YEAST VOL. 12: 349-359 (1996)

Effects of a Novel DNA-Damaging Agent on the Budding Yeast Saccharomyces cerevisiae Cell Cycle LAURA POPOLO*, F E R D I N A N D 0 VIGANO, EUGENIO ERBAt, NICOLA MONGELLIZ A N D MAURIZIO D'INCALCI?

Dipartirnento di Fisiologia e Biocliinrica Geiiernli, Universita drgli Studi di Mihno,',Sezione di Biocliitnicn Coriiparata. Via Celoria, 26.20133 Milano. Italj~ tLaborutorio di Clieriiioterapia Antiturnorole, Istituto di Ricerche Farniacologiche Mario Nrgri. Via Eritrea, 62.20157 Milano, Itulr 1 Pliarr?iacia R. & D. -B. A, , E.xperitnenta1 Onc,ologj Deprirtnient. Via Giownrzi XXIII , 23.20014 Nerviano ( M I ) , Italy

Received 7 August 1995: accepted 4 October 1995

We have investigated the effects on Succliaronijxes cerevisirie of a novel antitumour agent (FCE245 17 or Tallimustine) which causes selective alkylations to adenines in the minor groove of DNA. Tallimustine, added to wild-type cells for short periods, reduced the growth rate and increased the percentage of budded cells and delayed the cell cycle in the late S+G,+M phases. In the rad9A null mutant cells, Tallimustine treatment did not affect growth rate and the percentage of budded cells but greatly reduced cell viability compared to isogenic cells. Consistent with a role of RAD9 in inducing a transient delay in G, phase which preserves cell viability, the potent cytotoxic effect of the drug on rad9A cells was alleviated by treatment with nocodazole. Tallimustine was also found to delay the resumption from G, arrest of wild-type but not of rad9A cells. These data indicate that the effects of Tallimustine on cell cycle progression in yeast are mediated by the RAD9 gene product. From our data it appears that yeast could be a valuable model system to study the mode of action of this alkylating drug and of minor groove alkylators in general.

KEY WORDS - Saecharomyces cermisiae; rud9 mutant; alkylating agents; cell cycle; checkpoints

INTRODUCTION Studies on the eukaryotic cell cycle have received a relevant contribution by the increasing use of the fission yeast Schizosaccharoniyces poinbe and of the budding yeast Saccharoniyces cerevisiae. It has been demonstrated that the mechanisms of cell cycle regulation are highly conserved among eu- karyotic organisms (Murray and Hunt, 1993). In the S. cerevisiae cell cycle two restriction points are present: the first is in GI, at a regulatory area called START (Reed, 1980) and the second one is in the G, phase. The transitions to S and M phases involve activation of the CDC28 encoded-protein kinase, homologue of ~34'~'?, and different GI and B-type cyclins (Nasmyth, 1993).

Moreover in yeast, as in a large variety of eukaryotic cells, checkpoints are present at which

*Corresponding author.

cells monitor the completion of some events before initiating new ones (Hartwell and Weinert, 1989; Murray, 1995). One of the best understood check- points is located in the G2 phase and directs the coordination between the completion of DNA replication and entry into mitosis (Nurse, 1994). Among the signals that are monitored at the G,-checkpoint is the presence of damaged DNA. A wild-type cell delays the cell cycle at the G2 phase until DNA is repaired, thus preserving the integrity of the genome. Several pieces of evidence suggest that some types of tumor cells have lost this control (Hartwell, 1992; Hartwell and Kastan, 1994). The control circuit operating in G2 has been the subject of genetic analysis in S. cerevisiae. The Rad9p has been the first checkpoint-associated gene product described to be required for delaying the cell cycle prior to mitosis for repair of DNA damaged by ionizing or UV-radiations (Weinert

CCC 0749-503W96/040349-11 C, 1996 by John Wiley & Sons Ltd

350 L. POPOLO ET AL.

haploid rud9A strain 7833- I a ( M A Ta rutl9:: T R P l his3 leri2-3,112 trpl-28Y rod-52) were kindly provided by Dr L. H. Hartwell (University of Washington, Seattle, Washington).

