A novel multifunctional factor involved in trans-splicing of chloropla
http://www.100md.com
《核酸研究医学期刊》
Department of Molecular Biology and Department of Plant Biology, University of Geneva 30 quai E. Ansermet, CH-1211 Geneva 4, Switzerland
*To whom correspondence should be addressed. Tel: +41 22 379 6188; Fax: +41 22 379 6868; Email: michel.goldschmidt-clermont@molbio.unige.ch
ABSTRACT
In the chloroplast of Chlamydomonas reinhardtii, psaA mRNA is spliced in trans from three separate precursors which assemble to form two group II introns. A fourth transcript, tscA, completes the structure of the first intron. Of the fourteen nucleus-encoded factors involved in psaA splicing, only two are required for splicing of both introns. We cloned and characterized the first of these more general factors, Raa1. Consistently with its role in psaA splicing, Raa1 is imported in the chloroplast where it is found in a membrane fraction and is part of a large ribonucleoprotein complex. One mutant, raa1-L137H, is defective for splicing of both introns, but another allelic mutant, raa1-314B, still expresses the 3' part of the Raa1 gene and is deficient only in splicing of intron 2. This observation and a deletion analysis indicate the presence of two domains in Raa1. The C-terminal domain is necessary and sufficient for processing of tscA RNA and splicing of the first intron, while the central domain is essential for splicing of the second intron. The combination of these two functional domains in Raa1 suggests that this new factor may coordinate trans-splicing of the two introns to improve the efficiency of psaA maturation.
INTRODUCTION
Group II introns are typically characterized by small elements of conserved primary sequence and a common secondary structure involving a central core and six radiating domains with helical stems. They also share many conserved tertiary structure interactions among the domains, as well as between the domains and the exons . Group II introns are prevalent in the organelle genomes of chloroplasts and mitochondria, and are also found in bacteria and archea . These introns are thought to be evolutionarily related to the introns in nuclear pre-mRNAs, where catalysis is mediated by the ribonucleoprotein complexes of the spliceosome, and proceeds by a similar mechanism involving two trans-esterification reactions and a lariat intermediate (4).
In the green alga Chlamydomonas reinhardtii, the psaA gene is composed of three exons which are at distant positions in the chloroplast genome and are transcribed separately (5,6). The precursor transcripts are spliced in trans to generate the mature psaA mRNA (Figure 1D). Flanking the exons, the sequences which constitute the split introns have characteristic features of group II introns. The second intron is assembled from the precursors of exon 2 and of exon 3. However the first intron is actually composed of three transcripts: the precursors of exon 1 and of exon 2, and a small non-coding RNA, transcribed from a separate locus, tscA (7). This intron in three pieces can be seen as an intermediate between the usual introns of group II, containing their own catalytic sequences, and the introns of nuclear pre-mRNAs, where part of the structure and catalytic activity have been proposed to reside in the trans-acting RNAs of the snRNPs.
Figure 1 Characterization of raa1-314B and raa1-L137H. (A) Immunoblot analysis of total cell extracts from the wild-type (WT), the mutants raa1-314B and raa1-L137H, and the mutants rescued with the genomic DNA (cosmid #2), or the midigene construct. Top, PsaA; bottom, small subunit of Rubisco. (B) Immunoblot analysis with anti-Raa1 antibodies (affinity purified) of total cell extracts from the WT and the mutants raa1-L137H and raa1-314B. The arrow labels a non-specific band. (C) RNA blot analysis with probes (indicated on the left) specific for psaA exons 1, 2, 3, the 5' part of intron 2 (5'i2), tscA RNA or psbA transcript as a control. Symbols correspond to the transcripts on the splicing scheme (panel D), except the white square which labels a non-specific hybridization signal. (D) Trans-splicing of psaA mRNA. The precursors of exon1 (black square, 0.4 kb), exon 2 (black circle, 3.6 kb) and exon 3 (open circle, 2.4 kb) are spliced in trans to produce intermediates (exon1–exon2, asterisk, 2.6 kb or exon2-exon3, diamond, 3.8 kb) and mature psaA mRNA (black triangle, 2.8 kb). Splicing of the first split intron requires tscA RNA (black star, 0.4 kb) which is processed from longer precursors (open stars) containing chlN.
While some group II introns are well-studied ribozymes that can self-splice in vitro, others have been refractory to self-splicing in the test tube. Both types require accessory protein factors for efficient splicing in vivo. Because the catalytic activity is carried by the RNA, these factors may function as RNA chaperones, may assist proper folding of the intron, or may stabilize it in the active conformation (8).
In Chlamydomonas, a set of at least fourteen nuclear loci are essential for trans-splicing the two split group II introns of psaA (9). Three of these factors have been characterized more in detail, Raa3, Raa2 and Rat1. Raa3 is necessary for trans-splicing of the first intron . This novel protein is part of a large complex which also contains the precursor of exon 1 and the tscA RNA. Raa2 is essential for trans-splicing of the second intron . Although Raa2 is related to pseudouridine synthases, this enzymatic activity is not required for its function in trans-splicing. Rat1 is required for processing of tscA RNA from a larger precursor and for splicing of the first intron (12). Rat1 shows sequence similarity to the NAD+ binding domain of poly(ADP-ribose) polymerase, but this domain can be mutated without affecting the function of Rat1 in splicing. These factors are highly specific since they are each required for splicing of only one of the introns. However, two nuclear loci encode factors that are required for splicing of both introns, and may thus have a more general role in splicing of group II introns . Here we report the characterization of the first of these factors, Raa1, which is required for trans-splicing of the two introns of psaA. We map two distinct domains of Raa1, one involved in processing of tscA and splicing of intron 1, the other in splicing of intron 2. Our results show that Raa1 is a multifunctional protein required, as a component of a large membrane-bound and RNA-containing complex, for the splicing of these group II introns.
MATERIALS AND METHODS
Strains and media
Procedures for growing C.reinhardtii and media were described (13) (TAP: Tris Acetate Phosphate; HSM: High Salt Minimal). The strain arg2 cwd also has a plastome mutation conferring spectinomycin resistance and was obtained from Dr R. Loppes (University of Liège, Belgium). Mutant L137H was obtained by mutagenesis with 5-fluoro-deoxy-uridine and UV, as described previously (6,9).
Oligonucleotides
See Supplementary Table I.
Generation of the 314B mutant strain
arg2 cwd cells were transformed with plasmid pARG7.8 (14) and selected on TAP medium without arginine in the dark. Screening of ca. 4 x 103 colonies yielded six strains with the fluorescence induction kinetics of mutants deficient in photosystem I or the b6f complex and a psaA RNA profile characteristic of trans-splicing mutants (9). Mutant 314B was chosen for further study because analysis of fourteen progenies from a back-cross to the wild-type showed co-segregation of the pARG7.8 insertion with the trans-splicing mutant phenotype, suggesting that the mutation was tagged.
Transformation of Chlamydomonas
For transformation of raa1-314B, 3 x 107 cells were treated with 1 μg of the appropriate plasmid DNA by the glass bead method (15) and spread on a single agar plate (HSM or TAP plus 10 μg/ml paromomycin).
The raa1-L137H strain was transformed by electroporation . A total of 108 cells of the mutant were transformed with 1 μg of the appropriate plasmid DNA in presence of 50 μg of salmon sperm DNA and spread on a single plate as above. The electric pulse was applied in the absence of a shunt resistor with 700 V and 25 μF.
RNA blot hybridization
RNA blot analysis of psaA transcripts and the hybridization probes were described previously (11). The tscA probe was a 720 bp PCR fragment (7).
Cloning of the Raa1 gene
For cloning of Raa1, a genomic library of BamHI-digested DNA fragments from raa1-314B was constructed in bacteriophage lambda EMBL4 (17). The library was screened by hybridization with radiolabelled pBR329, the plasmid vector of pARG7.8 (14). Restriction mapping of one of the positive phage (P23) and DNA blot hybridizations with the pARG7.8 probe, or with total genomic DNA from C.reinhardtii as a probe, allowed the identification of a 1.8 kb HindIII fragment that contained neither parts of the pARG7.8 plasmid, nor repetitive sequences. This fragment was used to screen a cosmid library of wild-type C.reinhardtii DNA (18). Three positive cosmids were obtained and used for transformation of the raa1-314B mutant and selection on minimal medium (HSM). All three were capable of rescuing phototrophic growth. Cosmid #2 was used for restriction mapping and subcloning. A minimal region was identified spanning the ClaI to ScaI sites. Probes from this region were used to screen cDNA libraries (see below).
Raa1 was also independently identified by transformation of L137H with an ordered cosmid library (19) and selection on HSM in the light. One cosmid was found to restore the wild-type phenotype. This cosmid (30 kb insert) was used for restriction mapping and subcloning, and was found to contain the Raa1 gene.
Complementation analysis in vegetative diploids
Vegetative diploids were obtained by fusion of either raa1-L137H or raa2-A18 with raa1-314B. To allow selection of diploids, paromomycin-resistant derivatives of raa1-L137H and raa2-A18 strains were obtained by transformation with the aphVIII gene . The raa1-314B strain has a chloroplast mutation conferring resistance to spectinomycin.
Removal of the cell wall was achieved by incubating loopfuls of each strain with autolysin in dim light for 1 h at 25°C. Vegetative fusion was induced by spreading the cells on selective plates in fusion solution or a control solution . Diploids were selected in dim light on TAP plates containing paromomycin (10 μg/ml) and spectinomycin (120 μg/ml).
The genotype of raa1-L137H/raa1-314B diploids was verified by PCR amplification of the aphVIII gene (characteristic of raa1-L137H) with oligos paro2-5' and paro2-3' and of the pARG 7.8 plasmid (characteristic of raa1-314B) with oligos pBRBam2 and ArgSau2. The genotype of raa2-A18/raa1-314B diploids was verified by amplification of aphVIII and of Raa2 with oligos A18/rbcS and A18/Nci1.
Immunoblot analysis
Protein samples were analyzed on 5% polyacrylamide gels reinforced with 0.15% linear poly(acrylamide-co-acrylic acid) (Aldrich). Proteins were transferred to nitrocellulose for 16 h at 25 V. Raa1 rabbit antibodies were obtained by immunization with a mixture of three histidine-tagged polypeptides corresponding to residues 1009–1349, 1510–1829 and 1830–2103 of Raa1. They were affinity purified using the three antigens immobilized on Ni NTA. The monoclonal HA-11 antibody was obtained from Eurogentec.
Raa1 genomic and midigene constructs
The subclone gHi (116b) was obtained by inserting the 12 kb HindIII fragment from cosmid #2 into Bluescript KS (+) digested with HindIII. Subclone gSc was obtained by inserting the 11 kb EcoRI-ScaI fragment from 116b into Bluescript KS (+) digested with EcoRI and SmaI. Subclone gSa was derived from 116b by partial digestion with SalI and religation.
The cDNA clone C11 (Supplementary Figure 1) was isolated by screening a library constructed in bacteriophage lambda gt10 (21) with the 0.8 kb SalI–BstXI genomic fragment containing exons 11 and 12. The cDNA clone 405 (Supplementary Figure 1) was found by screening a library constructed in lambda NM1149 (H. Sommer, personal communication) with the 1.5 kb SalI–SphI fragment from C11 (exons 6–14). The cDNA clone 551 (Supplementary Figure 1) was obtained from a dedicated cDNA library. This library was prepared by priming synthesis of the first strand of cDNA with oligos i2 and m3 and cloning the double-stranded cDNA with EcoRI-NotI adaptors (Amersham-Pharmacia) into lambda gt10. This library was screened with a 0.68 kb EcoRI–BglII from 405 (exons 4–6). All cDNA inserts were excised with EcoRI and cloned into Bluescript KS (+). The open reading frame (ORF) that the cDNA clones defined encodes a predicted polypeptide of 2103 amino acid. It is preceded by stop codons in every frame, implying that the entire coding sequence has been obtained.
The HA-tagged midigene, mWT (also called pI2), was constructed as follows. Subclone B04 was derived by cloning the 9 kb BscI (ClaI) fragment from gHi into Bluescript KS (+) digested with BscI. A 3.4 kb SalI fragment from cDNA C11 was cloned into SalI-digested B04, to yield plasmid 269 (the SalI site used for this fusion is in exon 6). The C11 cDNA (recloned in pET15b, Novagen) was tagged with a His6-Myc-BstEII-Kanr-BstEII sequence introduced in the SmaI site in exon 6 (5'-ctcatgcgcggctcccaccaccaccaccaccacctcggcgagcagaagctgatctccgaggaggacctgggtaacc-Kanr-gtaacctggtc-3', the Kanr cassette was derived from pUC4K, Amersham-Pharmacia). This tagged cDNA sequence was then transferred as a SalI–StuI fragment into plasmid 269, to yield plasmid 349. The triple HA epitope fragment (22) was then inserted as a NruI–NaeI fragment into 349 digested with BstEII to replace the Kanr cassette and obtain plasmid 490. Finally the 3.5 kb SacI fragment of 490 (one SacI site in the polylinker, the other in exon 3) was replaced with the 7 kb fragment from 116b to provide the HA-tagged midigene pI2.
The aphVIII gene, conferring paromomycin resistance (20) was amplified by PCR from plasmid pSL17 (S. Lemaire and J. D. Rochaix, unpublished data). The construct mBg was derived from pI2 by digestion with BglII (at the intron 6 and exon 7 junction of Raa1) and KpnI (in the polylinker) and ligation to the aphVIII gene (amplified with Paro Bgl 5'and Paro 3' and digested with BglII and KpnI). The construct mKp was derived from pI2 by digestion with KpnI (in exon 16 of Raa1 and in the poly linker) and ligation to aphVIII (amplified with Paro Kpn 5' and Paro 3' and digested with KpnI). The construct mAM was derived from pI2 by digestion with AgeI, blunting with mung bean nuclease, digestion with MscI (in exon 2 of Raa1) and religation. In a second step, the aphVIII gene was amplified with Paro Mlu 5' and Paro Mlu 3', digested with MluI and inserted in the MluI site, upstream of Raa1. The clone pI4 was similarly derived from pI2 by inserting the aphVIII gene in the MluI site. The construct mNS was derived from pI4 by digestion with NruI and SmaI (in exons 2 and 4 of Raa1, respectively) and religation.
RT–PCR analysis of Raa1 transcripts
Total RNAs were isolated as described (23) and then treated with RQ1 DNase (Promega). Reverse transcription was performed at 50°C with M-MLV reverse transcriptase (Promega). To analyse three different regions (exons 1 and 2, exons 12 and 13 and exons 17 and 18) of the Raa1 transcript, the corresponding first strand cDNA was synthesized with oligos AUG 2, middle 4 or oligo dT, respectively. PCR amplification was performed with the following pairs of oligos: AUG5'-1 and AUG3'-1; m5'-3 and m3'-3; STOP5'-6 and STOP3'-6, respectively. The PCR products span two exons so that amplification of the cDNA can be distinguished from amplification of any contaminating genomic DNA which would include the intervening intron. As control, RT–PCR analysis was also performed for Cbl transcripts (24), using oligo dT to synthesize the first strand cDNA and oligos UP and DOWN for the PCR. The Arg7–Raa1 fusion transcript in raa1-314B was revealed by synthesizing the first strand cDNA with oligo middle 4 (in exon 14 of Raa1) and PCR amplification with oligos Arg7ex11 and m3'-3 (in exon 13 of Raa1).
