The Plant Cell, Vol. 13, 585–598, March 2001, www.plantcell.org © 2001 American Society of Plant Physiologists

Cell Cycle in the Fucus Zygote Parallels a Somatic Cell Cycle but Displays a Unique Translational Regulation ofCyclin-Dependent Kinases

Florence Corellou,

a

Colin Brownlee,

b

Lenaick Detivaud,

c

Bernard Kloareg,

a

and François-Yves Bouget

a,1

a

Unité Mixte de Recherche 1931 (Centre National de la Recherche Scientifique and Laboratoires Goëmar), Station Biologique, 29680 Roscoff, France

b

Marine Biological Association, The Laboratory, Citadel Hill, Plymouth PL1 2PB, United Kingdom

c

Cell Cycle Group, Station Biologique, 29680 Roscoff, France

In eukaryotic cells, the basic machinery of cell cycle control is highly conserved. In particular, many cellular events dur-ing cell cycle progression are controlled by cyclin-dependent kinases (CDKs). The cell cycle in animal early embryos,however, differs substantially from that of somatic cells or yeasts. For example, cell cycle checkpoints that ensure thatthe sequence of cell cycle events is correct have been described in somatic cells and yeasts but are largely absent inembryonic cells. Furthermore, the regulation of CDKs is substantially different in the embryonic and somatic cells. Inthis study, we address the nature of the first cell cycle in the brown alga Fucus, which is evolutionarily distant from themodel systems classically used for cell cycle studies in embryos. This cycle consists of well-defined G1, S, G2, and Mphases. The purine derivative olomoucine inhibited CDKs activity in vivo and in vitro and induced different cell cycle ar-rests, including at the G1/S transition, suggesting that, as in somatic cells, CDKs tightly control cell cycle progression.

The cell cycle of Fucus zygotes presented the other main features of a somatic cell cycle, such as a functional spindle as-sembly checkpoint that targets CDKs and the regulation of the early synthesis of two PSTAIRE CDKs, p32 and p34, andthe associated histone H1 kinase activity as well as the regulation of CDKs by tyrosine phosphorylation. Surprisingly, thesynthesis after fertilization of p32 and p34 was translationally regulated, a regulation not described previously for CDKs.Finally, our results suggest that the activation of mitotic CDKs relies on an autocatalytic amplification mechanism.

INTRODUCTION

During animal early embryogenesis, the fertilized egg under-goes rapid cell cycles without growing. Interphases (Sphase) and mitoses (M phase) alternate in quick successionwithout intervening G1 and G2 phases (King et al., 1994).The oscillation between S and M phases is driven by suc-cessive activation and inactivation of a key protein complexcalled the M phase–promoting factor (MPF) (Masui andMarkert, 1971; Kishimoto, 1988). MPF is composed of cdc2,a cyclin-dependent kinase (CDK), and its positively regulat-ing protein, cyclin B. Periodic synthesis and degradation ofcyclin B is the major mechanism of CDK regulation in em-bryos (Arion et al., 1988; Dunphy and Newport, 1988). Incontrast, additional mechanisms, such as phosphorylation,regulate CDK activity in somatic cells. In both embryos andsomatic cells, mitotic CDKs operate at the G2/M boundary,where their activation triggers, through the phosphorylation

of their substrate proteins, several major cellular events, in-cluding nuclear envelope breakdown, chromosome conden-sation, spindle assembly, and cytokinesis (Peter et al., 1990;Kishimoto, 1994).

In somatic cells and yeasts, however, the cell cycle ap-pears to be more tightly controlled by CDKs than in em-bryos (Nurse, 1994; Chevalier and Blow, 1996). A restrictionpoint controls cell cycle progression in G1 in response toextracellular and/or intracellular cues (Peter and Herskowitz,1994; Planas-Silva and Weinberg, 1997; Neufeld and Edgar,1998), and CDK activity promotes the assembly of replica-tion complexes in S phase. In contrast, in the sea urchin zy-gote, DNA replication can occur while CDKs are maintainedinactive (Moreau et al., 1998), suggesting that CDKs maynot control all aspects of cell cycle progression in animalembryos.

The differences between cell cycle control in somatic andembryonic cells are reflected by different mechanisms ofregulation of CDKs. Whereas the reentry of quiescent so-matic cells into the cell cycle in G1 is mediated by the suc-cessive expression of G1 and mitotic CDKs (Elledge et al.,

1

To whom correspondence should be addressed. E-mail bouget@sb-roscoff.fr; fax 33-2-98-29-23-24.

586 The Plant Cell

1992), constant levels of CDKs are detected in animal earlyembryos. Finally, cell cycle checkpoints of somatic cells dif-fer greatly from those of embryonic cells. Somatic cells dis-play stringent surveillance mechanisms that check thesuccess of various cell cycle events such as DNA replicationand the integrity of the spindle, preventing mitosis until ac-curate repairs have been performed (Rudner and Murray,1996; Paulovich et al., 1997). CDKs are a major target ofthese checkpoints, and their regulation after checkpoint ac-tivation is usually correlated with changes in their phosphor-ylation status as well as with associations with regulatorymolecules such as CDK inhibitors (Lew and Kornbluth, 1996;Rudner and Murray, 1996; Hardwick, 1998). These check-points are usually partial in or absent from early embryoniccells (reviewed in Hartwell and Weinert, 1989; Heichmanand Roberts, 1994; Nurse, 1994; Edgar, 1995).

Although much less is known about the in vivo regulationof the cell cycle in plants, evidence strongly suggests thatthe cell cycle machinery is well conserved between plantsand animals (Mironov et al., 1999). In particular, functionalhomologs of the yeast

cdc2

gene, which contains the fullyconserved PSTAIRE hallmark, have been identified in sev-eral plant species, and these are required for progressionthrough mitosis (reviewed in Dudits et al., 1998). The lack ofcell cycle mutants in plants has been partially overcome byusing a family of specific inhibitors of CDKs, such as olo-moucine, that specifically and reversibly inhibit cdk2 andcdc2 in many cell types (Vesely et al., 1994; Alessi et al.,1998). Olomoucine and other purine derivatives were foundto arrest cell cycle progression at both the G1/S and G2/Mtransitions in Petunia and Arabidopsis somatic cells (Glab etal., 1994; Planchais et al., 1997). Finally, tyrosine phosphoryla-tion has been proposed to regulate CDK activity in responseto plant hormones in somatic cells (Bell et al., 1993; Zhanget al., 1996; McKibbin et al., 1998).

Microspectrophotometric analysis of the DNA content insperm nuclei has suggested specificities of the cell cycle inhigher plant gametes, such as the initiation of DNA replica-tion before gametic fusion (Friedman, 1999). Furthermore, inin vitro–fertilized zygotes from maize, the transcripts of threecyclin genes are detected only after fertilization, suggestingthat there may be different patterns of cell cycle activity inhigher plant embryos (Sauter et al., 1998). However, higherplant embryos are embedded deeply in maternal tissue andare largely inaccessible to experimentation, and, to ourknowledge, in vitro fertilization systems do no provideenough material for biochemical approaches. Consequently,little is known about the molecular mechanisms of cell cycleregulation in these embryos.

In fucoid algae, including the genera

Fucus

and

Pelvetia

,fertilization is external, and large populations of synchro-nously developing zygotes can be obtained (Brownlee andBouget, 1998). This characteristic, coupled with the lengthof the first cell cycle (24 hr), makes fucoid algae powerfulexperimental systems for cellular and biochemical inves-tigations of cell cycle regulation. In a previous study, we

demonstrated the presence in fucoid zygotes of a DNA rep-lication checkpoint, a common feature of somatic cells butalso of a few animal embryos (Corellou et al., 2000a). To de-termine more precisely the nature of the first cell cycle in Fu-cus (embryonic versus somatic), we investigated theregulation of CDKs and their involvement in controlling thefirst embryonic cell cycle of Fucus zygotes during normaldevelopment and after various cell cycle arrests. Our resultsindicate that in Fucus zygotes, many events of the cell cycleand cell cycle arrest are tightly controlled by CDKs, whichthemselves appear to be regulated, as in somatic cells, bothat the level of synthesis and by reversible phosphorylation.The presence of a functional spindle assembly checkpointfurther confirmed that the cell cycle in Fucus is somatic innature. Interestingly, CDKs appear to be synthesized frommaternal mRNAs, suggesting a unique regulation of CDKsat the translational level. Our results also suggest the pres-ence of a mechanism of autocatalytic amplification ofmitotic cyclin/CDK activity similar to those described previ-ously in animals and yeasts. We propose a model of thepossible roles and regulations of CDKs throughout the firstcell cycle of fucoid algae.

