Cell death and the subsequent post-mortem changes, called necrosis, are integral parts of normal development and maturation cycle. Despite the importance of this process, the mechanisms underlying cell death are still poorly understood. In the recent literature, cell death is said to occur by two alternative, opposite modes: apoptosis, a programmed, managed form of cell death, and necrosis, an unordered and accidental form of cellular dying. The incorrect consequence is the overlapping of: a) the process whereby cells die, cell death; and b) the changes that the cells and tissues undergo after the cells die. Only the latter process can be referred to as necrosis and represents a process in cell life. In this review, we discuss the excellent basic research developed in this field during last decades and problems that remain to be resolved in defining both experimentally and mechanicistically the events that lead to and characterize cell death.

Over the past few decades, there has been a tremendous increase in cancer biology data and treatment. Cancer research has opened exciting new areas of cellular and molecular biology. Month by month, new genes which regulate the carcinogenesis process are being discovered. The result is an incredible knowledge of cancer: what makes a cancer cell a cancer cell, what cancer cells need to develop, and how cancer cells behave, interact, overgrow and die. In parallel, gene manipulation within cells lets us foresee future possibilities of new cancer treatments. On the other hand, this combination of increased knowledge and powerful new techniques has provided no effective cancer therapy. As it has been quoted during the 'Update and Intensive Review of Internal Medicine' meeting held in New York, August 1999: '...The success in treating Hodgkin's disease means that patients now live enough to develop complications related to the treatment'. Thus, after dedicated decades of excellent research, cancer remains a significant human, clinical, and economical burden. The purpose of this review is 2-fold. First, to analyze areas of basic cancer research that still await adequate scientific explanations. Second, to stress that, for its continuing advancement, cancer research is dependent upon close relationships among many disciplines; an intimate alignment of oncologists with biochemists, geneticists, immunologists, experimental pathologists, and pharmacologists is needed. In light of the great success registered at the basic science level but lack of effective therapies, it would be wise to establish human and economical resources addressed to a multidisciplinary collaborative effort in cancer research.

As part of a study to explore further the biochemical pathology of chemical hepatocarcinogenesis, cytoplasmic tRNA patterns have been studied in ethionine-induced rat putative preneoplastic nodules in comparison to surrounding non-nodular liver and control liver. A new h.p.l.c. methodology, able to resolve contemporaneously the numerous components in a tRNA population, has been used. The results obtained indicate the presence of marked differences in the chromatographic profiles of hepatocyte nodules, non-nodular surrounding liver and control liver. The differences are both quantitative and qualitative, with some chromatographic peaks showing increases and some decreases or absences. Also, new peaks are seen reproducibly in the nodules. The data show major changes in the tRNA population during hepatocarcinogenesis that might have mechanistic implications.

The effect of age and peroxidative stress on the concentration of a deoxyguanosine malondialdehyde adduct (dG-MDA) in rat tissues was investigated. Vitamin E deficiency had no effect on the dG-MDA content of liver DNA in rats fed a diet containing 10% corn oil. When 2% cod liver oil was added to this diet, the dG-MDA content of liver DNA doubled in the positive controls fed a high level of vitamin E (100 ppm dl-alpha-tocopherol), and there was a further increase when vitamin E was deleted. Neither iron nitrilotriacetate administration nor choline deficiency had any effect on the dG-MDA content of liver DNA. Carbon tetrachloride had a lowering effect. The failure of iron or carbon tetrachloride administration and of vitamin E deficiency to increase liver dG-MDA is consistent with their failure in previous experiments to affect the urinary excretion of dG-MDA. In contrast, these forms of peroxidative stress produce large increments in the urinary excretion of MDA adducts with lysine, reflecting increased formation and degradation of MDA-modified proteins. DNA appears to be protected from modification by MDA produced at extranuclear sites. The frequency of dG-MDA in different tissues of 4-month-old rats varied markedly: brain >> liver > kidneys and testes. Higher concentrations of dG-MDA were found in the liver and kidneys, but not the testes, of 25-month-old rats. The determinants of the concentration of dG-MDA in DNA merit further investigation.

The sites and times of appearance of preneoplastic foci in rat liver acinus during the first 11 days after initiation with diethylnitrosamine and promotion with 2-acetylaminofluorene plus partial hepatectomy was observed in three separate experiments. Foci appeared as alterations of hepatocytes followed by focal proliferation in each of the three zones before any ductular epithelial cell ('oval cell') proliferation. The dissociation between foci and 'oval cell proliferation' is strong evidence against a role of the latter in generating preneoplastic hepatocytes in hepatocellular carcinogenesis.

