Search

As a first step toward elucidating the biochemical basis of gene regulation at the G1-S boundary of the cell cycle, we have identified regions of the murine thymidine kinase (TK) promoter sufficient to confer appropriately growth-responsive expression to a heterologous gene. Using a series of TK promoter-chloramphenicol acetyltransferase (CAT) gene fusion constructs, we have identified sequences located between -174 base pairs upstream and +159 base pairs downstream of the TK translation initiation site that are sufficient to drive efficient S phase-specific expression of the CAT reporter gene in transfected murine fibroblasts. Both deletion analysis and site-specific mutagenesis experiments indicated that an Sp1 consensus binding site is critical to the activity of this promoter. Synchronized populations of BALB/c 3T3 cells stably transfected with either TK promoter-CAT fusion constructs or TK promoter-beta-globin fusion constructs expressed their respective reporter genes in an S phase-specific manner following serum stimulation. In each case, reporter gene expression was reduced during quiescence and G1 and rose upon entry of cells into S phase. The TK sequences included in these constructs therefore contained information sufficient to confer S phase-specific regulation to these two reporter genes. These results set the stage for a more detailed analysis of the sequences and trans-acting factors responsible for regulating murine TK gene expression and may lead to insights into the control of proliferation in normal and transformed cells.

To identify previously undetected genes that might be involved in later stages of the transition from a quiescent state (G0) to the DNA synthetic phase (S) of murine cells, we set out to isolate cDNA clones derived from mRNAs that appear late in G1 phase in serum-stimulated cells. A lambda-cDNA library was prepared using poly(A)+ RNA from chemically transformed Balb/c 3T3 cells (BP/A31) that had been brought to quiescence and subsequently stimulated for 12 h with serum. From the first screening of approximately 21,000 recombinant phage plaques, about 100 clones were isolated that hybridized to a single-stranded cDNA pool derived from stimulated-cell RNA but not to DNAs made from resting-cell RNA. Eventually, six different clones were identified. The mRNAs from five of these genes increased gradually during the G0 to S transition, in contrast to the "immediate-early" rise of c-myc mRNA or the later rise of thymidine kinase mRNA. These six clones were sequenced and compared to the GenBank database. Clones LG-80, LG-2, and LG-69 are highly homologous to beta-actin, lactate dehydrogenase, and alpha-tubulin. Clones LG-5, LG-61, and LG-74 had no significant homologies to known sequences. A subtractive cDNA library was used to isolate two additional clones, Sub-S1 and Sub-S2; these have homologies to enolase and ribosomal protein L32. Additional studies that examine the function and regulation of these newly identified "late response" genes in the pre-DNA synthesis pathway are in progress.

Tissues in adults can be maintained at constant mass or they can increase or decrease in size because of imbalances of synthetic and degradative processes acting at the cellular and molecular levels. Some size changes are caused by physiologic conditions to which the tissue must adjust. Alternatively, the balance may be distorted in favor of net tissue increase in pathologic situations such as cancer. Strict regulatory mechanisms are required to keep proliferation responsive to the organism's needs; these mechanisms may be defective in disease. Net tissue proliferation requires repeated rounds of cell duplication in excess of that necessary to counterbalance cell death. Duplication of a cell requires a net doubling of its every molecule and structure. The myriad of molecular events required for cell proliferation such as DNA duplication and its partitioning at mitosis are tightly regulated in normal cells. One may conceive of two classes of molecules: those required for "housekeeping," which constitute the cell's structural and functional machinery, and those such as growth factors, their receptors, and second messengers involved in signal transduction responsible for regulating the activities of the housekeeping molecules. These molecular events and the cascade of processes that control them can be organized within the sequence of the cell cycle. In this brief overview, we illustrate these issues with a few examples taken from very recent discoveries of novel proteins that appear to have major regulatory roles. Most of these results have been obtained with mammalian fibroblasts, but some have originated with discoveries made using two very different yeasts.(ABSTRACT TRUNCATED AT 250 WORDS)

A DNA binding activity to an upstream region of the murine thymidine kinase gene is regulated differently in a transformed and nontransformed cell line pair. Differences in regulation were observed (i) after serum levels were reduced, (ii) when serum levels were returned to initial high levels, and (iii) while protein synthesis was inhibited. After reduction of serum levels, the binding activity was unstable in nontransformed BALB/c 3T3 clone A31 cells but was significantly more stable in benzo[a]pyrene-transformed BALB/c 3T3 cells. After serum concentration was returned to high levels, the kinetic pattern of the binding activity differed between nontransformed and transformed cells. While protein synthesis was inhibited, the binding activity was unstable in nontransformed cells and stable in transformed cells. Partial inhibition of protein synthesis--a more stringent condition to test instability--prevented the induction of the binding activity in nontransformed cells. Previously, the labile protein hypothesis set forth the criterion that a protein regulating the onset of DNA synthesis should be unstable in nontransformed cells and stable in transformed cells. The DNA binding activity described here satisfies this criterion.

