This Article Abstract Full Text (PDF) Supplemental Data All Versions of this Article: 144/4/1946    most recent pp.107.103341v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via ISI Web of Science (4) Citing Articles via Google Scholar Google Scholar Articles by Eisenhut, M. Articles by Hagemann, M. Search for Related Content PubMed PubMed Citation Articles by Eisenhut, M. Articles by Hagemann, M. Agricola Articles by Eisenhut, M. Articles by Hagemann, M.

BIOENERGETICS AND PHOTOSYNTHESIS

Long-Term Response toward Inorganic Carbon Limitation in Wild Type and Glycolate Turnover Mutants of the Cyanobacterium Synechocystis sp. Strain PCC 6803^1,[W]

Universität Rostock, Institut für Biowissenschaften, Abteilung Pflanzenphysiologie (M.E., H.B., M.H.) und Ökologie (H.S.), D–18059 Rostock, Germany; University of Amsterdam, Institute for Biodiversity and Ecosystem Dynamics, NL–1018WS Amsterdam, The Netherlands (E.A.v.W., H.C.P.M.); Universität Rostock, Institut für Pathologie, Elektronenmikroskopisches Zentrum, D–18055 Rostock, Germany (L.J.); Netherlands Institute of Ecology, Centre for Limnology, Department of Foodweb Studies, NL–3631AC Nieuwersluis, The Netherlands (B.W.I.); and Eawag, Swiss Federal Institute of Aquatic Sciences and Technology, Centre of Ecology, Evolution, and Biogeochemistry, CH–6047 Kastanienbaum, Switzerland (B.W.I.)

	ABSTRACT

TOP ABSTRACT RESULTS AND DISCUSSION CONCLUSION MATERIALS AND METHODS LITERATURE CITED

 
Concerted changes in the transcriptional pattern and physiological traits that result from long-term (here defined as up to 24 h) limitation of inorganic carbon (C_i) have been investigated for the cyanobacterium Synechocystis sp. strain PCC 6803. Results from reverse transcription-polymerase chain reaction and genome-wide DNA microarray analyses indicated stable up-regulation of genes for inducible CO₂ and HCO₃^– uptake systems and of the rfb cluster that encodes enzymes involved in outer cell wall polysaccharide synthesis. Coordinated up-regulation of photosystem I genes was further found and supported by a higher photosystem I content and activity under low C_i (LC) conditions. Bacterial-type glycerate pathway genes were induced by LC conditions, in contrast to the genes for the plant-like photorespiratory C2 cycle. Down-regulation was observed for nitrate assimilation genes and surprisingly also for almost all carboxysomal proteins. However, for the latter the observed elongation of the half-life time of the large subunit of Rubisco protein may render compensation. Mutants defective in glycolate turnover (glcD and gcvT) showed some transcriptional changes under high C_i conditions that are characteristic for LC conditions in wild-type cells, like a modest down-regulation of carboxysomal genes. Properties under LC conditions were comparable to LC wild type, including the strong response of genes encoding inducible high-affinity C_i uptake systems. Electron microscopy revealed a conspicuous increase in number of carboxysomes per cell in mutant glcD already under high C_i conditions. These data indicate that an increased level of photorespiratory intermediates may affect carboxysomal components but does not intervene with the expression of majority of LC inducible genes.

In higher plants, changes in C_i concentrations strongly influence the carbon-fixing rate of Rubisco, since under LC its second substrate, oxygen, is increasingly used by Rubisco. Oxygenase activity forms large equimolar amounts of 2-phosphoglycolate (2PG) and the Calvin cycle intermediate, 3-phosphoglycerate. The 2PG by-product inhibits the Calvin cycle enzyme triosephosphate isomerase (Husic et al., 1987); it is rapidly metabolized by the action of at least 10 different enzymes in the glycolate pathway, also known as the photorespiratory C2 cycle (Tolbert, 1997). Though it is widely believed that the sophisticated CCM should inhibit oxygenase activity of Rubisco in cyanobacteria making a photorespiratory C2 cycle unnecessary, we have recently shown that an active 2PG metabolism operates in cyanobacteria. A Synechocystis mutant impaired in the central step of this metabolism, the glycolate dehydrogenase (GlcD) converting glycolate into glyoxylate, accumulated the photorespiratory intermediate glycolate already at high concentrations of CO₂ (HC), indicating a lower efficiency of CCM than generally assumed. It was suggested that the glycolate formed could be metabolized either by a plant-like C2 cycle or a bacterial-like glycerate pathway with domination of the plant-like C2 cycle (Eisenhut et al., 2006).

Despite great progress in understanding the dynamic reactions that relate to C_i availability, one interesting and central question is still open: What is the primary signal that induces the response toward C_i limitation? Different hypotheses have been put forward, some of which include a photorespiration-based mechanism (Kaplan and Reinhold, 1999; Woodger et al., 2005a). If the CO₂ concentration becomes limiting, the O₂/CO₂ concentration ratio and the oxygenase activity of Rubisco increase. As a result, intermediates of the photorespiratory C2 cycle may accumulate. It might be possible that cyanobacteria use alterations in the levels of these metabolites to sense C_i limitation. Two transcriptional factors, CmpR (Omata et al., 2001) and NdhR (Wang et al., 2004), were found to be involved in LC-induced up-regulation of many but not all LC-regulated genes; however, the signal transduction processes leading to their activation are still not known.

Using genome-wide DNA microarrays, it has been shown that shortly after a stress treatment numerous genes are transiently up-regulated, many of them encoding general stress proteins. However, after long-term acclimation to suboptimal conditions only a few genes remain transcribed at the elevated level, among those often functionally important proteins specific for a given stress condition were found to behave in this way (e.g. light acclimation by Hihara et al., 2001, salt acclimation by Marin et al., 2004). In this work, we examined the long-term response with regard to transcriptional and physiological alterations in Synechocystis to LC. Furthermore, two mutants defective in photorespiratory glycolate turnover of Synechocystis, glcD, and gcvT (Eisenhut et al., 2006) were included in the study to reassess the assumption that photorespiratory intermediates could be signal for sensing C_i limitation.

