Calmodulin-Dependent Cyclic Nucleotide Phosphodiesterase

[CANCER RESEARCH 64, 2568–2571, April 1, 2004] Calmodulin-Dependent Cyclic Nucleotide Phosphodiesterase (PDE1) Is a Pharmacological Target of Differentiation-Inducing Factor-1, an Antitumor Agent Isolated from Dictyostelium

Kasumi Shimizu,1 Taku Murata,1 Toshiro Tagawa,1 Katsunori Takahashi,2 Ryoki Ishikawa,3 Yumiko Abe,4 Kohei Hosaka,2 and Yuzuru Kubohara5 1Department of Oral and Maxillofacial Surgery, Faculty of Medicine, Mie University, Mie, Japan; 2Department of Basic Sciences for Medicine, Gunma University School of Health Sciences, Maebashi, Japan; 3Departments of Pharmacology, and 4Obstetrics and Gynecology, Gunma University School of Medicine, Maebashi, Japan; and 5Biosignal Research Center, Institute for Molecular and Cellular Regulation (IMCR), Gunma University, Maebashi, Japan

ABSTRACT In this study, we have tried to identify the target molecule(s) of DIF-1 in mammalian cells, and we show here that pharmacological The differentiation-inducing factor-1 (DIF-1) isolated from Dictyoste- concentrations of DIF-1 inhibit the calmodulin (CaM)-dependent cy- lium discoideum is a potent antiproliferative agent that induces growth clic nucleotide phosphodiesterase (PDE1) in vitro. Because DIF-1 did arrest and differentiation in mammalian cells in vitro. However, the specific target molecule(s) of DIF-1 has not been identified. In this study, not significantly affect the activity of some other PDEs or CaM- we have tried to identify the target molecule(s) of DIF-1 in mammalian dependent enzymes and, in addition, an isomer of DIF-1 was a less cells, examining the effects of DIF-1 and its analogs on the activity of some potent inhibitor, we have concluded that PDE1 is a pharmacological candidate enzymes. DIF-1 at 10–40 ␮M dose-dependently suppressed cell and specific target of DIF-1 in mammalian cells. growth and increased the intracellular cyclic AMP concentration in K562 leukemia cells. It was then found that DIF-1 at 0.5–20 ␮M inhibited the MATERIALS AND METHODS calmodulin (CaM)-dependent cyclic nucleotide phosphodiesterase (PDE1) in vitro in a dose-dependent manner. Kinetic analysis revealed that DIF-1 Reagents and Enzymes. DIF-1 and its analogs were synthesized by acted as a competitive inhibitor of PDE1 versus the substrate cyclic AMP. Toyama Chemical Co. Ltd. (Toyama, Japan) following the method of Masento Because DIF-1 did not significantly affect the activity of other PDEs or et al. (14). Bovine brain calcineurin, CaM, PDE1, p-nitrophenylphosphate, and CaM-dependent enzymes and, in addition, an isomer of DIF-1 was a less bis-p-nitrophenylphosphate were obtained from Sigma (St. Louis, MO), snake potent inhibitor, we have concluded that PDE1 is a pharmacological and venom PDE (svPDE) was from the Worthington Biochemical Corporation specific target of DIF-1. (Lakewood, NJ), and calf-intestinal alkaline phosphatase (AP) was from New England BioLabs (Beverly, MA). Smooth-muscle myosin and myosin light- chain kinase were purified from chicken gizzard as described previously (15, INTRODUCTION 16). The definitions of units (U) for enzymatic activities used in this study are The differentiation-inducing factor-1 [DIF-1; 1-(3,5-dichloro-2,6- in accordance with the manufacturer’s descriptions. dihydroxy-4-methoxyphenyl)hexan-1-one; Fig. 1] is a signal mole- Assay for Cell Growth. The human leukemia K562 cells were maintained at 37°C (5% CO ) in a growth medium (RPMI 1640 with 10% fetal bovine cule that induces stalk cell differentiation in Dictyostelium discoideum 2 serum; designated RPMI), and cell growth was assessed as described previ- (1, 2). DIF-3 (Fig. 1), a dechlorinated form of DIF-1, is the first ously (8). Briefly, cells were incubated in a multi(12)-well plate, each well metabolite of DIF-1, whose stalk-inducing activity is very low in D. containing 1 ml of RPMI (5 ϫ 104 cells/ml) in the presence or absence of discoideum (2, 3); on the other hand, 2-methoxy isomer of DIF-1 DIF-analogs or inhibitors for CaM or PDEs. On day 3, a 1:20 volume (50 ␮l) (2-MIDIF-1) and 6-methoxy isomer of DIF-3 (6-MIDIF-3; Fig. 1) are of Alamar Blue (cell number indicator) was added to each well, and after a ␮ the artificially synthesized isomers of DIF-1 and DIF-3, respectively. 1–2-h incubation at 37°C (5% CO2), 150 l of each of the sample solutions Recently, it has been shown that DIF-1 and DIF-3 exhibit antipro- were transferred into a 96-well plate, and absorbance at 570 nm (reference at liferative activities and occasionally induce cell differentiation in 595 nm) was measured with a microplate reader (Bio-Rad; Model 550). A cell mammalian cells (4–12). Some results obtained by our research number was given as % of the control absorbance. ϩ ϫ 6 2 Assay for [cAMP]i. Cells were suspended (2 10 cells/ml) in an assay group, namely, that DIF-1 increases [Ca ]i in some tumor cells buffer [20 mM HEPES-NaOH (pH 7.4), 137.5 mM NaCl, 5 mM KCl, 1.5 mM (6–9), activates Akt/protein kinase B human leukemia K562 cells (9), CaCl2, 0.8 mM MgCl2, 5.5 mM glucose, 0.6 mM NaHCO3, and 0.1% (w/v) inactivates signal transducer and activator of transcription 3 (STAT3) BSA]. One ml of the cell suspension in a tube was incubated at room in gastric cancer cells (11), and down-regulates G1 cyclins in vascular temperature with or without 10–40 ␮M DIF-1 for the indicated minutes (Fig. smooth muscle cells (10), are significant in the elucidation of the 2), and the cells were collected by quick centrifugation (3500 rpm, 30 s), molecular mechanism of the actions of DIF-1 in mammalian cells. resuspended in 100 ␮lof0.2N HCl, and destroyed by mild sonication. The Moreover, we have found that DIF-1 may block a decrease in the samples were neutralized and assayed for cAMP content as described previ-

