biological disposition of morphine andits surrogates-3*

Bull. Org. mond. Santg 1962, 26, 261-284Bull. Wld Hlth Org.

The Biological Disposition of Morphineand its Surrogates-3*

E. LEONG WAY, Ph.D.' & T. K. ADLER, Ph.D.'

CONTENTSPage

SYNTHETIC SURROGATES OF MORPHINE ... ....... 261Morphinan derivatives ............... 261Methadone .................... 266Acetylmethadol ...... . . . . . . . . . . . . 272Propoxyphene . . . .. . . . . . . . . ..............274Pethidine .................... . 275Anileridine ...... . . . . . . . . . . . . . . 280Other phenylpiperidine derivatives . . . . . . . . . .281Ethoheptazine .................. . 282

REFERENCES ..................... . 283

SYNTHETIC SURROGATES OF MORPHINE

MORPHINAN DERIVATIVES

Levorphanol, dextrorphan and racemorphanThe compound 3-hydroxy-N-methylmorphinan

was first described by Schnider & Gruessner in 1948.The product obtained by synthesis is a racemate,which is known under the name of racemorphan.Racemorphan was separated into its two opticalisomers by means of the tartrates (Schnider &Gruessner, 1951). The i-isomer is available com-mercially as the tartrate dihydrate salt and has thegeneric name of levorphanol; the d-isomer has thegeneric name of dextrorphan. The analgesic and

* This study on the biological disposition of morphineand its surrogates is being published in the Bulletin of theWorld Health Organization in four instalments. The firstinstalment was devoted to morphine per se (Bull. WId HlthOrg., 1961, 25, 227), the second to derivatives of morphine(Bull. Wld HIth Org., 1962, 26, 51). This-the third-dealswith synthetic surrogates of morphine; and the final instal-ment will discuss general considerations. The four instal-ments will eventually be available as a joint reprint.

1 Department of Pharmacology, University of CaliforniaMedical Center, San Francisco, Calif., USA. The authorswere aided in their preparation of this report respectively byGrant RG-1839 from the National Institutes of Health andby Senior Research Fellowship SF 271 from the PublicHealth Service, US Department of Health, Education, andWelfare.

respiratory effects of racemorphan can be largelyaccounted for by the i-isomer, the d-isomer beingvirtually inactive in these respects (Fromherz, 1951;Benson, Stefko & Randall, 1953). Likewise, levor-phanol possesses addiction liability but dextrorphanis devoid of this property (Isbell & Fraser, 1953).However, the disposition of the three optical formswill be discussed under a single heading, since thereappear to be many similarities in the manner inwhich the compounds are handled by the body.Methods of estimation. The methyl orange pro-

cedure for estimating organic bases (Brodie & Uden-friend, 1945; Brodie, Udenfriend & Dill, 1947) wasadapted by Fisher & Long (1953) and by Shore,Axelrod, Hogben & Brodie (1955) to the estimationof levorphanol in biological media. For the extrac-tion of levorphanol Fisher & Long used chloroform,which tends to yield high tissue blanks; no appraisalof the specificity of the procedure was given. Shoreand co-workers used the less polar benzene, whichgenerally extracts lesser amounts of interferingmethyl orange reactants, and furnished satisfactoryevidence for the validity of the procedure. Severalprocedures for the forensic detection of levorphanolhave been reported (Vidic, 1953b, 1953c; Breinlich,

1102 - 261 -

E. LEONG WAY & T. K. ADLER

1953a, 1953b; Curry &Powell, 1954; Jatzkewitz, 1954;Kaiser & Jori, 1954) but the methods have not beenapplied to investigation of the fate of levorphanol.Paper chromatography has also been used to estim-ate levorphanol in urine and faeces (Brossi, Hafliger& Schnider, 1955) but it appears at best only semi-quantitative. N-14C-methyl-labelled levorphanol wasdetermined in biological fluids after its extraction inconcentrations as low as 0.01 j,g/ml with a precisionbetter than ±10% (Woods, Mellett & Andersen,1958).

Absorption. Absorption of the compounds afteradministration by the usual routes is rapid in allspecies studied. Peak effects of levorphanol andracemorphan in man were noted within one hourafter subcutaneous administration and at two hoursafter oral ingestion (Isbell & Fraser, 1953). Plasmalevels reached a peak and declined rapidly within30 minutes after intraperitoneal administration ofeither levorphanol or dextrorphan in dogs (Shore,Axelrod, Hogben & Brodie, 1955), indicating rapidabsorption of the compounds before the end of thisinterval. Similarly, peak plasma levels of levor-phanol in the monkey and dog were attained within30 minutes after subcutaneous administration(Woods, Mellett & Andersen, 1958). The rate ofabsorption of racemorphan after oral administrationis rapid, as judged by the fact that 50% of the drugdisappeared from the gastro-intestinal tract of fastedrats in less than 30 minutes (Fisher & Long, 1953).However, despite the initial rapid uptake of thecompound, appreciable amounts were still found inthe gastro-intestinal tract after four hours.

It is probable that the rate of absorption of race-

morphan is even more rapid than that observed byFisher & Long (1953) since it has been shown thatboth the 1- and the d-isomer are secreted into thegastric juice (Shore, Axelrod, Hogben & Brodie,1955). In studies in the rat, Shore and co-workersfound high concentrations of levorphanol in thestomach after parenteral administration. In more-

extended studies on anaesthetized dogs, primed withhistamine to enhance gastric secretory activity, theadministration of levorphanol by continuous intra-venous infusion yielded concentrations of the drug(12 ,ig/ml) in the gastric juice that were about40 times greater than that in plasma (0.3 ,ug/ml).Similar results were reported for dextrorphan. Thesefindings would appear to explain the persistent pre-

sence of the compound in the gastro-intestinalsystem despite its rapid absorption from the site.

Distribution. Once levorphanol is absorbed itrapidly leaves the blood and localizes in tissues.Shore, Axelrod, Hogben & Brodie (1955) estimatedfrom the rate of decay of plasma levels in the dogthat the apparent biological half-life of the compoundwas less than 1 hour, the plasma levels of levor-phanol dropping from 1 jig/ml at extrapolated zerotime to less than 0.2 ,tg/ml at 90 minutes. Similarresults were noted with the d-isomer. Woods,Mellett & Andersen (1958) found the biologicalhalfAife of levorphanol in the dog and the monkeyto be approximately 75-90 minutes. A slowerdecline in plasma levels was noted by Fisher & Long(1953) after intravenous administration, but theirvalues are at the lower limits of sensitivity of theirmethod.Organ levels of levorphanol measured at various

time intervals in the dog indicate that the compoundis metabolized at a fairly rapid rate and that extensiveaccumulation does not occur. After administrationof 10 mg/kg of levorphanol intraperitoneally, lowconcentrations of the drug were found at 45 minutes,the highest concentrations (between 3 and 6 ,tg/g)being present in the lung, spleen, kidney, liver andheart. Only traces were found in the muscle andbrain. The plasma level at this time was 0.4 ,ug/ml.At 90 minutes the plasma level had declined to0.2 ,ug/ml and only the spleen showed detectableconcentrations of the drug (Shore, Axelrod, Hogben& Brodie, 1955).The distribution of levorphanol in the various

cellular fractions of the brain, liver and kidney wascompared 30 minutes after administration of30 mg/kg of N-l"C-methyl levorphanol base intra-peritoneally. In all three organs the concentrationsof levorphanol were highest in the soluble andmicrosomal, intermediate in the nuclear and lowestin the mitochondrial fractions. In these organsbetween 69% and 82% of the levorphanol presentwas found in the soluble fraction. The in vivo uptakeof levorphanol by the first nuclear fraction of allthree organs was greater than the uptake when thedrug was added in vitro, suggesting that the in vivodifferences found among the various cellular frac-tions were not merely artefacts reflecting a differentialpartitioning of the drug during disruption of thetissues. The microsomal and mitochondrial frac-tions of the liver demonstrated a greater capacity tobind levorphanol than the corresponding fractionsof the brain, but the reverse appears to be the casewith respect to the first nuclear fraction (Mellett &Woods, 1959).

262

BIOLOGICAL DISPOSITION OF MORPHINE AND ITS SURROGATES-3

Radioautographs of the kidney and liver obtainedafter administration of labelled levorphanol (Mellett& Woods, 1959) duplicated earlier findings by Miller& Elliott (1955) with labelled morphine. The kidneyshowed decreasing amounts of radioactivity as one

proceeded from the cortex towards the collectingduct. The distribution of radioactivity in the liverwas uniform throughout the organ.

Metabolism. Levorphanol and dextrorphan appear

to be metabolized in vivo in a similar manner. Bothisomers are converted in the body primarily by con-

jugation, presumably with glucuronic acid. Depend-ing on the species, both may be N-demethylated tosome extent.

It has been established that levorphanol or

dextrorphan is excreted as a conjugate. Fisher &Long (1953) obtained, after acid hydrolysis, in-increased amounts of racemorphan, levorphanol or

dextrorphan from the urine of dogs receiving eachcompound. These findings were subsequently con-

firmed by two other groups of workers (Shore,Axelrod, Hogben & Brodie, 1955; Woods, Mellett &Andersen, 1958). Shore and co-workers demonstra-ted conclusively that the bound product in dog urinecontained the parent substance which was admi-nistered to the animal. Levorphanol and dextror-phan were isolated in the crystalline state after beingreleased from the bound form by acid hydrolysis.Separation and characterization of the two isomerswere effected by counter-current distribution and theidentity of each was established by mixed-melting-point determinations with the authentic compound.Evidence suggesting that the conjugate is a glucuro-nide was furnished by Fisher & Long (1953), whofound that incubation of the bound product inurine with /-glucuronidase resulted in an increasedyield of racemorphan. However, actual isolationof the glucuronide has not been accomplished.There is suggestive but inconclusive evidence that

levorphanol or dextrorphan may be N-demethylatedto the corresponding nor-derivative by certainspecies in vivo. In the monkey about 20% of thedose of N-'4CH3-labelled levorphanol could beaccounted for as 14CO2 in the expired air. In the ratabout 5% of the dose was eliminated as 14CO2, butin the dog the 14CO2 eliminated accounted for only1-2% (Woods, Mellett & Andersen, 1958). Attemptsto identify norlevorphanol or its conjugate in dogurine after administration of levorphanol have beenunsuccessful, although 3-OH morphinan is easilydetected when present in urine. For example, the

nor-compound can be found in dog urine afteradministration of the 3-methoxy analogue (Brossi,Hafliger & Schnider, 1955) or the N-allyl analogue(Mannering & Schanker, 1958). Furthermore, whenthe nor-compound itself was administered to dogsit was easily detected in the urine in the free andbound forms (Shore, Axelrod, Hogben & Brodie,1955). Moreover, the free and bound forms of thenor-compound are found in the urine and faeces ofrats after injection of levallorphan (Mannering &Schanker, 1958).

In vitro studies indicate that liver microsomalenzymes of rats or mice catalyse N-demethylation oflevorphanol, with the formation of formaldehydeand, presumably, the nor-derivative. Axelrod(1956a) reported that the d-isomer was demethylatedmuch less readily than the i-isomer by rat prepara-tions, but Takemori & Mannering (1958) found thatthe isomers were demethylated with equal facilityby rat preparations and mouse preparations. Noformaldehyde was detected when levorphanol wasincubated with dog liver microsomal preparations(Shore, Axelrod, Hogben & Brodie, 1955).The mechanisms concerned with N-demethylation

are considered in the first instalment of this study inconnexion with morphine and the pharmacologicalimplications of this reaction will be discussed in thefourth and final instalment.

Excretion. The excretion of levorphanol, dex-trorphan and racemorphan appears to be quitesimilar. Relatively little of any of the three com-pounds is excreted unchanged. The compounds areexcreted chiefly in the urine as a conjugate ofglucuronic acid. Practically none is excreted in thefaeces in either the free or the bound form. Thesegeneralizations are based on experiments performedmostly on the dog and to a lesser extent on themonkey and the rat. Excretion studies on man havenot been reported.

