- Email: [email protected]
Simultaneous and Sensitive Determination of Levodopa and Carbidopa in Pharmaceutical Formulation and Human Serum by High Performance Liquid Chromatography with On-Line Gold Nanoparticles-Catalyzed Luminol Chemiluminescence Detection
CHINESE JOURNAL OF ANALYTICAL CHEMISTRY Volume 45, Issue 6, June 2017 Online English edition of the Chinese language journal
Cite this article as: Chin J Anal Chem, 2017, 45(6), e1726–e1733.
RESEARCH PAPER
Simultaneous and Sensitive Determination of Levodopa and Carbidopa in Pharmaceutical Formulation and Human Serum by High Performance Liquid Chromatography with On-Line Gold NanoparticlesCatalyzed Luminol Chemiluminescence Detection MU Chun-Lei, WU Dong, LU Hai-Feng, XIE He, ZHANG Qun-Lin* School of Pharmacy, Anhui Medical University, Hefei 230032, China
Abstract: Levodopa, the metabolic precursor of dopamine, is usually administrated in combination with carbidopa to control dopamine levels in an appropriate manner and reduce side effects in the treatment of Parkinson's disease. In this study, a selective and sensitive high performance liquid chromatography coupled with on-line gold nanoparticles-catalyzed luminol chemiluminescence method for simultaneous determination of levodopa and carbidopa was developed. This method was based on the strongly enhanced chemiluminescence signal of on-line gold nanoparticles-catalyzed luminol-H2O2 system by levodopa and carbidopa. The possible enhancement mechanism was attributed to that levodopa and carbidopa could promote the on-line formation of a large number of gold nanoparticles, which catalyzed the luminol-H2O2 chemiluminescence reaction. The good separation of levodopa and carbidopa was achieved with isocratic elution using a mixture of methanol and 0.2% aqueous phosphoric acid (5:95, V/V) within 10.5 min. Under the optimal conditions, the linear ranges of levodopa and carbidopa were 2.24–448 ng mL–1 and 4.32–1080 ng mL–1 with the detection limits of 0.89 and 1.08 ng mL–1 (S/N = 3), corresponding to 17.92 and 21.60 pg for 20 μL sample injection, respectively. The validated method was successfully applied to simultaneous quantification of levodopa and carbidopa in controlled-release tablets (Sinemet®) and human plasma. The average recoveries of levodopa and carbidopa in the tablets were 100.5% and 103.1% with the precisions (RSDs) of 2.4% and 4.0%. The recoveries of levodopa and carbidopa in human plasma ranged from 97.0% to 103.5% with RSDs of no more than 3.3%. Key Words:
1
Levodopa; Carbidopa; Chemiluminescence; On-line; Gold nanoparticles; Luminol
Introduction
Parkinson’s disease is a neurodegenerative disorder of the central nervous system. The loss of neurons in the substantia nigra, a region of the midbrain, leads to a deficiency of dopamine neurotransmitter in the brain, which may lead to Parkinson’s disease[1]. Dopamine is not effectively used for the treatment of Parkinson’s disease, because it could not
penetrate the blood-brain barrier (BBB)[2]. Levodopa (LD, Fig.1A), the metabolic precursor of dopamine, has been
Fig.1 Structures of (A) levodopa and (B) carbidopa
________________________ Received 8 April 2016; accepted 13 April 2017 *Corresponding author. Email: [email protected] This work was supported by the National Natural Science Foundation of China (No. 30973674), and the Science and Technological Fund of Anhui Province for Outstanding Youth of China (No. 1308085JGD10). Copyright © 2017, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences. Published by Elsevier Limited. All rights reserved. DOI: 10.1016/S1872-2040(17)61021-1
MU Chun-Lei et al. / Chinese Journal of Analytical Chemistry, 2017, 45(6): e1726–e1733
regarded as the standard for treating Parkinson’s disease, which is a condition precipitated by dopamine depletion in the central nervous system[3]. Unlike dopamine, LD can cross the BBB via a saturable transporter and is converted to dopamine by l-aromatic amino acid decarboxylase in the brain[4]. For the better therapeutic effect and lower toxicity, carbidopa (CD, Fig.1B) is administered in association with LD in the pharmaceutical formulations containing 10%‒25% of CD. CD is a competitive peripheral l-aromatic amino acid decarboxylase inhibitor with little or no pharmacological activity when given alone in usual dosage to Parkinson’s disease patients[5]. CD does not cross the BBB and contributes to the production of effective brain concentrations of dopamine from lower doses of LD by inhibiting the peripheral decarboxylation of LD to dopamine. In addition, the reduced peripheral formation of dopamine decreases the peripheral side effects such as nausea, vomiting and cardiac arrhythmia[6]. To release the fluctuating symptoms in clinical, the most commonly used method is to administrate with controlled-release LD/CD tablet (Sinemet®, specifications 200/50 mg), which is developed to prolong the therapeutic plasma level of LD[7,8]. Therefore, it is very important to establish a selective and sensitive method for simultaneous detection of LD and CD in pharmaceutical formulations and human serum, because of the coexistence of LD and CD in pharmaceutical formulations and the rapid fluctuations of plasma drug concentration of LD. However, the accurate and sensitive analysis of LD and CD is challenging due to their poor stability, low molecular mass, and high polarity[9]. Many analytical methods including chemiluminescence (CL)[10,11], spectrophotometry[12‒14], amperometric and voltammetric determination[2,15] have been reported for determination of LD or CD alone in various pharmaceutical preparations and biological samples. Simultaneous analysis of LD and CD was commonly achieved by high performance liquid chromatography (HPLC), because of the merits of high resolution and short analysis time. Fluorescent[16], mass spectrometric[17‒19] and electrochemical[20,21] detection have been used to couple with HPLC method for sensitive quantification of LD and CD. Among them, electrochemical detection method had the limitation of bad reproducibility, and mass spectrometry had the disadvantage of high instrument cost. The combination of HPLC method with CL detector is a high-efficiency technique and has many advantages such as high sensitivity, rapidity and superb reproducibility. To our knowledge, there is only one published literature that reported the luminol-potassium ferricyanide CL system coupled with HPLC method for determination of LD[22]. Luminol-H2O2 CL reaction, a popular CL system, has been widely applied to the detection of various substances[23‒25]. Gold nanoparticles (AuNPs) have been widely applied in CL reactions as
catalysts because of their excellent catalytic activities and facile synthesis. Zhang et al[26] first applied AuNPs to enhance the CL of luminol-H2O2 system, and the enhancement was supposed to originate from the catalysis of AuNPs, which facilitated the radical generation and electron-transfer processes taking place on the surface of AuNPs. In our previous work, a novel on-line AuNPs-catalyzed luminol-H2O2 CL detector for HPLC method was established to simultaneously determine eight phenolic compounds in red wine and catecholamines (norepinephrine, epinephrine and dopamine) at trace levels in rat brain[27,28]. In this work, it was found that LD and CD could strongly enhance the CL intensity of on-line AuNPs-catalyzed luminal-H2O2 reaction. This phenomenon allowed us to couple the highly sensitive on-line AuNPs-catalyzed luminol-H2O2 CL detection with HPLC (HPLC-nanoCL) for simultaneous quantification of LD and CD for the first time. This method exhibited the advantages such as low cost, short time and simple sample handling process. The experimental conditions for the good HPLC separation and maximal and stable CL intensities of LD and CD were systematically optimized. The possible mechanism of on-line AuNPs-catalyzed luminol-H2O2 CL enhanced by LD and CD was studied by UV-visible absorption sepctroscopy and transmission electron microscopy. The proposed HPLC-nanoCL method was applied to detect LD and CD in human serum and pharmaceutical formulation with satisfactory results.
