`
`524
`
`Deuterium isotope
`effects in studies of
`drug metabolism
`Allan B. Foster
`
`TIPS- December 1984
`
`the C-H bond with a consequent
`increase in bond stability. The ratio (kH/
`kD) of the rate constants for the protium
`and deuterium forms defines the primary
`observed isotope effect. The situation
`for enzyme reactions is complex in that
`such
`the maximal velocity of most
`reactions is dependent on several rate-
`rate-limiting
`contributing or partially
`steps. Thus, an enzyme reaction could
`involve the following sequence of events:
`
`Cancer Research Campaign Laboratory, Drug Development Section, Institute of Cancer Research,
`Clifton A venue, Sutton, Surrey SM2 5PX, UK.
`
`- EAB : *EAB =
`E + A; EA + B
`*EPQ,-- EPQ --* P + EQ--). Q +E
`
`Substitution of one or more carbon-bonded hydrogen atoms in a drug molecule
`with deuterium imposes negligible steric effects and influences physio-chemical
`properties minimally. Yet the resultant increased bond stability may cause dramatic
`changes in biological properties, particularly by retarding certain pathways of its
`metabolism. Allan Foster describes the potential of deuteration in modifying drug
`action. A specific biological action of a drug may be enhanced, reduced or
`prolonged; its metabolism may be diverted along a pathway which promotes
`formation of highly active metabolites, or avoids the formation of toxic metabolites.
`So why have deuterium isotope effects (DIEs) not been exploited? Cost of
`toxicology testing and clinical trials may underline the pharmaceutical industry's
`reluctance to synthesize deuterated analogues. Nonetheless, deuteration remains a
`promising possibility for future drug design.
`
`The majority of drugs and xenobiotics,
`following administration to, or inges-
`tion by, humans and animals, are
`metabolized, often rapidly and exten-
`sively'. Metabolism, which can occur in
`many organs of the body but takes
`liver, has been
`in the
`place mainly
`as a defence mechanism
`regarded
`are
`substances
`whereby exogenous
`converted into more polar derivatives
`that are excreted more rapidly than the
`parent compound. In the case of drugs,
`this defence mechanism can be coun-
`terproductive in that metabolism may
`limit plasma levels and half-lives and
`hence efficacy. Also, the usual, but not
`invariable, consequence of drug meta-
`bolism is deactivation, since metabolites
`usually have an affinity for the target
`lower
`(receptor, enzyme, membrane)
`than that of the parent drug, or may
`have properties which limit access to,
`the
`interaction with,
`therefore
`and
`target. Moreover, for some drugs, the
`metabolites have a biological activity
`different from that of the parent drug
`or may be toxic or carcinogenic.
`Control of the metabolism of a drug
`can give information on the mode of
`action and the role of metabolism in the
`In
`expression of biological activity.
`indicate
`this information could
`turn,
`the importance of metabolism control
`as a parameter in the design of more
`efficacious drugs. It is in this general
`context that deuterium isotope effects
`are now considered.
`
`Hydroxylation and the effect of
`deuteration
`Of the wide variety of drug meta-
`bolism pathways which have been
`identified, perhaps the most important
`are those mediated by the cytochrome
`P-450 group of enzymes 2 (mono-oxy-
`oxidases)
`function
`genases, mixed
`which are haem proteins commonly
`found as clusters of membrane-bound
`isoenzymes in the endoplasmic reticulum
`of many types of cell, and are particularly
`prevalent in liver cells. These enzymes
`utilize molecular oxygen and NADPH to
`effect, inter alia, the overall reaction:
`
`--L C-H -- " C-OH + H 20.
`
`For aromatic compounds the reaction
`usually involves the initial formation of
`re-
`an arene oxide and subsequent
`arrangement into a phenol. However, for
`aliphatic compounds and moieties, hydro-
`gen abstraction occurs first to give a
`radical -C- which is then hydroxylated.
`Deuterium isotope effects (DIEs) would
`type of
`this
`latter
`for
`be expected
`reaction.
