Practical and Scalable Synthesis of Cisatracurium Besylate

Abstract

Cisatracurium besylate (1) is a potent nondepolarizing neuromuscular blocking agent utilized in clinical anesthesia. Its primary synthetic challenges arise from its four chiral centers and its unstable properties. We herein attempt to explore a practical and scalable method for the enantioselective synthesis of 1. In this work, 1-(3,4-dimethoxybenzyl)-6,7-dimethoxy-3,4-dihydroisoquinoline (8) was used as a starting material. R-tetrahydropapaverine (R-6) was generated through a Ru-catalyzed asymmetric transfer hydrogenation reaction with 77.0% yield, 99.45% purity, and 99.67% ee. This was a key step. The control of the chiral nitrogen can be achieved through an N-methylation process to produce (R-cis)-10 with 41.6% yield, 99.08% purity, and gratifying enantioselectivity (100% de) through a simple and repeatable recrystallization. The process can be utilized to produce 1 without chromatography purification in five steps with an HPLC purity of 98.71% on a 100-g scale. The process was easy to implement and suitable for manufacturing-scale production.

 

MissingLinkText

Keywords

Cisatracurium besylate - enantioselective synthesis - asymmetric transfer hydrogenation

Introduction

Cisatracurium besylate is a potent nondepolarizing neuromuscular blocking agent and is widely utilized in clinical anesthesia due to its improved pharmacological profile compared with Atracurium besylate.[1] [2] As a R-cis stereoisomer, Cisatracurium besylate possesses 4-fold higher potency than Atracurium besylate. Cisatracurium besylate is primarily metabolized through Hofmann elimination in the body, making it suitable for patients with liver and kidney dysfunction and reducing histamine release-related cardiovascular side effects.[3] [4] Since Food and Drug Administration approval in 1995, this agent has been clinically essential for tracheal intubation and intraoperative muscle relaxation during general anesthesia.

The primary synthetic challenges of Cisatracurium besylate (1, [Fig. 1]) arise from its four chiral centers. Chemically, this compound contains two chiral carbon atoms and two chiral nitrogen atoms. Due to its symmetrical structure, the compound has nine stereoisomers, allowing strict quality control in the final product. Additionally, this compound features two ester structures and two quaternary ammonium salt structures, rendering it susceptible to degradation through ester hydrolysis and Hofmann elimination ([Fig. 1]).[5] [6] Given the above, the instability of Cisatracurium besylate complicates the production process and makes purification challenging.

Fig. 1 Cisatracurium besylate (1) chemical structure and its degradation pathway.

Initially, Hill and Turner reported a four-step route to obtain compound 1 with >98% purity ([Scheme 1]).[7] In the synthetic route, tetrahydropapaverine (6) was employed to prepare R-tetrahydropapaverine (R-6) to enhance the chiral purity of the chiral carbon atoms, but it has a low enantioselectivity (97% ee, 41.0% yield) and necessitated a considerable amount of solvent in the resolution process. Then, R-6•N-acetyl-L-leucine underwent an N-alkylation reaction, and the subsequent N-methylation process to form compound 1's symmetrical structure, generating two isomeric impurities: (1R-cis-1'R-trans)- 1 and (1R-trans-1'R-trans)- 1. The ratio of 1: (1R-cis-1'R-trans)- 1: (1R-trans-1'R-trans)- 1) in crude 1 was 58:34:6. Column chromatography purification was used to separate the final product with a low yield (37.5%). Currently, the primary production methods for Cisatracurium besylate are conducted using similar techniques. Therefore, there is an urgent need to explore a strategy with higher yields and easier quality control.

Scheme 1 The reported synthetic route of compound 1. Reagents and conditions: (a) aqueous ammonia; (b) N-acetyl-L-leucine, MeOH, diethyl ether; (c) acetone; (d) acetone; (e) aqueous ammonia; (f) 1,5-pentamethylene diacrylate, toluene, glacial acetic acid, silica gel (70-230 mesh); (g) acetone, oxalic acid; (h) water, sodium carbonate, toluene; (i) methyl benzenesulfonate; (j) column chromatography, DCM/MeOH/benzenesulfonic acid (4,000:500:0.25).

