Total Synthesis of Eliglustat via Diastereoselective Amination of Chiral para-Methoxycinnamyl Benzyl Ether

Eliglustat (Cerdelga®, Genzyme Corp. Cambridge, MA, USA) is an approved drug for a non-neurological type of Gaucher disease. Herein, we describe the total synthesis of eliglustat 1 starting from readily available 1,4-benzodioxan-6-carbaldehyde via Sharpless asymmetric dihydroxylation and diastereoselective amination of chiral para-methoxycinnamyl benzyl ethers using chlorosulfonyl isocyanate as the key steps. Notably, the reaction between syn-1,2-dibenzyl ether 6 and chlorosulfonyl isocyanate in the mixture of toluene and hexane (10:1) afforded syn-1,2-amino alcohol 5 at a 62% yield with a diastereoselectivity > 20:1. This observation can be explained by competition between the SNi and the SN1 mechanisms, leading to the retention of stereochemistry.

In particular, eliglustat is a medication used to treat a non-neurological type of Gaucher disease and was approved by the United States FDA in 2014 [4][5][6][7]. Due to its potent biological activity and unique structural features, several synthetic approaches for the preparation of eliglustat have been developed [16][17][18][19][20][21]. For example, the Genzyme company reported the synthesis of eliglustat from 1,4-benzodioxan-6-carbaldehyde through a chiral aldol reaction with the chiral-pooled intermediate of phenylglycinol [19]. Xu and coworkers described the total synthesis of eliglustat through a Crimmins aldol reaction of 1,4-benzodioxan-6carbaldehyde used with Evans auxiliary [20]. Moreover, the synthesis of eliglustat using an organocatalytic asymmetric Henry reaction as a key step was also reported [21]. As part of an ongoing research program aimed at the total synthesis of pharmacologically active compounds via stereoselective amination of chiral benzylic ethers using chlorosulfonyl isocyanate (CSI) [22][23][24][25][26], we herein report the total synthesis of eliglustat from commercially available 1,4-benzodioxan-6-carboxaldehyde via diastereoselective amination of chiral para-methoxycinnamyl benzyl ethers using CSI.

Results and Discussion
The retrosynthetic analysis of eliglustat (1) is shown in Scheme 1. It shows that 1 was synthesized through ozonolysis, reductive amination, the removal of protected groups, and acylation from the intermediate 5. The required syn-1,2-amino alcohol motif of 5 was prepared via the diastereoselective installation of a NHCbz moiety into syn-1,2dibenzyl ether 6 using CSI, which in turn was easily derived from 2,3dihydrobenzo[b] [1,4]dioxine-6-carbaldehyde (7) as a starting material. Our initial study focused on the efficient construction of syn-1,2-dibenzyl ether 6, which was subjected to a stereoselective amination methodology to give the protected syn-1,2-amino benzyl ether 5 (Scheme 2). The Horner-Wadsworth-Emmons olefination of 7 with trimethyl phosphonoacetate in the presence of sodium hydride afforded olefin 8 at a 98% yield. The Sharpless asymmetric dihydroxylation [27] of olefin 8 with ADmix-β furnished the chiral syn-1,2-diol 9 at an 82% yield. Next, the syn-1,2-diol 9 was protected with acetone to produce 10 at an 83% yield. The reduction of an ester group on 10 to a primary alcohol followed by Swern oxidation and Wittig olefination provided the para-methoxycinnamyl derivative 11 (E:Z ratio of 1:3) at a 67% yield over three steps.

Results and Discussion
The retrosynthetic analysis of eliglustat (1) is shown in Scheme 1. It shows that 1 was synthesized through ozonolysis, reductive amination, the removal of protected groups, and acylation from the intermediate 5. The required syn-1,2-amino alcohol motif of 5 was prepared via the diastereoselective installation of a NHCbz moiety into syn-1,2-dibenzyl ether 6 using CSI, which in turn was easily derived from 2,3-dihydrobenzo[b] [1,4]dioxine-6-carbaldehyde (7) as a starting material.

