Molecular Microbiology (2001) 42(1), 257–267

KIN241: a gene involved in cell morphogenesis in Paramecium tetraurelia reveals a novel protein family of cyclophilin –RNA interacting proteins (CRIPs) conserved from fission yeast to man Anna Krzywicka,1 2Janine Beisson,1 Anne-Marie Keller,1 Jean Cohen,1 Maria Jerka-Dziadosz2 and Catherine Klotz1* 1 Centre de Ge´ne´tique Mole´culaire, Centre National de la Recherche Scientifique, 91198 Gif-sur-Yvette Cedex, France. 2 Department of Cell Biology, Nencki Institute of Experimental Biology, Polish Academy of Sciences, 02 –093 Warsaw, Poland. Summary In this study, we report cloning, by functional complementation of the KIN241 gene involved in Paramecium cell morphogenesis, cortical organization and nuclear reorganization. This gene is predicted to encode a protein of a novel type, comprising a cyclophilin-type, peptidyl-prolyl isomerase domain, an RNA recognition motif, followed by a region rich in glutamate and lysine (EK domain) and a C-terminal string of serines. As homologues of this protein are present in the genomes of Schizosaccharomyces pombe, Caenorhabditis elegans, Drosophila melanogaster, Arabidopsis thaliana and Homo sapiens, the Kin241p predicted sequence defines a new family of proteins that we propose to call ‘CRIP’, for cyclophilin– RNA interacting protein. We demonstrate that, in Paramecium, Kin241p is localized in the nucleus and that deletion of some nuclear localization signals (NLSs) decreases transport of the protein into the nucleus. No Kin241-1 protein is present in mutant cells, suggesting that the C-terminal serine-rich region is responsible for protein stability.

Introduction Ciliates are unicellular organisms characterized by an elaborate cortical pattern based upon the precise, asymmetric and polarized arrangement of several thousands of cortical units, each centred on ciliary basal Accepted 20 July, 2001. *For correspondence. E-mail klotz@ cgm.cnrs-gif.fr; Tel. (133) 169823157; Fax (133) 169823150.

Q 2001 Blackwell Science Ltd

bodies, homologues of centrioles. Cell division proceeds by binary fission and results in the reconstruction, from two half-cells of different shape and organization, of two identical daughter cells. Division thus implies a complex developmental programme, which has been well analysed in Paramecium. It reveals that the cell is, in fact, a mosaic of territories with different morphogenetic properties that contribute differentially to the surface of the daughter cells (Iftode et al., 1989). The surface of a dividing cell is invaded by waves of morphogenetic signals that trigger the duplication of basal bodies successively and independently and the reorganization/assembly of the cortical cytoskeletal structures (Sperling et al., 1991). As metazoa evolved from unicellular eukaryotes and develop from a single cell, the egg, it would not be surprising that some of the mechanisms that play a role in ciliate morphogenesis might have been conserved and be involved in early steps of metazoan development. Here, we report the characterization of a gene, KIN241, identified by a morphological mutation in Paramecium, which has homologues of still unknown function in various organisms ranging from Schizosaccharomyces pombe to man. The mutation kin241-1 was selected on the basis of large cell size and thermolethality and shown to display highly pleiotropic effects on cortical organization, generation time and nuclear reorganization during sexual processes (Beisson et al., 1976; Grandchamp and Beisson, 1981). Detailed cytological analysis of basal body duplication and cytoskeleton dynamics throughout the cell cycle revealed anomalies at several levels: in basal body duplication and progression of the morphogenetic waves, yielding a modified body plan; in basal body polarities generating disorganized patches and rows of basal bodies with reversed polarities; in the nucleation activity of basal bodies leading to formation of supernumerary ciliary rootlets; and, finally, in occasional abnormal ultrastructure of basal bodies that display extramicrotubules (Jerka-Dziadosz et al., 1992). In contrast with other mutations also affecting cell shape, but more specifically basal body duplication, such as the sm19-1 mutation (Ruiz et al., 2000), the kin241-1 mutation with its highly pleiotropic effects probably resulted from a defect in some general upstream process.

258 A. Krzywicka et al. The KIN241 gene was cloned by functional complementation of the kin241-1 mutant and found to encode a predicted protein of a novel type, comprising a cyclophilintype peptidyl-prolyl isomerase domain, an RNA recognition motif, followed by a region rich in glutamate and lysine (EK domain) and a C-terminal string of serines. As homologues of this protein are present in the genomes of S. pombe, Caenorhabditis elegans, Drosophila melanogaster, Arabidopsis thaliana and man, the Kin241p predicted sequence defines a new family of proteins that we propose to call ‘CRIP’, for cyclophilin –RNA interacting protein. We demonstrate that Paramecium Kin241p, which possesses several nuclear localization signals (NLSs), is localized in the nucleus and that deletion of some NLSs decreases transport of the protein into the nucleus. The C-terminal serine-rich region seems to be responsible for protein stability as Kin241-1 protein is not present in mutant cells. We discuss the possibility that CRIP could be a new, conserved enzyme that can influence morphogenesis via its possible involvement in the processing of particular RNA species. Results KIN241 gene encodes a novel cyclophilin with an RNA recognition motif Complementation cloning of the wild-type gene KIN241 was done by sib-selection (Haynes et al., 1996; Skouri and Cohen, 1997) on kin241-1 mutant cells (Jerka-Dziadosz et al., 1992) using an indexed Paramecium genomic library (Keller and Cohen, 2000). The rescued cells displayed wild-type size, morphology and generation time and normal growth at the restrictive temperature of 358C, whereas kin241-1 mutant cells die at this temperature within 24 h. The complementing plasmid, p103o9, contains a 3.2 kb insert with a 2.1 kb open reading frame (ORF; EMBL database accession number AJ409115) interrupted by a short 28 nucleotide (nt) intron (bp 229 –253) typical of Paramecium (Dupuis, 1992; Russell et al., 1994).

