Xenobiotica the fate of foreign compounds in biological systems
ISSN: 0049-8254 (Print) 1366-5928 (Online) Journal homepage: http://www.tandfonline.com/loi/ixen20
Advantages of achiral h.p.l.c. as a preparative step for chiral analysis in biological samples and its use in toxicokinetic studies A. P. Beresford, K. Caswell, R. Chambers & I. P. Kirk To cite this article: A. P. Beresford, K. Caswell, R. Chambers & I. P. Kirk (1992) Advantages of achiral h.p.l.c. as a preparative step for chiral analysis in biological samples and its use in toxicokinetic studies, Xenobiotica, 22:7, 789-798, DOI: 10.3109/00498259209053141 To link to this article: http://dx.doi.org/10.3109/00498259209053141
Published online: 22 Sep 2008.
Submit your article to this journal
Article views: 5
View related articles
Citing articles: 1 View citing articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ixen20 Download by: [RMIT University Library]
Date: 25 April 2016, At: 10:31
XENOBIOTICA,
1992, VOL. 22,
NO.
7, 789-798
Advantages of achiral h.p.1.c. as a preparative step for chiral analysis in biological samples and its use in toxicokinetic studies A. P. BERESFORD, K. CASWELL, R. CHAMBERS and I. P. K I R K Drug Metabolism Department, Glaxo Group Research Limited, Greenford Road, Greenford, Middlesex UB6 OHE, UK
Downloaded by [RMIT University Library] at 10:31 25 April 2016
Received I2 June 1991; accepted 21 February 1992 1. Achiral reverse-phase h.p.1.c. with semi-automated post-column fraction collection and solid-phase sample reconcentration, has been applied as the purification procedure during the enantiomeric quantification of two widely differing experimental drugs; an HMG-CoA reductase inhibitor (I) and an a,-adrenoceptor antagonist (11). 2. The robust and specific achiral methodologies were available prior to the need for chiral analyses and recovery of drug from the fractions provided clean samples from a variety of biological matrices, without the need to develop compatible achiral/chiral mobile phases. 3. Compared with direct chiral chromatography of plasma extracts, this approach decreased the potential for metabolites and endogenous components to interfere or impair the performance of the chiral stationary phase.
4. The availability of quantitative data from achiral analysis of samples negated the need for internal standardization of the chiral analyses, helped confirm assay specificity and provided potential to determine enantiomeric ratios where only one isomer could be accurately measured.
5. Routine enantiomeric analyses were successfully carried out on samples taken from animals dosed orally with the racemic drugs, providing important data on the possible levels of exposure to individual enantiomers during toxicity testing.
Introduction Many drug substances under investigation contain one or more asymmetric centres, and it is frequently necessary to obtain pharmacokinetic data on individual isomers at an early stage of development. T h e outcome of such analyses may influence further progression of the compound as a single isomer or as a mixture. Samples may be supplied from early toxicology studies and in a variety of biological matrices, and at this stage rapid assay development could be advantageous. T h e enantioselective binding properties of plasma proteins have been well documented over the years (Muller and Wollert 1975, Brown et al. 1977, Albani et al. 1984) and the use of a,-acid glycoprotein and serum albumin in h.p.1.c. stationary phases offers a wide range of enantiomeric separation capabilities to the analyst (Schill et al. 1986, Allenmark 1986). Although the limited capacity of such columns for drug molecules (Allenmark 1989) may not be a problem in analytical methods, the detrimental effects on column performance of co-injected biological residues certainly can be (Enquist and Hermansson 1989). Consequently some chiral separations require extensive sample preparation and others may never become applicable to biological samples. A further problem during analysis of in viwo samples on chiral stationary phases is the unpredictable elution characteristics, not only of endogenous materials, but also of metabolites of the compound of interest (Chu and Wainer 1988). 0049-8254/92 $3.00 0 1992 Taylor & Francis Ltd.
A. P . Beresford et al.
790
II
I
Figure 1 .
Structures of the HMG-CoA reductase inhibitor, I, and the a,-adrenoceptor antagonist, fluparoxan (11).
Downloaded by [RMIT University Library] at 10:31 25 April 2016
* Indicates chiral carbon atoms.
These problems of endogenous interference and lack of specifity can be successfully overcome by isolating the components for analysis on a robust, achiral h.p.1.c. system and transferring only the compounds of interest to the chiral system. Because of the fairly restricted mobile-phase choice of many chiral stationary phases, column-switching techniques can be difficult to apply, and require the achiral stage to be developed with this mind (Walhagen and Edholm 1989). With many compounds for which a chiral separation is under investigation, a routine, wellvalidated and specific achiral h.p.1.c. method already exists for analysis of total drug in biological samples. Post-detector fraction collection of the achiral h.p.1.c. eluent can easily be performed, and samples reconcentrated for chiral chromatography as previously described for chloroquine and desethylchloroquine (Ofori-,4djei et al. 1986). ( - )-erythro-(E)-3,5-Dihydroxy-7-[4,5-bis(4-fluorophenyl)-2-( 1-methylethyl)1H-imidazol-l-yl]-6-heptenoic-acid(figure 1, I, GR92.549) is a potent, orally active, cholesterol lowering agent designed as a competitive inhibitor of the microsomal enzyme 3-hydroxy-3-methylglutaryl-coenzymeA (HMG-CoA) reductase (Nakaya et al. 1986). T h e compound entered early development as a racemic mixture of the 3R,5S ( + ) and 3S,5R ( - ) forms. However, the HMG-CoA reductase inhibitor activity was subsequently shown to reside only in the 3R,5S enantiomer and a method for the quantification of the individual isomers in biological samples was therefore required in order to confirm exposure of test animals to the active enantiomer. Fluparoxan, (trans)-( -)-5-fluoro-2,3,3a,9a-tetrahydro-lH-[l,4]-benzodioxino-[2,3-c] pyrrole (figure 1,II) is an a,-adrenoceptor antagonist (Halliday et al. 1988) for which the two enantiomers have shown similar in vivo potencies. However, as one isomer may have shown a superior pharmacokinetic profile, measurement of individual isomers, during testing of the racemate, was considered important. This paper describes the application of achiral fraction collecting to the quantification of the enantiomers of the HMG-CoA reductase inhibitor (I) and the a,-antagonist, fluparoxan (11), in samples from drug absorption studies in animals. While this approach initially appeared to be an indirect solution to the analytical problems, in practice the method showed a number of advantages which should be of benefit to any chiral analysis in biological samples, and these are illustrated and discussed.