2 2.1
Experimental Materials and reagents
A stock solution of luminol (10 mM) was prepared by dissolving luminol (Merck, Darmstadt, Germany) in sodium hydroxide solution (0.1 M) and stored at least 7 days before dilution. The buffer solutions of NaHCO3-Na2CO3 were prepared by mixing 0.1 M NaHCO3 and Na2CO3 aqueous solution. NaHCO3 and Na2CO3 were obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). The fresh H2O2 working solution was prepared daily from 30% (w/w) H2O2 (Suzhou Chemical Reagent Company, China). A stock solution (1%, w/w) of HAuCl4 was prepared by dissolving HAuCl4·4H2O (Shanghai Chemical Reagent Company, China) in water. The reference compounds of LD and CD were purchased from the Chinese National Institute for the Control of Pharmaceutical and Biological Products. Stock solutions of LD and CD were individually prepared in water containing H3PO4 (0.2%, V/V, Sinopharm Chemical Reagent Co., Ltd., Shanghai, China) at final concentration of 0.1 mg mL–1, respectively. All stock solutions and working solutions were stored in the dark at 4 ºC before use. Methanol was of HPLC grade, and all other chemicals were of analytical-reagent grade. Ultrapure water (18.3 MΩ cm,
MU Chun-Lei et al. / Chinese Journal of Analytical Chemistry, 2017, 45(6): e1726–e1733
Millipore, USA) was used throughout the experiment. The mobile phase of HPLC was prepared fresh daily and filtered through a 0.22-μm membrane filter (Bandao, Shanghai, China). Levodopa and carbidopa controlled-release tablets (Sinemet®), each with a nominal content of 50 mg CD and 200 mg LD, were obtained from Merck Sharp & Dohme (Australia) Pty. Ltd.
CL solutions were delivered by a peristaltic pump. The CL emission was monitored by PMT. The high potential of PMT was set as ‒600 V. The quantitative determination was based on the relative CL intensity ΔI = IS − I0, where IS and I0 (blank signal) were the CL intensities in the presence and absence of LD and CD. 2.3
2.2
Optical measurements
HPLC-CL detection
The HPLC-CL detection system was consisted of a HPLC system and a CL detection system, as shown in Fig.2. The HPLC system was Shimadzu LC-20A series (Shimadzu Corporation, Japan), including a quaternary pump, a vacuum degasser, a thermostated column compartment, a prominence diode array detector (DAD), a manual sample valve injector with a 20-μL loop, and an analytical column (Shim-pack ODS, 250 mm × 4.6 mm, 5 μm; Shimadzu, Japan). The CL detection was performed with a flow injection CL system (Remax, Xi’an, China), which was composed of a model IFFM-E peristaltic pump, a mixing tee, and a model IFFS-A CL detector equipped with a flat glass coil (used as reaction coil and detection cell) and a photomultiplier (PMT). The sample separation was performed on a ODS column at 40 ºC with gradient elution at a flow rate of 1.0 mL min–1. The mobile phase was composed of aqueous phosphoric acid (0.2%, V/V) and methanol (95:5, V/V). Detection wavelength of DAD was set at 275 nm. The column effluent from DAD was first mixed with HAuCl4 solution via a PEEK tube, then combined with luminol dissolved in NaHCO3-Na2CO3 buffer solution, and finally with H2O2 solution in a mixing tee, respectively. All
The UV-visible spectra were measured on a model 8453 spectrophotometer (Agilent, USA). The morphology of gold nanoparticles was examined on transmission electron microscope (TEM, JEOL JEM-2010, Japan) at 200 kV. 2.4 2.4.1
Preparation of sample solution Preparation of pharmaceutical sample
Six controlled-release tablets (Sinemet®, specification of 50 mg per tablet CD and 200 mg per tablet LD) were weighed up and ground to fine powder. Fine powder (7.19 mg, equivalence of about 5 mg of LD and 1.25 mg of CD) was accurately taken and dissolved in 0.1% phosphoric acid. The sample solution was subjected to ultrasonic treatment for 5 min and filtered through a 0.22-μm membrane filter prior to HPLC analysis. 2.4.2
Preparation of human serum sample
Human serum samples were donated by the First Affiliated Hospital of Anhui Medical University (Hefei, China). The C18
Fig.2 Schematic diagram of HPLC-CL system for simultaneous determination of levodopa and carbidopa
MU Chun-Lei et al. / Chinese Journal of Analytical Chemistry, 2017, 45(6): e1726–e1733
solid phase extraction (SPE) column was preconditioned with 6 mL of methanol and followed by 6 mL of water. Serum samples were centrifuged at 3000 rpm for 5 min at 4 ºC. 200 μL of supernatant in a 1.5-mL centrifuge tube was mixed with 200 μL of HClO4 (5%), shaken vigorously for 2 min, and then centrifuged at 10000 rpm for 15 min at 4 ºC. Finally, the supernatant was passed through SPE column slowly, washed by ultrapure water to remove soluble impurities, and then eluted by 3 mL of acetonitrile. The eluent was evaporated to dryness under a stream of nitrogen. The residue was dissolved in 400 μL of HPLC mobile phase for injection.
3 3.1
Results and discussion Mechanism of CL reaction
It was reported that HCO3− could reacted with H2O2 to form HCO4−, which is a strong reductant, and the reaction of HCO4− with AuCl4− resulted in the formation of (10 ± 2) nm AuNPs[29]. In our previous work, the reaction between luminol and H2O2 could be catalyzed by AuNPs which were produced by on-line reaction of H2O2, NaHCO3-Na2CO3 (buffer solution of luminol) and HAuCl4[28]. In the present CL system, when LD or CD solution was injected into the luminol-H2O2HAuCl4 mixture, the strongly enhanced CL signals were observed. To study the enhancement mechanism of LD and CD on the on-line AuNPs-catalyzed luminol-H2O2 CL, UV-visible absorption spectra were recorded, as shown in Fig.3. CD was chosen as a model compound. The absorption of CD at 370 nm decreased after the addition of H2O2, suggesting that CD was oxidized by H2O2 (curve b). When H2O2 was added into HAuCl4 in the NaHCO3-Na2CO3 buffer solution, a new absorption band at around 580 nm appeared (curve c), indicating the possible formation of AuNPs. Interestingly, when CD was added to the mixture of H2O2 and HAuCl4 in the NaHCO3-Na2CO3 buffer solution, the absorbance at 580 nm significantly increased (curve e), revealing that the presence of CD could promote the formation of AuNPs, which was further conformed by TEM observation. The inset in Fig.3 shows the TEM image of the shape and size of AuNPs with the average diameter of (46.8 ± 8.2) nm. Both LD and CD contain the orthodihydroxyphenyl (catechol) functional group, and their chemical structures are similar to those of catecholamines. Based on the above UV-visible absorption spectra and the reported CL mechanism of catecholamines[28], the possible enhancement mechanism of LD and CD on the on-line AuNPs-catalyzed luminol-H2O2 CL is proposed, as shown in Scheme 1. HCO3− was oxidized by H2O2 to form HCO4−, and HCO4− reacted with AuCl4− to produce a few AuNPs, which catalyzed the CL reaction between luminol and H2O2, leading to a weak CL signal. However, when LD or CD solution was added to the mixture of H2O2, luminol and HAuCl4 in the NaHCO3-Na2CO3 buffer
Fig.3 UV-visible spectra of AuNPs produced by the on-line reaction in NaHCO3-Na2CO3 buffer solution: (a) carbidopa; (b) H2O2 and carbidopa; (c) H2O2 and HAuCl4; (d) HAuCl4 and carbidopa; (e) H2O2, HAuCl4 and carbidopa. Inset shows the TEM image of 47-nm AuNPs produced by the on-line reaction. Conditions: carbidopa, 0.1 mg mL–1; H2O2, 0.1 M; HAuCl4, 0.1 mg mL–1; NaHCO3-Na2CO3 buffer, pH 10.28
Scheme 1 Possible enhancement mechanism of LD or CD on the on-line AuNPs-catalyzed luminol-H2O2 CL system
solution, a strong CL signal was observed. The enhanced CL can be explained as that LD or CD promotes the on-line formation of a large number of AuNPs, which will better catalyze the luminol-H2O2 CL reaction. 3.2
Optimization of HPLC conditions
