A role for p53 in maintaining and establishing the quiescence growth arrest in human cells

The p53 tumor suppressor protein induces transient growth arrest or apoptosis in response to genotoxic stress and mediates the irreversible growth arrest of cellular senescence. We present evidence here that p53 also contributes to the reversible, growth factor-dependent arrest of quiescence (G(0)). Microinjection of expression vectors encoding either MDM2 or a pRb-binding mutant of SV40 T antigen, both of which abrogate p53 function, stimulated quiescent normal human fibroblasts to initiate DNA synthesis and were 40-70% as effective as wild-type T antigen. Electrophoretic mobility shift and p53 transactivation assays showed that p53 activity was higher in quiescent and senescent cells compared with proliferating cells. As proliferating cells entered G(0) after growth factor withdrawal, the p53 mRNA level increased, followed by transient accumulation of the protein. Shortly thereafter, the expression (mRNA and protein) of p21, a p53 target gene and effector of cell cycle arrest, increased. Finally, stable expression of the HPV16 E6 oncogene or dominant negative p53 peptide, GSE-22, both of which inhibit p53 function, delayed entry into quiescence following growth factor withdrawal. Our data indicate that p53 is activated during both quiescence and senescence. They further suggest that p53 activity contributes, albeit not exclusively, to the quiescent growth arrest.     

p53R2-dependent ribonucleotide reduction provides deoxyribonucleotides in quiescent human fibroblasts in the absence of induced DNA damage

Human fibroblasts in culture obtain deoxynucleotides by de novo ribonucleotide reduction or by salvage of deoxynucleosides. In cycling cells the de novo pathway dominates, but in quiescent cells the salvage pathway becomes important. Two forms of active mammalian ribonucleotide reductases are known. Each form contains the catalytic R1 protein, but the two differ with respect to the second protein (R2 or p53R2). R2 is cell cycle-regulated, degraded during mitosis, and absent from quiescent cells. The recently discovered p53-inducible p53R2 was proposed to be linked to DNA repair processes. The protein is not cell cycle-regulated and can provide deoxynucleotides to quiescent mouse fibroblasts. Here we investigate the in situ activities of the R1-p53R2 complex and two other enzymes of the de novo pathway, dCMP deaminase and thymidylate synthase, in confluent quiescent serum-starved human fibroblasts in experiments with [5-(3)H]cytidine, [6-(3)H]deoxycytidine, and [C(3)H(3)]thymidine. These cells had increased their content of p53R2 2-fold and lacked R2. From isotope incorporation, we conclude that they have a complete de novo pathway for deoxynucleotide synthesis, including thymidylate synthesis. During quiescence, incorporation of deoxynucleotides into DNA was very low. Deoxynucleotides were instead degraded to deoxynucleosides and exported into the medium as deoxycytidine, deoxyuridine, and thymidine. The rate of export was surprisingly high, 25% of that in cycling cells. Total ribonucleotide reduction in quiescent cells amounted to only 2-3% of cycling cells. We suggest that in quiescent cells an important function of p53R2 is to provide deoxynucleotides for mitochondrial DNA replication.     

Effects of prolonged quiescence on neuclei and chromatin of WI-38 fibroblasts.

DNA synthesis and cell division are markedly reduced in confluent mono-layers of WI-38 diploid fibroblasts, but resume again if the depleted medium is replaced by fresh medium containing 10% fetal calf serum. If the cells are kept quiescent for prolonged periods of time after confluence (1 or 2 weeks), the fraction of cells that can be stimulated to proliferate by fresh serum decreases and the length of the prereplicative phase increases. The template activity of isolated nuclei decreases with increasing time of quiescence, and parallel changes occur in chromatin as evidenced by circular dichroism spectra and capacity to bind the intercalating dye, ethidium bromide. When WI-38 cells are stimulated to proliferate after prolonged quiescence, the increase in template activity of nuclei is delayed by several hours in comparison to cells stimulated after short periods of quiescence. Two distinct steps, both requiring serum, can be identified in the prereplicative phase of cells stimulated to proliferative after prolonged quiescence. We interpret the results as indicating that, during prolonged quiescence, WI-38 fibroblasts go into a deeper GO state from which they can be rescued only after prolonged stimulation. In this respect, prolonged quiescence may bear some resemblance to the process of aging.     

Growth state and cell cycle dependent phosphorylation of DNA topoisomerase II in Swiss 3T3 cells.

