Caffeine-Mediated Override of Checkpoint Controls: A Requirement for rhp6 (Schizosaccharomyces pombe)

Corresponding author: Roy Rowley, Experimental Oncology, Department of Radiation Oncology, University of Utah Medical Center, 50 North Medical Drive, Salt Lake City, UT 84132., roy.rowley{at}hsc.utah.edu (E-mail)

Communicating editor: P. G. YOUNG

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Cells exposed to inhibitors of DNA synthesis or suffering DNA damage are arrested or delayed in interphase through the action of checkpoint controls. If the arrested cell is exposed to caffeine, relatively normal cell cycle progression is resumed and, as observed in checkpoint control mutants, loss of checkpoint control activity is associated with a reduction in cell viability. To address the mechanism of caffeine's action on cell progression, fission yeast mutants that take up caffeine but are not sensitized to hydroxyurea (HU) by caffeine were selected. Mutants 788 and 1176 are point mutants of rhp6, the fission yeast homolog of the budding yeast RAD6 gene. Mutant rhp6-788 is slightly HU sensitive, radiosensitive, and exhibits normal checkpoint responses to HU, radiation, or inactivation of DNA ligase. However, the addition of caffeine does not override the associated cell cycle blocks. Both point and deletion mutations show synthetic lethality at room temperature with temperature-sensitive mutations in cyclin B (cdc13-117) or the phosphatase cdc25 (cdc25-22). These observations suggest that the rhp6 gene product, a ubiquitin-conjugating enzyme required for DNA damage repair, promotes entry to mitosis in response to caffeine treatment.

IN a wild-type cell, entry to mitosis is blocked if the cell harbors DNA damage or if DNA replication is incomplete. The mechanisms that prevent cell cycle progress under these circumstances are termed checkpoint controls (for reviews see HARTWELL and WEINERT 1989 ; ELLEDGE 1996 ). Such controls are activated by irradiation or treatment with hydroxyurea (HU), an inhibitor of ribonucleotide reductase. Two conditions allow mitosis to occur without repair of DNA damage or completion of DNA synthesis: mutational inactivation of the checkpoint control (WEINERT and HARTWELL 1988 ; ENOCH and NURSE 1990 ; AL-KHODAIRY et al. 1992; ROWLEY et al. 1992 ) and exposure to caffeine or a caffeine-like agent (WALTERS et al. 1974 ; SCHLEGEL and PARDEE 1986 ; DAS 1987 ; ROWLEY 1992 ). The affected cell is rendered sensitive to killing by the clastogen or DNA synthesis inhibitor in question, presumably through failure to complete restitution of the genome before defects are rendered permanent during DNA replication or chromosome segregation (BOSE et al. 1978 ; LUCKE-HUHLE 1982 ; SAWECKA et al. 1987 ).

The process by which caffeine modifies checkpoint functions remains unknown. Most attempts to define mechanisms have involved monitoring the effects of agents with more or less well-defined points of action on cell cycle progression after exposure to radiation or chemotherapeutic agents [summarized in ROWLEY 1992 and WANG et al. 1996 ]. Agents that reduce the duration of mitotic delay consequent to DNA damage include caffeine, theophylline, theobromine (WALTERS et al. 1974 ; KIMLER et al. 1978 ), pentoxyfylline (FAN et al. 1995 ), staurosporine and 7-hydroxystaurosporine (WANG et al. 1996 ), okadaic acid (YAMASHITA et al. 1990 ), fostriecin (ROBERGE et al. 1994 ), calyculin A (NAKAMURA and ANTOKU 1994 ), and 2-aminopurine (ANDREASSEN and MARGOLIS 1992 ; ROWLEY 1992 ; TAM and SCHLEGEL 1992 ). All affect the state of protein phosphorylation, and 2-aminopurine shows some specific activity against the Cdc2-related kinase PITSLREß (XIANG et al. 1994 ). Caffeine may therefore inhibit the phosphorylation of a protein or proteins that is/are normally phosphorylated in response to a signal to halt progression into mitosis, i.e., a checkpoint protein. This has been observed. Chk1 is a protein kinase, initially identified in fission yeast, required for the induction of the DNA-damage-dependent checkpoint control (WALWORTH et al. 1993 ). After irradiation, Chk1 is phosphorylated and activated (WALWORTH and BERNARDS 1996 ), in turn phosphorylating and inactivating the tyrosine phosphatase Cdc25, causing the latter to be sequestered. In the unirradiated cell, Cdc25 dephosphorylates and activates the mitotic regulator Cdc2. Chk1-mediated inactivation of Cdc25 thus blocks entry to mitosis (FURNARI et al. 1997 ; PENG et al. 1997 ; SANCHEZ 1997). Using in vitro egg extracts, KUMAGAI et al. 1998 examined the effect of caffeine on mitotic delay and phosphorylation of Xenopus Chk1. Mitotic delay induced by UV irradiation or the presence of unreplicated DNA was abolished, and Chk1 phosphorylation was inhibited; however, it was only possible to conclude that caffeine disrupts the checkpoint pathway responsible for Chk1 phosphorylation, and no specific caffeine target was identified. Previous efforts to identify a caffeine target have focussed on Cdc2. Also using Xenopus egg extracts, SMYTHE and NEWPORT 1992 showed that caffeine both reversed aphidicolin-induced inhibition of DNA synthesis and decreased the rate of Cdc2 tyrosine phosphorylation that usually accompanies induction of the S-phase checkpoint. By then using a kinase-inactive Cdc2 substrate (lacking cyclin B), the tyrosine kinase targetted was shown not to be Cdc2 itself, but rather an unidentified kinase upstream. After X irradiation in V79 hamster cells, caffeine-mediated override of the DNA-damage checkpoint also correlated with an increase in Cdc2 kinase activity, but whether this was a cause or consequence of checkpoint override was not established (HAIN et al. 1993 ).

Caffeine thus acts by modifying the function of a cell cycle progression control, raising the possibility that, with a suitable selection procedure, a mutant for that control might be identified. Using the fission yeast, a genetic approach was selected accordingly. Mutagen-treated cells were exposed to a cytotoxic combination of HU plus caffeine, using concentrations that are separately tolerated by the parental strain. Four mutants have been identified, defining three genes, which cannot be sensitized to HU by caffeine, but which take up caffeine and HU and retain a normal checkpoint response to HU in both the absence and presence of caffeine. We have identified a genetic defect common to two mutants (788 and 1176) and report data to support the hypothesis that caffeine overrides checkpoint control functions by activation of a mitotic promoter, Rhp6.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

See Table 1 for a list of strains used in this work.

Cell culture:
Fission yeast cells (Schizosaccharomyces pombe) were routinely maintained in YPD (2% glucose, 1% yeast extract, 2% Bacto-peptone) or on YPD-agar (YPD plus 2% Bacto-agar). Cells growing with dependence on a plasmid were maintained in Edinburgh minimal medium (EMM2) or in EMM2 plus 2% agar, formulated as described in ALFA et al. 1993 . Genetic procedures were performed according to this manual. To make crosses between the rhp6 deletion mutant and cdc mutants, the deletion mutant was first transformed with pWH5rhp6⁺ to restore normal mating efficiency, and then the spores were plated on YPD to allow plasmid loss.

