CHIRALITY 41342-348 (1992)
Formation of Glycine Conjugate and ( - )- (R) -Enantiomer From ( + )-(S)-2-PhenylpropionicAcid Suggesting the Formation of the CoA Thioester Intermediate of ( + )- (S)-Enantiomer in Dogs Y. TANAKA, Y. SHIMOMURA, T. HIROTA, A. NOZAKI, M. EBATA, W. TAKASAKI, E. SHIGEHARA, R. HAYASHI, AND J. CALDWELL Andyticd and Metabolic Research Laboratories, Sankyo Co., Ltd., Tokyo, JaDan (Y.T., Y.S., T.H., Afl., ME,, W.T., E.S., R.H.) and DeDarbnent of Pharmacology and Toxicolo~,St. Mary’s Hospital Medical School, Imperial College of Science, Technology and Medicine, Norfolk Pluce, London, Englund U.C.)
ABSTRACT It has been proposed that the chiral inversion of the 2-arylpropionicacids is due to the stereospecific formation of the ( - )-R-profenyl-CoA thioesters which are putative intermediates in the inversion. Accordingly, amino acid conjugation,for which the CoA thioesters are obligate intermediates, should be restricted to those optical forms which give rise to the ( - )-R-profenyl-CoA,i.e., the racemates and the ( - )-(R)-isomers.We have examined this problem in dogs with respect to 2-phenylpropionicacid(2-PPA).Regardless of the optical configuration of 2-phenylpropionicacid administered, the glycine conjugate was the major urinary metabolite and this was shown to be exclusively the (+)-(S)-enantiomerby chiral HPLC. Both ( - )-(R)-and ( + )-(S)-2-phenylpropionicacid were present in plasma after the administration of either antipode, and further evidence of the chiral inversion of both enantiomers was provided by the presence of some 25% of the opposite enantiomer in the free 2-phenylpropionicacid and its glucuronide excreted in urine after administration of (-)-(R)- and (+ )-(S)-2-phenylpropionic acid. The ( + )-(S)-enantiomerunderwent chiral inversion to the ( - )-(R)-antipodewhen incubated with dog hepatocytes. These data suggests that both enantiomers of 2-phenylpropionic acid are substrates for canine hepatic acyl CoA ligase(s) and thus undergo chiral inversion, but that the CoA thioester of only (+)-(S)-2-phenylpropionicacid is a substrate for the glycine N-acyl transferase. These studies are presently being extended to the structure and species specificity of the reverse inversion and amino acid conjugation of profen NSAIDs. 0 1992 Wiley-Liss, Inc.
KEY WORDS: arylpropionic acid, 2-phenylpropionic acid, glycine conjugation, stereoselectivity, chiral inversion, reverse chiral inversion, glycine N-acyl transferase, dog hepatocytes, chiral HPLC INTRODUCTION It is a well known phenomenon that the ( - )-(R)-enantiomers of 2-arylpropionic acids, the “profen” nonsteroidal antiinflammatory drugs(NSAIDs),are metabolically inverted to their ( + )-@-antipodes. The in vitro inhibitory activity of these agents against cyclooxygenase exclusively resides in the ( + )-(S)-enantiomers. Their ( - )-(R)-isomers, however, show pharmacological effects in vivo as a result of this unique chiral inversion reaction. Nakamura et al.4 proposed the chiral inversion mechanism of ibuprofen to be as follows (Fig. la):(l) ( - )-(R)-ibuprofenis stereospecifically activated to its CoA thioester by acyl CoA ligase(s)in microsomes and/or mitochondria, (2)(- )-(R)-ibuprofenyl-CoAis then epimerized by a cytosolic epimerase, and (3) the ( - )-(R)-and ( + )-(S)-CoAthioesters are hydrolyzed to yield free acids. Amino acid conjugation is one of the important metabolic reactions for compounds containing the carboxylic acid group and this proceeds by the mechanism shown in Figure lb5: (1) the carboxylic acid is transformed to its activated CoA thioester via an intermediate acyl adenylate and (2) the CoA thio@
1992 Wiley-Liss, Inc.
ester is conjugated with an amino acid by amino acid N-acyl transferases specific for the amino acid involved. Consideringthese mechanisms of chiral inversion and amino acid conjugation,one would predict that only the ( - )-(R)-enantiomers of 2-arylpropionic acids will give rise to amino acid conjugates. This hypothesis emerged from the detection of the taurine conjugate as a major urinary metabolite of loxoprofen sodium in dogs6 We have examined this problem systematically using 2-phenylpropionic acid, which is the simplest congener of the 2-arylpropionicacids and which has been reported to produce a glycine conjugate in dogs7 MATERIALS AND METHODS Chemicals 2-Phenylpropionicacid (2-PPA), its ( - )-(R)-and ( + )-(S)-enantiomers (optical purity W%),(+)-(R)-1-[naphthen-1-yllethylReceived for publication March 11, 1992; accepted May 20, 1992. Address reprint requests to Y. Tanaka, Analytical and Metabolic Research Laboratories, Sankyo Co.,Ltd., 1.2-58,Shinagawa-ku,Tokyo 140, Japan.
343
REVERSE CHIRAL INVERSION OF 2-PPA IN DOGS
(a) Chiral inversion mechanism : Acyl CoA ligase
(-)-R-IBU-COA Eplmerase
(+)-S-IBU
Hydrolase
(-)-R-i BU
+ *
(+)-S-IBU-CoA
(-)-R-IBU
(+)-S-IBU
Hydrolase
(b) Amino acid conjugation mechanism :
-
RCO-AMP RCO-SCoA -RCONHCH(R')COOH
RCOOH +ATP RCO-AMP + COASH RCO-SCoA + HSNCH(R')COOH
+ PP + AMP + CoASH
-
(c) From these two mechanisms, it would be predicted that: Amino acid conjugate (-)-R-profen NSAID (+)+profen
NSAID
+ D
Amino acid conjugate
Fig. 1. Proposed mechanisms for (a) chiral inversion" and (b) amino acid conj~gation.~ IBU, ibuprofen; NSAID, non-steroidal anti-inflammatory drug.
amine (NEA) with 99% optical purity, 1-hydroxybenzotriazole (HBT), dicyclohexylcarbodiimide (DCC), benzylthioacetic acid, 4-phenyl-n-butyricacid, and benzoic acid were all purchased from Tokyo Kasei Kogyo Co., Ltd. (Tokyo,Japan). Collagenase was obtained from Boehringer Mannheim GmbH (Germany).N[2-Hydroxyethyllpiperazine-N'-[2-ethanesulfonicacid] (HEPES) was from Sigma Chemical Co. (St. Louis, MO). All other reagents and solvents were of reagent grade. Glycine conjugates of (RS)- and ( + )-(S)-2-phenylpropionic acid were synthesized according to the procedure described by Dixon et a1.8 The identity of glycine conjugates synthesized and isolated from urine as described below were confirmed by mass spectrometry (MS). The optical purity of the glycine conjugate synthesized from (+)-(S)-2-PPAwas 99%. Animals and Dosing Three male beagle dogs (body weight W 1 2 kg) were used after fasting overnight. 2-Phenylpropionic acid dissolved in PEG-400/water (1:l)was orally administered by stomach tube at a dose of 200 mg/kg for the study of urinary metabolites and at a dose of 20 mg/kg for the plasma concentration study. Each enantiomer of 2-phenylpropionic acid was given at half the dose of the racemate. The animals were housed in metabolism cages to collect 24 h urine and feces, separately. Blood samples were taken from a foreleg vein into heparinized tubes at 10,20, 30 min, 1, 2, 4, and 6 h after dosing. Plasma was obtained by centrifugation. Urine and plasma were stored at - 20°C until analyzed. Isolation of Glycine Conjugate of 2-Phenylpropionic Acid Dog urine (50 ml) after oral administration of (-)-(R)-2phenylpropionic acid was adjusted to pH 2 with 4 M HCl and extracted with ethyl acetate (100 ml x 2). The extracts were concentrated to dryness in vacuo. The residue was subjected to preparative TLC purification: thin-layer, Kieselgel 60 F254 plate (Merck, thickness 0.25 mm); solvent system, benzene/ ethyl acetate/acetic acid (10:5:1). The zone of Rf 0.4 was
scraped off and extracted with methanol, and concentrated in vacuo. The sample obtained was further purified by semipreparative HPLC column, pBondapak C18 (Waters, 8 x 300 mm); mobile phase, methanol/water (50:50); flow rate, 4 ml/ min; detection, W 220 nm. Determination of Metabolites and Enantiomeric Composition Urine To urine (0.2 ml) were added 50 pg 4-phenyl-n-butyricacid (20 pl of 2.5 mg/ml solution)as an internal standard for the free acid and 500 pg benzoic acid (20 p1 of 25 mg/ml solution) for the glycine conjugate, followed by 0.8 m10.2 MHCl. The whole was extracted with 5 ml ethyl acetate on a vortex mixer, the layers separated by centrifugation, and the organic layer evaporated to dryness in vacuo. The residue was taken up in 0.5 ml methanol and analysed using HPLC system A for free 2-phenylpropionic acid and B for the glycine conjugate. The ester glucuronide of 2-PPA was isolated and identified as a major urinary product in several animal species, and hydrolyzed readily on warming with M NaOH for 1 h or incubating with P-glucuronidase at pH 5 Thus, the residual aqueous layer after extraction with ethyl acetate was mixed with 0.5 m12 M NaOH and incubated at 37°C for 30 min. To the reaction mixture were added 500 pg 4-phenyl-n-butyricacid and 0.5 ml 3 M HCl. The 2-phenylpropionic acid released was analysed by the same procedure as described above. The Calibration curves constructed for each metabolite were linear over the range G250 pg/ml. Quantitation was performed by reference to the calibration curves. The free and released 2-phenylpropionic acid and glycine conjugate were collected from the reversed-phase HPLC assay and the eluates were evaporated to dryness in vacuo. In the case of 2-phenylpropionic acid, the residue was treated with NEA, HBT, and DCC as described by Nagashima et al." to convert the acids to their respective ( + )-(R)-1-[naphthen-1-yllethylamides, which were then analysed using HPLC system C. Separation of the enantiomers of the glycine conjugate was achieved using HPLC system D without derivatization. *y9
344
TANAKA ET AL.
