«314 SYMPOSIUM 40: 450-458; GOLDBERG, L. J. AND LEIF, W. R. 1950. The use of radioactive isotopes in determining the retention and initial ...»
40: 450-458; GOLDBERG, L. J. AND LEIF, W. R. 1950. The use of radioactive isotopes
in determining the retention and initial distribution of airbone bacteria in the mouse.
Science, 112: 299-300; SONKIN, L. S. 1951. The role of particle size in experimental
airborne infection. Am. J. Pub. Health. 53: 337-354.
GOWEN, J. W. AND CALHOUN, M. L. 1943. Factors affecting genetic resistance of mice
to mouse typhoid. J. Infect. Dis. 73: 40-56.
IRWIN, M. R. 1929. The inheritance of resistance to the Danysz bacillus in the rat.
Genetics. 14: 337-365.
4. LURIE, M. B. 1941. Heredity, constitution and tuberculosis, an experimental study.
Amer. Rev. Tuberc. 44: No. 3, suppl. 1-125.
LURIE, M. B., ABRAMSON, S., HARRIS, T. N., ALLISON, M. J. AND HEPPLESTON, A. G.
1949. Constitutional factors in resistance to infection. I, II and III. Am. Rev. Tuberc.
59: 168-185, 186-197, 198-218.
LURIE, M. B., HEPPLESTON, A. G., ABRAMSON, S. AND SWARTZ, I. B. 1950. An evaluation of the method of quantitative airborne infection and its use in the study of the pathogenesis of tuberculosis. Am. Rev. Tuberc. 61: 765-797.
LURIE, M. B., ZAPPASODI, P., DANNENBERG, A. M., JR. AND SWARTZ, I. B. 1951. Constitutional factors in resistance to infection. The effect of cortisone on the pathogenesis of tuberculosis. Science. 113: 234-237.
LURIE, M. B., ABRAMSON, S. AND HEPPLESTON, A. G. 1952. On the response of genetically 8.
resistant and susceptible rabbits to the quantitative inhalation of human type tubercle bacilli and the nature of resistance to tuberculosis. J. Exp. Med. 95: 119-134.
LURIE, M. B., ZAPPASODI, P., CARDONA-LYNCH, E. AND DANNENBERG, A. M., JR.
1952. The response to the intracutaneous inoculation of BCG as an index of native resistance to tuberculosis. J. Immunol. 68: 369-387.
LURIE, M. B., ZAPPASODI, P., DANNENBERG, A. M., JR. AND CARDONA-LYNCH, E.
1952. Constitutional factors in resistance to infection. The effect of cortisone on the pathogenesis of tuberculosis. Symposium on the Effect of ACTH and Cortisone Upon Infection and Resistance. Columbia University Press. (in press) WEBSTER, L. T. 1933. Inherited and acquired factors in resistance to infection. J. Exp.
Med. 57: 793-817.
12. WRIGHT, S. AND LEWIS, P. A. 1921. Factors in the resistance of guinea pigs to tuberculosis, with special regard to inbreeding and heredity. Am. Naturalist. 55: 20-50.
4. The Bearing of Mouse Genetics on Our Understanding of Human Cancer, W. E. HESTON National Cancer Institute, Bethesda 14, Maryland THE problems involved in our understanding ofthe cause, origin and develop ment of cancer are basically problems in biology. The study of neoplasia con cerns living cells and the factors acting through the physiology of the living ' Paper presented as part of Symposium on "Light from Animal Experimentation on Human Heredity," at fifth annual meeting of The American Society of Human Genetics at Ithaca, New York, September 9, 1952.
ANIMAL EXPERIMENTATION AND HUMAN HEREDITYorganism and of the cell to cause it to take on a new characteristic of uncontrolled growth and to support this growth. Cell differentiation, cell nutrition, growth and biochemistry are involved, and, in that all of these are under the control of the genotype of the organism and of the cell, genetics is likewise involved. It is, therefore, through studies in genetics that much of our fundamental understanding of cancer is derived.
