Induction and inactivation of cytochrome P-450 by macrolide antibiotics

Macrolide antibiotics, which are N-demethylated, may generate a metabolite which binds to cytochrome P-450 (P-450) forming a metabolic-intermediate (MI) complex. The complex inhibits further P-450 activity and is spectrally detectable at 456 nm. Compounds which form complexes also induce hepatic P-450. Among seven macrolide antibiotics tested, troleandomycin was the most potent inducer and MI complex former of rat hepatic P-450. After four days of troleandomycin (500 mg/kg/day; i.g.), hepatic P-450 was tripled with 65% sequestered as a MI complex. The P-450 subpopulations induced and complexed by troleandomycin eluted in fractions I and II after DEAE cellulose column chromatography. SDS polyacrylamide gel electrophoresis revealed two unique subpopulations in troleandomycin-induced microsomes not found in uninduced or phenobarbital-induced (PB) microsomes. Their apparent molecular weights were 52,000 and 58,000 daltons. Troleandomycin induced ethylmorphine N-demethylase activity more than p-nitroanisole O-demethylase activity, whereas phenobarbital induced both equally. The induced P-450 subpopulations and MI complex were persistent in vivo; four days after the cessation of drug treatment, 38% of total P-450 was complexed and the associated mixed-function oxidase (MFO) activities still inhibited. By several criteria, the nitrogenous macrolide antibiotics constitute a new class of MI complex-forming substrates. Formation of MI complex requires an induced state of P-450. The quantity of complex formed in liver is much more than that from the amines, SKF 525-A or norbenzphetamine. No macrolide antibiotic MI complex was formed in lung microsomes. The quantity of MI complex formed from macrolide antibiotics differed among PB animals; rat liver formed more MI complex from troleandomycin than rabbit liver, yet both animals formed complex from erythromycin to equivalent extents. The troleandomycin MI complex was stable to changes in pH, temperature and ionic strength during solubilization. The complex was not stable in the presence of potassium ferricyanide or high oxygen tension. Troleandomycin induces select subpopulations of P-450 and subsequently, during metabolism, forms a MI complex. The persistence of elevated hepatic P-450 and MI complex in vivo and the consequent alterations in MFO activities suggests exercising caution when troleandomycin is administered concommitantly with other therapeutic agents whose metabolism is P-450 dependent.

INDUCTION AND INACTIVATION OF P-450 BY MACROLIDE CYTOCHRO~m A1~TIBIOTICS by Lynn Kennard Pershing A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Pharmacology Department of Biochemical Pharmacology and Toxicology The University of Utah }~rch 1983 Copyright 0 Lynn Kennard Pershing 1983 All rights reserved THE UNIVERSITY OF UTAH GRADUATE SCHOOL SUPERVISORY COMMITTEE APPROVAL of a dissertation submitted by Lynn Kennard Pershing This dissertation has been read by each member of the following supervisory committee and by majority vote has been found to be satisfactory. December 20, 1982 Chainnan: December 20, Michael R. Franklin, Ph.D. 1982 December 20, 1982 December 20, 1982 December 20, 1982 ������ St�/�Rofr3b.� THE UNIVERSITY OF UTAH GRADUATE SCHOOL FINAL READING APPROVAL To the Graduate Council of The University of Utah: I have read the dissertation of Lynn Kennard Pershing In lts final form and have found .that (1) its format, citations, and bibliographic style are consistent and acceptable; (2) its illustrative materials including figures, tables, and charts are in place; and (3) the final manuscript is satisfactory to the Supervisory Committee and is ready for submission to the Graduate School. December 20, 1982 Dat� Michael R. Franklin, Ph.D. Member. Supervisory Committee Approved for the Major Department Approved for the Graduate Council ABSTRACT Troleandomycin is a macrolide antibiotic used in the treatment of Mycoplasma pneumonia, Legionnaire's disease, penicillin- resistant respiratory patients. Many infections drug and interactions for penicillin-allergic involving troleandomycin or another macrolide antibiotic, erythromycin, have been cited in the recent medical literature. These drugs, which are oxidatively N-demethylated, may generate a metabolite, which binds to cytochrome P-450, forming a metabolic-intermediate (MI) complex. inhibits further cytochrome detectable at 456 nm. P-450 activity and The complex is spectrally In addition to inhibiting cytochrome P-450, troleandomycin induces hepatic cytochrome P-450. The selective induction of cytochrome P-450 subpopulations by troleandomycin and its characteristics as a substrate for MI complex formation were examined in the present study. Troleandomycin was the most former of rat antibiotics conc~ntration hepatic tested in potent cytochrome vivo. inducer P-450 Although among the and MI complex seven macrolide cytochrome P-450 was tripled after four days of treatment, an amount of cytochrome P-450 equivalent to that induced was sequestered as a HI complex (65% of total cytochrome P-450). and HI complex were persistent The induced subpopulations in vivo; four days after the cessation of drug treatment, 38% of total cytochrome P-450 was still complexed and the associated mixed-function oxidase activities still inhibited. The cytochrome complexed by troleandomycin, P-450 subpopulations induced eluted in fractions I and II after diethylaminoethyl (DEAE) cellulose column chromatography. dodecyl sulfate polyacrylamide gel and Sodium electrophoresis revealed two unique subpopulations iD troleandomycin-induced microsomes not found in uninduced apparent or phenobarbital-induced molecular weights were liver 52,000 microsomes. and 58,000 Their daltons. Troleandomycin induced ethylmorphine N-demethylase activity to a greater extent than p-nitroanisole O-demethylase activity, whereas phenobarbital induced both troleandomycin-induced equally. The subpopulations formed degree a to which complex from norbenzphetamine was similar to control cytochrome'P-450 (30%), but much less than phenobarbital-induced cytochrome P-450 (60%). By several criteria, the -nitrogenous macrolide antibiotics constitute a new class of HI complex forming substrates. of HI complex in vitro from macrolide antibiotics induced state of cytochrome P-450. Formation requires an The extent of the complex formed in liver was much less, however, than that from other amines such as norbenzphetamine and chloride (SKF diethylaminoethyl 2,2-diphenylvalerate:hydro- 525-A). Like SKF 525-A, and in contrast to norbenzphetamine, no macrolide antibiotic HI complex was formed in lung micro somes. The quantity of complex formed differed among phenobarbitalinduced animals; rat liver formed more HI complex from troleandov mycin than rabbit liver, yet both animal livers formed complex from erythromycin to equivalent extents. The troleandomycin MI complex, like other nitrogenous complexes, such as that from SKF 525-A, was rela tively stable to changes in pH, temperature and ionic strength during solubilization. The complex was unstable in the presence of potassium ferricyanide and unlike other nitrogenous complexes, under the influence of high oxygen tension. Macrolide antibiotics, in particular troleandomycin, appear to induce select subpopulations of cytochrome P-450 and subsequently, during metabolism, form a MI complex. The persistence of elevated hepatic cytochrome P-450 and MI complex concentrations in vivo and the consequent alterations in mixed-function oxidase activities suggests that caution should be exercised when macrolide antibiotics are administered concommitantly with other therapeutic agents whose metabolism is cytochrome P-450 dependent. vi CONTENTS ABSTRACT . • • LIST OF TABLES iv ..... ·· .viii LIST OF FIGURES ix ACKNOWLEDGEMENTS xii Chapter 1. INTRODUCTION • • • • • 2. METHODS . • • Animal Treatments and Microsome Preparation • Cytochrome P-450 and Metabolic-Intermediate Complex Quantification. . . • • . . . • • . Diethylaminoethyl Cellulose Chromatography and Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis . . . . . . • . • Stability of Solubilized Cytochrome P-450 and Metabolic-Intermediate Complex. • . • • • Metabolic-Intermediate Complex Persistence In Vivo • . • • • . Statistics . . Materials • 3. 4. 5. 6. 7. 8. 1 DOSE RESPO~SE OF CYTOCHROME P-450 INDUCTION BY TROLEANDOMYCIN IN RAT LIVER • • • . . . . • • . 16 .··. .. 16 17 19 20 21 22 22 ..··. 24 MACROLIDE ANTIBIOTIC INDUCTION OF CYTOCHROME P-450 AND METABOLIC-INTERMEDIATE COMPLIDC FORMATION IN VIVO. 28 tffiTABOLIC-INTERMEDIATE COlWLEX FOID1ATION IN VITRO FROM TROLEANDOMYCIN . • . . . • . . . • . • . . • . . • 33 INDUCTION OF CYTOCHROME P-450 SUBPOPULATIONS BY TROLEANDOMYCIN . • . . • • . . • . • . . . • . • 46 PERSISTENCE OF THE TROLEANDOMYCIN METABOLIC-INTEPJ1EDIATE COMPLEX IN VIVO 79 DISCUSSION . . . • 87 APPENDIX: NllfERICAL DATA • REFERENCES • • VITA. • • • 91 113 . • 119 LIST OF TABLES Table I. II. III. IV. A classification of cytochrome P-450 metabolicintermediate complexes . • • • • • . . . 8 The ability of amine containing antibiotics to form metabolic-intermediate complexes with phenobarbital-induced rat liver microsomes in vitro. . • • . • • • • • • • • • 43 Properties of the classes of cytochrome P-450 metabolic-intermediate complexes • • • • . • . . . . . . . . 44 Ability of induced and uninduced microsomal cytochrome P-450 to form a metabolic-intermediate complex from norbenzphetamine in vitro • • • • • • • • • • . . • 53 LIST OF FIGURES Figure 1. 2. Phase I and II hepatic metabolism eliminating xenobiotics from the body . . • 2 Hepatic mixed-function oxidase system and the mechanism of metabolic-intermediate complex formation . • • . 4 3. Structure of tro1eandomycin 4. ~~jor 5. routes of metabolic biotransformation of tro1eandomycin. . . . . ..•• . 10 • • • • . . • . . 12 The effect of tro1eandomycin dose on rat hepatic cytochrome P-450 concentration. •• ••.• . • • 25 6. Comparison of macrolide antibiotic induction and metabo1icintermediate complex formation in rat liver in vivo • . . . 29 7. Extent of metabolic-intermediate complex formation from tro1eandomycin, erythromycin base, SKF 8742-A and norbenzphetamine in vitro with uninduced, phenobarbita1- and S-naphthof1avone-induced rat liver microsomes. . • • • . . • . • . • • • • • • . . • • . 34 8. The extent of metabolic-intermediate complex formation in vitro with phenobarbital-induced rat and rabbit liver microsomes . . . • . • • • . . • • • • . . 37 9. The extent of metabolic-intermediate complex formation in vitro with phenobarbital-induced rabbit liver and lung microsomes. • . . • . . . 40 10. Time course of rat hepatic cytochrome P-450 induction by tro1eandomycin in vivo . • • • . • • . . • • . . . • . 47 11. Effect of phenobarbital and S-naphthof1avone pretreatment on metabolic-intermediate complex formation from tro1eandomycin in vivo . • • • . • • . . . • 12. • • . . • 50 DEAE cellulose column chromatography of uninduced, phenobarbital-induced and tro1eandomycininduced rat hepatic microsomes . . . • . . • . . . . • • . . . . 55 13. Diagrammatic representation of the microsomal proteins present after SDS polyacrylamide gel electrophoresis of DEAE eluate fractions . . . . . 58 14. DEAE cellulose column chromatography of troleandomycininduced and phenobarbital-induced-eight hourtroleandomycin-treated rat hepatic microsomes . . . . . . . . . . 61 15. The effect of solubilization and DEAE cellulose chromatography on the total cytochrome P-450 and metabolic-intermediate complex concentrations from uninduced, phenobarbital- and troleandomycininduced rat liver. • • . . . . . . . • . . . . • . • . . . . 64 16. The effect of temperature on the stability of total cytochrome P-450 and metabolic-intermediate complex during solubilization. . . . . . . . . . . . . . . . . . . . . . 67 17. The effect of salt concentration in the solubilizing buffer on the stability of total cytochrome P-450 and metabolic-intermediate complex . . •. . . . . . . . . . 70 18. The effect of pH of the solubilizing buffer on total cytochrome P-450 and metabolic-intermediate complex stability. .......... . . . . • . . 72 19. The effect of oxygen and nitrogen gassing of the solubilizing buffer on the stability of total cytochrome P-450 and metabolic-intermediate complex. . . . . . . 74 20. Comparison of nitrogen pregassing and ambient air on the DEAE cellulose column chromatography of solubilized troleandomycin-induced rat liver microsomes. . . . . . . . . 77 21. The duration of induced cytochrome P-450, metabolicintermediate complex and mixed-function oxidase activities after troleandomycin and phenobarbital induction of rat liver . . . . . . . . . . . . . . . . . . . . . 80 22. The hexobarbital sleep times during and following induction by phenobarbital or troleandomycin. . . . . . . . . . 84 ACKNOWLEDGEMENTS I express my sincere appreciation to the members of my cODDllittee, Drs. Michael R. Franklin, Bryan S. Finkle, Douglas E. Rollins, Stewart C. Harvey and Milton J. guidance and careful review of this work. Zmijewski, for their I am especially grateful to Dr. Michael R. Franklin, who has been instrumental in my graduate school experience at the University of Utah. A special note of gratitude to my friends, professional and nonprofessional, who have supported me and broadened my interests. Gratitude is also expressed to Dr. Zahra Parandoosh and Ms. Julie Davis Krall for their technical assistance. Finally, I am indebted to my husband, David, for his encouragement, patience and confidance throughout this endeavor. This work was supported by United States Public Health Service Grants CA 15760 and GM 07579. CHAPTER 1 INTRODUCTION A foreign compound once ingested and absorbed must be eliminated to prevent accumulation and possible adverse reactions. Elimination may be enhanced by Phase I metabolism, where the compound is metabolically biotransformed via oxidation, reduction or hydrolysis (Figure 1). Biotransformation provides the compound with a reactive center facilitating its conjugation with glucuronide, sulfate, amino acids or glutathione during Phase II metabolism. The addition the of conjugating groups increases the polarity of compound and thus promotes its rapid excretion from the body. The biological system responsible for much of the oxidative metabolism of xenobiotics is the mixed-function oxidase system, also referred to as cytochrome P-450, which is found most abundantly in the endoplasmic xenobiotic reticulum metabolism of the liver. (Figure 2), native In the schema ferricytochrome for P-450 (Fe-III) binds to a substrate (1) and then accepts an electron (2) from NADPH (via a flavoprotein) to become a ferrocytochrome P-450 (Fe-II)-substrate complex. The binding of molecular oxygen (3) and the addition of another electron (4) from NADPH (via a flavoprotein) splits molecular oxygen, oxidatively metabolizing the foreign compound (substrate). releasing the oxidized substrate (5), water and regenerating the ferricytochrome P-450 (Fe-III) (1). Figure 1. Phase I and II hepatic metabolism eliminating xenobiotics from the body. EXCRETION ~ ~ CHEMICAL COMPOUND i OXIDATION REDUCTION HYDROLYSIS PHASE I ,METABOLIC TRANSFORMATION) Jl ~ (CONJUGATION) PHASE II 1 ....... GLUCURONIDATION SULFATION ACETYLATION METHYLATION AMINO ACID CONJUGATION GLUTATHIONE CONJUGATION EXCRETION w Figure 2. Hepatic mixed-function oxidase system and the mechanism of metabolic-intermediate complex formation. (2) Cytochrome f- 450 I +CO ~- __ .... ~ __ vI SUb:trate oxidized substrate + Fpf--NAOPH K3Fe(CN)~ H20 fj'\Cytochrome P-450-substrate \V I ++ Fe '1 o o -70 Cytochrome P-450-substrate \.J ~IFe+++ 1 : product F +++ . _1_ oT ~Cytochrome PI-450-substrate-++ Cytochrome P-450-substrate j ++ ~ te CO CO , I Fe -02 (450 nm) K3 Fe (CN)6 . Fp .... NADPH f7\ Cytochrome '-.J P-450 a Ami nes - - - - - - - I ~-­ Fe++- MI""""""-I' (455 nm) ~. b 'Methyl ened1 omhen.v I 02;K3~ • '8' Cytochrome P-450--> {cri}6~ te i I Fe +++ .