Southern blot analysis, using the central part of the gene as a probe, showed that the KIN241 gene is unique in the genome. This was confirmed by the presence of a single band, of < 400 kb, visible on a blot of Paramecium chromosomes, separated by pulsed-field gel electrophoresis (PFGE). Northern blots probed with the same region of the gene indicate the existence of a 2.2 kb transcript. To verify that we cloned the KIN241 gene, and not a multicopy suppressor, the region corresponding to the KIN241 gene was sequenced in the mutant kin241-1 using polymerase chain reaction (PCR) products from three independent amplifications of DNA from the mutant strain. The amplified region corresponds to the 3.2 kb insert of the originally isolated plasmid p103o9 complementing the kin241-1 mutation. The kin241-1 mutation was found to be a two-nucleotide insertion (at position 1523) that generates a frameshift leading to the introduction of a stop codon at position 1658, i.e. 590 nucleotides upstream of its wild-type position. The predicted wild-type Kin241 protein contains 695 amino acid residues, with a predicted molecular weight of 82.7 kDa and an isoelectric point of 9.12. A BLAST search of the public database revealed that the gene encodes a multidomain protein (Figs 1 and 2A). The first domain (aa 6–150) is homologous to cyclophilin-type peptidyl-prolyl cis–trans isomerase. It is followed (aa 225 –303) by an RNA recognition motif domain (RRM). The next part (aa 340–598) of the protein is rich in charged residues, mainly lysine alternating with glutamic acid (EK domain), and has a predicted coiled-coil structure. The C-terminal region is composed of 34 amino acids of which 28 are serine residues. In addition, the protein contains six bipartite-type NLSs rich in lysine and arginine residues. Such NLSs, contrary to monopartite NLSs that are based on a single cluster of basic amino acids, are composed of two basic amino acids, a spacer region of 10 –12 amino acids and a basic cluster of three to five amino acids (Conti et al., 1998). All but one are localized within the EK domain (Fig. 2A).

Fig. 1. Sequence of predicted Kin241 protein in Paramecium: cyclophilin-type cis/trans peptidyl-prolyl isomerase homology domain is underlined (CY), the RNA recognition motif homology domain (RRM) is double underlined, the domain rich in charged amino acids is underlined with a dotted line, and the C-terminal serine-rich region is boxed.

Q 2001 Blackwell Science Ltd, Molecular Microbiology, 42, 257–267

A novel cyclophilin–RNA interacting protein

259

Fig. 2. A. Schematic diagram of the structure of predicted variants of the Kin241 protein in Paramecium: Kin241p from wild-type cells, kin241-1p from mutant cells and Kin241-rev1p and Kin241-rev2p from revertant lines: rev-1, rev-2 respectively; asterisk, mutation site (insertion of two nucleotides: GA) introducing a stop codon; open triangle, deletion (55 or 8 amino acids); open box, a deleted protein fragment; closed circle, nuclear localization signal (NLS1 at position 318, NLS2; 397, NLS3, 400; NLS4, 532; and NLS 5/6, 564/566); CY ,cyclophilin domain; RRM, RNA recognition motif. B. Northern blot with total RNA isolated from wild-type KIN241, mutant kin241-1 and revertant kin241-rev1 and kin241-rev2 lines and hybridized with a probe for the KIN241 gene.

Kin241p defines a novel family of cyclophilin-RRM proteins Sequence comparison revealed the existence of Kin241p homologues in the genomes of other organisms: C. elegans (T22008), D. melanogaster (AF132148), S. pombe (T39728) and A. thaliana (AAG21471). All these predicted proteins contain similar cyclophilin, RRM and charged domains (Fig. 3A and B). Only the cyclophilinRRM protein in A. thaliana has a supplementary domain, a CCHC -type Zn-finger (345 –358). A still partial sequence (CAB88111) obtained from the human genome project revealed a 50 fragment of the gene encoding the N-terminal 290 first amino acids of a KIN241-related protein with the same cyclophilin-RRM organization. Interestingly, no Kin241p homologue has been found in the genome of the budding yeast Saccharomyces cerevisiae. The N-terminal parts of all these proteins share significant identity with (above 40%) and similarity to (above 60%) the first 300 amino acids, i.e. within the cyclophilin and RRM domains and the surrounding regions (Fig. 3C). The downstream parts (Fig. 3A) are less related but present a similar coiled-coil structure, especially within the region rich in charged amino acids. Nuclear localization signals are also found in this region as in Paramecium. Drosophila melanogaster and A. thaliana have six and four bipartite-type NLSs. There are one monopartite and two bipartite NLSs in C. elegans, and S. pombe has three monopartite-type NLSs. In D. melanogaster, the charged region (422 –552) is composed of alternating charged and uncharged amino acids: mainly arginine and serine, with five imperfect repeats of RSRSRX and at least nine RS dipeptides. A region rich in serines follows this domain. The motif rich in alternating serine –arginine residues, known as the RS or RS-like domain, is characteristic of a great number of proteins involved in mRNA processing (Graveley, 2000). In C. elegans, this part of the protein has a short domain Q 2001 Blackwell Science Ltd, Molecular Microbiology, 42, 257–267

(347–381) composed of charged amino acids, mainly arginine, with few arginine –serine repetitions and without any distal string of serines. The orthologue of the Kin241p protein in Arabidopsis displays a charged region rich in arginine. In fission yeast, the second part of the protein is also composed of charged amino acids, mainly arginine. To summarize, the Kin241 protein from Paramecium and its homologues defines a new, conserved family of proteins that we propose to call CRIPs for cyclophilin– RNA interacting proteins whose cellular functions remain to be established.

