Drug Testing and Analysis Research article Received: 22 December 2012 Revised: 25 January 2013 Accepted: 17 February 2013 Published online in Wiley Online Library (www.drugtestinganalysis.com) DOI 10.1002/dta.1472 Quantification of curcumin, demethoxycurcumin, and bisdemethoxycurcumin in rodent brain by UHPLC/ESI-Q-TOF-MS/MS after intra-nasal administration of curcuminoids loaded PNIPAM nanoparticles Niyaz Ahmad,a Musarrat Husain Warsi,a Zeenat Iqbal,a Mohd Samimb and Farhan Jalees Ahmada* An ultra high performance liquid chromatography-electrospray ionization-synapt mass spectrometric method (UHPLC/ESIQTOF-MS/MS) for the analysis of curcumin (Cur), demethoxycurcumin (DMC), bisdemethoxycurcumin (BDMC) in Wistar rat brain homogenate was developed and validated. The chromatographic separation was achieved on a Waters ACQUITY UPLC™ BEH C18 (2.1mm
100 mm; 1.7mm) column using isocratic mobile phase, consisting of acetonitrile: 10mM ammonium formate: formic acid (90:10:0.05v/v/v), at a flow rate of 0.2 ml min-1. The transitions occurred at m/z 367.0694/217.0598, 337.0717/173.0910, 307.0760/187.0844 for Cur, DMC, BDMC and m/z 307.0344/229.0677 for the IS (Nimesulide) respectively. The recovery of the analytes from Wistar rat brain homogenate was optimized using liquid-liquid extraction technique (LLE) in (ethyl acetate: chloform) mixture. The total run time was 3.0 min and the elution of Cur, DMC, BDMC occurred at 1.6, 1.75, 1.70 min, and for the IS 1.87 min, respectively. The linear dynamic range was established over the concentration range of 1.00 ng mL-1 to 1000.0 ng mL-1(r2; 0.9909  0.0011, 0.9911  0.003, and 0.9919  0.0013) for Cur, DMC, and BDMC, respectively. The intra and inter-assay accuracy in terms of % CV for Cur, DMC, and BDMC was in the range 0.47–2.20, 0.47–1.65, and0.44–2.70, respectively. The lower limit of detection (LOD) and quantitation (LOQ) for Cur, DMC, and BDMC were 0.46, 0.05, 0.16 ng mL-1 and 0.153, 0.015, 0.052 ng mL-1, respectively. Analytes were stable and the method proved to be accurate (recovery, >85%), specific and was applied to evaluate the Cur, DMC, BDMC loaded PNIPAM NPs as vehicles for nose to brain drug delivery. Copyright © 2013 John Wiley & Sons, Ltd. Keywords: curcumin; demethoxycurcumin; bisdemethoxycurcumin; UHPLC-MS/MS-ESI-Q-TOF; PNIPAM nanoparticles Introduction Curcuminoids are obtained from Indian spice turmeric (Curcuma longa L.), which are the members of the ginger family (Zingiberaceae). Plant extracts of Curcuminoids are the mixtures of curcumin (diferuloylmethane), demethoxycurcumin (p-hydroxycinnamoyl, feruloylmethane), and bisdemethoxycurcumin (di-p-hydroxycinnamoylmethane). Turmeric is a food-colouring agent, and it has been found to be a rich source of phenolic compounds.[1] Curcumin is known to exhibit antioxidant, antiinflammatory, anti-microbial, and anti-carcinogenic activities.[2,3] Curcumin is also used in the treatment of cerebral ischemia.[4] In addition, curcumin has been shown to have the possibility of slowing the progress of Alzheimer’s disease by reducing amyloid b, [5] delays the onset of kainic acid-induced seizures[6] and inhibits the formation of brain tumours.[7] Due to its poor solubility, absorption, extensive metabolism, and rapid elimination, curcumin exhibits low serum and tissue levels.[8,9] Therefore, the therapeutic efficacy of curcumin is restricted due to its short systemic retention in circulation and hence in tissues. To enhance the bioavailability of curcuminoids various strategies have been applied in terms of formulations and route of administration. For the treatment of cerebral ischemia, Drug Test. Analysis (2013) we have formulated PNIPAM (Polymeric N-isopropyl acryl amide: vinyl Pyrilidone: acrylic acid) nanoparticles loaded with curcumin (Cur), demethoxycurcumin (DMC), bisdemethoxycurcumin (BDMC) as nose-to-brain delivery. Their concentration measurements were done on ultra high performance liquid chromatography-electrospray ionization-synapt mass spectrometric method (UHPLC/ESI-Q-TOFMS/MS), and also their comparative studies have done. There are so many analytical methods including gas chromatography-mass spectrometry (GC-MS), high performance liquid chromatography (HPLC), and its coupling to mass spectrometry (LC–MS), thin layer chromatography (TLC), and capillary electrophoresis (CE), have been used to analyze the chemical content of various turmeric samples. Various methods also have * Correspondence to: Dr Farhan Jalees Ahmad, Department of Pharmaceutics, Faculty of Pharmacy, Hamdard University, New Delhi-110062, India. E-mail: farhanja_2000@yahoo.com a Department of Pharmaceutics, Faculty of Pharmacy, Department of Pharmaceutics, New Delhi-110062, India b Department of Chemistry, Faculty of Science, Hamdard University, New Delhi110062, India Copyright © 2013 John Wiley & Sons, Ltd. Drug Testing and Analysis N. Ahmad et al. been developed for the analysis of Cur in plasma, and urine samples.[9–15] Among them one method is reported for estimation of Cur in different organs.[16] But, to the best of our knowledge there is no sensitive bioanalytical method yet reported for separate estimation of Cur, DMC, and BDMC, in a picogram level in tissues and plasma, mainly in brain tissue. Therefore a hyphenated chromatographical technique with advanced feature is desired to compensate the aforesaid loopholes. However, UHPLC is a novel chromatographic technique utilizing high linear velocities, which is based on concept using columns with smaller packing (1.7–1.8 mm porous particles) and operated under high pressure (up to 15 000 psi). This is an extremely powerful approach which dramatically improves peak resolution, sensitivity and speed of analysis.