Search

The cells in nearly pure (96-98%) primary cultures of hepatocytes from neonatal rat liver in high (1.0 mM)-Ca2+, serum-free, synthetic HiWo5Ba2000 medium initiated DNA synthesis and entered mitosis between 11 and 30 h after the addition of 10 ng/ml EGF. During the 10-h prereplicative period, the cultured hepatocytes, like regenerating rat liver cells, generated two large cyclic AMP transients, one peaking between 30 min and 2 h and the other around 6 h. Hepatocytes stimulated by the same concentration of EGF in low (0.02 mM)-Ca2+ medium increased cyclic AMP synthesis as much as the EGF-treated hepatocytes in high-Ca2+ medium, but they released the additional cyclic AMP into the medium and could not generate prereplicative internal cyclic AMP surges, initiate DNA replication, or enter mitosis. These results suggest that one of the ways external Ca2+ controls prereplicative development of hepatocytes is to restrain the release of cyclic AMP and thus enable the cell to accumulate enough internal cyclic AMP to stimulate events required to initiate DNA replication.

We have identified in mammalian cells a novel cyclic AMP response element (CRE)-binding protein of molecular mass 47 kDa. This protein was not recognized by either the CREB-327/341 or c-Jun antisera, and its tissue distribution did not overlap with those of the CREB and Jun families. For example, hepatoma and placental tissue did not contain the 47-kDa DNA-binding protein, but did contain the CREB isoforms. On the other hand, S49 lymphoma cells contained a high level of the 47-kDa DNA-binding protein but did not contain a 47-kDa Jun-related protein which was found in normal liver and hepatoma. This new 47-kDa factor bound to the CRE in the dephosphorylated form, and phosphorylation of the protein by the catalytic subunit of protein kinase A completely abolished its DNA-binding activity. The isoforms of the CREB-327/341 family, on the other hand, bound to DNA in the phosphorylated form, and alkaline phosphatase treatment reduced significantly their interaction with CRE sequence. This reverse effect of phosphorylation/dephosphorylation on the DNA-binding property of this new 47-kDa protein in particular distinguishes it from other known CREB factors and suggests that the protein might play a unique role in the regulation of cAMP-mediated transcription.

Using a combination of immunoblotting, double immunoprecipitation, immunoglobulin-affinity chromatography, and isoelectrofocusing, we have been able to identify a group of proteins that display CDP-reductase activity and contain antigenic epitopes recognized by anti-ribonucleotide reductase M1 subunit and anti-ubiquitin antibodies. In the cytoplasm of rat liver cells, we could detect a total of five proteins with molecular masses of 92, 89, 56, 45, and 37 kilodaltons which reacted with the anti-M1 subunit serum. All of them, except the 89-kilodalton protein (the nascent unmodified M1), were also recognized by the anti-ubiquitin antibody. In normal liver cells, all of the apparently ubiquitinated species of the M1 protein were found in the cytoplasm, but not in the nuclear envelope associated pool of the enzyme. However, we did not detect ubiquitinated M1 protein fragments in the cytoplasm of Morris hepatoma 5123tc. The level of the apparently ubiquitinated fragments of the M1 subunit increased in parallel to the DNA-synthetic activity of normal liver cells, suggesting that ubiquitination plays a key role in the regulation of the activity of the enzyme during the cell cycle.

We applied Southwestern and Western blotting and gel retardation techniques to investigate the changes that occur in the cyclic adenosine monophosphate (cAMP)-responsive element (CRE) binding (CREB) proteins in rapidly growing, chemically induced 5123tc and 5123D Morris hepatomas. Using the CRE sequences from the c-fos, E2A, and somatostatin gene promoters, we identified in the nuclear proteins from normal unstimulated or proliferating rat liver cells six different protein factors of Mr 34,000, 36,000, 40,000, 47,000, 56,000, and 72,000 capable of binding to the element. The Mr 47,000 protein had the highest specificity for the core CRE, suggesting its importance in cAMP-mediated gene expression. We could not find the Mr 47,000 CREB protein in the 5123tc and 5123D hepatomas. Our efforts to detect this protein in the tumors by (a) using the CRE sequence from different gene promoters, (b) altering the protocol for extracting nuclear proteins, or (c) attempting to restore its DNA-binding property by phosphorylation [with endogenous protein kinase(s), a catalytic subunit of cAMP-dependent protein kinase, and protein kinase C/dephosphorylation (with alkaline phosphatase)] were unsuccessful. The loss of tje Mr 47,000 CREB protein from solid tumors of the Morris hepatoma is likely to be related to the neoplastic properties of the tumor cell rather than to cell growth because the level of this protein remained unchanged during a 6-day period of liver regeneration. The nuclear extract from the Morris hepatoma that did not have the Mr 47,000 CRE-binding factor contained proteins immunologically related to the CREB, c-Jun, and c-Fos proteins. We conclude that the Mr 47,000 factor represents a distinct member of the CRE-binding protein family and that its absence from the hepatomas may lead to aberrant expression of cAMP-inducible genes.