Mutational analysis: essential role of Kin241 C-terminal region As the kin241-1 mutation is a two-nucleotide insertion that indirectly induced the appearance of a stop codon 590 nucleotides upstream of the wild-type position, the mutant protein truncates its C-terminus, i.e. part of the charged domain and the serine-rich region (Fig. 2A). Northern blot analysis demonstrated the presence of the 2.2 kb mRNA transcript that corresponds to the size of the wild-type mRNA. The quantity of mRNA was similar to that found in wild-type cells (Fig. 2B). The KIN241 or kin241-1 mRNA level also remained constant after 9 h incubation at the restrictive temperature 358C (not shown). These results indicate that the mutation kin241-1 does not affect mRNA expression or stability. The four revertant lines (kin241-rev-a, kin241-rev-b, kin241-rev-d and kin241-rev-g; M. Rossignol and J. Beisson, unpublished), presumed to contain intragenic suppressors, were sequenced. Three of these lines (kin241-rev-a, -b, -d) display the same mutation, kin241rev1, a deletion of 161 nucleotides (D nt 1501–1662) corresponding to a deletion of 55 amino acids (497 –550) (Fig. 2A). The fourth line (kin241-rev-g) reveals another mutation, kin241-rev2, a deletion of 23 nucleotides (D nt

260 A. Krzywicka et al.

Fig. 3. The CRIP family: sequence and organization of CRIPs from different organisms. A. Comparison of the overall organization of the predicted proteins. Closed circles, nuclear localization signal (NLS); CY, cyclophilin; RRM, RNA recognition motif; ZF, zinc finger. B. Alignment of the first 300 amino acids of CRIP proteins from D. melanogaster (accession number AF132148), H. sapiens (CAB88111), C. elegans (T22008), S. pombe (T39728), A. thaliana (AAG21471) and P. tetraurelia (AJ409115). These sequences were identified by a BLAST search in public databases. Identical amino acid positions are shaded in black, conservative changes in grey, introduced gaps are represented by dots. The cyclophilin type peptidyl-prolyl cis/trans isomerase domain (CY) and RNA recognition motif (RRM) domains were identified using the PROSITE program and underlined. The RNP-2 and RNP-1 consensus of the RRM is boxed. Sequence alignments were done using CLUSTAL W (Thompson et al., 1994) and edited with BOXSHADE 3.21. C. Table summarizing percentage of identity and similarity for the first 300 aa of CRIP performed using the DNA STRIDER program (Marck, 1988).

1508–1531) corresponding to a deletion of eight amino acids (505– 512) (Fig. 2A and B). Strikingly, in both cases, the deletions remove the kin241-1 mutation and restore the original reading frame; hence, the predicted C-terminal string of serines. These results clearly show the importance of the C-terminal region. Silencing of the CRIP gene produces a phenocopy of the kin241-1 mutation Gene-specific silencing can be achieved in Paramecium

by microinjection of the coding sequence of a gene without flanking promoter and terminator sequences into the vegetative nucleus at high copy number (Ruiz et al., 1998). In wild-type cells silenced for the KIN241 gene, the first defects, slower growth rate, larger cell size and thermolethality (at 358C after 24 h), appear after 10 –15 divisions in the clones issued from injected cells. In the majority of injected clones, the nuclear reorganization occurring during the sexual process of autogamy showed the same abnormalities as those described in the kin241-1 mutant. In a few, randomly chosen cell lines, examined by Q 2001 Blackwell Science Ltd, Molecular Microbiology, 42, 257–267

A novel cyclophilin–RNA interacting protein immunostaining with an anticiliary rootlet serum, we observed abnormalities in cortical organization similar to those described for kin241-1 mutant cells (Fig. 4). We can conclude that silencing causes the whole spectrum of changes characteristic of the kin241-1 mutation. The GFP –Kin241p fusion protein is localized in the nucleus A GFP –KIN241 fusion gene was prepared by N-terminal fusion of GFP with the KIN241 gene. In order to test whether the introduction of the green fluorescent protein (GFP) affects the function of the protein, the GFP-KIN241 plasmid was injected into kin241-1 mutant cells to assess its rescuing ability. Progeny of injected kin241-1 cells displayed a wild-type phenotype for all traits (cell size, growth rate, number and organization of basal bodies and nuclear reorganization) except thermolethality. All injected clones with rescued phenotypes had their vegetative nucleus, i.e. macronucleus, labelled, indicating a nuclear localization of the GFP –CRIP fusion protein. The GFPKIN241 plasmid was also injected into the nucleus of wildtype cells. Progenies of 40 injected cells were tested for GFP fluorescence. The hybrid protein GFP-Kin241p was

261

also localized in the vegetative nucleus throughout the cell cycle (Fig. 5A). To investigate whether the mutant phenotype could be related to an abnormal localization of kin241-1p, we constructed an N-terminal GFP fusion with the mutant kin241-1 gene. Progenies of cells transformed with this construct did not show any fluorescence, either nuclear or cytoplasmic. To ascertain whether the absence of fluorescence depended on a lack of mRNA or a lack of protein, we probed total RNA from injected cells for the presence of GFP –kin241-1 mRNA. Results showed that this mRNA was present in the injected cell lines without any sign of degradation and that the mRNA level is similar to that observed in wild-type cells injected with GFP – KIN241 (Fig. 5B). As no GFP fluorescence could be detected, the protein carrying the kin241-1 mutation was not present. This fact suggests a possible function of the C-terminal serine-rich region in protein stability. The next step was to check the stability of the protein in the revertant lines in which the C-terminal part of the protein was restored. The GFP –kin241 –rev1 fusion was injected into wild-type cells. The progenies of injected cells displayed fluorescence within the nucleus but also in the cytoplasm (Fig. 5A). The presence of fusion mRNA was

Fig. 4. Cortical pattern of the ventral side of a Paramecium cell as revealed by immunostaining of ciliary rootlets: a cytoskeletal structure associated with the basal body (Sperling et al., 1991) in wild-type, mutant kin241-1 and silenced cells. A. Wild-type cell. The ciliary rows shown by means of the ciliary rootlets are well oriented to the anterior of the cell. B. kin241-1 mutant cell. Global morphological abnormalities, such as increased number of ciliary rows, their disorder and inverted orientation (arrows), are visible. C. The silenced cell. Showing the changes characteristic of the kin241-1 mutation as disorder, inverted orientation and incomplete inclusion of ciliary rows (arrows). OA, oral apparatus. Scale bar 20 mm.