Evidence that the body excretes the morphinanisomers in a similar manner was obtained byFisher & Long (1953), who compared the urinaryexcretion of racemorphan, levorphanol, dextrorphanand a mixture of the 1- and d-isomers in dogsafter giving equivalent amounts of each drug sub-cutaneously. The percentage of each preparationaccounted for in the urine in the free and boundforms was quite similar, the total (free and bound)recovery for each substance being between 42%and 62% of the dose administered. This study issupported by the observations of Shore, Axelrod,

263


Hogben & Brodie (1955), who found that the 1- andd-isomers yielded virtually identical amounts ofthe bound form in the urine after intraperitonealadministration of 5 mg/kg, the percentage recoveredaveraging 56% for levorphanol and 59% for dex-trorphan.There is general agreement that a minor fraction

of administered levorphanol or dextrorphan isexcreted in the urine as the free alkaloid. However,the reported data on dogs indicate a range ofexcretion values from negligible (Shore, Axelrod,Hogben & Brodie, 1955) to as much as 30% of thedose (Fisher & Long, 1953). Shore and co-workers(1955) reported negligible amounts of free levor-phanol and free dextrorphan in the 48-hour urineof dogs injected with 5 mg/kg of each drug intra-peritoneally. On the other hand, Fisher & Long(1953) found that about 5% of the dose of race-morphan hydrobromide was excreted in the 24-hoururine of a dog that received a single intramuscularinjection of 10 mg/kg. Continued administrationof the compound generally elevated the excretionof the unchanged drug to about 15%, and in oneinstance with the d-isomer a value of 30% wasrecorded. In confirmation of the findings of Fisher& Long, Woods, Mellett & Andersen (1958) reportedthat they were able to account for between 3%and 8% of the dose in the 24-hour urine of dogsgiven 0.2-2 mg/kg of levorphanol subcutaneously;and a similar study in the monkey indicated that2-3 % of the levorphanol dose was excreted inthe free form. In both species most of the freelevorphanol excreted was present in the urinewithin 5 hours. Both Fisher & Long and Woods,Mellett & Andersen detected traces of the freealkaloid in urine several days after the admini-stration of the compound had been discontinued.

Levorphanol, dextrorphan or racemorphan isexcreted chiefly in the urine as a conjugate. Shoreadministered the 1- and the d-isomer to dogs (5 mg/kgsubcutaneously) and found that the average excre-tion of the conjugate of each compound was 60%of the administered dose. The conjugates of bothisomers were excreted at about the same rate, themajor part of the excretion occurring within 6 hoursafter injection. Fisher & Long (1953), using variousdosage schedules of all three optical forms, reportedvalues generally between 25% and 55% of theadministered dose. Woods, Mellett & Andersen(1958) found that about 40% of the dose of levor-phanol was excreted as the conjugate in the 24-hoururine of dogs given 0.2-2 mg/kg subcutaneously;

less than 0.1 % was found in the faeces. A compar-able experiment in the monkey indicated thatabout 35% of the dose was excreted in the urinein the bound form and less than 0.7% in thefaeces. Brossi, Hafliger & Schnider (1955) reportedtotal levorphanol and dextrorphan urinary excre-tion values for the dog which appear to befar too low. It seems probable that their condi-tions for hydrolysis of the bound isomers werenot optimal.The conjugate of levorphanol is fairly labile in

urine not collected by catheterization. Woods,Mellett & Andersen (1958) found that cage-collectedurine samples occasionally increased markedly infree levorphanol content with time and showed aloss in bound levorphanol content.

Shore, Axelrod, Hogben & Brodie (1955) foundno evidence for demethylation as a metabolic path-way for levorphanol or dextrorphan. Neither thecorresponding nor-derivatives of each isomer northeir respective conjugates were found in the urineafter administration of the compounds to dogs; afortified homogenate of dog liver failed to metabolizelevorphanol.The dog rat and monkey all excrete some 14C-

methyl-labelled levorphanol as 14CO2, but there is amarked species difference in the quantity of 14CO2eliminated. Only about 1-2% of the dose of levor-phanol is eliminated in this way by the dog 24 hoursafter 2 mg/kg subcutaneously. In the rat about 5%and in the monkey about 20% is eliminated as14CO2 after the same dose of the labelled drug.The maximum rate of elimination of 14CO2 in themonkey was observed in the 3-hour sample. Two-thirds of the total 14CO2 to be eliminated wasexcreted during the first 6 hours after admini-stration, but significant amounts of 14CO2 were stillbeing excreted at 24 hours (Woods, Mellett &Andersen, 1958).Development of tolerance does not appear to

affect the excretion of racemorphan. Fisher & Long(1953) studied the daily excretion of racemorphanin non-tolerant dogs that received a single intra-muscular injection of the drug (10 mg/kg) onthree consecutive days and were able to accountfor between 1 % and 17% of the dose in the freeform and 29-55% in the bound form. Two dogsmade tolerant by daily injections of 10 mg/kg overa 6-month period excreted in the urine between6% and 16% of the dose as the free alkaloid and27-55% as the conjugate (Shore, Axelrod, Hogben& Brodie, 1955).

264


After intravenous administration of levorphanoland dextrorphan, the drug is secreted into the gastricjuice in concentrations many times higher than thosein the plasma (Shore, Axelrod, Hogben & Brodie,1955).

Dextromethorphan

Dextromethorphan or d-3-methoxy-N-methylmor-phinan is available commercially as the hydro-bromide. It is used exclusively as an antitussiveand apparently does not produce analgesia (Benson,Stefko & Randall, 1953; Pellmont & Bachtold, 1954)or have addiction liability (Isbell & Fraser, 1953).

Little is known concerning the fate of this com-pound. Brossi, Hafliger & Schnider (1955) investi-gated the urinary excretion of dextromethorphan indogs after injecting 10 mg/kg of the hydrobromideonce or twice daily. Using paper chromatographicprocedures, they were able to establish that thecompound is N-dealkylated and 0-dealkylated.Urine samples collected for 3-4 days after injectionyielded 4-5% dextromethorphan, 1-4% 3-methoxy-morphinan, 2-6% 3-hydroxy-N-methylmorphinanand 2-3% 3-hydroxymorphinan. It was not estab-lished whether these products were present inthe free or bound forms, since estimations were

made for total alkaloid content only. The actualquantities of each substance excreted may be greaterthan indicated, since by using the same methods theauthors obtained values for total levorphanol anddextrorphanol excretion which are lower than thosefound by other workers (Fisher & Long, 1953;Shore, Axelrod, Hogben & Brodie, 1955; Woods,Mellett & Andersen, 1958).

Liver microsomes of rats or mice demethylatedextromethorphan less readily than its i-isomer(Axelrod, 1956a; Takemori & Mannering, 1958), butin both cases appreciable amounts of formaldehydeare obtained. The relative contributions of theN-methyl and 0-methyl groups to the formaldehydepool were not determined, but if these are com-

parable with codeine metabolism in vitro it may beassumed that the N-demethylation shows a markedpredominance over the 0-demethylation (Takemori& Mannering, 1958).

LevallorphanRacemic 3-hydroxy-N-allylmorphinan was syn-

thesized in 1950 by Schnider & Hellerbach (1950),who described it as having less analgesic action thanracemorphan. In 1952 the i-isomer, levallorphan,was reported to be a potent antagonist oflevorphanol

and of morphine (Benson, O'Gara & Van Winkle,1952; Fromherz & Pellmont, 1952) and, undercomparable experimental conditions, more potentin this respect than nalorphine (Costa & Bonnycastle,1955). The d-isomer of 3-hydroxy-N-allylmorphinanis not active as an antagonist of morphine or oflevorphanol in vivo (Benson, O'Gara & Van Winkle,1952; Fromherz & Peilmont, 1952) but its in vitroinhibition of demethylation of morphinan deriva-tives is equal to that of levallorphan (Takemori &Mannering, 1958).There is much information on the pharmacology

of levallorphan which has been amply reviewedrecently (Reynolds & Randall, 1957), but compara-tively little is known of the disposition of thecompound.

Methods of estimation. Brossi, Haflinger &Schnider (1955) used a semi-quantitative colori-metric method for determining the concentration oflevallorphan in dog urine and measured the totalhydrolysed alkaloids after preliminary extraction atpH 9.2 with ether and chromatographic separationusing buffered filter-paper. That the extraction con-ditions for levallorphan in urine may not be optimalis suggested by the work of Mannering & Schanker(1958), who found that levallorphan as well as itsdealkylated metabolite were extracted from variousbiological media at pH 11-12 with chloroform, andthat another metabolite of unknown structure wasextracted at pH 8-8.5 with a 10% solution of ethanolin chloroform. After extraction the alkaloids wereseparated by paper chromatography and eluted, andthe alkaloid concentrations were then quantified byreaction with methyl orange. The qualitative identi-fication of levallorphan rested on the melting-pointof the sublimed crystalline alkaloid and the mixedmelting-point with authentic material as well as onits characteristic Rf. The dealkylated metabolite(3-OH-morphinan) was identified by its melting-point,by the mixed melting-point of the hydrobromide, byinfra-red spectrum, by countercurrent distributionand by its Rf. The metabolite of unknown structurewas obtained in crystalline form by sublimation, itsempirical formula was established by elementalanalysis, and its melting-point and ultraviolet andinfra-red spectra were determined.

Absorption and distribution. There appear to beno quantitative data on the rate of absorption ortissue distribution of levallorphan. Brossi, Hafliger& Schnider (1955) were unable to detect the drugin dog blood one hour after subcutaneous injection

265


of 10 mg/kg of levallorphan tartrate. It is possiblethat levallorphan leaves the blood and enters thetissues rapidly, and shortly thereafter either leavesthe tissues, as is true of nalorphine, or is inactivatedin situ. Thus, as in the case of nalorphine comparedwith morphine, levallorphan appears to have amuch shorter duration of action than levorphanol,since after subcutaneous injection of a 1:1 mixtureof the drugs there is only a temporary suppressionof the hyperglycemic response of dogs to levor-phanol (Pittinger, Gross & Richardson, 1955).

Metabolism and excretion. The metabolism oflevallorphan appears to be much more complicatedthan that of the N-methyl analogue, levorphanol,or of the corresponding morphine derivative,nalorphine. Semi-quantitative data presented byBrossi, Hafliger & Schnider (1955) for the dogindicate that only 1.6-3.0% of the dose can berecovered, and this percentage is present as levallor-phan in urine collected for 6 days after 2 subcu-taneous doses of 10 mg/kg of the tartrate given7 hours apart. Certain objections to this work havebeen discussed above, More complete as well asquantitative data, furnished by Mannering &Schanker (1958) for the rat, show that about 85% ofthe drug is unaccounted for under either in vivoor in vitro conditions. Furthermore, the 15% thatcan be accounted for consists of several metabolicproducts, including conjugated levallorphan, thefree and conjugated forms of the N-dealkylateddrug, i.e., 3-hydroxymorphinan, and the free andconjugated forms of an oxidation product of un-known structure, designated Metabolite I. Thelast-mentioned compound is a phenolic alkaloidwith a high melting-point (254-256°C) and has theempirical formula C19H2502. It is not known justwhere the additional atom of oxygen is attached,but by a series of physico-chemical tests the authorshave established that the following positions in thelevallorphan molecule have apparently not beensubject to oxidation: the allyl side-chain; the piperi-dine nitrogen; carbons 2, 4 and 10 of the phenan-threne skeleton. Metabolite I is formed in vivo andin vitro by liver slices of rats, mice and rabbits,although the metabolite may be slightly modifiedin the case of the rabbit. Guinea-pigs and dogs failto produce this oxidation product either in vivoor in vitro. All species studied, however, do excrete3-hydroxymorphinan after injection of levallorphan,and all but the rabbit produce the dealkylated meta-bolite in vitro in the presence of liver slices.

Quantitative aspects of the in vitro metabolismof levallorphan were studied using rat liver slices.During the first hour of incubation approximatelytwo-thirds of the added levallorphan disappearedand could not be accounted for. At this time about8% of the drug was in the form of the oxidationproduct, another 2% was in the form of 3-hydroxy-morphinan, and about 22% was in the form ofunchanged levallorphan. After the first hour ofincubation there is relatively little additional increasein the amounts of metabolites recovered, althoughthere is a continued appreciable disappearance oflevallorphan itself. At the end of 3 hours theamounts of recovered compounds represent 12%of the initial amount as Metabolite I, 3.4% as3-hydroxymorphinan, and 3% as levallorphan.Practically no conjugation of these alkaloids takesplace in vitro.

Quantitative data on the excretion of levallorphanin the rat after a single intraperitoneal injection of250 mg/kg of levallorphan tartrate indicate strikingsimilarities between in vivo and in vitro metabolism,including the fact that most of the drug cannot beaccounted for and that, of the identified compounds,the oxidation product is present in greatest amount.However, in contrast to in vitro metabolism, thein vivo changes include appreciable conjugation ofall three identified alkaloids. The actual recoveryof material in the 24-hour urine of 5 rats amountedto an average of 12.0% of the administered dose.In another series of 3 rats, the recovery averaged13.5% of the dose and included 1.3% as free levallor-phan, 3.8% as bound levallorphan, 3.6% as freeMetabolite I, 3.3% as bound Metabolite I, 0.3%as free 3-hydroxymorphinan, and 1.2% as bound3-hydroxymorphinan. Neither faecal excretion nordelayed urinary excretion could account for thelarge missing fraction of the dose, since they repre-sented, respectively, only 1.1% and 0.8% of thedose as the identifiable alkaloids. Thus the fateof most of the dose of levallorphan is unknown(Mannering & Schanker, 1958).