intracellular cyclic AMP concentration ([cAMP]i) induced by proges- ously (17). terone and thereby inhibit progesterone-induced oocyte maturation in Assay for PDE1 Activity. Bovine PDE1 (0.75 mU), CaM (10 mU), and Xenopus laevis (13). Yet, the precise signaling system of DIF-1 and, CaCl2 (0.2 mM) were incubated at 30°C for 10 min in 0.3 ml of a reaction buffer {50 mM HEPES-NaOH (pH 7.5), 0.1 mM EGTA, 8.3 mM MgCl , 0.5 ␮M especially, the target molecule(s) of DIF-1 are unknown in Dictyos- 2 [3H]cAMP (18,000 cpm)} containing DIF-analogs or inhibitors. In studies telium, Xenopus, and mammalian cells. measuring PDE1 activity in the absence of CaM, PDE1 (3 mU) was incubated at 30°C for 15 min in a reaction buffer containing DIF-analogs or inhibitors. Received 11/12/03; revised 1/19/04; accepted 1/29/04. PDE1 activity was assayed by a modification of a previously described Grant support: Supported in part by grants from the Ministry of Education, Science, procedure (18). Sports and Culture of Japan (Y. Kubohara). The costs of publication of this article were defrayed in part by the payment of page Assay for PDE3A, PDE3B, and PDE8A. Baculoviruses of the human charges. This article must therefore be hereby marked advertisement in accordance with full-length PDE3A, human PDE3A NH2-terminal deletion mutant (amino 18 U.S.C. Section 1734 solely to indicate this fact. acids 511-1141; Ref. 19), human full-length PDE3B (20), and rat PDE3B Note: K. Shimizu and T. Murata contributed equally to this work. NH -terminal deletion mutant (amino acids 577-1108; Ref. 21) were gifts of Requests for reprints: Kubohara, Biosignal Research Center, Institute for Molecular 2 and Cellular Regulation (IMCR), Gunma University, Maebashi 371-8512, Japan. Phone: Dr. V. C. Manganiello (NIH, Bethesda, MD). Mouse PDE8A was prepared as 81-27-220-8866; Fax: 81-27-220-8897; E-mail: [email protected]. follows. Total RNA was isolated from mouse osteoblast MC3T3-E1 cells using 2568

