₃), 126.9, 127.7, 128.2, 128.4, 129.4, 129.5 (5 C, C_phenyl), 144.7*, 145.0 (C‐1_phenyl), 174.2 174.6* (C=O). Signals of trans‐14 are marked with *. Purity (HPLC): 84.4 %, t_R=13.9 min.

Ethyl cis‐ and trans‐2‐(1‐methyl‐2‐phenyl‐piperidin‐ 4‐yl)acetate (15a)

NaBH(OAc)₃ (2.46 g, 11.6 mmol, 3.0 eq) was added to a stirred solution of formalin (37 %, 866 μL, 11.6 mmol, 3.0 eq) and amine 14 (960 mg, 3.88 mmol) in CH₂Cl₂ (25 mL). The reaction mixture was stirred over night at rt. A saturated solution of NaHCO₃ (20 mL) was added. The organic layer was separated and the aqueous layer was extracted with CH₂Cl₂ (3×20 mL). The combined organic layers were dried (NaSO₄), filtered and concentrated in vacuo. The crude product was purified by fc (d=3 cm, l=16 cm, V=270 mL, cyclohexane : ethyl acetate=3 : 1+1 % dimethylethylamine, R_f=0.31). Colorless oil, yield 671 mg (66 %). C₁₆H₂₃NO₂ (261.4). HR‐MS (APCI): m/z=262.1823 (calcd. 262.1802 for C₁₆H₂₄NO₂ [M+H]⁺). ¹H NMR (400 MHz, CDCl₃): δ [ppm]=1.23 (t, J=7.1 Hz, 2.25H, CH₃ ^#), 1.26 (t, J=7.1 Hz, 0.75H, CH₃*), 1.30–1.42 (m, 0.75H, 3‐H_ax ^#), 1.43–1.55 (m, 0.75H, 5‐H_ax ^#), 1.61 (m, 0.5H, 3‐H_ax*, 5‐H_ax*), 1.75–1.83 (m, 1.5H, 3‐H_eq ^#, 5‐H_eq ^#), 1.91–2.00 (m, 0.75H, 4‐H^#), 2.01 (s, 2.25H, N‐CH₃ ^#), 2.03 (s, 0.75H, N‐CH₃*), 2.15–2.26 (m, 2.25H, CH₂ COOR^#, 6‐H_ax ^#), 2.26–2.40 (m, 0.25H, 6‐H_ax*), 2.38–2.52 (m, 0.25H, 4‐H*), 2.51–2.55 (m, 0.5H, CH₂ COOR*), 2.85 (d, J=11.4 Hz, 1H, 2‐H_ax ^#, 6‐H_eq*), 3.04 (m, 1H, 2‐H_ax*, 6‐H_eq ^#), 4.04–4.17 (m, 2H, CH₂ CH₃ ^#*), 7.23 (dt, J=8.5/4.2 Hz, 1H, 4‐H_phenyl ^#*), 7.31 (d, J=4.8 Hz, 4H, 2‐H_phenyl ^#*, 3‐H_phenyl ^#*, 5‐H_phenyl ^#*, 6‐H_phenyl ^#*). The ratio of diastereomers cis‐15a:trans‐15a is 75 : 25. Signals of cis‐15a are marked with ^#, signals of trans‐15a with *. ¹³C NMR (101 MHz, cdcl₃): δ [ppm]=14.39, 14.43* (CH₃), 28.9*, 33.7 (C‐4), 32.3, 39.6* (C‐5), 36.6*, 41.4 (CH₂COOR), 42.1 (C‐3), 44.2, 44.5* (N‐CH₃), 51.7*, 57.0 (1 C, C‐6), 60.4, 60.5* (CO₂ CH₂CH₃), 65.3*, 70.5 (C‐2), 127.3, 127.6, 128.6 (5 C, C_phenyl), 144.1 (C‐1_phenyl), 172.7, 173.1* (C=O). Signals of trans‐15 a are marked with *. Purity (HPLC): 98.9 %, t_R=14.5 min.

Ethyl cis‐ and trans‐2‐(1‐ethyl‐2‐phenyl‐piperidin‐4‐yl)acetate (15b)

NaBH(OAc)₃ (58.6 mg, 0.28 mmol, 1.6 eq) was added to a solution of acetaldehyde (12.3 mg, 0.28 mmol, 1.6 eq) and amine 14 (40 mg, 0.17 mmol, 1.0 eq) in CH₂Cl₂ (5 mL). The reaction mixture was stirred for 18 h at rt, before a saturated solution of NaHCO₃ (10 mL) and CH₂Cl₂ (5 mL) was added. The organic layer was separated and the aqueous layer was extracted with CH₂Cl₂ (3×10 mL). The combined organic layers were dried (NaSO₄), filtered and concentrated in vacuo. The crude product was purified by fc (d=3 cm, l=12 cm, V=30 mL, CH₂Cl₂:MeOH=97 : 2+1 % Et₃N, R_f=0.23). Pale yellow oil, yield 30 mg (68 %). C₁₅H₂₁NO₂ (275.4). HR‐MS (APCI): m/z=276.1972 (calcd. 276.1958 for C₁₅H₂₃NO [M+H]⁺). ¹H NMR (400 MHz, CDCl₃): δ [ppm]=00.92 (t, J=7.0 Hz, 3H, N‐CH₂‐CH₃ ), 1.23 (t, J=7.1 Hz, 3H, CO₂CH₂CH₃ ), 1.25–1.35 (m, 0.8H, 5‐H_ax ^#), 1.35–1.48 (m, 0.8H, 3‐H_ax ^#), 1.53–1.62 (m, 0.4 H, 3‐H_ax*, 5‐H_ax*), 175 −1.88 (m, 1.8H, 3‐H_eq ^#, 5‐H_eq ^#, 3‐H_eq*), 1.89–2.07 (m, 1.4H, 4‐H^#, 5‐H_eq*, N‐CH₂ CH₃*), 2.08–2.18 (m, 0.8H, 6‐H_ax ^#), 2.17–2.28 (m, 2H, CH₂ COOR^#*), 2.32– 2.46 (m, 0.4 H, 4‐H*, 6‐H_ax*) 2.47–2.61 (m, 1.6H, N‐CH₂ CH₃ ^#), 2.93–3.03 (m, 0.2H, 6‐H_eq*), 3.09 (dd, J=2.6/11.2 Hz, 0.8H, 2‐H_ax ^#), 3.18 (dt, J=3.6/11.6 Hz, 0.8H, 6‐H_eq ^#), 3.29 (dd, J=11.1/2.8 Hz, 0.2 H, 2‐H_ax*), 4.02–4.18 (m, 2H, CO₂ CH₂ CH₃ ^#*), 7.22 (m, 1H, 4‐H_phenyl ^#*), 7.27–7.34 (m, 4H, 2‐H_phenyl ^#*, 3‐H_phenyl ^#*, 5‐H_phenyl ^#*, 6‐H_phenyl ^#*). The ratio of diastereomers cis‐15b:trans‐15b is 80 : 20. Signals of cis‐15b are marked with ^#, signals of trans‐15b with *. ¹³C NMR (101 MHz, CdCl₃): δ [ppm]=11.3, 11.5* (N‐CH₂‐CH₃ ), 14.4, 14.5* (CO₂CH₂ CH₃ ), 29.1*, 33.8 (C‐4), 29.5*, 42.9 (C‐5), 32.4, 36.9* (C‐3), 41.5 (CH₂COOR), 46.7*, 52.0 (C‐6), 48.8, 49.0* (N‐CH₂‐CH₃), 60.3, 60.4* (CO₂ CH₂CH₃), 62.9*, 68.2 (C‐2), 127.1, 127.6, 127.61, 128.5, 128.6 (5 C, C_phenyl), 144.9 (C‐1_phenyl), 172.8 (C=O). Signals of trans‐15b are marked with *.Purity (HPLC): 80.8 %, t_R=15.4 min.

