Document ypYGnDbrrZQNLmK7VpmqBdmRX

V The Role of the Liver in Drug Metabolism H. REMMER, M.D. Tubingen, Germany From the Institute of Toxicology, University of Tubingen, TQbingen, Germany. Requests for reprints should be addressed to Or. H. Remmer, Institute of Toxicology, University of Tubingen, Tubingen. Germany. Enzymes located in the endoplasmic reticulum of liver cells pro tect the organism against an accumulation of lipid-soluble exogenous and endogenous compounds by converting them to water-soluble metabolites which can be easily excreted by the kidney. But only few drugs possess suitable groups which are conjugated with glucuronic or sulfuric acid. Most compounds have to be hydroxylated first. For this purpose the endoplasmic reticulum has at its disposal an enzymic system, completely unspecific, which activates molecular oxygen for the oxidation of lipid-soluble compounds. This takes place at a cytochrome, P1M* which is available in the endoplasmic membranes abun dantly. The oxidation rate, however, is extremly slow and de pendent on the chemical configuration of the compound and on genetically determined differences of the protein moiety of the enzyme. - Since more specific enzymes located in liver cells metabolize most of the endogenous compounds, such as steroids, at a much higher rate, the slow hydroxylation by the unspecific en doplasmic enzyme does not play an important role in their con version to inactive compounds. Drugs, however, are mainly converted to less active or in active compounds. Their effectiveness and their duration of action depend on the rate of metabolism in the endoplasmic reticulum. An overload of the liver cell with numerous lipid- soluble drugs increases drug metabolizing enzymes in-Hhe endoplasmic reticulum and augments the smooth membranes in the hepatocytes with the result that all lipid-soluble compounds reacting with cytochrome-P450 are oxidized more rapidly. Because of the lack of specificity of this enzyme, drugs com pete for the binding sites if high concentrations of several drugs are present in the liver cells. A slower metabolism of these drugs with less affinity is the result. Metabolism of drugs by this enzyme system leads sometimes to more active and toxic compounds which produce liver in jury, e.g., in the case of carbon tetrachloride. Drug metabolism is inhibited only in severe hepatitis, and exceptionally in liver cirrhosis. 24348001 Volume 49, November 1970 permission oMhe'^il'pi'oQn?^ ^Produced tniS mate ^ j|n 617 BFG36617 ROLE OF UVER IN DRUG METABOLISM--REMMER eicrtlton in urine Figure 1. Elimination of drugs. TABLE I Metabolism of Drugs by Enzymes Located in the Endoplasmic reticulum Metabolism Enzyme Oxidations of aliphatic and aromatic groups: barbiturates, diazepoxides, phenothiazines, meprobamate, phenytoine, etc. antihistamics, phenacetin, aminopyrine, antipyrine and congeners synthetic steroids, contraceptives, anabolic androgens (digitoxin) Cytochrome P4S0 Reductions of azo- and nitrogroups: azo-dyes, (chloramphenicol), etc. Flavin enzymes Hydrolyses of esters and Esterases acidamides: procaine, lidocaine, meperidine, atropine Conjugations of glucuronic acid with following groups: (a) alcoholic and phenolic (b) (carboxyl) (c) (amine) Transferases NOTE: Parentheses indicate that the metabolic pathway mentioned plays a minor role in conversion. To protect the organism against intoxications is one cf the main functions of the liver. Significant amounts of foreign compounds altering, if not to say, polluting man's normal environment invade our body as smoke particles and toxic gases, as dyes, preservatives and disinfectants, and last but not least as drugs, strictly defined. The character particularly of sedatives and stimulants has changed completely; formerly used as drugs for therapeutic purposes, now they have to be re garded as mass-products consumed in a similar manner as food. All these and many other compounds foreign to the organism should be regarded as drugs which, broadly defined, are chemical agents interfering with biochemical reactions in living cells. They are excreted by the kidney in unchanged form only very slowly, if they are lipid soluble. If taken daily they accumulate in the organism and would achieve a dangerous level if the liver had no capacity to convert lipid-soluble drugs into more water-soluble compounds (Figure 1). Described here is the role played by the liver in "drug metabolism, the properties of the enzyme sys tems involved, and the dependence of the dura tion and extent of drug actions on the function of an unspecific hydroxylase located in the endoplas mic reticulum. With few exceptions, the wellknown drugs are lipid-soluble and are able to pene trate cell membranes. This is why the drugs, after glomerular filtration of the portion not bound to plasma proteins, return to the blood from the ultrafiltrate, passing into the tubule cells by dif fusion. The driving force is the concentration gradient produced during reabsorption of water and solutes. The quantity of a substance which is not excreted depends mainly on its ratio of lipidto-water solubility. In general, the more lipidsoluble a drug is, the greater the proportion re tained in the organism. DRUG-METABOLIZING ENZYMES IN THE ENDOPLASMIC RETICULUM Fortunately, the endoplasmic reticulum in liver cells prevents accumulation of drugs. Enzymes are attached to the lipid layers of the membranes catalyzing the metabolism of drugs to more water soluble compounds (Table I) [1,2]. It is possible that metabolites formed by the enzymes in the in terior of the endoplasmic reticulum do not gain direct access to the cytosol and are discharged either into the bile canalifculi or into the extra cts The American Journal of Medicine BFG36618 note OF UVER IN ORUG METABOLISM--REMMER conjugation: inactive Figure 2. Metabolism of phenytoin (Dilantin) by two con secutive steps shown in the frames. oxidative desmethytacion: less active conjugation at either oneor two hydroxyl groups: inactive Figure 3. Metabolism of morphine. N-oxidation: noctivt? N 'A.HjC' CH, 0 oxidative desmethytation oc&v* N-oautje- reduction, active Figure 4. Metabolism of phenothiazines (Perazin): differ ent types of oxidations shown in the frames. cellular fluid where the endoplasmic membranes are in contact with the cell membrane. But this is still a moot question. The most important step is an oxidation of nearly all lipid-soluble compounds, whether formed in the body or consumed as drugs, to more polar and water-soluble substances. This may be fol lowed by a conjugation which converts the com pound to a highly polar acid that is inactive and readily excreted in the bile or urine. In this article several enzymatically catalyzed drug conversions which occur in the cytosol of the liver cells will not be considered. These reactions are of minor im portance, only two playing a