DNA macroarray and real-time PCR analysis of two nuclear photosystem I mutants from Chlamydomonas reinhardtii reveal downregulation of Lhcb genes but different regulation of Lhca genes

Biochimica et Biophysica Acta 1732 (2005) 62 – 68 http://www.elsevier.com/locate/bba

DNA macroarray and real-time PCR analysis of two nuclear photosystem I mutants from Chlamydomonas reinhardtii reveal downregulation of Lhcb genes but different regulation of Lhca genes Carsten Balczun 1 , Astrid Bunse 1,2 , Minou Nowrousian, Alexandra Korbel, Stephanie Glanz, Ulrich Kück⁎ Lehrstuhl für Allgemeine und Molekulare Botanik, Ruhr-Universität Bochum, D-44780 Bochum, Germany Received 25 August 2005; received in revised form 4 November 2005; accepted 8 November 2005 Available online 20 December 2005

Abstract In photoautotrophic organisms, the expression of nuclear genes encoding plastid proteins is known to be regulated at various levels. In this study, we present the analysis of two non-photosynthetic mutants (CC1051 and TR72) from the unicellular green alga Chlamydomonas reinhardtii. Both mutant strains show a defect in the processing of chloroplast psaA mRNA, and therefore they are assumed to be defective in photosystem I (PSI) assembly. We have performed macroarray experiments with trans-splicing mutants CC1051 and TR72 in order to analyse putative pleiotropic effects of nuclear-located mutations leading to a non-functional PSI. To the best of our knowledge, this is the first example of Chlamydomonas cDNA macroarray analysis comparing the transcriptional regulation of nuclear genes in wild-type and photosystem I mutants. The macroarray results demonstrated a transcriptional downregulation of members of the Lhcb gene family more than 2-fold in both mutant strains. In addition, real-time RT-PCR experiments found a 4- to 16-fold reduction in transcript levels of several Lhca genes in TR72; whereas in CC1051, no significant change in transcript levels was observed. Taken together, our data suggest that a signal is transmitted from the chloroplast to the nucleus that serves to regulate the level of light harvesting polypeptides in the organelle. © 2005 Elsevier B.V. All rights reserved. Keywords: Chlamydomonas reinhardtii; Photosystem I mutant; Transcriptome analysis; LHC; light-harvesting complex; Chloroplast-to-nucleus signaling

1. Introduction The unicellular green alga Chlamydomonas reinhardtii is a valuable eukaryotic model system that enables the genetic regulation of photosynthesis to be studied due to the huge set of non-photosynthetic mutant strains that are available [1,2]. Chloroplast gene expression is highly dependent on nuclear factors influencing all steps of organellar gene expression [3,4]. Transcription, RNA processing as well as translation and assembly of the photosynthetic apparatus are controlled by nuclear-encoded factors that are imported into the chloroplast. ⁎ Corresponding author. Tel.: +49 234 3226212; fax: +49 234 3214184. E-mail address: [email protected] (U. Kück). 1 These authors contributed equally to this work. 2 Present address: QIAGEN GmbH, QIAGEN Straße1, D-40724 Hilden, Germany. 0167-4781/$ - see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.bbaexp.2005.11.006

Mutations in such nuclear genes often give rise to distinct phenotypes, and molecular analysis of some of these mutant strains has led to the identification of the molecular function of the corresponding genes [5]. Photosystem I mutants of Chlamydomonas are mostly affected in the trans-splicing process of the chloroplast psaA transcript that encodes a subunit of photosystem I and exhibits a unique genomic organisation [6]. The three exon sequences of the gene are scattered around the chloroplast genome and transcribed as separate precursor molecules that are then joined together to form the mature transcript [7,8]. The analysis of several mutants defective in trans-splicing has revealed the direct or indirect participation of at least 15 nuclear factors in this complex molecular process [9]. The assembly of the photosynthetic apparatus in land plants and green algae is a balanced process. In psaA mutants in Chlamydomonas, the lack of the PsaA polypeptide was shown