Pentoxifylline, a methylxanthine, is an effective modulator of alkylating agents in tissue culture and in human tumor explants in mice. In this Phase I trial, escalating dose controlled release pentoxifylline was administered p.o. 3 times daily for 5 days (15 doses) with a constant dose of thiotepa, 40 mg/m2 i.v. on day 2. Forty-four courses of escalating doses of pentoxifylline varying from 400 to 2400 mg were administered to 22 patients with refractory malignancies. Gastrointestinal toxicity, consisting mainly of nausea and vomiting, was dose limiting at 2400 mg pentoxifylline and subsided completely within 24 h of cessation of the drug. Nongastrointestinal toxicity of this thiotepa/pentoxifylline combination was infrequent and included bone marrow depression and supraventricular tachycardia. Increasing the dose of pentoxifylline did not increase the frequency of these rare toxic effects. Plasma concentrations of pentoxifylline and its major metabolites were determined by gas chromatography. Drug accumulation was noted within a cycle (i.e., by day 5) in only two patients and between cycles in no patient. The recommended Phase II dose of p.o. pentoxifylline is 1600 mg (four 400-mg tablets) when given 3 times daily for 5 days (15 doses) with 40 mg/m2 i.v. thiotepa. Based on an interspecies comparison, this dose exceeds that predicted from mouse models to enhance chemotherapy. This regimen can be safely administered on an outpatient basis, with adequate control of gastrointestinal symptoms achieved by standard antiemetics and intermittent dosing with meals. Phase II trials are required to determine the activity of alkylator/modulator combinations.

We have previously described a defective herpes simplex virus (HSV-1) vector system that permits the introduction of virtually any gene into nonmitotic cells. pHSVlac, the prototype vector, stably expresses Escherichia coli beta-galactosidase from a constitutive promoter in many human cell lines, in cultured rat neurons from throughout the nervous system, and in cells in the adult rat brain. HSV-1 vectors expressing other genes may prove useful for studying neuronal physiology or performing human gene therapy for neurological diseases, such as Parkinson disease or brain tumors. A HSV-1 temperature-sensitive (ts) mutant, ts K, has been used as helper virus; ts mutants revert to wild type. In contrast, HSV-1 deletion mutants essentially cannot revert to wild type; therefore, use of a deletion mutant as helper virus might permit human gene therapy with HSV-1 vectors. We now report an efficient packaging system for HSV-1 vectors using a deletion mutant, D30EBA, as helper virus; virus is grown on the complementing cell line M64A. pHSVlac virus prepared using the deletion mutant packaging system stably expresses beta-galactosidase in cultured rat sympathetic neurons and glia. Both D30EBA and ts K contain a mutation in the IE3 gene of HSV-1 strain 17 and have the same phenotype; therefore, changing the helper virus from ts K to D30EBA does not alter the host range or other properties of the HSV-1 vector system.

Concomitant with the onset of S phase, a series of thymidine kinase (TK) splicing intermediates as well as mature TK mRNA accumulates in the nucleus of BALB/c 3T3 cells. Most of the TK splicing intermediates are retained by oligo(dT)-cellulose chromatography, and, therefore, 3' end formation and polyadenylation probably precede the splicing of TK pre-mRNAs. We have further characterized the TK pre-mRNAs that are present in the nuclei of S-phase cells by using specific probes derived from each of the six TK intervening sequences. Based on the sizes of the pre-mRNAs and their patterns of hybridization with these intron probes, we propose a pathway for intron removal from nascent TK transcripts. Intron excision occurred by a preferred, but not necessarily obligatory, order which appears to have been conserved in mouse and Chinese hamster cells.

In the preceding paper, we have identified a protein of Mr = 118,000 which is induced by stress conditions that lead to cessation of DNA synthesis and cell division (Verma, R., Iida, H., and Pardee, A.B. (1988) J. Biol. Chem. 263, 8569-8575). In the current study, we have investigated the possible role this protein may play in cellular proliferation by studying p118 expression in mutants of the cAMP metabolic pathway. The cyr 1-2 mutant gene encodes a thermolabile adenylate cyclase whose activity is only 7% of wild type even at permissive temperatures (23 degrees C). We have found that at 23 degrees C, the G1 period was 5-fold longer in cyr 1-2 than in CYR1+ cells and that p118 was constitutively expressed in these slow cycling mutants. Addition of 8-bromo-cAMP to cyr 1-2 mutants restored growth at both the restrictive and permissive temperatures and resulted in a shut-off in the synthesis of p118. The effect of the analog on p118 expression was rapid, preceding the increase in cell number and percentage-budded cells. In contrast to wild type cells, p118 synthesis was not induced by sulfur starvation in RAS2val19 mutants possessing high levels of adenylate cyclase activity and bcy1 mutants defective in the regulatory subunit of cAMP-dependent protein kinase. A large body of evidence exists supporting a role of cAMP in positive control of cell proliferation. It is therefore possible that conditions which decrease cAMP arrest growth through a chain of events that include p118 induction.

Introducing foreign genetic material into human cells is essential for the elucidation of the function of various human genes and has potential use in the treatment of human diseases by gene therapy. In this study we demonstrate that a defective herpes simplex virus type 1 vector, pHSVlac, can effectively transfer and express the Escherichia coli Lac Z gene in a variety of exponential and quiescent human cells. The human cells tested included representative cells derived from cancer-prone patients that presumably have various DNA repair deficiencies.

Schmid et al. (Cancer Treat. Rep., 60: 23-27, 1976) reported rapid emergence of resistance of L1210 leukemia cells in mice to two schedules of six antimetabolites and much slower development of resistance to a third schedule. Such rapid development of resistance to six drugs presents a striking puzzle, and one whose solution gives some insights into the basis for general emergence of drug resistance. Our approach was to examine the consequences of applying these drugs singly or in pairs and, from the results, to infer interactions in six-drug combinations. 6-Thioguanine (TG) and 6-mercaptopurine are the key drugs since, as shown by Schmid et al., resistance of leukemic cells appeared to six-drug combinations at the same time as did resistance to the purine analogs; sensitivity to the other drugs remained. We demonstrated that cells which emerged were resistant to both of the purine analogs, owing to a deficiency of the activating enzyme hypoxanthine-guanine phosphoribosyltransferase. TG resistance arose in the presence of TG because of an overgrowth of TG-resistance mutants that were present as one cell in 10(4) in the original L1210 population. L1210 cultures were prepared free of TG-resistant mutants. With these cells, TG administered shortly after inoculation was very effective in delaying their death. The cells that finally grew out were still TG sensitive. Simultaneous treatment with all the drugs greatly delayed appearance of TG resistance in vivo and in vitro. Methotrexate alone was responsible for this result, owing to its ability preferentially to kill TG-resistant cells. The other three drugs were not effective in delaying TG resistance. Methotrexate was effective only if it was added daily; one large injection was ineffective. Therefore, TG and methotrexate added daily for 6 days (simultaneous schedule) was the most effective drug regimen tested.