significant role in drug metabolism and being deserving of mention, i.e., the acetylation of drugs; such as sulfonamides and isoniazid, and the oxidation of alcohol by alcohol dehydrogenase [3]. MAJOR TYPES OF DRUG METABOLITES FORMED Only a few examples will be presented to illustrate the type of conversion of lipid-soluble molecules which occur in the endoplasmic reticulum. (1)Phenytoin is an example of many drugs which are hydroxylated at only one site of the molecule (Figure 2). This step converts the drug into a more water-soluble and inactive compound. When the hydroxyl group is attached, the sub stance can then easily be conjugated either with glucuronic or sulfuric acid [4], The hydroxylation is the rate-limiting step in the over-all reac tion and determines the duration of phenytoin ac tion. Similar reactions are involved in the metab olism of barbiturates and other hypnotic agents (5]. (2) Many drugs possessing a hydroxyl group can be conjugated directly to inactive, highly watersoluble metabolites, e.g., morphine (Figure 3). The oxidative demethylation that occurs can be regarded as a metabolic side-reaction. This con verts morphine to the less active normorphine [3]. (3) The oxidative enzyme system in the endo plasmic reticulum membranes is peculiar insofar as it can convert a drug to various metabolites oxi dized at different sites of the compound. Two or even three oxidation steps may occur in the same molecule. As a typical example, phenothiazine metabolism may be cited [61 (Figure 4). One of the main conversion steps involves hydroxybtion of the aliphatic methyl group attached to the nitrogen. This group can be readily split off as formaldehyde. The desmethyl-compound remains active. This and the parent drug undergo further oxidations to hydroxylated aromatic metabolites. Hydroxylation can occur as indicated in Figure 4 or at another site of the aromatic ring. However, not all sites are reactive. Other important metab olites are the sulfoxides and the N-oxides. The sulfoxides and the metabolites hydroxylated Volume 49, November 1970 BFG36619 619 24948003 ROLE OF LIVER IN DRUG METABOLISM--REMMER on the aromatic ring are inactive, but it is not yet known whether the N-oxide is an active or inactive compound [7]. It will be appreciated from Figure 4 that numerous metabolites may originate if one or two different oxidations involving the phenothiazine molecule or its desmethyl compound occur. All these conversion products can be found in the urine of man and all are produced by the same, enzyme system in the endoplasmic reticulum of the liver cells. The spectrum of the metabolites in the urine varies from species to species. Even in dividual differences are known, as has been shown in the urine of patients treated with phenothiazines. The composition of the metabolites in human urine also depends upon the phenothiazine compound prescribed. As an example, perazine has been selected [8] (Figure 4). since it is widely used as an antipsychotic drug in Europe. PROPERTIES OF THE MICROSOMAL HYDROXYLASE (CYTOCHROME,,,.) Discovery of the Drug-Hydroxylase. Brodie, Axel rod and co-workers [1.2] discovered liver microsomes able to hydroxylate numerous drugs. The nature of this enzyme system remained obscure until in 1963 Estabrook. Cooper and Rosenthal et al. [9.10] detected in the microsomes of the adrenal cortex cells a peculiar cytochrome, called cytochrome P,:.,,. which is able to hydroxylate progesteron at position C-1. Subsequent studies re vealed that the same type of cytochrome found in liver cells several years previously [11] is also responsible for drug hydroxylations [12.13]. It is rot present in all tissues. The highest content has been found in the liver, much smaller amounts m kidney > lung > intestinal mucosa > skin [14], The kidney contains, per gram tissue, only one seventh of the amount found in liver. Taking into consideration the 6 times greater weight of the liver, the kidney contributes only 2.5 per cent to the total hydroxylation of a drug in the body. The other tissues mentioned contribute even less. Comparison Between Oxidations in the Mitochon dria and in the Endoplasmic Reticulum. This en zyme system has in common with the mitochon drial enzyme system the capacity to transport elec trons to a final cytochroma! acceptor. Both con sume oxygen, but differ significantly in the type of oxygen activation [10,15] (Table II). The mito chondrial cytochrome oxidase needs four electrons for reduction of one molecule of oxygen. The micro somal hydroxylase located in the endoplasmic re ticulum of liver cells belongs to the group of mixed-function oxidases which use two electrons, one for the reduction of one atom 6f oxygen to water; the second oxygen atom is incorporated into the substrate. Electron transport in the mitochondria is coupled with the formation of high energy bonds (ATP) which furnish much of the driving force for biochemical reactions. There is no indication that a similar system operates in the endoplasmic membranes, implying that the energy produced by the oxidation process is wasted. The normal oxy gen consumption of an isolated and perfused rat liver increases, if a drug is metabolized, up to 40 per cent. This is the amount that is used for maximal drug hydroxylation [16,17]. Electron-Transfer Chains in the Endoplasmic Reticulum. Reduced NADP functions exclusively as an electron donor. The electrons are picked up by an FAD-containing flavoprotein known as NADPH-dependent cytochrome-c-reductase [18]. The electron flows from the reduced flavin enzyme TABLE II Electron Transfer to 02 in Mitochondria by Cytochrome Oxidase and in Endo-' plasmic Reticulum by Cytochrome P*so Cytochrome oxidase 4 e_ + 4 H+ + 02------------------------------------------ 2 l-^O Mixed function oxidase (Cytochrome Pso) X t- 2 e-- + 2 H+ + 02-------------------------- X : O + HjO p450 Figure 5. Scheme of electron transport from NADPH to cytochrome Po 620 The American Journal of Medicine ROLE OF LIVER IN DRUG METABOLISM--REMMER to the cytochrome P,M (Figure 5). The reduced heme of the cytochrome is now able to bind one molecule of oxygen, which is activated presum ably by picking up one electron from the hemeiron. How the second electron which is necessary for mixed function oxidation comes into play anfl how one atom of the oxygen reacts with the avail able substrate is nofyet known [15,19], However, the stoichiometry for this reaction has been elucidated and can be expressed in the following manner, where X stands for any drug as sub strate [9,10] H -f- NADPH -J- H- 4- (V-- X - OH NADP~ -p H:0 A second electron-transfer system is present in the endoplasmic reticulum. The cytochromal elec tron acceptor is cytochrome bs [15,20] (Figure 6). However, the significance of this system is still ob scure. Determination of Cytochrome P,i,,. It has been possible to measure cytochrome Ptr.,, in