[17] Consequently, UHPLC/ESI-Q-TOF-MS has been proved to be a powerful hyphenated technique for bioanalytical investigation.[17–22] In this paper, to the best of our knowledge, for the first time, a validated comparative assay of Cur, DMC, and BDMC loaded in PNIPAM nanoparticles was developed by UHPLC/ESI-Q-TOF-MS/MS and successfully implicated for bioanalytical investigations. The present method showed excellent performance for the bioanalysis of number of samples with respect to its high sensitivity and very low retention time. Material and methods Chemicals Cur, DMC, and BDMC (assigned purity >96%; Mol wt. 368.38, 338.36, 308.33, respectively) were purchased from LGC Promo Chem India Pvt. Ltd (Bangalore, India). HPLC-MS grade acetonitrile (assigned purity: 99.9%) was purchased from Sigma Aldrich (Steinheim, Germany) with lot no. 9052S. Nimesulide (Batch No. APL/WRS/16) as internal standard was obtained as a gift sample from Jubilant Clinsys Clinical Research Limited (Noida, India). Solvents used for UHPLC were of gradient grade purity and purchased from Merck (Darmstadt, Germany). MS-grade ammonium acetate (Lot No. 1411594) and ammonium formate (Batch No. T-835291) were obtained from Fluka analytical (Sigma-Aldrich, the Netherlands). Formic acid (Assigned purity >98%) was obtained from Fluka analytical (Steinheim, Germany) with lot no. 1439632. Deionized water was purified using a Milli-Q water purification system (Millipore, Bedford, MA, USA). UHPLC conditions UHPLC was performed with a Waters ACQUITY UPLCTM system (Waters Corp., MA, USA, Serial No. #F09UPB 920M) equipped with a binary solvent delivery system, and a tunable MS detector (Synapt; Waters, Manchester, UK). Chromatographic separation was performed on a Waters ACQUITY UPLCTM BEH C18 (2.1mm 100 mm; 1.7mm) column. The mobile phase for UHPLC analysis consisted of LCMS-grade acetonitrile: 10 mM ammonium formate: formic acid (90:10:0.05 v/v), which was degassed. For isocratic elution, the flow rate of the mobile phase was kept at 0.2 ml min-1 and 10 ml of sample solution was injected in each run. The total chromatographic run time was 3.0 min. Q-TOF-MS conditions MS was performed on a Waters Q-TOF Premier (Micromass MS Technologies, Manchester, UK, and Serial No. JAA 272) mass spectrometer. The Q-TOF Premier TM was operated in V mode with resolution over 32000 mass with 1.0 min scan time, and 0.02 s inter-scan delay. Argon was employed as the collision gas at a pressure of 5.3
10-5 Torr. Quantitation was performed using Synapt Mass Spectrometry (Synapt MS). The transitions occurred at m/z 367.0694/217.0598, 337.0717/173.0910, 307.0760/187.0844 for Cur, DMC, BDMC and m/z 307.0344/229.0677 for the internal standard (Nimesulide) respectively (Figures 1a, 1b, 1c and 2). The optimum values for compound-dependent parameters like trap collision energy (Trap CE) were set to 24.0, 26.0, 29.5, and 29.5 eV, respectively, for fragmentation information. The accurate mass and composition for the precursor ions and for the fragment ions were calculated using the MassLynx V 4.1 software. Quality control (QC) sample and standard sample preparation The standard stock solution of 100 mg mL-1 of Cur, DMC, and BDMC were prepared by dissolving requisite amount in methanol sonicated at 44 kHz/ 250W for 20 min. Calibration curve (CC) standards consisting of a set of ten non-zero concentrations (A–H) were prepared by 2% aqueous analytes spiking in blank rat brain homogenate (20 ml aqueous aliquots to 980 ml blank rat brain Figure 1a. Mass spectrum of (A) Curcumin parent ion (deprotonated precursor [M-H]- ions at m/z 367.0694) with chemical structure and (B) Curcumin product ion (major fragmented product ion at m/z 217.0598) showing fragmentation transitions. wileyonlinelibrary.com/journal/dta Copyright © 2013 John Wiley & Sons, Ltd. Drug Test. Analysis (2013) Quantification of curcumin, demethoxycurcumin and bisdemethoxycurcumin by UHPLC/ESI-Q-TOF-MS Drug Testing and Analysis Figure 1b. Mass spectrum of (A) Demethoxycurcumin parent ion (deprotonated precursor [M-H]- ions at m/z 337.0717) with chemical structure and (B) Demethoxycurcumin product ion (major fragmented product ion at m/z 173.0910) showing fragmentation transitions. Figure 1c. Mass spectrum of (A) Bisdemethoxycurcumin parent ion (deprotonated precursor [M-H]- ions at m/z 307.0760) with chemical structure and (B) Bisdemethoxycurcumin product ion (major fragmented product ion at m/z 187.0844) showing fragmentation transitions. Figure 2. Mass spectrum of, (A) Nimesulide (IS) precursor ion (deprotonated precursor [M-H]- ions at m/z 307.0344 with chemical structure and (B) IS product ion (major fragmented product ions at m/z 229.0677) showing fragmentation transitions. Drug Test. Analysis (2013) Copyright © 2013 John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/dta Drug Testing and Analysis N. Ahmad et al. homogenate) yielding concentration range from 1–1000 ng mL-1 for Cur, DMC, and BDMC. The final concentrations for each analyte were prepared to be 1, 2, 25, 210, 420, 640, 850, and 1000 ng mL-1. QC samples were prepared independently at four levels: 800 ng mL-1 (HQC, high quality control), 400 ng mL-1 (MQC, middle quality control), 2.9 ng mL-1 (LQC, low quality control), and 1.0 ng mL-1 (LLOQC). A 50 ng mL-1, internal standard working solution was prepared by diluting the stock solution in methanol-water (50 : 50 v/v). All the solutions were stored at 2–8  C until use. Sample preparation protocol All the solutions (CC standards, QC samples, and unknown brain homogenate samples) were freshly prepared before carrying out the experiments. The 500 ml aliquot of each sample was taken into glass tube, 50 ml of IS (50 ng mL-1) was added to each sample, and a further 200 ml of 10 mM ammonium formate solution was incorporated to mixture and vortexed at 300 rpm for 5 min. Extraction mixture (900 ml of Ethyl Acetate and 100 ml of chloroform previously separately prepared) of 5 ml was added and kept at reciprocating shaker for 20 min at 100 rpm. Spin the tubes in centrifuge at 4000 rpm for 10 min at 4  C. Transferred the approximately 4 ml of supernatant organic layer to another cleaned glass tubes and dried under a stream of nitrogen at pressure not more than 20 psi and at a temperature 50  2.0 C. The dried elute was reconstituted in 500 ml of mobile phase. Samples were transferred to vials for analysis and injected volume was 10 ml. the same instrument. The six replicates were run for LLOQC, LQC, MQC, and HQC samples. Matrix effect To study the effect of matrix on analyte quantification, six samples were prepared from six different batches of brain homogenate at LQC and HQC levels and checked for the % accuracy and precision (%CV) in both the QC samples. This was assessed by comparing the back calculated value from the QC’s nominal concentration. After specified storage conditions, samples were processed and analyzed. The matrix effect was investigated by post-extraction spike method. Peak area (A) of the analyte in spiked blank brain homogenate with a known concentration (MQC) was compared with the corresponding peak area (B) obtained by direct injection of standard in the mobile phase. The ratio (A/B
100) is defined as the matrix effect. LOD and LOQ The LLOD and LLOQ were calculated by the method based on the standard deviation (SD) of responses for triplicate blank injections of mobile phase and the slope (S) of the calibration plot. After that, LOD and LOQ were experimentally determined by diluting known concentrations of Cur, DMC, and BDMC until the average responses were approximately 3 or 10 times the standard deviation of the responses for triplicate. The values of LOD and LOQ were determined using the following formula: LOQ ¼ Std:Deviation
10 Slope (1) LOD ¼ Std:Deviation
3:3 Slope (2) Bioanalytical method validation The method validation of Cur, DMC, and BDMC in rodent brain homogenate was performed according to USFDA guidelines. [21] The linearity of the method was determined by analysis of three standard plots containing eight non-zero concentrations separately. Peak area ratios of analyte/IS were utilized for the construction of calibration curves, using weighted (1/x2) linear least squares regression of the brain concentrations and the measured peak area ratios. The lower limit of quantification (LLOQ) is the lowest concentration of the calibration curve, which could be measured with acceptable accuracy and precision. The LLOQ was determined based on the signal-to-noise ratio of 10:1. The extraction efficiency (recovery) of Cur, DMC, and BDMC were performed at LQC, MQC, and HQC levels separately. It was evaluated by comparing the mean area response of six replicates of extracted samples (spiked before extraction) to that of extracted drug free brain homogenate samples (spiked after extraction) at each QC level. The recovery of IS was similarly estimated. For determining the intra-day accuracy and precision, replicate analysis of brain samples of Cur, DMC, and BDMC were performed on the same day separately. The run consisted of a calibration curve and six replicates of LLOQC, LQC, MQC, and HQC samples. The inter-day accuracy and precision were assessed by analysis of six precision and accuracy batches on three consecutive validation days separately. Similarly, the robustness of the method was determined by making slight changes in operating conditions (Mobile Phase Composition, flow rate, and pH) using LQC, MQC, and HQC levels of QC samples separately. However, for evaluating the ruggedness of the method, one batch of precision and accuracy was run using a different column (same type) by a different analyst employing wileyonlinelibrary.com/journal/dta Ex vivo stability The stability of Cur, DMC, and BDMC in rodent brain was evaluated by analyzing six replicates of brain homogenate samples at the concentrations of 2.9 ng mL-1 (LQC) and 800 ng mL-1 (HQC) which were exposed to different conditions (time and temperature). Percentage stability was determined as; % Stability ¼ Mean reponse of stability stock
100 Mean response of standard stock (3) Long-term stability The long-term stability was assessed after storage of the standard spiked brain homogenate samples at deep freeze (-80  C) for one month. Six replicates of LQC and HQC were used for analysis. Freeze-thaw stability The freeze-thaw stability in brain homogenate was evaluated for three consecutive freeze-thaw cycles from -20  C to room temperature (+25  C). Six replicates of LQC and HQC were analyzed after undergoing three freeze-thaw cycles. Copyright © 2013 John Wiley & Sons, Ltd. Drug Test. Analysis (2013) Quantification of curcumin, demethoxycurcumin and bisdemethoxycurcumin by UHPLC/ESI-Q-TOF-MS Bench-top stability Bench-top stability was determined for 24 h storage in optimized conditions, using six sets each of LQC and HQC. The QC samples were quantified against the freshly spiked calibration curve standards. Post-processing stability Short-term stability was determined after the exposure (of processed samples) at 10  C for 24 h in an autosampler using six sets each of LQC and HQC. After specified storage conditions, samples were processed and analyzed. The analytes are considered to be stable when the precisions are below 15% and the accuracies are in the range of 85–115%, respectively, for both levels.