Insulin-like growth factor I (IGF-I) regulation of the sequence of transcriptional, translational, and postranslational events during G1 that are necessary for DNA synthesis, and the induction of thymidine kinase and in vivo thymidylate synthase activities was studied in synchronized confluent BALB/c 3T3 mouse fibroblasts. IGF-I was the only growth factor necessary from 6 to 2 1/2 hours before S phase for onset of DNA synthesis. 5,6 dichlororibofuranosylbenzimidazole (DRB), an inhibitor of transcription, was ineffective in blocking DNA synthesis and increase of both enzyme activities during an interval of about 2 1/2 hours before S phase. Cycloheximide, an inhibitor of translation, was ineffective in blocking DNA synthesis and the increase of in vivo thymidylate synthase activity during an interval of about 1 1/2 hours before S phase but was effective in blocking the increase of thymidine kinase activity up to the G1/S boundary. The results demonstrate that IGF-I is dispensable beyond the R-point (2-3 hours before S phase), the time when serum factors are no longer necessary for the initiation of DNA synthesis, and that IGF-I regulates transcriptional events necessary for both DNA synthesis and the induction of thymidine kinase and in vivo thymidylate synthase. The results also demonstrate a posttranslational interval for in vivo thymidylate synthase, suggesting posttranslational modification of this enzyme.

Permeabilized CHEF/18 hamster cells incorporate label from [3H] cytidine diphosphate (CDP) into material that we had designated as DNA. Spyrou and Reichard reported this label was incorporated only into RNA. To resolve this discrepancy the studies reported here were performed. We demonstrate incorporation into a DNA polymer with major interspersed RNA sequences. Incorporation into deoxycytidine isolated from this product was little diluted by a great excess of unlabeled dCTP, confirming channeling under these conditions of CDP into polymerized deoxyribonucleotides.

Newly synthesized mRNA from hamster cells was labeled in vivo with the thionucleoside analogs 4-thiouridine (4-TU) and 6-thioguanosine (6-TG). The thio-substituted RNA was selectively recovered by Affi-Gel 501 phenylmercury affinity chromatography. Following a 1-h labeling period, enrichment for newly transcribed RNA after a single round of chromatography ranged between 10- and 15-fold when compared with total RNA. Exposure of CHO UrdA- cells, a uridine auxotrophic line, to 50 microM 4-TU allowed for optimal recovery of newly transcribed RNA. Increasing the concentration of 4-TU to 100 microM or labeling with 6-TG at concentrations of 3 microM or greater resulted in similar recoveries from uridine-prototrophic hamster cell lines. For shorter term labeling, exposure of prototrophic cells to 500 microM 4-TU or 100 microM 6-TG for 15 min allowed newly synthesized RNA to be selectively recovered. As a specific test case, enrichment for histone H3.2 mRNA was analyzed after hamster cells were labeled with 4-TU under conditions in which the gene was highly transcriptionally active. Northern blot analysis and the specific activity of thio-substituted RNA revealed a 15-fold enrichment when compared to total RNA. In vivo labeling of cellular RNA with 4-TU or 6-TG should provide a useful method for studying inducible gene expression and for isolating and cloning specific mRNAs from mammalian cells.

Chinese hamster embryo fibroblast cells (CHEF/18) possess a plasma membrane-associated, amiloride-sensitive Na+/H+ antiporter that affects intracellular pH (pHi) and is activated by growth factor addition. Our results using 14C-benzoic acid distribution indicate that both epidermal growth factor (EGF) and thrombin are capable of causing rapid rises in the pHi of CHEF/18 cells. The maximal shift induced by these factors is 0.20 to 0.25 pH units above the basal unstimulated level. Distinctive differences were observed between the modes of action of these two growth factors. Sequential additions revealed that the rise in pHi due to EGF was additive with that caused by diacylglycerols (DAG), while that of thrombin was not. Furthermore, exposure of cells to the phorbol ester PMA for a prolonged period of time in order to down-regulate protein kinase C (pkC), or treatment with the pkC inhibitor H-7, abolished the pHi response to thrombin but not to EGF. In contrast, incubation of cells in nominally calcium-free medium or with the calmodulin antagonists W-7 or trifluoperazine (TFP) decreased only the ability of EGF to cause changes in pHi. These data suggest that there are two distinct mechanisms for activation of the Na+/H+ antiporter in CHEF/18 fibroblast cells and thus provide an example of the use of alternative modes for the modulation of intracellular processes.

Caffeine had been shown to induce mitotic events in Syrian hamster fibroblast (BHK) cells that were arrested during DNA replication (Schlegel and Pardee, Science 232:1264-1266, 1986). Inhibition of protein synthesis blocked these caffeine-induced events, while inhibition of RNA synthesis showed little effect. We now report that the protein(s) that are required for inducing mitosis in these cells were synthesized shortly after caffeine addition, the activity was very labile in the absence of caffeine, and the activity was lost through an ATP-dependent mechanism. Caffeine dramatically increased the stability of these putative proteins while having no effect on overall protein degradation. Experiments with an inhibitor of RNA synthesis indicated that mitosis-related RNA had accumulated during the suppression of DNA replication, and this RNA was unstable when replication was allowed to resume. These results suggest that the stability of RNA needed for mitosis is regulated by the DNA replicative state of the cell and that caffeine selectively stabilizes the protein product(s) of this RNA. Conditions can therefore be selected that permit mitotic factors to accumulate in cells at inappropriate times in the cell cycle. Two-dimensional gel electrophoresis has demonstrated several protein changes resulting from caffeine treatment; their relevance to mitosis-inducing activity remains to be determined.