Partial hepatectomy in the rat stimulated the phosphorylation of three proteins (Mr 100,000, 48,000, and 35,000) in the plasma membranes of the proliferatively activated cells of the liver remnant. The autophosphorylation of these plasma membrane proteins began to rise about 8 hours after surgery, peaked at 14 hours, and then returned to the original low level by 24 hours. This increase in autophosphorylation was not evident in isolated plasma membranes from other proliferatively activated liver cells, such as those in fetal liver or hepatomas. The protein kinase responsible for the phosphorylation of the three membrane proteins was activated by calcium but appeared both cyclic-AMP- and calmodulin-independent.

Low concentrations (e.g. 10(-12) and 10(-11) mol/l) of imidazole and indomethacin strongly stimulated DNA synthesis and mitosis of hepatocytes in 4-day-old primary cultures of neonatal rat liver. These agents seem to have acted by inducing quiescent hepatocytes to begin cycling rather than by affecting already cycling cells, because they did not shorten the total cell cycle time. Neither compound stimulated DNA synthesis by hepatocytes cultured in low (0.010 mol/l) calcium medium. Nevertheless, hepatocytes in calcium-deficient medium must have been mitogenically activated by these compounds and, hence, been able to reach a late stage of prereplicative development because they did initiate DNA synthesis very soon after the addition of calcium.

A protein kinase was purified from rat liver nuclei by affinity chromatography on poly(L-lysine)-agarose and protein kinase inhibitor-Sepharose 4B columns. The relative molecular weight of this protein kinase is 105000 (determined by 10% SDS-polyacrylamide slab gel electrophoresis, gel filtration on Sephadex G-200 and sedimentation velocity ultracentrifugation). Its pH optimum is between 8.0 and 9.0. It is active over a wide range of mg2+ concentrations, and its activity is stimulated by several small acidic proteins (calmodulin, lactalbumin, parvalbumin, protein kinase inhibitor and troponin C). The enzyme phosphorylates a variety of substrates including casein, histones, protamine and the synthetic basic polypeptides, poly(L-arginine) and poly(L-lysine).

Inhibitors of prostaglandin and thromboxane production such as mefenamic acid, hydrocortisone, and dexamethasone prevented a large proportion of the parenchymal cells of rat liver, proliferatively activated by a two-thirds partial hepatectomy, from entering mitosis without preventing them from initiating or completing DNA synthesis. This specific mitosis-inhibiting action was maximum when the drugs were present during the first few hours after partial hepatectomy. In contrast, indomethacin, another inhibitor of prostaglandin and thromboxane production, maximally inhibited both DNA synthetic and mitotic activities when present during the same early period of prereplicative development, which showed that it had an action not shared by the other inhibitors. Arachidonic acid completely reversed the hydrocortisone- or dexamethasone-induced inhibition of mitotic activity, but it did so only when it was injected between 2 and 3 hours after partial hepatectomy and glucocorticoid injection. Arachidonate's reversal of the hydrocortisone-induced inhibition did not occur in the presence of mefenamic acid. These observations indicate that there is an early, prostaglandin- or thromboxane-mediated, prereplicative process in proliferatively activated hepatocytes which determines their later entry into mitosis and which is separate from the early events leading to DNA synthesis.

Between 16 and 20 h after partial (70%) hepatectomy (HPX) in normal rats, the remaining liver cells accumulate ribonucleotide reductase subunits, assemble these into active holoenzyme, and initiate DNA replication. These late prereplicative activities did not occur in most of the liver cells remaining after HPX in rats which had been thyroparathyroidectomized (TPTX) 72 h previously. However, one intraperitoneal injection of 400 or 600 ng 1 alpha,25-dihydroxyvitamin D3/100 g body weight at the time of HPX enabled the remaining liver cells in such TPTX rats to make functional ribonucleotide reductase subunits, assemble these subunits into active CDP-reducing holoenzymes, and replicate their DNA, though they started to do so 4 to 16 h later than in normal animals.