[23] In vivo study Experimental animal The study was initiated after receiving approval from Animal Ethical Committee, Jamia Hamdard (New Delhi, India) and confirms to National Guidelines on the Care and Use of Laboratory Animals (protocol approval No: 847). Wistar rats (n = 6; 300–400 g, 8–10 weeks old) were kept in an environmentally controlled room (temperature: 25  2  C, humidity: 60  5%, 12 h dark–light cycle) for at least one week before the experiments. Animals were fed on a standard pelleted diet and water was provided ad libitum. The rats fasted overnight before the day of the experiment. Experimental protocol The animal protocol used in this study was approved by the Institutional Animal Ethics Committee, Jamia Hamdard. Rats were fasted for 12 h with free access to water prior to the pharmacokinetic investigation. The bioanalytical method was implicated for quantitative estimation of Cur, DMC, and BDMC in the Wistar rats’ striatum after intra-nasal administration of Cur, DMC, and BDMC solutions (100 mg kg-1 dissolved in water) and PNIPAM nanoparticles formulations (NNcur, NNDMC, NNBDMC, (100 mg kg-1)), separately. After 1 h, rats from different groups (control, API solution, and nanoparticles treated) were sacrificed and striatum was isolated. Further, striatum were homogenized (10% w/v in phosphate buffer, pH 7.4), centrifuged (2500
g; 10 min; 20  C) and the supernatant fractions were taken out. The collected brain striatum samples were preserved for investigation at -80 C until analysis. Drug Testing and Analysis tried for mobile phase selection but they didn’t provide good chromatographic resolution. Although various buffer systems were studied, the sharp peak with better signal response was observed for ammonium formate: formic acid (10:0.05% v/v). The MS full-scan spectra for Cur, DMC, and BDMC showed deprotonated daughter [M-H]- ions at m/z 367.06,337.07,307.07 and their daughter ion mass spectra at m/z 217.05,173.09,187.08, respectively (Figures 1a, 1b, 1c). During direct infusion, the mass spectra of IS showed precursor ion peaks at m/z 307.03 as [M-H]ions and most abundant product ions at m/z 229.06 (Figure 2). The optimum collision energies employed were 24.0, 26.0, 29.5, and 29.5 eV for Cur, DMC, BDMC, and IS, respectively. Quantification was done on the basis of main product ions. Identical capillary voltage of 4.3 kV was used for monitoring the precursor ions. The most widely employed biological sample preparation techniques are liquid-liquid extraction (LLE), protein precipitation (PPT), and solid-phase extraction (SPE). In the early stage of method development, a PPT method was employed to separate API but PPT separation is not an accurate method for separating the API from brain striatum samples, because strong ion suppression occurred from the endogenous substances in brain homogenate. Although it could be decreased by chromatographic separation, the run time would be sacrificed. Finally, LLE procedures were used to prepare Cur, DMC, and BDMC striatum samples in this study. To obtain optimum recovery, seven organic extraction solvents were evaluated including chloroform, ethyl acetate, dichloromethane, tertiary butyl methyl ether (TBME), diethyl ether and n-hexane. It was found that no solvent could yield the highest recovery alone, but the extraction mixture (900 ml of ethyl acetate and 100 ml of chloroform) showed the highest recovery (>85%) for Cur, DMC, and BDMC, and IS. Chromatogram of blank brain homogenate (extracted and reconstituted) is shown in Figure 3a. The elution of Cur, DMC, and BDMC spiked brain homogenate sample occurred at 1.6, 1.75, 1.70 min (Figures 3b, 3c, 3d) and IS (50 ng mL-1) at 1.87 min (Figure 3e). Bioanalytical method validation Linearity The calibration curves of Cur, DMC, and BDMC were linear over the concentration range of 1-1000 ng mL-1. The least squares regression analysis gave the linear r2 ≥ 0.99. The accuracy and precision (%CV) observed for the calibration curve standards of Cur, DMC, and BDMC ranged from 96.29–99.08, 95.38–99.03, and 94.22–99.06% and 0.47–2.20, 0.47–1.65, and 0.44–2.70, respectively. Accuracy and precision Results and discussion Cur, DMC, and BDMC are low molecular weight compounds (MW; 368.38, 338.35, 308.33, respectively), containing ketonic group in their structure. Due to the presence of ketone group in the molecule, Cur, DMC, and BDMC exhibited favourable sensitivity in negative ion mode detection. Optimum chromatographic separation of Cur, DMC, and BDMC, were achieved by acetonitrile: 10mM ammonium formate: formic acid (90:10:0.05 v/v/v), with a flow rate of 0.2 ml min-1. Baseline separation of Cur, DMC, and BDMC, and IS was obtained within runtime of 3.0 min, without any interference. Methanol and isopropyl alcohol have been also Drug Test. Analysis (2013) No endogenous peak was observed at the retention time of the both analytes and IS for any of the batches. Representative chromatogram (Figure 3a) of extracted blank brain homogenate fortified with IS and blank brain homogenate fortified with Cur, DMC, and BDMC demonstrates the selectivity of the method. The repeatability of the analytical method was good and precise, and the % CV values were obtained between 0.44–2.72 for intrabatch and 0.46–1.97 for inter-batch for all the QC levels of Cur, DMC, and BDMC. The accuracy results for intra-batch