Historical aspects

With the onset of research into the carcinogenicity of pure chemicals, the benz[a]anthracene (BA) nucleus was regarded, in these early studies, as a basic requirement for the demonstration of carcinogenicity. Kennaway (1930) had observed that the application of a pure polycyclic aromatic hydrocarbon (PAH) to mice produces tumours. At the same time Hieger (1930) was investigating the fluorescence characteristics of particular fractions obtained from coal tar which possessed carcinogenic activity. The fluorescence spectra of these fractions were similar to the fluorescence spectrum of BA. These observations led investigators to propose that the ring system of BA, a compound which is only weakly carcinogenic in itself, provides the basis for a potentially carcinogenic molecule and that cancer-producing properties are developed by substitution at suitable positions (Barry et al., 1935).

Initially, because of the strong carcinogenic activity of 3-methylcholanthrene (3-MC) and dibenz[a,h]anthracene, substituents at positions 8 and 9 on the BA nucleus were regarded as particularly favourable for carcinogenicity (see figure). The chemical synthesis of a large number of compounds based on BA, which varied in the position, number and type of substituents, was undertaken in England by Cook and his associates at the Royal Free Hospital, London, and in the USA by Fieser and co-workers in Harvard and Newman at Ohio State University. The biological activities of these compounds were examined under the supervision of Kennaway in London and Shear in the USA.

Reference titles contain original nomenclature.

1. Abell, C. W. & Heidelberger, C. (1962). Binding of tritium-labelled hydrocarbons to the soluble proteins of mouse skin. Cancer Res., 22, 931–46.

2. Ahlquist, K. A. (1966). Enzyme changes in rat testis produced by the administration of busulphan and of 7,12-dimethylbenz[a]anthracene. J. Reprod. Fertile12,377–9.

3. Allen, J. A. & Coombs, M. M. (1980). Covalent binding of polycyclic aromatic compounds to mitochondrial and nuclear DNA. Nature, 287,244–5.

4. Ames, B. N., Sims, P. & Grover, P. L. (1972). Epoxides of carcinogenic polycyclic hydrocarbons are frameshift mutagens. Science, 176,47–9.

5. Andervont, H. B. & Shimkin, M. B. (1940). Biologic testing of carcinogens. II. Pulmonary-tumour-induction technique. J. Nat. Cancer Inst., 1,225–39.

6. Arcos, J. C. & Argus, M. F. (1968). Molecular geometry and carcinogenic activity of aromatic compounds. Adv. Cancer Res., 11, 305–471.

7. Armstrong, B. & Doll, R. (1975). Environmental factors and cancer incidence mortality in different countries, with special reference to dietary practices. Int. J. Cancer, 15, 617–31.

8. Aylsworth, C. F., Jone, C, Trosko, J. E., Meites, J. & Welsch, C. W. (1984). Promotion of 7,12-dimethylbenz[a]anthracene-induced mammary tumorigenesis by high dietary fat in the rat: possible role of intercellular communication. J. Nat. Cancer Inst., 72, 637–45.

9. Bachmann, W. E., Kennaway, E. L. & Kennaway, N. M. (1938). The rapid production of tumours by two new hydrocarbons. Yale J. Biol. & Med., 11,97–102.

Introduction

The transformation of a normal cell into a malignant cell, after exposure to a chemical carcinogen, is the result of a complex series of biochemical events. It is generally believed that the initial step involves the reaction of an electrophile (formed from the carcinogen either directly or by metabolism) with cellular macromolecules such as DNA, RNA and protein. Which of these cellular macromolecules represents the critical target for the chemical carcinogen in the initiation of the carcinogenic process has been the subject of much theoretical discussion. The somatic mutation theory proposes that some permanent inheritable change in the nucleotide sequence occurs as the result of an alteration, deletion or rearrangement of the primary structure of DNA. Alternatively, epigenetic changes in cellular transcription and translation may bring about malignant cell transformation; such epigenetic changes also occur in normal development and differentiation (see Rubin, 1980; Marquardt, 1979a; and Barrett and Ts'o, 1978, for a fuller discussion of these hypotheses). The weight of evidence currently favours the somatic mutation theory, and underlies the widespread acceptance of the use of short-term mutation assays, such as the Ames Test, to indicate compounds with potential carcinogenic activity (many chemical carcinogens are mutagens). The DNA adducts formed in vivo and in vitro after attack by PAHs have been, in general, well characterized in terms of their chemical structure: it is often known which bases are involved, whereabouts on the base the PAH has bound and what type of functional group on the PAH is responsible for such binding.