Sf9 cells as described previously (22). Sf9 cells were maintained, propagated at 27°C in the TNM-FH Insect Medium (PharMingen), and were collected and sonicated in 1 ml of a homogenization buffer [100 mM Tris-HCl (pH 7.4) and

5mM MgSO4]. The homogenized cell samples containing full-length PDE3A or PDE3B were used for the assay for PDE activity. And the homogenized cells containing the truncated forms of PDE3A or PDE3B or the full-length PDE8A were centrifuged (100,000 ϫ g, 60 min, 4°C) to obtain soluble fractions, which were also used for the assay for PDE activity. Samples were incubated at 30°C for 10 min in 0.3 ml of a reaction buffer ␮ {50 mM HEPES-NaOH (pH 7.4), 0.1 mM EGTA, 8.3 mM MgCl2, 0.1 M [3H]cAMP (18,000 cpm)} in the presence of DIF-1 or inhibitors. PDE activity was assayed as described previously (18). Fig. 1. Chemical structure of DIF-analogs. DIF-1, differentiation-inducing factor-1 Assay for the Other Enzyme Activities. Three units each of bovine [1-(3,5-dichloro-2,6-dihydroxy-4-methoxyphenyl)hexan-1-one; DIF-3, differentiation-in- calcineurin (ϳ 0.1 ␮M) and CaM (ϳ 0.2 ␮M) were incubated at 30°C for 60 ducing factor-3 [1-(3-chloro-2,6-dihydroxy-4-methoxyphenyl)hexan-1-one]; 2-MIDIF-1, ␮ 2-methoxy isomer of DIF-1; 6-MIDIF-3, 6-methoxy isomer of DIF-3. min in 200 l of a reaction buffer in the presence or absence of various concentrations of DIF-1, and calcineurin activity was assayed as described previously (23). Chicken gizzard myosin (3.7 ␮M) was phosphorylated with 0.086 ␮M myosin light-chain kinase and 0.3 ␮M CaM in a reaction buffer (23) in the presence or absence of 10–20 ␮M DIF-1 or 10 ␮M calmidazolium (CZM) at 25°C for 20 min. After terminating the reaction by adding an equal volume of a sample buffer [6 M urea, 14 mM 2-mercaptoethanol, and 50 mM Tris-HCl (pH 6.8)], samples were analyzed with urea-glycerol PAGE as described previously (24). svPDE (0.1 unit) was incubated at 37°C for 15 min in 200 ␮l of an assay buffer [138 mM Tris-HCl (pH 8.7), 9 mM bis-p-nitrophenylphosphate], and the ␮ reaction was stopped by adding 800 lof1M Na2CO3. svPDE activity was quantified by measuring the A410 nm of the reaction mixtures. AP (26 mU) was incubated at 37°C for 5 min in 200 ␮l of an assay buffer