We have investigated the amount of DNA topoisomerase II and phosphorylation of the enzyme in Swiss 3T3 cells during the transition from cell quiescence to proliferation. A relatively high level of phosphorylation was observed with proliferating cells while no or a very low level of phosphorylation was observed with quiescent cells. Phosphoamino acid analysis of the phosphorylated topoisomerase II revealed that the phosphorylated aminoacyl residue was serine. When quiescent cells were stimulated to grow by the addition of serum, DNA synthesis began to increase at 9 h after serum addition, reaching a maximum at 15 h and then declining. The amount of topoisomerase II began to increase at 6 h and reached a maximum at 22-27 h, corresponding to the G2 phase. The phosphorylation of topoisomerase II measured by pulse-labeling gradually increased from 6 to 18 h and reached a maximum at 22 h when the amount of the enzyme was maximum. The level of phosphorylation measured by continuous-labeling increased gradually up to 12 h and markedly up to 28 h, and then declined. The increase in the rate of phosphorylation in the G2 phase was affected by inhibiting DNA synthesis, but the increase in the amount of the enzyme was not. Thus, it was suggested that the regulation of phosphorylation of topoisomerase II differs from that of the amount of the enzyme.     

Detection of a novel quiescence-dependent protein kinase

We have identified a cell quiescence-specific 33-kDa cytoplasmic protein kinase (p33(QIK), Quiescence-Induced Kinase) based on induction of p33(QIK)-specific kinase activity of cells growth-arrested in the quiescent phase and deactivation upon entry into the cell cycle. Blockage of macromolecular synthesis prevents p33(QIK) from deactivation, indicating a requirement of newly synthesized regulators for deactivation of p33(QIK) during G(0)/G(1) transition. Stress shock induces additional increases of p33(QIK) activity in a quiescence-dependent manner that correlates with induction of apoptosis. Using a specific antibody to Krs1/Mst2 protein, we found that p33(QIK) is related to p63(Krs1) and is distinguishable from a 36-kDa protein kinase, which is induced through proteolytic modification of activated p63(Krs1) in proliferating cells undergoing apoptosis. p33(QIK) is constantly expressed in quiescent, proliferating, and apoptotic quiescent cells. Regulation of p33(QIK) activity involves protein phosphorylation/dephosphorylation in a proteolysis-independent manner. Regulation of p33(QIK) and related p63(Krs1) and p36 appears to involve distinct pathways in quiescent and proliferating cells, respectively. Our results illustrate the relevance of p33(QIK) activity for cell quiescence that may provide a new insight into signaling pathways regulated in cells during quiescence and quiescence-related apoptosis.     

Co-operation between the p53 protein tumor antigen and platelet-poor plasma in the induction of cellular DNA synthesis

Plasmids containing DNA sequences coding for p53 were microinjected into quiescent Swiss 3T3 cells. Three constructs were used, carrying either the whole gene sequence, a full-length cDNA, or a hybrid between the gene and the cDNA. All of them stimulated DNA synthesis when cells were incubated with platelet-poor plasma (PPP) following injection. The p53 gene stimulated DNA synthesis to a lesser extent, also in the absence of PPP. Several negative results were obtained with different plasmids, including deletion mutants in the p53 coding region. However, a deletion mutant in which the p53 reading frame ended in the middle of the coding part of the p53 gene still stimulated DNA synthesis in co-operation with PPP. The stimulation of DNA synthesis induced by p53 cDNA was more synchronous and more limited than that induced by serum. The present data suggest that p53 may act as a competence factor in cell cycle progression.     

Potassium transport and content during G1 and S phase following serum stimulation of 3T3 cells.