Asymmetric transfer hydrogenation (ATH) reaction is a potential solution to meet the stereoselectivity challenge in the synthesis of compound R-6. We referred to other ATH reactions of 3,4-dihydroisoquinoline or its derivatives[8] [9] [10] [11] [12] [13] [14] [15] and chose to start at a Ru-catalyzed ATH reaction of 1-(3,4-dimethoxybenzyl)-6,7-dimethoxy-3,4-dihydroisoquinoline (8), which was synthesized through a one-pot method as we reported previously.[16] The N-methylation process was performed before the formation of the symmetrical structure; this avoids the column chromatography purification in the route. Fortunately, only one stereoisomer ((R-trans)-10) occurred in intermediates ([Scheme 2]), making crystallization possible. As we control the purity of all chiral centers within the intermediate, the quality control of the final product becomes straightforward, thereby avoiding operational difficulties that arise from unstable product properties.

Fig. 2 Ruthenium catalysts screened in asymmetric transfer hydrogenation.

Scheme 2 The improved synthetic route of compound 1.

 

MissingLinkText

Results and Discussion

The synthesis route of 1 presented in this work is depicted in [Scheme 2]. The route started with compound 8, which underwent a Ru-catalyzed ATH reaction to achieve R-6 D-tartrate (step 1), followed by a Michael addition reaction with tert-butyl acrylate to obtain compound 9 (step 2), an N-methylation process to obtain (R-cis)-10 (step 3), and transesterification to generate the target product (1) (steps 4 and 5). Consequently, 1 was synthesized through a 5-step reaction with a 21% overall yield without the need for column chromatography purification. Remarkably, the enantioselectivity of R-6 and (R-cis)-10 was 99.67% ee and 100% de, respectively. The entire process effectively demonstrated the safety and efficacy of this synthetic pathway.

Optimization of Asymmetric Transfer Hydrogenation

A series of ruthenium catalysts was screened for the ATH reaction ([Table 1], entries 1 − 6). Their chemical structures were shown in [Fig. 2]. Our data showed that when HCOOH/Et₃N (5:2) was used as a hydrogen source and EtOH as a solvent, RuCl[(S,S)-TsDPEN](p-cymene) (16) showed the best result with R-6 being obtained with a 96.56% ee and 97.82% conversion ([Table 1], entry 6). Next, we screened the solvents, including DCM, THF, DMF, i-PrOH, CH₃CN, and MeOH ([Table 1], entries 7 − 12). It was found that MeOH showed the best result with R-6 being obtained with a 97.00% ee and 98.34% conversion ([Table 1], entry 12). Then, the base and the ratio of the base to formic acid were screened when 16 was used as a catalyst and MeOH as a solvent ([Table 1], entries 13 − 15). To our disappointment, there were no significant differences in ee and conversion when using different bases and different ratios. At last, the hydrogen source was screened ([Table 1], entries 16 − 17). Our data showed that HCOONa was found to be suitable with higher ee, higher conversion, and shorter reaction time ([Table 1], entry 17). To improve the solubility of HCOONa, H₂O and MeOH/H₂O (1:1) were discussed as a solvent ([Table 1], entries 18 − 19), suggesting that MeOH/H₂O (1:1) was better than a single solvent ([Table 1], entry 19). Given the above, R-6 could be prepared by using RuCl[(S,S)-TsDPEN](p-cymene) (16, 0.8 mol% %) as a catalyst, HCOONa as the hydrogen source, and MeOH/H₂O (1:1, 10 V) as the solvent. The product had a yield of 77.0 and 99.67% ee when the process was conducted on a 1,096-g scale.