Fig. 2. Effects of differentiation-inducing factor-1 (DIF-1) on intracellular cyclic AMP ([cAMP]i) and cell growth in K562 cells. A, K562 cells were incubated for the indicated times with 20 ␮M DIF-1, and cAMP content was assessed. Values are the mean Ϯ SD of control (by ANOVA, postء P Ͻ 0.03 versus ,ءء .(three independent experiments (n ϭ 3 hoc Fisher’s protected least significant difference). B, K562 cells were incubated for 5 min with 10–40 ␮M DIF-1, and cAMP content was assessed. Values are the mean Ϯ one-half ϭ of the range of two independent experiments (n 2). DIF-1 increased [cAMP]i in a .controlء P Ͻ 0.005 versus ,ءءء ;controlء P Ͻ 0.05 versus ,ءء .dose-dependent manner C, K562 cells were incubated for 3 days with the indicated drugs, and the cell number was assessed. Drugs used are the DIF-analogs, DIF-1, DIF-3, 2-methoxy isomer of DIF-1 (2-MIDIF-1), and 6-methoxy isomer of DIF-3 (6-MIDIF-3); the calmodulin inhibitors, calmidazolium (CZM) and W-7; the PDE1 inhibitor, 8-methoxymethyl-3-isobutyl-1-methylxanthine (8-MIBMX); and the nonselective PDE inhibitor, 3-isobutyl-1-methylxanthine (IBMX). Values are the mean Ϯ SD of three independent experiments (n ϭ 3). Fig. 3. Effects of differentiation-inducing factor-1 (DIF-1) and its analogs on phosphodiesterase (PDE1) activity. A, bovine PDE1 (5 mU/tube) was incubated in vitro in an assay buffer with various concentrations of DIF-1 (F) or W-7 (E) in the presence (E, F, the RNeasy mini kit (Qiagen, Hilden, Germany). A specific oligonucleotide ϩ CaM) or absence (f, Ϫ CaM) of bovine calmodulin (CaM; 10 units/tube) for 10 min, primer set was synthesized: 5Ј-CCTAAATGTCTGCCTCGTTTGCTAGTG-3Ј and the samples were assayed for enzymatic activity. Every sample contains 1% (v/v) and 5Ј-GTGCCGCCGCCGCCAGTATGGGCTGCGCCCCG-3Ј; and reverse ethanol as a vehicle. Values are the mean Ϯ one-half of the range of two independent ϭ transcription-PCR was carried out using the Qiagen OneStep RT-PCR kit experiments (n 2). B, PDE1 (10 mU/tube) was incubated in an assay buffer with various concentrations of DIF-1(F), W-7(E), or calmidazolium (CZM; Œ) in the absence of CaM (Qiagen). The purified PCR product was introduced into the plasmid pCR (Ϫ CaM); and the samples were assayed for enzymatic activity. Values are the II-TOPO vector (Invitrogen, Carlsbad, CA) and was verified by DNA sequenc- mean Ϯ one-half of the range of two independent experiments (n ϭ 2). C, PDE1 (5 ing. The cDNA for mouse PDE8A was subcloned into the BamHI/XbaI sites mU/tube) was incubated in an assay buffer with 5 ␮M of DIF-1, DIF-3, 2-methoxy isomer of the pVL1393. Transfer of the cDNA from pVL1393 to the Autographa of DIF-1 [2-MIDIF-1 (2-MID-1)], or 6-methoxy isomer of DIF-3 [6-MIDIF-3 (6-MID-3)], or with 50 ␮M W-7, 10 ␮M CZM, or 1% ethanol (Vehicle) in the presence (ϩ) or absence California nuclear polyhedrosis virus was accomplished using the BaculoGold (-) of CaM (10 units/tube) for 10 min; and the samples were assayed for enzymatic ,ء .(Transfection kit (PharMingen, San Diego, CA). High-titer recombinant viral activity. Values are the mean Ϯ SD of three independent experiments (n ϭ 3 P Ͻ 0.005. D, PDE1 (10 mU/tube) was incubated in an assay buffer with ,ءء ;stocks encoding PDE8A were obtained and were used for the subsequent P Ͻ 0.001 ␮ ␮ infection of Sf9 insect cells (PharMingen). 5 M of DIF-1, DIF-3, 2-MIDIF-1 (2-MID-1), or 6-MIDIF-3 (6-MID-3), or with 50 M W-7, 10 ␮M CZM, or 1% ethanol (Vehicle) in the absence of CaM (Ϫ CaM) for 10 min; The membrane-integrated (full-length) and soluble (truncated) forms of and the samples were assayed for enzymatic activity. Values are the mean Ϯ SD of three .P Ͻ 0.05 ,ءء ;P Ͻ 0.001 ,ء .(PDE3A and PDE3B and the soluble (full-length) PDE8A were expressed in independent experiments (n ϭ 3 2569

