Note: All CDK and cyclin genes and their protein products from Paramecium tetraurelia mentioned in this report were named or renamed according to the newly proposed genetic nomenclature rules for ciliates (Allen et al., 1998).

 Introduction

The transitions between the phases of the cell cycle are controlled by a family of protein serine/threonine kinases termed cyclin-dependent protein kinases (CDKs) (for reviews see Fisher, 1997; Morgan, 1995; Pines, 1995). It is thought that CDKs phosphorylate a variety of substrates that are involved in the commitment of cells to cell cycle events such as DNA replication, nuclear envelope breakdown, chromosome condensation and cytokinesis (Peter et al., 1990; Chou et al., 1990; Satterwhite et al., 1992; Dutta et al., 1992). The kinase activity is regulated by at least three post-translational mechanisms: transient association with positive regulatory subunits known as cyclins, reversible phosphorylations/dephosphorylations at three conserved residues (Thr14, Tyr15 and Thr161 in human Cdk1) and binding to specific inhibitor proteins (CKIs) (Nigg, 1995; Toyoshima and Hunter, 1994; Xiong et al., 1993; Harper et al., 1993).

Cyclins were initially discovered in the cleaving eggs of marine invertebrates and were named after their pronounced synthesis and accumulation during interphase followed by abrupt destruction at metaphase/anaphase transition (Evans et al., 1983; Swenson et al., 1986; Standart et al., 1987). Cyclins have since been identified in a wide variety of species, and form part of the conserved mechanism of cell cycle regulation. They confer activity on the associated CDK partners at appropriate times during the cell cycle, and determine the subcellular localization and substrate specificity of these CDKs (Pines and Hunter, 1991; Peeper et al., 1993). Cyclins are diverse in sequence, sharing homology only over a region of about 100 amino acids that has been designated the ‘cyclin box’ (Pines and Hunter, 1989; Minshull et al., 1989). It is this region that is mainly involved in binding to the CDK subunits, although binding is also influenced by regions C-terminal to the cyclin box (Kobayshi et al., 1992; Lees and Harlow, 1993). So far, at least 14 cyclins have been described in vertebrates (Pines, 1996) and at least nine distinct cyclins are directly involved in the cell cycle regulation in the yeast Saccharomyces cerevisiae (Nigg, 1995). They are frequently classified as either G1 (START) or mitotic cyclins, according to sequence homology and expression pattern. While the G1 cyclins are involved in progression through G1 phase and at the G1/S transition, the mitotic A- and B-type cyclins are essential for progression through G2 phase and at the G2/M transition. In addition, A-type cyclins have been shown to play essential roles in DNA replication during S phase. A-type cyclins are synthesized and destroyed slightly earlier than B-type cyclins in M phase (Minshull et al., 1990; Pines and Hunter, 1990).

Under most circumstances, monomeric CDKs are inactive as protein kinases, depending on the binding of a cyclin for activation. The crystal structures of CDK2 monomer and CDK2/cyclin A complex, solved by De Bondt et al. (1993) and Jeffery et al. (1995), have revealed a molecular basis for this dependence. Cyclin binding results in reconfiguration of the bound ATP moiety, and allows part of the C-terminal lobe (T-loop) to move away from the active site where it blocks substrate binding.

Despite a substantial understanding of the molecular mechanisms of cell cycle regulation in yeast, invertebrates and plants, little is known about ciliates, a group of diverse unicellular organisms which emerged around a billion years ago, prior to the appearance of the higher eukaryotes, and which are phyletically well separated from the eukaryotic line that led to plants, animals and fungi. Like most ciliates, Paramecium tetraurelia contains two types of functionally distinct nuclei residing in the same cytoplasm: two transcriptionally silent diploid germ line micronuclei, and one transcriptionally active polygenomic somatic macronucleus. Because of this unique cellular organization, the Paramecium cell cycle is of considerable interest (Fig.1). Though in the same cytoplasm and of the same origin, the two types of nuclei undergo DNA synthesis and nuclear division at different time during the cell cycle by different modes. Macronuclear DNA synthesis begins at 0.25 of the cycle and persists through most of the cell cycle. In contrast, micronuclear DNA synthesis is restricted to a very short period immmediately preceding mitosis (Pasternak, 1967; Rasmussen and Berger 1982; Rasmussen et al., 1986; Adl and Berger, 1992). The macronucleus divides by amitosis while micronuclei divide by a typical ‘closed’ mitosis. Because of the relatively imprecise nature of the amitotic macronuclear division process, macronuclear DNA content is regulated through an incremental mechanism whereby all cells synthesize a certain amount of DNA regardless of their initial DNA content (Berger, 1979), suggesting that strict interdependence between M (or cytokinesis) and S (DNA replication) does not occur in vegetative Paramecium cells. [This previous sentence probably needs to be reworded and refined a bit] This differs from the situation of higher eukaryotic cells in which checkpoint controls act throughout the cell cycle to ensure that each DNA replicating origin can only be activated once per cycle and that cells do not enter mitosis until DNA replication is complete (Hayles et al., 1994). Furthermore, during the Paramecium sexual pathways, the DNA in the developing macronucleus undergoes approximately ten rounds of amplification without cytokinesis (Prescott, 1994 [Is this the right reference for Paramecium?]), a phenomenon which may contribute to the lack of dependence between macronuclear DNA synthesis and cytokinesis [Not clear to me how this would occur???]. However, critical cell cycle control points do occur in Paramecium, with the major one, the Point of Commitment to Division (PCD), located 90 minutes prior to fission during the preceding cell cycle. At this point, cells become irreversibly committed to division in the present cell cycle, the nature of the next cell cycle (meiotic or vegetative) is programmed, and the timing of IDS (Initiation of Macronuclear DNA Synthesis) is set if the following cell cycle is to be a vegetative cell cycle (reviewed by Berger, 1988).

[Don't you need to briefly review CDKs in Paramecium here? ...It is of interest to determine the molecular basis of these control events. Previous work has demonstrated the presence of three different CDKs in Parmecium() that seem to have activity patterns associated with different cell cycle stages. As CDKs are associated with cyclin regulatory units in virtually all other instances, it is of interest to analyse cyclins and their association with CDKs in Paramecium.] [I see that you have this material below. I think that slightly different transitions and some rearrangement of material would help here.]