The effect of serum stimulation on unidirectional and net K flux and their relationship to the initiation of DNA synthesis has been investigated in mouse 3T3 fibroblasts. Stimulation of quiescent 3T3 cells with 20% serum results in the initiation of S phase approximately ten hours after serum addition. During transition from G₁ to S phase distinct changes in K transport and cellular K content occur. Total unidirectional K influx undergoes an immediate 2-fold increase upon serum addition, an observation in qualitative agreement with previous results (Rozengurt and Heppel, 1975). This total increase in unidirectional K influx represents a proportional increase in the active, ouabain sensitive component and the K-K exchange component. The initial increase in total flux is followed by a gradual decline over a 16-hour period to levels approaching those of quiescent cells. Following the initial increase in unidirectional K influx is an approximately 75% increase in cell K on a per milligram protein basis or a 40% increase on a per volume basis. This increase peaks at four to five hours and then declines to initial levels at 10 to 14 hours. Populations of quiescent cells given 20% serum plus 0.5 mM ouabain simultaneously are totally blocked from entering S phase, as determined by the appearance of ³H-thymidine labeled nuclei. However, if the ouabain is removed after six hours these cells then undergo the same changes in unidirectional K influx and content as serum stimulated cells with entrance into S phase retarded by five to six hours. If ouabain is added to serum stimulated cells at six hours, after the increase in K transport and K content have occurred, entrance into S phase is not entirely blocked. In cells stimulated with serum and 0.5 mM dBcAMP plus 1 mM theophylline simultaneously, entrance into S phase is greatly reduced as compared to serum stimulation only. However, the early and late changes in K flux and K content are not substantially altered. This indicates that the K transport events associated with G₁ and early S phase are not directly regulated by changes in cAMP levels which follow serum stimulation.     

Stimulation by serum of the Na+/H+ antiporter in quiescent pig kidney epithelial (LLC-PK1) cells and role of the antiporter in the reinitiation of DNA synthesis

LLC-PK1 cells can be brought into a classical quiescent state by depriving them of serum for 6 days. At this time, pulse-labeling with [3H]-thymidine shows that only 3% of the cells are synthesizing DNA, but the quiescent cells can be stimulated with serum to re-enter the cell cycle at a point early in G1. The rate of amiloride-sensitive 22Na+ uptake (as a measure of the Na+/H+ antiporter) is relatively low during quiescence; it rises 2- to 3-fold within 4 h after serum addition. This increase in antiporter activity appears to be required for the resumption of DNA synthesis in the absence of bicarbonate, because ethylisopropylamiloride (EIPA) blocks [3H]-thymidine incorporation when serum is added to cells in bicarbonate-free medium. In the presence of bicarbonate, however, EIPA has no effect on [3H]-thymidine incorporation, indicating that another (bicarbonate-dependent) transport system can substitute for the antiporter under these conditions.     

Metabolic turnover of proliferation-related nuclear proteins in serum-stimulated Swiss mouse 3T3 cells

Stimulation of quiescent Swiss mouse 3T3 cells either by serum or by pure growth factors induces DNA synthesis after a lag period of about 15 h. Following restimulation by serum or by growth factors there is an overall increase of 2-4-fold in the rate of biosynthesis of nuclear proteins. Two nuclear polypeptides show specific temporal correlations with the transition from quiescence to proliferation. The synthesis of p30 (30 kDa, pI 5.2) is at a maximum within 5 h of restimulation, while the synthesis of p36 (36 kDa, pI 4.25) is first seen at 10-12 h after restimulation. The synthesis of p36 correlates well with the initiation of DNA biosynthesis. The metabolic turnover of both of these proteins has been estimated by pulse-chase and by cycloheximide inhibition experiments. They both have a half-life of 10-15 h and appear to be cell-cycle related.     

Hexose uptake and control of fibroblast proliferation

The role of glucose uptake in control of cell growth was studied by experimentally varying the rate of glucose uptake and examining the subsequent effect on initiation and cessation of cell proliferation. The rate of glucose uptake was varied by adjusting the concentration of glucose in the culture medium. This permitted analysis of two changes in rate of glucose uptake which are closely related to the regulation of cell growth: (1) the rapid increase in glucose uptake that can be detected within several minutes after mitogenic stimulation of quiescent fibroblasts and (2) the decrease in glucose uptake which accompanies growth to a quiescent state. Quiescent cultures of mouse 3T3, human diploid foreskin and secondary chick embryo cells were switched to fresh serum-containing medium with either the normal amount of glucose or a reduced level that lowered the rate of glucose uptake below the rate characteristic of quiescent control cells. The subsequent increases in cell number were equal in both media, demonstrating that the increase in glucose uptake, commonly observed after mitogenic stimulation, was not necessary for initiation of cell division. Measurements of intracellular D-glucose pools after serum stimulation of quiescent cells revealed that the increase in glucose uptake was not accompanied by a detectable change in the intracellular concentration of glucose. Nonconfluent growing cultures of mouse 3T3, human diploid foreskin and secondary chick embryo cells were switched to low glucose media, lowering the rate of glucose uptake below levels observed for quiescent cells. This did not affect rates of DNA synthesis or cell division over a several-day period. Thus, the decrease in glucose uptake, which usually occurs at about the same time as the decrease in DNA synthesis as cells grow to quiescence, does not cause the decline in cell proliferation. Experiments indicated that there was no set temporal relationship between the decline in glucose uptake and DNA synthesis as cells grew to quiescence. The sequence was variable and probably depended on the cell type as well as culture conditions. Measurements of intracellular D-glucose pools in secondary chick embryo cells demonstrated that the internal concentration of glucose in these cells did not significantly vary during growth to quiescence. Taken together, our results show that these fluctuations in the rate of glucose uptake do not lead to detectable changes in the intracellular concentration of glucose and that they do not control cell proliferation rates under usual culture conditions.     

