Chem. Res. Toxicol. 1995,8, 642-648
Chloramphenicol Oxamylethanolamine as an End Product of Chloramphenicol Metabolism in Rat and Humans: Evidence for the Formation of a Phospholipid Adduct Jean Pierre Cravedi,” Elisabeth Perdu-Durand, Maryse Baradat, Jacques Alary, Laurent Debrauwer, and Georges Bories Laboratoire des Xknobiotiques, INRA, B.P. 3, 31931 Toulouse Chdex, France Received January 9, 1995@
Chloramphenicol (CP) has been implicated as, although not proven to be, a causative agent of aplastic anemia in humans. Recent studies from our laboratory have presented evidence that CP-oxamylethanolamine was a n end product of CP biotransformation in birds. Because this novel metabolic pathway has never been reported in other species, we have now expanded these investigations to rat and humans. f3H1CPwas administered PO (10 mgkg) to adult male Wistar rats and to a human volunteer. Urine was collected and analyzed by HPLC and GCMS for CP metabolite determination. In rat, the two most important metabolites in 0-24 h urine were CP-base and CP-acetylarylamine which together accounted for about 50% of the ingested radioactivity. The remainder was due to unchanged CP, CP-oxamic acid, CP-alcohol, CP-glucuronide, and CP-oxamylethanolamine. The presence of these end products was also demonstrated in man. CP-oxamylethanolamine represented 0.74% and 1.37% of the ingested radioactivity in rat and human urine samples, respectively. CP-oxamylethanolamine formation was confirmed in vitro with isolated rat hepatocytes, suggesting the involvement of liver in the production of this metabolite. The origin of CP-oxamylethanolamine has been investigated with the use of hepatic liver microsomes from phenobarbital-treated rats. The incubation of [3H]CP with this subcellular fraction led to the binding of a radiolabeled compound to the microsomal lipids, whereas no binding occurred when CP-oxamic acid was incubated with the microsomes. Enzymatic hydrolysis of the microsome lipid fraction with phospholipase D from Streptomyces chromofiscus released CP-oxamylethanolamine. These data are consistent with a mechanism that involves an initial activation of CP t o CP-acyl chloride by cytochrome P450 dependent monooxygenases, a subsequent covalent binding of this reactive intermediate metabolite with phosphatidylethanolamine present in the microsomal membrane, followed by the breakdown of the phospholipid adduct and the elimination of CP-oxamylethanolamine in urine.
Introduction The broad spectrum antibiotic chloramphenicol (CP),’ D-(-)-threo-2-(dichloroacetamido)-l-(p-nitrophenyl~-l,3propanediol, has been used extensively in the practice of both human and veterinary medicine. However, toxic side effects in humans have been reported (1,2),resulting in a restriction on CP prescription. Nevertheless, CP remains the drug of choice in the treatment of bacterial meningitis in young children and of fulminating enteric fever caused by Salmonella typhi or Salmonella paratyphi (3). In addition to “gray baby syndrome”due to mitochondrial electron transport blocking and occurring at a dosage in excess of 100 mg/kg/day (41, CP produces two types of adverse reactions regarding bone marrow tissue: a dose-related reversible suppression of erythropoiesis due to the inhibition of mitochondrial protein synthesis (5)and a rare dose-independentidiosyncratic response resulting in aplastic anemia (6, 7).
* To whom correspondence should be addressed. @Abstractpublished in Advance ACS Abstracts, May 15, 1995. Abbreviations: CP, chloramphenicol; CP-oxamylethanolamine, chloramphenicol oxamylethanolamine;CP-base, chloramphenicol base; CP-acetylarylamine, chloramphenicol acetylarylamine; CP-oxamic acid, chloramphenicol oxamic acid; CP-alcohol, chloramphenicol alcohol; CPglucuronide, chloramphenicol glucuronide; CP-acyl chloride, chloramphenicol acyl chloride.
