Determination of (R,R)-glycopyrronium bromide and its related impurities by ion-pair HPLC

ORIGINAL ARTICLES Department of Medicinal Chemistry, Faculty of Life Sciences, University Vienna, Austria Determination of (R,R)-glycopyrronium bromide and its related impurities by ion-pair HPLC Dedicated to Professor Wilhelm Fleischhacker with the best wishes on occasion of his 75th birthday D. Nebiu, M. Walter, B. Lachmann, H. Kopelent, C. R. Noe Received October 12, 2006, accepted November 10, 2006 Prof. Dr. Christian R. Noe, Department of Medicinal Chemistry, University Vienna, Center of Pharmacy, Althanstr. 14, 1090 Vienna, Austria christian.noe@univie.ac.at Pharmazie 62: 406–410 (2007) doi: 10.1691/ph.2007.6.6220 A simple, rapid and specific ion-pair HPLC method for the determination of (R,R)-glycopyrronium bromide and its related impurities is presented, and parameters affecting the chromatographic properties of these compounds are discussed. Optimal analyte separation was achieved on base deactivated Nucleosil at 40
C, using phosphate buffer pH 2.30 with sodium-1-decanesulfonate (0.01 M)/methanol (35/65; v/v) as eluent for isocratic elution at a flow rate 1ml  min1. The analytical assay was validated according to international guidelines. The method is suitable for in-process control and as stability indicating assay. 1. Introduction 3-[(Cyclopentylhydroxyphenylacetyl)oxy]-1,1-dimethyl-pyrrolidinium bromide (INN: glycopyrronium bromide; market as a racemic drug, often under the USAN-term glycopyrrolate) belongs to the synthetic anticholinergic agents, a class of drugs which competitively antagonise the effects of acetylcholine competing with this neurohormone for cholinergic receptor sites (Rama Sastry 1996). Consequently, the drug has been investigated for the management of chronic obstructive pulmonary disease and bronchial asthma (Cydulka and Emerman 1995; Gilman et al. 1990; Hansel 2005; Villetti et al. 2006), hyperhydrosis (Wohlrab 2003), urine storage failure (Levin and Wein 1982), treatment of allergic rhinitis (Weinstein and Weinstein 2000), management of peptic ulcer (Miettinen et al. 1985) and as postoperative medication (Biwas et al. 2002). In order to increase muscarinic receptor subtype selectivity and to abolish unwanted side effects, enantiomerically pure (R,R)-glycopyrronium bromide has been designed (Czeche et al. 1997; Noe et al. 1999). A new, improved route of (R,R)-glycopyrronium bromide synthesis has been developed in our group (Noe and Walter 2002), consequently appropriate methods to determine its potential impurities and to monitor the reaction process (Scheme) are needed. These analytical steps are important, especially since (R,R)-glycopyrronium bromide is currently clinically developed for the treatment of chronic obstructive pulmonary disease (COPD) by MEDA Pharma (Guera et al. 2006). The final intermediate 5 as tertiary amine easily might pass the blood brain barrier and induce unwanted effects. Therefore, special emphasis should be placed on monitoring this potential impurity. Impurity 3 can result as decomposition product of (R,R)-glycopyrronium after hydro406 lytic cleavage of the ester functionality, and represents a biotransformation product of active compound as well. Previously reported analytical techniques for determination of glycopyrronium bromide were based on HPLC/MS (Matassa et al. 1992), GC/MS (Murray et al. 1984), CE/MS (Tang et al. 2001); high-flow ion spray HPLC/MS (Hopfgartner et al. 1993); normal and reverse phase TLC (Ojanpera et al. 1991); and spectrophotometry (Ebeid et al. 1986). The USP 25 monograph “glycopyrrolate injection” employs an RP-HPLC assay using sodium-1-pentanesulfonate as an ion-pairing agent (USP 25th Ed. 1995). Chromatography with b-cyclodextrine as chiral stationary phase has been