Eur. J. Biochem. 238, 121-128 (1996) 0 FEBS 1996
Characterization of centrin genes in Paramecium Luisa MADEDDU', Catherine KLOTZ', Jean-Pierre LE CAER2 and Janine BEISSON '
' Centre de GCnttique Moltculaire, Centre National de la Recherche Scientifique, Gif-sur-Yvette, France ' Institute Alfred Fessard, Centre National de la Recherche Scientifique, Gif-sur-Yvette, France (Received 12 January 1996)
-
EJB 96 0040/1
Centrins are highly conserved, ubiquitous cytoskeletal components which belong to the EF-hand superfamily of Caz+-modulated proteins. We report here the molecular characterization of new members of the centrin family, Paranzecium centrins. Previous studies described the organization of the infraciliary lattice (ICL), the innermost cortical cytoskeletal network of Paramecium, and showed that it was composed of a set of low-molecular-mass, Ca*+-binding polypeptides [Garreau de Loubresse, N., Klotz, C., Vigues, B., Rutin, J. & Beisson, J. (1991) Bid. Cell 71, 217-2251. In this paper we show that these polypeptides are recognized by specific anti-centrin polyclonal antibodies. Their microsequences revealed four distinct N-termini. For one of them, ICLI, N-terminal and internal peptide sequences were used for PCR amplification and cloning of a DNA fragment containing a complete centrin coding sequence. The deduced amino acid sequence presents about 50% identity with those of centrins from other species. Further molecular analysis allowed us to identify two additional closely related, co-expressed ICLl genes, providing the first example of a centrin multigenic family in a microorganism.
Keywords: Paramecium ; centrin; EF-hand Ca2+-bindingprotein; infraciliary lattice.
Within the large superfamily of EF-hand Ca2+-binding proteins, centrins constitute a subfamily of closely related, highly conserved cytoskeletal proteins. First identified in unicellular algae (Salisbury et al., 1984; Huang et al., 1988), centrins have now been Characterized at the molecular level from a broad range of eukaryotes, including yeast (Baum et al., 1986), higher plants (Bhattacharya et al., 1993) and humans (Lee and Huang, 1993 ; Errabolu et al., 1994). Analysis of their primary structure revealed that these small (-20 kDa), acidic polypeptides contain four similar domains of predicted helix-loop-helix EF-hand structure, like calmodulin and related Ca2+-modulated proteins (Moncrief et al., 1990). In contrast to other members of the EF-hand superfamily, which are known to be primarily regulatory proteins, centrins are characterized by unique cytoskeletal properties. In all the eukaryotic systems investigated so far, centrins or immunologically related proteins have been found in structures associated with microtubule-organizing centers: centrosomes, mitotic spindle poles, basal bodies (reviews: Bazinet et al., 1990; Melkonian et al., 1992; Schiebel and Bornens, 1995; Salisbury, 1995). While they appear to be restricted to centrosomes in mammalian cells and to spindle pole bodies in yeast, in diverse lower eukaryotes centrins assemble into Ca' -sensitive, contractile cy+
Correspondence to L. Madeddu, Centre de GCnCtique MolCculaire, Centre National de la Recherche Scientifique, F-91198 Gif-sur-Yvette, France Fax: + 33 1 69 82 31 50. Abbreviutions. 2D, two-dimensional; EEB, ecto-endoplasmic boundary; FITC antibody, fluorescein-isothiocyanate--antibodyconjugate; ICL, infraciliary lattice; RT-PCR, reverse transcription PCR. Enzymes. Trypsin (EC 3.4.21.4); endoproteinase AspN (EC 3.4.24.33). Note. The novel nucleotide sequence data published here has been deposited with the GenBank sequence data bank and is available under accession numbers U35344, U3.5397 and U35396 for ICLa, b and c, respectively.
toskeletal systems. These filamentous arrays are varied in their morphology and function, as exemplified by the striated rootlets and nucleus- basal body connectors of unicellular algae (Salisbury et al., 1984, 1988; Schulze et al., 1987; Taillon et al., 1992; Katsaros et al., 1993; Sanders and Salisbury, 1994), or the myonemes of the ciliate Eudiplodinum maggii (David and Viguks, 1994). The infraciliary lattice (ICL) of the ciliated protozoan Paramecium represents another example of a Ca"-modulated cytoskeletal structure. Constituting the innermost fibrous network of Paramecium cortical cytoskeleton, the ICL is formed of bundles of microfilaments organized in irregular polygonal meshes which run around the proximal ends of ciliary basal bodies (Allen, 1971 ; Garreau de Loubresse et al., 1988). Calcium-dependent contraction of the network has been observed in vivo, in response to physiological increases of cytosolic free Ca2+concentration, as well as in vitro (Garreau de Loubresse et al., 1988, 1991). Previous studies have shown that the ICL is composed of a set of immunologically related, Ca2+-binding polypeptides of low molecular mass (20-24 kDa) and acidic PI. The contractile properties of the ICL, and the physico-chemical characteristics of its major constituent polypeptides suggested that the ICL was formed of centnn-like molecules (Garreau de Loubresse et al., 1991). This hypothesis is confirmed here by the use of polyclonal antibodies raised against bacterially expressed Chlamydomonas centrin (Errabolu et al., 1994), which are shown to cross-react with the same major ICL components previously recognized as Ca2+-bindingproteins. Four different N-terminal sequences were obtained by microsequencing these polypeptides. Starting from N-terminal and internal peptide sequences of one of them (ICLl), we have amplified genomic DNA and obtained evidence for the existence in Paramecium of three distinct, co-expressed ICLl genes which code for nearly identical proteins. The com-
122
Madeddu et al. ( E m J. Biochem. 238)
plete deduced amino acid sequences unequivocally identified these proteins as centrins.