Sequence analysis of Raa1
The sequence of the Raa1 cDNA is deposited in the EMBL database under accession no. AJ605114 . The genomic sequence is now available (scaffold 33) in the draft of the C.reinhardtii genome, at the Joint Genome Institute of the US Department of Energy (http://genome.jgi-psf.org/chlre2/chlre2.home.html). Comparison of the cDNA and genomic sequences shows that the Raa1 gene is composed of 18 exons with very different sizes (from 60 to 2473 bp). The polyA tail is preceded 11 bp upstream by the TGTAA consensus signal typical of C.reinhardtii (25).
Transit peptide sequences were predicted using the algorithms ChloroP (26) (http://www.cbs.dtu.dk/services/ChloroP/) and Predotar (27) (version 0.5; http://www.inra.fr/predotar/). Protein secondary structure was predicted with GORIV (28) (http://npsa-pbil.ibcp.fr/cgi-bin/npsa_automat.pl?page=npsa_gor4.html)
Cell fractionation
Cells were lyzed with a nebulizer and chloroplasts were purified on Percoll gradients as described previously (10) in the presence of protease inhibitor mix A (PIM A: 1 mM -caproic acid, 0.5 mM benzamidine, 0.6 μg/ml leupeptin, 2 μM E64, 2 μM pepstatin and 0.5 mM AEBSF). Chloroplasts were resuspended in hypotonic buffer supplemented with protease inhibitor mix B (PIM B: 1 mM -caproic acid, 0.5 mM benzamidine, 0.6 μg/ml leupeptin, 2 mM ortho-phenantroline and 1/200 Protease Inhibitor Cocktail P8849 from Sigma). For further fractionation, the chloroplasts (250 μl, 7 mg protein/ml) were lyzed by sonication, the solution was adjusted to 20 mM HEPES (pH 7.8), 50 mM KCl and 10 mM MgCl2 (HepKM) and subjected to centrifugation at 100 000 g for 30 min at 4°C in a TLA45 rotor (Beckman). The supernatant was recovered for analysis, and the pellet was washed by resuspension in HepKMS (0.2 M sucrose, HepKM, PIM B) and centrifugation at 50 000 g for 15 min at 4°C in a TLA45 rotor. The supernatant was discarded and the pellet was resuspended in HepKM plus PIM B.
For analysis of the protein complexes, cells were collected by centrifugation, resuspended at a density of 2.5 108/ml in HepKM plus PIM A, and lyzed by freezing and thawing. The lysate (1 ml aliquot) was subjected to centrifugation at 18 000 g for 20 min at 4°C in a TLA45 rotor (Beckman). The pellet containing the membranes was resuspended in HepKM, 0.2 M (NH4)2SO4, PIM B. Where needed, 100 μl of RNaseA were added [10 mg/ml in 100 mM Tris–HCl (pH 7.5) and 10 mM Na acetate) or the same amount of buffer without RNase (mock treatment), followed by incubation at 16°C for 20 min. Membranes and insoluble material were removed by two rounds of centrifugation at 120 000 g for 20 min in a TLA45 rotor, and the supernatant was loaded on sucrose gradients (10 ml; 0.1–1.3 M sucrose, HepKM, PIM A). After centrifugation at 115 000 g for 17 h in the SW40 rotor (Beckman), 10 fractions (1.1 ml) were collected from the bottom by puncturing the tube. Sedimentation was calibrated with the High Molecular Weight Calibration Kit from Amersham-Pharmacia.
RESULTS
Characterization of the raa1-314B mutant deficient in splicing of psaA intron 2 and cloning of the Raa1 gene
Insertional mutants were generated by transformation of a Chlamydomonas arginine auxotroph with a plasmid containing the wild-type Arg7 gene. Mutant raa1-314B was unable to grow phototrophically (data not shown) and had no detectable PsaA protein (Figure 1A, lane 2). RNA blot hybridization (Figure 1C) showed that the pattern of transcripts that accumulate in raa1-314B is characteristic of a block in trans-splicing of the second intron of psaA . Probes specific for each of the three exons hybridize to mature psaA mRNA (black triangle, see Figure 1D) in the wild-type (Figure 1C, lane 1) but not in the mutant (lane 2). In the raa1-314B mutant, the probe for exon 3 only detects an unspliced precursor of exon 3 (open circle), which is also present in the wild-type. In the mutant, probes for exon 1 or for exon 2 reveal a splicing intermediate (asterisk) with exons 1 and 2 spliced together, which migrates slightly faster than mature psaA mRNA. This intermediate, with the 5' part of intron 2 still attached to exon 2 (5'i2 probe, lane 9), is not detected in the wild-type (lane 8). The precursor of exon 1 (black square) over accumulates in raa1-314B, as was observed previously for all other psaA splicing mutants (6). Because of these defects in splicing of psaA transcripts, we have called the mutant raa1-314B (RNA maturation of psaA).
Using the Arg7 insertion as a molecular tag, we cloned the raa1 mutant locus. We then obtained wild-type RAA1 cosmids that could rescue the raa1-314B mutant by transformation. These results also showed that the raa1-314B mutation is recessive to the wild-type. Additionally, Southern blot analysis of raa1-314B DNA indicated that the locus was rearranged (data not shown), confirming that the locus that we cloned was indeed RAA1. In the strains transformed with cosmid #2, PsaA protein is recovered (Figure 1A, 314B genomic, lane 3) and the fully spliced mRNA accumulates (Figure 1C, lane 4, black triangle). The splicing intermediate containing the 5' part of intron 2 is not detected any more (Figure 1C, lane 11, asterisk) and the levels of exon 1 precursor are much lower than in the parental mutant (Figure 1C, lanes 2 and 4, black square).
To map the gene within the 34 kb insert in cosmid #2, we tested subclones for their ability to restore phototrophic growth of raa1-314B by transformation and identified a minimal region of 7.4 kb. We obtained cDNA clones from this region which encode a predicted polypeptide of 2103 amino acid (See Materials and Methods). To ascertain that the appropriate cDNA had been cloned, a ‘midigene’ was constructed with the genomic sequence containing the 5' part of the gene (including the promoter) fused to the cDNA from the 3' part (Supplementary Figure 1D). This construct transformed the raa1-314B mutant to prototrophic growth as efficiently as cosmid #2 (data not shown) and restored normal levels of PsaA protein (Figure 1A, lane 4). The midigene also rescued trans-splicing of the psaA mRNA (black triangle, Figure 1C, lane 6), even if splicing intermediates are still detected with probes for exon 2 (black circle), exon 3 (diamond), exon 1 (black square) and 5'i2 (black circle, lane 13).
L137H, defective in splicing of both psaA introns, is an allele of Raa1
In parallel, we have characterized another mutant strain (L137H), that was previously shown to be affected in trans-splicing of both the psaA introns (class B) and was obtained by UV mutagenesis (6). L137H was unable to grow phototrophically (data not shown) and did not accumulate detectable levels of the PsaA protein (Figure 1A, lane 6, the mutant is labelled raa1-L137H in anticipation of evidence presented below which shows that L137H is allelic to raa1-314B). The pattern of transcripts in this mutant (Figure 1C) is characteristic of a block in trans-splicing of both introns (6). Probes specific for each of the three exons fail to detect mature psaA mRNA in the mutant (black triangle, lane 3), but revealed unspliced precursors of exon 1 (black square), exon 2 (black circle) and exon 3 (white circle). The exon 2 precursor, with the 5' part of intron 2 still attached to exon 2, was also detected with the 5'i2 probe (black circle, lanes 3 and 10). As noted above for raa1-314B and other trans-splicing mutants, the precursor of exon 1 over accumulates.
To clone the mutant locus, we transformed the L137H strain with an indexed genomic library (19) and screened the cosmids for their ability to restore phototrophic growth of the mutant cells. Surprisingly, we identified Raa1 as the nuclear gene that complements the photosynthetic defect of the L137H strain, suggesting that L137H and raa1-314B are allelic. These results also imply that the L137H mutation is recessive to the wild-type. However, Southern blot analysis did not show any alteration in the RAA1 locus of the raa1-L137H mutant (data not shown). In the L137H cells transformed with either the cosmid containing the Raa1 locus or the midigene, synthesis of the PsaA protein is recovered (Figure 1A, lanes 7 and 8) and fully spliced mRNA accumulates (black triangle, Figure 1C, lanes 5 and 7).
To confirm that raa1-314B and L137H are indeed allelic, we constructed vegetative diploids to test whether the two mutants could genetically complement (Supplementary Data). The observed lack of complementation, together with the fact that transformation with the Raa1 gene can rescue psaA splicing in both mutants, indicate that raa1-314B and L137H are allelic. Furthermore full-length Raa1 protein was not detectable by immunoblotting in the two mutant strains, raa1-314B and L137H (Figure 1B, lanes 2 and 3). We thus refer to the two mutants as raa1-314B and raa1-L137H.
The raa1-L137H strain is also defective in maturation of tscA RNA
HN31, another mutant defective in trans-splicing of both introns of psaA (class B), is allelic to L137H (9). Interestingly, HN31 was previously shown to be defective in the maturation of tscA (29), the small non-coding RNA required for splicing of intron 1 (Figure 1D). Mature tscA RNA could not be detected in raa1-L137H (Figure 1C, lane 10, black star). In the mutant, several larger transcripts were more prevalent (white stars). These precursor transcripts also contain sequences of chlN, which is downstream of tscA (29). However, normal levels of mature tscA RNA were recovered in the raa1-L137H strains transformed with the Raa1 cosmid or the ‘midigene’ (Figure 1C, lanes 12 and 14). Mature tscA RNA was observed in raa1-314B (lane 9), as expected for a strain competent in splicing of intron 1.
A truncated form of Raa1 is expressed as a fusion to Arg7 in raa1-314B
Our data indicate that the mutants are allelic, but their splicing defects are different: in raa1-L137H there is a block in maturation of the tscA RNA and in splicing of both psaA introns, while in raa1-314B only splicing of intron 2 is defective. This could be explained if Raa1 is partly functional in raa1-314B, and more severely impaired in raa1-L137H. To further investigate the basis for the different phenotypes, we analysed the expression of Raa1 transcripts by RT–PCR (Figure 2).
Figure 2 Expression of Raa1 in the mutants. (A) Schematic representation of the Raa1 mRNA with the fragments amplified in the RT–PCR analysis. (B) RT–PCR analysis with RNA from the wild-type (WT), raa1-314B and raa1-L137H, for three parts of Raa1 (panel A) and the Cbl transcript as a control. The + or – signs indicate whether reverse transcriptase was included (lanes 2–4) or omitted as a control (lane 1). (C) The 3' part of Raa1 is expressed as a fusion to Arg7 in raa1-314B. The structure of the genomic DNA in raa1-314B is shown with the last exons of Arg7 (open bars), a fragment of the cloning vector pBR329 from pARG7.8 (thick black line), the 3' part of Raa1 intron 9 (thin black line) and the exons of Raa1 (shaded bars). The sequence of a short hairpin-loop created at the insertion site by duplication of a 15 bp sequence of Raa1 is indicated above the scheme. A cryptic donor site in exon 13 of Arg7 (black arrow) and the acceptor site of exon 10 of Raa1 (black arrow) are used for splicing across the pBR329 sequence to generate an in-frame fusion transcript as indicated at the bottom.
We designed primers to amplify a 5' region (exons 1 and 2), an intermediate region (exons 12 and 13) and a 3' region (exons 17 and 18) of Raa1 mRNA (Figure 2A). In the wild-type strain the Raa1 transcript could be detected with the three primer pairs (Figure 2B, lane 2). In raa1-L137H (lane 3) the three parts of the transcript were also amplified, indicating the presence of the Raa1 mRNA. However, the full-length Raa1 protein was not detected by immunoblot analysis (Figure 1B, lane 2).
In raa1-314B (Figure 2B, lane 4) the intermediate and 3' regions of the Raa1 transcript could be amplified. However, no amplification of the 5' region (exons 1 and 2) was detected. This qualitative analysis suggested that the 5' part of the gene was altered but that a truncated form of Raa1 might be expressed. To investigate this further, we sequenced a genomic fragment from the raa1-314B mutant (P23, Supplementary Figure 1A) and found that the Arg7 gene and part of the pBR329 vector were joined to intron 9 of the Raa1 gene (Figure 2C). Indeed, the presence of an Arg7–Raa1 fusion transcript was revealed by RT–PCR with a 5' primer in exon 11 of Arg7 and a 3' primer in exon 13 of Raa1. Its sequence showed that the fused mRNA is spliced using a cryptic 5' donor site within the last exon of Arg7 (exon 13) and the 3' acceptor site for exon 10 of Raa1 (Figure 2C). The exons are fused in-frame, such that a polypeptide is predicted with all but the last 14 residues of the Arg7 polypeptide (i.e. arginino-succinate lyase) fused to the 644 C-terminal residues of Raa1. Arg7 is a nucleus-encoded plastid protein (30), so the fusion protein is also predicted to be targeted to the chloroplast. It thus appears that in raa1-314B the C-terminal domain of Raa1 is sufficient for processing of tscA RNA and splicing of intron 1, but not for splicing of intron 2. The predicted fusion protein of 116 kDa was not detected by immunoblotting probably because of low expression or high turnover, or because it was masked by non-specific bands.
Different domains of Raa1 are required for its functions in psaA trans-splicing
Since the C-terminal domain of Raa1 expressed in raa1-314B is sufficient for processing of tscA and for splicing of psaA intron 1, we sought to determine which parts of the protein are required for these functions or for splicing of intron 2. We conducted an initial deletion analysis using genomic fragments with 3' truncations of Raa1 (Figure 3C, construct names beginning with g for genomic). We transformed the two raa1 mutants and selected for phototrophy on minimal medium, an indication of whether psaA splicing is restored. However, in this approach, when phototrophy is not rescued, for instance because only one of the two introns is spliced, transformants are not recovered and the effect of the mutations on the excision of the individual introns cannot be analysed in detail. Therefore, we also constructed deletions of the Raa1 midigene carrying an aphVIII cassette, allowing selection on paromomycin (20) and included the HA epitope tag (Figure 3C, construct names beginning with m for midigene). Many of the transformants which were initially positive in anti-HA immunoblots gradually ceased to express Raa1 when they were sub-cultured on plates for a few weeks, most likely because of gene silencing. Therefore we used aliquots of the same culture for immunoblotting (Figure 3A) and for RNA analysis (Figure 3B). Some lines were particularly prone to silencing: only low levels of HA-tagged Raa1 could be detected in the transformant raa1-L137H mNS #4, and none in raa1-L137H mNS #7 (Figure 3A, lanes 8 and 9). Likewise only one line stably expressing Raa1 could be recovered for mAM in raa1-L137H (mAM #3, lane 10) or for mNS in raa1-314B (mNS #1, lane 14), and none for mBg in raa1-314B.
Figure 3 Deletion analysis of Raa1. (A) Immunoblot analysis with anti-HA antibodies of total cell extracts from transformants of raa1-L137H (lanes 1–10) or raa1-314B (lanes 11–16) with the midigene constructs depicted in panel C. (B) RNA blot hybridization of total RNA from the wild-type (lanes 1 and 12), the parental mutants (lanes 2, 3, 13 and 14) and the transformants of raa1-L137H (lanes 4–11) or raa1-314B (lanes 15–20). Symbols are as in Figure 1. (C) Schematic representation of the Raa1 polypeptide predicted for the wild-type or for the deletion derivatives (open bars, see Supplementary Materials and Methods). Names shown on the left begin with ‘g’ for genomic constructs, with ‘m’ for the midigene tagged with the HA epitope (white triangle). The part of Raa1 that is fused to Arg7 in raa1-314B is also represented as an open bar. Black lines at the bottom represent the domains involved in splicing of intron 1 or intron 2. In the columns marked ‘phototrophy’, growth of the two mutants on minimal medium is indicated by ‘+’, lack of growth by ‘–’. Splicing of intron 1 (i1) or of intron 2 (i2) is indicated by ‘+’, lack of splicing by ‘–’, as deduced from the RNA blot analysis in panel B. (nd: not determined, na: not applicable). The position of the restriction sites used for the deletions are shown at the top (they truncate the predicted polypeptide sequence after the following amino acid residues: AgeI, 85; NruI, 300; MscI, 731; SmaI, 944; SalI, 1136; BglII, 1173; ScaI, 1549; HindIII, 1785 and KpnI, 1853). Black triangles represent the 38 amino acid repeats (not drawn to scale).