METHADONE

The compound 6-dimethylamino-4,4-diphenyl-3-heptanone is obtained as a racemate and was firstsynthesized by Bockmuhl & Ehrhart (1949). Re-solution of the isomers of the compound has beenachieved and the pharmacology of each has beenintensively studied (Scott, Robbins & Chen, 1958;Thorp, 1949). The compound is available com-

266


mercially only as the racemate and hence most of thedisposition studies have been carried out on dl-methadone. Henceforth in this study, the term" methadone " refers to the racemic mixture.Although the i-isomer accounts for most, if not all,of the analgesic activity of the racemate, the datato date indicate that the 1- and d-isomers havecommon metabolic pathways.

Methods of estimation

Various procedures have been used to estimatemethadone in biological media, but none of themethods distinguishes between the d- and thei-isomers.

Three groups of workers have employed indicator-dye techniques that are similar in principle for esti-mating methadone; Scott & Chen (1946), bromo-thymol blue; Cronheim & Ware (1948), bromocresolpurple; and Way, Sung & McKelway (1949), methylorange. With these procedures, any basic aminewhich can form an organic-solvent-soluble dyecomplex will react as methadone and precautionsneed to be taken to eliminate any interfering sub-stances. Of the above methods, only the methylorange procedure meets this prerequisite. A highdegree of specificity for estimating the compoundin tissue was obtained by washing the solvent extractwith phosphate buffer to remove metabolic productsand tissue blank substances which reacted with themethyl orange reagent (Way, Sung & McKelway,1949). This treatment did not confer absolutespecificity on the method in respect of the determi-nation of methadone in faeces and urine, but it waslater shown that approximately 80% or more of themethyl orange reactants measured in urine wasessentially methadone (Way, Signorotti, March &Peng, 1951).Methadone has also been estimated by tracer

methods (Elliott, Chang, Abdou & Anderson,1949) by using methadone-2-14C (Tolbert, Christen-son, Chang & Sah, 1949) and observing the distribu-tion of radioactivity in tissues and excreta. Theassumption that radioactivity represented metha-done concentration was based on evidence that noradioactivity was detectable in the expired air andon the results of preliminary isotopic dilution experi-ments on rat bile. Rickards, Boxer & Smith (1950)and Schaumann (1960b) raised objections to thisinterpretation, pointing out correctly that theisotopic experiments as carried out did not excludethe degradation products of methadone.

Another method for determining methadone intissues, reported by Rickards, Boxer & Smith (1950),is based on the release of the compound after dis-integration of the tissue with strong alkali, extrac-tion into ether, nitration of the phenyl radicals inmethadone and the subsequent development ofcolour with ethyl methyl ketone. Recovery of thedrug from urine was 100% and the reproducibilitywas reported to be ±5% in the concentration rangeof 5-50 Htg per ml of urine; recovery of the drug fromvarious tissues was less complete and more variable,but it was adequate for the purpose of their experi-ments. The authors did not evaluate the specificityof the method since they felt that the essentialsteps of the procedure conveyed sufficient specifi-city. However, any methadone metabolic fragmentretaining the phenyl and amine groups will react asthe parent substance. Since evidence to this effecthas been presented (Way, Signorotti, March &Peng, 1951) and since N-demethylation has beenrecently reported to be a metabolic pathway forthe compound (Axelrod, 1956a; Pohland, Sullivan& Lee, personal communication), it would appearthat the methyl ketone procedure is subject to pos-sible interference by these unidentified metaboliteswhen they are present in significant quantities.An improved bromocresol purple procedure

(Soehring & ULhr, 1950) and forensic methods forthe detection of methadone have been reported,but to our knowledge they have not been appliedto metabolic studies (Vidic, 1951, 1955, 1957;Breinlich, 1953a; Jatzkewitz, 1954). A bio-assayprocedure, using the isolated guinea-pig gut, hasbeen reported by Schaumann (1960b) to be sensitiveto methadone at a concentration as low as 10-8 M.

Absorption

The onset of action of methadone is relativelyprompt and absorption studies bear this out. Mostof the quantitative studies have been carried outon the rat.Within 10 minutes after subcutaneous injectioni

of "4C-methadone appreciable concentrations of'4C appear in the plasma (Adler & Eisenbrandt,1949) and bile (Eisenbrandt, Adler, Elliott &Abdou, 1949). After subcutaneous injection oflabelled methadone, 10 mg/kg, into the left hind legsof rats, less than 15% of the radioactivity remainedat the injection site after 2-3 hours and virtually noactivity was present after 24 hours (Elliott, Chang,Abdou & Anderson, 1949). Rickards. Boxer &

9

267


Smith (1950) found that, after injection of methadone,15 mg/kg, subcutaneously in rats, 47% of the doseremained at the injection site after 1 hour, 10%after 3 hours and only 3% after 5 hours.

After administration of methadone, 20 mg/kg,by stomach tube to fasted rats, Way, Sung &McKelway (1949) found that within 2 hours 70%of the dose had disappeared from the gastro-intestinal tract. The rate of disappearance wasinitially very rapid, but levelled off after the 2-hourperiod. Decomposition of the drug in the enterictract was assumed to be negligible, on the basis thatno decrease in methadone levels was evident whenthe drug was incubated with the tissue. It was furtherconcluded that the rate of absorption was evengreater than indicated by these experiments, sincethe amount in the gastro-intestinal tract is due toexcreted material as well as to unabsorbed drug.

Distribution

Several studies on the distribution of methadonehave been reported, all on the rat. Although notall methods have been evaluated for specificity, thelevels found for various tissues, with the exceptionof the excretory organs, are roughly of the sameorder of magnitude. It appears from the reports thatthe methods utilizing photometric techniques (Way,Sung & McKelway, 1949; Rickards, Boxer & Smith,1950) measure methadone with greater specificitythan the method employing tracer procedures (Elliott,Chang, Abdou & Anderson, 1949). A comparisonof actual values is difficult because of differences inexperimental conditions (dosage, mode of administra-tion and time of sacrifice).There appears to be general agreement concerning

the affinity of a particular organ for methadone.On entering the systemic circulation, the compoundrapidly leaves the blood and localizes in the lungs,liver, kidneys and spleen, but the brain, muscle andheart contain only low concentrations (Way, Sung& McKelway, 1949; Elliott, Chang, Abdou &Anderson, 1949; Rickards, Boxer & Smith, 1950;Schaumann, 1960b). The adrenals and thyroid alsoconcentrate the compound (Adler & Eisenbrandt,1949; Elliott, Chang, Abdou & Anderson, 1949;Miller & Elliott, 1955). An attempt was made tocorrelate the concentration of the drug in the brainwith the analgesic effect, but the results were incon-clusive owing to the limitations of the analyticalmethod used for determining the agent (Rickards,Boxer & Smith, 1950). Subsequently, Miller &

Elliott (1955), using the more sensitive tracer tech-niques, showed that the levels (0.6-0.9 ug/g) whichwere found in various segments of the centralnervous system 30 minutes after subcutaneousadministration of 3 mg/kg correlated well with theintensity and duration of " analgesic " effect (reac-tion time of tail to thermal stimulus). Initial uptakeof methadone is rapid, with peak levels occurringwithin one or two hours after intraperitonealadministration (Way, Sung & McKelway, 1949).A large part of the drug present in the whole animalis found in the carcass (primarily skeleton, muscleand bone) (Elliott, Chang, Abdou & Anderson,1949; Rickards, Boxer & Smith, 1950). A highconcentration of methadone has been reportedin the gastro-intestinal tract after subcutaneousadministration (Elliott, Chang, Abdou & Anderson,1949), but other workers (Way, Sung & McKelway,1949; Rickards, Boxer & Smith, 1950) concludedthat a considerable portion of this was due to meta-bolic products reacting like the parent compoundin the analytical procedure. Radioactivity was alsonoted in the placentae and the foetuses of a pregnantrat after the administration of labelled methadone(Elliott, Chang, Abdou & Anderson, 1949).Although methadone appears to be firmly bound

to tissue protein (Sung, Way & Scott, 1953), accu-mulation of the drug does not occur to any greatextent. In rats given repeated high daily doses ofmethadone for several days, little or none of thedrug can be recovered from tissues 24 hours afterthe last dose, owing to metabolic alteration andexcretion of the compound (Way, Sung & McKel-way, 1949).

MetabolismIn vivo as well as in vitro evidence indicates that

methadone is rapidly and extensively metabolized.Recent evidence has established that N-demethyla-tion is a metabolic pathway for methadone. Severalinvestigators have reported finding biotransforma-tion products of methadone. At the present timeit is not known for certain whether these partiallypurified substances represent a common metaboliteor different compounds, but it appears that a des-methyl derivative of methadone may be the substancefound by the various groups of workers.

Considerable evidence indicates that methadonemay be N-demethylated to a des-methyl derivativeof methadone. Elliott, Chang, Abdou & Anderson(1949), using methadone labelled with 14C in posi-tion 2, found no 14CO2 eliminated by rats given the

268

BIOLOGICAL DISPOSITION OF MORPHINE AND ITS SURROGATEs-3

compound. This indicates that the first two carbonatoms of methadone are not removed by oxidation.Way and co-workers studied the biliary excretionof 1-, d-, and di-methadone in the rat and found atleast two bases excreted in the bile as methyl orangereactants. One of the organic bases had solubilityproperties different from those of methadone, beingpartitioned much more readily from the organicsolvent phase by acetate buffer (Way, Signorotti,March & Peng, 1951; Sung & Way, 1953b). Repeatedcounter-current transfer yielded a yellowish greenmaterial which turned red on standing in air. Theinfra-red spectrum of this unknown base yieldedabsorption consistent with the presence of mono-substituted benzene (Way, Sung & Peng, unpublisheddata). Miller & Elliott, using labelled methadone,confirmed these studies in 1955 and noted that theacetate-partitioned base was radioactive. Thus theevidence indicated that the unknown base hadretained many of the characteristics of methadoneand hence most of the methadone molecule wasstiU intact. Axelrod (1956a) noted that incubationof i-methadone with rat and rabbit liver micro-somal preparations resulted in the liberation offormaldehyde and the formation of a benzene-soluble ninhydrin-reacting substance. Vidic (1957),by means of paper chromatography, separated somerenal excretion products of methadone and foundevidence for the N-demethylation of methadone inthe presence of a primary and a secondary aminein his chromatogram. Finally, Pohland and hisassociates (personal communication) recently foundthat in attempting to synthesize des-N-methyl-methadone, a cyclic compound- ,5-dimethyl-3,3-diphenyl-2-ethylidinopyrrolidine-was obtained asa result of the ready dehydration of the intermediatehemiketal form of des-N-methylmethadone. Ametabolite isolated from incubates of rat liver sliceswith methadone or from the bile of a dog givenmethadone behaved like the cyclic product derivedfrom des-N-methylmethadone. A comparison ofthe infra-red absorption of 1,5-dimethyl-3,3-di-phenyl-2-ethylidinopyrrolidine and the base isolatedfrom rat bile by Way, Peng & Sung (unpublisheddata) indicates that the two compounds are verysimilar if not identical. Thus, there is strong evidencethat N-demethylation is an important metabolicpathway for methadone.The evidence also suggests that both methyl

groups may be split off to yield the correspondingprimary amine of methadone, since both Axelrod(1956a) and Vidic (1957) used reagents which are

generally known to react with primary amines.However, proof by isolation of this product incrystalline form remains still to be obtained.

Evidence that N-methylation is a possible meta-bolic pathway for methadone was reported bySchaumann (1960a). Incubation of methadone withliver slices of guinea-pigs in Krebs-Ringer solutionresulted in a disappearance of methadone and theappearance of methylmethadone. Approximately15% of the added methadone was accounted for atone hour as the quaternary compound. After onehour, although the methadone continued to dis-appear, little or no additional methylmethadonewas formed. Satisfactory identification of the bio-transformation product as methylmethadone wasaccomplished by paper chromatography and byseparate mixed-melting-point determinations of twocrystalline salts prepared from biosynthetic andauthentic methylmethadone. While evidence formethylation of methadone has not been establishedin vivo, Schaumann concluded that the unknownbound metabolite of methadone excreted in theurine of methadone addicts previously reported byVidic (1957) was methylmethadone.The liver appears to be chiefly responsible for

metabolizing methadone. Incubation of the com-pound with liver slices taken from rats caused therapid disappearance of methadone (Sung & Way,1950; Way, Sung & Fujimoto, 1954; Schaumann,1960a, 1960b), whereas slices from other tissuesshowed little or no activity (Sung & Way, 1950).The optimum pH for the reaction was observed tobe between 7.5 and 8.5. Oxygen was found to beessential. The presence of hydroxylamine, azide,cysteine, glutathione and ascorbic acid, or heatingstrongly, inhibited the reaction (Sung & Way, 1950).Partial hepatectomy caused rats to be more suscep-tible to the effects and prolonged the pharmacologicalactions of methadone (Sung & Way, 1950) as well asof i-methadone (Bonnycastle & Delia, 1950). Con-comitantly, higher levels of methadone were ob-tained in most tissues, especially in the heart, lungs,spleen and kidneys (Sung & Way, 1950). Liver slicesobtained from one woman at autopsy also metabol-ized methadone (Sung & Way, 1950).