As for the HPLC-CL detection, the mobile phase of HPLC should be not only suitable for the good separation of analytes but also should be compatible with CL reaction. Several kinds of HPLC mobile phase, such as methanol-formic acid, methanol-phosphoric acid and acetonitrile-phosphoric acid, were reported for the separation of LD and CD[16,20,21]. However, it was found that formic acid caused the quenching of CL signal, and methanol resulted in less baseline drift than acetonitrile. Finally, the mobile phase consisting of methanol and 0.2% aqueous phosphoric acid (5:95, V/V) was chosen for isocratic elution. The good HPLC separation was achieved within 10.5 min for LD and CD. 3.3 Optimization of CL conditions The effect of mixed order of CL reagents was investigated
MU Chun-Lei et al. / Chinese Journal of Analytical Chemistry, 2017, 45(6): e1726–e1733
at first. The stable and maximal signal-to-noise (S/N) ratios of CL signals for LD and CD were obtained when the effluent of HPLC column was first mixed with HAuCl4 solution, then with luminol in NaHCO3-Na2CO3 buffer solution, and finally with H2O2 solution, as shown in Fig.2. To obtain the optimal efficiency of CL system and stability of CL signal, the effects of pH value of NaHCO3-Na2CO3 buffer solution, concentrations of luminol, H2O2, and HAuCl4, and flow rate of CL reagents on the CL intensities of LD and CD were also investigated. The optimal concentration of LD and CD was found to be 200 ng mL–1, and the optimization results of CL reagents and conditions of on-line AuNPs-catalyzed luminol-H2O2 detector for HPLC were shown in Fig.4. The NaHCO3-Na2CO3 buffer solution is a vital factor affecting sensitivity of on-line AuNPs-catalyzed luminol CL system. The effect of buffer pH value used for preparing luminol solution was studied in the pH range of 9.40–11.83. The results in Fig.4A show that the optimum pH for both LD and CD is 10.28. The effect of luminol concentration on the S/N ratios of CL signals for LD and CD was also investigated in the range from 0.1 to 1.0 μM in the NaHCO3-Na2CO3 buffer solution (pH 10.28). It was found that the CL emission increased with the increase of luminol concentration until luminol concentration reached 0.2 μM and then decreased with further increasing luminol concentration. Therefore, 0.2 μM luminol was used in the following work (Fig.4B). The effect of concentration of H2O2 on the S/N ratio of CL
signal for LD and CD was examined over the range of 0.05–1.0 mM. The CL intensities increased with the increase of H2O2 concentration. However, when the H2O2 concentration was higher than 0.2 mM, the baseline noise increased dramatically, which caused the poor reproducibility of CL signal. Therefore, 0.2 mM H2O2 was chosen in the further studies (Fig.4C). The effect of HAuCl4 concentration in the range of 10–80 μg mL–1 on CL intensities was examined. The S/N ratio increased with increasing HAuCl4 concentration up to 40 μg mL–1, above which the S/N ratio decreased (Fig.4D). Thus, the HAuCl4 with concentration of 40 μg mL–1 was used for the further experiment research. The flow rate of CL solution is very important to the CL emission and should be carefully regulated. When flow rates were too slow or too high, no CL was emitted. Under the optimal CL conditions, the effect of flow rate was studied over the range of 0.5–3.5 mL min–1 in each stream. The S/N ratio for LD and CD increased with the increasing flow rate until it reached to 2.0 mL min–1. Therefore, a flow rate of 2.0 mL min–1 was selected to obtain a higher precision of CL signal and lower consumption of CL reagent. Under the optimized conditions above mentioned, the typical HPLC chromatograms of a mixture of LD and CD with DAD detection at 275 nm and CL detection were shown in Fig.5. The retention times of LD and CD were 4.9 and 10.4 min, respectively. By comparing the two kinds of chromatograms, it was concluded that the proposed on-line AuNPs-catalyzed
Fig.4 Effects of CL reagent conditions on the on-line AuNPs-catalyzed luminol-H2O2 detector for HPLC (A) Effect of pH value of NaHCO3-Na2CO3 buffer: luminol, 1.0 μM; H2O2: 0.3 mM; HAuCl4, 40 μg mL–1; (B) Effect of luminol concentration: NaHCO3-Na2CO3 buffer, pH 10.28; H2O2: 0.3 mM; HAuCl4, 40 μg mL–1; (C) Effect of H2O2 concentration: luminol, 0.2 μM dissolved in the NaHCO3-Na2CO3 buffer solution of pH 10.28; HAuCl4, 40 μg mL–1; (D) Effect of HAuCl4 concentration: luminol, 0.2 μM dissolved in the NaHCO3-Na2CO3 buffer solution of pH 10.28; H2O2: 0.2 mM. The concentration of LD and CD used for the optimization experiments was 200 ng mL–1
MU Chun-Lei et al. / Chinese Journal of Analytical Chemistry, 2017, 45(6): e1726–e1733
Fig.5
Chromatograms of a mixture of standard levodopa and carbidopa with (A) DAD at 275 nm and (B) on-line AuNPs-catalyzed luminal-H2O2 CL detection
Peaks: (1) levodopa; (2) carbidopa. Mobile phase of HPLC: methanol-0.2% aqueous phosphoric acid (5:95, V/V); Post-column CL reaction conditions: HAuCl4, 40 μg mL–1; luminol, 0.2 μM; NaHCO3-Na2CO3 buffer, pH 10.28; H2O2, 0.2 mM. The concentration of levodopa and carbidopa was 200 ng mL–1
2.24−448 ng mL–1 and 4.32−1080 ng mL–1, respectively. The detection limits of LD and CD (S/N = 3) were 0.89 and 1.08 ng mL–1, corresponding to 17.92 and 21.60 pg for 20 μL of sample injection, respectively. In comparison with the literature works (Table 2), the proposed HPLC-nanoCL method had wider linear ranges and lower detection limits. Therefore, this method is an alternative, sensitive and simple approach for simultaneous detection of LD and CD.
luminal-H2O2 detector for HPLC could indeed be employed for sensitive detection of LD and CD. 3.4 3.4.1
Method validation Linearity and detection limit
Under the optimal experimental conditions, the regression equations, linear ranges and detection limits of LD and CD were investigated and the results are summarized in Table 1. The calibration curves of LD and CD standards showed good linear relationship (r > 0.996, n = 10) between the logarithm of LD and CD concentration (C) and the logarithm of CL intensity (peak height, ΔI) in the concentration ranges of
3.4.2
Precision and recovery
The results of intra- and inter-day assay indicated that the present HPLC-nanoCL method had good accuracies and reproducibilities for LD and CD detection at three concentration
Table 1 Regression equations, linear ranges and detection limits of levodopa and carbidopa Regression equation
Linear range (ng mL–1)
Correlation coefficient (r)
Detection limit (ng mL–1)
Levodopa
lgΔI = 1.013lgC + 10.650
2.24−448
0.9960
0.89
Carbidopa
lgΔI = 1.093lgC + 10.818
4.32−1080
0.9964
1.08
Analyte
1
Table 2 Comparison of methods for detection of levodopa and carbidopa Analyte
Method
Linear range
Detection limit
Reference
Levodopa
HPLC-CLa CE-CLb Nanoelectrochemicalc Nanoelectrochemicalc
10−1000 ng mL–1 50−2500 nM 0.25−200 μM 0.10−900.0 μM
3 ng mL–1 20 nM 90 nM 41 nM
[22] [10] [30] [31]
Nanoelectrochemicalc HPLC-FLd HPLC-MS/MS HILIC-MS/MSe HPLC-nanoCLf
0.1−70 μM 5–500 ng mL–1 50–5000 ng mL–1 75–800 ng mL–1 2.24−448 ng mL–1
35 nM 0.3 ng mL–1 − 30 ng mL–1 0.89 ng mL–1
[32] [16] [19] [17] This work
Nanoelectrochemicac HPLC-FLd HPLC-MS/MS HILIC-MS/MSe HPLC-nanoCLf
20.0−900.0 μM 5–500 ng mL–1 3–600 ng mL–1 65–800 ng mL–1 4.32−1080 ng mL–1
0.38 μM 0.6 ng mL–1 − 25 ng mL–1 1.08 ng mL–1
[31] [16] [19] [17] This work
Carbidopa
a
HPLC-CL: luminal-potassium ferricyanide CL detector for HPLC; bCE-CL: luminal-potassium ferricyanide CL detector for capillary electrophoresis; cCarbon paste electrode modified with carbon nanotubes[30], TiO2 nanoparticles[31], and ZnO nanorods[32] as the nanostructured electrochemical sensors; dHPLC-FL: fluorescence detector for HPLC; eHILIC-MS/MS: hydrophilic interaction liquid chromatography-tandem mass spectrometry; fHPLC-nanoCL: on-line gold nanoparticle-catalyzed luminal-H2O2 CL detector for HPLC.