The growth medium was made up with 6.7 g/l of yeast nitrogenbase (YNB) without amino acids, 2%) glucose and with the appropriate supplements. We used as rich solid medium the YPDT (yeast extract 1% peptone 2% D-glucose 2'%, tryptophan 50 mg/l and agar 2%)). All cultures were grown at 30°C and the experiments were initiated at a cell density of about 2-4 x lo6 cells/ml. A 250-300 nil culture was divided in 50 ml subcultures and the drugs were added at the required concentrations. At the end of the treatment, cells were filtered on 1.2 pm nitrocellulose sterile filters or centrifuged. washed with prewarmed fresh medium and resus- pended in fresh medium i n the absence of the drug.

Tallimustine, previously indicated as FCE 2451 7: N-deformyl-N-(4-N-,N-bis( 2-chloroethylamino) benzoyl-distamycin A, and distamycin A were synthesized i n Pharmacia Laboratories (Nerviano. Italy). Immediately before use, the drugs were dissolved in a small volume of diinethylsulfoxide (DMSO) and then a 1 nig/ml stock solution in sterile water was prepared. To verify the actual concentration of the drugs, a 1 : 100 dilution of the stock solution in ethanol was prepared and the A,,, ,1,,1 was measured (E =0.674 for distamycin A and 0.744 for Tallimustine). Control culture were treated with a volume of DMSO equal to the volume of the stock solution added to the test culture. The percentage of DMSO never exceeded 0.5%.

For cell synchronization. u-factor (u-F) from Bacheni (Basel, Switzerland) was resuspended in 10 mM-HCI, 0.1 niM-EDTA and I mM-p- niercaptoethanol (0.3 inM stock solution) and added to the culture medium at a final concen- tration of 0.25 p ~ . Methyl-5-[2-thienylcarbonyl]- 1-H-benzimidazole-2-yl-carbamate (nocodazole) from Sigina Chemical Co. (St Louis, MO) was added to the medium of cultures at the final concentration of 15 pg/nil from a freshly prepared stock solution (3.3 nig/nil in DMSO).

Detertiiinutioti qf' cell nurher. percentugc~ qf' hudded cells arm' q f cells it'itli diflkretrt nuc-lrcir rnor plzologies

Cell iiuinber and volume distribution were de- termined as previously described (Vanoni et ul., 1983). The percentage of budded cells was deter- mined by counting at least 400 sonicated cells

and Hartwell. 1988). However, the biochemical activity and the physiological role of Rad9p are still unclear. The phenotype of rtid9A cells is similar to that of wild-type cells except that mutant cells lose chromosomes spontaneously at a higher rate and are hypersensitive to treatment with chemical or physical agents that damage DNA (Schiestl et al., 1989; Weinert and Hartwell, 1990).

I t is also known that agents that damage DNA cause a transient arrest in the GI phase. RADY and RAD24 have been found to be crucial for this response which has been defined as the G I - checkpoint (Wintersberger and Karwan, 1987: Siede et al., 1993). The involvement of RAD9 in transducing signals in G, and G3 phases suggests that it could function in a feedback circuit that ultimately inhibits the different p34 kinase-cyclins complexes. In mammalian cells, i t has been re- cently reported that y-irradiated GI cells arrest in G I phase via the p53-mediated induction of a Cdk inhibitor protein (Dulic et ul.. 1994).

The aim of the present study was to evaluate whether S. cew~isitie can be used as a model organism for the characterization of the cell cycle perturbations induced by Tallimustine, a benzoyl nitrogen mustard derivative of distamycin A (Arcamone et a/., 1989). Distamycin A. an anti- biotic originally isolated from Streptonzjws rlistol- licrrs, is a well-known minor-groove binder that has antiviral activity but is devoid of antiprolifera- t h e effects in animal cells (Kopka et ul., 1985; Broggini et ul., 199 1). In contrast, Tallimustine selectively alkylates very few adenines in a highly sequence-specific manner in the ininor groove of DNA (Broggini et d . . 1991, 1995) and it has been shown to have striking antiproliferative activity against both murine and human tumor cell lines (Broggini el ( / I . , 1991; Erba et ul., 1995) even against tumors resistant to other alkylating agents (Pezzoni et (11.. 1991). We report here that Tal- liniustine is also active in yeast and that the antiproliferative action of this agent is dependent upon functional GI and G,-checkpoint controls.