and interbatch were within the range of 94.22–99.08% and 94.72–99.07%, respectively for all QC levels of Cur, DMC, and BDMC. The detailed results are presented in (Tables 1a, 1b, 1c). Copyright © 2013 John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/dta Drug Testing and Analysis N. Ahmad et al. Figure 3. Typical chromatograms of (A) blank brain homogenate Extracted, (B) Brain Homogenate Extracted Curcumin, (C) Brain Homogenate Extracted Demethoxycurcumin, (D) ) Brain Homogenate Extracted Bisdemethoxycurcumin, and (E) ) Brain Homogenate Extracted Nimesulide IS (100 ng mL-1) extracted after spiking with Wistar rat-brain homogenate by selective reaction monitoring scan mode. Table 1a. Precision and Accuracy Data for Curcumin Intra-batch QC ID LOQQC LQC MQC HQC Inter-batch Theoretical content (ng mL-1) Mean concentration observedSD (ng mL-1)a Accuracy (%) CV (%) Mean concentration observedSD (ng mL-1)a Accuracy (%) CV (%) 1.00 2.90 400.00 800.00 0.970.012 2.830.061 393.503.824 792.603.741 96.29 97.41 98.37 99.08 1.17 2.20 0.97 0.47 0.970.011 2.840.051 393.053.922 792.534.134 96.33 98.02 98.26 99.07 1.21 1.74 1.00 0.52 Table 1b. Precision and Accuracy Data for Demethoxycurcumin(DMC) Intra-batch QC ID LOQQC LQC MQC HQC Inter-batch Theoretical content (ng mL-1) Mean concentration observedSD (ng mL-1)a Accuracy (%) CV (%) Mean concentration observedSD (ng mL-1)a Accuracy (%) CV (%) 1.00 2.90 400.00 800.00 0.960.012 2.820.041 393.463.804 792.603.691 95.38 97.18 98.36 99.03 1.36 1.65 0.97 0.47 0.970.013 2.840.042 392.533.661 792.244.061 95.87 97.90 98.13 99.03 1.30 1.47 0.93 0.51 Robustness The system suitability was checked by studying the effect of various parameters on the percentage CV and recovery of Cur, DMC, and BDMC. The parameters include the flow rate (0.199, 0.200, and 0.210, ml min-1), mobile phase ratio (acetonitrile: 10 mM ammonium formate: formic acid (90.9:9.9:0.04 v/ v/v), 90:10:0.05 v/v/v and 89.9:10.1:0.06 and pH of the mobile phase (6.3, 6.4 and 6.5).The low values of % CV of Cur, DMC, and BDMC wileyonlinelibrary.com/journal/dta (0.23 to 4.06, 0.28 to 2.19, and 0.29 to 2.31) obtained after introducing small deliberate changes in the developed UHPLC method indicated the robustness of the method (Tables 2a, 2b, 2c). Ruggedness One complete precision and accuracy batch for Cur, DMC, and BDMC was processed and analyzed by different analysts using Copyright © 2013 John Wiley & Sons, Ltd. Drug Test. Analysis (2013) Drug Testing and Analysis Quantification of curcumin, demethoxycurcumin and bisdemethoxycurcumin by UHPLC/ESI-Q-TOF-MS Table 1c. Precision and Accuracy Data for Bisdemethoxycurcumin(BDMC) Intra-batch QC ID Inter-batch Theoretical content (ng mL-1) Mean concentration observedSD (ng mL-1)a Accuracy (%) CV (%) Mean concentration observedSD (ng mL-1)a Accuracy (%) CV (%) 1.00 2.90 400.00 800.00 0.950.031 2.840.041 392.583.424 792.253.462 94.22 97.82 98.15 99.03 2.72 1.27 0.87 0.44 0.960.022 2.840.032 392.333.561 792.513.644 94.72 98.05 98.08 99.06 1.97 1.17 0.91 0.46 LOQQC LQC MQC HQC a Mean of six replicates at each concentration (n = 6); CV (%): Coefficient of variance (percentage) ; Theoretical contents; LLOQ: 1.00 ng mL-1, LQC: 2.90 ng mL-1; MQC: 400 ng mL-1; and HQC: 800 ng mL-1. Table 2a. Robustness of the Method for Curcumin (A) Robustness LQCSD (2.90 ng mL-1) Conditions MQCSD (400 ng mL-1) HQCSD (800 ng mL-1) Mobile Phase[ACN:10mM Ammonium Formate: Formic Acid (90:10:0.05v/ v/v)] Negative level (90.9:9.9:0.04, n=3) 2.650.12 (3.77%) 381.252.51 (0.66%) Zero level (90:10:0.05, n=3) 2.820.02 (0.71%) 394.222.15 (0.55%) Positive level (89.9:10.1:0.06, n=3) 2.710.04 (1.53%) 383.172.04 (0.53%) 776.873.79 (0.49%) 789.513.07 (0.39%) 779.523.11 (0.40%) Flow Rate (0.2 ml /min) Negative level (0.199, n=3) Zero level (0.2, n=3) Positive level (0.210, n=3) 2.640.11 (3.79%) 2.810.02 (0.54%) 2.700.04 (1.54%) 380.942.61 (0.69%) 392.931.99 (0.51%) 382.052.08 (0.54%) 775.832.50 (0.32%) 788.152.59 (0.33%) 778.222.94 (0.38%) pH of Mobile Phase(Default pH=6.4) Negative level (6.3, n=3) Zero level (6.4, n=3) Positive level (6.5, n=3) 2.610.11 (4.06%) 2.800.04 (1.25%) 2.700.04 (1.50%) 378.972.73 (0.72%) 391.502.53 (0.65%) 380.532.55 (0.67%) 774.181.78 (0.23%) 787.832.46 (0.31%) 776.603.06 (0.39%) (B) Ruggedness QC ID LOQQC LQC MQC HQC a Theoretical content (ng mL-1) 1.00 2.90 400.00 800.00 Mean concentration observed (ng mL-1)a 0.970.02 2.790.07 394.223.09 788.545.54 Accuracy (%) 95.54 96.09 97.38 98.57 CV (%) 1.94 2.49 0.78 0.70 Mean of six replicates at each concentration (n = 6); CV (%): Coefficient of variance (percentage) = standard deviation divided by mean concentration found
100; Theoretical contents; LOQQC: 1.00 ng mL-1, LQC: 2.90 ng mL-1; MQC: 400 ng mL-1; and HQC: 800 ng mL-1. different column and different sets of solutions. The mean accuracy (%) (n = 6) for drug ranged from (95.54–98.57, 93.67–98.67, and 94.33–98.73) and correlation of variance (%) from (0.32–4.09, 0.70–2.49, and 0.44–2.29), respectively (Tables 2a, 2b, 2c). Matrix effect Matrix effect is due to co-elution of some endogenous components present in biological samples which interferes the peak retention on actual position. The matrix effect (A/B
