• Aucun résultat trouvé

A pioneer protein is part of a large complex involved in trans-splicing of a group II intron in the chloroplast of Chlamydomonas reinhardtii

N/A
N/A
Protected

Academic year: 2022

Partager "A pioneer protein is part of a large complex involved in trans-splicing of a group II intron in the chloroplast of Chlamydomonas reinhardtii"

Copied!
43
0
0

Texte intégral

(1)

Article

Reference

A pioneer protein is part of a large complex involved in trans-splicing of a group II intron in the chloroplast of Chlamydomonas reinhardtii

LEGENDRE LEFEBVRE, Linnka, et al.

Abstract

Splicing of organellar introns requires the activity of numerous nucleus-encoded factors. In the chloroplast of Chlamydomonas reinhardtii, maturation of psaA mRNA encoding photosystem I subunit A involves two steps of trans-splicing. The exons, located on three separate transcripts, are flanked by sequences that fold to form the conserved structures of two group II introns. A fourth transcript contributes to assembly of the first intron, which is thus tripartite.

The raa7 mutant (RNA maturation of psaA 7) is deficient in trans-splicing of the second intron of psaA, and may be rescued by transforming the chloroplast genome with an intron-less version of psaA. Using mapped-based cloning, we identify the RAA7 locus, which encodes a pioneer protein with no previously known protein domain or motif. The Raa7 protein, which is not associated with membranes, localizes to the chloroplast. Raa7 is a component of a large complex and co-sediments in sucrose gradients with the previously described splicing factors Raa1 and Raa2. Based on tandem affinity purification of Raa7 and mass spectrometry, Raa1 and Raa2 were identified as [...]

LEGENDRE LEFEBVRE, Linnka, et al. A pioneer protein is part of a large complex involved in trans-splicing of a group II intron in the chloroplast of Chlamydomonas reinhardtii. The Plant Journal, 2016, vol. 85, no. 1, p. 57-69

DOI : 10.1111/tpj.13089 PMID : 26611495

Available at:

http://archive-ouverte.unige.ch/unige:86344

Disclaimer: layout of this document may differ from the published version.

1 / 1

(2)

A pioneer protein is part of a large complex involved in trans-splicing of a group II 1

intron in the chloroplast of Chlamydomonas reinhardtii 2

Linnka Lefebvre-Legendre1, Olga Reifschneider2, Laxmikanth Kollipara3, 3

Albert Sickmann3,4,5, Dirk Wolters6, Ulrich Kück2,7 and Michel Goldschmidt-Clermont1,7 4

1 Department of Botany and Plant Biology and Department of Molecular Biology, University of 5

Geneva, 30 quai Ernest Ansermet, 1211 Geneva 4, Switzerland 6

2Lehrstuhl für Allgemeintabe und Molekulare Botanik, Ruhr-University Bochum, 7

Universitätsstr. 150, 44801 Bochum, Germany 8

3Leibniz-Institut für Analytische Wissenschaften – ISAS – e.V., Otto-Hahn-Straße 6b, 44227 9

Dortmund, Germany 10

4Department of Chemistry, College of Physical Sciences, University of Aberdeen, Aberdeen, 11

Scotland, United Kingdom 12

5Medizinische Fakultät, Ruhr-University Bochum, Universitätsstr. 150, 44801 Bochum, 13

Germany 14

6Department of Analytical Chemistry, Ruhr-University Bochum, Universitätsstr. 150, 44801 15

Bochum, Germany 16

Runnning title: Chloroplast group II intron trans-splicing complex 17

Keywords: trans-splicing; plastid; chloroplast; splicing complex; psaA; Chlamydomonas 18

reinhardtii, mapped-based cloning, mass spectrometry, yeast two-hybrid.

19 20

7Corresponding authors:

21

Michel Goldschmidt-Clermont 22

Department of Botany and Plant Biology 23

University of Geneva, 24

30 quai Ernest Ansermet, 25

1211 Geneva 4, Switzerland 26

Contact email: michel.goldschmidt-clermont@unige.ch 27

phone: +41 22 3796188 28

Ulrich Kück 29

Department for General and Molecular Botany 30

Ruhr-Universität Bochum 31

Universitätsstr. 150 32

44801 Bochum, Germany 33

email: ulrich.kueck@rub.de 34

phone: +49 234 32 26212 35

(3)

Emails of other authors 36

Linnka Lefebvre-Legendre: Linnka.Legendre@unige.ch 37

Olga Reifschneider: olga.reifschneider@rub.de 38

Laxmikanth Kollipara: laxmikanth.kollipara@isas.de 39

Albert Sickmann: albert.sickmann@isas.de 40

Dirk Wolters: dirk.wolters@rub.de 41

42

Word count 43

Total: 8148

44

Summary: 248

45

Significance statement 74 46

Introduction: 802 47

Results: 2086

48

Discussion: 1548 49

Experimental Proc: 1880 50

Acknowledgements: 74 51

Figure legends: 1061 52

References: 1675 53

54 55

(4)

SUMMARY 56

Splicing of organellar introns requires the activity of numerous nucleus-encoded factors. In 57

the chloroplast of Chlamydomonas reinhardtii, the maturation of photosystem I subunit A 58

(psaA) mRNA involves two steps of trans-splicing. The exons, located on three separate 59

transcripts, are flanked by sequences that can fold to form the conserved structures of two 60

group II introns. A fourth transcript contributes to the assembly of the first intron which is thus 61

tripartite. The raa7 mutant (RNA of psaA 7) is deficient in trans-splicing of the second intron 62

of psaA, and can be rescued by transforming the chloroplast genome with an intron-less 63

version of psaA. Using mapped-based cloning, we identify the RAA7 locus which encodes a 64

pioneer protein with no previously-known protein domain or motif. The Raa7 protein, which is 65

not associated with membranes, localizes to the chloroplast. Raa7 is a component of a large 66

complex and co-sediments in sucrose gradients with the previously described splicing factors 67

Raa1 and Raa2. Based on tandem affinity purification of Raa7 and mass-spectrometry, Raa1 68

and Raa2 are identified as interacting partners of Raa7. Yeast two-hybrid experiments 69

indicate that the interaction of Raa7 with Raa1 and Raa2 may be direct. We conclude that 70

Raa7 is a component of a multimeric complex required for trans-splicing of the second intron 71

of psaA. The characterization of this psaA trans-splicing complex is also of interest in an 72

evolutionary perspective because the nuclear spliceosomal introns are thought to derive from 73

group II introns, with which they share mechanistic and structural similarity.

74 75 76 77

(5)

INTRODUCTION 78

During evolution, many of the ancestral cyanobacterial genes of the plastid have been 79

transferred to the nucleus or have been lost, so that only approximately one hundred proteins 80

are still encoded in the chloroplast genome, and most of the few thousand proteins of the 81

plastid proteome are nucleus-encoded and imported. The post-transcriptional steps of 82

chloroplast gene expression are highly intricate, and involve a large number of proteins that 83

are mostly nucleus-encoded (Barkan 2011, Stern et al. 2010).

84

In Chlamydomonas reinhardtii (Chlamydomonas), a particularly complex example of post- 85

transcriptional RNA maturation in the chloroplast is the trans-splicing of psaA (photosystem I 86

subunit A), which encodes a major subunit of photosystem I (PSI). The psaA gene is split in 87

three separate exons dispersed on the chloroplast genome (Kück et al. 1987). The exons are 88

transcribed separately and are flanked by sequences that can assemble to form the 89

conserved structures of group II introns. The mature psaA mRNA is formed by two steps of 90

splicing in trans (Choquet et al. 1988, Herrin and Schmidt 1988). Furthermore, the tscA locus 91

(trans-splicing chloroplast A), encodes a short non-coding RNA that completes the structure 92

of the first tripartite group II intron (Goldschmidt-Clermont et al. 1991).