microsomes of liver and adrenal cortical cells by ex ploiting its peculiar characteristics [9,10]. In the HfiSPsC. UPtOS I/ /////yx/nJporyrm fcfME daze soxidzs various / (ipkj sotuDle compounds phenYothia/zi ' / txxtJiturales pyndne CN nicotinamide orttine various basic con-pounds /^PEcnw-CrivsGe TYPE I TYPE I pindng to a hydrophobe rsg i on on proteins two soret bands 390 and 418 tm binding to tne 6*h ligand shift at sorst bond to longer wave length <.18 --> 426 nm Figure 6. Spectrums of cytochrome with a peak at 450 nm and cytochrcme bt with a maximum at 426 nm in suspensions of microsomes (2 mg protein/ml) isolated from control rats (------ ). benzpyrene-treated rats (3X, 20 mg/kg) (----------- ). phenobarbital-treated rats (3X, 80 mg/kg) 5 rats with a tody weight of 60 to 70 gm were starved lor forty-eight hours previous to de capitation. For method see (22, 24]. reduced form it reacts with carbon monoxide which has about the same affinity as oxygen, using the sixth ligand as a binding site. By competing with oxygen, carbon monoxide inhibits drug hydroxylations. The spectrum of the carbon mon oxide-cytochrome has a striking absorption peak at 450 nm which can easily be measured by dif ferential spectrophotometry in microsomal par ticles (Figure 6). This permits determination of the amount of cytochrome PKJt in the liver cells quantitatively. Besides cytochrome B.-, can be measured by utilizing the difference in absorption of the reduced and oxidized forms (Figure 6>. Unspecific Reaction of Cytochrome P,r.,, with Drugs. The oxidized form of cytochrome P4r,,,, the spec trum of which can be identified only with difficult procedures [21,22] reacts unspecifically with nearly all organic lipid-soluble compounds [23,24]. This reaction precedes the hydroxylation process (Figure 5). So far, two types of reversible binding can be distinguished spectrophotometrically (Fig ure 7) when lipid-soluble substrates are added to microsomal preparations. The difference in the ab sorption of microsomal suspensions without and then with added substrates, called binding spec trums, can be easily determined. This measure ment is widely used as a tool for studying drug metabolism. One enzyme or many? The question whether there is only one enzyme in the endoplasmic membranes or there are many using heme as a cofactor cannot be answered definitely, but the known facts speak for one or two [25,26]. It is also difficult to con ceive that a huge number of different protein moieties are formed, each of which reacts spe cifically with a particular type, of chemical con figuration. The idea of many enzymes is not com patible with the fact that drugs which have a high affinity to cytochrome Pji*, such as SKF-525 A, inhibit the metabolism of. nearly all compounds reacting with cytochrome [27]. Moreover, the available kinetic data describing the reactions of the hydroxylase with drugs offer no evidence for the presence of many enzymes. Rate of Hydroxylation. The hydroxylation rate is extremely low compared with well known enzymic reactions occurring in cells, and differs consider ably from substrate to substrate [24]. The ratelimiting step of the hydroxylation is the reduction of the heme iron, which increases if a substrate is bound to the hydrophobic region (type I) and de creases if a basic compound occupies the sixth ligand [28]. This indicates that the drug-cyto- 24348005 Volume 49, November 1970 BFG36621 621 ROLE OF LIVER IN DRUG METABOLISM--REMMER A Absorbance Figure 7. P,, [22]. Wovelength [m*i) Scheme of drug reactions with cytochrome chrome-Pi>.-complex with a soret band at 390 nm is reduced more easily than the unchanged cyto chrome. However, the binding of a drug to the sixth lizand inhibits electron transfer.- In intermediary metabolism the turnover num ber (number of substrate molecules converted/ mole of enzyme/minute) ranges from 1,000 to several millions in the case of catalase whereas the number of molecules converted by cytochrome P. of the rat varies between 0.2 and 15. A similar range seems also to be characteristic for the cytochrome present in the human liver [29,30]. Despite this unusually low turnover number most of the drugs seem to be hydroxylated at a reason able speed because of the surprisingly large amount of the hydroxylating system available. The liver of rats and rabbits contains about 30 to 40 nyM cytochrome Pir.n/gm [31]. If we assume that the molecular weight of this enzyme (which is not yet known) is in the same range as that of hemoglobin and albumin and comes to about 60,000,1 gm of liver would have at its disposal 1.8 to 2.4 mg cytochrome P0. More than 1 per cent of the total liver protein should be accounted as cytochrome P,r.. This is a very high content com pared with all the mitochondrial cytochromes, which are present in liver cells in smaller numbers. The high concentrations of the enzymes involved in drug hydroxylations signify the important role that cytochrome P,:.0 plays in the liver. The human liver contains less cytochrome Po [29.30]. From the figures available it can be estimated that about 10 to 20 nyM per 1 gm of liver are present. This is from one third to one half of the amount found in 1 gm of rat liver. If one also takes into consideration the much smaller relative liver weight in man, only 2 per cent of the body weight whereas rats have liver weigh ing 4 per cent of the body weight, the rat has at its disposal four to six times more cytochrome P4;.,,. This corresponds to the observations of many investigators indicating that man metabolizes drugs in vivo at a rate two to ten times slower than the rat [32]. Development of the Hydroxylation Enzyme During Ontogenesis. The enzyme in the endoplasmic reticulum is not fully developed before birth or several days and weeks afterward [33,34], This results in a low rate of drug metabolism which has caused serious and even lethal toxic reac tions in premature babies. The conjugation of drugs in particular is strikingly diminished. This was discovered after chloramphenicol was intro duced as an antibiotic in the treatment of pre mature babies. The increase in the mortality rate Figure 8. Variation in the plasma steady-state level of phenytoin in five patients who received 0.3 gm daily. 