Sustainable Synthesis of a Potent and Selective 5-HT7 Receptor Antagonist Using a Mechanochemical Approach

A mechanochemical procedure was developed to obtain PZ-1361, a potent and selective 5-HT7 receptor antagonist, with antidepressant properties in rodents. The elaborated protocol offered several advantages over classical batch synthesis, including improvement of the overall yield (from 34% to 64%), reduction of reaction time (from 60 to 5.5 h), limitation of the use of toxic solvents, and the formation of byproducts. This approach represents a rare example of the synthesis of biologically active compounds exclusively performed using mechanochemical reactions.

S ynthetic organic chemistry represents a key component of many drug discovery programs. In the last decades, different techniques have been developed, for example, microwave-assisted organic chemistry and flow chemistry, which might reduce the time required to generate compound libraries for biological screening or might ensure more efficient production of active pharmaceutical ingredients (APIs). 1−4 Recently, mechanochemical synthesis has also been recognized as an innovative methodology, 5 with a wide range of practical applications in both academic and industrial research. In particular, mechanochemistry has been used to produce various families of compounds. 6−10 The primary driving force underlying the rediscovery of mechanochemistry is green chemistry, 11−13 in particular, the need of chemical and pharmaceutical industries for the development of more sustainable synthetic protocols characterized by the energy efficiency of chemical transformations and reduction of solvent use. The use of such approaches offers additional advantages of mechanosynthesis over classical organic chemistry, in terms of excellent selectivity and the possibility to perform previously unknown reactions. 14−16 Interestingly, an increasing number of mechanochemical procedures for generating pharmaceutically relevant fragments and functionalities have been reported thus far. 17−19 This novel mechanochemical application led to coining the term "medicinal mechanochemistry." 20,21 We have recently developed a novel class of a potent and selective 5-HT 7 receptor (5-HT 7 R) antagonist, namely, an arylsulfonamide derivative of (aryloxy)alkyl alicyclic amine, and identified several lead structures that exhibit significant in vivo antidepressant and pro-cognitive properties in rodents ( Figure 1). 22 −26 The classical "in batch" synthetic pathway of this class of derivatives consists of four steps involving the alkylation of commercially available phenols in biphasic conditions, nucleophilic substitution of Boc-protected alicyclic amines, removal of the protecting group, and sulfonylation of the resulting primary amine in an alkaline environment (Scheme 1). The critical step of the entire process is the alkylation of phenol, as this reaction should be performed in the presence of a large excess of halogeno-alkanes (from 3 to 6 equiv) to avoid unwanted dimerization or opening of the epoxide ring. Additionally, apart from the deprotection of amine function, column chromatography purification is required in all of the remaining steps together with the use of a large amount of organic solvents (in particular, the highly toxic dichloromethane). 27,28 To overcome these issues and simultaneously extend the concept of medicinal mechanochemistry, we adapted the synthetic pathway by using a mechanochemical approach for the synthesis of the potent and selective 5-HT 7 R antagonist PZ-1361. 29 To demonstrate the versatility of this method, we subsequently increased the diversity of building blocks by conducting experiments using 2-substituted phenols, different central amine cores (e.g., piperazine, 3-aminotropane, 3-aminopyrrolidine), and differently substituted arylsulfonyl chlorides. This allowed proposing mechanochemistry as a promising synthetic strategy in medicinal chemistry, which would enable the preparation of lead compounds for preclinical development in a more sustainable and greener manner. 30 The optimization of the synthetic pathway started with the alkylation of commercially available 2-phenylphenol with racemic epichlorohydrin (1 equiv). The reaction was initially performed in a 10 mL PTFE jar with a 1 cm diameter stainless steel ball by using a vibratory ball mill (vbm) operated at 30 Hz. A thorough study of the different parameters was performed and is summarized in Table 1 (for more  information, see Tables S3−5).