93

While some group II introns from bacteria or fungal mitochondria are capable of autocatalytic 94

self-splicing in vitro, splicing of these introns in vivo requires splicing factors. The group II 95

introns of plant and algal organelles appear to have degenerated during evolution and 96

require a large cohort of protein factors to assist in their cis- or trans-splicing (de Longevialle 97

et al. 2010b, Glanz and Kück 2009, Stern, et al. 2010). Several mutants affected in trans- 98

splicing of psaA have been identified and assigned to three classes depending on their 99

splicing deficiencies (Choquet, et al. 1988, Herrin and Schmidt 1988). Class A mutants are 100

defective in splicing of exons 2 and 3, class C mutants are affected in splicing of the 101

transcripts of exons 1 and 2, and class B mutants are defective in both steps of trans- 102

splicing. Genetic analysis of these mutants revealed that at least 14 nucleus-encoded factors 103

are involved in the process (Goldschmidt-Clermont et al. 1990). To date, 6 of these proteins 104

have been identified and characterized. Rat2 (RNA maturation of psaA tscA 2) is necessary 105

for the processing of tscA RNA from a polycistronic precursor and is directly or indirectly 106

required for splicing of intron 1 (Balczun et al. 2005). In Raa1 (RNA maturation of psaA 1), 107

the C-terminal part is necessary for processing of tscA RNA and for splicing of intron 1, while 108

a central domain is involved in splicing of intron 2 (Merendino et al. 2006). Raa2 is involved 109

in splicing of intron 2 (Perron et al. 1999), and together with Raa1 is part of a 500 kDa 110

multiprotein complex (Perron et al. 2004). Raa3, and the recently discovered Raa8, are 111

necessary for the splicing of intron 1 (Marx et al. 2015, Rivier et al. 2001) and are part of a 112

(6)

large 1700 kDa complex that includes the tscA RNA and the precursor of psaA exon1. Raa4 113

is also necessary for the splicing of intron 1, and was shown to bind tscA RNA in vitro. In 114

addition, two-hybrid experiments in yeast suggested an interaction between Raa1, Raa3, 115

Raa4 and Rat2 in a complex involved in splicing of intron 1 (Jacobs et al. 2013).

116

Interestingly, Raa2 is similar to pseudouridine synthases but the respective enzymatic 117

activity is not necessary for its function in trans-splicing (Balczun, et al. 2005, Perron, et al.

118

1999). Raa1 displays octatricopeptide repeats (OPR) which are thought to mediate RNA 119

binding (Boulouis et al. 2015, Merendino, et al. 2006, Rahire et al. 2012).The three other 120

trans-splicing factors that have been identified show no previously described RNA-binding 121

domains or homology to known proteins. In addition, three proteins that bind tscA or psaA 122

precursor RNAs in vitro have been identified. Cpn60 (Chaperonin 60) binds psaA intron1 123

(Balczun et al. 2006), while cNAPL (chloroplast Nucleosome Assembly Protein Like) and 124

Rab1 (Raa4 binding protein 1) bind the tscA RNA (Glanz et al. 2006, Glanz et al. 2012, 125

Jacobs, et al. 2013). Furthermore Rab1 interacts with four other splicing factors, Raa1, Raa3, 126

Raa4 and Rat2.

127

It was shown that the 500 kDa complex containing Raa1 and Raa2 is absent in the L121G 128

mutant that belongs to class A (Perron, et al. 2004). These data suggested that the L121G 129

mutant could be deficient in a further protein involved in the splicing of intron 2. To test this 130

hypothesis, we characterized the L121G mutant and thus identified a factor that we named 131

Raa7, which is required for the trans-splicing of exons 2 and 3 of psaA and which is 132

associated with a large multimeric complex of approximately 500 kDa also containing Raa1 133

and Raa2. Raa7 is a pioneer protein as it lacks any identifiable sequence motif.

134 135

RESULTS 136

The raa7 mutant is affected in trans-splicing of psaA mRNA 137

The mutant L121G was part of a collection of PSI-deficient mutants, obtained by UV-light 138

mutagenesis (Girard et al. 1980). It fails to accumulate normal levels of psaA mRNA and 139

belongs to class A mutants that are unable to assemble exons 2 and 3 (Goldschmidt- 140

Clermont, et al. 1990). To confirm these observations, the L121G mutant was back-crossed 141

three times to the wild type, leading to a strain that we called raa7. Northern-blot analysis 142

(Figure 1A) showed a strong decrease in the level of the mature psaA mRNA (exons 1-2-3) 143

in raa7 compared to the wild type, while the precursor of exon 3 (pre-exon 3) was present at 144

normal levels and the precursor of exon 1(pre-exon 1) over-accumulated. These data are 145

(7)

consistent with previous results and indicate that the absence of Raa7 leads to a defect in 146

splicing and accumulation of the mature psaA mRNA.

147

In order to confirm that the defect in the raa7 mutant is specific to PSI, an analysis of 148

photosynthetic proteins was performed by SDS-PAGE and immunoblotting (Figure 1B). The 149

amounts of PSI subunits PsaA, PsaD and PsaF were reduced below detection levels in the 150

raa7 mutant, whereas subunits representative of photosystem II, the cytochrome b6f 151

complex, the light-harvesting antennae LHCI and LHCII, the ATP synthase and Rubisco 152

accumulated to normal levels in the mutant compared to the wild type, suggesting that the 153

raa7 mutant is only affected in PSI.

154

It was shown previously that trans-splicing of psaA can be by-passed by transforming the 155

chloroplast genome with an intron-less version of the psaA gene (psaA-Δi) where exons 1, 2 156

and 3 of psaA are fused and placed under the control of the promoter and 5’UTR of exon 1 157

(Lefebvre-Legendre et al. 2014). Its flanking sequences direct the replacement of the 158

resident psaA exon 3 by the intron-less construct, and a selectable marker (aadA), which 159

confers resistance to spectinomycin, is placed downstream (Goldschmidt-Clermont 1991).

160

The psaA-Δi construct was introduced by biolistic transformation in the chloroplast genome 161

of raa7, the transformants were selected on spectinomycin-containing medium. Figure 2A 162

shows the growth properties of one of these transformants (raa7/psaA-Δi) compared to raa7 163

and the WT control strain. While raa7 was unable to grow on minimal medium and was 164

sensitive to moderate light intensities on acetate-containing medium, raa7/psaA-Δi grew like 165

the wild-type strain in all conditions. This indicates that expression of PsaA and assembly of 166

PSI are rescued in the mutant when carrying the intron-less copy of psaA. Immunoblotting 167

analysis confirmed the normal amount of PsaA protein in raa7/psaA-Δi compared to the wild 168

type (Figure 2B), showing that Raa7 is required for trans-splicing of psaA, and that this is its 169

only essential function under the conditions tested.

170

Identification of the RAA7 gene 171

First, we crossed the raa7 mutant with the Chlamydomonas grossii strain S1-D2 for linkage 172

analysis based on genetic polymorphism (Rymarquis et al. 2005). We obtained 57 tetrads 173

and from each selected one clone having inherited the raa7 mutation, as identified by its 174

defective photoautotrophic growth. Using markers based on PCR amplification with primers 175

described by (Rymarquis, et al. 2005) and listed in Table S1, we first observed linkage of the 176

RAA7 locus to the kinesin A marker located at the extremity of the short arm of chromosome 177

III. To map RAA7 more accurately, we used additional markers (Figure 3A). The marker 178

STS14 showed 100% linkage (n = 57) with RAA7, and groups of markers on either side 179

cosegregated at 98% (STS8, STS10, kinesin A, and CAPS1 on the proximal side and 180

(8)

CAPS2, STS20 and STS15 on the distal side). RAA7 thus mapped to the 405 kb interval 181

between CAPS1 (position 8’322’000 of chromosome III) and CAPS2 (position 8 727 000). In 182

a third step, Bacterial Artificial Chromosome genomic clones (BACs) covering the region 183

comprised between these two markers were used to transform the mutant and select for 184

photoautotrophic growth (Lefebvre and Silflow 1999). Only BAC 3K11 was able to 185

complement the raa7 mutant, but BAC 2F17, corresponding to the 5’ part of the BAC 3K11, 186

was unable to save the mutant. Furthermore, internal deletions of BAC 3K11 as schematized 187

in Figure 3A did not prevent complementation, showing that the RAA7 gene was contained in 188

the distal part of the BAC. Finally, transformation of raa7;cw15 with the 7.3 kb SpeI sub- 189

fragment of 3K11 gave rise to photoautotrophic colonies with high efficiency. In the 190

Chlamydomonas genome only a single gene is annotated in this fragment, identifying RAA7 191

as Cre03.g201103.t1.1 (version 5.5, Phytozome 192

10; http://phytozome.jgi.doe.gov/pz/portal.html#!info?alias=Org_Creinhardtii ).

193

Since the genomic sequence and the BAC library were obtained from a strain (CC-503) that 194

shows polymorphism with our laboratory strain (another derivative of 137c, (Harris 2009)), 195

we also screened a cosmid library constructed from the latter and recovered a cosmid clone 196

(14C2) covering the region of interest (Depege et al. 2003, Purton and Rochaix 1994).

197

Cosmid 14C2 and its 7.3 kb SpeI subfragment were capable of rescuing the raa7;cw15 198

mutant with high efficiency, confirming the identification of the RAA7 gene. The annotated 199

gene model comprises 7 exons, but sequencing of RT-PCR fragments covering the complete 200

coding sequence revealed only 6 exons (Figure 4B), a difference that was due to the fusion 201

of the predicted exon 1, intron 1 and exon 2 into a single exon. The deduced RAA7 open 202

reading frame encodes a protein of 1289 amino acids (130 kDa). We could not identify any 203

putative conserved domain in the protein. In particular a search of Interpro returned no 204

predicted family membership, domain or repeat (http://www.ebi.ac.uk/interpro/).

