Hyperoxic-induced alterations in lung cell structure and function. Effects on cellular cyclic AMP content and comparison to alterations produced by exogenous cyclic AMP

Hyperoxic exposure in vitro of two lung-derived cell types (the epithelial-derived L2 cells and WI-38 fibroblasts) inhibits cellular replication, produces striking morphologic changes and may result in cell death; these effects have been observed consistently in other cell types. Hyperoxic exposure of L2 cells is associated with an increase in cellular cyclic AMP content (cellular cyclic AMP content 454 +/- 115 fmol/micrograms DNA in cells exposed to pO2 677 Torr for 96 h compared to 136 +/- 17 fmol/microgram DNA in air-grown cells). Hyperoxic exposure of WI-38 fibroblasts is not associated with increased cyclic AMP content. Although cultivation of L2 cells in the presence of exogenous dibutyryl cyclic AMP does inhibit replication and produce morphologic alterations, similar effects are produced by sodium butyrate alone. Hyperoxic exposure alters cyclic AMP metabolism in some cell types, but the structural and functional alterations observed in L2 cells and WI-38 fibroblasts following hyperoxic exposure are not produced by changes in cellular cyclic AMP content.     

Effect of cyclic AMP on chromatin-bound protein kinases in WI-38 fibroblasts stimulated to proliferate.

When resting confluent monolayers of WI-38 fibroblasts are stimulated to proliferate by serum, DNA synthesis begins to increase between 15-18 h after stimulation. Chromatin-bound protein kinase activity increases in stimulated cells within 1 h after the nutritional change, concomitant with an increase in the template activity of nuclear chromatin. Addition of dibutyryl 3' : 5'-cyclic adenosine monophosphate (dibutyryl cyclic) AMP to the stimulating medium inhibits the entrance of cells into S phase, but only if dibutyryl cyclic AMP (5-10(-4) M) is added before the onset of DNA synthesis. The increases in chromatin template activity and in the chromatin-bound kinase activity are not inhibited by dibutyryl cyclic AMP in the early hours after stimulation, but are completely inhibited after the 5th hour from the nutritional change. This seems to indicate that in stimulated WI-38 cells, dibutyryl cyclic AMP exerts its inhibitory action somewhere between 5 and 12 h after stimulation. A number of protein kinase activities were extracted from chromatin with 0.3 M NaCl and partially resolved on a phosphocellulose column. Two distinct peaks of protein kinase activity appeared to be markedly increased in WI-38 cells 6 h after serum stimulation. Both peaks of increased activity were inhibited by dibutyryl cyclic AMP in vivo. Adenosine, sodium butyrate and adenosine 5'-monophosphate (AMP) do not inhibit the increase in DNA synthesis nor the increase in protein kinase activity. The results suggest that stimulation of cell proliferation in confluent monolayers of WI-38 cells causes an increase (or the new appearance) of certain chromatin-bound protein kinases, and that this increase is inhibited by cyclic AMP in vivo.     

Prostaglandin E1 induced changes in cyclic AMP levels in WI-38 human fibroblasts: Two different effects of cell density

WI-38 lung diploid fibroblasts repond to protaglandin E₁ with increased levels of cyclic adenosine 3′:5′-monophosphate. This increase is affected by cell density in two ways: (a) The initial rate of accumulation of intracellular cyclic AMP increases with increasing cell density. (b) However, the elevated levels of cyclic AMP are more stably maintained in lower-density cells, and this stability decreases with increasing cell density. Cyclic AMP phosphodiesterase activities, as well as the efflux of intracellular cyclic AMP into the medium are simelar at all cell densities.     

Prolonged prostaglandin E1 stimulation of cyclic AMP production in transformed and normal WI-38 fibroblasts