Cell death remains poorly understood, despite its obvious importance in every organ and tissue in a wide variety of biological processes, including, of course, the many pathological. The past few years have seen an amazing expansion of interest in cell death in normal development and maturation, in the pathogenesis of many acute and chronic diseases, and in the therapy of some diseases, especially malignant neoplastic diseases and some hyperplastic diseases such as psoriasis. This expansion has included an unusual interest in a supposedly new form of cell death, a "programmed cell death," designated "apoptosis." This is proposed as a hitherto undescribed form of cell death in contrast to the classical cell death, necrosis. Apoptosis is considered by some, especially by nonpathologists, to represent quite a different type of cell death. A review of the literature on on apoptosis, programmed cell death, necrosis, etc. indicates that there is no field of basic cell biology and cell pathology that is more confusing and more unintelligible than the area of apoptosis versus necrosis. If any degree of clarity is to develop in our understanding of the fundamental principles underlying cell death of any type, it is incumbent upon us to rethink "from square one" the scientific analysis of how cells die and how can we assess cell death in a reasonably rational manner.

Changes in the degree of methylation of cytosine in DNA are considered to be mechanistically important in modulating gene expression. To gain a better understanding of the relationship(s) linking onco‐proliferative processes and enzymatic DNA methylation, a study has been carried out on the hepatic DNA methylation pattern during DNA replication following partial hepatectomy (PH), mitogen treatment and N‐methyl‐N‐nitrosourea (MNU) administration in rats. The following results were obtained: (i) DNA hypomethylation was seen during DNA synthesis, with each of the 3 stimuli, namely MNU administration, partial hepatectomy, and hepatomitogen treatment; (ii) the level of DNA hypomethylation was not quantificatively related to the extent of DNA replication as measured by incorporation of [3H]thymidine into hepatic DNA: (iii) MNU administration under conditions conducive to carcinogenic development, i.e. during the S phase of compensatory cell proliferation, caused hypermethylation of replicating hepatic DNA, as shown by Hpall and Mspl restriction patterns.

Levels of various cytochrome P450 proteins have been reported to be decreased to varying degrees in chemically induced hepatocyte nodules and following partial hepatectomy (PH). By screening a rat liver lambda ZAP cDNA expression library with antibodies raised against a partially purified preparation of cytochrome P450 isolated from untreated male Fischer 344 rats, we have isolated a 1.1-kb cDNA. This cDNA was sequenced for 139 bases from the 5' end of the sense strand and comparison of the resulting sequence with the sequences in Gene Man DNA data bank revealed 95% homology of the sequenced portion with male-specific rat cytochrome P450 (M-1, CYP IIC11). The 32P-labeled cDNA was used as a hybridization probe on RNA blots (Northern blots) prepared with total RNA from rat livers obtained post PH, from aflatoxin B1(AFB1)-induced rat liver tumors and from rat liver nodules induced with a combination of diethylnitrosamine/acetylaminofluorene/PH (DEN/AFF/PH). At 36 and 72 hr post PH, the mRNA level was decreased by > 93%. Relative to the corresponding control livers, the mRNA level was also decreased by 97% in the liver nodules and by 57% in AFB1-induced liver tumors. The RNA blots derived from the liver nodules and AFB1-induced liver tumors were also probed with a cDNA probe (R17) that recognizes other cytochromes P450 (CYP IIB1/CYP IIB2). The mRNA corresponding to CYP IIB1/CYP IIB2 was also depressed 92% in the nodules and 65% in the tumors. These results clearly indicate that the depression of both CYP IIC11 and IIB1/IIB2 in the hepatic nodules and the tumors is related to the inhibition of transcription and/or enhanced degradation of the mRNA.

The dominant dogma concerning the essential cellular changes during preneoplasia and precancer considers these as abnormal or foreign that evoke a basic "host-parasite" response. An alternative view of how cancer develops, here briefly outlined, views the early and intermediate cellular changes as essentially physiologic and adaptive. This different concept introduces clonal adaptation as a basic response to many genotoxic carcinogenic stimuli including chemicals, radiations, and some viruses. The evidence in support of this new view of the carcinogenic process is summarized.

Thus, the pathologic consequences of feeding a CD diet are fatty liver, liver cell death, liver cell proliferation, and liver cell cancer. The fatty liver with CD is similar to that with other types of fatty liver in that the most attractive current hypothesis is based on some interference with the production and output of VLDL by the liver. The induction of cell death appears to be consistent with quite a different hypothesis, genesis and/or increase in liver free radicals leading to both acute necrosis and initiation of carcinogenesis. Especially noteworthy is the low incidence of liver cirrhosis, even after 2 years of exposure to the CD diet. The feeding of the CD diet reproducibly induces severe and persistent fatty liver coupled with extensive cell death, a combination that is frequently considered to be appropriate for the induction of "micronodular" (fatty) cirrhosis in humans. The findings with the LD diet, the high incidence of cirrhosis, with severe persistent fatty liver without significant cell death, together with the low incidence of cirrhosis with the CD diet, stand out as unpredictable and strange, according to current concepts of the pathogenesis of human cirrhosis. The CD model offers an unusual opportunity to explore in increasing detail the possible roles of free radicals in two important problems in pathology and medicine-acute cell injury and neoplasia. The challenges include mechanistic studies on how the free radicals are generated and how they relate to the biological consequences. The relatively slow sequential changes in the induction of cell injury and neoplasia makes the CD model one of the best for mechanistic studies relating to free radicals.

