THE EFFECT OF THE CC1 MUTATION ON CELL CYCLEMORPHOGENESIS IN PARAMECIUM TETRAURELIA SINA M. ADLB.Sc., University of British Columbia, 1990A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCEinTHE FACULTY OF GRADUATE STUDIES(Department of Zoology)We accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLOMBIAMay 1992© Sina M. Adl, 1992In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)Department of Z.The University of British ColumbiaVancouver, CanadaDateDE-6 (2/88)Page iiABSTRACTThe ccl mutant is temperature sensitive and arrests the cell cycle at 34.4° C.The transition point for division under restrictive conditions defines the pointof commitment to division. We have examined the point of commitment to divisionin relation to the time course of stomatogenesis and micronuclear mitosis inboth synchronous and asynchronous cell samples. Our analysis shows thatstomatogenesis begins well before the point of commitment to division, whichoccurs at the 6-rowed oral primordium stage. At this stage micronuclei are inlate anaphase. Oral morphogenesis is arrested up to this stage in mutant cellsat the restrictive temperature, as are later stages of micronuclear division.These results are interesting because they suggest that micronuclear DNAsynthesis and initiation of mitosis occur normally in ccl mutants, even thoughmacronuclear DNA synthesis stops immediately on transfer to restrictiveconditions. The meiotic cell cycles do not require the ccl function. Thissuggests that ccl is required exclusively for the vegetative cell cycle and notfor micronuclear division or stomatogenesis during the sexual pathways.Page iiiTABLE OF CONTENTSABSTRACT^ iiLIST OF TABLES^ viLIST OF FIGURES viiACKNOWLEDGEMENT^ ixINTRODUCTION 1MATERIALS AND METHODS^ 7Stocks and culture^ 7Statistical procedures 7Experimental samples^ 8Chemostat cultures 9Cytological procedures^ 10Scoring stages of morphogenesis^ 12RESULTS^ 13A. Location of the point of commitment to division^131. Cell cycle location of the PCD in synchronous samples^132. Cell cycle location of the PCD in asynchronous samples 13B. Oral morphogenesis^ 141. Stages of oral morphogenesis^ 142. Timing of somatic and oral morphogenesis insynchronous samples^ 15Page iv3. Timing of oral and somatic morphogenesis inasynchronous cultures^ 164. Location of the PCD in relation to oral morphogenesisin synchronous samples^ 165. Relation between the PCD and oral morphogenesis inasynchronous samples^ 176. Oral morphogenesis in cells at the restrictivetemperature^ 18C. Micronuclear mitosis 211. Description of micronuclear mitosis^ 212. Timing of mitotic stages in synchronous samples^223. Timing of mitotic stages in asynchronous samples^234. Correlation of micronuclear stages with oralmorphogenesis^ 235. Micronuclear mitosis in cells at the restrictivetemperature^ 246. Does ccl interfere with micronuclear DNA synthesis^26D. The function of the ccl mutation in the sexual pathway^27DISCUSSION^ 311. Summary of results^ 312. The PCD and the duration of the committed interval^313. Timing of oral morphogenesis^ 334. Stabilization of the oral anlage 355. Micronuclear mitosis^ 36Page v6. The ccl gene function and stabilization of micronuclearmitosis^ 397. The ccl gene function in the meiotic cell cycles^408. Nature of the ccl mutation^ 40APPENDIX: Estimation of the cell cycle duration insynchronous cells^ 431. The effect of cell manipulation on division time^432. The timing of cell division in cells manipulatedonce and twice^ 433. Effect of cell cycle age and cell manipulation ondivision synchrony^ 444. Summary^ 455. The sister cell method as an estimator of cellcycle duration^ 456. Conclusion 46BIBLIOBRAPHY^ 75Page viLIST OF TABLESDifference in time of cell fission betweensister cells manipulated once and twiceDifference in median time of cell fissionbetween synchronous cell samples manipulatedonce and twiceEffect of cell age on fission delayTiming of stages of oral morphogenesis insynchronous samplesTiming of stages of oral morphogenesis inasynchronous samplesRelationship between stage V and PCDTiming of stages of mitosis in synchronoussamplesTiming of stages of mitosis in asynchronoussamplesThe micronuclear cycleMean diameter of micronucleiCumulative extinction of micronucleiin ccl cellsTable 1Table 2Table 3Table 4Table 5Table 6Table 7Table 8Table 9Table 10Table 114849505152535455565758Page viiLIST OF FIGURESFigure 1^The Paramecium cell cycle^ 59Figure 2^Selection and treatment of synchronous cellsamples^ 60Figure 3^Effect of cell manipulation on time of celldivision^ 61Figure 4^Timing of mitosis as estimated by the sistercell method^ 62Figure 5^The PCD as a function of cell cycle length^63Figure 6^Oral morphogenesis^ 64Figure 7^Stages of oral morphogenesis^ 65Figure 8^Location of PCD in relation to oralmorphogenesis^ 66Figure 9^Effect of the restrictive temperature on oralmorphogenesis^ 67Figure 10 Stages of mitosis 68Figure 11^Timing of stages of mitosis in synchronoussamples^ 69Figure 12 Comparison of staging of mitosis andstomatogenesis from the synchronous andasynchronous experiments^ 70Figure 13^The meiotic cell cycle and the first vegetativecell cycle^ 71Figure 14 Relationship between mitosis, stomatogenesisPage viiiand PCD^ 72Figure 15^Diagram of a cell immediately after thecommitment to division^ 73Figure 16 Activation of P34 cdc2 and initiation ofcytoskeletal re-organization at commitmentto division^ 74Page ixACKNOWLEDGEMENTI express my sincere gratitude to Dr. J.D. Berger for his support, guidanceand critique over the years. I also thank Dr. T. Grigliatti for hisencouragements and the members of my committee for help with themanuscript.Adl - Thesis^ Page 1INTRODUCTION:The cell cycle is a fundamental unit of cell growth and development. Itconsists of discrete events such as DNA replication and continuous processessuch as growth. The discrete events of the cell cycle are superimposed upon acontinuous growth period, in preparation for cell division. At cell divisionduplicated chromosomes and organelles are distributed to a pair ofmorphologically identical sister cells.During the growth period, aside from a general increase in cytoplasmiccontent (dictyosomes, ribosomes, endo-membrane network, mitochondria,proteins, etc.), certain morphogenetic events take place in sequence,including replication of the nuclear DNA, duplication of specializedorganelles (flagella, cytostome) and of microtubular organizing centres.Successive morphogenetic processes are tightly coupled by molecularinteractions regulating progression through the cell cycle. The end result isa duplication of the cell contents. Cell division is initiated at the end ofthe growth period. This final segment of the cell cycle involves separationof the duplicated set of chromosomes by mitosis into two daughter nuclei anddistribution of the cytoplasm and its organelles into two daughter cells bycytokinesis. The separation of the sister cells at the end of mitosis andcytokinesis is termed cell fission.Historically, the cell growth period has been sub-divided into threephases. The primary growth period during which cells accumulate cytoplasm isthe G1 phase. Sometime after the beginning of G1 cells replicate theirnuclear DNA. The period during which cell growth is accompanied by DNAreplication is termed S phase. Finally, the growth period following S phaseAdl - Thesis^ Page 2but prior to cell division is termed G2. At the end of G2, cell division(mitosis and cytokinesis) is initiated leading to fission of the sistercells.Two main cell cycle control points have been identified in variousorganisms (Berger, 1989; Enoch and Nurse, 1991). One occurs at the end of G1and regulates initiation of DNA synthesis (IDS) in the nucleus. The secondoccurs at the end of G2 and regulates initiation of cell division.Initiation of cell division in all eukaryotes involves activation of P34cdc2 protein kinase (Nurse, 1985; Enoch and Nurse, 1991; Pines and Hunter,1991). The kinase is maintained in an inactive state, throughout the growthperiod of the cell cycle, by a phosphorylated tyrosine residue. A complexsequence of positive and negative regulators (apparently protein kinases)maintains the P34 cdc2 protein in its phosphorylated state. At the transitionbetween G2 and cell division, P34 cdc2 is de-phosphorylated at the tyrresidue and associates with cyclins. This association forms a very activeprotein kinase which is thought to be responsible for the cascade of specificprotein phosphorylation observed during cell division.Cyclins form a group of related proteins whose concentration increasesduring the growth period (G1 to end of G2), but they are rapidly degradedafter the onset of cell division. Cyclins A and B associated with P34 cdc2are believed to have different distributions in the cell during cell division(Ackhurst et al., 1989; Bailly et al., 1989; Booher et al., 1989). Thesignificance of the two different complexes is not clear, but they may havedifferent substrates (Pines and Hunter, 1991).Adl - Thesis^ Page 3More recently, cyclins have been implicated at IDS, the transition fromG1 to S phase. In Saccharomyces cerevisiae G1 specific cyclins associate withP34 cdc2 at this transition point (Wittenberg and Reed, 1988) termed "start".Several criteria must be met for proper execution of "start" (Pringle andHartwell, 1981). These include accumulation of cell mass to a threshold leveland the presence of a threshold level of nutrient concentration. Execution of"start" commits the cell to DNA replication. If the nutrient concentration islow or if mating pheromones are present, "start" commits cells to meiosis orconjugation respectively. Although a "start" function is recognized in mostorganisms (Berger, 1989), the involvement of active P34 cdc2 with G1 specificcyclins has only been demonstrated in S. cerevisiae (Pines and Hunter, 1991).The G1 specific cyclins are not related to cyclins A and B. It is possiblethat IDS specific cyclins exist in other organisms as well (Pines and Hunter,1991).In Paramecium, the cell cycle is regulated from a single control pointthat occurs at about 90 min before separation of sister cells at cell fission(Berger, 1988). The main cell cycle events of Paramecium are illustrated inFig. 1. There are two independent S phases, one for each type of nucleus. Themacronucleus is a non-chromosomal nucleus and is the transcriptionally activenucleus of the cell. At G1 the macronucleus contains 865 copies of shortenedchromosome fragments rather than whole chromosomes (Freiburg, 1988). Once DNAsynthesis has begun it continues until cell division. The macronucleusdivides by amitosis, whereby the nucleus pinches in half and its contents areroughly equally distributed to daughter nuclei (Tucker et al., 1980). The twodiploid micronuclei are transcriptionally inactive and divide by closedAdl - Thesis^ Page 4mitosis prior to cell division. Before cell division the somatic componentsof the cell must also be duplicated. These include synthesis of a new oralapparatus (the oral anlage), formation of a second cytoproct, insertion ofnew basal bodies along the rows of basal bodies (kineties), and duplicationof the two contractile vacuoles. The cell cycle control point, located nearthe beginning of oral morphogenesis, commits the cell to division (Fig. 1).Several key decisions are made at this control point. Wild type cellstreated with metabolic inhibitors interfering with protein synthesis and DNAsynthesis show a transition point to drug sensitivity at 1.5 hrs beforedivision (Rasmussen, 1967). Cells exposed to inhibitors of protein or DNAsynthesis at regular intervals during the cell cycle, may or may not feed,but they never begin DNA synthesis, somatic morphogenesis or cell division.Cells treated within 1.5 hrs of cell fission proceed with cell division,although more slowly. Daughter cells of this division arrest in Gl. Theexperiments indicated that neither protein nor DNA synthesis are essentialduring the final 1.5 hrs segment of the cell cycle. Similarly, wild typecells subjected to nutrient downshift prior to 1.5 hrs before division,extend the duration of the growth phase. However, cells subjected to thedownshift within the last 1.5 hr of the cell cycle proceed with cell divisionunaffected, but the subsequent cell cycle is extended (Berger, 1986).Analysis of the cell mass and DNA content of cells subjected to nutrient downshifts at regular intervals throughout the cell cycle, indicates that themagnitude of the DNA increment to be synthesized in the next cell cycle isset at a point concordant