MATERIALS AND METHODS Cells and culture conditions. The wild-type strain d4-2 of Paramecium tetraurelia (Sonneborn, 1974) was used. Cells were grown at 27°C in an infusion of wheat grass powder (Pines International, Lawrence KA), inoculated with Enterobacter aerogenes and supplemented with 0.4 pg/ml p-sitosterol. Antibodies. The rabbit antiserum 26/14-1, raised against bacterially expressed Chlamydomonas centrin (Errabolu et al., 1994), was kindly provided by J. Salisbury (Department of Biochemistry and Molecular Biology, Mayo Clinic, Rochester). Immunofluorescence.Immunolabeling of Paramecium ICL was carried out essentially as previously described (Cohen and Beisson, 1988). After a 2-min permeabilization in buffer A (60 m M Pipes, 25 mM Hepes, 10 mM EGTA, 2 m M MgCl,, pH 6.9; Schliwa and Van Blerkom, 1981) containing 1 % Triton X-l 00, cells were fixed in 2 % paraformaldehyde in buffer A for 20-30 min. After washing, the cells were briefly incubated in 10 mM Tris/HCI pH 7.4, 0.15 M NaCl, 0.01 9k Tween-20 containing 3% bovine serum albumin and 5 mM CaCI,. The latter buffer was used at all subsequent steps: incubation of the primary antibody (IS min), washes, incubation in the secondary anti-rabbit FITC antibody (15 min ; Jackson ImmunoResearch Labs, West Grove PA) and two final washes. Cells were mounted in Citifluor (Citifluor Ltd, England), observed with a Zeiss epifluorescence microscope, and photographed with Kodak TMAX 400 film. Preparation of the ICL polypeptides. An ICL-enriched fraction was prepared as previously described (Garreau de Loubresse et al., 1988). Cell pellets were lysed in 1 mM Tris/HCI pH 8.5 containing 1.5% Triton X-100, 1 mM CaCI,, 0.1 mM MgCI,, 0.005 % heparin, 0.25 M sucrose, 0.02 M 2-mercaptoethanol and 1.3 M urea. The insoluble fraction, containing large pieces of ICL, was washed in 10 mM Tris/HCl pH 7.4, 0.25 M sucrose. The last step consisted in the specific solubilization of the ICL fragments by addition of 4 mM EGTA. The ICL polypeptides were recovered in the EGTA supernatant of a IS min, 20000 g centrifugation. Protease inhibitors (phenylmethylsulfonyl fluoride, 0.05 mg/ml ; leupeptine, 0.005 mg/ml) were present at all steps. Electrophoresis and blotting procedures. Isoelectric focusing was performed according to O’Farrell (1975) with 25 mM Mops as chemical spacer (Tindall, 1986). The second-dimension analysis (SDSPAGE) was performed according to Laemmli (1970) on 12 o/o acrylamide (37.5 : 1 acrylamides/bisacrylamide) gels. Proteins were electrophoretically transfered to membrane (Problott, Applied Biosystems) in 0.05 M Tris, 0.05 M boric acid (Bauw et al., 1989). For protein microsequencing, the spots to be sequenced (usually from eight duplicate filters) were cut after staining with 0.1 9k amidoblack in 45 % methanol and 7 7 0 acetic acid. For immunoreactions, the filters were incubated in 10 mM Tris/HCI pH 7.4, 0.15 M NaCI, 0.1 % Tween-20 containing 5 70non-fat dried milk. The secondary anti-rabbit antibodies (coupled to horseradish peroxidase ; Biosys, France) were used according to the recommendations of the supplier. N-terminal and internal protein sequences. Polypeptides adsorbed on Problott membranes were sequenced on a 470A protein sequencer (ABI, Perking Elmer) using a Problott cartridge. Phenylthiohydantoins were identified using an on-line analyser 120A (ABl). For the determination of internal sequences, spots from four two-dimensional (2D) gels were digested by trypsin or the endoproteinase AspN according to Ro-
senfeld et al. (1992). Peptides were purified by reverse-phase HPLC on a C8 30-nm column (220X2.1 mm). Peptides were eluted with a gradient of solvents A (0.1 % trifluoroacetic acid in water) and B (80% acetonitrile 20% water and 0.09% trifluoroacetic acid). The gradient was applied at a flow rate of 200 pl/min, solvent B was linearly increased by 1%/min. PCR amplification and sequence analysis. Paramecium tetruurelia genomic DNA was isolated from exponential phase cells as described by Dupuis (1992). For the first direct PCR amplification, partially degenerate oligonucleotide primers were designed on the basis of the ICLl peptide microsequences (shown in Fig. 2 C below). The nucleotide sequences, incorporating either an Xbal (sense primers) or an EcoRI (antisense primers) restriction enzyme site (underlined), were: ICLl :
5’-GCTCTAGACCWCCWCCWYARYARGC-3’ S’-GCGAATTCTRWGTWCCRTCWGTRTC-3’
PCR reactions (50 pl) contained 50-200 pmol of each primer, 50 ng genomic DNA, 0.1 mM dNTPs, and 2 U Tuq DNA polymerase (Boehringer, Mannheim). Reactions were overlayed with SO pl mineral oil, and performed through five cycles of denaturation at 90°C for 30 s, annealing at 40°C for 45 s, and extension at 72°C for 1.5 min, followed by 25 cycles of denaturation at 90°C for 30 s, annealing at 48 “C for 45 s, and extension at 72°C for 1.5 min. The amplification products were recovered using the QIAEX gel extraction kit (QIAGEN), then cloned into the Xbul and EcoRI sites of the pUC18 plasmid, according to standard protocols (Sambrook et al., 1989). DNA sequences were determined by the dideoxynucleotide chain-termination method using the T7 sequencingTMkit (Pharmacia). Internal, outward-facing oligonucleotides were then designed on the basis of the sequences determined by direct PCR and used to prime inverse PCR amplifications on templates consisting of circularized genomic DNA fragments (Ochman et al., 1990). The oligonucleotide sequences were : sense primer :
S’-GACCGAAGAGGAAGTTTTGG-3’
antisense primer:
5’-TTGATTCTTTTAAGCAGGGGG-3’
Genomic DNA was digested by RsaI, diluted to a final concentration of -1 ng/pl and incubated for = I 8 h at 14°C in the presence of T4 DNA ligase (0.02 U/pl; BRL). PCR arnplification was then performed on these circularized molecules (DNA final concentration: 0.2-0.4 ng/pl) as described above. The inverse PCR products obtained were cloned into the SmaI site of pUCl8 and sequenced on both strands. To verify the sequence obtained by inverse PCR, and to avoid errors introduced by Taq DNA polymerase infidelity, direct PCR amplification was carried out on undigested genomic DNA with the following oligonucleotide primers : sense primer: S’-GGCACGAAGAGGATAGTAACCACCACCC-3’ antisense primer:
S'-GC A AAGGTCTTTTTTGTCATAATGTTGTAG-3' After cloning into the SmuI site of pUC18, the different ICLl -coding DNA fragments generated were completely sequenced on both strands. The same oligonucleotides were then used to prime reversetranscription (RT) PCR, which was performed with total Paramecium RNA according to standard procedures, as described by Madeddu et al. (1995). The PCR products obtained were cloned and sequenced as described above.