Next, to investigate the kinetics of the alkylation, the experiment was performed using potassium carbonate and different time periods ranging from 40 to 220 min. A reaction time of approximately 2 h was sufficient to obtain around 40% conversion (Table 1, entry 5), while a prolonged milling time slightly accelerated the formation of desired product 1a without reaching full conversion (Table 1, entries 6 and 7). Increasing the milling time promoted the formation of the unwanted side products, 2a, resulting from nucleophilic addiction of 2-phenylphenol to the oxirane derivative 1a. Of note, a similar epoxide ring opening with carboxylic acids was previously reported. 31 On the basis of the assumption that more energetic collisions might be achieved by increasing the diameter of the milling ball, the use of a 1.5 cm diameter stainless steel ball significantly improved the formation of the monoalkylated product 1a, while restricting the formation of 2a to 2.5% (Table 1, entry 8). Additionally, the use of a slight excess of epichlorohydrin (Table 1, entry 9) enabled to achieve a 90% conversion of 2-phenylphenol into 1a (88% isolated yield) without increasing the formation of 2a. After identifying the optimal reaction conditions for alkylation on a small scale, the influence of different substituents at the phenol was subsequently determined using a 35 mL PTFE jar. The choice of substituent (e.g., 2-isopropyl and 2-iodo) was based on the biological data for this group of derivatives, wherein it was confirmed that only compounds bearing a sterically hindered substituent in 2-position preferentially bind to 5-HT 7 R. 23,24 All of the tested substituted phenols showed higher conversion rates under milling than the unsubstituted one, with the following reactivity rank order: 2-Ph ≈ 2-I > 2-iPr > H ( Table  2). These results are in line with yields and purities obtained from the classical organic synthesis pathway using biphasic conditions (for more information, see Supporting Information).
Because the conversion rate was slightly lower than that obtained in the 10 mL milling jar ( Table 2, entry 1), ethyl acetate and isopropanol as nontoxic liquid assistants were evaluated to enhance the overall mixing and reaction kinetics. 32−36 The results showed that, regardless of the nature of the phenol tested, liquid-assisted grinding (LAG) enabled to increase the conversions into 1a−d up to 11%, while limiting the formation of side products 2a−d, with isopropanol being the best assistant. A simple basic extraction allowed to remove Reaction conditions: 2-phenylphenol (1 equiv), epichlorohydrin (1 equiv), vbm 30 Hz, 10 mL PTFE jar, ϕ ball = 1 cm, total mass of reagents = 100 mg. b Conversions of 2-phenylphenol into 1a and 2a were determined by UPLC/MS analysis. c The ball used was 1.5 cm in diameter. d 1.2 equiv of epichlorohydrin was used. unreacted substrates and side products and afforded intermediates 1a−d in high purity and yields (65−85%). As a comparison, the same reactions performed in the solution for 48 h provided a higher amount of side products (20−35%), which required column chromatography separation. The next step involved the alkylation of Boc-4-aminopiperidine with the [1,1′-biphenyl] oxirane derivative 1a in a 35 mL PTFE jar using vbm at 30 Hz (Scheme 2). This reaction was originally performed in refluxing ethanol for 4 h, with an excess of reagents and required purification on silica gel (see Supporting Information). To our delight, the mechanochemical approach resulted in a complete conversion of the starting materials, reduced the reaction time to 70 min, limited the amount of solvent (from 15 mL to 33 μL, η = 0.1 μL mg −1 ), avoided the excess of an alkylating agent, and eliminated the purification step on silica gel. The desired compound 3a was obtained in 90% yield (Scheme 2). Encouraged by these findings, we extended this protocol to the alkylation of amine function by testing different Bocprotected alicyclic amines. The kinetics of the reaction strongly depended on the type of amine used as the alkylation of 3-Bocamino-8-azabicyclooctane (tropane), and 4-Boc-piperazine required a higher milling time (140 and 120 min, respectively) to achieve high conversions. The corresponding intermediates 3b and 3c were isolated in 90% and 89% yields, respectively. When 3-Boc-amino-pyrrolidine was used as a nucleophile, a prolonged milling time afforded the desired compound 3d in 83% yield. However, 15% of the dehydrated byproduct 3e, which was not generated in solution, was observed. In the case of other alicyclic amines, a lower percentage of the respective dehydrated side products (5−10%) were observed when the milling time exceeded 160 min.
Various accessible methods to form a sulfonamide bond in conventional solution or on solid support have been extensively reported in literature. 37−42 However, most of them involve the use of a strong organic base, excess of a sulfonylating agent, and low environmentally friendly solvents (in particular dichloromethane). To develop a more sustainable and efficient methodology, sulfonylation of intermediate 4 with 3-chlorobenzenesulfonyl chloride was optimized under milling conditions. Prior to the reaction, primary amine 4 was obtained by the treatment of Boc-derivative 3a in the presence of gaseous HCl (for more information, see Supporting Information).