•...... (438 nm) HI lFI 6 In the laboratory, this pathway can be manipulated to quantify cytochrome P-450. The introduction of carbon monoxide gas into the microsomal suspension containing reduced cytochrome P-450 (Fe-II) (3) , resul ts in a carbon monoxide-cytochrome P-450 complex (6) rather than an oxygen-cytochrome P-450 complex. P-450 ligand is carbon monoxide, maximum absorbance at 450 nm. When the cytochrome the complex is detectable by a Treatment with an oxidizing agent, potassium ferricyanide [K Fe(CN)6] oxidizes the iron, dissociates 3 the ligand, and releases carbon monoxide, thus regenerating cytochrome P-450 (Fe-III) (2). With some compounds, the second electron causes the formation of a reactive intermediate which is not released as a product. intermediate binds to the reduced cytochrome P-450 The (Fe-II), resulting in the formation of a metabolic-intermediate (MI) complex (7). These complexes are stable and prevent further participation of cytochrome P-450 in monooxygenase activity. compounds, With some foreign especially those with unsaturated carbon-carbon bonds, the heme is either modified or destroyed (De Matteis, 1973; Levin et al., 1973; Reynolds et al., 1975; Guengerich and Strickland, 1977; White, 1978 ; Ivanetich et al., 1978; De Matteis and Cantoni, 1979; Ortiz de Honte11ano et al., 1979; 1980). For other foreign compounds, the heme can be freed of its ligand and the cytochrome regenerated in complexes, this a functional can be (Elcombe et a1., 1976). accomplished by oxidizing state. With accomplished by non-nitrogenous displacing the HI ligand With nitrogenous complexes, this can be the heme to the ferric state, which 7 releases the unstable metabolic-intermediate from the stabilizing influence of the heme. Metabolism of amine groups can produce a nitroso metabolic-intermediate (Mansuy et al., 1977), which results in a HI complex with an absorbance maximum at 455 nm. with K3Fe(CN)6 dissociates the MI complex (7), Treatment releases the metabolic-intermediate product and regenerates ferricytochrome P-450 (Fe-III) (1). Compounds containing the methylenedioxyphenyl group differ from the amine compounds in that the intermediate forming the HI complex is a carbene (Ullrich et al., 1975). Such a complex remains intact even with the cytochrome in the ferric state. treating these HI complexes with the oxidizing agent K 3 Thus, Fe(CN)6 does not dissociate the complex, but changes the absorbance maximum from 455 nm to 438 nm. The dramatic increase in the identification of compounds able to inhibit cytochrome P-450 reactions by the formation of HI complexes has been facilitated by the ease of detection of the cytochrome P-450 HI complex. Wi th this plethora of compounds (Franklin, 1977), a need for classification has arisen. working classification is shown in Table I. the eight classes are easily discernible, A current Some delineations of the separation of the non-nitrogen containing compounds (where a carbene is believed to be the metabolic-intermediate, Ullrich et al., 1975) from those where the amine is believed to intermediate, dithionite, (Mansuy being et good be al., the source of a ni troso reac tive 1977) examples. and Other the stability delineations, towards such as whether the compound induces cytchrome P-450, or the extent to which Table I. A classification of cytochrome P-450 metabolic-intermediate complexes Class of MI Complex Substrate Extensively Studied Nitrogen Containing Cytochrome P-450 Inducing Ability Absorbance Maximum Fe++ !nm} Stability to Dithionite + 427, 455 + - (438) 427 + + Dissocation Fe++ . . . Fet++ Methylenedioxybenzene isosafro1e Dioxolanea 4-n-buty1dioxolane Amphetamine norbenzphetamine + 455 + + Oxidized Alkylamine N-hydroxyamphetamine + 455 + + SKF 525-A SKF 525-A + 452 + + Ary1amine p-chloroaniline + 448 + Hydrazineb N-aminopiperadine + 449 - (438) Macrolide Antibiotic C Tro1eandomycin + + + 456 + + aFrom Dahl and Hodgson (1979) b From Hines and Prough (1980) cFrom Pessayre et a1. (1981) Other information from M.R. Franklin (1977) 00 9 the cytochrome present in the liver micro somes of an induced or uninduced animal will form an MI complex, are less definite. Recent reports have widened the scope of nitrogenous MI complex forming agents to include several therapeutically useful macrolide antibiotics (Pessayre et al., 1981a; 1982; Danan et al., 1981). term macrolide refers to the structure of The those compounds which contain a many membered lactone ring (A) with deoxy sugars (B and C) attached via glycosidic linkages (Figure 3). Troleandomycin also contains three acetyl groups (positions 1, 2, and 3 in Figure 3). The desosamine sugar contains a tertiary amine (4), which is a site of metabolism of these compounds. The macrolide antibiotics are metabolized by cytochrome P-450 dependent N-demethylation (Figure 4) . The removal of one or both methyl groups from the tertiary amine results in metabolites, which after further oxidative metabolism, could produce a nitroso metabolic-intermediate (Mansuy, 1977). The intermediate may then bind to reduced cytochrome P-450 (Fe-II) forming a MI complex. Another minor metabolic pathway in humans, which is not cytochrome P-450-dependent, is deacetylation, the final product of which is oleandomycin (Celmers, 1957). The amine in macrolide antibiotics is attached to a "lactone" ring, thus differing from the attachment to an aromatic ring as in the aryl amines, and the attachment to a straight aliphatic chain, as in the amphetamine and SKF 525-A classes. Like the SKF 525-A class (Anders and Mannering, 1966; Schenkman et al., 1972; Buening and Franklin, 1976), but unlike the amphetamine macrolide antibiotics induce cytochrome P-450. classes, the Whether they induce Figure 3. Structure of troleandomycin. The molecular structure includes three distinct parts: Alactone nucleus, B - desosamine sugar containing the tertiary amine and C • cladinose sugar. Biotransformation by hepatic cytochrome P-450 occurs via N-dem.ethylation of the tertiary amine (position @ 1.. Un~ue to troleandomycin are the acetyl groups at positions <.!) , \.V and 0). 11 ® Figure 4. domycin. Major routes of metabolic biotransformation of trolean- Double arrow steps are cytochrome P-450 dependent (Ndemethylation). Single arrow steps are not cytochrome P-450 dependent and represent a minor metabolic pathway (deacetylation) in humans. Troleandomyc1n /~ N-demethylation ~ desmethYltrol~ . oXldation~ II omYcln N-demethylation ~ didesmethyl~leand~Cln I ~ 1,3-d1acetyloleandomycin I l 3-monoacetyloleandomycin deacetylation I deacetylation "" /R [---N"'o_ deacetylation •••• cytochrome P-450 (Fell)] ~ oleandomYcin MI COMPLEX ~ w 14 the same rat cytochrome P-450 subpopulations as SKF 525-A, which has been described as inducing like phenobarbital 1981), is a part of the present study. with new experimental results from (Thomas et al., This information together this study are intended to determine whether the macrolide antibiotics can be defined as a new class of compounds which form MI complexes. delineations arises their methylenedioxybenzene effectiveness insecticides and pharmacological toxicological Formation of stable cytochrome P-450 MI complexes implications. from from The importance of such and synergists. in derivatives blocking hence their Similarly, undoubtedly the oxidative widespread blocking use accounts for metabolism of as cytochrome insecticide P-450-dependent metabolism by MI complexes from amines, may impair the metabolism of other drugs and HI interactions. different thus produce complexes extents with the untoward effects, from different cytochrome i.e., classes P-450. drug-drug interact Since the to total cytochrome P-450 concentration is comprised of subpopulations with a different spectrum of monooxygenase activities (Lu and West, 1981), the differences in the extent of HI complex formation may represent interactions inhibiting with only different one or a subpopulations. limited number of By selective ly cytochrome P-450 subpopulation(s) and the monooxygenase reactions they catalyze, the remaining subpopulations may be free to metabolize other foreign compounds. In this manner, inhibiting cytochrome P-450 subpopulations that catalyze the metabolism of compounds to more toxic products, while allowing others that are unaffected to 15 catalyze reactions to non-toxic products, could be used to reduce the toxicity of otherwise useful chemicals. In the laboratory, this selectivity enables MI complexes to be used as probes for detecting cytochrome p-450 subpopu1ations. It also facilitates their quantitation without the need for an immunological method, which would require the isolation of highly purified subpopu1ations in order to prepare antibodies. There is also a need to understand their selective inducing properties of cytochrome P-450, in order to anticipate drug interactions or drug induced toxicities and to appreciate possible manipulation of cytochrome subpopu1ations for beneficial effects. CHAPTER 2 METHODS An±mai-~reatments and-Microsome Preparation Male Sprague-Dawley rats (180-280 grams) and male New Zealand White rabbits (2.5-3.0 kilograms) were used in these studies. Induction of rat liver cytochrome P-4S0 concentrations was achieved by administering phenobarbital [(80 mg/kg, intraperitoneally (i.p.) daily for four days)l, S-naphthoflavone in corn oil (80 mg/kg, i.p. daily for three days), or macrolide antibiotic in 1% methylcellulose [(500 mg/kg intragastrically (i.g.) daily for four days)j. rats received the appropriate vehicle for Control each drug treatment (saline, corn oil or 1% methy!cellulose, respectively). Rats and rabbits were starved 24 hours after the last dose and sacrificed by decapitation. In rabbits, a lower dose of phenobarbital (40 mg/kg) was used to induce liver cytochrome P-450 concentration. Rabbits were sacrificed by carbon dioxide inhalation. Livers were perfused in situ via cold 0.15 the hepatic portal vein with M sodium chloride, then removed from the animal, blotted and weighed. Liver microsomes were prepared by differential centrifugation (Franklin and Estabrook, 1971) where the tissue was first homogenized with 0.25 M sucrose to produce a 20% homogenate and centrifuged for 17 fifteen minutes at 9000 g, followed by recentrifugation of the supernatant at 19,000 supernatant was then g for another centrifuged at fifteen 100,000 minutes. g for The fifty-five minutes, resuspended in 0.15 M potassium chloride (20% homogenate) and recentrifuged. The potassium chloride wash removed excess heme from the microsomal preparation. The final pellet was resuspended either in 50 mM tris-O .25 M sucrose buffer Na HP0 (pH 7.4), 2 4 microsomal (pH 7 .4) or 0.01 depending on the experiment, protein concentration of to give a final approximately Microsomes were used immediately after react with alkaline phosphomolybdic-phosphotungstic reagent. blue color which is 30 preparation. protein was assayed according to Lowry et al. aromatic amino acids M Microsomal (1951), copper mg/ml. in which and reduce the This reaction produces a detected spectrophotometrically at 700 nm. Bovine serum albumin was used as the reference standard. Cytochrome P-450 and MetabolicIntermediate Complex Quantification Microsomes were diluted to 2.0 mg/ml in 50 mM tris-chloride buffer pH 7.4, containing magnesium The chloride. determined from dithionite reduced 150 mM potassium chloride and 10 mM the cytochrome carbon liver P-450 monoxide microsomes concentration difference and from the spectrum was of dithionite difference spectrum of carbon monoxide gassed lung microsomes, using extinction coefficients of 91 -1 mM -1 cm and respectively, for the 450 versus 490 nm absorbance difference (Omura and Sato, 1964; Estabrook et al., 1972). All spectra were recorded 18 with an Aminco DW-2a spectrophotometer. The MI complex formed in vivo after macrolide antibiotic treatment was quantified from the difference spectrum. observed three minutes after the addi tion of potassium ferricyanide (50 reference cuvette. ~M) to the microsomal suspension in the An extinction coefficient of 64 mM- 1 cm- 1 for the absorbance difference between 456 and 490 nm was determined (n = 14). The maximum extent of cytochrome P-450 HI complex formation in control, phenobarbital- or S-naphthoflavone-induced rat-liver, rabbit-liver or rabbit-lung microsomes in vitro was determined from repeated scanning of microsomal suspensions (2 mg protein/ml), which were incubated at 25 0 C with oxygenated tris buffer (pH 7 .4), containing 2 m.M NADPH and the MI complex forming substrate. The optimum substrate concentration for the macrolide antibiotics, the N-demethylated metabolites of erythromycin base, aminoglycosides, lucanthone, hycanthone, doxycycline, tetracycline, and chloroquine was 133 11 M. For norbenzphetamine, it was 100 ~ M and for ethylaminoethyl 2,2-diphenylvalerate hydrochloride (SKF 8742-A), 3311M. Extinction coefficients of the cytOChrome P-450 MI complexes were 65 mM- 1cm- 1 for norbenzphetamine (455 minus 490 nm) (Franklin, 1974), SKF 8742-A (452 minus 490 nm) unpublished data) and 64 mM- 1 cm- macrolide antibiotics. 