Madeddu et al. (Eur: J. Biochem. 238)
123
TAG-3'. 10 pmol oligonucleotide was incubated with 10 pCi [y-"P]ATP (Amersham) for 1 h at 37"C, in the presence of T4 kinase (Biolabs). Southern blots. Genomic DNA (4 pg/lane) was restrictiondigested, fractionated by electrophoresis on 1 % agarose gels and transferred to Hybond-N' filters (Amersham) in 0.4 M NaOH. Hybridizations were performed according to Church and Gilbert (1984) either at 50°C (PCR-generated probes) or at 37°C (ICLla intron oligonucleotide probe). The membranes were then washed at the same temperatures utilized for hybridization, with decreasing concentrations of NaCKit (NaCl/Cit = 1.50 mM NaC1, 1.5mM trisodium citrate, pH 7.2), in the presence of 0.1 % SDS: 2XNaCKit for 20 min, followed by 0.2XNaCVCit for 20 min. Filters hybridized with PCR-generated probes were further washed in 0.2XNaCKit at 60°C for 1.5 min. Autoradiograms were obtained by exposing the filters to Hyperfilm-MP films (Amersham) at -80°C with intensifying screens. Fig. 1. Immunofluorescence images of the infraciliary lattice (ICL) labeled by the polyclonal 26/14-1 anti-centrin antibodies. (A) Ventral face; (B) dorsal face of the same Paramecium cell; oa, oral apparatus; cy, cytoproct. Bar: 10 pm.
Radioactive probes. PCR-generated, ICLl '*P-labelled probes were generated by amplification of cloned DNA fragments as previously described (Madeddu et al., 199.5). Briefly, DNA fragments corresponding to the ICLla, b or c coding regions were amplified in the presence of [(I-'~P]ATP(Amersham), using as templates recombinant pUCl8 plasmids containing the appropriate insert DNA. The primers used were the same as those described above for the original PCR. The ICLla intron probe was prepared using the following specific oligonucleotide (which corresponds to the second intron of the ICLl a gene) : S-GTATATTATATCTTAAATCATAA-
RESULTS Infraciliary lattice polypeptides are recognized by anticentrin antibodies. The Paramecium infraciliary lattice (ICL) is a contractile fibrous meshwork that underlies the whole cell cortex. Previous work (Garreau de Loubresse et al., 1988, 1991) showed that this cytoskeletal network could be isolated by taking advantage of its Ca*+-dependent assembly and contractility properties. Electrophoretic analysis established that the ICL is composed of several small (apparent molecular mass 2024 kDa), acidic polypeptides. The six most acidic 2D gel spots (of a total of 10 major spots; see Fig. 2A), were shown to bind Ca2+and to react with a polyclonal antiserum raised against the ecto-endoplasmic boundary (EEB) of the ciliate Isotricha (Viguks et al., 1984), which is, like the ICL, a Ca2+-modulated cytoskeletal network. Their biochemical characteristics (molecu-
Fig. 2. Characterization of the ICL polypeptides. (A) 2D gel (Coomassie blue stained) showing the low-molecular-mass (20-24 kDa) components of the ICL. The four less acidic spots, always present, have been shown to be neither Ca2+-bindingnor recognized by the antibodies which decorate the ICL. The numbers refer to the six major Caz+-binding polypeptides designated ICLl -1CL6. (B) Immunoblot corresponding to (A), reacted with the polyclonal 26/14-I antiserum. (C) N-terminal amino acid sequences (see labelling of spots in A) of ICL1, ICL2, ICL3, ICL5, ICL6, and internal peptide sequences for ICLl. One of the spots, ICL4, very weak both on the Coomassie-blue-stained gel and the immunoblot, was not microsequenced. The regions used to design the partially degenerated oligonucleotide primers for direct PCR amplifications of ICLl genomic sequences are underlined (see Materials and Methods).
