Precolumn o -Phthalaldehyde- N -acetyl-L-cysteine Derivatization Followed by RP-HPLC Separation and Fluorescence Detection of Sitagliptin Enantiomers in Rat Plasma

CHIRALITY 25:883–889 (2013)

Precolumn o-Phthalaldehyde-N-acetyl-L-cysteine DerivatizationFollowed by RP-HPLC Separation and Fluorescence Detection of

Sitagliptin Enantiomers in Rat PlasmaR. NAGESWARA RAO,1* B. SRAVAN,1 K. RAMAKRISHNA,1 SHAIK SAIDA,1 AND RAJU PADIYA2

1Analytical Chemistry Division, Discovery Laboratory, Indian Institute of Chemical Technology, Tarnaka, Hyderabad, India2Pharmacology Division, Indian Institute of Chemical Technology, Tarnaka, Hyderabad, India

© 2013 Wiley Perio

ABSTRACT An indirect reversed-phase high-performance liquid chromatographic separationand fluorescence detection of sitagliptin enantiomers in rat plasma was developed and validated.Deproteinized rat plasma containing racemic sitagliptin was derivatized with o-phthalaldehydeand N-acetyl-L-cysteine under alkaline conditions, converted to diastereomers, and separated ona Lichrospher 100 RP-18e column using 20mM phosphate buffer and methanol (45:55 v/v) as amobile phase under isocratic mode of elution at a flow rate of 1.0mL/min. Fluorescence detectionwas performed at 330 and 450nm as excitation and emission wavelengths, respectively. Themethod was linear in the range of 50–5000ng/ mL for both enantiomers. The intra- and interdayaccuracy and precision were within the predefined limits of 15% at all concentrations. The methodwas successfully applied to a pharmacokinetic study of sitagliptin after 5mg/kg oral administrationto Wistar rats. Robustness of the method was evaluated using design of experiments. Chirality25:883–889, 2013. © 2013 Wiley Periodicals, Inc.

KEY WORDS: precolumn derivatization; sitagliptin; diabetes mellitus; enantiomer separation;rat plasma; fluorescence detection

*Correspondence to: R. Nageswara Rao, Indian Institute of ChemicalTechnology, Tarnaka, Hyderabad 500 607, India. E-mail: [email protected],[email protected] for publication 11 February 2013; Accepted 16 July 2013DOI: 10.1002/chir.22229Published online 12 September 2013 in Wiley Online Library(wileyonlinelibrary.com).

INTRODUCTIONDipeptidyl peptidase-IV (DPP-IV) inhibitors represent a

novel class of oral antihyperglycemic agents for treatment oftype 2 diabetes mellitus. Sitagliptin (STG) ( (2R)-4-oxo-4-[3-(trifluoromethyl)-5,6-dihydro[ 1,2,4]triazolo[4,3-a] pyrazin-7(8H)-yl]- 1-(2,4,5- rifluorophenyl)butan-2- amine) is an orallyactive, potent, and selective DPP-IV inhibitor for the treat-ment of type 2 diabetes mellitus. It enhances the levels of ac-tive glucagon-like peptide, other incretins, and facilitatesglucose-dependent insulin secretion.1,2 It also improves gly-cemic control and regulates the growth of insulin-producingβ-cells in pancreatic islets. STG has a bioavailability of 87%and exhibits low reversible binding to plasma proteins.The pharmacological activity of STG is assigned specificallyto R-enantiomer and the plasma binding capacity of twoenantiomers differs significantly influencing its efficacy.Thus, monitoring of undesired S-enantiomer of STG inbiological matrices is of great importance not only becauseof different pharmacological/toxicological properties butalso the possibility of different metabolism.3–5

A thorough literature search has revealed that liquidchromatography tandem mass spectroscopy (LC-MS/MS)has been widely used to determine STG in biological fluidssuch as plasma, urine, and hemodialysate of rats and humans.Zeng et al. developed a turbulence flow online extractionmethod followed by LC-MS/MS for determination of STG inhuman plasma, urine, and hemodialysate to support theclinical pharmacokinetic studies.6,7 However, the onlineextraction using high turbulence liquid chromatography re-quires trained personnel and dedicated instrumentation formethod transfer between laboratories. In order to overcomesuch problems, an offline liquid–liquid extraction followedby LC-MS/MS determination of STG in human plasma wasreported by Nigori et al.8 However, the liquid–liquid extrac-tion method requires large volumes of plasma and not a

dicals, Inc.