1 (Gumbrecht (456 minus and 490 nm) Franklin, for the 19 Diethylaminoethyl Cellulose Chromatography and Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis Microsomes were diethylaminoethyl column at room solubilized (DBAE) and chromatographed cellulose (Whatman DE 52) temperature by a modified method Franklin, 1982) of Warner and Neims (1979). on a ion exchange (Bornheim and Microsomes in 0.01 M sodium phosphate buffer (pH 7.4) were solubilized for one hour with the addition of a solubilizing buffer {0.66 ml/gram liver (original wet weight)] containing 10 mM ethylenediaminetetraacetic acid, sodium 0.75% phosphate, (w/v) 0.15 m.M sodium cholate, 30% (v/v) glycerol, 0.3% (v/v) Emulgen 911 and adjusted to pH 7.4. The solubilized microsomes containing 300 nmoles of cytochrome P-450 were applied to the column (30 X 2.5 cm) and developed with 100 ml of buffered detergent (10 mM ethylenediamine-tetraacetic acid, sodium phosphate, 0.5% (w/v) 0.1 mM sodium cholate, 20% (v/v) glycerol and 0.2% (v/v) Emulgen 911) followed by a 500 ml linear sodium chloride gradient (0-0.25 M) in buffered detergent. Eluted fractions were collected at and analyzed spectrophotometrically for the presence of MI complexes and total cytochrome P-450 (after dissociation of the MI complex by potassium ferricyanide). their apparent electrophoresis Proteins in the eluted fractions were assayed for molecular in the weights presence by of 0.1% polyacrylamide sodium dodecyl (SDS-PAGE) by the method of Laemmli (1970). gel sulfate A five milliliter 20 sample of the collected fractions was dialyzed overnight in cellulose membrane tubing overnight in a buffer (pH 6.8) containing 60 mM Tris-chloride and 1% sodium dodecyl sulfate. day, the dialyzed sucrose. combined samples were concentrated The following over crystalline A volume equivalent to approximately 30 mg protein was with 1% sodium dodecyl 15% sulfate, glycerol, bromophenol blue and 5% B-mercaptoethanol in a 400 and heated in boiling water for three minutes. ~L 0.1% capped vial The cooled sample was applied to a I-em stacking gel of 3% acrylamide followed by an 8-em running gel of 7.5% acrylamide. Gels were electrophoresed at 40 mamp through the stacking gel and 70 mamp through the running gel. The gels were then stained overnight with 0.05 % Coumassie Brilliant Blue in methanol:acetic acid:water (2:1:7), according to the method of Fairbanks et al. (1971). Stability of Solubilized Cytochrome P-450 and Metabolic-Intermediate Complex The stability of the troleandomycin MI complex was tested in vitro by varying conditions during solubilization, including: (7.0, 7.4 and 7.8), salt concentration (0, 0.25 and 0.5 pH M), temperature (4 oC and 25 0 C) and continual gassing of the sample in the solubilizing buffer with nitrogen, experiment, oxygen or air. In each a microsomal sample was divided into equal aliquots, such that a "control, n representing normal conditions, was tested against any change in a particular variable. treated with four days of saline, Microsomes from rats troleandomycin or 21 diethylaminoethyl 2,2-diphenylvalerate hydrochloride (SKF 525-A) were monitored in each study for total cytochrome P-450 and MI complex concentrations after 0, 1, 4, 8, 12 and 24 hours of solubilization. Investigation of the influence of oxygen on the stability of MI complexes during DEAE cellulose column chromatography was performed with the gradient nitrogen. All buffer, buffers and continuously gassed with column support were humidified pregassed with nitrogen before the experiment to insure a continuous oxygen-free environment. Microsomal preparation and column operating conditions were identical to those previously described. Metabo-lic-Intenped.iate Complex Persistence In Vivo The presence of the troleandomycin MI complex in vivo, following the cessation of drug administration, was studied in the rat liver by preparing microsomes 12, 24, 48 and 96 hours after maximal induction with troleandomycin (500 mglkg i.g. daily for a duration of 4 days). were Total cytochrome P-450 and MI complex concentrations measured and quantitated as previously described. Mixed-function oxygenase activities were also monitored at the same time points. the formation ethylmorphine p-Nitroanisole O-demethylase activity was measured by of p-nitrophenol (Netter and Seidel, 1964) and N-demethylase activity was quantitated by monitoring the formation of formaldehyde (Nash, 1953). NADPH cytochrome ..£. reductase activity was quantitated by measuring the reduction of exogenous cytochrome S at 550 nm (Masters et al., 1971). The in 22 vivo oxidative metabolism of hexobarbital (100 mg/kg, i.p.) was studied in male Sprague-Dawley rats (160-200 g) after 1, 2, 3 and 4 days of daily troleandomycin treatments (500 mg/kg, i.g.) as well as each of the administration. with a towel hypothermia. four days following the cessation of drug After hexobarbital injection, all rats were covered and placed under an incandescent lamp to reduce The sleep time was determined as that from the time of hexobarbital injection to the time when the treated, dorsally placed rat was able to right itself three consecutive times. Stati-stics Sample means from the treatment groups were analyzed for statistical difference by Student's t test for independent data. Materials S-Naphthoflavone, oleandomycin, tobramycin, kanamycin, tetracycline, doxycycline, spiramycin, streptomycin, cytochrome S neomycin, reductase gentamycin C , 1 chloroquine, (type VI) and NADPH (sodium salt, type IV) were purchased from Sigma Chemical Co., St. Louis, MO. DEAE (DE 52) cellulose support was purchased from Whatman, Inc., Clifton, NJ. Troleandomycin was a generous gift from Pfizer Co., Brooklyn, NY. Erythromycin base and its N-demethylated metabolites were gifts from Abbott Laboratories, North Chicago, IL., and the erythromycin estolate was a gift from Eli Lilly and Co., Indianapolis, IN. The Toyo Jozo Co., Ltd., leucomycins A3 and A5 for our research purposes. SKF 525-A were gifts from Japan, donated SKF 8742-A and Smith Kline and French Laboratories, 23 Philadelphia, PA., and N-benzy1-a1pha-methy1-phenethy1amine:hydrochloride (norbenzphetamine:HCL, purity> 99%) was synthesized in our laboratory by Z. Parandoosh. Lucanthone hydrochloride was donated by Research Burroughs We11come Co., hycanthone methanesu1fonate was Triangle supplied by Park, N.C., and Sterling Winthrop Research Institute, Rensselaer, N. Y.. Emu1gen 911 was a gift from Kao-At1as Co., Ltd., Japan. grade quality. All other chemicals were of analytical CHAPTER 3 DOSE RESPONSE OF CYTOCHROME P-450 INDUCTION BY TROLEANDOMYCIN IN RAT LIVER A dose response curve for induction was constructed using 250, 500 and 750 mg/kg of troleandomycin administered i.p_ daily, for 1, 4, or 7 days (Figure SA). The high dose, 750 mg/kg, was lethal to greater than 50% of the animals receiving tro1eandomycin for more than one day. A dose of 500 mg/kg induced cytochrome P-450 1.6 fold after both four and seven days, with no fatalities. Since seven daily doses of troleandomycin (500 mg/kg) produced no more induction than the four day treatment, the shorter treatment was chosen for subsequent studies. Intraperitoneal administration of 500 mg/kg of troleandomycin was compared with the intragastric administration methylcellulose) of the drug (Figure 5B). (in 1% Although the percent of metabolic-intermediate (MI) complex was similar after four days of 500 mg/kg troleandomycin, total cytochrome P-4s0 the intragastric administration induced to a greater extent. Thus, a daily intragastric administration of troleandomycin at 500 mg/kg for four days became the standard dose for the study. The extent of cytochrome P-450 induction by troleandomycin in the present study agreed with the data of Mansuy et ala (1981) (500 mg/kg/day x 3 days in corn oil; i.p.) and Pessayre et ala (1981a), Figure 5. The effect of troleandomycin dose on rat hepatic cytochrome P-450 concentration. A: The induction of hepatic cytochrome P-450 by troleandomycin administered intraperitoneally. Troleandomycin in doses of 250, 500 and 750 mg/kg/ day for 1 (unshaded bar), 4 {stippled bar} and 7 (hatched bar) days are compared. The data represent the mean ~ S.E.M. for 3 rats. B: A comparison of the routes of administration of troleandomycin (500 mg/kg day x 4 days) on cytochrome P-450 induction and metabolic-intermediate complex formation. Concentrations (nmoles/mg microsomal protein) of HI complex {shaded bar} and uncomplexed cytochrome P-450 after i.p. administration {unshaded bar} and i.g. administration (stippled bar) are plotted. Total cytochrome P-450 concentration is represented by MI complex plus the uncomplexed cytochrome P-4S0. The data represents the mean ~ S.E.M. for 3 rats. 26 A. ci E ...... • .! o E c: o ~ • Go •E e I/:. Co) o )Ii! Co) o 250 B. t •• 1c: ...... 0.0 .............CON TAO 500 150 mg/kg TAO 27 except that in the latter study a much larger dose (814 mg/kg/day x 4 days), which was above the LDSO in our rats, was required to elicit the same response. CHAPTER 4 MACROLIDE ANTIBIOTIC INDUCTION OF CYTOCHROME P-450 AND METABOLICINTERMEDIATE COMPLEX FORMATION IN VIVO For several classes of compounds which form metabolic- intermediate (MI) complexes, it has been shown that alteration or substitution of the functional groups on a skeletal structure can alter their ability to induce cytochrome P-450 and/or form (Franklin, complexes 1971; activity relationships for with several macrolide 1976). Elucidation of HI structure- these two properties was investigated antibiotics related (structures and numerical data in the Appendix). treatment with each macrolide antibiotic to troleandomycin After four days of (500 mg/kg/day), total cytochrome P-450 (uncomplexed + HI) and MI complex concentrations were determined (Figure 6). Troleandomycin was the most potent inducer of cytochrome P-450 and formed the greatest amount of MI antibiotics tested in vivo. appeared to troleandomycin, be far The determinants removed erythromycin complex of from the estolate was amine the macrolide that made moiety. esterified at it so Like the R1 position, but did not induce cytochrome P-450 or form MI complex significantly. Oleandomycin had functional groups similar to those of troleandomycin, except that it was not acetylated at the Rl and the two R3 positions and did not induced cytochrome P-450 Figure 6. Comparison of macrolide antibiotic induction and metabolic-intermediate complex formation in rat liver in vivo. Animals received 500 mg/kg/day x 4 days, i.g. of tro1eandomycin (TAO) or erythromycin esto1ate (EE) or erythromycin base (EB) or oleandomycin (OLD) or spiramycin (SPR) or 1eucomycin ~ (LEU ~) or 1eucomycin A (LEU AS) in a suspension of 1% methylcellurose. Control animats (C) received the vehicle, 1% methy1ce11u10se only. Uncomp1exed cytochrome P-450 (unshaded bars) and metabolic-intermediate complex (shaded bars) concentrations were determined. The data represents the mean ± S.E.M. for 3 rats from each drug treatment. * and ** are the significant differences in total cytochrome P-4S0 p < 0.05 and p < 0.001, respectively, from control (t test for independent data). 30 ~Il) ~e( :::l(f) ....We( a: a. en 9 0 CD W W W * ....~ *'" (J ~ o N ~ ( -IOJd -6wjSelowu J 09""d eWOJ4 ooJ,(O d 31 significantly. Thus, the uniqueness of troleandomycin appears to reside in the acetylation of the R3 positions. Spiramycin and leucomycins A3 and AS have a different ring structure (shown in the Appendix) compared with the other macrolide antibiotics, but shared in common the desosamine moiety. None of these compounds with the differing ring structure produced HI complex formation, but the leucomycins appeared to induce cytochrome P-450 slightly. Although nei ther erythromycin nor oleandomycin induced cytochrome P-450 to the same extent as troleandomycin when administered in identical doses, Danan et ale (1981) and Pessayre et ale (1982) showed that by increasing the erythromycin dose to 1468 mglkg and oleandomycin dose to 2752 mglkg, cytochrome P-450 concentrations were induced two-fold over uninduced concentrations. Even with high doses, however, the extent of cytochrome P-450 induction was much lower than that seen with These troleandomycin. latter data further support the difference in potencies observed among macrolide antibiotics with respect to induction of hepatic cytochrome P-450. Pessayre et ale (1982) suggested that the weak induction response to oleandomycin compared with solubility, identical. troleandomycin because the liver was related concentration to of its lower both lipid drugs was Acetylation of the alcohol groups of oleandomycin, which produces troleandomycin, enhances the lipid solubility of the latter compound (Weinstein, 1965). Although a gross relationship between lipid solubility and ability to induce microsomal enzymes has been observed (Conney, 1967), the rate of N-demethylation of erythromycin derivatives (base, esters and N-oxides) did not correlate with their 32 overall lipid solubilities (Mao and Tardew, 1965). In the present study, ester erythromycin estolate, which contains an in the desosamine sugar, has a lipid solubility similar to troleandomycin and was therefore predicted to induce cytochrome P-450 to the same extent. Experimentally, however, erythromycin estolate was significantly inferior to troleandomycin in its ability to induce cytochrome P-450. Therefore, lipid solubility of these compounds alone does not predict the ability to induce cytochrome P-450 or form a MI complex. The ester group, however, may influence the localized binding of the compound to the active site of specific cytochrome P-450 subpopulations, which catalyze the N-demethylation reaction. CHAPTER 5 METABOLIC-INTERMEDIATE COMPLEX FORMATION IN VITRO FROM TROLEANDOMYCIN The nature of metabolic-intermediate from troleandomycin was further (HI) complex formation studied in vi tro with uninduced, S -naphthoflavone-induced rat liver microsomes. phenobarbital- and In this way, the importance of an induced state of cytochrome P-450, as well as the type of cytochrome induced, (B-naphthoflavone induces cytochrome P-448, phenobarbital induces cytochrome P-450) was ascertained. The generation of cytochrome p-450 HI complexes· from troleandomycin and erythromycin (base) in a variety of microsomal fractions is shown in Figure 7 (numerical data in Appendix). The extent of MI complex formation from the macrolide antibiotics was compared with that seen in the same microsomes from norbenzphetamine and SKF 8742-A. These are compounds that classes of compounds which form HI complexes. represent two other It was found that neither macrolide antibiotic formed a HI complex in uninduced or S- naphthoflavone-induced rat liver microsomes and only about 10% of the cytochrome was complexed in phenobarbital-induced microsomes. In contrast, SKF 8742-A complexed 40% and norbenzphetamine 30% of cytochrome P-450 in uninduced microsomes and greater amounts of Figure 7. Extent of metabolic-intermediate complex formation from troleandomycin, erythromycin base, SKF 8742-A and norbenzphetamine in vitro with uninduced, phenobarbital- and B-naphthoflavone-induced rat liver microsomes. Cytochrome P-450 concentrations for control (C), phenobarbital (PB)- and S-naphthoflavone (BNF)-induced rat liver were 0.94 Z 0.03, 1.96 ± 0.11 and 1.81 ± 0.05 nmoles/mg microsomal protein, respectively. The extent of metabolic- intermediate complex formed with cytochrome P-450 is shown for control (unshaded bar), phenobarbital-induced (stippled bar), and S-naphthoflavone-induced (hatched bar) rat liver microsomes. Optimum substrate concentrations for troleandomycin (TAO) and erythromycin base (EB) was 133 lJ M, for SKF 8742-A, 33 lJ M and for norbenzphetamine (NB) , 100 lJM. The data represents the mean ± S.E.M. for 4 rats. 35 •......................................:.:.:.:.:.:...:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:...:.:.:........ . ...--...:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.: ... . ........ ... .................. ..:..:•.. A-a::t a:I Z (J <C N• .:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.: :.