Q 2001 Blackwell Science Ltd, Molecular Microbiology, 42, 257–267

262 A. Krzywicka et al. Discussion A novel family: the cyclophilin–RNA interacting proteins (CRIPs)

Fig. 5. A. Cellular localization of GFP – Kin241p fusion proteins. Noninjected wild-type cells are shown in phase contrast with their vegetative nuclei visualized by DAPI staining. No fluorescence was observed. When transformed with the wild-type GFP – KIN241 fusion gene, the wild-type cells display strong green fluorescence in the vegetative nucleus. In contrast, cells transformed with the revertanttype GFP – KIN241 – rev1 gene display strong fluorescence both in the nucleus and in the cytoplasm. In the cells transformed with the mutanttype GFP – kin241-1 gene, no fluorescence is visible, just as in the case of non-injected cells (data not shown). Scale bar 40 mm. B. Northern blot of total RNA isolated from lines injected by different GFP constructs and hybridized with a probe for the KIN241 gene. a, KIN241 gene mRNA; b, GFP – KIN241 fusion gene mRNA.

confirmed by Northern blot (Fig. 5B). These results confirmed the importance of the serine-rich C-terminal domain, as its presence seems to be correlated with protein stability. The partial cytoplasmic localization of the protein could be related to the absence, in the kin241-rev1 gene, of part of the EK domain, containing three (i.e. NLS4 and NLS5/6) of the six NLS motifs present in the wild-type protein.

The Kin241-1 mutation in Paramecium tetraurelia causes highly pleiotropic effects, affecting cell size and shape, generation time, nuclear differentiation, thermoresistance and cortical organization at both local and global levels (Jerka-Dziadosz et al., 1992). The KIN241 gene could be expected to be an upstream factor in cellular morphogenesis. The KIN241 gene encodes a multidomain protein that we propose to name CRIP, which defines a novel family of proteins with similar organization, conserved in fission yeast, plant and metazoans. All these proteins share three domains: a cyclophilin domain, an RNA recognition motif (RRM) and a domain rich in charged amino acids, and they are all predicted to be localized in the nucleus, as was demonstrated for the Paramecium CRIP. The first domain, characterized as a cyclophilin, belongs to the large group of peptidyl-prolyl isomerases. These enzymes catalyse the cis/trans conformational changes of the peptide bond preceding proline residues in protein substrates. They are thought to accelerate this often ratelimiting step in the folding of a number of proteins in vitro and in vivo (for a review, see Galat and Metcalfe, 1995; Fischer et al., 1998). Prokaryotic and eukaryotic cells contain many different types of isomerases that are often tissue specific or involved in specific processes. The subgroup of cyclophilins shares a conserved core domain of about 110 amino acids surrounded by divergent N- and C-terminal extensions. Cyclophilins are known to be involved in protein folding as chaperones, in assembly/ disassembly of protein complexes and in signal transduction or direct regulation of protein activity (Rassow and Pfanner, 1995; Leverson and Ness, 1998). The second domain (RRM) is a region of about 80 amino acids, containing several well-conserved residues, some of which cluster into two short submotifs, RNP-1 and RNP2. This domain is able to bind single-stranded RNA, and one or more RRMs are found in a variety of RNA-binding proteins (Bandziulis et al., 1989; Birney et al., 1993). The C-terminal part of the CRIP family of proteins differs in size and is less conserved than the N-terminal region. Nevertheless, each of the C-terminal domains generally consists of a region of predicted coiled-coil structure, rich in charged residues. All the described CRIPs possess a few predicted nuclear localization signals within this domain. Their number and type vary slightly, but most of them are arginine rich. The NLSs are responsible for the localization in the nucleus of the protein of interest, but their further subnuclear localization depends on other factors, such as RS repetitions or the co-localization with other proteins in specific complexes (Hedley et al., 1995). How can a nuclear protein, with a cyclophilin domain Q 2001 Blackwell Science Ltd, Molecular Microbiology, 42, 257–267

A novel cyclophilin–RNA interacting protein generally engaged in chaperone function and an RNAbinding domain supposed to act in RNA metabolism, be engaged in morphogenetic processes affecting the whole cell? Two hypotheses can be put forward. This protein could be a mosaic of independent domains, each acting in distinct pathways; the cyclophilin domain could affect a morphogenetic pathway independently of any effects on RNA processing. Alternatively, the juxtaposition of these two domains (which precisely characterizes the CRIPs) is essential for the function of the protein; in this case, the RNA recognition motif could bind the CRIP to an RNA molecule or a nucleoprotein complex in which the cyclophilin domain could play its role. Circumstantial evidence is in favour of CRIPs in RNA processing. Indeed, several cyclophilins have been reported to be involved in RNA processing and localized in the nucleus, as predicted for all CRIPs and demonstrated in Paramecium. The small cyclophilin SnuCyP-20 was shown to be tightly associated with the U4 – U6 nuclear ribonucleoprotein in HeLa cells and was proposed to act as a chaperone in the folding steps, or to assist in the assembly/disassembly of spliceosomes (Horowitz et al., 1997; Teigelkamp et al., 1998). Recently, it has been shown that the peptidyl-prolyl isomerase ESS1 of S. cerevisiae is involved in mRNA 30 end processing and mitotic regulation (Morris et al., 1999). The cyclophilin CyP-40 induces conformational changes in the transcriptional factor v-Myb, changing its DNA-binding activity (Leverson and Ness, 1998), and another cyclophilin with a serine –arginine-rich domain interacts with the C-terminal domain of mammalian RNA polymerase II in a yeast two-hybrid system (Bourquin et al., 1997). Among the already known proteins possessing both RRM and cyclophilin domains, the first described cyclophilin with an RNA-binding domain was hCyP33 of unknown function in human T cells (Mi et al., 1996). In the genome of C. elegans, the CyP-13 gene encodes a protein with a cyclophilin domain preceded by an RRM that is proposed to be engaged in mRNA formation (Zorio and Blumenthal, 1999). As all described cyclophilins localized in the nucleus are involved in the processes of RNA transcription, splicing or regulation, we are tempted to think that the ‘CRIP’ family could also be engaged in RNA metabolism. This hypothesis is supported by the existence of the RRM domain and the nuclear localization of Kin241p in Paramecium. Role of the C-terminal domain for localization and stability of the CRIP The predicted nuclear localization of Paramecium CRIP was confirmed by experiments using GFP fusions. The fluorescent wild-type protein is localized precisely in the vegetative nucleus of Paramecium, in which all transcription events take place (Prescott, 1994). Similar Q 2001 Blackwell Science Ltd, Molecular Microbiology, 42, 257–267