205

Interestingly, revertant strains which we called raa7-rev arose spontaneously when plating 206

raa7 cultures on minimal medium and selecting for photoautotrophic growth. Four of these 207

revertants were back-crossed to the raa7 mutant or to the wild-type strain. Analysis of 10 208

progeny from each cross indicated that the suppressor mutations segregated as Mendelian 209

nuclear markers and were tightly linked to raa7, suggesting a true reversion or an intragenic 210

suppression. Using RT-PCR we compared the cDNA sequences obtained from the RAA7 211

wild type, the raa7 mutant and the raa7-rev-1 revertant (Fig 3B; Figure S1). This revealed a 212

deletion of 13 bp in the raa7 mutant (nt 2864-2876) which shifted the reading frame from 213

amino acid residue 955 and led to a stop codon at position 1104. In the revertant strain, the 214

insertion of 1 bp (after nt 3045) restored the correct reading frame from amino acid residue 215

1016 to the end of the predicted protein (Fig 3B; Figure S1). These data confirm the 216

(9)

identification of RAA7 and show that the C-terminal part of the protein is essential for its 217

function.

218

An RAA7-HA “midigene” carrying an HA epitope (HaemAgglutinin) at the C-terminus (pL42, 219

Figure 3B), was constructed as a fusion within exon 2 of a genomic fragment including intron 220

1 to the 3’ part of the cDNA. A sequence encoding a triple HA epitope was inserted upstream 221

of the stop codon. Transformation of the raa7-cw15 mutant with this construct restored 222

photoautotrophic growth on minimal medium (Figure S2A ) and wild-type levels of PsaA 223

(Figure S2B), showing that the RAA7-HA midigene is functional. Using an HA antiserum for 224

immunoblotting, a band at around 175 kDa was detected in protein extracts from the 225

raa7;cw15;RAA7-HA strain (Figure 5B). This is larger than the predicted mass of 130 kDa, 226

but comparable slow migration was previously observed for other Chlamydomonas proteins 227

such as TAA1 (Lefebvre-Legendre et al. 2015), TAB1 (Rahire, et al. 2012) or RAA1 228

(Merendino, et al. 2006). Anomalous migration could be due to the presence of domains of 229

low sequence complexity (Figure S1) and the large size of these proteins.

230

Raa7 is a chloroplast protein 231

To investigate the localization of the Raa7 protein, we used the raa7;cw15;Raa7-HA 232

strain for cell fractionation experiments. Chloroplasts were prepared by Percoll gradient 233

centrifugation. Raa7 was found in the chloroplast fraction, together with the chloroplast 234

proteins PsaA and phosphoribulokinase (PRK), whereas the cytosolic protein Rpl37 was 235

detectable only in the total extract (Figure 4A). A mitochondrial fraction was also prepared, 236

which was highly enriched in the mitochondrial alternative oxidase mAOX1. Compared to the 237

total extract, Raa7-HA was depleted from this fraction in parallel with the chloroplast marker 238

PsaA. These data suggest that Raa7 is a chloroplast protein, consistent with its role in the 239

trans-splicing of psaA, but do not exclude that it could also localize to another compartment.

240

To determine whether Raa7 is a stromal or a membrane protein, we fractionated total cell 241

membranes from soluble proteins. Raa7 protein was found in the soluble fraction, where the 242

PRK protein was similarly enriched (Figure 4C), but was not detected in the crude membrane 243

fraction containing the thylakoid protein PsaA.

244

Raa7 is part of a high molecular weight complex 245

Raa2, a 40 kDa protein also involved in trans-splicing of exons 2 and 3 of psaA, is part of a 246

500 kDa complex that also contains the Raa1 protein, which is required for trans-splicing of 247

the psaA mRNA (Merendino, et al. 2006, Perron, et al. 2004). Moreover, this complex was 248

affected in the raa7 mutant (formerly L121G) (Perron, et al. 2004). To determine whether 249

Raa7 is part of this protein complex, a soluble extract of raa7;cw15;RAA7-HA was 250

(10)

fractionated by centrifugation in a sucrose gradient, and the fractions were analyzed by SDS- 251

PAGE and immunoblotting (Figure 5A). Raa7 was present mainly in fraction 4, corresponding 252

to an apparent molecular mass of approximately 500 kDa. As the molecular mass of Raa7 is 253

130 kDa, this result suggested that the protein was indeed part of a multimeric complex.

254

Furthermore, Raa2 could be detected in the same fraction as Raa7 (Figure 5A). Because the 255

anti-Raa2 antibody was not monospecific, fraction 4 was run alongside the wild type and the 256

raa2 mutant to validate the identification of Raa2 (Figure 5B). A parallel sedimentation 257

analysis of Raa1 tagged with an HA epitope in the background of the raa1 mutant 258

(raa1/RAA1-HA) showed two peaks as previously observed (Merendino, et al. 2006, Perron, 259

et al. 2004). The smaller complex containing Raa1-HA co-sedimented with Raa7 and Raa2 260

in fraction 4 (Figure 5A and B).

261

The sucrose sedimentation data suggested that Raa7 is part of a complex together with 262

Raa1 and Raa2. To confirm this interpretation, we used the previously described TAP 263

(tandem affinity purification) method coupled to mass spectrometry to purify native Raa7 264

protein complexes from chloroplasts of C. reinhardtii (Jacobs et al. 2013). A sequence 265

encoding the TAP tag (Jacobs, et al. 2013, Rigaut et al. 1999) was fused to the end of RAA7 266

in place of the HA tag in the midigene plasmid (RAA7::TAP, pL52). The TAP sequence, 267

which was optimized for the codon usage of C. reinhardtii, encodes protein A, a TEV 268

(Tobacco Etch Virus) protease cleavage site, and a calmodulin-binding peptide (Jacobs, et 269

al. 2013). When transformed with RAA7::TAP, the raa7 mutant recovered the ability to grow 270

photoautotrophically on minimal medium indicating the functionality of the fusion protein and 271

restoration of photosynthesis. Transformant T7.2 was selected for subsequent TAP 272

purification and analysis. The expression of the TAP tag in this strain was confirmed by RT- 273

PCR and by immunoblotting, which showed the expected Raa7::TAP fusion protein of 274

approximately 150 kDa (Figure S3).

275

For TAP purification of Raa7 and associated components, a total of three independent 276

experiments were performed in order to identify interactors with high confidence. To 277

discriminate between specific and non-specific interactions, control experiments using the 278

same purification parameters were conducted in parallel. Three RST-1 cultures, harboring a 279

TAP-tagged small subunit of RubisCO , and one non-tagged culture of the arg-cw15 were 280

used for TAP experiments (C1-C4). Proteins detected in these purifications were considered 281

as unspecific. A total of 53 specific proteins were detected in all three replicates of 282

Raa7::TAP, but not in control experiments. In order to further distinguish between positive 283

and false-positive candidates, subcellular localization was estimated using different in silico 284

prediction tools, namely ChloroP (Emanuelsson et al. 1999), PredAlgo (Tardif et al. 2012) 285

and TargetP (Emanuelsson et al. 2007). Compared to higher plants, C. reinhardtii chloroplast 286

(11)

targeting signals often show unique features and therefore are misinterpreted as 287

mitochondrial proteins. A total of 32 proteins showed at least two predictions of a chloroplast 288

or mitochondrial transit peptide and were assigned as putative Raa7 complex components 289

(Table 1).

290

Amongst these, Raa7 as bait protein was detected with a high number of unique peptides 291

(Table 1, “Σ ≠ peptides”) and a sequence coverage of26 %. Moreover, splicing factors Raa1 292

and Raa2 were identified in all three Raa7-TAP experiments. Thus, our TAP-MS results 293

confirm that Raa7 is part of the Raa1/Raa2 complex, which is required for splicing of the 294

second psaA group II intron.