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:. CD A- rt CO (J LI. ~ U) u. z CD a::t A- CD W (J u. Z o o U') ...... d I UI8JOJd ·.:.:.:.:.:.:.:.:.:. ········· CD CD :.:.:.:.:.:.:.:.:.: A- U') o o d o ·SW/S810 WU ) ~ OSp-d .WOJ4~oJA:) 0 ~ 36 the cytochrome, 46% and 55%, respectively, in phenobarbital-induced micro somes • Thus, HI complex formation from macrolide antibiotics required an induced subpopulation of cytochrome P-450. Other amine substrates readily formed a complex with both phenobarbital-induced and uninduced rat norbenzphetamine, liver however, than P-450 cytochrome microsomes. Both complexed much the more SKF 8742-A of the and induced antibiotics. macrolide That troleandomycin complexed the cytochrome P-450 it induced in vivo and formed HI an complex with phenobarbital-induced rat liver microsomes in vitro, suggests that common subpopulations are induced by both troleandomycin and phenobarbital. Differences in oxidative drug metabolism among various organs and animal species are well documented. Phenobarbital-induced rabbit and rat liver were therefore compared for their ability to form HI complexes from the macrolide antibiotics (Figure 8, numerical data in Appendix). Erythromycin base complexed about 10% of the cytochrome P-450 from both phenobarbital-induced rabbit and rat liver microsomes. These two species differ, however, in their ability HI to form a phenobarbital-induced complex total from cytochrome troleandomycin; P-450 microsomes and 10% in rat liver microsomes. in rabbit 2% of liver The lack of HI complex formation in uninduced animals of either species clearly indicates that the macrolide antibiotics were not metabolized to HI complexes by the same cytochromes norbenzphetamine or SKF 8742-A. that were able to metabolize After phenobarbital induction, the Figure 8. The extent of metabolic-intermediate complex formation in vitro with phenobarbital-induced rat and rabbit liver microsomes. Cytochrome P-450 concentrations for rat and rabbit phenobarbital- induced liver were 1.96 ~ 0.11 and 2.56 ~ 0.28 nmoles/ mg microsomal protein, respectively. The extent of cytochrome P-450 complexation (% MI complex of total cytochrome P-450) from troleandomycin (TAO), erythromycin base (EB) and its two metabolites, desmethylerythromycin base (EB-1CH ) 3 and didesmethylerythromycin base (EB-2CH ), SKF 8742-A, and 3 norbenzphetamine (NB) were compared using phenobarbital-induced rat (unshaded bar) and rabbit (stippled bar) liver microsomes. Optimum substrate concentrations for the in vitro testing were 133 ~M for the macrolide antibiotics, 33 UK for SKF 8742-A and 100 ~M for NB. The data represent the mean'~ S.E.M. for 3 animals. 38 ..•••••••••••••••••••••••••••••••• .............................. ................................................•...•...•...•...••... CD Z ...... ...... ....-..-. ..... .......... :.:.:.:.:. ................ :.:.:.:.:. CD w o ...c o ,.. (OStp-d -.LAO 1Y.LO.L :JO %J X:a1dWOO IW o 39 amount of cytochrome P-450 able to form a MI complex with the macrolide antibiotics was much smaller than that able to complex with norbenzphetamine or cytochrome (s) induced by SKF 8742-A in both B-naphthof1avone was species. (were) The not able to form MI complex from the macrolide antibiotics. Erythromycin is known to concentrate in the lung (Fraschini et a1., 1980), and hence, it was of interest to compare the lung and liver from the same animal for their ability to complex cytochrome P-450 in vitro (Figure 9). Unlike the liver, the lung from a phenobarbital-induced rabbi t was unable to form any detectable MI complexes from Appendix). tro1eandomycin or erythromycin (numerical data in Thus, the macrolide antibiotics resembled the SKF 525-A class of compounds, but differed from the amphetamine class, in which norbenzphetamine complexes about 35% of lung cytochrome P-450. The primary and secondary amine derivatives of erythromycin base behaved similar to the parent amine, in both phenobarbital-induced rat liver, rabbit lung and rabbit liver. the It was concluded that the possible absence of a cytochrome able to N-demethy1ate erythromycin was not the reason for the absence of MI complex formation in the uninduced liver microsomal fractions. That both tro1eandomycin and erythromycin base behaved almost identically in forming MI complexes in vitro was surprising, since in the present study, erythromycin base induced much less cytochrome P-450 and formed much less MI complex in vivo than tro1eandomycin when given at the same dose. The in vivo data agree with previous studies (Danan et a1., 1981; Pessayre et a1., 1981a), in which much Figure 9. The extent of metabolic-intermediate complex formation in vitro with phenobarbital-induced rabbit liver and lung microsomes. Cytochrome P-450 concentrations in the phenobarbital-induced rabbit liver and lung were 2.56 ± 0.28, and 0.24 ± 0.04 nmoles/mg microsomal protein, respectively. The extent of cytochrome P-450 complexation (% MI complex of total cytochrome P-450) from troleandomycin (TAO), erythromycin base (EB) and its two metabolites, degmethylerythromycin bas~ (EB-1CH ) 3 and didesmethylerythromycin base (EB-2CH ), SKF 8742-A, and 3 norbenzphetamine (NB) were compared uS1ng the ability of phenobarbital-induced rabbit (left) and lung microsomes (right) to form a metabolic-intermediate complex. Optimum substrate concentrations for all macrolide antibiotics was 133 ~M, for SKF 8742-A, 33 ~M and for NB, 100 ~M. The data represents the mean ± S.E.M. for 4 animals. LIVER LUNG 4 0 It) ~ a.• .,.: >- 30 0 ..J ~ 0 t- u.. 0 20 ~ )( W ..J a. 2 0 10 0 ~ O--------~----------------~------~------------------------~~-L------TAO EB EB.. EB- SKF NB TAO EB EBEB- SKF NB 1CH3 2CHa 8742-A 1CH3 2CH3 8742-A ~ ....... 42 larger doses of erythromycin (1468 mg/kg) compared with troleandomycin were required to produce maximal MI complex formation and induction in vivo. Other antibiotics containing amine groups were also tested for MI complex microsomes. formation with phenobarbital-induced rat liver Five aminoglycosides, the antischistosomal antibiotics lucanthone and hycanthone, as well as chloroquine, doxycycline, and tetracycline (structures in Appendix) were investigated. None of these drugs were able to form MI complexes in vitro (Table II). Viability of the microsomes was established with norbenzphetamine, which complexed about 70% and troleandomycin, which complexed 10% of the phenobarbital-induced cytochrome P-450. Macrolide antibiotics appear containing antibiotics to be unique among the amine investigated in their ability to form MI complexes with phenobarbital-induced rat liver microsomes. A compilation of these data together with other information (Hines and Prough, 1980; Franklin, 1977) highlights the differences between the classes of cytochrome P-450 MI complexes (Table III). The extent to which the macrolide antibiotics complex phenobarbital-induced rat liver cytochrome P-450 in vitro is less than the other classes of MI complex forming substrates. Further differentiation of macrolide antibiotics from other classes of MI complex forming substrates was possible by using their inability to form complexes in vitro with uninduced liver and/or lung microsomes; no other class failed to show MI complex formation in both. 43 Table II. The ability of amine containing antibiotics to form metabolic-intermediate complexes with phenobarbital-induced a rat liver microsomes in vitro. MI Complex Formation (nmoles/mg microsomal protein) Substrate b Norbenzphetamine Troleandomycin c C 1.36 .±. 0.09 0.20 .±. 0.02 Aminoglycosides gentamycin C 1 0.00 tobramycin 0.00 kanamycin 0.00 streptomycin 0.00 neomycin 0.00 Hycanthone 0.00 Lucanthone 0.00 Chloroquine 0.00 Doxycycline 0.00 Tetracycline 0.00 acytochrome P-450 concentrations were 1.96 .±. 0.11 nmoles/mg microsomal protein for 4 animals. bthree concentrations of the substrates (33,100 and 133 uM) were tested to optimize the formation of metabolic-intermediate. complex. COptimum substrate concentrations for reference substrates were 100 uM for norbenzphetamine and 133 uM for troleandomycin. Table III. Properties of the classes of cytochrome P-450 metabolicintermediate complexes Class of HI Complex Substrate ExtensivelY Studied Nitrogen Containing Cytochrome P-450 Inducing Ability + Absorbance Maximum Pe++ (RII) Stability to Ditbionite 427, 455 + - (438) 427 + + Dissocation Fe+++Fe+++ Methy1enedioxybenzene isosafro1e Dioxo1anea 4-n-buty1dioxo1ane Amphetamine norbenzphetamine + 455 + + Oxidized Alkylamine N-hydroxyamphetamine + 455 + + SKF 525-A SKF 525-A + 452 + + Ary1amine p-ch1oroani1ine + 448 + Hydrazine b N-aminopiperadine + 449 - (438) Macrolide Antibiotic C Tro1eandomycin + + + 456 + + aFrom Dahl and Hodgson (1979) b From Hines and Prough (1980) cProm Pessayre et a1. (1981) Other information from M.R. Franklin (1977) .p.. .p.. 45 Three classes compounds of which form HI complexes, methylenedioxyphenyl compounds, SKF 525-A and macrolide antibiotics, were compared for their ability to form an in vivo HI complex in untreated or phenobarbital- or S-naphthoflavone-pretreated animals. Troleandomycin formed less HI complex than isosafrole and SKF 525-A in untreated pretreated rat (9; 18 ;29%, liver respectively) (1;44;12%, and respectively). S -naphthoflavoneTroleandomycin, however, formed about the same amount of MI complex as SKF 525-A in phenobarbital-pretreated rats in vivo (about 25%), which was greater than that seen with isosafrole (7%). The rabbit lung did not support HI complex formation in vitro from troleandomycin or SKF 525-A, but a complex from piperonyl butoxide can be detected. It is clear from these data that although the macrolide antibiotics have some similarities to other established classes of HI complexes, significant differences exist which can be used to define these drugs as anew, distinct class of HI complex forming substrates. CHAPTER 6 INDUCTION OF CYTOCHROME P-450 SUBPOPULATIONS BY TROLEANDOMYCIN The nature of the cytochrome P-450 induced by troleandomycin was compared with that induced by SKF 525-A and the classical inducers Troleandomycin induction phenobarbital of and cytochrome S-naphthoflavone. P-450 and metabolic-intermediate (HI) complex formation in vivo in rat liver microsomes is shown in Figure 10 (numerical data in Appendix). Total cytochrome P-450 (free cytochrome + MI complex) was not increased significantly during the first three to eight hours. After 24 hours, however, the cytochrome concentration was increased 100% over control, with an amount equal to 75% of the induced cytochrome sequestered as a MI complex. Four days of daily troleandomycin administration produced a three-fold increase in total cytochrome P-450 concentration, again with 80% of the increased amount of cytochrome in the complexed state. Thus, there was very little increase in the amount of cytochrome present in the uncomplexed state. The time course of induction by SKF 525-A differed from that seen with troleandomycin. P-450 Twenty-five percent of the cytochrome was sequestered as a SKF 525-A MI complex within one hour, and by eight hours, about 25% induction of total cytochrome P-450 had occurred (Bornheim, Peters and Franklin, unpublished Figure 10. Time course of rat hepatic cytochrome P-4S0 induction by troleandomycin in vivo. Animals were sacrificed 3, 8, 12, or 24 hours after a single dose or 24 hours after 4 daily doses of troleandomycin (SOO mg/kg/day, i.g.). Uncomplexed cytochrome P-4S0 (unfilled circles) and total cytochrome P-4S0 (uncomplexed + metabolic-intermediate complex, filled circles) were monitored in all liver microsomes. The shaded area represents the metabolic-intermediate complex present in the samples. The data represents the mean ~ S.E.M. for S rats at each t~e point. * and ** represent significant difference from control, p. < O.OS and p < 0.001, respectively (t test for independent data). 48 •......................... :.::::::::=:.:.:.::::::::.: .:.:.. ....... ....:.... :.:.:.:.:.:..:............. :.:.:.:.:.:.:. :.:............ ............ :.: ............ .......... :.............. :...:.:... ...:.:.:.: ........ :.:.:.:..... :.:.:.:.::::::.:.:.:.:.:::: •• I •••••••••• ............................. :.:.:.:......:.:.:.:.:.... ........................... .......................... ......................... ......................... ......................... :.:.:.:.::.:.:.:.:.:.:. .....:::.:.: ..........::.:.: :.:. .:.:.:.:.:.:. ...... :.:.:.:.:..... ................ :.:............ :.:... : .... ...... ....... .:.:.:.:..:.:.:.: :.:.:.:.:.:. ...... :.:.:.:. .................... :.:.:....... ::.:.:.:.:.:::: ............. ............. .:::::::::::: .....•...." .:.:.:.:.: ........... .......... ..... :.:.....:.. ..... .......... ........ ...:.:.: ....:.. ......... . ........ ...... .:.:.: ....... .:.:.:. ~ -- -~ Q -'C ........ Q e ~ N 0 0 an ->Z ( .) :e 0 .···.·.......... ····..··......... ·....·..... ·.....--:: Q Z <C w ,.. ..J 0 a: t- a: W I- u.. <C (/) a: ~ 0 :I: ----~--------~--------~~------~o o 0 o o cw) N (·IOJd -6W/S810WU ) . o 49 observations). This induction increased to 50% at 24 hours. The induction caused by SKF 525-A is seen at earlier time periods than with troleandomycin. After four days of treatment, SKF 525-A, like troleandomycin, produced a three-fold increase in total cytochrome and sequestered an amount equal to 65% (slightly less than troleandomycin) of the amount of cytochrome P-450 induced as a MI complex (Buening and Franklin, 1976). Another method of establishing the uniqueness of troleandomycin induction is the inducers, such as troleandomycin. which are utilization of combined phenobarbital or treatment of known 8 -naphthoflavone with If the two agents induce forms of cytochrome P-450 under separate control mechanisms, then the co-administration of both compounds at maximally inducing doses, should produce concentrations of cytochrome P-450 that are equal to the sum of that seen with each agent individually. In contrast, if the two agents induce the same subpopulations, the administration of the two drugs together in maximally inducing doses should not increase the cytochrome P-450 concentration above that seen from each agent alone. Induction by troleandomycin after maximal induction by B-naphthoflavone followed an essentially similar time course to that seen in uninduced animals, there troleandomycin induction after being three little MI or eight hours, complex or but large quantities of both after 24 hours (Figure 11, numerical data in Appendix). After maximal induction by phenobarbital, the presence of MI complex three hours after troleandomycin administration, a Figure 11. Effect of phenobarbital and S-naphthoflavone pretreatment on metabolic-intermediate complex formation from troleandomycin in vivo. Uninduced (TAO), phenobarbital-induced (PB + TAO) and 8naphthoflavone-induced (BNF + TAO) rats received a single 500 mg/kg dose, i.g., of troleandomycin (TAO). Rats were sacrificed 3, 8 or 24 hours after troleandomycin administration, and monitored for uncomplexed (unshaded bar) and complexed {stippled bar} cytochrome P-450 concentrations. The data represents the mean ± S.E.M. for 5 rats at each time point. 51 ~ .. __~____W-________________~N co o ~ + LI. o co Z a::a o ~ + a:I o a. o C I- o 0 c-i N lUl810Jd -fSwjselOwU J OS'l--d o c:i 52 period too short for significant induction to occur, demonstrated that a cytochrome subpopu1ation induced by phenobarbital was able to form a MI complex from tro1eandomycin. the in vitro study (Figure 7), about 10% of in which tro1eandomycin comp1exed induced tro1eandomycin MI complex formed in the three hour time period with phenobarbital induced If the cytochrome P-4S0 phenobarbital microsomes. the These data correlated with in concentration of microsomes is subtracted from the amount of MI complex present at later time periods, the increase in MI complex formation with time parallels that seen in previously uninduced (naive) animals. In addition to MI complex formation, induction of cytochrome P-4S0 by tro1eandomycin also occurred. Thus, induction by tro1eandomycin appeared to be independent of previous induction, with most of the tro1eandomycin-induced cytochrome present as an MI complex. The uniqueness of induction of hepatic cytochrome P-4S0 by tro1eandomycin was also investigated by comparing the ability of tro1eandomycin-induced cytochrome(s) norbenzphetamine uninduced with induced by phenobarbital. induced by to form cytochrome MI complex populations or from those The ability of the cytochrome P-4S0( s) tro1eandomycin to form an norbenzphetamine in vitro is shown in Table IV. MI complex from Since most of the cytochrome P-4S0 induced by tro1eandomycin after either one or four days was sequestered as an MI complex, the amount of cytochrome able to form an MI complex from norbenzphetamine in vitro in native micro somes Destruction did not change from control. of the Table IV. Ability of induced and uninduced microsomal cytochrome P-450 to form a metabolic-intermediate complex from norbenzphetamine in vitro Native Microaomea Treatment Cytochrome P-450 a Microsomes Treated with K,Fe(CN)6 Maximum No rbenzphetamine MI Complexa"d Cytochrome P-45OS Maximum No rbenzphetamine HI Comp lexS. d 0.23 + 0.01 (24%) Control 0.86 ± 0.02 0.31 + 0.05 (36%) 0.94 + 0.03 Troleandomycinb (1 day) 0.96 ± 0.01 0.21 + 0.04 (22%) 1.19 ± 0.04* 1t 0.42 + 0.05 (23%)* Troleandomycinb (4 days) 1.12 ± 0.101- 0.33 + 0.06 (30%) 2.79 ± 0.08* * 0.89 ± 0.04 (32%)** t 1. 