	RESULTS AND DISCUSSION

TOP ABSTRACT RESULTS AND DISCUSSION CONCLUSION MATERIALS AND METHODS LITERATURE CITED

 
Synechocystis wild-type or mutant cells were shifted after precultivation with buffered BG11 at pH 7 from HC (5% CO₂) aeration to LC (0.035% CO₂). At this pH, a large amount of C_i is available as CO₂. Long-term alterations in transcript prevalence and physiological patterns following the transfer to LC conditions were investigated. Samples were taken at different time points during 24 h after the transfer to LC to follow the different stages of response as established by Wang et al. (2004). In this study, we applied semiquantitative reverse transcription (RT)-PCR and a newly developed genome-wide DNA microarray for Synechocystis, which is composed of custom-selected 60-mer synthetic oligo probes on an Agilent technology platform (E. Aguirre von Wobeser, B.W. Ibelings, J. Huisman, and H.C.P. Matthijs, unpublished data). The array comprises up to four different, nonoverlapping probes for each single gene. Only changes, which were reproducibly detected at each gene-specific spot and in each biological replicate, were taken into consideration. In the following, we present and discuss only selected data (the complete microarray data set is given as Supplemental Table S1).

Long-Term Transcriptional Changes of Genes Coding for CCM Components

Different types of C_i transporters are crucial for CCM in cyanobacteria. The expression of their genes is known to be regulated by C_i. Therefore, their expression data may verify the stringency of our C_i limitation and the reliability of the microarray data obtained with the new platform, also with reference to the work of Wang et al. (2004). Genes encoding all five C_i uptake systems in Synechocystis BCT1, SbtA, NDH-1₃, NDH-1₄, and BicA were selected (Table I ). In both studies, the operons encoding BCT1 (cmp operon, slr0040–slr0044), SbtA (sbt operon, slr1512, slr1513), and NDH-1₃ (ndhF3 operon, sll1732–sll1736) were found to be similarly up-regulated, with the strongest increase of expression for the cmp genes. Note that despite its high induction the BCT1 transporter is not essential for acclimation toward LC (Omata et al., 1999). The genes of these C_i uptake systems are known to be induced by LC conditions within 15 min (McGinn et al., 2003; Woodger et al., 2003; Wang et al., 2004). As expected from literature data (Shibata et al., 2001), the NDH-1₄ uptake system showed constitutive expression (Table I). Interestingly, for the open reading frame (ORF) sll0834, encoding the Na⁺-dependent HCO₃^– transporter BicA, a decline in transcript abundance (57%) could be detected after 24 h LC conditions. Since BicA expression was highly up-regulated in response to LC conditions in Synechococcus PCC 7002 (Price et al., 2004), a constitutive expression was postulated earlier for Synechocystis (Price et al., 2004; Wang et al., 2004). However, it should be noted that these are results for short-term responses to LC conditions, while we report here on gene expression after relatively long-term growth under LC.

View larger version (66K): [in this window] [in a new window] [as a PowerPoint slide]   Figure 1. Semiquantitative RT-PCR of Ci responsive genes sbtA, bicA, rbcL, ccmK2, and flv3. Wild-type cells were grown in BG11 medium at pH 7 and bubbled with air supplemented with 5% CO2. Samples were taken 0, 1, 2, 3, 12, and 24 h after transfer to pure air. The transcript level of the constitutive expressed gene rnpB served as control.

Genes Encoding Enzymes for 2PG Metabolism

Recently, metabolism of 2PG, generated by oxygenase activity of Rubisco, has been reinvestigated by our group in Synechocystis. Genes coding for enzymes of the plant-like photorespiratory C2 cycle and bacterial-type glycerate pathway were assigned (Eisenhut et al., 2006). Almost all the genes involved in a plant-like C2 cycle were repressed in cells shifted to LC conditions (Table II ). Exceptions are glcD (sll0404), gcl (sll1981), and tsr (slr0229), which form the bypassing bacterial-type glycerate pathway and showed elevated transcription. Though, transcript abundances suggest that the increased glycolate metabolism under LC mainly proceeds via the bacterial-type glycerate pathway, our former physiological investigations of defined Synechocystis mutants in phosphoglycolate metabolism rather pointed to a dominant role of the plant-like C2 cycle (Eisenhut et al., 2006). Possibly, in wild-type cells, the contribution of the glycerate pathway could be of higher importance than assumed from the artificial situation in mutants. Overall, the data support the view of a cooperative action of C2 cycle and glycerate pathway to avoid accumulation of toxic intermediates such as 2PG or glycolate in cyanobacteria.

View this table: [in this window] [in a new window]   Table II. List of genes encoding proteins putatively involved in phosphoglycolate metabolism (Eisenhut et al., 2006) Data represent fold ratios of transcript levels in cells exposed to LC for 24 h relative to that observed under HC. For comparison the data set for 12 h by Wang et al. (2004) is shown. ORF numbers, genes, and functions according to CyanoBase (http://www.kazusa.or.jp) or Eisenhut et al. (2006).