throughput procedure. Later, a high throughput proteinprecipitation / filtration followed by LC-MS/MS determina-tion of STG was reported by Zeng et al.9 To avoid matrixeffects contributed by plasma phospholipids, a molecularlyimprinted polymer based solid-phase extraction followedby high-performance liquid chromatography (HPLC) deter-mination of STG in rat plasma and urine was reported byRao et al.10 LC, LC-MS, and LDTD-MS/MS methods forsimultaneous identification and quantification of oral hypo-glycemic drugs including metformin in pharmaceuticalpreparations, plasma, and dried blood spots were alsoreported.11–13 Beconi et al. evaluated the pharmacokinetics,metabolism, and excretion of STG in rats and dogs byLC-MS/MS.14 In addition to LC-MS, spectrofluorimetryusing fluorescamine as the derivatization reagent was alsoused to determine STG in tablets and serum samples.15

However, all these methods were achiral in nature anddo not address the separation of STG enantiomers.Sohajda et al. studied the interaction of STG enantiomerswith β-cyclodextrin derivatives for their separation by capillaryelectrophoresis and optimized the mobile phase compositionconsisting of β-cyclodextrin and sulfobutylether- β-cyclodextrinin phosphate buffer.16 However, to the best of the authorsknowledge, no LC method either by direct or indirect analysisof STG enantiomers in plasma has been reported so far in theliterature.The present article describes a simple, rapid, and sensitive

indirect reverse-phase (RP)-HPLC method for analysis of STGenantiomers in rat plasma. It involves precolumn derivatization

NAGESWARA RAO ET AL.884

of STG with o-phthalaldehyde (OPA) and N-acetyl-L-cysteine(NAC) to form diasteriomeric iso-indoles which were thenseparated by RP-HPLC on a C18 column connected to afluorescence detector. The developed method was validatedaccording ICH guidelines17 and successfully applied to pharma-cokinetic studies of STG in rats.

EXPERIMENTALChemicals and Reagents

R-STG, R,S-STG were gift samples from Aurobindo Pharma (Hydera-bad, India), OPA and NAC (Sigma–Aldrich, St. Louis, MO), boric acid, so-dium hydroxid,e and disodium hydrogen phosphate (S.D. Fine-Chemical,Mumbai, India), methanol, and acetonitrile (Merck, Mumbai, India),glass-distilled and deionized water (Nanopure, Barnstead, Dubuque, IA)were used.

Instrumentation and Chromatographic ConditionsThe HPLC system consisted of two LC-20AD pumps, an RF- 10AXL

fluorescence detector, an SIL-20AC auto sampler, a DGU-20A5 degasser,and CBM-20A controller (all from Shimadzu, Kyoto, Japan),. An Elico,model LI 120, pH meter equipped with a combined glass–calomelelectrode was used for pH measurements. It was calibrated usingstandard buffers of pH 4.0, 7.0, and 9.2. The chromatographic and theintegrated data were recorded using an HP-Vectra (Hewlett Packard,Waldron, Germany) computer system using LC-Solution data-acquiringsoftware (Shimadzu,). Before delivering the mobile phase into thesystem, it was filtered through 0.45μm PTFE filters and degassed.Chromatographic separation was achieved on a Lichrospher 100 RP-18e(250 × 4.0mm, 5μm; Merck, Darmstadt, Germany) column using mobilephase consisting of 20mM disodium hydrogen phosphate buffer (pH7.2adjusted with ortho-phosphoric acid): methanol (45:55, v/v) in anisocratic elution mode at a flow rate of 1.0mL/min at 25°C. Fluorescencedetection was performed at 330 and 450 nm as excitation and emissionwavelengths, respectively. The injection volume was 20μL.