In raa1-314B, constructs mKp, gHi and gSc rescued phototrophy (Figure 3C). Indeed, spliced psaA mRNA was detected in the transformants mKp #4 and #7 (Figure 3B, black triangle, lanes 16 and 17). These results define a domain upstream of ScaI which is sufficient for splicing of intron 2. This part of Raa1 and the domain expressed as an Arg7–Raa1 fusion in raa1-314B are thus capable of intragenic complementation. However in raa1-314B, mBg and gSa did not rescue phototrophy (Figure 3C), indicating that a region between the BglII and ScaI sites is essential for splicing of intron 2. In the raa1-L137H mutant, even the shortest 3' deletion, mKp, failed to rescue phototrophy (Figure 3C). In these raa1-L137H mKp transformants, no psaA mRNA was detected (Figure 3B, exon 3, black triangle, lanes 7 and 8), but the intermediate with exon 2 spliced to exon 3 accumulated (exon 3, diamond). This confirms that mKp constructs can still support intron 2 splicing and also implies that the deletions at the C-terminal end of Raa1 affect splicing of intron 1 (Figure 3C). In raa1-L137H, mBg and gSa did not rescue phototrophy (Figure 3C), and in mBg transformants, the pattern was identical to the raa1-L137H mutant (Figure 3B, lanes 5 and 6 versus lane 2). Thus the domain required for intron 2 splicing is affected in this deletion, confirming the analysis of the raa1-314B transformants.
To analyse the N-proximal part of the protein, we constructed internal deletions that leave the putative transit peptide intact. The mAM deletion rescued phototrophy in both mutants (Figure 3C), and restored splicing of both introns to produce mature psaA mRNA (Figure 3B, exon 3, black triangle, lanes 11, 19 and 20). Thus a large domain in the N-terminal part of Raa1 is not essential for splicing of either intron. Transformation of raa1-314B with mNS does not rescue phototrophy or splicing of intron 2: the mature psaA mRNA is not made (Figure 3B, exon 3, black triangle, lane 18) and the exon1–exon2 intermediate is still accumulated (Figure 3B, 5'i2, asterisk, lane 18). Comparison of the mNS and mAM raa1-314B transformants indicate that part of the MscI to SmaI segment is necessary for intron 2 splicing. In the raa1-L137H mNS transformant #4, the intermediate with exon 2 spliced to exon 1 (Figure 3B, 5'i2, asterisk, lane 9 versus lane 2) indicates that splicing of intron 1 is restored, in spite of the low levels of Raa1 protein expressed in this strain (Figure 3A, lane 8). This confirms that the C-terminal domain of Raa1 is sufficient for splicing of intron 1. Only a very faint band is observed for raa1-L137H mNS #7 (Figure 3B, 5'i2, asterisk, lane 10) which does not detectably express Raa1 (Figure 3A, lane 9).
In this analysis, the domain of Raa1 required for splicing of intron 1 is not distinguishable from the domain required for tscA processing: in the mKpn and mBg transformants of raa1-L137H where intron 1 is not spliced, mature tscA fails to accumulate (Figure 3B, tscA, black star, lanes 5–8). Conversely intron 1 is spliced and mature tscA RNA is formed in the raa1-314B mutant, or in the mNS and mAM transformants of raa1-L137H (Figure 3B, tscA, black star, lanes 3, 9 and 11).
Taken together, these results delineate two functional domains of Raa1 (Figure 3C). A central region involved in splicing of intron 2 extends from MscI to somewhere between BglII and ScaI, with a possible additional contribution from the N-terminal region upstream of AgeI. A C-terminal region involved in splicing of intron 1 and tscA maturation starts at the breakpoint of the Arg7–Raa1 fusion in raa1-314B (Figure 3C, last open bar at the bottom) and ends somewhere between KpnI and the C-terminus. The two domains may partly overlap in a short region to the left of the ScaI site.
Raa1, a novel protein with internal repeats
The Predotar and ChloroP algorithms predict that Raa1 is imported into the chloroplast with an N-terminal transit peptide of 54 residues (Figure 4). This localization was confirmed by cell fractionation experiments (see below). A striking feature of the predicted Raa1 polypeptide (Figure 4A) is the occurrence of numerous stretches of repeated amino acids, such as (Ala)9, (Gly)9, (Gln)8 or (Pro)4. Raa1 also contains repetitive sequences like (Pro-Gln)4 or (Ser-Ala-Ala)5. Similar repetitive stretches also occur in other C.reinhardtii proteins, notably in another psaA splicing factor, Raa3 (10) and in other proteins involved in chloroplast gene expression (see Discussion). Raa1 harbours five copies of a 38 amino acid repeat (Figure 4B). The first three repeats are in tandem arrangement (residues 1423–1536), the other two are separate, further on the C-terminal side (1772–1809 and 1826–1863). They are related to the PPPEW repeats that are found in Tbc2, a nucleus-encoded protein of C.reinhardtii involved in translation of the chloroplast psbC mRNA . Four of the 38 amino acid repeats (repeats 2 to 5) are retained in raa1-314B, which can process tscA RNA and splice intron 1. The last repeat is disrupted in the mKpn deletion which can splice intron 2 but not intron 1, and two of the repeats (2 and 3) are in the region where the two functional domains of Raa1 may overlap upstream of the ScaI site. No other significant similarity was found between the deduced Raa1 polypeptide and sequences deposited in public databases.
Figure 4 The predicted Raa1 polypeptide. (A) The putative transit peptide (54 amino acids) is shown in italics and underlined (thin wavy line). Repeats of four or more identical consecutive amino acid residues are shaded, as well as repeats of alternating residues. The 38 amino acid repeats are underlined (thick black line or double line). (B) Alignment of the 38 amino acid repeats. Residues that are conserved in at least three of the five repeats are shaded. Segments predicted to form -helices are underlined (28). The consensus shows residues that are conserved in 3 repeats (uppercase), and 4 or 5 repeats (bold uppercase). The lowercase w represents a position where the aromatic residue W is present twice and Y once. The consensus for the 38 amino acid repeat (PPPEW) of Tbc2 (31) is shown below, with asterisks marking identical residues. The 34 amino acid TPR repeat with its two helical domains (underlined) and the 35 amino acid PPR repeat (presumptive helical domains underlined) are shown for comparison (51).
Raa1 is a membrane-associated chloroplast protein
To investigate the subcellular localization of Raa1, we used a strain expressing the HA-tagged wild-type midigene. In this construct, the genomic fragment, which extends 3.8 kb upstream of the ATG and also includes the 5' part of the gene with the first five introns, should preserve normal expression and regulation. As observed for the native protein revealed with the Raa1 antibody (Figure 1B), the HA-tagged protein also migrates more slowly than expected for its calculated molecular weight of 205 kDa (Figure 3A, lanes 1 and 11). This may be due to the unusual sequence features of the protein and its large size.
In order to determine the subcellular localization of the Raa1 protein, chloroplasts were prepared by Percoll gradient centrifugation. Raa1 was found in the chloroplast fraction, which also contains the other chloroplast proteins PsaA and Rubisco, but only trace levels of the cytosolic translation factor eiF4A (Figure 5A, lane 2). When isolated chloroplasts were further fractionated, Raa1 was enriched in the crude membrane pellet along with the integral thylakoid protein PsaA (lane 4), and depleted from the supernatant which contains the soluble protein Rubisco (lane 3). Raa1 thus fractionates as a chloroplast protein, as expected from the genetic data, although its presence in other organelles, such as mitochondria, is not ruled out. The membrane association of Raa1 was confirmed by sucrose gradient equilibrium centrifugation (data not shown).
Figure 5 Raa1 is part of a large RNase sensitive complex. (A) Subcellular localization of Raa1. Fractions were prepared from a transformant of raa1-314B expressing the HA-tagged Raa1 protein. The four lanes contain total cell extracts (Tot), the chloroplast fraction (Cp) and a chloroplast lysate fractionated by centrifugation into a soluble fraction (Sup) and a crude membrane fraction (Pellet). Immunoblots with antibodies against the HA epitope (Raa1::HA), PsaA (an integral membrane protein of Photosystem I), the small subunit of Rubisco (a soluble stromal protein) and eIF4A (a cytoplasmic translation factor, the nature of the sligthly larger cross-reacting band in the chloroplast fraction is not known). Equal amounts of protein were loaded in each lane (60 μg). (B) Raa1 is part of a large RNase-sensitive complex. A soluble extract was prepared by salt treatment of a total membrane fraction, and subjected to sucrose gradient sedimentation. Ten fractions were collected and analyzed by immunoblotting with a monoclonal anti-HA antibody. The extract was either untreated, mock treated for 20 min at 16°C, or treated with RNase before loading on the gradient. The sedimentation of protein markers of known size is indicated (in kDa) above the lanes.
Raa1 is part of a large RNase sensitive complex
The Raa1 protein can be released from the membrane fraction by treatment with moderate concentrations of salts such as 0.2 M ammonium sulphate (data not shown). An extract was prepared by washing membranes with salt and was fractionated by sedimentation on sucrose gradients (Figure 5B). Part of Raa1 was found in the lower part of the gradient (fractions 2–4), as a large complex which sediments just ahead of the 670 kDa molecular mass marker. Part of Raa1 was also found in a complex that sediments more slowly (fractions 6–8), where it is associated with Raa2 (32).
When the ammonium sulphate extract was incubated with RNase before sucrose gradient sedimentation, the position of the large complex shifted towards the top of the gradient. This shift was not observed when the extract was mock-incubated in the absence of RNase. The sensitivity to RNase shows the presence of RNA in the large Raa1 complex.
DISCUSSION
Raa1 is a multifunctional splicing factor
We have characterized a new mutant, raa1-314B, defective in trans-splicing of only the second intron of psaA (class A), and a mutant that fails to splice both introns, raa1-L137H (class B). The two mutants are clearly allelic since they fail to complement in genetic tests and both can be rescued by the wild-type Raa1 gene. Thus Raa1 has at least two distinct functions, in splicing of intron 1 and of intron 2. In raa1-HN31, which is an allele of raa1-L137H (9), there is no detectable mature tscA (29), the small RNA required for splicing intron 1 (7). The same defect is also observed in raa1-L137H, but in contrast raa1-314B splices intron 1 and accumulates mature tscA RNA. Thus the function of Raa1 in splicing of intron 1 may be direct (in this case Raa1 would have three distinct roles), or may be indirectly mediated by tscA. Two other factors which are required for intron 1 splicing and for processing of tscA were described recently, Rat1 and Rat2 (12). Like for Raa1, it is still unclear whether they are directly or indirectly involved in intron 1 splicing. The pattern of longer tscA transcripts that accumulate in raa1-L137H and raa1-HN31 and in the rat1 and rat2 mutants are similar (12,29). Thus a complex processing pathway requiring at least three factors (Raa1, Rat1 and Rat2) is involved in the maturation of tscA RNA from longer precursors.
The partial defect in raa1-314B can be explained by our observation that in raa1-314B the 3' part of Raa1 is expressed as a fusion to Arg7: this domain is thus sufficient to promote tscA maturation and splicing of intron 1. Our deletion analysis of Raa1 confirms that this 3'-terminal domain is required for tscA maturation and splicing of intron 1, while a more central domain is required for splicing of intron 2. However we did not reveal separate domains required for intron 1 splicing and for tscA RNA maturation. This is the first time that in Chlamydomonas a factor involved in splicing of the two introns of psaA (class B) is cloned and characterized. Interestingly, of the fourteen nuclear loci that are required for maturation of psaA transcripts, only two belong to this class. The other factors described to date, Raa2 (11), Raa3 (10), Rat1 and Rat2 (12) were shown to be required for either one of the two intervening sequences (classes A and C). The presence of the two functional domains in the same factor suggests that Raa1 may coordinate splicing of the two psaA introns. Mutants defective in splicing of either intron 1 or 2 show that the intervening sequences can be excised independently from each other and in either order (6). However, a mechanism that allows coordinate splicing of the two introns may improve the efficiency of maturation of psaA transcripts. Chloroplast factors with dual targets were found in Arabidopsis where PGR3 is required for expression of the petL operon and of an ndh gene (33) and in maize, where CRP1 functions in processing of petD RNA, and in translation of petA (34).
Raa1 is part of a large RNA–protein complex associated with membranes
We have cloned the Raa1 gene, which encodes a large predicted polypeptide of 206 kDa. We have shown that Raa1 is found in the chloroplast fraction, as expected from its predicted transit peptide and from its role in plastid gene expression, although its presence in other organelles cannot be ruled out. Surprisingly it is associated with membranes, and shares this property with Raa2 (formerly Maa2), a protein that is required for trans-splicing of the second intron of psaA (11), and Rat1, a factor required for processing of tscA and splicing of the first intron (12). This membrane association may be relevant because PsaA is a core subunit of photosystem I embedded in the thylakoid membrane. However Raa3, a factor required for splicing of the first intron of psaA, is mostly a stromal protein (10), like proteins involved in splicing of maize chloroplast group II introns, CRS1 and CRS2 (35,36). Chloroplast polyribosomes and some RNA-binding proteins are associated with membranes, and this association may facilitate the assembly of the photosystems and other membrane complexes . Whether the membrane association of Raa1, Raa2 and Rat1 reflects a coupling between RNA splicing, translation and protein assembly into the membrane complexes remains a matter of speculation.
When Raa1 is released from membranes with moderate concentrations of salt, it sediments in large complexes which can be resolved into two peaks in sucrose gradients. Raa1 is in part associated with Raa2, with which it can be co-immunoprecipitated (32). The complex containing Raa1 and Raa2 corresponds to the first peak. The faster sedimenting component of the second peak (700 kDa) is sensitive to treatment with RNase, implying that this Raa1 complex contains RNA. This finding fits with the role of Raa1 in splicing. However the precursor transcripts are broadly distributed in the gradients (data not shown), and we have not been able to conclusively determine whether they are part of the complex. The RNase-resistant fraction still sediments rapidly, indicating that Raa1 is in a multimeric complex, perhaps in association with other proteins. Raa3 is also part of a large ribonucleoprotein complex that contains exon 1 and the tscA RNA (10). It will be interesting to investigate interactions of Raa1 with Rat1 in splicing of intron 1. Although Raa3 is mainly in the soluble fraction, and Raa1 and Rat1 in the membrane fraction, it is possible that they could also interact transiently.
Raa1 is a novel splicing factor with repeats
Because of the limited sequence homology with other known proteins, it is difficult to predict the mode of action of Raa1. As already suggested for other proteins found to be essential for group II intron splicing, we expect that Raa1 may function as an RNA chaperone or a stabilizer of active RNA conformations (8). In maize, the CRS1 protein is specifically required for splicing of a group II intron in atpF, it binds the intron and promotes its folding (38). CRS2, together with CAF1 or CAF2, is required for splicing nine other group II introns (35,39). CRS1, CAF1 and CAF2 contain conserved CRM domains which may be RNA-binding modules (40). In plants, genes for proteins similar to intron-encoded maturases are found in the chloroplast and in the nucleus (41,42). Other proteins which associate in the chloroplast with group II introns have been described, but their role in splicing has not yet been established (43,44).