263

experiments with the revertant Kin241-rev1 protein allowed us to confirm that the nuclear localization of CRIP is related to the predicted NLSs. The protein from the revertant line, which lacks three out of the six NLSs, is localized within the nucleus but also in the cytoplasm. This suggests that the three remaining NLSs can target the protein to the nucleus but less efficiently than in the wildtype protein. This is in agreement with observations made in other organisms, that all NLS signals within the protein are necessary for optimal nuclear import and that multiple NLSs can increase transport to the nucleus (Zhang et al., 2000). A probable function of the distal serine-rich region of the CRIP can be deduced from correlation of the sequences of the wild-type and mutant forms of the KIN241 gene with the GFP detection level. The kin241-1 mutation creates a premature stop codon that results in deletion of the C-terminal serine-rich region. Whereas mRNA levels corresponding to the GFP –kin241-1 fusion are similar to GFP –KIN241 levels, no GFP fusion protein can be detected. These results strongly suggest that the serinerich region is necessary for the stability of the protein. The analysis of revertant lines on which the final domain is restored confirmed the stabilizing function of the string of serines, as the protein Kin241-rev1 tagged with GFP is easily detectable Therefore, we can conclude that the kin241-1 phenotype is due to the absence of the Kin241 protein in the mutant, as is expected in the case of silenced cells. Differences and similarities among C-terminal regions of CRIPs The C-terminal parts of Kin241 homologue proteins are divergent in amino acid sequence and length (Fig. 3A). Nevertheless, independently of their differences, all C-terminal domains consist of a region of predicted coiled-coil structure rich in charged amino acids. Arrangement and percentage of each of the amino acids depends on the organism. In general, arginine residues alternate with serine, glutamate or aspartate residues. Such a structural element was found to be a common feature of factors involved in pre-mRNA processing (Neugebauer et al., 1995). The important feature of such sequences is an alternation of negatively and positively charged amino acids, in which serine is thought to be negatively charged as a result of phosphorylation (Neugebauer et al., 1995). Indeed, all CRIPs have positively and negatively charged residues that alternate regularly. In the Paramecium CRIP, the charged region consists of alternation of glutamate and lysine (EK domain) with, in addition, a considerable number of arginine residues. Recently, a region rich in glutamate and lysine alternation, similar to the Paramecium EK domain, was also found

264 A. Krzywicka et al. within the splicing factor SRrp86 from rat (Barnard and Patton, 2000). However, the best studied element presenting alternation of charged amino acids is the RSrich domain, with alternating arginine and serine residues. The Drosophila CRIP within its charged region possesses such a sequence motif with imperfect repeats of RSRSRX and of RS dipeptides, and such sequence patterns are present in the Drosophila splicing regulators Tra and SWAP and in several other SR proteins (for a review, see Blencowe et al., 1999; Graveley, 2000). The CRIP from Drosophila could thus be a member of the superfamily of SR-related proteins. However, in the other organisms, the number of RS repetitions within the charged region of CRIP is too small to be called an SR-related protein, although alternating positively and negatively charged residues are clearly present. CRIPs and morphogenesis Within the CRIP family, the function of only two members, from Paramecium and C. elegans, has been approached. In C. elegans, the homologue of the KIN241 gene was one of the targets of functional genomic analysis of C. elegans chromosome I by systematic RNA-mediated interference (RNAi). Inactivation of the CRIP gene in C. elegans, encoding the predicted F39H2.2 protein, did not give any phenotype, as was the case for about 86% of the genes localized on this chromosome (Fraser et al., 2000). This negative result may have different explanations, one of them being that CRIP could be redundant with another gene in C. elegans. In Paramecium, the CRIP gene is involved in cell morphogenesis as shown by the effect of its mutation/inactivation. Besides, the protein structure with its RRM and cyclophilin domains points to a possible role in RNA processing in agreement with its nuclear localization. The macromolecular, nucleoproteic complexes involved in RNA maturation are potential targets for cyclophilins that can regulate protein –protein interactions and function through conformational changes. Although we observed that, in kin241-1 cells, neither the constitutive splicing of two abundantly expressed genes nor the mRNA stability was disturbed (results not shown), it remains possible that the CRIP could be involved in the processing of some particular mRNAs belonging to the morphogenesis pathway. Experimental procedures Strains and cell culture The wild-type strain d4-2 of Paramecium tetrautrelia, a derivative of stock 51 (Sonneborn, 1974), and the kin241-1 mutant derived from this wild-type stain (Beisson et al., 1976) were used in this study. kin241-1 is a monogenic, recessive nuclear mutation that confers thermolethality at 358C, slow

growth and affects most aspects of cellular morphogenesis as described previously (Grandchamp and Beisson, 1981; Jerka-Dziadosz et al., 1992). In particular, nuclear reorganization after sexual processes (conjugation or autogamy) is perturbed: the zygotic nucleus may undergo more than the two normal post-zygotic divisions, yielding an increased number of new nuclei, which, in turn, seem to differentiate randomly into sexual or vegetative nuclei. Four revertant lines of wild-type phenotype, selected for growth at 358C after mutagenesis of kin241-1 cells, were also used. Genetic analysis of these independent revertants (kin241-rev-a, kin241-rev-b, kin241-rev-d, kin241-rev-g) suggested that they correspond to intragenic suppressors. (J. Beisson and M. Rossignol, unpublished results). Cells were grown at either 278C or 358C in buffered wheat grass powder (Pines International) infusion containing 0.4 mg ml21 b-sitosterol (Merck), bacterized the day before use with Klebsiella pneumoniae according to the usual procedures (Sonneborn, 1970).