18 ± 0.08* 1.11 + 0.08 (62%) * 1.96 + 0.13 it 1.00 ± 0.10 (51%) t Phenobarbita1 C (4 days) anmoles/mg microsomal protein; mean b ± S.E.H., n = 4. 500 mg/kg in 1% methylcellulose, i.g. c 80 mg/kg, l.p. d Parenthesis indicate % of cytochrome in the HI complexed state. * p < 0.05 significant difference from native microsomes ** p < 0.001 significant difference from native microsomes t p < 0.05 significant difference from control * p < 0.001 significant difference from control Ln W 54 troleandomycin HI complex with potassium. £erricyanide revealed the full extent of cytochrome P-450 induction by troleandomycin. Some cytochrome P-450 released by ferricyanide was able to form a HI complex from norbenzphetamine, but the percentage of total cytochrome P-450 able to form that complex (23-32%) was similar to that seen from microsomes without ferricyanide treatment (20-30%) or all the cytochrome present in uninduced (36%) microsomes. It was much less than the amount of cytochrome P-450 that was able to form a norbenzphetamine HI complex after induction by phenobarbital (60%). To confirm the difference between phenobarbital- and troleandomycin-induced cytochrome P-450 subpopula tions, solub il ized microsomes were subj ected to DEAE cellulose column chromatography (Figure 12). Phenobarbital-induced cytochrome P-450 subpopulations eluted in DEAE fractions I, II and III. Troleandomycin, which induced cytochrome P-450 to a greater extent than phenobarbital, induced subpopulations which eluted in only two fractions, I and II (numerical data in Appendix). which induced (Franklin, subpopulations Pershing and This was in contrast to SKF 525-A, in eluted Bornheim, in 1982). fractions II Induction and III by tro- 1eandomycin was also different from that seen with 8-naphthoflavone, in which the increased amount of cytochrome P-450 was also found in fractions II and III (Franklin et al., 1982). While SKF 525-A produced HI complexes which eluted in fractions II and III, the HI complexes from troleandomycin were found in fractions I and II. Inspection of the elution of both MI complexes revealed that Figure 12. DEAE cellulose column chromatography of uninduced, phenobarbital-induced and tro1eandomycin-induced rat hepatic microsomes. Microsomes from uninduced (filled circle), phenobarbital-induced (filled square) or troleandomycin-induced (filled triangle) rats were solubilized and subjected to chromatography. Cytochrome P-450 was eluted in 5 m1 increments from the DE!! celluose column with a 0 - 0.25 M sodium chloride (NaCl) gradient and was pooled into four fractions (I, II, III, IV). The data represents the mean ± S.E.M. of total cytochrome P-450 (umoles/mg microsomal protein) from 3 microsomal preparations in each treacment. 56 r> o CO - Z w Q c o -*--- t== t- ... <0 a: C!) (j co z :E It) Co'! o _ . _ ..... * ..rt::-._. -- ............. --- . -0_.a:-..--.. - ~ ~ g _ .-.-.--:~ ...... o u. I I o a: ~ w 2 ",~/ III ",,"'./ I(~,. :> z ' ....... ,","", w .~ ~ III :) I- ,...". ,... (U!810Jd ·Sw/S810 W U J OS" - d 0. o 8WOJl.I:)OI~:) 57 they eluted in the DEAE fractions which showed induction of cytochrome P-450. Closer fractions examination after of the troleandomycin subpopulations induction within the undertaken was SDS-PAGE (Figure 13, numerical data in Appendix). DEAE using After induction by troleandomycin, the major protein band in DEAE fraction I was 52,000 daltons. This major protein band did not correspond to any of the major bands that were present either after phenobarbital induction (49,000, 51,000, and 55,000) or in uninduced (control) microsomes (50,000 and 53,000). In fraction II, the major molecular weight bands were 49,000, 52,000 and 55,000 in uninduced, a doublet of 52,000 and 55,000 in phenobarbital-induced and 50,000, 54,000 and 58,000 in troleandomycin-induced microsomes. troleandomycin, After induction by the 52,000 molecular weight region band that was present in DEAE fraction II of control microsomes was no longer apparent, and an additional high molecular weight band at 58,000 was evident. In fraction III, contained a 50,000 the troleandomycin-induced microsomes molecular region band like control and phenobarbial-induced, a 55,000 band like phenobarbital-induced, and a 58,000 band like control microsomes. The troleandomycin-induced microsomes lacked, however, the 52,000 molecular weight band that was present in control microsomes. Thus, changes in molecular weight occurred despite no change in total cytochrome P-450 eluting in fraction III. after error) In fraction IV, the molecular weight bands present troleandomycin induction corresponded to bands that were present in (within experimental both control and Figure 13. DiagraDm8tic representation of the microsomal proteins present after SDS polyacrylamide gel electrophoresis of DEAE eluate fractions. The fractions eluting upon DEAE cellulose chromatography of solubilized microsomes from uninduced (CON), phenobarbital (PB)- and troleandomycin (TAO)-induced rat livers were subjected to SDS polyacrylamide gel electrophoresis. The molecular weights were calculated in each slab gel from standards run in the same gel. The average molecular weight for each protein band from six gels was then determined. The intensity of the protein staining is indicated as most intense (solid), intermediate (stippled) or least intense (open). Only the protein bands between 49,000 and 60,000 are shown. 59 ~ co Cl.. 0 F > - 0 0 Z 0 () -- 0 0 ~ ~ 0 0 .0 0 0 0 0 co Cl.. ~ () 0 ~ B 0 I 0 O. 0 00 g oI I 0 0 I m ~ 0 0 g 0 co a. () g co a.. z 0 () I oCD t II I I I I CDI I I I I I I I ~ N 0 co II) II) II) II) OOOpc:lM"10W II) 60 phenobarbital-induced induction, microsomes. Thus, after troleandomycin several additional proteins not seen in controls were present in the liver microsomes, and these eluted in three of the four fractions. In addition, there was at least one protein band present after troleandomycin induction which was not apparent after phenobarbital induction, in all fractions except fraction IV. In rabbit, a single constitutive form of cytochrome P-450 (1M3 ) appeared to be induced by troleandomycin (Bonfils et al., 1982). In rat, however, the largest increases in cytochromes were seen in DEAE fractions I and II with at least two major new protein bands in these fractions (52,000 in I, and 58,000 in II). The rat does not therefore, appear to respond to troleandomycin in as singular a manner as does the rabbit. Since troleandomycin was shown to form a HI complex in vitro and in vivo chromatographic with phenobarbital-induced studies were rat performed with phenobarbital-induced-troleandomycin-treated rats which fractions the HI complex eluted. that produced no induction of liver microsomes, microsomes to from determine in A time point (eight hours) hepatic cytochrome P-450 by troleandomycin, yet produced sufficient detectable MI complex (22% of the total cytochrome P-450), was chosen for the evaluation. The troleandomycin MI complex formed in phenobarbital-induced liver eluted in DEAE column fraction II (Figure 14, numerical data in Appendix), which was similar to the elution of the MI complex from troleandomycin-induced liver. It therefore, appears that troleandomycin induces a subpopulation(s) similar to that induced by Figure 14. DEAE cellulose column chromatography of troleandomycin-induced and phenobarbital-induced-eight hour-troleandomycin-treated rat hepatic microsomes. Microsomes were solubilized and subjected to a DEAE cellulose column. Cytchrome P-450 was eluted from the column in 5 m1 increments with a 0 - 0.25 M sodium chloride (NaCl) gradient, and pooled into four fractions (I, II, III, IV). Total cytochrome P-450 (filled symbols) and metabolic-intermediate complex (unfilled symbols) (nmoles/mg microsomal protein) from troleandomycininduced (triangle) and phenobarbital-induced-eight-hourtroleandomycin-treated (hexagon) rat liver microsomes were quantitated. The data represent the mean.::!:. S.E.M. for 3 micro.somal preparations from each treatment group. 62 T -> 0 CO to- Z w s-c 1---- t-+ - 0 tD a: "ca (j z ::I ~. _. - . -- --- . ....... ,--. -.--.. .................... 0 ~ -.-. It) N d ~ -. 0 LI- 0 0 N a: W a:J ::I :;) z J. w 0 CD :;) to- " a. o d (UI8JOJd ·SW/S810WU J OS.,· d eWOJ4ooJAo 63 phenobarbital. found Since there were no common molecular weight bands following SDS-PAGE, they would appear properties, but are not identical proteins. to have similar The MI complex which eluted in fraction I when troleandomycin-induced microsomes were chromatographed appeared to be a cytochrome subpopulation unique to troleandomycin induction. Losses in cytochrome p-450 upon solubilization were greater in the induced microsomes, phenobarbital (23%) and (27%), than uninduced microsomes (7%) (Figure 15). solubilization, losses after chromatography were troleandomycin In contrast to sLnilar in all three types of microsomes (19, 14 and 16%, respectively). During solubilization, the precent loss of MI complex in troleandomycininduced microsomes was the same as the percent loss of cytochrome P-450, so that the MI complex still represented 65% of the total. After chromatography, however, the MI complex represented only 30% of the total cytochrome P-450, indicating a greater lability of MI complex compared wi th cytochrome P-450 during the chromatgraphic procedure. Some of the cytochrome originally existing as a MI complex was present after chromatography as free cytochrome. DEAE fractions III and IV of With troleandomycin-induced microsomes containing no MI complex, it could be argued that MI complexes of the cytochrome subpopulations eluting in these fractions the troleandomycin-induced chromatography. microsomes, Quantitatively, however, but existed in dissociated there is during insufficient cytochrome P-450 in fractions III and IV to account for the amount of MI complex lost. Figure 15. The effect of solubilization and DEAE cellulose chromatography on the total cytochrome P-450 and metabolicintermediate complex concentrations from uninduced, phenobarbitaland troleandomycin-induced rat liver. Origi.nal microsomal total cytochrome P-450 concentrations were 0.92 ± 0.16, 2.00 ± 0.09 and 2.93 ± 0.15 nmoles/mg microsomal protein for uninduced (C), phenobarbital (PB) and troleandomycin (TAO) treatments, respectively. Metabolic-intermediate complex (stippled bar) from the troleandomycin-induced microsomes was originally 1.95 ± 0.07 nmoles/mg microsomal protein. Data represents the mean of 3 experiments in each treatment. 65 M icroson1es Solubilization Chromatography 100- i--- .. 0 75- I-- It) .··...... •..... ... ~ • G- •E .. 0 :~::: ·.. :::::: ...... :.:.:. :=:=:: it·.. • • 50- .t:. CJ 0 .., -.. ~~ I-~ •••• i~~ ie~···· ,,~ 25- ,,~ ,,~ ~ 0 .:.:.:. ..... ~ .. ...... ~::::: :::::. .... :.:.:. ~. 0 C PB TAO ....'....................:.... .:.:.:.. ..... ....... ...... ~ ... ........ ...... ~ ~ - t:«: .~:~:~: ..... •..... .... .... "..... .... .... ..... ..... >0. CJ ca c:: r-- ~ ~ ~ ~. ' 'mm C PB TAO ' C PB TAO 66 The lability of MI complex resulted in the quantitation of only 30% of the original MI complex after chromatography. This overall low yield of MI complex could result in a misinterpretation of MI complex formed in vivo from troleandomycin. Microsomes were normally solubilized for one hour and subjected to DEAE cellulose column chromatography. The chromatography process continued for approximately ten hours. was started, the Seventeen hours after the chromatography eluted fractions were analyzed. Since the microsomes were in solubilizing buffer throughout the chromatography and collection period, the possibility existed that solubilization of the MI complex continued over 24 hours. Therefore, parameters that may influence MI complex stability, such as temperature, ionic strength, pH and oxygen tension were investigated in vitro after 4, 8, 12 and 24 hours of solubilization. 1, MI complexes from both troleandomycin and SKF 525-A were studied using micro somes from rats induced with these agents. Reduction of the temperature at which the microsomal solubilization was performed, from 25 0 e to 4 0 e, reduced the loss of cytochrome P-450 and/or MI complex • At 25 o e, there was a 50% loss of hours cytochrome P-450 after 24 of solubilization of micro somes , but not in uninduced or SKF troleandomycin-induced 525-A-induced microsomes (Figure 16, numerical data in Appendix). While MI from complexes solubilization at 25 0 e., both treatments were lost upon only troleandomycin MI complex was lost at Figure 16. The effect of temperature on the stability of total cytochrome P-450 and metabolic-intermediate complex during solubilization • Total cytochrome P-450 (filled symbols) and metabolicintermediate complex concentrations (unfilled symbols) (nmoles/mg microsomal protein) from troleandomycin-induced (circle), SKF 525-A-induced (square), and uninduced (triangle) rat liver microsomes were compared for stability during solubilization (hours). The data represents the mean ± S.E.M. for 2 microsomal preparations from each treatment group. 68 3. c: 2. 'it -e a. 1. 'i .,0E 0.0 e 0.0 0 0 J i 24 Co) i a e ...... ., • '0 E c: . 0 II) t CL • E 2. 2.0 -2 .c: -8 )I'll Co) -Ii" ~ ,,~ O.o-t-..,...-r----r~-......- - o HOURS OF SOLUBILIZATION 24 69 The influence of salt concentration on the loss of cytochrome P-450 and/or MI complex during solubilization was also investigated. A 0.25 Molar sodium chloride gradient was routinely utilized to elute four fractions of cytochrome P-450 from the DEAE cellulose column. Increasing salt concentrations to 0.5 Molar sodium chloride did not significantly alter the stability of total cytochrome P-450 or MI complexed cytochrome P-450 from that seen in the absence of salt (Figure 17, numerical data in Appendix). Altering the pH of the solubilizing buffer from 7.4 to 7.0 and 7.8 did not affect the stability of total cytochrome P-450 in any of the microsomes (Figure 18, numerical data in Appendix). pH did, however, Increased enhance the stability of the troleandomycin MI complex, over short periods of time, but not the MI complex from SKF 525-A. The processes of solubilization normally performed in ambient air. buffers and throughout DEAE cellulose the experiment, and chromatography were Pregassing with oxygen of all support, with continuous gassing increased the rate of degradation of total and complexed cytochrome P-450 from troleandomycin-induced microsomes, but not from SKF 525-A-induced and uninduced (Figure 19, numerical data in Appendix). microsomes Nitrogen-gassing improved the stability of total and complexed cytochrome P-450 from both troleandomycin- and SKF 525-A-induced microsomes. Of the variables stability of cytochrome P-450 tested, nitrogen troleandomycin-induced to the gassing total and greatest extent; however, improved MI the complexed the greatest Figure 17. The effect of salt concentration in the solubilizing buffer on the stability of total cytochrome P-450 and metabolicintermediate complex. Three concentrations of soa1um chloride (NaCI) in the solubilizing buffer (0, 0.2, 0.5 M) were compared for effects on the stability of total cytochrome P-450 (filled symbols) and metabolic-intermediate complex (unfilled symbols) (nmoles/mg microsomal protein) in troleandomycin-induced (circle), SKF 525-A-induced (square) and uninduced (triangle) microsomes during solubilization (hours). The data represents the mean ± S.E.M. for 2 microsomal preparations from each treatment group. 