124
Madeddu et al. ( E m J. Biochem. 238)
actggtattgttattaaattg
aacctaaccatcctactttgttagctgcaaaggaaggtataatccataggctcaatatcc acatctctatcactaatattcattattattaatttatatttaaattatagtttttattaa ttgcaaataattggaatcttctccttcctcaaagtattcccaaaactggatgaccttaag aggtggtccaaaaacgaattttttattattgtttttattctaataagagaaaaataaata
a
ttagaaatacagaaataaattagtgaatatcttttaataaataattatatttcttaataa
TAACAA A A A C A T CAACAA C A A...C A T 1 ATGGCACGAAGAGGATAGTACCACCACCCT A A . . . . . . TAAGCTCCCCCT~TAAAAGAATCAA W
61
A
R
R
G
Q
Q
P
P
P
Q
Q
Q
Q
A
P
P
A
Q
K
N
Q
G C T T G G A A A A G G T C C G A A C A ~GGAAAATTCAATCCAGCCGAATTTGTTAAACCTGGTTTGACCGAAGAGGAAGTTTTG T
G
K
F
N
P
A
E
F
V
K
P
G
L
T
E
E
E
V
181
I
K
A
F
D
L
F
D
T
D
Q
S
I
D
P
K
E
L
K
A
A
W
T
S
L
G
F
K
N
Q
T
I
Y
Q
W
I
S
D
L
D
T
D
G
S
G
T T T C C G T G T T C C G C AATTGATTTTGCAGAATTCTTGAAATTGATGACTGCCAGAATTAGTG~GAGATTCTAA I
361
D
F
A
E
L
K
L
W
T
A
R
I
S
E
R
D
S
T T T A T TA
T T
G G
T T
C C
A A
A A
T
GATTCTGAAAGAGCAGGAGTC-ACCTTMGGACCTTAGAAAGGTAGCAAAAGAATTG
D
S
E
R
A
G
V
V
T
L
K
D
L
R
K
V
A
K
E
T C AT C T T G T T T T G T T 481 GGAGAAACCATGGACGACTCAGAATTGC~G~TGATTGATAGAGCTGACTCAGATGGA E
T
M
D
D
S
E
L
Q
E
W
I
D
R
A
D
S
D
A
Q
V
T
F
E
D
F
Y
N
I
W
T
K
K
T
attcaatatcacataaaacataaaaacatattcaatatatattataagt
F
A
C
b a
C
b a
G
C T C T 541 GATGCTTAAGTGACATTTGAAGACTTCTACTTCTACAACATTATGAC~GACCTTTGCATGA D
C
b a
L
A
G
C
b a
K
T C TC gtatctaaaataataataattttcattag C C T T T C T gtacataaatatcactcaattcattctag C CC T AGCCGATATCTAAgtatat...tatatcttaaatcataatagAAAGTCTTTAATTTATTC A D I Q K V F N L F
A A 42 1
F
C
b a
Q
C
301
C
b a
E
C A C T C G C A TC G T AC C CT AGCAAAAAATCAAACCATTTACTAAATGATTAGTGATTTAGATACAGATGGAAGTGGTTA A
C
b a
G
tagT T T C G T A A C tagT TC G C A A aagCACCTAATCCATTGATCCAAAGGAATTAAAAGCTGCTGACTTCTTT~GATTTGA T
241
E
C
b a
L
T G G T T A T gtatgtattatttagataatct A C T A T gtataatatcatacttatttat 121 GAAATCAAAGAAGCATTTGACTTATTTGACACTGACGGgtatttaattgta.tcattatc E
C
b a
C
b a
*
a
Fig. 3. Nucleotide sequence of ICLla, which contains the entire protein coding sequence. The corresponding amino acids, in single letter code, are noted below. The 5’ and 3’ sequences, flanking the initial methionine and the TGA stop codon, respectively, as well as the introns interrupting the coding sequences, are in small characters. For ICLlb and ICLlc, whose 5’ and 3’ untranslated regions were not determined, only the nucleotides differing from the ICLla coding sequence are indicated. The sites which generate amino acid substitutions are underlined: the three dots at position 46-48 of the ICLl b sequence indicate the absence of the corresponding codon compared to the other two sequences. We note that the nucleotide sequences code for five additional amino acids compared to the N-terminal peptide sequence, and that initial methionine residues were not detected for either of the other three ICL microsequenced spots (Fig. 2). We do not know whether this reflects non-specific degradation of the ICL polypeptides during their isolation or physiological protein processing. Other microsequence ambiguities (see Fig. 2) could be due to the hetereogeneity of the ICLl components, and to the unusual richness in Gln and Pro residues of their N-termini.
Madeddu et al. (Eul: J. Biochem. 238)
lar mass, isoelectric point, Ca2+-binding capacity), and their ability to form filamentous structures with Caz'-dependent contractile properties, previously suggested that these proteins might be centrins (Garreau de Loubresse et al., 1991), which also form contractile fibrous arrays in other protists such as ChZamydomonas (Salisbury et al., 1988; Taillon et al., 1992; Sanders and Salisbury, 1994). Direct evidence that these proteins may well be centrins was obtained using polyclonal antibodies raised against bacterially expressed Chlamydo~onascentrin (Errabolu et al., 1994). In immunofluorescence experiments (Fig. I), the same staining pattern originally found with the anti-EEB serum was obtained. The antibodies were also capable of recognizing ICL proteins on Western blots. As shown on Fig. 2, the same six major spots previously found to bind calcium and react with the anti-EEB antibodies are also recognized by the anti-centrin antibodies. As a first step toward cloning genes coding for Paramecium centrins, we obtained N-terminal microsequences for the 2D gel spots recognized by the anti-centrin antibodies. After electrophoretic separation, the polypeptides were transferred to membranes for microsequencing. For one of the spots (ICLl), internal sequences were also generated from proteolytic fragments. Fig. 2 C shows the N-terminal sequences determined for five of the ICL polypeptides. Four distinct N-terminal sequences were found, a first indication that the complexity of the 2D gel pattern reflects differences in the primary structure of the polypeptides rather than post-translational modifications. ICLl and ICL2, however, have the same N-terminal sequence (at least over the first seven amino acids). It has been shown that the two isoforms of TetraseZmis centrin differ by their phosphorylation state (Martindale and Salisbury, 1990). The possibility that ICLl and ICL2 could also represent different phosphorylated isoforms of the same polypeptide is currently under investigation. The ICLl polypeptide was chosen for further molecular characterization, as enough microsequence information had been generated to design primers for PCR experiments.