Hydroxyurea and caffeine treatments:
Stock solutions of HU and caffeine were made up in medium at 200 mM concentration and diluted appropriately into liquid medium or medium plus agar before gelling (i.e., at 50°).

Irradiation:
For viability experiments, cells were exposed to gamma rays in a cesium irradiator (Shepherd and Associates, San Fernando, CA) at a dose rate of 29.5 Gy/min on ice. For cell cycle experiments, cells were exposed at a dose rate of 7.3 Gy/min, maintaining cells at growth temperature with insulation.

Heat treatments:
To measure cell sensitivity to killing by incubation at 36°, 50 ml of cell suspension in appropriate medium held in a 150-ml conical flask was transferred to a water bath at 36°. Samples were withdrawn at intervals and plated at the appropriate density on solid medium.

Cell cycle progression assays:
To score the fraction of cells undergoing septation, 0.5 ml of cell suspension was fixed and stained by the addition of 0.5 ml iodine in 0.075 M potassium iodide. The cells were then resuspended in fixative and viewed by bright-field optics at x400 magnification. To score the fraction of binucleate cells, cell samples were fixed in 70% ethanol, washed with water, and then stained with DAPI and viewed by UV-induced fluorescence at x1000 magnification.

Survival assays:
Cells were plated on solid medium, and the fraction of the plated population able to form macroscopic colonies was scored. Survival data are presented without normalization to the colony-forming ability of the control (untreated) population.

Mutagen treatment:
Cells were suspended in 2% (w/v) ethyl methanesulfonate (EMS) dissolved in YPD. After 3 hr at room temperature, the cells were washed three times then allowed to recover for 2 hr before plating on selective medium.

Caffeine uptake:
Cells, >10⁷ in exponential growth, were suspended in 1 ml of temperature-adjusted YPD containing 1 µCi of [¹⁴C]caffeine (Dupont/New England Nuclear). At intervals after suspension, the cells were then pelleted at 10,000 x g, the supernatant was aspirated off, and the pellet was dissolved for liquid scintillation counting of isotopic decays.

Gene cloning:
Mutant 788 was transformed with a fission yeast genomic library in plasmid pWH5 (kindly supplied by D. Beach) using the lithium acetate procedure (ALFA et al. 1993 ). Transformants were plated on EMM2, and the plates were incubated at 36°. Surviving cells (colonies) were picked, subcloned (x3), and retested for survival at 36°. Plasmid was rescued from mutant 788 and retransformed into the mutant to retest for complementation. The plasmid contained a 3.4-kb genomic DNA insert that was sequenced in its entirety by the Huntsman Cancer Institute sequencing facility. The sequence was matched to those in the EMBL and Genbank DNA sequence databases using Intelligenetics DNA sequence analysis software. The rhp6 gene was deleted as described in ALFA et al. 1993 . For this purpose, a construct (pRR394) was kindly provided by L. Prakash, as described in REYNOLDS et al. (1990). The plasmid pRR394 contains the 3.2-kb HindIII complementing DNA fragment in which the rhp6 gene has been replaced with the ura4⁺ gene from nucleotides -93 to +795 (relative to the first ATG in the rhp6 open reading frame). The deletion was verified by PCR amplification using primers that flank the deleted region. A 1.5-kb product was amplified from the parent strain, and a 2.4-kb product from the deletion mutant, consistent with the size of the deletion construct. An overexpression construct was made by PCR amplification of the rhp6 gene from nucleotides -177 to +1329 using primers GTGTTGGGTGCGCCCCATT and CGATGGC TCATTAGCTTCACC and the Expand long template PCR system (Boehringer Mannheim, Indianapolis, IN), which contains both Taq and Pwo polymerases to provide high-fidelity replication. The 1.3-kb fragment produced was blunt ended with pfu polymerase (Stratagene, La Jolla, CA) and cloned into the SmaI site of the thiamine-regulatable expression vector pREP1 (MAUNDRELL 1990 ). To determine the nucleotide sequence of the rhp6 gene in mutants rhp6-788 and rhp6-1176, genomic DNA was extracted (ALFA et al. 1993 ) and the rhp6 gene was amplified by PCR, using the aforementioned primers and high-fidelity polymerase. The products of three separate reactions were then sequenced.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Effect of caffeine on the response of S. pombe to HU:
As previously shown (ENOCH and NURSE 1990 ), HU blocks DNA synthesis in S. pombe and arrests cell cycle progress to mitosis (Figure 1). In the parental strain exposed to 10 mM HU in YPD, the fraction of cells with septa decreases over a period of 4–6 hr (i.e., cells stop dividing), and cell length increases consistent with arrested cell progress. After exposures of 4 hr or more, a few cells attempt to divide without completing DNA synthesis then arrest during septum deposition. To demonstrate the effect of caffeine on the HU-induced cell cycle block, 5 mM caffeine was added to cells either at the same time as HU (not shown) or 5 hr after the start of HU treatment (Figure 1). The latter procedure provides a clearer demonstration of caffeine's effect. Cells first accumulate at the checkpoint, then are released by caffeine treatment as a semisynchronous wave, and arrest attempting division without completing DNA synthesis. The terminal phenotype of these cells is similar to that of fission yeast checkpoint mutants exposed to HU alone (ENOCH and NURSE 1990 ; AL-KHODAIRY and CARR 1992 ; ROWLEY et al. 1992 ). Caffeine alone (5 mM) does not significantly alter cell cycle progression during the period of observation (Figure 1).

View larger version (29K): In this window In a new window Download PPT slide   Figure 1. The percentage of cells undergoing septation (dividing), monitored at intervals after the start of treatment. Cells were maintained and treated in YPD. Solid line, untreated control; dashed line, 5 mM caffeine; solid circles, 10 mM HU; open circles with dashed line, 10 mM HU followed at 5 hr by addition of 5 mM caffeine (arrow). Data are shown for the parent cell line (wt) and three mutants, 788, 1460 and 1519, as indicated.

To test the effect of caffeine on HU-induced cell killing, cells were plated on YPD containing concentrations of HU from 0 to 10 mM, with or without 5 mM caffeine, and incubated for colony formation. Caffeine causes a marked increase in HU sensitivity (Figure 2). Caffeine alone (5 mM) is not cytotoxic (see the zero HU dose point on the HU plus caffeine curve in Figure 2).

View larger version (27K): In this window In a new window Download PPT slide   Figure 2. The fraction of cells surviving treatment (able to form colonies) for cells grown on YPD-agar containing hydroxyurea (HU) alone () or HU plus 5 mM caffeine (). Data are shown for the parent cell line (wt) and three mutants, 788, 1460, and 1519, as indicated.

