Biochemical and cellular effects of roscovitine, a potent and selective inhibitor of the cyclin-dependent kinases cdc2, cdk2 and cdk5.

Cyclin-dependent kinases (cdk) play an essential role in the intracellular control of the cell division cycle (cdc). These kinases and their regulators are frequently deregulated in human tumours. Enzymatic screening has recently led to the discovery of specific inhibitors of cyclin-dependent kinases, such as butyrolactone I, flavopiridol and the purine olomoucine. Among a series of C2, N6, N9-substituted adenines tested on purified cdc2/cyclin B, 2-(1-ethyl-2-hydroxyethylamino)-6-benzylamino-9-isopropylpurine (roscovitine) displays high efficiency and high selectivity towards some cyclin-dependent kinases. The kinase specificity of roscovitine was investigated with 25 highly purified kinases (including protein kinase A, G and C isoforms, myosin light-chain kinase, casein kinase 2, insulin receptor tyrosine kinase, c-src, v-abl). Most kinases are not significantly inhibited by roscovitine. cdc2/cyclin B, cdk2/cyclin A, cdk2/cyclin E and cdk5/p35 only are substantially inhibited (IC50 values of 0.65, 0.7, 0.7 and 0.2 microM, respectively). cdk4/cyclin D1 and cdk6/cyclin D2 are very poorly inhibited by roscovitine (IC50 > 100 microM). Extracellular regulated kinases erk1 and erk2 are inhibited with an IC50 of 34 microM and 14 microM, respectively. Roscovitine reversibly arrests starfish oocytes and sea urchin embryos in late prophase. Roscovitine inhibits in vitro M-phase-promoting factor activity and in vitro DNA synthesis in Xenopus egg extracts. It blocks progesterone-induced oocyte maturation of Xenopus oocytes and in vivo phosphorylation of the elongation factor eEF-1. Roscovitine inhibits the proliferation of mammalian cell lines with an average IC50 of 16 microM. In the presence of roscovitine L1210 cells arrest in G1 and accumulate in G2. In vivo phosphorylation of vimentin on Ser55 by cdc2/cyclin B is inhibited by roscovitine. Through its unique selectivity for some cyclin-dependent kinases, roscovitine provides a useful antimitotic reagent for cell cycle studies and may prove interesting to control cells with deregulated cdc2, cdk2 or cdk5 kinase activities.

Pertussis toxin facilitates the progesterone-induced maturation of Xenopus oocyte. Possible role of protein phosphorylation.

Progesterone triggers the first meiotic cell division of Xenopus oocyte and inhibits cAMP synthesis. The effect of pertussis toxin purified from Bordetella pertussis was tested on the maturation of Xenopus oocyte. The toxin did not inhibit progesterone-induced resumption of meiosis or the hormone-induced drop in cAMP level. This indicates that progesterone action is not mediated by the Ni subunit of the oocyte adenylate cyclase. Furthermore, pertussis toxin caused a reduction in the time course of maturation correlated with the precocious appearance of an alkali stable 47 kDa phosphoprotein, a marker of the maturation promoting factor (MPF) activity. Pertussis toxin effects mimicked those of 2-glycerophosphate suggesting that both agents act on the steady-state level of phosphorylation implicated in MPF activity.

In vivo effects of microinjected alkaline phosphatase and its low molecular weight substrates on the first meiotic cell division in Xenopus laevis oocytes.

Xenopus laevis oocytes were microinjected with low molecular weight phosphoesters such as 2-glycerophosphate, phosphotyrosine, phosphoserine, phosphothreonine, 4-nitrophenyl phosphate, and orthophosphate. These compounds were able to induce a considerable reduction in the time course of progesterone-induced maturation, with 2-glycerophosphate being the most effective. The basal level of cAMP and its drop during maturation were not affected by the microinjection of 2-glycerophosphate. The injection of alkaline phosphatase (EC 3.1.3.1.) from calf intestine at a low concentration (10 ng per oocyte) was able to decrease or abolish the effect of 2-glycerophosphate. At higher concentration (25 ng per oocyte) this enzyme totally blocked progesterone- or maturation-promoting factor-induced maturation. Alkaline phosphatase might behave in vivo as a phosphoprotein phosphatase active towards phosphotyrosine-containing proteins. In addition, our results indicate that phosphate or phosphoester-containing buffers should be avoided in the course of maturation-promoting factor purification.

[A 45 kDa phosphorylated protein, resistant to alkaline treatment, appears at the time of rupture of the nuclear envelope during the 1st meiotic division of the Xenopus oocyte]. / Une protéine phosphorylée de 45 kDa, résistante à un traitement alcalin, apparaît à l'époque de la rupture de l'enveloppe nucléaire au cours de la première division méïotique de l'ovocyte de Xénope.

Xenopus oocytes were prelabeled with 32PO4 and induced to mature by progesterone treatment (1 microM). At the time of the breakdown of the germinal vesicle (nucleus), an alkali stable 45 kDa phosphoprotein appears in the 165 000 X g oocyte supernatant. Phospho-amino acid analysis shows that the 45 kDa protein is phosphorylated at threonine residues.

cAMP-dependent protein kinase regulates in ovo cAMP level of the Xenopus oocyte: evidence for an intracellular feedback mechanism.

