Preprophase bands of microtubules and the cell cycle: Kinetics and experimental uncoupling of their formation from the nuclear cycle in onion root-tip cells

We have studied the timing of preprophase band (PPB) development in the division cycle of onion (Allium cepa L.) root-tip cells by combinations of immunofluorescence microscopy of microtubules, microspectrophotometry of nuclear DNA, and autoradiography of [³H]thymidine incorporation during pulse-chase experiments. In normally grown onion root tips, every cell with a PPB had the G2 level of nuclear DNA. Some were in interphase, prior to chromatin condensation, and some had varying degrees of chromatin condensation, up to the stage of prophase at which the PPB-prophase spindle transition occurs. In addition, autoradiography showed that PPBs can be formed in cells which have just finished their S phase, and microspectrophotometry enabled us to detect a population of cells in G2 which had no PPBs, these presumably including cells which had left the division cycle. The effects of inhibitors of DNA synthesis showed that the formation of PPBs is not fully coupled to events of the nuclear cycle. Although the mitotic index decreased 6-10-fold to less than 0.5% when roots were kept in 20 g·ml^-1 aphidicolin for more than 8 h, the percentage of cells containing PPBs did not decrease in proportion: the number of cells in interphase with PPBs increased while the number in prophase decreased. Almost the same phenomena were observed in the presence of 100 g·ml^-1 5-aminouracil and 40 g·ml^-1 hydroxyurea. In controls, all cells with PPBs were in G2 or prophase, but in the presence of aphidicolin, 5-aminouracil or hydroxyurea, some of the interphase cells with PPBs were in the S phase or even in the G1 phase. We conclude that PPB formation normally occurs in G2 (in at least some cases very early in G2) and that this timing can be experimentally uncoupled from the timing of DNA duplication in the cell-division cycle. The result accords with other evidence indicating that the cytoplasmic events of cytokinesis are controlled in parallel to the nuclear cycle, rather than in an obligatorily coupled sequence.Abbreviations APC aphidicolin - 5-AU 5-aminouracil - DAPI 4, 6-diamidino-2phenylindole - HU hydroxyurea - MI mitotic index - MT microtubule - PMSF phenylmethyl-sulfonyl fluoride - PPB preprophase band - %PPB percentage of cells with PPBs     

Preprophase bands in cultured multinucleate soybean protoplasts

Summary Multinucleate cells derived from soybean protoplasts were used to investigate the effect of increased nuclear number on the development and frequency of preprophase bands (PPBs) of microtubules (MTs). The results do not support the assumption that one nucleus establishes one PPB because the majority of multinucleate cells had only one large PPB. However, nuclear number or ploidy level has some influence on PPB development since double PPBs occurred more often in multinucleate than uninucleate cells. Double (divergent) PPBs were present at early and late stages of PPB development, suggesting that they are not a transient stage. PPBs in multinucleate cells developed in a similar fashion to those in uninucleate cells. In multinucleate cells, each dividing nucleus had its own spindle and phragmoplast. Subsequent phragmoplast development was frequently uncoupled from PPB distribution. Most multinucleates contained a single large PPB but at telophase, multiple phragmoplasts oriented in different planes.Abbreviations MT microtubule - MtSB microtubule stabilizing buffer - PBS phosphate buffered saline - PNF perinuclear fluorescence - PPB preprophase band     

Rearrangements of Microtubules Involved in Establishing Cell Division Planes Start Immediately after DNA Synthesis and Are Completed just before Mitosis

This work concerns an aspect of spatial regulation of cell division, the development of the preprophase band (PPB) of microtubules. The PPB is significant in plant development because its position in the dividing cell indicates where the new cell wall will be inserted[mdash]an important site for control of histogenesis. We have categorized and determined the durations of stages in the development of PPBs, and have established their timing relative to the S-, G2-, and mitotic phases of the cell cycle. Roots of wheat seedlings were supplied with bromodeoxyuridine in continuous and pulse-chase treatments. Cells that were in the S-phase were identified and changes in their microtubule arrays were monitored by double immunolabeling. PPB initiation was detectable as early as the end of the S-phase as a narrowing of the preceding interphase array of microtubules. Development continued throughout G2 to a mature, narrow PPB, which existed only briefly and then eroded during the transition to the prophase mitotic spindle. The microtubule rearrangements of PPB development showed that preparation of the future site and plane of division in higher plant cells begins just after DNA replication and is completed just before mitosis.     

