Enhanced selectivity approach for fast analysis of enalaprilat , lisinopril and benazepril in pharmaceutical dosage forms and spiked human plasma by ion chromatography

A simple and highly selective method for determination of enalaprilat (ENT), lisinopril (LSP) and benazepril hydrochloride (BZP) in pharmaceutical dosage forms as well as spiked human plasma by ion chromatography with UV detection has been developed. ENT and LSP are compounds with little inherent hydrophobicity that make their analysis using HPLC essentially sophisticated necessitating either complex extraction procedures or especial detection methods. The method proposed is based upon ionization of those two drugs in basic medium to be well retained and readily resolved as dicarboxylic acid anions using IonPac AS11 (13 μm particle size, 4 × 250 mm, Dionex) anion exchange column adopting a very simple clean-up procedure via a single-step protein precipitation with UV detection at 215 nm. The three drugs were successfully determined in pharmaceutical dosage forms as well as in spiked human plasma with tyrosine as internal standard with a recovery approaching 100%. Adopting this proposed procedure, the analytes produce well shaped peaks with good linear relationship over the investigated concentration ranges and values of (r) higher than 0.998 for all drugs. ion level of200 ng ml -1 of the three drugs, therefore being satisfactory for their purposed analysis. The method was validated with respect to specificity, recovery, accuracy, precision and linearity. Moreover, the method could be applied to the determination of the drugs in pharmaceutical preparations. The method was validated with respect to specificity, recovery, accuracy, precision and linearity and the method proved applicable for routine fast analysis of the studied drugs.


INTRODUCTION
Angiotensin converting enzyme (ACE) inhibitors have achieved widespread usage in the treatment of cardiovascular and renal diseases.ACE inhibitors alter the balance between the vasoconstrictive, salt-retentive, and hypertrophic properties of angiotensin II and the vasodilatory and natriuretic properties of bradykinin and alter the metabolism of a number of other vasoactive substances.They are especially important because they have been shown to prevent early death resulting from hypertension, heart failure or heart attacks (Brown et al., 1998).
Enalaprilat (ENT), lisinopril (LSP) and benazepril hydrochloride (BZP) are important members of this group that is nowadays being massively administrated for a variety of cardiovascular and renal disorders increasing the need for more convenient methods for their investigation and monitoring in their dosage forms and biological fluids.A wide range of methods were reported in the literature for their analysis in pharmaceutical formulations and biological fluids.These include spectrophotometric (Belal et al., 2000;Bonazzi et al., 1997;El-Emam et al., 2004;El-Gindy et al., 2001;El-Yazbi et al., 1999), *Corresponding author.E-mail: zhuyan@zju.edu.cn.Tel: +86 571 88273637.Fax: +86 571 882736.
Although HPLC is a commonly used method in bioanalytical laboratories (Ahmad et al., 2011;Khan et al., 2011;Nama et al., 2011;Prakash et al., 2011), it is difficult to develop an HPLC method for ENT and LSP pharmacokinetic application, due to their low plasma concentration and endogenous interference.

Regents and materials
Benazepril HCl, benazeprilat, enalaprilat, and lisinopril were all purchased from China's National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China).HPLCgrade methanol was obtained from Tedia Company (Fairfield, USA).Analytical reagent grade sodium hydroxide was obtained from Sinopharm Chemical Reagent Co., Ltd (China).Acetic acid (HAC) was delivered from Jinlu Chemicals (Shanghai, China).Ammonia solution was obtained from Jinsha Chemicals (Shantou, China).Tyrosine internal standard (I.S.) was bought from Kang DaAmino Acids Co. (Shanghai, China).Ultra-pure water was prepared by the Millipore Milli-Q Academic water-purification system (Molsheim, France).Drug products were bought from the local market.Drug-free human plasma was supplied by the Blood Bank of Zhejiang Province.Waters Oasis® (3 cc/60 mg, 30 µm) HLB (hydrophilic-lipophilic balanced) reversed-phase sorbent solid phase extraction cartridges (Milford, USA) was used to isolate the drugs from plasma samples.

Preparation of drugs stock solutions, calibration standards, quality-control samples and other reagents
Standard stock solutions of each analyte and the IS were prepared separately as 20 mg% in distilled water and stored at 4°C for a maximum of two weeks.Working solutions were prepared by appropriate dilution of stock solutions using distilled water and stored at 4°C.Calibration standards in the concentration range from 0.2 to 4 μg ml -1 for the three drugs were prepared by spiking 0.45 ml of blank plasma with 50 μl of the standard drugs mixture solution.QC samples were prepared in bulk at the concentration of 0.2, 1 and 4 μg/ml and stored at -20°C.50 mM HAC and ammonia solution were prepared by appropriate dilution of the calculated volumes into water

Drug formulations
A powdered sample of each drug tablets was extracted for 15 min by ultrasonication using distilled water and then completed to volume.For ENT injection, an accurately measured volume was directly diluted with deionized distilled water.The content of the drug samples was calculated by referring to its linear regression equation obtained from analysis of their respective standards.