One of the striking features of the predicted Raa1 protein is the presence of numerous stretches of repeated amino acids. Such repetitive stretches occur occasionally in other proteins, but in Raa1 they are particularly long and numerous, and involve diverse residues (Ala, Gln, Ser, Pro and Arg). Several other proteins that share this feature with Raa1 are also involved in chloroplast gene expression, such as Raa3 which is also a psaA trans-splicing factor (10), Nac2, Mbb1 and Mca1 which are involved in RNA processing and stability (45–47), or Tbc2 which is implicated in translation (31). The functional significance of these repetitive stretches is unknown. They may play a structural role and may occur in parts of the protein where the precise sequence is not important so that it has drifted during evolution. There are five copies of a 38 amino acid repeat in Raa1, related to the 38 amino acid PPPEW repeats of Tbc2 (Figure 4). Four of the repeats are retained in the Arg7–Raa1 fusion of raa1-314B which is sufficient for tscA processing and intron 1 splicing, and the last repeat is interrupted in the mKpn deletion which disrupts these activities (Figure 3C). Two of the repeats are in the region of possible overlap with the domain required for intron 2 splicing, just upstream of the ScaI site.
We speculate that although they have a different sequence, the 38 amino acid repeats in Chlamydomonas might be a counterpart of the repeats in the PPR and TPR proteins, many of which are also implicated in RNA maturation. The PPR proteins form a large family in plants where a majority are predicted to be located in the organelles and many are known to interact with RNA . The TPR-like repeats are found in Mbb1 and Nac2 of C.reinhardtii (45,46) and in HCF107 of Arabidopsis (49). Each TPR repeat unit folds back to form two anti-parallel -helices which are stacked with the helices of the adjacent repeats to generate a superhelical structure (50). The related PPR repeats could form a similar structure (51), and one side of the superhelix could provide a surface for protein–RNA interactions, analogous to the protein–RNA interactions between the Puf repeats in Pumilio and their RNA target (52). Thus the repeat-containing chloroplast factors may have been recruited during evolution, by adaptation of a modular protein fold, to recognize new partners or substrates. It will be of interest to determine whether the repeats in Raa1 actually bind RNA, and specifically the psaA precursor transcripts or the tscA RNA.
In conclusion, in this report we describe a novel factor, Raa1, which is required for trans-splicing of both psaA introns in Chlamydomonas. Distinct domains in Raa1 are responsible for at least two functions, processing of tscA (the small RNA which is required for splicing of psaA intron 1) and splicing of psaA intron 2. Raa1 may thus play a role in coordinating the excision of the two psaA introns. Our analysis of Raa1 provides a basis to further investigate how different events of chloroplast RNA maturation are connected.
SUPPLEMENTARY DATA
Supplementary Data are available at NAR Online.
ACKNOWLEDGEMENTS
The authors wish to thank Nicolas Roggli for preparing the figures, the members of the Rochaix laboratory for helpful discussions, Hans Sommer for providing a cDNA library, Cris Kuhlemeier for a gift of eIF4A antibody and Frédy Barneche, Stéphane Bellafiore, Paula Duek, Angela Falciatore and Frédéric Iseni for helpful comments on the manuscript. L.M. was supported by a long-term HFSP fellowship. This work was supported by a grant of the Swiss National Fund for Scientific Research (3100-0667763.02). Funding to pay the Open Access publication charges for this article was provided by the Swiss National Fund for Scientific Research.
REFERENCES
Qin, P.Z. and Pyle, A.M. (1998) The architectural organization and mechanistic function of group II intron structural elements Curr. Opin. Struct. Biol, . 8, 301–308 .
Bonen, L. and Vogel, J. (2001) The ins and outs of group II introns Trends Genet, . 17, 322–331 .
Toro, N. (2003) Bacteria and Archaea Group II introns: additional mobile genetic elements in the environment Environ. Microbiol, . 5, 143–151 .
Villa, T., Pleiss, J.A., Guthrie, C. (2002) Spliceosomal snRNAs: Mg(2+)-dependent chemistry at the catalytic core? Cell, 109, 149–152 .
Kück, U., Choquet, Y., Schneider, M., Dron, M., Bennoun, P. (1987) Structural and transcription analysis of the two homolgous genes for the P700 chlorophyll a-apoprotein in Chlamydomonas reinhardtii: evidence for in vivo trans-splicing EMBO J, . 6, 2185–2195 .
Choquet, Y., Goldschmidt-Clermont, M., Girard-Bascou, J., Kuck, U., Bennoun, P., Rochaix, J.D. (1988) Mutant phenotypes support a trans-splicing mechanism for the expression of the tripartite psaA gene in the C.reinhardtiichloroplast Cell, 52, 903–913 .
Goldschmidt-Clermont, M., Choquet, Y., Girard-Bascou, J., Michel, F., Schirmer-Rahire, M., Rochaix, J.D. (1991) A small chloroplast RNA may be required for trans-splicing in Chlamydomonas reinhardtii Cell, 65, 135–143 .
Schroeder, R., Grossberger, R., Pichler, A., Waldsich, C. (2002) RNA folding in vivo Curr. Opin. Struct. Biol, . 12, 296–300 .
Goldschmidt-Clermont, M., Girard-Bascou, J., Choquet, Y., Rochaix, J.D. (1990) Trans-splicing mutants of Chlamydomonas reinhardtii Mol. Gen. Genet, . 223, 417–425 .
Rivier, C., Goldschmidt-Clermont, M., Rochaix, J.D. (2001) Identification of an RNA–protein complex involved in chloroplast group II intron trans-splicing in Chlamydomonas reinhardtii EMBO J, . 20, 1765–1773 .
Perron, K., Goldschmidt-Clermont, M., Rochaix, J.D. (1999) A factor related to pseudouridine synthases is required for chloroplast group II intron trans-splicing in Chlamydomonas reinhardtii EMBO J, . 18, 6481–6490 .
Balczun, C., Bunse, A., Hahn, D., Bennoun, P., Nickelsen, J., Kuck, U. (2005) Two adjacent nuclear genes are required for functional complementation of a chloroplast trans-splicing mutant from Chlamydomonas reinhardtii Plant J, . 43, 636–648 .
Harris, E.H. The Chlamydomonas sourcebook. A comprehensive guide to biology and laboratory use, (1989) San Diego, CA Academic Press, Inc .
Debuchy, R., Purton, S., Rochaix, J.D. (1989) The argininosuccinate lyase gene of Chlamydomonas reinhardtii: an important tool for nuclear transformation and for correlating the genetic and molecular maps of the ARG7 locus EMBO J, . 8, 2803–2809 .
Kindle, K.L. (1990) High-frequency nuclear transformation of Chlamydomonas reinhardtii Proc. Natl Acad. Sci. USA, 87, 1228–1232 .
Shimogawara, K., Fujiwara, S., Grossman, A., Usuda, H. (1998) High-efficiency transformation of Chlamydomonas reinhardtii by electroporation Genetics, 148, 1821–1828 .
Sambrook, J., Fritsch, E.F., Maniatis, T. Molecular Cloning. A laboratory manual, (1989) Cold Spring Harbor, NY Cold Spring Harbor Laboratory Press .
Purton, S. and Rochaix, J.D. (1994) Complementation of a Chlamydomonas reinhardtii mutant using a genomic cosmid library Plant Mol. Biol, . 24, 533–537 .
Zhang, H., Herman, P.L., Weeks, D.P. (1994) Gene isolation through genomic complementation using an indexed library of Chlamydomonas reinhardtii DNA Plant Mol. Biol, . 24, 663–672 .
Sizova, I., Fuhrmann, M., Hegemann, P. (2001) A Streptomyces rimosus aphVIII gene coding for a new type phosphotransferase provides stable antibiotic resistance to Chlamydomonas reinhardtii Gene, 277, 221–229 .
Goldschmidt-Clermont, M. and Rahire, M. (1986) Sequence, evolution and differential expression of the two genes encoding variant small subunits of ribulose bisphosphate carboxylase/oxygenase in Chlamydomonas reinhardtii J. Mol. Biol, . 191, 421–432 .
Silflow, C.D., LaVoie, M., Tam, L.W., Tousey, S., Sanders, M., Wu, W., Borodovsky, M., Lefebvre, P.A. (2001) The Vfl1 Protein in Chlamydomonas localizes in a rotationally asymmetric pattern at the distal ends of the basal bodies J. Cell. Biol, . 153, 63–74 .
Merendino, L., Falciatore, A., Rochaix, J.D. (2003) Expression and RNA-binding properties of the chloroplast ribosomal protein S1 from Chlamydomonas reinhardtii Plant Mol. Biol, . 53, 371–382 .
Schloss, J.A. (1990) A Chlamydomonas gene encodes a G protein beta subunit-like polypeptide Mol. Gen. Genet, . 221, 443–452 .
Silflow, C., Rochaix, J.D., Goldschmidt-Clermont, M., Merchant, S. The Molecular Biology of Chloroplasts and Mitochondria in Chlamydomonas, (1998) Dordrecht, The Netherlands Kluwer Academic Publishers pp. 25–40 Vol. 7 .
Emanuelsson, O., Nielsen, H., von Heijne, G. (1999) ChloroP, a neural network-based method for predicting chloroplast transit peptides and their cleavage sites Protein Sci, . 8, 978–984 .
Small, I., Peeters, N., Legeai, F., Lurin, C. (2004) Predotar: a tool for rapidly screening proteomes for N-terminal targeting sequences Proteomics, 4, 1581–1590 .
Garnier, J., Gibrat, J.F., Robson, B. (1996) GOR method for predicting protein secondary structure from amino acid sequence Methods Enzymol, . 266, 540–553 .
Hahn, D., Nickelsen, J., Hackert, A., Kuck, U. (1998) A single nuclear locus is involved in both chloroplast RNA trans-splicing and 3' end processing Plant J, . 15, 575–581 .
Baginsky, S., Siddique, A., Gruissem, W. (2004) Proteome analysis of tobacco bright yellow-2 (BY-2) cell culture plastids as a model for undifferentiated heterotrophic plastids J. Proteome Res, . 3, 1128–1137 .
Auchincloss, A.H., Zerges, W., Perron, K., Girard-Bascou, J., Rochaix, J.D. (2002) Characterization of Tbc2, a nucleus-encoded factor specifically required for translation of the chloroplast psbC mRNA in Chlamydomonas reinhardtii J. Cell. Biol, . 157, 953–962 .
Perron, K., Goldschmidt-Clermont, M., Rochaix, J.D. (2004) A multiprotein complex involved in chloroplast group II intron splicing RNA, 10, 704–710 .
Yamazaki, H., Tasaka, M., Shikanai, T. (2004) PPR motifs of the nucleus-encoded factor, PGR3, function in the selective and distinct steps of chloroplast gene expression in Arabidopsis Plant J, . 38, 152–163 .
Fisk, D.G., Walker, M.B., Barkan, A. (1999) Molecular cloning of the maize gene crp1 reveals similarity between regulators of mitochondrial and chloroplast gene expression EMBO J, . 18, 2621–2630 .
Till, B., Schmitz-Linneweber, C., Williams-Carrier, R., Barkan, A. (2001) CRS1 is a novel group II intron splicing factor that was derived from a domain of ancient origin RNA, 7, 1227–1238 .
Jenkins, B.D. and Barkan, A. (2001) Recruitment of a peptidyl-tRNA hydrolase as a facilitator of group II intron splicing in chloroplasts EMBO J, . 20, 872–879 .
Zerges, W. (2000) Translation in chloroplasts Biochimie, 82, 583–601 .
Ostersetzer, O., Cooke, A.M., Watkins, K.P., Barkan, A. (2005) CRS1: a chloroplast group II intron splicing factor, promotes intron folding through specific interactions with two intron domains Plant Cell, 17, 241–255 .
Ostheimer, G.J., Williams-Carrier, R., Belcher, S., Osborne, E., Gierke, J., Barkan, A. (2003) Group II intron splicing factors derived by diversification of an ancient RNA-binding domain EMBO J, . 22, 3919–3929 .
Ostheimer, G.J., Hadjivasiliou, H., Kloer, D.P., Barkan, A., Matthews, B.W. (2005) Structural analysis of the group II intron splicing factor CRS2 yields insights into its protein and RNA interaction surfaces J. Mol. Biol, . 345, 51–68 .
Mohr, G. and Lambowitz, A.M. (2003) Putative proteins related to group II intron reverse transcriptase/maturases are encoded by nuclear genes in higher plants Nucleic Acids Res, . 31, 647–652 .
Liere, K. and Link, G. (1995) RNA-binding activity of the matK protein encoded by the chloroplast trnK intron from mustard (Sinapis alba L.) Nucleic Acids Res, . 23, 917–921 .
Nakamura, T., Ohta, M., Sugiura, M., Sugita, M. (1999) Chloroplast ribonucleoproteins are associated with both mRNAs and intron-containing precursor tRNAs FEBS Lett, . 460, 437–441 .
Bunse, A.A., Nickelsen, J., Kuck, U. (2001) Intron-specific RNA-binding proteins in the chloroplast of the green alga Chlamydomonas reinhardtii Biochim. Biophys. Acta, . 1519, 46–54 .
Boudreau, E., Nickelsen, J., Lemaire, S.D., Ossenbuhl, F., Rochaix, J.D. (2000) The Nac2 gene of Chlamydomonas encodes a chloroplast TPR-like protein involved in psbD mRNA stability EMBO J, . 19, 3366–3376 .
Vaistij, F.E., Boudreau, E., Lemaire, S.D., Goldschmidt-Clermont, M., Rochaix, J.D. (2000) Characterization of Mbb1, a nucleus-encoded tetratricopeptide-like repeat protein required for expression of the chloroplast psbB/psbT/psbH gene cluster in Chlamydomonas reinhardtii Proc. Natl Acad. Sci. USA, 97, 14813–14818 .
Lown, F.J., Watson, A.T., Purton, S. (2001) Chlamydomonas nuclear mutants that fail to assemble respiratory or photosynthetic electron transfer complexes Biochem. Soc. Trans, . 29, 452–455 .
Lurin, C., Andres, C., Aubourg, S., Bellaoui, M., Bitton, F., Bruyere, C., Caboche, M., Debast, C., Gualberto, J., Hoffmann, B., et al. (2004) Genome-wide analysis of Arabidopsis pentatricopeptide repeat proteins reveals their essential role in organelle biogenesis Plant Cell, 16, 2089–2103 .
Felder, S., Meierhoff, K., Sane, A.P., Meurer, J., Driemel, C., Plucken, H., Klaff, P., Stein, B., Bechtold, N., Westhoff, P. (2001) The nucleus-encoded HCF107 gene of Arabidopsis provides a link between intercistronic RNA processing and the accumulation of translation-competent psbH transcripts in chloroplasts Plant Cell, 13, 2127–2141 .
Das, A.K., Cohen, P.W., Barford, D. (1998) The structure of the tetratricopeptide repeats of protein phosphatase 5: implications for TPR-mediated protein–protein interactions EMBO J, . 17, 1192–1199 .
Small, I.D. and Peeters, N. (2000) The PPR motif: a TPR-related motif prevalent in plant organellar proteins Trends Biochem. Sci, . 25, 46–47 .