MU Chun-Lei et al. / Chinese Journal of Analytical Chemistry, 2017, 45(6): e1726–e1733
levels (11.2, 89.6 and 336.0 ng mL–1 for LD and 10.8, 86.4 and 864.0 ng mL–1 for CD). The intra-day assay was tested with 7 repeated injections of LD and CD solution, while the inter-day assay was studied by analyzing LD and CD solution injected 3 times every day, on seven consecutive days. The accuracy (relative error, RE) was calculated from the nominal concentration (Cnom) and the mean value of observed concentration (Cobs) as follows: RE (%) = [(Cobs − Cnom)/(Cnom)] × 100 The precision (RSD, %) was calculated from the observed concentrations as follows: RSD (%) = [SD/Cobs] × 100. SD is standard deviation. The intra-day and the inter-day precisions were no more than 6.7%. The intra-day accuracy varied from −2.4% to 4.0%, while the inter-day accuracy ranged from −2.9% to 2.7%. 3.5 Practical applications 3.5.1
Determination of levodopa and carbidopa in pharmaceutical formulation
The proposed HPLC–nanoCL method was applied to direct determination of LD and CD in the controlled-release tablets (Sinemet®, each tablet contains 200 mg of LD and 50 mg of CD). The average recoveries of LD and CD tablets are 100.5% and 103.1% with the precisions of 2.4% and 4.0%, respectively. The results for determination of LD and CD tablets were shown in Table 3, indicating that the values obtained by the proposed method were in good agreement with labels of pharmaceutical preparations. To further validate
analytical results, HPLC method with DAD at 275 nm was employed to measure the content of LD and CD in tablets. A good agreement was achieved, and there was no significant difference between the results of two methods. Therefore, the proposed HPLC-nanoCL method could be applied for the quality control of pharmaceutical formulations contained LD and CD. 3.5.2
Determination of levodopa and carbidopa in human serum samples
To demonstrate the applicability of HPLC-nanoCL method in complex biological sample, the proposed method was used to analyze LD and CD in human serum samples. As shown in Fig.6, no peak was observed in the chromatogram of blank plasma (Fig.6A) and two peaks were found in the chromatogram of serum spiked with LD and CD (Fig.6B), indicating that none of endogenous compounds in human serum would interfere with the determination of LD and CD. The high selectivity for detection of LD and CD in human serum was obtained by a sample clean-up procedure performed on C18 SPE column. The recoveries of LD and CD in human serum by SPE purification at three spiked concentration levels (10, 40 and 160 ng mL–1 for LD; 20, 80 and 320 ng mL–1 for CD) were 61.4% and 63.5%, respectively. The analytical results of three human serum samples were summarized in Table 4. Good recoveries between 97.0% and 103.5% with RSDs of no more than 3.3% indicated that the present method could meet the requirements for simultaneous and sensitive quantification of LD and CD in human serum samples.
Table 3 Comparison of determination results of LD and CD in controlled-release tablets (Sinemet®) by two methods (n = 5) Batch No. J013155 J005774 J011060
This method
HPLC−DAD
Analyte
Declared (mg per tablet)
Found (μg)
RSD (%)
Found (μg)
RSD (%)
LD CD LD CD LD CD
200 50 200 50 200 50
207.2 50.6 196.8 48.6 204.6 52.3
2.4 4.0 3.1 2.5 2.0 4.0
198.8 51.0 203 49.0 202.4 51.8
1.9 1.2 1.8 1.6 1.0 1.5
Fig.6 Chromatograms of (A) blank human serum sample and (B) human serum spiked with 100 ng mL–1 levodopa and carbidopa. Peaks: (1) levodopa; (2) carbidopa. Chromatographic and CL conditions are the same as those of Fig.5
MU Chun-Lei et al. / Chinese Journal of Analytical Chemistry, 2017, 45(6): e1726–e1733
Table 4 Results for determination of LD and CD in human serum by using the proposed HPLC-nanoCL method (n = 3)
4
Human serum
Spiked (ng mL–1)
Found (ng mL–1)
LD
CD
LD
CD
LD
CD
LD
CD
1 2 3 4
0 8.0 40.0 200.0
0 10.0 50.0 250.0
ND 7.9 39 207
ND 9.7 51.1 252.3
− 98.7 97.5 103.5
− 97.0 102.2 100.9
− 3.3 1.6 1.1
− 2.5 2.0 0.5
Conclusions
Recovery (%)
RSD (%)
1998, 46(1): 39−44 [14] Pistonesi M, Centurión M E, Band B S F, Damiani P C,
In this work, based on the enhancing effect of LD and CD on the on-line AuNPs-catalyzed luminal-H2O2 CL reaction, a new HPLC-nanoCL method was developed for simultaneous and selective determination of LD and CD. The high sensitivity and reproducibility of on-line AuNPs-catalyzed CL responses, together with the ease of sample pretreatment, make the proposed method very useful for quantification of LD and CD in pharmaceutical formulations and human serum. It is believed that the presented HPLC-nanoCL method had promising potential in monitoring LD and CD at trace levels in serum and urine samples in Parkinson’s patients.
Olivieri A C. J. Pharm. Biomed. Anal., 2004, 36(3): 541–547 [15] Teixeira M F, Marcolino-Junior L H, Fatibello-Filho O, Dockal E R, Bergamini M F. Sens. Actuators B, 2007, 122(2): 549−555 [16] Raut P P, Charde S Y. Luminescence, 2014, 29(7): 762−771 [17] Vilhena R D O, Pontes F L D, Marson B M, Ribeiro R P, de Carvalho K A T, Cardoso M A, Pontarolo R. J. Chromatogr. B, 2014, 967(15): 41−49 [18] Junnotula V, Licea-Perez H. J. Chromatogr. B, 2013, 926(1): 47−53 [19] César I C, Byrro R M D, Mundim I M, Teixeira L S, Gomes S A, Bonfim R R, Pianetti G A. J. Mass Spectrom., 2011, 46(9): 943−948
References
[20] Sagar K A, Smyth M R. J. Pharm. Biomed. Anal., 2000, 22(3): 613−624
[1]
Dauer W, Przedborski S. Neuron, 2003, 39(6): 889−909
[2]
Hu G, Chen L, Guo Y, Wang X, Shao S. Electrochim. Acta, 2010, 55(16): 4711−4716
[3]
Cotzias G C, van Woert M H, Schiffer L M. New Engl. J. Med., 1967, 276(7): 374−379
[4]
Okereke C S. J. Pharm. Pharm. Sci., 2002, 5(2): 146−161
[5]
Reynolds J E F (ed), Martindale: The Extra Pharmacopoeia, The Pharmaceutical Press, London, 1993
[6]
Gomes P, Soares-da-Silva P. Neuropharmacology, 1999, 38(9): 1371−1380
[7]
Le Witt P A, Nelson M V, Berchou R C, Galloway M P, Kesaree N, Kareti D, Schlick P. Neurology, 1989, 39(11 Suppl 2): 45−53
[8]
Hammerstad J P, Woodward W R, Nutt J G, Gancher S T, Block G A, Cyhan G. Clin. Neuropharmacol., 1994, 17(5): 429−434
[9]
Titus D C, August T F, Yeh K C, Eisenhandler R, Bayne W F, Musson D G. J. Chromatogr., 1990, 534: 87−100
[10] Zhao S, Bai W, Wang B, He M. Talanta, 2007, 73(1): 142−146 [11] Marques K L, Santos J L, Lopes J A, Lima J L. Anal. Sci., 2008, 24(8): 985–991 [12] Nagaraja P, Vasantha R A, Sunitha K R. J. Pharm. Biomed. Anal., 2001, 25(3): 417–424 [13] Nagaraja P, Murthy K C, Rangappa K S, Gowda N M. Talanta,
[21] Karimi M, Carl J L, Loftin S, Perlmutter J S. J. Chromatogr. B, 2006, 836(1-2): 120−123 [22] Chen F N, Zhang Z J, Zhang Y X, He D Y. Anal. Bioanal. Chem., 2005, 382(1): 211−215 [23] Chen J, Guo Y Y, Chen Q, Qiu H D. Chin. J. Anal. Chen., 2015, 43(10): 1513−1519 [24] Chen Z Q, Chen H, Xie X, Yang M. Chinese J. Anal. Chem., 2015, 43(7): 1075−1079 [25] Zhang L, Niu W X, Gao W Y, Qi L M, Lai J P, Zhao J M, Xu G B. ACS Nano, 2014, 8(6): 5953−5958 [26] Zhang Z F, Cui H, Lai C Z, Liu L J. Anal. Chem., 2005, 77(10): 3324−3329 [27] Zhang Q L, Wu L, Lv C, Zhang X Y. J. Chromatogr. A, 2012, 1242(15): 84−91 [28] Mu C L, Zhang Q, Wu D, Zhang Y J, Zhang Q L. Biomed. Chromatogr., 2015, 29(1): 148−155 [29] Lu C, Li Y, Yang J M. Talanta, 2010, 82(4): 1676−1580 [30] Beitollahi H, Mostafavi M. Electroanal., 2014, 26(5): 1090−1098 [31] Mazloum-Ardakani M, Khoshroo A. Anal. Chim. Acta, 2013, 798(10): 25−32 [32] Molaakbari E, Mostafavi A, Beitollahi H, Alizadeh R. Analyst, 2014, 139(17): 4356−4364
Cite this article as: Chin J Anal Chem, 2017, 45(6), e1726–e1733.