Elimination of proliferating cells unmasks the shift from senescence to quiescence caused by rapamycin

Background

Depending on cellular context, p53-inducing agents (such as nutlin-3a) cause different outcomes including reversible quiescence and irreversible senescence. Inhibition of mTOR shifts the balance from senescence to quiescence. In cell lines with incomplete responses to p53, this shift may be difficult to document because of a high proportion of proliferating cells contaminating arrested (quiescent and senescent) cells. This problem also complicates the study of senescence caused by minimal levels of p21 that are capable to arrest a few cells.

Methodology

During induction of senescence by low levels of endogenous p53 and ectopic p21, cells were co-treated with nocodazole, which eliminated proliferating cells. As a result, only senescent and quiescent cells remained.

Results and Discussion

This approach revealed that rapamycin efficiently converted nutlin-induced-senescence into quiescence. In the presence of rapamycin, nutlin-arrested MCF-7 cells retained the proliferative potential and small/lean morphology. Using this approach, we also unmasked senescence in cells arrested by low levels of ectopic p21, capable to arrest only a small proportion of HT1080-p21-9 cells. When p21 did cause arrest, mTOR caused senescent phenotype. Rapamycin and high concentrations of nutlin-3a, which inhibit the mTOR pathway in these particular cells, suppressed senescence, ensuring quiescence instead. Thus, p21 causes senescence passively, just by causing arrest, while still active mTOR drives senescent phenotype.     

Protein synthesis in 3T3 cells stimulated to proliferate at various rates

Reproducible conditions were defined for using rates of leucine incorporation as a valid measure of rates of de novo protein synthesis in mouse 3T3 cells. Upon stimulation of quiescent cultures, rates of de novo synthesis of proteins increased and pool levels of amino acids decreased in proportion to the concentration of serum in the stimulating medium. Rates of de novo protein synthesis (per cell) exhibited a biphasic pattern of increase. These rates approached a plateau value at the end of the lag phase and increased again as cells entered S phase. This pattern of behaviour helps to explain the observed relationships between cell growth (increase in mass) and cell proliferation (increase in cell number).     

Expression of growth-regulated genes in tsJT60 cells, a temperature-sensitive mutant of the cell cycle

We have investigated the expression of growth-regulated genes in tsJT60 cells, a temperature-sensitive (ts) mutant of Fischer rat cells, which, on the basis of its kinetic behavior, can be classified as a G0 mutant. It grows normally at 34 degrees C and also at 39.5 degrees C if shifted to the higher temperature during exponential growth. However, if the cell population is first made quiescent by serum deprivation, subsequent stimulation by serum induces the cells to enter S phase at 34 degrees C but not at 39.5 degrees C. A panel of growth-regulated genes was used that included three protooncogenes (c-fos, c-myc, and p53), several genes that are induced in G0 cells stimulated by growth factors (beta-actin, 2A9, 2F1, vimentin, JE-3, KC-1, and ornithine decarboxylase), and an S-phase gene (histone H3). The expression of these growth-regulated genes was studied in both tsJT60 cells and its parental cell line, rat 3Y1 cells. All the genes tested, except histone H3, are similarly induced when quiescent tsJT60 cells are stimulated by serum at either permissive or restrictive temperatures. These results raise intriguing questions on the nature of quiescence and the relationship between G0 and G1 in cells in culture.     