24948006 622 The American Journal of Medicine ROLE OF LIVER IN DRUG METABOLISM--REMMER observed could be related to unusually high plasma levels of chloramphenicol. The same glucuronidation system located in the endoplasmic reticulum conjugates bilirubin, and the insufficiency of the system is also responsible for the appearance of neonatal jaundice. Dependence of Activity on Nutrition. The ef ficacy of drug metabolism in the liver is lowered if animals receive a calorically sufficient but pro tein-free diet [35]. This is also true during com plete starvation, in which the liver weight de creases considerably [36]. A limited supply of the reducing agent NADPH in vivo significantly enhances the decrease in the hydroxytation rate [16]. This may provide one explanation for the many observations on incompatibilities of drugs in undernourished patients. Because of the dimin ished metabolism the level of numerous drugs may rise and produce toxic side effects. VARIATIONS IN DRUG HYDROXYLATIONS Differences Between Species. In addition to quan titative differences in drug-metabolizing enzymes in different species, qualitative species variation in drug metabolism makes any prediction of the fate of a compound in man difficult even though it may be known how the same drug is handled in several laboratory animals [3,37]. The speed and type of oxidation may differ considerably. It is therefore difficult to devise any rule which may be useful if the fate of a drug is under consideration. Species X, which converts-an administered drug A at a higher rate than species Y, may metabolize drug B at a lower speed than Y. Pentobarbital is metabolized in dogs four times more rapidly than in man, but the opposite is true for thiopental, a closely related barbiturate [1]. Also, the meta bolites formed may not be the same because species hydroxylate preferentially at different sites of the drug molecule [3,37]. Many drug conver sions in the liver are sometimes dependent partly or exclusively on enzymes "not located in the endoplasmic reticulum, a circumstance which may modify the type and speed of metabolism even more. As an example tolbutamide may be cited [38]. Individual Variations of Drug Hydroxylation. What makes studies of drug metabolism even more com plicated and rational therapy so difficult is the fact that variations in the speed of drug metabo lism have also been found when the hydroxyla- fion rates of inbred strains of inice and rats are compared [32]. This is true also for different sub jects in a population consisting of different strains, e.g., in mongrel dogs [39]. The magnitude of individual variation and its significance for therapy may be explained by presenting as an example determinations of phenytoin (Dilantin*) in the plasma of epileptic subjects receiving the same dose, 4 to 5 mg/kg (0.3 gm/daily) [40]. This drug was selected for the study because it is ab sorbed completely, and only 2 to 5 per cent of the dose administered is excreted unchanged by the kidney. It is eliminated as a conjugated oxida tion product [4] (Figure 2). The rate-limiting step is the extremely slow hydroxylation in the endo plasmic reticulum, with a turnover number of less than 0.1. The long plasma half-life varies between eight and fifty hours. Consequently phenytoin belongs to that category of drugs which accumulate during therapy until a steady state is reached, reflected by a constant plasma level if the daily dose ingested equals the amount hydroxylated. The large variation in the plasma level (Figure 8) chiefly reflects differences in the hydroxylation rate and is still greater in the fifty patients receiving the same recommended stan dard dose of 0.3 gm daily studied to date. The range observed varied between 4 and 60 pg/ml. This is why phenytoin is ineffective in some pa tients if the dose is not increased, and why toxic reactions occur in others if the dosejs not lowered. Below a plasma concentration of 10 \lg/ml pheny toin exerts no significant antiepileptic action whereas toxic side effects appear if levels of about 20 jig/ml are achieved [41,42]. Differences in the amount of cytochrome Plao available in the liver and in the affinity of the - enzyme for the substrate may explain the large variation in the hydroxylation rate. A true genetic basis for the striking variation in phenytoin hy droxylation has been found in one family, of which several members have been treated with phenytoin [42]. Determinations of drug levels in the blood of identical and fraternal twins have provided the best evidence that drug metabolism is governed by genetic factors. The individual differences in the handling of drugs which were obvious in fraternal twins were not observed in identical twins [43]. ACTIVATION AND INHIBITION OF DRUG METABOLISM A great variety of lipid-soluble drugs increase their own metabolism and the metabolism of other 24948007 Volume 49, November 1970 623 ROLE OF LIVER IN DRUG METABOLISM--REMMER compounds not related pharmacologically or chemically by inducing the drug-hydroxylating enzyme system in the endoplasmic reticulum of the liver [44,45]. This nonspecific phenomenon, which has been observed in all mammalian species so far investigated, can be viewed as an adaptive process which protects the organism against an overload of foreign compounds [46], Induction of the Hydroxylating System. The first enzyme that increases during induction is S-aminolevulinic synthetase, which is involved in the syn thesis of heme [49]. This is followed by cyto chrome P,-., and the specific flavin enzyme for electron transport. All the other enzymes not in voked in drug hydroxylation increase later on to a much smaller extent. This is the case for cytochrome b:. and the relevant unspecific esterase [12]. After one injection of phenobarbital the maxi mal induction of cytochrome P4Sn is achieved after twenty-four hours. The newly formed enzyme dis appears very quickly, with a half-life of about one day. This seems to be the normal half-life of cyto chrome P,M [46.47]. If treatment with an inducing agent is not terminated the increase in cytochrome P4Jo con tinues until a maximal amount is formed after three to five days (Figure 6), and a new steady state is achieved. So long as the inducing agent is present in the liver cells it enhances the rate of synthesis of the heme as well as of the protein moiety by causing an increase in messenger-RNA formation in the nucleus [48], This occurs pre sumably by reaction with a repressor compound. The protein breakdown seems passively to follow the increased synthesis rate [47]. Increase in Smooth Endoplasmic Membranes. A significant enhancement of drug-metabolizing enzymes is visible with the electron microscope [12.50], The smooth (but not the rough) mem branes in the liver cells increase considerably. With the growth of the endoplasmic reticulum the hver b^^^s-.Jarg^^y^p^ipjicithif^r^r^hypertrophy. This does not mean that any enlargement of the eP#l^.hilci.f^^^sJgcfeSsarilya^&ciated with an increase in drug-metabolizing en zymes, -but.this point has to be elucidated further. Age-Dependence of Induction. The magnitude of induction is also highly dependent on the age of the animals. One tenth of the phenobarbital dose in weanling rats produces nearly the same induc tion as the full dose in adult animals [51]. This, of course, seems to be the reason also why it is possible to induce the glucuronidation system in babies with small doses of phenobarbital and to prevent icterus of the newborn, a method now