The level of the cytoplasmic ribonucleotide reductase nonheme-iron-containing L2 subunit in regenerating rat liver cells began rising about 2 h before the onset of DNA synthesis, rose sharply to a maximum level about 4 h before the DNA-synthetic activity reached its peak, and then stayed at this high level even after the cells had finished replicating their DNA. The cytoplasmic level of the CDP-specific, effector-binding L1 subunit and the holoenzyme activity began rising together about 2 h after the L2 subunit began increasing and at the same time as the DNA-synthetic activity, but subsequently rose much more slowly than the L2 subunit and continued rising even after the cells had finished making DNA. The nuclear level of the L2 subunit did not rise in the regenerating liver cells, but the nuclear level of the L1 subunit and the holoenzyme activity began rising together about the same time as the DNA-synthetic activity, peaked briefly 4-6 h before the peak DNA-synthetic activity, and dropped sharply back to the basal levels by the time the DNA-synthetic activity reached its peak, but then rose again slowly as the cells finished making DNA. Thyroparathyroidectomy 72 h before partial hepatectomy prevented the cytoplasmic and nuclear subunits and holoenzyme activity from rising and prevented most of the remaining liver cells from initiating DNA synthesis.

Two nonidentical subunits of mammalian ribonucleotide reductase, L1 and L2, from regenerating rat liver have been extensively purified for the first time. They were separated by dATP-Sepharose affinity chromatography. Subunit L1, which bound to dATP-Sepharose, was eluted with 50 mM ATP and purified to homogeneity (as demonstrated by sodium dodecyl sulfate (SDS) - polyacrylamide gel electrophoresis) by molecular exclusion high-pressure liquid chromatography (HPLC). This subunit had an apparent relative mass (Mr) of 45 000 and a Km of 0.9 X 10(-4) for CDP. Subunit L2, which did not bind to dATP-Sepharose, was purified by pH 5.2 precipitation followed by chromatography on CM-Sephadex, molecular exclusion HPLC, and DEAE-cellulose. This subunit contained iron and had an apparent Mr of 120 000 by HPLC molecular exclusion chromatography, but showed two bands (Mr 75 000 and Mr 47 000) on SDS-polyacrylamide gel electrophoresis. Neither L1 nor L2 separately had any enzyme activity but when combined they reduced CDP to dCDP.

Conditions have been found for the isolation of rat liver nuclei which maintain chromatin in its native state and suppresses endogenous nuclease activity. Chromatin prepared in this way can be dispersed into buffers containing various concentrations of monovalent or divalent cations so that the 30-nm fiber is either totally or partially decondensed. When probed with micrococcal nuclease, the digestion profiles show that although the 30-nm fiber can be cleaved periodically to generate superbead-like particles this only occurs under certain ionic conditions when the fiber is partially decondensed. It is likely that this cleavage pattern reflects the transient exposure of specific nuclease sensitive sites as the 30-nm fiber condenses, rather than the existence of a specific subunit of a beaded 30-nm fiber. The periodicity of these nuclease-sensitive sites appear to be related to the asymmetric distribution of histone H1 molecules along the length of the fiber.

The G0----G1 and G1----S transitions (but not the intervening events) in the G1 phase of T51B rat liver epithelial cells in serum-stimulated confluent cultures required a high concentration of extracellular Ca2+ and were accompanied or immediately preceded by increases in the amount of EDTA-extractable protein kinase C, a Ca2+/phospholipid-dependent enzyme. Involvement of this Ca2+-dependent enzyme in the two Ca2+-dependent transitions was further indicated by the facts that 12-O-tetradecanoyl phorbol-13-acetate (TPA), a compound that stimulated protein kinase C from T51B cells even in the absence of Ca2+, enabled these cells to transit G1 in Ca2+-deficient medium, while a TPA analogue (4 alpha-phorbol-12, 13-didecanoate (4 alpha-PDD) that did not stimulate the enzyme in cell-free preparations did not promote G0----G1 or G1----S transit in Ca2+-deficient medium.