RESULTS

We studied three different aspects of the Fucus first cell cy-cle to determine more precisely its nature (embryonic versussomatic): (1) the requirement for CDKs for cell cycle pro-gression, particularly at the G1/S transition; (2) the regula-tion of CDKs; and (3) the presence of a spindle assemblycheckpoint.

Olomoucine Induces Different Cell Cycle Arrests in Fucus Zygotes

Previous results have shown that in Fucus zygotes, the pu-rine derivative olomoucine specifically inhibits mitotic CDKsin vitro and blocks cell cycle progression from G2 to mitosis(Corellou et al., 2000a). Zygotes treated with olomoucinefrom 6 hr after fertilization (AF) display a decondensed nu-cleus and a spindle with an aligned centrosomal axis that istypical of zygotes in premitosis (Corellou et al., 2000a). Tak-ing advantage of this specific inhibitor of CDKs, we furtherinvestigated the involvement of CDKs in the control of theFucus first cell cycle. Zygotes were treated with variousconcentrations of olomoucine from 2 hr until 36 hr AF. Con-centrations of

z

35

m

M were required to block cytokinesis inmost embryos (data not shown). At these concentrations,zygotes exhibited condensed chromosomes that were oftenscattered in the cytoplasm (Figure 1B). Between 50 and 100

m

M olomoucine was required to arrest and maintain all zy-gotes with a decondensed nucleus. A concentration of 100

m

M olomoucine therefore was used to block chromatin con-

Cell Cycle Control in Fucus Zygotes 587

Figure 1. Effect of Olomoucine on Cell Cycle Progression in Zygotes.

Olomoucine (100 mM) was added at various times until 36 hr AF. Cells were fixed and subsequently stained with mithramycin A ([A] to [D]).(A) to (D) Zygotes treated with 100 mM olomoucine from 12 (A), 16 (B), 20 (C), and 24 (D) hr until 36 hr AF. Bars 5 25 mm.(E) Zygotes treated with 100 mM olomoucine at various times AF (x axis) displayed either one or two sets of dispersed chromosomes (white dotsand black dots, respectively) or one or two decondensed nuclei (black bars and white bars, respectively). The data shown are representative ofthe results of two independent experiments.(F) RNA gel blot analysis with a Fucus histone H3 DNA probe (left). Equal amounts of RNA (10 mg), extracted at the times indicated, were loadedin each lane, and RNA integrity was checked by staining with ethidium bromide (data not shown). Corresponding quantifications (right) were per-formed with a phosphorimager in control zygotes (open circles) as well as in zygotes treated from 3 hr AF with either 20 mM aphidicolin (closedsquares) or 100 mM olomoucine (closed triangles) until the times indicated on the x axis. The data shown are representative of the results of twoindependent experiments. AF, after fertilization.

588 The Plant Cell

densation (data not shown) and was added at various timesAF. When added in early G2, that is, when in most cells divi-sion was no longer sensitive to the DNA replication inhibitoraphidicolin (Corellou et al., 2000a), 100

m

M olomoucineblocked nuclei before mitosis, that is, at the G2/M transition(Figures 1A and 1E). In contrast, olomoucine treatmentsstarted just before mitosis, at 16 hr AF, allowed

z

40% ofthe chromosomes to condense (Figures 1B and 1E), like thearrest induced by continuous treatment with 35

m

M olomou-cine. Treatment of zygotes with olomoucine at 18 hr AF hadlittle effect on nuclear events of the first cell cycle (Figures1C to 1E), even though cytokinesis was still inhibited (datanot shown). Olomoucine (100

m

M) induced the same suc-cession of arrests during the second cell cycle (Figures 1Cto 1E) but more rapidly, consistent with the faster pace ofthis second cell cycle. These findings indicate that the CDK

inhibitor olomoucine induces various cell cycle arrests, in-cluding one at the G2/M transition and one in mitosis, andthat the targets of olomoucine display different degrees of invivo sensitivity to this drug.

It proved impossible to investigate directly the effect of ol-omoucine on the G1/S transition by quantifying DNA levelsin zygotes stained with mithramycin A or with any other DNAdye (Corellou et al., 2000a). In the one-celled zygote the nu-cleus is decondensed and occupies a central position, andthe cell is large and filled with organelles with high autofluo-rescence, such as plastids and polyphenol-containing phys-odes. These features make it difficult to directly measure thenuclear DNA levels in zygotes; therefore, we used an indi-rect method to monitor the G1/S progression and the effectof cell cycle inhibitors. In most eukaryotic cells, transcriptionof certain histone genes, including

histone H3

, begins at

Figure 2. Regulation of the Expression and Activity of CDKs during Early Development.

Total proteins were extracted from Fucus zygotes at the times indicated, and CDKs were purified by affinity on p9CKShs1–Sepharose beads. Theresults shown are representative of the results of three independent experiments. Purified proteins were assayed for their ability to phosphory-late histone H1 in vitro ([A] and [C]). Eluted proteins were immunoblotted with an anti-PSTAIRE antibody ([B] and [D]).(A) Histone H1 kinase activity during normal development.(B) Synthesis of PSTAIRE CDKs during normal development.(C) Histone H1 kinase activity in extracts from zygotes treated from 1 hr AF with either 16 mM actinomycin D (gray bars) or 0.2 mM cycloheximide(black bars), compared with control cells (white bars).(D) Expression of PSTAIRE CDKs in extracts from zygotes treated from 1 hr AF with either 16 mM actinomycin D or 0.2 mM cycloheximide.

Cell Cycle Control in Fucus Zygotes 589

the onset of S phase (Kapros et al., 1992; Chaubet andGigot, 1998). We monitored

histone H3

transcription as an Sphase–specific marker using RNA gel blot analysis with afragment of the coding region from a Fucus

histone

H3

gene. In Fucus zygotes,

histone H3

transcripts were clearlydetected at 6 hr AF, and they accumulated progressivelyuntil 12 hr AF (Figure 1F). This is unlikely to be due to a gen-eral increase of transcription, because the transcript of

ac-tin

, a constitutively expressed gene, displays constant levelsof transcript throughout the first cell cycle (Bouget et al.,1995). Although the transcription of

histone H3

still occurredin zygotes in which DNA replication was blocked by aphidi-colin (Figure 1F), no

histone H3

transcripts were detected inzygotes incubated with 100

m

M olomoucine from 3 hr AF.These results suggest that in Fucus zygotes, S phase be-gins at 4 to 6 hr AF, and 100

m

M olomoucine induced a cellcycle arrest before the arrest induced in early S phase byaphidicolin, that is, in G1 or at the G1/S transition.

Histone H1 Kinase Activity and the Synthesis of PSTAIRE CDKs Increase Dramatically AF and Are Regulated at the Translational Level, but They Do Not Depend on Karyogamy

CDKs were purified using their affinity for the human suc1homolog p9

CKShs1

, and their histone H1 kinase activity wasmeasured in vitro. The levels of two PSTAIRE CDKs boundto p9

CKShs1

, p32 and p34, were monitored using a mono-clonal anti-PSTAIRE antibody (Corellou et al., 2000a). Ex-tracts from unfertilized eggs exhibited very low levels ofhistone H1 kinase activity (Figure 2A), and both p32 and p34were not detectable or only barely detectable before fertili-zation (Figure 2B). Histone H1 kinase activity increased

z

10-fold between 0 and 2 hr AF (Figure 2A), which corre-lated with a significant increase in the synthesis of both p34and p32 (Figure 2B). This initial increase in histone H1 ki-nase activity was followed by a steady and continuous in-crease (Figure 2A; see also controls in Figures 2C and 6A),and the synthesis of PSTAIRE CDKs increased in a similarmanner (Figure 2B; see also control in Figure 2D). The spe-cific activity of histone H1 kinase relative to total protein incrude zygote extracts gave a similar profile (data notshown), suggesting that these postfertilization changes inhistone H1 kinase activity did not result from a general in-crease in protein synthesis.