with the point of commitment to cell division(PCD), about 90 min before fission (Berger and Ching, 1989). Similar toAdl - Thesis^ Page 5"start" in Yeasts, in Paramecium the PCD is also associated with commitmentto meiosis (Berger, 1989).Two alternative meiotic pathways exist in Paramecium. Conjugationinvolves pairing of cells of complementary mating types, followed by exchangeof gametic nuclei at fertilization. In absence of complementary mating typeor in older clones autogamy takes place. Autogamy (self-fertilization) isdifferent from conjugation only in that gametic nuclei are not exchanged, butfertilized within the cell (Wichterman, 1986). Cells challenged to entermeiosis by nutrient downshift before the PCD proceed with the sexual pathway.If cells commit to meiosis after the PCD, they will proceed with celldivision before entering meiosis (Berger, 1986; Berger et al., unpubl. obs.)Two non-allelic temperature sensitive mutations with a transition pointat the PCD have been obtained. Both mutations, ccl and cc2, prevent progressof the cell cycle "clock" at their restrictive temperature (34.4°) (Petersonand Berger, 1976). The cells feed and grow without progressing towards celldivision. DNA synthesis is not initiated, somatic morphogenesis does notoccur and cell division is not reached. However, if cells are incubated attheir restrictive temperature within 90 min of cell division, the cell cycleis not arrested, and cells proceed with division. This indicates that thewild type gene products have an essential function in cell cycle progression.The genes are required throughout the cell cycle until commitment todivision. The ccl mutation behaves differently from the classic cell cyclemutations (Hartwell, 1974). There is no sharply defined execution point atwhich the gene function is required. Under restrictive conditions ccl cellsdo not progress to a block point before the cell cycle is arrested; insteadAdl - Thesis^ Page 6all functions which require the ccl+ activity cease prior to the transitionpoint.This study set out to investigate the relationship between thepre-fission morphogenetic events and the PCD. The location of the PCD wasdefined as the transition point for the ccl mutation. The approach was toobtain synchronous ccl cell samples which were assayed for commitment todivision and for morphogenetic stages. Subsequently, the effect of theccl mutation on the relationship between the PCD, cell morphogenesis andmicronuclear mitosis were investigated. The results were compared to similarexperiments with unperturbed asynchronous ccl cultures. ccl cells at thepermissive temperature (27°) and wild type cells at 27° and at 34.4° wereused as controls.The results showed that morphogenetic processes are essentially completeprior to commitment to cell division. The micronuclei are in telophase,morphogenesis of the oral anlage is almost completed and the contractilevacuoles are duplicated when the PCD is reached. The ccl mutation interfereswith elongation of the micronuclear spindle, with stabilization of new basalbodies in the oral anlage, and with elongation of the cell prior to celldivision. Curiously, although macronuclear DNA synthesis does not proceed inthe mutant (Rasmussen and Berger, 1984) micronuclear DNA synthesis is notaffected. Morever, the mutation is not required during the sexual pathwayeven though the same processes take place.Adl - Thesis^ Page 7MATERIALS AND METHODS:Stocks and culture: Paramecium tetraurelia (Sonneborn, 1975) stock 51-Sand D4-1001 were grown in phosphate buffered cerophyl grass infusion (pH6.9), inoculated with R2ebsiella pneumoniae as the food organism (Sonneborn,1970). Stock d4-1001 was derived from wild type stock 51-S and carries thetemperature-sensitive mutation ccl (line 2A2 (Peterson and Berger, 1976)).Under restrictive conditions (34.4°C) macronuclear DNA synthesis and cellcycle progression are reversibly blocked (Peterson and Berger, 1976;Rasmussen and Berger, 1984; Rasmussen et al., 1986). Incubation of cellsunder restrictive conditions was carried out in a water-bath regulated to ±0.02 °C of the set temperature. Cells were placed in 0.5 ml of medium indisposable plastic microcentrifuge tubes which were totally submerged in thewater bath. All other experimental cells were kept at 27°C in an airincubator.Cell lines were transferred daily in isolation lines (Sonneborn, 1970).Post-autogamous cells from a flask culture were transferred to a petri plateat low density the night before each experiment to provide a culture inexponential growth phase for the next morning. Cells for mating experimentswere grown in tube cultures according to Sonneborn (1970). All other culturesand cell manipulations were carried out as described in Sonneborn (1970).Statistical Procedures: The median time of entry to developmental stageswas calculated by probit analysis using a recursive, variance-weightedregression procedure to obtain a maximum likelihood estimator for the medianand its variance (Finney, 1972). Standard errors and other statisticalAdl - Thesis^ Page 8procedures followed Zar (1984).Population density was estimated by manually counting cells in threerandom samples of known volume (0.05 ml) from the culture and calculating themean.Experimental cell samples: Synchronous cell samples were obtained byselecting dividing cells just before cell separation (Fig. 2). Eachsynchronous sample contained at least 30 dividers selected within 5 min.These cells separate into sister cells within one minute from selection. Thefirst and last sample selected were not used for experiments, but kept ascontrols to estimate synchrony of division. These cells were manipulated asecond time and placed individually in depression slide wells of medium, onecell per well, within one hour after division. They were examined frequentlyduring the latter part of the cell cycle and the cumulative number of cellscompleting division was plotted against time since fission. The median timeof fission was estimated by probit analysis. If the median times of fissionof these two synchronous samples were not within 10-15 min of each other theexperiment was discarded and repeated until consistant cell cycle durationswere obtained. The process of placing cells into drops produced a 10 minextension of the cell cycle (Appendix). The actual time of cell fission usedwas that of a synchronous cell sample manipulated once. The cumulativefrequency of divided cells in a control sample is the time of cell fissionfor the synchronous cell samples.A similar approach was used to estimate mean time of commitment to celldivision. The point of commitment to division was operationally defined asAdl - Thesis^ Page 9the point after which a shift to restrictive temperature did not blockdivision in ccl cells (Rasmussen and Berger, 1984; Rasmussen et al., 1986).Synchronous samples of 50 cells each were collected from exponential growthphase cultures. These were shifted to restrictive conditions at 20 minintervals during the latter half of the cell cycle. The fraction of cellsable to divide in each sample was scored one hour after a parallel sample hadcompleted division under permissive conditions. These values were plottedagainst time after fission. The median time of commitment to division, ofstages in oral morphogenesis and of micronuclear mitosis and their errorswere estimated by probit analysis.Chemostat cultures: The chemostat was used to obtain equilibriumcultures with cell cycles over 8 hrs in duration. Chemostat cultures wereestablished in a one-stage continuous culture apparatus (500 ml, Bellco). Theapparatus was assembled aseptically, inoculated with an exautogamousParamecium culture and continuously stirred. Antibiotics (penicillin, 125pg/ml; streptomycin sulfate, 125 pg/ml; gentamycin 5 pg/ml) were added tothe chemostat medium to suppress bacterial growth. Cell cycle duration ofcells in the chemostat culture was assumed to be equal to the reciprocal ofthe dilution rate, so that cell cycle duration is adjusted by changing therate of flow of medium. The chemostat culture is assumed to be in equilibriumwhen cell density reached a constant level.The point of commitment to division was determined from the fraction ofccl cells completing division under restrictive conditions. Twenty groups of40 cells each were withdrawn from the chemostat culture and immediatelyAdl - Thesis^ Page 10incubated at the restrictive temperature in 0.5 ml of fresh medium. Thenumber of cells at the end of a 2 hrs incubation was counted. The increase incell number was used to estimate the fraction of cells past the point ofcommitment to division at the time of the temperature shift. The fraction ofthe cell cycle between the point of commitment to division and fission wasthen estimated from the following equation:T=ln(1+freq)/1n2^(equation 1)whereT is the relative duration of the interval between commitment todivision and the end of the cell cycle, and freq is the frequency of cellsable to divide at the restrictive temperature (Mitchinson, 1971; Ching andBerger, 1986b).Cytological procedures: Basal bodies at the cell surface were stainedusing the Chatton-Lwoff silver impregnation procedure as modified by Frankeland Heckmann (1968). Synchronous cell samples were treated in depressionslide wells and were transferred by micropipette. Asynchronous cell sampleswere handled in a small centrifuge tube. The duration of several treatmentswas critical. Cells were killed by the addition of Champy's fixative. Thefixative was removed within 2 min and cells were kept in Da Fano's fixativeover 5 hrs but less than 20 hrs. The period of silver nitrate treatment was5-6 hrs. The slides were finally exposed to light from a bank of PhillipsTL-05 sun lamps at 25 cm for several hours.Routine staining of cells to determine the presence of ex-autogamouscells was carried out with Dippel stain (Sonneborn, 1970). Nuclei wereobserved by drying cells on albumen coated slides, followed by staining withAdl - Thesis^ Page 11Azure A (Berger, 1971). Micronuclear diameters were measured using amicrometer scale accurate to ±0.25 pm. A Zeiss CONTRON image analysissystem was used to estimate cumulative extinction of micronuclei stained withAzure A.Radiolabelling of nuclei followed Berger (1971).a) Preparation of labelled bacteria: E. coli strain 15T- are thymineauxotrophic. The bacteria were grown in standard M9 medium supplemented withthymine. An inoculum was added to M9(thy-) medium and the culture was grownover night to deplete thymine stores. These bacteria are centrifuged intofresh M9(thy-) to an O.D. of 0.1 and supplemented with 0.05 pCi/m1 ofH3-thymine (Amersham). When the culture reaches an 0.D. of 0.4, the bacteriaare washed with unlabelled M9(thy+) medium, by repeated centrifugations. Thelabelled bacteria are stored in cerophyll medium at 5°C.b) Labelling Paramecium nuclei: Paramecium cells were transferred bymicropipette into a depression well containing labelled cerophyll medium for1.5 hrs. Labelled bacteria were then washed off the cells by transferringeach Paramecium cell through three wells of unlabelled cerophyll medium.Finally the cells are dried on albumen coated microscope slides, the DNAstained with acriflavine, coated with photographic emulsion (Ilford K-5emulsion) and developed. The labelled nuclei show dense silver grains overthe macronucleus. There is a low background level of grains over thecytoplasm.Adl - Thesis^ Page 12Scoring stages of morphogenesis: The time course of oral morphogenesis,micronuclear mitosis and meiotic stages in synchronous samples was obtainedfrom plots of the cumulative frequency of cells having reached eachsuccessive stage against time. The median and its error were estimated byprobit analysis.Timing of prefission morphogenetic events in asynchronous cell sampleswas determined from the fraction of cells at each stage using the equationabove, to estimate the corresponding fraction of the cell cycle. Theseestimates were based on 500-1000 cells selected randomly from the chemostatculture. The relative duration of the terminal stage of morphogenesis wascalculated first. The terminal stage was then expanded to include eachsuccessive earlier stage to estimate the relative position in the cell cycleat which each stage began.Adl - Thesis^ Page 13RESULTS:Controls for the effect of cell manipulation on cell cycle duration arepresented in an appendix (Figs. 3,4 and Tables 1,2,3).A- LOCATION OF THE POINT OF COMMITMENT TO DIVISION:1- Cell cycle location of the point of commitment to division insynchronous