Ri
0E - y w x 2 -
R3
R2
Figure 1. Structural formulas of chloramphenicol (CP) and main metabolites. CP: R1 = NO2, Rz = NHCOCHC12, and R3 = OH. CP-acetylarylamine: R1= NHCOCH3. CP-oxamicacid: Rz = NHCOCOOH. CP-alcohol: Rz = NHCOCHzOH. CP-base: Rz = NH2. CP-oxamylethanolamine:Rz = NHCOCONHCHzCH2OH. CP-glucuronide: RB= 0 C & 1 0 6 .
Although the mechanism inducing this rare and fatal anemia is still unknown, it has been suggested that the nitro group in CP may be reduced in vivo to a nitroso group, resulting in a compound shown to bind irreversibly (8)and to be toxic (9, 10)to hematopoietic stem cells in vitro. Another hypothesis was proposed by Pohl and Krishna (11)and Halpert et al. (12),who showed in vitro and in vivo covalent binding of CP to rat liver microsomal proteins after activation of CP into an oxamyl chloride. The end product of the dechlorination process, Le., CPoxamic acid (Figure l),was isolated from the urine of various species including man (23). These data suggest that detailed knowledge of the metabolic fate of CP may contribute t o a better understanding of its toxic properties. The main in vivo metabolic pathways of CP in mammals were initially
0893-228~f95I27Q8-0642$Q9.QOlQ 0 1995 American Chemical Society
Phospholipid Adduct of Chloramphenicol Metabolites
Chem. Res. Toxicol., Vol. 8, No. 5, 1995 643
225
1
Figure 2. Mass spectrum (EI) of trimethylsilylated CP-oxamylethanolaminestandard (A, full spectrum) and correspondingmetabolite isolated from urine of rat treated with CP (B, single ion monitoring technique).
described by the Glazko group (14-16) 40 years ago and then completed by several authors during the last two decades (see ref 17 for review). These studies gave evidence for CP, CP-base, CP-alcohol, CP-oxamic acid, and CP-glucuronide (Figure 1) as end metabolites in urine and demonstrated an extensive biotransformation of the drug. Recently, we identified CP-oxamylethanolamine in excreta of duck treated with CP (18), but the origin of this end product remained unclear. The aim of this study was to investigate whether this new metabolite was present in rat and human urine and to determine the biotransformation pathway leading to this unexpected compound.
Experimental Section Chemicals. [rzng-3,5-3HlChloramphenicol was purchased from NEN (Dupont NEN Research Products, Les Ulis, France). The drug, which had a specific activity of 1184 GBq/mmol and was at least 99% pure based on HPLC and TLC analysis, was diluted with unlabeled CP (Sigma Chemical Co., St Louis, MO) to achieve the required dose. (1R, 2R)-l-(4-Nitrophenyl)-2amin0-1,3-propanediol(CP-base) and CP-glucuronide were obtained from Sigma. CP-oxamic acid and its methyl ester were synthesized according to Pohl et al. (19).N-Chloramphenicol oxamylethanolamine was synthesized from chloramphenicol oxamic acid methyl ester by the addition of ethanolamine and with the use of sodium methoxide as base (20). The identity of CP-oxamylethanolamine was confirmed by GC-MS as TMS derivative (Figure 2A). CP-alcohol was synthesized as previously described (21). In Vivo Experiments. Three male Wistar rats weighing 200 g and placed in metabolic cages were force-fed a single dose of
200 kBq of [3H]CP mixed with 10 mg of nonradioactive CP in 500 pL of propylene glycol. Urine was collected for 24 h, and an aliquot was taken for radioactivity counting. Samples were pooled, and an aliquot (100pL) was directly submitted to HPLC analysis for total metabolic profile. Metabolites from the remaining part were extracted with a Sep Pak C18 cartridge (22). Analysis of the methanol eluate was carried out by HPLC as described below. One human volunteer ingested a single dose of 500 mg (9.25MBq) of L3H1CPin a gelatin capsule. Urine was collected for 8 h and treated as indicated for the rat.