cis‐ and trans‐2‐(1‐Methyl‐2‐phenylpiperidin‐4‐yl)ethan‐1‐ol (16a)

A solution of ester 15a (400 mg, 1.53 mmol) in THF (15 mL) was added dropwise to an ice‐cooled suspension of LiAlH₄ (123 mg, 3.25 mmol, 2.1 eq) in THF (20 mL). The mixture was stirred for 30 min at 0 °C. Ice cooling was removed and the reaction mixture was stirred for 2 h at rt. H₂O was added under ice cooling until the gas formation has stopped and the mixture was heated to reflux for 30 min. After cooling to rt, the organic layer was separated and the aqueous layer was extracted with EtOAc (3 x 10 mL). The combined organic layers were dried (Na₂SO₄), filtered and concentrated in vacuo. The crude product was purified by fc (d=3 cm, l=15 cm, V=140 mL, CH₂Cl₂:MeOH=97 : 2+1 % Et₃N, R_f=0.23). Colorless oil, yield 287 mg (85 %). C₁₄H₂₁NO (219.3). HR‐MS (APCI): m/z=220.1701 (calcd. 220.1696 for C₁₄H₂₂NO [M+H]⁺). ¹H NMR (600 MHz, CDCl₃): δ [ppm]=1.33 (m, 0.8H, 3‐H_ax ^#), 1.40–1.48 (m, 1H, 5‐H_ax ^#*), 1.48–1.58 (m, 2H, CH₂ CH₂OH), 1.58–1.69 (m, 1H, 4‐H^#, 3‐H_ax*), 1.75–1.83 (m, 1.6H, 3‐H_eq ^#, 5‐H_eq ^#), 1.83–1.91 (m, 0.2H, 5‐H_eq*), 2.01 (d, J=6.8 Hz, 3H, N‐CH₃ ^#*), 2.17 (t, J=11.9 Hz, 0.8H, 6‐H_ax ^#), 2.37 (t, J=11.7 Hz, 0.2H, 6‐H_ax*), 2.81 (m, 1H, 2‐H_ax ^#, 6‐H_eq*), 3.01–3.10 (m, 1H, 6‐H_eq ^#, 2‐H_ax*), 3.68 (dt, J=11.7/5.3 Hz, 2H, CH₂CH₂ OH^#*), 7.24 (tt, J=5.9/3.1 Hz, 1H, 4‐H_phenyl ^#*), 7.31 (d, J=5.2 Hz, 4H, 2‐H_phenyl ^#*, 3‐H_phenyl ^#*, 5‐H_phenyl ^#*, 6‐H_phenyl ^#*). The ratio of diastereomers cis‐16a:trans‐16a is 80 : 20. Signals of cis‐16a are marked with ^#, signals of trans‐16a with *. ¹³C NMR (151 MHz, CDCl₃): δ [ppm]=29.8*, 42.5 (C‐3), 32.7 (C‐5), 33.1, 34.2* (C‐4), 39.6, 39.9* (CH₂CH₂OH), 44.3, 44.5* (N‐CH₃), 51.9*, 57.3 (C‐6), 60.5, 61.5* (CH₂ CH₂OH), 65.5*, 68.5 (C‐2), 127.2, 127.56, 127.62, 128.6, 128.6 (5 C, C_phenyl), 144.5 (C‐1_phenyl). Signals of trans‐16a are marked with *. Purity (HPLC): 99.6 %, t_R=10.4 min and 10.5 min.

cis‐ and trans‐2‐(1‐Ethyl‐2‐phenylpiperidin‐4‐yl)ethan‐1‐ol (16b)

