- Email: [email protected]
Stereoselective separation of β-adrenergic blocking agents containing two chiral centers by countercurrent chromatography
Journal of Chromatography A, 1513 (2017) 235–244
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Short communication
Stereoselective separation of -adrenergic blocking agents containing two chiral centers by countercurrent chromatography Liqiong Lv, Zhisi Bu, Mengxia Lu, Xiaoping Wang, Jizhong Yan, Shengqiang Tong ∗ College of Pharmaceutical Science, Zhejiang University of Technology, Hangzhou, 310032, China
a r t i c l e
i n f o
Article history: Received 23 March 2017 Received in revised form 11 July 2017 Accepted 13 July 2017 Available online 14 July 2017 Keywords: -Adrenergic blocking agents Countercurrent chromatography Di-n-hexyl l-tartrate Preparative stereoselective separation
a b s t r a c t Four -adrenergic blocking agents, including 1-[(1-methylethyl)amino]-3-phenoxy-2-propanol (1), 1-[(1-methylethyl)amino]-3-(3-methylphenoxy)-2-propanol (2), 1,1 -[1,4-phenylenebis(oxy)] bis[3-[(1-methylethyl)amino]-2-propanol (3) and 1,1 -[(4-methyl-1,2-phenylene)bis(oxy)]bis[3[(1-methylethyl)amino]-2-propanol (4), were stereoselectively separated by countercurrent chromatography using di-n-hexyl l-tartrate and boric acid as chiral selector. The compounds (3) and (4) have four optical isomers since they contained two chiral centers. A two-phase solvent system composed of chloroform-0.05 mol L−1 of acetate buffer containing 0.10 mol L−1 of boric acid (1:1, v/v) was selected, in which 0.10 mol L−1 of di-n-hexyl l-tartrate was added in the organic phase as chiral selector. 20–42 mg of each racemate was stereoselectively separated by countercurrent chromatography in a single run with high purity of 96–98%, and the recovery of each separated compound reached around 87–93%. This is the first time report on successful stereoselective separation of optical isomeric compounds containing two chiral centers by countercurrent chromatography. At the same time, a chiral stationary phase was screened for analytical stereoselective separation of compounds (3) and (4) by high performance liquid chromatography. © 2017 Elsevier B.V. All rights reserved.
1. Introduction -Adrenergic blocking agents have been widely used for the treatment of hypertension for the past 50 years, and continue to be recommended as a mainstay of therapy in many national guidelines. They have also been used in a variety of cardiovascular conditions commonly complicating hypertension, including angina pectoris, myocardial infarction, acute and chronic heart failure, as well as conditions like essential tremor and migraine [1,2]. Most of the -adrenergic blocking agents have one chiral carbon and so they have at least two optical isomers. As shown in Fig. 1, compound (1) and (2) are -adrenergic blocking agents with one chiral center. Compound (2) was also known under the name of (±)-toliprolol. In order to investigate the structure-activity relationship, two additional -adrenergic blocking agents containing two chiral centers, compound (3) and compound (4), were synthesized in our lab. Big difference in pharmacologic activities might be found between enantiomers and diastereoisomers due to the stere-
∗ Corresponding author at: College of Pharmaceutical Science, Zhejiang University of Technology, Chaowang Road 18, Chaohui District 6, Hangzhou, China. E-mail addresses: [email protected], [email protected] (S. Tong). http://dx.doi.org/10.1016/j.chroma.2017.07.051 0021-9673/© 2017 Elsevier B.V. All rights reserved.
ospecific characteristics of chemical structures, which necessitate a method for preparative stereoselective separation. No literature about preparative stereoselective separation of -adrenergic blocking agents containing two chiral centers are available, though quite a few of literature could be found which is about analytical enantioseparation of -blockers with one chiral center [3–9]. In our previous work, countercurrent chromatography and pH-zone-refining countercurrent chromatography were successfully used for enantioseparation of three -adrenergic blocking agents containing one chiral centers, propranolol, pindolol and alprenolol, based on borate coordination complex when di-n-hexyl l-tartrate was selected as chiral ligand [10]. Herein we want to report our recent study on stereoselective separation of -adrenergic blocking agents with two chiral centers by countercurrent chromatography. Meanwhile, a chiral column was selected for analytical stereoselective separation of the two synthesized adrenergic blocking agents containing two chiral centers. Drugs with more than one chiral center are of high medicinal values. However, the stereoselective separation of such type of racemate by various kinds of chromatographic technique is still a very challenging task. There is a limited number of scientific reports about stereoselective separation of racemate with more than one
236
L. Lv et al. / J. Chromatogr. A 1513 (2017) 235–244
hai Jinda Biotechnology Co., Ltd., Shanghai, China), and SEPU3000 workstation (Hangzhou Puhui Technology, Hangzhou, China) was employed to record the chromatogram. The high performance liquid chromatography (HPLC) used was a CLASS-VP Ver.6.1 system (Shimadzu, Japan) comprised of a Shimadzu SPD10Avp UV detector, a Shimadzu LC-10ATvp Multisolvent Delivery System, a Shimadzu SCL-10Avp controller, a Shimadzu LC pump, and a CLASS-VP Ver.6.1 workstation. The pH value was determined with a PB-10 pH meter (Sartorius, Germany). 2.2. Reagents l-Tartaric acid was purchased from Lanxi Shengda tartaric acid limited company, Zhejiang, China. Glycidyl phenyl ether and potassium hexafluorophosphate were purchased from J&K chemical scientific Co., Ltd, Shanghai, China. Hydroquinone, m-cresol, 4-methylcatechol, epichlorohydrin, isopropyl amine, triethylbenzylammonium chloride (TEBA) and boric acid were purchased from Huipu Chemical, Hangzhou, China. All organic solvents used for countercurrent chromatography were of analytical grade. Acetonitrile, methyl tert-butyl ether, n-hexane and ethanol used for HPLC analysis were of chromatographic grade. n-Butyl l-tartrate, isobutyl l-tartrate, n-hexyl l-tartrate, n-octyl l-tartrate and isooctyl l-tartrate were prepared according to the literature [21], and their structures were confirmed by 1 H NMR. 2.3. Preparation of ˇ-adrenergic blocking agents Fig. 1. Chemical structures of four synthesized -adrenergic blocking agents. (1) 1-[(1-methylethyl)amino]-3-phenoxy-2-propanol. (2) 1-[(1-methylethyl)amino]-3-(3-methylphenoxy)-2-propanol. (3) 1,1 -[1,4-phenylenebis(oxy)]bis[3-[(1-methylethyl)amino]-2-propanol. 1,1 -[(4-methyl-1,2-phenylene)bis(oxy)]bis[3-[(1-methylethyl)amino]-2(4) propanol.
chiral center by chromatography and capillary electrophoresis [11–17]. Countercurrent chromatography has been widely used for separation of chemical components from natural products [18]. However, much smaller number of literature concerning enantioseparations and stereoselective separations by countercurrent chromatography is available compared with traditional separation methods because it is difficult to find a suitable biphasic solvent system along with a chiral selector with high enantiorecognition [19]. All the applications of countercurrent chromatography in stereoselective separations in the past decades are about enantioseparation of racemates with one chiral center. Therefore, this is the first time report on successful stereoselective separation of racemates with two chiral centers by countercurrent chromatography. 2. Experimental section 2.1. Apparatus A model of TBE-200V preparative multilayer coil planet centrifuges (Shanghai Tauto Biotechnique, Shanghai, China) was used in the present work, each equipped with a set of three multilayer coils. The parameters for this apparatus have been described in our previous literature [20]. The separation columns were installed in a vessel that maintains column temperature by a model SDC6 constant-temperature controller (Ningbo Scientz Biotechnology Co. Ltd., Ningbo, China). The solvents were pumped into the column with a model TBP 5002 constant-flow pump (Shanghai Tauto Biotechnique, Shanghai, China). Continuous monitoring of the effluent was achieved with a model UVD-200 detector (Shang-
Compounds (1)–(4) were prepared according to the literature [22] and their chemical structures were determined by 1 H NMR. Compound (1): A solution of 2 g (13 mmol) of glycidyl phenyl ether and 3 mL (35 mmol) of isopropyl amine was stirred and refluxed at 50 ◦ C for 10 h. Excess amine was evaporated under reduced pressure, and it was further purified by column chromatography (dichloromethane: methanol = 95:5, v/v), yielding 2.15 g (79.13%) of compounds (1). Compound (2): A solution of 7.56 g (0.19 mol) of sodium hydroxide dissolved in 20 mL of water and 5 mL (0.048 mol) of m-cresol was stirred at room temperature for 40 min, and 0.5 g of triethylbenzylammonium chloride (TEBA) was added as phase transfer catalyst in the mixture and 15.04 mL of (0.192 mol) epichlorohydrin was added in 30 min. After stirring at room temperature for 5 h, the aqueous layer was extracted twice with ether, and the combined organic layers were washed, concentrated and purified by silica column chromatography (ethyl acetate: petroleum ether = 1:24, v/v), yielding 2.5 g (31.76%) of intermediate product. A solution of 2.5 g (15 mmol) of intermediate and 3 mL (35 mmol) of isopropyl amine was stirred and refluxed at 50 ◦ C for 14 h. Excess amine was evaporated under reduced pressure, and it was further purified by silica column chromatography (dichloromethane: methanol = 95:5, v/v), yielding 1.1645 g (34.26%) of compounds (2). Compound (3): A solution of 11.0 g (0.1 mol) of hydroquinone, 37.0 g (0.4 mol) of epichlorohydrin, and 0.4 mL of 10 mol L−1 sodium hydroxide was stirred under nitrogen gas under 40 ◦ C for 48 h. After cooling, 42.0 mL of 5 mol L−1 sodium hydroxide, saturated with sodium carbonate, was added and the mixture stirred vigorously at room temperature for 20 h. The aqueous layer was extracted twice with chloroform and the combined organic layers were washed, concentrated and purified by silica column chromatography (ethyl acetate: petroleum ether = 1:6, v/v), yielding 0.6 g (2.70%) of intermediate product. A solution of 0.6 g (2.7 mmol) intermediate, 4 mL (50 mmol) of isopropyl amine and 20 mL of absolute ethanol was stirred and refluxed at 50 ◦ C for 8 h. Excess amine was evaporated under reduced pressure, and it was further purified by silica column
L. Lv et al. / J. Chromatogr. A 1513 (2017) 235–244
Fig. 2. Chromatogram of HPLC analyses of racemic compound (1) and (2). HPLC conditions: chiral column: Chiralcel OD-R, 10 m particle size of the packing material (250 mm × 6 mm I.D.); mobile phase: 0.2 mol L−1 potassium hexafluorophosphate aqueous solution: acetonitrile (60:40, v/v); flow rate: 0.6 mL min−1 ; UV wavelength: 254 nm; column temperature: 30 ◦ C.