+
+
Achiral H.P.L.C. in toxicokinetic studies
791
Methods
Downloaded by [RMIT University Library] at 10:31 25 April 2016
Reagents The racemic HMG-CoA reductase inhibitor (figure 1, I, GR92549) and its (+)-enantiomer, racemic fluparoxan (figure 1, 11, GR50360) and its separate enantiomers, and the internal standard for the fluparoxan achiral assay (des-fluoro-N-2-propenyl 11, GR35996) were synthesized at Glaxo Group Research, Greenford, Middlesex. Solid-phase extraction cartridges (C2 Bond Elut, 1 ml and Bond Elut Certify, 3 ml, Analytichem International Harbor City, CA, USA) were obtained from Jones Chromatography, Hengoed, Wales. Acetonitrile (h.p.1.c. grade) was supplied by Rathburn Chemicals Ltd, Walkerburn, Scotland; diethyl ether (Pronalys AR) by May and Baker, Dagenham, Essex; orthophosphoric acid (HiPerSolv) by BDH Ltd, Poole, Dorset, and all other solvents and reagents (Analar grade) were from FSA Laboratory Supplies, Loughborough, Leicestershire. Animal studies Blood samples were taken from male Beagle dogs during toxicological studies on the HMG-CoA reductase inhibitor, I, and in an experiment designed to characterize the oral absorption of the enantiomers. Animals were given single daily doses (1,7 or 50 mg/kg) of the racemic drug as a powder, without excipient, in hard gelatin capsules. In a study to compare the concentrations of I in rat plasma and liver tissue (the site of pharmacological action), male and female random-bred hooded rats were given single oral doses (5 mg/kg) of the racemic d r ~ g ( ' ~ C - l a b e l l at e dthe 4-position) in solution. T h e compound was dissolved in 1 M-NaOH and adjusted to pH7.5 with 1 M-HC1and administered by gavage (lOml/kg). Blood samples were obtained by vena caval exsanguination and the livers excised from each animal (three per time-point) and immediately , frozen over solid-COJacetone. Oral absorption studies on fluparoxan were carried out in rats and rabbits at dose levels used in toxicological studies. Doses were prepared as solutions of the racemic drug in distilled water and administered by gavage. A single dose (5 mg/kg) was administered (4 ml/kg) to each of four female Dutch rabbits and serial blood samples collected from the lateral ear veins, directly into heparinized microfuge tubes. Single daily doses (50mg/kg) were administered (5ml/kg) to two groups of male, random-bred hooded rats. Blood samples were obtained (as above) from one group (three animals per time-point) after the first dose and from the second group after eight doses. Blood was transferred to heparinized tubes and centrifuged to yield plasma as soon as practicable after collection, and this was stored at about -20°C whilst awaiting analysis. Hooded rats and Beagle dogs were supplied by Glaxo Group Research Animal Breeding Unit. Dutch rabbits were obtained from J . and G. Phillips, Froxfield, Hampshire. Drug analyses HMG-CoA reductase inhibitor ( I ) . C2 Bond Elut cartridges were primed with methanol (1 ml) and distilled water (1 ml), and plasma samples (1 ml) were centrifuged to remove particulate material. Plasma (0.9 ml) was drawn through the cartridges under vacuum and a water wash (2 ml) was followed by elution of the drug with 0.5 ml of acetonitril4.1 M ammonium acetate (70 : 30 v/v). T h e eluents were dried at ) achiral h.p.1.c. 50°C under a stream of nitrogen and the residues dissolved in a small volume ( 1 0 0 ~ 1of ) injected onto an mobile-phase (acetonitrile-0.1 M ammonium acetate, 28 :72 v/v). Aliquots ( 6 0 ~ 1were ODS-Hypersil column (100mm x 4.6 mm int. diam. with 20mm guard column, Capital H P L C Specialists, Bathgate, Scotland) and run for 12 min at a flow rate of 1 ml/min. T h e drug was detected in the column eluent by U.V.absorption at 265 nm for dog samples and 285 nm for rat (wavelength determined by differing pattern of co-eluting endogenous components). Samples of liver (2 g) were homogenized (Potter-Elvehjem) in distilled water (10ml). Aliquots (0.75 ml) of homogenate were mixed with acetonitrile (0.75 ml), centrifuged and the supernatants mixed with distilled water (5 ml). T h e resulting samples were analysed for I, as in plasma. T o reduce the number of livers processed, the tissue having the median level of radioactivity at each time-point was assayed. Single fractions of h.p.1.c. eluent were collected, post-detector, from each chromatogram of the achiral analysis, using a Gilson 201 fraction-collector (Anachem, Luton, Bedfordshire). Collection was initiated 0.5 min prior to the start of the peak for I and continued for a total of 3.0min (figure 2A). T h e fractions were mixed with an equal volume of distilled water and reconcentrated, using a Gilson ASPEC system (Anachem), on C2 Bond-Elut cartridges as in the plasmaextraction step, with methanol (0.5 ml) as elution solvent. Following evaporation of the solvent, the residues were dissolved in a small volume ( 1 3 0 ~ 1of ) mobile phase for chiral analysis (2% propan-2-01 in 0.02 M ammonium acetate). Separation of enantiomers was achieved using an a,-acid glycoprotein column (100 mm Chiral-AGP supplied by Technicol, Stockport, Lancashire) at 35°C. with a mobile phase flow rate of 0.8 ml/min (injection volume 1 0 0 ~ 1U.V. , detection at 285 nm). Fluparoxan (IZ). Internal standard (GR35996) was added to plasma samples (1 ml) and the compounds extracted (rotary mixing, approx. 30 rpm, 5 min) into diethyl ether (5 ml). After centrifugation to separate the layers, the ether was transferred to clean tubes and evaporated to dryness at 45°C. T h e residues were redissolved in a small volume ( 1 0 0 ~ 1of ) the mobile phase for achiral h.p.1.c. (water-1.0~ammonium
A. P. Beresford et al.
792
I
(ii)
B(i)
”
--
0
4
8
1 2 0
4
8
12
r
0
5
10
0
;
10
Downloaded by [RMIT University Library] at 10:31 25 April 2016
min
Figure 2.
Achiral h.p.1.c. chromatograms of the HMG-CoA reductase inhibitor, I, and of fluparoxan (11). A:Analysis of rac-I in dog plasma (1 ml), (i) blank plasma (ii) plasma with added rac-I (250 ngiml). B: Analysis of rac-fluparoxan in dog plasma (1 ml) blank plasma, (ii) plasma with added racfluparoxan (0.5 pg/ml) and internal standard, GR35996. __ indicates fraction collected for subsequent chiral analysis.
dihydrogen orthophosphateacetonitrilephosphoric acid, 62.48 : 20 : 17.5 :0.02, by vol.) and aliquots (85 PI) injected onto a 5 pm ODS 11-Spherisorb column (100mm x 4.6mm int. diam. with 20mm guard column, Capital HPLC Specialists). The drug was detected in the column eluent by U.V.absorption (220nm). Collection of achiral h.p.1.c. fractions (figure 2B) was carried out using the Gilson 201, as for the HMG-CoA reductase inhibitor (I)and reconcentration performed using the Gilson ASPEC system. T h e untreated fractions were applied to Bond Elut Certify cartridges, washed with methanol (2ml) and fluparoxan recovered using 1 ml of conc. ammonia s o h (sp. gr. 0,88)-methanol (1 : 99vjv). After drying under nitrogen the residues were redissolved in a small volume (120~1)of mobile phase for chiral h.p.1.c. (0.25 M-phosphate buffer, p H 6.2). Separation of enantiomers was achieved using a bovine serum albumin column (Macherey-Nagel 150mm x 4mm, Resolvosil-7, supplied by Technicol) at 3 5 T , with a mobile , detection at 220nm). phase flow rate of 0.8ml/min (injection volume 1 0 0 ~ 1U.V.
Quantification For achiral analyses, calibration curves were prepared by weighted regression of the peak height or peak height ratio, against the drug concentration for seven standards, using a reciprocal weighting factor (concentration’). Analysis of the HMG-CoA reductase inhibitor was performed using a calibration range from 5 to 2500 ng/mI of plasma and 25 to 2000 ng/ml of liver homogenate. For fluparoxan the range was 0.1 25 to 8.00 pg/ml of plasma. For routine quality control of achiral assays, duplicate standards, at three concentrations within the calibration range and prepared from a separately weighed stock of compound, were processed with each batch of samples. Assay performance was considered acceptable provided that four of these, with one at each concentration, were measured to within 15% of expected values. Following chiral h.p.l.c, enantiomeric ratios were determined from peak area measurements and the concentration of each enantiomer was derived from the ratio and the equivalent achiral result. It was considered that performance of the chiral analyses was satisfactory provided that ratios for four of the quality control samples, with one at each concentration, were within 5% of the racemic value of 1.00 for the HMG-CoA reductase inhibitor and within 10% for fluparoxan. Pharmacokinetics Pharmacokinetic analyses of data were performed using the program SIPHAR (V3.3, Simed, Creteil, France) running on Vax/VMS. Areas under plasma concentration/time curves were determined using the linear trapezoidal rule.
