Translate this page into:
Detection, isolation and characterization of principle synthetic route indicative impurity in telmisartan
⁎Corresponding author at: Analytical Development Laboratory, Piramal Healthcare Ltd., Ennore, Chennai 600 057, India. vsvnreddy@yahoo.com (V. Srinivasan) srinivasan.viswanathan@piramal.com (V. Srinivasan)
This article was originally published by Elsevier and was migrated to Scientific Scholar after the change of Publisher.
Peer review under responsibility of King Saud University.
Abstract
An unknown impurity was detected in the telmisartan bulk drug (active pharmaceutical ingredient – API) using an isocratic reversed-phase high performance liquid chromatography (HPLC). This impurity was isolated by preparative HPLC. Spectral data of the isolated impurity were collected. Based on the spectral data deriving from two dimensional nuclear magnetic spectroscopy (2D-NMR) and mass spectrometry (MS), the impurity was characterized as “methyl 4′,4′-dibromo methyl biphenyl-2-carboxylate”. The arrived structure was further confirmed by theoretical studies.
Keywords
Telmisartan
Impurity
Characterization
HPLC
MS
NMR
1 Introduction
Telmisartan, chemically 4′-[(1,4′-dimethyl-2′-propyl[2,6′-bi-1H-benzimidazol]-1′-yl)methyl][1,1′-biphenyl]-2-carboxylicacid (Merck Index, 2006) is a member of the family of drugs called angiotensin receptor blockers (ARBs), which includes losartan (Cozaar), valsartan (Diovan), irbesartan (Avapro), and candesartan (Atacand). ARBs block the ability of the chemical angiotensin II to constrict or squeeze arteries and veins. As a result, the arteries and veins enlarge and blood pressure falls. It is also used for reducing the risk of heart attack, stroke, or death from cardiovascular causes (Sharma et al., 2002; Lindholm et al., 2002; Young et al., 2004). It is believed that telmisartan’s dual mode of action may provide protective benefits against the vascular and renal damage caused by diabetes and cardiovascular disease (CVD) (Benson et al., 2004; Yusuf et al., 2008). Telmisartan was shown to be better tolerated and associated with higher treatment compliance than ramipril (Engeli et al., 2005). We now wish to report the structural characterization of this impurity. The different analytical techniques reported so far for the determination of this drug along with corresponding impurities and degradation products in biological samples and pharmaceutical formulations are by HPTLC (Potale et al., 2010; Shah et al., 2007), HPLC (Wankhede et al., 2007; Torrealday et al., 2003; Zhang et al., 2009) and LC–MS/MS (Yan et al., 2008). The general considerations for the isolation of impurities by LC-NMR and other techniques have been reviewed for the isolation of unknown impurities observed.
A sample of telmisartan which was synthesized in our laboratory when subjected to HPLC analysis by chromatographic conditions published in the USP monograph of telmisartan (USP-NF, 2010), an unknown impurity at 1.59 RRT was observed to a level of 0.09%. As per regulatory guidelines, it is necessary to know the structure of the impurity to develop a synthetic process to remove the impurity and also the knowledge of the structure and the source of the new impurity are necessary in order to develop a more robust synthetic process. It is therefore, essential to isolate and characterize unidentified impurities present in the drug sample. The structure elucidation of this impurity is object of this work.
The sample was taken for the isolation of unknown impurity by using preparative HPLC and the isolated impurity was characterized by using MS, NMR and IR. To the best of our knowledge, the impurity detected at 1.59 RRT was established for the first time.
2 Experimental
2.1 Chemicals
HPLC grade acetonitrile, HPLC grade methanol, HPLC grade monobasic potassium phosphate, HPLC grade formic acid and HPLC grade ortho phosphoric acid were purchased from Merck (Germany). Sodium-1-pentanesulfonate was purchased from BDH (UK).