Microinjection experiments The complementation cloning was carried out by sib-selection screening with microinjection into the vegetative nucleus of Paramecium of the DNA from the indexed genomic library as described previously (Keller and Cohen, 2000; Froissard et al., 2001). Each DNA sample was injected into at least 20 cells, and their progeny were cultured, observed and tested at 358C after 8 –10 divisions, and surviving or dividing cells were assessed as rescued cells. In some cases, the phenotype of complemented cells was ascertained by immunocytochemistry. Injection steps were repeated until a single library clone capable of wild-type phenotype restoration was isolated. The other microinjection experiments were performed as described (Ruiz et al., 1998; Froissard et al., 2001) using plasmid constructs or specific polymerase chain reaction (PCR) amplification products.

Fluorescence microscopy Immunofluorescence experiments were performed as described previously (Ruiz et al., 1998). The ciliary rootlets were labelled with a specific rabbit antiserum (Sperling et al., 1991) at a 1:500 dilution. Immunolabelled cells were observed under a Zeiss epifluorescence microscope and photographed using Kodak TMAX 400 film. Living cells were immobilized and observed under a film of mineral oil (Nujol) as described previously (Ruiz et al., 1998). Cells were treated by DAPI 5 min before observation in order to stain nuclei. Fluorescence microscopy on living cells expressing green fluorescent protein (GFP) reporter fusions were performed with a Leica MRD epifluorescence microscope PowerHAD, with phasecontrast or epifluorescence illumination. Images were captured by a 3CCD colour video camera system (Sony) and processed using PEGASUS software (2i System).

Sequencing and sequence analysis The region corresponding to the 3.2 kb insert containing the KIN241 gene was amplified on genomic DNA templates from Q 2001 Blackwell Science Ltd, Molecular Microbiology, 42, 257–267

A novel cyclophilin–RNA interacting protein each described cell line, and the products from at least three independent PCR reactions were selected for sequencing using the ABI Prism BigDye Primer cycle sequencing ready reaction kit on an ABI 310 sequencing instrument (PerkinElmer Biosystems). Initial characterization of the DNA and protein sequences, as well as amino acid percentage, were performed with the DNA STRIDER program (Marck, 1988). Sequence comparisons with public databases were performed using the BLAST programs (Zhang and Madden, 1997). Potential domains and motifs contained in the protein were detected and analysed with the programs INTERPROSCAN (Apweiler, 2000), SMART (Schultz et al., 2000) and EXPASY database (Bjellqvist et al., 1994). Signal sequences were searched for using PSORT II on the National Institute for Basic Biology server, Japan (Nakai and Kanehisa, 1991). Sequence alignments were performed using the CLUSTAL W program (Thompson et al., 1994) with BLOSUM 62 parameters and presented using the BOXSHADE 3.21 program (http://www.ch. embnet.org/software/BOX_form.html).

Molecular biology techniques Genomic DNA was extracted from exponential-phase cultures and fractionated by electrophoresis as described previously (Ruiz et al., 1998). Hybridized filters were autoradiographed using a PhosphorImager (Molecular Dynamics) and quantified by IMAGEQUANT software (Molecular Dynamics). Total RNA was prepared using the Trizol reagent (Gibco Life Technologies) according to the supplier’s protocol, exept that the cells were lysed by vortexing in the presence of glass beads. Total RNA (20 mg per lane) was electrophoretically separated, transferred and hybridized as described previously (Galvani and Sperling, 2000). RNA on all Northern blots was quantified subsequently using a probe for the ICL1 gene (Madeddu et al., 1996). The analysis of chromosomal DNA was prepared, separated by contour-clamped homogeneous electric field (CHEF) electrophoresis and hybridized as described previously (Madeddu et al., 1995). The KIN241 radioactive probe (corresponding to coding sequence of the KIN241 gene) was generated by PCR amplification using the same primers as for the silencing experiments and was synthesized by [32P]-dATP incorporation using Random Primers labelling system (Gibco Life Technologies), according to the supplier’s protocol.

PCR for microinjection and sequencing PCR reactions (50 ml) contained 10 ng of plasmid or 50 ng of genomic DNA and 25 pmol of 50 and 30 primer were carried out according to the recommended protocol for Advantage cDNA polymerase mix (ClonTech) for 35 cycles: denaturation (30 s at 948C), annealing (30 s at 588C) and extension (1.5 min to 3 min at 728C, depending on product size). The primers used were: 50 p103-GAAGAAATCCTAACAATGG; 30 p103-GACCA ATCAATAGTTCAATG for allelic sequencing and 50 kinAT G-GTCAATTGTATTAGAAACCTCATTGGG; and 30 kinTGATGAGGATGATGATGAACTACT for silencing experiments or blotting probe. After primer removal with Qiaquick PCR purification kit (Qiagen), the DNA was used for sequencing or filtered and concentrated to 20 mg ml21 for microinjection experiments (Froissard et al., 2001). Q 2001 Blackwell Science Ltd, Molecular Microbiology, 42, 257–267