We wish to know whether CDK/cyclin, the major cell cycle regulator in higher eukaryotes, is also conserved in Paramecium, and if so, what is its rolein the process of cell division and its coordination

In the last few years, CDK homologues have been identified in two ciliates. Three classes of p34^cdc2 homologues have thus far been identified in Paramecium tetraurelia, and the genes encoding two of them have been cloned (Tang et al., 1994, 1996, 1997; Zhang and Berger, in preparation), and two classes of anti-PSTAIRE-reactive polypeptides have been reported in Tetrahymena thermophila (Roth et al., 1991: Fujishima et al., 1992). In Paramecium, the histone H1 kinase activity peak at IDS and its exclusive localization in macronucleus suggest a role for Cdk1 in the macronuclear DNA synthesis while the kinase peak of Cdk3, which is associated with p13^suc1, coincides with the PCD (Tang et al., 1994, 1997). Cdk2 activity peaks during cytokinesis (Zhang and Berger, in preparation). However, we know little about the regulation of the activity of these proteins. The regulation of Cdk1 and Cdk2 kinase activity during the vegetative cell cycle does not involve oscillations in their transcript or protein levels. Major regulatory phosphorylation sites equivalent to Thr 14, Tyr 15 and Thr 161 in human CDK1, and which are shared among most CDKs, are conserved in both Cdk1 and Cdk2 in Paramecium. Regions previously shown to be involved in cyclin binding such as PSTAIRE region in protein kinase domain III and the threonine residue (Thr 161 in human CDK1) in domain VIII (Pines and Hunter, 1990) can also be found in both sequences. It seems possible that the regulation of the kinase activity of the Paramecium p34^cdc2 homologues might include binding of cyclins and phosphorylation/dephosphorylation. Therefore, we sought to determine whether cyclin homologues are present in Paramecium, and if so, whether and how they are involved in the Paramecium cell cycle regulation in association with CDK partners.

In this report, we describe the identification of two mitotic cyclin homologues, Cyc1 and Cyc2 from Paramecium tetraurelia, and present evidence for their roles in the vegetative cell cycle regulation.

Paramecium tetraurelia wild type stock 51-S was used in all experiments. Cells were grown in phosphate-buffered Cerophyl medium (Pines, Lawence, KS) supplemented with stigmasterol (5 m g/ml), and inoculated with Klebsiella pneumoniae as the food organism one day before use (Sonneborn, 1970). Synchronized cells were obtained by centrifugal elutriation in a Beckman JE.6-B centrifuge with a JE.5 rotor and 30 ml chamber (Beckman, Palo Alto, CA), as described by Tang et al. (1994, 1997). The synchronous cells were maintained at 27° C and sampled at one hour intervals starting 30 minutes post elutriation.

Unless other specified, chemicals were purchased from Sigma (St. Louis, MO.)

The following degenerate oligonucleotides were used to amplify the Paramecium mitotic cyclins by PCR: Sense primer: 5’-ATGA/CGAGCA/TATT/AT/CTA/GG/ATA/TGA-3’; Antisense primer: 5’-ATT/CTCT/CTCA/GTAT/C TTT/AG/CA/TT/AGC-3’. They correspond to two highly conserved regions within the mitotic ‘cyclin box’, MRAILV and ASKYEEI, respectively. Paramecium genomic DNA (1 m g) was used in a 50 m l PCR reaction consisting of 50 mM Tris-HCl (pH 8.0), 2.0 mM MgCl₂, 0.05% Tween 20, 0.05% NP-40, 200 m M each of dATP, dTTP, dCTP, dGTP, 100 pmol of each primer, and 2.5 U Taq DNA polymerase (Gibco-BRL, Gaithersburg, MD). After an initial 5 min denaturation at 94° C, 30 cycles of amplification were carried out (1 min at 94° C, 2 min at 50° C and 3 min at 72° C). PCR products of the predicted length (~200 bp) were isolated by agarose electrophoresis, purified with Qiagen Gel Purification Kit (Qiagen, Santa Clarita, CA), and subcloned into the plasmid pBluescript II KS +/- (Stratagene, Cambridge, UK). The resulting plasmids were subjected to DNA sequencing by the dideoxy-mediated chain-termination method (Sambrook et al., 1989).

Two libraries were employed to obtain the full-length sequence of CYC1a: a l gt11 cDNA library constructed using poly(A)⁺ RNA from vegetative Paramecium cells, kindly provided by Dr. James Forney (Purdue University); a genomic library constructed in l EMBL3 with Sau3A partially digested Paramecium genomic DNA which was a gift from Dr. Eric Meyer (Laboratoire de Genetique Moleculaire, Ecole Normale Superieure, Paris). Approximately 1,000,000 plaques from each library were screened with a ³²P-labeled DNA probe corresponding to the initial PCR fragment, as described by Sambrook et al. (1989).

3’ and 5’ RACE (Rapid Amplification of cDNA ends) (Frohman, 1990) were used to obtain both CYC1b and CYC2 cDNA sequences. For the 3’ end sequence, first strand cDNA was synthesized from total RNA by an oligo-dT₁₇ primer and then was used as template in the PCR reaction primed by oligo-dT₁₇ and a gene specific sense primer derived from the original PCR fragment. For the 5’ end sequence, first strand cDNA was made from total RNA by a gene-specific antisense primer, and then tailed with dAs with Terminal d Transferase (Gibco-BRL). The resulting cDNA was amplified with oligo-dT₁₇ primer and an internal gene-specific antisense primer.

The complete cDNA and genomic DNA sequences were finally isolated by PCRs using primer pairs derived from both ends of the sequence. DNA sequencing was carried out on both strands.

The ³²P-labeled DNA probe used in the library screen was prepared using a random hexamer priming kit from Boehringer Mannheim (Mannheim, Germany) in the presence of [a -³²P]dATP (Amersham, Arlington Heights, IL). DIG-labeled DNA probes used in genomic Southern blot analysis were prepared by PCR by including DIG-11-dUTP (Boehringer Mannheim) in the reaction mix, as described by An et al. (1992).

Thirty microgram of genomic DNA was digested with five restriction enzymes, EcoRI, HindIII, XbaI, PstI and BamH1 (Gibco-BRL), subjected to electrophoresis on 0.7% agarose gel and blotted onto Hybond-N⁺ membrane (Amersham) using downward transferring technique in 0.4 M NaOH (Koetsier et al., 1993). The membrane was hybridized to DIG-labeled DNA probe corresponding the ‘cyclin box’ region of CYC1a, CYC1b or CYC2, detected and stripped as described by Engler-Blum et al. (1993). The membrane was washed at 40° C and 65° C to provide low and high stringency conditions, respectively.

A BamH1-EcoRI CYC1a cDNA fragment, containing residues 1-68 of the coding region, was subcloned in-frame into a pGEX-2T vector (Pharmacia, Uppsala, Sweden), and the plasmids were transformed into bacteria. GST-Cyc1 protein synthesis was induced by IPTG, and the protein was purified from bacterial lysates by affinity chromatography using Glutathione Sepharose 4B (Pharmacia). GST-Cyc1 fusion protein was emulsified with RIBI adjuvant, and injected intramuscularly into rabbits. Approximately 0.7 mg of protein was used per injection. The antibody production was carried out as described by Harlow and Lane (1988).

A peptide derived from the carboxyl end of Cyc2, corresponding to resides 321-336 with an additional cysteine at its amino end (CQEVSRIRVERQIKQQK) was synthesized and coupled to KLH (keyhole limpet hemocyanin), and then used for injection into rabbits for generating anti-Cyc2 antibody.