ExcretionSeveral studies on the excretion of methadone in

man and the rat have been reported. The availableevidence indicates that less than 10% of the drugis excreted unchanged in the urine and in the faeces.A major fraction of the compound can be accounted

269


for in the urine and faeces as chiefly unknown bio-transformation products.Very little methadone is excreted unchanged in

the urine by man. Way, Signoretti, March & Peng(1951) reported that only 4% of the dose wasexcreted by a subject after receiving 20 mg orally;larger amounts of an unknown base different frommethadone were also present in the urine. Threeother groups of workers (Scott & Chen, 1946;Cronheim & Ware, 1948; Davis, Andres & King,1952) reported higher values for methadone excretedin the urine, but the percentages given in thesestudies should be considered maximum values, sinceevidence of the specificity of the method for metha-done was not furnished and it appears that theunknown base, presumably a demethylated deri-vative of methadone, would not be excluded bythese procedures. Scott & Chen (1946) reportedthat in 3 normal adult males about 20-35% of an

oral dose of 5-7.5 mg was excreted over a periodof 24 hours. Davis, Andres & King (1952) recovered12-25% from the urine of patients injected with10 mg subcutaneously after 36-80 hours. Cronheim& Ware (1948) reported 5.7-13.7% in the urine of5 subjects receiving several 10-mg doses of metha-done subcutaneously.

In the rat the excretion of unchanged methadonein the urine and faeces is also low. Elliott, Chang,Abdou & Anderson (1949), using '4C-labelledmethadone, reported that 70% of the dose wasexcreted as unchanged methadone in the intestinalcontents and faeces, and about 30% in the urine,after subcutaneous injection of 10 mg/kg, butRickards, Boxer & Smith (1950) pointed out thatthe radioactivity measured did not represent intactmethadone. The available evidence would indicatethat the urinary and faecal excretion of unchangedmethadone by each route is not nearly so high.Rickards, Boxer & Smith (1950) reported that onlyabout 10% of this dose was recovered respectivelyin the urine and faeces. Way, Sung & McKelway(1949) found that only 4-11 % of the total dosageof methadone administered to rats could be ac-

counted for in the 24-hour urine as the unchangedcompound; and 19-24% in the faeces. It was

indicated at the time, however, that since themethod was not absolutely specific for methadone inexcreta, some revision of values would be neces-

sary. Subsequent studies (Way, Signorotti, March& Peng, 1951), utilizing counter-current techniques,indicated that the previous values for urine were

only slightly too high, a reasonable correction

factor being about 0.8, whereas the values for faeceswere much too high, the correction factor beingabout 0.25. The high values for faeces representincomplete removal of an unknown organic basicmetabolite, which is probably an N-demethylatedderivative of methadone, and indicate that theamount of unchanged methadone excreted in faecesshould be about 5% rather than 20%.

Biliary excretion is an important avenue for theelimination of methadone and its biotransforma-tion products. Eisenbrandt, Adler, Elliott & Abdou(1949) established that the radioactivity present inthe small intestine of the rat after receiving labelledmethadone was largely the result of the inflow ofradioactive bile. As much as 17% of an injecteddose of 5 mg/kg was recovered from the bile duringa period of 3 hours. These findings were confirmedin the dog as well as in the rat (Way, Signorotti,March & Peng, 1951; Sung & Way, 1953b). Theseworkers also reported that an acetate-soluble methylorange reactant (demethylated methadone) wasexcreted in larger amounts in the bile than metha-done. Miller & Elliott (1955) reported the presenceof another methadone metabolite in the bile whichhad low solubility in acetate buffer and high solu-bility in ethylene dichloride. Schaumann (1960b)reported that in the rat only a small fraction of thematerial present in the bile which reacted as metha-done by a modified dye method behaved as metha-done on bio-assay.

Factors influencing the disposition ofmethadone(a) Isomerism. Although the d- and i-isomers of

methadone differ greatly in their pharmacologicalactions (Scott, Robbins & Chen, 1948; Thorp, 1949),it appears that they have certain metabolic pathwaysin common. After intraperitoneal administrationof each isomer, the tissue distribution of the d- andthe i-forms was found to follow the same patternas that of racemic methadone and the levels foundwere essentially of the same order of magnitude.Despite the greater analgesic potency of i-methadone,its gross distribution in the brain was similar to thatof the d-isomer. The difference in the urinary andfaecal excretion of d- and i-methadone was notstriking. The rate of metabolism of the two isomersby liver slices, as measured by the rate of disappear-ance of the isomer, is approximately the same. Thereaction is inhibited by hydroxylamine, azide, cys-teine, ascorbic acid, glutathione, fluoride, andcyanide, but none of the agents exhibited a differen-tial effect between the d- and i-forms. The optimal

270


pH for the reaction was also the same for bothisomers (Sung & Way, 1953b).These common requirements suggest that the

disappearance of either isomer when incubatedwith liver slices may be due to a reaction catalysedby a common enzyme system. Since apparentlylittle or no influence on the reaction rate is exertedby the asymmetric carbon atom on the 6-position,the results of these liver slice experiments appearto be inconsistent with those obtained with livermicrosomal preparations, where it has been shownthat N-demethylation of the i-isomer occurs muchmore readily than that of the d-form (Axelrod,1956a). These results could be reconciled by postulat-ing that in the presence of liver slices, but not withisolated microsomes, the predominant reactioninvolves alteration of the methadone molecule at aplace more distant from the site of asymmetry thanthe N-methyl position.The well-known marked differences in the pharma-

cological activity of the d- and i-forms of metha-done suggest that the receptors mediating thebiological action of the drug exhibit a differentsensitivity or affinity for the two isomers. In con-trast, the disposition studies suggest that the recep-tors concerned with the general uptake and themetabolism of the. d- and i-forms are relativelynon-specific, although it cannot be ruled out thateach set of receptors may exhibit similar activity.It would appear, therefore, that the receptors whichmediate the action of the drugs are different fromthose concerned with the inactivation of the drug(Sung & Way, 1953b).

(b) Tolerance to methadone. Development oftolerance to methadone cannot be satisfactorilyexplained on the basis of increased rate of metabolicdegradation, alteration in gross tissue distribution,increased degree of tissue binding and/or increasedrate of urinary or faecal excretion of the compound(Sung, Way & Scott, 1953). Only smaUl differences inthe distribution of methadone in the blood, brain,heart, kidney, liver, lung, muscle, and spleen werefound between a group of six tolerant and a group offive normal rats, and the variations in tissue levelsfrom animal to animal within each group weretoo great for any significance to be attached to thefindings. Over a 24-hour period normal and tolerantrats excreted approximately the same amount ofmethadone in the urine, but tolerant rats were foundto excrete a larger quantity of an organic basicmetabolite in the faeces. Liver slices from tolerantrats were found to metabolize methadone less rapidly

than those taken from normal animals. Musclehomogenates from normal and tolerant rats werefound to bind methadone to the same degree.The distribution of methadone in rats made toler-

ant to the compound was compared to that in normalrats by Rickards, Boxer & Smith (1950). Theyreported that the fraction of the injected methadonedose recovered from the whole animals was notsignificantly different in the normal and in the toler-ant animals, but that there was some difference inthe distribution of the drug in the two groups. It wasobserved that the percentage of the injected dose ofmethadone present in the intestines and the skin ofthe tolerant animals was greater than in the normalanimals. Nearly twice as much of the drug was pre-sent in the carcass (skeleton and muscle) of the latterthan in that of the former. Although it was notedthat the values obtained were at the lower limits ofsensitivity of the method used by the authors forestimating methadone, it was pointed out that thefraction of injected methadone found in the brain ofthe tolerant rats was decidedly lower than in that ofthe normal animals. Since three of the four valuesgiven are below the reported sensitivity (1 pg) ofthe method, such a statement appears questionable.

(c) Endocrine effects. The effect of thyroidectomyand ofthiouracil or thyroid feeding on the actions andfate of methadone was investigated by Sung & Way(1953a) after it had been noted that rats made toler-ant to methadone had enlarged thyroids and reducedability to concentrate radioactive iodine (Sung, Way& Scott, 1953). Thiouracil feeding and thyroid-ectomy were found to increase tolerance to metha-done, even though the metabolic breakdown of thecompound was retarded and tissue levels were eleva-ted. Thyroid feeding increased susceptibility tomethadone. This was found to be due in part to aslower rate of degradation of methadone by the liver(Sung & Way, 1953a).

Despite the widespread metabolic effects of testo-sterone, it apparently exerts but little influence onthe actions and fate of methadone. The drug appearsto localize in the kidneys of rats to a greater extent infemales and castrated males than in normal males.This can be prevented by administering testosterone.The effect cannot be satisfactorily explained and it isbelieved to be caused possibly by an unknown func-tion of testosterone (March, Gordan & Way, 1950).The adrenals apparently have profound effects on

the disposition of methadone, and this may markedlyalter the pharmacological effects of the compound.Winter & Flataker (1951) reported that cortisone or

271


ACTH antagonized the analgesic effects of metha-done in the rat, as well as the effects of toxic doses ofthe drug in mice. They suggest that the possiblemechanism may be related to the change in ionicbalance and/or the fluid shift effected by adreno-cortical hormones within nervous tissue. However,Elliott & Elison (1959), on the basis of distributionstudies, concluded that this antagonism was relatedto an enhancement by cortisone in the rate of pas-sage of the drug through the body. In experimentson cortisone-treated and on adrenalectomized rats,they measured concentrations of methadone at theinjection site and in the intestines, an importantexcretory site. At one hour after intrapoplitealinjection, cortisone-treated animals exhibited lowestlevels of methadone at the injection site and highestlevels at the excretory site; this pattern was reversedin adrenalectomized animals, while in normal ani-mals the levels of methadone at both sites wereintermediate between those of the cortisone-treatedand the adrenalectomized animals. Since differenttreatments did not appear to affect the rate ofmetabolism of methadone, it was concluded thatcortisone enhancement of methadone absorptionwas accompanied by an even greater enhancementof the excretion of the compound, with the resultthat the sojourn of methadone in the body and itspharmacological effects were decreased.

ACETYLMETHADOL

,r±-Acetylmethadol (6-dimethylamino-4,4-diphenyl-3-aceoxyheptane) is a closely related derivative ofmethadone. Owing to the presence of two asym-metric carbon atoms in the molecule, two pairs(a and ,) of diastereoisomers were obtained. Thea-isomers were found to be more potent analgesicsthan the ,-isomers. The letters 1- and d- refer onlyto the sign of rotation for the a-acetylmethadols,since the laevorotatory isomer is derived fromd-methadone and the dextrorotatory from i-metha-done (Eddy, May & Mosettig, 1952).Both the l- and the d-enantiomers are highly

potent pharmacological agents, but differ consider-ably in the onset and duration of effect. The 1-isomer of acetylmethadol has a delayed onset, butlong duration of action, compared with the dextro-rotatory isomer (Eddy, May & Mosettig, 1952).

Method of estimationThe a-acetyhmethadol isomers were each deter-

mined by a modification of the methyl orange tech-

nique. The distribution ratios of the basic substanceextracted from rat tissues 4 hours after administra-tion of each isomer subcutaneously showed solubilitycharacteristics practically identical with those of theparent compound which was administered (Sung &Way, 1954).

Absorption

In man, after subcutaneous or intravenous injec-tions of l-acetylmethadol, Fraser & Isbell (1952)found that the euphoric effects in former morphineaddicts did not appear for 4-6 hours. Once developed,the effects of a single dose persisted for 24-72 hours.After oral administration of the same compound,however, the morphine-like effects were observed in1 hour and persisted for 48 hours. Keats & Beecher(1952), studying post-operative patients, foundthat /-acetylmethadol given subcutaneously pro-duced analgesia within 90 minutes, but delayedtoxic effects were noted 12-30 hours after theadministration of the compound. On the otherhand, subcutaneous or intravenous doses ofthe d-isomer promptly induced a train of mor-phine-like effects which disappeared in less than24 hours.

Quantitative studies in the rat were carried out bySung & Way in 1954. Although the i-isomer appearsto give delayed morphine-like effects, the experimentsindicated that the 1- and d-isomers were absorbed atroughly the same rates. After subcutaneous injec-tion in the left leg, each compound disappeared fromthe site of injection at about the same rate. Over70% of each isomer was found to be absorbedwithin 1 hour and approximately 7% of the dose wasstill recoverable from the injection site at the end of13 hours. The rate of disappearance of each isomerwas much slower after oral administration. Morethan four-fifths of the dose of each isomer wasrecovered from the stomach at the end of one hour.However, at the end of 13 hours less than 2% of thed-isomer was recovered from the stomach, while50% of the administered dose of the i-isomer waspresent even at 24 hours. Subsequent experiments,described below, indicated that an appreciableamount of the compound present could be attri-buted to secretion of the drug into the stomach.The active secretion of l-acetylmethadol into thestomach would tend to promote a longer dura-tion of effect but a lessened intensity of response,since the stomach would be acting as a depot forthe drug.