2.2 Apparatus
2.2.1 Preparative high performance liquid chromatography
A Shimadzu preparative HPLC equipment with LC-8A pump, SCL-10A VP system controller, SPD-10AVP UV–VIS detector, FRC-10A fraction collector and Rheodyne injector with 5.0 mL loop (Make SHIMADZU, Japan) was used. The chromatographic separation was achieved on a Gemini C18 (Phenomenex, USA), 150 mm × 30.0 mm, 5 μm semi-preparative column using a mobile phase containing mixture of water and acetonitrile (70:30, v/v). The flow rate of the mobile phase was 20.0 mL/min and the wavelength was monitored at 230 nm. The injection volume was 1.0 mL. The test concentration for the analysis was 50 mg/mL. Mobile phase was used as diluent for the preparation of test solutions. The data were recorded using LC solution software. The eluent was monitored at 230 nm and the collected fractions were injected into analytical HPLC to confirm the retention times. The collected fractions from preparative HPLC were evaporated using a rotovapor.
2.2.2 High performance liquid chromatography (HPLC)
The HPLC system was equipped with quaternary gradient pumps, auto-sampler and auto-injector (Model LC2010CHT, Make SHIMADZU, Japan) connected to photo diode array detector controlled with LC Solution software (Make SHIMADZU, Japan). A C18 column, of 125 × 4.6 mm, 5 μm (Kromasil) and the mobile phase consist of a mixture of 2.0 g of monobasic potassium phosphate and 3.8 g of sodium-1-pentanesulfonate in 1000 mL of water, pH at 3.0 with 1 M phosphoric acid. The eluent was monitored at 230 nm and at a flow rate of 1.0 mL/min. The telmisartan drug substance was dissolved in the mobile phase as diluents and the sample was sonicated for about 5 min; the sample was further filtered through 0.2 μm syringe filter and then injected into HPLC.
2.2.3 Mass spectrometry (MS)
Direct ionization (DI) mass was performed on GCMS-2010 (Shimadzu, Japan) equipped with DI source. Sample was introduced into the ion source with detector voltage 1.25 kV, detector temperature 300 °C, ion source temperature 225 °C and ionization energy 70 eV. The data were recorded using GC solution software.
The high resolution mass spectrometer consisted of an Agilent 1200 series high performance liquid chromatography system and a micrOTOFQ mass spectrometer (High sensitivity orthogonal time-of-flight instrument; Bruker Daltonics, Germany). All samples were analyzed in the micrOTOFQ mass spectrometer equipped with an ESI source for accurate mass values. In-house compound was used as an internal reference compound, which was introduced via six port divert valve using the Hystar and micrOTOF control softwares. The mass range was calibrated with ESI tuning mix solution (Part no. G2421A) from Agilent Technologies using quadratic fit. The sample was dissolved in methanol, sonicated for 10 min and diluted further to get a concentration of 10 μg/mL. An injection volume of 5 μL of sample was introduced using an auto-sampler in the flow injection of 0.1% formic acid solution in water and acetonitrile (1:1) at a flow rate of 200 μL/min. The ESI tuning mix solution (1/10) in acetonitrile containing 10 μg/mL of in-house reference compound was infused via infusion syringe to fill the 20 μL injection loop on six port divert valve. Using micrOTOF control software, the elution of sample peak followed by that of ESI tuning mix was programed.
2.2.4 NMR and IR spectrometry
The 1H, 13C, and DEPT experiments were performed in AL 300 MHz JEOL FT NMR spectrometer. All 2D-NMR experiments, Correlation Spectroscopy (COSY), Heteronuclear Single Quantum Coherence (HSQC) and Heteronuclear Multiple Bond Coherence (HMBC) were performed using a gradient probe using CDCl3 solvent in Bruker Avance 500 mHz FT NMR spectrophotometer equipped with a 5 mm 1H/13C/X (15N) three-channel triple resonance (TXI) probe, the pulse program used was employed from the pulse program library of Bruker. The data obtained were processed and analyzed by using Bruker software. The 1H chemical shift values were reported on the δ scale in ppm, relative to TMS (δ – 0.00 ppm) and chemical shift values were reported relative to CDCl3 (δ = 77.00 ppm), as internal standards respectively. The hydrogen and carbon chemical shifts are referred to the internal tetramethylsilane (TMS). The coupling constants are expressed in Hertz.