295

To further investigate whether the interaction of Raa7, Raa2 and Raa1 is direct, we used the 296

two-hybrid method in the yeast Saccharomyces cerevisiae. We were unable to amplify the 297

full-length cDNA of Raa7 which has a high GC content, but cloned subfragments of Raa7 298

into vectors with the GAL4 activation domain (pGADT7) or the GAL4 DNA binding domain 299

(pGBKT7). A sequence encoding the C-terminal part of Raa1 and a sequence coding for the 300

nearly full length of Raa2 (except its first 15 amino acids which are predicted to be part of the 301

transit peptide) were cloned in the two vectors (Jacobs, et al. 2013). Yeast strains carrying 302

the desired plasmids were mated to obtain diploids, which were then spotted on selective 303

medium to monitor two-hybrid interactions. In order to rule out activation in the absence of an 304

interacting partner, all strains were mated against control strains carrying empty vectors with 305

either the GAL4 DNA-binding domain or the GAL4 activation domain. In none of these cases 306

was there any evidence of activation when the diploids were tested for growth on selective 307

medium. Of the different Raa7 constructs that were used, the only positive results were 308

obtained with the plasmids encoding amino acids 42-432 fused either to the activation or the 309

DNA-binding domain of GAL4. We observed that Raa7 interacts with itself and with the Raa1 310

and Raa2 proteins (Figure 5C). It should be noted that these interactions were tested on 311

medium with the highest stringency suggesting a strong interaction. However, the interaction 312

between Raa7 and Raa2 could only be detected when Raa7 was fused to the GAL4 313

activation domain and Raa2 to the GAL4 DNA-binding domain but not conversely. These 314

data strongly suggested that Raa7 interacts with Raa1 and Raa2, and that this interaction is 315

direct.

316

DISCUSSION 317

Our characterization of raa7 led to the identification of a factor that is required for trans- 318

splicing of exons 2 and 3 of psaA in the chloroplast of Chlamydomonas. As with other 319

mutants affected in trans-splicing of psaA, introduction of an intron-less version of the psaA 320

gene in the chloroplast genome (Lefebvre-Legendre, et al. 2014) was sufficient to fully 321

(12)

rescue the raa7 mutant. This observation indicates that Raa7 has no essential function other 322

than in trans-splicing of psaA under the conditions that were tested.

323

Because revertants of raa7 appeared spontaneously with a relatively high frequency, it was 324

not possible to identify the Raa7 gene by complementation with pooled clones from a library 325

of genomic DNA. We resorted to map-based cloning, and used progeny from a cross of raa7 326

to a polymorphic strain of Chlamydomonas to define genetically a 400 kb interval of 327

chromosome 3 that includes the Raa7 locus. Complementation with individual BAC and 328

cosmid clones representing this interval allowed the identification of the Raa7 gene. The 329

predicted Raa7 protein (130 kDa) has a putative transit peptide for chloroplast import, but 330

does not harbor any other previously described protein domain that we could identify. Raa7 331

is enriched in purified chloroplast but not in mitochondrial preparations, and is present in the 332

soluble fraction but is not detected in the membrane fraction.

333

Group II introns from plant and algal organelles appear to have degenerated during 334

evolution, so that for splicing they typically require protein factors, most of which are encoded 335

in the nucleus. These factors form large multimeric complexes that are thought to function as 336

spliceosomes for specific group II introns. Previous work showed that trans-splicing of exons 337

2 and 3 of psaA involves a ∼500 kDa complex that includes Raa1 and Raa2, two proteins 338

which are essential for this maturation step (Merendino, et al. 2006, Perron, et al. 1999). The 339

size of the complex was affected in the raa7 mutant (formerly L121G), suggesting that Raa7 340

could be part of the complex, or alternatively be necessary for its assembly (Perron, et al.

341

2004). To investigate the protein-protein interactions that engage Raa7, we combined three 342

different experimental approaches: sedimentation analyses, TAP purification followed by 343

mass-spectrometry and yeast two-hybrid assays. The combination of the three methods 344

offered converging evidence that Raa7 is indeed part of the ∼500 kDa complex that also 345

contains Raa1 and Raa2. These three splicing proteins have a molecular mass of 346

approximately 360 kDa, thus suggesting that further subunits involved in trans-splicing of 347

exons 2 and 3 (class A) are part of this complex. Alternatively, different stoichiometry of 348

distinct subunits may contribute to the predicted molecular weight (Goldschmidt-Clermont, et 349

al. 1990). Raa7 is a pioneer protein that does not carry any recognizable sequence motif, 350

and in particular any domain typical of RNA binding. Likewise, no RNA-binding motif was 351

detected in Raa4, even though EMSA analysis showed a direct interaction of Raa4 with tscA 352

RNA in vitro (Glanz, et al. 2012). Alternatively, Raa7 may not bind RNA directly, but be 353

indirectly associated with RNA through its interacting partners. Raa7 has several stretches of 354

repetitive amino-acid sequences called low-complexity regions (LCRs) (Glanz, et al. 2012). It 355

has been observed that proteins containing LCRs tend to have more interaction partners in 356

protein-protein interaction networks than proteins devoid of these repeats (Coletta et al.

357

(13)

2010). One of the partners of Raa7 in the 500 kDa complex is Raa1, which may contain as 358

many as 14 OPR (octratricopeptide repeat) motifs which are thought to mediate RNA binding 359

(Boulouis, et al. 2015, Merendino, et al. 2006, Rahire, et al. 2012). The other identified 360

partner, Raa2, belongs to the family of pseudouridine synthases and may hence also have 361

the ability to bind RNA (Perron, et al. 1999).

362

From this and other work, the simple picture that emerges for the trans-splicing machinery of 363

psaA includes at least two complexes (Figure 6). The first one is a ribonucleoprotein 364

complex, containing subunits that either promote maturation of tscA or trans-splicing of 365

exons 1 and 2. It contains at least 5 proteins: Raa1, Raa3, Raa4, Rat2 and Rab1 (Jacobs, et 366

al. 2013). We show here that a second complex (∼500 kDa), required for the trans-splicing of 367

exons 2 and 3, contains at least Raa1, Raa2 and Raa7. However, the trans-splicing 368

machinery may be more complex since Raa1 was previously found in another complex of 369

approximately 670 kDa (Merendino, et al. 2006). The presence of Raa1 in two different 370

complexes may be correlated with the fact that it harbors two different domains, each 371

involved in trans-splicing of one of the two split introns of psaA. Furthermore, we have shown 372

that Raa1, Raa2 and Raa7 interact although Raa7 is mainly in the soluble fraction, whereas 373

Raa1 and Raa2 were reported to be mostly associated with membranes (Merendino, et al.

374

2006, Perron, et al. 1999). Likewise, it has been shown that Raa1, Raa3 and Raa4 are part 375

of a complex, although Raa3 is a stromal protein, whereas Raa1 is mostly associated with 376

membranes (Glanz, et al. 2012, Rivier, et al. 2001). As previously suggested (Jacobs, et al.

377

2013), the composition and localization of the splicing complexes could be dynamic and 378

change with the addition or loss of some of the different component proteins. It is also 379

possible that some components are lost during the isolation of the complexes.

380

Apart from the three previously known psaA trans-splicing factors, 29 other proteins were 381

also specifically enriched in the purification of Raa7-TAP (Table 1). Amongst these, there 382

were three proteins of the small subunit of the plastid ribosome. This could indicate that there 383

is a coupling between splicing and translation. Whether there is a direct interaction of the 384

psaA trans-splicing complex with the ribosome should be the subject of future investigation.

385

Six protein components of the thylakoid photosynthetic electron transfer chain were also 386

found in the purification of Raa7-TAP that belong to three distinct complexes (Photosystem 387

II, cytochrome b6f and ATP synthase). It is not clear whether this is of physiological 388

significance, or whether it reflects contamination by very abundant proteins of the 389

chloroplast.

390

The complexity of psaA trans-splicing and the numerous factors involved are not an 391

idiosyncrasy of Chlamydomonas. The splicing machinery in the chloroplast of higher plants is 392

(14)

also similarly complex. In plants, most of the organellar introns belong to group II, do not 393

encode a maturase and depend for their splicing on a large cohort of nucleus-encoded 394

factors (Brown et al. 2014, de Longevialle et al. 2010a). Many of these factors have been 395

identified in Arabidopsis thaliana and in Zea mays (reviewed by (Germain et al. 2013, Jacobs 396

and Kück 2011). Some splicing factors are required for only one specific transcript. For 397

example, CRS1 in Z. mays and its orthologue AtCRS1 in A. thaliana are specifically involved 398

in splicing of atpF (Till et al. 2001). They contain typical CRM (Chloroplast RNA splicing and 399

ribosome Maturation) RNA-binding domains. Likewise, PPR4 is a PPR (pentatricopeptide 400

repeat) protein that acts specifically in the trans-splicing of rps12 in Z. mays (Schmitz- 401

Linneweber et al. 2006). Other PPR proteins, HCF152, PPR5, OTP51 and OPT70, are 402

required for the splicing of petB, trnG ycf3-2 and rpoC1 respectively in A. thaliana (Beick et 403

al. 2008, Chateigner-Boutin et al. 2011, de Longevialle et al. 2008, Meierhoff et al. 2003). On 404

the other hand, some splicing factors act on a larger set of introns (reviewed by: de 405

Longevialle, et al. 2010a, Germain, et al. 2013). For example the CRM proteins CAF1 and 406

CAF2 are part of a complex which also contains CRS2 and which is required for splicing of at 407

least nine introns in Z. mays (Barkan et al. 2007).