Paramecium cells were lysed in 4 volumes of lysis buffer (50 mM Tris-HCl pH 7.5, 250 mM NaCl, 50 mM NaF, 5 mM EDTA, 1 mM DTT, 0.1% NP-40), supplemented with a cocktail of protease inhibitors including 50 m g/ml PMSF, 2 m g/ml leupeptin, 4 m g/ml aprotinin, and 1 m g/ml pepstatin A. The lysate was incubated on ice for 15 min, and centrifuged at 18,400 g for 15 min at 4° C. The supernatant was kept at -20° C. An aliquot was taken for protein quantitation using the Bradford methods (Bradford, 1976).

The proteins were resolved on 12.5% SDS-polyacrylamide gel (SDS-PAGE) and transferred to Immobilon-P membrane (Millipore, Bedford, MA). The membrane was first incubated with a blocking buffer (5% non-fat milk and 0.1% Tween-20 in PBS) for 1 h, and then with the primary antibody (1:500 dilution) for 1 h. After three washes in blocking buffer, the membrane was incubated with horseradish peroxidase (HRP)-conjugated donkey anti-rabbit IgG (1:4000 dilution), and the signals were detected by enhanced chemiluminescence (ECL kit, Amersham). The intensity of the cyclin bands on the X-ray film was quantitated using the public domain NIH Image program (developed at the U.S. National Institutes of Health)

For immunoprecipitation experiments, protein lysate was first pre-cleared with 50 m l of protein A-agarose beads (Gibco-BRL) (50% v/v in lysis buffer) for 2 h at 4° C. The supernatant was then incubated for 4 h with antiserum (5 m l per ml of lysate) or the corresponding pre-immune serum, followed by 1 h of incubation with 40 m l of protein A-agarose (50% v/v) with constant rotation at 4° C. Beads were pelleted by centrifugation and washed three times with lysis buffer. For immunoblotting, the pellets were resuspended in 20 m l of 2´ Laemmli sample buffer (Laemmli, 1970) and heated at 100° C for 5 minutes.

For histone H1 kinase assay, the pellets were further washed twice with 1´ kinase assay mix (50 mM Tris, pH7.5, 10 mM MgCl_2, 1 mM DTT, 40 m M ATP). Then, 10 m l of kinase assay cocktail containing 0.2 mg/ml bovine histone H1 (StressGen, Victoria, Canada) and 320 pmol (10 m Ci) of [g -³²P] ATP (Amersham) was added and the reactions were incubated at 27° C for 20 minutes. The reaction was terminated by addition of equal volume of 2´ Laemmli sample buffer. Aliquots were separated on SDS-polyacrylamide gel. Phosphorylated histone H1 was detected by autoradiography and quantitated by liquid scintillation counting of excised phosphorylated histone H1 bands.

By comparison of the amino acid sequences of known mitotic cyclins in the respective ‘cyclin box’ regions, two degenerate primers corresponding to conserved regions within the cyclin box region, MRAILV and ASKYEE, respectively, were synthesizedand used for PCR with Paramecium genomic DNA as template. Paramecium codon usage was taken into account to limit redundancy (Martindale, 1989). A 198 bp DNA fragment of expected size was consistently obtained, which was then gel purified and subcloned into pBluescript II KS+/-. DNA sequencing analysis revealed two distinct clones. The derived amino acid sequences of both fragments exhibited extensive homology with other known mitotic cyclins. They share about 71% identity at the amino acid level with each other. These two presumptive cyclin sequences were named CYC1 and CYC2, respectively.

The CYC1 DNA fragment was radioactively labeled and used in turn to screen a l gt11 cDNA library derived from mRNA of vegetative cells and a l EMBL3 genomic DNA library. Restriction mapping and DNA sequence analysis of positive clones isolated from libraries revealed an open reading frame of 972 bp, expected to encode a protein of 324 amino acids with a predicted molecular mass of approximately 38 kDa. BLAST search of GeneBank showed that the predicted open reading frame exhibited a high degree of homology to the mitotic cyclins. Thus, the protein encoded by CYC1 was named Cyc1, standing for the mitotic cyclin homologue 1 from Paramecium. Cyc1 is small compared to mitotic cyclins from other eukaryotes (45-60 kDa), but displays all the cyclin structural hallmarks (Fig. 2A).

Two independent primer extension analyses, using two different antisense primers derived from CYC1, demonstrated a further 290 bp from the presumptive initiation codon AUG to the 5’ end of transcript. An in-frame stop codon UGA was present upstream of the translational initiation codon AUG, and no other in-frame AUGs upstream in between (Fig. 2A). This corresponds well to the size of the transcript revealed by Northern blot analysis with the CYC1 probe, in which a single ~1.3 kb transcript of CYC1 was detected in exponentially growing cells, but not in starved cells (data not shown). These results suggest that the CYC1 sequence obtained is full-length.

The full-length cDNA sequence of cyc2 was obtained by a combination of 5’ and 3’ RACE (Frohman et al., 1990). It contains an open reading frame of 1008 bp and it encodes a protein of 336 amino acids which is 12 amino acids longer than Cyc1 (Fig. 2B). Its predicted molecular mass is 40 kDa. Cyc2 exhibits 48.8% overall identity with Cyc1, and 65% identity within the ‘cyclin box’ region.

Comparison between respective cDNA and genomic DNA sequences revealed that CYC1 contains two small introns, one is 23 bp and the other 24 bp while cyc2 has only one intron of 23 bp (Fig. 2A, B). Such small introns are common in the Paramecium genome (Russell et al., 1994). All these introns have consensus 5’/3’ splice sites GT/AG, but no apparent consensus branch point. They resided just outside of ‘cyclin box’ region, the most conserved region within cyclin sequence. The location of cyc2 intron corresponds to that of the second intron in CYC1.

The abrupt destruction of cyclins at mitosis involves a ubiquitin-dependent proteolysis that requires a short sequence motif, R´ ´ L´ ´ I´ N, the so-called ‘destruction box’ (Glotzer et al., 1991), usually located close to the N-terminus. In Paramecium Cyc1, the sequence, RCFGKEIAN, corresponding to residues 7-15, bears a single amino acid substitution (Lysine to Glycine) with respect to the destruction box consensus sequence. Similarly, in Cyc2, RFFGKELVN, in the corresponding region also shows limited similarity to the destruction box. However, whether they are actually involved in the cyclin destruction during the Paramecium cell cycle needs further investigations.

Figure 3 shows an alignment of the Paramecium Cyc1 and Cyc2 sequences with A- and B-type cyclins from other eukaryotes in the cyclin box region. Cyc1 exhibits 42-51% identity with both A- and B-type cyclins within the cyclin box region (Table I). The Cyc1 sequence differs in 19/70 positions from the consensus established previously for A-type cyclins, as opposed to 25/69 positions from the consensus for B-type cyclins (Hata et al., 1991). As for Cyc2, it differs in 21/70 positions from A-type cyclin consensus sequence whereas 24/69 from B-type cyclin consensus sequence. Cyc1 and Cyc2 are almost equally related to cyclins of the types A and B in the cyclin box region. Therefore, Cyc1 and Cyc2 may define a novel and distinct type of cyclin (also see below). A similar situation was also noticed previously for plant cyclins (Hata et al., 1991; Hirt et al., 1992).