Plasma To specimens of plasma (0.1 ml) were added 5.0 pg benzylthioacetic acid (20 p1 of 250 mg/ml solution) as an internal standard and 0.9 m10.25 M HCl, and the whole was extracted with 4 ml hexane/ethyl acetate (6:4). After evaporation, the residue was derivatized and analysed as above.
B)HPLC system B
AIHPLC system A
Metabolism of ZPhenylpropionic Acid by Isolated Hepatocytes
Hepatocytes were isolated from male beagle dogs according to the technique of Moldeus et al.” Briefly, the dogs were anaesthetized by an intravenous injection of sodium pentobarbital (25 mg/kg). A small whole lobe of the liver was chosen and a vessel producing clearance of the maximum volume was cannulated with a tight-fitting cannula. The liver lobe was first perfused with a calcium-free Hanks’ buffer containing EGTA (0.5mM),NaHCO3 (25 mM), and HEPES (12.6 mM) at 37°C for 12 min at a pressure of 60-70 cm H20.The perfusion was then switched over to collagenase (0.1%) in Hanks’ buffer without EGTA but supplemented with CaCI2 (2.0 mM). After perfusing for 10 min, isolated cells were filtered through cotton gauze, washed three times by centrifugation, and resuspended in Krebs-Henseleit buffer containing HEPES (12.6 mM). Viability of the cells, as assessed by Trypan blue exclusion, was routinely greater than 85%. Incubations (4 x lo6 cells/ml) were carried out in KrebsHenseleit buffer containing HEPES in rotating 50 ml roundbottomed flasks at 37°C and were gassed continuously with 95% 02/5% COB.The final concentrationwas 0.1 mMand 0.05 mMfor the racemate and each enantiomer of 2-phenylpropionic acid, respectively. Aliquots (1.0 ml) of hepatocyte suspensions were taken at 0.25, 0.5, and 1 h after incubation, and centrifuged at 10,OOOrpm for 2 min. The supernatant obtained was analysed by the above method.
I
I
I
I
I
I
1
I
1
1
I
Fig. 2. HPLC profiles of urinary metabolites after oral administration of enantiomersof 2-phenylpropionic acid (2-PPA)to dogs. Internal standards (IS.) are 4-phenyl-n-butyricacid in HPLC system A for the detection of 2-phenylpropionic acid and benzoic acid in system B for the glycine conjugate. (a)Control urine, (b) urine of animals dosed with (-)-R-2-PPA and (c) urine of animals dosed with (+)-S-2-PPA.
High-pressure Liquid Chromatography (HPLC) The flow rate was 1.0 ml/min, except for 0.8 ml/min in firmed to be the unchanged free acid, and the other in system system C. Eluates were monitored by ultraviolet absorption at B. This was isolated from urine by successive procedures of 210 nm to detect 2-phenylpropionic acid and its glycine conju- ethyl acetate extraction, preparative TLC and HPLC. The retengate and at 282 nm to detect the (+)-(R)-1-[naphthen-1-ylle-tion time on HPLC and FAB/MS/MS (Fig. 3) of the metabolite thylamides. The other column and mobile phase conditions were identical to those of a synthetic sample of the glycine were as follows. A for 2-phenylpropionic acid column, ERC conjugate of 2-phenylpropionic acid. It was evident from the ODs-1282(ERMA CR. Inc., Tokyo, 8 x 250 mm); mobile phase, HPLC trace that the glycine conjugate was excreted as a major acetonitrile/O.l% trifluoroacetic acid (45:55). B for the glycine metabolite in urine after the administration of both enantiomconjugate of 2-phenylpropionic acid: column, Senshu PAK C8- ers of 2-phenylpropionic acid. 1202-P(Senshu Scientific Co., Ltd., Tokyo, 4.6 x 150 mm) linked Quantitation of Urinary Metabolites in series with ERC ODS-1282(8~250mm), mobile phase; The 24 h urinary metabolites were quantitated by HPLC acetonitrile/O.l% trifluoroacetic acid (22:78). C for ( + )-(R)-1[naphthen-1-yllethylamides:column, ERC Silica-l282(ERMA after oral administration of 2-phenylpropionicacid at a dose of CR. Inc., 8 x 250 mm); mobile phase, hexane/ethyl acetate(6:4). 200 mg/kg and its enantiomers at a dose of 100 mg/kg to dogs D for the enantiomers of the glycine conjugate: column, CHI- (Table 1). When given the ( - )-(R)-enantiomer, the main RALPAK WH (Daicel Chemical Industries Ltd., Tokyo, 4 x 250 metabolites were the conjugates with glucuronic acid and glycine which accounted for 23.8% and 21.4% of dose, respecmm); mobile phase, 0.3 mM copper sulfate. tively, in addition to the parent compound (11% of dose). A RESULTS similar excretion pattern was found after the (+)-(S)-enantiHPLC Profile of Urinary Metabolites omer: unchanged acid, 6.9%; glucuronic acid conjugate, 31.6%; Figure 2 shows the typical HPLC profile of urinary metabo- glycine conjugate, 23.1 YO.The glucuronide tended to be formed lites after oral administration of the enantiomers of 2-phenyl- in a slightly larger amount from the (+ )-(S)-isomerthan the propionic acid to dogs. The ethyl acetate-extractable fraction ( - )-(R)-antipode. Similarly, the racemate was primarily contained two peaks, one in HPLC system A which was con- metabolised to the glucuronic acid (40.3%)and glycine (20.6%)
345
REVERSE CHIRAL INVERSION OF 2-PPA IN DOGS
(+)-(S)-2-phenylpropionicacid, but not as to its ( - )-(R)-enantiomer. Consequently, the amount of glucuronide from both enantiomers was almost same at 60 min of incubation. The increase of glucuronide after 30 min of incubation may be partly contributed by the ( + )-(S)-enantiomer inverted from ( - )-(R)-2-PPA. The glycine conjugate was generated to similar extents from the ( + )-(S)-and ( - )-(R)-enantiomersof 2-phenylpropionicacid, suggesting no stereoselectivity in the conjugation reaction, compatible with in vivo data. However, the degree of glycine conjugate formation was about one-third that of the glucuronide in isolated hepatocytes, unlike the approximately equal excretion rates of the two conjugated metabolites in urine. The ( + )-(S)-enantiomer generated from ( - )-(R)-2-PPA could be used for the formation of glycine conjugate having (+)-(S)-configuration as described below.