Further Newly Identified C_i Responsive Genes after Long-Term Acclimation

Though the results of microarray analyses for 12 (Wang et al., 2004) and 24 h (this study) C_i limitation mainly correspond to each other, differences were observed. Beside the identification of up-regulated genes coding for proteins possibly involved in glycolate breakdown, a clear and concerted increase in the transcription of many PSI genes is noticed after 24 h at LC (Table III ), which was not seen before in the short-term experiments by Wang et al. (2004). Increased PSI activity is corroborated by the higher transcript levels for PetG and PetJ. The potential PS1 increase indicated by the transcriptional data was experimentally proven. Seventy-seven degrees Kelvin fluorescence emission spectra after 440 nm (chlorophyll a) excitation showed a higher PSI/PSII emission ratio with LC-grown cells (Fig. 2A ), indicating a preferential excitation energy transfer toward PSI and/or a higher PSI/PSII ratio. To test whether this effect is caused by excitation energy transfer alone or not, P700 redox state was determined by means of pulse-amplitude modulated absorbance change measurements at 820 nm (Schreiber et al., 1988). As can be seen from Figure 2, B and C, P700⁺ absorbance signal got saturated above approximately 20 W m^–2 far red irradiance for both cell types, however, the signals (expressed as relative absorbance change, I/I, white circles) were far stronger for LC-grown cells compared to HC-grown ones, indicating a higher P700 content. Activation of PSII by a strong white light pulse led to rereduction of oxidized P700, causing a drop in the absorbance signal measured for P700 (black circles). For HC-grown cells the amount of rereduction was about 65% of the value at saturating far-red light intensities, whereas for LC-grown cells just 45% of the P700⁺ got rereduced, the majority of the P700 was still locked in the oxidized state. The activation of PSI at transcriptional and activity level and unchanged PSII activity demonstrates an increase in cyclic electron flow, which helps cells to acquire the capacity for photophosphorylation that operates independently of the linear photosynthetic electron transfer pathway. In addition, cyclic electron flow plays a crucial role in the energy supply for CO₂ to HCO₃^– interconversion at specialized NDH complexes such as NDH-1₃ as part of the CCM (see Zhang et al., 2004). However, the operation of PSI in the cyclic mode may not necessarily require the involvement of NADP⁺ as electron acceptor per se, since it is known that the cyclic electron flow around PSI in cyanobacteria employs several connections between the cytoplasmic sites of PSI and plastoquinone or cytochrome b₆f complex (Hagemann et al., 1999; Matthijs et al., 2002; Bukhov and Carpentier, 2004; Zhang et al., 2004; Yeremenko et al., 2005). Furthermore, the up-regulated genes cydAB, which encode a cytochrome bd quinol oxidase (Howitt and Vermaas, 1998; Berry et al., 2002) that is localized mostly on the cytoplasmic membrane, could provide cells with a capacity for oxidation of any accumulating reductant at the cytoplasmic membrane, potentially including that from the proposed glycolate breakdown via formate (see Table II).

View larger version (11K): [in this window] [in a new window] [as a PowerPoint slide]   Figure 2. Characterization of relative PSI content by fluorescence emission spectra and PSI activity by P700 absorbance changes of wild-type cells grown in BG11 medium at pH 7 and bubbled with air supplemented with 5% CO2 (HC) or for 24 h with pure air (LC). A, 77 K fluorescence emission spectra of HC (thick line) and LC (thin line) acclimated wild-type cells at 440 nm excitation. Mean values of four independent samples of each culture are shown, which were standardized between 780 and 688 nm. SD is indicated by the vertical bars. B and C, Effects of different far-red (FR) irradiance on P700 redox state in the presence (black circles) and absence (white circles) of strong white light (+WL), exciting PSII. Mean values of five independent samples of wild-type cells grown at HC (B) or LC (C) conditions are shown (bars indicate SD).

As mentioned above, regulatory processes involved in C_i acclimation are mostly unknown. Therefore, in our data set we searched for genes encoding proteins with regulatory functions and that showed stable up-regulation under LC conditions. In addition to the known C_i depending transcriptional factors CmpR (Omata et al., 2001) and NdhR (Wang et al., 2004), whose transcript abundances were elevated (cmpR: 3.8-fold, ndhR: 5.2-fold; Table I), we could newly identify three putative regulatory proteins, which might be involved in C_i regulated gene expression. Besides slr2104 (2.2-fold increase in mRNA level) encoding Hik22, two other genes seemed to be very interesting with respect to possible regulatory processes, namely the cotranscribed genes slr1409 and slr1410 (4-fold increased at LC, Table III), encoding proteins of the WD-repeat protein family. Their exact function is not clear but most of these proteins are involved in the regulation of metabolic processes, realized by nonenzymatic protein-protein interactions (Neer et al., 1994; Neer and Smith, 2000). Though mainly known from eukaryotic organisms, the occurrence of those proteins was reported for the cyanobacteria Anabaena PCC 7120 and Synechocystis too (Hisbergues et al., 2001). Only one of the five annotated WD-repeat proteins in Synechocystis has been investigated so far. Different mutants defective in slr0143 (hat) were shown to react sensitively toward LC conditions, including a loss of the inducible high-affinity transport system for C_i. Hence, it was postulated that Hat should be involved in the control of the activity of this inducible C_i transport system (Bedu et al., 1995). We suggest that the WD-repeat proteins Slr1409 and Slr1410 might play a regulatory role in long-term response toward LC similar to Hat.

Another group of newly identified C_i responsive genes forms the rfb cluster, which comprises 44 genes (slr0976–0985, slr1610–1619, slr1062–1087) encoding enzymes putatively involved in the biosynthesis of cell wall or outer membrane polysaccharides (Table III). The transcription of all genes in this cluster, with the exception of an inserted gene for a transposase (slr1075), was induced (2-fold on average, data not shown) 24 h after transfer to LC, while its expression was not significantly changed or even repressed until 12 h after downshift (Wang et al., 2004). The same transcriptional kinetics were observed for sll0923 (4.6-fold), a gene whose product EpsB has similarity to an uncharacterized protein involved in exopolysaccharide biosynthesis. Our finding of C_i induced up-regulation of the rfb cluster and epsB corresponds to the observation of marked changes in the cyanobacterial cell wall structure, particularly in the L2 layer, during acclimation to LC conditions (Marcus et al., 1982). Moreover, it has been observed that Anabaena variabilis (Marcus et al., 1982), Synechocystis, or Synechococcus PCC 7942 cells exposed to LC are far less susceptible to lysozyme treatment than HC grown cells (A. Kaplan, personal communication). It is likely that such alterations of cell wall and outer membrane layers during long-term acclimation to LC conditions lower the diffusive permeability of cyanobacteria for CO₂ and HCO₃^–. Particularly, CO₂ entry was shown to be a passive diffusion involving aquaporins (Tchernov et al., 2001).