A solution of ester 15b (535 mg, 1.95 mmol) in THF (15 mL) was added dropwise to an ice‐cooled suspension of LiAlH₄ (158 mg, 4.17 mmol, 2.1 eq) in THF (15 mL). The reaction was stirred for 30 min at 0 °C. Ice cooling was removed and the reaction mixture was stirred for 2 h at rt. H₂O was added under ice cooling until the gas formation has stopped and the mixture was heated to reflux for 30 min. After cooling to rt, K‐Na‐tartrate (10 mL) was added to the mixture and the organic layer was separated and the aqueous layer was extracted with EtOAc (3×10 mL). The combined organic layers were dried (Na₂SO₄), filtered and concentrated in vacuo. The crude product was purified by automated flash column chromatography (Snap, 100 g, V=200 mL, CH₂Cl₂:EtOAc=3 : 1+1 % ethyldimethylamine, R_f=0.2). Colorless resin, yield 287 mg (85 %). C₁₅H₂₃NO (233.4). HR‐MS (APCI): m/z=234.1858 (calcd. 234.1852 for C₁₅H₂₄NO [M+H]⁺). ¹H NMR (600 MHz, CDCl₃): δ [ppm]=0.91 (t, J=7.1 Hz, 3H, N‐CH₂‐CH₃ ), 1.26–1.34 (m, 1H, 3‐H_ax), 1.38 (qd, J=12.5/3.9 Hz, 1H, 5‐H_ax), 1.43–1.58 (m, 2H, CH₂ CH₂OH), 1.62 (m, 1H, 4‐H), 1.73–1.84 (m, 2 H, 5‐H_eq, 3‐H_eq), 1.97 (dq, J=13.7/7.0 Hz, 1H, N‐CH₂ ‐CH₃), 2.10 (td, J=12.0/2.7 Hz, 0.85H, 6‐H_ax ^#), 2.25–2.35 (m, 0.15H, 6‐H_ax*), 2.54 (dq, J=13.7/7.4 Hz, 2H, N‐CH₂ ‐CH₃), 2.92–2.97 (m, 0.15 H, 6‐H_eq*), 3.05 (dd, J=11.2/2.8 Hz, 0.85H, 2‐H_ax ^#), 3.19 (dt, J=3.5/11.6 Hz, 0.85H, 6‐H_eq ^#), 3.27–3.32 (m, 0.15 H, 2‐H_ax*), 3.68 (t, J=6.7 Hz, 1.7H, CH₂CH₂ OH^#), 3.71 (t, J=6.7 Hz, 0.3H, CH₂CH₂ OH*), 7.22 (tt, J=6.3/2.2 Hz, 1H, 4‐H_phenyl), 7.27–7.35 (m, 4H, H‐H_phenyl, 3‐H_phenyl, 5‐H_phenyl, 6‐H_phenyl). The ratio of diastereomers cis‐16b:trans‐16b is 85 : 15. Signals of cis‐16b are marked with ^#, signals of trans‐16b with *.¹³C NMR (151 MHz, CDCl₃): δ [ppm]=11.3, 11.4* (N‐CH₂‐CH₃), 28.3*, 33.2 (C‐4), 29.8*, 32.7 (C‐5), 34.4*, 39.7 (CH₂CH₂OH), 40.4*, 43.3 (C‐3), 46.8*, 52.2 (1 C, C‐6), 48.9, 49.2* (N‐CH₂‐CH₃), 60.6, 61.5* (CH₂ CH₂OH), 63.0*, 68.5 (C‐2), 126.9, 127.0, 127.5, 127.6, 128.5, 128.6 (5 C, C_phenyl), 145.2 (C‐1_phenyl). Signals of trans‐16b are marked with *. Purity (HPLC): 91.1 %, t_R=11.3 min.

cis‐ and trans‐2‐(1‐Methyl‐2‐phenylpiperidin‐4‐yl)acetaldehyde (17a)

A solution of alcohol 16a (197 mg, 0.90 mmol, 1.0 eq) in CH₂Cl₂ (5 mL) was added to a solution of Dess‐Martin Periodinane (576 mg, 1.36 mmol, 1.5 eq) in CH₂Cl₂ (3 mL). The reaction mixture was stirred for 5 h at rt, before a solution of saturated NaHCO₃ and 10 % Na₂S₂O₃ (1 : 1, 8 mL) was added. The organic layer was separated and the aqueous layer was extracted with CH₂Cl₂ (3×10 mL). The combined organic layers were dried (Na₂SO₄), filtered and concentrated in vacuo. The crude product was purified by fc (d=2 cm, l=18 cm, V=90 mL, cyclohexane : ethyl acetate=3 : 2+1.5 % ethyldimethylamine, R_f=0.30 and 0.37 first and second diastereomer). Yellow resin, yield 122 mg (62 %). C₁₄H₁₉NO (217.3). HR‐MS (APCI): m/z=218.1533 (calcd. 218.1539 for C₁₄H₂₀NO [M+H]⁺). ¹H NMR (400 MHz, CDCl₃): δ [ppm]=1.33–1.68 (m, 2H, 3‐H_ax ^#*, 5‐H_ax ^#*), 1.80 (dt, J=13.6/3.0 Hz, 1.3H, 3‐H_eq ^#, 5‐H_eq ^#), 2.04 (s, 3H, N‐CH₃ ^#*), 1.89–2.18 (m, 1.35H, 4‐H^#, 3‐H_eq*, 5‐H_eq*), 2.19–2.37 (m, 1H, 6‐H_ax ^#*), 2.38 (ddd, J=6.4/3.3/1.8 Hz, 1.3H, CH₂CHO^#), 2.59 (m, 0.35H, 4‐H*), 2.62–2.73 (m, 0.7H, CH₂CHO*), 2.88 (t, J=12.3 Hz, 1H, 2‐H_ax ^#, 6‐H_eq*), 3.00 (d, J=11.7 Hz, 0.35H, 2‐H_ax*), 3.09 (d, J=11.7 Hz, 1H, 6‐H_eq ^#), 7.20–7.35 (m, 5H, H_phenyl), 9.77 (t, J=2.0 Hz, 0.65H, CHO^#), 9.79 (t, J=2.0 Hz, 0.35H, CHO*). The ratio of diastereomers cis‐17a:trans‐17a is 65 : 35. Signals of cis‐17a are marked with ^#, signals of trans‐17a with *. ¹³C NMR (101 MHz, CDCl₃): δ [ppm]=26.3*, 31.3 (C‐4) 29.6*, 32.4 (C‐5), 39.7*, 42.1 (C‐3), 44.1, 44.4* (N‐CH₃), 45.8*, 50.5 (CH₂CHO), 51.6*, 56.9 (C‐6), 65.4*, 70.5 (C‐2), 127.40, 127.47, 127.55, 127.6, 128.65, 128.70 (5 C, C_phenyl), 143.6 (C‐1_phenyl), 201.9, 202.1* (CHO). Signals of trans‐17a are marked with *. Purity (HPLC) Rkt 180: 98.3 %, t_R=10.3 min and 10.9 min.

cis‐ and trans‐2‐(1‐Ethyl‐2‐phenylpiperidin‐4‐yl)acetaldehyde (17b)