Preparation of ReagentsThe 0.1M borate buffer (pH10.0) was prepared by dissolving 0.6200g

of boric acid in ~80mL of water, adjusted pH to 10.0 using 2.0N NaOHand made up to 100.0mL with water. Stock solutions of 50mM OPA wereprepared by dissolving 0.0670g of OPA in 10mL methanol and 50mMNAC was prepared by dissolving 0.0408g of NAC in 10mL of methanol.The derivatization reagent was prepared by mixing 1.0mL each of50mM OPA and 50mM NAC in a 10mL volumetric flask and made upto 10mL with 0.1M borate buffer (pH10.0). The storage life of the stocksolution was at least 4weeks at 4°C. The working solution was preparedfresh at the time of derivatization. Stock solutions of the racemic-STG andR-STG were prepared in water to give a final concentration 1mg/mL.Working standards of STG were prepared by appropriate dilutions ofthe stock solution with water. All the solutions were stored in the darkat 4°C and brought to room temperature before use.

Fig. 1. Formation of diastereomers

Chirality DOI 10.1002/chir

RESULTS AND DISCUSSIONMethod Development

The indirect method of enantioresolution using chiralderivatization reagents has several advantages, such as excel-lent separation and detection possibilities of the resulting dia-stereomers, easy optimization of chromatographic conditions,and the prospect of using relatively inexpensive achiral col-umns.18 Nimura and Kinoshita demonstrated the precolumnderivatization of amino acid enantiomers with optically activethiol NAC in the presence of OPA.19 Since STG enantiomersalso possess primary amines, it is possible to convert theminto diastereomers on derivatization with OPA and NAC.20

Figure 1 represents the reaction of STG with NAC and OPAunder alkaline conditions resulting in the formation of diaste-reomers having different physicochemical properties. Thederivatization reaction was found to be complete within2min at room temperature and the products were separatedby RP-HPLC connected to a fluorescence detector. Duringmethod development a number of columns, i.e., PhenominexLuna C18 column, Zorbax Eclipse XDB-C18, and MerckLichrospher 100 RP-18e columns packed with octadecylsilane stationary phases were evaluated. The column thatgave the best results in terms of diastereomeric separationand endogenous compounds was Lichrospher 100 RP-18e.After varying methanol and buffer concentration at variouslevels, the mobile phase consisting of a mixture ofmethanol:20mM phosphate buffer pH 7.2 (55:45 v/v) wasselected and delivered in an isocratic mode at a flowrate of1.0mL/min at room temperature. Figure 2 shows the typicalchromatograms of A) a blank sample, rat plasma spiked withB) racemic STG and C) R-STG at HLOQ. It could be clearlyseen that the chiral impurity (S-STG) was well resolved fromdrug substance (R-STG) with a resolution factor >2.0. Typicalretention times of R- and S-STG enantiomers were 8.5 and10.3min, respectively.

METHOD VALIDATIONLinearity, Selectivity

A calibration curve was obtained by plotting the area underthe peak (AUP) against the nominal concentration. Linearitywas demonstrated by running plasma standards at five concen-trations over the range of 50–5000ng/mL. Themethod showedexcellent linearity for both enantiomers from 50–5000ng/mL.The linear regression line for (R)- STG was calculated asy = 41.314x-1865.1 with R2 = 0.9984 and for S-STG it wasy = 46.78x-2295.3 with R2 = 0.9995. It can be seen from Figure 2that no endogenous components extracted from blank plasma

of STG with OPA and NAC.

Fig. 2. Typical HPLC chromatograms of plasma (A) blank, (B) spiked with(R, S)-STG, and (C) spiked with R-STG at HLOQ.

PRECOLUMN DERIVATIZATION OF SITAGLIPTIN ENANTIOMERS IN RAT PLASMA 885

eluted at the retention times of the peaks of either R- and S-STG.The developedmethod was found to be selective for determina-tion of STG in rat plasma without interference from theendogenous constituents of plasma.

Accuracy and PrecisionThe data on intra- and interday accuracy and precision of

the method for determination of STG are summarized in

TABLE 1. Intra- and interday a

Nominal conc. (ng/mL) 50 5

Intraday R-STG S-S

Run 1 Mean concentration (ng/mL) 50.93 50.SD 0.55 0.Precision (%CV) 1.08 0.Accuracy(%bias) -1.86 -1.