Wang, X., McLachlan, J., Zamore, P.D., Hall, T.M. (2002) Modular recognition of RNA by a human pumilio-homology domain Cell, 110, 501–512 .(Livia Merendino, Karl Perron, Michèle Ra)
*To whom correspondence should be addressed. Tel: +41 22 379 6188; Fax: +41 22 379 6868; Email: michel.goldschmidt-clermont@molbio.unige.ch
ABSTRACT
In the chloroplast of Chlamydomonas reinhardtii, psaA mRNA is spliced in trans from three separate precursors which assemble to form two group II introns. A fourth transcript, tscA, completes the structure of the first intron. Of the fourteen nucleus-encoded factors involved in psaA splicing, only two are required for splicing of both introns. We cloned and characterized the first of these more general factors, Raa1. Consistently with its role in psaA splicing, Raa1 is imported in the chloroplast where it is found in a membrane fraction and is part of a large ribonucleoprotein complex. One mutant, raa1-L137H, is defective for splicing of both introns, but another allelic mutant, raa1-314B, still expresses the 3' part of the Raa1 gene and is deficient only in splicing of intron 2. This observation and a deletion analysis indicate the presence of two domains in Raa1. The C-terminal domain is necessary and sufficient for processing of tscA RNA and splicing of the first intron, while the central domain is essential for splicing of the second intron. The combination of these two functional domains in Raa1 suggests that this new factor may coordinate trans-splicing of the two introns to improve the efficiency of psaA maturation.
INTRODUCTION
Group II introns are typically characterized by small elements of conserved primary sequence and a common secondary structure involving a central core and six radiating domains with helical stems. They also share many conserved tertiary structure interactions among the domains, as well as between the domains and the exons . Group II introns are prevalent in the organelle genomes of chloroplasts and mitochondria, and are also found in bacteria and archea . These introns are thought to be evolutionarily related to the introns in nuclear pre-mRNAs, where catalysis is mediated by the ribonucleoprotein complexes of the spliceosome, and proceeds by a similar mechanism involving two trans-esterification reactions and a lariat intermediate (4).
In the green alga Chlamydomonas reinhardtii, the psaA gene is composed of three exons which are at distant positions in the chloroplast genome and are transcribed separately (5,6). The precursor transcripts are spliced in trans to generate the mature psaA mRNA (Figure 1D). Flanking the exons, the sequences which constitute the split introns have characteristic features of group II introns. The second intron is assembled from the precursors of exon 2 and of exon 3. However the first intron is actually composed of three transcripts: the precursors of exon 1 and of exon 2, and a small non-coding RNA, transcribed from a separate locus, tscA (7). This intron in three pieces can be seen as an intermediate between the usual introns of group II, containing their own catalytic sequences, and the introns of nuclear pre-mRNAs, where part of the structure and catalytic activity have been proposed to reside in the trans-acting RNAs of the snRNPs.
Figure 1 Characterization of raa1-314B and raa1-L137H. (A) Immunoblot analysis of total cell extracts from the wild-type (WT), the mutants raa1-314B and raa1-L137H, and the mutants rescued with the genomic DNA (cosmid #2), or the midigene construct. Top, PsaA; bottom, small subunit of Rubisco. (B) Immunoblot analysis with anti-Raa1 antibodies (affinity purified) of total cell extracts from the WT and the mutants raa1-L137H and raa1-314B. The arrow labels a non-specific band. (C) RNA blot analysis with probes (indicated on the left) specific for psaA exons 1, 2, 3, the 5' part of intron 2 (5'i2), tscA RNA or psbA transcript as a control. Symbols correspond to the transcripts on the splicing scheme (panel D), except the white square which labels a non-specific hybridization signal. (D) Trans-splicing of psaA mRNA. The precursors of exon1 (black square, 0.4 kb), exon 2 (black circle, 3.6 kb) and exon 3 (open circle, 2.4 kb) are spliced in trans to produce intermediates (exon1–exon2, asterisk, 2.6 kb or exon2-exon3, diamond, 3.8 kb) and mature psaA mRNA (black triangle, 2.8 kb). Splicing of the first split intron requires tscA RNA (black star, 0.4 kb) which is processed from longer precursors (open stars) containing chlN.
While some group II introns are well-studied ribozymes that can self-splice in vitro, others have been refractory to self-splicing in the test tube. Both types require accessory protein factors for efficient splicing in vivo. Because the catalytic activity is carried by the RNA, these factors may function as RNA chaperones, may assist proper folding of the intron, or may stabilize it in the active conformation (8).
In Chlamydomonas, a set of at least fourteen nuclear loci are essential for trans-splicing the two split group II introns of psaA (9). Three of these factors have been characterized more in detail, Raa3, Raa2 and Rat1. Raa3 is necessary for trans-splicing of the first intron . This novel protein is part of a large complex which also contains the precursor of exon 1 and the tscA RNA. Raa2 is essential for trans-splicing of the second intron . Although Raa2 is related to pseudouridine synthases, this enzymatic activity is not required for its function in trans-splicing. Rat1 is required for processing of tscA RNA from a larger precursor and for splicing of the first intron (12). Rat1 shows sequence similarity to the NAD+ binding domain of poly(ADP-ribose) polymerase, but this domain can be mutated without affecting the function of Rat1 in splicing. These factors are highly specific since they are each required for splicing of only one of the introns. However, two nuclear loci encode factors that are required for splicing of both introns, and may thus have a more general role in splicing of group II introns . Here we report the characterization of the first of these factors, Raa1, which is required for trans-splicing of the two introns of psaA. We map two distinct domains of Raa1, one involved in processing of tscA and splicing of intron 1, the other in splicing of intron 2. Our results show that Raa1 is a multifunctional protein required, as a component of a large membrane-bound and RNA-containing complex, for the splicing of these group II introns.
MATERIALS AND METHODS
Strains and media
Procedures for growing C.reinhardtii and media were described (13) (TAP: Tris Acetate Phosphate; HSM: High Salt Minimal). The strain arg2 cwd also has a plastome mutation conferring spectinomycin resistance and was obtained from Dr R. Loppes (University of Liège, Belgium). Mutant L137H was obtained by mutagenesis with 5-fluoro-deoxy-uridine and UV, as described previously (6,9).
Oligonucleotides
See Supplementary Table I.
Generation of the 314B mutant strain
arg2 cwd cells were transformed with plasmid pARG7.8 (14) and selected on TAP medium without arginine in the dark. Screening of ca. 4 x 103 colonies yielded six strains with the fluorescence induction kinetics of mutants deficient in photosystem I or the b6f complex and a psaA RNA profile characteristic of trans-splicing mutants (9). Mutant 314B was chosen for further study because analysis of fourteen progenies from a back-cross to the wild-type showed co-segregation of the pARG7.8 insertion with the trans-splicing mutant phenotype, suggesting that the mutation was tagged.
Transformation of Chlamydomonas
For transformation of raa1-314B, 3 x 107 cells were treated with 1 μg of the appropriate plasmid DNA by the glass bead method (15) and spread on a single agar plate (HSM or TAP plus 10 μg/ml paromomycin).
The raa1-L137H strain was transformed by electroporation . A total of 108 cells of the mutant were transformed with 1 μg of the appropriate plasmid DNA in presence of 50 μg of salmon sperm DNA and spread on a single plate as above. The electric pulse was applied in the absence of a shunt resistor with 700 V and 25 μF.
RNA blot hybridization
RNA blot analysis of psaA transcripts and the hybridization probes were described previously (11). The tscA probe was a 720 bp PCR fragment (7).
Cloning of the Raa1 gene
For cloning of Raa1, a genomic library of BamHI-digested DNA fragments from raa1-314B was constructed in bacteriophage lambda EMBL4 (17). The library was screened by hybridization with radiolabelled pBR329, the plasmid vector of pARG7.8 (14). Restriction mapping of one of the positive phage (P23) and DNA blot hybridizations with the pARG7.8 probe, or with total genomic DNA from C.reinhardtii as a probe, allowed the identification of a 1.8 kb HindIII fragment that contained neither parts of the pARG7.8 plasmid, nor repetitive sequences. This fragment was used to screen a cosmid library of wild-type C.reinhardtii DNA (18). Three positive cosmids were obtained and used for transformation of the raa1-314B mutant and selection on minimal medium (HSM). All three were capable of rescuing phototrophic growth. Cosmid #2 was used for restriction mapping and subcloning. A minimal region was identified spanning the ClaI to ScaI sites. Probes from this region were used to screen cDNA libraries (see below).
Raa1 was also independently identified by transformation of L137H with an ordered cosmid library (19) and selection on HSM in the light. One cosmid was found to restore the wild-type phenotype. This cosmid (30 kb insert) was used for restriction mapping and subcloning, and was found to contain the Raa1 gene.
Complementation analysis in vegetative diploids
Vegetative diploids were obtained by fusion of either raa1-L137H or raa2-A18 with raa1-314B. To allow selection of diploids, paromomycin-resistant derivatives of raa1-L137H and raa2-A18 strains were obtained by transformation with the aphVIII gene . The raa1-314B strain has a chloroplast mutation conferring resistance to spectinomycin.
Removal of the cell wall was achieved by incubating loopfuls of each strain with autolysin in dim light for 1 h at 25°C. Vegetative fusion was induced by spreading the cells on selective plates in fusion solution or a control solution . Diploids were selected in dim light on TAP plates containing paromomycin (10 μg/ml) and spectinomycin (120 μg/ml).
The genotype of raa1-L137H/raa1-314B diploids was verified by PCR amplification of the aphVIII gene (characteristic of raa1-L137H) with oligos paro2-5' and paro2-3' and of the pARG 7.8 plasmid (characteristic of raa1-314B) with oligos pBRBam2 and ArgSau2. The genotype of raa2-A18/raa1-314B diploids was verified by amplification of aphVIII and of Raa2 with oligos A18/rbcS and A18/Nci1.
Immunoblot analysis
Protein samples were analyzed on 5% polyacrylamide gels reinforced with 0.15% linear poly(acrylamide-co-acrylic acid) (Aldrich). Proteins were transferred to nitrocellulose for 16 h at 25 V. Raa1 rabbit antibodies were obtained by immunization with a mixture of three histidine-tagged polypeptides corresponding to residues 1009–1349, 1510–1829 and 1830–2103 of Raa1. They were affinity purified using the three antigens immobilized on Ni NTA. The monoclonal HA-11 antibody was obtained from Eurogentec.
Raa1 genomic and midigene constructs
The subclone gHi (116b) was obtained by inserting the 12 kb HindIII fragment from cosmid #2 into Bluescript KS (+) digested with HindIII. Subclone gSc was obtained by inserting the 11 kb EcoRI-ScaI fragment from 116b into Bluescript KS (+) digested with EcoRI and SmaI. Subclone gSa was derived from 116b by partial digestion with SalI and religation.
The cDNA clone C11 (Supplementary Figure 1) was isolated by screening a library constructed in bacteriophage lambda gt10 (21) with the 0.8 kb SalI–BstXI genomic fragment containing exons 11 and 12. The cDNA clone 405 (Supplementary Figure 1) was found by screening a library constructed in lambda NM1149 (H. Sommer, personal communication) with the 1.5 kb SalI–SphI fragment from C11 (exons 6–14). The cDNA clone 551 (Supplementary Figure 1) was obtained from a dedicated cDNA library. This library was prepared by priming synthesis of the first strand of cDNA with oligos i2 and m3 and cloning the double-stranded cDNA with EcoRI-NotI adaptors (Amersham-Pharmacia) into lambda gt10. This library was screened with a 0.68 kb EcoRI–BglII from 405 (exons 4–6). All cDNA inserts were excised with EcoRI and cloned into Bluescript KS (+). The open reading frame (ORF) that the cDNA clones defined encodes a predicted polypeptide of 2103 amino acid. It is preceded by stop codons in every frame, implying that the entire coding sequence has been obtained.
The HA-tagged midigene, mWT (also called pI2), was constructed as follows. Subclone B04 was derived by cloning the 9 kb BscI (ClaI) fragment from gHi into Bluescript KS (+) digested with BscI. A 3.4 kb SalI fragment from cDNA C11 was cloned into SalI-digested B04, to yield plasmid 269 (the SalI site used for this fusion is in exon 6). The C11 cDNA (recloned in pET15b, Novagen) was tagged with a His6-Myc-BstEII-Kanr-BstEII sequence introduced in the SmaI site in exon 6 (5'-ctcatgcgcggctcccaccaccaccaccaccacctcggcgagcagaagctgatctccgaggaggacctgggtaacc-Kanr-gtaacctggtc-3', the Kanr cassette was derived from pUC4K, Amersham-Pharmacia). This tagged cDNA sequence was then transferred as a SalI–StuI fragment into plasmid 269, to yield plasmid 349. The triple HA epitope fragment (22) was then inserted as a NruI–NaeI fragment into 349 digested with BstEII to replace the Kanr cassette and obtain plasmid 490. Finally the 3.5 kb SacI fragment of 490 (one SacI site in the polylinker, the other in exon 3) was replaced with the 7 kb fragment from 116b to provide the HA-tagged midigene pI2.
The aphVIII gene, conferring paromomycin resistance (20) was amplified by PCR from plasmid pSL17 (S. Lemaire and J. D. Rochaix, unpublished data). The construct mBg was derived from pI2 by digestion with BglII (at the intron 6 and exon 7 junction of Raa1) and KpnI (in the polylinker) and ligation to the aphVIII gene (amplified with Paro Bgl 5'and Paro 3' and digested with BglII and KpnI). The construct mKp was derived from pI2 by digestion with KpnI (in exon 16 of Raa1 and in the poly linker) and ligation to aphVIII (amplified with Paro Kpn 5' and Paro 3' and digested with KpnI). The construct mAM was derived from pI2 by digestion with AgeI, blunting with mung bean nuclease, digestion with MscI (in exon 2 of Raa1) and religation. In a second step, the aphVIII gene was amplified with Paro Mlu 5' and Paro Mlu 3', digested with MluI and inserted in the MluI site, upstream of Raa1. The clone pI4 was similarly derived from pI2 by inserting the aphVIII gene in the MluI site. The construct mNS was derived from pI4 by digestion with NruI and SmaI (in exons 2 and 4 of Raa1, respectively) and religation.
RT–PCR analysis of Raa1 transcripts
Total RNAs were isolated as described (23) and then treated with RQ1 DNase (Promega). Reverse transcription was performed at 50°C with M-MLV reverse transcriptase (Promega). To analyse three different regions (exons 1 and 2, exons 12 and 13 and exons 17 and 18) of the Raa1 transcript, the corresponding first strand cDNA was synthesized with oligos AUG 2, middle 4 or oligo dT, respectively. PCR amplification was performed with the following pairs of oligos: AUG5'-1 and AUG3'-1; m5'-3 and m3'-3; STOP5'-6 and STOP3'-6, respectively. The PCR products span two exons so that amplification of the cDNA can be distinguished from amplification of any contaminating genomic DNA which would include the intervening intron. As control, RT–PCR analysis was also performed for Cbl transcripts (24), using oligo dT to synthesize the first strand cDNA and oligos UP and DOWN for the PCR. The Arg7–Raa1 fusion transcript in raa1-314B was revealed by synthesizing the first strand cDNA with oligo middle 4 (in exon 14 of Raa1) and PCR amplification with oligos Arg7ex11 and m3'-3 (in exon 13 of Raa1).
Sequence analysis of Raa1
The sequence of the Raa1 cDNA is deposited in the EMBL database under accession no. AJ605114 . The genomic sequence is now available (scaffold 33) in the draft of the C.reinhardtii genome, at the Joint Genome Institute of the US Department of Energy (http://genome.jgi-psf.org/chlre2/chlre2.home.html). Comparison of the cDNA and genomic sequences shows that the Raa1 gene is composed of 18 exons with very different sizes (from 60 to 2473 bp). The polyA tail is preceded 11 bp upstream by the TGTAA consensus signal typical of C.reinhardtii (25).