`In a reaction in which the cleavage of a
`C-H bond is partially or wholly rate-
`then the reaction of the
`determining,
`corresponding C-D analogue will be
`retarded because the difference in mass
`between hydrogen and deuterium results
`in the zero-point energy (lowest ground
`level) for C-D being
`state vibrational
`1.2-1.5 kcal mole - ' lower than that of
`
`@ 1984, Elsevier Science Publishers BV.. Amsterdam
`
`0165.- 6147184'502A00
`
`where E is the enzyme, A and B are
`substrates or substrate and cofactor, P
`and Q are the products, and * connotes
`an activated complex. Northrop 3 has
`described a family of DIEs associated
`with various segments of this sequence.
`That associated with the *EAB --- *EPQ
`segment is the intrinsic DIE. (Dk) which
`is analogous to the DIEs for chemical
`reactions. Although the magnitude of Dk
`it will usually be
`may be large (10-15)
`masked to an extent depending on the
`character of the preceding and succeeding
`steps in the overall sequence shown
`above, and the observed DIE. (Dv =
`kH/kD) may be quite low (1-5); Dvvalues
`in metabolism studies are usually deter-
`mined from the relative rates (kH/kD) of
`the protium and
`disappearance of
`deuterium forms, or the relative rates of
`appearance of the corresponding meta-
`bolites. Thus, Lu et al. 4 found a Dk value
`the de-O-ethylation of
`for
`12
`of
`7-ethoxycoumain (Fig. 1. [i]) and its dz-
`derivative (Fig. 1. [ii], by purified cyto-
`chrome P-450 (phenobarbital induction)
`and P-448 (methylcholanthrene induction)
`from rat liver microsomes, but Dv values
`of approximately 3.8 and 2, respectively.
`The replacement of one, or a few
`hydrogens in an aliphatic moiety of a
`drug molecule by deuterium will have
`negligible steric consequences or influence
`on physico-chemical properties and
`hence the deployment of DIEs in drug
`metabolism studies is attractive. How-
`ever, it should be emphasized that it is
`the magnitude of the observed DIE (DV)
`which will determine the extent to which
`is
`a particular metabolism pathway
`retarded and the consequent effect on
`biological activity.
`There are numerous uses5 of deuter-
`ium-labelling in drug metabolism studies
`other than for DIEs, and mass spectro-
`metry is a key technique when deuter-
`ium-labelled compounds are used. For
`in metabolism
`the change
`example,
`profile caused by specific deuteration of
`a drug can be readily detected and
`
`Apotex Ex. 1023
`
`Apotex v. Auspex
`IPR2021-01507
`
`
`
`quantified by mass spectrometry after
`subjecting an equimolar mixture of
`protium and deuterium forms to in-vitro
`or in-vivo metabolism.
`
`Modification of the biological activity of
`drugs by deuterium isotope effects
`One of the earliest attempts to modify
`the biological properties of a drug by
`specific deuteration and thereby eluci-
`date the mode of action was described by
`Belleau et al.6; dideuteration of the
`adrenergic p-tyramine
`(Fig. 1.
`[iii])
`doubled the duration of the effect on
`arterial pressure and nictitating mem-
`brane contractions following i.v. admin-
`istration to the cat. Subsequently, a DIE
`of approximately 2.4 was found for the
`oxidative deamination of 82-p-tyramine
`(Fig. 1. [iii]) with monoamine oxidase
`from rat liver mitochondria7.
`Barbiturates were also an early subject
`of study. Thus, whereas trideuteration at
`position 4 of the butyl group of butethal
`(Fig. 1. [iv]) had no effect on biological
`activity, dideuteration at position 3
`doubled the sleep time in mice 8 because
`3-hydroxylation of the butyl group,
`which is a major metabolism pathway,
`was thereby retarded. A Dv of approx-
`imately 1.6 was found in in-vitro meta-
`bolism experiments.
`The foregoing examples, which showed
`that relatively small DIEs could modify
`the in-vivo properties of a drug, illustrate
`the enhancement or prolongation of
`drug activity by specific deuteration. The
`opposite effect, namely reduction in the
`duration of drug action has also been
`reported 9. N-Desmethyldiazepam (Fig.
