CHAPTER 1
Tracers for Metabolism
BY I. P. SWORD
1 Introduction
This chapter is a review of organic synthetic methods using (mostly) low-energy [beta-]emitting radio-isotopes, and it deals largely with labelled compounds which have been synthesized for the elucidation of their metabolism in mammals. Many organic radiochemicals are synthesized for other purposes, but most of the labelled intermediates involved are potentially useful for metabolically oriented syntheses, and reference is made to them where appropriate. Some mention is also made of work with stable isotopes and, in passing, to reported improvements in some established radiochemical syntheses. Biologically mediated radiochemical syntheses have been excluded, since they have been reviewed earlier elsewhere, and are outside the scope of this review.
The literature coverage is for 1970 and 1971, although some information from 1969 is included, if it only became accessible (e.g., in Chemical Abstracts) in the following year. Although this chapter is not intended to be encyclopaedic, it is hoped that no major representative syntheses have been excluded. The reactions reported range from the relatively trivial to the complex.
It is encouraging to note that the editorial board of the Journal of Labelled Compounds intends to re-introduce a quarterly literature survey on publications involving work with isotopes, and to publish abstracts of all such work since 1966, when the service ended.
Investigations of the absorption, distribution, excretion, and biotransformation of foreign compounds are facilitated by the use of a radio-labelled form of the compound under study (see Vol. 1, p.34). Consequently, the link between radiochemistry and metabolic studies is well established. In fact, the metabolic fate of a compound is often accounted for in terms of radioactivity; (for limitations of the radiochemical method, see ref. 4).
General aspects of radiochemical synthesis, radiochemical purity, and the philosophy of using radiolabels in pharmacological and metabolic studies have been reviewed (see Vol. 1, p.36), as has the use of radio-isotopically labelled analytical reagents up to the end of 1969.
2 General Considerations
Choice of Nuclide and its Molecular Location. — A majority of metabolic studies are made with 14C as tracer. The relative cheapness of tritium, (brought about by military research) and the ease of its incorporation into molecules for study sometimes outweigh the disadvantages conferred by its potential lability under physiological conditions. When the economics of tritium labelling are compared with the alternative, often protracted 14C synthesis, the former is sometimes selected. However, caution must be exercised in the preparation and equilibration of tritium-labelled material before use, and in the interpretation of results, if useful information is to result.
Choice of molecular location for the tracer atom(s) primarily concerns syntheses with 14C, and to a lesser extent those with 3H. The site for labelling must be chosen to take the biological stability of the labelled part of the molecule and the cost of synthesis into consideration. A knowledge of the biotransformations of related molecules in mammals is essential to a proper consideration of which labelled form would give definitive information. Generally, only limited metabolic data are available from, for example, N-[14C]methyl or O-[14C]acetyl labelling, since in biological systems these groups may be dissociated from the main part of the molecule. Labelling in rings is preferable but normally expensive with 14C; even this has limitations as to its usefulness. The separate use of two or more labelled forms of the same material with the label in different molecular locations, or the incorporation of different radioactive nuclides in the same molecule constitute useful, but more expensive, methods for the more detailed monitoring of metabolites.
Sulphur-35 and phosphorus-32 have been used to some extent, but obviously give metabolic information only about molecular fragments containing these elements: there is normally no choice of molecular site for their incorporation. The amount of work with other nuclides in metabolic studies is relatively small.
Techniques of Syntheses with Isotopes. — The majority of reported organic syntheses with isotopes are undertaken on millimolar or larger quantities of material. For the synthetic chemist unfamiliar with techniques for handling small quantities of potentially hazardous material (often in vacuum-manifold systems), a number of reference texts are available. Probably the most useful general handbooks on the subject are those by Murray and Williams, Catch, and Evans. The series of reviews published in pamphlet form by the Radiochemical Centre, Amersham is indispensable. A review in Russian is also avai1able. Generally, a good radiochemical (or stable isotope) synthesis is a combination of a minimum number of high-yield synthetic steps involving the expensive labelled materials, and a minimum number of transfer and isolation procedures. Thus, the preparation of tracers for metabolism falls within the broad aims of organic synthesis, and some notable syntheses have been accomplished. Guidelines for the practical worker approaching radiochemical synthesis for the first time have been reported and the logistics of establishing and running a radio-synthesis unit discussed.
Specific Activity. — Tritium labelling has the advantage over, for example, 14C in that high specific activities are attainable at moderate cost, and this is particularly favourable for detection purposes when the mammalian dose of material under investigation is very low (e.g. with steroidal oral contraceptives). Specific activities with 32P and 35S generally exceed those attainable with 14C, which in turn normally exceed those attainable with 36Cl. One millicurie of product with a specific activity of 1 — 10 mCi mmol-l appears to be a convenient quantity for a normal metabolic study, but transport studies require material of high specific activity.
Radiochemical Purity and Autoradio1ysis. — Meaningful metabolic data can only be obtained from materials that are radiochemically pure, the criteria of radiochemical purity being quite distinct from those of chemical purity. However, in many of the papers reviewed in this chapter, no indication of the radiochemical purity or of the systems used to monitor it are given.
