Document 3JJzj0EYG33knm599zo1QJqGn

FILE NAME: Shell (SHEL) DATE: 1945 July 2 DOC#: SHEL003 DOCUMENT DESCRIPTION: Unpublished Internal ReportCarcinogenic Hydrocarbons and Related Compounds FOR INFORMATION ONLY - NOT FOR PUBLICATION A CONTRIBUTION OF INFORMATION TO m B E R S OF THE A.P.I. L A I C A L ADVISORY C O m T T E E SHELL OIL COMPANY, INCORPORATED WOOD RIVER RESEARCH LABORATORIES REPORT HO. M-12^7 JULY 2, 19^5 SUBJECT; CARCINOGENIC HYDROCARBONS AND RELATED COMPOUNDS' A LITERATURE REVIEW AUTHOR: H. H. ZUH2EMA Exportation of This Report is Subject to License Under the National Defense Act SHELL OIL COMPANY, INCORPORATED WOOD RIVER RESEARCH LABORATORIES REPORT NO. M-12^7 JULYJ^J^ SUBJECT: CARCINOGENIC HYDROCARBONS AND RELATED COMPOUNDS A LITERATURE REVIEW AUTHOR: H. H. ZUIDEMA Industrial cancer was first recognized in England in the latter part of the eighteenth century, when it was established that chimney sweeps were particularly liable to cancer of the scrotum.^ This was caused by soot. Progress in this field of study was very slow at first, and it was not until 1915 that two Japanese investi gators announced the first case of carcinoma in an experimental animal. They produced first papillomas and then true cancer of the epidermis on the ears of rabbits by painting with tar over long periods of time. 35 During the thirty years that have elapsed since that discovery, a great deal of work has been done with various coal tar and petroleum fractions and pure compounds either occurring in these fractions or prepared synthetically.' Numerous species of test animals, including mice, rats, fowl, rabbits, and dogs have boon used, and many methods of application, Including oral administration, intramuscular, intravenous, and subcutaneous injection, and painting on the skin. The method that appears to be the most widely used at present consists of painting a solution on the skin of a special in bred strain of mice. Mice are preferred over animals having longer . life-spans since they respond more rapidly. The special strains are used to increase the precision of the test. Painting is preferred to other methods of application because there is less likelihood of interference by simultaneous spontaneous carcinomata. The carcino genicity of a given compound is influenced by many factors including the genetic consiitution of the animal species and strain, its age and sex, the diet, the physical condition of the animal, the purity of the compound, the dose, the physical state of the compound, the solvent used, and the route or site of application.25 While carcinogenic properties are generally associated with certain polynuclear aromatics and their derivatives, there are many substances entirely unrelated to these compounds which have been re ported as having similar cancer-producing ability. Among these may be mentioned asbestos314>37 , inorganic compounds of arsenic nnJ of zinc2 5, aqueous potassium hydroxide and aqueous hydrochloric acid25, ethyl alcohol25, glucose25, fructose25, and a number of nitrogen ' compounds, most of which contain ono or more benzene rings25. Radio active elements and compounds are not included in this discussion since the mechanism of their producing cancer is probably quito different, i.e., physical rathor than chemical. Materials which have been roported Report No. M-12^7 2 as carcinogenic and whose carcinogenicity is probably due to the presence of polynuclear aromatics include coal tar and pitch26'27 blue shale oil5, mineral oils and asphalt'*0, tobacco tar1^ , tars obtained in the destructive distillation of teaU2and coffee1*0, diesel fuel52, distillates from Borneo crude petroleum51*, combustion gases from fuel oil1*1, the SOp extract from a spindle oil produced from a paraffinic crude petroleum.51*, the products obtained by heating acetylene, isoprene, yeast, a non-carcinogenic petroleum fraction, or human shin to 700 - 900C-3o, and the products obtained by treating acetylene, xylene, naphthalene, or