widely used in therapy [52-54], Activation of Drug Metabolism During Therapy. Observations during therapy and clinical studies show that an increase in drug metabolism can oc cur in man also [45,54,55], However, the number of drugs known so far to act as inducers of the hydroxylating system during therapy is small: these include barbiturates, glutethimide, meprobamate, phenylbutazone, phenytoin and the antifungal agent griseofulvin [55-57], This list is far from complete. The inducing action of these drugs in man, however, is exceedingly inconsistent, and many patients treated with the same compound at the same dose levels did not respond with in creased drug metabolism [56,58,59], Species, strain and individual variations in the magnitude of induction make any prediction as to the appearance and extent of this phenomenon in human therapy very difficult. However, several rules can be stated which permit a forecast whether a drug possessing inducing properties in animal experiments will exert a similar action in man. Conditions Necessary for Induction of the Micro somal Hydroxylase in Man. (1) The majority of lipid-soluble compounds metabolized by the microsomal hydroxylase have some inducing properties. This is true even for ethanol and for those drugs such as barbital which cannot be hydroxylated in measurable quantities during the period of testing. Alkaloids, however, seem to have no inducing capacity. (2) As can be expected, the inducing action is dose-dependent. However, many drugs are active only at nearly toxic doses; a few, however, such as long-acting barbiturates can induce the hy droxylating enzyme system when therapeutic doses are taken [50]. -- (3) Most important is the maintenance of a , high.concentration jn.liver cells for a certain time. The dose of hexobarbital has to be increased several-times if induction is to be produced in smaller animals in which hexobarbital is metabo lized very rapidly (Table III). Besides, a drug with a short action, such as hexobarbital, has to be given repeatedly in order to cause a measur able increase in the enzyme. A similar rule has been found when dogs were treated with long- acting phenobarbital. The more rapidly a dog metabolized barbiturates, the higher the daily 624 The American Journal of Medicine BFG36624 8008^6*2 \ l ! ti ROLE OF LIVER IN DRUB METABOLISM--REMMER table ill Half-life of Hexobarbital in Various Species Compared with the Pretreatment Schedule of Hexobarbital for Increasing Its Own Metabolism Species Half-Life (min) Pretreatrtent - Daily Dose Days (mg/kg) Increase in Hexobarbital Metabolism (%) Mice Rabbits Dogs Men 55 81 180 360 6 3 3 1 too 50 30 10 + 40 + 35 + 60 + 70 dose which induces drug-metabolizing enzymes had to be [60]. These observations provide a clue to an under standing of so many inconsistent reports about the induction of drug metabolism in man. Only those drugs induce hydroxylating enzymes which have a long half-life and accumulate to a certain ex tent in the organism. The maintenance plasma level of phenobarbital which caused doubling of the hydroxylation rate in dogs was 15 to 30 jig/mt plasma [60]. The level producing induction in patients seems _ to be similar, but it will not be achieved in all patients receiving a daily dose of 0.1 to 0.2 gm phenobarbital. Taking into account the extremely large variation in plasma phenobarbital levels, which should be of the same order as those in patients treated with phenytoin [40-42], we should expect significant induction only in pa tients in whom barbiturates are metabolized slowly and phenobarbital reaches a plasma steady-state level of 10 to 30 /ig/ml. The large variation in phenytoin metabolism al ready mentioned may also be partially explained by an inconsistent increase in the amount of drug- metabolizing enzymes in some patients [56,59]. Inhibition of Drug Hydroxylations. Any drug which can be hydroxyiated competes with another drug for the enzyme. Depending on the affinity and on the concentration of each drug in the liver, in hibition of hydroxylation can in fact occur [27]. Even ethyl alcohol, with a very small affinity for cytochrome can lower the metabolism' of barbiturates in rats by competitive inhibition be cause of the extremely high concentration of alco hol which is reached in the liver, exceeding the level of barbiturates about 100 times if the alco hol concentration in the blood reaches 100 mg - per cent [61,62]. Any drug which is effective only in a high dosage range is a potential inhibitor of drug-metabolizing enzymes and may interfere with the metabolism of a second drug prescribed. This can lead to a prolonged and stronger effect of the second compound administered. Many re ports have been published about a decrease in drug action because of enhanced drug metabolism due to enzyme induction, but observations show ing the opposite effect, i.e., an increase in drug action due to inhibited metabolism of the drug have been scanty and inconsistent thus far. " IMPLICATIONS OF THE PRESENCE OF THE MICROSOMAL HYDROXYLASE Hydroxylation of Endogenous Compounds. The unspecific action of the microsomal hydroxylase extends to endogenous lipid-soluble compounds. All steroids are hydroxyiated at different sites of the molecule, but to a very small extent [45] be cause the liver has at its disposal more specific enzymes for the inactivation of steroid hormones. Two types of enzymes should be mentioned, those which reduce steroids to the tetrahydro com pounds. which are conjugated easily, and those which form 17-ketosteroids. However, synthetic steroids, e.g., those which possess an aikylgroup at C17 and have little or no affinity for the steroid metabolizing enzymes (such as the contraceptives) obviously are hydroxyiated by the microsomal hy droxylase to various metabolites which are then excreted as conjugates in the urine [63]. Another endogenous compound metabolized in endoplasmic reticulum is the heme formed in mitochrondria and oxidized to bilirubin by cyto- chrome-P45o (Marver. personal communication). It is not yet clear whether the well known hydroxyla tions during the formation of bile acids are also catalyzed by the same enzyme [64]. Hydroxylation may be impaired in cholestasis. Probably more en dogenous compounds will be discovered which are hydroxyiated by the unspecific enzyme system in the endoplasmic reticulum. -- Formation of Active-Compounds. Not every hy droxylation of a drug in the liver leads to ineffective or less active metabolites.'Some well known drugs are oxidized to compounds with nearly identical effects. Thus phenylbutazone is hydroxyiated to oxyphenbutazone, which is now also widely em ployed as an anti-inflammatory agent. This is also true for phenacetin, the ethyl-group of which is oxidatively split off giving rise to N-acetyl-p- aminophenol (acetaminophen), and for aminopy- rine which is demethylated to