Cloning and characterization of complete ICLl gene sequences. The ICLl peptide sequences underlined in Fig. 2 C were used to design partially degenerate primers for direct PCR amplification of genomic DNA. The amplification products were cloned in a plasmid vector, and sequence analysis of a number of clones led to the identification of a 141-bp DNA fragment that contained an open reading frame. The deduced amino acid sequence matched the peptide microsequences previously determined. To amplify the DNA regions flanking the gene fragment identified, the nucleotide sequence determined by direct PCR was then used to design primers for inverse PCR experiments. By amplification of circularized genomic DNA, we obtained a 967-bp PCR product which contained an entire ICLl coding sequence (Fig. 3; ICLla). Except for the short internal segment between the two outward-facing oligonucleotide primers, this sequence was identical, where overlapping, to the sequence previously amplified by direct PCR. In order to verify the inverse PCR sequence, direct PCR was performed on undigested genomic DNA using oligonucleotide primers designed to amplify DNA fragments corresponding to the complete coding sequence. Sequence analysis of several of these cloned DNA fragments revealed two additional nucleotide sequences. The three very similar centrin gene sequences (ICLla, b and c) are shown in Fig. 3. Evidence confirming that the three sequences do correspond to distinct genes was provided by Southern blot experiments performed on Paramecium genomic (macronuclear) DNA (Fig. 4). Specific PCR-generated probes corresponding to the entire ICLl
125
Fig. 4. Hybridization of genomic DNA with ICL1-specific probes. Paramecium genomic DNA was digested with BglII (B), HindIII (H), and XbaI (X). The same blot was probed with: (A) an ICLl 597-bp probe generated by PCR, using as templates recombinant pUC18 plasmids containing the genomic sequence corresponding to the ICLl b coding region ; (B) an oligonucleotide probe corresponding to the second
ICLla intron. The intron probe hybridizes with a single genomic fragment. Arrowheads indicate hybridization to the same restriction fragments. We note that a hybridization pattern equivalent to that in (A), even if with different relative band intensities, was found when plasmids containing ICLla or c genomic sequences were used as templates to generate the PCR probes. coding region recognize different restriction fragments for all digests (BgZII, HindIII and XbaI), consistent in each case with the existence of at least three distinct ICLl genes in the macronuclear genome. None of the ICLl sequences identified here contain BgZII or HindIII restriction sites, and only the ICLlb sequence contains an XbaI site. We note that a partial nucleotide sequence corresponding to a putative fourth ICLl gene was obtained by analyzing the short PCR products originally amplified using the degenerate oligonucleotide primers (not shown). Furthermore, the three coding sequences are interrupted, at the same positions, by extremely short introns (24-29 bp), typical of the Paramecium macronuclear genome (Dupuis, 1992; Russell et al., 1994). As for the genes encoding Chlamydomonas centrin and other proteins belonging to the EF-hand superfamily, the positions of the introns are not conserved and do not correspond to the boundaries of the EF-hand domains (Lee et al., 1991). Except for the conserved 5' and 3' splice sites, they present greater sequence variation than the surrounding coding regions, and thus provide gene-specific markers (Fig. 4), as previously observed for introns in Paramecium secretory protein genes (Madeddu et al., 1995). The nature of these elements was confirmed by RT-PCR experiments. As expected, the amplified cDNA fragments were =50 bp smaller than the genomic DNA fragments obtained with the same oligonucleotide primers (Fig. 5). Three distinct nucleotide sequences, identical to the ICLla, b and c coding sequences obtained by amplification of genomic DNA, were found by cloning and sequencing these cDNA PCR products. The absence of the introns in these sequences proves that cDNA and not contaminating genomic DNA has been amplified. The three ICLl genes are co-expressed. The 181-amino-acid deduced ICLl a sequence corresponds to a polypeptide with calculated molecular mass of 20300 Da and pl of 4.48, in agreement with the experimental data (Garreau de Loubresse et al., 1991). As identical pl values are predicted for ICLlb and ICLlc, it is unlikely that the microheterogeneity in ICLl polypeptide sequences can account for the ICL2 spot (see Fig. 2).
126
Madeddu et al. (Eul: J. Biochem. 238)
Fig.5. ICLl DNA fragments were generated by PCR amplification using as template either genomic DNA or cDNA generated by reverse transcription of total Paramecium RNA. The PCR products were separated on a 1.6% agarose gel and visualized by ethidium bromide staining. The cDNA amplification products (right lane) appear to be smaller than the genomic DNA amplification products (left lane) by 250 bp, approximately the sum of two ICLl introns. The weaker, lower (2400 bp) band in the left lane represents a distinct genomic amplification product unrelated to ICLl, which according to our partial sequence data is likely to correspond to the Pararnrcium homologue of ubiquitin thiolesterase.
The ICLl genes code for centrins. Comparison of the ICLla sequence with other EF-hand sequenced proteins confirms that ICLl is a centrin. The ICLla deduced amino acid sequence shares 55.6% identity with Chlamydomunas centrin (Huang et
identity with human centrins (50.6% with al., 1988) and ~ 5 0 % Hcenl, Lee and Huang, 1993; 49.3% with Hcen2, Errabolu et al., 1994) as compared to only 42.7% identity with Paramecium calmodulin (Kink et al., 1990). An alignment of the ICLla Paramecium centrin sequence with sequences of centrins from other species is shown in Fig. 6. Paramecium centrins display the highest percentage of identity with the centrin of Chlamydomolzas (which possesses contractile centrin-based organelles) and the lowest percentage of identity with yeast (where no centrinbased contractile structures are known). The conservation of these proteins is highest over the region of the four EF-hand domains, which are perfectly aligned with no gaps or insertions. Each of the potential ICLl Ca2'-binding site sequences is compatible with a functional helix-loop-helix EF-hand structure (Moncrief et al., 1990) except for residue 127 in the third site, which is occupied by a non-canonical Ala. The least conserved part of the molecule is the N-terminal domain. Not only is this region most divergent between Paramecium centrin and centrins from other species, but among the three ICLl sequences we have obtained, all of the non-conservative amino acid differences are concentrated in this variable Nterminal domain, the longest so far reported for a centrin. The amino acid sequences predicted for the genes b and c differ from that of ICLla by the insertion of two Gln residues (in italics in Fig. 3); in addition, the ICLla GCT/Ala codon is deleted in ICLlb, and substituted by an ACT/Thr codon in the ICLlc sequence, while nucleotides 61-63, coding for Thr in ICLla, specify an Ala in both ICLl b and c. All the other differences among the three genes, which share only ~ 8 5 % identity at the nucleotide level, are silent substitutions, except for a conservative change : position 130 of the protein sequences corresponds alternatively to Val (ICtla) or Ile (ICL1 b and c).