samples:The PCD was determined from synchronous samples of ccl cells shifted tothe restrictive temperature at 20 min intervals. The fraction of cells able todivide at restrictive conditions was assumed to have been committed todivision at the time of temperature upshift. The PCD was estimated as themedian of the plot of the percentage of the cells which divided at therestrictive temperature against time since division (Fig. 5). For cell cyclesshorter than 6 hrs there is a significant regression of the post-commitmentinterval against time of cell fission (r 2=0.85). There is no significantregression for cell cycles longer than 6.5 hrs. The mean time of PCD is at89.7 min before fission with a range of 77 to 104 min. These values aresimilar to those reported by Ching and Berger (1986b) for cell cycles of 6.5to 30 hrs duration.2- The cell cycle location of the PCD in asynchronous samples:The PCD was determined from random samples of ccl cells obtained from achemostat culture in equilibrium. The commitment interval is calculated fromequation 1 for a range of cell cycle durations from 4.83 hrs to 22.5 hrs (Fig.Adl - Thesis^ Page 145).The duration of the commitment interval obtained from the asynchronoussamples are consistent with the results from synchronous samples. There is ashort post-commitment interval of 50.5 min in the 4.8 hrs cell cycle. Thisinterval lengthens with longer cell cycles: 83.8 min for a cell cycle of 6.47hrs and 84.0 min in a 22.5 hrs cell cycle.B- ORAL MORPHOGENESIS:1- Stages of oral morphogenesis:The following experiments investigated the location of the PCD withrespect to the beginning of oral morphogenesis. Commitment to oralmorphogenesis and to cell division are reported to occur at approximately thesame time (Berger, 1988; Kaneda and Hanson, 1974). To demonstrate whether oneoccurs before the other, the beginning of oral morphogenesis must besubdivided into several stages. In this study oral morphogenesis wassubdivided into nine defined stages (Figs. 6,7):Cells not undergoing oral morphogenesis are assigned to Stage O. Stage Iis characterized by a darkly staining endoral kinety (E.K.) beyond itsinterphase range (>20 basal bodies (Jones, 1976; Roque, 1956)). There is thefirst indications of an anlage basal body proliferation at the posterior ofthe E.K. (Jones, 1976; Porter, 1960; Yusa, 1957). Stage II begins when basalbodies are extending the anlage field alongside the E.K. and the paroral sacforms (Jones, 1976; Porter,1960; Roque, 1956; Yusa, 1957). The oral anlage isnow obvious and by the end of this stage, it extends the length of the rightwall of the vestibule. The E.K. becomes becomes less obvious through thisAdl - Thesis^ Page 15stage. The beginning of Stage III is marked by the absence of the E.K. Theanlage field becomes narrower during this stage and the basal bodies becomeorganized (Jones, 1976; Porter, 1960; Roque, 1956; Yusa,1957) as the oralprimordium curves into a hook. At this point basal bodies of the oralprimordium are organized in 3 rows (Jones, 1976). This characterizes Stage IV.Stage V corresponds to the 6 rowed anlage and the re-appearance of the E.K.Stage VI is recognized by a 12 rowed anlage. During this stage the anlageattains a "C" shape and migrates posteriorad. The fission zone first appearsat this stage, just anterior of the anlage as it migrates. Stage VII beginswhen the oral anlage reaches the posterior of the vestibule and opens to thecell surface. The proter and opisthe have a common vestibule during this stageand the peniculi of the opisthe are distinct. In Stage VIII the quadrulusforms, the proter and opisthe oral apparatus are just separated and thefission furrow is obvious. Cells in late fission furrow until the point ofdivision are in Stage IX. Newly divided cells are assigned to Stage X.2- Timing of somatic and oral morphogenesis in synchronous samples:Synchronous cell samples were fixed for silver staining at 20 minintervals. Each cell sample was then scored for the distribution of oralmorphogenesis stages. The cumulative frequency of cells at each stage wasplotted against time since division. The median time of each stage and of celldivision was determined by probit analysis (Table 4).Stage I of oral morphogenesis begins at a mean time of 0.63 through thecell cycle. Relative position of stomatogenesis as a fraction of cell cycleduration does not vary for different cell cycle lengths. Contractile vacuolesAdl - Thesis^ Page 16duplicate during stage II. They are separated as the cell elongates during thesucceeding stages. Supernumerary contractile vacuoles appeared transiently aspreviously reported by King (1954) but they disappeared by stage VII. Thefirst visible sign of somatic morphogenesis is the duplication of the basalbodies in the fission zone, observed at the beginning of stage VI.3- Timing of oral and somatic morphogenesis in asynchronous cultures:A random cell sample was withdrawn from a chemostat culture inequilibrium or from a flask culture in exponential growth phase andimmediately fixed for silver staining. The cells were scored for oralmorphogenesis stages. The stage frequency was used to calculate the relativeposition in the cell cycle of the onset of each stage, using equation 1. Theprocedure was repeated for a wide range of cell cycle durations (Table 5).The results show that relative position of stomatogenesis does not changewith increasing cell cycle duration. Oral morphogenesis lasts 165 min in a4.83 hrs cell cycle, but lasts 787 min in a 30.5 hrs cell cycle. Comparison ofthe stage distribution from synchronous and asynchronous experiments is good,but early stages appear to occur earlier in the asynchronous cultures andlater stages appear shorter. The source of this difference is not obvious, butit may be a consequence of the transformation of data by equation 1 or of cellmanipulation.4- Location of the PCD in relation to oral morphogenesis in synchronouscell samples:Adl - Thesis^ Page 17Cells are well into oral morphogenesis (stages IV-V), at the time ofcommitment to division (Fig. 5, Table 4).To determine more precisely where in oral morphogenesis commitment todivision occurs, synchronous samples of ccl cells were divided in twosubsamples each. At regular intervals, one subsample is incubated at therestrictive temperature until after the division of control cells at thepermissive temperature. The fraction of cells that had divided were used toobtain the PCD as previously described. The other sub-sample was fixed at thetime that the first sub-sample was temperature shifted, and prepared forsilver staining. Cells were scored for oral morphogenesis stages. The datawere then graphed as a plot of the percentage cumulative frequency of cells ineach stage and of the PCD against time since division (Table 6; Fig. 8). Oralmorphogenesis begins before the cells are committed to division. The PCDcorrelates very closely with stage V of oral morphogenesis. At this point thecontractile vacuoles have duplicated and are approximately equidistant; theoral anlage has three rows of basal bodies but the elongation of the somatickineties has not begun and there is no fission zone.5- Relation between the PCD and oral morphogenesis in asynchronoussamples:A random sample of ccl cells was obtained from an equilibrium chemostatculture. Twenty groups of 40 cells were isolated and incubated at therestrictive temperature for 3 hrs to determine the PCD (Methods). In themeantime a second random sample of cells were fixed for silver staining. Thesecells were scored for oral morphogenesis stage distribution and the temporalAdl - Thesis^ Page 18location of each stage and of the PCD, as a fraction of the cell cycle, werecalculated using equation 1 (Fig. 5, Table 5). The PCD occured at stagesIII-V. Despite the high repeatability of the data, this approach has lessresolution than the synchronous cell approach because of the uncertaintyassociated in estimating cell cycle duration and because of cell mortality atthe restrictive temperature.6- Oral morphogenesis in cells at the restrictive temperature:Since the PCD occurs after oral morphogenesis has already begun, it wasof interest to determine what happens to the oral anlage if the ccl functionis interfered with after initiation of oral morphogenesis, but before thecells became committed to division.In this experiment synchronous ccl cell samples were divided into threesub-samples each. One sub-sample was killed for silver staining at the sametime that a second is incubated at the restrictive temperature. This secondsub-sample was killed for silver staining at the end of the incubation period.The third sub-sample was kept at the permissive temperature and was killed atthe same time as the second. This allows one to score oral morphogenesis atthe time of heat treatment (sub-sample 1), at the end of the heat treatment(sub-sample 2), and for progress of oral morphogenesis at the permissivetemperature until the end of the heat treatment (sub-sample 3). The experimentwas repeated with four different incubation durations of 15 min, 20 min, 40min, 90 min.Results of the 40 min incubation indicate that cells shifted to therestrictive temperature before stage I do not initiate oral morphogenesis.Adl - Thesis^ Page 19Cells shifted during stages I-III were blocked with no further proliferationof basal bodies. Moreover the number of stainable basal bodies in the anarchicfield decreased and those remaining became more obvious. Cells shifted duringstage IV lost the ordered 3 rows of basal bodies and appeared as adisorganized field with few darkly staining basal bodies. When cells wereshifted during stage V, some proceeded with oral morphogenesis, but others didnot. The latter showed a disorganized field of basal bodies in the paroral sacwith or without an endoral kinety and proceeded with oral morphogenesis moreslowly. This led to a group of cells lagging behind those already at stage VIat the time of upshift. This observation is consistent with the PCD occurringduring stage V, whereby some of the cells will have committed to proceed withdivision and others will not have (Fig. 9). The data were obtained from threesub-samples of a synchronous group which was manipulated at the PCD-stage Vduring the experiment. There is an accumulation of cells at stages II-IVduring the incubation period, represented in the bottom 50% of the curve, incomparison to the progress of the sub-sample at 27°. The other 50% of thecells have continued with oral morphogenesis.Cells shifted to the restrictive temperature before contractile vacuoleduplication show that contractile vacuoles cannot duplicate at the restrictivetemperature. If the contractile vacuoles have already divided at the time oftemperature shift, they do not proceed with the migration along the dorsalsurface. The concurrent cell elongation during the pre-fission morphogenesisdoes not occur if cells are at the restrictive temperature before the PCD, atstage V. When the experiment was repeated with a 20 min or 90 min incubation,the results did not differ from those described above for the 40 minAdi - Thesis^ Page 20incubation. There was no change in disorganization or resorbtion of the anlageand the fission zone was not initiated. The shorter incubation of 15 min wasnot sufficient to disrupt the oral anlage as seen with the silver stain.This experiment was repeated with wild type cells as control. Wild typecells incubated for 40 min at 34.4° were not affected by the temperatureshift. All sub-samples proceeded normally with oral morphogenesis. Cells at34.4° were 10 min faster than control cells at 27° because of the highertemperature.A sham control was carried out where ccl cells were not temperatureshifted but simply manipulated into sub-samples at 27° before fixation.Again, the pattern and timing of oral morphogenesis was not affected.We conclude that the abnormal anlage morphology at the restrictivetemperature is due to the ccl mutation, which arrests morphogenesis up to thePCD, the transition point for the gene function. Both the distribution of themorphogenetic stages and the detailed analysis of the morphogenetic changesoccurring in the oral anlage, following transfer to restrictive conditions ascompared to the control cells, suggest that the stabilization of the anlagebasal bodies is associated with stage V. The results of these experimentsreinforce the conclusion that stage V of oral morphogenesis is tightly coupledto the PCD.An interesting observation was made when calculating the effectiveduration of the block imposed by temperature treatment. The median time ofeach stage for cells at the restrictive temperature was substracted from themedian time of that stage in cells at the permissive temperature. The valuesobtained indicated the duration the cells at the restrictive temperature wereAdl - Thesis^ Page 21prevented from proceeding with morphogenesis. This time can be compared to thereal duration of the heat treatment. In every case there is a 10 mindifference. The duration of the effective blockage was 10 min shorter than thereal duration of the incubation period. This indicates that there is a 10 minlag before cells respond to the restrictive temperature block. Therefore inthe earlier experiment with cells incubated at the restrictive temperature fordifferent lengths of time, the failure of the 15 min incubation to disruptoral morphogenesis can be explained. If cells required 10 min for thetemperature block to effectively arrest morphogenesis, the cells would onlyshow a 5 min lag compared to the 27° control cells. Such a brief timedifference in the stage distribution cannot be resolved by this procedure (seeAppendix).C- MICRONUCLEAR MITOSIS:Synchronous cell samples were dried on slides at regular intervals duringthe second half of the cell cycle, then stained with Azure A. Cells werestudied and a description of micronuclear behaviour from interphase todivision was prepared. Mitosis was defined in ten stages for scoring thesubsequent experiments.1- Description of micronuclear mitosis:The morphology and stages of mitotic micronuclei are described below (seeFig. 