Inhibition of Cyclin Dl Expression and Phosphorylation of Retinoblastoma Protein by Phosmidosine, a Nucleotide Antibiotic1

[CANCER RESEARCH 58. 704-710. February 15. I998| Inhibition of Cyclin Dl Expression and Phosphorylation of Retinoblastoma Protein by Phosmidosine, a Nucleotide Antibiotic1

Hideaki Kakeya, Rie Onose, Phillip C.-C. Liu,2 Chizuko Onozawa, FuiniÂ«Matsumura, and Hiroyuki Osada3

Antibiotics Laboratory, The Insiliate of Physical ami Chemical Research (RIKEN), Saitama 35I-OI9N, Japan Â¡H.K., R. O., C. O.. H. O.Â¡,and Department of Environmental Toxicology, and Institute of Toxicology and Environmental Health. University of California. Davis. California 95695 {P. C.-C. L., F. M.J

ABSTRACT implicated in the phosphorylation of pRB. Ectopie expression of cyclin Dl induces phosphorylation of pRB and accelerates progres In this report, we studied the effect of phosmidosine, a proline-contain- sion through G, (14, 15). In addition, microinjection of either anti- ing nucleotide on the serum-induced cell cycle progression in human lung fihroblast YVI-38 cells. Phosmidosine suppressed S-phase entry and ar cyclin Dl antibodies or antisense plasmid prevents cells from entering rested cell cycle progression at the I., phase. In serum-stimulated cells, S phase (14, 16, 17). phosmidosine did not affect the activation of the mitogen-activated protein Thus, understanding the underlying biochemical processes on the kinase cascade. However, phosmidosine inhibited hyperphosphorylation pRB functions will advance our knowledge of cellular proliferation in of retinohlastoma (RB) protein by RB-kinases such as cyclin-dependent general and also will have implications for efficient cancer therapies. kinase 4 and cyclin-dependent kinase 2, probably as a result of the Compounds that arrest cells at a specific point in the cell cycle have inhibition of cyclin Dl expression. Furthermore, in tsFT210 cells, a tem proven to be extremely useful tools for unraveling cell cycle regula perature-sensitive <â€¢restriction point in <;, and is a useful drug that may contribute to the we show that phosmidosine treatment of a mouse carcinoma cell line understanding of the regulatory mechanisms of G, progression. as well as a human normal cell line arrested cell cycle progression at G,. Concomitantly, phosphorylation of pRB and expression of cyclin Dl are blocked. These effects are irreversible because cells from INTRODUCTION which phosmidosine has been washed out remain arrested at G,. The RB4 gene is an antioncogene, the inactivation of which is often Because phosphorylation of pRB is essential for the G,-to-S transi tion, our experiments suggest that phosmidosine blocks cell growth by observed in various kinds of tumors including RBs and osteosarcomas ( 1). The RB gene product (pRB) is a nuclear protein of Mr 110,000- inhibiting hyperphosphorylation of pRB through the suppression of the cyclin Dl expression. 115.(XX)that forms complexes with the transforming protein encoded by DNA tumor virus, such as the adenovirus EIA protein, the SV40 (SV40) large T antigen, or the papilloma virus E7 protein. Recent MATERIALS AND METHODS genetic studies have revealed that complex formation between pRB and oncoproteins is closely linked with viral transforming activities Materials. Phosmidosine was isolated from Streptomyces durhameusis as (2-4). pRB also interacts specifically with several cellular proteins, described previously (18, 20). Human lung fibroblast WI-38 cells were ob including the transcription factor E2F and the proto-oncogene product tained from the RIKEN Cell Bank (Tsukuba, Japan). tsFT210 cells, a temper ature-sensitive cdc2 mutant isolated from the mouse mammary carcinoma cell Myc (5. 6). It is well known that pRB is phosphorylated in a cell cycle-dependent manner: hypophosphorylated in the G,,-G, phase and line FM3A, were a kind gift from Dr. F. Hanaoka (RIKEN; Ref. 21). MBP and hyperphosphorylated in the G,-to-S transition. Phosphorylation of histone HI were purchased from Sigma Chemical Co. (St. Louis, MO); and anti-cyclin Dl, anti-CDK2, and anti-CDK4 antibodies were purchased from pRB results in the dissociation of pRB and transcription factor, which Upstate Biotechnology (Lake Placid, NY). Anti-pRB antibody (PMG-245) was induces the S-phase entry (5, 7-9). obtained from PharMingen (San Diego, CA), and [7-32P]ATP was from 1CN Concomitant with these findings, pRB was shown to be phospho Biochemicals, Inc. (Costa Mesa, CA). rylated by CDKs in vitro and in vivo (10-13). CDK2/cyclin E com Cell Synchronization. WI-38 cells were grown at 37Â°Cin 5% CO2 in plex as well as CDK4/cyclin D phosphorylate pRB. Recent analysis of DMEM containing 10% FCS. Exponentially growing WI-38 cells were ar phosphorylation sites both in vitro and in vivo indicated that pRB is rested at G0 by incubation in DMEM medium supplemented with 0.2% PCS phosphorylated on more than 10 serine or threonine residues but is not for 3 days. Then these quiescent cells were stimulated by the addition of 10% phosphorylated on tyrosine residue. Cyclins of the D class (Dl, D2, FCS. tsFT2IO cells were maintained in RPMI 1640 with 10% PCS at the per and D3) complexed with CDK.4 or CDK6 are most prominently missive temperature of 32Â°Cand cultured at the nonpermissive temperature of 39Â°Cfor 17 h to synchronize them at the G2-M boundary. They were incubated Received 9/3/97; accepted 12/16/97. at 32Â°Cfor 4 h to induce their entry into G, (22). The costs of publication of this article were defrayed in part by the payment of page Flow Cytometric Analysis. Flow cytometric analysis was performed es charges. This article must therefore he hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. sentially as described by Noguchi and Browne (23). The harvested cells were 1This work was supported in part by a Special Grant for Promotion of Research stained with solution containing 50 fig/ml propidium iodide. 0.1% sodium (RIKEN). a Grant for Multibioprobes (RIKEN), and a Gram from the Ministry of citrate, and 0.2% NP40 and analyzed for DNA contents using a flow cytometer Education. Science. Sports and Culture. Japan. - Present address: Department of Biological Chemistry, University of Michigan Med (Profile II: Coulter Co., Hialeah, FL). [3H]Thymidine Incorporation. WI-38 cells were plated in 48-well plates ical School. Ann Arbor. Ml 48109-0606. ' To whom requests for reprints should be addressed, at The Institute of Physical and at 3 X IO4 cells/200 /il/well and cultured for 3 days. Cells were washed twice Chemical Research. RIKEN. Hirosawa 2-1. Wako-shi, Saitama .151-01, Japan. Phone: with serum-depleted DMEM and were incubated in DMEM containing 0.2% 81-48-467-9541; Fax: 81-48-462-4669; E-mail: [email protected]. 4 The abbreviations used are: RB. retinoblastoma; pRB. RB protein; CDK. cyclin- PCS at 37Â°Cfor 3 days. These quiescent cells were pretreated with various dependent kinase; MAP kinase, mitogen-activated protein kinase: MBP. myelin basic concentrations of phosmidosine for 15 min and stimulated with 10% FCS. protein: BrdUrd. 5-bromo-2'-deoxyuridine; CAK. CDK-activating kinase. After 18 h incubation in media with 10% FCS. the cells were labeled for 2 h 704