Fig.6. Comparison of amino acid sequences of centrins from different species. The amino acid sequences of centrins from human (Hcenl, Lee and Huang, 1993 ; Hcen2, Errabolu et al., 1994), Cklumydonzonus reinhardtii (Chlamy, Huang et al., 198S), Scher8elia dubiu (Scher, Bhattacharya et al., 1993), Nuegleriu grziheri (Nae, Levy, Lai, Remillard, Heintzelinan and Fulton, GenBank accession number U21725), AfripLex nummularia (Atri, Bhattacharya et al., 1993), yeast (Baum et al., 1986), and Parunzecium fetruureliu (Para, this study) are aligned. Identical residues are highlighted as white type on a dark background. The EF-hand Caz+-bindingdomains (EF-hd) are indicated by the asterisks underneath the sequences.
Madeddu et al.
DISCUSSION
J. Biochem. 238)
127
whose crystalline matrix is composed of at least 30 distinct immunologically related polypeptides, which are encoded by a Paramecium infraciliary lattice is a fibrous array which sub- huge multigenic family organized in sub-families (Madeddu et tends the whole cell cortex. Previous work described the Caz+- al., 1995). dependent contractile properties of this cytoskeletal network, The origin of these multigene families in Paramecium is still and identified its major components as immunologically related, unclear. We consider gene duplication more likely than a past small Ca”-binding proteins (Garreau de Loubresse et al., 1991). step of polyploidization, as single-copy genes, like that of calWe have presented here conclusive evidence that ICL structural modulin (Kink et al., 1990), are present in the genome of Paraproteins are centrins. First, we have shown that all the ICL Ca2+- mecium tetraurelia. It is also conceivable that the different verbinding polypeptides are recognized by specific anti-centrin an- sions of the ICLl genes are created during the macronuclear tibodies. Second, we have obtained nucleotide sequences for one differentiation process, as in ciliates the transcriptionally active of the polypeptides (ICLl). The predicted amino acid sequence polyploid macronucleus (somatic line) derives from the tranindicates that ICLl has all of the characteristics of a member scriptionally inactive diploid micronucleus (germ line) through of the EF-hand superfamily of Ca2+-binding proteins. Sequence extensive DNA rearrangements (Blackburn and Karrer, 1986). comparison reveals that it is a centrin, as the sequence shares However, we can rule out this possibility: sequences corregreater identity with centrins of distantly related species than sponding to two of the macronuclear genes (ICLIa and b) have with Paramecium calmodulin. Finally, we have presented evi- been identified in the micronuclear genome (Madeddu, unpubdence that in Paramecium a number of genes code for closely lished results), in the course of screening a Paramecium microrelated centrin molecules. nuclear library (kindly provided by J. Preer; Steele et al., 1994). There does not seem to be any direct relationship between Why does Paramecium use multigene families to code for centrin isotypic polymorphism and the diversity of structures cytoskeletal proteins such as centrins? While simple gene redunformed by the corresponding polypeptides. The simplest situa- dancy cannot be excluded for the moment, a more attractive tion is found in yeast; the cdc31 gene product, a most divergent hypothesis would be that these families of genes have been gencentrin, is localized to the spindle pole body and required for erated to respond to functional requirements. In particular, for spindle pole body duplication (Baum et al., 1986). Both molecu- the ICLl polypeptides, it is worth pointing out that all the nonlar and genetic evidence indicates that cdc31 is a unique, and an conservative amino acid variations are in the N-terminal region essential, gene (Spang et al., 1993). In humans, although centrin of the protein. This domain, which is the most divergent among appears to be restricted to a centrosomal localization, two centrin centrins of different species, is generally supposed to be ingenes have been characterized (Lee and Huang, 1993; Errabolu volved in protein-protein interaction (Bhattacharya et al., 1993 ; et al., 1994). It is still not clear whether their patterns of expres- Errabolu et al., 1994; Salisbury, 1995). It is tempting to suggest sion are tissue-specific ; co-expression has been found in retina that protein microheterogeneity is required to build up dynamic (Wolfrum and Salisbury, 1994). In protists, centrin assembles structures with complex geometries. Such microheterogeneity into a variety of contractile fibres. The best characterized exam- can be generated by several non-exclusive mechanisms : isotypic ple is provided by the flagellar basal body complex of Chlamy- polymorphism, post-translational modifications and associated domonas. Three morphologically and functionally distinct struc- proteins, as well documented for tubulins. In Parumecium, isotures, the nucleus basal-body connector, the distal striated fibers, typic polymorphism generated by gene duplication and diversifiand the stellate fibers all require centrin. Yet both molecular and cation could have been the privileged strategy for the assembly, genetic analysis show that there is a single centrin gene (Huang dynamics and functions of a complex Ca”-sensitive, contractile et al., 1988). In the $2-220 mutant, a point mutation in the fiber system like the ICL network. centrin gene leads to defective assembly of each of the three We are especially indebted to J. Salisbury for his generous gift of fibrous structures (Taillon et al., 1992; Taillon and Jarvik, 1995). Yet another situation is found in Paramecium, where centrins the 26/14-1 anti-centrin serum. We thank Jer6me Lesobre for excellent appear to participate in distinct cytoskeletal structures, in partic- technical assistance, and Marie-Christine Gautier and Laurence Vayssie for help with Southern blot and RT-PCR experiments. We are particuular the infraciliary lattice and the ciliary basal bodies (Klotz, larly grateful to Linda Sperling for invaluable support and critical advice Ruiz, Garreau de Loubresse