Table 1 Screening of asymmetric transfer hydrogenation[a]

Entry	Ruthenium catalyst[b]	Hydrogen source	Solvent	ee (%)	Conversion (%)
1	11	HCOOH/Et3N (5:2)	EtOH	59.34	86.26
2	12	HCOOH/Et3N (5:2)	EtOH	31.60	83.18
3	13	HCOOH/Et3N (5:2)	EtOH	−1.20	75.71
4	14	HCOOH/Et3N (5:2)	EtOH	−2.54	77.63
5	15	HCOOH/Et3N (5:2)	EtOH	94.10	89.88
6	16	HCOOH/Et3N (5:2)	EtOH	96.56	97.82
7	16	HCOOH/Et3N (5:2)	DCM	86.94	90.90
8	16	HCOOH/Et3N (5:2)	THF	86.34	98.02
9	16	HCOOH/Et3N (5:2)	DMF	93.40	74.38
10	16	HCOOH/Et3N (5:2)	i-PrOH	91.08	98.29
11	16	HCOOH/Et3N (5:2)	CH3CN	93.80	98.15
12	16	HCOOH/Et3N (5:2)	MeOH	97.00	98.34
13	16	HCOOH/Et3N (1:1)	MeOH	97.19	98.11
14	16	HCOOH/DIPEA (5:2)	MeOH	96.84	98.17
15	16	HCOOH/DIPEA (1:1)	MeOH	96.72	97.81
16	16	HCOONH4	MeOH	98.10	89.42
17	16	HCOONa	MeOH	98.17	99.16
18	16	HCOONa	H2O	95.99	99.12
19	16	HCOONa	MeOH:H2O (1:1)	98.19	99.64

^a Reaction conditions: 8 (1 g, 2.03 mmol, 1.0 equiv.), 40°C. ee and conversion was measured by a chiral IC column with high performance liquid chromatography method.

^b 0.8 mol %.

 

MissingLinkText

Process for (R-cis)-10 and Crystallization

Michael addition reaction of R-6 with tert-butyl acrylate gave compound 9 in 95.80% yield. (R-cis)-10 was synthesized from compound 9 and methyl benzenesulfonate, forming only a single isomeric impurity, (R-trans)-10, with a ratio of (R-cis)-10/(R-trans)-10 being 62.36:20.27. Due to their difference in solubility, we attempted to separate (R-trans)-10 and (R-trans)-10 through a recrystallization purification.

Several common solvents, including EA, DCM, Et₂O, and CH₃CN, were screened for the recrystallization process ([Table 2], entries 1 − 4). Our data showed that DCM and CH₃CN dissolved (R-cis)-10 effectively and (R-trans)-10 moderately, whereas EA and Et₂O showed no solubility. Mixed solvent systems, like DCM/Et₂O (1:2) and DCM/EA (1:3) ([Table 2], entries 5, 6), were explored to optimize crystallization. The result showed that DCM/Et₂O (1:2) preferentially dissolved (R-cis)-10, whereas DCM/EA (1:3) selectively dissolved (R-trans)-10. Finally, crystallization was performed using DCM/Et₂O (1:2, 5.5 V) to remove most of (R-trans)-10, followed by further crystallization with DCM/EA (1:3, 4 V). The final product had a 41.6% yield and 100% de when the resolution process was scaled to 857.71 g.

Table 2 Solvents screening for recrystallization purification of (R-cis)-10 and (R-trans)-10

Entry	Solvent	(R-cis)-10	(R-trans)-10
		Solubility (mg/mL)	Solubility (mg/mL)
1	DCM	2058	167
2	CH3CN	2087	252
3	EA	Insoluble	Insoluble
4	Et2O	Insoluble	Insoluble
5	DCM/Et2O (1:2)	390	5
6	DCM/EA (1:3)	364	7

^a Reaction conditions: 9 (5 g, 8.90 mmol, 1.0 equiv.), methyl benzenesulfonate (3.07 g, 17.8 mmol, 2.0 equiv.), DCM (2 V), 35°C.