Effect of caffeine on the response of S. pombe to ionizing radiation:
Ionizing radiation exposures activate the DNA damage checkpoint, causing a temporary decrease in the fraction of cells undergoing septation (Figure 3 and ROWLEY et al. 1992 ). Addition of 10 mM caffeine at 60 min after a 100-Gy radiation exposure releases the checkpoint-mediated cell cycle block; however, the effect is not immediate and is difficult to demonstrate at lower caffeine concentrations. The effect of caffeine is also masked at lower radiation doses when the time at which cells recover naturally from the block becomes indistinguishable from the time at which cells recover in the presence of caffeine.

View larger version (24K): In this window In a new window Download PPT slide   Figure 3. The fraction of cells undergoing nuclear division (i.e., with two nuclei per cell), monitored at intervals after the time of irradiation. or , parent strain; or , mutant 788; or , 100 Gy alone; or , 100 Gy plus 10 mM caffeine added at 60 min after irradiation (arrow).

To demonstrate an effect of caffeine on radiation sensitivity for cell killing, cells were irradiated on ice then plated for colony formation on YPD containing 5 mM caffeine (Figure 4). A marked increase in radiosensitivity is observed, such that the survival curve D₀ (reciprocal of the slope constant for the exponential portion of the curve, LEA 1956 ) decreases from 930 to 320 Gy.

View larger version (19K): In this window In a new window Download PPT slide   Figure 4. The fraction of cells surviving irradiation. or , parent strain; or , mutant 788; or , cells plated on YPD alone; or , cells plated on YPD containing 5 mM caffeine.

Effect of caffeine on the temperature sensitivity of cdc17-K42:
The cdc17-42 mutant is temperature sensitive for DNA ligase function so that cell cycle progress is arrested at 36°, and the cells elongate and die. Loss of ligase function generates single-stranded DNA breaks (NASMYTH 1977 ) and activates the DNA-damage-dependent checkpoint control (WEINERT and HARTWELL 1988 ; ROWLEY et al. 1992 ). Caffeine treatment should therefore increase the rate of cell killing induced by holding cells at 36°. The following is observed: cells incubated in suspension at 36° lose colony-forming ability with a half-time of 2 hr, while cells held in the presence of 10 mM caffeine lose viability with a halftime of ~1 hr (Figure 5). Consistent with caffeine-induced checkpoint override, at 24 hr after shifting to 36°, cells are an average of 29.5 µm in length (±11.1 SEM) in the absence of caffeine and 15.8 ± 7.3 µm in length in the presence of caffeine.

View larger version (17K): In this window In a new window Download PPT slide   Figure 5. The fraction of cells surviving culture at 36°. or , the temperature-sensitive DNA ligase mutant cdc17-K42; or , a cross between 788 and cdc17-K42 (i.e., rhp6-788 cdc17-K42); or , cells held at 36° in YPD; or , cells held at 36° in YPD containing 10 mM caffeine.

Identification of mutants that are not sensitized to HU:
The preceding results indicate that caffeine overrides checkpoint control functions and sensitizes the fission yeast to killing induced by inhibitors of DNA synthesis or clastogens, as observed in other eukaryotes (WALTERS et al. 1974 ; KIMLER et al. 1978 ; SCHLEGEL and PARDEE 1986 ; DAS 1987 ). Accordingly, we have used the fission yeast as a model system to investigate how caffeine overrides checkpoint control functions. A particular advantage of using S. pombe is that it allows a genetic approach to the problem through the identification of relevant mutants. To this end, it was assumed that caffeine sensitizes cells to HU- or radiation-induced killing by overriding checkpoint controls. Therefore, mutants that remain resistant to HU- or radiation-induced killing in the presence of caffeine might also retain checkpoint control functions in the presence of caffeine. Such mutants are potentially defective for a caffeine-activated mitotic promoter, or they contain a modified checkpoint protein that is no longer inactivated by caffeine. Mutants were selected by resistance to HU plus caffeine rather than radiation plus caffeine because the magnitude of caffeine-induced sensitization is greatest for HU-induced killing, and because caffeine appears to override the HU-induced checkpoint control more readily and at a lower concentration than that required for the radiation-induced checkpoint control.

Mutagen-treated cells (leu1-32 h^-) were plated on YPD-agar containing 8 mM HU and 5 mM caffeine (HU/caffeine). This treatment combination is cytotoxic, while exposure to either agent alone at these concentrations is not (Figure 2). More than 1600 clones that survived HU/caffeine selection were obtained. As resistance to HU/caffeine treatment may result from resistance to HU alone, clones were patched onto YPD containing 12 mM HU, the lowest concentration that will discernibly block reproduction of the parental strain by this assay. Surviving clones were presumed to be resistant to HU alone through altered HU uptake or metabolism and were set aside. Five clones were resistant only to HU/caffeine: 788, 948, 1176, 1460, and 1519. All have been subcloned and tested for altered caffeine uptake, an additional potential cause of resistance to HU/caffeine treatment. Table 2 shows isotopic decays per minute in cells suspended in YPD liquid medium containing 1 µCi/ml of [¹⁴C]caffeine and then pelleted after 30 min. All five clones and the parent cell lines take up caffeine in approximate proportion to their cellular volume, as calculated from graticule measurements of length and diameter. The data suggest that none excludes caffeine. Caffeine uptake by the parent line and by mutant 788, the focus of this paper, was also measured at 10 min (30°) and with the cells held on ice, in the event that the level of radioactive label measured at 30 min reflects the accumulation of an inactive metabolite. Caffeine uptake was essentially unchanged from 10 to 30 min, thus label does not accumulate, and holding cells at a temperature where drug metabolism is likely to be slow did not reduce uptake.

Mutants 788 and 1176 show a reduced mating efficiency, and mutant 1519 is sterile; however, complementing genomic DNAs have been obtained for 788, 1176, and, very recently, 1519, allowing mating and standard genetic analyses. In the process, 1176 was found to be allelic to 788 (described below). Mutants 788, 948, 1460, and 1519 were crossed with the parent strain to determine whether the mutant phenotypes segregate as single gene defects. Asci (20) from crosses with 788, 1460, and 1519 yielded two wild-type and two HU/caffeine-resistant spores, consistent with single-locus defects. Asci from matings with mutant 948 yielded inconsistently viable spores and cells with at least two phenotypes. Further work on 948 has been suspended. To show that the mutants represent distinct genetic defects, mutants 788 and 1460, 788 and 1519, as well as 1460 and 1519 were crossed, and spores were plated. All three crosses yielded progeny with the wild-type morphology (788 and 1460 are markedly elongated, 1519 is near spherical) that were sensitive to the combination treatment of 8 mM HU plus 5 mM caffeine, i.e., the parental phenotype segregated out.