aminoantipyrine [3]. Another example is codeine, which is demethylated to the more active morphine, signifying that its Volume 49, November 1970 625 BFG36625 ROLE OF LIVER IN DRUG METABOLISM--REWMR action depends to a great extent on the formation of morphine [3]. However, all the active metabolites mentioned are oxidized by a second or third step to inactive compound, or they are conjugated directly, as is the case for morphine, acetaminophen and many other drugs. Sometimes the drug has to be oxi dized to become active, but this is exceptional. Only two examples may be cited: Cyclophospha mide itself is inactive, oxidation in the endoplasmic reticulum converts it to an active alkylating drug useful in cancer therapy [65]. The well known pesticide parathion owes its activity to conversion to paraoxon by the hydroxylating enzyme [3]. Toxic metabolites may arise during the conver sion of drugs. This is the case if small amounts of hydroxylamine are formed as side-products when aromatic amino-compounds such as sulfon amides are metabolized [66]. Liver Damage Caused by Carbon Tetrachloride. Nothing is perfect in nature. Thus, enzymic re actions which prevent accumulation of drugs may thereby have deleterious consequences. An ex ample is the metabolism of halogenated hydro carbons. especially carbon tetrachloride. Para doxically. newborn rats are much more resistant to the toxic effects of carbon tetrachloride than adults, due to the low hydroxylation rate in"newborn animals. In accord with this view is the ob servation that phenobarbital accelerates the metabolism of CCI4 and increases its toxicity [67,- 63.70]. Electron microscopic examinations of liver cells after exposure to carbon tetrachloride have shown that the lesion starts in the endoplasmic reticulum. However, even several hours after CCI4 intoxication, when the membranes begin to dis integrate and lose their ribosomes, no signifi cant morphologic changes may be visible in the mitochondria [67]. The loss of enzyme activity in the endopiasmic reticulum as well as the persist ing normal function of the mitochondria are in accord with the morphologic observations [69].' With the decrease in cytochrome the hy droxylation of crugs decreases [71]. Ribosomal protein synthesis declines similarly [72]. This is also true for glucose-6-phosphatase, an enzyme located in the endoplasmic reticulum. On the other hand the flavin enzyme called cytochrome-c- reductase. which transports electrons from NADPH to cytochrome P<r-.. cytochrome b.-, and the un specific esterase which hydrolyses drugs, is not impaired three hours after treatment with lethal doses of CCI, [73]. This specific impairment of microsomal enzymes confirms the hypothesis that a free radical is formed from CCI4 which reacts predominantly with the phospholipid moiety of cytochrome P4.-,0 and leaves the protein portion of the enzyme intact. Only enzymes which can be easily split off from the membranes of the endo plasmic reticulum and solubilized by detergents without any loss of activity resist the action of CCI4. However, enzymes such as cytochrome P430 hydroxylase and glucose-6-phosphatase lose their activity strongly bound to the phospholipids. Similar effects are caused by lipid peroxida tion which also leads to breakdown of the endo plasmic membranes and damages only those en zymes whose activity depends on the integrity of the phospholipids in the membranes [69,74], Lipid peroxidation and CCI4 action seem to be re lated [67], Free radicals react with phospholipids during the peroxidation process. Thus, it is likely that free radicals are also produced from CCI4 by an unknown mechanism in which the hydroxylating system in the membranes is involved. This con cept is supported by the fact that, thermody namically, CCI4 can be converted to a free radical more easily than any other organic solvent [75]; It is likely that related free radicals may be formed occasionally from chloroform and halothane, although this is much more difficult to es tablish thermodynamically. Although the metabolic steps in the conversion of CCI4 are not yet eluci dated, drug-metabolizing enzymes appear to be in volved in the formation of the reactive CCI4 me tabolite. A similar mechanism may be the reason for liver injury produced by a variety of drugs, e.g., phenothiazines, tetracyclines, sulfonamides and many other compounds. If one considers the enor mous variation in drug metabolism it is possible that small amounts of free radicals are formed during the metabolism of these compounds in the endoplasmic reticulum membranes, reacting some what like CCI4. It is even possible that drug me tabolites act as haptens with proteins, causing a hypersensitivity reaction locally in the liver. Drug Metabolism in Liver Diseases. The activity of the hydroxylase probably is decreased in acute hepatitis if the cell injury is comparable with the cell damage after administration of small doses of carbon tetrachloride. Because all sedatives, hyp notics, tranquilizers and similar drugs acting on the central nervous system belong to the category of drugs which are converted to inactive metabolites in the liver, their action would be intensified -- 626 The American Journal of Medicine x ^ ^ CE C2 Q 21948010 BFG36626 ROLE OF LIVER IN DRUG METABOLISM--REMMER and prolonged. This concept would agree with the clinical impression that patients with liver disease tolerate drugs less well than normal persons. The few reports published, however, do not fully demonstrate that drug metabolism, hydroxylations as well as conjugations, is generally impaired by hepatocellular diseases. Even severe hepatitis with high transaminase levels in the plasma does not affect drug metabolism consistently. On the other hand, several patients with mild forms of hepa titis metabolized drugs more slowly than one should expect [76-78]. Surprisingly, drug hydroxylations decrease only occasionally in patients with liver cirrhosis [79,80]. This behavior is very hard to understand in view of the decrease in the number of cells functioning in a cirrhotic liver. Taking into consideration the dependence of the hydroxylation rate on the amount of enzyme available in the liver, one would expect a fair correlation between the extent of the cirrhotic process and the effect on drug metabo lism. But this is not the case. The huge variation in drug metabolism even in normal livers makes any comparison of the conversion rate in normal subjects and in patients with liver diseases ex tremely difficult [78], The nearly normal rate of drug metabolism in most cirrhotic livers might be explained by an increase in drug-metabolizing en zymes in the parts of the liver still functioning. This might occur also in some types of hepatitis. It is known that the smooth endoplasmic reticulum in creases during inflammatory processes in the liver. So long as the membranes in one part of