Although they are proliferatively quiescent, the cells in the intact adult rat liver express the gene coding for the M1 subunit of ribonucleotide reductase. But since they do not need deoxyribonucleotides, they promptly inactivate the 88 to 90 kDa M1 products and degrade them into 40 kDa fragments. Partial hepatectomy signals the remaining cells to start proliferating. Two hours before the onset of DNA replication, around 16 to 18 hr after partial hepatectomy, the cells start accumulating a large pool of functional ribonucleotide reductase M2 subunits. Near the end of the G1 build-up the cells step up M1 gene expression, stop inactivating, and reduce the degradation of the M1 products. The accumulating functional 88 to 90 kDa M1 subunits, each with more than one catalytic site, couple with functional M2 subunits to produce active ribonucleotide reductase holoenzyme which accumulates in the outer nuclear membrane from which they supply deoxyribonucleotide precursors to intranuclear replication enzymes. At the end of the S phase, the cell reduces M1 gene expression and resumes degrading 88 to 90 kDa M1 subunits. At least some of the 40 kDa M1 fragments are still active and can form partially active "holoenzymes" when mixed with a standard preparation of functional M2 subunits. The M1 control mechanism appears not to operate in hepatoma cells and Ehrlich ascites tumor cells, both of which maintain a pool of undegraded 88 to 90 kDa M1 components.

Calcium and cyclic AMP control the proliferation of nontumorigenic liver cells. These agents seem to be cogenerators of a signal to start synthesizing deoxyribonucleotides, the earliest of the DNA synthetic processes. By contrast, calcium has little or no effect on the proliferation of tumorigenic cells in vitro. Some possible reasons for this loss of control are presented, and the usefulness of this property as a tool for the detection of carcinogens is discussed.

Diacylglycerols, such as 1,2-diolein, and tumor-promoting phorbol compounds, such as TPA (12-0-tetradecanoyl phorbol-13-acetate), stimulate the Ca2+/phospholipid-dependent protein kinase C from T51B rat liver cells, probably by sensitizing it to activation by Ca2+, and they reduce the liver cells' content of EDTA-extractable (i.e., soluble) protein kinase C activity. Evidence is presented that indicates that the glucocorticoid, dexamethasone, and the tumor-promoting artificial sweetener, saccharin, also trigger a Ca2+-dependent increase in the activity of the protein kinase C from T51B liver cells and reduce the cells' content of EDTA-extractable protein kinase C activity. However, these novel stimulators do not activate the enzyme by binding to the same site as diacylglycerols and TPA, although they do alter this site as indicated by an increase in the binding of the TPA analogue PDBu (phorbol 12,13-dibutyrate).

A dual-labelling technique has been used to establish that partial hepatectomy has no effect on the degradation of poly (A)+mRNA and confirms that the increased incorporation of precursor into mRNA during early prereplicative development reflects an actual increase in mRNA biosynthesis. Simultaneous studies on the changes in nuclear RNA metabolism support the conclusion that an increase in gene transcription does occur. Colchicine, at concentrations known to disrupt microtubules, has no effect on this increase in gene transcription but prevents the translation of the gene products by promoting polysome disaggregation transiently during a critical stage of prereplicative development. Studies with mefenamic acid and hydrocortisone, specific inhibitors of prostaglandin metabolism, have ruled out any involvement of prostaglandins in the induction of prereplicative mRNA synthesis.

The regulatory subunits of cytoplasmic and nuclear cyclic AMP-dependent protein kinases increased, while the holoenzyme activities fell in the livers of thyroparathyroidectomized (TPTX) rats. 1 alpha,25-dihydroxyvitamin D3 reversed this TPTX-induced decline in holoenzyme activities. TPTX also prevented the increase in holoenzyme activities, but not the increase in the concentration of regulatory subunits that normally occurs in partially hepatectomized animals. An injection of 1 alpha, 25-dihydroxyvitamin D3 into these TPTX animals, at the time of partial hepatectomy restored the increases in holoenzyme activities, without affecting the changes in regulatory subunits. Thus, the synthesis of regulatory and catalytic subunits can be uncoupled by TPTX and 1 alpha, 25-dihydroxyvitamin D3 appears to selectively control the activity of C subunits.

Non-neoplastic T51B rat liver epithelial cells cannot proliferate in Ca2+-deficient medium. This proliferative inhibition in Ca2+-deficient medium is accompanied by a large reduction in the amount of cellular EDTA-extractable Ca2+/phospholipid-dependent protein kinase (protein kinase C) activity. By contrast, tumorigenic epithelial cells from several Morris hepatomas proliferate in Ca2+-deficient medium and either maintain or greatly increase their content of EDTA-extractable protein kinase C.

Cells from neonatal rat livers were unable to maintain DNA-synthetic activity in calcium-deficient medium, but neoplastic hepatocytes from Morris hepatomas 5123 tc and 7795 synthesized DNA and proliferated indefinitely in this calcium-deficient medium. The calcium content of fresh hepatoma tissue from which these cultures were derived was as much as 10 times greater than that of normal liver; but this difference could not account for the insensitivity of neoplastic cells to extracellular calcium because it disappeared during subsequent cultivation in vitro.