272


DistributionTissue levels of l-acetylmethadol after oral and

subcutaneous (left leg) administration of 20 mg/kgwere compared in fasted adult Long-Evans rats(Sung & Way, 1954). Consistent with the findings inrespect of the absorption rate, the levels in all organs,except the gastro-intestinal tract, were higher andpersisted longer after subcutaneous than after oraladministration. The drug was found in most tissuesone hour after the administration of the compoundby either route. Except for the site of administration,the highest concentrations were found in the lung.The kidney, spleen, liver and fat usually containedappreciable levels, whereas very low levels were pre-sent in the heart, blood and brain. A large propor-tion of the drug was accounted for in the carcass(chiefly muscle and bone), although the concentra-tion of the drug was somewhat lower there than inother organs. Considerable levels were still presentin most tissues at 24 hours. High concentrationswere found in the stomach contents even after sub-cutaneous or intravenous injection. Despite theinitial rapid tissue uptake of the l-acetylmethadolafter parenteral administration, and the resultinghigh tissue levels of the compound, morphine-likeeffects did not appear until 4-6 hours after theinjection, at which time the animal became depressedand sometimes cataleptic for as long as 24 hours.Brain levels at this time were barely detect-able. Following oral administration, however,the catalepsy was observed in about one houreven though the tissue levels were lower. Sincethere appears to be little correlation between thelevels of l-acetylmethadol attained in tissue andthe morphine-like effects, it was concluded thatsome unknown metabolite of the compound may beresponsible for the morphine-like effects and thatthe formation of this metabolite may be facilitatedby oral administration.The distribution pattern of d-a-acetylmethadol

after subcutaneous and oral administration wassimilar to that of the i-isomer, although the corres-ponding tissue levels were lower. However, unlikethe i-isomer, organ levels of the d-isomer were notnearly as sustained. Twenty-four hours after sub,-cutaneous injection of the d-isomer, the drug washardly detectable in any tissues other than thegastric tract and the lungs. The d-isomer disap-peared from the stomach more rapidly than thei-isomer after oral administration and its localiza-tion in the stomach after subcutaneous injection wasless marked than with the i-isomer.

Both the isomers of a-acetylmethadol are boundto the same extent by the various tissue homogenates.The binding by the brain and other tissue was con-siderable. The recovery of each isomer from theultrafiltrate of various homogenates to which eachacetylmethadol had been added was between 10%and 25 %.

MetabolismMost tissues (especially liver, kidney and muscle)

slowly metabolize I-acetylmethadol to some degree.Although other factors may play a contributoryrole, the longer duration of action of the l-acetyl-methadol as compared with the d-isomer appearedto be due chiefly to a slower rate of metabolism.In experiments on mice, between 40% and 50%of the administered drug was still present after12 hours. Twenty-four hours after the injection,the l-acetylmethadol was still present in mostanimals in appreciable amounts, but it was barelydetectable after 48-72 hours. The decline in tissuelevels in the rat appeared to be more rapid, but thepharmacological effects were sustained.As mentioned above, there appears to be little

correlation between the observable morphine-likeeffects and the levels of 1-acetylmethadol attainedin the organs, and, as a consequence, it was suggestedthat some unknown metabolite may be responsiblefor the effects. This argument may also be supportedby the fact that l-acetylmethadol is anomalous inthe sense that it does not fit the stereoconfigurationspostulated by Beckett & Casy (1954) for analgesicactivity. The delayed action of the compound alsolends credence to the view that it exerts its actionin some other form (Sung & Way, 1954).

ExcretionOnly a negligible amount (average, less than 3%

of the dose) was found in the faeces and urine ofrats after subcutaneous administration of 1- or d-a-acetylmethadol. Bile taken continuously from thecannulated bile ducts yielded very low amounts ofeither isomer. High concentrations of both com-pounds appear to be actively secreted into thestomach after intravenous or subcutaneous adminis-tration, but the d-isomer disappeared more rapidly,as indicated by the higher gastric levels of thei-isomer found after 13 and 24 hours (Sung &Way, 1954). These investigators concluded that, sincebiliary excretion of acetylmethadol is not important,localization of the acetylmethadol in the stomachis not due to regurgitation from the duodenum as

273


is the case with methadone (Eisenbrandt, Adler,Elliott & Abdou, 1949). Since the total excretionof both isomers in the rat was negligible, even aftertissue levels had virtually disappeared, it was con-cluded that the drug is disposed of essentially bybiotransformation (Sung & Way, 1954).

PROPOXYPHENE

The compound a-dl-4-dimethylamino-1,2-di-phenyl-3-methyl-2-propinoxybutane was synthesizedby Pohland & Sullivan in 1953. In 1955, the sameauthors resolved the dl-mixture and reported thatthe analgesic activity resided in the d-isomer. Thed-isomer of the hydrochloride salt is the commer-cially available preparation. The pharmacologyof the compound in animals has been described byRobbins (1955) and its clinical efficacy as an anal-gesic by Gruber and his associates (Gruber, King,Best, Schieve, Elkus & Zmolek, 1955; Gruber,Miller, Finneran & Chemish, 1956; Gruber, 1957).

Methods of estimationTwo methods have been used to determine the

compound in biological fluids (Lee, Scott & Poh-land, 1959). Both procedures were found to measurepropoxyphene plus biotransformation products.The N-14CH3-labelled compound was synthesizedand its radioactivity measured, and the non-labelledcompound was determined by the methyl orangetechnique. In the latter procedure, in order to reducethe amount of interfering substances reacting withmethyl orange, solvent extraction of propoxyphenewas effected under strongly alkaline conditions.It was found that at pH 11 and above, althoughabsolute specificity was not attained, the methylorange values were much lower than at pH 9. Theincreased yield of methyl orange reactants at pH 9was attributed to the extraction from the urine ofsignificant amounts of the des-N-methyl metaboliteof propoxyphene. It was suggested that interferencewith the methyl orange procedure by des-N-methylpropoxyphene was less at a high pH because, understrongly basic conditions, the compound probablyundergoes structural rearrangement, possibly by a

migration of the propionyl group from 0 to N toyield a neutral amide which would fail to form an

organic-solvent-soluble complex with methyl orange.

Absorption and distributionQuantitative measurements of the rate of absorp-

tion of propoxyphene and of its distribution in

various organs have not been reported. In experi-ments on rats Robbins (1955) reported peak anal-gesic effects at 60-90 minutes after administrationof 5 and 10 mg/kg.

MetabolismIn vivo as well as in vitro experiments indicate that

N-demethylation of propoxyphene is an importantmetabolic pathway (Lee, Scott & Pohland, 1959).A dealkylated metabolite, des-N-methyl propoxy-phene, was isolated as a dinitrophenyl derivativefrom the pooled urine of six volunteers who tooklarge doses of the drug. The identity of the dinitro-phenyl derivative of a des-N-methyl propoxyphenewas established by elemental analysis and by infra-red absorption studies.The rate of N-demethylation of propoxyphene was

followed in the rat in vivo and in vitro, using theN-14CH3-labelled compound and measuring 14CO2at various times. After intravenous injection of4 mg/kg of labelled propoxyphene, approximately38% of the radioactivity was eliminated as '4CO2within 22 hours. Beyond the 22-hour period little orno radioactivity appeared in the expired air. Thehalf-time for elimination of 14CO2 was approximately130 minutes. When labelled propoxyphene wasincubated with rat liver slices there was a constantincrease in the rate of 14CO2 eliminated up to 60minutes, after which the rate diminished. Similarstudies with lung, brain, spleen, mammary glands,stomach and whole blood yielded no 14CO2 after30 minutes' incubation. The kidneys exhibited someN-demethylating ability, but the amount of 14CO2liberated was only about 0.5-2.5% of that foundwith liver slices.

ExcretionVery little propoxyphene is excreted unchanged in

man or in the rat (Lee, Scott & Pohland, 1959). Inman the urinary excretion of propoxyphene wasfound to be quite low, despite the fact that the methylorange procedure used was found to be non-specific.The values reported, therefore, represent maximumvalues. After administration of 100 mg orally every4 hours for 16 hours, the amount of the dose recov-ered in the 24-hour urine in 3 subjects respectivelywas only 3%, 10% and 3%; the amount accountedfor in the 24-48-hour urine was even lower.

In the rat the urinary and faecal excretion of pro-poxyphene was followed for 48 hours, using labelledpropoxyphene. Most of the radioactivity in theurine appeared during the first 22 hours. During the

274


first 6 hours about 2.9% of the radioactivity appearedin the urine; at 22 hours the urinary excretion was

about 13% and at 48 hours it was 15%. Roughly35 % of the dose appeared in the faeces at 48 hoursas radioactive material.

PETHIDINE

Pethidine, ethyl 1-methyl-4-phenylpiperidine-4-carboxylate, is known under a variety of synonyms,including meperidine and isonipecaine. Since itsintroduction (Eisleb&Schaumann, 1939; Schaumann,1940) considerable information has accumulatedconcerning its biological disposition. Studies on

this compound first established the importance ofN-demethylation as a metabolic pathway for thenarcotic analgesics.

Methods of estimation

The compound has been determined in variousbiological media, principally by the use of indicatordyes. Bromothymol blue and methyl orange havebeen employed to form organic-solvent-soluble com-

plexes with pethidine which can be measured photo-metrically. The bromothymol blue methods used byLehman & Aitken (1943) and by Oberst (1943) forpethidine in urine have not been evaluated for speci-ficity or modified to exclude completely interferingcontaminants. More recent findings reveal the pos-

sibility that these methods probably also measure a

pethidine metabolite, norpethidine.Methyl orange was found to be a satisfactory

indicator for measuring pethidine in biological fluidsby several groups of investigators (Way, Swanson& Gimble, 1947; Way, Gimble, McKelway, Ross,Sung & Ellsworth, 1949; Plotnikoff, Elliott & Way,1952; Burns, Berger, Lief, Wollack, Papper & Brodie,1955; Plotnikoff, Way& Elliott, 1956; Kazyak, 1959).Specificity was conferred on the method by subjectingthe alkaline solvent extracts from various tissues toa series of buffer washes to remove naturally occur-

ring metabolites and a pethidine metabolic productthat reacted with methyl orange (Way, Gimble,McKelway, Ross, Sung & Ellsworth, 1949; Burns,Berger, Lief, Wollack, Papper & Brodie, 1955;Kazyak, 1959) or by separation of methyl orange

reactants using counter-current distribution (Plot-nikoff, Elliott & Way, 1952; Plotnikoff, Way &Elliott, 1956).

Pethidine has also been determined by tracermethods. Plotnikoff, Elliott & Way (1952) usedlabelled pethidine, prepared with 14C in the N-methyl

position by Tarpey, Hauptman, Tolbert & Rapoport(1950), and measured the radioactivity directly fromashed tissues. Since the authors were primarilyinterested in finding clues to possible metabolic pro-ducts which might be revealed by tracing and ac-counting for all the radioactivity, no attempt wasmade to render the method specific.An ultraviolet method for estimating pethidine,

based on the absorbency of the drug at 257 m,p, hasbeen recently reported by Kazyak (1959). The pro-cedure appears adequate for forensic purposes whenhigh concentrations of the drug are present in tissues,a pethidine concentration of 1 mg/ml giving anoptical density reading of 0.885. Norpethidine givesa similar spectrum and a correction must be appliedalso for blank contribution.

Absorption

Absorption of pethidine may be concluded to berapid on the basis of the prompt onset of action ofthe drug. Peak plasma levels were obtained betweenthe first and the second hour after oral administrationof single doses of 150 mg in three patients (Burns,Berger, Lief, Wollack, Papper & Brodie, 1955).Attempts to quantify the rate of absorption ofpethidine have been made on the rat by Way,Gimble, McKelway, Ross, Sung & Ellsworth (1949).Pethidine appeared to be rapidly and almost com-pletely absorbed from the gastro-intestinal tractwhen administered intragastrically to fasted rats.Less than 20% of the compound was recovered inthe entire gastro-enteric tract one hour after adminis-tration of the drug and less than 10% after fourhours. Incubation of pethidine with the enteric tractgave no evidence that decomposition, bacterial orotherwise, was contributing to the rapid disap-pearance of the drug. The percentage recovery innon-fasted animals was somewhat greater. The pre-sence of food in the stomach appears to delaymarkedly the absorption of pethidine in rabbits,since non-fasted animals survived an oral dose ofthe drug that was usually lethal for fasted ones(Way, 1946). In the mouse rapid absorption as wellas metabolism of pethidine is indicated by the factthat 850% of an intraperitoneal dose of 75 mg/kgdisappeared within one hour (Bums, Berger, Lief,Wollack, Papper & Brodie, 1955).