The requirements for the synthesis of PSTAIRE CDKs andfor the associated histone H1 kinase activity were investi-gated using actinomycin D and cycloheximide, two inhibi-tors of transcription and translation, respectively, in Fucuszygotes (Quatrano, 1968). When added before 12 hr AF, ac-tinomycin D (16

m

M) and cycloheximide (0.2

m

M) inhibitedcell division (data not shown). When cells were treated withcycloheximide from 1 hr AF, the levels of both histone H1 ki-nase activity and PSTAIRE CDKs remained low for at least26 hr (Figures 2C and 2D). In contrast, and although cell di-

vision was fully inhibited for at least 26 hr, incubations withactinomycin D from 1 hr AF did not affect the synthesis ofPSTAIRE CDKs (Figure 2D), suggesting that the synthesis ofCDKs is translationally regulated. In these conditions, how-ever, histone H1 kinase activities remained at levels lowerthan those of 14-hr-old controls (Figure 2C), suggesting thatthe synthesis of mitotic CDK-activating proteins is transcrip-tionally regulated.

When added at 2 hr AF, the microtubule depolymerizingagent nocodazole prevented the fusion of pronuclei but hadno effect on early morphogenesis (i.e., germination; Figure3A). The male pronucleus remained condensed and alwaysentered mitosis after the female pronucleus, as describedpreviously in other fucoid algae (Brawley and Quatrano,1979; Swope and Kropf, 1993; Motomura, 1995). Similarly,when nocodazole was added after pronuclei fusion, zygoteswere arrested in mitosis with clustered, heavily condensed

Figure 3. Effect of the Inhibition of Pronuclei Fusion by Nocodazoleon the Activity of CDKs.

(A) A zygote treated with 0.33 mM nocodazole from 2 to 36 hr AFand then stained with mithramycin A.(B) A zygote treated with 0.33 mM nocodazole from 12 to 36 hr AFand then stained with mithramycin A.(C) Histone H1 kinase activity in extracts from cells treated with 0.33mM nocodazole from either 2 hr AF (closed circles) or 12 hr AF(closed triangles, arrow) until the times indicated on the x axis. His-tone H1 kinase activity in extracts from control zygotes is repre-sented by open squares. The data shown are representative of theresults of two independent experiments.Bar in (A) 5 20 mm for (A) and (B).

590 The Plant Cell

chromosomes (Figure 3B). The developmental course of his-tone H1 kinase activity was almost identical in zygotestreated with nocodazole before or after pronuclei fusion,and the mitotic activity was much higher than in control zy-gotes (Figure 3C). These results suggest that the synthesisof CDKs and the activation of their kinase activity is initiallytriggered by fertilization independently of karyogamy.

Olomoucine Inhibits Both in Vivo and in Vitro Histone H1 Kinase Activity and Induces an Accumulation of Tyrosine-Phosphorylated CDKs

The role of tyrosine phosphorylation in CDK regulation wasinvestigated after various cell cycle arrests. Histone H1 ki-nase activities were monitored after treatment of 3-hr-oldzygotes with either 100 or 35

m

M olomoucine for 36 hr in zy-gotes that were arrested before S phase and in mitosis, re-spectively. In control zygotes, histone H1 kinase activitydoubled from 4 to 14 hr AF, and the maximal value of themitotic peak of activity corresponded to an eightfold in-crease compared with the activity detected at 4 hr AF. Incontrast, no peak of activity was observed when zygoteswere treated with 100

m

M olomoucine, and the level of H1kinase remained similar to that of 6- to 8-hr-old control zy-gotes even at 36 hr AF (Figure 4A). The histone H1 kinaseactivity from zygotes incubated with 35

m

M olomoucine in-creased rapidly, reaching a level similar to that of control zy-gotes after 14 to 16 hr. This kinase activity remained stableuntil 24 hr AF, increased again between 24 and 36 hr AF (afour- to 20-fold increase depending on the experiment), andfinally remained stable for at least another 24 hr (Figure 4A).Staining with mithramycin A revealed that, after treatment

Figure 4.

Effect of Olomoucine on Histone H1 Kinase Activity andTyrosine Phosphorylation of CDKs.

(A)

Histone H1 kinase activities in extracts from control zygotes

(open squares) and zygotes treated from 3 hr AF with either 100

m

Molomoucine (closed circles) or 35

m

M olomoucine (open circles) untilthe times indicated on the

x

axis. The data shown are representativeof the results of three independent experiments.

(B)

Dose-dependent in vitro inhibition by olomoucine of histone H1kinase activity in extracts from cells treated from 3 to 36 hr AF witheither 35

m

M olomoucine (black bars) or 100

m

M olomoucine (whitebars). For each treatment, the kinase activity is reported as a per-centage of control activity (no olomoucine). The data shown are rep-resentative of the results of two independent experiments.

(C)

Protein gel blot analysis of the expression and phosphorylationof PSTAIRE CDKs after treatment with olomoucine. Immunoblottingwas performed with anti-PSTAIRE (PSTAIRE) or anti-phosphoty-rosine (PY) antibodies. Top two panels, zygotes were treated from 3hr AF with either 100 or 35

m

M olomoucine until the times indicated(data shown are from the same experiment represented in

[A]

). Bot-tom panel, zygotes were treated from 4 or 10 hr AF with 100

m

M olo-moucine or with 0.33

m

M nocodazole (Noco) until 36 hr AF. The datashown are representative of the results of three independent experi-ments.

Cell Cycle Control in Fucus Zygotes 591

with 35

m

M olomoucine, the nuclei were decondensed at 24hr AF, whereas condensed chromosomes were observed at36 hr AF (data not shown).

Because olomoucine is a reversible inhibitor of CDKs, itwas possible to investigate the in vitro effect of olomoucineon histone H1 kinase in cell extracts from zygotes treatedpreviously with olomoucine (Corellou et al., 2000a). In vitrotreatments with olomoucine inhibited histone H1 kinase inextracts from 36-hr-old zygotes arrested before S phasewith 100

m

M olomoucine or in mitosis with 35

m

M olomou-cine (Figure 4B). Even though the initial levels of kinase ac-tivities were dramatically different (Figure 4A), the patterns ofin vitro dose-dependent inhibition were similar in both typesof cell cycle arrests, suggesting that similar CDK/cyclinswere targeted in vivo by olomoucine. The higher kinase ac-tivities observed after mitotic arrest with lower concentra-tions of olomoucine (Figure 4A) therefore may arise from apartial activation of mitotic CDKs (see Discussion).

Immunological detection of PSTAIRE CDKs in olomou-cine-treated zygotes showed that until 12 to 16 hr AF, lev-els of both p34 and p32 increased normally as in controls(data not shown), suggesting that the synthesis of PSTAIRECDKs was not inhibited by olomoucine regardless of theconcentration of the drug (Figure 4C, which corresponds tothe kinase assay shown in Figure 4A). In contrast, the immu-nological detection of phosphotyrosine residues revealedthat tyrosine phosphorylation increased progressively onboth proteins from 8 to 36 hr AF. In control zygotes, tyrosinephosphorylation of these two proteins has been shown tobe cell cycle regulated, being maximal in G2 phase and ab-sent at the time of mitosis (Corellou et al., 2000a). Zygotestreated with 100

m

M olomoucine at 10 hr AF, that is, at theS/G2 transition, were arrested with a decondensed nucleusat the G2/M transition (Figure 1). In these cells, the expres-sion patterns of both PSTAIRE and phosphotyrosine epi-topes were similar to those observed in zygotes treated with100

m

M olomoucine from 3 hr AF, and the 32- and 34-kDproteins remained phosphorylated on tyrosine until at least36 hr AF (Figure 4C). In contrast, no phosphorylation was de-tectable in zygotes arrested in mitosis by nocodazole (Fig-ure 4C, bottom).