408

Some proteins involved in chloroplast splicing show similarity to proteins that contain 409

domains required for other functions. For example, Raa2 is related to pseudouridine 410

synthases (Balczun, et al. 2005, Perron, et al. 1999). Likewise in maize, CRS1 is related to 411

amino-acyl tRNA hydrolases, and RNC1 to RNase III (Jenkins and Barkan 2001, Watkins et 412

al. 2007). However, the enzymatic activities of these domains are not necessary for splicing, 413

and in some cases have been lost. These observations suggest that preexisting proteins that 414

have functions in RNA metabolism have been recruited for a new task in chloroplast RNA 415

splicing. This can be interpreted in the framework of the theory of constructive neutral 416

evolution (CNE), where the complexity of chloroplast RNA maturation would in part be a 417

consequence of mutations that arose in the chloroplast and that were compensated by pre- 418

existing nuclear suppressors, such as RNA-binding proteins. (Covello and Gray 1993, 419

Lefebvre-Legendre, et al. 2014, Lukes et al. 2011, Lynch 2007, Lynch et al. 2011). In this 420

perspective the evolutionary origin of Raa7 remains enigmatic, as it does not share 421

detectable sequence similarity with other known proteins.

422

The group II introns are proposed to be the ancestors of the nuclear spliceosomal introns 423

with which they share mechanistic and structural resemblances. The case of the split intron 1 424

of psaA in Chlamydomonas is particularly interesting in this respect, since it includes the tscA 425

RNA which does not contain any exon sequence, but is required for the assembly and trans- 426

splicing of the split group II intron (Goldschmidt-Clermont, et al. 1991, Sharp 1991). The 427

group II trans-splicing complexes of psaA are thus reminiscent of the nuclear spliceosome, 428

(15)

which is constituted of trans-acting snRNAs and numerous proteins that form transient, 429

dynamic assemblies.

430 431

(16)

EXPERIMENTAL PROCEDURES 432

Strains and Media 433

The Chlamydomonas reinhardtii strains were grown in Tris-acetate-phosphate 434

medium (TAP) or in high salt minimal medium (HSM) (Rochaix et al., 1988) to densities of 1–

435

2 × 106 cells mL-1 in the dark or under fluorescent lights (low light: 60 μE m-2 s-1; dim light: 6 436

μE m-2 s-1) at 25 °C. For growth tests, 10 μl of cell culture at 2 × 106 cells mL-1 were spotted 437

on TAP or HSM agar plates and grown under 6 or 60 μE m-2 s-1 light as indicated. Where 438

necessary, the TAP medium was supplemented with 100 μg mL-1 spectinomycin (Sigma 439

Aldrich).

440

Genetic analysis 441

Crosses were performed using standard protocols (Harris, 1989). The L121G mutant 442

(Girard, et al. 1980) was back-crossed 3 times to the wild type, leading to the strain called 443

raa7. For linkage analysis, the raa7 mutant was crossed with the interfertile species 444

Chlamydomonas grossii (S1-D2) (Rymarquis et al. 2005). For nuclear transformation, the 445

raa7 mutant was crossed to the cw15 strain to give raa7;cw15. The four raa7 revertant 446

strains (raa7-rev/1-4) appeared spontaneously after plating on HSM medium in 60 μE m-2 s-1 447

light.

448

Chlamydomonas cells were transformed by Helium-gun bombardment with the 449

chloroplast transformation vector psaA-Δi (pOS200) (Lefebvre-Legendre et al., 2014).

450

Transformants were selected on TAP plates supplemented with 100 μg/mL spectinomycin.

451

under illumination (60 μE m-2 s-1). Nuclear transformations were achieved using the glass 452

beads / vortex protocol (Kindle 1990) and transformants were selected on HSM plates for 453

photoautotrophic growth (60 μE m-2 s-1).

454

Cloning of the RAA7 gene 455

Raa7 was crossed to S1-D2 and progeny were dissected and tested for lack of 456

photoautotrophic growth by plating on HSM plates and otherwise maintained on TAP plates.

457

One mutant progeny from each of 57 tetrads was used to create the mapping population.

458

The polymorphic markers were based on PCR reactions performed on total DNA extracts 459

(Cao et al. 2009) using primers listed in http://chlamycollection.org/mapping-kit/ and 460

Rymarquis et al. (2005) (Table S1). Markers were scored for the percent of progeny that 461

contained the marker allele from the raa7 parent (Figure 3A). Two BAC clones, K11 and 462

2F17 covering this region were used to transform the raa7;cw15 mutant (Figure 3B). Internal 463

deletions of 3K11 were obtained by MluI or AscI digestion followed by religation. An ordered 464

(17)

cosmid library (Depege, et al. 2003, Purton and Rochaix 1994) was screened by PCR using 465

the primers 13Rev2 and 13For2, or 13Rev3 and 13For3, leading to the identification of 466

cosmid 14C2 (Table S2). The 7.3 kb SpeI fragment from 14C2 containing the RAA7 gene 467

including the 5’UTR (0.4 kb upstream the ATG start codon) was subcloned in Bluescript KS 468

(+) to yield plasmid pL39, which was sequenced by Fasteris (Geneva).

469

In order to determine the exon organization of the RAA7 gene and the nature of the 470

mutation in the raa7 mutant, total RNA was extracted from raa7, revertant raa7-rev/1, and 471

the wild type and subjected to RT-PCR with random hexamer primers following the 472

instructions of the manufaturer (Takara, PrimeScript First-Strand cDNA Synthesis) to obtain 473

four overlapping cDNA fragments using the following oligonucleotides : ACC Rev8 + ACC 474

For4 ; ACC Rev9 + ACC For6 ; ACC Rev5 + ACC For8 ; ACC 13ATG bis + ACC BAC3 Rev 475

(Table S2). After agarose gel electrophoresis, the PCR fragments were purified and 476

sequenced (Fasteris, Geneva). The HA-tagged midigene RAA7-HA (pL42), was constructed 477

as follows. A synthetic 2.1 kb MfeI-EcoRI fragment (Genscript) corresponding to the 3’ part of 478

the cDNA (from amino acid 613) and containing a triple HA epitope fragment inserted 479

between XhoI sites just before the stop codon, was used to replace the MfeI-EcoRI fragment 480

of pL39, yielding plasmid RAA7-HA (pL42).

481

The TAP-tagged Raa7 gene Raa7-TAP (pL52), was constructed as follows. The 482

cTAP sequence was amplified from pCM10 (Jacobs et al. 2013) with the oligonucleotides 483

TAP tag For and TAP tag Rev (Table S2) and used to replace the XhoI fragment of Raa7-HA 484

(pL42) to yield the plasmid Raa7-TAP (pL52) . 485

To construct the Raa7 plasmids for yeast two-hybrid experiments, a cDNA fragment 486

coding for amino acids 42-432 was amplified by RT-PCR using the oligonucleotides ACC 487

DH1 For and ACC DH1 Rev (Table S2). This fragment was digested with EcoRI and BamHI, 488

and inserted into pGBKT7; and pGADT7. The yeast two-hybrid vectors pGBKT7 and 489

pGADT7, containing the cDNA of Raa1, were described in Jacobs et al. (2013). An EcoRI- 490

PstI fragment of the cDNA of RAA2 corresponding to the amino acids 15-410, was inserted 491

into pGBKT7 to yield pL55 and into pGADc2 to yield pGAD-Raa2 (kindly provided by Dr. Karl 492

Perron).

493

RNA analysis 494

Total RNA was prepared as described previously (Lefebvre-Legendre, et al. 2015) 495

and analysed by formaldehyde agarose gel electrophoresis, transfer to Nylon membranes 496

and hybridization using probes labelled with 32P-dATP by random priming (Ausubel et al.

497

1998).

498

(18)

Protein analysis 499

Total proteins from Chlamydomonas cells (5 mL, 2 × 106 cells mL-1) were prepared as 500

described previously (Lefebvre-Legendre, et al. 2015) and analysed by SDS-PAGE (15%, 501

12% or 6% acrylamide) and immunoblotting. Labelling of the membranes with antisera 502

against PsaA (a gift of Kevin Redding), D1, PRK, anti-cytf , psaF, psaD, rbcL, LHCI (p17.2), 503

LHCII (CP29) (gifts of Jean-David Rochaix), RPL37 (Ramundo et al. 2013), anti-CF1 (gift of 504

Sabeeha Merchant) or the monoclonal antibody HA-11 (Covance) was carried out at room 505

temperature in 1× TBS (50 mM Tris–HCl pH 7.6, 150 mM NaCl), 0.1% Tween 20 and 5% w/v 506

nonfat powder milk. After washing the membranes, the antibodies were revealed with a 507

peroxidase-linked secondary antibody (Promega) and visualized by enhanced 508

chemiluminescence.