proposed for the determination of glycopyrronium bromide diastereomers (Demian and Gripshover 1990). Investigations aimed at the determination of (R,R)-glycopyrronium bromide and its related impurities have not H O HO O HO O O N 5 4 H3C N + Br CH3 CH3 R HO H O HO O R 1 O O H3C N + Br CH3 O HO 2 N CH3 OH 3 Pharmazie 62 (2007) 6 ORIGINAL ARTICLES Scheme 12 Impurities 2,4 Impurity 3 Glycopyrronium Impurity 5 OH 10 R O + L-4-Hydroxyproline N O CH3 Retention time (min) HO CH3 N-methylpyrrolidin-3-ol racemate O HO 8 6 4 O R 2 quaternization N 5 0 CH3 45 50 O HO R HO O O R 60 65 70 75 O R N + Br CH3 H3 C Fig. 1: Retension times of 1, 2, 3, 4 and 5 at different methanol content. Mobile phase: methanol (50–70%) : ammonium acetate buffer, 20 mM, pH 4.00; sodium pentanesulfonate (0.01 M) resolution (3R,2'R) and (3S,2'R) 8 1 55 Methanol content (vol%) N + Br CH3 H3 C Impurities 2,4 Impurity 3 Glycopyrronium Impurity 5 7 been published hitherto. The purpose of the present study was to develop a simple, rapid, and sensitive method for simultaneous determination of (R,R)-glycopyrronium bromide and its potential impurities which arise during the synthesis process, and as a stability indicating assay with respect to degradation product 3. Retention time (min) 6 5 4 3 2 1 2. Investigations, results and discussion 0 3.5 4 4.5 5 5.5 6 6.5 2.1. Optimization of the chromatographic conditions Considering the structure of the polar and basic analytes severe peak tailing could be presumed when using unmodified RP phases, therefore a base deactivated end-capped stationary phase was included in the investigation. Initial studies were performed with two types of mobile phase modifiers: acetonitrile and methanol. Acetonitrile, due to its high elutive power, in all cases provided very fast elution and poor selectivity for all compounds studied. Therefore, all subsequent studies were performed employing methanol. As aqueous component of the eluent ammonium acetate buffer, 20 mM, pH 4.00; 4.50; 5.00 and 6.00, and phosphate buffer, pH 2.30, respectively, were tested for their suitability. The influence of the addition of ion pair reagents was studied as well as different column temperatures. 2.1.1. Eluent composition Ammonium acetate buffer system pH 4.00; 4.50; 5.00; 6.00 Initial experiments indicated that the shape of peaks from (R,R)-glycopyrronium bromide and impurity 5 result as highly asymmetric, therefore, utilization of an ion pair additive was considered. Variations of the aqueous component of the eluent with respect to pH and to type of the ion pairing reagent proved that chromatographic retention Pharmazie 62 (2007) 6 Fig. 2: Influence of the pH value on the retention of 1, 2, 3, 4 and 5. Mobile phase: methanol : ammonium acetate buffer, 20 mM, 60 : 40 (V/V); sodium-1-pentanesulfonate (0.01 M) of (R,R)-glycopyrronium bromide was not affected by pH change but mainly by the lipophilicity of the ion pair created with the anionic reagent and/or the ion-exchange process involved with the sodium ions from sorbed reagent molecules at the stationary phase. Retention of compound 5 was strongly dependent from the ion-pair reagent type as well as the pH value. Naturally the influence of the ion pairing on the chromatographic behaviour of 5 was more distinct at lower pH, at pH values between 4.0 and 6.0 both factors were equally significant (Fig. 1). As evident, the retention of mandelic acid derivative 3 was affected by the pH value only (Fig. 2). The ion pairing with sodium1-pentanesulfonate 0.01 M lead to separation of the critical peaks of 1 and 5, nevertheless the peak symmetry was unsatisfactory. Though reasonable capacity factors for compounds 1 and 5 were found with eluents consisting of methanol: ammonium acetate; 60 : 40 (v/v), at pH 4.00 and 6.00, respectively, containing sodium-1-decanesulfonate, the assymetry factor did not satisfy. Mandelic acid derivative 3 eluted very fast (k0 < 2) under these conditions. 