NH2 RESULTS Phosmidosine Arrests WI-38 Cell Cycle Progression at G,. The cell cycle progression of human lung fibroblast WI-38 cells was monitored by flow cytometry. Exponentially growing WI-38 cells had two peaks: 2 C DNA (G0-G, phase) and 4 C DNA (G2-M phase). A saddle peak between 2 C and 4 C peak was S phase, which was confirmed by the incorporation of BrdUrd, a thymidine analogue, into replicating DNA (Fig. 2, A-a and B-a). Phosmidosine (10 /MM)appar ently decreased the saddle peak area between 2 C and 4 C and the incorporation of BrdUrd (Fig. 2, A-b and B-b). After washing out the phosmidosine and allowing the cells to recover for 24 h, the cell cycle profile did not change appreciably (Fig. 2, A-c and B-c) from cells that were not allowed to recover (Fig. 2, A-b and B-b). These data indicate that phosmidosine irreversibly blocks the progression of WI-38 cells in G,. This arrest was not due to toxicity of the drug, as determined Fig. l. Structure of phosmidosine. by trypan blue exclusion (data not shown). Because phosmidosine appeared to arrest cycling cells at G,, we examined its effect on S-phase entry by quiescent WI-38 cells. with 2 fiCi/ml ['HJthymidine (25 mCi/ml; Amersham International, Bucking [3H]Thymidine incorporation was detected 16 h after serum addi hamshire, England), washed twice with 5% trichloroacetic acid, and lysed with tion, peaked after 23 h, and then declined (data not shown). We l N NaOH. The radioactive content of the resulting trichloroacetic acid- pretreated quiescent WI-38 cells with several concentrations of insoluble fractions was measured with a liquid scintillation counter (TriCurve; phosmidosine for 15 min, stimulated the cells with serum, and Packard, Meriden, CT). added [3H]thymidine 18 h later. As shown in Fig. 1C, phosmido Incorporation Assay of BrdUrd into Replicating DNA. Incorporation sine inhibited the incorporation of [3H]thymidine in a dose-depen assay of BrdUrd was carried out by using a cell proliferation kit (Amersham). Briefly, BrdUrd, a thymidine analogue, was incorporated into replicating DNA dent manner, with 10 /AMphosmidosine completely suppressing and incubated with a specific monoclonal antibody. Detection of the bounding entry of the cells into S phase. antibody was performed using a peroxidase-conjugated antibody to mouse Effect of Phosmidosine on MAP Kinase (ERK1/ERK2) Activa immunoglobulin, polymerizing diaminobenzidine, in the presence of cobalt tion. Because it is well known thatMAP kinase (ERK1/ERK2)plays and nickel, giving blue-black staining at sites of BrdUrd incorporation. a key role upstream of S-phase entry through the phosphorylation In-Gel Kiiiast Assay for the Detection of MAP Kinase (ERK1/ERK2) cascade activated after serum stimulation in quiescent cells, we in Activity. In-gel kinase assay was performed as described previously (24). vestigated the effect of phosmidosine on MAP kinase activation after Briefly, the cells treated with various conditions were lysed in the extraction serum stimulation. Activation of MAP kinase was detected by an buffer, and the protein concentration was determined by Bio-Rad protein assay in-gel kinase assay using MBP as a substrate. When quiescent WI-38 kit (Rio-Rad, Hercules, CA). Equal amounts of protein (10 /Â¿g)were electro- cells were stimulated with serum, MAP kinase activity was observed phoresed in 10% SDS-polyacrylamide gel. The SDS was removed from the within 5 min and then gradually decreased until 120 min. Although a gel, and proteins were renatured in the gel. Kinase activity