Results and Discussion
The retrosynthetic analysis of eliglustat (1) is shown in Scheme 1. It shows that 1 was synthesized through ozonolysis, reductive amination, the removal of protected groups, and acylation from the intermediate 5. The required syn-1,2-amino alcohol motif of 5 was prepared via the diastereoselective installation of a NHCbz moiety into syn-1,2dibenzyl ether 6 using CSI, which in turn was easily derived from 2,3dihydrobenzo[b] [1,4]dioxine-6-carbaldehyde (7) as a starting material. Our initial study focused on the efficient construction of syn-1,2-dibenzyl ether 6, which was subjected to a stereoselective amination methodology to give the protected syn-1,2-amino benzyl ether 5 (Scheme 2). The Horner-Wadsworth-Emmons olefination of 7 with trimethyl phosphonoacetate in the presence of sodium hydride afforded olefin 8 at a 98% yield. The Sharpless asymmetric dihydroxylation [27] of olefin 8 with ADmix-β furnished the chiral syn-1,2-diol 9 at an 82% yield. Next, the syn-1,2-diol 9 was protected with acetone to produce 10 at an 83% yield. The reduction of an ester group on 10 to a primary alcohol followed by Swern oxidation and Wittig olefination provided the para-methoxycinnamyl derivative 11 (E:Z ratio of 1:3) at a 67% yield over three steps. Our initial study focused on the efficient construction of syn-1,2-dibenzyl ether 6, which was subjected to a stereoselective amination methodology to give the protected syn-1,2amino benzyl ether 5 (Scheme 2). The Horner-Wadsworth-Emmons olefination of 7 with trimethyl phosphonoacetate in the presence of sodium hydride afforded olefin 8 at a 98% yield. The Sharpless asymmetric dihydroxylation [27] of olefin 8 with AD-mix-β furnished the chiral syn-1,2-diol 9 at an 82% yield. Next, the syn-1,2-diol 9 was protected with acetone to produce 10 at an 83% yield. The reduction of an ester group on 10 to a primary alcohol followed by Swern oxidation and Wittig olefination provided the para-methoxycinnamyl derivative 11 (E:Z ratio of 1:3) at a 67% yield over three steps. The removal of acetal and the subsequent benzylation of syn-diol 12 furnished the syn-1,2-dibenzyl ether 6 at a 71% yield. Next, we screened the optimal reaction conditions for the stereoselective amination of syn-1,2-dibenzyl ether 6 using CSI ( Table 1). The coupling of 6 and CSI in dichloromethane at 0 °C furnished the corresponding carbamate 5 at a 29% yield with a diastereoselectivity of 1.2:1 ( Table 1, entry 1). An examination of the effect of solvents under various temperatures indicated that the improved yield and diastereoselectivity were achieved by using toluene, affording 5 (Table 1, entry 9). After further screening, we found that the highest diastereoselectivity (dr > 20:1) was obtained in a mixture of toluene and n-hexane (10:1) at −78 °C with an improved yield of 62% (Table 1, entry 11). Next, we screened the optimal reaction conditions for the stereoselective amination of syn-1,2-dibenzyl ether 6 using CSI ( Table 1). The coupling of 6 and CSI in dichloromethane at 0 • C furnished the corresponding carbamate 5 at a 29% yield with a diastereoselectivity of 1.2:1 ( Table 1, entry 1). An examination of the effect of solvents under various temperatures indicated that the improved yield and diastereoselectivity were achieved by using toluene, affording 5 (Table 1, entry 9). After further screening, we found that the highest diastereoselectivity (dr > 20:1) was obtained in a mixture of toluene and n-hexane (10:1) at −78 • C with an improved yield of 62% (Table 1, entry 11).