407 ORIGINAL ARTICLES Phosphate buffer pH 2.30 At all ratios of eluent components the cationic derivatives eluted very early without the use of ion pair reagents, peak tailing was observed especially with compounds 1 and 5. Impurity 3 naturally had a high retention time. In order to allow isocratic elution, ion pair reagents were used to selectively increase the retention of the cationic compounds relative to impurity 3. Sodium-1-pentanesulfonate increased the retention of all ionic and ionizable compounds, compound 3 naturally was not affected (Fig. 3). At higher methanol percentages critical peaks (1 and impurity 5) were not efficiently resolved, lower amounts of methanol led to unsuitably long retention times. Aiming for fast analysis with appropriate retention range, possibly 2  k0  5, ion-pairing reagents of higher molecular size as sodium-1-dodecanesulfonate and sodium-1-decanesulfonate were considered, the latter proved to be better suitable (Fig. 4). At 65% methanol, this ion-pair reagent afforded baseline resolution for 1, 3 and 5, impurities 2 and 4 co-eluted. Summarizing, it can be stated that lower pH of the eluent and higher modifier percentages together with a more hydrophobic ion pairing agent attained good resolution com- Table 1: Influence of stationary phase on peak asymmetry. Mobile phase: methanol : ammonium acetate buffer 20 mM, pH 4.00, 70 : 30 (v/v); sodium-1-pentanesulfonate (0.01 M) Column type Asymmetry factor Nucleosil 100-5C-18 HD LiChrospher 100-RP 18 (R,R)-Glycopyrronium Impurity 5 2.14 2.87 1.9 3.3 Table 2: Influence of stationary phase on peak asymmetry. Mobile phase: methanol : phosphate buffer, pH 2.30, 65 : 35 (v/v); sodium-1-decanesulfonate (0.01 M) Column type Asymmetry factor Nucleosil 100-5C-18 HD LiChrospher 100-RP 18 (R,R)-Glycopyrronium Impurity 5 1.22 1.99 1.31 1.91 bined with satisfactory peak shapes. Accordingly, methanol/ phosphate buffer pH 2.30 with sodium-1-decanesulfonate 0.01 M (65 : 35, v/v) was selected as the most suitable mobile phase solution which was used for further optimization studies. 18 Impurities 2 and 4 Impurity 3 Glycopyrronium Impurity 5 16 Retention time (min) 14 2.1.2. Stationary phase A common octadecyl silica, LiChrospher 100-RP 18 column was compared with a base deactivated Nucleosil 1005C-18 HD column. As expected, the reduction of the influence of the free silanol groups considerably improved peak shapes. The selection of a lower eluent pH led to additional improvement of the peak symmetry (Tables 1 and 2). 12 10 8 6 2.1.3. Column temperature 4 2 0 45 50 55 60 65 70 75 Methanol content (vol%) Fig. 3: Retension times of 1, 2, 3, 4 and 5 at different methanol content. Mobile phase: methanol (50–70%) : phosphate buffer, pH 2.30; sodium-1-pentanesulfonate (0.01 M) Retention time (min) 2.2. Method validation This method has been substantiated with the determination of specificity, linearity, precision, accuracy, limits of detection and limits of quantitation, according ICH guidelines (ICH Harmonised Tripartite Guidelines 1994, 1996). The 25 Impurities 2,4 Impurity 3 Glycopyrronium Impurity 5 20 The influence of temperature was studied in the range 20 to 60
C. A temperature of 40
C resulted as optimal affording a short analysis time (less than 7 min) and adequate resolution between critical peaks of glycopyrronium bromide 1 and impurity 5 (Rs > 2). All ensuing analyses for the quantification of impurities thereafter were carried out at 40
C (Table 3 and Fig.5). 15 Table 3: Influence of column temperature in chromatographic retention. (Mobile phase : methanol : phosphate buffer, pH 2.30, 65 : 35 (v/v); sodium-1-decanesulfonate (0.01 M) 10 Temp.