`1.
`[vi])
`is the major metabolite of
`diazepam (Fig. 1. [v]) and is further
`metabolized by 3-hydroxylation to give
`oxepam (Fig. 1. [viii]) which accum-
`ulates in the brains of diazepam-treated
`mice and is responsible for the prolonged
`anticonvulsant action of
`the drug.
`Dideuteration at position 3 (Fig. 1. [vii])
`reduced the duration of the anticon-
`vulsant action from 20-->5 hours and in-
`vitro experiments using mouse
`liver
`microsomes revealed an approximately
`7.5-fold reduction in the extent of 3-
`hydroxylation.
`The use of specific deuteration in an
`attempt to retard the formation of toxic
`products of metabolism can be illustrated
`with
`the anaesthetic methoxyflurane
`(CH 3OCF2CHC12) which is metabolized
`by two pathways involving hydroxylation
`of the methyl and dichloromethyl groups.
`The former pathway gives CHC12COOH
`and releases fluoride ion which can cause
`renal dysfunction. McCarty et al. '0 found
`a 33% decrease in the urinary excretion
`of fluoride for perdeuterated methoxy-
`
`TIPS - December 1984
`
`HO-a
`
`CHCDNH,
`
`I R = CH 2CH 3
`II R = CD 2 CH 3
`
`III
`
`0
`
`4
`3
`/CHCH 2 CHzCH 3
`
`HN
`
`Et
`
`H I
`
`V
`
`V R 1 = Me, R2= H
`VI R1 = R2 = H
`VII R1 = R2 = D
`
`HN
`
`N >
`H
`
`XIV
`
`S
`
`0 N
`
`N
`
`S
`
`H
`
`HN
`H
`
`>
`
`H
`
`XIIl
`
`0
`
`eBu
`
`S-u
`
`HN
`R
`
`R 2
`
`N
`H
`R2 = H
`R
`R = H, R2= D
`R= R2 = D
`D R2 = H
`R-
`
`XV R = CH 3
`XVI R = CD 3
`
`CH 2
`XVII
`
`* CH,
`
`CH 3
`
`N N'
`IP h
`
`IXX
`
`XX
`
`Fig. 1. Structures of drugs the metabolism of which is influenced by deuterium labelling.
`
`Apotex Ex. 1023
`
`
`
`TIPS - December 1984
`
`Almost a decade
`later the situation
`has not changed. For drugs iot intended
`for use in humans or for products such
`as insecticides, the advantages to be
`gained by specific or general deuter-
`ation in modifying biological activity
`and/or duration of action must signi-
`ficantly outweigh the additional costs
`associated with the synthesis of deuter-
`ated analogues. For drugs intended for
`use in humans there will be a substantial
`additional cost, namely, that associated
`with preclinical toxicology and clinical
`trials. It seems very unlikely that the
`regulatory authorities associated with
`the pharmaceutical industry would regard
`a deuterated drug designed to have a
`biological activity significantly different
`from that of the parent protium form as
`other than a new drug.
`The examples noted above, which
`represent a not unreasonable cross-
`section, show that the in-vivo biological
`activity of a drug can be modified as a
`consequence of altering the metabolism
`profile through a DIE but none of the
`in-vivo effects reported in the literature
`to date can be regarded as
`truly
`dramatic. However, the end of the story
`may not yet have been reached. Lu et
`al.4 reported recently that, whereas the
`DIE for the de-O-ethylation of d2-7-
`ethoxycoumarin (Fig. 1. [ii]) by liver
`microsomes from phenobarbital-treated
`rats was -3,
`a much larger value (-6)
`was found for human liver microsomes.