As an independent pursuit of the complete sequence of CYC1, 5’ and 3’ RACEs were carried out. DNA sequence analyses of RACE-PCR products revealed another sequence which displayed overall 92.3% homology at the amino acid level and 85.9% homology at the nucleotide level with CYC1 that was cloned by library screening approach. Of 124 altered codons, 90 were silent changes and 34 resulted in amino acid changes, of which 13 were conservative substitutions (the sequence is in GenBank, accession number AF052486). To distinguish them, we designated the gene cloned from the libraries as CYC1a and the other as CYC1b. The corresponding protein products were named Cyc1A and Cyc1B, respectively.

A Southern blot analysis of genomic DNA digested with five restriction enzymes, EcoRI, HindIII, PstI, XbaI and BamH1, was carried out with CYC1a and CYC1b probes from the cyclin box region under high- and low-stringency conditions (Fig. 4). No recognition sites for any of these enzymes were found in the sequences. Single hybridizing bands were detected in all five restriction digests under high-stringency conditions. Under low-stringency conditions, on the other hands, two bands were visible in each lane. The weaker hybridizing bands detected with CYC1a under low-stringency conditions corresponded to the bands detected with CYC1b under high-stringency conditions, and vice versa. Therefore, Paramecium tetraurelia appears to carry one copy for each of the CYC1a and CYC1b genes.

When the blot was stripped and re-probed with the CYC2 probe corresponding to the cyclin box region, single hybridizing bands were detected in each lane under both high-and low-stringency conditions. We conclude that the CYC2 genes is also present as single copy genes in the Paramecium genome. Moreover, the discrete patterns of hybridization for CYC1a, CYC1b and CYC2 indicate that they are not tandemly arranged in the genome.

As CYC1a and CYC1b have the same 5’ and 3’ flanking sequences and high degree of homology, it is likely that they represent two isoforms of the same gene. Whether CYC1a and CYC1b have distinct or redundant functions is not known yet. It is possible that CYC1a and CYC1b originate during the massive deletions, fragmentation that occur during the macronucleus development. Similarly, CDK1 from Paramecium was also observed to have two closely related isoforms (Tang et al.,1995).

No other hybridizing bands were observed with CYC1a, CYC1b and CYC2 probes when washes were done at 37° C (data not shown), indicating the absence of additional cyclins with strong cyclin box sequence similarity with CYC1 and/or CYC2. These results, however, do not exclude the possibility that there are other cyclin families in Paramecium that can not be detected by hybridization.

To characterize the cyclin gene products, antisera were prepared against Cyc1A and Cyc2, respectively, as described in Materials and Methods. Since Cyc1A and Cyc1B have high degree of homology with each other, we referred to the antibody against Cyc1A as anti-Cyc1 antibody. On an immunoblot of total lysate from exponentially growing Paramecium cells, anti-Cyc1 antibody recognized a protein of apparent molecular weight of 38 kDa, while anti-Cyc2 recognized a protein of 40 kDa. Both their sizes agreed with the predicted molecular masses determined from respective primary sequences. These proteins were not detected when pre-immune serum was used (Fig. 5A and D). The immunoreactivities were specific to their cognate antigens, since preincubations of the immune sera with antigens at 4° C for 3.5 h abolished the appearance of the bands (data not shown). Sometimes, anti-Cyc2 serum also detected a faint band of 45 kDa on the western blot (Fig. 7B, lysate lane), however, this band did not exhibit any cell cycle-dependent fluctuation in its abundance, indicating that it may be due to non-specific reaction of the anti-peptide antibody.

To determine whether Cyc1 and Cyc2 display a periodic pattern of synthesis and destruction during the vegetative cell cycle, Paramecium cells were synchronized by selecting a population of the smallest, newly divided daughter cells, predominantly in early G1, using a centrifugal elutriation rotor. As shown in previous work, the extent of synchrony obtained by elutriation was comparable to that of hand-selected dividers from the same culture (Tang et al., 1997). Following the elutriation, these cells were then re-cultured into fresh medium and allowed to proceed through one relatively synchronous doubling. Medium time of cell division in elutriation sample ranges from 8.5 to 9.5 h post elutriation (Fig.5A). Protein samples were prepared at 1 h intervals and subjected to immunoblot analysis. As shown in Figure 5, both Cyc1 and Cyc2 protein levels varied during the cell cycle. Cyc1 level was low immediately after elutriation, began to increase at about 5.5 h, reached a peak at ~7.5 h, and then decreased gradually upon the onset of cytokinesis (Fig. 5A, C). The timing of the peak was observed at 1~2 h before most cells entered cytokinesis, coinciding with the point of commitment to cell division (PCD). On the other hand, Cyc2 was almost not detectable until 6.5 h after elutriation. The maximal level of Cyc2 expression was found late in the cell cycle after most of cells had undergone cytokinesis (Fig. 5D, F). By this time, Cyc1 level has declined. By the time when cells finished division, Cyc2 dropped abruptly to an almost undetectable level.

As expected, both Cyc1 and Cyc2 protein levels decreased dramatically in the starved cells (Fig. 6), suggesting that the roles of Cyc1 and Cyc2 are associated with cell proliferation.

To determine whether a physical interaction occurred between Paramecium cyclins and CDKs, we first examined the ability of Cyc1 and Cyc2 antibodies to coimmunoprecipitate CDKs. Both Cyc1 and Cyc2 immunoprecipitates were separated by SDS-PAGE and analyzed by immunoblot analysis for the presence of Cdk1 and Cdk2 with specific antisera. We found that Cdk2, but not Cdk1 coprecipitated with Cyc2 (Fig.7A), and that neither Cdk1 nor Cdk2 coprecipitated with Cyc1. However, when the same blot was probed with anti-PSTAIRE antibody, a protein of ~34 kDa was identified in the Cyc1 immunoprecipitate, suggesting that Cyc1 associates with a CDK other than Cdk1 and Cdk2 (data not shown). Furthermore, a cell cycle-dependent histone H1 kinase activity was detected in the Cyc1 immunoprecipitate, confirming that a CDK was associated with Cyc1 (see below).

We found that Cyc1 was present in the p13^suc1 bound fraction (Fig. 7C). And since Cdk3 binds to p13^suc1 (Tang et al., 1994), it is inferred that Cyc1 associates with p13^suc1 by virtue of its association with Cdk3. However, no direct evidence is available so far to support this possibility.

Furthermore, the converses experiment was also carried out to confirm the association of Cyc2 with Cdk2. Cdk2 immunoprecipitate were probed with Cyc1 and Cyc2 antibodies, only Cyc2, not Cyc1, was detected (Fig. 7B).