was measured by dose of 10 /XMphosmidosine caused the complete inhibition of S- incubating the gels in buffer containing [y-32P]ATP. Gels were washed, and phase entry, it did not inhibit serum-induced activation of MAP kinase incorporated radioactivity was analyzed by autoradiography. Western Blotting Analysis of pRB and Cyclin Dl. WI-38 cells were at the same concentration (Fig. 3). Additionally, phosmidosine did not inhibit in vitro MAP kinase activity using MBP as a substrate (data harvested with PBS and sonicated in lysis buffer [50 mM HEPES (pH 7.0), not shown). Thus, phosmidosine blocked the serum-induced G, pro 150 mm NaCl, 0.1% NP40, 10 mM sodium PPÂ¡,1 mM EDTA, 2.5 mM EGTA, 1 mM phenylmethylsulfonyl fluoride, 2% aprotinin, 1 mM DTT, 10 gression without affecting MAP kinase activation; in other words, mM NaF, and 1 mM NaVOJ. The lysates were electrophoresed on 7% (for phosmidosine did not affect signaling events upstream of MAP ki pRB) or 10% SDS-polyacrylamide gel (for cyclin Dl) and then electro- nase. phoretically transferred to a polyvinylidene difluoride membrane. The filter Inhibition of pRB Phosphorylation by Phosmidosine. Although was blocked with 4% (for pRB) or 5% skim milk (for cyclin Dl), washed, the events leading from activation of MAP kinase by growth and incubated with a solution containing primary antibody. After washing factors to progression through the cell cycle are not fully under with PBS-Tween 20, the filter was incubated with horseradish peroxidase- stood, one key switch in regulating the G, to S phase transition is conjugated secondary antibody (Bio-Rad). Protein-antibody conjugates the phosphorylation of pRB. To ascertain whether phosmidosine were visualized with a Western Blot Chemiluminescence Reagent (DuPont affected the phosphorylation of pRB, we performed Western blot NEN, Boston, MA). ting on WI-38 cell extracts with a specific anti-pRB monoclonal CDK4 and CDK2 Assays. WI-38 cells were harvested with PBS and antibody. In serum-starved cells, pRB is detected as a single band sonicated in lysis buffer [50 HIMHEPES (pH 7.5), 150 mM NaCl, 1 nun EDTA, corresponding to the hypophosphorylated form. After serum stim 2.5 HIMEGTA, 1 mM DTT, 0.1% Tween 20, 10% glycerol, 1 mM phenylmeth ulation, only this band is detectable for the first 8 h; however, ylsulfonyl fluoride, 2% aprotinin, 10 mM ÃŸ-glycerophosphate, l mM NaF, and 0.1 mM NaVOJ. The lysates were incubated with anti-CDK4 or anti-CDK2 beginning with 9 h, a second band of lower mobility corresponding antibody for l h at 4Â°C.The immunocomplexes were adsorbed to protein to the hyperphosphorylated form appears and persists through 24 h A-Sepharose 4B and washed extensively with the same lysis buffer. Purified (data not shown). The phosphorylation of pRB occurs before the onset of DNA synthesis, as expected. When quiescent WI-38 cells RB protein (for CDK4) or HI histone (for CDK2) was incubated with the immunocomplexes in 50 /xl of kinase reaction buffer [50 mM HEPES (pH 7.5), were pretreated with various concentrations of phosmidosine for 2.5 mM EGTA, 1 mM DTT, 10 mM MgCl2, 10 mM ÃŸ-glycerophosphate, l mM 15 min then stimulated with 10% serum, pRB phosphorylation, as NaF, 0.1 mM NaVO4, and 20 /J.MATP] containing [y-32P]ATP for 10 min at determined by the appearance of the upper band, was inhibited in 25Â°C(25-28). The samples were analyzed by SDS-PAGE followed by auto- a dose-dependent manner (Fig. 4A). Phosmidosine (10 /MM)was radiography. effective in suppressing pRB phosphorylation when added up to 705