C. Balczun et al. / Biochimica et Biophysica Acta 1732 (2005) 62–68

not to influence the synthesis of PsaB, the second core polypeptide of photosystem I [9,10]. However, PsaA and PsaB assemble at an early step of PSI biogenesis, leading to the formation of the chlorophyll a–protein complex I (CPI) that binds most of the pigments and redox cofactors of PSI. Therefore, the absence of the PsaA polypeptide inhibits PSI biogenesis at an early stage [11]. Thus, mutants lacking a functional photosystem I are sensitive to high light intensities because of the generation of reactive oxygen species [12] and can therefore only be maintained under low light conditions. We have performed macroarray experiments with different nuclear trans-splicing mutants in order to investigate putative pleiotropic effects due to a defective photosystem I. A cDNA array representing 960 independent nuclear transcripts was generated to compare the transcriptome of the wild-type and the trans-splicing mutant strains TR72 and CC1051 [13–15]. The data from array hybridisations were verified by Northern hybridisations and demonstrated that in the photosystem I mutants, the Lhcb genes are transcriptionally downregulated. The investigation of gene expression of a subset of Lhca genes (Lhca1, Lhca2, Lhca3, Lhca7) by means of real-time RT-PCR revealed a completely different transcriptional expression profile for these transcripts in the different mutant strains. 2. Materials and methods 2.1. Strains and culture conditions For RNA isolation, the Chlamydomonas reinhardtii cell wall depleted, arginine auxotroph strain cw15arg− (Duke University, Durham, NC, USA), the trans-splicing mutants TR72 [14] and CC1051 [13] as well as the wild-type strain 137c [16] were cultivated in TAP medium [17] to a cell density of 2 × 106 cells per ml in continuous white dim light (50 μE/m2 s). When required, the medium was supplemented with 50 μg arginine per ml. The wild-type strain CC406 [18] grown to a cell density of 1–3 × 106 cells per ml was used for the generation of the cDNA library, which served as probe collection for the macroarray. RNA for library construction was isolated following the different culture conditions shown in Table 1 to ensure that the library consists of a wide variety of expressed sequences.

2.2. cDNA library construction Total RNA from Chlamydomonas strain CC406 was isolated as described before [8]. 4 mg of total RNA pooled from the different culture conditions (Table 1) were used for the isolation of poly(A) RNA with the PolyATtract® poly(A)RNA Isolation System II (Promega, Mannheim, Germany) for large scale poly (A) RNA isolation according to the manufacturer's protocol. Based on this poly (A) RNA preparation as template, the cDNA library was constructed by BD

63

Biosciences Clontech. After cDNA synthesis, EcoRI linker were ligated and the DNA fragments were cloned into the EcoRI restriction site of plasmid pGAD10 (BD Biosciences, Heidelberg, Germany).

2.3. DNA macroarray construction, target preparation, hybridisation and scanning In order to prepare the macroarray probe collection, 10 microtiter plates each containing 96 bacterial colonies, were picked and indexed with the picking device of the GeneTac™ G3 roboter (Genomic Solutions™, Cambridgeshire, UK) leading to a total number of 960 individual cDNA clones. The bacterial clones were cultivated in 1.1 ml LB medium, supplemented with 0.1 μg ampicillin per ml in deep well boxes for plasmid extraction and the plasmid DNA was isolated using QIAprep 96 Turbo Miniprep Kits (QIAGEN, Hilden, Germany) according to the manufacturer's recommendations with a Genesis Workstation 100 (Tecan, Crailsheim, Germany). The length and integrity of the cDNA inserts of the isolated plasmids were verified by restriction analysis and 96 cDNA inserts were sequenced in order to investigate the redundancy of the selected macroarray probe collection. The preparation of membrane-based macroarrays was done using the gridding device of the GeneTac™ G3 roboter. To increase precision, each cDNA was spotted twice onto Nytran+ membranes (Schleicher and Schuell, Dassel, Germany) in a 5 × 5 gridding scheme and each grid contained a duplicate of an α-tubulin control spot [19]. After spotting, the membrane-bound DNA was denaturated for 10 min in 0.5 M NaOH, 1.5 M NaCl, neutralised for 10 min in 1.5 M NaCl, 0.2 M Tris–HCl, pH 7.4 and washed in 2× SSC buffer containing 0.1% SDS. The DNA was subsequently cross-linked to the membranes (“Stratalinker”, Stratagene, Amsterdam, the Netherlands). For target preparation, 6 μg of DNaseI treated poly(A) RNA was used as starting material for each target and a mixture of 2 μg oligo d(T) and 2 μg random hexamer oligonucleotides was used as primers for reverse transcription with the Omniscript reverse transcriptase (QIAGEN, Hilden, Germany) according to the manufacturer's protocol. The targets were radioactively labelled with (α33P)dATP during reverse transcription. The labelled targets were purified with Nucleospin Colums (Ambion Inc., Austin, Texas, USA) following the manufacturer's recommendations and immediately used for hybridisation of macroarray membranes. Array membranes were prehybridised for 1 h at 50 °C in 20 ml 5× SSPE, 0.2% SDS, 5× Denhardt's solution, 50% formamide. The 33P-labelled targets were added and hybridisation was carried out over night at 50 °C. After washing with 5× SSPE, 0.2% SDS for 5 min at 50 °C, the membranes were dried and the signals were detected using Phospho-Imager plates and the Bio-Imager BAS 1500 scanner (Fujifilm, Düsseldorf, Germany) with the TINA software. Analysis of the tiff-files from arrays was done with AIDA (raytest, Straubenhardt, Germany). Predefined grids were manually adjusted to ensure optimal spot recognition. The resulting data files were further analysed in Excel (Microsoft): for final data analysis, data points were averaged from two replicates for each probe on each membrane array. To correct for differences between membranes or for uneven loss of samples during target preparation, normalisation was done by calculating the average signal intensity of all spots for each membrane within an experimental series and a normalisation factor was determined. After normalisation, the cut-off for regulation of the different transcripts was set from 2-fold to 1.6-fold change.