Replicative activity of isolated chromatin from proliferating and quiescent early passage and aging cultured mouse cells

Replicative activity of isolated chromatin from late passage cultured mouse cells has been compared to the activities of chromatin preparaions from dividing and quiescent early passage cells. Rates of endogenous DNA synthesis are similar for chromatin from growing or resting cells but this activity is stimulated 2.5-fold in senescent cell chromatin. Chromatin from growing young cells copies exogenously added single stranded DNA at the highest efficiency. Chromatin of senescent cells copies this template at a lower rate and resting young cell chromatin replicates single stranded DNA at the lowest efficiency. Similar relative rates are obtained when activated DNA is copied by the various chromatin preparations. Total activity of DNA polymerase extracted by salt from chromatin is similar for dividing and quiescent young cells but the proportion of DNA polymerase beta is higher in the latter. Elevated activities of DNA polymerases are extracted from chromatin of old cells. It is concluded, therefore, that chromatin-directed replication is differently arrested in non-dividing senescent cells and in quiescent early passage cells. The possible regulatory mechanisms of DNA replication in quiescence and aging are discussed.     

Evidence for differences in the mechanism of cell cycle arrest between senescent and serum-deprived human fibroblasts: Heterokaryon and metabolic inhibitor studies

It has previously been shown that serum-deprived, early passage quiescent human diploid fibroblastlike (HDFL) cells are able to inhibit cycling cells from entry into DNA synthesis upon cell fusion. We have found that the degree of inhibition of DNA synthesis in the heterokaryon correlates with the duration of serum deprivation, which is consistent with the suggestion that serum-deprived cells may enter progressively deeper stages of G₀ as they increase their time in quiescence. In contrast to fusions with senescent cells, in heterokaryons between serum-deprived early passage and cycling young cells transient inhibition of protein synthesis with cycloheximide or inhibition of RNA synthesis with 5–6-dichloro-1-β-D-ribofuranosyl benzimidazole (DRB) did not stimulate nuclear [₃H]-thymidine incorporation. These results suggest that differences may exist in the mechanisms responsible for inhibiting cell cycle progression in senescent vs early passage quiescent HDFL cells.     

The relative effects of different types of growth factors on DNA replication, mitosis, and cellular enlargement

It was recently demonstrated that growth in cell size can be dissociated from DNA synthesis and mitosis. 3T3 cells starved to quiescence in low serum concentration can be stimulated to undergo DNA synthesis and one cell division without growing in size (unbalanced growth) (42-44). We report here that in cells stimulated to undergo unbalanced growth, the cell nucleus undergoes balanced growth, i.e., nearly doubles in size prior to mitosis. The reduced ability to grow in cell size under unbalanced growth conditions is thus mainly ascribable to the cytoplasm. Furthermore, the extent to which cells grow in size prior to mitosis is dependent on the serum concentration in the tissue culture medium (44). This data suggests that some macromolecular factor or factors in serum are required for growth in cell size prior to mitosis. We report in this study that epidermal growth factor (EGF) alone exerts a small but significant stimulatory influence on DNA synthesis and mitosis but does not affect cellular enlargement. In contrast, insulin added at supraphysiological concentrations does not stimulate quiescent cells to enter S phase but instead stimulates growth in cell size in the small fraction of dividing cells. Furthermore, cells stimulated to proliferate by EGF could be induced to undergo balanced growth when insulin was added concomitantly. Finally, platelet-derived growth factor (PDGF) stimulates quiescent sparse 3T3 cells to undergo DNA synthesis and mitosis. PDGF also exerts a limited but significant effect on cellular enlargement. However, PDGF alone could not induce a complete balanced growth, i.e., a doubling in cell size prior to mitosis.     

Epidermal growth factor receptors lose ligand binding ability as WI-38 cells progress from short-term to long-term quiescence

WI-38 cells, density arrested for short periods of time, can be stimulated to re-enter the cell cycle by epidermal growth factor (EGF) alone. However, cells density arrested for longer periods have a prolonged prereplicative phase when serum stimulated and cannot be stimulated by EGF alone. Radio-ligand binding studies performed on WI-38 cells showed that actively growing cells bind [¹²⁵I]EGF at relatively low levels that increase to a maximum as the cells become contact inhibited. As the cells enter a state of deeper quiescence, EGF binding falls to one-third to one-fifth the short-term growth arrested levels, remaining constant thereafter. The EGF-receptor complexes internalize more slowly in long-term growth arrested cells, and the rate of ligand association to the receptor is lower than short-term growth arrested cells. The amount of EGF receptor protein in lysates of equal numbers of both short- and long-term quiescent cells remains the same. These results suggest that the failure of long-term growth arrested cells to respond to EGF is not due to dramatic changes in the amount of receptor protein during prolonged quiescence but more likely to an alteration in the ability of these receptors to bind ligand and/or activate the EGF signal transduction pathway. © 1993 Wiley-Liss, Inc.