C 5 Retention time, min 3 1 5 4.69 4.18 3.91 3.56 2.80 9.15 7.30 6.17 5.00 3.30 11.70 8.66 7.29 6.00 3.40 0 58 60 62 64 66 68 Methanol content (vol%) 70 72 Fig. 4: Retension times of 1, 2, 3, 4 and 5 at different methanol content. Mobile phase: methanol (60–70%) : phosphate buffer, pH 2.30; sodium-1-decanesulfonate (0.01 M) 408 20 25 30 40 60 Pharmazie 62 (2007) 6 ORIGINAL ARTICLES Table 6: Limits of detection and limits of quantitation 80 AU (mV) 70 Glycopyrronium (R,R)-Glycopyrronium bromide Impurity 3 Impurity 5 60 LOD, mg  ml1 LOQ, mg  ml1 3.5103 1.5103 2.7103 1.0102 4.5103 8.2103 50 Impurities 2,4 40 Impurity 3 36 Impurity 5 30 1 2 3 4 5 6 7 AU (mV) 0 time (min) Fig. 5: HPLC of (R,R)-glycopyrronium bromide 100 mg ml1 , spiked with 5% impurities (mobile phase: methanol : phosphate buffer, pH 2.30, 65 : 35 (v/v); sodium-1-decanesulfonate 0.01 M); Column temperature, 40
C) 35 Glycopyrronium Compound 3 34 Compound 5 33 0 selectivity of the assay with respect to (R,R)-glycopyrronium bromide was assessed with injecting blank solution and (R,R)-glycopyrronium bromide, and for impurities in spiking experiments. No other signals appeared at the retention times of (R,R)-glycopyrronium bromide and its impurities. Linear relationship was observed between detector signals as a function of concentration of glycopyrronium 1, compounds 3 and 5. Regression line equations with their corresponding errors and squared correlation coefficients are presented in Table 4. For impurity testing, spiking experiments with the known impurities have been used for the evaluation of accuracy, which is reported as percentage of recovery by the assay of known added amounts of compound 3 and compound 5. At each study, triplicates of samples are evaluated. The mean of triplicates was expressed as recovery %. For glycopyrronium and compounds 3 and 5 repeatability was evaluated by expressing the relative standard deviation of triplicate injections of samples. Intermediate precision was evaluated for impurities 3 and 5. Accuracy (recovery values) at two different analysis days has been used to indicate the intermediate precision of method (Table 5). LOD and LOQ were assessed on the basis of the standard deviation of the response and the slope of calibration curve (Table 6 and Fig. 6). 1 2 3 4 time (min) 5 6 7 Fig. 6: (R,R)-Glycopyrronium bromide (100 mg ml1 ) spiked with 0.1% impurities 3. Experimental 3.1. Materials and reagents (R,R)-Glycopyrronium bromide; compounds 2, 3, 4 and 5 were synthesized in our laboratory. Methanol and acetonitrile, HPLC grade, were purchased from Acros Organics, Belgium. Sodium-1-pentanesulfonate and sodium-1decanesulfonate, analytical grade, Fluka Chemica GmbH, Switzerland; sodium-1-dodecanesulfonate, Merck, Germany. Phosphoric acid, Mallinkrodt Baker B.V. Holland, ammonium acetate, Riedel-De Haen AG, Hannover Germany, potassium dihydrogenphosphate, E. Merck, Darmstadt, Germany. HPLC grade water was purchased from Fisher Chemicals, UK, ammonium hydroxide and acetic acid from E. Merck Darmstadt, Germany. 3.2. Standard and sample preparation A stock solution of (R,R)-glycopyrronium bromide (1 mg  ml1) was prepared by dissolving 100 mg (R,R)-glycopyrronium bromide standard in mobile phase (volumetric flask 100 ml). This solution was further diluted with mobile phase to result in the appropriate concentrations to establish the calibration function. The solution of highest concentration, 100 mg  ml1, was used as reference standard in spiking experiments with solutions of impurities (compounds 2, 3, 4 and 5). Stock solutions of impurities (0.2 mg  ml1, compound 3 and 0.53 mg  ml1, compound 5) were