 

MissingLinkText

Process for Cisatracurium Besylate (Compound 1)

In this study, (R-cis)-10 was hydrolyzed with benzenesulfonic acid to get 2, which was then reacted with 1,5-pentanediol (0.5 equiv.) to obtain the target product (1) through a one-pot method. The purification of the target product was straightforward, involving extraction with water and recrystallization using DCM/MTBE, due to the high de of (R-cis)-10. However, the removal of water generated in this step is critical to increase the rate of inversion. Initially, co-boiling was explored to remove the water using 2-butanone, toluene, DCE, chloroform, and DCM tested as solvents ([Table 3], entries 1 − 5). The result showed that 2-butanone, toluene, DCE, and chloroform were associated with lower inversion rates and more impurities ([Table 3], entries 1 − 4). DCM achieved a 70.22% inversion rate ([Table 3], entry 5). When 4Å molecular sieves were used instead, in combination with DCM (15 V) as the solvent, the best inversion rate was achieved ([Table 3], entry 6). The product had a yield of 67.0 and 98.71% purity when the process was conducted on a 129.41-g scale.

Table 3 Method to remove H₂O in the synthesis of Cisatracurium besylate[a]

Entry	Solvent	Method to remove H2O	Temperature (°C)	Inversion rate (%)[b]
1	2-Butanone	Co-boiling	85	n.d.
2	Toluene	Co-boiling	90	22.43
3	DCE	Co-boiling	75	57.22
4	Chloroform	Co-boiling	65	52.39
5	DCM	Co-boiling	40	70.22
6	DCM	Add 4Å molecular sieve	40	90.08

^a Reaction conditions: (R-cis)-10 (1.00 g, 0.155 mmol, 1.0 equiv.), benzenesulfonic acid (1.23 g, 0.777 mmol, 5.0 equiv.).

^b Inversion rate was measured by a high performance liquid chromatography method.

 

MissingLinkText

 

MissingLinkText

Conclusion

In summary, a new approach is designed to prepare Cisatracurium besylate, the first chiral center is controlled by ATH reaction with a ruthenium catalyst (from 8 to R-6), and the second chiral center is controlled by crystallization purification (from 9 to (R-cis)-10). Column chromatography purification was avoided. The overall operation is simple, with less solvent usage and high atom economy. First, R-tetrahydropapaverine D-tartrate (R-6 D-tartrate) is synthesized from compound 8 with the catalysis of RuCl[(S,S)-TsDPEN](p-cymene) (16) on a 1,096-g scale with 77% yield and 99.67% ee. Second, compound 9 is synthesized from R-6 D-tartrate on a 1,093.45-g scale with 95.80% yield and 95.64% purity by a Michael addition. Then, a practical and convenient crystallization process is developed to prepare compound (R-cis)-10 with 41.6% yield and 100% de on a 857.71-g scale. Finally, Cisatracurium besylate (1) is performed with 67.0% yield and 98.71% purity on a 129.41-g scale. Overall, the product exhibits excellent isolated yield and purity.

 

MissingLinkText

Experimental Section

General

All reagents were commercially available and used without further purification unless indicated otherwise. Nuclear magnetic resonance (NMR) spectra were recorded on an AVANCE III 400 MHz spectrometer (Bruker) in deuterated D₂O or CDCl₃, using tetramethylsilane as the internal reference. Chemical shifts are given in δ values (ppm), and coupling constants (J values) are given in Hz. High-resolution mass spectrometry (HRMS) spectra were recorded using a Waters quadrupole time-of-flight micromass spectrometer with an electrospray ionization (ESI) source. The reaction was monitored by a high-performance liquid chromatographic (HPLC) method using an Agilent 1260 Infinity II HPLC, and the ee was determined by a chiral Dionex UltiMate 3000 HPLC.

The HPLC method was performed using a Syncronis C18 column (250 mm × 4.6 mm, 5 μm). The mobile phase A was a mixture of H₂O/MeOH/CH₃CN (8:1:1) supplemented with 3.2% ammonium formate and 1.6% formic acid in water; the mobile phase B was a mixture of CH₃CN/MeOH (1:1). The gradient program was as follows: from 90:10 A/B to 60:40 A/B over 80 minutes, 60:40 A/B over 10 minutes, 60:40 A/B to 90:10 A/B over 0.1 minute, 90:10 A/B over 10 minutes. The detection wavelength was 280 nm; the flow rate was 0.8 mL/min; the column temperature was 35°C.