In order to investigate the possible mechanisms by which cellular alterations can start an altered onco-developmental gene expression, we studied tRNA distribution profiles during the early steps of 2-acetylaminofluorene-induced hepatocarcinogenesis. The finding of progressive and sequential alterations appears to support the hypothesis of a causal connection between tRNA changes and nodular cell proliferation, possibly through the disruption of the mechanism which regulates tRNA functional adaptation.

There is an increasing awareness that many peroxisome proliferators are being introduced into our environment and many of these have shown carcinogenic activity in some rodent species. The agents involved include drugs (e.g., hypolipdemic agents of several chemical structures including HMG CoA reductase inhibitors) and industrial chemicals. The neoplasms seen are mainly in the liver with a variable incidence in the pancreas. The association between peroxisome proliferation and neoplasms is impressive. Yet, there are several seemingly fundamental differences between carcinogenic peroxisome proliferators and mutagenic or genotoxic chemical carcinogens. Peroxisome proliferators in general are neither mutagenic nor genotoxic and do not induce precancerous hepatic lesions and liver cell cancer until they have reached the late stage. With many agents, it appears that a considerable degree of peroxisome proliferation must take place for them to be carcinogenic. One poorly documented speculation is that peroxisome proliferators induce cancer by acting mainly as promoters, presupposing that the animals being tested are "preinitiated," a conclusion that is scientifically indefensible. This presentation covers a comparison between the key requirements for cancer development with genotoxic agents and those seen with peroxisome proliferators, and discusses the methylation of genes for selective enzymes of the resistance phenotype induced by mutagenic carcinogens.

Primary cultures of hepatocytes isolated from normal F-344 rats or from F-344 rats with hepatocellular carcinomas generated by a 2-step model of chemical carcinogenesis were used to determine if dexamethasone (DEX) or alpha 2-macroglobulin (alpha 2M) modify the ability of transforming growth factors-beta type I (TGF-beta I) and type 2 (TGF-beta 2) to inhibit labelling index of hepatocytes cultured continuously with or without epidermal growth factor (EGF). Both TGF-beta 1 and beta 2 were equivalently potent inhibitors of S-phase DNA synthesis in normal and neoplastic hepatocytes as determined by 3H-thymidine autoradiography. Both DEX (1 to 100 microM) and alpha 2M (50-200 microM) partially counteracted the mito-inhibitory effect of both TGF-betas on the proliferation of normal and surrounding hepatocytes. In contrast, neoplastic hepatocytes cultured with DEX released much less immunoreactive alpha 2M and were less able to overcome the inhibitory effect of TGF-beta than normal or surrounding hepatocytes. Purified bovine alpha 2M partially counteracted the inhibition of TGF-beta 1 or beta 2 of both surrounding and neoplastic hepatocytes. Both DEX and alpha 2M were more effective against the mito-inhibitory activity of TGF-beta 2. Our data suggest that alpha 2M released by DEX-treated normal hepatocytes contributes to the counteraction of the TGF-beta effect by DEX. Our results support the hypothesis that glucocorticoids and growth-factor-binding proteins may have important roles in modulating the effects of TGF-beta on normal hepatocyte proliferation and suggest that under some conditions hepatocellular neoplasms can be more sensitive than normal hepatocytes to inhibition of proliferation by TGF-beta.

In the present study we have examined the effect of a single dose of the mitogen lead nitrate (75 μmol/kg body wt) on the methylation status of hepatic DNA in male Wistar rats. It was found that extensive hypomethylation of hepatic DNA occurs in mitogen-treated rat liver. This effect could be seen as early as 12 h after metal treatment and parallels the changes in liver weight. Probing with the methylation-sensitive enzymes HpaII, MspI, and HaeIII confirmed HPLC analyses and showed that methylation at these sites was affected by lead treatment. DNA hypomethylation has already been found in regenerating rat liver and in hepatic (pre)malignant lesions when compared to normal nondividing liver. Thus the lowering of the DNA 5-methylcytosine content appears to be a property characteristic of cellular proliferation, regardless of whether it is caused by partial hepatectomy, carcinogen treatments, or mitogen administration.

This study was designed as one test of the hypothesis that an early sequence of steps in hepatocarcinogenesis in the rat, with the production of hepatocyte nodules, may be a special form of adaptive response that has survival value for the host. Fischer 344 rats were initiated with a single dose of diethylnitrosamine. Hepatocyte nodules were rapidly generated by selecting for resistant hepatocytes by a brief exposure to 2-acetylaminofluorene coupled with partial hepatectomy, a procedure that leads to liver cancer without any further treatment. Most animals with hepatocyte nodules were completely resistant to single doses of CCl4 that induced 100% mortality in control animals. The demonstration of this protective effect is consistent with the proposed hypothesis.