the liver remain hyperactive we would expect that their en hanced function could compensate for the de creased metabolism in other parts of the liver in which the injurious process has progressed and converted the hypertrophic endoplasmic reticulum into a hypoactive form. This hypothesis is based on the course of hepatic damage produced by several drugs administered chronically [81]. An hyperac tive hypertrophic endoplasmic reticulum in hepatocytes induced by the action of the drug is pro gressively converted into an hypoactive form which, however, still appears in the electron mi croscope as smooth membrane hypertrophy. CONCLUSION The liver is the organ which prevents drug and other intoxications. This report attempts to explain how detoxication is accomplished in the endoplas mic reticulum of the liver cells. Calling these proc esses "detoxication," incidentally, is typical teleologic thinking. How questionable such a concept is when it is realized that the same enzymic reac tion which "detoxifies" one drug can convert an other to a more toxic compound which produceslesions in the liver or in other tissues! The scien tist who considers the reality of nature can only state that lipid-soluble compounds, endogenous as well as exogenous, are converted to more polar metabolites by the hydroxylating enzyme system, and can then be excreted more easily via the urine or bile. REFERENCES 1. Brodie BB. Gillette JR. La Du BN: Enzymatic me tabolism of drugs and other foreign com pounds. Ann Rev Biochem 27: 427. 1958. . 2. Axelrod J: Biochemical factors in the activation and inactivation of drugs. Arch Exp Path Pharmak 238:24.1960. 3. 'Williams RT: Detoxication Mechanisms. London, Chapman & Hail, 1959. 4. Dai WA. Kazenko A, Wolf LM, Glazko AJ: Studies on 5.5'diphenylhydantoin (Dilantin) in animals and man. J Pharmacc! Exp Ther 118: 270, 1956. 5. Gillette JR: Metabolism of drugs and other foreign compounds by enzymatic mechanism. Progr Drug Res 6:11.1963. 6. Beckett AH. Beaven MA, Robinson AE: Metabolism of chlorpromazine in human. Biochem Phar macol 12: 779, 1963. 7. Bicke* MH: The pharmacology and biochemistry of N-oxides. Pharmacol Rev 21: 325, 1969. 3. Breyer U: Urinary metabolites of 10-[3'-(4-methyl- piperazinyl)-propyl]-phenothiazine (Perazine) in psychiatric patients. I. Isolation, identification and determination of metabolites. Biochem Pharmacol 18: 777,1969. 9. Estabrook RW, Coojjer DY, Rosenthal O: The light reversible monoxide inhibition of the steroid C3-hydroxylase system of the adrenal cortex. Biochem Z 338: 741,1963. 10. Omura T, Sato R, Cooper DY, Rosenthal 0, Esta brook RW: Function of cytochrome P-450 of mierosomes. Fed Proc.`24: 1181, 1965. 11. Klingenberg M: Pigments of rat liver mierosomes. Arch Biochem 75: 376, 1958. 12. Remmer H, Merker HJ: Effect of drugs on the for mation of smooth endoplasmic reticulum and drug-metabolizing enzymes. Ann NY Acad Sci 123: 79.1965. 13. Emster L, Orrenius S: Substrate induced synthesis of the hydroxylating enzyme system of liver mi- crosomes. Fed Proc 24: 1190. 1965. Voiume 49, November 1970 627 BFG36627 ROL Or UV'tR IN ORUG METABOLISM--REMMER 14. Uehteke H: Extrahepatic microsomal drug metabolism. Proc Europ Soc Study Drug Toxic 10: 94. 1963. 13. Ssabrock RW. Hildebrandt A, Remmer H, Schenk- man J3. Rosenthal 0, Cooper DY: The role of Otochrome P-450 in microsomal mixed func tion oxidation reactions, Biochemie des Sauer- stiffs chap 5 (Hess B. Staudinger HJ, eds). Berlin. Heidelberg. Springer-Verlag, 1968, p 142. 15. Thurman RG. Scholz R: Mixed function oxidation in perfused rat liver. Europ J Biochem 10: 459. 135917. 5,'es H, 3rauser B: Interaction of mixed function oxi- dase with its substrates and the associated redox transitions of cytochrome P-450 and py ridine nucleotides in perfused rat liver. Europ J Biochem 15: 531. 1970. 13. Kamin H. Siler Masters BS, Gibson QuH. Williams ChH Jr. Microsomal TPNH-cytoehrome-c-reduciase. Fed Proc 24:1164, 1965. 19. Mason HS. North JC, Vennaste M: Microsomal mixedfunction oxidations: the metabolism of xenobiotics. Fed Proc 24: 1172, 1965. 20. Smttmatter P; Protein and coenzyme interactions in the NADH-cytochrome b-5 reductase system. Fed Proc 24:1156, 1965. 21. Kiicshita T, Horie S: Studies on P-450. III. On the absorption spectrum of P-450 in rabbit liver microsomes. J Biochem (Tokyo) 61: 26. 1967. 22. Rammer H, Estabrook RW, Schenkman J, Greim H: Reaction, of drugs with microsomal liver hy droxylase: its influence on drug action. Arch Exp Path Pharmak 259: 98, 1963. 23. Schenkman J. Estabrook RW. Sasame H. Gillette J. Narasimhulu S. Cooper DY, Rosenthal 0: Drug interaction with hepatic microsomal cyto chrome. Molec Pharmacol 2: 187. 1956. 24. Schenkman J3, Remmer H, Estabrook RW: Spectral studies of drug interaction with hepatic micro somal cytochrome. Molec Pharmacol 3: 113, 1957. 25. Schenkman J3. Greim K, Zange M. Remmer H: On the problem of possible other forms of cyto chrome P-450 in liver microsomes. 8iochim Biophys Acta 171: 23, 1969. 26. ri.debrandt A. Remmer H, Estabrook RW: Cyto chrome P-450 of liver microsomes--one pig ment cr many. Biochem Biophys Res Comraun 30: 607.1968. 27. Ar.dsrs MW. Mannering GJ: Inhibition of drug me tabolism. Molec Pharmacol 2: 319, 1966. 2S. G'gon PL, Gram TE, Gillette JR: Studies on the rate of reduction of hepatic microsomal cytochrome P-450 by reduced nicotinamide adenine dinudeitide phosphate. Molec Pharmacol 5: 109, 1569. 29. Ahrares AP, Schilling G, Levin W, Kuntzman R, Brand L, Mark L: Cytochromes P-450 and b-5 in hu man liver microsomes. Clin Pharmacol Ther 10:655. 1969. 32- Ackarmann E. Heinrich I: Die Aktivitat der N- und 0-0emethy!ase in der Leber des Menschen. Biochem Pharmacol 19: 327, 1970. 31. Greim H: Synthesesteigerung und Abbauhemmung bei der Vermehrung der mikrosomalen Cyto chrome P-450 und b-5 durch Phenobarbital. Arch Pharmak 266: 261. 1970. 32. Quinn GP, Axelrod J, Brodie BB: Species, strain and sex differences in metabolism of hexobarbi- tone, amidopyrine, antipyrine and aniline. Biochem Pharmacol 1: 152, 1958. 33. Brodie BB. Maickel RP: Comparative biochemistry of drug metabolism. Proceedings of the First In ternational Pharmacology Meeting, vol 6 (Brodie BB. Erdos EG, eds). New York, Perga- mon, 1962. p 299. 34. Dutton GF: Glucuronide conjugation. Proceedings of the First International Pharmacology Meeting, vol 6 (Brodie BB. Erdds EG, eds). New York. Pergamon, 1962. p 39. 35. Seawright AA, Me Lean AEM: The effect of diet on carbon tetrachloride metabolism. Biochem J 105: 1055.1967. 36. Kato R, Gillette JR: Effect of starvation on NADPH- dependent enzymes in liver microsomes of male and female rats. J Pharmacol Exp Ther 150: 279.1965. 37. Williams RT: The fate of foreign compounds in man and animals. Pure Appl Chem 18: 129, 1969. 3S. Remmer H, Merger HJ: Vermehrung arzneimittelabbauender Enzyme durch Tolbutamid. Arch Exp Path Pharmak 249: 71. 