One central regulatory point under LC conditions is the balancing of the carbon/nitrogen ratio. As expected, we could observe a striking transcriptional repression of genes encoding proteins crucial for nitrogen acquisition and assimilation. As found before (Wang et al., 2004), the nrt and the moa operons were strongly down-regulated (80%–90%, data not shown). They encode all proteins necessary for the uptake and assimilation of nitrate. Additionally, we found a similar strong decline in transcript abundances for the genes sll0783 to sll0787 (Table III). With exception of Sll0784, the nitrilase MerR (Heinemann et al., 2003), the functions of all other encoded proteins are unknown. By the action of nitrilases, which hydrolyze nitriles to the corresponding carboxylic acids and ammonium, many bacteria like Pseudomonas fluorescens EBC191 can use nitriles as nitrogen source (Kiziak et al., 2005). Obviously, C_i limitation down-regulates also nitrogen demand. The involvement of the general nitrogen regulatory protein PII in LC acclimation has been shown before (Lee et al., 1999).

Alterations in Transcript Levels in Mutants Defective in Glycolate Turnover

To reinvestigate the hypothesis that alterations in concentrations of photorespiratory metabolites could act as signal for sensing C_i limitation (Kaplan and Reinhold, 1999; Woodger et al., 2005a, 2005b), we examined the global C_i regulated expression pattern in cells of two mutants defective in this pathway. While cells of gcvT, a mutant with a defect in the T protein of the Gly decarboxylase complex, accumulated Gly, in cells of glcD, a mutant with a knockout in GlcD, elevated glycolate levels were found even under C_i saturated conditions (Eisenhut et al., 2006). The majority of genes were transcribed equally in mutant and wild-type cells including almost all of the differentially regulated genes between HC and LC conditions such as the genes for C_i uptake systems (see Supplemental Table S1). Therefore, the data set from the mutants fully supports the changes observed with wild-type cells including all the newly identified C_i-regulated genes mentioned and discussed above. Moreover, the wild-type-like induction of most C_i-regulated genes in the Gly or glycolate accumulating mutant strains indicate that the intermediates of the photorespiratory metabolism do not serve to regulate transcription of the majority of these genes.

However, several genes showing a changed behavior between mutant and wild-type cells were found. Particularly interesting are differences in gene expression, which overlap in the two mutant transcriptomes (Table IV ). A reduced transcript level of the rbcLXS and ccmK-N operon was detected for gcvT and glcD under HC conditions compared to HC-grown wild-type cells. Moreover, the expression of the mer operon was reduced in mutants too. RT-PCR analyses supported these microarray data (Fig. 3 ). Basically, these obvious differences in transcript abundances resembled some features of long-term C_i limitation in wild-type cells, where these operons were also down-regulated. Therefore, these few changes between transcript abundances in mutant and wild-type cells are characteristic for changes normally occurring in LC-grown wild-type cells. Probably, the accumulation of photorespiratory intermediates such as glycolate and/or Gly in mutants of glcD and gcvT may be an indication for C_i limitation in Synechocystis cells to regulate not the majority but a defined subset of C_i-responsive genes. In a former study, a Synechococcus PCC 7942 mutant with defect in subunit F of the GlcD (glcF) had been analyzed for altered transcript abundances of high-affinity C_i transporter genes (cmpA, sbtA, cupA) under various CO₂/O₂ conditions. As found here, the expression of these transporter genes was not changed, leading to the conclusion that intermediates are not significant in signaling (Woodger et al., 2005b). Obviously, the expression of most C_i responsive genes, including the high-affinity C_i transporter genes cmpA, sbtA, and cupA, is not regulated by the content of glycolate or Gly, while the changed transcript level of the rbcLXS, ccmK-N, and mer operon in mutant cells or LC-grown wild-type cells could be sensed via the changed levels of photorespiratory intermediates. Moreover, the different behavior of various groups of C_i-regulated genes gives evidence for the cooperation of different regulatory mechanisms in LC acclimation.

View this table: [in this window] [in a new window]   Table IV. Ci limitation phenotype in cells of mutants gcvT and glcD grown at 5% CO2 Data represent fold ratios of transcript levels in mutant cells relative to levels found in wild-type (WT) cells exposed to HC. ORF numbers, genes, and functions according to CyanoBase (http://www.kazusa.or.jp).

View larger version (37K): [in this window] [in a new window] [as a PowerPoint slide]   Figure 3. Semiquantitative RT-PCR and immunoblotting to determine effects of mutation in gcvT and glcD on the expression of rbcL. RNA was isolated from cells of wild type, gcvT, or glcD grown in BG11 medium at pH 7 and bubbled with air supplemented with 5% CO2. The transcript level of the constitutively expressed gene rnpB served as control.

Indications for Posttranscriptional Regulation in Long-Term C_i-Limited Cells

The observed down-regulation of rbcLXS and ccmK-N mRNA levels after long-term acclimation to LC conditions was contradictory to findings in short-term LC-acclimated cyanobacterial cells, which showed up-regulation of such genes (e.g. Woodger et al., 2003). Moreover, in Synechococcus PCC 7942, the raise for rbcL mRNA was also found at the protein level, indicating transcriptional control of Rubisco in this cyanobacterial strain (Harano et al., 1997). This prompted us to have a look at the RbcL amount in Synechocystis cells. Interestingly, though transcript abundances were diminished (Table I), immunoblot analyses revealed a distinct increase in RbcL protein after C_i downshift as expected for an increased CCM capacity (Fig. 4A ). The most obvious explanation for this contradictory finding would be a variation in the half-life of RbcL. Therefore, protein stability was investigated in cells treated with the translational inhibitor lincomycin. Lincomycin-treated cells showed a steady decline in RbcL content during 24 h at HC conditions, while only a slight RbcL decrease occurred under LC conditions. This investigation of the stability of RbcL indicated a doubling of Rubisco half-life in LC-grown cells versus HC-grown Synechocystis cells (Fig. 5 ). Overall, RbcL seems to have a low turnover that is likely due to the fact that Rubisco is sequestered in the carboxysome and hence is relatively protected against degradation. Definitely, the results of the stability experiment have to be discussed very carefully since the application of the protein biosynthesis inhibitor lincomycin over such a long time of 24 h may affect its stability, or alterations in cell envelope during LC acclimation make cells differentially susceptible to lincomycin. Despite these objections, we suggest that the Rubisco content is not controlled at the transcriptional level in Synechocystis but possibly by the half-life of its large subunit. Another possibility to explain the contradiction between increased Rubisco content and decreased rbcL mRNA level is related to changed growth rates under HC versus LC conditions. Under our culture conditions, HC cells grew about 3 times faster than under LC conditions (Eisenhut et al., 2006). During the 24 h period after the shift from HC to LC investigated here almost no growth was observed. Even assuming a constant half-life of Rubisco, the delayed dilution of Rubisco by delayed cell separation could also compensate for the reduced rbcL transcript level in LC-acclimated cells, resulting in finally increased Rubisco content.