Preparation of Hepatocytes and Incubation with [SHICP. Isolated hepatocytes were prepared from adult male W i s h rats (200-250 g) anesthetized with pentobarbital, by the twostep method of perfusion with collagenase A (Boehringer Mannheim, Meylan, France) described by Seglen (23). They were then washed and resuspended in Dulbecco's modified Eagle's medium (GIBCO BRL, Cergy-Pontoise, France). Viability was assessed by the trypan blue exclusion test, and the hepatocytes were diluted to give a final concentration of 3 x lo6viable cells/ mL. Incubations (final volume 3 mL) were carried out at 37 "C on 6-well plates (Nunclon, Roskilde, Denmark) with constant gyratory shaking (ca. 120 rpm). After a 5 min preincubation period, 10 kBq of L3H1CP (550 pg) was added in the medium in 5 pL of methanol. After 120 min, incubations were terminated by centrifugation (80gfor 5 min) and removal of the supernatant fluid. A 10 pL aliquot of the supernatant was assayed for radioactivity counting . The remaining supernatant was stored frozen at -20 "C until HPLC analysis. Microsomal Incubation Treatment. Adult male Wistar rats (150-250 g) were pretreated with phenobarbital (80 mpl kg in saline, ip) 72,48, and 24 h before the experiment. The rats were killed by decapitation, and the livers were removed immediately and washed in 0.9% saline. Microsomes were prepared as previously described (24)and stored frozen at -80
644 Chem. Res. Toxicol., Vol. 8, No. 5, 1995
I
Cravedi et al.
I
MICROSOMAL INCUBATIONS
CENTRIFUGATION
(SFCC, Neuilly-Plaisance, France) with a Kontron 420 solvent delivery system and a Spectra-Physics (Fremont, CA) Model W 100 detector set at 280 nm and connected with a 3390 integrator (Hewlett-Packard, Palo Alto, CA). A n isocratic elution was run at 0.8 mumin with the following mobile phase: methanol and 0.1 M ammonium acetate buffer adjusted at pH 4.4 (20:80 v/v). In these chromatographic conditions, the recovery rate of L3H1CP-derived radioactivity was a t least 90%. Radioactivity profiles were obtained by coupling the instrument with a Model 202 microcollector (Gilson Medical Electronics, Middleton, WI) and counting the collected fractions in a liquid scintillation counter. GC-MS analyses were obtained on a Nermag R. 10.lO.H (Delsi Nermag Instruments, Argenteuil, France) single quadrupole mass spectrometer coupled to a Delsi DI 200 (Delsi Nermag Instruments) gas chromatograph fitted with a BPX5 25 m x 0.22 mm i.d. x 0.25 pm capillary column (SGE, Villeneuve-Saint-Georges, France). The samples were injected in the splitless mode. Instrumental conditions were as previously described (18).For radioactivity measurement, solutions were mixed with Ultima Gold scintillator (Packard Instrument Co., Downers Grove, IL) and counted directly in a Packard Tricarb 2200CA liquid scintillation counter. Radioactivity was assessed using external quench correction. Pelleted material was combusted in a Packard Tricard Model 306 oxidizer prior to tritium measurement by liquid scintillation counting in an appropriate cocktail (Monophase S, Packard Instrument Co.).