71 ~ :t .... C'II :E N CO d z t= 0 oC N 2CD :::» 0 "'" 0 (I) I&. 0 (I) a: 0 :::» ::c :E 0 d ~~~~~~~-------+o o N o .,: ~ Figure 18. The effect of pH of the solubilizing buffer on total cytochrome P-450 and metabolic-intermediate complex stability • Solubilizing buffer at three pH's (7.0, 7.4 and 7.8) were compared for their influence on the stability of total cytochrome P-450 (filled symbols) and metabolic-intermediate complex (unfilled symbols) (nmoles/mg microsomal protein) in troleandomycin-induced (circle), SKF 525-A-induced (square) and uninduced (triangle) microsomes during solubilization (hours). The data represents the mean ± S.E.M. for 2 microsomal preparations from each treatment group. 7.4 7.0 7.8 _4.0 0. { : 3.0 I05 f6 2 . ..0.. ...; 1.0 )100 CJ 0.0 0 4 8 12 24 0 4 8 12 24 ==-tiJ 24 o 4 8 12 24 0 4 IS 12 24 o 4 8 12 24 ;2. .. ....... .! 2 .e . ~ 1. Q. ...; ~ 0.0 I o 4 i i " 8 12 HOURS OF SOLUBILIZATION '" W Figure 19. The effect of oxygen and nitrogen gassing of the solubilizing buffer on the stability of total cytochrome P-450 and metabolic-intermediate complex. Air, oxygen and nitrogen gassing of solubilizing buffer were compared for the effects on total cytochrome P-450 (filled symbols) and metabolic-intermediate complex (unfilled symbols) stability in troleandomycin-induced (circle), SKF 525-A-induced (square) and control (triangle) microsomes. The data represents the mean ~ S.E.M. for 2 microsomal preparations from each treatment group. °2 AIR N2 4.0 C 'i ... 0 "Go i ~ 1.0 2 0.0 i ali! "• .! 0 4 8 12 24 (I 4 8 12 24 0 4 8 12 24 0 4 8 12 24 0 4 8 12 24 0 4 8 12 24 i c ..,... •E 0 I A. e z:. u ...t' 0 1. o. ........ lJl HOURS OF SOLUBILIZATION 76 losses of MI complex did not occur during solubilization, but during chromatography. Four day mic;rosomes were therefore, treated during chromatography. cytochrome P-450 rat liver solubilized and chromatographed while being exposed to either ambient influence of an oxygen-free troleandomycin air or nitrogen to environment on MI compare the complex stability The elution profiles of total and complexed performed under nitrogen were similar to those exposed to ambient air (Figure 20, numerical data in Appendix). It was concluded that although nitrogen gassing protected MI complex from degradation during solubilization, it did not enhance MI complex recovery during chromatography. These data suggest that a physio-chemical interaction between the MI complex and DEAE cellulose support may occur during the chromatographic procedure, which results in the loss of MI complex. Figure 20. Comparison of nitrogen pregassing and ambient air on the DEAE cellulose column chromatography of solubilized troleandomycininduced rat liver microsomes. Microsomes were solubilized with the particular gassed atmosphere and subjected to a DEAE cellulose column under continual gassing. Cytochrome P-450 was eluted from the column in 5 ml increments with a 0 - 0.25 M sodium chloride (NaCl) gradient and pooled into four fractions (I, II, III, IV). Total cytochrome P450 (filled symbols) and metabolic-intermediate complex (unfilled symbols) concentrations were quantitated in the eluted fractions from air-treated-troleandomycin-induced microsomes (square) and nitrogen gas-treated-troleandomycin-induced microsomes (circle). The data represents the mean Z S.E.M. for 4 microsome preparations from each treatment group. 78 I- 0 eo I- zw Q c( -t-t- a: e!' - 0 CD U as Z :::& .." N d 0 e ~ 0 u. 0 0 N a: W £II 2 - :::::» z ~ W 0 £II :::l l- e!' Q. o M 0 N ~ ~ ~ 0 [U,81 0Jd ·6w/S81~WU) OS~ -d eWOJlI:»ol'<:) CHAPTER 7 PERSISTENCE OF THE TROLEANDOMYCIN METABOLIC-INTERMEDIATE COMPLEX IN VIVO During its hepatic metabolism, troleandomycin is able to complex significant amounts of cytochrome P-450 in vivo and has been reported as contributing to a variety of drug interactions (Hayton, 1969; Mesdjian et al., 1980; Miguet et al., 1980; Weinberger et al., 1977). It was of interest, therefore, to study the duration of the troleandomycin metabolic-intermediate (MI) complex in vivo and the influence that the complexed cytochrome P-450 may have an mixed-function oxidase activities. Column A in Figure 21 shows the time course of troleandomycin induction and column B that of phenobarbital induction (0-4 days of daily drug treatment) as well as the duration of that effect after cessation of treatment (numerical data in Appendix 13). With troleandomycin, maximal induction occured 12 hours after the last of four· daily administrations, at which time 93% of the cytochrome P-450 was present in the complexed state. P-450 diminished over the next four days. The The total cytochrome complexed cytochrome was preferentially lost relative to total cytochrome P-450 in the first two of the four days. The uncomplexed cytochrome P-450 returned to a concentration similar to that of uninduced rats (day 0) by the second day and remained there through day four. Induction Figure 21. The duration of induced cytochrome P-4S0, metabolicinteDDediate complex and mixed-function oxidase activities after troleandomycin and phenobarbital induction of rat liver. Uncomplexed cytochrome P-4S0 (unshaded bars) and complexed cytochrome P-4S0 (stippled bars) concentrations (nmoles/mg microsomal protein) were compared in troleandomycin-treated and phenobarbital-treated rat liver microsomes over a time course of induction (1-4 days) and after cessation of drug therapy (1, 2 and 4 days post). Cytochrome £ reductase, p-nitroanisole o-demethylase and ethylmorphine N-demethylase activities were compared in the above microsomes without K3 Fe(CN)6 treatment, which reflects the activity associated with uncomplexed cytochrome P-4S0 (unfilled circles), and with K Fe(CN) treatment, which reflects total 3 cytochrome P-4S0 activ1ty (fil\ed circles). The data represents the mean ± S.E.M. for at least 3 animals. * and ** are the significant differences from control, p <: 0.05 and p < 0.001, respectively (t test for independent data). 81 A 1.5 -E u::sOl ... B * * -aE !Ci * UI.! '0 ;E us. t., .! 3. 2 . c: -2 ~ G.• ..10 '" U O· day. of TAO 41 2 day. poet TAO 1 2 days of P8 3 ... 10.5 * '" by phenobarbital troleandomycin. followed a similar time course to that of However, the decline of total cytochrome P-450 to control values following maximal induction was complete four days after cessation of phenobarbital, whereas after troleandomycin, it was still twice that of control. NADPH cytochrome £ reductase, p-nitroanisole o-demethylase and N-demethylase ethy~orphine throughout the time course activities of were induction also and quantified recovery. With troleandomycin induction (column A), uncomplexed cytochrome P-450 decreased progressively throughout induction and later returned to control concentrations. p-Nitroanisole O-demethylase and ethylmorphine N-demethylase activities broadly correlated with the changes in free cytochrome P-4S0. remained unchanged throughout Cytochrome £ reductase the time sequence, activity indicating that this enzyme was not altered by troleandomycin induction and MI complex formation. When the ferricyanide, In this Ml complex was degraded in vitro with potassium the mixed-function oxidase activities were enhanced. situation, with all the cytochrome free, ethylmorphine N-demethylation correlated more nearly with the concentration of total cytochrome P-450 than did p-nitroanisole o-demethylation. Phenobarbital-induced animals were compared with troleandomycin animals for differences in the induction of specific mixed-function oxidase activities. Phenobarbital in vivo forms no MI complex; therefore, cytochrome P-450 concentrations measured with and without potassium ferricyanide should be the same. While the pattern and 83 extent of ethylmorphine N-demethylase activity was similar between troleandomycin- and phenobarbital-induced p-nitroanisole o-demethylase phenobarbital, in activity was contrast troleandomycin. to induced four less and Phenobarbital animals, than two the fold with fold troleandomycin with differed therefore, not only in the extent of cytochrome P-450 induction, but also in the mixed-function oxidase activities they influenced. The in vivo metabolism of hexobarbital was also studied during the time course of troleandomycin or phenobarbital induction. Effects of troleandomycin or phenobarbital treatment on hexobarbital sleep times (100 mg/kg hexobarbital, (numerical data in Appendix). treatment and 24 hours after i.p.) are shown in Figure 22 Animals were tested before each drug administration. drug Control animals received no treatment. Control animals did not differ significantly in hexobarbital sleep time throughout the study; therefore, there was no inductive effect on cytochrome P-450 by hexobarbital. Troleandomycin increased the hexobarbital sleep time after the the first day of treatment, but thereafter remained at control values throughout the study. Since the in vivo drug metabolism of hexobarbital after troleandomycin present, nmoles/mg the reflects lower hydroxylation of control animals the uncomp1exed concentrations of microsomal troleandomycin) only were protein still hexobarbital 12 hours able at free to cytochrome cytochrome after four catalyze approximately the the (1.00 nmoles/mg microsomal protein). P-450 P-450 days in (0.6 of vivo same rate as These data Figure 22. The hexobarbital sleep times during and following induction by phenobarbital or troleandomycin. Sleep times (minutes) in rats after hexobarbital (100 mglkg, i.p.) were determined throughout four days of induction with tro1eandomycin {filled circles} or phenobarbital {filled squares} and the following four days of recovery. Hexobarbital sleep ttmes of a group receiving no prior treatment (unfilled circles) were determined simultaneously to elucidate any daily variations or influence of mUltiple hexobarbital injections. Each data point represents the mean ± S.E.M. for at least 5 rats. * represents the significant difference from the untreated group, p<O.OS {t test for independent data}. 85 UJ ...::. ~ c E w - 20 :E t- a. LI.I w ..J UJ 10 * o,-------------------~----------------------1 o 1 2 4 2 3 4 DAYS OF TREATMENT DAYS POST TREATMENT 86 differ from the work by Pessayre et al. (198lb); they reported hexobarbital sleep times were increased significantly above control, in spite of unchanged free cytochrome P-450 concentrations. discrepancy between these data may be due to the The doses of troleandomycin used (814 mg/kg/day x 4 days byPessayre et al., versus 500 mg/kg/day x 4 days in the present investigation), since the hexobarbital dose (100 mg/kg; i.p.) and the concentration of free cytochrome P-450 were identical in both studies. It is also possible, that the higher dose of troleandomycin used by Pessayre et ale (1981b) influenced the induction of particular subpopulation(s) with less hexobarbital hydroxylase activity to a greater extent, or that more troleandomycin was present to act as a inhibitor at the time of hexobarbital administration. competitive Phenobarbital decreased hexobarbital sleep times significantly from control and troleandomycin treatments. Two and four days after the cessation of troleandomycin administration, hexobarbital changed from control values. sleep times were not This is in agreement with the free cytochrome P-450 concentrations measured in both treatment groups. It can cytochrome effects P-450 are selectivity. be inferred and different that troleandomycin mixed~function from oxidase phenobarbital in induces activity, both hepatic but its degree and CHAPTER 8 DISCUSSION The macrolide antibiotics are an important addition to the pharmacological agents used to study cytochrome P-4S0. By several criteria, they constitute a new class of nitrogenous compounds able to form cytochrome P-450 metabolic-intermediate (HI) complexes. Their limited reactivity (expressed by MI complex formation) with cytochrome P-4S0 subpopu1ations (none in uninduced and-only 20% of cytochromes induced by phenobarbital) make them useful probes for the investigation of the mUltiple forms of cytochrome P-450. their apparently unique induction characteristics add to Also the ubiquity of the induction process, the complexity of the control of gene expression for cytochrome P-450 and the increasing number of cytochrome subpopu1ations which differ in subtle ways. Antibiotics are frequently part of a mUltiple drug regime but are not generally considered to have a high incidence of adver se reactions with concommitant1y administered therapeutic agents. Two of the macrolide antibiotics, tro1eandomycin and erythromycin, have been cited in several drug interactions (Hayton, 1969; Bartle, 1980; Mesdjian et a1., 1980; Miguet et al., 1980). During combined therapy with theophylline and tro1eandomycin (Weinberger et a1., 1977) or erythromycin (La Force et a1., 1981), the macrolide antibiotics inhibited the clearance of plasma theophylline, 50 and 88 26%, respectively, and caused toxicity. transformation of The inhibited metabolic theophylline in the above studies may be the result of the formation of a macrolide antibiotic MI complex with a cytochrome P-450 subpopulation that metabolizes theophylline. The observation in the present study that troleandomycin complexes more cytochrome P-450 than erythromycin in vivo parallels the difference between the two antibiotics in the above clinical reports. another clinical study, troleandomycin, when administered In with corticosteroids to steroid dependent asthmatic patients, increased the plasma half-life of the corticosteroids and enhanced the clinical efficacy of corticosteroid therapy (Spector et al., 1974; Selenke et al., 1980). The inhibition of oxidative drug metabolism by drugs such as troleandomycin could occur by several mechanisms; a direct, but transient competition between the two therapeutic agents for the active site of troleandomycin cytochrome inhibition. to P-450 cytochrome a as P-450, or following metabolic-intermediate a In addition, complex, a that metabolism sequesters persistent of the noncompetitive the inductive properties of troleando- mycin could-result in replacement of existing cytochrome P-450 subpopulations, which. metabolize other therapeutic agent (s), with subpopulations, which either do not metabolize the other agent(s) or are readily sequestered an a MI complex. While the latter mechanism would only apply to those classes of compounds listed in Table 1 that induce cytochrome P-450, each of the first two mechanisms could occur with compounds from each of the classes. 