prepared dissolving accurately weighted substances in mobile phase and sub- Table 4: Linearity parameters for compounds 1, 3, and 5 (R,R)-Glycopyrronium Impurity 3 Impurity 5 Regression line Eq. Sr Sa Sb R2 A ¼ 6095.9XC  754.96 A ¼ 14208.13XC þ 1487.80 A ¼ 7863.7XC  225.93 1750.41 2334.47 634.56 21.10 362.43 35.54 967.7 1487.0 344.3 1.00 0.9981 0.9999 Table 5: Accuracy and repeatability of impurities 3 and 5 Impurity level, % Impurity 3 tR, min 5.0 2.5 0.5 0.25 0.1 0.05 Impurity 5 Concentration tR, min Concentration min RSD % Observed mg  CI RSD, % n¼3 %R %R Second day min RSD % Observed mg  CI RSD, % n¼3 %R %R Second day 3.61 3.60 3.59 3.60 3.71 3.55 0 0.14 0.14 0.22 1.74 0.73 4.92  0.13 2.52  0.12 0.4  0.02 0.28*  0.05 0.108*  0.03 0.056*  0.006 1.1 2.05 2.5 7.5 11.5 0.41 98.4 100.2 80.0 111.0 108.0 112.0 96.0 97.6 88.0 108.0 116.0 –– 5.99 5.95 5.88 5.84 6.24 –– 0.08 0.60 0.14 0.41 3.5 –– 5.24  0.13 2.68  0.36 0.6  0.035 0.29  0.03 0.124  0.08 –– 1.0 5.3 2.3 4.5 26 –– 104.6 106.9 120.0 116.8 124.0 –– 86.4 96.0 109.0 112.0 120.0 –– * Recovery values and respective areas were calculated directly on the basis of (R,R)-glycopyrronium bromide peaks Pharmazie 62 (2007) 6 409 ORIGINAL ARTICLES sequently diluting with the same solvent to 5; 2.5; 0.5; 0.25; 0.1 and 0.05%, relative to the amount of parent compound (R,R)-glycopyrronium bromide, 100 mg  ml1. For construction of the calibration functions of impurities 3 and 5 their stock solutions were diluted with mobile phase to obtain appropriate concentrations. 3.3. Instrumentation The HPLC consisted of dual pump LC-10AD, Shimadzu; Rheodyne Injector Cotati, California USA, SPD-10A Shimadzu UV-VIS detector, and column thermostat SpH 99, Holland. Nucleosil 100-5 C-18 HD, octadecyl base deactivated (5 mm, 1254 mm I.D) and LiChrospher 100-RP 18, octadecyl, (5 mm, 1254 mm I.D) columns were used. 3.4. Mobile phase and LC conditions Phosphate buffer pH 2.30 was prepared dissolving 4.80 g phosphoric acid 85% and 6.66 g potassium dihydrogenphosphate in deionized water, adjusting to pH 2.30 with the phosphoric acid 85%, after that 1000 ml volume was accurately obtained adding deionised water. Ammonium acetate buffer, 20 mM was prepared dissolving 1.542 g of ammonium acetate in deionised water, adjusting the pH to the respective value with concentrated acetic acid or ammonia, deionized water was added to give 1000 ml. Accurately weighed ion-pairing reagents were dissolved in aqueous portions of mobile phase which was further mixed with organic phase and this solution was finally filtered through a Millipore 0.5 mm filter and degassed in Ultrasonic Bath Starsonic 60 for 10 min before each run. The chromatographic runs were performed in isocratic mode, mobile phase flow rate 1.0 ml  min1, and volume of injection 20 ml. Acknowledgements: Mr. Nebiu acknowledges financial support by a grant of the OeAD/University of Vienna. References Biwas SC, Biwas BN, Rudra A, Pan A (2002) A comparative study of metoclopramide, glycopyrrolate and combination of metoclopramide and glycopyrrolate in control of intraoperative and postoperative nausea and vomiting after spinal anesthesia for cesarean section. J Obstetrics Gyn India 52: 73–75. Cydulka RK, Emerman CL (1995) Effects of combined treatment with glycopyrrolate and albuterol in acute exacerbation of chronic obstructive pulmonary disease. Ann Emerg Med 25: 470–473. Czeche S, Elgert M, Noe C, Waelbroeck M, Mutschler E, Lambrecht G (1997) Antimuscarinic properties of the stereoisomers of glycopyrronium bromide. Life Sci 60: 1167. Demian J, Gripshover DF (1990) High-performance liquid chromatographic separation of 3-[(cyclopentylhydroxyphenylacetyl)oxy]-1,1-dimethylpyrrolidinium bromide diastereoisomers. J Liq Chromatogr 13: 779–787. Ebeid MY, Moussa BA, Abdel Malak AA (1986) Analysis of some antispasmodic drugs: oxyphencyclamine and glycopyrronium bromide. Pharm Weekbl Sci Ed 8: 252–258. Gilman MJ, Meyer L, Carter J, Slovis C (1990) Comparison of aerosolized glycopyrrolate and metaproterenol in acute asthma. Chest 98: 1095–1098. 