Mutant responses to HU and HU plus caffeine:
To confirm the results of the mutant selection procedure, cells were tested both for sensitivity to HU with or without caffeine and for the effects of HU and caffeine on cell cycle progression. Mutants 788 and 1519 are slightly more sensitive to HU alone than the parent when grown on YPD containing HU, while mutant 1460 is significantly more sensitive; however, none is further sensitized by the additional presence of 5 mM caffeine, consistent with the procedure used for mutant selection (Figure 2). To test for checkpoint override by caffeine, cells were exposed to 10 mM HU in YPD and, as with the parent strain, 5 mM caffeine was added after 5 hr. HU alone causes a slow decrease in the fraction of cells with septa (Figure 1), and cells elongate. The mutants therefore retain the replication-dependent checkpoint control; however, caffeine exposure does not result in a resumption of cell cycle progress. To confirm this observation, mutant 788 was also exposed to 10 mM caffeine 5 hr after the start of HU treatment. Again, there was no resumption of cell cycle progress. The object of the selection procedure was to identify mutants that (1) exhibit normal checkpoint responses, (2) retain these controls in the presence of caffeine, and (3) take up caffeine at least as well as the parent. Mutants 788, 1460, and 1519 meet these criteria. We now report a characterization of mutant 788.

Effect of caffeine on the response of mutant 788 to ionizing radiation:
The dose vs. response curve for radiation-induced killing of mutant 788 is steeper than that of the parental cell (D₀ = 170 Gy) and is unaltered by incubating cells on 5 mM caffeine after exposure (Figure 4). The mutant is thus radiosensitive and cannot be further sensitized by caffeine treatment, as was observed for the mutant response to HU. Similarly, the radiation-induced mitotic block is not reversed by adding 10 mM caffeine to the cells 60 min after irradiation (Figure 3).

Effect of caffeine on the temperature sensitivity of mutant 788 in a cdc17-K42 background:
Mutant 788 was crossed with the temperature-sensitive DNA ligase mutant cdc17-K42, and the effect of 10 mM caffeine on the temperature sensitivity of the double mutant was determined by plating cells for colony formation at intervals after shifting a suspension culture in YPD to 36° with or without the addition of caffeine. Caffeine does not alter the rate at which cells lose viability (Figure 5).

Heat sensitivity of mutant 788:
Unlike the parent cell, growth of mutant 788 at 36° is lethal (not shown). The heated cell stops dividing, elongates, branches, and becomes misshapen. Viability decreases with a half-time of ~3 hr. This phenotype is apparently more marked for the deletion mutant, as it will not grow at 30°.

Synthetic lethality with cell cycle mutants:
Mutant 788 grows slowly (Table 3) and does not respond to caffeine treatment by movement past the HU-induced checkpoint (Figure 1). These characteristics suggest the mutant harbors a defect in cell cycle progression control, possibly in G₂, as the distribution of cells about the cell cycle is not significantly different from the parent as determined by flow cytometric estimation of DNA content (not shown). The nature of this possible defect was investigated by crossing 788 with known cell cycle or checkpoint mutants. In the fission yeast, progress from G₂ into mitosis is regulated by the serine/threonine kinase Cdc2. Kinase activation requires association with cyclin B (SOLOMON et al. 1990 ), phosphorylation at threonine-167 (GOULD et al. 1991 ), and dephosphorylation at tyrosine-15 by the phosphatase Cdc25 (RUSSELL and NURSE 1986 ; GAUTIER et al. 1988 ; DUNPHY and KUMAGAI 1991 ). Cdc25-mediated activation is counteracted by the tyrosine kinase wee1, which phosphorylates Cdc2 at tyrosine-15 and threonine-14 (RUSSELL and NURSE 1987 ; FEATHERSTONE and RUSSELL 1991 ; PIWNICA-WORMS et al. 1991 ). The damage-responsive G₂ checkpoint control blocks progress to mitosis by inactivation of this control pathway. Specifically, DNA damage activates the kinase Rad3, which in turn activates the Chk1 kinase to phosphorylate Cdc25, allowing sequestration, probably by the 14-3-3 proteins Rad24 and Rad25 (FURNARI et al. 1997 ; PENG et al. 1997 ; SANCHEZ et al. 1997 ). Through its phosphorylation, Cdc25 is blocked from interacting with and activating Cdc2. Mutant 788 was therefore crossed with mutant rad3-136; the temperature-sensitive mutants cdc25-22, cdc13-117 (cyclin B), cdc2-33, and wee1-50; and the cdc2 mutants cdc2-1w and cdc2-3w. The latter cdc2 mutants are constitutively "wee" and enter mitosis prematurely (RUSSELL and NURSE 1986 ). Novel phenotypes were observed only for crosses with cdc13-117 and cdc25-22, such that spores germinated at the permissive temperature formed microcolonies of elongated cells and then ceased growth. The colonies depicted (Figure 6) had been cultured at room temperature for >2 wk. The phenotype of the rhp6-788 cdc25-22 mutant closely resembles that of the mutant spc1 cdc25-22 (SHIOZAKI and RUSSELL 1995 ), a MAP kinase-like protein. Spc1 is potentially activated by the MAP kinase kinase Wis1 and is inactivated by the phosphatases Pyp1 and Pyp 2. Mutant 788 was therefore crossed with a pyp1 deletion mutation; however, no novel phenotype emerged.

View larger version (68K): In this window In a new window Download PPT slide   Figure 6. (a) A microcolony resulting from a cross between mutant 788 (rhp6-788) and cdc13-117. The colony was grown at room temperature (~21°) from a single spore on YPD-agar for ~2 wk. (b) A microcolony resulting from a cross between mutant 788 (rhp6-788) and cdc25-22. The colony was grown at room temperature (~21°) from a single spore on YPD-agar for ~2 wk. Bar, 50 µm.

Cloning a complementing genomic DNA fragment:
The heat sensitivity of mutant 788 provided a phenotype with which to clone a complementing genomic DNA sequence. A genomic DNA library in pWH5 was transformed into 788, and the transformants were plated on minimal medium at 36°. Six colonies were obtained, and plasmid was rescued from one. The plasmid contained a 3.4-kb HindIII fragment that was sequenced in its entirety. The fragment includes the complete open reading frame of the rhp6 gene (rad 6 homolog pombe; REYNOLDS et al. 1990 ). PCR amplification and sequencing of both the genomic DNA and 390-base cDNA from mutant 788 revealed a G-to-A base transition at nucleotide 165 (numbered sequentially from the first base of the first ATG in the rhp6 open reading frame), changing a tryptophan residue to a stop codon. Mutant 1176, identified using the same selection procedure, possesses the same phenotype and is complemented by transformation with pWH5rhp6⁺. The rhp6 gene from 1176 contains a G-to-A base transition at nucleotide 488, a splice donor site, and the cDNA is 495 bases in length, consistent with failure to excise the second intron. Mutants 788 and 1176 are, hence, termed rhp6-788 and rhp6-1176 from this point on.