Run 2 Mean concentration (ng/mL) 47.16 51.SD 1.04 0.Precision (%CV) 2.20 1.Accuracy(%bias) 5.66 -2.

Run 3 Mean concentration (ng/mL) 48.33 46.SD 1.25 0.Precision (%CV) 2.60 1.Accuracy(%bias) 3.33 7.

Interday Mean concentration (ng/mL) 51.83 50.SD 1.25 0.Precision (%CV) 2.42 1.Accuracy(%bias) -3.66 -1.

Table 1. All values obtained were well within recognizedacceptance criteria for assay validations and were withinthe predefined 15% limits. Replicate analyses (n = 5) ofquality control (QC) samples at three concentration levels(50, 1000, and 5000 ng/mL) were used for determining theaccuracy and precision of the method. Precision was calculatedas the coefficient of variation (CV %) within a single run(intraday) and among different runs (interday). The accuracywas calculated as percent bias, i.e., the deviation betweennominal and measured concentrations. Thus, the developedmethod possessed good accuracy and precision.

Limit of Detection (LOD) and Quantitation (LOQ)LOQ is defined as the lowest analyte concentration that gives

accuracy within ±15% (bias) of the nominal value and interassayprecision (CV) 15%. The LOD was defined as the lowestconcentration of the analyte that could be detected to producea significant response. LOD represents the concentration ofanalyte at an S/N ratio of 3 and LOQ at which S/N is 10 weredetermined experimentally. The LOQ for both (R)-STG and(S)-STG was found to be 50ng/mL.

RecoveryThe extraction recoveries of STG enantiomers from rat

plasma were determined. The internal standard was not triedbecause all quantitative measurements were carried out usingthe external standard method. Satisfactory results wereobtained using this method. The extraction recoveries of STGenantiomers were determined by comparing the mean peakareas of extracted plasma samples with direct injection of thestandard solutions at the same concentrations. The recoveriesof STG enantiomers were determined at three concentrationlevels of QC at low, medium, and high concentrations and aregiven in Table 2. The average extraction recoveries for R-STGand S-STG were 92.09 and 91.01%, respectively.

StabilityThe stability of STG was performed under all storage

conditions at three QC concentration levels (100, 1000,and 5000 ng/mL) in six replicates. The freeze/thaw stabil-ity test was performed after three complete freeze/thaw

ccuracy and precision data

0 1000 1000 5000 5000

TG R-STG S-STG R-STG S-STG

70 951.66 1050.0 4710.0 4776.6643 10.40 30.0 55.67 66.5886 1.09 2.85 1.18 1.3940 4.83 -5.0 5.80 4.4633 975.0 965.10 4853.33 4760.095 13.42 10.0 95.04 60.8285 1.37 1.03 1.95 1.266 2.50 3.5 2.93 4.2040 1063.33 1040.0 4833.33 4721.6685 14.43 20.0 59.65 62.5184 1.35 1.92 1.23 1.322 -6.33 -4.0 3.33 5.5693 1033.33 1046.66 4733.33 4586.6683 15.27 37.85 102.63 83.2663 1.47 3.61 2.13 1.8186 -3.33 -4.66 5.33 8.26


TABLE 2. Recovery data

Concentration(ng/mL)

Recovery (%)

R-STG S-STG

50 90.15 89.621000 92.65 90.475000 93.48 92.95

TABLE 3. Stability data

Storage conditions STG

Nominalconc.(ng/mL)

Mean conc.(ng/mL)

%Bias

Ambient at 25°C for 6 h 100 101.9 -1.91000 954.3 4.575000 4821 3.58

Freeze/thaw at -20°C–25°C for three cycles

100 101.2 -1.21000 933 6.75000 4756 4.88

Autosampler at 10°C for24 h

100 95.5 4.51000 1015 -1.55000 4785 4.3

Rat plasma at �20°C for30 days

100 103.2 -3.21000 955.3 4.475000 4657 6.86

TABLE 4. DOE data for robustn

(a) Selected parameters and their variations

Molarity of borate buffer (mM) (A)

pH of borate buffer (B)