2.4. Preparation and analysis of RNA Table 1 Culture conditions used for cDNA library construction Strain

Medium

Culture condition

CC406 CC406 CC406 CC406 CC406

TAP TAPS TAP TAP-N HS

130 rpm; 23 130 rpm; 23 130 rpm; 23 130 rpm; 23 130 rpm; 23

°C; high light °C; high light °C; darkness °C; high light °C; high light

TAPS, TAP medium supplemented with 1% sorbitol; TAP-N, nitrogen depleted TAP medium [35]; HS, high salt medium [36].

Total RNA was isolated, size fractionated on denaturating formaldehyde agarose gels and transferred to Nytran+ membranes (Schleicher and Schuell, Dassel, Germany) as described elsewhere [15]. Radioactively labelled EcoRI DNA fragments derived from selected cDNA clones of the macroarray probe collection were used as probes in Northern hybridisation experiments.

2.5. Quantitative real-time PCR Quantitative real-time PCR was carried out as described previously [20] with the following modifications: Reverse transcription of the RNA sample was done with 400 U Superscript II reverse transcriptase (Invitrogen) and 0.33 mM dNTPs

64

C. Balczun et al. / Biochimica et Biophysica Acta 1732 (2005) 62–68

Table 2 Oligonucleotide primer pairs used in real-time PCR experiments Gene

Forward/reverse primer, each 5′ → 3′

Lhca1

ACCAACATCAACATGGGTGGC/ GCACCTTGTTGCCCTTGAAGA AACATCGACTTCCCCACCCTG/ TGAACTCCTCGGTGAAGCCC TGGTGTTGCCGGTATCCTGAT/ CTGCACAGCCAGCAGAGAGC GCTTGGTGGGAGGCTGGC/ AAGCCGGAAGTGCCGGTC CTCGCTTCGCTTTGACGGTG/ CGTGGTACGCCTTCTCGGC TGCCGACTAGGGATTGGCAG/ GTGGTGCCCTTCCGTCAATT

Lhca2 Lhca3 Lhca7 Tuba1 SSU rRNA

in a total volume of 30 μl for 2 h at 42 °C. Oligonucleotide primers used for realtime PCR are given in Table 2.

3. Results 3.1. Evaluation of an array system for the transcriptome analysis of C. reinhardtii The C. reinhardtii mutants TR72 and CC1051 belong to the class C trans-splicing mutants of the chloroplast psaA precursor RNAs. TR72 was derived by insertional mutagenesis of wildtype strain cw15arg− using plasmid pARG7.8ϕ3 [21] and carries a deletion of two adjacent nuclear genes, Rat1 and Rat2 [14]. Both genes function in the processing of chloroplast tscA RNA, a co-factor of psaA trans-splicing. Mutant CC1051, in which the nuclear factor Raa3 that is implicated in splicing of intron 1 of the psaA transcript is non-functional [13,22], was derived by conventional mutagenesis of wild-type strain 137c [16]. Due to these mutations, both mutant strains lack a functional photosystem I. We have designed a membrane-based macroarray with 960 independent cDNA clones from C. reinhardtii for the analysis

of putative pleiotropic effects of these nuclear mutations. The cDNA library was originally constructed for yeast hybrid system screens, and therefore the cDNA was cloned into plasmid pGAD10. In order to guarantee that a wide range of transcripts is represented in the library, the poly(A) RNA preparations for cDNA synthesis were performed using cells from different culture conditions (Table 1). In this study, both sample (trans-splicing mutants) and control (wild-types) targets were prepared by reverse transcription of poly(A) RNA in the presence of (α33P)dATP. Therefore, the different targets had to be hybridised to different membrane arrays. Consequently, the reproducibility of the membranes used was assessed by dividing a labelled wild-type target into two aliquots and separate hybridisation with two filters of a set of membranes. The scatter plot analysis of the Q values (= log2 [ratio normalised signal 1/normalised signal 2]) is shown in Fig. 1 and indicates a sufficient reproducibility of the membrane quality as all signals, except for four data points, did not exceed the threshold of background variation (dashed line in Fig. 1). To further ensure constant quality of the different biological samples, α-tubulin control spots in each grid were examined. The comparison of the α-tubulin signal intensities of the different biological samples of the wild-type and the mutant strains in terms of logarithmised mean values did not show significant variations (data not shown). Thus, these analyses confirm the reliability and reproducibility of the membranebased array system. 3.2. Transcriptional regulation of Lhcb genes in C. reinhardtii photosystem I mutants Each comparative array hybridisation experiment (control versus sample) was replicated three times for TR72 or twice for CC1051, and each series was run independently from starting material to hybridisation. Hybridisation was conducted under high stringency conditions, which might lead to lower overall signal intensity, but reduces the possibility of unspecific

