In vivo hepatocyte proliferation is inducible through a TNF and IL-6-independent pathway.

Recent studies in mice harboring a targeted disruption of genes encoding TNF receptor 1 (TNFR-1) or Interleukin 6 (IL-6) suggested a critical role for TNF and IL-6 in initiation of liver regeneration after 2/3 partial hepatectomy. However, hepatocyte proliferation can also occur following treatment with agents that do not induce tissue loss (primary mitogens). To determine whether the above cytokines could also be involved in mitogen-induced liver cell proliferation, we studied the hepatocyte proliferative response after treatment with primary mitogens in mice knock-out for TNFR-1 or IL-6. Our results showed no difference in the proliferative response of the liver between the wild type and the knock-out mice following treatment with the mitogens 1,4-bis[2-(3,5-dichloropyridyloxy)]benzene (TCPOBOP), or the peroxisome proliferator, ciprofibrate, suggesting that TNF or IL-6 may not play a major role in this type of proliferation. Gel shift assay indicated that TCPOBOP-induced hepatocyte proliferation is not associated with activation of STAT3 transcription factor, a major target of IL-6 and other growth factors/cytokines. Our results thus indicate that hepatocyte proliferation can be induced by at least two different pathways; compensatory regeneration being TNF and IL-6-dependent, and mitogen-induced direct hyperplasia which does not require TNF or IL-6.

Ploidy and nuclearity of rat hepatocytes after compensatory regeneration or mitogen-induced liver growth.

The distribution pattern of rat liver parenchymal cells of different ploidy classes was investigated in Wistar rats following cell proliferation induced by surgical partial hepatectomy (compensatory regeneration) or primary mitogens (direct hyperplasia). Animals were killed at 1, 2, 3, 4 and 15 days after the proliferative stimulus, and ploidy and nuclearity were measured using a computer-assisted imaging system in hepatocytes isolated by collagenase perfusion. Analysis of hepatocytes from animals undergoing regeneration after partial hepatectomy revealed a large increase in tetraploid and octoploid mononucleate cells. The most striking feature was the almost complete disappearance of binucleate cells (from 20% to < 1%) at 3 days after partial hepatectomy. On the contrary, when hepatocytes were analyzed after treatment with the mitogen lead nitrate, a high number of binucleate cells (40%) was observed. The increase that was maximal at 3 days after treatment occurred mainly in 4 x 2c and in 8 x 2c compartments. This resulted in an overall increase in the ratio of binucleate/mononucleate cells as well as in the ratio (8c + 16c):(2c + 4c). The cytological changes induced by lead nitrate were not reversible 2 weeks after treatment. Because a massive elimination of excess liver cells occurred by apoptosis during this time period, it appears that polyploid cells are not preferentially eliminated. The hepatic content of DNA at the end of the regression phase was similar to control values. However, because of the higher ploidy state, the number of cells present in the liver 2 weeks after treatment appears to be lower than that of controls (approximately -16%). When liver growth was induced by a single treatment with another mitogen, the peroxisome proliferator nafenopin, a slight increase in the ploidy state of the liver was observed; because of the shift towards higher ploidy classes (8c), the increase in DNA content observed 3 days after a single treatment with nafenopin (+21%) appears to be almost entirely justified by polyploidy rather than by a hyperplastic event.

Pas1 is a common lung cancer susceptibility locus in three mouse strains.

Inherited predisposition to lung cancer is a phenotypic trait shared by different mouse inbred strains that show either a high or an intermediate predisposition. Other strains are instead genetically resistant. The Pas1 locus is the major determinant of lung cancer predisposition in the A/J strain (Gariboldi et al. 1993). To define the determinants of susceptibility to lung tumorigenesis in the highly susceptible SWR/J and in the intermediately susceptible BALB/c mice, we analyzed (BALB/c x SWR/J)F2 and (BALB/c x C3H/He)F2 crosses by genetic linkage experiments. The present results provide unequivocal evidence that the same Pas1/+ allele that leads to lung cancer predisposition is shared by A/J, SWR/J, and BALB/c strains. The intermediate susceptibility of the BALB/c strain would result by interaction of Pas1 locus with lung cancer resistance loci.

Lack of significant promoting activity by benzene in the rat liver model of carcinogenesis.