Transformation of rhp6-788 with pWH5rhp6 restored caffeine-induced HU sensitization (Figure 7A) and the ability to grow at 36°; however, it did not restore the ability of the HU-treated cell to resume cell cycle progression after the addition of caffeine. We suspected this might be a consequence of using minimal medium, which is required instead of YPD for plasmid maintenance, and indeed the same result was obtained using the parent cell carrying the empty plasmid (not shown). These experiments were therefore repeated, but with the exception that cells were switched to YPD immediately before treatment with HU. Under these conditions, both the parental strain carrying the plasmid alone (not shown) and rhp6-788 transformed with pWH5rhp6 are released from the HU-induced cell cycle block and attempt division after caffeine addition (Figure 7B). A similar set of observations was made using radiation. It is significant that caffeine can induce sensitivity to HU in minimal medium, without effects on the checkpoint-control-mediated cell cycle block. It implies that HU induces potentially lethal damage, presumably DNA damage, that is repaired by a caffeine-sensitive process. HU is known to induce DNA strand breaks, chiefly in the DNA of S-phase cells, but also in DNA that is not undergoing replication (LI and KAMINSKAS 1987 ; LONN and LONN 1988 ). Caffeine, on the other hand, is known to bind single-stranded DNA (TS'O and LU 1964 ) and to block excision repair in Escherichia coli (METZGER 1964 ), recombinational repair in the fission yeast (PHIPPS et al. 1985 ), and postreplication repair in both mammalian cells (LEHMANN 1972 ) and yeast (PRAKASH 1981 ).

View larger version (18K): In this window In a new window Download PPT slide   Figure 7. (a) The fraction of rhp6-788 leu1-32 pWH5rhp6+ cells surviving treatment (able to form colonies) for cells grown on minimal medium-agar containing HU alone () or HU plus 5 mM caffeine (). (b) The percentage of cells undergoing septation (dividing), monitored at intervals after the start of treatment. Cells were exposed throughout the experiment to 10 mM HU. Caffeine (5 mM) was added at hour 5 (indicated). Solid line, the parental strain grown and treated in YPD (data from Figure 1); , rhp6-788 leu1-32 pWH5rhp6 grown in minimal medium and then transferred to YPD for the duration of the experiment; , rhp6-788 leu1-32 pWH5rhp6+ grown and treated in minimal medium.

Gene deletion and over expression:
The rhp6 gene was deleted by replacing the wild-type rhp6 gene with the ura4 gene (REYNOLDS et al. 1990 ). The mutant rhp6::ura4 has a low plating efficiency (~10^-1), exhibits the same morphological abnormalities as rhp6-788, and is heat sensitive. The deletion mutant is also sensitive to killing by caffeine; for this reason, it has not been possible to test the effects of caffeine on HU sensitivity or radiosensitivity. When crossed with mutants cdc13-117 and cdc25-22, microcolonies identical to those described for rhp6-788 formed over a period of weeks.

Overexpression of rhp6 was achieved by inserting the gene downstream of the thiamine-regulatable promoter of nmt1 (MAUNDRELL 1990 ). Transcription induction was monitored by Northern analysis and was maximal by 14 hr after shifting cells into thiamine-free medium, although complementation of rhp6-788 occurred without transcription induction. If Rhp6 is a rate-limiting mitotic promoter, then high-level expression in mutants that lack mitotic or checkpoint controls might induce premature mitotic events (mitotic catastrophe); however, expression of rhp6 only slowed colony formation of the mutants wee1-50, cdc2-3w, or rad3-136 compared with the same mutants transformed with the plasmid alone, and did not alter viability or morphology.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The object of this investigation is to determine how caffeine treatment overrides checkpoint controls. Four mutants have been identified, representing three distinct genes, which remain resistant to HU toxicity in the presence of caffeine. None of the four excludes caffeine, and all retain the DNA-replication-responsive and DNA-damage-responsive checkpoint controls when exposed to caffeine. Such mutants are potentially defective for a cell component or pathway required for caffeine to override checkpoint controls. Two mutants harbor point mutations in rhp6, the fission yeast homolog of the budding yeast RAD6 gene (REYNOLDS et al. 1990 ). Mutant rhp6-788 is both radiosensitive and HU sensitive and, as indicated by its slow growth and genetic interactions with regulators of Cdc2, may be involved in cell cycle regulation.

Rhp6 is a ubiquitin-conjugating enzyme (JENTSCH et al. 1987 ; REYNOLDS et al. 1990 ). Ubiquitin-mediated proteolysis is a multistep process requiring activating enzymes (generically termed E1 enzymes), E2 or ubiquitin-conjugating enzymes, and E3 or ubiquitin ligases. Ubiquitin, a 76-amino-acid protein, is adenylylated by an E1 enzyme and transferred to a thiol group on the E1 enzyme to form a ubiquitin-E1 thioester. The "activated" ubiquitin is transferred to a cysteine residue on an E2 enzyme (cysteine-88 on Rad6), which then either acts alone to couple ubiquitin to the ultimate acceptor protein or acts in concert with an E3 enzyme. The E3 enzyme, or ubiquitin ligase, may confer substrate specificity. In budding yeast, for example, the ubiquitin ligase Ubr1 targets proteins with a basic amino acid residue at the amino terminal (N-end rule proteolysis). The protein so tagged is then degraded by the 26S proteasome (BACHMAIR et al. 1986 ; BARTEL et al. 1990 ; DOHMEN et al. 1991 ; VARSHAVSKY 1996 ). All known biological functions of Rad6 are lost if cysteine-88 is changed to alanine or valine (SUNG et al. 1990 ). Mutation of this residue results in a protein devoid of ubiquitin-conjugating activity; thus, all biological functions involve ubiquitination. Both the budding and fission yeast mutants grow slowly, mate poorly, and are sensitive to a wide range of clastogens, including ionizing radiation and UV. The mutants are unable to perform postreplication repair (PRR), an error-prone DNA repair process required to fill gaps that are left in the nascent DNA strand opposite sites of DNA damage in the parental strand (for definition see LEHMANN 1972 ). Biochemical functions are best characterized for the budding yeast protein; however, the fission yeast protein rhp6 exhibits 77% identity with RAD6 (REYNOLDS et al. 1990 ), completely complements the radiation sensitivity of the budding yeast mutant (PRAKASH 1994 ), and partially restores sporulation defects (REYNOLDS et al. 1990 ); thus, their functions are likely to be similar. Rad6 has been shown by coimmunoprecipitation to form a heterodimer with Rad18, a single-stranded DNA-binding protein also required for PRR (PRAKASH 1981 ; BAILLY et al. 1994 , BAILLY et al. 1997 ). The dimer retains ubiquitin-conjugating activity and, as suggested by BAILLY et al. 1997 , may degrade components of the DNA replication machinery stalled at sites of damage, thereby allowing PRR. Only two in vivo substrates of Rad6 are currently known: the G-alpha subunit of the budding yeast heterodimeric G protein and the RNA polymerase of the Sindbis virus (VARSHAVSKY 1996 ). Rad6 also binds to Ubr1, an E3 enzyme, but does so to the exclusion of Rad18 (BAILLY et al. 1994 ). Whether Ubr1 is also necessary for PRR has not been demonstrated, but the mutant is not UV sensitive, indicating that it is not (A. VARSHAVSKY, personal communication). Mutations in UBR1 interact lethally with mutations in SLN1, consistent with shared or overlapping functions. The SLN1 gene product is a two-component environmental sensor (OTA and VARSHAVSKY 1993 ) that reports to the HOG1 MAP kinase signal cascade in response to changes in osmolarity (MAEDA et al. 1995 ). Sln1 thus regulates cell cycle progression in response to stress, and a similar role is therefore implied for Ubr1. This conclusion is supported by the observation that a rad6 mutation in which proline-64 is altered to serine produces a mutant that is temperature sensitive for proliferation but normally sensitive to UV irradiation (ELLISON et al. 1991 ). Rad6 appears then to serve at least two functions: repair of DNA damage and regulation of cell cycle progression.