Introduction

‘There is only one unambiguous experimental method to determine cardnogenicity; it is to ascertain if your compound produces cancer in animals.’ (Quoted from Huggins, 1979.) Cancer induction is a complex process, still not completely understood, and the experimental demonstration of a chemical's carcinogenic properties is dependent on a number of variables such as the dose administered, the species, age and sex of the animal employed, the mode of administration, the use of cocarcinogens or promotors, the time period of experimental observation and ensuring that sufficient numbers of animals survive the experiment for adequate statistical analyses to be performed. Conflicting data are occasionally reported from laboratories in which different procedures are employed. Also difficulties arise when attempts are made to compare the relative potency of various chemicals as carcinogens. Iball (1939) proposed that there are two measurable parameters involved in determining the potency of carcinogenic compounds; one is the percentage incidence of tumours, the other is the average latent period for the appearance of tumours. Giving equal weight to both these factors, Iball constructed an index of carcinogenic potency for various chemical carcinogens; 7,12-DMBA was by far the most potent chemical tested. An additional indicator of carcinogenic potential, which is often reported from skin-painting experiments, is the number of tumours per animal. Comprehensive tables on the carcinogenicity of benz[a]anthracene and its derivatives have been reported by Dipple et al. (1984a).

Introduction

Polycyclic aromatic hydrocarbons are metabolized by the mixed function oxidase system to forms more readily excretable by the cell; usually by making the PAH more polar and therefore more water soluble. The principal enzymes involved in this process are those consisting of the haem-containing cytochrome P-450 (which requires both NADPH and molecular oxygen) and epoxide hydrolase. These enzymes are located mainly as membrane-bound proteins in the endoplasmic reticulum or nuclei of cells although some cytosolic forms are known. They exist in ‘multiple forms’ some of which have been isolated and purified. These isoenzymes, which are present in a variety of tissues, can be selectively induced by treatment of the animal with various chemicals (commonly used cytochrome P-450 inducers are 3-MC and phenobarbital), and these various induced states can result in different metabolic pathways for the compounds under study.

Cytochrome P-450 introduces a hydroxyl group into an alkyl group, or an epoxide group across the double bond of an aromatic system. Epoxide hydrolase can then metabolize the epoxide further to a trans-dihydrodiol. If this trans-dihydrodiol contains an adjacent double bond then cytochrome P-450 may act for a second time to form a diol epoxide. Although these events are designed as detoxification routes current evidence suggests that the formation of a diol epoxide, particularly a bay-region diol epoxide, are in fact steps of metabolic activation.

Friedel–Crafts reactions

The four carbons needed to convert phenanthrenes into BAs have been introduced by succinoylation, which takes place at the 3-position of phenanthrene and at the 2-position of 9,10-dihydrophenan threne, Scheme XXI. With succinic anhydride 153 was produced and with α-methylsuccinic anhydride 156 was formed. Reduction, usually by the Clemmensen route afforded the acids, 154 and 157, which on ring closure yielded the 8-keto-8,9,10,11-tetrahydroBAs 155 (Haworth and Mavin, 1933), and 158 (Cook and Haslewood, 1934), respectively. Although ring closures to ketones 155 and 158 are most often effected by conversion of acids to acid chlorides followed by AlCl3 or SnCl4 promoted cyclizations, the use of trifluoromethanesulphonic acid anhydride (Hulin and Koreeda, 1984), and methanesulphonic acid (Premasagar et al., 1981), for such closures has also been reported. By reduction of 155 and aromatization, BA was obtained. Similarly 158 yielded 9-MBA and by reaction of 158 with methylmagnesium iodide followed by aromatization, 8,9-DMBA was produced. The latter hydrocarbon was of interest in establishing the structure of methylcholanthrene (Cook and Haslewood, 1934). The condensation of succinic anhydride with 3-methylphenanthrene afforded β[6-(3-methylphenanthroyl)]-propionic acid, 159, from which 2-MBA was synthesized by conventional steps (Bachmann and Cortes, 1943). Similar reaction of succinic anhydride with 9,10-dihydrophenanthrene yielded β-(9,10-dihydro-2-phenanthroyl)propionic acid, 160, which was converted into BA by reduction of the keto group, cyclization to 11-keto-5,6,8,9,10,11-hexahydroBA, 161, further reduction to 5,6,8,9,10,11-hexahydroBA and aromatization by heating with selenium (Burger and Mosettig, 1937).

The numbers at the beginning of each Reference refer to citations in the Tables.

1. Adapa, S. R., Sheikh, Y. M., Hart, R. W. & Witiak, D. T. (1980). Preparation of site specifically deuterated 7,12-dimethylbenz[a]anthracene derivatives: mechanism of hydrogenolysis of aryl halides with lithium aluminium hydride. J. Org. Chem., 45, 3343–4.

2. Agranat, I., Rabinovitz, M., Selig, H. & Lin, C. (1977). Fluorination capabilities of xenon fluoride/graphite intercalates: introduction of fluorine into carcinogenic polycyclic aromatic hydrocarbons. Synthesis, 267–8.