Nuclear histone kinase activity, specifically histone H1 phosphotransferase activity, was shown to increase in synchronous Chinese hamster cells from the G1/S boundary to late G2/early M phase. Chromatin extracts purified by DEAE-Sephacel chromatography showed a cAMP-independent kinase activity that demonstrated cell cycle dependence and high specificity for histone H1 as the phosphate acceptor in the presence of [gamma-32P] ATP. This activity was purified approximately 40-fold. Using as substrates calf thymus histone H1 subfractions resolved by Bio-Rex 70 ion exchange chromatography, phosphorylation by the nuclear histone H1 kinase indicated that 32P incorporation into H1-2 was at least twice that for H1-1 and H1-3 subfractions. Both amino- and carboxy-terminal fragments generated by N-bromosuccinimide cleavage were phosphorylated. Phosphoamino acid analysis showed phosphothreonine to be approximately twice as abundant as phosphoserine. Histone H1 kinase activity was not activated by cyclic nucleotides, nor inhibited by cAMP-dependent protein kinase inhibitors or regulatory subunits. There was no effect on activity by Ca2+ alone or in the presence of calmodulin or diacylglycerol. Kinase activity was inhibited by nonhydrolyzable analogs of ATP such as adenyl-5'-yl imidodiphosphate, by 5'-p-fluorosulfonylbenzoyladenosine which binds to the ATP binding site of the enzyme, and by quercetin. Column fractions enriched in histone H1 kinase were labeled with 5'-p-fluorosulfonylbenzoyl[8-14C]adenosine, and peptides were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. One band, Mr 67,000, was specifically labeled and may represent the H1 kinase catalytic subunit.

We have discovered and report here a means of separating a mitotic "subcycle" from the G1- and S-phase events of the mammalian cell cycle. Time-lapse videomicroscopy of Syrian hamster fibroblast (BHK) cells revealed that caffeine could induce multiple entries into mitosis while cells were blocked in DNA synthesis. As with normal mitoses, the abundance of mitosis-specific phosphoproteins was coupled with the condensation of chromatin. The BHK temperature-sensitive mutant tsBN2 also completed multiple entries into mitosis while arrested during DNA replication and raised to the restrictive temperature. Periodic mitotic events occurred even when BHK cells were exposed to low concentrations of serum or cycloheximide, conditions that prevent the cycling of BHK cells by blocking their entry into S phase. These results suggest that an oscillation governing the activation and inactivation of mitotic factors can be generated in mammalian cells and uncoupled from the G1 and DNA replication events of the normal cell cycle. This system will be useful for examining the molecular nature of mitotic factors.

Many conditions affect the growth of animal cells in culture. Some, such as essential nutrients, are necessary for growth as precursors of macromolecules. Other substances appear to have regulatory functions. Their presence or absence determines whether the cell will continue to grow or move into a resting, quiescent state. In particular certain factors (usually provided by serum) are thought to have regulatory roles. To decide which factors are regulatory is a difficult problem. We propose here that no single criterion is sufficient to permit a decision and that the best course at present is to apply several criteria. Three such criteria are proposed. To the degree that the tests are satisfied we can tentatively decide whether or not a given factor is regulatory.

In order to further investigate the connection between transport and growth control, 3T3 cells, SV40 transformed 3T3 cells (SV101), and three revertant cell lines derived from SV101 which have regained certain manifestations of growth control were used. Transport rates of 2-amino-isobutyric acid and 3-O-methyl-D-glucose were measured in sparse, confluent, serum-starved, and serum-stimulated cultures. As shown before, cessation of 3T3 cell growth in G0 under conditions of confluence or serum deprivation was associated with reduced rates of transport for both compounds, whereas the density and serum dependence of growth and transport was largely eliminated in SV101. The density revertant F1SV101, which has regained density regulation of growth similar to 3T3 cells, has also regained density regulation of transport. Neither growth nor transport were serum dependent. The serum revertants AgammaSV7 and LsSV6 have regained both density and serum regulation of growth, but not according to the original mechanism of 3T3 cells of entry into a Go state. Transport was high under conditions of confluence or serum deprivation. Thus for these cells rates of transport were not reduced simply as a consequences of slower cell growth nor were low transport rates responsible for growth arrest. The data are consistent with the possibility that growth arrest specifically in the G0 state could shut off a number of cellular activities, including transport.

Total growth of transplanted human or rodent tumors in the subrenal capsule of mice was much improved by treatment with cyclosporine (CSA, cyclosporin A). Tumor size increased rapidly between days 6 and 12 after implantation, CSA injected on days 1-5 or 2-8 prevented tumor regression. In contrast, immunologic regression occurred after 6 days in absence of the drug. Tumor growth was comparable in CSA-treated mice, athymic nude mice with human tumors, or normal mice with syngeneic rodent tumors. Studies with rodent tumors in syngeneic mice showed that the CSA treatments had no antitumor effect. Inflammatory infiltration was seen on days 6-12 after tumor implantation into control mice. Immunoperoxidase staining showed murine T cells to be prominent in the infiltrate. In contrast, tumors in CSA-treated mice contained minimal inflammatory infiltrate even 12 days after implantation. Allogeneic tumors in CSA-treated mice caused neovascularization, metastases, and local invasion into the kidney. cis-Diamminedichloroplatinum showed highly significant activity against human tumors in CSA-treated mice during the period 6-10 days after tumor implantation but showed no statistically significant antitumor activity 0-6 days after implantation in mice not treated with CSA. We suggest that in CSA-treated mice the subrenal capsule assay for tumor growth provides a rapid, economical model for investigations in vivo of mouse or human tumor biology, for drug screening with a standard tumor, or for determination of optimal treatment of particular human tumors.