Absorption of pethidine after subcutaneousadministration in rats begins almost immediately,but it is prolonged. In experiments with labelledpethidine, the site of injection (left hind leg) main-tained high levels of radioactivity (12.3 %) even

275


after 12 hours, whereas the uninjected right legcontained less than 1% of the total radioactivity(Plotnikoff, Elliott & Way, 1952).

Distribution

In man, following intravenous injection of 100-150 mg of the drug in ten subjects, plasma levelsdeclined relatively rapidly for the first 1-2 hours.After this interval a diffusion equilibrium betweenplasma and tissues was apparently attained, as indi-cated by the fact that when the logarithm of theconcentration was plotted against time, a straightline was obtained. The fall in plasma level averaged17% per hour, the range being 10-22% per hour.Ten hours after repeated intramuscular injections ofpethidine, the compound could not be detected inplasma, indicating that it did not accumulate to anydegree in the body. The decline in plasma levels infour cancer patients tolerant to pethidine was thesame as in non-tolerant individuals. As a consequenceof these studies, Burns, Berger, Lief, Wollack, Papper& Brodie (1955) concluded that increased capacityfor inactivation of pethidine is not a prime factorin the development of tolerance to the drug. Evi-dence that pethidine crosses the placental barrier inhumans was indicated by the detection of small

quantities of the compound in the urine of malenewborn infants whose mothers had previouslyreceived the drug (Way, Gimble, McKelway, Ross,Sung & Ellsworth, 1949) and in cord blood (Apgar,Burns, Brodie & Papper, 1952).High concentrations of pethidine and norpethidine

were found at autopsy in the tissues and fluids of a

male whose death was suspected of being causedby injection of an excessive dose of pethidine(Kazyak, 1959). The reported combined concentra-tion of the two compounds in the lungs was fantasti-cally high, being 340 ,ug/g. The liver had an appa-

rent pethidine concentration of 170 ,ug/g; thekidneys, 80 [ug/g, and the muscle, 70 ,tg/g. Thiswould mean that, in the muscle alone, a minimumof 1.96 grams of the drug was present, assumingthat muscle comprises 40% of the body-weight andthat the individual weighed 70 kg. Thus, ignoringthe appreciable amount present in other tissues as

well as any that was excreted, the dose received bythis individual would have been at least 40 ml ofthe commercially available preparation (50 mg/ml)for injection and could easily have been twice thisamount !

In the dog after administration of pethidine highlevels of the compound were found in the brain,

kidney, lung, liver, spleen and adrenals (Way,Gimble, McKelway, Ross, Sung & Ellsworth, 1949;Burns, Berger, Lief, Wollack, Papper & Brodie,1955). At plasma levels between 0.6 and 6.0 ,g/ml,approximately 40% of the pethidine was bound tothe non-diffusible constituents of plasma. The rateof decline of plasma levels is considerably slowerin the dog than in humans, being about 70% perhour after injection of 10 mg/kg intravenously(Burns, Berger, Lief, Wollack, Papper & Brodie,1955).

In rats very little pethidine could be detected inthe blood after intraperitoneal administration, butthe compound was found in high concentrationsin the kidneys, spleen and lung. Appreciable locali-zation also occurred in the brain and liver and, to alesser extent, in the heart and muscle. Evidencethat cumulative storage of the drug does not occurwas indicated by the fact that after nine successiveinjections of 50 mg/kg over a 3-day period, nopethidine was detectable on the fourth day (Way,Gimble, McKelway, Ross, Sung & Ellsworth, 1949).This was attributed to the rapid metabolic altera-tion of the compound by the liver (Way, Swanson &Gimble, 1947).

After subcutaneous administration of N-14CH3-labelled pethidine in the rat, significant amounts ofradioactivity were found in the liver, stomach, smallintestine, large intestine and faeces (Plotnikoff,Elliott & Way, 1952). It was concluded that theradioactive material in the gastro-intestinal tractconsisted chiefly of metabolite(s) of pethidine, sinceit had previously been shown that pethidine dis-appeared rapidly from the gastro-enteric tractafter oral administration. (Way, Gimble, McKel-way, Ross, Sung & Ellsworth, 1949).

MetabolismA schematic representation of the metabolic

pathways of pethidine in man is presented in Fig. 1.Pethidine is hydrolysed to pethidinic acid, which ispartially conjugated. In addition, pethidine isdemethylated to norpethidine, which is hydrolysedto norpethidinic acid and partially conjugated. Theidentity of the conjugate forms has not been establi-shed. Bound forms of pethidinic acid and norpethi-dinic acid have not been found in the urine of rats.

Pethidine and the five mentioned biotransforma-tion products have been identified by counter-cur-rent distribution studies on the urine of individualsreceiving pethidine. Pethidine and norpethidinewere extracted directly from alkalinized urine into

276


FIG. IMETABOLIC PATHWAYS OF PETHIDINE

H2 c NCH2 HYDROLYSIS

N

CHs

PETHIDINE

IJN-DEMET/YLATION

N\ //C0\CH1C14

H2C 'CH HYDROLYSIS >

MaC\ ,CH.N

H

NORPETHIDINE

C-OH

\C/H2C/ 'CH2

N

CH3

PETHIDINIC ACID

011\ C-OH

H2C 'ICH2H2C CHZ,-N2

N

H

NORPETHIDINIC ACID

CONJUGATIONJOJI"

[iC| 0-o1: c-OR

\11c/H2C lCH2

H2 \ /CH'

CH3

"BOUND" PETHIDINIC ACID

0

c-OR

CON/JUGATION10IC'

HIC CH2H2C CH2

N

"BOUND" NORPETHIDINIC ACID

benzene and identified by their respective partitionratios, using the methyl orange technique to measurethe fractionated basic amines (Plotnikoff, Elliott &Way, 1952). The presence of pethidinic and nor-pethidinic acid was established by demonstratingthat the concentrations of pethidine and norpethidinein urine increased after esterification with absoluteethanol in the presence of sulfuric acid. An evengreater recovery of pethidine and norpethidineresulted when the urinary constituents were sub-jected to acid hydrolysis before esterification,indicating the presence of bound forms of pethidinicand norpethidinic acid (Plotnikoff, Way & Elliott,1956). Norpethidine is the only biotransformationproduct which has been isolated in the crystallinestate. Its structure was identified by mixed melting-point determination and comparison of its infra-redspectrum with authentic norpethidine (Bums, Berger,Lief, Wollack, Papper & Brodie, 1955). It has alsobeen identified by paper chromatography as apethidine metabolite in dog urine (Vidic, 1957).Carbon dioxide and ethanol may be consideredpethidine metabolites, although the presence ofethanol has not actually been established (Plotnikoff,Elliott & Way, 1952).

It is of interest to consider the immediate source ofnorpethidinic acid, which could conceivably resulteither from hydrolysis of norpethidine or from N-demethylation of pethidinic acid. It was found thatafter administering a hydrolysed solution of pethi-dine-N-14CH3 to rats, no 14CO2 was present in theexpired air. No 14CO2 was produced by rat liverslices after incubation with a solution of hydrolysedN-l4CH3-pethidine, whereas under similar conditionswith N-14CHs-labelled pethidine 14CO2 was evolved(Plotnikoff, Elliott & Way, unpublished data).Moreover, it has been reported that an enzymesystem in liver microsomes demethylates pethidinebut not pethidinic acid (Gaudette & Brodie, 1959).Finally, after intravenous administration of pethi-dinic acid to humans, no norpethidinic acid wasfound in the urine, whereas norpethidine administra-tion resulted in large amounts of norpethidinic acidin the urine (Bums, Berger, Lief, Wollack, Papper& Brodie, 1955). Thus it appears that norpethidinicacid results primarily from hydrolysis of norpethidineand that pethidinic acid is not N-demethylated to anysignificant degree in the body.

In man N-demethylation appears to be an impor-tant pathway for the disposal of pethidine, since

277


about one-third of the compound can be accountedfor in the urine as N-demethylated derivatives.From animal studies it appears that norpethidinereadily penetrates the central nervous system (Way,Peng & Sung, unpublished data) and is more toxicthan pethidine (Miller & Anderson, 1954), althoughthe "analgesic " (tail-flick method) potency ofnorpethidine is less than that of pethidine (Miller& Anderson, 1954). This reduced activity of thenor-compound constitutes a strong objection toBeckett's hypothesis (Beckett, Casy & Harper, 1956)that N-demethylation is a prerequisite for analgesicactivity. A full discussion of this hypothesis will bepresented in the final instalment of this study.The amount of demethylation of pethidine occur-

ring in rats varies and appears to be dependent inpart on the route of administration (Plotnikoff,Elliott & Way, 1952). After intravenous injection ofpethidine-N-14CH3 in rats, approximately 43% of thetotal dosage was accounted for as expired '4CO2within 24 hours, over half of this amount beingexhaled within the first hour. After subcutaneousadministration, however, only 15% was recovered in24 hours and the 14CO2 was slowly excreted at afairly constant level for at least 10 hours. It wasconcluded that the metabolite norpethidine under-goes further degradation, since the amount ofdemethylated pethidine recovered was less than thetotal amount of '4CO2 expired.The metabolic changes indicated in Fig. 1 occur

chiefly in the liver. Bernheim & Bemheim (1945)have reported that pethidine is hydrolysed in vitro bythe liver homogenates of puppies, rabbits, rats,guinea-pigs, cats, turtles and frogs, but not by thebrain, blood, kidneys, spleen or heart. Manometrictechniques were employed, on the assumption thatpethidine would be metabolized to yield ethanol andl-methyl-4-phenylisonipecotic acid, and that thelatter compound would cause evolution of carbondioxide from a bicarbonate solution. (Oxidation ofthe ethanol was blocked by the addition of cyanide.)The hydrolysis of pethidine was found to be inhibitedby physostigmine and fluoride. It was postulatedthat the enzyme which hydrolysed pethidine wasdifferent from the known tropine esterases, cholin-esterase and the esterases which hydrolyse aliphaticesters.The above findings were extended by Way,

Swanson & Gimble (1947) with the liver from rats,dogs and man, using concentrations of pethidinecompatible with the expected levels to be found invivo. Pethidine was found to disappear rapidly when

added to liver homogenized in phosphate buffer.Other tissues exhibited little or no activity. In vivoevidence of the importance of the liver as the site ofbiotransformation is based on the finding that thepharmacological effect (potentiation of thiopentaldepression) and the blood levels of pethidine areincreased in rats after partial hepatectomy. Ratblood, but not dog or human blood, exhibited adelayed ability to metabolize pethidine to a slightdegree.The data indicate that the liver is also important

for demethylating pethidine. When N-14CH3-labelledpethidine was incubated with rat liver slices 14CO2was evolved (Plotnikoff, 1955). La Du, Gaudette,Trousof & Brodie (1955) reported that the methylgroup of pethidine was removed as formaldehyde.They also reported that the demethylation ofpethidine as well as of other alkylamines wascatalysed by an enzyme system present in livermicrosomes and that reduced triphosphopyridinenucleotide (TPNH) and oxygen were required.These findings were confirmed by Axelrod (1956a,1956b), who found that the rate of demethylationof pethidine by rat liver microsomal preparations,as measured by formaldehyde formation, wasmore rapid than that of morphine or i-methadone.Liver microsomal preparations from the rabbit werefound to be more active and those from the guinea-pig less active than rat liver microsomal prepara-tions, while microsomes from mice exhibited noactivity at all. After pretreatment with 3,4-benz-pyrene, the ability of rat microsomes to N-demethyl-ate pethidine was decreased in striking contrast tothe increase observed for other drug substrates(Conney, Gillette, Inscoe, Trams & Posner, 1959).A more detailed consideration of the mechanisms ofN-demethylation may be found in the first instal-ment of this study, which is devoted to morphine.

Irradiation of rats decreases the ability of theliver to metabolize pethidine as evidenced by studieson homogenized and perfused livers (Grossman &Chaloupka, 1959). According to the last-mentionedauthors, the irradiation of animals given pethidineresulted in higher blood levels, lower brain levelsand greater amounts in the urine than were observedin untreated controls.

Excretion

Very little pethidine (less than 10%) is excretedunchanged after administration. A major portionof the dose of pethidine can be accounted for asbiotransformation products, chiefly as hydrolysed

278


and demethylated derivatives in the urine and as CO2in the expired air. The gastro-enteric tract alsoappears to be important for excreting metabolites ofpethidine. Low concentrations of pethidine aredetectable in the saliva, but not in milk.