RESEARCH PAPER
Simultaneous and Sensitive Determination of Levodopa and Carbidopa in Pharmaceutical Formulation and Human Serum by High Performance Liquid Chromatography with On-Line Gold NanoparticlesCatalyzed Luminol Chemiluminescence Detection MU Chun-Lei, WU Dong, LU Hai-Feng, XIE He, ZHANG Qun-Lin* School of Pharmacy, Anhui Medical University, Hefei 230032, China
Abstract: Levodopa, the metabolic precursor of dopamine, is usually administrated in combination with carbidopa to control dopamine levels in an appropriate manner and reduce side effects in the treatment of Parkinson's disease. In this study, a selective and sensitive high performance liquid chromatography coupled with on-line gold nanoparticles-catalyzed luminol chemiluminescence method for simultaneous determination of levodopa and carbidopa was developed. This method was based on the strongly enhanced chemiluminescence signal of on-line gold nanoparticles-catalyzed luminol-H2O2 system by levodopa and carbidopa. The possible enhancement mechanism was attributed to that levodopa and carbidopa could promote the on-line formation of a large number of gold nanoparticles, which catalyzed the luminol-H2O2 chemiluminescence reaction. The good separation of levodopa and carbidopa was achieved with isocratic elution using a mixture of methanol and 0.2% aqueous phosphoric acid (5:95, V/V) within 10.5 min. Under the optimal conditions, the linear ranges of levodopa and carbidopa were 2.24–448 ng mL–1 and 4.32–1080 ng mL–1 with the detection limits of 0.89 and 1.08 ng mL–1 (S/N = 3), corresponding to 17.92 and 21.60 pg for 20 μL sample injection, respectively. The validated method was successfully applied to simultaneous quantification of levodopa and carbidopa in controlled-release tablets (Sinemet®) and human plasma. The average recoveries of levodopa and carbidopa in the tablets were 100.5% and 103.1% with the precisions (RSDs) of 2.4% and 4.0%. The recoveries of levodopa and carbidopa in human plasma ranged from 97.0% to 103.5% with RSDs of no more than 3.3%. Key Words:
1
Levodopa; Carbidopa; Chemiluminescence; On-line; Gold nanoparticles; Luminol
Introduction
Parkinson’s disease is a neurodegenerative disorder of the central nervous system. The loss of neurons in the substantia nigra, a region of the midbrain, leads to a deficiency of dopamine neurotransmitter in the brain, which may lead to Parkinson’s disease[1]. Dopamine is not effectively used for the treatment of Parkinson’s disease, because it could not
penetrate the blood-brain barrier (BBB)[2]. Levodopa (LD, Fig.1A), the metabolic precursor of dopamine, has been
Fig.1 Structures of (A) levodopa and (B) carbidopa
________________________ Received 8 April 2016; accepted 13 April 2017 *Corresponding author. Email: [email protected] This work was supported by the National Natural Science Foundation of China (No. 30973674), and the Science and Technological Fund of Anhui Province for Outstanding Youth of China (No. 1308085JGD10). Copyright © 2017, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences. Published by Elsevier Limited. All rights reserved. DOI: 10.1016/S1872-2040(17)61021-1
MU Chun-Lei et al. / Chinese Journal of Analytical Chemistry, 2017, 45(6): e1726–e1733
regarded as the standard for treating Parkinson’s disease, which is a condition precipitated by dopamine depletion in the central nervous system[3]. Unlike dopamine, LD can cross the BBB via a saturable transporter and is converted to dopamine by l-aromatic amino acid decarboxylase in the brain[4]. For the better therapeutic effect and lower toxicity, carbidopa (CD, Fig.1B) is administered in association with LD in the pharmaceutical formulations containing 10%‒25% of CD. CD is a competitive peripheral l-aromatic amino acid decarboxylase inhibitor with little or no pharmacological activity when given alone in usual dosage to Parkinson’s disease patients[5]. CD does not cross the BBB and contributes to the production of effective brain concentrations of dopamine from lower doses of LD by inhibiting the peripheral decarboxylation of LD to dopamine. In addition, the reduced peripheral formation of dopamine decreases the peripheral side effects such as nausea, vomiting and cardiac arrhythmia[6]. To release the fluctuating symptoms in clinical, the most commonly used method is to administrate with controlled-release LD/CD tablet (Sinemet®, specifications 200/50 mg), which is developed to prolong the therapeutic plasma level of LD[7,8]. Therefore, it is very important to establish a selective and sensitive method for simultaneous detection of LD and CD in pharmaceutical formulations and human serum, because of the coexistence of LD and CD in pharmaceutical formulations and the rapid fluctuations of plasma drug concentration of LD. However, the accurate and sensitive analysis of LD and CD is challenging due to their poor stability, low molecular mass, and high polarity[9]. Many analytical methods including chemiluminescence (CL)[10,11], spectrophotometry[12‒14], amperometric and voltammetric determination[2,15] have been reported for determination of LD or CD alone in various pharmaceutical preparations and biological samples. Simultaneous analysis of LD and CD was commonly achieved by high performance liquid chromatography (HPLC), because of the merits of high resolution and short analysis time. Fluorescent[16], mass spectrometric[17‒19] and electrochemical[20,21] detection have been used to couple with HPLC method for sensitive quantification of LD and CD. Among them, electrochemical detection method had the limitation of bad reproducibility, and mass spectrometry had the disadvantage of high instrument cost. The combination of HPLC method with CL detector is a high-efficiency technique and has many advantages such as high sensitivity, rapidity and superb reproducibility. To our knowledge, there is only one published literature that reported the luminol-potassium ferricyanide CL system coupled with HPLC method for determination of LD[22]. Luminol-H2O2 CL reaction, a popular CL system, has been widely applied to the detection of various substances[23‒25]. Gold nanoparticles (AuNPs) have been widely applied in CL reactions as
catalysts because of their excellent catalytic activities and facile synthesis. Zhang et al[26] first applied AuNPs to enhance the CL of luminol-H2O2 system, and the enhancement was supposed to originate from the catalysis of AuNPs, which facilitated the radical generation and electron-transfer processes taking place on the surface of AuNPs. In our previous work, a novel on-line AuNPs-catalyzed luminol-H2O2 CL detector for HPLC method was established to simultaneously determine eight phenolic compounds in red wine and catecholamines (norepinephrine, epinephrine and dopamine) at trace levels in rat brain[27,28]. In this work, it was found that LD and CD could strongly enhance the CL intensity of on-line AuNPs-catalyzed luminal-H2O2 reaction. This phenomenon allowed us to couple the highly sensitive on-line AuNPs-catalyzed luminol-H2O2 CL detection with HPLC (HPLC-nanoCL) for simultaneous quantification of LD and CD for the first time. This method exhibited the advantages such as low cost, short time and simple sample handling process. The experimental conditions for the good HPLC separation and maximal and stable CL intensities of LD and CD were systematically optimized. The possible mechanism of on-line AuNPs-catalyzed luminol-H2O2 CL enhanced by LD and CD was studied by UV-visible absorption sepctroscopy and transmission electron microscopy. The proposed HPLC-nanoCL method was applied to detect LD and CD in human serum and pharmaceutical formulation with satisfactory results.
2 2.1
Experimental Materials and reagents
A stock solution of luminol (10 mM) was prepared by dissolving luminol (Merck, Darmstadt, Germany) in sodium hydroxide solution (0.1 M) and stored at least 7 days before dilution. The buffer solutions of NaHCO3-Na2CO3 were prepared by mixing 0.1 M NaHCO3 and Na2CO3 aqueous solution. NaHCO3 and Na2CO3 were obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). The fresh H2O2 working solution was prepared daily from 30% (w/w) H2O2 (Suzhou Chemical Reagent Company, China). A stock solution (1%, w/w) of HAuCl4 was prepared by dissolving HAuCl4·4H2O (Shanghai Chemical Reagent Company, China) in water. The reference compounds of LD and CD were purchased from the Chinese National Institute for the Control of Pharmaceutical and Biological Products. Stock solutions of LD and CD were individually prepared in water containing H3PO4 (0.2%, V/V, Sinopharm Chemical Reagent Co., Ltd., Shanghai, China) at final concentration of 0.1 mg mL–1, respectively. All stock solutions and working solutions were stored in the dark at 4 ºC before use. Methanol was of HPLC grade, and all other chemicals were of analytical-reagent grade. Ultrapure water (18.3 MΩ cm,
MU Chun-Lei et al. / Chinese Journal of Analytical Chemistry, 2017, 45(6): e1726–e1733
Millipore, USA) was used throughout the experiment. The mobile phase of HPLC was prepared fresh daily and filtered through a 0.22-μm membrane filter (Bandao, Shanghai, China). Levodopa and carbidopa controlled-release tablets (Sinemet®), each with a nominal content of 50 mg CD and 200 mg LD, were obtained from Merck Sharp & Dohme (Australia) Pty. Ltd.