10). Interphase micronuclei are designated Stage O. Stage 1 begins whenone of the micronuclei moves away from the macronucleus and becomes enlarged.By Stage 2 both micronuclei have moved away from the macronucleus and haveAdl - Thesis^ Page 22become enlarged. At Stage 3 the micronuclei have enlarged to more than twicethe diameter of stage 0 nuclei. The characteristic Stage 3 nuclei have adoughnut shape, with a darkly staining perimeter around a small pale centralregion. Stage 4 micronuclei are more elongated and no longer circular. Stage 5begins when the micronuclei have elongated to more than 4.0 pm. Themicronuclei continue to elongate and form two pairs of darkly staining, almostparallel bands. During Stage 6 the nuclei elongate from 5.5 pm to 12.0 pm.This stage corresponds to chromosome separation along the spindle.Consequently the dark bands disappear and only dots representing thechromosomes can be seen. Stage 7 begins when the chromatin masses have reachedthe opposite ends of the spindles. Elongation of the spindle continues duringStage 7 and the chromatin masses condense. This stage correlates with thebeginning of macronuclear elongation. Stage 8 begins when the macronucleus hasbegun elongating along the dorsal length of the cell. The two pairs ofmicronuclei reach the opposite ends of the cell. Stage 9 begins withmacronuclear amitosis and lasts until separation of the daughter cells aftercytokinesis. During Stage 9 the micronuclei appeared similar to Stage 2micronuclei. Nuclei in newly separated daughter cells are assigned Stage 10 asthey have not yet located themselves adjacent to the macronucleus.2- Timing of mitotic stages in synchronous samples:Synchronous cell samples were dried on slide at regular intervals andstained with Azure A. These were scored and the cumulative distribution ofstages was plotted against time after division. The median time of each stageand of cell division was estimated by probit analysis (Table 7; Fig. 11).Adl - Thesis^ Page 23Micronuclei first begin to enlarge at about 170 min before division.Elongation of the mitotic spindle (stages 4-6) corresponds with the PCD.Telophase occurs during macronuclear elongation (stages 7,8). Macronuclearamitosis corresponds to micronuclei stage 9. As with oral morphogenesis, theduration of mitosis is proportional to cell cycle duration, so that individualstages occur at the same relative stage in cell cycles of different duration.3- Timing of mitosis in asynchronous cell samples:A random sample of wild type cells was obtained from a flask culture inexponential growth phase. The cells were concentrated by gentlecentrifugation, placed on albumen coated microscope slides and stained withAzure A. The slides were scored for distribution of micronuclear stages. Therelative cell cycle position of the stages were calculated from the frequencyof each stage as described above (Table 8). There is a persistent differencebetween the position of the early stages in the synchronous cell cyclescompared to the asynchronous cell cycles. Early stages of mitosis map toearlier positions in the cell cycle in the asynchronous cell samples. Thisdifference with the synchronous results is consistent with that alreadyobserved in the analysis of oral morphogenesis stages.4- Correlation of micronuclear stages with oral morphogenesis:To obtain a more precise relationship between oral morphogenetic eventsand mitosis, synchronous cell samples were subdivided in two sub-samples of 30cells each at regular intervals of 20 min. One sub-sample was fixed for silverstaining at the same time the other was dried on slide for nuclear staining.Adl - Thesis^ Page 24The slides were scored for cumulative distribution of stages and plottedagainst cell cycle time. The median time of onset of each stage was estimatedby probit analysis. The results are presented as the pooled mean of the stagedistribution obtained from several repeats of the experiment (Fig. 12).The data show that Stages 2-3 of mitosis occur during oral morphogenesisStage I. Micronuclear Stages 5-7 occur during Stages IV and V of oralmorphogenesis. Micronuclei in Stage 7 correspond with the end of Stage V oforal morphogenesis and the beginning of the fission furrow. These resultsindicate that micronuclei proceed with mitosis up to spindle elongation andchromosome separation (Stage 6) before the cell becomes commited to divisionat oral morphogenesis stage V.The same correlation exists between the stages of oral morphogenesis andmicronuclear mitosis as was obtained with the asynchronous samples (Fig. 12).Stage I of oral morphogenesis falls in the latter part of Stage 2 of mitosis.Stage III of oral morphogenesis is associated with Stage 5 of mitosis andStage V of oral morphogenesis coincides with micronuclear Stage 7. However theearly stages appear to occur much earlier in the asynchronous experiments. Thereason for this discrepancy is not clear.5- Location of the micronuclear S phase in the cell cycle:It has been previously established that micronuclear S phase occursmid-way in the cell cycle, from 0.48 to 0.70 and lasts a constant duration of80 min (Pasternak, 1967), (Table 9). Stages 1-3 of mitosis described here,coincide with this time interval. Since macronuclear DNA synthesis is arrestedimmediately when ccl cells are shifted to the restrictive temperatureAdl - Thesis^ Page 25(Rasmussen and Berger, 1984), it is of interest to determine whethermicronuclear DNA synthesis is also arrested by the mutation, and to ascertainwhether stages 1 and 2 correspond to the micronuclear DNA synthesis period.The following experiments were designed to determine the location ofmicronuclear S phase with respect to our stages and to determine whethermicronuclear processes up to the PCD require the ccl gene product.Synchronous samples of ccl cells were divided into two subsets of 30cells each at regular intervals throughout one cell cycle. One subset wasdried on slides for staining with Azure A; at the same time the other wasincubated at the restrictive temperature for 1.5 hrs then dried on slides.Each subset is then scored for progress of mitosis. The experiment wasrepeated with wild type cells as a control.Mitosis was normal in wild type cells at 34.4°. They were, on average,10 min faster than the cells kept at 27°, presumably because of the fastercell cycle progression at the higher temperature. At the restrictivetemperature ccl cells traversed stages 1,2,3 unaffected. However, they couldnot proceed with stages 4-8. Cells shifted during these stages were arrested.Stages 4-7 correspond to the elongation of the micronuclei and separation ofthe chromosomes. Cells shifted towards the end of stage 8 proceededunaffected. This indicates that ccl+ is required during mitosis for progressthrough anaphase and spindle elongation during telophase. This corresponds intime to oral morphogenesis stages IV-V which are also arrested as previouslydescribed (Fig. 9).Adl - Thesis^ Page 266- Does ccl interfere with DNA synthesis in micronuclei ?a) The diameter of micronuclei in the previous experiments were measured(methods) and the mean diameter of micronuclei for each stage was calculated,for both subsets at 27° and 34.4°, in ccl and wild type cells, (Table 10).Micronuclear size increase occurs during the period of DNA synthesis asreported by (Pasternak,1967). The size increase also occurs in cells at therestrictive temperature, whether they were incubated before or during stages 1and 2. This indicates that micronuclear size increase during stages 1-3 occursdespite the ccl mutation and could be due to DNA replication.b) To obtain a more convincing measure of DNA synthesis in micronuclei ofccl cells at the restrictive temperature, cumulative extinction readings ofmicronuclei were obtained. The cumulative extinction of the micronuclei wasassumed to be proportional to the DNA content, based on the intensity ofstain. Comparison of the cumulative extinction of micronuclei before, duringand after DNA synthesis in both wild type and ccl cells at both 27° and atthe restrictive temperature, indicate that the ccl mutation does not blockmicronuclear DNA synthesis (Table 11). Increase in DNA content begins withstage 1 and continues into stage 3. Telophase nuclei have the same DNA contentas interphase nuclei. However, the readings do not show a doubling incumulative extinction as would be predicted by a doubling of the DNA content.This is possibly because the stain is opaque to light and micronuclei are nottwo dimensional sheets, so some shielding effect is observed in threedimensional objects (Gledhill et al., 1966).c) To obtain more direct evidence of the ccl mutation not interferingwith micronuclei S phase, synchronous sets of ccl cells were labelled withAdl - Thesis^ Page 27H3 Thymidine during micronuclear DNA replication.Although the macronucleus of the cells were clearly labelled, themicronuclei could not be assayed in most cells. It was impossible to determinewhether a micronucleus was labelled over the background level of silvergrains.D-THE FUNCTION OF THE ccl MUTATION IN THE SEXUAL PATHWAY:Two types of sexual pathways occur in Paramecium. One involves pairing ofcells of complementary mating types and exchange of gametic nuclei(conjugation) before fertilization. The other is an alternative pathwayinvolving self-fertilization of gametic nuclei (autogamy). During the sexualcell cycles, the macronucleus disintegrates, to be replaced by a new nucleus,derived from a mitotic product of the fertilization. The new macronucleus (themacronuclear anlage) undergoes a process of chromosome diminution and DNAamplification to replace the old macronucleus (Freiburg, 1988). Also theexisting oral apparatus is resorbed until completion of fertilization when neworal structures are synthesized anew (Ng and Newman, 1985).We have shown that the ccl mutation blocks the oral anlage and mitosisduring the vegetative cell cycle. Both oral morphogenesis and micronucleardivisions occur during the sexual pathway (meiosis, gametogenesis,fertilization, macronuclear development, oral morphogenesis). In this sectionwe examine whether the ccl mutation is necessary for these processes in thesexual cell cycles.ccl cells of complementary mating types were induced to become matingreactive. A sample of synchronous conjugants were selected. Each sample wasAdl - Thesis^ Page 28split in two sub-sets at regular intervals. One sub-sample was incubated atthe restrictive temperature while the other was dried on slide. The sub-sampleat the restrictive temperature was dried on a microscope slide at the 8th hrafter initiation of conjugation. This corresponds to the time of developmentof new macronuclei (macronuclear anlagen) from micronuclei in wild type cells(Fig.13). ccl cells incubated at the restrictive temperature duringconjugation completed meiosis and anlagen differentiation normally. In asecond experiment cell samples were shifted to the restrictive temperature atregular intervals for 2 hrs then dried