Microtubule pattern and the occurrence of pre-prophase bands in embryogenic cultures of black spruce (Pieca mariana Mill.) and non-embryogenic cultures of jack pine (Pinus banksiana Lamb.)

Summary The organization of microtubules during interphase and prophase in embryogenic cultures of black spruce (Picea mariana) was investigated by indirect immunofluorescence. Somatic embryos of black spruce possessed an extensively branched and interconnecting network of fine interphase cortical microtubules. The development of pre-prophase bands (PPBs) in embryogenic black spruce cultures was compared with that in non-embryogenic cell cultures of jack pine (Pinus banksiana). PPBs in both species were initially arranged as a very broad array of microtubules, later (early to mid-prophase) becoming narrower and more intensely fluorescent. The occurrence of pre-prophase bands in relation to the number of phragmoplasts (i.e. PPB index) of black spruce somatic embryos was significantly higher (p<0.01) than that found for jack pine cells.     

Nuclear-Cycle Dependence of the Development of the Preprophase Band in Tobacco BY-2 Cells

When tobacco BY-2 cells that had been treated with aphidicolinfor 24 h were cultured in the absence of aphidicolin, DNA synthesiswas initiated within 1 h. DNA synthesis was completed within4 h and then the preprophase band of microtubules (PPB) developed.However, when cells that had been treated with aphidicolin werecultured in the absence of aphidicolin for 1 h and then againin its presence, DNA synthesis, which was initiated during thehour in the absence of aphidicolin, was not completed withina further 10 h in the presence of aphidicolin. Moreover, thePPB did not develop even after the PPB had appeared and disappearedin cells that were cultured contemporaneously in the continuedabsence of aphidicolin. The development of the PPB seems to be causally associated withthe nuclear cycle of cell division in tobacco BY-2 cells. Thisprocess seems to require the completion of the replication ofall, or almost all, of the nuclear DNA. (Received July 25, 1992; Accepted November 24, 1992)     

Preprophase bands in a suspension culture of the monocot Spartina pectinata

Continuous suspension cultures of the marsh grass Spartina pectinata grow as either unorganized colonies or files of cells. Immunofluorescence of tubulin revealed microtubule (MT) structures similar to those encountered in meristematic cells, including cortical microtubule (MT) bands in some interphase cells and in all prophase cells. These MT bands were judged to be preprophase bands (PPBs) on the basis of their temporal appearance in the cell cycle and their position and orientation relative to division planes. Although PPBs are widely thought to be associated with organized tissues and polarized divisions, there are reports of PPBs in suspension cultures of four dicot species. This is the first report of a PPB in suspension cultures of a monocot species.     

Nuclear and microtubular cycles in heterophasic multinuclearTriticum root-tip cells induced by caffeine

Summary Nuclear and microtubular cycles were studied in large heterophasic multinuclear cells induced in root tips ofTriticum turgidum by caffeine treatment. Multinuclear cells and cells with polyploid nuclei exhibited various configurations of multiple and complex preprophase microtubule (Mt) bands (PPBs), including helical ones. The developmental stages of PPBs in some heterophasic cells did not comply with the cell cycle stages of the associated nuclei, a fact indicating that these events are not directly controlled by the associated nuclei. The heterophasic cells exhibited asynchronous nuclei at different stages of mitosis. In cells displaying prophase and interphase nuclei, the prophase spindle was either absent or developed around both of them or developed around the prophase nuclei earlier than around the interphase ones. During prometaphase-metaphase of the advanced nuclei the lagging interphase nuclei were induced to form prematurely condensed chromosomes (PCCs) along with spindle formation around them. These observations suggest that the mitotic transition in heterophasic cells is delayed but is ultimately achieved due to the effect of the advanced nuclei, which induces a premature mitotic entry of the lagging nuclei. Although kinetochore Mt bundles were found associated with PCCs, their metaphase and anaphase spindles were abnormal resulting in abnormal or abortive anaphases. In some heterophasic cells, metaphase-anaphase transition did not take place simultaneously in different chromosome groups, signifying that the cells do not exit from the mitotic state after anaphase initiation of the advanced nuclei. Asynchronous pace of mitosis of different chromosome groups was also observed during anaphase and telophase. Implications of these observations in understanding plant cell cycle regulation are discussed.Abbreviations cdk cyclin dependent kinase - Mt microtubule - PCC prematurely condensed chromosome - PPB preprophase band     