OPA concentration (mM) (C)

NAC concentration (mM) (D)

(b) Plackett–Burman experimental designExp. no. Run order A B

1 6 75 9.52 10 125 9.53 19 75 10.54 5 125 10.55 4 75 9.56 7 125 9.57 12 75 10.58 14 125 10.59 18 75 9.510 11 125 9.511 16 75 10.512 17 125 10.513 15 75 9.514 13 125 9.515 9 75 10.516 1 125 10.517 2 100 10.018 8 100 10.019 3 100 10.0



cycles (�20 to 25°C). The long-term stability was assessedafter storage of QC samples at �20°C for 30 days. Forshort-term stability, QC samples were kept at ambienttemperature (25°C) for 6 h and then analyzed. Stability ofprocessed samples was assessed by reinjection ofextracted QC samples after 24-h conservation in anautosampler (set at 10°C). The stock solutions of STGstored at �20°C for 1month were compared with freshlyprepared stock solutions. The results are given in Table 3.It can be seen from Table 3 that the results are well withinthe acceptance limits.

RobustnessA “Plackett–Burman” design was used to test the robust-

ness of chromatographic separation and sample preparation.The experimental design is useful for this kind of study as itfacilitates the investigation of several parameters by reducingthe number of experiments.21

Sample preparation. Three variables with upper and lowerlimits as shown in Table 4a were investigated. The designmatrix with the factor settings is shown in Table 4b. Theexperiments were run randomly with plasma sample spikedwith 1000 ng/mL. The selected response was %CV. Theresults are summarized as scaled and centered coefficientplots in Figure 3. The results reveal that different combina-tions of significant parameters will not drastically affect theresponse, so that the developed method for sample prepara-tion was considered robust.

ess of sample preparation

– 0 +

75 100 125

9.5 10.0 10.5

45 50 55

45 50 55

C D %CVR-STG S-STG

45 45 2.68 3.4845 45 4.97 4.9245 45 3.28 4.4645 45 2.95 5.8355 45 0.73 2.7555 45 2.01 3.7455 45 1.75 1.4155 45 2.62 1.8245 55 3.80 4.6245 55 1.05 4.4945 55 1.15 1.7345 55 2.91 1.3055 55 3.46 2.7655 55 5.42 3.2955 55 3.56 3.1655 55 3.82 5.2150 50 4.79 3.8350 50 1.23 4.2150 50 2.91 5.14

Fig. 3. Scaled and centered coefficients of variation (% ) of peak areas (A)R-STG and (B) S-STG.


Chromatographic separation. Five factors are likely to besignificant in practical use of the method: methanol contentof the mobile phase, pH and molarity of the phosphate buffer,flowrate, and column oven temperature and were investigatedusing three variables with upper and lower limits as shown inTable 5a. The experiments were run randomly with plasma

TABLE 5. DOE data for robustness

(a) Selected parameters and their variations

MeOH in mobile phase (vol.%) (A)Molarity of phosphate buffer (mM) (B)Final pH of mobile phase (C)Flow rate (mL/min) (D)Oven temperature (°C) (E)

(b) Plackett–Burman experimental designExp. no. Run order B A C D1 6 15 50 7.0 0.82 18 15 60 7.0 0.83 2 25 50 7.0 0.84 16 25 60 7.0 0.85 17 15 50 7.4 0.86 11 15 60 7.4 0.87 19 25 50 7.4 0.88 5 25 60 7.4 0.89 12 15 50 7.0 1.210 1 15 60 7.0 1.211 7 25 50 7.0 1.212 8 25 60 7.0 1.213 15 15 50 7.4 1.214 14 15 60 7.4 1.215 10 25 50 7.4 1.216 3 25 60 7.4 1.217 4 20 55 7.2 1.018 9 20 55 7.2 1.019 13 20 55 7.2 1.0

Rs: resolution; TfR: tailing factor of R-STG; TfS: tailing factor of S-STG; RtR: retention

sample spiked with 1000 ng/mL STG. The selected responseswere resolution (Rs), tailing factor of R-STG (TfR), tailingfactor of S-STG (TfS), retention time of R-STG (RtR), andretention time of S-STG (RtS). The design matrix with thefactor settings is shown in Table 5b. The scaled and centeredcoefficient plots (Fig. 4) for the above responses revealed thatdifferent combinations of significant parameters did not affectresponses, so that the developed method was consideredrobust.