on slide and scored. The progress ofmeiosis at the restrictive temperature was compared to control cells kept at27° and dried on slides at regular intervals. Again, progress of stages ofmeiosis was the same in cells maintained at the permissive and restrictivetemperatures, as previously reported (Berger, 1986; Ng and Newman, 1984). Thelatter experiment was repeated with a cell culture undergoing autogamy(self-fertilization). Cells were not affected by the restrictive temperatureand reached the stage of anlage differentiation normally.The above experiments demonstrate that the ccl function is not requiredduring the sexual pathway. The Qc.1 mutation does not interfere with meiosis,fertilization, or with gametic and post-fertilization mitotic divisions, anddifferentiation of anlagen from the micronuclei. This is, at first, anunexpected result. The ccl gene interferes with macronuclear DNA synthesis,stabilization of oral basal bodies and spindle elongation only in thevegetative cycles. Thus a new question arises: when is the ccl gene functionagain required, after meiosis?Adl - Thesis^ Page 29Since macronuclear anlagen differentiate at the restrictive temperature,the ER1 gene function cannot be required for macronuclear development untillater. The cell processes occuring during the first cell cycle, afterdifferentiation of the nuclei, are: DNA amplification in the macronuclearanlagen, somatic and oral morphogenesis, micronuclear mitosis and basal bodyproliferation at the fission zone before cell division (Fig. 13). Thefollowing experiments demonstrate that the ccl function is required during thecell cycle, in preparation for the first division. Progress of the secondvegetative cell cycle is fully inhibited as previously reported.Synchronous mating pairs of ccl cells were separated in two sub-samplesat intervals. One sub-sample was killed at the time the other was incubatedunder restrictive conditions. At the end of the incubation period, a controlsample kept at the permissive temperature was also dried on slide as acontrol. One synchronous sample was shifted to the restrictive temperature at9.5 hrs after initiation of mating and was dried on slide for staining at21.75 hrs, after completion of the first post-conjugation cell division incontrol cells (fig. 13). Experimental cells held at the restrictivetemperature had not divided and the anlagen were faint in contrast withanlagen of control cells which had accumulated much more DNA. This blockage ofDNA accumulation and pre-fission morphogenesis is similar to that describedearlier in vegetative cell cycles (Peterson and Berger, 1976; Rasmussen andBerger, 1984; Rasmussen et al., 1986).Another cell sample was shifted to the restrictive temperature at 26.5hrs after initiation of conjugation. This time point occurs late in the firstcell cycle after conjugation, not long before the PCD (fig. 13). This sampleAdl - Thesis^ Page 30was incubated until after the second post-conjugation cell fission in controlcells, then dried on slide for staining. This cell sample did not undergopre-fission morphogenesis, as previously described for vegetative cell cycles.These results demonstrate that the QQ1 gene function is not requiredduring the sexual pathway for meiosis, gametogenesis, fertilization,post-zygotic nuclear divisions, or for stomatogenesis which occurs followingfertilization. The ccl function is necessary for cell morphogenesis inpreparation for the first mitotic cell division as well as for DNA synthesisin the macronuclear anlagen. These further indicate that regulation of spindleelongation and stomatogenesis at oral replacement are different in the sexualand vegetative cell cycles.Adl - Thesis^ Page 31DISCUSSION:1- Summary of results:The ccl gene function is required during the vegetative cell cycle forproper execution of oral morphogenesis and micronuclear spindle elongation,but it is not required for similar processes in the sexual pathway. Inaddition, the gene is necessary for maintainance of macronuclear DNA synthesisbut not for micronuclear DNA replication. This study also revealed newfeatures of the temporal relationship of cell cycle events (Figs. 13, 14) andcompares different approaches to estimating the timing of cell cycle events.Oral morphogenesis begins before the cell becomes commited to celldivision. Oral anlagen and contractile vacuoles are duplicated before the PCD.The oral anlage is unstable in ccl mutant at the restrictive temperature,before the PCD. Micronuclei also complete DNA synthesis and reach telophasebefore the PCD. The ccl+ product is necessary for elongation of the spindle,prior to the PCD. These processes occur normally in ccl cells at therestrictive temperature during the sexual pathway. Comparison of theexperiments with synchronous and asynchronous cells indicated a persistentdifference in the estimated timing of the cell cycle events. The early stagesof morphogenesis were expanded in the asynchronous data, whereas the laterstages of the cell cycle appear closer to cell fission.2- The PCD and the duration of the committed interval:The duration of the morphogenetic period is proportional to cell cyclelength in cycles of 6.5 hr or less (Fig. 5). In longer cell cycles theduration of the morphogenetic period is almost constant, as Judged by theAdl - Thesis^ Page 32constant interval between the PCD and cell division. In these longer cellcycles, the appropriate time reference for events during the morphogeneticperiod is time before cell division rather than time since the previousdivision, or the fractional stage in the cell cycle. In the shorter cellcycles (<6.5 hrs), fractional stage in the cell cycle provides the mostconsistent time base for morphogenesis.Previous estimates of the interval between commitment to division andcell fission are reported to be 90 min (Berger and Ching, 1989; Ching andBerger, 1986; Peterson and Berger, 1976). In these studies, the shorter cellcycles during which the committed interval is not fixed, was not sampled.The occurrence of an interval of fixed duration in longer cell cycles hasimplications on the regulation of post-commitment morphogenesis (Fig. 5).Since an increase in cell cycle duration reflects a proportionate decrease inthe rate of cell growth (cycle length proportional to 1/rate), then the linearextension of the committed interval relative to cell cycle length reflects aslower accumulation of substrates for the post-commitment morphogeneticperiod. The plateau at 6-7 hrs indicates that morphogenesis is no longerrate-limited by an accumulation of substrates, but can proceed faster than theproportionate decrease in growth rate. In the longer cell cycles (>6.5 hrs)the timer-like interval indicates that morphogenesis is limited by a fasterand different rate-limiting step. This rate limiting step could be thedirected assembly of the component sub-units from a pre-synthesized pool ofsubstrates which accumulate during the pre-commitment growth phase. Amorphogenetic period of fixed duration has also been reported inYetrahymena species for cell cycle lengths over 150 min (Antipa, 1980; NelsonAdl - Thesis^ Page 33et al., 1981; Suhr-Jessen et al., 1977). Accumulation of a pool of proteinsfor morphogenesis, before the assembly of the components as suggested above,has been demonstrated in Tetrahymena (Frankel et al., 1976).In light of the foregoing considerations, it is not surprising that themorphogenetic period prior to the PCD appears to be growth driven. Thevariable most strongly associated with commitment to division inParamecium is the synthesis of a standard amount of DNA (Berger, 1988). Thisoccurs because the cell cycle control mechanisms gate cells into division whenthey have accumulated a pre-set DNA increment, the magnitude of which isestablished very close to the time of commitment to division, and whichdepends on nutrient level and thus growth rate (Berger and Ching, 1989). Underuniform nutrient conditions, the DNA increment synthesized during each cellcycle remains constant, and independent of the initial cell mass and DNAcontent (Berger, 1979; Ching and Berger, 1986; Rasmussen and Berger, 1982;Rasmussen et al., 1986). Consequently the interval between initiation of DNAsynthesis and cell division can vary considerably.3- Timing of oral morphogenesis:Previous studies on the timing of cell cycle processes in ParameciumDrelied on the sister cell approach for controls. Therefore the timing is onlyroughly correct and the deviation from the mean is great (appendix, Fig. 3),(Gill and Hanson, 1968; Rasmussen, 1967)). Their description of cell cyclemorphogenesis included several steps into few broad time intervals.Kaneda and Hanson (1974) noted the beginning of the anarchic field (theinitial proliferation of basal bodies of the oral primordium) at about 0.75 inthe cell cycle and the presence of longitudinal rows of basal bodies in theAdl - Thesis^ Page 34anterior part of the anarchic field at about 0.85. In the present study thesestages were noted substantially earlier (Tables 4,5); initiation of theanarchic field at 0.63 and formation of longitudinal rows of basal bodies(stage IV) at 0.77. Jones (1976) on the other hand reports initiation of theanarchic field at 0.61 in the cell cycle, in agreement with this study. Gilland Hanson (1968) observed a transition point for actinomysin-D blockage ofcontractile vacuole pore duplication at 0.69 in the cell cycle which isconsistent with our observation of contractile vacuole duplication at 0.70(stage II) of the cell cycle.Tucker et al. (1980) reported the timing of the later pre-fission stagesbased on individual selected dividing cells. It is possible to correlate ourstages with morphogenetic stages shown in their diagram (stages undescribed).Their early fission stage corresponds to the latter part of our stage VI andis reported at 16 min before cell division. A stage equivalent to the end ofour stage VII was reported to occur 10 min before cell division and a stageequivalent to our stage VIII was reported at 7 min before division. Althoughtheir study does not take into account the change in timing associated withthe longer cell cycles, these times fit very well with our data fromsynchronous cell samples of short duration and from unperturbed asynchronouscultures of equivalent cell cycle duration. Nonetheless, the earlier stages ofmorphogenesis are expanded artifactually in our asynchronous data, and do notcorroborate the single cell and synchronous cell sample observations.Differences in timing of oral morphogenesis in synchronous andasynchronous samples are significant and not easily explained (Fig. 12). Theymay be a consequence of manipulation of cells after collection of synchronousAdl - Thesis^ Page 35sets of dividers, or of the transformation of data by equation 1. The timingof pre-fission morphogenesis is repeatable and shows a consistent pattern ofchange as the cell cycle length increases (Tables 4,5). It is not likely to beartifactual. The repeatability of the results within each procedure indicatesthe artifact is inherent to one or both approaches. We do not know the sourceof the discrepancy. However, while there is a difference in the timingobtained between synchronous and asynchronous approaches, the former is moreuseful in experimental analysis of the cell cycle and provides much greaterresolution of discrete time intervals.4- Stabilization of the oral anlage:Observations made on oral anlage development at the restrictivetemperature in ccl cells show a decrease in the number of stained basal bodiesin the paroral sac and a disorganization of the existing basal bodies (Figs.6,9). It is reasonable to assume that basal bodies which have completedinsertion at the cell surface and have established cross-linkages to othercortical structures are stable and would not be resorbed under restrictiveconditions, while those not yet stabilized are. The disruption of the linearbasal bodies in stages IV-V anlagen may also be a consequence of loss underrestrictive conditions of cross-linking components of the basal bodies of theoral anlage, which had not yet stabilized. The oral anlage of wild typecontrol cells at the restrictive temperature were unaffected. Consequently,blockage of oral anlage development in ccl cells is not a consequence of theheat treatment per se, but is a consequence of the developmental block imposedby the absence of ccl+ product at the restrictive temperature.Adl - Thesis^ Page 36Gill and Hanson (1968) observed a transition point for actinomysin-Dblockage of oral morphogenesis at 