`This, apparently, is the first report of
`the use of human liver microsomes in a
`study of DIEs in drug metabolism and is
`of potential importance for evaluating
`the real scope for using a DIE to
`influence metabolism, since DIEs asso-
`ciated with animal liver microsomes,
`apart from possible species differences,
`may be misleadingly small. Further,
`since microsomes are fragments of the
`endoplasmic reticulum, usually of liver
`cells, the metabolism they mediate may
`be quantitatively,
`if not qualitatively,
`different from that which occurs
`in
`intact hepatocytes. The use of animal,
`especially rodent, hepatocytes in drug
`metabolism studies is well established
`it has now been shown 16
`and
`that
`human hepatocytes, when co-cultured
`with rat epithelial cells, retain their full
`metabolizing capability for many hours
`and can be deployed in drug metabolism
`studies. It will be of interest to see what
`use can be made of this new type of
`metabolizing
`system
`for evaluating
`DIEs and metabolic switching as para-
`meters in drug design.
`The phenomenon of metabolic switch-
`ing is likely to be of growing interest in
`the future. The example noted above
`
`flurane (CD 3OCF2CDCI2), but the serum
`fluoride levels and renal dysfunction after
`anaesthesia of rats for 2 h were still
`unacceptable" 1.
`The occurrence of a DIE in an in-
`vitro system does not necessarily mean
`that biological properties will be altered
`in vivo. A major metabolic detoxific-
`ation pathway for the antitumour agent
`6-mercaptopurine (Fig. 1. [ix]) is via 8-
`hydroxy-6-thiopurine (Fig. 1. [xiii]) to
`thiouric acid (Fig. 1. [xiv]) thought to be
`mediated by xanthine oxidase. A signi-
`ficant DIE (3.5) was found for the
`action of this enzyme in vitro on 8-dl-6-
`mercaptopurine (Fig. 1. [x]) but not for
`the 2-dl-derivative (Fig. 1. [xii]) suggest-
`ing (see above) a different balance of
`rate-limiting and/or partially rate-deter-
`mining steps 12. A DIE of 3.8 was found
`for the 2,8-d 2-6-mercaptopurine (Fig. 1.
`[xi]). These DIEs were not
`fully
`reflected in in-vivo experiments. Thus,
`following i.p. administration of the 2,8-
`d2-derivative to rats, 2.2-3.7 times as
`much unchanged drug was excreted in
`the urine and 54-70% of thiouric acid in
`comparison with
`6-mercaptopurine
`(Fig. 1. [ix]) reflecting retardation of the
`detoxification metabolism pathway
`in
`the dideuterated compound. However,
`although the potency of the 2,8-d2 -6-
`mercaptopurine (Fig. 1. [xi]) against the
`adenocarcinoma Ca755
`in mice was
`increased 3-5-fold, the 8-dl-derivative
`(Fig. 1. [x]), which also had a significant
`DIE in vitro, had the same potency as 6-
`mercaptopurine.
`Most in-vivo studies of DIEs in drug
`metabolism have involved analysis of
`urine for metabolites and unchanged
`drug. An alternative approach was used
`recently 13
`in
`a study of butylated
`hydroxytoluene (BHT; Fig. 1. [xv]).
`This compound, which is a widely used
`antioxidant, causes lung damage in mice
`and the covalent binding in lung tissue is
`
`probably mediated by
`reactive
`the
`quinone methide metabolite (BHT-QM;
`Fig. 1. [xvii]). Trideuteration of the
`methyl group in BHT (Fig. 1. [xvi])
`reduced the pulmonary
`toxicity and
`markedly reduced the level of quinone
`methide in lung tissue thereby reflecting
`the DIE observed in in-vitro experi-
`ments.
`
`Metabolic switching
`When a drug is metabolized by two or
`more alternative pathways, a possible
`consequence of deuteration is metabolic
`switching, i.e. suppression of one path-
`way and accentuation of an alternative
`pathway. The term was introduced by
`Horning et al. 4 who found that the
`metabolism of antipyrine (Fig. 1. [xviii]),
`after i.p. injection
`into rats, and as
`reflected by the urinary metabolites,
`was switched from oxidation of the C-3
`methyl group (Fig. 1. [xix]; normal
`major pathway)
`to de-N-methylation
`(normal minor pathway) on trideuter-
`ation of the former group. An even
`more marked effect was observed in
`vitro. Using the 10 000 g supernatant of
`homogenized rat liver, the ratio of the
`products of hydroxylation of the C-3
`methyl group (Fig. 1. [xix]) and position
`4 (Fig. 1. [xx]) was only slightly changed
`(1.3-1.6) when the N-methyl group was
`trideuterated, but dramatically (to <
`0.1) when the C-3-methyl group was
`trideuterated.