Dess‐Martin Periodinane (409 mg, 0.96 mmol, 1.5 eq) was added to a solution of alcohol 16b (149 mg, 0.64 mmol) in CH₂Cl₂ (4 mL). The reaction mixture was stirred for 1.5 h at rt. Then 10 % Na₂S₂O₃ (4 mL) and a saturated solution of NaHCO₃ (4 mL) was added and the mixture was stirred for 10 min. The organic layer was separated and the aqueous layer was extracted with ethyl acetate (3×10 mL). The combined organic layers were dried (Na₂SO₄) and concentrated in vacuo. The crude product was purified by automated fc (Snap, 25 g, V=100 mL, CH₂Cl₂:EtOAc=4 : 1+2 % ethyldimethylamine, R_f=0.32). Orange resin, yield 137 mg (93 %). C₁₅H₂₁NO (231.3). HR‐MS (APCI): m/z=232.1685 (calcd. 232.1695 for C₁₅H₂₂NO [M+H]⁺). ¹H NMR (600 MHz, CDCl₃): δ [ppm]=0.91 (m, 3H, N‐CH₂‐CH₃ ), 1.34 (q, J=12.2 Hz, 0.85H, 3‐H_ax ^#), 1.39–1.65 (m, 1.15H, 5‐H_ax ^#*, 3‐H_ax*), 1.73–1.84 (m, 1.7H, 3‐H_eq ^#, 5‐H_eq ^#), 1.83–1.94 (m, 0.3H, 3‐H_eq*, 5‐H_eq*), 1.98 (tt, J=13.7/6.7 Hz, 2H, N‐CH₂ ‐CH₃ ^#*), 2.01–2.10 (m, 1H, 4‐H^#*), 2.11–2.18 (m, 0.85H, 6‐H_ax ^#), 2.25–2.43 (m, 2.15H, CH₂CHO^#*, 6‐H_ax*), 2.49–2.57 (m, 1H, N‐CH₂ ‐CH₃ ^#*), 2.97 (dt, J=12.5/4.0 Hz, 0.15H, 6‐H_eq*), 3.07–3.12 (m, 0.85H, 2‐H_ax ^#), 3.19 (m, 0.85H, 6‐H_eq ^#) 3.22–3.28 (m, 0.15H, 2‐H_ax*), 7.22 (m, 1H, 4‐H_phenyl ^#*), 7.26–7.34 (m, 4H, H_phenyl ^#*), 9.73–9.76 (m, 0.85H, CHO^#), 9.73–9.76 (m, 0.15H, CHO*). The ratio of diastereomers cis‐17b:trans‐17b is 85 : 15. Signals of cis‐17b are marked with ^#, signals of trans‐17b with *.¹³C NMR (151 MHz, CDCl₃): δ [ppm]=11.2, 11.4* (N‐CH₂‐CH₃ ), 29.7*, 32.5 (C‐5), 31.5 (C‐4), 40.1*, 43.0 (C‐3), 46.0*, 48.8 (N‐CH₂ ‐CH₃), 46.6*, 51.9 (C‐6), 50.7 (CH₂CHO), 62.9*, 68.2 (C‐2), 127.0, 127.1, 127.5, 127.6, 128.5 (5 C, C_phenyl), 144.7 (C‐1_phenyl), 202.1, 202.3* (CHO). Signals of trans‐17b are marked with *. Purity (HPLC): 94.0 %, t_R=11.5 min.

cis‐ and trans‐N‐Benzyl‐2‐(1‐methyl‐2‐phenyl‐piperidin‐ 4‐yl)ethan‐1‐amine (18a)

Benzylamine (49.3 mg, 0.46 mmol, 2.5 eq) and aldehyde 17a (20.6 mg, 0.09, 0.5 eq) were solved in CH₂Cl₂ (3 mL) and the mixture was stirred for 1 h at rt. Then, additional 0.5 equivalents of aldehyde 17a dissolved in CH₂Cl₂ were added and the mixture was stirred for 1 h, before NaBH(OAc)₃ (97.5 mg, 0.46 mmol, 2.5 eq) was added to the solution. The reaction mixture was stirred over night at rt. Then a saturated solution of NaHCO₃ (6 mL) was added and the aqueous layer was extracted with ethyl acetate (3×10 mL). The combined organic layers were dried (Na₂SO₄), filtered and concentrated in vacuo. The crude product was purified twice by fc (1. d=1 cm, l=20 cm, V=12 mL, cyclohexane : ethyl acetate=1 : 1+1.5 % ethyldimethylamine; and 2. d=1 cm, l=20 cm, V=30 mL, cyclohexane : ethyl acetate=1 : 1+1 % ethyldimethylamine, R_f=0.26). Yellow resin, yield 18.2 mg (31 %). C₂₁H₂₈N₂ (308.5). MS (APCI): m/z=309.2337 (calcd. 309.2325 for C₂₁H₂₉N₂ [M+H]⁺). ¹H NMR (600 MHz, CDCl₃): δ [ppm]=1.24–1.36 (m, 1H, 3‐H_ax), 1.37–1.61 (m, 4H, 5‐H, 4‐H, CH₂ CH₂N), 1.73 (d, broad, J=13.3 Hz, 2H, 3‐H, 5‐H), 2.00 (m, 3H, N‐CH₃), 2.14 (td, J=12.2/2.6 Hz, 0.85H, 6‐H_ax ^#), 3.37 (t, J=11.0 Hz, 0.15H, 6H_ax*), 2.60–2.70 (m, 2H, CH₂ CH₂ N), 2.79 (d, broad, J=11.3 Hz, 1H, 2‐H_ax ^#, 6‐H_eq*), 3.04 (dt, J=12.0/3.2 Hz, 1H, 6‐H_eq ^#, 2‐H_ax*), 3.77 (s, 1.7H, NCH₂ ‐bnz^#), 3.81 (s, 0.3H, NCH₂ ‐bnz*), 7.20–7.25 (m, 2H, 4‐H_phenyl), 7.28–7.35 (m, 8H, H_phenyl). The ratio of diastereomers cis‐18a:trans‐18a is 85 : 15. Signals of cis‐18a are marked with ^#, signals of trans‐18a with *. ¹³C NMR (151 MHz, CDCl₃): δ [ppm]=29.7*, 32.8 (C‐5), 34.6, (C‐4), 37.1 (CH₂ CH₂N), 39.9*, 42.6 (C‐3), 44.3, 44.5* (N‐CH₃), 46.9, 47.9* (CH₂‐CH₂ ‐N), 54.3 (NCH₂‐bnz), 51.9*, 57.3 (C‐6), 65.4*, 70.8 (C‐2), 127.1, 127.2, 127.6, 128.26, 128.28, 128.5, 128.6, 128.9 (8 C, C_benzyl, C_phenyl), 140.5 (C‐1_phenyl), 144.6 (C‐1_benzyl). Signals of trans‐18a are marked with *. Purity (HPLC): 95.6 %, t_R=21.1 min.

