Richmond T. Prehn

Studies on tissue homotransplantation in mice, using diffusion-chamber methods.

It is generally accepted, from the work of Medawar’ and of others, that the reaction of the host against homografts is based on a mechanism of actively acquired immunity. It has long been debated, however, whether destruction of homografts is caused by cells, by antibodies or cytotoxins in the serum, or by both. This paper is a brief account of our studies on this problem. In order to facilitate observations on the possible role of these factors, we developed methods by means of which implanted cells, although isolated from contact with cells of the host by a porous filter, could survive and proliferate within living mice?’ a The earliest paper we have found on a method intended for this purpose was that by Rezzesi4 in 1932. In this procedure, as used by Bisceglie,6 a collodion bag containing the graft was tied a t the end and placed within the peritoneal cavity. It was reported that the Ehrlich carcinoma of mice survived for 12 days in a guinea pig under these conditions. The recent availability of filters of cellulose nitrate or of cellulose esters of graded porosity from several sources*, enabled us to construct two types of diffusion chambers. The average porosity of the Millipore filters used in our work, except as indicated, was stated by the manufacturers to be 0.45 p , small enough to prevent the passage of cells but presumably large enough to let through proteins. For observation of living cells, the filters were adapted for use with a transparent chamber inserted under the skin? The filters are relatively opaque, but a portion of the filter, 3 mm. in diameter, was made transparent so as to allow observation of the living cells both within the chamber and on the tissues beneath the chamber. A second type of chamber intended for histologic study was placed intraperitoneally and removed for fixation after varying intervals of time. Inbred strains of mice were used in these experiments. I shall refer to grafts within the same inbred strain as “isologous.” These, as is well known, usually survive. Grafts between mice of different inbred strains will be referred to as homologous. These, of course, are usually rejected by the host. It was known from the work of Merwin and Hill6 that small homografts, placed so as not to become vascularized, will survive indefinitely in the subcutane‘ous site because they do not initiate immunity in the host. It seemed likely, therefore, that homografts would survive in diffusion chambers in nonimmune hosts, and this proved to be the case. * Millipore Filter Corp Watertown Mass.; Carl Schleicher and Shuel! Co. Keene, N. H:; F. M. Himmelweit the Wright-Fleming Insti&e of Microhiology, St. Mary’s Hospital Medical School, Paddington, London, En& land.

THE RELATIONSHIP OF IMMUNOLOGY TO CARCINOGENESIS

Since psychophysiologic factors, particularly various forms of stress, have been repeatedly implicated as modifiers of immunologic reactions, it is logical to consider this pathway when evaluating possible effects of the psyche upon carcinogenesis. That immunologic reactions play an important role in carcinogenesis seems to have been established in a variety of animal experiments. In this paper I will review some of the more pertinent evidence for this role, and particularly more recent data from my laboratory. There is no doubt that many animal tumors are directly due to the actions of viruses. It will indeed be surprising if a similar situation does not exist in man, although definite evidence is still lacking. The oncogenic viruses are antigenic and can arouse strong immune responses; their immunology is similar to that of the better known and supposedly non-oncogenic viral agents. Thus, the immunology of the viral agents does not require special discussion in regard to oncology, except to point out that many of the oncologic viruses in lower animals are transmitted “vertically” from parent to offspring. This transmission may be via the ovum, the placenta, or, sometimes, immediately after birth. The important features of these modes of transmission are that ordinary isolation procedures are ineffective in the control of viral spread, and that the immune reactions of the host are markedly affected by the early age at which the animal is first exposed. The fetus and the very young of most species are immunologically deficient. If “imprinted” at this critical early time with a viral antigen, they may remain deficient throughout life with respect to that antigen.l The immunology of the oncogenic viruses has been extensively reviewed.24 In addition to viruses, many other types of carcinogenic agents have been identified, including hundreds of chemicals, various radiations, and physical (i.e., smooth surface) factors. Tumor cells of whatever varied etiology, viral included, appear to possess antigens capable of arousing a defensive cytologic i r n m ~ n i t y . ~ ~ It is the role of this antitumor-cell immunity, as distinct from antiviral immunity, that will constitute the principal substance of this paper. There is no need to recount the extensive researches that have established the existence of these apparently tumor-specific histocompatibility antigens. This has been the subject of numerous reviews.u Instead, I will discuss some of the evidence concerning their role in carcinogenesis and the possible implications for psychophysiologic mechanisms. I believe that some of the most impressive indirect evidence concerning the importance of these antigens in carcinogenesis comes from a consideration of the interrelationships of latent period, antigenicity, and tumor incidence. Very early in the study of these antigens it was found that the apparent antigenic titre, especially in the case of the chemically induced tumors, varied markedly from tumor to tumor.5 Some tumors appeared to be very antigenic and others only weakly so, if indeed they were antigenic at all. There appeared to be a marked relationship between the latent period, after application of a particular dosage of carcinogen,

Influence of immune status of host on immunogenicity of tumors induced with two doses of methylcholanthrene

SummaryPrevious studies by Prehn demonstrated a direct correlation between the dose of carcinogen used for tumor induction and the immunogenicity of the resulting tumors. The purpose of the present study was to determine the role of the hosts immune response and the influence of the carcinogen on immune function in this relationship. For that reason, a comparison was made of the immunogenicities of tumors induced with two doses of carcinogen in immunologically normal mice and in mice immunodepressed by adult thymectomy and irradiation. If the direct relationship between dose and immunogenicity demonstrated in normal mice was due to the degree of immunosuppression produced by the carcinogen, this correlation should not be apparent in mice already immunosuppressed. Although there was some increase in the immunogenicity of tumors induced in the immunosuppressed mice, the same relationship between carcinogen dose and immunogenicity was observed in both groups of mice. These results indicate that the degree of immunogenicity of tumors induced with both high and low doses of carcinogen was influenced by immunoselection, but in addition another, non-immunologic factor was significant in the relationship between carcinogen dose and immunogenicity.

Effect of different immunization and challenge procedures on in vivo tumor immunogenicity tests.

SummaryThree different methods of immunization and two methods of challenge were used to test a series of methylcholanthrene (MCA)-induced sarcomas for their immunogenicities in vivo. Although there was some variation in the degree of immunity produced, similar results were obtained with the three immunization procedures. Challenge with explants of tumor tissue was found to be as sensitive a measurement of the antitumor response as challenge with a single cell suspension. The results indicate that in a substantial series of tumors, the relative immunogenicities are characteristics of the tumors themselves and not artifacts due to the immunization or challenge procedures used.