`Numerous other examples of meta-
`bolic switching of cytochrome P-450-
`mediated reactions as a result of DIEs
`have been
`reported
`(although not
`always recognized) in the past decade.
`
`Conclusions and future trends
`In their excellent 1975 review, Blake
`et al. 15 commented, 'At the present time,
`there are no drugs on the market that
`contain deuterium in the molecule ...
`
`Apotex Ex. 1023
`
`
`
`TIPS - December 1984
`
`reflects metabolic switching caused by a
`DIE, but it is quite possible that other
`types of structural change could be
`used. The deliberate deployment of
`metabolic switching as a parameter in
`drug design remains to be explored.
`Two promising areas
`involve
`the
`accentuation of the metabolism path-
`ways which generate the active meta-
`bolites from prodrugs and, where over-
`all metabolism cannot be retarded, the
`deflection of metabolism away from
`pathways leading to metabolites with
`toxic properties or other undesirable
`biological activity towards
`innocuous
`pathways.
`
`Reading list
`1 Testa, B. and Jenner, P. (1976) Drug Meta-
`bolism: Chemical and Biochemical Aspects,
`Marcel Dekker, New York
`2 Schenkman, J. B. and Stutzman, L. A. (eds)
`(1982) Hepatic Cytochrome P450 Mono-oxy-
`genase System, Pergamon Press, Oxford
`3 Northrop, D. B. (1982) Methods Enzymol. 87,
`
`607-641
`4 Lu, A. Y. H., Harada, N. and Miwa, G. T.
`(1984) Xenobiotica 14, 19-26
`5 Haskins, N. 1. (1982) Biomed. Mass Spectrom.
`9, 269-277
`6 Belleau, B., Burba, J.,
`PindeUl, M. and
`Reiffenstein. (1961) Science 133, 102-104
`7 Yu, P. H., Barclay, S., Davis, B. Boulton,
`A. A. (1981) Biochem. Pharmacol. 30, 3083-
`3094
`8 Tanabe, M., Yasuda, D. M., LeValley, S. and
`Mitoma, C. (1969) Life Sci. 8, 1123-1128
`9 Marcucci, F., Mussinei, E., Martelli, P.,
`Guaitani, A. and Garratini, S. (1973) J. Pharm.
`Sci. 62, 1900-1902
`10 McCarty, L. P., Malek, R. S. and Larsen,
`E. R. (1979) Anesthesiology 51, 106-110
`11 Baden, J. M., Rice, S. A. and Mazze, R. I.
`(1982) Anesthesiology 56, 203-206
`12 Jarman, M., Kiburis, J. H., Elion, G. B.,
`Knick, V. C., Lambe. G., Nelson, D. J. and
`Tuttle, R. L.
`(1982)
`in Stable Isotopes
`(Schmidt, H. L., F6rstel, H. and Heinzinger,
`K., eds), pp. 217-222, Elsevier, Amsterdam
`13 Mizutani, T., Yamamoto, K. and Tajima, K.
`(1983) Toxicol. Appl. Pharmacol. 69, 283-290
`14 Horning, M. G., Haegele, K. D., Sommer,
`K. R., Nowlin, J., Stafford, M. and Thenot,
`
`J. P. (1976) in Proceedings of the Second Inter-
`national Conference on Stable Isotopes (Klein,
`E. R. and Klein, P. D., eds), pp. 41-54,
`NTIS, Springfield, Virginia
`15 Blake, M. I., Crespi, 14. L. and Katz, J. J.