NMR Spectral data of Impurity-1: 1H-NMR (CDCl3, 300 MHz), δ (ppm), J (Hz): 3.64 (s, 3H, OCH3), 6.71 (s, 1H, CHBr2, 7.36–7.38, (d, J = 8.42, 2H, Ar-H), 7.40–7.43 (t, J = 8.61, 1H, Ar-H), 7.47–7.52 (t, J = 7.69, 1H, Ar-H), 7.58–7.61 (d, J = 6.04, 1H, Ar-H), 7.65–7.68 (d, J = 8.42, 2H, Ar-H) and 7.92–7.95(d, J = 7.69, 1H, Ar-H). 13C-NMR, δ (ppm): 40.77 (–CHBr2, C-1), 51.92 (–OCH3, C-15), 126.18 (–CH, C-7&3), 127.62 (–CH, C-11), 128.56 (–CH, C-6&4), 130.02 (–CH, C-10), 130.57 (–C, C-9), 130.63 (–CH, C-13), 131.42 (–CH, C-12), 140.66 (–C, C-2), 141.37 (–C, C-5), 142.97 (–C, C-8) and 168.55 (C⚌O, C-14).
The IR spectra were recorded in the solid state in KBr dispersion medium using Perkin–Elmer Spectrum One FT IR spectrophotometer.
2.2.5 Theoretical methods
The theoretical calculations presented here were performed with the Gaussian-94/DFT (Pople, 1995) program on a Pentium (IV) computer system. The molecular geometry of the impurity was optimized using the HF method with the basis set 6-31G∗.
3 Results and discussion
The isolated impurity was analyzed by HPLC and the HPLC purity was found to be 97.4%. The analytical HPLC chromatograms of telmisartan along with impurities are shown in (Fig. 1). Telmisartan eluted at a retention time of 14.24 min and the impurity at 22.69 min. To get a preliminary structural insight, mass analysis was carried out on the isolated sample. The mass spectra obtained showed molecular ions of the impurity at m/z 384[(M)+] whereas the telmisartan displayed molecular ions at m/z 514[(M)+]. Thus impurity has 130 a.m.u. less than that of the molecular ion of telmisartan.
Typical chromatograms of (a) telmisartan with all impurities and (b) isolated impurity.
The impurity was separated from telmisartan by preparative HPLC so as to isolate a larger amount of impurity for suitable spectroscopic investigation. Analytical HPLC chromatogram after the isolation of impurity was shown in (Fig. 1). The NMR data of the isolated impurity were collected and the structure of the impurity has been assigned with the help of COSY, HSQC and HMBC data.
3.1 Structure elucidation of impurity
The isolated impurity was initially analyzed by IR spectrometry (Fig. 2). The important vibrational signatures are analyzed and given in Table 1. HR-MS data showed exact mass of the molecular ion as sodium adduct at m/z 406.9064 (corrected m/z 383.9166) (Calcd. 384.06 for C15H12O2Br2), which corresponds to the molecular formula C15H12O2Br2. The 1H-NMR data of impurity were compared with those of telmisartan. The 1H-NMR spectrum (Fig. 3) has a peak at 3.64 ppm integrating for 3 protons. The DEPT135 and DEPT90 experiments show a peak at 52.03 ppm which corresponds to –OCH3 group. There is a singlet at 6.71 ppm which may be an isolated –CH group since there is no splitting. There are two doublets and two triplets in the aromatic region. This shows that there could be an ortho substituted phenyl ring present in the moiety. Moreover there are two doublets integrating for two protons each. This shows clearly that there is a para-substituted phenyl ring present in the moiety. This is being assumed based on the coupling constant values. There are six signals in the aromatic region of 13C spectrum that correspond to –CH carbons and four quaternary carbons (Fig. 3). Among the six peaks two peaks having intensities almost double and this could be for four carbons thus there are eight –CH carbons which matches well with the 1H integral values of eight –CH groups. It also shows a change in the chemical shift value of C⚌O group which indicates that there could be a substitution and was not an acid (chemical shift and coupling constants are given in Section 2).
IR spectrum of impurity.
| S. no. | Compound | IR (KBr) absorption bands (cm−1) |
|---|---|---|
| 1 | Impurity | 3013(s) C–H stretch, 1714(s) C⚌O stretch, 1250(s) C–O stretch |
s – strong.
-
1H- and 13C-NMR spectra of impurity.