cis‐ and trans‐N‐Benzyl‐2‐(1‐ethyl‐2‐phenyl‐piperidin‐ 4‐yl)ethan‐1‐amine (18b)

A solution of aldehyde 17b (49.3 mg, 0.21 mmol, 1.0 eq) in CH₂Cl₂ (3 mL) was added dropwise over 30 min to a solution of benzylamine (46.0 mg, 0.43 mmol, 2.0 eq) in CH₂Cl₂ (1 mL) and the reaction mixture was stirred for 3 h at rt, before NaBH(OAc)₃ (91.1 mg, 0.43 mmol, 2.0 eq) was added to the solution. The reaction mixture was stirred over night at rt. Then a saturated solution of NaHCO₃ (6 mL) was added to the solution and the aqueous layer was extracted with CH₂Cl₂ (3×10 mL). The combined organic layers were dried (Na₂SO₄), filtered and concentrated in vacuo. The crude product was purified three times by fc. First by automated fc (Snap, 10 g, V=50 mL, CH₂Cl₂:EtOAc=7 : 3+2 % ethyldimethylamine). Second (d=1 cm, l=22 cm, V=28 mL, cyclohexane : ethyl acetate=7 : 3+1 % ethyldimethylamine) and third (d=1 cm, l=22 cm, V=7 mL, cyclohexane : ethyl acetate=2 : 1+1 % ethyldimethylamine, R_f=0.13). Yellow resin, yield 4.8 mg (7 %). C₂₂H₃₀N₂ (322.5). HR‐MS (APCI): m/z=323.2476 (calcd. 323.2482 for C₂₂H₃₁N₂ [M+H]⁺). ¹H NMR (400 MHz, CDCl₃): δ [ppm]=0.84–0.97 (m, 3H, N‐CH₂‐CH₃ ), 1.21–1.41 (m, 2H, 3‐H_ax, 5‐H_ax), 1.40–1.62 (m, 3H, 4‐H, CH₂ ‐CH₂‐N), 1.67–1.80 (m, 2H, 3‐H_eq, 5‐H_eq), 1.93–2.04 (m, 2H, N‐CH₂ ‐CH₃), 2.09 (td, J=11.7/2.0 Hz, 0.9H, 6‐H_ax ^#), 2.35 (t, J=12.1 Hz, 0.1H, 6‐H_ax*), 2.48–2.61 (m, 2H, N‐CH₂ ‐CH₃), 2.61–2.71 (m, 2H, CH₂‐CH₂ ‐N), 2.89–2.97 (m, 0.1H, 6‐H_eq*), 3.04 (dd, J=11.2/2.6 Hz, 0.9H, 2‐H_ax ^#), 3.19 (dt, J=11.4/3.5 Hz, 0.9H, 6‐H_eq ^#), 3.28–3.36 (m, 0.1H, 2‐H*), 3.77 (s, 1.8H, CH₂‐bnz^#), 3.81 (s, 0.2H, CH₂‐bnz*), 7.20–7.36 (m, 10H, H_arom.). The ratio of diastereomers cis‐18b:trans‐18b is 90 : 10. Where signals could be distinguished, signals of cis‐18b are marked with ^#, signals of trans‐18b with *.¹³C NMR (151 MHz, CDCl₃): δ [ppm]=32.7 (C‐5), 34.7 (C‐4), 37.2 (CH₂ ‐CH₂‐N), 43.4 (C‐3), 47.0 (CH₂‐CH₂ ‐N), 49.0 (N‐CH₂ ‐CH₃), 52.2 (C‐6), 54.3 (CH₂‐bnz), 68.6 (C‐2), 126.85, 126.95, 126.98 127.1, 127.56, 127.65, 128.2, 128.3, 128.5, 128.6 (10 C, C_phenyl, C_benzyl), 140.7 (C‐1_benzyl), 145.3 (C‐1_phenyl). Purity (HPLC): 96.1 %, t_R=12.7 min and 11.5 min first and second diastereomer.

cis‐N‐(2‐(1‐Methyl‐2‐phenylpiperidin‐4‐yl)‐ethyl)‐3‐ phenylpropan‐1‐amine (19a)