265

Plasmid construction The plasmids containing the mutant kin241-1 and the revertant kin241-rev1 genes were generated by inserting PCR-amplified 3.2 kb fragments corresponding to the region containing the KIN241 gene. PCR products were obtained on kin241-1 and kin241-rev1 genomic DNA template using specific primers 50 p103, 30 p103, gel-purified using Qiaquick gel purification kit (Qiagen) and cloned into pGEM-T plasmid (Promega) according to instructions from the manufacturers. A synthetic GFP (plasmid GFP2L#1) was constructed using a two-step PCR method (Dillon and Rosen, 1990) to adapt the codon usage to Paramecium and to manipulate restriction sites in the sequence to facilitate cloning (E. Meyer and J. Cohen, unpublished results). When the GFP gene is cloned between cis-acting transcriptional elements of the calmodulin gene in the pPXV plasmid and microinjected into Paramecium cells, GFP can be produced and gives a diffuse dosedependent green fluorescence all over the cytoplasm, excluding the macronucleus (E. Meyer and J. Cohen, unpublished results). This was also observed for other GFP constructs introduced in Paramecium cells (Hauser et al., 2000; Haynes et al., 2000). N-terminal fusions of the KIN241, kin241-1 and kin241-rev1 genes with the synthetic GFP were obtained by introducing the PCR-amplified GFP gene into the Mfe I site. The primers for GFP fragment amplification, containing restriction Eco RI sites, were designed 50 GFPAAAAGAGA ATTCCAGGAGGAATGTCTAAAGGAGAAG AATTA and 30 G FP-AATATATGAATTCCTTCACCTTTGTATAATTCATCCAT. PCR reactions (50 ml) contained 10 ng of GFP2L#1 plasmid and 25 pmol of 50 and 30 primer were carried out as described previously. All constructs were verified by DNA sequencing using the ABI 310 sequencing system (Perkin-Elmer Biosystems).

Acknowledgements We thank Joelle Marie, Linda Sperling, Francoise Ruiz and France Koll for many useful discussions and critical reading of the manuscript. This work was supported by the Centre National de la Recherche Scientifique and the Association pour la Recherche contre le Cancer (contract 5425 to J.B.). A.K. was supported by the French governmental programme for graduate students (Co-Tutelle), a fellowship from the CNRS within the framework of the Jumelage franco-polonais, and La Ligue Contre le Cancer (Bourse de relai 2001).

Note added in proof Since the acceptance of this paper, the sequence of the CRIP mouse gene has been published in the public database (accession number AK014406). Its 100% homology with the partial human sequence underscores the high conservation of the CRIP family.

References Apweiler, R. (2000) Protein sequence databases. Adv Protein Chem 54: 31– 71. Bandziulis, R.J., Swanson, M.S., and Dreyfuss, G. (1989)

266 A. Krzywicka et al. RNA-binding proteins as developmental regulators. Genes Dev 3: 431 –437. Barnard, D.C., and Patton, J.G. (2000) Identification and characterization of a novel serine-arginine-rich splicing regulatory protein. Mol Cell Biol 20: 3049– 3057. Beisson, J., Rossignol, M., Ruiz, F., Adoutte, A., and Grandchamp, S. (1976) Genetic analysis of morphogenetic process in Paramecium: a mutation affecting cortical pattern and nuclear organization (abstract). J Protozool 23: 3A. Birney, E., Kumar, S., and Krainer, A.R. (1993) Analysis of the RNA-recognition motif and RS and RGG domains: conservation in metazoan pre-mRNA splicing factors. Nucleic Acids Res 21: 5803– 5816. Bjellqvist, B., Basse, B., Olsen, E., and Celis, J.E. (1994) Reference points for comparisons of two-dimensional maps of proteins from different human cell types defined in a pH scale where isoelectric points correlate with polypeptide compositions. Electrophoresis 15: 529– 539. Blencowe, B.J., Bowman, J.A., McCracken, S., and Rosonina, E. (1999) SR-related proteins and the processing of messenger RNA precursors. Biochem Cell Biol 77: 277–291. Bourquin, J.P., Stagljar, I., Meier, P., Moosmann, P., Silke, J., Baechi, T., et al. (1997) A serine/arginine-rich nuclear matrix cyclophilin interacts with the C-terminal domain of RNA polymerase II. Nucleic Acids Res 25: 2055 – 2061. Conti, E., Uy, M., Leighton, L., Blobel, G., and Kuriyan, J. (1998) Crystallographic analysis of the recognition of a nuclear localization signal by the nuclear import factor karyopherin alpha. Cell 94: 193 –204. Dillon, P.J., and Rosen, C.A. (1990) A rapid method for the construction of synthetic genes using the polymerase chain reaction. Biotechniques 9: 298 –300. Dupuis, P. (1992) The beta-tubulin genes of Paramecium are interrupted by two 27 bp introns. EMBO J 11: 3713 –3719. Fischer, G., Tradler, T., and Zarnt, T. (1998) The mode of action of peptidyl prolyl cis/trans isomerases in vivo: binding vs catalysis. FEBS Lett 426: 17 – 20. Fraser, A.G., Kamath, R.S., Zipperlen, P., Martinez-Campos, M., Sohrmann, M., and Ahringer, J. (2000) Functional genomic analysis of C. elegans chromosome I by systematic RNA interference. Nature 408: 325– 330. Froissard, M., Keller, A.M., Cohen, J. (2001) ND9p, a novel protein with armadillo-like repeats involved in exocytosis: physiological studies using allelic mutants in Paramecium. Genetics 157: 611– 620. Galat, A., and Metcalfe, S.M. (1995) Peptidylproline cis/trans isomerases. Prog Biophys Mol Biol 63: 67 –118. Galvani, A., and Sperling, L. (2000) Regulation of secretory protein gene expression in Paramecium role of the cortical exocytotic sites. Eur J Biochem 267: 3226 –3234. Grandchamp, S., and Beisson, J. (1981) Positional control of nuclear differentiation in Paramecium. Dev Biol 81: 336–341. Graveley, B.R. (2000) Sorting out the complexity of SR protein functions. RNA 6: 1197 –1211. Hauser, K., Haynes, W.J., Kung, C., Plattner, H., and Kissmehl, R. (2000) Expression of the green fluorescent protein Paramecium tetraurelia. Eur J Cell Biol 79: 144–149.