[cAMP]i (Fig. 2A) in a dose-dependent manner (Fig. 2B), inducing growth arrest (Fig. 2C; Refs. 4, 8). Although the physiological sig-

nificance of the slight increase in [cAMP]i was not known, this finding seemed to provide a hint for the identification of the pharmacological target(s) of DIF-1 in mammalian cells; it was likely that DIF-1 might affect cAMP production and/or degradation. Moreover, because some inhibitors for CaM and cyclic nucleotide PDEs showed antiproliferative activity in K562 cells (Fig. 2C), we examined the effects of DIF-1 on some PDEs and CaM-dependent enzymes in vitro (Figs. 3–6). It was then found that DIF-1 at pharmacological concen- Fig. 4. Kinetic analysis of differentiation-inducing factor-1 (DIF-1)-induced inhibition trations inhibited the activity of CaM-dependent cyclic nucleotide of phosphodiesterase (PDE1) activity. PDE1 and cyclic AMP (cAMP) were incubated in PDE1 in a dose-dependent manner (Fig. 3A). an assay buffer with various concentrations (0.5–5 units) of calmodulin (CaM)inthe To assess whether DIF-1 binds to PDE1 or CaM, we examined the presence or absence of DIF-1 (A), or PDE1 and CaM were incubated with various concentrations (0.2–1 ␮M) of cAMP in the presence or absence of DIF-1 (B), and the effect of DIF-1 on PDE1 in the absence of CaM (Fig. 3B). W-7 did not samples were assayed for enzymatic activity. The data were plotted as 1/v versus 1/[CaM] inhibit PDE1 activity in the absence of CaM, whereas CZM, which (A) or 1/[cAMP] (B), and the fittest lines were determined by linear regression analysis. Each result is a representative of three independent experiments. The results indicate that can directly bind not only to CaM but also to PDE1 (25), inhibited DIF-1 is a noncompetitive inhibitor toward CaM and a competitive inhibitor toward PDE1 activity (Fig. 3B). In the absence of CaM, DIF-1 at the phar- cAMP. macological concentration range inhibited PDE1 activity dose-dependently (Fig. 3B), which suggested that PDE1 but not CaM should be the target of DIF-1. A comparative study using DIF-analogs showed that DIF-1 and DIF-3 (DIFs) were potent inhibitors for PDE1 both in the presence (Fig. 3C) and in the absence (Fig. 3D) of CaM, whereas the effects of 2-MIDIF-1 and 6-MIDIF-3 were weaker than those of DIFs (Fig. 3, C and D), suggesting that PDE1 inhibition by DIFs should be chemical structure-specific. Quite interestingly, kinetic analyses (Fig. 4) revealed that DIF-1 should act as a competitive

inhibitor of PDE1 toward cAMP, and the Ki value was calculated to be 4.5–5 ␮M. We then examined the effects of DIF-1 on some other cyclic nucleotide PDEs (Fig. 5). DIF-1 at up to 50 ␮M did not affect the activity of the cyclic-GMP-inhibited cAMP-specific PDEs, PDE3A, and PDE3B (Ref. 26; Fig. 5A). However, because the original forms of PDE3A and PDE3B are membrane-integrated proteins and the crude cell lysates in the assay mixture might have interfered with the action of DIF-1, which is a lipophilic substance, we examined the

Fig. 5. Effect of differentiation-inducing factor-1 (DIF-1) on the activity of PDE3A, PDE3B, and PDE8A. A, Sf9 cell lysates containing the original form (membrane-integrated form) of human PDE3A or rat PDE3B were incubated in an assay buffer in the presence of DIF-1 or cilostamide, and phosphodiesterase (PDE) activity was assessed. Values are the mean Ϯ one-half of the range of two independent experiments (n ϭ 2). B, cell lysates containing a soluble form of PDE3A or PDE3B were centrifuged, and the enzymatic activity in the supernatants was assessed in the presence of DIF-1 or cilostamide (PDE3 inhibitor). Values are the mean Ϯ SD of three independent experiments (n ϭ 3). C, cell lysates containing the original form (soluble form) of mouse PDE8A were centrifuged, and the enzymatic activity in the supernatant was assessed in the presence of DIF-1 or dipyridamole (PDE8 inhibitor). Values are the mean Ϯ SD of three independent experiments (n ϭ 3).