High specific activity materials and radiochemicals which have been stored for extended periods are particularly susceptible to radiation-induced decomposition (autoradiolysis), a phenomenon of radio-labelled compounds that cannot be completely overcome, although the degradative effects can be minimized by appropriate precautions.
Labelled Precursors. — Many [14C]syntheses have started from barium [14C]carbonate, but alternatively, commercially available radiochemicals of some complexity have been incorporated into an increasing number of syntheses. The number and variety of these intermediates are increasing steadily, but labelled forms of some common 'building bricks' for drugs are still unavailable.
Catch and others have listed the primary [14C]reagents available directly from barium [14C]carbonate for further elaboration into complex molecules.
The commonest forms of tritium used for incorporation into substrates are the gas itself, tritium oxide, and the mixed-metal tritide reducing agents (LiAl3H4, etc.).
The primary labelling procedures with tritium are:
(i) Homogeneous catalytic exchange,
(ii) Heterogeneous catalytic exchange,
(iii) Radiation-induced (Wilzbach) exchange,
(iv) Addition to centres of unsaturation,
(v) Substitution by reduction (e.g. R — X -> R — 3H).
As indicated in the sequel, some more sophisticated methods with specialized applications have been reported in the literature. As with 14C, the number of commercially available reagents and finished products containing 3H, 32P, 35S, and the commoner stable isotopes continues to grow steadily.
Nomenclature. — The system recommended by the Chemical Society for naming labelled compounds has been adopted here as far as possible, but many systems of nomenclature are presently used in the literature, making an overwhelming case for standardization. The possibility of distribution of the label(s) within a molecule engenders its own sub-nomenclature (used by commercial suppliers of labelled materials): an indication of popular usage of this sub-nomenclature is given:
Specifically labelled: the isotope location is defined in the full chemical name of the compound. If more than one position is labelled, uniform distribution of the label among the defined sites is implied.
Uniformly labelled (U): the label is evenly distributed throughout the molecule as a result of the mode of formation, e.g. [U-14C]benzene from [1,2-14C]acetylene.
Generally labelled (G): a random undetermined label distribution (most often applies to products of non-regioselective tritium-exchange procedures).
Nominally labelled (N): the label is ostensibly located at the position indicated, but there is a finite chance that it also occurs elsewhere.
Arrangement of Material. — Since many compounds are labelled with more than one isotope, the classification of syntheses according to the nuclide used is impractical. In consequence, outlines of syntheses involving 3H, 14C, 32P, 35S, and 36Cl have been combined and arranged according to whether the final product is acyclic, alicyclic, aromatic, heterocyclic, or polyalicyclic. Not all of the compounds fall readily into a particular group, and in these cases the author's personal preference has been exercised. The few syntheses involving other radioactive nuclides are reported separately, followed in turn by work of general applicability, and then by work with stable isotopes.
In chemical formulae an asterisk (*)has been used to denote the site of isotope incorporation.
3 Syntheses with 3H, 14C, 32P, 35S, and 36Cl
Acetohydroxamic acid, an agent which inhibits urease activity in vitro, and which may decrease blood ammonia concentration in cases of hepatic coma, has been prepared labelled form by treatment of hydroxylamine with [1-14C]- or [3H]-acetic anhydride.
Although not intermediates to products for eventual metabolic evaluation, some useful 14C-labelled intermediates, viz. 2-chloro [l-14C]ethanol, 2-cyano[l-14C]ethanol, and [3-14C]acrylonitrile, have been synthesized. The reaction sequence is indicated in Scheme 1.
A radio-synthesis of the important industrial solvent O-isopropylethanediol has been achieved by heating [14C]ethylene oxide and propan-2-ol in a sealed tube with boron trifluoride as catalyst.
The potential anti-tumour agent methylene dimethanesulphonate (1) has been prepared, labelled separately with 35S and with 14C (Scheme 2). The classical reaction of [1,3-14C]acetone gave [14C]iodoform, which was reduced with sodium arsenite to di-iodo [14C]methane, which with silver methanesulphonate in acetonitrile gave the product. The same reaction with methane [35S]sulphonate gave the corresponding 35S-labelled product. A 14C-labelled form of the insecticide dimethoate (2), OO-dimethyl- S(N]14C]methylcarbamoylmethyl)phosphorodithioate, has been prepared in low (3.9%) yield. It appears that [Me-14C]isocyanate prepared from [2-14C]acetyl chloride was added to dimethoate carboxylic acid in the presence of triethylamine. In an alternative procedure, with improved yield, [14C]methylamine and dimethoate carboxylic acid were heated in a sealed tube to give the product directly.
[14C]Thiourea has been incorporated into creatinol O-phosphate (3), according to an outline report, which contains only data on purification of the material, and not on its preparation (see Scheme 3).