a) Standard Sample 105
1111
I1
m/+
100
2 00
b) Isolated Sample
n
105
208
1
Enantiomeric Composition of ZPhenylpropionic Acid and Its Glycine Conjugate 76
100
200
m/z
Fig. 3. FARjMSIMS spectra of glycine conjugate of 2-phenylpropionicacid. (b) isolated sample from dog urine dosed with (-)-R-2-phenylpropionic acid. (a) Synthetic standard sample and
conjugates. These results revealed that there is no substrate stereoselectivity for the glycine conjugation of 2-phenylpropionic acid, which is a main metabolic route in dogs. Metabolism of ZPhenylpropionic Acid by Dog Hepatocytes The production of glucuronic acid and glycine conjugates from the enantiomers of 2-phenylpropionic acid was investigated using freshly isolated dog hepatocytes (Fig. 4).The formation of glucuronide was stereoselective for the ( + )-@)-enantiomer up to 30 min of incubation. It corresponds to the tendency of a larger urinary excretion rate of glucuronide from the ( + )-(S)-enantiomerthan the ( - )-(R)-isomer.After 30 min, however, a plateau was reached as to glucuronide formation of TABLE 1. Urinary metabolites after oral administrationof 2-phenylpropionicacid and its enantiomers to dogs
The enantiomeric composition of 2-phenylpropionic acid as determined by normal phase HPLC of the ( + )-(R)-1-[naphthen1-yllethylamides. The enantiomers of glycine conjugate of 2-phenylpropionicacid were quite well separated by a chiral HPLC using a ligand-exchangecolumn CHIRALPAK WE (Fig. 5). Those values obtained are summarized in Table 2. The enantiomeric ratios of free and glucuronidated 2-phenylpropionic acid after oral administration of the ( - )-(R)-enantiomer were essentially identical [( +)-(S)/(- )-(R) ratios 29:71, free; 33:67, glucuronide]. Comparable data were obtained after the (+)-(S)-enantiomer,the (+ )-(S)/(- )-(R) ratios being 77:23 for free and 78:22 for the glucuronide. Taken together, these results indicate the occurrence of an unusual and extensive inversion from ( + )-(S)-to ( - )-(R)-2-phenylpropionicacid. The stereochemistry of the glycine conjugate is very interesting in that while it is formed following the administration of both enantiomers of 2-phenylpropionic acid, chiral HPLC showed clearly that the glycine conjugate has the ( + )-(S)-configuration exclusively. The enantiomeric compositions in the in vitro experiments using isolated hepatocytes are given in Table 2. The ( + )-(S)/(- )-(R) ratios of the residual and glucuronidated acid were 19:81 and 20:80, respectively, when the ( - )-(R)-enanti-
TABLE 2. Enantiomeric composition of free, glucuronic acid and glycine conjugate produced in urine and hepatocytes from enantiomers of 2-phenylpropionicacid a
+
( )-(S)/( - )-(R) ratio
YOof Dose Metabolites
Racemateb
Unchanged Glucuronide Glycinate
5.8 40.3 20.6
( + )-(SIC
6.9 f 1.5 31.6 f 4.0 23.1 f2.7
(- )-(RY
10.6f2.2 23.8 f 2.9 21.4f 3.8
"Racemate was administered at a dose of 200 mg/kg and its enantiomers at 100 mg/kg. Twenty-four hour urine was analysed. *Mean of two animals. cMean f SE of three animals.
Sample Urine Hepatocytes
Used enantiomer
Free
Glucuronide
Glycinate
(-)-(R) (+)-(S) (-)-(R) (+)-(S)
29171 7723 19:81 46:54
33:67 78:22 20 :80 83:17
( + 1-6) ( + )-(S) ( + )-(S) ( + )-(S)
"The 24 h urine samples in Table 1 and 60 min incubation mixtures in Figure 4 were analysed. b( + )-(S)-Enantiomerwas exclusively detected.
346
TANAKA ET AL.
A) GI u c
ur o n i d e
10
5
.I
BIGlycine c o n j u g a t e
;-..--4
-0
/ / -/ --
I
0
15
Racemate (4-R-2 -PPA
A (+)-S- 2- PPA
I
60
30 Incubation time (min)
Fig. 4. Formation of (A) glucuronide and (B) glycine conjugate from racemic (O,.), (+)-S- ( A , A ) and (-)-R(O,l, enantiomers of 2-phenylpropionic acid (2-PPA) by isolated dog hepatocytes. Symbols represent mean of three experiments.
omer was employed as a substrate. The corresponding ratios were 46:54 and 83:17, when the (+)-(S)-isomerwas incubated. Again, only the glycine conjugate of ( + )-(S)-2-phenylpropionic acid was detected by the chiral HPLC. Thus, the reverse chiral inversion and the stereospecific formation of glycine conjugate mentioned before were confirmed to take place in hepatocytes. Plasma Concentration
The racemate (20 mg/kg) and enantiomers (10 mg/kg) of 2-phenylpropionicacid were orally administered to dogs. The plasma concentration-time curves of each enantiomer are shown in Figure 6. The peak levels of each enantiomer were at 30 min with a slightly, but not significantly, higher concentration of the ( + )-(S)-isomer.After administration of the racemate, the areas under the plasma concentration-time curves (AUC)of the ( + )-(S)-and ( - )-(R)-enantiomerswere 74.2 f 9.3 and 73.1 f 4.3 pgh/ml, respectively, which were not significantly different. This might suggest that a chiral inversion apparently does not occur with 2-phenylpropionic acid in dogs. Nevertheless, the ( t )-('$isomer gradually appeared in the plasma dosed with the ( - )-(R)-2-phenylpropionicacid and reached to a maximum level around 1-2 h later. The AUC of ( + )-6)-enantiomer
newly generated was 18.5f 7.6 pgh/ml, which corresponded to about 40% of that of (- )-(R)-enantiomer(44.1 f 6.6 pgh/ml) administered originally. In addition, the reverse phenomenon was observed after administration of ( + )-(S)-2-phenylpropionic acid: the antipode having ( - )-@)-configurationappeared in the plasma with the peak time at 1 h and the AUC was 29.1 f5.4 pgh/ml corresponding to 70% of the AUC of (+ )-(S)-enantiomer itself (41.2 f 10.3 pgh/ml). Judged from these facts, it is almost certain that both the usual and unusual chiral inversions occur when 2-phenylpropionic acid is given as a single enantiomer to dogs. DISCUSSION
The metabolism of 2-phenylpropionicacid in dogs was first examined by Kay and Raper,7 who found that the racemate gave rise to an ester glucuronide and a glycine conjugate. They also showed the isolated glycine conjugate was dextrorotatory. In the present experiments, the glycine conjugate was isolated from urine of dogs given with the ( - )-(R)- and ( + )-@)-enantiomers and the racemate, and confirmed by FAB/MS/MS. The enantiomeric composition of the amino acid conjugate was directly determined by chiral HPLC using a ligand exchange
347
REVERSE CHIRAL INVERSION OF 2-PPA IN DOGS 30
r
aIStandard Glycine conj.
b)(-)-R-2-PPA PO Glycine conj.
A
0
25
1
2
3
4 Time (hr)
5
6
1
2
3 4 Time (hr)
5
6
r
c)(+)-S-~-PPA PO Glycine conj.
Fig. 5. Enantiomeric separation of glycine conjugate of 2-phenylpropionic acid (2-PPA)by chiral HPLC. Column was CHIRALPAK WH and mobile phase was 0.3rnM CuSO4.
column, CHIRALPAK WH. It was revealed that the glycine conjugate excreted as a major metabolite was exclusively the ( + )-(S)-enantiomerafter administration of both enantiomers of 2-phenylpropionic acid. Carboxylic acids must be activated to CoA thioester intermediate during amino acid conjugation pathway, The formation of a glycine conjugate having ( + )-(S)-configurationfrom ( - )-(R)-2-phenylpropionicacid is consistent with the formation of the CoA thioester of the ( + )-(S)-enantiomerduring the chiral inversion reaction. However, it is difficult to explain the production of glycine conjugate of the ( + )-(S)-enantiomerafter dosing ( + )-(S)-Zphenylpropionic acid, because the (+)-(S)-isomers do not serve as a substrate for acyl CoA ligase(s)according to the accepted inversion mechanism. The conflict was thought to be resolved by conversion of the ( +)-(S)-enantiomerof 2-phenylpropionic acid to its CoA thioester intermediate in dogs. Once the CoA thioester of the (+)-(S)-enantiomer is formed it will be converted to the ( - )-(R)-enantiomerby the catalysis of racemase. In this connection, the enantiomeric composition of 2-phenylpropionic acid was analysed. The appearance of the (+ )-(S)-enantiomerin plasma after administration of the (- )-(R)-isomer is the result of chiral inversion. In addition to the usual R+S inversion, a reverse inversion (S+R) was shown by the detection of the ( - )-(R)-isomerin plasma when the ( + )-(S)-antipodewas administered. The opposite enantiomers had a delayed peak time around 1-2 h after oral dosing compared to the administered isomers in the same manner as ibuprofen.12 After the peak time, the plasma concentrations of each enantiomer were approximately the same following the administration of both the
0
Fig. 6. Plasma concentrations of enantiomers after oral administration of racemic (2hg/kg), ( - )-R. and ( + )-S-enantiomers(lOmg/kg) of 2-phenylpropionic acid (2-PPA) to dogs.