Fig. 1. Reproducibility of membrane quality. For the analysis of reproducible membrane quality, a 33P-labelled wild-type target was divided into two aliquots and used for hybridisation of two membranes containing 960 independent cDNAs. The ratios of signal intensities are displayed as Q values. The dashed line indicates the threshold of 2-fold induction or repression of gene activity.

C. Balczun et al. / Biochimica et Biophysica Acta 1732 (2005) 62–68

hybridisation events. This is important for cDNA macroarray experiments, since unspecific hybridisation events cannot easily be distinguished. The variation in gene expression was monitored by means of scatter plot analysis of the logarithmised averaged and normalised hybridisation intensity of the 960 different cDNA probes. Fig. 2 presents a representative data set of a comparative analysis of the wild-type cw15arg- and mutant strain TR72. None of the genes on the array displayed upregulated transcript levels in mutant TR72 compared to the wild-type. However, 2-fold or higher reduction of transcription in the replicated experiments was detected for several probes on the membrane array, and including probes with very close to two-fold reduction (ratio ≤0.6), a total of 20 probes were identified as downregulated in mutant TR72. After sequence analysis of the respective probes, the number of downregulated transcripts was determined as eight due to a redundancy of the used cDNA probe collection. Similar results were obtained in array experiments with mutant CC1051 compared to its wildtype 137c. By comparing the DNA sequences of the identified probes with databases, the cDNAs were determined to be Lhcb sequences (Table 3; nomenclature is according to Elrad and Grossman [23]). To verify the observed downregulation of the different Lhcb transcripts in mutant TR72, Northern hybridisation experiments with radioactively labelled cDNA inserts of the respective array probes were performed. The results confirmed the decreased transcript accumulation in the nonphotosynthetic strain (Fig. 3), whereby the magnitude of measured downregulation in the Northern hybridisation experiments was about 2-fold higher than in the array experiments when compared to the normalised signal intensity of the autoradiograms (data not shown). As an exemplary probe, D2_8 (Lhcbm2) was also used in a Northern hybridisation with RNA from CC1051, in order to confirm the data for mutant TR72. As expected, analysis of the autoradiogram showed that the Lhcbm2 transcript in this mutant strain was also downregulated (data not shown).

Fig. 2. Variation in gene expression between wild-type cw15arg- and photosystem I mutant TR72. Comparison of the hybridisation signals (log2 values) of mutant TR72 versus wild-type for 960 nuclear-encoded sequences. Triangles represent outliers below the 0.6 cut-off ratio for repression in the mutant strain.

65

Table 3 Summary of downregulated Lhcb transcripts in two different photosystem I mutants of C. reinhardtii Clone identity

A8_1 D2_8 H2_10 E7_6 E6_3 F12_10 B11_3 F8_6 a b c

Gene a

Lhcbm1 Lhcbm2 Lhcbm3 Lhcbm5 Lhcbm6 Lhcbm8 Lhcb4 Lhcb5

Average ratio of transcript levels comparing mutant/ wild-type TR72 b

CC1051 c

0.518 0.465 0.533 0.601 0.581 0.504 0.473 0.501

0.519 0.436 0.523 – 0.581 0.433 0.568 0.563

Number of independently identified clones on the array 4 2 1 1 2 1 3 2

Gene denotations are according to [23]. Values are averages of three independent experiments. Values are averages of two independent experiments.