Application to a Pharmacokinetic StudySix Wistar rats, (200–220 g) housed under standard condi-

tions with ad libitum access to water and standard laboratorydiet throughout the experiments were used in the presentstudy. After a single dose by oral administration of 5mg/kgof STG to healthy Wistar rats (n = 6), blood samples(0.5mL) were collected. Serial blood samples were collectedinto processed test tubes at 0, 0.5, 1, 2, 3, 4, 6, 8, 10, 12, and24 h postdose. Specimens were thawed and allowed to reachroom temperature, derivatized with OPA and NAC, thenanalyzed by HPLC. The concentrations of STG weredetermined from the calibration curve on the same day. Forin vitro studies blood was collected from Wistar rats withoutdosing and plasma was separated from blood by centrifuga-tion. Rat plasma (0.2mL) was taken into a 1.5mL centrifugetube, then 100μL of STG working solution and 100μL ofperchloric acid (35%, w/w) were added. After vortex mixingfor 2min it was centrifuged for 10min at 2500 g. Then100μL of the supernatant was transferred to an autosamplervial. To this vial 200μL of derivatization reagent and 700μLof borate buffer were added. After sonicating for 1min,20μL of the reaction mixture was injected into the HPLC

of chromatographic separation

– 0 +

50 55 6015 20 257.0 7.2 7.40.8 1.0 1.220 25 30

E Rs TfR TfS RtR RtS30 3.68 1.04 1.04 18.49 14.6020 1.68 1.11 1.09 7.14 6.2220 3.96 1.05 1.06 31.85 23.9330 1.69 1.12 1.10 7.36 6.4920 3.99 1.03 1.03 26.88 20.4730 1.51 1.10 1.13 6.39 5.6630 3.41 1.03 1.04 15.93 12.6520 1.50 1.04 1.10 6.44 5.6420 3.51 1.04 1.03 17.42 13.2730 1.29 1.05 1.16 4.06 3.6130 3.54 1.08 1.07 16.44 12.9320 1.68 1.09 1.06 6.31 5.4230 3.37 1.06 1.05 13.43 10.6020 1.52 1.08 1.05 5.23 4.5220 3.18 1.06 1.09 15.06 11.4930 1.25 1.13 1.04 3.61 3.2425 2.50 1.05 1.07 10.25 8.4225 2.48 1.05 1.08 10.26 8.4925 2.46 1.06 1.08 10.36 8.51

time of R-STG; RtS: retention time of S-STG.


Fig. 5. Concentration-time profiles of R- and S- STG enantiomers.

Fig. 4. Scaled and centered coefficients of (A) resolution, (B) tailing factor of R-STG, (C) tailing factor of S-STG, (D) retention time of R-STG, and (E) retentiontime of S-STG.


system. The concentration–time profiles of R- and S- STGenantiomers in rat plasma were plotted and shown in Figure 5.The pharmacokinetics of STG was enantioselective and thepharmacokinetic parameters of the two enantiomers of STG;time at which the concentration reached the maximum (tmax),maximum plasma concentration (Cmax), area under plasmaconcentration–time curve (AUC), and half life (t1/2) werecalculated and given in Table 6. Except AUC, the remainingparameters were calculated using Microsoft Excel 2000while the pharmacokinetic software Ramkin, based on anoncompartment model, was used to calculate the [AUC]from the plasma drug concentration vs. time profiles. It canbe seen from Table 6 that the plasma concentration ofR-STG was higher than that of S-STG. Further the AUC0–24values were found to be high for R-STG compared toS-STG. These results were compared with that of Beconiet al.,14 which did not address the enantioselectivity ofSTG. The maximum plasma concentration (Cmax) of R-STG

TABLE 6. Pharmacokinetic parameters

Parameter R-STG S-STG

Cmax (ng/mL) 545 380tmax (h) 3.0 3.0AUC0→24 (h ng /mL) 6160 3966t1/2 ( h) 8.5 8.0