Results Recovery of compound during achiral analyses was > 75% for the HMG-CoA reductase inhibitor, I, over the calibration range used, and > 70% for fluparoxan. For both compounds the limit of quantification was taken as the concentration of the lowest standard.
793
Achiral H.P.L.C. in toxicokinetic studies A (i)
B (i)
(it)
-7
Downloaded by [RMIT University Library] at 10:31 25 April 2016
0
4
8
120
4
-
8
1 1
12
0
4
8
12.0
4
8
12
rnin min Figure 3 . Chiral h.p.1.c. chromatograms showing separation of enantiomers for the HMG-CoA reductase inhibitor, I, in fractions collected during achiral analysis of dog plasma and rat liver tissue samples. A: Dog plasma (1 ml), (i) blank, (ii) with added mc-I (250ng/ml).B: Rat liver (125 mg), (i) blank, (ii) with added rac-I (1600nglg).
A typical calibration plot of peak height for I against concentration gave a linear plot with r2 = 0-998. Replicate analyses (n= 6) of plasma samples containing drug, added at each of the seven concentrations used for calibration, gave mean values within 7% of the nominal concentration, with a maximum coefficient of variation of 12.3% at 5 ng/ml. Calibration plots of peak height ratio against concentration for fluparoxan were linear, with ~ ' 2 0 . 9 9 8 . Replicate analyses ( n = 6 ) at each of the seven calibration concentration gave mean values within 2% of the nominal concentration, with a maximum coefficient of variation of 8.3% at 0.25 pg/ml. Typical achiral chromatograms for extracts of control dog plasma with added rac-I (250 mg/ml) or rac-fluparoxan (0.5 pg/ml) are shown in figure 2. Sample analysis for the HMG-CoA reductase inhibitor, I Figure 2A shows typical achiral chromatograms for extracts of control dog plasma and plasma with added rac-I (250ng/ml). T h e resulting chiral chromatograms from these samples are shown in figure 3A. T h e cleanness of the collected fractions allowed multiple re-use of the extraction cartridges for concentration prior to chiral analysis, and the method was found to be equally applicable to plasma and liver samples from rats (figure 3B). This was in marked contrast to earlier results obtained without the achiral h.p.1.c. stage (extraction of drug into methyl-tert-butyl ether by mixing with solvent and saturated ammonium sulphate soln), where a potentially useful dog plasma assay did not translate to rat plasma (figure 4A). T h e achiral stage of the method was considered to be specific and no significant interfering peaks were observed during routine achiral or chiral analysis of experimental samples, including samples from toxicology studies. Again this contrasted with the earlier solvent extraction method (figure 4B) where the monitoring of enantiomeric ratios during a dog toxicity trial was impaired by an increase in co-extracting endogenous materials or drug metabolites. Guard columns of cr,-acid glycoprotein (AGP) were not available at the time of this work, but more than 300 samples were analysed on a single chiral-AGP column
A. P . Beresford et al.
794
-- -5
10
150
5
10
15
0
10
5
Downloaded by [RMIT University Library] at 10:31 25 April 2016
min
150
5
10
15
min
Figure 4. Chiral h.p.1.c. chromatograms showing attempts to analyse enantiomers for the HMG-CoA reductase inhibitor, I, directly on a chiral-AGP column following solvent extraction from dog and rat plasma. Extraction was with methyl tert-butyl ether and satd ammonium sulphate. A: Blank plasma (1 ml), (i) dog, (ii) rat. B:Plasma (1 ml) obtained 1 h post-dose from a dog following single daily oral doses of rac-I, (i) day 1 of dosing, (ii) day 26 of dosing.
with little deterioration of its performance. A slow increase in pressure, with a concomitant decrease in resolution of the enantiomers, was seen with time, and was compensated by decreasing the flow rate to 0 6 5 ml/min. Despite the absence of an internal standard in this method, the combination of solid-phase extraction and automated fraction collection was sufficiently reproducible to give satisfactory calibration curves for each enantiomer. Whilst this duplication of calibration data was unnecessary in normal use, it provided increased confidence in the specificity of both the chiral and achiral assays. Comparing results obtained from achiral analyses with those obtained by adding the separate (+) and ( - ) enantiomer data, showed the majority of results to be within 10% of each other. I n practice the chiral analysis of plasma I was used over the range 25-2500ng total drug/ml, compared to 5-2500 ng/ml for the achiral assay, the limitation being the broader peaks in the chiral chromatograms. Below 12*5ng/ml for each enantiomer a ratio of 1.00 for a racemic mixture could not be reliably (within 5%) obtained. When analysis was performed on samples of dog plasma prepared at the three quality control levels, with an increasing enantiomeric excess of the (+)enantiomer, all but one enantiomeric ratio was found within 5% of the expected Table 1. Areas under the plasma concentration/time curves for enantiomers of the HMG-CoA reductase inhibitor, I, in six male beagles, following single oral doses of the racemic I. Area under curve (ng/ml*p hour) by trapezoidal rule (AUC)
AUC ratio
Dose (mgikg)
Dog no.
1
9GR5 9GS4
7
9GR6 9HB4
633 481.
847 616
0.75 0.78
50
9GR7 9GS6
18900 11500
29400 22800
0.64 0.50
(+)-enantiomer
49.1 55.1
(-)-enantiomer
70.6 74.6
(+)/( -)
0.70 0.74
795
Achiral H.P.L.C. in toxicokinetic studies
-
10000,
L
al
> -
.-
. 0,
c 0,
m
1000-
5
m -
Q
-
E .
-
0,
100-
0 .-
+a
F
c
Downloaded by [RMIT University Library] at 10:31 25 April 2016
C V al
0
u
10
-
I
0
I
I
1
I
I
2
I
1
3
dosage (hi Figure 5. Concentrations of enantiomers of the HMG-CoA reductase inhibitor, I , in plasma and liver tissue from rats following single oral doses of the racemic drug. Dose of roc-I was 5mg/kg. 3R,SS(+) in liver (0); 3S,5R(-) in liver (m). 3 R , 5 S ( t ) in plasma (0); 3S,SR(-) in plasma ( 0 ) . Tlrne after
value. However, this experiment suggested that a change of at least 10% in the enantiomeric ratio would be required for reliable detection by the method over the specified range. Application of the method to samples from both dog and rat studies showed that exposure to the active, ( )-enantiomer was consistently lower (approximately 25%) than that to the (-)-enantiomer. This is illustrated by the dog plasma AUC data in table 1 and the rat plasma and liver concentrations shown in figure 5.
+
Analysis of fluparoxan samples T h e chiral method for fluparoxan has now been applied to plasma from rat, rabbit, dog and man. I n man the chiral analysis described was only suitable for measuring the enantiomers over the first few hours after an oral dose, as concentrations rapidly fell below the limit for reliable enantiomeric measurement (Gristwood 1990). However, the procedure has been used successfully to monitor the time-course of each isomer in plasma samples from the animal studies. Figure 6 compares chiral chromatograms of samples from an oral dose rat study with those of rac-fluparoxan standards. Due to the nature of the solid-phase extraction cartridge material used in the reconcentration of the achiral fractions, the internal standard (GR35996), having a 2-propenyl group on the pyrrole nitrogen, is not recovered at this stage. Hence, extending the fraction collection time would not endanger the integrity of the chiral chromatogram. This additional specificity at the recovery step further decreases the potential for co-elution on the chiral stationary phase. As with the chiral-AGP column, a slow decrease in resolution of enantiomers on the bovin serum albumin column was seen with continued use, and was compensated by decreasing the flow rate.