[ SUPERNATANT + < I
3Hcounting
HPLC
3Hcounting
CHCl3 / CH 3 OH (2 : 1, V/V) EXTRACTION
3Hcounting
1 3Hcounting UNEXTRACTABLE
PHOSPHOLIPASE D HYDROLYSIS
I I 4 4
3Hcounting
HPLC
+
GC-MS
Figure 3. Flow diagram for microsomal incubation treatment. HPLC and GC-MS analyses were as described in the Experimental Section. "C in 0.1 M phosphate buffer (pH 7.4) containing 20% glycerol. The concentration of protein was determined by the method of Lowry et al. (25). Incubations were performed in triplicate. Each incubation vial usually contained 5 mg of microsomal protein, 3 mM NADP, 7.5 mM glucose 6-phosphate, 25 mM nicotinamide, 20 mM MgC12,2.6 units of glucose-6-phosphate dehydrogenase, and 100 pM r3H1CP (518.5 kBq) in a final volume of 1 mL of 0.1 M phosphate buffer (pH 7.4). The mixture was incubated for 1 h at 37 "C, and the reaction was stopped by freezing (-20 "C).As shown in Figure 3, the reaction mixture was centrifuged (106000g, 1h) and the resulting microsomal pellet was extracted twice with chloroform-methanol (2:l vh). Saline (0.9%) (0.2 volume) was added to the pooled extracts, and the samples were vigorously shaken on a Vortex mixer before centrifugation at 2500g for 10 min. An aliquot of the organic phase was stored frozen (-20 "C) until HPLC analysis. Half of the remaining fraction was evaporated, and the residues were dissolved in 2 mL of Tris-HC1 buffer (0.1 M., pH 8) and treated with phospholipase D from Streptomyces chromofuscus (type VI, Sigma) as described by Imamura and Horiuchi (26).The remaining half was evaporated and dissolved in 1 mL of 0.2 M acetate buffer (pH 5.6) and treated with phospholipase D from cabbage (type I, Sigma) as indicated by Kates (27). In both cases, the hydrolysis was stopped after 2 h by methanol addition (1 volume). The reaction mixture was centrifuged and the supernatant analyzed by HPLC. Identification of the end products was obtained by GC-MS. To monitor recovery after each step, aliquots of the various fractions obtained during this procedure were transferred to scintillation vials for radioactivity counting. The radioactivity associated with microsomes and microsomal proteins was measured on aliquots after total combustion of the samples as described below. Instrumentsand Analytical Methods. The HPLC system consisted of a 250 x 4.6 mm ODs2 (5 pm) reversed phase column
Results
In man, excretion of the radioactivity following a single oral dose of I3H1CPoccurred rapidly in urine. After 8 h, 65% of the ingested 3H was excreted via this route. This percentage represented only 32% after 24 h in rat (data not shown). The HPLC metabolic profiles obtained from rat and human urine showed seven radioactive peaks (Figure 4A,B). Their retention times corresponded to standard CP-oxamic acid, CP-base, CP-alcohol, CP-acetylarylamine, CP-oxamylethanolamine, CP-glucuronide, and CP. As previously described (18,211their identity was confirmed by electron impact (EI)GC-MS analysis as TMS derivatives for CP-oxamic acid, CP-base, CP-alcohol, CP-acetylarylamine, and CP, and by negative FAJ3-MS analysis for CP-glucuronide. CP-oxamylethanolamine accounted for 0.74% and 1.37% of the ingested radioactivity in rat and human urine samples, respectively. Because this metabolite was present in low amount, it was not possible to obtain a full mass spectrum of this compound. In addition to the molecular ion, selected ion monitoring (SIM) analysis (Figure 2B) revealed the presence of mass fragments with m Jz 391,319,229,225, and 208 at the GC retention time of standard TMS CPoxamylethanolamine derivative. Initial viabilities of isolated rat hepatocyte preparations, as estimated by the Trypan blue exclusion test, averaged 82%. During the 2 h incubation time, no significant changes in viability were detected. Part C of Figure 4 shows the metabolic profile obtained from the hepatocyte incubates after 2 h. On the basis of chromatographic behavior and MS analysis, and compared with standards, the metabolites were identified as CPoxamic acid, CP-base, CP-oxamylethanolamine, CP-glucuronide, and CP. Traces of CP-alcohol,representing ca. 1% of the total radioactivity, were detected in the incubates. CP-oxamylethanolamine amounted to 1.2% of the total radioactivity. The incubation of [3H]CPwith microsomes from phenobarbital-pretreated rats resulted in a partial biotrans-
Phospholipid Adduct of Chloramphenicol Metabolites 100-
4
3 ?