3.3. Regulation of the Lhca gene family members Macroarray analyses of photosystem I mutants has identified differential expression of Lhcb genes compared to the wild-type which indicates a regulation mechanism for the expression of photosystem II-associated LHCII complex genes. Lhca transcript regulation could not be detected by our macroarray experiments, most probably due to the lack of representative Lhca probes on the array. To examine expression of photosystem I-associated Lhca genes in the mutant strains, we conducted quantitative real-time PCR experiments. The nomenclature for Lhca genes and their corresponding sequences used in this work is according to the recommendation by Elrad and Grossman [23]. Wild-type (cw15arg- and 137c) and mutant strains (TR72 and CC1051) were grown in constant dim light

Fig. 3. Northern blot verification of regulation of Lhcb genes in photosystem I mutant TR72. For transcriptional analysis, total RNA (40 μg/lane) extracted from wild-type cw15 arg− and mutant TR72 was hybridised with the indicated probes. The hybridisation with the Rps18 probe (ribosomal polypeptide) was used as a loading control.

66

C. Balczun et al. / Biochimica et Biophysica Acta 1732 (2005) 62–68

Fig. 4. Expression of Lhca genes in mutants TR72 and CC1051. Real-time PCR results for different Lhca genes are given as logarithmic values of the ratio mutant versus wild-type (logarithm to the base 2 for mean of three independent experiments). The dashed lines display the cut-off for significant regulation. The expression value of α-tubulin (tuba) is given as an example for a not regulated gene.

(50 μE/m2 s) and total RNA was analysed by quantitative realtime PCR for precise quantification of mRNA abundance of the Lhca1, Lhca2, Lhca3 and Lhca7 genes. The transcript levels of all examined genes were significantly reduced in mutant TR72 compared to the wild-type. The highest decrease in expression was measured for Lhca1 that showed a 16-fold reduction of transcript abundance, whereas Lhca2, Lhca3 and Lhca7 were about 4-fold downregulated (Fig. 4). Expression of the αtubulin gene [19] was used as a non-regulated control gene in the wild-type and mutants (Fig. 4). In contrast to TR72, the examined Lhca transcripts are not regulated in mutant CC1051 compared to the wild-type 137c (Fig. 4), except for Lhca2 which seems to be similarly repressed in both mutant strains. 4. Discussion Global analysis of transcriptional profiles by means of highdensity DNA microarrays or membrane-based macroarrays has recently become a powerful tool to analyse how organisms modulate gene activity in mutant backgrounds or in response to environmental conditions. Whereas array technology is qualified to analyse transcriptional regulation of a multitude of genes simultaneously, Northern hybridisation and quantitative realtime PCR are gene specific and usually offer a higher sensitivity in comparison to array analyses. In this study, analyses of expression ratios using Northern blots also showed higher sensitivity compared to macroarray results. Recently, DNA arrays were performed to analyse gene regulation in the unicellular green alga Chlamydomonas reinhardtii. By applying this technique, transcriptional profiles of nuclear [24,25] as well as of chloroplast gene expression were generated [26,27]. We have used this technology to analyse how nuclear C. reinhardtii photosystem I mutants adjust nuclear gene expression in response to disturbed chloroplast function. To the best of our knowledge, this is the first example of Chlamydomonas cDNA macroarray analysis comparing the

transcriptional regulation of nuclear genes in wild-type and photosystem I mutants. The array elements were derived from cDNAs from a recombinant library from cells grown under different nutritional and light conditions. Therefore, the macroarray was likely to contain representative sequences of metabolism- and photosynthesis-related genes. Our control experiments have clearly confirmed the reliability and reproducibility of this membrane-based DNA macroarray. Moreover, DNA arrays are applicable for the analysis of individual members of gene families such as the Lhc genes since this technique is sufficiently specific to generate reliable results [28]. Recently, substantial studies were carried out using a cDNA microarray from Chlamydomonas which contains almost 2800 unique sequences [24]. Within this study, a genome-wide view of changes in mRNA abundance due to sulphur limitation of wild-type and the sac1 mutant that exhibits abnormal sulphate uptake was derived. Different generalisations were made by the authors from the data presented in that study; for instance, levels of transcripts for many genes substantially changed during sulphur deprivation. These differentially expressed genes comprise both genes involved in sulphur metabolism (specific response) as well as genes with other cellular function (general response). A striking finding of this study is the decline of transcript levels for genes encoding proteins involved in photosynthesis [24]. Such transcripts include those encoding polypeptides of photosystem I and II as well as LHC polypeptides. Zhang and co-workers [24] assumed that the organism reconstructs the photosynthetic apparatus to accommodate sulphur deprivation conditions and that this decrease in transcript levels is therefore a general response. In this study, the photosystem I mutants TR72 and CC1051 were grown in constant dim light in acetate containing TAP medium. Hence, the strains were equally provided with a carbon source and trace elements like sulphur. Therefore, the observed downregulation of Lhcb polypeptides-encoding transcripts in both mutant strains is more likely a specific response to the photosynthesis defect. Previous work has already shown that the Lhcb1 (or cabII-1) transcript levels together with transcript levels of different photosystem I subunits were decreased in mutant CC1051 [13]. Our more global analysis of transcript abundance in two different photosystem I mutant strains revealed the downregulation of a set of Lhcb transcripts. Moreover, quantitative real-time PCR experiments indicated that Lhca transcript levels are also decreased in mutant TR72. Lhc gene family members function in managing the absorption, distribution and dissipation of light energy [23,29,30]. They share sequence similarities and account for a huge fraction of the chlorophyll associated with the photosystems. Absorbed light energy by the light-harvesting complexes is transferred to the corresponding photosystems and subsequently electron transfer to further acceptors takes place. The proper docking of LHCI to photosystem I is essential because a missing connection would prevent the transfer of the excitation energy from LHCI to the reaction centre of photosystem I. Excess excitation energy can cause the formation of singlet oxygen (1O2) by energy transfer from excited triplet chlorophylls to