Summary Long-term (48-hr) incubations of either the fibroblast strain WI-48 or its SV40-transformed counterpart, WI-38-VA13-2RA, in growth medium containing 1 μm prostaglandin E₁ (PGE₁) resulted in a sustained production and release of cyclic AMP from the cells into the medium. Despite the steady production, intracellular levels of the nucleotide decreased, reaching steady-state values within 4 hr of the initial exposure to PGE₁. These values were maintained for the remainder of the 48-hr experimental period. The steadystate levels of intracellular cyclic AMP were higher than those observed in unstimulated cells, and cyclic AMP-dependent protein phosphokinase was in a highly activated state as compared to controls. Under these conditions little change in the growth or morphology of either the normal or transformed cells was observed. In contrast, inhibition of growth, apparent cell death, and unusual morphological changes were observed in both normal and transformed cells when high concentrations of either PGE₁ (10 μm) or the phosphodiesterase inhibitor 1-methyl, 3-isobutylxanthine (0.5mm to 2mm) were used, which was indicative of toxic effects of the drugs. It was concluded that cyclic AMP-mediated activation of protein phosphokinase does not completely inhibit growth in WI-38 cells or restore normal growth and morphology to the SV40-transformed cells. This work was supported by Grants AM 13904 and CA 21612 from the National Institutes of Health, Department of Health, Education and Welfare.     

Effects of delta 1-tetrahydrocannabinol on cyclic AMP in cultured human diploid fibroblasts.

(-)-trans-delta 1-Tetrahydrocannabinol (delta 1-THC) antagonized the cyclic AMP responses of WI-38 fibroblasts to both prostaglandin E1 (PGE1) and catecholamines. Both cellular cyclic AMP accumulation and cyclic AMP escape to the incubation medium were reduced, but the reduction of escape was much more dramatic at all concentrations of the drug. Conversely, long term incubations of cells with delta 1-THC alone resulted in substantial accumulations of cyclic AMP in the incubation medium. This effect was potentiated by the phosphodiesterase inhibitor 1-methyl, 3-isobutylxanthine and appeared to result from weak agonist activity of the cannabinoid as determined by a) stimulation of radioactivity incorporated into cyclic AMP using 3H-adenine prelabelled cells, and b) a rapid and pronounced increase in the activity ratio of cellular protein kinase. The antagonistic effect of delta 1-THC on the cellular response to PGE1 was greater in preconfluent cells than in confluent monolayers. Further, the increased sensitivity of preconfluent cultures to delta 1-THC was associated with the appearance of cytoplasmic vacuoles in the perinuclear region of the cells. Cannabidiol acted similar to delta 1-Thc in affecting cyclic AMP metabolis whereas cannabinol and cannabicyclol showed mixed effects on the various parameters studied.     

Prostaglandin E1 induced changes in cyclic AMP levels in WI-38 human fibroblasts. Two different effects of cell density.

WI-38 lung diploid fibroblasts respond to protaglandin E1 with increased levels of cyclic adenosine 3':5'-monophosphate. This increase is affected by cell density in two ways: (a) The initial rate of accumulation of intracellular cyclic AMP increases with increasing cell density. (b) However, the elevated levels of cyclic AMP are more stably maintained in lower-density cells, and this stability decreases with increasing cell density. Cyclic AMP phosphodiesterase activities, as well as the efflux of intracellular cyclic AMP into the medium are similar at all cell densities.     

Prolonged prostaglandin E1 stimulation of cyclic AMP production in transformed and normal WI-38 fibroblasts.

Long-term (48-hr) incubations of either the fibroblast strain WI-38 or its SV40-transformed counterpart, WI-38-VA13-2RA, in growth medium containing 1 micron prostaglandin E1 (PGE1) resulted in a sustained production and release of cyclic AMP from the cells into the medium. Despite the steady production, intracellular levels of the nucleotide decreased, reaching steady-state values within 4 hr of the initial exposure to PGE1. These values were maintained for the remainder of the 48-hr experimental period. The steady-state levels of intracellular cyclic AMP were higher than those observed in unstimulated cells, and cyclic AMP-dependent protein phosphokinase was in a highly activated state as compared to controls. Under these conditions little change in the growth or morphology of either the normal or transformed cells was observed. In contrast, inhibition of growth, apparent cell death, and unusual morphological changes were observed in both normal and transformed cells when high concentrations of either PGE1 (10 micron) or the phosphodiesterase inhibitor 1-methyl, 3-isobutylxanthine (0.5 mM to 2 mM) were used, which was indicative of toxic effects of the drugs. It was concluded that cyclic AMP-mediated activation of protein phosphokinase does not completely inhibit growth in WI-38 cells or restore normal growth and morphology to the SV40-transformed cells.     