Our data support similar roles for Rhp6 in that rhp6-788 is radiosensitive and HU sensitive, suggesting a DNA repair function, and the mutant exhibits a cell cycle defect, viz. it is slow growing, and crosses with mutants defective for the regulation of Cdc2 are synthetically lethal. A dual role for Rhp6 also fits well with the function of checkpoint controls. Activation of the repair function by the presence of DNA damage would switch the role of Rhp6 from progression to repair. This would facilitate the action of checkpoint controls when DNA damage is incurred. Conversely, when damage has been repaired, reactivation of the progression function would terminate checkpoint-mediated cell cycle delays. The notion that termination of a checkpoint-mediated cell cycle delay involves activation of a mitotic promoter is not new (for review see RUDERMAN 1993 ), e.g., fission yeast Slp1 is required for recovery from radiation- induced G₂ arrest (MATSUMOTO et al. 1997). How Rhp6 might promote cell cycle progress is unclear, but the genetic interactions observed here between mutations in rhp6 and mutations in cyclin B and the phosphatase Cdc25 suggest that Rhp6 regulates Cdc2. Certainly there is precedent for ubiquitin-mediated cell cycle regulation. The fission yeast E3 (ubiquitin ligase) enzyme Pub1 is required for degradation of the mitotic promoter Cdc25, and, while a role in the G₂ checkpoint control seems likely, it has not been reported (NEFSKY and BEACH 1996 ). Two other members of the ubiquitin proteolysis pathway do exhibit characteristics of checkpoint control components. The rad31 product is an E1 homolog (SHAYEGHI et al. 1997 ). The mutant is sensitive to radiation (UV and X) but not to hydroxyurea, and, though it retains the DNA-damage-responsive and DNA replication-responsive checkpoint controls, double mutants with wee1-50, cdc10-, and cdc22- exhibit the rapid death phenotype characteristic of checkpoint mutants (ENOCH and NURSE 1990 ; ROWLEY et al. 1992 ). The hus5 product, like Rhp6, is homologous to ubiquitin ligases (AL-KHODAIRY et al. 1995 ). Mutants retain checkpoint functions but are radiation and HU sensitive, and they exhibit rapid death in conjunction with mutations in wee1, cdc17, and cdc22. Epistasis analyses suggest that Hus5 functions in a DNA-damage-recovery pathway that includes Rad31, and the checkpoint proteins Rad17 and Rad26 (and presumably the other core checkpoint proteins Rad, Rad3, Rad9, and Hus1, AL-KHODAIRY et al. 1994 ). In view of the similarities between Hus5 and Rhp6 functions, we transformed rhp6-788 with pREP41hus5. Mutant growth and morphological defects were not suppressed.

It follows that a possible mechanism for caffeine override of the G₂ checkpoint control is by inappropriate activation of the cell cycle promotion function of Rhp6, though it is unlikely that this is a direct effect. As we have shown (Figure 7), caffeine may sensitize cells to killing by HU (and radiation, not shown) without significantly affecting cell cycle arrest. This implies that caffeine may mask DNA damage by directly blocking repair [notably, caffeine binds single-stranded DNA (TS'O and LU 1964 ) and is a potent inhibitor of PRR (LEHMANN 1972 ; PRAKASH 1981 )]. Termination of the repair function might then free the Rhp6 protein to promote cell cycle progression and override the checkpoint control. In the absence of Rhp6 function, the protein is not available to promote cell cycle progression or to participate in repair, so neither repair (strictly, radiation sensitivity) nor checkpoint functions are affected by caffeine treatment. This proposal is consistent with the observation that caffeine treatment of UV- or HU-treated Xenopus extracts also blocks activation of the checkpoint protein Chk1 (KUMAGAI et al. 1998 ).

	ACKNOWLEDGMENTS

Thanks to Anil Ganesh for help in the use of gene data banks and the manipulation of DNA sequence data. This investigation was supported in part by U.S. Public Health Service grant number CA40245 and by a grant from the Huntsman Cancer Institute at the University of Utah.

Manuscript received August 11, 1998; Accepted for publication January 18, 1999.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ALFA, C., P. FANTES, J. HYAMS, M. MCLEOD and E. WARBRICK, 1993 Experiments with Fission Yeast: A Laboratory Course Manual. Cold Spring Harbor Press, Cold Spring Harbor, New York.

AL-KHODAIRY, F. and A. M. CARR, 1992  DNA repair mutants defining G2 checkpoint controls in S. pombe.. EMBO J. 11:1343-1350[Medline].

AL-KHODAIRY, F., T. ENOCH, I. M. HAGAN, and A. M. CARR, 1995  The Schizosaccharomyces pombe hus5 gene encodes a ubiquitin conjugating enzyme required for normal mitosis. J. Cell Sci. 108:475-486[Abstract].

AL-KHODAIRY, F., E. FOTOU, K. S. SHELDRICK, D. J. GRIFFITHS, and A. R. LEHMANN et al., 1994  Identification and characterization of new elements involved in checkpoint and feedback controls in fission yeast. Mol. Biol. Cell 5:147-160[Abstract].

ANDREASSEN, P. R. and R. L. MARGOLIS, 1992  2-aminopurine overrides multiple cell cycle checkpoints in BHK cells. Proc. Natl. Acad. Sci. USA 89:2272-2276[Abstract/Free Full Text].

BACHMAIR, A., D. FINLEY, and A. VARSHAVSKY, 1986  In vivo half-life of a protein is a function of its amino-terminal residue. Science 234:179-186[Abstract/Free Full Text].

BAILLY, V., J. LAMB, P. SUNG, S. PRAKASH, and L. PRAKASH, 1994  Specific complex formation between yeast RAD6 and RAD18 proteins: a potential mechanism for targeting RAD6 ubiquitin-conjugating activity to DNA damage sites. Genes Dev. 8:811-820[Abstract/Free Full Text].