View larger version (70K): [in this window] [in a new window] [as a PowerPoint slide]   Figure 4. Effect of long-term Ci limitation on the transcript and protein level of RbcL (A) and CcmK (B). Wild-type cells were grown in BG11 medium at pH 7 and bubbled with air supplemented with 5% CO2. Samples were taken 0, 12, and 24 h after transfer to pure air. The rbcL and ccmK2 transcript levels were compared by semiquantitative RT-PCR (top sections) and the amount of RbcL or CcmK proteins were checked by immunoblotting using specific antibodies (bottom sections). In the immunoblotting experiments the same amount of total protein (5 µg) was applied to each lane.

View larger version (16K): [in this window] [in a new window] [as a PowerPoint slide]   Figure 5. Effect of long-term Ci limitation on the half-life of RbcL. Wild-type cells were grown in BG11 medium at pH 7 and bubbled with air supplemented with 5% CO2 (HC) or with pure air (LC), as indicated. At time point zero the translational inhibitor lincomycin (final concentration 250 µg mL–1) was added. Samples were taken 0, 3, 6, 9, 12, and 24 h after application of lincomycin. Same amount of total protein (5 µg) was applied to each lane. RbcL was detected by immunoblotting using a specific antibody.

View larger version (112K): [in this window] [in a new window] [as a PowerPoint slide]   Figure 6. Changes in cell morphology during acclimation toward LC conditions. A, Representative electron micrographs of wild type (a + d), gcvT (b + e), and glcD (c + f) cells, respectively, are shown. Cells were pregrown in BG11 medium at pH 7 and 5% CO2. Samples were taken before (top section) and 24 h after transfer to pure air (bottom row). Carboxysomes (C), polyphosphate bodies (P), and glycogen granules (G) are marked by arrows. B, Carboxysomes per cell were counted from 50 thin sections per strain and treatment, presumably cut in the middle of the cell. Means and confidence intervals are shown. *, Statistically significant difference in the carboxysome number compared to wild-type cells grown with 5% CO2 (P 0.05, n = 50).

Acclimation to alterations in C_i availability includes global changes in metabolism and obviously also morphology. Carboxysomes represent one central compound of CCM. In these hexagonically shaped microcompartments CO₂ is finally extracted from HCO₃^– and becomes fixed by Rubisco. As it can be seen in Figure 6, a significant rise in carboxysome number becomes evident for LC-grown cells. The average number for carboxysomes in a thin section of cells increased from about one per cell to 2.3 per cell in cultures shifted from HC to LC. Interestingly, cells from the mutant glcD possessed two carboxysomes per cell already under HC conditions. This is an additional feature characteristic for LC-grown wild-type cells, which was found in HC cells of mutants defective in photorespiratory metabolism, again supporting the assumption that glycolate, which accumulates in this mutant, might eventually act as signal for the perception of C_i limitation, at least for regulation of some aspects of CCM.

The electron micrographs showed further that the C-storage product, glycogen, is degraded during acclimation toward LC conditions (Fig. 6). HC-grown cells contain much more glycogen granules in the thylakoid space compared to LC-grown cells. The first step in glycogen degradation is catalyzed by glycogen phosphorylase (GlgP). The genome of Synechocystis harbors two glgP genes: sll1356 and slr1367. Recently, the functional divergence of the two homologs was shown. When grown photoautotrophically, Sll1356 seemed to play the dominant role in glycogen consumption (Fu and Xu, 2006). Correspondingly, the transcript abundance of sll1356 is raised under LC conditions (3.5-fold, see Table III), while that of the second GlgP (slr1367) is repressed (4-fold, Table III).

In our DNA microarray analyses, genes of the rfb cluster, whose products are most probably involved in the synthesis of cell envelope components, were newly identified to be C_i responsive (see Table III). Our electron microscopy study revealed only slight differences in the cell wall composition such as more thickness of the outermost stained layer, similar to what was reported before for acclimation to LC conditions in Anabaena (Marcus et al., 1982).

	CONCLUSION

TOP ABSTRACT RESULTS AND DISCUSSION CONCLUSION MATERIALS AND METHODS LITERATURE CITED

 
Our analysis of long-term response toward C_i limitation in Synechocystis showed a stable up-regulation of genes encoding inducible CO₂ and HCO₃^– uptake systems, which are already induced after short-term transfer to LC conditions. However, genes encoding structural and enzymatic components of the carboxysome were down-regulated, while the number of carboxysomes per cell and protein levels of Rubisco increased. The amount of the carboxysomal shell proteins CcmK1 to 4 remained unaltered. These contradictory results indicate that different components of the cyanobacterial CCM are probably differently regulated: C_i transporters mainly by transcriptional changes and carboxysome components by posttranscriptional mechanisms as was indicated for the half-life in the case of Rubisco.

Besides the finding of already known genes involved in LC acclimation, our relatively long-term experiments revealed also the importance of additional processes. The stable induction of genes encoding proteins necessary for cell wall synthesis implies changes in C_i permeability. Up-regulation of PSI certainly helps to produce extra energy for C_i uptake, particularly CO₂ conversion into HCO₃^– at NDH-1. Moreover, genes for proteins necessary to avoid overreduction and oxygen radical accumulation were found to be up-regulated in a later stage during LC acclimation, which supports the view of an overlapping response toward LC and high light (Hihara et al., 2001).