0.64-0.77 and for cell division at0.70-0.86. The broad range is a consequence of the sister cell method used toobtain estimates of cell cycle duration (Appendix). Nonetheless their valuesare within range of those obtained in this study. We observed oralmorphogenesis to stabilize during stages III-V (corresponding to 0.73-0.81)with a transition point for stability of the anlage at stage V (0.81).Similar studies with stabilization of the oral anlage have been carriedout in Tetrahymena species and Glaucoma chattonii (Frankel, 1964, 1966).Resorption of the oral anlage after a heat shock or under the influence ofmetabolic inhibitors has shown that stabilization of the oral anlage occurswhen there are already several ordered rows of basal bodies, and formation ofthe oral membranelles is complete. The relative stage in Tetrahymena isequivalent to the point of oral anlage stabilization observed here withParamecium (Frankel, pers. comm.).5- Micronuclear mitosis:Although oral morphogenesis in Paramecium had been well described byseveral authors, micronuclear mitosis remained little studied. The timing ofDNA replication in the micronuclei was described by Pasternak (1967) (Table9), but a descriptive morphology of mitosis was still lacking. Maupas (1889),studied mitosis extensively in a variety of Ciliates including Paramecium6species. He did not describe mitosis in the Paramecium aurelia group. Part ofthe difficulty is with the small size of the micronuclei of P. aurelia andtheir location adjacent to the large macronucleus which tends to "cloak" them.Adl - Thesis^ Page 37Hertwig (1889) drew diagramatically what he thought occured. His work wastranslated without modification by Sonneborn (1947). Stevenson and Lloyd(1971) carried out an electron microscopy study of mitosis in Parameciumaurelia. They proceeded without knowing the gross morphology of mitoticnuclei. Their photographs show chromosomes attached to a spindle within thenuclear envelope. Nonetheless they suggested that because of the unusualmorphology of closed mitosis in Paramecium, a modified terminology should besought.The description of mitosis presented here reveals an unusual morphology.It is difficult to assign a definite prophase and metaphase label to thestages. Although anaphase is clearly under way at stage 6 (Fig. 10), a truemetaphase plate and spindle are not observed. The peripheral distribution ofthe DNA in stages 3-5 suggests the chromosomes are closely associated with thenuclear envelope. This close association is lost at stage 6 when chromosomesare migrating to opposite poles of the spindle. It may be that metaphasesensu stricto does not occur in Paramecium. A closed mitosis without metaphaseplate and intra-nuclear spindle is called closed pleuromitosis. This would notbe unusual. The variety of mitotic behaviour in Protists is great. Even withinthe Ciliates, closely related species do not have an identical mitoticapparatus (Raikov, 1982).Pasternak's observations (1967) on the micronuclear DNA replicationperiod indicates that micronuclear S phase has a duration of 81 min andextends from 0.48 to 0.70 in cell cycles of 5.75-7.0 hrs long (Table 9). Theposition of the micronuclear S-period is proportional to cell cycle length.Adl - Thesis^ Page 38Comparison of his data with our results of micronuclei size increase (Table10) and photometric readings (Table 11) suggests that the micronuclei DNAreplication period begins Just before stage 1 and continues to stage 3. Thisis interesting because it indicates that there is no G2 period in themicronuclear cycle. The nuclei complete S phase and enter prophaseimmediately, since stages 4-6 correspond to elongation of the nuclei andchromosome separation. Historically, micronuclei were assumed to be inprophase when the nuclei are swollen (stage 2, this study) (Tucker at al.,1980). The evidence presented here indicates that a substantial part of theincrease in micronuclear size and stain density is likely a consequence of DNAaccumulation.Tucker at al. (1980) described briefly the relation between micronucleardivision and cell morphogenesis in individual randomly selected pairs ofsister cells (n=18). Their results differ substantially from those reportedhere. Their mitotic stages are diagrammed but undescribed. The micronuclei intheir stage 1 dividers could correspond to our stages 2 or 3 and were reportedat 22 min before cell separation. In our experiments mitosis was completeprior to 22 min before cell separation and stage 3 was observed to begin atabout 130 min before cell separation (Tables 7,8, Fig. 14). We do not know thebasis for this drastic difference. We repeated their procedure several timesand cells were stained for DNA with azure A. The micronuclei were repeatedlyfound to be in late telophase at 22 min before division. However we agree withTucker at al. (1980) on the timing of somatic and oral morphogenesis duringthe last 22 min of the cell cycle, as previously discussed.Adl - Thesis^ Page 396- The ccl gene function and stabilization of micronuclear mitosis:Nicronuclear mitosis is not arrested as abruptly as is oral morphogenesiswhen ccl cells are incubated at the restrictive temperature. The nucleicontinue to stage 4 at the restrictive temperature, although with slowerkinetics than control cells, before becoming blocked. The stage at which finalblockage of mitosis occurs coincides with stage V of oral morphogenesis and isthe point at which stabilization of the oral anlage and final commitment tocell division occur (Fig. 10). These observations suggest that linkage betweenmicronuclear mitosis and oral morphogenesis is fairly tight and may occurthrough processes mediated by the c_1+ gene product. Such a linkage betweenmicronuclear function and oral morphogenesis in Paramecium has beendemonstrated previously in amicronucleate cells (Ng and Newman, 1985).Amicronucleate cell lines form defective oral structures and cannot generatean oral apparatus of normal size. The defect is restored by re-injection of amicronucleus.A strong linkage between oral morphogenesis and the micronuclear cyclewas also shown in retrahymena where similar results have been obtained withmorphogenesis mutants (Frankel et al., 1976). Heat or cold shocks block ordelay mitosis when administered prior to anaphase (Gavin, 1965). Latertreatments are without effect. The basis for this instability to temperatureshocks is unclear. In this study both oral morphogenesis and micronuclearmitosis are linked prior to the transition point for ccl.Adl - Thesis^ Page 407- The ccl gene function in the meiotic cycles:Placing ccl cells at the restrictive temperature during the sexualpathway produces no abnormal phenotype. Cells proceed with meiosis andfertilization until differentiation of the nuclei into micronuclei andmacronuclear anlagen. Therefore the micronuclei DNA replication and spindleelongation are not affected during meiosis. The mutant cells are also capableof forming food vacuoles, indicating that a new oral apparatus is present andfunctioning. However the macronuclear anlagen cannot develop into amacronucleus subsequent to differentiation. The ccl gene function is thereforenot required during the sexual pathway. At restrictive temperature ccl cellscannot divide at the first mitotic fission and there is no macronucleardevelopment in this interval. It is reasonable to assume that the gene becomesnecessary for the cell cycle during this time (Fig. 13).8- Nature of the ccl mutation:The pleiotropic nature of the ccl mutation and its absolute requirementfor a number of major determinative events in the cell cycle suggest that thegene product is a regulative protein that participates in the control ofdifferent cell processes dependent on cell cycle progression. Protein kinaseshave been described in several organisms as having important regulatory rolesin the cell cycle. These include the p34 cdc2 kinase which functions duringGl, gating the cell into the vegetative replication pathway or alternativedevelopmental fates, and during G2 in commitment to division (Nurse, 1985).These regulative processes involve the interaction of a number of regulativemolecules and it is reasonable to assume that the ccl gene product inParamecium may be a component of such a regulatory system.Adl - Thesis^ Page 41It is possible to make less vague guesses concerning the nature of theccl gene. Despite the pleiotropic phenotype of the mutant gene, there is acommon link between the cell cycle events which are disrupted. The elongationof the mitotic spindle, basal body stability and alignment at stomatogenesisand at the fission zone all require microtubules and associated proteins.Duplication of the contractile vacuoles also requires microtubules. Migrationof the contractile vacuoles and of the stage VI oral anlage require formationof the cytospindle (Adoutte et al., 1989). The cytospindle is a sheet ofparallel microtubules and other cytoskeletal proteins which appears under thecell cortex before the fission furrow forms, during pre-fission cellelongation. Finally, there are microtubular and other fibrous cytoskeletalelements in the macronucleus (Tucker et al., 1980). The role of these tubulesin the macronucleus is not known but they may be essential for DNAreplication. The ccl gene may affect the stability or function of these andother microtubular structures.Upon commitment to division in yeasts and animal cells, the cellcytoskeleton disappears and reforms in a diffuse network (Barnes, 1990; Resh,1990). Some of these cytoskeletal elements become associated with the mitoticspindle for chromosome separation. Reorganization depends on prior commitmentto division. The molecular process leading to final commitment to celldivision involves phosphorylation of p34 cdc2 (Nurse, 1990) (Fig. 16). Anumber of substrates are activated by phosphorylation and cells proceed withmitosis. One of the direct substrates of the active p34 cdc2 complex is thecellular proto-oncogene of the Rous sarcoma virus (c-src) (Morgan et al.,1989; Moreno and Nurse, 1990; Parsons and Weber, 1989; Resh, 1990; Shenoy etAdl - Thesis^ Page 42al. 1989), which is known to phosphorylate calmodulin at mitosis (Fukami etal., 1986). The c-src protein is responsible for disassembly and re-assemblyof the cytoskeleton at mitosis and for maintaining the cytoskeleton anchoredto the cell membrane at interphase (Henderson et al., 1987; Parsons and Weber,1989). This may be mediated by calmodulin, since calmodulin is phosphorylatedby c-src, and calmodulin regulates microtubular (and consequentlycytoskeletal) stability (Dinsmore and Sloboda, 1988; Rasmussen and Means,1989. At mitosis, calmodulin, c-src and p34 cdc2 co-localize at the centromereduring spindle elongation (Rattner et al., 1990; Rasmussen and Means, 1989).Calmodulin mutants are known to arrest mitosis at metaphase (Rasmussen andMeans, 1989). c-src mutants prevent stable assembly of the mitotic spindle(Parsons and Weber, 1989). It is unlikely that ccl is a cyclin because thegene function is required throughout the cell cycle; cyclins accumulate foractivation only at particular stages of the cell cycle and they are rapidlydegraded afterwards (Nurse, 1985). It is also unlikely that ccl is P34 cdc2.This molecule has a key central role in initiating mitosis and shunting cellsinto vegetative or sexual pathways, and it is responsible for initiation ofDNA synthesis (Pines and Hunter, 1991). ccl is not such a regulatory molecule,but arrests these processes when in progress, as explained above. It is morelikely that ccl is closely associated with c-src or calmodulin-like calciumbinding protein because of their direct involvement with the cytoskeleton andmicrotubular stability.Adl - Thesis^ Page 43APPENDIXEstimation of the cell cycle duration in synchronous cells:1- The effect of cell manipulation on time of cell fission:The following experiment examines the effect of picking cells onsynchrony of cell fission. A divider was selected at random from a petri plateculture and placed in a depression well with medium (one manipulation). Thetime of the next fission of the sister cells was observed. The process wasrepeated with 20 cells and the mean difference between the fission times ofeach pair of sister cells was calculated. The experiment was repeated with 20more dividers selected at random but this time one of the two sisters wasmanipulated a second time, by being moved to a new depression well. The resultof the mean difference between fission after cell manipulation is shown inTable 1 and agree well with those