Agents that prevent repair of lesions in DNA increase cell death. Caffeine enhances lethality of agents such as HN2. We propose that damaged cells stop in G2 phase until repair is completed, and caffeine prevents this G2 arrest. Therefore caffeine lets damaged cells go through mitosis; their chromosomes are shattered and they die. This action of caffeine requires rapid synthesis of protein(s). Some classes of damage by alkylating agent (methyl methane sulfonate (MMS] are removed slowly in human cells, followed by rapid repair synthesis. Lethality results if gaps created during repair of DNA are present during semiconservative replication. Nicotinamide analogs prevent ligation, the gaps persist, and lethality is greater. This effect is S phase specific. Further experiments may specify conditions and agents giving better therapeutic results than can be obtained with anticancer drugs used alone.

[125I]Iododeoxycytidine incorporation was used to measure herpes virus (HSV-1) DNA synthesis following specific DNA damage. Xeroderma pigmentosum fibroblasts were less able to replicate UV-irradiated viral DNA than were normal fibroblasts, indicating the necessity for excision repair for the survival of UV-irradiated virus. Because of its rapidity and ease of quantitation, this assay had advantages over standard viral mediated assays of DNA excision repair. It was possible to monitor viral replication as a function of the cellular cell cycle. Other genetic defects which have been proposed to reflect deficiencies in DNA-repair capacity were not detected by this assay. DNA-repair inhibitors, caffeine and 3-aminobenzamide, also did not show synergistic lethal effects on the replication of damaged viral DNA.

Cells prepare for S phase during the G1 phase of the cell cycle. Cell biological methods have provided knowledge of cycle kinetics and of substages of G1 that are determined by extracellular signals. Through the use of biochemical and molecular biological techniques to study effects of growth factors, oncogenes, and inhibitors, intracellular events during G1 that lead to DNA synthesis are rapidly being discovered. Many cells in vivo are in a quiescent state (G0), with unduplicated DNA. Cells can be activated to reenter the cycle during G1. Similarly, cells in culture can be shifted between G0 and G1. These switches in and out of G1 are the main determinants of post-embryonic cell proliferation rate and are defectively controlled in cancer cells.

We have identified a novel stress-inducible protein in Saccharomyces cerevisiae by pulse-labeling with [35S]methionine and two-dimensional gel analysis. The protein was characterized biochemically to gain further insight into mechanisms regulating the stress response. It has a Mr = 118,000 and exists in two forms of pI = 4.2 (p118A) and pI = 4.3 (p118B). p118A and p118B are modified by N-glycosylation. Tunicamycin treatment revealed the presence of precursor proteins of Mr = 105,000, pI = 4.1 (p105A) and pI = 4.25 (p105B). The synthesis of p118A and p118B was almost completely shut off in cycling cells and was increased 11-fold following a mild heat shock. Both forms of p118 decayed in a biphasic manner under induced conditions. A tight correlation was observed in the kinetics of thermotolerance induction and p118A synthesis. Other forms of stress such as sulfur starvation which lead to arrest in the unbudded phase also resulted in enhanced synthesis of both p118A and p118B. However, in cell division cycle mutants blocked at various stages at the restrictive temperature, p118A and p118B had different synthetic patterns. Taken together, these data imply a role for induced p118 in proliferation arrest in the unbudded state.

A feedback loop for the regulation of the rec/lex-mediated DNA repair system is proposed. This model was formulated from experiments on the genetic and metabolic regulation of the rate of synthesis of protein X performed in this laboratory, and from genetic data obtained in other laboratories. Protein X is proposed to prevent DNA degradation by the recBC-coded exonuclease. The model states tht: (1) The lex (or exrA in E. coli B) gene codes for a repressor. (2) This repressor binds to an operator region of DNA consisting of the tif-zab region at 51 minutes on the E. coli chromosome. (3) The operator region controls the production of several proteins involved in DNA repair, including protein X. (4) The recA gene product is required to remove the lex-coded repressor from the operator. Thre recA gene could code for an antirepressor (inducer protein or a protease) or a modifer of recBC nuclease action; (5) Low molecular weight products of DNA degradation are effectors that activate the system. (6) Protein X limits recBC nuclease action by binding to single-stranded DNA.

We propose that protein X provides the connection between damage to Escherichia coli DNA and inhibition of septation and cell division. This connection is needed to guarantee that each new bacterium receives a complete DNA copy. We present several new experiments here which demonstrate that the degree to which septation is inhibited following damage to DNA is correlated with the amount of protein X that is produced. Rifampin selectively blocks protein X production. This drug was shown to allow cells whose DNA had been damaged by nalidixic acid to resume septation. Several mutants formed septa-less filaments and also produced protein X at 42 degrees C; rifampin both inhibited their production of protein X and permitted them to form septa and divide. Essentially complementary results were obtained with a dnaA mutant which at 42 degrees C stopped making DNA, did not produce protein X, and continued to divide; added bleomycin degraded DNA, induced protein X, and inhibited septation. These results, as well as previous observations, are all consistent with the proposal that protein X is produced as a consequence of DNA damage and is an inhibitor of septation. We suggest that septation could require binding of a single-stranded region of DNA to a septum site in the membrane. Protein X could block this binding by combining with the DNA. This control could provide an emergency mechanism in addition to the usually proposed coordination in which completion of DNA synthesis creates a positive effector for a terminal step of septation. Or it could be the sole coordinating mechanism, even under unperturbed growth conditions.