+
( )-(S)-and ( - )-(R)-enantiomers.This suggests an equilibrium
state between the enantiomers, i.e., a racemization of each. The chiral inversion was apparently not observed after administration of the racemate, possibly due to the similar extent of inversion to both directions. Consequently, it may be said that the enantiomers of 2-phenylpropionic acid racemize in dogs. The analysis of enantiomeric composition was also performed with respect to the free and glucuronidated 2-phenylpropionic acid in urine. The opposite enantiomer was present in the free acid to the extent of 29% after the ( - )-(R)- and 23% after the ( + )-(S)-enantiomer.These compositions were substantially reflected in those of their glucuronides. Thus, both directions of inversion, from ( - )-(R)- to ( + )-(S)- and from ( + )-(S)- to ( - )-(R)-2-phenylpropionicacid, occur in dogs. There have been debates, recently reviewed by Williams,l3 as to the organ(s) responsible for the chiral inversion of profen NSAIDs. Freshly isolated dog hepatocyes were used to examine the metabolism of 2-phenylpropionic acid. The results obtained were essentially similar to those in vivo: the stereochemistry of glycine and glucuronic acid conjugations and both
348
TANAKA ET AL.
S-(+)-2-PPA Glucuronlde
1.
S-(+)-2-PPA
R-(-)-2-PPA
__F
1
-
JS-(+)-2-ppA
S-(+)-2-PPA-SCoA
It
R-(-)-2-PPA-SCoA
S-(+)-2-PPA Glyclnate
R-(-)-2-PPA Glyclnate
\ R-(-)-2 - P PA
R-(-)-2-PPA Glucuronide Fig. 7. Possible metabolic mechanisms of enantiomers of 2-phenylpropionic acid (2-PPA) in dogs,
directional inversions of 2-phenylpropionic acid were reproduced, indicating the liver is a main site of these metabolic reactions. Based on the combined results, the stereoselective metabolism of the enantiomers of 2-phenylpropionic acid in dogs is proposed in Figure 7. Both enantiomers are converted to CoA thioesters, which then inverted to the respective antipode by the action of an epimerase. However, only the thioester of (+)-(S)-2-phenylpropionicacid is a substrate for glycine N-acyl transferase to form the glycine conjugate stereospecifically . The metabolic chiral inversion is usually thought of as unidirectional. But, the reverse inversion, from (+)-(S) to ( - )-(R), has been suggested for 2-phenylpropionicacid in rats by Hutt et al.I4 and Fournel et al.,9 who reported small but significant amounts of the (-)-(R)-enantiomer was excreted in urine after administration of the ( + )-(S)-enantiomer.The present report clearly revealed the occurrence of this reverse reaction in dogs by the detection of the (- )@)-enantiomer in plasma dosed with ( + )-(S)-2-phenylpropionicacid. However, there was a marked species differences in the reverse inversion of 2-phenylpropionic acid, which was found in the mouse but not in the rat, guinea pig, and rabbit (unpublished data). Fournel et al.9 showed that the chiral inversion was found to occur after administration of racemic 2-phenylpropionicacid to the rat and rabbit, but not in the mouse. The species differences reported by Fournel et al. may be explained on the basis of the reverse inversion shown here. The reason for the apparent absence of chiral inversion of racemic 2-phenylpropionicacid in the mouse may be due to the active reverse reaction in this species. Recently Chen et al. reported that guinea pigs could effect the inversion of both (- )-(R)-and (+)-(S)-ibuprofenwith nearly the same efficiency, and that ( + )-(S)-ibuprofenmetabolism gave rise to small but significant levels of (-)-(R)-antipode in both rats and rabbits. l5 The (+)-(S)-enantiomerof ketoprofen was also inverted to the (- )-(R)-antipodein the mouse (unpublished data), but not in the rat according to Iwakawa et a1.I6 The reverse chiral inversion of ( + )-(S)-ZPPA,however, was not shown in guinea pig as mentioned previously. Since there seems to be a large difference in this reverse inversion as to the species and chemical structures, we are now investigating these problems in vivo and in vitro. ACKNOWLEDGMENTS
The authors thank Dr. Kinoshita, in our laboratories, for the mass spectromeric analyses.
LITERATURE CITED 1. Hutt, A.J., Caldwell, J. The metabolic chiral inversion of 2-arylpropionic acids-a novel route with pharmacological consequences. J. Pharm.Pharmacol. 35693-704, 1983. 2. Hutt, A.J., Caldwell, J. The importance of stereochemistry in the clinical pharmacokinetics of the 2-arylpropionic acid nonsteroidal anti-inflammatory drugs. Clin. Pharmacokinet. 9371-373, 1984. 3. Caldwell, J., Hutt A.J., Fournel-Gileux,S. The metabolic chiral inversion and dispositional enantioselectivity of the 2-arylpropionicacids and their biological consequences. Biochem. Pharmacol. 37:105-114,1988. 4. Nakamura, Y., Yamaguchi, T., Takahashi, S., Hashimoto, S., Iwatani, K., Nakagawa, Y. Optical isomerisation mechanism of ( - )-R-hydratropic acid derivatives. J. Pharmacobio-Dyn. 4S1, 1981. 5. Caldwell. J. Structure-metabolism relationships in amino acid conjugation. In: Conjugation Reactions in Drug Biotransformation. Aitio, A,, ed. Amsterdam: Elsevier/North Holland, 1978 111-120. 6. Tanaka, Y., Nishikawa, Y., Hayashi R. Species differences in metabolism of sodium 2-[4-(2-oxocyclopentylmethyl)phenyllpropionatedihydrate (Loxoprofen sodium), a new anti-inflammatory agent. Chem. Pharm. Bull. 31: 365G3664.1983. 7. Kay, H.D., Raper, H.S. XXXVII. The mode of oxidation of fatty acids with branched chains. 11. The fate in the body of hydratropic, tropic, atrolactic and atropic acids together with phenylacetaldehyde. Biochem. J. 16465474, 1922. 8. Dixon, P.A.F., Caldwell, J., Smith, R.L. Metabolism of arylacetic acids 2. The fate of [ 14C]hydratropicacid and its variation with species. Xenobiotica 7~707-715,1977. 9. Fournel, S.,Caldwell, J. The metabolic chiral inversion of 2-phenylpropionic acid in rat, mouse and rabbit. Biochem. Pharmacol. 35:415%4159, 1986. 10. Nagashima, H., Tanaka, Y., Watanabe, H., Hayashi, R., Kawada, K. Optical inversion of (2R)- to (%)-isomers of 2-[4-(2-oxocyclopentylmethyl)phenyl]propionic acid (Loxoprofen), a new antiinflammatory agent, and its monohydroxy metabolites in the rat. Chem. Pharm. Bull. 32251-257,1984. 11. Moldeus, P.,Hogberg, J., Orrenius, S. Isolation and use of liver cells. Methods Enzymol. 51:&71, 1978. 12. Beck, W.S., Geisslinger, G., Engler H., Brune, K. Pharmacokinetics of ibuprofen enantiomers in dogs. Chirality 3:165-169, 1991. 13. Williams, K.M. Enantiomers in arthritic disorders. Pharmacol. Ther. 46272295,1990. 14. Hutt, A.J., Fournel, S.. Caldwell, J. Application of a radial compression column to the high-performance liquid chromatographic separation of the enantiomers of some 2-arylpropionic acids a s their diastereomeric ( - )-Sl(naphthen-1-y1)ethylamides. J. Chromatogr. 378:40%418, 1986. 15. Chen, C.-S., Shieh, W.-R., Lu,P.-H., Harriman, S. and Chen, C.-Y. Metabolic stereoisomeric inversion of ibuprofen in mammals. Biochim. Biophys. Acta. 1078411417, 1991. 16. Iwakawa, S., He, X., Hashimoto, S., Volland. C., Benet, L.Z. and Lin, E.T. Stereoselective disposition of ketoprofen in rats. Drug Metab. Dispos. 1 9 717-718, 1991.