reported by Gill and Hanson (1968).These results indicate that manipulation of cells during selection andtransfer into new wells of medium affect synchrony of cell fission and must betaken into account.2- Timing of cell fission in cells manipulated once and twice:To investigate further the effect of cell manipulation on cell fission,synchronous cell samples were manipulated after selection. Three synchronouscell samples were placed in depression wells (first manipulation). Onesynchronous cell sample was transferred to a new well after selection (secondmanipulation). Another cell sample was transferred at one cell per well intonew depression wells (second manipulation). The last cell sample was kept inthe same well and not manipulated a second time. The time of cell fission inAdl - Thesis^ Page 44each sample was obtained from a plot of the cumulative number of cells dividedat regular intervals. The experiment was repeated thrice. The result of atypical run is shown in Fig. 3. The pooled mean of the three runs are in Table2.There is clearly an induced fission delay in synchronous cell samplesmanipulated twice, and loss of synchrony as indicated by decrease in the slopeof the graph. The greater error in estimating the time of fission in cellsamples kept in a well is due to the difficulty in counting accurately thenumber of cells in the well at any one time. Although the cell samplemanipulated twice into one cell per well provides less error about the mediantime of cell fission, the error introduced by the fission delay makes it apoorer estimate of the true time of fission.3- Effect of cell cycle age and cell manipulation on division synchrony:The following experiment investigates whether cell cycle age affects theeffect of a second manipulation on a synchronous cell sample. Synchronous cellsamples (first manipulation) were transferred to new depression wells at onecell per well (second manipulation). The second manipulation were carried outat different times in the cell cycle, before the next fission. One cell samplewas kept as a control (manipulated once), and one sample of cells was placedat one cell per well as it was being selected as a second control. Theexperiment was repeated twice (Table 3).This experiment confirms the preceding finding that a second manipulationof synchronous cell samples has an important impact on the timing of cellfission. Furthermore, handling the cells late in the cell cycle, close to thetime of commitment to cell division, disrupts cells dramatically.Adl - Thesis^ Page 45An additional experiment was performed in which the cells of synchronoussamples were transferred to new wells with different amount of medium in thewells. There was no difference in the timing of cell fission in samples placedinto different volumes of medium.4- Summary:It is more accurate to determine the time of cell fission when cells areplaced singly, one cell per well. Unfortunately this second manipulationcauses an additional fission delay. The average delay caused is 10.2 min witha range of 4.5-16.0 min. The error in estimating the time of cell fission inonce manipulated samples is still far within acceptable limits to discriminatebetween cell cycle processes of 10-15 min duration. This latter methodprovides a more reliable estimate of cell cycle duration.5- The sister cell method as an estimator of cell cycle duration:Previously, work with timing of events in synchronous samples ofParamecium consisted of isolating one divider, keeping one of the two sistercells as a control to estimate the time of cell fission, while the other waskilled or manipulated. As demonstrated above, the mean difference in time ofcell fission between the two sisters is ±22 min, because one cannot knowwhich of the two cells is the faster one. The following experiment was carriedout to investigate the predictive validity of the sister cell method.Several groups of 21 synchronous cells each were obtained by selectingdividers. The two sister cells from each divider were separated and kept inadjacent wells. Thus, each sample consisted of 21 pairs of sister cells, eachin a separate well. The cells had been manipulated twice. At regular interval,Adl - Thesis^ Page 46one sister of each pair of cells in a sample was dried on slide for stainingwith azure A. The other sister was kept and the time of fission noted. Theslides are then scored for stages of mitosis, and the time before cell fissionof each stage for each cell in each sample was calculated from the time offission of its sister cell (Fig. 4).It is obvious that the sister cell method provides a very dispersedestimate of the exact cell cycle position of each stage. The estimates have arange of 1.5 hrs for stages which last only several minutes (eg. anaphase).Thus the sister cell method is not an accurate estimator of the timing of cellcycle events. A medium sample size as used here, was already quite labourintensive, requiring three people to carry out and is an organizationalnightmare to keep track of each cell's sister cell. Interestingly, the mediantime for each stage coincides roughly with those obtained more accurately inthis thesis with far less work.6-Conclusion:The sister cell method was inadequate and could not be used in estimatingthe cell cycle position of discrete events. The cell fission time obtainedfrom a random synchronous cell sample placed in wells singly is highlyaccurate, but causes a fission delay of 4.5 to 16.0 min. The time of fissionobtained by counting the number of cells dividing in a synchronous sample,with all the cells in the same depression well is less accurate (±8%) butdoes not cause a noticeable fission delay. The procedure adopted in this studyinvolves combining the timing of cell fission from twice manipulatedsynchronous cell samples (placed singly one cell per well), with the timing ofAdl - Thesis^ Page 47cell fission obtained from a once manipulated synchronous cell sample, asdescribed in the materials and methods.48error (t sin)NONCE TWICE ONCE TWICE13.1 23.4 13.2 22.20.20 0.80 2.51 6.3520 20 17 17CELLS MANIPULATED:difference in (min)at cell fissionbetween sister cellsTHIS STUDY^KANEDA AND HANSON (1975)Table 1: Difference in time of cell fission between sister cells'manipulated once and twice.49EXPERIMENT^Difference at cell fissionbetween cell samples (min.)A 16.5B 4.5C* 17.0Table 2: Difference in median time of cell fission between synchronous cellsamples manipulated once and twice. Cell samples were selected (firstmanipulation) and a sub-sample was manipulated a second time. The difference inmedian time of cell fission between both sub-samples is shown. C* : bothsub-samples were manipulated twice.50MANIPULATIONS^First^Second(control)TIME OF SECONDMANIPULATION(hrs)0.00 0.67 1.17 3.50 6.00INDUCED DELAY(min)0.00 9.50 8.10 8.10 57.0Table 3: Effect of cell cycle age on fission delay. A synchronous cell sample(first manipulation) was sub-sampled (second manipulation) at different timesduring one cell cycle. The difference in median time of cell fission betweensub-samples are compared to cells manipulated once.STAGE I STAGE II STAGE III STAGE IV STAGE V STAGE VIiSTAGE VII STAGE VIII STAGE IX FISSIONmin BD 71.28 44.41 31.36 25.33 19.47 14.95 8.64 3.00 4.45fraction 0.73 0.83 0.88 0.91 0.93 0.94 0.97 0.99min BD 89.75 69.52 53.78 43.70 33.57 20.43 12.51 9.00 6.00 4.65fraction 0.68 0.75 0.81 0.84 0.88 0.93 0.96 0.97 0.98min BD 84.12 78.00 72.08 71.73 55.20 32.98 24.00 18.00 12.00 5.05traction 0.72 0.74 0.76 0.76 0.82 0.89 0.92 0.94 0.96min BD 114.41 80.19 67.09 59.10 46.80 22.57 16.80 10.80 5.18fraction 0.63 0.74 0.78 0.81 0.85 0.93 0.95 0.97min BD 94.97 84.29 73.80 62.48 53.14 30.50 14.40 5.56fraction 0.72 0.75 0.78 0.81 0.84 0.91 0.96min BD 91.20 79.46 60.57 5.57fraction 0.73 0.76 0.82min BD 108.85 99.44 86.96 73.10 61.43 50.93 38.40 5.69fraction 0.68 0.71 0.75 0.79 0.82 0.85 0.89min BD 141.25 114.82 101.90 77.95 64.45 57.61 46.09 5.79fraction 0.59 0.67 0.71 0.78 0.81 0.83 0.87min BD 113.40 78.80 59.00 37.40 27.70 22.40 18.00 6.44fraction 0.71 0.80 0.85 0.90 0.93 0.94 0.95min BD 121.00 101.84 92.00 89.05 49.26 33.99 18.95 11.40 6.83fraction 0.70 0.75 0.78 0.78 0.88 0.92 0.95 0.97min BD 148.26 129.68 112.68 100.12 95.57 7.35fraction 0.66 0.71 0.74 0.77 0.78MEANfraction 0.67 0.71 0.75 0.79 0.82 0.87 0.91 0.95 0.97Table 4: Timing of stages of oral morphogenesis in synchronous samples.min BD: minutes before cell fission. Fraction: fractional age in cell cycle.STAGE I STAGE II STAGE III STAGE IV STAGE V STAGE VI STAGE VII STAGE VIII STAGE IX FISSIONmin BD 165.51 122.23 77.05 46.44 33.31 25.61 20.54 11.40 4.92 4.83fraction 0.43 0.58 0.73 0.84 0.89 0.91 0.93 0.96 0.98 n-1013min BD 176.57 112.23 67.00 47.91 37.46 28.04 22.30 15.85 7.33 5.63fraction 0.48 0.67 0.80 0.86 0.89 0.92 0.93 0.95 0.98 no1726min BD 377.12 252.33 183.97 122.15 79.14 51.46 38.32 19.31 5.83 14.12fraction 0.55 0.70 0.78 0.86 0.91 0.94 0.95 0.98 0.99 n-627min BD 786.92 504.03 327.07 178.92 141.58 120.61 99.48 48.04 21.95 30.50fraction 0.57 0.72 0.82 0.90 0.92 0.93 0.95 0.97 0.99 n-600Table 5: Timing of stages of oral morphogenesis in asynchronous samples.min BD: minutes before cell fission. Fraction: fractional age in cell cycle.CELL CYCLE^STAGE V^P.C.D.DURATION(hrs)^(min. before cell fission)4.65 34 334.85 49 455.05 55 445.73 61 765.69 61 76Table 6: Relationship between stage V and P.C.D.STAGE 1 STAGE 2 STAGE 3 STAGE 4 STAGE 5 STAGE 6 STAGE 7 STAGE 8 STAGE 9 FISSIONsin BD 167.4 136.7 113.0 5.62fraction 0.50 0.59 0.66min BD 174.9 145.7 139.8 116.1 96.9 83.1 76.4 50.6 5.70fraction 0.49 0.57 0.59 0.66 0.72 0.76 0.78 0.85min BD 163.0 127.5 79.8 5.72fraction 0.53 0.63 0.77sin BD 147.0 120.0 90.0 81.0 73.8 63.0 33.0 6.0 6.20fraction 0.60 0.68 0.76 0.78 0.80 0.83 0.91 0.98sin BD 94.0 82.4 73.4 66.0 54.8 18.2 6.44fraction 0.76 0.79 0.81 0.83 0.86 0.95sin BD 207.9 145.3 109.5 100.2 8.59fraction 0.60 0.72 0.79 0.81Table 7: Timing of stages of mitosis in synchronous samples. min BD:minutes before cell fission. Fraction: fractional age in cell cycle.STAGE 1 STAGE 2 STAGE 3 STAGE 4 STAGE 5 STAGE 6 STAGE 7 • STAGE 8 STAGE 9 FISSION■in BO 210.00 180.00 138.00 84.00 65.00 57.00 40.00 21.00 9.00 6.00fraction 0.30 0.40 0.54 0.72 0.78 0.81 0.87, 0.93 0.97 n=800Table 8: Timing of stages of mitosis in asynchronous samples. min BD:minutes before cell fission. Fraction: fractional age in cell cycle.56G1^So^Send Duration^Mitosis^G.T.(min.) (min.)^(min.)^(min.)^205.8^0.49^0.66^71.4^142.8^420187.2^0.48^0.70^85.8 117.0^390176.0^0.51^0.73^75.9^93.1 345162.2^0.47^0.71^82.8 100.0^345155.3^0.45^0.71^89.7^100.0^345195.0^0.50^0.71^81.9 113.1^390Table 9: The micronuclear cycle (modified from Pasternak, 1967)). G1: durationof growth phase; So and Send indicate the start and end of micronuclear DNAreplication, as a fraction of the generation time G.T. Mitosis: duration ofthe interval between Send and cell fission.57STAGE FRACTION0.170.200.320.340.380.40DIAMETERS^1.9910.18^1.1710.081.7410.07^1.0910.051.8010.06^1.3830.072.0710.09^1.3210.091.9710.08^1.4310.051.8710.07^1.3210.08AREA1.831.491.952.152.211.94N#1016221825150 0.45 2.1010.07 1.5030.09 2.47 181 0.47 1.9710.05 1.5010.07 2.30 211 0.47 2.2110.07 1.4810.07 2.57 281 0.49 2.14±0.09 1.3710.10 2.30 131 0.51 2.3810.10 1.7510.11 3.27 131 0.52 2.2510.08 1.4510.08 2.56 231 0.56 2.35±0.07 1.7910.07 3.30 282 0.60 2.4010.09 1.9410.07 3.65 212 0.60 2.1810.04 1.79±0.05 3.05 202 0.61 2.2910.13 1.5410.11 2.77 172 0.61 2.1810.08 1.6610.09 2.84 193 0.64 2.4810.08 1.83±0.09 3.64 283 0.67 2.2510.07 1.6710.08 2.95 224 0.72 3.1410.06 1.2310.03 3.01 285 0.75 3.9310.13 0.8810.04 2.73 98 0.91 2.0010.08 2.7610.08 2.76 289 0.92 1.5010.08 1.1710.06 1.40 2810 1.05 1.8910.04 1.26±0.03 1.26 56Table 10: Mean diameter of micronuclei (with standard error).58stage 0 (presynthetic)^stage 3 (postsynthetic)Temp.