Since all three CDKs in Paramecium have been demonstrated in the previous studies to have histone H1 kinase activity (Tang et al., 1994, 1997; Zhang and Berger, in preparation), we assayed for histone H1 kinase activity in Cyc1 and Cyc2 immunoprecipitates to determine if cyclin forms active complexes with CDK in Paramecium.

Both Cyc1 and Cyc2 immunoprecipitates displayed kinase activity towards bovine histone H1, whereas immunoprecipitates using preimmune sera did not have such activity (Fig. 5B, E). Since Cyc1 and Cyc2 do not contain any known protein kinase sequences, it is almost certain that associated p34^cdc2 homologues are the kinases per se (Hanks et al., 1988).

To determine if the histone H1 kinase activities in the cyclin immunoprecipitates were cell cycle-dependent, Cyc1 and Cyc2 immunoprecipitates were prepared from lysates of synchronous cells by elutriation as a function of position in the cell cycle and assayed for histone H1 kinase activity. As expected based on the periodicities of the Cyc1 and Cyc2 protein amounts, a kinase activity peak for the Cyc1 immunoprecipitate was observed at around PCD (Fig. 5B), while the Cyc2 associated kinase peak was at the end of the cell cycle (Fig. 5E).

Figures 5C and 5F show the quantitation of Cyc1 and Cyc2 protein levels and their respective histone H1 kinase activities. The Cyc1 immunoprecipitate showed a peak of kinase activity about 2 h earlier than that of the Cyc2 immunoprecipitate. The Cyc1 associated kinase activity had fallen to 15% of its maximal level by the time the kinase activity of Cyc2 immunoprecipitates reached its peak. These results suggest that the Paramecium cyclins can form active complexes with CDKs and distinct CDK/cyclin complexes may fulfill different functions in the cell cycle regulation.

Given that great evolutionary diversities among the ciliates, it would be very interesting to see if cyclin-like sequences present in other groups of ciliates than Paramecium. We chose Tetrahymena, Skerkiella (Oxytricha), Colpodia and Blepharisma, which are important representatives of the major ciliate groups. By untilizing the same degenerate primer pair used for cloning the Paramecium cyclins, PCRs were carried out with genomic DNAs of these ciliates as templates. Two classes of cyclin-like sequences were obtained from Blepharisma and one each from Tetrahymena, Skerkiella, Colpodia. A GenBank search indicated that they were all homologues of mitotic cyclins (Fig. 8A).

Based on the sequence information of this small, but most conserved region among cyclins, these ciliate cyclins fall into two major groups, with those from Paramecium and Sterkiella in one group and those from Blepharisma, Tetrahymena and Colpodia in the other (Fig. 8B). One of the Blepharisma cyclin (BLCyc2) singled out as an independent group, suggesting the early separation of the two Blepharisma cyclins.

Even though Cyc1 and Cyc2 from Paramecium are clearly members of the mitotic cyclin family, they are quite divergent from the cyclins of other species, and do not fit well into either type when aligned with A- or B-type cyclins using the CLUSTALW program, suggesting the early evolutionary separation of ciliate cyclins from those in higher eukaryotes (Higgins et al., 1994) (Fig. 8C).

In previous studies, we identified three classes of p34^cdc2 homologues in Paramecium tetraurelia, two of which have been cloned so far (Tang et al., 1994, 1995, 1997; Zhang and Berger, in preparation). They all displayed the characteristic CDK protein kinase activity towards bovine histone H1. Cdk1 activity was associated with macronuclear DNA synthesis (IDS) while Cdk2 had a kinase peak at cytokinesis. Cdk3 was the only kinase showing affinity for yeast p13^suc1 so far, and a p13^suc1-associated histone H1 kinase activity peak was observed at PCD, when the cell becomes irreversibly committed to division (Tang et al., 1994).

In this report, we show that cyclins, the activating subunits of CDK kinase activity, are also present in ciliate Paramecium tetraurelia. This is the first time that cyclin homologues have been unambiguously identified, cloned and characterized in ciliates, although a protein which exhibited a cell cycle-dependent oscillation in abundance was observed in Tetrahymena thermophila (Williams and Macey, 1991), and a sequence which displayed a limited degree of homology to cyclin B was reported earlier in Stylonychia lemnae (Maercker and Lipps, 1994). This study not only extends the omnipresence of cyclins to this unique, diverse, unicellular group, but also lays the foundations for further investigation of the molecular basis of unusual cell cycle control system in ciliates.

The assignment of Cyc1 and Cyc2 as Paramecium mitotic cyclins is based upon their sequence similarity to other mitotic cyclins and their cell cycle-dependent oscillation at both protein level and associated histone H1 kinase activity. The sequence similarity is restricted to the cyclin box regions, which exhibit 42~51% identity to both A- and B-type cyclins from other eukaryotes, and to the ubiquitin specific ‘destruction box’ sequences near the N-termini.

Cyclin box homologous sequences were also identified in other ciliate species and construction of generic trees using the CLUSTALW program (Higgins et al., 1994) demonstrated that Cyc1 and Cyc2 did not belong to either A- or B-type cyclins. These results are consistent with an early evolutionary divergence between ciliates and other eukaryotes and suggest the possibility of functional differences between ciliate cyclins and other eukaryotic cyclins.

However, like most of mitotic cyclins examined thus far, both Cyc1 and Cyc2 display periodical oscillation in their protein levels during the vegetative cell cycle, but with different kinetics. Cyc1 was maximally expressed at around PCD, 90 minutes before cytokinesis whereas Cyc2 was maximally expressed during cytokinesis. This difference in periodicity implies that Cyc1 and Cyc2 may have different functions during the Paramecium cell cycle.

Cyclin-associated histone H1 kinase activities showed similar fluctuations during the vegetative cell cycle: that is, Cyc1-associated histone H1 kinase peaked at PCD whereas Cyc2-associated kinase activity reached a maximum at the end of cell cycle. Coincidence of the cyclin associated kinase activity with previously described CDK histone H kinase activity suggested that Cyc2 should bind to Cdk2 whereas Cyc1 should bind to Cdk3. This is confirmed by co-immunoprecipitation experiments for Cyc2 and Cdk2. p13^suc1 bound histone H1 kinase activity peaks at PCD (Tang et al., 1994), and Cyc1 was found in association with p13^suc1. But there is thus far no specific antibody against Cdk3, it was not possible to verify the physical interaction between Cdk3 and Cyc1. However, we can not exclude the possibility that Cyc1 forms a complex with an unidentified CDK that also associates with p13^suc1. The parallel oscillation of the cyclin protein levels with kinase activity indicates that Cyc1 and Cyc2 function as Paramecium CDK activators.

The presence of multiple cyclins has been well documented in higher eukaryotes. Human cells have at least 14 cyclins with 8 partner kinases (Pines, 1996), and each cyclin appears to bind to specific kinase partner(s), resulting in cyclin/CDK complexes that play distinct roles during the cell cycle. This also seems to be the case for Paramecium tetraurelia. Furthermore, histone H1 kinase activities of the resulting cyclin/CDK complexes reach the maximum at different stages of the cell cycle, suggesting functional separation between the Paramecium cyclins (Table II). Given unique cellular structure of ciliates, with temporally and spatially different regulated cell cycle events in macronucleus, micronucleus and cytoplasm, it is perhaps not surprising to find the presence of multiple CDK/cyclin combinations with distinct functions in Paramecium. One of the central questions that needs to be addressed is how the CDK/cyclin complexes in different compartments of the cell are coordinated with one another.