Because of difficulties in the collection and analysis of data from human beings on such a complex character as cancer, the ground work for the study of the genetics of cancer must be done with some other species. Important contributions to our understanding of the problem have been made and will continue to be made by investigations with Drosophila, corn, Neurospora and other lower organisms. Because of its physiology approximating that of man and because of its wide variety of neoplasms occurring with such frequency necessary for adequate study, the mouse, however, has provided the most extensive approach to the problem. Certain superficial facts revealed obviously can be applied only to the mouse, but basic facts and concepts derived from studies on the mouse can also be applied to man.
IS CANCER INHERITED?
From genetic studies in mice one is forced to conclude that probably every type of cancer in man can be subject to genic influence so that under certain conditions the genotype of the individual may be the deciding factor in determining whether or not the individual develops a certain type of cancer. Through the manipulation of genes in the development of the inbred strains, and through hybridization of the strains, geneticists have been able to prevent or promote the occurrence of more different types of cancer and to influence the occurrence of each to a greater degree than has any other investigator with any other group of agents. Strains vary in respect to incidence of mammary gland tumors from strain C3H in which 95 per cent or more of the females have mammary tumors to strain C57 BL in which less than one per cent of the females develop mammary tumors (Andervont, 1941, Heston et al 1950). Extremes in regard to pulmonary tumors are strain A with an incidence of 90 per cent at 18 months of age and strain C57 L in which only 3 spontaneous pulmonary tumors have been observed in the thousands of mice throughout the history of the strain (Heston, 1942b). Strains C58, and AKR are known for their high incidences of lymphocytic leukemia. (MacDowell, et al. 1945, Law, 1948).
These have been reported as high as 90 per cent. Little (1941) has reported an incidence of non-epithelial tumors of about 20 per cent in strain C57 BL. Most of these were classified as lymphoblastomas. About 25 per cent of mice of strain C57 L develop a Hodgkins-like lesion. In contrast, in strains A and C3H neoplasms of the blood-forming organs are rarely found. Hepatomas occur in approximately 30 per cent of strain C3H males and 10 per cent of the females 316 SYMPOSIUM (Andervont 1941). Strong (1945) has developed a strain with a high incidence of tumors of the glandular stomach, and strain I has a high incidence of a lesion of the glandular stomach (Andervont and Stewart, 1937), whereas tumors of the gastrointestinal tract rarely occur in other strains. Pybus and Miller (1938) reported a strain of mice with a high incidence of bone tumors, a type rarely found in other strains. Gardner and Pan (1948) reported a strain PM derived from the original Pybus and Miller strain with a high incidence of spontaneous uterine, cervical and vaginal tumors, although none were observed in 6 other strains. Woolley and co-workers (1945) developed a strain CE with a high incidence of carcinoma of the adrenals following early castration. With the same treatment strain DBA mice showed only hyperplasia of the adrenal cortex and strain C57 BL mice gave little if any adrenal response.
Burdette and Strong (1943) reported genetic differences in the incidence of subcutaneous sarcomas following the injection of methylcholanthrene. Strain C3Hf mice without the milk agent and their hybrid derivatives have a higher incidence of spontaneous subcutaneous sarcomas than have other strains.
Mice of strain HR which carries the hairless gene have a relatively high incidence of spontaneous papillomas and when painted with methylcholanthrene develop a higher incidence of squamous-cell carcinomas than has been reported for any other strain (Deringer, 1951). Even in respect to very rare tumors such differences are seen. Myoepitheliomas, often of salivary gland origin, occur much more frequently in strain A than in any other strain. In hybrids between strains C3H and C57 BL a number of tumors of the Harderian gland were noted although this tumor has not been reported in any inbred strain or other type of hybrid.
From these observations of genetic differences in respect to these many varied types of tumors it is safe to assume that strains of mice differing in incidence of any type of tumor could be developed with the proper genes in the original population followed by the necessary selection.
One who has worked with these strains of mice, therefore, no longer asks if cancer is inhreited in man. Instead he asks how is cancer inherited in man and how important are the genetic factors as compared with the various non-genetic factors in determining whether or not cancer will occur.
CAN A GENERAL INHERITED SUSCEPTIBILITY TO CANCER IN MAN BE EXPECTED?