BAILLY, V., S. LAUDER, S. PRAKASH, and L. PRAKASH, 1997  Yeast DNA repair proteins Rad6 and Rad18 form a heterodimer that has ubiquitin conjugating, DNA binding, and ATP hydrolytic activities. J. Biol. Chem. 272:23360-23365[Abstract/Free Full Text].

BARTEL, B., I. WUNNING, and A. VARSHAVSKY, 1990  The recognition component of the N-end rule pathway. EMBO J. 9:3179-3189[Medline].

BOSE, S. K., R. W. JONES, and L. J. TOLMACH, 1978  The action of caffeine on X-irradiated HeLa cells: II. Synergistic lethality. Radiat. Res. 71:666-677.

DAS, S. K., 1987  Abrogation of the effects of aphidicolin on NIH3t3 and V79 cells by caffeine. Biochim. Biophys. Acta 928:29-35[Medline].

DOHMEN, R. J., K. MADURA, B. BARTEL, and A. VARSHAVSKY, 1991  The N-end rule is mediated by the UBC2(RAD6) ubiquitin-conjugating enzyme. Proc. Natl. Acad. Sci. USA 88:7351-7355[Abstract/Free Full Text].

DUNPHY, W. G. and A. KUMAGAI, 1991  The cdc25 protein contains an intrinsic phosphatase activity. Cell 67:189-196[Medline].

ELLEDGE, S. J., 1996  Cell cycle checkpoints: preventing an identity crisis. Science 274:1664-1672[Abstract/Free Full Text].

ELLISON, K. S., T. GWOZD, J. A. PRENDERGAST, M. PATERSON, and M. J. ELLISON, 1991  A site-directed approach for constructing temperature-sensitive ubiquitin-conjugating enzymes reveals a cell cycle function and growth function for RAD6. J. Biol. Chem. 266:24116-24120[Abstract/Free Full Text].

ENOCH, T. and P. NURSE, 1990  Mutation of fission yeast cell cycle control genes abolishes dependence on mitosis on DNA replication. Cell 60:665-673[Medline].

FAN, S., M. L. SMITH, D. J. RIVET, Q. ZHAN, and K. W. KOHN et al., 1995  Disruption of p53 sensitizes breast cancer MCF-7 cells to Cisplatin and pentoxyfylline. Cancer Res. 55:1649-1654[Abstract/Free Full Text].

FEATHERSTONE, C. and P. RUSSELL, 1991  Fission yeast p107^wee1 mitotic inhibitor is a tyrosine/serine kinase. Nature 349:808-881[Medline].

FURNARI, B., N. RHIND, and P. RUSSELL, 1997  Cdc25 mitotic inducer targeted by Chk1 DNA damage checkpoint kinase. Science 277:1495-1497[Abstract/Free Full Text].

GAUTIER, J., C. NORBURY, M. LOHKA, P. NURSE, and J. MALLER, 1988  Purified maturation promoting factor contains the product of a Xenopus homolog of the fission yeast cell cycle control gene cdc2⁺. Cell 54:433-439[Medline].

GAUTIER, J., M. J. SOLOMON, R. N. BOOHER, J. F. BAZAN, and M. W. KIRSCHNER, 1991  Cdc25 is a specific tyrosine phosphatase that directly activates p34^cdc2. Cell 67:197-211[Medline].

GOULD, K. L., S. MORENO, D. J. OWEN, S. SAZER, and P. NURSE, 1991  Phosphorylation at Thr167 is required for Schizosaccharomyces pombe p34^cdc2 function. EMBO J. 10:3297-3309[Medline].

HAIN, J., N. E. CROMPTON, W. BURKART, and R. JAUSSI, 1993  Caffeine release of radiation induced S and G2 phase arrest in V79 hamster cells: increase of histone messenger RNA levels and p34cdc2 activation. Cancer Res. 53:1507-1510[Abstract/Free Full Text].

HARTWELL, L. H. and T. A. WEINERT, 1989  Checkpoints: controls that ensure the order of cell cycle events. Science 246:629-634[Abstract/Free Full Text].

JENTSCH, S., J. P. MCGRATH, and A. VARSHAVSKY, 1987  The yeast DNA repair protein RAD6 encodes a ubiquitin-conjugating enzyme. Nature 329:131-134[Medline].

KIMLER, B. F., D. B. LEEPER, M. H. SNYDER, R. ROWLEY, and M. H. SCHNEIDERMAN, 1978  Modification of radiation-induced division delay by caffeine analogs and dibutyryl cyclic AMP. Int. J. Radiat. Biol. 41:47-58.

KUMAGAI, A., Z. GUO, K. H. EMAMI, S. X. WANG, and W. G. DUNPHY, 1998  The Xenopus Chk1 protein kinase mediates a caffeine-sensitive pathway of checkpoint control in cell-free extracts. J. Cell Biol. 142:1559-1569[Abstract/Free Full Text].

LEA, D. E., 1956 Actions of Radiations on Living Cells, Ed. 2. Cambridge University Press.

LEHMANN, A. R., 1972  Postreplication repair of DNA in ultraviolet-irradiated mammalian cells. J. Mol. Biol. 66:319-337[Medline].

LI, J. C. and E. KAMINSKAS, 1987  Progressive formation of DNA lesions in cultured Ehrlich ascites tumor cells treated with hydroxyurea. Cancer Res. 47:2755-2758[Abstract/Free Full Text].

LONN, U. and S. LONN, 1988  Progressive formation of DNA lesions during treatment with antimetabolites without incorporation of the drugs into DNA. Mutat. Res. 200:243-247[Medline].

LUCKE-HUHLE, C., 1982  Alpha-irradiation-induced G2 delay: a period of cell recovery. Radiat. Res. 89:298-308[Medline].

MAEDA, T., M. TAKEKAWA, and H. SAITO, 1995  Activation of yeast PBS2 MAPKK by MAPKKKs or by binding of an SH3-containing osmosensor. Science 269:554-558[Abstract/Free Full Text].

MATSUMOTO, T., 1997  A fission yeast homolog of CDC20/p55CDC/Fizzy is required for recovery from DNA damage and genetically interacts with p34^cdc2. Mol. Cell. Biol. 17:742-750[Abstract].

MAUNDRELL, K., 1990  nmt1 of fission yeast: a highly transcribed gene completely repressed by thiamine. J. Biol. Chem. 265:10857-10864[Abstract/Free Full Text].

METZGER, K., 1964  On the dark reactivation mechanism in ultraviolet irradiated bacteria. Biochem. Biophys. Res. Commun. 15:101-109.

NAKAMURA, K. and S. ANTOKU, 1994  Enhancement of X-ray cell killing in cultured mammalian cells by the protein phosphatase inhibitor calyculin A. Cancer Res. 54:2088-2090[Abstract/Free Full Text].

NASMYTH, K. A., 1977  Temperature sensitive lethal mutants in the structural gene for DNA ligase in the yeast Schizosaccharomyces pombe.. Cell 57:775-786.