chromatography (dichloromethane: methanol = 19:2, v/v), yielding 0.6 g (65.30%) of compounds (3). Compound (4): A solution of 12.4 g (0.1 mol) of 4methylcatechol, 37.0 g (0.4 mol) of epichlorohydrin, and 0.4 mL of 10 mol L−1 sodium hydroxide was stirred under nitrogen gas under 40 ◦ C for 48 h. After cooling, 42.0 mL of 5 mol L−1 sodium hydroxide, saturated with sodium carbonate, was added and the mixture stirred vigorously at room temperature for 20 h. The organic layer was washed, concentrated and purified by silica column chromatography (ethyl acetate: petroleum ether = 1:5, v/v), yielding 1.82 g (7.71%) of intermediate product. A solution of 1.82 g of (7.7 mmol) intermediate, 5 mL (58.7 mmol) of isopropyl amine and 15 mL of toluene was stirred and refluxed at 80 ◦ C for 8 h. Excess amine was evaporated under reduced pressure, and it was further purified by silica column chromatography (dichloromethane: methanol = 19:2, v/v), yielding 0.6 g (21.98%) of compounds (4). 2.4. Enantioselective liquid-liquid extractions The enantioselective liquid-liquid extraction of compound (1)–(3) was conducted to determine the distribution ratio and to investigate the effects of influence factors on distribution ratio and stereoselective separation factor. The aqueous phases were pre-
237
Fig. 3. Chromatogram of HPLC analyses of racemic compound (3) and (4). HPLC conditions: chiral column: Chiralpak IE, 5 m particle size of the packing material (250 mm × 4.6 mm I.D.); mobile phase: methyl tert-butyl ether: hexane: ethanol: diethylamine (60:30:10:0.1 for compound (3) and 60:39:1:0.1 for compound (4), v/v/v/v); flow rate: 0.5 mL min−1 ; UV wavelength: 289 nm; column temperature: 35 ◦ C. Table 1 Effect of L-tartrates on the distribution ratio (D) and enantioseparation factor (␣) of racemic compound (1). Chiral selectors
D−
D+
␣
di-n-butyl l-tartrate di-isobutyl l-tartrate di-n-hexyl l-tartrate di-n-octyl l-tartrate di-isooctyl l-tartrate
0.325 0.344 0.387 0.417 0.508
0.803 0.849 0.981 1.045 1.210
2.473 2.466 2.534 2.506 2.383
Aqueous phases: 0.05 mol L−1 acetate buffer pH = 6.1 containing 0.10 mol L−1 boric acid and 1.0 mmol L−1 racemic compound (1); organic phase: chloroform solution of 0.10 mol L−1 l-tartrate; equilibration temperature: 10 ◦ C.
pared by dissolving 0.10 mol L−1 of boric acid in a 0.05 mol L−1 of acetate buffer with pH 3.6–6.9, in which 1 mmol L−1 of racemic compound was added. The organic phase chloroform contains 0.10 mol L−1 of dialkyl l-tartrate. The equilibrium experiments were performed in a 10 mL glass-stoppered tube. 2.0 mL of organic phase and 2.0 mL of aqueous phase were placed in the stoppered glass tube and shaken vigorously for 4 times, 2 min-shaking for each time, and 30 min-settling down in a water bath at a constant temperature to reach phase equilibrium. The investigated temperature was in the range of 5–30 ◦ C. After phase separation, the concentration of enantiomer in the aqueous phase was determined by HPLC, and the concentration of enantiomer in the organic phase was cal-
238
L. Lv et al. / J. Chromatogr. A 1513 (2017) 235–244
Fig. 4. Effect of concentration of di-n-hexyl l-tartrate and boric acid on the distribution ratio (D) and enantioseparation factor (␣) for compound (1). Organic phase: (a)–(a’) chloroform added with different concentration of di-n-hexyl l-tartrate; (b)–(b’) chloroform added with 0.10 mol L−1 of di-n-hexyl l-tartrate; aqueous phase: 0.05 mol L−1 of acetate buffer at pH6.1 containing (a)–(a’) 0.10 mol L−1 boric acid; (b)–(b’) different concentration of boric acid; 1 mmol L−1 of racemate dissolved in the aqueous phase, and temperature was 10 ◦ C. D+ = 13.879, D− = 15.713 and ␣ = 1.132 when the concentration of di-n-hexyl l-tartrate was 0.00 mol L−1 , and D+ = 7.673, D− = 6.664 and ␣ = 1.152 when the concentration of boric acid was 0.00 mol L−1 , which are not shown in the figure.
culated by the subtraction method. The distribution ratio (D) of the enantiomers was expressed as the total concentration of analytes in the upper phase divided by the total concentration of analytes in the lower phase, and the separation factor was obtained with the following equation: ˛=
D2 D1
(1)
where D2 ≥ D1 . 2.5. Preparation of biphasic solvent systems and sample solutions Biphasic solvent systems consisting of chloroform added with 0.10 mol L−1 of di-n-hexyl l-tartrate and 0.05 mol L−1 of acetate buffer added with 0.10 mol L−1 boric acid (1:1, v/v) were used. The solvent mixture was thoroughly equilibrated in a separatory funnel, and the two phases were separated shortly before use. The sample solutions were prepared as follows: 20–42 mg of racemate
was dissolved in 9 mL of the aqueous phase for each stereoselective separation. 2.6. Countercurrent chromatography procedure The countercurrent chromatography was conducted with headto-tail elution mode. All separations were initiated by filling the column with the organic stationary phase. The aqueous mobile phases were pumped into the column while the column was rotated at 800 rpm (counter clockwise). Each of the sample solution was injected after the hydrodynamic equilibrium was reached, as indicated by a clear mobile phase eluting at the outlet. 2.7. Analytical method Chiral separation of compound (1) and (2) was performed by HPLC in reverse phase mode on a column of Chiralcel OD-R, with 10 m particle size of the packing material, 250 mm × 4.6 mm I.D.
L. Lv et al. / J. Chromatogr. A 1513 (2017) 235–244
239
Fig. 5. The distribution ratio (D) and enantioseparation factor (␣) of compound (1) under different pH and temperature. (a)–(a’) Organic phase: chloroform containing 0.10 mol L−1 of di-n-hexyl l-tartrate, Aqueous phase: 0.05 mol L−1 of acetate buffer containing 0.10 mol L−1 of boric acid and 1 mmol L−1 of racemate; temperature 10 ◦ C; (b)–(b’) Organic phase: chloroform added with 0.10 mol L−1 of di-n-hexyl l-tartrate, Aqueous phase: 0.05 mol L−1 of acetate buffer at pH6.1 containing 0.10 mol L−1 of boric acid and 1 mmol L−1 of compound (1). D+ = 55.731, D− = 37.458 and ␣ = 1.488 when the pH value of aqueous phase was 3.6, which is not shown in the figure.
(Daicel Chiral Technologies Co., Ltd., Shanghai, China). The UV detector was set at 254 nm. The mobile phase was composed of 0.2 mol L−1 of potassium hexafluorophosphate aqueous solution: acetonitrile (60:40, v/v) at a flow rate of 0.6 mL min−1 . The column temperature was 30 ◦ C. Analytical stereoselective separation of compound (3) and (4) was conducted on a normal phase Chiralpak IE, with 5 m particle size of the packing material, 250 mm × 4.6 mm I.D. (Daicel Chiral Table 2 The distribution ratio (D) and seteroselective separation factor (␣) for compound (1)-(4) under optimized separation condition. racemates
pH
D − /D −,−
D +,−or−,+
D + /D +,+
␣
(1) (2) (3) (4)
6.1 5.3 7.0 7.0
0.387 0.407 0.550 0.111
– – 1.147 0.208
0.981 1.009 2.538 0.500
2.534 2.479 2.085, 2.213 1.874, 2.404
Aqueous phases: 0.10 mol L−1 boric acid aqueous with different pH values and 1.0 mmol L−1 racemic compound (1); organic phase: chloroform solution of 0.10 mol L−1 l-tartrate; equilibration temperature: 10 ◦ C.