[14C]Methylamine, [14C]methanol, and [32P]phosphorus trichloride have been used as starting materials in the preparation of several labelled combinations of the insecticides 3-hydroxy-NN-dimethyl-cis-crotonamide dimethyl phosphate (4) and 3-hydroxy-N-methyl-cis-crotonamide dimethyl phosphate (5). The reaction sequences are indicated (Scheme 4); the bioactive cis-crotonamides were separated by liquid-liquid partition chromatography. The same author prepared dichlorvos (2,2-dichlorovinyl dimethyl phosphate) (6) labelled with 14C, 32P, and 35Cl in four molecular locations. Trimethyl phosphite, the previous sequence, labelled with 32P or 14C, and prepared as in was condensed with chloral in the presence of potassium to afford appropriately labelled dichlorvos. 36Cl-Labelled chloral for incorporation into the product was prepared by an interesting procedure involving exchange between Li 36Cl and unlabelled chloral in tetramethylene sulphone. [l-14C]Chloral was obtained by chlorination of [l-14C]ethanol.
Condensation of carbon [35S]disulphide and cysteine in ammonia gave the radioprotectant triammonium 2-[35S]dithiocarbamyl-3-[35S]dithiocarbonylthiopropanoate (7). 1-α-Methylallylthiocarbamoyl-2-methylthiocarbamoylhydrazine (8), an inhibitor of pituitary gonadotrophic function, has been formed from potassium [14C]thiocyanate and 1-chloro-2-butene through thermal isomerization to the methylallyl isothiocyanate (9) followed by reaction with 4-methylthiosemicarbazide (Scheme 5).
The carcinogen, 3,3-dimethyl-l-phenyltriazine, PhN=N·NMe2 (l0), was synthesized from [14C]dimethylamine and diazotized aniline. U.v. spectrometry was the only criterion of purity recorded. The metabolic fate of 3-chloromethylhept-1-yn-3-carbamate (1l), tritiated in the chloromethyl group, has been reported by Jones et al. Although no detailed synthetic data are given, the preparative route is outlined as shown (Scheme 6).
For absorption studies, the phospholipid (12) has been prepared in three separately labelled forms incorporating 3H, 14C, and 32P. The obvious advantages of triple labelling by admixture of three separately labelled components are discussed; specially relevant in this study is the possibility of replacing that component labelled with the relatively short-lived 32P after several half-lives. The metabolic fates of phenyldimethyl- and phenyltrimethyl-silane have been studied using 14C-labelled material, derived from [14C]methyl iodide by the synthetic route indicated below (Scheme 7).
Radiocarbon- and tritium-labelled forms of Cardison, DL-N-methyl-N-propargyl-1-phenyl-2-aminopropane hydrochloride (13), have been prepared, and their syntheses described in detail. A key intermediate, DL- N-methylphenylisopropylamine (14), was obtained by reductive condensation of phenylacetone and methylamine in the presence of sodium borohydride. Phenyl[2-14C]acetone and phenyl [3-14C]acetone were obtained from the appropriately labelled ethyl acetates by the method indicated (Scheme 8). Reaction of the isopropylamine (14) with propargyl bromide in alkali afforded the product. High specific activity tritium-labelled forms (l-3H and 2-3H) were prepared respectively by catalytic dehalogenation of DL-chloropseudoephedrine hydrochloride (15) in a tritium atmosphere and by using sodium borotritide in the sequence parallel to that described for the radiocarbon synthesis. Reduction of the Schiff base (16) with tritium and Adams' catalyst gave material having lower specific activity.
The rigidity and tremor controller orphenadrine citrate, NN-dimethyl-2-(o-methyl-α- phenylbenzy1oxy)ethylamine citrate (17), labelled specifically with tritium (as shown), has been reported by Ellison et al. Reduction of o-methylbenzophenone with sodium borotri tide gave o-methylbenz [3H]hydrol, which with thionyl chloride gave the corresponding chloro-compound. This with dimethylaminoethanol yielded labelled orphenadrine: the N-demethyl and NN-didemethyl analogues were also prepared.
The analgesic agent (2S,2R)-4-dimethylamino-1,2-diphenyl-3-methyl-2propionoxybutane (18) (d-propoxyphene) has been synthesized with 14C in the benzyi residue by the sequence shown (Scheme 9). Optical resolution was effected after the Grignard reaction, using d-10-camphor sulphonate. The plant phenol chlorogenic acid (19), which is a constituent of tobacco, has been prepared labelled (Scheme 10) with 14C and starting from [2-14C]malonic acid. Condensation of the latter with 3,4-dihydroxybenzaldehyde gave [α-14C]caffeic acid, which, after protection of the phenolic hydroxygroups as carbonates, was converted into the acid chloride (20) for reaction with the acetonide (21). Mild hydrolysis then gave the desired material (19).
Two 14C-labelled forms of the spasmolytic agent l-benzyl-l-(3'-dimethylanlinopropoxy) cycloheptane fumarate (Halidorj (22) have been synthesized. 39 Reaction of the N-normethyl derivative of (22) (as the free base) with [14C]methyl iodide gave N-[14C]methyl-labelled product, whereas the other labelled form was prepared from [7-14C]benzyl chloride and cycloheptanone via the Grignard reaction as indicated (Scheme 11).