The chiral HPLC method was performed using a CHIRALPAK IC-3 (250 × 4.6 mm, 3 μm) column, with a mobile phase (n-hexane/EtOH/1,4-dioxane/TFA/Et₂NH, 750:200:50:3:3) over 20 minutes. The flow rate was 0.7 mL/min; the column temperature was 25°C, and the detection wavelength was 235 nm.

 

MissingLinkText

Preparation of (R)-1-(3,4-dimethoxybenzyl)-6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline (2S,3S)-2,3-dihydroxysuccinate (R-6) D-tartrate

To a solution of 1-(3,4-dimethoxybenzyl)-6,7-dimethoxy-3,4-dihydroisoquinoline hydrogen chloride (8) (1.0 equiv., 2.885 mol, 1,090 g) in the MeOH/H₂O (5 L:5 L) mixture was added sodium formate (5.0 equiv., 14.43 mol, 981 g) and RuCl[(S,S)-TsDPEN](p-cymene) (16, 14.7 g, 0.023 mol, 0.8 mol%). The mixture was stirred at 40°C for 4 hours. The reaction was monitored by thin-layer chromatography. MeOH was removed under reduced pressure. The residue was extracted with DCM (5 L, 5 V). The organic phase was concentrated, and the residue was dissolved in EtOH (25 L, 25 V). Then, D-tartaric acid was added (1.0 equiv., 2.885 mol, 433 g). The mixture was stirred at 70°C for 1 hour, then cooled gradually to 0 to 5°C. Crystals were filtered and washed with ice–cold EtOH (1 L). The solid was dried to get R-6 D-tartrate as a pink solid (1,096 g, 99.45% purity, 99.67% ee, 77.0% yield). ¹H NMR (400 MHz, D₂O) δ 6.96 (d, J = 8.3 Hz, 1H), 6.87 (s, 1H), 6.81 (dd, J = 8.2, 2.0 Hz, 1H), 6.71 (d, J = 2.0 Hz, 1H), 6.33 (s, 1H), 4.70 (t, J = 7.3 Hz, 1H), 4.50 (s, 2H), 3.81 (d, J = 2.7 Hz, 6H), 3.71 (s, 3H), 3.58 (s, 3H), 3.54–3.36 (m, 2H), 3.19 (qd, J = 13.9, 7.4 Hz, 2H), 3.02 (t, J = 6.5 Hz, 2H). ¹³C NMR (101 MHz, D₂O) δ 176.29, 148.19, 147.92, 147.54, 146.38, 128.07, 124.17, 122.96, 122.58, 113.10, 111.98, 111.66, 110.16, 72.77, 55.63, 55.60, 55.49, 55.37, 38.88, 38.32, 24.04. HRMS (ESI-TOF) calcd. for C₂₀H₂₆NO₄ ⁺ [M + H]⁺ 344.18563, found: 344.18496.

 

MissingLinkText

Preparation of tert-butyl (R)-3-(1-(3,4-dimethoxybenzyl)-6,7-dimethoxy-3,4-dihydroisoquinolin-2(1H)-yl)propanoate Oxalate (9)