Decision of spindle poles and division plane by double preprophase bands in a BY-2 cell line expressing GFP-tubulin

The preprophase band (PPB) of microtubules is thought to be involved in deciding the future division site. In this study, we investigated the effects of double PPBs on spindle formation and the directional decision of cytokinesis by using transgenic BY-2 cells expressing green fluorescent protein (GFP)-tubulin. At prophase, most of the cells with double PPBs formed multipolar spindles, whereas all cells with single PPBs formed normal bipolar spindles, clearly implicating the PPB in deciding the spindle poles. At metaphase, however, both cell types possessed the bipolar spindles, indicating the existence of correctional mechanism(s) at prometaphase. From prometaphase to metaphase, the spindles in double PPB cells altered their directions to become oblique to the cell-elongating axis, and these orientations were maintained in the phragmoplast and resulted in the oblique division planes. These oblique cell plates decreased when actin microfilaments were disrupted, and double actin-depleted zones (ADZs) appeared where the double PPBs had existed. These results suggest that the information necessary for proper cytokinesis may be transferred from the PPBs to the ADZs, even in the case of the double PPBs.     

Actomyosin is Involved in the Organization of the Microtubule Preprophase Band in Arabidopsis Suspension Cultured Cells

The microtubule preprophase bands （PPBs） participate in the sequence of events to position cell plates in most plants. However, the mechanism of PPB formation remains to be clarified. In the present study, the organization of PPBs in Arabidopsis suspension cultured cells was investigated by confocal laser scanning microscopy combined with pharmacological treatments of reagents specific for the cytoskeleton elements. Double staining of F-actin and microtubules （MTs） showed that actin filaments were arranged randomly and no colocalization with cortical MTs was observed in the interphase cells. However, cortical actin filaments showed colocalization with MTs during the formation of PPBs. A broad actin band formed with the broad MT band in the initiation of PPB and narrowed down together with the MT band to form the PPB. Nevertheless, broad MT bands were formed but failed to narrow down in cells treated with the F-actin disruptor latrunculin A. In contrast, in the presence of the F-actin stabilizer phalloidin, PPB formation did not exhibit any abnormality. Therefore, the integrity, but not the dynamics, of the actin cytoskeleton is necessary for the formation of normal PPBs. Treatment with 2, 3-butanedine monoxime, a myosin inhibitor, also resulted in the formation of broad MT bands, indicating that actomyosin may be involved in the rearrangement of MTs to form the PPBs. Double staining of MTs and myosin revealed that myosin concentrated on the PPB region during PPB formation. It is suggested that the actin cytoskeleton at the PPB site may serve as a rack to transport cortical MTs by using myosin when the broad MT band narrows down to form the PPB.     

Effects of cycloheximide on preprophase bands and prophase spindles in onion (Allium cepa L.) root tip cells

Summary Effects of cycloheximide (CHM) on preprophase bands (PPBs) of microtubules (MTs) and on prophase spindle MTs in root tip cells of onion (Allium cepa L.) were examined. When root tip cells were treated with 36 M CHM for 0.5–4 h, the population of cells with a PPB did not decrease markedly although the population of mitotic cells and that of prophase cells with a PPB gradually decreased to half of the control root tips. In prophase cells treated with 11 and 36 M CHM for 2 h, the width of the PPB was 1.4 times broader than that in the prophase PPB without CHM. Electron microscopic observation on the cross section of the PPB showed that the number of MTs and the distance between adjacent MTs in prophase PPBs treated with CHM were similar to those in the early developmental stage of PPBs without CHM. The bipolar spindle, that appeared in late prophase was not seen in prophase cells treated with 11 M or higher concentrations of CHM for 2 h. In order to examine differences of perinuclear MT arrangement between CHM treated and non-treated prophase cells, arrangement of perinuclear MTs was examined by confocal laser scanning microscopy. In control cells without CHM, MTs appeared on the nuclear surface with several branched or cross over type MT foci in the cytoplasm when broad PPB formation started. These MT foci were replaced by the aster type MT foci, from which several MTs radiated along the nuclear surface. The aster type MT foci gradually gathered to form a bipolar spindle. MTs connecting the spindle pole region and the PPB were seen in late prophase. In CHM-treated cells (11-360 M for 2 h), branched and cross over type MT foci were prominent, even in prophase cells with well condensed chromosomes. Neither linkages of MTs between the spindle pole region and the PPB nor aster type MT foci were seen. These observations showed that CHM prevents the bundling of MTs in the PPB and also inhibits the formation of aster type MT foci that is essential for bipolar spindle development.     