Formation of Glycine Conjugate and ( - )- (R) -Enantiomer From ( + )-(S)-2-PhenylpropionicAcid Suggesting the Formation of the CoA Thioester Intermediate of ( + )- (S)-Enantiomer in Dogs Y. TANAKA, Y. SHIMOMURA, T. HIROTA, A. NOZAKI, M. EBATA, W. TAKASAKI, E. SHIGEHARA, R. HAYASHI, AND J. CALDWELL Andyticd and Metabolic Research Laboratories, Sankyo Co., Ltd., Tokyo, JaDan (Y.T., Y.S., T.H., Afl., ME,, W.T., E.S., R.H.) and DeDarbnent of Pharmacology and Toxicolo~,St. Mary’s Hospital Medical School, Imperial College of Science, Technology and Medicine, Norfolk Pluce, London, Englund U.C.)
ABSTRACT It has been proposed that the chiral inversion of the 2-arylpropionicacids is due to the stereospecific formation of the ( - )-R-profenyl-CoA thioesters which are putative intermediates in the inversion. Accordingly, amino acid conjugation,for which the CoA thioesters are obligate intermediates, should be restricted to those optical forms which give rise to the ( - )-R-profenyl-CoA,i.e., the racemates and the ( - )-(R)-isomers.We have examined this problem in dogs with respect to 2-phenylpropionicacid(2-PPA).Regardless of the optical configuration of 2-phenylpropionicacid administered, the glycine conjugate was the major urinary metabolite and this was shown to be exclusively the (+)-(S)-enantiomerby chiral HPLC. Both ( - )-(R)-and ( + )-(S)-2-phenylpropionicacid were present in plasma after the administration of either antipode, and further evidence of the chiral inversion of both enantiomers was provided by the presence of some 25% of the opposite enantiomer in the free 2-phenylpropionicacid and its glucuronide excreted in urine after administration of (-)-(R)- and (+ )-(S)-2-phenylpropionic acid. The ( + )-(S)-enantiomerunderwent chiral inversion to the ( - )-(R)-antipodewhen incubated with dog hepatocytes. These data suggests that both enantiomers of 2-phenylpropionic acid are substrates for canine hepatic acyl CoA ligase(s) and thus undergo chiral inversion, but that the CoA thioester of only (+)-(S)-2-phenylpropionicacid is a substrate for the glycine N-acyl transferase. These studies are presently being extended to the structure and species specificity of the reverse inversion and amino acid conjugation of profen NSAIDs. 0 1992 Wiley-Liss, Inc.
KEY WORDS: arylpropionic acid, 2-phenylpropionic acid, glycine conjugation, stereoselectivity, chiral inversion, reverse chiral inversion, glycine N-acyl transferase, dog hepatocytes, chiral HPLC INTRODUCTION It is a well known phenomenon that the ( - )-(R)-enantiomers of 2-arylpropionic acids, the “profen” nonsteroidal antiinflammatory drugs(NSAIDs),are metabolically inverted to their ( + )-@-antipodes. The in vitro inhibitory activity of these agents against cyclooxygenase exclusively resides in the ( + )-(S)-enantiomers. Their ( - )-(R)-isomers, however, show pharmacological effects in vivo as a result of this unique chiral inversion reaction. Nakamura et al.4 proposed the chiral inversion mechanism of ibuprofen to be as follows (Fig. la):(l) ( - )-(R)-ibuprofenis stereospecifically activated to its CoA thioester by acyl CoA ligase(s)in microsomes and/or mitochondria, (2)(- )-(R)-ibuprofenyl-CoAis then epimerized by a cytosolic epimerase, and (3) the ( - )-(R)-and ( + )-(S)-CoAthioesters are hydrolyzed to yield free acids. Amino acid conjugation is one of the important metabolic reactions for compounds containing the carboxylic acid group and this proceeds by the mechanism shown in Figure lb5: (1) the carboxylic acid is transformed to its activated CoA thioester via an intermediate acyl adenylate and (2) the CoA thio@
1992 Wiley-Liss, Inc.
ester is conjugated with an amino acid by amino acid N-acyl transferases specific for the amino acid involved. Consideringthese mechanisms of chiral inversion and amino acid conjugation,one would predict that only the ( - )-(R)-enantiomers of 2-arylpropionic acids will give rise to amino acid conjugates. This hypothesis emerged from the detection of the taurine conjugate as a major urinary metabolite of loxoprofen sodium in dogs6 We have examined this problem systematically using 2-phenylpropionic acid, which is the simplest congener of the 2-arylpropionicacids and which has been reported to produce a glycine conjugate in dogs7 MATERIALS AND METHODS Chemicals 2-Phenylpropionicacid (2-PPA), its ( - )-(R)-and ( + )-(S)-enantiomers (optical purity W%),(+)-(R)-1-[naphthen-1-yllethylReceived for publication March 11, 1992; accepted May 20, 1992. Address reprint requests to Y. Tanaka, Analytical and Metabolic Research Laboratories, Sankyo Co.,Ltd., 1.2-58,Shinagawa-ku,Tokyo 140, Japan.
343
REVERSE CHIRAL INVERSION OF 2-PPA IN DOGS
(a) Chiral inversion mechanism : Acyl CoA ligase
(-)-R-IBU-COA Eplmerase
(+)-S-IBU
Hydrolase
(-)-R-i BU
+ *
(+)-S-IBU-CoA
(-)-R-IBU
(+)-S-IBU
Hydrolase
(b) Amino acid conjugation mechanism :
-
RCO-AMP RCO-SCoA -RCONHCH(R')COOH
RCOOH +ATP RCO-AMP + COASH RCO-SCoA + HSNCH(R')COOH
+ PP + AMP + CoASH
-
(c) From these two mechanisms, it would be predicted that: Amino acid conjugate (-)-R-profen NSAID (+)+profen
NSAID
+ D
Amino acid conjugate
Fig. 1. Proposed mechanisms for (a) chiral inversion" and (b) amino acid conj~gation.~ IBU, ibuprofen; NSAID, non-steroidal anti-inflammatory drug.
amine (NEA) with 99% optical purity, 1-hydroxybenzotriazole (HBT), dicyclohexylcarbodiimide (DCC), benzylthioacetic acid, 4-phenyl-n-butyricacid, and benzoic acid were all purchased from Tokyo Kasei Kogyo Co., Ltd. (Tokyo,Japan). Collagenase was obtained from Boehringer Mannheim GmbH (Germany).N[2-Hydroxyethyllpiperazine-N'-[2-ethanesulfonicacid] (HEPES) was from Sigma Chemical Co. (St. Louis, MO). All other reagents and solvents were of reagent grade. Glycine conjugates of (RS)- and ( + )-(S)-2-phenylpropionic acid were synthesized according to the procedure described by Dixon et a1.8 The identity of glycine conjugates synthesized and isolated from urine as described below were confirmed by mass spectrometry (MS). The optical purity of the glycine conjugate synthesized from (+)-(S)-2-PPAwas 99%. Animals and Dosing Three male beagle dogs (body weight W 1 2 kg) were used after fasting overnight. 2-Phenylpropionic acid dissolved in PEG-400/water (1:l)was orally administered by stomach tube at a dose of 200 mg/kg for the study of urinary metabolites and at a dose of 20 mg/kg for the plasma concentration study. Each enantiomer of 2-phenylpropionic acid was given at half the dose of the racemate. The animals were housed in metabolism cages to collect 24 h urine and feces, separately. Blood samples were taken from a foreleg vein into heparinized tubes at 10,20, 30 min, 1, 2, 4, and 6 h after dosing. Plasma was obtained by centrifugation. Urine and plasma were stored at - 20°C until analyzed. Isolation of Glycine Conjugate of 2-Phenylpropionic Acid Dog urine (50 ml) after oral administration of (-)-(R)-2phenylpropionic acid was adjusted to pH 2 with 4 M HCl and extracted with ethyl acetate (100 ml x 2). The extracts were concentrated to dryness in vacuo. The residue was subjected to preparative TLC purification: thin-layer, Kieselgel 60 F254 plate (Merck, thickness 0.25 mm); solvent system, benzene/ ethyl acetate/acetic acid (10:5:1). The zone of Rf 0.4 was
scraped off and extracted with methanol, and concentrated in vacuo. The sample obtained was further purified by semipreparative HPLC column, pBondapak C18 (Waters, 8 x 300 mm); mobile phase, methanol/water (50:50); flow rate, 4 ml/ min; detection, W 220 nm. Determination of Metabolites and Enantiomeric Composition Urine To urine (0.2 ml) were added 50 pg 4-phenyl-n-butyricacid (20 pl of 2.5 mg/ml solution)as an internal standard for the free acid and 500 pg benzoic acid (20 p1 of 25 mg/ml solution) for the glycine conjugate, followed by 0.8 m10.2 MHCl. The whole was extracted with 5 ml ethyl acetate on a vortex mixer, the layers separated by centrifugation, and the organic layer evaporated to dryness in vacuo. The residue was taken up in 0.5 ml methanol and analysed using HPLC system A for free 2-phenylpropionic acid and B for the glycine conjugate. The ester glucuronide of 2-PPA was isolated and identified as a major urinary product in several animal species, and hydrolyzed readily on warming with M NaOH for 1 h or incubating with P-glucuronidase at pH 5 Thus, the residual aqueous layer after extraction with ethyl acetate was mixed with 0.5 m12 M NaOH and incubated at 37°C for 30 min. To the reaction mixture were added 500 pg 4-phenyl-n-butyricacid and 0.5 ml 3 M HCl. The 2-phenylpropionic acid released was analysed by the same procedure as described above. The Calibration curves constructed for each metabolite were linear over the range G250 pg/ml. Quantitation was performed by reference to the calibration curves. The free and released 2-phenylpropionic acid and glycine conjugate were collected from the reversed-phase HPLC assay and the eluates were evaporated to dryness in vacuo. In the case of 2-phenylpropionic acid, the residue was treated with NEA, HBT, and DCC as described by Nagashima et al." to convert the acids to their respective ( + )-(R)-1-[naphthen-1-yllethylamides, which were then analysed using HPLC system C. Separation of the enantiomers of the glycine conjugate was achieved using HPLC system D without derivatization. *y9
344
TANAKA ET AL.