Treatments with cdc25A Phosphatase Restore Mitotic Levels of Histone H1 Kinase in Extracts from Cells Arrested by Olomoucine

Various arrests were induced during cell cycle progressionby olomoucine and nocodazole (Figure 5). The earlier thecells were arrested by olomoucine during cell cycle progres-sion, the lower their levels of histone H1 kinase activities(Figure 5; see also Figure 4A). Histone H1 kinase activitiesfrom cells arrested by 100

m

M olomoucine added at 3 hr AFwere similar to those obtained from zygotes blocked in earlyS phase by aphidicolin (Corellou et al., 2000a; data notshown). Histone H1 kinase activities were lower in extracts

from zygotes arrested in mitosis by 35

m

M olomoucine thanin extracts from zygotes arrested in mitosis by nocodazole(Figure 5). However, treatments with the human phos-phatase glutathione

S

-transferase (GST)–cdc25A led to adramatic increase of histone H1 kinase activities from cellsarrested by olomoucine in their cell cycle progression butnot in cells arrested by nocodazole. The relative activationwas higher when the zygotes were arrested early in the cellcycle by 100

m

M olomoucine, that is, when the kinase activ-ities were initially low. Although the factors of activationwere variable (from 5.0 to 15.5), the final levels of kinase ac-tivity after phosphatase treatments were broadly similar tothe mitotic level of histone H1 kinase observed in extractsfrom cells arrested by nocodazole. Together, Figures 4 and5 demonstrate that, as in animal somatic cells, tyrosinephosphorylation is a major mechanism involved in CDK reg-ulation in Fucus zygotes.

A Spindle Assembly Checkpoint Targets Mitotic CDKs in Fucus Zygotes

Nocodazole was used to inhibit mitotic spindle formation.Zygotes that were treated continuously with 0.33

m

M no-codazole displayed heavily condensed chromosomes for atleast 36 hr and never proceeded through mitosis (Figure6C), suggesting the presence of a spindle assembly check-point. In zygotes treated with 0.33

m

M nocodazole from 12 hr

Figure 5. In Vitro Activation of Histone H1 Kinase by the HumanPhosphatase GST-cdc25A.

Protein extracts from zygotes treated with olomoucine or nocoda-zole from the times indicated until 36 hr AF were assayed for histoneH1 kinase activity after treatment with 62 units of GST-cdc25A for20 min (black bars) or after incubation in the dephosphorylationbuffer (gray bars). The relative degrees of activation by GST-cdc25Aare indicated above the black bars. The data shown are representa-tive of the results of two independent experiments.

592 The Plant Cell

AF, histone H1 kinase activity increased progressively from4 to 16 hr AF. This was followed, at the time of entry into mi-tosis at 16 to 18 hr AF, by a sharp increase in activity until22 hr AF (Figure 6A), and high levels of kinase activity wereobserved for at least another 12 hr (data not shown). Thiswas in contrast to the levels of histone H1 kinase activityduring a normal cell cycle, which decreased after the mitotic

peak at 18 hr AF (Figure 6A). The abrupt increase of histoneH1 kinase activity preceded chromosome condensation inboth controls (38 and 80% of cells in mitosis at 18 and 20hr, respectively) and zygotes treated with nocodazole (22and 68% of cells in mitosis at 18 and 20 hr, respectively), in-dicating that the peak or high levels of H1 kinase corre-sponded to kinase activities of mitotic CDKs. No significant

Figure 6. A Spindle Assembly Checkpoint Targets Mitotic CDKs.

(A) Histone H1 kinase activity in extracts from either control zygotes (open circles) or cells treated with 0.33 mM nocodazole from 4 hr AF (closedcircles) until the times indicated on the x axis. The proportion of cells in mitosis was determined by staining DNA with mithramycin A.(B) Protein gel blot analysis of PSTAIRE CDKs in extracts from either control zygotes or zygotes treated with 0.33 mM nocodazole from 4 hr AFuntil the times indicated.(C) A zygote arrested in mitosis with 0.33 mM nocodazole from 6 to 36 hr AF, fixed, and stained with mithramycin A. Note the highly condensedand clustered chromosomes (arrowhead).(D) A zygote arrested in mitosis with 0.33 mM nocodazole from 6 to 36 hr AF (e.g., as in [C]), subsequently treated with both 100 mM olomoucineand 0.33 mM nocodazole for 6 hr, and finally fixed and stained with mithramycin A. Note the decondensed nucleus at 42 hr AF (arrowhead).(E) Dose-dependent in vitro inhibition by olomoucine of histone H1 kinase activity in extracts from cells treated with 0.33 mM nocodazole from 4to 36 hr AF. Kinase activity is represented as a percentage of control activity (no olomoucine). The data shown are representative of the resultsof three independent experiments.Bar in (C) 5 10 mm for (C) and (D).

Cell Cycle Control in Fucus Zygotes 593

differences in the levels of the two PSTAIRE CDKs, p32 andp34, were observed between control zygotes and zygotesarrested in mitosis by nocodazole (Corellou et al., 2000a),suggesting that the mitotic kinase activity of PSTAIRE CDKsis not regulated by the synthesis and/or degradation ofthese proteins (Figure 6B).

Zygotes arrested in mitosis by 0.33 mM nocodazole (Fig-ure 6C) were then treated with 100 mM olomoucine for 6 hr.This allowed the chromatin to decondense in .90% of thecells (Figure 6D). Furthermore, olomoucine inhibited, in adose-dependent manner, the high histone H1 kinase activityfound in extracts from cells incubated with 0.33 mM no-codazole from 6 to 36 hr AF (Figure 6E), suggesting thatCDKs present in these extracts are a target of olomoucine invivo. Together, these results suggest strongly that in Fucuszygotes, the spindle assembly checkpoint operates bymaintaining high levels of activity of mitotic CDKs. Thesedata also suggest indirectly that the inactivation of CDKs isrequired for chromatin decondensation at the end of mitosis.

DISCUSSION

We have shown previously that the G2 phase in Fucus zy-gotes starts at 10 to 12 hr AF and ends at 16 to 18 hr AF,whereas cytokinesis is completed by 22 to 24 hr AF (Corellouet al., 2000a). Based on the increase of histone H3 tran-scripts, which are typically detected at the onset of S phase(Kapros et al., 1992), the S phase begins between 4 and 6 hrAF in Fucus zygotes, that is, after pronuclei fusion (3 to 4 hr AF),an event that is required for full DNA replication (Motomura,1995). Therefore, like the cell cycles of dividing somaticcells, the first cell cycle in Fucus encompasses four well-defined phases: G1, S, G2, and M (Figure 7).

CDK Activity Is Required for S Phase Entry inFucus Zygotes

Although in mammalian somatic cells and yeasts the onset ofS phase is controlled by G1/S CDKs (reviewed in Heichmanand Roberts, 1994), in animal embryos the role of G1/SCDKs in promoting DNA replication is unclear. In the sea ur-chin zygote, cdk2 activity is not required for DNA replication(Moreau et al., 1998). Furthermore, treatments with olomou-cine do not impede DNA replication (Moreau et al., 1998). Incontrast, immunodepletion of cdk2 from Xenopus egg ex-tracts prevents DNA replication, and a peak of cyclin E/cdk2activity correlates with the onset of S phase in frog early em-bryos (Fang and Newport, 1991). In Fucus zygotes, histoneH3 mRNA was detected even upon inhibition of DNA repli-cation by aphidicolin, but the transcription of histone H3 wasfully inhibited by olomoucine, suggesting that this drug in-duces a cell cycle arrest before S phase. This cell cycle arrestwas directly demonstrated in multicellular embryos by quan-

tifying the relative DNA levels in nuclei stained with mithramycinA (data not shown). Therefore, in contrast to the observationsin sea urchin zygote, CDK activity appears to be required totrigger S phase entry in the Fucus zygote.