MATERIALS AND METHODS

Yctist strains arid groli,tli cotzditions Succlrurotii~ces cereisisirre haploid strain S288c

( M A T u SUC.2 niul rnel gal2 C U P l ) was obtained froni the yeast Genetic Stock Center (Berkeley, CA, U.S.A.). The haploid Rad+ strain 7830-2-4a ( M A T a his3 leu2-3,112 trpl-289 uru3-52) and the

NOVEL DNA-DAMAGING AGENT 35 1

fixed in 4% formalin. Nuclear morphology was determined by 4',6'-diamino-2-phenylindole.HCl (DAPI) staining. Cells fixed with ethanol 70'% were washed in PBS-EDTA 0.2 mM, stained for 10 min in DAPI 0.125 pg/ml in PBS-EDTA. Cells were washed twice in PBS-EDTA and analysed under a Leitz, Dialux 22 fluorescence microscope at x 1250.

Cell sur vivd ussuys

For a qualititative cell viability assay, 10'. lo3, lo2 and 10 cells were plated on solid rich medium and incubated for 48 h at 30°C.

A quantitative assay was performed as follows: aliquots of the cultures were withdrawn, cells were sonicated under sterile conditions and counted. Cells were then diluted and plated in duplicate to have 400 celldplate (single colony plates) or 800- 4000 celldplate for the cultures treated with highest concentrations of the drug. After 48 h of incubation at 3O"C, the colonies present in each plate were counted and the percentage of survival was determined with respect to that of the control plates (100% of viability).

DNA analysis DNA content per cell was analysed by flow

cytometry by staining with propidium iodide as previously described (Vanoni et u/., 1983). Cells were analysed with a FACStar PLUS (Becton Dickinson & Co.). The percentage of cells in the different cell cycle phases was evaluated using the method of Krishan and Frei (1976).

RESULTS

Eflects of short-term exposures of S. cerevisiae cells to Tullirnustiiie

According to the protocol used for mammalian cell lines (Broggini et ul., 1991), the effect of Tallimustine on yeast was tested with short-term treatments. When exponentially growing yeast cells were treated for 1 h with 10 to 1 0 0 ~ ~ - Tallimustine, the growth rate was rapidly reduced and the decrease was concentration-dependent (Figure 1A). The percentage of budded cells in- creased to a value of about 85-90% at 3.5 h from filtration and this value was maintained for several hours (Figure IB). This suggests that an apparent new steady-state of growth was established, characterized by a low growth-rate and a high

percentage of budded cells. N o effects were present at the concentration of 1 PM (data not shown).

The analysis of cell volume distribution shown in Figure 1C reveals a striking increase in the volume of treated cells. This suggests that the drug does not inhibit cell mass increase. Cell viability does not appear to be affected from 10 to 1 0 0 ~ ~ (Figure IG).

DNA analysis indicated that 30 inin after re- moval of the drug no effects were present in cells treated with 10 pM-Talliniustine (Figure 1D). At 3 h cells treated with both concentrations tended to enrich the S phase compartment (Figure 1 E). At 5 h cells treated with 10 PM accumulate for the most part of the late S + G , + M whereas cells treated with 1 0 0 ~ ~ appear to enrich S phase (Figure 1F). No net accumulation of cells in a unique cell cycle compartment is revealed since the situation of these cells is a dynamic one.

10 pM-Tallimustine was found to be effective in yeast even if the exposure time was reduced to 10 min. DNA analysis again revealed an en- richment of cells in the late S + G 2 + M phases (data not shown). Distamycin A was tested dur- ing 10-min and 1-h treatment at concentrations of 10 and 1 0 0 p ~ and it did not perturb the growth. the percentage of budded cells or the DNA profiles, suggesting that also in yeast this molecule has no antiproliferative effects (data not shown).

Microscopic urzulysis of Tullir?zustine-trrcited cells In cultures treated with different concentrations

of Tallimustine (10, 50 and 100 p ~ ) , cells with very large buds, often very elongated, or cells with a bud that emerges from the side of the cell appeared (Figure 2a, c, e and g). In the right panels (b-h) the nuclei of the same cells were observed. Table 1 gives the percentage of cells with the nucleus close to the neck between mother and daughter (class I ) , cells with the nucleus migrated into the neck (class 2) and cells with two nuclei (class 3). In unper- turbed conditions, these cells roughly correspond to cells in the late S+G,, in M or in the GI* phase. Tallimustine increased the number of cells of class 1 and class 2, whereas the fraction of class 3 did not change. This suggests that Tallimustine affects the progression through S + G2 + M phases but does not interfere with the process of cytodieresis. Consistent with this. a high proportion of cells with elongated mitotic spindles were detected (data not shown).