The diastereomeric ratio was determined by 1 H NMR analysis. As shown in Figure 2 (case selected), a single proton at the chiral position was significantly distinguishable in the 1 H NMR spectra. As a result, higher diastereoselectivity was obtained in the mixed solvent system of toluene and n-hexane (entry 11) than in other single nonpolar solvent systems with CH 2 Cl 2 (entry 1 and 2), n-hexane (entry 5) and toluene (entry 9). of syn-1,2-dibenzyl ether 6 using CSI ( Table 1). The coupling of 6 and CSI in dichloromethane at 0 °C furnished the corresponding carbamate 5 at a 29% yield with a diastereoselectivity of 1.2:1 ( Table 1, entry 1). An examination of the effect of solvents under various temperatures indicated that the improved yield and diastereoselectivity were achieved by using toluene, affording 5 (Table 1, entry 9). After further screening, we found that the highest diastereoselectivity (dr > 20:1) was obtained in a mixture of toluene and n-hexane (10:1) at −78 °C with an improved yield of 62% (Table 1, entry 11). The diastereomeric ratio was determined by 1 H NMR analysis. As shown in Figure 2 (case selected), a single proton at the chiral position was significantly distinguishable in the 1 H NMR spectra. As a result, higher diastereoselectivity was obtained in the mixed solvent system of toluene and n-hexane (entry 11) than in other single nonpolar solvent systems with CH2Cl2 (entry 1 and 2), n-hexane (entry 5) and toluene (entry 9). The origin of the diastereoselectivity can be explained by competition between the SNi pathway leading to a retention of stereochemistry through a four-membered transition state and the SN1 pathway through a carbocation intermediate [28] (Scheme 3). The improved diastereoselectivity in mixed nonpolar solvents (toluene and n-hexane) at low  The origin of the diastereoselectivity can be explained by competition between the S N i pathway leading to a retention of stereochemistry through a four-membered transition state and the S N 1 pathway through a carbocation intermediate [28] (Scheme 3). The improved diastereoselectivity in mixed nonpolar solvents (toluene and n-hexane) at low temperatures (−78 • C) might be realized by the increased formation of a tight ion pair intermediate (IIA) rather than a carbocation intermediate (IIB). To complete the total synthesis of 1, an ozonolysis reaction of compound 5 was performed to afford the corresponding aldehyde, which was converted to the tertiary amine 13 by reductive amination with pyrrolidine (Scheme 4). The Pd-catalyzed hydrogenolysis of 13 afforded syn-1,2-amino alcohol 14 at an 81% yield. Finally, the acylation of amino alcohol 14 with n-octanoyl chloride furnished eliglustat (1) at a 53% yield. The analytical data ( 1 H NMR) and melting point of the synthesized eliglustat (1) were in full agreement with those reported in the literature [19]. To complete the total synthesis of 1, an ozonolysis reaction of compound 5 was performed to afford the corresponding aldehyde, which was converted to the tertiary amine 13 by reductive amination with pyrrolidine (Scheme 4). The Pd-catalyzed hydrogenolysis of 13 afforded syn-1,2-amino alcohol 14 at an 81% yield. Finally, the acylation of amino alcohol 14 with n-octanoyl chloride furnished eliglustat (1) at a 53% yield. The analytical data ( 1 H NMR) and melting point of the synthesized eliglustat (1) were in full agreement with those reported in the literature [19].
To complete the total synthesis of 1, an ozonolysis reaction of compound 5 was performed to afford the corresponding aldehyde, which was converted to the tertiary amine 13 by reductive amination with pyrrolidine (Scheme 4). The Pd-catalyzed hydrogenolysis of 13 afforded syn-1,2-amino alcohol 14 at an 81% yield. Finally, the acylation of amino alcohol 14 with n-octanoyl chloride furnished eliglustat (1) at a 53% yield. The analytical data ( 1 H NMR) and melting point of the synthesized eliglustat (1) were in full agreement with those reported in the literature [19].

General Information
Commercially available reagents were used without additional purification unless otherwise stated. All reactions were performed under an inert atmosphere of nitrogen. The nuclear magnetic resonance spectra ( 1 H and 13 C NMR) were recorded on an Agilent 400 MR 400 MHz Spectrometer with Oxford NMR AS400 Magnet spectrometers in CDCl 3 solution, and the chemical shifts were reported as parts per million (ppm). Resonance patterns were reported with the notations s (singlet), d (doublet), t (triplet), q (quartet), and m (multiplet). In addition, the notation of br was used to indicate a broad signal. The coupling constants (J) were reported in hertz (Hz). The IR spectra were recorded on a JASCO FT/IR-4600 spectrophotometer and were reported as cm
To a solution of oxalyl chloride (1.09 mL, 12.74 mmol) in dry dichloromethane (25 mL), dimethyl sulfoxide (1.81 mL, 25.48 mmol) in dichloromethane (25 mL) was added at −78 • C. The reaction mixture was stirred for 1 h at −78 • C. To the reaction mixture, a solution of the alcohol in dichloromethane (25 mL) was added slowly and stirred for another 1 h at the same temperature. Then, trimethylamine (5.92 mL, 42.47 mmol) was added, and the resulting mixture was stirred for an additional 30 min. The reaction mixture was carefully quenched with water. The aqueous layer was extracted with dichloromethane (2 × 100 mL). The combined organic extracts were dried over anhydrous sodium sulfate, filtered, and concentrated in vacuo to afford crude carboxaldehyde.