796 ill
-
Downloaded by [RMIT University Library] at 10:31 25 April 2016
1
1
0
10
20
0
10
20 0
20
10
10
0
20
Figure 6 . Chiral h.p.1.c. chromatograms showing separation of fluparoxan enantiomers injected oncolumn and from fractions collected during achiral analysis of plasma samples from rats following single daily oral doses of the racemic drug.
Dose of roc-fluparoxan was 50mg/kg: (i) 0.5 p g on-column, (ii) plasma (1 ml) with added rucfluparoxan (0.5pg/ml), (iii) pre-dose plasma (1 ml), (iv) post-dose plasma (1 ml).
The increase in limit of quantification due to the width of the enantiomer peaks seen for the HMG-CoA reductase inhibitor, I, was also apparent for fluparoxan analysis. Here a limit of 0.5pg/ml was set for the chiral assay, compared to 0 1 25 pg/ml for the achiral. This limit was the lowest concentration of rac-fluparoxan at which a 1 : 1 peak area ratio could be routinely obtained within 10% (compared with 5% for I). This translated to a limit of 0.25 pg/ml for either enantiomer, as a reliable ratio could not be obtained for a total fluparoxan concentration of 0.5 pg/ml comprised 0*4pgof (+)-isomer and 0.1 pg (-)-isomer in a sample. 0.1 pg (-)-isomer in a sample. T h e results from the single-dose rabbit study indicated that the ratio of circulating enantiomers was not quite racemic, with, on average, an approximately 15% excess of the (+)-isomer. There was no clear indication of a change in the enantiomeric ratio with time after dosing, and this was also true in the rat study, as illustrated for both day 1 and day 8 in figure 7.
0
Figure 7.
5 10 15 20 Time after dosage ( h )
25
if
,
I 0
5
I
I
10 15 20 Time after dosage (h)
I
25
Concentrations of fluparoxan enantiomers in plasma from rats following repeat single daily oral doses of the racemic drug. Daily dose of roc-fluparoxan was 50mg/kg: (+)-fluparoxan (0);(-)-fluparoxan (m): A: Day 1 of dosing. 8: Day 8 of dosing.
Achiral H.P.L.C. in toxicokinetic studies
797
Downloaded by [RMIT University Library] at 10:31 25 April 2016
Discussion During the early development of drugs containing one or more asymmetric centres it may be necessary to quantify individual isomers in a variety of biological samples generated during pharmacological/toxicological assessment of the compound. In such cases the use of an achiral h.p.1.c. step for purification of the compounds, prior to chiral separation, is recommended. This approach is equally applicable to methods using direct separation on chiral stationary phases or where diastereomers are prepared. T h e variety of achiral h.p.1.c. conditions available permit development of specific methods for quantification of total drug. Post-detector fraction collection can provide clean samples for the more demanding enantioselective analyses whilst circumventing potential problems of solvent incompatibility. As previously mentioned by Edholm et al. (1988) quantification of racemates during achiral h.p.l.c, with only enantiomeric ratios being measured at the chiral stage, precludes the need for internal standards in the latter analysis. However, quantification of both achiral and chiral chromatograms can help confirm assay specificities, and should permit the calculation of enantiomeric ratios in situations where only one isomer is reliably quantifiable at the chiral stage. This approach has been successfully applied in our laboratories to a number of compounds under pharmaceutical development. In each case different chiral stationary phases were required for the enantiomeric sepration, but achiral h.p.1.c. methods for total drug measurement were already available. As the results presented here show, the technique increases the likelihood that methods will be applicable to samples from different biological sources. Here the method was used to confirm that animals dosed during the toxicological testing of an HMG-CoA reductase inhibitor, I, were exposed to the active (+)enantiomer following administration of the racemate. The results also confirmed that the active enantiomer reached the site of pharmacological action, the liver, rapidly and at concentrations (on a ng/g basis) up to 30 times that of plasma. In the case of the a,-adrenoceptor antagonist, the approach was used to confirm that both enantiomers were absorbed from oral doses during toxicokinetic studies in animals, and showed that there was little difference in their plasma profiles.
Acknowledgements T h e authors are indebted to their colleagues M r I. M. Mutton (Chemical Analysis Dept) for advice on chiral stationary phase selection and Miss P. A. Williams for her help in preparation of the manuscript.
References ALBANI,F., RIVA,R., CONTIN,M., and BARUZZI, A , , 1984, Stereoselective binding of propanolol enantiomers to human al-acid glycoprotein and human plasma. British Journal of Clinical Pharmacology, 18, 244-246. ALLENMARK, S., 1986, Optical resolution by liquid chromatography on immobolised bovine serum albumin. Journal of Liquid Chromatography, 9, 425-442. ALLENMARK, S., 1989, Protein-bonded stationary phases, in Chiral Separations by h.p.1.c.:Applications to Pharmaceutical Compounds, edited by A. M. Krstulovic (Chichester, UK: Ellis Horwood), pp. 286-315.
BROWN, N. A. JAHNCHEN, E., MULLER, W. E., and WOLLERT, U., 1977, Optical studies on the mechanism of the interactionof the enantiomers of the anticoagulant drugs phenprocoumon and warfarin with human serum albumin. Molecular Pharmacology, 13, 70-79.
Downloaded by [RMIT University Library] at 10:31 25 April 2016
798
Achiral H.P.L.C. in toxicokinetic studies
CHU,Y.-Q,, and WAINER, I. W., 1988, Measurement of warfarin enantiomers in serum using coupled achiral/chiral high performance liquid chromatography (h.p.1.c.). Pharmaceutical Research, 5 (lo), 680-683. J., and WALHAGEN, A,, 1988, Determination of drug EDHOLM,L.-E., LINDBERG, C., PAULSON, enantiomers in biological samples by coupled column liquid chromatography and liquid chromatography-mass spectrometry. Journal of Chromatography, 424, 61-72. M., and HERMANSSON, J., 1989, Separation and quantitation of (R)- and (S)-atenolol in human ENQUIST, plasma and urine using an al-acid glycoprotein column. Chirality, 1 , 209-215. GRISTWOOD, W. E., 1990, Determination of fluparoxan (GR50360) in plasma by gas chromatography. Journal of Chromatography, 527, 436440. D. W., TYERS,M . B., WALSH,D. M., and WISE,H., 1988, HALLIDAY, C. A,, JONES,B. J . , STRAUGHAN, GR50360, a novel and highly selective a2-adrenoceptor antagonist. British Journal of Pharmacology, 95, 751P. W. E., and WOLLERT, U., 1975, High stereospecificity of the benzodiazipine binding site on MULLER, human serum albumin. Molecular Pharmacology, 1 1 , 52-60. NAKAYA, N., HOMMA, Y.,TAMACHI, H., and GOTO,Y.,The effect of CS-514, an inhibitor of HMG-CoA reductase, on serum lipids in healthy volunteers. Atherosclerosis, 61, 125-126. O., LINDSTROM, B., HERMANSSON, J., ADJEPON-YAMOAH, K., and SJOQVIST, OFORI-ADJEI, D., ERICSSON, F., 1986, Enantioselective analysis of chloroquine and desethylchloroquine after oral administration of racemic chloroquine. Therapeutic Drug Monitoring, 8, 457461. SCHILL, G., WAINER,I. W., and BARKAN, S. A,, 1986, Chiral separation of cationic drugs on an al-acid glycoprotein bonded stationary phase. Journal of Liquid Chromatography, 9 (2 & 3), 641-666. WALHAGEN, A,, and EDHOLM, L.-E., 1989, Coupled-column chromatography on immobilized protein phases for direct separation and determination of drug enantiomers in plasma. Journal of Chromatography, 473, 371-379.