C. Balczun et al. / Biochimica et Biophysica Acta 1732 (2005) 62–68

ground stated triplet O2 [31,32]. These reactive oxygen species provoke photo-oxidative damage in the chloroplast. Prevention of photo-oxidative damage by these reactive oxygen species that especially affect photosystem II is therefore essential for cell survival. Thus, mechanisms that balance light energy input are an important means for cell fitness. Photosynthetic organisms are able to respond to photo-oxidative stress in different ways: On the one hand the size of the antenna of photosystem II can be reduced on the transcriptional level as it was observed in this study or on the other hand part of the antenna of photosystem II is transferred to photosystem I, a process of protein dynamic which is called state transition [33]. In TR72 and CC1051, photosystem I is missing, and thus docking of LHCII to photosystem I is impossible. Therefore, the protection mechanism of photo-oxidative damage in the photosystem I mutants is limited to the transcriptional level. The observed downregulation seems to be specific for genes encoding the antenna polypeptides of photosystem II. The nuclear psb1 gene which encodes the oxygen-evolving enhancer protein 1 [34] is also present on the macroarray and is not down-regulated in the analysed mutant strains (data not shown). However, the regulation of other nuclear-encoded subunits of photosystem II has to be determined. In contrast to the similar levels of downregulation of Lhcb gene expression in the two mutants, different expression patterns of Lhca transcripts were observed. Moreover, the analysis of two further non-allelic trans-splicing mutants (T111, T11-3) by real-time PCR revealed that Lhca expression in these strains is also not downregulated compared to the wildtype (Glanz, S. and Kück, U., unpublished data). All analysed mutant strains in this study are trans-splicing mutants, but they are affected in different genes. In TR72, two genes are deleted, both of which are necessary for the correct processing of the chloroplast tscA RNA, a co-factor of trans-splicing [14]. CC1051 originally carried the M18 mutation which leads to a C-terminal truncated trans-splicing factor Raa3 [22]. It has also picked up a second mutation, named Cen [13], which results in the transcripts of different nuclear-encoded chloroplast polypeptides not accumulating in this strain. Takahashi et al. [35] have shown that a photosystem I-deficient ΔpsaB mutant accumulates normal amounts of Lhca polypeptides. Therefore, the downregulation of Lhca-encoding transcripts in TR72 seems to be unique to this photosystem I mutant. One might speculate that the lack of the two factors in TR72 produces a signal which downregulates the expression of Lhca genes and that the malfunctions of the trans-splicing factors or lack of psaB polypeptide in the other mutants do not lead to such a regulatory signal. If the changes in abundance of LHC polypeptides encoding transcripts are a response to the defective chloroplast electron transport, then there should be a sensor for recognising oxidative stress in the organelle as well as a mechanism for tuning nuclear gene expression in order to prevent the formation of further reactive oxygen species. Using quantitative real-time PCR, several Lhcb transcripts were shown to be co-ordinately regulated in a light-dependent manner in C. reinhardtii [36]. Moreover, studies on the promoter elements of the nuclear

