Cancer Immunol Immunother (1984) 17:147-153 Review ancer mmunolggy mmunotherapy © Springer-Verlag 1984 Membrane antigens associated with infection, transformation, and tumorigenesis by polyoma virus Isaac P. Witz 1. and Georges Meyer 2 i Department of Microbiology, The George S. Wise Faculty of Life Science, Tel Aviv University, Tel Aviv 69978 Israel 2 Unit6 119 Institut National de la Sant6 et de la Recherche Medicale, 27 Bd Lei Route, F-13009 Marseille, France Introduction The fact that polyoma (Py) virus does not induce tumors under natural circumstances has led several investigations to propose that immune surveillance of tumor outgrowth actually func- tions only in cases involving ubiquitous, horizontally trans- mitted, potentially oncogenic viruses [20]. Surprisingly, however, relatively little has been done to elucidate the mechanism of this successful experiment of nature. Those workers who investigated the nature of the functional components involved in the resistance of adult, immunologically intact, animals to the induction of tumors by Py virus reached the conclusion that a T-cell-dependent reactivity was the primary factor involved in this resistance. This was based on experiments involving immunologically compromised animals such as ALS-treated animals [2, 46]; neonatally thymectomized mice [3], or nude thymus-less mice [42, 47]. The antigen present on the surface of Py-virus-transformed cells evoking this T-cell-associated immune surveillance is probably the tumor-specific transplan~:ation antigen (TSTA) [41]. Another species of surface antigens, termed collectively tumor-associated surface antigens (TASA), are determinants defined by serological reagents. With the exclusion of the virally coded middle T [18], other Py-associated TASA are poorly defined. Although some reports indicate a correlative association between TSTA and TASA [e.g., 49], no firm relationship between these antigens has ever been established. Moreover, at present we are not in a position either to estimate the number of Py-associated TASA (if more than one does exist) or to determine their structure and function. One of the major difficulties in defining the membrane antigens associated with Py virus transformation and tumori- genesis is the heterogeneity of cells and antiserum reagents used in different laboratories. Polyoma-virus-induced tumors can be obtained by various procedures: (a) Inoculation of newborn animals with complete or defective virus [13]; (b) inoculation of immune-compro- mised animals with complete or defective virus; (c) inoculation of newborn or adult animals with in vitro-transformed cells that were not passaged in vivo, which could be either primary cultures [48] or established cell lines [24]; and (d) inoculation * Incumbent: The David Furman Chair of Cancer Immunobiology Offprint requests to: I. P. Witz of newborn or adult animals with in vivo-passaged tumor cells [which could have arisen in vivo (e.g. as in point a, above) or in vitro (e.g. as in point c, above)]. When the history of these types of tumors and their biology are considered the possibility that certain differences might exist between them should be borne in mind. 1) The progression towards malignancy of cells trans- formed in vivo and cells transformed in vitro takes place in completely different environments, selecting, most probably, for different phenotypic traits. 2) While one could expect that the first in vivo passages of tumor cells originating from in vitro-transformed cells maintain some (or most) of the antigens of their cultured predecessors, one cannot make this assumption for similar tumors passaged for prolonged periods in vivo [22]. In addition, established cell lines may have acquired new surface antigens (not present in primary cultures). These antigens may correspond to a first step of transformation which immortalizes the cell [38]. 3) Furthermore, prolonged in vivo passage may have a cumulative effect on the antigenic constitution of the cells. Thus, certain antigens may not be so accentuated in primary tumors as they are in passaged ones. For example, prolonged in vivo passage may increase the repertoire of membrane antigens [54]. 4) The location of the in vivo-growing tumor and its form of growth (solid versus ascites) may effect antigenicity [39]. In view of these observations, it may be hardly possible to compare data on the antigenicity of Py-virus-induced tumor cells if these data have been obtained by analyzing cells obtained by different induction procedures. A corresponding heterogeneity is revealed upon exami- nation of the antiserum reagents used. Such reagents may include: (a) antibodies from non-tumor-bearing animals actively immunized with Py virus, with Py-virus-infected cells, with Py-virus-transformed ceils, with Py-virus-induced tumor cells or sera from animals immunized both with cells and with virus; and (b) antibodies from animals bearing tumors origi- nally induced by Py virus. It is obvious that these reagents would reflect the antigenic constitution of the immunizing agent. In this paper we review the present state of knowledge on the membrane antigens associated with Py virus infection, transformation, and tumorigenesis with reference to the type of cells expressing the antigen in question and the antiserum reagent utilized. Unless otherwise established, we do not assume that the various cell types mentioned above are identical.