tetrahydronaphthaleno "with aluminum chloride . Ihe carcinogenicity of tars produced by heating various substances tends to increase with increasing temperature. -30, I t is probable that tho carcinogenic hydrocarbons present in certain coal tar and petroleum fractions were to a large extent formed during the process steps and were not present in the coal or crude petroleum from which they were derived. However, the carcinogenicity of a given fraction depends upon the source as well as the method of processing. According to the Manchester Committee on Cancer1*9, the following order of increasing carcinogenicity prevails: Russian, Pennsylvania, Texas, Mid-Continent, Mexican, California, Persia, Romanian, Borneo, Venezuela and shale. Brie corresponds, roughly to the order of increasing aromati city. A large number of compounds have been tested for carcino genicity. Hartwell has published a survey29 which includes 696 compounds tested. Of these, lh6 were carcinogenic, and 23 additional ones produced papillomas but no true cancers. While some of these compounds have been tested under only one or two sets of conditions, others have been studied very extensively. The three compounds that have received tho moat detailed study are the throe derivatives of 1,2-bonzanthracene: 1,2,5,6-dibenzanthacene, 3,^-benzpyreno, and 20- methylcholanthreno. All throe aro potent carcinogens. It has been estimated that they have been investigated in 60 different labora tories. 20 A list of most of the known carcinogenic compounds is given in Table l,pagos 51Q. Structures, complete with numbering systems, are included, since lack of uniformity among various in vestigators in this field has led to some confusion. There is fortunately less deviation in the current papers than in the case of those written ten to fifteen years ago. The numbering used in this report corresponds to that used by Hartwell25, it should bo understood that the carcinogenicity of the compounds shown in Table 1 varies considerably in degree and that the results have been verified by several laboratories in some cases whilo in others tho results of only one or two experiments are available. The reader Is referred to Hartwell for more detailed information and for re ference to the original experiments. It will be observed that most of the compounds are eithor hydrocarbons or their derivatives containing at least three benzene rings or nitrogen compounds con taining at least two bonzeno rings. More specifically, the first of these classes consists, with only threb exceptions, of derivatives B e p o rt Bfo^ M-12^7 3 of phenanthrene. The exceptions are tetraphenylmethane, triphenylbenzene, and triphenylethylene. In Table 1 the hydrocarbons are subclassLfied further as derivatives of 1 ,2-benzanthracene, 3 ,k-benzpyrene, cholanthrene, and other derivatives of phenanthrene. Examination of the structures shows that all of these are derivatives of phenanthrene. While the work that has been done on the production of cancer by exposure to various compounds has made it possible to draw certain conclusions regarding the effect of chemical structure, it is not as yet possible to predict with any degree of certainty the carcinogenicity of a given compound from its structure. A phenanthrene derivative may or may not, for example, be carcinogenic depending upon such factors as the length of an alkyl side chain, the position of a benzene ring, or the position of an alkyl group. However, a paraffinic or naphthenic hydrocarbon or one containing only one or two benzene rings may with a reasonable degree of confidence bo assumed to be non-carcinogenic. The amount of compound required to produce an experimental cancer on a test animal is surprisingly small. Doses of 1,2,5,6-di benzanthracene and 3,^-benzpyrene as small as 2.5 and k micrograms, respectively, give-positive results when applied to mice.1(* 5Dlscontinuou3 application, e.g., once a week, of a given total quantity of material is