Analysis of spiked human plasma samples
Adopting protein precipitation procedure: Aliquots of 500 μl plasma samples were mixed with 25 μl IS solution (10 μg ml -1 ) followed by 475 μl methanol being added.The mixture was vortex mixed for about 30 s before being centrifuged at 10000 G for about 20 min.Then, an aliquot of 800 μl of the clear supernatant were pipetted and was evaporated under vacuum in a stream of nitrogen at about 40°C and the residue was dissolved in 200 μl water.The sample was filtered through a 0.22 um syringe filter and an aliquot of 20 μl were directly injected to the system.

Adopting solid phase extraction (SPE) procedure:
To each 500 μl plasma samples, 50 μl IS solution (2.5 μg ml -1 ) and 550 μl 50 mM HAC were added in a 4 ml polypropylene tube.The mixtures were vortex-mixed for 30 s, and then loaded to the preconditioned SPE cartridge.The cartridge was preconditioned by washing with 2 ml water, 2 ml methanol then another 1 ml water and finally 1 ml 50 mM HAC.After loading the sample, the cartridge was washed by 1 to 2 ml 50 mM HAC then 1 ml 20% methanol in 50 mM HAC.The sample was then eluted using 1 ml 10% 50 mM ammonia in methanol.The collected eluate was evaporated under vacuum in a stream of nitrogen at about 40°C and the residue was dissolved in 125 μl water.The sample was filtered through a 0.22 um syringe filter and an aliquot of 20 μl were directly injected to the system.

Chromatographic system
The chromatographic system was Dionex Ultimate 3000 series (USA) equipped with a quaternary pump, a temperature controlled auto injector.The Ultimate 3000 quad-wavelength ultraviolet detector was set at 215 nm for all drugs.The separation was performed at ambient temperature on a Dionex IonPac AG11 guard column (13 μm particle size, 4 × 50 mm) and a Dionex IonPac AS11 analytical column (13 μm particle size, 4 × 250 mm).The data collection and analysis were run with Dionex Chromeleon Client Software (USA).The mobile phase consists of 30 mM NaOH, methanol and water (50:36:14) in isocratic mode at a flow rate of 1.0 ml/min.

Assay validation
The extraction recoveries of each drug and I.S. were determined by comparing the peak area of extracted spiked samples to those of standard solutions at same concentration.Calibration curves were constructed with plasma standards spiked with 0.2, 0.5, 1, 2 and 4 μg/ml of each drug.During the method validation, calibration standards were independently prepared and measured on 3 consecutive days.The intra-day and inter-day precision and accuracy were determined based on Quintuplicate measurements of QC samples at low, middle and high concentrations (0.2, 1 and 4 μg mL -1 ).All samples were spiked with the drug standards on Day 1, and then analyzed on five different days.The drugs concentrations in QC samples were calculated from the linear regression equation obtained on the same day.

RESULTS AND DISCUSSION
LSP per se and ENT as a metabolite of the prodrug enalapril are amphoteric in nature having both acid carboxylic and basic amino group functionalities in their structure (Figure 1).The pKa of their carboxylic moieties lying in the range of 2 to 4 (Lemke and Williams, 2008) indicate significantly ionizable acidic functions.In alkaline pH, they behave as dicarboxylic acids with substantial anionic nature that keep them highly retained on anionic exchange columns giving the chance for their selective retention and subsequent analysis.The previously mentioned drugs were sufficiently polar to be eluted in the first part of the chromatograms in most published HPLC methods even with a small percentage of organic modifiers.This makes their analysis a point of sophistication due to their subject of interference from endogenous components in relatively complex matrix and their significant loss from most like plasma with a very simple protein precipitation clean up procedure using methanol.Alternatively and for more elegant chromatograms, Water Oasis HLB sorbent cartridges were used for sample purification.The unique outcome of using this sorbent for extracting those relatively polar compounds is the promising recovery values as well as enhanced extraction selectivity with consequent optimum resolution.The procedure was so simple to be applied for their routine analysis in less than 5 min.