Thus far, a cyclin that associates with Cdk1, the kinase associated with the macronuclear DNA synthesis, has not been identified (Table II). Previous evidence from glycerol density gradient centrifugation experiment indicated that Cdk1 might function as monomer (Tang et al., 1997). Further study is necessary to clarify whether this kinase complexes with a cyclin which has very limited sequence identity to the primers that were used to isolate Cyc1 and Cyc2 or whether it does in fact function without a cyclin. It is known that human G1 cyclins exhibit great divergence region from mitotic cyclins in the cyclin box, and the identity between different classes of human cyclins can be as low as 18% (Cyclin C vs. Cyclin A) (Lew et al., 1991).

Our work here is just the first step towards elucidation of the roles of the Paramecium cyclins in the cell cycle regulation and/or other metabolic pathways. Even though a few cell cycle mutants have been identified in Paramecium, none are mutated in either CDK or cyclin. It may attribute to the existence of gene isoforms in both CDK and cyclin.

Combined with our previous studies (Tang et al., 1994, 1995, 1997), our results support the idea that the major gene products involving in the cell cycle regulation are evolutionarily conserved in all eukaryotes. Moreover, even though the cell cycle control in the ciliates appears so different from other eukaryotes, the observation that cyclin-like sequences are also present in other ciliates further suggests that cyclins have been conserved during evolution. The identification of CDK/cyclin complexes in Paramecium tetraurelia may provide the first clue for uncovering the regulatory system underlying the ciliate unique cell cycle process. Further work should not only help to elucidate the Paramecium cell cycle, but also assist the complete understanding of the cell cycle controls operating in higher eukaryotes.

We would like to thank Dr. Sina M. Adl for providing Skerkiella, Colpodia and Blepharisma cultures for genomic DNA isolation, Drs James Forney and Eric Meyer for the DNA libraries. We are very grateful to Drs Gerald Weeks, Sarb Neil and Liren Tang for helpful discussions and comments on the manuscript. This work was funded by the NSERC Canada grant to J.D.B.

Adl, M. S. and Berger, J. D. (1992). Timing of micronuclear mitosis and its relation to commitment to division in Paramecium tetraurelia. Dev. Genet. 13, 229-234.

Allen, S. L., Altschuler, M. I., Bruns, P. J., Cohen, J., Doerder, F. P., Gaertig, J., Gorovsky, M., Orias, E. and Turkewitz, A. (1998). Proposed genetic nomenclature rules for Tetrahymena thermophila, Paramecium primaurelia and Paramecium tetraurelia. Genetics 149, 459-462.

An, S. F., Franklin, D. and Fleming, K. A. (1992). Generation of dioxigenin-labelled double-stranded and single-stranded probes using the polymerase chain reaction. Mol. Cell. Probes 6, 193-200.

Berger, J. D. (1988). The cell cycle and regulation of cell mass and macronuclear DNA content. In Paramecium (ed. H. D. Gortz),. pp.97-119. Springer-Verlag, Berlin Heideberg.

Berger, J. D. (1979). Regulation of macronuclear DNA content in Paramecium tetraurelia. J. Protozool. 26, 18-28.

Chou, Y.-H., Bischoff, J. R., Beach, D. and Goldman, R. D. (1990). Intermediate filament reorganization during mitosis is mediated by p34^cdc2 phosphorylation of vimentin. Cell 62, 1063-1071.

De Bondt, H. L., Rosenblatt, J., Jancarik, J., Jones, H. D., Morgan, D. O. and Kim, S.-H. (1993). Crystal structure of cyclin-dependent kinase 2. Nature 363, 595-602.

Dutta, A. and Stillman, B. (1992). Cdc family kinases phosphorylate a human cell DNA replication factor, RPA, and activate DNA replication. EMBO J. 11, 2189-2199.

Engler-Blum, G., Meier, M., Frank, J. and Muller, G. A. (1993). Reduction of background problems in nonradioactive Northern and Southern blot analysis enables higher sensitivity than ³²P-based hybridizations. Anal. Biochem. 210, 235-244.

Evans, T., Rosenthal, E. T., Youngblom, J., Distel, D. and Hunt, T. (1983). Cyclin: a protein sepcified by maternal mRNA in sea urchin eggs that is destroyrd at each cleavage division. Cell 33, 389-396.

Fisher, R. P. (1997). CDKs and cyclins in transition(s). Curr. Opin. Gen. & Dev. 7, 32-38.

Frohman, M. A. (1990). RACE: rapid amplification of cDNA ends. In PCR protocols (ed. J. J. Snincky and T. J. White). pp.28-38. Academic Press, San Diego, CA..

Fujishima, M., Katsu, Y., Ogawa, E., Sakimura, M., Yamashita, M. and Nagahama, Y. (1992). Meiosis-reinitiation-inducing factor of Tetrahymena functions upstream of M-phase-promoting factor. J. Protozool. 39, 683-690.

Glotzer, M., Murray, A. W. and Kirschner, M. W. (1991). Cyclin is degraded by the ubiquitin pathway. Nature 349, 132-138.

Hanks, S. K., Quinn, A. M. and Hunter, T. (1988). The protein kinase family: conserved features and deduced phylogeny of the catalytic domains. Science 241, 42-52.

Harlow, E. and Lane, D. (1988). Antibodies, a laboratory manual. pp.471-510. Cold Spring Harbor Laboratory Press, New York.

Harper, J. W., Adami, G. R., Wei, N., Keyomarsi, K. and Elledge, S. J. (1993). The p21 Cdk-interacting protein Cip1 is a potent inhibitor of G1 cyclin-dependent kinases. Cell 75, 805-816.

Hata, S., Kouchi, H., Suzuka, I. and Ishii, T. (1991). Isolation and characterization of cDNA clones for plant cyclins. EMBO J. 10, 2681-2688.

Hayles, J., Fisher, D., Woollard, A. and Nurse, P. (1994). Temporal order of S phase and mitosis in fission yeast is determined by the state of the p34^cdc2-mitotic B cyclin complex. Cell 78, 813-822.

Higgins, D., Thompson, J. and Gibson, T. (1994). CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22, 4673-4680.

Hirt, H., Mink, M., Pfosser, M., Bogre, L., Gyorgyey, J., Jonak, V., Gartner, A., Dudits, D. and Herberle-Bors, E. (1992). Alfalfa cyclins: differential expression during the cell cycle and in plant organs. Plant Cell 4, 1531-1538.

Jeffrey, P. D., Russo, A. A., Polyak, K., Gibbs, E., Hurwitz, J., Massague, J. and Pavletich, N. P. (1995). Mechanism of CDK activation revealed by the structure of a cyclinA-CDK2 complex. Nature 376, 313-320.