moro effective than continuous exposure.15 A single application of 20mothylcholanthrono in 0.6$ benzono solution is capable of producing cancer on the skin of mice. U5 When different carcinogenic hydrocarbons, e.g., methylcholanthrene, benzpyrene, and dibenzanthracene, are applied successively the effect is additive 3J. The investigators who made this observation offer the hypothesis that carcinogenisis is essentially an accumulation of abnormal protein within the cell. Since so many compounds have been tested for carcinogenicity it is only natural that several attempts have been made to correlate this property with other properties, either physical or chemical. Such a correlation would be extremely useful in that the determination of the physical or chemical property would be much simpler than the rather long and tedious process of exposing experimental animals to the substance in question. Eluorescence spectra have received more study in this regard than any other property.5 .7 .23.^The author of a recent paper made the statement that all known cariinogonic chemicals are fluorescent. 5 However, it is obvious from a glance at Table 1 or Hartwell s tabulation that this is too broad a claim. However, many investigators have used fluorescence spectra to good advantage in this field, and there appears to be a definite correlation with carcinogenicity,- at least with certain types of compounds. The correlation may be due in part to the fact that fluorescence spectra have been used as a guide in selecting fractions from coal tar fop physiological tests; it is conceivable that other carcinogenic com pounds, which are not fluorescent, exist or could be synthesized. A- rough parallelism between the carcinogenicity of poly nuclear aromatics and the diazo coupling reaction has been observed.2 1 .22 Attempts have also been made to correlate refractive index or more specifically, specific refraction, with the carcinogenicity of oils. Eeport No. M-1247 h The Manchester Committee on Cancer119 recommended that spindle oils of various ranges of specific gravity have specific refraction values below cortain limits to assure the absence of carcinogenic properties. However, it is obvious that thore can be no general correlation between a physical property of this nature and carcinogenicity, for slight changes in structure, such as the position of a methyl group, which would have little effect upon optical properties, affect carcinogenic properties profoundly) furthermore the presence of a large concentra tion of a non-carcinogenic aromatic would raise the specific refraction of a mixture much more than would the presence of a small amount of a potent carcinogen. The probability of obtaining a general correlation between carcinogenicity and some more easily determined property would be greatly enhanced if there were a clearer understanding of the mechanism of the action of carcinogenic compounds. It is, of course, very difficult to obtain such an understanding, for so little is known about cancer itself. Whether or not carcinogenisis by hydrocarbons and related compounds is in any way connected with spontaneous cancer is a' question that has received much conjecture. That such a relationship does exist is suggested by the fact that certain derivatives of estradiol (see Table 1) which is closely related to the sex hormone estrone, are carcinogenic. Spontaneous cancer quite often occurs in the reproductive organs. Furthermore, the very potent carcinogen, 20-methylcholanthreno, can be produced by the degradation of des- oxycholic acid, cholic acid, or cholesterol, the first two of which are normal constituents of human bile and the last of which is present in all tissues of the human body.22 These degradation reactions were, however, conducted under drastic conditions which are in no manner related to those prevailing physiologically, and there is no evidence that 20-methylcholanthrene is formed in the body through the degradation of bile