a s E 3

O su Ia 4C 2C 4C

Fig. 2. Phosmidosine arrests WI-38 cells in G,. A, flow cytometric analysis of DNA contents by propidium iodide fluorescence, a, un treated asynchronized population of WI-38 cells; h, cells treated with 10 jaw of phosmidosine for 24 h; c, cells after 24-h wash-out of phosmidosine. H. photographs of WI-38 cells showing BrdUrd incor poration, a-t; same treatment groups described in A. C. dose-depen dent effect of phosmidosine on |lH]thymidine incorporation. Quies cent WI-38 cells were pretreated with phosmidosine for 15 min. stimulated with HKfr PCS for 18 h. and labeled for 2 h with [lH]lhy- midine. Values are the means of triplicate samples; bars. SD. 15000

10000-

5000-

phosmidosine (|iM)

6 h after serum stimulation, but when added after 9 h, there was no with 10 IJLMphosmidosine effectively blocked the kinase activity. inhibition, indicating that the target of phosmidosine-induced cell Phosmidosine did not alter kinase activity of immunoprecipitated cycle arrest occurs in early to mid-G, (Fig. 4B). CDKs from serum-stimulated cells when the drug was tested by in Effect of Phosmidosine on RB Kinases. Phosphorylationof pRB vitro assays (data not shown). Therefore, phosmidosine is not a direct is believed to occur through the action of cyclin-complexed CDK4 or inhibitor of CDKs. The kinase activities of CDK4 and CDK2 require CDK2. To determine the effect of phosmidosine on these RB kinases, association with cyclin D. The protein level of cyclin D is tightly we immunoprecipitated CDK4 or CDK2 from WI-38 cell lysates in regulated during the cell cycle and is increased in response to mito- the absence or presence of phosmidosine (10 /Â¿M)andperformed genie factors. We determined the effect of phosmidosine on cyclin D kinase assays using purified GST-RB or histone HI as substrates for protein levels in WI-38 cells stimulated with serum in the presence or CDK4 and CDK2, respectively. As shown in Fig. 5A, serum stimu absence of phosmidosine. As shown in Fig. 5ÃŸ,theamount of cyclin lation of quiescent cells resulted in increased levels of substrate Dl in cells treated with 10 /J.Mphosmidosine is less than the amount phosphorylation by CDK4 and CDK2, but pretreatment of these cells in control cells. The effect of the drug on cyclin Dl protein correlates

none phosmidosine Fig. 3. Effect of phosmidosine on serum-induced MAP kinase (ERKI/ERK2I aclivation in WI-38 cells. Quiescent WI-38 cells were treated with or without IO JXMphosmidosine for I.*Â»minand then 30 120 30 120 (min) stimulated with 10% FCS. Cells were harvested at the indicated times and analy/ed for MAP kinase activity (see "Materials and Methods"). ERK1 Results are representative of three independent experiments. ERK2

706

2.5 10 20 |iM phosinidosine

jhypcrphosphorylated pRb Fig. 4. Inhibition of pRB phosphorylation by phosmidosine in WI-38 ] hypophosphorylatcd pRb cells. In A. quiescent WI-38 cells were pretreated with various concen trations (>iM)of phosmidosine for 15 min and then stimulated with 10% FCS. After 18 h, the cells were collected and lysed. Samples were analyzed by Western blotting using anti-pRB antibody. Lane G0, no serum addition. Results are representative of three independent experi ments. B, quiescent WI-38 cells were stimulated with 10% FCS. Phos midosine (10 H.M)was added at the indicated lime after serum stimula B tion. After 18 h, the cells were collected and lysed. Samples were Time of phosmidosine addition analyzed by Western blotting using anti-pRB antibody. G(). no serum addition; -, no phosmidosine addition. Results are representative of (h after serum stimulation) three independent experiments. GÂ« 12 -

ÃŽhyperphosphorylatedpRb hypophosphorylated pRb

with the effect of phosmidosine on Rb phosphorylation in the same we treated cells with 5 Â¡J.Mphosmidosine (Fig. 6e) or with two other cells. cell cycle inhibitors, 300 /J.Mmimosine (Fig. 6/) or 3 H.Maphidicolin Effect of Phosmidosine on Cell Cycle Progression in Synchro (Fig. 6g), at the time of release from the temperature-dependent arrest nous tsFT210 Cells. In the temperature-sensitive cell line tsFT210 and cultured the cells for 12 h. All three drugs allowed passage with a mutation in the cdc2 gene, normal growth is observed at the through G2 and M phases and gave characteristic G, arrest profiles. permissive temperature of 32Â°C(Fig. 6a), but cells arrest at G,-M at After removal of the drugs by extensive washing, phosmidosine- the nonpermissive temperature of 39Â°Cas a result of inactivation of treated cells did not show any change in cell cycle profile (Fig. 6, h p34cdc2 (Fig. 6b). Arrested cells can reenter the cell cycle by culturing compared with e), whereas cells treated with either mimosine or at 32Â°C.After 4 h at the permissive temperature, a significant fraction aphidicolin reentered the cell cycle (data not shown). We next at of the cells have progressed through mitosis into G, (Fig. 6c), and tempted to establish a temporal relationship between the effective after 12 h, cells undergoing S phase can be observed (Fig. 6d). To arrest points by phosmidosine and the other inhibitors using combi determine whether phosmidosine affects the G2-M-to-G, transition. nations of the drugs in synchronization and release experiments. Cells

â€¢GST-Rb

Fig. 5. Effect of phosmidosine on Rb kinase activity and cyclin Dl expression by phosmidosine in WI-38 cells. A. quiescent WI-38 cells were pretreated with or without 10 JIM phosmidosine for 15 min and â€¢HI then stimulated with 10% FCS. After 21 h, the cells were collected and lysed. CDK4 activity was assayed using the CDK4 immune complexes and GST-Rb as a substrate (upper panel}. CDK2 activity was assayed using the CDK2 immune complexes and histone HI as a substrate (lower panel). Lane I, quiescent cells; Lane 2, serum-stimulated cells without phosmidosine; Lane 3, serum-stimulated cells with 10 Â¿IM phosmidosine. Results are representative of three independent experi ments. B, quiescent WI-38 cells were stimulated with 10% PCS at the time 0 h. Phosmidosine (10 JÃŒM)wasadded at the indicated time. Cells B were harvested at the indicated times, and the expression levels of pRB (upper panel} and cyclin DI (lower panel} were analyzed by Western none phosmidosine blotting using their antibodies, as described in "Materials and Methods." The indicated times of upper and lower panels show the harvested time 24 24 24 harvested time (h) (h) and the time of phosmidosine addition (h) after serum stimulation, respectively. Results are representative of three independent experi 2 6 time of phosmidosine addition (h) ments. â€”|hyperphosphorylatcd pRb â€¢<â€¢ â€¢* â€”Ihypophosphorvlatcd pRb