CL solutions were delivered by a peristaltic pump. The CL emission was monitored by PMT. The high potential of PMT was set as ‒600 V. The quantitative determination was based on the relative CL intensity ΔI = IS − I0, where IS and I0 (blank signal) were the CL intensities in the presence and absence of LD and CD. 2.3
2.2
Optical measurements
HPLC-CL detection
The HPLC-CL detection system was consisted of a HPLC system and a CL detection system, as shown in Fig.2. The HPLC system was Shimadzu LC-20A series (Shimadzu Corporation, Japan), including a quaternary pump, a vacuum degasser, a thermostated column compartment, a prominence diode array detector (DAD), a manual sample valve injector with a 20-μL loop, and an analytical column (Shim-pack ODS, 250 mm × 4.6 mm, 5 μm; Shimadzu, Japan). The CL detection was performed with a flow injection CL system (Remax, Xi’an, China), which was composed of a model IFFM-E peristaltic pump, a mixing tee, and a model IFFS-A CL detector equipped with a flat glass coil (used as reaction coil and detection cell) and a photomultiplier (PMT). The sample separation was performed on a ODS column at 40 ºC with gradient elution at a flow rate of 1.0 mL min–1. The mobile phase was composed of aqueous phosphoric acid (0.2%, V/V) and methanol (95:5, V/V). Detection wavelength of DAD was set at 275 nm. The column effluent from DAD was first mixed with HAuCl4 solution via a PEEK tube, then combined with luminol dissolved in NaHCO3-Na2CO3 buffer solution, and finally with H2O2 solution in a mixing tee, respectively. All
The UV-visible spectra were measured on a model 8453 spectrophotometer (Agilent, USA). The morphology of gold nanoparticles was examined on transmission electron microscope (TEM, JEOL JEM-2010, Japan) at 200 kV. 2.4 2.4.1
Preparation of sample solution Preparation of pharmaceutical sample
Six controlled-release tablets (Sinemet®, specification of 50 mg per tablet CD and 200 mg per tablet LD) were weighed up and ground to fine powder. Fine powder (7.19 mg, equivalence of about 5 mg of LD and 1.25 mg of CD) was accurately taken and dissolved in 0.1% phosphoric acid. The sample solution was subjected to ultrasonic treatment for 5 min and filtered through a 0.22-μm membrane filter prior to HPLC analysis. 2.4.2
Preparation of human serum sample
Human serum samples were donated by the First Affiliated Hospital of Anhui Medical University (Hefei, China). The C18
Fig.2 Schematic diagram of HPLC-CL system for simultaneous determination of levodopa and carbidopa
MU Chun-Lei et al. / Chinese Journal of Analytical Chemistry, 2017, 45(6): e1726–e1733
solid phase extraction (SPE) column was preconditioned with 6 mL of methanol and followed by 6 mL of water. Serum samples were centrifuged at 3000 rpm for 5 min at 4 ºC. 200 μL of supernatant in a 1.5-mL centrifuge tube was mixed with 200 μL of HClO4 (5%), shaken vigorously for 2 min, and then centrifuged at 10000 rpm for 15 min at 4 ºC. Finally, the supernatant was passed through SPE column slowly, washed by ultrapure water to remove soluble impurities, and then eluted by 3 mL of acetonitrile. The eluent was evaporated to dryness under a stream of nitrogen. The residue was dissolved in 400 μL of HPLC mobile phase for injection.
3 3.1
Results and discussion Mechanism of CL reaction
It was reported that HCO3− could reacted with H2O2 to form HCO4−, which is a strong reductant, and the reaction of HCO4− with AuCl4− resulted in the formation of (10 ± 2) nm AuNPs[29]. In our previous work, the reaction between luminol and H2O2 could be catalyzed by AuNPs which were produced by on-line reaction of H2O2, NaHCO3-Na2CO3 (buffer solution of luminol) and HAuCl4[28]. In the present CL system, when LD or CD solution was injected into the luminol-H2O2HAuCl4 mixture, the strongly enhanced CL signals were observed. To study the enhancement mechanism of LD and CD on the on-line AuNPs-catalyzed luminol-H2O2 CL, UV-visible absorption spectra were recorded, as shown in Fig.3. CD was chosen as a model compound. The absorption of CD at 370 nm decreased after the addition of H2O2, suggesting that CD was oxidized by H2O2 (curve b). When H2O2 was added into HAuCl4 in the NaHCO3-Na2CO3 buffer solution, a new absorption band at around 580 nm appeared (curve c), indicating the possible formation of AuNPs. Interestingly, when CD was added to the mixture of H2O2 and HAuCl4 in the NaHCO3-Na2CO3 buffer solution, the absorbance at 580 nm significantly increased (curve e), revealing that the presence of CD could promote the formation of AuNPs, which was further conformed by TEM observation. The inset in Fig.3 shows the TEM image of the shape and size of AuNPs with the average diameter of (46.8 ± 8.2) nm. Both LD and CD contain the orthodihydroxyphenyl (catechol) functional group, and their chemical structures are similar to those of catecholamines. Based on the above UV-visible absorption spectra and the reported CL mechanism of catecholamines[28], the possible enhancement mechanism of LD and CD on the on-line AuNPs-catalyzed luminol-H2O2 CL is proposed, as shown in Scheme 1. HCO3− was oxidized by H2O2 to form HCO4−, and HCO4− reacted with AuCl4− to produce a few AuNPs, which catalyzed the CL reaction between luminol and H2O2, leading to a weak CL signal. However, when LD or CD solution was added to the mixture of H2O2, luminol and HAuCl4 in the NaHCO3-Na2CO3 buffer
Fig.3 UV-visible spectra of AuNPs produced by the on-line reaction in NaHCO3-Na2CO3 buffer solution: (a) carbidopa; (b) H2O2 and carbidopa; (c) H2O2 and HAuCl4; (d) HAuCl4 and carbidopa; (e) H2O2, HAuCl4 and carbidopa. Inset shows the TEM image of 47-nm AuNPs produced by the on-line reaction. Conditions: carbidopa, 0.1 mg mL–1; H2O2, 0.1 M; HAuCl4, 0.1 mg mL–1; NaHCO3-Na2CO3 buffer, pH 10.28
Scheme 1 Possible enhancement mechanism of LD or CD on the on-line AuNPs-catalyzed luminol-H2O2 CL system
solution, a strong CL signal was observed. The enhanced CL can be explained as that LD or CD promotes the on-line formation of a large number of AuNPs, which will better catalyze the luminol-H2O2 CL reaction. 3.2
Optimization of HPLC conditions
As for the HPLC-CL detection, the mobile phase of HPLC should be not only suitable for the good separation of analytes but also should be compatible with CL reaction. Several kinds of HPLC mobile phase, such as methanol-formic acid, methanol-phosphoric acid and acetonitrile-phosphoric acid, were reported for the separation of LD and CD[16,20,21]. However, it was found that formic acid caused the quenching of CL signal, and methanol resulted in less baseline drift than acetonitrile. Finally, the mobile phase consisting of methanol and 0.2% aqueous phosphoric acid (5:95, V/V) was chosen for isocratic elution. The good HPLC separation was achieved within 10.5 min for LD and CD. 3.3 Optimization of CL conditions The effect of mixed order of CL reagents was investigated
MU Chun-Lei et al. / Chinese Journal of Analytical Chemistry, 2017, 45(6): e1726–e1733