[0.4 M glycine-NaOH (pH 10.5), 1 mM MgCl2, 0.1 mM ZnCl2,and6mM p-nitrophenylphosphate] in the presence of various concentrations of DIF-1, Fig. 6. Effect of differentiation-inducing factor-1 (DIF-1) on calcineurin and myosin light-chain kinase (MLCK). A, bovine calcineurin (CN) and calmodulin were incubated in and AP activity was assessed as described previously (23). an assay buffer with various concentrations of DIF-1 or calmidazolium (CZM) and assayed for CN activity. Values are the mean Ϯ SD of three independent experiments RESULTS (n ϭ 3). B, chicken gizzard myosin, MLCK, and CaM were incubated in an assay buffer in the presence or absence of 10–20 ␮M DIF-1 or 10 ␮M CZM, and the samples were analyzed by urea-glycerol PAGE. Arrows, the positions of myosin light chain (Myosin Lc) We first examined whether DIF-1 would affect [cAMP]i in K562 and phosphorylated myosin light chain (P-Myosin Lc). A representative result of three leukemia cells (Fig. 2). DIF-1 at 10–40 ␮M slightly increased independent experiments is shown. 2570

Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 2004 American Association for Cancer Research. DIF-1 INHIBITS PDE1 BY COMPETING WITH cAMP effects of DIF-1 on soluble (truncated) forms of PDE3A and PDE3B 2. Kay RR, Berks M, Traynor D. Morphogen hunting in Dictyostelium discoideum. in the absence of the debris of cell membranes (Fig. 5B). Again, DIF-1 Development (Camb) 1989; Suppl :81–90. 3. Morris HR, Masento MS, Taylor GW, Jermyn KA, Kay RR. Structure elucidation of did not affect the activity (Fig. 5B). On the other hand, DIF-1 at 50 ␮M two differentiation inducing factors (DIF-2 and DIF-3) from the cellular slime mould reduced the activity of PDE8A, a cAMP-specific PDE (27), by ϳ40% Dictyostelium discoideum. Biochem J 1988;249:903–6. 4. Asahi K, Sakurai A, Takahashi N, Kubohara Y, Okamoto K, Tanaka Y. DIF-1, (Fig. 5C), but the effect of DIF-1 on this enzyme was much weaker morphogen of Dictyostelium discoideum, induces the erythroid differentiation in than that on PDE1 (Fig. 3A). Thus, PDE8A inhibition by DIF-1 should murine and human leukemia cells. Biochem Biophys Res Commun be relatively nonspecific. 1995;208:1036–9. 5. Kubohara Y, Saito Y, Tatemoto K. Differentiation-inducing factor of D. discoideum We further examined the effects of DIF-1 on two other CaM- raises intracellular calcium concentration and suppresses cell growth in rat pancreatic dependent enzymes (calcineurin and myosin light-chain kinase), AR42J cells. FEBS Lett 1995;359:119–22. 6. Kubohara Y, Kimura C, Tatemoto K. Putative morphogen, DIF, of Dictyostelium svPDE, and AP. DIF-1 at up to 10 ␮M did not inhibit calcineurin discoideum induces apoptosis in rat pancreatic AR42J cells. Dev Growth Differ activity, and DIF-1 at 30 ␮M reduced the activity by ϳ20% (Fig. 6A), 1995;37:711–6. whereas myosin light-chain kinase activity was not affected by DIF-1 7. Kubohara Y. DIF-1, putative morphogen of D. discoideum, suppresses cell growth and promotes retinoic acid-induced cell differentiation in HL-60. Biochem Biophys ␮M ␮M at 20 (Fig. 6B). It was also found that DIF-1 at up to 30 did Res Commun 1997;236:418–22. not affect svPDE or AP (data not shown). All of the results strongly 8. Kubohara Y. Effects of differentiation-inducing factors (DIFs) of Dictyostelium suggest that PDE1 is a pharmacological and specific target of DIF-1. discoideum on the human leukemia K562 cells: DIF-3 is the most potent anti- leukemic agent. Eur J Pharmacol 1999;381:57–62. 9. Kubohara Y, Hosaka K. The putative morphogen, DIF-1, of Dictyostelium discoi- DISCUSSION deum activates Akt/PKB in human leukemia K562 cells. Biochem Biophys Res Commun. 