A
0-
m
A
P
E
v
t
t
L L, 5 0 -
2
1 '
0 0
10
20 T I M E ( min )
30
20 T I M E (min)
10
40
30
40
30
40
30
40
100
B
h
loo[ 6
0
B
m
v
7
0 0
10
20 T I M E (min)
30
6
C
n
20 T I M E (min)
10
40
C 2 L
5I
A 0
0
10
20 T I M E (min)
30
40
Figure 4. Typical HPLC profiles of r3H1CP metabolites from 0-24 h rat urine (A), 0-8 h human urine (B), and rat hepatocyte incubates (C). HPLC system was as described in the Experimental Section. Peak 1was identified as CP-oxamic acid, 2 as CP-base, 3 as CP-alcohol, 4 as CP-acetylarylamine, 5 as CPoxamylethanolamine, 6 as CP-glucuronide,and 7 as unchanged
CP.
formation of the antibiotic. The HPLC profile of the supernatant fraction (Figure 5A) exhibited two major metabolites in addition to unchanged L3H1CP: a polar compound having the retention time of CP-oxamic acid standard, and a metabolite having a retention time of 21 min which corresponded to the chromatographic behavior of p-nitrobenzyl alcohol standard. No further identification steps were carried out since this result has previously been described (28). ARer incubation, the radioactivity associated with microsomes amounted to 9.1% of the total radioactivity. This percentage corresponded to 3.8% extractable with
10
20 T I M E (min)
Figure 5. HPLC profiles of supernatant fraction of 1 h incubation mixture of [3H]CP with liver microsomes from phenobarbital-treated rats (A), and corresponding lipid extract from the pelleted microsomes before (B) and after (C) enzymatic hydrolysis with phospholipase D from S. chromofiscus.
chloroform, 4.0% present in the methanol phase, and 1.3%unextractable, and probably bound to microsomal proteins. When the incubations were conducted with radiolabeled CP-oxamic acid, the radioactivity linked to microsomes was less than 1%. The extraction of the lipids from the microsome pellet (chloroform extract) and subsequent HPLC analysis revealed two fractions of radioactivity associated with unchanged CP and an unknown metabolite having a retention time of 23.5 min (Figure 5B). Hydrolysis of the extracted microsomal lipids with phospholipase D from S. chromofiscus resulted in a compound having the same HPLC retention time as CP-oxamylethanolamine(Figure 5C) and representing 0.4% of the radioactivity used for microsomal incubations. Confirmation of the structure of this com-
Cravedi et al.
646 Chem. Res. Toxicol., Vol. 8, No. 5, 1995
t OH
I
CHzOH
OH
t
I
II I
I
0 CI
HC1
phosphatidylethanolamine from microsomal membrane
CP-oxamvlethanolamine Figure 6. Proposed metabolic pathways giving rise to CP-oxamylethanolamine as end product excreted in urine. pound was given by E1 GC-MS (data not shown). We failed to obtain CP-oxamylethanolamine or a related compound when hydrolysis was carried out with phospholipase D from cabbage (not shown), suggesting that the steric hindrance imposed by the aromatic moiety present in the phosphatidylethanolamine adduct of CP metabolite impedes the action of phospholipase D from cabbage but not from bacteria. Similar observations had also been reported earlier for N-(kazidobenzoyl) phosphatidylethanolamine (29).