Technologies Co., Ltd., Shanghai, China).The UV detector was set at 289 nm. The mobile phase was composed of methyl tert-butyl ether: n-hexane: ethanol: diethylamine (60:30:10:0.1 for compound (3) and 60:39:1:0.1 for compound (4), v/v/v/v) at a flow rate of 0.5 mL min−1 . The column temperature was 35 ◦ C.
3. Results and discussion 3.1. Development of analytical HPLC methods As known, the methods for enantioseparation of -adrenergic blocking agents containing one chiral center by liquid chromatography and capillary electrophoresis have been well documented. The reverse phase Chiralcel OD-R was proven to be a very effective chiral stationary phase for enantioseparation of some typical -blocker drugs, such as propranolol, pindolol, alprenolol and propafenone. It was found that racemic compounds (1) and (2) could be well enantioseparated by Chiralcel OD-R column, as shown in Fig. 2. However, it was difficult to separate stereoselectively com-
240
L. Lv et al. / J. Chromatogr. A 1513 (2017) 235–244
Fig. 6. Chromatogram of enantioseparation of racemic compound (1) and (2) by countercurrent chromatography. Biphasic solvent system: chloroform: 0.05 mol L−1 of acetate buffer pH6.1 for compound (1) and pH5.3 for compound (2) (1:1, v/v), in which organic phase contained 0.10 mol L−1 of di-n-hexyl l-tartrate and aqueous phase contained 0.10 mol L−1 of boric acid; stationary phase: lower organic phase; mobile phase: upper aqueous phase; sample solution: 20 mg of compound (1) and 30 mg of compound (2) dissolved in 10 mL of the aqueous mobile phase, respectively; flow rate: 3.0 mL min−1 for (1) and 2.0 mL min−1 for (2); revolution: 800 rpm; column temperature: 10 ◦ C; stationary phase retention: 50% for (1) and 60% for (2).
pounds (3) and (4) by the same chiral stationary phase Chiralcel OD-R. Various kinds of chiral stationary phases as well as chiral mobile phase additives had been tried in our lab and only the chiral stationary phase Chiralpak IE column, normal phase, was found to be suitable for analytical stereoselective separation of compounds (3) and (4). Fig. 3 was a typical chromatogram for stereoselective separation of compounds (3) and (4) by high performance liquid chromatography with Chiralpak IE as chiral stationary phase. The mobile phase was composed of methyl tert-butyl ether: hexane: ethanol: diethylamine with a volume ratio of 60:30:10:0.1 for compound (3) and 60:39:1:0.1 for compound (4). Complete stereoselective separation of compound (3) (two enantiomers plus a meso compound) could be achieved, while only partial stereoselective separation of compound (4) (two pairs of enantiomers) was obtained after optimization of chromatographic conditions, as shown in Fig. 3.
3.2. Stereoselective separation by countercurrent chromatography 3.2.1. Optimization of separation conditions Stereoselective separation of the present -adrenergic blocking agents was based on a stereospecific ternary borate coordination
complex formed by enantiomer, l-tartrate and boric acid. Separation factor could be greatly affected by the following factors: type of chiral selector, concentration of di-n-hexyl l-tartrate, concentration of boric acid, pH value of aqueous buffer and temperature. Therefore, they were investigated by enantioselective liquid-liquid extraction with regarding to distribution performance of compound (1). Table 1 shows the distribution ratio and enantioseparation factor of compound (1) under different type of l-tartrate (chiral selector). No much difference was found among all the chiral selectors investigated. However, di-n-hexyl l-tartrate shows a slightly higher enantioseparation factor than the others. So di-n-hexyl ltartrate was selected in the present work. Fig. 4(a) and (a’) shows the distribution ratio and enantioseparation factor under various concentration of di-n-hexyl l-tartrate, respectively. With an increase of the concentration of di-n-hexyl l-tartrate, the distribution ratio decreased greatly, while the enantioseparation factor increased up to a constant when the concentration of di-n-hexyl l-tartrate was around 0.20 mol L−1 . This was caused by the fact that an improved coordination complex would be formed with an increasing concentration of l-tartrate. Very low distribution ratio was found with the concentration of di-n-hexyl l-tartrate being greater than 0.20 mol L−1 , which indicated very long retention time
L. Lv et al. / J. Chromatogr. A 1513 (2017) 235–244
241
Fig. 7. Chromatogram of stereoselective separation of compound (3) and (4) by countercurrent chromatography. Biphasic solvent system: chloroform contained 0.10 mol L−1 of di-n-hexyl l-tartrate: 0.10 mol L−1 of boric acid at pH7.0 adjusted with triethylamine (1:1, v/v); stationary phase: lower organic phase; mobile phase: upper aqueous phase; sample solution: 40 mg of compound (3) and 42 mg of compound (4) dissolved in 10 mL of the aqueous phase individually; flow rate: 2.0 mL min−1 ; revolution: 800 rpm; column temperature: 10 ◦ C; stationary phase retention: 60%.
was necessary if the upper aqueous phase was used as the mobile phase. So the concentration of di-n-hexyl l-tartrate at 0.10 mol L−1 was selected since a suitable distribution ratio could be obtained. Fig. 4(b) and (b’) shows the distribution ratio and enantioseparation factor under various concentration of boric acid in the aqueous phase, respectively. Boric acid was necessary for the formation of a stereospecific ternary borate coordination complex among enantiomer, boric acid and dialkyl l-tartrate. Enantioseparation factor completely lost when there was no boric acid in the solvent system and most of racemic solutes would partition into the aqueous solution. When the concentration of boric acid was in the range of 0.1-0.2 mol L−1 , suitable distribution ratio and a high enantioseparation factor could be obtained. The effects on distribution ratio and enantioseparation factor by pH of aqueous phase as well as temperature were shown in Fig. 5. Fig. 5(a) indicated distribution ratio decreased sharply with the pH of the aqueous phase increased from pH = 4.5 to pH = 6.8, while the enantioseparation factor reached a highest value when pH was increased to pH = 5.4 and then it leveled off, as shown in Fig. 5(a’). This phenomenon could be explained by the fact that the compound (1) is ionized with low pH in the aqueous phase due to the amino alcohol group of its chemical structure. So pH = 6.1
was selected in light of suitable distribution ratio along with relatively higher enantioseparation factor. Fig. 5(b) shows the influence of temperature in the distribution behavior, which indicated that a slight increase in the distribution ratio of (−)-enantiomer was observed while a slight decrease in the distribution ratio of (+)enantiomer was observed with the increasing of the temperature within the range5–30 ◦ C. As a result, it led to a decrease of the enantioseparation factor ␣ (Fig. 5(b’)). The enantioseparation factor decreased greatly when the temperature was over 20 ◦ C. Therefore, the temperature ranged within 5–15 ◦ C was selected. As a result, the optimized separation condition for compound (1) was as followings: the biphasic solvent system was composed of chloroform and 0.05 mol L−1 acetate buffer pH6.1 (1:1, v/v), in which organic phase contained 0.10 mol L−1 di-n-hexyl l-tartrate and aqueous phase contained 0.10 mol L−1 boric acid. And the separation temperature was 10 ◦ C. The distribution ratio and separation factors of compound (1)–(4) under optimized separation conditions were summarized in Table 2. However, it was difficult to determine the distribution ratio for each of the four stereoisomers of compound (4) due to their partial resolution in HPLC analysis (as shown in Fig. 3, bottom). Therefore, the distribution ratio of compound (4) listed in Table 2 were calculated by the following Eq. (2) according
242
L. Lv et al. / J. Chromatogr. A 1513 (2017) 235–244
Fig. 8. Chromatogram of HPLC analyses of collected fractions from stereoselective separation of compound (3) and (4) by countercurrent chromatography. (a): fraction containing (+, +)-compound (3); (b): fraction containing (+, −) or (−, +)-compound (3); (c): fraction containing (−, −)-compound (3); (a’): fraction containing (+, +)-compound (4); (b’): fraction containing (+, −) or (−, +)-compound (4); (c’): fraction containing (−, −)-compound (4); HPLC conditions: chiral column: Chiralpak IE, 5 m particle size of the packing material (250 mm × 4.6 mm I.D.); mobile phase: methyl tert-butyl ether: hexane: ethanol: diethylamine 60:30:10:0.1 for (3) and 60:39:1:0.1 for (4), v/v/v/v; flow rate: 0.5 mL min−1 , UV wavelength: 289 nm; column temperature: 35 ◦ C.