3. Ahmed, Z. & Cava, M. P. (1981). A novel anthraquinone annelation. A new approach to alkavinones. Tetrahedron Lett., 5239–42.

4. Ahmed, F. U., Rangarajan, T., Eisenbraun, E. J., Keen, G. W. & Hamming, M. C. (1975). The synthesis of BA. Org. Prep. Proc. Int., 7,267–70.

5. Anet, F. A. L. & Bavin, P. M. G. (1960). Studies in the Wagner–Meerwein rearrangement. IV. Derivatives of benz[b]fluorene. Can. J. Chem.,38, 240–3.

6. Awad, S. B., Sakla, A. B., Abdul-Malik, N. F. & Ishak, N. (1979). Cycloaddition of some quinones to 1,1-diarylethylenes. Indian J. Chem., 17B, 219–21.

7. Azerbaev, I. N. (1945). Hydrogenation of 1-ethynylcyclohexanol and 1-ethynylcyclohexene. Synthesis of 1-vinylcyclohexene. J. Gen. Chem. U.S.S.R., 15,412—20. Chem. Abst., 40,4683.

8. Babayan, V. O., Zagorets, P. A. & Tatevosyan, G. T. (1953). Synthesis of hydrocarbons of the BA Series. Zh. Obshch. Khim., 23,1214–20. Chem. Abst., 47,12214d.

9. Bachmann, W. E. (1936).

Additions to 1,4-naphthoquinones

1-Vinylcyclohexene, 195, has been reacted with 1,4-naphtho-quinone, 196, to give 1,2,3,4,6,6a,12a,12b-octahydroBAQ, 197, in high yield (Backer and Bij, 1943; Azerbaev, 1945), Scheme XXIV. On air oxidation 197 was converted into 1,2,3,4-tetrahydroBAQ, 198 (Fieser and Hershberg, 1937b), which was reduced to 1,2,3,4-tetrahydroBA. With the exception of two other examples (Inbasekaran et al., 1980; Carothers and Coffman, 1932), no further work with the use of vinylcyclohexenes in BA synthesis has been reported.

Rather, the addition of styrenes, 199, to 196, has been studied and well developed. When the reactions were carried out in refluxing toluene containing chloranil and trichloroacetic acid, the following nine BAQs, 200, were obtained:4-Br-, 1-, 2-, 3-, 4-Cl-, 4-F-, 4-methoxy-, 2-methyl-, and 1,4-dimethyl- (Manning et al., 1977). Subsequently, 3,4-dimethoxy- BAQ, 201, was synthesized (Manning and Wilbur, 1980) by a similar route. When 5-hydroxy-1,4-naphthoquinone, 202, and 5-methoxy-1,4- naphthoquinone, 203, were used, mixtures of 8-hydroxy- and 11-hydroxyBAQs, 204 and 206, and 8-methoxy- and 11-methoxyBAQs, 205 and 207, resulted (Manning, 1979). By the use of 6-hydroxy- and 6-methoxy-1,4-naphthoquinones, 208 and 209, there were obtained mixtures of 9-hydroxy- and 10-hydroxyBAQs, 210 and 212, and 9-methoxy- and 10-methoxyBAQs, 211 and 213, respectively (Manning, 1979; Manning et al., 1979; Manning, 1981). The conversion of many of the BAQs into BAs and 7,12-DMBAs by conventional methods has been described (Muschik et al., 1979). However, difficulties were encountered in the synthesis of BAs bearing the 1-methoxy group because of steric factors.

In Part 1 of this book the syntheses of benz[a]anthracenes are covered from the middle 1930s to the end of 1984. Prior to the middle 1930s almost all of the references are to quinones of possible use as dyestuff intermediates and are not covered here. Early syntheses were reviewed by Cook (1931). The synthetic procedures of interest in early studies of the carcinogenic activity of polycyclic aromatic hydrocarbons were reviewed by Fieser (1937). The present state of the synthetic approaches to benz[a]anthracenes is organized into several different categories. Advantages and difficulties in various synthetic routes are pointed out. Often the poor yields in certain steps should be capable of improvement by the use of alternate reagents and methods. Since many of the older methods have not been restudied, it is left to the reader to decide the best course of action in any chosen case.

The syntheses of arene oxides, dihydrodiols, quinones, phenols and diphenols in the benz[a]anthracene series will not be covered in depth and the reader is referred to the monograph Polycyclic Hydrocarbons and Carcinogenesis by Harvey (1985), which addresses these topics more fully.

In Part 2 the biological properties of benz[a]anthracene and its substituted derivatives (such as the highly active 7,12-dimethylbenz[a]anthracene) are discussed in relation to literature available up to the end of 1985.