Urinary excretion studies on six individuals whoreceived therapeutic doses of pethidine orally orsubcutaneously have been reported by Lehman &Aitken (1943). The proportion of the total dosefound in the urine ranged from 5% to 20%. Eventhis low percentage is probably too high, since themethod used would presumably also measure nor-pethidine, a metabolic product which has beenfound to be present in urine (Plotnikoff, Elliott &Way, 1952; Burns, Berger, Lief, Wollack, Papper& Brodie, 1955; Plotnikoff, Way & Elliott, 1956).Oberst (1943) studied the urinary excretion of pethi-dine in ten former addicts, using a modificationof the method of Lehman & Aitken (1943); hence,the same criticism applies to his data. Oberst(1943) reported that after a total of 300-800 mgof pethidine administered daily in multiple doses,the average amount appearing in the urine was9.1%, with a range of 2.2-21.2%. About 72% ofthe total amount detected was present in the urineafter seven hours. After administration of single100-mg doses of pethidine, the urinary excretionof the drug after 24 hours was roughly of the sameorder as it was in those individuals who receivedthe larger dose; tolerance was not a factor influen-cing pethidine excretion.

In over twenty subjects only a very small percent-age of the administered dose of pethidine was foundto be excreted unchanged in the urine after oral orintramuscular administration of 100 mg (Way,Gimble, McKelway, Ross, Sung & Ellsworth, 1949).The range of values, measured by the methyl orangereaction, was 2-10%. Identification of the substanceas pethidine was made by demonstrating that thedistribution characteristics between two immisciblesolvents of the recovered material were identicalwith those of an authentic sample of pethidine.Subsequent reports (Burns, Berger, Lief, Wollack,Papper & Brodie, 1955; Plotnikoff, Elliott & Way,1952; Plotnikoff, Way & Elliott, 1956) generallysupport these studies. Less than 1 % of the dose ofpethidine administered to mothers during the firststage of delivery was found in the total 24-hoururine of male newborn infants (Way, Gimble,McKelway, Ross, Sung & Ellsworth, 1949).

In eight individuals given 100 mg of pethidineintramuscularly, low concentrations of the com-

pound, ranging from 3.5 to 6.0 mg/litre, were foundin the saliva one and/or two hours later. No pethi-dine (less than 0.5 mg/litre) could be demonstratedin the milk of six lactating subjects one to six hoursafter injection of 100 mg intramuscularly (Way,Gimble, McKelway, Ross, Sung & Ellsworth, 1949).

Burns, Berger, Lief, Wollack, Papper & Brodie(1955) reported that about 5% of a pethidine dosewas excreted unchanged, 5% as norpethidine, 12%as pethidinic acid and 12% as norpethidinic acid.They concluded that N-demethylation of pethidineto yield norpethidine appeared to be only a minorroute of metabolism. On the other hand, Plotnikoff,Way & Elliott (1956) were able to account for de-methylated pethidine products in the urine amountingto as much as 57 % of the dose administered andconcluded, therefore, that N-demethylation was animportant metabolic pathway for the disposal ofpethidine. Pethidine and its metabolites weredetermined on 24-hour urine specimens taken fromone normal volunteer and three hospitalized patientswho received a total of 200-1800 mg of pethidinedaily in divided doses for several days. The meanvalues, with their respective ranges, were found tobe: pethidine, 5 (2-8) %; norpethidine, 7 (2-15)%pethidinic acid, 22 (10-41) %; norpethidinic acid,11 (3-28) %; bound pethidinic acid, 5 (0-16%);bound norpethidinic acid, 12 (4-22) %. The averageover-all recovery for identifiable substances was53 (32-103)%.The 48-hour urine of dogs injected with 35 mg/kg

of pethidine contained roughly 5% of the dose aspethidine and/or norpethidine (Walkenstein, Mac-Mullen, Knehel & Seifter, 1958).

In two rats 50% of the injected dose of labelledpethidine could be accounted for in the urine asradioactive material within 24 hours after sub-cutaneous administration, but only a minor fractionof this amount was identified as unchanged pethi-dine. Significant amounts of radioactivity werefound in the faeces and the gastro-enteric tract,even after 24 hours, indicating the excretory roleof the latter (Plotnikoff, Elliott & Way, 1952).However, since it was previously established by themethyl orange method that pethidine disappearsrapidly from the gastro-enteric tract after oraladministration (Way, Gimble, McKelway, Ross,Sung & Ellsworth, 1949), it is unlikely that theradioactive material represents unaltered pethidine.The urinary excretion of pethidine was accountedfor as follows: unchanged pethidine, 2-7%; pethi-dinic acid, 4-12%; norpethidine, 9-35%. The total

279


recovery was 18-40%; bound forms of pethidineand norpethidine were not found (Plotnikoff, Way& Elliott, 1956).

ANILERIDINE

The compound anileridine, ethyl-1,4-aminophenyl-ethyl-4-phenylisonipecotate was synthesized by Weij-lard, Orahovats, Sullivan, Purdue, Heath & Pfisterin 1956 and its pharmacology was studied byOrahovats, Lehman & Chapin in 1957.

Method of estimation

Anileridine was determined in biological fluidscolorimetrically by diazotization and coupling withN(l-naphthyl)-ethylenediamine dihydrochloride byPorter in 1957. The compound was first extracted atpH 7-8 with ethylene dichloride and determinationswere then carried out on dried residues of the solventextracts. A detailed critique of the procedure was notreported, and the possibility that other aromaticamines arising from metabolic fragments may con-

tribute to a small degree to the readings should bekept in mind.

Absorption

Rapid absorption of anileridine is indicated by thefact that high levels of anileridine were found in thebrain within 15 minutes after subcutaneous adminis-tration of 8 mg/kg in the rat (Porter, 1957). " Anal-gesia" in the rat is maximal in 20-30 minutes aftersubcutaneous or oral administration. In the dogafter subcutaneous or oral administration the onsetof action is prompt, with peak effects occurringusually in under one hour (Orahovats, Lehman &Chapin, 1957).

Distribution

Following subcutaneous administration of anileri-dine in the rat, Porter (1957) found the compoundto be widely distributed in various tissues, beingespecially concentrated in the lungs, followed by thebrain, kidneys, liver and, finally, muscle. Peak con-

centrations occurred at about one hour after dosing.After 24 hours only the liver contained detectableamounts. The concentration of anileridine in thebrain was found to be a linear function of the sub-cutaneous dose given and could be correlated withthe degree of " analgesia ", as measured by the tail-flick method. After intravenous administration,

plasma levels were detectable only if the animal wasimmediately sacrificed. Maximum concentrations inthe brain were found 2-4 minutes after the injection.Thirty minutes later brain anileridine concentrationshad fallen to about 11 % of the peak values.A metabolite of anileridine, the N-acetyl deriva-

tive, appeared in tissues within an hour in high con-centrations; in fact, the concentrations generallyexceeded those of anileridine at the same time afterinjection. Relatively large concentrations of acetyl-anileridine appeared in the lungs, liver and kidneys,while in the brain the compound accumulated moreslowly and to a lesser extent. The acetyl derivativewas determined as anileridine after hydrolysis of theN-acetyl group under conditions which did noteffect hydrolysis of the ethyl ester grouping (Porter,1957).

MetabolismThe only study of metabolism reported is that of

Porter (1957), who, on the basis of studies conductedon the urine of the rat, the guinea-pig and man,proposed the following pathway of degradation (Fig.2). Anileridine is hydrolysed to anileridinic acid,which can undergo acetylation and be excreted asacetylanileridinic acid. Anileridine also conjugateswith acetic acid to yield acetylanileridine, part ofwhich is excreted and part de-esterified to yieldacetylanileridinic acid. Unfortunately, although thereport described a method for estimating the fourabove-mentioned compounds in biological fluids, noevaluation of the validity of the analytical procedurefor each compound and no data establishing theidentity of the anticipated metabolic products werepresented. Nor was evidence given to indicatewhether acetylanileridinic acid arises both fromacetylation of anileridinic acid and from de-esterifi-cation of acetylanileridine. While both pathwaysappear to be likely routes, it is quite possible thatone pathway predominates to the exclusion of theother. An analogous situation exists with respect tonorpethidinic acid formation discussed earlier, in thesection on the metabolism of pethidine.An additional metabolite was present as an un-

identified diazotizable substance found principally inthe free state in human urine and in a bound formthat is released by hydrolysis in rat and guinea-pigurine. Counter-current studies with this fractionfrom rat urine indicated that the substance might beacetylaminophenylacetic acid or a very closelyrelated compound, but the precise experimentaldetails of the isolation and characterization of the

280

BIOLOGICAL DISPOSMON OF MORPHINE AND ITS SURROGATES-3

FIG. 2KNOWN AND POSTULATED METABOLIC PATHWAYS OF ANILERIDINE

H2C CM2C"

CMH2

l H2 AZ

NH2

ANILERIDItNE

A

\/ "OH bC\ CH2C CM2 1

M2C' ,1CH2.N

CH2

NH2

NILERIDINIC ACID

AC

// 'O H

C O\CH2H2 L N,CH2N

CH2

CM2

NH-COCH3OETYLANILERIDINIC ACID

compound were not given. It was postulated thatthis metabolite was produced by degradation of theisonipecotic acid portion of the anileridine moleculerather than by scission in the nitrogen-ethylanilineportion. To the present reviewers this conclusion isdifficult to accept and the latter explanation appearsto be more logical since this would involve merelyN-dealkylation to norpethidine.

ExcretionIn two human subjects about 5% of an oral dose

of anileridine was excreted unchanged in the urine.Only a trace of the unchanged drug was found to beexcreted in the urine of rats and guinea-pigs, theamount accounted for being from 0.2% to 0.5% ofthe dose. All three species excreted 0.5-2.0% of thedose as acetylanileridine. Approximately 10-19%was identified in the urine of man, the rat and theguinea-pig as de-esterified derivatives, either freeanileridinic acid or acetylated anileridinic acid.Some species variation exists with respect to theexcretion of these two substances. In the rat and theguinea-pig only about 2% of the dose was excretedas free anileridinic acid, and about 17% was account-ed for as the conjugated anileridinic acid in the urine.

However, 7-14% of the dose of anileridine wasidentified as free anileridinic acid in the urine of twohuman subjects and only 1-2% as the conjugatedderivative. Thus, about 20% of the dose of anileri-dine can be accounted for in the urine as metabolicproducts of the parent substance. An additional15-35% can be accounted for as unidentified diazoti-zable substance(s), the nature of which has beendiscussed above under the metabolism of anileridine(Porter, 1957).

OTHER PHENYLPIPERIDINE DERIVATIVES

We have not encountered any quantitativestudies on the disposition of the following phenyl-piperidine derivatives: (a) alphaprodine (a-l,3-dimethyl-4-phenyl-4-propionoxypiperidine); (b)keto-bemidone (1-methyl-4(m-hydroxyphenyl)-4-piperidylethyl ketone); (c) properidine (1-methyl-4-phenyl-piperidine-4-carboxylic acid isopropyl ester); (d) tri-meperidine (1,2,5-trimethyl-4-phenyl-4-propionoxy-piperidine).

Limited studies have been made on the meta-bolism of ketobemidone (Breinlich, 1953a, 1953b).

281


On the basis of paper chromatographic studies thecompound has been reported to be N-demethylatedto its corresponding nor-derivative (Vidic, 1957).A method for estimating ketobemidone in urine,using bromocresol green, has been developed butno urinary excretion values have been reported(Vidic, 1953a, 1953c, 1955, 1957).

ETHOHEPTAZINE

Ethoheptazine (4-carbethoxy-1-methyl-4-phenyl-hexamethyleneimine) was synthesized by Diamond& Bruce in 1954. It differs structurally from pethi-dine only with respect to the heterocyclic ring,ethoheptazine having a seven-membered instead ofa six-membered ring. It is available commerciallyas the racemate, and reported to be a mild analgesicwithout addiction liability. Its pharmacology hasbeen studied by Glassmann & Seifter (1955). Thedescription which follows of the fate of ethohepta-zine is based on studies by Walkenstein, Mac-Mullen, Knehel & Seifter (1958).

Method of estimationEthoheptazine has been estimated by tracer

techniques, using a preparation labelled with 14Cat the 4-position. Carbon-14 assay of urine andtissue samples consisted in wet and dry combustion,respectively, followed by conversion of the labelledCO2 to BaCO3.The compound has also been measured by the

methyl orange procedure, using toluene as the solventand a borate buffer wash to remove interferingsubstances. Although no data were furnished, itwas indicated that the specificity of the method waschecked by infra-red, counter-current and chromato-graphic analysis of the extracts.

AbsorptionEthoheptazine appears to be rapidly absorbed.

In the dog a peak plasma level of 7.7 jtg/ml wasattained within one hour after intraperitonealadministration of 73.5 mg/kg of the base; plasmalevels were not detectable at 8 hours. In the ratorgan levels of radioactivity were higher at 30 min-utes than at 90 minutes after administration of thelabelled compound.

DistributionThe distribution of 14C following the intraperi-

toneal injection of an unspecified dose of labelled

ethoheptazine was studied in the rat. After 30 min-utes the liver showed the highest level of radio-activity and the kidney the next highest. At 90 min-utes the radioactivity of most organs droppedfrom one-third to one-half the values at 30 minutes.The activity in the intestine, however, increasedseveral-fold, indicating that this was an excretoryroute for ethoheptazine.