Detailed investigation of the regulation of genes encoding enzymes involved in the metabolism of the Rubisco oxygenase activity product phosphoglycolate showed also a differential regulation. The plant-like C2 cycle enzymes are obviously not transcriptionally regulated after HC to LC transfer, while the bacterial-type glycerate pathway is activated. Moreover, the DNA microarray data indicated a third route for phosphoglycolate metabolism involving its complete breakdown to CO₂ via oxalate and formate, since the genes for the necessary steps were induced in a coordinated way.

In addition to wild-type cells, the mutants gcvT and glcD with defects in phosphoglycolate metabolism were included in this study. Generally, the transcriptional changes in mutant cells during LC acclimation were comparable to those observed in wild-type cells. Under HC conditions early up-regulation of inducible C_i uptake systems could not be observed. However, some changes in gene expression characteristic for the LC status of wild-type cells were detected already in HC-grown mutant cells. Together with the finding of an increased carboxysome number per cell in mutant glcD cells these data can be taken as an indication that the accumulation of phosphoglycolate metabolism intermediates such as glycolate (mutant glcD) or Gly (mutant gcvT) may serve as chemical signals to trigger not the whole but specific aspects of LC acclimation.

	MATERIALS AND METHODS

TOP ABSTRACT RESULTS AND DISCUSSION CONCLUSION MATERIALS AND METHODS LITERATURE CITED

 

Strains and Culture Conditions

The cyanobacterial strains used in this work are listed in Table V . The Glc-tolerant strain of Synechocystis sp. PCC 6803 was obtained from Professor Murata (National Institute for Basic Biology, Okazaki, Japan) and served as the wild type. Cultivation of mutants was performed at 50 µg mL^–1 kanamycin (Km) or at 20 µg mL^–1 spectinomycin. For the experiments, axenic cultures of the cyanobacteria (about 10⁸ cells mL^–1) were grown photoautotrophically in batch cultures (3 cm glass vessels with 5 mm glass tubes for aeration) at 29°C under continuous illumination of 130 µmol photons s^–1 m^–2 (warm light, Osram L58 W32/3) bubbling (flow rate about 5 mL min^–1) with air enriched with CO₂ (5% defined as HC) in the BG11 medium at pH 7.0. C_i limitation was set by transferring exponentially growing cultures (OD₇₅₀ of 0.9, volume of 130 mL) from bubbling with CO₂-enriched air to bubbling with pure air (about 0.035% CO₂ defined as LC).

RNA Isolation and RT-PCR Analysis

Cells from 10 mL of culture were harvested by centrifugation at 4,000 rpm for 5 min at 4°C and were immediately frozen at –80°C. Total RNA was extracted after pretreatment with hot phenol and chloroform by the High Pure RNA isolation kit (Roche Diagnostics). The same RNA samples were used for labeling in DNA microarray experiments and RT-PCR.

For RT-PCR, 3 µg RNA was used for cDNA synthesis. RT was performed with random hexamers (Amersham Pharmacia), dNTP mix (MBI Fermentas), and RevertAid M-MuLV Reverse Transcriptase (MBI Fermentas). The generated cDNA was diluted 4-fold and used as template for RT-PCR. Amplification with gene-specific primers (see Table V) for 18 to 30 cycles depending on the abundance of the transcript was evaluated by electrophoresis in 1.5% agarose gels. The transcript abundance of the constitutively expressed rnpB encoding the RNA subunit of RNaseP served as a control (Wang et al., 2004).

DNA Microarray Experiments

Direct cDNA labeling was performed as follows: 10 µg of total RNA and 0.5 µg random hexamers were used in a total volume of 15 µL. After incubation at 70°C for 10 min the samples were allowed to chill for 10 min at 4°C. Six microliters Reverse Transcriptase buffer (5x), 3 µL dithiothreitol (0.1 M), 0.6 µL dNTP mix (25 mM), and 1.4 µL milliQ water were added and mixed. As fluorescent dyes served either 2 µL Cy3 or Cy5 (1 mM, Amersham Pharmacia). Two microliters Superscript II reverse transcriptase (200 units µL^–1, Invitrogen) were added, the mixtures incubated for 10 min at RT and to start synthesis of fluorescently labeled cDNA transferred for 110 min to 42°C. RNA was hydrolyzed by addition of 1.5 µL NaOH (1 M) at 70°C for 10 min. A total of 1.5 µL HCl (1 M) was added afterward to neutralize the alkali. The samples were purified using QIAquick PCR Purification kit (Quiagen) according to the protocol of the manufacturer. Finally, the concentration and fluorescent dye incorporation was checked photometrically on a Nano-drop instrument (Nano-drop Technologies). Labeled cDNA was hybridized to 60-mer oligonucleotide DNA microarrays (Agilent) designed from the complete Synechocystis genome sequence (Kaneko et al., 1996). The probes for the microarray were designed to avoid cross hybridization with other parts of the Synechocystis genome; in addition, probes with similar melting temperature were chosen. The microarray contains 8,091 probes, covering each of the 3,264 ORFs of the Synechocystis genome two to four times (E. Aguirre von Wobeser, B.W. Ibelings, J. Huisman, and H.C.P. Matthijs, unpublished data).

The hybridization was performed according to the 60-mer oligionucleotide microarray protocol of Agilent (http://www.chem.agilent.com). Aliquots of Cy3- or Cy5-labeled cDNA were denatured at 98°C for 3 min and mixed with 125 µL hybridization buffer and 25 µL control targets (Agilent). This hybridization solution was exposed to the microarray surface in Agilent hybridization chambers for 17 h in a rotating oven at 60°C. The slides were then washed according to the manufacturer's protocol with one exception: acetonitrile was used instead of wash solution 3 to avoid the formation of precipitates on the slide. The scanning was performed with an Agilent Microarray Scanner (model G2505B) with default settings. Fluorescence intensity values were extracted using Feature Extraction software version 8.1.1.1 (Agilent). The mean signal of the pixels on each spot was used for all subsequent analysis. The values obtained for the different probes corresponding to the same gene were averaged to obtain one value per gene per sample. Given values are the means and SDs of at least two independent experiments using RNA isolated from separate cultures.