100) for Cur, DMC, and BDMC at LQC was 1.88, -1.86, and 0.97 % (n = 6 each; % CV 3.89, 3.73, and 2.85) respectively, however at HQC, it was -3.57, 0.13, and -2.03 (n = 6 each; % CV 4.53, 3.05, and 2.38) respectively. The CV (%) <5 suggested that the method was free from matrix effect. No significant ion suppression or enhancement from brain matrix was shown for Cur, DMC, and BDMC in post-column infusion experiments with LLE. Here 10 mM ammonium formate solution was used as a protein precipitation agent. Drug Test. Analysis (2013) LOD and LOQ LOD and LOQ were experimentally estimated by analysis of samples spiked with brain homogenate serially diluted Cur, DMC, and BDMC standard until the signal-to-noise ratio reached 3 and 10, respectively. LOD and LOQ were found to be 0.46, 0.05, 0.16 ng mL-1 and 0.153, 0.015, 0.052 ng mL-1, respectively. Ex vivo stability The ex vivo stability evaluation of analyte was designed to cover anticipated conditions of handling of the experimental samples. Tables 3a, 3b, 3c summarizes the results of stability experiments, which showed that Cur, DMC, and BDMC were stable during all storage conditions (long term, freeze-thaw, bench-top, and postprocessing stability). The analytes stability in brain sample was investigated at two quality control levels (in LQC and HQC). The recovery of the analytes relative to that at time zero is reviewed. After one month of storage (long-term stability), the recovery of Copyright © 2013 John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/dta Drug Testing and Analysis N. Ahmad et al. Table 2b. Robustness of the Method for Demethoxycurcumin(DMC) (A) Robustness LQCSD (2.90 ng mL-1) Conditions MQCSD (400 ng mL-1) Mobile Phase[ACN:10mM Ammonium Formate: Formic Acid (90:10:0.05 v/v/v)] Negative level (90.9:9.9:0.04, n=3) 2.660.02 (0.65%) 380.793.06 (0.80%) Zero level (90:10:0.05, n=3) 2.840.02 (0.70%) 393.183.04 (0.77%) Positive level (89.9:10.1:0.06, n=3) 2.730.04 (1.52%) 381.994.00 (1.05%) Flow Rate (0.2 ml /min) Negative level (0.199, n=3) 2.630.06 (2.19%) 379.922.61 (0.69%) Zero level (0.2, n=3) 2.830.03 (1.08%) 392.522.24 (0.57%) Positive level (0.210, n=3) 2.720.05 (1.70%) 380.022.03 (0.53%) pH of Mobile Phase(Default pH=6.4) Negative level (6.3, n=3) 2.600.05 (1.82%) 378.522.14 (0.57%) Zero level (6.4, n=3) 2.820.04 (1.34%) 391.502.53 (0.65%) Positive level (6.5, n=3) 2.700.05 (1.87%) 380.532.55 (0.67%) (B) Ruggedness QC ID Theoretical content (ng mL-1) Mean concentration observed (ng mL-1)a LOQQC 1.00 0.940.04 LQC 2.90 2.780.04 MQC 400.00 394.52.82 HQC 800.00 789.372.55 a HQCSD (800 ng mL-1) 775.713.25 (0.42%) 789.843.24 (0.41%) 778.993.46 (0.44%) 774.632.22 (0.29%) 788.483.00 (0.38%) 777.772.19 (0.28%) 772.822.14 (0.28%) 788.112.89 (0.37%) 775.534.07 (0.52%) Accuracy (%) 93.67 95.98 97.48 98.67 CV (%) 4.09 1.28 0.71 0.32 Mean of six replicates at each concentration (n = 6); CV (%): Coefficient of variance (percentage) = standard deviation divided by mean concentration found
100; Theoretical contents; LOQQC: 1.00 ng mL-1, LQC: 2.90 ng mL-1; MQC: 400 ng mL-1; and HQC: 800 ng mL-1. Table 2c. Robustness of the Method for Bisdemethoxycurcumin(BDMC) (A)Robustness LQCSD (2.90 ng mL-1) Conditions MQCSD (400 ng mL-1) HQCSD (800 ng mL-1) Mobile Phase[ACN:10mM Ammonium Formate: Formic Acid (90:10:0.05 v/v/v)] Negative level (90.9:9.9:0.04, n=3) 2.650.05 (1.90%) 380.833.01 (0.79%) Zero level (90:10:0.05, n=3) 2.860.05 (1.76%) 393.922.25 (0.57%) Positive level (89.9:10.1:0.06, n=3) 2.720.04 (1.49%) 382.972.29 (0.60%) 777.123.40 (0.44%) 788.833.07 (0.39%) 780.522.25 (0.29%) Flow Rate (0.2 mL /min) Negative level (0.199, n=3) Zero level (0.2, n=3) Positive level (0.210, n=3) 2.640.04 (1.53%) 2.850.03 (1.01%) 2.710.04 (1.40%) 377.533.01 (0.80%) 390.192.17 (0.56%) 380.712.15 (0.57%) 775.572.51 (0.32%) 788.152.58 (0.33%) 778.363.19 (0.41%) pH of Mobile Phase(Default pH=6.4) Negative level (6.3, n=3) Zero level (6.4, n=3) Positive level (6.5, n=3) 2.610.06 (2.31%) 2.820.06 (2.02%) 2.650.06 (2.14%) 371.592.97 (0.80%) 389.152.06 (0.53%) 379.833.58 (0.94%) 773.842.37 (0.31%) 787.962.50 (0.32%) 776.862.81 (0.36%) Theoretical content (ng mL-1) 1.00 2.90 400.00 800.00 Mean concentration observed (ng mL-1)a 0.940.02 2.790.05 394.034.22 789.863.46 (B) Ruggedness QC ID LOQQC LQC MQC HQC a Accuracy (%) 94.33 96.26 97.36 98.73 CV (%) 2.29 1.80 1.07 0.44 Mean of six replicates at each concentration (n = 6); CV (%): Coefficient of variance (percentage) = standard deviation divided by mean concentration found
100; Theoretical contents; LOQQC: 1.00 ng mL-1, LQC: 2.90 ng mL-1; MQC: 400 ng mL-1; and HQC: 800 ng mL-1. Cur, DMC, and BDMC were 99.29, 97.89, and 97.20% (LQC); 98.80, 98.84, and 98.75% (HQC). After 1, 2, and 3 cycles of freeze-thaw (freeze-thaw stability), Cur, DMC, and BDMC were recovered in the range of 97.87–99.65, 96.84–98.95, 97.55–99.30% (LQC) and 99.17–99.72, 98.91–99.48, 98.81–99.66% (HQC). After 24 h (bench– top stability) the recovery of Cur, DMC, and BDMC were 99.65, 97.19, and 97.55% (LQC), and 99.67, 98.89, and 98.92% (HQC). The wileyonlinelibrary.com/journal/dta recovery (post-processing stability) of Cur, DMC, and BDMC were 99.65, 98.60, and 99.65% (LQC) and 99.77, 99.88, and 99.77% (HQC). In vivo study The developed and validated UHPLC/ESI-Q-TOF-MS/MS method described above had been applied successfully to estimate Cur, Copyright © 2013 John Wiley & Sons, Ltd. Drug Test. Analysis (2013) Quantification of curcumin, demethoxycurcumin and bisdemethoxycurcumin by UHPLC/ESI-Q-TOF-MS Drug Testing and Analysis Table 3a. Ex vivo stability data for Curcumin Conditions LQCSD (2.90 ng mL-1) Long term stability; recovery (ng) after storage (-80  C) Previous day 2.820.03 30th Day 2.800.03 (99.29%) Freeze–thaw stress; recovery (ng) after freeze–thaw cycles (-20  C to 25 C) Pre-Cycle 2.820.03 First Cycle 2.810.03 (99.65%) Second Cycle 2.790.02 (98.94%) Third Cycle 2.760.01 (97.87%) Heating-cooling stress; recovery (ng) after Heating-cooling cycles (50  C to 4 C) Pre-Cycle 