Normal cells cannot initiate DNA synthesis under inadequate external conditions, yet after growth has started they complete their division cycle under these conditions. The sensitive biochemical event for a growing cell is proposed to be accumulation of a labile protein which in adequate amounts permits entry into S phase, after about 2 hr, and completion of the cycle. Instability of this protein (half-life about 2.5 hr) creates a dynamic state so that its accumulation depends on rates of both synthesis and degradation. Neoplastic cells may show poorly regulated growth either by synthesizing this protein more rapidly or degrading it less rapidly, under conditions that limit normal cells' growth. Known mechanisms of overproduction include: more copies of the protein's structural gene per cell, an adjacent high-activity promoter, or autoproduction of growth factors. Less rapid degradation could result from less protease activity or from stabilizing modifications of the protein. Thus, derangement in the control of a labile growth-regulatory protein acting by any one of these diverse mechanisms could lead to neoplasia.

When cultured fibroblasts are deprived of serum, the degradation of long-lived proteins and RNA increases, the cells stop proliferating, and they decrease in size. To determine the role of the increased protein catabolism in these responses, we studied the effects of inhibitors of intralysosomal proteolysis in Balb/c 3T3 cells. When these cells were placed in serum-deficient medium (0.5% serum), the rate of degradation of long-lived proteins increased about twofold within 30 min. This increase was reduced by 50-70% with inhibitors of lysosomal thiol proteases (Ep475 and leupeptin) or agents that raise intralysosomal pH (chloroquine and NH4Cl). By contrast, these compounds had little or no effect on protein degradation in cells growing in 10% serum. Thus, in accord with prior studies, lysosomes appear to be the site of the increased proteolysis after serum deprivation. When 3T3 cells were deprived of serum for 24-48 hours, the rate of protein synthesis and the content of protein and RNA and cell volume decreased two- to fourfold. The protease inhibitor, Ep475, reduced this decrease in the rate of protein synthesis and the loss of cell protein and RNA. Cells deprived of serum and treated with Ep475 for 24-48 hours had about twice the rate of protein synthesis and two- to fourfold higher levels of protein and RNA than control cells deprived of serum. The Ep475-treated cells were also about 30% larger than the untreated cells. Thus, the protease-inhibitor prevented much of the atrophy induced by serum deprivation. The serum-deprived fibroblasts also stopped proliferating and accumulated in the G1 phase of the cell cycle. The cells treated with Ep475 accumulated in G1 in a manner identical to untreated serum-deprived cells. Other agents which inhibited protein breakdown in serum-deprived cells also did not prevent the arrest of cell proliferation. Thus the enhancement of proteolysis during serum deprivation appears necessary for the decrease in size and protein synthesis, but probably not for the cessation of cell proliferation. When cells deprived of serum in the presence or absence of Ep475 were stimulated to proliferate by the readdition of serum, the larger Ep475-treated cells began DNA synthesis 1-2 hours later than the smaller untreated cells. Thus, after treatment with Ep475, the rate of cell cycle transit following serum stimulation was not proportional to the cell's size, protein, or RNA content, or rate of protein synthesis.

The control of eucaryotic cell proliferation is governed largely by a series of regulatory events which occur in the G1 phase of the cell cycle. When stimulated to proliferate, quiescent (G0) 3T3 fibroblasts require transcription, rapid translation, and three growth factors for the growth state transition. We examined exponentially growing 3T3 cells to relate the requirements for G1 transit to those necessary for the transition from the G0 to the S phase. Cycling cells in the G1 phase required transcription, rapid translation, and a single growth factor (insulin-like growth factor [IGF] I) to initiate DNA synthesis. IGF I acted post-transcriptionally at a late G1 step. All cells in the G1 phase entered the S phase on schedule if either insulin (hyperphysiological concentration) or IGF I (subnanomolar concentration) was provided as the sole growth factor. In medium lacking all growth factors, only cells within 2 to 3 h of the S phase were able to initiate DNA synthesis. Similarly, cells within 2 to 3 h of the S phase were less dependent on transcription and translation for entry into the S phase. Cells responded very differently to inhibited translation than to growth factor deprivation. Cells in the early and mid-G1 phases did not progress toward the S phase during transcriptional or translational inhibition, and during translational inhibition they actually regressed from the S phase. In the absence of growth factors, however, these cells continued progressing toward the S phase, but still required IGF at a terminal step before initiating DNA synthesis. We conclude that a suboptimal condition causes cells to either progress or regress in the cell cycle rather than freezing them at their initial position. By using synchronized cultures, we also show that in contrast to earlier events, this final, IGF-dependent step did not require new transcription. This result is in contrast to findings that other growth factors induce new transcription. We examined the requirements for G1 transit by using a chemically transformed 3T3 cell line (BPA31 cells) which has lost some but not all ability to regulate its growth. Early- and mid-G1-phase BPA31 cells required transcription and translation to initiate DNA synthesis, although they did not regress from the S phase during translational inhibition. However, these cells did not need IGF for entry into the S phase.