â€” cyclin Dl

707

gelative Â°Cl,2C cellnuma32 Â°C^ Â°C

17 hbi32 4hC12

Fig. 6. Effect of phosmidosine in tsFT210 cells. Flow cytometry was used to monitor the cell cycle Â°C\dHâ€¢f 4C39 h, 32 arrest and release of lsFT210 cells. lsFT210 cells were cultured at the permissive temperature of nonefÃ L- phosmidosine1 mimosinefgt1 aphidicolintwashed, 32Â°C(a) and then incubated for 17 h at the non- 'Ã fwashed permissive temperature of 39Â°C(b}. After release from cell cycle arrest by shifting cells lo 32Â°C,they were incubated for 4 h (c) or 12 h in the absence (d) or presence of 5 JIM phosmidosine (r), 300 U.M mimosine (/), and 3 /XMaphidicolin (#)- The phosmidosine-arresied cells were washed with fresh medium and cultured for another 6 h at 32Â°C(h). The mimosine (/)- or aphidicolin (/(-arrested cells were washed with fresh medium and cultured for â€”eL J+ another 6 h at 32Â°Cin the presence of 5 /IM phosmidosine, respectively. Fluorescence corre washed, then added then added sponding to 2 C and 4 C DNA contents are repre 16h,32Â°Ch1.iI1+ phosmidosine phosmidosine r6h,32Â°CjkA sented by vertical bars. Results are representative Â°CLL-\\+6 h, 32 of three independent experiments.

synchronized by G2-M arrest at 39Â°Cwere released into either mi rylations, and can be constrained by at least two families of CDK mosine (300 /XM)- or aphidicolin (3 /nM)-containing medium and inhibitory proteins (32, 33). To date, at least seven mammalian CDKs incubated at 32Â°Cfor 12 h to produce G, arrest. The mimosine or have been identified (34, 35). CDK1 (also called CDC2) is involved aphidicolin was washed out and replaced with fresh medium contain in regulation of the G2-to-M transition, in association with cyclin B ing 5 Â¿IMphosmidosine. The addition of phosmidosine did not inhibit (36). CDK1 associates with Cyclin A, and apparently this complex the release of the cells from either mimosine (Fig. 6i> or aphidicolin also plays a role in the G2-to-M transition (37). Because phosmido- (Fig. 6/')-induced G, block. Therefore, phosmidosine irreversibly ar sine-treated cells were not arrested at the G2-M phase (Fig. 6), it rests these cells in G,, as with WI-38 cells, at a specific point that seems likely that CDK1 activity is not affected by phosmidosine. appears earlier in the cell cycle than the arrest point for mimosine or CDK2 is involved in regulating the G,-to-S transition by its associ aphidicolin. ation with cyclin E and cyclin A (37-41). CDK4is the major catalytic partner for cyclin Dl, and its activation is induced prior to that of DISCUSSION CDK2; as a result, this complex can phosphorylate the pRB (27). Phosmidosine decreased cyclin Dl protein levels, and CDK4 activity Previous investigation suggested that a nucleotide antibiotic, phos immunoprecipitated with anti-CDK4 antibody was lower in phosmidosine, was an inhibitor of the mammalian cell cycle at G, (20). midosine-treated extracts (Fig. 5). Moreover, phosmidosine greatly However, its mode of action was unclear. Here, we studied the mechanism of action of phosmidosine-induced G, arrest using the suppressed the histone HI kinase activity of CDK2 immunoprecipi tated with anti-CDK2 antibody (Fig. 5). Cyclin Dl also makes com diploid human fibroblast cell line and the cdc2 mutant cell line. Phosmidosine inhibited S-phase entry in both asynchronous and syn plexes with CDK5 (26) and CDK6 (35), but the function of these complexes is unclear. Complete activation of CDKs also requires chronous cultured cells (Fig. 2). The G, arrest was irreversible but not due to cell toxicity, as determined by trypan blue exclusion. phosphorylation of their conserved threonine residue by CAK. CAK The signal transduction mechanism evoked by serum or mitogen is a CDK complex consisting of a catalytic subunit, CDK7(MO15), and a regulatory subunit, cyclin H. Complexes of D-type cyclins and such as epidermal growth factor and nerve growth factor have been CDK4 are catalytically inactive until the complexes are phosphoryl- intensively investigated, and a phosphorylation cascade is thought to ated on a threonine residue (Thr-172) by CAK (41, 42). which resides play an essential role in the transduction of mitogenic signals. MAP kinase (ERK1/ERK2) is one of the key kinases involved in the in the same position and general context as Thr-161/160 in CDC2/ phosphorylation cascade activated by serum stimulation (24, 29). CDK2. We cannot exclude the possibility that phosmidosine acts However, phosmidosine did not affect the activation step of MAP either by preventing the formation of functional CAK or by interfering kinase (Fig. 3), consistent with our findings that addition of phos with its ability to access its substrate in addition to the decrease of midosine 6 h after serum addition still caused G, arrest (data not cyclin Dl protein level. shown). pRB and other pRB-like proteins (pi30 and p 107) control gene Passage through the restriction point and entry into S phase is expression mediated by a family of heterodimeric transcriptional controlled by CDKs that are sequentially regulated by mainly cyclins regulators, termed the E2Fs, which transactivate genes whose prod D, E, and A (30, 31). In general. CDK activity regulates cyclin ucts are important for S-phase entry (2, 4). In its hypophosphorylated binding, depends on both positive and negative regulatory phospho- form, RB binds to a subset of E2F complexes, converting them to 708