Kobayashi, H., Stewart, E., Poon, R., Adamczewski, J. P., Gannon, J. and Hunt, T. (1992). Identification of the domains in cyclin A required for binding to, and activation of, p34^cdc2 and p32^cdk2 protein kinase subunits. Mol. Biol. Cell 3, 1279-1294.

Koetsier, P.A., Schorr, J. and Doerfler, W. (1993). A rapid optimized protocol for downward alkaline Southern blotting of DNA. BioTechniques 15, 260-261.

Laemmli, U. K. (1970). Cleavage of structured proteins during the assembly of the head of the bacteriophage T₄. Nature 277, 680-685.

Lees, E. M. and Harlow, E. (1993). Sequences within the conserved cyclin box of human cyclin A are sufficient for binding to and activation of cdc2 kinase. Mol. Cell. Biol. 13, 1194-1201.

Lehner, C. F. and O'Farrell, P. H. (1989) Expression and function of Drosophila cyclin A during embryonic cell cycle progression. Cell 56, 957-968.

Lehner, C. F. and O'Farrell, P. H. (1990). The roles of Drosophila cyclins A and B in mitotic control. Cell 61, 535-547.

Lew, D. J., Dulic, V. and Reed, S. I. (1991). Isolation of three novel human cyclins by rescue of G1 cyclin (Cln) function in yeast. Cell 66, 1197-1206.

Luo, Q., Michaelis, C. and Weeks, G. (1994). Overexpression of a truncated cyclin B gene arrests Dictyostelium cell division during mitosis. J. Cell Sci. 107, 3105-3114.

Maercker, C. and Lipps, H. J. (1994). A gene from the hypotrichous ciliate Stylonychia lemnae coding for a protein with homology to cyclin B. Gene 141, 145-146.

Martindale, D. W. (1989). Codon usage in Tetrahymena and other ciliates. J. Protozool. 36, 29-34.

Minshull, J., Blow, J. J. and Hunt, T. (1989). Translation of cyclin mRNA is necessary for extracts of activated Xenopus eggs to enter mitosis. Cell 56, 947-956.

Minshull, J., Golsteyn, R., Hill, C. S. and Hunt, T. (1990). The A- and B-type cyclin associated cdc2 kinases in Xenopus turn on and off at different times in the cell cycle. EMBO J. 9, 2865-2875.

Morgan, D. O. (1995). Principles of CDK regulation. Nature 374, 131-134.

Nigg, E. A. (1995). Cyclin-dependent protein kinases: key regulators of the eukaryotic cell cycle. BioEssays 17, 471-480.

Page, R. D. M. (1996). TREEVIEW: An application to display phylogenetic trees on personal computers. Computer Applications in the Biosciences 12, 357-358

Pasternak, J. (1967). Differential gene activity in Paramecium aurelia. J. Exp. Zool.

165, 395-418.

Peeper, D. S., Parker, L. L., Ewen, M. E., Toebes, M., Hall, F. L., Xu, M., Zantema, A., Eb, A. J. and Piwnica-Worms, H. (1993). A- and B-type cyclins differentially modulate substrate specificity of cyclin-cdk complexes. EMBO J. 12, 1947-1954.

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

Pines, J. (1995). Cyclins and cyclin-dependent kinases: a biochemical view. Biochem. J. 308, 697-711.

Pines, J. (1996). Cyclin from sea urchins to Helas: making the human cell cycle. Biochem. Society Transactions. 24, 15-33.

Pines, J. and Hunter, T. (1989). Isolation of a human cyclin cDNA: evidence for cyclin mRNA and protein regulation in the cell cycle and for interaction with p34^cdc2. Cell 58, 833-846.

Pines, J. and Hunter, T. (1989). Isolation of a human cyclin cDNA: evidence for cyclin mRNA and protein regulation in the cell cycle and for interaction with p34^cdc2. Cell 58, 833-846.

Pines, J. and Hunter, T. (1990). Human cyclin A is adenovirus E1A-associated protein p60, and behaves differently from cyclin B. Nature 346, 760-763.

Pines, J. and Hunter, T. (1991). Human cyclin A and B1 are differentially located in the cell and undergo cell cycle -dependent nuclear transport. J. Cell Biol. 115, 1-17.

Prescott, D. M. (1994). The DNA of ciliated protozoa. Microbiol. Rev. 58, 233-267.

Rasmussen, C. D. and Berger, J. D. (1982). Downward regulation of cell size in Paramecium tetraurelia: effects of increased cell size, with or without increased DNA content, on the cell cycle. J. Cell Sci. 57, 315-329.

Rasmussen, C. D., Berger, J. D. and Ching, A. L. (1986). Effects of increased cell mass and altered gene dosage on the timing of initiation of macronuclear DNA synthesis in Paramecium tetraurelia. Exp. Cell Res. 165, 53-62.

Roth, S. Y., Collini, M. P., Draetta, G., Beach, D. and Allis, C. D. (1991). A cdc2-like kinase phosphorylates histone H1 in the amitotic macronucleus of Tetrahymena. EMBO, J. 10, 2069-2075.

Russell, C. B., Fraga, D. and Hinrichsen, R. D. (1994). Extremely short 20-33 nucleotide introns are the standard length in Paramecium tetraurelia. Nucl. Acids Res. 22, 1221-1225.

Sambrook, J., Fritsch, E. F. and Maniatis, T. (1989). Molecular cloning, a laboratory manual. Cold Spring Harbor laboratory Press, New York.

Satterwhite, L. L., Lohka, M. J., Wilson, K. L., Scherson, T. Y., Cisek, L. J., Corden, J. L. and Pollard, T. D. (1992). Phosphorylation of myosin-II rergulatory light chain by cyclin-p34cdc: a mechanism for the timing of cytokinesis. J. Cell Biol. 118, 595-605.

Sonneborn, T. M. (1970). Methods in Paramecium research. Methods Cell Physiol. 4, 241-339.

Standart, N., Minshull, J., Pines, J. and Hunt, T. (1987). Cyclin synthesis, modification and destruction during meiotic maturation of the starfish oocyte. Dev. Biol. 124, 248-254.

Swenson, K. I., Farrell, K. M. and Ruderman, J. V. (1986). The clam embryo protein cyclin A indices entry into M-phase and the resumption of meiosis in Xenopus oocytes. Cell 47, 861-870.

Swenson, K. I., Farrell, K. M. and Ruderman, J. V. (1986). The clam embryo protein cyclin A induces entry into M phase and the resumption of meiosis in Xenopus oocytes. Cell 47, 861-870.

Tang, L., Adl, M. S. and Berger, J. D. (1997). A cdc2-related kinase is associated with macronuclear DNA synthesis in Paramecium tetraurelia. J. Euk. Microbiol. 44, 269-275.

