Previous data obtained in vivo and in vitro suggest that both prostaglandins (PGs) and catecholamines may have a role in promoting hepatocyte proliferation, and PGE2 and PGF2 alpha have also been implicated as mediators of the mitogenic actions of epidermal growth factor (EGF) (and transforming growth factor alpha [TGF alpha]). We have studied the effects of PGs and norepinephrine on DNA synthesis in serum-free primary cultures of rat hepatocytes, and compared the PG effects with those of norepinephrine. PGE2, PGF2 alpha, PGD2, and the synthetic analog dimethyl-PGE2 markedly enhanced the DNA synthesis. A more quantitative analysis of the effects of PGE2 and PGF2 alpha on the DNA synthesis, in the presence and absence of EGF, indicated that these PGs interacted in an essentially multiplicative manner with the effect of EGF. The effects of PGE2 and PGF2 alpha showed almost complete additivity with the stimulation of DNA synthesis produced by maximally effective concentrations of norepinephrine. The data suggest a) that PGE2 and PGF2 alpha facilitate and synergize with, rather than mediate, the actions of EGF in hepatocytes, and b) that this effect of the PGs occurs by mechanisms that are at least partly distinct from those of norepinephrine.

Several agents that act through G-protein-coupled receptors and also stimulate phosphoinositide-specific phospholipase C (PI-PLC), including angiotensin II, vasopressin, norepinephrine, and prostaglandin (PG) F2alpha, activated the ERK1 (p44mapk) and ERK2 (p42mapk) members of the mitogen-activated protein (MAP) kinase family in primary cultures of rat hepatocytes, measured as phosphorylation of myelin basic protein (MBP) by a partially purified enzyme, immunoblotting, and in-gel assays. All these agonists induced a peak activation (two to threefold increase in MBP-phosphorylation) at 3-5 min, followed by a brief decrease, and then a sustained elevation or a second increase of the MAP kinase activity that lasted for several hours. Although all the above agents also stimulated PI-PLC, implicating a Gq-dependent pathway, the elevations of the concentration of inositol (1,4,5)-trisphosphate did not correlate well with the MAP kinase activity. Furthermore, pretreatment of the cells with pertussis toxin markedly reduced the MAP kinase activation by angiotensin II, vasopressin, norepinephrine, or PGF2alpha. In addition, hepatocytes pretreated with pertussis toxin showed a diminished MAP kinase response to epidermal growth factor (EGF). The results indicate that agonists acting via G-protein-coupled receptors have the ability to induce sustained activation of MAP kinase in hepatocytes, and suggest that Gi-dependent mechanisms are required for full activation of the MAP kinase signal transduction pathway by G-protein-coupled receptors as well as the EGF receptor.

While many observations indicate that prostaglandins may act as positive regulators of hepatocyte proliferation, the underlying mechanisms are not known. We have examined some of the signal pathways in the growth response induced by prostaglandins in hepatocytes, with particular focus on adenylyl cyclase and phosphoinositide-specific phospholipase C. Adult rat hepatocytes were cultured as primary monolayers in serum-free medium in the presence of EGF and insulin. PGE2 or PGF2 alpha (added 0-3 h after plating) enhanced the incorporation of [3H]-thymidine into DNA (measured at 50 h); at 100 microM the stimulation was about threefold PGI2 and PGD2 also showed significant but smaller stimulatory effects. No significant increase in the level of cyclic AMP (cAMP) was detected in response to any of the prostaglandins. Low concentrations of glucagon (0.1-10 nM), a potent activator of hepatic adenylyl cyclase, or 8-bromo-cAMP (0.1-10 microM) enhanced the DNA synthesis. When 8-bromo-cAMP was used in maximally effective concentrations, no further stimulation was obtained by combining it with glucagon, whereas the effects of PGE2 and 8-bromo-cAMP were completely additive. All the prostaglandins also showed additivity with the effect of glucagon on the DNA synthesis. PGE2, PGF2 alpha, PGI2, and PGD2 increased intracellular inositol-1,4,5-trisphosphate (InsP3), with a relative order of efficacy roughly corresponding to their activity as stimulators of DNA synthesis. Increases in cytosolic free Ca2+, as measured in single cells, were elicited in a majority of the hepatocytes by all these prostaglandins at 1 microM. Supramaximal concentrations of vasopressin, a strong activator of phospholipase C in hepatocytes, acted additively with PGE2 on the DNA synthesis. Pretreatment of the hepatocytes with a concentration of pertussis toxin that prevented the inhibitory effect of PGE2 on glucagon-induced cAMP accumulation did not abolish the ability of PGE2 to stimulate the DNA synthesis. The results do not support a role for adenylyl cyclase activation in the stimulatory effect of prostaglandins on hepatocyte growth. While the data are compatible with an involvement of phosphoinositide-specific phospholipase C in the growth-promoting effect of prostaglandins in cultured rat hepatocytes, they suggest this may not be the sole mechanism.

Previous studies suggest that catecholamines may be involved in the regulation of liver growth. Considerable evidence implicates alpha 1-adrenergic mechanisms in the initiation of hepatocyte proliferation, while the role of beta-adrenoceptors is less clear. We have examined further the adrenergic regulation of hepatocyte DNA synthesis, using primary monolayer cultures. In hepatocytes that were also treated with epidermal growth factor and insulin, epinephrine or norepinephrine added early after the seeding strongly accelerated the rate of S phase entry. The beta-adrenergic agonist isoproterenol and the alpha-adrenergic agonist phenylephrine also stimulated the DNA synthesis, but were less efficient than epinephrine and norepinephrine. Experiments with the alpha 1-receptor blocker prazosine and the beta-receptor blocker timolol showed that the stimulatory effect of norepinephrine consisted of both an alpha 1- and a beta-adrenergic component. The alpha 1-component was most prominent in terms of maximal response at high concentrations of the agonist, but the beta-component contributed significantly and predominated at low concentrations (less than 0.1 microM) of norepinephrine. At later stages (about 40 h) of culturing norepinephrine strongly but reversibly inhibited the cells, acting at a point late in the G1 phase. This inhibition was mimicked by isoproterenol and abolished by timolol but was unaffected by prazosine, suggesting a beta-adrenoceptor-mediated effect. The results confirm the alpha 1-adrenoceptor-mediated stimulatory effect, but also show that beta-adrenoceptors may contribute to the growth stimulation by catecholamines. Furthermore, catecholamines, via beta-adrenoceptors and cyclic AMP, inhibit the G1-S transition, and may thus play a role in the termination of hepatic proliferation.

Data presented indicate that in hepatocytes insulin and glucagon promote growth by acting in a relatively early part of the prereplicative period (G0 or early G1) whereas cells (if pretreated with insulin) become more sensitive to EGF at the later stages, ie, nearer the S phase entry. The data indicate that at least two effects of glucagon (cAMP) on hepatocyte proliferation exist; in addition to a growth-promoting modulation early in the prereplicative period, there is also an inhibitory effect of glucagon (as well as other cAMP-elevating agents) that is exerted at a point shortly before the G1-to-S transition. Because both effects occur dose-dependently in the normal range of glucagon concentrations in portal blood, it is conceivable that glucagon/cAMP is involved both when liver growth is initiated and terminated.