509

Cell fractionation 510

Cells were lyzed with a nebulizer and chloroplasts were purified on Percoll gradients 511

as described previously (Rivier, et al. 2001) in the presence of 0.1 mM of AEBSF.

512

Mitochondria were purified according to (Cardol et al. 2002).

513

Fractionation of membrane and soluble proteins, was performed as described by 514

(Boulouis et al. 2011) with the modification that the supernatant from the French press lysate 515

(1 mL) was layered on top of two layers of sucrose (1 mL of 1.5 M and 1 mL of 0.5 M). After 516

centrifugation, the soluble fraction was the supernatant at the top of the tube and the 517

membrane fraction was at the interface of the two sucrose layers.

518

Sucrose gradient sedimentation analysis was performed with the supernatant 519

corresponding to soluble proteins, described above. The samples (1 mL) were loaded on 520

sucrose gradients (10 mL; 5-45% sucrose, 20 mM HEPES (pH 7.2), 50 mM KCl, 10 mM 521

MgCl2). After centrifugation at 37 000 rpm for 18h in the SW40 rotor (Beckman), 11 fractions 522

(1 mL) were collected from the bottom by puncturing the tube. Sedimentation was calibrated 523

with the Gel Filtration Calibration Kit HMW (GE28-4038-42 SIGMA).

524 525

Yeast two-hybrid analysis 526

S. cerevisiae strains PJ69-4a and PJ69-4α were transformed with the different 527

versions of pGADT7 and pGBKT7 respectively (Clontech). Cells were grown overnight in 528

complete YPGA medium. Cells were diluted in 10 ml of complete medium at a final optical 529

density at 650 nm of 0.2 and incubated 5 hours at 28°C. Cells were centrifuged and washed 530

(19)

first with 1 mL of water and then with solution A (Lithium Acetate 100 mM, TE 1X) and 531

resuspended in 100 μL of solution A. 50 μL of cells were mixed to 50 μg of salmon sperm 532

DNA, 1 μg of plasmid DNA and 300 μL of solution B (PEG-4000 40%, Acetate 100 mM, TE 533

1X) and incubated 30 minutes at 28°C with shaking and then at 42°C for 15 min. Cells were 534

harvested by centrifugation, washed once with 1 mL water, resuspended in 100 μL water, 535

and plated on minimal medium lacking leucine for pGADT7 transformations and lacking 536

tryptophan for pGBKT7’s transformations. The transformed strains were mated to obtain 537

diploid cells containing two plasmids. For this, the transformed strains were separately grown 538

overnight in selective medium and 5 µl of the first transformant were spotted on minimal 539

medium lacking leucine and tryptophan. Once the spots were dry, 5 µl of the second 540

transformant were spotted over the first transformant. After an incubation of 3-5 days at 541

30°C, diploids were subcloned on double selective medium. Two independent diploid clones 542

were then spotted onto quadruple selective minimal medium lacking leucine, tryptophan, 543

adenine and histidine.

544 545

TAP purification of Raa7-interacting proteins 546

For solubilisation of chloroplast membranes, pelleted cells were resuspended in lysis buffer 547

(100 mM Tris, 150 mM NaCl, pH 8.0) containing 0.5 % n-Dodecyl β-D-maltoside and 548

protease inhibitors (Protease Inhibitor Mixture VI, Calbiochem, Germany). Cells were lysed 549

by sonication (4 times 60 s, 70-90% power ) and incubated on ice for 20 min. After 550

centrifugation (25.000xg, 30 min, 4°C), the supernatant was used for TAP purification as 551

described previously (Jacobs et al. 2013, Bloemendal et al. 2012).

552 553

LC-MS/MS analysis 554

For the Raa7::TAP replicate P1, sample preparation and MudPIT-MS/MS analyses were 555

performed as described previously (Jacobs, et al. 2013). Replicates P2 and P3 were 556

processed as follows. All chemicals for ultra-pure HPLC solvents such as, formic acid (FA), 557

trifluoroacetic acid (TFA) and acetonitrile (ACN) were obtained from Biosolve, Valkenswaard, 558

the Netherlands. Generated tryptic peptides were acidified with TFA to a final concentration 559

of 0.5% (v/v). Desalting was conducted with 10 µL C18 tips (OMIX, Varian) according to the 560

manufacturer’s protocol and the eluates were subsequently dried under vacuum. To each 561

sample, 15 µL of 0.1% (v/v) TFA was added and only 1/3rd fraction was used for the LC- 562

MS/MS analysis. Briefly, peptides were separated on an Ultimate 3000 system (Thermo 563

(20)

Scientific) with self-packed columns filled with Kinetex reversed phase C18 material (2.6 µm, 564

100 Å) coupled to LTQ Velos Orbitrap Pro mass spectrometer (Thermo Scientific). Peptide 565

solutions were preconcentrated on the trapping column 100 µm x 2 cm for 16 min using 0.1%

566

(v/v) TFA at a flow rate of 6 µL/min followed by separation on the main column 75 µm x 23 567

cm using a binary gradient (A: 0.1% formic acid (v/v); B: 0.1% formic acid (v/v), 84%

568

acetonitrile (v/v)) ranging from 3-42% B in 90 min at a flow rate of 250 nL/min. MS survey 569

scans were acquired in the Orbitrap from m/z 300 to 2000 at a resolution of 60,000 using the 570

polysiloxane ion at m/z 371.101236 as lock mass (Olsen et al. 2005). The ten most intense 571

signals were subjected to collision induced dissociation (CID) in the ion trap taking into 572

account a dynamic exclusion of 20 s. CID spectra were acquired with a normalized collision 573

energy (NCE) of 35%, a default charge state of 2 and an activation time of 30 ms. AGC 574

target values were set to 106 for Orbitrap MS and 104 for ion trap MSn scans.

575

Data evaluation was carried out with Proteome Discoverer 1.4 (Thermo Scientific) and the 576

MS raw data was searched against C. reinhardtii database

577

409_Creinhardtii_ed_Chlamydomonas reinhardtii_v5.5, which comprises 19,526 entries from 578

the nuclear genome downloaded on 18.09.14 and 69 entries from the C. reinhardtii 579

chloroplast genome using two search algorithms Mascot 2.4 (Matrix Science) and 580

SEQUEST. The following search settings were applied: trypsin as enzyme, two missed 581

cleavage sites allowed and oxidation of methionine as dynamic modification. MS and MS/MS 582

tolerances were set to 10 ppm and 0.5 Da respectively, PSMs with an FDR < 1 % (high 583

confidence Target/Decoy PSM Validator setting), and search engine rank 1 were selected.

584

Proteins identified with at least two different peptides were considered for analyses.

585

ACKNOWLEGEMENTS 586

We thank Jean-David Rochaix, Kevin Redding and Sabeeha Merchant for the gift of antisera, 587

Sylvain Lemeille and Gregory Theiler for help with bioinformatics analysis, Nicolas Roggli for 588

preparing the figures, Christina Marx and Jessica Jacobs for their help with some of the 589

experiments. This work was supported by the University of Geneva, the Swiss National Fund 590

(grants 31003A_146300 and 3100A0-117712), and a grant by the Deutsche 591

Forschungsgemeinschaft (Bonn Bad-Godesberg, Germany, UK (KU 517/13-1)). LK and AS 592

acknowledge the support by the Ministerium für Innovation, Wissenschaft und Forschung des 593

Landes Nordrhein-Westfalen.

594 595

(21)

SUPPORTING INFORMATION 596

Supplementary Table S1 597

Polymorphic markers used for mapping of RAA7.

598 599

Supplementary Table S2 600

Oligonucleotides used in this work.

601 602

Supplementary Figure S1 603

Sequence of Raa7 wild-type and mutants 604

605

Supplementary Figure S2.

606

Complementation of raa7 mutant with Raa7-HA 607

608

Supplementary Figure S3.

609

Expression of the RAA7-TAP construct 610

611

(22)

References 612

Ausubel, F.A., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A. and Struhl, K. (1998) 613 Current Protocols in Molecular Biology New York, NY: John Wiley & sons.

614 Balczun, C., Bunse, A., Hahn, D., Bennoun, P., Nickelsen, J. and Kück, U. (2005) Two adjacent nuclear 615 genes are required for functional complementation of a chloroplast trans-splicing mutant 616 from Chlamydomonas reinhardtii. Plant J, 43, 636-648.

617 Balczun, C., Bunse, A., Schwarz, C., Piotrowski, M. and Kück, U. (2006) Chloroplast heat shock 618 protein Cpn60 from Chlamydomonas reinhardtii exhibits a novel function as a group II intron- 619 specific RNA-binding protein. FEBS Lett, 580, 4527-4532.

620 Barkan, A. (2011) Expression of plastid genes: organelle-specific elaborations on a prokaryotic 621 scaffold. Plant Physiol, 155, 1520-1532.