Development of the IC method
The IonPac AS11 Analytical Columns are specifically designed to resolve a large number of inorganic anions and organic acid anions from a single sample injection in one run using hydroxide eluent systems.AS11 columns are stable between pH 0 and 14 and are compatible with eluents containing 0 to 100% organic solvents (Dionex Reference Library ver.33 (CD-ROM), 2008).At moderately high pH values, the amine functionality of the studied drugs is deprotonated and the carboxyl group is ionized and free for electrostatic interaction with available positively charged ionizable sites (alkanol quaternary ammonium groups) of the column stationary phase resin.Thus, NaOH solution was chosen as mobile phase providing the optimal experimental conditions in terms of pH and solution compatibility for acceptable resolution via anion-exchange chromatography and satisfactory detection.In order to evaluate the effect of the hydroxide concentration on the retention times, a standard mixture of the drugs was eluted isocratically with mobile phase containing various concentrations of NaOH.Only, ENT and LSP were eluted as broad markedly tailed peaks and their retention time as expected being decreased with increasing hydroxide concentration.Therefore, methanol was added to the mobile phase in order to modify the hydrophobic interactions of the analytes giving more symmetric Gaussian shaped peaks necessary for reproducible retention and optimum quantitation.Only 10% was found enough for that purpose; however, 36% was used to permit elution of BZP as a well resolved peak in a reasonable time together with the other drug peaks.BZP is mainly retained according to its high hydrophobic properties; nevertheless its anionic characters are also considered as could be indicated by the inability of 50 or even higher percentages of methanol in water mixtures to affect its elution from the column without a specific anion eluent like hydroxide.By performing the chromatographic separation using 30 mM NaOH, methanol, water (50:36:14) as mobile phase, the selected drugs were separated within less than 5 min with a good S/N ratio considering the column efficiency and UV detector reliability.The brilliant advantage lying in choosing AS11 column is its strong anion exchange capacity in accordance with very low hydrophobicity enhancing selectivity and affecting superior resolution of anionic drugs from hydrophobic and neutral compounds.
A typical chromatogram obtained with an IonPac AS11 analytical column and UV detection at 215 nm for the selected analytes together with tyrosine (I.S.) is shown in Figure 2.

Applications
The developed IC method was applied for the determination of the studied drugs in their commercial tablets and results were found in good agreement with the claimed content.Concentrations of the drugs were calculated by referring to the calibration equations obtained for pure standards shown in Table 1.Results for the analysis of different tablet dosage forms are shown in Table 1.The chromatograms obtained with BZP and LSP tablets as well as ENT injection were shown in Figure 3. On the other hand, the method can differentiate intact BZP dosage forms from degraded ones.Benazeprilat (BZT), the main degradation product of BZP (Gana et al., 2002), was not eluted within at least 10 min under the proposed experimental conditions.This should be explained on the basis of more anionic features of the degraded product with respect to intact one due to one additional carboxyl group being produced during degradation pathway.This could be a great advantage for quality control and drug evaluation protocols.Unfortunately, Diketopiperazines of ENT and LSP as their main degradation products were not available to be tried in order to investigate the capability of the method to evaluate the drugs stability in their dosage forms.Analysis of spiked human plasma was also successfully conducted.Calibration graphs were constructed for the studied drugs in spiked human plasma samples according to the proposed procedures detailed under the experimental section.Results are summarized in Table 2. Adopting the protein precipitation procedure, lower concentrations were also tried and better sensitivity and detection limits for the drugs in plasma samples could be achieved by reconstituting drug residues after extraction and evaporation in lower injection volumes such as 100 μl but for more convenience and better baseline 200 μl was chosen (Figure 4).In addition, adopting the same procedure, some endogenous compounds in plasma interfered with first eluted drug LSP, however due to the great difference in their inherent anionic characters, the drug could be easily resolved from these components by decreasing the percentage of NaOH used in the mobile phase from 50 to 25% (Figure 5).Still for other drugs, 50% NaOH was preferred for faster, sharper, easier and more accurately quantified peaks.Using 25% NaOH, there was interference with tyrosine peak while that of ENT became a little bit broader.
However, the three drugs can be determined simultaneously adopting the HLB SPE extraction procedure with even much better quantification limits.Oasis HLB is a Hydrophilic-Lipophilic-Balanced, water-wettable, reversed-phase sorbent.It is made from a specific ratio of two monomers, hydrophilic N-vinylpyrrolidone and lipophilic divinylbenzene.It provides superior reversedphase capacity with a neutral polar "hook" for enhanced retention of polar analytes (Waters Corporation, 2008).This extraordinary retention of polar compounds, compared to traditional silica-based SPE sorbents like C18 gives rise to optimal recovery values as well as much better confident sample clean-up as a result of an extra washing step using 20% methanol solution.This washing step is prohibited using C18 SPE sorbents due to serious loss of sample with little or even no drug recovery.A comparison between the drugs recovery from extraction extraction using HLB and C18 cartridges is shown in Figure 6.These outcomes, in addition to obviating the need for complex derivatization steps have a distinct impact on the method sensitivity with even much less quantification limits.The chromatogram obtained using waters Oasis HLB SPE cartridge was shown in Figure 7.