R-6 D-tartrate (1.0 equiv., 2.026 mol, 1,000 g) was dissolved in H₂O (10 L, 10V). The pH was adjusted to 10 with NaOH (aq) (4.4 L, 5.0 mol/L) at 0 to 5°C. The mixture was extracted with toluene (3.5 L, 3.5V) twice, and the organic phase was combined and concentrated to 2 L (2 V) under reduced pressure. Then, tert-butyl acrylate (1.2 equiv., 2.431 mol, 311.6 g) and acetic acid (0.5 equiv., 1.013 mol, 61 g) were added to the solution. The reaction mixture was stirred at 85°C for 5 hours until the proportion of R-6 was controlled to be less than 5% (monitored by HPLC). After cooling to room temperature, ethyl acetate (20 L, 20V) was added. The mixture was cooled to 0 to 5°C. Subsequently, a solution of oxalic acid (1.1 equiv., 2.244 mol, 202 g) in 1.2 L ethyl acetate was slowly added. The mixture was stirred at 0 to 5°C for 10 hours and filtered to obtain a white powder, which was washed with ice–cold ethyl acetate (3 L) and dried to constant weight to afford compound 9 (1,093.45 g, 95.64% purity, 99.79% ee, 95.8% yield) as a white solid. ¹H NMR (400 MHz, CDCl₃) δ 6.74 (d, J = 8.1 Hz, 1H), 6.64 (d, J = 2.9 Hz, 2H), 6.49 (dd, J = 8.1, 1.9 Hz, 1H), 5.65 (s, 1H), 4.39 (s, 1H), 3.86 (s, 3H), 3.83 (s, 3H), 3.80 (s, 3H), 3.72 (s, 1H), 3.63–3.49 (m, 2H), 3.43 (s, 5H), 3.19 (dt, J = 19.6, 9.4 Hz, 1H), 2.98 (dd, J = 17.9, 6.1 Hz, 1H), 2.88–2.74 (m, 3H), 1.42 (s, 9H). ¹³C NMR (101 MHz, CDCl₃) δ 169.26, 163.26, 149.24, 149.12, 148.35, 147.15, 127.80, 122.62, 113.06, 111.33, 111.12, 82.34, 55.99, 55.96, 55.93, 55.49, 47.96, 30.59, 27.98. HRMS (ESI-TOF) calcd. for C₂₇H₃₈NO₆ ⁺ [M + H]⁺ 472.26936, found: 472.27013.

 

MissingLinkText

Preparation of (1R,2R)-2-(3-(tert-butoxy)-3-oxopropyl)-1-(3,4-dimethoxybenzyl)-6,7-dimethoxy-2-methyl-1,2,3,4-tetrahydroisoquinolin-2-ium benzenesulfonate ((R-cis)-10)

Compound 9 (1.0 equiv., 3.21 mol, 1,800 g) was dissolved in a mixture of H₂O and DCM (5:5, 14.4 L, 8 V). NaOH (aq) (1.5 L, 5.0 mol/L) was added to adjust pH = 10 at 0 to 5°C. The aqueous phase was extracted with DCM (3 L, 1.67 V). The organic phase was concentrated to 1.44 L (0.8 V) under reduced pressure. Subsequently, methyl benzenesulfonate (1.2 equiv., 3.85 mol, 663 g) was added. The mixture was stirred at 35°C for 15 hours, and the proportion of 9 was controlled to be less than 5%, as monitored by an HPLC method.

Raising the reaction temperature to 25°C, the reaction mixture was diluted with DCM (360 mL, 0.2 V), followed by dropwise addition of ether (6.66 L, 3.7 V). The mixture was stirred for 1 hour and gradually cooled to 0 to 5°C. The resulting crystals were filtered to obtain (R-trans)-10 (97.27% purity). The mother liquor was collected and concentrated to obtain crude (R-cis)-10 (1,398.35 g, 88.94% purity) as a white powder.

Crude (R-cis)-10 was dissolved in DCM (2.8 L). Ethyl acetate (8.4 L) was added dropwise at 35°C. After stirring for 1 hour, the mixture was gradually cooled to 26°C, stirred for 2 hours, and filtered to get a white powder, which was washed with ice–cold DCM/EA (1:3, 600 mL) and dried to constant weight to afford (R-cis)-10 (857.71 g, 41.6% yield, 100% de, and 99.08% purity) as a white powder. ¹H NMR (400 MHz, CDCl₃) δ 7.92–7.81 (m, 2H), 7.31–7.27 (m, 3H), 6.67 (d, J = 8.1 Hz, 1H), 6.60 (s, 1H), 6.51 (d, J = 2.0 Hz, 1H), 6.45 (dd, J = 8.2, 2.0 Hz, 1H), 5.96 (s, 1H), 4.99–4.88 (m, 1H), 4.21–4.11 (m, 1H), 4.06–3.97 (m, 1H), 3.81 (d, J = 4.4 Hz, 7H), 3.69–3.55 (m, 5H), 3.43 (s, 3H), 3.21 (s, 4H), 3.11 (t, J = 7.2 Hz, 2H), 3.01 (dd, J = 18.2, 6.4 Hz, 1H), 2.90 (dd, J = 13.4, 9.3 Hz, 1H), 1.45 (s, 9H). ¹³C NMR (101 MHz, CDCl₃) δ 169.12, 149.32, 149.09, 148.36, 147.30, 146.81, 129.20, 127.97, 126.88, 125.91, 122.39, 121.63, 120.53, 113.27, 111.78, 111.17, 110.60, 82.54, 71.12, 58.77, 55.98, 55.94, 55.88, 55.65, 53.65, 46.79, 37.66, 28.79, 28.01, 23.37. HRMS (ESI-TOF) calcd. for C₂₈H₄₀NO₆ ⁺ [M]⁺ 486.28501, found: 486.28605.