3‐Phenylpropan‐1‐amine (62.2 mg, 0.46 mmol, 2.5 eq) and aldehyde 17 a (20.6 mg, 0.09, 0.5 eq) were dissolved in CH₂Cl₂ (3 mL) and the mixture was stirred for 1 h at rt. Then, additional 0.5 equivalents of aldehyde 17a, dissolved in CH₂Cl₂ were added and the reaction mixture was stirred for 1 h, before NaBH(OAc)₃ (97.5 mg, 0.46 mmol, 2.5 eq) was added to the solution. The reaction mixture was stirred over night at rt. Then a saturated solution of NaHCO₃ (6 mL) was added and the aqueous layer was extracted with ethyl acetate (3×10 mL). The combined organic layers were dried (Na₂SO₄), filtered and concentrated in vacuo. The crude product was purified twice by fc (d=1 cm, l=20 cm, V=75 mL, cyclohexane : ethyl acetate=1 : 1+1.5 % ethyldimethylamine) and (d=1 cm, l=22 cm, V=40 mL, cyclohexane : ethyl acetate=1 : 1+1.5 % ethyldimethylamine, R_f=0.09). The first diastereomer and dialkylated by‐product eluted together from the column. Just one diastereomer was isolated. Colorless resin, yield 7.1 mg (11 %). C₂₃H₃₂N₂ (336.5). HR‐ MS (APCI): m/z=337.2637 (calcd. 337.2638 for C₂₃H₃₃N₂ [M+H]⁺). ¹H NMR (600 MHz, CDCl₃): δ [ppm]=1.24–1.35 (m, 1H, 3‐H_ax), 1.37–1.49 (m, 4H, 5‐H_ax, 4‐H, CH₂ CH₂‐N), 1.70–1.76 (m, 2H, 3‐H_eq, 5‐H_eq), 1.82 (dq, J=9.1/7.1 Hz, 2H, N‐CH₂ CH₂ CH₂‐ph), 2.00 (s, 3H, N‐CH₃), 2.14 (td, J=12.0/2.4 Hz, 1H, 6‐H_ax), 2.58–2.69 (m, 6H, CH₂ CH₂ ‐N, N‐CH₂ CH₂ CH₂ ‐phenyl, CH₂‐ph), 2.78 (dd, J=11.3/2.6 Hz, 1H, 2‐H_ax), 3.04 (ddd, J=11.6/3.9/2.7 Hz, 1H, 6‐H_eq), 7.15–7.36 (m, 10H, H_arom.). ¹³C NMR (151 MHz, CDCl₃): δ [ppm]=31.7 (N‐CH₂ CH₂ CH₂‐phenyl), 32.8 (C‐5), 33.8 (CH₂‐ph), 34.6 (C‐4), 37.0 (CH₂ CH₂‐N), 42.7 (C‐3), 44.3 (N‐CH₃), 47.4 (CH₂ CH₂ ‐N), 49.7 (N‐CH₂ CH₂CH₂‐phenyl), 57.3 (C‐6), 70.8 (C‐2), 125.9, 127.2, 127.5, 128.8, 128.50, 128.57 (10 C, C_arom.), 142.2 (C‐1_benzyl), 144.7 (C‐1_phenyl). Purity (HPLC): 94.2 %, t_R=14.2 min.

cis‐N‐(2‐(1‐Ethyl‐2‐phenylpiperidin‐4‐yl)ethyl)‐3‐phenylpropan‐1‐amine (19b)

A solution of aldehyde 17b (50.0 mg, 0.21 mmol) in CH₂Cl₂ (3 mL) was added dropwise over 30 min to a solution of 3‐phenylpropan‐1‐amine (58.1 mg, 0.43 mmol, 2.0 eq)) in CH₂Cl₂ (1 mL) and the mixture was stirred for 3 h at rt, before NaBH(OAc)₃ (91.1 mg, 0.43 mmol, 2.0 eq) was added to the solution. The reaction mixture was stirred over night at rt. Then a saturated solution of NaHCO₃ (6 mL) was added and the aqueous layer was extracted with CH₂Cl₂ (3×10 mL). The combined organic layers were dried (Na₂SO₄), filtered and concentrated in vacuo. The crude product was purified by fc (d=1 cm, l=22 cm, V=35 mL, CH₂Cl₂ : ethyl acetate=7 : 3+1 % ethyldimethylamine, R_f=0.15). Yellow resin, yield 46.9 mg (62 %). C₂₄H₃₄N₂ (350.6). HR‐MS (APCI): m/z=351.2817 (calcd. 351.2795 for C₂₄H₃₅N₂ [M+H]⁺). ¹H NMR (400 MHz, CDCl₃): δ [ppm]=0.89 (t, J=7.1 Hz, 3H, N‐CH₂‐CH₃ ), 1.20–1.41 (m, 2H, 3‐H_ax, 5‐H_ax), 1.45 (m, 3H, 4‐H, CH₂ CH₂‐NH,), 1.73 (tt, J=12.9/2.8 Hz, 2H, 3‐H_eq, 5‐H_eq), 1.82 (ddd, J=15.0/8.6/6.8 Hz, 2H, HN‐CH₂ CH₂ CH₂), 1.95 (dd, J=12.9/6.9 Hz, 2H, N‐CH₂ ‐CH₃), 2.00–2.15 (m, 1H, 6‐H_ax), 2.52 (dq, J=12.8/7.4 Hz, 2H, N‐CH₂ ‐CH₃), 2.57–2.67 (m, 6H, HN‐CH₂ CH₂ CH₂ , CH₂ CH₂ ‐NH), 3.02 (dd, J=11.2/2.7 Hz, 1H, 2‐H_ax), 3.17 (dt, J=11.6/3.4 Hz, 1H, 6‐H_eq), 7.12–7.33 (m, 10H, H_phenyl). ¹³C NMR (151 MHz, CDCl₃): δ [ppm]=δ [ppm]=11.3 (CH₃), 31.5 (HN‐CH₂ CH₂ CH₂), 32.7 (C‐5), 33.8 (HN‐CH₂CH ₂ CH₂ , 34.7 (C‐4), 36.9 (CH₂ ‐CH₂‐NH), 43.3 (C‐3), 47.4 (CH₂ CH₂ ‐NH), 49.0 (N‐CH₂ CH₃), 49.6 (HN‐CH₂ CH ₂ CH₂), 52.2 (C‐6), 68.5 (C‐2), 125.9, 126.9, 127.5, 128.45, 128.48, 128.50 (10 C, C_phenyl), 142.2 (C‐1_propylphenyl), 145.2 (C‐1_phenyl). Purity (HPLC): 98.3 %, t_R=14.9 min.

cis‐ and trans‐N‐(Cyclohexylmethyl)‐2‐(1‐methyl‐ 2‐phenylpiperidin‐4‐yl)ethan‐1‐amine (20a)