Specificity and recovery
Representative chromatograms of blank plasma samples as well as their spiked parallels were shown in Figures 4,  (A) (B) Figure 4. Representative typical IC-UV chromatograms of a blank human plasma sample from a healthy human volunteer (A) and that spiked with a combined Drugs standard mixture solution, 1 μg mL -1 each, (B) prepared according to the protein precipitation procedure.Mobile phase composition: 30 mM NaOH, methanol and water (50:36:14).BZP, 1; ENT, 2; LSP, 3; I.S., 4.

Calibration graphs
Regression parameters for drug standard mixtures including the slopes intercepts and regression coefficients are summarized for all the analyzed drugs in Table 1.These regression parameters were used for calculation of drugs concentrations in their pharmaceutical tablets.
For the analysis of the studied drugs in plasma samples, an internal standard was necessary to compensate for any variability of the extraction recoveries during sample preparation.Calibration equations were obtained by plotting each drug peak area to IS ratio vs. the concentration of each drug.In all cases, the intercepts were not significantly different from zero and correlation coefficients were greater than 0.998.Lower limits of detection (LOD) and quantification (LOQ) were also calculated from the obtained chromatograms for a signal-to-noise ratio (S/N) of 3 and 10 (n=8), respectively.Results are shown in Table 2.This analytical performance in terms of detection limits is generally sufficient for analysis of LSP and BZP peak plasma levels after a single oral dose of 20 mg or ENT after its IV injection, especially if some procedure of enhancing the sensitivity like post extraction sample evaporation was applied.The method was superior to other methods applying extensive derivatization techniques with more advantages of intact dug detection and enhanced drug selectivity and resolution.Another advantage is that the linear ranges showed good correlation coefficients (that is, >0.998) giving rise to excellent accuracy results for all drugs, this was referred to the super signal stability of the UV detectors if compared with other amperometric sensors or photometric detectors after pre-or post-column derivatization.a , With respect to A = bC + a, where C is the concentration in (µg ml −1 ), A is the peak area, a is the intercept and b is the slope.b , Lower limit of detection.
c , Lower limit of quantification.

Precision and accuracy
The intra-and inter-day precision and accuracy of the developed method were evaluated with five replicates of samples at concentration of 0.2, 1 and 4 μg/ml, and on five different days.The precisions ranged from 2.89 to 7.63% RSD for the three studied drugs at the three concentration levels which should be satisfactory to , Expressed as relative standard deviation (RSD) (n=5) on the same day and five different days, respectively.
e , Calculated on the basis of a signal to noise ratio of 3. f , Calculated on the basis of a signal to noise ratio of 10 determine the drugs in the sample matrix.The method proved to be sufficiently accurate with mean percentage recovery in the range of 96.34 to 102.52 at the three concentration levels for all the studied drugs.The drugs concentrations in QC samples were calculated from the linear regression equation obtained on the same day.
Results of analysis are shown in Table 2.

Conclusion
The method proposed in this work permits selective confident measurements of BZP, ENT and LSP in pharmaceutical formulations as well as spiked human plasma with a very simple clean up procedure without any serious interference from other endogenous matrix components.The fundamental advantage for the procedure lies in its ability to enhance drugs retention allowing a confident analysis in complex matrix with highly simple, convenient chromatographic procedure within less than 5 min making it an easy decision about applying the method for routine analysis in large pharmaceutical companies or medical laboratories.The suitability of the proposed procedure for analysis of the studied drugs in biological fluids with a simple technique and equipment is a great outcome.Drug concentrations matching their pharmacokinetic blood levels can be efficiently conducted making the method a good alternative for many published methods.

Figure 1 .
Figure 1.Chemical structures of the studied drugs.

Table 1 .
Collective calibration data for analysis of the studied drugs in their pharmaceutical formulations by th e proposed method

Table 2 .
Results for the analysis of the studied drugs in their spiked human plasma samples., Results shown for a selected quality control concentrations of 0.2, 1 and 4 μg ml -1 for all the studied drugs.
a b , Calculated as mean % recovery (n=5); c,d