ISSN: 0049-8254 (Print) 1366-5928 (Online) Journal homepage: http://www.tandfonline.com/loi/ixen20
Advantages of achiral h.p.l.c. as a preparative step for chiral analysis in biological samples and its use in toxicokinetic studies A. P. Beresford, K. Caswell, R. Chambers & I. P. Kirk To cite this article: A. P. Beresford, K. Caswell, R. Chambers & I. P. Kirk (1992) Advantages of achiral h.p.l.c. as a preparative step for chiral analysis in biological samples and its use in toxicokinetic studies, Xenobiotica, 22:7, 789-798, DOI: 10.3109/00498259209053141 To link to this article: http://dx.doi.org/10.3109/00498259209053141
Published online: 22 Sep 2008.
Submit your article to this journal
Article views: 5
View related articles
Citing articles: 1 View citing articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ixen20 Download by: [RMIT University Library]
Date: 25 April 2016, At: 10:31
XENOBIOTICA,
1992, VOL. 22,
NO.
7, 789-798
Advantages of achiral h.p.1.c. as a preparative step for chiral analysis in biological samples and its use in toxicokinetic studies A. P. BERESFORD, K. CASWELL, R. CHAMBERS and I. P. K I R K Drug Metabolism Department, Glaxo Group Research Limited, Greenford Road, Greenford, Middlesex UB6 OHE, UK
Downloaded by [RMIT University Library] at 10:31 25 April 2016
Received I2 June 1991; accepted 21 February 1992 1. Achiral reverse-phase h.p.1.c. with semi-automated post-column fraction collection and solid-phase sample reconcentration, has been applied as the purification procedure during the enantiomeric quantification of two widely differing experimental drugs; an HMG-CoA reductase inhibitor (I) and an a,-adrenoceptor antagonist (11). 2. The robust and specific achiral methodologies were available prior to the need for chiral analyses and recovery of drug from the fractions provided clean samples from a variety of biological matrices, without the need to develop compatible achiral/chiral mobile phases. 3. Compared with direct chiral chromatography of plasma extracts, this approach decreased the potential for metabolites and endogenous components to interfere or impair the performance of the chiral stationary phase.
4. The availability of quantitative data from achiral analysis of samples negated the need for internal standardization of the chiral analyses, helped confirm assay specificity and provided potential to determine enantiomeric ratios where only one isomer could be accurately measured.
5. Routine enantiomeric analyses were successfully carried out on samples taken from animals dosed orally with the racemic drugs, providing important data on the possible levels of exposure to individual enantiomers during toxicity testing.
Introduction Many drug substances under investigation contain one or more asymmetric centres, and it is frequently necessary to obtain pharmacokinetic data on individual isomers at an early stage of development. T h e outcome of such analyses may influence further progression of the compound as a single isomer or as a mixture. Samples may be supplied from early toxicology studies and in a variety of biological matrices, and at this stage rapid assay development could be advantageous. T h e enantioselective binding properties of plasma proteins have been well documented over the years (Muller and Wollert 1975, Brown et al. 1977, Albani et al. 1984) and the use of a,-acid glycoprotein and serum albumin in h.p.1.c. stationary phases offers a wide range of enantiomeric separation capabilities to the analyst (Schill et al. 1986, Allenmark 1986). Although the limited capacity of such columns for drug molecules (Allenmark 1989) may not be a problem in analytical methods, the detrimental effects on column performance of co-injected biological residues certainly can be (Enquist and Hermansson 1989). Consequently some chiral separations require extensive sample preparation and others may never become applicable to biological samples. A further problem during analysis of in viwo samples on chiral stationary phases is the unpredictable elution characteristics, not only of endogenous materials, but also of metabolites of the compound of interest (Chu and Wainer 1988). 0049-8254/92 $3.00 0 1992 Taylor & Francis Ltd.
A. P . Beresford et al.
790
II
I
Figure 1 .
Structures of the HMG-CoA reductase inhibitor, I, and the a,-adrenoceptor antagonist, fluparoxan (11).
Downloaded by [RMIT University Library] at 10:31 25 April 2016
* Indicates chiral carbon atoms.
These problems of endogenous interference and lack of specifity can be successfully overcome by isolating the components for analysis on a robust, achiral h.p.1.c. system and transferring only the compounds of interest to the chiral system. Because of the fairly restricted mobile-phase choice of many chiral stationary phases, column-switching techniques can be difficult to apply, and require the achiral stage to be developed with this mind (Walhagen and Edholm 1989). With many compounds for which a chiral separation is under investigation, a routine, wellvalidated and specific achiral h.p.1.c. method already exists for analysis of total drug in biological samples. Post-detector fraction collection of the achiral h.p.1.c. eluent can easily be performed, and samples reconcentrated for chiral chromatography as previously described for chloroquine and desethylchloroquine (Ofori-,4djei et al. 1986). ( - )-erythro-(E)-3,5-Dihydroxy-7-[4,5-bis(4-fluorophenyl)-2-( 1-methylethyl)1H-imidazol-l-yl]-6-heptenoic-acid(figure 1, I, GR92.549) is a potent, orally active, cholesterol lowering agent designed as a competitive inhibitor of the microsomal enzyme 3-hydroxy-3-methylglutaryl-coenzymeA (HMG-CoA) reductase (Nakaya et al. 1986). T h e compound entered early development as a racemic mixture of the 3R,5S ( + ) and 3S,5R ( - ) forms. However, the HMG-CoA reductase inhibitor activity was subsequently shown to reside only in the 3R,5S enantiomer and a method for the quantification of the individual isomers in biological samples was therefore required in order to confirm exposure of test animals to the active enantiomer. Fluparoxan, (trans)-( -)-5-fluoro-2,3,3a,9a-tetrahydro-lH-[l,4]-benzodioxino-[2,3-c] pyrrole (figure 1,II) is an a,-adrenoceptor antagonist (Halliday et al. 1988) for which the two enantiomers have shown similar in vivo potencies. However, as one isomer may have shown a superior pharmacokinetic profile, measurement of individual isomers, during testing of the racemate, was considered important. This paper describes the application of achiral fraction collecting to the quantification of the enantiomers of the HMG-CoA reductase inhibitor (I) and the a,-antagonist, fluparoxan (11), in samples from drug absorption studies in animals. While this approach initially appeared to be an indirect solution to the analytical problems, in practice the method showed a number of advantages which should be of benefit to any chiral analysis in biological samples, and these are illustrated and discussed.