Transit peptide sequences were predicted using the algorithms ChloroP (26) (http://www.cbs.dtu.dk/services/ChloroP/) and Predotar (27) (version 0.5; http://www.inra.fr/predotar/). Protein secondary structure was predicted with GORIV (28) (http://npsa-pbil.ibcp.fr/cgi-bin/npsa_automat.pl?page=npsa_gor4.html)
Cell fractionation
Cells were lyzed with a nebulizer and chloroplasts were purified on Percoll gradients as described previously (10) in the presence of protease inhibitor mix A (PIM A: 1 mM -caproic acid, 0.5 mM benzamidine, 0.6 μg/ml leupeptin, 2 μM E64, 2 μM pepstatin and 0.5 mM AEBSF). Chloroplasts were resuspended in hypotonic buffer supplemented with protease inhibitor mix B (PIM B: 1 mM -caproic acid, 0.5 mM benzamidine, 0.6 μg/ml leupeptin, 2 mM ortho-phenantroline and 1/200 Protease Inhibitor Cocktail P8849 from Sigma). For further fractionation, the chloroplasts (250 μl, 7 mg protein/ml) were lyzed by sonication, the solution was adjusted to 20 mM HEPES (pH 7.8), 50 mM KCl and 10 mM MgCl2 (HepKM) and subjected to centrifugation at 100 000 g for 30 min at 4°C in a TLA45 rotor (Beckman). The supernatant was recovered for analysis, and the pellet was washed by resuspension in HepKMS (0.2 M sucrose, HepKM, PIM B) and centrifugation at 50 000 g for 15 min at 4°C in a TLA45 rotor. The supernatant was discarded and the pellet was resuspended in HepKM plus PIM B.
For analysis of the protein complexes, cells were collected by centrifugation, resuspended at a density of 2.5 108/ml in HepKM plus PIM A, and lyzed by freezing and thawing. The lysate (1 ml aliquot) was subjected to centrifugation at 18 000 g for 20 min at 4°C in a TLA45 rotor (Beckman). The pellet containing the membranes was resuspended in HepKM, 0.2 M (NH4)2SO4, PIM B. Where needed, 100 μl of RNaseA were added [10 mg/ml in 100 mM Tris–HCl (pH 7.5) and 10 mM Na acetate) or the same amount of buffer without RNase (mock treatment), followed by incubation at 16°C for 20 min. Membranes and insoluble material were removed by two rounds of centrifugation at 120 000 g for 20 min in a TLA45 rotor, and the supernatant was loaded on sucrose gradients (10 ml; 0.1–1.3 M sucrose, HepKM, PIM A). After centrifugation at 115 000 g for 17 h in the SW40 rotor (Beckman), 10 fractions (1.1 ml) were collected from the bottom by puncturing the tube. Sedimentation was calibrated with the High Molecular Weight Calibration Kit from Amersham-Pharmacia.
RESULTS
Characterization of the raa1-314B mutant deficient in splicing of psaA intron 2 and cloning of the Raa1 gene
Insertional mutants were generated by transformation of a Chlamydomonas arginine auxotroph with a plasmid containing the wild-type Arg7 gene. Mutant raa1-314B was unable to grow phototrophically (data not shown) and had no detectable PsaA protein (Figure 1A, lane 2). RNA blot hybridization (Figure 1C) showed that the pattern of transcripts that accumulate in raa1-314B is characteristic of a block in trans-splicing of the second intron of psaA . Probes specific for each of the three exons hybridize to mature psaA mRNA (black triangle, see Figure 1D) in the wild-type (Figure 1C, lane 1) but not in the mutant (lane 2). In the raa1-314B mutant, the probe for exon 3 only detects an unspliced precursor of exon 3 (open circle), which is also present in the wild-type. In the mutant, probes for exon 1 or for exon 2 reveal a splicing intermediate (asterisk) with exons 1 and 2 spliced together, which migrates slightly faster than mature psaA mRNA. This intermediate, with the 5' part of intron 2 still attached to exon 2 (5'i2 probe, lane 9), is not detected in the wild-type (lane 8). The precursor of exon 1 (black square) over accumulates in raa1-314B, as was observed previously for all other psaA splicing mutants (6). Because of these defects in splicing of psaA transcripts, we have called the mutant raa1-314B (RNA maturation of psaA).
Using the Arg7 insertion as a molecular tag, we cloned the raa1 mutant locus. We then obtained wild-type RAA1 cosmids that could rescue the raa1-314B mutant by transformation. These results also showed that the raa1-314B mutation is recessive to the wild-type. Additionally, Southern blot analysis of raa1-314B DNA indicated that the locus was rearranged (data not shown), confirming that the locus that we cloned was indeed RAA1. In the strains transformed with cosmid #2, PsaA protein is recovered (Figure 1A, 314B genomic, lane 3) and the fully spliced mRNA accumulates (Figure 1C, lane 4, black triangle). The splicing intermediate containing the 5' part of intron 2 is not detected any more (Figure 1C, lane 11, asterisk) and the levels of exon 1 precursor are much lower than in the parental mutant (Figure 1C, lanes 2 and 4, black square).
To map the gene within the 34 kb insert in cosmid #2, we tested subclones for their ability to restore phototrophic growth of raa1-314B by transformation and identified a minimal region of 7.4 kb. We obtained cDNA clones from this region which encode a predicted polypeptide of 2103 amino acid (See Materials and Methods). To ascertain that the appropriate cDNA had been cloned, a ‘midigene’ was constructed with the genomic sequence containing the 5' part of the gene (including the promoter) fused to the cDNA from the 3' part (Supplementary Figure 1D). This construct transformed the raa1-314B mutant to prototrophic growth as efficiently as cosmid #2 (data not shown) and restored normal levels of PsaA protein (Figure 1A, lane 4). The midigene also rescued trans-splicing of the psaA mRNA (black triangle, Figure 1C, lane 6), even if splicing intermediates are still detected with probes for exon 2 (black circle), exon 3 (diamond), exon 1 (black square) and 5'i2 (black circle, lane 13).
L137H, defective in splicing of both psaA introns, is an allele of Raa1
In parallel, we have characterized another mutant strain (L137H), that was previously shown to be affected in trans-splicing of both the psaA introns (class B) and was obtained by UV mutagenesis (6). L137H was unable to grow phototrophically (data not shown) and did not accumulate detectable levels of the PsaA protein (Figure 1A, lane 6, the mutant is labelled raa1-L137H in anticipation of evidence presented below which shows that L137H is allelic to raa1-314B). The pattern of transcripts in this mutant (Figure 1C) is characteristic of a block in trans-splicing of both introns (6). Probes specific for each of the three exons fail to detect mature psaA mRNA in the mutant (black triangle, lane 3), but revealed unspliced precursors of exon 1 (black square), exon 2 (black circle) and exon 3 (white circle). The exon 2 precursor, with the 5' part of intron 2 still attached to exon 2, was also detected with the 5'i2 probe (black circle, lanes 3 and 10). As noted above for raa1-314B and other trans-splicing mutants, the precursor of exon 1 over accumulates.
To clone the mutant locus, we transformed the L137H strain with an indexed genomic library (19) and screened the cosmids for their ability to restore phototrophic growth of the mutant cells. Surprisingly, we identified Raa1 as the nuclear gene that complements the photosynthetic defect of the L137H strain, suggesting that L137H and raa1-314B are allelic. These results also imply that the L137H mutation is recessive to the wild-type. However, Southern blot analysis did not show any alteration in the RAA1 locus of the raa1-L137H mutant (data not shown). In the L137H cells transformed with either the cosmid containing the Raa1 locus or the midigene, synthesis of the PsaA protein is recovered (Figure 1A, lanes 7 and 8) and fully spliced mRNA accumulates (black triangle, Figure 1C, lanes 5 and 7).
To confirm that raa1-314B and L137H are indeed allelic, we constructed vegetative diploids to test whether the two mutants could genetically complement (Supplementary Data). The observed lack of complementation, together with the fact that transformation with the Raa1 gene can rescue psaA splicing in both mutants, indicate that raa1-314B and L137H are allelic. Furthermore full-length Raa1 protein was not detectable by immunoblotting in the two mutant strains, raa1-314B and L137H (Figure 1B, lanes 2 and 3). We thus refer to the two mutants as raa1-314B and raa1-L137H.
The raa1-L137H strain is also defective in maturation of tscA RNA
HN31, another mutant defective in trans-splicing of both introns of psaA (class B), is allelic to L137H (9). Interestingly, HN31 was previously shown to be defective in the maturation of tscA (29), the small non-coding RNA required for splicing of intron 1 (Figure 1D). Mature tscA RNA could not be detected in raa1-L137H (Figure 1C, lane 10, black star). In the mutant, several larger transcripts were more prevalent (white stars). These precursor transcripts also contain sequences of chlN, which is downstream of tscA (29). However, normal levels of mature tscA RNA were recovered in the raa1-L137H strains transformed with the Raa1 cosmid or the ‘midigene’ (Figure 1C, lanes 12 and 14). Mature tscA RNA was observed in raa1-314B (lane 9), as expected for a strain competent in splicing of intron 1.
A truncated form of Raa1 is expressed as a fusion to Arg7 in raa1-314B
Our data indicate that the mutants are allelic, but their splicing defects are different: in raa1-L137H there is a block in maturation of the tscA RNA and in splicing of both psaA introns, while in raa1-314B only splicing of intron 2 is defective. This could be explained if Raa1 is partly functional in raa1-314B, and more severely impaired in raa1-L137H. To further investigate the basis for the different phenotypes, we analysed the expression of Raa1 transcripts by RT–PCR (Figure 2).
Figure 2 Expression of Raa1 in the mutants. (A) Schematic representation of the Raa1 mRNA with the fragments amplified in the RT–PCR analysis. (B) RT–PCR analysis with RNA from the wild-type (WT), raa1-314B and raa1-L137H, for three parts of Raa1 (panel A) and the Cbl transcript as a control. The + or – signs indicate whether reverse transcriptase was included (lanes 2–4) or omitted as a control (lane 1). (C) The 3' part of Raa1 is expressed as a fusion to Arg7 in raa1-314B. The structure of the genomic DNA in raa1-314B is shown with the last exons of Arg7 (open bars), a fragment of the cloning vector pBR329 from pARG7.8 (thick black line), the 3' part of Raa1 intron 9 (thin black line) and the exons of Raa1 (shaded bars). The sequence of a short hairpin-loop created at the insertion site by duplication of a 15 bp sequence of Raa1 is indicated above the scheme. A cryptic donor site in exon 13 of Arg7 (black arrow) and the acceptor site of exon 10 of Raa1 (black arrow) are used for splicing across the pBR329 sequence to generate an in-frame fusion transcript as indicated at the bottom.
We designed primers to amplify a 5' region (exons 1 and 2), an intermediate region (exons 12 and 13) and a 3' region (exons 17 and 18) of Raa1 mRNA (Figure 2A). In the wild-type strain the Raa1 transcript could be detected with the three primer pairs (Figure 2B, lane 2). In raa1-L137H (lane 3) the three parts of the transcript were also amplified, indicating the presence of the Raa1 mRNA. However, the full-length Raa1 protein was not detected by immunoblot analysis (Figure 1B, lane 2).
In raa1-314B (Figure 2B, lane 4) the intermediate and 3' regions of the Raa1 transcript could be amplified. However, no amplification of the 5' region (exons 1 and 2) was detected. This qualitative analysis suggested that the 5' part of the gene was altered but that a truncated form of Raa1 might be expressed. To investigate this further, we sequenced a genomic fragment from the raa1-314B mutant (P23, Supplementary Figure 1A) and found that the Arg7 gene and part of the pBR329 vector were joined to intron 9 of the Raa1 gene (Figure 2C). Indeed, the presence of an Arg7–Raa1 fusion transcript was revealed by RT–PCR with a 5' primer in exon 11 of Arg7 and a 3' primer in exon 13 of Raa1. Its sequence showed that the fused mRNA is spliced using a cryptic 5' donor site within the last exon of Arg7 (exon 13) and the 3' acceptor site for exon 10 of Raa1 (Figure 2C). The exons are fused in-frame, such that a polypeptide is predicted with all but the last 14 residues of the Arg7 polypeptide (i.e. arginino-succinate lyase) fused to the 644 C-terminal residues of Raa1. Arg7 is a nucleus-encoded plastid protein (30), so the fusion protein is also predicted to be targeted to the chloroplast. It thus appears that in raa1-314B the C-terminal domain of Raa1 is sufficient for processing of tscA RNA and splicing of intron 1, but not for splicing of intron 2. The predicted fusion protein of 116 kDa was not detected by immunoblotting probably because of low expression or high turnover, or because it was masked by non-specific bands.
Different domains of Raa1 are required for its functions in psaA trans-splicing
Since the C-terminal domain of Raa1 expressed in raa1-314B is sufficient for processing of tscA and for splicing of psaA intron 1, we sought to determine which parts of the protein are required for these functions or for splicing of intron 2. We conducted an initial deletion analysis using genomic fragments with 3' truncations of Raa1 (Figure 3C, construct names beginning with g for genomic). We transformed the two raa1 mutants and selected for phototrophy on minimal medium, an indication of whether psaA splicing is restored. However, in this approach, when phototrophy is not rescued, for instance because only one of the two introns is spliced, transformants are not recovered and the effect of the mutations on the excision of the individual introns cannot be analysed in detail. Therefore, we also constructed deletions of the Raa1 midigene carrying an aphVIII cassette, allowing selection on paromomycin (20) and included the HA epitope tag (Figure 3C, construct names beginning with m for midigene). Many of the transformants which were initially positive in anti-HA immunoblots gradually ceased to express Raa1 when they were sub-cultured on plates for a few weeks, most likely because of gene silencing. Therefore we used aliquots of the same culture for immunoblotting (Figure 3A) and for RNA analysis (Figure 3B). Some lines were particularly prone to silencing: only low levels of HA-tagged Raa1 could be detected in the transformant raa1-L137H mNS #4, and none in raa1-L137H mNS #7 (Figure 3A, lanes 8 and 9). Likewise only one line stably expressing Raa1 could be recovered for mAM in raa1-L137H (mAM #3, lane 10) or for mNS in raa1-314B (mNS #1, lane 14), and none for mBg in raa1-314B.
Figure 3 Deletion analysis of Raa1. (A) Immunoblot analysis with anti-HA antibodies of total cell extracts from transformants of raa1-L137H (lanes 1–10) or raa1-314B (lanes 11–16) with the midigene constructs depicted in panel C. (B) RNA blot hybridization of total RNA from the wild-type (lanes 1 and 12), the parental mutants (lanes 2, 3, 13 and 14) and the transformants of raa1-L137H (lanes 4–11) or raa1-314B (lanes 15–20). Symbols are as in Figure 1. (C) Schematic representation of the Raa1 polypeptide predicted for the wild-type or for the deletion derivatives (open bars, see Supplementary Materials and Methods). Names shown on the left begin with ‘g’ for genomic constructs, with ‘m’ for the midigene tagged with the HA epitope (white triangle). The part of Raa1 that is fused to Arg7 in raa1-314B is also represented as an open bar. Black lines at the bottom represent the domains involved in splicing of intron 1 or intron 2. In the columns marked ‘phototrophy’, growth of the two mutants on minimal medium is indicated by ‘+’, lack of growth by ‘–’. Splicing of intron 1 (i1) or of intron 2 (i2) is indicated by ‘+’, lack of splicing by ‘–’, as deduced from the RNA blot analysis in panel B. (nd: not determined, na: not applicable). The position of the restriction sites used for the deletions are shown at the top (they truncate the predicted polypeptide sequence after the following amino acid residues: AgeI, 85; NruI, 300; MscI, 731; SmaI, 944; SalI, 1136; BglII, 1173; ScaI, 1549; HindIII, 1785 and KpnI, 1853). Black triangles represent the 38 amino acid repeats (not drawn to scale).