The HMBC spectrum shows connectivity to –OCH3 to the carbonyl carbon which comes at 168.55 ppm which generally indicates the presence of ester function. This carbonyl carbon shows connectivity to –CH group and another quaternary carbon. The presence of para and ortho substituted phenyl rings makes us to believe that it does not have imidazole moiety in this impurity compound. The MS data show that there are two bromine atoms present in it. From the NMR data it is clear that no bromine atom is directly attached to phenyl ring. The presence of –CH group in the aliphatic region of 13C-NMR makes us to believe that both Br atoms are attached to same carbon which was N-methylene in the parent compound. The 1H and 13C-NMR data show very clearly that this impurity does not have –CH2 groups which are further confirmed by DEPT135 spectrum. All these informations lead to the conclusion of the absence of imidazole moiety in this impurity molecule. So the remaining moiety in the parent compound with –CH3 and two bromine atoms should be the impurity molecule.
The mass data of telmisartan and the impurity were compared. The data indicate that the molecular ion of telmisartan is at m/z 514[(M)+] and the impurity at m/z 384[(M)+] (Fig. 4) with a characteristic isotopic abundance for two bromine atoms. The abundance ratio of the molecular ion is 1:2:1 indicating the presence of two bromines. The immediate fragment of mass m/z 304 accounts for the loss of one bromine and the abundance ratio of 1:1 indicates the presence of one bromine. From the above data it is confirmed that two bromines are present in the moiety. The further fragment of mass m/z 224 indicates a loss of one more bromine and fragment at mass m/z 165 accounts for the loss of –COOCH3 group. The residual mass m/z 152 indicates the presence of biphenyl ring. Since the moiety is having two bromine groups and the mass is almost nearer to the mass of one of the intermediate, viz, 4′-(bromomethyl) biphenyl-2-carboxylic acid methyl ester with m/z 305, it is clear that the impurity is related to the intermediate.
Mass spectrum of impurity.
Based on the 2D-NMR and MS data the structure of the impurity-1 is characterized as “methyl 4',4′-dibromo methyl biphenyl-2-carboxylate” (Fig. 5).
Impurity structure with numbering of C atoms.
3.2 Theoretical studies on the impurity
The experimentally arrived structure is further confirmed by the theoretical calculation. HF level theoretical studies on synthetic route indicative impurity is done. The optimized structure of the impurity is given in Fig. 6. The computed IR spectrum is given in Fig. 7. The important vibrational frequencies are assigned with the help of available literature values as well as from the result of computation.
Optimized structure of the impurity.
Computed IR spectrum of impurity.
4 Conclusion
In the present study, an unknown impurity was detected by HPLC in the drug substance of telmisartan. The source for the formation of this impurity is from the key starting material of telmisartan (4-methyl biphenyl-2-carboxylate), which on bromination with 1,3-dibromo-5,5-dimethyl hydantoin in the presence of Azobis isobutyro nitrile as a catalyst yields 4′-(bromomethyl) biphenyl-2-carboxylic acid methyl ester and the impurity as by product. The attempt to isolate this impurity has been successful. The impurity was isolated by preparative HPLC and characterized by spectroscopic studies. The structure of this impurity was not reported earlier in literature. This study highlights the importance of impurity profiling in a drug substance.
References
- Hypertension. 2004;43:993-1002.
- Hypertension. 2005;45:356-362.
- Lancet. 2002;359:1004-1010.
- Merck Index, 14th ed. Merck & Co, Inc., USA, 2006, p. 1569.
- Pople, J., et al., 1995. Gaussian 94, Revision C.2.
- Int. J. Pharm. Sci. Rev. Res.. 2010;2(2):35-39.
- Hypertension. 2002;40:609-611.
- Ind. J. Pharm. Sci.. 2007;69:202-205.
- J. Pharm. Biomed. Anal.. 2003;32:847-857.
- United States Pharmacopoeia, 2010. USP 33, First Supplement to the USP-NF Reissue, R-982.
- Ind. J. Pharm. Sci.. 2007;69(2):298-300.
- J. Pharm. Biomed. Anal.. 2008;48:1225-1229.
- Circulation. 2004;110:2618-2626.
- N. Engl. J. Med.. 2008;358:1547-1559.
- J. Chromatogr. B. 2009;877:3729-3733.