NEFSKY, B. and D. BEACH, 1996  Pub1 acts as an E6-AP-like protein ubiquitin ligase in the degradation of Cdc25. EMBO J. 15:1301-1312[Medline].

OTA, I. M. and A. VARSHAVSKY, 1993  A yeast protein similar to bacterial two-component regulators. Science 262:566-569[Abstract/Free Full Text].

PENG, C.-Y., P. R. GRAVES, R. S. THOMA, Z. WU, and A. S. SHAW et al., 1997  Mitotic and G2 checkpoint control: regulation of 14-3-3 protein binding by phosphorylation of Cdc25 on serine-216. Science 277:1501-1505[Abstract/Free Full Text].

PHIPPS, J., A. NASIM, and D. R. MILLER, 1985  Recovery, repair and mutagenesis in Schizosaccharomyces pombe.. Adv. Genet. 23:1-72[Medline].

PIWNICA-WORMS, H., S. ATHERTON-FESSLER, M. S. LEE, S. OGG, and K. I. SWENSON et al., 1991  p107^wee1 is a serine/threonine and tyrosine kinase that promotes the tyrosine phosphorylation of the cyclin/p34^cdc2 complex. Cold Spring Harbor Symp. Quant. Biol. 56:567-576[Abstract/Free Full Text].

PRAKASH, L., 1981  Characterization of postreplication repair in Saccharomyces cerevisiae and effects of rad6, rad18, rev3 and rad52 mutations. Mol. Gen. Genet. 184:471-478[Medline].

PRAKASH, L., 1994  The RAD6 gene and protein of Saccharomyces cere-visiae.. Ann. NY Acad. Sci. 726:267-273[Medline].

REYNOLDS, P., M. H. M. KOKEN, J. H. J HOEIJMAKERS, S. PRAKASH, and L. PRAKASH, 1990  The rhp6⁺ gene of Schizosaccharomyces pombe: a structural and functional homolog of the RAD6 gene from the distantly related yeast Saccharomyces cerevisiae.. EMBO J. 9:1423-1430[Medline].

ROBERGE, M., C. TUDAN, S. M. F. HUNG, K. W. HARDER, and F. R. JIRIK et al., 1994  Antitumor drug fostriecin inhibits the mitotic entry checkpoint and protein phosphatases 1 and 2A. Cancer Res. 54:6115-6121[Abstract/Free Full Text].

ROWLEY, R., 1992  Reduction of radiation-induced G2 arrest by caffeine. Radiat. Res. 129:224-227[Medline].

ROWLEY, R., S. SUBRAMANI, and P. G. YOUNG, 1992  Checkpoint controls in Schizosaccharomyces pombe.. EMBO J. 11:1335-1342[Medline].

RUDERMAN, J. V., 1993  MAP kinase and the activation of quiescent cells. Curr. Opin. Cell Biol. 5:207-213[Medline].

RUSSELL, P. and P. NURSE, 1986  cdc25⁺ functions as an inducer in the mitotic control of fission yeast. Cell 45:145-153[Medline].

RUSSELL, P. and P. NURSE, 1987  Negative regulation of mitosis by wee1⁺, a gene encoding a protein kinase homolog. Cell 49:559-567[Medline].

SANCHEZ, Y., C. WONG, R. S. THOMA, R. RICHMAN, and Z. WU et al., 1997  Conservation of the Chk1 pathway in mammals: linkage of DNA damage to Cdk regulation through Cdc25. Science 277:1497-1501[Abstract/Free Full Text].

SAWECKA, J., B. GOLOS, and J. MALEC, 1987  Modification by caffeine of acute cytotoxic response of cultured L5178Y cells to hydroxyurea treatment. Neoplasma 34:369-377[Medline].

SCHLEGEL, R. and A. B. PARDEE, 1986  Caffeine-induced uncoupling of mitosis from the completion of DNA replication in mammalian cells. Science 232:1264-1266[Abstract/Free Full Text].

SHAYEGHI, M., C. L. DOE, M. TAVASSOLI, and F. Z. WATTS, 1997  Characterisation of Schizosaccharomyces pombe rad31, a UBA-related gene required for DNA damage tolerance. Nucleic Acids Res. 15:1162-1169.

SHIOZAKI, K. and P. RUSSELL, 1995  Cell-cycle control linked to extracellular environment by MAP kinase pathway in fission yeast. Nature 378:739-743[Medline].

SMYTHE, C. and J. NEWPORT, 1992  Coupling of mitosis to the completion of S phase in Xenopus occurs via modulation of the tyrosine kinase that phosphorylates p34^cdc2. Cell 68:787-797[Medline].

SOLOMON, M. J., M. GLOTZER, T. H. LEE, M. PHILLIPPE, and M. KIRSCHNER, 1990  Cyclin activation of p34^cdc2. Cell 63:1013-1024[Medline].

SUNG, P., S. PRAKASH, and L. PRAKASH, 1990  Mutation of cysteine-88 in the Saccharomyces cerevisiae RAD6 protein abolishes its ubiquitin-conjugating activity and its various biological functions. Proc. Natl. Acad. Sci. USA 87:2695-2699[Abstract/Free Full Text].

TAM, S. W. and R. SCHLEGEL, 1992  Staurosporine overrides checkpoints for mitotic onset in BHK cells. Cell Growth Diff. 3:811-817[Abstract].

TS'O, P. O. P. and P. LU, 1964  Interaction of nucleic acids. I. Physical binding of thymine, adenine, steroids and aromatic hydrocarbons to nucleic acids. Proc. Natl. Acad. Sci. USA 51:17-24[Free Full Text].

VARSHAVSKY, A., 1996  The N-end rule: functions, mysteries, uses. Proc. Natl. Acad. Sci. USA 93:12142-12149[Abstract/Free Full Text].

WALTERS, R. A., L. R. GURLEY, and R. A. TOBEY, 1974  Effects of caffeine on radiation-induced phenomena associated with cell cycle traverse of mammalian cells. Biophys. J. 14:99-118.

WALWORTH, N. C. and R. BERNARDS, 1996  rad-dependent response of the Chk1-encoded protein kinase at the DNA damage checkpoint. Science 271:353-356[Abstract].

WALWORTH, N., S. DAVEY, and D. BEACH, 1993  The fission yeast Chk1 protein links the rad checkpoint pathway to cdc2.. Nature 363:368-371[Medline].

WANG, Q., S. FAN, A. EASTMAN, P. J. WORLAND, A. SAUSVILLE, and P. M. O'CONNOR, 1996  UCN-01: a potent abrogator of G2 checkpoint function in cancer cells with disrupted p53. J. Natl. Cancer Inst. 88:956-965[Abstract/Free Full Text].

WEINERT, T. A. and L. H. HARTWELL, 1988  The RAD9 gene controls the cell cycle response to DNA damage in Saccharomyces cerevisiae.. Science 241:317-322[Abstract/Free Full Text].