2.820.03 First Cycle 2.800.02 (99.29%) Second Cycle 2.760.03 (99.87%) Third Cycle 2.700.03 (95.74%) Bench top stability; recovery (ng) at room temperature (25  C) 0 hr 2.820.03 24 hr 2.810.02 (99.65%) Post processing stability; recovery (ng) after storage in auto sampler (4  C) 0 hr 2.820.03 24 hr 2.810.02 (99.65%) HQCSD (800.00 ng mL-1) 797.931.40 788.372.36 (98.80%) 797.931.40 795.671.59 (99.72%) 794.801.15 (99.61%) 791.322.29(99.17) 797.931.40 793.673.80 (99.47%) 788.323.73 (98.80%) 781.403.56 (97.93%) 797.931.40 795.290.38 (99.67%) 797.931.40 796.111.06 (99.77%) Values (MeanSD) are derived from six replicates. Figures in parenthesis represent analyte concentration (%) relative to time zero. Theoretical contents; LOQQC: 1.00 ng mL-1, LQC: 2.9 ng mL-1; MQC: 400 ng mL-1; and HQC: 800 ng mL-1. Table 3b. Ex vivo stability data for Demethoxycurcumin Conditions LQCSD (2.90 ng mL-1) HQCSD (800.00 ng mL-1) Long term stability; recovery (ng) after storage (-80  C) Previous day 2.850.02 30th Day 2.790.03 (97.89%) 795.683.12 786.461.57 (98.84%) Freeze–thaw stress; recovery (ng) after freeze–thaw cycles (-20  C to 25 C) Pre-Cycle 2.850.02 First Cycle 2.820.02 (98.95%) Second Cycle 2.790.02 (97.89%) Third Cycle 2.760.01 (96.84%) 795.683.12 791.553.08 (99.48%) 788.742.32 (99.13%) 786.972.85 (98.91%) Heating-cooling stress; recovery (ng) after Heating-cooling cycles (50  C to 4 C) Pre-Cycle 2.850.02 First Cycle 2.830.02 (99.30%) Second Cycle 2.770.02 (97.19%) Third Cycle 2.730.03 (95.79%) 795.683.12 792.382.38 (99.59%) 787.640.97 (98.99%) 783.423.02 (98.46%) Bench top stability; recovery (ng) at room temperature (25  C) 0 hr 2.850.02 24 hr 2.770.02 (97.19%) 795.683.12 786.812.29 (98.89%) Post processing stability; recovery (ng) after storage in auto sampler (4  C) 0 hr 2.850.02 24 hr 2.810.04 (98.60%) 795.683.12 794.734.23 (99.88%) Values (MeanSD) are derived from six replicates. Figures in parenthesis represent analyte concentration (%) relative to time zero. Theoretical contents; LOQQC: 1.00 ng mL-1, LQC: 2.9 ng mL-1; MQC: 400 ng mL-1; and HQC: 800 ng mL-1. DMC, and BDMC concentrations in striatal tissue of Wistar rats. The mean value of Cur, DMC, and BDMC concentration (ng mg-1 protein) in striatal tissue of rats treated with curcuminoids loaded PNIPAM nanoparticles by intra-nasal administration is presented in Table 4. The level of Cur, DMC, and BDMC in saline control animal, was found to be 0.8370  0.0307 Drug Test. Analysis (2013) ng mg-1 of protein (3.67%) where as Cur, DMC, and BDMC nanoparticles treated showed a maximum level of 9.4493  0.1269**ng mg-1, 9.6600  0.4058** ng mg-1, and 9.5200  0.3114** ng mg-1 (P < 0.001). The highly significant differences in Cur, DMC, and BDMC somehow proved the effectiveness of intra-nasal route for brain drug delivery. Copyright © 2013 John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/dta Drug Testing and Analysis N. Ahmad et al. Table 3c. Ex vivo stability data for Bisdemethoxycurcumin(BDMC) Conditions LQCSD (2.90 ng mL-1) HQCSD (800.00 ng mL-1) Long term stability; recovery (ng) after storage (-80  C) Previous day 2.860.01 30th Day 2.780.03 (97.20%) 795.692.90 785.731.48 (98.75%) Freeze–thaw stress; recovery (ng) after freeze–thaw cycles (-20  C to 25 C) Pre-Cycle 2.860.01 First Cycle 2.840.01 (99.30%) Second Cycle 2.820.02 (98.60%) Third Cycle 2.790.03 (97.55%) 795.692.90 792.993.05 (99.66%) 789.853.33 (99.27%) 786.242.48 (98.81%) Heating-cooling stress; recovery (ng) after Heating-cooling cycles (50  C to 4 C) Pre-Cycle 2.860.01 First Cycle 2.820.02 (98.60%) Second Cycle 2.790.04 (97.55%) Third Cycle 2.750.03 (96.15%) 795.692.90 793.033.07 (99.67%) 788.462.47 (99.09%) 783.093.65 (98.42%) Bench top stability; recovery (ng) at room temperature (25  C) 0 hr 2.860.01 24 hr 2.790.04 (97.55%) 795.692.90 787.122.24 (98.92%) Post processing stability; recovery (ng) after storage in auto sampler (4  C) 0 hr 2.860.01 24 hr 2.850.01 (99.65%) 795.692.90 794.321.90 (99.83%) Values (MeanSD) are derived from six replicates. Figures in parenthesis represent analyte concentration (%) relative to time zero. Theoretical contents; LOQQC: 1.00 ng mL-1, LQC: 2.9 ng mL-1; MQC: 400 ng mL-1; and HQC: 800 ng mL-1. Table 4. Estimation of Cur, DMC, AND BDMC in brain (ng/mg of protein) Specification Controle API solution of Cur NNcur, , API solution of DMC NNDMC API solution of BDMC NNBDMC Mean value of Curcumin/DMC/BDMC (SD) 0.83700.0307 4.05330.1693* 9.44930.1269** 3.96530.1753* 9.66000.4058** 3.03500.0976* 9.52000.3114** %CV 3.67 4.1763 1.3425 4.3966 4.2007 3.2157 3.2710 The mean value (Mean  sd, n = 6) of Cur, DMC, BDMC was represented in nanogram (ng) per milligram (mg) of striatum protein. Any significant difference at each time point was calculated according to the Tukey-Kramer Multiple Comparison Test: * P<0.005, ** P<0.001 Conclusion References A sensitive, selective, and rapid method was developed for the quantification of Cur, DMC, and BDMC in rodent brain homogenate by UHPLC/ESI-Q-TOF-MS/MS. The detection limit in picogram levels shows the potential of this method. After the extraction procedures, the recovery of three analytes was greater than >85% in brain homogenate. The results obtained for linearity, accuracy and precision, long term, bench-top, post-processing stability and stability following freeze-thaw cycles and matrix effect were found to be within the acceptable limits. The assay was successfully employed for in vivo studies in Wistar-rat-brain homogenate with acceptable precision, adequate sensitivity and satisfied accuracy. It would be, furthermore, applicable for clinical studies. Acknowledgement The authors are thankful to Council for Scientific and Industrial Research (CSIR), Pusa Road, New Delhi for the granting CSIR-SRF which supported this study. wileyonlinelibrary.com/journal/dta [1] M.