Haynes, W.J., Ling, K.Y., Saimi, Y., and Kung, C. (1996) Toward cloning genes by complementation in Paramecium. J Neurogenet 11: 81 –98. Haynes, W.J., Ling, K.Y., Preston, R.R., Saimi, Y., and Kung, C. (2000) The cloning and molecular analysis of the pawn-B gene in Paramecium tetraurelia. Genetics 155: 1105 –1117. Hedley, M.L., Amrein, H., and Maniatis, T. (1995) An amino acid sequence motif sufficient for subnuclear localization of an arginine/serine-rich splicing factor. Proc Natl Acad Sci USA 92: 11524– 11528. Horowitz, D.S., Kobayashi, R., and Krainer, A.R. (1997) A new cyclophilin and the human homologues of yeast Prp3 and Prp4 form a complex associated with U4/U6 snRNPs. RNA 3: 1374 – 1387. Iftode, F., Cohen, J., Ruiz, F., Torres-Rueda, A., Chen-Shan, L., Adoutte, A., and Beisson, J. (1989) Development of surface pattern during division in Paramecium. I. Mapping of duplication and reorganization of cortical cytoskeletal structures in the wild type. Development 105: 191 –211. Jerka-Dziadosz, M., de Garreau Loubresse, N., and Beisson, J. (1992) Development of surface pattern during division in Paramecium. II. Defective spatial control in the mutant kin241. Development 115: 319 –335. Keller, A.M., and Cohen, J. (2000) An indexed genomic library for Paramecium complementation cloning. J Eukaryot Microbiol 47: 1– 6. Leverson, J.D., and Ness, S.A. (1998) Point mutations in v-Myb disrupt a cyclophilin-catalyzed negative regulatory mechanism. Mol Cell 1: 203–211. Madeddu, L., Gautier, M.C., Vayssie, L., Houari, A., and Sperling, L. (1995) A large multigene family codes for the polypeptides of the crystalline trichocyst matrix in Paramecium. Mol Biol Cell 6: 649–659. Madeddu, L., Klotz, C., Le Caer, K.P., and Beisson, J. (1996) Characterization of centrin genes in Paramecium. Eur J Biochem 238: 121– 128. Marck, C. (1988) ‘DNA Strider’: a ‘C’ program for the fast analysis of DNA and protein sequences on the Apple Macintosh family of computers. Nucleic Acids Res 16: 1829 –1836. Mi, H., Kops, O., Zimmermann, E., Jaschke, A., and Tropschug, M. (1996) A nuclear RNA-binding cyclophilin in human T cells. FEBS Lett 398: 201 –205. Morris, D.P., Phatnani, H.P., and Greenleaf, A.L. (1999) Phospho-carboxyl-terminal domain binding and the role of a prolyl isomerase in pre-mRNA 30 -end formation. J Biol Chem 274: 31583–31587. Nakai, K., and Kanehisa, M. (1991) Expert system for predicting protein localization sites in gram-negative bacteria. Proteins 11: 95 –110. Neugebauer, K.M., Stolk, J.A., and Roth, M.B. (1995) A conserved epitope on a subset of SR proteins defines a larger family of pre-mRNA splicing factors. J Cell Biol 129: 899– 908. Prescott, D.M. (1994) The DNA of ciliated protozoa. Microbiol Rev 58: 233 –267. Rassow, J., and Pfanner, N. (1995) Molecular chaperones and intracellular protein translocation. Rev Physiol Biochem Pharmacol 126: 199 –264. Ruiz, F., Vayssie, L., Klotz, C., Sperling, L., and Madeddu, L. Q 2001 Blackwell Science Ltd, Molecular Microbiology, 42, 257–267

A novel cyclophilin–RNA interacting protein (1998) Homology-dependent gene silencing in Paramecium. Mol Biol Cell 9: 931– 943. Ruiz, F., Krzywicka, A., Klotz, C., Keller, A.M., Cohen, J., Koll, F., et al. (2000) The SM19 gene, required for duplication of basal bodies in Paramecium, encodes a novel tubulin, htubulin. Curr Biol 10: 1451 –1454. Russell, C.B., Fraga, D., and Hinrichsen, R.D. (1994) Extremely short 20 –33 nucleotide introns are the standard length in Paramecium tetraurelia. Nucleic Acids Res 22: 1221 –1225. Schultz, J., Copley, R.R., Doerks, T., Ponting, C.P., and Bork, P. (2000) SMART: a web-based tool for the study of genetically mobile domains. Nucleic Acids Res 28: 231– 234. Skouri, F., and Cohen, J. (1997) Genetic approach to regulated exocytosis using functional complementation in Paramecium: identification of the ND7 gene required for membrane fusion. Mol Biol Cell 8: 1063 – 1071. Sonneborn, T.M. (1970) Methods in Paramecium research. Methods Cell Physiol 4: 241 –339. Sonneborn, T.M. (1974) Ciliate morphogenesis and its bearing on general cellular morphogenesis. Tsitologiia 16: 1063 –1088. Sperling, L., Keryer, G., Ruiz, F., and Beisson, J. (1991)

Q 2001 Blackwell Science Ltd, Molecular Microbiology, 42, 257–267

267

Cortical morphogenesis in Paramecium: a transcellular wave of protein phosphorylation involved in ciliary rootlet disassembly. Dev Biol 148: 205 –218. Teigelkamp, S., Achsel, T., Mundt, C., Gothel, S.F., Cronshagen, U., Lane, W.S., et al. (1998) The 20kD protein of human [U4/U6.U5] tri-snRNPs is a novel cyclophilin that forms a complex with the U4/U6-specific 60 kD and 90 kD proteins. RNA 4: 127– 141. Thompson, J.D., Higgins, D.G., and Gibson, T.J. (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. Zhang, F., White, R.L., and Neufeld, K.L. (2000) Phosphorylation near nuclear localization signal regulates nuclear import of adenomatous polyposis coli protein. Proc Natl Acad Sci USA 97: 12577– 12582. Zhang, J., and Madden, T.L. (1997) PowerBLAST: a new network BLAST application for interactive or automated sequence analysis and annotation. Genome Res 7: 649 –656. Zorio, D.A., and Blumenthal, T. (1999) U2AF35 is encoded by an essential gene clustered in an operon with RRM/cyclophilin in Caenorhabditis elegans. RNA 5: 487 –494.