352 L. POPOLO ET A L .

Channel number

G Tallimustine control 1 10 100 pM

a

b

C

353 NOVEL DNA-DAMAGING AGENT

a b Table 1. Distribution of cells with different nuclear morphologies. Nuclear morphologies of S288c S. cerevi- siae cells at 5 h after 1 h of treatment with different concentrations of Tallimustine.

d Classes*

f Untreated cells 16 10 11 Tallimustine 10 p~ 31 16 13 Tallimustine 50 J.LM nd 20 15 Tallimustine 100 p~ 21 22 12

g h *These values were determined by counting at least 500 cells by fluorescence microscopy. nd, not determined.

Figure 2. Microscopic analysis of S288c S. cerevisiae cells at 5 h after 1 h of treatment with Tallimustine. Left panel (a-g) shows the analysis of cell morphology; right panel ( b h ) shows nuclear morphology of the same cells observed by fluorescence microscopy. Untreated cells (a and b), 10 p~ (c and d), 50 PM (e and f ) and 100 p~ (g and h) Tallimustine-treated cells.

Efects of a short-term exposure of S. cerevisiae rad9A cells to Tallimustine

The effects of Tallimustine were monitored in a haploid mutant in which the RAD9 gene was disrupted and in its isogenic counterpart (Radf). Cells were treated for 10 min during exponential growth phase with 10 pM-Tallimustine. While the isogenic cells underwent growth arrest and the percentage of budded cells increased (Figure 3A and C), rad9A cells continued to grow for about 5 h after drug removal, without significant changes in the percentage of budded cells (Figure 3B and D). This result suggests that Tallimustine does not significantly affect the rate of division or budding in rad9A cells. In panels E and F the increase of cell size in treated cells is shown. The increase of cell mass in control cells is about 50% more than in rad9A cells.

Microscopic analysis indicated that rad9A cells treated with Tallimustine are less elongated and

more frequently show abnormal morphologies and dead cells. A quantitative analysis of cells with different nuclear morphologies has indicated that the percentage of cells with the nucleus migrated close to the neck between mother and daughter cells and of cells with an elongated nucleus in Rad+ cells treated with Tallimustine increases with respect to the untreated ones (from 26 to 44%). In rad9A treated cells the percentage of cells with these nuclear morphologies is roughly equal to the untreated cells (29%).

Since these data indicate that rad9A cells are not retarded in cell cycle progression, we reasoned that the premature entry into mitosis could lead to a decrease in cell viability. As shown in Figure 4, rad9A Tallimustine-treated cells were much less viable than isogenic cells, which appear as viable as the untreated cells.

Control experiments have verified that distamy- cin A was not toxic to rad9A cells (data not shown) .

Potent cytotoxic efect of Tallimustine on rad9A cells

Assays for cell survival are a useful tool to evaluate the cytotoxic effects of a drug on a cer- tain cell line. We therefore tested the effects of

Figure I . Effects of 1 h of treatment with Tallimustine on S288c S. cerevisiae cells. Growth curve (A), percentage of budded cells (B), cell volume distribution (C), DNA analysis (D-F) and cellular viability (G). The arrow indicates the moment of drug addition ( - 1 h). At zero time, cells were filtered and resuspended in fresh medium in the absence of the drug. (A and B) Untreated (O), 10 p~ (U) and 100 p~ (0) Tallimustine-treated cells. (C) Untreated (unbroken line), 10 p~ (dashed line) and 100 ~ L M (dotted line) Tallimustine-treated cells, at 5 h after drug washout. (D-F) Cellular DNA content distribution of untreated (a), 10 p~ (b) and 1 0 0 ~ ~ (c) Tallimustine-treated cells, at 30 min (D), 3 h (E) and 5 h (F) after drug washout. (G) Cellular viability assay of 1, 10, 100 pM-Tallimustine-treated cells, at 5 h after drug washout. Several dilutions of cells were spotted (5 p1) on YPTD plate (a: lo2; b: lo3; c: lo4 cells/spot).