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1998 American Association for Cancer Research. PHOSMIDOSINE EFFECTS ON CYCLIN Dl AND pRB repressers that constrain expression of E2F target genes. Phosphoryl- 7. Weintraub. S. J.. Prater, C. A., and Dean. C. Retinoblastoma protein switches the E2F ation of pRB causes dissociation of E2Fs and permits transcriptional site from positive to negative element. Nature (Lond.), 358: 259-261, 1992. 8. Helin, K., Lees, J. A.. Vidal, M., Dyson, N.. Harlow, E., and Fattaey, A. A cDNA activation of the same genes, a process initially triggered by the cyclin encoding a pRB-binding protein with properties of the transcription factor E2F. Cell, D-dependent kinases and then accelerated by cyclin E-CDK2 (9, 27). 70: 337-350, 1992. Phosmidosine suppressed hyperphosphorylation of pRB by pRB-ki- 9. Sherr, C. J. The ins and outs of RB: coupling gene expression to the cell cycle block. Trends Cell Biol.. 4: 15-18. 1994. nases such as CDK4 and CDK2 after serum stimulation and inhibited 10. Taya. Y., Yasuda. H.. Kamijo, M.. Nakaya. K., Nakamura. Y.. Ohba, Y., and S phase entry (Fig. 4). Nishimura. S. In vitro phosphorylation of Ihe tumor suppressor gene RB proiein by mitosis-specific histone Hl kinase. Biochem Biophys. Res. Commun.. 164: 580-586, Cell cycle inhibitors are useful to dissect the cell cycle progression. 1989. In our experiment, two G, arresters were used as reversible inhibitors: 11. Lees, J. A., Buchkovich, K. J.. Marshak. D. R., Anderson. C. W., and Harlow. E. The mimosine and aphidicolin. Mimosine, a plant amino acid, reversibly retinoblastoma protein is phosphorylated on multiple sites by human cdc2. EMBO J., 10: 4279-4290, 1991. blocks the cell cycle at a point that occurs before the reversible arrest 12. Lin, B. T-L., Gruenwald, S.. Moria. A. O.. Lee. W-H.. and Wang. J. Y. J. Retino mediated by aphidicolin, an inhibitor of DNA polymerase a activity, blastoma cancer suppressor gene product is a substrate of the cell regulator ctlc2 which defines the G.-S boundary (43-45). Mimosine exerts its effects kinase. EMBO J., SO: 857-864. 1991. 13. Hamel, P. A.. Gill, R. M., Phillips, R. A., and Gallic, B. L. Region controlling on cell cycle progression via the inhibition of histone HI kinase hyperphosphorylation and conformation of the retinoblastoma gene product are synthesis (45). The restriction or R point has been shown to occur 2-3 independent of domains required for transcriplional repression. Oncogene. 7: 693- 701, 1992. h prior to the start of S phase in serum-stimulated fibroblasts and to 14. Quelle, D. E., Ashmun. R. A., Shurtleff, S. J.. Hato. J., Bar-Sagi, D.. Roussel, M. F.. correspond to the point of commitment to initiate DNA synthesis (45). and Sherr, C. J. Overexpression of mouse D-type cyclins accelerates Gl phase in Phosmidosine blocked the cell cycle at an earlier point than that of rodent fibroblasts. Genes Dev., 7: 1559-1571, 1993. 15. Resnitzky. D.. Gossen. M., Bujard, H., and Reed, S. I. Acceleration of the G,-S phase mimosine or aphidicolin (Fig. 6). It is likely, therefore, that the transition by expression of cyclins Dl and E with an inducible system. Mol. Cell. phosmidosine arrest point is located more distally from the R point in Biol., 14: 1669-1679, 1994. 16. Baldin, V., Lukas. J.. MarcÃ³te. M. J.. Pagano, M., and Draetta. G. Cyclin D is a G,, although the temporal relationship between the two points will nuclear protein required for cell cycle progression in Gl. Genes Dev.. 7: 812-821, require further investigations. In addition, mimosine or aphidicolin 1993. arrested the cell cycle at the G,-S boundary after the phosphorylation 17. Lukas, J., Muller. H., Bartkova, J.. Spitkovsky, D., Kjerulff, A. A.. Jansen-Durr, P., Strauss, M.. and Bartek, J. DNA tumor virus oncoproteins and retinoblastoma gene of pRB because hyperphosphorylated pRB accumulated in mimosine- mutations share the ability to relative the cell's requirement for cyclin Dl function in or aphidicolin-treated cells (28, 35). In contrast, hyperphosphorylated Gl. J. Cell Biol., 125: 625-638, 1994. pRB did not accumulate by phosmidosine treatment (Figs. 4 and 5). 18. Uramoto, M.. Kim. C. J., Shinya, K., Kusakabe, H., and Isono, K. Isolation and characterization of phosmidosine a new antifungal nucleotide antibiotic. J. Antibiot.. These results suggest that the phosmidosine arrest point is distinct 44: 375-381. 1991. from the mimosine or aphidicolin arrest point. The inability of phos 19. Phillips, D. R.. Uramoto, K., Isono. K., and McCloskey. J. M. Structure of the midosine to block progression of cell cycle after release from mimo antifungal nucleotide antibiotic phosmidosine. J. Org. Chem., 58: 854-859. 1993. 20. Matsuura. N., Onose, R., and Osada. H. Morphology reversion activity of phosmido sine- or aphidicolin-induced G, block further supports this idea, sine and phosmidosine B, a newly isolated derivatives, on sre transformed NRK cells. although further experiment will be necessary to understand the more J. Antibiol., 49: 361-365. 