acids or cholesterol. The possible connection between cancer produced spontaneously and carcinogenisis by methylcholanthrene or related compounds therefore remains a matter of conjecture, at least for the present. There has, for obvious reasons, been little attempt to induce cancer in human beings through exposure to the various com pounds that are carcinogenic to mice. Thero is, therefore, no direct correlation between the susceptibility of man and the tost animals. There is, of course, an indirect correlation of occupational cancer largely through exposure to coal tar and dye intermediates caused by materials which are carcinogenic to mice. However, many of the compounds shown in Table 1 do not occur in any products known to have caused occupational cancers, and it is conceivable that some compounds harmful to mice are innocous to man. Until that can be .the only safe policy is to regard al 1 compounds which are harmful to mice as dangerous and to avoid exposure by humans. HHZrVAC 6/23A5 Report No. M-1247 5 TABLE 1 CARCINOGENIC COMPOUNDS All 'compounds listed by Hartwell2^ unless otherwise indicated. 1. The following derivatives of 1 ,2-benzanthracene: 3 -methyl 5- methyl 6- methyl 9- methyl 5 -ethyl 10- uethyl 5-n-propyl 5-n-butyl2 5 -n-amyl2 5- n-hexyl2 6- i-propyl 3-hydroxy 10-hydroxy 10-aldehyde, H -C 0 H 0 10-methyleneacetoxy, -i-O-C-CH^ H H 0 10-methylenecarbornethoxy, --Ci--C--0--CHo i j H H i 10-cyanomethyl, -C-CII H 3-methoxy 10-methoxy 4.9- dimethyl 4.10- dimethyl 5 .6- dimethyl 5 .9- dimethyl 5 .10-dimethyl 6.,7-dimethyl 9.10- dimethyl 6.7- dimethyl 9 .10 -dimethyl 5-chloro, 10 -methyl 7-chloro, 10-methyl 5-cyano, 10-methyl 7-cyano, 10-methyl 1 ,2,7,8-dibenzanthracene31 1 ,2 ,5 ,6-dibenzanthracene Report No. M-1247 6 B ig following derivatives of 1,2,5,6-dibenzanthracene: 92 '--mmeetthhoyxly ' 3' -methyl 4-methyl 9Y-m,9et-hmyolthylene H ' H H 4,10-dimethylene, --k- H 3,4 -dlmethylene (phenanthra- 8,9-dlmethylene acenaphthene) 2. Derivatives of 3,4-"benzpyrene J3. 5 ,6-cyclo pervteno - 4 61 ?' ,72-c'y,c3l'o,p4e'n-tteentorahydrq Na salt of 9,10-endo alpha,beta succinic acid: H-C-C-ONa 3,4-benzpyrene The following derivatives of 3,4-benzpyrene: | 4 Li 5-aldehyde 4'-methyl 5-methyl 1 ',2 '-dihydro, 4'-methyl Y ,2 ',3',47-tetrahydro 10-acetyl 3,4,8,9-dibenzpyrene 5,lO-dihydro-3,4,8,9-dd- benzpyrene 7-methyl,1,2,3,4-dibenzpyrene 3. Derivatives of Cholanthrac ene m Cholanthrene The following derivatives of cholanthrene: 20-methyl 20-ethyl 20-isopropyl 20-tert-butyluu 1 5 .20-dimethyl 1 6 .20-dimethyl 1 5 -hydroxy, 20-methyl 1 5 -keto, 20-methyl Li Keport No. M-12^7 7 4. Other Derivatives of Phenanthrene Colchicine 1 ,2 ,5 ,6-dihenzfluorene '& ' S.' 3,^-henzphenanthrene The following derivatives of 3 ,^-henzphenanthrene ' 1 - methyl2 2- methyl2 2 -ethyl2 2-isopropyl2 7- methyl^ 8-methyl2 2 ,9-diethyl 2-methyl, 1 ,2 ,3 ,1*- dihenzphenanthene Cholesterol oleate . ^ CH3 H H-C-(CH2 )?-C-(CH3 )2 1 ,2-dimethyl chrysene2 5-methyl chrysene17 I*,5-methylene chrysene17 5 ,6-dimethyl chrysene17 Report No. M-12^7 Equilonin "benzoate 8 o 1 ,2-cyclopentene phenanthrene2 5. Other Aromatic Compounds Tetraphenylmethano 1,3,5-triphenylbenzene Triphenylothyleno 53 Report No. M-12^7 6. Organic Compounds Containing Nitrogen , 2-amino 1 -naphthol Alpha naphthylamine13 Beta-naphthylamine - 0 OH Acetylcholine, CH3 -C-0-CH2-N-(CE3 )3 p-dimethyl aminoazohenzene (butter yellow) \ y - N ( C H ,)2 Scarlet red ^-CH, ww Indole / -n-CH3 ^ \ )-N=N- \ -/ H / \_/ Syyryl CH,-C-N Benzidinelt5 H3C 0-C-CH3 Report No. M-12Vf 10 l>2,7i8-dtbenzacridine 3^.-5, ^-ditenzcarbazole 7 Miscellaneous Arsenic trioxide Potassium arsenite Aqueous hydrochloric acid Aqueous potassium hydroxide Zinc chloride Zinc sulfate Ethyl alcohol Fructose Glucose Galactose a Report No. M-12k7 p fill i J t . Li ! .... . 1} J BIBLIOGRAPHY 1. Auld, S. J. M., J. Inst. Pet. Tech., 2k, 577-83 (1938). 