A solution of aldehyde 17a (51.3 mg, 0.24 mmol) and cyclohexylmethylamine (65.0 mg, 0.57 mmol, 2.5 eq) in CH₂Cl₂ (4 mL).was stirred for 10 min, before NaBH(OAc)₃ (97.5 mg, 0.46 mmol, 2.0 eq) was added. After 2 h, a solution of Na₂S₂O₃ (10 %) and sat. NaHCO₃ (1 : 1, 5 mL) was added. and the aqueous layer was extracted CH₂Cl₂ (4×10 mL). The combined organic layers were dried (Na₂SO₄), filtered and concentrated in vacuo. The crude product was purified by fc (d=1 cm, l=20 cm, V=28 mL, cyclohexane : ethyl acetate=3 : 2+1,5 % ethyldimethylamine, R_f=0.11). Colorless resin, yield 48.8 mg (66 %). C₂₁H₃₄N₂ (314.5). HR‐MS (APCI): m/z=315.2811 (calcd. 315.2795 for C₂₁H₃₅N₂ [M+H]⁺). ¹H NMR (600 MHz, CDCl₃): δ [ppm]=0.88 (qd, J=14.0/13.1/4.1 Hz, 2H, cyclohexane), 1.09–1.27 (m, 4.17H, 3‐H_ax*, 4× cyclohexane), 1.31 (dt, J=13.3/11.2 Hz, 0.83H, 3‐H_ax ^#), 1.37–1.62 (m, 5H, 4‐H, 5‐H_ax, CH₂ ‐CH₂‐N, 1× cyclohexane), 1.62–1.92 (m, 5H, 3‐H_eq, 5‐H_eq, 3× cyclohexane), 2.00 (d, J=5.7 Hz, 3H, N‐CH₃), 2.14 (td, J=12.1/2.6 Hz, 1H, 6‐H_ax ^#), 2.32–2.39 (m, 0.17H, 6‐H_ax*), 2.42 (dd, J=6.8/2.3 Hz, 1.66H, N‐CH₂‐cyclohexane^#), 2.46 (dd, J=6.8/2.3 Hz, 0.34H, N‐CH₂‐cyclohexane*) 2.56–2.65 (m, 2H, CH₂‐CH₂ ‐N), 2.78 (dd, J=11.3/2.7 Hz, 1H, 2‐H_ax ^#, 6‐H*), 3.04 (ddd, J=11.7/3.9/2.7 Hz, 1H, 6‐H_eq ^#, 2‐H_ax*), 7.23 (dt, J=8.6/4.5 Hz, 1H, 4‐H_phenyl), 7.30 (d, J=4.5 Hz, 4H, H_phenyl). The ratio of diastereomers cis‐20a:trans‐20a is 83 : 17. Signals of cis‐20a are marked with ^#, signals of trans‐20a with *. ¹³C NMR (151 MHz, CDCl3): δ [ppm]=28.7, 28.7*, 29.3, 32.4*, 34.1, 34.2* (5 C, cyclohexane), 35.3 (C‐5), 37.2 (C‐4), 39.4 (1 C, CH₂ CH₂‐N), 40.4, 42.5* (C‐1_cyclohexane), 45.2 (C‐3), 46.9, 47.1* (N‐CH₃), 50.1, 51.2* (CH₂‐CH₂‐N), 54.4*, 59.8 (C‐6), 59.5 (N‐CH₂‐cyclohexane), 67.9*, 73.3 (C‐2), 129.6*,129.7 (C‐4_phenyl), 130.0, 130.1*, 131.0, 131.1* (4 C, C_phenyl), 147.3 (C‐1_phenyl). Signals of trans‐20a are marked with *. Purity (HPLC): 96.6 %, t_R=13.3 min.

cis‐ and trans‐N‐Benzyl‐N‐methyl‐2‐(1‐methyl‐ 2‐phenylpiperidin‐4‐yl)ethan‐1‐amine (21a)

A mixture of aldehyde 17a (40.0 mg, 0.18 mmol, 1.0 eq), N‐methylbenzylamine (36.9 μL, 0.27 mmol, 1.5 eq) and NaBH(OAc)₃ (57.2 mg, 0.27 mmol, 1.5 eq) in CH₂Cl₂ (3 mL) was stirred for 3 h at rt. Then a solution of Na₂S₂O₃ (10 %) and sat. NaHCO₃ (1 : 1, 4 mL) was added to the reaction mixture and the aqueous layer was extracted with CH₂Cl₂ (4×6 mL). The combined organic layers were dried (Na₂SO₄), filtered and concentrated in vacuo. The crude product was purified twice by fc. First column (d=1 cm, l=25 cm, V=20 mL, cyclohexane : ethyl acetate=2 : 1+1.5 % ethyldimethylamine). Second column (d=1 cm, l=30 cm, V=20 mL, cyclohexane : ethyl acetate=3 : 1+2 % ethyldimethylamine, R_f=0.39). Colorless oil, yield 28.2 mg (49 %). C₂₂H₃₀N₂ (322.5). HR‐MS (APCI): m/z=323.2490 (calcd. 323.2482 for C₂₂H₃₁N₂ [M+H]⁺). ¹H NMR (600 MHz, CDCl₃): δ [ppm]=1.31 (d, J=12.2 Hz, 1H, 3‐H_ax), 1.38–1.64 (m, 2.65H, 5‐H_ax, 4‐H*, CH₂ CH₂‐N), 1.67–1.76 (m, 2.7H, 3‐H_eq, 5‐H_eq, CH₂ CH₂‐N*), 1.81–1.94 (m, 0.4H, 4‐H*), 2.03 (m, 3H, piperidine‐N‐CH₃), 2.17 (m, 3H, 6‐H_ax, N‐CH₃ ^#), 2.22 (s, 1H, N‐CH₃*), 2.33–2.46 (m, 2H, CH₂‐CH₂ ‐N), 2.78–2.85 (m, 1H, 2‐H, 6‐H_ax*), 3.00–3.10 (m, 1H, 6‐H_eq, 2‐H_ax*), 3.42–3.49 (m, 0.8H, N‐CH₂ ‐pheny*), 3.48–3.55 (m, 1.2H, N‐CH₂ ‐phenyl^#), 7.24 (tdd, J=8.7/4.0/2.1 Hz, 2H, 4‐H_phenyl), 7.27–7.36 (m, 8H, H_phenyl). The ratio of diastereomers cis‐21a:trans‐21a is 60 : 40. Signals of cis‐21a are marked with ^#, signals of trans‐21a with *.¹³C NMR (151 MHz, CDCl₃): δ [ppm]=29.8 (CH₂ CH₂‐N), 32.7 (C‐5), 34.6 (C‐4), 39.8*, 42.6 (C‐3), 42.4, 42.5* (N‐CH₃), 44.3, 44.4* (N‐CH₃‐piperidine), 52.0*, 57.3 (C‐6), 54.8, 55.9* (CH₂ CH₂ ‐N), 62.6 (N‐CH₂ ‐phenyl), 65.5*, 70.9 (C‐2), 127.07, 127.11, 127.2, 127.58, 127.65, 128.32, 128.32, 128.37, 128.6, 129.2 (10 C, C_phenyl), 139.1 (2 C, C‐1_phenyl). Signals of trans‐21a are marked with *.Purity (HPLC): 96.9 %, t_R=12.6 and 13.7 min.