`(1975) J. Pharm. Sci. 64, 367-391
`16 Begue, J. M., Le Bigot, J. F., Guguen-
`Guillouzo, C., Kiechel, J. R. and Guillouzo,
`A. (1983) Biochem. Pharmacol. 32,1643-1646
`
`Allan B. Foster, D.Sc. graduated B.Sc., Ph.D. in
`Chemistry at the University of Birmingham, UK
`and then did several years of post-doctoral work
`including a year (1953-54) as a Fellow of the
`Rockefeller Foundation with Melville Wolfrom at
`the Ohio State University. In 1955 he became a
`member of faculty in the Department of Chemistry
`of the University of Birmingham with research
`interests in carbohydrate chemistry. In 1966 he was
`appointed Professor of Chemistry at the Institute of
`Cancer Research (University of London) where his
`research interests turned to studies of the meta-
`bolism and mode of action of anticancer drugs. He
`is presently involved with the design, development,
`and evaluation of antiendocrine-type anticancer
`drugs.
`
`Transynaptic
`mechanisms in the
`action of
`antidepressant drugs
`Giorgio Racagni and Nicoletta Brunello
`
`Institute of Pharmacology and Pharmacognosy, University of Milan, Via Andrea del Sarto, 21, 20129
`Milan, Italy.
`
`Antidepressant drugs act on different neuronal systems and pre- and postsynaptic
`sites. Integrated transynaptic events are considered to be involved in those
`adaptive changes which seem to be operative after a prolonged administration.
`The authors explain how these long-term effects, rather than the acute pharmaco-
`logical actions, are most likely to represent the biochemical mechanism underlying
`the delayed onset of antidepressant therapeutic efficacy. Among the possible
`mechanisms responsible for the adaptation of central aminergic neurons,
`interactions between serotonergic and noradrenergic systems, chemico-physical
`properties of the membranes and the modulatory actions of hormones and
`cotransmitters are considered.
`
`ergic systems which can occur both pre-
`and post-synaptically. The findings
`reported below summarize the most
`significant data to appear in the litera-
`ture in the last few years. (For detailed
`information see Refs 1-4.)
`Radioligand binding studies have
`revealed that chronic administration of
`antidepressants produces changes in
`sensitivity of presynaptic ct-adreno-
`ceptors and a decrease in the number
`[3H]imipramine
`([ 3H]IMI)
`of
`sites
`located
`presynaptically
`on
`sero-
`tonergic
`(5HT)
`terminals. Central
`noradrenergic transmission is reduced
`by chronic antidepressants as indicated
`by (1) a decrease in the activity of tyro-
`sine hydroxylase; (2) a reduction in the
`firing rate of noradrenergic cell bodies
`in the locus coerules; (3) lower cortical
`levels of normetanephrine (NMN), the
`O-methylated metabolite or norepine-
`phrine
`(NE);
`(4)
`a
`reduction
`in
`3-methoxy-4-hydroxy-phenylethylene-
`glycol
`(MHPG) content. However
`increase or no major changes have
`been also reported on this NE metab-
`olite. This apparent discrepancy may
`be due to several factors such as the
`different methodology used
`in
`the
`measurement of NE
`turnover,
`the
`administration of antidepressants with
`multiple sites of action, non compara-
`ble time of treatment and washout
`period, different brain areas investi-
`gated.
`Effects elicited by long-term anti-
`depressant administration at the post-
`synaptic level include changes in the
`©K 1964. Elsevier Science Pulshr B V , Amster~dam 0165 -(6147/84/$02,05)
`
`It is now generally accepted that the
`mechanism of action of antidepressant
`drugs cannot be attributed
`to their
`acute pharmacological actions since
`agents possessing many of these effects
`lack antidepressant activity. Anti-
`depressant therapy is associated with a
`lag phase of 15-20 days before the
`onset of a beneficial activity. Thus, in
`order to elucidate the mode of action
`
`of antidepressants, only those effects
`elicited after
`long-term
`treatment
`should be considered.
`
`Pre- and postsynaptic biochemical
`modifications induced by long-term
`antidepressant treatment
`Chronic antidepressant administra-
`tion is associated with a number of
`adaptive changes in central monoamin-
`
`Apotex Ex. 1023
`
`