Freshly prepared 4-methoxybenzyltriphenylphosphonium chloride (3.56 g, 8.49 mmol) was placed in a dry round-bottomed flask under a nitrogen atmosphere and diluted with dry tetrahydrofuran (50 mL). To the reaction mixture, n-Buthillithium (4.08 mL, 10.19 mmol, 2.5 M in tetrahydrofuran) was added dropwise at 0 • C. The resulting reaction mixture was stirred for 30 min until the mixture turned to a deep red solution. A solution of aldehyde in dry tetrahydrfuran (25 mL) was added slowly, and the resultant mixture was stirred for 4 h at room temperature. The reaction mixture was carefully quenched with water and extracted with EtOAc (2 × 100 mL). The combined organic layers were dried over anhydrous sodium sulfate, filtered, and concentrated in vacuo. The residue was purified by flash chromatography on silica gel (0~30% EtOAc/n-hexanes) to afford 11 (2.1 g, E/Z = 1:3) at a 67.1% yield (three steps overall) as a colorless oil (R f = 0.33 (20% EtOAc/n-hexanes);  [1,4]di-oxine (6) A solution of 12 (0.40 g, 1.22 mmol) in THF (5 mL) was added to a mixture of sodium hydride (60% dispersed in mineral oil, 0.15 g, 3.78 mmol) and tetra n-butylammonium iodide (0.045 g, 0.12 mmol) in THF (10 mL) at room temperature and stirred for 10 min at room temperature. Benzyl bromide (0.43 mL, 3.65 mmol) was added, and the resulting mixture was stirred for 12 h at room temperature. The reaction mixture was quenched with a 50% aqueous NaHCO 3 solution and extracted with EtOAc (2 × 50 mL). The combined organic layers were dried over anhydrous sodium sulfate, filtered, and concentrated in vacuo. The residue was purified by flash chromatography on silica gel (0~20% EtOAc/nhexanes) to afford 6 (0.44 g) at a 71.0% yield as a yellow oil (R f = 0. 28 [1,4]dioxin-6-yl)-4-(4-Methoxyphenyl)-but-3-en-2-yl)carbamate (5) To a stirred solution of 6 (0.10 g, 0.20 mmol) and sodium carbonate (0.27 g, 2.56 mmol) in a mixture of anhydrous toluene/n-hexane (10:1, 4.0 mL), chlorosulfonyl isocyanate (0.34 mL, 3.93 mmol) was added at −78 • C under a N 2 atmosphere. The reaction mixture was stirred for 24 h at −78 • C and quenched with water very carefully (caution: highly exothermic). The aqueous layer was extracted with EtOAc (2 × 20 mL). The combined organic layers were placed into a 100 mL round-bottomed flask and 25% aqueous solution of sodium sulfite (10 mL) was added. The mixture was further stirred for 12 h at room temperature. The EtOAc layer was separated, and the remaining compound was further extracted with EtOAc (30 mL) from the aqueous layer. The combined organic layers were washed with brine (30 mL), dried over anhydrous sodium sulfate, and concentrated in vacuo. The residue was purified by flash column chromatography on silica gel (0~20% EtOAc/n-hexanes) to afford 5 (0.067 g) at a 62.0% yield as a colorless oil (R f = 0. 15 (13) To a flame-dried round-bottomed flask containing 5 (0.16 g, 0.30 mmol), dry dichloromethane (15 mL) was added. The solution was cooled to −78 • C, and a stream of O 3 /O 2 was passed through the reaction for 30 min, after which the reaction was purged with a stream of O 2 for 1 min. After that, triphenyl phosphine (0.08 g, 0.30 mmol) was added and stirred for 1 h at room temperature. To the reaction mixture, pyrrolidine (0.026 mL, 0.32 mmol) and triethylamine (0.061 mL, 0.44 mmol) were added, and then sodium cyanoborohydride (0.024 g, 0.38 mmol) was added. The reaction mixture was stirred for 16 h at room temperature. The reaction mixture was quenched with water. The aqueous layer was extracted with DCM (2 × 20 mL). The combined organic layers were dried over anhydrous sodium sulfate, filtered, and concentrated. The residue was purified by flash column chromatography on silica gel (0~10% MeOH/EtOAc) to afford 13 (0.