Plasma To specimens of plasma (0.1 ml) were added 5.0 pg benzylthioacetic acid (20 p1 of 250 mg/ml solution) as an internal standard and 0.9 m10.25 M HCl, and the whole was extracted with 4 ml hexane/ethyl acetate (6:4). After evaporation, the residue was derivatized and analysed as above.
B)HPLC system B
AIHPLC system A
Metabolism of ZPhenylpropionic Acid by Isolated Hepatocytes
Hepatocytes were isolated from male beagle dogs according to the technique of Moldeus et al.” Briefly, the dogs were anaesthetized by an intravenous injection of sodium pentobarbital (25 mg/kg). A small whole lobe of the liver was chosen and a vessel producing clearance of the maximum volume was cannulated with a tight-fitting cannula. The liver lobe was first perfused with a calcium-free Hanks’ buffer containing EGTA (0.5mM),NaHCO3 (25 mM), and HEPES (12.6 mM) at 37°C for 12 min at a pressure of 60-70 cm H20.The perfusion was then switched over to collagenase (0.1%) in Hanks’ buffer without EGTA but supplemented with CaCI2 (2.0 mM). After perfusing for 10 min, isolated cells were filtered through cotton gauze, washed three times by centrifugation, and resuspended in Krebs-Henseleit buffer containing HEPES (12.6 mM). Viability of the cells, as assessed by Trypan blue exclusion, was routinely greater than 85%. Incubations (4 x lo6 cells/ml) were carried out in KrebsHenseleit buffer containing HEPES in rotating 50 ml roundbottomed flasks at 37°C and were gassed continuously with 95% 02/5% COB.The final concentrationwas 0.1 mMand 0.05 mMfor the racemate and each enantiomer of 2-phenylpropionic acid, respectively. Aliquots (1.0 ml) of hepatocyte suspensions were taken at 0.25, 0.5, and 1 h after incubation, and centrifuged at 10,OOOrpm for 2 min. The supernatant obtained was analysed by the above method.
I
I
I
I
I
I
1
I
1
1
I
Fig. 2. HPLC profiles of urinary metabolites after oral administration of enantiomersof 2-phenylpropionic acid (2-PPA)to dogs. Internal standards (IS.) are 4-phenyl-n-butyricacid in HPLC system A for the detection of 2-phenylpropionic acid and benzoic acid in system B for the glycine conjugate. (a)Control urine, (b) urine of animals dosed with (-)-R-2-PPA and (c) urine of animals dosed with (+)-S-2-PPA.
High-pressure Liquid Chromatography (HPLC) The flow rate was 1.0 ml/min, except for 0.8 ml/min in firmed to be the unchanged free acid, and the other in system system C. Eluates were monitored by ultraviolet absorption at B. This was isolated from urine by successive procedures of 210 nm to detect 2-phenylpropionic acid and its glycine conju- ethyl acetate extraction, preparative TLC and HPLC. The retengate and at 282 nm to detect the (+)-(R)-1-[naphthen-1-ylle-tion time on HPLC and FAB/MS/MS (Fig. 3) of the metabolite thylamides. The other column and mobile phase conditions were identical to those of a synthetic sample of the glycine were as follows. A for 2-phenylpropionic acid column, ERC conjugate of 2-phenylpropionic acid. It was evident from the ODs-1282(ERMA CR. Inc., Tokyo, 8 x 250 mm); mobile phase, HPLC trace that the glycine conjugate was excreted as a major acetonitrile/O.l% trifluoroacetic acid (45:55). B for the glycine metabolite in urine after the administration of both enantiomconjugate of 2-phenylpropionic acid: column, Senshu PAK C8- ers of 2-phenylpropionic acid. 1202-P(Senshu Scientific Co., Ltd., Tokyo, 4.6 x 150 mm) linked Quantitation of Urinary Metabolites in series with ERC ODS-1282(8~250mm), mobile phase; The 24 h urinary metabolites were quantitated by HPLC acetonitrile/O.l% trifluoroacetic acid (22:78). C for ( + )-(R)-1[naphthen-1-yllethylamides:column, ERC Silica-l282(ERMA after oral administration of 2-phenylpropionicacid at a dose of CR. Inc., 8 x 250 mm); mobile phase, hexane/ethyl acetate(6:4). 200 mg/kg and its enantiomers at a dose of 100 mg/kg to dogs D for the enantiomers of the glycine conjugate: column, CHI- (Table 1). When given the ( - )-(R)-enantiomer, the main RALPAK WH (Daicel Chemical Industries Ltd., Tokyo, 4 x 250 metabolites were the conjugates with glucuronic acid and glycine which accounted for 23.8% and 21.4% of dose, respecmm); mobile phase, 0.3 mM copper sulfate. tively, in addition to the parent compound (11% of dose). A RESULTS similar excretion pattern was found after the (+)-(S)-enantiHPLC Profile of Urinary Metabolites omer: unchanged acid, 6.9%; glucuronic acid conjugate, 31.6%; Figure 2 shows the typical HPLC profile of urinary metabo- glycine conjugate, 23.1 YO.The glucuronide tended to be formed lites after oral administration of the enantiomers of 2-phenyl- in a slightly larger amount from the (+ )-(S)-isomerthan the propionic acid to dogs. The ethyl acetate-extractable fraction ( - )-(R)-antipode. Similarly, the racemate was primarily contained two peaks, one in HPLC system A which was con- metabolised to the glucuronic acid (40.3%)and glycine (20.6%)
345
REVERSE CHIRAL INVERSION OF 2-PPA IN DOGS
(+)-(S)-2-phenylpropionicacid, but not as to its ( - )-(R)-enantiomer. Consequently, the amount of glucuronide from both enantiomers was almost same at 60 min of incubation. The increase of glucuronide after 30 min of incubation may be partly contributed by the ( + )-(S)-enantiomer inverted from ( - )-(R)-2-PPA. The glycine conjugate was generated to similar extents from the ( + )-(S)-and ( - )-(R)-enantiomersof 2-phenylpropionicacid, suggesting no stereoselectivity in the conjugation reaction, compatible with in vivo data. However, the degree of glycine conjugate formation was about one-third that of the glucuronide in isolated hepatocytes, unlike the approximately equal excretion rates of the two conjugated metabolites in urine. The ( + )-(S)-enantiomer generated from ( - )-(R)-2-PPA could be used for the formation of glycine conjugate having (+)-(S)-configuration as described below.