L. Lv et al. / J. Chromatogr. A 1513 (2017) 235–244
to the chromatogram of stereoselective separation of compound (4) by countercurrent chromatography. D=
VR − Vm Vs
(2)
where VR represents the elution volume of analyte, Vm and Vs are the volume of mobile phase and volume of stationary phase in the separation column. 3.2.2. Stereoselective separation by countercurrent chromatography Enantioseparation of compound (1) and (2) with one chiral center by TBE–200 V apparatus was conducted using the two-phase solvent system composed of chloroform and 0.05 mol L−1 acetate buffer (1:1, v/v), in which organic phase contained 0.10 mol L−1 of di-n-hexyl l-tartrate and aqueous phase contained 0.10 mol L−1 of boric acid. The only difference in separation conditions between compounds (1) and (2) was the pH value of aqueous phase (at pH6.1 for compound (1) and pH5.3 for compound (2)). A relatively large lipophilicity of compound (2) was observed compared with that of compound (1), which was mainly caused by the methyl group on the benzene ring. Suitable distribution ratio of compound (2) could be obtained by decreasing the pH value of aqueous phase. So pH5.3 was selected for enantioseparation of compound (2). Fig. 6 shows the typical chromatogram of enantioseparation of compound (1) and (2) by TBE–200V countercurrent chromatographic apparatus. As shown in Fig. 6, complete resolution of two racemates was achieved. HPLC results demonstrated that the purity of the enantiomer collected from the elution of countercurrent chromatography reached more than 98%. Fig. 7 shows the typical chromatogram of stereoselective separation of compound (3) and (4) with two chiral centers by TBE-200V countercurrent chromatographic apparatus. The biphasic solvent system used here was the same as that used in enantioseparation of compound (1) and (2) except for different pH value of aqueous phase. Compounds (3) and (4) tend to partition into aqueous phase with the additional substitution of amino alcohol group on the benzene ring. Stereoisomer of compounds (3) and (4) would be easily eluted out with the aqueous phase as the mobile phase in countercurrent chromatography. In order to improve its retention factor in the separation column, pH of aqueous phase was adjusted to around 7.0. As shown in Fig. 7, three main fractions were obtained after stereoselective separation by countercurrent chromatography with single run. Each fraction was collected according to the peak profile and analyzed by chiral HPLC. Fig. 8 shows the HPLC analysis of each fraction collected from stereoselective separation by countercurrent chromatography. Results showed that as for compound (3), (+,+)-enantiomer was eluted from the first peak and (−,−)enantiomer was eluted from the third peak, while (+,−) and (−,+)-mesomer were eluted in the second peak. The same elution performance was also observed as for (+,+)-stereoisomer and (−,−)stereoisomer of compound (4) and the diastereoisomers, (+,−) and (−,+)-compound (4), were eluted between (+,+)-stereoisomer and (−,−)-stereoisomer, which could not be separated by countercurrent chromatography. Diastereoisomers may exhibit significant differences in physical properties, such as solubility, partition coefficient, and melting point. Differences in biologic activity between diasteroisomers may, therefore, be due to differences in physical properties. It would still be difficult for countercurrent chromatography to separate the pair of diastereoisomer of compound (4) due to their extremely similarity of chemical structure. Therefore, only three main peaks were obtained, as shown in Fig. 7. It was noticed that the elution sequence of (+,+)stereoisomer and (−,−)-stereoisomer were completely reversed
243
compared with that of HPLC analysis as for compound (3). Meanwhile, the diastereoisomeric (+,−) and (−,+)-compound (4) was eluted before (−,−)-enantiomer in HPLC chromatogram. Recovery of enantiomers, mesomer or diastereoisomer of compound (1)–(4) was around 87–93% with their purity of 96–98%, as determined by HPLC. 4. Conclusion A method for stereoselective separation of four -adrenergic blocking agents by countercurrent chromatography using borate coordination complex, in which two of them owning two chiral centers, was attempted for the first time. A two-phase solvent system was selected and separation conditions were optimized. 20–42 mg of each racemic -adrenergic blocking agent was separated by high speed countercurrent chromatography in a single run with 96–98% purity, and the recovery of each enantiomer reached 87–93%. The present research indicated that countercurrent chromatography might also be applied in stereoselective separation of components with two chiral centers. References [1] L. Poirier, Y. Lacourciere, The evolving role of - adrenergic receptor blockers in managing hypertension, Can. J. Cardiol. 28 (2012) 334–340. [2] J.M. Cruickshank,  blockers in hypertension, Lancet 376 (2010) 415–416. [3] Y.Z. Lin, J. Zhou, J. Tang, W.H. Tang, Cyclodextrin clicked chiral stationary phases with functionalities- tuned enantioseparations in high performance liquid chromatography, J. Chromatogr. A 1406 (2015) 342–346. [4] Y.N. Zou, L.J. Wang, Q. Liu, H.Y. Liu, F.N. Li, Enantioseparations of 11 amino alcohols using di- n- amyl l- tartrate- boric acid complex as chiral mobile phase additive by RP- HPLC, Chromatographia 78 (2015) 753–761. [5] X.L. Weng, Z.B. Bao, H.B. Xing, Z.G. Zhang, Q.W. Yang, B.G. Su, Y.W. Yang, Q.L. Ren, Synthesis and characterization of cellulose 3, 5- dimethylphenylcarbamate silica hybrid spheres for enantioseparation of chiral - blockers, J. Chromatogr. A 1321 (2013) 38–47. [6] I. Ali, A. Haque, M.F. Al Ajmi, A. Hussain, M. Marsin Sanagi, I. Hussain, H.Y. Aboul-Enein, Supramolecular chiro-biomedical aspect of -blockers in drug development, Curr. Drug Targets 15 (2014) 729–741. [7] I. Ali, S.D. Alam, J.A. Farooqi, N. Nagae, V.D. Gaitonde, H.Y. Aboul-Enein, A comparison of -blockers separation on C18 and new generation C28 columns in human plasma, Anal. Methods 5 (2013) 3523–3529. [8] I. Ali, Z.A. Al-Othman, A. Hussain, K. Saleem, H.Y. Aboul-Enein, Chiral separation of -adrenergic blockers in human plasma by SPE-HPLC, Chromatographia 73 (2011) 251–256. [9] I. Ali, V.D. Gaitonde, H.Y. Aboul-Enein, A. Hussain, Chiral separation of -adrenergic blockers on cellucoat column by HPLC, Talanta 78 (2009) 458–463. [10] S.Q. Tong, Y. Zheng, J.Z. Yan, Y.X. Guan, C.Y. Wu, W.Y. Lei, Preparative enantioseparation of -blocker drugs by counter-current chromatography using dialkyl l-tartrate as chiral selector based on borate coordination complex, J. Chromatogr. A 1263 (2012) 74–83. [11] I. Ali, M. Suhail, M.N. Lone, Z.A. Al-Othman, A. Alwarthan, Chiral resolution of multichiral center racemates by different modalities of chromatography, J. Liq. Chromatogr. Relat. technol. 39 (2016) 435–444. [12] I. Ali, M. Suhail, Z.A. Al-Othman, A. Al-Warthan, H.Y. Aboul-Enein, Enantiomeric resolution of multiple chiral centres racemates by capillary electrophoresis, Biomed. Chromatogr. 30 (2016) 683–694. [13] Z.A. Al-Othman, A. Al-Warthan, S.D. Alam, I. Ali, Enantio- separation of drugs with multiple chiral centers by chromatography and capillary electrophoresis, Biomed. Chromatogr. 28 (2014) 1514–1524. [14] W. Bragg, D. Norton, S.A. Shamsi, Optimized separation of - blockers with multiple chiral centers using capillary electrochromatography- mass spectrometry, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 875 (2008) 304–316. [15] H.Y. Aboul-Enein, High- performance liquid chromatographic enantioseparation of drugs containing multiple chiral centers on polysaccharide- type chiral stationary phases, J. Chromatogr. A 906 (2001) 185–193. [16] H.Y. Aboul-Enein, I. Ali, Studies on the effect of alcohols on the chiral discrimination mechanisms of amylose stationary phase on the enantioseparation of nebivolol by HPLC, J. Biochem. Biophys. Methods 48 (2001) 175–188. [17] H.Y. Aboul-Enein, I. Ali, HPLC enantiomeric resolution of nebivolol on normal and reversed amylose based chiral phase, Pharmazie 56 (2001) 214–216. [18] M. Bojczuk, D. Zyzelewicz, P. Hodurek, Centrifugal partition chromatography – a review of recent applications and some classic references, J. Sep. Sci. 40 (2017) 1597–1609.
244
L. Lv et al. / J. Chromatogr. A 1513 (2017) 235–244
[19] X.Y. Huang, D.L. Di, Chiral separation by counter- current chromatography, TrAC Trends Anal. Chem. 67 (2015) 128–133. [20] S.Q. Tong, Y. Zheng, J.Z. Yan, Enantioseparation of chiral aromatic acids by multiple dual mode counter-current chromatography using hydroxypropyl--cyclodextrin as chiral selector, J. Sep. Sci. 36 (2013) 2035–2042.
[21] E. Heldin, K.-J. Lindner, C. Pettersson, W. Lindner, R. Rao, Tartaric acid derivatives as chiral selectors in liquid chromatography, Chromatographia 32 (1991) 407–416. [22] J. Zaagsma, W.Th. Nauta, -Adrenoceptor studies. 1. In vitro -adrenergic blocking, antiarrhythmic, and local anesthetic activities of a new series of aromatic bis(2-hydroxy-3-isopropylaminopropyl) ethers, J. Med. Chem. 17 (1974) 507–513.