Microinjection of cAMP-dependent protein kinase inhibitor (1.8 microM) increases the cAMP level of Xenopus oocyte. Its effect was observed in full-grown (stage VI) as well as in vitellogenic (stage IV) oocytes. In contrast the inhibitor I1 of protein phosphatase-1 blocks cAMP accumulation. Progesterone (1 microM) decreases the cAMP level in control and in PKI-treated oocytes of both stages. These results show that cAMP concentration is regulated by a cAMP-dependent phosphorylation indicating the presence of a feedback mechanism. The feedback control is disrupted when oocyte is induced to mature by progesterone.

Calmodulin modulates the cyclic AMP level in Xenopus oocyte.

Progesterone decreases the cAMP level of Xenopus oocytes which had been pretreated with cholera toxin (6 nM) and IBMX (1 mM); its action is obtained either by exposure to external hormone (1 micro M) or by microinjection of 50 nl of a 1 mM progesterone solution in paraffin oil. The cAMP content can be decreased in hormone-free oocytes by the calcium ionophore A 23187 or by microinjection of calcium-calmodulin. Conversely when endogenous calcium-calmodulin is inhibited by microinjection of either anticalmodulin antibodies or fluphenazine the cAMP content is increased. In all experimental conditions (high or low levels of intracellular calmodulin), progesterone is always capable of decreasing the oocyte cAMP concentration. Our results favor the view that the cAMP content is negatively controlled, probably via an inhibition of the adenylate cyclase activity, by two parallel mechanisms: the first involves calmodulin, the second results in an action of progesterone which does not require the intermediary formation of the calcium-calmodulin complex.

Microinjected progesterone reinitiates meiotic maturation of Xenopus laevis oocytes.

Microinjection of progesterone dissolved in paraffin oil induces the reinitiation of meiotic maturation in the Xenopus oocyte; 50% maturation is obtained when 50 nl of a 50 microM solution is microinjected into the oocyte. The kinetics of the response to microinjected progesterone are similar to the kinetics of response to externally applied hormone. When an aqueous solution of progesterone is microinjected instead of an oil solution, maturation is never observed, a result which confirms previous work. Leakage of the steroid into the external medium was estimated to range from 1.6 pmol/hr when microinjection was performed in oil to 3.6 pmol/hr when it was performed in aqueous solution. Metabolism of the hormone microinjected in oil is weak (less than 20%) as compared to that after aqueous microinjection (greater than 80%). Progesterone microinjected in oil decreases the cAMP content as does externally applied hormone. We therefore conclude that progesterone acts initially on an intracellular site in order to trigger meiotic maturation of the Xenopus oocyte.

In vitro effects of progesterone and estradiol-17 beta on choleragen activated Xenopus oocyte adenylate cyclase.

The basal cAMP levels of full-grown defoliculated Xenopus oocytes were shown to average 1.2 pmol per oocyte. Follicle cAMP concentration was not significantly different from that of denuded oocyte and was found, when expressed as unit of volume to be independent of follicle size. In vitro gonadotropin treatment of follicle does not affect cAMP accumulation. During the course of progesterone induced maturation we failed to detect any modification of the oocyte cAMP content. Cholera toxin treatment increased cAMP levels; estradiol-17 beta potentialized cholera toxin action and progesterone antagonized it. Estradiol-17 beta was also found to reduce the amount of cholera toxin necessary to inhibit 50% of the progesterone induced meiosis maturation.

Cyclic AMP phosphodiesterase activities in Xenopus laevis oocytes.

Cyclic AMP phosphodiesterase activity has been identified in full-grown Xenopus oocytes in vivo and in vitro. About 50% of the in vitro phosphodiesterase activity was present in the solution fraction and 35% in a partially purified membrane fraction. Both activities exhibited high substrate affinity (Km about 10(-6) M). Sucrose gradient fractionation revealed two forms of phosphodiesterase: a 5 S form (peak I) and a 6.5 S form (peak II). Treatment with trypsin led to the activation of the soluble enzyme with the transformation of peak II into peak I. Ethylene glycol bis (beta-aminoethyl ether)-N,N'-tetraacetic acid, calcium dependent regulator, and Fluphenazine did not influence the enzyme activities suggesting that the oocyte phosphodiesterases were not Ca2+-dependent. Intact oocytes were induced to mature by exposure to progesterone; their phosphodiesterase activities and distribution tested in vitro were comparable to those of untreated oocytes.

Cyclic AMP synthesis in Xenopus laevis oocytes: inhibition by progesterone.

[alpha-32P]ATP was microinjected into Xenopus oocyte and neosynthesized cyclic AMP was isolated. Cholera toxin inhibited progesterone-induced maturation and stimulated after 3 h of preincubation the amount of neosynthesized cyclic AMP. Progesterone decreased the neosynthesis of cyclic AMP during the first hour following addition of the hormone.