Tang, L., Pelech, S. L. and Berger, J. D. (1995). Isolation of the cell cycle control gene cdc2 from Paramecium tetraurelia. Biochim. Biophys. Acta. 1265, 161-167.

Tang, L., Pelech, S. L. and Berger, J. D. (1994). A cdc2-like kinase associated with commitment to division in Paramecium tetraurelia. J. Euk. Microbiol. 41, 381-387.

Tang, L. (1995). A CDC2-related kinase from Paramecium tetraurelia. Ph. D. dissertation, University of British Columbia.

Toyoshima, H. and Hunter, T. (1994). P27, a novel inhibitor of G1 cyclin-cdk protein kinase activity, is related to p21. Cell 78, 67-74.

Wang, J., Chenivesse, X., Henglein, B. and Brechot, C. (1990). Hepatitis B virus integration in a cyclin A gene in a hepatocellular carcinoma. Nature 343, 555-557.

Westendorf, J. M., Swenson, K. I. and Ruderman, J. V. (1989). The role of cyclin B in meiosis I. J. Cell Biol. 108, 1431-1444.

Williams, N. E. and Macey, M. G. (1991). Is cyclin Zeuthen’s "division protein"?. Exp. Cell Res. 197, 137-139.

Xiong, Y., Zhang, H., Casso, D., Kobayashi, R. and Beach, D. (1993). P21 is a universal inhibitor of cyclin kinases. Nature 366, 701-704.

Fig.1. Paramecium tetraurelia cell cycle control model. Two control points, IDS and PCD, are shown with the macronuclear (MAC) and micronuclear (MIC) cell cycle events. IDS stands for the initiation of macronuclear DNA synthesis. PCD is the point of commitment to cell division and is the major control point. It occurs ~90 min before cytokinesis. M: mitosis; AM: amitosis (modified from Tang, 1995).

Fig.2. Genomic DNA sequences of Paramecium tetraurelia CYC1a (Panel A) and CYC2 (Panel B), and the predicted amino acid sequences of their gene products. ‘Cyclin box’ regions are single underlined. ‘Destruction box’ regions close to ‘R´ ´ L´ ´ I´ N’ consensus sequence are double underlined. Introns are shown in lower cases. Potential CDK phosphorylation sites (Thr 269 and Thr 274 in Cyc1A and Thr 269 in Cyc2) are indicated by ‘#’. Stop codons (TGA) in both sequences are indicated by ‘*’. These two sequences are available from GenBank under accession numbers AF052484 and AF052487, respectively.

Fig.3. Sequence comparison in the ‘cyclin box’ regions of Paramecium tetraurelia Cyc1 and Cyc2 with A-type (Panel A) and B-type (Panel B) cyclins from various species. SS: Spisula solidissima (Swenson et al., 1986; Westendorf et al., 1989). DM: Drosophila melanogaster (Lehner and O'Farrell, 1989, 1990); XL: Xenopus laevis (Minshull et al., 1989, 1990). HS: Homo sapiens (Wang et al., 1990; Pines and Hunter, 1989). Dashes indicate identity with Cyc1, and gaps have been introduced where necessary to maximize the alignment. Consensus residues shared by most A-type (A con.) or B-type (B con.) cyclins are in bold letters (Hata et al., 1991).

Fig.4. Genomic Southern blot analysis. Paramecium genomic DNA was digested with EcoRI, HindIII, PstI, XbaI, BamH1, and probed with DIG-labeled DNA probes, as indicated below, and washed under either low stringency (LS, 40° C) or high stringency (HS, 65° C). The 1-kb marker from Gibco-BRL is indicated on the left. Arrowheads highlighted faint bands detected by CYC1b under low stringency conditions.

Fig.5. Oscillation of Cyc1 and Cyc2 protein levels and their associated histone H1 kinase during the Paramecium vegetative cell cycle. Small cells in early G1 were selected by centrifugal elutriation and re-inoculated into fresh medium. After 0.5 h, aliquots were taken at 1 h intervals to determine cumulative percentage of cell division (Panels C and F); Panels A and D: immunoblot of 50 m g of lysate from each fraction, probed with anti-Cyc1 (A) or anti-Cyc2 (D) antibodies; pPanels B and E: histone H1 kinase activity present in Cyc1 or Cyc2 immunoprecipitates was visualized by resolving the kinase assay reactions on SDS-PAGE and exposing dried gels to X-ray films. Protein levels were further quantitated using NIH Image program, and quantitation of the kinase activities was carried out by scintillation counting of the excised phosphorylated histone H1 bands. Results are shown as percentage of maximum in Panels C and F. Note: this figure presents composite results derived from different batches of synchronized samples due to the limited amount of proteins available from each elutriation.

Fig.6. Western blot analysis of protein levels of Cyc1 and Cyc2 in exponentially growing and starved cells. Equal amounts of protein were loaded on each lane.

Fig.7. Interaction of Cyc2 with Cdk2. More than 2.5 mg of total proteins from exponentially growing Paramecium cells were used in each co-immunoprecipitation experiment. Panels A and B: interaction between Cyc2 and Cdk2 was demonstrated by reciprocal coimmunoprecipitation experiments. Cdk2 was detectable in the Cyc2 immunoprecipitates (A) while Cyc2 was detectable in the Cdk2 immunoprecipitates (B). * indicates a non-specific band detected by anti-Cyc2 peptide antibody. The heavy band above 47.7 kDa represents rabbit IgG. Panel C: Cyc1 was detectable in p13^suc1 bound fraction. After incubation with the lysate at 4° C for 5 h, yeast p13^suc1 Sepharose CL4B beads (Upstate Biotechnology, Lake placid, NY) were washed extensively with bead buffer (Tang et al., 1994). Both bound (P) and unbound (S) fractions were resolved on SDS-PAGE and probed with anti-Cyc1 or anti-Cyc2 antibodies. Several bands below 34 kDa might attribute to the degraded products from the proteins of interest.

Fig.8. Conservation of cyclin across different groups of ciliates. Panel A: alignment of cyclins from ciliates, Paramecium (PT), Tetrahymena (TE), Sterkiella (SK), Blepharisma (BL) and Colpodia (CO) in the most conserved region within the ‘cyclin box’. Identical residues with Cyc1A are replaced with dashes. Panel B: phylogenetic tree of ciliate cyclins constructed from an alignment as shown in Panel A. Panel C: evolutionary tree of cyclins from Paramecium tetraurelia (PT), Spisula solidissima (SS, Swenson et al., 1986; Westendorf et al., 1989), Drosophila melanogaster (DM, Lehner and O'Farrell, 1989, 1990), Xenopus laevis (XL, Minshull et al., 1989, 1990), Homo sapiens (HS, Wang et al., 1990; Pines and Hunter, 1989) and Dictyostelium discoideum (DD, Luo et al., 1994). Trees are constructed on the basis of the cyclin box regions as shown in Figure 3, using the neighbor joining method with the CLUSTALW program (Higgins et al., 1994), and viewed with the TreeView (Page, 1996). The scale bar of "0.1" means 0.1 nucleotide substitutions per site.