1999;263:790–6. As already mentioned, the target molecule(s) of DIF-1 had not been 10. Miwa Y, Sasaguri T, Kosaka C, et al. DIF-1, a morphogen of Dictyostelium, induces G1 arrest and differentiation of vascular smooth muscle cells. Circ Res 2000;86:68– identified in either mammals or the original organism, Dictyostelium, 75. despite efforts by researchers (28). Here, we have shown for the first 11. Kanai M, Konda Y, Nakajima T, et al. Differentiation-inducing factor-1 (DIF-1) inhibits STAT3 activity involved in gastric cancer cell proliferation via MEK-ERK time that PDE1 is a pharmacological and specific target of DIF-1 and dependent pathway. Oncogene 2003;22:548–54. DIF-3 (designated DIFs) in mammalian cells. It should be of impor- 12. Takahashi-Yanaga F, Taba Y, Miwa Y, et al. Dictyostelium differentiation-inducing tance to note here that some inhibitors for PDE1 are expected to have factor-3 activates glycogen synthase kinase-3␤ and degrades cyclin D1 in mammalian cells. J Biol Chem 2003;278:9663–70. therapeutic potential in the treatment of cancer (29, 30), which agrees 13. Kubohara Y, Hanaoka Y, Akaishi E, Kobayashi H, Maeda M, Hosaka K. DIF-1, an well with the fact that DIFs are potent antitumor agents (4–11). anti-tumor agent found Dictyostelium discoideum, inhibits progesterone-induced oo- The inhibitory effect of CZM on PDE1 is slightly weaker than that cyte maturation in Xenopus laevis. Eur. J Pharmacol 2003;460:93–8. 14. Masento MS, Morris HR, Taylor GW, Johnson SJ, Skapski AC, Kay RR. Differen- of DIF-1 (Fig. 3B), whereas CZM is the most potent inhibitor of cell tiation-inducing factor from the slime mould Dictyostelium discoideum and its ana- growth among the drugs tested in Fig. 2C, probably because CZM is logues. Biochem J 1988;256:23–8. a broad inhibitor for CaM-dependent enzymes. 3-isobutyl-1-methyl- 15. Ebashi S. A simple method of preparing actin-free myosin from smooth muscle. J. Biochem. (Tokyo) 1976;79:229–31. xanthine (IBMX) and 8-methoxymethyl-3-isobutyl-1-methylxanthine 16. Adelstein R, Klee C. Purification and characterization of smooth muscle myosin light (8-MIBMX) should be good inhibitors for PDEs and PDE1, respec- chain kinase. J Biol Chem 1981;256:7501–9. 17. Hattori M, Ozawa K, Wakabayashi K. Sialic acid moiety is responsible for the charge tively, but with the inhibitors at relatively higher concentrations, cell heterogeneity and the biological potency of rat lutropin. Biochem Biophys Res growth can be suppressed (Fig. 2C). The therapeutic advantage, if Commun 1985;127:501–8. any, of DIFs as compared with those of other PDE inhibitors is not 18. Murata T, Taira M, Manganiello VC. Differential expression of cGMP-inhibited cyclic nucleotide phosphodiesterases in human hepatoma cell lines. FEBS Lett known at present. However, in general, because membrane-permeable 1996;390:29–33. and more specific inhibitors of PDE1 should be good tools for both 19. Kenan Y, Murata T, Shakur Y, Degerman E, Manganiello VC. Functions of the experimental and therapeutic uses, DIFs and/or their derivatives may N-terminal region of cyclic nucleotide phosphodiesterase 3 (PDE 3) isoforms. J Biol Chem 2000;275:12331–8. be such drugs. 