3 54

- E

L. POPOLO ET AL.

; 3 - 1) /, lo-

5 Z

1 1

1"

(D

E

/ I I I I 1 I i

6 A '0°1

100-

75 -

50-

C - I /

251 0 1 - 2 - 1 0 1 2 3 4 5

Time (h)

'"1 D

251 0 1 - 2 - 1 0 1 2 3 4 5

Time (h)

Channel number Channel number Figure 3. Effects of 10 min of treatment with 10pM-Tallimustine on Rad' (7830-2-4a) and rad9A (7833-la) S. cerevisiue cells. Growth curve (A, B), percentage of budded cells (C, D) and cell volume distribution (E, F). The arrow indicates the moment of drug addition ( - 10 min). At zero time, cells were filtered and resuspended in fresh medium in the absence of the drug. (A and C) Untreated (0) and 10 ~ L M (I) Tallimustine-treated Rad' cells. (B and D) Untreated (0) and 10 p~ (I) Tallimustine-treated rad9A cells. (E) Untreated (unbroken line) and 10 p~ (broken line) Tallimustine-treated Rad+ cells. (F) Untreated (unbroken line) and 10 p~ (broken line) Tallimustine-treated rud9A cells.

Tallimustine on cell viability in rad9A cells. Expo- DMSO and 5 h later a quantitative assay of cell nentially growing cells were treated for 10 min with viability was performed. Rad' cells were found to different concentrations of the drug or with be at least two orders of magnitude more resistant

NOVEL DNA-DAMAGING AGENT 355

Rad' rad9A Tallimustine - + +

1

2

3

4

Figure 4. Cell viability assay of Rad' (7830-2-4a) and radYA (7833-la) S. cerevisiae cells at 5 h after 10 rnin of treatment with Tallimustine 10 p ~ . Several dilutions of cells were spotted (5 pl) on YPDT plates (1: lo4; 2: 10'; 3: lo2; 4: 10 cellskpot).

Table 2. Cytotoxicity of Tallimustine on rad9A cells. Percentage of viable cells determined at 5 h after 10 min of exposure to various concentrations of Tallimustine.

Rad' cells rad9A cells

DMSO 100 100 Tallimustine ( p ~ )

0.1 nd 92 1 98 63 5 88 13

10 nd 4 50 55 nd

100 31 0.02

For details on this assay, see Materials and Methods nd, not determined.

to this drug than mutant cells (Table 2). This strongly supports the idea that the lack of the execution of the G,-checkpoint determines the high toxicity of this drug in rad9A cells.

Nocoduzole abolishes cytotoxicity qf Tallimustine on rad9A cells

In order to determine if the sensitivity to Tal- limustine could be due to the function of RAD9 in delaying cell cycle progression rather than activat- ing a repair process, we artificially elongated the G, phase by treating with nocodazole, an inhibitor of mitosis, and tested the effects of the drug. As shown in Figure 5 , nocodazole partially alleviates the cytotoxic effects of Tallimustine in rad9A cells. This result shows that RAD9 function is required to delay cell cycle progression after the treatment

Tallimustine ----- lOpM 100pM Nocodazole - + - + - +

1

2

3

4

Figure 5. Cell viability assay of radYA (7833-la) S. terevisiue cells treated with Tallimustine in the presence of nocodazole. Exponentially growing radYA cells were treated for 10 rnin with Tallimustine 10 ~ L M and 100 PM and, after filtration, resus- pended in fresh medium containing 15 pg/ml of nocodazole. At 5 h after Tallimustine removal, several dilutions of cells were spotted (Spl) on YPTD plates (1: lo4; 2: lo3; 3: lo2; 4: 10 cellskpot).

with Tallimustine, thus preventing loss in cell viability.

The delay in resumption from a-factor synchronization o j Tullimustine-treated cells is dependent on the RAD9 gene

We have studied the effects of exposure to Tallimustine on rad9A cells after synchronization with a-F at START. Cells were treated with 50 PM- Tallimustine during the last 15 min of incubation with a-F. After filtration, cells were resuspended in fresh medium. The kinetics of bud emergence revealed a delay in isogenic cells treated with the drug with respect to the untreated ones (Figure 6A). In contrast, the percentage of budded cells did not appear to be affected by Tallimustine treat- ment in rad9A cells (Figure 6B). The kinetics of growth are shown in Figure 6C and D.

In Figure 7, DNA analysis of isogenic and rud9A cells treated with Tallimustine is shown. The exit from G, blockade of the isogenic cells treated with Tallimustine was slower than that of the untreated ones. In contrast, no differences were present between untreated and treated rud9A cells.