Synthesis of PSTAIRE CDKs and Early Increase in Histone H1 Kinase Activity Are Triggered by Fertilization but Do Not Rely on Karyogamy

As illustrated in mitogen-activated quiescent human cells,which reenter the cell cycle by promoting the successivetranscription of cdk2 and cdc2 genes (Pagano et al., 1993),the sequential expression of CDK genes plays an importantrole in regulating CDK activity and ordering cell cycle pro-gression in animal somatic cells. In rice, the phytohormonegibberellin was reported to promote the expression of cdc2and associated H1 kinase activity, and in both alfalfa and Ara-bidopsis, the expression of several cdc2-related genes is cellcycle regulated and correlates with their specific activation(Sauter et al., 1995; Magyar et al., 1997; Mironov et al., 1997).In contrast, in animal early embryos, regulation of MPF activ-ity relies primarily on the periodic synthesis of cyclin B. Cdc2is already present in the oocyte, and many rounds of cell divi-sion can occur when transcription is inhibited (Gerhart et al.,1984; Arion and Meijer, 1989). In Fucus eggs, the levels ofboth CDK activity and PSTAIRE CDKs are very low, and thepostfertilization synthesis of PSTAIRE CDKs correlates with adramatic increase of CDK activities. Although the inhibition oftranscription had no effect on either the synthesis of PSTAIRECDKs or histone H1 kinase activity until early G2 phase, theinhibition of translation completely prevented both events,suggesting an early regulation of CDK activity at the transla-tional level. To complete the first cell cycle, however, the Fu-cus zygote also requires transcription for z10 hr AF, or untilearly G2 phase. Furthermore, the mitotic peak of histone H1kinase is dependent on transcription, even though upon inhi-bition of transcription, the level of PSTAIRE CDKs is similar tothat of mitotic controls. Therefore, it is likely that the synthesisof fully active G1/S CDK relies on maternal transcripts,whereas the transcription of proteins such as cyclin B is re-quired to activate mitotic CDKs. Such a regulation of CDK ac-tivities represents an intermediate situation between the cellcycle control of animal early embryos and cell cycle reentry insomatic cells.

Interestingly, karyogamy is dispensable for the synthesisand activation of CDKs during the first cell cycle of Fucuszygotes, suggesting that, as in the sea urchin egg, plas-mogamy is sufficient to commit the egg to enter the firstcell cycle (Shatten et al., 1989). When pronuclei fusion is in-hibited, DNA replication occurs in the egg pronucleus butnot in the male pronucleus, which remains condensedthroughout the first cell cycle of Fucus zygotes (Motomura,1995). Furthermore, the female pronucleus always entersmitosis before the male pronucleus (Motomura, 1995). Wepropose that the transcription of cell cycle genes occurs

594 The Plant Cell

only in the decondensed female pronucleus, because thetranscription-dependent histone H1 kinase is still observedwhen pronuclei fusion is inhibited. The male pronucleus,which remains condensed at all times, would not be able toeither replicate DNA or transcribe the cell cycle genes re-quired for entry into mitosis; thus, its entry into mitosiswould depend on the female pronucleus.

Tyrosine Phosphorylation Is a Major Mechanism Involved in Regulating the Mitotic Activity of CDKs in Fucus Zygotes

The regulation of cdc2 during early animal embryo cell cy-cles usually relies on the periodic synthesis and degradation

of cyclin B. In addition to this regulatory mechanism, in fis-sion yeast and mammalian somatic cells the activation ofmitotic cdc2 operates by tyrosine dephosphorylation ofcdc2 by the phosphatase cdc25 (Lew and Kornbluth, 1996).In various plant taxa, several cdc2 homologs contain a con-served tyrosine residue, most often at position 15 (Zhang etal., 1996), and in tobacco cells, tyrosine phosphorylationwas shown to regulate cdc2-like kinase activity upon stimu-lation with cytokinin (Zhang et al., 1996). Tyrosine phosphor-ylation of the PSTAIRE CDKs p32 and p34 was shown tobe cell cycle regulated in Fucus zygotes, and dephosphory-lation of p34 and p32 correlates with entry into mitosis(Corellou et al., 2000a). Furthermore, the S/M DNA check-point was shown to operate by maintaining mitotic CDKs in-active by tyrosine phosphorylation (Corellou et al., 2000a),

Figure 7. The First Cell Cycle and the Possible Roles and Regulations of CDKs in Fucoid Algae.

This diagram is based on the results of this study and on those reported previously (Corellou et al., 2000a). Activating and inhibitory mechanismsare shown in red and green, respectively. Active and inactive CDKs are shown in orange and yellow, respectively. The first cell cycle compriseswell-defined G1, S, G2, and M phases. (1) PSTAIRE CDKs are synthesized from maternal mRNAs after fertilization. (2) The CDK activity that is re-quired for the G1/S transition is inhibited by olomoucine. The transcription of histone H3 in S phase is inhibited by olomoucine but not by aphid-icolin. (3) Transcription, before 10 hr AF, of the genes encoding activating proteins is required for mitotic activity of CDKs. (4) CDKs aremaintained inactive in G2 by inhibitory phosphorylation (P) on tyrosine residues (Y) and are activated in mitosis by a cdc25-like phosphatase. (5)Olomoucine, by inhibiting CDKs, prevents entry into and progression through mitosis. (6) A DNA replication checkpoint prevents mitosis, includ-ing chromatin condensation and spindle formation, through inactivation of mitotic CDKs by inhibitory phosphorylation. (7) The autocatalytic am-plification of mitotic CDK activity (1), which may rely on the activation of a cdc25-like protein by CDKs, is inhibited by olomoucine (5). (8) Aspindle assembly checkpoint prevents progression through mitosis, including chromatin decondensation, by inhibiting the inactivation of CDKsby an unknown mechanism (?). F, fertilization.

Cell Cycle Control in Fucus Zygotes 595

whereas after activation of the spindle assembly check-point, highly active mitotic CDKs were not phosphorylatedon tyrosine residues. Our results indicate that the variouscell cycle arrests induced by olomoucine also trigger ty-rosine phosphorylation of both p32 and p34 proteins. Olo-moucine treatments did not impede the synthesis ofPSTAIRE CDKs, and the timing of tyrosine phosphorylationduring the first cell cycle was similar in zygotes arrested inG1, at the G2/M transition, or in mitosis. Furthermore, treat-ment with cdc25A led to an activation of CDK activity, con-firming that phosphorylation on tyrosine residues negativelyregulates the activity of CDKs.

In Fucus zygotes, all of the samples treated with olomou-cine (i.e., arrested in G1/S, G2/M, or mitosis) displayed, afterdephosphorylation by cdc25A, kinase levels similar to the highmitotic levels observed after nocodazole treatment. There-fore, the inhibition of G1/S CDKs by olomoucine does not ap-pear to prevent the formation of potentially active mitoticcyclin/CDK complexes, but it likely inhibits their activation bypreventing their dephosphorylation on tyrosine residues.Such a downregulation of mitotic CDKs also occurs whenDNA replication is inhibited (Corellou et al., 2000a). How theinhibition of G1/S CDKs leads to a downregulation of mitoticCDKs remains unclear in Fucus zygotes. As in Xenopus eggextracts, G1/S CDKs may directly and positively regulate mi-totic CDKs through the reversible phosphorylation and acti-vation of cdc25-like proteins (Guadano and Newport, 1996).Alternately, the inhibition of G1/S CDKs, by preventing DNAreplication, may activate the DNA replication checkpoint,which in turn would lead to an inactivation of CDKs by ty-rosine phosphorylation (Corellou et al., 2000a).

Inhibition of mitosis by nocodazole yielded maximal CDKactivities associated with dephosphorylated p34 and p32,whereas zygotes arrested in mitosis by low doses of olomou-cine displayed higher CDK activities than did zygotes ar-rested at the G2/M transition by high doses of olomoucine.Conversely, the activation by cdc25A was much higher in zy-gotes arrested by olomoucine at the G2/M transition than incells arrested by olomoucine in mitosis. Together, our datasuggest that a gradual dephosphorylation, or activation, oc-curs between G2 phase and mitosis, as reported in starfishoocytes (Borgne and Meijer, 1999). This activation seems torequire CDK activity, because it is much lower after a cell cy-cle arrest at the G2/M transition than after an arrest in mitosis.These results suggest a possible inhibition by olomoucine ofan autocatalytic amplification of mitotic cyclin/CDK activitysimilar to those described previously in animals and yeasts(reviewed in Lew and Kornbluth, 1996).