at first. The stable and maximal signal-to-noise (S/N) ratios of CL signals for LD and CD were obtained when the effluent of HPLC column was first mixed with HAuCl4 solution, then with luminol in NaHCO3-Na2CO3 buffer solution, and finally with H2O2 solution, as shown in Fig.2. To obtain the optimal efficiency of CL system and stability of CL signal, the effects of pH value of NaHCO3-Na2CO3 buffer solution, concentrations of luminol, H2O2, and HAuCl4, and flow rate of CL reagents on the CL intensities of LD and CD were also investigated. The optimal concentration of LD and CD was found to be 200 ng mL–1, and the optimization results of CL reagents and conditions of on-line AuNPs-catalyzed luminol-H2O2 detector for HPLC were shown in Fig.4. The NaHCO3-Na2CO3 buffer solution is a vital factor affecting sensitivity of on-line AuNPs-catalyzed luminol CL system. The effect of buffer pH value used for preparing luminol solution was studied in the pH range of 9.40–11.83. The results in Fig.4A show that the optimum pH for both LD and CD is 10.28. The effect of luminol concentration on the S/N ratios of CL signals for LD and CD was also investigated in the range from 0.1 to 1.0 μM in the NaHCO3-Na2CO3 buffer solution (pH 10.28). It was found that the CL emission increased with the increase of luminol concentration until luminol concentration reached 0.2 μM and then decreased with further increasing luminol concentration. Therefore, 0.2 μM luminol was used in the following work (Fig.4B). The effect of concentration of H2O2 on the S/N ratio of CL
signal for LD and CD was examined over the range of 0.05–1.0 mM. The CL intensities increased with the increase of H2O2 concentration. However, when the H2O2 concentration was higher than 0.2 mM, the baseline noise increased dramatically, which caused the poor reproducibility of CL signal. Therefore, 0.2 mM H2O2 was chosen in the further studies (Fig.4C). The effect of HAuCl4 concentration in the range of 10–80 μg mL–1 on CL intensities was examined. The S/N ratio increased with increasing HAuCl4 concentration up to 40 μg mL–1, above which the S/N ratio decreased (Fig.4D). Thus, the HAuCl4 with concentration of 40 μg mL–1 was used for the further experiment research. The flow rate of CL solution is very important to the CL emission and should be carefully regulated. When flow rates were too slow or too high, no CL was emitted. Under the optimal CL conditions, the effect of flow rate was studied over the range of 0.5–3.5 mL min–1 in each stream. The S/N ratio for LD and CD increased with the increasing flow rate until it reached to 2.0 mL min–1. Therefore, a flow rate of 2.0 mL min–1 was selected to obtain a higher precision of CL signal and lower consumption of CL reagent. Under the optimized conditions above mentioned, the typical HPLC chromatograms of a mixture of LD and CD with DAD detection at 275 nm and CL detection were shown in Fig.5. The retention times of LD and CD were 4.9 and 10.4 min, respectively. By comparing the two kinds of chromatograms, it was concluded that the proposed on-line AuNPs-catalyzed
Fig.4 Effects of CL reagent conditions on the on-line AuNPs-catalyzed luminol-H2O2 detector for HPLC (A) Effect of pH value of NaHCO3-Na2CO3 buffer: luminol, 1.0 μM; H2O2: 0.3 mM; HAuCl4, 40 μg mL–1; (B) Effect of luminol concentration: NaHCO3-Na2CO3 buffer, pH 10.28; H2O2: 0.3 mM; HAuCl4, 40 μg mL–1; (C) Effect of H2O2 concentration: luminol, 0.2 μM dissolved in the NaHCO3-Na2CO3 buffer solution of pH 10.28; HAuCl4, 40 μg mL–1; (D) Effect of HAuCl4 concentration: luminol, 0.2 μM dissolved in the NaHCO3-Na2CO3 buffer solution of pH 10.28; H2O2: 0.2 mM. The concentration of LD and CD used for the optimization experiments was 200 ng mL–1
MU Chun-Lei et al. / Chinese Journal of Analytical Chemistry, 2017, 45(6): e1726–e1733
Fig.5
Chromatograms of a mixture of standard levodopa and carbidopa with (A) DAD at 275 nm and (B) on-line AuNPs-catalyzed luminal-H2O2 CL detection
Peaks: (1) levodopa; (2) carbidopa. Mobile phase of HPLC: methanol-0.2% aqueous phosphoric acid (5:95, V/V); Post-column CL reaction conditions: HAuCl4, 40 μg mL–1; luminol, 0.2 μM; NaHCO3-Na2CO3 buffer, pH 10.28; H2O2, 0.2 mM. The concentration of levodopa and carbidopa was 200 ng mL–1
2.24−448 ng mL–1 and 4.32−1080 ng mL–1, respectively. The detection limits of LD and CD (S/N = 3) were 0.89 and 1.08 ng mL–1, corresponding to 17.92 and 21.60 pg for 20 μL of sample injection, respectively. In comparison with the literature works (Table 2), the proposed HPLC-nanoCL method had wider linear ranges and lower detection limits. Therefore, this method is an alternative, sensitive and simple approach for simultaneous detection of LD and CD.
luminal-H2O2 detector for HPLC could indeed be employed for sensitive detection of LD and CD. 3.4 3.4.1
Method validation Linearity and detection limit
Under the optimal experimental conditions, the regression equations, linear ranges and detection limits of LD and CD were investigated and the results are summarized in Table 1. The calibration curves of LD and CD standards showed good linear relationship (r > 0.996, n = 10) between the logarithm of LD and CD concentration (C) and the logarithm of CL intensity (peak height, ΔI) in the concentration ranges of
3.4.2
Precision and recovery
The results of intra- and inter-day assay indicated that the present HPLC-nanoCL method had good accuracies and reproducibilities for LD and CD detection at three concentration
Table 1 Regression equations, linear ranges and detection limits of levodopa and carbidopa Regression equation
Linear range (ng mL–1)
Correlation coefficient (r)
Detection limit (ng mL–1)
Levodopa
lgΔI = 1.013lgC + 10.650
2.24−448
0.9960
0.89
Carbidopa
lgΔI = 1.093lgC + 10.818
4.32−1080
0.9964
1.08
Analyte
1
Table 2 Comparison of methods for detection of levodopa and carbidopa Analyte
Method
Linear range
Detection limit
Reference
Levodopa
HPLC-CLa CE-CLb Nanoelectrochemicalc Nanoelectrochemicalc
10−1000 ng mL–1 50−2500 nM 0.25−200 μM 0.10−900.0 μM
3 ng mL–1 20 nM 90 nM 41 nM
[22] [10] [30] [31]
Nanoelectrochemicalc HPLC-FLd HPLC-MS/MS HILIC-MS/MSe HPLC-nanoCLf
0.1−70 μM 5–500 ng mL–1 50–5000 ng mL–1 75–800 ng mL–1 2.24−448 ng mL–1
35 nM 0.3 ng mL–1 − 30 ng mL–1 0.89 ng mL–1
[32] [16] [19] [17] This work
Nanoelectrochemicac HPLC-FLd HPLC-MS/MS HILIC-MS/MSe HPLC-nanoCLf
20.0−900.0 μM 5–500 ng mL–1 3–600 ng mL–1 65–800 ng mL–1 4.32−1080 ng mL–1
0.38 μM 0.6 ng mL–1 − 25 ng mL–1 1.08 ng mL–1
[31] [16] [19] [17] This work
Carbidopa
a
HPLC-CL: luminal-potassium ferricyanide CL detector for HPLC; bCE-CL: luminal-potassium ferricyanide CL detector for capillary electrophoresis; cCarbon paste electrode modified with carbon nanotubes[30], TiO2 nanoparticles[31], and ZnO nanorods[32] as the nanostructured electrochemical sensors; dHPLC-FL: fluorescence detector for HPLC; eHILIC-MS/MS: hydrophilic interaction liquid chromatography-tandem mass spectrometry; fHPLC-nanoCL: on-line gold nanoparticle-catalyzed luminal-H2O2 CL detector for HPLC.