DISCUSSION

We have investigated the effects of the distamycin A-derivative Tallimustine on progression of yeast cells through the cell cycle. In each experiment the effects of the drug have been evaluated using multiple criteria (growth kinetics, budding, cell

356

100

- v) 75 8 - U al 73 2 50 m bs

25

0

100

W z

!! 10 8

&

X - E . -

c 0

I3

Z E,

1

A loo

75

50

25

0 -4 -2 0 2 4

100

10

L. POPOLO ET AL.

B

D

/ I I I I I 1-

-4 -2 0 2 4 -4 -2 0 2 4

Time (h) Time (h)

Figure 6 . Effects of 15 min of treatment with 50 pM-Tallimustine on Rad' (7830-2-4a) and md9A (7833-la) S. cerevisiue cells after release from a-factor (a-F) arrest. Percentage of budded cells (A, B), growth curve (C, D). Exponentially growing cells were G,-synchronized by adding to the cultures 0.25 p~ of a-F at - 2 h. When cells appeared synchronized under microscopic observa- tion, they were treated with 50 pM-Tallimustine (arrow: - 15 min). Finally, at zero time, cells were filtered and resuspended in fresh medium in the absence of a-F and drug. (A and C) Untreated (U) and 50 p~ ( W ) Tallimustine-treated Rad' cells. (B and D) Untreated (0) and 50 p~ ( W ) Tallimustine-treated rad9A cells.

volume, DNA analysis, cell and nuclei morphol- ogy, and cell viability). Tallimustine, during short-term treatment of yeast cells, induces a dose-dependent delay in cell cycle progression. The drug establishes an apparent new condition of steady-state growth which is characterized by a longer T, and a higher percentage of budded cells. The increase in the mean cell volume indicates that Tallimustine does not inhibit the accumulation of RNA and proteins. Progression through S + G, + M phases is significantly delayed by Tal- limustine, in agreement with the effects on in vitro murine and human cells (Broggini et al., 1991; Erba et al., 1995). We also analysed the cellular

and nuclear morphology of treated yeast cells, and a dose-dependent increase of cells with elongated nuclei in the neck between mother and daughter cells was observed. This aspect is typical of cells treated with chemical and physical agents that damage DNA, for example the alkylating com- pounds MMS, EMS, MNNG (Wintersberger and Karwan, 1987; Kupiec and Simchen, 1985), ana- logues of bases or nucleotides, such as araCMP (McIntosh et al., 1986), X-rays (Weinert and Hartwell, 1988) or y-radiation (Brunborg and Williamson, 1978). Similar effects are seen in cells failing to complete S-phase due to cdc mutations, for example cdc9, 8, 16, 17, 20, 21, 23, 30 (Culotti

NOVEL DNA-DAMAGING AGENT

A

351

B

58 188 158 1( T 0 min

DNA CONTENT DNA CONTENT

Figure 7. DNA analysis of Rad' (7830-2-4a) and rad9A (7833-la) Tallimustine- treated cells releasing from a-factor (a-F) arrest. S. cerevisiae cells were G,- synchronized by 2 h of treatment with a-F. During the last 15 min of a-F treatment, cells were exposed to 50 pM-Tallimustine. After filtration (zero time), cells were resuspended in fresh medium in the absence of a-F and Tallimustine and, at the indicated times, aliquots of the cultures were withdrawn to be analysed by flow cytometry. Column A: cellular DNA content distribution of untreated (left) and 50 ~ L M (right) Tallimustine-treated Rad+ cells. Column B: cellular DNA content distribution of untreated (left) and 50 p~ (right) Tallimustine-treated rad9A cells.

358 L. POPOLO ET AL.

and Hartwell, 1971; Hartwell et al., 1973). Whereas in unperturbed conditions elongated nuclei are diagnostic of M phase, in cases of perturbations of S-phase it is known that their appearance is not strictly related to the presence of mitosis. In fact, nuclei migration and elongation and DNA replication are on parallel pathways in S. cerevisiae (Pringle and Hartwell, 1981).

In the wild-type S288c strain used in this study, the drug is not cytotoxic to the cells. Thus only a reversible cytostatic effect was observed. The surveillance mechanism that operates in G,, as first hypothesized by Tobey in mammalian cells (Tobey, 1975) and known as the G,-checkpoint, allows the cells to return to cycle only if DNA damage has been repaired and this is important not only for the maintenance of the fidelity of chromosome transmission but also for the preser- vation of cell viability (Hartwell, 1992). In the radYA cells the drug is highly cytotoxic (Table 2). The finding that nocodazole, which causes a tran- sient mitotic block, reduces Tallimustine cyto- toxicity in vadYA cells, as previously observed with y-radiation, further corroborates the view that the delay in the cell cycle has a protective significance for the cell.