67

Lhcb1 gene in Chlamydomonas have revealed an influence of photo-oxidative damage in the chloroplast on the expression of Lhcb1 [37]. In Dunaliella tertiolecta, Escoubas et al. [38] demonstrated a light intensity-dependent regulation of cab gene transcription that is signaled from the chloroplast to the nucleus by the redox state of the plastoquinone pool. Our results together with the conclusions of former studies strongly suggest that feedback mechanisms monitoring the chloroplast status are transferring signals from chloroplast to the nucleus. Further studies are required to elucidate the nature of the regulatory mechanism of Lhc gene expression in C. reinhardtii and the factors involved in this process. Acknowledgements The authors are grateful to Andrea Wimbert and Swenja Ellßel for excellent technical assistant. CB received a fellowship from the Landesgraduiertenförderung NRW. This work was supported by the Deutsche Forschungs Gemeinschaft (DFG, Sonderforschungsbereich 480, project B3). References [1] J. Nickelsen, The Green Alga Chlamydomonas reinhardtii—A Genetic Model Organism, in: K. Esser, U. Lüttge, W. Beyschlag, F. Hellwig (Eds.), Progress in Botany, Springer-Verlag, Wien, 2005, pp. 68–89. [2] E.H. Harris, Chlamydomonas as a model organism, Annu. Rev. Plant Physiol. Plant Mol. Biol. 52 (2001) 363–406. [3] J.D. Rochaix, Post-transcriptional regulation of chloroplast gene expression in Chlamydomonas reinhardtii, Plant Mol. Biol. 32 (1996) 327–341. [4] R.A. Monde, G. Schuster, D.B. Stern, Processing and degradation of chloroplast mRNA, Biochimie 82 (2000) 573–582. [5] R.M. Dent, C.M. Haglund, B.L. Chin, M.C. Kobayashi, K.K. Niyogi, Functional genomics of eukaryotic photosynthesis using insertional mutagenesis of Chlamydomonas reinhardtii, Plant Physiol. 137 (2005) 545–556. [6] J.D. Rochaix, K. Perron, D. Dauvillee, F. Laroche, Y. Takahashi, M. Goldschmidt-Clermont, Post-transcriptional steps involved in the assembly of photosystem I in Chlamydomonas, Biochem. Soc. Trans. 32 (2004) 567–570. [7] M. Goldschmidt-Clermont, Y. Choquet, J. Girard-Bascou, F. Michel, M. Schirmer-Rahire, J.D. Rochaix, A small chloroplast RNA may be required for trans-splicing in Chlamydomonas reinhardtii, Cell 65 (1991) 135–143. [8] U. Kück, Y. Choquet, M. Schneider, M. Dron, P. Bennoun, Structural and transcriptional analysis of two homologous genes for the P700 chlorophyll a-apoproteins in Chlamydomonas reinhardtii: evidence for in vivo transsplicing. EMBO J. 6 (1987) 2185–2195. [9] M. Goldschmidt-Clermont, J. Girard-Bascou, Y. Choquet, J.D. Rochaix, Trans-splicing mutants of Chlamydomonas reinhardtii, Mol. Gen. Genet. 223 (1990) 417–425. [10] J. Girard-Bascou, Y. Choquet, M. Schneider, M. Delosme, M. Dron, Characterization of a chloroplast mutation in the psaA2 gene of Chlamydomonas reinhardtii, Curr. Genet. 12 (1987) 489–495. [11] K. Wostrikoff, J. Girard-Bascou, F.A. Wollman, Y. Choquet, Biogenesis of PSI involves a cascade of translational autoregulation in the chloroplast of Chlamydomonas, EMBO J. 23 (2004) 2696–2705. [12] H. Naver, E. Bourdreau, J.D. Rochaix, Functional studies of Ycf3: Its role in assembly of photosystem I and interactions with some of its subunits, Plant Cell 13 (2001) 2731–2745. [13] D. Hahn, P. Bennoun, U. Kück, Altered expression of nuclear genes encoding chloroplast polypeptides in non-photosynthetic mutants of Chlamydomonas reinhardtii: evidence for post-transcriptional regulation, Mol. Gen. Genet. 252 (1996) 362–370. [14] C. Balczun, A. Bunse, D. Hahn, P. Bennoun, J. Nickelsen, U. Kück, Two

68

[15]

[16]

[17] [18] [19]

[20]

[21] [22]

[23]

[24]

[25]

[26]