Beta-lapachone is a potent inhibitor of DNA repair in mammalian cells and activates topoisomerase I. We show that beta-lapachone can prevent the oncogenic transformation of CHEF/18A cells by ionizing radiation. Potentially lethal DNA damage repair (PLDR) occurs while x-irradiated cells are held in medium containing low serum prior to replating. PLDR processes permitted survival recovery but also drastically increased the number of foci per plate (i.e., transformation) of CHEF/18A cells. By blocking PLDR with beta-lapachone, both survival recovery and enhanced transformation were prevented. At equivalent survival levels, exposure of x-irradiated cells to beta-lapachone resulted in an 8-fold decrease in the number of foci per dish as compared to the number of transformants produced after PLDR. Early PLDR-derived increases in transformation may be the result of error-prone genetic rearrangements dependent on topoisomerase I, which are thereby prevented by beta-lapachone. Beta-lapachone exposure decreased the rejoining of DNA strand breaks and produced additional double-strand breaks in x-irradiated cells during PLDR. The activation of topoisomerase I by beta-lapachone may convert repairable single-strand DNA breaks into the more repair-resistant double-strand breaks, thereby preventing PLDR and neoplastic transformation. These results suggest a new direction for the development of anticarcinogenic agents.

This paper provides evidence that normal animal cells possess a unique regulatory mechanism to shift them between proliferative and quiescent states. Cells cease to increase in number under a diversity of suboptimal nutritional conditions, whereas a uniformity of metabolic changes follows these nutritional shifts. Evidence is given here that cells are put into the same quiescent state by each of these diverse blocks to proliferation and that cells escape at the same point in G(1) of the cell cycle when nutrition is restored. The name restriction point is proposed for the specific time in the cell cycle at which this critical release event occurs. The restriction point control is proposed to permit normal cells to retain viability by a shift to a minimal metabolism upon differentiation in vivo and in vitro when conditions are suboptimal for growth. Malignant cells are proposed to have lost their restriction point control. Hence, under very adverse conditions, as in the presence of antitumor agents, they stop randomly in their division cycle and die.

The identification of cellular proteins involved in the control of cell proliferation in normal cells is essential for understanding the mechanism underlying growth regulation and cellular transformation. A nuclear protein termed Ki antigen with a relative mobility of 32,000 (Mr 32K) and which is recognized by SLE patient antisera has been identified in cells of human, bovine, and murine origin. Recently, cDNA clones for the bovine and human Ki antigens have been isolated using SLE patient antisera (T. Nikaido, et al., in preparation). The nucleotide sequence predicted a protein of 239 amino acids with a possible nuclear localization signal resembling that identified in SV40 T antigen and other nuclear proteins. Here we show that the expression of Ki antigen is regulated in the normal cell, but not in the transformed cell. Furthermore, in the K-ras temperature-sensitive mutant cell line, ts 371 normal rat kidney (NRK), Ki antigen expression increases several-fold at the permissive temperature relative to the nonpermissive temperature. These results suggest that expression of Ki antigen might be correlated with cellular transformation as well as with cell growth regulation.

Analysis of exponential and synchronous cultures of Escherichia coli B/r after the addition of FL1060 indicates a block point for division by this agent some 15 to 20 min before the end of the preceding cell division cycle, a time corresponding to the beginning of the C period of the cell division cycle. Morphological examination of FL1060-treated synchronous cultures of E. coli /r was consistent with inhibition by FL1060 of a very early event in the cell division cycle. This event appears to be essential for normal cell surface elongation in a rod configuration. Temporary treatment of synchronous cultures of E. coli B/r with FL1060 resulted in division delay, the extent of which was a function of the duration of exposure to FL1060. However, even after relatively long times of FL1060 treatment the delayed divisions were still synchronous. Although FL1060 had no direct effect on deoxyribonucleic acid (DNA) synthesis, the synchronous delayed division occuring after temporary treatment with FL1060 were accompanied by a delay in the attainment of resistance of cell division to inhibitors of DNA, ribonucleic acid, and protein synthesis. These results suggest aht an FL1060-sensitive event initiates at the beginning of the C period of the cell division cycle of E. coli and is responsible for normal cell elongation. This cell elongation pathway procedes independently of DNA synthesis, but there is an interaction between this pathway and termination of a round of DNA replication in which a normal rod configuration is necessary to allow a signal for cell division to be generated upon completion of DNA replication.

Major proteins synthesized by Swiss 3T3 cells at different stages of the cell cycle have been analyzed using double isotope labeling and one-dimensional SDS-polyacrylamide slab gels. The synthesis of actin was previously shown to be markedly enhanced a few hours after quiescent cells initiated growth following addition of serum. In contrast, the synthesis of actin remained at a constant rate, similar to that in quiescent cells, relative to synthesis of other proteins during the entire cell cycle. We conclude that enhanced actin synthesis is a process specific for the G0 to S transit, and may serve as a marker event during this interval. In contrast, three other proteins (90,000, 57,000, and 33,000 daltons) were synthesized throughout the cell cycle at higher rates than in G0 cells, and thus, are markers characteristic of cells traversing the cell cycle. A transient increase, such as seen for actin synthesis, by cells emerging from quiescence, may represent a process that these cells must perform before they can enter the G1 portion of the cell cycle. A transient event such as this need not be a periodic event that occurs during each cycle.

Several of the new 4-quinolones significantly increase the incorporation of [3H]thymidine into the DNA of mitogen-stimulated human lymphocytes. This study suggests that ciprofloxacin inhibits de novo pyrimidine biosynthesis, thereby resulting in a compensatory increase in the uptake of pyrimidine precursors through salvage pathways, and that additional effects may affect eucaryotic cell growth. Incorporation of deoxyuridine, uridine, and orotic acid as well as thymidine was increased in the presence of ciprofloxacin, one of the antibacterially most active of the new 4-quinolones. In contrast, the uptake was decreased in very high concentrations of the drug. Culture in HAT (hypoxanthine, aminopterine, thymidine) medium, which blocks de novo thymidylate synthesis, abrogated the increase in [3H]thymidine incorporation induced by ciprofloxacin. Ciprofloxacin also failed to increase the uptake of [14C]hypoxanthine or leucine, indicating a selective effect on pyrimidine and not on purine nucleotide biosynthesis. N-(Phosphonacetyl)-L-aspartate, an inhibitor of pyrimidine nucleotide biosynthesis, also increased [3H]thymidine incorporation in phytohemagglutinin-stimulated lymphocytes in a fashion similar to ciprofloxacin. The growth of several cell lines was partially inhibited by ciprofloxacin at 20 micrograms/ml and completely inhibited at 80 to 160 micrograms/ml. Growth inhibition by ciprofloxacin could not be restored by the addition of uridine to the medium. Chromosome breaks, gene amplification, or other genetic alterations could not be detected in human lymphocytes incubated with up to 25 micrograms of ciprofloxacin per ml.