Arabidopsis TANGLED identifies the division plane throughout mitosis and cytokinesis

BACKGROUND: In premitotic plant cells, the future division plane is predicted by a cortical ring of microtubules and F-actin called the preprophase band (PPB). The PPB persists throughout prophase, but is disassembled upon nuclear-envelope breakdown as the mitotic spindle forms. Following nuclear division, a cytokinetic phragmoplast forms between the daughter nuclei and expands laterally to attach the new cell wall at the former PPB site. A variety of observations suggest that expanding phragmoplasts are actively guided to the former PPB site, but little is known about how plant cells "remember" this site after PPB disassembly. RESULTS: In premitotic plant cells, Arabidopsis TANGLED fused to YFP (AtTAN::YFP) colocalizes at the future division plane with PPBs. Strikingly, cortical AtTAN::YFP rings persist after PPB disassembly, marking the division plane throughout mitosis and cytokinesis. The AtTAN::YFP ring is relatively broad during preprophase/prophase and mitosis; narrows to become a sharper, more punctate ring during cytokinesis; and then rapidly disassembles upon completion of cytokinesis. The initial recruitment of AtTAN::YFP to the division plane requires microtubules and the kinesins POK1 and POK2, but subsequent maintenance of AtTAN::YFP rings appears to be microtubule independent. Consistent with the localization data, analysis of Arabidopsis tan mutants shows that AtTAN plays a role in guidance of expanding phragmoplasts to the former PPB site. CONCLUSIONS: AtTAN is implicated as a component of a cortical guidance cue that remains behind when the PPB is disassembled and directs the expanding phragmoplast to the former PPB site during cytokinesis.     

Phragmoplasts in the Absence of Nuclear Division

All land plants (embryophytes) use a phragmoplast for cytokinesis. Phragmoplasts are distinctive cytoskeletal structures that are instrumental in the deposition of new walls in both vegetative and reproductive phases of the life cycle. In meristems, the phragmoplast is initiated among remaining non-kinetochore spindle fibers between sister nuclei and expands to join parental walls at the site previously marked by the preprophase band of microtubules (PPB). The microtubule cycle and cell cycle are closely coordinated: the hoop-like cortical microtubules of interphase are replaced by the PPB just prior to prophase, the PPB disappears as the spindle forms, and the phragmoplast mediates cell plate deposition after nuclear division. In the reproductive phase, however, cortical microtubules and PPBs are absent and cytokinesis may be uncoupled from the cell cycle resulting in multinucleate cells (syncytia). Minisyncytia of 4 nuclei occur in microsporocytes and several (typically 8) nuclei occur in the developing megagametophyte. Macrosyncytia with thousands of nuclei may occur in the nuclear type endosperm. Cellularization of syncytia involves formation of adventitious phragmoplasts at boundaries of nuclear-cytoplasmic domains (NCDs) defined by radial microtubule systems (RMSs) emanating from non-sister nuclei. Once initiated in the region of microtubule overlap at interfaces of opposing RMSs, the adventitious phragmoplasts appear structurally identical to interzonal phragmoplasts. Phragmoplasts are constructed of multiple opposing arrays similar to what have been termed microtubule converging centers. The individual phragmoplast units are distinctive fusiform bundles of anti-parallel microtubules bisected by a dark mid-zone where vesicles accumulate and fuse into a cell plate.     

Onset of the DNA replication checkpoint in the early Drosophila embryo

The Drosophila embryo is a promising model for isolating gene products that coordinate S phase and mitosis. We have reported before that increasing maternal Cyclin B dosage to up to six copies (six cycB) increases Cdk1-Cyclin B (CycB) levels and activity in the embryo, delays nuclear migration at cycle 10, and produces abnormal nuclei at cycle 14. Here we show that the level of CycB in the embryo inversely correlates with the ability to lengthen interphase as the embryo transits from preblastoderm to blastoderm stages and defines the onset of a checkpoint that regulates mitosis when DNA replication is blocked with aphidicolin. A screen for modifiers of the six cycB phenotypes identified 10 new suppressor deficiencies. In addition, heterozygote dRPA2 (a DNA replication gene) mutants suppressed only the abnormal nuclear phenotype at cycle 14. Reduction of dRPA2 also restored interphase duration and checkpoint efficacy to control levels. We propose that lowered dRPA2 levels activate Grp/Chk1 to counteract excess Cdk1-CycB activity and restore interphase duration and the ability to block mitosis in response to aphidicolin. Our results suggest an antagonistic interaction between DNA replication checkpoint activation and Cdk1-CycB activity during the transition from preblastoderm to blastoderm cycles.     