In raa1-314B, constructs mKp, gHi and gSc rescued phototrophy (Figure 3C). Indeed, spliced psaA mRNA was detected in the transformants mKp #4 and #7 (Figure 3B, black triangle, lanes 16 and 17). These results define a domain upstream of ScaI which is sufficient for splicing of intron 2. This part of Raa1 and the domain expressed as an Arg7–Raa1 fusion in raa1-314B are thus capable of intragenic complementation. However in raa1-314B, mBg and gSa did not rescue phototrophy (Figure 3C), indicating that a region between the BglII and ScaI sites is essential for splicing of intron 2. In the raa1-L137H mutant, even the shortest 3' deletion, mKp, failed to rescue phototrophy (Figure 3C). In these raa1-L137H mKp transformants, no psaA mRNA was detected (Figure 3B, exon 3, black triangle, lanes 7 and 8), but the intermediate with exon 2 spliced to exon 3 accumulated (exon 3, diamond). This confirms that mKp constructs can still support intron 2 splicing and also implies that the deletions at the C-terminal end of Raa1 affect splicing of intron 1 (Figure 3C). In raa1-L137H, mBg and gSa did not rescue phototrophy (Figure 3C), and in mBg transformants, the pattern was identical to the raa1-L137H mutant (Figure 3B, lanes 5 and 6 versus lane 2). Thus the domain required for intron 2 splicing is affected in this deletion, confirming the analysis of the raa1-314B transformants.
To analyse the N-proximal part of the protein, we constructed internal deletions that leave the putative transit peptide intact. The mAM deletion rescued phototrophy in both mutants (Figure 3C), and restored splicing of both introns to produce mature psaA mRNA (Figure 3B, exon 3, black triangle, lanes 11, 19 and 20). Thus a large domain in the N-terminal part of Raa1 is not essential for splicing of either intron. Transformation of raa1-314B with mNS does not rescue phototrophy or splicing of intron 2: the mature psaA mRNA is not made (Figure 3B, exon 3, black triangle, lane 18) and the exon1–exon2 intermediate is still accumulated (Figure 3B, 5'i2, asterisk, lane 18). Comparison of the mNS and mAM raa1-314B transformants indicate that part of the MscI to SmaI segment is necessary for intron 2 splicing. In the raa1-L137H mNS transformant #4, the intermediate with exon 2 spliced to exon 1 (Figure 3B, 5'i2, asterisk, lane 9 versus lane 2) indicates that splicing of intron 1 is restored, in spite of the low levels of Raa1 protein expressed in this strain (Figure 3A, lane 8). This confirms that the C-terminal domain of Raa1 is sufficient for splicing of intron 1. Only a very faint band is observed for raa1-L137H mNS #7 (Figure 3B, 5'i2, asterisk, lane 10) which does not detectably express Raa1 (Figure 3A, lane 9).
In this analysis, the domain of Raa1 required for splicing of intron 1 is not distinguishable from the domain required for tscA processing: in the mKpn and mBg transformants of raa1-L137H where intron 1 is not spliced, mature tscA fails to accumulate (Figure 3B, tscA, black star, lanes 5–8). Conversely intron 1 is spliced and mature tscA RNA is formed in the raa1-314B mutant, or in the mNS and mAM transformants of raa1-L137H (Figure 3B, tscA, black star, lanes 3, 9 and 11).
Taken together, these results delineate two functional domains of Raa1 (Figure 3C). A central region involved in splicing of intron 2 extends from MscI to somewhere between BglII and ScaI, with a possible additional contribution from the N-terminal region upstream of AgeI. A C-terminal region involved in splicing of intron 1 and tscA maturation starts at the breakpoint of the Arg7–Raa1 fusion in raa1-314B (Figure 3C, last open bar at the bottom) and ends somewhere between KpnI and the C-terminus. The two domains may partly overlap in a short region to the left of the ScaI site.
Raa1, a novel protein with internal repeats
The Predotar and ChloroP algorithms predict that Raa1 is imported into the chloroplast with an N-terminal transit peptide of 54 residues (Figure 4). This localization was confirmed by cell fractionation experiments (see below). A striking feature of the predicted Raa1 polypeptide (Figure 4A) is the occurrence of numerous stretches of repeated amino acids, such as (Ala)9, (Gly)9, (Gln)8 or (Pro)4. Raa1 also contains repetitive sequences like (Pro-Gln)4 or (Ser-Ala-Ala)5. Similar repetitive stretches also occur in other C.reinhardtii proteins, notably in another psaA splicing factor, Raa3 (10) and in other proteins involved in chloroplast gene expression (see Discussion). Raa1 harbours five copies of a 38 amino acid repeat (Figure 4B). The first three repeats are in tandem arrangement (residues 1423–1536), the other two are separate, further on the C-terminal side (1772–1809 and 1826–1863). They are related to the PPPEW repeats that are found in Tbc2, a nucleus-encoded protein of C.reinhardtii involved in translation of the chloroplast psbC mRNA . Four of the 38 amino acid repeats (repeats 2 to 5) are retained in raa1-314B, which can process tscA RNA and splice intron 1. The last repeat is disrupted in the mKpn deletion which can splice intron 2 but not intron 1, and two of the repeats (2 and 3) are in the region where the two functional domains of Raa1 may overlap upstream of the ScaI site. No other significant similarity was found between the deduced Raa1 polypeptide and sequences deposited in public databases.
Figure 4 The predicted Raa1 polypeptide. (A) The putative transit peptide (54 amino acids) is shown in italics and underlined (thin wavy line). Repeats of four or more identical consecutive amino acid residues are shaded, as well as repeats of alternating residues. The 38 amino acid repeats are underlined (thick black line or double line). (B) Alignment of the 38 amino acid repeats. Residues that are conserved in at least three of the five repeats are shaded. Segments predicted to form -helices are underlined (28). The consensus shows residues that are conserved in 3 repeats (uppercase), and 4 or 5 repeats (bold uppercase). The lowercase w represents a position where the aromatic residue W is present twice and Y once. The consensus for the 38 amino acid repeat (PPPEW) of Tbc2 (31) is shown below, with asterisks marking identical residues. The 34 amino acid TPR repeat with its two helical domains (underlined) and the 35 amino acid PPR repeat (presumptive helical domains underlined) are shown for comparison (51).
Raa1 is a membrane-associated chloroplast protein
To investigate the subcellular localization of Raa1, we used a strain expressing the HA-tagged wild-type midigene. In this construct, the genomic fragment, which extends 3.8 kb upstream of the ATG and also includes the 5' part of the gene with the first five introns, should preserve normal expression and regulation. As observed for the native protein revealed with the Raa1 antibody (Figure 1B), the HA-tagged protein also migrates more slowly than expected for its calculated molecular weight of 205 kDa (Figure 3A, lanes 1 and 11). This may be due to the unusual sequence features of the protein and its large size.
In order to determine the subcellular localization of the Raa1 protein, chloroplasts were prepared by Percoll gradient centrifugation. Raa1 was found in the chloroplast fraction, which also contains the other chloroplast proteins PsaA and Rubisco, but only trace levels of the cytosolic translation factor eiF4A (Figure 5A, lane 2). When isolated chloroplasts were further fractionated, Raa1 was enriched in the crude membrane pellet along with the integral thylakoid protein PsaA (lane 4), and depleted from the supernatant which contains the soluble protein Rubisco (lane 3). Raa1 thus fractionates as a chloroplast protein, as expected from the genetic data, although its presence in other organelles, such as mitochondria, is not ruled out. The membrane association of Raa1 was confirmed by sucrose gradient equilibrium centrifugation (data not shown).
Figure 5 Raa1 is part of a large RNase sensitive complex. (A) Subcellular localization of Raa1. Fractions were prepared from a transformant of raa1-314B expressing the HA-tagged Raa1 protein. The four lanes contain total cell extracts (Tot), the chloroplast fraction (Cp) and a chloroplast lysate fractionated by centrifugation into a soluble fraction (Sup) and a crude membrane fraction (Pellet). Immunoblots with antibodies against the HA epitope (Raa1::HA), PsaA (an integral membrane protein of Photosystem I), the small subunit of Rubisco (a soluble stromal protein) and eIF4A (a cytoplasmic translation factor, the nature of the sligthly larger cross-reacting band in the chloroplast fraction is not known). Equal amounts of protein were loaded in each lane (60 μg). (B) Raa1 is part of a large RNase-sensitive complex. A soluble extract was prepared by salt treatment of a total membrane fraction, and subjected to sucrose gradient sedimentation. Ten fractions were collected and analyzed by immunoblotting with a monoclonal anti-HA antibody. The extract was either untreated, mock treated for 20 min at 16°C, or treated with RNase before loading on the gradient. The sedimentation of protein markers of known size is indicated (in kDa) above the lanes.
Raa1 is part of a large RNase sensitive complex
The Raa1 protein can be released from the membrane fraction by treatment with moderate concentrations of salts such as 0.2 M ammonium sulphate (data not shown). An extract was prepared by washing membranes with salt and was fractionated by sedimentation on sucrose gradients (Figure 5B). Part of Raa1 was found in the lower part of the gradient (fractions 2–4), as a large complex which sediments just ahead of the 670 kDa molecular mass marker. Part of Raa1 was also found in a complex that sediments more slowly (fractions 6–8), where it is associated with Raa2 (32).
When the ammonium sulphate extract was incubated with RNase before sucrose gradient sedimentation, the position of the large complex shifted towards the top of the gradient. This shift was not observed when the extract was mock-incubated in the absence of RNase. The sensitivity to RNase shows the presence of RNA in the large Raa1 complex.
DISCUSSION
Raa1 is a multifunctional splicing factor
We have characterized a new mutant, raa1-314B, defective in trans-splicing of only the second intron of psaA (class A), and a mutant that fails to splice both introns, raa1-L137H (class B). The two mutants are clearly allelic since they fail to complement in genetic tests and both can be rescued by the wild-type Raa1 gene. Thus Raa1 has at least two distinct functions, in splicing of intron 1 and of intron 2. In raa1-HN31, which is an allele of raa1-L137H (9), there is no detectable mature tscA (29), the small RNA required for splicing intron 1 (7). The same defect is also observed in raa1-L137H, but in contrast raa1-314B splices intron 1 and accumulates mature tscA RNA. Thus the function of Raa1 in splicing of intron 1 may be direct (in this case Raa1 would have three distinct roles), or may be indirectly mediated by tscA. Two other factors which are required for intron 1 splicing and for processing of tscA were described recently, Rat1 and Rat2 (12). Like for Raa1, it is still unclear whether they are directly or indirectly involved in intron 1 splicing. The pattern of longer tscA transcripts that accumulate in raa1-L137H and raa1-HN31 and in the rat1 and rat2 mutants are similar (12,29). Thus a complex processing pathway requiring at least three factors (Raa1, Rat1 and Rat2) is involved in the maturation of tscA RNA from longer precursors.
The partial defect in raa1-314B can be explained by our observation that in raa1-314B the 3' part of Raa1 is expressed as a fusion to Arg7: this domain is thus sufficient to promote tscA maturation and splicing of intron 1. Our deletion analysis of Raa1 confirms that this 3'-terminal domain is required for tscA maturation and splicing of intron 1, while a more central domain is required for splicing of intron 2. However we did not reveal separate domains required for intron 1 splicing and for tscA RNA maturation. This is the first time that in Chlamydomonas a factor involved in splicing of the two introns of psaA (class B) is cloned and characterized. Interestingly, of the fourteen nuclear loci that are required for maturation of psaA transcripts, only two belong to this class. The other factors described to date, Raa2 (11), Raa3 (10), Rat1 and Rat2 (12) were shown to be required for either one of the two intervening sequences (classes A and C). The presence of the two functional domains in the same factor suggests that Raa1 may coordinate splicing of the two psaA introns. Mutants defective in splicing of either intron 1 or 2 show that the intervening sequences can be excised independently from each other and in either order (6). However, a mechanism that allows coordinate splicing of the two introns may improve the efficiency of maturation of psaA transcripts. Chloroplast factors with dual targets were found in Arabidopsis where PGR3 is required for expression of the petL operon and of an ndh gene (33) and in maize, where CRP1 functions in processing of petD RNA, and in translation of petA (34).
Raa1 is part of a large RNA–protein complex associated with membranes
We have cloned the Raa1 gene, which encodes a large predicted polypeptide of 206 kDa. We have shown that Raa1 is found in the chloroplast fraction, as expected from its predicted transit peptide and from its role in plastid gene expression, although its presence in other organelles cannot be ruled out. Surprisingly it is associated with membranes, and shares this property with Raa2 (formerly Maa2), a protein that is required for trans-splicing of the second intron of psaA (11), and Rat1, a factor required for processing of tscA and splicing of the first intron (12). This membrane association may be relevant because PsaA is a core subunit of photosystem I embedded in the thylakoid membrane. However Raa3, a factor required for splicing of the first intron of psaA, is mostly a stromal protein (10), like proteins involved in splicing of maize chloroplast group II introns, CRS1 and CRS2 (35,36). Chloroplast polyribosomes and some RNA-binding proteins are associated with membranes, and this association may facilitate the assembly of the photosystems and other membrane complexes . Whether the membrane association of Raa1, Raa2 and Rat1 reflects a coupling between RNA splicing, translation and protein assembly into the membrane complexes remains a matter of speculation.
When Raa1 is released from membranes with moderate concentrations of salt, it sediments in large complexes which can be resolved into two peaks in sucrose gradients. Raa1 is in part associated with Raa2, with which it can be co-immunoprecipitated (32). The complex containing Raa1 and Raa2 corresponds to the first peak. The faster sedimenting component of the second peak (700 kDa) is sensitive to treatment with RNase, implying that this Raa1 complex contains RNA. This finding fits with the role of Raa1 in splicing. However the precursor transcripts are broadly distributed in the gradients (data not shown), and we have not been able to conclusively determine whether they are part of the complex. The RNase-resistant fraction still sediments rapidly, indicating that Raa1 is in a multimeric complex, perhaps in association with other proteins. Raa3 is also part of a large ribonucleoprotein complex that contains exon 1 and the tscA RNA (10). It will be interesting to investigate interactions of Raa1 with Rat1 in splicing of intron 1. Although Raa3 is mainly in the soluble fraction, and Raa1 and Rat1 in the membrane fraction, it is possible that they could also interact transiently.
Raa1 is a novel splicing factor with repeats
Because of the limited sequence homology with other known proteins, it is difficult to predict the mode of action of Raa1. As already suggested for other proteins found to be essential for group II intron splicing, we expect that Raa1 may function as an RNA chaperone or a stabilizer of active RNA conformations (8). In maize, the CRS1 protein is specifically required for splicing of a group II intron in atpF, it binds the intron and promotes its folding (38). CRS2, together with CAF1 or CAF2, is required for splicing nine other group II introns (35,39). CRS1, CAF1 and CAF2 contain conserved CRM domains which may be RNA-binding modules (40). In plants, genes for proteins similar to intron-encoded maturases are found in the chloroplast and in the nucleus (41,42). Other proteins which associate in the chloroplast with group II introns have been described, but their role in splicing has not yet been established (43,44).