and Beisson, unpublished results). throughout the present work. L. M. was supported by a senior fellowship Our data demonstrate that at least three different Paramecium of the EU Bridge Program and by a Poste Rouge from the Centre Numacronuclear genes code for centrin proteins. Furthermore the tional de la Recherche Scientifique (CNRS). This work was financed by other ICL polypeptides, given their immunological cross-reac- a grant (70194) from the Groupement d’Efudes et de Recherches sur le tivity with anti-centrin antibodies, are also likely to be centrins, Genome and by the CNRS. and each 2D gel spot may correspond to products of several genes. This is supported by preliminary data obtained for an- REFERENCES other ICL protein recognized by the centrin-specific antibody, ICL5 (see Fig. 2). Southern blot experiments with nucleotide Allen, R. D. (1971) Fine structure of membranous and microfibrillar systems in the cortex of Parumecium caudutum, J. Cell Riol. 49, probes corresponding to the N-terminal ICL5 peptide sequence 35-44. revealed hybridization patterns consistent with the existence of Baum, P., Furlong, C. & Byers, B. (1986) Yeast gene required for spindle three or four ICL5 genes (Madeddu, unpublished results). pole body duplication: homology of its product with Ca’+-binding The acquisition and maintenance of such multigene families proteins, Proc. Nutl Acud. Sci. USA 83, 5512-5516. may be a characteristic of Parumecium. In addition to the ICL Bauw, G., Van-Damme, J. M., Puype, M., Vandekerckhove, J., Gesser, B., Ratz, G. P., Lauridsen, J. B. & Celis, J. E. (1989) Proteinnetwork, other Paramecium subcellular structures appear to be electroblotting and microsequencing strategies in generating protein built up from sets of biochemically and immunologically related data bases from two-dimensional gels, Proc. Nutl Acud. Sci. USA 86, polypeptides. This is the case for the ciliary rootlets, which are 7701 -7705. involved in the alignment of the ciliary basal bodies (Sperling Bazinet, C. W., Baron, A. T. & Salisbury, J. L. (1990) Centrin: a et al., 1991) and for the epiplasm, a fibrous meshwork that uncalcium-binding protein associated with centrosomes, in Sfimulus rederlies the alveolar and plasma membranes of each cortical unit sponse coupling: the role ojintrucellulur cukium-binding proteins (Nahon et al., 1993). The most striking and best characterized (Smith, V. L. & Dedman, J . R., eds) pp. 39-56, CRC Press, Boca Raton FL. example is provided by the secretory organelles (trichocysts),
128
Madeddu et al. (Euv. J. Biochem. 238)
Bhattacharya, D., Steinkotter, J. & Melkonian, M. (1993) Molecular cloning and evolutionary analysis of the calcium-modulated contractile protein, centrin, in green algae and land plants, Plant Mol. Biol. 23, 1243-1254. Blackburn, E. H. & Karrer, K. M. (1986) Genomic reorganization in ciliated protozoans, Annu. Rev. Genet. 20, 501 -521, Church, G. M. & Gilbert, W. (1984) Genomic sequencing, Proc. Natl Acad. Sci. USA 81, 1991 -1995. Cohen, J. & Beisson, J. (1988) The cytoskeleton, in Paramecium (Gortz, H. D., ed.) pp. 363-391, Springer-Verlag, Berlin, Heidelberg. David, C. & Viguks, B. (1994) Calmyonemin: a 23 kD analogue of algal centrin occurring in contractile myonemes of Eudiplodinium maggii (ciliate), Cell Motil. Cytoskel. 27, 169-179. Dupuis, P. (1992) The P-tubulin genes of Paramecium are interrupted by two 27 bp introns, EMBO J . 11, 3713-3719. Errabolu, R., Sanders, M. A. & Salisbury, J. L. (1994) Cloning of a cDNA encoding human centrin, an EF-hand protein of centrosomes and mitotic spindle poles, J. Cell Sci. 107, 9-16. Garreau de Loubresse, N., Keryer, G., Viguks, B. & Beisson, J. (1988) A contractile cytoskeletal network of Paramecium: the infraciliary lattice, J. Cell Sci. 90, 351-364. Garreau de Loubresse, N., Klotz, C., Vigu&s,B., Rutin, J. & Beisson, J. (1991) Ca”-binding proteins and contractility of the infraciliary lattice in Paramecium, Biol. Cell. 71, 217-225. Huang, B., Mengerson, A. & Lee, V. D. (1988) Molecular cloning of cDNA for caltractin, a basal body-associated Caz+-binding protein : homology in its protein sequence with calmodulin and the yeast CDC31 gene product, J. Cell Biol. 107, 133-140. Katsaros, C . I., Maier, I. & Melkonian, M. (1993) Immunolocalization of centrin in the flagellar apparatus of male gametes of Ectocarpus siliculosus (Phaeophyceae) and other brown algal motile cells, J. Phycol. 29, 187-797. Kink, J. A,, Maley, M. E., Preston, R. R., Ling, K. Y., Wallen-Friedman, M. A., Saimi, Y. & Kung, C. (1990) Mutations in Paramecium calmoduiin indicate functional differences between the C-terminal and N-terminal lobes in vivo, Cell 62, 165-174. Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of the bacteriophage T4, Nature 227, 680-685. Lee, V. D., Stapleton, M. & Huang, B. (1991) Genomic structure of Ch/amydomonas caltractin. Evidence for intron inseltion suggests a probable genealogy for the EF-hand superfamily of proteins, J. Mol. Biol. 221, 175-191. Lee, V. D. & Huang, B. (1993) Molecular cloning and centrosomal localization of human caltractin, Proc. Natl Acad. Sci. USA 90, 11 03911043. Madeddu, L., Gautier, M.-C., VayssiC, L., Houari, A. & Sperling, L. (1995) A large multigene family codes for the polypeptides of the crystalline trichocyst matrix in Paramecium, Mol. Bid. Cell 6, 649659. Martindale, V. E. & Salisbury, J. L. (1990) Phosphorylation of algal centrin is rapidly responsive to changes in the external milieu, J. Cell Sci. 96, 395-402. Melkonian, M., Beech, P. L., Katsaros, C. & Schulze, D. (1992) Centrinmediated cell motility in algae, in Algal cell motility (Melkonian, M., ed.) pp. 179-221, Chapman and Hall, London. Moncrief, N. D., Kretsinger, R. H. & Goodman, M. (1990) Evolution of EF-hand calcium-modulated proteins. I. Relationships based on amino acid sequences, J . Mu/. Evol. 30, 522-562. Nahon, P., Coffe, G., Le Guyader, H., Darmanaden-Delorme, J., Jeanmaire-Wolfe, R., Cltrot, J.-C. & Adoutte, A. (1993) Identification of the epiplasmins, a new set of cortical proteins of the membrane cytoskeleton in Paramecium, J. Cell Sci. 104, 975-990.