Discussion Although the biotransformation of CP in animals and humans-involving the propanediol chain conjugation, the reduction of the nitrophenol moiety, the dehalogenation of the chloroacetamide group, and the hydrolysis of the amide bound-was widely studied (17), the link between these metabolic pathways and the toxic effects of C P such as aplastic anemia has never been clearly established. As far as [3HlCPurinary metabolites were
concerned, a few percent of the radioactivity eliminated by this route in rat and man remained unexplained and needed further identification (12,21,30). Recently, new metabolic pathways of CP were discovered in the duck (18)giving rise to end products such as CP-oxamylethanolamine, and the presence of this metabolite in other species was questioned. Our results unambiguously demonstrate that CPoxamylethanolamine was excreted in urine of rat and man treated per os with CP. This finding raised the question of the metabolic pathway leading to this compound as well as the localization of these biochemical reactions. Although the direct conjugation of CP-oxamic acid with free endogenous ethanolamine cannot be totally excluded, this hypothesis seems unlikely since this phase I1 pathway has never been described in xenobiotic metabolism studies. Moreover, in a preliminary studf we failed in our laboratory to form CP-oxamylethanolamine by incubating CP-oxamic acid and ethanolamine 2
J. P. Cravedi and E. Perdu-Durand, unpublished observations.
Phospholipid Adduct of Chloramphenicol Metabolites with various rat liver subcellular fractions such as mitochondria, submitochondria, or mitochondria plus microsomes plus cytosol. Another possible explanation of the origin of CP-oxamylethanolamine is an interaction of CP-oxamyl chloride-a reactive intermediate metabolite of CP (11)-with membrane phosphatidylethanolamine followed by the metabolic processing of the adduct as previously observed with the endogenous lipoperoxidation end product malondialdehyde (311. In order to test the latter hypothesis, several assays were conducted in vitro. In a first step, CP-oxamylethanolamine formation was confirmed by incubation of CP with isolated rat hepatocytes, suggesting the involvement of liver in the production of this metabolite. The low yield obtained probably explains the reason why this compound was not detected in similar studies previously undertaken (32). In a second part, we found that when the incubations were performed with microsomes from phenobarbitalpretreated rat, approximately 4% of the radioactivity was associated with the lipidic fraction of the microsomal pellet. The HPLC analysis of this fraction demonstrated that a substantial part of the radioactivity was due to an unknown compound which could correspond to a lipid adduct of a CP metabolite. This assumption was confirmed by the presence of CP-oxamylethanolamineand the concomitant disappearence of the unknown labeled compound in the lipidic fraction treated by phospholipase D from S. chromofiscus. Our data are consistent with the proposal of Pohl(l1) and Halpert (33)that the covalently bound material is derived from an acyl chloride intermediate formed during the cytochrome P450 dependent oxidative dechlorination of chloramphenicol (Figure 6). Such an intermediate could react either with water to produce CP-oxamic acid (19) or with the +amino group of one or more lysine residues in the cytochrome P450 protein (20)or with the amino group of the adjacent phosphatidylethanolamine in the microsomal membrane. This phospholipid is known to be one of the major components of the microsomal membrane, mainly located in the cytoplasmic surface (34). The presence of CP-oxamylethanolaminein rat and human urine as well as in rat isolated hepatocyte incubates reflects the turnover of phospholipids that have been modified by the covalent binding of the CP metabolite and the subsequent renal excretion of the biotransformation product still bound to ethanolamine. Van Dyke and Gandolfi (35)in studying the covalent binding of halothane with rat hepatic microsomal phospholipids demonstrated that the turnover of the phospholipids is rapid enough so that by 12 h most of the adducts have been broken down. This is in accordance with the rapid urinary elimination we have observed. Although direct evidence is not available, the toxicity of CP and the covalent binding of CP reactive metabolites with macromolecules have been related. This covalent binding has been shown to occur to hepatic proteins, but nothing was known regarding the eventual binding of C P reactive metabolites to lipids. Our results support the idea of an adduct of CP-acyl chloride with microsomal phospholipids, this adduct being responsible for the presence of CP-oxamylethanolamineamong CP urinary metabolites. However, the toxicological significance of this unusual biotransformation pathway remains to be investigated. The specific role of covalent drug-phospholipid interactions in hepatocytes is a matter of
Chem. Res. Toxicol., Vol. 8, No. 5, 1995 647 speculation. The incorporation of an exogenous compound into the microsomal membrane is expected to alter the fluidity of the membrane and thus might change membrane function as well (36).