20. Miki T, Taira M, Hockman S, et al. Characterization of the cDNA and gene encoding Although how much DIFs can penetrate the membrane and inhibit human PDE3B, the cGIP1 isoform of the human cyclic GMP-inhibited cyclic nucleotide phosphodiesterase family. Genomics 1996;36:476–85. intracellular PDE1 is unknown, the concentration range of DIFs for 21. Taira M, Hockman SC, Calvo JC, Taira M, Belfrage P, Manganiello VC. Molecular growth inhibition (Fig. 2C) roughly corresponds to the concentration cloning of the rat adipocyte hormone-sensitive cyclic GMP-inhibited cyclic nucleo- range of DIFs for PDE1 inhibition in vitro (Fig. 3, A–D). However, tide phosphodiesterase. J Biol Chem 1993;268:18573–9. 22. Leroy MJ, Degerman E, Taira M, et al. Characterization of two recombinant PDE3 because the correlation between growth inhibition (Fig. 2C) and PDE1 (cGMP-inhibited cyclic nucleotide phosphodiesterase) isoforms, RcGIP1 and Hc- inhibition (Fig. 3, C and D) by the DIF-analogs was not perfect, it is GIP2, expressed in NIH 3006 murine fibroblasts and Sf9 insect cells. Biochemistry still possible that there may be some other target(s) of DIFs. 1996;35:10194–202. 23. Takahashi K, Akaishi E, Abe Y, et al. Zinc inhibits calcineurin activity in vitro by As a next step, we will need to further verify whether the known competing with nickel. Biochem Biophys Res Commun 2003;307:64–8. actions of DIFs in mammalian cells are exerted via PDE1 inhibition 24. Nakamura S, Nonomura Y. A simple and rapid method to remove light chain by DIFs and whether DIFs affect some other PDEs and/or enzymes phosphatase from chicken gizzard myosin. J Biochem (Tokyo) 1984;96:575–8. 25. Gietzen K, Sadorf I, Bader H. A model for the regulation of the calmodulin-dependent relating to cAMP. In addition, it should be assessed in the future enzymes erythrocyte Ca2ϩ-transport ATPase and brain phosphodiesterase by activa- whether this is the case in Dictyostelium. At any rate, the present tors and inhibitors. Biochem J 1982;207:541–8. 26. Degerman E, Belfrage P, Manganiello VC. Structure, localization, and regulation of information that DIF-1 directly inhibits PDE1 in vitro by competing cGMP-inhibited phosphodiesterase (PDE3). J Biol Chem 1997;272:6823–6. with cAMP should provide a new insight into, and a hint for the 27. Soderling SH, Bayuga SJ, Beavo JA. Cloning and characterization of a cAMP- identification of, some other target(s), if any, of DIFs. specific cyclic nucleotide phosphodiesterase. Proc Natl Acad Sci USA 1998;95: 8991–6. 28. Insall R, and Kay RR. A specific DIF binding protein in Dictyostelium. EMBO J REFERENCES 1990;9:3323–8. 29. Kakkar R, Raju RV, Sharma RK. Calmodulin-dependent cyclic nucleotide phospho- 1. Morris HR, Taylor GW, Masento MS, Jermyn KA, Kay RR. Chemical structure of the diesterase (PDE1). Cell Mol Life Sci 1999;55:1164–86. morphogen differentiation inducing factor from Dictyostelium discoideum. Nature 30. Lerner A, Kim DH, Lee R. The cAMP signaling pathway as a therapeutic target in (Lond.) 1987;328:811–4. lymphoid malignancies. Leuk Lymphoma 2000;37:39–51.

2571

Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 2004 American Association for Cancer Research. Calmodulin-Dependent Cyclic Nucleotide Phosphodiesterase (PDE1) Is a Pharmacological Target of Differentiation-Inducing Factor-1, an Antitumor Agent Isolated from Dictyostelium

Kasumi Shimizu, Taku Murata, Toshiro Tagawa, et al.

Cancer Res 2004;64:2568-2571.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/64/7/2568

Cited articles This article cites 30 articles, 10 of which you can access for free at: http://cancerres.aacrjournals.org/content/64/7/2568.full#ref-list-1

Citing articles This article has been cited by 4 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/64/7/2568.full#related-urls

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/64/7/2568. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.