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Short communication
Stereoselective separation of -adrenergic blocking agents containing two chiral centers by countercurrent chromatography Liqiong Lv, Zhisi Bu, Mengxia Lu, Xiaoping Wang, Jizhong Yan, Shengqiang Tong ∗ College of Pharmaceutical Science, Zhejiang University of Technology, Hangzhou, 310032, China
a r t i c l e
i n f o
Article history: Received 23 March 2017 Received in revised form 11 July 2017 Accepted 13 July 2017 Available online 14 July 2017 Keywords: -Adrenergic blocking agents Countercurrent chromatography Di-n-hexyl l-tartrate Preparative stereoselective separation
a b s t r a c t Four -adrenergic blocking agents, including 1-[(1-methylethyl)amino]-3-phenoxy-2-propanol (1), 1-[(1-methylethyl)amino]-3-(3-methylphenoxy)-2-propanol (2), 1,1 -[1,4-phenylenebis(oxy)] bis[3-[(1-methylethyl)amino]-2-propanol (3) and 1,1 -[(4-methyl-1,2-phenylene)bis(oxy)]bis[3[(1-methylethyl)amino]-2-propanol (4), were stereoselectively separated by countercurrent chromatography using di-n-hexyl l-tartrate and boric acid as chiral selector. The compounds (3) and (4) have four optical isomers since they contained two chiral centers. A two-phase solvent system composed of chloroform-0.05 mol L−1 of acetate buffer containing 0.10 mol L−1 of boric acid (1:1, v/v) was selected, in which 0.10 mol L−1 of di-n-hexyl l-tartrate was added in the organic phase as chiral selector. 20–42 mg of each racemate was stereoselectively separated by countercurrent chromatography in a single run with high purity of 96–98%, and the recovery of each separated compound reached around 87–93%. This is the first time report on successful stereoselective separation of optical isomeric compounds containing two chiral centers by countercurrent chromatography. At the same time, a chiral stationary phase was screened for analytical stereoselective separation of compounds (3) and (4) by high performance liquid chromatography. © 2017 Elsevier B.V. All rights reserved.
1. Introduction -Adrenergic blocking agents have been widely used for the treatment of hypertension for the past 50 years, and continue to be recommended as a mainstay of therapy in many national guidelines. They have also been used in a variety of cardiovascular conditions commonly complicating hypertension, including angina pectoris, myocardial infarction, acute and chronic heart failure, as well as conditions like essential tremor and migraine [1,2]. Most of the -adrenergic blocking agents have one chiral carbon and so they have at least two optical isomers. As shown in Fig. 1, compound (1) and (2) are -adrenergic blocking agents with one chiral center. Compound (2) was also known under the name of (±)-toliprolol. In order to investigate the structure-activity relationship, two additional -adrenergic blocking agents containing two chiral centers, compound (3) and compound (4), were synthesized in our lab. Big difference in pharmacologic activities might be found between enantiomers and diastereoisomers due to the stere-
∗ Corresponding author at: College of Pharmaceutical Science, Zhejiang University of Technology, Chaowang Road 18, Chaohui District 6, Hangzhou, China. E-mail addresses: [email protected], [email protected] (S. Tong). http://dx.doi.org/10.1016/j.chroma.2017.07.051 0021-9673/© 2017 Elsevier B.V. All rights reserved.
ospecific characteristics of chemical structures, which necessitate a method for preparative stereoselective separation. No literature about preparative stereoselective separation of -adrenergic blocking agents containing two chiral centers are available, though quite a few of literature could be found which is about analytical enantioseparation of -blockers with one chiral center [3–9]. In our previous work, countercurrent chromatography and pH-zone-refining countercurrent chromatography were successfully used for enantioseparation of three -adrenergic blocking agents containing one chiral centers, propranolol, pindolol and alprenolol, based on borate coordination complex when di-n-hexyl l-tartrate was selected as chiral ligand [10]. Herein we want to report our recent study on stereoselective separation of -adrenergic blocking agents with two chiral centers by countercurrent chromatography. Meanwhile, a chiral column was selected for analytical stereoselective separation of the two synthesized adrenergic blocking agents containing two chiral centers. Drugs with more than one chiral center are of high medicinal values. However, the stereoselective separation of such type of racemate by various kinds of chromatographic technique is still a very challenging task. There is a limited number of scientific reports about stereoselective separation of racemate with more than one
236
L. Lv et al. / J. Chromatogr. A 1513 (2017) 235–244
hai Jinda Biotechnology Co., Ltd., Shanghai, China), and SEPU3000 workstation (Hangzhou Puhui Technology, Hangzhou, China) was employed to record the chromatogram. The high performance liquid chromatography (HPLC) used was a CLASS-VP Ver.6.1 system (Shimadzu, Japan) comprised of a Shimadzu SPD10Avp UV detector, a Shimadzu LC-10ATvp Multisolvent Delivery System, a Shimadzu SCL-10Avp controller, a Shimadzu LC pump, and a CLASS-VP Ver.6.1 workstation. The pH value was determined with a PB-10 pH meter (Sartorius, Germany). 2.2. Reagents l-Tartaric acid was purchased from Lanxi Shengda tartaric acid limited company, Zhejiang, China. Glycidyl phenyl ether and potassium hexafluorophosphate were purchased from J&K chemical scientific Co., Ltd, Shanghai, China. Hydroquinone, m-cresol, 4-methylcatechol, epichlorohydrin, isopropyl amine, triethylbenzylammonium chloride (TEBA) and boric acid were purchased from Huipu Chemical, Hangzhou, China. All organic solvents used for countercurrent chromatography were of analytical grade. Acetonitrile, methyl tert-butyl ether, n-hexane and ethanol used for HPLC analysis were of chromatographic grade. n-Butyl l-tartrate, isobutyl l-tartrate, n-hexyl l-tartrate, n-octyl l-tartrate and isooctyl l-tartrate were prepared according to the literature [21], and their structures were confirmed by 1 H NMR. 2.3. Preparation of ˇ-adrenergic blocking agents Fig. 1. Chemical structures of four synthesized -adrenergic blocking agents. (1) 1-[(1-methylethyl)amino]-3-phenoxy-2-propanol. (2) 1-[(1-methylethyl)amino]-3-(3-methylphenoxy)-2-propanol. (3) 1,1 -[1,4-phenylenebis(oxy)]bis[3-[(1-methylethyl)amino]-2-propanol. 1,1 -[(4-methyl-1,2-phenylene)bis(oxy)]bis[3-[(1-methylethyl)amino]-2(4) propanol.
chiral center by chromatography and capillary electrophoresis [11–17]. Countercurrent chromatography has been widely used for separation of chemical components from natural products [18]. However, much smaller number of literature concerning enantioseparations and stereoselective separations by countercurrent chromatography is available compared with traditional separation methods because it is difficult to find a suitable biphasic solvent system along with a chiral selector with high enantiorecognition [19]. All the applications of countercurrent chromatography in stereoselective separations in the past decades are about enantioseparation of racemates with one chiral center. Therefore, this is the first time report on successful stereoselective separation of racemates with two chiral centers by countercurrent chromatography. 2. Experimental section 2.1. Apparatus A model of TBE-200V preparative multilayer coil planet centrifuges (Shanghai Tauto Biotechnique, Shanghai, China) was used in the present work, each equipped with a set of three multilayer coils. The parameters for this apparatus have been described in our previous literature [20]. The separation columns were installed in a vessel that maintains column temperature by a model SDC6 constant-temperature controller (Ningbo Scientz Biotechnology Co. Ltd., Ningbo, China). The solvents were pumped into the column with a model TBP 5002 constant-flow pump (Shanghai Tauto Biotechnique, Shanghai, China). Continuous monitoring of the effluent was achieved with a model UVD-200 detector (Shang-
Compounds (1)–(4) were prepared according to the literature [22] and their chemical structures were determined by 1 H NMR. Compound (1): A solution of 2 g (13 mmol) of glycidyl phenyl ether and 3 mL (35 mmol) of isopropyl amine was stirred and refluxed at 50 ◦ C for 10 h. Excess amine was evaporated under reduced pressure, and it was further purified by column chromatography (dichloromethane: methanol = 95:5, v/v), yielding 2.15 g (79.13%) of compounds (1). Compound (2): A solution of 7.56 g (0.19 mol) of sodium hydroxide dissolved in 20 mL of water and 5 mL (0.048 mol) of m-cresol was stirred at room temperature for 40 min, and 0.5 g of triethylbenzylammonium chloride (TEBA) was added as phase transfer catalyst in the mixture and 15.04 mL of (0.192 mol) epichlorohydrin was added in 30 min. After stirring at room temperature for 5 h, the aqueous layer was extracted twice with ether, and the combined organic layers were washed, concentrated and purified by silica column chromatography (ethyl acetate: petroleum ether = 1:24, v/v), yielding 2.5 g (31.76%) of intermediate product. A solution of 2.5 g (15 mmol) of intermediate and 3 mL (35 mmol) of isopropyl amine was stirred and refluxed at 50 ◦ C for 14 h. Excess amine was evaporated under reduced pressure, and it was further purified by silica column chromatography (dichloromethane: methanol = 95:5, v/v), yielding 1.1645 g (34.26%) of compounds (2). Compound (3): A solution of 11.0 g (0.1 mol) of hydroquinone, 37.0 g (0.4 mol) of epichlorohydrin, and 0.4 mL of 10 mol L−1 sodium hydroxide was stirred under nitrogen gas under 40 ◦ C for 48 h. After cooling, 42.0 mL of 5 mol L−1 sodium hydroxide, saturated with sodium carbonate, was added and the mixture stirred vigorously at room temperature for 20 h. The aqueous layer was extracted twice with chloroform and the combined organic layers were washed, concentrated and purified by silica column chromatography (ethyl acetate: petroleum ether = 1:6, v/v), yielding 0.6 g (2.70%) of intermediate product. A solution of 0.6 g (2.7 mmol) intermediate, 4 mL (50 mmol) of isopropyl amine and 20 mL of absolute ethanol was stirred and refluxed at 50 ◦ C for 8 h. Excess amine was evaporated under reduced pressure, and it was further purified by silica column
L. Lv et al. / J. Chromatogr. A 1513 (2017) 235–244
Fig. 2. Chromatogram of HPLC analyses of racemic compound (1) and (2). HPLC conditions: chiral column: Chiralcel OD-R, 10 m particle size of the packing material (250 mm × 6 mm I.D.); mobile phase: 0.2 mol L−1 potassium hexafluorophosphate aqueous solution: acetonitrile (60:40, v/v); flow rate: 0.6 mL min−1 ; UV wavelength: 254 nm; column temperature: 30 ◦ C.