Cmax: maximum concentration of drug in rat plasma was observed at aparticular timepoint; tmax: time at which maximum concentration of drug inrat plasma was observed,; AUC: area under plasma concentration–time curve;t1/2: time required for decomposition of the drug into half of the concentration.


and the time at which the concentration reached the maxi-mum (tmax) were 545ng/mL and 3.0 h against 8.1 ng/mL and0.5-1.0 h, respectively, also very high as reported by Beconiet al. Area under plasma concentration–time curve (AUC) wasalso very high i.e. 6160ng/mL against 26.1 ng/mL.

CONCLUSIONThis is the first report on the application OPA-NAC

fluorescent probe in precolumn derivatization, separation, anddetermination of STG enantiomers in rat plasma. Thesignificant analytical advancement involves the enhancementof the quantification limits to 50ng/mL, which were almost 5times greater than those reported by El-Bagary et al.12 Further,the proposed method is simple and rapid, as the time requiredfor derivatization was <2min at room temperature. TheOPA-NAC-based reagent provided successful conversion ofSTG enantiomers completely into diastereomers and theirchromatographic resolution with good limits of detection. Thefluorescence detector (FL) is an attractive alternative to theconventional UV–Vis detector due to its high sensitivity.The method is simple, rapid, economical, and high-throughput,as it uses an inexpensive achiral column with short run time.The experimental design was found to be very helpful in testingthe robustness of sample preparation as well as chromato-graphic separation during the validation phase. The highsensitivity combined with high robustness of precolumn deriv-atization, good reproducibility, and high sample throughputmakes the method ideal for clinical as well as therapeutic drugmonitoring of R-STG in a bioanalytical laboratory.

ACKNOWLEDGMENTS

The authors thank the Director, IICT, for encouragement andpermission to submit the article for publication. B.S. and K.R.K. thank the Council of Scientific and Industrial Research(CSIR) and R.P. thanks the ICMR, New Delhi, India, forresearch fellowship.

LITERATURE CITED

1. Kim D, Wang LP, Beconi M, Eiermann GJ, Fisher MH, He HB, Hickey GJ,Kowalchick JE, Leiting B, Lyons K, Marsilio F, McCann ME, Patel RA,Petrov A, Scapin G, Patel SB, Roy RS, Wu JK, Wyvratt MJ, Zhang BB, ZhuL, Thornberry NA, Weber AE. (2R)-4-Oxo-4- 3-(trifluoromethyl)-5,6-dihydro1,2,4 triazolo 4,3-alpha p yrazin-7(8H)-yl -1-(2,4,5-trifluorophenyl)butan-2-amine: A potent, orally active dipeptidyl peptidase IV inhibitor for thetreatment of type 2 diabetes. J Med Chem 2005;48:141–151.

2. Herman GA, Stevens C, Van Dyck K, Bergman A, Yi BM, De Smet M,Snyder E, Hilliard D, Tanen M, Tanaka W, Wang AQ, Zeng W, MussonD, Winchell G, Davies M, Ramael S, Gottesdiener KM, Wagner JA.

Pharmacokinetics and pharmacodynamics of sitagliptin, an inhibitor ofdipeptidyl peptidase IV, in healthy subjects: Results from two randomized,double-blind, placebo-controlled studies with single oral doses. ClinPharm Ther 2005;78:675–688.

3. Herman GA, Stein PP, Thornberry NA, Wagner JA. Dipeptidylpeptidase-4 inhibitors for the treatment of type 2 diabetes: Focus onsitagliptin. Clin Pharm Ther 2007;81:761–767.

4. Kieffer TJ, Habener JL. The glucagon-like peptides. Endocr Rev1999;20:876–913.

5. Meier JJ, Nauck MA. Glucose-dependent insulinotropic polypeptide/gastric inhibitory polypeptide. Pract Res Clin Endocrinol Metab2004;18:587–606.

6. Zeng W, Musson DG, Fisher AL, Wang AQ. Determination of MK-0431 inhuman plasma using high turbulence liquid chromatography onlineextraction and tandem mass spectrometry. Rapid Commun MassSpectrom 2006;20:1169–1175.