One of the first questions to be answered by data from the inbred strains is whether or not a general susceptibility to cancer is inherited. The fact that the strains are susceptible to specific types of cancer and that almost no strain is resistant to all types has indicated that the various types are inherited as independent characteristics. For the most part this would be expected when one visualizes the physiologic paths through which the genic action must become manifest. It is difficult to perceive how the development of pulmonary
ANIMAL EXPERIMENTATION AND HUMAN HEREDITYtumors could be preceded by the same physiologic factors that affect mammary gland tumor development.
On the other hand it would seem possible that the development of tumors in different organs of the same physiological system might have a common genetic basis. Variations in hormonal balance might influence tumor development in a number of endocrine organs, the reproductive organs and the breast. There is some evidence of a relationship between adrenal hyperplasia and the genes influencing mammary tumors through the control of the hormonal stimulation in mice. (Smith 1948, Huseby and Bittner 1948). Data of Macklin (1952) suggest that such an association may be demonstrated in man in the relationship between breast cancer and cancer of the prostate. In the mouse specific genes have been shown to influence the development of more than one tumor, but, as will be discussed subsequently, the effect may be through separate physiologic paths and one gene may have opposite effects upon two different types of tumor.
MULTIPLE FACTOR INHERITANCE OF CANCERAnother major question to be answered is in regard to the type of inheritance observed in respect to cancer. It should be emphasized that in studies of the mouse where breeding experiments can be carried out and where F2 and backcross segregants can be tested to differentiate between single factor ratios and simulated single factor ratios due to fluctuations about physiologic thresholds, not one type of cancer has thus far been shown to be controlled by a single gene. This would demand caution in concluding single factor inheritance for cancer in man. Before inbred strains of mice were available and when the only estimate of the genotype was the observation of whether or not the individual mouse developed a tumor, results were obtained which were interpreted as indicating single factor recessive inheritance. Following the development of the inbred strains single factor dominant inheritance was indicated by certain simple crosses. This was true of both pulmonary tumors and mammary gland tumors in mice. The variation between the various inbred strains, however, could not be explained with such simple interpretations. For example, in regard to pulmonary tumors, strain A has an incidence of 90 per cent, strain Swiss 40 per cent, strain BALB/c 20 per cent, strain C3H between 5 and 10 per cent and C57 L and C57 BL less than one per cent. Furthermore, when strain A was outcrossed to three different low-tumor strains the degrees of susceptibility to induced pulmonary tumors observed in the three resultant groups of F1 hybrids were not the same but were relatively high, medium and low (Heston, 1940). Later by utilizing the number of nodules appearing in the lungs of each individual as a reliable quantitative measure of degree of susceptibility to induced pulmonary tumors and checking these results against those obtained by using latent period as a measure of susceptibility multiple factor inheritance 318 SYMPOSIUM was established for induced pulmonary tumors (Heston, 1942a). Through hybridization studies multiple factor inheritance has also been established for spontaneous pulmonary tumors (Heston, 1942b). Here, as with the genetic analysis of other spontaneous tumors, there is not a quantitative measure of degree of susceptibility and the problem is more comparable to that of polydactyly in the guinea pig described by Wright, (1934b), i.e. that of a multiple factor character with alternative expression depending upon whether or not the combined action of the genetic and non-genetic factors surpasses a physiologic threshold.
While a quantitative measure of degree of susceptibility to mammary gland tumors in mice has not been available, the vast amount of data on mammary tumors, both from tabulation of incidences in the various strains and from hybridization studies, can be explained only on the basis of multiple factors (Bittner, 1952, Heston, 1945). Burdette (1943) has demonstrated multiple factor inheritance for induced subcutaneous sarcomas, and since the F1's were intermediate he assumed that at least one factor was dominant and one was recessive. In crosses between high and low leukemic strains Cole and Furth (1941) observed that the incidence of leukemia in the hybrid groups was roughly correlated with the total heredity from the high leukemic strain.
Through breeding tests of backcross segregants MacDowell et al (1945) established the inheritance of leukemia as multiple factor. Data on other types of tumors in mice consists primarily of incidence data of the inbred strains, but none suggests single factor inheritance.