It is notable that the RAD9-mediated G,-phase checkpoint mechanism seems to be activated by different kinds of DNA lesions. To our knowledge agents of the class of minor groove alkylators have never been tested in yeast. Although Tallimustine is the most sequence-specific alkylating agent (of small size) so far identified as it alkylates only the N3 of adenines located in the sequence TTTTGA or TTTTAA and the total number of alkylations in cells is limited, we cannot exclude that being yeast DNA AT-rich a higher number of adducts could be induced (Broggini et al., 1995).

We have also shown that Tallimustine delays entry into S-phase of Rad’ cells but not of rad9A cells synchronized in G, by pheromone treatment. This suggests that a RAD9-dependent regulatory mechanism operates to prevent cells starting DNA replication in the presence of DNA damage caused by Tallimustine. A recent study has indicated the existence of (at least) two RAD9-dependent check- points for damaged DNA prior to S-phase: one at START, at the a-F-sensitive step, and one between CDC4- and CDC7-mediated steps (termed the GUS checkpoint; Siede et al., 1994).

It should be noted that a mammalian homo- logue of the yeast RADY gene has not yet been identified. It is therefore impossible to investigate

whether there are tumors which have mutations in this gene. Recently, however, a cancer cell line hypersensitive to nitrogen mustard has been described and the peculiar sensitivity appears re- lated to an inefficient G,-checkpoint mechanism following DNA damage (O’Connor et al., 1992). Finally, this study supports the view that yeast could be a potentially suitable model system to further explore the mechanism of action of minor groove alkylators and of anticancer drugs causing DNA damage (Belenguer et al., 1995). The avail- ability of several mutants in DNA repair in yeast could be useful to study the mechanisms of repair from the damage caused by this drug.

ACKNOWLEDGEMENTS

This work was partially supported by the generous contribution of the Italian Association for Cancer Research (AIRC), by CNR (National Research Council) Progetto Finalizzato ACRO n.9302352 and 9302357 PS39 to M.D. and by MPI 60% to Lilia Alberghina.

REFERENCES Arcamone, F. M., Animati, F., Barbieri, B., et al. (1989).

Synthesis, DNA-binding properties and antitumour activity of novel distamycin derivatives. J. Med. Chem. 32, 774-778.

Belenguer, P., Oustrin, M. L., Tiraby, G. and Ducommun, B. (1995). Effects of phleomycin-induced DNA damage on the fission yeast Schizosaccharo- myces pombe cell cycle. Yeast 11, 225-231.

Broggini, M., Erba, E., Ponti, M., et al. (1991). Selective DNA interaction of the novel distamycin derivative FCE 24517. Cancer Res. 51, 199-204.

Broggini, M., Coley, H. M., Mongelli, N., et al. (1995). DNA sequence-specific adenine alkylation by the novel antitumour drug tallimustine (FCE 24517), a benzoyl nitrogen mustard derivative of distamycin. Nucl. Acids Rex 23, 81-87.

Brunborg, G. and Williamson, D. H. (1978). The rel- evance of the nuclear division cycle to radiosensitivity in yeast. Mol. Gen. Genet. 162, 277-286.

Culotti, J. and Hartwell, L. H. (1971). Genes controlling nuclear division. Exp. Cell Res. 67, 389-397.

Dulic, V., Kaufmann, W. K., Wilson, J. W., et al. (1 994). p53-dependent inhibition of cyclin-dependent kinase activities in human fibroblasts during radiation induced G , arrest. Cell 76, 1013-1023.

Erba, E., Mascellani, E., Pifferi, A. and D’Incalci, M. (1995). Cell cycle phase perturbations induced by the DNA minor groove alkylator Tallimustine compared to Melphalan on the SW626 cell line. Int. J. Cancer, in press.

NOVEL DNA-DAMAGING AGENT

Hartwell, L. H., Culotti, J., Mortimer, R. K. and Culotti, M. (1973). Genetic analysis of cdc mutants. Genetics 74, 265-275.

Hartwell, L. H. and Weinert, T. A. (1989). Checkpoints: controls that ensure the order of cell cycle events. Science 246, 629-634.