7. Zeng W, Musson DG, Fisher AL, Chen L, Schwartz MS, Woolf EJ, WangAQ. Determination of sitagliptin in human urine and hemodialysate usingturbulent flow online extraction and tandem mass spectrometry. J PharmBiomed Anal 2008;46:534–542.

8. Nirogi R, Kandikere V, Mudigonda K, Komarneni P, Aleti R, BoggavarapuR. Sensitive liquid chromatography tandem mass spectrometry methodfor the quantification of sitagliptin, a DPP-4 inhibitor, in human plasmausing liquid-liquid extraction. Biomed Chromatogr 2008;22:214–222.

9. Zeng W, Xu Y, Constanzer M, Woolf EJ. Determination of sitagliptininhuman plasma using protein precipitation and tandemmass spectrometry.J Chromatogr B 2010;878:1817–1823.

10. Rao RN, Maurya PK, Khalid S. Development of a molecularly imprintedpolymer for selective extraction followed by liquid chromatographic deter-mination of sitagliptin in rat plasma and urine. Talanta 2011;85:950–957.

11. Hess C, Musshoff F, Madea B. Simultaneous identification and validatedquantification of 11 oral hypoglycaemic drugs in plasma by electrosprayionisation liquid chromatography-mass spectrometry. Anal Bioanal Chem2011;400:33–41.

12. El-Bagary RI, Elkady EF, Ayoub BM. Liquid chromatographic determina-tion of sitagliptin either alone or in ternary mixture with metformin andsitagliptin degradation product. Talanta 2011;85:673–680.

13. Swales JG, Gallagher RT, Denn M, Peter RM. Simultaneous quantitationof metformin and sitagliptin from mouse and human dried blood spotsusing laser diode thermal desorption tandem mass spectrometry. J PharmBiomed Anal 2011;55:544–551.

14. Maria GB, James RR, Yohannes T, Yuan-Qing X, Christopher JK, DavidQL, Shiyao X, Charles SE, Suzanne C, Donald FH, Ralph AS, Stella HV.Disposition of the dipeptidyl peptidase 4 inhibitor sitagliptin in rats anddogs. DMD 2007;35:525–532.

15. Caglar S, Onal A, Toker S. determination of sitagliptin with fluorescaminein tablets and spiked serum samples by spectrofluorimetry and a degrada-tion study. Curr Pharm Anal 2012;8:278–285.

16. Sohajda T, Hu WH, Zeng LL, Li H, Szente L, Noszál B, Béni S. Evaluationof the interaction between sitagliptin and cyclodextrin derivatives bycapillary electrophoresis and nuclear magnetic resonance spectroscopy.Electrophoresis 2011;32:2648–2654.

17. I.C.H. Anonymous. Topic Q2B. In: Validation of analytical methods –

methodology. Geneva: IFPMA; 1996.18. Bhushan R, Martens J. Amino acids —chromatographic separation

and enantioresolution. New York: HNB Publishing; 2010;1–322.(ISBN 978-0-9728061-3-8).

19. Nimura N, Kinoshita T. o-Phthalaldehyde-N-acetyl-L-cysteine as a chiralderivatization reagent for liquid chromatographic optical resolution ofamino acid enantiomers and its application to conventional amino acidanalysis. J Chromatogr 1986;352:169–177.

20. Lanchote VL, Santos VJ, Cesarino EJ, Dreossi SA, Mere Júnior Y, SantosSR. Enantioselective analysis of N-hydroxymexiletine glucuronide inhuman plasma for pharmacokinetic studies. Chirality 1999;11:85–90.

21. Rao RN, Maurya PK, Shinde DD, Khalid S. Precolumn derivatizationfollowed by liquid chromatographic separation and determination oftramiprosate in rat plasma by fluorescence detector: Application topharmacokinetics. J Pharm Biomed Anal 2011;55:282–287.


Documents

Precolumn o -Phthalaldehyde- N -acetyl-L-cysteine Derivatization Followed by RP-HPLC Separation and Fluorescence Detection of Sitagliptin Enantiomers in Rat Plasma