Ochman, H., Medhora, M., Garz, D. & Hartl, D. L. (1990) Amplification of flanking sequences by inverse PCR, in PCR protocols (Innis, M. A., Gelfand, D. H., Sninsky, J. J. &White, T. J., eds) pp. 219-227, Academic Press, San Diego CA. O’Farrell, P. H. (1975) High resolution two-dimensional electrophoresis of proteins, J. Biol. Chem. 250, 4007-4021. Rosenfeld, J., Capdevielle, J., Guillemot, J. C. & Ferrara, P. (1992) Ingel digestion of proteins for internal sequence analysis after one- or two-dimensional gel electrophoresis, Anal. Biochem. 203, 173-179. Russel, C. B., Fraga, D. & Hinrichsen, R. D. (1994) Extremely short 20- 30 nucleotides introns are the standard length in Paramecium tetraurelia, Nucleic Acids Res. 22, 1221- 1225. Salisbury, J. L. (1995) Centrin, centrosomes, and mitotic spindle bodies, Cur%Opin. Cell Biol. 7, 39-45. Salisbury, J. L., Baron, A,, Surek, B. & Melkonian, M. (1984) Striated flagellar roots : isolation and partial characterization of a calciummodulated contractile organelle, 1. Cell Biol. 99, 962-970. Salisbury, J. L., Baron, A. T. & Sanders, M. (1988) The centrin-based cytoskeleton of Chlamydomonas reinhardtii: distribution in interphase and mitotic cells, J. Cell Biol. 107, 635-641. Sambrook, J., Fritsch, E. & Maniatis, T. (1989) Molecular cloning: a iaboratoly manual, 2nd edn, Cold Spring Harbor Laboratory Press, Cold Spring Harbor NY. Sanders, M. A. & Salisbury, J. L. (1 994) Centrin plays an essential role in microtubule severing during flagellar excision in Chlamydomonas reinhardtii, J. Cell Biol. 124, 795-805. Schiebel, E. & Bornens, M. (1995) Centrin conserved centrosomal proteins: a search for their functions, Trends Cell Biol. 5, 197-201. Schliwa, M. & Van Blerkom, J. (1981) Structural interactions of cytoskeletal components, J . Cell Biol. 90, 222-235. Schulze, D., Robenek, H., McFadden, G. I. & Melkonian, M. (1987) Immunolocalization of Ca2+-modulatedcontractile protein in the flagellar apparatus of green algae : the nucleus-basal body connector, Euv. J. Cell Biol. 45, 51 -61. Sonneborn, T. M. (1974) Paramecium aurelia, in Handbook of genetics 111 (King, R. C., ed.), pp. 469-594, Plenum Press. New York. Spang, A., Courtney, I., Fackler, U., Matzner, M. & Schiebel, E. (1993) The calcium-binding protein cell division cycle 3 1 of Saccharomices cerevisiae is a component of the half bridge of the spindle pole body, J . Cell Biol. 123, 405-416. Sperling, L., Keryer, G., Ruiz, F. & Beisson, J. (1991) Cortical morphogenesis in Paramecium: a transcellular wave of protein phosphorylation involved in ciliary rootlet disassembly, Dev. Biol. 148, 20521 8. Steele, C. J., Barkocy-Gallagher, G. A,, Preer, L. B. & Preer, J. R. (1994) Developmentally excised sequences in micronuclear DNA of Paramecium, Proc. Nut1 Acad. Sci. USA 91, 2255 -2259. Taillon, B. E., Adler, S. A,, Suhan, J. P. & Jarvik, J. W. (1992) Mutational analysis of centrin: An EF-hand protein associated with three distinct contractile fibers in the basal body apparatus of Chlamydornunas, J. Cell Biol. 119, 1613-1624. Taillon, B. E. & Jarvik, J. W. (1995) Central helix mutations in the centrosome-associated EF-hand protein centrin, Protoplasma 189, 203 - 215. Tindall, S. H. (1986) Selection of chemical spacers to improve isoelectric focusing resolving power: implications for use in two-dimensional electrophoresis, Anal. Biochem. 159, 287-294. Viguks, B., MCtknier, G. & Groliere, C. A. (1984) Biochemical and im-, munological characterization of the microfibrillar ecto-endoplasmic boundary in the ciliate lsotricha prostoma, Bid. Cell. 51, 61-78. Wolfrum, U. & Salisbury, J. L. (1994) Centrin expression in mammalian retinae, Mul. Biol. Cell Suppl. 5, 452a.