chromatography (dichloromethane: methanol = 19:2, v/v), yielding 0.6 g (65.30%) of compounds (3). Compound (4): A solution of 12.4 g (0.1 mol) of 4methylcatechol, 37.0 g (0.4 mol) of epichlorohydrin, and 0.4 mL of 10 mol L−1 sodium hydroxide was stirred under nitrogen gas under 40 ◦ C for 48 h. After cooling, 42.0 mL of 5 mol L−1 sodium hydroxide, saturated with sodium carbonate, was added and the mixture stirred vigorously at room temperature for 20 h. The organic layer was washed, concentrated and purified by silica column chromatography (ethyl acetate: petroleum ether = 1:5, v/v), yielding 1.82 g (7.71%) of intermediate product. A solution of 1.82 g of (7.7 mmol) intermediate, 5 mL (58.7 mmol) of isopropyl amine and 15 mL of toluene was stirred and refluxed at 80 ◦ C for 8 h. Excess amine was evaporated under reduced pressure, and it was further purified by silica column chromatography (dichloromethane: methanol = 19:2, v/v), yielding 0.6 g (21.98%) of compounds (4). 2.4. Enantioselective liquid-liquid extractions The enantioselective liquid-liquid extraction of compound (1)–(3) was conducted to determine the distribution ratio and to investigate the effects of influence factors on distribution ratio and stereoselective separation factor. The aqueous phases were pre-
237
Fig. 3. Chromatogram of HPLC analyses of racemic compound (3) and (4). HPLC conditions: chiral column: Chiralpak IE, 5 m particle size of the packing material (250 mm × 4.6 mm I.D.); mobile phase: methyl tert-butyl ether: hexane: ethanol: diethylamine (60:30:10:0.1 for compound (3) and 60:39:1:0.1 for compound (4), v/v/v/v); flow rate: 0.5 mL min−1 ; UV wavelength: 289 nm; column temperature: 35 ◦ C. Table 1 Effect of L-tartrates on the distribution ratio (D) and enantioseparation factor (␣) of racemic compound (1). Chiral selectors
D−
D+
␣
di-n-butyl l-tartrate di-isobutyl l-tartrate di-n-hexyl l-tartrate di-n-octyl l-tartrate di-isooctyl l-tartrate
0.325 0.344 0.387 0.417 0.508
0.803 0.849 0.981 1.045 1.210
2.473 2.466 2.534 2.506 2.383
Aqueous phases: 0.05 mol L−1 acetate buffer pH = 6.1 containing 0.10 mol L−1 boric acid and 1.0 mmol L−1 racemic compound (1); organic phase: chloroform solution of 0.10 mol L−1 l-tartrate; equilibration temperature: 10 ◦ C.
pared by dissolving 0.10 mol L−1 of boric acid in a 0.05 mol L−1 of acetate buffer with pH 3.6–6.9, in which 1 mmol L−1 of racemic compound was added. The organic phase chloroform contains 0.10 mol L−1 of dialkyl l-tartrate. The equilibrium experiments were performed in a 10 mL glass-stoppered tube. 2.0 mL of organic phase and 2.0 mL of aqueous phase were placed in the stoppered glass tube and shaken vigorously for 4 times, 2 min-shaking for each time, and 30 min-settling down in a water bath at a constant temperature to reach phase equilibrium. The investigated temperature was in the range of 5–30 ◦ C. After phase separation, the concentration of enantiomer in the aqueous phase was determined by HPLC, and the concentration of enantiomer in the organic phase was cal-
238
L. Lv et al. / J. Chromatogr. A 1513 (2017) 235–244
Fig. 4. Effect of concentration of di-n-hexyl l-tartrate and boric acid on the distribution ratio (D) and enantioseparation factor (␣) for compound (1). Organic phase: (a)–(a’) chloroform added with different concentration of di-n-hexyl l-tartrate; (b)–(b’) chloroform added with 0.10 mol L−1 of di-n-hexyl l-tartrate; aqueous phase: 0.05 mol L−1 of acetate buffer at pH6.1 containing (a)–(a’) 0.10 mol L−1 boric acid; (b)–(b’) different concentration of boric acid; 1 mmol L−1 of racemate dissolved in the aqueous phase, and temperature was 10 ◦ C. D+ = 13.879, D− = 15.713 and ␣ = 1.132 when the concentration of di-n-hexyl l-tartrate was 0.00 mol L−1 , and D+ = 7.673, D− = 6.664 and ␣ = 1.152 when the concentration of boric acid was 0.00 mol L−1 , which are not shown in the figure.
culated by the subtraction method. The distribution ratio (D) of the enantiomers was expressed as the total concentration of analytes in the upper phase divided by the total concentration of analytes in the lower phase, and the separation factor was obtained with the following equation: ˛=
D2 D1
(1)
where D2 ≥ D1 . 2.5. Preparation of biphasic solvent systems and sample solutions Biphasic solvent systems consisting of chloroform added with 0.10 mol L−1 of di-n-hexyl l-tartrate and 0.05 mol L−1 of acetate buffer added with 0.10 mol L−1 boric acid (1:1, v/v) were used. The solvent mixture was thoroughly equilibrated in a separatory funnel, and the two phases were separated shortly before use. The sample solutions were prepared as follows: 20–42 mg of racemate
was dissolved in 9 mL of the aqueous phase for each stereoselective separation. 2.6. Countercurrent chromatography procedure The countercurrent chromatography was conducted with headto-tail elution mode. All separations were initiated by filling the column with the organic stationary phase. The aqueous mobile phases were pumped into the column while the column was rotated at 800 rpm (counter clockwise). Each of the sample solution was injected after the hydrodynamic equilibrium was reached, as indicated by a clear mobile phase eluting at the outlet. 2.7. Analytical method Chiral separation of compound (1) and (2) was performed by HPLC in reverse phase mode on a column of Chiralcel OD-R, with 10 m particle size of the packing material, 250 mm × 4.6 mm I.D.
L. Lv et al. / J. Chromatogr. A 1513 (2017) 235–244
239
Fig. 5. The distribution ratio (D) and enantioseparation factor (␣) of compound (1) under different pH and temperature. (a)–(a’) Organic phase: chloroform containing 0.10 mol L−1 of di-n-hexyl l-tartrate, Aqueous phase: 0.05 mol L−1 of acetate buffer containing 0.10 mol L−1 of boric acid and 1 mmol L−1 of racemate; temperature 10 ◦ C; (b)–(b’) Organic phase: chloroform added with 0.10 mol L−1 of di-n-hexyl l-tartrate, Aqueous phase: 0.05 mol L−1 of acetate buffer at pH6.1 containing 0.10 mol L−1 of boric acid and 1 mmol L−1 of compound (1). D+ = 55.731, D− = 37.458 and ␣ = 1.488 when the pH value of aqueous phase was 3.6, which is not shown in the figure.
(Daicel Chiral Technologies Co., Ltd., Shanghai, China). The UV detector was set at 254 nm. The mobile phase was composed of 0.2 mol L−1 of potassium hexafluorophosphate aqueous solution: acetonitrile (60:40, v/v) at a flow rate of 0.6 mL min−1 . The column temperature was 30 ◦ C. Analytical stereoselective separation of compound (3) and (4) was conducted on a normal phase Chiralpak IE, with 5 m particle size of the packing material, 250 mm × 4.6 mm I.D. (Daicel Chiral Table 2 The distribution ratio (D) and seteroselective separation factor (␣) for compound (1)-(4) under optimized separation condition. racemates
pH
D − /D −,−
D +,−or−,+
D + /D +,+
␣
(1) (2) (3) (4)
6.1 5.3 7.0 7.0
0.387 0.407 0.550 0.111
– – 1.147 0.208
0.981 1.009 2.538 0.500
2.534 2.479 2.085, 2.213 1.874, 2.404
Aqueous phases: 0.10 mol L−1 boric acid aqueous with different pH values and 1.0 mmol L−1 racemic compound (1); organic phase: chloroform solution of 0.10 mol L−1 l-tartrate; equilibration temperature: 10 ◦ C.