622

Barkan, A., Klipcan, L., Ostersetzer, O., Kawamura, T., Asakura, Y. and Watkins, K.P. (2007) The 623 CRM domain: an RNA binding module derived from an ancient ribosome-associated protein.

624 RNA, 13, 55-64.

625 Beick, S., Schmitz-Linneweber, C., Williams-Carrier, R., Jensen, B. and Barkan, A. (2008) The 626 pentatricopeptide repeat protein PPR5 stabilizes a specific tRNA precursor in maize 627 chloroplasts. Mol Cell Biol, 28, 5337-5347.

628 Boulouis, A., Drapier, D., Razafimanantsoa, H., Wostrikoff, K., Tourasse, N.J., Pascal, K., Girard- 629 Bascou, J., Vallon, O., Wollman, F.A. and Choquet, Y. (2015) Spontaneous Dominant 630 Mutations in Chlamydomonas Highlight Ongoing Evolution by Gene Diversification. Plant Cell.

631 Boulouis, A., Raynaud, C., Bujaldon, S., Aznar, A., Wollman, F.A. and Choquet, Y. (2011) The 632 nucleus-encoded trans-acting factor MCA1 plays a critical role in the regulation of 633 cytochrome f synthesis in Chlamydomonas chloroplasts. Plant Cell, 23, 333-349.

634 Brown, G.G., des Francs-Small, C.C. and Ostersetzer-Biran, O. (2014) Group II intron splicing factors 635 in plant mitochondria. Frontiers in Plant Science, 5.35, 1-13.

636 Cao, M., Fu, Y., Guo, Y. and Pan, J. (2009) Chlamydomonas (Chlorophyceae) colony PCR.

637 Protoplasma, 235, 107-110.

638 Cardol, P., Matagne, R.F. and Remacle, C. (2002) Impact of mutations affecting ND mitochondria- 639 encoded subunits on the activity and assembly of complex I in Chlamydomonas. Implication 640 for the structural organization of the enzyme. J Mol Biol, 319, 1211-1221.

641 Chateigner-Boutin, A.L., des Francs-Small, C.C., Delannoy, E., Kahlau, S., Tanz, S.K., de Longevialle, 642 A.F., Fujii, S. and Small, I. (2011) OTP70 is a pentatricopeptide repeat protein of the E 643 subgroup involved in splicing of the plastid transcript rpoC1. Plant J, 65, 532-542.

644 Choquet, Y., Goldschmidt-Clermont, M., Girard-Bascou, J., Kück, U., Bennoun, P. and Rochaix, J.D.

645 (1988) Mutant phenotypes support a trans-splicing mechanism for the expression of the 646 tripartite psaA gene in the C. reinhardtii chloroplast. Cell, 52, 903-913.

647 Coletta, A., Pinney, J.W., Solis, D.Y., Marsh, J., Pettifer, S.R. and Attwood, T.K. (2010) Low- 648 complexity regions within protein sequences have position-dependent roles. BMC systems 649 biology, 4, 43.

650 Covello, P.S. and Gray, M.W. (1993) On the evolution of RNA editing. Trends Genet, 9, 265-268.

651 de Longevialle, A.F., Hendrickson, L., Taylor, N.L., Delannoy, E., Lurin, C., Badger, M., Millar, A.H.

652 and Small, I. (2008) The pentatricopeptide repeat gene OTP51 with two LAGLIDADG motifs is 653 required for the cis-splicing of plastid ycf3 intron 2 in Arabidopsis thaliana. Plant J, 56, 157-

654 168.

655 de Longevialle, A.F., Small, I.D. and Lurin, C. (2010a) Nuclearly encoded splicing factors implicated in 656 RNA splicing in higher plant organelles. Molecular plant, 3, 691-705.

657 de Longevialle, A.F., Small, I.D. and Lurin, C. (2010b) Nuclearly encoded splicing factors implicated in 658 RNA splicing in higher plant organelles. Mol Plant, 3, 691-705.

659 Depege, N., Bellafiore, S. and Rochaix, J.D. (2003) Role of chloroplast protein kinase Stt7 in LHCII 660 phosphorylation and state transition in Chlamydomonas. Science, 299, 1572-1575.

661

(23)

Emanuelsson, O., Brunak, S., von Heijne, G. and Nielsen, H. (2007) Locating proteins in the cell using 662 TargetP, SignalP and related tools. Nat Protoc, 2, 953-971.

663 Emanuelsson, O., Nielsen, H. and Von Heijne, G. (1999) ChloroP, a neural network-based method for 664 predicting chloroplast transit peptides and their cleavage sites. Protein Science, 8, 978-984.

665 Germain, A., Hotto, A.M., Barkan, A. and Stern, D.B. (2013) RNA processing and decay in plastids.

666 Wiley interdisciplinary reviews. RNA, 4, 295-316.

667

Girard, J., Chua, N.H., Bennoun, P., Schmidt, G. and Delosme, M. (1980) Studies on mutants 668 deficient in the photosystem I reaction centers in Chlamydomonas reinhardtii. Curr Genet, 2,

669 215-221.

670 Glanz, S., Bunse, A., Wimbert, A., Balczun, C. and Kück, U. (2006) A nucleosome assembly protein- 671 like polypeptide binds to chloroplast group II intron RNA in Chlamydomonas reinhardtii.

672 Nucleic Acids Res, 34, 5337-5351.

673 Glanz, S., Jacobs, J., Kock, V., Mishra, A. and Kück, U. (2012) Raa4 is a trans-splicing factor that 674 specifically binds chloroplast tscA intron RNA. Plant J, 69, 421-431.

675 Glanz, S. and Kück, U. (2009) Trans-splicing of organelle introns - a detour to continuous RNAs.

676 Bioessays, 31, 921-934.

677 Goldschmidt-Clermont, M. (1991) Transgenic expression of aminoglycoside adenine transferase in 678 the chloroplast: a selectable marker of site-directed transformation of chlamydomonas.

679 Nucleic Acids Res, 19, 4083-4089.

680 Goldschmidt-Clermont, M., Choquet, Y., Girard-Bascou, J., Michel, F., Schirmer-Rahire, M. and 681 Rochaix, J.D. (1991) A small chloroplast RNA may be required for trans-splicing in 682 Chlamydomonas reinhardtii. Cell, 65, 135-143.

683 Goldschmidt-Clermont, M., Girard-Bascou, J., Choquet, Y. and Rochaix, J.D. (1990) Trans-splicing 684 mutants of Chlamydomonas reinhardtii. Mol Gen Genet, 223, 417-425.

685 Harris, E.H. (2009) The genus Chlamydomonas. In The Chlamydomonas sourcebook. Introduction to 686 Chlamydomonas and its laboratory use.

687

. Oxford, UK.: Academic Press, pp. 1-18.

688 Herrin, D.L. and Schmidt, G.W. (1988) trans-splicing of transcripts for the chloroplast psaA1 gene. In 689 vivo requirement for nuclear gene products. J Biol Chem, 263, 14601-14604.

690 Jacobs, J. and Kück, U. (2011) Function of chloroplast RNA-binding proteins. Cellular and molecular 691 life sciences : CMLS, 68, 735-748.

692 Jacobs, J., Marx, C., Kock, V., Reifschneider, O., Franzel, B., Krisp, C., Wolters, D. and Kück, U. (2013) 693 Identification of a chloroplast ribonucleoprotein complex containing trans-splicing factors, 694 intron RNA, and novel components. Molecular & cellular proteomics : MCP, 12, 1912-1925.

695 Jenkins, B.D. and Barkan, A. (2001) Recruitment of a peptidyl-tRNA hydrolase as a facilitator of group 696 II intron splicing in chloroplasts. EMBO J, 20, 872-879.

697 Kindle, K.L. (1990) High-frequency nuclear transformation of Chlamydomonas reinhardtii. Proc Natl 698 Acad Sci U S A, 87, 1228-1232.

699 Kück, U., Choquet, Y., Schneider, M., Dron, M. and Bennoun, P. (1987) Structural and transcription 700 analysis of the two homolgous genes for the P700 chlorophyll a-apoprotein in 701 Chlamydomonas reinhardtii: evidence for in vivo trans-splicing. EMBO J, 6, 2185-2195.

702 Lefebvre-Legendre, L., Choquet, Y., Kuras, R., Loubery, S., Douchi, D. and Goldschmidt-Clermont, 703 M. (2015) A nucleus-encoded helical-repeat protein which is regulated by iron availability 704 controls chloroplast psaA mRNA expression in Chlamydomonas. Plant Physiol, 167, 1527-

705 1540.