Protective effect of cyclic AMP against cisplatin-induced nephrotoxicity

We reported earlier that reactive oxygen species are implicated in necrotic injury induced by a transient exposure of cultured renal tubular cells to a high concentration of cisplatin but not in apoptosis occurring after continuous exposure to a low concentration of cisplatin. We report here the protective effect of cyclic AMP against cisplatin-induced necrosis in cultured renal tubular cells as well as cisplatin-induced acute renal failure in rats. Several pharmacological agents that stimulate cyclic AMP signaling, including the nonhydrolyzable cyclic AMP analogue dibutyryl cyclic AMP, forskolin, 3-isobutyl-1-methylxanthine, and a prostacyclin analogue, beraprost, prevented cisplatin-induced cell injury in a protein kinase A-dependent manner. Cisplatin enhanced lipid peroxidation, decreased CuZn superoxide dismutase (SOD) while enhancing MnSOD activity, and increased cellular tumor necrosis factor-alpha (TNF-alpha) content. The elevation of TNF-alpha content and cell injury induced by cisplatin were attenuated by p38 mitogen-activated protein kinase (MAPK) inhibitors including SB203580 and PD169316. Indeed, cisplatin increased the number of phosphorylated p38 MAPK-like immunoreactive cells. These intracellular events were all reversed by antioxidants such as N-acetylcysteine (NAC) and glutathione or cyclic AMP analogues. The in vivo acute renal injury after cisplatin injection was associated with the elevation of renal TNF-alpha content. The cisplatin-induced renal injury and the increase in TNF-alpha content were reversed by NAC or beraprost. These findings suggest that cyclic AMP protects renal tubular cells against cisplatin-induced oxidative injury by obliterating reactive oxygen species and subsequent inhibition of TNF-alpha synthesis through blockade of p38 MAPK activation.     

Cell cycle regulation of DNA repair in normal and repair deficient human cells

The regulation of nucleotide excision repair and base excision repair by normal and repair deficient human cells was determined. Synchronous cultures of WI-38 normal diploid fibroblasts and Xeroderma pigmentosum fibroblasts (complementation group D) (XP-D) were used to investigate whether DNA repair pathways were modulated during the cell cycle. Two criteria were used: (1) unscheduled DNA synthesis (UDS) in the presence of hydroxyurea (HU) after exposure to UV light or after exposure to N-acetoxy-acetylaminofluorene (N-AcO-AAF) to quantitate nucleotide excision repair or UDS after exposure to methylmethane sulfonate (MMS) to measure base excision repair; (2) repair replication into parental DNA in the absence of HU after exposure to UV light. Nucleotide excision repair after UV irradiation was induced in WI-38 fibroblasts during the cell cycle reaching a maximum in cultures exposed 14–15 h after cell stimulation. Similar results were observed after exposure to N-AcO-AAF. DNA repair was increased 2–4-fold after UV exposure and was increased 3-fold after N-AcO-AAF exposure. In either instance nucleotide excision repair was sequentially stimulated prior to the enhancement of base excision repair which was stimulated prior to the induction of DNA replication. In contrast XP-D failed to induce nucleotide excision repair after UV irradiation at any interval in the cell cycle. However, base excision repair and DNA replication were stimulated comparable to that enhancement observed in WI-38 cells. The distinctive induction of nucleotide excision repair and base excision repair prior to the onset of DNA replication suggests that separate DNA repair complexes may be formed during the eucaryotic cell cycle.     

Effects of prostaglandins and other drugs on the cyclic AMP content of cultured bone cells

Prostaglandins of the E-series (PGE₁ and PGE₂) may be involved in disease-related, localized loss of bone. E-prostaglandins increase the cyclic AMP content of many cells; and, to determine if their effects on bone are mediated by cyclic AMP, we examined the effects of E-prostaglandins and of other agents on the cyclic AMP content of cultured bone cells. PGE₂ produced a rapid, marked and dose-related increase in the cyclic AMP content of confluent monolayers of bone cells isolated from newborn rat calvaria. At 2.8 × 10⁻⁶ M, PGE₁ and PGE₂ had approximately the same effect, while the effect of PGF_2α was much less pronounced. In the presence of theophylline, PGE₂ had a more marked effect than parathyroid hormone (PTH) and the combination of PGE₂ and PTH had a synergistic effect. The divalent, cationic, ionophore, A23187, produced an increase in cellular cyclic AMP and had an additive effect in combination with PGE₂. Synthetic salmon calcitonin (CT), which inhibits the bone resorptive effect of PGE₂, increased cellular cyclic AMP and had an additive effect in combination with PGE₂. A prostaglandin antagonist, SC-19220, partially inhibited the resorptive effect of PGE₂ and reduced its effect on cellular cyclic AMP. The calcium antagonist, D600, inhibited the bone resorptive effects of PGE₂ but had no effect on increased cellular cyclic AMP produced by PGE₂.The marked effect of PGE₂ on bone cell cyclic AMP suggests that this action is involved in the mechanism of PGE₂-related bone loss. The fact that agents with different effects on PGE₂-induced increases in cellular cyclic AMP can inhibit its resorptive actions, suggests that PGE₂-induced changes in cyclic AMP may be related less to its resorptive actions than to its inhibitory effect on bone formation.     