+
+
Achiral H.P.L.C. in toxicokinetic studies
791
Methods
Downloaded by [RMIT University Library] at 10:31 25 April 2016
Reagents The racemic HMG-CoA reductase inhibitor (figure 1, I, GR92549) and its (+)-enantiomer, racemic fluparoxan (figure 1, 11, GR50360) and its separate enantiomers, and the internal standard for the fluparoxan achiral assay (des-fluoro-N-2-propenyl 11, GR35996) were synthesized at Glaxo Group Research, Greenford, Middlesex. Solid-phase extraction cartridges (C2 Bond Elut, 1 ml and Bond Elut Certify, 3 ml, Analytichem International Harbor City, CA, USA) were obtained from Jones Chromatography, Hengoed, Wales. Acetonitrile (h.p.1.c. grade) was supplied by Rathburn Chemicals Ltd, Walkerburn, Scotland; diethyl ether (Pronalys AR) by May and Baker, Dagenham, Essex; orthophosphoric acid (HiPerSolv) by BDH Ltd, Poole, Dorset, and all other solvents and reagents (Analar grade) were from FSA Laboratory Supplies, Loughborough, Leicestershire. Animal studies Blood samples were taken from male Beagle dogs during toxicological studies on the HMG-CoA reductase inhibitor, I, and in an experiment designed to characterize the oral absorption of the enantiomers. Animals were given single daily doses (1,7 or 50 mg/kg) of the racemic drug as a powder, without excipient, in hard gelatin capsules. In a study to compare the concentrations of I in rat plasma and liver tissue (the site of pharmacological action), male and female random-bred hooded rats were given single oral doses (5 mg/kg) of the racemic d r ~ g ( ' ~ C - l a b e l l at e dthe 4-position) in solution. T h e compound was dissolved in 1 M-NaOH and adjusted to pH7.5 with 1 M-HC1and administered by gavage (lOml/kg). Blood samples were obtained by vena caval exsanguination and the livers excised from each animal (three per time-point) and immediately , frozen over solid-COJacetone. Oral absorption studies on fluparoxan were carried out in rats and rabbits at dose levels used in toxicological studies. Doses were prepared as solutions of the racemic drug in distilled water and administered by gavage. A single dose (5 mg/kg) was administered (4 ml/kg) to each of four female Dutch rabbits and serial blood samples collected from the lateral ear veins, directly into heparinized microfuge tubes. Single daily doses (50mg/kg) were administered (5ml/kg) to two groups of male, random-bred hooded rats. Blood samples were obtained (as above) from one group (three animals per time-point) after the first dose and from the second group after eight doses. Blood was transferred to heparinized tubes and centrifuged to yield plasma as soon as practicable after collection, and this was stored at about -20°C whilst awaiting analysis. Hooded rats and Beagle dogs were supplied by Glaxo Group Research Animal Breeding Unit. Dutch rabbits were obtained from J . and G. Phillips, Froxfield, Hampshire. Drug analyses HMG-CoA reductase inhibitor ( I ) . C2 Bond Elut cartridges were primed with methanol (1 ml) and distilled water (1 ml), and plasma samples (1 ml) were centrifuged to remove particulate material. Plasma (0.9 ml) was drawn through the cartridges under vacuum and a water wash (2 ml) was followed by elution of the drug with 0.5 ml of acetonitril4.1 M ammonium acetate (70 : 30 v/v). T h e eluents were dried at ) achiral h.p.1.c. 50°C under a stream of nitrogen and the residues dissolved in a small volume ( 1 0 0 ~ 1of ) injected onto an mobile-phase (acetonitrile-0.1 M ammonium acetate, 28 :72 v/v). Aliquots ( 6 0 ~ 1were ODS-Hypersil column (100mm x 4.6 mm int. diam. with 20mm guard column, Capital H P L C Specialists, Bathgate, Scotland) and run for 12 min at a flow rate of 1 ml/min. T h e drug was detected in the column eluent by U.V.absorption at 265 nm for dog samples and 285 nm for rat (wavelength determined by differing pattern of co-eluting endogenous components). Samples of liver (2 g) were homogenized (Potter-Elvehjem) in distilled water (10ml). Aliquots (0.75 ml) of homogenate were mixed with acetonitrile (0.75 ml), centrifuged and the supernatants mixed with distilled water (5 ml). T h e resulting samples were analysed for I, as in plasma. T o reduce the number of livers processed, the tissue having the median level of radioactivity at each time-point was assayed. Single fractions of h.p.1.c. eluent were collected, post-detector, from each chromatogram of the achiral analysis, using a Gilson 201 fraction-collector (Anachem, Luton, Bedfordshire). Collection was initiated 0.5 min prior to the start of the peak for I and continued for a total of 3.0min (figure 2A). T h e fractions were mixed with an equal volume of distilled water and reconcentrated, using a Gilson ASPEC system (Anachem), on C2 Bond-Elut cartridges as in the plasmaextraction step, with methanol (0.5 ml) as elution solvent. Following evaporation of the solvent, the residues were dissolved in a small volume ( 1 3 0 ~ 1of ) mobile phase for chiral analysis (2% propan-2-01 in 0.02 M ammonium acetate). Separation of enantiomers was achieved using an a,-acid glycoprotein column (100 mm Chiral-AGP supplied by Technicol, Stockport, Lancashire) at 35°C. with a mobile phase flow rate of 0.8 ml/min (injection volume 1 0 0 ~ 1U.V. , detection at 285 nm). Fluparoxan (IZ). Internal standard (GR35996) was added to plasma samples (1 ml) and the compounds extracted (rotary mixing, approx. 30 rpm, 5 min) into diethyl ether (5 ml). After centrifugation to separate the layers, the ether was transferred to clean tubes and evaporated to dryness at 45°C. T h e residues were redissolved in a small volume ( 1 0 0 ~ 1of ) the mobile phase for achiral h.p.1.c. (water-1.0~ammonium
A. P. Beresford et al.
792
I
(ii)
B(i)
”
--
0
4
8
1 2 0
4
8
12
r
0
5
10
0
;
10
Downloaded by [RMIT University Library] at 10:31 25 April 2016
min
Figure 2.
Achiral h.p.1.c. chromatograms of the HMG-CoA reductase inhibitor, I, and of fluparoxan (11). A:Analysis of rac-I in dog plasma (1 ml), (i) blank plasma (ii) plasma with added rac-I (250 ngiml). B: Analysis of rac-fluparoxan in dog plasma (1 ml) blank plasma, (ii) plasma with added racfluparoxan (0.5 pg/ml) and internal standard, GR35996. __ indicates fraction collected for subsequent chiral analysis.
dihydrogen orthophosphateacetonitrilephosphoric acid, 62.48 : 20 : 17.5 :0.02, by vol.) and aliquots (85 PI) injected onto a 5 pm ODS 11-Spherisorb column (100mm x 4.6mm int. diam. with 20mm guard column, Capital HPLC Specialists). The drug was detected in the column eluent by U.V.absorption (220nm). Collection of achiral h.p.1.c. fractions (figure 2B) was carried out using the Gilson 201, as for the HMG-CoA reductase inhibitor (I)and reconcentration performed using the Gilson ASPEC system. T h e untreated fractions were applied to Bond Elut Certify cartridges, washed with methanol (2ml) and fluparoxan recovered using 1 ml of conc. ammonia s o h (sp. gr. 0,88)-methanol (1 : 99vjv). After drying under nitrogen the residues were redissolved in a small volume (120~1)of mobile phase for chiral h.p.1.c. (0.25 M-phosphate buffer, p H 6.2). Separation of enantiomers was achieved using a bovine serum albumin column (Macherey-Nagel 150mm x 4mm, Resolvosil-7, supplied by Technicol) at 3 5 T , with a mobile , detection at 220nm). phase flow rate of 0.8ml/min (injection volume 1 0 0 ~ 1U.V.
Quantification For achiral analyses, calibration curves were prepared by weighted regression of the peak height or peak height ratio, against the drug concentration for seven standards, using a reciprocal weighting factor (concentration’). Analysis of the HMG-CoA reductase inhibitor was performed using a calibration range from 5 to 2500 ng/mI of plasma and 25 to 2000 ng/ml of liver homogenate. For fluparoxan the range was 0.1 25 to 8.00 pg/ml of plasma. For routine quality control of achiral assays, duplicate standards, at three concentrations within the calibration range and prepared from a separately weighed stock of compound, were processed with each batch of samples. Assay performance was considered acceptable provided that four of these, with one at each concentration, were measured to within 15% of expected values. Following chiral h.p.l.c, enantiomeric ratios were determined from peak area measurements and the concentration of each enantiomer was derived from the ratio and the equivalent achiral result. It was considered that performance of the chiral analyses was satisfactory provided that ratios for four of the quality control samples, with one at each concentration, were within 5% of the racemic value of 1.00 for the HMG-CoA reductase inhibitor and within 10% for fluparoxan. Pharmacokinetics Pharmacokinetic analyses of data were performed using the program SIPHAR (V3.3, Simed, Creteil, France) running on Vax/VMS. Areas under plasma concentration/time curves were determined using the linear trapezoidal rule.
