The discoveries of chemical, physical, and biological carcinogenic agents and their actions have been among the most exciting and significant in our understanding of the causation of cancer and of many aspects of its prevention, but nothing has intrigued the biological scientist more than the molecular differences between cancer cells and normal cells. The first hint of this inter- est by biochemists in the cancer problem was shown during the first two decades of this century when the structures of nucleotides and sugar phosphates were just becoming known to biochem- ical scientists. In a review, Potter (1982) detailed much of the historical development of bio- chemical investigations into the cancer problem through 1975 (Table 15.1). Just as with the initial morphological and biological studies of the neoplastic process (see Chapter 1), the devel- opment of biochemistry and, more recently, cell and molecular biology has given rise to a series of hypotheses on the biochemical nature of neoplasia as well as the molecular mechanisms that result in the conversion of a normal cell to a neoplastic cell. This chapter considers, first, the historical development and bases for some of the better known theories of the biochemical na- ture and genesis of neoplasia in vivo. The evolution of such concepts into our more modern-day understanding of molecular lesions in neoplastic cells is then discussed.
BIOCHEMICAL THEORIES OF CARCINOGENESIS AND CANCER
Although some understanding of biochemical reactions occurring in living systems such as fer- mentation were known during the latter half of the nineteenth century, quantitative studies on such reactions were pioneered by the German chemist Otto Warburg. Warburg was recognized internationally for his investigations in photosynthesis; in addition, he made a very significant initial contribution to our understanding of biochemical reactions occurring in neoplastic cells.
Glycolysis of Cancer Cells: The Warburg Theory
During the 1920s, the predominant investigations of the biochemistry of cancer centered around the monumental studies of Otto Warburg. He observed that, in the absence of oxygen, tumor slices utilized glucose and produced lactic acid. Warburg termed this process anaerobic glycoly- sis. Generally, slices of cancer tissue were found to produce more lactic acid than did normal tissue slices. In addition, he observed that both normal and neoplastic tissue slices produced less lactic acid in the presence of oxygen (aerobic glycolysis) than in the presence of nitrogen. He called this latter phenomenon the Pasteur effect in reference to Pasteur’s earlier observation that
yeast ceased fermentation when exposed to oxygen. In 1930, Warburg published his book on the metabolism of tumors, in which he demonstrated that in a wide variety of benign and malignant neoplasms investigated, tumors of both humans and lower animals exhibited a significant if not a very high rate of glycolysis (Warburg, 1930). Warburg’s theory (based on his studies up to that time and reiterated in 1956) stated that cancer cells originate from normal cells as a result of an irreversible injury to their respiration, this injury to the normal cell being compensated for in the cancer cell by increased fermentation (glycolysis). Until 40 years ago, there were few if any exceptions to this generalization, although many normal tissues exhibited equally high and in some instances even higher rates of glycolysis than the vast majority of tumors studied; exam- ples are embryonic tissue, the retina, and the renal papilla. In view of these findings, two ques- tions arose that were never answered satisfactorily by proponents of the Warburg hypothesis. The first is the primary association of glycolysis with the growth rate of tumors rather than with the neoplastic transformation itself. A number of studies carried out in many laboratories dem- onstrated a reasonable degree of correlation of glycolytic rate with growth rate of tumors in many systems. Thus, glycolysis may be a secondary event as a result of the loss of control of cellular replication in most neoplasms. The second question concerns the validity of many com- parisons of neoplastic tissues with their cells of origin. This problem recalls our original defini- tion of relative autonomy, which was relative to the tissue from which the neoplasm arose. Normal tissues differ dramatically in their biochemistry as well as their cell and molecular biol- ogy, and neoplasms are found to do the same. This fact was not completely evident during the time of Warburg but now becomes a major factor in determining the significance of changes in neoplastic cells in relation to their normal counterparts.
Weinhouse (1976), in reviewing the Warburg theory on its fiftieth anniversary, summa- rized the experimental evidence pertaining to Warburg’s hypothesis (Table 15.2). It was apparent that tumors do exhibit the Pasteur effect, although quantitatively not as efficiently as normal tis- sues. Weinhouse concluded, “There is no evidence either from Warburg’s own observations or from those of his contemporaries that respiration in cancer is either quantitatively lower or fails to lower glycolysis.” Warburg, and later Greenstein (see below), attempted to generalize their findings to include all neoplasms. Today it is clear that comparisons of normal liver with highly differentiated hepatocellular carcinomas reveal little if any difference in the glycolytic capacities of the two tissues, although, as these neoplasms continue to be transplanted, they tend to in-
crease their glycolytic activity. Investigations in the 1950s actually demonstrated the existence of primary hepatomas with little or no increased glycolytic rate compared with that of normal liver (cf. Greenstein, 1954). Malignant lymphoblasts glycolyze at essentially the same rate as their normal counterparts, and it is likely that malignant teratomas do not exhibit a degree of glycolysis in excess of that found in embryonic tissues. Thus, in support of Warburg’s original hypothesis, most neoplasms do have relatively high rates of glycolysis, but, just as we saw nor- mal karyotypes in early neoplasia, normal glycolytic rates do exist in many neoplasms, espe- cially those that grow slowly and are well differentiated. Increased glycolysis may, therefore, be a characteristic of the stage of tumor progression, similar to the occurrence of many karyotypic changes in neoplasms.