Inhibition of cyclic nucleotide phosphodiesterase during exposure to WI-38 cells to prostaglandin E1.

Short term incubation of WI-38 cultures with 5.7 micron prostaglandin E1 (PGE1) caused cyclic AMP phosphodiesterase activity in fibroblast homogenates to fall by 25 to 35% as compared to controls. The PGE1-induced decline in phosphodiesterase activity coincided with a rapid increase in intracellular cyclic AMP levels in response to the hormone and was rapidly reversed by washing the cultures free of the prostaglandin before homogenizing the cells. The effect of PGE1 on WI-38 phosphodiesterase activity was localized to the enzyme form(s) present in 27,000 times g supernatant fractions of cell homogenates. These data suggest that the pattern of cyclic AMP accumulation in WI-38 fibroblasts exposed to PGE1 may be related, at least in part, to decreased phosphodiesterase activity during hormone stimulation.     

Dissociation of the regulation of fatty acid synthesis and cyclic AMP levels in cultured glial and neuronal cells.

Abstract— Cultured C-6 glia and neuroblastoma were utilized to investigate the relation of rates of fatty acid synthesis (from ³H₂O) to levels of cyclic AMP under conditions of short-term and long-term regulation. The data demonstrate a consistent dissociation of alterations in rates of fatty acid synthesis and levels of cyclic AMP. Thus, marked alterations in the rate of fatty acid synthesis occurred when serum or albumin-bound palmitic acid was present in the culture medium, but there were no accompanying alterations in levels of cyclic AMP. Similarly, when high intracellular and/or extracellular levels of cyclic AMP were induced by exposure of the cells to dibutyryl cyclic AMP or isoproterenol, no change in the rate of fatty acid synthesis occurred. Although the data raise serious doubt about an important role for cyclic AMP in the regulation of fatty acid synthesis, they do not rule out such a role. The findings do indicate that any such role must involve alterations in compartmentalization, metabolism or binding of the mononucleotide within the cell.     

Action of 50 Hz magnetic fields on cyclic AMP and intercellular communication in monolayers and spheroids of mammalian cells

To investigate the influence of physiological parameters such as cell density and three-dimensional cell contact on the biological action of a 2 mT/50 Hz magnetic field, mouse fibroblasts were exposed as monolayers and as multicellular spheroids. Changes in cyclic AMP content of cells and alterations in gap junction-mediated intercellular communication were measured immediately after 5 min of exposure to the field. In monolayers of intermediate cell density (1 × 10⁵ cells/cm²), the field treatment caused an increase in cAMP to 121% of the control level, whereas, at 3 × 10⁵ cells/cm² (near confluence), a decrease to 88% of the unexposed cells was observed. Furthermore, field exposure stimulated gap-junction communication to 160% of the control level as determined by Lucifer yellow dye exchange. In spheroids, alterations in the radial profile of cellular cAMP were observed that were due both to field-induced local cAMP changes and to increased gap-junction permeability for this second messenger, the latter causing radial cAMP gradients to be flattened. The results indicate a strong dependence of field action on physiological parameters of the system exposed. © 1995 Wiley-Liss, Inc.     