Results Recovery of compound during achiral analyses was > 75% for the HMG-CoA reductase inhibitor, I, over the calibration range used, and > 70% for fluparoxan. For both compounds the limit of quantification was taken as the concentration of the lowest standard.
793
Achiral H.P.L.C. in toxicokinetic studies A (i)
B (i)
(it)
-7
Downloaded by [RMIT University Library] at 10:31 25 April 2016
0
4
8
120
4
-
8
1 1
12
0
4
8
12.0
4
8
12
rnin min Figure 3 . Chiral h.p.1.c. chromatograms showing separation of enantiomers for the HMG-CoA reductase inhibitor, I, in fractions collected during achiral analysis of dog plasma and rat liver tissue samples. A: Dog plasma (1 ml), (i) blank, (ii) with added mc-I (250ng/ml).B: Rat liver (125 mg), (i) blank, (ii) with added rac-I (1600nglg).
A typical calibration plot of peak height for I against concentration gave a linear plot with r2 = 0-998. Replicate analyses (n= 6) of plasma samples containing drug, added at each of the seven concentrations used for calibration, gave mean values within 7% of the nominal concentration, with a maximum coefficient of variation of 12.3% at 5 ng/ml. Calibration plots of peak height ratio against concentration for fluparoxan were linear, with ~ ' 2 0 . 9 9 8 . Replicate analyses ( n = 6 ) at each of the seven calibration concentration gave mean values within 2% of the nominal concentration, with a maximum coefficient of variation of 8.3% at 0.25 pg/ml. Typical achiral chromatograms for extracts of control dog plasma with added rac-I (250 mg/ml) or rac-fluparoxan (0.5 pg/ml) are shown in figure 2. Sample analysis for the HMG-CoA reductase inhibitor, I Figure 2A shows typical achiral chromatograms for extracts of control dog plasma and plasma with added rac-I (250ng/ml). T h e resulting chiral chromatograms from these samples are shown in figure 3A. T h e cleanness of the collected fractions allowed multiple re-use of the extraction cartridges for concentration prior to chiral analysis, and the method was found to be equally applicable to plasma and liver samples from rats (figure 3B). This was in marked contrast to earlier results obtained without the achiral h.p.1.c. stage (extraction of drug into methyl-tert-butyl ether by mixing with solvent and saturated ammonium sulphate soln), where a potentially useful dog plasma assay did not translate to rat plasma (figure 4A). T h e achiral stage of the method was considered to be specific and no significant interfering peaks were observed during routine achiral or chiral analysis of experimental samples, including samples from toxicology studies. Again this contrasted with the earlier solvent extraction method (figure 4B) where the monitoring of enantiomeric ratios during a dog toxicity trial was impaired by an increase in co-extracting endogenous materials or drug metabolites. Guard columns of cr,-acid glycoprotein (AGP) were not available at the time of this work, but more than 300 samples were analysed on a single chiral-AGP column
A. P . Beresford et al.
794
-- -5
10
150
5
10
15
0
10
5
Downloaded by [RMIT University Library] at 10:31 25 April 2016
min
150
5
10
15
min
Figure 4. Chiral h.p.1.c. chromatograms showing attempts to analyse enantiomers for the HMG-CoA reductase inhibitor, I, directly on a chiral-AGP column following solvent extraction from dog and rat plasma. Extraction was with methyl tert-butyl ether and satd ammonium sulphate. A: Blank plasma (1 ml), (i) dog, (ii) rat. B:Plasma (1 ml) obtained 1 h post-dose from a dog following single daily oral doses of rac-I, (i) day 1 of dosing, (ii) day 26 of dosing.
with little deterioration of its performance. A slow increase in pressure, with a concomitant decrease in resolution of the enantiomers, was seen with time, and was compensated by decreasing the flow rate to 0 6 5 ml/min. Despite the absence of an internal standard in this method, the combination of solid-phase extraction and automated fraction collection was sufficiently reproducible to give satisfactory calibration curves for each enantiomer. Whilst this duplication of calibration data was unnecessary in normal use, it provided increased confidence in the specificity of both the chiral and achiral assays. Comparing results obtained from achiral analyses with those obtained by adding the separate (+) and ( - ) enantiomer data, showed the majority of results to be within 10% of each other. I n practice the chiral analysis of plasma I was used over the range 25-2500ng total drug/ml, compared to 5-2500 ng/ml for the achiral assay, the limitation being the broader peaks in the chiral chromatograms. Below 12*5ng/ml for each enantiomer a ratio of 1.00 for a racemic mixture could not be reliably (within 5%) obtained. When analysis was performed on samples of dog plasma prepared at the three quality control levels, with an increasing enantiomeric excess of the (+)enantiomer, all but one enantiomeric ratio was found within 5% of the expected Table 1. Areas under the plasma concentration/time curves for enantiomers of the HMG-CoA reductase inhibitor, I, in six male beagles, following single oral doses of the racemic I. Area under curve (ng/ml*p hour) by trapezoidal rule (AUC)
AUC ratio
Dose (mgikg)
Dog no.
1
9GR5 9GS4
7
9GR6 9HB4
633 481.
847 616
0.75 0.78
50
9GR7 9GS6
18900 11500
29400 22800
0.64 0.50
(+)-enantiomer
49.1 55.1
(-)-enantiomer
70.6 74.6
(+)/( -)
0.70 0.74
795
Achiral H.P.L.C. in toxicokinetic studies
-
10000,
L
al
> -
.-
. 0,
c 0,
m
1000-
5
m -
Q
-
E .
-
0,
100-
0 .-
+a
F
c
Downloaded by [RMIT University Library] at 10:31 25 April 2016
C V al
0
u
10
-
I
0
I
I
1
I
I
2
I
1
3
dosage (hi Figure 5. Concentrations of enantiomers of the HMG-CoA reductase inhibitor, I , in plasma and liver tissue from rats following single oral doses of the racemic drug. Dose of roc-I was 5mg/kg. 3R,SS(+) in liver (0); 3S,5R(-) in liver (m). 3 R , 5 S ( t ) in plasma (0); 3S,SR(-) in plasma ( 0 ) . Tlrne after
value. However, this experiment suggested that a change of at least 10% in the enantiomeric ratio would be required for reliable detection by the method over the specified range. Application of the method to samples from both dog and rat studies showed that exposure to the active, ( )-enantiomer was consistently lower (approximately 25%) than that to the (-)-enantiomer. This is illustrated by the dog plasma AUC data in table 1 and the rat plasma and liver concentrations shown in figure 5.
+
Analysis of fluparoxan samples T h e chiral method for fluparoxan has now been applied to plasma from rat, rabbit, dog and man. I n man the chiral analysis described was only suitable for measuring the enantiomers over the first few hours after an oral dose, as concentrations rapidly fell below the limit for reliable enantiomeric measurement (Gristwood 1990). However, the procedure has been used successfully to monitor the time-course of each isomer in plasma samples from the animal studies. Figure 6 compares chiral chromatograms of samples from an oral dose rat study with those of rac-fluparoxan standards. Due to the nature of the solid-phase extraction cartridge material used in the reconcentration of the achiral fractions, the internal standard (GR35996), having a 2-propenyl group on the pyrrole nitrogen, is not recovered at this stage. Hence, extending the fraction collection time would not endanger the integrity of the chiral chromatogram. This additional specificity at the recovery step further decreases the potential for co-elution on the chiral stationary phase. As with the chiral-AGP column, a slow decrease in resolution of enantiomers on the bovin serum albumin column was seen with continued use, and was compensated by decreasing the flow rate.