-H. Teiten, S. Eifes, M. Dicato, M. Diederich. Curcumin-The paradigm of a multi-target natural compound with applications in cancer prevention and treatment. Toxins 2010, 2, 128. [2] P. Anand, A.B. Kunnumakkara, R.A. Newman, B.B. Aggarwal. Bioavailability of curcumin: Problems and promises. Mol. Pharm. 2007, 4, 807. [3] M. Thiyagarajan, C.L. Kaul, S.S. Sharma. Neuroprotective efficacy and therapeutic time window of peroxynitrite decomposition catalysts in focal cerebral ischemia in rats. Br. J. Pharmacol. 2004, 142, 899. [4] J. Jiang, W. Wang, Y.J. Sun, M.H.F. Li, D.Y. Zhu. Neuroprotective effect of curcumin on focal cerebral ischemic rats by preventing blood– brain barrier damage. Eur. J. Pharmacol. 2007, 561, 54. [5] M.G. Alloza, L.A. Borrelli, A. Rozkalne, B.T. Hyman, B.J. Bacskai. Curcumin labels amyloid pathology in vivo, disrupts existing plaques, and partially restores distorted neurites in an Alzheimer mouse model. J. Neurochem. 2007, 102, 1095. [6] Y. Sumanont, Y. Murakami, M. Tohda, O. Vajragupta, H. Watanabe, K. Matsumoto. Effects of manganese complexes of curcumin and diacetylcurcumin on kainic acid-induced neurotoxic responses in the rat hippocampus. Biol. Pharm. Bull. 2007, 30, 1732. [7] S. Purkayastha, A. Berliner, S.S. Fernando, B. Ranasinghe, I. Ray, H. Tariq, et al. Curcumin blocks brain tumor formation. Brain Res. 2009, 1266, 130. Copyright © 2013 John Wiley & Sons, Ltd. Drug Test. Analysis (2013) Quantification of curcumin, demethoxycurcumin and bisdemethoxycurcumin by UHPLC/ESI-Q-TOF-MS [8] P. Anand, H.B. Nair, B. Sung, A.B. Kunnumakkara, V.R. Yadav, R.R. Tekmal, et al. Design of curcumin-loaded PLGA nanoparticles formulation with enhanced cellular uptake, and increased bioactivity in vitro and superior bioavailability in vivo. Biochem. Pharmacol. 2010, 79, 330. [9] D.D. Heath, M.A. Pruitt, D.E. Brenner, C.L. Rock. Curcumin in plasma and urine: Quantitation by high-performance liquid chromatography. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 2003, 783, 287. [10] Y. Pak, R. Patek, M. Mayersohn. Sensitive and rapid isocratic liquid chromatography method for the quantitation of curcumin in plasma. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 2003,796, 339. [11] D.D. Heath, M.A. Pruitt, D.E. Brenner, A.N. Begumc, S.A. Frautschy, C.L. Rocke. Tetrahydrocurcumin in plasma and urine: Quantitation by high performance liquid chromatography. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 2005, 824, 206. [12] H. Jiang, B.N. Timmermann, D.R. Gang. Use of liquid chromatography– electrospray ionization tandem mass spectrometry to identify diarylheptanoids in turmeric (Curcuma longa L.) rhizome. J. Chromatogr. A 2006, 1111, 21. [13] K.Y. Yang, L.C. Lin, T.Y. Tseng, S.C. Wang, T.H. Tsai. Oral bioavailability of curcumin in rat and the herbal analysis from Curcuma longa by LC–MS/MS. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 2007, 853, 183. [14] V. Kakkara, S. Singhb, D. Singlab, S. Sahwney, A.S. Chauhanc, G. Singhc, et al. Pharmacokinetic applicability of a validated liquid chromatography tandem mass spectroscopy method for orally administered curcumin loaded solid lipid Nanoparticles to rats. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 2010, 878, 3427. [15] A. Liu, H. Lou, L. Zhao, P. Fan. Validated LC/MS/MS assay for curcumin and tetrahydrocurcumin in rat plasma and application to Drug Test. Analysis (2013) [16] [17] [18] [19] [20] [21] [22] [23] Drug Testing and Analysis pharmacokinetic study of phospholipid complex of curcumin. J. Pharm. Biomed. Anal. 2006, 40, 720. Y.M. Tsaia, C.F. Chiena, L.C. Lina, T.H. Tsaia. Curcumin and its nanoformulation: The kinetics of tissue distribution and blood–brain barrier penetration. Int. J. Pharm. 2011, 416, 331. I.D. Wilson, J.K. Nicholson, J. Castro-Perez, J.H. Granger, K.A. Johnson, B.W. Smith, et al. High resolution “ultra performance” liquid chromatography coupled to q-TOF mass spectrometry as a tool for differential metabolic pathway profiling in functional genomic studies. J. Proteome Res. 2005, 4, 591. L. Nov´akov, L. Matysov, P. Solich. Advantages of application of UPLC in pharmaceutical analysis. Talanta 2006, 68, 908. R. Plumb, J. Castro-Perez, J. Gragner, I. Beattie, K. Joncour, A. Wright. Ultra-performance liquid chromatography coupled to quadrupoleorthogonal time-of-flight mass spectrometry. Rapid Commun. Mass Spectrom. 2004, 19, 2331. FDA. Guidance for Industry Bioanalytical Method Validation. Available at: http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatory Information/Guidances/UCM070107.pdf [Accessed on June 2010]. M. Faiyazuddin, N. Ahmad, Z. Iqbal, S. Talegaonkar, A. Bhatnagar, R.K. Khar, et al. Development and validation of UHPLC/ESI-Q-TOF-MS method for terbutaline estimations in experimental rodents: Stability effects and plasma pharmacokinetics. Curr. Pharm. Anal. 2012, 7, 189. M. Faiyazuddin, N. Ahmad, Z. Iqbal, S. Talegaonkar, A. Bhatnagar, R.K. Khar, et al. Stabilized terbutaline submicron drug aerosol for deep lungs deposition: Drug assay, pulmonokinetics and biodistribution by UHPLC/ESI-q-TOF-MS method. Int. J. Pharm. 2012, 434, 59. L.C. Dickson, J.D. MacNeil, S. Lee, A.C. Fesser. Determination of betaagonist residues in bovine urine using liquid chromatographytandem mass spectrometry. J. AOAC Int. 2005, 88, 46. Copyright © 2013 John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/dta