1996. 21. Mineo. C., Murakami. Y.. Ishimi. Y., Hanaoka. F.. and Yamada. M. Isolation and detailed molecular mechanism of phosmidosine G, arrest. analysis of a mammalian temperature-sensitive mutant defective in G2 functions. Recent studies revealed that D-type cyclins are implicated in can Exp. Cell Res., 167: 53-62, 1986. cer; gene amplification of cyclin Dl has been found in breast cancer, 22. Osada, H., Cui, C-B., Onose, R., and Hanaoka, F. Screening of cell cycle inhibitors from microbial metabolites by a bioassay using a mouse cdc2 mutant cell line, pancreatic carcinoma, esophageal carcinoma, and squamous cell car tsFT210. Bioorg. Med. Chem.. 5: 193-203, 1997. cinoma (47-50). Besides gene amplification of cyclin Dl, its accu 23. Noguchi, P. D., and Browne, W. C. The use of erythrocyte nuclei as a biological mulation in human pancreatic carcinoma was associated with short standard for flow microfluorometry. J. Histochem. Cytochem., 26: 761-763, 1978. 24. Gotoh, Y.. Nishida. E.. Yamashita, T., Hoshi, M., Kawakami, M.. and Sakai, H. ened survival (48). Moreover, cyclin Dl cooperates with other Microtubule-associated-protein (MAP) kinase activated by nerve growth factor oncogenes to transform fibroblasts, and in transgenic mice, overex- and epidermal growth factor in PC12 cells. Eur. J. Biochem.. 193: 661-669, pression of cyclin Dl in breast epithelium results in breast cancer 1990. 25. Matsushime. H., Quelle. D. E., Shurtleff, S. A., Shibuya, M., Sherr, C. J., and Kalo, (51-53). In conclusion, phosmidosine suppressed cyclin Dl protein J. D-type cyclin-dependent kinase activity in mammalian cells. Mol. Cell. Biol.. 14: levels and the phosphorylation of pRB in response to mitogen stim 2066-2076. 1994. 26. Xiong. Y., Zhang, H., and Beach. D. D-type cyclins associate with multiple protein ulation and irreversibly arrested the cell cycle in G,. Such an inhibitor kinases and the DNA replication and repair factor PCNA. Cell, 71: 505-514, 1992. will be useful for the understanding the regulatory mechanisms of G, 27. Kato. J. Matsushime, H.. Hiebert, S. W.. Ewen, M. E.. and Sherr, C. J. Direct binding progression and may contribute to a new approach to cancer chemo of cyclin D to the retinoblastoma gene product (pRB) and pRB phosphorylation by the cyclin D-dependent kinase CDK4. Genes Dev., 7: 331-342. 1993. therapy. 28. Kitagawa. M., Higashi, H.. Takahashi, I. S., Okabe, T., Ogino, H., Taya, Y., Nishimura, S., and Okayama, A. A cyclin-dependent kinase inhibitor, butyrolactone I, inhibits phosphorylation of RB protein and cell cycle progression. Oncogene. 9: ACKNOWLEDGMENTS 2549-2557, 1994. 29. Pelech, S. L.. and Sanghera. J. S. MAP kinases: charting the regulatory pathways. We are grateful to Dr. F. Hanaoka (RIKEN) for the gift of tsFT210 cells and Science (Washington DC), 257- 1355-1356, 1992. Dr. M. Imoto (Keio University) for helpful suggestions. 30. Norbury, C., and Nurse, P. A. Animal cell cycle and their control. Annu. Rev. Biochem., 61: 441-470, 1992. 31. Reed, S. I. The role of p34 kinases in the GI to S-phase transition. Annu. Rev. Cell REFERENCES Biol., 8: 529-561, 1992. 32. Elledge. S. J.. and Harper, J. W. Cdk inhibitors: on the threshold of checkpoints and 1. Weinberg, R. A. Tumor suppressor genes. Science (Washington DC), 254: 1138- development. Curr. Opin. Cell Biol.. 6: 847-852, 1994. 1146, 1991. 33. Sherr, C. J.. and Roberts. J. M. Inhibitors of mammalian G l cyclin-dependent kinases. 2. Nevins, J. R. E2F: a link between the Rb tumor suppressor protein and viral Genes Dev., 9: 1149-1163. 1995. oncoproteins. Science (Washington DC), 258: 424-429. 1992. 34. Sherr, C. Mammalian Gl cyclins. Cell, 73: 1059-1065. 1993. 3. Zalvide, J., and DeCaprio, J. A. Role of pRB-related proteins in simian virus 40 35. Meyerson, M.. and Harlow, E. Identification of a GI kinase activity for cdk6, a novel large-T-antigen-mediated transformation. Mol. Cell. Biol.. /5: 5800-5810. 1995. cyclin D partner. Mol. Cell. Biol., 14: 2077-2086. 1994. 4. La Thangue, N. B. RDTF1/E2F: an expanding family of heterodimeric transcription 36. Pines. J., and Hunter, T. Isolation of a human cyclin cDNA: evidence for cyclin factors implicated in cell-cycle control. Trends Biochem. Sci., 19: 108-114, 1994. mRNA and protein regulation in the cell cycle and for interaction with p34cdt'2. Cell, 5. Chellapan, S. P., Hiebert. S., Mudryj, M., Horowitz, J. M., and Nevins, J. R. The E2F 58: 833-846, 1989. transcription factor is a cellular target for the RB protein. Cell. 65: 1053-1061, 1991. 37. Pagano. M.. Pepperkok. R.. Verde. F., Ansorge. W.. and Draetta. G. Cyclin A is 6. Rustgi, A. K., Dyson. N-, and Bernards, R. Amino-terminal domains of c-mvc and required at two points in human cell cycle. EMBO J.. //.- 961-971, 1992. N-mvc proteins mediate binding to the retinoblastoma gene product. Nature (Lond.). 38. Ohtsubo. M.. and Roberts. J. M. Cyclin-dependent regulation of GI in mammalian 352: 541-544, 1991. fibroblasts. Science (Washington DC). 259.- 1908-1912, 1993.