89 This study antibiotics, has examined especially the from formation the of macrolide MI complex from antibiotic class. Formation of the macrolide antibiotic HI complex required an induced state of hepatic cytochrome P-450. No MI compounds were detected in lung microsomes. differed among complex these The quantity of complex an~al phenobarbital-induced from species; rat liver microsomes formed more HI complex from troleandomycin than rabbit liver microsomes in vitro. Both phenobarbital-induced an~als, however, formed HI complex from. erythromycin to the same extent. The stable nitrogenous to changes HI in complex pH, formed temperature concentration during solubilization, complexes, such as from SKF 525-A. troleandomycin was and sodium chloride similar to other nitrogenous The complex was unstable in an oxygen-rich environment. The induction administration of subpopulations of of cytochrome troleandomycin cytochrome P-450 was P-450 also induced after repeated examined. by troleandomycin differed from those of the classic inducer phenobarbital. ability to form a HI complex from norbenzphetamine, profiles apparent from DEAE cellulose column molecular weights were troleandomycin and phenobarbital all induced a Their their elution chromatography different. The and their While both subpopulation of cytochrome P-450 with a molecular weight of approximately 49,000 daltons, troleandomycin also induced two unique subpopulations with molecular weights of 52,000 and 58,000 daltons. 90 The induced cytochrome P-450 subpopulations and HI complex from troleandomycin were persistent in vivo. Four days after the cessation of drug treatment, total cytochrome P-450 was still 1.5 fold over state. control, with 38% of Troleandomycin cytochrome P-450 appeared to be more in a complexed selective than phenobarbital in its induction of mixed-function oxidase activity, enhancing onlrf ethylmorphine N-demethylase activity, whereas phenobarbital enhanced both this and p-nitroanisole O-demethylase activity. There was no difference between the two agents, however, in their ability to induce NADPH cytochrome £. reductase activity. The alteration of demethylase activity did not reflect the in vivo situation, since much of the cytochrome was in the complexed state and thus, had little or no metabolic activity. Metabolic enzyme activity associated with only the uncomplexed cytochrome P-450 in troleandomycin-induced microsomes was reduced with respect to both N-demethylation persistent after and o-demethylation. drug cessation. The These reduced findings activity was suggest that troleandomycin may influence the metabolism of therapeutic agents which are metabolized by cytochrome P-450 dependent pathways, not only during concurrent therapy, but for extended periods of time after the cessation of antibiotic administration. APPENDIX: NUMERICAL DATA Structure-activity relationships of macrolide antibiotics on the induction of rat hepatic cytochrome P-450 and metabolic-intermediate complex formation in vivo lUng Structure A o AS C~3R- CHJID.~ R3 CH.l"S "4 o Ring Structure 8 CH:. NC 3 ft.t 0 CH3 h lKR etta) 6joQ-N~ ft.t OCH CHO Cti J 3 o t'J 3Z OCH:l li;?y0 "4 0 Compound lUng Structure Rl R2 RJ R4 IS H R, None total cyt. P-4S0· 0.94 !. 0.03 Trole.ndOlll,)'c1n A OCOCII3 H OCOC"J CH J Erythromycin estohte b A OCOC 2"5 CH 3 Oft CZ"S Oft (rythrOllU'cin b4se A Oft C"l Oft (2"S Oft 01eandClll\)'ctn A Oft H 0" CH leucOIIU'cin AJ lellcOIII)'cin AS 8 Oft B ON Sp1rall\Vctn 8 OH ------------- -- anmo1es/mg microsomal protelf.. bdilauryl sulfate salt 3 H OCOCHZCH(CHl'Z oeOC") OCO(CHZ'ZCH Oft 3 OH r" OCOCH OCOC2~5 (CH .OCU2- 2.61 C"3 1.61 HI complexa IMI 0 ±. 0.12 ±. 0.20 0.15 !. 0.04 , 8 1.43 !. 0.16 Cttl 1.31!. 0.10 0.11 !. 0.03 -OCH2• 1.04 :!:. 0.14 0.04 :!:. 0.00 54 4 01. 1.21 ! 0.11 0.009 !.0.002 0It 1.28 ! 0.02 0.00 0 3'2"0CH l 0 0- 0.84 ! 0.09 0.02 !. 0.02 2 0.1 i .. SEH (or n=3. - 1..0 N Formation of cytochrome P-450 MI complexes (in vitro) with control, 8-naphthoflavone, and phenobarbital treated liver and lung microsomes Maximal HI complex formationS from: Animal Rat Rabbit Inducing Agent Tissue Microsomal Cytochrome P-450a none I.iver 0.94 TAO EB EB-lCU, EB-2CH, ± 0.03 0 0 0 0 BNFb 1.S1 + 0.05 0 0 0 0 PIc 1.96 ± 0.11 1.31 ± 0.13 2.56 ± 0.28 None Liver PB None I.ung PB Mean ± ± 0.02 0.20 0.04 ± 0.04 ± ± 0.03 0.17 o o O.lS + 0.003 0.24 l'lnmoles/mg microsomal protein; 0.20 0.01 0.21 ± 0.03 ± 0.03 0.22 o 0.23 ± 0.02 0.17 SKF S742-A !f! 0.31 ± 0.05 0.37 ± 0.05 ± 0.02 LOS ± 0.10 0.91 ± 0.04 o 0.34 ± 0.03 0.09 ± 0.03 ± 0.02 0.S9 ± 0.14 0.24 ± 0.06 o o o o 0.07 ± 0.01 o o o o o 0.06 ± 0.01 o S.E.M •• n~ 3. b ao mg/kg/day x 3 days in corn oil. i.p. cSO mg/kg/day x 4 days. i.p. Abbreviations: TAO = troleandomycln, EB = erythromycin. EB-IC") = desmethyltroleandomycin. EI-2CU3 - didesmethyltroleandomycin. phetamine. SKF 8742-A • ethylaminoethyl-2.2-diphenylvalerate. BNF = B-naphthoflavone. PB = phenobarbital. NB = norbenz\D W Structures of various aminoglycoside antibiotics A = gentamycin CI , B • tobramycin, C· kanamycin, D = streptomycin and E = neomycin. B. A. c. K2 "-il CH3C'HNHCH3 N~ A NH2 NH2 NH2 NH2 D. NHg:H E. 2 OH OH \0 1J1 Structures of some amine containing antibiotics A - lucanthone, B = hycanthone, C = chloroquine, D = doxycycline and E = tetracycline. 97 B. A. c. D. E. NICH3 J2 OH I CONH2 NH2 OH 0 Effects of phenobarbital or S-naphthoflavone pretreatment on metabolic-intermediate complex formation from troleandomycin in vivo Time After Troleandomycin Adminiatrationb (hra.) Animal Pretreatment Cytochrome P-450 a None Total 0 0.94 HI Complex B-Naphthoflavone Total 1.81 HI Complex Phenobarbital Total b 1.02 + 0.04 0 0.04 ± 0.05 it 0 1.96 + 0.11 it 0 HI Complex anmoles/mg microsomal protein; mean ± 0.03 ± S.E.M., n ± 0.01 24 ! .1 1.00 + 0.05 1. 79 ± 0.04* 0.09 ± 0.01 0.64 + 0.06 1.81 + 0.06 it L 71 2.62 0.02 + 0.002 0.07 ± 0.03 it ± 0.005 1. 74 + 0.081- 2.06 0.27 + 0.03 0./~5 ± 0.13 it ± 0.04 0.42 ± 0.18* ± 0.05 2.67 + 0.12** 1.42 + 0.10 t 2.46 + 0.13* t 0.81 + 0.05 = 5. 500 mg/kg in 1% methylcellulose, i.g. c 4 daily doses of TAO * p < 0.05 when microsomal values are compared with the 0 time point. ** p < 0.001 when microsomal values are compared with the 0 time point. t p < 0.05 when microsomal values are compared to untreated microsomal values. it p < 0.001 when microsomal values are compared to untreated microsomal values. \0 00 The fractionation of microsomal cytochrome P-450 by DEAE cellulose chromatography Animal a Pretreatment Microsomal Cytochrome P-450b None 0.92 + 0.16 Phenobarbital 2.00 Troleandomycin d Cytochrome P-450 c eluting in DEAE fractions: % loss on solubilization % loss on chromatography 7 16 0.16 ± 0.09* 23 19 0.4l + 0.04** (T) 2.93 ± 0.15** t 21 14 0.42 ± 0.10** 2.25 ± 0.10 ** 1t (HI) 1.95 ± 0.07** 11- 33 54 0.16 ± 0.05 0.71 ± 0.11 20 7 0.33 + 0.12 1.13 ± 0.12 28 88 0 0.07 ± 0.01 Phenobarbital + (T) 2.20 + 0.17* 8h Troleandomycind (MI) 0.45 ± 0.10 n 1. ± 0.02 ill .!!. 0.56 + 0.10 0.17 ± 0.03 0.05 0.86 + 0.02* 0.62 ± 0.03** 0.08 + 0.00* 0.18 ± 0.041t 0.05 0 0.65 ± 0.03** 0 ± 0.01 ± 0.01 t 0 0.08 ± 0.02* 0 afour daily doses b nmoles/mg microsomal proteinj .mean ± S.E.M., n - 4. cnmoles/mg microsomal protein, adjusted for overall losses of total cytochrome. d T - total cytochrome P-450, HI = cytochrome P-450 complexed with metabolic-intermediate. * p < 0.05 Significance from untreated microsomes (t test for independent data). ** p <0.001 significance from untreated microsomes (t test for independent data). t p < 0.05 ~ p < 0.001 significance from phenobarbital treated microsomes (t test for independent data). significance from phenobarbital treated microsomes (t test for independent data). \0 \0 Apparent molecular weights of microsomal proteins present after SDS polyacrylamide gel electrophoresis of DEAE eluate fractions Pooled Fraction Number a CON 50.963 I 54.075 49.364 II 52.661 54.740 IV ± 431*** 49.022 52,139 55,370 287*** ± 705** ± 680*** ± 598** .!!2 ± 611** ± 440*** 48,748 52.236 ± 884* ± 628** ± 54.832 57.786 ± 50,794 ± 511* 53.990 56,807 ± 844** ± 643* 59.734 ± 1482* 52.579 ± ± 54.809 ± 349* 57,555 ± 631* 49.838 238* ± 560*** 437** 394*** 1520* ± 1156* ± 398* 50,193 ± 512** 50.323 ± 699* 52,760 55.292 ± 307* 55.114 ± 54,755 58,040 ± ± 203* 251* 49.162 52.712 ± 185* 49,072 ± 875* 48.768 ± 1288* * * 51.304 57,407 ± 832* 51.205 ± 523* 58.694 ± 1306* ± 834* 50,047 III ± PB 57.147 562** a from DEAE cellulose column chromatography. = no beON PB ~ treatment; mean ± S.E.K., n > 3. 80 mg/kg/day x 4 days, i.p.; mean ± TAO" 500 mg/kg/day x 4 days, 1.g.; mean *. ** and *** S.E.M •• n > 4. ± S.E.M •• n > 3. superscript representing intensity of molecular weight band ranging from 1. 2 and 3. respectively. ....... o o The effect of temperature on the stability of total cytochrome P-450 during solubilization Total Qltochrome P-450 a A. Hours of Solubilization TemEerature 0 1 4 8 12 24 25°C 0.87 0.87 0.93 + 0.13 0.90 + 0.03 1.12 + 0.24 0.85 + 0.11 4°C 0.69 + 0.19 1.03 + 0.09 0.90 + 0.04 0.90 + 0.02 0.98 + 0.02 0.96 + 0.02 25°C 2.16 + 0.25 2.13 + 0.21 2.00 2.07 + 0.40 2.14 + 0.30 1.87 + 0.37 4°C 2.17 + 0.26 1.87 + 0.19 1.91 + 0.40 1.88 + 0.15 2.00 + 0.38 1.70 + 0.05 25°C 3.98 + 0.07 4.26 + 0.52 3.50 + 0.50 3.17 + 0.57 2.76 + 0.57 1.69 + 0.08 4°C 3.98 + 0.07 4.25 + 0.58 3.89 + 0.50 3.52 + 0.60 3.27 + 0.20 2.74 + 0.71 Control SKF 525-A TAO a nmoles/ml; mean ± S.E.M., n = 2 .. ..... a ..... The effect of temperature on the stability of metabolic-intermediate complex during solubilization Metabolic-Intermediate Complex a A. Hours of Solubilization Temeerature 0 4 8 12 24 Control 25°C 4°C SKF 52S-A 25°C 1.07 + 0.04 0.98 + 0.16 4°C 1.07 + 0.04 1.28 + 0.19 25°C 3.29 + 0.49 4°C 3.29 + 0.49 0.43 + 0.04 0.17 + 0.06 0.00 1.20 + 0.20 1.29 + 0.22 1.15 + 0.16 1.17 + 0.14 2.71 + 0.38 2.44 + 0.66 1.51 + 0.55 0.50 + 0.10 0.39 + 0.16 3.52 + 0.79 2.63 + 0.97 2.54 + 0.53 2.33 + 0.46 1.53 + 0.56 0.84 TAO a nmo1es/m1; mean ± S.E.M., n = 2. I-' o N The effect of salt concentration in the solubilizing buffer on the stability of total cytochrome P-450 a Total Cytochrome P-450b Salt Concentration of Solubilizing Hours of Solubilization Buffer 12 1 8 4 24 0 Control 0.0 M 1.19 + 0.19 0.22 + 0.29 1.17 + 0.18 1.13 + 0.20 0.96 + 0.10 0.79 + 0.03 0.2 M 1.19 + 0.19 1.19 + 0.16 1.09 + 0.23 1.09 + 0.19 0.98 + 0.12 0.88 + 0.08 0.5 M 1.19 + 0.19 1.24 + 0.25 1.08 + 0.17 1.04 + 0.14 1.04 + 0.15 0.92 + 0.11 0.0 M 1.79 + 0.39 1.44 + 0.24 1.41 + 0.19 1.43 + 0.13 1.34 + 0.14 1.26 + 0.22 0.2 M 1.79 + 0.39 1.90 + 0.58 1.46 + 0.19 1.73 + 0.03 1.53 + 0.15 1.57 + 0.25 0.5 M 1.79 + 0.29 1.50 + 0.22 1.54 + 0.08 1.48 + 0.73 1.48 + 0.15 1.58 + 0.19 0.0 M 2.41 + 0.07 1.67 + 0.07 1.42 + 0.18 1.43 + 0.17 0.72 + 0.13 0.60 + 0.11 0.2 M 2.41 + 0.07 1.76 + 0.15 1.61 + 0.10 1.47 + 0.07 1.07 + 0.11 0.88 + 0.02 0.5 M 2.41 + 0.07 1.76 + 0.12 1.52 + 0.10 1.43 + 0.17 1.36 + 0.25 0.99 + 0.16 SKF 525-A TAO a b Sodium chloride nmo1es/mlj mean I-' 0 ± S.E.M., w n = 2. The effect of salt concentration in the solubilizing buffer on the stability of metabolicintermediate complex Metabolic-Intermediate ComE1exb Salta Concentration of Solubilizing Buffer Hours of Solubilization 1 0 4 8 12 24 Control 0.0 M 0.2 M 0.5 M SKF 525-A 0.0 M 0.79 + 0.11 0.72 + 0.13 0.53 + 0.07 0.24 + 0.09 0.05 + 0.05 0.00 0.2 M 0.79 + 0.11 0.86 + 0.23 0.53 + 0.21 0.24 + 0.10 0.07 + 0.07 0.00 0.5 11 0.79 + 0.11 0.73 + 0.29 0.50 + 0.11 0.36 + 0.07 0.18 + 0.09 0.00 0.0 M 1.56 + 0.13 1.11 + 0.15 0.56 + 0.08 0.28 + 0.16 0.11 + 0.11 0.04 + 0.04 0.2 M 1.56 + 0.13 1.23 + 0.24 0.89 + 0.15 0.56 + 0.12 0.28 + 0.09 0.05 + 0.05 0.5 M 1.56 + 0.13 1.18 + 0.12 0.89 + 0.14 0.64 + 0.09 0.35 + 0.07 0.09 + 0.05 TAO aSodium chloride b I-' o .t::-- nmoles/m1; mean + S.E.M., n = 2. The effect of pH of the solubilizing buffer on the sta.bi1ity of total cytochrome P-450 Total pH of Solubilizing Buffer Cyt~chrome P-450 a Hours of Solubilization 0 1 4 8 24 Control 7.0 1.25 + 0.22 1.27 + 0.14 1.33 + 0.23 1.09 + 0.14 1.15 + 0.21 1.05 + 0.24 7.4 1.25 + 0.22 1.32 + 0.24 1.36 + 0.32 1.20 + 0.26 0.91 + 0.07 0.86 + 0.06 7.8 1.25 + 0.22 1.25 + 0.23 1.21 + 0.24 1.11 + 0.21 0.93 + 0.17 0.86 + 0.06 7.0 1.67 1.57 1.66 2.01 1.74 1.77 7.4 1.67 1.73 2.09 1.40 1.74 1.83 7.8 1.67 1.79 1.73 1.71 1.87 1.66 7.0 2.55 + 0.47 2.34 + 0.58 2.23 + 0.54 1.82 + 0.45 1.44 + 0.41 0.96 + 0.13 7.4 2.55 + 0.47 2.41 + 0.27 2.36 + 0.40 1.95 + 0.43 1.65 + 0.38 1.15 + 0.37 7.8 2.55 + 0.47 2.52 + 0.39 2.40 + 0.33 1.96 + 0.46 1.67 + 0.32 0.94 + 0.16 b SKF 525-A TAO a b nmo1es/m1; mean ± S.E.M., nmo1es/m1, n = 1. n = 2. I-' 0 Ln The effect of pH of the solubilizing buffer on the stability of metabolic-intermediate complex Metabolic-Intermediate pH of Solubilizing Buffer Com~lexa Hours of Solubilization 0 1 4 S 12 24 7.0 1.32 1.02 0.65 0.44 0.12 0.00 7.4 1.32 0.73 0.76 0.37 0.14 0.00 7.S 1.32 1.02 0.73 0.34 0.11 0.00 7.0 1.51 + 0.19 1.lS + 0.16 O.SO + 0.14 0.39 + 0.05 0.12 + 0.01 0.00 7.4 1.51 + 0.19 1.13 + 0.59 1.00 + 0.24 0.92 + 0.26 0.59 + 0.23 0.32 + 0.16 7.S 1.51 + 0.19 2.09 + 0.42 1.62 + 0.41 0.98 + 0.26 0.67 + 0.23 0.20 + 0.06 Control 7.0 7.4 7.S SKF 525-Ab TAO a nmo1es/ml; mean b nmoles/m1, n ± = 1. S.E.M., n = 2. j-I 0 0'\ Ibe effect of oxygen and nitrogen gassing of the solubilizing buffer on the stability of total cytochrome P-450 Total Cytochrome P-450a Gassing the Solubilization Hours of Solubilization 0 1 4 8 12 24 Air 0.97 + 0.03 1.40 + 0.10 1.48 + 0.06 0.69 + 0.13 0.60 + 0.01 0.73 + 0.03 02 0.97 + 0.03 1 .. 43 + 0.12 1.29 0.68 0.68 0.81 N2 0.97 + 0.03 1.02 + 0.22 1.06 + 0.15 0.61 + 0.15 0.78 + 0.14 0.92 + 0.30 Air 2.54 + 0.17 2.31 + 0.45 3.28 + 0.57 3.24 + 0.39 3.23 + 0.16 2.04 + 0.36 02 2.54 + 0.17 2.55 + 0.12 3.10 + 0.10 3.16 + 0.02 2.31 + 0.87 1.63 + 0.12 N2 2.54 + 0.17 2.42 + 0.29 3.66 + 0.12 3.32 + 0.40 3.40 + 0.23 2.44 + 0.46 Air 4.09 + 0.94 3.89 + 0.07 3.09 + 0.05 2.82 + 0.11 1.92 + 0.08 0.90 + 0.07 02 4.09 + 0.94 4.13 + 0.35 3.10 + 0.01 1.34 + 0.21 0.91 + 0.18 0.55 + 0.09 N2 4.09 + 0.94 4.59 + 0.03 4.50 + 0.42 4.01 + 0.02 3.62 + 0.06 3.09 + 0.08 Control SKF 525-A TAO a f-I nmo1es/m1; mean + S.E.M., n = 2. 0 ......... The effect of oxygen and nitrogen gassing of the solubilizing buffer on the stability of metabolic-intermediate complex Metabolic-Intermediate Com21exa Gassing the Solubilization Buffer Hours of Solubilization 0 1 4 8 12 24 Air 1.67 + 0.21 1.49 + 0.27 1.16 + 0.37 0.64 + 0.16 0.06 + 0.06 0.00 °2 N2 1.67 + 0.21 1.53 + 0.13 1.61 + 0 .. 14 0.46 + 0.09 0.18 + 0.09 0.00 1.67 + 0.21 1.47 + 0.38 2.03 + 0.29 1.60 + 0.42 1.31 + 0.34 0.76 + 0.19 Air 3.09 + 0.07 3.01 + 0.01 1.63 + 0.30 0.95 + 0.05 0.49 + 0.04 0.03 + 0.03 °2 N2 3.09 + 0.07 3.16 + 0.26 1.55 + 0.22 0.24 + 0.00 0.13 + 0.04 0.00 3.09 + 0.07 3.90 + 0.13 2.99 + 0.24 2.42 + 0.08 2.11 + 0.13 1.37 + 0.07 Control Air °2 N2 SKF 525-A TAO ~ a nmo1es/ml; mean ± S.E.M., n = 2. 0 00 Comparison of nitrogen pregassing and ambient air on the DEAE cellulose column chromatography of solubilized troleandomycin-induced rat liver microsomes None Nitrogen a Cytochrome P-45·0 c Eluting in DEAE Fractions: Hicrosomal Cytochrome P-450 Gas Treatment (T) 2.83 % loss on Solubilization ± 0.12 a a % loss on Chromatography I 15 20 0.30 + 0.09 II 2.30 ± 0.15 (HI) 1. 