^permissive restrictive^permissive restrictive14^16^26^2513^15 26^2215^14^21^3114^13 23^22Mean ±SE^14.0 ±0.4 14.5 ±0.6^24.0 ±1.2^25.0 ±2.1Table 11: Cumulative extinction of micronuclei in ggIcells.MEEI D 1TIME 0MACM iCO.M.05 1II^59cvpcdFigure 1: The Paramecium cell cycle. Black: DNA synthesis; hatched: oraland somatic morphogenesis; CV: duplication of contractile vacuoles; PCD: pointof commitment to cell division; D: duration of cell division. Time is fractionof one cell cycle. (Modified from Berger, 1988).60 CONTROL 27 0TREATMENT 34.40DIVIDERSSELECTEDSYNCHRONOUSCELL SAMPLEINDEPRESSIONSLIDE WELLFigure 2: Selection and treatment of synchronous cell samples.615100 5,25 5850 5.75 6,00 6825 6850100.0090.00$0.001)70.00H4)^60.0050.00t)^60.0030.00.ri20.0011310.00fi^Imo()Time after Fission (III)Figure 3: Effect of cell manipulation on time of cell fission insynchronous sample. Open squares: once manipulated sample. Filled squares:twice manipulated sample.6276w0 54A1^o^1^IIAInn^11 11011^1 1AI^II mu I IR^II^1 11 1 1 I^nA-II^I^IIII1^133^2^1^OTIME BEFORE DIVISION (hr)Figure 4: Timing of mitosis as estimated by the sister cell method. Eachbar represents one cell scored at the stage indicated. The time is obtainedfrom the fission time of its sister cell. The arrows show the position of themedians. The open circles show the position of the medians obtained fromsynchronous samples in the text.63110 .00loo .00 —z 90.00 —0to 80.00 —torsi 70800060.00 —50,0040.00 —1,1 30.000.00 10100^20.00CYCLE DURATION (hr)N011.^ ■■30.00Figure 5: The PCD as a function of cell cycle length. Open squaressynchronous experiments. Filled squares asynchronous experiments.Figure 6: Oral morphogenesis. Description of stages in text. R: Stage Vanlage at the restrictive temperature. Scale bar: 10.0 pm.•Stage II •s1*^Stage hi•Figure 7: Stages of oral aorphogenesis. BO: buccal overture; ER: endoralkinety; P: peniculi; Q: quadrulus; RW: ribbed wall of cytostoue; RV: rightwall of vestibule; AF: anarchic field of basal bodies.661-=‘ 90zU 70050< 302 10125 100 75 50 25TIME BEFORE FISSION (min)Figure 8: Location of PCD in relation to oral morphogenesis. Filledcircles: cumulative stages of oral morphogenesis; open circles: position ofthe PCD.67F 9070a- 50Q 30 102 3 4 5 6 7 8 9STAGEFigure 9: Effect of the restrictive teaperature on oral aorphogenesis.Open circles: stage distribution at 27°. Open circles with dot: stagedistribution at 27°, 40 min later. Filled circles: stage distribution ofcells at the restrictive teaperature during the 40 min interval.68Figure 10: Stages of mitosis. Description of stages in text. Scale bar:1.0 pm.0TIME BEFORE FISSION(min)69Figure 11: Timing of stages of mitosis in synchronous samples.7 0sync.micsoralasync. 2mics +oral323 4567 8++^+ 44-^+++++-I— ++I^II III IV V VI VII VIII IX4 56 7 8 9+ ++ + + ++^+ + +III IV V VI IXa^ I0.4 0.5 0.6 0.7 0.8 0.9 1.0CELL CYCLE STAGEFigure 12: comparison of staging of mitosis and stomatogenesis from thesynchronous and asynchronous experiments.71timeQ^i0^fp^20^25^30 hrsferti Idiff^I1st^l2ndMEIOSIS IFiSSLON^-FISSIONFigure 13: The meiotic cell cycle and the first vegetative cell cycle.Hatched area: aal function required; fert.: fertilization of gametic nuclei;diff.: differentiation of nuclei into micronuclei and macronuclear anlagen.72••■1cdDNA syna tPn5 23t=Q gs'xi^61-Q---14 5 7 1310I3I0 5wX0^50^100CELL CYCLE STAGE (%)Figure 14: Relationship between mitosis, stomatogenesis and PCD. Circles:mean diameter and error of short and long axis of micronuclei. DNA syn:location of the DNA replication period. P: prophase. M: metaphase. A:anaphase. cd: commitment to division. T: telophase. Roman numerals: stages ofmitosis. Arabic numerals: stages of oral morphogenesis.73Figure 15: Diagram of a cell immediately after the commitment todivision. CV: contractile vacuoles. mac: macronucleus. mic: micronuclei. OA:oral apparatus. FZ: fission zone. The diagram shows duplicated and spacedcontractile vacuole. Mac is at the anterior of the cell, before itselongation. Mic are in early telophase (stage 7). OA is at stage VI, the 12rowed anlage. The fission zone is about to initiate basal body duplicationsand the fission furrow.74CDC 25^P13 SUC 1P34 CDC 2WEE^CYCLINSC -SRC" CALMODULI N"REORGANIZATIONFigure 16: Activation of P34 cdc2 and initiation of cytoskeletal reorganizationat commitment to division.Page 75REFERENCESAckhurst RJ et al. 1989. Intracellular localization and expression ofmammalian cdc2 protein during myogenic differentiation. Differentiation:40: 36-41.Antipa GA. 1980. A temporal analysis of cell cycle morphogenetic events inTetrahymena pyriformis. Acta Protozool. 19: 1-12.Bailly EM et al. 1989. p34 cdc2 is located in both nucleus and cytoplasm:part is centrosomally associated at G2/M and enters vesicles at anaphase.EMBO J. 8: 3985-3996.Barnes G et al. 1990. The cytoskeleton of Saccharomyces cerevisiae. CurrentOpinion in Cell Biology 2: 109-113.Berger JD. 1971. Kinetics of incorporation of DNA precursors from ingestedbacteria into macronuclear DNA of Paramecium tetraurelia. J. Protozool.18: 419-430.Berger JD. 1979. Regulation of macronuclear DNA content in Parameciumtetraurelia. J. Protozool. 26: 18-27.Berger JD. 1986. Autogamy in Paramecium tetraurelia: cell cycle stagespecific determination of the process. Exp. Cell Res. 166: 475-483.Berger JD. 1988. The cell cycle and regulation of cell mass and DNA content.In: Paramecium. Ed. GOrtz HD. Springer-Verlag, London NY BerlinBerger JD. 1989. The cell cycle in lower eukaryotes. Current Opinion in CellPage 76Biology 2: 256-264.Berger JD and Ching AS-L. 1988. The timing of initiation of DNA synthesis isestablished during the preceeding cell cycle as cells become committed tocell division. Exp. Cell Res. 174: 355-363.Berger JD and Ching AS-L. 1989. Commitment to division in Paramecium: effectof nutrient level on the macronuclear DNA increment. Exp. Cell Res. 182:90-112.Booher RN et al. 1989. The fission yeast cdc 2/cdc 13/suc 1 protein kinase:regulation of catalytic activity and nuclear localization. Cell 58:485-497.Ching AS-L and Berger JD. 1986. The timing of initiation of macronuclear DNAsynthesis is set during the preceeding cell cycle in Parameciumtetraurelia: analysis of the effect of abrupt changes in nutrient level.Exp. Cell Res. 167: 177-185.Dinsmore JH and Sloboda RD. 1988. Ca2+, calmodulin dependent phosphorylationof a 62 kD protein induces microtubule depolymerization in sea urchinmitotic apparatus. Cell 53: 769-778.Enoch T and Nurse P. 1991. Coupling M phase and S phase: controls maintainingthe dependence of mitosis on chromosome replication. cell 65: 921-924.Finney DJ. 1972. Probit analysis. 3rd ed. Cambridge University Press, London.Frankel J. 1964. The effect of high temperature on the pattern of oralPage 77development in Tetrahymena pyriformis GL. J. Exp. Zool. 155: 403-413.Frankel J. 1965. The effect of nucleic acid antagonists on cell division andoral organelle development in Tetrahymena pyriformis. J. Exp. Zool. 159:118-129.Frankel J and Heckmann K. 1968. A simplified Chatton-Lwoff silverimpregnation procedure for use in experimental studies with ciliates.Trans. Amer. Microsc. Soc. 87: 317-321.Frankel J et al. 1976. Causal relations among cell cycle processes inTetrahymena pyriformis: an analysis employing temperature sensitivemutants. J. Cell Biol. 71: 242-255.Freiburg M. 1988. Organization and expression of the nuclear genome. In:Paramecium. Ed. Gortz HD. Springer-Verlag, London NY Berlin.Fukami Y et al.. 1986. Phosphorylation of tyrosine residues of calmodulin inRSV transformed cells. Proc. Nati. Acad. Sci. 83: 4190-4194.Gavin RH. 1965. The effects of cold and heat shocks on cellular developmentin synchronized Tetrahymena pyriformis WH6. J. Protozool. 12: 307-319.Gill KS and Hanson ED. 1968. Analysis of prefission morphogenesis inParamecium tetraurelia. J. Exp. Zool. 167: 219-237.Gledhill BL t al. 1966. Changes in deoxyribonucleoprotein duringspermiogenesis in the bull. Exp. Cell Res. 41: 652-658.Page 78Hartwell LH. 1974. Saccharomyces cerevisiae cell cycle. Bact. Rev. 38:164-198.Henderson D et al. 1987. Cytoskeletal associations of PP60-src. Exp. CellRes. 168: 411-416.Hertwig R. 1889. Ober die Conjugationen der Infusorien. Abk. Bayer. Akad.Wiss. 17: 151-233.Iftode F et al. 1989. Development of surface pattern during division inParamecium: mapping of duplication and reorganization of corticalcytoskeletal structures in the wild type. Development 105: 191-202.Jones WR. 1976. Oral morphogenesis during asexual reproduction inParamecium tetraurelia. Genetic Res. Camb. 27: 187-194.Kaneda M and Hanson ED. 1974. Growth patterns and morphogenetic events in thecell cycle of Paramecium aurelia. In: paramecium: a current survey. Ed.Wagtendonk WJ van. Elsevier/North Holland Biomedical Press Amsterdam NY.King RL. 1954. Origin and morphogenetic movements of the pores of thecontractile vacuoles in Paramecium aurelia. J. Protozool. 1: 121-127.Maupas E. 1889. Le rajeunissement karyogamique chez les cilies. Arch. Zool.Exp. Gen. 7: 149-517.Moreno S and Nurse P. 1990. Substrates for P34 cdc2. Cell 61: 549-552.Morgan DO et al. 1989. Mitosis specific phosphorylation of P60 src by P34Page 79cdc2 protein kinase. Cell 57: 775-782.Nelson EM et al. 1981. Development of the ciliature of Tetralumenatiler-mop/lila. I. Temporal coordination with oral development. Dev. Biol.88: 27-39.Ng SF and Newman A. 1985. The role of the micronucleus in stomatogenesis insexual reproduction of Paramecium tetraurelia: micronuclei andstomatogenic events. Protistologica 20: 43-50.Nurse P. 1985. Cell cycle control genes in yeast. Trends in Genet. 1:51-52.Nurse P. 1990. Universal control mechanisms regulating onset of M-phase.Nature 344: 503-504.Parsons A and Weber B. 1989. The genetics of src. In: Current Topics in Microbiol. Immunol. 147. Ed. Vogt PR.Pasternak J. 1967. Differential gene activity in Paramecium aurelia. J. Exp.Zool. 165: 395-403.Peterson EL and Berger JD. 1976. Mutational blockage of DNA synthesis inParamecium tetraurelia. Can. J. Zool. 54: 2089-3002.Pines J and Hunter T. 1991. p34 cdc2: the S and M phase kinase? The NewBiologist 2: 389-395.Porter ED. 1960. The buccal organelles in Paramecium aurelia during fissionand conjugation with special reference to the kinetosomes. J. Protozool.7: 211-217.Page 80Pringle JR and Hartwell LH. 1981. The Saccharomyres cell cycle. In: The molecular biology of the yeast Saccharomyces. Eds. Strathern JN, JonesEW, Broasch J. Cold Spring Harbor Lab. NY.Rasmussen L. 1967. Effect of DL-p-Fluorophenylalanine on Paramecium aureliaiduring the cell generation cycle. Exp. Cell Res. 45: 501-510.Rasmussen CD and Berger JD. 1982. Downward regulation of cell size inParamecium tetraurelia: effects of increased cell size, with or withoutincreased DNA content, on the cell cycle. J. Cell Sci. 57: 315-327.Rasmussen CD and Berger JD. 1984. A gene function required for cell cycleprogression during the G1 portion of the cell cycle and for maintenanceof macronuclear DNA synthesis in Paramecium tetraurelia. Exp. Cell Res.155: 593-602.Rasmussen CD et al. 1986. Effects of increased cell mass and altered genedosage on the timing of initiation of macronuclear DNA synthesis inParamecium tetraurelia. Exp. Cell Res. 165: 53-61.Rasmussen CD and Means AR. 1989. Calmodulin is required for cell cycleprogression during G1 and mitosis. EMBO J. 8: 73-85.Rattner JB et al. 1990. p34 cdc2 kinase is localized to distinct domainswithin the mitotic apparatus. Cell Motil. Cytosk. 17: 227-333.Raikov IB. 1982. The protozoan nucleus. Springer-Verlag, Berlin London NY.Resh MD. 1990. Membrane interactions of PP60 v-src. Oncogenes 5: 1437-1445.Page 81Roque M. 1956. La stomagenese pendant l'autogamie, la coniugaison et ladivision chez Paramecium aurelia. C. R. Acad. Sci. 243: 1564-1581.Shenoy S et al. 1989. Purified MPF phosphorylates PP60 c-src at sitesphosphorylated during fibroblast mitosis. Cell 57: 763-778.Sonneborn TM. 1947. Recent advances in the genetics of Paramecium andEuplotes. Adv. Gen. 1: 263-298.Sonneborn TM. 1950. Methods in the general biology and genetics ofParamecium aurelia. J. Exp. Zool. 113: 87-147.Sonneborn TM. 1970. Methods in Paramecium research. Methods Cell Physiol. 4:242-335.Sonneborn TM. 1975. The Paramecium aurelia complex of fourteen siblingspecies. Trans. Amer. Microsc. Soc. 94: 155-173.Stevenson I and Lloyd FP. 1971. Ultrastructure of nuclear division inParamecium aurelia. I. Mitosis in the micronucleus. Austral. J. Biol.Sci. 24: 963-973.Suhr-Jessen P et al. 1977. Timing and regulation of nuclear and corticalevents in the cell cycle of Tetrahymena. J. Protozool. 24: 299-314.Tucker JB et al. 1980. Microtubules and control of macronuclear amitosis inParamecium. J. Cell Sci. 44: 135-147.Wichterman R. 1986. The biology of Paramecium. 2nd ed. Plenum Press NYPage 82London.Wittenberg C and Reed SI. 1988. Control of the yeast cell cycle is associatedwith assembly of the cdc 28 protein kinase complex. Cell 54: 1061-1072.Yamashiro S and Matsumura F. 1991. Mitosis specific phosphorylation ofcaldesmon. Bioessays 13: 563-571.Yusa A. 1957. The morphology and morphogenesis of the buccal organelles inParamecium with reference to their systematic significance. J. Protozool.4: 128-139.Zar JH. 1984. Biostatistical analysis. Second edition. Prentice Hall Inc.,N.J.