Acknowledgment. We thank Professor D. Corpet (Ecole Nationale VBt6rinaire de Toulouse, France) for kindly supplying the urine samples from the human volunteer treated with L3H1CP.
hferences (1) Shalit, I., and Marks, M. I. (1984) Chloramphenicol in the 1980s. Drugs 28, 281-291. (2) Yunis, A. A. (1988) Chloramphenicol: Relation of structure to activity and toxicity. Annu. Rev. Pharmacol. Toxicol. 28,83-100. (3) Smith, A. L., and Weber, A. (1983) Pharmacology of chloramphenicol. Pediatr. Clin. North Am. 30, 209-236. (4) Lietman, P. S. (1979) Chloramphenicol and the neonate-1979 view. Clin. Pharmacokinet. 6, 151-162. (5) Oski, F. A. (1979) Hematologic consequences of chloramphenicol therapy. J . Pediatr. 94, 515-516. (6) Wallerstein, R. O., Condit, P. K., Kasper, C. K., Brown, J., and Morrison, F. R. (1969) Statewide study of chloramphenicol treatment and fatal aplastic anemia. JAMA, J.Am. Med. Assoc. 208, 2045-2050. (7) Yunis, A. A. (1973) Chloramphenicol-induced bone marrow suppression. Semin. Hematol. 10, 225-234. (8) Gross, B. J., Branchflower, R. V., Burke, T. R., Lees, D. E., and Pohl, L. R. (1982) Bone marrow toxicity in vitro of chloramphenicol and its metabolites. Toxicol. Appl. Pharmacol. 64, 557-565. (9) Murray, T., and Yunis, A. A. (1981) The cellular uptake and covalent binding of nitroso-chloramphenicol.J . Lab. Clin. Med. 98,396-401. (10) Yunis, A. A. (1984) Differential in vitro toxicity of chloramphenicol, nitroso-chloramphenicoland thiamphenicol. Sa.Trans. Dis. 11, 340-342. (11) Pohl, L. R., and Krishna, G. (1978) Study of mechanism of metabolic activation of chloramphenicol by rat liver microsomes. Biochem. Pharmacol. 27, 335-341. (12) Halpert, J., Naslund, B., and Betner, I. (1983) Suicide inactivation of rat liver cytochrome P-450 by chloramphenicol in vivo and in vitro. Mol. Pharmacol. 23, 445-452. (13) Corpet, D. E., and Bories, G. F. (1987) [3H]-chloramphenicol metabolism in human volunteer: oxamic acid as a new major metabolite. Drug Metab. Dispos. 15, 925-927. (14) Glazko, A. J., Wolf, L. M., Dill, W. A,, and Bratton, A. C. (1949) Biochemical studies on chloramphenicol. 11. Tissue distribution and excretion studies. J . Pharmacol. Exp. Ther. 96, 445-459. (15) Glazko, A. J.,Dill, W. A., and Rebstock, M. C. (1950) Biochemical studies on chloramphenicol (chloromycetin). 111. Isolation and identification of metabolic products in urine. J. Biol.Chem. 183, 679-691. (16) Dill, W. A,, Thompson, E. M., Fisken, R. A,, and Glazko, A. J. (1960) A new metabolite of chloramphenicol. Nature 185, 535536. (17) Bories, G. F., and Cravedi, J. P. (1994) Metabolism of chloramphenicol: a story of nearly 50 years. Drug Metab. Rev. 26,767783. (18) Cravedi, J. P., Baradat, M., Debrauwer, L., Alary, J., Tulliez, J., and Bories, G. (1994) Evidence for new metabolic pathways of chloramphenicol in the duck. Drug Metab. Dispos. 22,578-583. (19) Pohl, L. R., Nelson, S. D., and Krishna, G. (1978) Investigation of the mechanism of the metabolic activation of chloramphenicol by rat liver microsomes. Identification of a new metabolite. Biochem. Pharmacol. 