Measuring of PSI Content and Activity

Low temperature emission spectra were measured at 77 K in the low temperature unit of a Hitachi F4010 spectrofluorometer (Hitachi) using dark-adapted cells frozen in glass capillaries. The red-sensitive photomultiplier was protected from scattered excitation light by a Schott RG 620 cutoff filter. Emission was recorded at 2 nm intervals from 620 to 780 nm. Redox changes of P700 were measured at 820 nm with a pulse-amplitude-modulated system equipped with an ED-800T emitter-detector unit (PAM101-103, Walz GmbH) as described in Schreiber et al. (1988). Far-red radiation was provided by means of a cut on/cutoff filter system with transmission between 705 and 735 nm only. Dark-adapted cells were treated with increasing far-red intensities, at each intensity P700 oxidation state was measured as relative absorbance equilibrium signal (I/I, see Schreiber et al., 1988). After reaching equilibrium, a strong (approximately 12,000 µmol photons m^–2 s^–1) white light pulse (5 s) was applied to activate PSII, causing a drop in the P700 absorbance signal. The P700 absorbance signal in presence of white (PSII) light was used for analysis by averaging the readings between 1 and 4 s (240 values) after onset of the pulse. Far-red intensity was measured by means of a spectroradiometer (SR9910 Macam Photometrics Ltd.).

Protein Isolation and Immunoblotting

Cells from 20 mL of culture were harvested by centrifugation at 4,000 rpm for 5 min at 4°C and were immediately frozen at –20°C. For protein isolation the pellets were resuspended in 500 µL 0.01 M HEPES buffer (pH 7.3) supplemented with 10 mM phenylmethylsulfonyl fluoride. Under ice cooling the suspensions were sonicated (2 x 1 min, 35 W) and the homogenates were centrifuged at 4,000 rpm for 20 min at 4°C. Protein contents were estimated by the method of Bradford (1976). SDS-PAGE was performed applying 12% or 15% polyacrylamide gels containing 0.1% SDS. Exactly 5 µg denaturized protein was applied per lane. The polypeptides were stained with Coomassie Brilliant Blue G-250 (Serva) or transferred onto nitrocellulose membrane (Amersham) for immunoblot analyses. Specific antibodies were used that were raised against the Rubisco large subunit from Nicotiana tabacum (RbcL, dilution 1:3,000) or the carboxysome shell protein CsoS1 from Halothiobacillus neapolitanus (CsoS1, dilution 1:1,000), which is homologous to the cyanobacterial shell proteins CcmK1 to 4. It has to be noted that the CsoS1 antibody does not discriminate between the four different CcmK proteins. As secondary antibody served alkaline phosphatase conjugated anti-rabbit IgG (Sigma-Aldrich). Detection was accomplished via nitroblue tetrazolium/-phosphate system in reaction buffer (0.1 M Tris/HCl, 0.1 M NaCl, 0.05 M MgCl₂, pH 9.5).

Estimation of Half-Life

The half-life of the RbcL protein was analyzed by incubation of cells with lincomycin (final concentration 250 µg mL^–1; Sigma) as a specific inhibitor of translation. Wild-type cells were precultivated for 24 h under HC and LC conditions in BG11, pH 7, respectively. After adjusting optical density (OD₇₅₀ = 0.9), lincomycin (final concentration 250 µg mL^–1) was applied. Cultures were kept on bubbling with 5% CO₂ or with pure air for 24 h more. Soluble proteins were isolated from cells grown under HC conditions and from cells grown under LC conditions in the presence or absence of lincomycin. The specific RbcL protein content was estimated before and 3, 9, 6, 12, and 24 h after addition of lincomycin, respectively. The signal intensities were estimated by SigmaGel 1.0 and plotted versus time to calculate the half-life.

Electron Microscopy

Synechocystis cells were harvested by centrifugation at 4,000 rpm for 5 min at 4°C and fixed with glutaraldehyde (4%, w/v) in phosphate-buffered saline (PBS). After washing with PBS, cells were postfixed with 1% OsO₄ in PBS for 1 h, washed again, dehydrated in graded acetone concentrations (30%, 60%, 90%, 100%), and embedded in Araldite (Fluka). Ultrathin sections were prepared by an ultramicrotome (Ultracut E. Reichert, Optische Werke), placed on copper grids, stained by uranyl acetate and lead citrate, and studied with a transmission electron microscope (EM 902 A, Zeiss). Ultramicrographs were taken with a CCD camera (Proscan) and a Pentium computer, using the ITEM software (Soft Imaging Solutions, SIS).

Supplemental Data

The following materials are available in the online version of this article.

Supplemental Table S1. Log2 ratio of transcript levels in wild-type (WT) and mutant cells (glcD and gcvT) grown for 24 h under LC (0.03% CO₂) versus HC (5% CO₂) conditions.

Supplemental Table S2. Log2 ratio of transcript levels in mutant cells (glcD or gcvT) to that in wild-type (WT) cells under HC (5% CO₂).

Supplemental Table S3. Genes significantly affected by inactivation of gcvT and glcD, respectively, in Synechocystis grown at 5% CO₂.

	ACKNOWLEDGMENTS

 
We would like to thank Ms. Klaudia Michl (Plant Physiology, University of Rostock, Rostock, Germany) for excellent technical assistance. The help of Gerhard Fulda (Electron Microscopic Center of Medical Faculty, Rostock, Germany) during the ultrastructural studies and Jasper Bok (Netherlands Institute of Ecology, Nieuwersluis, The Netherlands) in work with DNA microarrays is greatly acknowledged. Specific antibodies were kindly donated by Dr. E. Pistorius (RbcL, University of Bielefeld, Bielefeld, Germany) and Dr. G.C. Cannon (CsoS1, University of Southern Mississippi, Hattiesburg, Mississippi).