cis‐ and trans‐1‐(2‐(1‐Methyl‐2‐phenyl‐ piperidin‐4‐yl)ethyl)‐4‐phenylpiperazine (22a)

NaBH(OAc)₃ (72.0 mg, 0.34 mmol, 1.5 eq) was added to a stirred solution of aldehyde 17a (50.0 mg, 0.23 mmol, 1.0 eq) and 1‐phenylpiperazine (55.9 mg, 0.34 mmol, 1.5 eq) in CH₂Cl₂ (3 mL). The reaction mixture was stirred for 3 h, then quenched with a solution of Na₂S₂O₃ (10 %) and sat. NaHCO₃ (1 : 1, 5 mL). The organic layer was separated and the aqueous layer was extracted with ethyl acetate (4×10 mL). The combined organic layers wren dried (Na₂SO₄), filtered and concentrated in vacuo. The crude product was purified by fc (d=1 cm, l=22 cm, V=12 mL, cyclohexane : ethyl acetate=2 : 1+1,5 % ethyldimethylamine, R_f=0.34). Colorless resin, yield 57.4 mg (69 %). C₂₄H₃₃N₃ (363.6). HR‐MS (APCI): m/z=364.2727 (calcd. 364.2747 for C₂₄H₃₄N₃ [M+H]⁺). ¹H NMR (600 MHz, CDCl₃): δ [ppm]=1.32–1.37 (m, 0.75H, 3‐H_ax ^#), 1.44–1.55 (m, 3.75H, 4‐H^#, 5‐H_ax, CH₂ CH₂‐N), 1.62 (dt, J=13.9/2.5 Hz, 0.25H, 3‐H_ax*), 1.77 (ddt, J=15.9/12.7/3.9 Hz, 1.75H, 5‐H, 3‐H_eq ^#), 1.90 (s, 0.5 H, 4‐H*, 3‐H_eq*), 2.00–2.06 (m, 3H, N‐CH₃), 2.13–2.20 (m, 1H, 6‐H_ax), 2.36–2.46 (m, 2H, CH₂ CH₂ ‐N), 2.55–2.60 (m, 3H, CH₂‐N‐(CH₂ ‐CH₂)₂‐N‐Ph^#), 2.60–2.68 (m, 1H, CH₂‐N‐(CH₂ ‐CH₂)₂‐N−Ph*), 2.78–2.84 (m, 1H, 2‐H_ax), 3.04–3.10 (m, 1H, 6‐H_eq), 3.17–3.22 (m, 3H, CH₂‐N‐(CH₂‐CH₂ )₂‐N‐Ph^#), 3.21–3.26 (m, 1H, CH₂‐N‐(CH₂‐CH₂ )₂‐N−Ph*), 6.85 (qt, J=7.3/1.1 Hz, 1H, 4‐H_{phenylpiperazine}), 6.90–6.97 (m, 2H, 2‐H_{phenylpiperazine}, 6‐H_{phenylpiperazine}) 7.22–7.36 (m, 7H, 3‐H_{phenylpiperazine}, 5‐H_{phenylpiperazine}, H_phenyl). The ratio of diastereomers cis‐22a:trans‐22a is 75 : 25. Signals of cis‐22a are marked with ^#, signals of trans‐22a with *. ¹³C NMR (151 MHz, CDCl3): δ [ppm]=32.8 (C‐5), 33.8 (CH₂CH₂‐N), 35.1 (C‐4), 42.6 (C‐3), 44.3 (N‐CH₃), 49.3, 49.3* (CH₂‐N‐(CH₂‐CH₂)₂‐N−Ph), 53.5, 53.6* (2 C, CH₂‐N‐(CH₂‐CH₂)₂‐N−Ph), 56.4, 57.5* (CH₂‐CH₂ ‐N), 57.3 (C‐6), 65.5*, 70.8 (C‐2), 116.1, 116.2* (2 C, C‐2_{phenylpiperazine}, C‐6 _{phenylpiperazine}), 119.77, 119.83 (C‐4_{phenylpiperazine}), 127.2, 127.56, 127.65, 128.6, 128.7, 129.22, 129.25 (7 C, C‐3_{phenylpiperazine}, C‐5 _{phenylpiperazine}, C_phenyl), 144.6 (C‐1_phenyl), 151.5 (C‐1_{phenylpiperazine}). Signals of trans‐22a are marked with *.Purity (HPLC): 99.7 %, t_R=13.8 min.

Receptor binding studies

Receptor binding studies were performed as previously described.[ ⁴⁵ , ⁴⁶ , ⁴⁷ ] Details are given in the Supporting Information.

Molecular dynamics simulations

All simulations were carried out using the Pmemd modules of Amber 20, ^[51] running on our own CPU/GPU calculation cluster. See Supporting Information for full computational details.

Analysis of the effects of σ₁ receptor ligand 4 a on proliferation and morphology of the human tumor cell line A427

The effects of the piperidine derivative 4a on the growth and morphology of human tumor cell lines A 427 were determined with IncuCyte® S3 Live Cell Analysis System (Essen BioScience, Ltd., Royston, Hertfordshire, UK). In particular the confluence and IC ₅₀ values were determined. Details are given in the Supporting Information.

DU145 cell growth inhibition

Details are given in the Supporting Information.

Supporting Information

Supporting Information contains the purity data of all test compounds, details of the receptor biding studies and computational details. Experimental details of the effects on A427 and DU145 tumor cell lines are given. Finally, ¹H and ¹³C NMR spectra are displayed.

Conflict of interest

The authors declare no conflict of interest.

Supporting information

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

Supporting Information

C. Holtschulte, F. Börgel, S. Westphälinger, D. Schepmann, G. Civenni, E. Laurini, D. Marson, C. V. Catapano, S. Pricl, B. Wünsch, ChemMedChem 2022, 17, e202100735.

Data Availability Statement

The data that support the findings of this study are available in the supplementary material of this article.