One of the striking features of the predicted Raa1 protein is the presence of numerous stretches of repeated amino acids. Such repetitive stretches occur occasionally in other proteins, but in Raa1 they are particularly long and numerous, and involve diverse residues (Ala, Gln, Ser, Pro and Arg). Several other proteins that share this feature with Raa1 are also involved in chloroplast gene expression, such as Raa3 which is also a psaA trans-splicing factor (10), Nac2, Mbb1 and Mca1 which are involved in RNA processing and stability (45–47), or Tbc2 which is implicated in translation (31). The functional significance of these repetitive stretches is unknown. They may play a structural role and may occur in parts of the protein where the precise sequence is not important so that it has drifted during evolution. There are five copies of a 38 amino acid repeat in Raa1, related to the 38 amino acid PPPEW repeats of Tbc2 (Figure 4). Four of the repeats are retained in the Arg7–Raa1 fusion of raa1-314B which is sufficient for tscA processing and intron 1 splicing, and the last repeat is interrupted in the mKpn deletion which disrupts these activities (Figure 3C). Two of the repeats are in the region of possible overlap with the domain required for intron 2 splicing, just upstream of the ScaI site.
We speculate that although they have a different sequence, the 38 amino acid repeats in Chlamydomonas might be a counterpart of the repeats in the PPR and TPR proteins, many of which are also implicated in RNA maturation. The PPR proteins form a large family in plants where a majority are predicted to be located in the organelles and many are known to interact with RNA . The TPR-like repeats are found in Mbb1 and Nac2 of C.reinhardtii (45,46) and in HCF107 of Arabidopsis (49). Each TPR repeat unit folds back to form two anti-parallel -helices which are stacked with the helices of the adjacent repeats to generate a superhelical structure (50). The related PPR repeats could form a similar structure (51), and one side of the superhelix could provide a surface for protein–RNA interactions, analogous to the protein–RNA interactions between the Puf repeats in Pumilio and their RNA target (52). Thus the repeat-containing chloroplast factors may have been recruited during evolution, by adaptation of a modular protein fold, to recognize new partners or substrates. It will be of interest to determine whether the repeats in Raa1 actually bind RNA, and specifically the psaA precursor transcripts or the tscA RNA.
In conclusion, in this report we describe a novel factor, Raa1, which is required for trans-splicing of both psaA introns in Chlamydomonas. Distinct domains in Raa1 are responsible for at least two functions, processing of tscA (the small RNA which is required for splicing of psaA intron 1) and splicing of psaA intron 2. Raa1 may thus play a role in coordinating the excision of the two psaA introns. Our analysis of Raa1 provides a basis to further investigate how different events of chloroplast RNA maturation are connected.
SUPPLEMENTARY DATA
Supplementary Data are available at NAR Online.
ACKNOWLEDGEMENTS
The authors wish to thank Nicolas Roggli for preparing the figures, the members of the Rochaix laboratory for helpful discussions, Hans Sommer for providing a cDNA library, Cris Kuhlemeier for a gift of eIF4A antibody and Frédy Barneche, Stéphane Bellafiore, Paula Duek, Angela Falciatore and Frédéric Iseni for helpful comments on the manuscript. L.M. was supported by a long-term HFSP fellowship. This work was supported by a grant of the Swiss National Fund for Scientific Research (3100-0667763.02). Funding to pay the Open Access publication charges for this article was provided by the Swiss National Fund for Scientific Research.
REFERENCES
Qin, P.Z. and Pyle, A.M. (1998) The architectural organization and mechanistic function of group II intron structural elements Curr. Opin. Struct. Biol, . 8, 301–308 .
Bonen, L. and Vogel, J. (2001) The ins and outs of group II introns Trends Genet, . 17, 322–331 .
Toro, N. (2003) Bacteria and Archaea Group II introns: additional mobile genetic elements in the environment Environ. Microbiol, . 5, 143–151 .
Villa, T., Pleiss, J.A., Guthrie, C. (2002) Spliceosomal snRNAs: Mg(2+)-dependent chemistry at the catalytic core? Cell, 109, 149–152 .
Kück, U., Choquet, Y., Schneider, M., Dron, M., Bennoun, P. (1987) Structural and transcription analysis of the two homolgous genes for the P700 chlorophyll a-apoprotein in Chlamydomonas reinhardtii: evidence for in vivo trans-splicing EMBO J, . 6, 2185–2195 .
Choquet, Y., Goldschmidt-Clermont, M., Girard-Bascou, J., Kuck, U., Bennoun, P., Rochaix, J.D. (1988) Mutant phenotypes support a trans-splicing mechanism for the expression of the tripartite psaA gene in the C.reinhardtiichloroplast Cell, 52, 903–913 .
Goldschmidt-Clermont, M., Choquet, Y., Girard-Bascou, J., Michel, F., Schirmer-Rahire, M., Rochaix, J.D. (1991) A small chloroplast RNA may be required for trans-splicing in Chlamydomonas reinhardtii Cell, 65, 135–143 .
Schroeder, R., Grossberger, R., Pichler, A., Waldsich, C. (2002) RNA folding in vivo Curr. Opin. Struct. Biol, . 12, 296–300 .
Goldschmidt-Clermont, M., Girard-Bascou, J., Choquet, Y., Rochaix, J.D. (1990) Trans-splicing mutants of Chlamydomonas reinhardtii Mol. Gen. Genet, . 223, 417–425 .
Rivier, C., Goldschmidt-Clermont, M., Rochaix, J.D. (2001) Identification of an RNA–protein complex involved in chloroplast group II intron trans-splicing in Chlamydomonas reinhardtii EMBO J, . 20, 1765–1773 .
Perron, K., Goldschmidt-Clermont, M., Rochaix, J.D. (1999) A factor related to pseudouridine synthases is required for chloroplast group II intron trans-splicing in Chlamydomonas reinhardtii EMBO J, . 18, 6481–6490 .
Balczun, C., Bunse, A., Hahn, D., Bennoun, P., Nickelsen, J., Kuck, U. (2005) Two adjacent nuclear genes are required for functional complementation of a chloroplast trans-splicing mutant from Chlamydomonas reinhardtii Plant J, . 43, 636–648 .
Harris, E.H. The Chlamydomonas sourcebook. A comprehensive guide to biology and laboratory use, (1989) San Diego, CA Academic Press, Inc .
Debuchy, R., Purton, S., Rochaix, J.D. (1989) The argininosuccinate lyase gene of Chlamydomonas reinhardtii: an important tool for nuclear transformation and for correlating the genetic and molecular maps of the ARG7 locus EMBO J, . 8, 2803–2809 .
Kindle, K.L. (1990) High-frequency nuclear transformation of Chlamydomonas reinhardtii Proc. Natl Acad. Sci. USA, 87, 1228–1232 .
Shimogawara, K., Fujiwara, S., Grossman, A., Usuda, H. (1998) High-efficiency transformation of Chlamydomonas reinhardtii by electroporation Genetics, 148, 1821–1828 .
Sambrook, J., Fritsch, E.F., Maniatis, T. Molecular Cloning. A laboratory manual, (1989) Cold Spring Harbor, NY Cold Spring Harbor Laboratory Press .
Purton, S. and Rochaix, J.D. (1994) Complementation of a Chlamydomonas reinhardtii mutant using a genomic cosmid library Plant Mol. Biol, . 24, 533–537 .
Zhang, H., Herman, P.L., Weeks, D.P. (1994) Gene isolation through genomic complementation using an indexed library of Chlamydomonas reinhardtii DNA Plant Mol. Biol, . 24, 663–672 .
Sizova, I., Fuhrmann, M., Hegemann, P. (2001) A Streptomyces rimosus aphVIII gene coding for a new type phosphotransferase provides stable antibiotic resistance to Chlamydomonas reinhardtii Gene, 277, 221–229 .
Goldschmidt-Clermont, M. and Rahire, M. (1986) Sequence, evolution and differential expression of the two genes encoding variant small subunits of ribulose bisphosphate carboxylase/oxygenase in Chlamydomonas reinhardtii J. Mol. Biol, . 191, 421–432 .
Silflow, C.D., LaVoie, M., Tam, L.W., Tousey, S., Sanders, M., Wu, W., Borodovsky, M., Lefebvre, P.A. (2001) The Vfl1 Protein in Chlamydomonas localizes in a rotationally asymmetric pattern at the distal ends of the basal bodies J. Cell. Biol, . 153, 63–74 .
Merendino, L., Falciatore, A., Rochaix, J.D. (2003) Expression and RNA-binding properties of the chloroplast ribosomal protein S1 from Chlamydomonas reinhardtii Plant Mol. Biol, . 53, 371–382 .
Schloss, J.A. (1990) A Chlamydomonas gene encodes a G protein beta subunit-like polypeptide Mol. Gen. Genet, . 221, 443–452 .
Silflow, C., Rochaix, J.D., Goldschmidt-Clermont, M., Merchant, S. The Molecular Biology of Chloroplasts and Mitochondria in Chlamydomonas, (1998) Dordrecht, The Netherlands Kluwer Academic Publishers pp. 25–40 Vol. 7 .
Emanuelsson, O., Nielsen, H., von Heijne, G. (1999) ChloroP, a neural network-based method for predicting chloroplast transit peptides and their cleavage sites Protein Sci, . 8, 978–984 .
Small, I., Peeters, N., Legeai, F., Lurin, C. (2004) Predotar: a tool for rapidly screening proteomes for N-terminal targeting sequences Proteomics, 4, 1581–1590 .
Garnier, J., Gibrat, J.F., Robson, B. (1996) GOR method for predicting protein secondary structure from amino acid sequence Methods Enzymol, . 266, 540–553 .
Hahn, D., Nickelsen, J., Hackert, A., Kuck, U. (1998) A single nuclear locus is involved in both chloroplast RNA trans-splicing and 3' end processing Plant J, . 15, 575–581 .
Baginsky, S., Siddique, A., Gruissem, W. (2004) Proteome analysis of tobacco bright yellow-2 (BY-2) cell culture plastids as a model for undifferentiated heterotrophic plastids J. Proteome Res, . 3, 1128–1137 .
Auchincloss, A.H., Zerges, W., Perron, K., Girard-Bascou, J., Rochaix, J.D. (2002) Characterization of Tbc2, a nucleus-encoded factor specifically required for translation of the chloroplast psbC mRNA in Chlamydomonas reinhardtii J. Cell. Biol, . 157, 953–962 .
Perron, K., Goldschmidt-Clermont, M., Rochaix, J.D. (2004) A multiprotein complex involved in chloroplast group II intron splicing RNA, 10, 704–710 .
Yamazaki, H., Tasaka, M., Shikanai, T. (2004) PPR motifs of the nucleus-encoded factor, PGR3, function in the selective and distinct steps of chloroplast gene expression in Arabidopsis Plant J, . 38, 152–163 .
Fisk, D.G., Walker, M.B., Barkan, A. (1999) Molecular cloning of the maize gene crp1 reveals similarity between regulators of mitochondrial and chloroplast gene expression EMBO J, . 18, 2621–2630 .
Till, B., Schmitz-Linneweber, C., Williams-Carrier, R., Barkan, A. (2001) CRS1 is a novel group II intron splicing factor that was derived from a domain of ancient origin RNA, 7, 1227–1238 .
Jenkins, B.D. and Barkan, A. (2001) Recruitment of a peptidyl-tRNA hydrolase as a facilitator of group II intron splicing in chloroplasts EMBO J, . 20, 872–879 .
Zerges, W. (2000) Translation in chloroplasts Biochimie, 82, 583–601 .
Ostersetzer, O., Cooke, A.M., Watkins, K.P., Barkan, A. (2005) CRS1: a chloroplast group II intron splicing factor, promotes intron folding through specific interactions with two intron domains Plant Cell, 17, 241–255 .
Ostheimer, G.J., Williams-Carrier, R., Belcher, S., Osborne, E., Gierke, J., Barkan, A. (2003) Group II intron splicing factors derived by diversification of an ancient RNA-binding domain EMBO J, . 22, 3919–3929 .
Ostheimer, G.J., Hadjivasiliou, H., Kloer, D.P., Barkan, A., Matthews, B.W. (2005) Structural analysis of the group II intron splicing factor CRS2 yields insights into its protein and RNA interaction surfaces J. Mol. Biol, . 345, 51–68 .
Mohr, G. and Lambowitz, A.M. (2003) Putative proteins related to group II intron reverse transcriptase/maturases are encoded by nuclear genes in higher plants Nucleic Acids Res, . 31, 647–652 .
Liere, K. and Link, G. (1995) RNA-binding activity of the matK protein encoded by the chloroplast trnK intron from mustard (Sinapis alba L.) Nucleic Acids Res, . 23, 917–921 .
Nakamura, T., Ohta, M., Sugiura, M., Sugita, M. (1999) Chloroplast ribonucleoproteins are associated with both mRNAs and intron-containing precursor tRNAs FEBS Lett, . 460, 437–441 .
Bunse, A.A., Nickelsen, J., Kuck, U. (2001) Intron-specific RNA-binding proteins in the chloroplast of the green alga Chlamydomonas reinhardtii Biochim. Biophys. Acta, . 1519, 46–54 .
Boudreau, E., Nickelsen, J., Lemaire, S.D., Ossenbuhl, F., Rochaix, J.D. (2000) The Nac2 gene of Chlamydomonas encodes a chloroplast TPR-like protein involved in psbD mRNA stability EMBO J, . 19, 3366–3376 .
Vaistij, F.E., Boudreau, E., Lemaire, S.D., Goldschmidt-Clermont, M., Rochaix, J.D. (2000) Characterization of Mbb1, a nucleus-encoded tetratricopeptide-like repeat protein required for expression of the chloroplast psbB/psbT/psbH gene cluster in Chlamydomonas reinhardtii Proc. Natl Acad. Sci. USA, 97, 14813–14818 .
Lown, F.J., Watson, A.T., Purton, S. (2001) Chlamydomonas nuclear mutants that fail to assemble respiratory or photosynthetic electron transfer complexes Biochem. Soc. Trans, . 29, 452–455 .
Lurin, C., Andres, C., Aubourg, S., Bellaoui, M., Bitton, F., Bruyere, C., Caboche, M., Debast, C., Gualberto, J., Hoffmann, B., et al. (2004) Genome-wide analysis of Arabidopsis pentatricopeptide repeat proteins reveals their essential role in organelle biogenesis Plant Cell, 16, 2089–2103 .
Felder, S., Meierhoff, K., Sane, A.P., Meurer, J., Driemel, C., Plucken, H., Klaff, P., Stein, B., Bechtold, N., Westhoff, P. (2001) The nucleus-encoded HCF107 gene of Arabidopsis provides a link between intercistronic RNA processing and the accumulation of translation-competent psbH transcripts in chloroplasts Plant Cell, 13, 2127–2141 .
Das, A.K., Cohen, P.W., Barford, D. (1998) The structure of the tetratricopeptide repeats of protein phosphatase 5: implications for TPR-mediated protein–protein interactions EMBO J, . 17, 1192–1199 .
Small, I.D. and Peeters, N. (2000) The PPR motif: a TPR-related motif prevalent in plant organellar proteins Trends Biochem. Sci, . 25, 46–47 .
Wang, X., McLachlan, J., Zamore, P.D., Hall, T.M. (2002) Modular recognition of RNA by a human pumilio-homology domain Cell, 110, 501–512 .(Livia Merendino, Karl Perron, Michèle Ra)