C. Balczun et al. / Biochimica et Biophysica Acta 1732 (2005) 62–68 adjacent nuclear genes are required for functional complementation of a chloroplast trans-splicing mutant from Chlamydomonas reinhardtii, Plant Journal 43 (2005) 636–648. D. Hahn, J. Nickelsen, A. Hackert, U. Kück, A single nuclear locus is involved in both chloroplast RNA trans-splicing and 3′ end processing, Plant J. 15 (1998) 575–581. J. Girard, N.H. Chua, P. Bennoun, G. Schmidt, M. Delosome, Studies on mutants deficient in photosystem I reaction center in Chlamydomonas reinhardtii, Curr. Genet. 2 (1980) 215–221. E.H. Harris, The Chlamydomonas Sourcebook, A Comprehensive Guide to Biology and Laboratory Use, Academic Press, San Diego, 1989. D.R. Davies, A. Plaskitt, Genetical and structural analysis of cell-wall formation in Chlamydomonas reinhardtii, Genet. Res. 17 (1971) 33–43. C.D. Silflow, R.L. Chisholm, T.W. Conner, L.P. Ranum, The two alphatubulin genes of Chlamydomonas reinhardtii code for slightly different proteins, Mol. Cell. Biol. 5 (1985) 2389–2398. M. Nowrousian, C. Ringelberg, J.C. Dunlap, J.J. Loros, U. Kück, Crossspecies microarray hybridization to identify developmentally regulated genes in the filamentous fungus Sordaria macrospora, Mol. Genet. Genomics. 273 (2005) 137–149. N.J. Gumpel, S. Purton, Playing tag with Chlamydomonas, Trends Cell Biol. 4 (1994) 299–301. C. Rivier, M. Goldschmidt-Clermont, J.D. Rochaix, Identification of an RNA–protein complex involved in chloroplast group II intron transsplicing in Chlamydomonas reinhardtii, EMBO J. 20 (2001) 1765–1773. D. Elrad, A.R. Grossman, A genome's-eye view of the light-harvesting polypeptides of Chlamydomonas reinhardtii, Curr. Genet. 45 (2004) 61–75. Z. Zhang, J. Shrager, M. Jain, C.W. Chang, O. Vallon, A.R. Grossman, Insights into the survival of Chlamydomonas reinhardtii during sulfur starvation based on microarray analysis of gene expression, Eukaryot. Cell 3 (2004) 1331–1348. C.S. Im, Z.D. Zhang, J. Shrager, C.W. Chang, A.R. Grossman, Analysis of light and CO2 regulation in Chlamydomonas reinhardtii using genomewide approaches, Photosynth. Res. 75 (2003) 111–125. B. Erickson, D.B. Stern, D.C. Higgs, Microarray analysis confirms the specificity of a Chlamydomonas reinhardtii chloroplast RNA stability mutant, Plant Physiol. 137 (2005) 534–544.

[27] J.W. Lilly, J.E. Maul, D.B. Stern, The Chlamydomonas reinhardtii organellar genomes respond transcriptionally and post-transcriptionally to abiotic stimuli, Plant Cell 14 (2002) 2681–2706. [28] T. Desprez, J. Amselem, M. Caboche, H. Hofte, Differential gene expression in Arabidopsis monitored using cDNA arrays, Plant J. 14 (1998) 643–652. [29] J. Minagawa, Y. Takahashi, Structure, function and assembly of photosystem II and its light-harvesting proteins, Photosynth. Res. 82 (2004) 241–263. [30] E.J. Stauber, A. Fink, C. Markert, O. Kruse, U. Johanningmeier, M. Hippler, Proteomics of Chlamydomonas reinhardtii light-harvesting proteins, Eukaryot. Cell 2 (2003) 978–994. [31] K. Asada, in: N.R. Baker (Ed.), Advances in Photosynthesis, Kluwer Academic Publishers, Dordrecht, The Netherlands, 1996, pp. 123–150. [32] K. Asada, in: N.R. Baker, J.R. Bowyer (Eds.), Photoinhibition of Photosynthesis: From Molecular Mechanisms to the Field, BIOS Scientific Publishers, Oxford, UK, 1994, pp. 129–142. [33] O. Kruse, Light-induced short-term adaptation mechanisms under redox control in the PS II–LHCII supercomplex: LHC II state transitions and PS II repair cycle, Naturwissenschaften 88 (2001) 284–292. [34] S.P. Mayfield, P. Bennoun, J.D. Rochaix, Expression of the nuclear encoded OEE1 protein is required for oxygen evolution and stability of photosystem II particles in Chlamydomonas reinhardtii, EMBO J. 6 (1987) 313–318. [35] Y. Takahashi, T.A. Yasui, E.J. Stauber, M. Hippler, Comparison of the subunit compositions of the PSI–LHCI supercomplex and the LHCI in the green alga Chlamydomonas reinhardtii, Biochemistry 43 (2004) 7816–7823. [36] H. Teramoto, A. Nakamori, J. Minagawa, T.A. Ono, Light-intensitydependent expression of Lhc gene family encoding light-harvesting chlorophyll-a/b proteins of photosystem II in Chlamydomonas reinhardtii, Plant Physiol. 130 (2002) 325–333. [37] D. Hahn, U. Kück, Identification of DNA sequences controlling light- and chloroplast-dependent expression of the lhcb1 gene from Chlamydomonas reinhardtii, Curr. Genet. 34 (1999) 459–466. [38] J.M. Escoubas, M. Lomas, J. LaRoche, P.G. Falkowski, Light intensity regulation of cab gene transcription is signaled by the redox state of the plastoquinone pool, Proc. Natl. Acad. Sci. U. S. A. 92 (1995) 10237–10241.