a) Standard Sample 105
1111
I1
m/+
100
2 00
b) Isolated Sample
n
105
208
1
Enantiomeric Composition of ZPhenylpropionic Acid and Its Glycine Conjugate 76
100
200
m/z
Fig. 3. FARjMSIMS spectra of glycine conjugate of 2-phenylpropionicacid. (b) isolated sample from dog urine dosed with (-)-R-2-phenylpropionic acid. (a) Synthetic standard sample and
conjugates. These results revealed that there is no substrate stereoselectivity for the glycine conjugation of 2-phenylpropionic acid, which is a main metabolic route in dogs. Metabolism of ZPhenylpropionic Acid by Dog Hepatocytes The production of glucuronic acid and glycine conjugates from the enantiomers of 2-phenylpropionic acid was investigated using freshly isolated dog hepatocytes (Fig. 4).The formation of glucuronide was stereoselective for the ( + )-@)-enantiomer up to 30 min of incubation. It corresponds to the tendency of a larger urinary excretion rate of glucuronide from the ( + )-(S)-enantiomerthan the ( - )-(R)-isomer.After 30 min, however, a plateau was reached as to glucuronide formation of TABLE 1. Urinary metabolites after oral administrationof 2-phenylpropionicacid and its enantiomers to dogs
The enantiomeric composition of 2-phenylpropionic acid as determined by normal phase HPLC of the ( + )-(R)-1-[naphthen1-yllethylamides. The enantiomers of glycine conjugate of 2-phenylpropionicacid were quite well separated by a chiral HPLC using a ligand-exchangecolumn CHIRALPAK WE (Fig. 5). Those values obtained are summarized in Table 2. The enantiomeric ratios of free and glucuronidated 2-phenylpropionic acid after oral administration of the ( - )-(R)-enantiomer were essentially identical [( +)-(S)/(- )-(R) ratios 29:71, free; 33:67, glucuronide]. Comparable data were obtained after the (+)-(S)-enantiomer,the (+ )-(S)/(- )-(R) ratios being 77:23 for free and 78:22 for the glucuronide. Taken together, these results indicate the occurrence of an unusual and extensive inversion from ( + )-(S)-to ( - )-(R)-2-phenylpropionicacid. The stereochemistry of the glycine conjugate is very interesting in that while it is formed following the administration of both enantiomers of 2-phenylpropionic acid, chiral HPLC showed clearly that the glycine conjugate has the ( + )-(S)-configuration exclusively. The enantiomeric compositions in the in vitro experiments using isolated hepatocytes are given in Table 2. The ( + )-(S)/(- )-(R) ratios of the residual and glucuronidated acid were 19:81 and 20:80, respectively, when the ( - )-(R)-enanti-
TABLE 2. Enantiomeric composition of free, glucuronic acid and glycine conjugate produced in urine and hepatocytes from enantiomers of 2-phenylpropionicacid a
+
( )-(S)/( - )-(R) ratio
YOof Dose Metabolites
Racemateb
Unchanged Glucuronide Glycinate
5.8 40.3 20.6
( + )-(SIC
6.9 f 1.5 31.6 f 4.0 23.1 f2.7
(- )-(RY
10.6f2.2 23.8 f 2.9 21.4f 3.8
"Racemate was administered at a dose of 200 mg/kg and its enantiomers at 100 mg/kg. Twenty-four hour urine was analysed. *Mean of two animals. cMean f SE of three animals.
Sample Urine Hepatocytes
Used enantiomer
Free
Glucuronide
Glycinate
(-)-(R) (+)-(S) (-)-(R) (+)-(S)
29171 7723 19:81 46:54
33:67 78:22 20 :80 83:17
( + 1-6) ( + )-(S) ( + )-(S) ( + )-(S)
"The 24 h urine samples in Table 1 and 60 min incubation mixtures in Figure 4 were analysed. b( + )-(S)-Enantiomerwas exclusively detected.
346
TANAKA ET AL.
A) GI u c
ur o n i d e
10
5
.I
BIGlycine c o n j u g a t e
;-..--4
-0
/ / -/ --
I
0
15
Racemate (4-R-2 -PPA
A (+)-S- 2- PPA
I
60
30 Incubation time (min)
Fig. 4. Formation of (A) glucuronide and (B) glycine conjugate from racemic (O,.), (+)-S- ( A , A ) and (-)-R(O,l, enantiomers of 2-phenylpropionic acid (2-PPA) by isolated dog hepatocytes. Symbols represent mean of three experiments.
omer was employed as a substrate. The corresponding ratios were 46:54 and 83:17, when the (+)-(S)-isomerwas incubated. Again, only the glycine conjugate of ( + )-(S)-2-phenylpropionic acid was detected by the chiral HPLC. Thus, the reverse chiral inversion and the stereospecific formation of glycine conjugate mentioned before were confirmed to take place in hepatocytes. Plasma Concentration
The racemate (20 mg/kg) and enantiomers (10 mg/kg) of 2-phenylpropionicacid were orally administered to dogs. The plasma concentration-time curves of each enantiomer are shown in Figure 6. The peak levels of each enantiomer were at 30 min with a slightly, but not significantly, higher concentration of the ( + )-(S)-isomer.After administration of the racemate, the areas under the plasma concentration-time curves (AUC)of the ( + )-(S)-and ( - )-(R)-enantiomerswere 74.2 f 9.3 and 73.1 f 4.3 pgh/ml, respectively, which were not significantly different. This might suggest that a chiral inversion apparently does not occur with 2-phenylpropionic acid in dogs. Nevertheless, the ( t )-('$isomer gradually appeared in the plasma dosed with the ( - )-(R)-2-phenylpropionicacid and reached to a maximum level around 1-2 h later. The AUC of ( + )-6)-enantiomer
newly generated was 18.5f 7.6 pgh/ml, which corresponded to about 40% of that of (- )-(R)-enantiomer(44.1 f 6.6 pgh/ml) administered originally. In addition, the reverse phenomenon was observed after administration of ( + )-(S)-2-phenylpropionic acid: the antipode having ( - )-@)-configurationappeared in the plasma with the peak time at 1 h and the AUC was 29.1 f5.4 pgh/ml corresponding to 70% of the AUC of (+ )-(S)-enantiomer itself (41.2 f 10.3 pgh/ml). Judged from these facts, it is almost certain that both the usual and unusual chiral inversions occur when 2-phenylpropionic acid is given as a single enantiomer to dogs. DISCUSSION
The metabolism of 2-phenylpropionicacid in dogs was first examined by Kay and Raper,7 who found that the racemate gave rise to an ester glucuronide and a glycine conjugate. They also showed the isolated glycine conjugate was dextrorotatory. In the present experiments, the glycine conjugate was isolated from urine of dogs given with the ( - )-(R)- and ( + )-@)-enantiomers and the racemate, and confirmed by FAB/MS/MS. The enantiomeric composition of the amino acid conjugate was directly determined by chiral HPLC using a ligand exchange
347
REVERSE CHIRAL INVERSION OF 2-PPA IN DOGS 30
r
aIStandard Glycine conj.
b)(-)-R-2-PPA PO Glycine conj.
A
0
25
1
2
3
4 Time (hr)
5
6
1
2
3 4 Time (hr)
5
6
r
c)(+)-S-~-PPA PO Glycine conj.
Fig. 5. Enantiomeric separation of glycine conjugate of 2-phenylpropionic acid (2-PPA)by chiral HPLC. Column was CHIRALPAK WH and mobile phase was 0.3rnM CuSO4.
column, CHIRALPAK WH. It was revealed that the glycine conjugate excreted as a major metabolite was exclusively the ( + )-(S)-enantiomerafter administration of both enantiomers of 2-phenylpropionic acid. Carboxylic acids must be activated to CoA thioester intermediate during amino acid conjugation pathway, The formation of a glycine conjugate having ( + )-(S)-configurationfrom ( - )-(R)-2-phenylpropionicacid is consistent with the formation of the CoA thioester of the ( + )-(S)-enantiomerduring the chiral inversion reaction. However, it is difficult to explain the production of glycine conjugate of the ( + )-(S)-enantiomerafter dosing ( + )-(S)-Zphenylpropionic acid, because the (+)-(S)-isomers do not serve as a substrate for acyl CoA ligase(s)according to the accepted inversion mechanism. The conflict was thought to be resolved by conversion of the ( +)-(S)-enantiomerof 2-phenylpropionic acid to its CoA thioester intermediate in dogs. Once the CoA thioester of the (+)-(S)-enantiomer is formed it will be converted to the ( - )-(R)-enantiomerby the catalysis of racemase. In this connection, the enantiomeric composition of 2-phenylpropionic acid was analysed. The appearance of the (+ )-(S)-enantiomerin plasma after administration of the (- )-(R)-isomer is the result of chiral inversion. In addition to the usual R+S inversion, a reverse inversion (S+R) was shown by the detection of the ( - )-(R)-isomerin plasma when the ( + )-(S)-antipodewas administered. The opposite enantiomers had a delayed peak time around 1-2 h after oral dosing compared to the administered isomers in the same manner as ibuprofen.12 After the peak time, the plasma concentrations of each enantiomer were approximately the same following the administration of both the
0
Fig. 6. Plasma concentrations of enantiomers after oral administration of racemic (2hg/kg), ( - )-R. and ( + )-S-enantiomers(lOmg/kg) of 2-phenylpropionic acid (2-PPA) to dogs.