There is a marked increase in the concentration of putrescine during the first ten hours following partial hepatectomy in rats. The concentration of spermidine also increases but to a smaller degree. Putrescine levels return to normal between 10 and 24 hours after the operation, whereas the increased spermidine level is maintained. The production of putrescine and spermidine appears to be initiated by the induction of ornithine decarboxylase which shows a single peak of activity at four hours after hepatectomy. The activity of S-adenosylmethionine decarboxylase shows little change following hepatectomy. The changes in polyamine levels and the activities of the enzymes of polyamine metabolism are not affected by thyroparathyroidectomy 72 hours prior to hepatectomy. Thus although these hypocalcemic conditions considerably reduce and delay DNA synthesis and mitosis, the prereplicative changes in polyamine metabolism still occur. These data suggest that the hepatocytes in hypocalcemic animals have become activated and moved to an advanced stage of prereplicative development before being blocked.

Extracellular Ca2+-deprivation inhibited the proliferation of normal T51B rat liver cells and reduced (2-4 fold) the amount of EDTA/EGTA-extractable protein kinase C activity. By contrast, preneoplastic and neoplastic T51B rat liver cells were able to proliferate in Ca2+-deficient medium and retained all of their extractable protein kinase C activity.

A low concentration (10(-11) mol/l) of epidermal growth factor (EGF) and/or an equimolar (10(-14) mol/l) mixture of glucagon and insulin stimulated DNA synthesis in hepatocytes in 4-day-old primary cultures of neonatal rat liver. EGF seems to have acted by inducing quiescent hepatocytes to begin cycling, while the glucagon-insulin combination seems to have acted mainly by shortening the cell cycle time. Incubation in low calcium medium blocked untreated hepatocytes in the G1 phase of their cycle and prevented EGF and the glucagon-insulin mixture from stimulating DNA synthesis. Nevertheless, hepatocytes in calcium-deficient medium did respond to these agents, as they reached a late stage of prereplicative development before being blocked: in fact, they initiated DNA synthesis soon after the addition of calcium. EGF, but not the glucagon-insulin combination, also enabled the already cycling hepatocytes (but not the newly activated ones) to overcome the block imposed by the extracellular calcium deficiency after a delay of several hours.

Confluent T51B rat liver epithelial cells promptly began accumulating cyclic AMP-binding sites on their surfaces when they were stimulated from quiescence by serum growth factors in medium containing 1.8 mM Ca2+, but they began losing the accumulated binding sites shortly before initiating DNA replication. When the medium contained only 0.02 mM Ca2+, the cells still accumulated surface cyclic AMP-binding sites, but they did not initiate DNA replication and tended to continue accumulating the binding sites. The cyclic AMP-binding sites were eliminated completely by treating intact cells for 5 minutes with 0.005% trypsin (which did not damage the cells), and cyclic AMP caused them to be released from intact, undamaged cells into the medium. The binding sites also comigrated electrophoretically with purified regulatory subunits of type I cyclic AMP-dependent protein kinase, and to a lesser extent the regulatory subunit of type II cyclic AMP-dependent protein kinase. Therefore, it is likely that a transient accumulation of cyclic AMP-dependent protein kinases on the outer surface of the plasma membrane is part of the T51B rat liver cell's prereplicate program.

Digestion of nuclei from normal or partially hepatectomised rat livers with endogenous nucleases liberated a pool of cryptic DNA polymerase-alpha activity which had previously gone unrecognised. Most of this activity is released into the supernatant as free enzyme molecules (11S), but a small fraction of it is released as a complex of 16S (probably with DNA primase). About 40% of the enzyme remains in the pellet, which contains undigested chromatin and components of the residual nuclear matrix and nucleoskeletal structures. Virtually all of this remaining activity is extracted by 2.0 M salt. The activity of DNA polymerase-alpha increases equally in all nuclear fractions during the period of DNA replication in regenerating rat liver. Lowering of the serum calcium level by thyroparathyroidectomy does not affect basal DNA polymerase-alpha activity, but prevents induction of the enzyme during the later stages of prereplicative development. However, an injection of 1 alpha,25-dihydroxy vitamin D3 into the rat during the first 6 h after parital hepatectomy restores the ability of the hepatocytes to induce DNA polymerase-alpha activity and initiate DNA synthesis. These results are discussed in terms of the role of calcium ions in the regulation of the critical stage of prereplicative development which commits the cells to DNA replication.