Mitogenic and antimitogenic effects of cholera toxin-mediated cyclic AMP levels in 3T3 cells

The effect of time-controlled exposures to cholera toxin (CT) on intracellular levels of cyclic AMP (cAMP) and on the proliferative response of serum-stimulated 3T3 cells was investigated. Continuous exposure to CT caused up to 8-fold raises in cAMP content and inhibited DNA replication by delaying G1-S transition and by reducing the fraction of cells committed to DNA replication. In contrast, short exposures to CT during G0-G1 transition increased the fraction of cells responding to serum stimulation and potentiated the serum-induced morphological changes in the cell monolayer. A short exposure during late G1 phase, however, inhibited the onset of DNA synthesis but had little effect on ongoing DNA replication. The results indicate that cAMP has diverse and opposite effects on two defined restriction points in cell cycle control. Cyclic AMP was positively involved in the acquisition of the state of competence by quiescent cells (G0-G1 transition) but antagonistic on the onset of DNA replication (G1-S transition) in committed cells. The observations reconcile a number of controversial conclusions regarding the role of cAMP in cell cycle control.     

Prostacyclin analogues inhibit canine parietal cell activity and cyclic AMP formation

To assess further the mechanism by which prostacyclin inhibits acid secretion, the actions of two stable prostacyclin analogues on parietal cell function and cyclic AMP formation were tested using enzymatically dispersed cells from canine fundic mucosa. Accumulation of ¹⁴C-aminopyrine (AP) was used as an index of parietal cell response to stimulation. The 16-phenoxy derivative of PGI₂ inhibited accumulation of AP stimulated by histamine (10 μM), with 50% inhibition (ID₅₀) at 10 nM. 6_β-PGI₁ also inhibited the action of histamine (ID₅₀ 0.5μM) but failed to block stimulation by carbachol or the dibutyryl derivative of cyclic AMP (dbcAMP). In similiar concentrations to those producing inhibition of histamine-stimulated AP accumulation, the 16-phenoxy analogue and 6_β-PGI₁ inhibited histamine-stimulated cyclic AMP generation by parietal cells. At 100 fold higher concentrations, 6_β-PGI₁ stimulated cyclic AMP formation, presumably in non-parietal cells. Even in high concentrations the 16-phenoxy analogue failed to increase cyclic AMP formation by mucosal cells. These data indicate that the stable prostacyclin analogues are potent, direct inhibitors of histamine-stimulated parietal cell function and that it is the inhibition, rather than the stimulation, of cyclic AMP formation that is linked to the antisecretory actions of these prostanoid compounds.     

Unsaturated fatty acid effect on cyclic AMP levels in human embryo lung fibroblasts.

The addition of arachidonic acid at 250 muM to cultures of human embryo lung fibroblasts (IMR-90) increases cellular cyclic AMP levels within 5 minutes to approximately 15-fold over basal. Other unsaturated fatty acids, 11, 14, 17-eicosatrienoic, linoleic, 8, 11, 14-eicosatrienoic and oleic also cause similar rapid elevation of cellular cyclic AMP. During this time interval, no detectable conversion of the added linoleic or arachidonic acids to prostaglandin is observed. These cells produce prostaglandins at measurable concentrations in response to treatment with ascorbic acid or bradykinin. Saturated fatty acids have no influence on cyclic AMP levels in these cells. This effect of unsaturated fatty acids on cellular cyclic AMP levels varies with the cell type. For example, smooth muscle and endothelial cells obtained from the calf pulmonary artery show very little or no increase in cellular cyclic AMP upon exposure to arachidonic acid.     

Effect of hormones on cyclic AMP levels in cultured human cells.

Cultured cells derived from human adipose tissue grew more slowly and had significantly higher basal levels of cyclic AMP than cultured fibroblasts. Cyclic AMP levels in cultured adipose tissue cells were unaffected by epinephrine and were elevated 15-fold by prostaglandin E₁ while fibroblast cyclic AMP levels were elevated 27-fold by epinephrine and 95-fold by prostaglandin E₁. These results support the postulate that the cultured adipose tissue cell is a distinct cell type which may represent an adipocyte or preadipocyte in culture.     

Cyclic AMP and growth of Ehrlich ascites tumor cells. Lack of cyclic AMP elevation in nutritionally deprived cells and mechanism of retardation of growth by dibutryl cyclic AMP.