796 ill
-
Downloaded by [RMIT University Library] at 10:31 25 April 2016
1
1
0
10
20
0
10
20 0
20
10
10
0
20
Figure 6 . Chiral h.p.1.c. chromatograms showing separation of fluparoxan enantiomers injected oncolumn and from fractions collected during achiral analysis of plasma samples from rats following single daily oral doses of the racemic drug.
Dose of roc-fluparoxan was 50mg/kg: (i) 0.5 p g on-column, (ii) plasma (1 ml) with added rucfluparoxan (0.5pg/ml), (iii) pre-dose plasma (1 ml), (iv) post-dose plasma (1 ml).
The increase in limit of quantification due to the width of the enantiomer peaks seen for the HMG-CoA reductase inhibitor, I, was also apparent for fluparoxan analysis. Here a limit of 0.5pg/ml was set for the chiral assay, compared to 0 1 25 pg/ml for the achiral. This limit was the lowest concentration of rac-fluparoxan at which a 1 : 1 peak area ratio could be routinely obtained within 10% (compared with 5% for I). This translated to a limit of 0.25 pg/ml for either enantiomer, as a reliable ratio could not be obtained for a total fluparoxan concentration of 0.5 pg/ml comprised 0*4pgof (+)-isomer and 0.1 pg (-)-isomer in a sample. 0.1 pg (-)-isomer in a sample. T h e results from the single-dose rabbit study indicated that the ratio of circulating enantiomers was not quite racemic, with, on average, an approximately 15% excess of the (+)-isomer. There was no clear indication of a change in the enantiomeric ratio with time after dosing, and this was also true in the rat study, as illustrated for both day 1 and day 8 in figure 7.
0
Figure 7.
5 10 15 20 Time after dosage ( h )
25
if
,
I 0
5
I
I
10 15 20 Time after dosage (h)
I
25
Concentrations of fluparoxan enantiomers in plasma from rats following repeat single daily oral doses of the racemic drug. Daily dose of roc-fluparoxan was 50mg/kg: (+)-fluparoxan (0);(-)-fluparoxan (m): A: Day 1 of dosing. 8: Day 8 of dosing.
Achiral H.P.L.C. in toxicokinetic studies
797
Downloaded by [RMIT University Library] at 10:31 25 April 2016
Discussion During the early development of drugs containing one or more asymmetric centres it may be necessary to quantify individual isomers in a variety of biological samples generated during pharmacological/toxicological assessment of the compound. In such cases the use of an achiral h.p.1.c. step for purification of the compounds, prior to chiral separation, is recommended. This approach is equally applicable to methods using direct separation on chiral stationary phases or where diastereomers are prepared. T h e variety of achiral h.p.1.c. conditions available permit development of specific methods for quantification of total drug. Post-detector fraction collection can provide clean samples for the more demanding enantioselective analyses whilst circumventing potential problems of solvent incompatibility. As previously mentioned by Edholm et al. (1988) quantification of racemates during achiral h.p.l.c, with only enantiomeric ratios being measured at the chiral stage, precludes the need for internal standards in the latter analysis. However, quantification of both achiral and chiral chromatograms can help confirm assay specificities, and should permit the calculation of enantiomeric ratios in situations where only one isomer is reliably quantifiable at the chiral stage. This approach has been successfully applied in our laboratories to a number of compounds under pharmaceutical development. In each case different chiral stationary phases were required for the enantiomeric sepration, but achiral h.p.1.c. methods for total drug measurement were already available. As the results presented here show, the technique increases the likelihood that methods will be applicable to samples from different biological sources. Here the method was used to confirm that animals dosed during the toxicological testing of an HMG-CoA reductase inhibitor, I, were exposed to the active (+)enantiomer following administration of the racemate. The results also confirmed that the active enantiomer reached the site of pharmacological action, the liver, rapidly and at concentrations (on a ng/g basis) up to 30 times that of plasma. In the case of the a,-adrenoceptor antagonist, the approach was used to confirm that both enantiomers were absorbed from oral doses during toxicokinetic studies in animals, and showed that there was little difference in their plasma profiles.
Acknowledgements T h e authors are indebted to their colleagues M r I. M. Mutton (Chemical Analysis Dept) for advice on chiral stationary phase selection and Miss P. A. Williams for her help in preparation of the manuscript.
References ALBANI,F., RIVA,R., CONTIN,M., and BARUZZI, A , , 1984, Stereoselective binding of propanolol enantiomers to human al-acid glycoprotein and human plasma. British Journal of Clinical Pharmacology, 18, 244-246. ALLENMARK, S., 1986, Optical resolution by liquid chromatography on immobolised bovine serum albumin. Journal of Liquid Chromatography, 9, 425-442. ALLENMARK, S., 1989, Protein-bonded stationary phases, in Chiral Separations by h.p.1.c.:Applications to Pharmaceutical Compounds, edited by A. M. Krstulovic (Chichester, UK: Ellis Horwood), pp. 286-315.
BROWN, N. A. JAHNCHEN, E., MULLER, W. E., and WOLLERT, U., 1977, Optical studies on the mechanism of the interactionof the enantiomers of the anticoagulant drugs phenprocoumon and warfarin with human serum albumin. Molecular Pharmacology, 13, 70-79.
Downloaded by [RMIT University Library] at 10:31 25 April 2016
798
Achiral H.P.L.C. in toxicokinetic studies
CHU,Y.-Q,, and WAINER, I. W., 1988, Measurement of warfarin enantiomers in serum using coupled achiral/chiral high performance liquid chromatography (h.p.1.c.). Pharmaceutical Research, 5 (lo), 680-683. J., and WALHAGEN, A,, 1988, Determination of drug EDHOLM,L.-E., LINDBERG, C., PAULSON, enantiomers in biological samples by coupled column liquid chromatography and liquid chromatography-mass spectrometry. Journal of Chromatography, 424, 61-72. M., and HERMANSSON, J., 1989, Separation and quantitation of (R)- and (S)-atenolol in human ENQUIST, plasma and urine using an al-acid glycoprotein column. Chirality, 1 , 209-215. GRISTWOOD, W. E., 1990, Determination of fluparoxan (GR50360) in plasma by gas chromatography. Journal of Chromatography, 527, 436440. D. W., TYERS,M . B., WALSH,D. M., and WISE,H., 1988, HALLIDAY, C. A,, JONES,B. J . , STRAUGHAN, GR50360, a novel and highly selective a2-adrenoceptor antagonist. British Journal of Pharmacology, 95, 751P. W. E., and WOLLERT, U., 1975, High stereospecificity of the benzodiazipine binding site on MULLER, human serum albumin. Molecular Pharmacology, 1 1 , 52-60. NAKAYA, N., HOMMA, Y.,TAMACHI, H., and GOTO,Y.,The effect of CS-514, an inhibitor of HMG-CoA reductase, on serum lipids in healthy volunteers. Atherosclerosis, 61, 125-126. O., LINDSTROM, B., HERMANSSON, J., ADJEPON-YAMOAH, K., and SJOQVIST, OFORI-ADJEI, D., ERICSSON, F., 1986, Enantioselective analysis of chloroquine and desethylchloroquine after oral administration of racemic chloroquine. Therapeutic Drug Monitoring, 8, 457461. SCHILL, G., WAINER,I. W., and BARKAN, S. A,, 1986, Chiral separation of cationic drugs on an al-acid glycoprotein bonded stationary phase. Journal of Liquid Chromatography, 9 (2 & 3), 641-666. WALHAGEN, A,, and EDHOLM, L.-E., 1989, Coupled-column chromatography on immobilized protein phases for direct separation and determination of drug enantiomers in plasma. Journal of Chromatography, 473, 371-379.