Technologies Co., Ltd., Shanghai, China).The UV detector was set at 289 nm. The mobile phase was composed of methyl tert-butyl ether: n-hexane: ethanol: diethylamine (60:30:10:0.1 for compound (3) and 60:39:1:0.1 for compound (4), v/v/v/v) at a flow rate of 0.5 mL min−1 . The column temperature was 35 ◦ C.
3. Results and discussion 3.1. Development of analytical HPLC methods As known, the methods for enantioseparation of -adrenergic blocking agents containing one chiral center by liquid chromatography and capillary electrophoresis have been well documented. The reverse phase Chiralcel OD-R was proven to be a very effective chiral stationary phase for enantioseparation of some typical -blocker drugs, such as propranolol, pindolol, alprenolol and propafenone. It was found that racemic compounds (1) and (2) could be well enantioseparated by Chiralcel OD-R column, as shown in Fig. 2. However, it was difficult to separate stereoselectively com-
240
L. Lv et al. / J. Chromatogr. A 1513 (2017) 235–244
Fig. 6. Chromatogram of enantioseparation of racemic compound (1) and (2) by countercurrent chromatography. Biphasic solvent system: chloroform: 0.05 mol L−1 of acetate buffer pH6.1 for compound (1) and pH5.3 for compound (2) (1:1, v/v), in which organic phase contained 0.10 mol L−1 of di-n-hexyl l-tartrate and aqueous phase contained 0.10 mol L−1 of boric acid; stationary phase: lower organic phase; mobile phase: upper aqueous phase; sample solution: 20 mg of compound (1) and 30 mg of compound (2) dissolved in 10 mL of the aqueous mobile phase, respectively; flow rate: 3.0 mL min−1 for (1) and 2.0 mL min−1 for (2); revolution: 800 rpm; column temperature: 10 ◦ C; stationary phase retention: 50% for (1) and 60% for (2).
pounds (3) and (4) by the same chiral stationary phase Chiralcel OD-R. Various kinds of chiral stationary phases as well as chiral mobile phase additives had been tried in our lab and only the chiral stationary phase Chiralpak IE column, normal phase, was found to be suitable for analytical stereoselective separation of compounds (3) and (4). Fig. 3 was a typical chromatogram for stereoselective separation of compounds (3) and (4) by high performance liquid chromatography with Chiralpak IE as chiral stationary phase. The mobile phase was composed of methyl tert-butyl ether: hexane: ethanol: diethylamine with a volume ratio of 60:30:10:0.1 for compound (3) and 60:39:1:0.1 for compound (4). Complete stereoselective separation of compound (3) (two enantiomers plus a meso compound) could be achieved, while only partial stereoselective separation of compound (4) (two pairs of enantiomers) was obtained after optimization of chromatographic conditions, as shown in Fig. 3.
3.2. Stereoselective separation by countercurrent chromatography 3.2.1. Optimization of separation conditions Stereoselective separation of the present -adrenergic blocking agents was based on a stereospecific ternary borate coordination
complex formed by enantiomer, l-tartrate and boric acid. Separation factor could be greatly affected by the following factors: type of chiral selector, concentration of di-n-hexyl l-tartrate, concentration of boric acid, pH value of aqueous buffer and temperature. Therefore, they were investigated by enantioselective liquid-liquid extraction with regarding to distribution performance of compound (1). Table 1 shows the distribution ratio and enantioseparation factor of compound (1) under different type of l-tartrate (chiral selector). No much difference was found among all the chiral selectors investigated. However, di-n-hexyl l-tartrate shows a slightly higher enantioseparation factor than the others. So di-n-hexyl ltartrate was selected in the present work. Fig. 4(a) and (a’) shows the distribution ratio and enantioseparation factor under various concentration of di-n-hexyl l-tartrate, respectively. With an increase of the concentration of di-n-hexyl l-tartrate, the distribution ratio decreased greatly, while the enantioseparation factor increased up to a constant when the concentration of di-n-hexyl l-tartrate was around 0.20 mol L−1 . This was caused by the fact that an improved coordination complex would be formed with an increasing concentration of l-tartrate. Very low distribution ratio was found with the concentration of di-n-hexyl l-tartrate being greater than 0.20 mol L−1 , which indicated very long retention time
L. Lv et al. / J. Chromatogr. A 1513 (2017) 235–244
241
Fig. 7. Chromatogram of stereoselective separation of compound (3) and (4) by countercurrent chromatography. Biphasic solvent system: chloroform contained 0.10 mol L−1 of di-n-hexyl l-tartrate: 0.10 mol L−1 of boric acid at pH7.0 adjusted with triethylamine (1:1, v/v); stationary phase: lower organic phase; mobile phase: upper aqueous phase; sample solution: 40 mg of compound (3) and 42 mg of compound (4) dissolved in 10 mL of the aqueous phase individually; flow rate: 2.0 mL min−1 ; revolution: 800 rpm; column temperature: 10 ◦ C; stationary phase retention: 60%.
was necessary if the upper aqueous phase was used as the mobile phase. So the concentration of di-n-hexyl l-tartrate at 0.10 mol L−1 was selected since a suitable distribution ratio could be obtained. Fig. 4(b) and (b’) shows the distribution ratio and enantioseparation factor under various concentration of boric acid in the aqueous phase, respectively. Boric acid was necessary for the formation of a stereospecific ternary borate coordination complex among enantiomer, boric acid and dialkyl l-tartrate. Enantioseparation factor completely lost when there was no boric acid in the solvent system and most of racemic solutes would partition into the aqueous solution. When the concentration of boric acid was in the range of 0.1-0.2 mol L−1 , suitable distribution ratio and a high enantioseparation factor could be obtained. The effects on distribution ratio and enantioseparation factor by pH of aqueous phase as well as temperature were shown in Fig. 5. Fig. 5(a) indicated distribution ratio decreased sharply with the pH of the aqueous phase increased from pH = 4.5 to pH = 6.8, while the enantioseparation factor reached a highest value when pH was increased to pH = 5.4 and then it leveled off, as shown in Fig. 5(a’). This phenomenon could be explained by the fact that the compound (1) is ionized with low pH in the aqueous phase due to the amino alcohol group of its chemical structure. So pH = 6.1
was selected in light of suitable distribution ratio along with relatively higher enantioseparation factor. Fig. 5(b) shows the influence of temperature in the distribution behavior, which indicated that a slight increase in the distribution ratio of (−)-enantiomer was observed while a slight decrease in the distribution ratio of (+)enantiomer was observed with the increasing of the temperature within the range5–30 ◦ C. As a result, it led to a decrease of the enantioseparation factor ␣ (Fig. 5(b’)). The enantioseparation factor decreased greatly when the temperature was over 20 ◦ C. Therefore, the temperature ranged within 5–15 ◦ C was selected. As a result, the optimized separation condition for compound (1) was as followings: the biphasic solvent system was composed of chloroform and 0.05 mol L−1 acetate buffer pH6.1 (1:1, v/v), in which organic phase contained 0.10 mol L−1 di-n-hexyl l-tartrate and aqueous phase contained 0.10 mol L−1 boric acid. And the separation temperature was 10 ◦ C. The distribution ratio and separation factors of compound (1)–(4) under optimized separation conditions were summarized in Table 2. However, it was difficult to determine the distribution ratio for each of the four stereoisomers of compound (4) due to their partial resolution in HPLC analysis (as shown in Fig. 3, bottom). Therefore, the distribution ratio of compound (4) listed in Table 2 were calculated by the following Eq. (2) according
242
L. Lv et al. / J. Chromatogr. A 1513 (2017) 235–244
Fig. 8. Chromatogram of HPLC analyses of collected fractions from stereoselective separation of compound (3) and (4) by countercurrent chromatography. (a): fraction containing (+, +)-compound (3); (b): fraction containing (+, −) or (−, +)-compound (3); (c): fraction containing (−, −)-compound (3); (a’): fraction containing (+, +)-compound (4); (b’): fraction containing (+, −) or (−, +)-compound (4); (c’): fraction containing (−, −)-compound (4); HPLC conditions: chiral column: Chiralpak IE, 5 m particle size of the packing material (250 mm × 4.6 mm I.D.); mobile phase: methyl tert-butyl ether: hexane: ethanol: diethylamine 60:30:10:0.1 for (3) and 60:39:1:0.1 for (4), v/v/v/v; flow rate: 0.5 mL min−1 , UV wavelength: 289 nm; column temperature: 35 ◦ C.