709

39. Tarn. S. W., Theodoras, A. M., Shay, J. W., Draetta. G. F., and Pagano, M. some 1Iql3, is amplified and expressed in squamous cell and mammary carcinoma Differential expression and regulation of cyclin DI protein in normal and tumor and linked to BCL-1. Oncogene, 6: 439-444, 1991. human cells: association with cdk4 is required for cyclin Dl function in Gl progres 48. Gansauge. S., Gansauge. F., Ramadani, M., Stobbe, H., Rau, B., l lar.nl.i. N., and sion. Oncogene, 9: 2663-2674, 1994. Beger, H. G. Overexpression of cyclin Dl in human pancreatic carcinoma is associ 40. Girard, F. Strusfeld. U., Fernandez. A., and Lamb, N. Cyclin A is required for the ated with poor prognosis. Cancer Res., 57: 1634-1637, 1997. onset of DNA replication in mammalian fibroblasts. Cell, 67: 1169-1179, 1991. 49. Jiang, W., Kahn, S. M., Tornita, N., Zhang, Y. J., Lu, S. H., and Weinstein, I. B. 41. Morgan, D. O. Principles of CDKs regulation. Nature (Lend.), 374: 131-134, 1995. Amplification and expression of the human cyclin D gene in esophageal cancer. 42. Makela, T. P., Tassan, J. P., Nigg, E. A., Frutiger, S., Hughes, G. J., and Weinberg, Cancer Res., 52: 2980-2983, 1992. R. A. A cyclin associated with the CDK-activating kinase MO 15. Nature (Lond.), 50. Shuuring, E., Verhoeven, E., Mooi, W. J., and Michalides, R. J. Identification and 371: 254-257, 1994. cloning of two overexpressed genes, U2IB31/PRADI and EMSI, within the amplified 43. Huberman, J. A., New views of the biochemistry of eukaryotic DNA replication chromosome Ilql3 region in human carcinomas. Oncogene, 7: 355-361, 1992. revealed by aphidicolin. an unusual inhibitor of polymerase a. Cell, 23: 647-648, 51. Hinds, P. W., Dowdy, S. F., Eaton, E. N., Arnold, A., and Weinstein, R. A. Function 1981. of a human cyclin gene as an oncogene. Proc. Nati. Acad. Sci. USA, 91: 709-713, 44. Lalande, M. A reversible arrest point in the late Gl phase of the mammalian cell cycle. Exp. Cell. Res., 186: 332-339, 1990. 1994. 45. Feldman, S. T., and Schonthal, A. Negative regulation of histone HI kinase expres 52. Lovec, H., Sewing, A., Lucibello, F. C., Muller, R., and Moroy, T. Oncogenic activity sion by mimosine, a plant amino acid. Cancer Res., 54: 494-498, 1994. of cyclin Dl revealed through cooperation with Ha-ras: link between cell cycle 46. Weintraub. S. J. Mechanism of active transcriptional repression by the retinoblastoma control and malignant transformation. Oncogene, 9: 323-326, 1994. protein. Nature (Lond.), 375: 812-815, 1995. 53. Wang, T. C., Cardiff, R. D., Zukerberg, L., Lees, E., Arnold, A., and Schmidt, E. V. 47. Lammie, G. A., Fanti, V., Smith, R., Schuuring, E., Brookes, S., Michalides, R., Mammary hyperplasia and carcinoma in MMTV-cyclin Dl transgenic mice. Nature Dickson, A., Arnold, A., and Petter, G. DIIS287, a putative oncogene on chromo (Lond.). 569:669-671, 1994.

710

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1998 American Association for Cancer Research. Inhibition of Cyclin D1 Expression and Phosphorylation of Retinoblastoma Protein by Phosmidosine, a Nucleotide Antibiotic

Hideaki Kakeya, Rie Onose, Phillip C.-C. Liu, et al.

Cancer Res 1998;58:704-710.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/58/4/704

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/58/4/704. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.