'R `E ' > 2. Badger, G.M., Cook, J.W., Hewett, C.L., Kennaway, E.L. ,Kennaway, 439 67 ^ Eobinson> A.M., Broc. Boy. Soc., B129, 3. Badger, G.M., Cook, J.W., Hewett, C.L., Kennaway, E.L., Kennaway, N.M., and Martin, R.H., Proc. Roy. Soc., B131, 170-82 (1942). 4. Barry, G., Cook, J.W., Haslewood, G.A.D., Hewett, C.L., Hieger, I . , and Kennaway, E.L., ibid, B-117, 318-51 (1935). 5. (1943^blUm* *> and Schoental, R., Brit. J. Exp. Path., 24, 232-9 6. Bottomley, A.C., and Twort, C.C., Am. J. Cancer, 21, 781-8 (1934). 7- Bruce, W.E., and Todd, F., J. Am. Chem. Soc., 6l, 157-61 (1939) 8. Campbell, J.A., Brit. J. Exp. Path., 1 5 , 287-9!*. (193k). 9. Campbell, J.A., ibid, 20, 122-32 (1939). 10 . Cook, J.W., Nature, lk5 , 335-8 (I9k0). 1 1 . Cook, J.W., Haslewood, G.A.D., Hewett, C.L., Hieger, I., Kennaway, E.L., and Mayneord, W.Y., Am. J. Cancer; 29, 219-59 (1937). 12* Cook, J.W., and Kennaway, E.L., ibid, 33, 50-97 (1938). 13. Cook, J.W., and Kennaway, E.L., Ibid, 39, 38I-U28, 521-82 (I9k0). 14. Cook, J.W., Hieger, I., Kennaway, E.L., and Mayneord, W.V. ,, Proc Boy. soc., Bill, k55-8k (1932). ' 15- Cramer, W . , and Stowell, R.E.,Cancer Research, 3, 668-81. (1943) 1 6 ' ^ b^Jolskaya-Zavadskaya, N., Compt. rend. soc. biol., 129- 1055-7 17* Dunlap, C.E., and Warren, S., Cancer Research, 3, 606-7 (d$43). 18. Evans, E.E., J. Urol., 38, 212-15 (1937). 19. Fioser, L-P., "Chemistry of Phonanthrone", Relnhold publishing Company, N.Y., (1937). Eeport No. M-1247 Bibliography (Continued) 20. Fieser, L.F., Am. J. Cancer, 34, 37-124 (1938). 21. L,F*' 811(1 Campbell, W.P., J. Am. Chem. Soc., 60, 1142-5 .22 Fieser, L.F., and Fleaer, M . , "Organic Chemistry", pages 814-26 Heath and Co., Boston, Mass., 1944. ' ^ 23. Gottschalk, R.G., Proc. Soc. Exp. Biol. Med., j>0, 369-73 (1942). 2h. ? i 9 ^ r A *' ^ RoblnBon> A -M -> Proc* Ey- soc., B127, 277-87 25 26. Hartven, J.L., U.S. Public Health Service, Nat. Inst. Health, ' Tested8?nr^p1118? " Tested for Carcinogenic "Survey of Activity", 3C7o1mppopu.nds vhlch Have Been Heller, I., j. ind. Hyg., 1 2 , 169-97 (1930). 27 ^ > 3 7 f1931)K'en n a va y> N -M `> andKennavay, E.L., J. Hyg., 31, 28. Hleger, I., Blochem. J., 24, 505-11 (1930). 29 Hueper, W.C., J. ind. Hyg., 16, 255-79 (1934). 30. Kennaway, E.L., Brit. Med. J., 1925, II 1-4. 31. -32. Kennaway, E.L., Blochem. J., 24, 497-504 (1930). 1044-6?7 ' E 'L *' Hieger' *> Brit* Med* J -> 1930, No. 3622, 33. S L h ; 3i ; 169-92 (i9i2)BUSh ' H 'P '' C -A -. Cancer 34. 30T?b387:to9J { 1 9 u t ' ,eaier' H '"'' teoh- Pati- * " " (Tlrchow'e), 35. ,L *> Alexander's "Colloid Chemistry", (1944), Eelnhold Publishing Co., N.Y. Vol. V,' 995-1050 5 36. ^1935?" A *A,> ClaPP' 811(1 B^and, C.F., Science, 82, 134 37- Nordmann, M., Z. Krebsforach., 47, 288-302 (1938). Report No. M-1247 Bibliography (Concluded) 38. Robson, J.M., and Bonser, G.M., Nature, 142, 836 (1938). 39- Roffo, A.H., Prensa Med. Argentina, 26, 721-38 (1939). k 0 ' (1939) A'H*' B01, InSt' MQd' EXP* Estud- Cancer, 15, 741-56 41. Roffo, A.H., ibid, 16, No. 52, 1-6 (1939). 42. Roffo, A.H., ibid, 1 7 , 661-98 (1940). .43. Roffo, A.H., and Roffo, A.E., Jr., ibid, 20, 143-87 (1943). hh' ^ 9 4 3 ) n ' W "L `' ^ 0,13 perrault' A - J - Kat- Oanoer C rsm er' W,> Cancer Research, 3, 6o4-5 46. Stallybrass, C.O., J. State Med., 34, 249-68 (1926). ^7. Stowell, (19^2). R.E , and Cramer, W., Cancer Research, 2, 193-7 48` (1930) C 'C ',^ ;Fulton, Path` Bact, 33, 119-43 ^9. Twort, C.C., and Lyth, B., J. Hyg., 33, 464-73 (1933). 5* (1931) C `C '' ^ T m T t' J `M ', J * Ind' m '* -- ' 20k~26 51. Twort, C.C., and Twort, J.M., J. Hyg., 29, 373-9 (1930). 52. Twort, C.C., and Twort, J.M., ibid, 35, 130-49 (1935). 53. Twort, C.C., and Twort, J.M., Lancet, 1935, n , 1226-8 . 5^. Twort, J.M., and Twort, C.C., J. Hyg., 39, l6l-9 (1939). 55. Woodhouse, D.L., J. Inst. Pet. Tech., 20, I057-63 (1934). thin vdsh4n acquire a more extensive background in a 7 i'le8er ls recommended as a'good starring Stalled ^ original papers by Cook and co-workers. For a be consulted01^ 1011 f carclnoSenic compounds Hartwell's book2 5 should