L. Lv et al. / J. Chromatogr. A 1513 (2017) 235–244
to the chromatogram of stereoselective separation of compound (4) by countercurrent chromatography. D=
VR − Vm Vs
(2)
where VR represents the elution volume of analyte, Vm and Vs are the volume of mobile phase and volume of stationary phase in the separation column. 3.2.2. Stereoselective separation by countercurrent chromatography Enantioseparation of compound (1) and (2) with one chiral center by TBE–200 V apparatus was conducted using the two-phase solvent system composed of chloroform and 0.05 mol L−1 acetate buffer (1:1, v/v), in which organic phase contained 0.10 mol L−1 of di-n-hexyl l-tartrate and aqueous phase contained 0.10 mol L−1 of boric acid. The only difference in separation conditions between compounds (1) and (2) was the pH value of aqueous phase (at pH6.1 for compound (1) and pH5.3 for compound (2)). A relatively large lipophilicity of compound (2) was observed compared with that of compound (1), which was mainly caused by the methyl group on the benzene ring. Suitable distribution ratio of compound (2) could be obtained by decreasing the pH value of aqueous phase. So pH5.3 was selected for enantioseparation of compound (2). Fig. 6 shows the typical chromatogram of enantioseparation of compound (1) and (2) by TBE–200V countercurrent chromatographic apparatus. As shown in Fig. 6, complete resolution of two racemates was achieved. HPLC results demonstrated that the purity of the enantiomer collected from the elution of countercurrent chromatography reached more than 98%. Fig. 7 shows the typical chromatogram of stereoselective separation of compound (3) and (4) with two chiral centers by TBE-200V countercurrent chromatographic apparatus. The biphasic solvent system used here was the same as that used in enantioseparation of compound (1) and (2) except for different pH value of aqueous phase. Compounds (3) and (4) tend to partition into aqueous phase with the additional substitution of amino alcohol group on the benzene ring. Stereoisomer of compounds (3) and (4) would be easily eluted out with the aqueous phase as the mobile phase in countercurrent chromatography. In order to improve its retention factor in the separation column, pH of aqueous phase was adjusted to around 7.0. As shown in Fig. 7, three main fractions were obtained after stereoselective separation by countercurrent chromatography with single run. Each fraction was collected according to the peak profile and analyzed by chiral HPLC. Fig. 8 shows the HPLC analysis of each fraction collected from stereoselective separation by countercurrent chromatography. Results showed that as for compound (3), (+,+)-enantiomer was eluted from the first peak and (−,−)enantiomer was eluted from the third peak, while (+,−) and (−,+)-mesomer were eluted in the second peak. The same elution performance was also observed as for (+,+)-stereoisomer and (−,−)stereoisomer of compound (4) and the diastereoisomers, (+,−) and (−,+)-compound (4), were eluted between (+,+)-stereoisomer and (−,−)-stereoisomer, which could not be separated by countercurrent chromatography. Diastereoisomers may exhibit significant differences in physical properties, such as solubility, partition coefficient, and melting point. Differences in biologic activity between diasteroisomers may, therefore, be due to differences in physical properties. It would still be difficult for countercurrent chromatography to separate the pair of diastereoisomer of compound (4) due to their extremely similarity of chemical structure. Therefore, only three main peaks were obtained, as shown in Fig. 7. It was noticed that the elution sequence of (+,+)stereoisomer and (−,−)-stereoisomer were completely reversed
243
compared with that of HPLC analysis as for compound (3). Meanwhile, the diastereoisomeric (+,−) and (−,+)-compound (4) was eluted before (−,−)-enantiomer in HPLC chromatogram. Recovery of enantiomers, mesomer or diastereoisomer of compound (1)–(4) was around 87–93% with their purity of 96–98%, as determined by HPLC. 4. Conclusion A method for stereoselective separation of four -adrenergic blocking agents by countercurrent chromatography using borate coordination complex, in which two of them owning two chiral centers, was attempted for the first time. A two-phase solvent system was selected and separation conditions were optimized. 20–42 mg of each racemic -adrenergic blocking agent was separated by high speed countercurrent chromatography in a single run with 96–98% purity, and the recovery of each enantiomer reached 87–93%. The present research indicated that countercurrent chromatography might also be applied in stereoselective separation of components with two chiral centers. References [1] L. Poirier, Y. Lacourciere, The evolving role of - adrenergic receptor blockers in managing hypertension, Can. J. Cardiol. 28 (2012) 334–340. [2] J.M. Cruickshank,  blockers in hypertension, Lancet 376 (2010) 415–416. [3] Y.Z. Lin, J. Zhou, J. Tang, W.H. Tang, Cyclodextrin clicked chiral stationary phases with functionalities- tuned enantioseparations in high performance liquid chromatography, J. Chromatogr. A 1406 (2015) 342–346. [4] Y.N. Zou, L.J. Wang, Q. Liu, H.Y. Liu, F.N. Li, Enantioseparations of 11 amino alcohols using di- n- amyl l- tartrate- boric acid complex as chiral mobile phase additive by RP- HPLC, Chromatographia 78 (2015) 753–761. [5] X.L. Weng, Z.B. Bao, H.B. Xing, Z.G. Zhang, Q.W. Yang, B.G. Su, Y.W. Yang, Q.L. Ren, Synthesis and characterization of cellulose 3, 5- dimethylphenylcarbamate silica hybrid spheres for enantioseparation of chiral - blockers, J. Chromatogr. A 1321 (2013) 38–47. [6] I. Ali, A. Haque, M.F. Al Ajmi, A. Hussain, M. Marsin Sanagi, I. Hussain, H.Y. Aboul-Enein, Supramolecular chiro-biomedical aspect of -blockers in drug development, Curr. Drug Targets 15 (2014) 729–741. [7] I. Ali, S.D. Alam, J.A. Farooqi, N. Nagae, V.D. Gaitonde, H.Y. Aboul-Enein, A comparison of -blockers separation on C18 and new generation C28 columns in human plasma, Anal. Methods 5 (2013) 3523–3529. [8] I. Ali, Z.A. Al-Othman, A. Hussain, K. Saleem, H.Y. Aboul-Enein, Chiral separation of -adrenergic blockers in human plasma by SPE-HPLC, Chromatographia 73 (2011) 251–256. [9] I. Ali, V.D. Gaitonde, H.Y. Aboul-Enein, A. Hussain, Chiral separation of -adrenergic blockers on cellucoat column by HPLC, Talanta 78 (2009) 458–463. [10] S.Q. Tong, Y. Zheng, J.Z. Yan, Y.X. Guan, C.Y. Wu, W.Y. Lei, Preparative enantioseparation of -blocker drugs by counter-current chromatography using dialkyl l-tartrate as chiral selector based on borate coordination complex, J. Chromatogr. A 1263 (2012) 74–83. [11] I. Ali, M. Suhail, M.N. Lone, Z.A. Al-Othman, A. Alwarthan, Chiral resolution of multichiral center racemates by different modalities of chromatography, J. Liq. Chromatogr. Relat. technol. 39 (2016) 435–444. [12] I. Ali, M. Suhail, Z.A. Al-Othman, A. Al-Warthan, H.Y. Aboul-Enein, Enantiomeric resolution of multiple chiral centres racemates by capillary electrophoresis, Biomed. Chromatogr. 30 (2016) 683–694. [13] Z.A. Al-Othman, A. Al-Warthan, S.D. Alam, I. Ali, Enantio- separation of drugs with multiple chiral centers by chromatography and capillary electrophoresis, Biomed. Chromatogr. 28 (2014) 1514–1524. [14] W. Bragg, D. Norton, S.A. Shamsi, Optimized separation of - blockers with multiple chiral centers using capillary electrochromatography- mass spectrometry, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 875 (2008) 304–316. [15] H.Y. Aboul-Enein, High- performance liquid chromatographic enantioseparation of drugs containing multiple chiral centers on polysaccharide- type chiral stationary phases, J. Chromatogr. A 906 (2001) 185–193. [16] H.Y. Aboul-Enein, I. Ali, Studies on the effect of alcohols on the chiral discrimination mechanisms of amylose stationary phase on the enantioseparation of nebivolol by HPLC, J. Biochem. Biophys. Methods 48 (2001) 175–188. [17] H.Y. Aboul-Enein, I. Ali, HPLC enantiomeric resolution of nebivolol on normal and reversed amylose based chiral phase, Pharmazie 56 (2001) 214–216. [18] M. Bojczuk, D. Zyzelewicz, P. Hodurek, Centrifugal partition chromatography – a review of recent applications and some classic references, J. Sep. Sci. 40 (2017) 1597–1609.
244
L. Lv et al. / J. Chromatogr. A 1513 (2017) 235–244
[19] X.Y. Huang, D.L. Di, Chiral separation by counter- current chromatography, TrAC Trends Anal. Chem. 67 (2015) 128–133. [20] S.Q. Tong, Y. Zheng, J.Z. Yan, Enantioseparation of chiral aromatic acids by multiple dual mode counter-current chromatography using hydroxypropyl--cyclodextrin as chiral selector, J. Sep. Sci. 36 (2013) 2035–2042.
[21] E. Heldin, K.-J. Lindner, C. Pettersson, W. Lindner, R. Rao, Tartaric acid derivatives as chiral selectors in liquid chromatography, Chromatographia 32 (1991) 407–416. [22] J. Zaagsma, W.Th. Nauta, -Adrenoceptor studies. 1. In vitro -adrenergic blocking, antiarrhythmic, and local anesthetic activities of a new series of aromatic bis(2-hydroxy-3-isopropylaminopropyl) ethers, J. Med. Chem. 17 (1974) 507–513.