Localization of the microtubule end binding protein EB1 reveals alternative pathways of spindle development in Arabidopsis suspension cells

In a previous study on Arabidopsis thaliana suspension cells transiently infected with the microtubule end binding protein AtEB1a-green fluorescent protein (GFP), we reported that interphase microtubules grow from multiple sites dispersed over the cortex, with plus ends forming the characteristic comet-like pattern. In this study, AtEB1a-GFP was used to study the transitions of microtubule arrays throughout the division cycle of cells lacking a defined centrosome. During division, the dispersed origin of microtubules was replaced by a more focused pattern with the plus end comets growing away from sites associated with the nuclear periphery. The mitotic spindle then evolved in two quite distinct ways depending on the presence or absence of the preprophase band (PPB): the cells displaying outside-in as well as inside-out mitotic pathways. In those cells possessing a PPB, the fusion protein labeled material at the nuclear periphery that segregated into two polar caps, perpendicular to the PPB, before nuclear envelope breakdown (NEBD). These polar caps then marked the spindle poles upon NEBD. However, in the population of cells without PPBs, there was no prepolarization of material at the nuclear envelope before NEBD, and the bipolar spindle only emerged clearly after NEBD. Such cells had variable spindle orientations and enhanced phragmoplast mobility, suggesting that the PPB is involved in a polarization event that promotes early spindle pole morphogenesis and subsequent positional stability during division. Astral-like microtubules are not usually prominent in plant cells, but they are clearly seen in these Arabidopsis cells, and we hypothesize that they may be involved in orienting the division plane, particularly where the plane is not determined before division.     

Morphogenesis and regulated gene activity are independent of DNA replication in Xenopus embryos.

Xenopus embryos were transferred into media containing aphidicolin at late blastula, mid-gastrula, and early neurula stages. In each case, embryos continued to differentiate in the absence of DNA replication. When the inhibitor was added at late blastula, embryos continued to develop for about 8 h. However, when aphidicolin was added at the early neurula stage, development could be seen for up to 40 h after addition. The influence of replication on embryonic gene activity was studied by RNA blot analysis. Of the genes we examined only histone gene expression was down regulated by the addition of aphidicolin. The expression of various embryo-specific genes was unaffected by the lack of DNA synthesis. Even after several hours of treatment with aphidicolin, replication-inhibited tailbud and tadpole stages showed the same levels of specific mRNAs as control embryos containing 4-5 times more DNA. We conclude that morphogenesis and embryo-specific gene activity are independent of both DNA replication and a precise amount of DNA per embryo.     

Genetic evidence that aphidicolin inhibits in vivo DNA synthesis in Chinese hamster ovary cells

Using a genetic approach, Chinese hamster ovary (CHO) cells sensitive (aph^S) and resistant (aph^R) to aphidicolin were grown in the presence or absence of various DNA polymerase inhibitors, and the newly synthesized DNA isolated from [³²P]dNMP-labelled, detergent-permeabilized cells, was characterized after fractionation by gel electrophoresis. The particular aph ^Rmutant CHO cell line used was one selected for resistance to aphidicolin and found to possess an altered DNA polymerase of the a-family. The synthesis of a 24 kb replication intermediate was inhibited in wild-type CHO cells grown in the presence of aphidicolin, whereas the synthesis of this replication intermediate was not inhibited by this drug in the mutant CHO cells or in the aphidicolin-resistant somatic cell hybrid progeny constructed by fusion of wild-type and mutant cell lines. Arabinofuranosylcytosine (ara-C), like aphidicolin, inhibited the synthesis of this 24 kb DNA replication intermediate in the wild-type CHO cells but not in the aph^R mutant cells. However, carbonyldiphosphonate (COMDP) inhibited the synthesis of the 24 kb replication intermediate in both wild-type and mutant cells. N²-(p-n-Butylphenyl)-2 deoxyguanisine-5-triphosphate (BuPdGTP) was found to inhibit the formation of Okazaki fragments equally well in the wild-type and mutant cell lines and thus led to inhibition of synthesis of DNA intermediates in both cases. It appears that aphidicolin and ara-C both affect a common target on the DNA polymerase, which is different from that affected by COMDP in vivo. These data also show that aphidicolin, ara-C and COMDP affect the elongation activity of DNA polymerase but not the initiation activity of the enzyme during DNA replication. This is the first report of such differentiation of the DNA polymerase activities during nuclear DNA replication in mammalian cells. The method of analysis described here for replication intermediates can be used to examine the inhibitory activities of other chemicals on DNA synthesis.     