Cellular oncogenes are DNA sequences implicated in the genesis of cancer, but their functions in the transformation process are not understood. Our experiments provide data linking expression of two well-studied proto-oncogenes, c-myc and c-rasKi, to current knowledge of proliferation control and its perturbation by differentiation and chemical transformation. Growth stimulation of quiescent cells by serum elevates expression of the myc proto-oncogene in Balb/c 3T3 (A31) cells. In two chemically transformed A31 derivatives (BPA31 and DA31), c-myc expression is constitutive. The levels of c-myc mRNA in quiescent and growing transformed cells are nearly the same, and are only slightly elevated compared to the level found in growing A31 cells. By contrast, c-rasKi expression is cell-cycle-dependent in BPA31 cells. The relative abundance of c-rasKi mRNA begins to increase in mid- to late G0/G1. During terminal differentiation of teratocarcinoma stem cells (F9) into nonproliferating endoderm, relative mRNA abundance is diminished more markedly for c-myc than for c-rasKi. These results demonstrate that expression of the myc and rasKi proto-oncogenes is dependent upon the cellular growth state, and that growth control exhibits growth-factor-dependent, cell-cycle-timed oncogene expression. In the case of the BPA31 cells, c-myc is not rearranged, amplified, or overexpressed. However, the oncogene has lost its cycle-dependent regulation in the chemically transformed cells.

Swiss 3T3 cells arrested in B0 (quiescent state) by reducing serum content of the medium all contain the same amount of DNA but vary in nuclear volume over approximately a twofold range. By use of flow microfluorimetry, scatterplots of nuclear volume versus DNA content were obtained in intervals after serum stimulation. The earliest cells to enter DNA synthesis were those with the largest nuclei, whereas cells with the smallest nuclei were among the latest. Regulation of cellular transit from G0 to the S phase was therefore, at least in part, deterministic, since all G0 cells did not have equal probabilities of entry into S at a given moment. All cells having the same nuclear volume did not initiate DNA synthesis at the same moment; therefore, factors other than nuclear volume must also influence this timing. Nuclear volume correlated with the maximum rate at which cells could enter S. The kinetic model of the cell cycle postulating a probabilistic event as solely responsible for entry into S thus appears too simple.

To characterize the system(s) responsible for degradation of short-lived and long-lived proteins in mammalian cells, we compared the concentrations of ATP required for the degradation of these classes of proteins in growing hamster fibroblasts. By treating CHEF-18 cells with increasing concentrations of dinitrophenol and 2-deoxyglucose, it was possible to reduce their steady-state ATP content by different amounts (up to 98%). These treatments caused a rapid decrease in the degradation of both short- and long-lived proteins. Removal of the inhibitors led to a prompt restoration of ATP and proteolysis. As ATP content fell below normal levels (about 3.1 mM), rates of proteolysis decreased in a graded biphasic fashion. Reduction in ATP by up to 90% (as may occur in anoxia or injury) decreased proteolysis up to 50%; and with further loss of ATP, protein breakdown fell more sharply. Degradation of both classes of proteins was inhibited by 80% when ATP levels were reduced by 98%. The levels of ATP required for the breakdown of short- and long-lived proteins were indistinguishable. Protein synthesis was much more sensitive to a decrease in ATP content than protein breakdown and fell by 50% when ATP was reduced by only 15%. Chloroquine, an inhibitor of lysosome function, did not reduce the degradation of either class of proteins in growing cells, but it did inhibit the enhanced degradation of long-lived proteins upon removal of serum (in accord with previous studies). Thus, in growing fibroblasts, an ATP-dependent nonlysosomal process appears responsible for the hydrolysis of both short- and long-lived proteins.

Effects of compounds that inhibit repair of DNA lesions in cells have been reported frequently. The consequences include altered incidence of carcinogenicity in vivo, tumorigenic transformation of cultured cells, mutations, and increased lethality as well as sister-chromatid exchanges and chromosome aberrations. This literature is reviewed here, with major emphasis on methylxanthines (caffeine in particular) and nicotinamide analogs. Existing information is also summarized on a novel potent repair inhibitor, beta-lapachone. Compounds that inhibit both DNA replication and repair are not discussed in detail since they have been reviewed often, but miscellaneous inhibitors of repair are summarized in a table. The relatively small number of experiments performed on the anticarcinogenic effects of methyl-xanthines and nicotinamide analogs gave very conflicting results. Some investigators report decreased carcinogenicity of DNA-damaging agents when caffeine was provided, but others obtained the opposite effect. The three studies with nicotinamide analogs all reported enhanced tumorigenicity of carcinogens. The data are too few to enable firm conclusions to be drawn regarding the possibility of using repair inhibitors to prevent cancer in humans. Variations of experimental conditions, carcinogens, cells, etc. have provided conflicting results. The possibility of cancer prevention is, nevertheless, so important that further investigations with DNA-repair inhibitors, particularly with human cells, seem very well justified.