MU Chun-Lei et al. / Chinese Journal of Analytical Chemistry, 2017, 45(6): e1726–e1733
levels (11.2, 89.6 and 336.0 ng mL–1 for LD and 10.8, 86.4 and 864.0 ng mL–1 for CD). The intra-day assay was tested with 7 repeated injections of LD and CD solution, while the inter-day assay was studied by analyzing LD and CD solution injected 3 times every day, on seven consecutive days. The accuracy (relative error, RE) was calculated from the nominal concentration (Cnom) and the mean value of observed concentration (Cobs) as follows: RE (%) = [(Cobs − Cnom)/(Cnom)] × 100 The precision (RSD, %) was calculated from the observed concentrations as follows: RSD (%) = [SD/Cobs] × 100. SD is standard deviation. The intra-day and the inter-day precisions were no more than 6.7%. The intra-day accuracy varied from −2.4% to 4.0%, while the inter-day accuracy ranged from −2.9% to 2.7%. 3.5 Practical applications 3.5.1
Determination of levodopa and carbidopa in pharmaceutical formulation
The proposed HPLC–nanoCL method was applied to direct determination of LD and CD in the controlled-release tablets (Sinemet®, each tablet contains 200 mg of LD and 50 mg of CD). The average recoveries of LD and CD tablets are 100.5% and 103.1% with the precisions of 2.4% and 4.0%, respectively. The results for determination of LD and CD tablets were shown in Table 3, indicating that the values obtained by the proposed method were in good agreement with labels of pharmaceutical preparations. To further validate
analytical results, HPLC method with DAD at 275 nm was employed to measure the content of LD and CD in tablets. A good agreement was achieved, and there was no significant difference between the results of two methods. Therefore, the proposed HPLC-nanoCL method could be applied for the quality control of pharmaceutical formulations contained LD and CD. 3.5.2
Determination of levodopa and carbidopa in human serum samples
To demonstrate the applicability of HPLC-nanoCL method in complex biological sample, the proposed method was used to analyze LD and CD in human serum samples. As shown in Fig.6, no peak was observed in the chromatogram of blank plasma (Fig.6A) and two peaks were found in the chromatogram of serum spiked with LD and CD (Fig.6B), indicating that none of endogenous compounds in human serum would interfere with the determination of LD and CD. The high selectivity for detection of LD and CD in human serum was obtained by a sample clean-up procedure performed on C18 SPE column. The recoveries of LD and CD in human serum by SPE purification at three spiked concentration levels (10, 40 and 160 ng mL–1 for LD; 20, 80 and 320 ng mL–1 for CD) were 61.4% and 63.5%, respectively. The analytical results of three human serum samples were summarized in Table 4. Good recoveries between 97.0% and 103.5% with RSDs of no more than 3.3% indicated that the present method could meet the requirements for simultaneous and sensitive quantification of LD and CD in human serum samples.
Table 3 Comparison of determination results of LD and CD in controlled-release tablets (Sinemet®) by two methods (n = 5) Batch No. J013155 J005774 J011060
This method
HPLC−DAD
Analyte
Declared (mg per tablet)
Found (μg)
RSD (%)
Found (μg)
RSD (%)
LD CD LD CD LD CD
200 50 200 50 200 50
207.2 50.6 196.8 48.6 204.6 52.3
2.4 4.0 3.1 2.5 2.0 4.0
198.8 51.0 203 49.0 202.4 51.8
1.9 1.2 1.8 1.6 1.0 1.5
Fig.6 Chromatograms of (A) blank human serum sample and (B) human serum spiked with 100 ng mL–1 levodopa and carbidopa. Peaks: (1) levodopa; (2) carbidopa. Chromatographic and CL conditions are the same as those of Fig.5
MU Chun-Lei et al. / Chinese Journal of Analytical Chemistry, 2017, 45(6): e1726–e1733
Table 4 Results for determination of LD and CD in human serum by using the proposed HPLC-nanoCL method (n = 3)
4
Human serum
Spiked (ng mL–1)
Found (ng mL–1)
LD
CD
LD
CD
LD
CD
LD
CD
1 2 3 4
0 8.0 40.0 200.0
0 10.0 50.0 250.0
ND 7.9 39 207
ND 9.7 51.1 252.3
− 98.7 97.5 103.5
− 97.0 102.2 100.9
− 3.3 1.6 1.1
− 2.5 2.0 0.5
Conclusions
Recovery (%)
RSD (%)
1998, 46(1): 39−44 [14] Pistonesi M, Centurión M E, Band B S F, Damiani P C,
In this work, based on the enhancing effect of LD and CD on the on-line AuNPs-catalyzed luminal-H2O2 CL reaction, a new HPLC-nanoCL method was developed for simultaneous and selective determination of LD and CD. The high sensitivity and reproducibility of on-line AuNPs-catalyzed CL responses, together with the ease of sample pretreatment, make the proposed method very useful for quantification of LD and CD in pharmaceutical formulations and human serum. It is believed that the presented HPLC-nanoCL method had promising potential in monitoring LD and CD at trace levels in serum and urine samples in Parkinson’s patients.
Olivieri A C. J. Pharm. Biomed. Anal., 2004, 36(3): 541–547 [15] Teixeira M F, Marcolino-Junior L H, Fatibello-Filho O, Dockal E R, Bergamini M F. Sens. Actuators B, 2007, 122(2): 549−555 [16] Raut P P, Charde S Y. Luminescence, 2014, 29(7): 762−771 [17] Vilhena R D O, Pontes F L D, Marson B M, Ribeiro R P, de Carvalho K A T, Cardoso M A, Pontarolo R. J. Chromatogr. B, 2014, 967(15): 41−49 [18] Junnotula V, Licea-Perez H. J. Chromatogr. B, 2013, 926(1): 47−53 [19] César I C, Byrro R M D, Mundim I M, Teixeira L S, Gomes S A, Bonfim R R, Pianetti G A. J. Mass Spectrom., 2011, 46(9): 943−948
References
[20] Sagar K A, Smyth M R. J. Pharm. Biomed. Anal., 2000, 22(3): 613−624
[1]
Dauer W, Przedborski S. Neuron, 2003, 39(6): 889−909
[2]
Hu G, Chen L, Guo Y, Wang X, Shao S. Electrochim. Acta, 2010, 55(16): 4711−4716
[3]
Cotzias G C, van Woert M H, Schiffer L M. New Engl. J. Med., 1967, 276(7): 374−379
[4]
Okereke C S. J. Pharm. Pharm. Sci., 2002, 5(2): 146−161
[5]
Reynolds J E F (ed), Martindale: The Extra Pharmacopoeia, The Pharmaceutical Press, London, 1993
[6]
Gomes P, Soares-da-Silva P. Neuropharmacology, 1999, 38(9): 1371−1380
[7]
Le Witt P A, Nelson M V, Berchou R C, Galloway M P, Kesaree N, Kareti D, Schlick P. Neurology, 1989, 39(11 Suppl 2): 45−53
[8]
Hammerstad J P, Woodward W R, Nutt J G, Gancher S T, Block G A, Cyhan G. Clin. Neuropharmacol., 1994, 17(5): 429−434
[9]
Titus D C, August T F, Yeh K C, Eisenhandler R, Bayne W F, Musson D G. J. Chromatogr., 1990, 534: 87−100
[10] Zhao S, Bai W, Wang B, He M. Talanta, 2007, 73(1): 142−146 [11] Marques K L, Santos J L, Lopes J A, Lima J L. Anal. Sci., 2008, 24(8): 985–991 [12] Nagaraja P, Vasantha R A, Sunitha K R. J. Pharm. Biomed. Anal., 2001, 25(3): 417–424 [13] Nagaraja P, Murthy K C, Rangappa K S, Gowda N M. Talanta,
[21] Karimi M, Carl J L, Loftin S, Perlmutter J S. J. Chromatogr. B, 2006, 836(1-2): 120−123 [22] Chen F N, Zhang Z J, Zhang Y X, He D Y. Anal. Bioanal. Chem., 2005, 382(1): 211−215 [23] Chen J, Guo Y Y, Chen Q, Qiu H D. Chin. J. Anal. Chen., 2015, 43(10): 1513−1519 [24] Chen Z Q, Chen H, Xie X, Yang M. Chinese J. Anal. Chem., 2015, 43(7): 1075−1079 [25] Zhang L, Niu W X, Gao W Y, Qi L M, Lai J P, Zhao J M, Xu G B. ACS Nano, 2014, 8(6): 5953−5958 [26] Zhang Z F, Cui H, Lai C Z, Liu L J. Anal. Chem., 2005, 77(10): 3324−3329 [27] Zhang Q L, Wu L, Lv C, Zhang X Y. J. Chromatogr. A, 2012, 1242(15): 84−91 [28] Mu C L, Zhang Q, Wu D, Zhang Y J, Zhang Q L. Biomed. Chromatogr., 2015, 29(1): 148−155 [29] Lu C, Li Y, Yang J M. Talanta, 2010, 82(4): 1676−1580 [30] Beitollahi H, Mostafavi M. Electroanal., 2014, 26(5): 1090−1098 [31] Mazloum-Ardakani M, Khoshroo A. Anal. Chim. Acta, 2013, 798(10): 25−32 [32] Molaakbari E, Mostafavi A, Beitollahi H, Alizadeh R. Analyst, 2014, 139(17): 4356−4364