+
( )-(S)-and ( - )-(R)-enantiomers.This suggests an equilibrium
state between the enantiomers, i.e., a racemization of each. The chiral inversion was apparently not observed after administration of the racemate, possibly due to the similar extent of inversion to both directions. Consequently, it may be said that the enantiomers of 2-phenylpropionic acid racemize in dogs. The analysis of enantiomeric composition was also performed with respect to the free and glucuronidated 2-phenylpropionic acid in urine. The opposite enantiomer was present in the free acid to the extent of 29% after the ( - )-(R)- and 23% after the ( + )-(S)-enantiomer.These compositions were substantially reflected in those of their glucuronides. Thus, both directions of inversion, from ( - )-(R)- to ( + )-(S)- and from ( + )-(S)- to ( - )-(R)-2-phenylpropionicacid, occur in dogs. There have been debates, recently reviewed by Williams,l3 as to the organ(s) responsible for the chiral inversion of profen NSAIDs. Freshly isolated dog hepatocyes were used to examine the metabolism of 2-phenylpropionic acid. The results obtained were essentially similar to those in vivo: the stereochemistry of glycine and glucuronic acid conjugations and both
348
TANAKA ET AL.
S-(+)-2-PPA Glucuronlde
1.
S-(+)-2-PPA
R-(-)-2-PPA
__F
1
-
JS-(+)-2-ppA
S-(+)-2-PPA-SCoA
It
R-(-)-2-PPA-SCoA
S-(+)-2-PPA Glyclnate
R-(-)-2-PPA Glyclnate
\ R-(-)-2 - P PA
R-(-)-2-PPA Glucuronide Fig. 7. Possible metabolic mechanisms of enantiomers of 2-phenylpropionic acid (2-PPA) in dogs,
directional inversions of 2-phenylpropionic acid were reproduced, indicating the liver is a main site of these metabolic reactions. Based on the combined results, the stereoselective metabolism of the enantiomers of 2-phenylpropionic acid in dogs is proposed in Figure 7. Both enantiomers are converted to CoA thioesters, which then inverted to the respective antipode by the action of an epimerase. However, only the thioester of (+)-(S)-2-phenylpropionicacid is a substrate for glycine N-acyl transferase to form the glycine conjugate stereospecifically . The metabolic chiral inversion is usually thought of as unidirectional. But, the reverse inversion, from (+)-(S) to ( - )-(R), has been suggested for 2-phenylpropionicacid in rats by Hutt et al.I4 and Fournel et al.,9 who reported small but significant amounts of the (-)-(R)-enantiomer was excreted in urine after administration of the ( + )-(S)-enantiomer.The present report clearly revealed the occurrence of this reverse reaction in dogs by the detection of the (- )@)-enantiomer in plasma dosed with ( + )-(S)-2-phenylpropionicacid. However, there was a marked species differences in the reverse inversion of 2-phenylpropionic acid, which was found in the mouse but not in the rat, guinea pig, and rabbit (unpublished data). Fournel et al.9 showed that the chiral inversion was found to occur after administration of racemic 2-phenylpropionicacid to the rat and rabbit, but not in the mouse. The species differences reported by Fournel et al. may be explained on the basis of the reverse inversion shown here. The reason for the apparent absence of chiral inversion of racemic 2-phenylpropionicacid in the mouse may be due to the active reverse reaction in this species. Recently Chen et al. reported that guinea pigs could effect the inversion of both (- )-(R)-and (+)-(S)-ibuprofenwith nearly the same efficiency, and that ( + )-(S)-ibuprofenmetabolism gave rise to small but significant levels of (-)-(R)-antipode in both rats and rabbits. l5 The (+)-(S)-enantiomerof ketoprofen was also inverted to the (- )-(R)-antipodein the mouse (unpublished data), but not in the rat according to Iwakawa et a1.I6 The reverse chiral inversion of ( + )-(S)-ZPPA,however, was not shown in guinea pig as mentioned previously. Since there seems to be a large difference in this reverse inversion as to the species and chemical structures, we are now investigating these problems in vivo and in vitro. ACKNOWLEDGMENTS
The authors thank Dr. Kinoshita, in our laboratories, for the mass spectromeric analyses.
LITERATURE CITED 1. Hutt, A.J., Caldwell, J. The metabolic chiral inversion of 2-arylpropionic acids-a novel route with pharmacological consequences. J. Pharm.Pharmacol. 35693-704, 1983. 2. Hutt, A.J., Caldwell, J. The importance of stereochemistry in the clinical pharmacokinetics of the 2-arylpropionic acid nonsteroidal anti-inflammatory drugs. Clin. Pharmacokinet. 9371-373, 1984. 3. Caldwell, J., Hutt A.J., Fournel-Gileux,S. The metabolic chiral inversion and dispositional enantioselectivity of the 2-arylpropionicacids and their biological consequences. Biochem. Pharmacol. 37:105-114,1988. 4. Nakamura, Y., Yamaguchi, T., Takahashi, S., Hashimoto, S., Iwatani, K., Nakagawa, Y. Optical isomerisation mechanism of ( - )-R-hydratropic acid derivatives. J. Pharmacobio-Dyn. 4S1, 1981. 5. Caldwell. J. Structure-metabolism relationships in amino acid conjugation. In: Conjugation Reactions in Drug Biotransformation. Aitio, A,, ed. Amsterdam: Elsevier/North Holland, 1978 111-120. 6. Tanaka, Y., Nishikawa, Y., Hayashi R. Species differences in metabolism of sodium 2-[4-(2-oxocyclopentylmethyl)phenyllpropionatedihydrate (Loxoprofen sodium), a new anti-inflammatory agent. Chem. Pharm. Bull. 31: 365G3664.1983. 7. Kay, H.D., Raper, H.S. XXXVII. The mode of oxidation of fatty acids with branched chains. 11. The fate in the body of hydratropic, tropic, atrolactic and atropic acids together with phenylacetaldehyde. Biochem. J. 16465474, 1922. 8. Dixon, P.A.F., Caldwell, J., Smith, R.L. Metabolism of arylacetic acids 2. The fate of [ 14C]hydratropicacid and its variation with species. Xenobiotica 7~707-715,1977. 9. Fournel, S.,Caldwell, J. The metabolic chiral inversion of 2-phenylpropionic acid in rat, mouse and rabbit. Biochem. Pharmacol. 35:415%4159, 1986. 10. Nagashima, H., Tanaka, Y., Watanabe, H., Hayashi, R., Kawada, K. Optical inversion of (2R)- to (%)-isomers of 2-[4-(2-oxocyclopentylmethyl)phenyl]propionic acid (Loxoprofen), a new antiinflammatory agent, and its monohydroxy metabolites in the rat. Chem. Pharm. Bull. 32251-257,1984. 11. Moldeus, P.,Hogberg, J., Orrenius, S. Isolation and use of liver cells. Methods Enzymol. 51:&71, 1978. 12. Beck, W.S., Geisslinger, G., Engler H., Brune, K. Pharmacokinetics of ibuprofen enantiomers in dogs. Chirality 3:165-169, 1991. 13. Williams, K.M. Enantiomers in arthritic disorders. Pharmacol. Ther. 46272295,1990. 14. Hutt, A.J., Fournel, S.. Caldwell, J. Application of a radial compression column to the high-performance liquid chromatographic separation of the enantiomers of some 2-arylpropionic acids a s their diastereomeric ( - )-Sl(naphthen-1-y1)ethylamides. J. Chromatogr. 378:40%418, 1986. 15. Chen, C.-S., Shieh, W.-R., Lu,P.-H., Harriman, S. and Chen, C.-Y. Metabolic stereoisomeric inversion of ibuprofen in mammals. Biochim. Biophys. Acta. 1078411417, 1991. 16. Iwakawa, S., He, X., Hashimoto, S., Volland. C., Benet, L.Z. and Lin, E.T. Stereoselective disposition of ketoprofen in rats. Drug Metab. Dispos. 1 9 717-718, 1991.