27,491-496. (20) Halpert, J. (1981) Covalent modification of lysine during the suicide inactivation of rat liver cytochrome P-450 by chloramphenicol. Biochem. Pharmacol. 30, 875-881. (21) Bories, G. F., Peleran, J. C., Wal, J. M., and Corpet, D. E. (1983) Simple and ion-pair high performance liquid chromatography as an improved analytical tool for chloramphenicol metabolic profiling. Drug Metab. Dispos. 11, 249-254. (22) Pallante, S. L., Stogniew, M., Colvin, M., and Liberato, D. J. (1982) Elution of disposable octadecylsilane cartridges with hydrophobic organic solvents. Application in determination of chloramphenicol in biological materials. Anal. Chem. 54, 2612-2613. (23) Seglen, P. 0. (1976) Preparation of isolated rat liver cells. Methods Cell Biol. 13, 29-83. (24) Perdu-Durand, E. F., and Tulliez, J. E. (1985) Hydrocarbon hydroxylation system in liver microsomes from four animal species. Food. Chem. Toxicol. 23, 363-366.
648 Chem. Res. Toxicol., Vol. 8, No. 5, 1995 (25)Lowry, 0.H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951)Protein measurement with the Folin phenol reagent. J . Biol. Chem. 193,265-275. (26)Imamura, S.,and Horiuchi, Y. (1979)Purification of Streptomyces chromofuscus phospholipase D by hydrophobic affinity chromatography on palmitoyl cellulose. J . Biochem. 86, 79-85. (27)Kates, M. (1972)Techniques of lipidology. In Laboratory techniques in biochemistry and molecular biology (Work, T. S., and Work, E., Eds.) pp 568-570,Elsevier, Amsterdam and New York. (28)Morris P. L., Burke, T. R., George, J. W., and Pohl, L. R. (1982)
A new pathway for the oxidative metabolism of chloramphenicol by rat liver microsomes. Drug Metab. Dispos. 10,439-445. (29)Rajasekharan, R.,and Kemp, J. D. (1994)Synthesis of photoreactive phosphatidylethanolamine and its interaction with phospholipase Az. J . Lipid. Res. 35, 45-51. (30)Wal, J. M., Corpet, D., Peleran, J . C., and Bories, G. F. (1983) Comparative metabolism of chloramphenicol in germfree and conventional rats. Antimicrob. Agents Chemother. 24, 89-94. (31)Hadley, M., and Draper, H. H. (1989).Identification of N42propenal) ethanolamine as a urinary metabolite of malondialde
Cravedi et al. hyde. Free Radical Biol. Med. 6, 49-52. (32)Cravedi, J. P., and Baradat, M. (1991)Comparative metabolic profiling of chloramphenicol by isolated hepatocytes from rat and trout (Oncorhynchus mykiss). Comp. Biochem. Physiol. lOOC, 649-552. (33)Halpert, J . (1982)Further studies on the suicide inactivation of purified rat liver cytochrome P-450by chloramphenicol. Mol. Pharmacol. 21, 166-172. (34)Nilsson, O.,and Dallner, G. (1977)Enzyme and phospholipid asymmetry in liver microsomal membranes. J. Cell. Bwl. 72,568583. (35)Van dyke, R.A., and Gandolfi, A.J. (1974)Studies on irreversible binding of radioactivity from [14Clhalothaneto rat hepatic microsomal lipids and protein. Drug Metab. Dispos. 2,469-476. (36)Jain, S.K. (1984)The accumulation of malonaldehyde, a product of fatty acid peroxidation, can disturb aminophospholipid organization in the membrane bilayer of human erythrocytes. J.Biol. Chem. 259, 3391-3394.
TX950002A