Metabolism

Ethoheptazine is metabolized by at least threeroutes. The pathways for the compound includehydrolysis to the corresponding acid, oxidation to ahydroxy derivative which may further undergohydrolysis, and possibly N-demethylation to thecorresponding nor-derivative, which may subse-quently be hydrolysed. The evidence was obtainedmainly from studies on the urine of dogs givenethoheptazine. Biotransformation of ethoheptazineby hydrolysis was established by identification of thecorresponding acid derivative by paper chromato-graphy, electrophoresis and infra-red absorptionstudies. Oxidation as a pathway was demonstratedby isolation of the hydroxy derivative and establish-ment of its identity by elemental and infra-redabsorption analyses. The latter studies also indicatedthat the hydroxyl group was present on the- hexa-methyleneimine ring, although the precise positionof the substitution was not determined. The evidencefor N-demethylation was only suggestive, being basedon a positive colour test for a secondary amine onurine paper chromatograms. Limited studies onrat urine and rabbit urine indicated the presence ofunchanged ethoheptazine and hydrolytic products.In addition, rabbit urine contained a secondary aminemetabolite as well as another unknown product notpresent in dog urine.

Excretion

Apparently very little ethoheptazine is excretedunchanged. In the dog the 24-hour urine accountedfor less than 4% of a 35 mg/kg intraperitoneal dose,and most of this amount appeared as unchangedethoheptazine in the urine in the first 4 hours afterinjection. The excretory pattern in the rat was quitesimilar to that in the dog; 2% of a 50 mg/kg intra-peritoneal dose was present in the 24-hour urine,the major fraction of this amount appearing within4 hours after administration. In the rabbit less than1.5% was recovered unchanged. At the same time,approximately 7-8% of the respective doses in the

282

BIOLOGICAL DISPOSITION OF MORPHINE AND iTS SuJRRoGATEs-3 283

dog and the rat, and less than 5% of the dose in therabbit, were recovered as de-esterified products,presumably the corresponding acids of ethohepta-zine and hydroxyethoheptazine. Nearly 9% wasaccounted for in the dog as a hydroxy derivativeof ethoheptazine.

The urinary excretion of the d- and i-forms ofethoheptazine was studied in the dog. Excretionof the unchanged ethoheptazine, hydroxyethohepta-zine and their hydrolysed residual products wasapproximately three times greater after injectionof the d-form than after injection of the i-form.

REFERENCES

Adler, T. K. & Eisenbrandt, L. L. (1949) Proc. Soc. exp.Biol. (N. Y.), 72, 347

Apgar, V., Bums, J. J., Brodie, B. B. & Papper, E. M.(1952) Amer. J. Obstet. Gynec., 64, 1368

Axelrod, J. (1956a) J. Pharmacol., 117, 322Axelrod, J. (1956b) Science, 124, 263Beckett, A. H. & Casy, A. F. (1954) J. Pharm. (Lond.),

6, 986Beckett, A. H., Casy, A. F. & Harper, N. J. (1956)

J. Pharm. (Lond.), 8, 874Benson, W. M., O'Gara, E. & Van Winkle, S. (1952)

J. Pharmacol., 106, 373Benson, W. M., Stefko, P. L. & Randall, L. 0. (1953)

J. Pharmacol., 109, 189Bernheim, F. & Bernheim, M. L. C. (1945) J. Pharmacol.,

85, 74Bockmuhl, M. & Ehrhart, G. (1949) Justus Liebigs Ann.

Chem., 561, 52Bonnycastle, D. D. & Delia, C. W. (1950) Proc. Soc.

exp. Biol. (N. Y.), 74, 589Breinlich, J. (1953a) Arzneimittel-Forsch., 3, 93Breinlich, J. (1953b) Arzneimittel-Forsch., 3, 212Brodie, B. B. & Udenfriend, S. (1945) J. biol. Chem.,

158, 705Brodie, B. B., Udenfriend, S. & Dill, W. (1947) J. biol.

Chem., 168, 335Brossi, A., Hafliger, 0. & Schnider, 0. (1955) Arzneimittel-

Forsch., 5, 62Bums, J. J., Berger, B. L., Lief, P. A., Wollack, A.,

Papper, E. M. & Brodie, B. B. (1955) J. Pharmacol.,114, 289

Conney, A. H., Gillette, J. R., Inscoe, J. K., Trams, E. R.& Posner, H. S. (1959) Science, 130, 1478

Costa, P. J. & Bonnycastle, D. D. (1955) J. Pharmacol.,113, 310

Cronheim, G. & Ware, P. A. (1948) J. Pharmacol., 92, 98Curry, A. S. & Powell, H. (1954) Nature (Lond.), 173,

1143Davis, M. E., Andres, G. J. & King, A. G. (1952) J. Amer.med. Ass., 148, 1193

Diamond, J. & Bruce, W. (1954) Substituted azacyclo-alkanes, US Pat. No. 2 666 050 (Summarized in Chem.Abstr., 1955, 49, 4031)

Eddy, N. B., May, E. L. & Mosettig, E. (1952) J. org.Chem., 17, 321

Eisenbrandt, L. L., Adler, T. K., Elliott H. W. &Abdou, I. A. (1949) J. Pharmacol., 98, 200

Eisleb, 0. & Schaumann, 0. (1939) Dtsch. med. Wschr.,65, 767

Elliott, H., W., Chang, F. N. H. Abdou, I. A. &Anderson, H. H. (1949) J. Pharmacol., 94, 494

Elliott, H. W. & Elison, C. (1959) Fed. Proc., 18 387Fisher, H. & Long, J. P. (1953) J. Pharmacol., 107, 241Fraser, H. F. & Isbell, H. (1952) J. Pharmacol., 105, 458Fromherz, K. (1951) Arch. int. Pharmacodyn., 85, 387Fromherz, K. & PeUlmont, B. (1952) Experientia (Basel),

8, 394Gaudette, L. E. & Brodie, B. B. (1959) Biochem.Pharmacol., 2, 89

Glassman, J. M. & Seifter, J. (1955) J. Pharmacol., 115, 21Grossman, V. & Chaloupka, Z. (1959) Arch. exp. Path.Pharmak., 236, 14

Gruber, C. M., jr (1957) J. Amer. med. Ass., 164, 966Gruber, C. M., jr, King, E. P., Best, M. M.,

Schieve, J. F., Elkus, F. & Zmolek, E. K. (1955) Arch.int. Pharmacodyn., 104, 156

Gruber, C. M., jr, Miller, C. L., Finneran, J. & Chemish,S. M. (1956) J. Pharmacol., 118, 280

Isbell, H. & Fraser, H. F. (1953) J. Pharmacol., 107, 524Jatzkewitz, H. (1954) Dtsch. med. Wschr., 79, 541Kaiser, H. & Jori, H. (1954) Arch. Pharm. (Berl.), 287,224

Kazyak, L. (1959) J. forensic Sci., 4, 264Keats, A. S. & Beecher, H. K. (1952) J. Pharmacol.,

105, 210LaDu, B. N., Gaudette, L., Trousof, N. & Brodie, B. B.

(1955) J. biol. Chem., 214, 741Lee, H. M., Scott, E. G. & Pohland, A. (1959) J. Phar-

macol., 125, 14Lehman, R. A. & Aitken, T. (1943) J. Lab. clin. Med.,

28, 787Mannering, G. J. & Schanker, L. S. (1958) J. Pharmacol.,

124, 296March, C. H., Gordan, G. S. & Way, E. Leong (1950)

Arch. int. Pharmacodyn., 83, 270Mellett, L. B. & Woods, L. A. (1959) J. Pharmacol.,

125, 97Miller, J. W. & Anderson, H. H. (1954) J. Pharmacol.,

112, 191

10

284 E. LEONG WAY & T. K. ADLER

Miller, J. W. & Elliott, H. W. (1955) J. Pharmacol.,113, 283

Oberst, F. W. (1943) J. Pharmacol., 79, 10Orahovats, P. D., Lehman, E. G. & Chapin, E. W.

(1957) J. Pharmacol., 119, 26Pellmont, B. & Bachtold, H. (1954) Schweiz. med. Wschr.,

84, 1368Pittinger, C. B., Gross, E. G. & Richardson, 0. (1955)

Fed. Proc., 14, 379Plotnikoff, N. P. (1955) J. Pharmacol., 113, 44Plotnikoff, N. P., Elliott, H. W. & Way, E. Leong (1952)

J. Pharmacol., 104, 377Plotnikoff, N. P., Way, E. Leong & Elliott, H. W.

(1956) J. Pharmacol., 117, 414Pohland, A. & Sullivan, H. R. (1953) J. Amer. chem.

Soc., 75, 4458Pohland, A. & Sullivan, H. R. (1955) J. Amer. chem.

Soc., 77, 3400Pohland, A., Sullivan, H. R. & McMahon, R. E. (1957)

J. Amer. chem. Soc., 79, 1442Porter, C. C. (1957) J. Pharmacol., 120, 447Reynolds, A. K. & Randall, L. 0. (1957) Morphine and

allied drugs, Toronto, University PressRickards, J. C., Boxer, G. E. & Smith, C. C. (1950)

J. Pharmacol., 98, 380Robbins, E. B. (1955) J. Amer. pharm. Ass., sci. Ed.,

44, 497Schaumann, 0. (1940) Arch. exp. Path. Pharmak., 196, 109Schaumann, 0. (1960a) Arch. exp. Path. Pharmak., 239,

311Schaumann, 0. (1960b) Arch. exp. Path. Pharmak., 239,

321Schnider, 0. & Gruessner, A. (1948) Helv. chim. acta,

32, 821Schnider, 0. & Gruessner, A. (1951) Helv. chim. acta,

34, 2211Schnider, 0. & Hellerbach, J. (1950) Helv. chim. acta,

33, 1437Scott, C. C. & Chen, K. K. (1946) J. Pharmacol., 87, 63Scott, C. C., Robbins, E. B. & Chen, K. K. (1948)

J. Pharmacol., 93, 282Shore, P. A., Axelrod, J., Hogben, C. A. M. & Brodie,

B. B. (1955) J. Pharmacol., 113, 192

Soehring, K. & Lohr, H. (1950) Pharmazie, 5, 569Sung, C.-Y. & Way, E. Leong (1950) J. Pharmacol.,

98, 72Sung, C.-Y. & Way, E. Leong (1953a) J. Pharmacol.,

108, 1Sung, C.-Y. & Way, E. Leong (1953b) J. Pharmacol.,

109, 244Sung, C.-Y. & Way, E. Leong (1954) J. Pharmacol.,

110, 260Sung, C.-Y., Way, E. Leong & Scott, K. G. (1953)

J. Pharmacol., 107, 12Takemori, A. E. & Mannering, G. J. (1958) J. Pharmacol.,

123, 171Tarpey, W., Hauptman, H., Tolbert, B. M. & Rapoport,H. (1950) J. Amer. chem. Soc., 72, 5126

Thorp, R. H. (1949) Brit. J. Pharmacol., 4, 98Tolbert, B. M., Christenson, F., Chang, F. N. H. &

Sah, P. P. T. (1949) J. org. Chem., 14, 525Vidic, E. (1951) Arch. exp. Path. Pharmak., 212, 339Vidic, E. (1953a) Arzneimittel-Forsch., 3, 34Vidic, E. (1953b) Arzneimittel-Forsch., 3, 428Vidic, E. (1953c) Arzneimittel-Forsch., 3, 490Vidic, E. (1955) Arzneimittel-Forsch., 5, 291Vidic, E. (1957) Arzneimittel-Forsch., 7, 314Walkenstein, S. S., MacMullen, J. A., Knehel, C. &

Seifter, J. (1958) J. Amer. pharm. Ass., sci. Ed., 47, 20Way, E. Leong (1946) J. Pharmacol., 87, 265Way, E. Leong, Gimble, A. I., McKelway, W. Ph,

Ross, H., Sung, C.-Y. & Ellsworth, H. (1949) J.Pharmacol., 96, 477

Way, E. Leong, Signorotti, B. T., March, C. H. & Peng,C. T. (1951) J. Pharmacol., 101, 249

Way, E. Leong, Sung, C.-Y. & Fujimoto, J. M. (1954)J. Pharmacol., 110, 51

Way, E. Leong, Sung, C.-Y. & McKelway, W. P. (1949)J. Pharmacol., 97, 222

Way, E. Leong, Swanson, R. & Gimble, A. I. (1947)J. Pharmacol., 91, 178

Weijlard, J., Orahovats, P. D., Sullivan, A. P., jr,Purdue, G., Heath, F. K. & Pfister, K. (1956) J. Amer.chem. Soc., 78, 2342

Winter, C. A. & Flataker, L. (1951) J. Pharmacol., 103, 93Woods, L. A., Mellett, L. B. & Andersen, K. A. (1958)

J. Pharmacol., 124, 1

biological disposition of morphine andits surrogates-3*

Documents