The promoting activity of benzene on rat liver carcinogenesis was investigated. The chemical was tested for its ability to enhance the growth of preneoplastic foci, as detected by gamma-glutamyl transpeptidase (GGT) staining in diethylnitrosamine (DENA) initiated hepatocytes. Two weeks after receiving a single ip dose of 200 mg/kg DENA, F344 rats were given daily oral doses of 400 mg/kg benzene (5 d/wk) for 6 wk. At wk 3 after the experiment began, all animals underwent partial hepatectomy, and at wk 8 were sacrificed. Following benzene treatment, no variation in the liver/body weight ratio was observed. After scoring of foci in liver slides, no significant difference in foci number and area could be observed between rats treated with DENA plus benzene and rats treated with DENA alone. Practically no foci were observed in the liver of rats treated only with benzene. The lack of benzene promoting activity in the liver model is discussed.

Possible roles of nonparenchymal cells in hepatocyte proliferation induced by lead nitrate and by tumor necrosis factor alpha.

A single intravenous injection of lead nitrate (LN) to rats induces liver cell proliferation without causing cell necrosis (direct hyperplasia). We suggested that liver cell proliferation in this model may be triggered by the induction of liver tumor necrosis factor alpha (TNF-alpha). Because administration of TNF-alpha in vivo has been shown to induce proliferation of both parenchymal and nonparenchymal cells of the liver, we analyzed the temporal sequences of DNA synthesis in both cell populations following LN and recombinant TNF-alpha treatment by 5-bromo-2-deoxyuridine (BrdU) immunohistochemistry. The patterns of cell proliferation induced by these agents were further compared with those induced by a single dose of nafenopin (NAF), a direct mitogen which does not induce liver TNF-alpha messenger RNA (mRNA). In male Wistar rats given a single dose of LN (100 micromol/kg), BrdU incorporation of hepatocytes and nonparenchymal cells (Kupffer cells, endothelial cells and periportal nondescript cells) became evident 12 hours after the treatment. The labeling of all cell types reached a peak after 36 hours and declined thereafter. Rats given a single intravenous injection of human recombinant TNF-alpha (46 microg/rat) showed an increase of BrdU labeling in nonparenchymal cells after 24 hours, whereas the labeling of hepatocytes became evident at 36 hours. A single intragastric administration of NAF resulted in a rapid increase in the number of labeled hepatocytes with no substantial labeling of nonparenchymal cells. These results add further support to the notion that LN-induced liver cell proliferation is mediated by TNF-alpha, and suggest that different cell populations are involved in the initial proliferative response of the liver to mitogens, depending on the capacity of the mitogens to stimulate TNF-alpha production.

Effects of cell proliferation and cell death (apoptosis and necrosis) on the early stages of rat hepatocarcinogenesis.

An experiment was performed to investigate whether, during regression of the liver hyperplasia induced by a direct mitogen, apoptosis differentially affects replicated and non-replicated hepatocytes. After a single dose of the direct mitogen lead nitrate (LN), male Wistar rats were given repeated injections of tritiated thymidine, and were killed either 3 days (time of maximal hepatic DNA increase) or 15 days (complete regression of the hyperplasia) after mitogen treatment. Determination of liver DNA radioactivities and labelling indices (LIs) at the two time points revealed an approximately 40% loss in total liver DNA radioactivity, a 20% decrease in the specific activity of DNA, and a 20% reduction in the cell LI. Three days after LN administration 64% of the apoptotic bodies contained thymidine grains in their nuclear fragments. The results indicated that apoptosis affects both hepatocytes that replicated, and those that did not replicate, the former being slightly more sensitive. A second experiment was then performed to investigate whether and to what extent different types of cell death (apoptosis versus necrosis) influence the growth of hepatocytes initiated by a chemical carcinogen. Male Wistar rats were given a single dose of diethylnitrosamine, and 2 weeks thereafter either a single dose of LN, or a necrogenic dose of carbon tetrachloride (CCl4). Bromodeoxyuridine was next infused for 5 days, and some of the animals were killed at this time point, and others after an additional 3 weeks. Administration of CCl4 resulted in an increase in both the average size and the percentage area occupied by placental glutathione S-transferase-positive lesions. In contrast, administration of lead nitrate resulted in a strong reduction (50%) in the number of positive lesions with no remarkable change in the percentage area occupied by them. These differential effects occurred even though comparable LIs were observed in rats treated with the two agents. The results suggest that lead nitrate leads to a loss of initiated hepatocytes, due to the apoptosis that occurs during regression of the LN-induced hyperplasia.