Received June 4, 2007; accepted June 24, 2007; published June 28, 2007.

	FOOTNOTES

 
¹ This work was supported by a scholarship from Consejo Nacional para la Ciencia y Tecnología (Mexico) and by a grant from the Earth and Life Sciences Foundation (to E.A.v.W.), which is subsidized by the Netherlands Organization for Scientific Research (grant to Dr. Jef Huisman, University of Amsterdam). Financial support for setting up the array facility and for purchase of the arrays used in this study was provided by the Bootsma fonds at the Royal Academy of Arts and Sciences (grant to B.W.I.). This work was also supported by a grant from the Deutsche Forschungsgemeinschaft (to M.H.).

The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) is: Martin Hagemann (martin.hagemann{at}uni-rostock.de).

^[W] The online version of this article contains Web-only data.

www.plantphysiol.org/cgi/doi/10.1104/pp.107.103341

^* Corresponding author; e-mail martin.hagemann{at}uni-rostock.de; fax 49–0–3814986112.

	LITERATURE CITED

TOP ABSTRACT RESULTS AND DISCUSSION CONCLUSION MATERIALS AND METHODS LITERATURE CITED

 
Badger MR, Price GD, Long BM, Woodger FJ (2006) The environmental plasticity and ecological genomics of the cyanobacterial CO₂ concentrating mechanism. J Exp Bot 57: 249–265[Abstract/Free Full Text]

Bedu S, Pozuelos P, Cami B, Joset F (1995) Uptake of inorganic carbon in the cyanobacterium Synechocystis PCC6803: physiological and genetic evidence for a high-affinity uptake system. Mol Microbiol 18: 559–568[CrossRef][ISI][Medline]

Berry S, Schneider D, Vermaas WFJ, Roegner M (2002) Electron transport routes in whole cells of Synechocystis sp. strain PCC 6803: the role of the cytochrome bd-type oxidase. Biochemistry 41: 3422–3429[CrossRef][Medline]

Bradford M (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248–254[CrossRef][ISI][Medline]

Bukhov N, Carpentier R (2004) Alternative photosystem I-driven electron transport routes: mechanisms and functions. Photosynth Res 82: 17–33[CrossRef][ISI][Medline]

Cannon GC, Bradburne CE, Aldrich HC, Baker SH, Heinhorst S, Shively JM (2001) Microcompartments in prokaryotes: carboxysomes and related polyhedra. Appl Environ Microbiol 67: 5351–5361[Free Full Text]

Eisenhut M, Kahlon S, Hasse D, Ewald R, Lieman-Hurwitz J, Ogawa T, Ruth W, Bauwe H, Kaplan A, Hagemann M (2006) The plant-like C2 glycolate cycle and the bacterial-like glycerate pathway cooperate in phosphoglycolate metabolism in cyanobacteria. Plant Physiol 142: 333–342[Abstract/Free Full Text]

Fu J, Xu X (2006) The functional divergence of two glgP homologues in Synechocystis sp. PCC 6803. FEMS Microbiol Lett 260: 201–209[CrossRef][ISI][Medline]

Hagemann M, Jeanjean R, Fulda S, Havaux M, Erdmann N (1999) Flavodoxin accumulation contributes to enhanced cyclic electron flow around photosystem I in salt-stressed cells of Synechocystis sp. strain PCC 6803. Physiol Plant 105: 670–678[CrossRef]

Harano Y, Suzuki I, Maeda S, Kaneko T, Tabata S, Omata T (1997) Identification and nitrogen regulation of the cyanase gene from the cyanobacteria Synechocystis sp. strain PCC 6803 and Synechococcus sp. strain PCC 7942. J Bacteriol 179: 5744–5750[Abstract/Free Full Text]

Heinemann U, Engels D, Burger S, Kiziak C, Mattes R, Stolz A (2003) Cloning of a nitrilase gene from the cyanobacterium Synechocystis sp. strain PCC6803 and heterologous expression and characterization of the encoded protein. Appl Environ Microbiol 69: 4359–4366[Abstract/Free Full Text]

Helman Y, Tchernov D, Reinhold L, Shibata M, Ogawa T, Schwarz R, Ohad I, Kaplan A (2003) Genes encoding A-type flavoproteins are essential for photoreduction of O2 in cyanobacteria. Curr Biol 13: 230–235[CrossRef][ISI][Medline]

Hihara Y, Kamei A, Kanehisa M, Kaplan A, Ikeuchi M (2001) DNA microarray analysis of cyanobacterial gene expression during acclimation to high light. Plant Cell 13: 793–806[Abstract/Free Full Text]

Hisbergues M, Gaitatzes CG, Joset F, Bedu S, Smith TF (2001) A noncanonical WD-repeat protein from the cyanobacterium Synechocystis PCC6803: structural and functional study. Protein Sci 10: 293–300[Abstract/Free Full Text]

Howitt CA, Vermaas WF (1998) Quinol and cytochrome oxidases in the cyanobacterium Synechocystis sp. PCC 6803. Biochemistry 37: 17944–17951[CrossRef][Medline]

Husic DW, Husic HD, Tolbert NE (1987) The oxidative photosynthetic carbon cycle or C2 cycle. CRC Crit Rev Plant Sci 5: 45–100[ISI]

Kaneko T, Sato S, Kotani H, Tanaka A, Asamizu E, Nakamura Y, Miyajima N, Hirosawa M, Sugiura M, Sasamoto S, et al (1996) Sequence analysis of the genome of the unicellular cyanobacterium Synechocystis sp. strain PCC6803. II. Sequence determination of the entire genome and assignment of potential protein-coding regions. DNA Res 3: 109–136[Abstract]

Kaplan A, Reinhold L (1999) CO₂ concentrating mechanisms in photosynthetic microorganisms. Annu Rev Plant Physiol Plant Mol Biol 50: 539–570[CrossRef][ISI][Medline]

Kerfeld CA, Sawaya MR, Tanaka S, Nguyen CV, Phillips M, Beeby M, Yeates TO (2005) Protein structures forming the shell of primitive bacterial organelle