1964. 39. Siegert M, Alsleben B, Liebenschutz W. Remmer H: Unterschiede in der mikrosomalen Oxydation und Acetylierung von Arzneimitteln bei verschiedenen Arten und Rassen. Arch Exp Path Pharmak 247:509,1964. 40. Remmer H. Hirschman J, Greiner I: Die Bedeutung von (Cumulation und Elimination fur die Dosierung von Phenytoin (Diphenylhydantoin). Deutsch Med Wschr 94: 1265, 1969. 41. Buchthat F, Svensmark O. Schiller PJ: Clinical and electroencephalographic correlations with se rum levels of diphenylhydantoin. Arch Neurol (Chicago) 2:624, 1960. 42. Kutt H. Winter W. Kokenge R, Me Dowell FM: Di phenylhydantoin metabolism, blood levels and toxicity. Arch Neurol (Chicago) 11: 642, 1964. 43. Alexanderson B, Evans DAP, Sj&qvist F: Steady-state plasma levels of nortriptyline in twins: influ ence of genetic factors and drug therapy. Brit Med J 4: 764.1969. 44. Remmer H: Drug tolerance, Ciba Foundation Sym posia on Enzymes and Drug Action, London, Churchill. 1962, p 276. 45. Conney AH: Pharmacological implications of micro somal enzyme induction. Pharmacol Rev 19: 317,1967. 46. Remmer H, Merker HJ: Orug-induced changes in the liver endoplasmic reticulum: association with drug-metabolizing enzymes. Science 142: 1657. 1963. 47. Greim H, Schenkman JB, Klotzbucher M, Remmer H: The influence of phenobarbital on the turn over of hepatic microsomal cytochrome b-5 and cytochrome P-450 hemes in the rat. Biochim Biophys Acta 201: 20, 1970. E2S The American Journal of Medicine ZTOSbSfrZ BFG36628 ROLEJJF LIVER IN DRUG METABOLISM--REMMER 43, Kunz W, Niessing J, Schnieders B. Sekeris CE: Char* acterization of rapidly labelled rat liver RNA showing high affinity to MAK columns. Biochem J (in press). 49. Marver HS: The role of heme in the synthesis and repression of microsomal protein. Microsomes and Drug Oxidations (Gillette J. Conney AH, Cosmides GJ, Estabrook RW, FoutsJR. Mannering GJ, eds) New York, Academic Press, 1969, p 495. 50. Remmer H, Merker MJ: Enzyminduction; und Vermehrung von endoplasmatischem Retlkulum in der Leberzelle wahrend der Behandlung mit Phenobarbital (Luminal). Klin Wschr 41: 276, 1963. 51. _ Remmer H: Induction of drug metabolizing enzymes in different animal species. Proc Europ Soc Study Drug Toxic 11:14. 1969. 52. Yaffe SJ. Levy G, Matsuzawa T, Baliah T: Enhance ment of glucuronide-conjugating capacity in a hyperbilirubinaemic infant due to apparent en zyme induction by phenobarbital. New Eng J Med 275:1461,1966. 53. Crigler JF, Gold Nl: Sodium phenobarbital induced decrease in serum bilirubin in an infant with congenital non-haemolytic jaundice and pru ritus. J Clin Invest 45:998,1966. 54. Trolle D: Decrease of total serum bilirubin concen tration in newborn infants after phenobarbitone treatment. Lancet 2: 705, 1963. 55. Chen W, Vrindten PA, Dayton PG, bums JJ: Ac celerated aminopyrine metabolism in human, subjects pretreated with phenylbutazone. Life Sci 1: 35,1962. 56. Cucinell SA, Conney AH, Sansur M, Bums JJ: Drug interactions in man. I. Lowering effect of pheno barbital on plasma levels of bishydroxyeoumarin (Dicumarol) and dipherylhydantoin (Di lantin). Clin Pharmacol Ther 6: 420. 1965. 57. Bums JJ. Cucinell SA, Koster R, Ccnney AH: Appli cation of drug metabolism to drug toxicity studies. Ann NY Acad Sci 123: 272. 1965. 58. Douglas JF, Ludwig BJ. Smith N: Studies on the metabolism of meprobamate. Proc Soc Exp Biol Med 112:436, 1963. 59. Arnold K, Gerber N: The rate of decline of diphenylhydantoin in human plasma. Clin Pharmacol Ther 11:121,1970. 6a Remmer H, Siegert M: Beschteunigter Arzneimit- telabbau- durch Enzyminduktion beim Hunde nach Behandlung mit Phenobarbital. Arch Exp Path Pharmak 247: 522, 1964. 61. Schuppel R: Hemmung der Hydroxylierung von Arzneimitteln in vivo unter Athanolbelastung. Arch Pharmak 265: 156. 1969. 62. Ueber CS. De Carli LM: Hepatic microsomes: a new site for ethanol oxidation. J Clin Invest 47: 62a. 1969. 63. Remmer H: The fate of drugs in the organism. Ann - Rev Pharmacol 5: 405. 1965. 64. Hutterer F. Bacchin PG. Raisfeld IH, Schenkman JB. Schaffner F, Popper H: Aleration of microsomal biotransformation in the liver in cholestasis. Proc Soc Exp Biol Med (in press). 65. Brock N, Hohorst HS: Ober die Aktivierung von Cy- clophosphamid in vivo und in vitro. Arzneimittelforschung 13: 1021,1963. 66. Uehleke H: Biologische Oxydation und Reduktion am Stickstoff aromatischer Amino- und Nitroderi- ' vate und ihre Folgen fur den Organismus. Progr Drug Res 8: 195, 1964. 67. Slater TF: Necrogenic action of carbon tetrachloride in the rat: a speculative mechanism based on activation. Nature (London) 209: 36, 1966. 68. Recknagel RO: CCI, hepatotoxicity. Pharmacol Rev 19:145.1967. 69. Neubert D, Maibauer D: Vergleichende Untersuch- ungen der oxydativen Leistungen von Mito- chondrien und Mikrosomen bei experimenteller LeberschSdigung. Arch Exp Path Pharmak 235: 291,1959. 70. Me Lean AEM, Me Lean EK: The effect of diet and l.l,l-trichloro-2,2-bis-(p-chlorophenyl)eth- ane (DDT) on microsomal hydroxylating en zymes and on sensitivity of rats to carbon tetra chloride poisoning. Biochem J 100: 564, 1966. 71. Sasame HA, Castro JA, Gillette JR: Studies on the destruction of liver microsomal cytochrome P450 by carbon tetrachloride administration. Biochem Pharmacol 17: 1759. 1968. 72. Smuckter EA, Iseri OA, Benditt EP: An intracellular defect in protein synthesis induced by carbon tetrachloride. J Exp Med 116: 55. 1962. 73. Heni N, Remmer H: Der Einfluss verschiedener Tetrachlorkohlenstoff-Dosen auf die mikrosomalen Enzyme der Rattenleber. Arch Toxik. 1970 (in press). __ 74. Orrenius S, Dallner G, Ernster L: Inhibition of the TPNH-linked lipid peroxidation of liver micro somes by drugs undergoing oxidative demethyl- ation. Biochem Biophys Res Commun 14: 329, 1964. 75. Wirtschafter ZT, Cronyn MW: Free radical mechanism for solvent toxicity. Arch Environ Health (Chi cago) 9:186.1964. 76. Held H, von Oldershausen HF: Zur Pharmakokinetik von Meprobamat bei chronischen Hepatopath- ien und Arzneimittelsucht Klin Wschr 47: 78. 1969. 77. Held H, von Oldershausen HF, Remmer H: Der Ab- bau von Phenobarbital bei Leberschaden. Klin Wschr, 1970 (in press). 78. Levi AJ, Sherlock S. Walker D: Phenylbutazone and isoniazid metabolism in patients with liver dis ease in relation to previous drug therapy. Lan cet 1:1275.1968. 79. Brodie BB, Bums JJ, Weiner M: Metabolism of drugs in subjects with Laennec's cirrhosis. Med Exp (Basel) 1:290.1959. 80. Kunin CM, Glazko AJ, Finland M: Chloramphenicol and its' metabolic products in the blood of pa tients with several renal diseases or hepatic cirrhosis. J Clin Invest 38: 1509, 1959. 81. Hutterer F, Klion FM, Wengraf A, Schaffner F, Pop per H: Hepatocellular adaptation and injury. Structural and biochemical changes following dieldrin and methyl butter yellow. Lab Invest 20: 455. 1969. Volume 49, November 1970 629 24948013 BFG36629