Fucus Zygotes Display a Functional SpindleAssembly Checkpoint

Mitotic cyclin B/cdc2 is known to promote the nucleation ofthe mitotic spindle (Verde et al., 1990; Kishimoto, 1994), andinactivation of the mitotic complex cyclin B/cdc2 at meta-

phase is required for a cell to exit from mitosis and reenter thenext cell cycle (King et al., 1994). Inactivation of mitotic CDKsis first achieved by the degradation of cyclin B (Hershko,1997). In animal somatic cells and in Schizosaccharomycespombe, the lack of a functional spindle prevents the exitfrom mitosis by inhibiting the degradation of key proteinsimplicated in chromosome cohesion as well as that of cyclinB, thus maintaining high levels of mitotic CDK activity(Nishimoto et al., 1992). However, the early embryos of sev-eral animal species, such as Xenopus, lack a fully activespindle assembly checkpoint and undergo abnormal mitosisin the absence of a fully functional spindle (Clute and Masui,1992; Murray, 1992; Sluder et al., 1994). Fucus zygoteslacking a functional spindle after treatment with nocodazolefeature high levels of mitotic CDK. These zygotes did not di-vide or rereplicate DNA, indicating the presence of a fullyactive spindle assembly checkpoint. The fact that the inacti-vation of CDK activity by olomoucine triggers a deconden-sation of chromatin in zygotes continuously treated withnocodazole strongly suggests that inactivation of CDKs isrequired for the exit from mitosis and that, as in animal so-matic cells, the spindle assembly checkpoint operates atleast in part by hampering the inactivation of mitotic CDKs.

These findings are summarized in Figure 7, which de-scribes the major steps of the Fucus first cell cycle togetherwith the possible roles and regulations of CDKs. Cell cycleprogression, including S phase entry, is tightly regulated byCDKs. Two functional DNA replication and spindle assemblycheckpoints block cell cycle progression by altering CDKactivities. CDKs are regulated at both the transcriptional andtransductional levels and by phosphorylation. Together, ourresults indicate that in Fucus zygotes, the zygotic cell cycleresembles a somatic cell cycle more than the typical cell cy-cles of animal embryos in which controls are reduced.

METHODS

Culture and Inhibitors

Sexually mature receptacles of Fucus spiralis were collected at LeDossen (Brittany, France) and stored at 48C for up to 14 days. Gameteswere released by standard osmotic shock procedures in filtered sea-water for 1 hr (Corellou et al., 2000b). The time of fertilization (0 hr)was taken to be 30 min after the first eggs were released. Zygotesand embryos were grown at 148C. Unfertilized eggs were obtainedby inducing receptacles to release in high-potassium artificial seawa-ter, according to Brawley (1987). Aphidicolin (20 mM; Sigma) wasused to inhibit DNA replication and subsequent inhibition of cell divi-sion. Aphidicolin was added at various times after fertilization to de-termine experimentally the end of S phase, that is, the beginning ofG2 phase (Corellou et al., 2000a). Olomoucine was used to inhibit cy-clin-dependent kinases (CDKs) in vivo and in vitro. The structural an-alog iso-olomoucine had no effect on cell division at concentrations ashigh as 500 mM. Nocodazole (0.33 mM; Sigma) was used to preventspindle formation, and it arrested the cells in mitosis. Actinomycin D

596 The Plant Cell

(16 mM) and cycloheximide (0.2 mM) were used to prevent transcrip-tion and translation (Quatrano, 1968; Bouget et al., 1996). Aphidi-colin, olomoucine, nocodazole, actinomycin D, and cycloheximidewere stored in DMSO at 10, 300, 1, 10, and 0.5 mg/mL, respectively,and further diluted in artificial seawater before use. Control experi-ments were performed in filtered seawater containing the same finalconcentrations of DMSO.

DNA Staining and Quantification

Zygotes and embryos were fixed for 12 hr in 0.2 M citric acid and0.2% Triton X-100 and kept in 100% methanol for long-term storage.Fixed cells were attached to poly-L-lysine–coated cover slips, pro-cessed and stained with 50 mg/mL mithramycin A, as describedpreviously (Corellou et al., 2000a), and then observed by epifluores-cence microscopy (540- to 590-nm excitation filter, 585-nm bandpass, a 32-nm band pass emission filter, centered at 585 nm). As re-ported earlier (Corellou et al., 2000a), one-celled zygotes did notyield reproducible results. DNA was therefore quantified by measur-ing mithramycin A fluorescence on two-celled and older embryos.Mithramycin A fluorescence was measured either with a confocal mi-croscope (model 1024; Bio-Rad, Hemel Hempstead, UK) or with aCCD camera (Spot-RT Slider Eurocam model 2.3.0; Diagnostic In-struments, Inc., Sterling Heights, MI). The camera was standardizedwith Fluoresbrite fluorescent beads (Polyscience Inc., Warrington,PA), and the images were analyzed with the Image Tool software (afreeware from UCSB computer, science supported software;www.cs.ucsb.edu). For each sample, an average of 50 to 75 nucleiwere scanned. In the first experiment, the highest fluorescence valueobtained in nocodazole-treated embryos was given the arbitraryvalue of 8.0 relative fluorescence units (RFU), and the average fluo-rescence value of nocodazole-treated cells was determined to be 5.7RFU. In the following experiment, this value was used as a standardto normalize the values of both experiments. The relative fluores-cence of nuclei was represented using a scale of 0 to 8.0 RFUdivided into eight equal 1-unit classes. Histograms with the percent-age of nuclei in each class were established. For each sample, theaverage relative fluorescence were also determined. The average rel-ative fluorescence of nuclei in nocodazole-treated embryos (5.7RFU) was close to the average value of mitotic figures in metaphase(5.3 RFU), corresponding to a 4C DNA content. The average relativefluorescence corresponding to a 2C DNA content, which was deter-mined in derivative nuclei in anaphase (2.7 RFU), was close to theaverage fluorescence value of nuclei treated with aphidicolin frommitosis (2.6 RFU).

Protein Extraction, Purification of CDKs, and Protein GelBlot Analysis

Protocols for protein extraction, CDK purification, and protein gelblot analysis have been described in detail by Corellou et al. (2000a,2000b). Briefly, embryos were harvested, frozen in liquid nitrogen,and stored at 2808C until extraction. Frozen samples were ground inliquid nitrogen. Purification of CDKs was performed on p9CKShs1–Sepharose beads. Eluted proteins were resolved on a 10 or 12% (w/v)SDS–polyacrylamide denaturing gel and electrotransferred onto a ni-trocellulose or a polyvinylidene difluoride membrane (AmershamPharmacia Biotech, Little Chalfont, UK) for enhanced chemilumines-

cence (ECL) or ECL1 detection. Protein gel blotting was performedwith either a monoclonal antiphosphotyrosine antibody (PY20; SantaCruz Biotechnology, Santa Cruz, CA) at a 1:20,000 dilution or a mono-clonal anti-PSTAIRE antibody (Sigma) at a 1:3000 dilution. The an-tiphosphotyrosine antibody could not be detected using the classicECL detection system. Therefore, the blots were probed first with thisantibody, revealed in ECL1, and reprobed with the anti-PSTAIRE an-tibody, which required short times of exposure in ECL detection.

Histone H1 Kinase Activity and in Vitro Dephosphorylation of Proteins Bound to p9CKShs1

The histone H1 kinase activity of proteins bound to p9CKShs1–Sepharose beads was measured at 308C for 30 min using g-32P-ATP,as reported previously (Corellou et al., 2000a). Quantification of ra-dioactive histone H1 was performed using a STORM phosphorimagerand Image QuanT software (Molecular Dynamics, Sunnyvale, CA).For each lane, a protein-free region of the area under quantificationwas used to determine the background according to the object aver-age method. Kinase activities are represented either as relativeSTORM units or as a percentage of control activity for in vitro inhibi-tion experiments. The glutathione S-transferase (GST)–cdc25A fu-sion protein was overproduced in Escherichia coli and purified byaffinity on glutathione–agarose beads, as described previously(Corellou et al., 2000a). Dephosphorylations were initiated by addingto the proteins bound to the p9CKShs1 beads 100 mL (62 units) of thepurified GST-cdc25A in buffer B (50 mM Tris-HCl, pH 8.0, 50 mMNaCl, 1 mM EDTA) containing 20 mM DTT. Dephosphorylation reac-tions were performed at 308C for 1 hr. Control samples were treatedidentically but without GST-cdc25A.