Compound PZ-1361 (5a) was finally isolated in a high yield after milling of the substrates in the presence of nontoxic K 2 CO 3 for 1 min (Table 3, entry 1). To illustrate the versatility of this mechanochemical approach, compounds 5b−e bearing electron-withdrawing or -donating substituents in different positions of the benzenesulfonamide moiety were prepared. It seems that the position of the substituent has more influence on sulfonylation rather than its electronic effect. All derivatives were obtained in good yields and purities comparable to those obtained in solution. In addition, the presence of a 2-fluoro or 3-methoxy substituent ( Table 3, entries 3 and 4) led to higher conversions in a relatively shorter time than 4-nitro and unsubstituted analogues (Table 3, entries 2 and 5).
In summary, mechanochemistry enabled the preparation of the biologically active compound PZ-1361 (5a) in 4 steps with an overall yield of 64%. This approach required a total milling time of 331 min; only a slight excess of the alkylating agent was required, and no chromatographic purification was necessary to isolate 5a in high purity ( Figure 2). The versatility of the protocol was confirmed by introducing diversification at the aryloxy fragment (1a−d), central amine core (3a−d), and the benzenesulfonamide moiety (5a−e). Compared to the classical thermal methods in solution, the mechanochemical approach offered several advantages including (i) improvement of the overall yield (from 34% to 64%), (ii) reduction of reaction time (from 60 to 5.5 h), and (iii) limitation of the use of toxic solvents and the formation of byproducts. Moreover, reaction conditions and workup procedures were simplified because intermediates and final compounds were obtained by simple extraction without the need for column chromatography purification. To assess the sustainability of the newly developed approach, the green chemistry metrics E factor and Ecoscale   The Journal of Organic Chemistry pubs.acs.org/joc Note score were calculated, 43,44 which confirmed the advantage of the mechanochemical strategy over the classical organic chemistry pathway (for more information, see Supporting Information). The E factor calculated for the 4-step synthesis of PZ-1361 is reduced by 2.7 (from 1932 to 715) when switching from classical solution synthesis to mechanochemistry.
In addition, Ecoscale scores were found to be excellent for each synthetic step (>70) in the solid state, while some of them were inadequate in solution (<50). To the best of our knowledge, this approach represents the first example of organic synthesis of a CNS-acting agent exclusively performed using the mechanochemical methodology. Further studies aimed to confirm the suitability of the mechanochemical approach for an efficient and sustainable synthesis of a focused library of biologically active compounds are under development.
■ EXPERIMENTAL SECTION General Remarks. The milling treatments were carried out in a vibratory ball-mill Retsch MM400 operated at 30 Hz. The milling load is defined as the sum of the mass of the reactants per free volume in the jar and was equal to 10 or 25 mg/mL. All of the reactions using vibratory ball mill were performed under air. 1 H and 13 C NMR spectra were recorded on a JEOL JNM-ECZR500 RS1 (ECZR version) at 500 and 126 MHz, respectively, and are reported in ppm using deuterated solvent for calibration (CDCl 3 or CD 3  Chromatographic separations were carried out using the Acquity UPLC BEH (bridged ethyl hybrid) C18 column; 2.1 mm × 100 mm, and 1.7 μm particle size, equipped with Acquity UPLC BEH C18 Van Guard precolumn; 2.1 mm × 5 mm, and 1.7 μm particle size. The column was maintained at 40°C and eluted under gradient conditions from 95% to 0% of eluent A over 10 min, at a flow rate of 0.3 mL min −1 . Eluent A: water/formic acid (0.1%, v/v), Eluent B: acetonitrile/formic acid (0.1%, v/v). HRMS analyses were performed on an UPLC Acquity H-Class from Waters hyphenated to a Synapt G2-S mass spectrometer with a dual ESI source from Waters.
Alkylation of Phenol in Ball Mill (General Procedure B). Different substituted phenol (1 equiv) and previously grinded K 2 CO 3 (3 equiv) were introduced in a 35 mL PTFE jar (milling load 10 mg/ mL) with one stainless steel ball (ϕ ball = 1.5 cm). Then epichlorohydrin (1.2 equiv) was added. The reaction was carried out for 120−140 min at rt. The mixture was solubilized in CH 2 Cl 2 (15 mL), and the organic phase was washed with 2 N NaOH aqueous solution (3 × 5 mL) and saturated NaCl solution (1 × 5 mL), dried over Na 2 SO 4 , and finally filtered and concentrated under reduced pressure.