410 Guerra A, Campillo NE, Paez JA (2006) Glycopyrrolate combinations for the treatment of respiratory diseases. Expert Opin Ther Patents 16: 723– 728. Hansel TT, Neighbour H, Erin EM, Tan AJ, Tennant RC, Maus JG, Barnes PJ (2005) Glycopyrrolate causes prolonged bronchoprotection and bronchodilatation in patients with asthma. Chest 128 : 1974–1979. Hopfgartner G, Wachts T, Bean K, Henion J (1993) High-flow ion spray liquid chromatography/mass spectrometry. Anal Chem 65: 439–446. ICH (1996) Harmonized Tripartite Guideline on Validation of Analytical Procedures: Methodology, International Conference on Harmonization. ICH (1994) Harmonized Tripartite Guideline Text on Validation of Analytical Procedures, International Conference on Harmonization. Levin RM, Wein RJ (1982) Direct measurement of the anticholinergic activity of a series of pharmacological compounds on the canine and rabbit urinary bladder. J Urol 128: 396–398. Matassa LC, Woodward D, Leavitt LK, Firby P, Beaumier P (1992) Solidphase extraction techniques for the determination of glycopyrrolate from equine urine by liquid chromatography-tandem mass spectrometry. J Chromatogr 573: 43–48. Miettinen P, Anttonen V, Aukee S, Huttunen R, Lahtinen J, Poikolainen E, Paakkonen M, Tammela T (1985) Ranitidine versus anticholinergic/antacid for duodenal ulcer. A randomized, endoscopically controlled, singleblind multicentre trial. Scand J Gastroenterol 20: 701–705. Murray GR, Calvey TN, Williams NE, Chan K (1984) Quantitative capillary gas chromatographic determination of glycopyrronium in human plasma. J Chromatogr 308: 143–151. Noe C, Mutschler E, Lambrecht G, Elgert M, Czeche S, Waelbroeck M (1999) Pure enantiomer basic-aryl-cycloalkyl-hydroxycarboxylic acid esters, process for preparing the same and their use in medicaments. PCT/ AT 97/00245/1999, 1–27. Noe C, Walter M, Verfahren zur Herstellung der R,R oder S,S konfigurierten Glykopyrronium-Isomere. Pat-Appl.: A 1896/2002 and Walter M, PhD Thesis in Progress. Ojanpera I, Vartiovaara J, Ruohonen A, Vuori E (1991) Combined use of normal and reversed-phase thin-layer chromatography in the screening for basic and quaternary drugs. J Liq Chromatogr 14: 1435–1446. Rama Sastry BV (1996) Anticholinergic drugs. In: Wolff ME (Ed), Burgers medicinal chemistry and drug discovery. Vol. 2, A Wiley Interscience Publication p. 59–117. Tang FP, Leung GN, Wan TS (2001) Analyses of quaternary ammonium drugs in horse urine by capillary electrophoresis-mass spectrometry. Electrophoresis 22: 2201–2209. The United States Pharmacopoeia, 25th ed, Rockville, MD, USA, 1995, pp. 811–812. Villetti G, Bergamaschi M, Bassani F, Bolzoni PT, Harrison S, Gigli PM, Janni A, Geppetti P, Civelli M, Patacchini R (2006) Pharmacological assessment of the duration of action of glycopyrrolate vs tiotropium and ipratropium in guinea-pig and human airways. Br J Pharmacol 148: 291–298. Weinstein RE, Weinstein AM (2000) Nonsedating formulations for allergic rhinitis which possess antihistaminic and anticholinergic activity. US Pat. No: US6086914, 3. Wohlrab J (2003) Topical formulation containing glycopyrronium bromide for the therapy and prophylaxis of hyperhidrosis. PCT Int Appl Nr: WO200301340, 24. Pharmazie 62 (2007) 6