Cyclic AMP levels in Ehrlich ascites tumor cells changed little after deprivation of cells of essential nutrients, serum, glucose and amino acids, deprival of each of which leads to marked inhibition of growth and protein synthesis. Cyclic AMP levels also changed little after the addition of these nutrients to deprived cells. Thus cyclic AMP is not likely to be the intracellular mediator for growth regulation by these three nutrients. Elevation of cyclic AMP levels for short periods by exposure of cells to choleratoxin or theophylline produced only slight changes in parameters of protein synthesis (polyribosome pattern and rate of [3H]leucine incorporation). An exposure for 1 day to dibutyryl cyclic AMP did not inhibit cell growth. However, prolonged exposure to dibutyryl cyclic AMP inhibited the multiplication of Ehrlich ascites cells both in suspension and in stationary cultures. No morphological effects were evident in the former; in the latter, cells attached firmly to the substratum and formed elongated cytoplasmic processes. Inhibition of cell multiplication by dibutyryl cyclic AMP was related to cell density and to serum concentration. Cells in dibutyryl cyclic AMP-containing media plated at low cell densities multiplied as rapidly as control cells. The final densities cells reached were determined by the serum concentration; in dibutyryl cyclic AMP-containing media these densities were about one-half those of respective control cells. Limitation of cell multiplication by dibutyryl cyclic AMP was reversed by the addition of serum, by resuspending cells at lower densities, or by resuspending cells in media without dibutyryl cyclic AMP. These findings suggested that dibutyryl cyclic AMP may affect the utilization of serum factors by cells. Dibutyryl cyclic AMP did not inactivate serum factors and did not change the rate at which cells depleted the growth medium of serum factors. Dibutyryl cyclic AMP may limit cell multiplication by increasing the cellular requirement for serum factors.     

Regulation of cyclic nucleotide phosphodiesterase activity in human lung fibroblasts.

Cyclic nucleotide phosphodiesterase activity (3', 5'-cyclic-nucleotide 5'-nucleotidohydrolase, 3.1.2.17) was studied in homogenates of WI-38 human lung fibroblasts using 0.1--200 microgram cyclic nucleotides. Activities were observed with low Km for cyclic AMP(2--5 micron) and low Km for cyclic GMP (1--2 micron) as well as with high Km values for cyclic AMP (100--125 micron) and cyclic GMP (75--100 micron). An increased low Km cyclic AMP phosphodiesterase activity was found upon exposure of intact fibroblasts to 3-isobutyl-1-methylxanthine, an inhibitor of phosphodiesterase activity in broken cell preparations, as well as to other agents which elevate cyclic AMP levels in these cells. The enhanced activity following exposure to 3-isobutyl-1-methylxanthine was selective for the low Km cyclic AMP phosphodiesterase since there was no change in activity of low Km cyclic GMP phosphodiesterase activity or in high Km phosphodiesterase activity with either nucleotide as substrate. The enhanced activity due to 3-isobutyl-1-methylxanthine appeared to involve de novo synthesis of a protein with short half-life (30 min), based on experiments involving cycloheximide and actinomycin D. This activity was also enhanced with increased cell density and by decreasing serum concentration. Studies of some biochemical properties and subcellular distribution of the enzyme indicated that the induced enzyme was similar to the non-induced (basal) low Km cyclic AMP phosphodiesterase.     

Variations in cyclic AMP level and specific activities of adenylate cyclase and cyclic AMP phosphodiesterase during the cell cycle of an Actinomycete (author's transl)

The variations in the concentrations of intra- and extracellular cyclic AMP and in he specific activities of adenylate cyclase (EC 4.6.1.1) and cyclic AMP phosphodiesterase (EC 3.1.4.17) have been monitored in synchronized cultures of Nocardia restricta, a prokaryote belonging to the group of Actinomycetes. At the beginning of the cell cycle, during a first period of RNA and protein synthesis, there is an increasing synthesis of adenylate cyclase which can be suppressed in the presence of chloramphenicol or rifampicin. Simultaneously, the specific activity of cyclic AMP phosphodiesterase decreases and the concentrations of intra- and extracellular cyclic AMP rise. After the end of DNA replication, during a second period of RNA and protein synthesis, the specific activity of cyclic AMP phosphodiesterase increases; during the same time, the specific activity of adenylate cyclase and the level of intracellular cyclic AMP drop. It appears that the overall metabolism of cyclic AMP is coordinated so that the cyclic AMP level will be high at the beginning of DNA replication and will fall thereafter. The results are discussed in comparison with known data about the variations of cyclic AMP during the cell cycle of mammalian cells in cultures.