DURATION OF THE NUCLEAR CYC LE IN TRADESCANTIA PALUDOSA ROOT TIPS AS MEASURED WITH H3-THYMIDINE

Wimber , Donald E. (Brookhaven National Lab., Upton, N. Y.) Duration of the nuclear cycle in Tradescantia paludosa root tips as measured with H³-thymidine. Amer. Jour. Bot. 47(10): 828–834. Illus. 1960.—The duration of the nuclear cycle and its various subdivisions were measured in Tradescantia root tips by autoradiographic techniques. H³-thymidine was used as a nuclear label and was supplied to the roots for 0.5 hr. After labeling, the roots were allowed to grow in the absence of label for periods up to 38 hr. By determining the percentage of divisions labeled at the various times of fixation, a reconstruction of the nuclear cycle could be made. The average cycle was determined as 20 hr. in duration, DNA synthesis 10.8 hr., presynthetic interphase 4 hr., postsynthetic interphase 2.7 hr., prophase 1.6 hr., metaphase 0.3 hr. and anaphase-telophase 0.6 hr. Approximate standard deviations for the duration of some of the subdivisions were calculated.     

Chinese hamster ovary cell mitosis and its response to ionizing radiation: a morphological analysis of the living cell

Repeated microscopic observations of exponentially growing Chinese hamster ovary cells were made and the times and mitotic stages were recorded in control and irradiated cultures at 37 degrees C. As determined by autoradiography, the time from the end of S phase to early prophase (the G2 phase) was 46 min, to breakdown of the nuclear envelope was 91 min, and to restoration of the nuclear envelope was 116 min. The time spent in morphologically distinguishable phases of mitosis and the effects of 0.5, 1.0, 1.5, 2.0, and 4.0 Gy of gamma or X radiation on cells at each phase were determined. Affected cells were found to be delayed without or with reversion to an earlier mitotic stage before recovering and advancing through mitosis. Cells were timed in the five steps comprising delay with reversion: inertia, cessation I, regression, cessation II, and reprogression. No cells treated in late prophase, i.e., within 8-10 min of nuclear envelope breakdown, were delayed by the doses used; therefore the critical or transition point must be situated in middle prophase. Cells irradiated in this stage were not delayed by 0.5 or 1.0 Gy, but suffered a dose-dependent delay with or without reversion after 1.5, 2.0, and 4.0 Gy. Cells irradiated in early prophase and very late interphase responded similarly, but a greater percentage of the latter reverted.     

Regulation of mitosis onset and thymidine kinase activity during the cell cycle of Physarum polycephalum plasmodia: effect of fluorodeoxyuridine

The effects of fluorodeoxyuridine were investigated during three events of the cell cycle: S-phase, mitosis, and the cyclic synthesis of thymidine kinase in the synchronous plasmodium of the myxomycete Physarum. DNA synthesis was inhibited, and there was limited action on other macromolecular syntheses. When DNA synthesis was slowed down, onset of the following increase of thymidine kinase synthesis occurred at approximately the same time as in the control, but mitosis was blocked in a very early prophase stage and metaphase was never observed. These effects were suppressed when the action of fluorodeoxyuridine was prevented by the addition of thymidine to the medium. In agreement with the action of aphidicolin and hydroxyurea, these observations show that: 1) perturbation of the S-phase does not prevent the nuclei from entering a very early prophase stage, but it does prevent them from proceeding through metaphase; 2) blockage of DNA synthesis does not perturb the normal timing of the triggering of thymidine kinase synthesis; and 3) the signal that triggers the arrest of thymidine kinase synthesis is postmitotic and does not require extensive DNA synthesis. In contrast with hydroxyurea and aphidicolin, in the presence of fluorodeoxyuridine metaphase was not observed. Thus, the triggering of thymidine kinase synthesis is unambiguously dissociated from metaphase and postmitotic events. Because synthesis of thymidine kinase remains under the control of temperature shifts from 22 to 32 degrees C, a simple model of the cell cycle involving two regulatory pathways could account for the triggering of thymidine kinase synthesis, early prophase stage, and metaphase.