Cloning of the histone H3 Gene and RNA Gel Blot Analysis

DNA was extracted from Fucus sperm using a standard protocol forbrown algae (Apt and Grossman, 1993). A representative genomic li-brary (5 3 105 clones) was established in Lambda DASH II, as recom-mended by the manufacturer (Stratagene, La Jolla, CA). This librarywas screened under low-stringency conditions (Bouget et al., 1995)with a Laminaria digitata histone H3 cDNA probe (kindly provided byDr. Florent Crépineau, Centre National de la Recherche ScientifiqueRoscoff; GenBank accession number AW400773). A 300-bp exonfragment encoding a putative histone H3 (GenBank accession numberAJ276797) .95% identical to the L. digitata deduced amino acid se-quence of histone H3 was subcloned and used as a probe for RNA gelblot analysis as follows. RNA was extracted, electrophoresed througha 1.2% formaldehyde denaturing gel, and transferred to a Hybond N1

membrane (Bouget et al., 1996). The Fucus histone H3 DNA probewas labeled with radioactive a-32P-dCTP using a random priming la-beling kit (Megaprime, Amersham Life Science). High-stringency hy-bridization was performed overnight at 428C in 50% deionizedformamide, 7% (w/v) SDS, 250 mM NaCl, and 120 mM Na(PO4)2, pH7.2. Hybridization washes were performed twice at 428C in 2 3 SSC(13 SSC is 0.15 M NaCl and 0.015 M sodium citrate) containing 0.1%SDS and once in 1 3 SSC containing 0.1% SDS. Membranes were ex-posed to a PhosphorScreen (Molecular Dynamics, Amersham Phar-macia Biotech) overnight. Quantification of bound radioactive probewas performed using a STORM phosphorimager and Image QuanTsoftware.

Cell Cycle Control in Fucus Zygotes 597

ACKNOWLEDGMENTS

We thank Blandine Baratte and Sophie Leclerc for producing GST-cdc25A and Florent Crépineau for providing us with the L. digitata his-tone H3 cDNA. Thanks also to Laurent Meijer for helpful criticism ofthe manuscript. F.C. was a recipient of a doctoral fellowship from theConseil Régional de la Région Bretagne, whose help is gratefullyacknowledged.

Received August 14, 2000; accepted January 10, 2001.

REFERENCES

Alessi, F., Quarta, S., Savio, M., Riva, F., Rossi, L., Stivala, L.A.,Scovassi, A.I., Meijer, L., and Prosperi, E. (1998). The cyclin-dependent kinase inhibitors olomoucine and roscovitine arresthuman fibroblasts in G1 phase by specific inhibition of CDK2kinase activity. Exp. Cell Res. 245, 8–18.

Apt, K.E., and Grossman, A.R. (1993). Characterization and tran-script analysis of the major phycobiliprotein subunit genes fromAglaothamnion neglectum (Rhodophyta). Plant Mol. Biol. 21, 27–38.

Arion, D., and Meijer, L. (1989). M-phase–specific protein kinasefrom mitotic sea urchin eggs: Cyclic activation depends on pro-tein synthesis and phosphorylation but does not require DNA orRNA synthesis. Exp. Cell Res. 183, 361–375.

Arion, D., Meijer, L., Brizuela, L., and Beach, D. (1988). cdc2 is acomponent of the M phase–specific histone H1 kinase: Evidencefor identity with MPF. Cell 55, 371–378.

Bell, M.H., Halford, N.G., Ormrod, J.C., and Francis, D. (1993).Tobacco plants transformed with cdc25, a mitotic inducer genefrom fission yeast. Plant Mol. Biol. 23, 445–451.

Borgne, A., and Meijer, L. (1999). Sequential dephosphorylation ofp34(cdc2) on Thr-14 and Tyr-15 at the prophase–metaphase tran-sition. J. Biol. Chem. 271, 27847–27854.

Bouget, F.Y., Kerbouc’h, C., Liaud, M.-F., Loiseaux de Goër, S.,Quatrano, R.S., Cerff, R., and Kloareg, B. (1995). Structural fea-tures and phylogeny of the actin gene of Chondrus crispus (Gigar-tinales, Rhodophyta). Curr. Genet. 28, 164–172.

Bouget, F.Y., Gertulla, S., Shaw, S.L., and Quatrano, R.S. (1996).Localization of actin mRNA during the establishment of cell polarityand early cell divisions in Fucus embryos. Plant Cell 8, 189–201.

Brawley, S.H. (1987). A sodium-dependent, fast block to polyspermyoccurs in eggs of fucoid algae. Dev. Biol. 124, 390–397.

Brawley, S.H., and Quatrano, R.S. (1979). Effect of microtubuleinhibitors on pronuclear migration and embryogenesis in Fucusdistichus (Pheophyta). J. Phycol. 15, 266–272.

Brownlee, C., and Bouget, F.Y. (1998). Polarity determination inFucus: From zygote to multicellular embryo. Semin. Cell. Dev.Biol. 9, 179–185.

Chaubet, N., and Gigot, C. (1998). Histone gene expression. InPlant Cell Division, Vol. 10, D. Francis, D. Dudits, and D. Inzé, eds(London: Portland Press), pp. 269–283.

Chevalier, S., and Blow, J. (1996). Cell cycle control of replicationinitiation in eukaryotes. Curr. Opin. Cell Biol. 8, 815–821.

Clute, P., and Masui, Y. (1992). Development of microtubule-dependence of the chromosome cycle at the midblastula transi-tion in Xenopus laevis embryos. Dev. Growth Differ. 34, 27–36.

Corellou, F., Bisgrove, S.R., Kropf, D.L., Meijer, L., Kloareg, B.,and Bouget, F.-Y. (2000a). A S/M DNA replication checkpointprevents nuclear and cytoplasmic events of cell division includingcentrosomal axis alignment and inhibits activation of cyclin-dependent kinase-like proteins in fucoid zygotes. Development127, 1651–1660.

Corellou, F., Potin, P., Brownlee, C., Kloareg, B., and Bouget, F.-Y.(2000b). Inhibition of zygotic polarity by protein tyrosine kinaseinhibitors leads to an alteration of embryo pattern in Fucus. Dev.Biol. 219, 165–182.

Dudits, D., Magyar, Z., Deak, M., Mészaros, P., Miskolczi, A.,Brown, S., Kondorosi, E., Athanasidis, A., Pongor, S., Bako, L.,Koncz, C.S., and Györgyey, J. (1998). Cyclin-dependent and cal-cium-dependent kinase families: Response of cell division cycleto hormone and stress signals. In Plant Cell Division, Vol. 10, D.Francis, D. Dudits, and D. Inzé, eds (London: Portland Press), pp.21–45.

Dunphy, W.G., and Newport, J.W. (1988). Mitosis-inducing factorsare present in a latent form during interphase in the Xenopusembryo. J. Cell Biol. 106, 2047–2056.

Edgar, B. (1995). Diversification of cell cycle controls in developingembryos. Curr. Opin. Cell Biol. 7, 815–824.

Elledge, S.J., Richman, R., Hall, F.L., Williams, R.T., Lodgson, N.,and Harper, J.W. (1992). CDK2 encodes a 33-kDa cyclin A–asso-ciated protein kinase and is expressed before CDC2 in the cellcycle. Proc. Natl. Acad. Sci. USA 89, 2907–2911.

Fang, F., and Newport, J.W. (1991). Evidence that the G1-S andG2-M transitions are controlled by different cdc2 proteins inhigher eukaryotes. Cell 66, 731–742.

Friedman, E.F. (1999). Expression of the cell cycle in sperm of Ara-bidopsis: Implications for understanding patterns of gametogene-sis and fertilization in plants and other eukaryotes. Development126, 1065–1075.

Gerhart, J., Wu, M., and Kirschner, M. (1984). Cell cycle dynamicsof an M-phase specific cytoplasmic factor in Xenopus laevisoocytes and eggs. J. Cell Biol. 98, 1247–1255.