95 ± 0 ..07 21 65 0.12 + 0.06 0.72 ± 0.13 (T) 2.83 ± O.lSb 20 29 0.12 + 0.30 2.47 ± 0.14 (HI) 2.17 ± 0.15b 21 65 0.00 nmoles/mg microsomal protein (mean ± S.E.H., n bnmoles/ma microsomal protein (mean ± S.E.H., n - 4) 1.06 + 0.07 IV III 0.17 ± 0.03 0.00 0.18 ± 0.05 0.00 0.07 ± 0.05 0.00 0.06 ± 0.01 0.00 = 6) cnmoles/mg microsomal prote:f,n. adjusted for overall losses of total cytochrome P-450 I-' a \0 The duration of induced cytochrome P-4S0, metabolic-intermediate complex and mixed-function oxidase activities after troleandomycin induction of the rat liver NADPII Cytochrome c Reductase- Cytochrome P-450a HI Complex Total Tota1b Paranitroaniso1e 0-demethy1ation Free c Tota1 b Free c Ethylmorphine N-demethylation Totalb Free c Days of Tro1eandomycin d ± 0.11 8.2 ± 1.5 8.8 ± 0.6 0.00 114 ± 9 121 ± 8 0.79 ± 0.16 1.46 ± 0.06* 0.44 ± 0.02 142 ± 18 97 ± 6 0.77 ± 0.09 0.42 ± 0.05. 2.0 1.88 ± 0.12** 1.07 ± 0.05 179 ± 19 113 + 6 0.78 ± 0.06 0.47 ± 0.05* 8.2 ± 1.2 7.4±1.1 3.0 2.25 ± 0.07** 1.64 ± 0.16 161 ± 29 126 ± 15 0.84 ± 0.04 0.68 ± 0.05* 8.3±1.5 5.2 ± 0.2* 0.5 3.08 ± 0.18** 2.90 ± 0.16 173 ± 29* 162 ± 1.08 ± 0.15 0.36 ± 0.07. 14.2 ± 1.8* 5.3 ± 0.50* 1.0 2.86 ± 0.20** 2.26 ± 0.36 131 1.25 ± 0.10 0.43 ± 0.05* 13.9 ± 1.6* 7.4±1.5 2.0 2.55 ± 0.02** 1.38 ± 0.26 186 ± 1 16.5 ± 2.5* 10.4 ± 1.4* 4.0 1.94 ± 0.14* 0.7J ± 0.13 188 0.0 0.95 1.0 0.93 ± 0.11 8.5 ± 1.2 11.1 ± 1.2* Days post Tro1eandomycin ± 17 ± 6 26 108 ± 7 158 ± 16 1.50 ± 0.27* 0.72 131 ± 1.39 ± 0.17* 0.99 16 ± 0.09 ± 0.11 12.5 ± 1.8* 7.0 ± 0.6 anmo1es/mg microsomal protein; mean ± S.E.H •• n = 3. bnmo1es/mg/min activity in microsomes treated with 50 pH KsFe(CN),; mean ± S.E.H •• n 2 3. cnmoles/mg/min activity in microsomes without K,Fe(CN),; mean ± S.E.H •• n ~ 3. d500 mg/kg/day x 4 days, i.g. edays post maximal 4 day induction by troleandomycin. * p < 0.05 significant difference from control (t test for independent data). ** P <0.001 significant difference from control (t test for independent data). I--' I--' o The duration of induced cytochrome P-450 and mixed-function oxidase activities after phenobarbital induction of the rat liver NADPH Cytochrome c lleductase- Cytochrome P-450a Paranitroanisole o-demethylation !1 Fe (CN), treatedb K3Fe(CN), unteated C untreated C treated b ± 0.62± 0.06 Ethylmorphine N-demethylation treated b KsFe(CN)6 untreated C Days of Phenobarbital d ± 0.13 ± 0.35* 2.18 ± 0.35* 160 ± 22 2.30 ± 0.31* 2.48 ± 0.30* ± 1.3 ILl ± 4.9 11.4 ± 2.1 177 ± 2.57 ;!: 0.42* 2.87 ± 0.35* 13.4 ± 2.4 15.3±2.5* 168 ± 18 3.32 ± 0.10* 3.63 ± 0.24** 18.3 ± O.g* 16.9 ± 1. 7* 196 ± 10* 3.92 ± 0.48* 4.00 ± 0.43** 20.6 ± 2.4* 19.6 ± 2.7* 156 ± 16 3.13 ± 0.18** 3.82 ± 0.44** 14.4 ± 0.1* 16.3 ± 1.5* 139 ± 10 1.89 ± 0.06** 2;03 ± 0.09** 6.9 ± 1.0 8.7 0.0 0.68 ± 0.15 113 ± 16 112 1.0 0.96 ± 0.07 154 ± 10 80 2.0 1.03 ± 0.16 142 ± 10 3.0 1.33 ± 0.06* 162 ± 21 218 ± 11* 9 ± 26* 28* 2.15 0.86 7.6 7.3 ± 1.5 11.5 ± 13.7 ± 2.7 2.8 Days post Phenobarbital e ±: 0.09** 0.5 1. 73 1.0 1. 70 ± 0.06** 1]8 2.0 1.50 ± 0.01** 185 4.0 0.95 ± 0.13 169 ~'nmo1es/mg .± 8 ± 19 microsomal protein; mean ± S.£.I1 •• n ~ ± 0.7 l. bnmo1es/mg/min actidty in microaomes in vitro with 50 &.iH K,Fe(CNh i mean ± S.!.H •• n ~3. cnmo1es/mg/min activity in microsomes in vitro without 50 pM K,Fe(CN),; mean ± S.!.H., n ~ 3. dSO mg/kg/day, i.p. eSO mg/kg/day x 4 days. i.p . • P < 0.05 ** p < 0.001 significant difference from control (t test for independent data). significant difference from control (t test for independent data). I-' I-' I-' 112 The hexobarbital sleep times during induction of rats with phenobarbital or tro1eandomycin Hexobarbital Sleeptimea Days of Treatment TroleandomIcinb Control Phenobarbital c ± 1.2 ± 2.5 ± 1.4 22.8 ± 1.0* 22.0 ± 1. 7 0.0 23.0 ± 1.2 25.3 1.0 31.9.:!: 3.0 34.8 2.0 29.4 ± 3.2 31.0 .:!: 1. 8 3.0 29.a ± 2.6 35.9 ± 3.1 17.6 ± 1.0* 4.0 25.3.:!: 1.4 31.2 ± 1. 7* 14.0 ± 3.7* 10.4 .:!: 4.5* 23.2 Days post Treatment d 1.0 3.04 ± 2.5 30.1 ± 2.0 29.a ± 2.3 23.1 ± 1.9* 12.6 ± 4.0 26.6 .:!: 2.2 25.7 ± 2.2 19.6 ± 2.2 2.1 aminutes after 100 mg/kg hexobarbital, i.p.; mean b 500 mg/kg/day; mean -fao ± S.E.M., n mg/kg/day, i.p.; mean ± ~ ± S.E.M., n ~ 5. 5. S.E.M., n,~ 5. d 4 days post maximal induction by tro1eandomycin or phenobarbital. * 3.2* p < 0.05 significant difference from respective control (t test for independent data). REFERENCES Anders, M.W.; Mannering, G.J.: Inhibition of drug metabolism. IV. Induction of drug metaboliSD by 2-diethylamino 2,2-diphenylvalerate HCl (SKF 525-A) and 2,4-dichloro-6-phenylphenoxydiethylamine HBr (Lilly 18947) and the effect of induction on the inhibiting properties of SKF 525-A type compounds. Malec. Pharmacol. 1: 341-346 (1966). Bartle, W.R.: Possible warfarin-erythromycin interaction. int. Med. 149: 985-987 (1980). Arch. Bonfils, C.; Dalet-Beluche, I.j Maurel, P.: Induction by triacetyloleandomycin and partial purification of a L~ form of cytochrome P-450 from rabbit liver microsomes. Biochem. biophys. Res. Commun. 194: 1011-1017 (1982). Bornheim, L.M.j Franklin, M.R.: Metabolic-intermediate complex formation reveals major changes in rat hepatic cytochrome P-450 subpopulations in addition to those forms previously purified after phenobarbital, S-naphthoflavone and isosafrole induction. Molec. Pharmacol. 21: 527-532 (1982). Buening, M.X.; Franklin, M.R.: SXF 525-A inhibition, induction and 452 nm complex formation. Drug Metab. Disposit. !: 244-255 (1976). Celmers, W.D.: Triacetyloleandomycin: Biochemical Correlations. Antibiot. Annu. 277-283, (1958-59). Conney, A.H.: Pharmacological implications of microsomal enzyme inductions. Pharmacol. Rev. 19: 317-366 (1967). Dahl, A.R.; Hodgson, E.: The interaction of aliphatic analogs of methylenedioxyphenyl compounds with cytochrome p-450 and P-420. Chem.-Biol. Interact. 21: 163-175 (1979). Danan, G.; Descatoire, V.; Pessayre, D.: Self-induction by erythromycin of its own transformation into a metabolite forming an inactive complex with reduced cytochrome P-450. J. Pharm. exp. Ther. 218: 509-514 (1981). De Matteis, F.: Drug-induced destruction of cytochrome P-450. Metab. Disposit. 1: 267-272 (1973). Drug 114 De Matteis, F.; Cantoni, L.: Alteration of the porphyrin nucleus of cytochrome P-450 caused in the liver by treatment with allyl-containing drugs. Biochem. J. 183: 99-103 (1979). Elcombe, C.R.; Bridges, J.W. j N~o-Smith, R.H.: Substrateelicited dissociation of a complex of cytochrome P-450 with a methylenedioxyphenyl metabolite. Biochem. biophys. Res. Commun.11: 915-919 (1976). Estabrook, R.W.; Peterson, J.; Baron, J.j Hildebrandt, A.: The spectrophotometric measurement of turbid suspensions of cytochromes associated with drug metabolism; in Chignell, Methods in Pharmacology, Vol. 2, pp.303-350 (Appleton, Century and Croft, New York 1972). Fairbanks, G.; Steck, T.Lj Wallach, D.F.H.: Electrophoretic analysis of the major polypeptides of the human erythrocyte membrane. Biochem. 10: 2606-2616 (1971). Franklin, M.R.: The enzymic formation of a methylenedioxyphenyl derivative exhibiting an isocyanide-like spectrum with reduced cytochrome P-450 in hepatic microsomes. Xenobiotica 1: 581-591 (1971). Franklin, M.R.: The formation of a 455 nm complex during cytochrome P-450-dependent N-hydroxyamphetamine metabolism. Molec. Pharmacol. 1Q: 975-985 (1974). Franklin, M.R.: Methylenedioxyphenyl insecticide synergists as potential human health hazards. Environm. Health Perspec. 14: 29-37 (1976). Franklin, M.R.: Ligand complexes generated during cytochrome P-450 dependent metabolism; in Snyder, Parke, Kocsis, Jollow, Gibson, Witmer, Biological Reactive Intermediates-II, Part A, pp. 165-177 (Plenum Publishing Corp., New York 1982). Franklin, M.R.; Estabrook, R.W.: On the inhibitory action of mersalyl on microsomal drug oxidation: a rigid organization of the electron transport-chain. Arch. Biochem. Biophys. 143: 318-329 (1971). . Franklin, M.R.; Pershing, L.K.; Bornheim, L.M.: Characterization of rat hepatic cytochrome P-450 subpopulations by anion exchange chromatography and metabolic-intermediate complex formation; in Hietanen, Laitenen and Hanninen, Cytochrome P-450, Biochemistry, Biophysics and Environmental Implications, pp. 307-310 (Elsevier Biomedical Press, Amsterdam 1982). 115 Fraschini, F.; Braga, P.; Copponi, V.; Gattei, G.; Guerrasco, E.; Scaglione, F.; Villa, F.; Scarpazza, G.: The tropism of erythromycin for the respiratory system. J. int. Med. Res. 36-40 (1980). ~: Guengerich, F.P.; Strickland, T.W.: Metabolism of vinyl chloride: destruction of the heme of a highly purified liver microsomal cytochrome P-450 by a metabolite. Malec. Pharmacol. ~: 993-1004 (1977). Hayton, A.C.: Precipitation of acute ergotism by triacetyloleandomycin. NZ Med. J. 69: 42 (196"9). Hemsworth, T.; Renton, K.: Depression of theophylline metabolism and elimination by troleandomycin. and erythromycin. Biochem. Pharmac. 39: 1299-1304 (1981). Hildebrandt, A.; Estabrook, R.W.: Evidence for the participation of cytochrome bS in hepatic microsomal mixed-function oxidation reactions. Arch. Biochem. Biophys. 143: 66-79 (1971). Hines, R.N.; Prough, R.A.: The characterization of an inhibitory complex formed with cytochrome P-450 and a metabolite of 1,1disubstituted hydrazines. J. Pharm. exp. Ther. ~i4: 80-86 (1980). lvanetich, K.M.; Lucas, S.; Marsh, J.A.; Ziman, M.R.; Katz, l.D; Bradshaw, J.J.: Organic compounds: their interaction with and degradation of hepatic microsomal drug-metabolizing enzymes in yitro. Drug Metab. Disposit. !: 218-225 (1978). LaForce, C.F.; Miller, M.F.; Chai, H.: Effect of erythromycin on theophylline clearance in'asthmatic children. J. Peds. 99: 153-156 (1981). Laemmli, U.K.: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, Lond. 2il: 680-685 (1970) • Levin, W.; Jacobson, M.; Sernatinger, E.; Kuntzman, R.: Breakdown of cytochrome P-450 heme by secobarbital and other allyl-containing barbiturates. Drug Metab. Disposit. 1: 275-284 (1973). Lowry, O.H.; Rosenbrough, N.J.; Farr, A.L; Randall, R.J.: Protein measurement with Folin Phenol reagent. J. bioI. Chem. 193: 265-275 (1951). Lu, A.Y.H.; West S.B.: Multiplicity of mammalian microsomal cytochrome P-450. Pharmacol. Rev. 31: 277-295 (1980). 116 Mansuy, D.; Delaforge, M; LeProvost, E.; Flinois, J.-P.; Columelli, S.; Beaune, Ph.: Induction of cytochrome P-450 in rat liver by the antibiotic troleandomcyin: partial purification and properties of cytochrome P-450-troleandomycin metabolite complexes. Biochem. biophys. Res. Commun. 193: 1201-1208 (1981). Mansuy, D.; Gans, P.; Chottard, J.-C.; Bartoli, J.-F.: Nitroso-alkanes as Fe (II) ligands in the 455-n~absorbing cytochrome P-450 complexes formed from nitroalkanes in reducing conditions. Eur. J. Biochem. 76: 607-615 (1977). Mao, J. C-H.; Tardew, P.L.: Demethylation of erythromycins by rabbit tissues in-vitro. Biochem. Pharmac. 14: 1049-1058 (1965). Masters, B.S.S.; Baron, J.; Taylor, W.E.; Isaacson, E.L.; Lo Spalluto, J.: Immunochemical studies on electron transport chains involving cytochrome P-450. I. Effects of antibodies to pig liver microsomal reduced triphosphopyridine nucleotide-cytochrome £ reductase and the non-heme iron protein from bovine adrenocortical mitochondria. J. bioI. Chem. 246: 4143-4150 (1971). Mesdjian, E.; Dravet, C.; Cenraud, B.; Roger, J.: Carbamazepine intoxication due to triacetyloleandomycin administration in epileptic patients. Epilepsia 11: 489-496 (1980). Miguet, J.P.; Vuitton, D.; Pessayre, D.; Allemand, H.; Metreau, J.-M.; Poupon, R.; Capron, J.-P.; Blanc, F.: Jaundice from troleandomycin and oral contraceptives. Ann. int. Med. 92: 434 (1980). Nash, T.: The colorimetric estimation of formaldehyde by means of the Hantzch reaction. Biochem. J. 21: 416-421 (1953). Netter, K.J.; Seidel, G.: An adaptively stimulated O-demethylating system in rat liver microsomes and its kinetic properties. J. Pharmacol. exp. Ther. 146: 61-65 (1964). Omura, T.; Sato, R.: The carbon monoxide-binding pigment of liver microsomes: I. Evidence for its hemoprotein nature. J. bioI. Chem. 239: 2370-2378 (1964). Ortiz de Montellano, P.R.; Mico, B.A.: Destruction of a cytochrome P-450 by ethylene and other olefins. Molec. Pharmacol. ~: 128-135 (1980). 117 Ortiz de Montellano, P.R.; Kunze, K.L.; Yost, G.S.; Mico B.A.: Self-catalyzed destruction of a cytochrome P-450:covalent binding of ethylnyl sterols to prosthetic heme. Proc. nat'l. Acad. Sci. U.S.A.l.2.: 746-749 (1979). Ortiz de Montellano, P.R.; Yost, G.S.; Mico, B.A.; Dinizio, S.E.; Correia, M.A.; Kumbara, H.: Destruction of cytochrome p-450 by 2-isopropyl-4-pentenamide and methyl 2-isopropyl-4-pentoate: mass spectrophotometric characterization of prosthetic heme adducts and nonparticipation of epoxide metabolites. Arch. Biochem. Biopyhys. 191: 524-533 (1979). Osol, A.: Remington's Pharmaceutical Sciences, 16th ed. (Mack Publishing Co., Easton 1980). Pessayre, D.; Descatoire, V.; Konstantinova-Mitcheva, M.; Wandscheer, J.-C.; Cobert, B.; Level, R.; Benhamou, J.-P.j Jaouen, M.; Mansuy, D.: Self-induction by triacetyloleandomycin of its own transformation into a metabolite forming a stable 456 nm-absorbing complex with cytochrome P-450. Biochem. Pharmac. 39: 553-558 (1981a). Pessayre, D.; Konstantinova-Mitcheva, M.; Descatoire, V.; Cobert, B.; Wandscheer, J.-C.; Level, R.; Feldman, G.; Mansuy, D.; Benhamou, J.-P.: Hypoactivity of cytochrome P-450 after triacetyloleandomycin administration. Biochem. Pharmac. 39: 559-564 (1981b). Pessayre, D.; Descatoire, V.; Tinel, M.; Larrey, D.: Selfinduction by oleandomycin of its own transformation into a metabolite forming an inactive complex with reduced cytochrome P-450. Comparison with troleandomcyin. J. Pharmco1. exp. Ther. 221: 215-221 (1982). Reynolds, E.S.; Maslen, M.T.; Szabo, S.; Gaeger, R.J.: Vinyl chloride-induced deactivation of cytochrome P-450 and other components of the liver mixed function oxidase system: an in vivo study. Res. COlmIlun. chem. Pathol. Pharmacol. ll: 685-694 (1975). Schenkman, J.B.; Wilson, B.J.j Cinti, D.L.: Diethylaminoethyl 2,2-diphenylvalerate HCl (SKF 525-A)-in-vivo and in-vitro effects of metabolism by rat liver microsomes-formation of an oxygenated complex. Biochem. Pharmac. bl: 2373-2383 (1972). Selenke, W.M.; Leung, G.W.; Townley, R.G.: Nonantibiotic effects of macrolide antibiotics of the oleandomycin-erythromycin group with special reference to their "steroid-sparing" effects. J. Allerg. clin. Immunol. 65: 454-464 (1980). 118 Spector, S.L.; Katz, F.R.; Farr, R.S.: Troleandomycin: effectiveness in steroid dependent asthma and bronchitis. J. Allerg. clin. Immunol. 54: 367-379 (1974). Sunshine, I.: Handbook of Analytical Toxicology (The Chemical Rubber Co., Cleveland 1969). Thomas, P.E.; Reik, L.M.; Ryan, D.E.; Levin, W.: Regulation of three forms of cytochrome P-450 and epoxide hydrolase in rat liver microsomes. J. bioI. Chem. 256: 1044-1052 (1981). Ullrich, V.; Nastainczyk, W.; Ruf, H.H.: Ligand reactions of cytochrome P-450. Biochem. Soc. Trans. d: 803-807 (1975). Warner, M.; Neims, A.H.: Multiple for.ms of ethoxyresorufin o-de-ethylase and benzphetamine N-demethylase in solubilized and partially resolved rat liver cytochrome P-450. Drug Metab. Disposit. 2: 188-193 (1979). Weinberger, M.; Budgel, D.; Spector, S.; Chidsey, C.: Inhibition of theophylline clearance by troleandomycin. J. Allergy clin. Immunol. 59: 228-231 (1977). Weinstein, L.: Antibotics V. Miscellaneous antimicrobial agents; in Goodman, Gilman, The Pharmacological Basis of Therapeutics, third ed., pp. 1260-1307 (Macmillan, New York 1965). White, I.N.H.: Metabolic activation of acetylenic substituents to derivatives in the rat causing the loss of hepatic cytochrome P-450 and haem. Biochem. J. 174: 853-861 (1978).