 

MissingLinkText

Preparation of Cisatracurium besylate (1)

To a solution of compound (R-cis)-10 (1.0 equiv., 310.7 mmol, 200 g) in DCM (400 mL, 2V) was added benzenesulfonic acid (5.0 equiv., 1,553.5 mmol, 246 g). The reaction was stirred at 37°C for 4 hours to completely consume (R-cis)-10, as monitored by HPLC.

The reaction temperature was raised to 40°C. To the mixture was added DCM (2.6 L, 13V), 4Å molecular sieve (1,000 g), and 1,5-pentanediol (0.5 equiv., 155.36 mmol, 16.2 g). The reaction was stirred at 40°C for 67 hours. When the total proportion of 2 and 3 was controlled to be less than 10% (as monitored by HPLC), the mixture was gradually cooled to room temperature, washed with H₂O (3 L, 15 V; pH = 3 adjusting with benzenesulfonic acid) four times. The organic phase was washed with H₂O (3 L, 15 V) twice and concentrated under reduced pressure to give a residue, which was dissolved in DCM (100 mL, 0.5 V). The residue solution was added dropwise to MTBE (3 L, 15 V) at 0 to 5°C. After stirring for 4 hours, the crystals were filtered under N₂ protection, washed with ice–cold MTBE (200 mL), and dried at 30°C in a vacuum to afford compound 1 (129.41 g, 67.0% yield, 98.71% purity) as white crystals. ¹H NMR (400 MHz, CDCl₃) δ 7.87–7.78 (m, 4H), 7.29 (m, 6H), 6.63 (d, J = 8.2 Hz, 2H), 6.52 (s, 2H), 6.49–6.37 (m, 4H), 5.91 (s, 2H), 4.91 (dd, J = 9.5, 3.9 Hz, 2H), 4.16 (m, 6H), 4.00 (td, J = 13.2, 11.3, 5.7 Hz, 2H), 3.79 (m, 14H), 3.65–3.48 (m, 10H), 3.37 (m, 10H), 3.17 (m, 8 H), 2.95–2.82 (m, 4H), 1.68 (dq, J = 11.5, 6.4 Hz, 4H), 1.58–1.49 (m, 2H). ¹³C NMR (101 MHz, CDCl₃) δ 170.33, 149.17, 148.99, 148.26, 147.14, 146.62, 129.38, 128.04, 126.93, 125.83, 122.48, 121.66, 120.56, 113.26, 111.84, 111.15, 110.50, 70.79, 65.08, 58.74, 55.93, 55.88, 55.87, 55.58, 53.40, 46.55, 37.60, 27.73, 27.60, 23.27, 22.48. HRMS (ESI-TOF) cald. for C₅₃H₇₂N₂O₁₂ ²⁺ [M]²⁺ 928.5074, found: 464.25537.

 

MissingLinkText

 

MissingLinkText

Supplementary Material

HPLC, ¹H NMR, ¹³C NMR spectra, and HRMS results of compounds 1, R-6, 9, and (R-cis)-10, as well as chiral HPLC results for R-6 and 9, are included in the [Supplementary Figs. S1–S18] (available in online version).