Glab, N., Labidi, B., Qin, L.X., Trehin, C., Bergounioux, C., andMeijer, L. (1994). Olomoucine, an inhibitor of the cdc2/cdk2kinases activity, blocks plant cells at the G1 to S and G2 to M cellcycle transitions. FEBS Lett. 353, 207–211.

Guadano, T.M., and Newport, J.W. (1996). Cdk2 kinase is requiredfor entry into mitosis as a positive regulator of Cdc2–cyclin Bkinase activity. Cell 84, 73–82.

Hardwick, K.G. (1998). The spindle checkpoint. Trends Genet. 14, 1–4.

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

Heichman, K.A., and Roberts, J.M. (1994). Rules to replicate by.Cell 79, 557–562.

Hershko, A. (1997). Roles of ubiquitin-mediated proteolysin cellcycle control. Curr. Opin. Cell Biol. 9, 788–799.

Kapros, T., Bögre, L., Németh, K., Bako, L., Györgyey, J., Wu,

598 The Plant Cell

S.C., and Dudits, D. (1992). Differential expression of histone H3gene variants during cell cycle and somatic embryogenesis inalfalfa. Plant Physiol. 98, 621–625.

King, R.W., Jackson, P.K., and Kirschner, M.W. (1994). Mitosis intransition. Cell 79, 563–571.

Kishimoto, T. (1988). Regulation of metaphase by maturation pro-moting factor. Dev. Growth Differ. 30, 105–115.

Kishimoto, T. (1994). Cell reproduction: Induction of M-phase eventsby cyclin-dependent cdc2 kinase. Int. J. Dev. Biol. 38, 185–191.

Lew, D.J., and Kornbluth, S. (1996). Regulatory roles of cyclindependent kinase phosphorylation in cell cycle control. Curr.Opin. Cell Biol. 8, 795–804.

Magyar, Z., et al. (1997). Cell cycle phase specificity of putativecyclin-dependent kinase variants in synchronized alfalfa cells.Plant Cell 9, 223–235.

Masui, Y., and Markert, C.L. (1971). Cytoplasmic control of nuclearbehavior during meiotic maturation of frog oocytes. J. Exp. Zool.177, 129–146.

McKibbin, R.S., Halford, N.G., and Francis, D. (1998). Expressionof fission yeast cdc25 alters the frequency of lateral root formationin transgenic tobacco. Plant Mol. Biol. 36, 601–612.

Mironov, V., Van Montagu, M., and Inze, D. (1997). Regulation ofcell division in plants: An Arabidopsis perspective. Prog. CellCycle Res. 3, 29–41.

Mironov, V., de Veylder, L., Van Montagu, M., and Inze, D. (1999).Cyclin-dependent kinases and cell division in plants: The nexus.Plant Cell 11, 509–522.

Moreau, J.L., Marques, F., Barakat, A., Schatt, P., Lozano, J.C.,Peaucellier, G., Picard, A., and Geneviere, A.M. (1998). Cdk2activity is dispensable for the onset of DNA replication during thefirst mitotic cycles of the sea urchin early embryo. Dev. Biol. 200,182–197.

Motomura, T. (1995). Premature chromosome condensation of thekaryogamy-blocked sperm pronucleus in the fertilization in zygotesof the brown alga Fucus distichus. J. Phycol. 31, 108–113.

Murray, A. (1992). Creative blocks: Cell-cycle checkpoints andfeedback controls. Nature 359, 599–604.

Neufeld, T.P., and Edgar, B.A. (1998). Connections betweengrowth and the cell cycle. Curr. Opin. Cell Biol. 10, 784–790.

Nishimoto, T., Uzawa, S., and Schlegel, R. (1992). Mitotic check-points. Curr. Opin. Cell Biol. 4, 174–179.

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

Pagano, M., Pepperkok, R., Lukas, J., Baldin, V., Ansorge, W.,Bartek, J., and Draetta, G. (1993). Regulation of the cell cycle bythe cdk2 protein kinase in cultured human fibroblasts. J. Cell Biol.121, 101–111.

Paulovich, A.G., Toczyski, D.P., and Hartwell, L.H. (1997). Whencheckpoints fail. Cell 88, 315–321.

Peter, M., and Herskowitz, I. (1994). Joining the complex: Cyclin-dependent kinase inhibitory proteins and the cell cycle. Cell 79,181–184.

Peter, M., Nakagawa, J., Dorée, M., Labbé, J.C., and Nigg, E.A.(1990). In vitro disassembly of the nuclear lamina and M phase–spe-cific phosphorylation of lamins by cdc2 kinase. Cell 61, 591–602.

Planas-Silva, M.D., and Weinberg, R.A. (1997). The restriction pointand control of cell proliferation. Curr. Opin. Cell Biol. 9, 768–772.

Planchais, S., Glab, N., Trehin, C., Perennes, C., Bureau, J.M.,Meijer, L., and Bergounioux, C. (1997). Roscovitine, a novelcyclin-dependent kinase inhibitor, characterizes restriction pointand G2/M transition in tobacco BY-2 cell suspension. Plant J. 12,191–202.

Quatrano, R.S. (1968). Rhizoid formation in Fucus zygotes: Depen-dence on protein and ribonucleic acid syntheses. Science 162,468–470.

Rudner, A.D., and Murray, A.W. (1996). The spindle assemblycheckpoint. Curr. Opin. Cell Biol. 8, 773–780.

Sauter, M., Mekhedov, S.L., and Kende, H. (1995). Gibberellin pro-motes histone H1 kinase activity and the expression of cdc2 andcyclin genes during the induction of rapid growth in deepwaterrice internodes. Plant J. 7, 623–632.

Sauter, M., von Wiegen, P., Lörz, H., and Kranz, E. (1998). Cellcycle regulatory genes from maize are differentially controlled dur-ing fertilization and first embryonic cell division. Sex Plant Reprod.11, 41–48.

Shatten, H.C., Simerly, G., Maul, G., and Shatten, G. (1989).Microtubule assembly is required for the formation of pronuclei,nuclear lamin acquisition and DNA synthesis during mouse, butnot sea urchin, fertilization. Gamete Res. 23, 309–322.

Sluder, G., Miller, F.J., Thompson, E.A., and Wolf, D.E. (1994).Feedback control of the metaphase–anaphase transition in seaurchin zygotes: Role of maloriented chromosomes. J. Cell Biol.126, 189–198.

Swope, R.E., and Kropf, D.L. (1993). Pronuclear positioning andmigration during fertilization in Pelvetia. Dev. Biol. 157, 269–276.

Verde, F., Labbé, J.-C., Dorée, M., and Karsenti, E. (1990). Regu-lation of microtubule dynamics by cdc2 protein kinase in cell-freeextracts of Xenopus eggs. Nature 343, 233–238.

Vesely, J., Havlicek, L., Strnad, M., Blow, J.J., Donella-Deana, A.,Pinna, L., Letham, D.S., Kato, J., Detivaud, L., Leclerc, S., andMeijer, L. (1994). Inhibition of cyclin-dependent kinases by purineanalogues. Eur. J. Biochem. 224, 771–786.

Zhang, K., Letham, D.S., and John, P.C. (1996). Cytokinin controls thecell cycle at mitosis by stimulating the tyrosine dephosphorylationand activation of p34cdc2-like H1 histone kinase. Planta 200, 2–12.

DOI 10.1105/tpc.13.3.585 2001;13;585-598Plant Cell

Florence Corellou, Colin Brownlee, Lenaick Detivaud, Bernard Kloareg and François-Yves BougetRegulation of Cyclin-Dependent Kinases

Cell Cycle in the Fucus Zygote Parallels a Somatic Cell Cycle but Displays a Unique Translational

 This information is current as of April 12, 2019

 

References /content/13/3/585.full.html#ref-list-1

This article cites 57 articles, 14 of which can be accessed free at:

Permissions https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298X

eTOCs http://www.plantcell.org/cgi/alerts/ctmain

Sign up for eTOCs at:

CiteTrack Alerts http://www.plantcell.org/cgi/alerts/ctmain

Sign up for CiteTrack Alerts at:

Subscription Information http://www.aspb.org/publications/subscriptions.cfm

is available at:Plant Physiology and The Plant CellSubscription Information for

ADVANCING THE SCIENCE OF PLANT BIOLOGY © American Society of Plant Biologists