Hartwell, L. H. (1992). Defects in a cell cycle checkpoint may be responsible for the genomic instability of cancer cells. Cell 71, 543-546.

Hartwell, L. H. and Kastan, M. B. (1994). Cell cycle control and cancer. Science 266, 1821-1828.

Kopka, M. L., Yoon, C., Goodsell, D., Pjura, P. and Dickerson, R. E. (1985). The molecular origin of DNA-drug specificity in netropsin and distamycin. Proc. Natl. Acad. Sci. USA 82, 1376-1380.

Krishan, A. and Frei, E. (1976). Effect of adriamicin on the cell cycle traverse and kinetics of cultured human lymphoblasts. Cancer Res 36, 163-174.

Kupiec, M. and Simchen, G. (1985). Arrest of the mitotic cell cycle and of meiosis in Saccharomyces cerevisiae by MMS. Mol. Gen. Genet. 201, 558-565.

McIntosh, E. M., Kunz, B. A. and Haynes, R. H. (1986). Inhibition of DNA replication in Saccharo- myces cerevisiae by araCMP. Curr. Gen. 10, 79-585.

Murray, A. W. and Hunt, T. (1993). The Cell Cycle. Oxford University Press, New York.

Murray, A. (1995). The genetics of cell cycle checkpoint. Curr. Opin. in Genet. and Develop. 5, 5-1 1.

Nasmyth, K. (1993). Control of the yeast cell cycle by the Cdc28 protein kinase. Curr. Opin. Cell Biol. 5, 166179.

Nurse, P. (1994). Ordering S phase and M phase in the cell cycle. Cell 79, 547-550.

O’Connor, P. M., Ferris, D. K., White, G. A,, et al. (1992). Relationships between cdc2 kinase, DNA cross-linking, and cell cycle perturbations induced by nitrogen mustard. Cell Growth & Differentiation 43, 43-52.

Pezzoni, G., Grandi, M., Biasoli, G., et al. (1991). Biological profile of FCE 24517, a novel benzoyl mustard analogue of distamycin A. Br. J. Cancer 64, 1047-1 050.

359

Pringle, J. R. and Hartwell, L. H. (1981). The Saccha- romyces cerevisiae cell cycle. In Strathern, J., Jones, E. and Broach, J. (Eds), Molecular Biology of the Sac- charomyces cerevisiae. Cold Spring Harbor Press, Cold Spring Harbor, NY. pp. 1/97-1/142.

Reed, S. I. (1980). The selection of S. cerevisiae mutants defective in the start event of the cell division. Genetics 95, 561-577.

Schiestl, R. H., Reynolds, P., Prakash, S. and Prakash, L. (1989). Cloning and sequence analysis of the Saccharomyces cerevisiae RAD9 gene and further evidence that its product is required for cell cycle arrest induced by DNA damage. Mol. Cell Biol. 9, 1882-1 896.

Siede, W., Friedberg, A. S. and Friedberg, E. C. (1993). RAD9-dependent GI arrest defines a second check- point for damaged DNA in the cell cycle of Saccha- romyces cerevisiae. Proc. Natl. Acad. Sci. USA 90,

Siede, W., Friedberg, A. S., Dianova, I. and Friedberg, E. C. (1994). Characterization of G, checkpoint con- trol in the yeast Saccharomyces cerevisiae following exposure to DNA-damaging agents. Genetics 138, 271-28 1.

Tobey, A. R. (1975). Different drugs arrest cells at a number of distinct stages in G,. Nature 254, 245-247.

Vanoni, M., Vai, M., Popolo, L. and Alberghina, L. (1983). Structural heterogeneity in populations of the budding yeast Saccharomyces cerevisiae. J. Bacteriol. 156, 1282-1291.

Weinert, T. A. and Hartwell, L. H. (1988). The RAD9 gene controls the cell cycle response to DNA damage in Saccharomyces cerevisiae. Science 241, 3 17-322.

Weinert, T. A. and Hartwell, L. H. (1990). Characteriz- ation of RAD9 of Saccharomyces cerevisiae and evidence that its function acts posttranslationally in cell cycle arrest after DNA damage. Mol. Cell Biol. 10, 6554-6564.

Wintersberger, U. and Karwan, A. (1987). Retardation of cell cycle progression in yeast cells recovering from DNA damage: a study at the single cell level. Mol. Gen. Genet. 207, 320-327.

7985-7989.