 

MissingLinkText

 

MissingLinkText

Conflict of Interest

None declared.

^# These authors contributed equally to this work.

• References

• 1 Lien CA, Belmont MR, Abalos A. et al. The cardiovascular effects and histamine-releasing properties of 51W89 in patients receiving nitrous oxide/opioid/barbiturate anesthesia. Anesthesiology 1995; 82 (05) 1131-1138
• 2 Belmont MR, Lien CA, Quessy S. et al. The clinical neuromuscular pharmacology of 51W89 in patients receiving nitrous oxide/opioid/barbiturate anesthesia. Anesthesiology 1995; 82 (05) 1139-1145
• 3 Wastila WB, Maehr RB, Turner GL, Hill DA, Savarese JJ. Comparative pharmacology of cisatracurium (51W89), atracurium, and five isomers in cats. Anesthesiology 1996; 85 (01) 169-177
• 4 Sparr HJ, Beaufort TM, Fuchs-Buder T. Newer neuromuscular blocking agents: How do they compare with established agents?. Drugs 2001; 61 (07) 919-942
• 5 Welch RM, Brown A, Ravitch J, Dahl R. The in vitro degradation of cisatracurium, the R, cis-R'-isomer of atracurium, in human and rat plasma. Clin Pharmacol Ther 1995; 58 (02) 132-142
• 6 Zhang H, Wang P, Bartlett MG, Stewart JT. HPLC determination of cisatracurium besylate and propofol mixtures with LC-MS identification of degradation products. J Pharm Biomed Anal 1998; 16 (07) 1241-1249
• 7 Hill DA, Turner GL. Neuromuscular blocking agents. WO Patent 1992000965A1. January 23, 1992
• 8 Noyori R, Ohta M, Hsiao Y. et al. Asymmetric synthesis of isoquinoline alkaloids by homogeneous catalysis. J Am Chem Soc 1986; 108: 7117-7119
• 9 Piwowarczyk K, Zawadzka A, Roszkowski P. et al. Enantiomers of (2R*,3R*)-1-methyl-5-oxo-2-phenyltetrahydro-1H-pyrrolidine-3-carboxylic acid as novel chiral resolving agents. Tetrahedron Asymmetry 2008; 19: 309-317
• 10 Kitamura K, Hsiao Y, Ohta M. et al. General asymmetric synthesis of isoquinoline alkaloids. enantioselective hydrogenation of enamides catalyzed by BINAP-Ruthenium(II) complexes. J Org Chem 1994; 59: 297-310
• 11 Nobuyuki U, Akio F, Shohei H. et al. Asymmetric transfer hydrogenation of imines. J Am Chem Soc 1996; 118 (20) 4916-4917
• 12 Kagan HB, Langlois N, Phat Dang TP. Reduction asymetrique catalysee par des complexes de metaux de transition IV. synthese d'amines chirales au moyen d'un complexe de rhodium et d'isopropylidene dihydroxy-2,3 bis(diphenylphosphino)-1,4 butane (diop) [in French]. J Organomet Chem 1975; 90: 353-365
• 13 Zhong M, Jiang Y, Chen Y. et al. Asymmetric total synthesis of (S)-isocorydine. Tetrahedron Asymmetry 2015; 26 (20) 1145-1149
• 14 Luo Z, Sun G, Wang G. et al. Efficient and scalable enantioselective synthesis of a key intermediate for rimegepant: an oral CGRP receptor antagonist. Pharmaceut Fronts 2024; 6: e62-e68
• 15 Qin Y, Fu L, Su L. et al. Preparation method of benzyl tetrahydro isoquinoline compound [in Chinese]. CN Patent 107778236 A. March 9, 2018
• 16 Peng P, Zheng Z, Yu J. et al. Practical process for synthesizing R-tetrahydropapaverine—a key intermediate of cisatracurium besylate (Nimbex). Org Process Res Dev 2022; 26 (11) 3106-3114

Publication History

Received: 25 March 2025

Accepted: 31 July 2025

Article published online:
21 August 2025