706 Lefebvre-Legendre, L., Merendino, L., Rivier, C. and Goldschmidt-Clermont, M. (2014) On the 707 Complexity of Chloroplast RNA Metabolism: psaA Trans-splicing Can be Bypassed in 708 Chlamydomonas. Mol Biol Evol, 31, 2697-2707.

709 Lefebvre, P.A. and Silflow, C.D. (1999) Chlamydomonas: the cell and its genomes. Genetics, 151, 9-

710 14.

711

(24)

Lukes, J., Archibald, J.M., Keeling, P.J., Doolittle, W.F. and Gray, M.W. (2011) How a neutral 712 evolutionary ratchet can build cellular complexity. IUBMB Life, 63, 528-537.

713 Lynch, M. (2007) The origins of genome architecture Sunderland, Mass., USA: Sinauer Associates,

714 Inc.

715 Lynch, M., Bobay, L.M., Catania, F., Gout, J.F. and Rho, M. (2011) The repatterning of eukaryotic 716 genomes by random genetic drift. Annual review of genomics and human genetics, 12, 347- 717

366.

718 Marx, C., Wunsch, C. and Kück, U. (2015) The Octatricopeptide Repeat Protein Raa8 Is Required for 719 Chloroplast trans Splicing. Eukaryot Cell, 14, 998-1005.

720 Meierhoff, K., Felder, S., Nakamura, T., Bechtold, N. and Schuster, G. (2003) HCF152, an Arabidopsis 721 RNA binding pentatricopeptide repeat protein involved in the processing of chloroplast psbB- 722 psbT-psbH-petB-petD RNAs. Plant Cell, 15, 1480-1495.

723 Merendino, L., Perron, K., Rahire, M., Howald, I., Rochaix, J.D. and Goldschmidt-Clermont, M.

724

(2006) A novel multifunctional factor involved in trans-splicing of chloroplast introns in 725 Chlamydomonas. Nucleic Acids Res, 34, 262-274.

726 Olsen, J.V., de Godoy, L.M., Li, G., Macek, B., Mortensen, P., Pesch, R., Makarov, A., Lange, O., 727 Horning, S. and Mann, M. (2005) Parts per million mass accuracy on an Orbitrap mass 728 spectrometer via lock mass injection into a C-trap. Molecular & cellular proteomics : MCP, 4,

729 2010-2021.

730 Perron, K., Goldschmidt-Clermont, M. and Rochaix, J.D. (1999) A factor related to pseudouridine 731 synthases is required for chloroplast group II intron trans-splicing in Chlamydomonas 732 reinhardtii. EMBO J, 18, 6481-6490.

733 Perron, K., Goldschmidt-Clermont, M. and Rochaix, J.D. (2004) A multiprotein complex involved in 734 chloroplast group II intron splicing. RNA, 10, 704-710.

735 Purton, S. and Rochaix, J.D. (1994) Complementation of a Chlamydomonas reinhardtii mutant using 736 a genomic cosmid library. Plant Mol Biol, 24, 533-537.

737 Rahire, M., Laroche, F., Cerutti, L. and Rochaix, J.D. (2012) Identification of an OPR protein involved 738 in the translation initiation of the PsaB subunit of photosystem I. Plant J, 72, 652-661.

739 Ramundo, S., Rahire, M., Schaad, O. and Rochaix, J.D. (2013) Repression of essential chloroplast 740 genes reveals new signaling pathways and regulatory feedback loops in chlamydomonas.

741 Plant Cell, 25, 167-186.

742 Rigaut, G., Shevchenko, A., Rutz, B., Wilm, M., Mann, M. and Seraphin, B. (1999) A generic protein 743 purification method for protein complex characterization and proteome exploration. Nature 744 biotechnology, 17, 1030-1032.

745 Rivier, C., Goldschmidt-Clermont, M. and Rochaix, J.D. (2001) Identification of an RNA-protein 746

complex involved in chloroplast group II intron trans-splicing in Chlamydomonas reinhardtii.

747 EMBO J, 20, 1765-1773.

748 Rymarquis, L.A., Handley, J.M., Thomas, M. and Stern, D.B. (2005) Beyond complementation. Map- 749 based cloning in Chlamydomonas reinhardtii. Plant Physiol, 137, 557-566.

750 Schmitz-Linneweber, C., Williams-Carrier, R.E., Williams-Voelker, P.M., Kroeger, T.S., Vichas, A. and 751 Barkan, A. (2006) A pentatricopeptide repeat protein facilitates the trans-splicing of the 752 maize chloroplast rps12 pre-mRNA. Plant Cell, 18, 2650-2663.

753 Sharp, P.A. (1991) "Five easy pieces". Science, 254, 663.

754 Stern, D.B., Goldschmidt-Clermont, M. and Hanson, M.R. (2010) Chloroplast RNA metabolism.

755 Annual review of plant biology, 61, 125-155.

756 Tardif, M., Atteia, A., Specht, M., Cogne, G., Rolland, N., Brugiere, S., Hippler, M., Ferro, M., Bruley, 757 C., Peltier, G., Vallon, O. and Cournac, L. (2012) PredAlgo: a new subcellular localization 758 prediction tool dedicated to green algae. Mol Biol Evol, 29, 3625-3639.

759 Till, B., Schmitz-Linneweber, C., Williams-Carrier, R. and Barkan, A. (2001) CRS1 is a novel group II 760 intron splicing factor that was derived from a domain of ancient origin. RNA, 7, 1227-1238.

761

(25)

Watkins, K.P., Kroeger, T.S., Cooke, A.M., Williams-Carrier, R.E., Friso, G., Belcher, S.E., van Wijk, 762 K.J. and Barkan, A. (2007) A ribonuclease III domain protein functions in group II intron 763 splicing in maize chloroplasts. Plant Cell, 19, 2606-2623.

764 765 766

(26)

Table 1 767

TAP purification of Raa7 and mass spectrometry analysis of trans-splicing factors copurified with Raa7 768

Three independent TAP experiments were performed using Raa7 as bait (P1-P3). Non-specific interactors were identified by control 769

purifications using three RST-1 cultures and one non-tagged wild type culture for TAP experiments (C1-C4). Proteins were analyzed with the 770

tools PredAlgo (PA), TargetP (TP), and ChloroP (CP) for the prediction of subcellular localization. Proteins which were found with at least two 771

peptides in all three Raa7-TAP experiments and exhibit a putative mitochondrial/chloroplast transit peptide were considered as specific (32 772

proteins). Amongst 32 proteins (Table S3), the trans-splicing factors Raa1, Raa2, and Raa7 were detected (Columns “Accession” and 773

“Protein”). The number of unique peptides is given in column “Σ ≠ peptides”. Column “kDa” represents the predicted molecular mass of 774

corresponding proteins in kDa.

775

Raa7-TAP Σ ≠ peptides

Accession Protein kDa PA TP CP P1 P2 P3

Trans-splicing factors

Cre03.g201103 Raa7 (bait) 130 M M C 8 12 14

Cre16.g665750 Raa2, pseudouridine synthase 45 M M - 4 4 4

Cre09.g394150 Raa1 130 M C C 17 19 22

Plastid ribosomal proteins

Cre12.g494750 Plastid ribosomal protein S20 18 C C M 5 3 3

Cre02.g118950 Plastid ribosomal protein S17 12 C C M 3 5 3

Cre12.g494450 Plastid ribosomal protein S16 14 C C M 3 3 2

Not annotated

Cre17.g728850 no functional annotations 58 C M C 8 7 7

Cre17.g698750 no functional annotations 92 C M M 7 10 11

Cre10.g426450 no functional annotations 67 C C C 6 6 5

Références

Documents relatifs

Accumulation of MDA1 (Flag), ATPase α subunit (CF1α) and cytochrome f (loading control) proteins in the same strains is shown below. E) MDA1 and TDA1 are soluble proteins.

In the presence of D2 and in the absence of D1, CP47 translation is specifically decreased by the control by epistasy of synthesis (CES). D1 absence induces the inhibition of CP47

reinhardtii psbA, the chloroplast gene for the 32-kDa protein, has revealed a single T-A to G-C transversion in the DCMU-4 mutant that results in a deduced amino acid change of

In contrast to typical maturases whose coding regions include both exon and intron sequences, the chloroplast ribosomal intron open reading frame is contained entire-ly within

The mutants belong to 14 nuclear complementation groups and one chloroplast locus that are required for the assembly of psaA mRNA.. The chloroplast locus, tscA, is remote from any

We have characterized a Chlamydomonas reinhardtii nuclear mutant that is unable to accumulate transcripts covering not only psbB (encoding P5, the 47 kDa chloro- phyll-a

eugametos, the finding that the region downstream from the psaA exon 1 and the 5' end of the putative tscA RNA cannot fold into the typical struc- ture described

It appears that the PsbI polypeptide plays a role in the stability of photosystem II and possibly also in modulating electron transport or energy transfer in this complex.